Function and Molecular Ecology Significance of Two Catechol-Degrading Gene Clusters in Pseudomonas putida ND6

Abstract

Many bacteria metabolize aromatic compounds via catechol as a catabolic intermediate, and possess multiple genes or clusters encoding catechol-cleavage enzymes. The presence of multiple isozyme-encoding genes is a widespread phenomenon that seems to give the carrying strains a selective advantage in the natural environment over those with only a single copy. In the naphthalene-degrading strain Pseudomonas putida ND6, catechol can be converted into intermediates of the tricarboxylic acid cycle via either the ortho- or meta-cleavage pathways. In this study, we demonstrated that the catechol ortho-cleavage pathway genes (catB_IC_IA_I and catB_IIC_IIA_II) on the chromosome play an important role. The cat_I and cat_II operons are co-transcribed, whereas catA_I and catA_II are under independent transcriptional regulation. We examined the binding of regulatory proteins to promoters. In the presence of cis-cis-muconate, a well-studied inducer of the cat gene cluster, CatR_I and CatR_II occupy an additional downstream site, designated as the activation binding site. Notably, CatR_I binds to both the cat_I and cat_II promoters with high affinity, while CatR_II binds weakly. This is likely caused by a T to G mutation in the G/T-N₁₁-A motif. Specifically, we found that CatR_I and CatR_II regulate catB_IC_IA_I and catB_IIC_IIA_II in a cooperative manner, which provides new insights into naphthalene degradation.

Introduction

A number of aerobic biodegradation pathways of aromatic compounds like benzoate, aniline, phenol, pyrene and naphthalene converge at the catechol ring cleavage reaction, and eventually generate intermediates of the tricarboxylic acid cycle through two different routes [1-3]. The catechol meta-cleavage pathway is usually encoded by the nah and xyl genes on plasmids, while the catechol ortho-cleavage pathway is usually determined by the chromosomal cat and ben genes [4-9]. Ortho-cleavage is followed by further steps of the β-ketoadipate pathway which is catalyzed by the three enzymes catechol 1,2-dioxygenase (C12O, E.C. 1.13.11.1) (catA), cis,cis-muconate-lactonizing enzyme I (catB), and muconolactone isomerase (catC). The expression of this operon is regulated by CatR, and is induced by cis,cis-muconate (CCM) [9-15].

Several research groups have found that bacteria possess multiple catechol ortho-cleavage genes or clusters, which might indicate the importance of keeping the intracellular concentration of catechol and its derivatives low [6,7, 16-20]. Notably, the strains with multiple catechol dioxygenase genes can usually grow on many more aromatic compounds than strains with one dioxygenase gene [14, 21]. Different enzymatic properties and induction patterns of C12O isozymes may be responsible for the metabolism of different substrates [22]. For example, three different C12O isozymes encoded by catA, catA_I and catA_II in the naphthalene-degrading strain Pseudomonas putida ND6 were found to have different enzymatic properties [17]. Burkholderia sp. strain TH2, a 2-chlorobenzoate (2CB)-degrading bacterium, metabolizes benzoate (BA) and 2CB via two catechol ortho-cleavage pathways (cat₁ and cat₂). Interestingly, the inducer of catA₁ was found to be BA, and not 2CB. It was also found that CCM or its metabolite acts as an inducer for catA₂. These results suggest that although cat₂ genes are not indispensable for growth of TH2 on 2CB, they are advantageous [20]. Murakami et al. reported that the aniline-assimilating bacterium Frateuria sp. ANA-18 has two cat clusters, which are respectively located on the chromosome and the large plasmid [19]. Two CatR proteins that differ in their binding affinity for the catB promoters control the expression of the two cat gene clusters, which is affected by the concentration of CCM. The catB₁ promoter was shown to be active in the presence of CCM, while the catB₂ promoter was activated only at low concentrations of CCM [23]. Hence, the presence of multiple catechol ortho-cleavage genes and pathways in the same cell seems to be a widespread phenomenon that gives the host strains a selective advantage in the natural environment. However, the research on function, regulation and evolutionary significance of these genes and clusters should be further intensified.

P. putida strain ND6, which was isolated by our group from industrial wastewater in Tianjin, China, was capable of growth on naphthalene as the sole carbon and energy source, and successfully degraded 98% of 2 g/l naphthalene in mineral medium in 48 h. Similar to other described strains [24-26], ND6 possesses an integrated naphthalene-degrading pathway, and catabolizes catechol through the meta-cleavage pathway. However, a unique feature of ND6 exists in the presence of two catechol ortho-cleavage pathways, encoded by catR_IB_IC_IA_I and catR_IIB_IIC_IIA_II on the chromosome, in addition to the typical catechol meta-cleavage pathway on the plasmid. We investigated the cellular responses of P. putida ND6 grown in medium with 2 (normal concentration) and 4 g/l naphthalene (high concentration) using a quantitative proteomics-based approach. Comparative analysis of the proteomic data indicated that the expression levels of CatA_I, CatB_I, and CatB_II, which are involved in the catechol ortho-cleavage pathway, were upregulated [27]. This demonstrated that P. putida ND6 is able to survive in the presence of high naphthalene concentrations mainly because of the duplicated catechol ortho-cleavage operons. The two ortho-cleavage operons are regulated by the LysR-type regulatory proteins CatR_I and CatR_II, respectively. The distance between the two clusters was about 754 kb, and their sequence identity was 73.57%. Based on the previous work in our laboratory, we hypothesized that these two gene clusters have different functions and evolutionary origins, and might regulate each other via genetic cross-talk. In this paper, we compared the function, regulation and evolutionary significance of the two catechol ortho-cleavage gene clusters in P. putida strain ND6.

Materials and Methods

Bacterial Strains, Culture Conditions, Plasmids, Primers and DNA Manipulations

P. putida strain ND6 and mutants were cultured at 30°C and 180 rpm in Luria-Bertani (LB), or in mineral medium (MMB) with 0.2% naphthalene or 0.2% glucose as the carbon source. Escherichia coli strains were grown in LB at 37°C and 180 rpm. Where appropriate, spectinomycin (Sp; 50 mg/l), streptomycin (Sm; 50 mg/l), chloramphenicol (Cm; 25 mg/l), ampicillin (Amp; 50 mg/l), kanamycin (Kan; 50 mg/l), carbenicillin (Cb; 75 mg/l), tetracycline (Tet; 10 mg/l), or gentamicin (Gm; 10 mg/l) was added for selection.

The bacterial strains, plasmids, and primers used in the present study are listed in Tables 1 and 2. All DNA manipulations were performed according to standard procedures [31]. Restriction enzymes, DNA polymerase and T4 DNA ligase were used in accordance with the manufacturers’ specifications.

Table 1.

Strains and plasmids used in this study.

Strains/plasmids		Genotype and description	Source/reference
E. coli	JM109	cloning strain	Novagen
	S17-1	recA pro hsdR RP4-2-Tc::Mu-Km::Tn7; donor strain for conjugation;	[28]
	BL21	Expression strain	Novagen
P. putida	ND6	Cbr	[16]
	ND6-∆catRI	catRI mutant of ND6；Cbr, Kanr	This study
	ND6-∆catRII	catRII mutant of ND6；Cbr, Kanr	This study
	ND6-∆catA	catA mutant of ND6；Cbr, Kanr	This study
	ND6-∆catAnahH	catA and nahH mutant of ND6；Cbr, Kanr, Cmr	This study
	NC	ND6 containing plasmid pDN19lacΩ; Cbr, Spr, Smr	This study
	NP1	ND6 containing plasmid pDB1; Cbr, Spr, Smr	This study
	NP2	ND6 containing plasmid pDB2; Cbr, Spr, Smr	This study
	ND1P1	ND6-∆catRI containing plasmid pDB1; Cbr, Kanr, Spr, Smr	This study
	ND1P2	ND6-∆catRI containing plasmid pDB2; Cbr, Kanr, Spr, Smr	This study
	ND2P1	ND6-∆catRII containing plasmid pDB1; Cbr, Kanr, Spr, Smr	This study
	ND2P2	ND6-∆catRII containing plasmid pDB2; Cbr, Kanr, Spr, Smr	This study
Plasmid	pUC18-T simple	Cloning vector; Ampr	TransGen
	pEX18Tc	Gene replacement; Tcr, oriT+ sacB+	[29]
	pDN19lacΩ	Broad host range shuttle vector; Spr, Smr	[30]
	pEASY-Blunt	Source of kan gene; Ampr, Kanr	TransGen
	pXMJ19	Source of cat gene; Ampr, Cmr	TransGen
	pEX18Tc-RIKm	pEX18Tc with ∆catRI::Kanr; Tetr, Kanr	This study
	pEX18Tc-RIIKm	pEX18Tc with ∆catRII::Kanr; Tetr, Kanr	This study
	pEX18Tc-AKm	pEX18Tc with ∆catA::Kanr; Tetr, Kanr	This study
	pEX18Tc-HCm	pEX18Tc with ∆nahH::Cmr; Tetr, Cmr	This study
	pEX5-catRI	CatRI protein expression vector; Kanr	This study
	pEX5-catRII	CatRII protein expression vector; Kanr	This study
	pUC19c-catBI	pUC19c with catBI promoter; Apr	This study
	pUC19c-catBII	pUC19c with catBII promoter; Apr	This study
	pDBI	catI promoter inserted between the EcoRI and BamHI sites of pDN19lacΩ; Spr, Smr	This study
	pDBII	catII promoter inserted between the EcoRI and BamHI sites of pDN19lacΩ; Spr, Smr	This study

*Cb^r, carbenicillin resistant; Kan^r, kanamycin resistant; Cm^r, Chloramphenicol resistant; Sp^r, spectinomycin resistant; Sm^r, streptomycin resistant; Amp^r, ampicillin resistant; Tet^r, tetracycline resistant; Gm^r, gentamicin resistant.

Table 2.

Primers used in this study.

Primer name	Sequence (5’–3’)	Description
RIL-F	GTGATGTGGCCGAATGCCTC	Used in the creation of ND6-∆catRI mutant
RIL-R	CGGCATGAGCATTCGTGTC	Used in the creation of ND6-∆catRI mutant
RIR-F	GGCCGAAGCCAATGTCGA	Used in the creation of ND6-∆catRI mutant
RIR-R	GCTTTCGATGCCGGACTTG	Used in the creation of ND6-∆catRI mutant
RIIL-F	GACGGCCAAGTCGATGGTG	Used in the creation of ND6-∆catRII mutant
RIIL-R	TAGGCCAGCCGATATCGCT	Used in the creation of ND6-∆catRII mutant
RIIR-F	CCTGATAGATAGCCGCGTCG	Used in the creation of ND6-∆catRII mutant
RIIR-R	AGCAAGCGAAGAATGACCAAGG	Used in the creation of ND6-∆catRII mutant
AL-F	GGGCCGCTTGATCAACGTCGT	Used in the creation of ND6-∆AH mutant
AL-R	ATCTCAGGCAGGTTGGAAATAG	Used in the creation of ND6-∆AH mutant
AR-F	CAGTCCGAATACAACCTGCGC	Used in the creation of ND6-∆AH mutant
AR-R	CCGAAGGCGACAAAGCTGGAG	Used in the creation of ND6-∆AH mutant
HL-F	GCGCCGCTGGACGTGACAATGC	Used in the creation of ND6-∆AH mutant
HL-R	TCCAGTACACGCAGTTGCACG	Used in the creation of ND6-∆AH mutant
HR-F	TTATTTCTTCGACCCGTCCGG	Used in the creation of ND6-∆AH mutant
HR-R	GGCCGGTCCGACTTTCCAGGT	Used in the creation of ND6-∆AH mutant
kan-F	GGGCGGTTTTATGGACAGC	Used in the creation of ND6-∆catRI, ND6-∆catRII, ND6-∆AH mutant
kan-R	CGGTGCTCAACGGGAATC	Used in the creation of ND6-∆catRI, ND6-∆catRII, ND6-∆AH mutant
cat-F	CTTAAAAAAATTACGCCCCGC	Used in the creation of ND6-∆AH mutant
cat-R	TGATCGGCACGTAAGAGGTTC	Used in the creation of ND6-∆AH mutant
catBIP-F	AGCACGGTGCAGGTCTGTTC	Used for construction catBI-lacZ fusions
catBIP-R	GCCCGAGTGATGCGTTTAC	Used for construction catBI-lacZ fusions
catBIIP-F	CAAGCTGTTTGAGCAAGGTGTC	Used for construction catBII-lacZ fusions
catBIIP-R	TAAGGCATCAATACCCATCG	Used for construction catBII-lacZ fusions
CatRI-F	ACTTGATTGTTGAAGGATT	Used for CatRI protein expression
CatRI-R	TCAGAGTGCCTGTTGCTC	Used for CatRI protein expression
CatRII-F	GTGGAGCTCAGGCAT	Used for CatRII protein expression
CatRII-R	TTATCGATTGATGGTCG	Used for CatRII protein expression
q16S-F	CGGATCGCAGTCTGCAACTC	16s gene for RT-qPCR
q16S-R	ACACCGTGGTAACCGTCCTCC	16s gene for RT-qPCR
qcatAI-F	AGGAATGCCTGGACCTGCTCG	catAI gene amplification for RT-qPCR
qcatAI-R	CGAATGACAACTCGGCAAAGCG	catAI gene amplificationfor RT-qPCR
qcatBI-F	CTGGCATTGCCTTGTACGG	catBI gene amplification for RT-qPCR
qcatBI-R	AGGCGCTGTTCGTCCAATG	catBI gene amplification for RT-qPCR
qcatCI-F	ACATGGACCCGGCCAAGG	catCI gene amplification for RT-qPCR
qcatCI-R	TCAGCGATCGTCGCTGTG	catCI gene amplification for RT-qPCR
qcatAII-F	AAGGGCTAATCGGCGAAGTG	catAII gene amplification for RT-qPCR
qcatAII-R	GCTGTTCGGCAGTCTCAACC	catAII gene amplification for RT-qPCR
qcatBII-F	GGTGCGGCTCAATCAACG	catBII gene amplification for RT-qPCR
qcatBII-R	GCCTGGGCTATCTGGGCAG	catBII gene amplification for RT-qPCR
qcatCII-F	CAAGTGGCTGCCCGCTTG	catCII gene amplification for RT-qPCR
qcatCII-R	AGCGCTTCGACACTGTCCACA	catCII gene amplification for RT-qPCR
catBICI-F	GGGCCTGACATTGGACGAAC	The region of catBI gene and catCI gene; RT-PCR
catBICI-R	ACACAGGCCGTCGACCTCA	The region of catBI gene and catCI gene; RT-PCR
catCIAI-F	CGTACATGGACATTGAGGTCGA	The region of catCI gene and catAI gene; RT-PCR
catCIAI-R	AGGTCGAGGAAGTGCTCGATG	The region of catCI gene and catAI gene; RT-PCR
catBIICII-F	GATCAGCTGCAATGGCACAC	The region of catBII gene and catC2 gene; RT-PCR
catBIICII-R	AGCGCTTCGACACTGTCC	The region of catBII gene and catCII gene; RT-PCR
catCIIAII-F	CAAGTGGCTGCCCGCTT	The region of catCII gene and catAII gene; RT-PCR
catCIIAII-R	CGATAAACAACGTGGTGGCAAC	The region of catCII gene and catAII gene; RT-PCR
M13F-11	CGCCAGGGTTTTCCCAGTCACGAC	TaqMan primer used for catBI or catBII EMSA and DNase I footprinting template
M13R-12	AGCGGATAACAATTTCACACAGGA	TaqMan primer used for catBI or catBII EMSA and DNase I footprinting template
catBI-pe	TGTGTTCAATCAGCGCGCTTGTC	CatBI gene-specific primer, Primer extension
catBII-pe	GCTGCTGTCATTTCAGGTTCCATC	CatBII gene-specific primer, Primer extension

Construction of P. putida ND6 Mutants

The suicide vector pEX18Tc was used for gene replacement in Pseudomonas. The catR_I replacement vector pEX18Tc-R₁Km was constructed by inserting the Kan^r cassette into the catR_I gene in the pEX18Tc vector to generate a partial deletion from 226 to 440 bp. Similarly, the catR_II replacement vector pEX18Tc-R_IIKm was constructed by replacing 335 to 407 bp, and the catA replacement vector pEX18Tc-AKm by replacing between 10 to 514 bp. The knockout plasmids pEX18Tc-RIKm, pEX18Tc-R_IIKm, and pEX18Tc-AKm were transferred from E. coli S17-1 into P. putida ND6 by intergeneric conjugation , followed by selection on LB + Cb + Kan medium with 20% sucrose to generate the catR_I, catR_II and catA mutants. The resulting mutant strains were named P. putida ND6-ΔcatR_I, ND6-ΔcatR_II, and ND6-ΔcatA, respectively.

The nahH replacement vector pEX18Tc-HCm was constructed by inserting the Cm^r cassette into the internally deleted nahH gene in the pEX18Tc vector, replacing the sequence between 204 to 742 bp. Finally, the P. putida strain ND6-ΔcatAnahH was obtained.

Time-Courses of Growth, Naphthalene Degradation, and Catechol Dioxygenase Activity

The wild-type ND6, as well as the mutants ND6-ΔcatR_I, ND6-ΔcatR_II and ND6-ΔcatAnahH, were individually grown in MMB medium with 0.2% naphthalene. Bacterial growth was monitored by measuring the optical density at 600 nm. Naphthalene in the culture broth was extracted twice with an equal volume of dichloromethane, and its concentration was measured according to a published method [32]. The activity of catechol dioxygenases (C12O and C23O) was assayed as described previously [17].

Quantitative Real-Time PCR (RT-qPCR)

The wild-type ND6, as well as the ND6-ΔcatR_I and ND6-ΔcatR_II mutants were cultured at 30°C in MMB with glucose or naphthalene. Cells were harvested at 25 h (end of the exponential growth phase) and washed with TE buffer. RNA was extracted using the RNAprep Pure Bacteria Kit (Tiangen Biotech, China) and reverse-transcribed into cDNA using the FastKing RT Kit (with gDNase) (Tiangen Biotech). The RT-qPCR was performed following the instructions of the Green Premix Ex Taq II Kit. All RT-qPCR reactions were conducted in three biological and three technical replicates.

Plasmid standards were used for absolute quantification of the corresponding gene fragments. The plasmid copy number was determined according to the molar mass derived from the plasmid and amplicon sequences [33]. For each standard sample, the RT-qPCR system was used to measure the cycle threshold values, which were used to draw a standard curve for each isoform by plotting the cycle threshold values versus the log value of the transcript copy number. Regression equations generated by the system software were used to calculate the transcript copy number for each isoform in each test sample, which was normalized to the value of the 16S rRNA and expressed as the absolute copy number per 1000 copies of 16S transcript. The 95% confidence interval was used to assess the significance of copy number differences, and analysis of variance (ANOVA) was used to assess the significance of the threshold cycle values.

Expression and Purification of CatR_I and CatR_II

E. coli BL21 carrying the CatR_I and CatR_II expression plasmids pEX5-CatR_I and pEX5-CatR_II were grown overnight in LB medium with kanamycin. The expression was induced overnight at 16°C by adding a final concentration of 0.5 mM isopropyl-β-D-thiogalactopyranoside (IPTG). To purify CatR_I and CatR_II, cell pellets obtained from 1 liter cultures were collected at 13,000 ×g for 3 min at 4°C and resuspended in 40 ml of binding buffer (50 mM sodium phosphate, 0.3 M NaCl, pH 7.4), and sonicated. The clarified lysate was applied to a nickel-nitrilotriacetic acid (NTA) affinity column and allowed to bind for 1 h, followed by washing with binding buffer containing 10 mM, 30 mM and 60 mM imidazole, and eluting with binding buffer containing 500 mM imidazole. The collected fractions were determined by sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE). The desalination column Sephadex G-25 (GE, USA) was used to remove the imidazole, and yielded the purified protein. The protein was stored in glycerol at -20°C.

Primer-Extension Assay

The primer extension assay was carried out according to the protocol of Fekete et al., as published previously [34].

Electrophoretic Mobility Shift Assay (EMSA)

The fluorescent 6-fluorescein amidite (FAM)-labeled probes were prepared by amplifying the promoter regions of pUC19c-catB_I and pUC19c-catB_II by PCR using the primers M13F-11 and M13R-12 (Table 2). The Wizard SV Gel and PCR Clean-Up System (Promega) was used to purify the resulting FAM-labeled probes, which were then quantified using a NanoDrop 2000C instrument (Thermo Fisher Scientifc, USA). The EMSA samples comprised 20 μl of 50 mM Tris-HCl [pH 8.0], 100 mM KCl, 2.5 mM MgCl2, 0.2 mM dithiothreitol (DTT), with 2 μg salmon sperm DNA and 10% glycerol, as well as 40 ng of the probe and the indicated DNA-binding proteins. After incubation for 30 min at 30°C, the mixture was loaded onto a 10% PAGE gel buffered with 0.5×Tris-borate-EDTA (TBE).

DNase I Footprinting

The fluorescent FAM-labeled probes were prepared as described above. The DNase I footprinting assays were conducted as published by Wang et al. [35]. For each assay, 400 ng of the probe in a total volume of 40 μl was mixed with different amounts of the indicated DNA-binding protein. The preparation of the DNA ladder, electrophoresis and data analysis were the same as described before [35], except that the GeneScan-LIZ500 size standard (Applied Biosystems) was used.

Construction of catB_I-lacZ and catB_II-lacZ Fusions and β-Galactosidase Activity Assays

A 309 bp DNA fragment corresponding to the catR_I-catB_I intergenic region (from -347 to -39 relative to the catB_I initiation codon) and a 330bp DNA fragment corresponding to the catR_II-catB_II intergenic region (from -358 to -29 relative to the catB_II initiation codon) were amplified by PCR to construct cat_I promoter-lacZ fusion and cat_II promoter-lacZ fusion, respectively. Two amplified fragments of cat_I and cat_II promoter were digested with EcoRI and BamHI, respectively, and inserted into pDN19lacΩ that had been previously cut with the same enzymes. The resulting plasmids, pDB_I and pDB_II, were transformed into E. coli S17-1 by chemical transformation. The complementation was performed through the conjugation between E. coli S17-1 with plasmid pDB_I and pDB_II and P. putida ND6, P. putida ND6-ΔcatR_I, and ND6-ΔcatR_II. Recombinant P. putida was selected on MMB plate containing 0.2% salicylate (w/v %) with Cb, Sp and Sm, and subcultured several times to ensure plasmid stabilization. The β-galactosidase activity was measured according to the method of Green [31].

Phylogenetic Tree and Analysis of Conserved Sequences

Database searches were performed using BLAST at the National Center for Biotechnology Information (NCBI) website. Multiple sequence alignments were performed using Clustal W. Then, a neighbor-joining (NJ) phylogenetic tree was constructed using MEGA software version 5.0, and the branching reliability was tested using bootstrap re-sampling (1,000 pseudo-replicates). Analysis of conserved motifs based on known sequences was conducted using WebLogo to generate LOGO diagrams.

Results

Functional Analysis of the Two Catechol-Degrading Gene Clusters

As mentioned above, the genome of P. putida ND6 encodes two catechol-degrading gene clusters (catR_IB_IC_IA_I and catR_IIB_IIC_IIA_II) (Fig. 1). To investigate the possible physiological role of these two orthologous clusters, catA (encoding C12O) and nahH (encoding C23O) from the large plasmid pND6-1, which are responsible for catechol ring cleavage [16], were knocked out. The obtained double-mutant strain ND6-ΔcatAnahH was still able to grow on naphthalene by employing the catechol-degrading gene clusters in the genome as a backup, although the C12O activity dropped from 74.60 to 43.02 U/mg (Fig. 2). Furthermore, to examine the relative roles of the cat_I and cat_II clusters in catechol metabolism, we disrupted the catR_I and catR_II regulatory genes and examined the properties of the resulting strains. The ability of these strains to use naphthalene as the sole carbon source was tested in growth and degradation experiments. Interestingly, inactivation of catR_I or catR_II had no obvious effect on the strains’ ability to utilize naphthalene (Fig. 3).

Fig. 1. Genetic, regulatory and biochemical connections of naphthalene catabolism in P. putida ND6.

The figure sketches the pathways involved in the naphthalene catabolism of P. putida ND6: the upstream operon (nahABCDEF) for the conversion of naphthalene into salicylate; the downstream operon for the conversion of salicylate into catechol and eventually into intermediates of the central carbon metabolism. Catechol metabolism can proceed via either the ortho- or the meta-cleavage pathway. Both pathways converge towards catechol and diverge at that point, as this compound can be cleaved between positions 1 and 2 by catA, whereas nahH cleaves between positions 2 and 3. The three catA genes are located on the chromosome and large plasmid, respectively.

Fig. 2. Naphthalene-degradation phenotype and catechol dioxygenase enzyme activities of P. putida ND6 and the ND6-ΔcatAnahH mutant.

Naphthalene (0.2%, w/v %) was provided as the sole carbon source. The small figure shows the C12O and C23O enzyme activity of P. putida ND6 and ND6-ΔcatAnahH.

Fig. 3. Growth and naphthalene-degradation curves of P. putida ND6 wild type, as well as the mutants ND6-ΔcatR_I and ND6-ΔcatR_II.

Naphthalene (0.2%, w/v) was provided as the sole carbon source.

Transcriptional Analysis of the Two Catechol-Degrading Gene Clusters

The cDNA of P. putida ND6 was used as template to amplify the intergenic regions of the catB_IC_IA_I and catB_IIC_IIA_II operons. The results indicated that either the catB_IC_IA_I or catB_IIC_IIA_II cluster can be co-transcribed (Fig. 4). However, the results of RT-qPCR revealed that the transcription of the downstream genes catC_I and catA_I was higher than that of catB_I (Fig. 5A), suggesting that catC_I and catA_I were independently transcribed, while the catB_IC_IA_I cluster was co-transcribed. A similar phenomenon was also observed in the cluster catB_IIC_IIA_II (Fig. 5A).

Fig. 4. Agarose gel plots were used to determine the transcription patterns of the two cat gene clusters.

M1: Tiangen DNA Marker I; M2: Tiangen DNA Marker II; Lane1: Band of the catB_I-catC_I intergenic region DNA fragment; Lane 2: Band of the catC_I-catA_I intergenic region DNA fragment; Lane 3: Band of the catB_I-catA_I intergenic region DNA fragment; Lane 4: Band of the catB_II-catC_II intergenic region DNA fragment; Lane 5: Band of the catC_II-catA_II intergenic region DNA fragment; Lane 6: Band of the catB_II-catA_II intergenic region DNA fragment; Lane 7: negative control.

Fig. 5. Copy number determination by absolute RT-qPCT using plasmid standards.

Transcript copy number of genes per 1000 mRNA transcript copies of the 16S rRNA in ND6 (A), ND6-ΔcatR_I (B), and ND6-ΔcatR_II (C). *p < 0.05, **p < 0.01, ***p < 0.001 between cells grown on naphthalene and on glucose. (D) Transcript copy number of two cat gene clusters per 1000 mRNA transcript copies of the 16S rRNA in ND6, ND6-ΔcatR_I and ND6-ΔcatR_II. Different superscript letters indicate statistically significant differences among experimental groups (p < 0.05; Duncan’s multiple range test). Either naphthalene (N-) or glucose (G-) was provided as the sole carbon source. Bars show the standard errors of the means.

In the naphthalene medium, the activity of transcription (Fig. 5A) and expression levels (Fig. 2, small figure) of three catA genes and C12O were much higher than nahH gene and C23O. Therefore, the naphthalene metabolism of P. putida ND6 proceeds via catechol cleavage in both the ortho and meta positions, whereby the ortho-cleavage pathway is the main cleavage pathway. The transcriptional levels of the ortho-cleavage pathways (catA_I, catB_I catC_I, catA_II, catB_II, catC_II, catA) were all increased in naphthalene medium, while that of meta-cleavage pathways (nahH) was not (Fig. 5A). It indicates that the ortho-cleavage pathways can be induced by naphthalene or its metabolites but the meta-cleavage pathway apparently cannot. Furthermore, we measured the transcriptional levels of cat_I and cat_II in the mutant strains ND6-ΔcatR_I and ND6-ΔcatR_II, following cultivation in the presence of glucose or naphthalene. For the ND6-ΔcatR_I strain (only catR_II is functional), the transcriptional levels of catB_I were significantly decreased, while the catB_II gene cluster could still be induced (Fig. 5B). For the ND6-ΔcatR_II strain (only catR_I is functional), the transcriptional level of catB_II was also decreased while the catB_I gene cluster could still be induced (Fig. 5C). This result demonstrated that CatR_I and CatR_II are the native regulatory proteins of the cat_I cluster and the cat_II cluster, respectively.

To investigate their possible regulatory roles, the impacts of catR_I and catR_II were further analyzed by determining the transcriptional levels of all six catechol-degradation genes in wild-type ND6 and the mutants when grown with naphthalene as the sole carbon source. When either catR_I or catR_II was knocked out, the transcription of the cat_I or cat_II cluster could still be detected (Fig. 5D). It indicates that CatR_I and CatR_II might cross-regulate the transcription of each other's target genes, and explains that the naphthalene degradation curves of ND6-ΔcatR_I and ND6-ΔcatR_II did not exhibit significant changes (Fig. 3).

Promoter Structure of the Two Catechol-Degrading Gene Clusters

The primer-extension assay indicated that the transcription start site (TSS) of cat_I is located at a G nucleotide (Fig. 6). Surprisingly, the primer extension assay of cat_II failed three times based on reliable methods, probably because the transcript abundance of the cat_II cluster is particularly low.

Fig. 6. Results of the primer-extension assay.

Characterization of the cat₁ transcription start site (TTS) using the primer extension assay. The primer extension experiments used the primer catB_I-pe (Table 2), which was designed to bind downstream of the catB_I initiation codon. Comparison of the sequencing results of the cDNA (above) and the genomic DNA (below) to determine the TSS (marked with a red asterisk).

The CatR_I and CatR_II binding regions were also studied using DNase I footprinting (Fig. 7). Purified CatR_I was incubated with a 408 bp fragment labeled with FAM, with a sequence corresponding to the promoter region of cat_I, and the resulting complexes were treated with DNase I, followed by DNA fragment analysis by capillary electrophoresis. The binding sequence of the CatR_II-catB_II complex was assessed using the same method, and a 403 bp FAM-labeled DNA fragment was incubated with purified CatR_II. In the absence of CCM, CatR_I protected a continuous 26 bp region that had also been determined to be located from -137 to -112 relative to the catB_I codon (Fig. 7A). CatR_II protected a continuous 25 bp region that had also been determined to be located from -137 to -113 relative to the catB_I initiation codon (Fig. 7C). In the presence of CCM, CatR_I protected a continuous 48 bp region that was determined to be located from -137 to -90 relative to the catB_I initiation codon (Fig. 7B). CatR_II protected a continuous 41 bp region that was determined to be located from -137 to -97 relative to the catB_I initiation codon (Fig. 7D). These results demonstrated that CCM has a significant effect on the DNA binding mode of CatR. In the absence of CCM, CatR bound the repression binding site (RBS), which might be associated with the negative regulation of itself. In the presence of CCM, CatR occupied an adjacent downstream site (Fig. 8), designated as the activation binding site (ABS) [36, 37]. Like most LysR proteins, the DNA-binding sites of CatR_I and CatR_II invariably contain incomplete inverted repeats (G-N₁₁-A) or inverted repeats (T-N₁₁-A), respectively [38]. The results also indicated that the -35 and -10 regions of the cat promoter were important for promoter activity but not for CatR binding.

Fig. 7. DNase I footprinting analysis of CatR_I (A, B) binding to the catB_I promoter and CatR_II (C, D) binding to the catB_II promoter with or without CCM.

Based on these results and earlier reports on bacterial promoter prediction [39], we identified the promoter structure of the two catechol-degrading gene clusters (Fig. 8). The putative TSS of cat_I was consistent with the primer-extension experiment, and the putative TSS of cat_II was located 67 bp upstream of the catB_II initiation codon. Other motifs of the cat_I and cat_II promoters were identified by comparing to the reported conserved sequences recognized by RNA polymerase [11, 13, 19, 23, 40, 41], including the –35 element (positions –35 to –30, if the transcriptional start site is denoted as +1), the –10 element (positions –12 to –7) and the discriminator element (Dis; –6 to –4) (Fig. 8).

Fig. 8. Structure of the catB_I and catB_II promoter regions.

(A) The promoter region of the cat_I gene cluster. (B) The promoter region of the cat_II gene cluster. The transcription start site is shown as +1. The predicted discriminator element (Dis), initiation codon (arrows, the direction represents the direction of transcription), and -10 and -35 regions of the promoter were also designated above the sequence. The promoter regions of cat_I and cat_II protected by CatR_I or CatR_II from DNase I digestion without CCM, called the repression binding site (RBS), are marked with black solid lines; the extended protected region in the presence of CCM, called the activation binding site (ABS), is marked with the black dotted line.

Transcriptional Cross-Regulation between the Two Catechol-Degrading Gene Clusters

The regulatory proteins CatR_I and CatR_II were cloned into a vector to introduce six histidine residues at the C-terminus of the protein, and were expressed and purified from E. coli as described in the Materials and Methods section. The binding abilities of CatR_I and CatR_II to the upstream regulatory regions of cat_I and cat_II were studied by EMSA using a concentration gradient of CCM (Fig. 9). As expected, the addition of 60 ng CatR_I to 40 ng of the labeled probe caused a strong shift in the mobility of the cat_I promoter fragment (Fig. 9A). Notably, CatR_I could also strongly bind to the cat_II promoter region (Fig. 9C). By contrast, CatR_II was able to specifically bind the cat_II and cat_I promoter regions, but the binding stability of the CatR_II-cat_II and CatR_II-cat_I complexes was obviously reduced (Figs. 9B and 9D). These results support the hypothesis that CatR_I and CatR_I might cross-regulate the transcription of each other's target genes. In addition, CCM did not affect the binding of CatR_I to the corresponding target regions, which was in agreement with previous reports [10, 13]. However, the binding regions in the presence and absence of CCM might be different. As can be seen in Figs. 9B anddding CCM to the reaction mixture decreased the intensity of the DNA band, indicating that CatR_II recognizes and binds to the cat_II and cat_I promoters when CCM is at a lower concentration.

Fig. 9. EMSA of CatR_I and CatR_II binding to the cat_I and cat_II promoters, respectively.

(A) EMSA of CatR_I binding to the cat_I promoter. (B) EMSA of CatR_II binding to the cat_II promoter. (C) EMSA of CatR_I binding to the cat_II promoter. (D) EMSA of CatR_II binding to the cat_I promoter. The first lane contains the free fragment, lanes 2-7 contain labeled probe and CatR_I or CatR_II protein, with 0, 0.1, 1.0, 10, 100, and 500 μM muconic acid, respectively. The concentration of the labeled probe was fixed at 40 ng.

Promoter Activity Analysis of the Two Catechol-Degrading Gene Clusters

The primer extension assay indicated that there might be significant differences in the abundance of transcripts between the cat_I and cat_II clusters. To test this idea, the promoter activities of cat_I and cat_II (transcriptionally fused to the lacZ gene in pDN19lacΩ) were examined using the classical reporter β-galactosidase (Table 3).

Table 3.

Determination of relative promoter activities via β-galactosidase expression (measured in Miller units).

Strains	NC	NP1	NP2	ND1P1	ND1P2	ND2P1	ND2P2
MMB + glucose	8.37	151.30	15.89	8.09	9.45	196.37	8.19
MMB + naphthalene	8.30	199.31	11.52	11.87	9.94	297.38	11.54

*The β-galactosidase activity was determined in the following strains: NC (ND6 containing pDN19lacΩ); NP1 (ND6 containing pDN19lacΩ + cat_I promoter); NP2 (ND6 containing pDN19lacΩ + cat_II promoter); ND1P1 (ND6-ΔcatR_I containing pDN19lacΩ + cat_I promoter); ND1P2 (ND6-ΔcatR_I containing pDN19lacΩ + cat_II promoter); ND2P1 (ND6-ΔcatR_II containing pDN19lacΩ + cat_I promoter); ND2P2 (ND6-ΔcatR_II containing pDN19lacΩ + cat_II promoter).

As expected, in strain NP2 (ND6 wild type containing the cat_II promoter-lacZ fusion), the β-galactosidase activity was low in both glucose and naphthalene media. By contrast, the β-galactosidase activity of strain NP1 (ND6 wild type containing the cat_I promoter-lacZ fusion) was very high, indicating that the cat_I promoter is much stronger than the cat_II promoter. The activity of the cat_II promoter in cells grown on glucose or naphthalene showed little variation, probably because the transcript abundance was too low. The β-galactosidase activity of NP1 cells grown on glucose was approximately 75.9% of the activity of cells grown on naphthalene, indicating that naphthalene or its metabolites can slightly induce the cat_I promoter but are not necessary for its activity.

To confirm the function of CatR_I, we introduced the plasmid pDB_I (pDN19lacΩ + cat_I promoter) and plasmid pDB_II (pDN19lacΩ+cat_II promoter) into ND6-ΔcatR_I and ND6-ΔcatR_II, respectively, resulting in the strains ND1P1 and ND1P2. A drastic reduction of β-galactosidase activity was observed in ND1P1, whereby the promoter activity of cat_I was almost equal to that of the control strain NC (ND6 containing pDN19lacΩ) in glucose-grown cells, but the cat_I promoter could be still induced slightly in naphthalene-grown cells. This result suggested CatR_II might function as a backup regulator. Interestingly, the promoter activity of cat_II was also decreased in the CatR_I mutant (ND1P2) compared with NP2, suggesting that CatR_I might competitively bind the promoter region of cat_II and initiate transcription more effectively in the wild-type ND6. On the other hand, we found that the β-galactosidase activity of ND2P1 (ND6-ΔcatR_II containing pDN19lacΩ + cat_I promoter) was obviously higher than that of NP1, both on glucose and on naphthalene. This result suggested that CatR_I could bind the cat_II promoter and initiate transcription more efficiently with or without the inducer when CatR_II is unavailable. All these results were in agreement with the EMSA results. Through in vitro and in vivo experiments, we have confirmed that two regulatory proteins CatR_I and CatR_II have interactive regulation on the cat_I and cat_II gene clusters, and CatR_I has a stronger regulatory and activation effect on the gene clusters than CatR_II.

Evolutionary Analysis of the Two Catechol-Degrading Gene Clusters and Their Promoter Regions

To analyze the diversity of the two catechol-degrading gene clusters in the ND6 genome, 24 sequences of different cat gene clusters were downloaded from NCBI to construct a phylogenetic tree (Fig. 10). Overall, the phylogenetic analysis showed that the cat gene clusters have significant phylogenetic diversity, and could be divided into two main groups depending on the genus. Most Pseudomonas spp. have two cat clusters, which constituted one group, including cat_I and cat_II of the ND6 strain. In the Pseudomonas spp. group, most second copies of the cat gene clusters, including catR_IIB_IIC_IIA_II of P. putida ND6, formed a monophyletic clade. By contrast, most cat₁ gene clusters, including catR_IB_IC_IA_I of P. putida ND6, formed a separate clade, indicating that cat_I and cat_II of ND6 have different evolutionary origins. Interestingly, in three Burkholderia sp. strains (GenBank: CP001052.1, GenBank: AB035483.1 and GenBank: CP026112.1), catA was found to be transcribed individually while catR-catB-catC was transcribed in the opposite direction, which is different from the common order catB-catC/catA-catA/catC. Therefore, the expression of C12O, encoded by catA, must be very adaptable to respond to the environment. However, gene clusters with different transcription sequences have a far-reaching relationship. Consistent with a previous study[38], analysis of an alignment of nine intergenic regions of catB downloaded from NCBI confirmed the presence of two types of binding motifs, with conserved sequences T-N₁₁-A and mutant sequences G-N₁₁-A (Fig. 11). In agreement with a report by Parsek et al. [37], the mutation of a T to a G resulted in an increase in the binding of CatR_I to both the catB_I and catB_II promoters (Fig. 9).

Fig. 10. Phylogenetic analysis of cat gene clusters.

The phylogenetic tree was generated using MEGA 5.0 with maximum parsimony and 1000 bootstrap replicates. Reference DNA sequences were selected by BLAST searches in the NCBI database.

Fig. 11. LOGO diagrams of the binding sites occupied by CatR.

The sequence was derived from the catR-B intergenic region of the reported catRBCA gene clusters, including P. putida PRS2000 [13], P. putida RB1 [11], Pseudomonas resinovorans strain CA10 [40], Acinetobacter lwoffii K24 [41], Frateuria sp. ANA-18 [19, 23], and P. putida ND6. The LOGO diagrams were built using all the predicted binding sites for the respective CatR in all studied genomes in all catR-B intergenic regions. The sequence between the two dashed lines is an inverted repeat (T-N₁₁-A ) or an incomplete inverted repeat (G-N₁₁-A ). The T/G and A residues of the T/G-N₁₁-A motif were marked in red and green, respectively.

Discussion

A number of aerobic biodegradation pathways of aromatic compounds like phenol, benzoate, or naphthalene converge into catechol ring cleavage [42]. Our previous study showed that each C12O isozyme in P. putida ND6, encoded by the separate catA genes on the chromosome and the large pND6-1 plasmid, belonged to independent branches of the phylogenetic tree and have different enzymatic properties [17]. The two catA genes located on the chromosome (Fig. 1) were found to significantly contribute to the fitness of the host strain that is adapted to high concentrations of naphthalene [27]. The phenotypic determination showed that the growth curve of P. putida ND6-ΔcatAnahH decreased slightly compared to the wild type, indicating that the two catechol ortho-cleavage clusters in the genome play an important role but are not the only genes with this function. In nature, many bacteria grow and develop by catechol ortho- and meta-cleavage pathways to adapt to the presence of aromatics in the environment [3, 43]. The results of RT-qPCR and the catechol dioxygenase enzyme activity assay supported this idea. In addition to P. putida ND6, many other strains, especially of Pseudomonas and Burkholderia spp., possess multiple cat gene clusters. In Burkholderia sp. strain TH2, although catA₂ is not indispensable for the growth on 2CB, the presence of the cat₂ gene cluster is beneficial [44]. Similar observations were made in Acinetobacter lwoffii K24, and Frateuria species ANA-18 [19]. Thus, the presence of multiple cat genes or clusters in the same cell is a widespread phenomenon[14] that may help the host adapt to changing environments by either reducing the intracellular concentration of catechol, which is a toxic intermediate of the degradation of various aromatic compounds, or by substituting for each other to circumvent harmful mutations [3].

It is interesting to note the different arrangement of cat genes in Fig. 10. Most of them are arranged in the order catB-catC/catA-catA/catC, while the catR gene is divergently transcribed from the operon. Furthermore, the catA gene is transcribed separately from catR-catB-catC as arranged in Paraburkholderia terrae strain DSM 17804, Burkholderia sp. strain TH2 and Paraburkholderia phytofirmans strain PsJN. At the same time, we found that catA is under independent transcriptional regulation in the ND6 strain, while the catBCA operon is co-transcribed. Consequently, we assumed that individual transcription of the catA gene might be important and not an accidental evolutionary event. Notably, catA encodes the rate-limited enzyme C12O in the catechol ortho-cleavage pathway, so the abundance of C12O should be regulated sensitively and quickly to adapt to the concentration of the substrate. Therefore, the independent transcription of catA might represent a positively selected genetic adaptation [3].

The intricate regulatory interplay of multiple cat gene clusters suggests an intimate mutual adaptation [45]. For instance, CatR is able to activate the clcABD promoter but ClcR cannot activate the catBCA promoter in P. putida PRS2000 [46]. As a result, transcription from the clc promoter is repressed by the TCA-cycle intermediates succinate, citrate and fumarate, while the presence of these organic acids does not affect the transcription from the cat promoter. This difference provides some flexibility to respond to different environmental signals in addition to the presence of the inducer [47]. Similarly, the question of how the two catBCA gene clusters are regulated in P. putida ND6 was investigated. We found that the core transcriptional activation mechanisms of the two ortho-cleavage operons are conserved. First, both can be induced by CCM. Second, CatR_I and CatR_II are the native regulatory proteins of the cat_I cluster and cat_II cluster, respectively. Third, in the absence of CCM, CatR_I and CatR_II bind to the RBS, which contains a T/G-N₁₁-A motif, presumably allowing CatR to negatively regulate its own expression, while in the presence of CCM, CatR_I and CatR_II occupy an adjacent downstream site, designated as the ABS. Based on these conserved characters, CatR_I and CatR_II share functional similarities which allow them to complement each other’s mutants at the transcriptional level. Different from CatR and ClcR of P. putida, CatR_I and CatR_II can activate the promoters of the other, which provides another level of flexibility to respond to harmful mutations of each regulator. Additionally, EMSA demonstrated that CatR_I binds to both the cat_I and cat_II promoters with high affinity, while CatR_II binds weakly. These observations were confirmed by lacZ transcriptional-fusion expression experiments, which indicated that CatR_I might competitively bind the promoter region of cat_II and initiate transcription more effectively (Table 3). A mutation in the binding motif from T-N₁₁-A (cat_II) to G-N₁₁-A (cat_I) may explain the difference of binding characteristics. In conclusion, the two cat gene clusters in P. putida ND6 can cross-regulate each other as a result of similar evolutionary origins, but they diverged due to the accumulation of mutations that appear to be evolutionarily advantageous.

In this study, we found that the catC_I/catC_II and catA_I/catA_II genes are transcribed independently, in addition to the co-transcription of the catBCA operon. We are conducting further experiments to demonstrate why and how the regulation of cat genes in P. putida ND6 is so complex.