Degradation of Fluorobenzene by Rhizobiales Strain F11 via ortho Cleavage of 4-Fluorocatechol and Catechol

The aerobic metabolism of fluorobenzene by Rhizobiales sp. strainF11 was investigated. Liquid chromatography-mass spectrometryanalysis showed that 4-fluorocatechol and catechol were formedas intermediates during fluorobenzene degradation by cell suspensions.Both these compounds, unlike 3-fluorocatechol, supported growthand oxygen uptake. Cells grown on fluorobenzene contained enzymesfor the ortho pathway but not for meta ring cleavage of catechols.The results suggest that fluorobenzene is predominantly degradedvia 4-fluorocatechol with subsequent ortho cleavage and alsopartially via catechol.

INTRODUCTION

Top

Introduction

During the last decades, environmental contamination by fluorinatedorganic compounds has received increasing attention becauseof their use as herbicides, fungicides, surfactants, refrigerants,intermediates in organic synthesis, solvents, and pharmaceuticals(11). Whereas the biodegradation of chlorinated compounds hasbeen studied quite extensively (19), little is known about thebacterial metabolism of fluoroaromatic compounds, even thoughthere have been several reports on the degradation of fluorobenzoicacids (5, 6, 7, 16). With chloroaromatics, most degradationroutes involve dioxygenase- and dehydrogenase-mediated conversionto the corresponding chlorocatechols, which are further metabolizedby a dioxygenase that cleaves the aromatic ring. Dehalogenationoccurs during metabolism of the ring-cleavage products (19).Most described strains degrade chlorocatechols via the ortho-cleavagepathway (14, 18, 19, 20), but meta cleavage of 3-chlorocatecholcan also occur (13), even though the meta-cleavage route isoften unproductive due to the formation of toxic or dead-endproducts (1, 19). Dehalogenation may in some cases occur priorto ring cleavage. For example, mutants of Pseudomonas sp. strainB13 and Alcaligenes eutrophus B9 that grow on 2-fluorobenzoateuse a dioxygenase to convert it to catechol, with concomitantdecarboxylation and defluorination (5). Pseudomonas putida strainCLB 250, which can use three different 2-halobenzoates, alsoconverts these substrates by initial dehalogenating dioxygenation(6), and a defluorinating 4-fluorobenzoate monooxygenase hasbeen reported as well (16).

The present paper describes a metabolic pathway for fluorobenzene(FB). Information about the bacterial metabolism of this compoundis scarce, despite studies on its chlorinated analogue (13,18). Lynch et al. (12) described the oxidation of FB to 3-fluorocatecholby a strain of Pseudomonas putida, but in this study FB wasnot used as a carbon source. Recently, FB was reported to becompletely degraded by a bacterial consortium (2) and by a purebacterial culture that utilized it as a sole carbon and energysource (3). This gram-negative bacterium, phylogenetically classifiedwithin the order Rhizobiales, was named strain F11 and was usedhere to investigate the metabolism of FB.

Intermediates produced during FB degradation.

Top

Intermediates produced during fb...

In order to obtain information about the degradation pathwayof FB, we tested which intermediates accumulated upon incubationof concentrated cell suspensions of strain F11 with FB. First,cells were grown in sealed flasks on FB in mineral medium asdescribed previously (3), harvested by centrifugation (10,000x g for 15 min at 4°C), washed twice with mineral medium,and resuspended in the same medium to give an optical densityat 600 nm of 0.3. Glucose (1 mM) was added, since it was foundin preliminary experiments that this enhanced degradation ofFB and stimulated accumulation of intermediates. The suspensionswere incubated in closed flasks with FB, and samples were takenat appropriate times, centrifuged, and subjected to high-performanceliquid chromatography (HPLC), gas chromatography, and fluoridemeasurements. Fluoride was measured with a Dionex Dx-120 ionchromatograph equipped with an Allsep A-2 anion column fromAlltech, and the eluent was a mixture of NaHCO₃ and Na₂CO₃ indeionized water. For FB analysis, culture samples were extractedwith diethylether and analyzed by gas chromatography as describedpreviously (3). It was observed that whole cells of strain F11completely removed 1.1 mM FB in 13 h, but stoichiometric fluoriderelease was seen only after 29 h (Fig. 1). This indicates thatfluorinated intermediates did temporarily accumulate but thatthere was no formation of high levels of fluorinated dead-endmetabolites.

View larger version (10K):
[in this window]
[in a new window]
FIG. 1. Accumulation of metabolites during degradation of 1.1 mM FB by strain F11. Symbols: ${blacksquare}$ , FB (calculated as if all substrate was present in the liquid phase); ${blacktriangleup}$ , fluoride; X, 4-fluorocatechol; •, catechol; ${blacklozenge}$ , optical density at 600 nm. Liquid phase FB is in equilibrium with gas-phase substrate according to the Henry coefficient (H = C_g/C_l = 0.258). The closed flasks had a volume of 1,200 ml and contained 250 ml medium.

Samples (20 µl) from the same culture fluid were alsoanalyzed by HPLC and liquid chromatography-mass spectrometry(LC-MS). HPLC analysis was carried out on a Lichrospher 100RP8 reversed-phase column in connection with Jasco PU-980 pumps,a Jasco MD-910 diode array detector, and a Jasco UV-2075 detector.Compounds were isocratically eluted at a flow rate of 1 ml/minwith a solution of water-acetonitrile (80:20) and 10 mM formicacid. LC-MS was carried out with a Micromass ZMD detector equippedwith a XTerra MS, a SymmetryShield C8 column (4.6 mm by 150mm), a 996 photodiode array detector, and a 2690 separationsmodule, all from Waters. While degradation proceeded, five metabolites(compounds I to V) appeared in the culture medium (Table 1).Four of these (compounds I, II, IV, and V) were completely consumedduring prolonged incubation. One minor metabolite (compoundIII) remained in the culture supernatant even after incubationfor 48 h.

View this table:
[in this window]
[in a new window]
TABLE 1. HPLC retention times and absorption maxima of metabolic intermediates formed by F11 cells exposed to fluorobenzene in the presence of glucose

Metabolite I, which appeared early, coeluted with catechol andhad a molecular mass (negative-mode MS) of m/z = 109 (M-H⁺).Metabolite II was identified as 4-fluorocatechol on the basisof cochromatography with a standard in HPLC analysis and itsnegative-mode mass spectrum with m/z = 126.91 (M-H⁺). The dead-endmetabolite III was identified as cis-dienelactone by cochromatographyand mass spectrometry (positive ionization, m/z = 140.04 [M+H⁺]).Metabolites IV and V could not be identified, since no ionizationwas obtained with LC-MS.

The occurrence of the two catechols during the initial 20 hof FB degradation (Fig. 1) suggests that strain F11 convertsFB partially to catechol and partially to 4-fluorocatechol duringthe first metabolic step. Approximately 0.6 mM of the FB thatwas converted transiently appeared as 4-fluorocatechol, andabout 0.3 mM was detected as catechol. This is in agreementwith the observation that significant fluoride release is takingplace already during the initial period of FB degradation, i.e.,when catechol is formed but that fluoride release is only completewhen the intermediate 4-fluorocatechol and possibly other fluorinatedmetabolites have been degraded. The formation of both catecholand 4-fluorocatechol from FB is also consistent with the abilityof strain F11 to grow on both of these catechols. The fact thatcatechol remained in the medium for quite a long time even thoughit is a better growth substrate than 4-fluorocatechol suggestsinhibition of the catechol pathway by the presence of 4-fluorocatechol.

Substrate-dependent oxygen consumption by whole cells.

Top

Substrate-dependent oxygen...

In order to test the inducibility of FB transformation activity,oxygen uptake measurements were done. Cells of strain F11 weregrown on FB, benzene, or citrate, harvested by centrifugation,washed, resuspended to a density of 0.43 mg of cellular proteinper ml, and transferred to a stirred vessel that was equippedwith a fiber optic oxygen sensor (MOPS-1; ProSense BV, Hanover,Germany). The rate of O₂ consumption was measured at room temperaturein the presence of different substrates (Table 2).

View this table:
[in this window]
[in a new window]
TABLE 2. Substrates oxidized by fluorobenzene-, benzene-, and citrate-grown cells of strain F11^a

Cells grown on FB oxidized fluorobenzene and cis-1,2-dihydrobenzenediolas well as 4-fluorocatechol and were also highly induced forthe oxidation of catechol. Rapid oxidation of catechol, 4-fluorocatechol,and cis-1,2-dihydrobenzenediol was also obtained with cellsthat were pregrown on benzene. With fluorobenzene and benzene,these cells showed even higher oxygen uptake rates than cellsgrown on FB. With cells grown on citrate, the aromatic substrateswere not oxidized, suggesting that the formation of the firstcatabolic enzyme was induced during growth on the aromatic substratesand repressed on citrate. The oxygen uptake rates with 3-fluorocatecholwere very low, independent of whether the cells were grown onFB, benzene, or citrate. The patterns of oxygen consumptionshow that oxidation of catechol, 4-fluorocatechol, and cis-1,2-dihydrobenzenediolwas induced by FB and benzene, whereas 3-fluorocatechol wasnever oxidized. When batch cultures of strain F11 growing with1 mM of FB were supplemented with 0.1 mM of 3-fluorocatechol,FB was no longer converted by the cells, and fluoride, measuredwith a fluoride-selective electrode (2), was not released.

Enzymes involved in the degradation of FB.

Top

Enzymes involved in the...

To investigate whether degradation of FB proceeds via meta orortho cleavage, the presence of several enzymes involved inthese routes was tested (Table 3). Late exponential FB-growncells were harvested by centrifugation, washed twice with 0.1M Tris-HCl buffer (pH 7.5) containing 0.1 mM 1,4-dithiothreitol,and disrupted by sonication in the same buffer. After centrifugation(90,000 x g for 60 min at 4°C) the clear supernatant wasused as the cell extract for enzyme assays. Its protein contentwas determined with Coomassie brilliant blue using bovine serumalbumin as the standard.

View this table:
[in this window]
[in a new window]
TABLE 3. Enzyme activities in cell extracts of strain F11 grown on FB

Catechol 2,3-dioxygenase was measured by determining the formationof 2-hydroxymuconic semialdehyde ( ${varepsilon}$ = 44,000 M^–1 cm^–1)at 375 nm, according to the method of Nozaki (15). (Fluoro)catechol 1,2-dioxygenase activity was measured similarly, asdescribed by Dorn and Knackmuss (4) ( ${varepsilon}$ _cis_,cis_-muconate = 16,800M^–1, ${varepsilon}$ _{2-fluoro-cis,cis-muconate} = 14,900 M^–1 cm^–1, ${varepsilon}$ _{3-fluoro-cis,cis-muconate} = 14,900 M^–1 cm^–1). Muconatecycloisomerase activity was measured by following the consumptionof cis,cis-muconate in an assay mixture containing 30 mM Tris-HCl(pH 8.0), 1 mM MnCl₂, and 0.1 mM cis,cis-muconate. Dienelactonehydrolase activity was determined by following at 280 nm ( ${varepsilon}$ =17,000 M^–1 cm^–1) the decrease in the level of 0.1mM cis-dienelactone that was incubated with enzyme in 10 mMhistidine-HCl (pH 6.5). Maleylacetate was prepared on the dayof its use by alkaline hydrolysis of cis-dienelactone (8), andthe reductase was measured by following maleylacetate (0.1 mM)-dependentNADH (0.2 mM) oxidation at 340 nm in 50 mM Tris-HCl (pH 7.5).Activities were corrected for substrate-independent NADH oxidation.3-Oxoadipate:succinyl-coenzyme A transferase was measured asdescribed by Mars et al. (13). One unity of activity was definedas the amount of enzyme required to convert 1 µmol ofsubstrate per min at 25°C.

Activities of the ortho pathway enzymes catechol 1,2-dioxygenase,muconate cycloisomerase, maleylacetate reductase, and 3-oxoadipate:succinyl-coenzymeA transferase were found in cell extracts of strain F11 grownon FB. Catechol 2,3-dioxygenase activity was not detected, indicatingthat strain F11 does not use a meta-cleavage pathway to degradeFB. A 1,2-dioxygenase activity was detected with both catecholand 4-fluorocatechol but was hardly detected with 3-fluorocatechol.Instead, the catechol 1,2-dioxygenase activity was reduced by70% and 90% in the presence of 0.1 and 0.5 mM 3-fluorocatechol,respectively. A muconate cycloisomerase activity with respectto cis,cis-muconic acid was also detected. No activity was foundfor cis-dienelactone hydrolase in extracts of strain F11. Theseobserved enzyme activities suggest that the catechols in theFB degradation pathway undergo ortho cleavage.

We judge it highly unlikely that 3-fluorocatechol is an intermediatein FB degradation pathway, since it was not used as a growthsubstrate and it strongly inhibited FB degradation and growthand since no dioxygenase activity with 3-fluorocatechol wasdetected in cell extracts of strain F11 grown on FB. This conclusionis in line with the previously described resistance of the expectedproduct 2-fluoro-cis,cis-muconic acid to enzymatic cycloisomerization(23), although it is risky to generalize such a finding to otherorganisms. Furthermore, in mutants of Alcaligenes eutrophusB9 and Pseudomonas sp. strain B13 that use 2-fluorobenzoatefor growth, the formation of toxic 3-fluorocatechol is preventedby loss of dihydrodihydroxybenzoate dehydrogenase activity,allowing growth on catechol that can be formed by initial dioxygenationof the aromatic ring (5). The absence of a 3-halocatechol intermediateclearly distinguishes the fluorobenzene pathway of strain F11from chlorobenzene catabolic pathways, which proceed via 3-chlorocatechol(13, 17, 18). This could explain the lack of growth of strainF11 on chlorobenzene.

Pathway of FB degradation.

Top

Pathway of fb degradation.

We propose the pathway for FB metabolism shown in Fig. 2. Theinitial attack of FB by dioxygenase activity yields two differentfluorinated dihydrodiols. The 4-fluoro-cis-benzene-1,2-dihydrodiolthat is produced is transformed into 4-fluorocatechol by a dihydrodioldehydrogenase. Conversion of the other product, 1-fluoro-cis-benzene-1,2-dihydrodiol,to catechol can proceed without involvement of dehydrogenasethat reduces a cofactor, since the electrons are transferredto the fluoride that is being released (Fig. 2). Simultaneousconversion of a fluorinated compound to catechol and a fluorinatedcatechol was described earlier for the degradation of 2-fluorobenzoateby Pseudomonas sp. strain B13 and strain FLB300 (5, 7).

View larger version (7K):
[in this window]
[in a new window]
FIG. 2. Proposed pathway for fluorobenzene metabolism by strain F11. The enzyme activities are denoted as follows: 1, fluorobenzene dioxygenase; 2, fluorobenzene dihydrodiol dehydrogenase; 3, fluorocatechol 1,2-dioxygenase; 4, fluoromuconate cycloisomerase; 5 and 6, possible side reactions to cis-dienelactone by fluoromuconate cycloisomerase (activity 5) or by slow spontaneous conversion (activity 6); 7, trans-dienelactone hydrolase; 8, maleylacetate reductase; 9, fluorobenzene dioxygenase; 10, nonenzymatic defluorination; 11, catechol 1,2-dioxygenase; 12, muconate cycloisomerase; 13, muconolactone isomerase; 14, 3-oxoadipate enol-lactone hydrolase.

The capability of strain F11 cells to use 4-fluorocatechol asa growth substrate, its transient accumulation in cell suspensionsto higher levels than catechol, and the fact that 4-fluorocatecholstimulated oxygen uptake by whole cells lead to the conclusionthat 4-fluorocatechol is the predominant intermediate. The occurrenceof 4-fluorocatechol as an intermediate has also been describedfor the aforementioned ${alpha}$ -proteobacterium strain FLB300, whichdegrades both benzoate and all monofluorosubstituted benzoates(7). The metabolism of 4-fluorocatechol is proposed to proceedthrough ortho cleavage by a (fluoro)catechol 1,2-dioxygenasethat yields 3-fluoro-cis,cis-muconate. This product could betransformed with concomitant defluorination into maleylacetatevia either 4-fluoromuconolactone or another lactone derivative.ortho cleavage of 4-fluorocatechol is also a key step in themetabolism of 3- and 4-fluorobenzoate by several bacterial strains(9, 21, 22, 24). These convert 4-fluorocatechol via 3-fluoro-cis,cis-muconateand 4-fluoromuconolactone (22, 24) or via a non-fluorinateddienelactone intermediate (24). If 4-fluorocatechol metabolismin strain F11 proceeds in the same way, this could also explainthe accumulation of cis-dienelactone since it can slowly beformed as a side product during spontaneous or enzyme-catalyzeddehydrodefluorination of 4-fluoromuconolactone (23, 24). Thispathway would also allow formation of protoanemonin.

Maleylacetate can be channeled into the tricarboxylic acid cyclevia 3-oxoadipate. Catechol, the minor product of the initialdioxygenation reaction, is proposed to be metabolized to cis,cis-muconate,converted to the lactone derivative, and then also channeledinto the 3-oxoadipate route (Fig. 2).

ACKNOWLEDGMENTS

M.F.C. acknowledges a research grant from Fundaçãopara a Ciência e Tecnologia (FCT), Portugal (BD/21839/99),and Fundo Social Europeu (III Quadro Comunitário de Apoio).This work was supported in part by the European Community'sHuman Potential Programme under contract HPRTN-CT-2002-00213[BIOSAP] and by a grant from the European Science Foundation.

We thank Paolo De Marco for help in the preparation of cellextracts and Filip Kaminski for valuable discussions and a kindgift of protoanemonin. Both 4-fluorocatechol and cis-dienelactonewere generous gifts of W. Reineke (Bergische Universität,Wuppertal, Germany). We thank Theodora Tiemersma for help withthe LC-MS analysis.

FOOTNOTES

* Corresponding author. Mailing address: Escola Superior de Biotecnologia, Universidade Católica Portuguesa, Rua Dr. António Bernardino de Almeida, 4200-072 Porto, Portugal. Phone: 351 225580059. Fax: 351 225090351. E-mail: plc{at}esb.ucp.pt.

Published ahead of print on 15 September 2006.

M.F.C. and M.I.M.F. contributed equally to this work.

REFERENCES

Top