1. Journal of Microbiological Methods 64 (2006) 250 – 265
www.elsevier.com/locate/jmicmeth
Alternative primer sets for PCR detection of genotypes involved
in bacterial aerobic BTEX degradation: Distribution of the
genes in BTEX degrading isolates and in subsurface
soils of a BTEX contaminated industrial site
Barbara Hendrickxa,b, Howard Juncac, Jolana Vosahlovad, Antje Lindnere, Irene Rueggf,
¨
f g f e
Margarete Bucheli-Witschel , Folkert Faber , Thomas Egli , Margit Mau ,
Michael Schlomanne, Maria Brennerovad, VLadimir Brennerd,
¨
Dietmar H. Pieperc, Eva M. Topb,1, Winnie Dejonghea,
Leen Bastiaensa, Dirk Springaela,h,*
a
Environmental and Process Technology (Vito), Flemish Institute for Technological Research, B-2400 Mol, Belgium
b
Laboratory of Microbial Ecology and Technology, University of Ghent (UG), B-9000 Gent, Belgium
c
Biodegradation Group, German Research Centre for Biotechnology (GBF), 38124 Braunschweig, Germany
d
Department of Cell and Molecular Microbiology, Institute of Microbiology, 14200 Prague 4, Czech Republic
e ¨
Interdisziplinares Okologisches Zentrum, TU Bergakademie Freiberg, D-09599 Freiberg, Germany
¨
f
Department of Microbiology, Swiss Federal Institute for Environmental Science and Technology (EAWAG), 8600 Dubendorf, Switzerland
¨
g
Department of Microbiology, Centre Ecological Evolutionary Studies, University of Groningen, 9750 AA Haren, The Netherlands
h
Laboratory of Soil and Water Management, Catholic University of Leuven, Kasteelpark Arenberg 20, B-3001 Heverlee, Belgium
Received 1 August 2004; received in revised form 6 April 2005; accepted 11 May 2005
Available online 8 June 2005
Abstract
Eight new primer sets were designed for PCR detection of (i) mono-oxygenase and dioxygenase gene sequences involved in
initial attack of bacterial aerobic BTEX degradation and of (ii) catechol 2,3-dioxygenase gene sequences responsible for meta-
cleavage of the aromatic ring. The new primer sets allowed detection of the corresponding genotypes in soil with a detection
limit of 103–104 or 105–106 gene copies gÀ 1 soil, assuming one copy of the gene per cell. The primer sets were used in PCR to
assess the distribution of the catabolic genes in BTEX degrading bacterial strains and DNA extracts isolated from soils sampled
from different locations and depths (vadose, capillary fringe and saturated zone) within a BTEX contaminated site. In both soil
DNA and the isolates, tmoA-, xylM- and xylE1-like genes were the most frequently recovered BTEX catabolic genes. xylM and
* Corresponding author. Present address: Catholic University of Leuven (KUL), Laboratory for Soil and Water Management, Kasteelpark
Arenberg 20, 3001 Heverlee, Belgium. Tel.: +32 16 321604; fax: +32 16 321997.
E-mail address: dirk.springael@agr.kuleuven.ac.be (D. Springael).
1
Present address: University of Idaho, Department of Biological Sciences, 83844-3051 Moscow, Idaho, USA.
0167-7012/$ - see front matter D 2005 Elsevier B.V. All rights reserved.
doi:10.1016/j.mimet.2005.04.018

2. B. Hendrickx et al. / Journal of Microbiological Methods 64 (2006) 250–265 251
xylE1 were only recovered from material from the contaminated samples while tmoA was detected in material from both the
contaminated and non-contaminated samples. The isolates, mainly obtained from the contaminated locations, belonged to the
Actinobacteria or Proteobacteria (mainly Pseudomonas). The ability to degrade benzene was the most common BTEX
degradation phenotype among them and its distribution was largely congruent with the distribution of the tmoA-like genotype.
The presence of tmoA and xylM genes in phylogenetically distant strains indicated the occurrence of horizontal transfer of
BTEX catabolic genes in the aquifer. Overall, these results show spatial variation in the composition of the BTEX degradation
genes and hence in the type of BTEX degradation activity and pathway, at the examined site. They indicate that bacteria
carrying specific pathways and primarily carrying tmoA/xylM/xylE1 genotypes, are being selected upon BTEX contamination.
D 2005 Elsevier B.V. All rights reserved.
Keywords: PCR detection; Aerobic BTEX biodegradation; Catabolic gene distribution; BTEX degrading isolates; BTEX contaminated site
1. Introduction further degraded into Krebs cycle intermediates (Har-
ayama and Rekik, 1993). Phylogenetic studies of
BTEX (benzene, toluene, ethylbenzene and amino acid sequences of the proteins involved show
xylenes) are frequently occurring groundwater con- that they can be divided into specific families and
taminants. BTEX can be biodegraded under both subfamilies showing significant sequence homology
aerobic and anaerobic conditions and in situ bioreme- and indicating a common ancestry which allowed the
diation, either passive or active, is increasingly ap- design of group-specific primer sets for detection by
plied for the elimination of BTEX in groundwater PCR (Baldwin et al., 2003; Eltis and Bolin, 1996). In
(Lovley, 2001; Barker et al., 1987). Decisions as to the past, several studies have reported PCR primers to
whether a site should be contained and monitored or detect and quantify the presence of specific genotypes
actively treated are largely made on an empirical encoding those key steps in BTEX biodegradation and
basis. Basic knowledge about the distribution, popu- their mRNA in environmental samples (Baldwin et
lation densities and activities of BTEX degrading al., 2003; Hallier-Soulier et al., 1996; Junca and Pie-
organisms at the polluted sites can contribute to per, 2003; Mesarch et al., 2000; Meyer et al., 1999;
rational decision-making (Baldwin et al., 2003). The Ogram et al., 1995; Okuta et al., 1998; Ringelberg et
ability to rapidly and accurately detect BTEX biode- al., 2001). However, the recent availability of a lot of
grading bacteria and their activity in the environment new sequence information on BTEX degradation
is therefore of major interest. This can be done by genes indicates that previously published primer sets
demonstrating the occurrence of catabolic genotypes are not always suitable (Baldwin et al., 2003; Junca
involved in BTEX degradation or their corresponding and Pieper, 2003). In addition, for some BTEX cata-
mRNA in the aquifer by employing sensitive PCR and bolic protein families there are no primers reported yet
RT-PCR detection methods (Baldwin et al., 2003). (Baldwin et al., 2003). Therefore, design and/or rede-
Initial oxidative attack of BTEX converting the sign of primers for PCR detection of BTEX catabolic
compound into a catechol structure and the cleavage genes is required.
of the catechol structure are key steps in aerobic Previously, we reported the design of a degenerate
BTEX degradation. As such, both activities are of primer set for the detection of tmoA-like genes, encod-
direct interest as monitoring objects. Initial oxidative ing a-subunit of subfamily 2 of the hydroxylase
attack consists of direct oxidation of the aromatic ring component of bacterial multi-component mono-oxy-
via a mono-oxygenase (Kahng et al., 2001) or a genases involved in BTEX degradation (Hendrickx
dioxygenase attack (Zylstra and Gibson, 1989; Furu- et al., submitted). In this study, we report (i) an alter-
kawa et al., 1993) or oxidation of the alkyl side chain native primer set for detection of genes encoding
which is catalyzed by mono-oxygenases (Burlage et subgroup I of subfamily 1 a-subunit of the hydroxy-
al., 1989). Ring cleavage occurs by catechol 2,3- lase component of bacterial multi-component mono-
dioxygenases (C23O) after which the structure is oxygenases, (ii) a new primer set for detecting genes

3. 252 B. Hendrickx et al. / Journal of Microbiological Methods 64 (2006) 250–265
involved in mono-oxygenase attack of the side chain and p-xylene which exceeded 10,000 times the regu-
of TEX, (iii) a new primer set for detecting genes latory criteria for groundwater concentrations of ben-
encoding for iron–sulfur a-subunit of dioxygenase zene in European countries. At A3, the concentration
complexes and (iv) 4 alternative or novel primer sets of benzene in the groundwater was 0.1 mg lÀ 1 while
for detection of the corresponding genes of the 4 major A4 was not contaminated. The contents of the BTEX
subfamilies of C23O proteins. The distribution of these compounds in the groundwater were chemically de-
genes in a set of BTEX degrading bacterial isolates and termined by means of the US EPA 8260 modified
total community DNA from subsurface soil samples of method. Subsurface soil sample GP48 was a sandy
a BTEX contaminated site was explored. aquifer sample obtained from a BTEX contaminated
site situated in Belgium. It originated from a sampling
point where the total BTEX concentration in the
2. Materials and methods groundwater was around 50 mg lÀ 1.
2.1. Bacterial strains and growth conditions 2.3. Isolation of BTEX degraders
The bacterial BTEX degrading reference strains Bacterial strains were isolated from the soil sam-
used in this study are described in Table 1 and were ples by either one of 5 different isolation proce-
routinely grown at 30 8C on 869 medium or on Tris dures. Procedure 1 consisted of direct plating on
minimal medium (Mergeay et al., 1985) supplemen- R2A agar plates in the presence of BTEX vapors as
ted with BTEX as described (Hendrickx et al., in described previously (Junca and Pieper, 2003). After
press). Phenanthrene and biphenyl were provided as incubation at 30 8C, colonies expressing meta-cleav-
a carbon source by placing crystals on the agar plates age activity were retained. Procedure 2 consisted of
or adding them to the liquid cultures. BTEX, phen- enrichment in chloride free mineral medium MM5
anthrene and biphenyl were purchased from Janssen (Hickey and Focht, 1990) supplemented with 50 mg
Chemica (Beerse, Belgium). lÀ 1 BTEX 1:1:1:1 mixture as a sole source of
carbon and energy. Growing cultures were plated
2.2. Subsurface soil samples used in this study on MM5 agar with BTEX and developing colonies
were retained. In procedure 3, BTEX degraders
Subsurface soil samples used for direct DNA ex- were isolated from a micro-aerophylic chemostat
traction were sampled in March 2001 at an oil-refin- inoculated with the soil samples (Faber, unpublished
ery site, situated in Northern Bohemia (Czech results). In procedure 4, BTEX degrading strains
Republic) containing a BTEX (mainly benzene) con- were isolated from an enrichment in Flavobacterium
taminated groundwater plume. Soil samples for isola- aquatile medium, containing 0.25 g lÀ 1 yeast ex-
tion of BTEX degrading isolates were obtained from tract, 0.5 g lÀ 1 peptone, 1 g lÀ 1 caseinate, 0.25 g
the same site in March and April 2001 and in April lÀ 1 K2HPO4, cycloheximide (75 mg lÀ 1) and dif-
and May 2002. The subsurface soil samples were ferent contents of soil (2.5%, 5%, and 10% wt/vol)
taken from different locations along the plume (A1, at 14 8C. Procedure 5 consisted of enrichment of
A2, A3 and A4) and from different depths designated BTEX degraders at 12 8C in mineral medium mod-
as X, Y and Z, in which X represents the vadose zone ified from Evans et al. (1970), diluted to 25% of its
(depth of 2.58 m), Y the capillary fringe (depth of original strength, adjusted to pH 7 by addition of a
2.91 m) and Z the saturated zone (depth of 3.84 m). potassium phosphate buffer (50 mM) and further
The tested soil samples were accordingly designated modified by adding 162 mg lÀ 1 CaCl2 d 2H2O, by
as A1X, A1Y, A1Z, A2X, A2Y, A2Z, A3X, A3Y, replacing citric acid with EDTA and a vitamin
A3Z and A4Z. All were mixtures of three soil samples solution (Egli et al., 1988) with a BTEX mixture
taken at the same place and depth within a distance of supplied via the vapour phase. After 6 days the
1 m. The groundwater at sampling point A1 and A2 culture was plated on a mineral medium plate and
was strongly contaminated by BTEX containing 100 growing colonies retained after incubation in a
mg lÀ 1 benzene, 10 mg lÀ 1 toluene and l mg lÀ 1 m- BTEX atmosphere at 30 8C.

4. Table 1
Description of used reference strains and results of PCR detection of BTEX degradation genes with the developed primer sets on reference strains
Organism BTEX degrading Reference(s) Initial attack Lower pathway (meta-cleavage)
capacity and relevant Mono-oxygenase Dioxygenase
catabolic genotypesa
B. Hendrickx et al. / Journal of Microbiological Methods 64 (2006) 250–265
TBMD TMOA TOL XYLA TODC1 XYLE1 XYLE2 CDO TBUE TODE
-F/-R -F/-R -F/-R -F/-R -F/-R -F/-R -F/-R -F/-R -F/-R -F/-R
B. sp. strain JS150 B, T, EB, tbmABCDEF, (Johnson and Olsen, 1997; + + À À + À À À À +
tbc2ABCDEF Kahng et al., 2001)
B. cepacia G4 B, T, tomA012345, tomB (Shields et al., 1995) + À À À À À À À À À
R. pickettii PKO1 B, T, EB, o-X, m-X, p-X, (Byrne et al., 1995) + + À À À À À À + À
tbuA1UBVA2C (tbuT),
tbuD (tbuR), tbuEFGKIHJ
(tbuS), tbuX
P. mendocina KR1 B, T, tmoXABCDEF (tmoST) (Yen et al., 1991) + + À À À À À À À À
P. aeruginosa JI104 B, bmoABCDEF (Kitayama et al., 1996a) + + À À + + À À À À
P. stutzeri OX1 B, T, o-X, touABCDEF (touR), (Bertoni et al., 1998) + + + + À + À À À À
xylAM
P. putida F1 B, T, EB, todFC1C2BADE, (Zylstra and Gibson, 1989) À À À À + À À À À +
todGIH, (todST)
P. putida mt-2 (PaW1) T, m-X, p-X, xylUWCMABN, (Burlage et al., 1989) À À + + À + À À À À
xylXYZLTEGFJQKIH,
(xylS, xylR)
P. putida MT15 T, m-X, p-X, xylUWCMABN, (Keil et al., 1985a) À À + À À + À + À À
xylXYZLTEGFJQKIH, cdo
P. putida MT53 T, m-X, p-X, xylUWCMABN, (Kok et al., 1999) À À + + À + À + À À
xylXYZLTEGFJQKIH, cdo
S. yanoikuyae B1 T, m-X, p-X, xylMAB, (Kim and Zylstra, 1999) À À À À À À + À À À
xylCXYFEGJQKIHT
P. putida JHR Biphenyl, Springael, unpublished À À À À + À À À À À
bphA1A2A3A4BCDEFG
B, T, EB, o-X, m-X, p-X: growth on benzene, toluene, ethylbenzene, o-xylene, m-xylene, p-xylene, respectively; + = PCR signal; À = no PCR signal.
a
The genes encoding the hydroxylase a-subunit of the aromatic ring mono-oxygenases, the terminal hydroxylase component and electron transfer component of the side chain
mono-oxygenases, the iron–sulfur oxygenase a-subunit of the dioxygenase complex and the catechol 2,3-dioxygenases are underlined.
253

5. 254 B. Hendrickx et al. / Journal of Microbiological Methods 64 (2006) 250–265
2.4. Extraction of total DNA from bacterial cultures TBUE-R and TODE-F/TODE-R were designated as
and from soil tbmD-, tmoA-, xylM-, xylA-, todC1-, xylE1-, xylE2-,
cdo-, tbuE- and todE-like genes, respectively. Novel
Total genomic DNA was extracted from bacterial generate and degenerate primers were designed based
strains grown in 5 ml Tris minimal medium exposed on Protein and DNA sequence alignments of the
to BTEX vapors or in 5 ml 869 medium as described appropriate proteins/genes, constructed using the
by Vanbroekhoven et al. (2004). Extraction of total GCG Wisconsin protein and DNA analysis program
DNA from aquifer material was performed as de- (version 7.0) (Genetics Computer Group, Madison,
scribed previously (Hendrickx et al., in press). Wisconsin).
2.5. BOX-PCR fingerprint analysis 2.7. PCR amplification of catabolic genes from pure
strain and soil DNA
PCR amplification of BOX repetitive sequences
and pattern analysis was performed as described pre- PCR amplification was carried out in a 50
viously (Vanbroekhoven et al., 2004). For all strains, Al reaction mixture containing 100 ng of pure strain
DNA concentrations were adjusted to 100 ng AlÀ 1 DNA or soil DNA as templates. A 475-bp xylM gene
and 1 Al was used in the PCR reaction. fragment was amplified using primer set TOL-F/TOL-
R as described by Baldwin et al. (2003). Using the
2.6. Used primers and primer design TBMD-F/TBMD-R, TMOA-F/TMOA-R, TOL-F/
TOL-R, XYLA-F/XYLA-R, TODC1-F/TODC1-R,
Fig. 1 shows a schematic overview of the catabolic TBUE-F/TBUE-R or TODE-F/TODE-R primer sets,
reactions of relevance in this paper and the involved the PCR mixture contained 1.25 U exTaq polymerase,
target genotypes. The used primer sets, their target 10 pmol of the forward primer, 10 pmol of the reverse
genes, locations in the gene and product lengths are primer, 200 AM of each dNTP and 5 Al of 10x exTaq
described in Table 2. For convenience, genotypes reaction buffer (20 mM MgCl2). The Taq polymerase,
amplified with primer sets TBMD-F/TBMD-R, dNTPs and PCR buffer were purchased from TaKaRa
TMOA-F/TMOA-R, TOL-F/TOL-R, XYLA-F/ (TaKaRa Ex Taqk, TaKaRa Shuzo Co. Biomedical
XYLA-R, TODC1-F/TODC1-R, XYLE1-F/XYLE1- Group, Japan). Using the XYLE1-F/XYLE1-R,
R, XYLE2-F/XYLE2-R, CDO-F/CDO-R, TBUE-F/ XYLE2-F/XYLE2-R or CDO-F/CDO-R primer set,
(CH3)
tmoA tmoA
(CH3 ) (CH3 )
tbmD tbmD OH
OH
OH
CH 3 CH 2 OH OH
xylA Mono-oxygenase
Initial attack xylM OH
CH 3
CH 3 CH 3 CH 3
todC1 OH OH
H
Dioxygenase
H
OH OH
xylE1
OH xylE2 O Catechol
COOH
Ring cleavage 2,3-dioxygenase
OH tbuE OH
cdo
todE
Fig. 1. Schematic presentation of BTEX catabolic enzyme reactions catalyzed by the proteins detected by the primer sets.

6. Table 2
Primer sets used in this study
Primer pair Proteins targeted Sequence Amplicon PCR Reference
size (bp) annealing
temperature
(8C)
TBMD-F/TBMD-R Subfamily 1 of a-subunits of 5V-GCCTGACCATGGATGC(C/G)TACTGG-3V 640 65.5 This study
hydroxylase component of 5V-CGCCAGAACCACTTGTC(A/G)(A/G)TCCA-3V
multi-component
mono-oxygenases
B. Hendrickx et al. / Journal of Microbiological Methods 64 (2006) 250–265
TMOA-F/TMOA-R Subfamily 2 of a-subunits of 5V-CGAAACCGGCTT(C/T)ACCAA(C/T)ATG-3V 505 61.2 Hendrickx
hydroxylase component of 5V-ACCGGGATATTT(C/T)TCTTC(C/G)AGCCA-3V et al.,
multi-component submitted
mono-oxygenases
TOL-F/TOL-R Subfamily 5 of hydroxylase 5V-TGAGGCTGAAACTTTACGTAGA-3V 475 55 (Baldwin
component of 5V-CTCACCTGGAGTTGCGTAC-3V et al., 2003)
two-component side
chain mono-oxygenases
XYLA-F/XYLA-R Electron transfer component 5V-CCAGGTGGAATTTTCAGTGGTTGG-3V 291 64 This study
of two-component side 5V-AATTAACTCGAAGCGCCCACCCCA-3V
chain mono-oxygenases
TODC1-F/TODC1-R Subfamilies D.1.B + D.1.C + 5V-CAGTGCCGCCA(C/T)CGTGG(C/T)ATG-3V 510 66 This study
D.2.A + D.2.B + D.2.C of 5V-GCCACTTCCATG(C/T)CC(A/G)CCCCA-3V
a-subunits of Type D
iron–sulfur multi-component
aromatic dioxygenases
XYLE1-F/XYLE1-R Subfamily I.2.A of catechol 5V-CCGCCGACCTGATC(A/T)(C/G)CATG-3V 242 61.5 This study
extradiol dioxygenases 5V-TCAGGTCA(G/T)CACGGTCA(G/T)GA-3V
XYLE2-F/XYLE2-R Subfamily I.2.B of catechol 5V-GTAATTCGCCCTGGCTA(C/T)GTICA-3V 906 64 This study
extradiol dioxygenases 5V-GGTGTTCACCGTCATGAAGCG(C/G/T)TC-3V
CDO-F/CDO-R cdo (U01826) of subfamily 5V-CATGTCAACATGCGCGTAATG-3V 255 58 This study
I.2.C of catechol extradiol 5V-CATGTCTGTGTTGAAGCCGTA-3V
dioxygenases
TBUE-F/TBUE-R tbuE (U20258) of subfamily 5V-CTGGATCATGCCCTGTTGATG-3V 444 60 This study
I.2.C of catechol extradiol 5V-CCACAGCTTGTCTTCACTCCA-3V
dioxygenases
TODE-F/TODE-R todE (Y18245), todE (Y18245), 5V-GGATTTCAAACTGGAGACCAG-3V 246 58 This study
tobE (AF180147) of subfamily 5V-GCCATTAGCTTGCAGCATGAA-3V
I.3.B of catechol extradiol
dioxygenases
27F/1492R Eubacterial 16S rRNA genes 5V-AGAGTTTGATCCTGGCTCAG-3V 1465 55 (Johnson,
5V-TACGGYTACCTTGTTACGACTT-3V 1994)
38F/518R Eubacterial 16S rRNA genes 5V-GATCTTGGCTCAGGTTGAACGCTG-3V 480 55 (Muyzer
5V-ATTACCGCGGCTGCTGG-3V et al., 1993)
255

7. 256 B. Hendrickx et al. / Journal of Microbiological Methods 64 (2006) 250–265
the PCR mixture contained 67 mM Tris/HCl (pH 8.8), Purification Kit from the Westburg company (West-
16.6 mM (NH4)2SO4, 0.45% Triton X-100, 0.2 mg burg, Leusden, The Netherlands). The purified PCR
gelatine mLÀ 1, 120 AM of each dNTP, 2 mM MgCl2, products of the BTEX catabolic gene sequences were
10 pmol primer-F, 10 pmol primer-R, 1.25 U Taq sequenced double stranded by the Westburg company
DNA polymerase (5 U AlÀ 1). Primers were synthe- (Westburg, Leusden, The Netherlands). A similarity
sized by Westburg (Westburg BV, Leusden, The Neth- analysis of the DNA sequences was obtained by using
erlands). The PCR temperature/time profile used for the Advanced Blast Search program BLASTX for the
all primer sets, except for the XYLE2-F/XYLE2-R BTEX catabolic gene sequences and BLASTN for the
pair, was an initial denaturation of 5 min at 95 8C, 16S rRNA gene sequences available from GenBank
followed by 35 cycles of denaturation for 1 min at 94 (GenBank, National Centre for Biotechnology Infor-
8C, annealing for 1 min at the temperature reported in mation, Rockville Pike Bethesda, USA). The BTEX
Table 2 and elongation for 2 min at 72 8C. The last catabolic gene sequences reported in this study are
step included an extension for 10 min at 72 8C. The available from GenBank under accession numbers
profile used with the XYLE2-F/XYLE2-R primer set AY504971 to AY504995, while the 16S rRNA gene
consisted of an initial denaturation step of 5 min at 95 sequences are available under accession numbers
8C, followed by 35 cycles of denaturation for 15 s at AY510158 to AY510165, AY512600 to AY512644
94 8C, annealing for 30 s at 64 8C and elongation for and AY517534 to AY517541.
45 s at 72 8C. The last step was an extension for 3 min
at 72 8C. 1465-bp or 480-bp eubacterial 16S rRNA
gene fragments were amplified using respectively 3. Results and discussion
primer set 27F/1492R and primer set 38F/518R as
described (Johnson, 1994; Muyzer et al., 1993). In 3.1. Design of novel PCR primer sets targeting BTEX
all cases, PCR was performed on Biometra (Biometra, initial attack mono- and dioxygenases
Gottingen, Germany) or Perkin Elmer (Perkin Elmer,
¨
Connecticut, USA) PCR machines. PCR products The first novel primer set designed was primer set
were analyzed by 2% agarose gel electrophoresis as TMBMD-F/TBMDR allowing detection of all genes
described (Vanbroekhoven et al., 2004). encoding a-subunits of subfamily 1 of the hydroxy-
lase complex of aromatic mono-oxygenases (for the
2.8. Sensitivity of PCR method phylogenetic tree see Baldwin et al., 2003). The
members in subfamily 1 are primarily phenol and
To examine the limit of PCR detection for the cresol hydroxylase a-subunits, but include the tolu-
different primer sets, a known amount of viable ene/benzene 2-hydroxylase a-subunit (Tb2m)
cells of different relevant BTEX degrading strains encoded by the tbmD gene in strain JS150 (Johnson
was added to two different sterilized soils, A2Z and and Olsen, 1995) and the toluene 2-hydroxylase a-
GP48, at different final cell concentrations (i.e. ap- subunit (TOM) encoded by the tomA3 gene in Bur-
proximately 108, 106, 104, 102 CFU gÀ 1) prior to kholderia cepacia G4 (Shields et al., 1995). Because
DNA-extraction. Inoculum cells were harvested of the strong protein and DNA homology within the
from liquid cultures, washed twice and added in 100 phenol/cresol hydroxylase a-subunits, it was not
Al suspension of appropriate cell densities to 1 g of possible to design a primer set targeting only the
sterilized soil. After extraction, total soil DNA was tbmD and tomA3 genes. The novel TMBMD-F/
subsequently used as template for PCR with the dif- TBMDR primer set is an alternative for primer set
ferent developed primer sets. PHE-F/PHE-R reported by Baldwin et al. (2003).
However, primers TBMD-F and TBMD-R are less
2.9. Sequence analysis of the PCR amplified catabolic degenerate than PHE-F and PHE-R, which is of
gene and 16S rRNA gene fragments interest for diversity studies by DGGE (Hendrickx
et al., submitted) and the TBMD-F/TBMD-R PCR
Amplicons resulting from PCR with the different product is 434-bp longer than the PHE-F/PHE-R
primer sets were purified with the QIAquick PCR fragment and yields more sequence information.

8. B. Hendrickx et al. / Journal of Microbiological Methods 64 (2006) 250–265 257
Genes encoding more distantly related second, third ducts of the expected size and sequence (data not
and fourth subfamily a-subunits, should not be am- shown) for most of the strains it was expected for
plified, as they show 14 to 19 mismatches with the (data not shown). Nevertheless, some unexpected
forward primer TBMD-F and 14 to 17 mismatches results were obtained with some strains. tbmD-like
with the reverse primer TBMD-R. amplicons were also obtained with Pseudomonas
The new primers for detecting the BTEX side mendocina KR1 and Pseudomonas aeruginosa
chain mono-oxygenases were based on the xylA JI104 DNA. The deduced protein sequences of the
gene. BTEX side chain mono-oxygenases are two- PCR products from KR1 and JI104 showed 95%
component enzyme systems consisting of a terminal identity with the phenol hydroxylase a-subunit of P.
hydroxylase component encoded by the xylM gene putida P-8 (Accession no. BAA74744) and 99%
and an electron transfer component encoded by the identity with the phenol hydroxylase a-subunit
xylA gene. More xylA gene sequences are available in PhhN (Accession no. CAA55663), respectively. Sim-
GenBank than xylM gene sequences, and xylA and ilar results were reported by Baldwin et al. (2003)
xylM are always found linked on TOL plasmids using primer pair PHE-F/PHE-R. Because both KR1
(Shaw and Harayama, 1992; Sentchilo et al., 2000). and JI104 produce methyl-substituted phenols from
Moreover, a suitable primer set, i.e., TOL-F/TOL-R, toluene, the existence of downstream phenol hydro-
was recently reported for PCR detection of xylM xylases in that pathway and hence tbmD-related genes
(Baldwin et al., 2003). in those strains would be likely (Baldwin et al., 2003).
The novel degenerate primer set, i.e., TODC1-F/ Furthermore, no xylA PCR amplicon was obtained
TODC1-R, for detecting genes involved in a direct from P. putida MT15, although it carries the
dioxygenase attack of BTE, was designed to target pWW15 TOL plasmid and although a xylM PCR
simultaneously the genes of all five subfamilies of the amplicon was recovered using primer set TOL-F/
iron–sulfur oxygenase a-subunits of the dioxygenase TOL-R (Table 1.). These results indicate that either
complex. The phylogenetic tree of the iron–sulfur xylA is not always linked with xylM or the xylA gene
oxygenase a-subunits of the dioxygenase complexes, might show more sequence heterogeneity than de-
reported by Baldwin et al. (2003), comprises two duced from reported sequences. Finally, PCR with
major types of a-subunits. The first type (N) consists primer set TODC1-F/TODC1-R yielded unexpectedly
primarily of naphthalene dioxygenase a-subunits. A also PCR products with Burkholderia sp. strain JS150
second type (D) comprises two families, i.e., D.1 and and P. aeruginosa JI104 DNA. The deduced amino
D.2, which both consist of 3 subfamilies. Subfamilies acid sequence of the PCR product of strain JS150
D.1.A and D.2.C include BTEX dioxygenase a-sub- showed 98% amino acid identity with the terminal
units. The novel primer set is an alternative for the 5 oxygenase large subunit McbAa of Ralstonia sp.
dioxygenase primer sets recently described by Bald- JS705 involved in hydroxylation of chlorobenzene
win and co-workers (2003), which each allow the (Accession no. CAA06970) (van der Meer et al.,
detection of the genes of a specific subfamily, and 1998), whereas the PCR product of strain JI104
upgrades the primer set reported by Ogram et al. showed 100% identity with the biphenyl dioxygenase
(1995) for detection of the todC1 gene of Pseudomo- a subunit of biphenyl 2,3-dioxygenase BphA1 of
nas putida F1. The ladder is a non-degenerate primer Pseudomonas pseudoalcaligenes KF707 (Accession
set and is not suitable for the detection of all other no. AAA25743) (Taira et al., 1992). Data by Robert-
members of subfamily D.2.C and those of the other son et al. (1992) indicate that catabolism of toluene in
subfamilies. strain JS150 can also be initiated by a broad-substrate
To test the theoretical assumptions concerning toluene/benzene dioxygenase similar to that of P.
primer selectivity, PCR amplification with the 3 putida F1, whereas Kitayama et al. (1996b) reported
novel primer sets was performed with DNA from a growth of strain JI104 on biphenyl suggesting that it
variety of well-characterized BTEX degrading bacte- contains a bph operon including todC1-like genes.
rial strains with and without the corresponding target The presence and detection of todC1-like genes in
gene as reported in Table 1. All strains were tested JS150 and JI104 agree with those observations and
with each primer set. PCR amplification yielded pro- can be explained by them.

9. 258 B. Hendrickx et al. / Journal of Microbiological Methods 64 (2006) 250–265
3.2. Design of PCR primer sets targeting (alkyl) H. M. Tan, unpublished results), all involved in tolu-
catechol 2,3-dioxygenases (C23O) ene degradation. As no primer sets to detect the
corresponding gene sequences exist, a suitable primer
Proteins involved in BTEX degradation can be set TODE-F/TODE-R was designed.
found in subfamilies I.2.A, I.2.B and I.2.C within PCR amplification with the primer sets targeting
family I.2 and in subfamily I.3.B within family I.3 the different C23O DNA sequences was performed
of the phylogenetic tree of C23O amino acid with DNA from all BTEX degrading bacterial strains
sequences reported by Eltis and Bolin (1996). Sub- reported in Table 1 including strains with and without
family I.2.A contains the C23O sequences of mainly the target gene (Table 1). All strains were tested with
fluorescent Pseudomonas bacteria, whereas subfamily all primer sets. PCR products of expected size were
I.2.B includes the C23O sequences of mainly Sphin- obtained for strains carrying the corresponding target
gomonas bacteria. Primer sets for the detection of gene and the fragments showed the corresponding
C23O genes of subfamilies I.2.A and I.2.B were gene sequences. However, use of the TODE-F/
previously reported by Hallier-Soulier et al. (1996), TODE-R primer set resulted into some aspecific frag-
Okuta et al. (1998), Meyer et al. (1999) and Mesarch ments in addition to the expected fragment. The trans-
et al. (2000). Since then, many new C23O sequences lated xylE1 amplicon sequences from strains MT15
became available. As those reported primers did not and MT53 showed 100% identity with other Pseudo-
always matched with all presently available monas C23O protein sequences including XylE of the
sequences, we adapted and/or designed new degener- pWWO TOL plasmid (Greated et al., 2002; Keil et al.,
ate primer sets, which would allow the detection of 1985b), demonstrating the presence of TOL-like xyl
genes encoding for subfamily I.2.A and subfamily genes in those strains. Unexpectedly, for Burkholdeia
I.2.B. C23O proteıns, i.e., primer set XYLE1-F/
¨ sp. strain JS150, a todE-like gene PCR amplicon was
XYLE1-R and primer set XYLE2-F/XYLE2-R. obtained. The deduced protein showed 90% amino
Subfamily I.2.C comprises mainly C23O acid sequence identity with TodE of P. putida F1
sequences involved in phenol degradation derived (Zylstra and Gibson, 1989) and P. putida DOT-T1
from a wide variety of bacterial genera (Pseudomo- (Mosqueda et al., 1999). Previously, we detected a
nas, Comamonas, Burkholderia and Ralstonia). In todC1-like gene in strain JS150. Those data strongly
addition, it contains two C23O proteıns involved in
¨ suggest the presence of a tod-like pathway in JS150.
BTEX degradation for which no primers for PCR
detection have been reported, i.e., the cdo gene coding 3.3. Sensitivity of detection of the PCR with the
for the C23OII Cdo in P. putida MT15 (Keil et al., different primer sets in soil
1985a) and the tbuE gene coding for the C23O TbuE
in Ralstonia pickettii PKO1 (Kukor and Olsen, 1996). To examine the sensitivity of the PCR to detect the
Since the members of subfamily I.2.C C23O protein target catabolic genes in soil, decreasing concentra-
sequences showed a low degree of similarity between tions of different bacterial strains, carrying relevant
each other, no subfamily I.2.C specific primer set BTEX degradation genes (P. putida mt-2 (xylA,
could be designed. Instead, non-degenerate Cdo and xylE1), S. yanoikuyae B1 (xylE2), P. putida MT15
TbuE C23O gene specific primer sets, named CDO-F/ (xylE1, cdo), Ralstonia pickettii PKO1 (tbuA1, tbuE)
CDO-R and TBUE-F/TBUE-R respectively, were and P. putida F1 (todC1, todE)) were simultaneously
designed. added to both sterilized soils A2Z and GP48. DNA
Family I.3 consist of subfamilies I.3.A and I.3.B, was extracted and subjected to PCR amplification.
and includes C23O sequences from different Pseudo- Except for primer sets TODC1-F/TODC1-R and
monas and Rhodococcus strains, which are mainly TODE-F/TODE-R with detection limits of 105–106
involved in biphenyl degradation. However, subfam- gene copies gÀ 1 soil, all primer sets detected the
ily I.3.B also contains the 3-methylcatechol 2,3-diox- corresponding catabolic gene sequences in both soils
ygenases TodE found in P. putida F1 (Zylstra and with a detection limit of ca. 103–104 gene copies gÀ 1
Gibson, 1989) and P. putida DOT-T1 (Mosqueda et soil, assuming one copy of the gene per cell. The
al., 1999) and TobE of P. putida PB4071 (W. Li and experiment was performed twice with both soils and

10. B. Hendrickx et al. / Journal of Microbiological Methods 64 (2006) 250–265 259
gave identical results (data not shown). No signals nated area both Actinobacteria and Proteobacteria
were obtained with DNA from sterilized uninoculated were found (Hendrickx et al., in press).
soils. The observed detection limits are commonly With a few exceptions, bacterial isolates of the
found for PCR detection of functional genes in envi- same BOX-PCR group utilized the same BTEX sub-
ronmental samples such as soil (Kowalchuk et al., strates. Most of the obtained isolates utilized benzene
1999), compost (Kowalchuk et al., 1999) or seawater as C-source, while only two isolates grew on o-xy-
(Sinigalliano et al., 1995). lene. The ability to utilize TEX compounds as a
carbon source is not always accompanied by the
3.4. Distribution of BTEX degradation genes in BTEX ability to utilize benzene and vice versa (Stapleton
degrading bacteria isolated from a BTEX polluted site and Sayler, 2000). In our study, benzene was the main
contaminant which probably resulted into a major
A total of 81 BTEX degrading isolates was obtained distribution of the ability to utilize benzene in the
from the site. BOX-PCR was used to discriminate bacterial community of that site. In contrast, o-xylene
between potential clonal isolates, recognizing 52 diffe- was not detected in the groundwater at the site and the
rent BOX-PCR groups (Fig. 2). From each BOX-PCR ability to degrade o-xylene was the least detected
group, the 16S rRNA gene of at least one isolate was BTEX degradative phenotype among the isolates.
partially sequenced for identification and all isolates The tmoA-like gene was the most represented
were examined for the presence of the BTEX degra- BTEX initial attack genotype among the isolates,
dation genotypes and growth on BTEX compounds. i.e., 59 and 39 of the isolates and BOX-PCR groups,
For each BTEX degradation gene, PCR products respectively, carried the tmoA-genotype, followed by
obtained from the isolates were randomly selected the xylM gene which was present in 39 of the isolates
for sequencing to examine if the recovered gene frag- and 20 of the BOX-PCR groups. The other initial
ments were indeed the corresponding target genes. attack genotypes tbmD, todC and xylA were present
For all contaminated locations, Actinobacteria and in 28, 8 and 3 isolates and 12, 4 and 2 BOX-PCR
Proteobacteria comprised the two main groups of groups, respectively. For all primer sets targeting gen-
resident culturable BTEX degrading bacteria, while otypes encoding initial attack proteins, no aspecific
from the uncontaminated location only Actinobacteria signals were obtained and all PCR products showed
were isolated. The Actinobacteria included mainly the expected sizes. Moreover, all sequences obtained
Rhodococcus and Arthrobacter strains. Proteobac- from randomly selected PCR products matched in
teria included mainly g-Proteobacteria, and more BlastX similarity analysis with the BTEX initial attack
specifically Pseudomonas. Similarly, Cavalca et al. genes expected to be detected by the respective primer
(2003) isolated from a BTEX-polluted aquifer, treated sets showing the specificity of the primer sets. The
by air-sparging, bacteria belonging to both the classes recovered and sequenced tbmD-like initial attack
of the Proteobacteria (Pseudomonas, Azoarcus and genes showed a very high similarity with mono-oxy-
Bradyrhizobium) and the Actinobacteria (Microbac- genases involved in phenol degradation instead of
terium and Mycobacterium). On the other hand, Sta- BTEX degradation indicating that at least the se-
pleton and Sayler (2000), mostly isolated quenced tbmD-like genes are rather involved in
Proteobacteria and no Actinobacteria. The prolifera- mono-oxygenase attack of a phenol compound than
tion of BTEX degrading Proteobacteria therefore of BTEX itself. The phenol compound might have
seems to be a major characteristic of adaptation to been derived from initial attack of the BTEX by a true
BTEX in BTEX contaminated sites. Recently, at least BTEX mono-oxygenase such as those encoded by
at the site examined in this study, this was confirmed tmoA-like genes. From 3 of the 8 isolates showing
by means of an in situ microcosm study in which todC1 genes, the todC1 PCR products were se-
aquifer material from an uncontaminated location was quenced. The deduced protein sequences were most
placed into the contaminated groundwater plume. In closely related to the BedC1 sequence. bedC1 is the
the microcosms, the aquifer community was clearly todC1-like gene of a benzene degrading isolate show-
developing into a community dominated by fluores- ing the specialized adaptation to benzene degradation
cent Pseudomonads. In contrast, at the non-contami- on the examined site.

11. 260 B. Hendrickx et al. / Journal of Microbiological Methods 64 (2006) 250–265
Isolation procedure
BOX-PCR group
BTEX substrate
16S rRNA gene
BTEX catabolic
Sampling date
Identification
Identification
Soil sample
Pearson correlation (Opt:1.86%) [0.0%-100.0%]
BOX-PCR BOX-PCR
Isolate
genes
100
50
60
70
80
90
1 A2 groundwater March 2001 A2w7 2 ND B tb mD, tmoA
1 A2 groundwater March 2001 A2w8 2 Pseudomonas putida B tb mD, tmoA, todC1
1
1 A2 groundwater March 2001 A2w4 2 ND B tb mD, tmoA
1 A2Y March 2001 A2Y25 2 Stenotrophomonas sp. m-X, p-X tb mD, tmoA
2 2 A1Y March 2001 A1Y11 2 Pseudomonas putida T, EB, m-X,p-X xylE1
3 A1 groundwater March 2001 A1w4' 2 Pseudomonas putida B tb mD, tmoA, xylM, todC1, xylE1
3 2
3 A1 groundwater March 2001 A1w3' ND B tb mD, tmoA, xylM, todC1, xylE1
4 4 A1 mix of all zones May 2 002 F 27 3 Agrobacterium sp. BTEX-mix tm oA
5 5 A1Y March 2001 Amico5 4 Arthrobacter sp. (B) tm oA
6 6 A1Y March 2001 Amico7 4 Arthrobacter sp. (B) tm oA
7 7 A1 mix of all zones May 2 002 E31 3 Georgenia sp. BTEX-mix tm oA
8 8 A1 mix of all zones May 2 002 C2 0 3 Hydrogenophaga sp. BTEX-mix tb mD, tmoA
9 A3X April 2002 A3X2 2 ND B tm oA, xylM, todC1, xylE1
9 9 A3X April 2002 A3X4 2 Pseudomonas jessenii B tm oA, xylM, todC1, xylE1
9 A3Z April 2002 A3Z2 2 ND B tm oA, xylM, todC1,xylE1
10 10 A2Y March 2001 Ami co27 4 Arthrobacter sp. B, BTEX-mix tm oA
11 11 A2Y March 2001 Ami co51 4 Rhodococcus sp. B, BTEX-mix tm oA
12 12 A1Y March 2001 Ami co42 4 Rhodococcus sp. B, BTEX-mix tm oA
13 13 A1X May 2 002 A1X B1-5 1 Rhodococcus sp. Catechol tm oA, xylM, xylE1, todE
2
14 14 A1Y March 2001 A1Y13 Pseudomonas sp. (BTEX-mix) tm oA
15 15 A1 mix of all zones April 2001 A1- 1 3 Arthrobacter sp. BTEX-mix tm oA, xylE1
16 16 A1 mix of all zones April 2001 A3- 104 3 Pseudomonas sp. BTEX-mix tm oA
17 17 A1 mix of all zones April 2001 A1- 13 3 Sphingomonas sp. BTEX-mix tm oA, xylE2
3
18 18 A1 mix of all zones April 2001 A1- 69 Arthrobacter polychromogenes BTEX-mix tm oA
19 19 A1 mix of all zones April 2001 A1- 10 3 ND BTEX-mix tm oA
20 20 A1X May 2 002 A1X dBTEX1-4 1 Pseudomonas veronii Catechol xylM, xyl 1
E
21 21 A1Y May 2 002 A1Y dBTEX1-5 1 Pseudomonas sp. Catechol xylM, xyl 1
E
22 22 A1X May 2 002 A1X B1-4 1 Pseudomonas fluorescens B xylM, xyl 1
E
23 A1Y May 2 002 A1Y B2-4 1 Pseudomonas veronii B tb mD, tmoA, xylM , xylE1
23 A1Y May 2 002 A1Y B1-4 1 ND B tb mD, tmoA, xylM , xylE1
23 23 A1Y May 2 002 A1Y B3-4 1 Pseudomonas veronii B tb mD, tmoA, xylM , xylE1
23 A1Y May 2 002 A1Y C2-5 1 Pseudomonas veronii B tb mD, xylM,xylE1
23 A1Y May 2 002 A1Y C3-5 1 ND B tb mD, xylM,xylE1
24 24 A1X May 2 002 A1X B2-5 1 Pseudomonas veronii B, T, EB xylM, xyl 1
E
25 A2Y April 2001 IA2YCDA 1 Pseudomonas putida B tb mD, xylM
25
26 A3Y May 2 002 A3Y dBTEX1-5 1 Pseudomonas sp. B xylM, xyl 1
E
26 A3Y May 2 002 A3Y Xyl3 -5 1 ND B xylM, xyl 1
E
26 26 A3Y May 2 002 A3Y Xyl2 -4 1 Pseudomonas sp. B xylM, xyl 1
E
26 A3Y May 2 002 A3Y Xyl1 -4 1 ND B, T xylM, xyl 1
E
26 A3Y May 2 002 A3Y dBTEX2-4 1 Pseudomonas sp. B, T, EB xylM, xyl 1
E
27 27 A1X May 2 002 A1X Xyl1 -5 1 Sphingomonas sp. B, T, EB tm oA, xylM, xylE2
28 28 A1X April 2001 IA1XBOX 1 Rhodococcus sp. B tm oA, xylM, xylE1
29 29 A2Y April 2001 IA2YCDO 1 Pseudomonas sp. B xylM, xyl 1
E
30 A1Y April 2001 IA1YICDB 1 Pseudomonas sp. T, m-X, p-X xylM, xyl 1
E
30 30 A1Y April 2001 IA1YICDA 1 ND T, m-X, p-X xylM, xyl 1, cdo
E
31 31 A2X April 2001 IA2XCDB 1 Pseudomonas putida T, m-X, p-X xylA, xylM, xylE1
32 32 A1 mix of all zones April 2001 A1- 8 3 Sphingomonas sp. BTEX-mix tm oA, xylE2
33 A1X April 2002 A1X/2A 2 Pseudomonas veronii B, T, EB tb mD, tmoA, xylE1
33 A1X April 2002 A1X/2B 2 ND B, T, EB tb mD, tmoA, xylE1
33 33 A1X April 2002 A1X/3B 2 ND (B), (m-X), (p-X) tb mD, tmoA, xylE1
33 A1X April 2002 A1X/4A 2 ND (X-mix), (m-X) tb mD, tmoA, xylE1
33 A1X April 2002 A1X/3A 2 Pseudomonas veronii B, T, EB tb mD, tmoA, xylE1
33 A1X April 2002 A1X/4B 2 Pseudomonas veronii B, T tb mD, tmoA, xylE1
34 A1Y May 2 002 A1Y C1-5 1 ND B, T, EB tb mD, xylM,xylE1
34
34 A1Y May 2 002 A1Y Xyl3-5 1 ND B, T, EB tb mD, xylM,xylE1
34 A1Y May 2 002 A1Y dBTEX2-5 1 Pseudomonas veronii B, T, EB tb mD, xylM,xylE1
35 A3Y April 2002 A3Y3 2 Pseudomonas gingerii B, T,m-X,p-X tm oA, xylA,xylM,xylE1
35
35 A3X April 2002 A3X3 2 ND B, T,m-X,p-X tm oA, xylA,xylM,xylE1
36 36 A2Y March 2001 Amico3 4 Arthrobacter sp. (X-mix) tm oA
37 37 EAWAG March 2001 KZ1 5 ND B,T, EB, (p-X),m-X, o-X tm oA, todE
38 38 A1 groundwater March 2001 A1w2 2 Stenotrophomonas maltophilia m-X, p-X tm oA
38 A1 groundwater March 2001 A1w3 2 ND m-X, p-X tm oA
39 39 A4Z March 2001 A4Z24 2 Arthrobacter sp. B, T tm oA
40 A3Z March 2001 A3Z19 2 Rhodococcus sp. T, EB, X-mix tm oA
40 40 A2Y March 2001 A2Y26 2 Rhodococcus sp. T, EB, X-mix tm oA
41 A2X March 2001 A2X9 2 ND T, EB, X-mix tm oA, xylM
41 A2Y March 2001 A2Y7' 2 Rhodococcus sp. T, EB, X-mix tm oA, xylM
41
41 A2Y March 2001 A2Y6 2 Rhodococcus sp. T, EB, X-mix tm oA, xylM
41 A2Y March 2001 A2Y5' 2 ND T, EB, X-mix tm oA, xylM
42 42 A3Y April 2002 A3Y2 2 Rhodococcus sp. B,T, EB, 0-X tm oA
43 43 A1Y March 2001 A1Y15 2 Ralstonia eutropha B, T, EB tb mD, tmoA
44 44 A2Y March 2001 Amico6 4 Variovorax sp. B,T, EB, BTEX-mix tb mD, cdo
45 45 A1 mix of all zones May 2 002 H8 3 Pseudomonas stutzeri ssp. BTEX-mix tm oA, xylM, xylE1
46 46 A1Y April 2002 A1Y/3A 2 Pseudomonas veronii (EB), (m-X) tb mD, xylE1
47 47 A3Z March 2001 A3Z17 2 Arthrobacter sp. B, EB, X-mix tm oA
48 48 A3Z March 2001 A3Z18 2 Arthrobacter sp. B, EB, X-mix tm oA
49 49 A1 mix of all zones May 2 002 B33 3 Aquaspirillum sp. BTEX-mix tm oA
50 50 A1 mix of all zones May 2 002 G1 4 3 Agrobacterium sp. BTEX-mix tm oA
51 A3Z April 2002 A3Z4 2 Pseudomonas marginalis B, T tb mD, tmoA, xylM , todC1 , xylE1
51
51 A3Z April 2002 A3Z1 2 Pseudomonas marginalis B, p-X tb mD, tmoA, xylM , todC1 , xylE1
52 52 A1 mix of all zones May 2 002 A23 3 Acidovorax sp. BTEX-mix tb mD, tmoA
Fig. 2. UPGMA clustering of BOX-PCR fingerprints of the 81 bacterial strains isolated from BTEX contaminated soil samples.

12. B. Hendrickx et al. / Journal of Microbiological Methods 64 (2006) 250–265 261
tmoA-like genes have been shown to be implicat- detected by the respective primer sets showing the
ed in benzene, toluene and o-xylene degradation specificity of the primer sets. As expected, xylE1
(Bertoni et al., 1998; Byrne et al., 1995; Kitayama was often found in combination with xylM. Howev-
et al., 1996a; Yen et al., 1991), todC1 in (ethyl- er, some unexpected combinations of initial attack
)benzene and toluene degradation (Zylstra and Gib- enzymes and C23O enzymes were observed. todE
son, 1989) and xylM/xylA in toluene, m- and p- methylcatechol 2,3-dioxygenase-like genes which
xylene degradation (Burlage et al., 1989; Keil et were only detected in two BOX-PCR groups were
al., 1985a; Kok et al., 1999). In this study, isolates never found in combination with a todC1-like initial
carrying tmoA-like genes and todC1-like genes attack gene. Moreover, cdo-like genes which up to
showed a BTEX degradation range as expected now were only detected in TOL plasmid containing
from previous observations. However, many isolates bacteria in combination with xylM and xylE1 (Keil et
containing xylM-like genes did not grow on either al., 1985a,b), were in this study only combined with
toluene, m- and p-xylene or ethylbenzene (BOX- xylM-, xylE1-like genes in one Pseudomonas strain
PCR groups: 3, 9, 22, 23, 25, 26, 28 and 29), but belonging to BOX-PCR group 30. The cdo C23OII-
instead only utilized benzene. Benzene cannot be like gene was also once combined with a tbmD-like
attacked by a XylM/XylA enzyme system since it initial attack gene in a Variovorax strain (BOX-PCR
does not carry an alkyl side chain. The bacteria be- group: 44), indicating the combination of cdo and
longing to those BOX-PCR groups carry in addition xylM as originally found in P. putida strains MT15
also tmoA- and/or todC1-like genes, which might be and MT53 is not always consistent.
implicated in benzene degradation in those strains. Interestingly, xylM-like genes were also detected in
The recovered xylM/xylA genes might be truncated one Sphingomonas strain (BOX-PCR group: 27) and
versions of the TOL plasmid pathway which would in some Rhodococcus strains (BOX-PCR groups: 13,
nevertheless allow them to initiate degradation of 28, and 41), whereas the tmoA-like gene was also
toluene and xylene for consumption of the metabolites detected in different a-Proteobacteria like Agrobac-
by synergistic communities. Many bacteria carry terium (BOX-PCR groups: 4 and 50) and Sphingo-
different initial attack genotypes. The presence of monas (BOX-PCR groups: 17, 27 and 32) and in
multiple catabolic genotypes with a similar catabolic Gram-positives like Arthrobacter (BOX-PCR groups:
activity in a single bacterial strain is not that re- 5, 6, 10, 15, 18, 36, 39, 47 and 48). This result might
markable and has been shown for Burkholderia sp. be an indication for the occurrence of horizontal
strain JS150 (Kahng et al., 2001) and in different. P. transfer of BTEX catabolic genes in the aquifer com-
putida strains (Cavalca et al., 2000). About the res- munity and moreover, that lateral gene transfer in the
pective role and activity of multiple initial attack aquifer occurred across phylogenetic boundaries. A
systems in BTEX degrading bacteria in the presence similar observation was done by Cavalca et al. (2003)
of single or mixed BTEX compounds is not much who suggested interspecies transfer of the tmoA genes
known. Interestingly, as found for some TOL plasmid in a BTEX contaminated aquifer based on the occur-
carrying reference strains, many of xylM carrying rence of tmoA-analogues in different BTEX degrading
BTEX degraders did not contain a xylA-analogue. isolates (Pseudomonas, Mycobacterium and Bradyr-
The xylE1-like genotype was the most recovered hizobium) from that aquifer.
C23O gene among the isolates, being present in 42
and 19 of the isolates and BOX-PCR groups, respec- 3.5. PCR detection of BTEX degradation genes in
tively, followed in order by the xylE2-, cdo-/todE- contaminated subsurface soil samples from a BTEX
and tbuE-genotype. As for the initial attack geno- polluted site
types, no aspecific signals were obtained with the
primers targeting C23O genes, except for primer pair tbmD-, tmoA-, xylM-, xylE1- and cdo-like genes
TODEF/TOER. All PCR products showed the were recovered from DNA extracts from all conta-
expected sizes and all sequences from randomly minated locations and this for al three zones X, Y,
selected PCR products matched in BlastX similarity Z, though the cdo- and tbmD-like gene signals were
analysis with the C23O genes expected to be weak for samples derived from sampling point A3