The clavulanic acid biosynthetic cluster of Streptomyces clavuligerus: genetic organization of the region upstream of the car gene

Encarnación Mellado1, Luis Miguel Lorenzana2, Marta Rodríguez-Sáiz1, Bruno Díez1, Paloma Liras2 and José Luis Barredo1

Área de Biotecnología, Antibióticos SA, Avenida de Antibióticos 59-61, 24009 León, Spain1
Área de Microbiología, Facultad de Biología, Universidad de León, 24071 León, Spain2

Author for correspondence: José L. Barredo. Tel: +34 987 895826. Fax: +34 987 895986. e-mail: jbarredo{at}antibioticos.it


   ABSTRACT
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS AND DISCUSSION
REFERENCES
 
The genetic organization of the region upstream of the car gene of the clavulanic acid biosynthetic gene cluster of Streptomyces clavuligerus has been determined. Sequence analysis of a 12·1 kb region revealed the presence of 10 ORFs whose putative functions, according to database searches, are discussed. Three co-transcriptional units are proposed: ORF10–11, ORF12–13 and ORF15–16–17–18. Potential transcriptional terminators were identified downstream of ORF11 (fd) and ORF15. Targeted disruption of ORF10 (cyp) gave rise to transformants unable to produce clavulanic acid, but with a considerably higher production of cephamycin C. Transformants inactivated at ORF14 had a remarkably lower production of clavulanic acid and similar production of cephamycin C. Significant improvements of clavulanic acid production, associated with a drop in cephamycin C biosynthesis, were obtained with transformants of S. clavuligerus harbouring multiple copies of plasmids carrying different constructions from the ORF10–14 region. This information can be used to guide strain improvement programs, blending random mutagenesis and molecular cloning, to optimize the yield of clavulanic acid.

Keywords: antibiotic biosynthesis, gene cluster, cytochrome P450, ferredoxin, PBP

Abbreviations: CA, clavulanic acid; PBP, penicillin-binding protein

The GenBank accession number for the 12162 bp sequence reported in this paper is AY034175.


   INTRODUCTION
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS AND DISCUSSION
REFERENCES
 
Streptomyces clavuligerus is used industrially for the production of clavulanic acid (CA), a ß-lactam compound with ß-lactamase-inhibitory properties, useful against ß-lactamase-producing micro-organisms when combined with conventional ß-lactam antibiotics (Reading & Cole, 1977 ). The biosynthetic pathway leading to cephamycin in S. clavuligerus has been well characterized and the corresponding genes have been identified (Paradkar et al., 1996a ; Liras, 1999 ). However, the CA biosynthetic route remains only partially understood. A pathway comprising a putative sequence of biochemical reactions leading to CA and clavams has been proposed (Fig. 1) (Liras & Rodríguez-García, 2000 ), but additional work needs to be done in order to determine the complete biosynthetic pathway.



View larger version (24K):
[in this window]
[in a new window]
 
Fig. 1. Biosynthetic pathway leading to CA (Liras & Rodríguez-García, 2000 ). The genes and enzymic activities characterized are the following: ceaS (also named pyc) encoding N2-carboxyethylarginine synthase (CEAS) (Khaleeli et al., 1999 ; Pérez-Redondo et al., 1999 ), bls encoding ß-lactam synthetase (BLS) (Bachmann et al., 1998 ; Bachmann & Towsend, 2000 ), cas2 encoding clavaminate synthase (CAS) (Elson et al., 1987 ; Hodgson et al., 1995 ), pah encoding proclavaminate amidinohydrolase (PAH) (Wu et al., 1995 ) and car (also named cad) encoding clavulanate-9-aldehyde reductase (CAR) (Nicholson et al., 1994 ; Pérez-Redondo et al., 1998 ).

 
A condensation reaction of arginine with D-glyceraldehyde 3-phosphate to form N2-(2-carboxyethyl)arginine (CEA) has been recently elucidated as the initial step of the CA pathway (Khaleeli et al., 1999 ). This biosynthetic step is catalysed by a TPP-dependent carboxyethylarginine synthase encoded by the ceaS (also named pyc) gene (Pérez-Redondo et al., 1999 ). The synthesis of the ß-lactam ring from CEA is catalysed by ß-lactam synthetase, a new type of ATP/Mg2+-dependent enzyme, encoded by the bls gene (Bachmann et al., 1998 ; Bachmann & Townsend, 2000 ). The crystal structure of this ß-lactam synthetase has been recently reported, revealing evolutionary changes with reference to its homologous class B asparagine synthetases, which are essential for ß-lactam biosynthesis (Miller et al., 2001 ). After ß-lactam ring formation, only three enzymes have been characterized: clavaminate synthase (CAS), proclavaminate amidinohydrolase (PAH) and clavulanate-9-aldehyde reductase (CAR). S. clavuligerus has two CAS isoenzymes (CAS1 and CAS2), encoded by the genes cas1 and cas2, which are at least 20 kb apart in the chromosome of S. clavuligerus (Marsh et al., 1992 ). CAS2 catalyses the hydroxylation of deoxyguanidinoproclavaminic acid to guanidinoproclavaminic acid and the further oxidative cyclization of proclavaminic acid to clavaminic acid (Elson et al., 1987 ). PAH, which hydrolyses the guanidic residue of the guanidinoproclavaminic acid to form proclavaminic acid, is encoded by the pah gene (Wu et al., 1995 ). CAR is likely to be involved in the direct conversion of the highly reactive intermediate clavulanate-9-aldehyde into CA (Nicholson et al., 1994 ) and is encoded by the car (also named cad) gene (Pérez-Redondo et al., 1998 ).

A total of 11 genes involved in CA biosynthesis have been located in the genome of S. clavuligerus in a region immediately contiguous to the cephamycin biosynthetic gene cluster (Jensen et al., 2000 ; Li et al., 2000 ). In addition to the five above-mentioned genes (ceaS, bls, cas2, pah and car), ORF6, ORF7, claR, cyp, fd and ORF12 have been described (Fig. 2). ORF6 and ORF7 encode respectively a putative ornithine acyltransferase and a protein involved in peptide transport (Hodgson et al., 1995 ). The claR gene, which encodes a regulatory protein similar to transcriptional activators of the LysR family, is involved in the regulation of the late steps of the pathway (Pérez-Redondo et al., 1998 ; Paradkar et al., 1998 ). While the genes cyp (encoding a cytochrome P450) and fd (encoding a ferredoxin) may be involved in an oxidative reaction late in the pathway, the function of the product of ORF12 is still unknown (Li et al., 2000 ). The cyp gene was disrupted in S. clavuligerus by insertion of the tsr gene, leading to the complete loss of CA production (Li et al., 2000 ). The enzymic complex cytochrome P450/ferredoxin is a common component in antibiotic gene clusters of other actinomycetes (Fouces et al., 1999 ). Ferredoxins have been shown to supply electrons to multicomponent monooxygenases in Streptomyces spp. (O’Keefe et al., 1991 ). Amplification of the pyc or claR genes in S. clavuligerus results in an increase of CA production; however, amplification of the car gene has no significant effect on production (Pérez-Redondo et al., 1999 ).



View larger version (14K):
[in this window]
[in a new window]
 
Fig. 2. Restriction map of the CA biosynthetic gene cluster. The region described in this work includes from ORF10 (cyp) to ORF19. The phages (f) or plasmids (p) described in the text are marked as thick or thin lines respectively. The ORFs revealed by Geneplot analysis and the direction of transcription are shown by arrows. The black arrow at the left represents pcbR, a PBP-encoding gene (Paradkar et al., 1996a ) located at the beginning of the cephamycin cluster. White arrows are previously described genes belonging to the CA cluster and black arrows are the genes described in this work. Note that ORF7 and ORF15 present high similarity. B, BglII, E, EcoRI, N, NotI, S, SacI.

 
Although preliminary studies in S. clavuligerus indicated the presence of the complete CA biosynthetic pathway in a discrete DNA region containing eight contiguous genes (Jensen et al., 1993 ), the existence of additional clustered genes required for CA production outside of this region is assumed (Li et al., 2000 ). Since the genes involved in antibiotic biosynthesis are clustered in most actinomycetes, we worked on the isolation and characterization of additional genes of the right region flanking the CA gene cluster. In this way, we have identified seven additional ORFs (from ORF13 to ORF19; Fig. 2). This information will contribute to clarifying the CA biosynthetic pathway and provides valuable tools for strain improvement of S. clavuligerus.


   METHODS
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS AND DISCUSSION
REFERENCES
 
Micro-organisms, phages and plasmids.
The CA-producing strain S. clavuligerus ATCC 27064 (Reading & Cole, 1977 ) was used as source of DNA. Escherichia coli SURE (Stratagene) and E. coli LE392 (Sambrook et al., 1989 ) were the hosts for {lambda}GEM12 phage derivatives. E. coli DH5{alpha} (Sambrook et al., 1989 ) was the recipient for high-frequency plasmid transformation. Phage {lambda}GEM12 (Promega) was used for the construction of the genomic library of S. clavuligerus. pBluescript I KS(+) and pBC KS(+) phagemids (Stratagene) were used for subcloning and sequencing. Single-stranded DNA for sequencing was produced with phagemids by standard techniques (Sambrook et al., 1989 ). pULVK99 (Chary et al., 1997 ) was used as E. coli-Streptomyces spp. shuttle vector and pHZ1351 (Pérez-Redondo et al., 1999 ) as targeted inactivation vector.

Transformation and fermentation.
Protoplast transformation of Streptomyces lividans and S. clavuligerus was as previously described (Hopwood et al., 1985 ; García-Domínguez et al., 1987 ). Chloramphenicol (30 µg ml-1) or ampicillin (100 µgml-1) were used for selection of transformants in E. coli. Transformants of S. clavuligerus were selected with thiostrepton (50 µg ml-1). Insertional inactivation was done as described by Pérez-Redondo et al. (1999) . CA production was tested by flask fermentation in TSB (tryptic soy broth 3%; Difco) every 12 h (Romero et al., 1984 ) and further quantified by derivatization with imidazole (Bird et al., 1982 ) and HPLC analysis (Mosher et al., 1999 ).

Construction and screening of a genomic library of S. clavuligerus.
The DNA of S. clavuligerus was purified as described by Hopwood et al. (1985) . Fragments of 17–22 kb were purified from Sau3AI partially digested DNA and ligated to {lambda}GEM12 according to standard procedures (Sambrook et al., 1989 ). Ligation products were packaged in vitro with the Gigapack II Gold kit (Stratagene), obtaining around 2x104 p.f.u. For the screening, the library was amplified in E. coli LE392, plated to obtain about 3x104 p.f.u., and hybridized with digoxigenin-labelled probes according to standard methods (Sambrook et al., 1989 ). Recombinant phages were amplified in liquid medium to purify their DNA (Sambrook et al., 1989 ).

Nucleic acid hybridization and sequencing.
Southern hybridization was carried out according to standard procedures (Sambrook et al., 1989 ) using digoxigenin-labelled probes. Sequencing clones were constructed with the Erase a base kit (Promega) and sequenced by the dideoxynucleotide method using Sequenase 2.0 (Amersham). The DNA sequence was analysed using the Dnastar and Winstar packages. Comparisons of the deduced polypeptides against the databases were accomplished using the BLASTP 2.2.1 program (Altschul et al., 1997 ). Protein alignments were done with the CLUSTAL V algorithm (Winstar).


   RESULTS AND DISCUSSION
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS AND DISCUSSION
REFERENCES
 
Cloning and sequence determination of the region upstream of the car gene
A region of DNA adjacent to the cephamycin gene cluster including a total of 11 genes (from ceaS to ORF12; Fig. 2) has been demonstrated to be involved in CA biosynthesis in S. clavuligerus (Jensen et al., 2000 ; Li et al., 2000 ). In order to analyse the genomic region upstream of the car gene, a {lambda}GEM12 library of S. clavuligerus was screened using as a probe a 600 bp EcoRI fragment containing the 5' flanking region of the car gene. Three recombinant phages (fALCL1, fALCL2 and fALCL3) were isolated. Since these phages covered only around 9 kb upstream of the car gene (Fig. 2), a further screening was performed with an 800 bp NotI probe located at the right edge of fALCL3. Three new recombinant phages (fALCL6, fALCL7 and fALCL11) were purified covering a region of around 25 kb upstream of the car gene (Fig. 2).

To construct ordered sets of deletion subclones, the NotI fragments of the recombinant phages previously purified were subcloned in pBluescript I KS(+) or pBC KS(+), generating the following plasmids: pALCL23 (insert of 12·5 kb), pALCL25 (insert of 12 kb) and pALCL26 (insert of 4·5 kb included in pALCL23) (Fig. 2). A nucleotide sequence of 12162 bp was obtained using pALCL23 and pALCL25 as template. Computer analysis with the Geneplot program, using the codon preference algorithm, revealed the presence of 10 ORFs designated from ORF10 to ORF19 (Fig. 3; Table 1). All of them showed the typical biased codon usage of Streptomyces and a mean G+C content of 72·5 mol%. Whereas from ORF14 to ORF18 were transcribed in the same direction, from ORF10 to ORF13 and ORF19 were oppositely oriented. The arrangement of the genes within the cluster suggests the possibility of three co-transcriptional units (Fig. 3): ORF10-ORF11 (spaced by 5 bp), ORF12-ORF13 (spaced by 0 bp), ORF15-ORF16-ORF17-ORF18 (spaced by 20 bp between ORF16-ORF17 and 0 bp between the rest). Moreover, two sequences recognized as potential transcriptional terminators with free energies of -75·6 kcal (-316·3 kJ) mol-1 and -58·6 kcal (-245·2 kJ) mol-1 were identified, with the aid of the mfold program (Mathews et al., 1999 ), downstream of ORF11 (fd) and ORF15, respectively (Fig. 4). The position of these terminators agrees to the above-proposed organization of the ORFs.



View larger version (43K):
[in this window]
[in a new window]
 
Fig. 3. Geneplot analysis revealing the ORFs in the 12162 bp sequenced region. The putative transcriptional units are indicated by arrows. Co-transcription of ORFs 10–11, ORFs 12–13 and ORFs 15–16–17–18 is proposed. The numbering above the ORFs indicates spacing in bp between each pair.

 

View this table:
[in this window]
[in a new window]
 
Table 1. Putative products and deduced functions of the ORFs

 


View larger version (13K):
[in this window]
[in a new window]
 
Fig. 4. RNA secondary structure prediction with the mfold program (Mathews et al., 1999 ). Two sequences acting as potential transcriptional terminators were identified downstream of (a) ORF11 (fd) (from 1636 to 1844 of the sequence) and (b) ORF15 (from 5236 to 5328 of the sequence), with free energies of -75·6 kcal (316·3 kJ) mol-1 and -58·6 kcal (245·2 kJ) mol-1, respectively.

 
Deduced products of the identified ORFs
The proteins deduced from the previously identified ORFs were used for searching the databases to obtain information about their putative functions. The results are summarized in Table 1.

ORF10 and ORF11. These ORFs correspond to the previously described cyp and fd genes encoding a cytochrome P450 and a ferredoxin, respectively (Jensen et al. 2000 ; Li et al., 2000 ). Although the role of this cytochrome/ferredoxin enzymic complex in the biosynthesis of CA is not clear, it could potentially be involved in some of the oxidation steps. Li et al. (2000) suggested a role for this complex in the oxidative reaction between clavaminic acid and clavaldehyde. However, other oxidative steps in the CA pathway, i.e. the transformation of deoxyguanidine proclavaminic acid to guanidine proclavaminic acid, the conversion of proclavaminic acid to dihydroclavaminic acid, and the conversion of dihydroclavaminic acid to clavaminic acid, have been described to be catalysed by the versatile enzyme CAS (Jensen & Paradkar, 1999 ). Further experiments need to be done to demonstrate the exact role of this oxidative complex in the CA biosynthetic pathway.

ORF12. This encodes a protein of 466 amino acids, with a molecular mass of 50607. A protein of 430 residues and 47081 Da was proposed by Li et al. (2000) . Using the translation start codon suggested by Geneplot analysis according to the codon bias, ORF12 starts at a GTG triplet located 108 nucleotides upstream of the start point proposed by Li et al. (2000) . The difference between the two sequences consists in the lack of a cytosine (CGACGACTCCCCCATG) 7 bp upstream of the translation start codon in the sequence under accession number AF200819. This means 36 additional residues at the N-terminal end of the protein. The predicted protein showed similarity to two uncharacterized proteins of Mycobacterium tuberculosis (LPQF protein and a hypothetical protein), to a secreted protein of Mycobacterium leprae and to ß-lactamases from Deinococcus radiodurans, Providencia stuartii, Streptomyces cacaoi, Streptococcus pyogenes and Pseudomonas aeruginosa. Although the protein includes a highly conserved SDN motif (residues 242–244), which plays a crucial role in the catalytic activity of class A ß-lactamases, other residues critical to the catalytic activity of ß-lactamases (STFK, EPELN and KGT) are absent.

ORF13. The deduced polypeptide presented significant similarity to an export pump for cysteine and other metabolites of the cysteine pathway such as N-acetyl-L-serine and O-acetyl-L-serine from E. coli and to an integral membrane protein from Streptomyces coelicolor. ORF13 could be involved in the transport of metabolites of the CA pathway, e.g. the N-acyl derivatives of clavaminic acid excreted into the medium (Elson et al., 1988 ). Whether the possible co-transcription of ORF12-ORF13 has any biological meaning is unclear.

ORF14. The deduced protein showed similarity to the following proteins: an acetyltransferase from Pseudomonas syringae responsible for tabtoxin resistance, a hypothetical protein from Deinococcus radiodurans, a hypothetical protein from Streptomyces coelicolor, and a phosphinothricin acetyltransferase from D. radiodurans. All these proteins contain conserved domains from the acetyltransferase (GNAT) family (Pfam00583). The modification of an antibiotic molecule by acetylation, methylation or phosphorylation of specific residues is one of the mechanisms conferring antibiotic resistance in Streptomyces (Fedorenko et al., 1985 ; Cundliffe, 1992 ). Thus, resistances to chloramphenicol (Murray et al., 1989 ), kanamycin (Joe & Goo, 1998 ) or aminoglycoside antibiotics (Zhu et al., 1999 ) are conferred by acetyltransferases. Since (i) apart from CA, S. clavuligerus produces other antibiotics, e.g. cephamycin C (Nagarajan et al., 1971 ), derivatives of olivanic acid (Brown et al., 1979 ), holomycin and an antibiotic related to tunicamycin (Kening & Reading, 1979 ), and (ii) the clustering of biosynthetic and resistance genes for the same antibiotic has been reported in several micro-organisms (Fouces et al., 1999 ), ORF14 could be involved in antibiotic resistance. Whether the product of ORF13, similar to an export pump for N-acetyl-L-serine and O-acetyl-L-serine, could be involved in the transport of the acetylated metabolites generated by the acetyltransferase encoded by ORF14 will be investigated.

ORF15. The ATG start codon overlaps the last codon of ORF16. The deduced polypeptide displayed a high similarity index to the product of the ORF7 also located in the CA biosynthetic cluster of S. clavuligerus. ORF7 encodes a putative peptide transport protein (Hodgson et al., 1995 ; Jensen et al., 2000 ). Moreover, it showed similarity to the oligopeptide-binding lipoprotein encoded by the SC2A11.11 gene of Streptomyces coelicolor. The polypeptides encoded by ORF15 and ORF7 share a conserved domain characteristic of bacterial extracellular solute-binding proteins (family 5): 84SLGESSEDGRVWTYRLREGLRY105 (ORF7) and 88GEGEVSDGGRTWTYRLRRGLRY109 (ORF15). The biological significance of the presence of two peptide transport proteins in the biosynthetic cluster is intriguing. The existence of two sets of paralogous genes, encoding a functionally equivalent protein, involved in the early steps of CA biosynthesis has been described (Jensen et al., 2000 ). While paralogous genes, expressed only in soy-based culture medium, are assumed for ceaS, bls, pah, cas2 and ORF6, individual disruption of ORF7, claR, car and cyp resulted in transformants defective in CA production, even in soy medium. However, all of them were able to produce other clavams (Jensen et al., 2000 ). The existence of ORF15 opens the question whether it is really a paralogous gene of ORF7 or whether it could be involved in clavam biosynthesis. The lack of CA production by the transformants inactivated in ORF7 described by Jensen et al. (2000) suggests that the second hypothesis may be more reasonable. Nevertheless, gene replacement of ORF15 resulted in loss of CA production, indicating that the protein encoded belongs to the CA biosynthetic pathway (L. M. Lorenzana & P. Liras, unpublished results).

ORF16.The product of ORF16 showed slight similarity to the hypothetical protein D, involved in the biosynthesis of rapamycin, from Streptomyces hygroscopicus and to the hypothetical protein SF1447 of Xylella fastidiosa. The polypeptide encoded by ORF16 includes the conserved hexapeptide 375LPRTGE380, which has been proposed to be responsible for a post-translational modification necessary for the proper anchoring of the proteins to the cell wall.

ORF17. The deduced polypeptide presented sequence similarity to the Y4RH protein (similar to biotin carboxylases) from Rhizobium sp., to a carboxylase from S. coelicolor and to a carboxylase and an argininosuccinate lyase from Mesorhizobium loti. It includes the carbamoyl phosphate synthase subdomain signature 2 (324IIEANPRT331). It is also noteworthy that ORF15, ORF16, ORF17 and ORF18 seem to be co-transcribed.

Two PBP-encoding genes are present at the right end of the sequenced region
ORF18. This ORF encodes a product with significant similarity to type A penicillin-binding proteins (PBP-A) from different bacteria, including Streptomyces coelicolor, S. griseus and Mycobacterium tuberculosis. A transpeptidase domain (Pfam00905) and an ATP/GTP binding site motif A (420GVTVGGKT427) are conserved in these proteins. The protein includes the conserved motifs of class A ß-lactamases 167KTG169, 220STKF223, 254STN256 (instead of SDN) and 426KTG428, but the consensus EPELN is absent. The name pbpA is proposed for ORF18. Two other genes (pcbR and pbp74) encoding PBPs have been described in the cephamycin cluster of S. clavuligerus (Paradkar et al., 1996b ; Pérez-Llarena et al., 1998 ). The similarity index of the protein encoded by pbpA to the products of pcbR (PBP type B involved in ß-lactam resistance) and pbp74 (high-molecular-mass PBP) was 22% and 35%, respectively, but lower in any case than the above-mentioned similarity to the PBPs from S. coelicolor and S. griseus.

ORF19. The truncated sequence of ORF19 starts at an ATG codon at position 11893 and encodes a protein which presented similarity to type 2 PBPs (PBP-2) from Streptomyces coelicolor, S. clavuligerus, Pasteurella multocida, Haemophilus influenzae and Rickettsia prowazekii. The name pbp2 is proposed for ORF18. The similarity index of the product encoded by pbp2 to the PBPs encoded by pcbR and pbp74 was 31% and 35%, respectively, being poorer than the above-stated similarity to the PBPs from S. coelicolor. The biological meaning of the presence of four PBP-encoding genes (pcbR, pbp74, pbpA and pbp2) in the contiguous biosynthetic clusters of CA and cephamycin could be related either to ß-lactam resistance (pcbR) or to cell wall synthesis. Remarkably, pcbR is located at one end (left in Fig. 2) of the CA cluster and pbpA-pbp2 are at the opposite end.

Targeted disruption of ORF10 and ORF14
To elucidate whether ORF10 and ORF14 were involved in CA biosynthesis, the plasmids pHZORF10:aph (Fig. 5a) and pHZORF14:aph (Fig. 5b), were constructed and transformed into S. clavuligerus. Both plasmids contain a kanamycin-disrupted copy of these ORFs in pHZ1351. Specifically, a 1·35 kb aph cassette was inserted in place of a 1195 bp EcoRI-SphI fragment of ORF10 and of a 495 bp SalI fragment of ORF14. After protoplast formation and regeneration of a selected transformant of each plasmid, the progeny was screened for a kanamycin-resistant/thiostrepton-sensitive phenotype, allowing the elimination of the plasmid.



View larger version (41K):
[in this window]
[in a new window]
 
Fig. 5. Restriction maps of the plasmids pHZORF10:aph (a) and pHZORF14:aph (b) used for inactivation of ORF10 and ORF14, respectively. (c-e) Genetic organization of the CA cluster in (c) the wild-type strain S. clavuligerus ATCC 27064, (d) S. clavuligerus/pHZORF10:aph and (e) S. clavuligerus/pHZORF14:aph. (f) Southern hybridization analysis of S. clavuligerus/pHZORF10:aph (NcoI, lane 1, and NruI, lane 2) and the parental untransformed strain (NcoI, lane 3, and NruI, lane 4) using as a probe the aph gene. (g) Southern hybridization analysis of the parental untransformed strain (BstXI, lane 1, and ApaLI, lane 3) and S. clavuligerus/pHZORF14:aph (BstXI, lane 2, and ApaLI, lane 4) using the aph gene as a probe. Expected patterns were obtained, confirming targeted disruption.

 
To verify the targeted inactivation, chromosomal DNA of both disrupted transformants was purified, digested with appropriate restriction enzymes and separately hybridized to aph-, ORF10- and ORF14-specific probes. Both disrupted transformants showed hybridizing bands of the expected size with the aph probe: (i) S. clavuligerus ORF10:aph gave two bands of 1·6 kb and 1·9 kb when it was digested with NcoI and a single band of 5·0 kb with NruI (Fig. 5d, f), and (ii) S. clavuligerus ORF14:aph showed single bands of 3·5 kb with BstXI and of 2·5 kb with ApaLI (Fig. 5e, g). The wild-type strain did not show any hybridizing signal (Fig. 5f, g). In the wild-type genome, the probe corresponding to ORF10 (0·7 kb SphI-NcoI) showed two hybridizing bands of 5·0 kb and 2·0 kb when it was digested with NruI and a single band of 3·4 kb with NcoI. Additionally, the ORF14-aph probe (1·0 kb ApaLI-BstEII) gave a band of 2·6 kb with BstXI and of 1·6 kb with ApaLI in the wild-type strain (not shown). These results confirmed the expected targeted integration patterns.

Once the targeted gene inactivation had been verified, the parental strain and the disrupted transformants were fermented in TSB medium. The ORF10-disrupted transformant failed to produce CA, whereas the inactivation of ORF14 generated a transformant with a remarkably lower production (33% of the parental strain; Fig. 6a). Although cephamycin C production was not affected by ORF14 gene disruption (data not shown), it was notably increased in the transformant inactivated in ORF10 (Fig. 6b). In this case, a part of the CA precursors could have been channeled toward cephamycin C. Insertional inactivation of ORF10 (cyp) and ORF12 was previously described (Li et al., 2000 ), leading to the complete loss of CA production and demonstrating the involvement of these genes in CA biosynthesis.



View larger version (17K):
[in this window]
[in a new window]
 
Fig. 6. CA (a) and cephamycin C (b) production by the parental untransformed strain S. clavuligerus ATCC 27064 ({bullet}) and transformants disrupted in ORF10 ({blacktriangleup}) or ORF14 ({blacksquare}). Inactivation of ORF10 or ORF14 gave rise to the lack of or a notable drop in CA production, respectively. Cephamycin C production was notably increased in the transformant disrupted in ORF10, but was not affected in the other transformant (this last result not shown). Fermentations were carried out in triplicate flasks. Vertical bars indicate standard deviations.

 
Taking into account that ORF12 and ORF13 could share the same transcript, it is likely that ORF13 belongs to CA cluster. Additionally, the reduction (but not loss) of CA production obtained by ORF14 disruption indicates its relationship to the biosynthetic pathway. In this case, the possibility to produce CA could be explained by the presence of a paralogous gene of ORF14 expressed in soy-based medium (TSB), as previously described for other genes of the CA cluster (Jensen et al., 2000 ). Moreover, both (i) the characterization of ORF15 as similar to ORF7, and (ii) the co-transcription of ORF15-ORF16-ORF17-ORF18, point to their likely association with the CA cluster. Assuming this hypothesis, the CA gene cluster could be located from the ceaS gene up to, at least, ORF18.

Overexpression of ORFs in S. clavuligerus
In order to determine whether any of the previously described ORFs has a positive effect on CA production, different plasmid constructions were introduced by transformation into S. clavuligerus ATCC 27064, testing CA and cephamycin C production in the recombinant strains.

The DNA fragments isolated from this region were cloned in pULVK99 (7·8 kb), generating the following plasmids (Fig. 2): pULVKcyp-fd (1·9 kb EcoRI-NcoI insert including ORF10-ORF11), pULVKORF12 (1·8 kb SphI-NruI insert including ORF12), pALCL66 (3·4 kb BglII-SalI insert including ORF12-ORF13), pULVKORF14 (1·7 kb NruI-BstEII insert including ORF14), pALCL36 (6·6 kb EcoRI-NotI insert including ORF10-ORF11-ORF12-ORF13-ORF14).

The effect of these genes on CA and cephamycin C production was tested by fermentation of a selected transformant of each plasmid (pULVK99, pULVKcyp-fd, pULVKORF12, pALCL66, pULVKORF14 and pALCL36) in triplicate flasks of TSB medium. In order to avoid interferences originated by the different rate of growth of the transformants, the production is shown as µg antibiotic per mg DNA (Fig. 7). Those transformants harbouring multiple copies of ORF10–11, ORF12 and ORF14 showed the higher improvements of CA production. At 34 h of fermentation, where the production level was maximum, increases of higher than 100% were obtained. Curiously, only a slight production increase was obtained with the transformant corresponding to pALCL36 (carrying from ORF10 to ORF14) (Fig. 7, top row). These results point to the involvement of the analysed genes in CA biosynthesis.



View larger version (32K):
[in this window]
[in a new window]
 
Fig. 7. CA and cephamycin C production by S. clavuligerus ATCC 27064 transformed with pULVK99 ({bullet}), pULVKcyp-fd ({circ}), pULVKORF12 ({triangleup}), pALCL66 ({blacktriangleup}), pULVKORF14 ({blacksquare}) and pALCL36 ({square}). Additional copies of the genes gave rise to increments of CA production, whereas cephamycin C production was lower in all the transformants. Fermentations were carried out in triplicate flasks. Vertical bars indicate standard deviations.

 
Cephamycin C production was higher in the transformant harbouring pULVK99 (without CA genes) (Fig. 7, bottom row), indicating that those transformants containing additional CA genes could have displaced their secondary metabolism toward CA production.

The complete characterization of the CA biosynthetic gene cluster will provide a very useful tool for the improvement of CA production. The development of some antibiotic-producing strains by increasing the copy number of the biosynthetic genes has been reported (Díez et al., 1997 ). This suggests that transforming CA-producing micro-organisms with these biosynthetic genes would improve CA productivity.


   ACKNOWLEDGEMENTS
 
The authors thank M. Sandoval, J. A. González, P. Merino, M. Mediavilla and M. T. García for their excellent technical assistance. L. M. Lorenzana was supported by a fellowship of the University of León. This work was partially supported by grant 1FD97-1419-CO2-02.


   REFERENCES
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS AND DISCUSSION
REFERENCES
 
Altschul, S. F., Madden, T. L., Schäffer, A. A., Zhang, J., Zhang, Z., Miller, W. & Lipman, D. J. (1997). Gapped BLAST and PSI-BLAST: a new generation of protein database search programs. Nucleic Acids Res 25, 3389-3402.[Abstract/Free Full Text]

Bachmann, B. O. & Townsend, C. A. (2000). Kinetic mechanism of the beta-lactam synthetase of Streptomyces clavuligerus. Biochemistry 39, 11187-11193.[Medline]

Bachmann, B. O., Li, R. & Townsend, C. A. (1998). ß-Lactam synthetase: a new biosynthetic enzyme. Proc Natl Acad Sci USA 95, 9082-9086.[Abstract/Free Full Text]

Bird, A. E. J., Bellis, J. M. & Gasson, B. C. (1982). Spectrophotometric assay of clavulanic acid by reaction with imidazole. Analyst 107, 1241-1245.

Brown, A. G., Corbett, D. F., Eglington, A. J. & Howarth, T. T. (1979). Structures of olivanic acid derivatives MM 22380, MM 22381, MM 22382 and MM 22383; four new antibiotics isolated from Streptomyces olivaceus. J Antibiot 32, 961-963.[Medline]

Chary, V. K., De La Fuente, J. L., Liras, P. & Martin, J. F. (1997). amy as a reporter gene for promoter activity in Nocardia lactamdurans: comparison of promoters of the cephamycin cluster. Appl Environ Microbiol 63, 2977-2982.[Abstract]

Cundliffe, E. (1992). Self-protection mechanisms in antibiotic producers. Ciba Found Symp 171, 208-214.

Díez, B., Mellado, E., Rodríguez, M., Fouces, R. & Barredo, J. L. (1997). Recombinant microorganisms for industrial production of antibiotics. Biotechnol Bioeng 55, 216-226.

Elson, S. W., Baggaley, K. H., Gillett, J., Holland, S., Nicholson, N. H., Sime, J. T. & Woroniecki, S. R. (1987). Isolation of two novel intracellular ß-lactams and a novel dioxygenase cyclising enzyme from Streptomyces clavuligerus. J Chem Soc Chem Commun 1987, 1736-1738.

Elson, W. E., Gillett, J., Nicholson, N. H. & Tyler, J. W. (1988). N-Acyl derivatives of clavaminic acid produced by a mutant of Streptomyces clavuligerus. J Chem Soc Chem Commun 1988, 979-980.

Fedorenko, V. A., Lomovskaia, N. D., Chinenova, T. A. & Emel’ianova, L. K. (1985). Genetic control of actinomycetes resistance to antibiotics. Mol Gen Mikrobiol Virusol 3, 3-14.[Medline]

Fouces, R., Mellado, E., Díez, B. & Barredo, J. L. (1999). The tylosin biosynthetic cluster from Streptomyces fradiae: genetic organization of the left region. Microbiology 145, 855-868.[Abstract]

García-Domínguez, M., Martín, J. F., Mahron, B., Demain, A. L. & Liras, P. (1987). Efficient plasmid transformation of the beta-lactam producer Streptomyces clavuligerus. Appl Environ Microbiol 53, 1376-1381.[Medline]

Hodgson, J. E., Fosberry, A. P., Rawlinson, N. S., Ross, H. N., Neal, R. J., Arnell, J. C., Earl, A. J. & Lawlor, E. J. (1995). Clavulanic acid biosynthesis in Streptomyces clavuligerus: gene cloning and characterization. Gene 166, 49-55.[Medline]

Hopwood, D. A., Bibb, M. J., Chater, K. F. & 7 other authors (1985). Genetic Manipulation of Streptomyces: a Laboratory Manual. Norwich, UK: John Innes Foundation.

Jensen, S. E. & Paradkar, A. S. (1999). Biosynthesis and molecular genetics of clavulanic acid. Antonie Leeuwenhoek 75, 125-133.[Medline]

Jensen, S. E., Alexander, D. C., Paradkar, A. S. & Aidoo, K. A. (1993). Extending the ß-lactam biosynthetic gene cluster in Streptomyces clavuligerus. In Industrial Microorganisms: Basic and Applied Molecular Genetics , pp. 169-176. Edited by R. H. Baltz, G. D. Hageman & P. L. Skatrud. Washington, DC:American Society for Microbiology.

Jensen, S. E., Elder, K. J., Aidoo, K. A. & Paradkar, A. S. (2000). Enzymes catalyzing the early steps of clavulanic acid biosynthesis are encoded by two sets of paralogous genes in Streptomyces clavuligerus. Antimicrob Agents Chemother 44, 720-726.[Abstract/Free Full Text]

Joe, Y. A. & Goo, Y. M. (1998). Kanamycin acetyltransferase gene from kanamycin-producing Streptomyces kanamyceticus IFO 13414. Arch Pharm Res 21, 470-474.[Medline]

Kening, M. & Reading, C. (1979). Holomycin and an antibiotic (MM 19290) related to tunicamycin, metabolites of Streptomyces clavuligerus. J Antibiot 32, 549-554.[Medline]

Khaleeli, N., Li, R. & Townsend, C. A. (1999). Origin of the ß-lactam carbons in clavulanic acid from an usual thiamine pyrophosphate-mediated reaction. J Am Chem Soc 121, 9223-9224.

Li, R., Khaleeli, N. & Townsend, C. A. (2000). Expansion of the clavulanic acid gene cluster: identification and in vivo functional analysis of three new genes required for biosynthesis of clavulanic acid by Streptomyces clavuligerus. J Bacteriol 182, 4087-4095.[Abstract/Free Full Text]

Liras, P. (1999). Biosynthesis and molecular genetics of cephamycins. Cephamycins produced by actinomycetes. Antonie Leeuwenhoek 75, 109-124.

Liras, P. & Rodríguez-García, A. (2000). Clavulanic acid, a ß-lactamase inhibitor: biosynthesis and molecular genetics. Appl Microbiol Biotechnol 54, 467-475.[Medline]

Marsh, E. N., Chang, M. D.-T. & Townsend, C. A. (1992). Two isoenzymes of clavaminate synthase central to clavulanic acid formation: cloning and sequencing of both genes from Streptomyces clavuligerus. Biochemistry 31, 12648-12657.[Medline]

Mathews, D. H., Sabina, J., Zuker, M. & Turner, D. H. (1999). Expanded sequence dependence of thermodynamic parameters improves prediction of RNA secondary structure. J Mol Biol 288, 911-940.[Medline]

Miller, M. T., Bachmann, B. O., Townsend, C. A. & Rosenzweig, A. C. (2001). Structure of beta-lactam synthetase reveals how to synthesize antibiotics instead of asparagine. Nat Struct Biol 8, 684-689.[Medline]

Mosher, R. H., Paradkar, A. S., Anders, C., Barton, B. & Jensen, S. E. (1999). Genes specific for the biosynthesis of clavam metabolites antipodal to clavulanic acid are clustered with the gene for clavaminate synthase 1 in Streptomyces clavuligerus. Antimicrob Agents Chemother 43, 1215-1224.[Abstract/Free Full Text]

Murray, I. A., Gil, J. A., Hopwood, D. A. & Shaw, W. V. (1989). Nucleotide sequence of the chloramphenicol acetyltransferase gene of Streptomyces acrimycini. Gene 85, 283-291.[Medline]

Nagarajan, R., Boeck, L. D., Gorman, M., Hamill, R. L., Higgens, C. E., Hoehn, M. M., Stark, W. M. & Whitney, J. G. (1971). Beta-lactam antibiotics from Streptomyces. J Am Chem Soc 93, 2308-2310.[Medline]

Nicholson, N. H., Baggaley, K. H., Cassels, R., Davison, M., Elson, S. W., Fulston, M., Tyler, J. W. & Woroniecki, S. R. (1994). Evidence that the immediate biosynthetic precursor of clavulanic acid is its N-aldehyde analogue. J Chem Soc Chem Commun 1994, 1281-1282.

O’Keefe, D. P., Gibson, K. J., Emptage, M. H., Lenstra, R., Romesser, J. A., Litle, P. J. & Omer, C. A. (1991). Ferredoxins from two sulfonylurea herbicide monooxygenases systems in Streptomyces griseolus. Biochemistry 30, 447-455.[Medline]

Paradkar, A. S., Jensen, S. E. & Mosher, R. H. (1996a). Comparative genetics and molecular biology of ß-lactam biosynthesis. In Biotechnology of Antibiotics , pp. 241-277. Edited by W. R. Strohl. New York:Marcel Dekker.

Paradkar, A. S., Aidoo, K. A., Wong, A. & Jensen, S. E. (1996b). Molecular analysis of a beta-lactam resistance gene encoded within the cephamycin gene cluster of Streptomyces clavuligerus. J Bacteriol 178, 6266-6274.[Abstract]

Paradkar, A. S., Aidoo, K. A. & Jensen, S. E. (1998). A pathway-specific transcriptional activator regulates late steps of clavulanic acid biosynthesis in Streptomyces clavuligerus. Mol Microbiol 27, 831-843.[Medline]

Pérez-Llarena, F. J., Rodriguez-Garcia, A., Enguita, F. J., Martin, J. F. & Liras, P. (1998). The pcd gene encoding piperidine-6-carboxylate dehydrogenase involved in biosynthesis of alpha-aminoadipic acid is located in the cephamycin cluster of Streptomyces clavuligerus. J Bacteriol 180, 4753-4756.[Abstract/Free Full Text]

Pérez-Redondo, R., Rodríguez-García, A., Martín, J. F. & Liras, P. (1998). The claR gene of Streptomyces clavuligerus, encoding a LysR-type regulatory protein controlling clavulanic acid biosynthesis, is linked to the clavulanate-9-aldehyde reductase (car) gene. Gene 12, 311-321.

Pérez-Redondo, R., Rodríguez-García, A., Martín, J. F. & Liras, P. (1999). Deletion of the pyc gene blocks clavulanic acid biosynthesis except in glycerol-containing medium: evidence for two different genes in formation of the C3 unit. J Bacteriol 181, 6922-6928.[Abstract/Free Full Text]

Reading, C. & Cole, M. (1977). Clavulanic acid: a beta-lactamase-inhibiting beta-lactam from Streptomyces clavuligerus. Antimicrob Agents Chemother 11, 852-857.[Medline]

Romero, J., Liras, P. & Martín, J. F. (1984). Dissociation of the cephamycin and clavulanic acid biosynthesis in Streptomyces clavuligerus. Appl Microbiol Biotechnol 20, 318-325.

Sambrook, J., Fritsch, E. F. & Maniatis, T. (1989). Molecular Cloning: a Laboratory Manual, 2nd edn, Cold Spring Harbor, NY: Cold Spring Harbor Laboratory.

Wu, T.-K., Busby, R. W., Houston, T. A., McIlwaine, D. B., Egan, L. A. & Townsend, G. A. (1995). Identification, cloning, sequencing, and overexpression of the gene encoding proclavaminate amidino hydrolase and characterization of protein function in clavulanic acid biosynthesis. J Bacteriol 177, 3714-3720.[Abstract]

Zhu, C. B., Sunada, A., Ishikawa, J., Ikeda, Y., Kondo, S. & Hotta, K. (1999). Role of aminoglycoside 6'-acetyltransferase in a novel multiple aminoglycoside resistance of an actinomycete strain #8: inactivation of aminoglycosides with 6'-amino group except arbekacin and neomycin. J Antibiot 52, 889-894.[Medline]

Received 14 August 2001; revised 12 November 2001; accepted 7 January 2002.