BTEX biodegradation by Bacillus amyloliquefaciens subsp. plantarum W1 and its proposed BTEX biodegradation pathways

Benzene, toluene, ethylbenzene and (p-, m- and o-) xylene (BTEX) are classified as main pollutants by several environmental protection agencies. In this study, a non-pathogenic, Gram-positive rod-shape bacterium with an ability to degrade all six BTEX compounds, employed as an individual substrate or as a mixture, was isolated. The bacterial isolate was identified as Bacillus amyloliquefaciens subsp. plantarum strain W1. An overall BTEX biodegradation (as individual substrates) by strain W1 could be ranked as: toluene > benzene, ethylbenzene, p-xylene > m-xylene > o-xylene. When presented in a BTEX mixture, m-xylene and o-xylene biodegradation was slightly improved suggesting an induction effect by other BTEX components. BTEX biodegradation pathways of strain W1 were proposed based on analyses of its metabolic intermediates identified by LC–MS/MS. Detected activity of several putative monooxygenases and dioxygenases suggested the versatility of strain W1. Thus far, this is the first report of biodegradation pathways for all of the six BTEX compounds by a unique bacterium of the genus Bacillus. Moreover, B. amyloliquefaciens subsp. plantarum W1 could be a good candidate for an in situ bioremediation considering its Generally Recognized as Safe (GRAS) status and a possibility to serve as a plant growth-promoting rhizobacterium (PGPR).

BTEX biodegradation in a liquid medium system. The ability of strain W1 to utilize each individual BTEX component as a sole carbon source was investigated in detail. As BTEX loss can be contributed not only by microbial activity, but also from evaporation, abiotic reactions and absorption onto the cells (both live and dead cells), the 'dead cell control' has been conducted in parallel in order to distinguish the loss by microbial activity from those of 'abiotic losses' . The percentage of remaining BTEX (%C/C 0 ) between live cells and dead cells were compared (Fig. 1). After 24 h, the remaining BTEX compounds in the live cells system were significantly lower than those of the dead cells control, indicating that B. amyloliquefaciens W1 was capable of degrading all BTEX compounds as a single substrate.
It should be noted that the ability to degrade all six BTEX compounds individually was not common even in the BTEX-degraders reported previously. Pseudomonas putida F1, a well-known aromatic degrader, was able to degrade only benzene, toluene and ethylbenzene 28 . Similarly, Pseudomonas sp. OX-1 can utilize only benzene, www.nature.com/scientificreports/ toluene and o-xylene as a sole carbon source 17 . Only a few bacterial strains were reported for their ability to degrade all BTEX compounds individually, for example, Rhodococcus rhodochrous 29 , Pseudoxanthomonas spadix BD-a59 30 , Comamonas sp. JB 31 and Microbacterium esteraromaticum SBS1-7 20 . The percentage of BTEX biodegradation by strain W1 (at 24 h) were 29 ± 2.7, 34 ± 7.1, 29 ± 1.4, 30 ± 0.6, 18 ± 5.0 and 14 ± 4.0% for benzene, toluene, ethylbenzene, p-xylene, m-xylene and o-xylene, respectively (Table S2, supplementary material online). Therefore, the substrate preference of B. amyloliquefaciens W1 can be ranked as follows: toluene > benzene, ethylbenzene, p-xylene > m-xylene > o-xylene. Ethylbenzene and/or o-xylene were frequently reported to be the most recalcitrant compounds in BTEX biodegradation 15,32,33 . Their presence frequently resulted in lower growth 15 and/or BTEX degradation rate 29,30 . Although the toxicity exhibited by o-xylene was not observed in this study, it should be noted that the concentration tested (10 mg/L) was lower than the inhibition concentration mentioned in the relevant literature (e.g. 100 mg/L reported for Ralsotonia sp. PHS1 34 ). Interestingly, strain W1 could effectively utilize ethylbenzene and approximately 29% of ethylbenzene could be degraded within 24 h (Fig. 1).
In order to investigate the synergistic/inhibitory effect in the presence of other BTEX compounds, the biodegradation period in the liquid medium system was extended over a period of 1 week (Fig. 2). When BTEX was presented as a single substrate, all compounds could be rapidly degraded by approximately 30% within the first 12 h of incubation. During this initial period of 12 h, biodegradation of all BTEX components decreased by 5-15% when BTEX were supplemented as a mixture ( Fig. 2; Table S3, supplementary material online). This result is consistent with the literature reporting the retardation of degradation rate in the presence of other BTEX compounds 34 .
It should be noted that the biodegradation of most BTEX compounds (except benzene) in a BTEX mixture became equal or even greater than those of a single substrate scenario once the adaptation period was over ( Fig. 2; Table S3, supplementary material online). Biodegradation of m-xylene and o-xylene were significantly enhanced in a system supplemented with a BTEX mixture (Fig. 2e,f), indicating by 12-15% improvement in the percentage of biodegradation after 24 h of incubation (Table S3, supplementary material online). This result is consistent with the previous observation that the presence of benzene, toluene and ethylbenzene (BTE) enhanced xylene degradation in Pseudomonas putida YNS1 14 . This is not surprising considering that toluene, m-xylene and o-xylene degradation generated a common metabolite in the form of 3-methylcatechol (3-MC). 3-MC was broken down further into cis,cis-2-hydroxyl-6-oxohepta-2,4-dienoate and later on into other metabolites entering into the TCA cycle. Therefore, the presence of toluene or other metabolites that can enhance the degradation of 3-MC would also benefit the degradation of m-xylene and o-xylene. One possible explanation for this may be an induction effect of toluene or its corresponding metabolites toward an oxygenase-related enzyme in strain W1. Toluene and benzoate (a product of toluene degradation) were, for example, reported as the inducers for xylene monooxygenase (XMO) which played important roles in toluene and m-xylene biodegradation 35 . The addition of o-xylene, on the other hand, was reported to upregulate an expression of naphthalene 1,2-dioxygeanse (NDO; nidABEF) which was responsible for o-xylene, m-xylene, p-xylene as well as ethylbenzene utilization in Rhodococcus opacus TKN14 36 .
The inhibition of other BTEX compounds toward benzene biodegradation by strain W1 was also observed in this study ( Fig. 2a; Table S3, supplementary material online). A similar phenomenon in which an inhibition of benzene degradation (approximately 70% reduction in benzene degradation rate) was observed in the presence of other BTEX compounds, especially ethylbenzene, was reported for Stenotrophomonas maltophilia T3-c 37 . The presence of ethylbenzene (at 20 mg/L) also resulted in a lower BTX biodegradation rate in a toluene-enriched microbial consortium containing Rhodocoocus rhodochrous 29 . Moreover, benzene and toluene degradation were also inhibited when p-xylene (50 mg/L) was added into the liquid medium system containing a pure culture of Pseudomonas sp. CFS-215 38 . Similarly, toluene degradation by a toluene-enriched consortium was retarded by 5-10 h when toluene was supplemented as a mixture with benzene, ethylbenzene and p-xylene 39 . A significant decrease in BTEX biodegradation rate was observed in P. spadix BD-a59 when ethylbenzene and o-xylene were presented in the liquid medium 30 . Interestingly, biodegradation of toluene, ethylbenzene and p-xylene by strain W1 seemed to be unaffected by the presence of other BTEX compounds ( Fig. 2b-d). Such effective ethylbenzene degradation in strain W1 is suspected to reduce the strong inhibition effect of ethylbenzene observed by several other researchers mentioned previously.
The first-order kinetic model represented the BTEX biodegradation in the liquid medium system in terms of specific degradation constant (k) and half-life (t 1/2 ) (  40 . In summary, B. amyloloquefaciens subsp. plantarum W1 could effectively degrade all BTEX compounds supplemented individually or as a mixture. Therefore, the strain would be a good candidate for a bioremediation of BTEX-contaminated industrial exhaust stream and/ or BTEX-contaminated wastewater which usually consists of several BTEX compounds in different combination. BTEX biodegradation in a soil slurry system. BTEX biodegradation activity in a liquid medium system and a soil slurry system were compared in a 30-day experiment supplemented with 60 mg/L of BTEX mixture. Generally, the reported concentration of BTEX are in the range of 0.1 to 100 µg/L (ppb) for typical groundwater. The BTEX concentration, however, could be as high as 3,500 µg/L for a contaminated groundwater 41 . Therefore, it should be noted that the range of BTEX concentrations employed in this study is significantly greater than actual BTEX concentrations found in the environment. www.nature.com/scientificreports/ www.nature.com/scientificreports/ By comparing these two systems, the liquid medium system exhibited greater level of overall BTEX degradation (Fig. 3). Catabolic repression from soil nutrients was suspected as a 'probable' cause, as the soil organic matters with hydrophobic characteristics could absorb BTEX compounds into the soil particles 42 . Evidence of catabolic repression by an alternative carbon source was also reported in benzene biodegradation by Ralstonia pickettii PKO1 43 . Nonetheless, it should be noted that the effects of soils on BTEX biodegradation is still inconclusive. While a decrease in BTEX biodegradation rate after an addition of soil has been reported 44 , an addition of organic compounds derived from soil provided a nutritional benefit for growth of P. spadix BD-a59 and thus contributed to the higher BTEX biodegradation rate observed 30 . It should be noted that the dependency of P. spadix BD-a59 on the organic nutrients (unidentified nutrients presented in the solid portion of soil as well as in a yeast extract) may contribute to the difference observed.
In this study, the viable cell count (CFU/mL) and spore count (spores/mL) in both systems were monitored during an entire experimental period. While a total count of approximately 10 8 CFU/mL could be maintained in both systems, a percentage of sporulation (percentage of cells with spore formation) increased drastically and reached 100% within 3 days. This observation agreed with the fact that BTEX degradation rate was fastest during the first 3 days (Fig. 3). Despite a complete sporulation, a continuous decrease in residual BTEX could be observed up to 30 days. Sporulation might be resulted from nutrient depletion in MM considering the fact that MM was a minimal medium containing only inorganic salts. Such nutrient limitation was reported to trigger sporulation in genus Bacillus 45,46 .
In the liquid medium system, most of the BTEX compounds except benzene and o-xylene could be effectively removed to less than 10% within 30 days. This result agreed with the lower preference toward o-xylene ( Fig. 1) as well as the previously described inhibitory effect of other BTEX compounds on benzene biodegradation ( Fig. 2). With a significantly lower degradation observed, the residual BTEX in the soil slurry system was between 13.7% (in case of p-xylene, Fig. 3d) and 49.0% (in case of benzene, Fig. 3a). Therefore, the presence of soil matrix resulted in a significantly slower BTEX degradation in B. amyloliquefaciens W1 (Table S4 and S5, supplementary material online).

Proposed BTEX biodegradation pathways in B. amyloliquefaciens W1. During an aerobic
BTEX biodegradation, aromatic compounds were incorporated with an oxygen atom via activity of mono-or dioxygenases 47 . Analyses of the metabolites produced during the biodegradation by solvent extraction and identification of the metabolites by LC-MS/MS allowed us to postulate the BTEX biodegradation pathways utilized by the target strain-of-interest. The pathway construction was performed based on the previously reported enzymatic reactions from the University of Minnesota Biocatalysis/Biodegradation Database (UMBBD, https :// umbbd .msi.umn.edu) 48 as well as from other relevant literature.
The ability to utilize BTEX in genera Bacillus is not fully elucidated as a result of its minority status in BTEXenriched culture 49 . Only a few Bacillus strains were reported as BTEX degraders, for example, B. subtilis DM-04 (able to degrade benzene, toluene and m-xylene) 50 , B. stratosphericus FLU-5 (capable of growing on toluene, ethylbenzene, o-xylene, m-xylene and p-xylene) 51 and B. cereus ATHH39 (able to degrade toluene) 52 . Thus far, only Bacillus pumilus MVSV3 has been reported for its mono-oxidation (for toluene and ethylbenzene) and dioxidation (for benzene and o-xylene) 22 . In this study, biodegradation pathways of all six BTEX compounds in B. amyloliquefaciens were proposed for the first time (Fig. 4).
Purified naphthalene 1,2-dioxygenase (NDO) from Pseudomonas sp. strain NCIB 9816-4 was reported to convert all three xylene isomers into the substituted benzyl alcohols and benzaldehyde derivatives 64 . Xylene monooxygenase (XMO) from P. putida mt-2, on the other hand, could catalyze only p-xylene and m-xylene 77 . The fact that the conversion of xylene into the substituted benzyl alcohol was observed in all xylene isomers suggested that the xylene degradation in strain W1 was most likely a result of NDO-like enzyme. Previously, evidence of NDO-like gene was reported in Bacillus megaterium strain 2 and strain 3 based on a PCR and hybridization analysis using ndoB gene. Naphthalene, however, was not degraded by those B. megaterium strains, indicating a knowledge gap for NDO-like enzyme in genus Bacillus 78 . Interestingly, a significantly higher signal of metabolites, including corresponding benzaldehydes and corresponding benzoates, was detected in m-xylene and p-xylene degradation by B. amyloliquefaciens W1. This observation suggested a high metabolic flux in the pathways which agreed well with the high degradation observed in m-xylene and p-xylene (Figs. 1, 2, 3). The fact that o-xylene was the least favorable substrate in strain W1 suggested the presence of a bottleneck in the upper part of o-xylene pathway (prior to 3-MC).

Conclusion
B. amyloliquefaciens W1, isolated from a petrochemical waste, was able to degrade all six BTEX compounds presented as an individual compound or a BTEX mixture. Based on the presence of amyE-like gene, strain W1 belongs to the subsp. plantarum which has been frequently reported in association with plants. Although the presence of soil matrix resulted in a significantly slower BTEX degradation, B. amyloliquefaciens strain W1 could effectively remove BTEX in both liquid medium and soil slurry system. To the best of our knowledge, this study was the first direct evidence of a complicated BTEX-biodegrading pathway in the genera Bacillus. Considering its safe status and the ability to thrive in most soil ecosystems, B. amyloliquefaciens W1 would be considered beneficial for in situ bioremediation applications. The well-documented ability of B. amyloliquefaciens subsp. plantarum to grow in association with plants also suggests that strain W1 should be evaluated for its biodegrading ability in the rhizosphere system. Additionally, strain W1 only requires simple cultivation conditions, and therefore can be easily produced on large-scale for scale-up study or commercialization.

Methods
Chemicals and media. Chemicals used in this study, including benzene (99.8% purity), toluene (99.5% purity), ethylbenzene (> 98% purity), p-xylene (> 98% purity), o-xylene (> 98% purity) and m-xylene (> 98% purity), were of GC grade (developed specifically for residue analysis of pesticides with very low content of nonvolatile matters) from well-known manufacturers. Mineral salts medium (MM) 20 was used for an enrichment and biodegradation test. It consisted of (per L) 0.91 g of KH 2  Isolation of BTEX-degrading bacteria. BTEX-degrading bacteria were isolated using an enrichment protocol. The enrichment experiment was conducted in a 25 mL-serum bottle sealed with a butyl rubber septum and an aluminum cap. Briefly, MM supplemented with 1,200 mg/L of a BTEX mixture (200 mg/L of each compound) was inoculated with 1% (v/v) of a liquid petrochemical waste and incubated at 30 °C, 120 rpm. Bacterial culture was transferred into a fresh MM supplemented with the same level of BTEX mixture every 3 days for 1 month. After that, the culture was plated onto a tryptic soy agar (TSA) plate. All bacterial colonies with different morphologies (e.g. form, elevation, margin, color and transparency) were collected and streaked on a fresh TSA plate repeatedly until the pure isolates (exhibiting only one colony morphology) were obtained. Glycerol stocks were prepared and kept at − 80 °C for a long-term storage.
The temperature profile used were as follows: 25 cycles of denaturation at 95 °C for 1 min, annealing at 55 °C for 1 min, and extension at 72 °C for 1.5 min followed by 1 cycle of a final extension at 72 °C for 5 min. The 16S rDNA fragment obtained was purified using a HiYield Gel/PCR DNA Fragments Extraction Kit (RBC Bioscience, Taiwan) and sequenced by Macrogen (Seoul, Korea). The sequence was then compared with the data in GenBank using the basic local alignment search tool (BLAST) of the National Center for Biotechnology Information (NCBI). As Bacillus amyloliquefaciens subsp. amyloliquefaciens and Bacillus amyloliquefaciens subsp. plantarum can be distinguished by the presence of a specific polysaccharide-degrading gene, 2 different sets of primers designed specifically for a starch-liquefying α-amylase (amyA) and a saccharifying enzyme (amyE) ( Table S1, supplementary material online) were used for subspecies identification. The temperature profile used were as follows: 25 cycles of denaturation at 95 °C for 1 min, annealing at 56 °C for 1 min, and extension at 72 °C for 2 min followed by 1 cycle of a final extension at 72 °C for 5 min. PCR products were visualized on 1% agarose gel with a post-staining using Fluorovue Nucleic Acid Gel Stain (SMOBIO, Taiwan).

Quantification of BTEX-biodegrading activity in a liquid medium system. BTEX-biodegrading
activity in a liquid medium system was performed according to the protocol described by Wongbunmak et al. 20  Quantification of BTEX biodegrading activity in a soil slurry system. BTEX-biodegrading activity in a soil slurry system was performed according to the protocol described by Wongbunmak et al. 20 . An uncontaminated soil sample (pristine soil; silt, pH 6.5) was collected from an agricultural area in Ratchaburi Province, Thailand, in March of 2014. The soil particles were passed through an 18 × 18 mesh (opening = 1.429 mm) and then sterilized by autoclaving at 121 °C for 15 min (3 times) before use. The experimental set up consisted of 0.6 mL of the cell suspension mixed with 0.6 g of sterile pristine soil. The cell suspension used in this experiment was concentrated so that the soil slurry system contained an equal number of cells to that of the liquid medium system described above. Then, 60 mg/L of BTEX mixture was supplemented and the experiment was conducted in a similar manner to that of the liquid medium system.

BTEX analysis by GC-FID.
To quantify the remaining BTEX concentration in the system, n-hexane and 1-hexanol were used as an extracting solvent and an internal standard, respectively. The extracting solvent (0.2 mL) and the cell suspension (1 mL) were mixed vigorously for 5 min and then centrifuged at 11,000×g, 4 °C for 10 min. The solvent layer (1 µL) was injected into a gas chromatography (GC-2014, Shimadzu, Japan) equipped with a capillary DB-wax column (J&W Scientific, CA; 30 m-length, 0.250 mm-inner diameter, 0.25 µm-film thickness) and a flame ionization detector (FID) with split mode (10:1). The oven temperature was controlled as follows: held at 50 °C for 5 min; raise to 120 °C at a rate of 5 °C/min and then raise to 240 °C at a rate of 30 °C/min. Helium was used as a carrier gas at 17 mL/min of total flow rate. The injector and detector were operated at 250 °C and 300 °C, respectively. Known concentrations of standard BTEX were supplemented into the MM and extracted in a similar manner to that of the experiment set up. The obtained peak area was used to construct a standard curve between 'Ratio of Area-BTEX and Area-internal standard' and 'BTEX concentration' . The standard curve of each BTEX compound (Fig. S2, supplementary material  www.nature.com/scientificreports/ tion (R-square) were 0.997 for benzene, 0.999 for toluene, 0.998 for ethylbenzene, 0.998 for p-xylene, 0.997 for m-xylene and 0.996 for o-xylene. Remaining BTEX (%) in the system was calculated as %C/C 0 when C was the actual BTEX concentration at a particular sampling time and C 0 was the initial BTEX concentration. The results were presented as an average value from at least 3 independent experiments.
First-order kinetics of BTEX biodegradation. The biodegradation of BTEX in a single substrate and a mixture condition could be described by the first-order kinetics 80 using Eq. (1): where C is the BTEX concentration (mg/L) at a certain time, C 0 is the initial concentration of BTEX (mg/L), k is the specific degradation rate constant (h −1 ) and t is the biodegradation period (h). Biodegradation half-life time was calculated using Eq. (2): where k is the specific degradation rate constant calculated from Eq. (1).
Viable cell and spore enumeration. At a specified sampling time, the cell suspension (200 µL) was taken from the system with a sterile syringe. The sample was serially diluted with sterile NSS. The viable cell count (CFU/mL) and spore count (spores/mL) were determined using a drop plate technique 81 . Briefly, 2 drops of the cell suspension (10 µL) from each dilution were dropped onto LB agar and allowed to dry. Plates were incubated at 30℃ and the number of colonies were counted after 16 h. For the spore count, the cell suspension was heated at 80℃ for 10 min to kill the vegetative cells prior to the enumeration.

Data analysis.
All experiments were conducted in triplicate. The data were presented as the mean of three replicates ± standard deviation of mean. ANOVA test and Tukey's multiple comparison were performed using Minitab software (Release 15, State College, PA, USA) for the biodegradation data to compare that the sample means differ at a significant level P < 0.1.

Metabolites tracking.
The metabolites tracking protocol (Fig. S3, supplementary material online) was modified from that reported previously 20 . Briefly, the cell suspension was prepared as described above. Then, an individual BTEX component was supplemented into 20 mL of the cell suspension at 10% (v/v). The experiment was performed in a 100 mL serum bottle sealed with a butyl rubber septum and an aluminum cap. The degradation was performed at 30 °C for 3 h under shaking condition at 200 rpm. A sterile MM supplemented with individual BTEX component, referred as ' Abiotic control' , was conducted in parallel. Metabolites were extracted by using 15 mL of ethyl acetate. After mixing by sonication for 10 min, the upper layer (ethyl acetate) was collected. The extraction was repeated twice and the solvent phase were pooled together. The residual water in the sample was removed by an addition of sodium sulfite. The sample was dried under vacuum (− 0.8 bar) for 1 h and then dried under nitrogen gas (flow rate of 15 L/min) for 5 min. The dried residue was resuspended in 200 µL of 5% (v/v) acetonitrile (ACN). The sample was filtered with 0.2 µm nylon filter before analysis. Metabolites were analyzed by LC-MS/MS (Dionex UltiMate 3000/Bruker MicrOTOF/maXis) operated in a positive ionization mode (collision energy 20 eV; scan range 40-1500 m/z). Acclaim 120 C18 reversed phase column (2.2 µm, 2.1 × 100 mm) was used for separation. A binary gradient consisting of water (mobile phase A) and ACN (mobile phase B) (both + 0.1% formic acid) at a constant flow rate of 0.3 mL/min was used. A gradient elution was applied as follows: 0-1 min, 5% B; 1-9 min, 5-60% B; 9-12 min, 90% B and 12-15 min, 5% B. The signal from the negative control served as a 'baseline' for identifying the metabolites formed as a result of microbial activity. Each identified metabolite (peak) was then subjected to MS/MS analysis for chemical identification. The fragmentation pattern was compared with the MS/MS database of METLIN (The Scripps Research Institute). Biodegradation pathways as well as possible responsible enzymes in the biodegradation of each BTEX compound by B. amyloliquefaciens W1 were proposed according to the detected metabolites.

Data availability
All data generated or analyzed during this study are included in this published article (and its Supplementary Information file online).