www.fems-microbiology.org

FEMS Microbiology Ecology 51 (2005) 375–388

Occurrence and community composition of fast-growingMycobacterium in soils contaminated with polycyclic

aromatic hydrocarbons

Natalie M. Leys a,b,1, Annemie Ryngaert a, Leen Bastiaens a, Pierre Wattiau c,2,Eva M. Top b,3, Willy Verstraete b, Dirk Springael a,d,*

a Environmental and Process Technology, Flemish Institute for Technological Research (Vito), Boeretang 200, 2400 Mol, Belgiumb Laboratory of Microbial Ecology and Technology, University of Gent (UG), Coupure Links 653, 9000 Gent, Belgium

c Bioengineering Unit, Catholic University of Louvain La Neuve (UCL), Place Croix du Sud 2 – Box 13, 1348 Louvain La Neuve, Belgiumd Laboratory of Soil and Water Management, Catholic University of Leuven, Kasteelpark Arenberg 20, 3001 Heverlee, Belgium

Received 1 February 2004; received in revised form 25 May 2004; accepted 28 September 2004

First published online 28 October 2004

Abstract

Fast-growing mycobacteria are considered essential members of the polycyclic aromatic hydrocarbons (PAH) degrading bacte-

rial community in PAH-contaminated soils. To study the natural role and diversity of the Mycobacterium community in contam-

inated soils, a culture-independent fingerprinting method based on PCR combined with denaturing gradient gel electrophoresis

(DGGE) was developed. New PCR primers were selected which specifically targeted the 16S rRNA genes of fast-growing mycobac-

teria, and single-band DGGE profiles of amplicons were obtained for most Mycobacterium strains tested. Strains belonging to the

same species revealed identical DGGE fingerprints, and in most cases, but not all, these fingerprints were typical for one species,

allowing partial differentiation between species in a Mycobacterium community. Mycobacterium strains inoculated in soil were

detected with a detection limit of 106 CFU g�1 of soil using the new primer set as such, or approximately 102 CFU g�1 in a nested

PCR approach combining eubacterial and the Mycobacterium specific primers. Using the PCR-DGGE method, different species

could be individually recognized in a mixed Mycobacterium community. This approach was used to rapidly assess the Mycobacte-

rium community structure of several PAH-contaminated soils of diverse origin with different overall contamination profiles, pollu-

tion concentrations and chemical-physical soil characteristics. In the non-contaminated soil, most of the recovered 16S rRNA gene

sequence did not match with previous described PAH-degrading Mycobacterium strains. In most PAH-contaminated soils, myco-

bacteria were detected which were closely related to fast-growing species such as Mycobacterium frederiksbergense and Mycobacte-

rium austroafricanum, species that are known to include strains with PAH-degrading capacities. Interestingly, 16S rRNA genes

related to M. tusciae sequences, a Mycobacterium species so far not reported in relation to biodegradation of PAHs, were detected

in all contaminated soils.

� 2004 Federation of European Microbiological Societies. Published by Elsevier B.V. All rights reserved.

Keywords: PAH biodegradation; Mycobacterium; 16S rRNA gene; PCR; DGGE

0168-6496/$22.00 � 2004 Federation of European Microbiological Societies. Published by Elsevier B.V. All rights reserved.

doi:10.1016/j.femsec.2004.09.015

* Corresponding author. Present address: Catholic University of Leuven (KUL), Laboratory for Soil and Water management, Kasteelpark

Arenberg 20, 3001 Heverlee, Belgium. Tel.: +32 16 321604; fax: +32 16 321997.

E-mail address: dirk.springael@agr.kuleuven.ac.be (D. Springael).1 Present address: Belgian Nuclear Research Centre (SCK/CEN), Laboratory of Microbiology, Boeretang, 2400 Mol, Belgium.2 Present address: Vetinary and Agrochemical Research Institute, Department of Bacteriology, Groesetenberg 99, B-1180 Brussels, Belgium.3 Present address: University of Idaho, Department of Biological Sciences, 83844-3051 Moscow, Idaho, USA.

376 N.M. Leys et al. / FEMS Microbiology Ecology 51 (2005) 375–388

1. Introduction

Polycyclic aromatic hydrocarbons (PAHs) are haz-

ardous environmental pollutants that are found in high

concentrations on sites of old gas factories and wood

preservation plants [1]. In spite of the limited bioavaila-bility and poor biodegradability of PAHs, different bac-

teria, often mycobacteria, have been isolated that are

able to use PAHs as sole sources of carbon and energy

[2–7]. So far, all PAH-biodegrading Mycobacterium iso-

lates [2–11] have been placed in the phylogenetic branch

of the �fast-growing mycobacteria�. In the Mycobacte-

rium phylogenetic tree, the �fast-growing mycobacteria�form a coherent line of descent, distinct from the morerecently evolved slow-growers, within which the well-

known mycobacterial pathogens are clustered [12–15].

The �fast-growing mycobacteria� constitute a group of

mycobacteria, mostly of environmental origin, that

are, based on growth, biochemical characteristics and

infectious properties (i.e., mycobacteria of Bio safety Le-

vel 1, growth within 7 days), highly different from path-

ogenic and facultative pathogenic more slowly growingspecies like Mycobacterium avium, Mycobacterium

tuberculosis, Mycobacterium leprae or Mycobacterium

ulcerans (i.e., mycobacteria of Bio safety levels 2 & 3,

growth after more than 7 days) [12–15].

The diversity of fast-growing mycobacteria in the

environment is still largely unknown, but could be of

major interest for bioremediation of PAH-contaminated

soils. Therefore, methods are needed for communityanalysis and monitoring of indigenous and/or inoculated

fast-growing PAH-degrading mycobacteria in soil. For

detection of mycobacteria, direct culture-independent

detection methods are preferable over indirect culture-

dependent techniques because: (i) a large fraction of cells

in soil is difficult to culture or even believed to be uncul-

turable (viable but non-culturable – VBNC state) [16–

18], (ii) the hydrophobic Mycobacterium cells are knownto adhere strongly to organic soil particles, resulting in

difficult recovery [19,20] and (iii)Mycobacterium species,

even �fast-growers�, are relatively slow-growing organ-

isms in comparison to other soil bacteria, which makes

them easily overgrown by other bacteria in culture med-

ia [21]. In addition, molecular methods based on PCR

amplification have been proven to be successful for the

diagnosis of mycobacterial diseases in humans [22–26]and fish [27] and for the identification of environmental

infection sources of Mycobacterium opportunistic path-

ogens such as M. avium and M. ulcerans in plants, water

and soil [28–30]. PCR amplification of variable 16S

rRNA gene-fragments combined with direct analysis

of amplicons by denaturing gradient gel electrophoresis

(DGGE) is a commonly used technique for rapid molec-

ular assessment of microbial communities. In all previ-ous studies [22–29] except one [30], however, the PCR

primers were designed to reveal the presence of slow-

growing mycobacteria in the tested samples and were

never combined with a method for direct community

diversity analysis such as DGGE. Recently, Cheung

and Kinkle [31] designed a Mycobacterium specific pri-

mer pair and analyzed the amplicons using temperature

gradient gel electrophoresis (TGGE) for fingerprinting.However, none of the described primer sets were specific

enough to preferentially target fast-growing mycobacte-

ria, belonging to non-pathogenic and non-opportunistic

species. Moreover, the advantages and limits of the

PCR-TGGE method were not previously addressed.

In this study, we describe the development of a new

set of non-degenerated primers that annealed, as exclu-

sively as possible, to 16S rRNA genes of fast-growingmycobacteria and that amplified a short fragment suita-

ble for DGGE-fingerprinting. The PCR-DGGE method

was theoretically and practically evaluated for the detec-

tion of fast-growing mycobacteria in soil samples and

applied to examine the Mycobacterium community com-

position in PAH-contaminated soils.

2. Materials and methods

2.1. Bacterial strains and growth conditions

In total, 80 different bacterial strains from different or-

ders, suborders and families were used in this study,

including many strains with reported pollutant-degrad-

ing capacities. The following strains of the Mycobacte-

rium genus were used: Mycobacterium aichiense 5545

(DSM44147T), Mycobacterium alvei CR-21 (DSM

44176T), Mycobacterium aurum 358 (DSM43999T),

Mycobacterium vanbaalenii PYR-1 (DSM7251T), Myco-

bacterium austroafricanum E9789 (DSM44191T), M.

austroafricanum-related strains VM0456, VM0450,

VM0451, VM0447, VM0452, VM0573 (Springael et al.,

unpublished), Mycobacterium chlorophenolicum PCP-1(DSM43826T), Mycobacterium diernhoferi SN1418

(DSM43524T), Mycobacterium frederiksbergense FAn9

(DSM44346T), M. frederiksbergense related strains

LB501T, VM0503, VM0531, VM0458, VM0579,

VM0585 (Springael et al., unpublished), Mycobacterium

gilvum strains SM35 (DSM44503T), BB1 (DSM9487),

HE5 (DSM44238),Mycobacterium gilvum related strains

LB307T, VM0505, VM0504, VM0552, VM 0442,LB208, VM0583 (Springael et al., unpublished), Myco-

bacterium hodleri EM12 (DSM44183T), Mycobacterium

komossense Ko2 (DSM44078T), Mycobacterium neoau-

rum 3503 (DSM44074T), Mycobacterium parafortuitum

311 (DSM43528T), Mycobacterium peregrinum 6020

(DSM43271T), Mycobacterium petroleophilum RF002

(Lloyd-Jones et al., unpublished), Mycobacterium vac-

cae-related strains VM0587, VM0588 (Springael et al.,unpublished), and Mycobacterium sp. strains WF2 [9]

and GP1 (Lloyd-Jones et al., unpublished). Besides

N.M. Leys et al. / FEMS Microbiology Ecology 51 (2005) 375–388 377

Mycobacterium strains, also other Coryneform bacteria

were used, i.e., Dietzia maris VM0283 [59], Dietzia maris

IMV195 (DSM43627T), Corynebacterium glutamicum

2247 (DSM20411T), Tsukumurella paurometabola

(DSM 20162T), Nocardia asteroides N3 [32], Nocardia

coeliaca AB.4.1.b (DSM44595T), Pseudonocardia hydro-

carbonoxydans (DSM43281T), Rhodococcus erythropolis

ICPB4417 (DSM43066T), Gordonia hydrophobica 1610/

1b (DSM44015T), and Gordonia amarae Se 6

(DSM43392T). Other Actinomycetales used, not belong-

ing to the Coryneform bacteria, were Actinomyces sp.

A1008 [32], Actinosynnema mirum 101 (DSM43827T),

Arthrobacter sufureus 8-3 (DSM20167T),Kineospora aur-

antiaca A/10312 (DSM43858T), Microbispora rosea

IMRU37485 (ATCC21946), Micromonospora chalcea

A0919, A2868, A2894 [32], Planomonospora parontos-

pora B-987 (DSM43869T), Promicromonospora citrea

INMI 18 (DSM43110T), Streptomyces albus A0818,

A1893, A2198, A3986 [32], Streptomyces aureofaciens

A-377 (DSM40127T), Streptomyces rutgersensis BJ-608

(DSM40830T), Streptomyces phaeofaciens T-23 (DSM

40367T), and Streptosporangium album A0958 [32].Strains used belonging to the Proteobacteria phylum

were Agrobacterium luteum A61 (DSM5889T), Brevundi-

monas diminuta 342 (DSM7234T), Sphingomonas chloro-

phenolica DSM43869T, Ralstonia metallidurans CH34

(DSM2839T), Burkholderia sp. JS150 (DSM8530T),

Aeromonas enteropelogenes J11 (DSM6394T), Acinetob-

acter calcoaceticus 46 (DSM30006T), Pseudomonas put-

ida DSM8368T, Desulfobacter latus AcRS2 (DSM3381T), Desulfonema magnum 4be13 (DSM2077T), and

Desulfolobus rhabdoformis M16 (DSM8777T). In addi-

tion, 1 strain from the Flavobacteria phylum, i.e., Flavo-

bacterium resinovorum (ATCC33545T), was included.

Most strains were obtained from the DSMZ culture col-

lection (DSMZ, Braunschweig, Germany), provided by

other laboratories [5,9,32,33] or selected from the in-

house collection of PAH-degrading strains previouslyisolated from PAH-contaminated soil (Springael et al.,

unpublished). For DNA-extraction purposes, strains

other than mycobacteria were cultivated in 869-broth

[34], while Mycobacterium strains were cultivated in

Middelbrook 7H9 Broth (DIFCO, Kansas City, USA).

For inoculation purposes, Mycobacterium strains were

cultivated in a phosphate-buffered minimal liquid med-

ium as described by Wick et al. [35], containing 2 g l�1

of anthracene or pyrene crystals (ACROS Organics,

Fisher Scientific, Boston, USA) floating in the medium

as the sole carbon and energy source. All cultures were

grown in the dark on an orbital horizontal shaker at

200 rpm at a constant temperature of 30 �C.

2.2. Soils used in this study

Soil samples were taken from different anthropo-

genic PAH-contaminated sites (Table 1). The soil tex-

ture, pH, total carbon content (TC), total inorganic

carbon (TIC) content and total organic carbon (TOC)

content (TC – TIC) of each soil sample were deter-

mined using standard methods (DIN Method 38414,

S4; ISO-CEN ENMethod 1484). The soils were chemi-

cally analyzed for the 16 PAHs legislated by the USEnvironmental Protection Agency [36]. PAHs were ex-

tracted through an accelerated solvent extraction (ASE

200 Accelerated Solvent Extractor, Dionex Corp., Sun-

nyval, CA) using a standard approach (EPA Method

3545). ASE-extracts were purified over an internal silica

phase in the extraction cell (i.e., in thimble clean up)

followed by an alumina column. For quantification,

capillary gas chromatography (GC, MFC 500, CarloErba Instruments, Milan, Italy) was used, coupled to

a mass-spectrophotometric detector operated in the se-

lected ion-monitoring mode (MS, QMD 100, Fisons

Instruments, Loughborough, UK) (EPA Method

8270). The total concentration of mineral oil present

in the soil sample was determined after an ultrasonic

tetrachloroethene extraction followed by a FLORISIL

clean up (FLORISIL, US Silica Company, BerkeleySprings, USA), using an infrared quantification at

2925, 2958 and 3030 cm�1 (NEN Method 5733).

2.3. Design of a 16S rRNA gene primer set specific for

fast-growing mycobacteria

Primer sequences were selected from a multiple align-

ment constructed with the Bionumerics software (Ver-sion 2.50, Applied Maths, Kortrijk, Belgium) of

approximately 200 16S rRNA genes (GenBank, NCBI)

[37], representing approximately 100 fast- and 100

slow-growing Mycobacterium species. The alignment

was further analyzed with the PLOTCON program

(Version 1.9.1, EMBOSS) [38] to identify conserved

and variable gene regions. Based on the alignment,

new Mycobacterium specific primers were selected ingene regions that are conserved within the group of

fast-growing Mycobacterium species, but as variable as

possible within slow-growing mycobacteria. In addition,

for optimal species differentiation duringDGGE-analysis

of the PCR-products (see below), the primers were se-

lected so that they amplified a region between 200 and

600 bp long with high variability. The selectivity of the

selected primers was evaluated by visual analysis ofthe primer region within the constructed alignment of

Mycobacterium rrn genes, by the Sequence Match pro-

gram (RDP II) [39] and by the Advanced Blast Search

program (GenBank, NCBI) [37,40]. The best primer

combination consisted of the forward primer MYCO66f

(5 0-CATGCAAGTCGAACGGAAA-3 0, Escherichia

coli position 66–84) and the reverse primer MYCO600r

(5 0-TGTGAGTTTTCACGAACA-3 0, E. coli position600–583). A 40-basepair long GC-clamp [41,42] was at-

tached to the 5 0 end of the MYCO600r primer for

Table 1

Soil samples used in this study

Soil Origin Soil type pH TOC (%) PAH conc.

(mg kg�1)

Oil conc.

(mg kg�1)

DNA conc.a

(lg g�1)

MYCOb Nestedc

S587 Corn field (Belgium) Sand 5.5 2.15 0.289 <50 31.00 + ND

S588 Horse pasture (Belgium) Sand 6.0 2.46 0.391 <50 18.00 + ND

S585 Pine tree forest (Belgium) Sand 5.8 3.19 0.673 <50 31.75 + ND

S589 Ditch in agricultural area (Belgium) Sand 5.8 4.24 0.721 <50 49.50 + ND

S592 Vegetable garden (Belgium) sand 7.0 3.16 1.011 <50 38.25 + ND

S584 Compost heap (Belgium) Sand 7.3 7.04 1.063 <50 27.75 + ND

S591 Non-paved land road (Belgium) Sand 9.0 0.76 3.357 <50 6.25 NP ND

TB3 Coal gasification plant (Belgium) Sand 8.23 1.52 14 <50 2.65 + +

K3840 Gasoline station site (Denmark) Sand 8.20 0.50 20 98 2.75 + +

B101 Coal gasification plant (Belgium) Sand 7.00 2.63 107 70 27.25 + +

E6068 Gasoline station site (Denmark) Sand 7.96 9.94 258 300 5.40 + +

TM Coal gasification plant (Belgium) Sand 8.00 3.85 506 4600 4.75 + +

Barl Coal gasification plant (Germany) Gravel 8.90 4.63 1029 109 6.15 NP NP

AndE Railway station site (Spain) Clay 8.10 2.35 3022 2700 3.40 NP +

a DNA recovery per g soil, mean value of 2 parallel extractions of 1 g of soil.b Result of direct PCR with Mycobacterium specific primers MYCO66f and GC40-MYCO600r on soil DNA extract: +, detectable PCR product;

NP, no detectable PCR product and ND, not determined.c Result of nested PCR with eubacterial primers 27f and 1492r followed by Mycobacterium specific primers MYCO66f and GC40-MYCO600r on

soil DNA extract: +, detectable PCR product; NP, no detectable PCR product and ND, not determined.

378 N.M. Leys et al. / FEMS Microbiology Ecology 51 (2005) 375–388

DGGE analysis of the Mycobacterium amplicons. This

new primer couple MYCO66f and GC40-MYCO600r

amplified a 538 bp sequence of the 16S rRNA gene,

resulting in a PCR-product of 578 bp.

2.4. DNA-extraction

Genomic DNA from pure bacterial cultures was ob-

tained as described by Belisle et al. [43]. The DNA

recovery was approximately 2.7–27.3 lg DNA g�1 soil.

For PCR purposes, the DNA-concentration was ad-

justed to a final concentration of 100 ng ll�1. Forfast-growing mycobacteria, 100 ng of template DNA

corresponds to approximately 1.2–1.9 · 107 cell equiva-

lents of genomic DNA and 2.4–3.8 · 107 copies of

PCR targets, assuming a genomic molecular weight of

3.13–5.20 · 109 Da per cell and two 16S rRNA gene

copies per genome [44]. DNA was extracted from 1 g

soil, using a protocol described by Boon et al. [45]. After

purification over a Wizard� column (Promega Corpora-tion, Madison, USA), the DNA concentration in the 50-

ll soil extract was measured spectrophotometrically. To

assure that the soil DNA was of PCR quality, dilution

series of soil DNA extracts were tested by PCR with

the universal eubacterial 16S rRNA gene primer pair

GC40-63f and 518r as previously described [46].

2.5. PCR amplification of pure strain and soil DNA

The PCR protocol used with the MYCO66f and MY-

CO600r primer pair consisted of a short denaturation of

15 s at 95 �C, followed by 50 cycles of denaturation for

3 s at 94 �C, annealing for 10 s at 50 �C, elongation for

30 s at 74 �C, and a final extension for 2 min at 74 �C.

Primers designed by Cheung and Kinkle (MycF and

MycR) were used for PCR amplification as described

[31]. PCR was performed on Biometra (Gottingen, Ger-

many) or Perkin Elmer (Connecticut, USA) thermal cy-clers. PCR mixtures contained 100 ng of pure strain

DNA or dilutions of soil DNA as templates, 1 U Taq

polymerase, 25 pmol of each primer, 10 nmol of each

dNTP and 1 · PCR buffer in a final volume of 50 ll. Inthe nested PCR approach, the eubacterial primers 27fC

(5 0 AGAGTTTGATCCTGGCTCAG 3 0) and 1492rC

(5 0 TACGGCTACCTTGTTTACGACTT 3 0) were used

in the first amplification round as described elsewhere[47], and 1 ll of the resulting unpurified PCR-product

was used as a template in the second PCR with the MY-

CO66f and MYCO600r primers. All primers were syn-

thesized by Westburg (Westburg BV, Leusden, The

Netherlands). The Taq polymerase, dNTPs and PCR

buffer were purchased from TaKaRa (TaKaRa Shuzo

Co., Ltd., Biomedical Group, Kyoto, Japan).

2.6. DGGE analysis

The PCR-products were examined on 1.5% agarose

gels (MetaPhor, BioWhittaker, Labtrade Inc., Miami,

FL) and directly used for DGGE analysis on polyacryl-

amide gels as previously described [48]. Optimal dena-

turing conditions were defined based on the theoretical

melting temperatures of amplification fragments, calcu-lated with the Melt Analysis Software (Version 1.0.1,

INGENY International BV, Goes, The Netherlands).

A 6% polyacrylamide gel with a denaturing gradient of

40–75% (100% denaturant gels contain 7 M urea and

40% formamide) was used for DGGE. Electrophoresis

was performed at a constant voltage of 130 V for 16.6

N.M. Leys et al. / FEMS Microbiology Ecology 51 (2005) 375–388 379

h in 1 · TAE running buffer at 60 �C in the DGGE-ma-

chine (INGENYphorU-2, INGENY International BV,

Goes, The Netherlands). After electrophoresis, the gels

were stained with 1 · SYBR Gold nucleic acid gel stain

(Molecular Probes Europe BV, Leiden, The Nether-

lands) and photographed under UV light, using a Phar-macia digital camera system (Image Master VDS,

Pharmacia Biotech, Cambridge, England) with Liscap

Image Capture software (Version 1.0, Pharmacia Bio-

tech, Cambridge, England). Photofiles were processed

and analyzed using Bionumerics software (Version

2.50, Applied Maths, Kortrijk, Belgium).

2.7. Sensitivity of the PCR-DGGE method

To study the sensitivity of the PCR-DGGE method,

a known amount of viable Mycobacterium cells was

added to white sand (a model soil matrix) or other soil

samples (Table 1) at different final cell densities (i.e., a

10-fold dilution series of approximately 108–101 cells

g�1), prior to DNA-extraction. Cells were harvested

from liquid cultures, washed twice and added in 100 llaqueous suspensions to 1 g of soil. One, two or three dif-

ferent Mycobacterium strains (LB501T, VM0552 and

DSM43524T) were separately or simultaneously added

in different cell densities to assess the effect of cell ratios

on the detection sensitivity for each single strain within a

Mycobacterium population. In parallel, DNA was ex-

tracted from serial dilutions of the cell cultures. The dif-

ferent DNA-extracts or dilutions thereof weresubsequently used as templates for direct PCR or the

nested PCR as described above.

2.8. Sequence analysis of amplified 16S rRNA gene

fragments

PCR products of Mycobacterium 16S rRNA genes

were cloned into plasmid vector pCR2.1-TOPO, usingthe TOPO Cloning Kit (N.V. Invitrogen SA, Merelbeke,

Belgium) as described by the manufacturer. DGGE pat-

terns of cloned fragments were compared with the fin-

gerprints of the parent soil Mycobacterium community

to identify which signals from the community fingerprint

were cloned. A 500-bp long fragment was sequenced

(Westburg BV, Leusden, The Netherlands) from a selec-

tion of cloned inserts with different DGGE-patterns.The sequences were analyzed with the �Chimera Check�program (RDPII) [39] to detect possible chimeras and

with the �Blast Search� program (GenBank, NCBI)

[40]. Cloned sequences were imported into an alignment

of Mycobacterium 16S rRNA genes and edited manually

to remove nucleotide positions of ambiguous alignment

and gaps. Sequence similarities were calculated over the

16S rRNA gene fragment between the MYCO-primers,corrected using Kimura�s two-parameter algorithm to

compensate for multiple nucleotide exchange and used

to construct a distance-based evolutionary tree with

the Neighbor-Joining algorithm [49]. The topography

of the branching order within the dendrogram was eval-

uated using the Maximum-Likelihood [50] and the Max-

imum-Parsimony [51] character-based algorithms in

parallel combined with bootstrap analysis with a roundof 500 reassemblings. An out-group of the closely re-

lated genera Rhodococcus and Dietzia was included to

root the tree.

2.9. Nucleotide sequence accession numbers

The partial 16S rRNA gene sequences of Mycobacte-

rium clones reported in this study are available fromGenBank under Accession Nos. AY148197 to

AY148217.

3. Results

3.1. Design of specific 16S rRNA gene primers for PCR

detection of fast-growing PAH-degrading mycobacteria

A new specific primer set was designed to amplify the

16S rRNA genes of fast-growingMycobacterium species.

Based on an alignment of approximately 200 sequences,

the 16S rRNA genes of fast- and slow-growing mycobac-

teria appeared highly conserved in comparison to other

bacteria, i.e., only a few short well-defined regions within

the gene were found to be highly variable (data notshown). A minimum similarity of 94% over the total

length of the 16S RNA gene was found for all fast-grow-

ing mycobacteria. The best possible primer combination

was selected from the alignment taking into account the

amplicon length and amplicon sequence variability and

the Blast and sequence match results of both primers.

The sequence of the forward primer MYCO66f (E. coli

locations 66–84) was conserved in 300 rrn gene sequencesof mainly fast-, but also of some slow-growing mycobac-

teria of the approximately 900Mycobacterium sequences

currently available in the GenBank database (NCBI)

(Table 2). The MYCO66f primer also aligned 100% with

Corynebacterium, Phytoplasma, Gordonia and Propioni-

bacterium 16S rRNA genes. The sequence of primer

MYCO600r (E. coli locations 600–583), however, was

100% conserved in only 165 sequences of exclusivelyfast-growing environmental Mycobacterium strains,

including all known PAH-degrading species (Table 2).

It clearly differed with respect to most other 16S rRNA

gene sequences from slow-growing mycobacteria and

non-mycobacterium strains with 1 to 7 mismatches of

the 18 bp long primer region (Table 2). In comparison

with the Mycobacterium specific primers MycF and

MycR described by Cheung and Kinkle [31], forwardprimers MYCO66f (this study) and MycF [31] were sim-

ilarly conserved in mycobacterial 16S rRNA genes, but

Table 2

DNA-sequence homology between the Mycobacterium genus-specific primers and the 16S rRNA gene sequence of some reference mycobacteria

Organism 16S rRNA genea Primersb

MYCO66f (E. coli 66–84)

50-C A T G C A A G TC G A A C G G A A A-30MYCO600r (E. coli 600–583)

5 0-T G T G A G T T T TC A C G A A C A-30

Fast-growing Mycobacterium species

M. aichiense DSM44147T X55598 - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -

M. alvei DSM44176T AF023664 - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -

M. aurum DSM43999T X55595 - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -

M. austroafricanum DSM44191T X93182 - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -

M. chlorophenolicum DSM43826T X79094 - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -

M. diernhoferi DSM43524T X55593 - - - - - - - - - - - - - - - - - - - - - - - - - A - - - - - - - - - - -

M. frederiksbergense DSM44346T AJ276274 - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -

M. gilvum DSM44503T X81996 - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -

M. hodleri DSM44183T X93184 - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -

M. komossense DSM44078T X55591 - - - - - - - - - - - - - - - - T - - - - - - - - - - - - - - - - - - - -

M. neoaurum DSM44074T M29564 region unsequenced - - - - - - - - - - - - - - - - - -

M. parafortuitum DSM43528T X93183 - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -

M. peregrinum DSM43271T AF058712 - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -

M. petroleophilum RF002 U90876 - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -

M. vaccae VM0587 AF44638 - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -

M. sp. DSM44238 AJ012738 - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -

M. sp. WF2 U90877 - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -

Slow-growing Mycobacterium species

M. gordonae DSM44183T X52923 - - - - - - - - - - - - - - - - T - - C - - - - - A - - - - - - - - - - -

M. genavense ATCC51233 X60070 - - - - - - - - - - - - - - - - - - - C C C C C G - A - - - - - - - - - -

M. branderi ATCC51789T X82234 - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - G - - -

M. leprae X55587 - - - - - - - - - - - - - - - - - - - C - - - - - A - - - - - - - - - N N

M. tuberculosis H37Rv NC_000962 - - - - - - - - - - - - - - - - - - - C - - - - - A - - - - - - - - - - -

M. ulcerans X88926 - - - - - - - - - - - - - - - - - - - C - - - - - A - - - - - - - - - - -

Non-Mycobacterium bacteria

Gordonia terrae AB111113 AF397061 - - - - - - - - - - - - - - - - - - - - - C - - A - - - C A C A - - C G -

Corynebacterium sp. ATCC43833 AF262996 - - - - - - - - - - - - - - - - - - - C C G - T A - - - C A C A - - C G -

Phytoplasma sp. AF500334 - - - - - - - - - - - - - - - - - - - - T G A - A - - A A A C T A G - G -

Rhodococcus globerulus NCIB12315 X77779 - - - - - - - - - - - - - - - - T - - - - - - - - - - - - - A - - - - G -

Dietzia maris DSM43672 X79290 - - - - - - - - - - - - G - - - T - - C - A - G A - - - - - A G - - - G -

a Accession No. of 16S rRNA gene sequence in the GenBank (GenBank, NCBI, Rockville, Pike Bethesda, USA).b N = A or T or G or C, Dashes indicate homologous sequences.

380

N.M

.Leyset

al./FEMSMicro

biologyEcology51(2005)375–388

Fig. 1. Mycobacterium species differentiation by DGGE analysis of

Mycobacterium DNA fragments amplified with the Mycobacterium

genus-specific primer pair MYCO66f and GC40-MYCO600r. Lanes: 1,

M. vaccae VM0587; 2, M. diernhoferi DSM43524T; 3, M. austroafri-

canum DSM7251; 4, M. austroafricanum DSM44191T; 5, M. aurum

DSM43999T; 6, M. chlorophenolicum DSM43826T; 7, Mycobacterium

sp. DSM44238; 8, M. hodleri DSM44183T; 9, M. gilvum DSM44503T;

10, M. petroleophilum RF002; 11, Mycobacterium sp. WF2; 12,

Mycobacterium sp. GP1; 13, M. peregrinum DSM43271T; 14, Myco-

bacterium sp. VM0585; 15, Mycobacterium sp. VM0579; 16, M.

frederiksbergense DSM44346T; 17, M. neoaurum DSM44074T; 18, M.

alvei DSM44176T; 19, M. komossense DSM44078T; 20, M. parafortu-

itum DSM43528T; 21, M. aichiense DSM44147T. DGGE fingerprints

of strains were compared by Bionumerics software based on co-

running standard (not shown). The symbol * indicates multiple band

DGGE patterns for single strains with arrows indicating the addition-

ally cloned and sequenced bands.

N.M. Leys et al. / FEMS Microbiology Ecology 51 (2005) 375–388 381

reverse primer MYCO600r (this study) was far more spe-

cific for fast-growing mycobacteria than the MycR

primer [31]. The primer couple MYCO66f and MY-

CO600r produced products of the appropriate size with

the DNA obtained from the 40 tested Mycobacterium

strains representing different fast-growing environmentaland PAH-degrading species, as listed in Table 1. Due to

the risks associated with most slow-growing Mycobacte-

rium species classified as �Biosafety Level 2 & 3 0-agents

[52], only fast-growing reference strains were tested.

No PCR-products were obtained with DNA from strains

belonging to other related genera such as Actinomyces,

Arthrobacter, Dietzia, Corynebacterium, Nocardia,

Sphingomonas, Burkolderia, Acinetobacter, Desulfob-

acter and more.

3.2. Differentiation of fast-growing Mycobacterium

species by DGGE-analysis

In order to examine the potential of a PCR-DGGE

method based on the new primer set to differentiate be-

tween species within a Mycobacterium community, purestrain DGGE-patterns of a variety of fast-growing

mycobacteria were compared. Different Mycobacterium

isolates belonging to the same species showed identical

DGGE-patterns. This was observed for 8 M. austroafri-

canum-related strains tested, 7 M. frederiksbergense-

related strains tested and 10 M. gilvum related strains

tested (data not shown). Usually, different species

showed different DGGE fingerprints (Fig. 1). For exam-

ple, the PCR-products obtained for the M. frederiksber-

gense-related strains clearly migrated differently from

the products from M. austroafricanum or M. gilvum re-

lated strains. However, in some cases the differences in

migration between different species were minor, suchas seen for M. spaghni,M. hodleri and M. gilvum strains.

The species-specific DGGE fingerprints usually dis-

played one single band. However, some strains revealed

extra bands in comparison to other members of the

same species (Fig. 1). For example, strains DSM7251

and VM0450 both showed a similar pattern with 5

bands, strains VM0579, VM0531 and VM0503 dis-

played 3 bands and VM0585 even 10 bands.

3.3. Sensitivity of the Mycobacterium-specific PCR-

DGGE protocol

To examine the sensitivity of the PCR-DGGE proto-

col for the detection of Mycobacterium strains in soil, a

known decreasing amount of Mycobacterium sp.

LB501T cells was added prior to DNA-extraction, bothto white sand as a model soil matrix and to different

PAH-contaminated soil samples (Table 1). For some

contaminated soils there was a clear inhibitory effect

of the soil matrix on the PCR amplification, so the soil

DNA template was diluted 1:10 or 1:100 prior to PCR.

In a PCR with primer pair MYCO66f and GC40-MY-

CO600r, LB501T cells could generally be detected until

a minimum cell concentration of 106–108 cells per gramof soil, depending on the soil used. Similar results were

also obtained when two or three different Mycobacte-

rium strains (LB501T and VM0552 or LB501T,

VM0552 and DSM43524T) were simultaneously added

to white sand in equal cell concentrations (ratios 1:1

or 1:1:1 ratio), i.e., all strains were detected equally well

until a concentration of 106–107 CFU g�1 soil (Fig.

2(a)). To assess the impact of unequal cell concentrationratios on the detection sensitivity, different concentra-

tions of VM0552 or VM0552 and DSM43524T (1:1) cells

were added to white sand in the presence of a constant

LB501T concentration of approximately 108 CFU g�1.

Decreasing cell amounts of VM0552 and DSM43524T

could be detected in the presence of 108 CFU g�1

LB501T cells, but only if the cell concentration was

higher than 106 CFU g�1 (Fig. 2(b)). Attempts to signif-icantly improve the detection limit by optimizing the

DNA extraction and purification protocol or by reduc-

ing the length of the GC-clamp were not successful (data

not shown). Similar detection limits were obtained with

DNA extracts from dilution series of pure cultures and

with dilution series of a pure culture DNA extract (data

not shown). However, via a nested PCR approach, con-

sisting of a first PCR with eubacterial primers followedby a second PCR with the MYCO-primers using the

product of the first PCR as template, the detection limit

Fig. 2. Detection efficiency of the PCR-(DGGE) method using the Mycobacterium genus-specific primer pair MYCO66f and GC40-MYCO600r in a

direct or nested PCR-approach. (a) PCR-DGGE detection of the simultaneously added M. frederiksbergense LB501T, M. gilvum VM0552 and M.

diernhoferi DSM43524T at cell concentrations of approximately 108, 107, 106, 105, 104, 103 and 102 CFU g�1 in white sand, using a direct PCR

approach. The symbol * indicates the detection limit on the figure. (b) PCR-DGGE detection of M. gilvum VM0552 and M. diernhoferi DSM43524T

added in declining cell concentrations together with a constant cell density of M. frederiksbergense LB501T of 108 CFU g�1 using the direct PCR

approach. The symbol * indicates the detection limit on the figure. (c) PCR detection of M. frederiksbergense LB501T at cell concentrations of

approximately 109, 108, 107, 106, 105, 104, 103 and 102 CFU g�1 in white sand using the nested PCR approach. In comparison to a marker (M), the

lower arrow indicates the 578 bp Mycobacterium 16Sr RNA gene amplicon, the upper arrow indicates a 1465 bp 16S rRNA gene fragment, a residue

originating from the first PCR with the 27fC/1492rC primer pair. No signal was obtained without added cells.

382 N.M. Leys et al. / FEMS Microbiology Ecology 51 (2005) 375–388

could be lowered to approximately 102 cells per gram of

soil (Fig. 2(c)). Using this nested PCR protocol, a min-

imum of 50 fg of pure Mycobacterium chromosomal

DNA could be detected in pure culture (data notshown).

3.4. Composition analysis of the fast-growing

Mycobacterium community in soil

The MYCO-primer PCR-DGGE method was used to

assess the Mycobacterium community in a set of PAH-

contaminated soil samples with different contaminationrecords and in uncontaminated soils (Table 1). Indige-

nous mycobacteria could be detected in 6 of the 7

uncontaminated soils tested and in 6 of the 7 PAH-

contaminated soils tested (Fig. 3). PCR-DGGE finger-

printing with eubacterial primers revealed the presence

of a heterogeneous bacterial soil community in soils neg-

ative by PCR with the MYCO-primer set. Moreover,DNA-extracts from parallel samples containing added

Mycobacterium cells were amplified well with the

MYCO-primer set, omitting PCR inhibition as a possi-

ble cause for the negative PCR results obtained with

the MYCO-primer set for the unseeded samples. The

DGGE fingerprints of the Mycobacterium community

in the positive soil samples were complex, comprising

several bands for each soil (Fig. 3(a)). The 16S rRNAgene PCR products from 1 uncontaminated and 3 con-

taminated samples were randomly cloned, and clones

representing different bands from one soil fingerprint

Table 3

Cloned 16S rRNA gene sequences retrieved from PAH-polluted soil samples

Origin Clones Accession No. Nearest match in blast analysis (Accession No.) Similarity (%)

Soil S589 S589/1 AY725804 M. alvei DSM44176T (AF023664) 97

S589/2 AY725805 M. moriokaense DSM44221T (AJ429044) 94

S589/3 AY725806 M. lacus (AF406783) 97

S589/4 AY725807 M. lacus (AF406783) 97

S589/5 AY725808 M. anthracenicuma (Y15709) 98

S589/6 AY725809 M. margeritense 1336 (AJ011335) 95

S589/7 AY725810 M. hodleri DSM44183T (X93184) 97

Soil K3840 K3840/1 AY148216 Mycobacterium sp. M0183 (AF055332) 99

K3840/2 AY148200 Mycobacterium sp. HXN1500a (AJ457057) 98

K3840/3 AY148207 M. tusciae DSM44338T (AF058299) 95

K3840/4 AY148201 M. frederiksbergense LB501Ta (AJ245702) 98

K3840/6 AY148214 M. austroafricanum DSM44191Ta (X93182) 98

K3840/7 AY148197 M. gadium ATCC27726 (X55594) 98

K3840/8 AY148210 M. tusciae DSM44338T (AF058299) 99

Soil B101 B101/1 AY148202 Mycobacterium sp. JKD2385 (AF221088) 98

B101/2 AY148208 M. isoniacini INA-I (X80768) 97

B101/3 AY148198 M. holsaticum 1406 (AJ310467) 97

B101/4 AY148204 M. tusciae DSM44338T (AF058299) 96

B101/5 AY148212 M. septicum HX1900 (AJ457056) 99

B101/6 AY148217 M. petroleophilum RF002a (U90876) 98

B101/7 AY148215 Mycobacterium sp. WF2a (U90877) 97

Soil TM TM/1 AY148211 M. tusciae DSM44338T (AF058299) 99

TM/2 AY148209 M. tusciae DSM44338T (AF058299) 99

TM/3 AY148203 M. tusciae DSM44338T (AF058299) 97

TM/4 AY148205 M. tusciae DSM44338T (AF058299) 97

TM/5 AY148206 M. tusciae DSM44338T (AF058299) 98

TM/6 AY148196 M. moriokaense MCR07 (AF058299) 97

TM/7 AY148213 M. septicum HX1900 (AJ457056) 98

TM/8 AY148199 M. tusciae DSM44338T (AF130308) 99

a Known oil- or PAH-degrading bacterium.

Fig. 3. PCR-DGGE fingerprint of indigenous Mycobacterium cells in soil samples using the Mycobacterium-specific primers MYCO66f and GC40-

MYCO600r (a) and phylogenetic analysis of detected 16S rRNA gene sequences (b). (a) DGGE-fingerprints of the Mycobacterium population in

different soils. Arrows indicate cloned �bands� within the soil fingerprint, based on the comparison of migration profiles of pure clones and the soil

profile. (b) Phylogenetic positioning of Mycobacterium 16S rRNA gene sequences detected in soil within the Mycobacterium genus. Arrows indicate

cloned Mycobacterium 16S rRNA gene sequences retrieved from the different polluted soils. The evolutionary tree was generated by the Neighbour-

Joining method, based on Kimara 2-parameter corrected similarity percentages of 538 bp 16S rRNA gene fragments between the MYCO primers,

and branching orders were evaluated using the Maximum-Parsimony algorithm. The topology was also evaluated by bootstrap analysis (500

reassemblings) and percentages of bootstrap support are indicated at the branch points, with values above 70% indicating reliable branches. An out-

group of the closely related genera Rhodococcus and Dietzia was included to root the tree. The bar at the top indicates the estimated evolutionary

distance, i.e., 1% indicating an average of 1 nucleotide substitution at any nucleotide position per 100 nucleotide positions. The evolutionary distance

between two strains is the sum of the branch lengths between them.

N.M. Leys et al. / FEMS Microbiology Ecology 51 (2005) 375–388 383

Fig. 3 (continued)

384 N.M. Leys et al. / FEMS Microbiology Ecology 51 (2005) 375–388

were sequenced. All sequences exhibited high levels of

similarity to 16S rRNA gene sequences of Mycobacte-

rium strains (Table 3) and could be placed within the

phylogenetic group of fast-growing Mycobacterium spe-

cies (Fig. 3(b)), confirming the specificity of the primer

set. In the uncontaminated soil mainly sequences most

similar to non-PAH-degrading mycobacteria were de-

tected. Sequences that were closely related to 16S rRNA

gene sequences of known PAH- and oil-degrading bacte-

ria belonging to the species M. frederiksbergense,

N.M. Leys et al. / FEMS Microbiology Ecology 51 (2005) 375–388 385

M. austroafricanum andM. petroleophilum were detected

in contaminated soils K3840 and B101, but not in soil

TM. However, the dominant 16S rRNA gene soil clone

sequences from all 3 PAH-contaminated soils showed

highest sequence similarities with the 16S rRNA gene

of the relatively unknown M. tusciae. M. tusciae relatedsequences were not retrieved from the non-contaminated

soil. Interestingly, the different soil fingerprints revealed

bands closely related to M. tusciae, but with different

migration profiles. In addition, the cloned sequences

showed a relatively high variation in similarity scores

(from 99% to 95%). The M. tusciae sequences isolated

in this study grouped with other unidentifiedMycobacte-

rium sequences cloned from DNA from petroleum-contaminated soils found by Cheung and Kinkle, using

the MycF and MycR primer pair [31] (Fig. 3(b)). Besides

the M. tuscia strains, only strains of the M. monacence

(AF107039) species, a fast-growing species represented

by a type strain of clinical origin and an atypical isolate

(U46146), seem to be closely linked to this cluster.

4. Discussion

We developed a specific PCR-DGGE method to rap-

idly monitor the community composition of fast-grow-

ing mycobacteria in PAH-contaminated soil. The

newly designed MYCO66f and MYCO600r primer set

is the first primer set that is specifically developed for

the detection of fast-growing Mycobacterium species.All previously described primer combinations were spe-

cific for the total Mycobacterium genus [22,24–31] or

exclusively for pathogenic and facultative pathogenic

slow-growing mycobacteria [23,28].

DGGE analysis of gene fragments amplified with the

MYCO primer set could differentiate between most fast-

growing Mycobacterium species, including all important

PAH-degrading species. For some very closely relatedspecies the DGGE fingerprints were identical, due to

the strong conservation of the Mycobacterium 16S

rRNA genes. For comparison, the 16S rRNA gene

amplicons obtained with primer pair MycF and GC40-

MycR, designed by Cheung and Kinkle [31], were also

analyzed by DGGE. Although a different fragment of

the 16S rRNA gene was amplified, PCR-DGGE with

both primer sets resulted in the same species differentia-tion degrees for all Mycobacterium strains tested (data

not shown). The amplified rrn gene fragments of species

that could not be differentiated by DGGE, such as M.

spaghni, M. hodleri and M. gilvum, had a similarity of

99–100%. Similarly, other studies using DGGE analysis

of 16S rRNA gene fragments could not discriminate be-

tween several species of Burkolderia [53] and Bifidobacte-

rium [54] or other Gram-positive coryneform soilbacteria such as Arthrobacter and Nocardoides [55],

due to the high conservation levels of the amplified

16S rRNA gene fragments. It is clear that the practical

resolution limit of the DGGE technique is at the species

or genus level or intermediate between the two, depend-

ing on the gene conservation level within the taxonomic

group that is under investigation.

Most Mycobacterium strains were characterized by asingle-band DGGE fingerprint. Only a few strains

showed satellite bands. The same numbers of DGGE

bands per strain were obtained when using the MycF

and MycR primers (designed by Cheung and Kinkle)

on the test strains [31] (data not shown). Others also re-

ported multiple-band DGGE patterns for pure strains of

species such as Paenibacillus polymyxa [56], Burkholde-

ria cepacia [53] and Bifidobacterium adolescentis [54],due to the presence of multiple 16S rRNA gene copies

with sequence heterogeneity. However, based on own

southern blotting and DNA–DNA hybridization results

(data not shown), other reports and analysis of the total

genome sequences from 3 mycobacteria from clinical

origin, slow-growing and fast-growing mycobacteria

possess only 1 and 2 copies, respectively, of rrn genes

with minor sequence variations [44,57], theoreticallyleading to a maximum of 2 homoduplex bands in a

DGGE gel. The additional bands in the upper part of

the DGGE gel are presumably the result of heteroduplex

formation during PCR between the different copies of

16S rRNA genes within one strain [54,58]. Such specific

DGGE fingerprints of 16S rRNA gene heteroduplexes

have even been used for the identification of Mycobacte-

rium strains [58]. Taking into account the possibleheteroduplex formation and the fact that some species

or strains may give more than one fragment in DGGE,

some caution must be exercised when interpreting the

DGGE community fingerprints, especially when esti-

mating strain and species numbers and diversity. On

the other hand, multiple bands were only observed for

some isolates, so these effects may be minor.

With the one-step PCR-DGGE method, using onlythe MYCO66f and GC40-MYCO600r primer pair,

mycobacteria could be detected with a detection limit

of approximately 106 cells per gram soil. None of the

other Mycobacterium genus specific primer sets devel-

oped in the past report on the Mycobacterium abun-

dance or detection limit in environmental samples for

comparison [28–31], but we found a similar detection

limit when using the primer set developed by Cheungand Kinkle [31]. The value reported for a similar direct

PCR-DGGE detection method for Burkholderia species

in soil was only slightly lower (detection limit 5 · 105

CFU g�1) [53], although more copies of the rrn genes

are present in the target bacterium (6 rrn copies in Burk-

holderia while 2 in mycobacteria). Nested PCR, using

eubacterial primers in the first round and the MYCO-

primers with GC-clamp in the second round, drasticallylowered the detection limit to approximately 102 cells

per gram soil. Another approach could be the use of

386 N.M. Leys et al. / FEMS Microbiology Ecology 51 (2005) 375–388

the more abundant rRNA molecules instead of the

rRNA gene as targets for the MYCO-primers in a re-

verse transcription PCR (RT-PCR) protocol. In a RT-

PCR set up using the MYCO-primer set, Mycobacte-

rium sp. LB501T was detected at a concentration as

low as 102 active cells per gram soil (Hendrickx et al.,unpublished data). Based on our results, all fast-growing

Mycobacterium strains are expected to be detected

equally well in a mixed Mycobacterium community.

Our results clearly suggest a wide distribution of fast-

growing mycobacteria in the environment, since fast-

growing mycobacteria could be detected in most con-

taminated and uncontaminated soils. Unlike Cheung

and Kinkle [31], we found no clear correlation betweenMycobacterium biodiversity (assessed by the number of

bands in the Mycobacterium DGGE fingerprints) and

the PAH-concentration of the soils. In contrast, strong

signals for Mycobacterium were especially obtained with

soils containing relatively low concentrations of PAHs

with mainly higher molecular weight and hence low-bio-

available PAHs. In comparison, weaker signals for

Mycobacterium were obtained with soils containing highconcentrations of mostly more bioavailable and more

easily degradable 3-ring PAHs such as phenanthrene.

In the latter case, mycobacteria might be out-competed

by more quickly growing PAH-degrading bacteria such

as Sphingomonas or Pseudomonas species. Interestingly,

in a parallel study examining the same soils as in this

study, high concentrations of Sphingomonas were espe-

cially encountered in the soils containing high phenanth-rene concentrations [64]. The presence of mycobacteria

in soils with lower PAH-concentrations may indicate a

natural selection of fast-growing Mycobacterium species

in PAH-polluted soil enriched in less bioavailable and

more recalcitrant, high(er)-molecular weight PAHs.

Mycobacterium species may be better adapted to harsh

oligothrophic soil conditions, as they have a low mainte-

nance energy demand and make use of several PAH bio-availability-enhancing mechanisms such as high-affinity

uptake systems and special adhesion to the substrate

[35,59].

Strains closely related to known, fast-growing PAH-

degrading isolates belonging to the M. frederiksbergense

and M. austroafricanum species were detected only in

PAH-contaminated soils. None of the detected se-

quences from contaminated soils seemed to originatefrom strains belonging to M. gilvum species, another

species known to comprise many PAH-degrading strains

[3,5,60]. Surprisingly, mainly sequences related to theM.

tusciae species were repeatedly detected in all PAH-con-

taminated soils, originating from different countries and

different industrial sites, but not in the uncontaminated

soil. The M. tusciae sequences isolated in this study

grouped with other unidentified Mycobacterium se-quences cloned from DNA from petroleum-contami-

nated soils and found by Cheung and Kinkle using the

MycF and MycR primer pair [31]. These results may

indicate an important role for M. tusciae and/or its re-

lated species in PAH-degradation processes in soil.

The M. tusciae species has never been isolated or de-

tected before in PAH-contaminated soils and the type

strain of this species, M. tusciae strain DSM44338T, isa facultative pathogenic clinical isolate from a sick child

[61]. However, recently, two unpublished vinyl chloride-

degrading soil isolates (Coleman et al., unpublished)

were identified as members of the M. tusciae species.

Based on the different DGGE bands and the varying

similarity of the clones to the type strain, we may have

detected two still unknown species relatively closely re-

lated to M. tusciae.The repeated detection of Mycobacterium cells in

soils with low PAH-concentrations support the natural

importance of fast-growing Mycobacterium species in

PAH-polluted soil. The developed PCR-DGGE detec-

tion system is an important tool to specifically monitor

the natural abundance, the diversity and the dynamics

of these bacteria in soil for optimization of bioremedia-

tion. The developed primer pair may also be useful in aRT-PCR approach with ribosomal RNA soil extracts to

analyze the diversity of the actively PAH-degrading

population of fast-growing mycobacteria. Eventually,

these primers could be combined with primers developed

for the detection of messenger RNA of the well-con-

served PAH-catabolic genes in mycobacteria [62,63].

This will contribute to a better understanding of the role

of mycobacteria in the biodegradation of PAHs in theenvironment.

Acknowledgments

This work was supported by the European Commis-

sion, through the contracts BIO4-CT97-2015 and

QLRT-1999-00326. We thank E.M.H. Wellington forproviding bacterial strains and S. Schioetz-Hansen, J.

Amor and J. Vandenberghe for providing soil samples.

References

[1] Cerniglia, C.E. (1992) Biodegradation of polycyclic aromatic

hydrocarbons. Biodegradation 3, 351–368.

[2] Guerin, W.F. and Jones, G.E. (1988) Mineralization of phen-

anthrene by a Mycobacterium sp.. Appl. Environ. Microbiol. 54,

937–944.

[3] Bastiaens, L. (1998) Isolation and characterization of polycyclic

aromatic hydrocarbon-degrading bacteria and evaluation of the

use of isolates in soil slurry experiments. PhD Thesis, Doctor-

aatsproefschrift nr 363, Faculteit Landbouwkundige en Toege-

paste Biologische Wetenschappen, KULeuven, Leuven.

[4] Churchill, S.A., Harper, J.P. and Churchill, P.F. (1999) Isolation

and characterization of a Mycobacterium species capable of

degrading three- and four-ring aromatic and aliphatic hydrocar-

bons. Appl. Environ. Microbiol. 65, 549–552.

N.M. Leys et al. / FEMS Microbiology Ecology 51 (2005) 375–388 387

[5] Bastiaens, L., Springael, D., Wattiau, P., Harms, H., deWachter,

R., Verachtert, H. and Diels, L. (2000) Isolation of adherent

polycyclic aromatic hydrocarbon (PAH)-degrading bacteria

using PAH-sorbing carriers. Appl. Environ. Microbiol. 66,

1834–1843.

[6] Solano-Serena, F., Marchal, R., Caseregola, S., Vasnier, C.,

Lebeault, J.-.M. and Vandecasteele, J.-.P. (2000) A Mycobacte-

rium strain with extended capacities for degradation of gasoline

hydrocarbons. Appl. Environ. Microbiol. 66, 2392–2399.

[7] Willumsen, P., Karlson, U., Stackebrandt, E. and Kroppenstedt,

R.M. (2001) Mycobacterium frederiksbergense sp. nov., a novel

polycyclic aromatic hydrocarbon-degrading Mycobacterium spe-

cies. Int. J. Syst. Evol. Microbiol. 51, 1715–1722.

[8] Briglia, M., Eggen, R.I., Van Elsas, D.J. and De Vos, W.M.

(1994) Phylogenetic evidence for transfer of pentachlorophenol-

mineralizing Rhodococcus chlorophenolicus PCP-I(T) to the genus

Mycobacterium. Int. J. Syst. Bacteriol. 44, 494–498.

[9] Poelarends, G.J., van Hylckama Vlieg, J.E.R., Marchesi, J.R.,

Freitas Dos Santos, L.M. and Janssen, D.B. (1999) Degradation

of 1,2-dibromoethane by Mycobacterium sp. strain GP1. J.

Bacteriol. 181, 2050–2058.

[10] Yagi, O., Hashimoto, A., Iwasaki, K. and Nakajima, M. (1999)

Aerobic degradation of 1,1,1-trichloroethane by Mycobacterium

spp. isolated from soil. Appl. Environ. Microbiol. 65, 4693–4696.

[11] Schrader, T., Schuffenhauer, G., Sielaff, B. and Andreesen, J.R.

(2000) High morpholine degradation rates and formation of

cytochrome P450 during growth on different cyclic amines by

newly isolated Mycobacterium sp. strain HE5. Microbiology 146,

1091–1098.

[12] Stalh, D.H. and Urbance, J.W. (1990) The division between fast-

and slow-growing species correspond to natural relationships

among the mycobacteria. J. Bacteriol. 172, 116–124.

[13] Rogall, T., Wolters, J., Flohr, T. and Bottger, E.C. (1990)

Towards a phylogeny and definition of species at the molecular

level within the genus Mycobacterium. Int. J. Syst. Bacteriol. 40,

323–330.

[14] Pitulle, C., Dorsch, M., Kazda, J., Wolters, J. and Stackebrandt,

E. (1992) Phylogeny of rapidly growing members of the genus

Mycobacterium. Int. J. Syst. Bacteriol. 42, 337–343.

[15] Tortoli, E. (2003) Impact of genotypic studies on mycobacterial

taxonomy: the new mycobacteria of the 1990s. Clin. Microbiol.

Rev. 16, 319–354.

[16] Staley, J.T. and Konopka, A. (1985) Measurement of in situ

activities of non-photosynthetic microorganisms in aquatic and

terrestrial habitats. Ann. Rev. Microbiol. 39, 321–346.

[17] Amann, R.I., Ludwig, W. and Schleifer, K.-.H. (1995) Phylo-

genetic identification and in situ detection of individual microbial

cells without cultivation. Microbiol. Rev. 59, 143–169.

[18] Ghezzi, J.I. and Steck, T.R. (1999) Induction of the viable but

non-culturable condition in Xanthomonas campestris pv. campes-

tris in liquid microcosms and sterile soil. FEMS. Microbiol. Ecol.

30, 203–208.

[19] Barry, C.E., Lee, R.E., Mdluli, K., Sampson, A.E., Schroeder,

B.G., Slayden, R.A. and Yuan, Y. (1998) Mycolic acids: structure,

biosynsthesis and physiological functions. Prog. Lipid Res. 37,

143–179.

[20] Draper, F. (1998) The outer parts of the mycobacterial envelope

as permeability barriers. Front. Biosci. 3, 1253–1261.

[21] Allen, B.W. (1998) Mycobacteria: general culture methodology

and safety considerations In: Mycobacteria Protocols (Parish, T.

and Stoker, N.G., Eds.), pp. 15–30. Humana Press Inc., Totowa,

NJ.

[22] Boddinghaus, B., Rogall, T., Flohr, T., Blocker, H. and Bottger,

E.C. (1990) Detection and identification of mycobacteria by

amplification of rRNA. Clin. Microbiol. 28, 1751–1759.

[23] De Beenhouwer, H., Liang, Z., De Rijk, P., Van Eekeren, C. and

Portaels, F. (1995) Detection and identification of mycobacteria

by DNA amplification and oligonucleotide-specific capture plate

hybridisation. J. Clin. Microbiol. 33, 2994–2998.

[24] Kox, L.F.F., van Leeuwen, J., Kuijper, S., Jansen, H.M. and

Kolk, A.H.J. (1995) PCR assay based on DNA coding for 16S

rRNA for detection and identification of mycobacteria in clinical

samples. J. Clin. Microbiol. 33, 3225–3233.

[25] Kox, L.F.F., Jansen, H.M., Kuijper, S. and Kolk, A.H.J. (1997)

Multiplex PCR assay for immediate identification of the infecting

species in patients with mycobacterial diseases. J. Clin. Microbiol.

35, 1492–1498.

[26] van der Heijden, I.M., Wilbrink, B., Schouls, L.M., Van Embden,

J.D.A., Breedveld, F.C. and Tak, P.P. (1999) Detection of

mycobacteria in joint samples from patients with arthritis using

a genus-specific polymerase chain reaction and sequence analysis.

Rheumatology 38, 547–553.

[27] Talaat, A.M., Reimschuessel, R. and Trucksis, M. (1997) Iden-

tification of mycobacteria infecting fish to the species level using

polymerase chain reaction and restriction enzyme analysis. Vet.

Microbiol. 58, 229–237.

[28] Schwartz, T., Kalmbach, S., Hoffmann, S., Szewzijk, U. and Obst,

U. (1998) PCR-based detection of mycobacteria in biofilms from

a drinking water distribution system. J. Clin. Microbiol. 33, 3225–

3233.

[29] Stinear, T., Davies, J.K., Jenkin, G.A., Hayman, J.A., Oppedis-

ano, F. and Johnson, P.D.R. (2000) Identification of Mycobac-

terium ulcerans in the environment from regions in southeast

Australia in which it is endemic with sequence capture-PCR.

Appl. Environ. Microbiol. 66, 3206–3213.

[30] Mendum, T.A., Chilima, B.Z. and Hirsch, P.R. (2000) The PCR

amplification of non-tuberculose mycobacterial 16S rRNA

sequences from soil. FEMS Microbiol. Lett. 185, 189–192.

[31] Cheung, P.-.Y. and Kinkle, B.K. (2001) Mycobacterium diversity

and pyrene mineralisation in petroleum-contaminated soils. Appl.

Environ. Microbiol. 67, 2222–2229.

[32] Heuer, H. and Smalla, K. (1997) Application of denaturing

gradient gel electrophoresis and temperature gradient gel

electrophoresis for studying soil microbial communities In:

Modern Soil Microbiology (van Elsas, J.D., Trevors, J.T. and

Wellington, E.N.M., Eds.), pp. 353–373. Marcel Dekker, New

York.

[33] Wattiau, P., Springael, D., Gemoets, J., Diels, L. and Cornelis, G.

(1999) Bacterial monitoring of a crude oil-contaminated soil

undergoing laboratory-scale bioremediation In: Novel

Approaches for Bioremediation of Organic Pollution (Fass, R.,

Flashner, Y. and Reuveny, S., Eds.). Plenum Publishing corp.,

New York.

[34] Mergeay, M., Nies, D., Schlegel, H.G., Gerits, J., Charles, P. and

Van Gijsegem, F. (1985) Alcaligenes eutrophus CH34 is a

facultative chemolithotroph with plasmid-bound resistance to

heavy metals. J. Bacteriol. 162, 328–334.

[35] Wick, L.Y., Colangelo, T. and Harms, H. (2001) Kinetics of mass

transfer-limited bacterial growth on solid PAHs. Environ. Sci.

Technol. 35, 354–361.

[36] USEPA (2004) 2002 Toxic Release Inventory, Public Data

Release, Office of Environmental Information, United States

Environmental Protection Agency, Washington, DC. Available

from: <http://www.epa.gov/triinter/tridata/tri02/index.htm>.

[37] Benson, D.A., Karsch-Mizrachi, I., Lipman, D.J., Ostell, J. and

Wheeler, D.L. (2004) GenBank: update. Nucleic Acids Res. 32,

D23–D26.

[38] Rice, P., Longden, I. and Bleasby, A. (2000) EMBOSS: The

European Molecular Biology Open Software Suite. Trends Genet.

16, 276–277.

[39] Cole, J.R., Chai, B., Marsh, T.L., Farris, R.J., Wang, Q., Kulam,

S.A., Chandra, S., McGarrell, D.M., Schmidt, T.M., Garrity,

G.M. and Tiedje, J.M. (2003) The Ribosomal Database Project

(RDP-II): previewing a new autoaligner that allows regular

388 N.M. Leys et al. / FEMS Microbiology Ecology 51 (2005) 375–388

updates and the new prokaryotic taxonomy. Nucleic Acids Res.

31, 442–443.

[40] Altschul, S.F., Gish, W., Miller, W., Myers, E.W. and Lipman,

D.J. (1990) Basic local alignment search tool. J. Mol. Biol. 215,

403–410.

[41] Muyzer, G., de Waal, E.C. and Uitterlinden, A.G. (1993)

Profiling of complex microbial populations by denaturing gradi-

ent gel electrophoresis analysis of polymerase chain reaction-

amplified genes encoding for 16S rRNA. Appl. Environ. Micro-

biol. 59, 695–700.

[42] Muyzer G., Hottentrager S., Teske A. and Wawer S. (1996)

Denaturing Gradient Gel Electrophoresis of PCR-amplified 16S

rDNA – A new molecular approach to analysis the genetic

diversity of mixed microbial population., 1-23 pp. NATO.

[43] Belisle, J.T. and Sonnenberg, M.G. (1998) Isolation of genomic

DNA from mycobacteria In: Mycobacteria Protocols (Parish, T.

and Stoker, N.G., Eds.), pp. 31–44. Humana Press Inc., Totowa,

New Jersey.

[44] Bercovier, H., Kafri, O. and Sela, S. (1986) Mycobacteria possess

a surprisingly small number of ribosomal RNA genes in relation

to the size of their genome. Biochem. Biophys. Res. Commun.

136, 1136–1141.

[45] Boon, N., Goris, J., De Vos, P., Verstraete, W. and Top, E.M.

(2000) Bioaugmentation of activated sludge by an indigenous 3-

chloroaniline-degrading Comamonas testosteroni strain I2gfp.

Appl. Environ. Microbiol. 66, 2906–2913.

[46] Marchesi, J.R., Takuichi, S., Weightman, A.J., Martin, T.A., Fry,

J.C., Hiom, S.J. and Wade, W.G. (1998) Design and evaluation of

useful bacterium-specific PCR primers that amplify genes coding

for bacterial 16S rRNA. Appl. Environ. Microbiol. 64, 795–799.

[47] Polz, M.F. and Cavanaugh, C.M. (1998) Bias in template-to-

product ratios in multitemplate PCR. Appl. Environ. Microbiol.

64, 3724–3730.

[48] Muyzer, G., Brinkhoff, T., Nubel, U., Santegoeds, C., Schafer, H.

and Wawer, C. (1998) Denaturing gradient gel electrophoresis

(DGGE) in microbial ecologyMol. Micro. Ecol. Manual, pp. 1–

27. Humana Press Inc., Totowa, New Jersey.

[49] Saitou, N. and Nei, M. (1987) The neighbor-joining method: a

new method for reconstructing phylogenetic trees. Mol. Biol.

Evol. 4, 406–425.

[50] Felsenstein, J. (1981) Evolutionary trees from DNA sequences: a

maximum likelihood approach. J. Mol. Evol. 17, 368–376.

[51] Kluge, A.G. and Farris, J.S. (1969) Quantitative phyletics and the

evolution of anurans. System. Zool. 18, 1–32.

[52] Centers for Disease Control and Prevention (CDC), Office of

Health and Safety (OHS) (1999) Biosafety in microbiological and

biomedical laboratories (BMBL). US Government Printing

Office., Washington.

[53] Falcao Salles, J., Adriano De Souza, F. and van Elsas, J.D.

(2002) Molecular method to asses the diversity of Burkholderia

species in environmental samples. Appl. Environ. Microbiol. 68,

1595–1603.

[54] Satokari, R.M., Vaughan, E.E., Akkermans, A.D.L., Saarela, M.

and de Vos, W.M. (2001) Bifidobacterial diversity in human feces

detected by genus-specific PCR and denaturing gradient gel

electrophoresis. Appl. Environ. Microbiol. 67, 504–513.

[55] Felske, A., Vancanneyt, M., Kersters, K. and Akkermans, A.D.

(1999) Application of temperature-gradient electrophoresis in

taxonomy of coryneform bacteria. Int. J. Syst. Bacteriol. 49, 113–

121.

[56] Nubel, U., Engelen, B., Felske, A., Snaidr, J., Wieshuber, A.,

Amann, R.I., Wolfgang, L. and Backhaus, H. (1996) Sequence

heterogeneities of genes encoding 16S rRNAs in Paenibacillus

polymyxa detected by Temperature Gradient Gel Electrophoresis.

J. Bacteriol. 178, 5636–5643.

[57] Klappenbach, J.A., Saxman, P.R., Cole, J.T. and Schmidt, T.M.

(2001) rrndb: The ribosomal RNA operon copy number database.

Nucleic Acids Res. 29, 181–184.

[58] Waleria-Aleixo, A., Kroon, E.G., Campos, M.A.S., Margutti-

Pinto, M.E., Bonjardim, C.A. and Ferreira, P.C.P. (2000)

Heteroduplex mobility assay for rapid, sensitive and specific

detection of mycobacteria. Diagnostic Microbiol. Infect. Disease

36, 225–235.

[59] Wick, L.Y., Ruiz de Munain, A., Springael, D. and Harms, H.

(2002) Responses of Mycobacterium sp. LB501T to the low

bioavailability of solid anthracene. Appl. Microbiol. Biotechnol.

58, 378–385.

[60] Boldrin, B., Thiem, A. and Fritzsche, C. (1993) Degradation of

phenanthrene, fluorene and pyrene by aMycobacterium sp.. Appl.

Environ. Microbiol. 59, 1927–1930.

[61] Tortoli, E., Kroppenstedt, R.M., Bartoloni, A., Caroli, G., Jan, I.,

Pawlowski, J. and Emler, S. (1999) Mycobacterium tusciae sp.

nov.. Int. J. Syst. Bacteriol. 49, 1839–1844.

[62] Khan, A.A., Wang, R.F., Cao, W.W., Doerge, D.R., Wenner-

strom, D. and Cerniglia, C.E. (2001) Molecular cloning, nucle-

otide sequence and expression of genes encoding a polycyclic

aromatic ring dioxygenase from Mycobacterium sp. strain PYR-1.

Appl. Environ. Microbiol. 67, 3577–3585.

[63] Krivobok, S., Kuony, S., Meyer, C., Louwagie, M., Willison, J.C.

and Jouanneau, Y. (2003) Identification of pyrene-induced

proteins in Mycobacterium sp. strain 6PY1: evidence for two

ring-hydroxylating dioxygenases. J. Bacteriol. 185, 3828–3841.

[64] Leys, N., Ryngaert, A., Bastiaens, L., Verstraete, W., Top, E.M.

and Springael, D. (2004) Occurrence and phylogenetic diversity of

Sphingomonas in soils contaminated with polycyclic aromatic

hydrocarbons (PAHs). Appl. Environ. Microbiol. 70, 1944–1955.