Materials and microorganisms

Two A. xylosoxidans strains used in this study were previously isolated from the UAE environment [18] A. x. C2 was isolated from soil samples in an old-planted forest in Al Wathbah, Abu Dhabi, UAE and A. x. KW38 from oil contaminated sites at the Khorfakkan Port, Sharjah, UAE. Microbial samples were collected from abandoned agricultural lands not under active management or private ownership. These former farmlands are publicly accessible with no access restrictions. No permits were required for sample collection as these sites are not classified as protected areas. The non-native tree species present at the sampling locations are common throughout the UAE and not classified as endangered. No plant material was used in this research. Pseudomonas paraeruginosa (ATCC 9027) (P. paraeruginosa) by Rudra et al. (2022) [19] and Escherichia coli (ATCC 8739) (E. coli) were purchased from the American Type Culture Collection (ATCC) (Manassas, VA, USA). All strains were revived from glycerol stock by streak plating on LB agar, followed by incubating at 37 ˚C for 24–48 hrs. Single colonies were picked and grown in LB broth at 37 ˚C for 48 hrs. The culture was subsequently diluted with fresh LB broth to an OD₆₀₀ of 0.2, representing the exponential growth phase, for use in subsequent experiments.

Glucose, LB nutrient broth, LB agar, agar-agar, copper (II) chloride, cobalt (II) chloride hexahydrate, boric acid, manganese (II) chloride, ethylenediaminetetraacetic acid (EDTA), zinc chloride, iron (III) chloride, thiamine-HCl, biotin, calcium chloride dihydrate, magnesium sulfate heptahydrate, di-Sodium hydrogen phosphate dodecahydrate, monopotassium phosphate, sodium chloride, ammonium chloride, sodium hydroxide, analytical grade bisphenol A (BPA), 4-hydroxybenzoic acid (4-HBA), 1-naphthalenemethanol (1-NM), and pyrene (PYR) were purchased from Sigma-Aldrich (Burlington, MA, USA). Acetone and toluene were obtained from Honeywell Inc. (Charlotte, NC, U.S.A).

Media and carbon source preparation

Minimal salts M9 media and agar plates were prepared according to the manufacturers’ protocol and supplemented with 2 mg/mL glucose [20]. Carbon sources were prepared by solubilizing 2 mg of each of BPA, 4-HBA, and 1-NM in 10 µL 100% acetone. Acetone was allowed to evaporate before adding them to 1 mL of 2 mg/mL of glucose-supplemented media. PYR was similarly prepared by adding 2 mg into 10 µL of 100% toluene, allowing it to evaporate before adding 1 mL of 2 mg/mL of glucose-supplemented media. Each prepared solution was filtered through a 0.22 μm Millipore Express PLUS Stericup® Quick Release-GP Vacuum Filtration System (MilliporeSigma, MA, USA).

Co-metabolic carbon source optimization

The growth of A. x. C2 and A. x. KW38 at varying concentrations of glucose and supplementary aromatic carbon sources optimum were monitored at OD₆₀₀ using a NanoDrop™ 2000/2000c spectrophotometer (Thermo Fisher Scientific, Waltham, MA, USA) to determine the optimal glucose-to-aromatic carbon ratio for degradation studies.

The concentration of supplementary glucose was determined by assessing the bacterial growth at four different ratios of glucose to aromatic hydrocarbons. The final concentration of carbon sources (glucose + aromatic hydrocarbons) was maintained constant at 4 mg/mL [21]. The growth was evaluated at varying glucose levels of 0, 1, 2, 3, 4 mg/mL mixed with corresponding aromatic hydrocarbon levels of 4, 3, 2, 1, 0 mg/mL, respectively. The bacterial growth level was determined in each sample after 10 days of incubation using absorbance measurement at OD₆₀₀. The lowest concentration of glucose that initiated the growth was selected as a supplementary carbon source concentration. Subsequently, the maximum concentration of individual aromatic hydrocarbons that the strains can tolerate was tested by growing the strains at increasing concentration of aromatic hydrocarbons (0, 1, 2, 3, 4 mg/mL) in glucose-enriched M9 minimal media [carbon-free M9 minimal media supplemented with 2 mg/mL glucose as the sole additional carbon source].

Aromatic hydrocarbon degradation assay

An aliquot of 10 µL from each bacterial strain was added into 2 mL of glucose-enriched M9 media containing 2 mg/mL of aromatic hydrocarbons. The samples were incubated in a shaker incubator at 130 rpm and 37 ˚C for 10 days in an orbital incubator/shaker (Model Stuart S1600C, Wagtech, UK). The bacterial growth was determined by measuring the turbidity of the culture at OD₆₀₀ for both bacterial strains on day 10 [22,23].

An aromatic hydrocarbon degradation assay was performed on agar plates to corroborate the results obtained in the liquid culture assay. A total of 1 mL of 2 mg/mL solubilized aromatic hydrocarbon was drop-cast on the surface of agar plate and spread using a sterile spreader, then allowed to evaporate. Once the solvent had evaporated, 100 µL of bacterial culture was spread on the agar plate followed by incubation at 37 ˚C for 10 days.

Scanning electron microscope (SEM)

The morphology of bacterial cells and colony structures were examined using SEM [24]. Briefly, 10 µL of bacterial culture was diluted to 0.1 OD and grown on a coverslip inside 12-well plate incubated in 37 ˚C incubator. Then the media was gently removed, and the coverslip was washed twice with phosphate buffered saline (PBS) to eliminate any residual contaminants. Then the bacteria were fixed with 2% glutaraldehyde at room temperature for 1 hr. Following fixation, sequential dehydration was performed using 60% ethanol for 10 min, followed by 70%, 80%, 90%, and 100% ethanol. The samples were air-dried, and gold coated for SEM visualization. Similarly, agar slides were prepared by cutting a 1 x 1 cm² piece of agar, followed by fixation and dehydration.

Mash-based analysis of Acromobacter genomes

At first, Similar Genome Finder from the Bacterial and Viral Bioinformatics Resource Center (BV-BCR), previously known as PATRIC [25] was used to identify genetic distances between the UAE indigenous strains and other members of the Achromobacter genus using the Mash/MinHash metric. Genetic distances were calculated using “mash distance”. The analysis resulted in a set of 100 genomes within a p-value threshold = 1, and Mash distance = 1. The results of the Similar Genome Finder were then loaded into a separate genome group, including A. x. C2 and A. x. KW38 and the 100 selected genomes. A phylogenetic tree analysis was conducted on the list of genomes, obtained from the Similar Genome Finder, using the Codon Tree method [26] and were performed using 500 genes for analysis, and zero allowed deletions and duplications (results of 1000 genes are similar).

Selected study group for comparative analysis

Six closely related Achromobacter species: A. arsenitoxydans SY8, A. piechaudii ATCC 43553, A. sp. DMS1, A. xylosoxidans A8, A. xylosoxidans subsp. xylosoxidans NCTC 10807, and A. denitrificans NBRC 15125 genomes along with A. x. C2 and A. x. KW38 were selected for comparative genomic analysis. Complete genomes of selected Achromobacter species were retrieved into the BV-BRC database.

Identifying enzymes involved in degradation of aromatic hydrocarbons

Annotation of putative aromatic carbon degradation pathways was performed for all 6 selected species bacteria using BV-BRC. Next, pathway information about adjacent enzymes was extracted from and pathway neighbors were defined as enzymes that are connected to each other through a common substrate. Using BLAST and the Kyoto Encyclopedia of Genes and Genomes (KEGG) database, we mapped coding sequences (CDS) into similar enzymes (<99% identity) in reference genomes and extracted EC numbers from reference genomes. Metabolic pathways absent in both A. x. C2 and A. x. KW38 were excluded from the analysis.

The putative enzyme functions analysis was performed manually by comparing selected enzyme domains to known protein structures in reference genomes using Pymol (pymol.org) [27] and BV-BRC. Multiple sequence alignment (MSA) was performed for each of the degradation enzymes separately to compare and manually curate the data before threading their structures.

We performed a global protein alignment using BV-BRC MSA tool. In some instances, there were multiple copies of the enzyme identified in the genomes. Protein sequences below 100 amino acids, sequences longer than the average length of occurrences, similar sequences (containing sequence members of different families), and alignments with inserts greater than 10 amino acids were excluded from the analysis. Additional MSA was performed using T-Coffee (tcoffee.crg.eu) [28] and MUSCLE alignment tools (ebi.ac.uk) [29]. Moreover, genome search via BLAST (blast.ncbi.nlm.nih.gov) [30] was performed using UniProt (uniprot.org) [31] and NCBI on the following species: Achromobacter, Acidovorax, Verminephrobacter, Comamanas, Variovorax, Ramlibacter, Hydrogenophaga, Rhizobacter, Pelomonas, Rubrivivax, Azohydromonas, Sphaerotilus, Thiomonas, Bordetella, Castellaniella, Bordetella, Pusillimonas, Candidatus, Burkholderia, Pandoraea, Capriavidus, Ralstonia, Cupriavidus, Massilia, Janthinobacterium, Massilia, Herbaspirillum, Collimonas, Thauera, Azoarcus, Dechlorosoma, Pseudoglubenkiania, Xanthomonas, Stenotrophomonas, Lysobacter, Arenimonas, Pseudomonas, and Sphingomonas-like bacterium. The results were aligned to the A. x. C2 and A. x. KW38 genomes to assist in refining gene structures.

Protein structure prediction by threading methods

All selected enzymes were passed through the Search Tool for the Retrieval of Interacting Genes (STRING) (string-db.org) [32] to obtain their protein-protein interactions (PPIs) network connection. STRING integrates known and predicted interactions from multiple sources. However, such computational predictions may include false positives or lack context-specific information, highlighting the inherent challenges in accurately interpreting PPI networks. Therefore, screening of the PPI network was performed to select the highest connected module with the core proteins having greater than 98% identity to proteins in A.xylosoxidans NBRC15126.

Secondary structure prediction was performed by threading the amino acid sequences to the library of known folds using Jpred4 [33], Phyre2 [34], and PSIPRED (bio.tools/psipred) [35,36]. The program PyMOL was used for visualization and comparison of protein structure alignments [37]. All enzymes were uploaded to the Search Tool for the Retrieval of Interacting Genes (STRING) [32] to construct the PPI network with no clustering option was chosen. Screening of the constructed modules was performed to select the highest connected module. Only proteins that had > 98% identity to proteins in A.xylosoxidans NBRC 15126 were used in the analysis.

Data analysis

All data obtained in the study were subjected to a t-test, p < 0.05 was considered statistically significant. The absorbance readings in the optimization of carbon source and the biodegradation assay represent the bacterial growth as the mean ± SD of three replicas. This article does not contain any studies with human participants or animals performed by any of the authors.

Optimizing growth conditions of A. x. C2 and A. x. KW38 in presence of aromatic hydrocarbons

Prior to the biodegradation assays, growth optimization experiments were conducted to determine the optimal growth conditions for A. x. C2 and A. x. KW38 in the presence of aromatic hydrocarbons. We observed that the bacterial strains used for aromatic hydrocarbon degradation did not exhibit significant growth when cultured directly on the aromatic compounds, Fig 2a, 2b). Supplementing the culture with the minimum amount of co-carbon source, like glucose, stimulated initial microbial growth and further induce the expression of degradation genes required to utilize the aromatic compounds [46–48]. In various bacteria, the consumption of aromatic compounds like benzene, toluene, and phenanthrene is supported by the initial availability of glucose, which subsequently activate key enzymes including catechol 1,2-dioxygenase, benzene/toluene/chlorobenzene dioxygenase, and phenanthrene monooxygenases, S3 Table in S1 File. The shift from glucose to aromatic compounds generally requires the activation of certain degradation pathways, facilitating the effective microbial adaptation and breakdown of aromatic compounds in environments with various carbon sources [49,50]. Therefore, the strains were cultivated in carbon-free M9 minimal media supplemented with various concentrations of glucose as a supplementary carbon source to determine the optimal growth conditions for subsequent degradation of BPA, 4-HBA, 1-NM, and PYR, S1 Fig in S1 File. The minimum threshold required to support microbial activity was 2 mg glucose-C/g soil. At lower glucose concentrations, the supplemental glucose was insufficient to initiate the bacterial growth. Our findings were consistent with the results reported in the study of Stephanie Reischke et al., 2015 [51], where they found that only concentrations of 2 mg glucose-C/g soil and higher met the energy demand for maintaining active biomass and resulted in an exponential increase in bacterial respiration.

Download:

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

Fig 2. Evaluating biodegradation of aromatic hydrocarbons by A.

x. C2, A. x. KW38, P. paraeruginosa, and E. coli under optimized growth conditions. Incubation conditions: 37 ˚C, pH 7.0, shaking at 130 rpm both in aqueous and solid-based degradation assays. a. Absorbance measurements quantifying bacterial growth after 10 days in M9 minimal media with 2 mg/mL glucose and 2 mg/mL of the indicated aromatic hydrocarbon. Chemical structures are shown. Error bars represent standard deviation (n = 9). b. Visual turbidity assay comparing growth of the four strains in BPA, glucose, and BPA + glucose conditions supported with SEM images. c. Schematic illustrating the solid agar assay setup with glucose and aromatic hydrocarbons coated onto the surface before adding bacterial cultures, supported with SEM images showing growth in form of fractal structures surrounding insoluble PYR particles on the agar surface. d. glucose-enriched agar plates, streaked with the four strains in the presence of different aromatic hydrocarbons and incubated for 10 days at 37 ˚C. Triplicates validate degradation potential, particularly for PYR. colors describing the scoring system for growth (in green), no growth (in blue) on the solid agar assay plates Figure created using BioRender.com [45].

https://doi.org/10.1371/journal.pone.0329515.g002

The effects of glucose concentration on growth were determined by testing glucose: aromatic hydrocarbon ratios of (4:0, 3:1, 2:2, 1:3, and 0:4) with a final concentration of both carbon sources at 2 mg/mL. For all four aromatics, the 1:1 ratio (2 mg/mL glucose) supported optimal growth after 10 days, as indicated by OD₆₀₀, see S1a–S1d Fig in S1 File. Lower glucose levels did not permit robust growth, likely due to aromatic inhibition, while higher levels elevated turbidity independently of aromatic catabolism. P. paraeruginosa (positive control) and E. coli (negative control) followed similar trends, though E. coli grew less, potentially from aromatic toxicity [52].

Keeping glucose concentration at 2 mg/mL, the individual effect of aromatic concentration was tested in the range of (0–4 mg/mL). For BPA, 4-HBA, 1-NM, and PYR, growth decreased progressively with increasing aromatic levels above 2 mg/mL, with complete inhibition at 4 mg/mL, S1e–S1h Fig in S1 File. Therefore, 2 mg/mL was selected as the optimized aromatic concentration, supporting moderate growth (average OD₆₀₀ ≈ 0.2) suitable for biodegradation studies. At 2 mg/mL of aromatics, the growth level of A. x. C2 and A. x. KW38 closely matched the level of P. paraeruginosa while E. coli grew slightly less, likely lacking efficient aromatic catabolism. In summary, supplementing the media with equal amounts of glucose and aromatic hydrocarbons (2 mg/mL each) resulted in a robust growth of A. x. C2 and A. x. KW38 comparable to P. paraeruginosa.

Exploring aromatic hydrocarbon degradation in the UAE-isolated bacterial strains, A. x. C2 and A. x. KW38

The bacterial strains A. x. C2 and A. x. KW38 were previously isolated from surface rhizosphere of desert plants and oil-contaminated site, respectively, as shown in Fig 1a. Given their isolation from ecologically degraded and contaminated environments, we looked at their potential to degrade aromatic hydrocarbons using both solid and aqueous-based culture assays. Previous work identified A. x. C2 and A. x. KW38 as gram-negative aerobic strains [18]. The SEM images Fig 1c revealed that A. x. C2 and A. x. KW38 are rod shaped, non-flagellated bacteria with a wrinkled, rough, and glossy surface. The cell dimensions ranged from 1.1–1.9 µm in length and 0.5–0.8 µm in width.

In the solid-based assay, agar plates, supplemented with 2 mg/mL glucose, were surface coated with solubilized aromatic hydrocarbons. The solvent was allowed to evaporate before introducing the fresh bacterial cultures. In the liquid assay, soluble aromatic hydrocarbons served as the sole carbon source in M9 minimal media, with 2 mg/mL of glucose provided as a supplementary carbon source. The growth level of both A. x. C2 and A. x. KW38 was promoted in these two assays compared with the growth level in glucose alone, indicating the ability of the strains to metabolize the respective aromatic compounds and thrive more in their presence, Fig 1b and 1e). P. paraeruginosa, a well-characterized aromatic hydrocarbon degrader, was used as a positive control [53–58], while E. coli, which lacks natural aromatic catabolic capabilities, served as the negative control [59].

Both strains exhibited growth levels comparable to P. paraeruginosa when cultured in M9 media abundant with aromatic hydrocarbons, as well as on agar coated with aromatic hydrocarbons. The aqueous-based assay, Fig 2a and 2b), demonstrated high growth level in the presence of aromatics compared to glucose alone for degrading strains A. x. C2, A. x. KW38, and P. paraeruginosa (supplementary figures S2–S4 Figs in S1 File for other aromatics). Likewise, the growth on agar showed results as evidenced by the similar colony size and number, Fig 2d, (supplementary figures S5–S8 Figs in S1 File for other aromatics). This suggests the potential ability of these strains to degrade the supplied aromatic substances in a performance resembling the degradative potential of Pseudomonas. In contrast, E. coli showed negligible growth in both solid and liquid media with aromatic hydrocarbons, as expected. However, it showed low growth level in BPA and PYR, these results support the findings reported by Bruce S. Hass et al., where BP was found to promote the growth of E. coli, and PYR has little to no effect on its growth [59].

These results show the capacity of the indigenous UAE strains A. x. C2 and A. x. KW38 to degrade a range of aromatic hydrocarbons through putative dedicated catabolic pathways. Their growth in the presence of these compounds highlights their metabolic capabilities and their potential use in bioremediation of organic aromatic pollutants.

Growth on aqueous and solid media revealed distinct degradation profiles for the four bacterial strains when challenged with different aromatic hydrocarbons in the presence of glucose as a co-metabolic substrate. For BPA, the strain A. x. KW38 and the model degrader, P. paraeruginosa displayed the highest growth capabilities based on culture turbidity, agar plate colonies, and biomass accumulation, S5 Fig in S1 File. This suggests that these strains possess efficient catabolic pathways and tolerance mechanisms to mineralize BPA as a primary carbon and energy source when supplemented with glucose. In contrast, E. coli exhibited moderate growth inhibition by BPA, only achieving partial degradation likely through general metabolic pathways instead of dedicated ones. Some engineered E. coli strains can degrade BPA via introduced bisdA and bisdB genes, but the wild-type strain used here lacked such specialized functions [60].

A divergent pattern emerged with 1-NM, where A. x. C2, A. x. KW38, and P. paraeruginosa demonstrated comparable and robust growth, while E. coli proliferation was completely abolished, S2 and S6 Figs in S1 File. This stark inhibition aligns with previous studies reporting E. coli’s extreme sensitivity to even trace 1-NM levels (0.05 mg/mL). The UAE isolates thus appear to possess distinct enzymatic machinery conferring tolerance and metabolic processing of this naphthol compound that E. coli naturally lacks.

For 4-HBA, S3 and S7 Figs in S1 File, all three degrader strains (A. x. C2, A. x. KW38, P. paraeruginosa) displayed significantly reduced growth compared to other aromatics. This decrease in substrate degradation over the time course of the experiment may relate to the structural features of 4-HBA, which can induce multidrug efflux pumps and unidentified resistance mechanisms in E. coli that completely inhibited its proliferation. Other aromatic acids like benzoic acid are known to act as potent antimicrobials against E. coli and other gram-negative pathogens [52].

The high-ring aromatic hydrocarbon PYR proved challenging for most strains except A. x. C2 and A. x. KW38, which effectively tolerated and grew on 2 mg/mL in both liquid and solid assays (S4 and S8 Figs in S1 File). In contrast, E. coli was unable to proliferate at this elevated PYR concentration. While some reports describe engineered PYR-degrading E. coli, the wild-type strain used here lacked intrinsic tolerance/catabolic capabilities for this heavily condensed polycyclic aromatic structure. P. paraeruginosa also showed reduced growth on PYR compared to the other aromatics tested. The robust degradation by the UAE isolates highlights their potential for remediating PYR and other high molecular weight PAH pollutants.

Interestingly, all three aromatic degraders (A. x. C2, A. x. KW38, P. paraeruginosa) formed distinctive fractal biofilm patterns surrounding insoluble aromatic clusters on the solid agar plates, Fig 2d. This unique morphology reflects the adaptive biological phenomena which is visual evidence of the ability of the aromatic degraders to utilize catabolic pathways and diffusion gradients to access, funnel, and mineralize solid-phase aromatic hydrocarbon pollutants as carbon/energy sources once supplemental glucose is depleted. Fractal biofilms are usually observed under conditions of diffusion limitation or nutrient gradients because they help maximize surface area and increase efficiency of expansion and resource capture. According to T. Vicsek et al. [61], when bacterial growth is dominated by slowly spreading nutrients they grow in diffusion-limited aggregation forming fractal structures. Furthermore, the presence of such complex morphologies provides evidence that the degraders are undergoing advanced catabolic mechanisms to utilize limited carbon sources. In other words, as glucose gets depleted, the branching growth spreads more to increase the metabolic reach and access to the insoluble clusters of aromatic hydrocarbons. Such phenomena has also been explored in other studies in which bacterial biofilm formed fractal patterns as a response to nutrient limitation [62–64].

Various mechanisms may contribute to bacterial survival when utilizing PAHs as a carbon source. Certain bacteria operate within microbial communities, producing exopolysaccharides that house enzymes capable of degrading PAHs. Specifically, in the case of A. xylosoxidans, research indicates that it releases biosurfactants, which enhance the bacteria’s ability to adhere to and break down PAHs effectively [65]. Additionally, other studies suggest that bacteria secrete specialized polysaccharides that facilitate efficient diffusive-based transport of PAHs from crystalline forms to bacterial cells [66]. Moreover, among the bacteria examined, biofilm formation on PAH sources emerged as a key strategy to address mass transfer challenges associated with poorly soluble PAHs. The synthesis of surfactants can lower both surface tension and interfacial tension, a process known as surfactant-enhanced remediation (SER) [67]. Based on this, we can propose that the combination of biosurfactant production and biofilm formation may be the primary mechanisms driving bacterial attachment to and utilization of PAHs in our study. Nevertheless, further multiomics investigations are needed to elucidate the precise molecular responses underlying these processes.

Although the UAE indigenous isolates A. x. C2 and A. x. KW38 demonstrated versatile degradation capacities across all aromatics tested, there were notable differences in their efficiencies relative to P. paraeruginosa and E. coli which correlated with compound structures and strains’ intrinsic metabolic networks. These results highlight their promise as bioremediators, but also reveal substrate-specific variations that warrant further investigation of underlying enzymatic and regulatory mechanisms governing their aromatic catabolic metabolism.

Interactomics investigation: Mapping molecular interactions through genome dynamics and comparative analysis

To validate the results of the degradation assays, subsequent bioinformatic analysis was utilized to determine the available metabolic pathways and specific enzymes that can be potentially active during the degradation of aromatic hydrocarbons. Initially, these strains went through high throughput sequencing and characterization processes [18]. Genome sequencing and annotation revealed that A. x. C2 and A. x. KW38 as closely related Achromobacter strains with 6.55 Mbp and 6.58 Mbp genomes, ~ 68% GC content, and 10 rRNA operons each [18]. To investigate their aromatic hydrocarbon degradation capabilities, a multi-omics bioinformatic analysis pipeline was employed, S9 Fig in S1 File. We developed a comprehensive methodological framework that strategically locates the pathways and enzymes involved in the degradation of aromatic hydrocarbons. The framework is based on integrating several key steps to systematically search for specific metabolic pathways and proteins, analyzing their sequences and structures to optimally improve future experimental work and design.

Comparative analysis via BLAST revealed significant sequence and functional similarity between the two strains. The current investigation focuses on a systematic methodology designed to assess the biodegradation capabilities of these bacterial strains involving aromatic hydrocarbon structures and identifying the metabolic pathways and specific enzymes associated with aromatic hydrocarbon degradation within their sequences.

Initially, a BLAST-based homology search against 500 public bacterial genomes identified six Achromobacter strains as closest relatives to A. x. C2 and A. x. KW38 based on whole genome distance estimation using Mash/MinHash approaches. Phylogenomic analysis positioned the UAE isolates in a distinct clade alongside these six reference strains (A. xylosoxidans A8, A. denitrificans NBRC 15125, A. arsenitoxydans SY8, A. piechaudii ATCC 43553, A. xylosoxidans NCTC 10807, A. sp. DMS 1), which were included for intraspecies comparative genomics, Fig 3a, details available in S1 Table in S1 File.

Download:

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

Fig 3. Comparative bioinformatic analysis, including phylogenetic, enzymatic, and PPI network analysis of BPA-degrading UAE-indigenous bacterial strains.

a. Phylogenetic tree analysis of UAE-indigenous strains (in blue and green) and other bacterial strains globally available from other regions (full phylogenetic tree is available in S10 Fig in S1 File) and the corresponding enzyme heatmap representing the number of occurrences of each enzyme involved in BPA breakdown pathway in all selected strains. b. Schematic representing the degradation of BPA using Achromobacter strain. c. The corresponding BPA degradation pathway active during the breakdown of structure highlighting the EC numbers of enzymes involved in each step. d. The number of enzymes present in each aromatic hydrocarbon degradation pathways in all the genomes analyzed in the phylogenetic tree (Fig 3a). e. PPI network complex and modular analysis of 25 curated enzymes that degrade aromatic hydrocarbons in A. x. KW38, PPI: number of nodes = 25, number of edges = 21, average node degree = 1.68, average local clustering coefficient = 0.353. f. PPI network complex for A. x. C2, PPI; number of nodes = 24, number of edges = 21, average node degree = 1.75, average local clustering coefficient = 0.368. *Here in Fig 3e and 3f), network nodes represent proteins, splice isoforms or post-translational modifications are collapsed, i.e., each node represents all the proteins produced by a single, protein-coding gene locus. Figure created using BioRender.com [38].

https://doi.org/10.1371/journal.pone.0329515.g003

Global pathway mapping revealed 6 out of 140 total pathways (details are available in S2 Table in S1 File) directly involved in aromatic hydrocarbon/PAH degradation: naphthalene/anthracene (NAP/ANT) (S11 Fig in S1 File), PAH (S12 Fig in S1 File), dioxin (S13 Fig in S1 File), BP (Fig 3c and S14 Fig in S1 File), benzoate hydroxylation (BNZ) (S15 Fig in S1 File), and dichlorodiphenyltrichloroethane (DDT) pathway (S16 Fig in S1 File). An additional 16 lipid metabolism pathways contributing to biosurfactant synthesis were found to indirectly support aromatic catabolism. Within these six core degradation routes, 37 unique enzyme entries were identified (details of each are available in S3 Table in S1 File), 25 of which (66%) were conserved across the studied Achromobacter genomes. Interestingly, one dehydrogenase (EC 1.14.13.1) was exclusively present in A. x. C2 and A. x. KW38. In contrast, 11 enzymes were unique to other reference strains. The benzoate hydroxylation pathway contained 80% of the total aromatic degradation enzymes mapped (details available in S18 and S22 Figs in S1 File). Despite overall conservation, manual curation of enzyme sequences unveiled strain-specific variations. Pairwise alignment analyses identified A. xylosoxidans NCTC 10807 as most closely related to the UAE isolates based on enzyme repertoires, while A. sp. DMS1 was the most divergent. Most enzymes yielded high-quality alignments (scores 995–1000) after multi-step sequence curation (details of each are available in S4 Table). Subsequently, the protein structures were threaded into 3D structures. The structures of the 25 enzymes are available in the supplementary document S23–S48 Figs in S1 File).

Structural modeling enabled construction of PPI networks for 25 curated aromatic degradation enzymes in A. x. KW38 and 24 in A. x. C2, Fig 3e and 3f). These interconnected networks were significantly enriched for biological processes related to aromatic catabolism like the beta-ketoadipate pathway. Certain enzymes served as network hubs potentially interacting with multiple partners, S18–S22 Figs in S1 File. The PPI network topology, Fig 3e and 3f), together with the genomic evidence of conserved aromatic catabolic pathways shared across Achromobacter species, substantiates the phenotypic observations of A. x. C2 and A. x. KW38 efficiently mineralizing diverse aromatic pollutants. Unique genetic elements like the exclusive dehydrogenase further point to specialized features underpinning these strains’ biodegradative versatility. In summary, this integrated multi-omics approach leveraging comparative genomics, pathway reconstruction, enzyme homology, and interactome mapping has provided a comprehensive systems-level perspective, correlating the indigenous isolates’ remarkable aromatic hydrocarbon degradation capabilities with their genomic and metabolic machinery. The insights gained highlighted their promising potential for future biotechnological applications in bioremediation and waste treatment.