Background Of The Invention

The present invention relates to novel poiyketides and methods and means for preparing them and specifically to novel erythromycins that are useful as antibacterial and antiprotozoal agents and other applications (e g anticancer, atherosclerosis gastric motility reduction etc ) in mammals including man. as well as in fish and birds This invention also relates to pharmaceutical compositions containing the novel compounds and to methods of treating bacterial and protozoal infections in mammals fish and birds by administering the novel compounds to mammals fish and birds requiring such treatment

Polyketide biosynthetic genes or portions of them, which may be derived from different polyketide biosynthetic gene clusters are manipulated to allow the production of novel erythromycins Poiyketides 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, polyether lonophores and FK506 In particular, poiyketides 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 poiyketides arises from the selection of (usually) acetate or propionate as "starter" or "extender" units; and from the differing degree of processing of the β-keto group observed after each condensation Examples of processing steps include reduction to p-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 biosynthesis of poiyketides is initiated by a group of chain-forming enzymes known as polyketide synthases Two classes of polyketide synthase (PKS) have been described

in actinomycetes However, the novel poiyketides and processes which are the subject of this invention are synthesised by Type I PKS's, represented by the PKS ^' s for the macrolides erythromycin, avermectm and rapamycin (Figure 1), and consist of a different set or "module" of enzymes for each cycle of polyketide chain extension (Figure 2A) (Cortes, J et al Nature (1990) 5 348 176-178, Donadio, S et al Science (1991) 252 675-679 MacNeii, D J et al Gene (1992), 1 15 1 19-125 Schwecke, T et al Proc Natl Acad Sci USA (1995) 92 7839-7843) Note The term "natural module" as used herein refers to the set of contiguous domains, from a fi- ketoacylsynthase ( 'KS") gene to the next acyl carrier protein ("ACP ) gene which accomplishes one cycle of polyketide chain extension The term 'combinatorial module" is used to refer to any l o group of contiguous domains (and domain parts) extending from a first point in a first natural module to a second equivalent point m a second natural module The first and second points will generally be in core domains which are present in all modules ι e both at equivalent points of respective KS AT (acyl transferase) ACP domains or in linker regions between domains

Figure 2 shows the organisation of the erythromycin producing PKS, (also known as 6-

15 deoxyerythronolide B synthase, DEBS) genes Three open reading frames encode the DEBS polypeptides The genes are organised in six repeated units designated modules The first open reading frame encodes the first multi-enzyme or cassette (DEBS1 ) which consists of three modules the loading module (erv-load) and two extension modules (modules 1 and 2) The loading module comprises an acyl transferase and an acyl carrier protein This may be contrasted 0 with Figure 1 of W093/13663 (referred to below) This shows ORF1 to consist of only two modules, the first of which is in fact both the loading module and the first extension module

In-frame deletion of the DNA encoding part of the ketoreductase domain of module 5 in DEBS has been shown to lead to the formation of erythromycin analogues 5.6-dιdeoxy-3- -mycarosyl-5oxoerythronolιde B 5 6-dιdeoxy-5-oxoerythronolιde B and 5,6-dιdeoxy-6,6 5 -epoxy-5-oxoerythronolιde B (Donadio S et al Science, (1991 ) 252 675-679) Likewise alteration of active site residues in the eπoylreductase 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-anhydroerythromyctn C (Donadio S et al Proc Natl Acad Sci USA (1993) 90 71 19-7123)

International Patent Application number WO 93/13663, which is incorporated herein by reference in its entirety, describes additional types of genetic manipulation of the DEBS genes that are capable of producing altered poiyketides However, many such attempts are reported to have been unproductive (Hutchmson C Ft and Fuμi I Annu Rev icrobiol (1995) 49 201-238 at p 231 ) The complete DNA sequence of the genes from Streptomyces hygroscopicus that encode the modular Type 1 PKS governing the biosynthesis of the macrocyc c immunosuppressant polyketide rapamycin has been disclosed (Schwecke T et al (1995) Proc Natl Acad Sci USA 92 7839-7843) (Figure 3) The DNA sequence is deposited in the EMBUGenbank Database under the accession number X86780

Although large numbers of therapeutically important poiyketides have been identified there remains a need to obtain novel poiyketides that have enhanced properties or possess completely novel bioactivity The complex poiyketides produced by modular Type I PKS ^' s are particularly valuable, in that they include compounds with known utility as anthelminthics, insecticides immunosuppressants antifungal, and/or antibacterial agents Because of their structural complexity, such novel poiyketides are not readily obtainable by total chemical synthesis or by chemical modifications of known poiyketides One aspect of the invention arises from our appreciation that a Type I PKS gene assembly encodes a loading module which is followed by extension modules It is particularly useful to provide a hybrid PKS gene assembly in which the loading module is heterologous to the extension modules and is such as to lead to a polyketide having an altered starter unit This is a concept quite unknown to the prior art since this does not recognise the existence of loading modules W093/13663 refers to altering PKS genes by inactivating a single function (i e a single enzyme) or affecting "an entire module" by deletion, insertion or replacement thereof The loading assembly in their terms, is not a module

If the loading module is one which accepts many different carboxylic acid units, then the hybrid gene assembly can be used to produce many different poiyketides For example, a hybrid

gene assembly may employ nucleic acid encoding an avr loading module with eiy extender modules A loading module may accept unnatural acid units and derivatives thereof: the ayr loading module is particularly useful in this regard (Dutton et al , (1991 ) J Antibiot , 44 357-365) In addition, it is possible to determine the specificity of the natural loading module for unnatural starter units and to take advantage of the relaxed specificity of the loading module to generate novel poiyketides Thus, another aspect of this invention is the unexpected ability of the §r loading module to incorporate unnatural carboxylic acids and derivatives thereof to produce novel erythromycins in erythromycin-producing strains containing only DEBS genes Of course one may also make alterations within a product polyketide particularly by replacing an extension module by one that gives a ketide unit at a different oxidation state and/or with a different stereochemistry It has generally been assumed that the stereochemistry of the methyl groups in the polyketide chain is determined by the acyltransferase, but it is in fact, a feature of other domains of the PKS and thus open to variation only by replacement of those domains individually or by module replacement Methyl and other substituents can be added or removed by acyltransferase domain replacement or total module replacement Consequently, it also becomes apparent to those skilled in the art that it is possible to combine the use of the relaxed substrate specificity of the erythromycin loading module with extension module replacement and hybrid loading module substitution with extension module replacement as a mechanism to produce a wide range of novel erythromycins Thus, this invention describes the production of novel erythromycins by non- transformed organisms and also such gene assemblies, vectors containing such gene assemblies, and transfσrmaπt organisms that can express them to produce novel erythromycins in transformed organisms Transformant organisms may harbour recombinant plasmids, or the plasmids may integrate A plasmid with an int sequence will integrate into a specific attachment site (aff) of a host's chromosome Transformant organisms may be capable of modifying the initial products, e g , by carrying out all or some of the biosynthetic modifications normal in the production of erythromycins (as shown in Figure 2B) However use may be made of mutant organisms such that some of the normal pathways are blocked e g to produce products without one or more natural ^"

hydroxy-groups or sugar groups, for instance as described in WO 91/16334 or in Weber et al (1985) J Bactenol 164 425-433 which are incorporated herein by reference in their entirety Alternatively, use may be made of organisms in which some of the normal pathways are overexpressed to overcome potential rate-limiting steps in the production of the desired product, for instance as described in WO 97/06266 which is incorporated herein by reference in its entirety This aspect of the method is largely concerned with treating PKS gene modules as building blocks that can be used to construct enzyme systems, and thus novel erythromycin products, of desired types This generally involves the cutting out and the assembly of modules and multi-module groupings Logical places for making and breaking intermodular connections are be in the linking regions between modules However, it may be preferable to make cuts and pins actually within domains (ι e , the enzyme-coding portions), close to the edges thereof The DNA is highly conserved here between all modular PKS's, and this may aid in the construction of hybrids that can be transcribed It may also assist in maintaining the spacing of the active sites of the encoded enzymes, which may be important For example, in producing a hybrid gene by replacing the gry loading module by an avr loading module, the ery. module together with a small amount of the following ketosynthase (KS) domain was removed The start of the KS domain (well spaced from the active site) is highly conserved and therefore provides a suitable splicing site as an alternative to the linker region between the loading domain and the start of the KS domain The excised gry module was then replaced by an ayr loading module In fact, when substituting a loading module, it may be desirable to replace not just the loading module domains (generally acyl transferase (AT) and acyl carrier protein (ACP)), but also the KS at the start of the following extension module Typically the excised loading module would have provided a propionate starter, and the replacement is intended to provide one or more different starters Propionate, however, may feed into the KS of the extension module from a propionate pool in the host cell, leading to dilution of the desired products This can be largely prevented by substituting an extended loading module including all or most of the KS domain

(The splice site may be in the end region of the KS gene, or early in the following AT gene or the linker region between them )

When replacing "modules", one is not restricted to "natural" modules For example, a "combinatorial module" to be excised and/or replaced and/or inserted may extend from the corresponding domain of two natural-type modules, e g , from the AT of one module to the AT of the next, or from KS to KS The splice sites will be in corresponding conserved marginal regions or in linker regions A combinatorial module can also be a 'double' or larger multiple, for adding 2 or more modules at a time

In a further aspect, the invention provides novel erythromycins obtainable by means of the previous aspects These include the following

(i) An erythromycin analogue (being a macrolide compound wrth a 14-membered ring) in which C-13 bears a side-chain other than ethyl, generally a straight chain C3-C6 alkyl group, a branched Cg-Cg alkyl group a Cg-Cg cycloalkyl or cycloalkenyl group (optionally substituted, e g ,

with one or more hydroxy, C, . ₄ alkyl or alkoxy groups or halogen atoms), or a 3-6 membered

heterocycle containing O or S. saturated or fully or partially unsaturated optionally substituted (as for cycloalkyl), or R, is phenyl which may be optionally substituted with at least one substitueπt selected from C,-C ₄ alkyl, C.-C ₄ alkoxy and C,-C ₄ alkylthio groups, halogen atoms, tπfluoromethyl, and cyano, or R, may be a group with a formula (a) as shown below

wherein X is O, S or -CH ₂ -, a. b, c, and d are each independently 0-2 and a + b + c + d < 5 Preferred candidates for the C-13 substituent R are the groups of carboxylate units RCOOR'. usable as substrates by an ayr starter module or rapamycin starter variants Preferred substrates are the

carboxylic acids RCOOH Alternative substrates that can be effectively used are carboxyiic acid salts, carboxylic acid esters, or amides Preferred esters are N-acetyl-cysteamine thioesters which can readily be utilised as substrates by the ayr starter module as illustrated by Dutton et al in EP 0350187 which is incorporated herein by reference in its entirety Preferred amides are N-acyl imidazoles Other alternative substrates that may be used are derivatives which are oxidative precursors for the carboxylic acids, thus for example suitable substrates would be ammo acids of the formula RCH(NH _? )COOH, glyoxylic acids of the formula RCOCOOH, methylamine derivatives of the formula RCH _? NH ₂ , methanol derivatives of the formula RCH ₂ OH, aldehydes of the formula RCHO or substituted alkanoic acids of the formula R(CH ₂ )„COOH wherein n is 2, 4 or 6 Thus examples of preferred substrates include isobutyrate (R=ι-Pr) and 2-methylbutyrate (R=1 - methylpropyl) Other possibilities include n-butyrate cyclopropyl carboxylate, cyclobutyl carboxylate, cyclopentyl carboxylate cyclohexyl carboxylate cycloheptanyl carboxylate, cyclohexenyl carboxylates, cycloheptenyl carboxylates and ring-methylated variants of the cyclic carboxylates and the aforementioned derivatives thereof The erythromycin analogue may correspond to the initial product of a PKS (6- deoxyerythronolide) or the product after one or more of the normal biosynthetic steps As shown in Figure 2b these comprise 6-hydroxylatιon. 3-0-glycosylatιon, 5-0-glycosylatιon 12- hydroxylation, and specific sugar methylation

Thus, the analogues may include those corresponding to 6-deoxyerythronolιde B, erythromycin A, and the various intermediates and alternatives (although not limited to those) shown in Figure 2b

(n) Erythromycin analogues differing from the corresponding 'natural' compound (Figure 2b) in the oxidation state of one or more of the ketide units (i e selection of alternatives from the group -CO- -CH(OH)-, =CH- and -CH ₂ -)

The stereochemistry of any -CH(OH)- is also independently selectable

(in) Erythromycin analogues differing from the corresponding natural' compound in the absence of a 'natural' methyl side-chain (This is achievable by use of a variant AT) Normal

extension modules use either C ₂ orC ₃ units to provide unmethylafed and methylated ketide

units One may provide unmethylated units where methylated units are natural (and vice versa in systems where there are naturally unmethylated units) and also provide larger units, e g , C ₄ to

provide ethyl substituents (iv) Erythromycin analogues differing from the corresponding ^' natural' compound in the stereochemistry of 'natural' methyl, and/or ring substituents other than methyl

(v) Erythromycin analogues having the features of two or more of sections (i) to (iv) (vi) Derivatives of any of the above which have undergone further processing by non-PKS enzymes, e g , one or more of hydroxylatiσn epoxidation glycosylation and methylation Methods are described for the production of the novel erythromycins of the present invention In the simplest method, unnatural starter units (preferably but not restricted to the carboxylic acid analogues of the unnatural starter units) are introduced to untransformed organisms capable of producing erythromycins A preferred approach involves introduction of the starter unit into fermentation broths of the erythromycin-producing organism an approach which is more effective for transformed organisms capable of producing erythromycins However the starter unit analogue can also be introduced to alternative preparations of the erythromycin-producing organisms, for example, fractionated or unfractionated broken-cell preparations Again, this approach is equally effective for transformed organisms capable of producing erythromycins In another method, one or more segments of DNA encoding individual modules or domains within a heterologous Type I PKS (the "donor" PKS) have been used to replace the DNA encoding, respectively, individual modules or domains within the DEBS genes of an erythromycin-producing organism Loading modules and extension modules drawn from any natural or non-natural Type I PKS, are suitable for this "donor" PKS but particularly suitable for this purpose are the components of Type I PKS's for the biosynthesis of erythromycin, rapamycin, avermectin, tetroπasin, oleandomycin, monensin, amphotencin, and rifamyciπ, for which the gene and modular organisation is known through gene sequence analysis, at least in part Particularly favourable examples of the loading modules of the donor PKS are those loading modules showing a relaxed

specificity, for example, the loading module of the avermectin (avr)-producιng PKS of Streptomyces avermitilis, or those loading modules possessing an unusual specificity, for example the loading modules of the rapamycin- FK506- and ascomycin-producing PKS's, all ot which naturally accept a shikimate-deπved starter unit Unexpectedly, both the untransformed and genetically engineered erythromycin-producing organisms when cultured under suitable conditions have been found to produce non-natural erythromycins and where appropriate, the products are found to undergo the same processing as the natural erythromycin

In a further aspect of the present invention, a plasmid containing "donor" PKS DNA is introduced into a host cell under conditions where the plasmid becomes integrated into the DEBS genes on the chromosome of the erythromycin-producing strain by homologous recombination to create a hybrid PKS A preferred embodiment is when the donor PKS DNA includes a segment encoding a loading module in such a way that this loading module becomes linked to the DEBS genes on the chromosome Such a hybrid PKS produces valuable and novel erythromycin products when cultured under suitable conditions as described herein Specifically, when the loading module of the DEBS genes is replaced by the loading module of the avermectin-producing (avr) PKS, the novel erythromycin products contain a starter unit typical of those used by the avr PKS Thus, when the loading module of the ery PKS is replaced by the avr loading module, Saccharopolyspora erythraea strains containing such hybrid PKS are found to produce 14-membered macrolides containing starter units typically used by the avr PKS It is unexpected that the 14-membered macrolide poiyketides produced by such recombinant ceiis of S erythraea are found to include derivatives of erythromycin A, showing that the several processing steps required for the transformation of the products of the hybrid PKS into novel and therapeutically valuable erythromycin A derivatives are correctly carried out A further aspect of the present invention is the unexpected and surprising finding that transcription of any of the hybrid erythromycin genes can be specifically increased when the hybrid genes are placed under the control of a promoter for a Type II PKS gene linked to a specific activator gene for that promoter It is particularly remarkable that when a genetically engineered cell containing hybrid

erythromycin genes under such control is cultured under conditions suitable for erythromycin production significantly enhanced levels of the novel erythromycin are produced Such specific increases in yield of a valuable erythromycin product are also seen for natural erythromycin PKS placed under the control of a Type II PKS promoter and activator gene In a preferred embodiment, desired genes present on an SCP2 ^* -derιved plasmid are placed under the control of the bidirectional actl promoter derived from the actinorhodin biosynthetic gene cluster of Streptomyces coelicolor, and in which the vector also contains the structural gene encoding the specific activator protein Act ll-orf 4 The recombinant plasmid is introduced into Saccharopolyspora erythraea, under conditions where either the introduced PKS genes, or PKS genes already present in the host strain, are expressed under the control of the actl promoter Such strains produce the desired erythromycin product and the activator gene requires only the presence of the specific promoter in order to enhance traπscriptional efficiency from the promoter This is particularly surprising in that activators of the Actll-orf4 family do not belong to a recognised class of DNA-binding proteins Therefore it would be expected that additional proteins or other control elements would be required for activation to occur in a heterologous host not known to produce actinorhodin or a related isochromanequinone pigment It is also surpnsing and useful that the recombinant strains can produce more than ten-fold erythromycin product than when the same PKS genes are under the control of the natural promoter, and the specific erythromycin product is also produced precociously in growing culture, rather than only during the transition from growth to stationary phase Such erythromycins are useful as antibiotics and for many other purposes in human and veterinary medicine Thus, when the genetically engineered cell is Saccharopolyspora erythraea, the activator and promoter are derived from the actinorhodin PKS gene cluster and the actl/actll-orf4-regulated ery PKS gene cluster is housed in the chromosome, following the site-specific integration of a low copy number plasmid vector, culturing of these cells under suitable conditions can produce more than ten-fold total 14-membered macrolide product than in a comparable strain not under such heterologous control When in such a genetically engineered cell of S erythraea the PKS genes under this heterologous control are

hybrid Type I PKS genes whose construction is described herein, more than ten-fold hybrid polyketide product can be obtained compared to the same hybrid Type I PKS genes not under such control Specifically when the hybrid Type I PKS genes are the ery PKS genes in which the loader module is replaced by the avr loading module, a ten-fold increase is found in the total amounts of novel 14-membered macrolides produced by the genetically engineered cells when cultured under suitable conditions as described herein

The suitable and preferred means of growing the untransformed and genetically- engineered erythromycin-producing cells and suitable and preferred means for the isolation identification and practical utility of the novel erythromycins are described more fully in the examples

Summary of the Invention

The present invention relates to compounds of the formula 1

and to pharmaceutically acceptable salts thereof wherein

R, is an alpha-branched C ₃ -C _β alkyl, alkenyl, alkynyl alkoxyalkyl or alkyft iσalkyl group any of which may be optionally substituted by one or more hydroxyl groups, a C ₅ -C ₈ cycloalkylalkyl group wherein the alkyl group is an alpha-branched C ₆ -C ₅ alkyl group a C ₃ -C _B cycloalkyl or C ₅ -C ₈ cycloalkenyl group either of which may optionally be substituted by methyl or one or more hydroxyl or one or more C.-C„ alkyl groups or halo atoms or a 3 to 6 membered oxygen or sulphur

containing heterocyclic ring which may be saturated, or fully or partially unsaturated and which may optionally be substituted by one or more C,-C ₄ alkyl groups or halo atoms, or R, is phenyl which may be optionally substituted with at least one substrtuent selected from C,-C ₄ alkyl, C,-C ₄ alkoxy and C,-C ₄ alkylthio groups, halogen atoms, tnfluoromethyl, and cyano, or R. may be a group with a formula (a) as shown below

wherein X is 0, S or -CH ₂ -, a, b, c, and d are each independently 0-2 and a + b + c + d < 5 R ₂ ιs H orOH, R ₃ -R;, are each independently H CH, or CH _? CH ₃ . R ₆ is H or OH and R ₇ is H,

CH ₃ , orCH ₂ CH ₃ , R ₈ is H or desosamine R ₉ is H CH ₃ or CH _? CH ₃ , R _m is OH mycarose (R ₁₃ is H), or cladinose (R ₁₃ is CH ₃ ), R,, is H, or R _ro = R,. = O, and R, ₂ is H CH ₃ or CH ₂ CH ₃

In the above definition, alkyl groups containing 3 or more carbon atoms may be straight or branched chain Halo means fluoro, chloro, bromo or lodo Alpha-branched means that the carbon attached to the C-13 position is a secondary carbon atom finked to two further carbon atoms, the remainder of the alkyl chain may be straight or branched chain

Preferred compounds of formula 1 are those wherein R ₃ -R ₆ , R ₇ , R ₉ , and R, ₂ are CH ₃ , and R, is isopropyl or sec-butyl, 2-buten-2-yl, 2-penten-2-yl, or 4-methyl-2-peπten-2-y! optionally substituted by one or more hydroxyl groups Also preferred are compounds of formula 1 wherein R ₃ -R ₅ , R ₇ R _B , and R, ₂ are CH ₃ , and R, is C ₃ -C _β cycloalkyl or cycloalkenyl, which may optionally be substituted by one or more hydroxyl groups or one or more C.-C, alkyl groups In a further group of preferred compounds, R, is a 5 or 6 membered oxygen or sulphur containing heterocyclic ring, particularly a 3-thιenyl or 3-furyl ring, which may be optionally substituted by one or more hydroxyl groups or one or more C,-C ₄ alkyl groups or halogen atoms In another group of preferred compounds, R, ιsaC ₃ -C ₈ alkylthιoalkyl group particularly a 1 -methylthioethyl group

Other specific embodiments of this invention include compounds of formula 2

and to pharmaceutically acceptable salts thereof wherein

R, is H C,-C _β alkyl C ₆ -C ₈ alkenyl, C ₆ -C ₆ alkynyl alkoxyalkyl or alkylthioalkyl containing from 1 to 6 carbon atoms in each alkyl or alkoxy group wherein any of said alkyl alkoxy, alkenyl or alkynyl groups may be substituted by one or more hydroxyl groups or by one or more halo atoms or a C ₃ - C ₈ cycloalkyl or C ₃ -C ₈ cycloalkenyl either of which may be optionally substituted by methyl or one or more C.-C ₄ alkyl groups or halo atoms, or a 3 to 6 membered oxygen or sulphur containing heterocyclic ring which may be saturated or fully or partially unsaturated and which may optionally be substituted by one or more C--C ₄ alkyl groups or halo atoms, or a group of the formula SR ₁₄ wherein R ₁₄ is C,-C ₈ alkyl, C ₂ ~C„ alkenyl, C ₂ -C„ alkynyl, C ₃ -C ₈ cycloalkyl, C ₅ -C„ cycloalkenyl, phenyl or substituted phenyl wherein the substituent is C,-C ₄ alkyl, C.-C ₄ alkoxy or halo, or a 3 to 6 membered oxygen or sulphur-containing heterocyclic ring which may be saturated, or fully or partially unsaturated and which may optionally be substituted by one or more C,-C ₄ alkyl groups or halo atoms

R ₂ is H or OH, R ₃ -R ₅ are each independently H, CH ₃ , or CH ₂ CH ₃ , R _β is H or OH and R, is H CH.,, or CHpCH,, R„ is H or desosamine, R ₉ is H CH ₃ , or CH _? CH ₃ , R ₁₀ is OH, mycarose (R, ₃ is H) or cladιnose (R ₁₃ ιs CH ₃ ), R., is H. or R, ₀ = R,, = O and R, ₂ ιs H CH ₃ , or CH ₂ CH ₃ , with the proviso that when R ₃ -R«,are CH ₃ , R ₇ ιs CH _n R ₉ ιs CH ₃ and R, _? ιsCH ₃ then R, is not H or C, alkyl

In the above definition, alkyl groups containing 3 or more carbon atoms may be straight or branched chain Halo means fluoro, chloro bromo or lodo

Preferred compounds of formula 2 are those wherein R ₃ -R ₅ are CH ₃ , R ₇ is CH ₃ , R ₉ is CH ₃ , and R, _? ιsCH ₃ , and R. is SR ₁₄ wherein R, ₄ is methyl or ethyl In another group of preferred compounds, R, is methyl, isopropyl, or sec-butyl, which may be substituted by one or more hydroxyl groups In a further group of preferred compounds R, is branched C ₃ -C„ alkyl group substituted by one or more hydroxyl groups or one or more halo atoms particularly 1- (trιfluoromethyl)ethyl

The invention also relates to a pharmaceutical composition for the treatment of a bacterial infection or a protozoal infection in a mammal, fish, or bird which comprises a therapeutically effective amount of a compound of formula 1 or formula 2, or a pharmaceutically acceptable salt thereof and a pharmaceutically acceptable carrier

The invention also relates to a method of treating a bacterial infection or a protozoal infection in a mammal, fish, or bird which compπses administering to said mammal, fish or bird a therapeutically effective amount of a compound of formula 1 or formula 2 or a pharmaceutically acceptable salt thereof

The term "treatment", as used herein, unless otherwise indicated, includes the treatment or prevention of a bacterial infection or protozoal infection as provided in the method of the present invention As used herein, unless otherwise indicated, the terms "bacterial ιnfectιoπ(s)" and

"protozoal ιnfectιon(s)" include bacterial infections and protozoal infections that occur in mammals, fish and birds as well as disorders related to bactenal infections and protozoal infections that may be treated or prevented by administering antibiotics such as the compounds of the present invention Such bacterial infections and protozoal infections, and disorders related to such infections, include the following pneumonia, otitis media, sinusitus bronchitis, tonsillitis, and mastoiditis related to infection by Streptococcus pneumoniae, Haemophtlus mfluβπzae, Moraxβlla catarrtialis Staphyiococcus aureus, or Peptostreptococcus spp . pharyngitis, rheumatic fever, and

glomeruloπephπtis related to infection by Streptococcus pyogenes Groups C and G streptococci, Clostπdium diptheπae or Actinobacillus haemolyticum, respiratory tract infections related to infection by Mycoplasma pneumoniae Legtonella pneumophila. Streptococcus pneumomae. Haemophilus influenzae or Chlamydia pneumoniae, uncomplicated skin and soft tissue infections, abscesses and osteomyelitis and puerperal fever related to infection by Staphyiococcus aureus coagulase-positive staphylococci (i e S epidermidis, S hemolyticus etc ) Streptococcus pyogenes Streptococcus agalactiae Streptococcal groups C-F (minute-colony streptococci) viridans streptococci Corynebacteπum mtnutissimum Clostndium spp , or Bartonella henselae uncomplicated acute urinary tract infections related to infection by Staphyiococcus saprophyticus or Enterococcus spp urethπtis and cervicitis and sexually transmitted diseases related to infection by Chlamydia trachomatis Haemophtlus ducreyi, Treponema pallidum, Ureaplasma urealyticum or Neiserna gonorrheae toxin diseases related to infection by S aureus (food poisoning and Toxic shock syndrome) or Groups A B, and C streptococci ulcers related to infection by Helicobacter pylori, systemic febrile syndromes related to infection by Borrelia recurrentis Lyme disease related to infection by Borrelia burgdodeπ, conjunctivitis keratitis, and dacrocystitis related to infection by Chlamydia trachomatis, Neissena gonorrhoeae S aureus, S pneumoniae S pyogenes H influenzae or Listena spp , disseminated Mycobacterium avium complex (MAC) disease related to infection by Mycobacterium avium or Mycobacterium mtracellulare, gastroenteritis related to infection by Campylobacter jejuni, intestinal protozoa related to infection by Cryptospondium spp , odontogenic infection related to infection by viridans streptococci, persistent cough related to infection by Bordetella pertussis, gas gangrene related to infection by Clostndium perfrmgens or Bacterotdes spp , and atherosclerosis related to infection by Helicobacter pylori or Chlamydia pneumoniae Bacteπal infections and protozoal infections and disorders related to such infections that may be treated or prevented in animals include the following bovine respiratory disease related to infection by P haem , P ultocida Mycoplasma bovis. or Bordetella spp cow enteric disease related to infection by E coli or protozoa (ι e , coccidia cryptospoπdia etc ), dairy cow mastitis related to infection by Staph aureus Strep

ubens, Strep agalacttae. Strep dysgalactiae, Klebsiella spp , Corynebacteπum, or Enterococcus spp , swine respiratory disease related to infection by A pleuro . P multocida or Mycoplasma spp , swine enteric disease related to infection by E coli Lawsonia mtracellularis, Salmonella, or Serpulma hyodyisinteπae, cow footrot related to infection by Fusobacterium spp , cow metntis related to infection by E coli, cow hairy warts related to infection by Fusobacterium necrophorum or Bacteroides nodosus. cow pink-eye related to infection by Moraxella bovis, cow premature abortion related to infection by protozoa (i e neospoπum), urinary tract infection in dogs and cats related to infection by E coli, skin and soft tissue infections in dogs and cats related to infection by Staph epidermidiε, Staph mtermedius, coagulase neg Staph or P multocida. and dental or mouth infections in dogs and cats related to infection by Alcaligenes spp Bacteroides spp Clostndium spp , Enterobacter spp , Eubactenum Peptostreptococcus. Porphyromonas. or Prevotella Other bacterial infections and protozoal infections and disorders related to such infections that may be treated or prevented in accord with the method of the present invention are referred to in J P Sanford et al , "The Sanford Guide To Antimicrobial Therapy." 26th Edition, (Antimicrobial Therapy, inc , 1996) ft is also becoming increasingly apparent that compounds of this invention can have considerable utility in the treatment of disease states (e g , cancer, AIDS, and atherosclerosis) not normally associated with bacterial or protozoal infections

When used to treat a bacterial infection or a disorder related to a bacterial infection or cancer in a mammal, such as a human, or a fish, or bird, a compound of formula 1 or formula 2 can be administered alone or in the form of a pharmaceutical composition comprising the compound and a pharmaceutically acceptable diluent or carrier Such compositions can be administered orally, for example, as tablets or capsules, or parenterally which includes subcutaneous and intramuscular injection The compounds of formula 1 or formula 2 may also be administered rectally such as through application of a suppository The pharmaceutically acceptable earner will depend on the intended mode of administration For example lactose sodium citrate, and salts of phosphoric acid together with disintegrating agents (such as starch) and lubricating agents (such as magnesium stearate, sodium laurel sulfate, and talc) can be used as the pharmaceutically

acceptable carrier in tablets Also, for use in capsules, useful pharmaceutically acceptable carriers are lactose and high molecular weight polyethylene glycols (e g , having molecular weights from 2,000 to 4,000) For parenteral use sterile solutions, or suspensions can be prepared wherein the pharmaceutically acceptable carrier is aqueous (e g , water, isotonic saline, or isotonic dextrose) or non-aqueous (e g , fatty oils of vegetable origin such as cottonseed or peanut oil, of polyols such as glycerol or propyiene glycol)

When used in vivo to treat a bacterial infection or orders related to a bacterial infection in a mammalian subject, or for treatment of various cancers in humans, (in particular non-small cell lung cancer) and other mammals such as dogs either orally or parenterally, the usual daily dosage will be in the range from 0 1-100 mg/kg of body weight especially 0 5-25 mg/kg of body weight, in single or divided doses

The phrase "pharmaceuticaly acceptable salt(s)". as used herein, unless otherwise indicded, includes salts of acidic or basic groups which may be present in the compounds of the present invention The compounds of the present inventDn that are basic in nature are capable of forming awide vanety of salts with vanous norganic aid organic acids The acids t at may be used to prepare pharmaceutically acceptable acid addition sattsof such basic compounds of are those that form non-toxic acid additDn salts i e , salts contain ng pharmacologically acceptabls anions, such as the hydrochloπde, hydrobromide, hydro iodide, nitrate, sulfate, bisulfate, phosphate, acid phosphate, tsoncotinale, acetate, lactate, salicylate, citrate, acid citrate, tartrate, pantothenate, brtartrate, ascorbate, succmate, maleate, geπtisinate.fumarate. gluconate.glucaronate, sa haate, formate, benzoate, glutamate, methanesulfonate, ethanesuffonate, benzenesulfonate, p- toluenesulfonateand pamoate [i e , 1 , l '-methytene-bιs-(2-hydroxy-3-naphthoa.e)] salts

Those compoundsof the present invention that are acidicin nature are capable of forming base salts with vanous pharmacologcally acceptable cations Examples of such salts include the alkali metal oralkaline earth metal salts and, particularly, the calcium, magnesium, sodium and potasaum saltsof the compounds of the present invention

Certari compounds of the present invention may have asymmetπc centers and therefore exist in different enantiomeπc and diastereomc forms This invention relc_tes to theuse of al optical isomers and stereoisomersof the compounds of the present invention and mixtures thereof and to all pharmaceutical compostions and methods of treatmentthat may employ orcontan them The present invention includes the compounds of the present invention, and the pharmaceutically acceptable salts thereof, wherein one or more hydrogen carbon o r other atoms are replaced by isotopes thereof Suc compoundsmay be useful as research anddiagnostictools in metabolism pharmacokinetic studies aid in binding assays

Compounds of the present nvention are produced by fermentation of an untransformed or transformed organism capable of producing erythromycins, including but not .mited to

Sacchaopolyspora species Streptomycesgπseoplanus Nocardiasp Micromonosporasp , Arthobactersp , and Streptomycesantibioticus but excluding S coelicolor Particularly suitable in this regard are untransformed and transformed strans of Saccharopolyspora erythraea forexample NRRL 2338, 18643, 21484 Particularly preferred transformed strans are those in which the erythromycin loading module has been replaced with the bajing module from the avermectin producer, Streptomyces avermitilis orthe rapamycin producer, Streptomyceshygroscopcus The preferred method of producing compounds of the current nvention B by fermentation of the appropπate organism in the presenceof the appropπate carboxylic acid of the formulaRlCOOH wheren R1 is aspreviously defined in formulae lorg., orasalt, ester (particularty preferable being the N-acetylcysteamne thioester) or amde thereof or oxidative precursorthereof The acid or derivative thereof is added to the fermentation either atthe timeof inoculation or atintervalsduπng the fermentation Production of the compounds of this invention may be monitored by removing samples from the fermentation extracting with an organc solventand following the appearance of the compounds of this invention by chromatography forexample usng high pressure liquid chromaograp y Incubation iscontnued untilthe yieldof the compound offσrmulae lor2 has been maximised generallyfor a perod of 4 to 10 days A preferred level of each addition of the carboxylic acidordeπvaϋve thereof is between 0 05 and 4 O g/L The bestyields of the compounds

from formulae 1 or 2 are generally by gradually adding the acid ordeπvative to the fermentation for example by dalyaddition over a period of several days The medum used forthe fermentation may be aconventional complexmedium contaning assimilable sources of carbon nitrogen and trace elements The suitable and preferred means of growing the untransformed and genetically- engineered erythromycin-producing cells, and suitable and preferred means for the isolation identification, and practical utility of the compounds of formulae 1 and 2 are described more fully in the Examples

Bnef Descnption of the Drawings

Some embodiments of the invention will now be described with reference to the accompanying drawings in which

Figure 1 gives the chemical formulae of three known poiyketides, Figure 2a is a diagram showing the functioning of 6-deoxyerythronolιde synthase B (DEBS), a PKS producing 6-deoxyerythronolιde B (6-DEB), a precursor of erythromycin A,

Figure 2b shows post-PKS biosynthesis of erythromycins including the conversion of 6-DEB to erythromycin A,

Figures 3a and 3b are diagrams showing the construction of plasmid pIGl , Figures 4a, 4b, and 4c are diagrams showing the construction of plasmid pND30,

Figure 5 is a diagram showing the construction of plasmid pAVLD,

Figure 6 shows the integration of pAVLD into the genome of S erythraea NRRL2338

Figure 7 is a diagram showing the biosynthesis of rapamyαn

Figure 8 is a diagram showing the construction of plasmid pM06 Figure 9 is a diagram showing the construction of plasmid pCJR26

Figure 10 is a diagram showing the construction of plasmid pC-ATX

Figure 11 is a diagram showing the construction of plasmid pC-AT12 Figure 12 is a diagram showing the construction of plasmid pCJR49

Detailed Description of the Invention

The wide range of starter units accepted by the loading module has been comprehensively established in previous studies (for example European Patent Applications 0 214 731 0 350 187, 0 317 148 which are incorporated herein in their entirety) Consequently, it should be understood that the invention is not limited to the specific detail of these examples and they simply serve to confirm the effectiveness of the ayr loading module Furthermore the examples using the pIGlor pND30 construct clearly demonstrate the capability of the actl promoter and its cognate activator gene act!l-orf4 to ehance the expression of the novel compounds of this invention when linked to the ayr loading module It is also apparent from the examples that untransformed strains of Saccharopolyspora erythraea are also readily capable of taking up exogenously-supplied substrates to generate novel erythromycin poiyketides Consequently it is also apparent to those skilled in the art that specific novel compounds of this invention can be readily produced by selection of the appropriate erythromycin producing strain (optionally incorporating the plG1 or pND30 plasmid into the desired strain), and supplementing the fermentation with the appropriate starter unit Thus, 6-deoxyerythromycιn and 6, 12- dideoxyerythromycin derivatives of the present invention can be readily produced using

Saccharopolyspora erythraea NRRL 18643 or NRRL 21484 as indicated in U S 5, 141 ,926 and WO 9706266 Similarly, use of the Saccharopolyspora erythraea strains described by Weber et al in J Bacteπol , 164 425-433, 1991 can also be employed to obtain the desired novel analogues of the present invention For example strain UW24 can be used (optionally transformed by plG1 or pND30) to obtain novel analogues of erythronolide B

UV spectra were recorded using a Hewlett-Packard 1090M diode-array spectrophotometer All NMR spectra were measured in CDCI3 by a Vaπan Unity 500 MHz

spectrometer unless otherwise indicated and peak positions are expressed in parts per million (ppm) downfield from tetramethysilane The peak shapes are denoted as follows s singlet, d, doublet t, triplet, q, quartet, m, multiplet, br, broad The atom number shown in the NMR structures is not representative of standard nomenclature but correlates NMR data to that particular example HPLC-MS data was acquired using a Hewlett-Packard 1090M liquid chromatograph interfaced to a VG Platform II mass spectrometer equipped with an APCI source (method A) or using a Hewlett-Packard 1050 liquid chromatograph intertaced to a VG Platform II mass spectrometer equipped with an APCI source (method B and method C)

HPLC method A

Column Beckman Ultrasphere 5 μm ODS 4 mm x 25 cm Flow 0 85 mL/mιn Mobile phase Gradient acetonitπle 0 05 M ammonium acetate (28 72) to acetonitπle 0 05 M ammonium acetate (50 50) over 22 minutes, maintain acetonitπle 0 05 M ammonium acetate

(50 50) 22-25 minutes return to initial conditions 25-30 minutes

HPLC method B

Column MetaChem inertsil 5 μm C83 mm x 150 mm Flow 0 5 mL/mιn Mobile phase Isocratic methanol 0 05 M ammonium acetate with 0 1 % tπfiuoroacetic acid (6040)

HPLC method C

Column Waters Symmetry 5μm C182 1 mm x 150 mm

Flow 022mϋ'mιn Mobile phase Gradient acetonitnle 005 M ammonium acetate (3070) to acetonitnle 005 M ammonium acetate (5050) over 30 minutes

Use is made of the following media and solutions Sucrose-Succmate Defined Medium sucrose 69 g

KNO 10 g

succinic acid 236g KH ₂ P0 ₄ 27g

MgS0 ₄ 7H ₂ Q 12g

ZnCI- 10 mg

MnCI ₂ 4H ₂ 0 62g

CuCI ₂ 2H ₂ 0 053 mg

CoCI 055 mg

FeS0 ₄ 7H ₂ 0 25mg

CaQ ₂ 2H ₂ 0 38 mg

milli-Q water to 10L KOH to pH 6-64

Tap water medium glucose 5g tryptone 5g yeast extract 25g EDTA 36 mg

Nutπsoy™ flour is purchased from British Arkady Group Skerton Road, Manchester, UK

The present invention is illustrated by the following examples

Example 1a - Construction of Plasmid plG1

Plasmid plG1 consists of an SCP2 ^* -deπved plasmid containing a hybrid Type I PKS gene comprising the avr loading module in place of the ery loading module, the first two extension modules of the ery PKS and the thioesterase of the ery PKS This is constructed via several intermediate plasmids as follows (Figure 3) (i) Construction of Plasmid PVE3 4 Plasmid pVE1446 which contains a portion of the avermectin (avr) PKS genes was obtained from E coli strain ATCC 68250 (MacNeii, D J et al Ann N Y Acad Sci (1994) 721 123-132) Plasmid pVE1446 was digested with BamHI and the 7 6 kbp fragment between coordinates 32 15 and 3 40 (MacNeii. D J et al Ann N Y Acad Sci (1994) 721 123-132) was purified by gel electrophoresis and recirculansed The mixture contained the desired plasmid pVE3 4 which was isolated after transformation of E coli strain TGI recO (constructed by Dr P Oliver, Dept of

Genetics, U Cambridge, Kolodner. R et al J Bactenol (1985) 163 1060-1066, T Gibson P D Thesis U Cambridge, 1985

(ιι) Construction of plasmid pNCQ12 Plasmid pBK25 (Bevitt, D J et al Eur J Biochem (1992) 20439-49) was digested with Ncol and the 12 kbp fragment was end-repaired and ligated into plasmid pUClβ which had been linearised with Smal The ligatioπ mixture was transformed into E coli TG1 recO and individual colonies were checked for their plasmid content The desired plasmid pNCOl2 was identified by its restriction pattern fm) Construction of Plasmid pCRabc Plasmid pCRabc (Figure 3) was constructed as follows Three separate PCR reactions were conducted First 20 pmol each of synthetic oligonucleotides A 1 (5'-CTCGTCGGTGGCTTT GCG-3') and A2 (5'-CCC GGG AAA AAC GAA GAC TAG TGG CGC GGA CGG CCG-3') were used to amplify a 1 0 kbp product from 100 ng pNCOI 2 template The PCR product was end-repaired, phosphoryiated and cloned into Smal-cut pUC18 to obtain plasmid pCRa Secondly, 20 pmol each of synthetic oligonucleotides C1 (5'-CACGCGCAGCGCGGCGGA-3') and C2 (5'-CGAA CCG CTA GCGGTCGTCGCG ATGGCCT-3') were used to amplify a 1 5 kbp product from 100 ng pNC012 template The product was end-repaired, phosphoryiated and cloned into Smal-cut pUC18 to obtain plasmid pCRc Thirdly, 20 pmol each of synthetic oligonucleotides B1 (5'-GTGGCX CX3GCC_X3TCCG∞CCACTAGTCTTCGTTTTT-3 ^* ) and B2 (5'-AAC AGCTAGCGGTTCGTCCGCCGCTGCCGTGCC-3') were used to amplify a 1 4 kbp product from 100 ng pVE3 4 template The product was end-repaired, phosphoryiated and cloned into Smal-cut PUC18 to obtain plasmid pCRb

Plasmid pCRa was digested with Hindlll and Spel and the 1 0 kbp insert was ligated with plasmid pCRb previously digested with Hindlll and Spel, to obtain plasmid pCRab Plasmid pCRc was digested with Nhel and EcoR1 and the 1 5 kbp insert was ligated with plasmid pCRab previously digested with Nhel and EcoR1 to obtain plasmid pCRabc

( v) Construction of Plasmid pNEWAVETE Plasmid pCRabc was digested with Mfel and Sfil and the DNA fragment containing the loading domain of the avr PKS was puπfied by gel electrophoresis and ligated with plasmid pNTEP2 which had been digested with Mfel and Sfil and the larger fragment purified by gel electrophoresis The ligation mixture was transformed into E coli TG1 recO and individual colonies were checked for their plasmid content The desired plasmid pNEWAVETE (13 7 kbp) was identified by its restriction pattern

(v) Construction of Plasmid pRM52 Plasmid pRM52 is a derivative of plasmid pRM5 (McDanιel R et al Science, (1993) 262 1546-1550) pRM5 was first linearised by digestion with Ndel, end-repaired and then religated to produce pRM51 pRM51 was cut with Pad and Nsil and the large Pacl-Nsil fragment was isolated and ligated to a short double-stranded oligonucleotide linker containing an Ndel site and constructed from the synthetic oligonucleotides 5'-TAAGGAGGACACATATGCA-3' and 5'-TAATTCCTCCTGTGTAT-3' which were annealed together The ligation mixture was transformed into E coli TGlrecO and isolated colonies were screened for their plasmid content The desired plasmid (19 6 kbp) was identified by its restriction map and was designated pRM52

(vi) Construction of Plasmid plG1 Plasmid pNEWAVETE was digested with Ndel and Xbal and the insert was purified by sedimentation on a sucrose gradient The purified insert was ligated into plasmid pRM52 (19 6 kbp) which had been digested with Ndel and Xbal, and the vector purified by sedimentation on a sucrose gradient The ligation mixture was used to transform E coli and individual colonies were checked for their plasmid content. The desired plasmid plG1 was identified by its restriction pattern Example 1b - Construction of Plasmid PND30 Plasmid pND30 consists of an SCP2 ^* -deπved plasmid containing a hybrid Type I PKS gene comprising the avr loading module in place of the ery loading module the first two extension

modules of the ery PKS and the thioesterase of the ery PKS This is constructed via several intermediate plasmids as follows (Figure 4) d) Construction of the Recombinant Vector pCJR101 pCJR101 (Figure 4) is a shuttle plasmid constructed to be used for expression of PKS genes in actmomycetes It includes a ColEi replicon to allow it to replicate in E coli, an SCP2 ^* low copy number Streptomyces replicon (Bibb, M J and Hopwood, D A J Gen Microbiol (1981) 126427) and the actlf-orf4 activator gene from the act cluster which activates transcription from the act promoter during the transition from growth phase to stationary phase in the vegetative mycelium It is constructed as follows an approximately 970 bp DNA fragment from pMF1015 (containing the actll orf4 activator gene) (Fernandez-Moreno M A et al Cell (1991) 66 769-780) is amplified by PCR using as primers the synthetic oligonucleotides 5 ^' -ACT AGT CCA CTG CCT CTC GGT AAA ATC CAG C-3 and 5'-CTT AAG AGG GGC TCC ACC GCG TTC ACG GAC-3'. which also introduces flanking Spel and Aflll restriction sites This fragment is cloned into the end- repaired Aatll site of plasmid pUCl9 to yield plasmid pC R18 An approximately 215 bp DNA fragment is amplified from pMV400 which contains the bi-directional promoter pair Pactlll/Pactl) (Parro, V et al Nucl Acids Res (1991) 19 2623-2627), using as primers the synthetic oligonucleotides 5'-ACA TTC TCT ACG CCT AAG TGT TCC CCT CCC TGC CTC-3' and 5 -GTG ATG TAT GCT CAT ATG TGT CCT CCT TAA TTA ATC GAT GCG TTC GTC CGG TG-3', which also introduces flanking Ndel and Aflll sites The PCR product is digested with Ndel and Aflll and ligated with the plasmid pCJR18 previously cut with Ndel and Aflll, to generate plasmid pCJR19 A 1 1 kbp Hindlll SphI fragment containing the tsr gene, which confers resistance to thiostrepton, is obtained by PCR from plasmid plJ922 (Lydiate, D J et al Gene (1985) 35 223-235) as template, using as pπmers the oligonucleotides 5'-TGA ACA CCA AGC TTG CCA GAG AGC GAC GAC TTC CCC-3' and 5 ^' -GAC AGA TTG CAT GCC CTT CGA GGA GTG CCC GCC CGG-3" which also introduces flanking Hindlll and SphI sties The PCR product is digested with Hindlll and SphI and ligated with plasmid pCJR19 cut with Hindlll and SphI to obtain plasmid pCJR24 The plasmid plJ922 is digested with BamHI and Sstl and the fragment containing a portion of the fertility locus and the

origin of replication (Lydiate, D J et al Gene (1985) 35 223-235) is ligated into pUC19 digested with BamHI and Sstl to generate the bifunctional plasmid pCJR16 (147 kbp) Plasmid pCJR24 is digested with Sail and SphI, the two larger fragments from the digest are purified by gel electrophoresis, and combined in a four-component ligation with plasmid pCJRl6 which has been digested with Xhol and SphI The ligation mixture is used to transform Streptomyces hvidans and colonies are selected in the presence of thiostrepton One such colony is shown to contain the desired plasmid pCJR101 (approx 12 4 kbp), identified by its restriction pattern

(n) Construction of Plasmid pCJR29 The construction of plasmid pCJR29 is illustrated in Figure 4 A 1 1 kbp Hindlll-Xhol fragment containing the tsr gene which confers resistance to thiostrepton. is obtained by PCR from plasmid plJ922 as template, using as primers the oligonucleotides 5'-TGA ACA CCA AGC TTG CCA GAG AGC GAC GAC TTC CCC-3 and 5 -GAC AG A TTC TCG AGC CTT CGA GGA GTG CCC GCC CGG-3 ^' which also introduces flanking Hindlll and Xhol sites The PCR product is digested with Hindlll and Xhol and ligated with plasmid pCJRl 6 which has been digested with Hindlll and Xhol, to generate plasmid pCJR25 Plasmid pCJR25 is digested with Hindlll and SphI and ligated with plasmid pCJR19 which has been digested with Hindlll and SphI, to produce the desired plasmid pCJR29 (approx 12 4 kbp), identified by its restriction pattern Plasmid pCJR29 differs from pCJR101 in the orientation of the tsr gene the actll-orf gene and the actl/actlll promoter, with respect to the SCP2 ^* -derιved origin of replication (in) Construction of Plasmid pND30

Plasmid pNEWAVETE was digested with Ndel and Xbal and the insert was puπfied by sedimentation on a sucrose gradient The punfied insert was ligated into plasmid pCJR29 (approx 12 4 kbp) which had been digested with Ndel and Xbal, and the vector purified by sedimentation on a sucrose gradient The ligation mixture was used to transform E coli and individual colonies were checked for their plasmid content The desired plasmid pND30 was identified by its restriction pattern

Example 1c - Construction of plasmid pCJR26

Plasmid pMOβ (Figure 8) was first constructed in several steps (i ) Construction of plasmid pMOl The approximately 1 3 kbp DNA segment of the eryAI gene of S erythraea extending from nucleotide 1948 to nucleotide 3273 of eryAI (Donadio S et al Science (1991) 252 675-679) was amplified by PCR employing as primers synthetic oligonucleotides 5' -CAT GCT CGA GCT CTC CTG GGA AGT-3' and 5'-CAA CCC TGG CCA GGG AAG ACG AAG ACG G-3' and plasmid pNTEP2 as a template The PCR product was end-repaired and ligated with plasmid pUC18 which had been linearised by digestion with Smal and then treated with alkaline phosphatase The ligation mixture was used to transform E coli TG1 recO and individual colonies were checked for their plasmid content The desired plasmid pMOI (3 9 kbp) in which the Stul site bordering the insert is adjacent to the Hindlll site in the polylinker was identified by tts restriction pattern

(II) Construction of plasmid pMQ2 The approximately 085 kbp DNA segment of the rapA gene of Streptomyces hygroscopicus extending from nucleotide 1643 to nucleotide 2486 of rapA was amplified by PCR employing as primers the following oligonucleotides 5'-TTC CCT GGC CAG GGG TCG CAG CGT G-3' and 5'-CAC CTA GGA CCG CGG ACC ACT CGA C-3', and the DNA from the recombinant bacteπophage λ-1 E (Schwecke T et al , Proc Natl Acad Sci USA (1995) 92 7839-7843) as the template PCR product was end-repaired and ligated with plasmid pUC18, which had been linearised by digestion with Smal and then treated with alkaline phosphatase The ligation mixture was used to transform E coli TGI recO and individual colonies were checked for their plasmid content The desired plasmid pM02 (3 5 kbp) was identified by its restriction pattern

(in) Construction of plasmid pMQ3 The approximately 1 7 kbp DNA segment of the eryAI gene o1 S erythraea extending from nucleotide 4128 to nucleotide 5928 of eryAI, was amplified by PCR employing as primers the synthetic oligonucleotides 5'-TGG CCA GGG AGT CGG TGC ACC TAG GCA-3' and 5'-GCC GAC AGC GAG TCG ACG CCG AGT T-3', and plasmid pNTEP2 as template The PCR product was end-

repaired and ligated with plasmid pUCl 8, which had been linearised by digestion with Smal and then treated with alkaline phosphatase The ligation mixture was used to transform E coli TG1 recO and individual colonies were checked for their plasmid content The desired plasmid pM03 (44 kbp), in which Ball and Avril sites are adjacent to the Hindlll site of the polylinker, was identified by its restriction pattern

(iv) Construction of plasmid pMQ4 Plasmid pMOl was digested with Hindlll and Ball and the 1 3 kbp insert was ligated with plasmid pM03 which had been digested with Hindlll and Ball The ligation mixture was used to transform E coli TG1 recO and individual colonies were checked for their plasmid content The desired plasmid pM04 (5 6 kbp) was identified by its restriction pattern (v) Construction of plasmid pMQ5 Plasmid pM04 was digested with Stul and the 3 0 kbp insert was ligated with plasmid pNTEP2 which had been digested with Stul and purified by gel electrophoresis to remove the 3 8 kbp insert The ligation mixture was transformed into E coli TG1 recO and individual colonies were checked for their plasmid content The plasmid pM05 (12 8 kbp) was identified by its restriction pattern

(vi) Construction of plasmid pMQ6 Plasmid pM02 was digested with Ball and Avrll and the insert was ligated with plasmid pMOδ which had been digested with Ball and Avrll The ligation mixture was used to transform E coli TG1 recO and individual colonies were checked for their plasmid content The desired plasmid pMO ^β (13 5 kbp) was identified by its restriction pattern

(vn ) Construction of plasmid pCJR26 Plasmid pCJR26 is an SCP2* based plasmid containing a PKS gene comprising the ery loading module, the first and second extension modules of the ery PKS and the ery chain-terminating thioesterase, except that the DNA segment encoding the methylmalonyl-CoA ACP acyltransferase within the first extension module has been specifically substituted by the DNA encodinα the malonyl-CoA ACP acyltransferase of module 2 of the rap PKS It was constructed as

follows (Figure 9) plasmid pM06 was digested with Ndel and Xbal and the insert was ligated with plasmid pCJR24 which had been digested with Ndel and Xbal and purified by gel electrophoresis The ligation mixture was transformed into E coli TG 1 recO and individual colonies were checked for their plasmid content The desired plasmid pCJR26 was identified by its restriction pattern

Example 1d - Construction of S erythraea JC2/pCJR26 and production of TKL derivatives Plasmid pC R26 was used to transform S erythraea JC2 protoplasts Thiostrepton resistant colonies were selected on R2T20 medium containing 10 μg/ml of thiostrepton Several clones were tested for the presence of pCJR26 integrated into the chromosome by Southern blot hybridisation of their genomic DNA with DIG-labelled DEBSl-TE gene

A clone with an integrated copy of pCJR26 was grown in SSM medium containing 5 μg/ml of thiostrepton and allowed to grow for seven days at 28-30°C After this time the broth was filtered to remove mycelia and the pH was adjusted to pH 3 The broth was extracted twice with two volumes of ethyl acetate and the combined ethyl acetate extracts were washed with an equal volume of saturated sodium chloride, dried over anhydrous sodium sulfate, and the ethyl acetate was removed under reduced pressure to give about 500 mg of crude product The products were shown to be (2S, 3R, 5R)-2-methyl-3.5-dιhydroxy-π-hexanoιc acid ft-lactone and (2S 3R 5R)-2- methyl-3.5-dιhydroxy-n-heptanoιc acid b -lactone

Example 1e - Construction of S erythraea NRRL 2338/pCJR26 and its use in production of 14- membered macrolides

Approximately 5 mg pCJR49 DNA was used to transform S erythraea NRRL2338 protoplasts to give a strain in which the plasmid is integrated into the chromosome From several colonies total DNA was obtained and analysed by Southern hybridisation to confirm that the plasmid has integrated in module 2 of EryAI to give a novel macrolide biosyntheticpathway Further integrations had occurred to give repeated plasmid sequences S erythraea NRRL 2338 /pCJR49 was inoculated into tryptic soy broth containing 5mg/ml thiostrepton and incubated at 30°C for three days 100 mL of this seed culture was used to inoculate 2 L of sucrose succinate defined medium containing 5mg/ mL thiostrepton in 5 x 2 L flasks each containing 500mL medium with 2 springs to aid dispersion and shaken at 300 rpm After a further 5 days of growth the cultures were centrifuged and the pH of the supernatant adjusted to pH 9 The supernatant was then extracted three times with an equal volume of ethyl acetate and the solvent removed by evaporation Products were analysed by HPLC/MS and two macr olides were identified as the erythromycin analogues

Example 1f - Construction of plasmid pC-ATX Plasmid pC-ATX is an SCP2' based plasmid containing a PKS gene comprising the ery loading module the first and second extension modules of the ery PKS and the erv chain-

terminating tnioesterase, except that the DNA segment encoding the methylmafonyl-CoA ACP acyltransferase within the first extension module has been specifically substituted by the DNA encoding the malonyl-CoA ACP acyltransferase from a putative type I PKS gene cluster cloned from Streptomyces cmnamoπensis ATCC 14513 (producer of the polyether polyketide monensin) It was constructed via several intermediate plasmids as follows (Figure 10) (i) Isolation of cosmid pSCIN02 Genomic library of Streptomyces cinnamonensis ATCC 14513 (the monensin producer) was constructed from size fractioπed 35 - 45 kbp Sau3A fragments of chromosomal DNA ligated into BamHl-lineaπsed and alkaline phosphatase- treated cosmid vector pWEl 5 The ligation mixture was packaged into Λ-partιcles using Gigapack packaging extracts and transfected into E coli NMIblue Approximately 600 colonies of the library were grown on the surface of a nylon membrane lysed and their DNA was crosslinked to the membrane by UV irradiation The membrane was subsequently used for the screening procedure The insert of pM08 comprising the ketosynthase domain from module 2 of DEBS was labelled by random priming in the presence of ^P < ATP and used as a probe for DNA hybridisation The probe was hybridised for I6h at 68°C in 4 OxSSC buffer and subsequently washed off for ih at 68°C in 0 8xSSC buffer Three positive clones were isolated DNA of the inserts of all three clones was end sequenced from T3 and T7 priming sites present in the vector pWE15 A region homologous to type I ketosynthase and malonyl-CoA ACP acyltransferase domains was discovered in the DNA sequence from the T7 priming site using clone 2 (named pSCIN02) as a template Partial DNA sequencing of the malonyl- CoA ACP acyltransferase domain (named ATX) revealed an unusual sequence motif in the putative substrate recognition part of the domain which was substantially different from previously described malonate- or methylmalonate-spec fic CoA ACP acyltransf erases (Haydock. S F et al FEBS (1995) 374 246-248) (ii) Construction of plasmid pM038

The approximately 0 9 kbp DNA segment of the ATX domain was amplified by PCR employing as primers the following oligonucleotides 5' CTG GCCAGG GCGCGCAATGGCCGAGCAT-3'and

5'-CCC TAG GAG TCG CCG GCA GTC CAG CGC GGC GCC C-3' using the DNA from the cosmid pSCIN02 as the template The PCR product was end-repaired and ligated with plasmid pUCl8 which had been linearised by digestion with Smal and then treated with alkaline phosphatase The ligation mixture was used to transform E coli TGI recO and individual colonies were checked for their plasmid content The desired plasmid pM038 (3 5 kbp) was identified by its restriction pattern

(in) Construction of plasmid PM034 Plasmid pM034 is a derivative of pM06 with a polycloning site inserted after the stop codon of the inserted D1 -AT2 gene Plasmid pM06 was digested with EcoRI and Hindlll and annealed with two oligonucleotides forming the double-stranded region of the polycloning site 5'-AAT TCA TAA CTA GTA GGA GGT CTG GCC ATC TAG A-3" and 5 -TCG AAG ATC TAC CGG TCT GGA GGA TG A TCA ATA C-3' The mixture was ligated and transformed into E coli TGl recO Individual colonies were checked for their plasmid content The desired plasmid pM034 (13 5 kbp) was identified by its restriction pattern (iv) Construction of plasmid pMQ35

Plasmid pM035 is a derivative of pM034 containing TKLS-AT2 gene and a translationally coupled crotonyl-CoA-reductase gene from Streptomyces collinus (Wallace et al , E J Biochem (1995) 233 954-962) The crotonyl-CoA-reductase gene was excised from the plasmid pZYB3 (the gift of Prof K Reynolds) as an Ndel - BamHI fragment, which was treated with mung bean nuclease to produce blunt ends and ligated into pM034 previously cut with Spel and likewise blunt-ended using mung bean nuclease The ligation mixture was used to transform E coli TG1 recO and individual colonies were checked for their plasmid content The desired plasmid pM035 (14 2 kbp), with the correct orientation of the crotonyl-CoA-ketoreductase gene was identified by its restriction pattern (v) Construction of plasmid pMQ36

Plasmid pM038 was digested with Ball and Avrll and the insert was ligated with plasmid pM035 which had been diαested with Ball and Avrll The liαation mixture was used to transform F roll TG1

recO and individual colonies were checked fon their plasmid content The desired plasmid pM036 (13 5 kbp) was identified by its restriction pattern

(vi) Construction of plasmid pC-ATX Plasmid pM036 was digested with Ndel and Xbal and the insert was ligated with plasmid pCJR29, which had been digested with Ndel and Xbal and purified by gel electrophoresis The ligation mixture was transformed into E coli TG1 recO and individual colonies were checked for their plasmid content The desired plasmid pC-ATX was identified by its restriction pattern

Example iq • Construction of S erythraea JC2/pC-ATX and production of TKL derivatives Plasmid pC-ATX was used to transform S erythraea JC2 protoplasts Thiostrepton resistant colonies were selected on R2T20 medium containing 10 μg/ml of thiostrepton Several clones were tested for presence of pC-ATX integrated into the chromosome by Southern blot hybridisation of their genomic DNA with DIG-labelled DNA encoding the DEBS1-TE gene

A clone with an integrated copy of pC-ATX was grown in SSM medium, containing 5μg/ml of thiostrepton, and allowed to grow for seven days at 28-30°C After this time the broth was filtered to remove myce a and the pH adjusted to pH 3 The broth was extracted twice with two volumes of ethyl acetate and the combined ethyl acetate extracts were washed with an equal volume of saturated sodium chloride, dried over anhydrous sodium sulfate, and the ethyl acetate was removed under reduced pressure, to give about 500 mg of crude product The products were characterised by gas chromatography, mass spectrometry and NMR, and were shown to be (2S, 3R, 4S, 5R)-2-methyl-4-ethyl-3,5-dιhydroxy-n-hexanoιc acid ό-lactone and (2S, 3R, 4S, 5R)-2- methyl-4-ethyl-3,5-dιhydroxy-n-heptanoιc acid ft-lactone

Example ih - Construction of S erythraea NRRL 2338/pC-ATX and its use in production of 14- membered macrolides

Approximately 5mg pC-ATX DNA was used to transform S erythraea NRRL 2338 protoplasts to give a strain in which the plasmid is integrated into the chromosome From several colonies total DNA was obtained and analysed by Southern hybridisation to confirm that the plasmid has integrated in module 2 of EryAI to give a novel macrolide biosyntnetic pathway

Further integrations had occurred to give repeated plasmid sequences S e ^r ythraea NRRL 2338

/pC-ATX was inoculated into tryptic soy broth containing 5 mg/mL thiostrepton and incubated at 30°C for three days 100 mL of this seed culture was used to inoculate 2 L of sucrose succinate defined medium containing 5 mg/mL thiostrepton in 5 x 2 L flasks each containing 500 mL medium with 2 springs to aid dispersion and shaken at 300 rpm After a further 5 days of growth the cultures were centrifuged and the pH of the supernatant adjusted to pH 9 The supernatant was then extracted three times with an equal volume of ethyl acetate and the solvent removed by evaporation Products were analysed by HPLC-MS and two macrolide products were identified

EΞxample 1ι - Construction of plasmid pC-AT12

Plasmid pC-ATl2 is an SCP2 ^* based plasmid containing a PKS gene comprising the ery loading module, the first and second extension modules of the ery PKS and the ery chain- terminating thioesterase, except that the DNA segment encoding the rnethylmalonyl-CoA ACP acyltransferase within the second extension module has been specifically substituted by the DNA encoding the malonyl-CoA ACP acyltransferase of module 2 of the rap PKS It was constructed via several intermediate plasmids as follows (Figure 11 ) (i ) Construction of plasmid pMQ25 The approximately 1 0 kbp DNA segment of the eryAI gene of S erythraea extending from nucleotide 6696 to nucleotide 7707 of eryAI (Donadio S et al Science (1991 ) 252, 675-679) was amplified by PCR employing as primers synthetic oligonucleotides 5'-GGCGGGTCCGGA GGTGTTCACCGAGTT-3' and 5' -ACC TTG GCC AGG GAA GAC GAA CAC TGA-3' and plasmid pNTEp2 as a template The PCR product was end-repaired and ligated with plasmid pUC18, which had been linearised by digestion with Smal and then treated with alkaline phosphatase The ligation mixture was used to transform E coli TGl ecO and individual colonies were checked for their plasmid content The desired plasmid pM025 (3 6 kbp) in which the Stul site bordering the insert is adjacent to the Hindlll site in the polylmker, was identified by its restriction pattern (n) Construction of plasmid pM026 The approximately 0 6 kbp DNA segment of the eryAI gene of S erythraea extending from nucleotide 8660 to nucleotide 9258 of eryAI, was amplified by PCR employing as primers the synthetic oligonucleotides 5'-TCCTAGGCCGGGCCGGACTGGTCGACCTGCCGGGTT-3' and 5 ^" -AAA CAC CGC GAC CTG GTC CTC CGA GC-3', and plasmid pNTEP2 as template The PCR product was end-repaired and ligated with plasmid pUClδ which had been linearised by digestion with Smal and then treated with alkaline phosphatase The ligation mixture was used to transform E coli TG 1 recO and individual colonies were checked for their plasmid content The desired

plasmid pM026 (3 2 kbp), in which the Avrll site is adjacent to the Hindlll site of the polyhnker, was identified by its restriction pattern

(in) Construction of plasmid pMQ27 Plasmid pM025 was digested with EcoRl and Ball and the 1 0 kbp insert was ligated with plasmid pM02 which had been digested with EcoRl and Ball The ligation mixture was used to transform E coli TG1 recO and individual colonies were checked for their plasmid content The desired plasmid pM027 (4 4 kbp) was identified by its restriction pattern

(iv) Construction of plasmid pMQ32 Piasmid pM026 was digested with Avrll and Hindlll and the 0 6 kbp insert was ligated with plasmid pM027 which had been digested with Avrll and Hindlll The ligation mixture was used to transform E coli TG1 recO and individual colonies were checked for their plasmid content The desired plasmid pM032 (5 1 kbp) was identified by its restriction pattern fv) Construction of plasmid pMQ33 Plasmid pM032 was digested with BspEI and SexAI and the 2 7 kbp insert was ligated with plasmid pNTEP2 which had been digested with the same two enzymes and purified by gel electrophoresis to remove the 2 8 kbp insert The ligation mixture was transformed into E coli TG1 recO and individual colonies were checked for their plasmid content The piasmid pM033 (12 8 kbp) was identified by its restriction pattern

(vfl Construction of p gmid PC-AT12 Plasmid pM033 was digested with Ndel and Xbal and the insert was ligated with plasmid pCJR29, which had been digested with Ndel and Xbal and purified by gel electrophoresis The ligation mixture was transformed into E coli TGl recO and individual colonies were checked for their plasmid content The desired plasmid pC-ATl 2 was identified by its restriction pattern

Example 1 J - Construction of S erythraea JC2/pC-AT 12 and production of TKL derivatives Plasmid pC-AT12 was used to transform S erythraea JC2 protoplasts Thiostrepton resistant colonies were selected on R2T20 medium containing 10 μg/ml of thiostrepton Several

A clone with an integrated copy of pC-ATl2 was grown in SSM medium, containing 5 μg/mL of thiostrepton and allowed to grow for seven days at 28-30°C After this time the broth was filtered to remove mycelia and the pH adjusted to pH 3 The broth was extracted twice with two volumes of ethyl acetate and the combined ethyl acetate extracts were washed with an equal volume of saturated sodium chloride, dried over anhydrous sodium sulfate, and the ethyl acetate was removed under reduced pressure, to give about 500 mg of crude product The products were shown to be (3R. 4S, 5R)-4-methy!-3.5-dihydroxy-n-hexanoic acid ft-lactone and (3R. 4S. 5R)-4- methyl-3 5-dιhydroxy-n-heptanoιc acid ft -lactone

Example Ik - Construction of S erythraea NRRL 2338/pC-AT12 and its use in production of 14- membered macrolides

Approximately 5μg pC-AT12 DNA was used to transform S. erythraea NRRL 2338 protoplasts to give a strain in which the plasmid is integrated into the chromosome. From several colonies, total DNA was obtained and analysed by Southern hybridisation to confirm that the plasmid has integrated 3' of module 2 of EryAI to give a novel macrolide biosynthetic pathway.

Further integrations had occurred to give repeated plasmid sequences. S erythraea NRRL 2338 /pC-ATl2 was inoculated into tryptic soy broth containing 5 mg/mL thiostrepton and incubated at 30°C for three days 100 mL of this seed culture was used to inoculate 2 L of sucrose succinate defined medium containing 5μg/ mL thiostrepton in 5 x 2 L flasks each containing 500 mL medium with 2 springs to aid dispersion and shaken at 300 rpm After a further 5 days of growth

the cultures were centrifuged and the pH of the supernatant adjusted to pH 9 The supernatant was then extracted three times with an equal volume of ethyl acetate and the solvent removed by evaporation Products were analysed by HPLC-MS and two macrolide products were identified

Example 11 - Construction of plasmid pCJR49 pCJR49 is a pCJR24-based plasmid containing a mutant DEBS1-TE gene which has no ketoreductase in module 2, and the AT domain in module 2 has been replaced py RAPS AT2 in order to incorporate a malonyl extender instead of a methylmalonyl extender in the second module

(Figure 12) pM032 was digested with BspEI and SexAI and the fragment containing the AT from RAP module 2 was cloned into pUC1-0 which had been previously digested with BspE I and SexA I, to yield the plasmid pCJR43 pCJR43 was digested with Ndel and Xbal and the fragment containing the mutant DEBS1 -

TE gene was cloned into pCJR24 which had previously been digested with Ndel and Xbal, to yield plasmid pCJR49 pCJR49 was confirmed by restriction enzyme mapping

Example 1m - Construction of S erythraea JC2/PCJR49 and production of TKL derivatives

Approximately 5μg pCJR49 DNA was used to transform S erythraea JC2 protoplasts to give a strain in which the plasmid is integrated into the chromosome From several colonies total DNA is obtained and analysed by Southern hybridisation to confirm that the plasmid has integrated into the eryTE S erythraea JC2/pCJR49 is inoculated into tryptic soy broth containing 5μg / _m |_ thiostrepton and incubated at 30°C for three days . OO mL of this seed culture was used to inoculate 2 L of sucrose succinate defined medium containing 5μg /mL thiostrepton in 5 x 2 L flasks each containing 500 mL medium with 2 springs to aid dispersion and shaken at 300 rpm After a further 5 days of growth the cultures were centrifuged and the pH of the supernatant was adjusted to pH 3 The supernatant was then extracted three times with an equal volume of ethyl acetate and the solvent removed by evaporation Products were dissolved in methanol and analysed by GCMS on a Finnegan-MAT GCQ System This analysis indicated that by comparison to synthetic standards two new lactones were present These products were (4S 5R)-4-methyl-3- keto-5-hydroxyhexanoιc acid ft lactone and (4S.5R)-4-methyl-3-kelo-5-hydroxyheptanoιc acid ft lactone

Example 1n - Construction of S erythraea NRRL 2338/pCJR49 and its use for production of 14- membered macrolides

5μg pCJR49 DNA was used to transform S erythraea NRRL 2338 protoplasts to give a strain in which the plasmid is integrated into the chromosome From several colonies total DNA is obtained and analysed by Southern hybridisation to confirm that the plasmid has integrated in module 2 of EryAI to give a novel macrolide biosynthetic pathway Further integrations had

occurred to give repeated plasmid sequences S erythraea /pCJR49 is inoculated into tryptic soy broth containing 5μσ /mL thiostrepton and incubated at 30°C for three days 100 mL of this seed culture was used to inoculate 2 L of sucrose succinate defined medium containing 5μg/mL thiostrepton in 5 x 2 L flasks each containing 500mL medium with 2 springs to aid dispersion and shaken at 300 rpm After a further 5 days of growth the cultures were centrifuged and the pH of the supernatant adjusted to pH 9 The supernatant was then extracted three times with and equal volume of ethyl acetate and the solvent removed by evaporation Products were analysed by HPLC-MS and two macrolides were identified

Example 2 - Construction of S erythraea ERMD1. Carrying a Hybrid PKS Gene in which the ayr Loading Didomain is Substituted for the ery Loading Didomain of S erythraea NRRL 2338 (ι) Construction of plasmid pAVLD Plasmid pCRabc (Example 1) was linearised with BamHI and ligated to plJ702 previously digested with Bglll The mixture contained the desired plasmid pAVLD (Figure 5) The ligation mixture was transformed into E coli TG1 recO and individual colonies were checked for their plasmid content The desired plasmid pAVLD was identified by its restriction pattern (Figure 5) (n) Construction of S ervthrea ERMD1

Approximately 5-10 μg of pAVLD, isolated from E coli TGIrecO(pAVLD) was transformed into S erythraea NRRL2338 and stable thiostrepton resistant colonies were isolated One of these colonies was selected and total DNA was digested with Pstl and analysed by Southern hybridisation employing as a probe the insert from plasmid pCRc which contains the fragment of the ery Al gene encoding the ketosynthase domain KS1 The analysis showed positively-hybridizing Pstl fragments of 8 5 kbp, 4 8 kbp and 33 kbp, indicating the presence of two taπdemly integrated copies of pAVLD (Figure 6)

Example 3 - Preparation of Isopropyl and sec-Butyl Erythromycins Using S erythraea ERMD1 A 50 mL fermentation of S erythraea ERMD1 was carried out on tap water medium and after 4 days at 30°C the mycelium was harvested and used to inoculate 1 5 L of sucrose-succmate medium containing thiostrepton (50μg/mL) After growth at 30°C for 4 days, the whole broth was extracted twice with an equal volume of ethyl acetate The combined extracts were concentrated under reduced pressure and subjected twice to preparative thin layer chromatography on silica plates (20 x 20cm) eluted with chloroform/methanol/ 88 ammonia 8 2001 (by vol) The products were further separated by HPLC on a PhaseSep C18 base-deactivated reversed-phase column S5 ODS (octadecylsilane) 6 (4 6mm x 25 cm), eluted with methanol/0 5% ammonium acetate (70.30 (vol vol)), at 1 mlJmin Fractions were collected between 7 and 11 minutes from three separate injections, and the pooled fractions were re-injected in ten separate injections The analogues containing an isopropyl side chain (isopropyl at R, of formula 1) derived from the incorporation of a 4-carbon (C-4, isobutyryl) starter unit eluted earlier, with the analogues containing a sec-butyl side chain (sec-butyl at R, of formula 1 ) derived from the incorporation of a 5-carbon (C-5, 2-methybutyryl) starter unit emerging several minutes later High resolution MS gave results for C-4 eryA, eryβ and eryD analogues and for C-5 eryA and eryB analogues, which correspond closely to those calculated

Measured Mass 762 5021 748 4820 748 5077

732 4933

In these experiments natural erythromycins were present only in low or undetectable amounts, and there were no detectable amounts of eryC analogues The overall concentration ratio of C-4/C-5 compounds in the fermentation broth as assessed by ESMS of ethyl acetate extracts of broths, was between 4 1 and 6 1 in favour of C-4 compounds The ratio of A B D analogues is variable, about 15 60 25, but with an increasing proportion of A analogues as the fermentation proceeds The total yield of erythromycins is about 400 μg/litre No supplementation with either isobutyπc or 2-methylbutyrιc acid was performed Thus, it would appear that the isobutyryl and 2-methylbutyryl starter units are derived from endogenously supplied precursors analogous to the synthesis of natural avermectins (e g , Hafner et al (1991 ), J Antibiot , 44349- 356)

Example 4a - Construction of S erythraea NRRL 2338/plG1

Approximately 5 μg of plasmid pIGl was transformed into protoplasts of S erythraea NRRL 2338 and stable thiostrepton resistant colonies are isolated From several such colonies, total DNA was obtained and analysed by Southern hybridisation, to confirm that the plasmid had integrated specifically into eryAI, and Southern analysis also showed that the site of integration was appropπate to generate a mutant capable of producing altered macrolides via the ave loading module

Example 4b - Construction of S erythraea NRRL 2338/pND30

Approximately 5 μg of plasmid pND30 was transformed into protoplasts of S erythraea NRRL 2338 and stable thiostrepton resistant colonies are isolated From several such colonies, total DNA was obtained and analysed by Southern hybridisation to confirm that the plasmid had integrated specifically into eryAI, and Southern analysis also showed that the site of integration was appropriate to generate a mutant capable of producing altered macrolides via the ave loading module

Example 5a - Preparation of 13-lsopropyl and 13-sec-Butyl Erythromycins Using S erythraea NRRL 2338/plG1 S erythraea NRRL 2338/plGl was inoculated into tap water medium containing 50 μg/mL thiostrepton and allowed to grow for four days at 30°C After this 20 mL of the mycelium was used to seed 500 mL of sucrose-succinate medium containing 50 μg /mL thiostrepton in a 2L flask with a single spring to reduce clumping, shaken at 280 rpm After between 3 5 and 6 days, the broth was filtered to remove mycelia and then extracted three times with a quarter volume of ethyl acetate The combined ethyl acetate extracts were dried over anhydrous sodium sulphate and solvent removed by evaporation Analysis of the product mixture using GC and electrospray MS revealed that of a total of 5-6 mg/L of 14-membered macrolide products the major component was sec-butyl-erythromycin D (about 1 5 mg/L), with other components present being sec-butyl-erythromycin B and sec-butyl-erythromycin A, isopropyl-erythromycins A B and D, and small amounts of natural erythromycins A, B and D S erythraea NRRL 2338/plGl produced approximately 10-15 times more of the novel isopropyl- and sec-butyl-erythromycins compared to the equivalent S erythraea ERMD1 construct (see Example 2) clearly demonstrating the capability of the actl promoter and its cognate activator gene actll-orf4 to enhance the expression of Type I PKS's Again, no supplementation with either isobutync or 2-methylbutyrιc acid was performed Thus, it would appear that the isobutyryl and 2-methylbutyryl starter units are derived from endogenously supplied precursors

Example 5b - Preparation of 13-lsopropyl and 13-sec-Butyl Erythromycins Using S erythraea NRRL 2338/DND30

S erythraea NRRL 2338/pND30 was inoculated into tap water medium containing 50 μg/mL thiostrepton and allowed to grow for four days at 30°C After this 20 mL of the mycelium was used to seed 500 mL of sucrose-succinate medium containing 50 μg /mL thiostrepton, in a 2L flask with a single spring to reduce clumping shaken at 280 rpm After between 3 5 and 6 days, the broth was filtered to remove mycelia and then extracted three times with a quarter volume of ethyl acetate The combined ethyl acetate extracts were dried over anhydrous sodium sulphate and solvent removed by evaporation Analysis of the product mixture using GC and electrospray MS revealed that of a total of 5-6 mg/L of 1 -membered macrolide products the major component was sec-butyl-erythromycin D (about 1 5 mg/L) with other components present being sec-butyl-erythromycin B and sec-butyl-erythromycin A isopropyl-erythromyαns A B and D and small amounts of natural erythromycins A B and D S erythraea NRRL 2338/pND30 produced approximately 10-15 times more of the novel isopropyl- and sec-butyi-erythromycins compared to the equivalent S erythraea ERMD1 construct (see Example 2) clearly demonstrating the capability of the actl promoter and its cognate activator gene actll-orf4 to enhance the expression of Type I PKS's Again, no supplementation with either isobutyπc or 2-m ethyl butyric acid was performed Thus it would appear that the isobutyryl and 2-methylbutyryl starter units are derived from endogenously supplied precursors

Example 6a - Preparation of 13-cvclopentyl-erythromycιn B using S erythraea NRRL 2338/plG1

The culture S erythraea NRRL 2338/plG1 was inoculated into 50 mL tap water medium in a 300 mL Erlenmeyer flask After 36 hours incubation at 28O, this flask was used to inoculate 3 5 L of ERY-P medium in a 5 L minijar The broth was incubated at 28°C with an aeration rate of 1 75 L/min Cyclopentane carboxylic acid (1 4 mL) was added after 24 hours and the fermentation was continued for 168 hours After this time the whole broth was adjusted to pH 8 5 with aqueous

sodium hydroxide and extracted with ethyl acetate (10 L) The ethyl acetate extract was concentrated to dryness giving the crude product as a gum (42 g) One gram of this extract was dissolved in ethyl acetate (5 mL) and added to a prepacked silica gel cartridge (10 g, International Sorbeπt Technology) previously conditioned with ethyl acetate (10 mL) The column was 5 sequentially eluted with ethyl acetate (4 x 10 mL) dichloromethane methanol (1 1) (2 x 10 mL), dichloromethane methanol ammonia (90 9 1) (1 x 10 mL), dichloromethane methanol ammonia (80 19 1) (1 x 10 mL) methanol (2 x 10 mL) Fractions 7-10 were combined and evaporated to dryness This fractionation was repeated on the remaining 3 2 g of gum This enrichment step yielded ca 920 mg of a gummy solid containing the desired product This was further purified by

I o preparative reversed-phase HPLC using a Zorbax 7 μm ODS column (21 2 mm x 25 cm) using a mobile phase of acetonitnle 0 05 M ammonium acetate (73) at 8 mlJmin Fractions, containing the product of interest from four separate injections were combined and evaporated to dryness before repeat preparative reversed-phase HPLC using a Beckman 5 μm Ultrasphere ODS column (10 mm x 25 cm) using a mobile phase gradient of acetonitnle 0 05 M ammonium acetate (28 72) to

15 acetonitnle 0 05 M ammonium acetate (50 50) over 18 minutes (flow rate 4 mL/min) Fractions containing the product of interest, from five separate injections were combined and evaporated to dryness to give a pure white solid (7 mg) The structure of the product was confirmed by mass spectrometry (MS) and nuclear magnetic resonance (NMR) spectroscopy as follows HPLC retention time - Method A - 26 0 minutes

0 APCI-MS - (M + H) ⁺ observed at m/e 758, required for C40H72NO12 - 758

NMR data

Atom number ³ C chemical shift from "C NMR spectrum Η chemical shift and multiplicity 1 1760

2 450 2891Hdq J = 94 71

3 804 4021Hdd J = 94, 17

4 392 208 IHmultiplet

5 838 3591Hd J = 74

6 754

7 380 2001H d J= 147, 108 ca 1651H muftiplet

8 450 2711Hdqd J= 106, 68, 26

9 ca 2200

10 389" 2981Hqd J = 68 15 11 694 3731Hdd J = 99 12 12 388 ^* 171 IHmultiplet 13 783 5191Hdd J = 105, 10 14 41 7 2151 H br sextet

15 304 ca 169 IHmultiplet ca 121 IHmultiplet

16 254 ca 1631 H multiplet ca 1521 H multiplet

17 251 ca 163 IHmultiplet ca 1521H multiplet

18 290 ca 169 IHmultiplet ca 1211H multiplet

19 158 1183Hd J = 71

20 924 1133Hd J = 70

21 274 1463HS 2 185 1143Hd J = 68 3 95 0993Hd J = 68 4 916 0863Hd J = 71 V 1032 4431Hd J = 73

2' 709 3241Hdd J = 103, 73

3' 654 251 IHddd J = 120 106 41

4' 290 ca 168 IHmultiplet ca 124 IHmultiplet 5' 688 3501 H br sextet

6' 215 1223Hd J = 61 ',8' 401 2322 3Hs

1" 965 4901Hd J = 46 " 351 2361Hd J = 152 + small (<1 Hz)

1581H multiplet " 726 " 780 302 IHd J = 91

5" 65 6 4 01 1 H multiplet

6" 18 7 1 29 3H d J = 6 3

7" 21 4 1 24 3H s

8" 49 5 3 31 3H s

^* - Assignments for signal with asterisks may be interchangeable

Example 6b - Preparation of 13-cvclopentvi-ervthromvcιn B using S erythraea NRRL 2338/PND30 An experiment similar to example 6a using the culture S erythraea NRRL 2338/pND30 produces the compound exemplified in example 6a

Example 7a - Preparation of 13-cvclobutyl-ervthromvcιn B using S erythraea NRRL 2338/plGl

The culture S erythraea NRRL 2338/plGl was inoculated into 50 mL tap water medium in 3 x 300 mL Erlenmeyer flasks After 72 hours incubation at 28°C this flask was used to inoculate 3 5 Lof ERY-P medιum ιπ 3 x 5 L minijars The broth was incubated at 28°C with an aeration rate of 2 0 LΛnin and stirring at 500 rpm Two feeds of cyclobutane carboxylic acid (1 4 mL) were added after 24 hours and 48 hours and the fermentation was continued for 168 hours After this time, the pH of the whole broth was adjusted to 8 5 with aqueous sodium hydroxide and then extracted with ethyl acetate (20 L) The ethyl acetate extract was concentrated to dryness giving the crude product as a gum (9 2g) A portion (2 3 g) of this extract was dissolved in ethyl acetate (12 5 mL) and added to a prepacked silica gel cartridge (10 g, International Sorbent Technology) previously conditioned with ethyl acetate (10 mL) The column was sequentially eluted with ethyl acetate (4 x 24 mL), dichloromethane methanol (9 1 ) (1 x 24 mL), dichloromethane methanol (8 2) (2 x 24 mL), dichloromethane methanol ammonia (80 19 1) (1 x 24 mL), methanol (1 x 24 mL) Fractions 6-8 were combined and evaporated to dryness This fractionation was repeated on a remaining sample of gum (4 7 g) This enrichment step yielded ca 41 5 mg of a solid containing the desired product This was further purified by preparative reversed-phase HPLC using a Zorbax 7 μm ODS column (21 2 mm x 25 cm) using a mobile phase of acetonitnle 0 05 M ammonium acetate (3 1 ) at 8

mL miπ Fractions, containing the product of interest, from five separate injections were combined and evaporated to dryness before repeat preparative reversed-phase HPLC using a Beckman 5 μm Ultrasphere ODS column (10 mm x 25 cm) using a mobile phase gradient of acetonitnle 0 05 M ammonium acetate (28 72) to acetonitnle 0 05 M ammonium acetate (50 50) over 18 minutes (flow rate 4 mL/min) Fractions, containing the product of interest were combined and evaporated to dryness to give a pure white solid (27 mg) The structure of the product was confirmed by MS and NMR as follows

HPLC retention time - Method A - 22 3 minutes

APCI-MS - (M + H) observed at m/e 744 required for C39H70NO1 2 - 744

'H chemical shift and multiplicity

2 87 1 H dq J = 9 1 , 7 1 4 03 1 H d J = 9 1 2 08 I H multiplet

5 839 359 1Hd J=74

6 754

7 377 199 1H multiplet

164 1Hdd J = 150,30

8 445 2731 H br doublet of pentets

J = ca 7030

9 2194 0 389 297 1Hqd J=69 11 1 691 375 1Hdd J = 102 + small 2 373 162 1H multiplet 3 774 539 1Hdd J = 93, 11 4 370 260 1 H br sextet 5 252 ca 197 1H multiplet ca 182 IHmultiplet 6 177 ca 183 2H multiplet 7 242 ca 193 1 H multiplet ca 175 1 H multiplet 8 155 117 3Hd J = 71 9 93 112 3Hd J = 75 0 270 146 3Hs 1 180 114 3Hd J = 71 2 91 098 3Hd J = 69 3 93 081 3Hd J = 71 V 1034 443 1Hd J = 73

2' 709 327 1Hdd J = 103, 73 ' 652 264 IHmultiplet

4' 291 ca 172 IHmultiplet ca 122 1 H multiplet

5' 687 352 1Hbr sextet J = ca 61

6' 210 123 3Hd J = 61

7" 8' 400 238 2 x 3H s

1" 968 489 1Hd J = 45

2" 347 238 IHmultiplet

157 1HddJ = 150 50

3" 725

4" 778 302 1Hd J = 92

5" 653 402 IHmultiplet

6" 182 129 3Hd J = 62

7" 212 124 3Hs

8" 491 331 3Hs

Example 7b - Preparation of 13-cvclobutyl-ervthromvcιn B using S erythraea NRRL 2338/pND30 An experiment similar to example 7a using the culture S erythraea NRRL 2338/pND30 produces the compound exemplified in example 7a

Example 8a - Preparation of 13-(3-furanyl)-ervthromvcιn B using S erythraea NRRL 2338/plG1

The culture S erythraea NRRL 2338/plG1 was inoculated into 50 mLtap water medium in a 300 mL Erlenmeyer flask After 72 hours incubation at 28°C, this flask was used to inoculate 35 L of ERY-P medium in a 5 L minijar The broth was incubated at 28°C with an aeration rate of 175 L/min 3-Furoιc acid (14 g in 6 mL methanol) was added filter sterilised after 24 hours and the fermentation was continued for 138 hours After this time the pH of the whole broth was adjusted to 85 with aqueous sodium hydroxide and then extracted with ethyl acetate (10 L) The ethyl acetate extract was concentrated to dryness giving the crude product as a gum (38 g) A portion

of this extract (1 9 g) was dissolved in ethyl acetate (10 mL) and added to a prepacked silica gel cartridge (10 g, International Sorbent Technology) previously conditioned with ethyl acetate (10 mL) The column was sequentially eluted with ethyl acetate (4 x 24 mL), dichloromethane methanol (9 1 ) (1 x 24 mL), dichloromethane methanol (8 2) (2 x 24 mL), dichloromethane methanol ammonia (80 19 1) (1 x 36 L) methanol (1 x 24 mL) Fractions 8 and 9 were combined and evaporated to dryness This fractionation was repeated on the remaining 1 9g of gum This enrichment step yielded a solid containing the desired product Thiswas further purified by preparative reversed-phase HPLC using a Zorbax 7 μm ODS column (21 2 mm x 25 cm) using a mobile phase of acetonitnle 0 05 M ammonium acetate (3 I ) at 8 mi_/mιn Fractions containing the product of interest from three separate injections were combined and evaporated to dryness before repeat preparative reversed-phase HPLC using a Beckman 5 μ Uttrasphere ODS column (10 mm x 25 cm) using a mobile phase gradient of acetonitnle 0 05 M ammonium acetate (28 72) to acetonrtπle 0 05 M ammonium acetate (50 50) over 18 minutes (flow rate 4 mL/min) Fractions, containing the product of interest, from three separate injections were combined and evaporated to dryness to give pure 13-(3-furanyl)-erythromycιn B as a white solid (9 mg) The structure of the product was confirmed by mass spectrometry HPLC retention time - Method A - 17 0 minutes

APCI-MS - (M + H) ⁺ observed at m/e 756, required for C39H66NO13 - 756

¹ H chemical shift and multiplicity

293 1Hdq J = 81,70

414 1Hd J = 81

218 1Hm J = 70,72

361 1Hd J = 72

2.07 1Hdd J = 146, 113

170 1H dd J = 146, not resolved

8 448 278 1Hbrm J = 113, 70, 22 9 2197 10 392 3.05 1 H dq J = 69, not resolved

11 695 395 1 H dd J = 100, not resolved 12 414 188 1Hdq J = 100 70 13 692 647 1H complex m

14 1383 731 1H complex m J= 07, 18

15 1243

16 1086 631 1H complex m J = 18.07

17 1428 739 iHt J = 18

18 155 121 3Hd J = 70 '

19 87 116 3Hd J = 70 0 270 148 3Hs 1 183 118 3Hd J = 70 2 84 103 3Hd J = 69 3 87 088 3Hd J = 70 1' 1033 446 1Hd J = 74 2' 71 1 3.29 1Hm

3' 651 260 1H broad m

4' 293 178 1Hm

124 1Hm

5' 688 3.54 1Hm

6' 207 124 3H d (obscured) ', 8' 400 239 2x3Hs " 966 488 1Hbrd J = 49 " 352 239 1Hm

159 1Hdd J = 150,49 " 717 " 778 303 1 H d (obscured) " 661 403 1Hdq J = 90, 61 " 183 130 3Hd J = 61 " 212 126 3Hs " 491 333 3H s

" Assignments indicated with an asterisk maybe reversed

Example 8b - Preparation of 13-(3-furanyl)-ervthromvcιn B using S erythraea NRRL 2338/PND30 An experiment similar to example 8a, using the culture S erythraea NRRL.2338/pND30, produces the compound exemplified in example 8a

Example 9a - Preparation of 13-cyclopropyl-ervthromycιn B using S erythraea NRRL 2338/plG1

The culture S erythraea NRRL 2338/plGl was inoculated into 50 mL tap water medium in a 300 mL Erlenmeyer flask. After 72 hours incubation at 28°C. this flask was used to inoculate 3.5 L of ERY-P medium Thiostreptone (105 mg) was added immediately after sterilisation The broth was incubated at 28°C with an aeration rate of 2 L/min and stirring at 500 rpm Cyclopropane carboxylic acid (1.2 mL) was added after 24 hours and the fermentation was continued for 144 hours After this time, the whole broth was adjusted to pH 8 5 with aqueous sodium hydroxide and then extracted with ethyl acetate (3 L). The ethyl acetate extract was concentrated to dryness giving the crude product as a gum (1 7 g). This extract (0.85 g) was dissolved in ethyl acetate (10 mL) and added to a prepacked silica gel cartridge (10 g, International Sorbent Technology) previously conditioned with ethyl acetate (20 mL) The column was sequentially eluted with ethyl acetate (4 x 24 mL); dιchloromethane:methanol (9.1) (1 x 24 mL), dichloro-methane:methanol (8:2) (1 x 24 mL); dichloromethane:methanol:ammonia (80 19 1 ) (3 x 24 m L), methanol (1 x 24 mL). Fractions 6-9 were combined and evaporated to dryness This fractionation was repeated on the remaining 0.85 g of gum. This enrichment step yielded a solid containing the desired product. This was further purified by preparative reversed-phase HPLC using a Zorbax 7 μm ODS column (21.2 mm x 25 cm) using a mobile phase of acetonitnle 0 05 M ammonium acetate (3 1) at 8 mlJmin Fractions, containing the product of interest, from 4 separate injections were combined and evaporated to dryness before repeat preparative reversed-phase HPLC using a Beckman 5 μm Ultrasphere ODS column (10 mm x 25 cm) using a mobile phase gradient of acetonitnle 0 05 M

ammonium acetate (28:72) to acetonιtrile:0.05 M ammonium acetate (50:50) over 18 minutes (flow rate 4 mL/min). Fractions, containing the product of interest, from three separate injections were combined and evaporated to dryness to give pure i3-cyclopropyl-erythromycin B as a white solid (9 mg) The structure of the product was confirmed by mass spectrometry. HPLC retention time - Method A - 17.9 minutes

APCI-MS - (M + H) ⁺ observed at m/e 730, required for C3βH68 ^N Oι 2 - 730

¹ H chemical shift and multiplicity

2.88 1H dq J = 8.5, 7.1

4.04 1 H dd J = 8.5, 1 9

2.11 1 H br pentet J = ca 7

3.58 1H d J = 7.5

2.02 1H dd J = 14 7, 10 9

166 1H dd J = 147, 28

8 451 274 IHdqd J = 109, 71,28 9 2200 10 390 301 IHmultiplet

11 698 376 1Hdd J = 100, ca 14 12 404 184 IHdqd J= 10.0, 6.9, 1.2 13 785 468 1Hdd J = 92, 12 14 132 1.09 iHmultipiet 15 41 051 IHmultiplet 042 IHmultiplet

16 27 051 IHmultiplet

029 IHmultiplet

17 151 119 3Hd J = 71

18 93 114 3Hd J = ca71

19 274 146 3H s 0 183 116 3Hd J = 71 1 93 1001 3Hd J = ca72" 2 93 0.9983H d J = ca 69 ^*

1' 103.3 443 1Hd J = 72

2' 710 3.25 1Hdd J= 10.3, 7.2

3' 656 254 IHbrddd J = ca 12.5, 103, 4

4' 290 169 IHmultiplet

1.25 IH ultiplet

5' 692 351 IHbrdq J = ca11,6 6' 209 122 3H d J = 62 ',8' 399 233 2x3Hs 1" 971 487 1H brd J=ca5

2" 35 0 2 37 1H br d J = ca 15

1 57 1 H d J = 15 1 , 5 0 3" 72 5

4" 78 1 3 01 1 H br d J = 9 2

5" 66 0 4 02 1H dq J = 9 2, 6 2

6" 18 3 1 28 3H d J = ca 6 2

7" 21 0 1 24 3H s

8" 49 3 3 32 3H s

Assignments indicated with an asterisk maybe reversed

Example 9b - Preparation of 13-cyclopropyl-ervthromvcιn B using S erythraea NRRL 2338/pND30 An experiment similar to example 9a using the culture S erythraea NRRL 2338/pND30 produces the compound exemplified in example 9a

Example 10a - Preparation of 13-M-methyKhιo-ethyl)-erythromycιn B using S erythraea NRRL 2338/PIG1 The culture s erythraea NRRL 2338/plGl was inoculated into 1 L tap water medium in a

2 8 L Fernbach flask Thiostreptone (50 mg) was added immediately after inoculation After 84 hours incubation at 29 ^C C, this flask was used to inoculate 8 L of supplemented ERY-P medium (50 g/L Dextrose, 30 g/L Nutπsoy flour 3 g/L ammonium sulfate, 5 g/L NaCI, 6 g/L CaCOj, 10 g L sucrose 5 g/L corn steep solids, 0 5 g/L MgS0 ₄ and 1 mUL of P2000) in a 14 L fermentor jar The broth was incubated at 28°C with an aeration rate of 8L/mιn, stirring at 800 rpm and with pH maintained between 69 and 7 3 with NaOH or H ₂ S0 ₄ ( 15%) Methylthiolactic acid (3 2 L) was added after 24 and 48 hours Additional methylthiolactic acid (1 6 mL) was added after 120 hours The fermentation was continued for 142 hours After this time the whole broth was centrifuged to yield centrate (34 L) which was loaded onto a XAD-16 resin column (600 mL Rohm and Haas) The

resin column was then washed with water (1 8 L) and eluted with ethyl acetate (2 5 L) The ethyl acetate was partially concentrated (to 250 mL) and then the product of interest was extracted into 100 mM sodium phosphate buffer, pH 3 5 (1 3 L) The product was transferred back into ethyl acetate by adjusting the water to pH 9 with sodium hydroxide and then mixing with ethyl acetate (450 mL) The erythromycin rich ethyl acetate layer was separated evaporated to a gum (5 0 g) and then resuspended into 20% methanol (120 mL) which was loaded onto a CG-161 resin column (100 mL, Toso Haas) The resin column was sequentially eluted with 20% methanol (3 x 100 mL) 40% methanol (3 x 100 mL), 60% methanol (3 x 100 mL), 80% methanol (3 x 100 mL) and neat methanol (4 x 100 mL) The neat methanol fractions 2 and 3 containing the product of interest were evaporated to a solid (220 mg) and further purified over a reversed-phase 10 urn Kromasil C18 HPLC column (75 mm x 25 cm) using a mobile phase of acetonrtrile 0 05 M ammonium acetate with 0 1% tπfluσroacetic acid gradient (32 68) to (38 62) over 60 minutes at a flow rate of 215 mUmin Fractions containing the product of interest were combined (1 7 L) adjusted to pH 9 with sodium hydroxide and extracted into methylene chloride (300 mL) The methylene chloride layer was separated and evaporated to dryness to yield partially pure product The preparative HPLC step and extraction step was repeated to obtain pure i3-(1 -methylthιo- ethyl)-erythromycιn B (31 mg) The structure of the product was confirmed by MS and NMR spectroscopy (BrukerDMX500 MHz spectrometer) as follows HPLC retention time - Method B - 14 9 minutes

APCI-MS-(M+H) ⁺ observed at m/e 764, required for C38H70NO12S - 764

NMR data

Atom# 'H(ppm) 1 2 3 449 4 494 5 363 6 406 7 307 8 542 9

10 11 332 12 377 13 356 14 406

Exam ^p le 10b - Preparation of 13-(1-methylthιo-ethyl)-erythromycιn B using S erythraea NRRL 2338/PND30

An experiment similar to example 10a, using the culture S erythraea NRRL 2338/pND30 produces the compound exemplified in example 10a

Example 11 - Preparation of 13-cyclobutyl-ervthromvcιn B using S erythraea NRRL 2338 5 The culture S erythraea NRRL 2338 was inoculated into 50 mL tap water medium in a 300 mL Erlenmeyer flask After 48 hours incubation at 28°C, this flask was used to inoculate 50 mL of ERY-P medium in a 300 mL Erlenmeyer flask The broth was incubated at 28°C Cyclobutane carboxylic acid (20 mL) was added after 24 hours and the fermentation was continued for 168 hours After this time the whole broth was adjusted to pH 8 5 with aqueous sodium hydroxide and it ⁾ then extracted with ethyl acetate (50 mL) The ethyl acetate was separated and concentrated to dryness The sample was redissolved in methanol (1 mL) for HPLC-MS analysis This confirmed the production of 13-cyclobutyl-erythromycιn B by the untransformed non-recombinant NRRL 2338 as described in example 7 for the genetically-engineered strain containing the ayr loading module (NRRL 2338/plGl construct) 5 HPLC retention time - Method A 22 3 minutes

APCI-MS - (M + H) ⁺ observed at m/e 744 required for C39H70NO12 - 744

Example 12 - Preparation of 13 -cvclopropyi -erythromycin B using S erythraea NRRL 2338

The culture S erythraea NRRL 2338 was inoculated into 50 mL tap water medium in a 300 0 mL Erlenmeyer flask After 48 hours incubation at 28°C, this flask was used to inoculate 50 mL of ERY-P medium in a 300 mL Erlenmeyer flask The broth was incubated at 28'C Cyclopropane carboxylic acid (20 mL) was added after 24 hours and the fermentation was continued for 168 hours After this time, the whole broth was adjusted to pH 8 5 with aqueous sodium hydroxide and then extracted with ethyl acetate (50 mL) The ethyl acetate was separated and concentrated to 5 dryness The sample was redissolved in methanol (1 mL) for HPLC-MS analysis

This confirmed the production of 13-cyclopropyl-erythromycιn B, by the untransformed non-recombinant NRRL 2338 as described in example 9 for the genetically-engineered strain containing the ayr loading module (NRRL 2338/plG1 construct)

HPLC retention time - Method A - 17 9 minutes

APCI-MS - (M + H) ⁺ observed at m/e 730, required for C38H68NO1 2 - 730

Example 13a - Preparation of 13-cvc.obutyl-ervthromvcιn A using S erythraea NRRL 2338/plGl

The culture S erythraea NRRL 2338/plGl was inoculated into 50 mL tap water medium in 3 x 300 L Erlenmeyer flasks After 72 hours incubation at 28°C each flask was used to inoculate 3 5 L of ERY-P medium in 3 x 5 L minijars The broth was incubated at 28°C with an aeration rate of 2 0 L/min and stirring at 500 rpm Two feeds of cyclobutane carboxylic acid (1 4 mL) were added after 24 hours and 48 hours and the fermentation was continued for 168 hours After this time the pH of the whole broth was adjusted to 8 5 with aqueous sodium hydroxide and then extracted with ethyl acetate (20 L) The ethyl extract was concentrated to dryness giving the crude product as a gum (9 2g) A portion (2 3 g) of this extract (2 3 g) was dissolved in ethyl acetate (12 5 mL) and added to a prepacked silica gel cartridge (10g International Sorbent Technology) previously conditioned with ethyl acetate (10 mL) The column was sequentially eluted with ethyl acetate (4 x

24 mL), dichloromethane methanol (9 1) (1 x 24 mL), dichloromethane methanol (8 2) (2 x 24 mL), dichloromethane methanol ammonia (80 19 1) (1 x 24 mL) methanol (1 x 24 mL) Fractions 6 - 8 were combined and evaporated to dryness This fractionation was repeated on the remaining 4 7g of gum This enrichment step yielded 415 mg of a solid containing the desired product This was further purified by preparative reversed-phase HPLC using a Zorbax 7μm ODS column (21 2 mm x

25 cm) using a mobile phase of acetonitnle 0 05 M ammonium acetate (3 1) at 8 mL/min Fractions containing the product of interest, from 5 separate injections were combined and evaporated to dryness before repeat preparative reversed-phase HPLC using a Beckman 5 μm Ultrasphere ODS column (10 mm x 25 cm) using a mobile phase gradient of acetonitnle 0 05 M ammonium acetate (28 72) to acetonitnle 0 05 M ammonium acetate (50 50) over 18 minutes (flow rate 4 mL/min)

Fractions, containing the product of interest, were combined and evaporated to dryness to give a pure white solid (4 mg) The structure of the product was confirmed by MS and NMR spectroscopy as follows

HPLC retention time - Method A - 17 5 minutes

APCI-MS - (M + H)' observed at m/e 760, required for C _M H _ro N0 ₁₃ - 760

Nmrdata

Atom number Approximate ^,3 C chemical shift, from 'H - ⁿ C Η chemical shift and multiplicity correlation data

1 176 1 2 44 7 288 IH multiplet 3 79 8 3 99 I H multiplet 4 39 3 1 97 I H multiplet 5 83 4 3 57 1 H d J = 7 6 6 75 4 7 38 3 1 92 I H multiplet

1 72 I H multiplet

66

8 451 270 IHmultiplet

9 2225 10 374

307 1HbrqJ = 70 11 691

377 1Hbrs 12 761 13 771

510 1Hd J = 71 14 348 286 IHmultiplet 15 267 198 1 H multiplet 180 IHmultiplet

16 188 188 IHmultiplet 171 IHmultiplet

17 248 189 2H multiplet

18 158 119 3Hd J = 70

20 269 147 3Hs

21 181 116 3Hd J = 70

1' 1030 442 1Hd J = 72

2' 710 324 1Hdq J = 104 72

3' 653 251 IHmultiplet

4' 287 168 IHmultiplet 126 1 H multiple!

5' 692 349 IHmultiplet

6' 213 123 3H d J = 62 '.8' 400 238 2x3Hs " 963 489 1Hd J = 45 " 347 238 IH ultiplet

157 1Hdd J = 15050 " 730

4" 78 0 3 02 1H d J = 9 4

5" 65 6 4 00 I H multiplet 6" 18 4 1 29 3H d J = 6 2 7" 21 3 1 24 3H s 8" 49 4 3 32 3H s

Example 13b - Preparation of 13-cvclobutyl-erythromvcιn A using S erythraea NRRL 2338/pND30

An experiment similar to example 13a using the culture S erythraea NRRL 2338/pND30 produces the compound exemplified in example 13a

Example 14a - Preparation of 13-cvclopropyl-ervthromvcιn A using S erythraea NRRL 2338/plG1 Six 2800 mL Fernbach flasks were inoculated with S erythraea NRRL 2338/plG1 Each flask contained 1 L of tap water medium with 50 mg of thiostreptone added to each flask After 24 hours of incubation at 28°C, 200 ppm of cyclopropane carboxylic acid was added to each flask The flasks were incubated for 90 hours and composited into a sterile 8 L aspirator bottle The aspirator bottle was used to inoculate 31 gallons of ERY-P medium in a 500 gallon pilot vessel The broth was incubated run at 27°C to 29°C at a pH ranging from 6 7 to 7 4 with an aeration rate of 20 standard cubic feet/mm and stirring at 175 rpm Cyclopropane carboxylic acid (200 mg/L) was added after 33 hours, 81 hours and 117 hours The fermentation was continued for 198 hours After this time, the whole broth was filtered over a 0 2 μm, ceramic filter (30 ft ² , U.S Filter) The filtrate was loaded onto an XAD-16 resin column ( 12 L Rohm and Haas) The resin column was then eluted with ethyl acetate (60 L) The ethyl acetate was concentrated to a gum (302 g) to which 2 L of methylene chloride was added The resulting methylene chloride solution was washed with 8 L of 250 mM sodium bicarbonate buffer, pH 9 The erythromycin rich methylene chloride layer was separated, evaporated to a gum (200 g) and then resuspended into 40% methanol (10 L) which was loaded onto a CG-161 resin column (9 L, Toso Haas) The resin column was washed with 40% methanol (30 L) and sequentially eluted with 75% methanol (8 x 10 L) and

neat methanol (3 x 10 L) The 75% methanol fractions 5 through 8, containing the product of interest, were combined and evaporated to 3 2 L The concentrate was adjusted to pH 9 and added to 0 95 L of methylene chloride The methylene chloride layer was separated and evaporated to yield 12 4 g of a gum Part of the gum (six grams) was further purified by preparative reversed-phase HPLC using a Kromasil 10 μm C18 HPLC column (75 mm x 25 cm) using a mobile phase of methanol 0 05 M ammonium acetate with 0 1 % tπfluoroacetic acid isocaratic (50 50) at a flow rate of 215 mL miπ Fractions containing the product of interest were combined (230 mL), evaporated to a concentrate (110 mL), adjusted to pH 9 with sodium hydroxide and extracted into methylene chloride (50 mL) The methylene chloride layer was separated and evaporated to dryness to yield 630 mg of partially pure product Another portion of the gum (one gram) was further purified by preparative reversed-phase HPLC using a MetaChem Inertsil 10 urn C8 column (50 mm x 25 cm) using a mobile phase of acetonitnle 0 05 M ammonium acetate with 0 1% trrfluoroacetic acid gradient (20 80) to (25 75) over 50 minutes at a flow rate of 125m!_/mιπ Fractions, containing the compound of interest (28-46 minutes) were combined saturated with sodium bicarbonate and extracted with methylene chloride The methylene chloride was separated and evaporated to dryness to yield 361 mg of partially pure product A portion of these partially pure materials was further purified by reversed-phase HPLC as exemplified by Phenomenex Prodigy 10 μm C18 column (50 x 50 mm) using a mobile phase of methanol 0 05 M ammonium acetate with 0 1 % trrfluroacetic acid isocratic (5050) at a flow rate of 100 mL min Fractions, containing the product of interest (27-31 minutes), were combined, saturated with sodium bicarbonate, and extracted with methylene chloride The methylene chloride was separated and evaporated to dryness to yield partially pure product Material was further purified by preparative reversed-phase HPLC using a Phenomenex Prodigy 10 μm Cl 8 column (50mm x 25 cm) using a mobile phase of methanol 0 05 M ammonium acetate with 0 1% tπfluro acetic acid isocratic (48 52) at aflow rate of 100mL/mιn Fractions containing the product of interest (41-45 minutes) were combined, saturated with sodium bicarbonate and extracted with methylene chloride The methylene chloride was separated and evaporated to dryness to yield 13-

cyclopropyl-erythromycin A as a solid (20 mg) The structure of the product was confirmed by MS and NMR spectroscopy (Bruker DMX 500 MHz spectrometer) as follows HPLC retention time - Method B - 5.6 minutes APCI-MS - (M+H)* observed at m/e 746, required for - 746

70

37 1 64 2 0 47/0 32

Example 14b - Preparation of 13-cvclopropyl-erythromvcιn A using S. erythraea NRRL

2338/PND30 An experiment similar to example 14a using the culture S erythraea NRRL 2338/pND30 produces the compound exemplified in example 14a

Example 15a - Preparation of l3- (3-thιenyl)-ervthromvcιn B using S erythraea NRRL 2338/plGl

The culture S erythraea NRRL 2338/plGl was inoculated into 50 mL tap water medium in a 300 mL Erlenmeyer flask After 48 hours incubation at 28°C, 5 mL of this inoculum was used to inoculate 50 mL of ERY-P mediun in a 300 mL Erlemneyer fiask The broth was incubated at 28°C The N-acetyl cysteamine thioester of 3-thιophene carboxylic acid (20mg in 0 5 mL of methanol) was added filter sterilised after 24 hours and the fermentation was continued for 168 hours After this time, the whole broth was adjusted to pH 8 5 with aqueous sodium hydroxide and then extracted with ethyl acetate (50 mL) The ethyl acetate was separated and concentrated to dryness The sample was redissolved in methanol (1 mL) for HPLC-MS analysis This confirmed the production of 13-(3-thιenyl)-erythromycιn B

HPLC retention time - Method A - 20 0 minutes APCI-MS - (M + H) ^* observed at m/e 772, required for CsHβ OoS - 772

Example 15b - Preparation of 13- (3-thιenyl)-erythromycιn B using S erythraea NRRL 2338/PND30

An experiment similar to example 15a, using the culture S erythraea NRRL 2338/pND30, produces the compound exemplified in example 15a

Example 16 - Preparation of 6-deoxy- 13-cvclopropyl-erythromycιn B using S erythraea NRRL 18643

The culture S erythraea NRRL 18643, an e/yFmutant of S erythraea (Science, 252 114 5 April 1991) was inoculated into 1 L tap water medium in a 2 8 L Fernbach flask Cyclopropane carboxylic acid (200 mL) was added after 24 hours incubation at 29°C with 200 rpm agitation After three (3) days total incubation, one flask was used to inoculate 8 L of supplemented ERY-P medium (60 g/L cerelose, 30 g/L Nutπsoy flour, 3 g/L (NH ₄ ) _? SO„ 5 g/L NaCI, 6 g L corn steep solids, 0 5 g/L MgS0 ₄ , and 1 mlJL P2000) in 14 L fermentor jars The broth was incubated at 28°C with an aeration rate of 8 L/min, stirring at 800 rpm and with pH maintained between 6 9 and 7 3 with NaOH or H ₂ S 0 _< ( 15%) Cyclopropane carboxylic acid ( 1 6 m L) was added after 24 and 48 hours The fermentation, done in duplicate, was harvested after 163 hours total incubation time The pH of the whole broth was adjusted to 9 with sodium hydroxide and the broth was extracted with ethyl acetate (16 L) The ethyl acetate was concentrated to an oil in a 20 L Buchi rotoevaporation unit The oil was redissolved with 500 mL of methylene chloride Five hundred mL of water was added to this liquid, and the pH of the aqueous phase was adjusted to 9 with 10% ammonium hydroxide After shaking vigorously the methylene chloride layer was collected and evaporated in a 1 L rotoevaporation flask to yield 11 0 g of oily residue This material was dissolved with 250 mL of a 4 6 methanol water solution, and loaded onto an 80 mL CG-161 resin column (Toso Haas) The column was washed with 350 mL of 4 6 methanol water solution The column was then brief ly washed at 8 mLmin with a 73 methanol water solution until coloured impurities began to elute from the column (approximately 2 bed volumes wash) At this time, a 1 hour gradient run was initiated, with the concentration of methanol changing from 70% to 100%, over a 1 hour period, at a flow of 8 mUmin The fraction containing the product of interest was evaporated to dryness and further purified on a reversed-phase 10 μm Kromasil C18 HPLC column (50 mm x 25 cm), using a mobile phase of acetonitnle buffer consisting of 0 01 M ammonium acetate, 0 02% tnfluoroacetic acid and 26% acetonitnle (5 95) for 50 minutes at a flow rate of 120 mL/min This was followed by a linear gradient from (5 95) to (33 67) over the next 40 minutes Fractions containing the product of interest were combined (530 mL), adjusted to a pH of 9 with 10% ammonium hydroxide, and extracted into methylene chloride (400 mL) The methylene chloride

layer was separated and evaporated to dryness to yield purified 6-deoxy-13-cyclopropyl- erythromycin B (12 mg) The structure of the product was confirmed by MS HPLC retention time - method B - 21 4 minutes

APCI mass spectroscopy - (M+H) ^* observed at m/e 714, required for C^H^NO^ - 714

Example 17 - Preparation of 6-deoxy-l3-propyl-ervthromycιn B using S erythraea NRRL 18643

S erythraea NRRL 18643 was inoculated from a three day patch on _YPD agar (0 5% Difco yeast extract, 0 5% Difco Bacto peptone, 0 25% dextrose, 0 5% MOPS, 1 7% Difco Basto agar, pH adjusted to 7 0) into 25 mL of _YPD broth (0 5% Difco yeast extract, 0 5% Difco Bacto peptone 0 25% dextrose, 0 5% MOPS pH adjusted to 7 0) in a 250 mL Erlenmeyer flask The flask was incubated at 225 rpm, 29°C, for 48 hours 2 5 mL were inoculated into 25 mL ERY-P medium (5% dextrose, 3% Nutπsoy flour, 0 3% (NH ₄ )jS0 ₄ , 0 5% NaCI, 0 6% CaC0 ₃ , pH adjusted to 7 0) and incubated at 225 rpm, 29°C, for a total of 6 days Butyric acid was added to the flask at 24, 72 and 120 hours (400 ppm, 400 ppm, 200 ppm, respectively) Whole broth pH was then adjusted to 9 1 using 1N NaOH The sample was extracted twice with an equal volume of ethyl acetate Ethyl acetate phases were concentrated to dryness under nitrogen (50°C water bath), then resuspended in 1 0 mL methanol for HPLC-MS analysis This confirmed the production of 6- deoxy-13-propyl-erythromycιn B

HPLC retention time - Method C - 23 5 minutes APCI-MS - (M+HV observed at m/e 716, required for CH^NO,, - 716

Example 18 - Evaluation of Antibacterial Activity

An in vitro antibacterial assay was performed in microtiter and interpreted according to Performance Standards for Antimicrobial Disk Susceptibility Tests - Sixth Edition, Approved Standard, published by The National Committee for Clinical Laboratory Standards (NCCLS) guidelines Minimum inhibitor concentratιons(MICs) were obtained versus various bacteria For

example, Staphyiococcus aureus 80CR5 (macrolide susceptible strain) afforded values generally ranging from < 0 1 to 1 56 μg/mL