INTRODUCTION

Top Abstract Introduction Materials and Methods Results Discussion References

Pseudomonas putida converts a variety of environmental pollutants and plant phenolic compounds to a small number of structurally simple aromatic compounds that are the starting points for pathways of aromatic ring fission (20). Ring fission is termed ortho cleavage when it occurs between two adjacent hydroxyl groups and meta cleavage when it occurs adjacent to a single hydroxyl group. P. putida can degrade the aromatic acid benzoate, after converting it to catechol, by either a meta ring cleavage pathway or an ortho ring cleavage pathway (20). The aromatic acid 4-hydroxybenzoate (4-HBA) is degraded by an ortho ring cleavage pathway after conversion to protocatechuate (Fig. 1).

View larger version (22K): [in this window] [in a new window]   FIG. 1.   Initial steps for the ortho and meta cleavage pathways used by P. putida to degrade 4-HBA, benzoate, and methylbenzoates. The meta cleavage pathway is encoded by the TOL catabolic plasmid. The methyl group can be present in either the 3 or 4 position of the ring.

The TOL (toluene degradation) catabolic plasmid (pWW0) from P. putida carries genes for the degradation of toluene and xylenes (34). On the plasmid are found two operons involved in aromatic compound degradation. One of the operons encodes enzymes for the conversion of toluene and xylenes to benzoate and methylated benzoates (the upper pathway), and the other encodes enzymes that catalyze a meta ring cleavage and subsequent reactions leading to the formation of tricarboxylic acid cycle intermediates (the meta pathway). Transcription of the meta pathway genes is regulated by XylS, a protein also encoded on the TOL catabolic plasmid (34). P. putida will also convert benzoate to catechol using chromosomally encoded enzymes. The genes for these enzymes have not yet been sequenced or fully characterized (25). Catechol, but not methylated catechols, is then further degraded to trichloroacetic acid cycle intermediates by an ortho ring cleavage pathway (Fig. 1) (20).

Benzoate induces the synthesis of the TOL plasmid-encoded enzymes of the meta fission pathway (11, 22) as well as synthesis of the chromosomally encoded enzymes that convert benzoate to catechol. Additionally, a surprising recent finding is that benzoate represses the utilization of 4-HBA (40). When P. putida is given a mixture of benzoate and 4-HBA, it degrades benzoate in preference to 4-HBA. Presumably, this is a reflection of the fact that benzoate supports a slightly higher rate of growth than 4-HBA. Benzoate was found to depress the levels of 4-HBA hydroxylase and protocatechuate dioxygenase activity, as well as the level of 4-HBA transport in cells grown on benzoate plus 4-HBA. Benzoate represses 4-HBA transport by preventing transcription of pcaK, the gene encoding the 4-HBA permease (40).

Here we described a chromosomally encoded cluster of eight genes from P. putida that is involved in conversion of benzoate to catechol and that includes a new regulatory gene, benR. We determined that BenR activates expression of benzoate dioxygenase genes in response to benzoate. It is also necessary for benzoate-dependent repression of 4-HBA transport gene expression. In addition, we demonstrate the likeness of BenR to XylS by showing that BenR activates expression of the meta cleavage pathway operon of the TOL catabolic plasmid. BenR thus has roles as an activator of benzoate degradation via ortho ring fission, as an activator of benzoate and methylbenzoate degradation via meta ring fission, and in repression of 4-HBA degradation.

	MATERIALS AND METHODS

Top Abstract Introduction Materials and Methods Results Discussion References

Bacterial strains and growth conditions. The plasmids and bacterial strains used in this study are listed in Table 1. P. putida was grown at 30°C in basal mineral (BM) medium [25 mM KH₂PO₄, 25 mM Na₂HPO₄, 0.1% (NH₄)₂SO₄, 1% Hutner mineral base (14) (final pH 6.8)] containing an appropriate carbon and energy source. Carbon sources were sterilized separately and added to final concentrations of 5 mM (benzoate or 4-HBA) and 10 mM (glucose or succinate). Plasmids were mobilized from E. coli DH5 into P. putida via a triparental mating system using E. coli HB101(pRK2013) as the mobilizing strain. The mating mixtures were incubated overnight on solid Luria-Bertani medium at 30°C. Unless specified otherwise, E. coli strains were grown at 37°C in Luria broth (LB) (7). The TOL plasmid (pWW0) was transferred from P. putida PaW1 to the P. putida benR mutant (strain 4157) by direct mating with selection on solid BM medium containing 5 mM 3-methylbenzoate plus kanamycin. Antibiotics were added to the following final concentrations (in micrograms per milliliter): ampicillin, 100; gentamicin, 5; kanamycin, 100; spectinomycin, 100; and tetracycline, 25. Solid media contained 1.5% agar.

Mutagenesis and screening for benzoate-nondegrading mutants. P. putida cells were randomly mutagenized with the transposon mini-Tn5 by mating P. putida PRS2000 cells with E. coli S17-1pir (pUTminiTn5-Km) at 30°C overnight on LB plates with a donor-to-recipient cell ratio of 1:5. Mating mixtures were suspended in BM and plated onto BM plates containing 5 mM benzoate, a low concentration of succinate (1 mM), and kanamycin. Small colonies were patched onto plates containing kanamycin and either benzoate, 4-HBA, or succinate as the sole carbon source to identify mutants that could not utilize benzoate as a sole carbon source.

Arbitrary PCR amplifications. The region of DNA flanking the transposon insertion in strain PRS4157 was amplified by arbitrary PCR (3) using arbitrary primers as described elsewhere (45). During the first round of amplification, a primer specific to the 5' end of the transposon and an arbitrary primer were used to amplify sequences flanking the upstream end of the inserted transposon. In a similar manner, a primer specific to the 3' end and an arbitrary primer were used to amplify sequences flanking the downstream end of the transposon. Product from the first-round reactions was used as template in second-round reactions with primers annealing to the 5' end of the arbitrary primers and the respective transposon-specific primers from the first-round reactions. PCR products were sequenced using the transposon-specific primers as sequencing primers.

Colony hybridization. A 335-bp probe specific to the region flanking the transposon insertion in strain PRS4157 was labeled with [³²P]dCTP using Ready To Go DNA labeling beads (dCTP) and purified with a ProbeQuant G-50 Micro column (Pharmacia Biotech, Piscataway, N.J.). E. coli colonies carrying EcoRI-generated fragments of P. putida genomic DNA in pUC19 were screened by colony hybridization (48) to identify wild-type DNA corresponding to the region that was indicated by transposon mutagenesis to be required for benzoate degradation. The probe hybridized to an approximately 12-kb fragment of DNA that was designated pCNN100. A 1.3-kb EcoRI-XmaI fragment containing the portion of the benF reading frame not contained on pCNN100 was subsequently cloned by inverse PCR (41). This clone was designated pCCH108.

DNA sequencing and analysis. DNA sequencing was performed by the University of Iowa DNA Sequencing Facility (Iowa City). Sequence assembly and analysis were done with GENE Inspector, version 1.0.1 (Textco Inc., West Lebanon, N.H.). The amino acid sequences of open reading frames were submitted to the National Center for Biotechnology Information (Bethesda, Md.) and analyzed using the BLASTp 2.0.9 algorithm (1). The sequence alignment was constructed using the CLUSTAL W multiple-sequence alignment program at the Baylor College of Medicine Human Genome Center (52), and the program BOXSHADE (version 3.21) was used to shade aligned sequences.

Cloning and DNA manipulations. Standard protocols were used for cloning and transformations. Restriction digests and ligations were performed using standard techniques. Plasmid DNA was prepared by using a QIAprep Spin Miniprep kit, and DNA restriction fragments were isolated from agarose gels using the QIAquick gel extraction kit (Qiagen Inc., Santa Clarita, Calif.).

-Galactosidase assays. -Galactosidase activities of P. putida cells carrying benA-lacZ or pcaK-lacZ transcriptional fusion plasmids were assayed according to Miller (36). For analysis of P. putida strains carrying pCCH101, cultures were grown to an A₆₆₀ of 0.1 with succinate as the sole carbon source, at which time compounds to be tested for the ability to induce gene expression were added to a final concentration of 1 mM. For analysis of P. putida strains carrying pHNN216, cultures were grown with the indicated concentrations of compounds. -Galactosidase activities were determined for cells harvested at an A₆₆₀ of 0.2. Six to eight independently grown cultures were assayed in triplicate, and the values were averaged. For analysis of gene expression in E. coli, cells were grown in LB at 30°C to an A₆₆₀ of 0.25, at which time benzoate and catechol, if added, were added to a final concentration of 1 mM. E. coli cells were harvested at a final A₆₆₀ of 0.5.

4-HBA uptake assays. Cells (50 ml) grown in BM with either 4-HBA or a mixture of benzoate (2.5 mM) and 4-HBA (2.5 mM) as carbon sources were harvested at mid-logarithmic phase, washed in 25 ml of phosphate buffer (25 mM KH₂PO₄, 25 mM Na₂HPO₄), and resuspended in 2 ml of phosphate buffer. Resuspended cells (300 µl) were added to an equal volume of reaction mixture (25 mM KH₂PO₄, 25 mM Na₂HPO₄, 4.0 mM succinate, 4.0 mM glucose, 127 µM ¹⁴C-labeled 4-HBA) to start the assay. At timed intervals, 100-µl samples were removed, filtered through Nucleopore polycarbonate membranes (0.2-µm pore size; Costar Corp., Cambridge, Mass.), and washed with 1.8 ml of phosphate buffer. Accumulated substrate was determined by scintillation counting of the cells retained on the filters.

Protein determinations. Whole cells were precipitated by addition of trichloroacetic acid to 5% and then boiled in 0.1 N NaOH for 10 min. Protein concentrations were determined using the Bio-Rad (Hercules, Calif.) protein assay, with bovine serum albumin as a standard.

Radiochemicals. [¹⁴C]uniformly-ring-labeled 4-HBA (33 Ci/mmol) and [³²P]dCTP were obtained from Amersham Corp. (Arlington Heights, Ill.).

Nucleotide accession number. The nucleotide sequence has been assigned GenBank accession number AF218267.

	RESULTS

Top Abstract Introduction Materials and Methods Results Discussion References

Identification of P. putida genes involved in benzoate degradation. To identify genes involved in benzoate degradation, we isolated a transposon mutant, PRS4157, that was unable to utilize benzoate as a sole carbon source but that grew at wild-type rates on succinate and 4-HBA. Sequencing of the region of DNA flanking the transposon revealed that the transposon was inserted in a gene predicted to encode a regulator with high similarity to XylS, a TOL plasmid-encoded activator of benzoate and methylbenzoate degradation (Fig. 2) (17). We named this gene benR. A 13.3-kb segment of P. putida DNA was subsequently cloned and sequenced and was found to include eight genes that can be inferred to be involved in benzoate degradation based on their sequence similarity to known genes (Fig. 3; Table 2).

View larger version (44K): [in this window] [in a new window]   FIG. 2.   Amino acid sequence alignment of BenR and XylS proteins. Identical residues are outlined in black. Similar residues are shaded gray.

View larger version (50K): [in this window] [in a new window]   FIG. 3.   (A) Map of chromosomally encoded benzoate degradation genes from P. putida. (B) Map of the benA promoter region, diagram of the region incorporated to construct the benA-lacZ fusion plasmid, pCCH101, and nucleotide sequence of the benA promoter region. The putative BenR binding sites are heavily underlined. Introduced restriction sites are lightly underlined. The transcriptional and translational start sites are in bold. (C) Determination of the 5' end of the benA transcript by primer extension. RNA was isolated from glucose-grown cells (lane 1) and glucose-benzoate-grown cells (lane 2) as described in the text. A sequence ladder was generated with the same primer (lanes C, T, A, and G). The first nucleotide in the transcript is shown in bold.

BenR activates the expression of benABC in response to benzoate. The benR mutant (strain PRS4157) was able to grow on benzoate when a plasmid-borne copy of benR was supplied in trans on pCCH107. The complemented mutant grew on benzoate with a generation time of 2.4 h, compared to about 1.8 h for the wild-type strain. The sequence of the benR gene and the growth phenotype of the benR mutant suggested that the BenR protein is probably involved in regulating expression of the benA, benB, and benC genes, predicted to encode benzoate 1,2-dioxygenase. Reverse transcription-PCR amplification of the regions between benA and benB and between benB and benC showed that these three genes are cotranscribed in benzoate-grown cells (results not shown). Primer extension analysis indicated that the 5' end of benA lies 30 bp upstream from its predicted translational start site (Fig. 3).

View larger version (16K): [in this window] [in a new window]   FIG. 4.   (A) -Galactosidase activities of the benA-lacZ fusion (pCCH101) in wild-type (PRS2000) and benR mutant (PRS4157) cells. Cells were grown on succinate (white), succinate plus benzoate (black), or succinate plus catechol (gray) and assayed as described in Materials and Methods. (B) -Galactosidase activities of E. coli cells harboring the benA-lacZ fusion (pCCH101) and/or the BenR expression construct (pCCH106) in the absence (white) or presence (black) of 1 mM benzoate. -Galactosidase activity was assayed as described in Materials and Methods.

BenR activates expression of the TOL plasmid-encoded meta-cleavage pathway operon. Pm, the promoter of the meta-cleavage operon from the TOL catabolic plasmid, is activated by the TOL plasmid-encoded regulator XylS when benzoate or methylbenzoates are present. A long-standing observation in studies of TOL plasmid gene regulation is that Pm can also be activated in the presence of benzoate by a chromosomally encoded regulator. The regulatory gene responsible for this activation was identified genetically in 1988 and given the name benR (6), but this gene was never sequenced. To determine whether the benR gene described here might be the regulatory gene that is responsible for XylS-independent activation of Pm, we expressed the BenR protein from pCCH107 in E. coli CC118 Pm-lacZ, a strain that has a Pm::lacZ fusion inserted in its chromosome. When BenR was present, 13,000 Miller units of -galactosidase was expressed from Pm over an undetectable background. A slight increase in -galactosidase production (17,000 Miller units) was seen when benzoate was included in the growth medium. The addition of catechol had no effect.

XylS responds to benzoate, but not 3-methylbenzoate, to modulate expression of benA. To determine whether XylS can restore benzoate-dependent regulation of the benA promoter in the absence of BenR, the TOL catabolic plasmid was introduced from P. putida PaW1 into the benR mutant (PRS4157) containing the benA-lacZ fusion plasmid (pCCH101). When the TOL plasmid was present, the observed level of benA-lacZ expression was fivefold higher when benzoate was included in the growth medium together with succinate than when cells were grown on succinate alone or on succinate plus 4-HBA or 3-methylbenzoate (Fig. 5). These results indicate that XylS, encoded on the TOL plasmid, can partially complement BenR function.

View larger version (40K): [in this window] [in a new window]   FIG. 5.   -Galactosidase activities of benR mutant (PRS4157) cells carrying the benA-lacZ fusion plasmid (pCCH101) and the TOL plasmid. Cells were grown on succinate, succinate plus 4-hydroxybenzoate, succinate plus benzoate, or succinate plus 3-methylbenzoate, as indicated, and assayed as described in Materials and Methods.

BenR is required for benzoate-mediated repression of 4-HBA degradation. When P. putida cells that are growing on 4-HBA as a sole carbon source are transferred to a medium that includes equal amounts of 4-HBA and benzoate, their rate of 4-HBA degradation decreases and they start to degrade benzoate at a high rate. Once all of the benzoate is depleted, rapid degradation of 4-HBA resumes (40). This preferred usage of benzoate over 4-HBA by P. putida can be partially explained by the observation that benzoate represses expression of pcaK, a gene that encodes a 4-HBA permease. To determine if BenR plays a role in this repression, we compared the levels of -galactosidase produced by wild-type and benR mutant strains carrying a pcaK-lacZ transcriptional fusion (Fig. 6). As was previously demonstrated (40), wild-type P. putida cells grown on a mixture of benzoate and 4-HBA expressed -galactosidase from the pcaK promoter to levels that were fivefold lower than those seen in cells grown on 4-HBA only. This repressive effect of benzoate on 4-HBA-induced pcaK expression was not seen when the pcaK-lacZ fusion was present in a benR mutant (Fig. 6). This indicates that BenR is involved in benzoate-mediated repression of pcaK expression. To determine whether BenR directly regulates pcaK, we examined the expression of the pcaK-lacZ fusion in E. coli BL21 cells in the presence and absence of BenR and benzoate. Low levels of pcaK-lacZ expression that were measured in E. coli were not influenced by BenR or benzoate (data not shown).

View larger version (30K): [in this window] [in a new window]   FIG. 6.   (A) Map of the chromosomally encoded pcaRKF cluster of 4-HBA degradation genes. (B) Map of the pcaK-lacZ fusion, pHNN216. The putative PcaR binding site is heavily underlined. Restriction sites introduced by cloning are lightly underlined. The transcriptional and translational start sites are in bold (36). (C) -Galactosidase activities of the pcaK-lacZ fusion in wild-type (PRS2000) and benR mutant (PRS4157) cells. Cells were grown on succinate (white), 4-HBA (black), or 4-HBA plus benzoate (gray), and -galactosidase activity was assayed as described in Materials and Methods.

View larger version (16K): [in this window] [in a new window]   FIG. 7.   Representative 4-HBA uptake assay of wild-type (PRS2000) and benR mutant (PRS4157) cells grown on 4-HBA or 4-HBA plus benzoate as indicated. Assays were performed as described in Materials and Methods.

	DISCUSSION

Top Abstract Introduction Materials and Methods Results Discussion References

A cluster of genes similar to the P. putida ben genes described here is also present in the gram-negative, nonmotile soil bacterium Acinetobacter sp. strain ADP1. This microbe has genes in the order benP benK benM benABDE (4, 26). Biochemical evidence indicates that the Acinetobacter benA, -B, and -C genes encode benzoate 1,2-dioxygenase and the benD gene encodes 2-hydro-1,2-dihydroxybenzoate dehydrogenase (39). The Acinetobacter benK gene encodes a benzoate transporter (5). We have assigned these functions to the homologous benABCD and -K genes from P. putida (Fig. 3). The benE genes of Acinetobacter and P. putida are homologous but do not resemble any known genes in the databases. benF from P. putida and benP from Acinetobacter each resemble porins in deduced amino acid sequence, although they do not resemble each other very much. The proposed porin function has yet to be demonstrated for either organism. A major difference between the two ben gene clusters is that they are controlled by members of two different families of regulatory proteins. Whereas expression of the Acinetobacter ben genes is controlled by BenM, a member of the LysR family of regulatory proteins (4), P. putida ben gene expression is regulated by BenR, an AraC/XylS family member.

BenR not only activates expression of the benABC genes in response to benzoate but also represses the 4-HBA-inducible expression of the 4-HBA transport protein, PcaK. Thus, BenR has the effect of shutting down the ability of cells to take up 4-HBA from their environment when benzoate is present. This provides a mechanism for the preferential degradation of benzoate by P. putida cells given a mixture of benzoate and 4-HBA. Regional regulation of aromatic compound degradation is not unique to P. putida; Acinetobacter sp. strain ADP1 also degrades benzoate in preference to 4-HBA (12). There is some evidence that BenM plays a role in mediating this preference (4).

Genes termed benR that regulated expression of chromosomal benABCD genes and that also activated Pm from the TOL plasmid were described for P. putida and P. aeruginosa some years ago (6, 25) but were not sequenced. Subsequently, Kessler et al. (30) concluded that very similar, if not identical, Pm sequence elements were recognized by the chromosomally encoded regulator BenR and the TOL plasmid-encoded regulator XylS. This prompted speculation that benR should be homologous to xylS; however, no hybridization between xylS and benR DNAs was detected in Southern hybridization experiments (25). This left open the possibility that two different types of regulators might be able to activate Pm. Results presented here show that the chromosomally encoded P. putida benR gene that activates Pm is, in fact, homologous to xylS.

Members of the AraC/XylS family of regulators are found widely distributed among bacteria. The family is characterized by a consensus sequence in the C-terminal 100 amino acids that includes two helix-turn-helix motifs that are proposed to mediate binding of the regulator to DNA (13). BenR has this consensus sequence. The activity of BenR and the benA promoter sequence have characteristics that match those of XylS activation of Pm transcription. The XylS protein is thought to bind to Pm as a dimer to a recognition sequence, TGCAN₆GGNTA, that is repeated between nucleotides 70 and 56 and between nucleotides 49 and 35 (15). The benA promoter contains a direct repeat sequence between nucleotides 68 and 34 that matches the experimentally determined XylS binding site almost exactly (Fig. 3). In interacting with the downstream binding site that overlaps the 35 binding site for RNA polymerase, BenR may compete with RNA polymerase for binding to DNA. This is consistent with our observation of a threefold-higher basal level of -galactosidase expression from the benA promoter under nonactivating conditions in the benR mutant compared to the wild type (Fig. 4A). Kaldalu et al. (28) have suggested that the N-terminal region of XylS interacts with its C-terminal domain to cause intramolecular repression of XylS function. On binding the benzoate effector, the N-terminal domain is proposed to undergo a change in conformation that then allows the C-terminal domain of XylS to function to allow initiation of transcription. The N-terminal regions of BenR and XylS share about 65% amino acid identity (Fig. 2), and BenR, like XylS, responds to benzoate as an effector. We found that BenR activates Pm expression to high levels in the absence of the benzoate effector when it is overexpressed in an E. coli background. A similar observation has been made with XylS (24, 28, 35). In the latter case, it has been proposed that a small amount of XylS is always present in a conformation that is active and able to stimulate transcription. When XylS is produced in large amounts, there is enough of the active form available to induce high levels of transcription in the absence of the effector (28, 35).

The effects of BenR in repressing transcription from the pcaK promoter may be indirect. We were unable to demonstrate an effect of BenR on pcaK expression in an E. coli background. Also, there are no detailed reports of an AraC/XylS family member responding to an effector molecule to mediate repression of gene transcription. Moreover, there is no recognizable XylS/BenR binding site in the pcaK promoter region. Clearly, much more study is required to determine the exact role of BenR in repressing pcaK transcription in response to benzoate.