APPLIED AND ENVIRONMENTAL MICROBIOLOGY, Apr. 1989, p. 946-952 Vol. 55, No. 40099-2240/89/040946-07$02.00/0Copyright © 1989, American Society for Microbiology

Bacterial Metabolism of Hydroxylated BiphenylsFRANK K. HIGSON AND DENNIS D. FOCHT*

Department of Soil and Environmental Sciences, University of California, Riverside, California 92521

Received 16 November 1988/Accepted 23 January 1989

Isolates able to grow on 3- or 4-hydroxybiphenyl (HB) as the sole carbon source were obtained by enrichmentculture. The 3-HB degrader Pseudomonas sp. strain FH12 used an NADPH-dependent monooxygenaserestricted to 3- and 3,3'-HBs to introduce an ortho-hydroxyl. The 4-HB degrader Pseudomonas sp. strain FH23used either a mono- or dioxygenase to generate a 2,3-diphenolic substitution pattern which allowed meta-fissionof the aromatic ring. By using 3-chlorocatechol to inhibit catechol dioxygenase activity, it was found that 2- and3-HBs were converted by FH23 to 2,3-HB, whereas biphenyl and 4-HB were attacked by dioxygenation. 4-HBwas metabolized to 2,3,4'-trihydroxybiphenyl. Neither organism attacked chlorinated HBs. The degradation of3- and 4-HBs by these strains is therefore analogous to the metabolism of biphenyl, 2-HB, and naphthalene inthe requirement for 2,3-catechol formation.

Polychlorinated biphenyls (PCBs) have found widespreadindustrial use in the past owing to their physical and chem-ical stability and their dielectric properties. Inadequatewaste disposal has led to their release into the environment,where they have been routinely detected in soil, water, andwildlife since the early 1960s. Their toxicity has causedPCBs to be prohibited in many countries. Originally believedrecalcitrant, PCBs have been shown to be degraded bothphotochemically and by microorganisms (7, 18).The monohydroxylation of biphenyl is a detoxification

process common to hepatic microsomes (25) and a range offungi and actinomycetes (27). The chlorinated derivativesare frequently hydroxylated as well, even though severalassays, such as erythrocyte lysis (T. L. Miller, Fed. Proc.,36:413, 1977) and mouse acute 50% lethal dose (36), demon-strate the greater toxicity of the hydroxylated PCBs over theparent isomers. This may be due to the increased solubilityand membrane interaction of the products.para-Hydroxylation of biphenyl normally predominates in

mammals, fungi, and actinomycetes; Streptomyces rimosus,an exception, generates only 2-hydroxybiphenyl (2-HB; 28).A cytochrome P450-linked monooxygenase (27) is responsi-ble for these transformations, and water-soluble conjugatesare generated from the phenylphenols (6, 35).A number of investigators have isolated bacterial strains

capable of utilizing biphenyl as a sole carbon source, includ-ing species of Achromobacter (1), Pseudomonas (5), andBeijerinckia (16), which oxidize biphenyl at the 2,3 positionto form a cis-dihydrodiol. This is then dehydrogenated toyield 2,3-dihydroxybiphenyl, which is metabolized throughthe metapyrocatechase pathway to form benzoic acid. Broadspecificity of the initiating enzymes effects cometabolism ofPCBs to hydroxylated intermediates that are not growthsubstrates of the biphenyl degraders Acinetobacter sp. strainP6 and Alcaligenes sp. strain Y42 (13). Neither organismgrew on 2- or 4-HB, although oxygen was consumed inresponse to both substrates with resting cells of strain P6.HBs may arise in the environment from the fungal oxida-

tion of biphenyl applied to fruit on a large scale as a fungicideand from the spraying of sodium 2-phenylphenolate to pre-vent apple and citrus decay (10). They have also been usedin household disinfectants, as preservatives for water-oil

* Corresponding author.

emulsions in textiles, and by metal industries. The orthoisomer is considered to be moderately toxic (17), causingcorneal injury and irritation when inhaled. Cats fed largedoses display severe internal hemorrhaging, and in the rat,the compound is stored primarily in the kidney (33). Inaddition to microbial cometabolism, hydroxylated PCBs canbe produced photochemically (together with products ofdechlorination [1]) by prechlorination in drinking-watertreatment (34) and by PCB pyrolysis (26).

Kohler et al. (22) recently isolated Pseudomonas sp. strainHBP1, which can use 2-HB but not biphenyl as the solecarbon source by monooxygenase attack. The strain did notsignificantly metabolize chlorinated analogs.We report here on the degradation of 3- and 4-HBs by new

strains of Pseudomonas sp. Their action on chlorinatedanalogs is also discussed.

MATERIALS AND METHODSIsolation and growth of strains used. The strains FH12 and

FH23 were isolated from activated sewage sludge by enrich-ment and serial transfer with 3- or 4-HB, respectively, as thegrowth substrate in the minimal salts medium of Kroeckeland Focht (23). The organisms were grown in Fernbachflasks agitated by a gyratory shaker to provide suitableaeration; biphenyl and its hydroxylated derivatives wereadded to the hot medium after autoclaving to minimizelosses due to volatilization. Growth did not occur in uninoc-ulated controls. The strains were characterized by the pro-cedures described by Starr et al. (32) and Gerhardt et al. (14).

Chemicals. 3,3'-Dihydroxybiphenyl was synthesized from3,3'-dimethoxybiphenyl by a procedure based on that ofDawson et al. (C. R. Dawson, D. Wasserman, and H. Keil,Letter, J. Am. Chem. Soc. 68:534-535, 1946). A mixture of10 g of the dimethoxy compound, 120 ml of chlorobenzene,and 12.5 g of aluminum chloride (all three from AldrichChemical Co., Inc., Milwaukee, Wis.) was refluxed in a1-liter three-neck round-bottom flask fitted with a condenserand thermometer for 3.5 h and cooled, and the solvent wasreduced by rotary evaporation. The solution was then acid-ified with HCI, and all organics were extracted with diethylether (Mallinckrodt, St. Louis, Mo.). A 5% (wt/vol) aqueoussolution of sodium carbonate (Fisher Scientific Co., Pitts-burgh, Pa.) was applied to the ether extract to remove thedesired product as the diphenolate salt, while leaving thestarting material largely in the organic phase. The aqueous

946

on June 16, 2017 by guesthttp://aem

.asm.org/

Dow

nloaded from

BACTERIAL METABOLISM OF HYDROXYLATED BIPHENYLS

extract was acidified with HCl and dihydroxybiphenyl wasagain extracted into diethyl ether. The ether phase was driedwith sodium sulfate (Fisher), and the solvent was evapo-rated. Gas chromatography-mass spectroscopy (GC-MS)and nuclear magnetic resonance (NMR) results confirmedthe identity of the product.

3-Chlorocatechol was synthesized from 2-chlorophenol ina two-step procedure; a Reimer-Tiemann addition of a CHOgroup (29) produced mostly 3-chlorosalicylaldehyde, whichwas converted to the chlorocatechol by the Dakin oxidation(8). After the first reaction (with chloroform in aqueoussodium hydroxide), diethyl ether was used to extract allorganics, and sodium bicarbonate solution was applied toremove any salicylate formed, while leaving the aldehyde inthe ether phase. A bisulfite addition compound was pro-duced and filtered out, and the aldehyde was regenerated onacidification and recovered by ethyl acetate extraction. Theproduct was purified by thin-layer chromatography on a1-mm-thick silica gel G plate (Analtech, Newark, Del.) byusing a solvent system of Knuutinen (17), viz., hexane-ethylacetate-acetic acid (80:15:5; all supplied by Fisher). Thecatechol band was identified by its formation of an orangecolor with tungstate and nitrite (12). The identity of the3-chlorocatechol fraction was confirmed by NMR, UV andinfrared spectroscopy, and GC-MS. The compound showedbroad absorption peaks at 213 and 277 nm in methanolsolution. The mass spectrum corresponded to that of theauthentic standard presented by Knuutinen and Korhonen(21).Methanol, ethyl acetate (high-performance liquid chroma-

tography grade), and phenol were obtained from Fisher; 3-and 3,4-HBs and diphenyl-methanol were from EastmanKodak Co., Rochester, N.Y.; 3,4,3',4'-HB and 2-chlorobi-phenyl were from K&K Labs, ICN Biomedical Corp.,Plainview, N.Y.; 2-chlorophenol was obtained from GeneralDynamics, South Plainfield, N.J.; 3-chlorobiphenyl wasfrom Chemical Procurement Co., College Point, N.Y.; 4-chlorobiphenyl was from Pfaltz and Bauer, Waterbury,Conn.; dichlorobiphenyls were from Analabs, Norwalk,Conn.; hydroxylated PCBs were from Chem Service, WestChester, Pa.; N,O-bis(trimethylsilyl)acetamide was fromPierce Chemical Co., Rockford, Ill.; 2,3-HB was a generousgift from Herman Finkbeiner, General Electric Research andDevelopment Center, Schenectady, N.Y. Other chemicalswere purchased from Aldrich.

Preparation of washed cell suspensions and extracts. Cellswere harvested at the end of the logarithmic growth phase bycentrifugation (15 min at 12,000 x g), washed with potassiumphosphate buffer (50 mM, pH 7.5), and resuspended in thesame buffer. Cell-free extracts were prepared from suspen-sions by passage through a French press (20,000 lb/in2) andcentrifugation at 40,000 x g for 40 min at 4°C. Protaminesulfate (20 mg/ml of phosphate buffer) was added to thecrude extract to precipitate DNA and histones, using 0.2ml/ml.

Derivatization. A sample of the ethyl acetate (triple) ex-tract from a cell transformation was clarified by centrifuga-tion in Eppendorf tubes (2 min in a microcentrifuge; Beck-man Instruments, Inc., Fullerton, Calif.), and the solventwas removed by using a stream of nitrogen gas. The residuewas treated with N,O-bis(trimethylsilyl)acetamide, whichsilylates phenolic groups, at 70°C for 15 min. Excess reagentwas evaporated, and the product was redissolved in metha-nol.

Purification of a trihydroxybiphenyl metabolite. The prod-uct of 4-HB transformation was separated by silica gel

preparative thin-layer chromatography by using a mobilephase of Davis et al. (9), viz., 9:1 (vol/vol) benzene-meth-anol. The metabolite was recognized as a catechol by itsfluorescence on spraying with para-toluene sulfonate (20%solution in chloroform; 31). The substrate showed no suchreaction under UV illumination. GC analysis showed theproduct to be >88% pure, eluting just after 2,3- and 3,4-HBs.

Analytical methods. Oxygen uptake rates were measuredwith an oxygen electrode (Yellow Springs Instrument Co.,Yellow Springs, Ohio), which was mounted to a reactionvessel held at a constant temperature (30°C). The assaymixture contained 0.1 ml of the cell suspension (opticaldensity at 600 nm, 10) and 1.9 ml of phosphate buffer.Reactions were started by injecting substrates as concen-trated methanol solutions (10 ,ul of a 10 mg/ml stock, except5 mg/ml for phenols and 100 mg/ml for benzoates), andspecific oxygen consumption rates were corrected for endog-enous uptake. Methanol had no effect in the concentrationrange used. Monooxygenase activity with 3-HB as thesubstrate was assayed either polarographically by measuringoxygen consumption or spectrophotometrically (Uvikon860; Kontron Instruments, Everett, Mass.) by following thedisappearance of NADH at 340. Addition of cell extractstarted the reaction; the reaction mixture contained 0.2 mMNADH and 0.1 mM substrate in the phosphate buffer de-scribed above. The rate was corrected for backgroundconsumption in the absence of HB.1H-NMR spectra were recorded in deuterated methanol

solution in a General Electric QE300 instrument. The NMRparameters used were 8-p.s flip angle with 23-s recycle time;eight scans were collected and coadded; 16k datum pointswere used in quadrature detection. A 0.3-Hz line-broadeningwas applied before Fourier transform; integration of theresonances was carried out at 7.35 and 6.6 to 6.8 ppm.Mass spectra were obtained on one of two high-resolution

spectrometers (ZAB 2FHF; VG Analytical, Ltd., Manches-ter, England, or model 4000; Finnigan, San Jose, Calif.)operated at 70 eV. Both instruments were interfaced to a gaschromatograph (model 5790; Hewlett-Packard Co., PaloAlto, Calif.) which contained a DB5 capillary column (J&WScientific, Folsom, Calif.). An oven temperature gradientfrom 50 to 250°C (100 rise per min) was used, while theinjector and detector were held at 220 and 250'C, respec-tively.

RESULTS

Characterization. A bacterium able to grow on 3-HB as thesole carbon source was isolated from activated sewagesludge by selective enrichment. The strain, designated Pseu-domonas sp. strain FH12, was a short gram-negative rod,motile by polar flagella, that was catalase- and cytochromeoxidase-positive. The organism hydrolyzed gelatin, reducednitrate, did not form levan from Luria medium containingsucrose, was nonfluorescent on King A or B or glutamatemedium, grew on citrate and fructose, and showed strictlyaerobic growth on glucose. FH12 was resistant in Luriamedium to nalidixate, erythromycin, chloramphenicol (all 50mg/I), penicillin G (400 mg/l), and ampicillin (100 mg/l) andsensitive to tetracycline, kanamycin, and streptomycin (all50 mg/I). In batch culture, the growth rate (p.) of FH12 on3-HB was 0.56/h (doubling time, 1.7 h), and the opticaldensity (600 nm) reached 0.46. A transient yellow coloration,suggesting a meta ring-cleavage product (Xmax, 375 nm,decolorized by acid), could be observed in this culture. Thestrain also grew on 3,3'-HB, benzoate, and 3- and 4-hy-

VOL. 55, 1989 947

on June 16, 2017 by guesthttp://aem

.asm.org/

Dow

nloaded from

948 HIGSON AND FOCHT

TABLE 1. Oxygen uptake responses of resting cells of3-HB-grown Pseudomonas sp. strain FH12 to

various substrates

Substrate Oxygen uptake rate(nmol/min per mg of protein)"

3-HB .......... .................... 94 (100)3,3'-HB ............. ................. 89 (95)2-HB .......... .................... 10 (11)4-HB..............................0Biphenyl .............. ................ 7 (7)2,3-HB ............ .................. 106 (99)3,4-HB .............................. 13 (12)Benzoate .............................. 53 (50)2-Hydroxybenzoate ........................... 17 (16)3-Hydroxybenzoate ........................... 17 (16)4-Hydroxybenzoate ........................... 8 (7)3'-Hydroxyacetophenone .................... 3 (3)

" The values in parentheses are relative to the growth substrate.

droxybenzoates. It failed to grow on biphenyl, 2- or 4-HB,2,2'- or 4,4'-HB, methylbiphenyls, salicylate, and chlori-nated biphenyls, HBs, and benzoates.Pseudomonas sp. strain FH23 was isolated by selective

enrichment with 4-HB as the sole carbon source. Theorganism displayed the same taxonomic characteristics asFH12 but was a more extended rod and was sensitive to 50mg of chloramphenicol per liter. The maximal growth rate(p>) in 4-HB medium was 0.35/h (doubling time, 2.8 h), and an

optical density (at 600 nm) of 0.69 was reached after 3 days.Strain FH23 grew on biphenyl and many HBs (2-, 3-, 4-,2,2'-, 3,3'-, 2,3-, and 2,5-HBs) and benzoate and hydroxy-benzoates (2-, 3-, 4-, 2,3-, 2,5-, and 3,4-). It also grew on4-hydroxymethyl and 4-methoxybiphenyls but not on theortho isomers, on 4,4'-HB, or on any of the chlorinatedbiphenyls or HBs tested. FH23 did not grow on toluene,naphthalene, indole, chlorinated benzoates, hydroxylateddiphenylmethanes or benzophenones, alkylphenols, or bi-phenylmethanols but did grow on phenol medium at aconcentration of 250 ppm or lower.

Degradation of 3-HB by Pseudomonas sp. strain FH12.Whole cells of Pseudomonas sp. strain FH12 showed en-

hanced oxygen uptake in response to HBs other than 3- and3,3'-HBs (Table 1), but the cell extract was restricted inNADH-dependent response to only these two substrates.3,3'-Dihydroxybiphenyl was one-third more active on amolar basis than 3-HB at 0.2 mM. Both NADH and oxygenconsumption rates fell off rapidly, but if similar time periodsare compared, the NADH/02 stoichiometry was 0.93 and0.89 for 3- and 3,3'-HBs, respectively. No activity (oxygenuptake, yellow color) was observed from 2,3-HB. The ratewith NADPH was 64% of that with NADH. Maximal NADHdisappearance was seen at a 3-HB concentration of about0.07 mM (Fig. 1). The profile suggests that substrate inhibi-tion occurs. This phenomenon has been described previ-ously (3) by the expression ViJx/v = K,,,/[S] + (1 + [S]IKSS)where Vmax is the maximum rate and v the observed rate innanomoles of NADH consumed per minute per milligram atprotein, Km is the Michaelis constant, KSS is the inhibitionconstant, and [S] is the 3-HB concentration (all three inmicromolar). The above equation can be rearranged to a

polynomial y = a + bx + Cx2, where y = [S]lv and x = [S].Here, a = K,n/Vmax, b = 1/Vmax, and c = 1/(KSs Vmax).Computer fitting of the data gave a Vmax of 2,600 nmol ofNADH consumed per min per mg of protein (K,,, = 1,026pLM; KSS = 3.795 ,uM.Cells of FH12 grown on 3-HB were induced for benzoate

A

C

8

0

0.

E 6

r'4

z

E 2c

B

E

C

I.

0

0.

E

c

Cl)

200 400 600

100 200

(S),I.tMFIG. 1. Dependence of Pseudomonas sp. strain FH12 monoox-

ygenase activity on 3-HB concentration, [SI, as determined by therate of disappearance of NADH (v).

oxidation; those grown on 3,3'-HB were induced for 3-hydroxybenzoate oxidation (Table 2).3-HB-grown FH12 readily oxidized 2,3-HB, producing a

bright-yellow color, while 3,4-HB evoked a minimal re-

sponse and no coloration. 3-Chlorocatechol was observed toinhibit the catechol 2,3-dioxygenase of Pseudomonas putidamt-2 (3). Since 3-chlorocatechol at 1 mM was found toprevent the oxygen uptake response with 2,3-HB, it wasused at this concentration in resting-cell incubations withseveral substrates (5 mg/20 ml) in order to accumulatecatecholic intermediates. Neutral ethyl acetate extracts wereanalyzed by GC-MS, both underivatized and after N,O-bis(trimethylsilyl)acetamide treatment. 3-HB generated 2,3-HB; 3,3'-HB produced a trihydroxybiphenyl, identified as

Bo A.0

-0

10

n

APPL. ENVIRON. MICROBIOL.

on June 16, 2017 by guesthttp://aem

.asm.org/

Dow

nloaded from

BACTERIAL METABOLISM OF HYDROXYLATED BIPHENYLS

TABLE 2. Oxygen uptake responses of resting cells ofPseudomonas sp. strains FH23 and FH12 to benzoate and

hydroxybenzoates when grown on different substrates

Oxygen uptake ratePseuido- (nmol/min per mg of protein)mnostias sp.

Growth with test substratestrain

~ substratestrain Benzo- Salic- 3-Hydrox- 4-Hydrox-

ate ylate ybenzoate ybenzoate

FH234-HB 0 4 0 20Biphenyl 91 5 0 02-HB 9 4 0 82,2'-HB 15 54 0 03,3'-HB 22 20 123 16Benzoate 3 0 0 10Salicylate 3 55 5 03-Hydroxybenzoate 0 7 67 04-Hydroxybenzoate 0 0 0 33

FH123-HB 164 19 13 333,3'-HB 9 8 387 12Benzoate 23 0 15 03-Hydroxybenzoate 5 0 51 3

loor

80

60

20

OHo

OH

63 94

145 174

II 184

202

L," I

50 100 150 200

FIG. 3. Mass spectrum of 2,3,3'-trihydroxybiphenyl, formed bythe oxidation of 3,3'-dihydroxybiphenyl by Pseudomonas sp. strainFH12.

2,3,3'-HB. The mass spectra of the silylated derivatives ofboth degradation products are given in Fig. 2 and 3.

Degradation of 4-HB by Pseudomonas sp. strain FH23.Cells of FH23 showed oxygen consumption in response to awide range of hydroxylated biphenyls and their analogs(Tables 2 and 3). With phenolic compounds, oxygen uptakeresponse became greater as the size of the alkyl group

increased. Certain chlorinated biphenyls effected oxygen

uptake, even though their hydroxylated derivatives wereinactive. 4-HB-grown cells were induced for oxidation of4-hydroxybenzoate rather than benzoate (Table 4), whilegrowth on 2- and 2,2'-HB induced for the oxidation ofbenzoate and salicylate, respectively.

100

80

186

OH

OH 170

6o_

40_

201

r,1

139115

78 129

105 1I ,-.Vl I,, I I I,,50 100 150 200

FIG. 2. Mass spectrum of 2,3-dihydroxybiphenyl, formed by theoxidation of 3-HB by Pseudomnonas sp. strain FH12.

When 4-HB-grown cells of FH23 were incubated with4-HB in the presence of 3-chloro-4-hydroxybenzoate to actas a competitive inhibitor, 4-hydroxybenzoate accumulated,as evidenced by coelution of the authentic compound with

TABLE 3. Oxygen uptake responses of resting cells of4-HB-grown Pseudomonas sp. strain FH23 to

a range of substrates

Substrate Oxygen uptake rate(nmol/min per mg of protein)'

4-HB.131 (100)2-HB.61 (47)3-HB.42 (32)2,2'-HB.43 (33)3,3'-HB.38 (29)4,4'-HB.48 (37)2,3-HB.298 (228)3,4-HB. 91 (70)2,5-HB.40 (31)3,4,3',4'-HB.16 (12)Biphenyl.77 (59)4-Hydroxydiphenylmethane 74 (57)4-Hydroxybenzophenone.29 (22)4-Biphenylmethanol.97 (74)4-Methoxybiphenyl.142 (109)2-Hydroxydiphenylmethane 73 (56)2-Hydroxybenzophenone.43 (33)2-Biphenylmethanol.25 (19)2-Methoxybiphenyl.37 (28)Diphenylmethanol.21 (16)2-Hydroxynaphthalene.25 (19)2-Chlorobiphenyl.12 (9)3-Chlorobiphenyl.48 (37)4-Chlorobiphenyl.84 (64)2,2'-Dichlorobiphenyl.3 (2)3,3'-Dichlorobiphenyl. 04,4'-Dich lorobiphenyl.10 (8)2,3-Dich lorobiphenyl.13 (10)3,4-Dich lorobiphenyl.31 (24)2,4'-Dichlorobiphenyl. 0

" The values in parentheses are relative to the growth substrate.

VOL. SS, 1989 949

40 _

on June 16, 2017 by guesthttp://aem

.asm.org/

Dow

nloaded from

950 HIGSON AND FOCHT

TABLE 4. Oxygen uptake responses of resting cells of4-HB-grown Pseudomonas sp. strain FH23 to

a range of phenols

Substrate Oxygen uptake rate(nmol/min per mg of protein)"

Phenol.............................. 13 (10)2-Methylphenol .............................. 25 (19)2-Ethylphenol .............................. 31 (24)2-Propylphenol .............................. 59 (45)2-sec-Butylphenol .............................. 42 (32)2-tert-Butylphenol ............................. 26 (20)2-Chlorophenol .............................. 17 (13)2-Bromophenol .............................. 22 (17)3-Methylphenol .............................. 9 (7)3-Ethylphenol .............................. 9 (7)3-Isopropylphenol ............................. 14 (11)3-tert-Butylphenol ............................. 18 (14)3-Chlorophenol .............................. 9 (7)4-Methylphenol .............................. 1 (1)4-Ethylphenol .............................. 14 (11)4-Isopropylphenol ............................. 14 (11)4-tert-Butylphenol ............................. 21 (16)4-Chlorophenol .............................. 14 (11)All PCB phenols tested"...................... 0

aThe values in parentheses are relative to the growth substrate.bThe congeners used were 4'-CI-4-HB, 2',3',4',5'-tetraCl-4-HB, 3-Cl-

2-HB, 5-CI-2-HB, 3,5-diCI-2-HB, 2',5'-diCI-2-HB, 2',5,5'-triC1-2-HB, 2',3',5'-triCI-2-HB, 2',3,3',4',5'-pentaCI-2-HB, 2',3',4',5,5'-pentaC1-2-HB, 2',5'-diCI-3-HB, 2',3',4',5'-tetraC1-3-HB, 4,4'-diCI-3-HB, and 3,3',5,5'-tetraCI-4,4'-HB.

the metabolite on high-performance liquid chromatography.Growth on four different substrates, 4-, 2-, and 2,2'-HBs andbiphenyl, strongly induced for oxygen uptake on 2,3-HB,generating a bright-yellow color (Xmax = 375 nm) that waslost on acidification. No such yellow color was seen with theother phenylcatechol, 3,4-HB.

Resting cells of FH23 were incubated with a number ofsubstrates in the presence of 3-chlorocatechol, and theextracts were prepared for GC-MS analysis as outlinedabove. The action on 4-HB, for example, produced a com-pound of molecular weight 202 which gave rise to threederivatives of molecular weights 274, 346, and 418. Thesevalues would correspond to sequential substitution of phe-nolic hydrogens with a trimethylsilyl group. Mass spectra ofunreacted and tris(trimethylsilyl) compounds are presentedin Fig. 4 and 5.

Further, the NMR spectrum (Fig. 6) of the metaboliteisolated from a resting-cell transformation was consistentwith its identification as 2,3,4'-trihydroxybiphenyl in view ofcorresponding signals reported (Sadtler Research Laborato-ries, NMR standard spectra no. 10 383 and no. 17 034, 1970)for catechol and 4-HB. In this interpretation, we assume thatthe two protons ortho to the single hydroxyl (3',5') producea fused signal (split by long-range coupling to the other ring)and that the meta pair (2',6') behave in the same way. Thefirst set would be upfield-shifted by the hydroxyl lone pairs.Free rotation would eliminate any differences in field due tothe catecholic ring groupings. The signal centered on 6.68ppm is split many times and may represent the three remain-ing protons since the signal is upfield. Moreover, integrationof the whole aromatic region gives a 1:1:1.5 ratio for signalsfrom the first pair, second pair, and cluster, respectively.The spectrum is not consistent with 2,3,4-HB, since anunsubstituted ring would generate signals from five relativelydownfield protons.

Finally, the purified metabolite evoked a strong oxygenuptake response in resting cells, with the formation of ayellow meta ring fission compound.

I00F

80F

OH

O H

60F

40F

20-

11 Ii 27145 1 310163

5

i IhKL L

202

50 100 150 200

FIG. 4. Mass spectrum of 2,3,4'-trihydroxybiphenyl, formed bythe oxidation of 4-HB by Pseludornonas sp. strain FH23.

Compounds of similar mass spectral fragmentation pat-terns and silylation products were also produced from othersubstrates (Table 5). Variation in retention times presumablyreflects alternative configurations of three hydroxyls aboutthe biphenyl nucleus. 2,3-HB was generated from biphenyl,2- and 3-HBs, a dihydroxy derivative, was from 4-chlorobi-phenyl, and no metabolites were observed from 4'-chloro-4-HB or 4-hydroxydiphenylmethane, or at time zero (<1min), or from cell-free controls.A cell extract of FH23 supplemented with NADH effected

rapid oxygen consumption on addition of 2- and 3-HBs,whereas 4-HB, biphenyl, or 4-chlorobiphenyl gave no re-sponse.

IOOr

80k

60k

40k

20-

~73

(CH 3)3 Si O

(CH 3)3-Si -0

O-Si - (CH3)3

13147 258133 165 287241 1274.L . .I . . i^. . I. .

418

315

330346 403

vT-7 7 7 TrT -

IIII rTIIII ITI-I-I1 I-r50 100 150 200 250 300 350 400 450 500

FIG. 5. Mass spectrum of the tris(trimethylsilyl) derivative of2,3,4'-trihydroxybiphenyl produced from 4-HB and Pseudomonassp. strain FH23.

APPL. ENVIRON. MICROBIOL.

on June 16, 2017 by guesthttp://aem

.asm.org/

Dow

nloaded from

BACTERIAL METABOLISM OF HYDROXYLATED BIPHENYLS

OH

3-HB

I ,p.

OH8.0 7.5 7.0 6.5 6.0

PPM

FIG. 6. NMR spectrum of purified 2,3,4'-trihydroxybiphenylformed by the oxidation of 4-HB by Pseudomonas sp. strain FH23.

DISCUSSION

Soil bacteria have been found to degrade aromatic hydro-carbons, such as benzene (2), toluene (37), naphthalene (11),and biphenyl (16) with the aid of dioxygenases that introduceboth of the atoms of molecular oxygen. Catechol or ananalog is produced as an intermediate and cleaved bymeta-pyrocatechase. The catechol has a 3- rather than a4-substituent in the toluene and biphenyl pathways. Thus,2,3-HB but not 3,4-HB is the isomer of choice in biphenyldegradation.The strains described in this study conform to this 2,3-

preference. A meta-hydroxyl led to the insertion of anadjacent ortho-hydroxyl in Pseudomonas sp. strain FH12,while a para-hydroxyl caused Pseudomonas sp. strain FH23to attack the unsubstituted ring, presumably using a dioxy-genase to form a 2,3,4'-HB from 4-HB. These actions aresummarized in Fig. 7.FH12 attacks 3-HB by monoxidation, while FH23 attacks

an HB by either mono- or dioxygenation. The latter organ-ism can therefore also metabolize 2- and 3-HBs, where onlyone hydroxyl completes the catechol pattern. While aro-matic dioxygenases are multicomponent and labile, the 2-HBmonooxygenase described by Kohler et al. (22) was readilymeasured in cell extracts. The loss of activity with 4-HB incells that retain the ability to oxidize 2-HB would be consis-

TABLE 5. Metabolites generated by Pseudomnonas sp. strainFH23 from biphenyl and substituted biphenyls in the presence of

3-chlorocatechol as determined by GC-MS

Parent ion inParention in Suggested structureSubstrate mass spectrum of Sumetructeintermediate ~ of metaboliteintermediate

4-HB 202 2,3,4'-HBBiphenyl 186 2,3-HB2-HB 186 2,3-HB

202 2,3,2'-HB3-HB 186 2,3-HB

202 2,3,3'-HB2,2'-HB 202 2,3,2'-HB3,3'-HB 202 2,3,3'-HB4,4'-HB NS"4-Chlorobiphenyl 220 2,3-Dihydroxy-4'-

chlorobiphenyl4'-Chloro-4-HB NS

- NS, No significant intermediates accumulated.

2,-OH

2,3-HB

OH

OH

OH

>C

OH 0

BENZOIC ACID

OH 0

OH

4-HB 2,3,4- HB 4-HYDROXY-BENZOIC ACID

FIG. 7. Proposed routes for the degradation of 3-HB by Pseiudo-monas sp. strain FH12 and of 4-HB by Pseutdomonas sp. strainFH23. Formation of benzoate or 4-HB proceeds along the conven-tional metapyrocatechase pathway.

tent with this differential stability. Spain and Gibson (30)recently reported that P. putida Fl carries both dioxygenaseactivity against benzene, toluene, and chlorobenzene and amonooxygenase activity that converts methyl-, chloro-, andnitro-substituted phenols to the corresponding catechols.The presumed FH23 dioxygenase has broad specificity, in

attacking biphenyl with a para-hydroxyl or chlorine substi-tuent. This can be compared with the Beijerinckia dioxygen-ase that oxidizes biphenyl (16), benz[a]pyrene (15), anthra-cene (19), and dibenzothiophene (24) to cis-dihydrodiols.

Neither organism showed significant activity with hydrox-ylated PCBs. Ring chlorination presumably prevents theaction of the monooxygenase system for steric and/or elec-tronic reasons.

ACKNOWLEDGMENTS

The work was funded in part by National Science Foundationgrant ECE 8419315 and by a grant from the Occidental Chemical Co.We are grateful for helpful discussion with Hans-Peter E. Kohler,

for mass spectroscopy assistance from Richard Kondrat, and forassistance with NMR spectrosocpy from Robert Lee.

LITERATURE CITED1. Ahmed, M., and D. D. Focht. 1973. Degradation of polychlori-

nated biphenyls by two species of Achromobacter. Can. J.Microbiol. 19:47-52.

2. Axcell, B. C., and P. J. Geary. 1975. Purification and someproperties of a soluble benzene-oxidizing system from a strainof Pseiudoinonas. Biochem. J. 146:173-183.

3. Bartels, I., H.-J. Knackmuss, and W. Reineke. 1984. Suicideinactivation of catechol 2,3-dioxygenase from Pseudoinonasputtida mt-2 by 3-halocatechols. Appl. Environ. Microbiol.47:500-505.

4. Bray, H. G., and K. White. 1966. Kinetics and thermodynamicsin biochemistry, p. 295. Academic Press, Inc., New York.

5. Catelani, D., and A. Colombi. 1974. Metabolism of biphenyl.Structure and physicochemical properties of 2-hydroxy-6-oxo-6-phenylhexa-2,4-dienoic acid, the meta-cleavage product from2,3-dihydroxybiphenyl by Pseiudomonas putida. Biochem. J.143:431-434.

6. Cerniglia, C. E., J. P. Freeman, and R. K. Mitchum. 1983.Glucuronide and sulfate conjugation in the fungal metabolism ofaromatic hydrocarbons. Appl. Environ. Microbiol. 43:1070-1075.

7. Crosby, D. G., and K. W. Moilanen. 1973. Photodecompositionof chlorinated biphenyls and dibenzofurans. Bull. Environ.Contam. Toxicol. 10:372-377.

8. Dakin, H. D. 1909. The oxidation of hydroxy derivatives of

VOL. 55, 1989 951

on June 16, 2017 by guesthttp://aem

.asm.org/

Dow

nloaded from

952 HIGSON AND FOCHT

benzaldehyde, acetophenone and related substances. Am.Chem. J. 42:477-498.

9. Davis, P. J., L. K. Jamieson, and R. V. Smith. 1978. Determi-nation of biphenyl and eight biphenols in microbial extracts bygas chromatography and thin layer chromatography. Anal.Chem. 50:736-740.

10. Eckert, J. W. 1977. Control of postharvest diseases. F. Phenols,p. 318-322. In M. R. Siegel and H. D. Sisler (ed.), Antifungalcompounds, vol. 1. Marcel Dekker, Inc., New York.

11. Ensley, B. D., D. T. Gibson, and A. L. Laborde. 1982. Oxidationof naphthalene by a multicomponent enzyme system fromPseudomonas sp. strain NCIB9816. J. Bacteriol. 149:948-954.

12. Evans, W. C. 1947. Oxidation of phenol and benzoic acid bysome soil bacteria. Biochem. J. 41:373-382.

13. Furukawa, K., F. Matsumara, and K. Tonomura. 1978. Alccali-genes and Acinetobacter strains capable of degrading polychlo-rinated biphenyls. Agric. Biol. Chem. 42:543-548.

14. Gerhardt, P., R. G. E. Murray, R. N. Costilow, E. N. Nester,W. A. Wood, N. R. Krieg, and G. B. Phillips (ed.). 1981. Manualof methods for general microbiology. American Society forMicrobiology, Washington, D.C.

15. 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.

16. Gibson, D. T., R. L. Roberts, M. C. Wells, and V. M. Kobal.1973. Oxidation of biphenyl by a Beijerinckia sp. Biochem.Biophys. Res. Commun. 50:211-219.

17. Gosselin, R. E., R. P. Smith, and H. C. Hodge. 1984. Clinicaltoxicity of commercial products, 5th ed., p. 11-189. TheWilliams & Wilkins Co., Baltimore.

18. Hankin, L., and B. L. Sawhney. 1984. Microbial degradation ofpolychlorinated biphenyls in soil. Soil Sci. 137:401-407.

19. Jerina, D. M., H. Selander, H. Yagi, M. C. Wells, J. F. Davey,V. Mahadevan, and D. T. Gibson. 1976. Dihydrodiols fromanthracene and phenanthrene. J. Am. Chem. Soc. 98:5988-5996.

20. Knuutinen, J. 1981. Thin layer chromatography of chlorinatedcatechols on a silica gel 60 layer. J. Chromatogr. 209:446-450.

21. Knuutinen, J., and I. 0. 0. Korhonen. 1983. Mass spectra ofchlorinated aromatics formed in pulp bleaching. I. Chlorinatedcatechols. Org. Mass Spectr. 18:438-444.

22. Kohler, H.-P. E., D. Kohler-Staub, and D. D. Focht. 1988.Degradation of 2-hydroxybiphenyl and 2,2'dihydroxybiphenylby Pseudomonas sp. strain HBP1. Appl. Environ. Microbiol.54:2683-2688.

23. Kroeckel, L., and D. D. Focht. 1987. Construction of chloroben-zene-utilizing recombinants by progenitive manifestation of arare event. Appl. Environ. Microbiol. 53:2470-2475.

24. Laborde, A. L., and D. T. Gibson. 1977. Metabolism of diben-zothiophene by a Beijerinckia sp. Appl. Environ. Microbiol.34:783-790.

25. Makary, M. H., and W. A. Brindley. 1983. Biphenyl hydroxy-lation and its induction differ between Montane moles and SwissWebster mice. Pestic. Biochem. Physiol. 19:23-30.

26. Paasivirta, J., R. Herzschuh, T. Humppi, E. Kantolahti, J.Knuutinen, M. Lahtipera, R. Laitinen, J. Salovaara, J. Tar-hanen, and L. Virkki. 1985. Pyrolysis products of PCBs. Envi-ron. Health Perspect. 60:269-278.

27. Smith, R. V., and P. J. Davis. 1980. Induction of xenobioticmono-oxygenases. Adv. Biochem. Eng. 14:61-100.

28. Smith, R. V., and J. P. Rosazza. 1974. Microbial models ofmammalian metabolism. Aromatic hydroxylation. Arch. Bio-chem. Biophys. 161:551-558.

29. Smith, B. V., and N. M. Waldron. 1980. Vogel's elementarypractical organic chemistry. vol. 1. Preparations, 3rd ed., p.319-320. Longman, London.

30. Spain, J. C., and D. T. Gibson. 1988. Oxidation of substitutedphenols by Pseudomonas putida Fl and Pseudomonas sp.strain JS6. Appl. Environ. Microbiol. 54:1399-1404.

31. Stahl, E. 1969. Thin layer chromatography. A laboratory hand-book. Springer-Verlag, New York.

32. Starr, M. P., H. Stolp, H. G. Trueper, A. Balows, and H. G.Schlegel (ed.). 1981. The prokaryotes. Springer-Verlag, NewYork.

33. Vettorazzi, G. 1979. International regulatory aspects for pesti-cide chemicals, vol. 1. Toxicity profiles, p. 79. CRC Press, Inc.,Boca Raton, Fla.

34. Voudrias, E. A., R. A. Larson, and V. L. Snoeylnk. 1985. Effectsof activated carbon on the reactions of free chlorine withphenols. Environ. Sci. Technol. 19:441-449.

35. Wiebkin, P., J. R. Fry, C. A. Jones, R. K. Lowing, and J. W.Bridges. 1978. Biphenyl metabolism in isolated rat hepatocytes:effect of induction and nature of the conjugates. Biochem.Pharmacol. 27:1899-1907.

36. Yamamoto, H., and H. Yoshimura. 1973. Metabolic studies onpolychlorinated biphenyls. Ill. Complete structure and acutetoxicity of the metabolites of 2,4,3',4'-tetrachlorobiphenyl.Chem. Pharm. Bull. 21:2237-2242.

37. Yeh, W.-K., D. T. Gibson, and T.-N. Liu. 1977. Toluenedioxygenase: a multicomponent system. Biochem. Biophys.Res. Commun. 78:401-410.

APPL. ENVIRON. MICROBIOL.

on June 16, 2017 by guesthttp://aem

.asm.org/

Dow

nloaded from