Biodegradation of binary mixtures of octane with benzene, toluene, ethylbenzene or xylene (BTEX): insights on the potential of Burkholderia, Ralstonia, Pseudomonas, and Cupriavidus isolates

The contamination of the environment by crude oil and its by-products, which mainly composed of aliphatic and aromatic hydrocarbons, is a widespread problem. Biodegradation by bacteria is one of the processes responsible for the removal of these pollutants. This study was conducted to determine the abilities of Burkholderia sp. B5, Ralstonia sp. B1, Pseudomonas sp. T1, and Cupriavidus sp. X5 to degrade binary mixtures of octane (representing aliphatic hydrocarbons) with benzene, toluene, ethylbenzene, or xylene (BTEX as aromatic hydrocarbons) at a nal concentration of 100 ppm under aerobic conditions. These strains were isolated from an enriched bacterial consortium (Yabase or Y consortium) that prefer to degrade aromatic hydrocarbon over aliphatic hydrocarbons. We found that B5 degraded all BTEX compounds more rapidly than octane. In contrast, B1, T1 and X5 utilized more of octane over BTX compounds. B5 also preferred to use benzene over octane with varying concentrations of up to 200 mg/l. B5 possesses alkane hydroxylase (alkB) and catechol 2,3-dioxygenase (C23D) genes, which are responsible for the degradation of alkanes and aromatic hydrocarbons, respectively. This study strongly supports our notion that Burkholderia played a key role in the preferential degradation of aromatic hydrocarbons over aliphatic hydrocarbons in the previously characterized Y consortium. The preferential degradation of more toxic aromatic hydrocarbons over aliphatics is crucial in risk-based bioremediation.


Introduction
Petroleum hydrocarbons are among the widely reported contaminants of surface, groundwater, and marine environments (McGenity et al. 2012;Murphy et al. 2016). These compounds pose health hazards to humans and other ecological receptors. Many aromatic hydrocarbons such as benzene, toluene, ethylbenzene and xylene (collectively known as BTEX) and polycyclic aromatic hydrocarbons (PAHs) are known to have carcinogenic, neurotoxic and mutagenic potentials (Samanta et al. 2002;Haritash and Kaushik 2009;Masekameni et al. 2019). Another groups of hydrocarbons, the aliphatic hydrocarbons, which composed of straight chain, branched and cycloalkanes are usually more abundant in oil, but relatively less toxic compared to aromatic hydrocarbons (Potter and Simons 1998;Bacosa et al. 2010).
The differences in the toxicities of these groups of hydrocarbons have attracted attention towards a riskbased remediation of hydrocarbon contaminated environments.
In polluted sites, hydrocarbons undergo biodegradation by actions of natural bacterial populations capable of utilizing these compounds as carbon and energy sources Adzigbli et al. 2018;Dominguez et al. 2019;. Upon the introduction of hydrocarbons, the natural micro ora shifts to favor the growth of oil-degrading bacterial populations and often result in a signi cant shift in community structure (Severin et al. 2016;Gemmell et al. 2018;Bacosa et al. 2018;Kamalanathan et al. 2019;Steichen et al. 2020). The ability of the bacterial community to degrade the hydrocarbons is affected by various factors that include the type of bacteria, nutrients, sunlight, temperature, shoreline energy, and oil concentration (Bacosa et al. 2015;Gemmell et al. 2016;Liu et al. 2017;Evans et al. 2017;Williams et al. 2017;Sun et al. 2018;Bacosa et al. 2021). The biodegradation of oil is a sequential process in which the aliphatic hydrocarbons are generally degraded faster over aromatic hydrocarbons (Greenwood et al. 2008). However, the interest in bacteria that utilize aromatic hydrocarbons faster that aromatic hydrocarbons has received great interest in the past decade in an attempt to achieve a low-cost and more effective remediation of contaminated sites (Bacosa et al. 2010).
Despite the many research studies conducted on the bacterial biodegradation of BTEX components, most studies have focused on the biodegradation of only one or two components per bacterial isolate (Singh et al. 2020).
We obtained a bacterial consortium through enrichment culturing from Yabase oil eld in Akita, Japan (called "Y" consortium) that preferred to degrade aromatic hydrocarbons over aliphatic hydrocarbons (Bacosa et al. 2010). This consortium is mainly composed of bacteria closely related to Achromobacter, Alcaligenes, Cupriavidus, Achromobacter, and Rhodanobacter. Real-time polymerase chain reaction (PCR) analysis using primers targeting these bacterial groups revealed that Burkholderia increased with the degradation of xylene, a representative aromatic hydrocarbon in a binary mixture with decane (Bacosa et al. 2011). However, Cupriavidus was strongly associated with the disappearance of decane as a representative aliphatic hydrocarbon. In this study, we aimed to isolate representatives from these genera to investigate their respective potential to degrade aromatic and aliphatic hydrocarbons. Speci cally, we were interested in isolating Burkholderia to determine its abilities to degrade binary mixtures of aliphatic and aromatic hydrocarbons.

Bacterial isolation
The bacterial strains used in this study were isolated from the bacterial consortium obtained from Yabase oil eld in Akita, Japan, called "Y" consortium that preferentially degrade aromatic hydrocarbons over aliphatic hydrocarbons in kerosene as described in Bacosa et al. (2010) and Bacosa et al. (2011). Isolates were obtained through a serial dilution of bacterial suspension from the enrichment culture and plated in petri plates containing Bushnell Haas Medium (BHM), 1.5% agarose, and 20 µl of kerosene as the sole carbon and energy source (Bacosa et al. 2013;Bacosa and Inoue 2020). The plates were incubated for two weeks at 30°C. Colonies with unique morphological characteristics were selected and puri ed by subsequent plating. Puri ed isolates were then cultured in test tubes containing BHM with 1% (v/v) kerosene incubated at 30°C with shaking.

Identi cation of bacterial isolates
The isolates were identi ed by sequencing a fragment of 16S rRNA gene following the previously established protocol (Bacosa et al. 2010;Bacosa and Inoue 2020). Brie y, the cells were pelleted by centrifugation and total genomic DNA was extracted using Proteinase K, TTNE buffer, and TE saturated phenol. The almost full-length fragment of 16S rRNA gene was ampli ed using universal primers 10F and 1500R (Takami et al. 1999). After con rmation of the desired PCR product by gel electrophoresis, the PCR products were puri ed using Gene Elute Mammalian Total RNA Miniprep Kit (Sigma Aldrich) according to the manufacturer's protocol. The puri ed DNA was then sequenced following the Big Dye Terminator v3.1 Cycle sequencing chemistry in Applied Biosystems Genetic Analyzer 3130 (Bacosa et al. 2013;Dominguez et al. 2019;Bacosa and Inoue 2020). The isolates were preliminarily sequenced using primer 10 F. To sequence a nearly full length 16S rRNA gene fragment, representative strains were sequenced using primers 907R and 1500R. DNA sequences were assembled by Genetyx software and compared to the sequences in the GenBank. The sequences were submitted to the GenBank under Accession Numbers AY741343, AB109778, EF424401, AB266610.
Detection of alkane-degrading and aromatic-degrading genes in B5 isolate The 238-bp fragment of catechol 2,3-dioxygenase gene (C23D) was ampli ed using primers 23CAT-F and 23CAT-R developed by Mesarch et al. (2000) following the PCR ampli cation conditions therein. C23D is a common pat for the degradation of aromatic hydrocarbons. A fragment (~ 100 bp) of alkane hydroxylase gene (alkB), which is responsible for the terminal oxidation of alkane was ampli ed using primers AlkB-F and AlkB-R by Powell et al. (2006) following established conditions.

Biodegradation experiment
The biodegradation experiment was conducted to determine which isolate would degrade aromatic hydrocarbons faster than aliphatic hydrocarbons under aerobic conditions. We used our established procedure on binary mixture of aliphatic and aromatic hydrocarbons (Bacosa et al. 2011). To accomplish this, the isolates were precultured in BHM medium with kerosene until reaching the late exponential phase of growth (Bacosa et al. 2010). The cells were then harvested by centrifugation, washed twice with phosphate saline buffer, and resuspended in BHM. Cell suspension (0.5 mL) was inoculated in 120-ml sterile serum vial containing 9.5 ml of BHM. The bacterial cells were added to attain an initial density of 1x10 7 cell per ml (Bacosa et al. 2011). The vial was then capped with Te on-line septum and sealed with an aluminum cap. n-Octane, benzene, toluene, ethylbenzene, and p-xylene were individually spiked into the vials using a microsyringe at a nal concentration of 100 mg/l. Each vial contains n-octane or any of the BTEX compound in what is termed as a binary mixture representing an aliphatic and an aromatic hydrocarbon compound. All hydrocarbons used have purities of > 99% and obtained from Wako Pure Chemicals (Osaka, Japan). The following were the sets of binary mixture prepared for the isolates: octane-benzene, octane-toluene, octane-ethylbenzene, and octane-xylene. Three replicate vials were prepared for each treatment, wherein vials were placed in an orbital shaker and incubated at 30ºC with shaking at 120 rpm. Control vials were not prepared in this experiment because our previous work showed no appreciable change in concentration in the uninoculated sealed vials (Bacosa et al. 2011;Bacosa et al. 2012).
Burkholderia sp.was further evaluated for the degradation of binary mixture of octane and benzene with varying initial concentrations. Benzene was chosen among the BTEX compounds because it has no substituted methyl or alkyl group that provides for alternate sites for initial oxidation of the aromatic ring.
The concentrations tested were 100 mg/l octane-100 mg/l benzene, 100mg/l octane-200 mg/l benzene, and 200 mg/l octane-200 mg/l benzene. Another treatment was prepared for the degradation of a mixture of octane, benzene, and toluene at a nal concentration of 100 mg/l. The experimental procedure was prepared using a new inoculum. Control vials with no microbial inoculants were prepared to account for the loss of hydrocarbons in this set of experiment.

Hydrocarbon analysis
The vials were analyzed periodically for the residual hydrocarbon compounds using headspace gas chromatography as previously described (Bacosa et al. 2011;Bacosa et al. 2012). Brie y, headspace (50 µl) was withdrawn from the vial using a 250 µl microsyringe (Hamilton, USA) and injected directly into a gas chromatograph (GC-4000, GL Sciences Inc., Japan) equipped with a ame ionization detector (FID).
The column used was an Inert-Cap 17 MS column (30 m long, 0.25 mm internal diameter, 0.25 µm lm thickness). The injector and detector temperatures were set at 350ºC. The column temperature was adjust based on the combination of hydrocarbons substrates and ranged between 80ºC and 100ºC.
Helium was used as a carrier gas at a ow rate of 1 ml/min. The quanti cation of hydrocarbons was achieved by plotting against a seven-point standard curve prepared according to the combination of the hydrocarbon substrates.

Growth in PAHs
The growth of the four isolates in selected PAHs compounds, namely: uorene, phenanthrene, anthracene, and pyrene (Wako, Osaka, Japan) was evaluated following established procedure Bacosa and Inoue 2020). Brie y, each isolate was incubated in 30 ml test tube containing 10 ml of BHM and 1000 mg/l of each PAH compound, for 14 days with shaking at 30ºC. The growth after 14 days of incubation was measured using spectrophotometer at an optical density of 600 nm and evaluated using standard criteria (Chaerun et al. 2004; Bacosa and Inoue 2020).

Results
Thirty bacterial strains were successfully isolated from the "Y" consortium. Preliminary sequencing using primer 10F revealed that these isolates were closely related to Burkholderia, Ralstonia, Pseudomonas and Cupriavidus. Among the isolates, four representative strains representing the identi ed genera were sequenced further using 907R and 1500R. These isolates are B5-closely related to Burkholderia cepacia ATCC 27515, B1-closely related to Ralstonia basilensis, T1-closely related to Pseudomonas panipatensis Esp1, and X5-closely related to Cupriavidus sp. KU-21 ( Table 1). All of these isolates have similarities of greater than 99.5%. Note that only Burkholderia and Cupriavidus were detected by clone library analysis in "Y" consortium (Bacosa et al., 2010).
All of the four isolates were tested in the degradation of binary mixture of octane with benzene, toluene, ethylbenzene or p-xylene. B5 degraded all BTEX compounds faster than octane (Fig. 1). It was observed that xylene was degraded the fastest (within 20 hours) while toluene the slowest (after 100 h). The signi cant degradation of octane commenced when octane was nearly totally depleted. Although ethylbenzene was degraded a bit faster than octane in ethylbenzene-octane mixture, the degradation of these two compounds were closely similar. Isolates B1, T1, and X5 showed a different pattern of action wherein octane was degraded faster than any of benzene, toluene and xylene. However, this is not the case in octane-ethylbenzene mixture, where octane and ethylbenzene were similarly degraded. Interestingly, isolates B1 hardly degraded toluene, while T1 and X5 was not able to utilize benzene even after prolonged incubation to 280 h.
Another set of experiment was performed to evaluate the response of B5 in varying concentrations of octane and benzene. Benzene was remarkably degraded more rapidly than octane regardless of concentration of octane or benzene (Fig. 2). A concentration of 100 mg/l benzene was completely degraded within 40 hours even in the presence of 100 mg/l and 200 mg/l of octane. When added at higher concentration (200 mg/l), benzene degradation occurred longer such that complete degradation happened in 60 h.
As shown in Fig. 3, isolate B5 harbors both C23D and alkB genes, which are associated with the oxidation of aromatic hydrocarbons and aliphatic hydrocarbons, respectively. When a mixture of three compounds was inoculated and incubated with B5, complete degradation of benzene happened in less than 40 h, followed by toluene at 50 h, and octane at 80 h (Fig. 2E). Moreover, B5 also showed good growth in representative PAHs such as uorene, phenanthrene, anthracene, and pyrene suggesting its abilities to degrade these PAHs compounds (Table 2).

Discussion
Crude oil and its by products are highly complex mixtures composed of more toxic aromatic hydrocarbons and more abundant yet less harmful aliphatic hydrocarbons. Generally, aliphatic hydrocarbons are consumed faster than aromatics by microorganisms (Greenwood et al. 2008;Liu et al. 2017). The search for microbial consortium or bacterial strain capable of degrading the more toxic components in pollutant mixture such as hydrocarbons is important in risk-based remediation. Here, we demonstrated that a strain of Burkholderia consumed the more toxic aromatic hydrocarbons rapidly.
The source consortium (Y or Yabase consortium) was composed of a variety of bacteria including Achromobacter, Alcaligenes, Cupriavius, Burkholderia, Herbaspirillum, Paucimonas, Rhodanobacter, Pseudoxanthomonas, Epilithonimonas, and Terrimonas (Bacosa et al. 2010). However, among the 30 unique colonies obtained during isolation, only Burkholderia and Cupriavidus were identi ed. Ralstonia and Pseudomonas, which were not detected by clone library, were isolated. Isolation using kerosene on agar plate could have favored the colony formation of these bacteria. Possibly, the other bacteria identi ed using 16S rRNA sequencing of clone library did not form colonies as more than 99% of microorganism are unculturable and cannot form colonies on agar plates (Nichols 2007).
Among all the isolates, B5 degraded all BTEX compounds faster than octane. Using primers designed to amplify the 16S rRNA genes, both Burkholderia and Cupriavidus increased with the degradation of pxylene as revealed by real-time PCR (Bacosa et al. 2012). Here, we clearly demonstrated that the B5 isolate degrade p-xylene faster than octane when these compounds were added together in the same bottle. This strongly supports our claim that Burkholderia were the key aromatic degraders in Y consortium Burkholderia is one of those bacterial species, whether alone or in consortium, known to degrade BTEX compounds, heavy oil, and various aliphatic and aromatic hydrocarbons (Lee et al. 2012;Lee et al. 2019: Bacosa andInoue 2020). In this study, it was observed that xylene was degraded the fastest while toluene the slowest. This contrasted with what was stated by El-Naas et al. (2014) that toluene is the most easily biodegradable among the BTEX compounds due to the presence of the substituent group on the ring that offers an alternative route of attack on the side chain or oxidize the aromatic ring.
There was a different pattern of biodegradation of isolates B1, T1, and X5 wherein octane was degraded faster while octane and ethylbenzene were similarly degraded. Ethylbenzene degradation can take place by directly oxidizing the aromatic ring or acting on the ethyl group (Chakraborty et al. 2005). Likely, the isolates oxidized octane and ethyl side chain in ethylbenzene concurrently resulting in simultaneous degradation in the binary mixture. Benzene is known to be more recalcitrant than its alkylated derivatives toluene, ethylbenzene and xylene isomers because benzene molecule is thermodynamically stable due to the symmetric π-electron system of the aromatic ring and the lack of potentially destabilizing or reactive substituents (Vogt et al. 2011). Unlike TEX compounds with substituted alkyl side chains, benzene degradation can only proceed with the initial attack of the ring so degradation often proceeds very slowly (Heider 2007;Vogt et al. 2011)). Probably, the genes responsible for benzene degradation were repressed by the metabolite of octane or these two isolates do not have benzene-degrading genes at all (Juhasz et al. 2002;Hennessee and Li 2016). The unique capability of B5 isolate to oxidized aromatic hydrocarbons much faster than octane signi es its innate characteristics to consume aromatic hydrocarbons for growth (Lee et al. 2019).
When B5 was exposed to varying concentrations of octane and benzene, benzene was remarkably degraded more rapidly than octane and when added with higher concentration of benzene, degradation occurred longer. It is possible that at increasing concentration, the consequent initial phase lag time is also increasing (Musat and Widdel 2008;Laban et al. 2009). Note that octane was characterized by more than 20 h of lag phase and complete degradation occurred in 80 hours regardless of initial concentration. Doubling the benzene concentration did not result in the inhibition of aromatic-preferring potential of B5 and its degradation pattern suggesting that the genes for aromatic hydrocarbons of this bacterium are expressed, not repressed. The C23D gene, which is an important gene and pathway for aromatic hydrocarbons degradation, was observed in B5. The dioxygenases activate the aromatic nucleus by introducing molecular oxygen to yield phenol or cis-benzene dihydrodiol, compounds that are further oxidized to catechol (Gibson and Parales 2000;Tao et al. 2004). The aromatic ring of catechol is nally cleaved by further dioxygenases in ortho-or meta-position (Vaillancourt et al. 2006). Also, a fragment of alkB gene which is responsible for the initial attack on terminal carbon in the alkane chain was detected in B5. These two genes harbored by B5 could be expressed for the degradation of both octane and BTEX compounds during the incubation period. Moreover, the degradation potential of B5 is not just limited to BTEX. It also showed good growth in different PAH compounds suggesting that it contains PAH dioxygenase genes likely responsible for the attack of the ring in these more complex and more recalcitrant compounds. This is consistent with our observation with "Y" consortium in which PAHs of up to 3-rings were preferentially degraded over alkanes (Bacosa et al. 2010) Overall, the ndings of the experiments presented here showed that Burkholderia sp.B5 has remarkable ability to preferentially degrade all BTEX compounds completely and much faster than octane. Even if the concentration of both benzene and octane varied and increased, the preferential degradation of benzene over octane occurred. B5 also exempli ed good growth in uorene, phenanthrene, anthracene and pyrene indicating their potential to degrade several PAHs compounds. B5 harbor alkane hydroxylase and C23D playing key roles in the oxidation of aliphatic and aromatic hydrocarbons, respectively. These sets of evidences support our previous ndings that Burkholderia in the Y consortium plays a crucial role in the preferential biodegradation of aromatic hydrocarbons over aliphatic hydrocarbons. Our results advance our understanding on the roles of aromatic-preferring bacteria towards a risk-based remediation. Bacteria that rapidly degrade the more toxic components of pollutant mixtures are useful in cost effective, e cient and successful bioremediation. Figure 1 Degradation of the binary mixtures of octane with benzene, toluene, ethylbenzene or xylene by Burkholderia sp. B5, Ralstonia sp. B1, Pseudomonas sp. T1 and Cupriavidus sp. X5  Detection of alkane hydroxylase (alkB) and catechol 2,3-dioxygenase genes in Burkholderia sp. B5

Supplementary Files
This is a list of supplementary les associated with this preprint. Click to download.