The benPK operon, proposed to play a role in transport, is part of a regulon for benzoate catabolism in Acinetobacter sp. strain ADP1

Todd J. Clark1, Cory Momany2 and Ellen L. Neidle1

Department of Microbiology, University of Georgia, Athens, GA 30602-2605, USA1
Department of Pharmaceutical and Biomedical Sciences, University of Georgia, Athens, GA 30602, USA2

Author for correspondence: Ellen L. Neidle. Tel: +1 706 542 2852. Fax: +1 706 542 2674. e-mail: eneidle{at}arches.uga.edu


   ABSTRACT
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES
 
BenM and CatM are distinct, but similar, LysR-type transcriptional regulators of the soil bacterium Acinetobacter sp. strain ADP1. Together, the two regulators control the expression of at least 14 genes involved in the degradation of aromatic compounds via the catechol branch of the ß-ketoadipate pathway. In these studies, BenM and CatM were each purified to homogeneity to test the possibility that they regulate the expression of two additional genes, benP and benK, that are adjacent to benM on the chromosome. Each regulator bound to a DNA fragment containing the benP promoter region. Additional transcriptional studies suggested that benP and benK are co-transcribed as an operon, and a site of transcription initiation was identified. Alignment of this initiation site with those of several CatM- and BenM-regulated genes revealed common regulatory motifs. Mutants lacking both CatM and BenM failed to activate benP transcription. The ability of each protein to regulate gene expression was inferred from strains lacking either CatM or BenM that were still capable of increasing benP expression in response to cis,cis-muconate. This compound has previously been shown to induce all enzymes of the catechol branch of the ß-ketoadipate pathway through a complex transcriptional circuit involving CatM and BenM. Thus, the regulated expression of the benPK operon in concert with other genes of the regulon is consistent with the model that BenP, a putative outer-membrane porin, and BenK, an inner-membrane permease, transport aromatic compounds in strain ADP1.

Keywords: porin, BenM, CatM, LTTR (LysR-type transcriptional regulators)

Abbreviations: CCM, cis,cis-muconate


   INTRODUCTION
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES
 
The adjacent benP and benK genes of the Gram-negative soil bacterium Acinetobacter sp. strain ADP1 are in a chromosomal region associated with aromatic compound degradation. A wide variety of aromatic compounds can serve as carbon and energy sources for ADP1 and are degraded through the ß-ketoadipate pathway. This multi-step catabolic route feeds into the tricarboxylic acid cycle and is encoded by genes that are grouped together in large supraoperonic clusters (Harwood & Parales, 1996 ; Young et al., 2001 ). The location and sequence of benP and benK suggest that they participate in the uptake of compounds, such as benzoate, that are subsequently converted to catechol, a key metabolite of the ß-ketoadipate pathway. BenK, an inner-membrane permease, plays a role in the cellular entry of benzoate and other aromatic compounds, although it is not essential for growth on benzoate as a sole carbon source (Collier et al., 1997 ). As described in this report, we tested the effect of inactivating benP, which encodes a putative outer-membrane porin. Moreover, the expression of benP and benK was investigated to determine whether transcriptional regulation of these genes is coordinated with that of other genes of the associated pathway.

Upstream of benP are genes involved in the degradation of alkyl salicylates (salA, -R, -E, -D) and the degradation of alkanoate esters of benzyl alcohols (areA, -B, -C, -R) (Jones et al., 1999 , 2000 ; Jones & Williams, 2001 ). The SalE and AreABC enzymes convert various aromatic compounds to salicylate (2-hydroxybenzoate) or benzoate. Salicylate is the substrate of SalA, a hydroxylase that mediates the production of catechol. Genes for the degradation of catechol and those that enable benzoate to be converted to catechol are immediately downstream of benK (Fig. 1) (Collier et al., 1998 ). Thus, benP and benK are sandwiched between genes that help funnel a wide array of aromatic compounds into catechol, the substrate of a ring-cleaving dioxygenase (Fig. 1). This genetic arrangement may reflect the participation of BenP and BenK in aromatic compound transport.



View larger version (25K):
[in this window]
[in a new window]
 
Fig. 1. Benzoate degradation in ADP1. The Ben and Cat enzymes (a) are encoded by genes and operons in a 21 kbp chromosomal region that is not drawn to scale (b). Arrows below genes indicate their relative transcriptional orientation. The BenM and CatM regulators control expression of their own genes and activate transcription from the promoters of the benABCDE operon (PbA), catA (PcA) and the catBCIJFD operon (PcB). Regulation of the benPK promoter (PbP) was investigated in these studies. In an alignment of the promoter regions (c), known or proposed binding sequences for the regulatory proteins are boxed, and transcriptional start sites are underlined (+1). Identities in three (:) or four (|) of the aligned sequences are indicated. Arrows (*A and *B) indicate the sites of benP transcription initiation that would generate primer-extension products of the size shown by corresponding arrows in Figs 3 and 4.

 
BenK is required for wild-type rates of benzoate uptake and for the unimpaired use of benzoate or benzaldehyde as a sole carbon source. In addition, the expression of a chromosomal benK::lacZ fusion increases in response to cis,cis-muconate (CCM), an inducer that regulates all genes known to participate in catechol degradation (Collier et al., 1997 ). CCM, which is produced from catechol ring cleavage, interacts with CatM, a LysR-type transcriptional regulator, to activate expression of the catA gene and the catBCIJFD operon (Fig. 1) (Romero-Arroyo et al., 1995 ). One goal of the current studies was to determine whether CatM is responsible for the CCM-inducibility of benK expression. A second LysR-type transcriptional regulator, BenM, activates expression of the benABCDE operon needed for the conversion of benzoate to catechol (Collier et al., 1998 ). This regulator, encoded by a gene adjacent to benK, is similar to CatM in sequence and in its ability to respond to CCM and activate catA expression. Therefore, a possible role for BenM in benK expression was also investigated.

The function and expression of benP have not previously been investigated. Porin-like genes similar to benP have been identified near several genetic regions involved in bacterial aromatic compound degradation (Cowles et al., 2000 ; Segura et al., 1999 ). For example, phaK, essential for the assimilation of phenylacetic acid in a Pseudomonas putida strain, appears to encode a specific channel-forming protein (Olivera et al., 1998 ). The presence of genes likely to encode porins in different catabolic regions suggests that protein channels can facilitate the entry of aromatic compounds into Gram-negative bacteria. In ADP1, the location of the benP gene immediately upstream of benK suggested that both might be co-expressed. It also seemed likely that CatM and/or BenM would control gene expression. To test these possibilities, RT-PCR and primer-extension methods were used to study the transcriptional regulation of benP and benK. In addition, the CatM and BenM proteins were purified to enable further investigation of their roles in regulating the ß-ketoadipate pathway for aromatic compound dissimilation.


   METHODS
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES
 
Strains and growth conditions.
Descriptions of strains and plasmids used in this study are listed in Table 1. Acinetobacter strains are derivatives of Acinetobacter sp. ADP1, originally designated Acinetobacter calcoaceticus BD413 (Juni & Janik, 1969 ). Plasmids were maintained in Escherichia coli DH5{alpha} (Gibco-BRL). All bacterial cultures were grown in Luria–Bertani (LB) broth or minimal medium (MM) at 37 °C as previously described (Sambrook et al., 1989 ; Shanley et al., 1986 ). Carbon sources were supplemented to LB or MM at the following final concentrations: anthranilate, 3 mM; benzoate, 3 mM; cis,cis-muconate (CCM), 2·5 mM; and succinate, 10 mM. Benzaldehyde, benzyl acetate, benzyl alcohol, ethyl salicylate and salicylate were provided as sole carbon sources in a range of concentrations from 0·5 to 2·5 mM. Volatile compounds, such as the esters, were spotted onto filter paper in the lids of inverted Petri dishes containing minimal medium. Antibiotics were added as necessary at the following final concentrations: ampicillin, 150 µg ml-1; kanamycin, 25 µg ml-1; streptomycin, 25 µg ml-1; and spectinomycin, 25 µg ml-1. Bacterial growth was monitored by spectrophotometric measurement of the OD600 of cultures (Sambrook et al., 1989 ).


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

 
DNA manipulations, plasmid construction, and interposon mutagenesis.
Standard methods were used for plasmid DNA purifications, restriction enzyme digestions, PCR, electrophoresis, ligations and E. coli transformations (Sambrook et al., 1989 ). Oligonucleotides were purchased from Genosys (Table 2). Plasmid pBAC84, carrying benP, was constructed by ligating a 3·4 kbp SmaI–EcoRI fragment from pBAC68 into HincII/EcoRI-digested pUC19. Plasmid pBAC370, used for interposon mutagenesis of benP, was constructed by inserting an omega cassette conferring resistance to streptomycin and spectinomycin ({Omega}S), into the centre of benP. A 2·0 kbp PstI fragment from pHP45, with {Omega}S followed by translational and transcriptional stop signals, was ligated into a 500 bp deletion in pBAC84 created by ClaI digestion at benP sites, at nucleotide positions 2323 and 2719 (numbering system of Fig. 2).


View this table:
[in this window]
[in a new window]
 
Table 2. Oligonucleotide primers

 


View larger version (40K):
[in this window]
[in a new window]
 
Fig. 2. Co-expression of the benP and benK genes. The relative positions of genes (arrows), primers (triangles) and RT-PCR products (lines) are indicated (a). Numbers correspond to DNA sequence such that position 1 corresponds to the last position of the stop codon of the complementary sequence of areR. RT-PCR products were separated on an agarose gel (b) from reactions with the BENP-FOR and BENK-REV primers and total RNA from ADP1 grown on succinate (lanes 1 and 2), benzoate (lanes 3 and 4) or CCM (lanes 5, 6 and 7). An RT-PCR product from a previous reaction was digested with XbaI (lane 7). No reverse transcriptase was added to reactions in the even-numbered lanes. An arrow shows the 1015 bp product. The positions and sizes of DNA size standards (STD) are indicated on the right.

 
The CatM expression plasmid, pBAC381, was constructed after PCR-amplifying catM from plasmid pIB25 with the primers CatM-NDEI and CatM-BAMHI (Table 2). The PCR product was electrophoresed on a 1·0% agarose gel, excised, and purified with the QIAquick gel extraction kit (Qiagen). At sites introduced by the amplification primers, the PCR product was digested with NdeI and BamHI and ligated into the similarly digested pET-21b expression vector (Novagen). The BenM expression plasmid pBAC382 was similarly constructed with primers BenM-NDEI and BenM-ECORI (Table 2) and plasmid pBAC14 as the template for PCR amplification. The DNA sequences of the entire catM and benM genes of pBAC381 and pBAC382, respectively, were verified.

To generate an Acinetobacter strain with an inactivated chromosomal benP gene, plasmid pBAC370 was digested with NdeI to yield a linear fragment with the disrupted benP::{Omega}S5450 allele. This fragment was purified from a 1% agarose gel using the Qiaquick purification kit (Qiagen). As described previously (Neidle et al., 1989 ), the linearized DNA fragment was used to transform and to replace the corresponding chromosomal region of strain ADP1, generating ACN450. The correct chromosomal configuration of ACN450 was confirmed by Southern hybridization analysis as previously described (Gregg-Jolly & Ornston, 1990 ).

Expression and purification of CatM and BenM.
E. coli strain BL21-Gold (DE3) (Novagen) was transformed with pBAC381 or pBAC382 and plated on solid medium with ampicillin. An isolated colony was used to inoculate 5 ml LB broth with ampicillin, and the culture was grown at 37 °C with agitation until reaching an OD600 between 0·4 and 0·8. This 5 ml culture was used to inoculate 1 litre of fresh medium and allowed to grow to an OD600 of 0·4–0·8, at which point the culture was put on ice. After 10 min on ice, the culture was transferred to a 16 °C incubator, and protein expression was induced with the addition of IPTG to a final concentration of 1 mM. After overnight incubation (10–14 h), each 1 litre culture was divided into 100 ml batches that were harvested by centrifugation at 7000 g for 10 min at 4 °C. After removal of the supernatant fluid, cell pellets were immediately stored at -70 °C.

Cell pellets containing CatM or BenM were suspended on ice in 10 ml buffer A1 (50 mM Tris/HCl pH 6·0, 50 mM NaCl, 5%, v/v, glycerol, 0·5 mM EDTA and 0·5 mM DTT) containing PMSF at a final concentration of 100 µg ml-1. Cells were lysed by two passages through a 4 °C chilled French pressure cell at 15000 p.s.i. (103·5 MPa). The resulting lysate was centrifuged at 15000 g for 15 min at 4 °C. All column purification steps were done with an FPLC system from Amersham-Pharmacia. A 10 ml sample of the supernate was loaded onto a 5 ml HITRAP SP cation-exchange column that had been equilibrated with 25 ml buffer A1. Protein was eluted from the column at a flow rate of 1 ml min-1 over a linear gradient of buffer B1 (buffer A1 with 1 M NaCl) and immediately placed on ice. Fractions containing CatM (300–350 mM NaCl) or BenM (250–350 mM NaCl) were pooled following analysis by 12% SDS-PAGE (Sambrook et al., 1989 ).

To lower the NaCl concentration and allow subsequent binding of CatM or BenM to additional columns, pooled samples were diluted fivefold with buffer A2 (50 mM Tris/HCl pH 7·0, 50 mM NaCl, 5%, v/v, glycerol, 0·5 mM EDTA and 0·5 mM DTT). The mixture was concentrated to a 10 ml volume with an Ultrafree S-10 centrifuge concentrator (Millipore). Samples were loaded onto a 5 ml HITRAP heparin-agarose affinity column that had been equilibrated with 25 ml buffer A2. Protein was eluted over a linear gradient of buffer B2 (Buffer A2 with 1 M NaCl) at 1 ml min-1. Fractions containing CatM (500–600 mM NaCl) or BenM (300–400 mM NaCl) were identified by SDS-PAGE. Samples containing pure protein were pooled, diluted fivefold with buffer A2, and concentrated to 1–3 ml using an Ultrafree S-10 centrifuge concentrator. Samples were stored in 20 µl aliquots at -70 °C, at a final concentration of 0·5–1·0 mg ml-1.

Protein concentrations were determined by the method of Bradford (1976) with bovine serum albumin as the standard. The molecular mass of CatM was estimated by gel-filtration chromatography. A 5 mg sample of purified CatM (1·0 mg ml-1) was loaded onto a HiPrep Sephacryl S-200 gel-filtration column that had been equilibrated with buffer A2. Protein was eluted and fractions analysed by SDS-PAGE to determine the elution volume for CatM. Known proteins served as standards according to the directions of the Molecular mass-Gel Filtration-200 kit (Sigma).

RNA extraction, RT-PCR, and primer-extension analysis.
Total RNA was extracted from mid-exponential-phase (OD600 0·3–0·5) Acinetobacter cultures by a hot phenol extraction method as described by Williams & Rogers (1987) . RNA samples were treated with DNase I (Promega). The Dnase I was later removed from the RNA with RNeasy columns (Qiagen) in accordance with the manufacturer’s specifications. Total RNA concentrations were estimated spectrophotometrically and rRNA bands were visualized on a 1·0% agarose gel (Sambrook et al., 1989 ).

The Omniscript reverse transcriptase kit (Qiagen) was used for both RT-PCR and primer-extension methods. Taq polymerase (Fisherbrand) and RNasin ribonuclease inhibitor (Promega) were used for RT-PCR reactions. Products were analysed on 1·0% agarose gels and sometimes purified with the QIA-quick gel extraction kit (Qiagen). For primer-extension reactions, the BENP-PE primer (Table 2) was end-labelled with T4 polynucleotide kinase (Promega) and [{gamma}-32P]ATP (ICN Biomedicals) according to the manufacturer’s instructions. G-25 spin columns (Amersham-Pharmacia) were used to remove free [{gamma}-32P]ATP from the labelling reactions. One microlitre of the end-labelled primer was combined with 2 µg total RNA in a final volume of 6 µl and incubated at 65 °C for 5 min to allow primer annealing. A DNA sequencing ladder was generated with the fmol DNA Cycle Sequencing kit (Promega), {gamma}-32P end-labelled BENP-PE primer, and plasmid pBAC84 containing the wild-type benP region as template. Primer-extension products and sequencing reactions were analysed by 6% denaturing PAGE and visualized by autoradiography (Sambrook et al., 1989 ).

Gel-retardation assays.
A double-stranded DNA fragment with the benP operator–promoter region (PbP) was radiolabelled as previously described (Parsek et al., 1994 ). The BENPO/P-FOR and BENPO/P-REV primers, after being 5' end-labelled with [{gamma}-32P]ATP and T4 polynucleotide kinase (Promega), were used to PCR-amplify a 317 bp DNA fragment from pBAC84. This labelled fragment was purified following separation by 5% PAGE. The conditions for DNA-binding reactions were previously described (Romero-Arroyo et al., 1995 ). Binding reactions were initiated by mixing 5–1000 ng CatM or BenM with 1000 c.p.m. of the 32P-labelled benP fragment (approx. 10 nM) in the presence or absence of 50 mM CCM. The DNA–protein complexes were analysed as described by Parsek et al. (1994) .

Unlabelled DNA fragments were used in competition studies with the labelled fragment to verify the specificity of CatM and BenM binding. For use as a specific competitor, a DNA fragment with PbP was generated by PCR as described above except that the BENPO/P-FOR and BENPO/P-REV primers were not labelled with 32P. To generate a fragment for non-specific DNA competition studies, the M13 forward and M13 reverse universal primers (Promega) were used in PCR to amplify a 656 bp fragment from pBAC539 (Table 1), which contains the pobRA operator–promoter region. This region of the ADP1 chromosome regulates the catabolism of p-hydroxybenzoate (DiMarco et al., 1993 ) and does not contain a recognizable binding sequence for CatM or BenM. Competition studies were done by adding an increasing range of competitor DNA, from 1- to 50-fold molar excess, to the reaction as previously described (Tobiason et al., 1999 ).


   RESULTS
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES
 
Co-expression of benP and benK
The orientation of other genes in the vicinity suggested that benP and benK would be transcribed either individually or as a biscistronic operon (Fig. 1). To understand the regulation of gene expression, we first explored the possibility that the two adjacent genes were co-transcribed. Total RNA was isolated from the wild-type strain, ADP1, grown with sole carbon sources that are degraded via the catechol branch of the ß-ketoadipate pathway (benzoate or CCM). In addition, RNA was isolated from ADP1 grown with succinate as the sole carbon source, a growth condition that does not induce expression of genes associated with the ß-ketoadipate pathway.

RT-PCR analysis was carried out with the BENP-FOR primer in benP and the BENK-REV primer in benK (Fig. 2, Table 2). With RNA from ADP1 grown on benzoate or CCM, but not succinate, a 1015 bp product of the size expected for these primers was obtained (Fig. 2b). Control reactions lacking reverse transcriptase did not yield a product, indicating no significant DNA contamination in the RNA samples. RT-PCR products were isolated and digested with XbaI to confirm the presence of a known recognition sequence in benK. Although complete digestion was not always achieved, the sizes of the cleavage products were consistent with the correct fragments having been amplified, as shown for CCM-grown cells in Fig. 2(b), lane 7. Additional RT-PCR experiments with the same BENK-REV primer and a primer closer to the 5' end of benP (BENP2-FOR, Table 2) yielded a product of the expected size with RNA from cells grown on CCM (data not shown). Thus, there were benK transcripts that initiate more than 700 nt upstream of its translational start codon. These results indicate that benP and benK are co-transcribed and that expression of the benPK operon occurs during growth on CCM or benzoate, but not succinate.

Transcript initiation and inducible expression of benPK
Expression of benP was further explored with primer-extension methods. The primer used for these experiments (BENP-PE, Table 2) annealed to a region downstream of the predicted benP AUG start codon. In extension reactions with total RNA from ADP1 grown on benzoate or CCM as the carbon source, two prominent products were generated. Their sizes would correspond to transcripts that initiate at sites 49 and 40 nt upstream of the start codon, labelled A and B in Fig. 3(a). No extension products were detected in reactions with RNA from succinate-grown cells, despite multiple repetitions with independently isolated RNA samples. Spectral analysis of total RNA and electrophoretic detection of rRNA revealed no obvious differences among the RNA samples from cells grown on different carbon sources. Consistent with the RT-PCR results, these studies indicated that benP expression is inducible.



View larger version (34K):
[in this window]
[in a new window]
 
Fig. 3. Primer-extension analysis of benP. (a) RNA was extracted from the wild-type strain ADP1 grown on succinate (SUC), benzoate (BEN) or CCM as a sole carbon source. Lanes C, T, A and G indicate the DNA sequencing ladder. Arrows A and B mark major products of sizes corresponding to transcripts initiating 49 and 40 nucleotides upstream of the predicted benP translational start site. Nucleotides at these positions are highlighted in bold and marked with an asterisk. (b) RNA was analysed from LB-grown mutants. Strain ISA25 has a large cat-region deletion ({Delta}cat) that prevents CCM catabolism and strain ACN32 has an insertion (in benA) that prevents benzoate catabolism. CCM was added to the growth medium of the former and benzoate (BEN) to the latter. Arrows showing major products (A and B) are of identical sizes in panels (a) and (b).

 
Mutants were used to determine whether benzoate or CCM, in the absence of catabolism, could induce gene expression. RNA was isolated from ISA25, a strain with a large cat-gene deletion that prevents CCM catabolism. Primer-extension analysis indicated that benP is expressed when this strain is grown on rich medium in the presence of CCM (Fig. 3b). Therefore, CCM itself, rather than a catabolite generated during growth on CCM, appears to cause increased benP expression. CCM similarly stimulated benP expression in ACN32, which has a large insertion in benA that prevents benzoate catabolism (data not shown). However, when ACN32 was grown on rich medium supplemented with benzoate, no benP expression was detected (Fig. 3b). Consistent with this result, benzoate does not increase expression of a chromosomal benK::lacZ transcriptional fusion in the absence of benzoate catabolism (Collier et al., 1997 ). Therefore, transcription from the benPK promoter, PbP, in benzoate-grown wild-type cells most likely results from the endogenous generation of CCM during benzoate degradation.

Roles of CatM and BenM in regulating PbP
The identification of CCM as an inducer raised the possibility that CatM or BenM can activate transcription from PbP. To test this possibility, benP expression was studied in mutants (listed in Table 1) lacking CatM (strain ISA13), BenM (strain ISA36), or both of these transcriptional regulators (strain ACN9). Total RNA was isolated from the wild-type and from each of these mutants grown on rich medium supplemented with CCM. Primer-extension reactions with the benP primer (BENP-PE, Table 2) indicated that the loss of both transcriptional regulators significantly reduced transcription from PbP (Fig. 4). In contrast, the absence of either CatM or BenM alone had little effect on transcription relative to that in the wild-type strain under the same growth conditions (Fig. 4). These results suggested that both BenM and CatM are individually able to regulate PbP in response to CCM. The overlap of their regulatory capabilities may account for the requirement that both BenM and CatM be absent in order for the loss of regulated gene expression to observed. A similar overlap in regulation by BenM and CatM occurs for the expression of catA from its promoter region PcA (Romero-Arroyo et al., 1995 ).



View larger version (54K):
[in this window]
[in a new window]
 
Fig. 4. Primer-extension analysis of benP in mutants. RNA was extracted from LB-grown wild-type (wt) strain ADP1 and mutant strains ISA13 (CatM-), ISA36 (BenM-) and ACN9 (CatM- BenM-). The growth medium was supplemented with CCM except where noted for the wild-type control. Two major products marked by arrows are of the identical size to those labelled A and B in Fig. 3.

 
Protein purification following expression of catM and benM in E. coli
To study the interactions of BenM and CatM with DNA in the vicinity of PbP in vitro, the two regulators were purified. Expression vectors pBAC381 and pBAC382 (Table 1) were constructed for the production of CatM and BenM, respectively, in E. coli. When protein production was induced at 37 °C, problems were encountered with inclusion body formation. As described in Methods, conditions were identified for purification following protein induction at 16 °C. CatM was purified by cation-exchange and heparin-agarose affinity chromatography. From 1 g dry cell mass, CatM protein (approx. 2 mg total) was obtained with an estimated purity of >95% (Fig. 5a). BenM was similarly purified, yielding approximately 1·5 mg (g dry cell mass)-1 (Fig. 5b). Gel-filtration analysis indicated that the sole oligomeric form of CatM was approximately 140 kDa. Since denaturing SDS-PAGE analysis was consistent with the prediction that the monomeric protein is 35·5 kDa, the CatM protein was inferred to be tetrameric in solution. A similar conclusion was drawn from gel-filtration analysis of BenM (Bundy, 2001 ).



View larger version (56K):
[in this window]
[in a new window]
 
Fig. 5. Purification of CatM (a) and BenM (b) after induction in E. coli. Coomassie-stained 12% SDS-polyacrylamide gel of protein from: lanes 1, soluble extract of uninduced E. coli cells with plasmid-borne catM or benM; lanes 2–4, soluble extract from IPTG-induced E. coli cells with plasmid-borne catM or benM (lanes 2) after cation-exchange (SP) chromatography (lanes 3) and after elution from the heparin agarose column (lanes 4). The positions and molecular masses of protein standards are indicated to the left of each panel. The arrows indicate the approximately 36 kDa CatM or BenM.

 
Binding of CatM and BenM to the benP regulatory region
Based on studies of other LysR-type transcriptional regulators (McFall et al., 1998 ), we predicted that if CatM and BenM activate benP expression, then they should bind to DNA in the region of PbP in the presence or absence of the inducer, CCM. Gel-retardation assays were used to test this prediction. Samples of purified CatM and BenM were incubated with a radiolabelled fragment carrying PbP (Fig. 6). Increasing amounts of either protein increased the proportion of DNA with retarded electrophoretic mobility. To confirm that the changes in mobility patterns resulted from BenM or CatM binding specifically to the PbP region, competition experiments were done with unlabelled DNA (described in Methods). In the presence of 100 ng of either BenM or CatM, reactions containing the unlabelled PbP fragment in 50-fold molar excess to the 32P-labelled probe resulted in the absence of detectable fragments with retarded mobility (data not shown). In contrast, reactions containing the non-specific competitor DNA fragment in 50-fold molar excess did not reduce the ability of CatM or BenM to retard the mobility of the fragments with PbP (data not shown).



View larger version (45K):
[in this window]
[in a new window]
 
Fig. 6. Gel-retardation assays of CatM or BenM bound to the PbP region. Purified CatM (a) or BenM (b) was incubated with the 32P-labelled PbP fragment in the absence of CCM (lanes 1–7) or in presence of 50 mM CCM (lanes 8–14). Lanes 1 and 8 contained no protein. The following amounts of CatM or BenM were used in the binding reactions: lanes 2 and 9, 1000 ng; 3 and 10, 500 ng; 4 and 11, 100 ng; 5 and 12, 50 ng; 6 and 13, 10 ng; 7 and 14, 5 ng. The arrow marked F indicates the mobility of free DNA in the absence of protein. The retarded mobility of fragments at positions labelled C1, C2, and B presumably results from the binding of CatM or BenM to form protein–DNA complexes. The position of a putative binding site for these regulators is schematically indicated by a box in (c). Also indicated are the primers and the position of the 32P-labelled fragment (PbP Probe) used in these studies. The numbering system is that used in Fig. 2.

 
The presence of CCM in the binding reaction with CatM resulted in altered patterns of fragment mobility at the higher protein concentrations (Fig. 6a: compare lanes 2, 3 and 4 with lanes 9, 10 and 11). In the presence of CCM, labelled fragments were observed at the position labelled C2, whereas in the absence of CCM additional fragments were observed at the position labelled C1. In the reactions with BenM, the presence or absence of CCM did not alter the position of the fragments with retarded mobility. However, in the absence of CCM, BenM may have a higher affinity for the PbP probe. Although the significance of the specific patterns is difficult to infer, the ability of the inducer to alter the mobility of DNA–protein complexes has been observed for similar LysR-type regulators (McFall et al., 1998 ; Parsek et al., 1992 ; Schell, 1993 ).

The function of benP
The regulation of PbP by BenM and CatM should coordinate the transcription of the benPK operon with the expression of additional ben and cat genes involved in benzoate degradation (Fig. 1). Therefore, the ability to degrade benzoate and related compounds was tested in ACN450, a strain in which the chromosomal copy of benP was disrupted by insertional inactivation (Table 1, Methods). ACN450 was capable of growing on solid medium using benzoate, benzaldehyde, benzyl alcohol, benzyl acetate, ethyl salicylate, salicylate or anthranilate as the sole carbon source (data not shown). With these substrates, possible differences in growth rates between ACN450 and the wild-type strain were not characterized. The mutation in ACN450 should also prevent expression of benK, a gene known to affect the rate of growth on some of these compounds (Collier et al., 1997 ). These results indicate that neither BenP nor BenK is essential for the cellular entry of this set of aromatic compounds under the laboratory conditions tested. Different experimental conditions, different substrates and/or different genetic backgrounds may be needed to reveal the phenotypic effects caused by the loss of BenK and BenP.

Initial analysis of BenP detected relatively low, but significant, sequence identity in pairwise alignments with known porins. However, computer-based programs failed to optimize the alignment of BenP and other proteins throughout their entire lengths. As shown in Fig. 7, generated by a combination of computer-based methods and hand-made alignments, there are segments of BenP with strong sequence similarity to known or putative porins that are interspersed with regions, of variable size, containing little or no identity to the other sequences. According to the transporter classification (TC) system of Saier, BenP and several other putative porins involved in aromatic compound transport form a family that includes the OprD porin of Pseudomonas aeruginosa (TC# 1.B.25; http://tcdb.ucsd.edu/tcdb/search2.php) (Saier, 2000 ). The substrates of OprD include cationic amino acids, peptides and an analogous compound, the antibiotic imipenem (Trias & Nikaido, 1990 ).



View larger version (47K):
[in this window]
[in a new window]
 
Fig. 7. Modified alignment of the deduced BenP protein sequence with the sequences of known or putative porins from Acinetobacter and Pseudomonas. Boxes A–F display the regions of greatest similarity and asterisks mark residues (shown in bold) identical in all sequences. These regions are connected by less conserved stretches of amino acids (aa) of variable length. The total number of amino acids is shown parenthetically. In the VanORF and BenF sequences there are insertions of 10 and 4 amino acids, respectively, that do not resemble those aligned in other peptide sequences. Sequences in OprD that may form transmembrane ß-strands are underlined and numbered to correspond with the 16-ß-stranded OprD topology model of Huang et al. (1995) that was recently modified (Ochs et al., 2000 ). According to this model, short loops between strands 2 and 3 (box B), 4 and 5 (box C), 10 and 11 (box D) are in the periplasmic space, and the residues of box A are part of the first of eight loops external to the outer membrane. The GenBank accession numbers for each sequence are as follows: OprE1 (D12711), OprE3 (AB006797), OprD3 (AF033849), VanORF (AF009672), PhaK (AF029714), BenF (AAF63454), BenP (AF150928) and OprD (X63152).

 
No protein structure has yet been determined for a BenP-like porin, although various bacterial porins have a ß-barrel structure (Koebnik et al., 2000 ). A 16 ß-strand topology model proposed for OprD has been tested by deletion mutagenesis (Huang et al., 1995 ; Ochs et al., 2000 ). The predicted OprD ß-strands with sequences similar to those of BenP are underlined in Fig. 7. Furthermore, 23 amino acid residues that are conserved among all sequences in the alignment, marked by asterisks, are likely to have a role in protein structure or function. Proteins in the alignment that may be involved in aromatic compound catabolism include PhaK and BenF of Pseudomonas putida. PhaK was inferred to form an outer-membrane channel with narrow substrate specificity since inactivation of its gene prevents the use as a sole carbon source of phenylacetate but not 4-hydroxyphenylacetate or additional phenylacetate-related compounds (Olivera et al., 1998 ). A predicted role for BenF in aromatic compound catabolism is based on the co-expression of benF with several genes in an operon for benzoate dissimilation (Cowles et al., 2000 ).


   DISCUSSION
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES
 
Regulation of transcription initiation
A region of transcription initiation for the benPK operon was identified (Figs 3 and 4). Alignment of this PbP region with other CatM- and BenM-controlled promoters helped identify potential regulatory sequences (Fig. 1c). Recent studies with DNase I footprinting demonstrated that CatM and BenM bind strongly to the PbA region, which controls expression of the divergently transcribed benM and benA genes. CatM and BenM each recognize a consensus LysR-type binding sequence in PbA (T-11 nt-A, with a small region of dyad symmetry), boxed in Fig. 1(c) (Collier, 2000 ). An identical sequence to that in PbA, ATAC-7 nt-GTAT, is recognized by CatM in the PcB region, which controls expression of the divergently transcribed catM and catB genes (Romero-Arroyo et al., 1995 ). Similar sequences are recognized by a subfamily of LysR-type transcriptional regulators involved in bacterial aromatic compound catabolism, to which CatM and BenM belong (Coco et al., 1994 ). Therefore, the boxed regions of PbP and PcA are also likely to serve as BenM and CatM binding sites (Fig. 1c). Furthermore, six additional nucleotides, which are highlighted in Fig. 1(c), are identical in all these promoter regions and occur in locations shown to be important for the regulation of PbA (Bundy, 2001 ).

This sequence alignment places the previously identified transcriptional start sites of the benA, catA and catB genes in an identical position (+1) (Fig. 1c) (Collier, 2000 ; Romero-Arroyo et al., 1995 ). This position of benP would correspond to a primer-extension product of the size denoted as ‘A’ in Figs 3 and 4. Thus, the placement of putative regulatory sequences and the observation that ‘A’ is the largest observed primer-extension product suggest that benP transcription initiates at a position corresponding to that of the other CatM- and BenM-regulated genes. The multiple products in the benP primer-extension reactions may result from RNA secondary structure, RNA processing and/or degradation, multiple transcription initiation sites, or some combination of these factors. The significance of the experimental variation in relative intensities of the products is not evident. Nevertheless, the presence of the primer-extension products clearly indicated the conditions under which benP is expressed.

The regulon controlled by CatM and BenM
CCM caused transcription from PbP to increase in strains with CatM or BenM (Figs 3 and 4). The absence of both regulators significantly reduced benP expression, although the resultant expression was higher than that of the wild-type strain grown in the absence of CCM (Fig. 4). Similarly, in a mutant lacking both BenM and CatM, the expression of benA is approximately fourfold higher than in strains with the regulators when no inducer is present (Cosper et al., 2000 ). This increased expression results from the ability of CatM or BenM, in the absence of inducers, to bind to a segment of PbA and repress basal benA expression by preventing access to RNA polymerase (Bundy, 2001 ). BenM and CatM, which are able to bind to the PbP region in the absence of CCM (Fig. 6), may similarly repress basal benP expression.

The gel-retardation studies indicate that CatM can form at least two different complexes with the PbP region. The presence of CCM may alter the interactions between the regulatory proteins and one or more binding site in this region. Multiple binding sites for BenM and CatM have been demonstrated in the PbA region (Bundy, 2001 ). Multiple binding sites in the target genes have also been identified in Pseudomonas putida for the CatR and ClcR regulators, which control catechol and chlorocatechol degradation, respectively (McFall et al., 1998 ). Based on structural studies of transcriptional regulators, the putative CatM/BenM binding site in the PbP region, boxed in Fig. 1(c), should bind a protein dimer (Branden & Tooze, 1999 ). However, BenM and CatM, like many LysR-type regulators, were tetrameric in solution (Schell, 1993 ). Additional studies are needed to determine if the presence of CCM affects the oligomeric structure of CatM or BenM and to characterize the specific binding of these regulators to the PbP region.

While CatM and BenM both regulate transcription from PbP, PbA, PcA and PcB (depicted in Fig. 1), the specific regulation at each locus varies considerably. For example, benzoate interacts with BenM to activate transcription from PbA, and most likely from PcA, as well (Collier et al., 1998 ; Romero-Arroyo et al., 1995 ). In contrast, despite the ability of BenM to regulate transcription from PbP, benzoate did not serve as an inducer in the absence of its conversion to CCM (Figs 3 and 4). Regulated expression from PbP appeared to be mediated equally well by CatM or BenM in response to CCM (Fig. 4). Similarly, CCM enables CatM and BenM each to activate high-level transcription from PcA (Romero-Arroyo et al., 1995 ). In contrast, at PbA or PcB, BenM or CatM, respectively, acts as the principal regulator of the expression of a multi-gene operon (Collier et al., 1998 ; Romero-Arroyo et al., 1995 ). Thus CatM and BenM control the expression of numerous genes with related functions. The demonstration in this report that the benPK operon is part of this complex regulon strongly suggests a role for BenP and BenK in the degradation of aromatic compounds.

Transporters in aromatic compound catabolic pathways
The proposed porin functions have yet to be demonstrated for BenP or for similar proteins likely to be involved in aromatic compound uptake. However, open reading frames encoding hypothetical bacterial porins are frequently located in the vicinity of genes known to participate in aromatic compound degradation. For example, two different BenP/PhaK-like hypothetical proteins of Pseudomonas sp. strain CA10 are encoded in genetic regions involved in the degradation of the heterocyclic aromatic compound carbazole via the catechol branch of the ß-ketoadipate pathway (Nojiri et al., 2001 ). Moreover, the putative porin-encoding genes are often located near genes encoding proteins resembling inner-membrane permeases such as BenK, PcaK and MucK that are involved in the uptake of compounds degraded via the ß-ketoadipate pathway (Nichols & Harwood, 1997 ; Williams & Shaw, 1997 ). Taken collectively, the data support a model in which a porin, such as BenP, and a permease, such as BenK, function together to facilitate the cellular entry of some substrates of bacterial aromatic compound catabolic pathways.


   ACKNOWLEDGEMENTS
 
We are grateful to Drs Lauren S. Collier, Becky M. Bundy and James E. Posey for their contributions to protein purification and primer-extension analysis. We also thank Drs Frank Gherardini, Timothy R. Hoover, Anna C. Glasgow Karls and Mark A. Schell for many helpful discussions. This research was supported by NSF grant MCB-9808784 (to E.L.N).


   REFERENCES
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES
 
Bradford, M. M. (1976). A rapid and sensitive method for the quantitation of microgram quantities of protein utilizing the principle of protein-dye binding. Anal Biochem 72, 248-254.[Medline]

Branden, C. & Tooze, J. (1999). Introduction to Protein Structure, 2nd edn. New York: Garland Publishing.

Brzostowicz, P. C. (1997). Carbon-source dependent expression of the pobA gene needed for 4-hydroxybenzoate degradation in Acinetobacter sp. strain ADP1. MS thesis, University of Georgia.

Bundy, B. M. (2001). Transcriptional regulation of the benABCDE operon of Acinetobacter sp. strain ADP1: BenM-mediated synergistic induction in response to benzoate and cis,cis-muconate. PhD thesis, University of Georgia.

Coco, W. M., Parsek, M. R. & Chakrabarty, A. M. (1994). Purification of the LysR family regulator, ClcR, and its interaction with the Pseudomonas putida clcABD chlorocatechol operon promoter. J Bacteriol 176, 5530-5533.[Abstract]

Collier, L. S. (2000). Transcriptional regulation of benzoate degradation by BenM and CatM in Acinetobacter sp. strain ADP1. PhD thesis, University of Georgia.

Collier, L. S., Nichols, N. N. & Neidle, E. L. (1997). benK encodes a hydrophobic permease-like protein involved in benzoate degradation by Acinetobacter sp. strain ADP1. J Bacteriol 179, 5943-5946.[Abstract]

Collier, L. S., Gaines, G. L.III & Neidle, E. L. (1998). Regulation of benzoate degradation in Acinetobacter sp. strain ADP1 by BenM, a LysR-type transcriptional activator. J Bacteriol 180, 2493-2501.[Abstract/Free Full Text]

Cosper, N. J., Collier, L. S., Clark, T. J., Scott, R. A. & Neidle, E. L. (2000). Mutations in catB, the gene encoding muconate cycloisomerase, activate transcription of the distal ben genes and contribute to a complex regulatory circuit in Acinetobacter sp. strain ADP1. J Bacteriol 182, 7044-7052.[Abstract/Free Full Text]

Cowles, C. E., Nichols, N. N. & Harwood, C. S. (2000). BenR, a XylS homologue, regulates three different pathways of aromatic acid degradation in Pseudomonas putida. J Bacteriol 182, 6339-6346.[Abstract/Free Full Text]

DiMarco, A. A., Averhoff, B. & Ornston, L. N. (1993). Identification of the transcriptional activator pobR and characterization of its role in the expression of pobA, the structural gene for p-hydroxybenzoate hydroxylase in Acinetobacter calcoaceticus. J Bacteriol 175, 4499-4506.[Abstract]

Gaines, G. L.III, Smith, L. & Neidle, E. L. (1996). Novel nuclear magnetic resonance spectroscopy methods demonstrate preferential carbon source utilization by Acinetobacter calcoaceticus. J Bacteriol 178, 6833-6841.[Abstract]

Gregg-Jolly, L. A. & Ornston, L. N. (1990). Recovery of DNA from the Acinetobacter calcoaceticus chromosome by gap repair. J Bacteriol 172, 6169-6172.[Medline]

Harwood, C. S. & Parales, R. E. (1996). The ß-ketoadipate pathway and the biology of self-identity. Annu Rev Microbiol 50, 553-590.[Medline]

Huang, H., Jenateur, D., Pattus, F. & Hancock, R. E. W. (1995). Membrane topology and site-specific mutagenesis of Pseudomonas aeruginosa porin OprD. Mol Microbiol 16, 931-941.[Medline]

Jones, R. M. & Williams, P. A. (2001). areCBA is an operon in Acinetobacter sp. strain ADP1 and is controlled by AreR, a {sigma}54-dependent regulator. J Bacteriol 183, 405-409.[Abstract/Free Full Text]

Jones, R. M., Collier, L. S., Neidle, E. L. & Williams, P. A. (1999). areABC genes determine the catabolism of aryl esters in Acinetobacter sp. strain ADP1. J Bacteriol 181, 4568-4575.[Abstract/Free Full Text]

Jones, R. M., Pagmantidis, V. & Williams, P. A. (2000). sal genes determining the catabolism of salicylate esters are part of a supraoperonic cluster of catabolic genes in Acinetobacter sp. strain ADP1. J Bacteriol 182, 2018-2025.[Abstract/Free Full Text]

Juni, E. & Janik, A. (1969). Transformation of Acinetobacter calcoaceticus (Bacterium anitratum). J Bacteriol 98, 281-288.[Medline]

Keen, T., Tamaki, S., Kobayashi, D. & Trollinger, D. (1988). Improved broad-host-range plasmids for DNA cloning in Gram-negative bacteria. Gene 70, 191-197.[Medline]

Koebnik, R., Locher, K. P. & Van Gelder, P. (2000). Structure and function of bacterial outer-membrane proteins: barrels in a nutshell. Mol Microbiol 37, 239-253.[Medline]

McFall, S. M., Chugani, S. A. & Chakrabarty, A. M. (1998). Transcriptional activation of the catechol and chlorocatechol operons: variations on a theme. Gene 223, 257-267.[Medline]

Neidle, E. L., Hartnett, C. & Ornston, L. N. (1989). Characterization of Acinetobacter calcoaceticus catM, a repressor gene homologous in sequence to transcriptional activator genes. J Bacteriol 171, 5410-5421.[Medline]

Nichols, N. N. & Harwood, C. S. (1997). PcaK, a high-affinity permease for the aromatic compounds 4-hydroxybenzoate and protocatechuate from Pseudomonas putida. J Bacteriol 179, 5056-5061.[Abstract]

Nojiri, H., Sekiguchi, H., Maeda, K., Urata, M., Nakai, S., Yoshida, T., Habe, H. & Omori, T. (2001). Genetic characterization and evolutionary inplications of a car gene cluster in the carbazole degrader Pseudomonas sp. strain CA10. J Bacteriol 183, 3663-3679.[Abstract/Free Full Text]

Ochs, M. M., Bains, M. & Hancock, R. E. W. (2000). Role of putative loops 2 and 3 in imipenem passage through the specific porin OprD of Pseudomonas aeruginosa. Antimicrob Agents Chemother 44, 1983-1985.[Abstract/Free Full Text]

Olivera, E. R., Minambres, B., Garcia, B., Muniz, C., Moreno, M. A., Ferrandez, A., Diaz, E., Garcia, J. L. & Luengo, J. M. (1998). Molecular characterization of the phenylacetic acid catabolic pathway in Pseudomonas putida U: the phenylacetyl-CoA catabolon. Proc Natl Acad Sci USA 95, 6419-6424.[Abstract/Free Full Text]

Parsek, M. R., Shinabarger, D. L., Rothmel, R. K. & Chakrabarty, A. M. (1992). Roles of CatR and cis,cis-muconate in activation of the catBC operon, which is involved in benzoate degradation in Pseudomonas putida. J Bacteriol 174, 7798-7806.[Abstract]

Parsek, M. R., Coco, W. M. & Chakrabarty, A. M. (1994). Gel-shift assay and Dnase I footprinting analysis of transcriptional regulation of biodegradation genes. Methods Mol Genet 3, 273-290.

Prentki, P. & Krisch, H. M. (1984). In vitro insertional mutagenesis with a selectable DNA fragment. Gene 29, 303-313.[Medline]

Romero-Arroyo, C. E., Schell, M. A., Gaines, G. L.III & Neidle, E. L. (1995). catM encodes a LysR-type transcriptional activator regulating catechol degradation in Acinetobacter calcoaceticus. J Bacteriol 177, 5891-5898.[Abstract]

Saier, M. H. (2000). Families of proteins forming transmembrane channels. J Membr Biol 175, 165-180.[Medline]

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

Schell, M. A. (1993). Molecular biology of the LysR family of transcriptional regulators. Annu Rev Microbiol 47, 597-626.[Medline]

Segura, A., Bunz, P. V., D’Argenio, D. A. & Ornston, L. N. (1999). Genetic analysis of a chromosomal region containing vanA and vanB, genes required for conversion of either ferulate or vanillate to protocatechuate in Acinetobacter. J Bacteriol 181, 3494-3504.[Abstract/Free Full Text]

Shanley, M. S., Neidle, E. L., Parales, R. E. & Ornston, L. N. (1986). Cloning and expression of Acinetobacter calcoaceticus catBCDE genes in Pseudomonas putida and Escherichia coli. J Bacteriol 165, 557-563.[Medline]

Tobiason, D. M., Lenich, A. G. & Glasgow, A. C. (1999). Multiple DNA binding activities of the novel site-specific recombinase, Piv, from Moraxella lacunata. J Biol Chem 274, 9698-9706.[Abstract/Free Full Text]

Trias, J. & Nikaido, H. (1990). Protein D2 channel of the Pseudomonas aeruginosa outer membrane has a binding site for basic amino acids and peptides. J Biol Chem 265, 15680-15684.[Abstract/Free Full Text]

Williams, M. G. & Rogers, M. (1987). Expression of the arg genes of Escherichia coli during arginine limitation dependent upon stringent control of translation. J Bacteriol 169, 1644-1650.[Medline]

Williams, P. A. & Shaw, L. E. (1997). mucK, a gene in Acinetobacter calcoaceticus ADP1 (BD413) encodes the ability to grow on exogenous cis,cis-muconate as sole carbon source. J Bacteriol 179, 5935-5942.[Abstract]

Yanisch-Perron, C., Vieira, J. & Messing, J. (1985). Improved M13 phage cloning vectors and host strains: nucleotide sequences of the M13mp18 and pUC19 vectors. Gene 33, 103-119.[Medline]

Young, D. M., Parke, D., D’Argenio, D. A., Smith, M. A. & Ornston, L. N. (2001). Evolution of a catabolic pathway. ASM News 67, 362-369.

Received 27 September 2001; revised 21 November 2001; accepted 23 November 2001.