Australasian Biotechnology (backfiles)
AusBiotech
ISSN: 1036-7128
1997
Australasian Biotechnology,
Volume 7 Number 6, November/December 1997, pp.350-354

Pyrene Degradation and Metabolite Formation by Burkholderia cepacia Strain VUN 10,003

Albert L. Juhasz^1,

Centre for Bioprocessing and Food Technology, Victoria University of Technology, PO Box 14428 MCMC, Melbourne, Australia, 8001
^1 Current address: Division of Land and Water, CSIRO, Private Bag 2, Glen Osmond, Adelaide, Australia, 5064


Code Number: AU97048
Size of Files:
    Text: 21.4K
    Graphics: Line drawings (gif) - 7K

The degradation of pyrene, a polycyclic aromatic hydrocarbon containing four fused benzene rings, was evaluated using Burkholderia cepacia strain VUN 10,003. Over 70% of the ^14C-pyrene was mineralised after 120 hours. GC-MS analysis of TLC-isolated pyrene metabolites identified a pyrene dihydrodiol and pyrenol as ring oxidation products of pyrene and 4-hydroxynaphthenone and 4-phenanthroic acid as ring fission products.

Introduction

Although the microbial degradation of PAHs has been studied extensively, the biochemical principles underlying their degradation has been studied to a lesser extent. For bioremediation to be considered a viable technology for the decontamination of PAH-polluted sites, an understanding of the microorganisms, enzymatic processes and metabolic pathways is necessary to optimise the degradation of these compounds (Cerniglia, 1992). The ultimate aim of bioremediation is to reduce the potential toxicity of environmental contaminants by degrading them to harmless constituents such as carbon dioxide and water (Wilson and Jones, 1993). However, the biological degradation of PAH compounds often results in the incomplete degradation of the compounds, resulting in the production and accumulation of intermediate products. Metabolites including dihydrodiols, phenols and arene oxides have been identified as having carcinogenic and mutagenic properties (Datta and Samanta, 1988). As such, these metabolites pose an even greater risk to the environment than the parent compounds due to their increased polarity, water solubility and mobility. The determination of PAH metabolic pathways may help to predict the fate of PAH compounds in the environment and how they are detoxified. This will assist in assessing microorganisms for their potential for bioremediation. This study reports the degradation of pyrene by Bu. cepacia strain VUN 10,003 and the identification of metabolites produced from the degradation of the compound.

Materials and Methods

Chemicals

[4, 5, 9, 10-^14C] Pyrene (58.7 mCi/mmol), unlabelled pyrene, pyrenol and chemical reagents were purchased from Sigma. Solvents were purchased from Lab Chem, Ajax Chemicals. All of the solvents and chemicals were high purity grade reagents.

Microorganisms

Bu. cepacia strain VUN 10,003 was previously isolated from PAH-contaminated soil (Juhasz et al., 1995). VUN 10,003 could grow on fluorene, phenanthrene or pyrene (100 mg/l) as the sole carbon and energy source; the complete degradation of these compounds occurred within 7-10 days (Juhasz et al., 1997). Resting cell experiments performed with high density cell inocula of VUN 10,003 demonstrated that benzo[a]pyrene, dibenz[a,h]anthracene and coronene could also be degraded (Juhasz et al., 1996).

Stock Solutions, PAH-Containing Media and Growth Conditions

Unlabelled pyrene was prepared in dimethylformamide (DMF) at concentrations of 25 and 100 mg/ml. A basal salts medium (BSM) (Juhasz et al., 1996) was supplemented with individual PAHs to achieve final concentrations of 250 mg/l or 500 mg/l of pyrene. Unless otherwise stated, cultures were incubated in the dark at 30 C and 175 rpm.

Pyrene Degradation

Bacterial inocula were prepared as follows: VUN 10,003 was grown in four, 1.5 litre volumes of BSM containing 250 mg/l pyrene. Following the complete degradation of pyrene, cells were harvested by centrifugation at 5,000 rpm for 10 minutes. The cell pellet was washed twice in Ringers solution and resuspended in BSM to achieve a ten-fold concentration in cell biomass. Aliquots (20 ml, 5 x 108 cells/ml) of the cell suspension were inoculated into biometer flasks (Bellco Glass) to evaluate mineralisation of pyrene. Each flask was supplemented with 1.0 uCi of [4, 5, 9, 10- ^14C] pyrene (58.7 mCi/mmol) and 250 mg/l of unlabelled pyrene (0.2 ml). Mineralisation of pyrene was determined by monitoring the distribution of ^14C in the culture media, cell pellets and gaseous phases. Uninoculated PAH-containing media and mercuric chloride killed cells served as the controls. For the isolation of pyrene metabolites, VUN 10,003 (500 ml, 5 x 108 cells/ml) was inoculated into BSM (3.5 litres) containing unlabelled pyrene at a concentration of 500 mg/l (20 ml). Pyrene cultures were incubated for 120 hours and samples were taken routinely over the time course period.

Analytical Procedures

^14CO2 was collected in 0.1 M NaOH (5.0 ml). At various time intervals, the NaOH was removed from the flask side arm and replaced with fresh NaOH. After the final sample, 10 M HCl (0.5 ml) was added to the culture medium to release dissolved CO2. At each time point, dilutions of the NaOH were prepared with cytoscint scintillation cocktail (ICN) and the ^14C radioactivity was quantified using a Pharmacia, Wallac 1410 scintillation counter. At the end of the incubation period, the cultures were centrifuged (15,000 rpm for 10 minutes; Beckman J2-HS centrifuge with JA 21 rotor) and the supernatants were assayed for radioactivity. Cell pellets were extracted with dichloromethane (DCM) and the beta emissions measured. To determine the amount of ^14C incorporated into cellular material, the cell debris, after extraction with DCM, was suspended in water (5.0 ml), diluted in scintillation fluid and the radioactivity assayed.

Pyrene metabolites were extracted from cell-free culture fluids with DCM by the method of Heitkamp et al. (1988). Pyrene metabolites were isolated and purified by thin layer chromatography (TLC) (Guerin and Jones, 1988). PAH crude extracts (5-10 ul) were applied to the TLC plates. TLC analyses were performed with silica gel 60 plates (Merck) using a three phase solvent system. Separation was achieved with benzene:hexane (1:1, v/v), hexane:acetone (8:2, v/v) and benzene:acetone:acetic acid (85:15:5 v/v/v). After solvent development, PAH metabolites were visualised under UV light (302 nm) (LKB 2011 Macrovue transilluminator, Bromma). Individual pyrene metabolites were isolated from TLC plates by removing bands of silica gel containing the metabolites and extracting with methanol. The
purified metabolites were stored in methanol at -20 C until further analysis.

Pyrene metabolites were identified using a Varian Star 3400 gas chromatograph equipped with a Varian Saturn II mass spectrometer (MS) and a BPX-5 (25 m x 0.22 mm, SGE, Melbourne, Australia) capillary column. The MS was operated in electron impact mode with an electron energy of 70 eV over a scan range of 45-400 Da. The column temperature was programmed at 100 C for one minute, followed by a linear increase of 10 C/min to 300 C, holding at 300 C for 9 minutes. The injector and transfer line temperatures were maintained at 250 C and 300 C respectively. Spectra were analysed using Star Chromatography software.

Results and Discussion

Pyrene was rapidly mineralised to ^14CO2 by VUN 10,003 after an initial lag period of 10 hours. The pyrene-induced culture mineralised 70.5% of the pyrene after 120 hours incubation (Figure 1A). The distribution of the remaining recovered labelled carbon was as follows: 5.5% of the ^14C was recovered in the organic phase, 4.1% in the aqueous phase and 17.8% of the ^14C was recovered from the cell debris (Figure 1B). These results are indicative of the low amount of polar and non-polar metabolites produced by VUN 10,003. The higher amount of labelled carbon detected in the cell debris is an indication of the proportion of pyrene carbon that was incorporated into the cellular material. Abiotic degradation of pyrene in control flasks containing no cells or mercuric chloride killed cells was minimal; small amounts of label were detected in the aqueous phase (0.9-2.1%) and as 14CO2 (0.5-0.6%). Pyrene has previously been shown to be mineralised by nocardiaform bacteria. Mycobacterium sp. strains PYR1 (Heitkamp et al., 1988) and RJGII 135 (Schneider et al., 1996) mineralised approximately 50% of added pyrene, while Rhodococcus sp. strain UW1 (Walter et al., 1991) mineralised 72% of added pyrene after 14 days incubation.

    Figure 1: Mineralisation of pyrene (A) by Burkholderia cepacia strain VUN 10,003. The distribution of the labelled carbon (B) in the organic phase, aqueous phase, gaseous phase and cell debris is also shown.

Over 20 different metabolite bands were resolved by TLC from the crude pyrene supernatant extracts of VUN 10,003. The chromatographic mobility (Rf) of individual metabolites varied from 0.22, for the most polar compound to 0.96 for ring oxidation products or non-polar metabolites (data not shown). The metabolite banding profiles of VUN 10,003 changed over the incubation period with the appearance and disappearance of various compounds.

Four pyrene metabolites of VUN 10,003 were identified by GC-MS (Figure 2). Metabolite 1 had a molecular ion (M+) at m/z 236 and a base peak at m/z 218. This represented a loss of an H2O unit (M^+-18). Fragmentation ions were detected at m/z 189, m/z 176 and m/z 94. The mass spectral fragmentation pattern suggests that metabolite 1 was a pyrenedihydrodiol, however, it is not clear whether the ring cleavage occurred at the 1,2- or the 4,5- position.

    Figure 2: Metabolites isolated from the microbial degradation of pyrene by Burkholderia cepacia strain VUN 10,003. The compounds in brackets represent the most likely structure of the dihydrodiol, however, the absolute stereochemistry could not be determined.

Heitkamp et al., (1988) proposed that the initial oxidation of pyrene by a Mycobacterium sp. occurred at the 4,5-position, resulting in the formation of 4,5-dihydroxy-4,5-dihydropyrene. Although pyrene-1,2-dihydrodiol was not detected as a ring oxidation metabolite, Heitkamp et al. (1988) suggested that the formation of 4-hydroxy-perinaphthenone, an isolated ring fission metabolite, probably resulted from the ring oxidation and cleavage of 1,2-dihydroxy- 1,2-dihydropyrene. Two analogous pathways were proposed for the initial oxidation and ring fission of pyrene by Rhodococcus sp. strain UW1 (Walter et al., 1991). Walter et al. (1991) were unable to determine the configuration of metabolite I (C16H10O4), however, they proposed that the initial oxidation of pyrene occurred at either the 1,2- or the 4,5- position. Oxidation of pyrene at the 1,2- position seems likely since PAHs with similar structural configurations are attacked at this position. A number of studies have indicated that bacteria initially oxidise naphthalene and phenanthrene by incorporating molecular oxygen into the aromatic molecule to form 1,2- dihydrodiols (Kelly et al., 1990; Cerniglia and Heitkamp, 1989; Cerniglia, 1984; Cox and Williams, 1980).

The formation of dihydrodiols is indicative of dioxygenase enzyme systems (Cerniglia, 1992). Prokaryotes are known to utilise dioxygenase enzymes to incorporate two atoms of oxygen into aromatic hydrocarbons, which results in the production of dihydrodiols with a cis formation. Both cis and trans pyrene dihydrodiols were detected from the degradation of pyrene by Mycobacterium sp. (Heitkamp et al., 1988), suggesting that the organism was capable of multiple pathways for the initial oxidative attack on the compound. Heitkamp et al. (1988) proposed that in addition to the dioxygenase enzyme system, a monooxygenase catalysed reaction was responsible for the formation of trans pyrene dihydrodiol.

Metabolite 2 had a molecular ion (M^+) at m/z 218 and a fragment ion at m/z 189 (M^+-29), representing the loss of a -CHO group. The GC-MS retention time and the mass spectral fragmentation pattern were identical to those of authentic 1-hydroxypyrene. The position of the hydroxyl moiety was unable to be determined, however, 1-hydroxypyrene has been reported as a fungal metabolite of pyrene (Cerniglia et al., 1986). In addition to the oxidative metabolism of pyrene by VUN 10,003, the formation of pyrenol may have resulted from the non-enzymatic dehydration of pyrene dihydrodiols (Heitkamp et al., 1988).

GC-MS analysis of metabolite 3 showed a molecular ion (M+) at m/z 196 and fragment ions at m/z 168 (M^+-28) and m/z 139 (M^+-57). The fragment ions indicated losses of a -CO group (M^+-28) as well as a -CO group plus a -COH group (M^+-57). The mass spectral analysis is consistent with a molecular formula of C13H8O2 and an aromatic hydrocarbon containing single keto and hydroxyl moieties (Heitkamp et al., 1988). The chromatographic characteristics, molecular weight and mass spectral fragmentation pattern indicated that metabolite 3 was 4-hydroxyperinaphthenone.

Metabolite 4 had a molecular ion (M+) at m/z 222 and fragment ions at m/z 205 and m/z 177. The major ion fragments represented probable losses of an -OH group (M+-17) and a -COOH group (M+-45). Minor fragment ions were also observed at m/z 194 (M+-28), m/z 165 and m/z 151 (M^+-71), representing probable losses of a -CH2=CH2 from an aromatic ring (M^+-28), the loss of a -C from m/z 177 and the loss of -CHCCOOH plus a -H (M^+-71) from an accompanying hydrogen shift respectively (Heitkamp et al., 1988). The derivatised metabolite (methylated) had a M^+ at m/z 236, representing a mass increase of 14 mass units over the underivatised compound. Fragment ions were observed at m/z 221 (M^+-15), m/z 205 (M^+-31) and m/z 177 (M^+-59). The fragment ions represented the loss of a -CH3 group (M+-15), the loss of a -OCH3 unit (M^+-31) and the loss of a -COOCH3 unit (M^+-59). Metabolite M15 was given the molecular formula of C15H10O2. The molecular formula, chromatographic characteristics and mass spectral fragmentation pattern of metabolite 4 is indicative of 4-phenanthroic acid.

Cis dihydrodiols are further metabolised by bacteria after rearomatisation through cis dihydrodiol dehydrogenase (Cerniglia, 1984). This yields dihydroxylated derivatives which are further metabolised by the enzymatic cleavage of the aromatic ring. Two ring fission products (metabolites 3 and 4) were isolated from the culture supernatants of VUN 10,003 and identified as 4-hydroxyperinaphthenone and 4-phenanthroic acid. It appears that the formation of these metabolites resulted from the ring fission of two separate ring oxidation products. 4-Hydroxyperinaphthenone probably resulted from the ring oxidation and cleavage of the alpha ring of pyrene (Heitkamp et al., 1988). 1,2-Dihydroxy-1,2-dihydropyrene was not detected in crude extracts which may be due to ring fission of the compound occurring at a fast rate, and as such pyrene 1,2-dihydrodiol may not accumulate in the culture medium.

The ring fission of 4,5-dihydroxy-4,5-dihydropyrene probably resulted in the formation of 4-phenanthroic acid. Both Heitkamp et al. (1988) and Schneider et al. (1996) isolated 4-phenanthroic acid from the degradation of pyrene by Mycobacterium species. It was proposed that the formation of 4-phenanthroic acid resulted from a 1-carbon excision from the K-region of pyrene, however, the mechanism of this reaction is not known. A similar mechanism was observed during the degradation of fluoranthene by Alcaligenes denitrificans WW1 (Weissenfels et al., 1991). After the initial hydroxylation of fluoranthene at the 9,10- position, 7-hydroxy-8-acenaphthylenealdehyde was expected as the result of an aldolase reaction on the ring fission product. Instead, 7-hydroxy-acenaphthylene occurred by a 1-carbon excision from the aromatic aldehyde. Weissenfels et al. (1989) also observed this reaction during fluorene degradation by Pseudomonas paucimobilis. Further metabolism of 4-hydroxyperinaphthenone and 4-phenanthroic acid may result in the formation of cinnamic and phthalic acids, however, these compounds were not detected in the supernatant extracts.

Conclusions

The Bu. cepacia strain reported in this study readily degraded pyrene to CO2. The results of the metabolite study extends the current information on the microbial metabolism of pyrene which may be useful in predicting the fate of this compound in the environment.

Acknowledgments

The research was funded by Ph.D scholarships from Australian Research Award (Industry), in conjunction with Australian Defence Industries, the Victoria Education Foundation and Centre for Bioprocessing and Food Technology, Victoria University of Technology. I wish to thank Prof. Margaret Britz and Dr Grant Stanley for their contribution to the research.

References

Cerniglia, C.E. (1984). Microbial metabolism of polycyclic aromatic hydrocarbons. Adv. Appl. Microbiol. 30:31-71.

Cerniglia, C. E. (1992). Biodegradation of polycyclic aromatic hydrocarbons. Biodegradation 3:351-368.

Cerniglia, C. E., Kelly, D. W., Freeman, J. P. and Miller, D. W. (1986). Microbial metabolism of pyrene. Chem-Biol. Interact. 57:203-216.

Cox, D. P. and Williams, L. (1980). Biological process for converting naphthalene to cis-1,2-dihydroxy-1,2,dihydronaphthalene. Appl. Environ. Microbiol. 39:320-326.

Datta, D and Samanta, T. B. (1988). Effects of inducers on metabolism of benzo[a]pyrene in vivo and in vitro: Analysis by high pressure liquid chromatography. Biochem. Biophys. Res. Comm. 155:493-502.

Guerin, W. F and Jones, G. E. (1988). Mineralisation of phenanthrene by a Mycobacterium sp.. Appl. Environ. Microbiol. 54(4):937-944.

Heitkamp, M. A. and Cerniglia, C. E. (1989). Polycyclic aromatic hydrocarbon degradation by a Mycobacterium
FLORIGENE STRIP AD
sp. in microcosms containing sediment and water from a pristine ecosystem. Appl. Environ. Microbiol. 55:1969-1973.

Heitkamp, M. A., Freeman, J. P., Miller, D. W. and Cerniglia, C. E. (1988). Pyrene degradation by a Mycobacterium sp.: Identification of ring oxidation and ring fission products. Appl. Environ. Microbiol. 54:2556-2565.

Juhasz, A. L., Britz, M. L. and Stanley, G. A. (1996). Degradation of high molecular weight polycyclic aromatic hydrocarbons by Pseudomonas cepacia. Biotech. Letters 18:577-582.

Juhasz, A. L., Britz, M. L. and Stanley, G. A. (1997). Degradation of fluoranthene, pyrene, benz[a]anthracene and dibenz[a,h]anthracene by Burkholderia cepacia. Accepted J. Appl.Microbiol.

Juhasz, A. L., Stanley, G. A. and Britz, M. L. (1995). Isolation of polycyclic aromatic hydrocarbon degrading microorganisms. Third International Symposium; In Situ and On-Site Bioreclamation, San Diego.

Kelly I., Freeman, J. P. and Cerniglia, C. E. (1990). Identification of metabolites from the degradation of naphthalene by a Mycobacterium sp.. Biodegradation 1:283-290.

Schneider, J., Grosser, R., Jayasimhulu, K., Xue, W and Warshawsky, D. (1996). Degradation of pyrene, benz[a]anthracene and benzo[a]pyrene by Mycobacterium sp. strain RJGII-135, isolated from a former coal gasification site. Appl. Environ. Microbiol. 62:13-19.

Walter, U., Beyer, M., Klein, J. and Rehm, H. J. (1991). Degradation of pyrene by Rhodococcus sp. UW1. Appl. Microbiol. Biotechnol. 34:671-676.

Weissenfels, W. D., Beyer, M. and Klein, J. (1989). Bakterieller abbau von phenanthren, fluoren und fluoranthen. Forum Mikrobiol. 104:1-2

Weissenfels, W. D., Beyer, M., Klein, J. and Rehm, H. J. (1991). Microbial metabolism of fluoranthene: Isolation and identification of ring fission products. Appl. Microbiol. Biotechnol. 34:528-535.

Wilson, S. C. and Jones, K. C. (1993). Bioremediation of soils contaminated with polynuclear aromatic hydrocarbons (PAHs): A review. Environ. Poll. 81:229-249.

Copyright 1997 Australian Biotechnology Association Ltd.


The following images related to this document are available:

Line drawing images

[au97048b.gif] [au97048a.gif]
© Bioline International, 1989 - 2004, Site last up-dated on 03-Ago-2006.