Transcriptional Regulation of the ant Operon, Encoding Two-Component Anthranilate 1,2-Dioxygenase, on the Carbazole-Degradative Plasmid pCAR1 of Pseudomonas resinovorans Strain CA10

Bacterial strains and growth conditions.The bacterial strains and plasmids used in this study are listed in Table 1. Pseudomonas strains were routinely grown on nutrient broth (Eiken Chemical Co., Ltd., Tokyo, Japan) or carbon- and nitrogen-free mineral medium (CNFMM) supplemented with carbazole or sodium anthranilate as the sole source of carbon, nitrogen, and energy at a final concentration of 1 mg/ml at 30°C as described previously (26). P. resinovorans strains CA10dm1 and CA10dm3 were obtained from successive cultures of strain CA10 on nutrient broth. Strain CA10dm1 harbors plasmid pCAR1Δ1, which lacks the 13-kb DNA region spanning ORF9 and ORF37 (between two copies of ISPre1, including only car genes), whereas strain CA10dm3 harbors plasmid pCAR1Δ3, which lacks the 120-kb DNA region spanning ORF136 and ORF50 (between two copies of ISPre3, including ant genes, car genes, and the putative AraC/XylS-type regulatory genes): please see the physical map of pCAR1 (24).

Escherichia coli strain DH5α (TOYOBO Co., Ltd., Tokyo, Japan) used for cloning procedures as a host strain was grown on 2× YT medium (31) at 37°C. Ampicillin (50 μg/ml), gentamicin (15 μg/ml), or kanamycin (50 μg/ml) was added for selective media. For plate cultures, the above media solidified with 1.6% (wt/vol) agar were used.

DNA manipulation.The plasmid was prepared from the E. coli host strain by the alkaline lysis method. DNA fragments were extracted from agarose gel with an E.Z.N.A. gel extraction kit (Omega Bio-tek, Inc., Doraville, Ga.). Other DNA manipulations were done according to standard protocols (31).

Northern hybridization analysis.According to the method described previously (26), total RNA was prepared from strain CA10 cells just after starvation on CNFMM and after growth on nutrient broth or CNFMM supplemented with carbazole or anthranilate. Probes for the antA, antB, and antC genes were prepared from a 200-bp SmaI fragment of pBCA711 (3′ portion of antA gene) (26), a 953-bp SmaI fragment of pBCA731 (3′ portion of the antB gene), and a 553-bp SmaI fragment of pBCA731 (3′ portion of the antC gene), respectively. A probe for the carAa gene was prepared from a 359-bp fragment amplified by PCR using pUCA1 as a template with the primer set CARAA-F (5′-TTATTGGCGAACATGGGGTC-3′) and CARAA-R (5′-GCCTTTCTCATCGGCGTAGA-3′). The ³²P-labeling reaction, prehybridization, hybridization, washing, and detection were performed as described previously (26).

RT-PCR analysis.As a template, 0.1 μg of total RNA of strain CA10 prepared as described above was used. The primer set used was ANTA-F (5′-TGCGCAACCTGAACATATAC-3′) and ANTC-R (5′-GGCAGCAAACGGATCAAGCC-3′) or ORF9-F (5′-AGGTCAACAGCGAAAAAGGC-3′) and CARAA-R. Reverse transcription-PCR (RT-PCR) was performed with a One Step RNA PCR kit (Takara Shuzo Co., Ltd., Kyoto, Japan). When ANTA-F and ANTC-R were used, after the RT reaction at 50°C for 30 min, PCR was performed according to the following conditions: 94°C for 10 min and 20 cycles of 94°C for 30 s, 55°C for 30 s, and 72°C for 5 min. When ORF9-F and CARAA-R were used, the RT-PCR conditions were the same as described above, except there were 25 cycles and the temperature of 72°C lasted for 10 min. For a negative control, reverse transcriptase was not mixed in the reaction mixture. The RT-PCR product was subjected to electrophoresis with 0.9% agarose gel.

Primer extension analysis.Primer extension was performed with the Li-Cor model 4200L-2 auto-DNA sequencer (Li-Cor, Inc., Lincoln, Nebr.) according to the manufacturer's instructions. Oligonucleotides ANTA-1 (5′-TCCGGCTCGGTGAATATAT-3′) and ANTA-2 (5′-GCGGGTAATAATCATCGGCT-3′) were 5′ end labeled with IRD₈₀₀ (Aloka, Ltd., Tokyo, Japan). The ANTA-1 primer was a complement only to the DNA region of antA gene, and its 5′-end thymine base corresponded to position +113 from the translation start point of antA. The 5′-end guanine base of the ANTA-2 primer corresponded to both position +243 from the antA translation start point and position +54 from the translation start point of ORF9. The primer extension reaction was done with 1.5 pmol of labeled primer and 20 μg of total RNA of strain CA10 prepared as described above. The sequence ladder for the upstream region of antA or ORF9 was obtained with pBCA731 or pBCA721 as a template, respectively.

Construction of transcriptional fusions.Transcriptional fusions for luciferase reporter analyses were prepared by amplifying appropriate DNA fragments corresponding to the upstream region of antA gene by PCR using adequately designed primer sets and pUCA741 as a template. Forward primers ANTA-F-BamHI-103 (5′-GGATCCCACCGATAAAGCGGATAGA-3′), ANTA-F-BamHI-113 (5′-GGATCCGGCCTGTGTTCACCGATAA-3′), ANTA-F-BamHI-123 (5′-GGATCCCCAATTTTATGGCCTGTGTTCA-3′), ANTA-F-BamHI-133 (5′-GGATCCCGCCGGGCGGCCAATTTTATG-3′), ANTA-F-BamHI-143 (5′-GGATCCGCCAGGGCCGCGCCGGGCGGCCAATTTT-3′), ANTA-F-BamHI-153 (5′-GGATCCATCGCCCACGGCCAGGGCCGCGCCGGGCGGCCAAT-3′), ANTA-F-BamHI-203 (5′-GGATCCGACCATCCCTAGCCTGTTA-3′), and ANTA-F-BamHI-253 (5′-GGATCCTTTTAAAATCGGCAGGGGC-3′) and reverse primer ANTA-R-HindIII (5′-AAGCTTCCTTTAATCGATACCTCG-3′) were designed to introduce artificial BamHI and HindIII restriction sites, the ribosomal binding site, and a stop codon preventing the fusion with the reporter gene (indicated by an underline, boldface type, and double underline, respectively). The amplified DNA fragments were sequenced and confirmed to be accurate. The resultant fragments were digested with BamHI and HindIII and then inserted into the broad-host-range vector pBBR1MCS-5 together with the HindIII-EcoRI fragment containing the engineered firefly luciferase gene, luc+NF, from pSP-luc+NF fusion vector (Promega, Madison, Wis.).

Luciferase reporter assay. Pseudomonas strains were grown on 100 ml of appropriate medium overnight, and the resultant cells were gathered by centrifugation (3,300 × g), washed twice with chilled sterile water, and then resuspended with 500 μl of chilled sterile water containing 10% (vol/vol) glycerol. To electrotransform 50 μl of the cells with appropriate plasmids, GENE PULSER II (Bio-Rad, Hercules, Calif.) was used under the following conditions: 200 Ω, 25 μF, and 1.8 kV for 4.7 to 5.0 ms in a 0.1-cm cuvette.

The resultant transformants were grown on nutrient broth overnight, and cells from 4 ml of the culture were washed twice with CNF buffer (26). The washed cells were suspended in 2 ml of the same buffer and then starved by incubation at 30°C with reciprocal shaking at 300 strokes/min for 3 h. Five milliliters of succinate mineral medium (SMM) containing 2.2 g of Na₂HPO₄, 0.8 g of KH₂PO₄, 3.0 g of NH₄NO₃, 2.7 g of Na₂C₄H₄O₄ · 6H₂O, 0.2 g of MgSO₄ · 6H₂O, 0.01 g of FeSO₄ · 7H₂O, 0.01 g of CaCl₂ · 2H₂O, and 0.05 g of yeast extract per liter, supplemented with sodium anthranilate or catechol (at a final concentration of 1 mg/ml), was inoculated with 500 μl of cell suspension after starvation. If necessary, isopropyl-β-d-thiogalactopyranoside (IPTG) was added to SMM at a final concentration of 1 mM. After 2 h of incubation under the above-mentioned condition, the resultant cells from 2 ml of incubation mixture were harvested by centrifugation (2,400 × g, 5 min, 4°C). The pellets obtained were resuspended in sonication buffer containing 25 mM Tris-HCl (pH 8.0), 2 mM EDTA, and 10% (vol/vol) glycerol, and then the cells were broken by ultrasonic treatment. The crude extract was collected after centrifugation (21,600 × g, 10 min, 4°C). On a 96-well microtiter plate, 10 μl of crude extract (diluted with sonication buffer to a 0.1-μg/μl protein concentration) and 10 μl of Picagene LT2.0 (Toyo Ink Co., Ltd., Tokyo, Japan) were added, the mixture was shaken for 10 s, and then luciferase activities were measured for 10 s with Centro LB960 (Berthold Technologies GmbH & Co. KG, Bad Wildbad, Germany). Each sample was assayed at least three times independently.

Construction of the expression vectors of ORF22 and ORF23.To express the putative regulatory genes from the tac promoter induced by the addition of IPTG, pMK derived from the expression vector pMMB66HE was used. ORF22 and ORF23 were separately amplified by PCR using pUCA741 as a template with primer sets ORF22-F (5′-GTCGACGAATTCAAGGAGATTGAGCCATGGCCGGGTTAGCGGGGG-3′) and ORF22-R (5′-GGATCCTTACTGGGCAATAGTCTGGTGCGGCAGCTCGCCGAAGCG-3′) and ORF23-F (5′-GTCGACGAATTCAAGGAGAGTGCCCCGTGATGAGTACAAGCC-3′) and ORF23-R (5′-GGATCCTCAAAGCGACCGGTTGCGGCGGGCTTCGTTGCGTTGC-3′), respectively. Primers were designed to introduce artificial restriction sites for SalI, EcoRI, and BamHI (indicated by underline) and ribosomal binding site (indicated by boldface type). Furthermore, a DNA fragment containing both ORF23 and ORF22 was similarly generated by PCR using ORF23F and ORF22R. These fragments were sequenced and confirmed to have precise sequences. The SalI-BamHI-digested fragment containing ORF22, ORF23, or both ORF22 and ORF23 was ligated into the corresponding site of pMK to form pMK22, pMK23, or pMK2322, respectively.

Alignment of amino acid sequences.Alignment of amino acid sequences of AraC/XylS family regulatory proteins was performed by the CLUSTAL W multiple alignment program (35).

Transcriptional analyses of the antABC gene cluster.Transcription of the antA gene is induced when P. resinovorans strain CA10 is grown on carbazole (26). Using the same probe used previously (26), we detected the specific transcript of antA from the total RNA of anthranilate-grown strain CA10 cells, as well as carbazole-grown cells (Fig. 2A, lanes 3 and 4). No hybridization was detected when total RNA extracted just after the starvation on CNFMM or after the growth on nutrient broth was used (Fig. 2A, lanes 1 and 2). Similar hybridization patterns were detected when the probes prepared from antB and antC genes were used (data not shown). These results clearly showed that the transcription of antABC genes is induced at least when strain CA10 is grown on anthranilate. Although the hybridized band was smeared, the maximum size of the transcript was estimated at about 3 kb.

• Open in new tab
• Download powerpoint

FIG. 2.

(A) Northern hybridization analysis using a 200-bp probe inside of the antA gene. Total RNA was prepared from strain CA10 cells grown on nutrient broth (lane 2), anthranilate (lane 3), and carbazole (lane 4). As a control, total RNA was similarly extracted from the cells just after starvation on CNFMM (lane 1). (B) RT-PCR amplification of the region spanning antA to antC. Total RNA extracted from the strain CA10 cells grown on anthranilate (lane 1) or carbazole (lane 2) was used as a template.

In order to confirm that the antABC genes are cotranscribed, RT-PCR analysis was performed with a primer set designed to amplify across the antABC genes. As shown in Fig. 2B, amplification of the DNA fragments with the expected size (1,506 bp) was observed, using the RNA template prepared from the cells grown on anthranilate or carbazole. No amplified bands were detected when the template prepared from the cells just after the starvation or grown on nutrient broth was used (data not shown). Along with the results of Northern hybridization analyses, we concluded that antABC genes are transcribed as a single transcriptional unit.

Determination of the transcription start point of the antA gene.We performed primer extension using 5′-end-labeled ANTA-1 primer and the same RNA samples used in Northern hybridization and RT-PCR analyses. The ANTA-1 primer anneals to a specific site inside the antA gene. A common single transcription start point was found 53 bp upstream of the antABC translation start point when anthranilate or carbazole was supplied as the growth substrate (Fig. 3, lanes 1 and 2). As shown in Fig. 3, the −35 and −10 regions corresponding to the transcription start point (TAGACC-N₁₇-TTTAAT) were highly consistent with the conserved sequence of σ⁷⁰ promoter of E. coli (TTGACA-N_16-18-TATAAT) (30): 4 out of 6 and 5 out of 6 matches, respectively. Therefore the promoter playing a central role in the inducible transcription of antABC of strain CA10 was designated P_ant.

• Open in new tab
• Download powerpoint

FIG. 3.

Determination of the transcription start point of antA by primer extension. Total RNA was isolated from strain CA10 grown on anthranilate (lane 1) and carbazole (lane 2). The other lanes correspond to a sequence ladder obtained with pBCA731 as a template, and the sequence pattern is shown on the right. The arrows indicate the primer extension product and the corresponding transcription start point in the nucleotide sequence shown below. The nucleotide sequence of the P_ant promoter is shown with the −35 and −10 boxes underlined.

The DNA region required for the inducible expression of the ant operon.Deletion analyses of the region upstream of P_ant promoter were performed with luciferase as a reporter. Strain CA10 was transformed by each pBRC plasmid harboring transcriptional fusions (Fig. 4), and the luciferase activity of the resultant transformants was measured in the presence or absence of anthranilate as described in Materials and Methods.

• Open in new tab
• Download powerpoint

FIG. 4.

Deletion analysis of the 5′ region upstream of the P_ant promoter. The DNA region fused with the luc gene is shown to the left. The numbers indicate the position of the 5′ termini of the transcriptional fusions relative to the transcription start point of antA (+1). Black boxes upstream of antA indicate −35 and −10 boxes of the P_ant promoter, respectively. Strain CA10 was transformed by the reporter plasmids of pBRC series indicated at the middle. Luciferase activity of the cells incubated on SMM with (black bars) or without (white bars) anthranilate is shown to the right. Values and error bars represent averages and standard deviations of at least three independent experiments. RLU, relative light units.

In the reaction mixtures without anthranilate, the luciferase activities of transformants were generally low (Fig. 4). On the other hand, in the presence of anthranilate, the transformants having the DNA fragments larger than (or equal to) 123 bp showed significantly higher luciferase activities than that having pBRCantA113 and pBRCantA103, although luciferase activity was somehow decreased with pBRCantA203 and pBRCantA253. This result clearly indicated that the cis-activating region necessary for the inducible expression of ant operon in strain CA10 is located within the region up to at least 70 bp from the transcription start point of antA.

Identification of the inducer of the ant operon.Using strain CA10 harboring pBRCantA253, we investigated the inducer of the ant operon. Because the P_ant promoter governs the expression of anthranilate 1,2-dioxygenase, which catalyzes the conversion of anthranilate to catechol (Fig. 1A), the effect of the presence of anthranilate (substrate) or catechol (product) on the induction of P_ant promoter was determined. The level of luciferase activity was clearly increased by the presence of anthranilate, while the activity detected in the presence of catechol was nearly negligible, as was found in the absence of either compound (Fig. 5). This result strongly indicated that anthranilate is the inducer for transcription of the ant operon in strain CA10.

• Open in new tab
• Download powerpoint

FIG. 5.

Expression profile of the P_ant promoter. Strain CA10 harboring pBRCantA253 was inoculated on SMM containing succinate plus anthranilate (•), succinate plus catechol (▪), or succinate (▴) at time zero. Luciferase activity of the cells harvested periodically was measured. Values and error bars represent averages and standard deviations of at least three independent experiments. RLU, relative light units.

Identification of the transcriptional regulator of the ant operon.The nucleotide sequence of carbazole-degradative plasmid pCAR1 containing both ant and car genes has been completely determined (24). On pCAR1, there exist only three ORFs whose deduced amino acid sequences show homologies to known transcriptional regulators. Two of them, ORF22 and ORF23, encode putative AraC/XylS family transcriptional regulators containing the conserved helix-turn-helix DNA binding motif in their C termini (18). These ORFs are tandemly located with each other in the same orientation at the region 3.2 kb upstream of antA, although ISPre1 and ORF24 are interposed (Fig. 1B). In deduced amino acid sequences, ORF23 showed 58.6% identity to PA2511 (accession no. 251201 ), a probable AraC/XylS-type regulatory gene located divergently just upstream of the antA gene of P. aeruginosa strain PAO1, but the overall lengthwise identity to other members of AraC/XylS was below 30%. On the other hand, ORF22 did not have any particularly close relatives (<39%) within the AraC/XylS family.

To determine whether the product of ORF23 or that of ORF22 takes part in the inducible expression of the ant operon, we performed reporter gene analyses using the transcriptional fusion plasmid pBRCantA253 in the first two types of Δcar mutants of strain CA10, designated P. resinovorans strains CA10dm1 and CA10dm3, as host strains. Strains CA10dm1 and CA10dm3 harbor pCAR1-derived plasmids pCAR1Δ1 and pCAR1Δ3, respectively. While both ORF22 and ORF23 were located on pCAR1Δ1, pCAR1Δ3 lost both ORFs. In the absence of anthranilate, both of these strains harboring pBRCantA253 showed significantly low activity of luciferase as well as strain CA10 (Table 2). In the presence of inducer, while strain CA10dm1(pBRCantA253) expressed almost the same level of luciferase as strain CA10, the expression level in strain CA10dm3(pBRCantA253) was decreased to about 76% of that in strain CA10 cells (Table 2).

Second, P. putida strains HS01 (Shintani et al., submitted) and DS1 (14) were used as host strains in reporter gene analyses. Strain HS01 was generated by the mating of carbazole-degrader Pseudomonas sp. strain K23 (20) with P. putida strain DS1 and harbors carbazole-degradative plasmid pCAR2, the genetic structure of which is similar to that of pCAR1 and which contains not only ant and car catabolic genes but also ORF22 and ORF23 (Shintani et al., submitted). Strain HS01 can grow on carbazole and anthranilate as the sole source of carbon, nitrogen, and energy, but strain DS1 cannot. We detected a significant induction of luciferase expression by anthranilate in strain HS01 harboring pBRCantA253, although the expression of the reporter was not induced in strain DS1 harboring pBRCantA253 (Table 2). This clearly indicated that the putative transcriptional regulator for the promoter P_ant was encoded on plasmid pCAR2 (or pCAR1) but not on the genome of strain DS1.

Then, to elucidate whether the product of ORF22, ORF23, or both can activate the expression from P_ant, we used P. putida strain DS1 transformed by a pMK-based expression vector for ORF22, ORF23, and both ORFs. Expression of these ORFs was induced by the addition of IPTG. Reporter plasmid pBRCantA253 was simultaneously introduced into strain DS1 with each pMK series, and the luciferase activities of respective transformants were determined. When both were expressed using pMK2322, the luciferase activity in response to inducer was significantly increased (Table 2). The expression of luciferase from the P_ant promoter was clearly induced by anthranilate in strain DS1 expressing only the ORF23 product, and the luciferase activity detected was as high as that of strain DS1 expressing both ORF22 and ORF23 products. On the other hand, no induction was observed in strain DS1 expressing only ORF22 (Table 2). Hence, it was concluded that the product of ORF23, designated AntR, is required for the stimulation of expression from P_ant, but that ORF22 is not involved in the regulation of anthranilate degradation.

Another copy of the P_ant promoter drives transcription of the upper car gene cluster.ORF9, which is located upstream of the upper car gene cluster (Fig. 1B), is proposed to be formed via the transposition of ISPre1 along with the 5′ portion of the antA gene (26). Because the intergenic region between ISPre1 and antA is identical in nucleotide sequence to that between ISPre1 and ORF9 (Fig. 6A), the second P_ant promoter upstream of ORF9 (about 2.1 kb upstream of the carAa gene) was assumed to be functional. Here we used the 5′-end-labeled ANTA-2 primer that is specific to both antA and ORF9 for primer extension analysis. Two primer extension products were detected with total RNA prepared from the cells grown on anthranilate or carbazole. The larger band represented the primer extension product transcribed from the antA transcription start point as determined by the sequence ladder made from pBCA731 containing ISPre1-antA intergenic region (data not shown). On the other hand, the position of the shorter band corresponded to a thymine base 79 bp upstream of the translation start point of ORF9, as determined by the sequence ladder obtained from pBCA721 containing the ISPre1-ORF9 intergenic region (Fig. 6B). This transcription start point is definitely consistent with that formerly found upstream of the antA gene in the nucleotide sequence (Fig. 3 and 6B). This result clearly indicated that the second copy of P_ant promoter upstream of ORF9 is functional.

• Open in new tab
• Download powerpoint

FIG. 6.

(A) Comparison of the downstream regions of two ISPre1s on ant and car loci. The white box in the antA gene represents the deleted region in the sequence of ORF9 as described previously (26). Nearly identical regions are shown by shading. Small arrows represent the primers used in primer extension analyses. Transcription start points of two copies of P_ant promoter are marked by arrows. (B) Transcription start point at the ORF9 locus determined by primer extension analysis. Total RNA was isolated from strain CA10 cells grown on anthranilate (lane 1) and carbazole (lane 2). The other lanes correspond to a sequence ladder obtained with pBCA721 as a template, and the sequence pattern is shown on the right.

Then, we proceeded to confirm that the upper car gene cluster is transcribed from P_ant promoter. The specific transcript of carAa from the total RNA of anthranilate-grown cells, as well as carbazole-grown cells, was detected through Northern hybridization, while a weak hybridization was detected with total RNA extracted from cells just after starvation or grown on nutrient broth (Fig. 7A). This result indicated that transcription of the carAa gene is induced when strain CA10 was grown on carbazole or anthranilate, but the lower level of constitutive expression was detected when the strain was starved or grown on nutrient medium. Next, we performed RT-PCR with primer set ORF9-F and CARAA-R. Amplification of the DNA fragments of the expected size (2,359 bp) was observed with total RNA prepared from the cells grown on anthranilate or carbazole as a template. This result revealed that carAa and ORF9 are cotranscribed. These facts indicated that the upper car gene cluster is transcribed from the second copy of the P_ant promoter and that AntR simultaneously regulates the transcription of both the ant operon and the upper car gene cluster, while it was suggested that the other constitutive promoter is also involved in the transcription of the upper car gene cluster.

• Open in new tab
• Download powerpoint

FIG. 7.

(A) Northern hybridization analysis using a 359-bp probe inside of the carAa gene. Total RNA was prepared from strain CA10 cells grown on nutrient broth (lane 2), anthranilate (lane 3), and carbazole (lane 4). As a control, total RNA was similarly extracted from the cells just after starvation on CNFMM (lane 1). (B) RT-PCR amplification of the region spanning ORF9 and carAa. Total RNA extracted from the strain CA10 cells grown on anthranilate (lane 1) or carbazole (lane 2) was used as a template.