Production of polyketides

Abstract

Recombinant host cells of the suborder Cystobacterineae containing recombinant expression vectors that encode heterologous PKS genes can produce polyketides synthesized by the PKS enzymes encoded on those vectors at high levels.

Description

CROSS-REFERENCE TO RELATED APPLICATIONS [0001] This application claims priority to U.S. patent application Ser. No. 09 / 443,501, filed 19 Nov. 1999; PCT patent application US99 / 27438, filed 19 Nov. 1999; and U.S. provisional application Ser. Nos. 60 / 130,560, filed 22 Apr. 1999; 60 / 122,620, filed 3 Mar. 1999; 60 / 119,386, filed 10 Feb. 1999; and 60 / 109,401, filed 20 Nov. 1998, each of which is incorporated herein by reference.REFERENCE TO GOVERNMENT FUNDING [0002] This invention was supported in part by SBIR grant 1R-43-CA79228-01. The U.S. government has certain rights in this invention.FIELD OF THE INVENTION [0003] The present invention provides recombinant methods and materials for producing polyketides in recombinant host cells. The recombinant host cells are from the suborder Cystobacterineae, preferably from the genera Myxococcus and Stigmatella that have been transformed with recombinant DNA expression vectors of the invention that encode modular or iterative polyketide synthase...

AI Technical Summary

Benefits of technology

[0117] In a preferred and illustrative embodiment, the recombinant host cell of the invention produces epothilone or an epothilone derivative. The epothilones (epothilone A, B, C, D, E, and F) and compounds structurally related thereto (epothilone derivatives) are potent cytotoxic agents specific for eukaryotic cells. These compounds have application as anti-fungals, cancer chemotherapeutics, and immunosuppressants. The epothilones are produced at very low levels in the naturally occurring Sorangium cellulosum cells in which they have been identified. Moreover, S. cellulosum is very slow growing, and fermentation of S. cellulosum strains is difficult and time consuming. One important benefit conferred by the present invention is the ability simply to produce an epothilone or epothilone derivative in a non-S. cellulosum host cell. Another advantage of the present invention is the ability to produce the epothilones at higher levels and in greater amounts in the recombinant host cells provided by the invention than possible in the naturally occurring epothilone producer cells. Yet another advantage is the ability to produce an epothilone derivative in a recombinant host cell. Thus, the present invention provides recombinant host cells that produce a desired epothilone or epothilone derivative. In a preferred embodiment, the host cell produces the epothilones at equal to or greater than 10 mg / L. In one embodiment, the invention provides host cells that produce one or more of the epothilones or epothilone derivatives at higher levels than produced in the naturally occurring organisms that produce epothilones. In another embodiment, the invention provides host cells that produce mixtures of epothilones that are less complex than the mixtures produced by naturally occurring host cells that produce epothilones.

[0118] In an especially preferred embodiment, the host cells of the invention produce less complex mixtures of epothilones than do naturally occurring cells that produce epothilones. Naturally occurring cells that produce epothilones typically produce a mixture of epothilones A, B, C, D, E, and F. The table below summarizes the epothilones produced in different illustratrive host cells of the invention. Cell TypeEpothilones ProducedEpothilones Not Produced1A, B, C, DE, F2A, CB, D, E, F3B, DA, C, E, F4BA, C, D, E, F5DA, B, C, E, F Thus, the recombinant host cells of the invention also include host cells that produce only one desired epothilone or epothilone derivative.

[0119] An analysis of the domains of the epothilone PKS suggests that the PKS enzyme catalyzes the production of epothilones G and H, which differ from one another in that epothilone G has a hydrogen at C-12 and epothilone H has a methyl group at that position. The variance at the C-12 position is predicted to arise from the ability of the corresponding AT domain (extender module 4) of the PKS to bind either malonyl CoA, leading to hydrogen, or methylmalonyl CoA, leading to methyl. However, epothilones G and H have not been observed in nature or in the recombinant host cells of the invention; instead, the products of the PKS appear to be epothilones C and D, which differ from epothilones G and H, respectively, by having a C-12 to C-13 double bond and lacking a C-13 hydroxyl substituent. Thus, the dehydration reaction that would form epothilones C and D from epothilones G and H may be carried out by the epothilone PKS itself or by another enzymatic activity that is present in the host cells in which the epothilones have been produced to date. Epothilones A and B are formed from epothilones C and D, respectively, by epoxidation of the C-12 to C-13 double bond by the epoK gene product. Epothilones E and F are formed from epothilones A and B, respectively, by hydroxylation of the C-21 methyl group.

[0120] Thus expression of the epothilone PKS genes and the epoK gene in a host cell of the invention leads to the production of epothilones A, B, C, and D. If the epoK gene is not present or is rendered inactive by mutation, then only epothilones C and D are produced. If the AT domain of extender module 4 is replaced by an AT domain specific for malonyl Co A, then epothilones A and C only are produced, and if there is no functional epoK gene, then only epothilone C is produced. If the AT domain of extender module 4 is replaced by an AT domain specific for methylmalonyl Co A, then epothilones B and D only are produced, and if there is no functional epoK gene, then only epothilone D is produced.

[0121] The epothilone PKS and modification enzyme genes were cloned from the epothilone producing strain, Sorangium cellulosum SMP44. Total DNA was prepared from this strain using the procedure described by Jaoua et al., 1992, Plasmid 28:157-165, incorporated herein by reference. A cosmid library was prepared from S. cellulosum genomic DNA in pSupercos (Stratagene). The entire PKS and modification enzyme gene cluster was isolated in four overlapping cosmid clones (deposited with the American Type Culture Collection (ATCC), Manassas, Va., USA, and assigned ATCC accession numbers as follows pKOS35-70.1A2 (ATCC 203782), pKOS35-70.4 (ATCC 203781), pKOS35-70.8A3 (ATCC 203783), and pKOS35-79.85 (ATCC 203780)) and the DNA sequence determined, as set forth in U.S. patent application Ser. No. 09 / 443,501, filed 19 Nov. 1999, incorporated herein by reference. DNA sequence analysis revealed a PKS gene cluster with a loading module and nine extender modules. Downstream of the PKS sequence is an open reading frame (ORF), designated epoK, that shows strong homology to cytochrome P450 oxidase-genes and encodes the epothilone epoxidase modification enzyme.

[0122] The PKS genes are organized in 6 ORFs. At the polypeptide level, the loading module and extender modules 1 (an NRPS), 2, and 9 appear on individual polypeptides; their corresponding genes are designated epoA, epoB, epoC and epoF respectively. Modules 3, 4, 5, and 6 are contained on a single polypetide whose gene is designated epoD, and modules 7 and 8 are on another polypeptide whose gene is designated epoE. It is clear from the spacing between ORFs that epoC, epoD, epoE and epoF constitute an operon. The epoA, epoB, and epoK gene may be also part of the large operon, but there are spaces of approximately 100 bp between epoB and epoC and 115 bp between epoF and epoK which could contain a promoter. The epothilone PKS gene cluster is shown schematically below. Immediately downstream of epoK, the P450 epoxidase gene, is ORF1, which encodes a polypeptide that appears to include membrane spanning domains and may be involved in epothilone transport. This ORF is followed by a number of ORFs that include genes that may encode proteins involved in transport and regulation.

Problems solved by technology

While such methods are valuable and highly useful, certain polyketides are expressed only at very low levels or are toxic to the heterologous host cell employed.

However, the production of epothilone was only about 50 to 100 μg / L and appeared to have a deleterious effect on the producer cells.

Despite the success of these efforts, the chemical synthesis of the epothilones is tedious, time-consuming, and expensive.

Indeed, the methods have been characterized as impractical for the full-scale pharmaceutical development of an epothilone.

Images