TrustArc Webinar - Stay Ahead of US State Data Privacy Law Developments
Junca Pieper2003 Jmm
1. Journal of Microbiological Methods 55 (2003) 697 – 708
www.elsevier.com/locate/jmicmeth
Amplified functional DNA restriction analysis to determine
catechol 2,3-dioxygenase gene diversity in soil bacteria
Howard Junca, Dietmar H. Pieper *
Department of Environmental Microbiology, GBF-German Research Centre for Biotechnology, Mascheroder Weg 1,
D-38124 Brunswick, Germany
Received 31 March 2003; received in revised form 22 July 2003; accepted 22 July 2003
Abstract
To determine phylogenetic diversity of a functional gene from strain collections or environmental DNA amplifications, new
and fast methods are required. Catechol 2,3-dioxygenase (C23O) subfamily I.2.A genes, known to be of crucial importance for
aromatic degradation, were used as a model to adapt the amplified ribosomal DNA restriction analysis to functional genes.
Sequence data of C23O genes from 13 reference strains, representing the main branches of the C23O family I.2.A phylogeny, were
used for simulation of theoretical restriction patterns. Among other restriction enzymes, Sau3A1 theoretically produce
characteristic profiles from each subfamily I.2.A member and their similarities reassembled the main divergent branches of C23O
gene phylogeny. This enzyme was used to perform an amplified functional DNA restriction analysis (AFDRA) on C23O genes of
reference strains and 19 isolates. Cluster analyses of the restriction fragment profiles obtained from isolates showed patterns with
distinct similarities to the reference strain profiles, allowing to distinguish four different groups. Sequences of PCR fragments from
isolates were in close agreement with the phylogenetic correlations predicted with the AFDRA approach. AFDRA thus provided a
quick assessment of C23O diversity in a strain collection and insights of its gene phylogeny affiliation among known family
members. It cannot only be easily applied to a vast number of isolates but also to define the predominant polymorphism of a
functional gene present in environmental DNA extracts. This approach may be useful to differentiate functional genes also for
many other gene families.
D 2003 Elsevier B.V. All rights reserved.
Keywords: Amplified functional DNA restriction analysis; AFDRA; Catechol 2,3-dioxygenase catabolic gene diversity; Soil bacteria; BTEX
biodegradation
1. Introduction mental microbial community composition, to detect
new community members, activities, and functions,
It is well documented that only a small fraction of have experienced a fast development in the last deca-
environmental microorganisms can be cultured to date, des. Culture-independent approaches, usually focused
and thus, culture-independent methods (Amann et al., on 16S rRNA phylogeny, have been applied for ana-
1995; Staley and Konopka, 1985) to describe environ- lysing populations in diverse environments (Torsvik
and Ovreas, 2002), and various molecular fingerprint-
* Corresponding author. Tel.: +49-531-6181-467; fax: +49-531-
ing methods like denaturing gradient gel electrophore-
6181-411. sis (DGGE), temperature-gradient gel electrophoresis
E-mail address: dpi@gbf.de (D.H. Pieper). (TGGE), ribosomal intergenic spacer analysis (RISA),
0167-7012/$ - see front matter D 2003 Elsevier B.V. All rights reserved.
doi:10.1016/S0167-7012(03)00214-8
2. 698 H. Junca, D.H. Pieper / Journal of Microbiological Methods 55 (2003) 697–708
terminal restriction fragment length polymorphism aromatic ring fission activity, which play an essential
(T-RLFP), upstream-independent ribosomal RNA am- role in the degradation of a wide range of aromatic
plification analysis (URA), and polymerase chain pollutants. A large collection of genes coding for such
reaction – single-strand conformation polymorphism an activity have been cloned and sequenced in the last
(PCR-SSCP) have been optimised to obtain a fast years, being classified as a diverse gene family (Eltis
and reliable overview on microbial community com- and Bolin, 1996). The gene phylogeny of these very
positions and their shifts by changing environmental closely related sequences does not follow strictly a
conditions (Kent and Triplett, 2002). Other fingerprint- taxonomical relation with the bacterial hosts, since
ing methods like amplified ribosomal DNA restriction these genes are mainly found on plasmids, and their
analysis (ARDRA) (Khetmalas et al., 2002) or PCR – evolution and conservation rates are heavily affected
multiple enzyme restriction fragment length polymor- by traits like selection pressures, horizontal transfer,
phism (PCR-MERFLP) (Porteous et al., 2002) have and mobile genetic elements (Williams et al., 2002).
been developed to distinguish single 16S rDNA PCR Conserved regions of the C23O genes have been
fragments from isolates or gene libraries to get a rapid selected in several studies as suitable PCR targets in
overview on taxonomical diversity in cultivable or culture-independent approaches, showing a correlation
noncultivable microbial fractions. in relative C23O abundances dependent of pollutants
However, a number of bacterial genes such as those levels (Erb and Wagner-Dobler, 1993; Mesarch et al.,
involved in antibiotic resistance, antimicrobial produc- 2000; Meyer et al., 1999; Ringelberg et al., 2001;
tion or pollutant degradation are selected or present Wikstrom et al., 1996).
independently of the rate of evolution and taxonomy of In this study, we developed and evaluated an
the bacterial host (Davison, 1999). Thus, assessment amplified DNA restriction analysis targeting a func-
of community functions needs reliable tools to analyse tional gene (amplified functional DNA restriction
those functions rather than taxonomical composition. analysis, AFDRA) in a collection of strains sharing
PCR-based techniques have been used to detect func- a phenotypic character, meta-cleavage activity. We
tional/catabolic genes in environmental isolates or examined the suitability of this method to directly
environmental DNA, and diversity is usually assessed analyse soil DNA by a combination of PCR serial
by sequencing of genes from isolates or PCR clone dilution assays and AFDRA analyses. This ap-
libraries (Buchan et al., 2001; Duarte et al., 2001; proach allowed us to determine C23O gene copy
Hamelin et al., 2002; Yeates et al., 2000). Recent numbers and predominant C23O gene variants in
reports showed that PCR-amplified fragments of cat- soil samples.
abolic genes from environmental DNA can be sepa-
rated by DGGE (Henckel et al., 1999; Nicolaisen and
Ramsing, 2002), and RFLP analyses were used to 2. Materials and methods
select distinctive restriction patterns of single ampli-
cons for further sequence determinations (Bakermans 2.1. Microorganisms, samples, isolation, and culture
and Madsen, 2002; Braker et al., 2000; Yan et al., conditions
2003). A more detailed knowledge on catabolic genes,
retrieved by culture-independent methods and from Reference strains P. stutzeri AN10 (Bosch et al.,
isolates, can significantly improve our understanding 2000), P. putida CF600 (Bartilson and Shingler, 1989),
of microbial functioning and degradation processes in P. putida G7 (Ghosal et al., 1987), P. putida H
the environment, which would help to design new (Herrmann et al., 1995), Pseudomonas sp. IC (Car-
bioremediation strategies (Widada et al., 2002). How- rington et al., 1994), P. aeruginosa JI104 (Kitayama et
ever, no fast and reliable tools are available to rapidly al., 1996), Sphingomonas sp. KF711 (Moon et al.,
analyse gene phylogeny in respective culture collec- 1996), P. putida mt-2 (Nakai et al., 1983), P. putida
tions or to determine abundant gene variants and their mt53 (Keil et al., 1985), P. stutzeri OM1 (Ouchiyama
phylogeny in the environment. et al., 1998), P. stutzeri OX1 (Arenghi et al., 2001), P.
Catechol 2,3-dioxygenases (C23O) comprise a fam- putida HS1 (Benjamin et al., 1991), P. putida 3,5X
ily of genes coding for a group of enzymes with (Hopper and Kemp, 1980), and isolated bacterial
3. H. Junca, D.H. Pieper / Journal of Microbiological Methods 55 (2003) 697–708 699
strains were cultured on R2A agar (DIFCO) at 30 jC 0.25 AM of each primer (synthesized by Invitrogen)
for 3 days. and 0.3 U/Al Taq DNA polymerase (Promega).
Bacterial strains were isolated by direct plating on For amplification of soil DNA, PCR was per-
R2A agar of appropriate soil dilutions from samples formed in a PCRExpress gradient thermocycler
obtained from four sampling points of a BTEX- (Hybaid) as follows: an initial step at 95 jC for 3
contaminated area in Czech Republic. After 3 days min, followed by 35 cycles of denaturation at 94 jC
of incubation at 30 jC, colonies were tested by for 45 s, annealing at 55 jC for 45 s, and elongation at
spraying with a 100 mM aqueous catechol solution, 72 jC for 90 s. For amplification of DNA from
and those exhibiting meta-cleavage activity, indicated isolates, the PCR program comprised an initial step
by yellow coloration of the medium, were selected, at 95 jC for 3 min, 10 cycles of denaturation at 94 jC
streaked, and purified. for 45 s, annealing for 45 s at a starting temperature of
For culture-independent analyses, soil samples 60jC with a decrease in annealing temperature of 1
were collected from the capillary fringe zone (zone jC per cycle, elongation for 90 s at 72 jC, followed
of essentially water-saturated soil just above the water by 25 PCR cycles at a constant annealing temperature
table) and the saturated zone (zone below the water of 55 jC, final elongation step at 72 jC for 8 min, and
table) of a highly contaminated area (BTEX concen- further storage of the reactions at À 20 jC.
trations in the groundwater of 320, 97, 6, and 13 mg/ To determine the minimum number of C23O
l of benzene, toluene, ethylbenzene, and xylenes, copies detected by PCR with this set of primers,
respectively, as determined by Aquatest Prague). Total an equimolar mixture of 15 C23O PCR amplicons
DNA was extracted in triplicates from 500 mg of each (from 13 reference strains and from isolate numbers
soil sample and purified with the Fast Prep Soil DNA 9 and 19), was serially diluted and aliquots of these
Extraction Kit (BIO101). Concentration of DNA was dilutions applied in triplicates to PCR reactions. The
determined by using PicoGreen dsDNA Quantitation maximum dilution at which a PCR signal of the
Kit (Molecular Probes) and a microtiter plate reader as expected size was detected in one of the triplicates
described previously (Weinbauer and Hofle, 2001).
¨ was determined analysing 3 Al of the PCR reactions
by gel electrophoresis (1.5% agarose, 10 cm length,
2.2. Primer design and PCR conditions 1 Â TAE running buffer, 1 h at 95 V and visualized
by ethidium bromide staining) (Sambrook et al.,
The primers C23O-ORF-F 5VAGG TGW CGT 1989). As the average molecular mass of the 934-
SAT GAA MAA AGG 3Vand C23O-ORF-R 5VTYA bp C23O fragments is 570 kDa, it can be calculated
GGT SAK MAC GGT CAK GAA 3Vwere designed that a single PCR amplicon molecule corresponds to
to amplify 934 bp comprising the complete open approximately 9.5 Â 10À 10 ng of DNA. The mini-
reading frames of the subfamily I.2.A C23O genes. mum number of C23O copies necessary to produce a
Degenerations are placed according to all the possible detectable PCR amplification with the primers
variable positions in the alignment of 20 C23O assayed was used to extrapolate the number of
reference sequences (EMBL/GenBank/DBBJ acces- copies per milligram of soil (Ringelberg et al.,
sion numbers M33263, AY112717, S77084, 2001) by determination of the maximum dilution
X80765, D83057, X77856, M65205, X60740, of soil DNA at which a PCR signal is detected in
J C 4 8 8 5 , A F 0 3 9 5 3 4 , A B 0 0 1 7 2 2 , V 0 11 6 1 , six independent experiments. The inhibitory effect of
AF226279, AF102891, AJ496739, D83042, JC5654, soil DNA extracts on C23O PCR amplification
U01825, AY228547, X06412). As template DNA, 4- efficiency was determined by spiking soil DNA
Al aliquots of a total of 50-Al supernatant from (5 Al containing a total of 10 ng of DNA) extracted
colonies boiled for 10 min (Kanakaraj et al., 1998) from a control sample of the saturated zone of the
or 5 Al of serial dilutions of soil DNA (prepared from study area with no previous BTEX contamination
50 Al of a total DNA extraction of 500 mg of soil) and devoid of amplifiable C23O genes, with 1 ng of
were applied in a PCR mixture containing 1 Â PCR an equimolar mixture of 15 C23O PCR amplicons
Buffer (Promega) supplemented with 1.5 mM MgCl2, (equivalent to 107 C23O gene copies). By serial
200 AM of each deoxyribonucleotidetriphosphate, dilutions in triplicates, the maximum dilution at
4. 700 H. Junca, D.H. Pieper / Journal of Microbiological Methods 55 (2003) 697–708
which a C23O PCR signal of the expected size is 2.5. DNA sequencing and phylogenetic analyses
detected was determined.
Nucleotide sequencing of PCR fragments or plas-
2.3. In silico PCR-RFLP analyses mids with cloned inserts was carried out on both
strands using Taq dye –deoxy terminator in an ABI
Predictions and simulations of the restriction frag- 373A automatic DNA sequencer (Perkin-Elmer Ap-
ment length of 13 C23O DNA sequences from refer- plied Biosystems) following the protocols provided by
ence strains was performed by a restriction mapping the manufacturer. Primers used for sequence reactions
software (Heiman, 1997) with around 300 different were the same as for PCR. Alignments were per-
endonuclease recognition sites. Enzymes that poten- formed with CLUSTAL X 1.8 windows interface of
tially produce phylogenetic informative fragments the CLUSTALW program using default values
were selected after detailed inspection and comparison (Thompson et al., 1997). DNA alignments were edited
of the predicted restriction positions and calculation of and translated with the GeneDoc program (Nicolas,
the produced fragment lengths. Matrices to represent 1997). Phylogenetic trees were obtained with the
discrete values for presence or absence of a predict- option available on the CLUSTAL program through
able restriction fragment size in the sequence data set the Neighbour Joining (N-J) algorithm method. Dis-
were generated for each selected enzyme. Distance tances were generated using Kimura Matrix, and tree
estimations on these matrices were performed with stability was supported through Bootstrap analysis. NJ
Treecon software (Van de Peer and De Wachter, 1993) trees were visualized with the NJplot program (Per-
by Simple Matching or by Nei and Liu methods. Tree riere and Gouy, 1996). Values of more than 50% of
topology was inferred by the UPGMA clustering 1000 replications (seed value 111) are shown on
method. Those results were compared with C23O appropriate branches. A similarity matrix to compare
protein phylogeny. PCR-RFLP gel patterns was calculated using Bio1D
v.99.02 Software (Vilber Lourmat), with Jaccard coef-
2.4. Amplified functional DNA restriction analysis ficients with 3% confidence applied to both compared
bands. A dendrogram of similarity values in the
Selected restriction enzymes (NEB) were used in matrix was calculated using the UPGMA algorithm.
reactions in a final volume of 20 Al containing the 1 Â The sequences of gene fragments reported in this
buffer recommended by the manufacturer, 3 U of the study are available under the EMBL/ GenBank/ DBBJ
enzyme, and approximately 200 ng of PCR product, accession numbers AJ544921 to AJ544938.
incubated at optimal temperature for 4 h.
The restriction fragments patterns were resolved by
gel electrophoresis in a 5.5% Nusieve 3:1 (FMC 3. Results and discussion
Bioproducts) agarose matrix (14 Â 11 cm) in 1 Â
TBE buffer (Sambrook et al., 1989), at 120 V (80 3.1. Targeting the functional gene by PCR
mA), until the bromophenol blue dye in the loading
buffer was reaching the front edge (approximately From soil samples contaminated with BTEX, a
5 h). The loading buffer comprised glycerol (30%, v/ broad set of microorganisms exhibiting catechol 2,3-
v) in water, xylene cyanol, and bromophenol blue at dioxygenase activity, identified based on a simple test
final concentrations of 0.025%. To correctly detect all for induction of this enzyme activity during growth on
bands and intensities onto the gel, the loading buffer agar plates, was isolated. All isolates referred here
contained dyes at 1/10 strength of the standard con- could use benzene and/or toluene as sole source of
centrations (Sambrook et al., 1989), avoiding their carbon and energy.
collateral effect of dark background. For staining, the We hypothesized that these isolates could carry
gel was placed in ethidium bromide solution (10 mg/l) C23O genes members of the I.2.A subfamily (Eltis
for 30 min. The gel was visualised, images acquired and Bolin, 1996), because this group of enzymes is
and stored on a gel documentation system (Vilber known to be involved in the degradation of aromatic
Lourmat). compounds in several environmental strains. A prim-
5. H. Junca, D.H. Pieper / Journal of Microbiological Methods 55 (2003) 697–708 701
er set was designed to amplify the complete C23O acid sequencing of the genes. The theoretical simula-
gene of members in the I.2.A subfamily, annealing at tion was performed through the generation of discrete
the start and stop codon positions. For the forward matrices of restriction fragment sizes produced in
primer, annealing was optimised by placing its 5V C23O sequences deposited on EMBL/GenBank/
side 10 bases upstream of the C23O starting codon. DDBJ databases. These matrices were further analysed
This primer set was tested on 13 reference strains to find restriction enzymes producing specific frag-
(see Materials and methods) known to carry C23O ment profiles for different genes, and a restriction
subfamily I.2.A gene variants and on a soil bacterial pattern related to gene phylogeny. In this search, out
strain collection. From all the reference strains and of approximately 300 enzymes tested on each of the 13
from 30 out of 37 bacterial isolates tested, a signal of selected sequences, some enzymes, among them AluI,
the expected size was successfully amplified, where- BsiHKAI, BstUI, HhaI, HinP1I, MspR9I, PalI,
as 7 isolates did not give any amplification product, Sau3AI, and TaqI, were found as theoretically produc-
indicating that they probably harbour a C23O of ing characteristic profiles from each subfamily I.2.A
another subfamily. Nineteen of those isolates, show- member.
ing an amplification product of 934-bp in size, were
randomly selected for further analysis. 3.3. Characterization and comparison of C23O
diversity in bacterial isolates by AFDRA and DNA
3.2. Selection of restriction enzymes producing sequencing
patterns clustering as the divergent branches of
C23O protein phylogeny by using predictions and After a meticulous comparison of the predictions,
simulations on reported DNA sequences we selected the restriction enzyme Sau3A1 for the
experiments. This enzyme could cleave the C23O
To assess the diversity of the C23O genes, a fast genes in such a way that the resulting clustering of
screening method to identify potential redundant or the matrix (37 Â 14), representing discrete values for
highly similar genes based on a restriction fragment presence or absence of a predictable restriction frag-
simulation, was developed. This method was com- ment size (37 theoretical restriction fragments from 11
pared with results obtained by a conventional nucleic reference strains; Fig. 1), reassembled the main diver-
Fig. 1. Cluster dendrogram of theoretical Sau3AI restriction fragments produced from C23O genes of P. stutzeri AN10, P. putida CF600,
P. putida G7, P. putida H, Pseudomonas sp. IC, P. aeruginosa JI104, P. putida mt-2, P. putida mt53, P. stutzeri OM1, P. putida HS1, P. putida
3,5X. The corresponding discrete matrix of presence (1) or absence (0) of a specific restriction fragment size is given to the right. Columns 1 –
37 represent, respectively, the following sizes in base pairs: 10, 14, 18, 19, 21, 23, 30, 34, 35, 39, 44, 55, 58, 88, 92, 104, 111, 112, 126, 133,
138, 141, 152, 153, 156, 157, 163, 174, 177, 232, 239, 269, 285, 306, 321, 443, and 447.
6. 702 H. Junca, D.H. Pieper / Journal of Microbiological Methods 55 (2003) 697–708
gent branches of the C23O phylogeny (Eltis and Bolin, tion patterns present in the isolates. A selection of
1996). Furthermore, the predicted fragment size range those representative patterns found with distinct sim-
produced by this enzyme (10 – 447 bp) could be ilarities to the reference strain profiles are shown in
separated by high-resolution agarose gel electrophore- Fig. 2. The most dominant pattern, observed in 13 of
sis. Sau3A1 restriction assays were performed using 19 isolates of C23O PCR-RFLP, was equal to that
PCR C23O gene amplifications from seven reference obtained from P. stutzeri AN10. Another common
strains (Fig. 2) selected to cover the main divergent pattern, which shared some restriction fragment sizes
branches of the protein gene phylogeny. Comparisons with P. putida G7, was present in four isolates. The
of the restriction patterns under optimised electropho- PCR fragment from a single isolate produced a
retic conditions resolved the informative restriction restriction pattern closely related to that from P. putida
fragment patterns of reference strain amplifications mt-2. Another PCR fragment from a single isolate
in the >35- to < 600-bp range, effectively discriminat- showed a restriction pattern sharing, to a lesser extent,
ing down to 5-bp size differences, as evidenced by the band sizes with the P. putida 3,5X C23O gene. It can
comparison of 133-bp and 138-bp fragments. The thus be hypothesized that the isolates are dominated
same procedure was performed on 19 PCR-amplified by catechol 2,3-dioxygenase genes closely related to
products of bacterial isolates, which were producing the P. stutzeri AN10 gene, whereas other isolates
distinctive restriction fragment patterns. Digestions harbour far divergent evolutionary variants of catechol
were usually complete under the conditions used, 2,3-dioxygenases.
and only in case of high concentrations of amplifica- To confirm the reliability of this assay, the frag-
tion product subjected to digestions, faint bands due to ments from all the isolates mentioned above were
incomplete digests were observed. completely sequenced, and the resulting sequences
Cluster analyses of the restriction fragment profiles were aligned against C23O DNA sequences of refer-
obtained allowed to distinguish four groups of restric- ence strains. In the phylogenetic tree of the deduced
Fig. 2. Agarose gel-generated cluster dendrogram illustrating the relationship of C23O genes based on similarity of restriction fragment patterns
produced by Sau3AI digestions. Digestions of the 934-bp C23O PCR fragments amplified from the reference strains P. stutzeri AN10, P. putida
CF600, P. putida G7, Pseudomonas sp. IC, P. putida mt-2, P. putida HS1, P. putida 3,5X, and isolated bacterial strains of group A: isolates
number 2, 9, 4, 12, and 13; group C: isolate number 14; group B: isolate number 18; and group D: isolate number 1. Marker lane (MWM) shows
DNA marker V (Roche), in the upper side are written the corresponding base pair length values of each DNA fragment.
7. H. Junca, D.H. Pieper / Journal of Microbiological Methods 55 (2003) 697–708 703
C23O protein sequences from reference strains and to P. putida mt-2 C23O protein numbering) to tyro-
isolates (Fig. 3), the sequences obtained from isolates sine (11 isolates, subgroup A1) or histidine (2 isolates,
are clustering in five different branches. This closely subgroup A2) (Fig. 3). This difference does not affect
corresponds to the results obtained by the restriction a Sau3AI recognition site and therefore could not be
analysis. In addition, all the sequences that were discriminated by PCR-RFLP. In such cases where
supposed to carry the same or a closely related gene small differences like point mutations should be
polymorphism are, in fact, at sequence level sharing detected, a screening technique not retrieving gene
this feature (Figs. 2 and 3). In the case of group A, phylogeny information but with higher sensitivity for
comprising 13 isolates, a single point mutation dif- single nucleotide substitutions like SSCP (Orita et al.,
ference is present, producing a difference in the 1989) or DGGE (Muyzer et al., 1993) could be then
coding of the amino acid at position 218 (referred applied.
Fig. 3. Phylogenetic tree of deduced amino acid sequences of C23O gene fragments from isolates (shaded) and from reference strains (indicated
by their EMBL/GenBank/DDBJ accession number, abbreviated organism name, and strain designation). Alignment of a 268-amino-acid length
block was performed with CLUSTAL W using default values. N-J phylogenetic tree was generated using the option available in CLUSTAL W
program. Bootstrap values above 50% from 1000 neighbor-joining trees are indicated to the left of the nodes. Bar represents 2 amino acid
changes per 100 amino acids.
8. 704 H. Junca, D.H. Pieper / Journal of Microbiological Methods 55 (2003) 697–708
3.4. AFDRA of new C23O sequences follow the struction of gene phylogeny. AFDRA proved thus to
expected relation with gene phylogeny be a valuable tool to screen PCR fragments of func-
tional genes from isolates, avoiding the redundant work
The PCR-RFLP discrete matrix (40 Â 18) of C23O of repeated sequencing of the same gene from different
members, including the four new sequences observed isolates or colonies, saving time and resources, and
in isolates and comprising three new PCR-RF sizes giving a fast and reliable overview of the predominance
only present in these isolates is shown in Fig. 4. The and phylogenetic affiliation of the genes in a group of
thereby generated dendrogram reassemble the main environmental bacterial isolates. It can be particularly
divergent clustering topology of the C23O protein useful in studies determining functional gene diversity
phylogeny (Fig. 3). The application of PCR-RFLP in in collections of numerous environmental isolates, as
a closely related group of sequences diminishes the well as to screen PCR fragments of functional genes
potential limitations of the RFLP technique (Hillis et amplified from environmental DNA.
al., 1996), as neither larger insertion/deletions or rear-
rangement events nor the convergent fragments were 3.5. Quantitative and qualitative applications of the
observed in the sequences analysed. However, the AFDRA on soil DNA
cluster analysis of the bands resolved on the gel showed
an outgrouping restriction pattern of the P. putida The efficiency and the minimum amount of C23O
CF600 C23O gene, and its 141-bp fragment was copies detected with our primers designed to target the
incorrectly assigned as clustering with fragments C23O subfamily I.2.A were determined, as a prereq-
corresponding to 138 bp. That is an expected limitation uisite for a quantitative and qualitative application of
of the technique due to the experimental error in AFDRA to DNA extracted from soil. PCR products
calculation of very similar fragment sizes. Neverthe- were observed in all triplicates when the PCR mixture
less, the generation and detection of a high number of contained 10À 8 ng of DNA (corresponding to approx-
discrete characters allowed an approximated recon- imately 10 template molecules) and usually in one of
Fig. 4. Cluster dendrogram and corresponding discrete matrix of presence – absence of a restriction fragment size of theoretical restrictions of
C23O genes with Sau3AI from reference strains and isolates. Reference strains are P. stutzeri AN10, P. putida CF600, P. putida G7, P. putida H,
Pseudomonas sp. IC, P. aeruginosa JI104, P. putida mt-2, P. putida mt53, P. stutzeri OM1, P. putida HS1, P. putida 3,5X. Group A includes
sequences from isolates number 2, 9, 4, 12, and 13, group C includes isolate number 14, group B includes isolate number 18, and group D
includes isolate number 1. The resultant 40 columns represent the following sizes in base pairs (1 – 40, respectively): 10, 14, 18, 19, 21, 23, 27,
30, 34, 35, 39, 44, 55, 58, 88, 92, 104, 111, 112, 126, 133, 138, 141, 147, 152, 153, 156, 157, 163, 174, 177, 232, 239, 269, 285, 306, 321, 353,
443, and 447.
9. H. Junca, D.H. Pieper / Journal of Microbiological Methods 55 (2003) 697–708 705
the triplicates when it contained 10À 9 ng of DNA can be assumed that the soil extract decreased the
(corresponding to approximately one template mole- C23O amplification efficiency by a factor of 100. This,
cule), independent of the complexity of the applied in turn, indicates that the contaminated soils analysed
experimental C23O gene composition (Fig. 5A). Soil would contain approximately 106 C23O copies/g of
DNA extractions yielded 100 – 300 ng of DNA/g of soil. To determine if the amplification affects the
soil for the samples analysed. C23O PCR products diversity recovered, AFDRA was applied on the PCR
were observed from as low as 20 pg of soil DNA as products obtained from artificial C23O gene mixtures
template (Fig. 5B), indicating that 1 g of soil of the containing 10– 104 gene copies. The primers did not
sampling point analysed would contain at least 104 show evidence for preferential amplification of any
C23O gene copies, independent if the soil was collect- specific polymorphism under the conditions tested
ed from the capillary fringe or the saturated zone. (Fig. 5C; lanes 17 and 18). The pattern obtained from
However, a control soil DNA extract devoid of C23O an equimolar mixture of 15 C23O gene variants was
genes and spiked with 107 C23O gene copies showed very complex and did not reassemble clearly any of the
an amplification signal only, when the serial dilution single C23O gene profiles, even when the applied
contained at least 100 copies of C23O genes. Thus, it template mixture contains only approximately 10
C23O copies. On the other hand, the primers amplified
preferentially the most abundant polymorphism out of
Fig. 5. Detection limit, diversity comprised, and performance in
artificial mixtures or environmental DNA of the primers targeting
C23O subfamily I.2.A and assessment of predominant gene variants
in soil DNA by AFDRA approach. (A) Detection limit of primers
designed to target C23O subfamily I.2.A. Lanes 1 – 5 show PCR
products obtained from 10À 6, 10À 8, 10À 10, 10À 11, and 10À 12 ng,
respectively, of an equimolar mixture of 15 C23O genes. Lanes 6 – 10
show PCR products obtained from 10À 5, 10À 7, 10À 9, 10À 10, and
10À 11 ng, respectively, of an equimolar mixture of 14 C23O gene
variants supplemented with a 10-fold excess of the P. putida CF600.
Lanes M were loaded with 200 ng of DNA Molecular Weight Marker
VIII (Roche). (B) Determination of the C23O copy number in
contaminated soils. Total DNA extracted from 500 mg of soil (100
ng) was resuspended in 50 Al of H2O with a DNA concentration of 2
ng/Al and appropriate dilutions subjected to PCR amplification. PCR
products obtained from DNA extracted from a soil sample of the
capillary fringe zone (lanes 11 – 13) or from the saturated zone (lanes
14 – 16) were analysed by agarose gel electrophoresis. PCR products
correspond to those obtained from 200, 20 or 2 pg of soil DNA,
respectively. Lanes 1kb were loaded with 100 ng of GeneRuler 1 kb
DNA ladder (MBI Fermentas). (C) AFDRA patterns obtained from
PCR-amplified artificial C23O mixtures or direct amplifications of
soil DNA. Lanes 17 and 18 show AFDRA of the PCR amplification
products obtained from 10À 6 or 10À 8 ng of an equimolar mixture of
15 C23O subfamily I.2.A gene variants. Lanes 19 and 20 show
AFDRA of the PCR amplification product obtained from 10À 5 or
10À 7 ng of an equimolar mixture of 14 C23O gene variants
supplemented with a 10-fold excess of the P. putida CF600 gene.
Lanes 21 and 22 show AFDRA of the PCR amplification product
obtained from 200 or 20 pg of soil DNA of the capillary fringe zone,
lanes 23 and 24 AFDRA of the PCR amplification product obtained
from 200 or 20 pg of soil DNA of the saturated zone, lanes 25 – 28
AFDRA of the C23O genes from P. stutzeri AN10, P. putida CF600,
P. putida G7, and P. putida mt-2, respectively. Lanes M were loaded
with 120 ng of DNA Molecular Weight Marker V (Roche).
10. 706 H. Junca, D.H. Pieper / Journal of Microbiological Methods 55 (2003) 697–708
a complex template mixture containing a 10-fold Gallegos, Prof. Dr. M. Tsuda, Dr. A. Kitayama, Dr. R.
excess of the P. putida CF600 C23O gene, (Fig. 5C; Bosch, Prof. Y. Kim, Prof. V. Shingler, Dr. D. Young,
lanes 19, 20, and 26). These results show the potential and Prof. D. Kunz for kindly providing the reference
use of AFDRA to rapidly determine the predominant strains used in this work. We also thank Dr. Andreas
gene polymorphism in complex environmental DNA Felske for critical reading of the manuscript and
mixtures. When AFDRA was applied to PCR products Robert Witzig for soil DNA quantification. This work
obtained from 200 pg of soil DNA, characteristic was supported by grant QLK3-CT-2000-00731 from
patterns were observed. When PCR-amplified DNA the European Community.
extracted from the capillary fringe zone was analysed
(lane 21), a characteristic pattern identical to that
obtained from P. putida AN10 (lane 25) or group A References
isolates was observed. This result perfectly matches
with the information obtained by screening of a PCR Amann, R.I., Ludwig, W., Schleifer, K.H., 1995. Phylogenetic iden-
clone library constructed with a 527-bp C23O frag- tification and in situ detection of individual microbial cells with-
out cultivation. Microbiol. Rev. 59, 143 – 169.
ment amplified from this soil sample (Junca and
Arenghi, F.L., Berlanda, D., Galli, E., Sello, G., Barbieri, P., 2001.
Pieper, unpublished data). The C23O gene composi- Organization and regulation of meta cleavage pathway genes for
tion in the saturated zone differed qualitatively from toluene and o-xylene derivative degradation in Pseudomonas
that of the capillary fringe horizon (Fig. 5C; lane 23), stutzeri OX1. Appl. Environ. Microbiol. 67, 3304 – 3308.
evidencing the coexistence of two most abundant Bakermans, C., Madsen, E.L., 2002. Diversity of 16S rDNA and
C23O polymorphisms: one closely related to group naphthalene dioxygenase genes from coal – tar – waste – contami-
nated aquifer waters. Microb. Ecol. 44, 95 – 106.
C isolates and another equal to that obtained from P. Bartilson, M., Shingler, V., 1989. Nucleotide sequence and expres-
putida AN10 –group A isolates. sion of the catechol 2,3-dioxygenase-encoding gene of phenol-
catabolizing Pseudomonas CF600. Gene 85, 233 – 238.
3.6. Concluding remarks Benjamin, R.C., Voss, J.A., Kunz, D.A., 1991. Nucleotide sequence
of xylE from the TOL pDK1 plasmid and structural comparison
with isofunctional catechol-2,3-dioxygenase genes from TOL,
In this report, we demonstrated that the ARDRA pWW0 and NAH7. J. Bacteriol. 173, 2724 – 2728.
approach can be transferred to assess functional di- Bosch, R., Garcia-Valdes, E., Moore, E.R., 2000. Complete nucleo-
versity of environmental isolates in a very convenient tide sequence and evolutionary significance of a chromosomally
first selection providing preliminary phylogenetic in- encoded naphthalene-degradation lower pathway from Pseudo-
monas stutzeri AN10. Gene 245, 65 – 74.
formation, for further focused and dedicated biochem-
Braker, G., Zhou, J., Wu, L., Devol, A.H., Tiedje, J.M., 2000.
ical characterization, point mutation determinations or Nitrite reductase genes (nirK and nirS) as functional markers
sequencing work, instead of random or massive to investigate diversity of denitrifying bacteria in Pacific north-
screenings. The method proved to be also appropriate, west marine sediment communities. Appl. Environ. Microbiol.
coupled to a quantitative survey, to assess predomi- 66, 2096 – 2104.
nance of gene variant(s) in DNA extracted directly Buchan, A., Neidle, E.L., Moran, M.A., 2001. Diversity of the ring-
cleaving dioxygenase gene pcaH in a salt marsh bacterial com-
from environmental samples. Future works will de- munity. Appl. Environ. Microbiol. 67, 5801 – 5809.
termine if the technique can be optimised to refine its Carrington, B., Lowe, A., Shaw, L.E., Williams, P.A., 1994. The
phylogenetic resolution power, by additional testing lower pathway operon for benzoate catabolism in biphenyl-uti-
of multiple restrictions of a single PCR fragment lizing Pseudomonas sp. strain IC and the nucleotide sequence of
(Porteous et al., 2002) or by integrated comparisons the bphE gene for catechol 2,3-dioxygenase. Microbiology 140,
499 – 508.
of profiles obtained in independent restrictions with Davison, J., 1999. Genetic exchange between bacteria in the envi-
different enzymes (Sikorski et al., 2002). ronment. Plasmid 42, 73 – 91.
Duarte, G.F., Rosado, A.S., Seldin, L., de Araujo, W., van Elsas,
J.D., 2001. Analysis of bacterial community structure in sulfur-
Acknowledgements ous-oil-containing soils and detection of species carrying di-
benzothiophene desulfurization (dsz) genes. Appl. Environ.
Microbiol. 67, 1052 – 1062.
We thank Prof. D.J. Hopper, Prof. H. Herrmann, Eltis, L.D., Bolin, J.T., 1996. Evolutionary relationships among
Prof. Dr. T. Omori, Prof. P.A Williams, Dr. M.T. extradiol dioxygenases. J. Bacteriol. 178, 5930 – 5937.
11. H. Junca, D.H. Pieper / Journal of Microbiological Methods 55 (2003) 697–708 707
Erb, R.W., Wagner-Dobler, I., 1993. Detection of polychlorinated Alcaligenes sp. KF711: overexpression, enzyme purification,
biphenyl degradation genes in polluted sediments by direct and nucleotide sequencing. Arch. Biochem. Biophys. 332,
DNA extraction and polymerase chain reaction. Appl. Environ. 248 – 254.
Microbiol. 59, 4065 – 4073. Muyzer, G., de Waal, E.C., Uitterlinden, A.G., 1993. Profiling of
Ghosal, D., You, I.S., Gunsalus, I.C., 1987. Nucleotide sequence complex microbial populations by denaturing gradient gel elec-
and expression of gene nahH of plasmid NAH7 and homology trophoresis analysis of polymerase chain reaction-amplified
with gene xylE of TOL pWWO. Gene 55, 19 – 28. genes coding for 16S rRNA. Appl. Environ. Microbiol. 59,
Hamelin, J., Fromin, N., Tarnawski, S., Teyssier-Cuvelle, S., 695 – 700.
Aragno, M., 2002. nifH gene diversity in the bacterial commun- Nakai, C., Kagamiyama, H., Nozaki, M., Nakazawa, T., Inouye, S.,
ity associated with the rhizosphere of Molinia coerulea, an oli- Ebina, Y., Nakazawa, A., 1983. Complete nucleotide sequence
gonitrophilic perennial grass. Environ. Microbiol. 4, 477 – 481. of the metapyrocatechase gene on the TOL plasmid of Pseudo-
Heiman, M., 1997. Webcutter 2.0. http://www.firstmarket.com/ monas putida mt-2. J. Biol. Chem. 258, 2923 – 2928.
cutter/cut2.html. Nicolaisen, M.H., Ramsing, N.B., 2002. Denaturing gradient gel
Henckel, T., Friedrich, M., Conrad, R., 1999. Molecular analyses of electrophoresis (DGGE) approaches to study the diversity of am-
the methane-oxidizing microbial community in rice field soil by monia-oxidizing bacteria. J. Microbiol. Methods 50, 189 – 203.
targeting the genes of the 16S rRNA, particulate methane mono- Nicolas, K.B., 1997. GeneDoc 2.6.02. http://www.psc.edu/biomed/
oxygenase, and methanol dehydrogenase. Appl. Environ. Mi- genedoc/.
crobiol. 65, 1980 – 1990. Orita, M., Iwahana, H., Kanazawa, H., Hayashi, K., Sekiya, T.,
Herrmann, H., Muller, C., Schmidt, I., Mahnke, J., Petruschka, L., 1989. Detection of polymorphisms of human DNA by gel elec-
Hahnke, K., 1995. Localization and organization of phenol deg- trophoresis as single-strand conformation polymorphisms. Proc.
radation genes of Pseudomonas putida strain H. Mol. Gen. Natl. Acad. Sci. U. S. A. 86, 2766 – 2770.
Genet. 247, 240 – 246. Ouchiyama, N., Miyachi, S., Omori, T., 1998. Cloning and nucleo-
Hillis, D.H., Moritz, H., Mable, B.K., 1996. Molecular System- tide sequence of carbazole catabolic genes from Pseudomonas
atics. Sinauer Associates, Sunderland, MA, USA, pp. 240 – 279. stutzeri strain OM1, isolated from activated sludge. J. Gen.
Hopper, D.J., Kemp, P.D., 1980. Regulation of enzymes of the 3,5- Appl. Microbiol. 44, 57 – 63.
xylenol-degradative pathway in Pseudomonas putida: evidence Perriere, G., Gouy, M., 1996. WWW-query: an on-line retrieval
for a plasmid. J. Bacteriol. 142, 21 – 26. system for biological sequence banks. Biochimie 78, 364 – 369.
Kanakaraj, R., Harris, D.L., Songer, J.G., Bosworth, B., 1998. Porteous, L.A., Widmer, F., Seidler, R.J., 2002. Multiple enzyme
Multiplex PCR assay for detection of Clostridium perfringens restriction fragment length polymorphism analysis for high res-
in feces and intestinal contents of pigs and in swine feed. Vet. olution distinction of Pseudomonas (sensu stricto) 16S rRNA
Microbiol. 63, 29 – 38. genes. J. Microbiol. Methods 51, 337 – 348.
Keil, H., Keil, S., Pickup, R.W., Williams, P.A., 1985. Evolutionary Ringelberg, D.B., Talley, J.W., Perkins, E.J., Tucker, S.G., Luthy,
conservation of genes coding for meta pathway enzymes within R.G., Bouwer, E.J., Fredrickson, H.L., 2001. Succession of phe-
TOL plasmids pWW0 and pWW53. J. Bacteriol. 164, 887 – 895. notypic, genotypic, and metabolic community characteristics
Kent, A.D., Triplett, E.W., 2002. Microbial communities and their during in vitro bioslurry treatment of polycyclic aromatic hydro-
interactions in soil and rhizosphere ecosystems. Annu. Rev. carbon-contaminated sediments. Appl. Environ. Microbiol. 67,
Microbiol. 56, 211 – 236. 1542 – 1550.
Khetmalas, M.B., Egger, K.N., Massicotte, H.B., Tackaberry, L.E., Sambrook, J., Fritsch, E.F., Maniatis, T., 1989. Molecular Cloning:
Clapperton, M.J., 2002. Bacterial diversity associated with sub- A Laboratory Manual. Cold Spring Harbor Laboratory Press,
alpine fir (Abies lasiocarpa) ectomycorrhizae following wildfire Cold Spring Harbor, NY.
and salvage-logging in central British Columbia. Can. J. Micro- Sikorski, J., Mohle, M., Wackernagel, W., 2002. Identification of
biol. 48, 611 – 625. complex composition, strong strain diversity and directional
Kitayama, A., Achioku, T., Yanagawa, T., Kanou, K., Kikuchi, M., selection in local Pseudomonas stutzeri populations from marine
Ueda, H., Suzuki, E., Nishimura, H., Nagamune, T., Kawaka- sediment and soils. Environ. Microbiol. 4, 465 – 476.
mi, Y., 1996. Cloning and characterization of extradiol aro- Staley, J.T., Konopka, A., 1985. Measurement of in situ activities of
matic ring-cleavage dioxygenases of Pseudomonas aeruginosa nonphotosynthetic microorganisms in aquatic and terrestrial
JI104. J. Ferment. Bioeng. 82, 217 – 223. habitats. Annu. Rev. Microbiol. 39, 321 – 346.
Mesarch, M.B., Nakatsu, C.H., Nies, L., 2000. Development of Thompson, J.D., Gibson, T.J., Plewniak, F., Jeanmougin, F.,
catechol 2,3-dioxygenase-specific primers for monitoring biore- Higgins, D.G., 1997. The CLUSTAL_X windows interface:
mediation by competitive quantitative PCR. Appl. Environ. Mi- flexible strategies for multiple sequence alignment aided by
crobiol. 66, 678 – 683. quality analysis tools. Nucleic Acids Res. 25, 4876 – 4882.
Meyer, S., Moser, R., Neef, A., Stahl, U., Kampfer, P., 1999. Differ- Torsvik, V., Ovreas, L., 2002. Microbial diversity and function in
ential detection of key enzymes of polyaromatic-hydrocarbon- soil: from genes to ecosystems. Curr. Opin. Microbiol. 5,
degrading bacteria using PCR and gene probes. Microbiology 240 – 245.
145, 1731 – 1741. Van de Peer, Y., De Wachter, R., 1993. TREECON: a software
Moon, J., Min, K.R., Kim, C.K., Min, K.H., Kim, Y., 1996. Char- package for the construction and drawing of evolutionary trees.
acterization of the gene encoding catechol 2,3-dioxygenase of Comput. Appl. Biosci. 9, 177 – 182.
12. 708 H. Junca, D.H. Pieper / Journal of Microbiological Methods 55 (2003) 697–708
Weinbauer, M.G., Ho ¨fle, M.G., 2001. Quantification of nucleic Williams, P.A., Jones, R.M., Shaw, L.E., 2002. A third transpos-
acids from aquatic environments by using green-fluorescent able element, ISPpu12, from the toluene – xylene catabolic plas-
dyes and microtiter plates. In: Akkermans, A., van Elsas, J., mid pWW0 of Pseudomonas putida mt-2. J. Bacteriol. 184,
de Bruijn, F. (Eds.), Molecular Microbial Ecology Manual, 6572 – 6580.
5th Supplement 2.1.3. Kluwer, Dordrecht. Yan, T., Fields, M.W., Wu, L., Zu, Y., Tiedje, J.M., Zhou, J., 2003.
Widada, J., Nojiri, H., Omori, T., 2002. Recent developments in Molecular diversity and characterization of nitrite reductase
molecular techniques for identification and monitoring of xeno- gene fragments (nirK and nirS) from nitrate- and uranium-con-
biotic-degrading bacteria and their catabolic genes in bioreme- taminated groundwater. Environ. Microbiol. 5, 13 – 24.
diation. Appl. Microbiol. Biotechnol. 60, 45 – 59. Yeates, C., Holmes, A.J., Gillings, M.R., 2000. Novel forms of
Wikstrom, P., Wiklund, A., Andersson, A.C., Forsman, M., 1996. ring-hydroxylating dioxygenases are widespread in pristine
DNA recovery and PCR quantification of catechol 2,3-dioxyge- and contaminated soils. Environ. Microbiol. 2, 644 – 653.
nase genes from different soil types. J. Biotechnol. 52, 107 – 120.