INTRODUCTION

Top Abstract Introduction Materials and Methods Results Discussion References

The expression of bacterial catabolic pathways for aromatic compounds is often controlled by one or more regulatory proteins, and the effectors of these regulatory proteins are usually either the initial substrates or catabolic intermediates of the pathways (35). The acquisition of an efficient transcriptional regulation system appears to be a major asset for a microorganism which enables it to thrive in an ever-changing environment, as the constitutively expressed pathway enzymes may impose an energy burden.

The expression of multicomponent phenol hydroxylase (mPH) (14, 19, 20, 31, 32, 49, 50) is thought to be controlled by a regulator of the XylR/DmpR subclass within the NtrC-type family of transcriptional regulators, resulting in the expression of phenol-metabolizing enzymes only in the presence of the pathway substrates or structural analogs (2, 20, 24, 30, 31, 36, 38, 42-45, 50). The regulators of this subclass are activated by direct interaction with an effector molecule which is normally the substrate for the catabolic pathway the regulators control (41). The physiological status of regulation is also thought to be overimposed on the system of the XylR/DmpR regulator-⁵⁴-dependent promoter (7, 8, 13, 22, 23, 26, 47, 48). Another type of transcriptional regulator for the expression of mPH, belonging to the GntR family of transcriptional regulators, has been identified in Comamonas testosteroni TA441 (1). This regulator, named AphS, repressed the transcription of mPH genes and caused the inability of the bacterium to grow on phenol (1).

C. testosteroni R5 has been shown to exhibit high activity for both phenol oxygenation (55) and trichloroethylene degradation (18). We have cloned the DNA fragment encoding mPH (PhcKLMNOP) and its cognate transcriptional activator (PhcR) of the XylR/DmpR subclass from strain R5 (50). Our previous work (50) and sodium dodecyl sulfate-polyacrylamide gel electrophoretic analysis (unpublished data) indicated that the high activity of strain R5 was due to the high level of Phc mPH expression, leading us to investigate its transcriptional machinery. In this study, we found one open reading frame (phcS) downstream of phcR. The physiological role of PhcS on the expression of phenol-metabolizing enzymes in strain R5 was studied.

	MATERIALS AND METHODS

Top Abstract Introduction Materials and Methods Results Discussion References

Bacterial strains, plasmids, and oligonucleotides. The bacterial strains, plasmids, and oligonucleotides used in this study are listed in Table 1.

Media and growth conditions. The culture media used in this study were Luria-Bertani (LB) medium (37), M9 medium (3), and an inorganic medium called MP (50). The Escherichia coli strains were grown at 37°C, while the C. testosteroni and Pseudomonas strains were grown at 30°C, unless stated otherwise. When required, the media were supplemented with the following antibiotics at the indicated concentrations: tetracycline (TET), 12 µg/ml; ampicillin, 100 µg/ml; carbenicillin, 500 µg/ml; kanamycin (KAN), 30 µg/ml (E. coli) or 400 µg/ml (C. testosteroni); and chloramphenicol (CHL), 20 µg/ml (E. coli) or 80 µg/ml (C. testosteroni).

Genetic techniques. Plasmid isolation, restriction endonuclease digestion, and transformation of the E. coli strains were conducted by the methods of Sambrook et al. (37). The Pseudomonas aeruginosa and C. testosteroni strains were transformed by the method of Chakrabarty et al. (9).

Nucleotide sequencing and computer analysis. To determine the nucleotide sequence of the 0.8-kb SalI-Sse8387I fragment downstream of phcR (Fig. 1), subfragments were cloned into the multicloning site of pBluescript II KS() (Toyobo). The nucleotide sequences of the subfragments were determined in both orientations by using M13 primers (Takara), a DNA sequencing kit (Dye Terminator Cycle Sequence; Perkin-Elmer), and a model 377 DNA sequencer (Perkin-Elmer) according to the manufacturers' instructions. The templates for the dideoxy chain-termination reaction mixtures were prepared by using Wizard minipreps (Promega). The DNA sequence data were aligned by using version 1.7 of CLUSTAL W (51).

View larger version (26K): [in this window] [in a new window]   FIG. 1.   Genetic organization of the regulatory and structural genes for mPH in C. testosteroni R5. phcS (in black) was identified in this study. The other genes were identified in our previous study (50). The arrows indicate the direction of transcription. The ability (+) or inability () of plasmids pROR502 and pROR504 to allow P. putida PAO1c to grow on phenol as the sole carbon source is indicated. pKT231S is a derivative of pKT231 and carries a MunI-EcoRV fragment which contains the phcS gene, while pRC50Ps is a derivative of pRC50 and carries an Sse8387I-NheI fragment which contains phcR. pRC50Pk is also a derivative of pRC50 and carries an EcoRV-BglII fragment which contains a phcK promoter region. The two small arrows indicate the direction of lacZ on plasmids pRC50Ps and pRC50Pk. pSK02S carries the phcS gene which was disrupted by the insertion of a tetracycline resistance gene (Tcr) cassette, while pBS02RS carries the phcS and phcR genes which were disrupted by the insertion of a Tcr cassette.

Construction of the phcS and phcRS knockouts. An appropriate 3.9-kb KpnI-SmaI DNA fragment containing phcS on pROR501 (Fig. 1) was first subcloned into the multiple cloning site of pMT5059 (pSK1). The phcS gene was disrupted by inserting a 1.7-kb EcoRV fragment of pMT5056, which carried a Tc^r gene, into the blunted ApaI-SacI site of pSK1 (pSK01S). A NotI fragment containing the mobilization cassette of pMT5071 was subsequently inserted into pSK01S. The plasmid thus constructed, pSK02S, was conjugally mobilized (10) from E. coli S17-1 to strain R5, and Tc^r selection was done on an M9 agar plate containing 200 mg of phenol per liter, 5% (wt/vol) sucrose, and TET. The transconjugants were chosen for their sensitivity to carbenicillin, and their chromosomal DNAs were analyzed by PCR to confirm that gene replacement had occurred (data not shown).

Continuous culture conditions. The culture and sampling conditions were as described by Teramoto et al. (50), unless stated otherwise. The culture (working volume of 1 liter) was started at 25°C by inoculating cells grown in 100 ml of an LB medium into 1 liter of MP medium. The MP medium containing 1,000 mg of phenol per liter or 2,615 mg of sodium acetate per liter as the sole carbon source was then continuously supplied at the rate of 0.35 liter per day. For C. testosteroni P1 cells transformed with a pKT231 derivative, all media were supplemented with KAN.

Assay for phenol-oxygenating activity. The phenol-oxygenating activity (oxygen uptake rate) was measured at 25°C with a Clark-type oxygen electrode (5/6 Oxygraph; Gilson) as described previously (50). The activity, which was measured in the presence of 10 mM potassium cyanide following the addition of phenol (final concentration of 10 µM), represents the amount of oxygen consumed equally by PH and catechol 2,3-dioxygenase (C23DOase). A previous study had indicated that the phenol-oxygenating activity was double the PH activity, showing that the activity of C23DOase is higher than that of PH (55). The cell weight (dry weight) was determined as described previously (50). Cells from a continuous culture were sampled immediately before the activity was measured.

Assay for C23DOase activity. Cells (about 50 ml) from a continuous culture were assayed for C23DOase activity. They were collected by centrifugation and washed with 10 ml of ED buffer (10 mM ethylendiamine-H₂SO₄ [pH 7.5] and 10% isopropyl alcohol) before being stored at 80°C. They were then resuspended in 1 ml of the ED buffer and disrupted by sonication (Sonifier 250; Branson). The crude extract was centrifuged at 14,000 × g for 5 min at 4°C, and the resulting supernatant was used for the enzyme assay. This assay was performed by using a 100 mM potassium phosphate buffer (pH 7.4) at 25°C with 330 µM catechol as a substrate. The amount of the ring cleavage product from catechol (2-hydroxymuconic semialdehyde; the molar absorption coefficient at 375 nm is 33,000) was determined spectrophotometrically (33). The protein concentration was determined by the method of Bradford (5) with a protein assay kit (Bio-Rad), using bovine serum albumin as the standard.

Induction experiment. The expression of mPH under batch culture conditions was examined next. C. testosteroni strains were grown in LB medium to the stationary phase, harvested, washed with MP medium, resuspended in the original culture volume of MP medium, and exposed to different substrates, phenol or acetate, at a final concentration of 2 mM at 30°C for the indicated periods of time while shaking at 100 rpm. Before the phenol-oxygenating activity was measured (see above), the culture was washed with MP medium and then resuspended in the same medium. C. testosteroni strains transformed with a pKT231 derivative, a pMMK67HE derivative, or a pRC50 derivative were grown in LB medium containing KAN (for the pKT231 and pMMK67HE derivatives) or CHL (for the pRC50 derivative) to the stationary phase and then washed and suspended in MP medium containing phenol or acetate as described above. After each experiment, maintenance of the plasmid was checked by using nonselective and selective plates (supplemented with KAN or CHL).

Other methods. A growth test on P. aeruginosa PAO1c derivatives containing pROR502 or pROR504 was performed at 25°C with MP medium supplemented with 200 mg of phenol per liter as the sole carbon source, and the optical density at 660 nm of the culture was monitored by an automated recording system (TN2612; Advantec). The concentration of phenol in each culture medium was measured with a kit (color reagent tablets of Phenol-test Wako; Wako Pure Chemical Industries, Osaka, Japan) as described previously (50). The activity of -galactosidase, the lacZ gene product, was determined by the protocol described by Miller (27).

Nucleotide sequence accession number. The nucleotide sequence of the 0.8-kb SalI-Sse8387I region was deposited in the DDBJ/EMBL/GenBank database under accession no. AB050891.

	RESULTS

Top Abstract Introduction Materials and Methods Results Discussion References

Identification of the phcS gene. We analyzed the DNA region downstream of phcR cloned in pLAFRR501 (50). The sequencing of the 0.8-kb SalI-Sse8387I region identified a 732-bp open reading frame, named phcS, preceded by a putative Shine-Dalgarno sequence (39) (Fig. 1). The phcS gene was immediately downstream of phcR, with only 22 bp between the two genes. The phcS gene encodes a protein of 244 amino acid residues with a predicted molecular mass of 27 kDa. The deduced product has 96% identity and 98% similarity to AphS, a transcriptional repressor for the expression of mPH in C. testosteroni TA441 (1). PhcS also shows 15% identity and 52% similarity to negative transcriptional regulator GntR of Bacillus subtilis (15, 16, 28). A possible helix-turn-helix (HTH) motif that is likely to be involved in DNA binding was thought to be located in the N-terminal region of the sequence (amino acid residues 43 to 64) (12). Members of the GntR family of transcriptional regulators have the putative HTH DNA-binding motif at their N terminus (11), and the residues conserved in the N-terminal domains of GntR-like proteins shown by Mouz et al. (29) were well conserved in the PhcS sequence, suggesting that the phcS gene product belongs to the GntR family of transcriptional regulators. Two AphS-binding sequences reported by Arai et al. (1) were identified in the corresponding regions in and around the phcR-phcK intervening sequence, indicating that PhcS may also bind to these two sites.

Analysis of phcS in PAO1c cells. C. testosteroni TA441 has been reported to be unable to grow on phenol, as AphS represses transcription of the aph genes. In contrast, C. testosteroni R5 can grow on phenol and shows high phenol-oxygenating activity even in the presence of phcS. To investigate whether phcS encodes such a transcriptional repressor, the region encoding PhcS was deleted from the DNA fragment coding for mPH, and plasmids carrying the deleted and undeleted fragments (pROR502 and pROR504) were introduced into P. aeruginosa PAO1c (phenol-negative, catechol-positive) cells (Fig. 1). The PAO1c derivative with phcS was unable to grow on phenol. These data and the observations with AphS suggest that PhcS caused transcriptional repression of the phc mPH genes in the presence of phenol. Consequently, strain R5 should involve a mechanism to overcome the PhcS-mediated repression, as this strain expressed mPH in the presence of phenol even when a functional phcS gene was present.

Repression by PhcS of the gratuitous expression of phenol-metabolizing enzymes. We constructed a phcS knockout of C. testosteroni strain R5 (R5S) to examine the physiological role of phcS. R5S and parental strain R5 were continuously cultured with phenol or acetate as the sole carbon source, and the phenol-oxygenating activity and C23DOase activity of the cultures were measured (Fig. 2). When the R5S strain was grown on phenol, its phenol-oxygenating activity was the same as that of strain R5. However, although acetate-grown R5 cells did not show any detectable phenol-oxygenating activity, acetate-grown R5S cells showed 66% of the activity of the phenol-grown cells.

View larger version (44K): [in this window] [in a new window]   FIG. 2.   Effect of disrupting phcS on expression of mPH in C. testosteroni strain R5 grown on phenol or acetate in a continuous culture. (A) Phenol-oxygenating activity of strains R5 and R5S. (B) C23DOase activity of strains R5 and R5S. Each value is the mean ± standard error from two or three independent cultures.

View larger version (20K): [in this window] [in a new window]   FIG. 3.   mPH expression of C. testosteroni strain R5S in response to the presence of phenol or acetate. Stationary-phase cultures grown on LB medium were exposed to MP medium with 2 mM phenol (circles) or acetate (triangles) or without growth substrate (squares). Each value is the mean ± standard error from three independent experiments.

View larger version (23K): [in this window] [in a new window]   FIG. 4.   Transcriptional activity of the phcK promoter in C. testosteroni strain R5S in response to the presence of phenol or acetate. The strain was transformed with pRC50Pk, which carries a transcriptional fusion of phcKL::lacZ (white symbols) or pRC50 (control vector [black symbols]). Stationary-phase cultures were transferred to MP medium with 2 mM phenol (circles) or acetate (triangles) or without growth substrate (squares). Each value is the mean ± standard error from two or three independent experiments.

Requirement of PhcR for the gratuitous expression. To investigate the involvement of PhcR in the gratuitous expression of the Phc phenol-metabolizing enzymes in the phcS knockout, a phcRS knockout of C. testosteroni R5 (R5RS) and the knockout complemented by phcR in trans, R5RS(pRC50Ps), were both constructed (Table 2). Cells of R5RS(pRC50) containing a control vector did not express any phenol-oxygenating activity after being incubated with acetate, while cells of R5RS(pRC50Ps) incubated with acetate expressed a higher level of phenol-oxygenating activity than R5S did. This high level of activity is thought to be due to a copy number effect of phcR. These results show that the PhcR protein was essential for the gratuitous expression of the Phc phenol-metabolizing enzymes.

1

1

Phenol-oxygenating activity in a chemostat culture with acetate and with other phenol-degrading bacteria. The phenol-oxygenating activity during growth on acetate was also examined with other phenol-degrading bacteria, Pseudomonas putida CF600 and C. testosteroni P1. Strain CF600 is a well-known phenol-degrading bacterium in which no gene homologous to phcS has been found. CF600 and P1 cells were continuously cultured in a chemostat of phenol or acetate as the sole carbon source, and the cultures were assayed. No phenol-oxygenating activity was detected in strain CF600 when it was grown on acetate, and the activity of the phenol-grown CF600 cells was 15 µmol min¹ g (dry weight) of cells¹. On the other hand, the phenol-oxygenating activity of strain P1 grown on acetate, 2 µmol min¹ g (dry weight) of cells¹, was 11% of that of the phenol-grown cells, 18 µmol min¹ g (dry weight) of cells¹.

Complementation of strain R5S with aphS and of strain P1 with phcS. To investigate the functional equivalence of PhcS and AphS, C. testosteroni strains R5S and P1 were complemented with either aphS or phcS. Strains R5S and P1 were transformed with either a plasmid carrying aphS (pHAK1102) or its control plasmid (pMMK67HE), and the growth rates of P1(pHAK1102) and R5S(pHAK1102) on phenol were significantly slower than those of P1(pMMK67HE) and R5S(pMMK67HE), respectively (data not shown). Strains R5S and P1 were then transformed with either a plasmid carrying phcS (pKT231S) or its control plasmid (pKT231), and the growth rates of R5S(pKT231S) and P1(pKT231S) on phenol were not significantly different from those of R5S(pKT231) and P1(pKT231), respectively (data not shown). These results suggest that AphS may have had a strong repressive effect on the transcription of the phenol-metabolizing operon even in the presence of phenol and that PhcS may have had no apparent effect on this transcription in C. testosteroni strains growing on phenol.

1

1

	DISCUSSION

Top Abstract Introduction Materials and Methods Results Discussion References

In this study we demonstrated that phcS encodes a functional transcriptional repressor of the GntR family for the phc phenol-metabolizing operon. Binding of GntR to the operator region of the gnt operon, which is responsible for gluconate metabolism, resulted in negative regulation, and this binding was specifically inhibited in vitro only by the substrates of the pathway, such as gluconate and glucono--lactone (28). However, the activity of PhcS was not affected by phenol, as PAO1c(pROR504) was unable to grow on phenol. A comparison between the data with P. aeruginosa PAO1c and C. testosteroni R5 indicated that R5 incorporated an unknown mechanism by which the PhcS-mediated repression was overcome in the presence of the genuine substrate for Phc mPH. C. testosteroni TA441 described by Arai et al. (1) seems to have also incorporated such a mechanism, as C. testosteroni P1(pKT231S) was able to grow on phenol at a rate similar to the growth rate of P1(pKT231). However, this mechanism in TA441 may have been blocked by the presence of amino acid residues in AphS which are different from those present in PhcS, which may have resulted in the inability of TA441 to grow on phenol.

Our results show that the PhcS protein was important in repressing the gratuitous expression of the Phc phenol-metabolizing enzymes which occurs in minimal media. This type of expression did not occur in LB (rich) medium (at 0 h [Fig. 3]), showing that the expression was subject to regulation that was determined by the physiological status. It is possible that some amino acids contained in the LB medium may have triggered the repression, as has been seen in the inducer-activated XylR-⁵⁴-dependent Pu and Ps promoter systems (26).

Regulators of the XylR/DmpR subclass have been thought to be activated by direct interaction with a substrate for the catabolic pathway they control, resulting in the expression of the pathway enzymes only in the presence of the substrate or a structural analog (36, 41-45). However, our data indicate that PhcR of the XylR/DmpR subclass caused the expression of the phenol-metabolizing operon in both the presence and absence of the genuine substrate and that the gratuitous expression which occurred in the absence of the genuine substrate was repressed by PhcS. C. testosteroni strain P1 which did not possess the PhcS-like protein (AphS) also demonstrated gratuitous expression. The regulation mechanisms for the expression of mPH in C. testosteroni R5 and P1 are therefore similar to each other but different from that in P. putida CF600 in terms of the gratuitous expression. The details of the mechanism for gratuitous expression and the reason why strain CF600 did not show such gratuitous expression are not known.

Acquiring phcS must have been crucial for C. testosteroni R5 to repress the highly unfavorable expression of phenol-metabolizing enzymes for survival in the natural environment. Strain R5 has in fact been found to be a major population in an oil refinery activated-sludge microbial community, its original habitat, where phenolic compounds were loaded intermittently as carbon and energy sources (54). Further investigation of the molecular mechanisms which regulate PhcS-mediated repression in strain R5 and control its high phenol-oxygenating activity will provide substantial information regarding bacterial survival strategies in the natural environment.