This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrowReprints and Permissions
Right arrow Copyright Information
Right arrow Books from ASM Press
Right arrow MicrobeWorld
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Nowak-Thompson, B.
Right arrow Articles by Ligon, J. M.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Nowak-Thompson, B.
Right arrow Articles by Ligon, J. M.

 Previous Article  |  Next Article 

Journal of Bacteriology, February 2003, p. 860-869, Vol. 185, No. 3
0021-9193/03/$08.00+0     DOI: 10.1128/JB.185.3.860-869.2003
Copyright © 2003, American Society for Microbiology. All Rights Reserved.

2,5-Dialkylresorcinol Biosynthesis in Pseudomonas aurantiaca: Novel Head-to-Head Condensation of Two Fatty Acid-Derived Precursors

Brian Nowak-Thompson,{dagger} Philip E. Hammer,{ddagger} D. Steven Hill, Jill Stafford,§ Nancy Torkewitz, Thomas D. Gaffney, Stephen T. Lam, István Molnár,* and James M. Ligon

Syngenta Biotechnology, Inc., Research Triangle Park, North Carolina 27709

Received 7 August 2002/ Accepted 12 November 2002


arrow
ABSTRACT
 
2-Hexyl-5-propylresorcinol is the predominant analog of several dialkylresorcinols produced by Pseudomonas aurantiaca (Pseudomonas fluorescens BL915). We isolated and characterized three biosynthetic genes that encode an acyl carrier protein, a ß-ketoacyl-acyl carrier protein synthase III, and a protein of unknown function, all of which collectively allow heterologous production of 2-hexyl-5-propylresorcinol in Escherichia coli. Two regulatory genes exhibiting similarity to members of the AraC family of transcriptional regulators are also present in the identified gene cluster. Based on the deduced functions of the proteins encoded by the gene cluster and the observed incorporation of labeled carbons from octanoic acid into 2-hexyl-5-propylresorcinol, we propose that dialkylresorcinols are derived from medium-chain-length fatty acids by an unusual head-to-head condensation of ß-ketoacyl thioester intermediates. Genomic evidence suggests that there is a similar pathway for the biosynthesis of the flexirubin-type pigments in certain bacteria belonging to the order Cytophagales.


arrow
INTRODUCTION
 
The fluorescent pseudomonads produce a variety of biologically active natural products (14, 38), many of which have an ecological function in these gram-negative bacteria. Some of these natural products contribute to the suppression of plant-pathogenic fungi (20, 49, 50), whereas others are important virulence factors of certain plant-pathogenic Pseudomonas species (12). Because secondary metabolism significantly contributes to the molecular ecology of various Pseudomonas species and has also provided lead compounds for crop protection applications, several gene clusters that encode secondary metabolic pathways in this genus have been sequenced and characterized. Despite the structural simplicity of many Pseudomonas secondary metabolites, both the gene clusters and the metabolic pathways for the biosynthesis of these compounds often reveal surprising sophistication and divergence compared to the model systems studied in the gram-positive actinomycetes (13). For example, some Pseudomonas biosynthetic pathways have functionally combined modular, dissociated, or chalcone synthase-like polyketide synthases with adenylating enzymes (pyoluteorin, mupirocin, coronatine) or with components of fatty acid synthases (2,4-diacetylphloroglucinol) (13). Other Pseudomonas pathways are composed of enzymes that rearrange and modify primary metabolic products to produce secondary metabolites (pyrrolnitrin, phenazine 1-carboxylic acid). Thus, Pseudomonas natural product research provides an opportunity to study not only the ecological function of secondary metabolism but also the potential diversity found in secondary metabolic pathways. In the study described here we further illustrated the metabolic versatility of the pseudomonads by characterizing the biosynthetic pathway and gene cluster responsible for dialkylresorcinol production in Pseudomonas aurantiaca (previously identified as Pseudomonas fluorescens BL915).

A systematic antifungal screening program of the Syngenta natural products research group in Switzerland previously demonstrated that P. aurantiaca produces various antifungal compounds, including 2-hexyl-5-propyl-alkylresorcinol (HPR). HPR was first isolated from an unidentified Pseudomonas species in 1975 and was reported to possess moderate antifungal and antibacterial properties (31, 35). Later studies described an HPR analog, resorstatin, and demonstrated that both compounds can act as free radical scavengers to protect against superoxide-induced tissue damage (32).

Concurrent with the initial discovery of HPR, several polyene metabolites that are collectively referred to as flexirubins were shown to possess the generalized 2,5-dialkylresorcinol (DAR) structure present as an aromatic ester, a moiety that is, in effect, an HPR analog (6-8). The major structural variations in the flexirubins are due to the alkyl substituents of this DAR moiety, which sometimes have terminally branched alkyl chains (1, 4). The flexirubins are characteristic taxonomic markers for certain Cytophaga, Flexibacter, and Flavobacterium species (5, 43, 44).

In an interesting series of publications, two groups of workers have described various biosynthetic studies of both the flexirubin DAR moiety and HPR (Fig. 1). Achenbach et al. initially described incorporation of radiolabeled acetate into the DAR moiety and speculated about the possibility that orsellinic acid is a DAR precursor (2). Sankawa et al. subsequently demonstrated incorporation of [1,2-13C2]acetate into HPR and, based on the observed labeling pattern, proposed that HPR arises by an unusual head-to-head condensation of two polyketide chains (45). Shortly thereafter, Achenbach et al. reported identical 13C-labeling patterns in the flexirubin DAR moiety (3); however, they also presented further evidence obtained from in vivo competitive labeling studies that is consistent with the hypothesis that orsellinic acid is a direct precursor of the DAR moiety. In addition, orsellinic acid was detected in culture extracts by isotope dilution techniques, further strengthening arguments that it is involved as a pathway precursor.



View larger version (20K):
[in this window]
[in a new window]
  		FIG. 1. Proposed pathways for the biosynthesis of resorcinol moieties in natural products. The HPR pathway is the Pseudomonas sp. pathway proposed by Sankawa et al. (45), the flexirubin pathway is the Flexibacter elegans pathway suggested by Achenbach et al. (3), and the pyoluteorin pathway is the P. fluorescens pathway(40). The thick lines represent incorporation of intact acetate units as determined by feeding experiments.

The formation of the 2-acylresorcinol ring of the phytotoxic and antifungal Pseudomonas metabolite pyoluteorin (48) has been elucidated previously. The pattern of [1,2-13C2]acetate incorporation (18) and the deduced functions of the genes in the cloned biosynthetic gene cluster (40) demonstrated that the resorcinol moiety of pyoluteorin is assembled by a type I polyketide synthase that uses a proline-derived pyrrole ring as the starter unit. Despite the incorporation of acetate into the resorcinol moieties of both HPR and pyoluteorin, the resultant labeling patterns were found to be distinct (Fig. 1), making it unlikely that the biosynthetic pathways are analogous.

In the present study, we obtained nucleotide sequence and precursor incorporation data suggesting that the DAR moiety in HPR produced by P. aurantiaca is neither derived directly from orsellinic acid nor assembled from polyketide precursors. Instead, it is formed by a novel head-to-head condensation of two fatty acid-derived precursors. Furthermore, in this paper we present genomic evidence suggesting that there is a homologous biosynthetic pathway for the flexirubin DAR moiety. More importantly, however, we believe that the results of this study further develop our view of Pseudomonas secondary metabolism and the ingenuity of this genus in exploiting the aldol condensation as a means of natural product biosynthesis.


arrow
MATERIALS AND METHODS
 
Bacterial strains. Escherichia coli DH5{alpha} and DH10B were used for routine cloning. P. aurantiaca strain BL915 (= ATCC 55169) was used as the HPR producer strain. E. coli S17-1 was used as a donor host for introducing plasmids into Pseudomonas by conjugation (47). The different plasmids used in this study are listed in Table 1.


View this table:
[in this window]
[in a new window]
  		TABLE 1. Plasmids used in this study

Mutagenesis and cloning of the dar (dialkylresorcinol) gene cluster. Transposon mutagenesis of P. aurantiaca strain BL915 was achieved by using plasmid pCIB116 (36) that contained a modified Tn5 element as described by Lam and Torkewitz (37). Mutants were screened for altered antifungal activity by surveying in vitro inhibition of Neurospora crassa (37). Mutants with changed antifungal activity were screened for the production of several antifungal compounds, including pyrrolnitrin, HPR, chitinase, and HCN. The Tn5-generated mutation in strain 2215 that confers an HPR-deficient phenotype was then complemented by shuttling a BL915 cosmid library, made up in pHC79 (29), into strain 2215.

The Tn5 element with flanking chromosomal DNA was rescued from strain 2215 by in vivo marker exchange between the complementing cosmid, pBL3610, and the chromosome of strain 2215. Cosmid pBL3610 was conjugated into mutant 2215, and after a period of growth, it was transferred back into E. coli DH10B by conjugation. Exconjugants were subsequently selected for kanamycin resistance. Thus, the Tn5 insertion in mutant strain 2215 was transferred by homologous recombination into cosmid pBL3610, yielding cosmid pBL3610Tn.

To define the site of the Tn5 insertion, EcoRI fragments of pBL3610Tn were subcloned in E. coli, and each clone was tested for its ability to impart kanamycin resistance. The sequence of the region flanking the insertion was generated by using DNA sequencing primers directed outward from the ends of the Tn5 element. Heterologous expression was also used to further characterize the DAR biosynthetic gene cluster by subcloning BglII fragments derived from pBL3610 into an E. coli host strain expressing the Pseudomonas pleiotropic response regulator gac*3 (24). pB3, containing an 11.3-kb BglII fragment derived from pBL3610, was capable of transferring HPR production to E. coli. Although a functional gac3 allele is strictly required for production of DAR in Pseudomonas, the presence of the gac*3 gene in E. coli hosts was later proven not to be necessary for heterologous production of DAR.

DNA and protein sequence analysis. DNA sequence analysis was performed by using the dideoxynucleotide chain termination method with Applied Biosystems model 377 sequencers. Primer sites in the double-stranded, 11,269-bp cloned BglII DNA fragment were introduced by using an EZ::TN <Tet-1> transposon insertion kit (Epicentre, Madison, Wis.). The results of the individual DNA sequencing runs were assembled by using the PHRED, PHRAP, and CONSED programs (22, 23, 26). The University of Wisconsin Genetics Computer Group programs (19) were used for sequence analysis. The codon frequency table used for CodonPreference analysis was generated from coding regions of the 16S rRNA of P. aurantiaca (GenBank accession no. AB021412) and both the hydrogen cyanide synthase cluster and the mannitol operon of P. fluorescens (GenBank accession no. AF053760 and AF007800, respectively).

Deduced protein sequences encoded by the open reading frames (ORFs) identified in the dar cluster were compared to the GenBank database sequences maintained by the National Center for Biotechnology Information (National Institutes of Health, Bethesda, Md.) with the BLAST algorithm and the standard parameters of Altschul et al. (9).

Production and isolation of HPR. HPR was routinely produced by either P. aurantiaca BL915 or E. coli DH10B carrying an appropriate plasmid in 15-ml cultures in Luria-Bertani medium containing the appropriate antibiotics in 125-ml baffled flasks. The cultures were incubated at 37 and 28°C for E. coli and P. aurantiaca BL915, respectively. HPR was extracted into 10 ml of ethyl acetate and dried under a vacuum. For high-performance liquid chromatography (HPLC) or thin-layer chromatography (TLC) analysis, a sample was dissolved in a minimal volume of methanol and analyzed under conditions described below. For isolation of pure HPR, a sample was dissolved in ethyl acetate, adsorbed on a small quantity of silica gel, and separated on a flash silica column eluted with a mixture containing 92% hexane and 8% ethyl acetate. The purified HPR was crystallized from hot hexane. The 1H and 13C nuclear magnetic resonance (NMR) spectra for the purified HPR were identical to the previously reported data (31, 35, 45).

HPR analysis by HPLC, TLC, or GC-MS. HPR was detected by HPLC on a CC125/4 Kromasil 100-5 C18 column (Macherey-Nagel Inc., Easton, Pa.) by using the following gradient at a rate of 1.5 ml/min: initial conditions, 50:50 methanol-H2O; 15 min with 75:25 methanol-H2O; 16 min with 100% methanol; 21 min with 100% methanol; and 22 min with 50:50 methanol-H2O (both solvents contained 0.1% acetic acid). Compounds were detected at 270 nm. The retention time for HPR under these conditions was 10.3 min. HPR was qualitatively detected by TLC on normal-phase silica gel (60Å; Sigma, St. Louis, Mo.) developed in 100% toluene and was visualized as a magenta derivative with vanUrk's reagent (2% 4-dimethylaminobenzaldehyde in a 50:50 mixture of ethanol and concentrated HCl). Separation of HPR was also achieved by gas chromatography (GC) on an HP-5MS cross-linked 5% phenylmethyl Siloxane column (30 m by 0.25 mm by 0.25 µm) with the following temperature program: initial conditions, 50°C; temperature increased at a rate of 10°C/min to 250°C; and temperature increased at a rate of 25°C/min to 300°C. Separation was monitored by mass spectrometry (MS), and HPR was detected at 19.49 min as the M+H ion m/z 237. Positive ion chemical ionization-MS (CH4): m/z 237 (M+H), 221 (M-CH3), 207 (M-C2H5), 193 (M-C3H7), 179 (M-C4H9), 165 (M-C5H11).

[13C]octanoic acid labeling study. Ten 1-liter flasks containing 200 ml of Luria-Bertani medium were each inoculated with 200 µl of an overnight P. aurantiaca BL915 culture. After 5 h of growth at 28°C, 500 µl of a suspension of [1,2,3,4-13C4]octanoic acid (132 mg in 5 ml of 5% NaHCO3; Cambridge Isotope Laboratories, Andover, Mass.) was added to each of the flasks. The cultures were grown for an additional 9 h, and another 500 µl of a suspension of [1,2,3,4-13C4]octanoic acid (125 mg in 5 ml 5% NaHCO3) was added to each flask. To limit degradation of the labeled compound, 800 µl of a 250 mM 3-mercaptopropionate solution (Sigma) was also added to each of the flasks at the same time (39). The cultures were grown for a total of 22 h at 28°C, and each entire culture was extracted twice with 500 ml of ethyl acetate. The organic layer was dried over anhydrous MgSO4 and evaporated to dryness. HPR was isolated with a flash silica column by using the protocol described above (yield, 5.5 mg), dissolved in CDCl3, and analyzed by standard 1D-13C and 2D-INADEQUATE NMR.

Nucleotide sequence accession number. The annotated nucleotide sequence of the dar gene cluster and the surrounding region has been deposited in the National Center for Biotechnology Information GenBank database under accession number AY135187.


arrow
RESULTS
 
Mutagenesis and cloning of the dar gene cluster. Transposon mutagenesis of P. aurantiaca strain BL915 was carried out by using a modified Tn5 element as described in Materials and Methods. A single mutant that failed to produce HPR (strain 2215) was identified. The HPR-deficient phenotype of mutant 2215 was complemented by conjugating a cosmid library of strain BL915 genomic DNA into the mutant strain. One cosmid clone, pBL3610, was found to be capable of restoring HPR production in strain 2215. Subcloning different restriction fragments of pBL3610 resulted in identification of an 11.3-kb BglII fragment (cloned as pB3) that was sufficient to direct HPR production in E. coli as a host, suggesting that it contained the biosynthetic gene cluster necessary for HPR production.

Sequence analysis of the dar gene cluster and flanking regions. We defined 12 ORFs within the sequenced 11.3-kb DNA fragment containing the dar gene cluster (Fig. 2A). All of the ORFs identified exhibited a G+C bias in the third codon position ranging from 75 to 85%. Putative ribosome binding sites were identified for orf1, orf3, orf5, and orf11 based on the presence of at least four of the eight nucleotide bases of the AAGGAGGT consensus sequence approximately 8 to 12 bases upstream of the identified start codon. Less convincing ribosome binding sites, comprised of a purine-rich region extending as far as 17 bases upstream from the initiation codon, were identified for orf2, orf9, orf10, and orf13. Only 7 bp separate orf1 and orf2, and less than 50 bp separate orf11, orf12, and orf13; it is possible, therefore, that these ORFs are translationally coupled. From the sequence analysis, we could not justify the translation of either orf4 or orf7, neither of which appeared to possess any pretense of a ribosome binding site or to be organized in a manner that would allow translational coupling.



View larger version (64K):
[in this window]
[in a new window]
  		FIG. 2. (A) Dialkylresorcinol gene cluster and flanking regions. The ORFs represented by the black arrows (orf1 to orf5) have been demonstrated to be involved in dialkylresorcinol production and have been designated as follows: orf1, darA; orf2, darB; orf3, darC; orf4, darR; and orf5, darS. The ORFs represented by the gray arrows are not involved in dialkylresorcinol production, and their presumed functions are described in Table 2. (B) Alignment of KAS III protein sequences: FabH proteins from E. coli (FabH Escol) (GI:16129054) and M. tuberculosis (FabH Mytub) (GI:15607673), DarB from P. aurantiaca (DarB Psaur) (this study), and protein from C. hutchinsonii homologous to DarB (DarB Cyhut) (http://www.jgi.doe.gov/JGI_microbial/html/cytophaga/cytoph_homepage.html). A gray background indicates sequence conservation in all four proteins. A black background indicates significant amino acids mentioned in the text; the positions of these amino acids are indicated above the sequence. Insertions in the DarB proteins (compared to E. coli FabH) are indicated by thick lines above the sequence.

BLAST search analysis revealed homologous proteins for all but three of the predicted peptides encoded by the ORFs identified. Although the peptides encoded by orf8 and orf9 have database homologs, no putative functions have been described for any of the database sequences. The BLAST results are summarized in Table 2, but only those genes functionally demonstrated to play a role in HPR biosynthesis are described here in detail.


View this table:
[in this window]
[in a new window]
  		TABLE 2. BLAST search results for each ORF identified in Fig. 2

(i) Biosynthetic enzymes. The predicted protein encoded by orf3 exhibits high levels of similarity to many acyl carrier proteins (ACPs). A PROFILESCAN analysis identified the diagnostic motif LGxDS 33 amino acids from the N terminus, which contains the invariant Ser residue required for covalent binding of the 4'-phosphopantetheine cofactor required for ACP function (17). An alternative start codon is present 6 bp upstream from the putative start codon; however, a ribosome binding site could not be identified to support translation from the alternative start site. Furthermore, alignment of the orf3 protein sequence with database sequences suggested that protein synthesis initiates from the assigned start codon. The orf3 coding sequence was designated darC (Fig. 2A).

BLAST search results identified the orf2-encoded protein as a homolog of 3-ketoacyl-ACP synthase III (KAS III) enzymes, which are collectively represented by the E. coli FabH protein. Multiple-sequence alignment with other FabH proteins identified several conserved residues in the deduced Orf2 protein (Fig. 2B). These residues include the catalytic nucleophile Cys112 (E. coli FabH numbering); the general base His244, which also binds the carbonyl oxygen of the malonyl-ACP extender unit; Asn274, which promotes the decarboxylation of malonyl-CoA via stabilization of the enol intermediate; Gly306, which stabilizes the oxyanion; and Arg42, which takes part in maintaining the structural integrity of the ligand binding domain (28). Residues that are implicated in interactions either with the pantheteine arm of the holo-ACP (i.e., Asn247, Gly209, Ile156, Met207, Val212, Phe213, and Ile250) or with the adenine ring of the coenzyme A (CoA) (Trp32 and Arg151) are not conserved. In addition, of the residues that are known to form the active site hydrophobic pocket of E. coli FabH (Leu142, Phe157, Leu189, Leu205, and Phe87'), only Leu142 is conserved in the orf2 protein. Most interestingly, the Orf2 protein also contains seven sequence insertions that vary from 3 to 19 amino acids long and are distributed along the entire peptide (Fig. 2B). The orf2 coding sequence was designated darB (Fig. 2A).

The protein encoded by orf1 had no related sequences in the National Center for Biotechnology Information protein database but did exhibit 32% identity to a hypothetical protein encoded in the Cytophaga hutchinsonii genome. The uncharacterized Cytophaga gene is adjacent to another gene encoding a KAS III homolog that is very similar to darB (BLAST expect value, 3e-99; high-scoring segment pair, 381 amino acids), suggesting that a dar-like pathway exists in this bacterium. Other computational methods failed to reveal any recognizable sequence motifs; thus, no putative function could be assigned to this deduced protein based solely on sequence analysis. Nevertheless, functional analysis of orf1 demonstrated that it is required for HPR production (see below). Therefore, the orf1 coding sequence was designated darA (Fig. 2A).

(ii) Regulatory proteins and promoter region. BLAST searches for the deduced proteins encoded by both orf4 and orf5 suggested that these proteins are members of the AraC/XylS family of transcriptional regulators. Analysis of the orf4 protein with HTHScan identified two helix-turn-helix (HTH) domains, one between amino acids 238 and 257 (score, 22.0; probability, 6.5E-6) and the other between amino acids 303 and 322 (score, 9.2; probability, 4.5E-3). Furthermore, this protein also contains 16 of the 19 amino acid residues that are highly conserved in the HTH region of AraC/XylS proteins (21, 51). Although the peptide sequences encoded by orf5 and orf4 exhibit 41% identity over the first 188 amino acid residues, the coding region of orf5 is 306 bp shorter than that of orf4 and does not contain the AraC/XylS-type HTH domains. Conceptual translations downstream of orf5 for each possible reading frame eliminated the possibility that a sequencing error or a recent mutation prematurely truncated the gene sequence. These observations suggest that both orf4 and orf5 might be involved in transcriptional regulation of the dar cluster; however, it is likely that only the orf4-encoded protein is capable of directly binding to a promoter region. The coding sequences of orf4 and orf5 were designated darR and darS, respectively (Fig. 2A).

Identification of a putative transcriptional regulatory protein prompted us to search for promoter elements that might be targeted by DarR. Promoters regulated by AraC/XylS proteins like DarR generally contain tandem repeat sequences (21, 51), although there is no conserved motif among the repeat sequences. There is an 859-bp region between darA and orf13 that has a significantly lower G+C content than the entire dar cluster (51 versus 64%), suggesting that it possesses promoter activity. We identified a region 553 to 491 bp upstream from the darA start codon that has a high degree of repeat character and may serve as the target for the inferred DNA binding activity of DarR.

Heterologous expression and functional analysis in E. coli. It was observed early during the course of this work that HPR was produced by E. coli strains containing the dar gene cluster. We therefore initially characterized gene function through heterologous expression in E. coli. Chemical analysis of culture extracts by either TLC or HPLC revealed that pPEH148, which contains only darA, darB, and darC, conferred HPR production on E. coli hosts. HPR was not produced, however, by strains containing a kanamycin resistance marker disrupting either darA (pPEH143) or darB (pPEH144). Furthermore, Ptac-driven expression of neither darA (pPEH161) nor darB (pPEH162) nor darC (pPEH163) was sufficient for HPR production in E. coli. However, plasmid pBT3028, which contains only darA and darB, conferred some HPR production in the heterologous host. These results indicate that both darA and darB are necessary for heterologous production of HPR but that a native E. coli protein(s) may complement the function of DarC. Indeed, the E. coli fatty acid synthase (FAS) ACP has been shown to be capable of interacting with FAS components from Bacillus subtilis, Streptomyces collinus, and even avocado (27) and also with the type II polyketide synthase (PKS)-associated KAS III homolog DpsC (11).

HPR production in E. coli strains carrying plasmid pPEH145, which contains darR and darS in addition to darABC, was approximately 3.5-fold greater than HPR production in strains with pPEH148 (containing only darABC). This observation is consistent with the predicted regulatory function of DarR and/or DarS. The level of HPR production by a darR-disrupted strain (pPEH140) was indistinguishable from that observed with cultures of a darABC strain (pPEH148). Surprisingly, HPR production by the strain with the darABC({Delta}S)R plasmid pRB1068 was only about 1.7-fold greater than HPR production by the darABC strain. Therefore, it appears that DarR is able to increase HPR production on its own but that both DarS and DarR are required for maximal HPR production in E. coli.

Functional analysis of the dar cluster in P. aurantiaca BL915. To assess gene functions in the native host, each identified dar gene either was deleted completely or had an in-frame deletion constructed in the coding sequence. The resulting constructs were subsequently used to replace the native coding sequence by homologous recombination in the BL915 chromosome. HPR production was completely eliminated by disruption of either darA or darB but could be fully restored by Ptac-mediated expression of the corresponding gene (Fig. 3). A {Delta}darC strain, however, still produced HPR, although at a much lower level than the wild type. Surprisingly, complementation of the darC deletion could only partially restore HPR production. These results demonstrate that there is a requirement for both darA and darB for HPR production and also suggest that the function of darC can be replaced by one or more endogenous proteins in Pseudomonas, mirroring the results described above for the heterologous production of HPR in E. coli with darA, darB, and darC.



View larger version (42K):
[in this window]
[in a new window]
  		FIG. 3. TLC analysis of dialkylresorcinol production in darABC deletion constructs. The location of dialkylresorcinol is indicated by the arrow. The spot with greater mobility corresponds to the antifungal compound pyrrolnitrin. Lanes {Delta}, culture extracts of P. aurantiaca containing chromosomal deletions in the genes indicated; lanes +, culture extracts of trans-complemented deletion strains. darB1 and darB2 are two independent isolates with the same chromosomal deletion in the darB gene.

Deletion analysis of darS and darR was also performed by constructing markerless gene replacements of each coding sequence and analyzing the production of HPR in culture extracts. Deletion of either darS or darR resulted in an approximately 50% reduction in the HPR titer compared to that in the wild-type strain.

Incorporation of octanoic acid into HPR. Previous labeling studies suggested that HPR is formed by the head-to-head condensation of two polyketide chains (45). The presence of a KAS III homolog in the dar cluster indicated, however, that HPR is assembled by a modified fatty acid synthase, an explanation that can also be reconciled with the incorporation data. To establish whether HPR is biosynthesized de novo from acetyl-CoA or whether medium-chain-length fatty acids from primary metabolism are used as intermediates in this pathway, [1,2,3,4-13C4]octanoic acid was added to actively growing cultures of P. aurantiaca BL915, and HPR was isolated and analyzed by 13C NMR. The existence of adjacent labeled nuclei was demonstrated by complex splitting patterns within the proton-decoupled 13C NMR spectrum (Fig. 4A). A 33-Hz coupling between carbon resonances at {delta} 29.2 (d) and 20.6 (dd) ppm was evidence that there was labeling at the C-1' and C-2' positions of the hexyl side chain. The resonance at {delta} 20.6 ppm was further split into two signals with a coupling constant of 43 Hz, which correlated with a second 43-Hz coupling observed in the C-2 alkyl-substituted carbon resonance at {delta} 112.5 (dd) ppm. These observed couplings established that at least three contiguous carbons from the labeled octanoic acid were incorporated into HPR.



View larger version (24K):
[in this window]
[in a new window]
  		FIG. 4. (A) Observed 13C NMR couplings for 13C-labeled hexylpropylresorcinol. (B) Positive chemical ionization mass spectrum of 13C-labeled hexylpropylresorcinol. The molecular ion peak was detected as the M+H ion at m/z 237. Molecular ions of the M+2, M+4, and M+6 multiply labeled isotopomers indicated were absent from unlabeled samples. The resulting loss of a C5H11 fragment as indicated on the chemical structure was detected as a daughter ion at m/z 165. The corresponding isotopomers were reduced by 1 mass unit and detected as M+3 and M+5 fragment ions. The deduced incorporation of [1,2,3,4--13C]octanoic acid is indicated by the thick lines in the chemical structure.

Additional couplings were also observed among the aromatic resonances at {delta} 154.3, 112.5, and 108.1 ppm. It was not possible, however, to definitively assign the labeled carbons due to the symmetry of the resorcinol ring and the resulting overlap of the carbon resonances. Nevertheless, the complexity of the splitting patterns suggested that a population of isotopomers existed in which both the C-1C-2 and C-3C-4 centers were labeled. GC-MS analysis identified molecular ions for MH+2, MH+4, and MH+6 isotopomers (m/z 239, 241, and 243, respectively) (Fig. 4B), demonstrating that there was incorporation of up to six 13C nuclei within a single molecule. Furthermore, loss of a C5H11 fragment from the hexyl side chain resulted in a prominent daughter ion of m/z 165 and the corresponding isotopomer ions of MH+3 and MH+5. This result indicates that one of the 13C labels in the C5H11 fragment was lost and that the remaining fragment ion contained label at either three or five distinct positions. We therefore concluded that both the C-1C-2 and the C-3C-4 positions within the ring were labeled by incorporation of octanoic acid into HPR. The observed labeling of the C-1C-2C-1'C-2' fragment suggests that there was direct incorporation of octanoic acid into HPR. In contrast, labeling of the C-3C-4 fragment can be explained by the ß-oxidation of octanoic acid, followed by incorporation of a labeled hexanoic acid derivative.


arrow
DISCUSSION
 
We cloned and characterized the biosynthetic gene cluster for HPR biosynthesis in P. aurantiaca BL915 and identified three biosynthetic genes that are sufficient for heterologous production of HPR. These genes encode a ß-ketoacyl-ACP synthase III homolog and a novel protein with no functionally characterized homolog in the data banks, both of which are necessary for HPR production, and an ACP, whose function can be at least partially supplemented by endogenous proteins both in Pseudomonas and in E. coli. Based on biosynthetic incorporation studies, Sankawa et al. (45) suggested that HPR is derived from a polyketide pathway. Despite the presence of a gene that encodes a type II ACP homolog in the gene cluster, this possibility can now be discounted since it has been found that the dar gene cluster does not contain a gene that encodes a KAS II homolog like the PKS ketosynthases and consequently HPR biosynthesis does not require a KAS II homolog like the PKS ketosynthases. Furthermore, this gene cluster does not encode additional proteins typically found in type II PKS gene clusters, such as the chain length factor (KSQ) (30, 34).

Instead, we propose that HPR biosynthesis represents a unique offshoot of fatty acid metabolism in which medium-chain-length fatty acid-derived precursors are further modified by DarABC (Fig. 5). In this model, DarA catalyzes a head-to-head condensation between two ß-ketoacyl thioester intermediates. Although both thioester intermediates may originate from medium-chain-length fatty acids that have undergone a single, chain-extending, aldol condensation (Fig. 5, step 1) catalyzed by DarBC (i.e., ß-keto-ACPDarC adducts), it is also possible that DarA accepts ß-keto-CoA thioester intermediates of fatty acid catabolism as substrates (step 2). Formation of the C-2C-3 bond likely results in thioester hydrolysis of the shorter-chain precursor, followed by C-5C-6 bond formation and subsequent dehydration to form the dioxocyclohexene intermediate (step 3). Decarboxylation at C-6 is expected to occur following thioester hydrolysis of the longer-chain precursor (step 4), and tautomerization of the decarboxylated intermediate forms the final aromatic resorcinol moiety of HPR (step 5). This proposed pathway rationally explains the function of each protein encoded in the dar gene cluster. In addition, it is consistent with the previously observed incorporation of acetate (3, 45), as well as the reported detection of a carboxylated DAR derivative in culture extracts of a flexirubin-producing Flexibacterium species (3).



View larger version (13K):
[in this window]
[in a new window]
  		FIG. 5. Proposed pathway for dialkylresorcinol production in P. aurantiaca BL915. Precursors are thought to be derived from both fatty acid biosynthesis and degradation. Pathway intermediates are enclosed by brackets. It is not known whether the ß-ketohexanoyl thioester intermediate is an ACP or acyl-CoA derivative, and this intermediate is indicated as an R functionality. The final aromatization of the ring is believed to be a spontaneous tautomerization and not enzyme dependent. See the text for a detailed description of the reaction steps. n-Bu, n-butyl.

We demonstrated the validity of the proposed pathway through incorporation of 13C-labeled octanoic acid into HPR. Predictably, two routes for incorporation exist; there is a direct route involving the catalytic activity of DarBC, and there is an indirect route in which octanoate is first partially degraded to a 3-ketohexanoate intermediate of the ß-oxidation pathway. Both routes of incorporation were observed and show that HPR is, in fact, derived from medium-chain-length fatty acid precursors. Furthermore, GC-MS analysis of partially purified HPR detected identical labeling patterns for other dialkylresorcinol derivatives having different alkyl chain lengths produced by P. aurantiaca (Nowak-Thompson, unpublished data), suggesting that DarB and DarA are able to tolerate chain length variation in their substrate. Although the variation in chain length in HPR analogs parallels the structural variations in the DAR moiety of the flexirubins, our labeling results contradict previous data suggesting that the flexirubin DAR moiety originates from the direct alkylation of orsellinic acid (3). The evidence suggesting that orsellinic acid is a DAR precursor is, however, indirect and does not unequivocally demonstrate incorporation into the resorcinol moiety. In contrast, the observed 13C coupling between the C-2 of the aromatic ring and C-1' of the hexyl side chain can only be explained if labeled octanoic acid is incorporated intact, thereby excluding late-stage alkylation of the aromatic ring. While it could be argued that the flexirubin DAR moiety and HPR originate from two distinct bacteria that may utilize different biosynthetic routes for production of these structurally related compounds, it should be noted that both a darA homolog and a darB homolog were found linked in the genomic sequence of C. hutchinsonii, a species which tests positive for the presence of flexirubin-type pigments (43). We believe, therefore, that the biosynthetic pathway proposed here is broadly applicable to DAR production in both the pseudomonads and the Cytophagales. It is also interesting, however, that the resorcinol moiety in an another Pseudomonas metabolite, pyoluteorin, is formed from a single polyketide chain assembled by a modular PKS, which results in a different acetate labeling pattern (Fig. 1).

In light of the proposed role of fatty acyl thioester precursors in HPR biosynthesis, we considered two possible sources for these compounds from primary metabolism. One possible supply mechanism is by direct shunting of FAS ACPs, charged with ß-ketoacyl intermediates, from fatty acid biosynthesis. The proposed involvement of an ACP adduct and the partial complementation of DarC by endogenous proteins in both E. coli and P. aurantiaca support this possibility; however, we believe that this is unlikely because it would require a mechanism that prevents ketoreduction of fatty acid biosynthetic intermediates having specific chain lengths, while intermediates with shorter and longer chains are easily reduced. Nevertheless, it is possible that the DarB KAS III commits medium-chain-length fatty acyl thioesters for HPR biosynthesis by providing unreduced ß-ketoacyl-ACP for DarA in a manner analogous to prodiginine biosynthesis in Streptomyces coelicolor. Prodiginine biosynthesis proceeds by single-chain extension of dodecoyl-ACP by pathway-specific KAS II domains that provides ß-ketomyristoyl-ACPRedN, but since RedN does not interact with FAS ketoreductases, the ß-ketoacyl thioester avoids reduction (15). Alternatively, fatty acid degradation can also serve as a direct source for ß-ketoacyl thioesters for HPR biosynthesis, as exemplified by the observed incorporation of ketohexanoate derived from [1,2,3,4-13C4]octanoic acid. Despite various attempts to physiologically and genetically determine the relative contributions of fatty acid biosynthesis and fatty acid catabolism in supplying HPR precursors in either P. aurantiaca or E. coli, we were unable to identify a predominant contribution of one pathway over the other.

The importance of KAS III enzymes in both primary and secondary metabolic pathways should not be overlooked considering the ability of these enzymes to discriminate among available pathway precursors. In primary metabolism, KAS III enzymes, such as FabH of E. coli, initiate fatty acid biosynthesis by catalyzing the condensation of specific acyl-CoA primers with malonyl-ACP extenders. The substrate specificities of FabH enzymes largely determine the fatty acid profiles of microorganisms. For example, both E. coli and Streptococcus pneumoniae synthesize only straight-chain fatty acids. The corresponding FabH enzymes for these bacteria accept acetyl-CoA, propionyl-CoA, and, to a small extent, butyryl-CoA (but not long- or branched-chain acyl-CoA) primers, despite the apparent sequence divergence between the substrate binding pockets of the two enzymes (33). In contrast, FabH from Staphylococcus aureus, which produces a relatively high percentage of branched-chain fatty acids, accepts acetyl-CoA and butyryl-CoA as primers, but the most preferred substrate is isobutyryl-CoA. Although the substrate specificities of the S. aureus and E. coli FabH proteins are dramatically different, amino acids within the substrate binding pockets of these enzymes are highly conserved (28).

The amino acids that constitute the FabH substrate binding pockets in DarB and its Cytophaga homolog (Leu142, Phe157, Leu189, Leu205, and Phe89') (Fig. 2B) are also substantially divergent compared to the amino acids that constitute the FabH substrate binding pockets in the E. coli enzyme (41). The paradoxical relationship between the conserved residues within the substrate binding pockets and the substrate specificity of the FabH enzymes described above does not allow one to rationalize catalytic function on this basis alone. Nevertheless, a comparison of the DarB sequences with the Mycobacterium tuberculosis FabH sequences required for meromycolic acid (C-50 to C-56) biosynthesis provided some insight. The M. tuberculosis FabH contains a second hydrophobic channel adjacent to the active site that accommodates the bulky substrate myristoyl-CoA (46). Access to this channel in the E. coli enzyme is sterically blocked by Phe87' (from the other monomer), yet this residue has been replaced by Thr and Met in M. tuberculosis FabH and DarB, respectively. In addition, other residues that sterically block the channel access in E. coli FabH are displaced by structural changes induced by a 4-amino-acid insertion (amino acids 201 to 204) in M. tuberculosis FabH; both DarB and its Cytophaga homolog also possess a 4-amino-acid insertion at the same position. While structural similarities between DarB and M. tuberculosis FabH are certainly consistent with the proposed function of DarB, definitive proof awaits further biochemical characterization.

KAS III homologs also play important roles in secondary metabolite biosynthesis. During prodiginine and methylenomycin biosynthesis in S. coelicolor and diacetylphloroglucinol biosynthesis in Pseudomonas KAS III homologs are used to synthesize acetoacetyl starter units for subsequent condensations (10, 15, 16). KAS III homologs also play a role in some type II PKS clusters to define or modulate starter unit selection; DpsC in daunorubicin biosynthesis, for example, serves as a fidelity factor for selection of the propionate starter unit (42). It is not unprecedented, therefore, that HPR biosynthesis involves a KAS III homolog tailored specifically for selection of medium-chain-length fatty acid precursors.

Secondary metabolite production in Pseudomonas is controlled by complex regulatory networks. Global regulators like the LemA/GacA two-component regulatory system or sigma factors like {sigma}S and {sigma}70 exert their pleiotropic effects through pathway-specific regulators that are customary features in Pseudomonas secondary metabolite biosynthetic gene clusters. phzR in phenazine biosynthetic clusters and mupR in mupirocin biosynthetic clusters encode transcriptional activators that belong to the LuxR family (13). pltR in the pyoluteorin cluster encodes a LysR-type transcriptional activator, and pchR in the pyochelin cluster codes for an AraC-type transcriptional activator (13). Using heterologous expression in E. coli, we demonstrated that darSR also encode regulatory proteins capable of increasing dialkylresorcinol production. The deduced proteins DarS and DarR are both members of the AraC/XylS family of transcriptional regulators; however, only DarR contains the HTH domains required for DNA binding, suggesting that there is a direct interaction of DarR with the darABC promoter region. Similarly, in the coronatine biosynthetic gene cluster of Pseudomonas syringae, corR and corP code for response regulators that both have (almost identical) receiver domains, but only CorR has an HTH DNA binding domain (13). When darR was coexpressed in E. coli with the darABC operon, production of HPR was increased 1.7-fold. However, it was only when both darR and darS were present that production of dialkylresorcinol was maximized. The role of DarS is less obvious, however, since inclusion of darS on darABC plasmids without darR did not increase production. Given the dimeric nature of the AraC transcriptional regulators and the general requirement for an effector molecule for AraC-mediated transcription (25), it is possible that DarR and DarS act cooperatively to induce transcription. In such a system, DarR could provide the DNA recognition domain and DarS could contribute to recognition of the effector molecule. One consequence of such a system might be to regulate dialkylresorcinol production in response to two different effector molecules, one that is recognized by the DarR-DarS complex and another that is recognized by the DarR homodimer.

Characterization of HPR biosynthesis in P. aurantiaca further demonstrated that the pseudomonads have successfully harnessed the synthetic utility of the aldol condensation within their secondary metabolic pathways, an observation that has also been borne out through characterization of the gene clusters responsible for pyoluteorin, 2,4-diacetylphloroglucinol, and coranatine biosynthesis. Whereas the structural complexity of secondary metabolites produced by the pseudomonads is rather limited compared to the structural complexity of actinomycete secondary metabolites, it nevertheless appears that by integrating aldol condensations within different biochemical contexts, the pseudomonads are exploring the molecular diversity of hybrid biosynthetic pathways. If this is a characteristic of Pseudomonas secondary metabolism, it presents a compelling argument to continue efforts to understand secondary metabolism in this genus, in the hope that we may begin to exploit the natural diversity of the aldol condensation chemistry and incorporate it into engineered pathways.


arrow
ACKNOWLEDGMENTS
 
We extend our gratitude to Makoto Ono and An Hu of Syngenta Biotechnology, Inc. for their expert assistance in DNA sequencing and assembly and to Pierre Rouviere of DuPont Central Research and Development for bringing the flexirubin structures to our attention.


arrow
FOOTNOTES
 
* Corresponding author. Mailing address: Syngenta Biotechnology, Inc., 3054 Cornwallis Rd., Research Triangle Park, NC 27709. Phone: (919) 541-8640. Fax: (919) 541-8585. E-mail: istvan.molnar{at}syngenta.com. Back

{dagger} Present address: Northland College, Ashland, WI 54806. Back

{ddagger} Present address: Athenix Corp., Durham, NC 27702. Back

§ Present address: Kean University, Union, NJ 07083. Back


arrow
REFERENCES
 
    1
  1. Achenbach, H., A. Boettger-Vetter, E. Fautz, and H. Reichenbach. 1982. On the origin of the branched alkyl substituents on ring B of flexirubin-type pigments. Arch. Microbiol. 132:241-244.[CrossRef]
    2. 2
  2. Achenbach, H., A. Bottger, W. Kohl, E. Fautz, and H. Reichenbach. 1979. Understanding the biogenesis of the flexirubins—origin of benzene ring A and the aromatic C-methyl group. Phytochemistry 18:961-963. (Translated.)[CrossRef]
    3. 3
  3. Achenbach, H., A. Bottger-Vetter, D. Hunkler, E. Fautz, and H. Reichenbach. 1983. Investigations on the biosynthesis of flexirubin—the origin of benzene ring B and its substituents. Tetrahedron 39:175-185.[CrossRef]
    4. 4
  4. Achenbach, H., and W. Kohl. 1979. Elucidation of the structure of pigments of the flexirubin-type—mass spectrometry studies. Chem. Ber. 112:209-217. (Translated.)
    5. 5
  5. Achenbach, H., W. Kohl, and H. Reichenbach. 1978. The flexirubin-type pigments—a novel class of natural pigments from gliding bacteria. Rev. Latinoam. Quim. 9:111-124.
    6. 6
  6. Achenbach, H., W. Kohl, and H. Reichenbach. 1976. Investigation on metabolites of microorganisms. XI. Flexirubin, a novel pigment from Flexibacter elegans. Chem. Ber. 109:2490-2502.
    7. 7
  7. Achenbach, H., W. Kohl, and H. Reichenbach. 1977. The principle pigments from Cytophaga johnsonae—two new pigments of the flexirubin type. Tetrahedron Lett. 12:1061-1062.
    8. 8
  8. Achenbach, H., W. Kohl, W. Wachter, and H. Reichenbach. 1978. Investigations of the pigments from Cytophaga johnsonae Cy jl. New flexirubin-type pigments. Arch. Microbiol. 117:253-257.[CrossRef][Medline]
    9. 9
  9. Altschul, S. F., W. Gish, W. Miller, E. W. Meyes, and D. J. Lipman. 1990. Basic local alignment search tool. J. Mol. Biol. 215:403-410.[CrossRef][Medline]
    10. 10
  10. Bangera, M. G., and L. S. Thomashow. 1999. Identification and characterization of a gene cluster for synthesis of the polyketide antibiotic 2,4-diacetylphloroglucinol from Pseudomonas fluorescens Q2-87. J. Bacteriol. 181:3155-3163.[Abstract/Free Full Text]
    11. 11
  11. Bao, W., P. J. Sheldon, and C. R. Hutchinson. 1999. Purification and properties of the Streptomyces peucetius DpsC beta-ketoacyl:acyl carrier protein synthase III that specifies the propionate-starter unit for type II polyketide biosynthesis. Biochemistry. 38:9752-9757.[CrossRef][Medline]
    12. 12
  12. Bender, C. L., F. Alarcon-Chaidez, and D. C. Gross. 1999. Pseudomonas syringae phytotoxins: mode of action, regulation, and biosynthesis by peptide and polyketide synthetases. Microbiol. Mol. Biol. Rev. 63:266-292.[Abstract/Free Full Text]
    13. 13
  13. Bender, C. L., V. Rangaswamy, and J. Loper. 1999. Polyketide production by plant-associated pseudomonads. Annu. Rev. Phytopathol. 37:175-196.[CrossRef][Medline]
    14. 14
  14. Budzikiewicz, H. 1993. Secondary metabolites from fluorescent pseudomonads. FEMS Microbiol. Rev. 10:209-228.[Medline]
    15. 15
  15. Cerdeno, A. M., M. J. Bibb, and G. L. Challis. 2001. Analysis of the prodiginine biosynthesis gene cluster of Streptomyces coelicolor A3(2): new mechanisms for chain initiation and termination in modular multienzymes. Chem. Biol. 8:817-829.[CrossRef][Medline]
    16. 16
  16. Challis, G., and K. Chater. 2001. Incorporation of [U-13C]glycerol defines plausible early steps for the biosynthesis of methylenomycin A in Streptomyces coelicolor A3(2). Chem. Commun. 10:935-936.[CrossRef]
    17. 17
  17. Cortes, J., S. F. Haydock, G. A. Roberts, D. J. Bevitt, and P. F. Leadlay. 1990. An unusually large multifunctional polypeptide in the erythromycin-producing polyketide synthase of Saccharopolyspora erythraea. Nature 348:176-178.[CrossRef][Medline]
    18. 18
  18. Cuppels, D. A., C. R. Howell, R. D. Stipanovic, A. Stoessl, and J. B. Stothers. 1986. Biosynthesis of pyoluteorin: a mixed polyketide-tricarboxylic acid cycle origin demonstrated by [1,2-13C2]-acetate incorporation. Z. Naturforsch. Sect. C Biosci. 41:532-536.
    19. 19
  19. Devereux, J., P. Haeberli, and O. Smithies. 1984. A comprehensive set of sequence analysis programs for the VAX. Nucleic Acids Res. 12:387-395.
    20. 20
  20. Dowling, D. N., and F. O'Gara. 1994. Metabolites of Pseudomonas involved in the biocontrol of plant disease. Trends Biotechnol. 12:133-144.
    21. 21
  21. Egan, S. M. 2002. Growing repertoire of AraC/XylS activators J. Bacteriol. 184:5529-5532.[Free Full Text]
    22. 22
  22. Ewing, B., and P. Green. 1998. Base-calling of automated sequencer traces using phred. II. Error probabilities. Genome Res. 8:186-194.[Abstract/Free Full Text]
    23. 23
  23. Ewing, B., L. Hillier, M. C. Wendl, and P. Green. 1998. Base-calling of automated sequencer traces using phred. I. Accuracy assessment. Genome Res. 8:175-185.[Abstract/Free Full Text]
    24. 24
  24. Gaffney, T. D., S. T. Lam, J. Ligon, K. Gates, A. Frazelle, J. Di Maio, S. Hill, S. Goodwin, N. Torkewitz, A. M. Allshouse, et al. 1994. Global regulation of expression of antifungal factors by a Pseudomonas fluorescens biological control strain. Mol. Plant-Microbe Interact. 7:455-463.[Medline]
    25. 25
  25. Gallegos, M. T., R. Schleif, A. Bairoch, K. Hofmann, and J. L. Ramos. 1997. Arac/XylS family of transcriptional regulators. Microbiol. Mol. Biol. Rev. 61:393-410.[Abstract]
    26. 26
  26. Gordon, D., C. Abajian, and P. Green. 1998. Consed: a graphical tool for sequence finishing. Genome Res. 8:195-202.[Abstract/Free Full Text]
    27. 27
  27. Han, L., S. Lobo, and K. A. Reynolds. 1998. Characterization of beta-ketoacyl-acyl carrier protein synthase III from Streptomyces glaucescens and its role in initiation of fatty acid biosynthesis. J. Bacteriol. 180:4481-4486.[Abstract/Free Full Text]
    28. 28
  28. He, X., and K. A. Reynolds. 2002. Purification, characterization, and identification of novel inhibitors of the beta-ketoacyl-acyl carrier protein synthase III (FabH) from Staphylococcus aureus. Antimicrob. Agents Chemother. 46:1310-1318.[Abstract/Free Full Text]
    29. 29
  29. Hohn, B., and J. Collins. 1980. A small cosmid for efficient cloning of large DNA fragments. Gene 11:291-298.[CrossRef][Medline]
    30. 30
  30. Hutchinson, C. R., and I. Fujii. 1995. Polyketide synthase gene manipulation: a structure-function approach in engineering novel antibiotics. Annu. Rev. Microbiol. 49:201-238.[CrossRef][Medline]
    31. 31
  31. Kanda, N., N. Ishizaki, N. Inoue, M. Oshima, and A. Handa. 1975. DB-2073, a new alkylresorcinol antibiotic. I. Taxonomy, isolation and characterization. J. Antibiot. (Tokyo) 28:935-942.[Medline]
    32. 32
  32. Kato, S., K. Shindo, H. Kawai, M. Matsuoka, and J. Mochizuki. 1993. Studies on free radical scavenging substances from microorganisms. III. Isolation and structural elucidation of a novel free radical scavenger, resorstatin. J. Antibiot. (Tokyo) 46:1024-1026.[Medline]
    33. 33
  33. Khandekar, S. S., D. R. Gentry, G. S. Van Aller, P. Warren, H. Xiang, C. Silverman, M. L. Doyle, P. A. Chambers, A. K. Konstantinidis, M. Brandt, R. A. Daines, and J. T. Lonsdale. 2001. Identification, substrate specificity, and inhibition of the Streptococcus pneumoniae beta-ketoacyl-acyl carrier protein synthase III (FabH). J. Biol. Chem. 276:30024-30030.[Abstract/Free Full Text]
    34. 34
  34. Khosla, C., R. S. Gokhale, J. R. Jacobsen, and D. E. Cane. 1999. Tolerance and specificity of polyketide synthases. Annu. Rev. Biochem. 68:219-253.[CrossRef][Medline]
    35. 35
  35. Kitahara, T., and N. Kanda. 1975. DB-2073, a new alkylresorcinol antibiotic. II. The chemical structure of DB-2073. J. Antibiot. (Tokyo) 28:943-946.[Medline]
    36. 36
  36. Lam, S. T., D. M. Ellis, and J. M. Ligon. 1991. Genetic approaches for studying rhizosphere colonization, p. 43-50. In D. L. Keister and P. B. Cregan (ed.), The rhizosphere and plant growth. Kluwer Academic Press, Amsterdam, The Netherlands.
    37. 37
  37. Lam, S. T., and N. Torkewitz. March 1996. Pseudomonas biocontrol strains. U.S. patent 5496547.
    38. 38
  38. Leisinger, T., and R. Margraff. 1979. Secondary metabolites of the fluorescent pseudomonads. Microbiol. Rev. 43:422-442.[Free Full Text]
    39. 39
  39. Li, Z., F. M. Martin, and J. C. Vederas. 1992. Biosynthetic incorporation of labeled tetraketide intermediates into dehydrocurvularin, a phytotoxin from Alternaria cinerariae, with assistance of beta-oxidation inhibitors J. Am. Chem. Soc. 114:1531-1533.[CrossRef]
    40. 40
  40. Nowak-Thompson, B., N. Chaney, J. S. Wing, S. J. Gould, and J. E. Loper. 1999. Characterization of the pyoluteorin biosynthetic gene cluster of Pseudomonas fluorescens Pf-5. J. Bacteriol. 181:2166-2174.[Abstract/Free Full Text]
    41. 41
  41. Qiu, X., C. A. Janson, A. K. Konstantinidis, S. Nwagwu, C. Silverman, W. W. Smith, S. Khandekar, J. Lonsdale, and S. S. Abdel-Meguid. 1999. Crystal structure of beta-ketoacyl-acyl carrier protein synthase III. A key condensing enzyme in bacterial fatty acid biosynthesis. J. Biol. Chem. 274:36465-36471.[Abstract/Free Full Text]
    42. 42
  42. Rajgarhia, V. B., N. D. Priestley, and W. R. Strohl. 2001. The product of dpsC confers starter unit fidelity upon the daunorubicin polyketide synthase of Streptomyces sp. strain C5. Metab. Eng. 3:49-63.[CrossRef][Medline]
    43. 43
  43. Reichenbach, H. 1989. Cytophagales Leadbetter 1974, 99, p. 2013-2071. In J. T.Staley, M. P. Bryant, N. Pfennig, and J. G. Holt (ed.), Bergey's manual of systematic bacteriology, vol. 3. Williams & Wilkins, Baltimore, Md.
    44. 44
  44. Reichenbach, H., W. Kohl, A. Boettger-Vetter, and H. Achenbach. 1980. Flexirubin-type pigments in Flavobacterium. Arch. Microbiol. 126:291-293.[CrossRef]
    45. 45
  45. Sankawa, U., H. Shimada, and K. Yamasaki. 1981. Biosynthesis of 2-hexyl-5-propylresorcinol: biosynthetic incorporation of deuterium from [2-13C, 2-2H3]-acetate. Chem. Pharm. Bull. 29:3601-3605.
    46. 46
  46. Scarsdale, J. N., G. Kazanina, X. He, K. A. Reynolds, and H. T. Wright. 2001. Crystal structure of the Mycobacterium tuberculosis beta-ketoacyl-acyl carrier protein synthase III. J. Biol. Chem. 276:20516-20522.[Abstract/Free Full Text]
    47. 47
  47. Simon, R., U. Priefer, and A. Pühler. 1983. A broad host range mobilization system for in vivo genetic engineering: transposon mutagenesis in gram negative bacteria. Bio/Technology 1:784-791.[CrossRef]
    48. 48
  48. Takeda, R. 1958. Pseudomonas pigments. I. Pyoluteorin, a new chlorine-containing pigment produced by Pseudomonas aeruginosa. Hakko Kogaku Zasshi 36:281-290.
    49. 49
  49. Thomashow, L., and D. Weller. 1995. Current concepts in the use of introduced bacteria for biological control: mechanisms and antifungal metabolites, p. 187-235. In G. Stacey and N. T. Keen (ed.), Plant-microbe interations, vol. 1. Chapman & Hall, New York, N.Y.
    50. 50
  50. Thomashow, L. S. 1996. Biological control of plant root pathogens. Curr. Opin. Biotechnol. 7:343-347.[CrossRef][Medline]
    51. 51
  51. Tobes, R., and J. L. Ramos. 2002. AraC-XylS database: a family of positive transcriptional regulators in bacteria Nucleic Acids Res. 30:318-321.[Abstract/Free Full Text]

Journal of Bacteriology, February 2003, p. 860-869, Vol. 185, No. 3
0021-9193/03/$08.00+0     DOI: 10.1128/JB.185.3.860-869.2003
Copyright © 2003, American Society for Microbiology. All Rights Reserved.



This article has been cited by other articles:

  • McBride, M. J., Xie, G., Martens, E. C., Lapidus, A., Henrissat, B., Rhodes, R. G., Goltsman, E., Wang, W., Xu, J., Hunnicutt, D. W., Staroscik, A. M., Hoover, T. R., Cheng, Y.-Q., Stein, J. L. (2009). Novel Features of the Polysaccharide-Digesting Gliding Bacterium Flavobacterium johnsoniae as Revealed by Genome Sequence Analysis. Appl. Environ. Microbiol. 75: 6864-6875 [Abstract] [Full Text]  

This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrowReprints and Permissions
Right arrow Copyright Information
Right arrow Books from ASM Press
Right arrow MicrobeWorld
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Nowak-Thompson, B.
Right arrow Articles by Ligon, J. M.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Nowak-Thompson, B.
Right arrow Articles by Ligon, J. M.

Copyright © 2009 by the American Society for Microbiology.   For an alternate route to Journals.ASM.org, visit: http://intl-journals.asm.org | More Info»