A specific role of the Saccharopolyspora erythraea thioesterase II gene in the function of modular polyketide synthases

Zhihao Hu1, Blaine A. Pfeifer2, Elizabeth Chao2, Sumati Murli1, Jim Kealey1, John R. Carney1, Gary Ashley1, Chaitan Khosla2 and C. Richard Hutchinson1

1 Kosan Biosciences, Hayward, CA 94545, USA
2 Department of Chemical Engineering, Chemistry and Biochemistry, Stanford University, Stanford, CA 94305, USA

Correspondence
Zhihao Hu
hu{at}kosan.com


   ABSTRACT
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES
 
Bacterial modular polyketide synthase (PKS) genes are commonly associated with another gene that encodes a thioesterase II (TEII) believed to remove aberrantly loaded substrates from the PKS. Co-expression of the Saccharopolyspora erythraea ery-ORF5 TEII and eryA genes encoding 6-deoxyerythronolide B synthase (DEBS) in Streptomyces hosts eliminated or significantly lowered production of 8,8'-deoxyoleandolide [15-nor-6-deoxyerythronolide B (15-nor-6dEB)], which arises from an acetate instead of a propionate starter unit. Disruption of the TEII gene in an industrial Sac. erythraea strain caused a notable amount of 15-norerythromycins to be produced by utilization of an acetate instead of a propionate starter unit and also resulted in moderately lowered production of erythromycin compared with the amount produced by the parental strain. A similar behaviour of the TEII gene was observed in Escherichia coli strains that produce 6dEB and 15-methyl-6dEB. Direct biochemical analysis showed that the ery-ORF5 TEII enzyme favours hydrolysis of acetyl groups bound to the loading acyl carrier protein domain (ACPL) of DEBS. These results point to a clear role of the TEII enzyme, i.e. removal of a specific type of acyl group from the ACPL domain of the DEBS1 loading module.


Abbreviations: ACP, acyl carrier protein; AT, acyltransferase; 6dEB, 6-deoxyerythronolide B; DEBS, 6-deoxyerythronolide B synthase; KS, ketosynthase; NRPS, non-ribosomal peptide synthetase; PKS, polyketide synthase; TE, thiosesterase; Carb, carbenicillin; Strep, streptomycin; Tet, tetracycline


   INTRODUCTION
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES
 
Complex polyketides are a large family of bacterial natural products, many of which have considerable medicinal importance. Although complex polyketides have diverse structures (O'Hagan, 1993), the carbon framework of these compounds is created by one type of enzyme called modular polyketide synthases (PKSs). Each of these large, multifunctional proteins, such as the ones responsible for the biosynthesis of erythromycin (Cortes et al., 1990; Donadio & Katz, 1992; Donadio et al., 1991), rapamycin (Aparicio et al., 1996; Molnar et al., 1996; Schwecke et al., 1995), rifamycin (August et al., 1998) and FR-008 (Hu et al., 1994), consists of sets of modules and each module contains two or more enzymic domains that catalyse a particular round of polyketide chain extension from simple acyl-coenzyme A (CoA) substrates.

The well-studied 6-deoxyerythronolide B synthase (DEBS) system is responsible for the biosynthesis of 6-deoxyerythronolide B (6dEB), the aglycone of the erythromycins, and consists of three large proteins – DEBS1, DEBS2 and DEBS3 (Caffrey et al., 1992; Cortes et al., 1990; Donadio et al., 1991) (Fig. 1). The biosynthesis of 6dEB starts with the acyltransferase domain (ATL) of the loading module selecting and loading propionyl-CoA onto the acyl carrier protein (ACPL) in the same module. After the ACPL-bound propionyl group is transferred to the first ketosynthase (KS1) active site of DEBS1, KS1 can catalyse the decarboxylative condensation between the propionate thioester, transferred onto KS1 of DEBS from the ACPL of the didomain module, and a 2-methylmalonate thioester attached to the 4'-phosphopantetheinyl group of the ACP1 domain, which has been loaded by the AT1 domain of module 1. The resulting formation of a 2-methyl-3-ketopentanoyl-ACP1 thioester (Fig. 1a, R=H) represents the typical reaction catalysed by the three basic domains in a module. The intermediate product can be passed onto the KS domain in another module or, as in the case of module 1 of DEBS1, can be reduced by the ketoreductase (KR1) domain before its transfer (Fig. 1). The other modules repeat the basic process step-by-step, with or without reduction of {beta}-carbonyl, dehydration of {beta}-hydroxy and reduction of {alpha},{beta}-C=C, in the manner dictated by the domain organization of modules and the architecture of DEBS. For the latter, the product 6dEB is synthesized from one propionyl-CoA starter unit and six methylmalonyl-CoA extender units through six rounds of decarboxylative condensation.



View larger version (27K):
[in this window]
[in a new window]
 
Fig. 1. (a) The model for 6dEB and 15-nor-6dEB production by the DEBS system. In heterologous hosts like Str. coelicolor and Str. lividans, both propionyl-CoA and acetyl-CoA are used by the PKS as starter units, whereas in the native host, Sac. erythraea, propionyl-CoA is preferentially used to produce the aglycone for the biosynthesis of the erythromycins. (b) Loading of the preferred substrate, propionyl-CoA, onto the ACPL of DEBS does not invoke the editing function of the ery-ORF5 TEII enzyme, whereas (c) non-specific loading of other acyl-CoA substrates results in hydrolysis of the acyl-ACPL thioester by this enzyme.

 
Studies have shown that selection of the primer and extender units is primarily controlled by the AT domains within each module (Donadio et al., 1991; Haydock et al., 1995; Lau et al., 2000; Marsden et al., 1994). Extender AT domains have a high degree of substrate specificity, whereas the ATL domain in DEBS1 exhibits considerable flexibility. Acetyl-CoA, butyryl-CoA and isobutyryl-CoA can serve as starter units in addition to propionyl-CoA in a heterologous host or in vitro (Kao et al., 1994; Lau et al., 2000; Wiesmann et al., 1995). When the DEBS genes are heterologously expressed in Streptomyces coelicolor CH999, 6dEB and 15-nor-6dEB (Fig. 1a) are produced in an approximately 4 : 1 ratio in R2YE liquid medium (Kao et al., 1994). The ratio of 6dEB to 15-nor-6dEB is increased further when a modified medium containing 0·96 g propionate l-1 is used (unpublished results). The reason for the production of 15-nor-6dEB in non-natural hosts remains unclear, although it is likely to be the result of a lower intracellular concentration of propionyl-CoA in Str. coelicolor compared to the native erythromycin producer, Saccharopolyspora erythraea (Kao et al., 1994), because there is a large kinetic preference for utilization of propionyl-CoA over acetyl-CoA by DEBS (Pieper et al., 1996).

Small type II thioesterase (TEII) proteins are encoded by genes found in most modular PKS and non-ribosomal peptide synthetase (NRPS) gene clusters. Usually, two types of thioesterases are involved in PKS and NRPS function. A type I TE domain is usually found at the carboxyl terminus of the last module to act in the sequence of events catalysed by a PKS or NRPS, whereas the TEII enzymes are separate, single proteins. The TEI is responsible for release of the acyl chain from the PKS (Gokhale et al., 1999), NRPS (Kohli et al., 2001; Schwarzer et al., 2001) or NRPS/PKS hybrid (Tang et al., 2000), whereas TEII is thought to serve an editing function to remove aberrant intermediates from the PKS and NRPS systems (Schwarzer et al., 2002). TEII loss-of-function mutations in some bacterial PKS, NRPS and NRPS/PKS gene clusters have been reported to result in greatly reduced polyketide or oligopeptide production (Butler et al., 1999; Doi-Katayama et al., 2000; Schneider & Marahiel, 1998; Xue et al., 1998). Co-expression of a cognate TEII gene with the pikromycin PKS genes increased production of both narbonolide and 10-deoxymethynolide in a Streptomyces host two- to sevenfold, compared with the amount produced by a strain without the TEII gene (Tang et al., 1999). In a study of model systems in vitro (Heathcote et al., 2001), the results support the idea that the TEII enzyme associated with the tylosin PKS can remove aberrant short-chain fatty acids from ACP domains that have been mis-loaded with a fatty acid as a consequence of the erroneous decarboxylation of ACP-bound {alpha}-carboxythioesters. In more recent work, the purified pikromycin pikAV TEII was shown to hydrolyse propionyl and butyryl derivatives of different ACP domains (including ACPL of DEBS1) about threefold faster than acetyl derivatives (Kim et al., 2002). Hence, although there is still limited mechanistic understanding, it is generally believed that the role of such TEII enzymes is to edit the modular PKS during the process of polyketide biosynthesis. Long-chain fatty acid synthesis in Escherichia coli also depends to some extent on the TesA and TesB TEII enzymes, although their exact role has not been established (Cronan & Charles, 1996).

Here we report the results of a study in which the Sac. erythraea ery-ORF5 TEII gene directly influenced the levels of 6dEB or a 15-substituted analogue produced in the natural Sac. erythraea host, in two different Streptomyces species and in E. coli. The specificity of the TEII enzyme was also evaluated via direct biochemical experiments. We found that this TEII not only boosted the production of 6dEB, but also dramatically increased the ratio of 6dEB to 15-nor-6dEB in the Streptomyces hosts bearing the eryA PKS genes. The enzyme also approximately doubled the production of 6dEB in E. coli. Further support for the ability of this TEII to edit the ACPL domain was obtained from biochemical analysis of the purified TEII, where it could be shown that it removed acyl groups from mis-primed ACPs in the loading module. This observation is supported by the discovery that a Sac. erythraea strain bearing a disrupted ery-ORF5 TEII gene produced a considerable amount of 15-norerythromycins, which were not found in culture extracts of the parent strain, due to utilization of acetyl-CoA instead of propionyl-CoA as the starter unit.


   METHODS
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES
 
Strains and plasmids.
Strains and plasmids made and used in this study are listed in Table 1.


View this table:
[in this window]
[in a new window]
 
Table 1. Bacterial strains and plasmids used in this study

 
Media and chemicals.
R5 (Kieser, 2000) and Luria Broth (LB) solid medium were used for protoplast generation and strain propagation. TSB (Kieser, 2000) and R6 liquid medium (see below) were used for seed and fermentation media for Streptomyces strains. IT plate medium (see below) was used to maintain K41-135 and its derivatives. R6 medium is composed of (l-1): 103 g sucrose, 0·25 g K2SO4, 10·12 g MgCl2.6H2O, 0·96 g sodium propionate, 0·1 g Difco Casamino acids, 5·0 g yeast extract, 28·2 g Bis-Tris propane (Sigma) and 2·0 ml trace elements solution (the same as used in R5 medium). After autoclaving, the following aqueous solutions were added (l-1): 10 ml 0·5 % (w/v) KH2PO4, 8 ml 2·5 M CaCl2.H2O and 15 ml 20 % (w/v) L-proline. IT plate medium contains (l-1): 5 g anhydrous glucose, 5 g tryptone, 0·5 g betaine hydrochloride, 5 g starch, 1 g corn steep liquor (50 %), 200 mg MgSO2.7H2O, 2 mg ZnSO4.7H2O, 0·8 mg CuSO4.5H2O, 0·2 mg CoCl2.6H2O, 4 mg FeSO4.7H2O, 80 mg CaCl2.6H2O, 10 g NaCl, 150 mg KH2PO4 and 20 g agar, adjusted to pH 7 with 20 % NaOH.

For the E. coli experiments, antibiotics at the following concentrations were added to LB media: carbenicillin (Carb), 100 µg ml-1; streptomycin (Strep), 20 µg ml-1; tetracycline (Tet), 7·5 µg ml-1. E. coli fermentation medium was supplemented with 5 mM sodium propionate, 50 mM monosodium glutamate and 50 mM succinic acid purchased from Sigma and prepared as stock solutions adjusted to pH 7·0.

Co-expression of ery-ORF5 TEII with eryA DEBS genes in Streptomyces strains.
For expression in Streptomyces strains the ery-ORF5 clone was prepared as follows. Using cosmid pKOS79-170 DNA (see Table 1) as the template, the ery-ORF5 gene was amplified by PCR with the forward primer 5'-d(TATGCATGAGCACCTGGCTGCGGCGG)-3', designed to introduce an NsiI site overlapping the start codon, and the reverse primer 5'-d(GGCCGGCCTCGACTTCGTGATCGCCTGA)-3', designed to introduce an FseI site downstream of the stop codon (the restriction sites are shown in bold type). The PCR product was cloned into ZERO-Blunt (Invitrogen), then a 0·7 kb NsiI–NsiI (one NsiI site is from the vector) fragment containing the ery-ORF5 gene was transferred into NsiI-cut pKOS146-83A (Table 1) to give plasmid pKOS146-101A. pKOS146-83A was made from pUC119 (Vieira & Messing, 1987), in which the HindIII–EcoRI polylinker was replaced by a HindIII–EcoRI fragment containing the actII-ORF4 gene and the divergent actI and actIII promoters from pWHM467 (Wohlert et al., 2001). Three fragments, the EcoRI–PacI fragment of pKOS146-101A, the HindIII–PacI fragment of pKOS146-88A (Table 1) and the EcoRI–HindIII fragment of pKOS146-87B (Table 1) were ligated together and packaged using a GigapackIII-plus (Stratagene) in vitro packaging kit. pKOS146-103A, identified from Carb-resistant E. coli transformants infected by the packaged mixture, contains the eryA DEBS genes and the ery-ORF5 TEII gene under the control of the actI and actIII promoters, respectively. Replacement of the BglII–PacI fragment of pKOS146-103A with the BglII–PacI fragment of pJRJ2 (Jacobsen et al., 1997) gave pKOS146-109, a DEBS KS1° version of pKOS146-103A. pKOS146-103A and pKOS146-109 were introduced by transformation into Str. lividans K4-114 and Str. coelicolor CH999 separately. The titres of erythromycin aglycones were measured by HPLC-MS analysis of culture extracts, as described below.

Construction of the Sac. erythraea ery-ORF5 TEII mutant strain.
This mutant was obtained by gene disruption as follows. A BamHI–BglII fragment (3 kb in size) containing eryF, ery-ORF5 and eryG genes from cosmid pKOS79-170 (Table 1) was subcloned into pLitmus 28 (BioLabs), previously cut with BamHI, to make pKOS146-119. The kanamycin resistance gene (kan) from Supercos1 (Stratagene) was removed as a SmaI–StuI fragment and inserted into the PshAI site of pKOS146-119 to give pKOS146-129B. Then the XbaI–EcoRI fragment from pKOS146-129B containing eryF, ery-ORF5, eryG and the kan genes was ligated with pKOS97-49B, previously digested with SpeI and EcoRI, to make pKOS146-129C.

To construct the ery-ORF5 disruptant strain, plasmid pKOS146-129C was introduced by transformation into E. coli ET12567(pUB307) (Flett et al., 1997), then mobilized into Sac. erythraea K41-135 by conjugation from ET12567 transformants, selecting for apramycin-resistant colonies on R5 plates (60 µg apramycin ml-1). After sporulation and propagation of the apramycin-resistant exconjugants of the K41-135(pKOS146-129C) recombinant strain on IT medium plates containing 50 µg kanamycin ml-1, kanamycin-resistant, apramycin-sensitive clones were chosen as potential double-crossover recombinants. The desired ery-ORF5 TEII disruptant strains were verified by Southern blot hybridization against pKOS146-129C as explained in the text.

Production and quantification of 6dEB and its analogues produced by Streptomyces strains.
Streptomyces transformants were picked into 6 ml TSB liquid medium with 50 mg thiostrepton l-1 and grown in culture tubes with shaking (250 r.p.m.) at 30 °C. After sufficient growth (normally 3–4 days), 2 ml portions of the cultures were transferred to 250 ml Erlenmeyer flasks containing 40 ml R6 medium (supplemented with appropriate antibiotics, and in the case of strains containing the DEBS KS1° expression plasmid, the propyl diketide was fed at a final concentration of 1 g l-1). The flasks were shaken at 30 °C and 250 r.p.m. for about 7 days, after which 1 ml culture was withdrawn and centrifuged (1200 g, 5 min). Samples of the supernatants were analysed by on-line extraction by LC-MS using a system composed of a 10-port, 2-position switching valve/injector, Beckman System Gold HPLC, an Alltech evaporative light scattering detector (ELSD) and a PE-SCIEX API100LC MS-based detector configured with an atmospheric pressure chemical ionization source. Clarified whole broth (50 or 100 µl) was loaded onto a Metachem Metaguard Inertsil ODS-3 guard column (particle size 5 µm, column 4·6x30 mm) that had been pre-equilibrated for 1 min with H2O (0·1 % HOAc) at 1 ml min-1, with the eluate being diverted to waste. At 30 s post-injection, a linear gradient to 15 % MeCN (0·1 % HOAc) over 1 min was initiated. At 2 min the direction of flow through the guard column was reversed and the eluate was diverted onto a Metachem Inertsil ODS-3 column (5 µm, 4·6x150 mm) pre-equilibrated with 15 % MeCN (0·1 % HOAc). Compounds were eluted using a linear gradient from 15 to 100 % MeCN (0·1 % HOAc) at 1 ml min-1 over 6 min, then 100 % MeCN (0·1 % HOAc) for 3 min. The eluate stream was split equally between the ELSD and MS detectors. Under these conditions, 6dEB elutes at 9·9 min and 15-nor-6dEB at 9·3 min. Titres were determined by comparing the ELSD response of samples to a standard curve constructed from a power fit to 15-nor-6dEB standards.

Production and quantification of erythromycins and analogues produced by the Sac. erythraea ery-ORF5 TEII mutant.
The Sac. erythraea K41-135 parental and ery-ORF5 mutant strains were cultured in TSB for about 3 days with shaking (250 r.p.m.) at 34 °C, then 2 ml of the culture was transferred into 40 ml F1 medium (Brunker et al., 1998) in 250 ml flasks and fermentation was continued for 9 days. Titres of erythromycins were determined as described by Carreras et al. (2002).

Authentic standards of 15-norerythromycins were prepared by bioconversion of 15-nor-6dEB using the methods described by Carreras et al. (2002). Authentic standards of 15-nor-6-deoxyerythromycins were prepared by the same methods using a mutant strain of Sac. erythraea having a defective eryF gene encoding the C6-hydroxylase. The fermentations were performed under the conditions described above, after which 1·5 ml culture was withdrawn and centrifuged (1200 g, 5 min). Samples of the supernatants were analysed by on-line extraction by LC-MS. For LC-MS analyses, 20 µl clarified broth was chromatographed on a Phenomenex Develosil column (5 µm, 4·6x150 mm) with a mobile phase of 39 % 3 : 2 (v/v) MeCN/EtOH (5 mM NH4OAc) at 1 ml min-1. The eluate was split 1 : 1 between a Sedex 55 ELSD and an Applied Biosystems Mariner time-of-flight mass spectrometer equipped with a turbo-ion spray source (source temp. 400 °C; spray tip potential 5500 V; nozzle potential 175 V). Under these conditions, standards of erythromycins C, A, D and B eluted at 5·8, 9·3, 10·1 and 17·8 min, respectively. For exact mass measurements by LC-MS, the [M+H]+ and [M-C8H14O3]+ ions of erythromycin B in samples were used to calibrate the mass scale.

15-nor-6-deoxyerythromycin B and 15-norerythromycin B were characterized by NMR (1H, 13C, COSY, HSQC and HMBC) and MS analyses:

15-nor-6-deoxyerythromycin B: 13C-NMR (CDCl3, 100 MHz); {delta} 217·6 (C9), 177·0 (C1), 104·3 (C1'), 97·0 (C1''), 84·0 (C5), 79·2 (C3), 78·0 (C4''), 72·5 (C3''), 70·6 (C2'), 70·4 (C13), 70·3 (C11), 69·2 (C5'), 65·6 (C3'), 65·6 (C5''), 49·3 (3''OMe), 45·3 (C8), 44·7 (C2), 43·4 (C4), 41·9 (C12), 41·1 (C10), 40·3 (NMe2), 35·7 (C6), 35·1 (C2''), 34·2 (C7), 28·5 (C4'), 21·4 (C6'), 21·2 (C3''Me), 19·6 (Me6), 18·2 (C6''), 18·0 (C14), 16·7 (Me8), 14·2 (Me2), 9·4 (Me4), 8·7 (Me12), 7·9 (Me10). HRMS: calculated for C36H66NO11, 688·4630; found, 688·4591.

15-norerythromycin B: 13C-NMR (CDCl3, 100 MHz); {delta} 219·3 (C9), 176·0 (C1), 103·3 (C1'), 96·8 (C1''), 83·4 (C5), 78·0 (C3), 77·9 (C4''), 75·6 (C6), 72·6 (C3''), 70·8 (C2'), 69·8 (C13), 69·7 (C11), 69·1 (C5'), 65·7 (C3'), 65·5 (C5''), 49·5 (3''OMe), 45·0 (C8), 44·7 (C2), 41·1 (C12), 40·3 (NMe2), 40·3 (C4), 39·6 (C10), 38·4 (C7), 35·0 (C2''), 28·6 (C4'), 27·3 (Me6), 21·4 (C6'), 21·4 (Me3''), 18·6 (Me8), 18·5 (C6''), 18·2 (C14), 14·7 (Me2), 9·4 (Me4), 9·1 (Me10), 8·8 (Me12). HRMS: calculated for C36H66NO12, 704·4580; found, 704·4565.

Co-expression of ery-ORF5 TEII and eryA DEBS genes in E. coli.
The E. coli K207-3 host strain for 6dEB production has been described previously (Murli et al., 2003). Briefly, this strain has four T7-promoter-regulated genes integrated in the chromosome: sfp (required to pantetheinylate the DEBS proteins), prpE (required to convert propionate to propionyl-CoA) and accA1/pccB [required to convert propionyl-CoA to (2S)-methylmalonyl-CoA)]. Plasmids pKOS207-129 and pBP130 expressing the DEBS1, and DEBS2 and DEBS3 proteins, respectively, from T7 promoters have been described previously (Murli et al., 2003; Pfeifer et al., 2001). Plasmid pKOS207-142a is similar to pKOS207-129 except that the NdeI–SpeI fragment encoding the DEBS1 PKS in pKOS207-129 is replaced by the NdeI–SpeI fragment encoding the DEBS1 module 2 only from pRSG64 (Gokhale et al., 1999). To generate an E. coli expression vector for ery-ORF5 that was compatible with the DEBS plasmids, the ery-ORF5 PCR fragment used to generate pKOS146-124B described below was cloned as a blunt PCR fragment into pCR-Blunt (Invitrogen), generating pKOS146-124, and sequenced. The NdeI–NsiI ery-ORF5-containing fragment was cloned from pKOS146-124 into pKOS116-172a (Dayem et al., 2002), generating pKOS149-159g92. Finally, the ery-ORF5-containing NdeI–AvrII fragment from pKOS149-159g92 was cloned into the backbone of pKOS207-15a (Murli et al., 2003) between the T7 promoter and the T7 terminator. The resulting plasmid, pKOS285-93, is a pACYC derivative containing T7 promoter-ery-ORF5-T7 terminator.

For polyketide analysis in E. coli, fresh transformants of production strains carrying the indicated plasmids were grown overnight in LB medium supplemented with the appropriate antibiotics (Carb, Strep and Tet). These cultures were diluted 1 : 50 into 25 ml fresh LB medium with Tet only in 250 ml shake flasks and grown at 37 °C until the OD600 reached 0·4–0·5. The cultures were cooled to room temperature (25 °C), induced with 0·5 µM IPTG and the following media supplements were added: 5 mM sodium propionate, 50 mM succinate and 50 mM monosodium glutamate. When necessary, propyl diketide was fed at a final concentration of 0·5 g l-1. The cultures were grown for an additional 48 h at 22 °C. At the end of the fermentation, the OD600 was determined and the cells were collected by centrifugation. Five millilitres of cell-free supernatant was extracted with an equal volume of ethyl acetate. The organic fraction (top layer) was removed and dried under vacuum. The residue was resuspended in 500 µl methanol. An appropriate dilution was analysed by LC-MS and quantified by ELSD as described previously (Dayem et al., 2002; Murli et al., 2003). Polyketides were quantified by comparing the ELSD peak area to a standard curve of peak areas generated from an authentic sample. Polyketide titres are reported as means with standard errors of duplicate or triplicate samples, determined from independent colonies of the strains analysed.

TEII protein purification and in vitro analysis.
The ery-ORF5 gene was expressed in E. coli using pKOS146-124B, a pET16b derivative that was constructed as follows. The ery-ORF5 gene in an NdeI–BamHI cassette was amplified from pKOS79-170 (Table 1) with primer pair 5'-d(TCATATGAGCACCTGGCTGCGGCG)-3' and 5'-d(TGGATCCGACTTCGTGATCGCCTGAGC)-3', then the NdeI–BamHI fragment containing ery-ORF5 was cloned between the NdeI and BamHI sites of pET-16b (Novagen) to make pKOS146-124b. LB cultures (100 ml) of BL21(DE3)/pKOS146-124B containing 100 µg Carb ml-1 were incubated at 37 °C to an OD600 between 0·5 and 1 and induced with 100 µM IPTG. Cultures were incubated overnight at 30 °C and pelleted, resuspended in 100 mM Tris buffer (pH 8·4) and sonicated. The TEII (~30 kDa) protein was purified at 4 °C using Ni-NTA affinity chromatography. For this, the lysate was mixed with Ni-NTA resin for 1 h at 4 °C and loaded into a column for elution. TEII eluted at between 70 and 100 mM imidazole. The purified protein was subsequently buffer-exchanged into 100 mM Tris (pH 8·4) with 10 % glycerol, frozen with liquid nitrogen and stored at -80 °C. Typical protein concentrations were 0·3 mg l-1.

To test whether TEII could hydrolyse acyl-CoA substrates, the Ellman reagent (5,5'-dithio-2-nitrobenzoic acid, DTNB) was used as described earlier (Heathcote et al., 2001). Briefly, assays contained 200 mM potassium phosphate (pH 7·4), 1·3 µM TEII and 1 mM CoA substrates and were quenched with 10 µl 20 mM Ellman's reagent solution (in 200 mM potassium phosphate, pH 7·4). The final assay volume was 300 µl and assays were monitored spectrophotometrically at 412 nm.

For the acyl-ACP hydrolysis assays, the following components were used: 100 pmol apo-ACP2 or AT6-ACPL, 50 pmol TEII, 1 µl MgCl2 (1 M), 1 µl [1-14C]acetyl-CoA (100 µCi ml-1, 50 mCi mmol-1), [1-14C]propionyl-CoA (100 µCi ml-1, 50 mCi mmol-1) or [1-14C]-butyryl-CoA (100 µCi ml-1, 50 mCi mmol-1) and Sfp (1 µM). The final reaction volume was brought up to 100 µl with buffer containing 50 mM HEPES, 100 mM NaCl and 10 mM EDTA, pH 7·0. The reaction temperature was 37 °C. Initially, samples were mixed without TEII or Sfp and one sample was analysed (as described below) to ensure that subsequent samples contained much higher radioactivity than that found in the background sample. Sfp was then added to further samples and they were pre-incubated for 5 min to allow Sfp to convert the apo-ACPs into the corresponding acyl-ACPs. (The 5 min incubation period was chosen based on separate experiments to verify the time required for complete Sfp labelling.) The reaction was initiated by the addition of TEII (or buffer to act as a negative control). After 2 min, 15 µl BSA (15 mg ml-1) was added and samples were subjected to TCA (10 % w/v, 800 µl) precipitation. Samples were incubated for 30 min on ice and centrifuged at 4 °C for 30 min at 13 000 r.p.m. The samples were then washed once with an additional 800 µl TCA, dissolved in 400 µl formic acid (98 %), added to scintillation cocktail and counted. As mentioned above, a sample without Sfp was analysed to obtain background radioactivity counts and successful Sfp labelling was indicated by samples, without TEII, that showed a significant increase in isolated radioactivity (typically about an order of magnitude difference). Therefore, a significant drop in the isolated radioactivity with the inclusion of TEII would indicate a successful hydrolysis of the acyl-ACP substrate by the TEII. Successful (or unsuccessful) reactions showed no change or showed no significant change in d.p.m. counts past the first 2 min time point, indicating that each reaction was essentially an ‘end point’ assay.


   RESULTS
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES
 
Co-expression of the ery-ORF5 TEII and eryA DEBS genes in Streptomyces hosts
The function of the ery-ORF5 TEII gene was initially examined by studying how it affected the production of 6dEB by Streptomyces host strains carrying the eryA DEBS genes on a low-copy-number plasmid vector. Co-expression of the TEII and DEBS genes was achieved by placing each of them under the control of the divergently oriented actIII and actI promoters, respectively, on a pRM5-derived vector (McDaniel et al., 1993) (see Methods) where the promoters are regulated by the positively acting actII-ORF4 gene to ensure expression in the early stationary phase of growth. Plasmid pKOS146-103A (Table 1) carrying the ery-ORF5 TEII and DEBS genes was introduced by transformation into Str. lividans K4-114 (Ziermann & Betlach, 1999) and Str. coelicolor CH999 (Kao et al., 1994). In each host the combined yield of 6dEB and 15-nor-6dEB in R6 growth medium, which contains 0·96 g propionate l-1 but no glucose, was increased approximately twofold over the amount produced by a comparable strain without the ery-ORF5 TEII gene (75–89 vs 30–40 mg l-1). Only a trace amount of 15-nor-6dEB was produced (less than 2 mg l-1), whereas the ratio of 6dEB to 15-nor-6dEB produced in the same medium by the strain without the TEII gene was 2 : 1, as shown in Fig. 2(a). In FM6-1 medium, which is not supplemented with propionate, 77·6 % of the total polyketide product was 6dEB in the extract from the CH999/pKOS146-103A strain tested, whereas only 33 % was 6dEB in the extract from the CH999/pKOS11-26* (Table 1) control strain. These data show that co-expression of the TEII gene approximately doubles the 6dEB titre and considerably lowers the amount of 15-nor-6dEB produced relative to 6dEB, even though the presence of this gene does not eliminate 15-nor-6dEB production as it does in a Sac. erythraea strain (see below).



View larger version (24K):
[in this window]
[in a new window]
 
Fig. 2. (a) LC-MS data for the Streptomyces control (upper panel) and TEII+ (lower panel) strains, and (b) LC-MS data for the Sac. erythraea control (K41-135; upper panel) and ery-ORF5 TEII mutant (KOS146-171; lower panel) strains.

 
Co-expression of the ery-ORF5 TEII and eryA DEBS KS1° genes in Streptomyces hosts
The increased production of 6dEB in the presence of the TEII gene observed in the above experiments might have resulted from the decreased formation of 15-nor-6dEB due to an effect of TEII on the loading module only, an increased DEBS productivity due to the editing effect of TEII on other modules, or a combination of both types of activity. To study these possibilities, the TEII gene was co-expressed with the DEBS KS1° mutant that produces 15-methyl-6dEB when the racemic (2S,3R)-2-methyl-3-hydroxylhexanoate N-propionyl cysteamine thioester (‘propyl diketide’) is fed to the culture. In this case, the substrate is loaded onto the KS2 domain, bypassing both the loading module and module 1 of DEBS (Jacobsen et al., 1997). The pKOS146-109 plasmid (Table 1) was constructed from the DEBS KS1° genes in the same manner as pKOS146-103A and introduced into K4-114 and CH999 by transformation. Twice as much 15-methyl-6dEB (30–40 mg l-1) was produced in R6 medium by both the K4-114/pKOS146-109 and CH999/pKOS146-109 compared with the amount produced by control strains carrying pJRJ2 that lacks the TEII gene. This result shows that the positive effect of TEII on polyketide production can be due to more than just its effect on the loading module in these host strains.

The observation that expression of the ery-ORF5 TEII gene in Streptomyces hosts boosted the production of 6dEB and its analogues at least twofold has been confirmed in E. coli (Pfeifer et al., 2002). Yet, in both types of bacteria studied here, the ery-ORF5 TEII gene clearly is not essential for DEBS to function properly, in contrast to the situation in some other actinomycetes where TEII mutations have caused nearly complete loss of macrolide biosynthesis (Butler et al., 1999; Doi-Katayama et al., 2000; Xue et al., 1998).

Erythromycin production in a Sac. erythraea ery-ORF5 TEII mutant
To ascertain whether the effect of the ery-ORF5 TEII gene noted above was indicative of its role in the normal host, we carried out a gene replacement experiment targeted at the chromosomal ery-ORF5 gene in an erythromycin-producing strain of Sac. erythraea. The kanamycin resistance gene (from Tn5) was inserted into the PshAI site of ery-ORF5 in place of the 3·1 kb BamHI–BglII segment, as described in Methods, to give pKOS146-129C (Table 1). This plasmid was introduced into Sac. erythraea K41-135 by interspecies conjugation and the transformants resistant to both kanamycin and apramycin were serially transferred on to solid media without selection to isolate kanamycin-resistant, apramycin-sensitive strains. Southern analysis of genomic DNA isolated from such strains was used to identify the ery-ORF5 mutants. pKOS146-129C hybridized to 3·1 kb fragments in BglII+BamHI- and XhoI+BglII-digested DNA from K41-135 (Fig. 3a, lanes 5 and 1). Hybridization to a BglII fragment larger than 4 kb was seen also (data not shown). In a strain with the mutated ery-ORF5 gene, this probe hybridized to 2·3 and 2·1 kb fragments in BglII+BamHI-digested DNA, 1·9 and 2·5 kb fragments in XhoI+BglII-digested DNA and a 2·3 kb fragment in BglII-digested DNA (Fig. 3a, lanes 6, 2 and 3). These data, when analysed as shown in Figs 3(b) and 3(c), allowed us to conclude that the TEII gene had been disrupted in the KOS146-171 strain (Table 1) by insertion of the kanamycin resistance gene.



View larger version (32K):
[in this window]
[in a new window]
 
Fig. 3. (a) Southern hybridization of genomic DNA from the ery-ORF5 TEII mutant KOS146-171. pKOS146-129c was used as probe. Lanes: 1, genomic DNA from K41-135 digested with XhoI and BglII; 2, genomic DNA from KOS146-171 digested with XhoI and BglII; 3, genomic DNA from K41-135 digested with BglII; 4, genomic DNA from KOS146-171 digested with BglII; 5, genomic DNA from K41-135 digested with BamHI and BglII; 6, genomic DNA from KOS146-171 digested with BamHI and BglII. (b, c) Genes, restriction enzyme sites and predicted size of restriction fragments surrounding the ery-ORF5 TEII genes in the KOS146-171 and K41-135, respectively.

 
Fermentation of the KOS146-171 TEII mutant strain in F1 medium optimized for erythromycin production followed by LC-MS analysis of the products recovered in the culture extract (see Methods) showed that the mutant strain produced the expected mixture of the erythromycins, along with notable amounts of three 15-norerythromycins that were not seen in the extract of the K41-135 parental strain that had been carried through the same procedures. Fig. 2(b) shows the relative mobility of two of these three compounds. Elemental compositions of the three new compounds were determined by high resolution mass spectrometry (HRMS). Under the electrospray ionization conditions used, erythromycins display prominent [M+H]+ and [M+Na]+ quasimolecular ions along with fragments corresponding to loss of the neutral sugar cladinose ([M-159]+) or mycarose ([M-145]+), and a fragment corresponding to desosamine (m/z 158; HRMS gives C8H16NO2). These data can be used to tentatively identify the new compounds as shown in Table 2. The identity of the compound eluting at 24·3 min was confirmed as 15-nor-6dEB by co-injection with an authentic sample. The compound eluting at 10·5 min was similarly identified as 15-norerythromycin B based on co-injection with an authentic standard prepared by bioconversion of 15-nor-6dEB using a mutant strain of Sac. erythraea containing inactivated eryA and eryF genes. The compound eluting at 23·5 min gave identical MS data to 15-norerythromycin B, suggesting it is an isomer of that compound such as 15-nor-6-deoxyerythromycin A. Co-injection with 15-nor-6-deoxyerythromycin A revealed it to be a different, as yet unidentified compound, however. Weber et al. (1991) have reported the production of 15-nor-6-deoxyerythromycins by a strain of Sac. erythraea in which the eryF gene had been inactivated by an insertion, but 15-norerythromycin B has not been reported as a product of Sac. erythraea. The combined yield of all erythromycins produced by the TEII mutant was decreased less than 20 % from the titre determined for the K41-135 strain.


View this table:
[in this window]
[in a new window]
 
Table 2. LC-MS data for selected erythromycins produced by 129CM

 
Co-expression of the ery-ORF5 TEII and eryA DEBS genes in E. coli
Consistent with the effect seen in Streptomyces, expression of the ery-ORF5 TEII gene has been shown to increase 6dEB titres twofold in E. coli (Pfeifer et al., 2002). The results presented above suggesting a role for TEII in editing the loading domain ACP in Streptomyces and Sac. erythraea led us to examine if a similar effect was detectable in E. coli, in an attempt to understand the observation of Pfeifer et al. (2002). We analysed 6dEB production by the DEBS enzymes, and 15-methyl-6dEB production by the DEBS module 2, DEBS2 and DEBS3 enzymes, in the presence and absence of ery-ORF5 expression. Consistent with the report of Pfeifer et al. (2002), an approximately twofold increase in 6dEB titres in E. coli was found when ery-ORF5 was co-expressed (Table 3). However, in the absence of a loading domain, i.e. when 15-methyl-6dEB was produced by diketide feeding to the strain with DEBS module 2, DEBS2 and DEBS3, ery-ORF5 co-expression did not increase titres significantly (Table 3). Analyses of protein extracts from these strains showed that ery-ORF5 co-expression did not affect the levels of the DEBS proteins (data not shown). These data support the suggested role of the ery-ORF5 TEII in editing the loading domain ACP of DEBS and thereby increasing titres in E. coli. We are unable to explain why, because for unknown reasons production of 15-nor-6dEB has not been observed in this host when the eryA DEBS genes are expressed under the growth conditions used in spite of ample acetyl-CoA levels (unpublished results). We speculate that the lack of an effect on 15-methyl-6dEB titres in E. coli is due to differences between the acyl-CoA pools in E. coli and Streptomyces sp. or Sac. erythraea that lead to less mis-acylation of the ACP domains in the extender modules in E. coli. Another factor may be the absence of the ACPL domain in the E. coli system compared with the DEBS KS1° mutant used for diketide feeding experiments in Str. coelicolor.


View this table:
[in this window]
[in a new window]
 
Table 3. Effect of ery-ORF5 TEII expression on polyketide production in E. coli

All plasmids were introduced into the production host K207-3. The DEBS1 gene was expressed from pKOS207-129, the DEBS module 2 gene from pKOS207-142a and the DEBS2 and DEBS3 genes from pBP130. The pACYC vector control used was pKOS164-185 and ery-ORF5 TEII gene was expressed from pKOS285-93.

 
Biochemical analysis of ery-ORF5 TEII specificity using acyl-thioester substrates
To investigate the substrate specificity of TEII, ery-ORF5 was expressed in E. coli BL21(DE3) using pKOS146-124B (Table 1), and the recombinant C-terminally hexa-His-tagged protein was purified to homogeneity using a Ni-NTA affinity column (Fig. 4a). A variety of acyl-thioesters were tested as candidate substrates for the enzyme, including acetyl-CoA, propionyl-CoA, malonyl-CoA, methylmalonyl-CoA and butyryl-CoA, and selected acyl-ACP substrates derived from the ACP domain of module 2 (expressed as an individual polypeptide; Xue et al., 1998) and the loading ACP domain (ACPL), expressed as a 60 kDa didomain protein containing the extender AT domain from DEBS module 6 fused to ACPL (Liou et al., 2003).



View larger version (50K):
[in this window]
[in a new window]
 
Fig. 4. (a) SDS-PAGE analysis showing purified TEII. (b) Liquid scintillation counting data for different TEII catalysed reactions (‘+’, with; ‘-’, without). For details, see text.

 
Using Ellman's reagent as an end point colorimetric indicator, in the presence of 1·3 µM TEII and a substrate concentration of 1 mM, hydrolysis of none of the acyl-CoA substrates was observed within detectable limits over a time-course of 20–30 min. This result suggested that one or more acyl-ACPs, rather than acyl-CoA compounds, were the primary physiological substrates of the TEII enzyme. To test this hypothesis, apo-forms of the ACP2 and ACPL proteins were converted into acyl-ACP substrates in vitro using the Sfp enzyme, a phosphopantetheinyl transferase with broad substrate specificity (Quadri et al., 1998). 14C-labelled propionyl-, acetyl- and butyryl-CoA were thus used to prepare 14C-labelled propionyl-, acetyl- and butyryl-ACPs, respectively (see Methods). As shown in Fig. 4(b), TEII effectively hydrolysed the acetyl group from acetyl-ACPL. However, TEII did not recognize (or recognized to a much weaker extent) acetyl-ACP2, propionyl-ACPL and butyryl-ACPL. [In the latter two cases, a slight increase in radiolabelling intensity was observed after TEII addition, presumably due to the fact that the Sfp catalysed reaction did not go to completion during the pre-incubation period (see Methods).] These results demonstrate that TEII has preferential specificity towards acetyl chains attached to the loading ACP domain of DEBS, and are consistent with our in vivo results described above.


   DISCUSSION
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES
 
Our initial decision to examine the effect of the ery-ORF5 gene on 6dEB production was motivated by the literature reports that TEII homologues are important for the production of polyketide, oligopeptide and related antibiotics: tylosin (Butler et al., 1999), pikromycin (Xue et al., 1998), rifamycin (Doi-Katayama et al., 2000) and surfactin (Schneider & Marahiel, 1998). Targeted mutations that resulted in lack of a particular TEII caused a major decrease in antibiotic production in all these cases. Conversely, inclusion of the cognate TEII gene increased production of the pikromycin aglycones considerably when the pik PKS genes were heterologously expressed in Str. coelicolor (Tang et al., 1999). In the latter case and in other heterologous systems involving expression of the eryA DEBS genes in Str. coelicolor or Str. lividans, the macrolide products can still be produced in amounts greater than 160 mg l-1 (in R6 medium) even in the absence of the cognate TEII gene. These observations led us to test if co-expression of the ery-ORF5 and eryA DEBS genes in Streptomyces sp. could boost 6dEB production.

This assumption has been validated by the results of the present study and we propose the following explanation. First and foremost, the ery-ORF5 TEII can cause a major decrease in the use of acetyl-CoA as a chain starter unit by DEBS, as reflected in the greatly decreased amount of 15-nor-6dEB produced in an ery-ORF5+ background versus that produced in the ery-ORF5 mutant. This observation is consistent with editing of the ACPL in the loading module of DEBS1 to remove acetate selectively and is supported by the qualitative biochemical analysis. In this analysis, the TEII enzyme shows a clear preference for an acetylated ACPL domain. [In contrast, the pikromycin TEII enzyme encoded by pikAV exhibits an approximately threefold kcat/Km preference for hydrolysis of the propionyl- versus acetyl-ACPL derivative of the same didomain protein (Kim et al., 2002).] This editing feature most likely derives from the ability of the loading AT to incorporate acetyl-CoA in lieu of the natural propionyl-CoA starter unit. Lack of such editing is the most likely reason for the appearance of the 15-norerythromycins in culture extracts of the Sac. erythraea ery-ORF5 mutant. Detection of 6-deoxyerythromycins in this experiment simply reflects the chance that EryF occasionally is unable to act on a compound before it is glycosylated. [Certain 15-nor-6-deoxyerythromycins have been seen previously in the extracts of a Sac. erythraea strain in which eryF was disrupted by insertion of plasmid DNA sequences (Weber et al., 1991), most likely because the insertion blocked expression of the ery-ORF5 gene immediately downstream of eryF.] Interestingly, the in vitro data show that butyryl-ACPL is not an ery-ORF5 TEII substrate, even though this same species is hydrolysed by the PikAV TEII (Kim et al., 2002). Although incorporation of an acetyl starter unit by DEBS certainly occurs in vivo, 6dEB derivatives resulting from utilization of a butyryl-CoA starter unit are not observed. That this is not observed to any significant extent in vivo is most likely because of insufficient butyryl-CoA.

Others (Heathcote et al., 2001; Kim et al., 2002) have proposed that the ery-ORF5 and PikAV TEII enzymes may also edit the ACP domains in other modules of DEBS to remove aberrantly loaded or decarboxylated substrates. We also found that the yield of 15-methyl-6dEB was increased twofold by co-expression of the ery-ORF5 and DEBS KS1° genes. Here, editing of the loading module ACPL is irrelevant because the KS1° mutation prevents processing of the substrates loaded onto the loading module. In contrast, co-expression of the ery-ORF5 TEII gene in the E. coli system had no effect on the titre of 15-methyl-6dEB produced by diketide feeding, but did increase the titre of 6dEB made by DEBS, suggesting again that the major role of the TEII enzyme is effected on the loading domain ACP. Consequently, the results of the present study favour the idea that an editing activity of the ery-ORF5 TEII in the initial step of 6dEB biosynthesis in both a heterologous and homologous genetic background is the main function of this enzyme.

It is tempting to extend the conclusion that the primary role of the Sac. erythraea TEII enzyme is editing the loading domain ACP to all TEII enzymes that are associated with the biosynthesis of macrolide, oligopeptide and related hybrid antibiotics. We caution against such broad interpretation because there are cases where the ATL and ACPL domains of a modular PKS normally choose more than one acyl-CoA substrate and produce two or more polyketide products; e.g. the avermectin (Burg et al., 1979) and lankamycin (Egan & Martin, 1970) PKSs. In fact, replacement of the DEBS1 loading module with that of the avermectin PKS in the wild-type Sac. erythraea NRRL 2338 strain is known to result in production of different 13-substituted erythromycins from {alpha}-branched chain fatty acids (Marsden et al., 1998; Pacey et al., 1998). The ery-ORF5 TEII obviously cannot effectively edit the ACPL of the avermectin PKS in this case. Nevertheless, mis-charging of the ACPL loading domain, in the absence of TEII editing, could greatly reduce the productivity of a modular PKS that strongly prefers one acyl-CoA starter unit.


   ACKNOWLEDGEMENTS
 
We thank Rick Desai for assaying the Str. coelicolor CH999/146-103A strain in FM medium, Nina Viswanathan for assistance with LC-MS analyses and Ralph Reid for bioinformatics studies. This research was supported in part by a grant from the NIH (GMO67937) to C. K.


   REFERENCES
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES
 
Aparicio, J. F., Molnar, I., Schwecke, T., Konig, A., Haydock, S. F., Khaw, L. E., Staunton, J. & Leadlay, P. F. (1996). Organization of the biosynthetic gene cluster for rapamycin in Streptomyces hygroscopicus: analysis of the enzymatic domains in the modular polyketide synthase. Gene 169, 9–16.[CrossRef][Medline]

August, P. R., Tang, L., Yoon, Y. J. & 9 other authors (1998). Biosynthesis of the ansamycin antibiotic rifamycin: deductions from the molecular analysis of the rif biosynthetic gene cluster of Amycolatopsis mediterranei S699. Chem Biol 5, 69–79.[Medline]

Brunker, P., Minas, W., Kallio, P. T. & Bailey, J. E. (1998). Genetic engineering of an industrial strain of Saccharopolyspora erythraea for stable expression of the Vitreoscilla haemoglobin gene (vhb). Microbiology 144, 2441–2448.[Abstract]

Burg, R. W., Miller, B. M., Baker, E. E. & 12 other authors (1979). Avermectins, new family of potent anthelmintic agents: producing organism and fermentation. Antimicrob Agents Chemother 15, 361–367.[Medline]

Butler, A. R., Bate, N. & Cundliffe, E. (1999). Impact of thioesterase activity on tylosin biosynthesis in Streptomyces fradiae. Chem Biol 6, 287–292.[CrossRef][Medline]

Caffrey, P., Bevitt, D. J., Staunton, J. & Leadlay, P. F. (1992). Identification of DEBS 1, DEBS 2 and DEBS 3, the multienzyme polypeptides of the erythromycin-producing polyketide synthase from Saccharopolyspora erythraea. FEBS Lett 304, 225–228.[CrossRef][Medline]

Carreras, C., Frykman, S. & Ou, S. (2002). Saccharopolyspora erythraea-catalyzed bioconversion of 6-deoxyerythronolide B analogs for production of novel erythromycins. J Biotechnol 92, 217–228.[CrossRef][Medline]

Cortes, J., Haydock, S. F., Roberts, G. A., Bevitt, D. J. & Leadlay, P. F. (1990). An unusually large multifunctional polypeptide in the erythromycin-producing polyketide synthase of Saccharopolyspora erythraea. Nature 348, 176–178.[CrossRef][Medline]

Cronan, J. E., Jr & Charles, O. (1996). Biosynthesis of member lipids. In Escherichia coli and Salmonella typhimurium: Cellular and Molecular Biology, pp. 629–630. Edited by E. C. C. Lin, K. B. Low, B. Magasanik, W. S. Reznikoff, M. Riley, M. Schaechter & H. E. Umbarger. Washington, DC: American Society for Microbiology.

Dayem, L. C., Carney, J. R., Santi, D. V., Pfeifer, B. A., Khosla, C. & Kealey, J. T. (2002). Metabolic engineering of a methylmalonyl-CoA mutase-epimerase pathway for complex polyketide biosynthesis in Escherichia coli. Biochemistry 41, 5193–5201.[CrossRef][Medline]

Doi-Katayama, Y., Yoon, Y. J., Choi, C. Y., Yu, T. W., Floss, H. G. & Hutchinson, C. R. (2000). Thioesterases and the premature termination of polyketide chain elongation in rifamycin B biosynthesis by Amycolatopsis mediterranei S699. J Antibiot (Tokyo) 53, 484–495.[Medline]

Donadio, S. & Katz, L. (1992). Organization of the enzymatic domains in the multifunctional polyketide synthase involved in erythromycin formation in Saccharopolyspora erythraea. Gene 111, 51–60.[CrossRef][Medline]

Donadio, S., Staver, M. J., McAlpine, J. B., Swanson, S. J. & Katz, L. (1991). Modular organization of genes required for complex polyketide biosynthesis. Science 252, 675–679.[Medline]

Egan, R. S. & Martin, J. R. (1970). Structure of lankamycin. J Am Chem Soc 92, 4129–4130.[Medline]

Evans, P. D., Cook, S. N., Riggs, P. D. & Noren, C. J. (1995). LITMUS: multipurpose cloning vectors with a novel system for bidirectional in vitro transcription. Biotechniques 19, 130–135.[Medline]

Flett, F., Mersinias, V. & Smith, C. P. (1997). High efficiency intergeneric conjugal transfer of plasmid DNA from Escherichia coli to methyl DNA-restricting streptomycetes. FEMS Microbiol Lett 155, 223–229.[CrossRef][Medline]

Gokhale, R. S., Hunziker, D., Cane, D. E. & Khosla, C. (1999). Mechanism and specificity of the terminal thioesterase domain from the erythromycin polyketide synthase. Chem Biol 6, 117–125.[CrossRef][Medline]

Haydock, S. F., Aparicio, J. F., Molnar, I. & 7 other authors (1995). Divergent sequence motifs correlated with the substrate specificity of (methyl)malonyl-CoA : acyl carrier protein transacylase domains in modular polyketide synthases. FEBS Lett 374, 246–248.[CrossRef][Medline]

Heathcote, M. L., Staunton, J. & Leadlay, P. F. (2001). Role of type II thioesterases: evidence for removal of short acyl chains produced by aberrant decarboxylation of chain extender units. Chem Biol 8, 207–220.[CrossRef][Medline]

Hu, Z., Bao, K., Zhou, X., Zhou, Q., Hopwood, D. A., Kieser, T. & Deng, Z. (1994). Repeated polyketide synthase modules involved in the biosynthesis of a heptaene macrolide by Streptomyces sp. FR-008. Mol Microbiol 14, 163–172.[Medline]

Hu, Z., Hopwood, D. A. & Hutchinson, C. R. (2003). Enhancing heterologous polyketide production in Streptomyces by exploiting plasmid co-integration. J Ind Microbiol Biotechnol (in press).

Jacobsen, J. R., Hutchinson, C. R., Cane, D. E. & Khosla, C. (1997). Precursor-directed biosynthesis of erythromycin analogs by an engineered polyketide synthase. Science 277, 367–369.[Abstract/Free Full Text]

Kao, C. M., Katz, L. & Khosla, C. (1994). Engineered biosynthesis of a complete macrolactone in a heterologous host. Science 265, 509–512.[Medline]

Kieser, T., Bibb, M. J., Buttner, M. J., Chater, K. F. & Hopwood, D. A. (2000). Practical Streptomyces Genetics. Norwich: John Innes Centre.

Kim, B. S., Cropp, T. A., Beck, B. J., Sherman, D. H. & Reynolds, K. A. (2002). Biochemical evidence for an editing role of thioesterase II in the biosynthesis of the polyketide pikromycin. J Biol Chem 277, 48028–48034.[Abstract/Free Full Text]

Kohli, R. M., Trauger, J. W., Schwarzer, D., Marahiel, M. A. & Walsh, C. T. (2001). Generality of peptide cyclization catalyzed by isolated thioesterase domains of nonribosomal peptide synthetases. Biochemistry 40, 7099–7108.[Medline]

Lau, J., Cane, D. E. & Khosla, C. (2000). Substrate specificity of the loading didomain of the erythromycin polyketide synthase. Biochemistry 39, 10514–10520.[CrossRef][Medline]

Liou, G. F., Lau, J., Cane, D. E. & Khosla, C. (2003). Quantitative analysis of loading and extender acyltransferases of modular polyketide synthases. Biochemistry 42, 200–207.[CrossRef][Medline]

MacNeil, D. J., Gewain, K. M., Ruby, C. L., Dezeny, G., Gibbons, P. H. & MacNeil, T. (1992). Analysis of Streptomyces avermitilis genes required for avermectin biosynthesis utilizing a novel integration vector. Gene 111, 61–68.[CrossRef][Medline]

Marsden, A. F., Caffrey, P., Aparicio, J. F., Loughran, M. S., Staunton, J. & Leadlay, P. F. (1994). Stereospecific acyl transfers on the erythromycin-producing polyketide synthase. Science 263, 378–380.[Medline]

Marsden, A. F., Wilkinson, B., Cortes, J., Dunster, N. J., Staunton, J. & Leadlay, P. F. (1998). Engineering broader specificity into an antibiotic-producing polyketide synthase. Science 279, 199–202.[Abstract/Free Full Text]

McDaniel, R., Ebert-Khosla, S., Hopwood, D. A. & Khosla, C. (1993). Engineered biosynthesis of novel polyketides. Science 262, 1546–1550.[Medline]

Molnar, I., Aparicio, J. F., Haydock, S. F., Khaw, L. E., Schwecke, T., Konig, A., Staunton, J. & Leadlay, P. F. (1996). Organisation of the biosynthetic gene cluster for rapamycin in Streptomyces hygroscopicus: analysis of genes flanking the polyketide synthase. Gene 169, 1–7.[CrossRef][Medline]

Murli, S., Kennedy, J., Dayem, L. C., Carney, J. R. & Kealey, J. T. (2003). Metabolic engineering of Escherichia coli for improved 6-deoxyerythronolide B production. J Ind Microbiol Biotechnol (in press).

O'Hagan, D. (1993). The Polyketide Metabolites. Chichester: Ellis Horwood.

Pacey, M. S., Dirlam, J. P., Geldart, R. W. & 7 other authors (1998). Novel erythromycins from a recombinant Saccharopolyspora erythraea strain NRRL 2338 pIG1. I. Fermentation, isolation and biological activity. J Antibiot (Tokyo) 51, 1029–1034.[Medline]

Pfeifer, B. A., Admiraal, S. J., Gramajo, H., Cane, D. E. & Khosla, C. (2001). Biosynthesis of complex polyketides in a metabolically engineered strain of E. coli. Science 291, 1790–1792.[Abstract/Free Full Text]

Pfeifer, B., Hu, Z., Licari, P. & Khosla, C. (2002). Process and metabolic strategies for improved production of Escherichia coli-derived 6-deoxyerythronolide B. Appl Environ Microbiol 68, 3287–3292.[Abstract/Free Full Text]

Pieper, R., Ebert-Khosla, S., Cane, D. & Khosla, C. (1996). Erythromycin biosynthesis: kinetic studies on a fully active modular polyketide synthase using natural and unnatural substrates. Biochemistry 35, 2054–2060.[CrossRef][Medline]

Quadri, L. E., Weinreb, P. H., Lei, M., Nakano, M. M., Zuber, P. & Walsh, C. T. (1998). Characterization of Sfp, a Bacillus subtilis phosphopantetheinyl transferase for peptidyl carrier protein domains in peptide synthetases. Biochemistry 37, 1585–1595.[CrossRef][Medline]

Schneider, A. & Marahiel, M. A. (1998). Genetic evidence for a role of thioesterase domains, integrated in or associated with peptide synthetases, in non-ribosomal peptide biosynthesis in Bacillus subtilis. Arch Microbiol 169, 404–410.[CrossRef][Medline]

Schwarzer, D., Mootz, H. D. & Marahiel, M. A. (2001). Exploring the impact of different thioesterase domains for the design of hybrid peptide synthetases. Chem Biol 8, 997–1010.[CrossRef][Medline]

Schwarzer, D., Mootz, H. D., Linne, U. & Marahiel, M. A. (2002). Regeneration of misprimed nonribosomal peptide synthetases by type II thioesterases. Proc Natl Acad Sci U S A 99, 14083–14088.[Abstract/Free Full Text]

Schwecke, T., Aparicio, J. F., Molnar, I. & 7 other authors (1995). The biosynthetic gene cluster for the polyketide immunosuppressant rapamycin. Proc Natl Acad Sci U S A 92, 7839–7843.[Abstract]

Tang, L., Fu, H., Betlach, M. C. & McDaniel, R. (1999). Elucidating the mechanism of chain termination switching in the picromycin/methymycin polyketide synthase. Chem Biol 6, 553–558.[CrossRef][Medline]

Tang, L., Shah, S., Chung, L., Carney, J., Katz, L., Khosla, C. & Julien, B. (2000). Cloning and heterologous expression of the epothilone gene cluster. Science 287, 640–642.[Abstract/Free Full Text]

Vieira, J. & Messing, J. (1987). Production of single-stranded plasmid DNA. Methods Enzymol 153, 3–11.[Medline]

Weber, J. M., Leung, J. O., Swanson, S. J., Idler, K. B. & McAlpine, J. B. (1991). An erythromycin derivative produced by targeted gene disruption in Saccharopolyspora erythraea. Science 252, 114–117.[Medline]

Wiesmann, K. E., Cortes, J., Brown, M. J., Cutter, A. L., Staunton, J. & Leadlay, P. F. (1995). Polyketide synthesis in vitro on a modular polyketide synthase. Chem Biol 2, 583–589.[Medline]

Wohlert, S., Lomovskaya, N., Kulowski, K., Fonstein, L., Occi, J. L., Gewain, K. M., MacNeil, D. J. & Hutchinson, C. R. (2001). Insights about the biosynthesis of the avermectin deoxysugar L-oleandrose through heterologous expression of Streptomyces avermitilis deoxysugar genes in Streptomyces lividans. Chem Biol 8, 681–700.[CrossRef][Medline]

Xue, Y., Zhao, L., Liu, H. W. & Sherman, D. H. (1998). A gene cluster for macrolide antibiotic biosynthesis in Streptomyces venezuelae: architecture of metabolic diversity. Proc Natl Acad Sci U S A 95, 12111–12116.[Abstract/Free Full Text]

Ziermann, R. & Betlach, M. C. (1999). Recombinant polyketide synthesis in Streptomyces: engineering of improved host strains. Biotechniques 26, 106–110.[Medline]

Received 23 September 2002; revised 24 April 2003; accepted 25 April 2003.



This Article
Abstract
Full Text (PDF)
Alert me when this article is cited
Alert me if a correction is posted
Citation Map
Services
Email this article to a friend
Similar articles in this journal
Similar articles in PubMed
Alert me to new issues of the journal
Download to citation manager
Google Scholar
Articles by Hu, Z.
Articles by Hutchinson, C. R.
Articles citing this Article
PubMed
PubMed Citation
Articles by Hu, Z.
Articles by Hutchinson, C. R.
Agricola
Articles by Hu, Z.
Articles by Hutchinson, C. R.


HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
INT J SYST EVOL MICROBIOL MICROBIOLOGY J GEN VIROL
J MED MICROBIOL ALL SGM JOURNALS
Copyright © 2003 Society for General Microbiology.