identification of enzymes involved in anaerobic benzene ... · identiﬁcation of enzymes involved...

Identification of enzymes involved in anaerobicbenzene degradation by a strictly anaerobiciron-reducing enrichment cultureemi_2248 2783..2796

Nidal Abu Laban,1 Draženka Selesi,1 Thomas Rattei,2

Patrick Tischler2 and Rainer U. Meckenstock1*1Helmholtz Zentrum München – German ResearchCenter for Environmental Health, Institute ofGroundwater Ecology, Ingolstädter Landstraße 1,D-85764, Neuherberg, Germany.2Chair for Genome-oriented Bioinformatics, TechnischeUniversität München, Life and Food Science CenterWeihenstephan, Am Forum 1, D-85354Freising-Weihenstephan, Germany.

Summary

Anaerobic benzene degradation was studied with ahighly enriched iron-reducing culture (BF) composedof mainly Peptococcaceae-related Gram-positivemicroorganisms. The proteomes of benzene-, phenol-and benzoate-grown cells of culture BF were com-pared by SDS-PAGE. A specific benzene-expressedprotein band of 60 kDa, which could not be observedduring growth on phenol or benzoate, was subjectedto N-terminal sequence analysis. The first 31 aminoacids revealed that the protein was encoded by ORF138 in the shotgun sequenced metagenome of cultureBF. ORF 138 showed 43% sequence identity to phe-nylphosphate carboxylase subunit PpcA of Aroma-toleum aromaticum strain EbN1. A LC/ESI-MS/MS-based shotgun proteomic analysis revealed otherspecifically benzene-expressed proteins with encod-ing genes located adjacent to ORF 138 on the metage-nome. The protein products of ORF 137, ORF 139and ORF 140 showed sequence identities of 37% tophenylphosphate carboxylase PpcD of A. aromaticumstrain EbN1, 56% to benzoate-CoA ligase (BamY) ofGeobacter metallireducens and 67% to 3-octaprenyl-4-hydroxybenzoate carboxy-lyase (UbiD/UbiX) of A.aromaticum strain EbN1 respectively. These genesare proposed as constituents of a putative ben-zene degradation gene cluster (~17 kb) composedof carboxylase-related genes. The identified gene

sequences suggest that the initial activation reactionin anaerobic benzene degradation is probably a directcarboxylation of benzene to benzoate catalysed byputative anaerobic benzene carboxylase (Abc). Theputative Abc probably consists of several subunits,two of which are encoded by ORFs 137 and 138, andbelongs to a family of carboxylases including phe-nylphosphate carboxylase (Ppc) and 3-octaprenyl-4-hydroxybenzoate carboxy-lyase (UbiD/UbiX).

Introduction

The aromatic hydrocarbon benzene is a compo-nent of crude oil and gasoline and is intensively usedin the chemical industry. Due to its high solubility(1.78 g l-1 in water at 25°C), chemical stability(DGf° = +124.4 KJ mol-1), and impact on human health,benzene is considered as one of the most hazardouspollutants of groundwater and sediments (Johnsonet al., 2003). Therefore, it is of great interest to under-stand the fate and the biodegradability of benzene in theenvironment.

For aerobic degradation of benzene well studiedpathways are known (Gibson et al., 1968). Furthermore,enrichment cultures and microcosm studies indicatedthat benzene can be effectively degraded in the absenceof molecular oxygen under iron-reducing (Lovley et al.,1996; Anderson et al., 1998; Rooney-Varga et al., 1999;Kunapuli et al., 2008), sulfate-reducing (Lovley et al.,1995; Kazumi et al., 1997; Phelps et al., 1998; Caldwelland Suflita, 2000; Abu Laban et al., 2009), denitrifying(Burland and Edwards, 1999; Coates et al., 2001; Ulrichand Edwards, 2003) and methanogenic (Ulrich andEdwards, 2003; Chang et al., 2005) conditions. So far,only two anaerobic benzene-degrading strains were iso-lated to purity, which were described as denitrifyingmembers of the genera Dechloromonas and Azoarcus(Coates et al., 2001; Kasai et al., 2006). Despite thecultivation of benzene-degrading cultures, a profoundknowledge of the biochemical mechanism and theenzymes and genes of anaerobic benzene degradation isstill lacking.

In the presence of molecular oxygen, benzene isactivated by oxygenases introducing one or two oxygen

Received 13 October, 2009; accepted 5 April, 2010. *For Correspon-dence. E-mail: [email protected]; Tel.(+49) 89 3187 2561; Fax (+49) 89 3187 3361.

Environmental Microbiology (2010) 12(10), 2783–2796 doi:10.1111/j.1462-2920.2010.02248.x

© 2010 Society for Applied Microbiology and Blackwell Publishing Ltd

mailto:[email protected]

atoms from O2 yielding phenol, benzene epoxide orcatechol respectively (Bugg, 2003). Under anaerobicconditions, the reactive oxygen is missing and degrada-tion requires other biochemical reactions to overcomethe extreme chemical stability of benzene. Examples ofanaerobic activation reactions of hydrocarbons includeoxygen-independent hydroxylation of alkyl substituent via,e.g. ethylbenzene dehydrogenase (Rabus and Widdel,1995; Ball et al., 1996; Kniemeyer and Heider, 2001), orin p-cresol degradation (Hopper et al., 1991; Peters et al.,2007), fumarate addition by glycyl radical enzymes tomethyl carbon atoms in toluene (Biegert et al., 1996;Beller and Spormann, 1998; Leuthner et al., 1998), m-, o-,p-xylene (Krieger et al., 1999; Achong et al., 2001;Morasch et al., 2004; Morasch and Meckenstock, 2005)and p-cresol (Müller et al., 2001), and anaerobic carbo-xylation of phenol (biological Kolbe-Schmitt reaction)(Schmeling et al., 2004; Schühle and Fuchs, 2004;Narmandakh et al., 2006).

Although the mechanism of benzene activation underanaerobic conditions is still unclear, putative reactionswere proposed based on metabolite detection using 13C6-benzene, H2

18O or 13C-labelled bicarbonate buffer. Suchreactions included (ii) hydroxylation of benzene to phenol(Vogel and Grbic-Galic, 1986; Grbic-Galic and Vogel,1987; Caldwell and Suflita, 2000; Chakraborty andCoates, 2005), (ii) direct carboxylation to benzoate

(Caldwell and Suflita, 2000; Phelps et al., 2001; Kunapuliet al., 2008; Abu Laban et al., 2009) or (iii) a methylationto toluene (Ulrich et al., 2005) (Fig. 1). Recent studiesshowed an abiotic formation of phenol from benzene inculture media for iron- and sulfate-reducing organisms bycontact with air during sampling (Kunapuli et al., 2008;Abu Laban et al., 2009). This indicated that phenol as aputative intermediate of benzene degradation has to beinterpreted with caution.

Recently, the genomes of five anaerobic aromatics-degrading microorganisms, including the phototrophicorganism Rhodopseudomonas palustris strain CGA009(Larimer et al., 2004), the denitrifying organisms Mag-netospirillum magneticum strain AMB-1 (Matsunaga et al.,2005) and Aromatoleum aromaticum strain EbN1 (Rabuset al., 2005), the iron reducer Geobacter metallireducensstrain GS-15 (Butler et al., 2007) and the fermenter Syn-trophus aciditrophicus strain SB (McInerney et al., 2007),have been sequenced and several operons coding forenzymes of anaerobic aromatic hydrocarbon degradationhave been identified (Carmona et al., 2009). The overallorganization of anaerobic catabolic gene clusters wasconserved across a wide variety of microorganisms.However, genus- and species-specific variations accountfor differences in gene arrangements, substrate specifi-cities and regulatory elements. Although the genomesequence of the benzene-degrading Dechloromonas

CH3

OH

COOH

COO-

-OOC

COO-

P O-

OH

O

COSCoA

C-

O

ATP AMP+Pi OH-

Pi CO2

PpsABC PpcABCD PpcABCD

Ohb1

BssABCD

BadA, BclA, BzdABamY

COSCoA

OH

COO-

OH

CoA+ATP AMP+PPi

2H+ +

2Fdred

H2O +2Fdox

CoA+ATP AMP+PPi COSCoA

Acetyl-CoA CO2

2ATP2[H]

2ADP2Pi

BcrCBADBzdNOPQ

2[H] BamB-I

CO2

HbaAHcrL

HbaBCDHcrCABPcmRST

CO2

Benzene

Toluene Benzylsuccinate Benzylsuccinyl-CoA

Benzoate

Phenol

Phenylphosphate

4-hydroxybenzoate 4-hydroxybenzoyl-CoA

Benzoyl-CoAa

b

c

COO-

+

COSCoA

COO-

O

BbsGHCDBbsAB

O

Fig. 1. Proposed options for anaerobic biodegradation of benzene. (a) Methylation to toluene followed by fumarate addition to formbenzylsuccinate that is subsequently metabolized to benzoyl-CoA. (b) Carboxylation to benzoate and ligation of CoA forming benzoyl-CoA.(c) Hydroxylation to phenol, and subsequent degradation via 4-hydroxybenzoate and benzoyl-CoA. The names of the enzymes in the differentorganisms are: BssABC, benzylsuccinate synthase; BbsEF, succinyl-CoA:(R)-benzylsuccinate CoA-transferase; BbsG (R)-benzylsuccinyl-CoAdehydrogenase; BbsH, phenylitaconyl-CoA hydratase; BbsCD, 2-[hydroxy(phenyl)methyl]-succinyl-CoA dehydrogenase; BbsAB,benzoylsuccinyl-CoA thiolase; BadA, BclA, BzdA and BamY, benzoate-CoA ligase; BcrCBAD, BzdNOQP and BamB-I, benzoyl-CoA reductase;PpsABC, phenylphosphate synthase; PpcABCD, phenylphosphate carboxylase; HbaA, HcrL, 4-hydroxybenzoate-CoA ligase; and HbaBCD,HcrCAB and PcmRST, 4-hydroxybenzoyl-CoA reductase.

2784 N. Abu Laban et al.

© 2010 Society for Applied Microbiology and Blackwell Publishing Ltd, Environmental Microbiology, 12, 2783–2796

aromatica strain RCB has been elucidated (accessionnumber NC_007298), no genes coding for putativeenzymes involved in anaerobic benzene degradationwere indentified (Salinero et al., 2009).

In the present study, we describe for the first time acombined proteomic and genomic approach to elucidatethe biochemical mechanism of anaerobic benzene degra-dation. Whole proteomes of benzene-, phenol- andbenzoate-grown cells were compared by a LC/ESI-MS/MS-based shotgun proteomic analysis and were corre-lated to the high-throughput sequenced metagenomeinformation of the iron-reducing, benzene-degradingenrichment culture BF. Additionally, specific protein bandsexpressed in benzene-grown cells were subjectedto N-terminal sequence analysis. A putative benzenecarboxylase-related protein was identified as key proteinthat might be responsible for the initial activation ofbenzene, and thus provide new evidence for carboxyla-tion of benzene under iron-reducing condition.

Results and discussion

Community structures of the benzene-degrading cultureBF with different growth substrates

In the present study, we investigated the anaerobicbenzene degradation pathways of the iron-reducingenrichment culture BF isolated from soil at a formermanufactured gas plant site. The highly enriched iron-reducing culture BF was dominated by Gram-positivePeptococcaceae-related microorganisms (Kunapuli et al.,2007). Besides benzene, the microorganisms were ableto grow with phenol (1 mM), 4-hydroxybenzoate (1 mM)and benzoate (1 mM), but not with toluene (1 mM), ethyl-benzene (1 mM) or xylene isomers (1 mM) as sole sourceof carbon (Kunapuli et al., 2007).

With the help of differential expression analysis weaimed at identifying proteins specifically expressedwith benzene as growth substrate but not with putativemetabolites of benzene degradation such as phenol orbenzoate. By comparing the different expression profileswe expected benzoate degradation genes to be inducedwith all three substrates. However, culture BF is not pureand a problem for proteome analysis might arise if differ-ent community members grew up when the culture istransferred from benzene to phenol or benzoate.

Two different full-length (1524 bp) 16S rRNA genesequences clustering with the Peptococcaceae (T-RF289) and Desulfobulbaceae (T-RF 162) were identifiedfrom the metagenome of the benzene-grown enrichmentculture BF. The T-RFLP analysis indicated the dominanceof Peptococcaceae (T-RF 289) during growth on benzene(Fig. S1A). To ensure that the microbial composition of theiron-reducing enrichment culture BF did not change when

the culture was transferred from benzene to a differentsubstrate, the community structure of benzene-, phenol-and benzoate-grown cultures was assessed by T-RFLPanalysis when approximately 25 mM ferrous iron was pro-duced (Fig. S1). With all substrates, the analysis showedthe same dominance of the Peptococcaceae-relatedmicroorganisms forming a T-RF of 289 bp (Fig. S2). Thisallowed a direct comparison of the expressed proteins ofcells grown on the respective substrates.

The metagenome of the iron-reducing culture BF

The draft metagenomic sequences (10.13 Mb) of theiron-reducing culture BF contained 14 270 open read-ing frames (ORFs) and 5832 contigs, while the DNAsequence length of the contigs ranged between 0.064 and537.470 kb. BLASTP search of the translated ORFs indi-cated that 21.3% of the total identified ORFs could beassigned with more than 50% identity to genes in theNCBI non-redundant protein sequence database, andaround 40.7% of the total ORFs did not show any relevantidentity to genes with known function. Details aboutgeneral genomic features are summarized in Table S1.

About 205 genes were identified to have closestsequence similarity to genes encoding enzymes and tran-scriptional regulators known to be involved in anaerobicaromatic hydrocarbon degradation indicating the impor-tance of aromatic compounds as carbon source for thegrowth of culture BF. Colocalized gene clusters werediscovered for some aromatic hydrocarbon-degradingproteins. However, the majority of genes were scatteredacross 63 different contigs of the metagenome (Table S2).Moreover, around 90 genes were discovered to havesequence similarity to genes encoding proteins related toxenobiotics transporters, e.g. ATP-binding cassette (ABC)and major facilitator superfamily (MFS-1).

An interesting feature of the metagenome was the pres-ence of about 112 genes similar to genes encodingphage-like proteins related to phage attachment and inte-gration, such as terminase, recombinase, integrase andresolvase. In addition, 18 ORFs similar to genes presentin transposable elements (e.g. transposase) indicatinga potential mobility of the genetic elements within themetagenome were identified. Some of the correspondingtransposable and phage-like genes were located withinthe contigs that contained putative aromatic-degradinggenes (Table S3).

Differential comparative proteome analysis of benzene-,phenol- and benzoate-grown cells under iron-reducingcondition by mass spectrometric analysis andN-terminal sequencing

The soluble proteomes of benzene-, phenol- andbenzoate-grown cells of the iron-reducing culture BF were

Enzymes involved in anaerobic benzene degradation 2785


separated by SDS-PAGE and differentially identified bymass spectrometric analysis. In total, 4020 proteins wereidentified for all tested substrates. A total of 276 proteinswere exclusively expressed in benzene-grown cells, 20of which could be assigned to aromatic hydrocarbondegradation pathways (Table 1). In addition, proteomes ofphenol-grown cells revealed the exclusive expression of90 proteins, 3 of which were similar to proteins involved inphenol degradation (Table 1). Further 200 proteins wereidentified to be specific for benzoate-grown cells, 13 ofwhich were similar to putative Bam and Bzd proteinsinvolved in the anaerobic degradation of benzoate(Table 1).

Benzene metabolism

Three different pathways have been proposed for anaero-bic benzene degradation, including a direct methylationto toluene and subsequent degradation via the benzylsuc-cinate pathway to benzoyl-CoA (Ulrich et al., 2005), ahydroxylation of benzene to phenol and further degrada-tion by the well-known phenol pathway to benzoyl-CoA(Caldwell and Suflita, 2000; Chakraborty and Coates,2005), and a direct carboxylation to benzoate and activa-tion by CoA to benzoyl-CoA (Caldwell and Suflita, 2000;Phelps et al., 2001; Kunapuli et al., 2008; Musat andWiddel, 2008; Abu Laban et al., 2009). If benzene degra-dation would proceed via a direct methylation as the firstactivation reaction of benzene we would expect thebenzoyl-CoA pathway enzymes to be expressed, theenzymes of the benzylsuccinate degradation pathway,plus an unknown methylase. If benzene would be acti-vated by a direct hydroxylation to phenol, one wouldexpect the same phenol degradation enzymes expressedas in phenol grown cells plus some unknown hydroxylaseenzyme. Finally, for a carboxylation as the first activationreaction of benzene, we would expect all benzoyl-CoAdegradation enzymes to be expressed plus the initial car-boxylase but no phenol degradation enzymes.

Benzoate metabolism-benzoyl-CoA pathway

In facultative anaerobes, benzoate degradation tobenzoyl-CoA involved benzoate activation to benzoyl-CoA catalysed by benzoate-CoA ligase (Schühle et al.,2003), and ring reduction of benzoate catalysed by thekey enzyme benzoyl-CoA reductase (Bcr) (Boll andFuchs, 1995; Boll, 2005). A Bcr enzyme has been isolatedand studied in the denitrifying Thauera aromatica (Bolland Fuchs, 1995). It is an ATP-dependent oxygen-sensitive enzyme with a abgd heterotetramere structure(Boll et al., 2000). In the strictly anaerobic organism G.metallireducens, the membrane protein complex BamBCDEFGI (BamB-I) was proposed to be responsible for

benzoyl-CoA reduction. BamY was identified as thebenzoate-CoA ligase (Wischgoll et al., 2005; Heintz et al.,2009).

Genes of benzoate metabolism in culture BF. Openreading frames homologous to putatively benzoateanaerobic metabolism (bam) genes that encode proteinsinvolved in reductive dearomatization of benzoyl-CoA to acyclic, conjugated diene during benzoate degradation inG. metallireducens (Wischgoll et al., 2005), were found tobe scattered over 19 different contigs of the metagenomesequence (Table S2). A large contig with 561 ORFs(contig BF_11345) contained genes similar to bamCBH-FEDI (ORFs 48–49, 52–54 and 57–58; Fig. 2B) encodingproteins responsible for benzoyl-CoA reduction in G. met-allireducens (Kung et al., 2009). The predicted geneproducts showed a high sequence identity (> 60%) tothe respective proteins described in Geobacter sp. FRC-32 (YP_002535701, YP_002535686, YP_002535687,YP_002535704, YP_0025-35705 and YP_002535691;Table S2). Furthermore, the metagenome of culture BFcontained one gene cluster putatively encoding proteinscatalysing subsequent reactions in benzoate degradation.The formation of a hydrolytic ring cleavage product inA. aromaticum strain EbN1 involves the enoyl-CoAhydratase (BzdW), a short-chain alcohol dehydrogenase(BzdX) and the ring opening hydrolase (BzdY), which areprobably encoded by ORFs 95–96 and 98 in culture BF(Table 1).

Benzoate specific protein expression. In benzoate-growncells different putatively bam-related ORFs probablyencoding benzoate-CoA ligase (ORF 194; Table 1) andbenzoyl-CoA reductase (ORFs 53, 56, 86, 152, 155, 157,159, 168 and 178; Table 1) were expressed. Theexpressed bam-like genes were located on differentcontigs of the metagenome sequence. Some of thesegenes were also expressed in benzene- and phe-nol-grown cells (ORFs 56, 86, 155, 168, 178 and 194). Inaddition, specific genes probably encoding enzymesinvolved in the lower, ring-opening pathway of anaerobicbenzoate degradation were found to be expressed inthe culture BF. bzdY- and bzdW-like genes (ORFs 95–97)encoding a putative ring-opening hydrolase and dienoyl-CoA hydratases were expressed in benzoate-grown cells,whereas one of the bzdW genes (ORF 96) was inducedwith all substrates. However, no protein products of thebzdX could be identified from phenol- and benzoate-grown cultures probably because expression of bzdXgenes was below the detection limit of the mass spectro-metric analysis. It is not clear from the obtained datawhy the bzdX gene is so little expressed in phenol andbenzoate-grown cells that the protein could not bedetected by mass spectrometric analysis. A possible



Tab

le1.

Ben

zene

-,ph

enol

-,an

dbe

nzoa

te-e

xpre

ssed

OR

Fs

ofcu

lture

BF

rela

ted

tokn

own

gene

sin

volv

edin

anae

robi

car

omat

ichy

droc

arbo

nde

grad

atio

nan

dtr

ansp

ort.

Gen

epr

oduc

tC

ontig

No.

Exp

ress

edpr

otei

nsa

Rel

ated

gene

prod

uct

(BLA

ST

Pan

nota

tion)

c

Leng

th(a

a)S

ize

(kD

a)

MS

/MS

pept

ides

iden

tifica

tionb

Gen

eP

rote

inE

-val

ueId

entit

y(%

)A

cces

sion

No.

Org

anis

md

Ben

zene

Phe

nol

Ben

zoat

e

OR

F10

BF

_588

718

521

.72

(7)

––

Ace

toph

enon

eca

rbox

ylas

e,ga

mm

asu

buni

t1.

E-3

850

ZP

_034

9485

0A

liac

OR

F25

BF

_596

619

921

.82

(5)

2(5

)–

Pro

babl

eU

biX

-like

carb

oxyl

ase

2.E

-72

65Y

P_1

5878

1A

zose

OR

F29

BF

_597

847

953

.2–

2(6

)2

(5)

Phe

nylp

hosp

hate

carb

oxyl

ase,

alph

asu

buni

t4.

E-1

7764

YP

_158

784

Azo

seO

RF

33B

F_5

983

397

44.3

–2

(10)

–ba

mI

For

mat

ede

hydr

ogen

ase,

alph

asu

buni

t1.

E-1

3061

ZP

_043

5307

3N

AO

RF

34B

F_6

004

289

30.9

13(5

0)–

2(1

1)pc

mR

Mol

ybdo

pter

inde

hydr

ogen

ase,

FAD

-bin

ding

subu

nit

1.E

-84

58Y

P_0

0253

5583

Geo

sfO

RF

35B

F_6

004

114

12.2

2(1

8)–

–pc

mS

(2F

e-2S

)-bi

ndin

gdo

mai

npr

otei

n9.

E-4

471

YP

_385

089

Geo

mg

OR

F53

BF

_113

4514

315

.9–

–2

(5)

bam

FM

ethy

l-vio

loge

n-re

duci

nghy

drog

enas

e,de

ltasu

buni

t2.

E-4

556

YP

_002

5357

05G

eosf

OR

F56

BF

_113

4576

8.0

–2

(5)

2(5

)ba

mE

4Fe-

4Sfe

rred

oxin

,iro

n-su

lfur

bind

ing

prot

ein

2.E

-26

85Y

P_4

6144

2S

ynas

OR

F70

BF

_113

4532

637

.62

(8)

––

Ben

zoyl

-CoA

redu

ctas

e/2-

hydr

oxyg

luta

ryl-C

oAde

hydr

atas

e1.

E-1

3267

YP

_001

2113

38P

elts

OR

F77

BF

_113

4895

10.1

–2

(16)

2(1

6)pc

mS

(2F

e-2S

)-bi

ndin

gdo

mia

npr

otei

n6.

E-3

373

YP

_002

5355

82G

eosf

OR

F83

BF

_113

4882

9.4

–2

(28)

–pp

cDP

heny

lpho

spha

teca

rbox

ylas

e,de

ltasu

buni

t2.

E-1

444

YP

_158

783

Azo

seO

RF

86B

F_1

1348

652

74.2

2(5

)–

2(5

)ba

mB

Ald

ehyd

efe

rred

oxin

oxid

ored

ucta

se0.

E+0

073

YP

_002

5356

85G

eosf

OR

F87

BF

_113

5251

457

.0–

––

bam

YB

enzo

ate-

CoA

ligas

e2.

E-1

5754

YP

_385

097

Geo

mg

OR

F90

BF

_113

7058

766

.8–

2(1

5)2

(10)

Ubi

D-li

keca

rbox

ylas

e0.

E+0

058

ZP

_028

4998

2D

prot

OR

F95

BF

_113

9337

642

.4–

–2

(9)

bzdY

6-ox

ocyc

lohe

x-1-

ene-

1-ca

rbon

yl-C

oAhy

drat

ase

3.E

-148

66A

AQ

0880

5A

zoev

OR

F96

BF

_113

9317

118

.55

(43)

2(1

2)2

(44)

bzdW

Put

ativ

edi

enoy

l-CoA

hydr

atas

e3.

E-4

251

CA

D21

636

Azo

evO

RF

97B

F_1

1393

708.

02

(36)

–2

(36)

bzdW

Die

noyl

-CoA

hydr

atas

e7.

E-0

335

CA

D21

628

Azo

evO

RF

98B

F_1

1393

183

20.1

2(1

5)–

–bz

dX6-

hydr

oxyc

yloh

ex-1

-ene

-1-c

arbo

xyl-C

oAde

hydr

ogen

ase

4.E

-51

61C

AI7

8830

NA

OR

F99

BF

_113

9313

815

.22

(18)

––

bzdX

6-hy

drox

ycyl

ohex

-1-e

ne-1

-car

boxy

l-CoA

dehy

drog

enas

e4.

E-3

245

CA

I788

30N

AO

RF

115

BF

_114

0648

153

.2–

2(5

)–

ppcA

Phe

nylp

hosp

hate

carb

oxyl

ase,

alph

asu

buni

t4.

E-1

6059

YP

_158

784

Azo

seO

RF

117

BF

_114

1119

921

.72

(18)

––

3-po

lypr

enyl

-4-h

ydro

xybe

nzoa

tede

carb

oxy-

lyas

e8.

E-6

764

YP

_160

255

Azo

seO

RF

120

BF

_114

1530

233

.22

(9)

––

Pro

babl

eU

biD

-like

carb

oxyl

ase

6.E

-52

41Y

P_1

5878

2A

zose

OR

F12

1B

F_1

1416

421

47.4

18(3

8)–

–pp

cAP

heny

lpho

spha

teca

rbox

ylas

e,al

pha

subu

nit

1.E

-45

31Y

P_1

5878

4A

zose

OR

F12

2B

F_1

1416

206

23.0

5(3

0)–

–3-

poly

pren

yl-4

-hyd

roxy

benz

oate

deca

rbox

y-ly

ase

1.E

-19

31Z

P_0

4376

160

NA

OR

F12

3B

F_1

1416

286

32.5

7(2

3)–

–pp

cAP

heny

lpho

spha

teca

rbox

ylas

e,al

pha

subu

nit

7.E

-24

30Y

P_1

5878

4A

zose

OR

F12

6B

F_1

1418

498

56.1

11(1

7)–

–H

ypot

hetic

alpr

otei

n1.

E-4

429

ZP

_043

7558

9C

atad

OR

F12

7B

F_1

1418

309

34.2

2(9

)–

–M

RP

prot

ein

hom

olog

ue,

cons

erve

dA

TP

ase

8.E

-57

44A

Y_0

0143

6025

NA

OR

F13

2B

F_1

1418

224

25.4

3(1

3)–

–H

ypot

hetic

alpr

otei

n3.

E-1

848

NP

_143

765

Pyr

hoO

RF

133

BF

_114

1813

115

.27

(51)

––

Hyp

othe

tical

prot

ein

3.E

-00

28A

AP

3507

4D

erfa

OR

F13

6B

F_1

1418

317

34.2

2(8

)–

–M

ultid

rug

resi

stan

cepr

otei

n,M

RP

fam

ily3.

E-5

542

YP

_002

3524

37D

ictd

OR

F13

7B

F_1

1418

126

14.7

13(8

5)–

9(4

8)pp

cDP

heny

lpho

spha

teca

rbox

ylas

e,de

ltasu

buni

t2.

E-0

237

YP

_158

783

Azo

seO

RF

138

BF

_114

1850

857

.430

(85)

–15

(43)

pp

cAP

hen

ylp

ho

sph

ate

carb

oxy

lase

,al

ph

asu

bu

nit

2.E

-106

43Y

P_1

5878

4A

zose

OR

F13

9B

F_1

1418

541

60.7

7(1

2)–

–ba

mY

Ben

zoat

e-C

oAlig

ase

1.E

-172

56Y

P_3

8509

7G

eom

gO

RF

140

BF

_114

1815

316

.85

(29)

––

Pro

babl

eU

biX

-like

carb

oxyl

ase

2.E

-52

68P

5776

7T

haar

OR

F15

2B

F_1

1447

102

11.5

––

2(8

)ba

mF

Met

hyl-v

iolo

gen-

redu

cing

hydr

ogen

ase,

delta

subu

nit

5.E

-09

50Y

P_0

0253

5689

Geo

sfO

RF

155

BF

_114

5265

474

.317

(29)

–8

(7)

bam

BA

ldeh

yde

ferr

edox

inox

idor

educ

tase

0.E

+00

73Y

P_0

0253

5685

Geo

sfO

RF

157

BF

_114

5218

420

.7–

–4

(30)

bam

C4F

e-4S

ferr

edox

in,

iron-

sulfu

rbi

ndin

gpr

otei

n1.

E-5

457

YP

_384

757

Geo

mg

OR

F15

9B

F_1

1460

637

72.6

––

2(1

0)ba

mB

Ald

ehyd

efe

rred

oxin

oxid

ored

ucta

se0.

E+0

070

YP

_002

5356

85G

eosf

OR

F16

1B

F_1

1460

289

30.9

––

5(1

0)pc

mR

Mol

ybdo

pter

inde

hydr

ogen

ase,

FAD

-bin

ding

subu

nit

1.E

-54

57Y

P_3

8475

7G

eom

gO

RF

163

BF

_114

6040

545

.22

(7)

––

bam

MA

cyl-C

oAde

hydr

ogen

ase

dom

ain

prot

ein

1.E

-118

51Z

P_0

4150

428

Bac

11O

RF

168

BF

_114

92_

365

40.5

10(3

2)–

25(3

1)ba

mI

4Fe-

4Sfe

rred

oxin

,iro

n-su

lfur

bind

ing

prot

ein

8.E

-38

29Y

P_0

0243

0370

Geo

mg

OR

F17

8B

F_1

1515

732

82.0

22(2

0)–

25(3

1)ba

mF

Ace

tyl-C

oAde

carb

onyl

ase,

beta

subu

nit

0.E

+00

74Y

P_3

6006

0C

arhz

OR

F18

4B

F_1

1550

339

36.6

–2

(12)

–hb

aD4-

hydr

oxyb

enzo

yl-C

oAre

duct

ase,

beta

subu

nit

1.E

-47

37N

P_9

4602

5R

hop2

OR

F18

6B

F_1

1550

543

58.7

––

–pc

mT

4-hy

drox

yben

zoyl

-CoA

redu

ctas

e,al

pha

subu

nit

3.E

-109

40Y

P_3

8509

0G

eom

gO

RF

194

BF

_115

5452

157

.710

(20)

–2

(5)

bam

YB

enzo

ate-

CoA

ligas

efa

mily

7.E

-168

55Y

P_0

0253

5674

Geo

sfO

RF

197

BF

_115

5477

485

.018

(32)

–9

(12)

pcm

T4-

hydr

oxyb

enzo

yl-C

oAre

duct

ase,

alph

asu

buni

t0.

E+0

077

YP

_002

5355

81G

eosf

OR

F19

9B

F_1

1555

167

17.8

2(1

4)–

–pc

mS

4-hy

drox

yben

zoyl

-CoA

redu

ctas

e,ga

mm

asu

buni

t2.

E-6

770

ZP

_030

2330

8G

eosf

OR

F20

1B

F_1

1556

773

84.9

2(5

)–

–pc

mT

4-hy

drox

yben

zoyl

-CoA

redu

ctas

e,al

pha

subu

nit

0.E

+00

76Y

P_0

0253

5581

Geo

sf

a.‘a

a’is

the

num

ber

amin

oac

ids;

‘kD

a’is

the

size

ofpr

otei

nsin

kilo

Dal

ton.

b.

Num

bers

with

out

pare

nthe

sis

indi

cate

uniq

uepe

ptid

es.

Num

bers

with

pare

nthe

sis

give

the

perc

enta

geof

cove

rage

(%)

ofth

eam

ino

acid

sequ

ence

sof

the

iden

tified

prot

eins

byth

ede

tect

edpe

ptid

es.

c.H

itsw

ere

obta

ined

from

BLA

ST

Pco

mpa

rison

ofpr

edic

ted

prot

eins

with

NC

BI

data

bank

(http

://bl

ast.n

cbi.n

lm.n

ih.g

ov/B

last

.cgi

).T

heE

-val

ue,

give

sex

pect

atio

nva

lue.

Iden

tity,

give

sth

epe

rcen

tage

ofam

ino

acid

sth

atar

eid

entic

albe

twee

nth

eO

RF

and

the

next

rela

ted

prot

ein

inth

eda

taba

nk.A

cces

sion

no.,

the

acce

ssio

nnu

mbe

rfr

omN

CB

Ida

taba

se.

d.

Spe

cies

nam

esar

eab

brev

iate

dac

cord

ing

toth

elis

tof

orga

nism

iden

tifica

tion

code

(SW

ISS

-PR

OT

):A

zose

,Aro

mat

oleu

mar

omat

icum

stra

inE

bN1;

Pel

ts,

Pel

otom

acul

umth

erm

opro

pion

icum

stra

inS

1;G

eosf

,G

eoba

cter

sp.

stra

inF

RC

-32;

Geo

mg,

Geo

bact

erm

etal

lired

ucen

sst

rain

GS

-15;

Alia

c,A

licyc

loba

cillu

sac

idoc

alda

rius;

Dpr

ot,

Del

ta-p

rote

obac

teriu

m;

Azo

ev,

Azo

arcu

sev

ansi

i;C

arhz

,C

arbo

xydo

ther

mus

hydr

ogen

ofor

man

s;R

hop2

,R

hodo

pseu

dom

onas

palu

stri;

Syn

as,

Syn

trop

hus

acid

itrop

hicu

s;T

haar

,Tha

uera

arom

atic

a;B

ac11

,Bac

illus

sp.;

Cat

ad,C

aten

ulis

pora

acid

iphi

la;P

yrho

,Pyr

ococ

cus

horik

oshi

i;D

erfa

,Der

mat

opha

goid

esfa

rinae

;Dic

td,D

icty

oglo

mus

turg

idum

;NA

,Nam

eof

the

orga

nism

isno

tava

ilabl

ein

SW

ISS

-PR

OT.

The

spec

ifica

llybe

nzen

e-ex

pres

sed

prot

ein

of60

kDa

iden

tified

byth

edi

rect

N-t

erm

inal

sequ

enci

ngfr

omth

eS

DS

gels

isde

pict

edin

bold

.



http://blast.ncbi.nlm.nih.gov/Blast.cgi

explanation would be that for benzene degradation fastprocessing of down stream metabolites in the pathwayis required to enhance the initial reaction. A higher expres-sion of such down stream enzymes would thereforeincrease the kinetics of the pathway and help removingmetabolites of the first reaction steps such as benzoate.

The identification of bam-like proteins suggests thatculture BF utilizes a benzoyl-CoA pathway whose activa-tion and reduction steps are analogous to the one pro-posed for strictly anaerobic microorganisms, such asG. metallireducens or Syntrophus gentiani (Wischgollet al., 2005; McInerney et al., 2007; Heintz et al., 2009).Moreover, the expression of the putative bzdXYW-likesgenes suggests that the genes encoding the modifiedb-oxidation of the benzoate degradation pathway usingdienoyl-CoA as substrate and finally generating3-hydroxypimelyl-CoA (Laempe et al., 1998; Laempeet al., 1999) appear to be more closely related to thoseof denitrifying bacteria (Carmona et al., 2009). As wepresent here for the first time a putative benzoyl-CoApathway of a Gram-positive organism, this might indicate

that this phylogenetic linage uses a combination of thebenzoyl-CoA pathways of strict anaerobes and facultativeanaerobes.

Methylation as a possible initial activationreaction of benzene

Putative genes of toluene metabolism. On one contig(BF_5902; Table S2) of the metagenome of culture BFwe identified ORFs (ORFs 11–13), which are similar tobssCAB toluene catabolic genes encoding benzylsucci-nate synthase of A. aromaticum strain EbN1 (YP_158059to YP_1580561) and bssD (ORF 18) encoding the acti-vating enzyme. These genes were part of a large toluene-like cluster which contained also ORFs similar to bssF(ORFs 15 and 17) and bssE (ORF 14) encoding proteinswith so far unknown functions in toluene degradation ofA. aromaticum strain EbN1 (Fig. 2A). In addition to theputative bss genes, a further contig (BF_11494; Table S2)containing 123 ORFs harboured putative bbsCABEFDgenes whose products might be involved in the

3-octaprenyl-4-hydroxybenzoate carboxy-lyase

Benzylsuccinate synthase

4-hydroxybenzoyl-CoA reductase

Phenylphosphate carboxylase

Phenylphosphate synthase

Benzoate-CoA ligase

Benzoate degradation bamB-I

Sigma 54 specific transcriptional regulator

B

C

A

5 10 15 Kb

Putative uncharacterized protein

ORF 72 73 80797877 85838281

104 114105 106 107 110 111 112 115 116

48 49 52 5453 55 56 57 58

11 12 13 18

ORF

ORF

ORF

170 171 172 173 174 175

14 15 17

47

β-oxidation of benzylsuccinate

Contig BF_ 5887

Contig BF_ 11494

Contig BF_11345

Contig BF_11348

Contig BF_ 11406

bssC bssA bssB bssE bssFbssF bssD

bbsA bbsB bbsD bbsE bbsG bbsF

bamC bamB bamFbamH bamE bamE bamE bamD bamI

ubiX ubiX pcmS pcmT ppsA ppcB ppcC ppcA ppcD pcmR

84

ppcD ppcA ppcC ppcB CcppAcppDcppAspp

Fig. 2. Genes identified in the metagenome of the iron-reducing culture BF that encode putative proteins of anaerobic aromatic degradation:(A) toluene, (B) benzoate and (C) phenol and 4-hydroxybenzoate. Predicted functions of ORFs were based on sequence similarities to knowngenes. The gene clusters were discovered on different contigs of the metagenome of culture BF (Table S2).



b-oxidation of benzylsuccinate or some analogue (ORFs170–175; Fig. 2A).

Protein expression of toluene metabolism. Althoughgenes were identified in the metagenome that mightencode the toluene degradation enzymes, gene productscould be identified from protein extracts of neitherbenzene-, nor phenol- or benzoate-grown cells. Thisstrongly suggests that benzene is not degraded viatoluene as an intermediate, which would be the firstproduct of a putative methylation reaction. Moreover,toluene was not utilized as growth substrate by culture BF(Kunapuli et al., 2008). Recent studies of benzene degra-dation by other sulfate-reducing enrichment cultures alsoindicated that toluene could neither be used as a sub-strate nor as a co-substrate with benzene (Ulrich et al.,2005; Musat et al., 2008; Oka et al., 2008; Abu Labanet al., 2009). Furthermore, benzylsuccinate as the specificmetabolite of toluene degradation was also detectedneither in the iron-reducing culture BF nor in the sulfate-reducing culture BPL (Kunapuli et al., 2008; Abu Labanet al., 2009). Therefore, our proteomic data support therecent metabolite analyses that excluded the formation oftoluene during benzene degradation, and we thereforeexclude methylation as an initial activation mechanism forthe anaerobic benzene degradation in our iron-reducingculture BF. Nevertheless, we propose that the putativebss and bbs genes identified in the metagenome of BFculture are involved in catalysing a fumarate additionreaction of non-identified alkylated aromatic compoundsother than toluene.

Hydroxylation as a possible initial activationreaction of benzene

The hydroxylation of benzene to phenol was previouslyproposed as an initial activation mechanism based on theidentification of phenol as metabolite in culture super-natants (Caldwell and Suflita, 2000; Chakraborty andCoates, 2005). Anaerobic phenol catabolism by the deni-trifying microorganisms T. aromatica and A. aromaticumstrain EBN1, and the iron-reducing organism G. metallire-ducens, has been described in detail (Breinig et al., 2000;Rabus et al., 2005; Schleinitz et al., 2009), and proceedsvia phenylphosphate synthase (PpsABC), phenylphos-phate carboxylase (PpcABCD), 4-hydroxybenzoate-CoAligase (HcrL) and 4-hydroxybenzoyl-CoA reductase(HcrCAB/PcmRST) to benzoyl CoA.

Genes of phenol metabolism. Open reading framescoding for enzymes similar to phenylphosphate synthaseand phenylphosphate carboxylase in A. aromaticumstrain EbN1 were found in two different gene clusterslocated on separate contigs of the metagenome of culture

BF (BF_11348 and BF_11406; Table S2). The respectivecontigs contained 72 and 60 ORFs respectively. Openreading frames similar to the putative a-subunit of thephenylphosphate synthase (PpsA) were located in bothclusters (ORFs 79 and 110; Fig. 2C) and contained thecharacteristic conserved His-522 residue, which is thespecific binding site for ATP. ORFs similar to genes codingfor the phenylphosphate synthase b- and g-subunits(PpsB and PpsC) were absent in both gene clusters.However, such sequences were identified on othercontigs of the metagenome sequence (ORFs 36-37, 89and 156; Table S2). Adjacent to ORFs 79 and 110, whichare putatively encoding phenylphosphate synthasea-subunit (PpsA), two clusters of ORFs were found,respectively, that are similar to genes encoding phe-nylphosphate carboxylase ppcBCAD (ORFs 80–83 and104–107; Fig. 2C). The identified subunits of the putativephenylphosphate carboxylases showed more than 45%sequence identity to the respective proteins of A. aromati-cum strain EbN1 (YP_158783 to YP_158786).

Degradation of 4-hydroxybenzoate as an intermediatecompound of anaerobic phenol degradation requires4-hydroxybenzoate-CoA ligase and 4-hydroxybenzoyl-CoA reductase producing benzoyl-CoA. No ORFs similarto genes encoding 4-hydroxybenzoate-CoA ligase couldbe identified in the shotgun metagenome of culture BF.However, several ORFs orthologous to the gene codingfor benzoate-CoA ligase (BamY) of G. metallireducens(YP_385097) were found within a cluster of putativephenol degradation genes mentioned above (ORF 111)and another gene cluster (ORFs 189–198) related to4-hydroxybenzoate degradation (ORF 194). In general,amino acid sequences of aromatic CoA-ligases arevery similar to each other and it is difficult to differentiatesuch ligases only based on the sequence (Butleret al., 2007; Peters et al., 2007). Open reading framessimilar to genes pcmRST encoding three subunits of4-hydroxybenzoyl-CoA reductase (ORFs 77–78 and 85;Fig. 2C) were identified within one of the two putativephenol degradation gene clusters (contig BF_11348;Table S2), showing 73% (ORF 77: putative PcmS), 70%(ORF 78: putative PcmT) and 58% (ORF 85: putativePcmR) amino acid sequence identities to the respectiveproteins in G. metallireducens (Peters et al., 2007).Moreover, like in G. metallireducens the PcmR subunit ofthe putative 4-hydroxybenzoyl-CoA reductase lacks theinsertion of 40 amino acids, which carries the additional[4Fe-4S] cluster loop responsible for an inverted electronflow in facultative anaerobes. Additionally, several ORFsorthologous to 4-hydroxybenzoyl-CoA reductase encod-ing genes were discovered on 9 different contigs ofthe BF metagenome sequence (ORFs 34, 35, 62, 63, 64,103, 161, 162, 183, 186, 197, 198, 199, 201, 202 and203 respectively; Table S2).



Phenol-specific protein expression. The majority of pro-teins specifically expressed in phenol-grown cells wererelated to phenylphosphate carboxylase (Ppc; ORFs29, 83, and 115), 4-hydroxybenzoyl-CoA reductase(Hba/Pcm; ORFs 77 and 184) and 3-octaprenyl-4-hydroxybenzoate carboxy-lyase (UbiD/UbiX; ORF 90)enzymes. In addition, ORFs 77 and 184 encoding PcmS-and HbaD-like proteins were expressed. Noteworthy,ORFs 77, 83 and 115 were located in the putative phenol-degrading clusters that are described in Fig. 2C. We con-clude that these clusters are responsible for phenoldegradation in culture BF. Unfortunately, the other phenol-expressed ORFs 29 and 184 are located on very shortcontigs, and therefore putative information concerningthe phenylphosphate synthase (PpsABC) and carboxy-lase (PpcBC) is not available.

Our proteomic data revealed that the phenol degra-dation genes described above were not expressed inbenzene-grown cells. This finding supports recent studiesshowing that phenol detected in benzene-degradingcultures was formed abiotically during sampling due tocontact of reduced ferrous iron in the culture medium withoxygen (Kunapuli et al., 2008; Abu Laban et al., 2009).

A theoretical possibility for the hydroxylation of ben-zene could be by xanthine-oxidase-like proteins (Enrothet al., 2000), which showed high sequence identityto 4-hydroxybenzoyl-CoA reductase PcmRST subunitsof Geobacter sp. strain FRC-32 (YP_002535581 toYP_002535583) (ORFs 34, 35, 197, 199 and 201).PcmRST of G. metallireducens is a special proteinbelonging to the xanthine-oxidase family of molybdenumcofactor containing enzymes that catalyse the irreversiblereductive dehydroxylation of 4-hydroxybenzoyl-CoA yield-ing benzoyl-CoA and water (Peters et al., 2007). It is veryunlikely that such an enzyme could catalyse the directhydroxylation of benzene to phenol, which would formallybe a reversal of the dehydroxylation reaction. Never-theless, if the gene products of pcmRST catalyse the4-hydroxybenzoyl-CoA reduction, their expression inbenzene-grown cells would indicate an active phenoldegradation pathway while the initial hydroxylationof benzene to phenol would be performed by otherenzymes. Furthermore, the pcmRST-like ORFs in cultureBF were unspecifically expressed with all three growthsubstrates. This finding was in accordance with the resultsof other studies, showing that pcmRST genes were con-stitutively expressed with all tested aromatic substratesutilized by G. metallireducens. On the other hand,4-hydroxybenzoyl-CoA reductase activity could onlybe measured with cells grown on p-cresol and4-hydroxybenzoate as substrates. Therefore, the authorssuggested a posttranscriptional regulation of the PcmRSTenzyme activity (Peters et al., 2007). These observationsmight indicate that the expressed gene products of ORFs

34, 35, 197, 199 and 201 in benzene-grown cells ofculture BF (Table 1), which are similar to PcmRST mightnot necessarily be active.

Our data provide no direct indications that theexpressed pcmRST-like genes are involved in phenoldegradation in culture BF. At present we can thereforenot propose a clear function for the expressed pcm-likegenes in benzene-grown cells. Together with the lack ofproteomic indications that benzene might be hydroxylatedand further processed via a phenol degradation pathway,we do not support a direct hydroxylation of benzene asthe initial activation mechanism.

Carboxylation as a possible initial activationreaction of benzene

Benzene specific protein expression. Electrophore-tic separation of benzene-, phenol- and benzoate-expressed proteins revealed a very prominent proteinband with a mass of about 60 kDa specifically expressedwith benzene as growth substrate (Fig. 3). TheN-terminal sequence of the first 31 amino acids of thepolypeptide was determined by Edman sequencing andwas identical to the amino acid sequence predicted fromORF 138 (contig BF_11418; Table 1). The gene productof ORF 138 [molecular mass (Mw) 57.4 kDa] showed43% sequence identity to genes coding for the a-subunitof phenylphosphate carboxylase (PpcA) in A. aromati-cum strain EbN1 (YP_158784). These data were alsoconfirmed by mass spectrometric identification of pep-tides of the 60 kDa band. Whereas no expression ofORF 138 could be detected in protein extracts fromphenol-grown cultures and no pronounced band wasvisible in the SDS gels, the much more sensitive massspectrometric analysis showed that ORF 138 was as well

200

kDa

120

BA C M

100

50

40

30

25

20

15

6070

200

kDa

120

BA C M

100

50

40

30

25

20

15

6070

Fig. 3. Coomassie-stained SDS-PAGE of proteins extracted fromthe iron-reducing culture BF grown on either benzene (lane A),phenol (lane B) or benzoate (lane C). The arrow indicates thespecific benzene-induced protein of 60 kDa. Lane M representsthe molecular mass standard.



expressed with benzoate as growth substrate. As thisband was strongly expressed only during growth onbenzene and not on phenol, the protein is probably notinvolved in phenol metabolism. We rather propose thatthis enzyme is involved in the direct carboxylation ofbenzene. Furthermore, the absence of the 60 kDa geneproduct of ORF 138 in SDS gels from benzoate growncells might indicate that the gene is expressed to a minorextent but not upregulated. Thus, the polypeptide wasdetected by the qualitative mass spectrometry but notvisible in the less sensitive protein gel. However, a quan-titative proteomic analysis could give further insight inthe specific regulation of the genes involved in benzenedegradation.

Furthermore, a product of ORF 139 (Mw 60.7 kDa;Table 1), which is specifically expressed during growthon benzene, encodes a protein with 56% sequence iden-tity to benzoate-CoA ligase of G. metallireducens(BamY). This protein was also identified by mass spec-trometry analysis from a gel slice containing peptideswith masses around 60 kDa. However, this polypeptidewas not detected in the Edman sequencing of the60 kDa band probably because it was present in onlyminor amounts as compared with the prominent ORF138 gene product. We propose that the benzene-inducedgene product of ORF 139 constitutes a benzoate-CoAligase channelling the product of benzene carboxylation,benzoate, into the benzoyl-CoA reduction pathway. Wename the enzyme BzlA for benzoate-CoA ligase ofculture BF.

Additionally, a product of ORF 137 encoding a protein(Mw 14.7 kDa; Table 1) similar to the d-subunit of phe-nylphosphate carboxylase (PpcD) was identified by massspectrometry. However, the later ORF was not visible inSDS-PAGE of benzene-grown cells because of the lowresolution for small-molecular-weight proteins in the SDSgel. Moreover, other carboxylase related proteins wereexpressed with benzene. These were encoded by ORFs121 and 123, which are located on another contig(BF_11416; Table 1). The later ORFs might be homo-logues to ORF138 and showed low similarity (30%sequence identity) to PpcA of A. aromaticum strain EbN1(YP_158784).

Other proteins specifically expressed on benzenewere UbiD/UbiX-like carboxylase (3-octaprenyl-4-hydroxybenzoate carboxy-lyase; ORFs 122, 124 and140; Table 1) and an MRP family ATP-binding protein-homologue (multidrug-resistant protein; ORFs 127 and136; Table 1).

Moreover, proteins similar to 4-hydroxybenzoyl-CoA reductase subunits (PcmRST; ORFs 34, 35, 197,199 and 201) were identified from benzene-grown cells(Table 1). Open reading frames 34 and 197 were alsoexpressed in benzoate but not in phenol-grown cells.

Putative genes of benzene metabolism. Open readingframes coding for the putative subunits PpcD (ORF 137)and PpcA (ORF 138) of phenylphosphate carboxylasewere discovered on a 47-ORF-containing contig (BF_11418; Table S2). The two ORFs clustered togetherwith genes encoding an UbiD/UbiX-like carboxylase(3-octaprenyl-4-hydroxybenzoate carboxy-lyase; ORFs124 and 140; Table S2) (Fig. 4A). The predicted geneproducts of ORF 137 (14.7 kDa) and ORF 138 (57.4 kDa)displayed markedly reduced sequence identities 35% and37% to putative ppcD (ORF 83 and ORF 104) and 43%and 46% to putative ppcA (ORF 82 and ORF 105) genesof the other putative phenol degradation gene clustersmentioned above. As the gene cluster containing ORFs137 and 138 did not contain ORFs similar to ppcB andppcC needed for phenol carboxylation or ORFs similar toppsABC genes encoding for phenylphosphate synthase(PpsABC), we conclude that this gene cluster is notinvolved in anaerobic phenol metabolism but rathercontained the putative anaerobic benzene carboxylase(Abc) genes (Fig. 4A).

The location in one gene cluster and the simultaneousexpression of ORFs 126–140 (Table 1), all related toeither carboxylation or transport, together with the specificbenzoate-CoA ligase (BzlA), strongly suggest that wehave identified the best candidates for genes encodingan enzyme, which is directly carboxylating benzene tobenzoate. Based on the genomic and proteomic datadiscussed above we propose a direct carboxylation ofbenzene to benzoate as the initial activation mechanismin anaerobic benzene degradation (Fig. 4B). We namethis enzyme Abc, which most likely belongs to a carboxy-lase family including phenylphosphate carboxylase andUbiX (3-octaprenyl-4-hydroxybenzoate carboxy-lyase).Anaerobic benzene carboxylase probably consists ofseveral subunits two of which are encoded by ORFs 137and 138, whereas the other potential subunits mightbe encoded by, e.g. ORFs 124, 126, 132, 133 and 140(Fig. 4A). The direct association of the genes to the dif-ferent subunits and functions of the enzyme has to besolved by measuring the enzyme reaction and subse-quent purification of Abc. Nevertheless, the anaerobiccarboxylation of benzene constitutes a novel enzymereaction, which is unprecedented in biochemistry. A par-allel in chemistry would be a Friedl–Crafts acylation ofaromatic compounds, but it is out of the scope of such astudy to speculate about possible reaction mechanisms.

Enzyme assays of putative benzene carboxylase,4-hydroxybenzoate-CoA ligase and4-hydroxybenzoyl-CoA reductase

Enzymes activity tests of putative benzene carboxylase,4-hydroxybenzoate-CoA ligase and 4-hydroxybenzoyl-



CoA reductase were performed with extracts of cellsgrown on benzene and phenol as control. However, wewere not able to measure any activity so far, which mightbe related to several reasons. First, culture BF is mainlycomposed of Gram-positive bacterial cells, which are noteasily disrupted. We might have failed to open these cells,e.g. due to the presence of ferric iron minerals. Second,the yield of cells biomass and proteins (25–40 mg l-1) ofthe culture BF was extremely low and might not havebeen sufficient for the enzyme assay. Third, the presenceof insoluble ferrihydrite in the cell extract might have inter-fered with the enzyme activity.

Possible horizontal gene transfer of the putativearomatic-degrading genes of culture BF

The metagenome analysis indicated that most of the geneproducts likely to be involved in phenol carboxylation(e.g. PpsA, PpcBCAD) and a modified b-oxidation of thebenzoyl-CoA pathway (e.g. BzdYWX) are similar to thoseof denitrifying Betaproteobacteria. However, gene prod-ucts likely to be involved 4-hydroxybenzoyl-CoA reduction(e.g. PcmSTR), benzoate activation (BamY) and dearo-matization (e.g. Bam-like) appear to be more similar tothose of strict anaerobic Deltaproteobacteria (Table 1).These data may suggest that the aromatic-degradinggene clusters of culture BF might be acquired from twophylogenetically different microorganisms through hori-zontal gene transfer. This might be explained by the iden-

tification of phage-related and transposable elementgenes within contigs containing aromatic-degrading gene(Table S3).

Experimental procedures

Growth conditions of anaerobic benzene-degradingenrichment cultures

The iron-reducing enrichment culture BF was cultivated aspreviously described (Kunapuli et al., 2007). For differentialprotein expression analyses, cultivation was performed in 4 lgrowth medium. Inocula (10%, v/v) from benzene-grown pre-cultures of culture BF were transferred into separate bottlescontaining either benzene (1 mM), phenol (1 mM) or ben-zoate (1 mM) as growth substrates in the presence of XAD7as substrate reservoir (Morasch et al., 2001) and 50 mMferrihydrite as electron acceptor. All culture bottles were incu-bated at 30°C in the dark.

Microbial community structure analysis of theiron-reducing enrichment culture BF

The total genomic DNA of benzene-, phenol- and benzoate-grown cells of the iron-reducing enrichment culture BF wasextracted during the exponential growth phase with theFastDNA Spin Kit for Soil (MP Biomedicals, Illkirch, France)according to the manufacturer’s protocol. The microbial com-munity structure was analysed by terminal restriction frag-ment length polymorphism (T-RFLP) analysis as previouslydescribed by Winderl and colleagues (2008). The analysiswas carried out in three biological replicates for each aro-matic substrate.

Contig BF_11418αδ

Carboxylase-like

Putative benzene carboxylase

Benzoate-CoA ligase

Transcriptional regulator, MarR family

2 Kb

Putative uncharacterized protein

124 126* 127* 132*133* 136*137* 138*

139* 140* 141 142ORF

Multidrug resistance protein MRP homologue, conserved ATPase

abcD abcA bzlA

A

B

COOH COSCoA

BzlA

CoA + ATP AMP + PPi

AbcDABenzene Benzoate

CO2

Benzoyl-CoA

ubiX ubiX

Fig. 4. Putative benzene degradation gene cluster of the iron-reducing culture BF.A. Genes that encode proteins specifically expressed during anaerobic benzene degradation are indicated with stars. Putative genes involvedin the initial activation of benzene are shown in a rectangle box. Predicted functions of ORFs are based on sequence similarities to knowngenes.B. Proposed roles of predicted genes in the initial activation of anaerobic benzene degradation to benzoate and benzoyl-CoA. Thecarboxylation step catalysed by anaerobic benzene carboxylase AbcDA might involve further, not specified gene products, whereas thebenzoate-ligation step is probably catalysed by benzoate-CoA ligase BzlA.



Shotgun genomic DNA sequencing of theiron-reducing culture BF

The whole metagenome of the iron-reducing culture BFgrown on benzene was extracted with the FastDNA Spin Kitfor Soil (MP Biomedicals) according to the manufacturer’sprotocol. Genomic information was obtained by shotgunsequencing using a GS FLX sequencer (performed by Euro-fins MWG Operon, Ebersberg, Germany). The sequencereads have been automatically assembled into contigs.Automated annotation of the assembled sequences was per-formed using the PEDANT (PMID 18940859) softwaresystem. The prediction of coding sequences was performedwith a combination of Genemark 2.6r (PMID 11410670) andGlimmer 3.02 (PMID 17237039). Homology to already pub-lished proteins in Uniref100 was used to decide about thebest gene models and in order to decide about the best genestarts. Coding sequences were automatically assigned byPEDANT to functional categories according to the functionalrole catalogues FunCat (PMID 15486203) and Gene Ontol-ogy (PMID 14681407). All sequence data from this studywere deposited at GenBank under the accession numbersGU357855 to GU358059.

Protein extraction and SDS-PAGE

Cells were harvested from 4 l culture bottles by centrifugationat 9000 g for 15 min. The cell pellet was washed three timeswith 50 mM Tris-HCl buffer (pH 7.5) and incubated overnightat -20°C. One gram of cell pellet was resuspended in 1 mllysis buffer (9 M urea, 2% CHAPS, 1% DTT; GE HealthcareEurope GmbH, Freiburg, Germany). The cell extracts weretreated with a 7¥ stock solution (167 ml ml-1 lysis buffer) ofComplete Mini EDTA-free Protease Inhibitor Cocktail Tablet(Roche Diagnostics GmbH, Penzberg, Germany) and incu-bated at 15°C for 30 min. Extracts were transferred intomatrix lysis tubes B (MP Biomedicals) and were treated in abead beating FastPrep Instrument 120 (MP Biomedicals) at6.0 m s-1 for 35 s. After centrifugation for 1 min at 20 000 gthe supernatant was incubated with nuclease mix (1 ml per100 ml of supernatant; GE Healthcare) at 20°C for 30 min andagain purified by centrifugation (20 000 g, 1 h).

Quantification of protein concentrations was performedusing 2D Quant Kit according to the manufacturer’s instruc-tions (GE Healthcare). Proteins were precipitated withacetone for 10 min at -20°C, and collected by centrifugationat 15 000 g for 30 min. Some 30 mg of proteins was resus-pended in 20 ml lysis buffer and subsequently separated bysodium dodecyl sulfate-polyacrylamide gel electrophoresis(SDS-PAGE; 4% stacking gel, 12 % separating gel) with arunning buffer containing glycine 14.4 g l-1, Tris-base 3 g l-1

(pH 8.3) and SDS 1 g l-1. After electrophoresis, the gels werestained with colloidal Coomassie brilliant blue (Zehr et al.,1989). Protein extracts were prepared from three biologicalreplicates for each substrate. Three parallel gels for eachprotein extract were analysed to account for technical varia-tions in proteomic analysis.

Protein identification by ESI/LC-MS/MS

For identification of proteins from the SDS-PAGE ofbenzene-, phenol- and benzoate-grown cells, the complete

lane was cut into 10 slices, washed with H2O for 30 min,reduced with 5 mM DTT (15 min, 20°C), and acetylated with25 mM iodacetamide (15 min, 20°C). Then, the gel sliceswere washed twice with 40% acetonitrile and once for 5 minwith 100% acetonitrile. Subsequently, proteins were digestedwith trypsin (0.03 mg ml-1 of 50 mM ammonium bicarbonate)overnight at 37°C. The digested peptides were separatedby reversed phase chromatography [PepMap, 75 mm (insidediameter) ¥ 250 mm, LC Packings] operated on a nano-HPLC (Ultimate 3000, Dionex) with a non-linear 170 mingradient using 2% acetonitrile and 0.1% formic acid in water(A) and 0.1% formic acid in 98% acetonitrile (B) as eluentswith a flow rate of 250 nl min-1. The gradient settings weresubsequently: 0–140 min: 2–30% B, 140–150 min: 30–60%B, 150–160 min: 60–99% B, 160–170 min: stay at 99% B.The nano-LC was connected to a linear quadrupole ion trap-Orbitrap (LTQ Orbitrap) mass spectrometer (ThermoElectron,Bremen, Germany) equipped with a nano-ESI source. Themass spectrometer was operated in the data-dependentmode to automatically switch between Orbitrap-MS and LTQ-MS/MS acquisition. Survey full scan MS spectra (from m/z300 to 1500) were acquired in the Orbitrap with resolutionR = 60 000 at m/z 400 (after accumulation to a target of1 000 000 charges in the LTQ). The method used allowedsequential isolation of the most intense ions, up to five,depending on signal intensity, for fragmentation on the linearion trap using collisionally induced dissociation at a targetvalue of 100 000 charges. Target ions already selectedfor MS/MS were dynamically excluded for 30 s. Generalmass spectrometry conditions were: electrospray voltage,1.25–1.4 kV; no sheath and auxiliary gas flow. The ion selec-tion threshold was 500 counts for MS/MS. Furthermore,an activation Q-value of 0.25 and activation time of 30 mswere applied for MS/MS. The resulted peptide MS/MSspectra were identified by Mascot search (http://www.matrixscience.com). Mascot was searched with a fragmention mass tolerance of 1.00 Da and a parent ion tolerance of10.0 p.p.m. Iodacetamide derivatives of cysteine were speci-fied in Mascot as a fixed modification. Protein identificationwas carried out by blast of peptides fragments against NCBI-BLASTP (http://blast.ncbi.nlm.nih.gov/Blast.cgi) or againstthe translated metagenome sequence of the iron-reducingculture BF in a pedant database (Munich information centrefor protein sequences) by using the Scaffold 2.02.03 software(Proteome Software, Portland, OR, USA). Protein identi-fication was accepted if the probability was greater than95% and when at least two peptides were identified. Proteinprobabilities were assigned by the Protein Prophet algorithm(Nesvizhskii et al., 2003).

N-terminal sequence analysis

A 60 kDa protein band specifically expressed in benzene-grown cells was transferred from SDS gels onto a PVDFmembrane using a semi-dry blotting apparatus (Bio-RADTrans-Blot SD) at 1 mA cm-2 for 2 h. The PVDF membranewas stained with Coomassie brilliant blue and washed with50% methanol. Coomassie-stained proteins were excisedfrom the PVDF membrane and collected fractions were sub-jected to N-terminal sequence analysis using a 492-proteinsequencer (Applied Biosystems, Darmstadt, Germany)



http://www.matrixscience.com

http://www.matrixscience.com

http://blast.ncbi.nlm.nih.gov/Blast.cgi

according to the manufacturer’s instructions. The obtainedN-terminal sequence was used for protein identification bysearch against the translated metagenome sequence of theiron-reducing culture BF. The analysis was carried out withtwo separate biological replicates.

Acknowledgements

The authors would like to thank Hakan Sarioglou for theassistance in proteomic analysis, and Reinhard Mentele forthe Edman sequencing. This project was supported by theDAAD-Helmholtz PhD exchange program through an operat-ing grant awarded to Nidal Abu Laban and by priority program1319 funded by the German Research Foundation (DFG).

References

Abu Laban, N., Selesi, D., Jobelius, C., and Meckenstock, R.(2009) Anaerobic benzene degradation by Gram-positivesulfate-reducing bacteria. FEMS Microbio Ecol 68: 300–311.

Achong, G., Rodríguez, A., and Spormann, A. (2001)Benzylsuccinate synthase of Azoarcus sp. strain T: cloning,sequencing, transcriptional organization, and its role inanaerobic toluene and m-xylene mineralization. J Bacteriol183: 6763–6770.

Anderson, R.T., Rooney-Varga, J.N., Gaw, C.V., and Lovley,D.R. (1998) Anaerobic benzene oxidation in the Fe(III)reduction zone of petroleum-contaminated aquifers.Environ Sci Technol 32: 1222–1229.

Ball, H.A., Johnson, H.A., Reinhard, M., and Spormann, A.M.(1996) Initial reactions in anaerobic ethylbenzene oxidationby a denitrifying bacterium, strain EB1. J Bacteriol 178:5755–5761.

Beller, H., and Spormann, A. (1998) Analysis of the novelbenzylsuccinate synthase reaction for anaerobic tolueneactivation based on structural studies of the product.J Bacteriol 180: 5454–5457.

Biegert, T., Fuchs, G., and Heider, J. (1996) Evidence thatanaerobic oxidation of toluene in the denitrifying bacteriumThauera aromatica is initiated by formation of benzylsuc-cinate from toluene and fumarate. Eur J Biochem 238:661–668.

Boll, M. (2005) Dearomatizing benzene ring reductases.J Mol Microbiol Biotechnol 10: 132–142.

Boll, M., and Fuchs, G. (1995) Benzoyl-CoA reductase(dearomatizing), a key enzyme of anaerobic aromaticmetabolism. ATP-dependence of the reaction, purificationand some properties of the enzyme from Thauera aro-matica strain K 172. Eur J Biochem 234: 921–933.

Boll, M., Fuchs, G., Meier, C., Trautwein, A., and Lowe, D.J.(2000) EPR and Mossbauer studies of benzoyl-CoA reduc-tase. J Biol Chem 275: 31857–31868.

Breinig, S., Schiltz, E., and Fuchs, G. (2000) Genes involvedin anaerobic metabolism of phenol in the bacteriumThauera aromatica. J Bacteriol 182: 5849–5863.

Bugg, T. (2003) Dioxygenase enzymes: catalytic mecha-nisms and chemical models. Tertahedro 65: 7075–7101.

Burland, S.M., and Edwards, E.A. (1999) Anaerobic benzenebiodegradation linked to nitrate reduction. Appl EnvironMicrobiol 65: 529–533.

Butler, J., He, Q., Nevin, K., He, Z., Zhou, J., and Lovley, D.(2007) Genomic and microarray analysis of aromaticsdegradation in Geobacter metallireducens and comparisonto a Geobacter isolate from a contaminated field site. BMCGenomics 8: 180.

Caldwell, M., and Suflita, J.M. (2000) Detection of phenol andbenzoate as intermediates of anaerobic benzene biodeg-radation under different terminal electron-accepting condi-tions. Environ Sci Technol 34: 1216–1220.

Carmona, M., Zamarro, M., Blazquez, B., Durante-Rodriguez, G., Juarez, J., Valderrama, J., et al. (2009)Anaerobic catabolism of aromatic compounds: a geneticand genomic view. Microbiol Mol Biol Rev 73: 71–133.

Chakraborty, R., and Coates, J. (2005) Hydroxylation andcarboxylation – two crucial steps of anaerobic benzenedegradation by Dechloromonas strain RCB. Appl EnvironMicrobiol 71: 5427–5432.

Chang, W., Um, Y., and Pulliam Holoman, T. (2005) Molecularcharacterization of anaerobic microbial communities frombenzene-degrading sediments under methanogenic condi-tions. Biotechnol Prog 21: 1789–1794.

Coates, J., Chakraborty, R., Lack, J., O’Connor, S.,Cole, K.A., Bender, K.S., and Achenbach, L.A. (2001)Anaerobic benzene oxidation coupled to nitrate reductionin pure culture by two strains of Dechloromonas. Nature411: 1039–1043.

Enroth, C., Eger, B., Okamoto, K., Nishino, T., Nishini, T., andPai, E. (2000) Crystal structures of bovine milk xanthinedehydrogenase and xanthine oxidase: structure basedmechanism of conversion. Proc Natl Acad Sci USA 97:10723–10728.

Gibson, D., Koch, J., and Kallio, R. (1968) Oxidative degra-dation of aromatic hydrocarbons by microorganisms. I.Enzymatic formation of catechol from benzene. Biochem-istry 7: 2653–2662.

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

Heintz, D., Gallien, S., Wischgoll, S., Ullmann, A.,Schaeffer, C., Kretzschmar, A., et al. (2009) Differentialmembrane proteome analysis reveals novel proteinsinvolved in the degradation of aromatic compounds inGeobacter metallireducens. Mol Cell Proteomics 8: 2159–2169.

Hopper, D., Bossert, I., and Rhodes-Roberts, M. (1991)p-Cresol methylhydroxylase from a denitrifying bacteriuminvolved in anaerobic degradation of p-cresol. J Bacteriol173: 1298–1301.

Johnson, S., Woolhouse, K., Prommer, H., Barry, D., andChristofi, N. (2003) Contribution of anaerobic microbialactivity to natural attenuation of benzene in groundwater.Engin Geol 70: 343–349.

Kasai, Y., Takahata, Y., Manefield, M., and Watanabe, K.(2006) RNA-based stable isotope probing and isolationof anaerobic benzene-degrading bacteria from gasoline-contaminated groundwater. Appl Environ Microbiol 72:3586–3592.

Kazumi, J., Caldwell, M.E., Suflita, J.M., Lovley, D.R., andYoung, L.Y. (1997) Anaerobic degradation of benzene indiverse anoxic environments. Environ Sci Technol 31:813–818.



Kniemeyer, O., and Heider, J. (2001) Ethylbenzene dehy-drogenase, a novel hydrocarbon-oxidizing molybdenum/iron-sulfur/heme enzyme. J Biol Chem Geol 276: 21381–21386.

Krieger, C.J., Beller, H.R., Reinhard, M., and Spormann, A.M.(1999) Initial reactions in anaerobic oxidation of m-xyleneby the denitrifying bacterium Azoarcus sp. strain T.J Bacteriol 181: 6403–6410.

Kunapuli, U., Lueders, T., and Meckenstock, R. (2007) Theuse of stable isotope probing to identify key iron-reducingmicroorganisms involved in anaerobic benzene degrada-tion. ISME J 1: 643–653.

Kunapuli, U., Griebler, C., Beller, H., and Meckenstock, R.U.(2008) Identification of intermediates formed duringanaerobic benzene degradation by an iron-reducingenrichment culture. Environ Microbiol 10: 1703–1712.

Kung, J., Löffler, C., Dörner, K., Heintz, D., Gallien, S.,Van Dorsselaer, A., et al. (2009) Identification and charac-terization of the tungsten-containing class of benzoyl-coenzyme A reductases. Proc Natl Acad Sci USA 106:17687–17692.

Laempe, D., Eisenreich, W., Bacher, A., and Fuchs, G.(1998) Cyclohexa-1,5-diene-1-carbonyl-CoA hydratase, anenzyme involved in anaerobic metabolism of benzoyl-CoAin the denitrifying bacterium Thauera aromatica. Eur JBiochem 255: 618–627.

Laempe, D., Jahn, M., and Fuchs, G. (1999)6-Hydroxycyclohex-1-ene-1-carbonyl-CoA dehydroge-nase and 6-oxocyclohex-1-ene-1-carbonyl-CoA hydrolase,enzymes of the benzoyl-CoA pathway of anaerobic aro-matic metabolism in the denitrifying bacterium Thaueraaromatica. Eur J Biochem 263: 420–429.

Larimer, F.W., Chain, P., Hauser, L., Lamerdin, J., Malfatti, S.,Do, L., et al. (2004) Complete genome sequence ofthe metabolically versatile photosynthetic bacteriumRhodopseudomonas palustris. Nat Biotechnol 22: 55–61.

Leuthner, B., Leutwein, C., Schulz, H., Hörth, P., Haehnel, W.,Schiltz, E., et al. (1998) Biochemical and genetic charac-terization of benzylsuccinate synthase from Thauera aro-matica: a new glycyl radical enzyme catalysing the firststep in anaerobic toluene metabolism. Mol Microbiol 28:615–628.

Lovley, D.R., Coates, J.D., Woodward, J.C., and Phillips, E.J.(1995) Benzene oxidation coupled to sulfate reduction.Appl Environ Microbiol 61: 953–958.

Lovley, D.R., Woodward, J.C., and Chapelle, F.H. (1996)Rapid anaerobic benzene oxidation with a variety of che-lated Fe(III) forms. Appl Environ Microbiol 62: 288–291.

McInerney, M., Rohlin, L., Mouttaki, H., Kim, U., Krupp, R.,Rios-Herna’ndez, L., et al. (2007) The genome of Syntro-phus aciditrophicus: life at the thermodynamic limit ofmicrobial growth. Proc Natl Acad Sci USA 104: 7600–7605.

Matsunaga, T., Okamura, Y., Fukuda, Y., Wahyudi, A.,Murase, Y., and Takeyama, H. (2005) Complete genomesequence of the facultative anaerobic magnetotactic bac-terium Magnetospirillum sp. strain AMB-1. DNA Res 12:157–166.

Morasch, B., and Meckenstock, R.U. (2005) Anaerobic deg-radation of p-xylene by a sulfate-reducing enrichmentculture. Curr Microbiol 51: 127–130.

Morasch, B., Annweiler, E., Warthmann, R., and Mecken-stock, R. (2001) The use of a solid adsorber resin forenrichment of bacteria with toxic substrates and to identifymetabolites: degradation of naphthalene, o-, and m-xyleneby sulfate-reducing bacteria. J Microbiol Meth 44: 183–191.

Morasch, B., Schink, B., Tebbe, C.C., and Meckenstock, R.U.(2004) Degradation of o-xylene and m-xylene by a novelsulfate-reducer belonging to the genus Desulfotomaculum.Arch Microbiol 181: 407–417.

Müller, J.A., Galushko, A.S., Kappler, A., and Schink, B.(2001) Initiation of anaerobic degradation of p-cresol byformation of 4-hydroxybenzylsuccinate in Desulfobacte-rium cetonicum. J Bacteriol 183: 752–757.

Musat, F., and Widdel, F. (2008) Anaerobic degradation ofbenzene by a marine sulfate-reducing enrichment culture,and cell hybridization of the dominant phylotype. EnvironMicrobiol 10: 10–19.

Musat, F., Galushko, A., Jacob, J., Widdel, F., Kube, M.,Reinhardt, R., et al. (2008) Anaerobic degradation ofnaphthalene and 2-methylnaphthalene by strains ofmarine sulfate-reducing bacteria. Environ Microbiol 11:209–219.

Narmandakh, A., Gad’on, N., Drepper, F., Knapp, B.,Haehnel, W., and Fuchs, G. (2006) Phosphorylation ofphenol by phenylphosphate synthase: role of histidinephosphate in catalysis. J Bacteriol 188: 7815–7822.

Nesvizhskii, A..I., Keller, A., Kolker, E., and Aebersold, R.(2003) A statistical model for identifying proteins by tandemmass spectrometry. Anal Chem 75: 4646–4658.

Oka, A.R., Phelps, C.D., McGuinness, L.M., Mumford, A.,Young, L.Y., and Kerkhof, L.J. (2008) Identification ofcritical members in a sulfidogenic benzene-degradingconsortium by DNA stable isotope probing. Appl EnvironMicrobiol 74: 6476–6480.

Peters, F., Heintz, D., Johannes, J., Dorsselaer, A., and Boll,M. (2007) Genes, enzymes, and regulation of para-cresolmetabolism in Geobacter metallireducens. J Bacteriol 189:4729–4738.

Phelps, C., Kerkhof, L., and Young, L. (1998) Molecularcharacterization of a sulfate-reducing consortium whichmineralizes benzene. FEMS Microbiol Ecol 27: 269–279.

Phelps, C.D., Zhang, X., and Young, L.Y. (2001) Use of stableisotopes to identify benzoate as a metabolite of benzenedegradation in a sulphidogenic consortium. Environ Micro-biol 3: 600–603.

Rabus, R., and Widdel, F. (1995) Anaerobic degradationof ethylbenzene and other aromatic hydrocarbons by newdenitrifying bacteria. Arch Microbiol 163: 96–103.

Rabus, R., Kube, M., Heider, J., Beck, A., Heitmann, K.,Widdel, F., and Reinhardt, R. (2005) The genomesequence of an anaerobic aromatic-degrading denitrifyingbacterium, strain EbN1. Arch Microbiol 183: 27–36.

Rooney-Varga, J., Anderson, R., Fraga, J., Ringelberg, D.,and Lovley, D.R. (1999) Microbial communitiesassociated with anaerobic benzene degradation in apetroleum-contaminated aquifer. Appl Environ Microbiol65: 3056–3063.

Salinero, K., Keller, K., Feil, W., Feil, H., Trong, S., Di Bartolo,G., and Lapidus, A. (2009) Metabolic analysis of the soilmicrobe Dechloromonas aromatica str. RCB: indications



of a surprisingly complex life-style and cryptic anaerobicpathways for aromatic degradation. BMC Genomics 10:351.

Schleinitz, K., Schmeling, S., Jehmlich, N., von Bergen, M.,Harms, H., Kleinsteuber, S., et al. (2009) Phenol degra-dation in the strictly anaerobic iron-reducing bacteriumGeobacter metallireducens GS-15. Appl Environ Microbiol75: 3912–3919.

Schmeling, S., Naramandakh, A., Schmitt, O., Gad’on, N.,Schüle, K., and Fuchs, G. (2004) Phenylphosphate syn-thase: a new phosphotransferase catalyzing the first stepin anaerobic phenol metabolism in Thauera aromatica.J Bacteriol 186: 8044–8057.

Schühle, K., and Fuchs, G. (2004) Phenylphosphate car-boxylase: a new C-C lyase involved in anaerobic phenolmetabolism in Thauera aromatica. J Bacteriol 186: 4556–4567.

Schühle, K., Gescher, J., Feil, U., Paul, M., Jahn, M., Schäg-ger, H., and Fuchs, G. (2003) Benzoate-coenzyme A ligasefrom Thauera aromatica: an enzyme acting in anaerobicand aerobic pathways. J Bacteriol 185: 4920–4924.

Ulrich, A.C., Beller, H., and Edwards, E.A. (2005) Metabolitesdetected during biodegradation of 13C6-benzene in nitrate-reducing and methanogenic enrichment cultures. EnvironSci Technol 39: 6681–6691.

Ulrich, A.C., and Edwards, E.A. (2003) Physiologicaland molecular characterization of anaerobic benzene-degrading mixed cultures. Environ Microbiol 5: 92–102.

Vogel, T.M., and Grbic-Galic, D. (1986) Incorporation ofoxygen from water into toluene and benzene duringanaerobic fermentative transformation. Appl EnvironMicrobiol 52: 200–202.

Winderl, C., Anneser, B., Griebler, C., Meckenstock, R.U.,and Lueders, T. (2008) Depth-resolved quantificationof anaerobic toluene degraders and aquifer microbial

community patterns in distinct redox zones of a tar oilcontaminant plume. Appl Environ Microbiol 74: 792–801.

Wischgoll, S., Heintz, D., Peters, F., Erxleben, A., Sar-nighausen, E., Reski, R., et al. (2005) Gene clustersinvolved in anaerobic benzoate degradation of Geobactermetallireducens. Mol Microbiol 58: 1238–1252.

Zehr, B., Savin, T., and Hall, R. (1989) A one-step, low back-ground coomassie staining procedure for polyacrylamidegels. Anal Biochem 182: 157–159.

Supporting information

Additional Supporting Information may be found in the onlineversion of this article:

Fig. S1. Formation of Fe(II) during growth of enrichmentculture BF on benzene ( ), phenol (�) and benzoate (�).Arrows indicate the time points of sampling for proteomicanalysis.Fig. S2. Terminal restriction fragment length polymorphism(T-RFLP) fingerprinting of the iron-reducing culture BF grownon (a) benzene, (b) phenol and (c) benzoate. Numbersrepresent the length of major T-RFs in base pairs (bp).Table S1. General features of the benzene-grown, iron-reducing culture metagenome.Table S2. Properties of the aromatic-degrading genes andtranscriptional regulators of the iron-reducing culture BF.Table S3. Genes present in putative mobile elements identi-fied in the aromatic-degrading contigs of the culture BFmetagenome.

Please note: Wiley-Blackwell are not responsible for thecontent or functionality of any supporting materials suppliedby the authors. Any queries (other than missing material)should be directed to the corresponding author for the article.



identification of enzymes involved in anaerobic benzene ... · identiﬁcation of enzymes involved...

Documents