Salicylate degradation by a cold-adapted Pseudomonas sp.

Abstract

Pseudomonas sp. strain MC1 was characterized as a cold-adapted, naphthalene-degrading bacterium that is able to grow in a broad temperature range of 5–30°C. MC1 harbors a catabolic plasmid, designated pYIC1, which is almost identical to the archetypal NAH7 plasmid from the mesophilic bacterium Pseudomonas putida G7. On pYIC1, the catabolic genes for naphthalene degradation are clustered in two operons: nahAa-Ab-Ac-Ad-B-F-C-Q-E-D encoding the conversion of naphthalene to salicylate, and nahG-T-H-I-N-L-O-M-K-J encoding the conversion of salicylate through meta-cleavage pathway to pyruvate and acetyl CoA. NahH, the bona fide extradiol dioxygenase in MC1 salicylate metabolism, is thermolabile and is a cold-adapted enzyme. The thermal profiles of the NahH enzyme activities expressed in different hosts indicate the presence of a factor(s) or mechanism(s) to protect the thermolabile NahH enzyme (100% aa identity with MC1 counterpart) in G7. Overall, the results reported in the present work suggest that the thermolabile NahH might be a product of the cold-adaptation process of MC1 and thus contribute to the survival and growth ability of MC1 on salicylate and naphthalene in cold environments.

Introduction

During the past half century, much research has centered on elucidating the aerobic degradation of naphthalene by mesophilic pseudomonads at the biochemical and molecular levels, and the details of those metabolic pathways have been well documented (Simon et al. 1993). In many cases, the degradative genes are found on plasmids with sizes ranging from approximately 80 kb to 175 kb, of which only about 30 kb is responsible for naphthalene degradation (Yen and Serdar 1988; Dennis and Zylstra 2004; Sota et al. 2006). As a representative example, the catabolic genes of the NAH7 plasmid from Pseudomonas putida G7 are clustered in two operons, the nah and the sal operons. The former encodes the conversion of naphthalene to salicylate, while the latter encodes the conversion of salicylate through the meta-pathway to pyruvate and acetaldehyde. In contrast, little in-depth work has been reported for naphthalene degradation by pseudomonads in cold environments, although pseudomonads have been most commonly isolated as naphthalene degraders from oil-contaminated environments (Aislabie et al. 2000; Whyte et al. 2002; Eriksson et al. 2003; Flocco et al. 2009).

Pseudomonas sp. strain MC1 (GenBank accession number KC470078) was originally isolated for its ability to grow on naphthalene as the sole carbon and energy source at temperatures below 10 °C from the wastewater treatment facilities at Korea’s King Sejong station (62°13′S 58°47′W) on King George Island, Antarctica. MC1 was deposited with the Polar and Alpine Microbial Collection (PAMC) established by Korea Polar Research Institute under the accession number PAMC 26573. Preliminary growth substrate-range experiments coupled with data from ring-cleavage enzyme assays suggest that MC1 initially metabolizes naphthalene to salicylate, with subsequent degradation through the meta-ring-cleavage pathway. Also, as shown in Fig. 1, MC1 growth curves using a temperature-gradient incubator (Advantec, Tokyo, Japan) demonstrated that MC1 (1) grows well over a wide range of temperatures (5–25 °C), (2) is unable to grow (although it can survive) at 37 °C, and (3) reaches highest growth yield at 5 °C. These observed growth characteristics are apparently unique to cold-adapted microorganisms. A preliminary report of this work has been presented elsewhere (Ahn et al. 2012). In this work, we characterize the ability of MC1 to degrade salicylate at the molecular and physiological levels.

Fig. 1

Growth profiles of Pseudomonas sp. MC1 on salicylate at various temperatures

Materials and methods

Bacterial strains, growth conditions, and chemicals

Pseudomonas sp. strain MC1 was grown on mineral salts basal medium (Stanier et al. 1966) containing 20 mM succinate or 5 mM salicylate at 15 °C for use in subsequent experiments. Escherichia coli TOP10 [F⁻ mcrA Δ(mrr-hsdRMS-mcrBC) Φ80lacZΔM15 ΔlacX74 recA1 deoR araD139 Δ(ara-leu)7697galU galK rpsL (Str^R) endA1 nupG] (Invitrogen, Carlsbad, CA) was used as the host for the general cloning of PCR products in pEXP5/CT-TOPO and for the maintenance of the resulting recombinant plasmids. E. coli BL21(DE3) [F⁻ ompT hsdS_B ®_B ⁻m_B ⁻) gal dcm (DE3)] (Invitrogen) was used as the host strain for the expression of recombinant plasmids that were constructed by cloning PCR fragments in pEXP5/CT-TOPO. E. coli strains for cloning were grown on Luria-Bertani (LB) medium at 37 °C. Ampicillin was added to the medium when needed at 100 μg/ml.

DNA manipulation

Total genomic DNA was isolated from MC1 following standard procedures (Asturias and Timmis 1993) with slight modification. Plasmid DNA was purified using the DNA-spin Plasmid DNA Purification Kit (iNtRON, Kyungki-Do, Korea). Agarose-gel electrophoresis was performed in TAE buffer (40 mM Tris, 20 mM acetate, 2 mM EDTA). Agarose plugs containing genomic DNA for pulsed field gel electrophoresis (PFGE) were prepared as described by Saeki et al. (1999) with slight modification. PFGE was performed using the Bio-Rad Laboratories CHEF-DR III system (Bio-Red, Hercules, CA). Gels [1.0% agarose in 0.5× TBE buffer (1× TBE is 89 mM Tris borate, 2.5 mM EDTA, pH 8.0)] were run at 6 V/cm at 14 °C. The pulse duration was increased from 1 s to 8 s during a 16 h run.

Heterologous expression of the NahH and NahT-NahH proteins

The coding sequence of nahH was amplified by PCR using primers nahH-F (5′-ATG AAC AAA GGT GTA ATG-3′) and nahH-R (5′-TTA GGT CAT AAC GGT CAT-3′). The coding sequence of nahT to that of nahH was amplified by PCR using primers nahT-F (5′-ATG TCA GAG GTC TTT GAA ATC ACT-3′) and nahH-R (5′-TTA GGT CAT AAC GGT CAT-3′). MC1 genomic DNA was used as the template. The PCR was performed in a 20-μl reaction mixture containing approximately 100 ng template DNA and 10 pmol each primer set with PicoMaxx High Fidelity PCR Master Mix (Stratagene, La Jolla, CA) according to the instructions of the manufacturer. The thermal cycling program was: 2 min hot start (95 °C), 35 cycles of 40 s denaturation (95 °C), 30 s annealing (nahH: 43 °C, nahT to nahH: 47 °C), 1 min/1 kb extension (72 °C), and a final 10 min extension (72 °C). The PCR products were cloned using the pEXP5/CT-TOPO TA expression kit (Invitrogen) according to the manufacturer’s instructions. Each recombinant plasmid was transformed into competent cells of E. coli strain TOP10 (Invitrogen) for propagation. The insertional orientation of the cloned fragment was confirmed by PCR. Properly oriented clones were then sequenced for confirmation and transformed into E. coli BL21(DE3) for expression.

Enzyme assay

Pre-cultures of E. coli BL21(DE3) containing the nahH and nahTH genes were prepared by inoculating three or four colonies, respectively, into 4 ml LB medium supplemented with ampicillin (100 μg/ml) and incubated overnight at 37 °C. All of the overnight culture was transferred to 200 ml LB supplemented with ampicillin and further incubated at 37 °C. To overexpress nahH and nahTH, respectively, in the E. coli host system, 1.0 mM (final concentration) IPTG was added when the bacterial cells reached the exponential phase (OD₆₀₀ = 0.5–0.8). The incubation was prolonged at 15 °C overnight, and the induced cells were harvested and washed in one-half volume 1× PBS [phosphate buffered saline: 140 mM NaCl, 2.7 mM KCl, 10 mM NaHPO₄, and 1.8 mM KH₂PO₄ (pH 7.4)]. The cells were chemically homogenized in Bugbuster Master Mix (Novagen, Madison, WI) according to the manufacturer’s instructions. Unbroken cells and cell debris were removed by centrifugation at 14,000 g for 30 min. The resulting supernatant was used as the crude enzyme solution. The extradiol dioxygenase activity was assayed spectrophotometrically by measuring, at the corresponding wavelengths, the increase in absorbance of each meta-cleavage product formed from the following substrates: catechol, λ_max = 375 nm, ε = 33,400 cm⁻¹ M⁻¹; 3-methylcatechol, λ_max = 388 nm, ε = 13,800 cm⁻¹ M⁻¹; and 4-methylcatechol λ_max = 382 nm, ε = 28,100 cm⁻¹ M⁻¹. The reaction mixture contained 100 mM phosphate buffer (pH 7.4) and the appropriate substrate at a final concentration of 0.4 mM. One unit of enzyme activity was defined as the amount of enzyme required to form 1 μmol product per minute. The protein content was determined using the Bradford method with bovine serum albumin as the standard.

Nucleotide sequencing and sequence analysis

The pYIC1 plasmid was isolated from MC1 cells grown in MSB liquid containing 20 mM succinate using QIAGEN Large-Construct Kit (QIAGEN, Hilden, Germany). DNA sequencing of the plasmid was outsourced to Cosmogenetech (Seoul, Korea). The annotation was done by merging the results obtained from the Rapid Annotation using Subsystem Technology (RAST) server (Aziz et al. 2008) and the Cluster of Orthologous Groups database (Tatusov et al. 2003).

Results

Identification, sequencing, and analysis of a catabolic plasmid from MC1

PFGE and subsequent sequencing analyses revealed that MC1 harbors a catabolic plasmid of 81,814 bp, designated pYIC1 (GenBank accession number NC_016644), that encodes naphthalene degradation (Fig. 2a,b). The catabolic genes of plasmid pYIC1 for naphthalene degradation were clustered in two operons consisting, respectively, of the genes (nahAa-Ab-Ac-Ad-B-F-C-Q-E-D; bases 29,510–38,903) encoding the conversion of naphthalene to salicylate and the genes (nahG-T-H-I-N-L-O-M-K-J; bases 44,137–53,412) encoding the conversion of salicylate through the meta-cleavage pathway to pyruvate and acetyl CoA (Fig. 3). Interestingly, the entire sequence of pYIC1 is almost identical to that of the archetypal NAH7 plasmid (82,232 bp, Fig. 2b) from the mesophilic bacterium P. putida G7 (Sota et al. 2006), although the former is 418 bp shorter than the latter. Careful examination of the 418 bp size difference, calculated simply by comparing the number of base pairs between NAH7 and pYIC1, revealed the following: (1) the first 54 and the last 65 bases of NAH7 were missing from pYIC1; (2) a total of eight bases of NAH7 were missing from pYIC1, while a total of four bases of pYIC1 were missing from NAH7; and (3) three short-size DNA stretches (28 bp, 32 bp, and 235 bp) of pYIC1 were tandemly duplicated in NAH7.

Fig. 2

a Pulsed field gel electrophoresis (PFGE) separation of pYIC1 plasmid from Pseudomonas sp. MC1 genome. Lanes: 1 λ ladder standard, 2 strain MC1. b Gene maps of pYIC1 and NAH7 plasmids constructed using the BASys web server (Van Domselaar et al. 2005). The 83 CDSs on pYIC1 are color-coded according to the Clusters of Orthologous Groups functional category

Fig. 3

Organization of the genes encoding naphthalene-degrading enzymes on pYIC1. The naphthalene degradation pathway is depicted briefly below the gene organization

Comparative analysis of the MC1 pYIC1 and the G7 NAH7 plasmids

In the NAH7 plasmid, nucleotide position 1 is set to the adenine of the start codon of parA, encoding a putative plasmid partitioning protein. Accordingly, the lack of the first 54 nucleotides in pYIC1 would result in the truncation of the ParA protein by 18 amino acids in the N-terminal region in MC1. An NCBI BLAST search revealed that the conceptually translated amino acid sequence of the MC1 ParA has significant sequence identity with the homologous protein Soj from Thermus thermophiles HB27 (Leonard et al. 2005), the crystal structure of which has been determined (Cordell and Löwe 2001; Hayashi et al. 2001; Sakai et al. 2001). According to the structure of the Soj protein, the truncated amino acids of the MC1 ParA belong to both structurally and functionally important regions of the protein, including the first β-strand, which is positioned in the middle of the β-sheet of the Rossmann fold protein, and a P-loop for ATP binding. It is therefore unlikely that the pYIC1-encoded parA gene produces a functional protein. In contrast, the lack of the last 65 bases in pYIC1 seems to cause no significant physiological changes in MC1, because it belongs to a non-coding region.

Except for the deletions corresponding to the T and C at positions 43,472 and 43,504, respectively, in NAH7, all of the single-base deletions occurred in non-coding regions of pYIC1. Indeed, the two coding deletions in pYIC1 caused a frameshift resulting in a premature stop codon at amino acid position 301, producing a protein that is shorter than the G7 NahR by 20 amino acids. NahR and NahR-like proteins belong to the large LysR family of transcriptional regulators, and are between 300 and 350 amino acid residues in length. Except for the size, little variation has been found in these proteins among the naphthalene-degrading bacteria investigated so far. In fact, the 300 amino acid NahR protein of MC1 is suspected to be functional based on the following observations: (1) the presence of many NahR proteins of the same size as revealed by BLAST search (data not shown), and (2) the compact C-terminal receiving domain of transcriptional regulator DntR of Burkholderia sp. DNT (Smirnova et al. 2004). DntR regulatory proteins, also belonging to the LysR family, consist of an N-terminal helix-turn-helix (HTH) region around 60–70 amino acid residues in length, which is involved in DNA binding, and a much larger C-terminal regulatory domain, where inducer binding occurs.

Thermolability of the catechol 2,3-dioxygenase encoded by nahH

The sequence analysis clearly showed that there are no genetic differences in the operons for naphthalene and salicylate degradation between MC1 and the mesophile G7. MC1 employs the meta (catechol 2,3-dioxygenase, C23O) pathway for catechol degradation during salicylate metabolism. Hence, in order to confirm that the pYIC1-encoded nah genes are bona fide genes for naphthalene degradation by MC1, the C23O encoded by nahH was chosen for a comparison of its activity with the meta-cleavage dioxygenase detected in MC1 when growing on salicylate. The nahH gene was amplified by PCR, cloned, and expressed in E. coli. Assessment of the C23O enzyme activity expressed in E. coli demonstrated that the enzyme activity profile of NahH was within a standard deviation of that of the C23O previously detected in MC1 cells during growth in the presence of salicylate. The C23O in MC1 displayed maximum activity against catechol (∼4600 U/mg protein, considered as 100%), compared to 4-methylcatechol (∼2990 U/mg protein, 65%) and 3-methylcatechol (∼2300 U/mg protein, 50%).

Because this indicates that the NahH enzyme is the bona fide C23O for salicylate metabolism by MC1, its thermal stability was examined by pre-incubating the assay mixture for 3 h without substrate at 0, 15, and 30 °C. As shown in Fig. 4a, the higher the incubation temperature, the more rapid was the observed decay of the enzyme activity. Pre-incubation at 0 °C showed a slight negative effect on the initial enzyme activity against catechol, while pre-incubations at 15 and 30 °C resulted in approximately 40% and 90% loss of enzyme activity, respectively, after 3 h. When pre-incubated at 30 °C, the NahH enzyme retained less than 60% of the initial value after only 30 min.

Fig. 4

Thermostability of the NahH enzyme when a expressed alone and b coexpressed with NahT. Relative activity against catechol was measured in sequential 30 min periods during 3 h of incubation of each enzyme solution at 0, 15, and 30 °C

It is generally known that C23O enzymes, including NahH, are vulnerable to inactivation by the oxidation of the active-site ferrous iron, and many aromatic-degrading bacteria have a chloroplast-type ferredoxin that specifically reactivates C23O (Hugo et al. 2000). Among the open reading frames identified on pYIC1, nahT, which is present immediately upstream of nahH, was annotated to encode such a putative ferredoxin. Accordingly, the contiguous nahTH genes were amplified, cloned, and expressed in E. coli to examine the effect of NahT on the NahH stability. Indeed, when coexpressed with NahT, the NahH enzyme displayed almost constant levels of activity over the 3 h pre-incubations at 0 and 15 °C, respectively. At 30 °C, however, NahT exerted only a marginal protective effect (Fig. 4b). These results indicate that the thermolability of the NahH enzyme is a major cause of the observed decay in enzyme activity at 30 °C.

Discussion

Pseudomonas sp. strain MC1, originally isolated as a naphthalene degrader, was characterized as having a broad temperature range of 5–30 °C for growth on salicylate, showing the highest growth yield at 5 °C, albeit with a longer lag phase (Fig. 1). When the partial 16S rRNA gene sequence of MC1 was analyzed using the NCBI database, it showed high homology with those of Pseudomonas spp. strains (above 98% similarity) from various isolation sites in different environments. Further phylogenetic analysis of the 16S rRNA sequences showed that the Pseudomonas isolates from cold regions such as permafrost, glacier, and polar seawater clustered close together, producing a distinct group. In this group, MC1 constituted a subgroup of strains mostly from the Antarctic (Fig. 5). Accordingly, one can reasonably hypothesize that MC1 is a cold-adapted bacterium that uses a thermolabile C23O enzyme to metabolize salicylate.

Fig. 5

Bacterial phylogenetic tree constructed based on partial 16S rRNA genes of Pseudomonas strains. The sequences were obtained from the 16S ribosomal RNA sequence database on NCBI. The bacterial strain names and information about their isolation sites are indicated in parentheses, with those from the Antarctic marked with black dots. The phylogenetic tree was constructed using MEGA software (version 4.0) with the neighbor-joining method, and bootstrap values (> 50) from 1000 replicates are shown for each node. Scale bar 0.001 substitution per site and unit time

If the above hypothesis is true, the next question is about how the mesophilic G7 is able to grow normally on salicylate at 30 °C using the same NahH enzyme. In order to address this issue, NahH enzyme activities were examined in the cell-free extracts of strains MC1 and G7 grown on salicylate under the same pre-incubation conditions. Indeed, 3 h pre-incubation led to a 40% loss of the enzyme activity in MC1, while the enzyme activities were maintained at almost the same level in G7 (Fig. 6a,b). Overall, the thermal profiles of the NahH enzyme activities expressed in different hosts strongly suggest the presence of a factor(s) or mechanism(s) to protect the thermolabile NahH enzyme in G7.

Fig. 6

Thermostability of catechol 2,3-dioxygenase activity in crude extracts from a MC1 and b G7. Relative activity against catechol was measured in sequential 30 min periods during 3 h of incubation of each enzyme solution at 0, 15, and 30 °C

Because cold-adapted bacteria are known to play a key role in the degradation of aromatic compounds in cold environments (Margesin et al. 2003; Labbé et al. 2007), information about the cold-adaptation mechanisms of those bacteria could be the basis for a better understanding of the in situ biodegradation of organic pollutants. For growth and survival in the cold, psychrophilic bacteria should adapt all of their cell components, both structurally and functionally, to the cold as follows: outer membranes (lipid composition), intra cellular machines (ribosomes), and enzymes and nucleic acids (tRNA) (Cavicchioli 2006). It has been suggested that they likely possess a unique set of genes encoding psychrophilic enzymes and a collection of synergistic changes in their overall genome content and amino acid composition (Methé et al. 2005). The synthesis of cold-adapted catabolic enzymes is believed to be one of the main bacterial cold-adaptation mechanisms (Collins et al. 2008). In the same context, the MC1 C23O appears to serve as supporting evidence for the hypothesis of a cold-adaptation mechanism at the enzyme level.