**1. Introduction**

Environmental spills with recalcitrant pollutants such as mono- and polycyclic aromatic hydrocarbons (MAHs, PAHs) are a problem worldwide, threatening the environment and human

health [1]. Many of the diverse group of aromatic hydrocarbons are classified as priority pollutants by the US Environmental Protection Agency [2]. Naphthalene has also been classified as a group C possible human carcinogen. When oil spill accidents occur on land, degradation of petroleum hydrocarbons by indigenous microorganisms is often a slower process to remediate contaminated areas in comparison to traditional physical and chemical remediation treatments, often due to low microbial numbers and activity [3]. Microbial bioremediation, the process of pollutant degradation by microorganisms, is a green, cheap, and safe approach to cleaning up polluted sites [4,5]. This process can be improved by promoting the growth of endogenous bacteria in the polluted soil itself (biostimulation) or by introducing hydrocarbon-degrading bacteria (bioaugmentation) [6,7]. Frequently, due to strict legislations that impede the introduction of non-indigenous microorganisms at the site of interest, bioaugmentation is implemented by the use of native strains, or strains which carry the degradative genes on plasmids which can be transferred to the indigenous population though means of natural gene transfer. Studies demonstrated how this approach is considered the best choice for bio-detoxification of soils with a low degradation potential by the indigenous bacterial communities [8,9].

Undoubtedly, it is a challenge to find an e ffective and e fficient method to remediate polluted soils [10], especially because of the complexity and specificity of each site. Oil spillages cause profound and persistent changes in soil properties. pH may vary, as well as nutrient availability (e.g., carbon, nitrogen), with high concentrations of mono- and polycyclic aromatic hydrocarbons, often in combination with heavy metal pollution (nickel, lead, chromium, copper, zinc, cadmium). All these factors a ffect and re-shape the structure of indigenous microbial communities [11,12]. In this highly complex framework, the study and isolation of novel native ring hydroxylating bacterial strains constitutes an approach with high potential and a powerful alternative to the traditional physical and chemical remediation approaches.

In this study we combined traditional microbiological methods to isolate and characterize potential degraders from an ancient oilfield, combined with a bacterial genomic investigation aiming to reveal the genetic backgrounds of two interesting hydrocarbon-degrading bacteria belonging to the genus *Pseudomonas*. Our study aims to elucidate (1) the degradation pathways of linear and aromatic hydrocarbons, (2) plant growth promotion traits (hormones and stress enzyme reduction), and (3) the adaptation of the two identified *Pseudomonas* spp. to life in soil, tolerance to various stresses, and extraordinary degradation capacities of micropollutants.

#### **2. Materials and Methods**

#### *2.1. Isolation of Hydrocarbon-Degrading Bacteria Using a Sandwich Di*ff*usion System*

In June 2016, we sampled the historical oilfield forest soil in Bóbrka, Poland (Latitude: 49.616449; Longitude: 21.710454; Altitude: 350.45 m). The Bóbrka sampling site is unique in the world since it has one of the oldest oil wells still in production since 1854, and still pumps one barrel a day. The black top soil layer near an oil pump (Figure 1) was sampled with a sterilized shovel and stored in self-sealed bags. These bags were kept in temperature-controlled boxes and shipped to Hasselt University.

In the lab, one gram of soil was added into 15-mL sterile falcon tubes containing 5 mL of 0.1 M phosphate-bu ffered saline (PBS, pH = 7.0). After vortexing, tubes were incubated for one hour on an orbital shaker at 120 rpm to allow the microorganisms to e ffectively being released from the soil particles. In the meantime, Teflon components, steel screws, and hydrophilic membranes of a custom-made sandwich di ffusion system (Figure S1) were sterilized by autoclaving (120 psi, 30 min). Each sterilized "perforated plate" was dipped into a mixture of Bushnell Haas medium [13] with 0.7% w/v Gelzan (G1910; Sigma-Aldrich, St. Louis, Missouri, USA) supplied with 0.1% CaCl2 and the soil dilution to obtain a 10-<sup>5</sup> concentration of the soil suspension. This dilution was chosen to obtain the growth of one or few bacterial cells per well. The mixture was left to solidify under a sterile laminar flow after which the membranes were placed on both sides of the agar-filled plates (WHA111103; Nucleopore ®polycarbonate track-etched membrane; 47-mm diameter, 0.05-μm pore

size, Whatman, Maidstone, UK). The membranes allow the exchange of nutrients and molecules from the soil to the water-agar containing bacterial cells but not the passage of bacterial cells. Finally, the sandwich diffusion systems were closed with screws and buried into the wetted soil and incubated at 20 ◦C for one month.

**Figure 1.** Sampling site of Bóbrka with spilled black crude oil.

After one month of incubation, the sandwich systems were disassembled and each agar plug was pushed out with sterile toothpicks into a sterile deep-well masterblock filled with Bushnell Haas medium and incubated overnight at 30 ◦C on an orbital shaker. The next day, 100 μL of each bacterial suspension were plated onto 1/10 869 rich medium [14] supplemented with 2 mM indole according to Nagayama et al. 2015 [15]. After incubation for seven days at 30 ◦C, the colonies producing a dark-blue pigment were selected as positive for aromatic compound ring hydroxylase activity (conversion of the colourless indigo to dark-blue indole by dioxygenase enzymes) [15]. *Pseudomonas putida* f1 [16] and *Pseudomonas aeruginosa* WatG [17] were used as positive controls.

#### *2.2. Growth Conditions and Genotypic Characterization of the Bacterial Strains*

Bacteria stored in a 96 well masterblock (Greiner Bio-One, Kremsmünster, Austria) and those scoring positive on the indole test were grown in 5 mL of 869 medium [14] and incubated for 24 h at 30 ◦C while shaking at 200 rpm. Two mL of culture broth were used for total DNA extraction using the Qiagen Blood and Tissue kit (Qiagen, Venlo, Netherlands). The quality of the purified genomic DNA was checked using a NanoDrop 1000 UV-Vis spectrophotometer (Thermo Fisher Scientific, Wilmington, DE, USA). The 27F (5 AGAGTTTGATCMTGGCTCAG 3) and 1492R (5 TACGGYTACCTTGTTACGACTT 3) primers were used for the amplification of the 16S rRNA gene [18]. The PCR master mix consisted of: DNA template (10 ng μL−1), 1 × High Fidelity PCR buffer (Invitrogen, Carlsbad, CA, USA), 0.2 mM dNTPs, 2 mM MgCl2, 0.2 μM each of the forward and reverse primers, and 1 U High Fidelity Platinum Taq DNA polymerase (Invitrogen, Carlsbad, CA, USA) per 50 μL. PCR conditions were set as follow: denaturation at 94 ◦C for 5 min, 30 cycles of 94 ◦C for 1 min, 54 ◦C for 45 s, and 72 ◦C for 1.5 min, followed by a final extension of 10 min at 72 ◦C. Confirmation of PCR product amplification was tested by running a 1% agarose gel electrophoresis. The PCR products for each strain were sent to Macrogen (Amsterdam, The Netherlands) for 16S rRNA gene Sanger sequencing. Sequences were quality checked using Geneious v4.8.5 (Biomatters ApS, Aarhus, Denmark) and blasted against the nucleotide sequences present in the Ribosomal Database Project (RDP Release 11).

#### *2.3. BOX-Fingerprinting of Potential Hydrocarbon Degraders and nahAc PCR Amplification*

BOX-fingerprinting was performed on the dark blue pure colonies from the indole test. The BOX\_A1R primer (5-CTACGGCAAGGCGACGCTGACG-3), as described by Rademaker (1997), was used in the PCR reaction mixture described above. The reaction started with denaturation for 10 min at 95 ◦C, followed by 30 cycles of denaturation for 1 min at 95 ◦C, annealing for 1 min at 50 ◦C, and elongation for 8 min at 68 ◦C, with the final elongation for 8 min at 68 ◦C. PCR products were visualized on a 1.5% agarose gel stained with gel-red [19].

Naphthalene dioxygenase Fe-S large subunit (NahAc) genes were amplified with the primers nah-F (5 CAAAARCACCTGATTYATGG 3) and nah-R (5 AYRCGRGSGACTTCTTTCAA 3) using PCR conditions as described by Baldwin et al. 2003 [20]. PCR products of the correct size were purified and sent to Macrogen (Amsterdam, The Netherlands) for Sanger sequencing. Multiple sequence alignment was performed with Clustal Omega.

#### *2.4. Genome Sequencing and Assembly*

RNA-free DNA was extracted from stationary phase cells of strains *Pseudomonas* sp. VI4.1 and *Pseudomonas veronii* VI4.1 grown in 869 rich broth prior to digesting and ligating sequencing adaptors and barcodes using an Ion Xpress Plus Fragment Library Kit (Life Technologies Inc., Burlington, ON, Canada). Adaptor-ligated DNA was size-selected to 480 bp on a 2% E-Gel SizeSelect agarose gel, and Agencourt MAPure XP beads (Beckman Coulter, Mississauga, ON, Canada) were used for purification. The library dilution factor was determined using an Ion Library Quantitation Kit prior to amplification and enrichment with an Ion PGM Template Hi-Q OT2 kit on an Ion OneTouch 2 system. The enriched Ion Sphere Particles were quantified using an Ion Sphere Quality Control Kit prior to sequencing on a 316v2 chip with an Ion PGM Hi-Q View Sequencing Kit on an Ion Torrent PGM (Life Technologies Inc., Carlsbad, CA, USA).

Sequencing of strain VI4T1 generated a total of 1.1 million reads (mean length 300 bases, or 331 Mb of data (>307 M Q20 bases) in Torrent Suite 5.0.5. For strain VI4.1, 1.24 million reads (mean length 299 bases) generated 371 Mb data (>344 M Q20 bases). Reads were assembled using SPAdes 3.11.1 (uniform coverage mode; kmers 21, 33, 55, 77, 99, 127) into contigs greater than 1000 bp [21]. The assembly of strain VI4T1 resulted in 211 contigs with consensus length of 7,150,343 bp (>1000 bp) at 27x coverage (60.54% G+C content, largest contig of 237,158 bp; and N50 = 68,401 bp). Assembly of VI4.1 resulted in 185 contigs with a consensus length of 7,346,306 bp at 26.0X coverage (60.02% G+C content; largest contig of 285,953 bp; N50=73,645 bp). Plasmids were predicted using plasmidSPAdes, with kmers 21, 33, 55, 77, 99, 127 [22].

#### *2.5. Genome Annotation and Phylogenetic Tree of the 16S rRNA Gene Sequences*

Open Reading Frame (ORF) prediction and gene annotation were performed using the RAST annotation system (Overbeek et al. 2014) [23], Prokaryotic Genome Automatic Annotation Pipeline (PGAP) of the National Center for Biotechnology Information (NCBI) [24] and the platform MicroScope using the tool Magnifying Genomes, MaGe (http://www.genoscope.cns.fr) [25]. Clusters of Orthologous Genes (COGs) [26] and metabolic pathway reconstruction was carried out using the databases of Kyoto Encyclopedia of Genes and Genomes (KEGG) [27] and MetaCyc [28].

Comparative genome analyses were performed in SimpleSynteny, which uses NCBI BLAST, BioRuby and RMagick to map genes onto genomes and generate figures [29]. Default settings were used with BLAST E-value of 0.001 and 50% minimum query coverage.

The in silico DNA–DNA hybridisation (DDH) values between VI4.1 and VI4T1 and closely related strains was calculated using the online Genome-to-Genome distance calculator version 2.1 (GGDC) (http://ggdc.dsmz.de/ggdc\_background.php#) [30,31]. Circular maps of the chromosomes and plasmids were generated in MaGe and in Bandage version 0.8.0 [32]. Clustered regularly interspaced short palindromic repeats (CRISPRs) were identified using the web service (http://crispr.u-psud.fr/ Server/CRISPRfinder.php).

The bioinformatics prediction of genes associated with the biodegradation of naphthalene, BTEX (Benzene, Toluene, Ethylbenzene, Xylene), and aliphatic hydrocarbons was based on Basic Local Alignment Search Tool (BLAST) searches, Protein family (PFAM) queries and conserved

domain searches, using the databases MetaCyc, UniProt database (http://www.uniprot.org/) and NCBI. Visualisations of the plasmid and operon structures was performed with SnapGene version 3.2.1.

A phylogenetic tree with VI4.1, VI4T1, and closely related strains evaluated against the Ribosomal Database was built in MEGA7 [33]. The evolutionary distances were computed using the Maximum Composite Likelihood method [34].

#### *2.6. Biochemical and Chemotaxonomic Identification*

The biochemical metabolite profile of VI4T1 and VI4.1 was determined using Gen-III MicroPlates (Biolog, Hayward, USA). Bacteria were grown in 869 rich medium overnight, subsequently resuspended in sterile 10 mM sodium phosphate bu ffer (pH 7.2) to an Optical Density (OD) of one checked at 600 nm. Next day, 100 μL of each strain was inoculated into each well of the Gen-III microplate and incubated at 30 ◦C for five days. Each microplate allows to carry out 94 phenotypic tests: 71 carbon source utilization assays and 23 chemical sensitivity assays. Each well contains a redox tetrazolium dye that changes colour as a result of cellular respiration providing a reliable metabolic fingerprint, which can be evaluated by measuring absorbance at 595 nm.

In addition, a GC Analysis of total fatty acid methyl esters (FAMEs) was performed at EMSL Analytical (Cinnaminson, NJ, USA) to identify the strains based on the FAME fingerprint and using the Sherlock Microbial Identification system (MIS).

#### *2.7. In Vitro Estimation of Diesel Oil Degradation Capabilities*

The bacterial strains were tested in triplicates for their capability to use diesel as the only carbon source using the 2,6-dichlorophenol indophenol (DCPIP) assay [35]. DCPIP is a compound with high affinity for electrons which when oxidized by metabolic reactions turns from blue to colourless. In brief, bacteria were pre-cultured in 5 mL of 869 rich medium at 30 ◦C and 160 rpm on an orbital shaker until OD660 nm = 1.0. Cells were pelleted by centrifugation (4000 g for 20 min), washed three times with 10 mM MgSO4 and incubated overnight at 30 ◦C to allow the bacteria to use all remaining carbon source traces (starvation). Subsequently, 750 μL of Bushnell and Haas medium supplemented with 50 μL DCPIP solution (100 μg ml−1) were added to a 2-mL sterile microcentrifuge tube. Subsequently, 80 μL of cell suspension and 5 μL filter sterilized diesel were added. Cells were cultivated in the dark to avoid photodegradation of the redox dye (30 ◦C; 120 rpm) for 1 week. The colour of the reaction medium was compared with three controls: two negative controls with respectively no diesel and no bacteria and a positive control with *Pseudomonas aeruginosa* WatG strain [17]. Strains were scored as positive for microbial hydrocarbon degradation capability if the solution looked clear, and negative if it persisted in blue.

#### *2.8. Analysis of Naphthalene Degradation Using High-Resolution Proton Transfer Reaction Time-of-Flight Mass Spectrometry (PTR-TOF-MS)*

The capabilities of the strains to use naphthalene as sole carbon source was assessed by measuring naphthalene degradation using PTR-TOF-MS. 20 mL sterile glass vials with 1.5 g of sterilised sand were inoculated with 200 μL of bacteria suspension (O.D.=1) in Bushnell Haas medium without carbon source. Vials were spiked with 0.3 μL of naphthalene against the inner glass wall to saturate the head-space and incubated for 1 week at 25 ◦C. A set of vials with the same conditions but no bacteria was used as a control. All experiments were carried out in triplicate.

For PTR-TOF-MS analyses, the headspace of the vials was sampled through the Teflon septum with a 50-mL glass syringe (Z314587, Fortuna Optima, Sigma-Aldrich) equipped with a three-way stopcock attached to the syringe Luer lock tip. The sample was injected into the air inlet of the PTR-TOF-MS instrument (PTR-TOF 8000, Ionicon, Innsbruck, Austria). The instrument drift tube was operated at a field density ratio (E/N) of ≈130 Td, resulting from 2.2 mbar pressure, 80 ◦C temperature, and 530 V of electric potential. In short, the air sample met a rich mixture of H3O+ that protonated the

volatiles contained in the air sample. The reaction rate coe fficient between the hydronium ion (H3O+) and naphthalene was assumed 2.45 × 10−<sup>9</sup> cm<sup>3</sup> s<sup>−</sup><sup>1</sup> [36].

#### *2.9. In Vitro Plant Growth Promotion Activity of the Bacterial Strains*

Bacteria were screened in vitro for plant growth promoting traits including 1-aminocyclopropane-1-carboxylate (ACC)-deaminase, siderophore production, acetoin, organic acid, and indole-3-acetic acid (IAA) production, phytate mineralization and nitrogen fixation. The production of ACC-deaminase was estimated by monitoring the amount of α-ketobutyrate generated by the enzymatic hydrolysis of ACC [37]. Siderophore release was evaluated via Chrome Azurol S (CAS) assay [38]. The production of the phytohormone IAA was estimated by a colorimetric assay using Salkowski's reagen<sup>t</sup> [39]. The production of the volatile plant growth promoting compound acetoin was assessed using the Voges-Proskauer assay [40]. The evaluation of the results was based on the observations of the colorimetric reactions after 5 days of incubation at 30 ◦C. Phytate mineralization was evaluated after 12 days of incubation by observing the halo produced around the colonies growing on solid medium supplemented with Na-phytate [41]. Organic acids produced by plant-associated bacteria facilitate the solubility of nutrients in soil, thus facilitating the uptake of nutrients by the plant. The method of Cunningham and Kuiack was used to check their production [42]. Bacteria able to fixate nitrogen possess the enzyme nitrogenase which catalyses the reduction of atmospheric nitrogen to ammonium that can be detected by a colour change on semisolid medium without any nitrogen source [43].

#### *2.10.* Pseudomona *VI4.1 and VI4T1 Motility*

Bacterial motility was tested in sterile glass tubes previously filled with 869 semi-solid agar medium. Inoculation was performed with a sterile straight wire, making a single stab down the centre of the tube to about half the depth of the medium. Subsequently, the tubes were incubated at 30 ◦C and observed for bacterial growth at regular intervals (1 until 7 days). Non-motile bacteria generally grow into the stab-line, have sharply defined margins and leave the surrounding medium clearly transparent while motile bacteria typically grow by di ffusing throughout the medium, rendering it slightly opaque.

#### *2.11. Sequence Database Accession Numbers*

The genome sequencing projects have been deposited at DDBJ/EMBL/GenBank with strain VI4T1, Accession Number MULN00000000 and BioProject PRJNA369437, and for strain VI4.1, Accession Number NZ\_MULM00000000 and BioProject PRJNA224116. The Genbank accession numbers for the sequenced naphthalene dioxygenase genes are MN030640-MN030657 with associated partial 16S rRNA gene sequences of the isolates: MN006582-MN006610. The other bacterial isolates with 16S rRNA gene sequence identification were also deposited in NCBI Genbank with accession numbers: MN030267-MN030332.

#### **3. Results and Discussion**

#### *3.1. Genotypic and Functional Characterization of the Cultivable Hydrocarbon-Degrading Bacterial Community from the Oilfield of Bóbrka*

In search for novel hydrocarbon and plant growth promoting strains, we performed an isolation and selective enrichment approach to identify and characterize novel microorganisms capable of degrading oil-related pollutants. Seventy bacterial strains were isolated by using the sandwich di ffusion system which had incubated for one month in polluted soil from Bóbrka with crude oil (Figure 2A). 16S rRNA Sanger sequencing showed a dominance of *Pseudomonas* spp. (35%) followed by *Achromobacter* (18%), *Mycobacterium* (12%), *Caulobacter* (6%), *Burkholderia* (4%), *Enterobacter* (3%), *Ralstonia* (3%), and *Stenotrophomonas* spp. (3%) (Figure 2A). A high percentage (60%) of the strains were able to use diesel as sole carbon source. A majority of the strains produced ACC-deaminase (53%). ACC-deaminase

is an important trait of many soil and plant-associated bacteria as they can facilitate growth and development of the plant by lowering stress ethylene levels [44]. 33% of the strains produced the hormone indole-3-acetic acid. Strains producing IAA can have a phyto-stimulation effect on the host plant or pathogenic, depending on the concentration [45]. Bacteria use this hormone to interact with the host plants as part of their colonization strategy and to circumvent the basal plant defence mechanisms. It can also serve as signalling molecule between bacteria. Siderophore production and organic acid production was present in 18% and 36% of the isolated strains, respectively (Figure 2B). This can aid in the solubilization of iron and phosphorous in soil, making it more available to microbial and plant uptake. Only 6% of the strains were scored as positive for acetoin production. This ketone showed to enhance plant growth by stimulating root development and increasing resistance against pathogens and drought stress. An overview of the traits per species is shown in Figure 2C. The PGP and diesel use as sole carbon source traits were diversely spread across the different taxa. Strains with multiple PGP traits are most interesting for in vivo plant growth promotion tests, such as *Pseudomonas veronii* YOB01, *Burkholderia cepacia* YOAG01, *Mycobacterium* YOBG04, and *Achromobacter* YOBB03, amongs<sup>t</sup> others.
