p-Aminobenzoic acid and chloramphenicol biosynthesis in Streptomyces venezuelae: gene sets for a key enzyme, 4-amino-4-deoxychorismate synthase

Z. Changa,1, Y. Suna,1, J. Heb,1 and L. C. Vining1

Department of Biology, Dalhousie University, Halifax, Nova Scotia, CanadaB3H 4J11

Author for correspondence: L. C. Vining. Tel: +1 902 494 2040. Fax: +1 902 494 3736. e-mail: Leo.Vining{at}Dal.Ca


   ABSTRACT
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES
 
Amplification of sequences from Streptomyces venezuelae ISP5230 genomic DNA using PCR with primers based on conserved prokaryotic pabB sequences gave two main products. One matched pabAB, a locus previously identified in S. venezuelae. The second closely resembled the conserved pabB sequence consensus and hybridized with a 3·8 kb NcoI fragment of S. venezuelae ISP5230 genomic DNA. Cloning and sequence analysis of the 3·8 kb fragment detected three ORFs, and their deduced amino acid sequences were used in BLAST searches of the GenBank database. The ORF1 product was similar to PabB in other bacteria and to the PabB domain encoded by S. venezuelae pabAB. The ORF2 product resembled PabA of other bacteria. ORF3 was incomplete; its deduced partial amino acid sequence placed it in the MocR group of GntR-type transcriptional regulators. Introducing vectors containing the 3·8 kb NcoI fragment of S. venezuelae DNA into pabA and pabB mutants of Escherichia coli, or into the Streptomyces lividans pab mutant JG10, enhanced sulfanilamide resistance in the host strains. The increased resistance was attributed to expression of the pair of discrete translationally coupled p-aminobenzoic acid biosynthesis genes (designated pabB/pabA) cloned in the 3·8 kb fragment. These represent a second set of genes encoding 4-amino-4-deoxychorismate synthase in S. venezuelae ISP5230. In contrast to the fused pabAB set previously isolated from this species, they do not participate in chloramphenicol biosynthesis, but like pabAB they can be disrupted without affecting growth on minimal medium. The gene disruption results suggest that S. venezuelae may have a third set of genes encoding PABA synthase.

Keywords: p-aminobenzoate synthase genes, primary and secondary metabolism

Abbreviations: ADC, 4-amino-4-deoxychorismic acid; PABA, p-aminobenzoic acid; PAPA, p-aminophenylalanine; Am, apramycin; Cm, chloramphenicol; Ts, thiostrepton; Vio, viomycin

The GenBank accession number for the sequence reported in this paper is AF189258.

a Present address: Microbiology Department, University of Minnesota, 1030 Mayo Building, 420 Delaware Street SE, Minneapolis, MN 55455, USA.

b Present address: Pharmaceutics Department, Pharmacy University of Shenyang, 103 Wenhua Rd, Shenyang, P.R. China.


   INTRODUCTION
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES
 
In their biosynthetic origins, p-aminobenzoic acid (PABA) and chloramphenicol (Cm) share the early steps of the shikimate pathway leading to chorismic acid. At the chorismate branch point, separate routes have evolved to supply the proteinogenic aromatic amino acids (phenylalanine, tyrosine, tryptophan) and the folic acid cofactor component PABA. Although the p-nitrophenylserinol moiety of Cm bears obvious similarity to phenylalanine and tyrosine, it is not derived metabolically from them, but is synthesized de novo in the branch of the shikimate pathway that leads to PABA. The initial reaction, catalysed by an enzyme complex formed from the association of pabA and pabB gene products, converts chorismic acid to 4-amino-4-deoxychorismic acid (ADC). In turn, ADC is a metabolic branch point, leading either to PABA via a lyase reaction catalysed by the pabC product (Green et al., 1992 ), or to 4-amino-4-deoxyprephenic acid by a mutase-catalysed reaction (Teng et al., 1985 ; Fig. 1). Conversion of 4-amino-4-deoxyprephenic acid to p-aminophenylpyruvic acid gives Cm via the known precursor p-aminophenylalanine (PAPA) (Siddiquellah et al., 1967 ; Jones & Vining, 1976 ). An alternative series of reactions from ADC via 4-amino-4-deoxyarogenic acid to PAPA and Cm is also feasible. The dichloroacetyl moiety of Cm is contributed by a still undetermined pathway that does not involve shikimic acid, but creates intermediates derived ultimately from glucose (Munro et al., 1975 ).



View larger version (22K):
[in this window]
[in a new window]
 
Fig. 1. Biosynthesis of primary and secondary aromatic metabolites in the shikimic acid pathway.

 
Because genes associated with the biosynthesis of antibiotics in streptomycetes are commonly clustered on the chromosome of the producing organism, participation of ADC in Cm biosynthesis suggested that a cluster of Cm biosynthetic genes might be located in the vicinity of ADC synthase genes involved in PABA formation. Cloning PABA synthase genes could then offer a potential route to cloning a Cm biosynthesis gene cluster. Consistent with these expectations, Brown et al. (1996) demonstrated that Cm production in Streptomyces venezuelae ISP5230 was severely impaired when an indigenous gene that complemented a Streptomyces lividans pab mutant was replaced with a disrupted allele. The locus has not yet been genetically linked to the mapped Cm biosynthesis gene cluster in S. venezuelae ISP5230 (Doull et al., 1986 ; Vats et al., 1987 ), but the DNA that complemented the S. lividans pab lesion has been shown by cloning and sequence analysis to be a gene consisting of fused pabA and pabB domains. In their deduced amino acid sequences, these domains closely resemble the products of discrete pabA and pabB genes in a variety of bacteria (Brown et al., 1996 ). A strain of S. venezuelae in which the fused pabAB had been disrupted showed increased sensitivity to sulfanilamide, a metabolic inhibitor that competes with PABA for the active site of a key enzyme required in the synthesis of folate cofactors (Swerdberg & Skold, 1980 ). Sulfanilamide is normally included in culture media to counteract contaminants that obscure PABA assays based on a growth response. Because the interaction is competitive, the increased sensitivity to sulfanilamide observed after pabAB disruption implied reduced PABA synthesis. However, loss of pabAB did not completely prevent growth of the S. venezuelae strain in the absence of a PABA supplement. The results suggested that the ADC synthase encoded by pabAB acts predominantly in the secondary metabolic pathway and that S. venezuelae ISP5230 might contain another set of genes expressing ADC synthase to form PABA for primary metabolism.


   METHODS
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES
 
Bacterial strains, plasmids and vectors.
These are listed in Table 1.


View this table:
[in this window]
[in a new window]
 
Table 1. Strains and vectors

 
Culture conditions.
Escherichia coli was grown as described by Sambrook et al. (1989) . For Micrococcus luteus the medium used was GNY (Malik & Vining, 1970 ). Streptomyces venezuelae was maintained on MYM agar (Stuttard, 1982 ) supplemented with antibiotics as needed to select strains with plasmids; Streptomyces lividans JG10 was maintained on K1 agar (Aidoo, 1989 ), except when TOY agar (20 g each of Heinz tomato paste, Pablum and agar, and 2·5 g yeast extract in 1 litre water, pH 6·8), was required to enhance sporulation. For long-term maintenance of streptomycetes, spores in 20% (v/v) glycerol were stored at -70 °C. Growth and sensitivity to sulfanilamide of streptomycete strains was assessed on minimal medium (MM; Hopwood, 1967 ). Sensitivity of E. coli strains to sulfanilamide was determined on minimal agar (MA; Clowes & Hayes, 1968 ) supplemented with vitamin-free Casamino acids (0·2 mg ml-1) and thiamine (1·0 mg ml-1). Sulfanilamide inhibition of E. coli and streptomycetes was measured by plating cells or spores at densities that gave well-separated colonies with no cross-feeding.

Transformation.
Competent E. coli cells were prepared and transformed as described by Sambrook et al. (1989) . Protoplasts of S. lividans were prepared, transformed and regenerated by standard methods for streptomycetes (Hopwood et al., 1985 ); for S. venezuelae these methods were modified (Aidoo et al., 1990) . Procedures for conjugal transfer of plasmids from E. coli to streptomycetes were essentially those of Flett et al. (1997) .

DNA manipulations.
{lambda} phage DNA, E. coli plasmid DNA and single-stranded DNA templates for sequencing were prepared as described by Sambrook et al. (1989) . Genomic and plasmid DNA were isolated from streptomycetes by the methods of Hopwood et al. (1985) . For rapid screening of plasmids in E. coli the Sekar (1987) procedure was adopted. DNA fragments excised in agarose gel slices were extracted with the QIAEX II kit (Qiagen).

Amplification of S. venezuelae chromosomal DNA by PCR.
The primers used were 5'-ATGATCGTGGACCTSGACCGSAACGAC-3' (sun-3) and 5'-CTGCGGCTCSAGCTCGTCGATGATCTCCAT-3' (sun-4). The reaction mixture (total volume 100 µl) consisted of: dNTPs, 10 µl (44x1·25 mM); primer sun-3, 1 µl (268 pmol); primer sun-4, 1 µl (182 pmol); chromosomal DNA, 100 ng; 10xPromega PCR buffer, 10 µl; and water to 100 µl. It was aliquotted (20 µl) in thin-walled 0·5 ml microcentrifuge tubes and layered with mineral oil (ca 20 µl). Each aliquot was mixed with 1 U Taq DNA polymerase and heated to 80 °C, then thermal cycling was started (model PTC-100; MJ Research). The programme for the first 7 cycles was: 96 °C for 1 min (denaturing), 67 °C for 1 min (annealing) and 72 °C for 1 min (extension). For the next 30 cycles denaturing and annealing times were each reduced to 45 s. Amplified S. venezuelae DNA was inserted into the SmaI site of pUC18 (SureClone Kit; Pharmacia Biotech) and sequenced by the dideoxy chain-termination method (Sanger et al., 1977 ).

Construction of an S. venezuelae genomic library in {lambda} GEM-11.
Genomic DNA (50–100 µg) from S. venezuelae was partially digested at 37 °C in a 100 µl reaction mixture containing 3 U Sau3AI. The incubation time was adjusted to optimize the yield of 9–23 kb fragments; digestion was terminated with EDTA (20 mM). The DNA recovered was fractionated by sucrose-gradient centrifugation and 3'-CTAG-5' overhangs of the 9–23 kb fragments were partially filled-in by incubation with the Klenow fragment of DNA polymerase I in the presence of dGTP and dATP. {lambda} GEM-11 arms digested with XhoI (giving 3'-AGCT-5' overhangs) were partially filled in with dTTP and dCTP; the modified fragments and arms, after incubation with T4 DNA ligase and a {lambda} phage packaging system (Promega), were used to infect E. coli LE392. The phage library was amplified as described by Sambrook et al. (1989) and stored at -70 °C in SM buffer containing 7% (v/v) DMSO.

Construction of disruption vectors.
The vector used to disrupt pabAB has been described by Brown et al. (1996) . To disrupt the putative pabB in pJV305, a 1·5 kb DNA fragment containing an apramycin resistance (AmR) gene (Paradkar & Jensen, 1995 ) was inserted into the Eco72I site of ORF1 to give pJV307. The disrupted ORF1 in pJV307 was then recloned as pJV323 in pHJL400. To construct a disruption vector containing the viomycin resistance (VioR) gene, the 2·0 kb Streptomyces vinaceus DNA fragment containing the gene (Thompson et al., 1982 ) was excised from pJV230 (Chang, 1999 ) and introduced into pJV305 at the Eco72I site in ORF1, giving pJV308. The 5·8 kb BamHI fragment containing ORF1 disrupted with the VioR gene was recloned from pJV308 into pHJL400, giving pJV324. Since protoplast procedures failed to give VioR transformants of VS629 with pJV324, the 5·8 kb BamHI insert of pJV324 was subcloned in the conjugal vector pJV326 to give pJV325. Conjugal transfer of pJV325 from E. coli into S. venezuelae VS629 using methodology developed by Mazodier et al. (1989) and Flett et al. (1997) yielded single- and double-crossover mutants VS1003 and VS1004.

DNA sequencing and analysis.
The 3·8 kb NcoI fragment of S. venezuelae ISP5230 DNA from recombinant phage YSB1 was subcloned (both orientations) into pBluescript II SK(+), giving pJV305 and pJV306. Nested overlapping deletions were introduced into the inserts by the Henikoff (1984) procedure and the DNA was used to transform E. coli DH5{alpha}. Inserts in plasmid DNA extracted from the transformants were sequenced (ABI Prism model 373 DNA Sequencer). To detect ORFs by codon third position mol% G+C bias and codon usage, the CODONPREFERENCE programme (GCG Wisconsin Package, version 9.0; Devereux et al., 1984 ) was used. Restriction enzyme sites were located with GeneRunner (Hastings Software) and similarities between derived amino acid sequences and proteins in GenBank were assessed from BLAST searches (Altschul et al., 1997 ). Sequences were aligned using CLUSTAL W (Higgins et al., 1996 ); the MacVector software of the Oxford Molecular Group was used for phylogenetic analysis.

Hybridization.
For phage library screening (Hopwood et al., 1985) and Southern analyses (Southern, 1975 ), DNA samples bound to nylon membranes were incubated with 32P-labelled probes at 65 °C overnight in hybridization solution (Sambrook et al., 1989 ). The membranes were washed at room temperature with 2xSSC, and then twice at 65 °C with 0·1xSSC (each solution containing 0·1% SDS).

Cm production.
Strains of S. venezuelae were grown and analysed for Cm production by methods described by Brown et al. (1996) . The Cm content of filtered broths was determined by HPLC.


   RESULTS
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES
 
Cloning a second ADC synthase gene set from S. venezuelae
To isolate S. venezuelae ADC synthase homologues different from the pabAB cloned by Brown et al. (1996) , candidate genomic sequences were amplified by PCR using primers chosen from consensus regions in an alignment of GenBank PABA synthase genes. The pair of oligonucleotides, sun-3 and sun-4 (Fig. 2a), was designed from a streptomycete codon usage table (Wright & Bibb, 1992 ) to prime the amplification of 250 bp of S. venezuelae ISP5230 genomic DNA lying between two such consensus regions. Conditions for the PCR were optimized to obtain a dominant 250 bp product. Amplified genomic DNA was cloned in pUC18 and used to transform E. coli DH5{alpha}. Plasmid DNA was extracted from 21 individual transformants purified by plating on LB agar and the cloned insrts were sequenced. CLUSTAL W alignments showed four distinguishable patterns (YH1, YH2, YH3 and YH5; Fig. 2b) among the nucleotide sequences of the inserts. A BLASTX search of GenBank indicated that the deduced amino acid sequence of the type-YH1 amplicon matched the ADC synthase encoded by pabAB from S. venezuelae; a CLUSTAL W alignment confirmed the identity of the two sequences. With YH5 a BLASTX search detected no significant similarities between deduced amino acid sequences and GenBank proteins, but YH2 showed 30–45% similarity to bacterial anthranilate synthases, while the deduced amino acid sequence of the YH3 amplicon gave much closer matches (66–78% similarity) to PABA synthases. By using [32P]dCTP-labelled YH3 to probe an S. venezuelae ISP5230 genomic library constructed in {lambda} GEM-11, a hybridizing recombinant phage (YSB1) was recovered. Restriction mapping and Southern hybridization of the DNA insert in YSB1 (Fig. 3) located a 3·82 kb NcoI fragment of DNA that hybridized with the YH3 probe. The fragment was excised, blunt-ended and subcloned (both orientations) in the EcoRV site of pBluescript II SK(+) to give pJV305 and pJV306 (Fig. 4). Both strands of the YSB1 insert were sequenced and the sequence was deposited in GenBank (accession no. AF189258). The YH3 probe sequence was present at nt 1285–1496 in the cloned fragment.



View larger version (54K):
[in this window]
[in a new window]
 
Fig. 2. (a) Alignment of deduced amino acid sequences for pabB from the GenBank database. The amino acids boxed and underlined with arrows are the conserved sequences used to design oligonucleotide primers sun-3 and sun-4. Identical and similar amino acids are indicated below the alignment by asterisks and periods, respectively. (b) Nucleotide sequences of fragments amplified by PCR from an S. venezuelae ISP5230 genomic DNA template using sun-3 and sun-4 as primers. Primer sequences recognized in the amplicons are boxed. Asterisks and periods below the alignment signify nucleotides present in all or most sequences, respectively.

 


View larger version (9K):
[in this window]
[in a new window]
 
Fig. 3. Top: restriction map of S. venezuelae ISP5230 DNA cloned in {lambda} phage YSB1. # indicates that not all SacI sites are shown. The expanded region below shows the 3·82 kb NcoI–NcoI fragment blunt-ended with S1 nuclease and subcloned (alternative orientations) in the EcoRV site of pBluescript II SK(+) as pJV305 and pJV306 (the orientation shown); it also locates ORF1, ORF2 and the incomplete ORF3 in the cloned 3·82 kb DNA fragment.

 


View larger version (27K):
[in this window]
[in a new window]
 
Fig. 4. Plasmids constructed for this study. Solid filled regions represent S. venezuelae DNA from YSB1. The two regions of ORF1 separated by disruption with apr or vph at Eco72I are distinguished in pJV323 and pJV325 as ORF1- and -ORF1. amp, apr, tsr and vph designate genes for ampicillin, Am, Ts and Vio resistance.

 
Sequence analysis of the cloned 3·82 kb NcoI fragment
Analysis of the pJV306 insert sequence located ORF1 between nt 668 and 1843 in frame +2; a second ORF (ORF2) was present between nt 1840 and 2472 in frame +1. (see Fig. 3). Both ORF1 and ORF2 showed high G+C content (74 and 73·9 mol%, respectively) and a strong G+C bias (>94%) in the third position of codons, typical of streptomycete genes. The two ORFs, which are transcribed in the same direction but in different reading frames, overlap by 3 nt. An incomplete ORF (ORF3) was located from nt 2289 to the end of the sequenced region in frame -3. The ORF3 sequence overlapped one end of ORF2 (in frame +1) by 183 nt, but was transcribed in the opposite direction. Using BLASTX to search GenBank for proteins with sequences similar to the deduced amino acid sequences of ORF1 and ORF2 showed 30–40% identity between the ORF1 product and PabB or TrpE from a variety of organisms; similar close matches were found between the ORF2 product and PabA or TrpG proteins. The ORF1 product resembled (40% identity, 53% similarity) PabB of Bacillus subtilis and the PabB domain in products of the pabAB-type genes in Streptomyces griseus and Streptomyces pristinae-spiralis (35% identity, 51–52% similarity). CLUSTAL W alignments confirmed that the ORF1 and ORF2 products corresponded to the C- and N-terminal regions, respectively, in pabAB products (Fig. 5).



View larger version (79K):
[in this window]
[in a new window]
 
Fig. 5. Alignment with CLUSTAL W of the deduced amino acid sequences of the pJV306 ORF1 (pabB) product (top line) and ORF2 (pabA) product (bottom line) with those of the pabAB product (centre line). Identical amino acids in comparisons of either the PabA and PabAB, or the PabB and PabAB sequences are highlighted by reverse printing; similar amino acids are outlined.

 
In a BLASTP search of proteins encoded by the S. coelicolor A3(2) genome (http://www.sanger.ac.uk/Projects/S_coelicolor) using translated pabB from S. venezuelae as the query sequence, the two closest matches were the anthranilate synthase component I products of trpE2 in cosmid E8 (46% identical, 59% similar amino acids) and trpE3 in cosmid 4G6 (39% identical, 54% similar amino acids). The ORF1 and ORF2 products were also related (41–43% identity, 59–60% similarity) to the trpE and trpG products, respectively, of several plants (e.g. Arabidopsis thaliana, Ruta sp.), as well as to trp products of the extreme bacterial thermophile Aquifex aeolicus (ca 37% identity, 51% similarity). A BLASTX search of the GenBank database with the partial ORF3 sequence showed no similarity to shikimate pathway gene products, but instead a strong resemblance to the sequence of MocR, a GntR-type transcriptional regulator in Rhizopus species.

Expression of genes for PABA biosynthesis
The 3·82 kb NcoI fragment of S. venezuelae ISP5230 DNA excised from {lambda} YHB1 was blunt-ended and subcloned in the EcoRV site of pBluescript II SK(+) as pJV305 and pJV306 (alternative orientations; see Fig. 3). To determine whether their inserts expressed enzymes for PABA biosynthesis, each of these plasmids was used to transform the E. coli PABA auxotrophs AB3292 and AB3295, which are defective in pabA and pabB functions, respectively. Plasmid pJV305 contained ORFs 1 and 2 oppositely oriented from the vector lacZ, whereas in pJV306, ORFs 1 and 2 were in the same orientation as lacZ (see Fig. 4). To assess whether PABA synthase genes were expressed, the growth response of the E. coli pab mutants and transformants on medium supplemented with sulfanilamide was determined. Expression was inferred from enhanced sulfanilamide resistance. The higher sulfanilamide resistance in mutants transformed with pJV306 than in those transformed with pJV305 (Table 2) implied that the cloned DNA was transcribed more efficiently when ORF1/ORF2 and lacZ were oriented in the same direction. This would be consistent with expression of the cloned E. coli insert from the vector lacZ promoter and with little or no expression from a native promoter. Comparable results have been reported for S. griseus pabAB, which was shown by Gil & Hopwood (1983) to be expressed in E. coli only from the vector promoter.


View this table:
[in this window]
[in a new window]
 
Table 2. Effect of sulfanilamide on growth of E. coli pab auxotrophs alone or transformed with pJV305 and pJV306

 
When the S. venezuelae DNA in pJV305 was subcloned in both orientations as a BamHI fragment in the bifunctional E. coli/Streptomyces vector pJV326, the resulting plasmids (pJV327 and pJV328; see Fig. 4) differed in their ability to confer sulfanilamide resistance on S. lividans PABA auxotrophic strain JG10. Cultures transformed with pJV328, in which the cloned fragment and vector promoter were oriented in the same direction, grew in a medium containing 10 µg sulfanilamide ml-1, whereas cultures transformed with pJV327 (where the insert was oppositely oriented) were, like the auxotrophic parent, completely inhibited by 2 µg sulfanilamide ml-1 (Table 3). Again the higher resistance to sulfanilamide of the pJV328 transformant is attributed to expression from the vector promoter of the PABA synthase genes cloned from S. venezuelae. Because the results in both E. coli and S. lividans indicate that the 3·82 kb ADC synthase gene set cloned from S. venezuelae ISP5230 can express enzymes supporting PABA biosynthesis, we propose to designate the relevant gene pair pabB/pabA.


View this table:
[in this window]
[in a new window]
 
Table 3. Effect of sulfanilamide on growth of Streptomyces lividans JG10 auxotroph transformed with pJV327 or pJV328

 
Disruption of the second ADC gene set in S. venezuelae
Transformation of wild-type S. venezuelae protoplasts with the ORF1 disruption vector pJV323 (see Fig. 4) yielded the AmR and thiostrepton-resistant (TsR) strain VS1001 containing both the free and integrated plasmid. Plasmid derivatives of pHJL400 are segregationally unstable in streptomycetes (Larson & Hershberger, 1996 ) and are lost during propagation without selection for TsR; thus cultivation of VS1001 on unsupplemented MYM agar yielded a double-crossover transformant (VS1002), the identity of which was confirmed by Southern hybridization (Fig. 6a). Strain VS1002, in which ORF1 had been insertionally inactivated, was slightly more sensitive to sulfanilamide than S. venezuelae ISP5230 (Table 4), but produced a wild-type titre of Cm.



View larger version (73K):
[in this window]
[in a new window]
 
Fig. 6. (a) Southern hybridization of S. venezuelae genomic DNA from strains ISP5230 (lane 1) and VS1002 (lane 2) digested with NcoI and probed with the 3·8 kb BamHI fragment of pJV305. The increased size of the hybridizing fragment in VS1002 is accounted for by integration of the 1·5 kb apr (see Fig. 4). (b) Southern hybridization of S. venezuelae genomic DNA from strains ISP5230 (lane 1) and VS1004 (lane 2) digested with MluI and probed with the 5·8 kb BamHI fragment of pJV325. The increased size of the 1·6 kb hybridizing fragment in VS1004 is accounted for by integration of the 2·0 kb vph (see Fig. 4).

 

View this table:
[in this window]
[in a new window]
 
Table 4. Effect of sulfanilamide on growth of Streptomyces venezuelae ISP5230 and strains disrupted in ADC synthase genes

 
To inactivate both pabAB and the putative pabB (ORF1) in the same wild-type strain by the methods used to disrupt the genes individually (see above), it was necessary to construct a conjugal disruption vector carrying a different resistance gene. Transformation of the pabAB-disrupted S. venezuelae strain VS629 (Brown et al., 1996 ) with the new vector (pJV325; see Fig. 4) gave single-crossover transformants (VS1003) from which a double-crossover strain (VS1004) resistant to both Am and Vio, but sensitive to Ts, was isolated. Disruption of ORF1 in VS629 by insertion of the VioR gene, giving strain VS1004, was confirmed by Southern hybridization (Fig. 6b). When the growth of S. venezuelae strains ISP5230, VS629, VS1002 and VS1004 was compared in minimal medium supplemented with increasing concentrations of sulfanilamide (see Table 4), the difference in sensitivity to sulfanilamide between ORF1-disrupted strains and their parents was marginal (compare strains VS1004 and VS629). It was also noted that the Cm titre in strain VS1004 was not significantly different from that in strain VS629.


   DISCUSSION
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES
 
Functions of ADC gene sets in S. venezuelae
The loss of over 97% in Cm production when the native pabAB was replaced with a disrupted allele, and restoration of production when disrupted cultures were supplemented with PAPA have provided strong evidence that this ADC synthase gene set encodes an enzyme converting chorismate to ADC as a precursor of PAPA, and ultimately of Cm (Brown et al., 1996 ; Halpern, 1997 ). Since pabAB-disrupted strains retained their prototrophic phenotype and were only slightly more sensitive than the wild-type to inhibition by sulfanilamide, specific conversion of ADC formed by the pabAB product was inferred and S. venezuelae ISP5230 was postulated to contain a second set of ADC synthase genes needed to synthesize PABA for folate cofactors used in primary metabolism (Brown et al., 1996 ). In exploring S. venezuelae ISP5230 for the presence of genes encoding PABA synthase, we have discovered a new set of ADC synthase genes, cloned it in pJV305 and investigated its properties. This set differs in organization from pabAB and in this respect is more like the PABA synthase genes in the fol operon of B. subtilis (Slock et al., 1990 ), which not only contains discrete genes, but also contains its pabB component upstream of its pabA equivalent (which in B. subtilis is the amphibolic trpG). A noteworthy difference from B. subtilis is the absence of pabC immediately downstream of the amphibolic trpG. The organization of both the pabAB and pabB/pabA sets of ADC synthase genes in S. venezuelae contrasts with that in Mycobacterium tuberculosis, E. coli and other bacteria where pabA and pabB are separated on the chromosome by up to 1500 kb.

The probable identity of ORF1 and ORF2 with pabB and pabA, respectively, is indicated by comparisons of their deduced amino acid sequences with proteins in the GenBank database and by the ability of the ORFs to complement pab mutants in E. coli and S. lividans. However, the absence of a decisive loss of sulfanilamide resistance in S. venezuelae ISP5230 when pabB is disrupted implies that this gene set is not, as anticipated, primarily associated with PABA biosynthesis. The conclusion is supported by the negligible change in sulfanilamide resistance following disruption of pabB in the pabAB mutant VS629 to yield a strain inactivated in both of its known ADC synthase genes. It is possible that the weak PABA synthase activity in VS629 is due, at least in part, to incomplete inactivation of pabAB, so the effect of disrupting pabB is masked by the host’s residual PabAB activity. Contributing to this might be the high normal level of pabAB expression implied by the exceptionally high resistance of S. venezuelae to sulfanilamide. Aidoo et al. (1990) postulated that this high sulfanilamide resistance reflects a large ADC pool associated with the level of pabAB expression needed to support Cm biosynthesis. Spillover from the pool in the wild-type, or residual activity in the pabAB mutant might outweigh a relatively weak contribution to PABA synthesis from pabB and obscure the response to pabB disruption. However, the presence in S. venezuelae ISP5230 of a third gene set involved in supplying most of the PABA needed to support growth must also be considered.

Anthranilate and ADC synthase component I genes of S. venezuelae ISP5230 and S. coelicolor A3(2)
Information currently available from the S. coelicolor A3(2) genome sequencing project (http://www.sanger.ac.uk/Projects/S_coelicolor) has located four genes encoding putative anthranilate synthase component I proteins on cosmids cloned from the chromosome: cosmid 6E10 contains trpE1, cosmid E8 contains trpE2, cosmid 4G6 contains trpE3 and cosmid L11 contains an undesignated putative trpE. No PABA synthase component I genes have yet been detected. In a BLASTP search of GenBank for sequence similarity to the translated TrpE1, the highest score was obtained with the anthranilate synthase of S. venezuelae ISP5230 associated with the primary metabolic pathway for tryptophan biosynthesis (Lin et al., 1998 ). This result is consistent with similarities between genes of the primary metabolic tryptophan biosynthetic pathways in S. venezuelae and S. coelicolor, and with the presence of another primary metabolic trp gene (trpD) on the overlapping cosmid 6G1 (Redenbach et al., 1996 ). Similar BLASTP searches of GenBank with the translated TrpE2 and TrpE3 sequences ranked only anthranilate synthases in the 30 highest scoring protein sequences. Evidence that trpE2 and trpE3 are tryptophan biosynthesis genes associated with secondary metabolism supports these search results: trpE2 is involved in biosynthesis of tryptophan residues for the lipopeptide calcium-dependent antibiotic of S. coelicolor A3(2) (Chong et al., 1998 ), while the presence of three other trp genes (trpA, trpB and trpC) distinct from the primary metabolic trp genes on the same cosmid (4G6) as trpE3 implies an involvement with secondary metabolic tryptophan biosynthesis activity. Therefore, three of the four genes predicted to encode anthranilate synthases appear to be unambiguously assigned, but assignment of the cosmid L11 gene as trpE is less certain. A BLASTP search of GenBank with the translated sequence of this gene ranked the PabB sequences of Mycobacterium tuberculosis, Xanthomonas albilineans and Pseudomonas aeruginosa ahead of TrpE sequences in order of similarity and also gave high scores to the PabBs of enteric bacteria and Vibrio cholerae. The gene on S. coelicolor cosmid L11 may, therefore, encode an ADC synthase rather than an anthranilate synthase.

In support of this conclusion, BLASTP searches of the S. coelicolor A3(2) genome with deduced amino acid sequences of the YH1, YH2 and YH3 PCR products amplified from the S. venezuelae chromosome showed that the cosmid L11 gene product resembled the YH1 and YH3 sequences more closely (63% similarity, 53% identity) than it resembled the YH2 sequence (53% similarity, 35% identity). Since the deduced YH1 and YH3 amino acid sequences matched PabAB and PabB, respectively, whereas the deduced YH2 sequence matched TrpE, the results indicate that the S. coelicolor gene in cosmid L11 is more likely to be pabB than trpE. In agreement with the assignments of trpE genes in the other S. coelicolor cosmids, the product of trpE3 in cosmid 4G6 closely resembled (75% similarity, 70% identity) the deduced amino acid sequence of the YH2 PCR product, indicating that YH2 is amplified from a trpE3 counterpart in S. venezuelae. The product of trpE2 in cosmid E8 was less similar (65% similar, 55% identical amino acids) to the YH2 PCR product, confirming that YH2 is unlikely to have been amplified from a trpE2 counterpart. Although these sequence comparisons suggested that the pabB cloned from S. venezuelae corresponds to the S. coelicolor A3(2) gene in cosmid L11, a preferential relationship could not be demonstrated through BLASTP matching of S. coelicolor A3(2) genomic proteins with the translated sequence of S. venezuelae pabB. The highest scoring sequence was anthranilate synthase component I encoded by trpE2 in cosmid E8 (59% similar, 46% identical amino acids), followed by the anthranilate synthase of trpE3 in cosmid 4G6 (54% similar, 39% identical amino acids), and then the product of the putative trpE in cosmid L11 (48% similar, 37% identical amino acids). However, similar rankings were obtained when the products of B. subtilis pabB or S. venezuelae pabAB were used to query the S. coelicolor A3(2) genomic proteins, suggesting that such sequence comparisons are not reliable indicators of functional similarity. The indication that the putative pabB/pabA gene set cloned from S. venezuelae encodes proteins with weak PABA synthase activity, but appreciable anthranilate activity is, nevertheless, noteworthy in that it could account for the observation (Lin et al., 1998 ) that disruption of the primary metabolic trpE(G) in S. venezuelae ISP5230 did not create an auxotrophic requirement for tryptophan. The analysis of S. coelicolor and S. venezuelae sequences presented above bears on the evidence from gene disruptions that implicates a third and so far undetected set of genes encoding PABA biosynthesis in S. venezuelae ISP5230 and suggests that this set may be related to the putative trpE cloned from S. coelicolor A3(2) in cosmid L11.

Phylogenetic analysis of genes in the pabB/trpE superfamily
The shared evolutionary origin of genes for PABA and anthranilate synthases, postulated from sequence data by Goncharoff & Nichols (1984) , was demonstrated further in the results of a CLUSTAL W alignment of 35 PabB and TrpE sequences. The dendrogram showing sequence relationships (Fig. 7) indicated a superfamily derived from an ancestral sequence that diverged to form three initial groups. One (Group A) contains six TrpEs, the second (Group B) contains two TrpEs and the third (Group C) has both TrpEs and PabBs. The product of pabB cloned from S. venezuelae ISP5230 is present in Group C; its sequence shows early divergence from subgroups containing either TrpE or other PabB sequences and it has no closely related family member of recent origin.



View larger version (30K):
[in this window]
[in a new window]
 
Fig. 7. Sequence relationships between 35 gene products in the PabB/TrpE superfamily. Sequences were aligned with CLUSTAL W in the MacVector software of the Oxford Molecular Group. Numbers in the dendrogram are distance measurements given as changes per residue. The sequences used in the alignment and their protein IDs are: Aquifex aeolicus PabB (AAC07173); Bacillus subtilis PabB (AAA22694), TrpE (NNBS1); Bacillus halodurans PabB (BAB03809); Campylobacter jejuni PabB (CAB73127); Clostridium thermocellum TrpE (P14953); Corynebacterium glutamicum TrpE (B24723); Escherichia coli PabB (BAB35944), TrpE (AAC74346); Haemophilus influenzae PabB (AAC22834), TrpE (AAC22033); Klebsiella aerogenes PabB (AAA88207); Lactococcus lactis PabB (AAK05412); Mycobacterium tuberculosis EntC (CAB08301), PabB (CAB08140), TrpE (CAB08903), TrpE2 (CAB03739); Pseudomonas aeruginosa PchA (S58229), TrpE (20580); Pseudomonas fluorescens PhzE (Q5 1791); Sinorhizobium meliloti TrpEG (AAA26370); Saccharomyces cerevisiae PabAB (P37254): Salmonella typhimurium PabB (AAA88618); Streptomyces coelicolor cosmid E8, TrpE2 (CAB385), cosmid L11, TrpE (CAB76066), cosmid 4C2, TrpE (CAC17508), cosmid 4G6 TrpE3 (CAB51435), cosmid 6E10, TrpE1 (CAB51965); Streptomyces griseus PabAB (P32438); Streptomyces pristinae-spiralis PabAB (AAC44866); Streptomyces venezuelae PabAB (AAB30312), PabB (AAF01062), TrpEG (AAC35404); Sulfolobus solfataricus TrpE (AAA7379); Thermus thermophilus TrpE (CAA30566).

 
In general, genes associated with secondary metabolic biosynthetic pathways diverge relatively early from homologues with roles in primary metabolism. The dendrogram implies the possible divergence of a cluster of streptomycete pabABs involved in antibiotic production from the Saccharomyces cerevisiae pabAB, which is dedicated to PABA biosynthesis (Edman et al., 1993 ). The entire pabAB cluster shares ancestry with well separated clusters consisting of one or more discrete primary metabolic pabBs and with trpE from S. coelicolor cosmid L11. Only a small number of antibiotic-producing streptomycetes have PABA synthase genes that exhibit a correlation between fusion of pabA/pabB domains and a role in secondary metabolism. A second example where Pab/Trp superfamily genes associated with secondary metabolism have diverged from primary metabolic homologues can be found in Pseudomonas fluorescens. Here the gene encoding anthranilate synthase (PhzE) for phenazine biosynthesis is related to anthranilate synthase genes for tryptophan biosynthesis in S. coelicolor and S. venezuelae. The dendrogram also supports the sequence similarity, noted in the preceding section, between S. coelicolor A3(2) TrpE in cosmid L11 and a PabB of M. tuberculosis. This protein appears to be the only gene product that might provide the PABA synthase activity required for primary metabolism in S. coelicolorA3(2), strengthening the possibility that the gene is distributed generally in streptomycetes and is indeed the third ADC synthase gene set in S. venezuelae ISP5230.


   ACKNOWLEDGEMENTS
 
We are grateful to Dr J. Gil, Oviedo University, for the culture of S. lividans JG10, to Dr A. S. Paradkar, University of Alberta, for the NcoI cassette containing the AmR gene, to Dr C. L. Hershberger, Eli Lilly, for the vector pHJL400, and to Dr D. J. MacNeil for a culture of E. coli ET 12567. We thank Dr M. Paget for advice on conjugal transfer of plasmids from E. coli to S. venezuelae This work was supported by the Natural Sciences and Engineering Research Council of Canada.


   REFERENCES
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES
 
Aidoo, D. A. (1989). Approaches to the cloning of genes for chloramphenicol biosynthesis in Streptomyces venezuelae ISP5230. PhD thesis, Dalhousie University, Halifax, NS, Canada.

Aidoo, D. A., Barrett, K. & Vining, L. C. (1990). Plasmid transformation of Streptomyces venezuelae: modified procedures used to introduce the genes for p-aminobenzoate synthetase. J Gen Microbiol 136, 657-662.[Medline]

Altschul, S. F., Madden, T. L., Schaffer, 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]

Brown, M. P., Aidoo, K. A. & Vining, L. C. (1996). A role for pabAB, a p-aminobenzoate synthase gene of Streptomyces venezuelae ISP5230, in chloramphenicol biosynthesis. Microbiology 142, 1345-1355.[Abstract]

Chang, Z. (1999). Genes for cysteine biosynthesis and metabolism in Streptomyces venezuelae ISP5230: cloning, sequencing, functional analysis and relevance to chloramphenicol biosynthesis. PhD thesis, Dalhousie University, Halifax, NS, Canada.

Chong, P. P., Podmore, S. M., Kieser, H. M., Redenbach, M., Turgay, K., Marahiel, M., Hopwood, D. A. & Smith, C. P. (1998). Physical identification of a chromosomal locus encoding biosynthetic genes for the lipopeptide calcium-dependent antibiotic (CDA) of Streptomyces coelicolor A3(2). Microbiology 144, 193-199.[Abstract]

Clowes, R. C. & Hayes, W. (1968). Experiments in Microbial Genetics. New York: Wiley.

Devereux, J., Haeberli, P. & Smithies, O. (1984). A comprehensive set of sequence analysis programs for the VAX. Nucleic Acids Res 12, 387-395.[Abstract]

Doull, J. L., Vats, S., Chaliciopoulos, M., Stuttard, C., Wong, K. & Vining, L. C. (1986). Conjugational fertility and location of chloramphenicol biosynthesis genes on the chromosomal linkage map of Streptomyces venezuelae. J Gen Microbiol 132, 1327-1338.

Edman, J. C., Goldstein, A. L. & Erbe, J. G. (1993). Para-aminobenzoate synthase gene of Saccharomyces cerevisiae encodes a bifunctional enzyme. Yeast 9, 669-675.[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.[Medline]

Gil, J. A. & Hopwood, D. A. (1983). Cloning and expression of a p-aminobenzoic acid synthase gene of the candicidin-producing Streptomyces griseus. Gene 25, 119-132.[Medline]

Goncharoff, P. & Nichols, B. P. (1984). Nucleotide sequence of Escherichia coli pabB indicates a common evolutionary origin of p-aminobenzoate synthetase and anthranilate synthetase. J Bacteriol 159, 57-62.[Medline]

Green, J. M., Merkel, W. K. & Nichols, B. P. (1992). Characterization and sequence of Escherichia coli pabC, the gene encoding aminodeoxychorismate lyase, a pyridoxal phosphate-containing enzyme. J Bacteriol 174, 5317-5323.[Abstract]

Halpern, T. (1997). Role of p-aminophenylalanine in the biosynthesis of chloramphenicol. BSc Honours thesis. Dalhousie University, Halifax, NS, Canada.

Henikoff, S. (1984). Unidirectional deletion with exonuclease III creates targeted breakpoints for DNA sequencing. Gene 28, 351-359.[Medline]

Higgins, D. G., Thompson, J. D. & Gibson, T. J. (1996). Using CLUSTAL for multiple sequence alignments. Methods Enzymol 266, 383-402.[Medline]

Hopwood, D. A. (1967). Genetic analysis and genome structure in Streptomyces coelicolor. Bacteriol Rev 31, 373-403.[Medline]

Hopwood, D. A., Bibb, M. J., Chater, K. F., Kieser, T., Bruton, C. J., Kieser, H. M., Lydiate, D. J., Smith, C. P., Ward, J. M. & Schrempf, H. (1985). Genetic Manipulation of Streptomyces: a Laboratory Manual. Norwich: John Innes Foundation.

Jones, A. & Vining, L. C. (1976). Biosynthesis of chloramphenicol in Streptomyces sp. 3022a. Identification of p-amino-L-phenylalanine as a product from the action of arylamine synthetase on chorismic acid. Can J Microbiol 22, 237-244.[Medline]

Larson, J. L. & Hershberger, C. L. (1986). The minimal replicon of a streptomycete plasmid produces an ultrahigh level of plasmid DNA. Plasmid 15, 199-209.[Medline]

Lin, C., Paradkar, A. S. & Vining, L. C. (1998). Regulation of an anthranilate synthase gene in Streptomyces venezuelae by a trp attenuator. Microbiology 144, 1971-1980.[Abstract]

MacNeil, D. J., Gewain, K. M., Rudy, 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.[Medline]

Malik, V. S. & Vining, L. C. (1970). Metabolism of chloramphenicol by the producing organism. Can J Microbiol 16, 173-179.[Medline]

Maniatis, T., Fritsch, E. F. & Sambrook, J. (1982). Molecular Cloning: a Laboratory Manual. Cold Spring Harbor, NY: Cold Spring Harbor Laboratory.

Mazodier, P., Petter, R. & Thompson, C. (1989). Intergeneric conjugation between Escherichia coli and Streptomyces species. J Bacteriol 171, 3583-3585.[Medline]

Munro, M. H., Taniguchi, G. M., Rinehart, K. L. & Gottlieb, D. (1975). A cmr study of the biosynthesis of chloramphenicol. Tetrahedron Lett 2659–2662.

Paradkar, A. S. & Jensen, S. E. (1995). Functional analysis of the gene encoding the clavaminate synthase 2 isoenzyme involved in clavulanic acid biosynthesis in Streptomyces clavuligerus. J Bacteriol 177, 1307-1314.[Abstract]

Redenbach, M., Kieser, H. M., Denapaite, D., Eichner, A., Cullum, L., Kinashi, H. & Hopwood, D. A. (1996). A set of ordered cosmids and a detailed genetic and physical map for the 8 Mb Streptomyces coelicolor A3(2) chromosome. Mol Microbiol 21, 77-96.[Medline]

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

Sanger, F., Nicklen, S. & Coulson, A. R. (1977). DNA sequencing with chain-terminating inhibitors. Proc Natl Acad Sci USA 74, 5463-5467.[Abstract]

Sekar, V. (1987). A rapid screening procedure for identification of recombinant bacterial clones. Biotechniques 5, 11-13.

Siddiquellah, M., McGrath, R., Vining, L. C., Sala, F. & Westlake, D. W. S. (1967). Biosynthesis of chloramphenicol II. p-Aminophenylalanine as a precursor of the p-nitrophenylserinol moiety. Can J Biochem 45, 1881-1889.[Medline]

Slock, J., Stahly, D. P., Han, C. Y., Six, E. W. & Crawford, I. P. (1990). An apparent Bacillus subtilis folic acid biosynthetic operon containing pab, an amphibolic trpG gene, a third gene required for synthesis of para-aminobenzoic acid, and the dihydropteroate synthase gene. J Bacteriol 172, 7211-7226.[Medline]

Southern, E. M. (1975). Detection of specific sequences among DNA fragments separated by gel electrophoresis. J Mol Biol 98, 503-517.[Medline]

Stuttard, C. (1982). Temperate phages of Streptomyces venezuelae: lysogeny and host specificity shown by SV1 and SV2. J Gen Microbiol 128, 115-121.

Swerdberg, G. & Skold, O. (1980). Characterization of different plasmid-borne dihydropteroate synthases mediating bacterial resistance to sulfonamides. J Bacteriol 142, 1-7.[Medline]

Teng, C. Y. P., Ganem, B., Doktor, S. Z., Nichol, B. P., Bhatnagar, R. K. & Vining, L. C. (1985). Total biosynthesis of 4-amino-4-deoxychorismic acid: a key intermediate in the biosynthesis of p-aminobenzoic acid and L-p-aminophenylalanine. J Am Chem Soc 107, 5008-5009.

Thompson, C. J., Skinner, R. H., Thompson, J., Ward, J. M., Hopwood, D. A. & Cundliffe, E. (1982). Biochemical characterization of resistance determinants cloned from antibiotic-producing streptomycetes. J Bacteriol 151, 678-685.[Medline]

Vats, S., Stuttard, C. & Vining, L. C. (1987). Transductional analysis of chloramphenicol biosynthesis genes in Streptomyces venezuelae. J Bacteriol 169, 3809-3813.[Medline]

Vining, L. C. & Stuttard, C. (1994). Chloramphenicol. In Genetics and Biochemistry of Antibiotic Production , pp. 505-530. Edited by L. C. Vining & C. Stuttard. Boston:Butterworth-Heinemann.

Wright, F. & Bibb, M. J. (1992). Codon usage in the G+C-rich Streptomyces genome. Gene 113, 55-65.[Medline]

Received 7 November 2000; revised 29 March 2001; accepted 26 April 2001.