AU4611699A

AU4611699A - Genes for the biosynthesis of epothilones

Info

Publication number: AU4611699A
Application number: AU46116/99A
Authority: AU
Inventors: Devon Cyr; Jorn Gorlach; James Madison Ligon; Istvan Molnar; Thomas Schupp; Ross Zirkle
Original assignee: Novartis AG
Current assignee: Novartis AG
Priority date: 1998-06-18
Filing date: 1999-06-16
Publication date: 2000-01-05
Anticipated expiration: 2019-06-16
Also published as: IL139735A0; HUP0102186A2; KR100511233B1; NZ508326A; NO20006195L; IL190391A0; WO1999066028A3; CN100374565C; WO1999066028A2; CN1305530A; TR200003759T2; PL345579A1; NO20091055L; BR9911349A; HUP0102186A3; JP2008092958A; JP2002518004A; ID29128A; IL139735A; KR20010052962A

Description

WO 99/66028 PCT/EP99/04171 GENES FOR THE BIOSYNTHESIS OF EPOTHILONES FIELD OF THE INVENTION The present invention relates generally to polyketides and genes for their synthesis. In particular, the present invention relates to the isolation and characterization of novel poly ketide synthase and nonribosomal peptide synthetase genes from Sorangium cellulosum that are necessary for the biosynthesis of epothilones A and B. BACKGROUND OF THE INVENTION Polyketides are compounds synthesized from two-carbon building blocks, the carbon of which always carries a keto group, thus the name polyketide. These compounds include many important antibiotics, immunosuppressants, cancer chemotherapeutic agents, and other compounds possessing a broad range of biological properties. The tremendous structural diversity derives from the different lengths of the polyketide chain, the different side-chains introduced (either as part of the two-carbon building blocks or after the poly ketide backbone is formed), and the stereochemistry of such groups. The keto groups may also be reduced to hydroxyls, enoyls, or removed altogether. Each round of two-carbon addition is carried out by a complex of enzymes called the polyketide synthase (PKS) in a manner similar to fatty acid biosynthesis. The biosynthetic genes for an increasing number of polyketides have been isolated and sequenced. For example, see U.S. Patent Nos. 5,639,949, 5,693,774, and 5,716,849, all of which are incorporated herein by reference, which describe genes for the biosynthesis of soraphen. See also, Schupp et al., FEMS Microbiology Letters 159: 201-207 (1998) and WO 98/07868, which describe genes for the biosynthesis of rifamycin, and U.S. Patent No. 5,876,991, which describes genes for the biosynthesis of tylactone, all of which are incorpo rated herein by reference. The encoded proteins generally fall into two types: type I and type 11. Type I proteins are polyfunctional, with several catalytic domains carrying out diffe rent enzymatic steps covalently linked together (e.g. PKS for erythromycin, soraphen, rifa mycin, and avermectin (MacNeil et al., in Industrial Microorganisms: Basic and Applied Mo lecular Genetics, (ed.: Baltz et al.), American Society for Microbiology, Washington D. C.

WO 99/66028 PCT/EP99/04171 -2 pp. 245-256 (1993)); whereas type 11 proteins are monofunctional (Hutchinson et al., in Industrial Microorganisms: Basic and Applied Molecular Genetics, (ed.: Baltz et aL.), American Society for Microbiology, Washington D. C. pp. 203-216 (1993)). For the simpler polyketides such as actinorhodin (produced by Streptomyces coelicolor), the several rounds of two-carbon additions are carried out iteratively on PKS enzymes encoded by one set of PKS genes. In contrast, synthesis of the more complicated compounds such as erythromycin and soraphen involves PKS enzymes that are organized into modules, whereby each module carries out one round of two-carbon addition (for re view, see Hopwood et al., in Industrial Microorganisms: Basic and Applied Molecular Gene tics, (ed.: Baltz et al.), American Society for Microbiology, Washington D. C., pp. 267-275 (1993)). Complex polyketides and secondary metabolites in general may contain substructu res that are derived from amino acids instead of simple carboxylic acids. Incorporations of these building blocks are accomplished by non-ribosomal polypeptide synthetases (NRPSs). NRPSs are multienzymes that are organized in modules. Each module is respon sible for the addition (and the additional processing, if required) of one amino acid building block. NRPSs activate amino acids by forming aminoacyl-adenylates, and capture the acti vated amino acids on thiol groups of phophopantheteinyl prosthetic groups on peptidyl car rier protein domains. Further, NRPSs modify the amino acids by epimerization, N-methyla tion, or cyclization if necessary, and catalyse the formation of peptide bonds between the enzyme-bound amino acids. NRPSs are responsible for the biosynthesis of peptide secon dary metabolites like cyclosporin, could provide polyketide chain terminator units as in rapa mycin, or form mixed systems with PKSs as in yersiniabactin biosynthesis. Epothilones A and B are 16-membered macrocyclic polyketides with an acylcyste ine-derived starter unit that are produced by the bacterium Sorangium cellulosum strain So ce90 (Gerth et al., J. Antibiotics 49: 560-563 (1996), incorporated herein by reference). The structure of epothilone A and B wherein R signifies hydrogen (epothilone A) or methyl (epo thilone B) is: WO 99/66028 PCT/EP99/04171 -3 R O S HO N OH The epothilones have a narrow antifungal spectrum and especially show a high cytotoxicity in animal cell cultures (see, H6fle et al., Patent DE 4138042 (1993), incorpo rated herein by reference). Of significant importance, epothilones mimic the biological effects of taxol, both in vivo and in cultured cells (Bollag et al., Cancer Research 55: 2325 2333 (1995), incorporated herein by reference). Taxol and taxotere, which stabilize cellular microtubules, are cancer chemotherapeutic agents with significant activity against various human solid tumors (Rowinsky et al., J. Nat. Cancer Inst. 83: 1778-1781 (1991)). Competi tion studies have revealed that epothilones act as competitive inhibitors of taxol binding to microtubules, consistent with the interpretation that they share the same microtubule-bin ding site and possess a similar microtubule affinity as taxol. However, epothilones enjoy a significant advantage over taxol in that epothilones exhibit a much lower drop in potency compared to taxol against a multiple drug-resistant cell line (Bollag et al. (11995)). Further more, epothilones are considerably less efficiently exported from the cells by P-glycoprotein than is taxol (Gerth et al. (1996)). In addition, several epothilone analogs have been syn thesized that have a superior cytotoxic activity as compared to epothilone A or epothilone B as demonstrated by their enhanced ability to induce the polymerization and stabilization of microtubules (WO 98/25929, incorporated herein by reference). Despite the promise shown by the epothilones as anticancer agents, problems per taining to the production of these compounds presently limit their commercial potential. The compounds are too complex for industrial-scale chemical synthesis and so must be produ ced by fermentation. Techniques for the genetic manipulation of myxobacteria such as Sorangium cellulosum are described in U.S. Patent No. 5,686,295, incorporated herein by reference. However, Sorangium cellulosum is notoriously difficult to ferment and production levels of epothilones are therefore low. Recombinant production of epothilones in hetero logous hosts that are more amenable to fermentation could solve current production pro blems. However, the genes that encode the polypeptides responsible for epothilone bio- WO 99/66028 PCT/EP99/04171 -4 synthesis have heretofore not been isolated. Furthermore, the strain that produces epo thilones, i.e. So ce90, also produces at least one additional polyketide, spirangien, which would be expected to greatly complicate the isolation of the genes particularly responsible for epothilone biosynthesis. Therefore, in view of the foregoing, one object of the present invention is to isolate the genes that are involved in the synthesis of epothilones, particularly the genes that are involved in the synthesis of epothilones A and B in myxobacteria of the Sorangium/ Polyangium group, i.e., Sorangium cellulosum strain So ce90. A further object of the invention is to provide a method for the recombinant production of epothilones for application in anticancer formulations. SUMMARY OF THE INVENTION In furtherance of the aforementioned and other objects, the present invention unex pectedly overcomes the difficulties set forth above to provide for the first time a nucleic acid molecule comprising a nucleotide sequence that encodes at least one polypeptide involved in the biosynthesis of epothilone. In a preferred embodiment, the nucleotide sequence is isolated from a species belonging to Myxobacteria, most preferably Sorangium cellulosum. In another preferred embodiment, the present invention provides an isolated nucleic acid molecule comprising a nucleotide sequence that encodes at least one polypeptide in volved in the biosynthesis of an epothilone, wherein said polypeptide comprises an amino acid sequence substantially similar to an amino acid sequence selected from the group con sisting of: SEQ ID NO:2, amino acids 11-437 of SEQ ID NO:2, amino acids 543-864 of SEQ ID NO:2, amino acids 974-1273 of SEQ ID NO:2, amino acids 1314-1385 of SEQ ID NO:2, SEQ ID NO:3, amino acids 72-81 of SEQ ID NO:3, amino acids 118-125 of SEQ ID NO:3, amino acids 199-212 of SEQ ID NO:3, amino acids 353-363 of SEQ ID NO:3, amino acids 549-565 of SEQ ID NO:3, amino acids 588-603 of SEQ ID NO:3, amino acids 669-684 of SEQ ID NO:3, amino acids 815-821 of SEQ ID NO:3, amino acids 868-892 of SEQ ID NO:3, amino acids 903-912 of SEQ ID NO:3, amino acids 918-940 of SEQ ID NO:3, amino acids 1268-1274 of SEQ ID NO:3, amino acids 1285-1297 of SEQ ID NO:3, amino acids 973 1256 of SEQ ID NO:3, amino acids 1344-1351 of SEQ ID NO:3, SEQ ID NO:4, amino acids 7-432 of SEQ ID NO:4, amino acids 539-859 of SEQ ID NO:4, amino acids 869-1037 of SEQ ID NO:4, amino acids 1439-1684 of SEQ ID NO:4, amino acids 1722-1792 of SEQ ID WO 99/66028 PCT/EP99/04171 -5 NO:4, SEQ ID NO:5, amino acids 39-457 of SEQ ID NO:5, amino acids 563-884 of SEQ ID NO:5, amino acids 1147-1399 of SEQ ID NO:5, amino acids 1434-1506 of SEQ ID NO:5, amino acids 1524-1950 of SEQ ID NO:5, amino acids 2056-2377 of SEQ ID NO:5, amino acids 2645-2895 of SEQ ID NO:5, amino acids 2932-3005 of SEQ ID NO:5, amino acids 3024-3449 of SEQ ID NO:5, amino acids 3555-3876 of SEQ ID NO:5, amino acids 3886 4048 of SEQ ID NO:5, amino acids 4433-4719 of SEQ ID NO:5, amino acids 4729-4974 of SEQ ID NO:5, amino acids 5010-5082 of SEQ ID NO:5, amino acids 5103-5525 of SEQ ID NO:5, amino acids 5631-5951 of SEQ ID NO:5, amino acids 5964-6132 of SEQ ID NO:5, amino acids 6542-6837 of SEQ ID NO:5, amino acids 6857-7101 of SEQ ID NO:5, amino acids 7140-7211 of SEQ ID NO:5, SEQ ID NO:6, amino acids 35-454 of SEQ ID NO:6, amino acids 561-881 of SEQ ID NO:6, amino acids 1143-1393 of SEQ ID NO:6, amino acids 1430-1503 of SEQ ID NO:6, amino acids 1522-1946 of SEQ ID NO: 6, amino acids 2053-2373 of SEQ ID NO:6, amino acids 2383-2551 of SEQ ID NO:6, amino acids 2671 3045 of SEQ ID NO:6, amino acids 3392-3636 of SEQ ID NO:6, amino acids 3673-3745 of SEQ ID NO:6, SEQ ID NO:7, amino acids 32-450 of SEQ ID NO:7, amino acids 556-877 of SEQ ID NO:7, amino acids 887-1051 of SEQ ID NO:7, amino acids 1478-1790 of SEQ ID NO:7, amino acids 1810-2055 of SEQ ID NO:7, amino acids 2093-2164 of SEQ ID NO:7, amino acids 2165-2439 of SEQ ID NO:7, SEQ ID NO:8, SEQ ID NO:10, SEQ ID NO:11, and SEQ ID NO:22. In a more preferred embodiment, the present invention provides an isolated nucleic acid molecule comprising a nucleotide sequence that encodes at least one polypeptide in volved in the biosynthesis of an epothilone, wherein said polypeptide comprises an amino acid sequence selected from the group consisting of: SEQ ID NO:2, amino acids 11-437 of SEQ ID NO:2, amino acids 543-864 of SEQ ID NO:2, amino acids 974-1273 of SEQ ID NO:2, amino acids 1314-1385 of SEQ ID NO:2, SEQ ID NO:3, amino acids 72-81 of SEQ ID NO:3, amino acids 118-125 of SEQ ID NO:3, amino acids 199-212 of SEQ ID NO:3, amino acids 353-363 of SEQ ID NO:3, amino acids 549-565 of SEQ ID NO:3, amino acids 588 603 of SEQ ID NO:3, amino acids 669-684 of SEQ ID NO:3, amino acids 815-821 of SEQ ID NO:3, amino acids 868-892 of SEQ ID NO:3, amino acids 903-912 of SEQ ID NO:3, amino acids 918-940 of SEQ ID NO:3, amino acids 1268-1274 of SEQ ID NO:3, amino acids 1285-1297 of SEQ ID NO:3, amino acids 973-1256 of SEQ ID NO:3, amino acids 1344-1351 of SEQ ID NO:3, SEQ ID NO:4, amino acids 7-432 of SEQ ID NO:4, amino acids 539-859 of SEQ ID NO:4, amino acids 869-1037 of SEQ ID NO:4, amino acids 1439-1684 WO 99/66028 PCT/EP99/04171 -6 of SEQ ID NO:4, amino acids 1722-1792 of SEQ ID NO:4, SEQ ID NO:5, amino acids 39 457 of SEQ ID NO:5, amino acids 563-884 of SEQ ID NO:5, amino acids 1147-1399 of SEQ ID NO:5, amino acids 1434-1506 of SEQ ID NO:5, amino acids 1524-1950 of SEQ ID NO:5, amino acids 2056-2377 of SEQ ID NO:5, amino acids 2645-2895 of SEQ ID NO:5, amino acids 2932-3005 of SEQ ID NO:5, amino acids 3024-3449 of SEQ ID NO:5, amino acids 3555-3876 of SEQ ID NO:5, amino acids 3886-4048 of SEQ ID NO:5, amino acids 4433-4719 of SEQ ID NO:5, amino acids 4729-4974 of SEQ ID NO:5, amino acids 5010 5082 of SEQ ID NO:5, amino acids 5103-5525 of SEQ ID NO:5, amino acids 5631-5951 of SEQ ID NO:5, amino acids 5964-6132 of SEQ ID NO:5, amino acids 6542-6837 of SEQ ID NO:5, amino acids 6857-7101 of SEQ ID NO:5, amino acids 7140-7211 of SEQ ID NO:5, SEQ ID NO:6, amino acids 35-454 of SEQ ID NO:6, amino acids 561-881 of SEQ ID NO:6, amino acids 1143-1393 of SEQ ID NO:6, amino acids 1430-1503 of SEQ ID NO:6, amino acids 1522-1946 of SEQ ID NO: 6, amino acids 2053-2373 of SEQ ID NO:6, amino acids 2383-2551 of SEQ ID NO:6, amino acids 2671-3045 of SEQ ID NO:6, amino acids 3392 3636 of SEQ ID NO:6, amino acids 3673-3745 of SEQ ID NO:6, SEQ ID NO:7, amino acids 32-450 of SEQ ID NO:7, amino acids 556-877 of SEQ ID NO:7, amino acids 887-1051 of SEQ ID NO:7, amino acids 1478-1790 of SEQ ID NO:7, amino acids 1810-2055 of SEQ ID NO:7, amino acids 2093-2164 of SEQ ID NO:7, amino acids 2165-2439 of SEQ ID NO:7, SEQ ID NO:8, SEQ ID NO:10, SEQ ID NO:11, and SEQ ID NO:22. In yet another preferred embodiment, the present invention provides an isolated nucleic acid molecule comprising a nucleotide sequence that encodes at least one polypep tide involved in the biosynthesis of an epothilone, wherein said nucleotide sequence is substantially similar to a nucleotide sequence selected from the group consisting of: the complement of nucleotides 1900-3171 of SEQ ID NO:1, nucleotides 3415-5556 of SEQ ID NO:1, nucleotides 7610-11875 of SEQ ID NO:1, nucleotides 7643-8920 of SEQ ID NO:1, nucleotides 9236-10201 of SEQ ID NO:1, nucleotides 10529-11428 of SEQ ID NO:1, nucleotides 11549-11764 of SEQ ID NO:1, nucleotides 11872-16104 of SEQ ID NO:1, nucleotides 12085-12114 of SEQ ID NO:1, nucleotides 12223-12246 of SEQ ID NO:1, nucleotides 12466-12507 of SEQ ID NO:1, nucleotides 12928-12960 of SEQ ID NO:1, nucleotides 13516-13566 of SEQ ID NO:1, nucleotides 13633-13680 of SEQ ID NO:1, nucleotides 13876-13923 of SEQ ID NO:1, nucleotides 14313-14334 of SEQ ID NO:1, nucleotides 14473-14547 of SEQ ID NO:1, nucleotides 14578-14607 of SEQ ID NO:1, nucleotides 14623-14692 of SEQ ID NO:1, nucleotides 15673-15693 of SEQ ID NO:1, WO 99/66028 PCT/EP99/04171 -7 nucleotides 15724-15762 of SEQ ID NO:1, nucleotides 14788-15639 of SEQ ID NO:1, nucleotides 15901-15924 of SEQ ID NO:1, nucleotides 16251-21749 of SEQ ID NO:1, nucleotides 16269-17546 of SEQ ID NO:1, nucleotides 17865-18827 of SEQ ID NO:1, nucleotides 18855-19361 of SEQ ID NO:1, nucleotides 20565-21302 of SEQ ID NO:1, nucleotides 21414-21626 of SEQ ID NO:1, nucleotides 21746-43519 of SEQ ID NO:1, nucleotides 21860-23116 of SEQ ID NO:1, nucleotides 23431-24397 of SEQ ID NO:1, nucleotides 25184-25942 of SEQ ID NO:1, nucleotides 26045-26263 of SEQ ID NO:1, nucleotides 26318-27595 of SEQ ID NO:1, nucleotides 27911-28876 of SEQ ID NO:1, nucleotides 29678-30429 of SEQ ID NO:1, nucleotides 30539-30759 of SEQ ID NO:1, nucleotides 30815-32092 of SEQ ID NO:1, nucleotides 32408-33373 of SEQ ID NO:1, nucleotides 33401-33889 of SEQ ID NO:1, nucleotides 35042-35902 of SEQ ID NO:1, nucleotides 35930-36667 of SEQ ID NO:1, nucleotides 36773-36991 of SEQ ID NO:1, nucleotides 37052-38320 of SEQ ID NO:1, nucleotides 38636-39598 of SEQ ID NO:1, nucleotides 39635-40141 of SEQ ID NO:1, nucleotides 41369-42256 of SEQ ID NO:1, nucleotides 42314-43048 of SEQ ID NO:1, nucleotides 43163-43378 of SEQ ID NO:1, nucleotides 43524-54920 of SEQ ID NO:1, nucleotides 43626-44885 of SEQ ID NO:1, nucleotides 45204-46166 of SEQ ID NO:1, nucleotides 46950-47702 of SEQ ID NO:1, nucleotides 47811-48032 of SEQ ID NO:1, nucleotides 48087-49361 of SEQ ID NO:1, nucleotides 49680-50642 of SEQ ID NO:1, nucleotides 50670-51176 of SEQ ID NO:1, nucleotides 51534-52657 of SEQ ID NO:1, nucleotides 53697-54431 of SEQ ID NO:1, nucleotides 54540-54758 of SEQ ID NO:1, nucleotides 54935-62254 of SEQ ID NO:1, nucleotides 55028-56284 of SEQ ID NO:1, nucleotides 56600-57565 of SEQ ID NO:1, nucleotides 57593-58087 of SEQ ID NO:1, nucleotides 59366-60304 of SEQ ID NO:1, nucleotides 60362-61099 of SEQ ID NO:1, nucleotides 61211-61426 of SEQ ID NO:1, nucleotides 61427-62254 of SEQ ID NO:1, nucleotides 62369-63628 of SEQ ID NO:1, nucleotides 67334-68251 of SEQ ID NO:1, and nucleotides 1-68750 SEQ ID NO:1. In an especially preferred embodiment, the present invention provides a nucleic acid molecule comprising a nucleotide sequence that encodes at least one polypeptide involved in the biosynthesis of an epothilone, wherein said nucleotide sequence is selected from the group consisting of: the complement of nucleotides 1900-3171 of SEQ ID NO:1, nucleotides 3415-5556 of SEQ ID NO:1, nucleotides 7610-11875 of SEQ ID NO:1, nucleotides 7643 8920 of SEQ ID NO:1, nucleotides 9236-10201 of SEQ ID NO:1, nucleotides 10529-11428 of SEQ ID NO:1, nucleotides 11549-11764 of SEQ ID NO:1, nucleotides 11872-16104 of WO 99/66028 PCT/EP99/04171 -8 SEQ ID NO:1, nucleotides 12085-12114 of SEQ ID NO:l, nucleotides 12223-12246 of SEQ ID NO:1, nucleotides 12466-12507 of SEQ ID NO:1, nucleotides 12928-12960 of SEQ ID NO:1, nucleotides 13516-13566 of SEQ ID NO:1, nucleotides 13633-13680 of SEQ ID NO:1, nucleotides 13876-13923 of SEQ ID NO:1, nucleotides 14313-14334 of SEQ ID NO:1, nucleotides 14473-14547 of SEQ ID NO:1, nucleotides 14578-14607 of SEQ ID NO:1, nucleotides 14623-14692 of SEQ ID NO:1, nucleotides 15673-15693 of SEQ ID NO:1, nucleotides 15724-15762 of SEQ ID NO:1, nucleotides 14788-15639 of SEQ ID NO:1, nucleotides 15901-15924 of SEQ ID NO:1, nucleotides 16251-21749 of SEQ ID NO:1, nucleotides 16269-17546 of SEQ ID NO:1, nucleotides 17865-18827 of SEQ ID NO:1, nucleotides 18855-19361 of SEQ ID NO:1, nucleotides 20565-21302 of SEQ ID NO:1, nucleotides 21414-21626 of SEQ ID NO:1, nucleotides 21746-43519 of SEQ ID NO:1, nucleotides 21860-23116 of SEQ ID NO:1, nucleotides 23431-24397 of SEQ ID NO:1, nucleotides 25184-25942 of SEQ ID NO:1, nucleotides 26045-26263 of SEQ ID NO:1, nucleotides 26318-27595 of SEQ ID NO:1, nucleotides 27911-28876 of SEQ ID NO:1, nucleotides 29678-30429 of SEQ ID NO:1, nucleotides 30539-30759 of SEQ ID NO:1, nucleotides 30815-32092 of SEQ ID NO:1, nucleotides 32408-33373 of SEQ ID NO:1, nucleotides 33401-33889 of SEQ ID NO:1, nucleotides 35042-35902 of SEQ ID NO:1, nucleotides 35930-36667 of SEQ ID NO:1, nucleotides 36773-36991 of SEQ ID NO:1, nucleotides 37052-38320 of SEQ ID NO:1, nucleotides 38636-39598 of SEQ ID NO:1, nucleotides 39635-40141 of SEQ ID NO:1, nucleotides 41369-42256 of SEQ ID NO:1, nucleotides 42314-43048 of SEQ ID NO:1, nucleotides 43163-43378 of SEQ ID NO:1, nucleotides 43524-54920 of SEQ ID NO:1, nucleotides 43626-44885 of SEQ ID NO:1, nucleotides 45204-46166 of SEQ ID NO:1, nucleotides 46950-47702 of SEQ ID NO:1, nucleotides 47811-48032 of SEQ ID NO:1, nucleotides 48087-49361 of SEQ ID NO:1, nucleotides 49680-50642 of SEQ ID NO:1, nucleotides 50670-51176 of SEQ ID NO:1, nucleotides 51534-52657 of SEQ ID NO:1, nucleotides 53697-54431 of SEQ ID NO:1, nucleotides 54540-54758 of SEQ ID NO:1, nucleotides 54935-62254 of SEQ ID NO:1, nucleotides 55028-56284 of SEQ ID NO:1, nucleotides 56600-57565 of SEQ ID NO:1, nucleotides 57593-58087 of SEQ ID NO:1, nucleotides 59366-60304 of SEQ ID NO:1, nucleotides 60362-61099 of SEQ ID NO:1, nucleotides 61211-61426 of SEQ ID NO:1, nucleotides 61427-62254 of SEQ ID NO:1, nucleotides 62369-63628 of SEQ ID NO:1, nucleotides 67334-68251 of SEQ ID NO:1, and nucleotides 1-68750 SEQ ID NO:1.

WO 99/66028 PCT/EP99/04171 -9 In yet another preferred embodiment, the present invention provides an isolated nucleic acid molecule comprising a nucleotide sequence that encodes at least one polypeptide involved in the biosynthesis of an epothilone, wherein said nucleotide sequence comprises a consecutive 20, 25, 30, 35, 40, 45, or 50 (preferably 20) base pair nucleotide portion identical in sequence to a respective consecutive 20, 25, 30, 35, 40, 45, or 50 (preferably 20) base pair portion of a nucleotide sequence selected from the group consisting of: the complement of nucleotides 1900-3171 of SEQ ID NO:1, nucleotides 3415 5556 of SEQ ID NO:1, nucleotides 7610-11875 of SEQ ID NO:1, nucleotides 7643-8920 of SEQ ID NO:1, nucleotides 9236-10201 of SEQ ID NO:1, nucleotides 10529-11428 of SEQ ID NO:1, nucleotides 11549-11764 of SEQ ID NO:1, nucleotides 11872-16104 of SEQ ID NO:1, nucleotides 12085-12114 of SEQ ID NO:1, nucleotides 12223-12246 of SEQ ID NO:1, nucleotides 12466-12507 of SEQ ID NO:1, nucleotides 12928-12960 of SEQ ID NO:1, nucleotides 13516-13566 of SEQ ID NO:1, nucleotides 13633-13680 of SEQ ID NO:1, nucleotides 13876-13923 of SEQ ID NO:1, nucleotides 14313-14334 of SEQ ID NO:1, nucleotides 14473-14547 of SEQ ID NO:1, nucleotides 14578-14607 of SEQ ID NO:1, nucleotides 14623-14692 of SEQ ID NO:1, nucleotides 15673-15693 of SEQ ID NO:1, nucleotides 15724-15762 of SEQ ID NO:1, nucleotides 14788-15639 of SEQ ID NO:1, nucleotides 15901-15924 of SEQ ID NO:1, nucleotides 16251-21749 of SEQ ID NO:1, nucleotides 16269-17546 of SEQ ID NO:1, nucleotides 17865-18827 of SEQ ID NO:1, nucleotides 18855-19361 of SEQ ID NO:1, nucleotides 20565-21302 of SEQ ID NO:1, nucleotides 21414-21626 of SEQ ID NO:1, nucleotides 21746-43519 of SEQ ID NO:1, nucleotides 21860-23116 of SEQ ID NO:1, nucleotides 23431-24397 of SEQ ID NO:1, nucleotides 25184-25942 of SEQ ID NO:1, nucleotides 26045-26263 of SEQ ID NO:1, nucleotides 26318-27595 of SEQ ID NO:1, nucleotides 27911-28876 of SEQ ID NO:1, nucleotides 29678-30429 of SEQ ID NO:1, nucleotides 30539-30759 of SEQ ID NO:1, nucleotides 30815-32092 of SEQ ID NO:1, nucleotides 32408-33373 of SEQ ID NO:1, nucleotides 33401-33889 of SEQ ID NO:1, nucleotides 35042-35902 of SEQ ID NO:1, nucleotides 35930-36667 of SEQ ID NO:1, nucleotides 36773-36991 of SEQ ID NO:1, nucleotides 37052-38320 of SEQ ID NO:1, nucleotides 38636-39598 of SEQ ID NO:1, nucleotides 39635-40141 of SEQ ID NO:1, nucleotides 41369-42256 of SEQ ID NO:1, nucleotides 42314-43048 of SEQ ID NO:1, nucleotides 43163-43378 of SEQ ID NO:1, nucleotides 43524-54920 of SEQ ID NO:1, nucleotides 43626-44885 of SEQ ID NO:1, nucleotides 45204-46166 of SEQ ID NO:1, nucleotides 46950-47702 of SEQ ID WO 99/66028 PCT/EP99/04171 -10 NO:1, nucleotides 47811-48032 of SEQ ID NO:1, nucleotides 48087-49361 of SEQ ID NO:1, nucleotides 49680-50642 of SEQ ID NO:1, nucleotides 50670-51176 of SEQ ID NO:1, nucleotides 51534-52657 of SEQ ID NO:1, nucleotides 53697-54431 of SEQ ID NO:1, nucleotides 54540-54758 of SEQ ID NO:1, nucleotides 54935-62254 of SEQ ID NO:1, nucleotides 55028-56284 of SEQ ID NO:1, nucleotides 56600-57565 of SEQ ID NO:1, nucleotides 57593-58087 of SEQ ID NO:1, nucleotides 59366-60304 of SEQ ID NO:1, nucleotides 60362-61099 of SEQ ID NO:1, nucleotides 61211-61426 of SEQ ID NO:1, nucleotides 61427-62254 of SEQ ID NO:1, nucleotides 62369-63628 of SEQ ID NO:1, nucleotides 67334-68251 of SEQ ID NO:1, and nucleotides 1-68750 SEQ ID NO:1. The present invention also provides a chimeric gene comprising a heterologous pro moter sequence operatively linked to a nucleic acid molecule of the invention. Further, the present invention provides a recombinant vector comprising such a chimeric gene, wherein the vector is capable of being stably transformed into a host cell. Still further, the present invention provides a recombinant host cell comprising such a chimeric gene, wherein the host cell is capable of expressing the nucleotide sequence that encodes at least one poly peptide necessary for the biosynthesis of an epothilone. In a preferred embodiment, the recombinant host cell is a bacterium belonging to the order Actinomycetales, and in a more preferred embodiment the recombinant host cell is a strain of Streptomyces. In other embo diments, the recombinant host cell is any other bacterium amenable to fermentation, such as a pseudomonad or E. coli. Even further, the present invention provides a Bac clone comprising a nucleic acid molecule of the invention, preferably Bac clone pEPO15. In another aspect, the present invention provides an isolated nucleic acid molecule comprising a nucleotide sequence that encodes an epothilone synthase domain. According to one embodiment, the epothilone synthase domain is a @-ketoacyl-syn thase (KS) domain comprising an amino acid sequence substantially similar to an amino acid sequence selected from the group consisting of: amino acids 11-437 of SEQ ID NO:2, amino acids 7-432 of SEQ ID NO:4, amino acids 39-457 of SEQ ID NO:5, amino acids 1524-1950 of SEQ ID NO:5, amino acids 3024-3449 of SEQ ID NO:5, amino acids 5103 5525 of SEQ ID NO:5, amino acids 35-454 of SEQ ID NO:6, amino acids 1522-1946 of SEQ ID NO: 6, and amino acids 32-450 of SEQ ID NO:7. According to this embodiment, said KS domain preferably comprises an amino acid sequence selected from the group consisting of: amino acids 11-437 of SEQ ID NO:2, amino acids 7-432 of SEQ ID NO:4, amino acids 39-457 of SEQ ID NO:5, amino acids 1524-1950 of SEQ ID NO:5, amino acids WO 99/66028 PCT/EP99/04171 - 11 3024-3449 of SEQ ID NO:5, amino acids 5103-5525 of SEQ ID NO:5, amino acids 35-454 of SEQ ID NO:6, amino acids 1522-1946 of SEQ ID NO: 6, and amino acids 32-450 of SEQ ID NO:7. Also, according to this embodiment, said nucleotide sequence preferably is substantially similar to a nucleotide sequence selected from the group consisting of: nucleo tides 7643-8920 of SEQ ID NO:1, nucleotides 16269-17546 of SEQ ID NO:1, nucleotides 21860-23116 of SEQ ID NO:1, nucleotides 26318-27595 of SEQ ID NO:1, nucleotides 30815-32092 of SEQ ID NO:1, nucleotides 37052-38320 of SEQ ID NO:1, nucleotides 43626-44885 of SEQ ID NO:1, nucleotides 48087-49361 of SEQ ID NO:1, and nucleotides 55028-56284 of SEQ ID NO:1. According to this embodiment, said nucleotide sequence more preferably comprises a consecutive 20, 25, 30, 35, 40, 45, or 50 (preferably 20) base pair nucleotide portion identical in sequence to a respective consecutive 20, 25, 30, 35, 40, 45, or 50 (preferably 20) base pair portion of a nucleotide sequence selected from the group consisting of: nucleotides 7643-8920 of SEQ ID NO:1, nucleotides 16269-17546 of SEQ ID NO:1, nucleotides 21860-23116 of SEQ ID NO:1, nucleotides 26318-27595 of SEQ ID NO:1, nucleotides 30815-32092 of SEQ ID NO:1, nucleotides 37052-38320 of SEQ ID NO:1, nucleotides 43626-44885 of SEQ ID NO:1, nucleotides 48087-49361 of SEQ ID NO:1, and nucleotides 55028-56284 of SEQ ID NO:1. In addition, according to this embo diment, said nucleotide sequence most preferably is selected from the group consisting of: nucleotides 7643-8920 of SEQ ID NO:1, nucleotides 16269-17546 of SEQ ID NO:1, nucle otides 21860-23116 of SEQ ID NO:1, nucleotides 26318-27595 of SEQ ID NO:1, nucleo tides 30815-32092 of SEQ ID NO:1, nucleotides 37052-38320 of SEQ ID NO:1, nucleotides 43626-44885 of SEQ ID NO:1, nucleotides 48087-49361 of SEQ ID NO:1, and nucleotides 55028-56284 of SEQ ID NO:1. According to another embodiment, the epothilone synthase domain is an acyltrans ferase (AT) domain comprising an amino acid sequence substantially similar to an amino acid sequence selected from the group consisting of: amino acids 543-864 of SEQ ID NO:2, amino acids 539-859 of SEQ ID NO:4, amino acids 563-884 of SEQ ID NO:5, amino acids 2056-2377 of SEQ ID NO:5, amino acids 3555-3876 of SEQ ID NO:5, amino acids 5631 5951 of SEQ ID NO:5, amino acids 561-881 of SEQ ID NO:6, amino acids 2053-2373 of SEQ ID NO:6, and amino acids 556-877 of SEQ ID NO:7. According to this embodiment, said AT domain preferably comprises an amino acid sequence selected from the group consisting of: amino acids 543-864 of SEQ ID NO:2, amino acids 539-859 of SEQ ID NO:4, amino acids 563-884 of SEQ ID NO:5, amino acids 2056-2377 of SEQ ID NO:5, amino WO 99/66028 PCT/EP99/04171 -12 acids 3555-3876 of SEQ ID NO:5, amino acids 5631-5951 of SEQ ID NO:5, amino acids 561-881 of SEQ ID NO:6, amino acids 2053-2373 of SEQ ID NO:6, and amino acids 556 877 of SEQ ID NO:7. Also, according to this embodiment, said nucleotide sequence pre ferably is substantially similar to a nucleotide sequence selected from the group consisting of: nucleotides 9236-10201 of SEQ ID NO:1, nucleotides 17865-18827 of SEQ ID NO:1, nucleotides 23431-24397 of SEQ ID NO:1, nucleotides 27911-28876 of SEQ ID NO:1, nucleotides 32408-33373 of SEQ ID NO:1, nucleotides 38636-39598 of SEQ ID NO:1, nucleotides 45204-46166 of SEQ ID NO:1, nucleotides 49680-50642 of SEQ ID NO:1, and nucleotides 56600-57565 of SEQ ID NO:1. According to this embodiment, said nucleotide sequence more preferably comprises a consecutive 20, 25, 30, 35, 40, 45, or 50 (preferably 20) base pair nucleotide portion identical in sequence to a respective consecutive 20, 25, 30, 35, 40, 45, or 50 (preferably 20) base pair portion of a nucleotide sequence selected from the group consisting of: nucleotides 9236-10201 of SEQ ID NO:1, nucleotides 17865 18827 of SEQ ID NO:1, nucleotides 23431-24397 of SEQ ID NO:1, nucleotides 27911 28876 of SEQ ID NO:1, nucleotides 32408-33373 of SEQ ID NO:1, nucleotides 38636 39598 of SEQ ID NO:1, nucleotides 45204-46166 of SEQ ID NO:1, nucleotides 49680 50642 of SEQ ID NO:1, and nucleotides 56600-57565 of SEQ ID NO:1. In addition, accor ding to this embodiment, said nucleotide sequence most preferably is selected from the group consisting of: nucleotides 9236-10201 of SEQ ID NO:1, nucleotides 17865-18827 of SEQ ID NO:1, nucleotides 23431-24397 of SEQ ID NO:1, nucleotides 27911-28876 of SEQ ID NO:1, nucleotides 32408-33373 of SEQ ID NO:1, nucleotides 38636-39598 of SEQ ID NO:1, nucleotides 45204-46166 of SEQ ID NO:1, nucleotides 49680-50642 of SEQ ID NO:1, and nucleotides 56600-57565 of SEQ ID NO:1. According to still another embodiment, the epothilone synthase domain is an enoyl reductase (ER) domain comprising an amino acid sequence substantially similar to an ami no acid sequence selected from the group consisting of: amino acids 974-1273 of SEQ ID NO:2, amino acids 4433-4719 of SEQ ID NO:5, amino acids 6542-6837 of SEQ ID NO:5, and amino acids 1478-1790 of SEQ ID NO:7. According to this embodiment, said ER do main preferably comprises an amino acid sequence selected from the group consisting of: amino acids 974-1273 of SEQ ID NO:2, amino acids 4433-4719 of SEQ ID NO:5, amino acids 6542-6837 of SEQ ID NO:5, and amino acids 1478-1790 of SEQ ID NO:7. Also, ac cording to this embodiment, said nucleotide sequence preferably is substantially similar to a nucleotide sequence selected from the group consisting of: nucleotides 10529-11428 of WO 99/66028 PCT/EP99/04171 -13 SEQ ID NO:1, nucleotides 35042-35902 of SEQ ID NO:1, nucleotides 41369-42256 of SEQ ID NO:1, and nucleotides 59366-60304 of SEQ ID NO:1. According to this embodiment, said nucleotide sequence more preferably comprises a consecutive 20, 25, 30, 35, 40, 45, or 50 (preferably 20) base pair nucleotide portion identical in sequence to a respective consecutive 20, 25, 30, 35, 40, 45, or 50 (preferably 20) base pair portion of a nucleotide sequence selected from the group consisting of: nucleotides 10529-11428 of SEQ ID NO:1, nucleotides 35042-35902 of SEQ ID NO:1, nucleotides 41369-42256 of SEQ ID NO:1, and nucleotides 59366-60304 of SEQ ID NO:1. In addition, according to this embodiment, said nucleotide sequence most preferably is selected from the group consisting of: nucleotides 10529-11428 of SEQ ID NO:1, nucleotides 35042-35902 of SEQ ID NO:1, nucleotides 41369-42256 of SEQ ID NO:1, and nucleotides 59366-60304 of SEQ ID NO:1. According to another embodiment, the epothilone synthase domain is an acyl carrier protein (ACP) domain, wherein said polypeptide comprises an amino acid sequence substantially similar to an amino acid sequence selected from the group consisting of: amino acids 1314-1385 of SEQ ID NO:2, amino acids 1722-1792 of SEQ ID NO:4, amino acids 1434-1506 of SEQ ID NO:5, amino acids 2932-3005 of SEQ ID NO:5, amino acids 5010-5082 of SEQ ID NO:5, amino acids 7140-7211 of SEQ ID NO:5, amino acids 1430 1503 of SEQ ID NO:6, amino acids 3673-3745 of SEQ ID NO:6, and amino acids 2093 2164 of SEQ ID NO:7. According to this embodiment, said ACP domain preferably comprises an amino acid sequence selected from the group consisting of: amino acids 1314-1385 of SEQ ID NO:2, amino acids 1722-1792 of SEQ ID NO:4, amino acids 1434 1506 of SEQ ID NO:5, amino acids 2932-3005 of SEQ ID NO:5, amino acids 5010-5082 of SEQ ID NO:5, amino acids 7140-7211 of SEQ ID NO:5, amino acids 1430-1503 of SEQ ID NO:6, amino acids 3673-3745 of SEQ ID NO:6, and amino acids 2093-2164 of SEQ ID NO:7. Also, according to this embodiment, said nucleotide sequence preferably is substan tially similar to a nucleotide sequence selected from the group consisting of: nucleotides 11549-11764 of SEQ ID NO:1, nucleotides 21414-21626 of SEQ ID NO:1, nucleotides 26045-26263 of SEQ ID NO:1, nucleotides 30539-30759 of SEQ ID NO:1, nucleotides 36773-36991 of SEQ ID NO:1, nucleotides 43163-43378 of SEQ ID NO:1, nucleotides 47811-48032 of SEQ ID NO:1, nucleotides 54540-54758 of SEQ ID NO:1, and nucleotides 61211-61426 of SEQ ID NO:1. According to this embodiment, said nucleotide sequence more preferably comprises a consecutive 20, 25, 30, 35, 40, 45, or 50 (preferably 20) base pair nucleotide portion identical in sequence to a respective consecutive 20, 25, 30, 35, 40, WO 99/66028 PCT/EP99/04171 -14 45, or 50 (preferably 20) base pair portion of a nucleotide sequence selected from the group consisting of: nucleotides 11549-11764 of SEQ ID NO:1, nucleotides 21414-21626 of SEQ ID NO:1, nucleotides 26045-26263 of SEQ ID NO:1, nucleotides 30539-30759 of SEQ ID NO:1, nucleotides 36773-36991 of SEQ ID NO:1, nucleotides 43163-43378 of SEQ ID NO:1, nucleotides 47811-48032 of SEQ ID NO:1, nucleotides 54540-54758 of SEQ ID NO:1, and nucleotides 61211-61426 of SEQ ID NO:1. In addition, according to this embodi ment, said nucleotide sequence most preferably is selected from the group consisting of: nucleotides 11549-11764 of SEQ ID NO:1, nucleotides 21414-21626 of SEQ ID NO:1, nucleotides 26045-26263 of SEQ ID NO:1, nucleotides 30539-30759 of SEQ ID NO:1, nucleotides 36773-36991 of SEQ ID NO:1, nucleotides 43163-43378 of SEQ ID NO:1, nucleotides 47811-48032 of SEQ ID NO:1, nucleotides 54540-54758 of SEQ ID NO:1, and nucleotides 61211-61426 of SEQ ID NO:1. According to another embodiment, the epothilone synthase domain is a dehydratase (DH) domain comprising an amino acid sequence substantially similar to an amino acid se quence selected from the group consisting of: amino acids 869-1037 of SEQ ID NO:4, ami no acids 3886-4048 of SEQ ID NO:5, amino acids 5964-6132 of SEQ ID NO:5, amino acids 2383-2551 of SEQ ID NO:6, and amino acids 887-1051 of SEQ ID NO:7. According to this embodiment, said DH domain preferably comprises an amino acid sequence selected from the group consisting of: amino acids 869-1037 of SEQ ID NO:4, amino acids 3886-4048 of SEQ ID NO:5, amino acids 5964-6132 of SEQ ID NO:5, amino acids 2383-2551 of SEQ ID NO:6, and amino acids 887-1051 of SEQ ID NO:7. Also, according to this embodiment, said nucleotide sequence preferably is substantially similar to a nucleotide sequence selected from the group consisting of: nucleotides 18855-19361 of SEQ ID NO:1, nucleotides 33401 33889 of SEQ ID NO:1, nucleotides 39635-40141 of SEQ ID NO:1, nucleotides 50670 51176 of SEQ ID NO:1, and nucleotides 57593-58087 of SEQ ID NO:1. According to this embodiment, said nucleotide sequence more preferably comprises a consecutive 20, 25, 30, 35, 40, 45, or 50 (preferably 20) base pair nucleotide portion identical in sequence to a respective consecutive 20, 25, 30, 35, 40, 45, or 50 (preferably 20) base pair portion of a nucleotide sequence selected from the group consisting of: nucleotides 18855-19361 of SEQ ID NO:1, nucleotides 33401-33889 of SEQ ID NO:1, nucleotides 39635-40141 of SEQ ID NO:1, nucleotides 50670-51176 of SEQ ID NO:1, and nucleotides 57593-58087 of SEQ ID NO:1. In addition, according to this embodiment, said nucleotide sequence most pre ferably is selected from the group consisting of: nucleotides 18855-19361 of SEQ ID NO:1, WO 99/66028 PCT/EP99/04171 - 15 nucleotides 33401-33889 of SEQ ID NO:1, nucleotides 39635-40141 of SEQ ID NO:1, nucleotides 50670-51176 of SEQ ID NO:1, and nucleotides 57593-58087 of SEQ ID NO:1. According to yet another embodiment, the epothilone synthase domain is a s-keto reductase (KR) domain comprising an amino acid sequence substantially similar to an ami no acid sequence selected from the group consisting of: amino acids 1439-1684 of SEQ ID NO:4, amino acids 1147-1399 of SEQ ID NO:5, amino acids 2645-2895 of SEQ ID NO:5, amino acids 4729-4974 of SEQ ID NO:5, amino acids 6857-7101 of SEQ ID NO:5, amino acids 1143-1393 of SEQ ID NO:6, amino acids 3392-3636 of SEQ ID NO:6, and amino acids 1810-2055 of SEQ ID NO:7. According to this embodiment, said KR domain pre ferably comprises an amino acid sequence selected from the group consisting of: amino acids 1439-1684 of SEQ ID NO:4, amino acids 1147-1399 of SEQ ID NO:5, amino acids 2645-2895 of SEQ ID NO:5, amino acids 4729-4974 of SEQ ID NO:5, amino acids 6857 7101 of SEQ ID NO:5, amino acids 1143-1393 of SEQ ID NO:6, amino acids 3392-3636 of SEQ ID NO:6, and amino acids 1810-2055 of SEQ ID NO:7. Also, according to this embo diment, said nucleotide sequence preferably is substantially similar to a nucleotide sequen ce selected from the group consisting of: nucleotides 20565-21302 of SEQ ID NO:1, nucle otides 25184-25942 of SEQ ID NO:1, nucleotides 29678-30429 of SEQ ID NO:1, nucleo tides 35930-36667 of SEQ ID NO:1, nucleotides 42314-43048 of SEQ ID NO:1, nucleotides 46950-47702 of SEQ ID NO:1, nucleotides 53697-54431 of SEQ ID NO:1, and nucleotides 60362-61099 of SEQ ID NO:1. According to this embodiment, said nucleotide sequence more preferably comprises a consecutive 20, 25, 30, 35, 40, 45, or 50 (preferably 20) base pair nucleotide portion identical in sequence to a respective consecutive 20, 25, 30, 35, 40, 45, or 50 (preferably 20) base pair portion of a nucleotide sequence selected from the group consisting of: nucleotides 20565-21302 of SEQ ID NO:1, nucleotides 25184-25942 of SEQ ID NO:1, nucleotides 29678-30429 of SEQ ID NO:1, nucleotides 35930-36667 of SEQ ID NO:1, nucleotides 42314-43048 of SEQ ID NO:1, nucleotides 46950-47702 of SEQ ID NO:1, nucleotides 53697-54431 of SEQ ID NO:1, and nucleotides 60362-61099 of SEQ ID NO:1. In addition, according to this embodiment, said nucleotide sequence most preferably is selected from the group consisting of: nucleotides 20565-21302 of SEQ ID NO:1, nucle otides 25184-25942 of SEQ ID NO:1, nucleotides 29678-30429 of SEQ ID NO:1, nucleo tides 35930-36667 of SEQ ID NO:1, nucleotides 42314-43048 of SEQ ID NO:1, nucleotides 46950-47702 of SEQ ID NO:1, nucleotides 53697-54431 of SEQ ID NO:1, and nucleotides 60362-61099 of SEQ ID NO:1.

WO 99/66028 PCT/EP99/04171 -16 According to an additional embodiment, the epothilone synthase domain is a methyltransf erase (MT) domain comprising an amino acid sequence substantially similar to amino acids 2671-3045 of SEQ ID NO:6. According to this embodiment, said MT domain preferably comprises amino acids 2671-3045 of SEQ ID NO:6. Also, according to this embodiment, said nucleotide sequence preferably is substantially similar to nucleotides 51534-52657 of SEQ ID NO:1. According to this embodiment, said nucleotide sequence more preferably comprises a consecutive 20, 25, 30, 35, 40, 45, or 50 (preferably 20) base pair nucleotide portion identical in sequence to a respective consecutive 20, 25, 30, 35, 40, 45, or 50 (preferably 20) base pair portion of nucleotides 51534-52657 of SEQ ID NO:1. In addition, according to this embodiment, said nucleotide sequence most preferably is nucleo tides 51534-52657 of SEQ ID NO:1. According to another embodiment, the epothilone synthase domain is a thioesterase (TE) domain comprising an amino acid sequence substantially similar to amino acids 2165 2439 of SEQ ID NO:7. According to this embodiment, said TE domain preferably comprises amino acids 2165-2439 of SEQ ID NO:7. Also, according to this embodiment, said nucleo tide sequence preferably is substantially similar to nucleotides 61427-62254 of SEQ ID NO:1. According to this embodiment, said nucleotide sequence more preferably comprises a consecutive 20, 25, 30, 35, 40, 45, or 50 (preferably 20) base pair nucleotide portion iden tical in sequence to a respective consecutive 20, 25, 30, 35, 40, 45, or 50 (preferably 20) base pair portion of nucleotides 61427-62254 of SEQ ID NO:1. In addition, according to this embodiment, said nucleotide sequence most preferably is nucleotides 61427-62254 of SEQ ID NO:1. In still another aspect, the present invention provides an isolated nucleic acid mole cule comprising a nucleotide sequence that encodes a non-ribosomal peptide synthetase, wherein said non-ribosomal peptide synthetase comprises an amino acid sequence substantially similar to an amino acid sequence selected from the group consisting of: SEQ ID NO:3, amino acids 72-81 of SEQ ID NO:3, amino acids 118-125 of SEQ ID NO:3, amino acids 199-212 of SEQ ID NO:3, amino acids 353-363 of SEQ ID NO:3, amino acids 549 565 of SEQ ID NO:3, amino acids 588-603 of SEQ ID NO:3, amino acids 669-684 of SEQ ID NO:3, amino acids 815-821 of SEQ ID NO:3, amino acids 868-892 of SEQ ID NO:3, amino acids 903-912 of SEQ ID NO:3, amino acids 918-940 of SEQ ID NO:3, amino acids 1268-1274 of SEQ ID NO:3, amino acids 1285-1297 of SEQ ID NO:3, amino acids 973 1256 of SEQ ID NO:3, and amino acids 1344-1351 of SEQ ID NO:3. According to this WO 99/66028 PCT/EP99/04171 -17 embodiment, said non-ribosomal peptide synthetase preferably comprises an amino acid sequence selected from the group consisting of: SEQ ID NO:3, amino acids 72-81 of SEQ ID NO:3, amino acids 118-125 of SEQ ID NO:3, amino acids 199-212 of SEQ ID NO:3, amino acids 353-363 of SEQ ID NO:3, amino acids 549-565 of SEQ ID NO:3, amino acids 588-603 of SEQ ID NO:3, amino acids 669-684 of SEQ ID NO:3, amino acids 815-821 of SEQ ID NO:3, amino acids 868-892 of SEQ ID NO:3, amino acids 903-912 of SEQ ID NO:3, amino acids 918-940 of SEQ ID NO:3, amino acids 1268-1274 of SEQ ID NO:3, amino acids 1285-1297 of SEQ ID NO:3, amino acids 973-1256 of SEQ ID NO:3, and amino acids 1344-1351 of SEQ ID NO:3. Also, according to this embodiment, said nucleotide sequence preferably is substantially similar to a nucleotide sequence selected from the group con sisting of: nucleotides 11872-16104 of SEQ ID NO:1, nucleotides 12085-12114 of SEQ ID NO:1, nucleotides 12223-12246 of SEQ ID NO:1, nucleotides 12466-12507 of SEQ ID NO:1, nucleotides 12928-12960 of SEQ ID NO:1, nucleotides 13516-13566 of SEQ ID NO:1, nucleotides 13633-13680 of SEQ ID NO:1, nucleotides 13876-13923 of SEQ ID NO:1, nucleotides 14313-14334 of SEQ ID NO:1, nucleotides 14473-14547 of SEQ ID NO:1, nucleotides 14578-14607 of SEQ ID NO:1, nucleotides 14623-14692 of SEQ ID NO:1, nucleotides 15673-15693 of SEQ ID NO:1, nucleotides 15724-15762 of SEQ ID NO:1, nucleotides 14788-15639 of SEQ ID NO:1, and nucleotides 15901-15924 of SEQ ID NO:1. According to this embodiment, said nucleotide sequence more preferably comprises a consecutive 20, 25, 30, 35, 40, 45, or 50 (preferably 20) base pair nucleotide portion identical in sequence to a respective consecutive 20, 25, 30, 35, 40, 45, or 50 (preferably 20) base pair portion of a nucleotide sequence selected from the group consisting of: nucle otides 11872-16104 of SEQ ID NO:1, nucleotides 12085-12114 of SEQ ID NO:1, nucleo tides 12223-12246 of SEQ ID NO:1, nucleotides 12466-12507 of SEQ ID NO:1, nucleotides 12928-12960 of SEQ ID NO:1, nucleotides 13516-13566 of SEQ ID NO:1, nucleotides 13633-13680 of SEQ ID NO:1, nucleotides 13876-13923 of SEQ ID NO:1, nucleotides 14313-14334 of SEQ ID NO:1, nucleotides 14473-14547 of SEQ ID NO:1, nucleotides 14578-14607 of SEQ ID NO:1, nucleotides 14623-14692 of SEQ ID NO:1, nucleotides 15673-15693 of SEQ ID NO:1, nucleotides 15724-15762 of SEQ ID NO:1, nucleotides 14788-15639 of SEQ ID NO:1, and nucleotides 15901-15924 of SEQ ID NO:1. In addition, according to this embodiment, said nucleotide sequence most preferably is selected from the group consisting of: nucleotides 11872-16104 of SEQ ID NO:1, nucleotides 12085 12114 of SEQ ID NO:1, nucleotides 12223-12246 of SEQ ID NO:1, nucleotides 12466- WO 99/66028 PCT/EP99/04171 -18 12507 of SEQ ID NO:1, nucleotides 12928-12960 of SEQ ID NO:1, nucleotides 13516 13566 of SEQ ID NO:1, nucleotides 13633-13680 of SEQ ID NO:1, nucleotides 13876 13923 of SEQ ID NO:1, nucleotides 14313-14334 of SEQ ID NO:1, nucleotides 14473 14547 of SEQ ID NO:1, nucleotides 14578-14607 of SEQ ID NO:1, nucleotides 14623 14692 of SEQ ID NO:1, nucleotides 15673-15693 of SEQ ID NO:1, nucleotides 15724 15762 of SEQ ID NO:1, nucleotides 14788-15639 of SEQ ID NO:1, and nucleotides 15901 15924 of SEQ ID NO:1. The present invention further provides an isolated nucleic acid molecule comprising a nucleotide sequence that encodes a polypeptide comprising an amino acid sequence selected from the group consisting of SEQ ID NOs:2-23. In accordance with another aspect, the present invention also provides methods for the recombinant production of polyketides such as epothilones in quantities large enough to enable their purification and use in pharmaceutical formulations such as those for the treat ment of cancer. A specific advantage of these production methods is the chirality of the molecules produced; production in transgenic organisms avoids the generation of popu lations of racemic mixtures, within which some enantiomers may have reduced activity. In particular, the present invention provides a method for heterologous expression of epothi lone in a recombinant host, comprising: (a) introducing into a host a chimeric gene compri sing a heterologous promoter sequence operatively linked to a nucleic acid molecule of the invention that comprises a nucleotide sequence that encodes at least one polypeptide in volved in the biosynthesis of epothilone; and (b) growing the host in conditions that allow biosynthesis of epothilone in the host. The present invention also provides a method for producing epothilone, comprising: (a) expressing epothilone in a recombinant host by the aforementioned method; and (b) extracting epothilone from the recombinant host. According to still another aspect, the present invention provides an isolated polypep tide comprising an amino acid sequence that consists of an epothilone synthase domain. According to one embodiment, the epothilone synthase domain is a s-ketoacyl synthase (KS) domain comprising an amino acid sequence substantially similar to an amino acid sequence selected from the group consisting of: amino acids 11-437 of SEQ ID NO:2, amino acids 7-432 of SEQ ID NO:4, amino acids 39-457 of SEQ ID NO:5, amino acids 1524-1950 of SEQ ID NO:5, amino acids 3024-3449 of SEQ ID NO:5, amino acids 5103 5525 of SEQ ID NO:5, amino acids 35-454 of SEQ ID NO:6, amino acids 1522-1946 of SEQ ID NO: 6, and amino acids 32-450 of SEQ ID NO:7. According to this embodiment, WO 99/66028 PCT/EP99/04171 -19 said KS domain preferably comprises an amino acid sequence selected from the group consisting of: amino acids 11-437 of SEQ ID NO:2, amino acids 7-432 of SEQ ID NO:4, amino acids 39-457 of SEQ ID NO:5, amino acids 1524-1950 of SEQ ID NO:5, amino acids 3024-3449 of SEQ ID NO:5, amino acids 5103-5525 of SEQ ID NO:5, amino acids 35-454 of SEQ ID NO:6, amino acids 1522-1946 of SEQ ID NO: 6, and amino acids 32-450 of SEQ ID NO:7. According to another embodiment, the epothilone synthase domain is an acyltrans ferase (AT) domain comprising an amino acid sequence substantially similar to an amino acid sequence selected from the group consisting of: amino acids 543-864 of SEQ ID NO:2, amino acids 539-859 of SEQ ID NO:4, amino acids 563-884 of SEQ ID NO:5, amino acids 2056-2377 of SEQ ID NO:5, amino acids 3555-3876 of SEQ ID NO:5, amino acids 5631 5951 of SEQ ID NO:5, amino acids 561-881 of SEQ ID NO:6, amino acids 2053-2373 of SEQ ID NO:6, and amino acids 556-877 of SEQ ID NO:7. According to this embodiment, said AT domain preferably comprises an amino acid sequence selected from the group consisting of: amino acids 543-864 of SEQ ID NO:2, amino acids 539-859 of SEQ ID NO:4, amino acids 563-884 of SEQ ID NO:5, amino acids 2056-2377 of SEQ ID NO:5, amino acids 3555-3876 of SEQ ID NO:5, amino acids 5631-5951 of SEQ ID NO:5, amino acids 561-881 of SEQ ID NO:6, amino acids 2053-2373 of SEQ ID NO:6, and amino acids 556 877 of SEQ ID NO:7. According to still another embodiment, the epothilone synthase domain is an enoyl reductase (ER) domain comprising an amino acid sequence substantially similar to an ami no acid sequence selected from the group consisting of: amino acids 974-1273 of SEQ ID NO:2, amino acids 4433-4719 of SEQ ID NO:5, amino acids 6542-6837 of SEQ ID NO:5, and amino acids 1478-1790 of SEQ ID NO:7. According to this embodiment, said ER do main preferably comprises an amino acid sequence selected from the group consisting of: amino acids 974-1273 of SEQ ID NO:2, amino acids 4433-4719 of SEQ ID NO:5, amino acids 6542-6837 of SEQ ID NO:5, and amino acids 1478-1790 of SEQ ID NO:7. According to another embodiment, the epothilone synthase domain is an acyl carrier protein (ACP) domain, wherein said polypeptide comprises an amino acid sequence substantially similar to an amino acid sequence selected from the group consisting of: ami no acids 1314-1385 of SEQ ID NO:2, amino acids 1722-1792 of SEQ ID NO:4, amino acids 1434-1506 of SEQ ID NO:5, amino acids 2932-3005 of SEQ ID NO:5, amino acids 5010 5082 of SEQ ID NO:5, amino acids 7140-7211 of SEQ ID NO:5, amino acids 1430-1503 of WO 99/66028 PCT/EP99/04171 -20 SEQ ID NO:6, amino acids 3673-3745 of SEQ ID NO:6, and amino acids 2093-2164 of SEQ ID NO:7. According to this embodiment, said ACP domain preferably comprises an amino acid sequence selected from the group consisting of: amino acids 1314-1385 of SEQ ID NO:2, amino acids 1722-1792 of SEQ ID NO:4, amino acids 1434-1506 of SEQ ID NO:5, amino acids 2932-3005 of SEQ ID NO:5, amino acids 5010-5082 of SEQ ID NO:5, amino acids 7140-7211 of SEQ ID NO:5, amino acids 1430-1503 of SEQ ID NO:6, amino acids 3673-3745 of SEQ ID NO:6, and amino acids 2093-2164 of SEQ ID NO:7. According to another embodiment, the epothilone synthase domain is a dehydratase (DH) domain comprising an amino acid sequence substantially similar to an amino acid se quence selected from the group consisting of: amino acids 869-1037 of SEQ ID NO:4, ami no acids 3886-4048 of SEQ ID NO:5, amino acids 5964-6132 of SEQ ID NO:5, amino acids 2383-2551 of SEQ ID NO:6, and amino acids 887-1051 of SEQ ID NO:7. According to this embodiment, said DH domain preferably comprises an amino acid sequence selected from the group consisting of: amino acids 869-1037 of SEQ ID NO:4, amino acids 3886-4048 of SEQ ID NO:5, amino acids 5964-6132 of SEQ ID NO:5, amino acids 2383-2551 of SEQ ID NO:6, and amino acids 887-1051 of SEQ ID NO:7. According to yet another embodiment, the epothilone synthase domain is a p-keto reductase (KR) domain comprising an amino acid sequence substantially similar to an ami no acid sequence selected from the group consisting of: amino acids 1439-1684 of SEQ ID NO:4, amino acids 1147-1399 of SEQ ID NO:5, amino acids 2645-2895 of SEQ ID NO:5, amino acids 4729-4974 of SEQ ID NO:5, amino acids 6857-7101 of SEQ ID NO:5, amino acids 1143-1393 of SEQ ID NO:6, amino acids 3392-3636 of SEQ ID NO:6, and amino acids 1810-2055 of SEQ ID NO:7. According to this embodiment, said KR domain prefer ably comprises an amino acid sequence selected from the group consisting of: amino acids 1439-1684 of SEQ ID NO:4, amino acids 1147-1399 of SEQ ID NO:5, amino acids 2645 2895 of SEQ ID NO:5, amino acids 4729-4974 of SEQ ID NO:5, amino acids 6857-7101 of SEQ ID NO:5, amino acids 1143-1393 of SEQ ID NO:6, amino acids 3392-3636 of SEQ ID NO:6, and amino acids 1810-2055 of SEQ ID NO:7. According to an additional embodiment, the epothilone synthase domain is a methyl transferase (MT) domain comprising an amino acid sequence substantially similar to amino acids 2671-3045 of SEQ ID NO:6. According to this embodiment, said MT domain preferab ly comprises amino acids 2671-3045 of SEQ ID NO:6.

WO 99/66028 PCT/EP99/04171 - 21 According to another embodiment, the epothilone synthase domain is a thioesterase (TE) domain comprising an amino acid sequence substantially similar to amino acids 2165 2439 of SEQ ID NO:7. According to this embodiment, said TE domain preferably comprises amino acids 2165-2439 of SEQ ID NO:7. Other aspects and advantages of the present invention will become apparent to those skilled in the art from a study of the following description of the invention and non-limiting examples. DEFINITIONS In describing the present invention, the following terms will be employed, and are intended to be defined as indicated below. Associated With / Operatively Linked: Refers to two DNA sequences that are related physically or functionally. For example, a promoter or regulatory DNA sequence is said to be "associated with" a DNA sequence that codes for an RNA or a protein if the two sequen ces are operatively linked, or situated such that the regulator DNA sequence will affect the expression level of the coding or structural DNA sequence. Chimeric Gene: A recombinant DNA sequence in which a promoter or regulatory DNA sequence is operatively linked to, or associated with, a DNA sequence that codes for an mRNA or which is expressed as a protein, such that the regulator DNA sequence is able to regulate transcription or expression of the associated DNA sequence. The regulator DNA sequence of the chimeric gene is not normally operatively linked to the associated DNA sequence as found in nature. Coding DNA Sequence: A DNA sequence that is translated in an organism to pro duce a protein. Domain: That part of a polyketide synthase necessary for a given distinct activity. Examples include acyl carrier protein (ACP), P-ketosynthase (KS), acyltransferase (AT), ketoreductase (KR), dehydratase (DH), enoylreductase (ER), and thioesterase (TE) domains. Epothilones: 16-membered macrocyclic polyketides naturally produced by the bacte rium Sorangium cellulosum strain So ce9O, which mimic the biological effects of taxol. In this application, "epothilone" refers to the class of polyketides that includes epothilone A and epothilone B, as well as analogs thereof such as those described in WO 98/25929.

WO 99/66028 PCTIEP99/04171 - 22 Epothilone Synthase: A polyketide synthase responsible for the biosynthesis of epo thilone. Gene: A defined region that is located within a genome and that, besides the afore mentioned coding DNA sequence, comprises other, primarily regulatory, DNA sequences responsible for the control of the expression, that is to say the transcription and translation, of the coding portion. Heterologous DNA Sequence: A DNA sequence not naturally associated with a host cell into which it is introduced, including non-naturally occurring multiple copies of a natu rally occurring DNA sequence. Homologous DNA Sequence: A DNA sequence naturally associated with a host cell into which it is introduced. Homologous Recombination: Reciprocal exchange of DNA fragments between homologous DNA molecules. Isolated: In the context of the present invention, an isolated nucleic acid molecule or an isolated enzyme is a nucleic acid molecule or enzyme that, by the hand of man, exists apart from its native environment and is therefore not a product of nature. An isolated nucleic acid molecule or enzyme may exist in a purified form or may exist in a non-native environment such as, for example, a recombinant host cell. Module: A genetic element encoding all of the distinct activities required in a single round of polyketide biosynthesis, i.e., one condensation step and all the P-carbonyl pro cessing steps associated therewith. Each module encodes an ACP, a KS, and an AT activity to accomplish the condensation portion of the biosynthesis, and selected post condensation activities to effect the p-carbonyl processing. NRPS: A non-ribosomal polypeptide synthetase, which is a complex of enzymatic activities responsible for the incorporation of amino acids into secondary metabolites in cluding, for example, amino acid adenylation, epimerization, N-methylation, cyclization, peptidyl carrier protein, and condensation domains. A functional NRPS is one that cata lyzes the incorporation of an amino acid into a secondary metabolite. NRPS gene: One or more genes encoding NRPSs for producing functional secon dary metabolites, e.g., epothilones A and B, when under the direction of one or more com patible control elements.

WO 99/66028 PCT/EP99/04171 -23 Nucleic Acid Molecule: A linear segment of single- or double-stranded DNA or RNA that can be isolated from any source. In the context of the present invention, the nucleic acid molecule is preferably a segment of DNA. ORF: Open Reading Frame. PKS: A polyketide synthase, which is a complex of enzymatic activities (domains) responsible for the biosynthesis of polyketides including, for example, ketoreductase, dehy dratase, acyl carrier protein, enoylreductase, ketoacyl ACP synthase, and acyltransf erase. A functional PKS is one that catalyzes the synthesis of a polyketide. PKS Genes: One or more genes encoding various polypeptides required for produ cing functional polyketides, e.g., epothilones A and B, when under the direction of one or more compatible control elements. Substantially Similar: With respect to nucleic acids, a nucleic acid molecule that has at least 60 percent sequence identity with a reference nucleic acid molecule. In a preferred embodiment, a substantially similar DNA sequence is at least 80% identical to a reference DNA sequence; in a more preferred embodiment, a substantially similar DNA sequence is at least 90% identical to a reference DNA sequence; and in a most preferred embodiment, a substantially similar DNA sequence is at least 95% identical to a reference DNA sequence. A substantially similar DNA sequence preferably encodes a protein or peptide having substantially the same activity as the protein or peptide encoded by the reference DNA sequence. A substantially similar nucleotide sequence typically hybridizes to a reference nucleic acid molecule, or fragments thereof, under the following conditions: hybridization at 7% sodium dodecyl sulfate (SDS), 0.5 M NaPO 4 pH 7.0, 1 mM EDTA at 500C; wash with 2X SSC, 1% SDS, at 500C. With respect to proteins or peptides, a substantially similar amino acid sequence is an amino acid sequence that is at least 90% identical to the amino acid sequence of a reference protein or peptide and has substantially the same activity as the reference protein or peptide. Transformation: A process for introducing heterologous nucleic acid into a host cell or organism. Transformed / Transgenic / Recombinant: Refers to a host organism such as a bac terium into which a heterologous nucleic acid molecule has been introduced. The nucleic acid molecule can be stably integrated into the genome of the host or the nucleic acid mo lecule can also be present as an extrachromosomal molecule. Such an extrachromosomal molecule can be auto-replicating. Transformed cells, tissues, or plants are understood to WO 99/66028 PCT/EP99/04171 - 24 encompass not only the end product of a transformation process, but also transgenic pro geny thereof. A "non-transformed", "non-transgenic", or "non-recombinant" host refers to a wild-type organism, i.e., a bacterium, which does not contain the heterologous nucleic acid molecule. Nucleotides are indicated by their bases by the following standard abbreviations: adenine (A), cytosine (C), thymine (T), and guanine (G). Amino acids are likewise indicated by the following standard abbreviations: alanine (ala; A), arginine (Arg; R), asparagine (Asn; N), aspartic acid (Asp; D), cysteine (Cys; C), glutamine (Gln; Q), glutamic acid (Glu; E), glycine (Gly; G), histidine (His; H), isoleucine (lie; I), leucine (Leu; L), lysine (lys; K), methionine (Met; M), phenylalanine (Phe; F), proline (Pro; P), serine (Ser; S), threonine (Thr; T), tryptophan (Trp; W), tyrosine (Tyr; Y), and valine (Val; V). Furthermore, (Xaa; X) represents any amino acid. DESCRIPTION OF THE SEQUENCES IN THE SEQUENCE LISTING SEQ ID NO:1 is the nucleotide sequence of a 68750 bp contig containing 22 open reading frames (ORFs), which comprises the epothilone biosynthesis genes. SEQ ID NO:2 is the protein sequence of a type I polyketide synthase (EPOS A) encoded by epoA (nucleotides 7610-11875 of SEQ ID NO:1). SEQ ID NO:3 is the protein sequence of a non-ribosomal peptide synthetase (EPOS P) encoded by epoP (nucleotides 11872-16104 of SEQ ID NO:1). SEQ ID NO:4 is the protein sequence of a type I polyketide synthase (EPOS B) encoded by epoB (nucleotides 16251-21749 of SEQ ID NO:1). SEQ ID NO:5 is the protein sequence of a type I polyketide synthase (EPOS C) encoded by epoC (nucleotides 21746-43519 of SEQ ID NO:1). SEQ ID NO:6 is the protein sequence of a type I polyketide synthase (EPOS D) encoded by epoD (nucleotides 43524-54920 of SEQ ID NO:1). SEQ ID NO:7 is the protein sequence of a type I polyketide synthase (EPOS E) encoded by epoE (nucleotides 54935-62254 of SEQ ID NO:1). SEQ ID NO:8 is the protein sequence of a cytochrome P450 oxygenase homologue (EPOS F) encoded by epoF (nucleotides 62369-63628 of SEQ ID NO:1). SEQ ID NO:9 is a partial protein sequence (partial Orf 1) encoded by orfl (nucleotides 1-1826 of SEQ ID NO:1).

WO 99/66028 PCT/EP99/04171 -25 SEQ ID NO:10 is a protein sequence (Orf 2) encoded by orf2 (nucleotides 3171-1900 on the reverse complement strand of SEQ ID NO:1). SEQ ID NO:11 is a protein sequence (Orf 3) encoded by orf3 (nucleotides 3415-5556 of SEQ ID NO:1). SEQ ID NO:12 is a protein sequence (Orf 4) encoded by orf4 (nucleotides 5992-5612 on the reverse complement strand of SEQ ID NO:1). SEQ ID NO:13 is a protein sequence (Orf 5) encoded by or/5 (nucleotides 6226-6675 of SEQ ID NO:1). SEQ ID NO:14 is a protein sequence (Orf 6) encoded by orf6 (nucleotides 63779 64333 of SEQ ID NO:1). SEQ ID NO:15 is a protein sequence (Orf 7) encoded by orf7 (nucleotides 64290 63853 on the reverse complement strand of SEQ ID NO:1). SEQ ID NO:16 is a protein sequence (Orf 8) encoded by or/B (nucleotides 64363 64920 of SEQ ID NO:1). SEQ ID NO:17 is a protein sequence (Orf 9) encoded by or/9 (nucleotides 64727 64287 on the reverse complement strand of SEQ ID NO:1). SEQ ID NO:18 is a protein sequence (Ori 10) encoded by orl0 (nucleotides 65063 65767 of SEQ ID NO:1). SEQ ID NO:19 is a protein sequence (Orf 11) encoded by orfl1 (nucleotides 65874 65008 on the reverse complement strand of SEQ ID NO:1). SEQ ID NO:20 is a protein sequence (Orf 12) encoded by orf12 (nucleotides 66338 65871 on the reverse complement strand of SEQ ID NO:1). SEQ ID NO:21 is a protein sequence (Orf 13) encoded by orf13 (nucleotides 66667 67137 of SEQ ID NO:1). SEQ ID NO:22 is a protein sequence (Orf 14) encoded by orf14 (nucleotides 67334 68251 of SEQ ID NO:1). SEQ ID NO:23 is a partial protein sequence (partial Orf 15) encoded by orf15 (nucleotides 68346-68750 of SEQ ID NO:1). SEQ ID NO:24 is the universal reverse PCR primer sequence. SEQ ID NO:25 is the universal forward PCR primer sequence. SEQ ID NO:26 is the NH24 end "B" PCR primer sequence. SEQ ID NO:27 is the NH2 end "A" PCR primer sequence. SEQ ID NO:28 is the NH2 end "B" PCR primer sequence.

WO 99/66028 PCT/EP99/04171 -26 SEQ ID NO:29 is the pEPO15-NH6 end "B" PCR primer sequence. SEQ ID NO:30 is the pEPO15-H2.7 end "A" PCR primer sequence. DEPOSIT INFORMATION The following material has been deposited with the Agricultural Research Service, Patent Culture Collection (NRRL), 1815 North University Street, Peoria, Illinois 61604, under the Budapest Treaty on the International Recognition of the Deposit of Microorganisms for the Purposes of Patent Procedure. All restrictions on the availability of the deposited material will be irrevocably removed upon the granting of a patent. Deposited Material Accession Number Deposit Date pEPO15 NRRL B-30033 June 11, 1998 pEPO32 NRRL B-30119 April 16, 1999 DETAILED DESCRIPTION OF THE INVENTION The genes involved in the biosynthesis of epothilones can be isolated using the techniques according to the present invention. The preferable procedure for the isolation of epothilone biosynthesis genes requires the isolation of genomic DNA from an organism identified as producing epothilones A and B, and the transfer of the isolated DNA on a suitable plasmid or vector to a host organism that does not normally produce the polyketide, followed by the identification of transformed host colonies to which the epothilone-producing ability has been conferred. Using a technique such as X::Tn5 transposon mutagenesis (de Bruijn & Lupski, Gene 27: 131-149 (1984)), the exact region of the transforming epothilone conferring DNA can be more precisely defined. Alternatively or additionally, the transfor ming epothilone-conferring DNA can be cleaved into smaller fragments and the smallest that maintains the epothilone-conf erring ability further characterized. Whereas the host organism lacking the ability to produce epothilone may be a different species from the orga nism from which the polyketide derives, a variation of this technique involves the transfor mation of host DNA into the same host that has had its epothilone-producing ability disrup ted by mutagenesis. In this method, an epothilone-producing organism is mutated and non epothilone-producing mutants are isolated. These are then complemented by genomic DNA isolated from the epothilone-producing parent strain.

WO 99/66028 PCT/EP99/04171 -27 A further example of a technique that can be used to isolate genes required for epo thilone biosynthesis is the use of transposon mutagenesis to generate mutants of an epothi lone-producing organism that, after mutagenesis, fails to produce the polyketide. Thus, the region of the host genome responsible for epothilone production is tagged by the transpo son and can be recovered and used as a probe to isolate the native genes from the parent strain. PKS genes that are required for the synthesis of polyketides and that are similar to known PKS genes may be isolated by virtue of their sequence homology to the biosynthetic genes for which the sequence is known, such as those for the biosynthesis of rifamycin or soraphen. Techniques suitable for isolation by homology include standard library screening by DNA hybridization. Preferred for use as a probe molecule is a DNA fragment that is obtainable from a gene or another DNA sequence that plays a part in the synthesis of a known polyketide. A preferred probe molecule comprises a 1.2 kb Smal DNA fragment encoding the ketosyntha se domain of the fourth module of the soraphen PKS (U.S. Patent No. 5,716,849), and a more preferred probe molecule comprises the p-ketoacyl synthase domains from the first and second modules of the rifamycin PKS (Schupp et al., FEMS Microbiology Letters 159: 201-207 (1998)). These can be used to probe a gene library of an epothilone-producing microorganism to isolate the PKS genes responsible for epothilone biosynthesis. Despite the well-known difficulties with PKS gene isolation in general and despite the difficulties expected to be encountered with the isolation of epothilone biosynthesis genes in particular, by using the methods described in the instant specification, biosynthetic genes for epothilones A and B can surprisingly be cloned from a microorganism that produ ces that polyketide. Using the methods of gene manipulation and recombinant production described in this specification, the cloned PKS genes can be modified and expressed in transgenic host organisms. The isolated epothilone biosynthetic genes can be expressed in heterologous hosts to enable the production of the polyketide with greater efficiency than might be possible from native hosts. Techniques for these genetic manipulations are specific for the different available hosts and are known in the art. For example, heterologous genes can be expres sed in Streptomyces and other actinomycetes using techniques such as those described in McDaniel et al., Science 262: 1546-1550 (1993) and Kao et a., Science 265: 509-512 (1994), both of which are incorporated herein by reference. See also, Rowe et aL., Gene WO 99/66028 PCT/EP99/04171 -28 216: 215-223 (1998); Holmes et al., EMBO Journal12(8): 3183-3191 (1993) and Bibb et al., Gene 38: 215-226 (1985), all of which are incorporated herein by reference. Alternately, genes responsible for polyketide biosynthesis, i.e., epothilone biosynthe tic genes, can also be expressed in other host organisms such as pseudomonads and E. coli. Techniques for these genetic manipulations are specific for the different available hosts and are known in the art. For example, PKS genes have been sucessfully expressed in E. coli using the pT7-7 vector, which uses the T7 promoter. See, Tabor et al., Proc. Nat. Acad. Sci. USA 82: 1074-1078 (1985), incorporated herein by reference. In addition, the expression vectors pKK223-3 and pKK223-2 can be used to express heterologous genes in E. coli, either in transcriptional or translational fusion, behind the tac or trc promoter. For the expression of operons encoding multiple ORFs, the simplest procedure is to insert the operon into a vector such as pKK223-3 in transcriptional fusion, allowing the cognate ribo some binding site of the heterologous genes to be used. Techniques for overexpression in gram-positive species such as Bacillus are also known in the art and can be used in the context of this invention (Quax et al., in: Industrial Microorganisms: Basic and Applied Mo lecular Genetics, Eds. Baltz et al., American Society for Microbiology, Washington (1993)). Other expression systems that may be used with the epothilone biosynthetic genes of the invention include yeast and baculovirus expression systems. See, for example, "The Expression of Recombinant Proteins in Yeasts," Sudbery, P. E., Curr. Opin. Biotechnol. 7(5): 517-524 (1996); "Methods for Expressing Recombinant Proteins in Yeast," Mackay, et al., Editor(s): Carey, Paul R., Protein Eng. Des. 105-153, Publisher: Academic, San Diego, Calif (1996); "Expression of heterologous gene products in yeast," Pichuantes, et al., Editor(s): Cleland, J. L., Craik, C. S., Protein Eng. 129-161, Publisher: Wiley-Liss, New York, N. Y (1996); WO 98/27203; Kealey et al., Proc. Nat/. Acad. Sci. USA 95: 505-509 (1998); "Insect Cell Culture: Recent Advances, Bioengineering Challenges And Implications In Protein Production," Palomares, et al., Editor(s): Galindo, Enrique; Ramirez, Octavio T., Adv. Bioprocess Eng. Vol. II, Invited Pap. Int. Symp., 2nd (1998) 25-52, Publisher: Kluwer, Dordrecht, Neth; "Baculovirus Expression Vectors," Jarvis, Donald L., Editor(s): Miller, Lois K., Baculoviruses 389-431, Publisher: Plenum, New York, N. Y. (1997); "Production Of He terologous Proteins Using The Baculovirus/Insect Expression System," Grittiths, et al., Me thods Mol. Biol. (Totowa, N. J.) 75 (Basic Cell Culture Protocols (2nd Edition)) 427-440 (1997); and "Insect Cell Expression Technology," Luckow, Verne A., Protein Eng. 183-218, WO 99/66028 PCT/EP99/04171 - 29 Publisher: Wiley-Liss, New York, N. Y. (1996); all of which are incorporated herein by refe rence. Another consideration for expression of PKS genes in heterologous hosts is the re quirement of enzymes for posttranslational modification of PKS enzymes by phosphopante theinylation before they can synthesize polyketides. However, the enzymes responsible for this modification of type I PKS enzymes, phosphopantetheinyl (P-pant) transferases are not normally present in many hosts such as E. coli. This problem can be solved by coexpres sion of a P-pant transferase with the PKS genes in the heterologous host, as described by Kealey et al., Proc. Natl. Acad. Sci. USA 95: 505-509 (1998), incorporated herein by re ference. Therefore, for the purposes of polyketide production, the significant criteria in the choice of host organism are its ease of manipulation, rapidity of growth (i.e. fermentation), possession or the proper molecular machinery for processes such as posttranslational modification, and its lack of susceptibility to the polyketide being overproduced. Most preferred host organisms are actinomycetes such as strains of Streptomyces. Other pre ferred host organisms are pseudomonads and E. coli. The above-described methods of polyketide production have significant advantages over the technology currently used in the preparation of the compounds. These advantages include the cheaper cost of production, the ability to produce greater quantities of the compounds, and the ability to produce com pounds of a preferred biological enantiomer, as opposed to racemic mixtures inevitably ge nerated by organic synthesis. Compounds produced by heterologous hosts can be used in medical (e.g. cancer treatment in the case of epothilones) as well as agricultural applica tions.

WO 99/66028 PCT/EP99/04171 - 30 EXPERIMENTAL The invention will be further described by reference to the following detailed examples. These examples are provided for purposes of illustration only, and are not inten ded to be limiting unless otherwise specified. Standard recombinant DNA and molecular cloning techniques used here are well known in the art and are described by Ausubel (ed.), Current Protocols in Molecular Biology, John Wiley and Sons, Inc. (1994); T. Maniatis, E. F. Fritsch and J. Sambrook, Molecular Cloning: A Laboratory Manual, Cold Spring Harbor laboratory, Cold Spring Harbor, NY (1989); and by T.J. Silhavy, M.L. Berman, and L.W. Enquist, Experiments with Gene Fusions, Cold Spring Harbor Laboratory, Cold Spring Harbor, NY (1984). Example 1: Cultivation of an Epothilone-Producing Strain of Sorangium cellulosum Sorangium cellulosum strain 90 (DSM 6773, Deutsche Sammlung von Mikroorganis men und Zellkulturen, Braunschweig) is streaked out and grown (30"C) on an agar plate of SolE medium (0.35% glucose, 0.05% tryptone, 0.15% MgSO 4 x 7H 2 0, 0.05% ammonium sulfate, 0.1% CaC 2 , 0.006% K 2

HPO

4 , 0.01% sodium dithionite, 0.0008% Fe-EDTA, 1.2% HEPES, 3.5% [vol/vol] supernatant of sterilized stationary S. cellulosum culture) pH ad. 7.4. Cells from about 1 square cm are picked and inoculated into 5 mis of G51t liquid medium (0.2% glucose, 0.5% starch, 0.2% tryptone, 0.1% probion S, 0.05% CaCl 2 x2H 2 0, 0.05% MgSO 4 x7H 2 0, 1.2% HEPES, pH ad. 7.4) and incubated at 30'C with shaking at 225 rpm. After 4 days, the culture is transferred into 50 mis of G51t and incubated as above for 5 days. This culture is used to inoculate 500 mis of G51t and incubated as above for 6 days. The culture is centrifuged for 10 minutes at 4000 rpm and the cell pellet is resuspended in 50 mis of G51t. Example 2: Generation of a Bacterial Artificial Chromosome (Bac) Library To generate a Bac library, S. cellulosum cells cultivated as described in Example 1 above are embedded into agarose blocks, lysed, and the liberated genomic DNA is partially digested by the restriction enzyme HindIll. The digested DNA is separated on an agarose gel by pulsed-field electrophoresis. Large (approximately 90-150 kb) DNA fragments are WO 99/66028 PCT/EP99/04171 - 31 isolated from the agarose gel and ligated into the vector pBelobacll. pBelobacIl contains a gene encoding chloramphenicol resistance, a multiple cloning site in the lacZ gene provi ding for blue/white selection on appropriate medium, as well as the genes required for the replication and maintenance of the plasmid at one or two copies per cell. The ligation mix ture is used to transform Escherichia coli DH1 OB electrocompetent cells using standard electroporation techniques. Chloramphenicol-resistant recombinant (white, lacZ mutant) colonies are transferred to a positively charged nylon membrane filter in 384 3X3 grid for mat. The clones are lysed and the DNA is cross-linked to the filters. The same clones are also preserved as liquid cultures at -80'C. Example 3: Screening the Bac Library of Sorangium cellulosum 90 for the Presence of Type I Polyketide Synthase-Related Sequences The Bac library filters are probed by standard Southern hybridization procedures. The DNA probes used encode B-ketoacyl synthase domains from the first and second modules of the rifamycin polyketide synthase (Schupp et al., FEMS Microbiology Letters 159: 201-207 (1998)). The probe DNAs are generated by PCR with primers flanking each ketosynthase domain using the plasmid pNE95 as the template (pNE95 equals cosmid 2 described in Schupp et al. (1998)). 25 ng of PCR-amplified DNA is isolated from a 0.5% agarose gel and labeled with 32 P-dCTP using a random primer labeling kit (Gibco-BRL, Bethesda MD, USA) according to the manufacturer's instructions. Hybridization is at 650C for 36 hours and membranes are washed at high stringency (3 times with 0.1x SSC and 0.5% SDS for 20 min at 650C). The labeled blot is exposed on a phosphorescent screen and the signals are detected on a Phospholmager 445SI (screen and 445SI from Molecular Dynamics). This results in strong hybridization of certain Bac clones to the probes. These clones are selected and cultured overnight in 5 mIs of Luria broth (LB) at 370C. Bac DNA from the Bac clones of interest is isolated by a typical miniprep procedure. The cells are resuspended in 200 pl lysozyme solution (50mM glucose, 10 mM EDTA, 25 mM Tris-HCI, 5mg/ml lysozyme), lysed in 400 p.1 lysis solution (0.2 N NaOH and 2% SDS), the proteins are precipitated (3.0 M potassium acetate, adjusted to pH5.2 with acetic acid), and the Bac DNA is precipitated with isopropanol. The DNA is resuspended in 20pI of nuclease-free distilled water, restricted with BamHI (New England Biolabs, Inc.) and separated on a 0.7% agarose gel. The gel is blotted by Southern hybridization as described above and probed WO 99/66028 PCT/EP99/04171 - 32 under conditions described above, with a 1.2 kb Smal DNA fragment encoding the ketosyn thase domain of the fourth module of the soraphen polyketide synthase as the probe (see, U.S. Patent No. 5,716,849). Five different hybridization patterns are observed. One clone representing each of the five patterns is selected and named pEPO15, pEPO20, pEPO30, pEPO31, and pEPO33, respectively. Example 4: Subcloning of BamHI Fragments from pEPO15, pEPO20, pEPO30, pEPO31, and pEPO33 The DNA of the five selected Bac clones is digested with BamHl and random frag ments are subcloned into pBluescript Il SK+ (Stratagene) at the BamHl site. Subclones car rying inserts between 2 and 10 kb in size are selected for sequencing of the flanking ends of the inserts and also probed with the 1.2 Smal probe as described above. Subclones that show a high degree of sequence homology to known polyketide synthases and/or strong hybridization to the soraphen ketosynthase domain are used for gene disruption experi ments. Example 5: Preparation of Streptomycin-Resistant Spontaneous Mutants of Sorangium cellulosum strain So ce90 0.1 ml of a three day old culture of Sorangium cellulosum strain So ce90, which is raised in liquid medium G52-H (0.2% yeast extract, 0.2% soyameal defatted, 0.8% potato starch, 0.2% glucose, 0.1% MgSO4 x7H20, 0.1% CaCl2 x2H20, 0.008% Fe-EDTA, pH ad 7.4 with KOH), is plated out on agar plates with SolE medium supplemented with 100 jig/ml streptomycin. The plates are incubated at 30 0 C for 2 weeks. The colonies growing on this medium are streptomycin-resistant mutants, which are streaked out and cultivated once more on the same agar medium with streptomycin for purification. One of these strepto mycin-resistant mutants is selected and is called BCE28/2.

WO 99/66028 PCT/EP99/04171 - 33 Example 6: Gene Disruptions in Sorangium cellulosum BCE28/2 Using the Subcloned BamHl Fragments The BamHI inserts of the subclones generated from the five selected Bac clones as described above are isolated and ligated into the unique BamHI site of plasmid pCIB132 (see, U.S. Patent No. 5,716,849). The pCIB132 derivatives carrying the inserts are trans formed into Escherichia coli ED8767 containing the helper plasmid pUZ8 (Hedges and Matthew, Plasmid 2: 269-278 (1979). The transformants are used as donors in conjugation experiments with Sorangium cellulosum BCE28/2 as recipient. For the conjugation, 5-10 x 10 cells of Sorangium cellulosum BCE28/2 from an early stationary phase culture (reaching about 5 x 108 cells/ml) grown at 300C in liquid medium G51b (G51b equals medium G51t with tryptone replaced by peptone) are mixed in a 1:1 cellular ratio with a late-log phase culture (in LB liquid medium) of E. coli ED8767 containing pCIB1 32 derivatives carrying the subcloned BamHI fragments and the helper plasmid pUZ8. The mixed cells are then centri fuged at 4000 rpm for 10 minutes and resuspended in 0.5 ml G51b medium. This cell sus pension is then plated as a drop in the center of a plate with Sol E agar containg 50 mg/l kanamycin. The cells obtained after incubation for 24 hours at 300C are harvested and res uspended in 0.8 ml of G51b medium, and 0.1 to 0.3 ml of this suspension is plated out on a selective Sol E solid medium containing phleomycin (30 mg/), streptomycin (300 mg/I), and kanamycin (50 mg/I). The counterselection of the donor Escherichia coli strain takes place with the aid of streptomycin. The colonies that grow on this selective medium after an in cubation time of 8-12 days at a temperature of 300C are isolated with a plastic loop and streaked out and cultivated on the same agar medium for a second round of selection and purification. The colony-derived cultures that grow on this selective agar medium after 7 days at a temperature of 300C are transconjugants of Sorangium cellulosum BCE28/2 that have acquired phleomycin resistance by conjugative transfer of the pCIB132 derivatives carrying the subcloned BamHI fragments. Integration of the pClB132-derived plasmids into the chromosome of Sorangium cellulosum BCE28/2 by homologous recombination is verified by Southern hybridization. For this experiment, complete DNA from 5-10 tranconjugants per transferred BamHI frag ment is isolated (from 10 ml cultures grown in medium G52-H for three days) applying the method described by Pospiech and Neumann, Trends Genet. 11: 217 (1995). For the Southern blot, the DNA isolated as described above is cleaved either with the restriction WO 99/66028 PCT/EP99/04171 - 34 enzymes BgAI, Clal, or Nol, and the respective BamHl inserts or pC!B132 are used as 32P labelled probes. Example 7: Analysis of the Effect of the Integrated BamHI Fragments on Epothilone Production by Sorangium cellulosum After Gene Disruption Transconjugant cells grown on about 1 square cm surface of the selective So1 E plates of the second round of selection (see Example 6) are transferred by a sterile plastic loop into 10 ml of medium G52-H in an 50 ml Erlenmeyer flask. After incubation at 300C and 180 rpm for 3 days, the culture is transfered into 50 ml of medium G52-H in an 200 ml Erlenmeyer flask. After incubation at 30*C and 180 rpm for 4-5 days, 10 ml of this culture is transfered into 50 ml of medium 23B3 (0.2 % glucose, 2 % potato starch, 1.6 % soya meal defatted, 0.0008 % Fe-EDTA Sodium salt, 0.5 % HEPES (4-(2-hydroxyethyl)-piperazine-1 ethane-sulfonic-acid), 2 % vol/vol polysterole resin XAD16 (Rohm & Haas), pH adjusted to 7.8 with NaOH) in an 200 ml Erlenmeyer flask. Quantitative determination of the epothilone produced takes place after incubation of the cultures at 300C and 180 rpm for 7 days. The complete culture broth is filtered by suction through a 150 p.m nylon filter. The resin remaining on the filter is then resuspended in 10 ml isopropanol and extracted by shaking the suspension at 180 rpm for 1 hour. 1 ml is removed from this suspension and centrifuged at 12,000 rpm in an Eppendorff Microfuge. The amount of epothilones A and B therein is determined by means of an HPLC and detection at 250 nm with a UVDAD detector (HPLC with Waters -Symetry C18 column and a gradient of 0.02 % phosphoric acid 60%-0% and acetonitril 40%-i 00%). Transconjugants with three different integrated BamHl fragments subcloned from pEPO15, namely transconjugants with the BamHl fragment of plasmid pEPO15-21, trans conjugants with the BamHI fragment of plasmid pEPO15-4-5, and transconjugants with the BamHl fragment of plasmid pEPO1 5-4-1, are tested in the manner described above. HPLC analysis reveals that all transconjugants no longer produce epothilone A or B. By contrast, epothilone A and B are detectable in a concentration of 2-4 mg/I in transconjugants with BamHI fragments integrated that are derived from pEPO20, pEPO30, pEPO31, pEPO33, and in the parental strain BCE28/2.

WO 99/66028 PCT/EP99/04171 - 35 Example 8: Nucleotide Sequence Determination of the Cloned Fragments and Construction of Contigs A. BamHI Insert of Plasmid pEPO15-21 Plasmid DNA is isolated from the strain Escherichia coli DH1OB [pEPO15-21], and the nucleotide sequence of the 2.3-kb BamHI insert in pEPO15-21 is determined. Automa ted DNA sequencing is done on the double-stranded DNA template by the dideoxynucleo tide chain termination method, using Applied Biosystems model 377 sequencers. The pri mers used are the universal reverse primer (5' GGA AAC AGC TAT GAC CAT G 3' (SEQ ID NO:24)) and the universal forward primer (5' GTA AAA CGA CGG CCA GT 3' (SEQ ID NO:25)). In subsequent rounds of sequencing reactions, custom-synthesized oligonucle otides, designed for the 3' ends of the previously determined sequences, are used to extend and join contigs. Both strands are entirely sequenced, and every nucleotide is se quenced at least two times. The nucleotide sequence is compiled using the program Sequencher vers. 3.0 (Gene Codes Corporation), and analyzed using the University of Wisconsin Genetics Computer Group programs. The nucleotide sequence of the 2213-bp insert corresponds to nucleotides 20779-22991 of SEQ ID NO:1. B. BamHI Insert of Plasmid pEPO1 5-4-1 Plasmid DNA is isolated from the strain Escherichia coli DH10B [pEPO15-4-1], and the nucleotide sequence of the 3.9-kb BamHl insert in pEPO15-4-1 is determined as descri bed in (A) above. The nucleotide sequence of the 3909-bp insert corresponds to nucleo tides 16876-20784 of SEQ ID NO:1. C. BamHI Insert of Plasmid pEPO15-4-5 Plasmid DNA is isolated from the strain Escherichia coli DH10B [pEPO15-4-5], and the nucleotide sequence of the 2.3-kb BamHI insert in pEPO15-4-5 is determined as described in (A) above. The nucleotide sequence of the 2233-bp insert corresponds to nucleotides 42528-44760 of SEQ ID NO:1.

WO 99/66028 PCT/EP99/04171 - 36 Example 9: Subcloning and Ordering of DNA Fragments from pEPO15 Containing Epothilone Biosynthesis Genes pEPO15 is digested to completion with the restriction enzyme Hindill and the resul ting fragments are subcloned into pBluescript I SK- or pNEB193 (New England Biolabs) that has been cut with Hindlll and dephosphorylated with calf intestinal alkaline phospha tase. Six different clones are generated and named pEPO15-NH1, pEPO15-NH2, pEPO15-NH6, pEPO15-NH24 (all based on pNEB193), and pEPO15-H2.7 and pEPO15 H3.0 (both based on pBluescript II SK-). The BamHI insert of pEPO15-21 is isolated and DIG-labeled (Non-radioactive DNA labeling and detection system, Boehringer Mannheim), and used as a probe in DNA hybri dization experiments at high stringency against pEPO15-NH1, pEPO15-NH2, pEPO15 NH6, pEPO15-NH24, pEPO15-H2.7 and pEPO15-H3.0. Strong hybridization signal is de tected for pEPO15-NH24, indicating that pEPO15-21 is contained within pEPO15-NH24. The BamHl insert of pEPO15-4-1 is isolated and DIG-labeled as above, and used as a probe in DNA hybridization experiments at high stringency against pEPO15-NH1, pEPO15-NH2, pEPO15-NH6, pEPO15-NH24, pEPO15-H2.7 and pEPO15-H3.0. Strong hybridization signals are detected for pEPO1 5-NH24 and pEPO1 5-H2.7. Nucleotide se quence data generated from one end each of pEPO15-NH24 and pEPO15-H2.7 are also in complete agreement with the previously determined sequence of the BamHI insert of pEPO1 5-4-1. These experiments demonstrate that pEPO1 5-4-1 (which contains one inter nal Hindill site) overlaps pEPO15-H2.7 and pEPO15-NH24, and that pEPO15-H2.7 and pEPO15-NH24, in this order, are contiguous. The BamHI insert of pEPO15-4-5 is isolated and DIG-labeled as above, and used as a probe in DNA hybridization experiments at high stringency against pEPO1 5-NH1, pEPO15-NH2, pEPO15-NH6, pEPO15-NH24, pEPO15-H2.7 and pEPO15-H3.0. Strong hybridization signal is detected for pEPO15-NH2, indicating that pEPO15-21 is contained within pEPO15-NH2. Nucleotide sequence data is generated from both ends of pEPO15-NH2 and from the end of pEPO1 5-NH24 that does not overlap with pEPO1 5-4-1. PCR primers NH24 end "B": GTGACTGGCGCCTGGAATCTGCATGAGC (SEQ ID NO:26), NH2 end "A": AGCGGGAGCTTGCTAGACATTCTGTTTC (SEQ ID NO:27), and NH2 end "B": GACGCGCCTCGGGCAGCGCCCCAA (SEQ ID NO:28), pointing towards the HindIll sites, WO 99/66028 PCT/EP99/04171 - 37 are designed based on these sequences and used in amplification reactions with pEPO15 and, in separate experiments, with Sorangium cellulosum So ce90 genomic DNA as the templates. Specific amplification is found with primer pair NH24 end "B" and NH2 end "A" with both templates. The amplimers are cloned into pBluescript II SK- and completely se quenced. The sequences of the amplimers are identical, and also agree completely with the end sequences of pEPO15-NH24 and pEPO15-NH2, fused at the Hindill site, estab lishing that the HindIll fragments of pEPO15-NH2 and pEPO15-NH24 are, in this order, contiguous. The HindIllI insert of pEPO15-H2.7 is isolated and DIG-labeled as above, and used as a probe in a DNA hybridization experiment at high stringency against pEPO15 digested by Notl. A Noti fragment of about 9 kb in size shows a strong a hybridization, and is further subcloned into pBluescript Il SK- that has been digested with Not and dephosphorylated with calf intestinal alkaline phosphatase, to yield pEPO15-N9-16. The Notd insert of pEPO15-N9-16 is isolated and DIG-labeled as above, and used as a probe in DNA hybri dization experiments at high stringency against pEPO15-NH1, pEPO15-NH2, pEPO15 NH6, pEPO15-NH24, pEPO15-H2.7 and pEPO15-H3.0. Strong hybridization signals are detected for pEPO1 5-NH6, and also for the expected clones pEPO1 5-H2.7 and pEPO1 5 NH24. Nucleotide sequence data is generated from both ends of pEPO15-NH6 and from the end of pEPO15-H2.7 that does not overlap with pEPO15-4-1. PCR primers are de signed pointing towards the HindIll sites and used in amplification reactions with pEPO15 and, in separate experiments, with Sorangium cellulosum So ce90 genomic DNA as the templates. Specific amplification is found with primer pair pEPO15-NH6 end "B": CACCGAAGCGTCGATCTGGTCCATC (SEQ ID NO:29) and pEPO15-H2.7 end "A": CGGTCAGATCGACGACGGGCTTTCC (SEQ ID NO:30) with both templates. The ampli mers are cloned into pBluescript 11 SK- and completely sequenced. The sequences of the amplimers are identical, and also agree completely with the end sequences of pEPO15 NH6 and pEPO15-H2.7, fused at the HindillI site, establishing that the Hindill fragments of pEPO15-NH6 and pEPO15-H2.7 are, in this order, contiguous. All of these experiments, taken together, establish a contig of HindIll fragments covering a region of about 55 kb and consisting of the Hindlil inserts of pEPO15-NH6, pEPO15-H2.7, pEPO15-NH24, and pEPO15-NH2, in this order. The inserts of the re maining two Hindill subclones, namely pEPO15-NH1 and pEPO15-H3.0, are not found to be parts of this contig.

WO 99/66028 PCT/EP99/04171 - 38 Example 10: Further Extension of the Subclone Contig Covering the Epothilone Biosynthesis Genes An approximately 2.2 kb BamHl - HindIll fragment derived from the downstream end of the insert of pEPO1 5-NH2 and thus representing the downstream end of the subclone contig described in Example 9 is isolated, DIG-labeled, and used in Southern hybridization experiments against pEPO15 and pEPO15-NH2 DNAs digested with several enzymes. The strongly hybridizing bands are always found to be the same in size between the two target DNAs indicating that the Sorangium cellulosum So ce90 genomic DNA fragment cloned into pEPO15 ends with the HindllI site at the downstream end of pEPO15-NH2. A cosmid DNA library of Sorangium cellulosum So ce90 is generated, using establi shed procedures, in pScosTriplex-ll (Ji, et al., Genomics 31: 185-192 (1996)). Briefly, high molecular weight genomic DNA of Sorangium cellulosum So ce90 is partially digested with the restriction enzyme Sau3Al to provide fragments with average sizes of about 40 kb, and ligated to BamHl and Xbal digested pScosTriplex-II. The ligation mix is packaged with Gigapack IlIl XL (Stratagene) and used to transfect E. coli XL1 Blue MR cells. The cosmid library is screened with the approximately 2.2 kb BamHI - Hindlll frag ment, derived from the downstream end of the insert of pEPO15-NH2, used as a probe in colony hybridization. A strongly hybridizing clone, named pEPO4E7 is selected. pEPO4E7 DNA is isolated, digested with several restriction endonucleases, and probed in Southern hybridization experiments with the 2.2 kb BamHI - HindIllI fragment. A strongly hybridizing Not fragment of approximately 9 kb in size is selected and subcloned into pBluescript Il SK- to yield pEPO4E7-N9-8. Further Southern hybridization experiments reveal that the approximately 9 kb Nofl insert of pEPO4E7-N9-8 overlaps pEPO15-NH2 over 6 kb in a Noti - Hindlll fragment, while the remaining approximately 3 kb Hindill - Nol fragment would extend the subclone contig described in Example 9. End sequencing re veals, however, that the downstream end of the insert of pEPO4E7-N9-8 contains the BamHI - Noti polylinker of pScosTriplex-Il, thereby indicating that the genomic DNA insert of pEPO4E7 ends at a Sau3AI site within the extending Hindill - Notn fragment and that the Noti site is derived from pScosTriplex-l. An approximately 1.6 kb Psti - Sal fragment derived from the approximately 3 kb extending Hindill - Notl subfragment of pEPO4E7-N9-8, containing only Sorangium WO 99/66028 PCTIEP99/04171 - 39 cellulosum So ce90-derived sequences free of vector, is used as a probe against the bacterial artificial chromosome library described in Example 2. Besides the previously isolated EPO15, a Bac clone, named EP032, is found to strongly hybridize to the probe. pEPO32 is isolated, digested with several restriction endonucleases, and hybridized with the approximately 1.6 kb Psti - SaA probe. A Hindlll - EcoRV fragment of about 13 kb in size is found to strongly hybridize to the probe, and is subcloned into pBluescript 11 SK digested with Hindill and Hincil to yield pEPO32-HEV15. Oligonucleotide primers are designed based on the downstream end sequence of pEPO15-NH2 and on the upstream (Hindill) end sequence derived from pEPO32-HEV15, and used in sequencing reactions with pEPO4E7-N9-8 as the template. The sequences reveal the existence of a small Hindill fragment (EPO4E7-HO.02) of 24 bp, undetectable in standard restriction analysis, separating the Hindill site at the downstream end of pEPO15 NH2 from the Hindlll site at the upstream end of pEPO32-HEV15. Thus, the subclone contig described in Example 9 is extended to include the Hindill fragment EPO4E7-HO.02 and the insert of pEPO32-HEV15, and constitutes the inserts of: pEPO15-NH6, pEPO15-H2.7, pEPO15-NH24, pEPO15-NH2, EPO4E7-HO.02 and pEPO32 HEV15, in this order. Example 11: Nucleotide Sequence Determination of the Subclone Contig Covering the Epothilone Biosynthesis Genes The nucleotide sequence of the subclone contig described in Example 10 is determined as follows. pEPO15-H2.7. Plasmid DNA is isolated from the strain Escherichia co/iDH10B [pEPO15-H2.71, and the nucleotide sequence of the 2.7-kb BamHl insert in pEPO15-H2.7 is determined. Automated DNA sequencing is done on the double-stranded DNA template by the dideoxynucleotide chain termination method, using Applied Biosystems model 377 sequencers. The primers used are the universal reverse primer (5' GGA AAC AGC TAT GAC CAT G 3' (SEQ ID NO:24)) and the universal forward primer (5' GTA AAA CGA CGG CCA GT 3' (SEQ ID NO:25)). In subsequent rounds of sequencing reactions, custom synthesized oligonucleotides, designed for the 3' ends of the previously determined sequences, are used to extend and join contigs.

WO 99/66028 PCT/EP99/04171 - 40 pEPO15-NH6, pEPO15-NH24 and pEPO15-NH2. The Hindlli inserts of these plas mids are isolated, and subjected to random fragmentation using a Hydroshear apparatus (Genomic Instrumentation Services, Inc.) to yield an average fragment size of 1-2 kb. The fragments are end-repaired using T4 DNA Polymerase and Klenow DNA Polymerase en zymes in the presence of desoxynucleotide triphosphates, and phosphorylated with T4 DNA Kinase in the presence of ribo-ATP. Fragments in the size range of 1.5-2.2 kb are isolated from agarose gels, and ligated into pBluescript It SK- that has been cut with EcoRV and de phosphorylated. Random subclones are sequenced using the universal reverse and the universal forward primers. pEPO32-HEV15. pEPO32-HEV15 is digested with Hindill and Sspl, the approxima tely 13.3 kb fragment containing the -13 kb HindIll - EcoRV insert from So. cellulosum So ce90 and a 0.3 kb Hincl - Sspl fragment from pBluescript Il SK- is isolated, and partially digested with Haelll to yield fragments with an average size of 1-2 kb. Fragments in the size range of 1.5-2.2 kb are isolated from agarose gels, and ligated into pBluescript |1 SK- that has been cut with EcoRV and dephosphorylated. Random subclones are sequenced using the universal reverse and the universal forward primers. The chromatograms are analyzed and assembled into contigs with the Phred, Phrap and Consed programs (Ewing, et al., Genome Res. 8(3): 175-185 (1998); Ewing, et al., Genome Res. 8(3): 186-194 (1998); Gordon, et al., Genome Res. 8(3): 195-202 (1998)). Contig gaps are filled, sequence discrepancies are resolved, and low-quality regions are resequenced using custom-designed oligonucleotide primers for sequencing on either the original subclones or selected clones from the random subclone libraries. Both strands are completely sequenced, and every basepair is covered with at least a minimum aggregated Phred score of 40 (confidence level of 99.99%). The nucleotide sequence of the 68750 bp contig is shown as SEQ ID NO:1.

WO 99/66028 PCT/EP99/04171 -41 Example 12: Nucleotide Sequence Analysis of the Epothilone Biosynthesis Genes SEQ ID NO:1 is found to contain 22 ORFs as detailed below in Table 1: Table 1 ORF Start codon Stop codon Homology of deduced protein Proposed function of deduced protein orfl outside of 1826 sequenced range orf2 * 3171 1900 Hypothetical protein SP: Q1 1037; DD-peptidase SP:P15555 o3 3415 5556 Na/H antiporter PID: DI017724 Transport orf4* 5992 5612 or50 6226 6675 epoA 7610 11875 Type I polyketide synthase Epothilone synthase: Thiazole ring formation epoP 11872 16104 Non-ribosomal peptide synthetase Epothilone synthase: Thiazole ring formation epoB 16251 21749 Type I polyketide synthase Epothilone synthase: Polyketide backbone formation epoC 21746 43519 Type I polyketide synthase Epothilone synthase: Polyketide backbone formation epoD 43524 54920 Type I polyketide synthase Epothilone synthase: Polyketide I_ backbone formation epoE 54935 62254 Type I polyketide synthase Epothilone synthase: Polyketide backbone formation epoF 62369 63628 Cytochrome P450 Epothilone macrolactone oxidase orf6 63779 64333 or/7 * 64290 63853 orf8 64363 64920 orI9 * 64727 64287 orfl0 65063 65767 orfll * 65874 65008 orfl2* 66338 65871 orfl3 66667 67137 orfl4 67334 68251 Hypothetical protein GI:3293544; Transport Cation efflux system protein GI:2623026 orfl5 68346 outside of sequenced range * On the reverse complementer strand. Numbering according to SEQ ID NO:1. epoA (nucleotides 7610-11875 of SEQ ID NO:1) codes for EPOS A (SEQ ID NO:2), a type I polyketide synthase consisting of a single module, and harboring the following do mains: p-ketoacyl-synthase (KS) (nucleotides 7643-8920 of SEQ ID NO:1, amino acids 11- WO 99/66028 PCT/EP99/04171 - 42 437 of SEQ ID NO:2); acyltransferase (AT) (nucleotides 9236-10201 of SEQ ID NO:1, amino acids 543-864 of SEQ ID NO:2); enoyl reductase (ER) (nucleotides 10529-11428 of SEQ ID NO:1, amino acids 974-1273 of SEQ ID NO:2); and acyl carrier protein homologous domain (ACP) (nucleotides 11549-11764 of SEQ ID NO:1, amino acids 1314-1385 of SEQ ID NO:2). Sequence comparisons and motif analysis (Haydock, et al. FEBS Lett. 374: 246 248 (1995); Tang, et al., Gene 216: 255-265 (1998)) reveal that the AT encoded by EPOS A is specific for malonyl-CoA. EPOS A should be involved in the initiation of epothilone bio synthesis by loading the acetate unit to the multienzyme complex that will eventually form part of the 2-methylthiazole ring (C26 and C20). epoP (nucleotides 11872-16104 of SEQ ID NO:1) codes for EPOS P (SEQ ID NO:3), a non-ribosomal peptide synthetase containing one module. EPOS P harbors the following domains: * peptide bond formation domain, as delineated by motif K (amino acids 72-81 [FPLTDIQESY] of SEQ ID NO:3, corresponding to nucleotide positions 12085-12114 of SEQ ID NO:1); motif L (amino acids 118-125 [VVARHDML] of SEQ ID NO:3, correspon ding to nucleotide positions 12223-12246 of SEQ ID NO:1); motif M (amino acids 199 212 [SIDLINVDLGSLSI] of SEQ ID NO:3, corresponding to nucleotide positions 12466 12507 of SEQ ID NO:1); and motif 0 (amino acids 353-363 [GDFTSMVLLDI] of SEQ ID NO:3, corresponding to nucleotide positions 12928-12960 of SEQ ID NO:1); * aminoacyl adenylate formation domain, as delineated by motif A (amino acids 549 565 [LTYEELSRRSRRLGARL] of SEQ ID NO:3, corresponding to nucleotide positions 13516-13566 of SEQ ID NO:1); motif B (amino acids 588-603 [VAVLAVLESGAAYVPI] of SEQ ID NO:3, corresponding to nucleotide positions 13633-13680 of SEQ ID NO:1); motif C (amino acids 669-684 [AYVIYTSGSTGLPKGV] of SEQ ID NO:3, corresponding to nucleotide positions 13876-13923 of SEQ ID NO:1); motif D (amino acids 815-821 [SLGGATE] of SEQ ID NO:3, corresponding to nucleotide positions 14313-14334 of SEQ ID NO:1); motif E (amino acids 868-892 [GQLYIGGVGLALGYWRDEEKTRKSF) of SEQ ID NO:3, corresponding to nucleotide positions 14473-14547 of SEQ ID NO:1); motif F (amino acids 903-912 [YKTGDLGRYL] of SEQ ID NO:3, corresponding to nucle otide positions 14578-14607 of SEQ ID NO:1); motif G (amino acids 918-940 [EFMGREDNQIKLRGYRVELGEIE] of SEQ ID NO:3, corresponding to nucleotide positions 14623-14692 of SEQ ID NO:1); motif H (amino acids 1268-1274 [LPEYMVP] of SEQ ID NO:3, corresponding to nucleotide positions 15673-15693 of SEQ ID NO:1); and WO 99/66028 PCT/EP99/04171 - 43 motif I (amino acids 1285-1297 [LTSNGKVDRKALR] of SEQ ID NO:3, corresponding to nucleotide positions 15724-15762 of SEQ ID NO:1); * an unknown domain, inserted between motifs G and H of the aminoacyl adenylate formation domain (amino acids 973-1256 of SEQ ID NO:3, corresponding to nucleotide positions 14788-15639 of SEQ ID NO:1); and * a peptidyl carrier protein homologous domain (PCP), delineated by motif J (amino acids 1344-1351 [GATSIHIV] of SEQ ID NO:3, corresponding to nucleotide positions 15901-15924 of SEQ ID NO:1). It is proposed that EPOS P is involved in the activation of a cysteine by adenylation, binding the activated cysteine as an aminoacyl-S-PCP, forming a peptide bond between the en zyme-bound cysteine and the acetyl-S-ACP supplied by EPOS A, and the formation of the initial thiazoline ring by intramolecular heterocyclization. The unknown domain of EPOS P displays very weak homologies to NAD(P)H oxidases and reductases from Bacillus species. Thus, this unknown domain and/or the ER domain of EPOS A may be involved in the oxida tion of the initial 2-methylthiazoline ring to a 2-methylthiazole. epoB (nucleotides 16251-21749 of SEQ ID NO:1) codes for EPOS B (SEQ ID NO:4), a type I polyketide synthase consisting of a single module, and harboring the following do mains: KS (nucleotides 16269-17546 of SEQ ID NO:1, amino acids 7-432 of SEQ ID NO:4); AT (nucleotides 17865-18827 of SEQ ID NO:1, amino acids 539-859 of SEQ ID NO:4); dehydratase (DH) (nucleotides 18855-19361 of SEQ ID NO:1, amino acids 869-1037 of SEQ ID NO:4); P-ketoreductase (KR) (nucleotides 20565-21302 of SEQ ID NO:1, amino acids 1439-1684 of SEQ ID NO:4); and ACP (nucleotides 21414-21626 of SEQ ID NO:1, amino acids 1722-1792 of SEQ ID NO:4). Sequence comparisons and motif analysis reveal that the AT encoded by EPOS B is specific for methylmalonyl-CoA. EPOS A should be in volved in the first polyketide chain extension by catalysing the Claisen-like condensation of the 2-methyl-4-thiazolecarboxyl-S-PCP starter group with the methylmalonyl-S-ACP, and the concomitant reduction of the b-keto group of C17 to an enoyl. epoC (nucleotides 21746-43519 of SEQ ID NO:1) codes for EPOS C (SEQ ID NO:5), a type I polyketide synthase consisting of 4 modules. The first module harbors a KS (nucle otides 21860-23116 of SEQ ID NO:1, amino acids 39-457 of SEQ ID NO:5); a malonyl CoA specific AT (nucleotides 23431-24397 of SEQ ID NO:1, amino acids 563-884 of SEQ ID NO:5); a KR (nucleotides 25184-25942 of SEQ ID NO:1, amino acids 1147-1399 of SEQ ID NO:5); and an ACP (nucleotides 26045-26263 of SEQ ID NO:1, amino acids 1434-1506 of WO 99/66028 PCT/EP99/04171 - 44 SEQ ID NO:5). This module incorporates an acetate extender unit (Cl 4-C13) and reduces the r-keto group at C15 to the hydroxyl group that takes part in the final lactonization of the epothilone macrolactone ring. The second module of EPOS C harbors a KS (nucleotides 26318-27595 of SEQ ID NO:1, amino acids 1524-1950 of SEQ ID NO:5); a malonyl CoA specific AT (nucleotides 27911-28876 of SEQ ID NO:1, amino acids 2056-2377 of SEQ ID NO:5); a KR (nucleotides 29678-30429 of SEQ ID NO:1, amino acids 2645-2895 of SEQ ID NO:5); and an ACP (nucleotides 30539-30759 of SEQ ID NO:1, amino acids 2932-3005 of SEQ ID NO:5). This module incorporates an acetate extender unit (C12-C11) and reduces the P-keto group at C13 to a hydroxyl group. Thus, the nascent polyketide chain of epothi lone corresponds to epothilone A, and the incorporation of the methyl side chain at C12 in epothilone B would require a post-PKS C-methyltransf erase activity. The formation of the epoxi ring at C13-C12 would also require a post-PKS oxidation step. The third module of EPOS C harbors a KS (nucleotides 30815-32092 of SEQ ID NO:1, amino acids 3024-3449 of SEQ ID NO:5); a malonyl CoA-specific AT (nucleotides 32408-33373 of SEQ ID NO:1, amino acids 3555-3876 of SEQ ID NO:5); a DH (nucleotides 33401-33889 of SEQ ID NO:1, amino acids 3886-4048 of SEQ ID NO:5); an ER (nucleotides 35042-35902 of SEQ ID NO:1, amino acids 4433-4719 of SEQ ID NO:5); a KR (nucleotides 35930-36667 of SEQ ID NO:1, amino acids 4729-4974 of SEQ ID NO:5); and an ACP (nucleotides 36773-36991 of SEQ ID NO:1, amino acids 5010-5082 of SEQ ID NO:5). This module incorporates an ace tate extender unit (C10-C9) and fully reduces the p-keto group at C11. The fourth module of EPOS C harbors a KS (nucleotides 37052-38320 of SEQ ID NO:1, amino acids 5103 5525 of SEQ ID NO:5); a methylmalonyl CoA-specific AT (nucleotides 38636-39598 of SEQ ID NO:1, amino acids 5631-5951 of SEQ ID NO:5); a DH (nucleotides 39635-40141 of SEQ ID NO:1, amino acids 5964-6132 of SEQ ID NO:5); an ER (nucleotides 41369-42256 of SEQ ID NO:1, amino acids 6542-6837 of SEQ ID NO:5); a KR (nucleotides 42314-43048 of SEQ ID NO:1, amino acids 6857-7101 of SEQ ID NO:5); and an ACP (nucleotides 43163 43378 of SEQ ID NO:1, amino acids 7140-7211 of SEQ ID NO:5). This module incorporates a propionate extender unit (C24 and C8-C7) and fully reduces the p-keto group at C9. epoD (nucleotides 43524-54920 of SEQ ID NO:1) codes for EPOS D (SEQ ID NO:6), a type I polyketide synthase consisting of 2 modules. The first module harbors a KS (nucleotides 43626-44885 of SEQ ID NO:1, amino acids 35-454 of SEQ ID NO:6); a methylmalonyl CoA-specific AT (nucleotides 45204-46166 of SEQ ID NO:1, amino acids 561-881 of SEQ ID NO:6); a KR (nucleotides 46950-47702 of SEQ ID NO:1, amino acids WO 99/66028 PCT/EP99/04171 - 45 1143-1393 of SEQ ID NO:6); and an ACP (nucleotides 47811-48032 of SEQ ID NO:1, ami no acids 1430-1503 of SEQ ID NO:6). This module incorporates a propionate extender unit (C23 and C6-C5) and reduces the p-keto group at C7 to a hydoxyl group. The second mo dule harbors a KS (nucleotides 48087-49361 of SEQ ID NO:1, amino acids 1522-1946 of SEQ ID NO: 6); a methylmalonyl CoA-specific AT (nucleotides 49680-50642 of SEQ ID NO:1, amino acids 2053-2373 of SEQ ID NO:6); a DH (nucleotides 50670-51176 of SEQ ID NO:1, amino acids 2383-2551 of SEQ ID NO:6); a methyltransferase (MT, nucleotides 51534-52657 of SEQ ID NO:1, amino acids 2671-3045 of SEQ ID NO:6); a KR (nucleotides 53697-54431 of SEQ ID NO:1, amino acids 3392-3636 of SEQ ID NO:6); and an ACP (nucleotides 54540-54758 of SEQ ID NO:1, amino acids 3673-3745 of SEQ ID NO:6). This module incorporates a propionate extender unit (C21 or C22 and C4-C3) and reduces the P-keto group at C5 to a hydoxyl group. This reduction is somewhat unexpected, since epo thilones contain a keto group at C5. Discrepancies of this kind between the deduced reduc tive capabilities of PKS modules and the redox state of the corresponding positions in the final polyketide products have been, however, reported in the literature (see, for example, Schwecke, et al., Proc. Nat/. Acad. Sci. USA 92: 7839-7843 (1995) and Schupp, et al., FEMS Microbiology Letters 159: 201-207 (1998)). An important feature of epothilones is the presence of gem-methyl side groups at C4 (C21 and C22). The second module of EPOS D is predicted to incorporate a propionate unit into the growing polyketide chain, providing one methyl side chain at C4. This module also contains a methyltransferase do main integrated into the PKS between the DH and the KR domains, in an arrangement simi lar to the one seen in the HMWP1 yersiniabactin synthase (Gehring, A.M., DeMoll, E., Fetherston, J.D., Mori, I., Mayhew, G.F., Blattner, F.R., Walsh, C.T., and Perry, R.D.: Iron acquisition in plague: modular logic in enzymatic biogenesis of yersiniabactin by Yersinia pestis. Chem. Biol. 5, 573-586, 1998). This MT domain in EPOS D is proposed to be responsible for the incorporation of the second methyl side group (C21 or C22) at C4. epoE (nucleotides 54935-62254 of SEQ ID NO:1) codes for EPOS E (SEQ ID NO:7), a type I polyketide synthase consisting of one module, harboring a KS (nucleotides 55028 56284 of SEQ ID NO:1, amino acids 32-450 of SEQ ID NO:7); a malonyl CoA-specific AT (nucleotides 56600-57565 of SEQ ID NO:1, amino acids 556-877 of SEQ ID NO:7); a DH (nucleotides 57593-58087 of SEQ ID NO:1, amino acids 887-1051 of SEQ ID NO:7); a pro bably nonfunctional ER (nucleotides 59366-60304 of SEQ ID NO:1, amino acids 1478-1790 of SEQ ID NO:7); a KR (nucleotides 60362-61099 of SEQ ID NO:1, amino acids 1810-2055 WO 99/66028 PCT/EP99/04171 -46 of SEQ ID NO:7); an ACP (nucleotides 61211-61426 of SEQ ID NO:1, amino acids 2093 2164 of SEQ ID NO:7); and a thioesterase (TE) (nucleotides 61427-62254 of SEQ ID NO:1, amino acids 2165-2439 of SEQ ID NO:7). The ER domain in this module harbors an active site motif with some highly unusual amino acid substitutions that probably render this do main inactive. The module incorporates an acetate extender unit (C2-Cl), and reduces the p-keto at C3 to an enoyl group. Epothilones contain a hydroxyl group at C3, so this reduc tion also appears to be excessive as discussed for the second module of EPOS D. The TE domain of EPOS E takes part in the release and cyclization of the grown polyketide chain via lactonization between the carboxyl group of C1 and the hydroxyl group of C15. Five ORFs are detected upstream of epoA in the sequenced region. The partially se quenced orfl has no homologues in the sequence databanks. The deduced protein pro duct (Orf 2, SEQ ID NO:10) of or/2 (nucleotides 3171-1900 on the reverse complement strand of SEQ ID NO:1) shows strong similarities to hypothetical ORFs from Mycobacterium and Streptomyces coelicolor, and more distant similarities to carboxypeptidases and DD peptidases of different bacteria. The deduced protein product of orf3 (nucleotides 3415 5556 of SEQ ID NO:1), Orf 3 (SEQ ID NO:1 1), shows homologies to Na/H antiporters of different bacteria. Orf 3 might take part in the export of epothilones from the producer strain. orf4 and or/5 have no homologues in the sequence databanks. Eleven ORFs are found downstream of epoE in the sequenced region. epoF (nucle otides 62369-63628 of SEQ ID NO:1) codes for EPOS F (SEQ ID NO:8), a deduced protein with strong sequence similarities to cytochrome P450 oxygenases. EPOS F may take part in the adjustment of the redox state of the carbons C12, C5, and/or C3. The deduced pro tein product of orf14 (nucleotides 67334-68251 of SEQ ID NO:1), Orf 14 (SEQ ID NO:22) shows strong similarities to GI:3293544, a hypothetic protein with no proposed function from Streptomyces coelicolor, and also to GI:2654559, the human embrionic lung protein. It is also more distantly related to cation efflux system proteins like GI:2623026 from Methano bacterium thermoautotrophicum, so it might also take part in the export of epothilones from the producing cells. The remaining ORFs (or6-orfl3 and orfl5) show no homologies to entries in the sequence databanks. Example 13: Recombinant Expression of Epothilone Biosynthesis Genes WO 99/66028 PCT/EP99/04171 -47 Epothilone synthase genes according to the present invention are expressed in hete rologous organisms for the purposes of epothilone production at greater quantities than can be accomplished by fermentation of Sorangium cellulosum. A preferable host for hetero logous expression is Streptomyces, e.g. Streptomyces coelicolor, which natively produces the polyketide actinorhodin. Techniques for recombinant PKS gene expression in this host are described in McDaniel et al., Science 262: 1546-1550 (1993) and Kao et al., Science 265: 509-512 (1994). See also, Holmes et al., EMBO Journal12(8): 3183-3191 (1993) and Bibb et al., Gene 38: 215-226 (1985), as well as U.S. Patent Nos. 5,521,077, 5,672,491, and 5,712,146, which are incorporated herein by reference. According to one method, the heterologous host strain is engineered to contain a chromosomal deletion of the actinorhodin (act) gene cluster. Expression plasmids contai ning the epothilone synthase genes of the invention are constructed by transferring DNA from a temperature-sensitive donor plasmid to a recipient shuttle vector in E. coli (McDaniel et al. (1993) and Kao et al. (1994)), such that the synthase genes are built-up by homolo gous recombination within the vector. Alternatively, the epothilone synthase gene cluster is introduced into the vector by restriction fragment ligation. Following selection, e.g. as described in Kao et al. (1994), DNA from the vector is introduced into the act-minus Streptomyces coelicolor strain according to protocols set forth in Hopwood et al., Genetic Manipulation of Streptomyces. A Laboratory Manual (John Innes Foundation, Norwich, United Kingdom, 1985), incorporated herein by reference. The recombinant Streptomyces strain is grown on R2YE medium (Hopwood et al. (1985)) and produces epothilones. Alternatively, the epothilone synthase genes according to the present invention are ex pressed in other host organisms such as pseudomonads, Bacillus, yeast, insect cells and/or E. coli. PKS and NRPS genes are preferably expressed in E. coli using the pT7-7 vector, which uses the T7 promoter. See, Tabor et al., Proc. Natl. Acad. Sci. USA 82: 1074-1078 (1985). In another embodiment, the expression vectors pKK223-3 and pKK223-2 are used to express PKS and NRPS genes in E. coli, either in transcriptional or translational fusion, behind the tac or trc promoter. Expression of PKS and NRPS genes in heterologous hosts, which do not naturally have the phosphopantetheinyl (P-pant) transferases needed for post translational modification of PKS enzymes, requires the coexpression in the host of a P pant transferase, as described by Kealey et al., Proc. Natl. Acad. Sci. USA 95: 505-509 (1998).

WO 99/66028 PCT/EP99/04171 - 48 Example 14: Isolation of Epothilones from Producing Strains Examples of cultivation, fermentation, and extraction procedures for polyketide isola tion, which are useful for extracting epothilones from both native and recombinant hosts ac cording to the present invention, are given in WO 93/10121, incorporated herein by referen ce, in Example 57 of U.S. Patent No. 5,639,949, in Gerth et al., J. Antibiotics 49: 560-563 (1996), and in Swiss patent application no. 396/98, filed February 19, 1998, and U.S. patent application no. 09/248,910 (that discloses also preferred mutant strains of Sorangium cellulosum), both of which are incorporated herein by reference. The following are pro cedures that are useful for isolating epothilones from cultured Sorangium cellulosum strains such as So ce90, and may also be used for the isolation of epothilone from recombinant hosts. A: Cultivation of epothilone-producing strains: Strain: Sorangium cellulosum Soce-90 or a recombinant host strain according to the present invention. Preservation of the strain: In liquid N 2 . Media: Precultures and intermediate cultures: G52 Main culture: 1B12 G52 Medium: yeast extract, low in salt (BioSpringer, Maison Alfort, France) 2 g/l MgSO 4 (7 H 2 0) 1 g/l CaCl 2 (2 H 2 0) 1 g/l soya meal defatted Soyamine 50T (Lucas Meyer, Hamburg, Germany) 2 g/I potato starch Noredux A-150 (Blattmann, Waedenswil, Switzerland) 8 g/l glucose anhydrous 2 g/I EDTA-Fe(Ill)-Na salt (8 g/l) 1 mI/I WO 99/66028 PCTIEP99/04171 - 49 pH 7.4, corrected with KOH Sterilisation: 20 mins. 120 *C 1B12 Medium: potato starch Noredux A-150 (Blattmann, Waedenswil, Switzerland) 20 g/l soya meal defatted Soyamine 50T (Lucas Meyer, Hamburg, Germany) 11 g/l EDTA-Fe(lll)-Na salt 8 mg/I pH 7.8, corrected with KOH Sterilisation: 20 mins. 120 0 C Addition of cyclodextrins and cyclodextrin derivatives: Cyclodextrins (Fluka, Buchs, Switzerland, or Wacker Chemie, Munich, Germany) in different concentrations are sterilised separately and added to the 1B12 medium prior to seeding. Cultivation: 1 ml of the suspension of Sorangium cellulosum Soce-90 from a liquid N 2 am poule is transferred to 10 ml of G52 medium (in a 50 ml Erlenmeyer flask) and incubated for 3 days at 180 rpm in an agitator at 300C, 25 mm displacement. 5 ml of this culture is added to 45 ml of G52 medium (in a 200 ml Erlenmeyer flask) and incubated for 3 days at 180 rpm in an agitator at 300C, 25 mm displacement. 50 ml of this culture is then added to 450 ml of G52 medium (in a 2 litre Erlenmeyer flask) and incubated for 3 days at 180 rpm in an agi tator at 300C, 50 mm displacement. Maintenance culture: The culture is overseeded every 3-4 days, by adding 50 ml of culture to 450 ml of G52 medium (in a 2 litre Erlenmeyer flask). All experiments and fermentations are carried out by starting with this maintenance culture. Tests in a flask: (1) Preculture in an agitatinq flask: WO 99/66028 PCT/EP99/04171 - 50 Starting with the 500 ml of maintenance culture, 1 x 450 ml of G52 medium are seeded with 50 ml of the maintenance culture and incubated for 4 days at 180 rpm in an agitator at 300C, 50 mm displacement. (ii) Main culture in the agitating flask: 40 ml of 1B12 medium plus 5 g/I 4-morpholine-propane-sulfonic acid (= MOPS) powder (in a 200 ml Erlenmeyer flask) are mixed with 5 ml of a 1Ox concentrated cyclodextrin solution, seeded with 10 ml of preculture and incubated for 5 days at 180 rpm in an agitator at 300C, 50 mm displacement. Fermentation: Fermentations are carried out on a scale of 10 litres, 100 litres and 500 litres. 20 litre and 100 litre fermentations serve as an intermediate culture step. Whereas the pre cultures and intermediate cultures are seeded as the maintenance culture 10% (v/v), the main cultures are seeded with 20% (v/v) of the intermediate culture. Important: In contrast to the agitating cultures, the ingredients of the media for the fermentation are calculated on the final culture volume including the inoculum. If, for example, 18 litres of medium + 2 litres of inoculum are combined, then substances for 20 litres are weighed in, but are only mixed with 18 litres. Preculture in an agitating flask: Starting with the 500 ml maintenance culture, 4 x 450 ml of G52 medium (in a 2 litre Erlen meyer flask) are each seeded with 50 ml thereof, and incubated for 4 days at 180 rpm in an agitator at 300C, 50 mm displacement. Intermediate culture, 20 litres or 100 litres: 20 litres: 18 litres of G52 medium in a fermenter having a total volume of 30 litres are seeded with 2 litres of the preculture. Cultivation lasts for 3-4 days, and the conditions are: 300C, 250 rpm, 0.5 litres of air per litre liquid per min, 0.5 bars excess pressure, no pH control. 100 litres: 90 litres of G52 medium in a fermenter having a total volume of 150 litres are seeded with 10 litres of the 20 litre intermediate culture. Cultivation lasts for 3-4 days, and the conditions are: 300C, 150 rpm, 0.5 litres of air per litre liquid per min, 0.5 bars excess pressure, no pH control.

WO 99/66028 PCT/EP99/04171 -51 Main culture, 10 litres, 100 litres or 500 litres: 10 litres: The media substances for 10 litres of 1B12 medium are sterilised in 7 litres of water, then 1 litre of a sterile 10% 2-(hydroxypropyl) -p-cyclodextrin solution are added, and seeded with 2 litres of a 20 litre intermediate culture. The duration of the main culture is 6 7 days, and the conditions are: 30*C, 250 rpm, 0.5 litres of air per litre of liquid per min, 0.5 bars excess pressure, pH control with H 2

SO

4 IKOH to pH 7.6 +/- 0.5 (i.e. no control between pH 7.1 and 8.1). 100 litres: The media substances for 100 litres of 1B12 medium are sterilised in 70 litres of water, then 10 litres of a sterile 10% 2-(hydroxypropyl) -p-cyclodextrin solution are added, and seeded with 20 litres of a 20 litre intermediate culture. The duration of the main culture is 6-7 days, and the conditions are: 300C, 200 rpm, 0.5 litres air per litre liquid per min., 0.5 bars excess pressure, pH control with H 2 SOSKOH to pH 7.6 +/- 0.5. The chain of seeding for a 100 litre fermentation is shown schematically as follows: maintenance culture (500ml) G52 medium 10 intermediate preculturesculture (e.g. 20 ) 10 % (4 x 500 ml) G52 medium G52 medium maintenance culture (500 ml) G52 medium main culture (e.g. 10 1) medium + HP-p-CD 500 litres: The media substances for 500 litres of 1 B1 2 medium are sterilised in 350 litres of water, then 50 litres of a sterile 10% 2-(hydroxypropyl) -p-cyclodextrin solution are added, and seeded with 100 litres of a 100 litre intermediate culture. The duration of the main culture is 6-7 days, and the conditions are: 300C, 120 rpm, 0.5 litres air per litre liquid per min., 0.5 bars excess pressure, pH control with H 2 SOSKOH to pH 7.6 +/- 0.5. Product analysis: Preparation of the sample: WO 99/66028 PCT/EP99/04171 - 52 50 ml samples are mixed with 2 ml of polystyrene resin Amberlite XAD16 (Rohm + Haas, Frankfurt, Germany) and shaken at 180 rpm for one hour at 300C. The resin is subsequently filtered using a 150 pm nylon sieve, washed with a little water and then added together with the filter to a 15 ml Nunc tube. Elution of the product from the resin: 10 ml of isopropanol (>99%) are added to the tube with the filter and the resin. Afterwards, the sealed tube is shaken for 30 minutes at room temperature on a Rota-Mixer (Labinco BV, Netherlands). Then, 2 ml of the liquid are centrifuged off and the supernatant is added using a pipette to HPLC tubes. HPLC analysis: Column: Waters-Symetry 018,100 x 4 mm, 3.5 pm WAT066220 + preliminary column 3.9 x 20 mm WAT054225 Solvents: A: 0.02 % phosphoric acid B: Acetonitrile (HPLC-Quality) Gradient: 41% B from 0 to 7 min. 100% B from 7.2 to 7.8 min. 41% B from 8 to 12 min. Oven temnp.: 300C Detection: 250 nm, UV-DAD detection Injection vol.: 10 pl Retention time: Epo A: 4.30 min Epo B: 5.38 min B: Effect of the addition of cyclodextrin and cyclodextrin derivatives to the epothilone concentrations attained. Cyclodextrins are cyclic (a-1,4)-linked oligosaccharides of a-D-glucopyranose with a relatively hydrophobic central cavity and a hydrophilic external surface area. The following are distinguished in particular (the figures in parenthesis give the number of glucose units per molecule): a-cyclodextrin (6), P-cyclodextrin (7), y- cyclodextrin (8), S-cyclodextrin (9), E- cyclodextrin (10), -cyclodextrin (11), TI-cyclodextrin (12), and 6 cyclodextrin (13). Especially preferred are 8-cyclodextrin and in particular c-cyclodextrin, @ cyclodextrin or y-cyclodextrin, or mixtures thereof.

WO 99/66028 PCT/EP99/04171 - 53 Cyclodextrin derivatives are primarily derivatives of the above-mentioned cyclodex trins, especially of o-cyclodextrin, P-cyclodextrin or y-cyclodextrin, primarily those in which one or more up to all of the hydroxy groups (3 per glucose radical) are etherified or este rified. Ethers are primarily alkyl ethers, especially lower alkyl, such as methyl or ethyl ether, also propyl or butyl ether; the aryl-hydroxyalkyl ethers, such as phenyl-hydroxy-lower-alkyl, especially phenyl-hydroxyethyl ether; the hydroxyalkyl ethers, in particular hydroxy-lower alkyl ethers, especially 2-hydroxyethyl, hydroxypropyl such as 2-hydroxypropyl or hydroxy butyl such as 2-hydroxybutyl ether; the carboxyalkyl ethers, in particular carboxy-lower-alkyl ethers, especially carboxymethyl or carboxyethyl ether; derivatised carboxyalkyl ethers, in particular derivatised carboxy-lower-alkyl ether in which the derivatised carboxy is etherified or amidated carboxy (primarily aminocarbonyl, mono- or di-lower-alkyl-aminocarbonyl, mor pholino-, piperidino-, pyrrolidino- or piperazino-carbonyl, or alkyloxycarbonyl), in particular lower alkoxycarbonyl-lower-alkyl ether, for example methyloxycarbonylpropyl ether or ethyloxycarbonylpropyl ether; the sulfoalkyl ethers, in particular sulfo-lower-alkyl ethers, especially sulfobutyl ether; cyclodextrins in which one or more OH groups are etherified with a radical of formula -O-[alk-O-],-H wherein alk is alkyl, especially lower alkyl, and n is a whole number from 2 to 12, especially 2 to 5, in particular 2 or 3; cyclodextrins in which one or more OH groups are etherified with a radical of formula R' 1 0 (Alk-O) Alk z7Y wherein R' is hydrogen, hydroxy, -O-(alk-O)z-H, -O-(alk(-R)-O-)p-H or -O-(alk(-R)-O-),-alk-CO-Y; alk in all cases is alkyl, especially lower alkyl; m, n, p, q and z are a whole number from 1 to 12, preferably 1 to 5, in particular 1 to 3; and Y is OR 1 or NR 2

R

3 , wherein R 1 , R 2 and R 3 independently of one another, are hydrogen or lower alkyl, or R 2 and

R

3 combined together with the linking nitrogen signify morpholino, piperidino, pyrrolidino or piperazino; or branched cyclodextrins, in which etherifications or acetals with other sugar molecules are present, especially glucosyl-, diglucosyl- (G 2 -p-cyclodextrin), maltosyl- or dimaltosyl cyclodextrin, or N-acetylglucosaminyl-, glucosaminyl-, N-acetylgalactosaminyl- or galactosaminyl-cyclodextrin.

WO 99/66028 PCT/EP99/04171 - 54 Esters are primarily alkanoyl esters, in particular lower alkanoyl esters, such as acetyl esters of cyclodextrins. It is also possible to have cyclodextrins in which two or more different said ether and ester groups are present at the same time. Mixtures of two or more of the said cyclodextrins and/or cyclodextrin derivatives may also exist. Preference is given in particular to a-, P- or y-cyclodextrins or the lower alkyl ethers thereof, such as methyl--cyclodextrin or in particular 2,6-di-O-methyl-B-cyclodextrin, or in particular the hydroxy lower alkyl ethers thereof, such as 2-hydroxypropyl-o-, 2-hydroxy propyl-p- or 2-hydroxypropyl-y-cyclodextrin. The cyclodextrins or cyclodextrin derivatives are added to the culture medium preferably in a concentration of 0.02 to 10, preferably 0.05 to 5, especially 0.1 to 4, for example 0.1 to 2 percent by weight (w/v). Cyclodextrins or cyclodextrin derivatives are known or may be produced by known processes (see for example US 3,459,731; US 4,383,992; US 4,535,152; US 4,659,696; EP 0 094 157; EP 0 149 197; EP 0 197 571; EP 0 300 526; EP 0 320 032; EP 0 499 322; EP 0 503 710; EP 0 818 469; WO 90/12035; WO 91/11200; WO 93/19061; WO 95/08993; WO 96/14090; GB 2,189,245; DE 3,118,218; DE 3,317,064 and the references mentioned the rein, which also refer to the synthesis of cyclodextrins or cyclodextrin derivatives, or also: T. Loftsson and M.E. Brewster (1996): Pharmaceutical Applications of Cyclodextrins: Drug Solubilization and Stabilisation: Journal of Pharmaceutical Science 85 (10):1017-1025; R.A. Rajewski and V.J. Stella(1996): Pharmaceutical Applications of Cyclodextrins: In Vivo Drug Delivery: Journal of Pharmaceutical Science 8 (11): 1142-1169). All the cyclodextrin derivatives tested here are obtainable from the company Fluka, Buchs, CH. The tests are carried out in 200 ml agitating flasks with 50 ml culture volume. As controls, flasks with adsorber resin Amberlite XAD-1 6 (Rohm & Haas, Frankfurt, Germany) and without any adsorber addition are used. After incubation for 5 days, the following epothilone titres can be determined by HPLC: Table 2: Addition order Conc Epo A [mg/] Epo B [mg/] No. [%w/v] Amnberlite XAD-16 (v/v) 2.0 (%vlv) 9.2 7 3.8 WO 99/66028 PCT/EP99/04171 - 55 Addition order Conc Epo A [mg/] Epo B [mg/] No. [%w/v]' 2-hydroxypropyl-p-cyclodextrin 56332 0.1 2.7 1.7 2-hydroxypropyl-p-cyclodextrin " 0.5 4.7 3.3 2-hydroxypropyl-p-cyclodextrin 1.0 4.7 3.4 2-hydroxypropyl-p-cyclodextrin 2.0 4.7 4.1 2-hydroxypropyl-p-cyclodextrin 5.0 1.7 0.5 2-hydroxypropyl- a-cyclodextrin 56330 0.5 1.2 1.2 2-hydroxypropyl- ca-cyclodextrin 1.0 1.2 1.2 2-hydroxypropyl- a-cyciodextrin 5.0 2.5 2.3 p-cyclodextrin 28707 0.1 1.6 1.3 P-cyclodextrin " 0.5 3.6 2.5 P-cyclodextrin " 1.0 4.8 3.7 P-cyclodextrin " 2.0 4.8 2.9 P-cyclodextrin " 5.0 1.1 0.4 methyl-p-cyclodextrin 66292 0.5 0.8 <0.3 methyl-p-cyclodextrin 1.0 <0.3 <0.3 methyl-p-cyclodextrin 2.0 <0.3 <0.3 2,6 di-o-methyl-p-cyclodextrin 39915 1.0 <0.3 <0.3 2-hydroxypropyl-y-cyclodextrin 56334 0.1 0.3 <0.3 2-hydroxypropyl-y-cyclodextrin 0.5 0.9 0.8 2-hydroxypropyl-y-cyclodextrin 1.0 1.1 0.7 2-hydroxypropyl-y-cyclodextrin 2.0 2.6 0.7 2-hydroxypropyl-y-cyclodextrin 5.0 5.0 1.1 no addition 0.5 0.5 ) Apart from Amberlite (%v/v), all percentages are by weight (%w/v). Few of the cyclodextrins tested (2,6-di-o-methyl-p-cyclodextrin, methyl-p-cyclodextrin) display no effect or a negative effect on epothilone production at the concentrations used. 1-2% 2-hydroxy-propyl-p-cyclodextrin and P-cyclodextrin increase epothilone production in the examples by 6 to 8 times compared with production using no cyclodextrins.

WO 99/66028 PCT/EP99/04171 - 56 C: 10 litre fermentation with 1% 2-(hydroxypropyl)-B-cyclodextrin): Fermentation is carried out in a 15 litre glass fermenter. The medium contains 10 g/l of 2-(hydroxypropyl)-p-cyclodextrin from Wacker Chemie, Munich, Germany. The progress of fermentation is illustrated in Table 3. Fermentation is ended after 6 days and working up takes place. Table 3: Progress of a 10 litre fermentation duration of culture [d] Epothilone A [mg/l] Epothilone B [mg/] 0 0 0 1 0 0 2 0.5 0.3 3 1.8 2.5 4 3.0 5.1 5 3.7 5.9 6 3.6 5.7 D: 100 litre fermentation with 1% 2-(hydroxypropyl)--cyclodextrin): Fermentation is carried out in a 150 litre fermenter. The medium contains 10 g/l of 2 (Hydroxypropyl)-B-cyclodextrin. The progress of fermentation is illustrated in Table 4. The fermentation is harvested after 7 days and worked up. Table 4: Progress of a 100 litre fermentation duration of Epothilone A Epothilone B culture [d] [mg/] [mg/l] 0 0 0 1 0 0 2 0.3 0 WO 99/66028 PCT/EP99/04171 - 57 3 0.9 1.1 4 1.5 2.3 5 1.6 3.3 6 1.8 3.7 7 1.8 3.5 E: 500 litre fermentation with 1% 2-(hydroxypropyl)-B-cyclodextrin): Fermentation is carried out in a 750 litre fermenter. The medium contains 10 g/Il of 2 (Hydroxypropyl)-p-cyclodextrin. The progress of fermentation is illustrated in Table 5. The fermentation is harvested after 7 days and worked up. Table 5: Progress of a 500 litre fermentation duration of culture [d] Epothilone A Epothilone B [mg/] [mg/] 0 0 0 1 0 0 2 0 0 3 0.6 0.6 4 1.7 2.2 5 3.1 4.5 6 3.1 5.1 F: Comparison example 10 litre fermentation without adding an adsorber: Fermentation is carried out in a 15 litre glass fermenter. The medium does not contain any cyclodextrin or other adsorber. The progress of fermentation is illustrated in Table 6. The fermentation is not harvested and worked up. Table 6: Progress of a 10 litre fermentation without adsorber.

WO 99/66028 PCTIEP99/04171 - 58 duration of culture [d] Epothilone A Epothilone B [mg/] [mg/l] 0 0 0 1 0 0 2 0 0 3 0 0 4 0.7 0.7 5 0.7 1.0 6 0.8 1.3 G: Working up of the epothilones: Isolation from a 500 litre main culture: The volume of harvest from the 500 litre main culture of example 2D is 450 litres and is separated using a Westfalia clarifying separator Type SA-20-06 (rpm = 6500) into the liquid phase (centrifugate + rinsing water = 650 litres) and solid phase (cells = ca. 15 kg). The main part of the epothilones are found in the centrifugate, The centrifuged cell pulp contains < 15% of the determined epothilone portion and is not further processed. The 650 litre centrifugate is then placed in a 4000 litre stirring vessel, mixed with 10 litres of Amberlite XAD-16 (centrifugate:resin volume = 65:1) and stirred. After a period of contact of ca. 2 hours, the resin is centrifuged away in a Heine overflow centrifuge (basket content 40 litres; rpm = 2800). The resin is discharged from the centrifuge and washed with 10-15 litres of deionised water. Desorption is effected by stirring the resin twice, each time in portions with 30 litres of isopropanol in 30 litre glass stirring vessels for 30 minutes. Separation of the isopropanol phase from the resin takes place using a suction filter. The isopropanol is then removed from the combined isopropanol phases by adding 15-20 litres of water in a vacuum-operated circulating evaporator (Schmid-Verdampfer) and the resulting water phase of ca. 10 litres is extracted 3x each time with 10 litres of ethyl acetate. Extraction is effected in 30 litre glass stirring vessels. The ethyl acetate extract is concentrated to 3-5 litres in a vacuum-operated circulating evaporator (Schmid-Verdampfer) and afterwards concentrated to dryness in a rotary evaporator (BQchi type) under vacuum. The result is an ethyl acetate extract of 50.2 g. The ethyl acetate extract is dissolved in WO 99/66028 PCT/EP99/04171 - 59 500 ml of methanol, the insoluble portions filtered off using a folded filter, and the solution added to a 10 kg Sephadex LH 20 column (Pharmacia, Uppsala, Sweden) (column diameter 20 cm, filling level ca. 1.2 m). Elution is effected with methanol as eluant. Epothilone A and B is present predominantly in fractions 21-23 (at a fraction size of 1 litre). These fractions are concentrated to dryness in a vacuum on a rotary evaporator (total weight 9.0 g). These Sephadex peak fractions (9.0 g) are thereafter dissolved in 92 ml of acetonitrile:-water:-methylene chloride = 50:40:2, the solution filtered through a folded filter and added to a RP column (equipment Prepbar 200, Merck; 2. 0 kg LiChrospher RP-1 8 Merck, grain size 12 im, column diameter 10 cm, filling level 42 cm; Merck, Darmstadt, Germany). Elution is effected with acetonitrile:water = 3:7 (flow rate = 500 ml/min.; retention time of epothilone A = ca. 51-59 mins.; retention time of epothilone B = ca. 60-69 mins.). Fractionation is monitored with a UV detector at 250 nm. The fractions are concentrated to dryness under vacuum on a Bjchi-Rotavapor rotary evaporator. The weight of the epothilone A peak fraction is 700 mg, and according to HPLC (external standard) it has a content of 75.1%. That of the epothilone B peak fraction is 1980 mg, and the content according to HPLC (external standard) is 86.6%. Finally, the epothilone A fraction (700 mg) is crystallised from 5 ml of ethyl acetate:toluene = 2:3, and yields 170 mg of epothilone A pure crystallisate [content according to HLPC (% of area) = 94.3%]. Crystallisation of the epothilone B fraction (1980 mg) is effected from 18 ml of methanol and yields 1440 mg of epothilone B pure crystallisate [content according to HPLC (% of area) = 99.2%]. m.p. (Epothilone B): e.g. 124-125 0C; 1 H-NMR data for Epothilone B: 500 MHz-NMR, solvent: DMSO-d6. Chemical displacement 8 in ppm relative to TMS. s = singlet; d = doublet; m = multiplet 6 (Multiplicity) Integral (number of H) 7.34 (s) 1 6.50 (s) 1 5.28 (d) 1 5.08 (d) 1 4.46 (d) 1 4.08 (m) 1 WO 99/66028 PCT/EP99/04171 -60 3.47 (m) 1 3.11 (m) 1 2.83 (dd) 1 2.64 (s) 3 2.36 (m) 2 2.09 (s) 3 2.04 (m) 1 1.83 (m) 1 1.61 (m) 1 1.47 - 1.24 (m) 4 1.18 (s) 6 1.13 (m) 2 1.06 (d) 3 0.89 (d + s, overlapping) 6 Z = 41 Example 15: Medical Uses of Recombinantly Produced Epothilones Pharmaceutical preparations or compositions comprising epothilones are used for example in the treatment of cancerous diseases, such as various human solid tumors. Such anticancer formulations comprise, for example, an active amount of an epothilone together with one or more organic or inorganic, liquid or solid, pharmaceutically suitable carrier materials. Such formulations are delivered, for example, enterally, nasally, rectally, orally, or parenterally, particularly intramuscularly or intravenously. The dosage of the active ingredient is dependent upon the weight, age, and physical and pharmacokinetical condition of the patient and is further dependent upon the method of delivery. Because epothilones mimic the biological effects of taxol, epothilones may be substituted for taxol in compositions and methods utilizing taxol in the treatment of cancer. See, for example, U.S.

WO 99/66028 PCT/EP99/04171 - 61 Patent Nos. 5,496,804, 5,565,478, and 5,641,803, all of which are incorporated herein by reference. For example, for treatments, epothilone B is supplied in individual 2 ml glass vials formulated as 1 mg/1 ml of clear, colorless intravenous concentrate. The substance is formulated in polyethylene glycol 300 (PEG 300) and diluted with 50 or 100 ml 0.9% Sodium Chloride Injection, USP, to achieve the desired final concentration of the drug for infusion. It is administered as a single 30-minute intravenous infusion every 21 days (treatment three-weekly) for six cycles, or as a single 30-minute intravenous infusion every 7 days (weekly treatment). Preferably, for weekly treatment, the dose is between about 0.1 and about 6, preferably about 0.1 and about 5 mg/m2, more preferably about 0.1 and about 3 mg/m 2 , even more preferably 0.1 and 1.7 mg/m 2 , most preferably about 0.3 and about 1 mg/m 2 ; for three-weekly treatment (treatment every three weeks or every third week) the dose is between about 0.3 and about 18 mg/m2, preferably about 0.3 and about 15 mg/m2 , more preferably about 0.3 and about 12 mg/m2 , even more preferably about 0.3 and about 7.5 mg/m 2 , still more preferably about 0.3 and about 5 mg/m 2 , most preferably about 1.0 and about 3.0 mg/m2 . This dose is preferably administered to the human by intravenous (i.v.) administration during 2 to 180 min, preferably 2 to 120 min, more preferably during about 5 to about 30 min, most preferably during about 10 to about 30 min, e.g. during about 30 min. While the present invention has been described with reference to specific embodiments thereof, it will be appreciated that numerous variations, modifications, and embodiments are possible, and accordingly, all such variations, modifications and embodiments are to be regarded as being within the spirit and scope of the present invention.

WO 99/66028 - 62 - PCT/EP99/04171 BUDAPEST TREATY ON THE INTERNATIONAL RECOGNITION OF THE DEPOSIT OF MICROORGANISMS FOR THE PURPOSE OF PATENT PROCEDURES INTERNATIONAL FORM To RECEIPT IN THE CASE OF AN ORIGINAL DEPOSIT Novartin AG issued pursuant to Rule 7.1 by the Novartis Corporation INTERNATIONAL DEPOSITARY AUTHORITY Patent and Trademark Dept. Identified at the bottom of this page 3054 Cornwallis Rd. Research Triangle Park, NC 27709 NAME AND ADDRESS OF DEPOSITOR I. IDENTIFICATION OF THE MICROORGANISM Identification reference given by the Accession number given by the DEPOSITOR: INTERNATIONAL DEPOSITARY AUTHORITY: Escherichla coi DH10B [pEP015) NRRL B-30033 II. SCIENTIFIC DESCRIPTION AND/OR PROPOSED TAXONOMIC DESIGNATION The microorganism identified under I. above was accompanied byr M a scientific description a proposed taxonomic designation (Mark with a cross where applicable) III. RECEIPT AND ACCEPTANCE This International Depositary Authority accepts the microorganism identified under I. above, which was received by it on June 11, 1998(date of the original deposit)' IV. RECEIPT OF REQUEST FOR CONVERSION The microorganism identified under I. above was received by this International Depositary Authority on (date of the original deposit) and a request to convert the original deposit to a deposit under the Budapest Treaty was received by it on (date of receipt of request for conversion). V. INTERNATIONAL DEPOSITARY AUTHORITY Name; Agricultural Research Culture Signature(s) of person(s) having the power Collection (NRRL) to represent the International Depositary International Depositary Authority Authority or of uthorized official(s): Addremso 1815 N. University Street / Peoria, Illinois 61604 U.S.A. Date Where Rule 6.4(d) applies, such date is the date on which the status of international depositary authority was acquired.

WO 99/66028 - 63 - PCT/EP99/04171 BUDAPEST TREATY ON THE IXTaRNM'I0Makh RECOGNITION OF THE DEPOSIT OF MICROORGANISMS FOR THE PURPOSE OF PATENT PROCEDURES INTERNATIONAL FORM TO RECEIPT IN TE CASE OF AN ORIGINAL DEPOSIT Novatis AG issued pursuant to Rule 7.1 by the c/o Novartis Agricultural Biotechnology INTB NATIONAL DEPOSITARY AUTHORITY Research, Int. identified at the bottom of this page Patent & Trademark Department 3054 Cornwallis Road Research Triangle Park, NC 27709 NAME AND ADDRESS OF DEPOSITOR I. IDENTIFICATION OF THE MICROORGANISM Identification reference given by the Accession number given by the DEPOSITOR: INTERNATIONAL DEPOSITARY AUTHORITY: Escherichia coli DH10B [EPO32] NRRL B-30119 II. SCIENTIFIC DESCRIPTION AND/OR PROPOSED TAXONOMIC DESIGNATION The microorganism identified under I. above was accompanied by: a scientific description L: a proposed taxonomic designation (Mark with a cross where applicable) III. RECEIPT AND ACCEPTANCE This International Depositary Authority accepts the microorganism identified under I. above, which was received by it on April 16, 1999 (date of the original deposit)' IV. RECEIPT OF REQUEST FOR CONVERSION The microorganism identified under I. above was received by this International Depositary Authority on (date of the original deposit) and a request to convert the original deposit to a deposit under the Budapest Treaty was received by it on (date of receipt of request for conversion). V. INTERNATIONAL DEPOSITARY AUTHORITY '_ Name: Agricultural Research Culture Signature(s) of person(s) having the power Collection (NRRL) to represent the International Depositary Internatirmal Depositary Authority Authority or of authorized officialss; Address: 1815 N. University Street 4,I, Peoria, Illinois 61604 U.S.A. Date: -Ti here Rule 6.4(d) applies, such date is the date on which the status of international depositary authority was acquired.

Claims

1. An isolated nucleic acid molecule comprising a nucleotide sequence that encodes at least one polypeptide involved in the biosynthesis of epothilone.

2. An isolated nucleic acid molecule according to claim 1, wherein said nucleotide sequence is isolated from a myxobacterium.

3. An isolated nucleic acid molecule according to claim 2, wherein said myxobacterium is Sorangium cellulosum.

4. A chimeric gene comprising a heterologous promoter sequence operatively linked to a nucleic acid molecule according to claim 1.

5. A recombinant vector comprising a chimeric gene according to claim 4.

6. A recombinant host cell comprising a chimeric gene according to claim 4.

7. The recombinant host cell of claim 6, which is a bacteria.

8. The recombinant host cell of claim 7, which is an Actinomycete.

9. The recombinant host cell of claim 8, which is Streptomyces.

10. A Bac clone comprising a nucleic acid molecule according to claim 1.

11. The Bac clone of claim 10, which is pEPO15.

12. An isolated nucleic acid molecule according to claim 1, wherein said polypeptide comprises an amino acid sequence substantially similar to an amino acid sequence selected from the group consisting of: SEQ ID NO:2, amino acids 11-437 of SEQ ID NO:2, amino acids 543-864 of SEQ ID NO:2, amino acids 974-1273 of SEQ ID NO:2, amino acids

1314-1385 of SEQ ID NO:2, SEQ ID NO:3, amino acids 72-81 of SEQ ID NO:3, amino acids WO 99/66028 PCT/EP99/04171 - 65 118-125 of SEQ ID NO:3, amino acids 199-212 of SEQ ID NO:3, amino acids 353-363 of SEQ ID NO:3, amino acids 549-565 of SEQ ID NO:3, amino acids 588-603 of SEQ ID NO:3, amino acids 669-684 of SEQ ID NO:3, amino acids 815-821 of SEQ ID NO:3, amino acids 868-892 of SEQ ID NO:3, amino acids 903-912 of SEQ ID NO:3, amino acids 918-940 of SEQ ID NO:3, amino acids 1268-1274 of SEQ ID NO:3, amino acids 1285-1297 of SEQ ID NO:3, amino acids 973-1256 of SEQ ID NO:3, amino acids 1344-1351 of SEQ ID NO:3, SEQ ID NO:4, amino acids 7-432 of SEQ ID NO:4, amino acids 539-859 of SEQ ID NO:4, amino acids 869-1037 of SEQ ID NO:4, amino acids 1439-1684 of SEQ ID NO:4, amino acids 1722-1792 of SEQ ID NO:4, SEQ ID NO:5, amino acids 39-457 of SEQ ID NO:5, amino acids 563-884 of SEQ ID NO:5, amino acids 1147-1399 of SEQ ID NO:5, amino acids 1434-1506 of SEQ ID NO:5, amino acids 1524-1950 of SEQ ID NO:5, amino acids

2056-2377 of SEQ ID NO:5, amino acids 2645-2895 of SEQ ID NO:5, amino acids 2932 3005 of SEQ ID NO:5, amino acids 3024-3449 of SEQ ID NO:5, amino acids 3555-3876 of SEQ ID NO:5, amino acids 3886-4048 of SEQ ID NO:5, amino acids 4433-4719 of SEQ ID NO:5, amino acids 4729-4974 of SEQ ID NO:5, amino acids 5010-5082 of SEQ ID NO:5, amino acids 5103-5525 of SEQ ID NO:5, amino acids 5631-5951 of SEQ ID NO:5, amino acids 5964-6132 of SEQ ID NO:5, amino acids 6542-6837 of SEQ ID NO:5, amino acids

6857-7101 of SEQ ID NO:5, amino acids 7140-7211 of SEQ ID NO:5, SEQ ID NO:6, amino acids 35-454 of SEQ ID NO:6, amino acids 561-881 of SEQ ID NO:6, amino acids 1143 1393 of SEQ ID NO:6, amino acids 1430-1503 of SEQ ID NO:6, amino acids 1522-1946 of SEQ ID NO: 6, amino acids 2053-2373 of SEQ ID NO:6, amino acids 2383-2551 of SEQ ID NO:6, amino acids 2671-3045 of SEQ ID NO:6, amino acids 3392-3636 of SEQ ID NO:6, amino acids 3673-3745 of SEQ ID NO:6, SEQ ID NO:7, amino acids 32-450 of SEQ ID NO:7, amino acids 556-877 of SEQ ID NO:7, amino acids 887-1051 of SEQ ID NO:7, amino acids 1478-1790 of SEQ ID NO:7, amino acids 1810-2055 of SEQ ID NO:7, amino acids 2093-2164 of SEQ ID NO:7, amino acids 2165-2439 of SEQ ID NO:7, SEQ ID NO:8, SEQ ID NO:10, SEQ ID NO:11, and SEQ ID NO:22.

13. An isolated nucleic acid molecule according to claim 12, wherein said polypeptide comprises an amino acid sequence selected from the group consisting of: SEQ ID NO:2, amino acids 11-437 of SEQ ID NO:2, amino acids 543-864 of SEQ ID NO:2, amino acids 974-1273 of SEQ ID NO:2, amino acids 1314-1385 of SEQ ID NO:2, SEQ ID NO:3, amino acids 72-81 of SEQ ID NO:3, amino acids 118-125 of SEQ ID NO:3, amino WO 99/66028 PCT/EP99/04171 - 66 acids 199-212 of SEQ ID NO:3, amino acids 353-363 of SEQ ID NO:3, amino acids 549 565 of SEQ ID NO:3, amino acids 588-603 of SEQ ID NO:3, amino acids 669-684 of SEQ ID NO:3, amino acids 815-821 of SEQ ID NO:3, amino acids 868-892 of SEQ ID NO:3, amino acids 903-912 of SEQ ID NO:3, amino acids 918-940 of SEQ ID NO:3, amino acids 1268-1274 of SEQ ID NO:3, amino acids 1285-1297 of SEQ ID NO:3, amino acids 973 1256 of SEQ ID NO:3, amino acids 1344-1351 of SEQ ID NO:3, SEQ ID NO:4, amino acids 7-432 of SEQ ID NO:4, amino acids 539-859 of SEQ ID NO:4, amino acids 869-1037 of SEQ ID NO:4, amino acids 1439-1684 of SEQ ID NO:4, amino acids 1722-1792 of SEQ ID NO:4, SEQ ID NO:5, amino acids 39-457 of SEQ ID NO:5, amino acids 563-884 of SEQ ID NO:5, amino acids 1147-1399 of SEQ ID NO:5, amino acids 1434-1506 of SEQ ID NO:5, amino acids 1524-1950 of SEQ ID NO:5, amino acids 2056-2377 of SEQ ID NO:5, amino acids 2645-2895 of SEQ ID NO:5, amino acids 2932-3005 of SEQ ID NO:5, amino acids 3024-3449 of SEQ ID NO:5, amino acids 3555-3876 of SEQ ID NO:5, amino acids 3886 4048 of SEQ ID NO:5, amino acids 4433-4719 of SEQ ID NO:5, amino acids 4729-4974 of SEQ ID NO:5, amino acids 5010-5082 of SEQ ID NO:5, amino acids 5103-5525 of SEQ ID NO:5, amino acids 5631-5951 of SEQ ID NO:5, amino acids 5964-6132 of SEQ ID NO:5, amino acids 6542-6837 of SEQ ID NO:5, amino acids 6857-7101 of SEQ ID NO:5, amino acids 7140-7211 of SEQ ID NO:5, SEQ ID NO:6, amino acids 35-454 of SEQ ID NO:6, amino acids 561-881 of SEQ ID NO:6, amino acids 1143-1393 of SEQ ID NO:6, amino acids 1430-1503 of SEQ ID NO:6, amino acids 1522-1946 of SEQ ID NO: 6, amino acids 2053-2373 of SEQ ID NO:6, amino acids 2383-2551 of SEQ ID NO:6, amino acids 2671 3045 of SEQ ID NO:6, amino acids 3392-3636 of SEQ ID NO:6, amino acids 3673-3745 of SEQ ID NO:6, SEQ ID NO:7, amino acids 32-450 of SEQ ID NO:7, amino acids 556-877 of SEQ ID NO:7, amino acids 887-1051 of SEQ ID NO:7, amino acids 1478-1790 of SEQ ID NO:7, amino acids 1810-2055 of SEQ ID NO:7, amino acids 2093-2164 of SEQ ID NO:7, amino acids 2165-2439 of SEQ ID NO:7, SEQ ID NO:8, SEQ ID NO:10, SEQ ID NO:11, and SEQ ID NO:22.

14. An isolated nucleic acid molecule according to claim 12, wherein said nucleotide sequence is substantially similar to a nucleotide sequence selected from the group consisting of: the complement of nucleotides 1900-3171 of SEQ ID NO:1, nucleotides 3415 5556 of SEQ ID NO:1, nucleotides 7610-11875 of SEQ ID NO:1, nucleotides 7643-8920 of SEQ ID NO:1, nucleotides 9236-10201 of SEQ ID NO:1, nucleotides 10529-11428 of SEQ WO 99/66028 PCT/EP99/04171 - 67 ID NO:1, nucleotides 11549-11764 of SEQ ID NO:1, nucleotides 11872-16104 of SEQ ID NO:1, nucleotides 12085-12114 of SEQ ID NO:1, nucleotides 12223-12246 of SEQ ID NO:1, nucleotides 12466-12507 of SEQ ID NO:1, nucleotides 12928-12960 of SEQ ID NO:1, nucleotides 13516-13566 of SEQ ID NO:1, nucleotides 13633-13680 of SEQ ID NO:1, nucleotides 13876-13923 of SEQ ID NO:1, nucleotides 14313-14334 of SEQ ID NO:1, nucleotides 14473-14547 of SEQ ID NO:1, nucleotides 14578-14607 of SEQ ID NO:1, nucleotides 14623-14692 of SEQ ID NO:1, nucleotides 15673-15693 of SEQ ID NO:1, nucleotides 15724-15762 of SEQ ID NO:1, nucleotides 14788-15639 of SEQ ID NO:1, nucleotides 15901-15924 of SEQ ID NO:1, nucleotides 16251-21749 of SEQ ID NO:1, nucleotides 16269-17546 of SEQ ID NO:1, nucleotides 17865-18827 of SEQ ID NO:1, nucleotides 18855-19361 of SEQ ID NO:1, nucleotides 20565-21302 of SEQ ID NO:1, nucleotides 21414-21626 of SEQ ID NO:1, nucleotides 21746-43519 of SEQ ID NO:1, nucleotides 21860-23116 of SEQ ID NO:1, nucleotides 23431-24397 of SEQ ID NO:1, nucleotides 25184-25942 of SEQ ID NO:1, nucleotides 26045-26263 of SEQ ID NO:1, nucleotides 26318-27595 of SEQ ID NO:1, nucleotides 27911-28876 of SEQ ID NO:1, nucleotides 29678-30429 of SEQ ID NO:1, nucleotides 30539-30759 of SEQ ID NO:1, nucleotides 30815-32092 of SEQ ID NO:1, nucleotides 32408-33373 of SEQ ID NO:1, nucleotides 33401-33889 of SEQ ID NO:1, nucleotides 35042-35902 of SEQ ID NO:1, nucleotides 35930-36667 of SEQ ID NO:1, nucleotides 36773-36991 of SEQ ID NO:1, nucleotides 37052-38320 of SEQ ID NO:1, nucleotides 38636-39598 of SEQ ID NO:1, nucleotides 39635-40141 of SEQ ID NO:1, nucleotides 41369-42256 of SEQ ID NO:1, nucleotides 42314-43048 of SEQ ID NO:1, nucleotides 43163-43378 of SEQ ID NO:1, nucleotides 43524-54920 of SEQ ID NO:1, nucleotides 43626-44885 of SEQ ID NO:1, nucleotides 45204-46166 of SEQ ID NO:1, nucleotides 46950-47702 of SEQ ID NO:1, nucleotides 47811-48032 of SEQ ID NO:1, nucleotides 48087-49361 of SEQ ID NO:1, nucleotides 49680-50642 of SEQ ID NO:1, nucleotides 50670-51176 of SEQ ID NO:1, nucleotides 51534-52657 of SEQ ID NO:1, nucleotides 53697-54431 of SEQ ID NO:1, nucleotides 54540-54758 of SEQ ID NO:1, nucleotides 54935-62254 of SEQ ID NO:1, nucleotides 55028-56284 of SEQ ID NO:1, nucleotides 56600-57565 of SEQ ID NO:1, nucleotides 57593-58087 of SEQ ID NO:1, nucleotides 59366-60304 of SEQ ID NO:1, nucleotides 60362-61099 of SEQ ID NO:1, nucleotides 61211-61426 of SEQ ID NO:1, nucleotides 61427-62254 of SEQ ID NO:1, nucleotides 62369-63628 of SEQ ID NO:1, nucleotides 67334-68251 of SEQ ID NO:1, and nucleotides 1-68750 SEQ ID NO:1. WO 99/66028 PCT/EP99/04171 - 68

15. A nucleic acid molecule according to claim 12, wherein said nucleotide sequence is selected from the group consisting of: the complement of nucleotides 1900 3171 of SEQ ID NO:1, nucleotides 3415-5556 of SEQ ID NO:1, nucleotides 7610-11875 of SEQ ID NO:1, nucleotides 7643-8920 of SEQ ID NO:1, nucleotides 9236-10201 of SEQ ID NO:1, nucleotides 10529-11428 of SEQ ID NO:1, nucleotides 11549-11764 of SEQ ID NO:1, nucleotides 11872-16104 of SEQ ID NO:1, nucleotides 12085-12114 of SEQ ID NO:1, nucleotides 12223-12246 of SEQ ID NO:1, nucleotides 12466-12507 of SEQ ID NO:1, nucleotides 12928-12960 of SEQ ID NO:1, nucleotides 13516-13566 of SEQ ID NO:1, nucleotides 13633-13680 of SEQ ID NO:1, nucleotides 13876-13923 of SEQ ID NO:1, nucleotides 14313-14334 of SEQ ID NO:1, nucleotides 14473-14547 of SEQ ID NO:1, nucleotides 14578-14607 of SEQ ID NO:1, nucleotides 14623-14692 of SEQ ID NO:1, nucleotides 15673-15693 of SEQ ID NO:1, nucleotides 15724-15762 of SEQ ID NO:1, nucleotides 14788-15639 of SEQ ID NO:1, nucleotides 15901-15924 of SEQ ID NO:1, nucleotides 16251-21749 of SEQ ID NO:1, nucleotides 16269-17546 of SEQ ID NO:1, nucleotides 17865-18827 of SEQ ID NO:1, nucleotides 18855-19361 of SEQ ID NO:1, nucleotides 20565-21302 of SEQ ID NO:1, nucleotides 21414-21626 of SEQ ID NO:1, nucleotides 21746-43519 of SEQ ID NO:1, nucleotides 21860-23116 of SEQ ID NO:1, nucleotides 23431-24397 of SEQ ID NO:1, nucleotides 25184-25942 of SEQ ID NO:1, nucleotides 26045-26263 of SEQ ID NO:1, nucleotides 26318-27595 of SEQ ID NO:1, nucleotides 27911-28876 of SEQ ID NO:1, nucleotides 29678-30429 of SEQ ID NO:1, nucleotides 30539-30759 of SEQ ID NO:1, nucleotides 30815-32092 of SEQ ID NO:1, nucleotides 32408-33373 of SEQ ID NO:1, nucleotides 33401-33889 of SEQ ID NO:1, nucleotides 35042-35902 of SEQ ID NO:1, nucleotides 35930-36667 of SEQ ID NO:1, nucleotides 36773-36991 of SEQ ID NO:1, nucleotides 37052-38320 of SEQ ID NO:1, nucleotides 38636-39598 of SEQ ID NO:1, nucleotides 39635-40141 of SEQ ID NO:1, nucleotides 41369-42256 of SEQ ID NO:1, nucleotides 42314-43048 of SEQ ID NO:1, nucleotides 43163-43378 of SEQ ID NO:1, nucleotides 43524-54920 of SEQ ID NO:1, nucleotides 43626-44885 of SEQ ID NO:1, nucleotides 45204-46166 of SEQ ID NO:1, nucleotides 46950-47702 of SEQ ID NO:1, nucleotides 47811-48032 of SEQ ID NO:1, nucleotides 48087-49361 of SEQ ID NO:1, nucleotides 49680-50642 of SEQ ID NO:1, nucleotides 50670-51176 of SEQ ID NO:1, nucleotides 51534-52657 of SEQ ID NO:1, nucleotides 53697-54431 of SEQ ID NO:1, nucleotides 54540-54758 of SEQ ID WO 99/66028 PCT/EP99/04171 - 69 NO:1, nucleotides 54935-62254 of SEQ ID NO:1, nucleotides 55028-56284 of SEQ ID NO:1, nucleotides 56600-57565 of SEQ ID NO:1, nucleotides 57593-58087 of SEQ ID NO:1, nucleotides 59366-60304 of SEQ ID NO:1, nucleotides 60362-61099 of SEQ ID NO:1, nucleotides 61211-61426 of SEQ ID NO:1, nucleotides 61427-62254 of SEQ ID NO:1, nucleotides 62369-63628 of SEQ ID NO:1, nucleotides 67334-68251 of SEQ ID NO:1, and nucleotides 1-68750 SEQ ID NO:1.

16. A chimeric gene comprising a heterologous promoter sequence operatively linked to a nucleic acid molecule according to claim 12.

17. A recombinant vector comprising a chimeric gene according to claim 16.

18. A recombinant host cell comprising a chimeric gene according to claim 16.

19. The recombinant host cell of claim 18, which is a bacteria.

20. The recombinant host cell of claim 19, which is an Actinomycete.

21. The recombinant host cell of claim 20, which is Streptomyces.

22. An isolated nucleic acid molecule according to claim 1, wherein said nucleotide sequence comprises a consecutive 20 base pair nucleotide portion identical in sequence to a consecutive 20 base pair portion of a nucleotide sequence selected from the group consisting of: the complement of nucleotides 1900-3171 of SEQ ID NO:1, nucleotides 3415 5556 of SEQ ID NO:1, nucleotides 7610-11875 of SEQ ID NO:1, nucleotides 7643-8920 of SEQ ID NO:1, nucleotides 9236-10201 of SEQ ID NO:1, nucleotides 10529-11428 of SEQ ID NO:1, nucleotides 11549-11764 of SEQ ID NO:1, nucleotides 11872-16104 of SEQ ID NO:1, nucleotides 12085-12114 of SEQ ID NO:1, nucleotides 12223-12246 of SEQ ID NO:1, nucleotides 12466-12507 of SEQ ID NO:1, nucleotides 12928-12960 of SEQ ID NO:1, nucleotides 13516-13566 of SEQ ID NO:1, nucleotides 13633-13680 of SEQ ID NO:1, nucleotides 13876-13923 of SEQ ID NO:1, nucleotides 14313-14334 of SEQ ID NO:1, nucleotides 14473-14547 of SEQ ID NO:1, nucleotides 14578-14607 of SEQ ID NO:1, nucleotides 14623-14692 of SEQ ID NO:1, nucleotides 15673-15693 of SEQ ID WO 99/66028 PCT/EP99/04171 - 70 NO:1, nucleotides 15724-15762 of SEQ ID NO:1, nucleotides 14788-15639 of SEQ ID NO:1, nucleotides 15901-15924 of SEQ ID NO:1, nucleotides 16251-21749 of SEQ ID NO:1, nucleotides 16269-17546 of SEQ ID NO:1, nucleotides 17865-18827 of SEQ ID NO:1, nucleotides 18855-19361 of SEQ ID NO:1, nucleotides 20565-21302 of SEQ ID NO:1, nucleotides 21414-21626 of SEQ ID NO:1, nucleotides 21746-43519 of SEQ ID NO:1, nucleotides 21860-23116 of SEQ ID NO:1, nucleotides 23431-24397 of SEQ ID NO:1, nucleotides 25184-25942 of SEQ ID NO:1, nucleotides 26045-26263 of SEQ ID NO:1, nucleotides 26318-27595 of SEQ ID NO:1, nucleotides 27911-28876 of SEQ ID NO:1, nucleotides 29678-30429 of SEQ ID NO:1, nucleotides 30539-30759 of SEQ ID NO:1, nucleotides 30815-32092 of SEQ ID NO:1, nucleotides 32408-33373 of SEQ ID NO:1, nucleotides 33401-33889 of SEQ ID NO:1, nucleotides 35042-35902 of SEQ ID NO:1, nucleotides 35930-36667 of SEQ ID NO:1, nucleotides 36773-36991 of SEQ ID NO:1, nucleotides 37052-38320 of SEQ ID NO:1, nucleotides 38636-39598 of SEQ ID NO:1, nucleotides 39635-40141 of SEQ ID NO:1, nucleotides 41369-42256 of SEQ ID NO:1, nucleotides 42314-43048 of SEQ ID NO:1, nucleotides 43163-43378 of SEQ ID NO:1, nucleotides 43524-54920 of SEQ ID NO:1, nucleotides 43626-44885 of SEQ ID NO:1, nucleotides 45204-46166 of SEQ ID NO:1, nucleotides 46950-47702 of SEQ ID NO:1, nucleotides 47811-48032 of SEQ ID NO:1, nucleotides 48087-49361 of SEQ ID NO:1, nucleotides 49680-50642 of SEQ ID NO:1, nucleotides 50670-51176 of SEQ ID NO:1, nucleotides 51534-52657 of SEQ ID NO:1, nucleotides 53697-54431 of SEQ ID NO:1, nucleotides 54540-54758 of SEQ ID NO:1, nucleotides 54935-62254 of SEQ ID NO:1, nucleotides 55028-56284 of SEQ ID NO:1, nucleotides 56600-57565 of SEQ ID NO:1, nucleotides 57593-58087 of SEQ ID NO:1, nucleotides 59366-60304 of SEQ ID NO:1, nucleotides 60362-61099 of SEQ ID NO:1, nucleotides 61211-61426 of SEQ ID NO:1, nucleotides 61427-62254 of SEQ ID NO:1, nucleotides 62369-63628 of SEQ ID NO:1, nucleotides 67334-68251 of SEQ ID NO:1, and nucleotides 1-68750 SEQ ID NO:1.

23. A chimeric gene comprising a heterologous promoter sequence operatively linked to a nucleic acid molecule according to claim 22.

24. A recombinant vector comprising a chimeric gene according to claim 23.

25. A recombinant host cell comprising a chimeric gene according to claim 23. WO 99/66028 PCT/EP99/04171 - 71

26. The recombinant host cell of claim 25, which is a bacteria.

27. The recombinant host cell of claim 26, which is an Actinomycete.

28. The recombinant host cell of claim 27, which is Streptomyces.

29. An isolated nucleic acid molecule comprising a nucleotide sequence that encodes at least one epothilone synthase domain.

30. An isolated nucleic acid molecule according to claim 29, wherein said epothilone synthase domain is a P-ketoacyl-synthase domain comprising an amino acid sequence substantially similar to an amino acid sequence selected from the group consisting of: amino acids 11-437 of SEQ ID NO:2, amino acids 7-432 of SEQ ID NO:4, amino acids 39 457 of SEQ ID NO:5, amino acids 1524-1950 of SEQ ID NO:5, amino acids 3024-3449 of SEQ ID NO:5, amino acids 5103-5525 of SEQ ID NO:5, amino acids 35-454 of SEQ ID NO:6, amino acids 1522-1946 of SEQ ID NO: 6, and amino acids 32-450 of SEQ ID NO:7.

31. An isolated nucleic acid molecule according to claim 30, wherein said P ketoacyl-synthase domain comprises an amino acid sequence selected from the group consisting of: amino acids 11-437 of SEQ ID NO:2, amino acids 7-432 of SEQ ID NO:4, amino acids 39-457 of SEQ ID NO:5, amino acids 1524-1950 of SEQ ID NO:5, amino acids 3024-3449 of SEQ ID NO:5, amino acids 5103-5525 of SEQ ID NO:5, amino acids 35-454 of SEQ ID NO:6, amino acids 1522-1946 of SEQ ID NO: 6, and amino acids 32-450 of SEQ ID NO:7.

32. An isolated nucleic acid molecule according to claim 30, wherein said nucleotide sequence is substantially similar to a nucleotide sequence selected from the group consisting of: nucleotides 7643-8920 of SEQ ID NO:1, nucleotides 16269-17546 of SEQ ID NO:1, nucleotides 21860-23116 of SEQ ID NO:1, nucleotides 26318-27595 of SEQ ID NO:1, nucleotides 30815-32092 of SEQ ID NO:1, nucleotides 37052-38320 of SEQ ID NO:1, nucleotides 43626-44885 of SEQ ID NO:1, nucleotides 48087-49361 of SEQ ID NO:1, and nucleotides 55028-56284 of SEQ ID NO:1. WO 99/66028 PCT/EP99/04171 - 72

33. An isolated nucleic acid molecule according to claim 30, wherein said nucleotide sequence comprises a consecutive 20 base pair nucleotide portion identical in sequence to a consecutive 20 base pair portion of a nucleotide sequence selected from the group consisting of: nucleotides 7643-8920 of SEQ ID NO:1, nucleotides 16269-17546 of SEQ ID NO:1, nucleotides 21860-23116 of SEQ ID NO:1, nucleotides 26318-27595 of SEQ ID NO:1, nucleotides 30815-32092 of SEQ ID NO:1, nucleotides 37052-38320 of SEQ ID NO:1, nucleotides 43626-44885 of SEQ ID NO:1, nucleotides 48087-49361 of SEQ ID NO:1, and nucleotides 55028-56284 of SEQ ID NO:1.

34. An isolated nucleic acid molecule according to claim 30, wherein said nucleotide sequence is selected from the group consisting of: nucleotides 7643-8920 of SEQ ID NO:1, nucleotides 16269-17546 of SEQ ID NO:1, nucleotides 21860-23116 of SEQ ID NO:1, nucleotides 26318-27595 of SEQ ID NO:1, nucleotides 30815-32092 of SEQ ID NO:1, nucleotides 37052-38320 of SEQ ID NO:1, nucleotides 43626-44885 of SEQ ID NO:1, nucleotides 48087-49361 of SEQ ID NO:1, and nucleotides 55028-56284 of SEQ ID NO:1.

35. An isolated nucleic acid molecule according to claim 29, wherein said epothilone synthase domain is a an acyltransferase domain comprising an amino acid sequence substantially similar to an amino acid sequence selected from the group consisting of: amino acids 543-864 of SEQ ID NO:2, amino acids 539-859 of SEQ ID NO:4, amino acids 563-884 of SEQ ID NO:5, amino acids 2056-2377 of SEQ ID NO:5, amino acids 3555-3876 of SEQ ID NO:5, amino acids 5631-5951 of SEQ ID NO:5, amino acids 561-881 of SEQ ID NO:6, amino acids 2053-2373 of SEQ ID NO:6, and amino acids 556-877 of SEQ ID NO:7.

36. An isolated nucleic acid molecule according to claim 35, wherein said acyltransferase domain comprises an amino acid sequence selected from the group consisting of: amino acids 543-864 of SEQ ID NO:2, amino acids 539-859 of SEQ ID NO:4, amino acids 563-884 of SEQ ID NO:5, amino acids 2056-2377 of SEQ ID NO:5, amino acids 3555-3876 of SEQ ID NO:5, amino acids 5631-5951 of SEQ ID NO:5, amino acids 561-881 of SEQ ID NO:6, amino acids 2053-2373 of SEQ ID NO:6, and amino acids 556 877 of SEQ ID NO:7. WO 99/66028 PCT/EP99/04171 - 73

37. An isolated nucleic acid molecule according to claim 35, wherein said nucleotide sequence is substantially similar to a nucleotide sequence selected from the group consisting of: nucleotides 9236-10201 of SEQ ID NO:1, nucleotides 17865-18827 of SEQ ID NO:1, nucleotides 23431-24397 of SEQ ID NO:1, nucleotides 27911-28876 of SEQ ID NO:1, nucleotides 32408-33373 of SEQ ID NO:1, nucleotides 38636-39598 of SEQ ID NO:1, nucleotides 45204-46166 of SEQ ID NO:1, nucleotides 49680-50642 of SEQ ID NO:1, and nucleotides 56600-57565 of SEQ ID NO:1.

38. An isolated nucleic acid molecule according to claim 35, wherein said nucleotide sequence comprises a consecutive 20 base pair nucleotide portion identical in sequence to a consecutive 20 base pair portion of a nucleotide sequence selected from the group consisting of: nucleotides 9236-10201 of SEQ ID NO:1, nucleotides 17865-18827 of SEQ ID NO:1, nucleotides 23431-24397 of SEQ ID NO:1, nucleotides 27911-28876 of SEQ ID NO:1, nucleotides 32408-33373 of SEQ ID NO:1, nucleotides 38636-39598 of SEQ ID NO:1, nucleotides 45204-46166 of SEQ ID NO:1, nucleotides 49680-50642 of SEQ ID NO:1, and nucleotides 56600-57565 of SEQ ID NO:1.

39. An isolated nucleic acid molecule according to claim 35, wherein said nucleotide sequence is selected from the group consisting of: nucleotides 9236-10201 of SEQ ID NO:1, nucleotides 17865-18827 of SEQ ID NO:1, nucleotides 23431-24397 of SEQ ID NO:1, nucleotides 27911-28876 of SEQ ID NO:1, nucleotides 32408-33373 of SEQ ID NO:1, nucleotides 38636-39598 of SEQ ID NO:1, nucleotides 45204-46166 of SEQ ID NO:1, nucleotides 49680-50642 of SEQ ID NO:1, and nucleotides 56600-57565 of SEQ ID NO:1.

40. An isolated nucleic acid molecule according to claim 29, wherein said epothilone synthase domain is an enoyl reductase domain comprising an amino acid sequence substantially similar to an amino acid sequence selected from the group consisting of: amino acids 974-1273 of SEQ ID NO:2, amino acids 4433-4719 of SEQ ID NO:5, amino acids 6542-6837 of SEQ ID NO:5, and amino acids 1478-1790 of SEQ ID NO:7.

41. An isolated nucleic acid molecule according to claim 40, wherein said enoyl reductase domain comprises an amino acid sequence selected from the group consisting WO 99/66028 PCT/EP99/04171 - 74 of: amino acids 974-1273 of SEQ ID NO:2, amino acids 4433-4719 of SEQ ID NO:5, amino acids 6542-6837 of SEQ ID NO:5, and amino acids 1478-1790 of SEQ ID NO:7.

42. An isolated nucleic acid molecule according to claim 40, wherein said nucleotide sequence is substantially similar to a nucleotide sequence selected from the group consisting of: nucleotides 10529-11428 of SEQ ID NO:1, nucleotides 35042-35902 of SEQ ID NO:1, nucleotides 41369-42256 of SEQ ID NO:1, and nucleotides 59366-60304 of SEQ ID NO:1.

43. An isolated nucleic acid molecule according to claim 40, wherein said nucleotide sequence comprises a consecutive 20 base pair nucleotide portion identical in sequence to a consecutive 20 base pair portion of a nucleotide sequence selected from the group consisting of: nucleotides 10529-11428 of SEQ ID NO:1, nucleotides 35042-35902 of SEQ ID NO:1, nucleotides 41369-42256 of SEQ ID NO:1, and nucleotides 59366-60304 of SEQ ID NO:1.

44. An isolated nucleic acid molecule according to claim 40, wherein said nucleotide sequence is selected from the group consisting of: nucleotides 10529-11428 of SEQ ID NO:1, nucleotides 35042-35902 of SEQ ID NO:1, nucleotides 41369-42256 of SEQ ID NO:1, and nucleotides 59366-60304 of SEQ ID NO:1.

45. An isolated nucleic acid molecule according to claim 29, wherein said epothilone synthase domain is an acyl carrier protein domain comprising an amino acid sequence substantially similar to an amino acid sequence selected from the group consisting of: amino acids 1314-1385 of SEQ ID NO:2, amino acids 1722-1792 of SEQ ID NO:4, amino acids 1434-1506 of SEQ ID NO:5, amino acids 2932-3005 of SEQ ID NO:5, amino acids 5010-5082 of SEQ ID NO:5, amino acids 7140-7211 of SEQ ID NO:5, amino acids 1430 1503 of SEQ ID NO:6, amino acids 3673-3745 of SEQ ID NO:6, and amino acids 2093 2164 of SEQ ID NO:7.

46. An isolated nucleic acid molecule according to claim 45, wherein said acyl carrier protein domain comprises an amino acid sequence selected from the group consisting of: amino acids 1314-1385 of SEQ ID NO:2, amino acids 1722-1792 of SEQ ID WO 99/66028 PCT/EP99/04171 - 75 NO:4, amino acids 1434-1506 of SEQ ID NO:5, amino acids 2932-3005 of SEQ ID NO:5, amino acids 5010-5082 of SEQ ID NO:5, amino acids 7140-7211 of SEQ ID NO:5, amino acids 1430-1503 of SEQ ID NO:6, amino acids 3673-3745 of SEQ ID NO:6, and amino acids 2093-2164 of SEQ ID NO:7.

47. An isolated nucleic acid molecule according to claim 45, wherein said nucleotide sequence is substantially similar to a nucleotide sequence selected from the group consisting of: nucleotides 11549-11764 of SEQ ID NO:1, nucleotides 21414-21626 of SEQ ID NO:1, nucleotides 26045-26263 of SEQ ID NO:1, nucleotides 30539-30759 of SEQ ID NO:1, nucleotides 36773-36991 of SEQ ID NO:1, nucleotides 43163-43378 of SEQ ID NO:1, nucleotides 47811-48032 of SEQ ID NO:1, nucleotides 54540-54758 of SEQ ID NO:1, and nucleotides 61211-61426 of SEQ ID NO:1.

48. An isolated nucleic acid molecule according to claim 45, wherein said nucleotide sequence comprises a consecutive 20 base pair nucleotide portion identical in sequence to a consecutive 20 base pair portion of a nucleotide sequence selected from the group consisting of: nucleotides 11549-11764 of SEQ ID NO:1, nucleotides 21414-21626 of SEQ ID NO:1, nucleotides 26045-26263 of SEQ ID NO:1, nucleotides 30539-30759 of SEQ ID NO:1, nucleotides 36773-36991 of SEQ ID NO:1, nucleotides 43163-43378 of SEQ ID NO:1, nucleotides 47811-48032 of SEQ ID NO:1, nucleotides 54540-54758 of SEQ ID NO:1, and nucleotides 61211-61426 of SEQ ID NO:1.

49. An isolated nucleic acid molecule according to claim 45, wherein said nucleotide sequence is selected from the group consisting of: nucleotides 11549-11764 of SEQ ID NO:1, nucleotides 21414-21626 of SEQ ID NO:1, nucleotides 26045-26263 of SEQ ID NO:1, nucleotides 30539-30759 of SEQ ID NO:1, nucleotides 36773-36991 of SEQ ID NO:1, nucleotides 43163-43378 of SEQ ID NO:1, nucleotides 47811-48032 of SEQ ID NO:1, nucleotides 54540-54758 of SEQ ID NO:1, and nucleotides 61211-61426 of SEQ ID NO:1.

50. An isolated nucleic acid molecule according to claim 29, wherein said epothilone synthase domain is a dehydratase domain comprising an amino acid sequence substantially similar to an amino acid sequence selected from the group consisting of: WO 99/66028 PCTIEP99/04171 - 76 amino acids 869-1037 of SEQ ID NO:4, amino acids 3886-4048 of SEQ ID NO:5, amino acids 5964-6132 of SEQ ID NO:5, amino acids 2383-2551 of SEQ ID NO:6, and amino acids 887-1051 of SEQ ID NO:7.

51. An isolated nucleic acid molecule according to claim 50, wherein said dehydratase domain comprises an amino acid sequence selected from the group consisting of: amino acids 869-1037 of SEQ ID NO:4, amino acids 3886-4048 of SEQ ID NO:5, amino acids 5964-6132 of SEQ ID NO:5, amino acids 2383-2551 of SEQ ID NO:6, and amino acids 887-1051 of SEQ ID NO:7.

52. An isolated nucleic acid molecule according to claim 50, wherein said nucleotide sequence is substantially similar to a nucleotide sequence selected from the group consisting of: nucleotides 18855-19361 of SEQ ID NO:1, nucleotides 33401-33889 of SEQ ID NO:1, nucleotides 39635-40141 of SEQ ID NO:1, nucleotides 50670-51176 of SEQ ID NO:1, and nucleotides 57593-58087 of SEQ ID NO:1.

53. An isolated nucleic acid molecule according to claim 50, wherein said nucleotide sequence comprises a consecutive 20 base pair nucleotide portion identical in sequence to a consecutive 20 base pair portion of a nucleotide sequence selected from the group consisting of: nucleotides 18855-19361 of SEQ ID NO:1, nucleotides 33401-33889 of SEQ ID NO:1, nucleotides 39635-40141 of SEQ ID NO:1, nucleotides 50670-51176 of SEQ ID NO:1, and nucleotides 57593-58087 of SEQ ID NO:1.

54. An isolated nucleic acid molecule according to claim 50, wherein said nucleotide sequence is selected from the group consisting of: nucleotides 18855-19361 of SEQ ID NO:1, nucleotides 33401-33889 of SEQ ID NO:1, nucleotides 39635-40141 of SEQ ID NO:1, nucleotides 50670-51176 of SEQ ID NO:1, and nucleotides 57593-58087 of SEQ ID NO:1.

55. An isolated nucleic acid molecule according to claim 29, wherein said epothilone synthase domain is a @-ketoreductase domain comprising an amino acid sequence substantially similar to an amino acid sequence selected from the group consisting of: amino acids 1439-1684 of SEQ ID NO:4, amino acids 1147-1399 of SEQ ID NO:5, amino WO 99/66028 PCT/EP99/04171 - 77 acids 2645-2895 of SEQ ID NO:5, amino acids 4729-4974 of SEQ ID NO:5, amino acids 6857-7101 of SEQ ID NO:5, amino acids 1143-1393 of SEQ ID NO:6, amino acids 3392 3636 of SEQ ID NO:6, and amino acids 1810-2055 of SEQ ID NO:7.

56. An isolated nucleic acid molecule according to claim 55, wherein said ketoreductase domain comprises an amino acid sequence selected from the group consisting of: amino acids 1439-1684 of SEQ ID NO:4, amino acids 1147-1399 of SEQ ID NO:5, amino acids 2645-2895 of SEQ ID NO:5, amino acids 4729-4974 of SEQ ID NO:5, amino acids 6857-7101 of SEQ ID NO:5, amino acids 1143-1393 of SEQ ID NO:6, amino acids 3392-3636 of SEQ ID NO:6, and amino acids 1810-2055 of SEQ ID NO:7.

57. An isolated nucleic acid molecule according to claim 55, wherein said nucleotide sequence is substantially similar to a nucleotide sequence selected from the group consisting of: nucleotides 20565-21302 of SEQ ID NO:1, nucleotides 25184-25942 of SEQ ID NO:1, nucleotides 29678-30429 of SEQ ID NO:1, nucleotides 35930-36667 of SEQ ID NO:1, nucleotides 42314-43048 of SEQ ID NO:1, nucleotides 46950-47702 of SEQ ID NO:1, nucleotides 53697-54431 of SEQ ID NO:1, and nucleotides 60362-61099 of SEQ ID NO:1.

58. An isolated nucleic acid molecule according to claim 55, wherein said nucleotide sequence comprises a consecutive 20 base pair nucleotide portion identical in sequence to a consecutive 20 base pair portion of a nucleotide sequence selected from the group consisting of: nucleotides 20565-21302 of SEQ ID NO:1, nucleotides 25184-25942 of SEQ ID NO:1, nucleotides 29678-30429 of SEQ ID NO:1, nucleotides 35930-36667 of SEQ ID NO:1, nucleotides 42314-43048 of SEQ ID NO:1, nucleotides 46950-47702 of SEQ ID NO:1, nucleotides 53697-54431 of SEQ ID NO:1, and nucleotides 60362-61099 of SEQ ID NO:1.

59. An isolated nucleic acid molecule according to claim 55, wherein said nucleotide sequence is selected from the group consisting of: nucleotides 20565-21302 of SEQ ID NO:1, nucleotides 25184-25942 of SEQ ID NO:1, nucleotides 29678-30429 of SEQ ID NO:1, nucleotides 35930-36667 of SEQ ID NO:1, nucleotides 42314-43048 of SEQ ID WO 99/66028 PCT/EP99/04171 - 78 NO:1, nucleotides 46950-47702 of SEQ ID NO:1, nucleotides 53697-54431 of SEQ ID NO:1, and nucleotides 60362-61099 of SEQ ID NO:1.

60. An isolated nucleic acid molecule according to claim 29, wherein said epothilone synthase domain is a methyltransferase domain comprising an amino acid sequence substantially similar to amino acids 2671-3045 of SEQ ID NO:6.

61. An isolated nucleic acid molecule according to claim 60, wherein said methyltransf erase domain comprises amino acids 2671-3045 of SEQ ID NO:6.

62. An isolated nucleic acid molecule according to claim 60, wherein said nucleotide sequence is substantially similar to nucleotides 51534-52657 of SEQ ID NO:1.

63. An isolated nucleic acid molecule according to claim 60, wherein said nucleotide sequence comprises a consecutive 20 base pair nucleotide portion identical in sequence to a consecutive 20 base pair portion of nucleotides 51534-52657 of SEQ ID NO:1.

64. An isolated nucleic acid molecule according to claim 60, wherein said nucleotide sequence is nucleotides 51534-52657 of SEQ ID NO:1.

65. An isolated nucleic acid molecule according to claim 29, wherein said epothilone synthase domain is a thioesterase domain comprising an amino acid sequence substantially similar to amino acids 2165-2439 of SEQ ID NO:7.

66. An isolated nucleic acid molecule according to claim 65, wherein said thioesterase domain comprises amino acids 2165-2439 of SEQ ID NO:7.

67. An isolated nucleic acid molecule according to claim 65, wherein said nucleotide sequence is substantially similar to nucleotides 61427-62254 of SEQ ID NO:1.

68. An isolated nucleic acid molecule according to claim 65, wherein said nucleotide sequence comprises a consecutive 20 base pair nucleotide portion identical in sequence to a consecutive 20 base pair portion of nucleotides 61427-62254 of SEQ ID NO:1. WO 99/66028 PCT/EP99/04171 - 79

69. An isolated nucleic acid molecule according to claim 65, wherein said nucleotide sequence is nucleotides 61427-62254 of SEQ ID NO:1.

70. An isolated nucleic acid molecule comprising a nucleotide sequence that encodes a non-ribosomal peptide synthetase, wherein said non-ribosomal peptide synthetase comprises an amino acid sequence substantially similar to an amino acid sequence selected from the group consisting of: SEQ ID NO:3, amino acids 72-81 of SEQ ID NO:3, amino acids 118-125 of SEQ ID NO:3, amino acids 199-212 of SEQ ID NO:3, amino acids 353-363 of SEQ ID NO:3, amino acids 549-565 of SEQ ID NO:3, amino acids 588-603 of SEQ ID NO:3, amino acids 669-684 of SEQ ID NO:3, amino acids 815-821 of SEQ ID NO:3, amino acids 868-892 of SEQ ID NO:3, amino acids 903-912 of SEQ ID NO:3, amino acids 918-940 of SEQ ID NO:3, amino acids 1268-1274 of SEQ ID NO:3, amino acids 1285-1297 of SEQ ID NO:3, amino acids 973-1256 of SEQ ID NO:3, and amino acids 1344-1351 of SEQ ID NO:3.

71. An isolated nucleic acid molecule according to claim 70, wherein said non ribosomal peptide synthetase comprises an amino acid sequence selected from the group consisting of: SEQ ID NO:3, amino acids 72-81 of SEQ ID NO:3, amino acids 118-125 of SEQ ID NO:3, amino acids 199-212 of SEQ ID NO:3, amino acids 353-363 of SEQ ID NO:3, amino acids 549-565 of SEQ ID NO:3, amino acids 588-603 of SEQ ID NO:3, amino acids 669-684 of SEQ ID NO:3, amino acids 815-821 of SEQ ID NO:3, amino acids 868-892 of SEQ ID NO:3, amino acids 903-912 of SEQ ID NO:3, amino acids 918-940 of SEQ ID NO:3, amino acids 1268-1274 of SEQ ID NO:3, amino acids 1285-1297 of SEQ ID NO:3, amino acids 973-1256 of SEQ ID NO:3, and amino acids 1344-1351 of SEQ ID NO:3.

72. An isolated nucleic acid molecule according to claim 70, wherein said nucleotide sequence is substantially similar to a nucleotide sequence selected from the group consisting of: nucleotides 11872-16104 of SEQ ID NO:1, nucleotides 12085-12114 of SEQ ID NO:1, nucleotides 12223-12246 of SEQ ID NO:1, nucleotides 12466-12507 of SEQ ID NO:1, nucleotides 12928-12960 of SEQ ID NO:1, nucleotides 13516-13566 of SEQ ID NO:1, nucleotides 13633-13680 of SEQ ID NO:1, nucleotides 13876-13923 of SEQ ID NO:1, nucleotides 14313-14334 of SEQ ID NO:1, nucleotides 14473-14547 of SEQ ID WO 99/66028 PCT/EP99/04171 - 80 NO:1, nucleotides 14578-14607 of SEQ ID NO:1, nucleotides 14623-14692 of SEQ ID NO:1, nucleotides 15673-15693 of SEQ ID NO:1, nucleotides 15724-15762 of SEQ ID NO:1, nucleotides 14788-15639 of SEQ ID NO:1, and nucleotides 15901-15924 of SEQ ID NO:1.

73. An isolated nucleic acid molecule according to claim 70, wherein said nucleotide sequence comprises a consecutive 20 base pair nucleotide portion identical in sequence to a consecutive 20 base pair portion of a nucleotide sequence selected from the group consisting of: nucleotides 11872-16104 of SEQ ID NO:1, nucleotides 12085-12114 of SEQ ID NO:1, nucleotides 12223-12246 of SEQ ID NO:1, nucleotides 12466-12507 of SEQ ID NO:1, nucleotides 12928-12960 of SEQ ID NO:1, nucleotides 13516-13566 of SEQ ID NO:1, nucleotides 13633-13680 of SEQ ID NO:1, nucieotides 13876-13923 of SEQ ID NO:1, nucleotides 14313-14334 of SEQ ID NO:1, nucleotides 14473-14547 of SEQ ID NO:1, nucleotides 14578-14607 of SEQ ID NO:1, nucleotides 14623-14692 of SEQ ID NO:1, nucleotides 15673-15693 of SEQ ID NO:1, nucleotides 15724-15762 of SEQ ID NO:1, nucleotides 14788-15639 of SEQ ID NO:1, and nucleotides 15901-15924 of SEQ ID NO:1.

74. An isolated nucleic acid molecule according to claim 70, wherein said nucleotide sequence is selected from the group consisting of: nucleotides 11872-16104 of SEQ ID NO:1, nucleotides 12085-12114 of SEQ ID NO:1, nucleotides 12223-12246 of SEQ ID NO:1, nucleotides 12466-12507 of SEQ ID NO:1, nucleotides 12928-12960 of SEQ ID NO:1, nucleotides 13516-13566 of SEQ ID NO:1, nucleotides 13633-13680 of SEQ ID NO:1, nucleotides 13876-13923 of SEQ ID NO:1, nucleotides 14313-14334 of SEQ ID NO:1, nucleotides 14473-14547 of SEQ ID NO:1, nucleotides 14578-14607 of SEQ ID NO:1, nucleotides 14623-14692 of SEQ ID NO:1, nucleotides 15673-15693 of SEQ ID NO:1, nucleotides 15724-15762 of SEQ ID NO:1, nucleotides 14788-15639 of SEQ ID NO:1, and nucleotides 15901-15924 of SEQ ID NO:1.

75. A method for heterologous expression of epothilone in a recombinant host, comprising: (a) introducing a chimeric gene according to claim 4 into a host; and (b) growing the host in conditions that allow biosynthesis of epothilone in the host. WO 99/66028 - 81 - PCTIEP99/04171

76. A method for producing epothilone, comprising: (a) expressing epothilone in a recombinant host by the method of claim 75; and (b) extracting epothilone from the recombinant host.

77. An isolated polypeptide comprising an amino acid sequence that consists of an epothilone synthase domain.

78. An isolated polypeptide according to claim 77, wherein said epothilone synthase domain is a P-ketoacyl-synthase domain comprising an amino acid sequence substantially similar to an amino acid sequence selected from the group consisting of: amino acids 11 437 of SEQ ID NO:2, amino acids 7-432 of SEQ ID NO:4, amino acids 39-457 of SEQ ID NO:5, amino acids 1524-1950 of SEQ ID NO:5, amino acids 3024-3449 of SEQ ID NO:5, amino acids 5103-5525 of SEQ ID NO:5, amino acids 35-454 of SEQ ID NO:6, amino acids 1522-1946 of SEQ ID NO: 6, and amino acids 32-450 of SEQ ID NO:7.

79. An isolated polypeptide according to claim 78, wherein said P-ketoacyl-synthase domain comprises an amino acid sequence selected from the group consisting of: amino acids 11-437 of SEQ ID NO:2, amino acids 7-432 of SEQ ID NO:4, amino acids 39-457 of SEQ ID NO:5, amino acids 1524-1950 of SEQ ID NO:5, amino acids 3024-3449 of SEQ ID NO:5, amino acids 5103-5525 of SEQ ID NO:5, amino acids 35-454 of SEQ ID NO:6, amino acids 1522-1946 of SEQ ID NO: 6, and amino acids 32-450 of SEQ ID NO:7.

80. An isolated polypeptide according to claim 77, wherein said epothilone synthase domain is an acyltransferase domain comprising an amino acid sequence substantially similar to an amino acid sequence selected from the group consisting of: amino acids 543 864 of SEQ ID NO:2, amino acids 539-859 of SEQ ID NO:4, amino acids 563-884 of SEQ ID NO:5, amino acids 2056-2377 of SEQ ID NO:5, amino acids 3555-3876 of SEQ ID NO:5, amino acids 5631-5951 of SEQ ID NO:5, amino acids 561-881 of SEQ ID NO:6, amino acids 2053-2373 of SEQ ID NO:6, and amino acids 556-877 of SEQ ID NO:7.

81. An isolated polypeptide according to claim 80, wherein said acyltransferase domain comprises an amino acid sequence selected from the group consisting of: amino WO 99/66028 - 82 - PCT/EP99/04171 acids 543-864 of SEQ ID NO:2, amino acids 539-859 of SEQ ID NO:4, amino acids 563 884 of SEQ ID NO:5, amino acids 2056-2377 of SEQ ID NO:5, amino acids 3555-3876 of SEQ ID NO:5, amino acids 5631-5951 of SEQ ID NO:5, amino acids 561-881 of SEQ ID NO:6, amino acids 2053-2373 of SEQ ID NO:6, and amino acids 556-877 of SEQ ID NO:7.

82. An isolated polypeptide according to claim 77, wherein said epothilone synthase domain is an enoyl reductase domain comprising an amino acid sequence substantially similar to an amino acid sequence selected from the group consisting of: amino acids 974 1273 of SEQ ID NO:2, amino acids 4433-4719 of SEQ ID NO:5, amino acids 6542-6837 of SEQ ID NO:5, and amino acids 1478-1790 of SEQ ID NO:7.

83. An isolated polypeptide according to claim 82, wherein said enoyl reductase domain comprises an amino acid sequence selected from the group consisting of: amino acids 974-1273 of SEQ ID NO:2, amino acids 4433-4719 of SEQ ID NO:5, amino acids 6542-6837 of SEQ ID NO:5, and amino acids 1478-1790 of SEQ ID NO:7.

84. An isolated polypeptide according to claim 77, wherein said epothilone synthase domain is an acyl carrier protein domain, wherein said polypeptide comprises an amino acid sequence substantially similar to an amino acid sequence selected from the group consisting of: amino acids 1314-1385 of SEQ ID NO:2, amino acids 1722-1792 of SEQ ID NO:4, amino acids 1434-1506 of SEQ ID NO:5, amino acids 2932-3005 of SEQ ID NO:5, amino acids 5010-5082 of SEQ ID NO:5, amino acids 7140-7211 of SEQ ID NO:5, amino acids 1430-1503 of SEQ ID NO:6, amino acids 3673-3745 of SEQ ID NO:6, and amino acids 2093-2164 of SEQ ID NO:7.

85. An isolated polypeptide according to claim 84, wherein said acyl carrier protein domain comprises an amino acid sequence selected from the group consisting of: amino acids 1314-1385 of SEQ ID NO:2, amino acids 1722-1792 of SEQ ID NO:4, amino acids 1434-1506 of SEQ ID NO:5, amino acids 2932-3005 of SEQ ID NO:5, amino acids 5010 5082 of SEQ ID NO:5, amino acids 7140-7211 of SEQ ID NO:5, amino acids 1430-1503 of SEQ ID NO:6, amino acids 3673-3745 of SEQ ID NO:6, and amino acids 2093-2164 of SEQ ID NO:7. WO 99/66028 PCT/EP99/04171 - 83

86. An isolated polypeptide according to claim 77, wherein said epothilone synthase domain is a dehydratase domain comprising an amino acid sequence substantially similar to an amino acid sequence selected from the group consisting of: amino acids 869-1037 of SEQ ID NO:4, amino acids 3886-4048 of SEQ ID NO:5, amino acids 5964-61.32 of SEQ ID NO:5, amino acids 2383-2551 of SEQ ID NO:6, and amino acids 887-1051 of SEQ ID NO:7.

87. An isolated polypeptide according to claim 86, wherein said dehydratase domain comprises an amino acid sequence selected from the group consisting of: amino acids 869-1037 of SEQ ID NO:4, amino acids 3886-4048 of SEQ ID NO:5, amino acids 5964-6132 of SEQ ID NO:5, amino acids 2383-2551 of SEQ ID NO:6, and amino acids 887 1051 of SEQ ID NO:7.

88. An isolated polypeptide according to claim 77, wherein said epothilone synthase domain is a p-ketoreductase domain comprising an amino acid sequence substantially similar to an amino acid sequence selected from the group consisting of: amino acids 1439 1684 of SEQ ID NO:4, amino acids 1147-1399 of SEQ ID NO:5, amino acids 2645-2895 of SEQ ID NO:5, amino acids 4729-4974 of SEQ ID NO:5, amino acids 6857-7101 of SEQ ID NO:5, amino acids 1143-1393 of SEQ ID NO:6, amino acids 3392-3636 of SEQ ID NO:6, and amino acids 1810-2055 of SEQ ID NO:7.

89. An isolated polypeptide according to claim 88, wherein said P-ketoreductase domain comprises an amino acid sequence selected from the group consisting of: amino acids 1439-1684 of SEQ ID NO:4, amino acids 1147-1399 of SEQ ID NO:5, amino acids 2645-2895 of SEQ ID NO:5, amino acids 4729-4974 of SEQ ID NO:5, amino acids 6857 7101 of SEQ ID NO:5, amino acids 1143-1393 of SEQ ID NO:6, amino acids 3392-3636 of SEQ ID NO:6, and amino acids 1810-2055 of SEQ ID NO:7.

90. An isolated polypeptide according to claim 77, wherein said epothilone synthase domain is a methyltransferase domain comprising an amino acid sequence substantially similar to amino acids 2671-3045 of SEQ ID NO:6.

91. An isolated polypeptide according to claim 90, wherein said methyltransf erase domain comprises amino acids 2671-3045 of SEQ ID NO:6. WO 99/66028 PCT/IEP99/04171 - 84

92. An isolated polypeptide according to claim 77, wherein said epothilone synthase domain is a thioesterase domain comprising an amino acid sequence substantially similar to amino acids 2165-2439 of SEQ ID NO:7.

93. An isolated polypeptide according to claim 77, wherein said thioesterase domain comprises amino acids 2165-2439 of SEQ ID NO:7.