Login| Sign Up| Help| Contact|

Patent Searching and Data


Title:
HYBRID GLYCOSYLATED PRODUCTS AND THEIR PRODUCTION AND USE
Document Type and Number:
WIPO Patent Application WO/2001/079520
Kind Code:
A1
Abstract:
The present invention relates to hybrid glycosylated products, and in particular, to natural products such as polyketides and glycopeptides, and to processes for their preparation. The invention is particularly concerned with recombinant cells in which a cloned microbial glycosyltransferase can be conveniently screened for its ability to generate specific glycosylated derivatives when supplied with polyketide, peptide, or polyketide-peptides as substrates. The invention demonstrates that cloned glycosyltransferases when rapidly screened for their ability to attach a range of activated sugars to a range of exogenously supplied or endogenously generated aglycone templates, show a surprising flexibility towards both aglycone and sugar substrates, and that this process allows the production of glycosylated polyketides in good yield. This overcomes the problem not only of supplying novel sugar attachments to individual polyketides, including polyketides altered by genetic engineering, but also of increasing the diversity of polyketide libraries by combinatorial attachment of sugars.

Inventors:
LEADLAY PETER FRANCIS (GB)
STAUNTON JAMES (GB)
GAISSER SABINE (GB)
Application Number:
PCT/GB2001/001743
Publication Date:
October 25, 2001
Filing Date:
April 17, 2001
Export Citation:
Click for automatic bibliography generation   Help
Assignee:
BIOTICA TECH LTD (GB)
LEADLAY PETER FRANCIS (GB)
STAUNTON JAMES (GB)
GAISSER SABINE (GB)
International Classes:
C12N15/09; A61K31/7048; A61P31/04; A61P31/10; A61P37/02; C07H17/08; C12N1/21; C12N9/10; C12N15/52; C12N15/54; C12P19/62; C12R1/01; (IPC1-7): C12P19/00; C12P19/62; C07H17/08; C12N1/00; C12N1/21; C12N15/54; C12N15/76
Domestic Patent References:
WO1997023630A21997-07-03
WO1996040968A11996-12-19
WO1991016334A11991-10-31
WO1998049315A21998-11-05
WO1993010248A11993-05-27
Foreign References:
US3697547A1972-10-10
US4439426A1984-03-27
EP0098732A11984-01-18
GB2256197A1992-12-02
Other References:
M. DOUMITH ET AL.: "Interspecies complementation in Saccharopolyspora erythraea.", MOL. MICROBIOL., vol. 34, no. 5, 1999, pages 1039 - 1048, XP001021052
S. GAISSER ET AL.: "A defined system for hybrid macrolide biosynthesis in Saccharopolyspora erythraea.", MOL. MICROBIOL., vol. 36, no. 2, 2000, pages 391 - 401, XP001021072
K. MILLER ET AL.: "Purification, cloning and heterologous expression of a catalytically efficient flavonol 3-O-galactosyltransferase.", J. BIOL. CHEM., vol. 274, no. 48, 1999, pages 34011 - 34019, XP001021051
See also references of EP 1278881A1
Attorney, Agent or Firm:
Kiddle, Simon J. (Mewburn Ellis York House 23 Kingsway London Greater London WC2B 6HP, GB)
Download PDF:
Claims:
Claims :
1. A process for producing a hybrid glycosylated product by transferring one or more sugar moieties to an aglycone template, the process comprising : transforming microorganism host cells with nucleic acid encoding a glycosyltransferase (GT) ; and, providing an aglycone template to the GT so that GT transfers one or more sugar moieties to the aglycone template to produce a hybrid glycosylated product ; wherein one or more of the sugar moiety or moieties, the aglycone template, the GT or the host cells are heterologous to the other components.
2. The process of claim 1, wherein the aglycone template and the sugar moiety or moieties are heterologous to each other.
3. The process of claim 1, wherein the aglycone template and/or the sugar moiety or moieties are heterologous to the host cells.
4. The process of claim 1, wherein the aglycone template, the sugar moiety or moieties and the GT are heterologous to the host cells.
5. The process of claim any one of claims 1 to 4, wherein the host cell is transformed with a gene or genes for producing the sugar moiety.
6. The process of any one of claims 1 to 5, wherein the glycosyltransferase is : (a) from the erythromycin pathway of Saccharopolyspora erythraea, desosaminyltransferase eryCIII or mycarosyltransferase eryBV ; (b) from the megalomycin pathway of Micromonospora megalomicea, desosaminyltransferase megCIII, mycarosyltransferase megBV or megosaminyltransferase ; (c) from the oleandomycin pathway of Streptomyces antibioticus, oleandrosyltransferase oleG2 (also transfers rhamnose and olivose) or desosaminyltransferase oleGl ; (d) from the tylosin pathway of Streptomyces fradiae, mycaminosyltransferase tyIMII, deoxyallose transferase tylN or mycarosyltransferase tylCV ; (e) from the midecamycin pathway of Streptomyces mycarofaciens, mycaminosyltransferase midI, deoxyallose transferase or mycarosyltransferase ; (f) from the pikromycin/narbomycin pathway of Streptomyces venezuelae, desosaminyltransferase desVII ; (g) from the spinosyn pathway of Saccharopolyspora spinosa, rhamnosyltransferase or forosaminyltransferase ; (h) from the amphotericin pathway of Streptomyces nodosus, mycaminosyltransferase amphDI ; (i) from the avermectin pathway of Streptomyces avermitilis, oleandrosyltransferase ; (j) from the nystatin pathway of Streptomyces, mycaminosyltransferase, (k) from the polyene 67121 C pathway of Actinoplanes caerulens, mycosaminyltransferase, mannosyltransferase (transferring to the mycosamine) ; (1) from the elloramycin pathway of Streptomyces olivaceaous Tu2353, rhamnosyltransferase elmGT ; (m) from the mithramycin pathway of Streptomyces argillaceus, olivosyltransferase mtmGIV ; (n) from the daunomycin pathway of Streptomyces peucetius, daunosaminyltransferase dnrS ; (o) from the urdamycin pathway of Streptomyces fradiae Tu2717, rhodinosyltransferase urdGTlc, olivosyltransferase urdGTlb, rhodinosyltransferase urdGTla and olivosyltransferase urdGT2.
7. The process of any one of the preceding claims, further comprising employing an enzyme for modifying the sugar moiety and/or the aglycone template, either before or after attachment of the sugar moiety to the aglycone template.
8. The process of claim 7, wherein the enzyme is a methyltransferase or a P450.
9. The process of claim 7 or claim 8, wherein the host cells is transformed with a heterologous gene encoding said enzyme.
10. The process of any one of the preceding claims, further comprising transforming the host cells with a recombinant polyketide synthase gene or genes so that the aglycone template is produced by the host cells.
11. The process of claim 10, wherein the recombinant PKS genes are selected from the group consisting of natural PKS genes, mutated versions of natural PKS genes, hybrid PKS genes consisting of portions from at least two different natural Type I PKS gene clusters, natural Type II PKS gene clusters and a library of hybrid PKS genes of either Type I or Type II.
12. The process of claim 12, wherein the macrolide aglycone template is 6deoxy erythronolide B, erythronolide B or tylactone, or derivatives thereof.
13. The process of any one of the preceding claims, wherein the aglycone template is selected from the group consisting of a polyketide, a mixed polyketidepeptide and a peptide.
14. The process of any one of the preceding claims, wherein the aglycone template is a polyketide.
15. The process of claim 15 + wherein the polyketide is selected from the group consisting of a Type I and Type II polyketides.
16. The process of claim 14 or claim 15, wherein the PKS gene or genes produce a macrolide aglycone template.
17. The process of any one of the preceding claims which comprises supplying one or more exogenous aglycone templates to the cells.
18. The process of any one of the preceding claims which comprises employing one or more endogenous aglycone templates.
19. The process of any one of the preceding claims, further comprises deleting or inactivating one or more genes in the microorganism host cells involved in the production of the aglycone template and/or in its subsequent processing, thereby to suppress or alter the production of the natural aglycone template or product.
20. The process of any one of claims 1 to 20, comprising the steps of : producing one of the aglycone template or the sugar moiety in first host cells as a first product ; optionally, purifying the first product from a culture of the first host cells ; and, adding the first product to a second host cell comprising one or more genes encoding the other of the aglycone template or the sugar moiety and one or more glycosyltransferases, so that the sugar moiety is transferred to the glycosyltransferase, to produce the hybrid glycosylated product.
21. The process of claim 21, wherein the first product is a filtered supernatant from the culture of the first host cells.
22. The process of claim 21, wherein the first product is isolated from the culture of the first host cells.
23. A hybrid glycosylation product as obtainable by the process. of any one of claims 1 to 23.
24. A hybrid glycosylated products which comprises : (a) one or more natural sugars linked to an erythronolide at the 7position ; or, (b) one or more rhamnose or substituted (e. g. methyl) rhamnose sugars linked to an erythronolide ; or, (c) one or more mycarose or substituted mycarose sugars linked to an erythronolide ; and combinations of (a), (b) and (c) sugar substituents on an erythronolide.
25. A hybrid glycosylated products which comprises : (a) one or more natural sugars linked to erythromycin at the 7position ; or, (b) one or more mycarose or substituted mycarose sugars linked to an erythromycin ; or, (c) one or more mycaminose or substituted mycaminose sugars linked to an erythromycin ; and combinations of (a), (b) and (c) sugar substituents on an erythromycin.
26. A hybrid glycosylated products which comprises : (a) one or more glucose or substituted glucose sugars linked to a tylactone ; or, (b) one or more desosaminose or substituted desosaminose sugars linked to a tylactone ; (c) one or more mycaminose or substituted mycaminose sugars linked to a tylactone ; or (d) one or more rhamnose or substituted rhamnose sugars linked to a tylactone ; and combinations of (a), (b), (c) and (d) sugars substituents such as a rhamnose and a mycaminose sugar linked to a tylactone.
27. A compound which is : 30 (2'Omethylrhamnosyl) erythronolide B, 30 (2', 3'bisOmethylrhamnosyl) erythronolide B, 30 (2', 3', 4'trisOmethylrhamnosyl) erythronolide B, 3Omycarosylerythronolide B, 8ahydroxy3Omycarosyl erythronolide B, 8, 8aepoxy3Omycarosyl erythronolide B, 8, 8adehydro6deoxyerythronolide B, 8hydroxy6deoxyerythronolide B, <BR> <BR> <BR> 30(2'Omethylrhamnosyl) erythromycin D,<BR> <BR> <BR> <BR> <BR> 30 (2', 3'bis0methylrhamnosyl) erythromycin D, 30 (2', 3', 4'trisOmethyl rhamnosyl) erythromycin D, 5Omycaminosylerythromycin A, 5Omycaminosyl4"Omycarosyl erythromycin A, <BR> <BR> <BR> 5Oglucosyltylactone,<BR> <BR> <BR> <BR> <BR> 5Odesosaminyltylactone, 23Orhamnosyl 50mycaminosyl tylactone, 50 (2'O)bisglucosyltylactone, 3Orhamnosyl8, 8adehydro6deoxyerythronolide B, 3Orhamnosyl8, 8adihydroxyerythronolide B, or 3, 5 diOmycarosyl erythronolide B.
28. A host cell transformed with nucleic acid encoding a glycosyltransferase (GT), wherein the GT is heterologous to the host cells and transfers one or more sugar moieties to an aglycone template within the cells to produce a hybrid glycosylated product.
29. The host cell of claim 29, wherein the host cell is further transformed with one or more auxiliary genes.
30. The host cell of claim 30, wherein the auxiliary gene is a sugar pathway gene encoding a protein involved in the biosynthesis of a sugar moiety, thereby enabling a host cell transformed with the expression cassette to produce sugar moieties for subsequent transfer to an aglycone template.
31. The host cell of any one of claims 29 to 31 which is a strain of actinomycete,.
32. The host cell of claim 32, wherein the actinomycete strain is selected from the group consisting of Saccharopolyspora erythraea, Streptomyces coelicolor, Streptomyces avermitilis, Streptomyces griseofuscus, Streptomyces cinnamonensis, Streptomycesfradiae, Streptomyces longisporoflavus, Streptomyces hygroscopicus, Micromonospora griseorubida, Streptomyces lasaliensis, Streptomyces venezuelae, Streptomyces antibioticus, Streptomyces lividans, Streptomyces rimosus, Streptomyces albus, Amycolatopsis mediterranei, and Streptomyces tsukubaensis.
33. A process for producing a hybrid glycosylated product, the process comprising culturing the host cell of any one of claims 29 to 33 and isolating the product thus produced.
34. A process for producing a library which comprises a plurality of hybrid glycosylated products, the process comprising : transforming microorganism host cells with nucleic acid encoding one or more glycosyltransferases (GT) ; and, providing one or more aglycone templates to the GTs so that the GTs transfer one or more sugar moieties to the aglycone templates to produce said plurality of hybrid glycosylated products ; wherein one or more of the sugar moiety or moieties, the aglycone template, the glycosyltransferase or the host cells are heterologous to the other components.
35. The process of claim 35, wherein the host cell is further transformed with one or more auxiliary genes.
36. The host cell of claim 36, wherein the auxiliary gene is a sugar pathway gene encoding a protein involved in the biosynthesis of a sugar moiety, thereby enabling a host cell transformed with the expression cassette to produce sugar moieties for subsequent transfer to an aglycone template.
37. The process of any one of claims 35 to 37, further comprising screening the library for a hybrid glycosylated product having a desired characteristic.
38. The process of any one of claims 35 to 38, wherein the library comprises at least two different hybrid glycosylated products.
39. The process of any one of claims 35 to 38, wherein the library comprises at least 10 different hybrid glycosylated products.
40. The process of any one of claims 35 to 38, wherein the library comprises at least 100 different hybrid glycosylated products.
41. The process of any one of claims 35 to 41, further comprising isolating a host cell producing a desired hybrid glycosylated product.
42. The process of claim 42, further comprising culturing the host cells and isolating the hybrid glycosylated product thus produced.
43. The process of any one of claims 35 to 43, wherein endogenous polyketide biosynthesis is suppressed.
44. The process of claim 44, wherein the cells are cultured in the presence of a suppressor of endogenous polyketide synthesis.
45. The process of claim 45, wherein endogenous polyketide biosynthesis is suppressed by mutating, deleting or inactivating one or more of the PKS genes naturally present within the cells.
46. An expression cassette comprising one or more glycosyltransferase genes and one or more auxiliary genes, operably linked under the control of a promoter.
47. The expression cassette of claim 47, wherein the auxiliary gene is a sugar pathway gene encoding a protein involved in the biosynthesis of a sugar moiety, thereby enabling a host cell transformed with the expression cassette to produce sugar moieties for subsequent transfer to an aglycone template.
48. The expression cassette of claim 48, wherein the auxiliary gene encodes an enzyme involved in the processing of a sugar moiety or an aglycone template, either before or after the sugar moiety is transferred to the aglycone by the glycosyltransferase.
49. The expression cassette of claim 49, wherein the enzyme is a methyltransferase or a P450 enzyme.
50. The expression cassette of any one of claims 47 to 50, wherein the genes are linked in a contiguous head to tail assembly.
51. The expression cassette of any one of claims 47 to51, wherein the gene or genes are introduced into the cassette by : introducing Xbal restriction sites at the 3'and 5'ends of a PCR fragment comprising the gene or genes ; and cloning the XbaI flanked fragment into a host strain with an active Dam methylase.
52. The expression cassette of any one of claims 47 to 52, wherein the genes are under the control of a single promoter.
53. The expression cassette of claim 53, wherein the promoter is a strong promoter.
54. The expression cassette of claim 54, wherein the genes are under the control of the actIIOrf4 regulator.
55. The expression cassette of any one of claims 47 to 55, wherein the cassette comprises a nucleic acid sequence encoding a histidine tag adjacent the terminal gene in the expression cassette.
56. A process of producing an expression cassette of any one of claims 47 to 56, the process comprising operably linking the genes together under the control of a promoter.
57. The process of claim 57, further comprising transforming a host cell with the expression cassette and expressing the genes comprised within it to produce the GT and proteins encoded by the auxiliary genes.
58. A host cell produced by the process of claim 58.
Description:
Hybrid Glycosylated Products and Their Production and Use Field of the Invention The present invention relates to hybrid glycosylated products, and in particular, to natural products such as polyketides and glycopeptides, and to processes for their preparation. The invention is particularly concerned with recombinant cells in which a cloned microbial glycosyltransferase can be conveniently tested for its ability to generate specific glycosylated derivatives when supplied with polyketide, peptide, or polyketide-peptides as substrates.

Background to the Invention Glycosylation is important for the bioactivity of many natural products, including antibacterial compounds such as the polyketide erythromycin A and the glycopeptide vancomycin, and antitumour compounds such as the aromatic polyketide daunorubicin and the glycopeptide-polyketide bleomycin. Polyketides are a large and structurally diverse class of natural products that includes many compounds possessing antibiotic or other pharmacological properties, such as erythromycin, tetracyclines, rapamycin, avermectin, monensin, epothilones and FK506. In particular, polyketides are abundantly produced by Streptomyces and related actinomycete bacteria. They are synthesised by the repeated stepwise condensation of acylthioesters in a manner analogous to that of fatty acid biosynthesis. The greater structural diversity found among natural polyketides arises from the selection of (usually) acetate or propionate as"starter"or"extender"units ; and from the differing degree of processing of the p-keto group observed after each condensation. Examples of processing steps include reduction to (3-hydroxyacyl-, reduction followed by dehydration to 2-enoyl-, and complete reduction to the saturated acylthioester. The stereochemical outcome of these processing steps is also specified for each cycle of chain extension. The polyketide chains are usually cyclised in specific ways and subject to further enzyme-catalysed modifications to produce the final polyketide.

Naturally-occurring peptides produced by non-ribosomal peptide synthetases are likewise synthesised by repeated stepwise assembly, in this case of activated amino

acids, and the chains produced are similarly subject to further modifications to produce the fully bioactive molecules. Mixed polyketide-peptide compounds, hereinafter defined as incorporating both ketide and amino acid units, are also known and their bioactivity is also influenced by their pattern of glycosylation and other modification. The compounds so produced are particularly valuable because they include large numbers of compounds of known utility, for example as anthelminthics, insecticides, immunosuppressants, antifungal or antibacterial agents.

Streptomyces and closely-related genera of filamentous bacteria are abundant producers of polyketide metabolites. Although large numbers of therapeutically important polyketides have been identified, there remains a need to obtain novel polyketides that have enhanced properties or that possess completely novel bioactivity. The inexorable rise in the incidence of pathogenic organisms with resistance to antibiotics such as 14-membered macrolides or glycopeptides represents a significant threat to human and animal health. Current methods of obtaining novel polyketide metabolites include large-scale screening of naturally-occurring strains of Streptomyces and other organisms, either for direct production of useful molecules, or for the presence of enzymatic activities that can bioconvert an existing polyketide, which is added to the growth medium, into specific derivatives. These procedures are time-consuming and costly, and biotransformation using whole cells may in addition be limited by side-reactions or by a low concentration or activity of the intracellular enzyme responsible for the bioconversion. Given the complexity of bioactive polyketides, they are not readily amenable to total chemical synthesis in large scale.

Chemical modification of existing polyketides has been widely used, but many desirable alterations are not readily achievable by this means.

Meanwhile, methods have been developed for the biosynthesis of altered polyketides and non-ribosomally-synthesised polypeptides by the engineering of the corresponding genes encoding the polyketide synthases and polypeptide synthetases respectively. The biosynthesis of polyketides is initiated by a group of chain-forming enzymes known as polyketide synthases. Two classes of polyketide synthase (PKS)

have been described in actinomycetes. One class, named Type I PKSs, represented by the PKSs for the macrolides erythromycin, oleandomycin, avermectin and rapamycin, consists of a different set or"extension module"of enzymes for each cycle of polyketide chain extension (Cortes, J. et al. Nature (1990) 348 : 176-178). The term "extension module"as used herein refers to the set of contiguous domains, from a ß-ketoacyl-ACP synthase ("KS") domain to the next acyl carrier protein ("ACP") domain, which accomplishes one cycle of polyketide chain extension.

In-frame deletion of the DNA encoding part of the ketoreductase domain in module 5 of the erythromycin-producing PKS (also known as 6-deoxyerythronolide B synthase, DEBS) has been shown to lead to the formation of erythromycin analogues 5, 6-dideoxy-3-a-mycarosyl-5-oxoerythronolide B, 5, 6-dideoxy-5-oxoerythronolide B and 5, 6-dideoxy, 6 ß-epoxy-5-oxoerythronolide B (Donadio, S. et al. Science (1991) 252 : 675-679). Likewise, alteration of active site residues in the enoylreductase domain of module 4 in DEBS, by genetic engineering of the corresponding PKS-encoding DNA and its introduction into Saccharopolyspora erythraea, led to the production of 6, 7-anhydroerythromycin C (Donadio, S. et al. Proc Natl. Acad. Sci.

USA (1993) 90 : 7119-7123). WO 93/13663 describes additional types of genetic manipulation of the DEBS genes that are capable of producing altered polyketides.

However many such attempts are reported to have been unproductive (Hutchinson, C.

R. and Fujii, I. Annu. Rev. Microbiol. (1995) 49 : 201-238, at p. 231).

WO 98/01546 describes the engineering of hybrid Type I PKS genes which utilise portions of PKS genes derived from more than one natural PKS, particularly derived from different organisms, and the use of such recombinant genes for the production of altered polyketide metabolites.

The second class of PKS, named Type II PKSs, is represented by the synthases for aromatic compounds. Type II PKSs contain only a single set of enzymatic activities for chain extension and these are re-used in successive cycles (Bibb, M. J. et al.

EMBO J. (1989) 8 : 2727-2736 ; Sherman, D. H. et al. EMBO J. (1989) 8 : 2717-2725 ;

Fernandez-Moreno, M. A. et aL J. Biol. Chem. (1992) 267 : 19278-19290). The "extender"units for the Type II pKSs are usually acetate units, and the presence of specific cyclases dictates the preferred pathway for cyclisation of the completed chain into an aromatic product (Hutchinson, C. R. and Fujii, 1. Annu. Rev. Microbiol.

(1995) 49 : 201-238). Hybrid polyketides have been obtained by the introduction of clones Type II PKS gene-containing DNA into another strain containing a different Type II PKS gene cluster, for example by introduction of DNA derived from the gene cluster for actinorhodin, a blue-pigmented polyketide from Streptomyces coelicolor, into an anthraquinone polyketide-producing strain of Streptomyces galileus (Bartel, P.

L. et al. J. Bacteriol. (1990) 172 : 4816-4826).

The minimal number of domains required for polyketide chain extension on a Type II PKS when expressed in a Streptomyces coelicolor host cell (the"minimal PKS") has been defined for example in WO 95/08548 as containing the following three polypeptides which are products of the act I genes : first KS ; secondly a polypeptide termed the CLF with end-to-end amino acid sequence similarity to the KS but in which the essential active site residue of the KS, namely a cysteine residue, is substituted either by a glutamin residue, or in the case of the PKS for a spore pigment such as the whiE gene product (Chater, K. F. and Davis, N. K. Mol. Microbiol. (1990) 4 : I679-1691) by a glutamic acid residue ; and finally an ACP. The CLF has been stated for example in WO 95/08548 to be a factor that determines the chain length of the polyketide chain that is produced by the minimal PKS. However, it has been found (Shen, B. et al. J. Am. Chem. Soc. (1995) 117 : 6811-6821) that when the CLF for the octaketide actinorhodin is used to replace the CLF for the decaketide tetracenomycin in host cells of Streptomyces glaucescens, the polyketide product is not found to be altered from a decaketide to an octaketide. An alternative. nomenclature has been proposed in which KS is designated KSa and CLF is designated KS (3, to reflect this lack of confidence in the correct assignment of the function of CLF (Meurer, G. et al. Chemistry and Biology (1997) 4 : 433-443).

International Patent Application WO 00/00618 has recently shown that CLF and its counterpart in Type I PKS multienzymes, the so-called KSQ domain, are involved in

initiation of polyketide chain synthesis. WO 95/08548 for example describes the replacement of actinorhodin PKS genes by heterologous DNA from other Type II PKS gene clusters, to obtain hybrid polyketides.

This ability to engineer PKS genes of both Type I and Type II raises the possibility of combinatorial biosynthesis of polyketides to produce diverse libraries of novel natural products which may be screened for desirable bioactivities. However, the aglycones produced by the recombinant PKS genes may be only partially, or not at all, processed by glycosyltransferases and other modifying enzymes into analogues of the mature polyketides. There is therefore an additional need to provide processes for efficient conversion of such novel aglycones into specific glycosylated products. Further, the invention of efficient processes for glycosylation would provide a new means to increase very significantly the diversity of combinatorial polyketide libraries, by utilisation of recombinant cells containing alternative cloned glycosyltransferases and alternative complements of activated sugars.

The well-known influence of glycosylation on biological activity has encouraged intensive research into the genes and enzymes governing the synthesis and attachment of specific sugar units to polyketide and polypeptide metabolites (for a review see Trefzer, A. et al. Natural Products Reports (1999) 16 : 283-299). Surveys of such metabolites have revealed a high diversity in the type of glycosyl substitution that is found, including a very large number of different deoxyhexoses and deoxyaminohexoses (see for a review Liu, H.-W. and Thorson, J. S. Annu. Rev.

Microbiol. (1994) 48 : 223-256) review). The sequencing of biosynthetic gene clusters for numerous glycosylated polyketides and peptides has revealed the presence of such sugar biosynthetic genes, and also genes encoding the glycosyltransferases that transfer the glycosyl group from an activated form of the sugar, eg dTDP-or dUDP- forms, to the aglycone acceptor. For example, the eryB genes and the eryC genes of the erythromycin biosynthetic gene cluster in Saccharopolyspora erythraea have been identified as involved in the biosynthesis and attachment of respectively, L-mycarose and D-desosamine to the aglycone precursor of erythromycin A (Dhillon, N. et al.,

Mol. Microbiol. (1989) 3 : 1405-1414 ; Haydock et al. Mol. Gen. Genet. (1991) 230 : 120-128 ; Salah-Bey, K. et al. Mol Gen. Genet (1998) 257 : 542-553 ; Gaisser, S. et al., Mol. Gen. Genet. (1998) 258 : 78-88 ; Gaisser, S. et al. (1997) Mol. Gen. Genet.

256 : 239-251 ; Summers, D. et al. Microbiology (1997) 143 : 3251-3262). Both WO 97/23630 and WO 99/05283 describe the preparation of an altered erythromycin by deletion of a specific sugar biosynthetic gene, so that an altered sugar becomes attached to the aglycone. Thus WO 99/05283 describes low but detectable levels of erythromycins in which for example desosamine is replaced by mycaminose (eryCIV knockout), or desmethylmycarosyl erythromycins (eryBIII knockout) are produced.

Meanwhile methymycin analogues have been produced in which desosamine has been replaced by D-quinuvose (Borisova, S. A. et al. Org. Lett. (1999) 1 : 133-136), or through the incorporation of the calH gene of the calicheamycin gene cluster from Micrornonospora echinospora into the methymycin producing strain (Zhao, L. et al.

J. Amer. Chem. Soc. (1999) 121 : 9881-9882). Similarly, hybrid glycopeptides have been produced by using cloned glycosyltransferases from the vancomycin-producing Amycolatopsis orientais to add D-xylose or D-glucose to aglycones of closely-related glycopeptides according to US Patent 5, 871, 983 (1999) (Solenberg, P. et al. Chem.

Biol (1997) 4 : 195-202). Hybrid aromatic polyketides have also been produced, by interspecies complementation of a mutant individual sugar biosynthetic gene-with a similar gene with a different stereospecificity. Thus instead of the natural daunosamine, 4'epi-daunosamine is produced in recombinant Streptomyces peucetius and attached by the daunosamine glycosyltransferase to the aglycone to yield the antitumour derivative epirubicin in place of doxorubicin (Madduri, K. et al. Nature Biotechnology (1998) 16 : 69-74). In all these cases, the specificity of the glycosyltransferase allowed the substitution of an alternative activated sugar, but the aglycone and glycosyltransferase were not heterologous to each other. It has been found that when oleandrose glycosyltransferase oleG2 of Streptomyces antibioticus is cloned into the erythromycin-producing Saccharopolyspora erythraea, in addition to other products, the novel erythromycin in which cladinose/mycarose at C-3 is replaced by rhamnose, was obtained (Doumith, M. et al, Mol. Microbiol. (1999) 34 : 1039- 1048). It was assumed that the activated rhamnose is produced by the host cells, and

is recruited by the oleG2 glycosyltransferase in competition with the activated mycarose known to be present.

Summary of the Invention The present invention shows that cloned glycosyltransferases when rapidly screened for their ability to attach a range of activated sugars to a range of exogenously supplied or endogenously generated aglycone templates, show a surprising flexibility towards both aglycone and sugar substrates, and that this process allows the production of glycosylated polyketides in good yield. This overcomes the problem not only of supplying novel sugar attachments to individual polyketides, including polyketides altered by genetic engineering, but also of increasing the diversity of polyketide libraries by combinatorial attachment of sugars. It is particularly surprising that new glycosylated products can be produced in systems in which We or more of the components are heterologous to each other, the components being selected from the aglycone template, the sugar moiety or moieties, the glycosyltransferase, the host cell and/or genes encoding enzymes capable of modifying the sugar moiety, either before or after attachment to the aglycone template. In preferred embodiments, two, three, four or all of the components are heterologous to each other.

Accordingly, in a first aspect, the present invention provides a process for producing a hybrid glycosylated product by transferring one or more sugar moieties to an aglycone template, the process comprising : transforming microorganism host cells with nucleic acid encoding a glycosyltransferase (GT) ; and, providing an aglycone template to the GT so that the GT transfers one or more sugar moieties to the aglycone template to produce a hybrid glycosylated product ; wherein one or more of the sugar moiety or moieties, the aglycone template, the glycosyltransferase or the host cells are heterologous to the other components.

Preferably, the hybrid glycosylated product is other than compounds Ml to M4 disclosed in Doumith et al (supra), e. g. an erythromycin in which cladinose/mycarose

at C-3 is replaced by rhamnose.

In a further aspect, the present invention provides host cells transformed with nucleic acid encoding a glycosyltransferase (GT), wherein the GT is heterologous to the host cells and transfers one or more sugar moieties to an aglycone template within the cells to produce a hybrid glycosylated product.

In a further aspect, the present invention provides a process for producing a hybrid glycosylated product, the process comprising culturing the host cells defined above and isolating the product thus produced. In embodiments in which the aglycone template is supplied to the host cells, rather than being produced by the host cell, the process may comprise the additional step of supplying the aglycone template to the cells.

In further aspects, the present invention provides hybrid glycosylation products as obtainable by any of the processes disclosed herein.

In some embodiments of the present invention, a"hybrid glycosylated products one in which the aglycone template and the sugar moiety or moieties are heterologous to each other. In the processes described herein, one or more of the components of the system used to produce or modify the hybrid glycosylated product may be heterologous to one another. These components include the aglycone template, the sugar moiety or moieties, the microorganism strain/host cells, the glycosyltransferase which catalyses the attachment of the sugar moiety to the aglycone template and modifying genes capable of modifying the sugar moiety either before or after attachment to the aglycone template. The hybrid glycosylated product may also be subject of further processing or derivatisation, either by the strain or after isolating from culture medium.

In the present invention, an"aglycone template"is a polyketide, a peptide or a mixed polyketide-peptide which is capable of further processing, e. g. by a

glycosyltransferase, to transfer one or more activated sugar moieties to the template.

The aglycone template may include forms of glycosylation other than that introduced by the heterologous GT. The cells may additionally contain one or more heterologous modifying genes, including but not limited to genes encoding enzymes or other proteins capable of carrying out methyl transfer, hydroxylation, or epoxidation reactions, on sugar moiety, before or after attachment to the aglycone template.

Alternatively or additionally, further diversity in hybrid products can be obtained by deleting or modifying one or more homologous modifying genes.

In some embodiments, the aglycone template may be produced by the microorganism strain, either naturally or by transforming the strain with one or more genes or gene clusters capable of producing the template. By way of example, where a microorganism strain naturally produces a polyketide, the process may employ the polyketide aglycone template endogenously produced by the strain or the strain may be engineered to delete or inactivate the production of this template. In this latter case, the cells may be transformed with one or more PKS genes or a PKS gene cluster for the production of a heterologous template or, additionally or alternatively, one or more templates can be exogenously supplied to the host cells, e. g. in the screening method described below.

As mentioned above, the aglycone may be produced by the host cells by additionally cloning into the cell a recombinant polyketide synthase gene or genes either of type I or type II. The recombinant PKS genes may consist either of natural PKS genes or of mutated versions of natural PKS genes, or of hybrid PKS genes consisting of portions from at least two different natural type I PKS gene clusters, or natural type II PKS gene clusters, and may consist of a library of hybrid PKS genes of either type I or type II. Examples of PKS gene assemblies include those which produce the type I polyketide macrolides rifamycin, avermectin, rapamycin, immunomycin, or erythromycin, narbomycin, oleandomycin, pikromycin, spiramycin or tylosin ; polyenes such as amphotericin B, candicidin, nystatin or pimaricin ; polyethers such as monensin, salinomycin, semduramycin or tetronasin ; and type II polyketides such as

actinorhodin, daunorubicin, oxytetracycline or tetracycline.

A preferred host cell strain is actinomycete, more preferably strains such as Saccharopolyspora erythraea, Streptomyces coelicolor, Streptomyces avermitilis, Streptomyces griseofuscus, Streptomyces cinnamonensis, Streptomyces fradiae, streptomyces longisporoflavus, Streptomyces hygroscopicus, Micromonospora griseorubida, Streptomyces lasaliensis, Streptomyces venezuelae, Streptomyces antibioticus, Streptomyces lividans, Streptomyces rimosus, Streptomyces albus, Amycolatopsis mediterranei, and Streptomyces tsukubaensis. Examples of preferred strains and preferred modifications to those strains to adapt them for use in the present invention are set out below.

Examples of glycosyltransferases suitable for use in accordance with the present invention (listing the GTs, their normal biosynthetic contexts and normal substrate specificities) : From the erythromycin pathway of Saccharopolyspora erythraea : desosaminyltransferase eryCIII and mycarosyltransferase eryBV.

From the megalomycin pathway of Micromonospora megalomicea : desosaminyltransferase megCIII, mycarosyltransferase megBV and megosaminyltransferase.

From the oleandomycin pathway of Streptomyces antibioticus : oleandrosyltransferase oleG2 (also transfers rhamnose and olivose) and desosaminyltransferase oleGl.

From the tylosin pathway of Streptomyces fradiae : mycaminosyltransferase tylMII deoxyallose transferase tylN and mycarosyltransferase tylCV.

From the midecamycin pathway of Streptomyces mycarofaciens : mycaminosyltransferase midI, deoxyallose transferase and mycarosyltransferase.

From the pikromycin/narbomycin pathway of Streptomyces venezuelae : desosaminyltransferase desVII.

From the spinosyn pathway of Saccharopolyspora spinosa : rhamnosyltransferase and forosaminyltransferase.

From the amphotericin pathway of Streptomyces nodosus : mycaminosyltransferase amphDI.

From the avermectin pathway of Streptomyces avermitilis : oleandrosyltransferase.

From the nystatin pathway of Streptomyces : mycaminosyltransferase.

From the polyene 67-121C pathway of Actinoplanes caerulens : mycosaminyltransferase, mannosyltransferase (transferring to the mycosamine).

From the elloramycin pathway of Streptomyces olivaceaous Tü2353 : rhamnosyltransferase elmGT.

From the mithramycin pathway of Streptomyces argillaceus : olivosyltransferase mtmGIV.

From the daunomycin pathway of Streptomyces peucetius : daunosaminyltransferase dnrS.

From the urdamycin pathway of Streptomyces fradiae Tu2717 : rhodinosyltransferase urdGTlc, olivosyltransferase urdGTlb, rhodinosyltransferase urdGTla and olivosyltransferase urdGT2.

Preferably, the process further comprises the step of deleting or inactivating one or more genes in the microorganism host cells involved in the production of the aglycone

template and/or in its subsequent processing, thereby to suppress or alter the production of the natural aglycone template or product.

In a further aspect, the present invention provides a process for producing a library capable of producing a plurality of hybrid glycosylated products, the process comprising : transforming microorganism host cells with nucleic acid encoding one or more glycosyltransferases (GT) ; and, providing one or more aglycone templates to the GTs so that the GTs transfer one or more sugar moieties to the aglycone template to produce said plurality of hybrid glycosylated products ; wherein one or more of the sugar moiety or moieties, the aglycone template, the glycosyltransferase or the host cells are heterologous to the other components.

In further aspects, the present invention provides a process which further comprises screening the library for a hybrid glycosylated product having a desired characteristic.

In preferred embodiments, the library comprises 2 hybrid glycosylated products, more preferably at least 10 hybrid glycosylated products, more preferably at least 50 hybrid glycosylated products and still more preferably at least 100 hybrid glycosylated products.

In one embodiment, the present invention provides a process for screening for a hybrid glycosylated product, the process comprising : producing one or more different microorganism host cells, each host cell being transformed with nucleic acid encoding a glycosyltransferase (GT), wherein the GT is heterologous to the microorganism strain, to form a library of host cells ; supplying the library with one or more aglycone templates ; screening the library for hybrid glycosylated products produced by the GTs transferring one or more sugar moieties to the aglycone templates.

Preferably, the processes described herein further comprise isolating a host cell producing a desired hybrid glycosylated product, and treating it further (e. g. culturing the cells and isolating the product produced) so that the product can be made in bulk.

Preferably, in order to maximise diversity in the hybrid products, the screening method employs at least two different host cells and/or aglycone templates and/or glycosyltransferases and/or activated sugar moieties and/or heterologous modifying genes capable of modifying the sugar moiety before or after transfer to the aglycone template, more preferably at least 3, more preferably at least 5, more preferably at least 10, more preferably at least 20 and most preferably at least 50 different cells and/or templates.

This screening method allows a large number of novel hybrid products to be generated and screened maximising the number and variety of products that can be made and tested. Desired hybrid products can be detected by their biological activity (e. g. as antibiotics).

It will be evident to those skilled in the art that production of hybrid glycosides may be done in a number of alternative ways using the present invention, e. g. : (1) by including all the required genes in the same cell, whether introduced separately or as a single cassette ; or (2) stepwise, the product of one fermentation with a recombinant cell containing some of the expressed genes (either after purification or used as a filtered supernatant) in its turn being fed to a second bioconversion strain containing the rest of the expressed genes.

In the latter case, the process of the invention may comprise the steps of : producing one of the aglycone template or the sugar moiety in first host cells as a first product ; optionally, purifying the first product from a culture of the first host cells ; and, adding the first product to a second host cell comprising one or more genes encoding the other of the aglycone template or the sugar moiety and one or more

glycosyltransferases, so that the sugar moiety is transferred to the glycosyltransferase, to produce the hybrid glycosylated product.

The first product may be in various degrees of purification, i. e. it may be employed as a filtered supernatant, in an isolated form or in any degree of purification compatible with production in the second host cells.

In some embodiments, the host cells may additionally contain cloned and expressed genes for sugar biosynthesis, either an entire group of genes needed to furnish a naturally occurring or novel deoxysugar, for example all the specific eryB genes from Saccharopolyspora erythraea or elsewhere required to make activated dTDP-mycarose from metabolic intermediates common to actinomycete cells, or such gene sets missing certain genes so that altered activated sugars are provided, or certain individual genes only that modify the type of activated sugars produced by endogenous deoxyhexose biosynthetic pathways. Thus, the present invention includes the possibility of transforming the host cells with one or more genes for modifying the aglycone template, e. g. to provide alternative positions at which glycosylation can be introduced, to modify existing functionality in the template or which are involved in the downstream processing of the hybrid glycosylation products.

The cells may be additionally cultivated in the presence of cerulenin, which specifically suppresses endogenous polyketide biosynthesis, or additionally or alternatively mutated to delete or otherwise inactivate one or more of the PKS genes naturally present within the cells, in either case the result is to decrease competition with the supplied aglycone.

In a further aspect, the present invention provides the hybrid glycosylated products produced by any one of the processes described herein.

In a further aspect, the present invention provides novel hybrid glycosylated products resulting from the attachment of sugar moieties to aglycone templates. Examples of

hybrid products include those comprising : (a) one or more natural sugars linked to an erythronolide at the 7-position.

(b) one or more rhamnose or substituted (e. g. methyl) rhamnose sugars linked to an erythronolide (c) one or more mycarose or substituted mycarose sugars linked to an erythronolide ; and combinations of (a), (b) and (c) sugar substituents on an erythronolide (e. g. erythronolide B).

Examples of hybrid products based on an erythromycin template include those comprising : (a) one or more natural sugars linked to erythromycin at the 7-position ; (b) one or more mycarose or substituted mycarose sugars linked to an erythromycin ; (c) one or more mycaminose or substituted mycaminose sugars linked to an erythromycin ; and combinations of (a), (b) and (c) sugar substituents on an erythromycin (e. g. erythromycin A).

Examples of hybrid products based on an tylactone template include those comprising : (a) one or more glucose or substituted glucose sugars linked to a tylactone ; (b) one or more desosaminose or substituted desosaminose sugars linked to a tylactone ; (c) one or more mycaminose or substituted mycaminose sugars linked to a tylactone ; (d) one or more rhamnose or substituted rhamnose sugars linked to a tylactone ; and combinations of (a), (b), (c) and (d) sugars substituents such as a rhamnose and a mycaminose sugar linked to a tylactone (e. g. 23-0-rhamnosyl 5-O- mycaminosyl tylactone).

Specific examples of hybrid products of the invention are : 3-0-(2'-0-methylrhamnosyl) erythronolide B 3-0- (2', 3'-bis-0-methylrhamnosyl) erythronolide B 3-O-(2',3',4'-tris-O-methylrhamnosyl)erythronolide B 3-0-mycarosyl-erythronolide B 8a-hydroxy-3-0-mycarosyl erythronolide B 8, 8a-epoxy-3-0-mycarosyl erythronolide B 8, 8a-dehydro-6-deoxyerythronolide B 8-hydroxy-6-deoxyerythronolide B 3-0-(2'-0-methylrhamnosyl) erythromycin D 3-0- (2', 3'-bis-0-methylrhamnosyl) erythromycin D 3-O-(2', 3', 4'-tris-0-methyl rhamnosyl) erythromycin D 5-0-mycaminosyl-erythromycin A 5-0-mycaminosyl-4"-O-mycarosyl erythromycin A 5-O-glucosyl-tylactone<BR> 5-O-desosaminyl-tylactone 23-0-rhamnosyl 5-0-mycaminosyl tylactone 5-0 (2'-O)-bis-glucosyl-tylactone 3-0-rhamnosyl-8, 8a-dehydro-6-deoxyerythronolide B 3-0-rhamnosyl-8, 8a-dihydroxy-erythronolide B 3, 5 di-O-mycarosyl erythronolide B.

In a further aspect, the present invention provides a process for assembling a gene set in a cassette for transformation of a host cell for carrying out the processes described herein. The strategy to prepare gene cassettes with different combinations of glycosyltransferase-and methyltransferase genes is adapted from a technique previously described (WO 00/77181A2) to build gene cassettes expressed under the control of the actIl-Orf4 regulator. The expression of these gene cassettes in a suitable strain background is a powerful approach to generate novel post-PKS modified polyketides in a random or directed fashion. The method is based upon the introduction of XbaI restriction sites at the 3'and 5'-end of the PCR fragments. The

introduction of a XbaI site at the 5'-end of the PCR fragment which is sensitive to the Dam methylase of the strain background will protect this site from further XbaI digest.

To retain the Shine Dalgarno sequence 5'of the respective gene the pSG142 derived constructs which contain these genes were used as a template. Using plasmid DNA isolated from dam~ host strains (such as E. coli ET 12567-McNeil et al, (1992) Gene, 111, 61-68), the amplified genes can be isolated as XbaI fragments. Using a host strain with an active Dam methylase such as E. coli DH10B these fragments can be sequentially cloned into gene cassettes. This technique provides the means to build gene cassettes of different length and different order using the same strategy over and over again. An overview of the strategy described here and the isolated gene cassettes is depicted in Fig. 18. In some cases, expression of the terminal gene surprisingly was increased when nucleic acid encoding a histidine tag was fused to the 3'end of the gene cassette.

Where more than one gene is required to be introduced and expressed in the said host cells, many ways will readily occur to the person skilled in the art as to how this goal may be achieved in a way that ensures that there will be coordinated expression of all the required gene products. However, the present invention further provides a novel process and expression cassette for achieving this goal which, in one embodiment, allows the stepwise contiguous head to tail assembly of individual sugar pathway genes, or of heterologous modifying genes, or of both types of gene, and thereby not only places them on a single region of DNA under the control of a common promoter, but also thereby facilitates their further genetic manipulation together as a single unit or cassette. Each gene to form part of such a cassette assembly can be either a natural or modified gene, or synthetic versions of a natural genes, and such natural genes may be obtained either from the said host cells or may be heterologous to the said host cells.

Accordingly, in a still further aspect, the present invention provides an expression cassette comprising one or more glycosyltransferase genes and one or more auxiliary genes, operably linked under the control of a promoter. As described above, the

auxiliary genes may be genes encoding proteins involved in the biosynthesis of one or more sugars (a'sugar pathway gene') to enable a host cell transformed with the expression cassette to produce one or more sugar moieties for subsequent transfer to an aglycone template. Other examples of heterologous auxiliary genes include enzymes involved in the processing of sugar moieties or the aglycone template, either before or after the sugar moiety is transferred to the aglycone by the GT. Examples of these enzymes include methyltransferases and P450s which are responsible for hydroxylation of the aglycone template. Preferably, the genes in the cassette are under the control of a single, preferably strong, promoter. Further, the present inventors have found that by incorporating a nucleic acid sequence encoding a histidine tag adjacent to (or immediately 3') the terminal gene in the expression cassette so that expression of genes in the cassette which are distal to the promoter is improved.

In a further aspect, the present invention provides a process of producing an expression cassette comprising one or more glycosyltransferase genes and one or more auxiliary genes, the process comprising operably linking the genes together under the control of a promoter. The process may comprise the further step of transforming a host cell with the expression cassette and expressing the genes comprised within it to produce the GT and proteins encoded by the auxiliary genes.

In a further aspect, the present invention provides a host cell transformed with such an expression cassette.

Embodiments of the present invention will now be described by way of example and not limitation with reference to the accompanying figures.

Brief Description of the Figures Figure 1A : Scheme for the isolation of S. erythaea Aorfl4 containing a 1247-bp deletion in eryB :. tsr, thiostrepton resistance.

Figure 1B : Scheme for the isolation of S. erythaea strain DM containing both a

deletion in eryBV and a 1191-bp deletion in eryCIII. tsr, thiostrepton resistance.

Figure 2 : Structure of expression plasmid pSG142. EryRHS denotes a DNA fragment from the ermE-distal flank of the erythromycin biosynthetic gene cluster.

Figure 3 : General scheme for the screening of the cloned glycosyltransferase (glctr) for their ability to glycosylate aglycones in vivo.

Figure 4 : Structure of 3-0-rhamnosyl-erythronolide B + 3-0-rhamnosyl-6-deoxy- erythronolide B.

Figure 5A : Structure of 5-0-desosaminyltylactone.

Figure 5B : Structure of 5-0-glucosyltylactone.

Fig. 6 : Overview of the plasmids pSGSpnI, pSGSpnK and pSGSpnH.

Fig. 6B : Comparison of the DNA (top) and amino acid (bottom) sequence of spnl/ SpnI and the published sequence accession AY007564. The changes included into the corrected nucleotide and amino acid sequence are underlined.

Fig. 7 : Analysis of spni using chromosomal DNA of SGT2pSGSpnI as a template.

Fig. 8 : Analysis of spnK using chromosomal DNA of SGT2pSGSpnK as a template.

Fig. 9 : Analysis of spnH using chromosomal DNA of SGT2pSGSpnH as a template.

Fig. 10 : Analysis of the culture supernatants of SGT2 and SGT2pSGSpnI after feeding of rhamnosyl-erythronolide B.

Fig. 11 : Structure of3-0- (2'-0-methylrhamnosyl) erythronolide B.

Fig. 12 : Analysis of the culture supernatants of SGT2 and SGT2pSGSpnK after feeding with 3-0- (2'-0-methylrhamnosyl) erythronolide B.

Fig. 13 : Overview of the plasmids pSGSpnHK and pSGSpnHKI.

Fig. 14 : Structure of 3-0- (2', 3'-bis-0-methylrhamnosyl) erythronolide B.

Fig. 15 : Structure of 3-O-(2', 3', 4'-tris-0-methylrhamnosyl) erythronolide B.

Fig. 16 : Structures of 3-0-rhamnosyl-erythromycin D (A) ; 3-O-(2'-O- methylrhamnosyl) erythromycin D (B) ; and 3-0- (2', 3'-bis-0- methylrhamnosyl) erythromycin D (C).

Fig. 17 : Bioassay of 3-0-(2'-0-methylrhamnosyl) erythromycin D and erythromycin A.

Fig. 18 : Strategy to isolate gene cassettes of oleG2, synI, spnH, spnK and eryCIII.

Fig. 19 : Results of the analysis of the culture supernatant of SGT3pSGcasoleG2spnIspnK.

Fig. 20 : Results of the analysis of the culture supernatant of SGT3pSGcasoleG2spnIspnKeryCIII.

Figure 21 : Structures of 8a-hydroxy-3-0-mycarosyl erythronolide B and 8, 8a-epoxy- 3-0-mycarosyl erythronolide B.

Figure 22 : Structures of 8, 8a-dehydro-6-deoxyerythronolide B and 8-hydroxy-6- deoxyerythronolide B.

Figure 23 : Scheme for the construction of gene cassettes and transformation into S. erythraea.

Figure 24 : Structures of 3-0-rhamnosyl-8, 8a-dehydro-6-deoxyerythronolide B and 3- O-rhamnosyl-8, 8a dihydroxy-6-deoxyerythronolide B.

Figure 25 : DNA sequence of the oleGl start region (accession number AJ002638).

Figure 26 : Results of the complementation of the eryCIII mutation in S. erythraea SGT2 and SGT2pSGOleGl. 3-0-mycarosyl erythronolide B was fed to the cultures as described previously (Gaisser et al., 2000). The presence of 3-0-mycarosyl erythronolide B is indicated by a peak of a retention time of 20 min. The peak of a retention time of 15. 7 and 734 m/z indicates the presence of erythromycin A.

Figure 27 : Results of the feeding of 6-deoxyerythronolide B to SGQ1, SGQIpSGOleGl, SGQlpSGOleG2 and SGQIpSGOleP. Structures of the major compounds are indicated.

Figure 28 : Results of the feeding of 3-0-rhamnosyl-6-deoxyerythronolide B to SGQ1, SGQ1pSGO1eG1, SGQIpSGOleG2 and SGQIpSGOleP. Structures of the major compounds are indicated.

Figure 29 : Structure of 23-0-rhamnosyl-5-0-mycaminosyl-tylactone.

Figure 30 : Structure of 23-0-rhamnosyl-5-0-desosaminyl-tylactone.

Figure 31 : Structure of 5-0- (2'-O-) bis-glucosyl-tylactone.

Figure 32 : Structures of 3-0-mycarosyl-5-0-mycaminosyl-erythromycin A and 3-0- mycarosyl-5-O-mycaminosyl-4"-O-mycarosyl-eryhromycin A.

Figure 33 : Sequence of the genes tylMI, /M3 and tylB, showing 8 amino acid changes in type as compared to the published sequence.

Figure 34 : Assembly of gene cassettes in pUC18 to produce constructs pUClStylMN- tylB and pUC18w1MIII-tylB-tylMI.

Figure 35 : Restriction maps for pSGCIII, pSGTYLM2, pSGDESVII and pSGTYLCV.

Figure 36 : Restriction map of plasmid pGGl.

Figure 37 : Mass spectra of novel compounds 3-O-mycarosyl-5-O-mycaminosyl- erythromycin A and 3, 5-di-O-mycarosyl-erythronolide B.

Materials and Methods Escherichia coli XLI-Blue MR (Stratagene) and E. coli DH10B (GibcoBRL) were grown in 2xTY medium as described by Sambrook et al. (Molecular Cloning, A Laboratory Manual, 2nd Edition (1989) Cold Spring Harbor, NY, Cold Spring Harbor Press). Vectors Litmus28 and pUC18 were obtained from New England Biolabs and vector pQE-16 from QIAGEN. Vector pIJ702 (Katz et al. J. Gen. Microbiol. (1983) 129 : 2703-2714) was kindly provided by D. A. Hopwood (John Innes Institute, Norwich, UK). Vector pUC2G which contains a 6. 2 kb BgllI fragment from cosAB35 cloned into pUC18 was kindly provided by J. Salas (University of Oviedo, Spain).

Vector pLQDl which contains a 5076 bp PvuII fragment from cosmid cos25G8 cloned into pUC18 was kindly provided by J. Salas (University of Oviedo, Spain).

Cosmid no 7 which has been isolated from a Streptomyces fradiae cosmid library, was kindly provided by J. Cortes. E. coli transformants were selected with 100, ug/ml ampicillin. The Saccharopolyspora erythraea NRRL 2338-red variant strain (Hessler et al., Appl. Microbiol. Biotechnol. (1997) 47 : 398-404) was kindly provided by J. M.

Weber and was routinely maintained on M1-102 agar (Kaneda et al. J. Biol. Chem.

(1962) 237 : 322-327), R2T20 (Yamamoto et al. J. Antibiot. (1986) 34 : 1304-1313), R2T2 (same as R2T (Weber et al. J. Bacterial. (1985) 164 : 425-433), but without peptone), and TSB (Difco) for liquid cultures at 30°C. Bacillus subtilis ATCC 6633 was used in bioassays to assess erythromycin production as described (Gaisser et al.

(1997), supra). These assays were modified to investigate erythromycin production after feeding with erythronolide B or 3-a-mycarosyl-erythronolide B kindly provided by J.-M. Michel (Hoechst Marion Roussel, Romainville, France). Both metabolites (10, ul of lOmM stock solution) were applied to agar wells cut into the S. erythraea lawn and incubated at 30°C overnight as described (Gaisser et al. (1997) supra) and the development of zones of inhibition in the B. subtilis lawn around S. erythraea colonies was assessed. Expression vectors in S. erythraea were derived from plasmid pCJR24 (Rowe et al., Gene (1998) 216 : 215-223). Plasmid-containing S. erythraea were selected with 25/g/ml thiostrepton. To investigate the production of antibiotics, S. erythraea strains were grown in sucrose-succinate medium (Caffrey et al. (1992) FEBS Lett. 304 : 225-228) as described (Gaisser et al., (1997) supra) and the cells were harvested by centrifugation. Tylactone was kindly provided by B.

Wilkinson of Glaxo Group Research, Stevenage, UK. 150/, zl of a 100 mg/ml stock solution was added to a 500 ml S. erythraea expression culture.

DNA manipulation and sequencing DNA manipulations, PCR and electroporation procedures were carried out as described in Sambrook et al. (1989) supra. Protoplast formation and transformation procedures of S. erythraea were as described (Gaisser et al., (1997) supra). Southern hybridizations were carried out with probes labelled with digoxigenin using the DIG DNA labelling kit (Boehringer Mannheim). DNA sequencing was performed by the method of Sanger et al. (P. N. A. S. USA (1977) 74 : 5463-5467), using automated DNA sequencing on double stranded DNA templates with an Applied Biosystems 373A sequencer. Sequence data were analysed using the Staden Programs (Staden, R. Nucl.

Acids Res. (1984) 12 : 521-528) and the Genetics Computer Group (GCG, version 10) software package (Devereux et al., Nucl. Acids Res. (1984) 12 : 387-395).

Extraction and mass spectrometry 10 ml of each fermentation broth was centrifuged and the pH of the supernatant was adjusted to pH 9. The supernatant was extracted twice with an equal volume of ethyl acetate. The organic layer was dried over Na2SO4, evaporated to dryness and then

redissolved in 0. 3 ml acetonitrile/water (1 : 1 v/v). Mass spectrometry was performed on a BioQ (Micromass, Manchester, UK) or a Finnigan LCQ (Finnigan, CA) instrument. High resolution spectra were obtained on a Bruker BioApex II FT-ICR (Bruker, Bremen, FRG).

For NMR analysis, the bacterial broth was centrifuged and the pH of the supernatant was adjusted to about pH 9. The supernatant was extracted with three equal volumes of ethyl acetate, the extracts were combined, dried (Na2SO4) and evaporated under reduced pressure to yield a yellow solid. Final purification was achieved using reversed phase preparative HPLC on a Gilson 315 System using a 21 mm x 250 mm Prodigy ODS3 column (Phenomenex, Macclesfield, UK.). The mobile phase was pumped at a flow rate of 21 ml/min as a binary system consisting of 45% CH3CN, 55% 20 mM NH40Ac [pH 5. 5 with HCOOH] increasing linearly to 95% CH3CN over 25 min.

'H Nuclear Magnetic Resonance (NMR) spectra were acquired at 800 MHz on a Bruker Avance DRX800 and at 500 MHz on a Bruker Avance DRX500."C NMR spectra were acquired at 100 MHz on a Bruker Avance DRX400 spectrometer.

Samples for NMR analysis were dissolved in CD30D and the experiments were performed at 300 K. High Performance Liquid Chromatography (HPLC) was performed on a Hewlett Packard HP 1100 liquid chromatograph. Liquid Chromatography Mass Spectrometry (LC-MS), Tandem Mass Spectrometry (MS/MS) and Mus'spectra were obtained on a Finnigan MAT (San Jose, CA) LCQ. High resolution Quadrupole Time Of Flight MS/MS data were obtained on a Micromass (Macclesfield, U. K.) QTOF. High resolution MS were obtained on a Bruker Daltonics BioApex II 4. 7 T Fourier Transform Ion Cyclotron Resonance mass spectrometer using PEG as external calibrant.

Chromosomal deletion of eryBV (Fig. 1A) Vector pNC028 (Gaisser et al, 1997) (Fig. 1A) was digested withXbaI/NheI and XbaIlSpeI, a 3. 8 kb band and a 0. 4 kb fragment were isolated, ligated and transformed

into E. coli DH 1 OB. A vector construct pNCOA containing the deletion of eryBV was isolated and the SpellNheI ligation site was verified by sequencing. After digestion with BglII, the plasmid was ligated with BgIII-cut pIJ702 and used to transform E. coli DH10B. Plasmid pNCOApIJ702 was isolated and used to transform S. erythraea NRRL 2338 (red variant). Colonies were selected for thiostrepton resistance and integration into the chromosome was confirmed by Southern analysis. To allow for the second recombination event, integrants were subcultured at least four times in TSB medium (Difco) at 30°C. Single colonies were screened for thiostrepton sensitivity and for erythromycin production. The 1247 nt chromosomal deletion of eryBV was verified by using the 1. 5 kb NcoI fragment of plasmid pNCOA to probe ClaI/PstI-and NcoI-digested chromosomal DNA of the wild type strain and of mutant Aorfl4. Analysis of the wild type S. erythraea showed the expected 2. 9 kb ClaI/PstI and 2. 7 kb NcoI band after hybridization. When chromosomal DNA of Aorfl 4 was treated similarly, only a 1. 6 kb ClallPstI and a 1. 5 kb NcoI fragment were detected indicating that eryBV had been removed. The mutant Aorfl4 was tested in a bioassay.

No zones of inhibition were observed around the S. erythraea colonies in the B. subtilis lawn, indicating that no erythromycin A is produced by the mutant strain.

Chromosomal deletion of eryCIII (Fig. 1B) Plasmid #SE55 (Haydock et al., Mol. Gen. Genet. (1991) 230 : 120-128) was digested with NcoVXhoI (Fig. 1B) and a 2. 4 kb fragment was cloned into NcoVXhoI digested vector Litmus28. Plasmid pLitNX was isolated and its identity verified by restriction digestion and by sequencing. Using SE55 as a template and primers SG10 5'- GGCGATGTGCCAGCCCGCGAAGTT-3'and SG11 5'- AGCCGTCACCGGCCATGGTCGTCGGCATCT-3', a 573 nt fragment was amplified using PCR, treated with T4 polynucleotide kinase and cloned into SmaI-cut pUC18. The sequence was checked, and this plasmid construct was then digested with NcoI and a 0. 5 kb fragment was isolated, ligated into NcoI-digested pLitNX and used to transform E. coli DH10B. Plasmid pLitNXP was isolated and the correct insert was verified, by restriction digestion and sequencing, as bearing a 1191 nt deletion in eryCIII. The construct pLitNXP was digested with BglII and ligated into

pIJ702 previously treated with BgIII, and the mixture was used to transform E. coli DH10B. Plasmid pLitNXPIJ was isolated and used to transform the S. erythraea mutant Aorfl4 as described. Integration was verified by Southern blot hybridization.

After subculturing, single colonies were screened for thiostrepton sensitivity as described above. Thiostrepton sensitive colonies were grown in small patches and agar plugs taken from well-grown areas were placed on bioassay plates containing 3 alpha-mycarosyl-erythronolide B. No inhibition zone was observed around colonies of the isolated S. erythraea strain DM (Aorfl4Aorf8). The 1247 nt chromosomal deletion was verified in Southern blot hybridizations as described for the mutant Aorfl4. The 1191 nt chromosomal eryCIII deletion was verified using a digoxygenin-labelled 573 nt DNA fragment, amplified by primers SG10 and SG11, to probe NcoI-and ClaI/BglII-digested chromosomal DNA of the wild type strain and of the mutant DM (Aorfl4Aorf8). Analysis of the wild type S. erythraea showed the expected 1. 7 kb NcoI and 12. 4 kb BglII bands after hybridization. When chromosomal DNA of DM was treated similarly, only an 11 kb BgIII and a 0. 5 kb Ncol fragment were detected indicating that 1. 2 kb of the eryCIII gene had been removed.

Chromosomal deletion of eryA To allow the feeding of different aglycones to a S. erythraea mutant strain housing heterologous glycosyltransferase genes, the eryA deletion previously described for JC2/Del60 (Rowe et al., (1998) supra) was also introduced into the S. erythraea mutant strain DM. S. erythraea SGT2 was isolated (Fig. 2) and all three deletions (AeryAABVACCII) were verified by Southern blot analysis. Culture broths of S. erythraea SGT2 were analysed by electrospray mass spectrometry. No peaks corresponding to either erythromycin A or precursor metabolites were found.

Construction of expression plasmidfor eryBV The gene eryBV was amplified by PCR using the primers 1518 5'-GGGGGATCCCATATGCGGGTACTGCTGACGTCCTTCG-3'and 1519 5'- GAAAAGATCTGCCGGCGTGGCGGCGCGTGAGTTCCTC-3', which introduce a

BamHI and a MM site at the 5'end and a Bg/II site at 3'end of eryBV. After treatment with T4 polynucleotide kinase the PCR product was cloned into SmaI-cut pUC18 and the resulting plasmid used to transform E. coli DH1OB. The sequence of eryBV in the isolated plasmid was checked, and this plasmid was then digested using the restriction enzymes BamHI and BglII, a 1. 2 kb fragment was isolated and ligated into the identically-digested vector fragment of pQE-16, which introduced a C-terminal His6-tag into EryBV. The pQE-16 derived plasmid was digested with NdeI andXbaI and a 2. 2 kb fragment was isolated and ligated into vector pCJR24 previously cut with NdeI and XbaI, to give plasmid pSG2414. To allow the recombination of this plasmid into the genome of S. erythraea, a 1. 7 kb NcoI fragment from cosmid cos6B (Gaisser et al., (1997) supra) from the ermE distal flank of the erythromycin biosynthetic cluster (Pereda et al., Gene (1997) 193 : 65-71) was isolated and cloned into the pQE-16 derived NcoI site. This final construct was named pSG142.

Construction of expression plasmidfor eryCIII For expression of eryCIII, primers SG14 5'- GAAAAGATCTTCGTGGTTCTCTCCTTCCTGCGGCCAG-3'and SG15 5'- GGGGGATCCCATATGCGCGTCGTCTTCTCCTCCAT-3'were used to amplify eryCIII with BSE55 as template. The 1287 bp DNA fragment was isolated, treated with T4 polynucleotide kinase and cloned into SmaI cut pUC18. After transformation into E. coli DH10B the construct was isolated and the sequence of eryCIII was verified. After digestion with NdeIlBglII, a 1. 2 kb fragment was isolated, ligated with the vector fragment of NdeIlBglII digested pSG142 and used to transform E. coli DH10B. Plasmid pSGCIII was isolated.

Construction of expression plasmidfor oleG2 For expression of oleG2, the primers Ole3 5'- GGCGGATCCCATATGCGCGTACTGCTGACCTGCTTCGCC-3'and Ole4 5'- CCAGATCTGCCCGCATGGTTCCCGCCTCCTCGTCC-3'were used to amplify oleG2 using plasmid pUC2G or chromosomal DNA of Streptomyces antibioticus as a

template. The PCR fragment was isolated, treated with T4 polynucleotide kinase and cloned into SmaI cut pUC 18. After transformation into E. coli DH10B the construct was isolated and the sequence of oleG2 was verified. After digestion with NdeVBgllI, a 1. 3 kb fragment was isolated, ligated with the vector fragment of NdeI/BglII digested pSG142 and used to transform E. coli DH10B. Plasmid pSGOLEG2 was isolated.

Construction of expression plasmidfor tyIM2 For expression of tyIM2 (Gandecha et al., Gene (1997) 184 : 197-203) the primers Tyll 5'-GTGGAGATCTCCTTTCCGGCGCGGATCGGGACCG-3'and Tyl2 5'- GGGGGATCCCATATGCGGGTACTGCTGACCTGTATCG-3'were used to amplify tylM2 using cosmid no 7 as a template or chromosomal DNA of Streptomyces fi » adiae. Primer Tyl2 was chosen on the basis of sequence comparisons of known glycosyltransferases, which indicated the methionine at position 21 of the published sequence (accession no X81885) to be the start codon. The changes to the DNA sequence of tylM2 noted in a recent update (X81885/June 1999) were confirmed independently in this work. The PCR DNA fragment was isolated, treated with T4 polynucleotide kinase and cloned into SmaI cut pUC18. After transformation into E. coli DH1OB the construct was isolated and the sequence of tylM2 was verified. After digestion with NdeVBglII, a 1. 3 kb fragment was isolated, ligated with the vector fragment of NdellBglII digested pSG142 and used to transform E. coli DH1OB.

Plasmid pSGTYLM2 was isolated.

Construction of expression plasmidfor desVII For expression of desVII (Xue et al., (1998) P. N. A. S. USA, 95 : 12111-12116) the primers Pikl 5'-GGAGGATCCCATATGCGCGTCCTGCTGACCTCGTTCG-3' and Pik2 5'-GGGGTGCAGATCTGTGCCGGGCGTCGGCCGGCGGG-3'were used to amplify v desVII using genomic DNA of Streptomyces venezuelae as a template. The PCR DNA fragment was isolated, treated with T4 polynucleotide kinase and cloned into SmaI cut pUC18. After transformation into E. coli DH10B the construct was isolated and the sequence of desVII was verified. After digestion with NdeVBglII, a 1. 3 kb fragment was isolated, ligated with the vector fragment of

NdeI/BglII digested pSG142 and used to transform E. coli DH10B. Plasmid pSGDESVII was isolated.

Construction of expression plasmidfor tylH For expression of tylH (Fouces et al., Microbiology, (1999) 145 : 855-868) the primers TylH1 5'-CCGCCCGGCCCAGATCTCCGCGGCCCTCATGCGT-3'and TylH2 5'- TTGAGGCCGCAGCGACATATGTCCTCGTCCGGGGA-3'were used to amplify tylH using genomic DNA of Streptomyces fradiae as a template. The PCR DNA fragment was isolated, treated with T4 polynucleotide kinase and cloned into SmaI cut pUC18. After transformation into E. coli DH10B the construct was isolated. After digestion with NdeIlBgllI, a 1. 6 kb fragment was isolated, ligated with the vector fragment ofNdeIlBgIII digested pSG142 and used to transform E. coli DH10B.

Plasmid pSGTYLH1 was isolated.

Construction of expression plasmidfor tylN For expression of tylN (Fouces et al., 1999) the primers Tyl5 5'-GGGCATATGCGCATAGCGTTGCTGACCATGGGCT-3'and Tyl4 5'- GGCCAGATCTGCCGGGGGTGTGTGCCGTGGTCCGGG-3'were used to amplify tylN using genomic DNA of Streptomyces fradiae as a template. The PCR DNA fragment was isolated, treated with T4 polynucleotide kinase and cloned into SmaI cut pUC18. After transformation into E. coli DH1OB the construct was isolated. After digestion with NdellBglII, a 1. 3 kb fragment was isolated, ligated with the vector fragment of NdeI/BglII digested pSG142 and used to transform E. coli DH10B.

Plasmid pSGTYLN was isolated.

Construction of expression plasmids for both tylH and tylN Plasmid pSGTYLH was digested with BglII and pSGTyIN was digested with AflIIlNheI. The fragments were submitted to fill-in reactions using Klenow polymerase (Sambrook et al., (1989) supra). The pSGTYLH vector derived DNA was isolated and ligated with the 1. 5 kb fragment encoding TyIN. E. coli DH1OB was transformed with the ligation mixture. Plasmid pSGTYLHN was isolated.

Construction of expression plasmidfor oleD For expression of oleD (Hernandez et al., Gene (1993) 134 : 139-140) the primers OleDl 5'-CCGGATCCCATATGACCACCCAGACCACTCCCGCCCACATC-3' and OLED2 5'-CGAGATCTCAAAGCGGATCTCTGCCGGTCGGAACGGA-3' were used to amplify oleD using pLQD1 (Luis M. Quiros) or chromosomal DNA of Streptomyces antibioticus as a template. The PCR DNA fragment was isolated, treated with T4 polynucleotide kinase and cloned into SmaI cut pUC 18. After transformation into E. coli DH1OB the construct was isolated. After digestion with NdeIlBgllI, a 1. 3 kb fragment was isolated, ligated with the vector fragment of NdeIlBgllI digested pSG142 and used to transform E. coli DH10B. Plasmid pSGOLED was isolated.

Isolation of 3-0-rhamnosyl-erythronolide B and 3-0-rhamnosyl-6-deoxy-. erythronolide B The plasmid pSGOLEG2 was transformed into S. erythraea mutant DM cells and culture broths of the transformed strains were analysed as described by Gaisser et al.

(1997) and by Gaisser et al. (1998). Analysis of the S. erythraea mutant DM (pSGOLEG2) by electrospray mass spectroscopy revealed the presence of only two distinct new peaks, at m/z 555 and 571 respectively, which were not present in the culture broth of the S. erythraea strain DM examined under the same conditions of growth, extraction and analysis. MS/MS experiments revealed that the ion with m/z of 571 fragmented into an ion with m/z of 425 (corresponding to the sodium salt of erythronolide B, EB-Na+) corresponding to the loss of m/z 146 (rhamnose). The fragmentation pattern of the ion of m/z 555 was identical to that of the fragment of m/z 571, but shifted lower by 16 mass units, indicating a missing hydroxy group.

This was evidence that the compound with m/z 555 represents rhamnosyl-6- deoxyerythronolide B. To confirm these structures, 1. 5 litres of culture broth were used to purify 4. 6 mg of 3-O-rhamnosyl-erythronolide B and 2. 7 mg of 3-O- rhamnosyl-6-deoxyerythronolide B. Both compounds were analysed and the structures were fully confirmed by lH and 13C NMR.

Table 1 :'HNMR Datafor 3-O=rhamnosyl-erythronolide B Proton 5H multiplicity coupling 2-H 2. 84 dq 10. 4, 6. 8 3-H 3. 71 d 10. 4 4-H 2. 16 dd 7. 4, 3. 6 5-H 3. 52 d 3. 6 7-Ha 1. 94 dd 14. 7, 10. 3 7-Hb 1. 43 dd 14. 7, 2. 6 8-H 2. 70 m 10-H3. 03qd 6. 9, 1. 7 11-H 3. 96 dd 9. 7, 1. 5 12-H 1. 65 qd 9. 7, 7. 1 13-H 5.44 ddd 9. 7, 4. 4, 0. 6 14-Ha 1. 71 m 14-Hb 1. 49 m 15-H3 0.88 dd 7. 4, 7. 4 16-H3 1.20 d 6. 8 17-H3 1. 02 d 7. 4 18-H3 1. 30 s 19-H3 1.13 d 7.1 20-H3 0. 96 d 6. 8 21-H3 0. 94 d 7. 1 1'-H 4.88 d 1.9 2'-H 3. 94 dd 3. 2, 1. 9 3'-H 3. 64 dd 9. 5, 3. 2 4'-H 3.42 dd 9.5, 9. 5 5'-H 3. 85 dq 9. 5, 6. 2 6'-H3 1.28 d 6.2

Table 2: 13C NMR Data for 3-O-rhamnosyl-erythronolide B Carbon bc C1 175. 4 C2 44. 3 C3 87. 5 C4 36. 1 C5 80. 6 C6 74. 4 C7 36. 3 C8 44. 7 C9 219. 9 C10 39. 3 C11 69. 5 C12 39. 8 C13 74. 5 C14 25. 5 C15 9. 3 C16 14. 5 C17 7. 6 C18 16. 4 C19 17. 4 C20 8. 1 C21 8. 1 C1'103. 0 C2'70. 8 C3'70. 7 C4'72. 2 C5'69. 3 C6'16. 5

Table 3 :'HNMR Datafor 3-0-rhamnosyl-6-deoxyerythronolide B Proton bH multiplicity coupling 2-H 2. 90 m 3-H 3. 69 m 4-H 1. 66 m 5-H 3. 52 d 8. 7 6-H 1. 63 m 7-Ha 1. 89 m 7-Hb 0. 96 m 8-H 2. 68 m 10-H 2. 89 m 11-H 3. 75 dd 10. 0, 1. 8 12-H 1. 69 m 13-H 5. 25 dd 9. 1, 4. 5 14-Ha 1. 78 m 14-Hb 1.53 m 15-H3 0.92 dd 9. 4, 9. 4 16-H3 1.97 d 6. 8 17-H3 1. 07 d 6. 9 18-H3 1.17 d 6. 8 19-H3 1. 11 d 6. 6.

20-H3 1. 24 d 6. 9 21-H3 0. 91 d 7.0 1'-H 4.79 d 1. 6 2'-H 3.97 dd 3. 2, 1. 8 3'-H 3. 61 dd 9. 6''3. 2 4'-H 3. 42 dd 9. 6, 9. 6 5'-H 3.69 dq 9. 6, 6. 2 6'H3 1. 28 d 6. 2

Table 4 :'3C NMR Data for 3-O-rhamnosyl-6-deoxyerythronolide B Carbon8 C1 177. 0 C2 42. 6 C3 83. 1 C4 41. 5 C5 76. 4 C6 36. 2 C7 33. 6 C8 44. 4 C9 213. 3 C10 44. 9 C11 70. 0 C12 40. 9 C13 76. 0 C14 25. 2 C15 5. 9 C16 13. 8 C17 8. 5 C18 19. 5 C19 14. 8 C20 14. 1 C21 8. 3 C1' 103. 0 C2'70. 6 C3'70. 8 C4'72. 2 C5'69. 4 C6'16. 6

Construction of expression plasmid for spnl, The gene spnI was amplified by PCR using the primers SpnIl 5'- CTTCATATGAGTGAGATCGCAGTTGCCCCCTGGTCG-3'and SpnI2 5'- AACAGATCTGCCGCCCTCGACGCCGAGCGCTTGCC-3', which introduce a MM site at the 5'end and a BglII site at 3'end of spnI. Chromosomal DNA of Saccharopolyspora spinosa was used as a template. After treatment with T4 polynucleotide kinase the PCR product was cloned into SmaI-cut pUC18 and the resulting plasmid was used to transform E. coli DH10B. The sequence of spnl in the isolated plasmid was checked, and this plasmid was then digested using the restriction enzymes NdeI and BglII. A 1. 2 kb fragment was isolated and ligated into the identically-digested vector fragment of pSG142, which introduced a C-terminal His6- tag into SpnI. This final construct was named pSGSpnI (Fig. 6). Differences to the published DNA sequence accession AY007564 were detected (Fig. 6B).

Construction of expression plasmidfor spnK For expression of spnK, primers SpnKI 5'- TCATCCATATGTCCACAACGCACGAGATCGAAACCGT-3'and SpnK2 5'- TCTGCAGATCTCTCGTCCTCCGCGCTGTTCACGTCGGCCA-3'were used to amplify spnK with chromosomal DNA of Saccharopolyspora spinosa as a template.

The 1. 2 kb DNA fragment was isolated, treated with T4 polynucleotide kinase and cloned into SmaI cut pUC18. After transformation into E. coli DH10B the construct was isolated and the sequence of spnK was verified. After digestion with NdeIlBglII, a 1. 2 kb fragment was isolated, ligated with the vector fragment of NdeIlBglU digested pSG142 and used to transform E. coli DH10B. Plasmid pSGSpnK was isolated (Fig.

6). A C-terminal His6-tag was introduced into SpnK.

Construction of expression plasmidfor spnH For expression of the primers SpnHl 5'- TTCTAGAGATCTACCACAACCTGGTATTCGTGGAGAA-3'and SpnH2 5'- AACATATGCCCTCCCAGAACGCGCTGTACCTGG-3'were used to amplify synth using chromosomal DNA of Saccharopolyspora spinosa as a template. The

PCR fragment was isolated, treated with T4 polynucleotide kinase and cloned into SmaI cut pUC18. After transformation into E. coli DH10B the construct was isolated and the sequence of spnH was verified. After digestion with NdeIlBglII, a 0. 9 kb fragment was isolated, ligated with the vector fragment of NdeIlBglII digested pSG142 and used to transform E. coli DH1OB. Plasmid pSGSpnH was isolated (Fig.

6).

Isolation of the bioconversion strains SGT2pSGSpnI, SGT2pSGSpnk and SGT2pSGSpnH Saccharopolyspora erythraea SGT2 (Gaisser et al., 2000) was transformed with the plasmid constructs pSGSpnI, pSGSpnK and pSGSpnH. The transformants were verified by isolating chromosomal DNA followed by PCR analysis. The PCR products were assessed by restriction digests and the pattern of DNA fragments for spnI, spnK and synSwas as expected (Fig. 7, 8 and 9).

Preparation of 3-O-(2'-O-methylrhamnosyl)erythronolide B Feeding experiments using 3-O-rhamnosyl-erythronolide B (Gaisser et al., 2000) were carried out as described (Gaisser et al., 1997). The cultures of the strains SGT2 and SGT2pSGSpnI, were fed with 3-0-rhamnosyl-erythronolide B, incubated at 30 °C for 3 to 5 days and analysed using electrospray mass spectrometry (Fig. 10). A novel peak was visible in supernatants of SGT2pSGSpnI with a retention time of 17. 5 minutes and m/z 545 ([M-H2O] H+) and m/z 585 ([M] Na+).

Isolation of 3-O-(2'-O-methylrhamnosyl)erythronolide B 1 1 of DMpSGOleG2 culture supernatant containing 3-0-rhamnosyl-erythronolide B was filter sterilised and fed to cultures of SGT2pSGSpnI using standard microbiological techniques as described above. The new compound with the retention time of 17. 5 minutes and m/z 545 ([M-H2O] H+) and m/z 585 ( [M] Na+) was isolated from the supernatant of these cultures as previously described (Gaisser et al., 2000).

The novel compound was characterised as 3-0- (2'-0-methylrhamnosyl) erythronolide B (Fig. 11).

Table 5: 1H and 13C NMR data for 3-O-(2'-O-methylrhamnosyl)erythronolide B <BR> <BR> <BR> <BR> <BR> <BR> <BR> <BR> <BR> Position #H Multiplicity Coupling #C 1 176. 8 2 2. 85 dq 10. 2, 7. 0 45. 9 3 3. 75 d 10. 6 88. 9 4 2. 15 m 37. 8 5 3. 51 d 3. 9 81. 8 6 75. 8 7 1. 93 dd 14. 5, 10. 2 37. 9 1. 43 dd 14. 5, 2. 7 8 2. 71 m 46. 2 9 220. 8 10 3. 04 m 40. 9 11 3. 96 dd 10. 4, 1. 6 71. 0 12 1. 65 m 41. 3 13 5. 44 dd 9. 8, 4. 7 76. 1 14 1. 73 m 27. 1 1. 49 m 15 0. 88 dd 7. 4, 7. 4 10. 8 16 1. 21 d 7. 0 16. 1 17 1. 01 d 7. 4 9. 1 18 1. 33 s 26. 6 19 1. 13 d 7. 0 19. 0 20 0. 96 d 6. 7 9. 6 21 0. 94 d 7. 0 9. 7 1'4. 93 d 1. 4 100. 8 2' 3.54 dd 3. 3, 1. 7 82. 3 3'3. 68 dd 9. 4, 3. 2 72. 1 4'3. 34 dd 9. 4, 9. 4 73. 9 5'3. 82 dq 9. 4, 6. 3 70. 7 6' 1.25 d 6. 3 17. 9 7'3. 44 s 59. 1

Feeding of 3-0-(2cO-methylrhamnosyl) erythronolide B 3-O-rhamnosyl-erythronolide B was fed to cultures of SGT2pSGSpnI followed by an incubation at 30°C. The supernatants containing 3-0- (2'-0- methylrhamnosyl) erythronolide B were centrifuged, filter sterilised and added to cultures of the strains SGT2 and SGT2pSGSpnK using standard microbiological techniques. After incubation at 30°C for several days the supernatants were analysed by electrospray mass spectrometry (Fig. 12). A new peak with a retention time of 20. 7 minutes and m/z of 559 ([M-H2O] H) and m/z of 599 ([M]Na+) was detected which indicates the presence of 3-0-(2', 3'-bis-O-methylrhamnosyl) erythronolide B in the culture supernatant of the strain SGT2pSGSpnK. To prepare sufficient amounts of this novel compound for NMR analysis, plasmid pSGSpnIKH was isolated.

Construction of expression plasmid pSGSpnIKH An expression plasmid which contains both genes, spnH and spunk, was isolated after digesting plasmid pSGSpnH with BglII and isolating the vector DNA. Plasmid pSGSpnK was digested with AflII/NheI and the 1. 5 kb DNA band was isolated. Fill- in reactions were performed using the isolated DNA fragments as described in Sambrook et al., 1989 followed by ligation and transformation of E. coli DH1OB.

Plasmid pSGSpnHK was isolated (Fig. 13). Plasmid pSGSpnHK was digested with Xbal and the vector DNA was isolated. Plasmid pSGSpnI was digested with AfElI/ NheI and the 1. 5 kb DNA band was isolated. Fill-in reactions were performed using the isolated DNA fragments as described in Sambrook et al., 1989 followed by ligation and transformation of E. coli DH1OB. Plasmid pSGSpnIKH was isolated (Fig.

13) and S. erythraea SGT2 was transformed.

Preparation of 3-0- (2', 3'-bis-O-rraethylrhamnosyl) erythronolide B 1 1 of DMpSGOleG2 culture supernatant containing 3-0-rhamnosyl-erythronolide B was filter sterilised and fed to cultures of SGT2pSGSpnIKH using standard microbiological techniques. The new compound was isolated from the supernatant of these cultures and analysed by NMR using the methods described for the preparation of 3-O-(2'-O-methylrhamnosyl) erythronolide B. The novel compound was characterised as 3-0- (2', 3'-bis-O-methylrhamnosyl) erythronolide B (Fig. 14).

Table 6: 1H and 13C NMR data for 3-O-(2', 3'-bis-O-methylrhamnosyl)erythronolide B <BR> <BR> <BR> <BR> <BR> <BR> <BR> Position 5F4 Multiplicity Coupling<BR> <BR> <BR> <BR> 1 175. 3 2 2. 87 dq 10. 2, 6. 8 44. 4 3 3. 37 br. d 10. 2 87. 5 4 2. 15 qdd 7. 3, 3. 9, 0. 9 36. 3 5 3. 51 d 3. 8 80. 4 6 74. 4 7 1. 93 dd 14. 8, 10. 4 36. 4 1. 44 dd 14. 7, 2. 7 8 2. 71 dqd 13. 1, 7. 1, 2. 8 44. 7 9 219. 3 10 3. 04 qd 6. 8, 1. 7 39. 4 11 3. 97 dd 10. 2, 1. 7 69. 6 12 1. 65 dqd 10. 4, 7. 2, 0. 9 39. 8 13 5. 44 ddd 9. 8, 4. 7, 0. 9-74. 7 14 1. 72 ddq 14. 0, 9. 6, 7. 3 25. 6 1. 50 dqd 14. 0, 7. 5, 4. 6 15 0. 88 dd 7. 4, 7. 4 9. 4 16 1. 23 d 6. 9 14. 6 17 1. 03 d 7. 4 7. 7 18 1. 34 s 25. 1 19 1. 14 d 7. 1 17. 5 20 0. 96 d 6. 8 8. 2 21 0. 94 d 7. 1 8. 2 1'4. 95 d. 1. 7 99. 5 2'3. 75 dd 3. 2, 1. 9 76. 8 3'. 3.37 dd 9. 4, 3. 0 80. 5 4'3. 43 dd 9. 4, 9. 4 71. 3 5'3. 85 dq 9. 4, 6. 4 69. 3 6'1. 26 d 6. 2 16. 5 7'3. 45 s 57. 6 8'3. 48 s 56. 5 A small peak with m/z 613 was detected and the MS/MS analysis indicated that this peak represented 3-O-(2', 3', 4'-tris-0-methylrhamnosyl) erythronolide B (Fig. 15).

Formation of 3-0-rhamnosylerythromycins and 3-0-rhamnosyl-6- deoxyerythromycins The plasmid pSGCIII was transformed into S. erythraea SGT2 cells to produce mutant strain SGT2, pSGCIII and culture broths of the transformed strain were analysed as described by Gaisser et al. (1997) and by Gaisser et al. (1988).

Supernatants from culture broths of S. erythraea mutant DM (pSGOLEG2) containing 3-0-rhamnosyl-erythronolide B were fed to SGT2pSGCIII cells. Analysis of the supernants using electrospray mass spectrometry, revealed the presence of a new peak at m/z 706 corresponding to 3-0-rhamnosyl-erythromycin D.

Both compounds were isolated using a gene cassette approach as described below.

Isolation of 3-0-(2'-0-methylrhamnosyl) erythromycin D 11 of DMpSGOleG2 culture supernatant which contained 3-0-rhamnosyl- erythronolide B was filter sterilised and fed to cultures of SGT2pSGSpnI using standard microbiological techniques. The culture supernatant was analysed for 3-O- (2'-O-methylrhamnosyl) erythronolide B and extracted as described in Materials and Methods. The crude extract was dissolved in 1 ml methanol and added to cultures of SGT2pSGeryCIII followed by an incubation at 30°C for four days. The supernatant was analysed and a major peak at m/z 720 was detected. The novel compound was analysed using the same methods as described for the preparation of 3-O-(2'-O- methylrhamnosyl) erythronolide B. The novel compound was identified as 3-0- (2'-0- methylrhamnosyl) erythromycin D (Fig. 16B).

Table 7: 1H and 13C NMR data for 3-O-(2'-O-methylrhamnosyl)erythromycin D <BR> <BR> Position #H Multiplicity Coupling #C 1 175. 9 2 2. 93 dq 9. 4, 7. 3 44. 6 3 4. 19 d 9. 0 82. 1 4 2. 16 m 39. 4 5 3. 59 overlaps with 2'83. 7 6 73. 8 7 1. 99 dd 14. 9, 8. 5 37. 7 1. 54 dd 14. 9, 4. 7 8 2. 80 m 42. 8 9 218. 7 10 2. 98 qd 6. 8, 1. 7 40. 4 11 3. 99 dd 10. 2, 1. 3 69. 2 12 1. 65 dq 9. 8, 7. 3 40. 1 13 5.35 ddd 9. 4, 4. 7, 0. 9 74. 8 14 1.74 ddq 14. 1, 9. 4, 7. 3 25. 2 1. 51 dqd 14. 1, 7. 3, 4. 7 15 0. 89 dd 7. 3, 7. 3 9. 2 16 1. 25 d 6. 8 14. 8 17 1. 12 d 7. 3 8. 3 18 1. 42 s 26. 1 19 1. 12 d 6. 8 8. 3 20 0. 97 d 6. 8 7. 6 21 0.93 d 7.3 8. 2 1' 5.01 d 1.3 98. 0 2' 3.59 overlaps with 5 80. 6 3' 3.71 dd 9. 4, 3. 4 70. 6 4'. 3. 40 dd 9. 4, 9. 4 72. 1 5'3. 75 dq 9. 4, 6. 0 69. 8 6'1. 32 d 6. 4 17. 3 7'3. 46 s 57. 6 1" 4.38 d 7.3 103. 0 2"3. 27 dd 10. 7, 7.3 70. 7 3 2. 80 m 63. 9 4"1. 77 m 30. 2 1. 28 m 5'3. 75 dq 9. 0, 6. 0 67. 7 6" 1.20 d 6.0 20. 2 7"2. 43 s 38. 9

Bioactivity of 3-0- (2'-O-methylrhamnosyl) erythromycin D Bacillus subtilis ATCC 6633 was used in bioassays as described previously (Gaisser et al., 1998). To assess the bioactivity of 3-0- (2'-O-methylrhamnosyl) erythromycin D, 1. 1 mg aliquots of erythromycin A (Sigma) and of3-0- (2'-0- methylrhamnosyl) erythromycin D were each dissolved in 200 ul of methanol, and series of 10-fold dilutions were prepared. Filter discs were soaked with 10 ul of these solutions and placed on 2 x TY plates overlaid with agar inoculated with an overnight culture of B. subtilis as described previously (Gaisser et al., 1997). The development of zones of inhibition in the B. subtilis lawn was assessed. The size of the halos in the bacterial lawn around the filter discs indicated that the bioactivity of3-0- (2'-0- methylrhamnosyl) erythromycin D was about 100-fold less compared to erythromycin A with Bacillus subtilis ATCC 6633 as indicator strain (Fig. 17).

Isolation of 3-0-(2'3'-bis-O-methylrharnnosyl) erythromycin D 11 of DMpSGOleg2 culture supernatant which contained 3-0-rhamnosyl- erythronolide B was filter sterilised and fed to cultures of SGT2pSGSpnHKI using standard microbiological techniques. The culture supernatant was analysed for 3-O- (2', 3'-bis-O-methylrhamnosyl) erythronolide B and extracted as described in Materials and Methods. The fraction which contained 3-0-(2', 3'-bis-O- methylrhamnosyl) erythronolide B was isolated. The dried extract of3-0- (2', 3'-bis-0- methylrhamnosyl) erythronolide B was dissolved in 1 ml methanol and added to cultures of SGT2pSGeryCIII followed by an incubation at 30°C for four days. The supernatant was analysed. A major peak with m/z of 734 was detected. The compound was isolated and analysed using the methods described for the preparation of 3-0- (2'-0-methylrhamnosyl) erythronolide B. The novel compound was identified as 3-0- (2', 3'-bis-O-methylrhamnosyl) erythromycin D (Fig. 16C).

Table 8: 1H and 13C NMR data for 3-O-(2',3'-bis-O-methylrhamnosyl)erythromycin D Multiplicity Coupling 1 176. 0 2 2. 93 dq 9. 4, 6. 8 44. 8 3 4. 20 dd 9. 4, 0. 9 82. 5 4 2. 18 dq 7. 7, 7. 3 39. 2 5 3. 61 d 7. 7 84. 0 6 74. 2 7 1. 99 dd 14. 9, 8. 5 37. 8 1. 52 dd 14. 9, 4. 7 8 2. 81 m 42. 8 9 219. 2 10 2. 96 qd 6. 8, 1. 7 40. 5 11 3. 99 dd 10. 2, 1. 7 69. 4 12 1. 65 dqd 10. 2, 7. 3, 0. 9 39. 9 13 5.36 ddd 9. 4, 4. 7, 1. 3 74. 9 14 1. 74 ddq 14. 1, 9. 4, 7. 3 25. 4 1. 50 dqd 14.1, 7. 3, 4. 7 15 0. 90 dd 7.3, 7. 3 9. 3 16 1. 27 d 7.3 14. 8 17 1. 13 d 7. 3 8. 4 18 1. 43 s 26. 3 19 1. 11 d 6. 8 17. 2 20 0. 97 d 6. 8 7. 7 21 0. 92 d 7. 3 8. 3 1' 5.02 d 2.1 98. 4 2'3. 80dd2. 6, 2. 6 76. 6 3' 3.39 dd 9.0, 3. 0 80. 4 4' 3.49 dd 9.0, 9. 0 70. 9 5'3. 75 dq 9.0, 6. 4 70. 2 6'1. 32 d 6. 4 17. 2 7'3. 49 s 57. 5 8'3. 46 s 56. 3 1"4. 44 d 7. 3 102. 5 2" 3.37 dd 7. 3, 3. 4 69. 9 3" 3.21 m 65. 0 4"1. 91 m 29. 9 1. 42 m 5"3. 75 m 70. 0 6 1. 25 d 6. 4 20. 1 7"2. 68 s 38. 5

Construction of the Saccharopolyspora erythraea strain SGT 3 (heryCIIIzJeryB VZl eryBVl) To prevent contamination with mycarosyl-erythronolide B in feeding assays, the Saccharopolyspora erythaea strain SGT3 (#eryCIII#eryBV#eryBVI) was isolated using plasmid pHhol (Gaisser et al., 1997). The transformation of the S. erythraea strain DM and the isolation of the mutant SGT3 were performed as described (Gaisser et al., 1997). To investigate the thiostrepton sensitive mutants no 31, 33, 34 and 25, chromosomal DNA was analysed using PCR analysis. Chromosomal DNA was subjected to PCR using the primers as described earlier (Gaisser et al., 1997). The expected 360 bp fragment was amplified from wild type DNA and two bands of roughly 100 and 300 bp of size were detected after PstI restriction digest. In samples with the chromosomal DNA of SGT3 as a template, a 300 bp fragment was amplified which was found to be resistant to digestion by PstI. This result indicates the introduction of a 60 bp deletion in eryBVI into the genome of SGT3. This strain is used as background for the expression of the gene cassettes. described below.

Construction of expression plasmidfor oleP For expression of oleP, the primers Ole 15- CTCCAGCAAAGGACACACCCATATGACCGATACGCACA-3'and OleP2 5'- CGGCAGATCTGCCGGCCGTCACCAGGAGACGATCTGG-3'were used to amplify oleP using plasmid 3gh2 as a template. The PCR fragment was isolated, treated with T4 polynucleotide kinase and cloned into Smal cut pUC18. After transformation into E. coli DH1OB the construct was isolated and the sequence of oleo was verified. After digestion with NdeVBgllI, a 1. 3 kb fragment was isolated, ligated with the vector fragment of NdeVBglJI digested pSG142 and used to transform E. coli DH10B. Plasmid pSGOleP was isolated.

Feeding of erythronolides to strain SGT2pSGOleP Saccharopolyspora erythraea SGT2 (Gaisser et al., 2000) was transformed with the plasmid construct pSGOleP as described in Materials and Methods. The transformants were verified by isolating chromosomal DNA followed by PCR analysis. Feeding

experiments using 6-deoxyerythronolide B, erythronolide B, 3-O-mycarosyl- erythronolide B, 3-O-rhamnosyl-erythronolide B (Gaisser et al., 2000), and erythromycin A were carried out as described (Gaisser et aL, 1997). The cultures of the strains SGT2 and SGT2pSGOleP were fed with these compounds, incubated at 30 °C for 3 to 5 days and analysed using electrospray mass spectrometry. Novel peaks were visible in supernatants of SGT2pSGOleP fed with 6-deoxyerythronolide B (m/z 434 [M] NH4+), 3-O-mycarosyl-erythronolide B (m/z 578 [M] NH4+ and m/z 580 [M] NH4+) and 3-0-rhamnosyl-erythronolide B (m/z 580 [M] NH4+ and m/z 582 [M] NH4+). Novel peaks could not be detected in supernatants that contained erythronolide B and erythromycin A. MS/MS analysis of these new compounds indicated the presence of 8, 8a-epoxy- or 8, 8a-dihydroxy derivatives of 6-deoxy- erythronolide B, 3-O-mycarosyl-erythronolide B, and 3-0-rhamnosyl-erythronolide B.

Preparation of 8a-hydroxy-3-O-mycarosyl erythronolide B and 8-epoxy-3-0- mycarosyl erythronolide B 2. 5 1 of culture supernatant of SGT2pSGOleP fed with 60 mg of 3-0-mycarosyl- erythronolide B were grown and the novel compounds were isolated using methods described in Materials and Methods. The structures of these compounds were confirmed by NMR analysis using the methods described for the preparation of 3-O- (2'-O-methylrhamnosyl) erythronolide B. The 3-0-mycarosyl erythronolide B derived compounds were identified as 8a-hydroxy-3-0-mycarosyl erythronolide B and 8, 8a- epoxy-3-O-mycarosyl erythronolide B (Fig 21).

Table 9: 1H NMR Data for 8a-hydroxy-3-O-mycarosyl erythronolide B Proton #H multiplicity coupling (Hz) 2-H 2. 87 dq 10. 1, 7. 0 3-H 3. 75 dd 10. 0, 1. 2 4-H 2. 15 m 5-H 3. 55 d 3. 5 7-Ha 1. 49 dd 14. 6, 3. 0 7-Hb 1. 84 dd 14. 6, 10. 3 8-H 2. 83 m 8a-Ha 3. 66 dd 11. 0, 6. 0 8a-Hb 3. 71 dd 11. 0, 9. 1 10-H 3. 07 qd 6. 9, 1. 8 11-H 3. 93 dd 10. 2, 1. 8 12-H 1. 66 m 13-H 5. 44 ddd 9. 5, 4. 9, 1. 2 14-Ha 1. 49 dqd 16. 9, 7. 3, 4. 8 14-Hb 1. 73 ddq 16. 9, 9. 5, 7. 3 15-H0. 88d 7. 3- 16-H3 1. 18 d 6. 9 17-H3 1. 02 d 7. 3 18-H3 1. 33 s 19-H3 0. 96 d 6. 8 20-H3 0. 94 d 7. 1 1'-H 5.02 dd 3. 9, 1. 1 2'-Ha 1. 84 dd 14. 5, 4. 2 2'-Hb 2. 08 dd 14. 5, 1. 2 4'-H 2. 98 d 9. 7 5'-H4. 06dq 9. 8, 6. 2 6'-H3 1. 28 d 6. 2 7'-H3 1. 22 s

Table 10 :"C NMR Datafor 8a-hydroxy-3-0-mycarosyl erythronolide B Carbon 1 175. 3 2 44. 4 3 86. 4 4 36. 3 5 80. 1 6 74. 4 7 31. 0 8 52. 9 8a 63. 5 9 217. 3 10 40. 6 11 69. 6 12 39. 6 13 74. 6 14 25. 4 15 9. 3 16 14. 4 17 7. 5 18 24. 7 19 7. 2 20 8. 2 1'99. 2 2'40. 8 3'69. 5 4'76. 3 5'65. 5 6'16. 8 7'24. 7 Table 11: 1H NMR Data for 8,8a-epoxy-3-O-mycarosyl erythronolide B Proton 6 multiplicity coupling (Hz) 2-H 2. 89 dq 10. 4, 7. 0 3-H 3. 74 dd 10. 4, 1. 3 4-H 2. 21 m 5-H 3. 48 d 3. 4 7-Ha 1. 50 d 14. 9 7-Hb 2. 62 d 14. 9 8a-Ha 2. 52 d 5.5 8a-Hb 2. 67 d 5. 5 10-H 3.12 qd 6. 8, 1. 9 11-H 4. 22 dd 10. 4, 1. 9 12-H 1. 67 qd 7. 0, 1. 3 13-H 5. 47 ddd 9. 8, 4. 7, 1. 3 14-Ha 1. 51 m 14-Hb 1. 75 m 15-H3 0. 88 dd 7. 3, 7. 3 16-H3 1. 20 d 7. 3 <BR> <BR> 17-H3 1. 03 d 7. 3<BR> <BR> <BR> <BR> 18-H3 1. 43 s 19-H3 0. 97 d 6. 8 20-H3 0. 95 d 7. 0 1'-H 5.02 dd 4. 2, 1. 3 2'-Ha 1. 84 dd 14. 5, 4. 3 2'-Hb 2. 08 dd 14. 5, 1. 3 4'-H 2. 98 d 9. 8 5'-H 4. 06 dq 9. 8, 6. 2 6'-H3 1. 27 d 6. 2 7'-H3 1. 22 s

Table 12: 13C NMR Data for 8,8a-epoxy-3-O-mycarosyl erythronolide B Carbon 1 175. 4 2 44. 4 3 86. 7 4 36. 2 5 81. 0 6 75. 2 7 35. 0 8 62. 5 8a 50. 0 9 210. 5 10 44. 1 11 68. 9 12 39. 5 13 74. 6 14 25. 5 15 9. 4 16 14. 6 17 7. 5 18 25. 9 19 7. 7 20 8. 2 1'99. 4 2'40. 7 3'69. 5 4'76. 2 5'65. 3 6'16. 8 7'24. 6

Preparation of 8, 8a-dehydro-6-deoxyerythronolide B and 8-hydroxy-6- deoxyerythronolide B Plasmid pSGOleP was used to transform the 6-deoxyerythronolide B producer strain S. erythraea SGT1 (AeryBV, AefyC7Z/, AeryF) as described in Materials and Methods.

The transformants were verified by isolating chromosomal DNA followed by PCR analysis. Cultures of SGT1 and SGTIpSGOleP were grown as described previously (Gaisser et al., 2000) and the supernatants were analysed using electrospray mass spectrometry as described in Materials and Methods. Two major compounds in the supernatant were purified and analysed using NMR techniques as described for the preparation of 3-0- (2'-O-methylrhamnosyl) erythronolide B. The product with m/z 401 [M-H2O] H+was identified as the 8, 8a-dihydroxy derivative of 6-deoxyerythronolide B recently disclosed (Shah et al., 2000). The compound with m/z of 385 [M] H was confirmed as 8, 8a-dehydro-6-deoxyerythronolide B (Fig 22). The structure of a further, minor compound of the culture supernatant was identified as 8-hydroxy-6- deoxyerythronolide B (Fig 22).

Table 13 :'HNMR Datafor 8, 8a-dehydro-6-deoxyerythronolide B Proton #H multiplicity coupling (Hz) 2-H 2. 70 dq 9. 8, 6. 8 3-H 3. 55 ovrlp 4-H 1. 69 ovrlp <BR> <BR> 5-H 3. 55 ovrlp<BR> <BR> <BR> <BR> <BR> <BR> 6-H 1. 91 m 7~Ha 2. 11 dd 17. 1, 7. 7 7-Hb 2. 45 dd 17. 1, 3. 8 8a-Ha 5. 38 s 8a-Hb 5. 66 s 10-H3. 23dq 6. 8, 1. 7 11-H 3. 60 dd 10. 2, 1. 7 12-H 1. 70 ovrlp 13-H 5. 24 ddd 9. 4, 4. 7, 2. 1 14-Ha 1. 52 ddq 14. 1, 9. 4, 7. 3 14-Hb 1. 77 dqd 14. 1, 7. 3, 4. 7 15-H3 0. 90 dd 7. 3, 7. 3 16-H3 1. 18 d 6. 8 17-H3 1. 03 d 6. 8 18-H3 1. 12 d 6. 8 19-H3 0. 97 d 6. 8 20-H3 0. 96 d 7. 3

Table 13: 13C NMR Data for 8,8a-dehydro-6-deoxyerythronolide B Carbon 6 1 178. 7 2 45. 3 3 75. 5 4 42. 3 5 77. 7 6 35. 9 7 33. 2 8 150. 4 8a 120. 6 9 208. 8 10 44. 9 11 71. 9 12 42. 1 13 76. 9 14 26. 7 15 10. 9 16 15. 2 17 8. 9 18 20. 2 19 6. 5 20 10. 0

Table 14 :'HNMR Datafor 8-hydroxy-6-deoxyerythronolide B (CDCI) Proton #H multiplicity coupling (Hz) 2-H 2. 63 dq 10. 2, 6. 8 3-H 3. 61 dd 10. 2, 3. 0 4-H 1. 56 m 5-H 3. 53 br. s 6-H 1. 40 m 7-Ha 1. 74 m 7-Hb 1. 96 dd 14. 9, 3. 4 10-H 3. 05 qd 6. 8, 0. 9 11-H 3.45 d 9. 8 12-H 1. 72 m 13-H 5. 45 ddd 9. 4, 4. 7, 0. 9 14-Ha 1. 50 m 14-Hb 1. 72 m 15-H3 0. 90 t 7. 3, 7. 3 16-H3 1. 25 d 6. 8 17-H3 1. 05 d 7. 3 18-H3 1. 15 d 6. 8 19-H3 1. 42 s 20-H3 1. 11 d 6. 8 21-H3 0. 90 d 7. 3

Table 15: 13C NMR Data for 8-hydroxy-6-deoxyerythronolide B (CDCl3) Carbon bx 1 175. 6 2 43. 9 3 77. 8 4 41. 3 5 79. 4 6 36. 0 7 39. 2 8 79. 7 9 218. 7 10 38. 8 11 69. 4 12 40. 1 13 75. 2 14 25. 6 15 10. 3 16 14. 8 17 7. 7 18 20. 2 19 26. 9 20 9. 6 21 8. 9

Strategy to isolate gene cassettes The strategy to prepare gene cassettes with different combinations of glycosyltransferase-and methyltransferase genes is adapted from a technique previously described (WO 077181 A2) to build gene cassettes expressed under the control of the actII-Orf4 regulator. The expression of these gene cassettes in a suitable strain background is a powerful approach to generate novel post-PKS modified polyketides in a random or directed fashion. The method is based upon the introduction of XbaI restriction sites at the 3'and 5'-end of the PCR fragments. The introduction of a XbaI site at the 5'-end of the PCR fragment which is sensitive to the Dam methylase of the strain background will protect this site from furthered digest.

To retain the Shine Dalgarno sequence 5'of the respective gene the pSG142 derived constructs which contain these genes were used as a template. Using plasmid DNA isolated from dam~ host strains such as E. coli ET12567, the amplified genes were isolated as XbaI fragments. Using a host strain with an active Dam methylase such as E. coli DM1OB these fragments were sequentially cloned into gene cassettes. This technique provides the means to build gene cassettes of different length and different order using the same strategy over and over again. An overview of the strategy described here and the isolated gene cassettes is depicted in Fig. 18.

The following example of this methodology was based upon the isolation of a PCR fragment of oleG2 into which a HindIII and a NdeI restriction site was introduced at the 5'-end of the fragment and a XaI, BglII and EcoRI site at the 3'-end of the PCR fragment. This fragment was digested using the restriction enzymes HindIII and EcoRI and it was cloned into pUC 19 which was digested identically. Plasmid pSGcasoleG2 was isolated. The genes synI, spunk, spnH and eryCIII were amplified using PCR techniques. A XhaI restriction site was introduced at the 5'-end which is sensitive to methylation by the Dam methylase of the strain background. At the 3'- end aXbaI site was introduced. The pSG142 derived constructs which contain these genes were used as a template. The PCR fragments were treated with T4 polynucleotide kinase as described above and cloned into SmaI cut pUC18. The DNA sequences of these clones were confirmed by sequencing analysis. After transforming

the constructs into a dam-strain background, the DNA was isolated and digested using XbaI. The Xbal fragments of the inserts of around 0. 8-1. 3 kb of size were isolated and ligated into thé goal cut pSGcasOleG2. After building the gene cassettes in pUC19, each construct was digested using the restriction enzymes NdeI/BglII and the DNA fragment encoding the gene cassette was isolated and cloned into the NdeI/ BglII digested vector DNA of pSG142. These plasmids were transformed into SGT3. The transformants were analysed as described above.

The following primers were used : <BR> <BR> <BR> <BR> casoleG21 5'GGGGAAGCTTGCCGACGATGACGACGACCACCGGACGAACGCATCGATTAATTAAG&l t;BR> <BR> <BR> <BR> <BR> <BR> casoleG22 5'GGGGAATTCAGATCTGGTCTAGAGGTCAGCCCGCATGGTTCCCGCCTCCTCGTCCGC GTCCGCCGCT casspnl3 5'GGGTCTAGATCCGGACGAACGCATCGATTAATTAAGGAGGACAGA. TATGAGTGAGA, TCGCAGTTGCCCC casspnI4 5'GGGGTCTAGAGGTCAGCCGCCCTCGACGCCGAGCGCTTGCCGGGGCACGAACCCCGG CGCGGCAGGCT <BR> <BR> <BR> <BR> <BR> <BR> <BR> casspnK1 5'GGGTCTAGATCCGGACGAACGCATCGATTAATTAAGGAGGACAGATATGTCCACAAC GCACGAGATCG casspnK2 5'GGGGTCTAGAGGTCACTCGTCCTCCGCGCTGTTCACGTCGGCCAGGTGCAATATGTC <BR> <BR> <BR> <BR> casewCIII1 5'GGGTCTAGATCCGGACGAACGCATCGATTAATTAAGGAGGACAGATATGCGCGTCGT& lt;BR> <BR> <BR> <BR> <BR> CTTCTCCTC<BR> <BR> <BR> <BR> <BR> <BR> <BR> <BR> <BR> <BR> caseryCIII2 5'GGGGTCTAGAGGTCATCGTGGTTCTCTCCTTCCTGCGGCCAGTTCCTCGCA Analysis of SGT3pSGcasoleG2spnI The clone SGT3pSGcasoleG2spnI was isolated using the approach described above.

The cells were grown as described in Materials and Methods and the culture supernatant was analysed. As expected, 3-0- (2'-0-methylrhamnosyl) erythronolide B and 3-0-(2'-O-methylrhamnosyl)-6-deoxyerythronolide B were detected.

Analysis of SGT3pSGcasoleG2spnIspnK

SGT3pSGcasoleG2spnIspnK was isolated using the approach described above. The cells were grown as described in Materials and Methods and the culture supernatant was analysed. As expected, 3-0-(2'-O-methylrhamnosyl) erythronolide B and 3-0- (2', 3'-bis-O-methylrhamnosyl) erythronolide B were detected (Fig. 19). The culture supernatant also contained a further novel compound which was isolated and the structure was confirmed by NMR analysis using the methods described for the preparation of 3-0-(2'-O-methylrhamnosyl) erythronolide B. The structure of the novel compound was characterised as 3-0-(2, 3-bis-O-methylrhamnosyl)-6- deoxyerythronolide B (Fig. 19).

Table 16: 1H and 13C NMR data for 3-O-(2',3'-bis-O-methylrhamnosyl)-6-dEB <BR> <BR> <BR> <BR> <BR> <BR> Position 814 Multiplicity Coupling 8c 1 178. 4 2 2. 89 dq 8. 1, 7. 3 46. 2 3 3. 37 overlap 84. 1 4 1. 68 m 43. 1 5 3. 49 d 9. 2 77. 8 6 1. 66 m 37. 1 7 1. 01 m 35 1. 86 m 8 2. 68 m 45. 8 9 218 10 2. 95 qd 6. 8, 1. 9 43. 8 11 3. 73 dd 10. 2, 1. 7 72. 1 12 1. 65 m 42 13 5. 25 ddd 9. 2, 5. 3, 1. 3 77. 4 14 1. 54 m 26. 5 1. 80 m 15 0. 92 dd 7. 3, 7. 3 10. 8 16 1. 27 d 6. 4 15. 3 17 1. 08 d 7. 0 10 18 1.16 d 6.4 20. 6 19 1. 13 d 6. 6 16. 4 20 0. 98 d 6. 8 7. 5 21 0.92 d 7.0 9. 7 1'4. 97 d 1.5 100.7 2' 3.77 overlap 78. 1 3' 3.33 dd 9.4, 3. 0 81. 8 4'3. 42 dd 9.4, 9.4 72. 8 5'3. 67 dq 9.4, 6.2 70. 9 6'1. 25 d 6. 0 18. 1 7'3. 45 s 58. 9 8' 3.47 s 57. 9

Analysis of SGT3pSGoleG2spnlspnKeryCIII SGT3pSGoleG2spnIspnKeryCIII was isolated using the methods described above.

The cells were grown as described in Materials and Methods. The culture supernatants of SGT3pSGcasoleG2spnIspnKeryCIII contained only small amounts of compounds with an attached desosamine sugar residue.

Analysis of SGT3pSGoleG2spnIspnKeryCIIIhis The PCR product using the primer combination of caseryCIII2 and SG14 was isolated using the methods described for the construction of expression plasmid for eryCIII.

The gene cassette pSGcasoleG2spnIspnKeryCIIIhis was created using the approach described above. Strain SGT3pSGcasoleG2spnIspnKeryCIIIhis was isolated and cells were grown as described in Materials and Methods. The culture supernatant of the clone was assessed using techniques described in Materials and Methods.

Compounds 3-O-mannosyl erythromycin D, 3-0-(2'-O-methyl rhamnosyl erythromycin D, 3-0- (2', 3'-bis-0-methyl rhamnosyl erythromycin D), 3-0- (2'-0- methyl rhamnosyl)-6-deoxyerythromycin D and 3-0-(2', 3'-bis-O-methyl rhamnosyl)- 6-deoxy erythromycin D were detected (Fig. 20). The introduction of the his6-tag at the C-terminus of EryCIII therefore seems to improve glycosyl transfer of the desosamine sugar residue to its substrates. This result indicates that the expression of the last gene of the gene cassette can be improved by introducing the his6-tag fusion at the C-terminal end of the protein.

OIeP cassette To include oleP into the arrangements of gene cassettes, primers OlePcassl 5'- GGGTCTAGATCCGGACGAACGCATCGATTAATTAAGGAGGACAGATATGA CCGATACGCACACCGGACCGACACC-3'and OlePcass2 5'- GGGGTCTAGAGGTCACCAGGAGACGATCTGGCGTTCCAGTCCGCGGATCA -3'were used. The PCR fragment-using plasmid pSGOleP as template-was isolated, treated with T4 polynucleotide kinase and cloned into SmaI cut pUC18. After transformation into E. coli DH10B the construct was isolated and verified. Plasmid pSGOlePcass was used to transform the dam-Escherichia coli strain ET12567. The

plasmid DNA was isolated and after digestion withxbal, a 1. 3 kb fragment was isolated, ligated with the vector fragment ofxbal digested constructs and used to transform E. coli DH10B. An overview over these constructs is given in Fig. 23.

Preparation of 3-0-rhamnosyl-8, 8a-dehydro-6-deoxyerythronolide B S. erythraea strain SGTIpSGcasoleG2oleP was isolated using the methods described above and 4 1 of cells were grown as described in Material and Methods. The culture supernatant were isolated and analysed as described for the preparation of 3-0- (2'-0- methylrhamnosyl) erythronolide B. Two novel compounds, 3-O-rhamnosyl-8, 8a-dehydro-6-deoxyerythronolide B and 3-O-rhamnosyl-8, 8a- dihydroxy-6-deoxyerythronolide B were detected (Fig. 24). The structure of 3-O- rhamnosyl-8, 8a-dehydro-6-deoxyerythronolide B was confirmed by NMR analysis.

Table 17 :'HNMR Datafor 3-0-rhamnosyl-8, 8a-dehydro-6-deoxyerythronolide B Proton #H multiplicity coupling (Hz) 2-H 2. 80 dq 7. 3, 7. 3 3-H 3. 76 dd 7. 5, 2. 6 4-H 1. 66 m 5-H 3. 48 dd 7. 9, 2. 6 6-H 2. 00 m 7-Ha 2. 18 m 7-Hb 2. 39 d 16. 4 8a-Ha 5. 42 s 8a-Hb 5. 68 s 10-H 3.21 m 11-H 3. 68 m 12-H 1. 72 m 13-H 5. 18 ddd 9. 2, 4. 9, 1. 3 14-Ha 1. 53 m 14-Hb 1. 77 m 15-H3 0. 90 dd 7. 3, 7. 3 16-H3 1. 20 d 7. 3 17-H3 1. 06 d 6. 8 18-H3 1. 12 d 6. 8 19-H3 0. 98 d 6. 8 20-H3 0. 96 d 7. 3 1'-H 4. 83 obscured 2'-H 3. 96 dd 3. 2, 1. 7 3'-H 3. 62 dd 4'-H 3. 42 dd 9. 6, 9. 6 5'-H 3. 69 m 6'-H3 1.28 d 6. 2

Table 18 :/3C NMR Data for 3-0-rhamnosyl-8, 8a-dehydro-6-deoxyerythronolide B Carbon 1 177. 3 2 45. 0 3 81. 1 4 42. 0 5 75. 5 6 33. 3 7 31. 2 8 148. 4 8a 119. 3 9 210. 2 10 43. 8 11 71. 1 12 40. 5 13 76. 0 14 25. 1 15 9. 3 16 13. 7 17 8. 6 18 18. 3 19 8. 8 20 8. 8 1'102. 4 2'70. 5 3'70. 9 4'72. 2 5'69. 3 6'16. 6

Costruction of expression plasmidfor oleGl To establish which of the various possible start codons in the published sequence (accession number AJ002638) is used for the expression of oleGl (Fig. 25), various constructs were tested by measuring the complementation of the eryCIII mutation in S. erythraea SGT2 after feeding with 3-0-mycarosyl erythronolide B using techniques described in Materials and Methods. Complementation indicated by the production of small amounts of erythromycin A was only observed when vector pSGOleGl was used (Fig. 26). Plasmid pSGOleGl was isolated using the primers 7390 5'-CCGCCATATGAGCATCGCGTCGAACGGCGCGCGCTCGGC-3' Ole2 5'-TCAGATCTCCGCCTTCCCGCCATCGCGCCGGTGGCAT-3'to amplify oleGl. The cloning procedure was as described for the construction of the expression plasmid for oleG2. Expression vectors using the published start codon or one of the following ATG codons indicated in Fig. 25 did not complement the eryCIII mutation of SGT2 after feeding with 3-0-mycarosyl erythronolide B. This result indicates that the correct start codon which is required for the expression of oleGl is the ATG overlapping with the olePl stop codon (Fig. 25).

6-deoxyerythronolide B as a substrate for oleG2, but not oleGl The S. erythraea mutant SGQ1 (SGT2DeryF) was created starting with SGT2 by introducing a deletion in the eryF gene as described above using standard microbiological techniques. SGQ1 was transformed with the plasmid constructs pSGOleG2, pSGOleP and pSGOleGl and feeding experiments using the sterile filtered culture supernatants of SGTI containing 6-deoxyerythronolide B were carried out as described in Materials and Methods, The results indicate, that both, OleG2 and to a smaller extent OleP, accept 6-deoxyerythronolide B as a substrate (Fig. 27). 6- deoxyerythronolide B is not a substrate for OleGl.

3-0-y°hamnosyl-6-deoxyerythronolide B as a substrate for oleP Culture supernatants of SGTIpSGOleG2 containing 3-O-rhamnosyl-6- deoxyerythronolide B was sterile filtered and were added to cultures of SGQ1, SGQIpSGOleG2, SGQIpSGOleP and SGQIpSGOleG1 using standard

microbiological techniques. The analysis of these culture supernatants indicates that 3- O-rhamnosyl-6-deoxyerythronolide B is a substrate for OleP but not for OleGl (Fig 28).

Feeding of 8, 8a-epoxy-3-0-mycarosyl erythronolide B Cell cultures of SGTIpSGOleG1 were grown as described in Material and Methods and fed with 8, 8a-epoxy-3-O-mycarosyl erythronolide B as described earlier (Gaisser et al., 2000). The culture supernatant was analysed using techniques described above and the results indicate that 8, 8a-epoxy-3-O-mycarosyl erythronolide B is a substrate for OleGl.

Expression of pSGcassOleG2EryCIII in SGT3 The gene oleG2 was amplified using the primers casoleG21 and casoleG22 (see above) and the DNA of plasmid pSGOleG2 as a template (Gaisser et al., 2000). The PCR product was ligated into SmaI cut pUC 18 and transformed into the Escherichia coli strain DH10B. The sequence of the PCR product was verified. The resulting plasmid was digested using the restriction enzymes EcoRI and HindIII followed by a ligation into EcoRIHindIII digested pUC19 and transformation of E. coli DH10B.

The resulting plasmid was named pSGcassOleG2.

The gene eryCIII was amplified using the primers caseryCIII and caseryCIII2 (see above) and the DNA of plasmid pSGEryCIII as a template. The PCR product was ligated into SmaI cut pUC 18 and transformed into the Escherichia coli strain DH1 OB.

The sequence of the PCR produce was verified. The resulting plasmid was transformed into the dam Escherichia coli strain ER12567. The DNA of the transformant was isolated and digested using the restriction enzyme Xbai. The 1. 3 kb DNA fragment was isolated and ligated into XbaI digested pSGcassOleG2 and transformed into E. coli DH10B. The correct orientation of eryCIII was assessed using restriction digests and plasmid pOleG2EryCIII was isolated. A DNA band of about 2. 8 kb was isolated after a restriction digest using NdeI, BgIII and DraI followed by a ligation into the NdeIlBglII digested expression vector pSG142 and

transformation of E. coliDHlOB. Plasmid pSGcassOleG2EryCIII was isolated and used to transform the S. erythraea mutant SGT3. Thiostrepton resistant colonies were selected. Culture supernatants of these strains were isolated as described (Gaisser et al., 1997) and analysed using electrospray mass spectrometry. A peak with the retention time of 9. 2 and the m/z of 706 was detected which indicates the presence of rhamnosyl-erthyromycin D in the supernatant. Another peak with m/z 690 was also found which indicates the presence of rhamnosyl-6-deoxyerythromycin D in the supernatant (see Fig 16).

Anti-microbial activity of the resulting erythromycin analogues was demonstrated through development of zones of inhibition in a lawn of erythromycin-sensitive Bacillus subtilis around plugs of the transformed cells in a standard bioassay.

Isolation of 5-0-glucosyl-and 5-0-desosaminyl-tylactone Analysis of the tylactone standard using electrospray mass spectrometry showed a major peak at m/z 377 and minor peaks at m/z 359 and m/z 417. When tylactone was supplied to strain SGT2, in which both eryBV and eryCIII are deleted, peaks at m/z 557 and 579 were detected in the culture supernatants which would correspond to glucosylated derivatives of tylactone. This confirmed the presence of another glycosyltransferase in S. erythraea which accepts tylactone as a substrate. However, the analysis of the culture supernatant of S. erythraea SGT2 (pSGTYLM2) fed with tylactone revealed a major peak at m/z 552, which fragmented into peaks at m/z 158 and m/z 359 in MS/MS experiments, indicating the presence of 5-O-desosaminyl-tylactone in the expression medium of S. erythraea SGT2 (pSGTYLM2). The putative glucosyl-and desosaminyl-tylactones (2. 2 mg and 2. 0 mg respectively from 1. 5 1 of culture broth) were analysed and the structures were fully confirmed as 5-O-glucosyl-and 5-O-desosaminyl-tylactone by using'H and 13C NMR (see Figs 5A and 5B respectively).

Table 19 :'HNMR Datafor 5-desosaminyl-tylactone Proton 5H Multiplicity Coupling 2-Ha 2. 49 dd 17. 4, 9. 6 2-Hb 2. 04 d 17. 4 3-H 3. 72 d 9. 6 4-H 1. 70 m 5-H 3. 77 d 9. 6 6-H 1. 11 m 7-Ha 1. 72 m 7-Hb 1. 45 m 8-H 2. 64 m 10-H 6.47 m 15. 4 11-H 7. 27 d 15. 4 13-H 5. 70 d 10. 4 14-H 2. 81 ddq 10. 2, 10. 2, 6. 5 15-H 4. 72 ddd 9. 7, 9. 7, 2. 5 16-Ha 1. 90 m 16-Hb 1. 60 m 17-H3 0. 96 dd 7. 4, 7. 4 18-H3 1. 04 D 6. 8 19-Ha 1. 60 m 19-Hb 1. 42 m 20-H3 0. 85 dd 7. 2, 7. 2 21-H3 1. 19 d 6. 9 22-H3 1. 84 br. s 23-H3 1. 08 d 6. 5 1'-H 4. 28 d 7. 3 2'-H 3. 37 dd 10. 5, 7. 3 3'-H 3. 20 ddd 11. 4, 11. 4, 4. 0 4'-Ha 1. 93 m 4'-Hb 1.44 m 5'-H 3. 61 dq 10. 2, 6. 1 6'-H3 1. 26 d 6. 1 7'(CH3)2 2.70 S

Table 20.'3C NMR Data for 5-desosaminyl-tylactone Carbon bc C1 173. 6 C2 39. 7 C3 67. 0 C4 40. 9 C5 78. 6 C6 39. 4 C7 33. 3 C8 44. 7 C9 205. 5 C10 118. 2 Cll 148. 1 C12 133. 3 C13 146. 0 C14 38.3 C15 78. 4 C16 23. 9 C17 8. 4 C18 8. 3 C19 20. 5 C20 10. 8 C21 16. 3 C22 11. 4 C23 14. 3 C'103. 3 C2'69. 3 C3'65. 0 C4'29. 7 C5' 67. 8 C6'19. 5 N7'-(CH3)2 38. 5

Table 21 :'HNMR Datafor 5-glucosyl-tylactone Proton bH Multiplicity Coupling 2-Ha 2. 49 dd 17. 4, 9. 6 2-Hb 2. 05 d 17. 4 3-H 3. 74 br. d 9. 9 4-H 1. 68 m 5-H 3. 77 br. d 9. 4 6-H 1. 11 m 7-Ha 1. 72 m 7-Hb 1. 48 m 8-H 2. 66 dqd 11. 8, 6. 9, 3. 6 10-H 6. 44 d 15. 5 11-H 7. 23 d 14. 9 13-H 5. 67 d 10 14-H 2.78 ddq 10.3, 10. 3, 6. 6 15-H 4. 68 ddd 9. 1, 9. 1, 2. 7 16-Ha 1. 86 m 16-Hb 1. 58 m 17-H3 0. 94 dd 7. 3, 7. 3 18-H3 1. 04 d 6. 9 19-Ha 1. 43 m 19-Hb 1. 64 m 20-H3 0. 87 t 7. 3 21-H3 1. 19 d 6. 9 22-H3 1.85 br. s 23-H3 1. 07 d 6. 6 1'-H 4. 28 d 7. 8 2'-H 3.17 dd 9. 2, 7. 8 3'-H 3. 33 m 4'-H 3.34 m 5'-H 3. 19 m 6'-Ha 3. 82 dd 11. 6, 2. 6 6'-Hb 3. 72 dd 11. 6, 4. 8

Table 22: 13C NMR Data for 5-glucosyl-tylactone Carbon bc C1 173. 6 C2 39. 8 C3 66. 9 C4 41. 1 C5 79. 1 C6 39. 6 C7 33. 4 C8 45. 1 C9 205. 7 C10 118. 6 Cll 148. 2 C12 133. 9 C13 146. 3 C14 38. 4 C15 78. 6 C16 24. 3 C17 8. 6 C18 8. 2 C19 21. 3 C20 10. 9 C21 16. 5 C22 11.1 C23 14. 9 C'103. 1 C2'74. 3 C3'76. 7 C4'70. 2 C5' 75. 9 C6'61. 2

Production of 23-hydroxy 5-O-mycaminosyl tylactone The S. erythraea strain SGT2pSGTylM2 was grown as described previously (Gaisser et al., 1997). Tylactone was fed to these cultures after 48h (compared to 24h in previous feedings). Analysis of the supernatants using electrospray mass spectroscopy revealed the presence of a new peak at m/z 552 which was identified as 5-O- desosaminyl-tylactone. A second peak with m/z 568 was also detected in this supernatant and MS/MS analysis of this compound confirmed the presence of 5-O- mycaminosyl-tylactone in the expression medium of S. erythaea strain SGT2pSGTylM2. An aliquot of this supernatant was used to feed cultures of SGT2pSGTYLH. Analysis of the supernatants revealed a shift of the 5-O- mycaminosyl-tylactone peak of m/z 568 to 584 m/z. This result indicates that 5-O- mycarninosyl-tylactone was further processed by the expression of the TyIH locus (Fouces et al., 1999). The TylH locus consists of two genes, tylHl (3Fe-4S-type ferrodoxin) and tylH2 (P450 type cytochrome), postulated to form the oxidoreduction system involved in C23 oxidation) to produce 23-hydroxy 5-O-mycaminosyl tylactone in the culture supernatant.

Production of 23-0-rhamnosyl S-0-mycaminosyl tylactone An aliquot of the supernatant containing 5-O-mycaminosyl-tylactone was used to feed cultures of S. erythraea SGT2pSGTYLHN. Analysis of the supernatants using electrospray mass spectrosopy revealed the presence of a new peak at rn/z 730. The shift by m/z 146 indicated the presence of 23-rhamnosyl 5-O-mycaminosyl tylactone in the culture supernatant.

Isolation of 5-0-(2'-0)-bis-glucosyl-tylactone Cultures of SGT2pSGOLED are fed with tylactone. The supernatants of these cultures contain a new product with m/z consistent with the structure of diglucosyl- tylactone. The compound was purified as described above and the structure fully confirmed by'H and 3C NMR

Table 23 :'HNMR Data for 5-0 (2'-0)- bis-glucosyl-tylactone Proton 6, multiplicity coupling 2-Ha 2. 05 d 17. 5 2-Hb 2. 48 dd 17. 5, 9. 6 3-H 3. 73 br. d 9. 8 4-H 1. 74 m 5-H 3. 78 d 9. 4 6-H 1. 11 m 7-Ha 1. 45 m 7-Hb 1.75 m 8-H 2. 65 m 10-H 6.46 d 15. 3 11-H 7. 23 d 15. 3 13-H 5. 66 d 10. 2 14-H 2. 78 ddq 10. 2, 10. 2, 6. 4 15-H 4. 68 ddd 10. 2, 9. 4, 2. 6 16-Ha 1. 58 m 16-Hb 1. 86 m 17-H3 0. 94 dd 7. 3, 7. 3 18-H3 1. 07 d 6. 8 19-Ha 1. 43 m 19-Hb 1. 63 m 20-H3 0. 88 dd 7.3, 7. 3 21-H3 1. 21 d 6. 8 22-H3 1. 85 s 23-H3 1. 07 d 6. 6 1'-H 4. 42 d 7. 7 2'-H 3.43 m 3'-H 3.52 dd 9. 0, 9. 0 4'-H 3.37 m 5'-H 3.21 m 6'-Ha 3. 70 m 6"-Hb 3. 82 dd 11. 5, 2. 6 1"-H 4.56 d 7. 7 2"-H 3.27 m 3"-H 3.37 m 4"-H 3.28 m 5"-H 3.29 m 6"-Ha 3. 71 m 6"-Hb 3.89 br. d 11. 9

Table 24 : 13C NMR Data for 5-O(2'-O)-bis-glucosyl-tylactone Carbon 8 C1 173. 4 C2 40. 0 C3 66. 9 C4 40. 9 C5 78. 7 C6 39. 6 C7 33. 5 C8 45. 2 C9 205. 7 C10 118. 8 Cll 148. 5 C12 133. 9 C13 146. 4 C14 38. 5 C15 78. 5 C16 24. 4 C17 8. 6 C18 8. 3 C19 21. 1 C20 11. 1 C21 16. 6 C22 11. 8 C23 15. 0 C1' 101. 0 C2'81. 7 C3'76. 2 C4'70. 1 C5'75. 8 C6'61. 2 Cl"104. 3 C2"74. 2 C3"76. 6 C4"70. 0 C5"77. 1 C6"61. 6

Using the approach described for the creation of gene cassettes a strategy was developed to isolate 5-O-mycaminosyl-erythromycin A and 5-O-mycaminosyl- (4"-O- mycarosyl)-erythromycin A (Fig. 32).

Isolation of a gene cassette encoding the mycaminose biosynthetic pathway A gene cassette was isolated encoding the genes responsible for the synthesis of TDP- D-mycaminose by amplifying tylMIII, tylB and tylMI from the tylosin biosynthetic gene cluster (accession numbers sf08223 and x81885) using chromosomal DNA of Streptomyces fradiae and the following primers : <BR> <BR> <BR> <BR> TylM3 1<BR> <BR> <BR> <BR> <BR> 5'GGCGGGGAGAGAGGAGAGCATATGAACACGGCAGCCGGCCCGACC<BR> <BR> <BR> <BR> <BR> <BR> <BR> <BR> <BR> <BR> Ty1M32<BR> <BR> <BR> <BR> <BR> 5'CCCCCTCTAGAGGTCACTCGGGGACATACGGGGCGACGGGCAGCCG<BR> <BR> <BR> <BR> <BR> <BR> <BR> <BR> <BR> <BR> TyIMIl<BR> <BR> <BR> <BR> <BR> 5'GGGGGTCTAGATCTTAATTAAGGAGGACAACCATGGCCCATTCATCCGCCACGGCCGG ACCGCAGGCCGA TylMI2 5'GGGGGTCTAGAGGCATATGTGTCCTCCTTAATTAATCACCGGGTTTCTCCCTTCGCTC CG GGGAGCCCGGT TylBI 5'CCCCCTCTAGATCTTAATTAAGGAGGACACCCATGACAGGGCTGCCGCGGCCCGCCGT C CGGGTG TylB2 5'GGGGGTCTAGAGGTCACGGGCCTTCCTCCCAGGAGTCCAGCGCGGCGGA The PCR fragments were cloned into SmaI cut pUC18 using standard cloning techniques as described in Materials and Methods. The sequences of the cloned fragments were verified by DNA sequence analysis. No difference to the published sequence was detected for tylMl or tylM3, but changes were detected in tylB which

resulted in the change of 8 amino acid regions compared to the published sequence of TyIB (Fig. 33).

The gene cassettes were assembled in pUC 18 using the approach described above (Fig. 34). The constructs pUC18tylMIII-tylB and pUC18tylMIII-tylB-tylMI were isolated and confirmed by restriction digests. Plasmid pUC 18tylMIIl tylB-tylMl was digested with NdeI and the insert of about 3. 5 kb was isolated and ligated into NdeI digested pSGCIII, pSGTYLM2, pSGDESVII and pSGTYLCV (Fig. 35). The correct orientation was confirmed using restriction digests.

Isolation of S. erythraea GG1 Plasmid pNC062 (Gaisser et al., 1997) was isolated from a dani Escherichia coli host strain and digested with the restriction enzymes BalI l BclI. To introduce a 0. 9 kb deletion into eryCIV as previously described (Salah-Bey et al., 1998) the ends of the DNA fragment were filled-in using standard microbiological techniques followed by a ligation step and electroporation of E. coli DH1OB. Plasmid pGG17 was isolated and confirmed by sequence analysis and restriction digest. To introduce a selectable marker into this construct, a 1. 1 kb fragment containing the thiostrepton resistance gene was isolated using plasmid pIB060 and ligated into pGG17 to generate pGGl (Fig. 36). This plasmid was used to introduce the eryCIV deletion into the-genorne of S. erythraea wild type. To isolate the S. erythraea strain GG1 techniques described previously (Gaisser et al., 1998) were used.

Isolation c f S. erythraea SGQ2 Plasmid pGGl was used to introduce a 0. 9 kb deletion in eryCIV (Salah-Bey et al., 1998) into the S. erythraea mutant strain SGT2 to create the quadruple mutant SGQ2, using the microbiological techniques described previously (Gaisser et al., 1998).

To verify the mutant, plasmid pSGCIII was used to transform SGQ2 and SGQ2pSGCIII was isolated. The cells were grown as described in Material and Methods and feeding with 3-0-mycarosyl erythronolide B was carried out as described (Gaisser et al., 2000). The supernatant of the cell culture was assessed

using techniques described in Materials and Methods and two novel peaks with 750 m/z and 713 m/z were detected (Fig. 37). Using MS/MS techniques described in Materials and Methods these novel compounds were identified as 5-O-mycaminosyl- erythromycin A and 3, 5-di-O-mycarosyl-erythronolide B.

Improved poduction of 5-0-mycaminosyl-erythronaycin A The plasmid produced by the cloning of the mycaminose gene cassette tyIMIII-tylB- tylM1 in correct orientation into the NdeI site of the plasmid pSGCIII was used to transform SGQ2 and strain SGQ2p (mycaminose) CIII was isolated. The cells were grown as described in Materials and Methods and feeding with mycarosyl- erythronolide B was carried out as described (Gaisser et al., 2000). The supernatant of the cell culture was analysed using HPLC-MS as desribed in Materials and Methods and peaks with 750 m/z and 713 m/z were detected, but the amount of the material with 750 m/z, corresponding to 5-O-mycaminosyl-erythromycin A, was significantly increased relative to the other peaks.

Construction of expression plasmidfor tylCV For expression of tylCVthe primers TylCV1 5'- GCCTGACGAAGGGTCCTGCCATATGGCTCATATTGCATT and TylCV2 5'- GCGTGGGCCGGCCGGAGATCTGGCCGCGGGGGACAGCA were used to amplify tylCV using genomic DNA of S. fradiae as template. The PCR fragment was isolated and cloned as described for the construction of expression plasmid of eryCIII.

After digestion with NdeI I BgllI a 1. 2 kb fragment was isolated, ligated into pSG142 digested with the same restriction enzymes and used transform E. coli DH10B as described above. Plasmid pSGTYLCV was isolated.

Production of 5-0-mycaminosyl-(4"-0-mycarosyl) erythromycinA The plasmid pSGTylCV was used to transform SGQ2 and strain SGQ2pSGTyICV was isolated. The cells were grown as described in Materials and Methods and a filtered supernatant from strain SGQ2p (mycaminose) CIII, containing 5-O- mycaminosyl-erythromycin A, was carried out as described for similar experiments

previously (Gaisser et al., 2000). The supernatant of the cell culture of strain SGQ2pSGTylCV was analysed using HPLC-MS as described in Materials and Methods and a novel peak with 894 m/z was detected, corresponding to 5-O- mycaminosyl- (4"-O-mycarosyl) erythromycin A.

References The references cited herein are all incorporated by reference.

Caffrey, P. et al., (1992) FEBS 304 : 225-228.

Devereux, J. et al., (1984) Nucl Acids Res 12 : 387-395.

Fouces, R. et al., (1999) Microbiol 145 : 855-868.

Gaisser, S. et al., (1997) Mol Gen Genet 256 : 239-251.

Gaisser, S. et al., (2000) Mol Microbiol 36 : 391-401.

Gandecha, A. R et al., (1997) Gene 184 : 197-203.

Haydock, S. F. et al., (1991) Mol Gen Genet 230 : 120-128..

Hernandez, C. et al., (1993) Gene 134 : 139-140.

Hessler, P. E. et al., (1997) Appl Microbiol Biotechnol 47 : 398-404.

Kaneda, T. et al., (1962) JBiol Chem 237 : 322-327.

Katz, E. et al., (1983) JGen Microbiol 129 : 2703-2714.

Pereda, A. et al., (1997) Gene 193 : 65-71.

Sambrook, J. et al., (1989) 2nd ed. Cold Spring Harbor Laboratory Press, N. Y.

Sanger, F. et al., (1977) Proc Natl Acad Sci USA 74 : 5463-5467.

Staden, R. (1984) Nucl Acids Res 12 : 521-528.

Weber, J. M. et al., (1985) JBacteriol 164 : 425-433.

Yamamoto, H. et al., (1986) JAntibiot34 : 1304-1313.

Xue, Y. et al., (1998) Proc Natl Acad Sci USA 95 : 12111-12116.

Olano et al., (1998) Mol Gen Genet 259 : 299-308.

Rodriguez et al., (1995) FEMS Microbiol. Letters 127 : 117-120.

Shah et al., (2000) J Antibiot 53 : 502-508.

Spagnoli et al., (1983) JAntibiot36 : 365-375.