N. pentaromativorans US6-1 was cultured according to a method described previously [10]. A starter culture of bacteria was prepared by growing cells in marine broth (MB) at 30°Cto an optical density at 600 nm (OD₆₀₀) of 0.8. Bacteria were harvested aseptically and equal amounts of bacteria (culture of 500 ml) were added to fresh Bushnell-Hass broth (incorporating 30 g NaCl/L) (BD, USA) containing phenanthrene (50 ppm), pyrene (50 ppm), benzo(a)pyrene (50 ppm), benzoate (50 ppm), and p-hydroxybenzoate (50 ppm). OD₆₀₀ values of all cultures were checked at every six hours until 48 hours using spectrophotometer (Beckman coulter proteome Lab Du 800, USA). Bacteria were harvested after 12 or 36 h before suspending in 20 mM Tris-HCl buffer (pH 8.0) and then disrupting twice in a French pressure cell (SLM AMINCO, Urbana, IL, USA) at 20,000 lb/in². Supernatants (crude cell extracts) were collected by centrifugation (15,000×g, 45 min) and subjected to oxygenase activity assay and proteomic analysis. Protein concentrations were determined by the Bradford method [13]. Enzyme activity assay and proteome analysis was conducted on the basis of same protein amount.

The activities of catechol 1,2-dioxygenase (CD1,2), catechol 2,3-dioxygenase (CD2,3), protocatechuate 3,4-dioxygenase (PCD3,4), and protocatechuate 4,5-dioxygenase (PCD4,5) were determined using a UV spectrometer (Beckman Coulter Proteome Lab DU800, USA), as reported previously [14]. The activities of CD1,2 and CD2,3 were assayed by monitoring increases in concentration of cis, cis-muconate at A₂₆₀ and 2-hydroxymuconic semialdehyde at A₃₇₅, respectively. Activities of PCD3,4 was determined by monitoring the increase in concentration of β-carboxymuconate at A₂₉₀ (absorbance decreased as β-carboxymuconate concentrations increased) and PCD4,5 activity was measured through the increase in 2-hydroxy-4-carboxy-muconic semialdehyde at A₄₁₀, respectively. For each assay, one unit (U) of enzyme activity is defined as the amount of enzyme producing 1 mmol of product per min. Activity assay of each sample was conducted at least three times for each sample.

SDS-PAGE and in-gel digestion was performed as reported previously [15]. After electrophoresis and Coomassie blue staining, 1D-gels were divided into 10 fractions according to molecular weight. Sliced gels containing fractionated protein samples were digested with trypsin (Promega, Madison, WI, USA) for 16 h at 37°C after reduction with 10 mM dithiothreitol (DTT) and alkylation of cysteines with 55 mM iodoacetamide. The digested peptides were recovered with extraction solution (50 mmol/L ammonium bicarbonate, 50% acetonitrile, and 5% trifluoroacetic acid). The extracted tryptic peptides were dissolved in 0.5% trifluoroacetic acid prior to further fractionation by LC-MS/MS analysis.

LC MS/MS analysis was performed according to a modified version of a previously published method [16]. Tryptic peptide samples were loaded onto a 2G-V/V trap column (Waters, USA) for the enrichment of peptides and the removal of chemical contaminants. Concentrated tryptic peptides were eluted from the column and directed onto a 10 cm ×75 µm i.d. C18 reverse phase column (PROXEON, Denmark) at a flow rate of 300 nl/min. Peptides were eluted using a gradient of 0−65% acetonitrile for 80 min. All MS and MS/MS spectra were acquired in a data-dependent mode using a LTQ-Velos ESI Ion Trap mass spectrometer (Thermo Scientific, Germany). Each full MS (m/z range of 300 to 2,000) scan was followed by three MS/MS scans of the most abundant precursor ions in the MS spectra. For protein identification, MS/MS spectra were analyzed by MASCOT (Matrix Science, http://www.matrix science.com). The genome sequence database of N. pentaromativorans US6-1 (GI:359402640) was downloaded from NCBI and used for protein identification. Mass tolerance of the parent or fragment ion was 0.8 Da. Cabamidomethylation of cysteine and oxidation of methionine were considered to be variable modifications of tryptic peptides in the MS/MS analysis.

An alternative MS/MS analysis was conducted using a nano-ACQUITY Ultra Performance LC Chromatography™ equipped Synapt™ HDMS System (Waters Corporation, MA, USA) as described previously [17]. The flow-through peptides were applied to a nano-ACQUITY UPLC BEH300 C18 RP column (180 µm ×250 mm, particle size, 1.7 µm). Trypsin-digested peptide mixtures were loaded onto the enrichment column (180 µm ×20 mm, particle size, 5 µm), which was equilibrated with mobile phase A (3% acetonitrile in water with 0.1% formic acid) to remove salts and concentrate the peptides. Flow-through peptides were directly applied to a nano-ACQUITY UPLC BEH300 C18 RP column (180 µm ×250 mm, particle size, 1.7 µm) at a flow rate of 300 nl/min. The step gradient was as follows: 3−40% mobile phase B (97% acetonitrile in water with 0.1% formic acid) for 95 min, 40−70% mobile phase B for 20 min, and then a sharp increase to 80% mobile phase B for 10 min. MS data analysis was carried out with the continuum LC-MS^E data using ProteinLynx GlobalServer (PLGS) version 2.3.3 (Waters Corporation). The criteria for protein identification used in the PLGS search engine were applied with a peptide tolerance of 100 ppm, a fragment tolerance of 0.2 Da, and a missed cleavage allowance of 1. Analysis of quantitative changes in protein abundance (>95% confidence based on peptide ion peak intensities observed in low collision energy mode (MS) in a triplicate set) was conducted using Expression^™ Software version 2(Waters Corporation).

The emPAI values were used in the cluster analysis of all analyzed proteins, and the proteome dataset was z-transformed and median-centered normalized. Analysis of the proteome dataset was performed by Pearson correlation and average linkage hierarchical clustering by Cluster and TreeView [18]. All proteins identified by proteomic analysis were classified according to the cluster of orthologous groups (COG) functions, and their subcellular localization of the identified proteins was predicted by Cello (v. 2.0; http://cello.life.nctu.edu.tw/) [19]. Prediction of trans-membrane topology was performed using Phobius (http://phobius.sbc.su.se/) [20].

N. pentaromativorans US6-1 was grown in LB broth containing 10 g/L NaCl. Plasmid DNA was isolated using a standard alkaline lysis procedure, and purified on NucleoBond columns (NucleoBond Plasmid BAC Maxi kit, Clontech, USA). Genomic DNA was shared by careful pipetting of the DNA solution up and down several times with a 200 µl Pipetman tip. Purified plasmid DNA was analyzed in a 1% (w/v) agarose gel by clamped homogeneous electrical field (CHEF) gel electrophoresis (Bio-Rad, USA). CHEF parameters were set according to the manufacturer's protocol, including a circulating temperature of 14°C, electric current of 350 mA, and a pulse time of 35 s for 28 h. After electrophoresis, the gel was stained with ethidium bromide solution for 30 min, washed with TBE for 30 min, and DNA bands were detected using the Gel-Doc System (Bio-Rad, USA).

N. pentaromativorans US6-1 were pre-cultured in MB to obtain sufficient biomass to determine dioxygenase activity and proteomic analysis. Approximately equal quantities of cells were transferred into PAH and MAH culture media. Although all culture has same cell mass, each has different OD₆₀₀ values because of different absorbance of PAHs and MAHs. OD values are as follow; benzo(a)pyren (0.5146 at OD₆₀₀), pyren (0.5612 at OD₆₀₀), phenanthren (0.604 at OD₆₀₀), benzoate (0.4242 at OD₆₀₀), and p- hydroxybenzoate (0.4366 at OD₆₀₀). After 12 h incubation, delta OD values of benzo(a)pyren, pyren, phenanthren, benzoate, and p-hydroxybenzoate were +0.0101, +0.1461, +0.0628, −0.0347, −0.0199, respectively. After 36 h incubation, delta OD values of benzo(a)pyren, pyren, phenanthren, benzoate, and p-hydroxybenzoate were +0.0013, +0.0254, −0.008, −0.0492, +0.5695, respectively. The bacteria were harvested after 12 h and 36 h incubation for enzyme activity assay and proteome analysis. Highest delta OD values of benzo(a)pyren and pyren were +0.0101 and +0.1461 at 12 hours incubation, respectively. Phenanthren was continually increased until 48 hours (+0.0628). Unexpectedly, cell mass under benzoate culture condition was not increased, whereas degrading enzyme activities were significantly increased. The activities of four major dioxygenases (CD1,2, CD2,3, PCD3,4 and PCD4,5) were assayed using protein extracts from N. pentaromativorans US6-1 cultured in the presence of three PAHs and two MAHs to determine which biodegradation pathways were induced. Analysis of the N. pentaromativorans US6-1 genome indicated the presence of only extradiol oxygenase genes; however, many unspecified oxygenase genes were identified. In attempting to determine if N. pentaromativorans US6-1 also expressed intradiol oxygenase activities, we selected the four dioxygenase enzymes (CD1,2, CD2,3, PCD3,4 and PCD4,5) that we considered would cover most degradation pathways in aerobic cultures. No activity of the intradiol dioxygenases CD1,2 and PCD3,4 was detected. Activity of CD2,3 was high in cells cultured in media containing phenanthrene (1.59 U/mg at 12 h cultivation), benzoate (0.53 U/mg at 36 h cultivation), and p-hydroxybenzoate (0.31 U/mg at 36 hr cultivation), whereas, CD2,3 activity was detected to some degree in cells cultured under all conditions (0.02−0.04 U/mg), including MB media (Fig. 1). Activity of PCD4,5 was only detected in cells cultured in p-hydroxybenzoate (0.20 U/mg at 36 hr cultivation) media (Fig. 1). This suggests that the CD2,3 pathway could be a primary pathway for the degradation of PAHs and benzoate, while p-hydroxybenzoate is broken-down via the PCD4,5 or CD2,3 pathway in N. pentaromativorans US6-1.

Cytosolic proteins were prepared for shotgun proteomics using 1-DE/LC MS/MS from cells cultured in different PAHs. Approximately 650−718 proteins were identified in cells grown in the presence of phenanthrene, pyrene, and benzo(a)pyrene. Comparative analysis between the three PAHs was made with cells grown in MB as a control (Table S1 and Fig. S1). Analysis revealed that 494 proteins were commonly induced by all culture media, with 26−32 unique to the different aromatic hydrocarbons used as sole carbon sources. The identified proteins were divided into six groups (C1−C6) according to cluster analysis with each protein group arranged according to COG functions (Fig 2). Enzymes of the PAH and MAH catabolic pathways were clustered in group C2 and group C6, and were expressed at higher levels in the presence of phenanthrene. This was particularly noticeable for the degradation enzymes associated with ‘secondary metabolite biosynthesis, transport and catabolism [Q]’ (Table S2). Proteins in group C6 were strongly induced in those cells grown in MB and the primary up-regulated COG protein group was ‘Translation, ribosomal structure and biogenesis [J]’. The ribosomal proteins induced during growth in MB increased by more than 1.69-fold compared to PAHs. ‘Proteins uncharacterized or unknown [R or S]’ were relatively higher in C3 group proteins, which were abundant under benzo[a]pyrene culture conditions.

A notable outcome of the proteomic analysis was the strong induction of genes originating from pLA1. This large plasmid encodes 199 genes, and approximately 27 are clustered, considered to be a putative coding region for LMW aromatic-hydrocarbon degradation. These genes could regulate biodegradation of bicyclic aromatic ring compounds through to tricarboxylic acid cycle metabolites, such as acetaldehyde. Approximately 27−36 proteins coded by genes located on pLA1 were induced by three PAHs. Among the putative biodegradation genes, 24, 19, and 12 proteins were up-regulated in the presence of phenanthrene, pyrene, and benzo(a)pyrene, respectively, compared to MB (Table 1). Comparative results showed that the greatest amount of protein induction occurred in the presence of phenanthrene, which is consistent with the results of the dioxygenase activity assay (Fig. 1). Enzymes catalyzing the breakdown of bicyclic aromatic compounds to acetyl CoA were strongly induced. Ring-hydroxylating dioxygenase, dihydrodiol dehydrogenase, putative biphenyl-2,3-diol-1,2-dioxygenase, and catechol 2,3-dixoygenase were included. Although the activity of catechol 2,3-dixoygenase in those cultures containing pyrene was low, proteomic results showed a strong induction of both this enzyme and others involved in related biodegradation steps (Table 1). The signal intensity for those enzymes produced by plasmid pLA1 in the presence of benzo(a)pyrene was lower than that for the other PAHs phenanthrene and pyrene.

These results were verified by proteomic analysis using Nano-UPLC MS, confirming the up-regulation of N. pentaromativorans US6-1 biodegradation genes by phenanthrene, pyrene, and benzo(a)pyrene (Table 2). These results showed that the biodegradation genes located on pLA1 play a major role in the utilization of PAHs as sole carbon sources in N. pentaromativorans US6-1.

In response to the differential induction of extradiol dioxygenases of N. pentaromativorans US6-1 in the presence of benzoate and p-hydroxybenzoate, a comparative proteome analysis was conducted. Approximately 1,475 proteins were identified. 126 were induced by benzoate, and 85 by p-hydroxybenzoate (Table S1 and Fig. S1.). Genomic analysis revealed the presence of a CD2,3 gene and a ring-hydroxylating gene on pLA1, which are responsible for the initial step of the lower catabolic pathway of aromatic hydrocarbons. Notably, four PCD4,5 genes (small and large subunits) were found to be localized on the chromosome. Proteome analysis showed that when N. pentaromativorans US6-1 was cultured in p-hydroxybenzoate, two PCD4,5 and p-hydroxybenzoate degradation enzymes were induced (Table 3 and Fig. 3). These genes (gene number NSU_3623−NSU_3811) were clustered on contig 58 of the chromosome (Table S3). However, none of the three PAHs or benzoate induced the expression of PCD4,5 and p-hydroxybenzoate degradation enzymes, suggesting that the chromosomal-born protocatechuate pathway plays a role in the breakdown of p-hydroxybenzoate, but not in the degradation of PAHs or benzoate. Unexpectedly, the biodegradation genes on pLA1 were expressed in response to p-hydroxybenzoate, despite having no direct role in the breakdown of this compound.