Vol. 56, No. 6APPLIED AND ENVIRONMENTAL MICROBIOLOGY, June 1990, p. 1858-18640099-2240/90/061858-07$02.00/0Copyright © 1990, American Society for Microbiology

Anaerobic Oxidation of Toluene, Phenol, and p-Cresol by theDisssimilatory Iron-Reducing Organism, GS-15

DEREK R. LOVLEY* AND DEBRA J. LONERGAN

Water Resources Division, U.S. Geological Survey, 430 National Center, Reston, Virginia 22092

Received 7 December 1989/Accepted 4 April 1990

The dissimilatory Fe(III) reducer, GS-15, is the first microorganism known to couple the oxidation ofaromatic compounds to the reduction of Fe(III) and the first example of a pure culture of any kind known toanaerobically oxidize an aromatic hydrocarbon, toluene. In this study, the metabolism of toluene, phenol, andp-cresol by GS-15 was investigated in more detail. GS-15 grew in an anaerobic medium with toluene as the soleelectron donor and Fe(IH) oxide as the electron acceptor. Growth coincided with Fe(III) reduction.[ring-.4C]toluene was oxidized to '4C02, and the stoichiometry of '4C02 production and Fe(III) reductionindicated that GS-15 completely oxidized toluene to carbon dioxide with Fe(III) as the electron acceptor.Magnetite was the primary iron end product during toluene oxidation. Phenol and p-cresol were alsocompletely oxidized to carbon dioxide with Fe(III) as the sole electron acceptor, and GS-15 could obtain energyto support growth by oxidizing either of these compounds as the sole electron donor. p-Hydroxybenzoate wasa transitory extracellular intermediate of phenol and p-cresol metabolism but not of toluene metabolism. GS-15oxidized potential aromatic intermediates in the oxidation of toluene (benzylalcohol and benzaldehyde) andp-cresol (p-hydroxybenzylalcohol and p-hydroxybenzaldehyde). The metabolism described here provides amodel for how aromatic hydrocarbons and phenols may be oxidized with the reduction of Fe(III) incontaminated aquifers and petroleum-containing sediments.

Aromatic hydrocarbons are among the most commongroundwater contaminants (28, 34, 36, 45). Under aerobicconditions, microorganisms can readily degrade monoaro-matic hydrocarbons such as benzene, xylenes, and toluene,and the pathways for this metabolism have been studiedintensively (10, 15). However, most groundwaters pollutedwith organic compounds are anaerobic. Geochemical evi-dence has indicated that aromatic hydrocarbons can beoxidized in anaerobic groundwater in which nitrate reduc-tion, methane production, or Fe(III) reduction is the termi-nal electron accepting process (21, 23, 28, 34; I. M. Cozza-relli, R. P. Eganhouse, and M. J. Baedecker, Environ. Geol.Water Sci., in press). Laboratory studies have indicated thatthe oxidation of aromatic hydrocarbons under these anaer-obic conditions is the result of microbial metabolism (17,21-23, 28, 44-47).Although anaerobic microorganisms capable of metaboliz-

ing aromatic acids and phenols under denitrifying, photosyn-thetic, sulfate-reducing, or methanogenic conditions havebeen described previously (for reviews, see references 5 and13), until recently there were no microorganisms in pureculture that were known to anaerobically oxidize aromatichydrocarbons. Two microbial isolates which can anaerobi-cally oxidize toluene have now been described. The dissim-ilatory Fe(III)-reducing microorganism, GS-15, was found togrow in an anaerobic medium with toluene as the electrondonor and a poorly crystalline Fe(III) oxide as the electronacceptor (23). The extent of carbon dioxide production in thecultures as well as the stoichiometry of carbon dioxideproduction and Fe(III) reduction indicated that GS-15 couldcompletely oxidize toluene to carbon dioxide with Fe(III) asthe sole electron acceptor. More recently, a Pseudomonassp. was found to oxidize [ring-'4C]toluene to '4Co2 withnitrate or N20 as the potential electron acceptor (46). It wasnot demonstrated whether the Pseudomonas sp. could ob-

* Corresponding author.

tain energy for growth from toluene oxidation, and the endproduct(s) of nitrate or N20 reduction was not determined.

In this report, we detail the growth and metabolism ofGS-15 on toluene as well as on two' other important aromaticcontaminants, p-cresol and phenol. These results indicatethat GS-15 may be a useful organism with which to studypotential mechanisms for the anaerobic oxidation of aro-matic hydrocarbons and to model the metabolism of aro-matic contaminants in the Fe(III)-reducing zone of sedimen-tary environments.

MATERIALS AND METHODSCulture conditions. As previously described (26), strict

anaerobic culturing and sampling techniques were usedthroughout. The basic growth medium for GS-15 was thesame as that previously described for growth of GS-15 onacetate (26). The medium contained (in grams per liter ofdeionized water): NaHCO3 (2.5), CaCI2 2H20 (0.1), KCI(0.1), NH4Cl (1.5), and NaH2PO4 H20 (0.6), as well as amixture of vitamins and trace minerals. The medium con-tained ca. 100 mmol of Fe(III) in the form of a poorlycrystalline Fe(III) oxide. This was synthesized, as previ-ously described (24), by neutralizing a solution of FeCl3 andcollecting the Fe(III) oxide precipitate. The gas phase wasN2-C02 (80:20). The pH was 6.7. In order to examine theformation of magnetite during toluene metabolism, withoutpotential interference of siderite formation (26), GS-15 wasgrown in a modified medium in which the NaHCO3 wasomitted and the gas phase was N2.The medium was dispensed in 10-ml volumes in anaerobic

pressure tubes or 80- or 100-ml volumes in 160-ml serumbottles. The medium was bubbled for at least 6 (pressuretubes) or 15 (serum bottles) min with the appropriate gasphase (N2-C02 or N2) to remove dissolved oxygen. Theculture vessels were then sealed with thick butyl rubberstoppers (Bellco Glass, Inc., Vineland, N.J.) and an alumi-num crimp. The medium was sterilized by autoclaving

1858

on January 17, 2020 by guesthttp://aem

.asm.org/

Dow

nloaded from

ANAEROBIC TOLUENE OXIDATION 1859

(121°C, 20 min). Unless noted, no reducing agent was added,because the Fe(II) that was transferred with the inoculumwas more than sufficient to remove any traces of oxygen thatmight not have been removed by sparging the medium (27).However, metabolism of toluene, phenol, and p-cresol was

also studied in the presence of added sulfide to furtherensure anaerobic conditions. In these instances, sodiumsulfide was added to sterilized medium from an anaerobicstock solution to provide a final concentration of 0.5 g ofsodium sulfide per liter.

Phenol, p-cresol, benzoate, benzylalcohol, benzaldehyde,p-hydroxybenzoate, and p-hydroxybenzaldehyde were add-ed to sterilized medium from anaerobic stock solutions (5 or50 mM) to provide an initial concentration of ca. 0.5 mM.Toluene was added to the medium with a microsyringe toprovide either 1 or 10 mmol of toluene per liter of medium.Because of the low solubility of p-hydroxybenzylalcohol, itwas added directly to the medium prior to bubbling andsterilization in order to provide an initial concentration of 0.5mM.

In order to test the ability of GS-15 to metabolize toluene,phenol, and p-cresol, an inoculum of GS-15 that had beengrown with benzoate as the electron donor and Fe(III) oxideas the electron acceptor was added to media containing thevarious electron donors. GS-15 was grown for nine transfers(10% inoculum) on toluene or p-cresol and for four transferson phenol prior to conducting the cell growth, Fe(III)reduction, and radiotracer studies reported here. All incuba-tions were at 30°C in the dark.To determine whether the reduction of Fe(III) with tolu-

ene resulted from the oxidation of the aromatic ring, GS-15was grown in the presence of [ring-'4C]toluene. Medium (80ml) was inoculated with 8 ml of a culture of GS-15 which hadbeen grown on toluene, and 8 p.l of [ring-`4C]toluene (10mCi/ml; Du Pont, Boston, Mass.) was added to initiallyprovide 1 mmol of toluene per liter of medium. As outlinedbelow, '4C02production was measured after 7 weeks ofincubation.

In order to measure '4C02 production from [U-'4C]phe-nol-, [U-4C]phenol (213 mCi/mmol; Sigma Chemical Co., St.Louis, Mo.) was dissolved in a solution of 50 mM phenol.The resultant solution contained 85 ,uCi per mmol of phenol.Dissolved oxygen was removed from the solution by bub-bling it with N2. The phenol solution (0.1 ml) was then addedto medium (10 ml) to provide an initial phenol concentrationof 0.5 mM.

Analytical techniques. Cell numbers were monitored withepifluorescence microscopy (18). Samples (1 ml) were anaer-

obically removed over time with a syringe and needle andwere fixed with glutaraldehyde (final concentration, 2.5%).The iron forms were dissolved with an acidic oxalate solu-tion as previously described (26), and the cells were treatedwith an acridine orange solution to give a final acridineorange concentration of 0.01%. After 2 min, the sample was

filtered onto a black Nuclepore filter (0.2-p.m pore diameter),and the filters were observed under oil immersion (x 1,000)with a Zeiss epifluorescence microscope. All fluorescentcells were counted.

Fe(III) reduction was monitored by measuring the accu-

mulation of Fe(II) over time. As previously described (25),subsamples were extracted in 0.5 N HCl for 15 min todissolve Fe(II) minerals, and Fe(II) was determined withferrozine.To measure aromatic compounds, a subsample of the

culture was filtered (Gelman filter; 0.45-p.m pore diameter)and the aromatics were separated on a SupelCosil LC-18

100

E

a.0

Io

0

80

60

40

20

0

0 1 0 20 30 40 50 60

50

40

30

20

0

10 E=L

0

Days of Incubation

FIG. 1. Cell growth (open symbols) and Fe(II) (closed symbols)when GS-15 was inoculated into medium with no added electrondonor (triangles) or either 10 (squares) or 1 (circles) mmol of tolueneadded per liter of medium.

column (3 cm x 4.6 mm; particle size diameter, 3 p.m;Supleco, Inc., Bellefonte, Pa.) with an eluant of 18 mMsodium acetate and 8% (vol/vol) acetonitrile at a flow rate ofeither 1 or 2 ml/min. The aromatics were detected with a

variable-wavelength UV detector set at 275 nm.To quantify the 14C02 produced from the metabolism of

[ring-14C]toluene, the bottles were connected to a flushingtrain in which N2 was flushed through the bottle, bubbledthrough a scintillation vial containing toluene (10 ml), andthen bubbled through three successive vials containing 10 mlof 0.4 N NaOH. The medium was acidified with 8 ml of 5 NH2SO4 to convert inorganic carbon forms to CO2. The

headspace and culture medium were sparged with N2 (20ml/min) for 30 min. Preliminary studies indicated that tolu-ene flushed from the culture bottles accumulated in the

tubing leading to the toluene trap and in the toluene trapitself and did not contaminate the NaOH traps. The 14C02 in

the NaOH traps was quantified by subsampling (3 ml) each

trap into Ecolume (15 ml; ICN Biomedical, Inc., Costa

Mesa, Calif.) and counting on a scintillation counter.14C02 production from [U-14C]phenol metabolism was

quantified in a manner similar to that for toluene except that

the toluene trap was omitted. Preliminary studies indicated

that the [U-14C]phenol was not flushed into the NaOH traps.

RESULTS

Growth of GS-15 coincided with Fe(III) reduction in

medium with toluene as the sole electron donor (Fig. 1).There was more Fe(III) reduction and cell growth in medium

that received 10 mmol of toluene per liter of medium than

there was in medium that received only 1 mmol of toluene

per liter of medium. However, growth and metabolism

lagged at the higher toluene concentration. There was no

Fe(III) reduction in medium not inoculated with GS-15 (datanot shown). When GS-15 was inoculated into medium which

did not contain added toluene, there was little growth or

Fe(III) reduction (Fig. 1).However, it was not possible to accurately measure the

loss. of toluene during these growth studies because of the

poor water solubility and high volatility of toluene as well as

significant absorption of toluene into the rubber stoppersduring the course of the incubation (17). Therefore, in order

to determine the stoichiometry of toluene oxidation coupledto Fe(III) reduction, 14CO2 and Fe(II) production were

monitored when GS-15 was grown on [ring-14C]toluene.

{-UCells-Toluene 10

p._-~ ,'~/ A,/ Fe(ll)-Toluene 10

_ 'p ~-' Cells-Toluene 1

/ - Fe(il)-Tolue/- i Cells and Fe(ll)-Toluene 0

VOL. 56, 1990

on January 17, 2020 by guesthttp://aem

.asm.org/

Dow

nloaded from

1860 LOVLEY AND LONERGAN

TABLE 1. Oxidation of [ring-14C]toluene and Fe(III) reduction by GS-15

Addition(s) to toluene-Fe(III) 14CO d Toluene trap Culture medium Toluene oxidized Fe(II) producedoxide mediuma 2 (dpm) 14C (dpm)b 14C (dpm) (Lmol)c (,umol)

GS-15 101,652 ± 6,317d 1,991 ±13d 12,160 ± 5,866 45.79 1,707 ± 276dGS-15 and Na2S 93,092 ± 10,423 2,692 ± 91 17,600 ± 554 41.93 1,751 + 215None 269 + 114 7,217 ± 2,021 8,640 ± 7,498 0.12 0GS-15 [Fe(III) oxide omitted] 6,518 ± 738 6,224 ± 599 8,320 ± 8,658 2.93 70 ± 63

a Each culture bottle (80 ml of medium) received 8 p.l (177,600 dpm) of [ring-14C]toluene to initially provide 1 mmol of toluene per liter of medium. Except wherenoted, each bottle also contained ca. 100 mmol of Fe(III) per liter of medium.

b 14C recovered in the vial of toluene through which the N2 exiting the culture bottles was bubbled prior to bubbling through the NaOH traps for 14 CO2.Calculated from '4Co2 produced and the specific activity of the [ring-14C]toluene of 1 ,Ci per mmol.

d Mean + standard deviation; n = 3.

GS-15 oxidized the [ring-14C]toluene to '4CO2 (Table 1).The ratio of Fe(II) produced to toluene oxidized to carbondioxide closely approximated the ratio of 36 that was ex-

pected for the complete oxidation of toluene to carbondioxide with Fe(III) as the sole electron acceptor (Table 2,reaction 1). In these studies, the Fe(II) transferred with theinoculum served as a reductant to ensure that the mediumwas anaerobic (27). Furthermore, toluene was oxidized tocarbon dioxide in a similar manner in a medium to whichsulfide had also been added as a reductant (Table 1). Toluenewas not oxidized in the absence of GS-15 (Table 1). WhenGS-15 was inoculated into a medium that contained toluenebut no Fe(III) oxide, there was only a slight oxidation oftoluene, which was in proportion to the small amount ofFe(III) that was carried over with the inoculum and subse-quently reduced (Table 1).

In the toluene cultures that were inoculated with GS-15and contained Fe(III), 55% of the label added as [ring-14C]toluene was recovered as 14CO2 and 10% was recoveredin the toluene trap or the culture medium (Table 1). In theuninoculated treatment, only 9% of the added label wasrecovered in the carbon dioxide traps, toluene trap, or themedium. It is assumed that the label not recovered in themedium or the traps was absorbed into the stoppers as

rubber stoppers will absorb toluene over time (17).Growth of GS-15 on toluene resulted in the production of

copious quantities of a magnetic mineral (Fig. 2), whichX-ray diffraction analysis identified as magnetite.GS-15 metabolized phenol with the concomitant reduction

of Fe(III), and this metabolism was associated with cellgrowth (Fig. 3). Low concentrations of p-hydroxybenzoateaccumulated and then were metabolized. After three cul-tures had metabolized an initial phenol concentration of 0.42mM, the ratio of phenol oxidized to Fe(II) produced was 29.1

1.3 (mean ± standard deviation, n = 3). This compares

with 28 mol of Fe(II) theoretically reduced during the

oxidation of phenol to carbon dioxide (Table 2, reaction 2).Oxidation of phenol to carbon dioxide was confirmed with[U-'4C]phenol (Table 3) and within the errors of the mea-

surements was consistent with that in reaction 2 (Table 2).Phenol was oxidized in a similar manner in medium whichcontained sulfide as an added reductant. There was no

Fe(III) reduction (Table 3), loss of phenol (data not shown),or significant oxidation of phenol (Table 3) when the mediumwas not inoculated with GS-15. In contrast to the studieswith [ring-'4C]toluene, the label in '4CO2 and the labelremaining in the culture medium after flushing out the 14CO2accounted for 90% or more of the amount of label added inall treatments (Table 3). The label remaining in the mediumwas presumably primarily in the form of unmetabolizedphenol or p-hydroxybenzoate.GS-15 also metabolized p-cresol with the reduction of

Fe(III), and this metabolism yielded energy to supportgrowth (Fig. 4A). p-Hydroxybenzoate accumulated duringp-cresol metabolism (Fig. 4A) but was further metabolizedwith extended incubation (data not shown). Medium whichcontained sulfide had similar rates of p-cresol metabolismand Fe(III) reduction (Fig. 4B). There was no loss ofp-cresolover time if the p-cresol-Fe(III) oxide medium was notinoculated with GS-15 or if the Fe(III) oxide was omittedfrom the inoculated medium (Fig. 4B).A stoichiometry for p-cresol oxidation to carbon dioxide

coupled to Fe(III) reduction could be approximated from thedata shown in Fig. 4 and the corresponding duplicates ofeach treatment. After 7 days of incubation for the fourreplicates, 0.34 + 0.01 mmol (mean + standard deviation, n

= 4) of p-cresol was metabolized, 0.11 + 0.05 mmol ofp-hydroxybenzoate had accumulated, and 9.1 + 1.0 mmol ofFe(II) had been produced per liter. The incomplete oxidationof a mole of p-cresol to p-hydroxybenzoate is expected toresult in the reduction of 6 mol of Fe(III) (Table 2, reaction7). When the moles of p-cresol metabolized and Fe(III)

TABLE 2. Stoichiometry and standard free energies of reactions related to toluene, phenol, and p-cresol metabolism by GS-15

Reaction no. Reactants Products AGo'(k/reaction)a

1 Toluene + 36Fe3+ + 21H20 36Fe2+ + 7HC03- + 43H+ -3,6302 Phenol + 28Fe3+ + 17H20 28Fe2+ + 6HC03- + 34H+ -2,8773 p-Cresol + 34Fe3+ + 20H20 34Fe2+ + 7HC03- + 41H+ -3,4924 Toluene + 108Fe(OH)3 36Fe3O4 + 7HC03 + 7H+ + 159H20 -3,1745 Acetate- + 8Fe3+ + 4H20 8Fe2+ + 2HC03- + 9H+ -8096 Phenol + HC03- p-Hydroxybenzoate + H20 -267 p-Cresol + 6Fe3+ + 2H20 p-Hydroxybenzoate- + 6Fe2+ + 7H+ -6418 p-Hydroxybenzoate- + 2Fe2+ + 2H+ Benzoate- + 2Fe3+ + H20 +3559 p-Hydroxybenzoate + H2 Benzoate- + H20 +127

a Free energy calculated from the standard free energies of the reactants and products (from reference 41), except for values for iron oxides (from reference38), and by assuming standard conditions except for pH 7.

APPL. ENVIRON. MICROBIOL.

on January 17, 2020 by guesthttp://aem

.asm.org/

Dow

nloaded from

ANAEROBIC TOLUENE OXIDATION 1861

I

t-

0rv

E=L

10

5

0

0 10 20

Days of Incubation

FIG. 2. Magnetite accumulation in a culture of GS-15 grown inphosphate-buffered medium, with toluene as the sole electron donorand Fe(III) oxide as the electron acceptor.

reduced associated with the accumulated p-hydroxyben-zoate were subtracted from the total amount of Fe(III)reduced and p-cresol metabolized, the ratio of Fe(III) re-

duced to p-cresol metabolized was 37. This is only slightlyhigher than the theoretical ratio of 34 for the completeoxidation ofp-cresol to carbon dioxide (Table 2, reaction 3).GS-15 could metabolize a number of potential aromatic

intermediates in the oxidation of toluene, phenol, or cresol.These included benzoate (23) as well as benzylalcohol,benzaldehyde, p-hydroxybenzoate, p-hydroxybenzylalco-hol, and p-hydroxybenzaldehyde (data not shown). GS-15could obtain energy for growth from the oxidation of each ofthese compounds as evidenced by continued metabolism ofthe compounds after repeated sequential transfers.

DISCUSSION

GS-15 provides the first example of an organism in pure

culture which can anaerobically oxidize an aromatic hydro-carbon, toluene (23). The [ring-14C]toluene experimentsreported here confirm that, with Fe(III) as the electronacceptor, GS-15 can oxidize the aromatic ring of toluene tocarbon dioxide under strict anaerobic conditions. This me-

tabolism was found to yield energy to support cell growth.The results also demonstrate that GS-15 can obtain energyfor growth by completely oxidizing phenol and p-cresol tocarbon dioxide, with Fe(III) reduction as the electron ac-

cepting process.

FIG. 3. Phenol (A), Fe(II) (0), cell number (*), and p-hydroxy-benzoate (x) when GS-15 was inoculated into medium with phenol as

the sole electron donor and Fe(III) oxide as the electron acceptor.

Radiotracer studies and measurements of substrate uptakeand Fe(III) reduction suggest that the complete oxidation oftoluene, phenol, and p-cresol to carbon dioxide proceededaccording to the reactions in Table 2 (reactions 1 to 3). Ofcourse, the actual metabolism of these compounds is morecomplicated than the equations indicate because some of theorganic carbon may be incorporated into cells, and Fe(III)and Fe(II) exist in various forms (26). Thermodynamiccalculations indicate that these are all energetically favorablereactions that can potentially yield energy to support cellgrowth. This is true whether soluble or solid iron forms are

considered to be the reactants and products (Table 2, com-pare reactions 1 and 4). It is not surprising that GS-15 canoxidize toluene under anaerobic conditions when it is con-sidered that, per Fe(III) reduced, the oxidation of toluenehas a potential energy yield comparable to acetate oxidationcoupled to Fe(III) reduction (Table 2, reaction 5). Thenumber of cells produced per Fe(III) reduced with thearomatic compounds was at least as high as the ca. 1.5 x 106cells produced per Fe(III) reduced that was previouslyobserved during acetate metabolism with Fe(III) oxide as theelectron acceptor (26). However, with aromatics, the rate ofFe(III) reduction is typically slower, with a longer initial lagperiod than with acetate. Detailed comparisons of physio-logical parameters during growth of GS-15 on various or-

ganic acids and aromatic compounds are needed.Initial steps in phenol, p-cresol, and toluene metabolism.

GS-15 oxidizes benzoate to carbon dioxide (23) without theaccumulation of extracellular aromatic or fatty acid interme-diates (unpublished data), and no accumulation of aromaticintermediates has been detected during toluene metabolism(23). However, p-hydroxybenzoate temporarily accumu-lated during growth on phenol or p-cresol. This suggests thatthe first step in phenol metabolism is carboxylation of the

TABLE 3. Oxidation of [U-'4C] phenol and Fe(III) reduction by GS-15

Addition(s) to phenol-Fe(III) 14 Culture medium Phenol oxidized Fe(II) producedoxide mediuma CO2 (dpm) 14C (dpm) (>mol)b (,mol)

GS-15 466,299 ± 77,986c 418,400 + 88,221c 2.47 99 ± 18cGS-15 + Na2S 440,528 + 84,986 407,720 ± 69,054 2.33 75 ± 19None 1,705 ± 195 999,520 ± 104,815 0.01 0GS-15 [Fe(III) oxide omitted] 10,184 ± 4,803 934,280 ± 50,695 0.05 0

a Each culture tube (10 ml of medium) received 943,500 dpm of [U-_4C]phenol to initially provide 0.5 mM phenol. Except where noted, each tube also containedca. 100 mmol of Fe(III) per liter of medium.

b Calculated from "4CO2 produced and the specific activity of the [U-_4C]phenol of 85 ,Ci per mmol.c Mean + standard deviation; n = 3.

E

0.50 :0c

100.25 -IF

C

0.00 0.i

VOL. 56, 1990

on January 17, 2020 by guesthttp://aem

.asm.org/

Dow

nloaded from

1862 LOVLEY AND LONERGAN

AE0

a.S

=t-

0

L-0

00E±L

10

B0.5 2E

0.4

0

0.3 =>01

0.2.0

0.1 .

x

2-a

o

0.0

12

IEL

I-.G-0

0

E±L

10

0 2 4 6 8

0.6 aE

0.5 _a0

0.4 @0

0.3°

0.2 ^0Zo

0.1 00

0.0 ?0.

Days of Incubation Days of Incubation

FIG. 4. (A) p-Cresol (-), Fe(II) (0), cell number (*), and p-hydroxybenzoate (x) when GS-15 was inoculated into medium with p-cresolas the sole electron donor. (B) p-Cresol concentrations over time in uninoculated medium (A) or in medium inoculated with GS-15 with theFe(III) oxide omitted (O), as well as p-cresol (A), Fe(II) (0), and p-hydroxybenzoate (x) when GS-15 was inoculated into a medium withp-cresol as the sole electron donor and sulfide added as an extra reductant.

aromatic ring and that the methyl group of p-cresol isoxidized prior to metabolism of the aromatic ring (Fig. 5).These potential pathways are the same as those previouslyproposed for phenol and p-cresol metabolism in denitrifyingmicroorganisms (6-8, 42). p-Hydroxybenzoate also ap-peared to be an intermediate ofp-cresol oxidation in slurriesof aquifer sand in which p-cresol was oxidized to carbondioxide with the reduction of sulfate (37). p-Hydroxyben-zoate was not reported as an extracellular intermediateduring phenol or p-cresol oxidation coupled to sulfate reduc-tion by Desulfobacterium phenolicum, but this organism wasfound to metabolize p-hydroxybenzoate (2). In methano-genic consortia metabolizing phenol, phenol is carboxylatedand dehydroxylated to yield benzoate (14, 20). p-Hydroxy-benzoate was not identified as a free intermediate and, in oneconsortium, added p-hydroxybenzoate was decarboxylatedto phenol (20).

It has been proposed for denitrifying cultures (7, 8, 42) andsulfate-reducing sediments (37) that the oxidation ofp-cresolto p-hydroxybenzoate proceeded via p-hydroxybenzylal-cohol and p-hydroxybenzaldehyde (Fig. 5). The finding thatGS-15 can also oxidize p-hydroxybenzylalcohol and p-hy-droxybenzaldehyde demonstrates that GS-15 could poten-tially metabolize p-cresol by a similar pathway. However,neither of these potential intermediates has been observedduring p-cresol metabolism. It has been proposed that deni-trifiers reduce p-hydroxybenzoate to benzoate prior to me-tabolizing the aromatic ring (13, 31, 40). If GS-15 has asimilar metabolism, then the accumulation ofp-hydroxyben-zoate during phenol and p-cresol metabolism may be be-cause the steps leading to p-hydroxybenzoate (Table 2,reactions 6 and 7) are energetically favorable whereas thereduction of p-hydroxybenzoate to benzoate is not energet-ically favorable with potentially available extracellular re-ductants such as Fe2+ or H2 (Table 2, reactions 8 and 9).Anaerobic toluene oxidation has been previously ob-

served under mixed culture conditions in which denitrifica-tion (22, 47) or methane production (17, 44, 45) was theterminal electron accepting process. Furthermore, after thecompletion of the studies reported here, a Pseudomonas sp.was reported to oxidize [ring-14C]toluene to 14CO2 (46) whennitrate or N20 was provided as a potential electron acceptor.Two potential pathways for toluene oxidation under denitri-fying or methanogenic conditions have been hypothesized

(17, 22). One proposed pathway is hydroxylation of thearomatic ring to form p-cresol, with subsequent oxidation ofthe p-cresol to p-hydroxybenzoate (Fig. 5). Although thelack of p-hydroxybenzoate accumulation during toluenemetabolism by GS-15 suggests that toluene metabolism doesnot involve hydroxylation of the aromatic ring, it is prema-ture to rule out this potential pathway. Even if p-hydroxy-benzoate is an intermediate, it might not accumulate duringthe slow metabolism of toluene as it does during the fastermetabolism ofp-cresol and phenol. The alternative potentialpathway for toluene metabolism is the sequential oxidationof the methyl group to form benzylalcohol, benzaldehyde,and then benzoate (Fig. 5). GS-15 can readily metabolize allof these potential intermediates. However, this does notdemonstrate that these compounds are, in fact, intermedi-ates in toluene metabolism. Further studies to elucidate thepathways for anaerobic toluene oxidation are required.Environmental implications. Although most pristine shal-

low aquifers are aerobic, when they are contaminated withorganics, anaerobic conditions typically develop (1, 39).When soils or freshwater sediments become anaerobic,Fe(III) is often the most abundant potential electron accep-tor for microbial metabolism (9, 33, 43). Studies in an aquiferpolluted with aromatic hydrocarbons have demonstratedthat the microbial oxidation of aromatic contaminants cou-pled to Fe(III) reduction can be an important process for theremoval of aromatic contaminants from groundwater (23).Such Fe(III) reduction appears to be a widespread phenom-enon in other aquifers contaminated with aromatics, asevidenced by the accumulation of dissolved Fe(II) (11, 36;I. M. Cozzarelli, M. J. Baedecker, J. A. Hopple, and B. J.Franks, Abstr. Geol. Soc. Am. 1988, 19, p. 629) or theleaching of sediment Fe(III) oxides (F. Chapelle, personalcommunication). Aromatic hydrocarbons, phenols, andcresols are among the most prevalent aromatic contaminantsin groundwater (3, 4, 11, 16, 19, 34-36, 45). Thus, themetabolism of GS-15 provides a model mechanism for theoxidation of several important aromatic contaminants cou-pled to Fe(III) reduction in sedimentary environments.

Fe(III)-reducing microorganisms may also play an impor-tant role in sediments in which hydrocarbons are a naturalcomponent. Geological evidence (29, 32) suggests that re-duction of Fe(III) in the presence of hydrocarbons is acommon phenomenon in sedimentary environments. The

APPL. ENVIRON. MICROBIOL.

on January 17, 2020 by guesthttp://aem

.asm.org/

Dow

nloaded from

ANAEROBIC TOLUENE OXIDATION 1863

CH3

benzylalcohol

. TOLUENE 2Fe+3benzaldehyde

'I

2Fe+3

p-hydroxybenzylalcohol

p-hydroxybenzaldehyde

3OFe+2

-CO2Or 3OFe+3

BENZOATE

p-hydroxybenzoate

2Fe+2COOH

[2H]

PHENOL

HCO3

OHFIG. 5. Potential pathways for the oxidation of toluene, p-cresol, and phenol coupled to Fe(III) reduction. GS-15 can oxidize all of the

compounds shown. However, compounds in brackets have not been observed as extracellular intermediates during growth on otheraromatics. The proposed pathways are based on those proposed for pure cultures of denitrifiers (7, 8, 42) and mixed culture systems in whichnitrate reduction (22), sulfate reduction (37), or methane production (17) was the terminal electron accepting process, as well as the resultspresented here for GS-15. However, none of the pathways for the metabolism of these compounds by GS-15 has been unequivocallydemonstrated, and the pathways are shown merely as an aid in following the discussion.

accumulation of magnetite around hydrocarbon seeps hasbeen attributed to microbial oxidation of hydrocarbon com-ponents coupled to Fe(III) reduction (12, 27, 30). Theproduction of copious quantities of magnetite during theoxidation of toluene by GS-15 lends support to this hypoth-esis.

In summary, the results presented here document that,under strict anaerobic conditions, GS-15 can completelyoxidize toluene, phenol, and p-cresol to carbon dioxide withFe(III) as the sole electron acceptor. Previous studies havedemonstrated that, in addition to the reduction of Fe(III) toFe(II), GS-15 can oxidize acetate with the reduction ofMn(IV) to Mn(II) or nitrate to ammonia (26). Althoughdetailed stoichiometric studies have not been conducted,preliminary studies have demonstrated that nitrate orMn(IV) may also serve as the electron acceptor for benzoateor toluene oxidation (E. J. P. Phillips and D. R. Lovley,unpublished data). Further studies on aromatic metabolismby GS-15 are expected to contribute to an understanding ofthe oxidation of aromatic contaminants under anaerobic

conditions and to aid in the development of bioremediationstrategies for contaminated groundwater.

ACKNOWLEDGMENTS

We thank Elizabeth Phillips for technical assistance and RonOremland and Isabelle Cozzarelli for helpful suggestions on themanuscript.

This study was supported by the U.S. Geological Survey ToxicWaste-Ground-Water Contamination Program.

LITERATURE CITED1. Baedecker, M. J., and W. Back. 1979. Modem marine sediments

as a natural analog to the chemically stressed environment of a

landfill. J. Hydrol. (Amst.) 43:393-414.2. Bak, F., and F. Widdel. 1986. Anaerobic degradation of phenol

and phenol derivatives by Desulfobacterium phenolicum sp.nov. Arch. Microbiol. 146:177-180.

3. Barber, L. B., II, E. M. Thurman, and M. P. Schroder. 1988.Long-term fate of organic micropollutants in sewage-contami-nated groundwater. Environ. Sci. Technol. 22:205-211.

4. Barker, J. F., G. C. Patrick, and D. Major. 1987. Natural

P-CRESOL

2Fe-2

VOL. 56, 1990

on January 17, 2020 by guesthttp://aem

.asm.org/

Dow

nloaded from

1864 LOVLEY AND LONERGAN

attenuation of aromatic hydrocarbons in a shallow sand aquifer.Ground Water Monit. Rev. 7:64-71.

5. Berry, D. F., A. J. Francis, and J.-M. Boliag. 1987. Microbialmetabolism of homocyclic and heterocyclic aromatic com-pounds under anaerobic conditions. Microbiol. Rev. 51:43-59.

6. Bossert, I. D., M. D. Rivera, and L. Y. Young. 1986. p-Cresolbiodegradation under denitrifying conditions: isolation of abacterial coculture. FEMS Microbiol. Ecol. 38:313-319.

7. Bossert, I. D., G. Whited, D. T. Gibson, and L. Y. Young. 1989.Anaerobic oxidation of p-cresol mediated by a partially purifiedmethylhydroxylase from a denitrifying bacterium. J. Bacteriol.171:2956-2962.

8. Bossert, I. D., and L. Y. Young. 1986. Anaerobic oxidation ofp-cresol by a denitrifying bacterium. Appl. Environ. Microbiol.52:1117-1122.

9. Bostrom, K. 1967. Some pH-controlling redox reactions innatural waters, p. 286-311. In R. E. Gould (ed.), Equilibriumconcepts in natural waters. American Chemical Society, Wash-ington, D.C.

10. Dagley, S. 1984. Microbial degradation of aromatic compounds.Dev. Ind. Microbiol. 25:53-65.

11. Ehrlich, G. G., E. M. Godsy, D. F. Goerlitz, and M. F. Hult.1983. Microbial ecology of a creosote-contaminated aquifer atSt. Louis Park, Minnesota. Dev. Ind. Microbiol. 24:235-245.

12. Elmore, R. D., M. H. Engel, L. Crawford, K. Nick, S. Imbus,and Z. Sofer. 1987. Evidence for a relationship between hydro-carbons and authigenic magnetite. Nature (London) 325:428-430.

13. Evans, W. C., and G. Fuchs. 1988. Anaerobic degradation ofaromatic compounds. Annu. Rev. Microbiol. 42:289-317.

14. Genthner, B. R. S., G. T. Townsend, and P. J. Chapman. 1989.Anaerobic transformation of phenol to benzoate via para-carboxylation: use of fluorinated analogues to elucidate themechanism of transformation. Biochem. Biophys. Res. Com-mun. 162:945-951.

15. Gibson, D. T., and V. Subramanian. 1984. Microbial degrada-tion of aromatic hydrocarbons, p. 181-252. In D. T. Gibson(ed.), Microbial degradation of organic compounds. MarcelDekker, Inc., New York.

16. Goerlitz, D. F., D. D. Troutman, E. M. Godsy, and B. J. Franks.1985. Migration of wood-preserving chemicals in contaminatedgroundwater in a sand aquifer at Pensacola, Florida. Environ.Sci. Technol. 19:955-961.

17. Grbic-Galic, D., and T. M. Vogel. 1987. Transformation oftoluene and benzene by mixed methanogenic cultures. Appl.Environ. Microbiol. 53:254-260.

18. Hobbie, J. E., R. J. Daley, and S. Jasper. 1977. Use ofNuclepore filters for counting bacteria by fluorescence micros-copy. Appl. Environ. Microbiol. 33:1225-1228.

19. Humenick, M. J., and C. F. Mattox. 1978. Groundwater pollu-tants from underground coal gasification. Water Res. 12:463-469.

20. Knoll, G., and J. Winter. 1989. Degradation of phenol viacarboxylation to benzoate by a defined, obligate syntrophicconsortium of anaerobic bacteria. Appl. Microbiol. Biotechnol.30:318-324.

21. Kuhn, E. P., P. J. Colberg, J. L. Schnoor, 0. Wanner, A. J. B.Zehnder, and R. P. Schwarzenbach. 1985. Microbial transforma-tions of substituted benzenes during infiltration of river water togroundwater: laboratory column studies. Environ. Sci. Tech-nol. 19:961-968.

22. Kuhn, E. P., J. Zeyer, P. Eicher, and R. P. Schwarzenbach.1988. Anerobic degradation of alkylated benzenes in denitrify-ing laboratory aquifer columns. Appl. Environ. Microbiol. 54:490-496.

23. Lovley, D. R., M. J. Baedecker, D. J. Lonergan, I. M. Cozzarelli,E. J. P. Phillips, and D. I. Siegel. 1989. Oxidation of aromaticcontaminants coupled to microbial iron reduction. Nature (Lon-don) 339:297-299.

24. Lovley, D. R., and E. J. P. Phillips. 1986. Organic mattermineralization with the reduction of ferric iron in anaerobicsediments. Appl. Environ. Microbiol. 51:683-689.

25. Lovley, D. R., and E. J. P. Phillips. 1986. Availability of ferriciron for microbial reduction in bottom sediments of the fresh-water tidal Potomac River. Appl. Environ. Microbiol. 52:751-757.

26. Lovley, D. R., and E. J. P. Phillips. 1988. Novel mode ofmicrobial energy metabolism: organic carbon oxidation coupledto dissimilatory reduction of iron or manganese. Appl. Environ.Microbiol. 54:1472-1480.

27. Lovley, D. R., J. F. Stolz, G. L. Nord, Jr., and E. J. P. Phillips.1987. Anaerobic production of magnetite by a dissimilatoryiron-reducing microorganism. Nature (London) 330:252-254.

28. Major, D. W., C. I. Mayfield, and J. F. Barker. 1987. Biotrans-formation of benzene by denitrification in aquifer sand. GroundWater 26:8-14.

29. McBride, E. F. 1974. Significance of color in red, green, purple,olive, brown, and gray beds of difunta group, NortheasternMexico. J. Sediment. Petrol. 44:760-773.

30. McCabe, C., R. Sassen, and B. Saffer. 1987. Occurrence ofsecondary magnetite within biodegraded oil. Geology 15:7-10.

31. Nozawa, T., and Y. Maruyama. 1988. Anaerobic metabolism ofphthalate and other aromatic compounds by a denitrifyingbacterium. J. Bacteriol. 170:5778-5784.

32. Picard, M. D. 1965. Iron oxides and fine-grained rocks of RedPeak and Crow Mountain sandstone members, Chugwater (Tri-assic) Formation, Wyoming. J. Sediment. Petrol. 35:464-479.

33. Ponnamperuma, F. N. 1972. The chemistry of submerged soils.Adv. Agron. 24:29-96.

34. Reinhard, M., N. L. Goodman, and J. F. Barker. 1984. Occur-rence and distribution of organic chemicals in two landfillleachate plumes. Environ. Sci. Technol. 18:953-961.

35. Sawhney, B. L., and R. P. Kozloski. 1984. Organic pollutants inleachates from landfill sites. J. Environ. Qual. 13:349-352.

36. Schwille, F. 1976. Anthropogenically reduced groundwaters.1976. Hydrol. Sci. Bull. 21:629-645.

37. Smolenski, W. J., and J. M. Suflita. 1987. Biodegradation ofcresol isomers in anoxic aquifers. Appl. Environ. Microbiol.53:710-716.

38. Stumm, W., and J. J. Morgan. 1981. Aquatic chemistry. JohnWiley & Sons, Inc., New York.

39. Suflita, J. M., S. A. Gibson, and R. E. Beeman. 1988. Anaerobicbiotransformations of pollutant chemicals in aquifers. J. Ind.Microbiol. 3:179-194.

40. Taylor, B. F., W. L. Campbell, and I. Chinoy. 1970. Anaerobicdegradation of the benzene nucleus by a facultatively anaerobicmicroorganism. J. Bacteriol. 102:430-437.

41. Thauer, R. K., K. Jungermann, and K. Dekker. 1977. Energyconservation in chemotropic anaerobic bacteria. Bacteriol. Rev.41:100-180.

42. Tschech, A., and G. Fuchs. 1987. Anaerobic degradation ofphenol by pure cultures of newly isolated denitrifyingpseudomonads. Arch. Microbiol. 148:213-217.

43. Van Breeman, N. 1988. Effects of seasonal redox processesinvolving iron on the chemistry of periodically reduced soils, p.797-809. In J. W. Stucki, B. A. Goodman, and U. Schwertmann(ed.), Iron in soils and clay minerals. D. Reidel PublishingCompany, Boston.

44. Vogel, T. M., and D. Grbic-Galic. 1986. Incorporation of oxygenfrom water into toluene and benzene during anaerobic fermen-tative transformation. Appl. Environ. Microbiol. 52:200-202.

45. Wilson, B. H., G. B. Smith, and J. F. Rees. 1986. Biotransfor-mations of selected alkylbenzenes and halogenated aliphatichydrocarbons in methanogenic aquifer material: a microcosmstudy. Environ. Sci. Technol. 20:997-1002.

46. Zeyer, J., P. Eicher, J. Dolfing, and R. P. Schwarzenbach. 1990.Anaerobic degradation of aromatic hydrocarbons, p. 33-40. InD. Kamely, A. Chakrabarty, and G. S. Omenn (ed.), Biotech-nology and biodegradation. Gulf Publishing Company, Hous-ton.

47. Zeyer, J., E. P. Kuhn, and R. P. Schwarzenbach. 1986. Rapidmicrobial mineralization of toluene and 1,3-dimethylbenzene inthe absence of molecular oxygen. Appl. Environ. Microbiol.52:944-947.

APPL. ENVIRON. MICROBIOL.

on January 17, 2020 by guesthttp://aem

.asm.org/

Dow

nloaded from