Taxonomic relationship.

Phylogenetic relationship of CSV86 was established using 16S rRNA gene sequences of 38 completely sequenced Pseudomonas species from NCBI database. The alignment was carried out using ClustalW and the phylogenetic tree was constructed using the maximum likelihood algorithm (Hasegawa-Kishino-Yano model) with MEGA 5.2 [17]. In addition, MEGA 5.2 was also used for alignment and constructing phylogenetic tree using promoter sequences of naphthalene degradation genes.

Sequence variation in metabolic genes.

Primary DNA and protein sequences of CSV86 were compared with other closely related species for similarity in catabolic pathways using NCBI blast tools such as megaBLAST and BLASTp, respectively. Genes involved in the degradation of aromatic compounds in CSV86 were identified using RAST and NCBI PGAAP annotation along with available literature, KEGG [18] and Metacyc [19] databases.

Comparative genome analysis using BRIG and Mauve.

BRIG (BLAST Ring Image Generator) [20] software was used for the circular representation of multiple genome comparison. The draft genome of CSV86 was used as the reference genome and was compared with genome of P. putida S16 (NC_015733), P. putida KT2440 (NC_002947), P. entomophila L48 (NC_008027) and P. stutzeri CCUG 29243 (NC_018028).

Progressive alignment function of Mauve software with default settings was used to compare the homology among naphthalene degradation pathway genes reported from various Pseudomonas genomes. The draft genome of CSV86 was aligned against complete genome of P. stutzeri CCUG 29243 (NC_018028), Pseudomonas sp. ND6 plasmid pND6-1 (NC_005244), P. putida plasmid NAH7 (NC_007926) and P. fluorescens strain PC20 plasmid pNAH20 (AY887963).

Growth.

Strain CSV86 was grown on 150 ml minimal salt medium (MSM) [5] in 500 ml capacity baffled Erlenmeyer flasks at 30°C on a rotary shaker (200 rpm) supplemented aseptically with vanillin, veratraldehyde, ferulic acid, phenylalanine or tyrosine (0.1%). The cell growth was observed spectrophotometrically at 540 nm.

Preparation of cell-free extracts.

CSV86 cells grown on phenylalanine (0.1%) or glucose (0.25%) till late-log phase were harvested and washed twice with Tris-malaete buffer (200 mM, pH 6.0). Cells were re-suspended in ice-cold Tris-malaete buffer (1∶4 [wt/vol]) and sonicated at 4°C with four cycles of 15 pulses each (1 s pulse, 1 s interval, cycle duration 30 s, output 15 W) using an Ultrasonic processor (GE130). The cell lysate was centrifuged at 37,000 ×g for 30 min. The clear supernatant obtained was referred to as the cell-free extract and used as the source of enzyme. Protein was estimated by the method of Bradford [24] using bovine serum albumin as the standard.

Enzyme assay.

Homogentisate 1,2-dioxygenase activity was monitored by measuring the rate of O₂ consumption at 30°C using an oxygraph (Hansatech) fitted with a Clarke's O₂ electrode. The reaction mixture (1 ml) contained Tris-malaete buffer (200 mM, pH 6.0), homogentisate (2.5 mM) and an appropriate amount of enzyme. The enzyme activity was calculated as nmol of O₂ consumed per min. The specific activity is reported as nmol of O₂ consumed min⁻¹ mg⁻¹ protein.

Heavy metal resistance.

CSV86 was grown on 150 ml modified minimal salt medium (MSM, medium contained Tris, 8 g; KH₂PO₄,0.2 g; NH₄NO₃, 1 g; MgSO₄.7H₂O, 100 mg; MnSO₄.H₂O, 1 mg; CuSO₄.5H₂O, 1 mg; FeSO₄.7H₂O, 5 mg; H₃BO₃, 1 mg; CaCl₂.2H₂O , 1 mg; NaMoO₄, 1 mg; pH 7.5) in 500 ml capacity baffled Erlenmeyer flasks at 30°C on a rotary shaker (200 rpm) supplemented aseptically with naphthalene (0.1%) or glucose (0.25%) and appropriate concentration of heavy metals such as copper, cadmium or cobalt (0.5 or 1 mM) and the growth was monitored.

Naphthalene degradation pathway.

CSV86 can utilize naphthalene and its derivates such as 1- and 2-methylnaphthalenes as the sole source of carbon and energy via ring-hydroxylation pathway (Figure 3), while side-chain hydroxylation pathway leads to its detoxification [4], [5]. In CSV86, naphthalene catabolic pathway is initiated by naphthalene 1,2-dioxygenase (a three-component system) which catalyzes the hydroxylation of the aromatic ring to yield 1,2-dihydroxynaphthalene as a upper pathway (contig 105). This diol is further sequentially oxidized to catechol via lower pathway (contig 69), which enters the tricarboxylic acid cycle (TCA) after meta ring-cleavage (Figure 3) [4]. Using BLASTp, amino acid sequences of CSV86 naphthalene degrading upper and lower pathway genes (Table S2) were compared with that of other reported bacteria. It was observed that upper pathway amino acid sequences shares higher homology and hence are more conserved than that in lower pathway (Table S3).

Both the nah and sal operon of CSV86 showed similarity with Pseudomonas putida NCIB 9816-4 plasmid pDTG1 and Pseudomonas sp. ND6 plasmid pND6-1 (Figure 4A & 4B; Figures S2 & S3). Interestingly, in contig 105, the arrangement of genes observed was tRNA^Gly, integrase followed by nah operon in the order nahAa,Ab,Ac,Ad, BFCED. This arrangement is a characteristic feature of a GI for e.g. clc element [29]. The sal operon consists of 9 genes organized as nahGTHILMOKJ with the regulatory gene nahR present downstream of nahG gene. The regulation of nah genes is controlled by nahR, which is in turn induced by salicylate [30], [31]. In CSV86 a transposase encoding gene is present upstream of nahR gene which is missing in the plasmids being compared (Figure 4B & Figure S3).

Regulation of nah operon.

The genome sequence analysis of CSV86 revealed that the naphthalene pathway is under the control of LysR family of transcription regulators (LTTRs) [32], [33]. We have analyzed the differential regulation of naphthalene degradation pathway in CSV86 with P. stutzeri CCUG 29243 (NC_018028) (chromosomally coded) and Pseudomonas putida plasmid NAH7 DNA, strain G7 (AB237655). NahR protein is essential for the activation of both the upper and lower operon of naphthalene pathway in the presence of salicylate [34]. The binding site for NahR with the promoter for Pnah and Psal are reported to be located at 60 bp upstream to transcriptional start site [34], [35]. Therefore, the promoter data for nahAa (upper pathway), nahG (lower pathway) and also the coding sequences for NahR protein was analyzed. The consensus binding sequences of NAH7 promoter has two cis-acting elements situated 6 bp apart that interact with NahR protein [36]. The nahAa promoter of CSV86 and P. stutzeri are identical with the reported NAH7 binding site for NahR protein. Both these promoters have an additional cis-acting element with one base pair substitution and 4 bp spacing between the cis-acting elements. Whereas, nahG promoter has a base pair substitution (TGAT is changed to TAGT) in both the chromosomal promoters, with 4 bp separating the two cis-acting elements (Table 2). A phylogenetic tree of nahR, nahAa and nahG promoter sequences was also constructed. In all the three promoter sequence comparisons, CSV86 and P. stutzeri were grouped in same cluster (Figure S4).

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 2. Consensus sequence of nahR binding site in nahAa and nahG gene obtained from P. putida plasmid NAH7.

https://doi.org/10.1371/journal.pone.0084000.t002

The NahR of CSV86 interestingly showed 100% identity with protein from P. stutzeri (chromosomally located) as compared to 81% identity with NAH7 (plasmid encoded) protein. The substitution of methionine to isoleucine in NahR protein of NAH7 altered the specificity of protein to salicylate and allowing salicylate analog like benzoate to act as an inducer [37]. In CSV86 and P. stutzeri, NahR protein at 116^th position has isoleucine (Figure S5). However in CSV86, the enzymes responsible for naphthalene and salicylate degradation are inducible in nature. Benzoate does not induce these operons as the benzoate grown cells failed to respire on naphthalene or salicylate and showed no activity of enzymes involved in naphthalene or salicylate degradation [5].

Benzoate degradation pathway.

In CSV86 benzoate degradation is initiated with the incorporation of molecular oxygen by benzoate dioxygenase (encoded by benABC genes, a two-component system) to yield catechol which enters the central carbon metabolism via β-ketoadipate pathway after ortho-cleavage (Figure 3) [5], [38]. The details of the genes for benzoate degradation that are present in contigs 175, 103, 118 and 116 are described in Table S2. The CSV86 genome has the presence of complete benzoate utilization system including the regulatory option of the benABC operon i.e., transcriptional activator BenR with benzoate as effector; and utilization of catechol regulated via CatR with cis,cis-muconate as effector [38]. In P. fluorescnes Pf0-1 the catBC and catR genes are located between genes encoding benzoate MFS transporter and catechol 1,2-dioxygenase while, in CSV86 these genes are located upstream to the benzoate cluster. The catBC genes are absent in this cluster of P. putida KT2440 (Figure 4C, Figure S6).

The optional genes for benzoate degradation were also identified in CSV86 genome, which encode p-hydroxybenzoyl-CoA thioesterase (catalyses the conversion of p-hydroxybenzoyl-CoA to p-hydroxybenzoate) and 5-carboxymethyl-2-hydroxymuconate Delta-isomerase (catalyses the conversion of 5-carboxymethyl-2-hydroxymuconate to form 5-carboxy-2-oxohept-3-enedioate) located in contig 60 and 103, respectively. Besides salicylate hydroxylase in contig 69 (sal operon), contig 103 also showed the presence of an additional salicylate hydroxylase with 23% identity.

Aromatic alcohol degradation pathway.

Although the detoxification pathway of methylnaphthalenes closely resembles to the side-chain hydroxylation of toluene degradation, CSV86 failed to utilize toluene or xylene as the sole source of carbon and energy. Interestingly, strain could grow on benzyl alcohol, 2- and 4-hydroxy benzyl alcohol (Figure 3) [5]. The key enzymes of the aromatic alcohol metabolism, benzyl alcohol dehydrogenase (BADH) and benzaldehyde dehydrogenase (BZDH), have been purified and were found to be wide-substrate specific and shown to catalyze the conversion of 1- and 2-hydroxymethylnaphthalene to respective naphthoic acids (dead end products) [5], [6].

In CSV86, the gene cluster encoding BZDH and BADH was located in contig 119. The proposed BZDH or NAD⁺-dependent aryl aldehyde dehydrogenase gene was located adjacent to BADH, aryl alcohol dehydrogenase gene in CSV86 and shares homology with Pseudomonas putida DOT-T1E (aldehyde dehydrogenase, 87%). In CSV86 gene encoding for transcriptional regulator (AraC family) was located downstream to the gene encoding putative benzaldehyde dehydrogenase oxidoreductase protein, which is absent in the same cluster of Pseudomonas putida GB-1 and Burkholderia sp. (Figure 4D, Figure S7, Table S2).

Phenylacetic and p- hydroxyphenylacetic acid degradation pathway.

CSV86 metabolizes phenylacetic acid (PA) and p-hydroxyphenylacetic acids (HPA) using an unconventional ‘aerobic hybrid’ pathway (Figure 3). PA is first activated to its CoA-thioester (phenylacetyl CoA) by phenylacetyl-CoA ligase, which through a series of CoA-thioester intermediates ultimately leads to the formation of succinyl-CoA and acetyl-CoA that enters the central metabolic pathway (Figure 3) [7], [39]. In CSV86 the PA catabolic genes (contig 88, Table S2) were organized in the order as observed in Pseudomonas putida W619 and P. putida GB-1 (Figure 4E, Figure S8).

4-HPA degradation pathway in CSV86 follows homoprotocatechuate route which involves initial hydroxylation of 4-HPA followed by ring cleavage (Figure 3) [7]. The pathway genes of 4-HPA in CSV86 were present in contig 7, 120, 175 and 216 (Table S2). There were two copies of each hpaD (contig 7 and 120) and homoprotocatechuate degradative operon repressor gene (contig 7 and 216). The genes present in CSV86 contig 7 were organized as hpaAGEDFHI while, in Pseudomonas fluorescens SBW25 homoprotocatechuate degradative operon repressor gene was present upstream of this hpa cluster (Figure 4F, Figure S9).

p-Hydroxybenzoate degradation pathway.

In CSV86, p-hydroxybenzoate degradation is initiated by ring-hydroxylating 4-hydroxybenzoate 3-monooxygenase (PobA) to yield 3,4-dihydroxybenzoate (protocatechuate) which is further metabolized by ortho ring-cleavage to yield carboxy-cis,cis-muconate by protocatechuate dioxygenase (PcaGH, Figure 3). The later part of the pathway was catalyzed by the pcaBCDIJF gene products which are also involved in the degradation of benzoate.

The genes for 4-hydroxybenzoate pathway were distributed in contig 99 (pcaHG), 107 (pobA), 118 (pcaRKFTBDC) and 175 (pcaIJ, Table S2) in CSV86. Like Pseudomonas fluorescens Pf-5 and P. putida GB-1, pobA gene in CSV86 (contig 107) was located downstream to the gene encoding transcriptional regulator PobR (AraC family) (Figure 4G, Figure S10A). The pcaHG genes were located in contig 99 and present downstream of zinc metalloprotease superfamily, as observed in P. entomophila L48 and P. putida KT2440 (Figure 4H, Figure S10B). A transporter, PcaK, is involved in the transport of 4-hydroxybenzoate across the membrane and reported to be located in between pcaR and pcaF in P. putida, as can also be seen in contig 118 in CSV86 (Figure 4I, Figure S10C). The expression of pcaK gene has been shown to be repressed by benzoate, suggesting cells prefer benzoate instead of 4-HPA when given together [40]. The pca genes have been shown to be arranged in a single cluster in P. fluorescens [41], while in CSV86 they were segregated in different contigs (Figure 4, Table S2, Figure S10).

Phenylpropanoid degradation pathway.

Although lignin degradation pathways are shown in fungi by enzymes such as lignin peroxidases, manganese peroxidases, laccases etc [42]; aromatic degrading bacterial species are also reported to metabolize lignin [43]. Vanillin, ferulic acid, veratraldehyde, coniferyl aldehyde, β-coniferylether are identified as degradation intermediates of lignin. In Pseudomonas sp. strain HR199, the feruloyl-CoA synthetase (encoded by fcs) activates ferulic acid to its CoA-thioester followed by hydration and non-oxidative cleavage by enoyl-CoA hydratase/aldolase (encoded by ech) to form vanillin and acetyl Co-A [44]. Vanillin is further converted to vanillate and later protocatechuate by vanillin dehydrogenase (encoded by vdh) and vanillin monooxygenase (encoded by vanAB) or vanillate-O-demethylase (encoded by vanA and vanB), respectively [41], [44]. In CSV86, genes involved in the phenylpropanoid degradation were segregated in different contigs (Figure 5A&B, Table S2, Figure S11). The fcs-vdh-ech genes formed a cluster in contig 115 (Figure 5A, Figure S11A), vanAB genes in contig 119; and Vanillate O-demethylase oxygenase subunit, vanA and vanillate O-demethylase oxidoreductase, vanB along with the transcriptional regulator for ferulate or vanillate catabolism in contig 220 (Figure 5B, Figure S11B).

Strain CSV86 showed good growth on MSM supplemented with vanillin, veratraldehyde, or ferulic acid (0.1%). These observations suggest that the lignin degradation intermediates can be used as the sole source of carbon and energy and reflects the existence of functional phenylpropanoid metabolic pathway in CSV86. However, cells failed to utilize lignin (lignin sulphonic acid) as the sole source of carbon and energy (data not shown).

Homogentisate degradation pathway.

Homogentisate is a metabolic intermediate of aromatic amino acid pathways. Phenylalanine via tyrosine is converted to 4-hydroxyphenylpyruvate (by PhhABC). The generated 4-hydroxyphenylpyruvate, is then transformed to homogentisate (2,5-dihydroxyphenylacetic acid) by 4-hydroxyphenylpyruvate dioxygenase (encoded by hpd gene) [41]. Homogentisate 1,2-dioxygenase (HmgA) is the first enzyme of the homogentisate pathway which catalyses the transformation of homogentisate to 4-maleylacetoacetate. Isomerization of 4-maleylacetoacetate by maleylacetoacetate isomerase (HmgC) leads to the formation of fumarylacetoacetate, which is finally hydrolyzed by fumarylacetoacetase (HmgB) generating fumarate and acetoacetate [45], [46]. The hpd gene is present along with hmg genes in Pseudomonas syringae, Pseudomonas stutzeri and Pseudomonas mendocina, whereas in P. putida these are scattered; in P. aeruginosa the hpd gene is clustered with phh genes [41]. In CSV86, phh, hpd and hmg genes were segregated in different clusters (Figure 5C, D & E; Figure S12,Table S2). The phh genes were present in the contig 177 with the arrangement of phhR (phenylalanine hydroxylase transcriptional activator) and phhABC genes similar to P. putida KT2440 and P. putida F1 (Figure 5C, Figure S12C). The hpd gene was present in contig 99 (Figure 5D, Figure S12B) and 134. The gene coding for transcriptional regulator, TetR family, was present upstream to hpd gene in CSV86 (contig 99), P. fluorescens Pf-5 and SBW25. The clustering of hmg genes (contig 127) was similar to P. putida F1 and KT2440, with the gene coding for transcriptional regulator (IclR family) being transcribed in reverse direction to hmgABC genes (Figure 5E, Figure S12A).

Strain CSV86 showed good growth on MSM supplemented with phenylalanine or tyrosine (0.1%) as the sole source of carbon and energy. Cell-free extract prepared from the cells grown on phenylalanine showed the activity of homogentisate dioxygenase (specific activity 49.9 nmol min⁻¹ mg⁻¹protein) while glucose grown cells failed to do so. These results suggest that the homogentisate pathway is functional in CSV86 and the enzyme is inducible in nature (data not shown).