Functional adaptation of microbial communities from jet fuel-contaminated soil under bioremediation treatment: simulation of pollutant rebound

R E S E A R C H A R T I C L E

Functional adaptationofmicrobial communities from jet fuel-contaminated soil under bioremediation treatment: simulationofpollutant reboundOlesya Korotkevych1, Jirina Josefiova1, Martina Praveckova1, Tomas Cajthaml1, Monika Stavelova2 &Maria V. Brennerova1

1Department of Cell Molecular Microbiology, Institute of Microbiology, Prague, Czech Republic; and 2AECOM CZ Ltd, Prague, Czech Republic

Correspondence: Maria V. Brennerova,

Department of Cell Molecular Microbiology,

Institute of Microbiology v. v. i., Videnska

1083, 142 20 Prague, Czech Republic. Tel.:

1420 241 062 781; fax: 1420 241 722 257;

e-mail: [email protected]

Received 5 December 2010; revised 21 June

2011; accepted 27 June 2011.

Final version published online 1 August 2011.

DOI:10.1111/j.1574-6941.2011.01169.x

Editor: Michael Schloter

Keywords

mesocosms; biodegradation;

phytoremediation; qPCR of catabolic genes;

DGGE.

Abstract

To investigate the link between the functionality and the diversity of microbial

communities under strong selective pressure from pollutants, two types of meso-

cosms that simulate natural attenuation and phytoremediation were generated using

soil from a site highly contaminated with jet fuel and under air-sparging treatment.

An increase in the petroleum hydrocarbon concentration from 4900 to

18 500 mg kg�1 dw soil simulated a pollutant rebound (postremediation pollutant

reversal due to residual contamination). Analysis of soil bacterial communities by

denaturing gradient gel electrophoresis of PCR-amplified 16S rRNA gene fragments

showed stronger changes and selection for a phylogenetically diverse microbial

population in the mesocosms with pollutant-tolerant willow trees. Enumeration of

the main subfamilies of catabolic genes characteristic to the site detected a rapid

increase in the degradation potential of both systems. A marked increase in the

abundance of genes encoding extradiol dioxygenases with a high affinity towards

various catecholic substrates was found in the planted mesocosms. The observed

adaptive response to the simulated pollutant rebound, characterized by increased

catabolic gene abundance, but with different changes in the microbial structure, can

be explained by functional redundancy in biodegrading microbial communities.

Introduction

Subsurface spills of petroleum compounds that have disas-

trous consequences for the biotic and abiotic components

of ecosystems are the most frequently cited cause of

groundwater contamination (Okoh, 2006; Glick, 2010; Das

& Chandran, 2011). The large family of several hundred

hydrocarbon compounds that originally come from crude

oil is described by the term total petroleum hydrocarbons

(TPH), which is defined as the measurable amount of

hydrocarbons (Thompson & Nathanail, 2003). In addition

to aliphatic hydrocarbons, petroleum hydrocarbons contain

benzene, toluene, ethylbenzene and xylenes (BTEX). These

hazardous substances are major components of gasoline and

jet fuels, and they are regulated by many nations. The US

Environmental Protection Agency and the Agency for Toxic

Substances and Disease Registry have compiled a list of the

most frequently observed toxic compounds, wherein TPH

components such as n-hexane, monoaromatic and polyaro-

matic hydrocarbons, fuel oils, gasoline and hydraulic fluids

are cataloged (Glick, 2010).

Biodegradation is a process whereby microorganisms play

a major role in the biological conversion of hazardous

pollutants to innocuous products and has been reported to

be one of the primary mechanisms by which petroleum and

other hydrocarbon pollutants can be removed from the

environment (Das & Chandran, 2011). Biodegradation and

the use of plants to remediate polluted soils (phytoremedia-

tion) are believed to be noninvasive, effective and inexpen-

sive technologies (Glick, 2010; Tang et al., 2010).

Over the past three decades, investigations of the bio-

chemical pathways, genes and proteins responsible for the

enzymatic degradation of mono- and polyaromatic pollu-

tants have focused on cultivation techniques and the isola-

tion of bacterial strains useful for biodegradation (Yeates

et al., 2000; Witzig et al., 2006). Aerobic bacteria have been

FEMS Microbiol Ecol 78 (2011) 137–149 c� 2011 Federation of European Microbiological SocietiesPublished by Blackwell Publishing Ltd. All rights reserved

MIC

ROBI

OLO

GY

EC

OLO

GY

extensively studied with regard to their ability to utilize

aromatic compounds as the sole carbon and energy sources

and their environmental plasticity (Harwood & Parales,

1996; Pieper, 2005). The organization and regulation of

catabolic genes has been characterized, and catabolic operon

prototypes have been described (Harayama et al., 1987; Eltis

& Bolin, 1996; Jimenez et al., 2002; Parales & Harwood,

2002; McLeod & Eltis, 2008). The metagenomic approach

has further enabled the direct and detailed evaluation of

microbial community networks without cultivation (Rie-

senfeld et al., 2004) and the identification of enzymes

involved in the metabolism of aromatic compounds (Sue-

naga et al., 2007; Brennerova et al., 2009). These achieve-

ments have led to new horizons for monitoring natural and

anthropogenically endangered environments for new genes

and attenuation processes.

Hradcany is a former military airbase (5013709.95500N,

14143054.53100E) that occupies a region that is the most

intensively polluted with petroleum hydrocarbons in North-

ern Bohemia, the Czech Republic. More than 28 ha were

contaminated by military activities from 1940 to 1991. Jet

fuel comprised 70% of the contamination. TPH concentra-

tions varied from 5000 to 55 000 mg kg�1 dry soil and BTEX

was detected at 1000 mg kg�1 (Machackova et al., 2008).

Based on risk assessments, the Czech Ministry of Environ-

ment dictated the TPH clean-up target limit for soil to be

5000 mg kg�1 dry soil. A remediation project was initiated in

1997 that involved in situ soil vacuum extraction, injection

of contaminant-free air into the subsurface saturated zone,

known as air sparging, and additional stimulation of the soil

microorganisms by nutrient amendment. At the end of

2008, the decontamination measures were terminated. Re-

vitalization and utilization of the site with the introduction

of plants as facilitators of residual contaminant removal was

planned by 2012. Many crop plants and grasses (maize, rice,

legumes, sorghum, ryegrass, Bermuda grass and beggar

ticks) are effective in degrading TPH (Shirdam et al.,

2008; Glick, 2010; Tang et al., 2010). Willows and poplar

trees are alternative phytoremediation species. Their reme-

diation potential is proportional to survival, fast growth,

height, deep root system and volume production, which are

measures of potential benefit for long-term petroleum

degradation by the trees and their rhizosphere-associated

microorganisms (Trapp et al., 2001; Zalesny et al., 2005).

Willows (Salix euxina, also known as Salix fragilis) are part

of the native vegetation of Hradcany. Their use for revitali-

zation of the site requires the collection of missing experi-

mental evidence under conditions that are similar to the

field environment (Glick, 2010).

Pollutants in the soil can frequently return to levels

detected before remediation initiation (Switzer & Kosson,

2007). The contamination reversal, called pollutant re-

bound, is a post-treatment phenomenon that follows the

interruption of soil vacuum extraction and air sparging, and

is caused by alterations in the pollution equilibrium between

groundwater, soil and soil gas. This postremediation pollu-

tant concentration increase may impose environmental

stress on the microbial communities present in the soil and

rhizosphere. On the other hand, the in situ stimulation of

the natural microbial communities at the Hradcany site has

enabled strong selection for meta-cleavage degradation

pathways (Brennerova et al., 2009); thereby, the soil’s

microbial populations evolving after a bioremediation treat-

ment are ideal model systems to study the pollutant

rebound related to changes in microbial biodegradation

potential and adaptability.

The main objectives of this study were (1) to define the

level of tolerance of willow trees to the contaminant, (2) to

compare changes in the soil microbial populations in two

different mesocosm systems modeling natural attenuation

and phytoremediation, (3) to study the ability of microbial

populations to adapt their catabolic machinery to the

biodegradation process following a sudden increase in

pollutant concentration and (4) to investigate the possible

dependence of catabolic potential on shifts in the structure

of the microbial community. This work furthers the devel-

opment of phytoremediation techniques by undertaking a

methodical assessment of phytoadaptation to petroleum

hydrocarbons and provides new insight into the relationship

between the dynamics of microbial populations and their

degradation potential.

Materials and methods

Soil analysis

The contaminated soil used in the study was excavated at a

depth of 1.5–2.5 m from an area in Hradcany (field G) that

had been intensively cleaned for 7.5 years. The analysis of

grain sizes and petroleum hydrocarbons in the soil was

carried out by the accredited laboratory ALS Czech Republic

Ltd. Fractions of soil particles from 4 2 to 0.063 mm were

assessed using the sieving method, and other fractions were

identified from the o 0.063-mm fraction by a laser particle

size analyzer using the liquid dispersion mode (test method

CZ_SOP_D06_07_N11). The TPH concentration in the soil

was estimated as the average of nine samples measured using

Fourier transform infrared spectroscopy (Czech standards

CSN 757505 and CSN 757506). In light non-aqueous-phase

liquid (LNAPL), the analysis of hydrocarbon fractions with

carbon numbers C10 through C40 was performed by GC

(CSN EN ISO 9377-2). Additional BTEX analyses, in both

soil and LNAPL, were conducted according to methods for

the chromatographic/mass spectrometric (GC–MS) detec-

tion of volatile organic compounds (VOC) in EPA 624 and

EPA 8260. PAHs were analyzed as semi-VOC by GC–MS

FEMS Microbiol Ecol 78 (2011) 137–149c� 2011 Federation of European Microbiological SocietiesPublished by Blackwell Publishing Ltd. All rights reserved

138 O. Korotkevych et al.

according to methods EPA 8270, EPA 8131, EPA 8091 and

Czech normative CSN EN ISO 6468.

Soil was suspended in water at a 1 : 1 ratio by shaking in the

dark for 1 h, and the pH was measured using a PH114 (Snail

Instruments, Czech Republic). The water content was deter-

mined by drying the soil at 105 1C for 10 h and was expressed

as a percentage of the sample weight (CSN ISO 11465).

The numbers of cultivable heterotrophic bacteria were

determined to monitor the influence of pollutant rebound

on the soil microbiota. Soil samples were homogenized in

0.9% NaCl and agitated for 1 h at 4 1C. Appropriately

diluted aliquots were spread on R2A agar (BD DifcoTM)

and incubated at 24 1C in closed glass jars in the presence of

LNAPL vapors. Three independent determinations of the

CFU g�1 dry soil were performed in triplicate.

Mesocosm experiments

Mesocosm cultivation systems were constructed outdoors

using six 45-cm� 45-cm� 50-cm cultivation boxes built

from white polypropylene (Fig. 1). The boxes were filled

with a 5-cm layer of keramzite gravel and 60 kg of

Fig. 1. Mesocosm setup: (a) graphic representation of mesocosm systems that model natural bioremediation (pot N) and phytoremediation (pot P).

(b) Picture of the mesocosms on day 30 of the experiment.

FEMS Microbiol Ecol 78 (2011) 137–149 c� 2011 Federation of European Microbiological SocietiesPublished by Blackwell Publishing Ltd. All rights reserved

139Functional redundancy in biodegrading microbial communities

contaminated soil mixed with LNAPL. Three mesocosm

boxes were assigned as models for natural attenuation (N).

Another three mesocosms were used for a comparative

analysis of rhizosphere-enhanced phytoremediation (P).

Pilot experiments were carried out to determine the thresh-

old concentration of pollutants tolerated by the young trees.

Three-month-old hydroponically rooted willow cuttings

were compared with cuttings rooted in a garden substrate.

Direct transfer into soil containing TPH at 13 000 mg kg�1 dw

was lethal for the hydroponically pregrown willows and had

detrimental effects on the young trees when their rhizospheres

had developed in soil. In parallel mesocosms with a twofold

lower soil TPH concentration, the organic phase had a toxic

impact on the hydroponic plants, but was tolerated by the

willow cuttings with soil-grown rhizospheres. Therefore, slow

adaptation of the trees was chosen for the final mesocosm

experiment. Crack willow (S. euxina, also known as S. fragilis)

cuttings that were 30–35 cm in length and approximately 1 cm

in diameter and had 3-month-old root systems developed in

1.5 L of garden substrate were gradually adapted to the

contaminant by transfer to pots containing 3 L of soil from

field G. They were grown for an additional month under

conditions of a gradually increasing TPH concentration by

weekly admixing of LNAPL aliquots into the substrate to a

final concentration of 13 000 mg kg�1 dry soil. Four adapted

young trees were transferred into each P container, and the

water content of the soil was maintained between 8.3% and

14.7%. The sandy soil, which quickly drained and dried out,

was mixed at the sides and the center of the pots once every 2

weeks, thus providing additional air access. Soil sampling was

carried out on days 1, 15, 32, 62, 90 and 126. Bulk soil was

collected at five sampling points in each pot by augering (100 in

diameter) to a depth of 20–25 cm and, in the case of P pots,

10–13 cm from the plants. The control sample, termed ‘day 0’,

was taken before mixing the soil with the LNAPL. An

additional control, termed ‘rhizosphere’, was collected from

the rhizosphere-associated soil at the termination of meso-

cosm P cultivation. All soil samples were homogenized, stored

in airtight sterile glass bottles, placed on ice and then used for

DNA isolation, enumeration of cultivable soil microorganisms

and analyses of TPH concentration, pH and water content.

Changes in the microbial community structure and in cata-

bolic gene concentrations in both the N and the P systems

were followed over a period of 4 months during the growing

season (June 15, 2006–November 2, 2006).

Soil DNA extraction

High-molecular-weight DNA was recovered from 10 g of soil

as described previously (Brennerova et al., 2009). The DNA

was purified further by humic acid removal using the

PowerClean DNA Clean-Up Kit (MoBio Laboratories,

Carlsbad, CA). The concentration of DNA was determined

using a spectrophotometer (BioMate 5, Thermo Spectronic)

and a NanoDrop ND-1000 UV-VIS (NanoDrop Technolo-

gies). DNA was stored at 4 1C during the course of the study.

Denaturing gradient gel electrophoresis (DGGE)analysis of partial 16S rRNA genes

The V3 region of the 16S rRNA gene was amplified by a PCR

using the 338f forward primer, which is specific for a region

conserved among members of the domain Bacteria (Ovreas

et al., 1997), and the 518r universal primer (Muyzer et al., 1993)

to amplify a product with an average length of 196 bp. A 40-bp

GC clamp (50-CGC CCG CCG CGC CCC GCG CCC GGC

CCG CCG CCG CCG CCG C-30) was attached to the 50 end of

the forward primer (Myers et al., 1987) for DGGE system

adaptation. For sequencing, 16S rRNA gene sequences from the

V3-V4 hypervariable regions were retrieved by PCR with the

primers GC-338f (Ovreas et al., 1997) and 802r (Tamura &

Hatano, 2001), which span positions 802-785 of the Escherichia

coli 16S rRNA gene. Amplification was performed using a

Mastercycler ep (Eppendorf AG, Hamburg, Germany). The 50-

mL reaction mixture contained 50 ng of soil DNA, 10 pmol each

primer, 200mM each dNTP, 2.5 mM MgCl2, 5mL 10�PCR

buffer and 1.25 U Thermo-Starts DNA Polymerase (ABgene,

UK). The enzyme was activated at 95 1C for 15 min. To increase

the specificity of amplification and reduce the formation of

spurious byproducts, a ‘touchdown’ hotstart PCR was per-

formed. The initial annealing temperature (61 1C) was de-

creased by 1 1C during every cycle until an annealing

temperature of 55 1C was reached, and 28 additional cycles

were carried out at 94 1C (denaturing), 55 1C (annealing) and

72 1C (elongation) for 45 s at each temperature. The final

primer extension was performed at 72 1C for 10 min. The PCR

products, together with pBR322 DNA/AluI Marker (Fermentas,

Lithuania), were separated on 2% agarose gels, visualized by

ethidium bromide staining and quantified using KODAK mole-

cular imaging software v4.5 (Rochester). PCR amplicons were

then separated by DGGE using a BioRad D-Code System

(BioRad, Hercules). Sample volumes containing 500–600 ng of

PCR product were applied to an 8% w/v polyacrylamide–bisa-

crylamide (37.5 : 1) gel in 1�TAE. The PCR amplicons

obtained with the primers GC-338f and 802r were separated

in a 7% gel. Optimal separation was achieved at 60 1C with a

parallel 45–60% urea–formamide denaturing gradient increas-

ing in the direction of electrophoresis. Denaturant at 100%

corresponded to 7 M urea and 40% v/v formamide. The gels

were electrophoresed for 15 min at 20 V, then for 24 h at 55 V

and stained with SYBRs Green I (1 : 10 000; Molecular Probes,

Eugene) in 25 mM Tris-HCl (pH 8) for 20 min and photo-

graphed using a Kodak EDAS 290 camera. PCR runs for DGGE

and DGGE fingerprinting were performed in triplicate, and the

resulting fingerprints were analyzed using GELCOMPAR II software

v5.1 (Applied Maths, Belgium).

FEMS Microbiol Ecol 78 (2011) 137–149c� 2011 Federation of European Microbiological SocietiesPublished by Blackwell Publishing Ltd. All rights reserved

140 O. Korotkevych et al.

Sequencing of DNA from DGGE bands andphylogenetic analysis

DGGE separation of amplicons with approximate lengths of

456 bp was achieved under the conditions described above.

Selected resolved bands were excised from the polyacryl-

amide gels under UV illumination and incubated in 20 mL

nanopure water at 4 1C for 20 h. Aliquots (5 mL) served as

templates for PCR reamplification with the same primers

(without GC clamp), followed by purification using the

Wizards SV Gel and PCR Clean-Up System (Promega,

Madison, WI) and cloning into the pGEM-T easy vector

system (Promega). Transformant colonies were suspended

in water and incubated at 95 1C for 10 min, and plasmid

inserts were amplified by PCR with vector-specific M13

forward and reverse primers. Restriction fragment length

polymorphism (RFLP) analysis of inserts from 20 transfor-

mants was carried out by restriction with BsuRI, FspBI and

Tru1I purchased from Fermentas, in 10-mL reaction mix-

tures, with subsequent separation on 3.0% SeaKems LE

agarose gels (Cambrex Bio Science Rockland). Distinct

RFLP types were selected for sequencing of both strands

using M13 primers by BigDyes Terminator v3.1 chemistry

according to the manufacturer’s instructions (Applied Bio-

systems, Foster City, CA) on an ABI PRISM 3100 Genetic

Analyzer (Applied Biosystems).

DNA similarity searches of GenBank databases were per-

formed using the BLASTN program from the National Center for

Biotechnology Information (NCBI) website. Vector sequences

were removed using the KODON program v3.6 (Applied Maths),

and the sequences were oriented 50–30 relative to known 16S

rRNA genes using ORIENTATIONCHECKER (Bioinformatics Toolkit,

Cardiff School of Biosciences, UK). The sequences were

analyzed for potential chimeric sequences using the services

available at the Ribosomal Database Project II (Cole et al.,

2009). Additional potential chimeras were assessed using the

program BELLEROPHON (Huber et al., 2004). Alignments were

generated using the CLUSTALX 1.83 Windows interface of the

CLUSTALW program running default values. A block of sequence

alignments was selected using the GENEDOC multiple sequence

alignment editor software (Nicholas et al., 1997). Phylogenetic

trees were obtained by MEGA 4.1 software (Tamura et al., 2007)

using the neighbor-joining algorithm (Saitou & Nei, 1987)

with a p-distance model and pairwise deletion of gaps and

missing data. A consensus tree was inferred from a total of

1000 bootstrap trees generated for each data set. The sequences

of the 16S rRNA gene PCR fragments obtained in this study

are available under the EMBL/GenBank/DBBJ accession num-

bers GU560728–GU560730 and GU568253–GU568336.

Quantitative real-time PCR (qPCR) analysis ofcatabolic genes in soil samples

A real-time PCR method for the absolute quantification of the

main catabolic genes characteristic of the Hradcany site

(Brennerova et al., 2009) was developed. The primers intro-

duced were specific for four groups of dioxygenase genes

(Table 1). The forward and reverse primers bphAf668-3 and

bphAr1153-2 (Witzig et al., 2006) target conserved regions of

genes encoding the catalytic a-subunits of the Rieske non-

heme iron oxygenases, which are members of the toluene/

biphenyl oxygenase subfamily of aromatic ring-hydroxylating

dioxygenases (RHDO). The primer set EXDO-Dbt-F/EXDO-

Dbt-R allowed the amplification of metagenomic gene frag-

ments that encode peptides showing 55–64% sequence iden-

tity to extradiol dioxygenases involved in the degradation of

naphthalene, specifically to DbtC (Brennerova et al., 2009).

The group of targeted genes was named EXDO-Dbt after their

phylogenetic affiliation with dbtC (AF404408) of Burkholderia

sp. DBT1 (Di Gregorio et al., 2004). The primer set EXDO-

K2-F/EXDO-K2-R targets a third group of genes (EXDO-K2

genes) encoding proteins with 78–81% identity to the I.3.A

extradiol dioxygenases (Eltis & Bolin, 1996), which is reported

to be involved in the degradation of isopropylbenzene and

ethylbenzene. EXDO-K2 genes include ipbC (U53507) in

Pseudomonas sp. JR1 (degrading isopropylbenzene) (Pflugma-

cher et al., 1996) and ebdC (AJ293587) in P. putida 01G3

(degrading ethylbenzene) (Chablain et al., 2001). In our

previous study, the highly degenerate primer pair EXDO-D-

F/EXDO-D-R was used to amplify phylogenetically diverse

genes from the EXDO-D group, which corresponds to the

I.2.C subfamily of catechol 2,3-dioxygenases (Brennerova

Table 1. Oligonucleotide primers used in this study

Primers Sequence (50–30)

Amplicon

size (bp)

Annealing

temperature ( 1C)

Elongation

time (s) References

bphAF668-3 GTTCCGTGTAACTGGAARTWYGC 535 58 55 Witzig et al. (2006)

bphAR1153-2 CCAGTTCTCGCCRTCRTCYTGHTC

EXDO-K2-F GAAAAAGTGGGTTTGATGGAGG 810 63 70 Brennerova et al. (2009)

EXDO-K2-R CGCTTATGCCKCGTCATCACCC

EXDO-Dbt-F TCCGCATGGATTACAACC 423 58 45 Brennerova et al. (2009)

EXDO-Dbt-R GATCTGTGGAACGGGCAA

EXDO-D2-F GCTGGATCATTGCCTGTTG 314 64 35 This study

EXDO-D2-R GCGTGGGGGTGACATCM

FEMS Microbiol Ecol 78 (2011) 137–149 c� 2011 Federation of European Microbiological SocietiesPublished by Blackwell Publishing Ltd. All rights reserved

141Functional redundancy in biodegrading microbial communities

et al., 2009). The formation of primer-dimers during qPCR

was avoided by designing the more specific primer set EXDO-

D2-F/EXDO-D2-R. OLIGO EXPLORER and OLIGO ANALYZER soft-

ware (Teemu Kuulasmaa, Finland) were used for primer

construction and for confirming both the optimal DG values

and the lack of hairpin structures formation. In contrast to the

originally used highly degenerate EXDO-D primer set that was

used for screening of metagenomic libraries (Brennerova et al.,

2009), the new EXDO-D2-F primer was not degenerate and

the EXDO-D2-R primer had only one degenerate nucleotide

at the 30-end position (Supporting Information, Fig. S4). The

new primers allowed the amplification of an abundant cluster

of genes (EXDO-D2 genes) showing similarity to btxH

(DQ834383) in Ralstonia sp. strain PHS1 (which is able to

utilize toluene, ethylbenzene, o-xylene and cresols) and to tbuE

(U20258) in Ralstonia pickettii strain PK01 (which degrades

toluene, benzene, phenol and cresols) (Kukor & Olsen, 1991).

An absolute calibration model was developed from a

series of DNA standards that utilizes recombinant fosmids

of known copy number (ABI, 2006; Sivaganesan et al., 2010)

and their associated quantification cycle (Cq). For the

purpose of generating standard curves, DNA was isolated

from the metagenomic fosmid clones s115, s37, s79 and s76,

which harbor the RHDO, EXDO-D2, EXDO-K2 and

EXDO-Dbt genes, respectively (Brennerova et al., 2009).

Fosmid DNA was isolated under conditions that induce

high-copy amplification and was purified using the Qiagen

Plasmid Mini Kit system (Qiagen, Hilden, Germany). The

plasmids were linearized by NotI digestion and their size was

determined by pulse-field gel electrophoresis. The standards

were divided into small aliquots, stored at � 20 1C and

thawed only once before use. The standard curve was

generated from Cq measurements of 10-fold serial dilutions

of the standard (103–108 gene copies). Soil DNA with an

unknown concentration of the target genes was analyzed in

the same instrument run. Real-time PCR was conducted on

a Mastercycler ep Realplex Thermocycler (Eppendorf AG).

Amplifications were performed in 25-mL reaction volumes

containing 1� Express SYBR GreenER qPCR SuperMix

(Invitrogen, Carlsbad, CA), 0.2 mM each primer and 2mL

soil DNA (10–20 ngmL�1) or 5 mL of plasmid DNA used as

the qPCR standard. A mixture of all PCR reagents without

any DNA was used as a no-template control (NTC).

Temperature gradient qPCR with subsequent melting curve

DNA analysis and agarose gel electrophoresis was performed

to determine the optimal conditions (annealing temperature

and extension time; Table 1) required for the amplification

of a specific product with each primer set. The final qPCR

runs were carried out under the following conditions: step

one, 50 1C for 1 min (UDG incubation); step two, 95 1C for

2 min; and step three, 40 cycles of 95 1C for 15 s, for 20 s at

the primer-specific annealing temperature and elongation

time at 72 1C (Table 1). SYBR Green I fluorescence signal

intensity was measured at 0.5 1C increments every 15 s, from

60 to 95 1C. The amplified products were analyzed in

ethidium bromide-stained agarose gels (Fig. S5) and by

real-time fluorescence incorporation. The phylogenetic af-

filiation with the expected dioxygenases was confirmed by

sequencing of the PCR products as described previously

(Brennerova et al., 2009). The postrun data analyses were

performed using EPPENDORF REALPLEX software v1.5 (Eppen-

dorf AG). Triplicate Cq measurements were collected to

generate a standard curve. Instrument runs with NTCs Cq

values Z35 were further evaluated. Potential PCR inhibition

due to the coextraction of humic substances with the soil

DNA (Van Doorn et al., 2009) was assessed by comparing

gene copy quantification in serial dilutions of the template.

The final number of target genes, reported as gene co-

pies ng�1 DNA, was an average of triplicate measurements

from two independent DNA extractions made from each

soil sample. For both standard and target DNA, the ampli-

fication efficiency (E) was estimated at 95% for qPCR with

the primer set bphAf668-3/bphAr1153-2, 88% for EXDO-

K2-, 90% for EXDO-Dbt- and 87% for EXDO-D2-. The

correlation coefficient values (r2) were 0.99 for all runs.

Results

Mesocosm simulation of pollutant rebound

Soil microbial communities and their degradation potential

were examined in natural remediation (N) and phytoreme-

diation (P) systems under conditions of pollutant rebound.

The test was performed outdoors to simulate real growing

conditions (Fig. 1). The soil consisted of 78% medium- and

coarse-grained sand particles, 9% silt and 2% clay. Its TPH

concentration was estimated at 4900 mg kg�1 dry soil. After

mixing with LNAPL, its TPH concentration was measured at

18 500 mg kg�1 dry soil. Thus, at the start of the degradation

experiment, the TPH content of the soil was elevated 3.8-

fold. Analyses of the LNAPL revealed that C6–C20 hydro-

carbons consisting of 59.9% C10–C12 and 38.2% C12–C16

hydrocarbons (characteristic of kerosene-type jet fuel) were

predominant among the C10–C40 analytical fractions

(69 800mg L�1) (Fig. S1). In addition, BTEX (8270mg L�1),

190.08 mg L�1 PAHs, such as naphthalene (184mg L�1), ace-

naphthalene (0.789mg L�1), acenaphthene (2.08mg L�1),

fluorene (1.89mg L�1), phenanthrene (0.736mg L�1), fluor-

anthene (0.136mg L�1), pyrene (0.141mg L�1), benzanthra-

cene (0.021mg L�1), chrysene (0.023mg L�1) and

benzo(b)fluoranthene (0.036mg L�1) were present. Willows

that had been gradually adapted to high TPH concentra-

tions (13 000 mg kg�1 dw) tolerated subsequent transfer to

the P mesocosms, where the pollutant content was brought

up to 18 500 mg kg�1 dw soil. Ten of the 12 plants survived

the first 10 days of cultivation. The average pH value of the

FEMS Microbiol Ecol 78 (2011) 137–149c� 2011 Federation of European Microbiological SocietiesPublished by Blackwell Publishing Ltd. All rights reserved

142 O. Korotkevych et al.

soil in all the mesocosms was 5.8 during the first 62 days and

decreased to 5.0 by the end of cultivation.

DGGE assay of changes in the bacterialcommunity structure

Control 16S rRNA gene fingerprints of the control DNA

(day 0) were compared with the community profiles of bulk

soil samples taken at days 15, 32, 62, 90 and 126 of

mesocosm development without (pot N) or with (pot P)

young willows adapted to LNAPL. GELCOMPAR II software

analysis of the bacterial fingerprints is shown in Fig. 2. The

3.8-fold increase in the pollutant concentration to

18 500 mg kg�1 dry soil did not cause significant changes in

the microbial community profiles during the first month of

mesocosm cultivation. The DGGE band patterns on days 15

and 32 of the P mesocosms showed 50% similarity and those

of the N mesocosms showed 60% similarity. In the later

phases of mesocosm development, more pronounced mi-

crobial shifts were detected in the phytoremediation systems

(P). Between days 62 and 90, the microbial fingerprints of P

soil samples showed 30.8% similarity. Further changes were

observed in the microbial fingerprint of bulk soil from the P

mesocosms on day 126. Thus, at the end of the experiment,

the bacterial population in the P pots retained 20.2%

similarity to the other DGGE patterns. On the other hand,

the phylogenetic profile of the control ‘rhizosphere’ (soil

adhering to roots of the willows) shared only 7.7% of the

bands of the bulk soil community patterns of the N and P

mesocosms. In parallel with the described phylogenetic

changes, the growth of cultivable bacteria was also observed

in both systems using the CFU-detection method, which

indicated that the sudden increase in the TPH concentration

did not have an adverse effect on population density.

CFU g�1 dw soil increased 100-fold by the third month of

cultivation (as shown in Fig. S2). At the end of the experi-

ment, the average values for cultivable soil bacteria remained

20–50-fold higher than the initial values. At this point, the

final average TPH concentration in the N (4700 mg kg�1 dry

soil) and P (5000 mg kg�1 dry soil) mesocosms was compar-

able to the TPH content of the soil before simulation of the

rebound (4900 mg kg�1 dw soil).

Phylogenetic affiliations of the dominant 16SrRNA gene sequences

Pollutant rebound-driven changes in soil microbial popula-

tions were assessed by cloning the bands from the DGGE

fingerprints that displayed the most intense shifts on days 90

and 126 (Fig. S3). DNA was extracted from 20 bands from the

N mesocosms and 22 bands from the P mesocosms. RFLP

screening of reamplified inserts distinguished 71 N- and 61 P-

derived clones that contained sequences with unique restric-

tion patterns. After removing chimeras, the sequences were

grouped into 46 phylotypes (defined as having a minimum of

91% similarity) from the N mesocosms and 40 phylotypes

from the P mesocosms. Notably, up to seven different 16S

rRNA gene sequences were retrieved from a single DGGE

band (e.g. band P51 in Figs 3 and S3), each having different

annotations in the NCBI database. In addition, different bands

containing sequences (clones N23a, N24f, N27a and N28e)

assigned to the same bacterial genus were observed, which can

be attributed to the divergence and redundancy of 16S rRNA

gene sequences in genomes with multiple rrn operons (Acinas

et al., 2004; Lee et al., 2009).

Figure 3 presents a comparative phylogenetic analysis of

the predominant bacterial representatives of the N and P soil

microbial communities. In both types of mesocosm, the

priority proteobacterial classes of Alpha-, Beta- and Gamma-

proteobacteria comprised 75–76% of all phylotypes. The

major group of sequences retrieved from the N mesocosm

was affiliated with Betaproteobacteria (34.8%), followed by

Fig. 2. Dynamic changes in microbial

community structure at different stages of

mesocosm cultivation. DGGE analysis-generated

profiles of 16S rRNA gene products amplified

from DNA extracted from control soil and from

bulk soil sampled from the N and P mesocosms.

The DGGE patterns were compared using

GELCOMPAR II v5.1 software. Percent similarities

were calculated using the band-based Dice

coefficient. The numbers at the nodes represent

the degree of similarity.

FEMS Microbiol Ecol 78 (2011) 137–149 c� 2011 Federation of European Microbiological SocietiesPublished by Blackwell Publishing Ltd. All rights reserved

143Functional redundancy in biodegrading microbial communities

Fig. 3. Phylogenetic trees of the predominant bacteria retrieved by 16S rRNA gene V3-V4 region PCR-DGGE from natural remediation (N) and

phytoremediation (P) mesocosm systems on days 90 and 126 of cultivation. The retrieved partial 16S rRNA gene sequences are designated in bold, with

N or P preceding the code of the clone. The trees were rooted with the 16S rRNA gene sequence of Methanoplanus petrolearius DSM 11571. The

percentages of 1000 bootstrap resamplings are shown above or near the relevant nodes. A 5% scale bar indicates the estimated sequence divergence.

FEMS Microbiol Ecol 78 (2011) 137–149c� 2011 Federation of European Microbiological SocietiesPublished by Blackwell Publishing Ltd. All rights reserved

144 O. Korotkevych et al.

Gammaproteobacteria (23.9%) and Alphaproteobacteria

(17.4%). In the P mesocosm, Gammaproteobacteria were the

most abundant, accounting for 37.5% of all phylotypes, and

Betaproteobacteria were the third most abundant bacterial

group (15%).

Degradation potential dynamics duringmesocosm development

Soil DNA was extracted at different phases of mesocosm

cultivation and examined by an absolute qPCR assay using

primers that target four groups of catabolic genes and the

optimized quantification conditions shown in Table 1.

Changes in gene copy number per ng soil DNA were used

to assess shifts in the degradation potential in situ (Fig. 4).

The sudden increase in the TPH concentration from 4900 to

18 500 mg kg�1 dry soil failed to negatively influence the

catabolic potential for the aerobic biodegradation of aro-

matic hydrocarbons in either mesocosm system. Moreover,

all four assessed groups of dioxygenase genes increased in

number and reached maximum levels after the second

month of mesocosm cultivation (Fig. 4). The quantification

of RHDO genes and EXDO-K2 genes showed common

dynamic patterns that can be ascribed to their colocalization

(Brennerova et al., 2009). In N mesocosm soil, the EXDO-K2

and RHDO genes reached maximum copy numbers of

5.57� 104 and 6.76� 104 gene copies ng�1 total DNA, respec-

tively, on day 90. In P mesocosm soil, the highest numbers,

4.19� 104 for EXDO-K2 genes and 8.61� 104 for RHDO

genes, were detected on day 126. At the end of cultivation, both

mesocosm systems showed a greater than threefold increase in

biodegradation potential relative to the untreated soil values

(1.46� 104 for EXDO-K2 and 2.11� 104 for RHDO).

The strongest shifts were observed for EXDO-D2 genes,

which encode proteins with an exceptionally high affinity

for various catecholic substrates (Brennerova et al., 2009).

The newly designed qPCR primers identified up to

8.15� 103 gene copies ng�1 DNA in the N soil on day 90

and 1.83� 104 gene copies ng�1 DNA in the P soil on day 62.

Hence, in a simulation of the rebound effect, biodegradation

potential dependent on EXDO-D2 genes increased 6.5- and

14.6-fold in the N and P microbial populations, respectively.

A similar tendency for the gene copy number to increase

was observed for the least abundant catabolic gene EXDO-

Dbt. The encoded enzyme exhibits broad specificity for

different substrates, including 1,2-dihydroxynaphthalene,

dihydroxybiphenyl, 3-methylcatechol and catechol (Bren-

nerova et al., 2009). On day 62, soil DNA from the P

mesocosms produced a maximum amplification signal of

1.02� 104 copies, which is three times higher than the peak

signal detected on day 32 in the N mesocosms and 8.4 times

higher than the gene concentration in the control soil before

mixing with the LNAPL fraction.

Discussion

In the current study, we adopted a metagenomic approach

to monitoring changes in the degradation potential of soil

microbial communities under conditions that simulate

pollutant rebound. Little is known about (1) the ability of

soil microbiota to resist sudden increases in petroleum

hydrocarbon concentrations, (2) the relationship between

microbial diversity and the functionality of microbial eco-

systems and (3) the influence of plants on the behavior and

biodegradation potential of bulk soil bacteria that are not in

close association with the rhizosphere. Understanding these

relationships is necessary for the design and implementation

of effective and economical remediation and revitalization

measures on sites with a long history of environmental

contamination.

We initiated this investigation by searching for conditions

under which TPH concentrations are several times higher

than the remediation limit (5000 mg kg�1 dw soil), but are

still not lethal to willow trees. To our knowledge, there is a

lack of evidence regarding the pollutant tolerance of plant

species that can be used for in situ remediation. A field study

Fig. 4. The abundance of four groups of catabolic genes during the

development of the N and P mesocosms. Gene copy numbers were

determined by qPCR of soil DNA extracted before the addition of LNAPL

(control) and on days 15, 32, 62, 90 and 126 of mesocosm cultivation.

Each bar represents an average of triplicate PCR runs performed on two

independent DNA extractions (n = 6). Error bars indicate the SD.

FEMS Microbiol Ecol 78 (2011) 137–149 c� 2011 Federation of European Microbiological SocietiesPublished by Blackwell Publishing Ltd. All rights reserved

145Functional redundancy in biodegrading microbial communities

using hydroponically rooted willow and poplar trees found

that weathered gasoline and diesel at a concentration of

4000 mg kg�1 reduced transpiration by 50%, and it was

concluded that at high TPH concentrations, for example,

4 5000 mg kg�1 dw soil, willows cannot be used for phytor-

emediation due to toxic effects (Trapp et al., 2001). Our

pilot mesocosm experiments confirmed the aforementioned

results. Although 6500 mg kg�1 dw soil had a lethal effect on

young willows prerooted in water, the control cuttings

developed larger root systems in a garden substrate; their

vitality was not affected by transfer to contaminated soil

with subsequent increases in TPH up to 18 500 mg kg�1 dw

soil. Thus, their successful adaptation to increasing pollu-

tant concentrations and their development of large roots

that reach over 6 m in depth make willow trees suitable for

the remediation and revitalization of soil contaminated with

petroleum hydrocarbons.

We were specifically interested in comparing microbial

populations and their catabolic properties in the bulk soil of

two systems. In the N mesocosm, the main factor that

defined specific changes in the bacterial community was the

rapid increase in hydrocarbon contamination from 4900 to

18 500 mg kg�1 dw soil. An additional determinant of phy-

logenetic changes in the P mesocosm was the rhizosphere of

pollutant-tolerant willows. Comparative 16S rRNA gene

DGGE analysis of the bacterial communities revealed a clear

change in their structure, reflecting a net adaptation to the

new conditions in the soil (Fig. 2). Distinct microbial

populations developed in the N and P mesocosms during

the 4 months of active vegetative growth under outdoor

conditions. Changes in the microbial community structure

in the N pots were the most intense during the first month

of mesocosm development. The DGGE fingerprints of the

15- and 32-day-old P mesocosms clustered together with

those of the 15- and 32-day-old N mesocosms. In the later

phases of cultivation, stronger shifts were detected for the P

mesocosms, indicating phylogenetic diversification that can

be ascribed to the influence of the willows.

The soil used in our study originated in field G of

Hradcany, which is directly adjacent to the HRB-1 site, well

characterized by biomolecular tools. Phospholipid fatty acid

profiling and 16S rRNA gene library analysis revealed a

highly diverse soil microbial population (Kabelitz et al.,

2009). The latter study, together with a metagenomic

characterization of the diversity and abundance of active

meta-cleavage pathways (Brennerova et al., 2009), demon-

strate that the majority of biodegradation genes are encoded

by Betaproteobacteria species, which represent the largest

phylogenetic group in the soil microbial community.

Although PCR-DGGE does not allow the full complexity of

the system to be determined, we were able to monitor

dynamic changes in the soil microbial communities and

compare differences between the two types of mesocosms.

Bacterial population shifts in the plant-free N mesocosms

were less intense, and the predominant phylotypes were

affiliated mainly with Betaproteobacteria at the final stages of

cultivation. More dynamic changes in the DGGE finger-

prints of the mesocosms with pollutant-tolerant willow trees

indicated selection for phylogenetically diverse bacterial

populations. Sequence analysis confirmed that Gamma-

proteobacteria species comprised the major and predomi-

nant phylotype groups in the P mesocosms. However, the N

and P systems shared eight common phylotypes showing

at least 91% similarity to isolates and uncultured

bacteria previously reported in hydrocarbon-contaminated

environments (Palleroni et al., 2004). We assume that the

representatives affiliated with the Sphingomonadaceae

(Novosphingobium acidiphilum), Acetobacteraceae (Acidocel-

la sp.), Rhodocyclaceae, Alcanivoracaceae, Sinobacteraceae

(Hydrocarboniphaga effusa), Spirochaetaceae and Geobacter-

aceae (Geobacter sp.) families are phylogenetic signatures of

the biodegrading community. A characteristic of the N

mesocosms was 16S rRNA gene sequences showing the

presence of Acidovorax species-related patterns in the soil

microbial community, sequences with similarity to the

gammaproteobacterium Pseudoxanthomonas spadix isolated

from oil-contaminated soil and to the gram-positive Ther-

moanaerobacteriaceae clone found in tar oil-impacted aqui-

fers where BTEX degradation depends mainly on sulfate

reduction (Winderl et al., 2008). Bacteria characteristic of

the soil community in the P mesocosms were phylogeneti-

cally affiliated to (1) Brevundimonas sp. 39, which was

isolated from contaminated river sediment, (2) aerobic

benzene-degrading Hydrogenophaga sp. Rs71, (3) Citrobac-

ter farmeri from offshore oil fields and (4) alicyclic hydro-

carbon-degrading Bacteroides sp. ECP-C1, which was

isolated in sulfate-reducing conditions from a gas conden-

sate-contaminated aquifer (Rios-Hernandez et al., 2003).

Direct evidence that TPH removal is a primary effect of

bioremediation processes occurring in the soil can be

obtained by specifically targeting genes responsible for the

functionality of the autochthonous microbial community.

The majority of culture-independent surveys of catabolic

gene diversity in contaminated environments have used

conserved nucleotide sequences to design primers to detect

the presence, abundance and diversity of catabolic genes that

encode a defined group of enzymes thought to be critical in

the target environment. Our previous study improved upon

this by adopting a metagenomic approach in combination

with functional selection and PCR screening with newly

designed primers to identify key catabolic gene groups in

soil (Brennerova et al., 2009). Detailed enumeration of the

degradation genes characteristic to the locality revealed that

simulated pollutant rebound failed to impair the overall

degradative capacity of the soil microbiota in either the N or

the P mesocosm. Moreover, an increase in gene copy per ng

FEMS Microbiol Ecol 78 (2011) 137–149c� 2011 Federation of European Microbiological SocietiesPublished by Blackwell Publishing Ltd. All rights reserved

146 O. Korotkevych et al.

soil DNA was demonstrated for all EXDO and related

RHDO genes in the microbial community at the contami-

nated site (Fig. 4). It may be that during cultivation of the P

mesocosms, the evolution of taxonomically distinct popula-

tions was concomitant with the remedial response to the

hydrocarbon contamination, resulting in an increase of the

catabolic potential, understood as an increase of gene copies

per genome equivalence at the microbial community. In

addition, the abundance of EXDO-D2 and EXDO-Dbt

genes in the P soil was over two and three times higher than

that in the natural attenuation system. The latter two groups

of EXDO genes encode enzymes that have an exceptionally

high affinity for various catecholic substrates (Brennerova

et al., 2009) and could promote the environmental fitness of

microbial community members. Thus, the relative increases

in EXDO-D2 and EXDO-Dbt gene copy numbers in the P

community would confer the bacterial hosts with enhanced

biodegradative capabilities.

According to contemporary ecological theory, the stabi-

lity of soil-based biotic components is of fundamental

importance to the functionality of ecosystems (Wardle &

Giller, 1996). The ability of ecosystems to resist changes in

environmental conditions is positively correlated with spe-

cies diversity and functional redundancy (Gitay & Wilson,

1996; Griffiths et al., 2000). It was found that the relative

abundances of genes with similar functions are more highly

conserved than are the relative abundances of phylotypes

(Turnbaugh et al., 2009). Turnbaugh and colleagues sug-

gested that the microbial communities exist at the level of

shared genes and different species assemblages converge on

shared core functions provided by distinctive components.

The microbial community of the site under study includes

members of at least 14 bacterial orders (Kabelitz et al., 2009).

Under the selective pressure of a chronic and massive

pollutant spill, clusters of ecologically equivalent specialized

microbial species play overlapping roles in the maintenance

and regulation of the ecosystem. The rapid increase in the

degradation potential of both mesocosm systems in response

to pollutant rebound can be explained by the functional

redundancy of biodegrading microbial communities. Thus,

the increased concentration of organic matter served as a

primary environmental cue that triggered the expansion of

microbial density and catabolic gene abundance. The more

intense microbial shifts that took place in the mesocosms

containing plants could be attributable to the presence of the

rhizosphere as an additional factor supporting the activity of

the biodegrading microbial community.

The current work demonstrates that the taxonomical

affiliations of the microbial communities are of little

descriptive power in terms of catabolic degradation per-

formance in complex in situ conditions. Our results support,

in a culture-independent manner and directly in environ-

mental samples, the observations made in well-studied

cultured bacterial pollutant degraders, where there is no

clear association between the extradiol dioxygenases gene

phylogeny and the taxonomy of the host (Vilchez-Vargas

et al., 2010), indicating once more the importance of

detecting catabolic genes in studies defining the influence

of pollutants in the environment.

Acknowledgements

We wish to thank V. Reimannova for excellent technical

assistance; Jirina Machackova, Stanislava Proksova and J. Jurak

from AECOM CZ for providing soil samples, analytical data

and helpful discussions; and M. Havelkova for graphic work.

We would like to thank Howard Junca and Miroslav Patek for

their suggestions and advice. We also thank the anonymous

reviewers for their constructive comments and criticism. This

work was supported by the Czech Ministry of Education

research programs 2B06156 and 1M06011 and by the Institu-

tional Research Concept # AVOZ50200510.

Statement

GenBank accession numbers: GU560728–GU560730,

GU568253–GU568336.

Authors’contribution

O.K. and J.J. contributed equally to this work.

References

ABI (2006) Creating Standard Curves with Genomic DNA or

Plasmid DNA Templates for Use in Quantitative PCR. Applied

Biosystems, Carlsbad, California.

Acinas SG, Marcelino LA, Klepac-Ceraj V & Polz MF (2004)

Divergence and redundancy of 16S rRNA sequences in

genomes with multiple rrn operons. J Bacteriol 186:

2629–2635.

Brennerova MV, Josefiova J, Brenner V, Pieper DH & Junca H

(2009) Metagenomics reveals diversity and abundance of

meta-cleavage pathways in microbial communities from soil

highly contaminated with jet fuel under air-sparging

bioremediation. Environ Microbiol 11: 2216–2227.

Chablain PA, Zgoda AL, Sarde CO & Truffaut N (2001) Genetic

and molecular organization of the alkylbenzene catabolism

operon in the psychrotrophic strain Pseudomonas putida

01G3. Appl Environ Microb 67: 453–458.

Cole JR, Wang Q, Cardenas E et al. (2009) The Ribosomal

Database Project: improved alignments and new tools for

rRNA analysis. Nucleic Acids Res 37: D141–D145.

Das N & Chandran P (2011) Microbial degradation of petroleum

hydrocarbon contaminants: an overview. Biotechnol Res Int

2011, doi:10.4061/2011/941810.

FEMS Microbiol Ecol 78 (2011) 137–149 c� 2011 Federation of European Microbiological SocietiesPublished by Blackwell Publishing Ltd. All rights reserved

147Functional redundancy in biodegrading microbial communities

Di Gregorio S, Zocca C, Sidler S, Toffanin A, Lizzari D & Vallini G

(2004) Identification of two new sets of genes for

dibenzothiophene transformation in Burkholderia sp. DBT1.

Biodegradation 15: 111–123.

Eltis LD & Bolin JT (1996) Evolutionary relationships among

extradiol dioxygenases. J Bacteriol 178: 5930–5937.

Gitay H & Wilson JBB (1996) Species redundancy: a redundant

concept? J Ecol 84: 121–124.

Glick BR (2010) Using soil bacteria to facilitate

phytoremediation. Biotechnol Adv 28: 367–374.

Griffiths BS, Ritz K, Bardgett RD et al. (2000) Ecosystem response

of pasture soil communities to fumigation-induced microbial

diversity reductions: an examination of the biodiversity–

ecosystem function relationship. Oikos 90: 279–294.

Harayama S, Mermod N, Rekik M, Lehrbach PR & Timmis KN

(1987) Roles of the divergent branches of the meta-cleavage

pathway in the degradation of benzoate and substituted

benzoates. J Bacteriol 169: 558–564.

Harwood CS & Parales RE (1996) The beta-ketoadipate pathway

and the biology of self-identity. Annu Rev Microbiol 50:

553–590.

Huber T, Faulkner G & Hugenholtz P (2004) Bellerophon: a

program to detect chimeric sequences in multiple sequence

alignments. Bioinformatics 20: 2317–2319.

Jimenez JI, Minambres B, Garcıa JL & Dıaz E (2002) Genomic

analysis of the aromatic catabolic pathways from Pseudomonas

putida KT2440. Environ Microbiol 4: 824–841.

Kabelitz N, Machackova J, Imfeld G, Brennerova M, Pieper DH,

Heipieper HJ & Junca H (2009) Enhancement of the microbial

community biomass and diversity during air sparging

bioremediation of a Northern Bohemia soil highly

contaminated with kerosene and BTEX. Appl Microbiol Biot

82: 565–577.

Kukor JJ & Olsen RH (1991) Genetic organization and regulation

of a meta cleavage pathway for catechols produced from

catabolism of toluene, benzene, phenol, and cresols by

Pseudomonas pickettii PKO1. J Bacteriol 173: 4587–4594.

Lee ZM-P, Bussema C III & Schmidt TM (2009) rrnDB:

documenting the number of rRNA and tRNA genes in bacteria

and archaea. Nucleic Acids Res 37: D489–D493.

Machackova J, Wittlingerova Z, Vlk K, Zima J & Linka A (2008)

Comparison of two methods for assessment of in situ jet-fuel

remediation efficiency. Water Air Soil Poll 187: 181–194.

McLeod MP & Eltis LD (2008) Genomic insights into the aerobic

pathways for degradation of organic pollutants. Microbial

Biodegradation. ISBN: 978-1-904455-17-2 (Dıaz E, ed), pp.

1–23. Caister Academic Press, Norfolk, UK.

Muyzer G, de Waal EC & Uitterlinden AG (1993) Profiling of

complex microbial populations by denaturing gradient gel

electrophoresis analysis of polymerase chain reaction-

amplified genes coding for 16S rRNA. Appl Environ Microb 59:

695–700.

Myers RM, Maniatis T, Lerman LS & Ray W (1987) Detection and

localization of single base changes by denaturing gradient gel

electrophoresis. Method Enzymol 155: 501–527.

Nicholas KB, Nicholas HB Jr & Deerfield DW II (1997) GeneDoc:

analysis and visualization of genetic variation.

EMBNET.NEWS 4: 1–4.

Okoh AI (2006) Biodegradation alternative in the cleanup of

petroleum hydrocarbon pollutants. Biotechnol Mol Biol Rev 1:

38–50.

Ovreas L, Forney L, Daae FL & Torsvik V (1997) Distribution of

bacterioplankton in meromictic Lake Saelenvannet, as

determined by denaturing gradient gel electrophoresis of PCR-

amplified gene fragments coding for 16S rRNA. Appl Environ

Microb 63: 3367–3373.

Palleroni NJ, Port AM, Chang H-K & Zylstra GJ (2004)

Hydrocarboniphaga effusa gen. nov., sp. nov., a novel member

of the gamma-proteobacteria active in alkane and aromatic

hydrocarbon degradation. Int J Syst Evol Micr 54: 1203–1207.

Parales RE & Harwood CS (2002) Bacterial chemotaxis to

pollutants and plant-derived aromatic molecules. Curr Opin

Microbiol 5: 266–273.

Pflugmacher U, Averhoff B & Gottschalk G (1996) Cloning,

sequencing, and expression of isopropylbenzene degradation

genes from Pseudomonas sp. strain JR1: identification of

isopropylbenzene dioxygenase that mediates trichloroethene

oxidation. Appl Environ Microb 62: 3967–3977.

Pieper DH (2005) Aerobic degradation of polychlorinated

biphenyls. Appl Microbiol Biot 67: 170–191.

Riesenfeld CS, Goodman RM & Handelsman J (2004)

Uncultured soil bacteria are a reservoir of new antibiotic

resistance genes. Environ Microbiol 6: 981–989.

Rios-Hernandez LA, Gieg LM & Suflita JM (2003)

Biodegradation of an alicyclic hydrocarbon by a sulfate-

reducing enrichment from a gas condensate-contaminated

aquifer. Appl Environ Microb 69: 434–443.

Saitou N & Nei M (1987) The neighbor-joining method: a new

method for reconstructing phylogenetic trees. Mol Biol Evol 4:

406–425.

Shirdam R, Zand AD, Bidhendi GN & Mehrdadi N (2008)

Phytoremediation of hydrocarbon-contaminated soils with

emphasis on the effect of petroleum hydrocarbons on the

growth of plant species. Phytoprotection 89: 21–29.

Sivaganesan M, Haugland RA, Chern EC & Shanks OC (2010)

Improved strategies and optimization of calibration models

for real-time PCR absolute quantification. Water Res 44:

4726–4735.

Suenaga H, Ohnuki T & Miyazaki K (2007) Functional screening

of a metagenomic library for genes involved in microbial

degradation of aromatic compounds. Environ Microbiol 9:

2289–2297.

Switzer C & Kosson DS (2007) Soil vapor extraction performance

in layered vadose zone materials. Vadose Zone J 6: 397–405.

Tamura K, Dudley J, Nei M & Kumar S (2007) MEGA4:

Molecular Evolutionary Genetics Analysis (MEGA) software

version 4.0. Mol Biol Evol 24: 1596–1599.

Tamura T & Hatano K (2001) Phylogenetic analysis of the genus

Actinoplanes and transfer of Actinoplanes minutisporangius

Ruan et al. 1986 and ‘Actinoplanes aurantiacus’ to

FEMS Microbiol Ecol 78 (2011) 137–149c� 2011 Federation of European Microbiological SocietiesPublished by Blackwell Publishing Ltd. All rights reserved

148 O. Korotkevych et al.

Cryptosporangium minutisporangium comb. nov. and

Cryptosporangium aurantiacum sp. nov. Int J Syst Evol Micr 51:

2119–2125.

Tang JC, Wang RG, Niu XW, Wang M, Chu HR & Zhou QX

(2010) Characterisation of the rhizoremediation of

petroleum-contaminated soil: effect of different influencing

factors. Biogeosciences 7: 3961–3969.

Thompson C & Nathanail P (2003) Chemical Analysis of

Contaminated Land. ISBN: 1-84127-334-1. Blackwell

Publishing, Oxford, UK, 305 pp..

Trapp S, Kohler A, Larsen LC, Zambrano KC & Karlson U (2001)

Phytotoxicity of fresh and weathered diesel and gasoline to

willow and poplar trees. J Soils Sediments 1: 71–76.

Turnbaugh PJ, Hamady M, Yatsunenko T et al. (2009) A core gut

microbiome in obese and lean twins. Nature 457: 480–484.

Van Doorn R, Klerks MM, van Gent-Pelzer MPE, Speksnijder

AGCL, Kowalchuk GA & Schoen CD (2009) Accurate

quantification of microorganisms in PCR-inhibiting

environmental DNA extracts by a novel internal amplification

control approach using Biotrove OpenArrays. Appl Environ

Microb 75: 7253–7260.

Vilchez-Vargas R, Junca H & Pieper DH (2010) Metabolic

networks, microbial ecology and ‘omics’ technologies: towards

understanding in situ biodegradation processes. Environ

Microbiol 12: 3089–3104.

Wardle DA & Giller KE (1996) The quest for a contemporary

ecological dimension to soil biology. Soil Biol Biochem 28:

1549–1554.

Winderl C, Anneser B, Griebler C, Meckenstock RU & Lueders T

(2008) Depth-resolved quantification of anaerobic toluene

degraders and aquifer microbial community patterns in

distinct redox zones of a tar oil contaminant plume. Appl

Environ Microb 74: 792–801.

Witzig R, Junca H, Hecht H-J & Pieper DH (2006) Assessment of

toluene/biphenyl dioxygenase gene diversity in benzene-

polluted soils: links between benzene biodegradation and

genes similar to those encoding isopropylbenzene

dioxygenases. Appl Environ Microb 72: 3504–3514.

Yeates C, Holmes AJ & Gillings MR (2000) Novel forms of ring-

hydroxylating dioxygenases are widespread in pristine and

contaminated soils. Environ Microbiol 2: 644–653.

Zalesny RS, Bauer EO, Hall RB, Zalesny JA, Kunzman J, Rog CJ &

Riemenschneider DE (2005) Clonal variation in survival and

growth of hybrid poplar and willow in an in situ trial on soils

heavily contaminated with petroleum hydrocarbons. Int J

Phytoremediat 7: 177–197.

Supporting Information

Additional Supporting Information may be found in the

online version of this article:

Fig. S1. GC chromatograms of LNAPL and kerosene.

Fig. S2. Changes in the concentrations of cultivable bacteria

and TPH in bulk soil from mesocosms N (natural biode-

gradation) and P (phytoremediation model).

Fig. S3. Negative image of DGGE gel used for band excision

and 16S rRNA gene sequencing analysis.

Fig. S4. The fosmid clones used to design the primers

EXDO-D2-F (a) and EXDO-D2-R (b).

Fig. S5. Agarose-gel electrophoresis for verification the

specificity of qPCR products after using the primers target-

ing (a) RHDO, (b) EXDO-K2, (c) EXDO-D2 and (d)

EXDO-Dbt genes (Table 1).

Please note: Wiley-Blackwell is not responsible for the

content or functionality of any supporting materials sup-

plied by the authors. Any queries (other than missing

material) should be directed to the corresponding author

for the article.

FEMS Microbiol Ecol 78 (2011) 137–149 c� 2011 Federation of European Microbiological SocietiesPublished by Blackwell Publishing Ltd. All rights reserved

149Functional redundancy in biodegrading microbial communities