APPLIED AND ENVIRONMENTAL MICROBIOLOGY, Feb. 1984, p. 395-4020099-2240/84/020395-08$02.00/0Copyright C 1984, American Society for Microbiology

Vol. 47, No. 2

Microbial Metabolism of Haloaromatics: Isolation and Properties ofa Chlorobenzene-Degrading BacteriumWALTER REINEKElt* AND HANS-JOACHIM KNACKMUSS2

Gesellschaft fur Strahlen- und Umweltforschung mbH, Munich,1 and Lehrstuhl fur Chemische Mikrobiologie derUniversitat-Gesamthochschule-Wuppertal, D-5600 Wuppertal 1,2 Federal Republic of Germany

Received 23 June 1983/Accepted 17 October 1983

A chlorobenzene-degrading bacterium was isolated by continuous enrichment from a mixture of soil andsewage samples. This organism, strain WR1306, was grown in a chemostat on a mineral medium withchlorobenzene being supplied through the vapor phase with a critical DC value at a dilution rate of 0.55 h-1.Maximum growth rates in batch culture were accomplished at substrate concentrations of -0.5 mM in theculture medium. During growth on chlorobenzene, stoichiometric amounts of chloride were released.Respiration data and enzyme activities in cell extracts as well as the isolation of 3-chlorocatechol from theculture fluid are consistent with the degradation of chlorobenzene via 3-chloro-cis-1,2-dihydroxycyclohexa-3,5-diene, 3-chlorocatechol, 2-chloro-cis,cis-muconate, trans-4-carboxymethylenebut-2-en-4-olide, maleyl-acetate, and 3-oxoadipate.

Of the numerous chemical substances that enter theenvironment with wastewater and exhaust, a great numberare benzene derivatives and other nonpolar aromatics. Themicrobial oxidation of the parent compound, benzene, wasfirst reported by Wagner (52). Since that time severalmicroorganisms have been described which are able to usebenzene as the sole carbon source. The first information onthe mechanism of the initial reactions of benzene degrada-tion was obtained from sequential induction experimentsindicating catechol as an intermediate (37, 38, 53). Furtherevidence for catechol as the ring fission substrate in thepathway of benzene degradation was supplied by Marr andStone (38), who detected this intermediate chromatographi-cally in the culture medium of benzene-grown Pseudomonasaeruginosa. The initial reactions of benzene oxidation bybacteria have now been clarified by the thorough studies ofGibson and his colleagues. They established that cis-1,2-dihydroxycyclohexa-3,5-diene is the initial metabolite ofbenzene degradation by Pseudomonas putida which is rearo-matized, yielding catechol (23, 26). The same sequence ofreactions has also been shown to occur in a species ofMoraxella (31). Further breakdown of catechol has beenreported to proceed via the meta- (11, 22, 50) or ortho-cleavage pathway (1, 31, 37, 38).However, details of halobenzene degradation by microor-

ganisms are scarce. Erikson observed some type of growthof Micromonospora strains at the expense of 1,4-dichloro-benzene during a 6-week incubation (16). More recently,Bacillus polymyxa has been reported to grow with bromo-benzene (48). Haider et al. described benzene-grown micro-organisms that readily cooxidize chlorobenzene (29). How-ever, little is known about the reactions involved in thedegradation of chlorobenzene. The present paper describesthe isolation, mode of cultivation, and metabolic propertiesof a chlorobenzene-degrading bacterium.

MATERIALS AND METHODSIsolation, maintenance, and culture of microorganism. Theorganism strain WR1306 described in this study was isolated

* Corresponding author.t Present address: Universitat-Gesamthochschule-Wuppertal,

Fachbereich 9, D-5600 Wuppertal 1, Federal Republic of Germany.

from an inoculum derived from a mixture of many differentsoil and sewage samples, which were collected in the Got-tingen area in the Federal Republic of Germany. Benzenewas used as the substrate of enrichment and was graduallyreplaced by chlorobenzene. For enrichment, a 200-ml che-mostat (dilution rate, 50.0075 h-1) was used as describedpreviously by Hartmann et al. (30). The continuous culturewas initially supplied with benzene through the vapor phase(5500 ppm [s500 mg/ml] by vol at 20°C, 0.6 liters/h). Thishydrocarbon was stepwise replaced by chlorobenzene (5200ppm, 0.6 liters/h). As soon as chlorobenzene was used as thesole growth substrate, cells were plated on nutrient agar.Individual colonies were picked and checked for purity. Theisolates were maintained on agar plates and grown routinelyat 30°C on a mineral medium described by Dom et al. (13) byincubation in a desiccator. Chlorobenzene was supplied asthe carbon source through the vapor phase (3 ,ul of liquidchlorobenzene per liter of gas volume in the desiccator)without direct contact of the hydrocarbon to the medium.

Small quantities of cells were grown at 30°C in 500- or3,000-ml sealed Erlenmeyer flasks with baffles containing 30or 250 ml of growth medium by shaking on a rotary shaker at150 rpm. Chlorobenzene was introduced with a syringe intoa side arm equipped with a septum, which allowed thehydrocarbon to evaporate. Large-scale growth of biomasswas carried out in a 2-liter fermentor (500 rpm; Multigen F2000 from New Brunswick Scientific Co., Inc., Edison,N.J., modified with a Waldhof system) containing 1 liter ofmedium. For continuous cultivation, fresh medium contain-ing antifoaming agent P 2000 (0.05 ml/liter) was fed by aperistaltic pump (Multiperpex 2115; LKB Instruments Inc.,Bromma, Sweden). Chlorobenzene was supplied to thefermentor with the incoming air at a rate of 0.25 liter/min.Chlorobenzene was admixed to the vapor phase at a con-trolled rate by passing air through a container of liquidchlorobenzene and via a bypath. The gas flow was measuredand adjusted with flowmeters (Fisher and Porter, Gottingen,Federal Republic of Germany). The chlorobenzene concen-tration was determined by monitoring its presence in theinflow gas spectrophotometrically at 210 nm with a 100-mmlight path. An extinction of 0.03 corresponds to 10 ppm ofchlorobenzene. The calibration curve was obtained afterevaporation of amounts of liquid chlorobenzene, measured

395

396 REINEKE AND KNACKMUSS

in microliters, in a sealed cuvette with a 100-mm light path.For cooxidation of toluene, the hydrocarbon was intro-

duced at a controlled rate by the same equipment.Measurement of growth. Growth of the cultures was

monitored turbidimetrically at 546 nm with a Zeiss PM4spectrophotometer (Oberkochen, Federal Republic of Ger-many) or a Klett-Summerson colorimeter (New York)equipped with a 520- to 580-nm filter.

Preparation of washed suspensions of bacteria. Organismswere harvested by centrifugation (30,000 x g for 20 min at28°C), washed twice in 100 mM K-phosphate buffer (pH 7.5),and resuspended in the same buffer.

Preparation of extracts. Washed suspensions were disrupt-ed by one passage through a French pressure cell (Aminco,Md.; 140 MPa, 0°C) before centrifugation (100,000 x g for 60min at 4°C) to remove cell debris. The clear supernatantsolution was used as a source of crude cell extract.

Protein estimation. The protein content of extracts wasdetermined by the method of Bradford (7), using bovineserum albumin as the standard.Measurement of oxygen uptake. Oxygen consumption was

measured at 30°C by conventional Warburg manometry.Incubation mixtures contained (in 2.2 ml): 200 ,umol of K-phosphate buffer (pH 7.5), 5 ,umol of substrate, and a portionof cell suspension. The substrates were added as a solutionin 0.2 ml of solvent (N,N-dimethylformamide plus 10%water [vol/vol]). Oxygen uptake rates were corrected forendogenous respiration of the cell suspension. Whole cellswere unable to oxidize N,N-dimethylformamide, nor did thissolvent inhibit enzymatic activity.Enzyme assays. All enzyme assays were done at 25°C. cis-

1,2-Dihydroxycyclohexa-3,5-diene (NAD+) oxidoreductasewas measured by determining NAD reduction at 340 nm.

Reaction mixtures contained (in 1 ml): 150 ,umol of K-phosphate buffer (pH 7.5), 2 ,umol of NAD, 1.25 ,umol of 3-chloro-cis-1,2-dihydroxycyclohexa-3,5-diene, and extract(0.01 to 0.5 mg of protein).Catechol 1,2-dioxygenase was measured as described pre-

viously by Dorn and Knackmuss (14). Reaction mixturescontained (in 3 ml): 100 ,umol of Tris-hydrochloride buffer(pH 8.0), 1 ,umol of catechol or the chloro-substitutedanalogs, 4 ,umol of EDTA, and extract (0.1 to 2 mg ofprotein).

Catechol 2,3-dioxygenase was measured by determiningproduct formation at 375 nm (41). Reaction mixtures con-

tained (in 3 ml): 150 ,umol of K-phosphate buffer (pH 7.5), 1,umol of catechol, and extract (0.1 to 2 mg of protein), whichhad been treated at 55°C for 10 min.

4-Carboxymethylenebut-2-en-4-olide hydrolase was mea-

sured at 280 nm by determining the decrease of substrateconcentration (46). Reaction mixtures contained (in 1 ml): 50,umol of K-phosphate buffer (pH 6.5), 1 ,umol of cis- or trans-

4-carboxymethylenebut-2-en-4-olide (E280 = 17,000 or 15,625liters mol-1 cm-', respectively), and extract (0.01 to 0.5 mgof protein).Maleylacetate reductase was measured by determining the

maleylacetate-dependent NADH oxidation at 340 nm. Reac-tion mixtures contained (in 1 ml): 50 ,umol of K-phosphatebuffer (pH 7.5), 0.2 ,umol of NADH; 3 ,umol of maleylace-tate, and extract (0.01 to 0.5 mg of protein).

Analytical methods. Metabolites in the culture fluid were

detected by high-pressure liquid chromatography on a re-

verse-phase column as described by Hartmann et al. (30).Metabolites were identified by comparison of retention timewith authentic samples and by in situ scanning of the UVspectra after the flow had been stopped. Samples of culture

fluids (5 to 20 ,ul) were injected after cells had been removedby centrifugation. For the detection of low concentrations ofmetabolites the cell-free medium was extracted with anexcess of diethyl ether. After evaporation of the solvent theresidue was dissolved in water before injection.The concentration of the chloride ions in the medium was

measured with an ion-selective combination chloride elec-trode (model 16/17; Orion Research, Inc., Cambridge,Mass.), which was calibrated with NaCl (0.1 to 50 mM) inmineral medium.

Chemicals. 4-Chlorocatechol was prepared by the methodof Willstatter and Muller (54), and 3-chloro-, 3-fluoro-, and 4-fluorocatechol were prepared as described by Schreiber etal. (47). 3,5-Dichlorocatechol was synthesized by chlorina-tion of 2-hydroxybenzaldehyde by the method of Biltz andStepf (5) and subsequent Dakin reaction (12). 3,4-Dichloro-,3,6-dichloro-, and 5-chloro-3-methylcatechol were a gener-ous gift from Juha Knuutinen, University of Jyvaskyla,Finland. All catechols were purified by vacuum sublimationbefore use. 2-Methyl- and 4-methyl-2,5-dihydro-5-oxo-furan-2-acetic acid were prepared as described by Hartmann et al.(30). 3-Chloro-cis-1,2-dihydroxycyclohexa-3,5-diene waskindly supplied by David T. Gibson, University of Texas atAustin. Chlorobenzene was obtained from Fluka AG,Buchs, Switzerland, whereas bromo-, 1,2-dichloro-, 1,3-dichloro-, and 1,4-dichlorobenzene were purchased fromMerck-Schuchard, Hohenbrunn, Germany. Biochemicalswere obtained from Boehringer Mannheim Biochemicals,Mannheim, Germany. All other materials were of the highestpurity commercially available and were used without furtherpurification.The isomeric cis- and trans-4-carboxymethylenebut-2-en-

4-olides were prepared as follows, using a cell extract of 3-chlorobenzoate-grown Pseudomonas sp. strain B13. A stocksolution of 4-chlorocatechol (50 mM [pH 6.5], supplementedwith 35 mM 2-mercaptoethanol, kept at 0°C) was pumped toa stirred solution of cell extract in 50 mM K-phosphate buffer(pH 7.5) plus 12.5 mM EDTA which was incubated at 30°C.The pH was kept constant at 7.5 by the automatic addition of0.1 M NaOH. The turnover of 4-chlorocatechol was moni-tored by high-pressure liquid chromatography. The pumpingrate was adjusted in such a manner that 4-chlorocatecholwas not detectable in the solution during the entire course ofthe reaction. The resulting 3-chloro-cis,cis-muconic acid wascyclo-isomerized overnight at 25°C by acidification of thesolution to pH 3 by the addition of H3PO4. The butenolideswere extracted repeatedly with diethyl ether after the solu-tion had been both acidified to pH 2 and centrifuged forprecipitation of the protein. The products were purified byuse of preparative thin-layer chromatography plates, pre-coated with silica gel 60 and the solvent system diisopropylether-formic acid-water (200:7:3 [vol/vol/vol]). Bands werelocated by the use of UV light and extracted with ethylacetate. The trans isomer was obtained as a pure compoundby crystallization from ethyl acetate. Crystallization fromdichloromethane yielded the pure cis isomer. Maleylacetatewas prepared by hydrolysis of cis-4-carboxymethylenebut-2-en-4-olide at pH 11 at room temperature (18). During hydrol-ysis the absorption peak at 277 nm was shifted to a smallerpeak at 243 nm.

RESULTSEnrichment and isolation. A chlorobenzene-degrading bac-

terium was obtained by continuous enrichment from abenzene-degrading population originating from a mixture ofsoil and sewage samples from the Gottingen area, Federal

APPL. ENVIRON. MICROBIOL.

MICROBIAL METABOLISM OF HALOAROMATICS 397

C,)

c4-

4-(1

200150

100

50

0.5

_0

1.3 3.9 5.2

0 2 4i [h]

FIG. 1. Effect of chlorobenzene concentration on the lag time in batch culture. Chlorobenzene-grown cells of strain WR1306 wereinoculated into 25 ml of mineral medium in sealed 500-ml Erlenmeyer flasks. Various amounts of chlorobenzene were added to a side arm by asyringe which allowed the hydrocarbon to evaporate. Concentrations of chlorobenzene (mM) were calculated as if the substrate was totallyabsorbed by the medium.

Republic of Germany. Benzene was supplied to the culturewith the incoming air loaded with variable concentrations ofthe hydrocarbon. Within 2 weeks the concentration ofbenzene was increased stepwise to 500 ppm in the vaporphase, after which chlorobenzene was added to the culture.Stepwise-increasing amounts of chlorobenzene (to 200 ppmin the incoming air) were introduced into the culture over aperiod of 9 months. Simultaneously, the concentration ofbenzene in the vapor phase was gradually decreased (to 50ppm). A massive appearance of eucaryotes which periodical-ly eliminated most of the bacteria from the culture wassuppressed by repeated doses of cycloheximide (10 mg/liter).After ca. 9 months of operation a culture was obtained whichwas able to grow in batch culture with chlorobenzene as thesole carbon source (as described above, sealed Erlenmeyerflasks with side arms were used which allowed the additionof portions of liquid chlorobenzene [2 ,ul/100 ml of mineralmedium]). The culture was subcultured several times inbatch culture with chlorobenzene as the sole carbon source.By streaking the culture on nutrient agar to obtain singlecolonies within 2 days, a pure culture was obtained whichwas able to utilize chlorobenzene as the growth substrate.Growth on chlorobenzene. Several studies have confirmed

that the cell envelope of gram-negative bacteria is damagedby toluene (8, 33, 55). Gibson et al. (23) found that a P.putida strain, although able to use benzene, toluene, andethylbenzene as growth substrates, would not grow on thehydrocarbons when these compounds were added directly tothe culture medium. However, when toluene or ethylben-zene was introduced into the vapor phase rapid growth wasobserved. Benzene under the same conditions did not allowgrowth (23).

In the present investigation, the addition of variousamounts of liquid chlorobenzene (0.1- to 0.45 g/liter) directly

to the culture medium did not result in the growth of strainWR1306. Correspondingly, when this organism was inocu-lated into a liquid culture with a vapor phase saturated withchlorobenzene (11,900 ppm, by vol at 20°C [35]) growth wasnot observed. However, chlorobenzene allowed growthwhen supplied into the vapor phase at a concentration of 660ppm (3 ,ul of liquid chlorobenzene per liter of gas volume inthe desiccator). But even under these conditions the strainwas highly sensitive toward higher concentrations of chloro-benzene (Fig. 1). With increasing amounts of chlorobenzenea prolonged lag phase was observed. Correspondingly, asudden increase in substrate concentration (change in sub-strate concentration, >0.5 mM) severely disturbed exponen-tial growth with chlorobenzene (Fig. 2). However, a linearrelationship between the chlorobenzene added and the in-crease in turbidity was obtained when the carbon source wasintroduced in small portions (change in substrate concentra-tion, c0.5 mM) into the vapor phase of the cultures (Fig. 3).The data in Fig. 4 indicates total degradation of the chlorinat-ed hydrocarbon because growth at the expense of chloroben-zene was accompanied by the elimination of stoichiometricamounts of chloride.By use of a chemostat (schematic diagram of the single-

state chemostat, Fig. 5), undisturbed growth of the strainwith chlorobenzene as the sole source of carbon and energywas accomplished. Figure 6 shows the steady state relation-ship in the continuous culture. The critical value DC wasreached at a dilution rate of 0.55 h-1, which corresponds to adoubling time of 1.26 h.

Oxidation of aromatic hydrocarbons by whole cells. Thewashed cell suspension of strain WR1306, grown with chlo-robenzene as the source of carbon, oxidized benzene andchlorobenzene at approximately equal rates. Bromobenzenewas oxidized at approximately half the rate observed with

200 r

cn

4-4-

1501-1001-

50

0.5

~~~/-~~1.3

,i,5.23.3

f

0 2 4 [h]FIG. 2. Impaired growth of strain WR1306 in batch culture during discontinuous addition of chlorobenzene. Chlorobenzene-grown cells of

strain WR1306 were inoculated into 25 ml of mineral medium in sealed 500-ml Erlenmeyer flasks. The substrate was added to a side arm andcorresponded to a concentration of 2.5 mM (calculated as if the substrate was totally absorbed by the medium). After growth for 4 h an

additional portion of chlorobenzene was added as indicated (concentration in mM).

VOL. 47, 1984

2.6f 0 -

398 REINEKE AND KNACKMUSS

60 -

40a) 4

20 -

I I I ,1 2 3 4 5

chlorobenzene [mM]FIG. 3. Increase in turbidity versus chlorobenzene consumption

in batch culture. Strain WR1306 was grown in five separate culturesin 25 ml of mineral medium in sealed 500-ml Erlenmeyer flasks.Liquid chlorobenzene was added to a side arm in small portions overa period of 1 week. The concentration of the substrate was calculat-ed as if it was being totally absorbed by the medium.

benzene, whereas toluene (methylbenzene) was only slowlyoxidized (Table 1). The cells failed to oxidize fluorobenzene,the isomeric dichlorobenzenes, the isomeric xylenes (di-methylbenzenes), and 4-fluorotoluene at measurable rates.When cell suspensions were incubated with phenol or mono-chlorophenols, which have been postulated to be intermedi-ates of benzene or chlorobenzene metabolism (4), no oxygenuptake at the expense of these compounds was measured.

Catabolic enzyme activities in cell extracts. Enzyme activi-ties of cis-1,2-dihydroxycyclohexa-3,5-diene (NAD+) oxido-reductase, catechol 1,2-dioxygenase, muconate cycloiso-

8

rY6.E.a)V 40-C0 2

FIG. 5. Schematic diagram of the single-stage chemostat forgrowth with chlorobenzene supplied via the vapor phase. Thesubstrate was supplied with the incoming air. Chlorobenzene wasadmixed to the vapor phase at a controlled rate by passing airthrough a container of liquid chlorobenzene (1) and via a bypath (acontainer of water) (2). The flow rate of gas was adjusted byflowmeters (3 and 4). The concentration of the chlorobenzene in theincoming (5) and outgoing (6) air was monitored spectrophotometri-cally at 210 nm. UV cuvettes with a 10-cm light path were used. Theconcentration of chlorobenzene in the incoming air was adjusted inproportion to the dilution rate such that the supply of fresh medium(peristaltic pumps [7 and 8]) always corresponded to the feed ofchlorobenzene.

merase, 4-carboxymethylenebut-2-en-4-olide hydrolase, andNADH-dependent maleylacetate reductase were induced incells of strain WR1306 when grown on chlorobenzene (Table2). However, catechol 2,3-dioxygenase activity was notdetectable. In contrast to Pseudomonas sp. strain B13 whichinduced a pyrocatechase I- and pyrocatechase II-type en-

0.8

cs(0t 0.6Lo

0.4.

' 0.2

2 4 6 8chlorobenzene [mM]

FIG. 4. Chlorobenzene consumed and chloride released duringgrowth of strain WR1306. Strain WR1306 was grown in five separatecultures in 200 ml of mineral medium in sealed 3-liter Erlenmeyerflasks. Liquid chlorobenzene was added to a side arm in portionsover a period of 1 week.

Ea

3000 £Qa)a)

2000 Na)

1000 0.0

dilution rate [h-1]FIG. 6. Steady state relationship in a continuous culture of strain

WR1306 with chlorobenzene as the substrate. The steady statevalues of turbidity (U) and substrate concentration in the incoming(0) and outgoing (0) air at different dilution rates were obtainedafter six exchanges of the total volume after each change of thedilution rate. The substrate in the incoming air (20 liters h-') waschanged in proportion to the dilution rate. For additional data seethe legend to Fig. 5.

APPL. ENVIRON. MICROBIOL.

MICROBIAL METABOLISM OF HALOAROMATICS 399

TABLE 1. Relative oxygen uptake at the expense of variousbenzenes and phenols by washed cell suspensions of

chlorobenzene-grown strain WR1306'Substrate Relative rate of oxygen uptakeb

Benzene ........................... 100Chlorobenzene ..................... 119Bromobenzene ..................... 45Fluorobenzene ..................... <2Toluene............................ 18Isomeric dichlorobenzenes ........... <2Isomeric xylenes.................... <24-Fluorotoluene..................... <2Phenol............................. <2Isomeric chlorophenols ....... ....... <2

a Strain WR1306 was grown in a chemostat (D value = 0.33-h1;20 liters of air h-1 loaded with chlorobenzene [1,750 ppm]). Cellswere washed with 100 mM K-phosphate buffer (pH 7.5) andresuspended so that the cell density corresponded to an opticaldensity at 546 nm of 12.

b Values for oxygen uptake are expressed relative to that forbenzene taken as 100%. With benzene as the substrate, 100 ,u1 of 02was consumed per min by 2 ml of the cell suspension.

zyme for ring fission of catechol and chlorocatechols whengrown on 3-chlorobenzoate (14, 44), strain WR1306 inducedonly the pyrocatechase II type of enzyme when grown onchlorobenzene. Accordingly, only one pyrocatechase peakwas eluted from a DEAE-cellulose column (W. Reineke,unpublished data). The enzyme resembles pyrocatechase IIfrom Pseudomonas sp. strain B13 with respect to its highspecific and relative activities with substituted substrates(Table 3).

Isolation of metabolites. When chlorobenzene was continu-ously fed to a batch culture of strain WR1306 by theincoming air, an almost linear growth was obtained. At ahigher cell density corresponding to an E546 of 3 to 5, anincreasing light-blue to violet coloration of the medium wasobserved which progressively changed to brown and finallyto black. From the slightly colored culture a metabolitecould be extracted through diethyl ether without acidifica-tion (pH 7). By use of high-pressure liquid chromatographyand an authentic compound, the metabolite was identified as3-chlorocatechol. The isomeric 4-chlorocatechol was notpresent.

When chlorobenzene-grown cells cooxidized toluene, thedead-end metabolite 2,5-dihydro-4-methyl-5-oxo-furan-2-acetic acid was accumulated in the culture fluid. This was

identified as an authentic sample by comparison of theretention time during high-pressure liquid chromatographyand by in situ scanning of the UV spectrum. The isomeric2,5-dihydro-2-methyl-5-oxo-furan-2-acetic acid which wouldresult from toluene dioxygenation in the 3,4-position with 4-methylcatechol as an intermediate was not detectable. How-ever, in a control experiment chlorobenzene-grown cellsreadily converted not only 3-methylcatechol to 4-methyl-2,5-dihydro-5-oxo-furan-2-acetic acid but also 4-methyl-catecholto the 2-methyl-substituted 2,5-dihydro-5-oxo-furan-2-aceticacid.

DISCUSSIONIn general, degradation of nonphenolic hydrocarbons by

bacteria is initiated by double hydroxylation (22, 45). Theresulting cis-dihydrodiols are subject to dehydrogenation toyield catechols which are subsequently oxygenated to ring-cleavage products. Studies with purified preparations ofbenzene dioxygenase (2) and cis-1,2-dihydroxycyclohexa-3,5-diene (NAD+) oxidoreductase (1) as well as the isolationof metabolites with mutants (26) clearly demonstrated thatthe initial metabolites in benzene degradation by bacteria arecis-1,2-dihydroxycyclohexa-3,5-diene and catechol.Analogous initial reactions for the degradation of chloro-

benzene via 3-chloro-cis-1,2-dihydroxycyclohexa-3,5-dieneand 3-chlorocatechol must be proposed for strain WR1306.This is based on the detection of cis-1,2-dihydroxycyclo-hexa-3,5-diene (NAD+) oxidoreductase activity in cell ex-

tracts of chlorobenzene-grown cells and the isolation andidentification of 3-chlorocatechol in the culture fluid of a

chlorobenzene-utilizing population.From these data it can be deduced that an alternative

pathway via chlorophenol by the action of a benzene mono-

oxygenase, as postulated by Ballschmiter et al. (4), was notused by strain WR1306. This corresponds to the observationthat chlorobenzene-grown cells did not oxidize phenol or

chlorophenols. Ballschmiter et al. (4) have identified chloro-phenols as bacterial degradation products of chlorobenzenesby use of gas chromatographic-mass spectrometric analysis.However, these products can readily be interpreted as

artifacts derived from the cis-1,2-dihydroxycyclohexa-3,5-

TABLE 2. Specific activities of catabolic enzymes in cell extracts of strain WR1306Sp act (U/mg of protein)

Activity Assay substrateChlorobenzenea Acetateb

cis-1,2-Dihydroxycyclohexa-3,5-diene dehydrogenase 3-Chloro-cis-1,2-dihydroxycyclohexa-3,5-diene 0.12 <0.01

Catechol 1,2-dioxygenase Catechol 0.06 <0.013-Chlorocatechol 0.1 <0.01

Catechol 2,3-dioxygenase Catechol <0.002 <0.002

Muconate cycloisomerase cis,cis-Muconate 0.06 NDC2-Chloro-cis,cis-muconate 0.07 ND

4-Carboxymethylenebut-2-en-4-olide hydrolase trans-4-Carboxymethylenebut-2-en-4-olide 0.62 <0.005

Maleylacetate reductase Maleylacetate 0.94 <0.005

a Chlorobenzene-grown cells were obtained from chemostat cultures (D value = 0.33 h-1; 20 liters of air h-1; concentration ofchlorobenzene in the incoming air, 1,750 ppm).

b Cells were grown overnight in batch culture with 20 mM acetate.c ND, Not done.

VOL. 47, 1984

400 REINEKE AND KNACKMUSS

TABLE 3. Relative activity of catechol 1,2-dioxygenase fromchlorobenzene-grown cells of strain WR1306 with various

halogenated catecholsSubstrate Relative activity (%)a

Catechol ................................... 1003-Chlorocatechol............................ 1704-Chlorocatechol............................ 1223,4-Dichlorocatechol ........................ 103,5-Dichlorocatechol ........................ 1353,6-Dichlorocatechol ........................ 475-Chloro-3-methylcatechol ................... 4403-Fluorocatechol............................ 414-Fluorocatechol............................ 178

a Values are expressed as percentages of that for catechol takenas 100%. Specific activities are given in Table 2.

dienes as the initial metabolites from arene double hydroxyl-ation. It is well known that cis-dihydrodiols readily rearoma-tize with formation of phenols under mild acid conditions(28).

Monosubstituted benzenes, independent from the type ofsubstituent (chloro, methyl, ethyl, or other alkyl groups),were found to be exclusively dioxygenated in the 2,3 posi-tion yielding 3-substituted catechols as reaction products ofthe respective cis-dihydrodiols (3, 9, 24, 25, 27, 34). Thesame specificity in dioxygenation of the benzene dioxygen-ase was found in strain WR1306, since 3-chlorocatechol wasdetected as the exclusive chlorocatechol in culture filtrates.This supposition was confirmed by demonstrating 2,5-dihy-dro-4-methyl-5-oxo-furan-2-acetic acid to be the exclusivedead-end metabolite during cooxidation of the structurallyanalogous toluene.The entire degradative pathway of chlorobenzene in strain

WR1306 proposed on the basis of the enzyme activitiesfound is presented in Fig. 7. 3-Chlorocatechol is subject toortho cleavage with formation of 2-chloro-cis,cis-muconicacid. This is cycloisomerized with coincident or subsequentelimination of chloride yielding 4-carboxymethylenebut-2-en-4-olide, which is further converted by use of a hydrolase.The resulting maleylacetate is reduced in an NADH-depen-dent reaction to 3-oxoadipate. Such an enzyme was shownto participate in the metabolism of 3-chlorobenzoate inPseudomonas sp. strain B13 (Reineke, unpublished data), ofresorcinol in P. putida (10) and Trichosporon cutaneum (19),and of L-tyrosine in T. cutaneum (49).Whereas pure cultures able to utilize benzene or toluene

or both can readily be isolated from soil and sewage samples(39, 51), the isolation of chlorobenzene-degrading strainsseems to be rather difficult (32). One explanation of thedifficulty in isolating chlorobenzene-grown bacteria might bethat ring cleavage is a crucial reaction in the metabolism ofhaloaromatics. This is documented by the accumulation ofhalocatechols when arene-degrading strains act on the halo-substituted analogs (43). ortho-Pyrocatechases from ordi-nary aromatic-degrading organisms are rather inefficient forring cleavage of halocatechols (15). For 3-halocatechols,relative activity is -1% in comparison to that for catechol.The inefficiency towards 3-chlorocatechol has also beenshown for meta-pyrocatechases which were inactivated byeither 3-chlorocatechol itself (36) or its ring-cleavage prod-uct, 5-chloroformyl-2-hydroxypenta-2,4-dienoic acid (I. Bar-tels, H.-J. Knackmuss, and W. Reineke, unpublished data).For total degradation and utilization of chlorobenzene, accu-mulation of 3-chlorocatechol must be avoided by ortho-cleavage enzymes exhibiting high rates for halocatechols.

Such enzymes have been found in Pseudomonas sp. strainB13 and derivatives of it (44) as well as in Brevibacteriumfuscum (40) or in organisms that are capable of using 4-chloro- or 2,4-dichlorophenoxyacetate as the sole source ofcarbon and energy (6, 17, 18, 20, 21). Correspondingly, cellextracts of strain WR1306 grown on chlorobenzene containthis essential enzyme activity as well as the completesubsequent sequence of enzymes which convert the ringcleavage product 2-chloro-cis,cis-muconate to metabolitesof the tricarboxylic acid cycle: muconate cycloisomerase, 4-carboxymethylenebut-2-en-4-olide hydrolase, and maleyla-cetate reductase.A second explanation for chlorobenzene being a critical

growth substrate can be derived from its highly lipophiliccharacter. When present at a high concentration, this com-pound exhibits inhibitory effects even on strain WR1306which harbors all essential degradative enzymes for chloro-benzene.

XCI

02COOH

COOH

CI

t~ol0

COOH

COOHY COOH

COOHY COOH

TCCFIG. 7. Proposed catabolic pathway of chlorobenzene by strain

WR1306.

APPL. ENVIRON. MICROBIOL.

MICROBIAL METABOLISM OF HALOAROMATICS 401

Recently, a third adaptive response has been observedwhen ordinary catabolic functions for aromatics take on a

new role in chloroaromatic degradation. Utilization of ben-zene and chlorobenzene is somewhat incompatible. Initially,during adaptation to chlorobenzene, growth with this sub-strate was rather poor (data not shown). A culture derivedfrom this early state of adaptation could still grow withbenzene through the induction of a meta-cleavage pathway.Prolonged growth with chlorobenzene, however, confersselective advantage to populations which have lost the meta-cleavage activity. Consequently, the pure culture obtainedafter several passages of growth with chlorobenzene was no

longer able to use benzene as the growth substrate. Obvious-ly benzene when present as a growth substrate counteractstotal suppression of its meta pathway. This pathway, howev-er, which is unproductive for haloaromatics must be pre-

vented to allow productive breakdown of chlorocatecholsvia the ortho pathway. This has clearly been shown withhybrid strains derived from Pseudomonas sp. strain B13with chlorobenzoates and toluates as the model compounds(42).Although the present isolate as well as hybrid strains

obtained by conjugation between Pseudomonas sp. strainB13 and a benzene-degrading P. putida (Reineke, unpub-lished data) fulfill essential biochemical and physiologicalprerequisites for the utilization and rapid mineralization ofchlorobenzene, their suitability for the removal of chloro-benzene from industrial sewage or exhaust air would belimited by their sensitivity to shock loads.

LITERATURE CITED1. Axcell, B. C., and P. J. Geary. 1973. The metabolism of benzene

by bacteria. Purification and some properties of the enzyme cis-1,2-dihydroxycyclohexa-3,5-diene (nicotinamide adenine dinu-cleotide) oxidoreductase (cis-benzene glycol dehydrogenase).Biochem. J. 136:927-934.

2. Axcell, B. C., and P. J. Geary. 1975. Purification and some

properties of a soluble benzene-oxidizing system from a strainof Pseudomonas. Biochem. J. 146:173-183.

3. Baggi, G., D. Catelani, E. Gauli, and V. Treccani. 1972. Themicrobial degradation of phenylalkanes: 2-phenylbutane, 3-phenylpentane, 3-phenyldodecane and 4-phenylheptane. Bio-chem. J. 126:1091-1097.

4. Ballschmiter, K., C. Unglert, ajid P. Heizmann. 1977. Bildungvon Chlorphenolen durch mikrobielle Umwandlung von Chlor-benzolen. Angew. Chem. 89:680-681.

5. Biltz, H., and K. Stepf. 1904. Uber die Chlorierung des Salicylal-dehyds. Ber. Dtsch. Chem. Ges. 37:4022-4031.

6. Bollag, J.-M., G. G. Briggs, J. E. Dawson, and M.Alexander.1968. 2,4-D metabolism. Enzymatic degradation of chlorocate-chols. J. Agric. Food Chem. 16:829-833.

7. Bradford, M. M. 1976. A rapid and sensitive method for thequantitation of microgram quantities of protein utilizing theprinciple of protein-dye binding. Anal. Biochem. 72:248-254.

8. Bucksteeg, W. 1942.Uber den Einflu, und die Wirkungsweisevon Toluol auf Bakterienzellen. Zentralbl. Bakteriol. Parasi-tenkd. Infektionskr. Hyg. Abt. 2 105:209-213.

9. Catelani, D., A. Colombi, C. Sorlini, and V. Treccani. 1977.Metabolism of quaternary carbon compounds: 2,2-Dimethyl-heptane and terbutylbenzene. Appl. Environ. Microbiol.34:351-354.

10. Chapman, P. J., and D. W. Ribbons. 1976. Metabolism ofresorcinylic compounds by bacteria: alternative pathways forresorcinol catabolism in Pseudomonas putida. J. Bacteriol.125:985-998.

11. Claus, D., and N. Walker. 1964. The decomposition of tolueneby soil bacteria. J. Gen. Microbiol. 36:107-122.

12. Dakin, H. D. 1909. The oxidation of hydroxy derivatives ofbenzaldehyde, acetophenone and related substances. J. Am.Chem. Soc. 42:477-498.

13. Dorn, E., M. Hellwig, W. Reineke, and H.-J. Knackmuss. 1974.Isolation and characterization of a 3-chlorobenzoate degradingpseudomonad. Arch. Microbiol. 99:61-70.

14. Dorn, E., and H.-J. Knackmuss. 1978. Chemical structure andbiodegradability of halogenated aromatic compounds. Two cat-echol 1,2-dioxygenases from a 3-chlorobenzoate-grown pseudo-monad. Biochem. J. 174:73-84.

15. Dorn, E., and H.-J. Knackmuss. 1978. Chemical structure andbiodegradability of halogenated aromatic compounds. Substitu-ent effects on 1,2-dioxygenation of catechol. Biochem. J.174:85-94.

16. Erikson, D. 1941. Studies on some lake-mud strains of Micro-monospora. J. Bacteriol. 41:277-300.

17. Evans, W. C., B. S. W. Smith, H. N. Fernley, and J. I. Davies.1971. Bacterial metabolism of 2,4-dichlorophenoxyacetate. Bio-chem. J. 122:543-551.

18. Evans, W. C., B. S. W. Smith, P. Moss, and H. N. Fernley.1971. Bacterial metabolism of 4-chlorophenoxyacetate. Bio-chem. J. 122:509-517.

19. Gaal, A., and H. Y. Neujahr. 1979. Metabolism of phenol andresorcinol in Trichosporon cutaneum. J. Bacteriol. 137:13-21.

20. Gaunt, J. K., and W. C. Evans. 1971. Metabolism of 4-chloro-2-methylphenoxyacetate by a soil pseudomonad. Preliminaryevidence for the metabolic pathway. Biochem. J. 122:519-526.

21. Gaunt, J. K., and W. C. Evans. 1971. Metabolism of 4-chloro-2-methylphenoxyacetate by a soil pseudomonad. Ring-fission,lactonizing and delactonizing enzymes. Biochem. J. 122:533-542.

22. Gibson, D. T. 1971. The microbial oxidation of aromatic hydro-carbons. Crit. Rev. Microbiol. 1:199-223.

23. Gibson, D. T., G. E. Cardini, F. C. Maseles, and R. E. Kallo.1970. Incorporation of oxygen-18 into benzene by Pseudomonasputida. Biochemistry 9:1631-1635.

24. Gibson, D. T., B. Gschwendt, W. K. Yeh, and V. M. Kobal.1973. Initial reactions in the oxidation of ethylbenzene byPseudomonas putida. Biochemistry 12:1520-1528.

25. Gibson, D. T., M. Hensley, H. Yoshioka, and T. J. Mabry. 1970.Formation of (+)-cis-2,3-dihydroxy-1-methylcyclohexa-4,6-di-ene from toluene by Pseudomonas putida. Biochemistry9:1626-1630.

26. Gibson, D. T., J. R. Koch, and R. E. KallHo. 1968. Oxidativedegradation of aromatic hydrocarbons by microorganisms. I.Enzymatic formation of catechol from benzene. Biochemistry7:2653-2662.

27. Gibson, D. T., J. R. Koch, C. L. Schuld, and R. E. Kallio. 1968.Oxidative degradation of aromatic hydrocarbons by microorga-nisms. II. Metabolism of halogenated aromatic hydrocarbons.Biochemistry 7:3795-3802.

28. Gibson, D. T., V. Mahadevan, D. M. Jerina, H. Yagi, andH. J. C. Yeh. 1975. Oxidation of the carcinogens benzo[a]pyrene and benzo[a] anthracene to dihydrodiols by a bacterium.Science 189:295-297.

29. Haider, K., G. Jagnow, R. Kohnen, and S. U. Lim. 1974. Abbauchlorierter Benzole, Phenole und Cyclohexan-Derivate durchBenzol und Phenol verwertende Bodenbakterien unter aerobenBedingungen. Arch. Microbiol. 96:183-200.

30. Hartmann, J., W. Reineke, and H.-J. Knackmuss. 1979. Metabo-lism of 3-chloro-, 4-chloro-, and 3,5-dichlorobenzoate by apseudomonad. Appl. Environ. Microbiol. 37:421-428.

31. Hogn, T., and L. Jaenicke. 1972. Benzene metabolism of Morax-ella species. Eur. J. Biochem. 30:369-375.

32. Hutzinger,O., M. T. M. Tulp, and V. Zitko. 1978. Chemicalswith pollution potential, p. 13-31. In Aquatic pollutants: trans-formation and biological effects. Proceedings of the SecondInternational Symposium on Aquatic Pollutants, Noordwijker-hout (Amsterdam), The Netherlands. Pergamon Press, Inc.,Elmsford, N.Y.

33. Jackson, R. W., and J. A. DeMoss. 1965. Effects of toluene onEscherichia coli. J. Bacteriol. 90:1420-1425.

34. Jigami, Y., T. Omori, and Y. Minoda. 1975. The degradation ofisopropylbenzene and isobutylbenzene by Pseudomonas sp.Agric. Biol. Chem. 39:1781-1788.

35. Kao, C.-I., and N. Poffenberger. 1979. Chlorinated benzenes, p.

VOL. 47, 1984

402 REINEKE AND KNACKMUSS

797-808. In Kirk and Othmer, Encyclopedia of chemical tech-nology, 3rd ed., vol. 5. John Wiley & Sons, Inc., New York.

36. Klecka, G. M., and D. T. Gibson. 1981. Inhibition of catechol2,3-dioxygenase from Pseudomonas putida by 3-chlorocate-chol. Appl. Environ. Microbiol. 41:1159-1165.

37. Kleinzeller, A., and Z. Fencl. 1952. Oxydace benzenfi mikroor-ganismem Z odpadnich vod. Chem. Listy 46:300-303.

38. Marr, E. K., and R. W. Stone. 1961. Bacterial oxidation ofbenzene. J. Bacteriol. 81:425-430.

39. McKenna, E. J., and R. E. Kallio. 1965. The biology of hydro-carbons. Annu. Rev. Microbiol. 19:183-208.

40. Nakagawa, H., H. Inone, and Y. Takeda. 1963. Characteristicsof catechol oxygenase from Brevibacterium fuscum. J. Bio-chem. (Tokyo) 54:65-74.

41. Nozaki, M. 1970. Metapyrocatechase (Pseudomonas). MethodsEnzymol. 17A:522-525.

42. Reineke, W., D. J. Jeenes, P. A. Williams, and H.-J. Knack-muss. 1982. TOL plasmid pWW0 in constructed halobenzoate-degrading Pseudomonas strains: prevention of meta pathway. J.Bacteriol. 150:195-201.

43. Reineke, W., and H.-J. Knackmuss. 1979. Construction ofhaloaromatics utilising bacteria. Nature (London) 277:385-386.

44. Reineke, W., and H.-J. Knackmuss. 1980. Hybrid pathway forchlorobenzoate metabolism in Pseudomonas sp. B13 deriva-tives. J. Bacteriol. 142:467-473.

45. Reineke, W., W. Otting, and H.-J. Knackmuss. 1978. cis-Dihydrodiols microbially produced from halo- and methylben-zoic acids. Tetrahedron 34:1707-1714.

46. Schmidt, E., and H.-J. Knackmuss. 1980. Chemical structure

and biodegradability of halogenated aromatic compounds. Con-version of chlorinated muconic acids into maleoylacetic acid.Biochem. J. 192:339-347.

47. Schreiber, A., M. Hellwig, E. Dorn, W. Reineke, and H.-J.Knackmuss. 1980. Critical reactions in fluorobenzoic acid degra-dation by Pseudomonas sp. B13. Appl. Environ. Microbiol.39:58-67.

48. Shelat, Y. A., and K. S. Patel. 1973. Utilization of bromo-benzene as a sole source of carbon by Bacillus polymyxa. Curr.Sci. 42:368.

49. Sparnins, V. L., D. G. Burbee, and S. Dagley. 1979. Catabolismof L-tyrosine in Trichosporon cutaneum. J. Bacteriol. 138:425-430.

50. Tausson, W. 0. 1929. Uber die Oxydation der Benzolkohlen-wasserstoffe durch Bakterien. Planta (Berl.). 7:735-758.

51. van der Linden, A. C., and G. J. E. ThiJsse. 1965. The mecha-nisms of microbial oxidations of petroleum hydrocarbons. Adv.Enzymol. Relat. Subjects Biochem. 27:469-546.

52. Wagner, R. 1914. Uber Benzol-Bakterien. Z. Garungsphysiol.4:289-319.

53. Wieland, T., G. Griss, and B. Haccius. 1958. Untersuchungenzur mikrobiellen Benzoloxydation. I. Nachweis und Chemismusdes Benzolabbaus. Arch. Mikrobiol. 28:383-393.

54. Willstfitter, R., and H. E. Muller. 1911. Uber Chlorderivate desBrenzcatechins und des ortho-Chinons. Ber. Dtsch. Chem. Ges.44:2182-2191.

55. Woldringh, C. L. 1973. Effects of toluene and phenethyl alcoholon the ultrastructure of Escherichia coli. J. Bacteriol. 114:1359-1361.

APPL. ENVIRON. MICROBIOL.