Introduction

Benzene is a known carcinogenic organic compound, classified as a priority pollutant by the US Environmental Protection Agency. Its occurrence in gasoline contaminated subsurface environments is general, and due to its relatively high solubility in water (1.79 g/L), considerably threatens subsurface water reservoirs. Thus, the removal of benzene from polluted sites is crucial. Since a plethora of microorganisms can use benzene as a sole source of carbon and energy, bioremediation is one of the most frequently applied processes of benzene removal. Still, bioremediation of benzene contaminated subsurface ecosystems is challenging due to the fact that the vast majority of the degraders use aerobic metabolic pathways for the activation and cleavage of the benzene ring. Under aerobic conditions these key steps are catalysed by mono- and dioxygenases, which require molecular oxygen as a co-substrate. However, availability of oxygen in subsurface environments, even in pristine ones is usually limited. In highly contaminated environments, the aerobic degraders relatively quickly deplete oxygen, giving chance to anaerobic degraders (nitrate-, iron-, or sulphate-reducers, methanogens) to cope with the contamination Meckenstock et al. (2016). Nevertheless, benzene is very slowly degraded without molecular oxygen and this fact makes air sparging technique a preferred and successful bioremediation process in case of large-scale subsurface BTEX-contaminations Kabelitz et al. (2009). As cost effectiveness is becoming an increasingly important factor in the case of bioremediation processes as well, it is a valid aim to look for bacteria, which can aerobically degrade BTEX-compounds even in the presence of tiny amount of oxygen. It has been known for more than two decades, that aerobic benzene degradation is feasible even under microaerophilic conditions (~ 0.2 mg/L dissolved oxygen) and that phenol is the major intermediate metabolite during the degradation of benzene at low concentrations of dissolved oxygen Yerushalmi et al. (2002). Recently, it has been shown that toluene and xylene can be degraded under microaerobic conditions and mainly a well-defined group of Betaproteobacteria in the families Comamonadaceae, Zoogloeaceae and Rhodocyclaceae were reported as the key toluene- or xylene-degraders under such conditions (Táncsics et al. 2018, 2023). Moreover, it has been shown that meta ring-cleaving catechol 2,3-dioxygenases belonging to I.2.C subfamily of extradiol dioxygenases (EDOs) are functional under hypoxic conditions and play prominent role in the degradation of methyl-substituted aromatics under microaerobic conditions (Kukor and Olsen 1996). Nevertheless, benzene can be degraded through the ortho-cleavage pathway as well, and an ortho ring-cleaving catechol 1,2-dioygenase, adapted to low oxygen availability has already been reported earlier Balcke et al. (2007). Still, it is clear that our knowledge on the diversity of bacteria and ring-cleavage enzymes playing a key role in the degradation of benzene under microaerobic conditions is still in its infancy (Bedics et al. 2022). Therefore, the aims of the present study were to reveal differences between microbial communities degrading benzene under aerobic versus microaerobic conditions and to clearly identify microaerobic benzene-degraders by a microcosm experiment combined with DNA-stable isotope probing.

Materials and methods

Sampling site and sample acquisition

Groundwater sediments were taken from the deeply studied Siklós BTEX-contaminated site of Hungary (Táncsics et al. 2012, 2013, 2023). Previously, sediment sample originated from the same groundwater well (designated as ST2) was used successfully in a SIP study, which aimed to reveal diversity of bacteria capable of degrading toluene under microaerobic conditions (Táncsics et al. 2018). Sediment rich groundwater was pumped out from the bottom of the monitoring well and collected into sterile, 1 L brown coloured glass bottles. Samples were stored at 4 °C and transported to the laboratory immediately.

Incubation of sediments

Triplicates of 5 g (wet weight) homogenously mixed sediment material were transferred into sterile 100-mL serum bottles containing 50 mL of mineral salts (MS) medium. The MS medium was prepared with 100 ml each of solutions A and B, 1 ml of solution C and 800 ml of water. Solution A: 5 g MgSO4·7H2O and 1 g CaCl2·H2O l−1 (filtered). Solution B: 11.1 g Na2HPO4, 2.5 g KH2PO4 and 6.7 g NH4Cl l−1 (autoclaved). Solution C: 10 g FeSO4·7H2O, 0.64 g Na2EDTA·3H2O, 0.1 g ZnCl2, 0.015 g H3BO3, 0.175 g CoCl2·6H2O, 0.15 g Na2MoO4·2H2O, 0.02 g MnCl2·4 H2O and 0.01 g NiCl2·6 H2O l−1 (filtered). Additionally, the medium was supplemented with 1 mg/L thiamine, 15 μg/L biotin, and 20 μg/L vitamin B12. Microaerobic microcosms were sparged aseptically with N2/CO2 (80:20, v/v) for 10 min, after which the desired volume of sterile (0.2-µm-pore-size-filtered) air was injected into the bottles through butyl rubber stoppers. Dissolved oxygen concentration in the microaerobic microcosms was set to 0.5 mg/L and kept between 0.5 and 0 mg/L throughout the experiment. In the aerobic microcosms dissolved oxygen concentration was kept between 6–8 mg/L. Oxygen was replenished once every 24 h. A 5 µL of fully labeled (13C6) benzene (Sigma-Aldrich) was injected into the microcosms. Abiotic control bottles (autoclaved three times) amended with unlabelled toluene were also prepared to exclude abiotic benzene loss. The bottles were incubated at 16 °C in a rotary shaker at 145 rpm.

Process measurements

Dissolved oxygen concentration in the liquid phase of the microcosms was measured by using planar oxygen sensor spots and a Fibox 3 Oxygen Meter (PreSens, Germany). At each sampling spot, dissolved oxygen concentrations were registered every second for 1 min, and the results were displayed by using the OxyView-PST3 software (V7.01, PreSens). Benzene concentrations were determined by headspace analysis on an ISQ Single Quadrupole GC–MS (ThermoFischer Scientific) via a SLB-5 ms fused silica capillary column (Supelco Analytical). The oven temperature was set to 40 °C for 3 min, then ramped at a rate of 20 °C/min to 190 °C, and held for 1 min. The mass spectrometer (MS) was operated at 250 °C in full scan mode.

Nucleic acid extraction and ultracentrifugation

Sediments were collected from sacrificed microcosms after 4 d (aerobic microcosms) or 7 d (microaerobic microcosms) of incubation by centrifugation at 2360 g at 4 °C for 10 min using a Rotanta 460 R (Hettich). DNA was isolated from the sediment pellets by using the NucleoSpin Soil Kit (Macherey–Nagel) according to the instructions of the manufacturer. DNA samples were stored frozen at −80 °C until downstream analyses. For isopycnic density gradient ultracentrifugation, equal volumes (30 µL) of eluted DNA from the triplicate aerobic and microaerobic microcosms were pooled, respectively. This was necessary to get enough amount of DNA for the ultracentrifugation. Subsequently, approximately 1 µg of NanodropOne quantified (Thermo Fisher Scientific) DNA extract was loaded onto 18.5 mL gradient medium of CsCl (average density 1.72 g/mL, Fisher Chemical) in gradient buffer (0.1 M Tris–HCl at pH 8, 0.1 M KCl, 1 mM EDTA) and centrifuged (39,000 rpm, ~ 122,000 RCFav, 72 h) by using a Sorvall Combi ultracentrifuge (DuPont). Centrifugation was performed by using 18.5 mL centrifuge tubes (DuPont) and vertical rotor (Sorvall TV-865B). Following centrifugation, eighteen gradient fractions (1 mL in volume) were collected into sterile 1.5 mL Eppendorf tubes from “heavy” to “light” fractions. Bouyant density (BD) of the fractions was calculated based on the mass of 0.5 mL of each of the gradient fractions. Subsequently, recovery of the DNA from the gradient fractions was performed by using 30% (m/V) polyethylene glycol 6000 (PEG 6000) (Fluka AG) as described by Griffiths et al. (2000).

16S rRNA gene targeted qPCR, Illumina amplicon sequencing and genome-resolved metagenomics

Twelve DNA fractions (from 3rd to 14th) of each gradient were selected for bacterial 16S rRNA gene-targeted qPCR, which was performed as described earlier (Kunapuli et al. 2007; Pilloni et al. 2011).

Illumina 16S rRNA gene amplicon sequencing was carried out both in case of non-density resolved, unique DNA samples of triplicate aerobic and microaerobic microcosms, and selected “light”, “medium” and “heavy” DNA fractions resulted after the isopycnic density gradient centrifugation. The variable V3 and V4 regions of the 16S rRNA gene were amplified by using primers recommended by Klindworth et al. (2013). PCR amplification was carried out using the KAPA HiFi HotStart Ready Mix (KAPA Biosystems) according to the 16S metagenomics sequencing library preparation guide of Illumina. Paired-end fragment reads were generated on an Illumina MiSeq sequencer using a MiSeq Reagent Kit v2 (500 cycles) by Eurofins BIOMI Kft. (Gödöllő, Hungary). Primary data analysis (base-calling) was conducted with Bbcl2fastq^ software (v2.17.1.14, Illumina). Finally, the sequences were processed using Mothur v1.41.1 software Schloss et al. (2009) following the recommendations of the MiSeq SOP manual (http://www.mothur.org/wiki/MiSeq_SOP) Kozich et al. (2013). Based on the sequence alignment as determined by the SILVA 132 SSURef NR99 database Quast et al. (2013), sequences were then assigned. The detection of chimeras was performed with Mothur’s uchime command and the ‘split.abund’ command was also used to remove singleton reads (Edgar et al. 2011; Kunin et al. 2010). A 97% similarity threshold level was used to assign sequences into Operational Taxonomic Units (OTUs) as suggested by Tindall et al. (2010) for prokaryotic species delineation. Amplicon sequence data were deposited at NCBI under the BioProject number PRJNA704528 (BioSamples: SAMN39829928 and SAMN39829929).

The process of metagenome sequencing and reconstruction of bacterial genomes (genome binning) has been described earlier Táncsics et al. (2023). Briefly, paired-end fragment reads (2 × 250 nucleotides) were sequenced using the MiSeq Reagent Kit v2 (500-cycles) on an Illumina MiSeq sequencer by Seqomics Kft. (Mórahalom, Hungary). Quality-controlled (BBDuk and sickle v1.33) reads were assembled using the MetaSPAdes pipeline, version 3.15.0. Nurk et al. (2017). Subsequently, Abawaca Brown et al. (2015) was run twice, first to include reads with a minimal length of 3000 base pairs (bps) and sequences split after 5000 bps and second to set the minimal scaffold length of 5000 bps and sequences split after 10,000 bps. MaxBin2 Wu et al. (2016) was also run on both marker sets. All bins created in the four runs were optimized and filtered using the DAS Tool Sieber et al. (2018) and hand curated with uBin Bornemann et al. (2023). Phylogenetic affiliation of the bins was ascertained using the pipeline of the MiGA–Microbial Genomes Atlas Rodriguez-R et al. (2018). The Azovibrio-related bin genome analyzed in the present study can be found at NCBI under the BioProject number PRJNA818156.

Phylogenetic analyses

Reference genomes for comparison purposes were retrieved from the GenBank database (https://www.ncbi.nlm.nih.gov/genbank). The phylogenomic tree shown in the study was constructed using a concatenated alignment of 92 core genes with UBCGs software Na et al. (2018). The catechol 2,3-dioxygenase-based phylogenetic tree was created with MEGA version 7 Kumar et al. (2016) by using the neighbor-joining algorithm Saitou and Nei (1987). Orthologous average nucleotide identity (OrthoANI) value reported in the study was calculated by using the OAT software Lee et al. (2016), while average amino-acid identity value was determined by the MiGA pipeline Rodriguez-R et al. (2018).

Results and discussion

Benzene degradation in the aerobic vs microaerobic microcosms

Both aerobic and microaerobic degradation of benzene proved to be a rather rapid process. Under clear aerobic conditions, benzene was depleted after 4 days of incubation (96 h), while in the microaerobic microcosms the degradation was a bit slower process and additional 3 days were needed (a total of 7 days) for the complete degradation of benzene (Fig. S1). This is at least because anoxic periods occurred in the microaerobic microcosms, which slowed down the degradation process. A similar observation was made in our earlier SIP study, in which a seven days-long period was needed for the complete microaerobic degradation of toluene (1 mM) Táncsics et al. (2018). In earlier degradation experiments, which we performed with pure strains and where microaerobic conditions prevailed throughout the whole experiment (due to a much smaller amount of benzene and bacterial biomass) no significant difference in terms of degradation time was observed between aerobic and microaerobic degradation of benzene (Banerjee et al. 2022; Bedics et al. 2024).

Bacterial community composition of aerobic and microaerobic benzene-degrading microcosms

At first, microbial communities were investigated by using the non-density resolved DNA samples obtained from each of the enrichments. In the aerobic enrichments members of the genus Pseudomonas were overwhelmingly dominant (with relative abundance values between 40–51%), Rhizobium (10–15%), Thauera (12–15%) (Fig. 1). The major Pseudomonas OTUs were related to P. aromaticivorans (35–45%) and members of the P. stutzeri complex (4–5%) (Table 1). Both lineages contain prominent aromatic hydrocarbon degrading microorganisms and P. aromaticivorans was described by us previously as a benzene-degrading bacterium (Banerjee et al. 2022; Li et al. 2022). The genus Thauera was represented by T. aminoaromatica, which was described as an aromatic compound-degrading bacterium (Mechichi et al. 2002). The genus Rhizobium was mainly represented by an OTU most closely related to a lineage containing R. alamii and R. mesosinicum. The relatively high abundance of Rhizobium in a benzene-degrading enrichment was unexpected, however, recently Rhizobium cremeum was described as an aromatic compound-degrading bacterium, capable of degrading benzene as well (Yang et al. 2022). Notable minor community members detectable in all of the aerobic enrichments were Zoogloea (2–3%), Rugosibacter (~ 2%) and Malikia (~ 1.5%). In the case of the microaerobic enrichments, members of genera Malikia (26–40%) and Azovibrio (20–28%) dominated the communities, which later lineage was completely missing from the aerobic enrichments. The genus Malikia was represented by an OTU most closely related to M. spinosa. Although the type strain of this bacterium was isolated from a pristine freshwater, it has been shown by us earlier that this bacterium can acquire the ability to degrade aromatic hydrocarbons and can play major role in aerobic benzene degradation Révész et al. (2020). The Azovibrio genus was represented by an OTU most closely related A. restrictus, but only at 95.7% 16S rRNA gene sequence similarity. The same Azovibrio-like bacterium was enriched by us earlier in microaerobic xylene-degrading enrichment cultures Táncsics et al. (2023). Nevertheless, the role of Azovibrio in the degradation of BTEX has never been reported earlier. Moreover, the ecological role of this bacterium often remains hidden in relevant environmental microbiological studies Abbas et al. (2019). It was also observed that the relative abundance of genus Pseudomonas considerably decreased and became a minor player of the microaerobic communities. This decrease was due to the fact that the abundance of the P. aromaticivorans OTU was only 0.5–1% in the microaerobic enrichment communities, while the abundance of the other Pseudomonas OTU remained between 4–6%. Another minor community players of the microaerobic enrichments were Geobacter (4–6%), Extensimonas (2–3%) and Rhizobium with slightly decreased abundance (3–8%). Overall, non-density resolved DNA samples indicated that highly different benzene-degrading microbial communities evolved during the experiment period.

Fig. 1
figure 1

Genus level bacterial community structure of aerobic (A) and microaerobic (M) 13C6 benzene-degrading microcosms as revealed by Illumina paired-end 16S rRNA gene amplicon sequencing, using the unresolved DNA samples. Only taxa contributing more than 1% abundance are depicted

Full size image
Table 1 The TOP20 operational taxonomic units (OTUs) detected in the microcosms (ND, not detected)
Full size table

Analysis of density resolved DNA samples

In order to clearly identify the bacterial lineages which were degrading the isotopically labelled benzene (13C6-benzene), community DNA samples were density resolved by isopycnic gradient ultracentrifugation. As a result, “heavy” and “light” DNA fractions were differentiated (Fig. S2) and analysed further by Illumina 16S rRNA gene amplicon sequencing. In case of the aerobic enrichments, genera Pseudomonas and Rhizobium were clearly identified as benzene-degraders, due to the notable increase in the abundance of their 16S rDNA fragments in the heavy DNA fraction compared to that of the light fraction (Fig. 2). In case of the P. aromaticivorans OTU, its 36.8% abundance observed in the light fraction increased to 43% in the heavy DNA fraction. However, in case of the P. stutzeri OTU, an opposite trend was observable, as its abundance was higher in the light fraction (4.0% vs 2.7%). Accordingly, P. aromaticivorans was identified as a prominent benzene degrader under clear aerobic conditions. On the other hand, this OTU was hardly detectable in the heavy DNA fraction of the microaerobic enrichments, reaching only 0.15% abundance. Nevertheless, this bacterium was described as a microorganism capable of degrading benzene even under microaerobic conditions Banerjee et al. (2022) in pure culture. Consequently, the fact that a bacterium can degrade aromatic compounds under microaerobic conditions in pure culture does not mean that it will be a competitive microaerobic degrader as a member of a community. Interestingly, the abundance of Thauera was lower in the heavy DNA fraction than in the light DNA fraction (12.0% vs 16.6%). On the other hand in the medium fractions its abundance was considerably higher, e.g. in fraction A8 (BD = 1.715) it was ~ 23% (data not shown). Similar observation was made by us earlier in case of a microarobic, 13C7-toluene–degrading enrichment community, where the DNA of an aromatic compound-degrading Rhodoferax bacterium was primarily found in medium DNA fractions Táncsics et al. (2018). As mentioned above, Rhizobium was identified as benzene-degrader, since its abundance was considerably higher in the heavy DNA fraction than in the light one (23.0% vs 12.5%). In case of the microaerobic enrichments the two major players, Malikia and Azovibrio were clearly identified as benzene-degraders. DNA of both lineages showed considerably higher abundance in the heavy fraction than in the light fraction (42% vs 33% and 33% vs 26%, respectively). Regarding minor community members of the microaerobic enrichments, Sulfuricurvum kujiense-related DNA was detected only in the light DNA fraction. This result is not surprising, as this bacterium was described as facultatively anaerobic, chemolithoautotrophic, sulfur-oxidizing bacterium, which was first isolated from an underground crude-oil storage (Kodama and Watanabe 2004). Similarly, Geobacter-related DNA was detected only in the light DNA fraction. Although some of these strictly anaerobic bacteria can degrade benzene or toluene anaerobically (Kunapuli et al. 2010; Zhang et al. 2013), they have no role in the aerobic degradation of aromatic compounds. Consequently, this lineage was not labelled during the experiment. The abundance of Rhizobium-related DNA was slightly enriched in the heavy DNA fraction, while Extensimonas-related DNA was almost equal in the heavy and the light DNA fractions. This means that the genus Rhizobium-related bacteria could have some minor role in the degradation under microaerobic conditions, while the role of Extensimonas remained unknown. This later genus belongs to the family Comamonadaceae, which contains numerous genera involved in aromatic hydrocarbon degradation (e.g. Acidovorax, Comamonas, Hydrogenophaga, Polaromonas, Variovorax etc.). On the other hand, only handful of studies link the genus Extensimonas to hydrocarbon degradation Wang et al. (2021).

Fig. 2
figure 2

Relative read abundance of major taxa in Illumina 16S rRNA gene amplicon libraries of heavy and light DNA fractions of 13C-benzene SIP gradients. A, aerobic microcosms; M, microaerobic microcosms. All genera contributing more than 1% abundance were depicted

Full size image

Investigation of a relevant metagenome-assembled Azovibrio genome

As mentioned above, an Azovibrio bacterium was highly abundant in the microaerobic benzene-degrading microcosms. Besides, in one of our previous enrichment experiments of the Siklós sediment, which aimed to investigate xylene degradation under microaerobic conditions, the same Azovibrio lineage was dominant Táncsics et al. (2023). Although the whole genome of this most probably aromatic hydrocarbon-degrading bacterium was assembled earlier by using metagenomic data, it has not been analysed yet. The results of the present study directed our attention to this metagenome-assembled genome, designated as Azovibrio sp. XYLBin2 (WGS accession number JANLJF000000000). Thus, its phylogenomic and functional analysis was performed. Accordingly, it was revealed by the MiGA pipeline, that this Azovibrio-related genome is 2926 790 bp, while its completeness is 99.1%, with low level of contamination (3.8%). The UBCG phylogenomic tree indicated that Azovibrio sp. XYLBin2 clustered most closely to Azovibrio restrictus DSM 23866 T (Fig. 3). On the other hand, the low OrthoANI and AAI values (82.2% and 72.6%, respectively) between the two genomes indicated that the XYLBin2 genome represents a yet uncultured member of the genus Azovibrio.

Fig. 3
figure 3

Phylogenomic tree constructed using UBCGs (concatenated alignment of 92 core genes) showing the phylogenetic position of Azovibrio sp. XYLBin2. Bar, 0.05 substitution per nucleotide position

Full size image

Functional analysis of the metagenome-assembled genome revealed a gene cluster encoding all necessary elements of multicomponent phenol hydroxylase (mPH) system. Activation of the aromatic ring is often performed by benzene/toluene monooxygenases or dioxygenases. However, if these are not available in the genome, the mPH system performs this task in the same way. Although its primary substrate is phenol, it can also catalyze the transformation of benzene into catechol Cafaro et al. (2004). The activity of the enzyme requires the coordinated operation of the dmpKLMNOP genes, which are responsible for the expression of the P0, P1, P2, P3, P4 and P5 components of the enzyme system. As mentioned above, all of these genes were found in the corresponding gene cluster, assuming that benzene was converted to catechol through the activity of the mPH system in the case of Azovibrio sp. XYLBin2. The next step in the aerobic degradation of benzene is the cleavage of the aromatic ring, either through ortho- or meta-cleavage. In the XYLBin2 genome a catechol 2,3-dioxygenase (C23O) gene was found, thus benzene is degraded through the meta-cleavage pathway by this Azovibrio-related bacterium. Phylogenetic analysis of the C23O gene revealed that it encodes an extradiol dioxygenase (EDO), which belongs to the recently described EDO subfamily I.2.I (Fig. 4) Táncsics et al. (2023). It showed the highest similarity to a C23O gene of Dechloromonas denitrificans strain 110 (85% similarity at nucleotide level), followed by C23O genes of Thauera and Azoarcus strains, e.g. Azoarcus olearius DQS-4 T, which was isolated from oil-contaminated soil, but primarily described as a nitrogen-fixing plant growth-promoting bacterium (Chen et al. 2013; Faoro et al. 2016). Besides the C23O gene, genes encoding the oxalocrotonate branch of the lower meta-cleavage pathway enzymes (e.g. 2-hydroxymuconate semialdehyde dehydrogenase; 4-oxalocrotonate tautomerase, 4-oxalocrotonate decarboxylase etc.) were also found in the XYLBin2 genome, suggesting that benzene was degraded through the meta-pathway.

Fig. 4
figure 4

Neighbor-joining phylogenetic tree of the deduced amino acid sequences of catechol 2,3-dioxygenase (C23O) genes encoding class I EDO enzymes. The Azovibrio sp. XYLBin2-related catechol 2,3-dioxygenase revealed by the present study is highlighted with boldface type. Class I EDO subfamilies are indicated at the main branches of the tree. Only bootstrap values > 50 are indicated. Scale bar, 0.2 substitutions per amino acid position

Full size image

It has to be noted here, that the genus Azovibrio contains only one single species, A. restrictus, which was reported to have limited metabolic abilities (Reinhold-Hurek and Hurek 2000). Members of the genus were shown to grow on a limited number of carbon sources (e.g. L-malate, DL-lactate, succinate, fumarate, acetate, propionate and ethanol, and amino acids such as L-glutamate), and they cannot grow on mono- or disaccharides. Their most widely known characteristic is that they can fix atmospheric nitrogen. Besides, some strains show strict microaerophilic lifestyle (Reinhold-Hurek and Hurek 2000) Overall, increasing number of evidence suggests that these potentially microaerophilic bacteria of the Rhodocyclales, aquiring subfamily I.2.I-type EDO genes have another ecological role, namely the degradation of monoaromatic hydrocarbons in petroleum hydrocarbon contaminated subsurface ecosystems.

Conclusion for future biology

The main conclusion of the present study, which potentially affects future bioremediation procedures of petroleum hydrocarbon contaminated subsurface ecosystems, is that distinctly different bacterial communities took part in the degradation of benzene under aerobic versus microaerobic conditions. While under clear aerobic conditions members of the genera Pseudomonas, Thauera and Rhizobium were the key degraders, their role under microaerobic conditions was rather marginal. On the other hand, under microaerobic conditions, members of the genera Malikia and Azovibrio mostly replaced these bacteria. Consequently, results confirmed the previous observation, that Malikia species can be prominent benzene degraders in contaminated aquifers. Additionally, the present study provided further evidence on the assumption that members of the genus Azovibrio, which are potentially microaerophilic bacteria, can play a significant role in the degradation of aromatic hydrocarbons under limited oxygen availability. Whether the subfamily I.2.I-type extradiol dioxygenase provides further competitive advantage to these bacteria over other aerobic hydrocarbon degraders (e.g. over Pseudomonas spp.), remains a question and needs to be further elucidated.