NRPS-PKS GENE CLUSTER AND ITS MANIPULATION AND UTILITY

The present invention relates to the cloning and sequencing of the gene cluster encoding the biosynthetic machinery for the synthesis of the polyketide macrolactam BE- 14106, which includes a both a non-ribosomal peptide synthetase (NRPS) adenylation domain and a modular polyketide biosynthetic enzyme or enzyme complex (PKS; polyketide synthase enzyme or enzyme complex). The biosynthesis machinery thus comprises a hybrid NRPS-PKS enzyme system. The invention accordingly relates to novel genes and nucleic acid molecules encoding the biosynthetic machinery for the synthesis of the macrolactam BE-14106, including a modular NRP S -polyketide biosynthetic enzyme or enzyme system involved in BE-14106 biosynthesis and the biosynthetic machinery including the modular NRPS-polyketide synthase enzyme system or complex itself (as well as components thereof). The invention further relates to the use of these genes, nucleic acid molecules, the machinery, enzymes and enzyme systems or complexes thereof both in facilitating BE-14106 biosynthesis and in the synthesis of BE-14106 derivatives and novel macrolactam structures.

Polyketides or polyketide-based or related structures are, or form the basis of, natural products synthesized by bacteria, fungi, plants, and animals, many of which have applied potential as pharmaceuticals or as agricultural or veterinary products, e.g. as antibiotics, antifungals, cytostatics, anticholesterolemics, antiparasitics, coccidiostatics, animal growth promoters and natural insecticides.

The Gram-positive bacteria Streptomyces are the main producers of polyketides and polyketide-based molecules, and the genetics and biochemistry of polyketide biosynthesis in these organisms are relatively well characterized (McDaniel R, et al; Chem Rev. 2005 Feb;105(2):543-58.)

Other producers include other actinomycetes. A range of different polyketide-based (or polyketide-related) molecules are known, of which macrolactams represent one class. The biosynthetic gene clusters for synthesis of the macrolactams vicenistatin and salinilactam have been reported Ogasawara Y. et al\ Chem Biol. 2004 Jan; 1 l(l):79-98, and Udwary et al\ Proc Natl Acad Sci USA 2007 Jun 19; 104(25):10376-81, respectively.

BE-14106 (alternative name GT-32A) is a macrolactam antibiotic having a chemical structure as set out in Figure 1. It has been isolated from a strain of Streptomyces spheroides and has been shown to have cytotoxic effects on leukemia cell lines, as well as antimicrobial activity against a range of tested organisms, antiproliferative activity against a H-ras transformed BALB3T3 cell line and inhibitory activity against mixed lymphocyte reaction (JP4001179, Kojiri et al 1992 Journal of Antibiotics, 868-74, Takahashi et al 1997, Journal of Antibiotics 186-8). An 8-deoxy analogue (GT-32B) has also been isolated from an unspecified Streptomyces species and this was shown to share many of the activities of BE- 14106 (Takahashi et al, supra).

Macrolactam compounds such as BE-14106 can be formed via activation and priming of the PKS system with an activated amino acid and extension of the amino acid residue (aminoacyl chain) by repeated condensations of simple carboxylic acids by polyketide synthases (PKS) in a manner similar to fatty acid biosynthesis. Thus, unlike the case with a simply polyketide chain where the "starter unit" is a carboxylic acid residue, in this case, the starter unit for the PKS is an aminoacyl intermediate synthesized from an amino acid and an acyl chain. PKSs can be organised as iterative PKSs which re-use domains in a cyclic fashion or as modular (Type I) PKSs which contain a sequence of separate modules (or repeated units) and do not re-use domains. Each module is responsible for one condensation cycle in the synthesis of the polyketide chain and contains various enzyme domains. In the case of BE-14106 the "polyketide" chain is strictly speaking a hybrid amino acid- polyketide chain, or an aminoacyl chain, but is referred to herein as a "polyketide chain". Thus, besides domains for the condensation of the next carboxylic acid onto the growing polyketide chain, catalysed by the β-ketoacyl synthase (KS) domain, modules of type I PKS may contain domains with β-ketoreductase (KR), dehydratase (DH) or enoyl reductase (ER) activities, which determine the reduced state of incorporated extender units. The acyltransferase (AT) and acyl carrier protein (ACP) domains present in each module are responsible for the choice of extender unit and retention of the growing polyketide chain on the PKS, respectively. Upon completion of synthesis, the polyketide chain is released from PKSs via action of a thioesterase (TE), that is probably also involved in cyclization

of the final product. Thus, PKSs type I represent an assembly line for polyketide biosynthesis, that can be manipulated by changing the number of modules, their specificities towards carboxylic acids, and by inactivating or inserting domains with reductive activities (Weissman and Leadlay, Nat. Rev. Microbiol. 2005 Dec;3(12):925-36.). After the polyketide moiety is synthesized and cyclized to form a macrolactone (or macrolactam) ring, it may be modified via hydroxylation, glycosylation, methylation and/or acylation. These modifications may be important for the biological activities of certain polyketide-based product. As will be described in more detail below, in work leading up to the present invention the genes encoding the BE- 14106 NRPS-PKS enzyme system (the BE- 14016 "gene cluster") have been cloned and sequenced and it has been determined that the BE- 14106 NRPS-PKS enzyme system contains several type I PKSs, each of which is organized in the modular way, and is made up of repeated units (modules).

The genes for polyketide biosynthesis in Streptomyces are generally organized in clusters, and a number of such clusters have already been identified, responsible for the synthesis of various natural products. The molecular cloning and complete DNA sequencing of several macrolide antibiotic gene clusters of Streptomyces has been described, including those for avermectin, pikromycin and rapamycin (Ikeda H., Omura S. (2002). Biosynthesis, Regulation , and Genetics of Macrolide Production. In: Macrolide Antibiotics: Chemistry, Biology and Practice, 2 ^nd Ed. (ed. S. Omura), pp. 286-326, Academic Press, New York.) As mentioned above, gene clusters for the biosystems of certain macrolactam antibiotics have also been reported.

As noted above and described below, the present invention is based on the identification, cloning and sequencing of a novel gene cluster for biosynthesis of BE-14106 which has not heretofore been available. Analysis of the cloned genes has further allowed the elucidation of the biosynthetic pathway for BE-14106. Accordingly it is now proposed that the normal process of synthesis of BE-14106 is initiated through the synthesis of a starter unit (C17-C25), where an acyl moiety is synthesised from 1 proprionate and 2 acetate units. Synthesis of the starter unit continues with the activation of a glycine molecule by an NRPS adenylation domain and loading of the activated glycine on to a peptidyl carrier protein. The oxidative

- A -

deamination of glycine releases ammonium, which makes a nucleophilic attack on the C- 17 carbonyl to form a C- 17 imino group, which is subsequently reduced to an amino group. Release of the aminoacyl chain from the peptidyl carrier protein, results in the formation of a carboxylic acid, which is then adenylated and ligated to coenzyme A (CoA). The resultant activated aminoacyl-CoA is transferred to the ACP domain of a PKS by an acyltransferase and extended and modified by the sequential action of the enzymes in the PKS system as described in more detail below. The β-ketoacyl synthase (KS) enzyme domain in each module catalyses the condensation of the appropriate carboxylic acid (e.g. acetate or propionate) as determined by the acyltransferase (AT) module. Enzyme domains with β- ketoreductase (KR) or dehydratase (DH) activity determine the reduced state of incorporated extender units.

The C20-C25 hydrocarbon side chain of BE- 14106 is comprised from part of the starter unit and results from the cyclisation of the macrolactam ring. Finally, further modification of the macrolactam ring occurs via hydroxylation.

The BE- 14106 biosynthetic gene cluster also encodes or includes various regulatory elements and proteins for the transport of the synthesized molecules.

Since the chemical synthesis of compounds such as this is highly complex, a biosynthetic route in practice needs to be used and accordingly the isolation or purification of the compounds from appropriate hosts is desirable. As has been recognised in the art, this affords the opportunity of manipulating genes of the PKS gene cluster in order to change the biosynthesis and thereby result in the synthesis of new or modified polyketide or polyketide-based compounds. Whilst the modification of a number of PKS gene clusters has been described resulting in the synthesis of various new compounds, there remains a need and desire to increase the repertoire of available compounds, especially antibiotics, and/or to improve upon the properties (e.g. efficacy, toxicity, solubility in water etc.) of existing drugs. The present invention is directed to these aims, and is based on the cloning and DNA sequencing of the BE-14106 biosynthetic gene cluster. This provides the first sequence for these antibiotic biosynthetic genes, as well as a tool for genetic manipulation in order to modify the expression levels or properties of BE-14106 and/or the producing organism, or to obtain novel potentially useful compounds. In

this respect, whilst the antibiotic BE- 14106 is known, in the background of a plurality of polyketide-based molecules synthesised in Streptomyces and corresponding plurality of biosynthetic gene clusters, it was not a straightforward matter to identify and clone the correct gene cluster for BE-14106; a considerable effort and ingenuity in terms of sequence analysis and design or selection of probes was required.

The present inventors have isolated and purified BE-14106 from a previously unknown source, bacterial isolate MP28-13, believed to be a novel strain of Streptomyces (deposited under the deposit number DSM21069, on 25 January 2008, at the Deutsche Sammlung von Mikroorganismem und Zellkulturen GmbH

(DSMZ)) which was isolated from shallow water sediment in the Trondheim fjord, Norway. The isolation of this novel microorganism has enabled the inventors to clone and sequence the entire BE-14106 biosynthetic gene cluster. This cluster contains 22 genes that encode proteins presumed to be involved in the biosynthesis of the BE- 14106 molecule (see Table 1 ).

To perform this cloning, specially designed oligonucleotide primers representing sequences encoding parts of the ketosynthase (KS) domain were used for amplification of KS domain coding regions from isolate MP28-13. Once amplified sequences had been obtained and characterised, based on complex and extensive bioinformatic analysis, one of the sequences was chosen as a probe. This probe was used for screening the genomic library that was constructed for MP28-13. The cosmids that were identified in this manner were analysed and sequenced to provide the whole biosynthetic cluster. The sequence has been fully annotated and a two-part pathway for BE-14106 biosynthesis has been elucidated, as set out in Figures 2A and 2B. The first part of the pathway for biosynthesis of the starter aminoacyl unit is shown in Figure 2A and the second part, elongation of the aminoacyl chain, its cyclisation resulting in the formation of macrolactam ring and post PKS modification, is depicted in Figure 2B. Thus, it is proposed that the BE- 14106 biosynthetic gene cluster encodes a first enzyme system or complex comprising PKS and other enzymes or proteins for synthesis of the aminoacyl chain and an additional PKS enzyme system or enzyme complex for elongation of said

aminoacyl chain, as well as an enzyme for post PKS modification of the molecule, proteins for regulation of the pathway, and resistance/efflux proteins.

Based on the knowledge of the sequence, a method for genetic manipulation of Streptomyces species MP28-13 was developed, hi this way it was possible to show that the novel sequence was indeed responsible for BE-14106 biosynthesis.

Furthermore, as will be described in more detail below, manipulation of functional DNA sequences within the novel biosynthetic gene cluster which has been identified, can lead to the synthesis of novel molecular structures, e.g. BE- 14106 derivatives or analogues with the altered, e.g. improved function or properties. As such the BE-14106 gene cluster can be manipulated to obtain not only beneficial new BE-14106 derivatives or analogues, but also to improve and facilitate the biosynthetic production process (for example to improve yield, or production conditions, or to expand the range of available host cells) or more preferably to provide novel compounds with new activities and/or properties. The complete coding sequence for (i.e. the complete nucleotide sequence encoding) the BE-14106 biosynthetic gene cluster is shown in SEQ ID No. 1. This has been shown to contain a number of genes or ORFs, which encode the various proteins and polypeptides which are responsible for the activities that are required for BE-14106 biosynthesis. The biosynthetic gene cluster contains genes and ORFs that are believed to encode all of the proteins and polypeptides that are required for normal BE-14106 biosynthesis. However, not all of the encoded proteins and polypeptides have yet been ascribed a role in the biosynthesis and so it may be that not all of the encoded proteins or polypeptides of the cluster are essential for BE-14106 biosynthesis. The various genes and ORFs may encode enzymes that catalyse one or more biochemical reactions, or proteins that do not have catalytic activity but instead are involved in other processes such as the regulation of the process of BE-14106 synthesis, or BE- 14106 transport, for example.

Several of the enzymes are polyketide synthases (PKSs), and it is possible that a number of these PKSs may physically associate to form an enzyme complex, although this has not yet been established. Such a group or set of enzymes is referred to herein as a polyketide biosynthetic enzyme system or complex or PKS

enzyme system or complex, although not all of the enzymes/proteins in the system/complex need be actual polyketide synthases, i.e. have polyketide synthase activity; they may have other activities or functional roles in BE-14106 synthesis. For example, a discrete adenylation domain of non-ribosomal peptide synthetase (NRPS) (BecL), along with some other accessory proteins (e.g. BecJ, BecS, BecU) encoded by the cluster are involved in the synthesis of the starter unit for biosynthesis of BE-14106 by activating an amino acid (presumed to be a glycine) and its loading on one of several BE-14106 PKS modules for further elongation. Other proteins, such as BecO, perform hydroxylation of the macrolactam ring at C- 8. A group or set of enzymes comprising such a NRPS domain and PKS enzymes may be referred to as a hybrid NRPS-PKS enzyme system or enzyme complex. The group of proteins and polypeptides encoded by the gene cluster as a whole are collectively referred to as the biosynthetic machinery for the biosynthesis of BE- 14106. Thus in one aspect, the present invention provides a nucleic acid molecule comprising:

(a) a nucleotide sequence as shown in SEQ ID No. 1 ; or

(b) a nucleotide sequence which is the complement of SEQ ID No. 1 ; or

(c) a nucleotide sequence which is degenerate with SEQ ID No. 1 ; or (d) a nucleotide sequence having at least 85% sequence identity (preferably at least 87%, 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98% or 99% sequence identity) with SEQ ID No. 1; or

(e) a part of any one of (a) to (d) wherein said part preferably comprises a sequence which corresponds to a BE-14106 biosynthetic gene or open reading frame (ORF), or is complementary thereto or degenerate therewith.

More particularly such a nucleic acid molecule encodes (or comprises a nucleotide sequence encoding) one or more polypeptides, or comprises one or more genetic elements, having functional activity in the synthesis of a polyketide or macrolactam molecule, more particularly the synthesis of BE- 1406 or a derivative thereof, or BE- 1406 related molecule, or which is the complement of such a nucleic acid molecule. Such functional activity may be enzymatic activity, e.g. an activity involved in the synthesis or transport or transfer of a polyketide or macrolactam

molecule (this can be polyketide chain or macrolactam ring synthesis or any step contributory thereto, or macrolactam ring or polyketide chain modification etc) and/or it may be a regulatory activity, e.g. regulation of the expression of the genes (e.g. a transcriptional regulator) or proteins involved in the synthesis, or regulation of the synthetic process, and/or it may be a "transporter activity". Thus, included generally are also transport proteins involved in the transfer or transport of polyketide or macrolactam moieties e.g. in the transport or efflux of the synthesised molecule within or out of the cell.

Whilst nucleotide sequences encoding a desired product are preferred according to the invention, also encompassed are nucleotide sequences comprising functional genetic elements such as promoters, promoter-operator regions, enhancers, other regulatory sequences etc. Thus, the nucleic acid molecule of the invention need not comprise the entire PKS gene cluster but may comprise a portion or part of it e.g. a part encoding a polypeptide having a particular function or a regulatory sequence. This may comprise one or more genes, and/or regulatory sequences, and/or one or more modules or, enzymatic domains, or non-coding or coding functional genetic elements (e.g. elements controlling gene expression, transcription, translation etc). Generally speaking, a nucleic acid molecule of the invention will comprise a number of different genes and/or regulatory sequences leading to the synthesis of a polyketide-based or macrolactam molecule, e.g. a BE- 14106 derivative or a modified BE- 14106 molecule.

A "BE-14106 biosynthetic gene or ORF" is defined further below, but briefly in the context of section (e) above means a gene or ORF which encodes a protein or polypeptide that is functional in the biosynthetic process of BE- 14106 or a BE- 14106 derivative or analogue or BE-14106-related molecule. As noted above, this could be an enzyme that is involved in the activation of the starting amino acid, transfer of the activated amino acid/aminoacyl chain to a PKS enzyme, generation of the polyketide chain or modification thereof, or a protein that is required for regulation, or for transport or transfer of the molecule at any stage of its biosynthesis.

A nucleic acid molecule of the invention may be an isolated nucleic acid molecule (in other words isolated or separated from the components with which it is

normally found in nature) or it may be a recombinant or a synthetic nucleic acid molecule.

As discussed elsewhere herein, the BE- 14106 biosynthetic gene cluster is a large nucleic acid molecule which contains the various genetic elements or different genes or ORFs that encode the proteins or peptides that are required for the biosynthesis of the BE- 14106 molecule or a BE-14106 derivative or analogue or BE-14106-related molecule. Each BE-14106 biosynthetic gene or ORF encodes a single polypeptide chain (which can alternatively be described as a protein) that has or is believed to have a function in the biosynthesis of the BE-14106 molecule or a BE- 14106 derivative or analogue or BE- 14106-related molecule. 22 such genes or ORFs have been identified (see Table 1). As shown in Figures 2 A and 2B, 14 of these are ascribed a direct role in the biosynthesis of BE-14106. As explained further below, certain others are also believed or proposed to play a role in BE- 14106 biosynthesis. Thus for example, becH and Mare believed to encode regulators, BecL is believed to be involved in the glycine activation, BecU is believed to mediate the protein interaction between the ACP of BecC and PCP BecS, BecN is believed to be involved in efflux and/or resistance, BecP is thought to assist the cyclisation of the macrolactam ring and BecQ in the release of the initiating aminoacyl chain from the BecC-BecU-BecS complex. Certain of the proteins have enzymatic activity and can thus be defined as being enzymes. Various of these enzymes can be described as polyketide synthases (PKSs). Such enzymes contain one or more than one module, and each module can contain from one to six, preferably two, three, four or five enzyme domains, each of which is responsible for a different activity in the biosynthesis of the BE-14106 molecule or a BE- 14106 derivative or analogue or BE- 14106-related molecule. As such, in these PKSs multiple active sites can be present in a single polypeptide or enzyme.

For example, the enzyme BecB is a PKS and has three modules; module 1 (the "loading module" in Figure 2B) has a single active site or domain (ACP), and each of modules 2 and 3 (Modules 1 and 2 in Figure 2B) have five active sites or domains having KS, AT, DH, KR and ACP activities. Other PKSs encoded by the gene cluster are BecA, BecC, BecD, BecG, BecF and BecE. Such PKSs can contain

numerous domains, each possessing catalytic activity to extend and/or alter the structure of the polyketide. The polyketide passes along the protein such that the different acitivities are carried out sequentially on the growing polyketide chain. As discussed above, various of the PKSs encoded by the gene cluster may associate to form a biosynthetic enzyme complex.

The nucleic acid molecule of the invention encodes (or comprises a nucleotide sequence encoding) some, or more preferably all, of the polypeptides or proteins that are involved in the biosynthesis of the molecule BE- 14106 or a BE- 14106 derivative or analogue or BE-14106-related molecule. For example, the nucleic acid molecule may contain each of the 22 genes or ORFs and thus encode each of the proteins that are involved in the biosynthesis of the molecule BE-14106 as set out in Table 1, or it may comprise a portion or part of the nucleotide sequence of SEQ ID No. 1 e.g. a sequence encoding a single protein or polypeptide encoded by a single gene or ORF within the BE-14106 biosynthetic gene cluster. Parts comprising e.g. at least 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, or up to 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, (e.g.1-21, 2-20, 3-19, 4-18, 5-17, 6-16, 7-15, 8-14, 9-13, 10-12 ) genes or ORFs are contemplated. Preferably the nucleic acid molecule of the invention encodes all of the proteins that are involved in the biosynthesis of the molecule BE-14106 as set out in Table 1. Alternatively it may comprise all of the ORFs/genes as set out in Table 1 except any one or more of becR and ORF6. Since Table 1 sets out all of the genes or ORFs which have been characterised, a nucleic acid molecule encoding all of the proteins that are involved in the biosynthesis of the molecule BE-14106 as set out in Table 1 can be defined as a sequence which comprises the BE-14106 biosynthetic gene cluster. The nucleic acid molecule of the invention thus encodes one or more polypeptides involved in the biosynthesis of or having functional activity in the synthesis of BE-14106 or a BE-14106 derivative or analogue or BE-14106-related molecule. Alternatively it may encode one or more functionally equivalent variants or functional equivalents thereof. As defined above, the nucleic acid molecules of the invention may comprise functionally equivalent variants of SEQ ID No. 1 and such variants may include parts, degenerate sequences, or homologues defined by a % sequence identity to SEQ ID No. 1. Such functionally equivalent variants encode

proteins/polypeptides having functional activity as defined above. Such functional activity may be enzymatic activity e.g. an activity involved in the synthesis or transport or transfer of a polyketide moiety or a macrolactam molecule (this can be chain or ring synthesis or any step contributory thereto, or modification etc at any stage of biosynthesis, e.g. BecA, BecU, BecB, BecJ, BecK, BecS, BecO, BecD, BecG, BecF, BecE, BecT, BecQ, BecP, BecC, Bed, BecL) and/or it may be a regulatory activity, e.g. regulation of the expression of the genes or proteins involved in the synthesis, or regulation of the synthetic process e.g. BecH, BecM, and/or it may be a "transporter activity" or resistance e.g. BecN. Thus, included generally are also transport proteins involved in the transfer, transport or efflux of the synthesised molecule within or out of the cell. Also contemplated are sequences that encode one or more modules or enzymatic domains.

Such molecules may be at least 200 bases in length, more preferably at least, 300, 500, 600, 700, 800, 900, 1000, 1500, 2000, 3000, 5000, 10000, 15000, 20000, 30000, or 50000 bases. Representative fragment lengths thus include fragments that are lOObp to 18000bp in length, e.g. 100-3000bp, 200-2500bp, 2000-8000bp, 3000- 5000bp, 4000-17000bp, 7000-12000bp or 8000-1 lOOObp in length. As mentioned above, a number of genes and ORFs have been identified within SEQ ID NO:1 and parts or fragments which comprise such genes or ORFs represent preferred "parts" or fragments of SEQ ID NO: 1. These are tabulated in Table 1 below:

Table 1

In the above Table, "C" indicates that the protein is encoded by the complement strand

The sequences set out above thus represent BE- 14106 biosynthetic genes or ORFs. In other words, such genes/ORFs are found within the BE- 14106 biosynthetic gene cluster and encode proteins or polypeptides which have or are proposed to have a role in the biosynthesis of BE- 14106 in Streptomyces. The term "BE-14106 biosynthetic gene" or "BE-14106 biosynthetic ORF" also includes genes and ORFs which encode proteins that share activity or function with the above proteins, and for example share high levels of sequence identity, as discussed elsewhere herein. They can alternatively be described as "functionally equivalent variants" or "functional equivalents". In general the term "gene" includes the ORF which encodes the protein, together with any regulatory sequences such as promoters,' whereas the term'ORF" refers only to the part of the gene which is responsible for encoding the protein.

As referred to herein "functionally equivalent variants" or "functional equivalents" retain at least one function of the entity to which they are related (or from which they are derived), e.g. encode a protein with substantially the same properties, or exhibit substantially the same regulatory or other functional properties or activities. The properties or activities can be tested for using standard techniques that are known in the art.

Whilst nucleotide sequences encoding a desired product (e.g. ORFs and genes) are preferred according to the invention, also encompassed are nucleotide sequences comprising functional genetic elements such as promoters, promoter- operator regions, enhancers, other regulatory sequences etc. Thus, the nucleic acid molecule of the invention need not comprise the entire gene cluster but may comprise a portion or part of it e.g. a part encoding a polypeptide having a particular function or a regulatory sequence. This preferably comprises one or more genes, and/or regulatory sequences. Also contemplated are sequences that in general will be smaller than this and encode one or more modules or, enzymatic domains, or non-coding or coding functional genetic elements (e.g. elements controlling gene expression, transcription, translation etc). The invention thus extends to a nucleic acid molecule comprising a nucleotide sequence selected from SEQ ID NO:2, 4, 6, 8, 10, 12, 14, 16, 18, 20, 22, 24, 26, 28, 30, 32, 34, 36, 38, 40, 42 and 44 (as identified by reference to nucleotide

start and end positions in SEQ ID No. 1 as shown in Table 1) or a nucleotide sequence which is complementary thereto or degenerate therewith.

Also provided are nucleic acid molecules comprising nucleotide sequences which exhibit at least 80% (preferably at least 85%, 87%, 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98% or 99%) sequence identity with any one of SEQ ID

NO:2, 4, 6, 8, 10, 12, 14, 16, 18, 20, 22, 24, 26, 28, 30, 32, 34, 36, 38, 40, 42 and 44 or a nucleotide sequence which is complementary thereto or degenerate therewith.

The invention further relates to a nucleic acid molecule comprising a nucleotide sequence encoding one or more amino acid sequences selected from SEQ ID NO:3, 5, 7, 9, 11, 13, 15, 17, 19, 21, 23, 25, 27, 29, 31, 33, 35, 37, 39, 41, 43 or 45 or a nucleotide sequence which is complementary thereto or degenerate therewith.

Also provided are nucleic acid molecules comprising nucleotide sequences which encode one or more amino acid sequences (i.e. polypeptides) which exhibit at least 80% (preferably at least 85%, 87%, 90%, 91 %, 92%, 93%, 94%, 95%, 96%, 97%, 98% or 99%) sequence identity with any one of SEQ ID NO:3, 5, 7, 9, 11, 13, 15, 17, 19, 21, 23, 25, 27, 29, 31, 33, 35, 37, 39, 41, 43 or 45 or a nucleotide sequence which is complementary thereto or degenerate therewith.

In each case, the nucleic acid molecule is preferably a BE-14106 biosynthetic gene or ORF, as defined herein.

Nucleotide or amino acid sequence identity may be assessed by any convenient method. However, for determining the degree of sequence identity between sequences, computer programs that make multiple alignments of sequences are useful, for instance Clustal W (Thompson, J. D et al., 1994, "CLUSTAL W: Improving the sensitivity of progressive multiple sequence alignment through sequence weighting, position- specific gap penalties and weight matrix choice". Nucleic Acids Res 22: 4673-4680). Programs that compare and align pairs of sequences, like ALIGN (E. Myers and W. Miller, 1988, "Optical Alignments in Linear Space", CABIOS 4: 11-17), FASTA (W.R. Pearson and DJ. Lipman, 1988, "Improved tools for biological sequence analysis", PNAS 85:2444-2448, and W.R. Pearson, 1990, "Rapid and sensitive sequence comparison with FASTP and FASTA"Methods in Enzymology 183:63-98) and gapped BLAST (Altschul, S.F., et

al., 1997, "Gapped BLAST and PSI-BLAST: a new generation of protein database search programs". Nucleic Acids Res. 25: 3389-3402) are also useful for this purpose. Furthermore, the DaIi server at the European Bioinformatics institute offers structure-based alignments of protein sequences (Holm, 1993, J. of MoI. Biology, 233: 123-38; Holm, 1995, Trends in Biochemical Sciences, 20: 478-480; Holm, 1998, Nucleic Acid Research, 26: 316-9).

For example, nucleotide sequence identity may be determined using the BestFit program of the Genetics Computer Group (GCG) Version 10 Software package from the University of Wisconsin. The program uses the local homology algorithm of Smith and Waterman with the default values: Gap creation penalty = 50, Gap extension penalty = 3, Average match = 10,000, Average mismatch = - 9.000.

Nucleotide sequences according to the invention may exhibit at least 85%, 87%, 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98% or 99% sequence identity with SEQ ID NO: 1 and such sequences will preferably encode or are complementary to a sequence which encodes some or all of the proteins that are involved in the biosynthesis of the molecule BE-14106. Nucleotide sequences meeting the % sequence identity criteria defined herein may be regarded as "substantially identical" sequences. A further aspect of the invention provides polypeptides encoded by a nucleic acid molecule of the invention as defined herein.

As discussed above, SEQ ID NO:1 encodes several proteins or polypeptides and as such this aspect of the invention provides a polypeptide comprising:

(a) all or part of an amino acid sequence as shown in any one or more of SEQ ID Nos. 3, 5, 7, 9, 11, 13, 15, 17, 19, 21, 23, 25, 27, 29, 31, 33, 35, 37, 39, 41, 43 or 45; or

(b) all or part of an amino acid sequence which has at least 80% sequence identity with any one or more of SEQ ID Nos. 3, 5, 7, 9, 11, 13, 15, 17, 19, 21, 23, 25, 27, 29, 31, 33, 35, 37, 39, 41, 43 or 45. In particular the amino acid sequence may exhibit at least 80%, 85%, 87%,

90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98% or 99% identity with the polypeptide of any one of SEQ ID Nos. 3, 5, 7, 9, 11, 13, 15, 17, 19, 21, 23, 25, 27,

29, 31, 33, 35, 37, 39, 41, 43 or 45. Alternatively, the amino acid sequence may exhibit at least 80%, 85%, 87%, 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98% or 99% similarity with the polypeptide of any one of SEQ ID Nos. 3, 5, 7, 9, 11, 13, 15, 17, 19, 21, 23, 25, 27, 29, 31, 33, 35, 37, 39, 41, 43 or 45. Amino acid (polypeptide) sequences meeting the % sequence identity or similarity criteria herein are regarded as "substantially identical".

The polypeptide of the invention may be an isolated, purified or synthesized polypeptide. The term "polypeptide" is used herein to include any amino acid sequence of two or more amino acids i.e. both short peptides and longer lengths (i.e. polypeptides or proteins) are included.

Programs for determining amino acid sequence identity are mentioned above, for example amino acid sequence identity or similarity may be determined using the BestFit program of the Genetics Computer Group (GCG) Version 10 Software package from the University of Wisconsin. The program uses the local homology algorithm of Smith and Waterman with the default values: Gap creation penalty - 8, Gap extension penalty = 2, Average match = 2.912, Average mismatch = -2.003. A "part" of the amino acid sequence of any one of SEQ ID Nos. 3, 5, 7, 9, 11, 13, 15, 17, 19, 21, 23, 25, 27, 29, 31, 33, 35, 37, 39, 41, 43 or 45 (or of a "substantially identical" sequence as defined above) may comprise at least 20 contiguous amino acids, preferably at least 30, 40, 50, 70, 100, 150, 200, 300, 400, 500, 1,000, 2,000, 5,000 or 10,000 contiguous amino acids.

The polypeptide, and preferably also the part thereof, is functionally active according to the definitions given above, e.g. is enzymatically active or has a regulatory or transport functional activity in the biosynthesis of BE-14106 or a derivative of BE- 14106 or a modified version thereof. The part may thus correspond to or comprise an active site or domain or a module as discussed above.

The nucleotide and polypeptide sequences of the invention have been characterised and various functional regions within them have been identified. Such functional regions form separate aspects of the invention. "Parts" as defined herein thus preferably correspond to at least one module or enzymatic domain, or non- coding or coding functional genetic element. Table 2 below shows the functional

regions identified with the translation products of the ORPs identified in SEQ ID No. 1 which encode PKS enzymes.

Table 2

Domain boundaries in BE-14106 PKS proteins

BecA (SEO ID No. 5)

Molecule: BecA, 5582 aas Protein

Molecule Features :

Type Start End Name Description

REGION 17 436 KSq KS-like domain, loading modu

REGION 543 858 ATO AT domain, loading module, propionate

REGION 933 1004 ACPO ACP domain, loading module

REGION 1027 1450 KSl KS domain, module 1

REGION 1561 1879 ATI AT domain, acetate, module 1

REGION 1892 2095 DHl DH domain, module 1

REGION 2415 2662 KRl KR domain, module 1

REGION 2698 2771 ACPI ACP domain, module 1

REGION 2795 3220 KS2 KS domain, module 2

REGION 3331 3649 AT2 AT domain, acetate, module 2

REGION 3663 3868 DH2 DH2 domain, module 2

REGION 4189 4435 KR2 KR domain, module 2

REGION 4472 4545 ACP2 ACP domain, module 2

REGION 4572 4995 KSx KS domain, module 3

REGION 5103 5416 ATx AT domain, acetate, module 3

BecB (SEQ ID No .13)

Molecule : BecB, 3527 aas Protein

Molecule Features :

Type Start End Name Description

REGION 10 90 ACPl ACP domain, , module 1

REGION 112 537 KS2 KS domain, module 2

REGION 645 961 AT2 AT domain, acetate, m module 2

REGION 975 1169 DH2 DH domain, module 2

REGION 1424 1672 KR2 KR domain, module 2

REGION 1709 1782 ACP2 ACP domain, , module 2

REGION 1801 2225 KS3 KS domain, module 3

REGION 2332 2651 AT3 AT domain, propionate module 3

REGION 2665 2865 DH3 DH domain, module 3

REGION 3132 3374 KR3 KR domain, module 3

REGION 3411 3485 ACP3 ACP domain, module 3

BecC rSEO ID No.9)

Molecule: BecC, 694 aas Protein Molecule Features:

Type Start End Name Description

REGION 346 600 KR KR domain, module 3 REGION 615 689 ACP ACP domain, module 3

BecD (SEO ID No.29)

Molecule: BecD, 3372 aas Protein Molecule Features:

Type Start End Name Description

REGION 38 447 KS4 KS domain, module 4

REGION 561 861 AT4 AT domain, acetate, module 4

REGION 875 1070 DH4 DH domain, module 4

REGION 1321 1570 KR4 KR domain, module 4

REGION 1586 1659 ACP4 ACP domain, module 4

REGION 1678 2089 KS5 KS domain, module 5

REGION 2173 2466 AT5 AT domain, acetate, module 5

REGION 2480 2672 DH5 DH domain, module 5

REGION 2934 3178 KR5 KR domain, module 5

REGION 3217 3290 ACP5 -ACP domain, module 5

BecE (SEO ID No.37)

Molecule: BecE, 1631 aas Protein Molecule Features:

Type Start End Name Description

REGION 34 445 KS6 KS domain, module 6

REGION 529 822 AT6 AT domain, acetate, module 6

REGION 836 984 DH6i DH domain, module 6, C-term deletion, probably inactive

REGION 1201 1448 KR6 KR domain, module 6

REGION 1476 1550 ACP6 ACP domain , module 6

BecF (SEO IDNo. 35)

Molecule: BecF, 3377 aas Protein

Molecule Features:

Type Start End Name Description

REGION 37 447 KS7 KS domain, module 7

REGION 558 864 AT7 AT domain, propionate, module

REGION 878 1079 DH7 DH domain, module 7

REGION 1341 1585 KR7 KR domain, module 7

REGION 1618 1691 ACP7 ACP domain, , module 7

REGION 1710 2121 KS8 KS domain, module 8

REGION 2203 2496 AT8 AT domain, acetate, module 8

REGION 2510 2702 DH8 DH domain, module 8

REGION 2944 3186 KR8 KR domain, module 8

REGION 3221 3294 ACP8 ACP domain, . module 8

BecG (SEO ID No. 33)

Molecule: BecG, 1986 aas Protein Molecule Features:

Type Start End Name Description

REGION 35 445 KS9 KS domain, module 9

REGION 553 847 AT9 AT domain, acetate, module 9

REGION 861 1066 DH9 DH domain, module 9

REGION 1341 1586 KR9 KR domain, module 9

REGION 1616 1690 ACP9 ACP domain, module 9

REGION 1767 1986 TE TE domain

The pathway for biosynthesis of BE- 14106 has been elucidated as follows and is shown in Figures 2A and 2B.

Biosynthesis of BE-14106 starts with assembly of a C17-C25 acyl moiety by the proteins BecA (which has the sequence SEQ ID NO:5 and is encoded by SEQ ID NO:4) and BecC (which has the sequence SEQ ID NO:9 and is encoded by SEQ ID NO:8) from 1 propionate and 2 acetate units (Figure 2A). The KR domain in BecC is most likely inactive, leaving a carbonyl group at Cl 9 of the synthesized polyketide chain. Biosynthesis of that aminoacyl starter continues with the activation of glycine by the discrete NRPS adenylation domain BecL (which has the sequence of SEQ ID NO:21 and is encoded by SEQ ID NO: 20) and loading of the activated glycine onto discrete peptidyl carrier protein BecS (which has the

sequence of SEQ ID NO: 19 and is encoded by SEQ ID NO: 18). BecU (which has the sequence of SEQ ID NO: 11 and is encoded by SEQ ID NO: 10) is thought to mediate the interaction between the ACP of BecC and BecS, bringing the substrates into proximity to each other. D-amino acid oxidase Bed presumably catalyzes the oxidative deamination of glycine, releasing ammonium, which makes a nucleophilic attack on the C- 17 carbonyl to afford a C- 17 imino group, which is subsequently reduced to an amino group. The latter reduction supposedly leads to the oxidation of the acyl chain and migration of double bonds. Thioesterase BecQ (which has the sequence of SEQ ID NO:41 and is encoded by SEQ ID NO:40) releases the aminoacyl chain from BecC-BecU-BecS complex, resulting in the formation of a aminoacyl-carboxylic acid. Putative acyl-CoA ligase, BecJ (which has the sequence of SEQ ID NO: 15 and is encoded by SEQ ID NO: 14), is assumed to activate the resulting aminoacyl-carboxylic acid through adenylation and subsequent ligation with CoA, making the aminoacyl-CoA an acceptable substrate for the acyl transferase BecK (which has the sequence of SEQ ID NO: 17 and is encoded by SEQ ID NO: 16). A discrete acyltransferase BecK transfers the activated aminoacyl chain to the ACP domain in module 1 on PKS BecB, which has the sequence of SEQ ID NO: 13 and is encoded by SEQ ID NO: 12. The latter PKS lacks all domains in module 1 except for ACP (loading module of Figure 2B). Modules 2 and 3 (modules 1 and 2 in Figure 2B) of BecB elongate the aminoacyl moiety from Cl 7 with 1 acetate and 1 propionate unit (C 16- 15 and C 14- 13), respectively.

The growing chain is then passed to BecD, which has the sequence SEQ ID NO:29 and is encoded by SEQ ID NO:28, for further elongation and modification BecD contains modules 4 and 5 (modules 3 and 4 in Figure 2B), which elongate the chain with 2 acetate units (C 12- 11 and C 10-9).

The chain is then passed to BecE which has the sequence SEQ ID NO:37 and is encoded by SEQ ID NO:36. Module 6 (module 5 in Figure 2B) of BecE PKS elongates the chain with one acetate unit (C8-7). The fact that the DH domain in this module is inactive is responsible for appearance of the C-9 hydroxy group. (The DH domain is believed to contain a deletion at the C-terminal region which eliminates the activity).

The chain is then passed to BecF which has the sequence SEQ ID NO:35 and is encoded by SEQ ID NO:34. Modules 7 and 8 (modules 6 and 7 in Figure 2B) of BecF PKS elongate the chain with one propionate (C6-5) and one acetate (C4-3) unit, respectively. The chain is then passed to BecG which has the sequence SEQ ID NO:33 and is encoded by SEQ ID NO:32. Module 9 (module 8 in Figure 2B) of the BecG PKS extends the chain with one acetate unit and the TE domain of BecG is responsible for the hydrolysis of the thioester bond and the release of the completed aminated polyketide chain from the PKS. This causes the cyclisation of the macrolactam ring, probably assisted by the putative homolog of proline iminopeptidase BecP (which has the sequence SEQ ID NO:31 and is encoded by SEQ ID NO:30). The biosynthesis is completed by P450 monooxygenase BecO (which has the sequence SEQ ID NO:27 and is encoded by SEQ ID NO:26), which hydroxylates 8-deoxy BE- 14106 at C-8. No clear role in the biosynthesis of BE- 14106 could be defined for BecT

(which has the sequence SEQ ID NO:39 and is encoded by SEQ ID NO:38). BecR (which has the sequence SEQ ID NO:43 and is encoded by SEQ ID NO:42) is not involved in the biosynthesis of BE- 14106, as proven by the gene inactivation experiment (Example 10). Proteins that are thought to be involved in the regulation of the pathway include the LuxR-type protein BecH (which has the sequence SEQ ID NO:3 and is encoded by SEQ ID NO:2) and the TetR-type regulator BecM (which has the sequence SEQ ID NO:23 and is encoded by SEQ ID NO:22).

The MFS-type efflux pump BecN (which has the sequence SEQ ID NO:25 and is encoded by SEQ ID NO:24)is thought to be responsible for BE- 14106 efflux/resistance .

The nucleotide sequences of the present invention provide important tools and information which can be utilised in a number of ways to manipulate BE- 14106 biosynthesis, to synthesise new BE- 14106 derivatives or analogues or novel polyketide-based or macrolactam molecules or structures, and to provide novel or modified PKS biosynthetic machinery for the biosynthesis of such novel polyketide or macrolactam molecules or structures. By PKS biosynthetic machinery is meant a

group of proteins (e.g. encoded by a gene cluster) that comprises one or more PKSs that may form a protein complex, collection or assembly, which is functional in polyketide synthesis, but which is not necessarily restricted only to the presence of PKS enzymes or enzymatic domains, and which may contain also other functional activities, e.g. other enzymatic (e.g. modificatory) or transporter or regulatory functional proteins). The proteins encoded by the gene cluster may thus be viewed as an "enzyme system" or "enzyme complex" or "protein system" or "protein complex" without necessarily implying that the proteins in the system are physically associated in any way. They are "functionally" associated in the sense of constituting the biosynthetic machinery for BE- 14106. They may alternatively be termed a "biosynthetic system" or "biosynthetic complex" or "assembly".

Thus, for example, the entire BE-14106 biosynthetic gene cluster or biosynthetic machinery or a constituent enzyme or enzyme complex as provided herein, or a portion thereof, may be subjected to modification. The modification takes place by modifying one or more genes or ORFs in the gene cluster so as to cause the modification of one or more encoded proteins or peptides (e.g. enzymes or modules, or enzymatic domains, or functional sequences within the encoded protein/peptide or enzyme). Thus, enzyme activity may be altered or inactivated so as to result in modification to the molecule (e.g. polyketide or macrolactam structure) which is synthesised. Such modified or derivatised NRPS-PKS biosynthetic machinery may thus be used to synthesize novel or modified polyketide or macrolactam moieties, as will be described in more detail below. In this situation, the BE-14106 biosynthetic gene cluster or enzyme complex or groups of enzymes provided herein, or a fragment or portion thereof, may function as an "origin" or "template" or "source" system or sequence for modification. Thus the NRPS-PKS biosynthetic machinery may be seen as a NRPS-PKS biosynthetic system, or "NRPS-PKS system".

As described in more detail below, in one embodiment the modification of the gene cluster may take place in situ. In other words, the endogenous gene cluster as contained in a microorganism which produces BE-14106 may be modified, for example by gene replacement or gene inactivation. Thus, the native gene cluster as it naturally occurs in a microorganism may be modified. Whilst recombinant

expression of a nucleic acid molecule of the invention is a possibility (i.e. the introduction of the nucleic acid molecule into a host cell (e.g. a heterologous host cell) and the culturing (or growth) of that host cell under conditions which allow the nucleic acid molecule to be expressed and the polyketide or macrolactam molecule to be produced (i.e. conditions which allow the expression product of the nucleic acid molecule to synthesise the polyketide/macrolactam molecule), this is less preferred. In such a recombinant expression system, the nucleic acid molecule may be subject to modification before being introduced into the host cell and expressed. . According to the invention and as further described below, the non- functional parts (e.g., non-biologically active parts, for example non-coding parts) of said system (i.e. of the gene cluster or protein complex or assembly or biosynthetic machinery) may be utilised as a "scaffold", and left unmodified and the functional parts (e.g. sequences encoding enzymatic portions) may be modified to yield the derivative or modified NRPS-PKS system. In preferred embodiments only a single selected, or few selected functional (e.g. enzymatic) regions, modules or domains may be modified, leaving the remaining sequence or structure largely intact.

Included within the scope of the invention are synthetic or recombinant polyketide synthase or other enzymes and complexes or systems containing such enzymes, or other proteins of the biosynthetic machinery, i.e. enzymes or proteins or complexes or systems derived from the scaffold encoded by the BE- 14106 biosynthetic gene cluster which are modified in order to change the properties of at least one protein encoded by the BE- 14106 biosynthetic gene cluster.

For example such a modification could be to include sequences encoding one or more functional units (e.g. modules or domains or even whole genes/ORFs) derived from other modular enzymes. Alternatively, such a modification could be to introduce sequences encoding one or more functional units derived from the BE- 14106 biosynthetic gene cluster but which are found at a different location in the naturally occurring sequence. The modification could also be modification which results in the inactivation or deletion of an encoded domain, module, enzyme or other functional unit in the BE- 14106 biosynthetic machinery.

Such functional units may be catalytic or a transport or regulatory protein domain. To perform such manipulations, the sequences that are used can be from the

nucleic acid molecule of the invention, or appropriate sequences encoding domains can be derived from nucleotide molecules encoding other polyketide synthesising enzymes, peptide synthesising enzymes, hybrid peptide polyketide synthesising enzymes, fatty acid synthesising enzymes or other enzyme domains known in the art.

Thus, in a very general sense, the present invention provides the use of the nucleic acid molecules of the invention as defined herein in the preparation of a modified BE-14106 biosynthetic gene cluster, the encoded biosynthetic machinery (or protein system) and the resultant modified molecules that are synthesised therefrom.

The nucleotide sequence of the nucleic acid molecules of the invention may be utilised in this way according to the present invention in a random or directed or designed manner, e.g. to obtain and test a particular predetermined or pre-designed structure, or to create random molecules, for example libraries of polyketide structures, e.g. for screening.

By modified BE-14106 molecule or BE-14106 derivative it is meant that the chemical structure of BE-14106 has been changed relative to that which is depicted in Figure 1. Such modifications can alter the functional properties of BE-14106 such that the strength or potency of the molecule (e.g. cytotoxic effects or antiproliferative activity) is enhanced or reduced, for example, or they can influence selective toxicity, solubility, pharmacokinetics, affinity to cellular target(s) or other properties of the compound.

Alternatively, the modifcations described herein can be used to change the way that the BE-14106 molecule is transported or its biosynthesis regulated. Such a novel or modified biosynthetic gene cluster, the encoded biosynthetic machinery including the constituent proteins or protein system and the resultant modified molecules that are synthesised therefrom each form a separate aspects of the present invention. Also included as part of the invention are cells into which a biosynthetic gene cluster has been introduced or cells containing or comprising such a modified biosynthetic gene cluster or the modified encoded biosynthetic machinery.

The genes or genetic elements which can be modified include not only the actual PKS genes or ORFs (which encode BecA, BecB, BecC, BecD, BecE, BecF, BecG) or the individual modules or domains thereof, but also genes or ORFs encoding other enzymes or functional proteins involved in BE-14106 biosynthesis (e.g. which encode BecU, Bed, BecP, BecJ, BecK, BecS, BecL, BecO, BecT, BecQ) and transport (e.g. which encode BecN) or regulation (e.g. which encode BecH, BecM), all of which are referred to herein collectively as "BE-14106 biosynthetic genes".

As regards the actual PKS genes, as will be described in more detail below, these may be modified to change the nature of a loading module domain which determines the nature of the starter unit, the number of modules, the nature of the extender, as well as the various dehydratase, reductase and synthase activities which determine the structure of the polyketide chain.

Thus, for example, the number of modules of a PKS enzyme may be increased or decreased, e.g. one or more modules of a PKS enzyme may be deleted; KS, DH and/or KR domains of a PKS enzyme may be deleted, inactivated or introduced (e.g. from the same PKS gene or another PKS gene), an AT domain of a PKS enzyme may be modified to alter its specificity for a starter or extender unit; the KS ^Q domain of a BecA (SEQ ID No. 4/5) may be modified (e.g. inactivated) to alter the nature of the starter unit which will constitute part of the side chain.

Change of specificity of BecO hydroxylase may provide for hydroxylation at carbon atoms of BE-14106 (or its analogues) other than C-8.

In addition to modification of the NRPS or PKS enzymes to change the nature of the synthesised molecules, the nucleic acid molecules of the invention can also be utilised to manipulate or facilitate the biosynthetic process, for example by extending the host range or increasing yield or production efficiency etc.

To enable recombinant expression of a nucleic acid molecule of the invention, the invention also provides a vector, for example a cloning or expression vector, comprising the nucleic acid of the invention and a host cell containing such a molecule. However, as noted above, this aspect of the invention is less preferred.

In practice, the modification advantageously can be carried out by manipulating the BE-14106 biosynthetic gene cluster sequence in situ in a cell

(which may be viewed herein as a "host cell") e.g. to alter the nucleotide sequence encoding an enzyme activity, so as to inactivate or modify that activity or to introduce an activity. Appropriate genetic constructs can be generated which contain sequences having the required modifications that are to be introduced into the BE-14106 biosynthetic gene cluster sequence (e.g. into SEQ ID No. 1) as contained in an endogenous nucleic acid molecule (in other words a nucleic acid molecule contained in the host cell). Introduction of a vector (e.g. a plasmid) with this sequence into the appropriate host cell can then be carried out. This ultimately leads to the integration of the modified sequence into the gene cluster (e.g. within the genome of the host cell) near the corresponding (e.g. endogenous or naturally occurring) portion of the biosynthetic gene cluster via homologous recombination. Upon a second recombination event, the endogenous portion of the gene cluster is replaced by the modified version and the vector is eliminated.

The resultant modified host cell will therefore contain a modified BE-14106 biosynthetic gene cluster, which encodes a modified BE-14106 enzyme system. The modified BE-14106 biosynthetic machinery thus synthesises a modified BE-14106 molecule.

As noted above, the gene cluster which is modified may be the native gene cluster which is naturally present in a host cell (microorganism) which produces BE- 14106 or a derivative thereof (and hence a native nucleic acid or endogenous nucleic acid molecule is modified). Less preferably, but still encompassed by the present invention, the nucleic acid molecule of the invention may be introduced into the host cell before modification. Alternatively, the nucleic acid molecule is introduced into the host cell after modification. Hence an exogenous nucleic acid molecule may be modified.

Given that the invention provides the sequence of the full length BE-14106 gene cluster, this gene replacement strategy is of general application to modify the BE-14106 gene cluster. The strategy can e.g. be used to delete a portion of the gene cluster, to introduce activities or to substitute activities found within the modules in the native or wild-type sequence. The strategy requires knowledge of the gene cluster sequence but does not necessarily require the complete gene cluster to be isolated from a host cell in order to carry out the manipulation.

Thus, in a further aspect the present invention provides a method for preparing a nucleic acid molecule encoding a modified BE-14106 NRPS-PKS biosynthetic machinery (or modified BE-14106 NRPS-PKS system), said method comprising modifying a nucleic acid molecule, as hereinbefore defined, encoding said BE- 14106 NRPS-PKS biosynthetic machinery (or modified BE- 14106 NRPS- PKS system).

The nucleic acid molecule is modified by modifying its sequence, and more particularly the nucleic acid molecule may be modified by introducing, mutating, deleting, replacing or inactivating a sequence encoding one or more activities (or proteins) encoded by said nucleic acid molecule. Thus, one or more sequences may be modified that encode enzymatic or other functional activities. Such modification results in a nucleic acid molecule which encodes a BE-14106 NRPS-PKS biosynthetic machinery (or BE-14106 NRPS-PKS system) which has altered function or activity or altered properties as compared to the native or wild-type BE- 14106 NRPS-PKS biosynthetic machinery (or BE-14106 NRPS-PKS system). Thus the modified biosynthetic machinery (or NRPS-PKS system) may have one or more altered or modified enzymatic activities and may result in the synthesis of a molecule (e.g polyketide or macrolactam molecule) which is other than (or different to) the molecule synthesised from the native (i.e unmodified) biosynthetic machinery (or NRPS-PKS system). Alternatively, as noted above modification of the biosynthetic machinery may result in an improvement in the biosynthetic process e.g. increased yield etc.

The nucleic acid which is modified may be contained within a cell or organism, which may be the cell or organism used for production of the polyketide/macrolactam molecule which is synthesised by the modified biosynthetic machinery.

As noted above, the nucleic acid molecule which is modified may advantageously be the nucleic acid molecule which is endogenously (or naturally) present in an organism which produces BE-14106 (or a derivative thereof). Thus, the method of the invention may involve modifying in situ a native nucleic acid molecule (more particularly modifying the sequence of a nucleic acid molecule) within a cell or organism (generally a microbial cell or a microorganism) which

produces BE- 14106 or a derivative thereof. The nucleic acid molecule is or represents the BE-14106 biosynthetic gene cluster or a part thereof and may thus be seen as a nucleic acid molecule encoding the BE-14106 biosynthetic machinery or BE-14106 NRPS-PKS system or a part thereof. A number of different microorganisms may produce BE- 14106 and it is further known that some microrganisms may produce naturally occurring derivatives of BE-14106 such as the 8-deoxy derivative (Takahashi et al., supra). Reference to a BE-14106 gene cluster or biosynthetic machinery (or NRPS-PKS system etc) is intended to include gene clusters or biosynthetic machinery etc that produce not only BE-14106 itself but also naturally occurring derivatives such as the 8-deoxy derivative (designated GT32-B).

The invention further provides a method for preparing a modified BE-14106 NRPS-PKS biosynthetic machinery (or modified BE-14106 NRPS-PKS system), said method comprising expressing a modified nucleic acid molecule prepared (or obtained) as defined above. This may be achieved simply by modifying the nucleic acid molecule contained in the cell, as described above, and allowing the cell to grow under conditions in which the modified nucleic acid molecule is expressed. Thus, for example, the native nucleic acid molecule in the cell may be modified in situ and the cell may be allowed to grow. This aspect of the invention may also provide the modified BE- 14106

NRPS-PKS biosynthetic machinery (or modified BE-14106 NRPS- PKS system) obtained by such a method.

The invention also provides a method for preparing a modified polyketide or macrolactam molecule, said method comprising expressing a modified nucleic acid molecule prepared (or obtained) as defined above.

Generally speaking, the nucleic acid molecule will be expressed in a host cell under conditions in which the modified biosynthetic machinery may be expressed. As outlined above, this may be achieved by introducing the nucleic acid molecule into a host cell, but generally the "host cell" will be the cell or organism in which the nucleic acid molecule is naturally or endogenously present and in which the nucleic acid molecule is modified. The host cell will be grown or cultured under conditions which allow the modified nucleic acid molecule and biosynthetic machinery to be

expressed, and the molecule produced from the biosynthetic machinery to be synthesised.

Thus, the nucleic acid molecule may be expressed in any desired host cell, but preferably it will be expressed in the cell or microorganism from which it was (or from which it may be) derived and in which the (unmodified) molecule is natively present.

The method of the invention for preparing a modified polyketide or macrolactam molecule may include the further step of recovering (e.g. isolating or purifying) the molecule e.g. from the culture medium in which the host cell was grown or from the host cell. Thus, this aspect of the invention may also provide the modified polyketide or macrolactam molecule obtained by such a method

A further aspect of the present invention thus provides a cell or microorganism containing a nucleic acid molecule encoding a modified BE- 14106 NRPS-PKS biosynthetic machinery (or modified BE-14106 NRPS-PKS system) obtained by a method as hereinbefore defined. Alternatively, the cell or microorganism may contain a modified BE-14106 NRPS-PKS biosynthetic machinery (or modified BE-14106 NRPS-PKS system) obtained by a method as defined above.

In an alternative but less preferred embodiment the invention may also provide a host cell containing a nucleic acid molecule as defined above, wherein said molecule has been introduced into said host cell.

By way of representive example, it is envisaged that such manipulations of the gene cluster can be performed with the aim of generating a modified BE-14106 molecule in which a hydroxyl group is introduced in any one or more of positions C- 2, 3, 4, 7, 11, 13, 15, 17, 21 and 23 of BE-14106 or a derivative or modified version thereof. This can be achieved by inactivating or deleting the appropriate DH domain(s) in the BE-14106 biosynthetic gene cluster. According to the biosynthetic pathway proposed herein the following modifications should be made:

Table 3

In addition to or as an alternative, it is envisaged that such manipulations of the gene cluster can be performed with the aim of generating a modified BE- 14106 molecule in which an oxo (keto) group is introduced in any one or more of positions C-2, 3, 4, 7, 9, 11, 13, 15, 17,21 and 23 of BE-14106 or a derivative or modified version thereof. This can be achieved by inactivating or deleting the appropriate KR domain(s) in the BE-14106 biosynthetic gene cluster. According to the biosynthetic pathway proposed herein the following modifications should be made:

Table 4

Thus, a further aspect of the present invention provides a BE- 14106 analogue which may be produced by modifying the genes/proteins of the BE-14106 biosynthetic pathway. A BE-14106 analogue of the invention includes, but is not limited to, a molecule comprising a modification selected from any one or more of the group comprising 8-deoxy BE-14106, 3-, 5-, 7-, 11-, 13-, 15-, 17- or 23-hydroxy BE-14106 and 3-, 5-, 7-, 9-, H-, 13-, 15-, 17- or 23-oxo BE-14106 or combinations thereof. A selection of the structures of such representative BE-14106 analogues of the invention, and modules which require inactivation to generate said analogues, are shown in table 5.

Table 5 - BE-14106 analogues that can be produced upon engineering of the polyketide synthase genes

In addition to or as an alternative to the above modifications, the C20-C25 side chain can be shortened by deletion of the DNA regions in becA encoding entire module(s) of BecA PKS. Alternatively, one or more molecules of BecA may be deleted to shorten the side chain. For example, deletion of one entire module of BecA may shorten the side chain.

In addition to or as an alternative to the above modifications, one or more of the PKS modules (e.g. of BecA, BecB, BecD, BecE, BecF, BecG) can be deleted in order to produce analogues having macrolactam rings having fewer members (e.g. 18-, 16-, 14-, 12-, 10- and 8- membered rings). As shown in Table 5, and in addition to or as an alternative to other modifications described or tabulated above, BecO can be inactivated or deleted so as to eliminate the hydroxyl group at position 8.

Further modifications which may be made in addition to or as an alternative to the above modifications include substitution of the C-I carbonyl of BE- 14106 with thio-carbonyl (C=S) or carboxamide (C=NH) to increase the stability of the molecule (for example against protease degradation to which BE- 14106 may be susceptible due to the similarity of the -N-C=O linkage in BE- 14106 to a peptide bond); glycosylation of BE-14106 or a modified BE-14106 molecule, e.g. with monosaccharide moieties; acylation of free hydroxyl groups, and reduction of the C21-C22 double bond is the side chain of BE-14106.

Regulator proteins can be overexpressed in order to increase production of the molecule synthesised, e.g. BE-14106 or a derivative or analogue thereof.

As such, a method of producing a modified BE-14106 biosynthetic gene cluster is provided, in which one or more of the modifications as set out above is made to the nucleic acid molecule of the invention as defined herein. Conveniently, said modifications may be made by carrying out gene replacement in a host cell which endogenously contains the nucleic acid molecule of the invention.

By "gene replacement" it is meant any method in which a gene or ORF or portion thereof (e.g. module, domain or other functional unit) as found in the nucleic acid molecule of the invention is effectively replaced with or substituted by a modified version thereof. The modified version thereof can be one in which the gene or ORF or portion thereof (e.g. the sequence encoding the protein or peptide or

domain or module thereof) is changed or altered (e.g. mutated such as by substitution, deletion or insertion of one or more nucleotide residues). Such changes could result in changes to the activity of the encoded polypeptide, but also includes changes which result in one or more proteins or polypeptides or domains or modules thereof being deleted or inactivated. Preferably, the modified version is one in which the sequence encoding said protein or polypeptide or one or more modules or domains thereof is inactivated or deleted.

It should be noted that the method does not require that the whole gene or ORF is physically removed and replaced, but rather that some or all of the sequence as found in the nucleic acid molecule of the invention is replaced with a modified version thereof.

Gene replacement can be performed according to techniques that are known in the art (see for example Sekurova et al., 1999 FEMS Microbiol. Lett., 177, 297- 304). A host cell (e.g. microorganism) which endogenously contains the nucleic acid molecule of the invention is preferably subjected to modification of the nucleic acid molecule e.g. by gene replacement, so as to change or alter the BE-14106 biosynthetic gene cluster that is present in said host cell. A host cell endogenously contains the nucleic acid molecule of the invention if said nucleic acid molecule of the invention is normally found in that host cell or naturally occurs in that host cell (i.e. the nucleic acid molecule is present in the genetic material of the host cell). In other words, the nucleic acid molecule of the invention is not present in the host cell merely by virtue of it having been introduced (e.g. transfected or transferred) into the host cell e.g. in a recombinant DNA molecule such as a plasmid. Whether or not a host cell contains the nucleic acid molecule of the invention can be ascertained by determining whether the host cell can synthesise BE-14106 or a derivative thereof. Alternatively or additionally, genetic techniques to analyse the sequences of the nucleic acid molecules present in the host cell .can be carried out (e.g. PCR, Southern Blotting and other standard techniques that are known in the art). Such genetic techniques can be used to determine whether the nucleic acid molecule is present endogenously.

The host cell for use in the methods of the invention may be any desired cell or organism, prokaryotic or eukaryotic, but generally it will be a microorganism particularly a bacterium. More particularly, the host cell will be an actinomycete.

Preferred host cells include Streptomyces strains, preferably that endogenously contain the nucleic acid molecule of the invention. A suitable example is Streptomyces strain espi-A14106 (FERMP-11378, as referred to in JP4001179). The novel isolated strain referred to above, from which the gene cluster was sequenced (isolate MP28-13) as deposited under deposit number DSM21069 at the DSMZ is particularly preferred. Further, a method of producing a modified polypeptide encoded by the BE-

14106 biosynthetic enzyme system is provided, in which a modified BE-14106 biosynthetic gene cluster obtainable by the above described methods is expressed in a host cell. Once the appropriate modifications have been made to the BE-14106 biosynthetic gene cluster, e.g. by the methods set out above, the host cell can be grown or cultured under conditions which allow the expression of polypeptides and proteins from the modified gene cluster so as to produce the modified BE-14106 biosynthetic polypeptide(s).

As such, the modified host cell containing the modified BE-14106 biosynthetic gene cluster will express the polypeptides and proteins encoded by the gene cluster and this will lead to the production of a modified polypeptide or protein, as well as the other proteins encoded by the gene cluster. As a result of the presence of one or more modified polypeptides or proteins in the biosynthetic machinery, the properties of the biosynthetic machinery for BE-14106 will be changed. Depending on the nature of the modification(s) that was made to the BE- 14106 biosynthetic gene cluster, the properties of the encoded polypeptides, proteins and enzymes will be different to those of the wild type.

For most purposes, the production of the BE-14106 biosynthetic machinery within the cell will suffice. Alternatively, the proteins which make up the BE-14106 biosynthetic machinery can be purified from the cell in which it was expressed. A further aspect of the invention provides a polypeptide encoded by the nucleic acid molecule of the invention which has been modified as defined above.

A method of producing a modified BE- 14106 molecule is also provided. According to this method, a modified BE-14106 biosynthetic gene cluster obtainable by the above described methods is expressed in a host cell. This is achieved by growing or culturing a host cell in which a modified BE-14106 biosynthetic gene cluster obtainable by the above described methods is present under conditions which allow the expression of the polypeptides, proteins and enzymes encoded by the modified BE-14106 biosynthetic gene cluster. The cell will thus contain the biosynthetic machinery necessary for the biosynthesis of the modified BE-14106 molecule and synthesis of this molecule will ensue. The method may further comprise the step of recovering, e.g. isolating or purifying the modified BE-14106 molecule. This can be isolated or purified from the cell culture medium into which it has been transported or secreted if appropriate, or otherwise from the host cell in which it has been included. Thus, for example, the cells of the producing organism may be harvested, e.g. by centrifugation, and may be extracted, for example with organic solvent(s) (e.g., methanol or other alcohols). The molecules may be recovered from such an extract, for example by precipitation. Further purification of a crude product obtained in this way may include e.g. chromatography, e.g. HPLC.

In order to enable practice of the invention according to the principles above, the invention also provides a host cell containing the nucleic acid molecule as herein defined, and more particularly a cell containing a nucleic acid moleclue of the invention modified as defined above.

In general, the methods of the invention can be carried out on any host cell which contains the nucleic acid molecule of the invention, preferably host cells which endogenously contain the molecule. As noted above, preferred host cells include cells of Streptomyces spp. More preferred cells are Streptomyces cells which have the antibiotic resistance and sensitivity characteristics as set out in Table 6.

In a highly preferred embodiment, the methods are carried out using the novel strain of Streptomyces MP28-13 deposited under number DSM21069 at the DSMZ) or a mutant or modified strain thereof which produces BE-14106 or a derivative thereof.

The invention thus further provides a microorganism, particularly a bacterium and especially a Streptomyces, obtainable by the methods as described herein.

In an important aspect the invention also provides a strain of Streptomyces as deposited under number DSMZ 21069 at the DSMZ, or a mutant or modified strain thereof which produces BE- 14106 or a derivative thereof.

The methods of the invention may be seen to result in the production of derivative nucleic acid molecules, derived from the nucleic acid molecules of the invention defined above. The derivative nucleic acid molecule may be formed in situ in host cells or microorganisms in which the unmodified nucleic acid molecules are contained. Alternatively, but less preferably, they may be formed in host cells in which the nucleic acid molecules are introduced. Such "modified" microoganisms containing the derivative or modified nucleic acid molecules according to the present invention may be used to form libraries, for example libraries of polyketides or polyketide- based or macrolactam molecules wherein the members of the library are synthesized by modified BE-14106 biosynthetic machineries or NRPS-PKS systems derived from the naturally occurring BE- 1406 system provided herein. Generally, many members of these libraries may themselves be novel compounds, and the invention further includes novel members (compounds) of these libraries. The methods of the invention may thus be directed to the preparation of an individual compound. The compound may or may not be novel, but the method of preparation permits a more convenient method of preparing it. The resulting compounds (e.g. polyketides) may be further modified, for example, to convert them to further active molecules, e.g. antibiotics, for example by glycosylation. The invention also includes methods to recover novel compounds with desired activities by screening the libraries of the invention.

The invention provides libraries or individual modified forms, ultimately of polyketide-based, or macrolactam molecules, by generating modifications in the BE- 14106 NRPS-PKS gene cluster so that the proteins produced by the cluster have altered activities in one or more respects, and thus produce compounds other than the natural product of the NRPS-PKS system encoded by the gene cluster (i.e. BE-

14106). Novel compounds may thus be prepared, or compounds in general prepared more readily, using this method. By providing a large number of different genes or gene clusters derived from the naturally occurring BE- 14016 gene cluster, each of which has been modified in a different way from the native cluster, an effectively combinatorial library of compounds can be produced as a result of the multiple variations of these activities. The modified NRPS-PKS encoding sequences and biosynthesis machinery/systems used in the present invention thus represent encoding sequences and enzyme/protein machinery or systems "derived from" a naturally occurring BE-14106 NRPS-PKS biosynthetic machinery or system. By a biosynthetic machinery or NRPS-PKS system "derived from" the BE-

14016 biosynthetic machinery is meant a biosynthetic machinery or NRPS-PKS system in which at least one enzymatic or functional activity is mutated, deleted, inactivated or replaced, so as to alter the activity. Alteration results when these activities are deleted or are replaced by a different version of the activity, or simply mutated in such a way that a compound (e.g. a polyketide or macrolactam) other than the natural product results from these collective activities. This occurs because there has been a resulting alteration of the starter unit and/or extender unit, and/or stereochemistry, and/or chain length or cyclization and/or reductive or dehydration cycle outcome at a corresponding position in the product compound. Where a deleted activity is replaced, the origin of the replacement activity may come from a corresponding activity in a different naturally occurring NRPS or polyketide synthase or from a different region of the same NRPS-PKS system/machinery. Modification or manipulation of the modular NRPS-PKS may involve truncation, e.g. gene or domain or module deletion or domain/gene/module swapping, addition or inactivation, which may involve insertion or deletion.

Alternatively, random or directed modifications (i.e. mutations) may be made in the nucleotide sequence of the selected portion (e.g. in a gene/domain/module etc).

Advantageously, a biosynthetic machinery or NRPS-PKS system "derived from" the BE-14106 machinery or system contains at least an NRPS adenylation domain and at least two modules of a PKS enzyme and may optionally contain mutations, deletions, or replacements of one or more of the activities of these functional domains or modules so that the nature of the resulting compound is

altered. This definition applies both at the protein and genetic levels. Particular preferred embodiments include those wherein a KS, AT, KR or DH has been inactivated or deleted or replaced by a version of the activity from a different machinery or PKS/NRPS system or from another location within the same machinery or NRPS-PKS system. Also preferred are derivatives where at least one noncondensation cycle enzymatic activity (e.g. KR or DH) has been deleted or wherein any of these activities has been mutated so as to change the ultimate compound synthesized.

Thus, there are five degrees of freedom for constructing a PKS biosynthetic machinery or system in terms of the compound that will be produced. First, the polyketide chain length will be determined by the number of modules in the machinery or system. Second, the nature of the carbon skeleton of the polyketide will be determined by the specificities of the acyl transferases which determine the nature of the extender units at each position, e.g. malonyl, methyl malonyl or ethyl malonyl, etc. Third, the loading domain specificity will also have an effect on the resulting carbon skeleton of the polyketide. Thus, the loading domain may use a different starter unit, such as acetyl, propionyl, and the like. Fourth, the oxidation state at various positions of the polyketide will be determined by the dehydratase and reductase portions of the modules. This will determine the presence and location of ketone, alcohol, double bonds or single bonds in the polyketide.

Finally, the stereochemistry of the resulting polyketide is a function of various aspects of the synthase. The first aspect is related to the AT/KS specificity associated with substituted maloyls as extender units, which affects stereochemistry only when the reductive cycle is missing or when it contains only a ketoreductase since the dehydratase would abolish chirality. Second, the specificity of the ketoreductase will determine the chirality of any β-OH.

By modifying the PKS involved in the biosynthesis of the aminoacyl "starter", the compound that is produced can be altered.

Thus the modified machinery or NRPS-PKS system may permit a wide range of compounds to be synthesized.

The size of the synthesized product can be varied by varying the number of modules.

The polyketide/macrolactam products of the modified biosynthetic machinery may be further modified for example by glycosylation or other derivatisation, in order to exhibit or improve activity e.g. antibiotic activity. Methods for glycosylating polyketides are generally known in the art; the glycosylation may be effected intracellularly by providing the appropriate glycosylation enzymes or may be effected in vitro using chemical synthetic means.

In order to obtain nucleic acid molecules encoding a variety of derivatives (or analogues) of the naturally occurring BE- 14016 NRPS-PKS system, and thus a variety of polyketides macrolactam-based compounds, a desired number of constructs can be obtained by "mixing and matching" enzymatic activity-encoding portions, and mutations can be introduced into the native host nucleic acid molecule/gene cluster or portions thereof.

Mutations can be made to the native sequences using conventional techniques. The substrates for mutation can be an entire cluster of genes or only one or two of them; the substrate for mutation may also be portions of one or more of these genes. Techniques for mutation are well known in the art and described in the literature. Such techniques include preparing synthetic oligonucleotides including the mutation(s) and inserting the mutated sequence into the gene using restriction endonuclease digestion. Alternatively, the mutations can be effected using a mismatched primer (generally 15-30 nucleotides in length) which hybridizes to the native nucleotide sequence, at a temperature below the melting temperature of the mismatched duplex. The primer can be made specific by keeping primer length and base composition within relatively narrow limits and by keeping the mutant base centrally located. Primer extension is effected using DNA polymerase, the product cloned and clones containing the mutated DNA, derived by segregation of the primer extended strand, selected. The technique is also applicable for generating multiple point mutations. PCR mutagenesis will also find use for effecting the desired mutations.

The vectors used to perform the various operations to replace the enzymatic activity in the host genes or ORFs or to support mutations in these regions of the host genes or ORFs may be chosen to contain control sequences operably linked to

the resulting coding sequences in a manner that expression of the coding sequences may be effected in the host. However, simple cloning vectors may be used as well.

The invention will now be described in more detail in the following non- limiting Examples with reference to the drawings in which:

Figure 1 shows the chemical structure of the macrolactam antibiotic BE- 14106;

Figure 2A shows the proposed initiation of biosynthetic pathway for BE-14106;

Figure 2B shows the proposed completion of the BE- 14106 biosynthesis.

Example 1- Characterisation of isolate MP28-13 (Streptomyces strain DSM 21069)

On ISP2 agar growth medium (Difco, USA): The substrate mycelium is pale yellow, the same color as the ISP2-agar plates. The aerial mycelium is white and the spores are almost white, just slightly greenish.

On SFM agar growth medium (soya flour, 20g/l; mannitol, 20 g/1; agar, 20 g/1): The substrate mycelium is more beige than on ISP2, aerial mycelium and spores the same as on ISP2.

On ISP2 plates growth is visible after 2 days, but sporulation takes about 20 days. Sporulation is quite poor on both media.

Growth in liquid media: Grows well in TSB liquid growth medium (Oxoid, UK), with shaking at 225 rpm and glass beads (3 mm). 2 days at 25°C is necessary to obtain sufficient mycelium.

The strain grows at 20°C, 25 ⁰ C and 30 ⁰ C, but the optimal temperature is around 25°C. At 30 ⁰ C the sporulation is affected.

The 16S RNA gene sequence of strain DSM 21069 (isolate MP28-13) is shown in SEQ ID No. 46.

Table 6 shows the antibiotic resistance characteristics of strain DSM 21069

Table 6 Antibiotic resistance

Antibiotic 5 μg/ml 10 μg/ml 20 μg/ml 50 μg/ml

Apramycin sensitive sensitive sensitive sensitive

Kanamycin sensitive sensitive sensitive sensitive Neomycin sensitive sensitive sensitive sensitive

Rifamycin resistant resistant resistant sensitive

Streptomycin resistant resistant resistant resistant

Thiostrepton sensitive sensitive sensitive sensitive

Example 2 - Generation of a probe for the BE-14106 biosynthesis gene cluster

Total DNA was isolated from DSM 21069 (MP28-13) using the DNeasy Blood & Tissue Kit (QIAGEN). β-ketoacyl synthase (KS) domains were amplified using the degenerate primers KSMA-F (5'-TS GCS ATG GAC CCS CAG CAG-3' [SEQ ID No. 47]) and KSMB-R (5'-CC SGT SCC GTG SGC CTC SAC-3' [SEQ ID No. 48]) described by Izumikawa et al. ((2003) Bioorg. Med. Chem., 11 , 3401- 3405). The 50 μl reaction mix contained total DNA isolated from MP28-13 (10-20 ng), Ix ThermoPol Reaction Buffer (New England Biolabs), 400 nM of each primer, 200 μM of each dNTP and 2,5 U of Taq DNA Polymerase (New England Biolabs). The reaction was run at 95°C for 5 min, then 35 cycles of 1 min at 95°C, 1 min at 60°C and 2 min at 72°C, and then a final 5 min extension at 72°C.

The 50 μl reaction mix was subjected to a gel electrophoresis and the resulting DNA fragment of about 700 bp was purified using the QIAEX II Suspension (QIAGEN). The purified PCR-product was cloned in the pDrive vector (QIAGEN) in E.coli EZ-cells using the QIAGEN PCR Cloning Kit (QIAGEN). Plasmid DNA from the transformants was isolated using the Wizard® Plus SV Minipreps DNA Purification System (Promega).

8 recombinant plasmids were sequenced using the pDrive-specific primers Ml 3 forward (-20) (5 ¹ GTA AAA CGA CGG CCA GT 3' [SEQ ID No. 49]) and Ml 3 reverse (5 ¹ AAC AGC TAT GAC CAT G 3' [SEQ ID No. 50]) described in the QIAGEN PCR Cloning Handbook (QIAGEN, 2001 ). The sequencing was performed using BigDye® Terminator vl .1 Cycle Sequencing Kit (Applied Biosystems). Of the 8 sequences obtained 5 were different from each other.

Translation into protein sequences and BLAST searches gave a match for PKS type I for all of the sequences (see Table 7).

The most interesting sequences were no. 1, 3, 6, 7 and 8. Sequence no. 1 matched a PKS involved in the biosynthesis of meridamycin in 5. violaceusniger (Sun et al, 2006 Microbiol., 152, 3507-3515). Sequences no. 3, 7, 8 gave a match to LnmJ involved in the biosynthesis of leinamycin in S. atroolivaceus (Cheng et al, 2003 Proc. Natl. Acad. Sci. USA, 100, 3149-3154). Sequence no. 6 gave a strong match to VinPl involved in the biosynthesis of the macrolactam vicenistatin in S.halstedii (Ogasawara et al, 2004 Chem. & Biol., 11, 79-86) and also to AVES 2 involved in the biosynthesis of the macrolide avermectin in S. avermitilis (Ikeda et al, 1999 Proc. Natl. Acad. Sci. USA, 96, 9509-9514).

Sequence no. 6 (SEQ ID No. 51) was chosen as a probe for screening the genomic library constructed for DSM 21069 (MP28-13). A digoxygenin (DIG) labeled probe was generated using the PCR DIG Probe Synthesis Kit (Roche Applied Science) and the Ml 3 primers described above. The plasmid containing sequence no.6 was used as a template. The reaction was run at 95°C for 3 min, then 30 cycles of 45 sec at 95°C, 1 min at 44°C and 3 min at 68°C, and then a final 7 min extension at 68°C. The resulting PCR product was subjected to a gel electrophoresis and the DNA fragment purified with the QIAEX II Suspension (QIAGEN).

Table 7. Sequencing of PCR amplified KS domains from DSM 21069

Example 3 - Optimisation of the conjugation procedure

To establish a procedure for genetically modifying strain DSM 21069 (MP28-13), conjugation with E. coli strain ET12567 (pSOK804 + pUZ8002) (Sekurova et al., 2004, J. Bacteriol., 186, 1345-1354) was tested following the procedure described by Flett et al. (1997 FEMS Microbiol. Lett., 155, 223-229). A few modifications to the procedure were made. The E. coli donor was grown to an OD _6O o of 0.4-0.5. Only fresh spore suspension of MP28-13 was used and the spore suspension was made with 2 ^χ YT (Sambrook et al., 2000 Molecular cloning: a laboratory manual. Cold Spring Harbor Laboratory Press, New York, NY). Spores and the donor cells were mixed and pelleted by centrifugation, and the pellet was resuspended in a smaller volume and spread on two SFM plates. Conjugation plates were incubated 24 h before addition of antibiotics (0.9 mg/ml nalidixic acid and 1.5 mg/ml apramycin). The incubation temperature and the temperature and time of the heat shock were varied to find the optimum. The best results were obtained at an incubation temperature of 25°C and a heat shock at 5O ⁰ C for 5 min.

Example 4 - Gene inactivation experiment

The PKS sequence no. 6 from strain DSM 21069 (MP28-13) cloned in the pDrive vector was excised from the plasmid using restriction enzymes BamH I and Hind III, and ligated with the 3.1 kb BamH I/Hind III fragment from the vector pSOK201 (Zotchev et al., 2000 Microbiol., 146, 611-619) and transformed into E.coli DH5α. The new construct was checked by restriction analysis, and then transformed into E.coli ET12567 (pUZ8002). The construct was transferred into DSM 21069 by conjugation following the procedure described above. The 3.1 kb BamH I/Hind III fragment from the vector pSOK201 does not contain genetic elements needed for autonomous replication in Streptomyces. Therefore, the transconjugants can only be obtained if this part of the vector is ligated with a fragment having high level of homology to the chromosomal DNA fragment in DSM 21069. Such homology allows for recombination leading to integration of the entire vector into the corresponding chromosomal region. If the cloned fragment

does not contain start or stop codons of the gene, such integration will lead to gene disruption, effectively inactivating chromosomal copy of the gene.

A single transconjugant was obtained and analyzed for BE-14106 production. No production of BE- 14106 was observed, verifying that sequence no. 6 belongs to the BE- 14106 biosynthetic gene cluster.

Example 5 - Construction of the genomic library

The genomic library for DSM 21069 was constructed in the cosmid vector SuperCos 1 (Stratagene) according to manufacturer's instructions (Stratagene, 2005). Genomic DNA was isolated from DSM 21069 following the Kirby mix procedure (Kieser et al., 2000 Practical Streptomyces Genetics, The John Innes Foundation, Norwich, England.), partially digested with Mbo I and dephosphorylated before ligation with Xba I-, CIAP- and BamH I-treated SuperCos 1. E.coli XLl -Blue MR (Stratagene) was used as a host for the construction of the library.

Example 6 - Screening of the genomic library

The library was plated out on Luriana-Bertani (LB) agar plates (Corning® Low Profile Square Bio Assay Dish) containing 100 μg/ml ampicillin to give -2000 colonies per plate. 2304 colonies were picked using a Genetix QPixII Colony Picker and transferred to 24 96 well plates (Nunc) containing LB broth and 24 96 well plates containing Reduced Hi+YE-medium (120 μl in each well). The well plates were incubated with shaking (900 rpm) at 30°C overnight. Glycerol were added to the 24 LB plates to give a final concentration of 15 % (v/v) and then stored at -80°C. Culture from the 24 Reduced Hi+YE plates were transferred to 6 384 well plates (Nunc) using a Tecan Genesis RSP 200 robotic liquid handling system and then stamped on a filter (4 replica stampings) using the Genetix QpixII Colony Picker. The process was repeated 4 times to give 4 replica filters. The filters were dried for 20 min under a sterile hood. Cultures were lysed by placing the filters on 3MM Whatman filter paper saturated with 10% SDS for 5 min. DNA was denatured by placing the filters on 3MM Whatman filter paper saturated with NaOH/Chloride Buffer (1.5 M NaCl, 0.5

M NaOH) for 10 min and neutralized with Tris/NaCl Buffer (3 M NaCl, 1 M Tris- Cl, pH 7.4) for 10 min. Finally the filters were submerged in 2X SSCP Buffer (2X SSC + 0.1 % (w/v) Sodium Pyrophosphate) to remove colony debris and baked at 80°C for 2 hrs. The filters were stored at 4°C until hybridization was started. Hybridization was carried out as described for the DIG System (Roche Applied Science) using the probe obtained from DSM 21069. By exposing the filter to an X- ray film, 3 candidate cosmids were identified and their corresponding hosts could be restreaked from the LB-plates stored at -80 ⁰ C.

Cosmid DNA was isolated from overnight cultures using the Wizard® Plus SV Minipreps DNA Purification System (Promega) and end-sequenced using primers designed for the cosmid regions flanking the insert site (SuperCos forw; 5' GGC CGC AAT TAA CCC TCA C 3' [SEQ ID No. 52] and SuperCos rev; 5' GGC CGC ATA ATA CGA CTC AC 3' [SEQ ID No.53]). The sequencing was performed using BigDye® Terminator vl.l Cycle Sequencing Kit (Applied Biosystems). Results are given in Table 8.

Table 8: End-sequencing of cosmid 1, 2 and 3.

The results indicated that cosmid 1 might contain one end of the cluster and cosmid 3 the other end. The 3 cosmids were tested to see if they contained any overlapping sequences by designing primers for the end-sequences and using those primers for sequencing the other cosmids. From these results it was concluded that cosmid 2 and cosmid 3 were overlapping, but cosmid 1 did not have any overlap with cosmid 2 and 3. Primers were designed for the forward primer end-sequence of cosmid 2 for amplifying a new probe for the missing part of the cluster. A digoxygenin (DIG) labelled probe was generated using the PCR DIG Probe

Synthesis Kit (Roche Applied Science) with cosmid 2 as a template. The reaction was run at 95°C for 5 min, then 35 cycles of 1 min at 95°C, 1 min at 60°C and 2 min at 72°C, and then a final 5 min extension at 72°C. The resulting PCR product was subjected to a gel electrophoresis and the DNA fragment purified with the QIAEX II Suspension (QIAGEN). One of the replica filters was used for hybridization with the new probe and 2 new candidate cosmids were identified (cosmid 4 and 5). The process with end-sequencing and cross-sequencing with the other cosmids was repeated and cosmid 4 was found to overlap with both cosmid 1 and 2.

Example 7 Verification of BecA function

To verify that the gene cluster contained in cosmids 1-4 was responsible for the BE-14106 production, another gene inactivation experiment was carried out. PKS fragments were amplified from cosmid 2 and 3 using one cosmid primer (SuperCos forw [SEQ ID No. 52] or SuperCos rev [SEQ ID No. 53]) and one degenerate primer for the KS domain (KSMA-F [SEQ ID No. 47] or KSMB-R [SEQ ID No. 48]). A 1.2 kb fragment was obtained for cosmid 2 and a 3.7 kb fragment for cosmid 3. Both fragments were cloned in pDrive and pSOK201 as described above and the construct was transferred into MP28-13 by conjugation. For the cosmid 2 fragment only one transconjugant was obtained. For the cosmid 3 fragment several transconjugants were obtained and 6 were chosen for further analysis.

Table 9 Analysis of BE-14106 production in knock-out mutants

Based on the results from the cross-sequencing of the cosmids and the gene inactivation experiment, cosmids 1, 2, 3 and 4 were sequenced.

Example 8 - Production, purification and identification of BE-14106

Cultivation of MP28- 13

Preparation of standard inoculum

Inoculum: Spores from an agar plate was transferred to a shake flask (250 ml, baffled) with 50 ml modified TSB-medium supplemented with glucose (composition given in Table 9). To increase the shear forces in the shake-flask, 3 g of 3 mm glass beads was added.

Incubation: The culture was incubated at 25 °C for 3 days at 225 rpm (Infors Multitron shaking incubator, orbital movement, amplitude 2.5 cm).

Preservation: Glycerol was added to the culture to a concentration of 15 %. The mixture was transferred to cryo vials and stored at -80°C.

Preparation of pre-culture for production

Inoculum: 1.5 ml standard inoculum was transferred to a shake flask (250 ml, baffled) with 50 ml modified TSB-medium supplemented with glucose (composition given in Table 10) and 3 g of 3 mm glass beads. Incubation: The culture was incubated at 25 °C for 2 days at 200 rpm (Infors

Multitron shaking incubator orbital movement, amplitude 2.5 cm

Production culture ,

Inoculum: 3 ml pre-culture for production was transferred to a shake flask (500 ml, baffled) with 100 ml 0.3 x BPS-medium supplemented with glucose (composition given in Table 11) and 5 g of 3 mm glass beads. Incubation: The culture was incubated at 25 °C for 2 days at 200 rpm (Infors

Multitron shaking incubator orbital movement, amplitude 2.5 cm.

Composition of media used for production

Table 10. Composition of modified TSB- medium supplemented with glucose

a) Autoclaved separately.

Table 11. Composition of 0.3 x BPS medium supplemented with glucose

a ⁾ Autoclaved separately. b ⁾ 10 mg/ml phenol red solution, pH-adjusted to 8.2 with NaOH.

The components were added to pre-heated water and the components were swelling for 10 minutes before the medium was autoclaved. pH of the medium was adjusted after autoclaving with HCl or NaOH until orange color was obtained which occurs at pH =7 when phenol red is used as pH-indicator.

Purification of BE- 14106

Harvesting and homogenization of cell mass

The cell mass in the production culture was harvested by centrifugation and freeze dried. Freeze dried pellet was homogenized with magnetic iron beads until fine pellet.

Crude purification of BE- 14106

The freeze dried cell pellet was extracted with 240 ml methanol/g for 1 hour. Glass beads were added to increase shear forces. Cell pellet was removed by centrifugation followed by filtration to remove all insoluble matter. The clear supernatant was added water and kept on ice for approximately 30 minutes in order to precipitate BE-14106. The precipitate was collected by centrifugation, washed with water to remove remaining methanol and freeze-dried. The freeze dried product represents a crude product.

Preparative HPLC purification of crude product

The crude product was dissolved in DMSO and the purification was performed on a reverse-phase column.

Preparative method: BIOP PREP BE14106_KFD.M HPLC system: Agilent 1100 series preparativ HPLC with fraction collection system

Column: PREP-C18, 10 μm, 50x250 mm (PN410910)

Column temperature: Ambient

Mobile phase: 10 mM ammonium acetate pH 4.0 (A) and methanol (B).

Time % B

0.00 85

8.50 85

8.60 100

10.60 100

10.70 85

13.00 85

Eluent flow: 85 ml/min

The fractions were added 1 % of a 2 M NH3 solution and stored at -2O ⁰ C.

Concentration of product in the preparative HPLC-fractions

Most of the methanol in the fractions was vaporized at a rotational vacuum evaporator at 5O ⁰ C. The remaining water phase containing BE-14106 was frozen to increase the precipitation yield and the precipitate was pelleted by centrifugation. The pellet was washed with water and freeze dried to give the final product.

LC-DAD-TOF analysis of BE14106 purified by preparative HPLC Calculation of purity by LC-DAD-TOF

After purification by preparative LC, BE-14106 was shown to constitute > 99 % of compounds in the sample which is absorbing at 291 nm as determined by UV and TOF data. It is assumed that the extinction coefficient for the contaminants is the same as for BE-14106.

TOF-MS data of BE-14106

The TOF-MS enable a LCTOF plot of purified BE-14106 to be obtained. From this a theoretical accurate m/z (negative ion) of BE-14106 (C ₂₇ H ₇₃ NO ₃ ) is 422.2701. This m/z was observed with acceptable accuracy and the 422 peak correlates well with the heptaene UV-peak.

LCTOF method: BIOP BE14106 SE.M

Column: Zorbax Bonus-RP 2.1χ50mm, 3.5μm (Agilent Technologies). Mobile phase A: 10 mM ammonium acetate (Riedel-de-Haen Cat#: 34674), Mobile phase B: 100% acetonitrile supergrade (Labscan UN 1648) Flow: 0.3 ml/min Column temperature: Ambient

Time %JB

0.00 40

10.00 70

10.10 90

12.00 90

12.10 40

17.00 40

TOF-MS parameters:

Negative API-ES ionization Drying gas: 10 1/min Nebulizer pressure: 40 psig

Drying gas temp.: 350°C Capillary voltage: 3000 V Fragmentor: 200 V

Example 9 - Characterisation of strain DSM 21069 antifungal activity

The strain DSM 21069 was investigated for production of antifungal activity. Growth conditions of 25 ⁰ C on medium PM2 (Bredholt et. al, 2008, Marine Drugs, 6(1) pp. 12-24) for 7 days, resulted in strong antifungal activity. After incubation the medium was dried and then extracted with DMSO. After filtration, the DMSO extracts were used as samples in a robotic bioassay procedure with the strains

Candida albicans CCUG3943 and C. glabrata CCUG3942 as indicator organisms.

The latter strain has a high level of resistance against polyene antibiotics, while the C. albicans strain is sensitive to polyenes. The medium used in the bioassay was AM19(B) (9.4 g/1 peptone [Oxoid], 4.7 g/1 yeast extract [Oxoid], 2.4 g/1 beef extract [Difco], 10 g/1 glucose [BDH], distilled water). Samples of DMSO-extracts with interesting bioactivity were fractionated using an Agilent 1100 series HPLC system equipped with a diode array detector (DAD) and a fraction collector. Each sample was fractionated in parallel using 2 different types of LC-columns: Agilent ZORBAX Eclipse XDB-Cl 8, 5 um, 4.6 x 150 mm and Agilent SB-CN 3.5 um, 4.6 x 75 mm. For both types of columns, a flow of 1 ml/min of a mixture of 0,005% formic acid in deionized water and acetonitrile was used as mobile phase. In both cases the concentration of acetonitrile was kept at 40% the first minute, then increased linearly from 40 to 95 % during the next 9 minutes and kept at a concentration of 95% for the rest of the run. The fraction collector was used to collect 12 fractions of the eluent from 1 minute until 13 minutes from injection.

The samples were dryed in a SpeedVac instrument (Thermo Scientific), dissolved in DMSO and the bioactivity of the fractions was measured (assay described above).

The fractions with bioactivity was analysed using an Agilent 1100 series HPLC system connected to a diode array detector (DAD) and a time of flight (TOF) mass spectrometer. The same columns and buffers were used in this analysis as described above for the fractionation step. Electrospray ionization was performed in the negative (ESI-) mode. The DAD plots were used to identify the approximate retention time of the bioactive compounds in the fractionation runs compounds and in the LC-MS-TOF analysis. Molecular masses corresponding to significant peaks identified in bioactive samples from parallel fractionations (C 18 and CN columns) were compared and molecular masses common to fractions from the Cl 8 and CN columns were identified. These molecular masses (10 ppm window) were submitted to the online version of the Dictionary of Natural Products (http://dnp.chemnetbase.com/) in order to search for previously characterized compounds with bioactivity.

In LCMS analysis of the bioactive fractions, significant peaks with a molecular mass corresponding to Antibiotic BEl 4106 was identified. The molecular mass observed in the LC-MS-TOF analysis was within 1 ppm of the molecular mass given in DNP (Accurate mass 423.277344). In addition, the DAD-profiles of these extracts were compared to the information given in DNP about the UV-absorbance spectra of Antibiotic BEl 4106. A good correlation was observed between the data given in DNP and the DAD profile of the compound identified in the extracts.

Example 10 - Characterisation of BecL BecO, BecR, BecC and BecP functions In order to verify the roles of certain genes in the biosynthesis of BE- 14106, a series of gene inactivation experiments was carried out. As described above (Example 7), a gene inactivation experiment using PCR amplified fragments from cosmids 2 and 3 has also been accomplished. Sequencing of these fragments and comparison with the BE-14106 cluster showed both of these fragments to be a part of the becA gene, the 1.1 kb fragment encoding parts of the KS and AT domains of module 2 and the 3.7 kb fragment encoding the KS, AT and DH domains of module 2. The production of BE-14106 was clearly affected in both mutants (Table 9).

Construction of Vectors for Gene Inactivation Experiments bed replacement vector: The 3.63 kb BgI ll-Kpn I fragment from cosmid 2 was cloned into pGEM3Zf(-) digested with BamH l-Kpn I, resulting in construct pBIRl . From this construct a 0.8 kb Nru 1-FspA I fragment was removed and the construct was religated, resulting in construct pBIR2. From the new construct a 2.86 kb EcoR \-Hind III fragment was excised and ligated with a 3.11 kb EcoR l-Hind III fragment of pSOK201 , resulting in the bed replacement vector, pBIR3.

becO replacement vector: The 14.77 kb Sph I fragment from cosmid 4 was cloned into pGEM3Zf(-) digested with Sph I, resulting in construct pBORl A. From pBORl A a 3.71 kb Sph \-Xba I fragment was excised and ligated into the Sph \-Xba I digested pGEM3Zf(-), yielding construct pBORIB. The 6.37 kb EcoN l-Age I fragment from pBORIB was treated with Klenow to fill in ends and religated as construct pBOR2. A 3.23 kb EcoR l-Hind III fragment was excised from construct

pBOR2 and ligated with the 3.11 kb EcoR l-Hind III fragment from pSOK201, resulting in the becO replacement vector pBOR3.

becR replacement vector: The 4.35 kb Hind Ul-B amH I fragment from cosmid 1 was cloned into pGEM3Zf(-) digested with Hind UI-BamH I, resulting in construct pBRRl. A 0.4 kb SnaB l-BsaB I fragment was removed from pBRRl and religation resulted in the new construct, pBRR2. The 3.96 kb EcoR l-Hind III fragment from construct pBRR2 was ligated with the 3.11 kb EcoR l-Hind III fragment of pSOK201, resulting in the becR replacement vector, pBRR3.

becC replacement vector: The 5.24 kb BamH l-Sac I fragment from cosmid 4 was cloned into pGEM3Zf(-) digested with BamH l-Sac I, yielding construct pBCRl . The 7.48 kb Not 1-Acc65 I fragment from pBCRl was treated with Klenow to fill in ends and religated as construct pBCR2. A 4.33 kb EcoR l-Hind III fragment was excised from construct pBCR2 and ligated with the 3.11 kb EcoR l-Hind III fragment from pSOK201 , resulting in the becC replacement vector pBCR3.

becP replacement vector: The 6.49 kb Sac l-Sph I fragment from cosmid 4 was ligated into pGEM3Zf(-) digested with Sac l-Sph I, resulting in construct pBPRl A. A 3.74 kb Hind Hl-BcI I fragment was excised from pBPRl A and ligated into pLITMUS28 digested with Hind lll-BamH I, resulting in construct pBPRIB. The 5.85 kb Xmn l-BbvC I fragment from pBPRIB was treated with Klenow to fill in ends and religated as construct pBPR2. The 1.82 kb EcoR l-Apa I fragment and 1.23 kb Hind lll-Apa I fragment from pBPR2 was ligated with the 3.11 kb EcoR l-Hind III fragment from pSOK201 , resulting in the becP replacement vector pBPR3.

All replacement vectors were introduced into ET12567(pUZ8002) and then used for conjugation with Streptomyces sp. DSM 21069 following the procedure described by Flett et al, 1997 (FEMS Microbiol. Lett., 155, pp, 223,-229), but with the donor cells grown to an OD ₆₀₀ of 0.4-0.5 and the heat shock time reduced to 5 min. Antibiotics were added after 24 hrs incubation.

BecR was initially assigned a putative role of linking together the C20-C25 acyl side chain made by BecA and the macrolactam ring. The mutant, verified by a

00759

- 57 -

Southem blot analysis, was tested for BE-14106 production by LC-MS of fermentation extracts. The BE-14106 production was not affected in the AbecR mutant, implying that it is not involved in the biosynthesis.

In addition to the role of BecR there were also questions about the roles of Bed, BecC and BecP in the biosynthesis of BE-14106, and the suggested role of BecO as a C-8 hydroxylase needed to be verified. Using the above vectors, second crossover mutants were obtained for all genes and verified by Southern blot analyses. The BE-14106 production for each mutant strain was tested by LC-MS of fermentation extracts. For the AbecO mutant, the expected mass corresponding to the stoichiometric formula (C27H37NO ₂ ) of the suggested 8-deoxy BE- 14106 was found with 1.0 ppm difference from the theoretical mass, thereby confirming the role of BecO as a P450 monooxygenase hydroxylating BE-14106. The C-8 carbon represents the only likely target of BecO, since the hydroxyl group at the C-9 appears due to the lack of the DH domain activity in BecE PKS involved in the biosynthesis of the macrolactam ring.

LC-MS analyses of fermentation extracts from the Abed, AbecC and AbecP mutants all showed complete absence of BE-14106 production, and no putative BE- 14106 analogues/precursors could be identified. This might indicate that all three enzymes function very early in the biosynthesis, presumably being involved in the synthesis of the starter aminoacyl unit.

Example 11 - Feeding studies to determine amino acid incorporation to BE-14106 Feeding studies of DSM 21069 were performed using a defined production medium based on the ¹⁵ N Silantes OD2 medium (Silantes pr. No. 103202). The composition of the media was ¹⁵ N Silantes OD2 medium 536 ml/1 and in addition g per 1: MgSO ₄ x7H ₂ O, 0.4; CaCO ₃ , 5.0; ( ¹⁵ NH ₄ ) ₂ SO ₄ , 0.54; KH ₂ PO4, 0.2 and glucose, 10. The medium was supplemented with trace mineral solution TMSl (Borgos et al., 2006, Arch. Microbiol., 185, pp. 165-171) 3 ml/1. Incorporation of unlabeled amino acids was tested by adding 0.14 g/1 D-asparagine or 0.06 g/1 glycine or 0.10 g/1 Na- glutamate or no addition of unlabeled amino acid. The production cultures were inoculated with 3 % of a 0.5 x TSB pre-culture cultivated as described above, except that the cells were washed with the production medium once before inoculation to

remove components from the pre-culture. Both the production culture and the pre- culture were cultivated in baffled shake flasks with glass beads as described above. Quantitative and qualitative LC-MS analyses of BE- 14106 and 8-deoxy BE- 14106 were performed on methanol extracts from culture pellets using an Agilent 1100 series HPLC system connected to a diode array detector (DAD) and a TOF mass spectrometer. Electrospray ionization was performed in the negative (ESI-) or positive (ESI+) mode, essentially as described previously (Bruheim et al, 2004, Antimicrob. Agents Chemother, 48, pp. 4120-4129), but with the following modifications: LC separation were performed on an Agilent ZORBAX Bonus-RP 2.1x50 mm column. The acetonitrile concentration was increased linearly from 40 to 70% for the first 10 min and was then kept at a concentration of 90 % for the rest of the run. Concentrations of BE-14106 were determined by UV peak absorption at 291 nm using BE-14106 purified by preparative HPLC as a standard.

All nitrogens in the media before addition of the amino acids to be tested for incorporation in the biosynthesis were ¹⁵ N isotope labeled. The amino acids to be tested were added as ¹⁴ N. In addition to D-asparagine and glycine, addition of L- glutamate and no addition of extra amino acids were used as controls in the experiment. The concentration of relevant ¹⁵ N amino acids in the media was: asparagine 0 μM (aspartate 54 μM), glycine 29 μM (serine 11 μM) and glutamate 29 μM. The addition of the unlabeled amino acids were added to 20 x concentration of ¹⁵ N aspartate, 27 x concentration of ¹⁵ N glycine and 20 x concentration of ¹⁵ N glutamate for D-aspraragine, glycine and glutamate respectively. The production cultures were extracted in DMSO and the extracts were analyzed on LC-DAD-TOF. The BE-14106 molecule contains one nitrogen atom and it is therefore expected that if N from one of the added ¹⁴ N amino acids is incorporated, a BE-14106 molecule with an accurate mass (M-H) of 422.27 will be observed in a LC-TOF-spectra whereas ¹⁵ N labeled BE-14106 will have the accurate mass (M-H) of 423.28. The production cultures were extracted in DMSO and the extracts were analyzed on LC- DAD-TOF. The TOF -mass spectra showed that addition of unlabeled glycine resulted in 20 % incorporation of ¹⁴ N, whereas the incorporation ratio by addition of D-aspragine and glutamate were 3 %, which is the same as for the control.