Downloaded from www.microbiologyresearch.org by

IP: 54.160.113.5

On: Sun, 29 May 2016 21:00:06

IncP-1b plasmids of Comamonas sp. and Delftiasp. strains isolated from a wastewater treatmentplant mediate resistance to and decolorization ofthe triphenylmethane dye crystal violet

Yvonne Stolze,13 Felix Eikmeyer,13 Daniel Wibberg,1 Gerrit Brandis,1

Christina Karsten,1 Irene Krahn,1 Susanne Schneiker-Bekel,1

Prisca Viehover,1 Aiko Barsch,2 Matthias Keck,3 Eva M. Top,4

Karsten Niehaus3 and Andreas Schluter1

Correspondence

Andreas Schluter

aschluet@CeBiTec.Uni-Bielefeld.

DE

Received 12 March 2012

Revised 24 May 2012

Accepted 28 May 2012

1Institute for Genome Research and Systems Biology, CeBiTec, Bielefeld University,PO Box 100131, D-33501 Bielefeld, Germany

2Bruker Daltonik GmbH, Fahrenheitstr. 4, D-28359 Bremen, Germany

3Proteome and Metabolome Research, Faculty of Biology, Bielefeld University, PO Box 100131,D-33501 Bielefeld, Germany

4Department of Biological Sciences, Initiative for Bioinformatics and Evolutionary Studies,University of Idaho, Moscow, ID 83844-3051, USA

The application of toxic triphenylmethane dyes such as crystal violet (CV) in various industrial

processes leads to large amounts of dye-contaminated sludges that need to be detoxified. Specific

bacteria residing in wastewater treatment plants (WWTPs) are able to degrade triphenylmethane

dyes. The objective of this work was to gain insights into the genetic background of bacterial strains

capable of CV degradation. Three bacterial strains isolated from a municipal WWTP harboured

IncP-1b plasmids mediating resistance to and decolorization of CV. These isolates were assigned

to the genera Comamonas and Delftia. The CV-resistance plasmid pKV29 from Delftia sp. KV29

was completely sequenced. In addition, nucleotide sequences of the accessory regions involved in

conferring CV resistance were determined for plasmids pKV11 and pKV36 from the other two

isolates. Plasmid pKV29 contains typical IncP-1b backbone modules that are highly similar to those

of previously sequenced IncP-1b plasmids that confer antibiotic resistance, degradative capabilities

or mercury resistance. The accessory regions located between the conjugative transfer (tra) and

mating pair formation modules (trb) of all three plasmids analysed share common modules and

include a triphenylmethane reductase gene, tmr, that is responsible for decolorization of CV.

Moreover, these accessory regions encode other enzymes that are dispensable for CV degradation

and hence are involved in so-far-unknown metabolic pathways. Analysis of plasmid-mediated

degradation of CV in Escherichia coli by ultra-high-performance liquid chromatography-

electrospray ionization-quadrupole-time-of-flight MS revealed that leuco crystal violet was the first

degradation product. Michler’s ketone and 4-dimethylaminobenzaldehyde appeared as secondary

degradation metabolites. Enzymes encoded in the E. coli chromosome seem to be responsible for

cleavage of leuco crystal violet. Plasmid-mediated degradation of triphenylmethane dyes such as

CV is an option for the biotechnological treatment of sludges contaminated with these dyes.

3These authors contributed equally to this work/paper.

Abbreviations: BPC, base peak chromatogram; CV, crystal violet; EIC, extracted ion chromatogram; GMC oxidoreductase, glucose-methanol-cholineoxidoreductase; leuco-CV, leuco crystal violet; MK, Michler’s ketone; UHPLC, ultra-high-performance liquid chromatography; UHPLC-ESI-Q-TOF-MS,ultra-high-performance liquid chromatography-electrospray ionization-quadrupole-time-of-flight MS; WWTP, wastewater treatment plant.

The GenBank/EMBL/DDBJ accession numbers for the nucleotide sequences of plasmid pKV29 and of the accessory gene regions of plasmids pKV11and pKV36 are JN648090 (pKV29), JN648091 (pKV36) and JN648092 (pKV11).

Microbiology (2012), 158, 2060–2072 DOI 10.1099/mic.0.059220-0

2060 059220 G Printed in Great Britain

Downloaded from www.microbiologyresearch.org by

IP: 54.160.113.5

On: Sun, 29 May 2016 21:00:06

INTRODUCTION

The triphenylmethane dye crystal violet (CV) is used inseveral industrial processes and for different applications(Docampo & Moreno, 1990), e.g. in the textile industry tostain wool, cotton, silk and other raw materials, in paperprinting (Azmi et al., 1998), as a reagent for bacterialclassification (Kwasniewska, 1985), as an antifungal agent(Docampo & Moreno, 1990) and in cosmetic products (Janget al., 2005). CV is harmful in different ways to humans andanimals (Docampo & Moreno, 1990). In solution it is acation able to interact with nucleophilic compounds such asnegatively charged DNA (Littlefield et al., 1985). Binding ofCV to DNA can lead to chromosomal damage and reducedexpression of genes (Au et al., 1978), as well as increasedgrowth of tumour cells in mice (Littlefield et al., 1985) andsome fish (Chen et al., 2007). Moreover, CV can even causeblindness and kidney damage in humans (Mittal et al.,2010). The extensive application of CV and other triphe-nylmethane dyes, especially in the textile industry, leads tolarge amounts of contaminated wastewater, which need tobe cleaned and detoxified before disposal to avoid environ-mental contamination, and animal and human healthproblems (Chen et al., 2008).

Several chemical and physical methods have been deve-loped to degrade toxic compounds such as CV and othertriphenylmethane dyes, but these processes are expensiveand lead to large amounts of sludge (Azmi et al., 1998).Decolorization of triphenylmethane dyes and in particularCV by certain bacteria such as Nocardia corallina (Yatomeet al., 1993), Mycobacterium avium (Jones & Falkinham,2003), Aeromonas hydrophila (Ren et al., 2006) and Kurthiasp. (Sani & Banerjee, 1999), as well as by some fungisuch as the white rot fungus Phanaerochete chrysoporium(Bumpus & Brock, 1988), represents an alternative bio-logical solution. In the case of N. corallina (Yatome et al.,1993) and Bacillus subtilis (Yatome et al., 1991), Michler’sketone (MK) was detected as a CV degradation pro-duct. Analysis of CV degradation products released byShewanella sp. also revealed MK and four other products(Chen et al., 2008). The enzyme triphenylmethane reduc-tase, encoded by the tmr gene in Citrobacter sp. strainKCTC 18061P, was found to be responsible for decol-orization of triphenylmethane dyes such as CV (Jang et al.,2005). A triphenylmethane reductase gene is also encodedin an accessory region of the IncP-1b plasmid pGNB1,which was isolated from a bacterium from a wastewatertreatment plant (WWTP) (Schluter et al., 2007). Thisplasmid was found to confer resistance to and decoloriza-tion of CV.

To further analyse plasmid-mediated decolorization oftriphenylmethane dyes and to elucidate correspondingdegradation pathways, new CV-resistance plasmids wereisolated from activated sludge bacteria from a WWTP.Sequencing of these plasmids revealed their geneticstructure and putative genes that might function indegradative pathways. Metabolites arising in the course

of plasmid-mediated CV degradation were further analysedby means of ultra-high-performance liquid chromatogra-phy-electrospray ionization-quadrupole-time-of-flight MS(UHPLC-ESI-Q-TOF-MS).

METHODS

Isolation of bacteria that are non-susceptible to CV from an

activated sludge bacterial community. Activated sludge samples

were taken from an activated sludge basin of the WWTP Bielefeld-

Heepen (Germany) in March 2009. Activated sludge bacteria were

cultivated on Luria broth (LB) agar medium containing CV (10 mg

ml21) for 40 h at 25 uC. Growing bacteria were screened for clones

able to decolorize CV (Schluter et al., 2007). Positive clones appear

white and are surrounded by a colourless halo, indicating degradation

of CV. In contrast, bacterial clones that are non-susceptible to CV

feature an intensely violet colour.

Total plasmid DNA was isolated from positive clones by means of the

NucleoBond PC 100 kit on AX 100 columns according to the protocol

supplied by the manufacturer (Macherey-Nagel). Subsequently,

plasmid DNA preparations were used to transform CaCl2-competent

Escherichia coli KAM3 (Morita et al., 1998) and DH5a cells.

Transformants were selected on LB medium supplemented with CV

(10 mg ml21). E. coli DH5a strains (Grant et al., 1990) containing

plasmid pKV11, pKV29, pKV36, pGNB1 or pIK405 were grown in LB

medium supplemented with CV (10 mg ml21) at 37 uC. For

decolorization assays these strains were grown in 10 ml M9 minimal

medium containing CV (10 mg ml21) for 24 h at 37 uC.

Standard DNA techniques. The plasmid content of candidate

isolates and E. coli strains was analysed in Eckhardt gels as described

elsewhere (Hynes et al., 1985). Plasmid DNA was isolated from E. coli

strains by means of the NucleoBond PC 100 kit (Macherey-Nagel),

whereas recombinant plasmids containing subcloned restriction

fragments were isolated by using the QIAprep Spin Miniprep kit

(Qiagen). Restriction enzyme digestion, agarose gel electrophoresis,

DNA cloning and transformation of E. coli were carried out according

to Sambrook et al. (1989).

Amplification of 16S rRNA gene fragments and part of the

triphenylmethane reductase gene tmr. Crude total DNAs from

activated sludge bacterial isolates were prepared as described

previously (Schluter et al., 2007). 16S rDNA fragments were amplified

from crude total DNA preparations by using the primers 1385r and

27f (Schluter et al., 2007). Taxonomic classification of 16S rDNA

sequences was done by means of the RDP-Classifier (Wang et al., 2007)

and by comparison with the NCBI nucleotide sequence database using

the BLASTN algorithm. The primer pair tmr-L1 (59-CCGTCATCA-

AACAAACGTGA-39) and tmr-R1 (59-TCGTTTATTTTTGGCTT-

CCG-39) was used to generate an amplicon containing only the

complete pKV29 tmr gene encoding the enzyme triphenylmethane

reductase. PCRs were done under standard conditions using the

BIOTAQ DNA polymerase (Bioline). Amplicons were verified by

sequencing. The amplicon containing the tmr gene from plasmid

pKV29 was cloned into the T-cloning vector pGEM-T Easy (Promega).

The corresponding construct was named pIK405.

Complete sequencing of the CV-resistance plasmid pKV29.Plasmid DNA for sequence determination was isolated from E. coli

DH5a (pKV29) using the Plasmid Midi kit (Qiagen) as for low-

copy-number plasmids. The DNA sequence of pKV29 was

determined by pyrosequencing. The shotgun sequencing protocol

using the Roche/454 platform has been described elsewhere

(Margulies et al., 2005). The plasmid was sequenced together with

Crystal violet resistance IncP-1b plasmids

http://mic.sgmjournals.org 2061

Downloaded from www.microbiologyresearch.org by

IP: 54.160.113.5

On: Sun, 29 May 2016 21:00:06

29 other multiplex identifier (MID)-tagged libraries in a pool on a

GS FLX platform with Titanium sequencing chemistry (Roche/454

Life Sciences). The number of reads obtained for plasmid pKV29 was

11 522 with an average read length of ~320 bp. The generated

amount of sequence information accounted for a 60-fold coverage of

the plasmid genome. Sequence reads were assembled using the

Newbler assembly software (Roche/454 Life Sciences). Assembly of

pKV29 reads resulted in two contigs. To complete the pKV29

sequence, a PCR-based strategy was applied, which involved PCR

amplification of fragments covering gaps between contigs and

sequencing of these amplicons.

Annotation of plasmid sequences. Annotation of plasmid

sequences was done within the Annotation Platform GenDB 2.0

(Meyer et al., 2003) essentially as described previously (Eikmeyer

et al., 2012; Heinl et al., 2012; Szczepanowski et al., 2011). Intergenic

regions were analysed manually for missed coding sequences by

applying BLAST programs (Altschul et al., 1997). Automatic annota-

tions were curated manually if necessary. Plasmid DNA and amino

acid sequences were compared with related plasmids deposited in the

NCBI database by applying the tools BLASTN, BLASTP and BLASTX

(Altschul et al., 1997).

Restriction fragment library construction, sequencing and

sequence analysis of the pKV11 and pKV36 CV-resistance

gene regions. Purified pKV11 and pKV36 plasmid DNA was

restricted with the restriction endonucleases BamHI, EcoRI, PstI, SalI,

SphI and NotI. Restriction fragments representing the CV-resistance

gene region were separately subcloned into the sequencing vectors

pUC18 (Yanisch-Perron et al., 1985), pBluescript-II-KS (Stratagene)

and pZErO-2 (Invitrogen). Sequencing of recombinant plasmids was

done as described previously (Szczepanowski et al., 2007). Computer-

assisted assembly and sequence quality control were carried out with

the Consed/Autofinish software tool (Gordon et al., 1998, 2001). For

annotation of the finished sequence, the annotation tool GenDB was

applied (Meyer et al., 2003).

Comparative analysis of accessory plasmid regions. To

determine homologous regions of the plasmids pKV11, pGNB1,

pKV29 and pKV36, pairwise BLASTN analyses were performed. The

sequences of accessory regions were extracted from plasmid sequences

between the inverted repeat of the insertion sequence IS1071 and

the stop codon of the second ORF located downstream of the

triphenylmethane reductase gene tmr.

Spectrophotometric measurement of plasmid-mediated decol-

orization of CV. To follow decolorization of CV by E. coli DH5a

(pIK405, tmr), 6 ml of overnight cultures (OD600 0.96) were

separately transferred into 100 ml glucose M9 minimal medium

(supplemented with thiamine) containing CV (5 mg ml21) and

were shaken (150 r.p.m.) at 37 uC. Aliquots (1 ml) of the cultures

were taken every 2 h beginning at time point zero (t50) until

decolorization of CV had been completed (t510). Sampled aliquots

were centrifuged and the absorbance of cell-free supernatants was

measured from 200 to 700 nm (Ultrospec 3000 pro UV/visible

spectrophotometer, GE Healthcare). A plasmid-free E. coli DH5a

culture in M9 medium without CV was taken as control; M9

medium containing CV but no bacteria was taken as reference. For

additional decolorization assays, the bacterial strains Comamonas

sp. KV11, Delftia sp. KV29, Comamonas sp. KV36 and E. coli DH5a

(pKV29) were cultivated in triplicate experiments in LB medium at

room temperature. Samples were taken every 1.5 h and the absor-

bance of the supernatant was measured at 590 nm using LB me-

dium without CV as a reference. CV concentrations were calculated

from a calibration curve with known CV concentrations. CV

degradation rates were calculated for each period between

samplings.

Extraction of CV degradation products by solid-phase extrac-

tion (SPE). Aliquots (3 ml, OD600 1.1) of decolorized overnight

cultures were transferred into 100 ml M9 minimal medium contain-

ing CV (80 mg ml21) and shaken (150 r.p.m.) at 37 uC until complete

decolorization was achieved (~10 h). Cultures were centrifuged

(4 uC, 8000 r.p.m.) and supernatants were filtered using 0.2 mm

pore-size Whatman filters. CV degradation products were extracted

by SPE. Supernatants were applied onto columns (Discovery DSC-

C18, 1 ml, 100 mg, Supelco). After washing with water, the

compounds were eluted with 1 ml acetronitrile (containing 0.1 %

formic acid) and stored in the dark at 4 uC.

Analysis of CV degradation products by UHPLC-ESI-Q-TOF-

MS. Liquid chromatography (LC) separations of aliquots (3 ml) of the

extracted degradation products were performed using an ACQUITY

UHPLC system (Waters) combined with an ACQUITY UHPLC BEH

C18, 1.7 mm column (2.1650 mm, Waters). Eluent solvents were

0.1 % formic acid in water (A) and 0.1 % formic acid in acetonitrile

(B). The following gradient was applied at a flow rate of 0.35 ml

min21: 0 to 0.5 min isocratic at 95 % A; 0.5 to 6.5 min linear

gradient from 95 to 5 % A; 6.5 to 7.0 min isocratic at 95 % A; 7.0 to

7.5 min linear gradient from 5 to 95 % A. The total runtime was

10 min.

The UHPLC was hyphenated with an Apollo II ESI source operated in

positive ion mode to a micrOTOF-QII ESI-Q-TOF mass spectrometer

(Bruker Daltonics). Mass spectrometer source settings were as

follows: dry gas temperature 200 uC, dry gas flow 8 l min21, nebulizer

gas pressure 2 bar, capillary voltage 4500 V, end plate offset 2500 V.

Mass spectra were detected from 75 to 1000 m/z at 1 Hz acquisition

speed.

The data were evaluated using the DataAnalysis 4.0 software (Bruker

Daltonics). Target compounds were confirmed by comparison with

purified standards and by calculating the sum formulae using the

SmartFormula module in DataAnalysis 4.0, which takes into account

accurate mass and isotopic pattern information.

RESULTS AND DISCUSSION

Isolation of bacterial strains able to decolorize CVfrom an activated sludge community of a WWTP

To isolate bacteria that were able to decolorize the tri-phenylmethane dye CV, aliquots of diluted activated sludgesamples from a municipal WWTP were cultivated on LBagar medium containing CV to a final concentration of10 mg ml21. Approximately 2.5 % of activated sludgebacteria that were able to grow on LB medium were notsusceptible to the CV concentration applied. Moreover,about 1.4 % of those bacteria that could grow on LB me-dium supplemented with CV decolorized the dye. De-colorizing colonies were white, in contrast to dark violetcolonies, which were not susceptible to CV but were unableto decolorize it. Thirty-six colonies that showed reducedsusceptibility to CV and the decolorization phenotype werechosen for further analysis. Since in the first instance wewere interested in plasmid-encoded decolorization of CV,the obtained isolates were initially tested for their plasmid

Y. Stolze and others

2062 Microbiology 158

Downloaded from www.microbiologyresearch.org by

IP: 54.160.113.5

On: Sun, 29 May 2016 21:00:06

content by applying the Eckhardt gel lysis method. Eigh-teen out of 36 isolates appeared to contain plasmids.Plasmid-free isolates were discarded. To exclude putativepathogenic bacteria from further analysis, partial 16S rRNAgene sequences were determined for each isolate. Basedon the analysis of these sequences, seven isolates werepotentially pathogenic bacteria (Acinetobacter, Aeromonas,Klebsiella and Stenotrophomonas) and were thereforeexcluded from further analysis. The plasmids of five otherisolates appeared to be unstable, so that analysis of theseisolates was not feasible. The remaining six plasmid-containing isolates were named KV10, KV11, KV16, KV17,KV29 and KV36. The isolates KV11, KV29 and KV36contained plasmids of about 60–70 kb in size, whereas theisolates KV10, KV16 and KV17 harboured large plasmidsthat had sizes of at least 150 kb (data not shown). At thisstage it was not clear whether the CV decolorizationphenotype of the isolates was plasmid or chromosomallyencoded.

Bacterial isolates able to decolorize CV aremembers of the family Comamonadaceae

To taxonomically classify the activated sludge isolates thatwere able to decolorize CV, their 16S rRNA genes werepartially sequenced. All isolates were allocated to the classb-Proteobacteria, the order Burkholderiales and the familyComamonadaceae (Table 1). The isolates KV11 and KV36belonged to the genus Comamonas, whereas the other fourisolates were affiliated to the genus Delftia. Many Delftiaand Comamonas strains are able to degrade xenobiotics andother man-made compounds. Moreover, members of thegenera Delftia and Comamonas are known to harbourplasmids involved in degradation of different recalcitrantsubstances.

Isolates KV11, KV29 and KV36 containCV-resistance plasmids

To analyse whether the plasmids of the strains KV10,KV11, KV16, KV17, KV29 and KV36 were responsible forthe observed CV-resistance and decolorization phenotype,plasmid DNA was isolated from each strain and subse-quently used to transform E. coli KAM3 and E. coli DH5a.E. coli KAM3 carries a deletion in the multidrug effluxsystem genes acrAB and therefore is markedly sensitiveto triphenylmethane dyes (Morita et al., 1998). E. colitransformants were selected on LB medium containing 10 mgCV ml21. Transformants were obtained for the plasmidspKV11, pKV29 and pKV36 of the isolates KV11, KV29 andKV36, respectively. Accordingly, these three plasmids con-ferred CV resistance to E. coli, suggesting that they carry acorresponding resistance gene. Moreover, E. coli derivativescarrying the plasmids pKV11, pKV29 and pKV36 were ableto decolorize CV, which was apparent as a colourless halosurrounded growing colonies on agar plates containing CV(10 mg ml21). The plasmids of the isolates KV10, KV16 andKV17 could not be transformed into E. coli. Either theplasmids were too large to be efficiently transformed orreplication of these plasmids was not compatible with the E.coli background. It is also possible that the CV-resistancephenotype of strains KV10, KV16 and KV17 is encoded onthe chromosome and not on the plasmids. Since theinvolvement of plasmids pKV10, pKV16 and pKV17 in CVdecolorization was unclear, further analyses focused on theplasmids pKV11, pKV29 and pKV36.

IncP-1b plasmids pKV11, pKV29 and pKV36 arerelated to the reference plasmid pGNB1

The CV-resistance reference plasmid pGNB1 (Schluteret al., 2007) belongs to the IncP-1b incompatibility group.

Table 1. Taxonomic affiliation of activated sludge isolates with the ability to decolorize CV

Isolate RDP-Classifier

assignment*D

Best BLASTN hit Identity Origin of reference strain GenBank

accession no.

KV10 Delftia (100 %) Delftia sp. XYJ6 1248/1249 Sediment of wastewater treatment in a pesticide

plant, Beijing, China

EU707799

KV11 Comamonas (100 %) Comamonas sp. H4-3 1244/1245 Community using poly(3-hydroxybutyrate-co-3-

hydroxyvalerate) in solid-phase denitrification

processes

AB277848

KV16 Delftia (100 %) Delftia sp. JDC-3 1249/1249 – FJ378038

KV17 Delftia (100 %) Delftia sp. XYJ6 1248/1249 Sediment of wastewater treatment in a pesticide

plant, Beijing, China

EU707799

KV29 Delftia (100 %) Bacterium SE4 1248/1249 Gastrointestinal tract of Epinephelus coioides EU520344

KV36 Comamonas (100 %) Comamonas sp. H4-3 1244/1245 Community using poly(3-hydroxybutyrate-co-3-

hydroxyvalerate) in solid-phase denitrification

processes

AB277848

*All isolates were classified as belonging to the class b-Proteobacteria, the order Burkholderiales and the family Comamonadaceae by means of the

RDP-Classifier.

DAll RDP-Classifier results got the confidence value 100 %.

Crystal violet resistance IncP-1b plasmids

http://mic.sgmjournals.org 2063

Downloaded from www.microbiologyresearch.org by

IP: 54.160.113.5

On: Sun, 29 May 2016 21:00:06

Therefore, we tested whether the plasmids pKV11, pKV29and pKV36 also belong to this group. IncP-1-specificreplication initiation trfA fragments were amplified bymeans of the trfA-specific primers trfA-1 and trfA-2 (Gotzet al., 1996). Amplicons were obtained for all threeplasmids, and sequencing of the trfA amplicons revealedthat the fragments of all plasmids are almost identical tocorresponding sequences of the CV-resistance referenceplasmid pGNB1 and the Burkholderia ambifaria AMMDplasmid 1 (see Table 2). The latter plasmid also is anIncP-1 plasmid, but its precise function has not yet beendetermined (GenBank accession no. CP000443). Thus, theplasmids pKV11, pKV29 and pKV36 clearly belong to theIncP-1b incompatibility group and their trfA nucleotidesequences cluster within the subgroup of plasmidsrepresented by the catabolic plasmids pGNB1, pJP4(Trefault et al., 2004), pUO1 (Sota et al., 2003) andpADP-1 (Martinez et al., 2001), the antibiotic-resistanceplasmids pB8 (Schluter et al., 2005) and pB10 (Schluteret al., 2003), and the mercury-resistance plasmid pTP6(Smalla et al., 2006).

As plasmids pKV11, pKV29 and pKV36 are of differentsizes, it is assumed that they mainly differ in their accessorymodules.

Complete sequencing of the IncP-1b plasmidpKV29 reveals that it contains an accessory CV-resistance gene region

Plasmid pKV29 was sequenced at the DOE Joint GenomeInstitute (JGI) on the GS FLX platform. The finishedplasmid genome appeared to be 61 667 bp in size, with anaverage GC content of 64.0 %. It contains 69 predictedprotein-coding sequences (see Fig. 1).

Plasmid pKV29 consists of eight backbone modules involvedin replication initiation (ssb–trfA), plasmid stabilization,maintenance and control (klu, klc, kle, incC–korB and kfr),conjugative plasmid transfer (tra) and mating pair forma-tion (trb) (Fig. 1). These backbone modules are very similaror identical to corresponding modules of other IncP-1bplasmids such as the atrazine catabolic plasmid pADP-1(NC_004956), the B. ambifaria AMMD plasmid 1 (NC_008385) and different antibiotic-resistance plasmids (Table3). Plasmid pKV29 harbours two accessory regions locateddownstream of the replication initiation gene trfA and

between the tra and trb modules, respectively. The smalleraccessory region downstream of the trfA gene contains sevengenes and is flanked by inverted repeats related to those oftransposon Tn5501 (inverted repeat Tn5501 in Fig. 1). Theelement inserted into pKV29 is homologous to the Tn5501-like element of the reference plasmid pGNB1 (Schluter et al.,2007) and encodes a conserved barrel protein belonging tothe Cupin 2 family and a putative addiction system (parE–orf1).

The second accessory region on pKV29 contains the tmrgene, encoding a putative triphenylmethane reductase(Tmr). The tmr gene region features a particularly lowGC content (minimum ~34 %) compared with other partsof the plasmid, suggesting that this segment was acquiredby the plasmid only recently. ORFs located up- anddownstream of tmr encode hypothetical proteins that areidentical to corresponding gene products of the referenceplasmid pGNB1 (Fig. 2). The second pKV29 accessoryregion is flanked by a conserved module that consists of acopy of the insertion sequence IS1071, a truncated 9lpdAgene (dihydrolipoamide dehydrogenase) and a transposasegene (tnpA) of another insertion sequence element(ISGNB1-2). This module is also conserved in the referenceplasmid pGNB1. The central part of the second pKV29accessory region is composed of six coding sequences. Mostof the corresponding gene products are hypothetical; onlythe product of the cobA gene was predictable: it encodes aputative uroporphyrinogen-III C-methyltransferase thatis 84 % similar to a corresponding enzyme of the Hydramagnipapillata symbiont Curvibacter sp. (Comamona-daceae) (Chapman et al., 2010).

pKV11 and pKV36 accessory gene regionsencoding the triphenylmethane reductase genetmr are related but not identical

As deduced from partial sequencing of plasmid-specificgenes, the backbone regions of the CV-resistance plasmidspKV11 and pKV36 are very similar in comparison withplasmid pKV29. Accordingly, only the accessory regions ofthese plasmids were sequenced. The pKV11 CV-resistanceregion was subcloned on an 11 kb BamHI and an over-lapping 5.5 kb SphI restriction fragment. Both fragmentsmediated CV resistance and decolorization in E. coliDH5a. Therefore it was decided to determine the complete

Table 2. Comparison of replication initiation trfA amplicon sequences for the plasmids pKV11, pKV29 and pKV36 withcorresponding database entries

Plasmid Best BLASTN hit Identity Origin of reference plasmid Reference or accession no.

pKV11 pGNB1 232/234 Uncultured bacterium from activated sludge Schluter et al. (2007)

Plasmid 1 232/234 B. ambifaria AMMD CP000443

pKV29 pGNB1 233/234 Uncultured bacterium from activated sludge Schluter et al. (2007)

Plasmid 1 233/234 B. ambifaria AMMD CP000443

pKV36 pGNB1 233/234 Uncultured bacterium from activated sludge Schluter et al. (2007)

Plasmid 1 233/234 B. ambifaria AMMD CP000443

Y. Stolze and others

2064 Microbiology 158

Downloaded from www.microbiologyresearch.org by

IP: 54.160.113.5

On: Sun, 29 May 2016 21:00:06

nucleotide sequences of both restriction fragments. TheSphI fragment contains the IncP-1-specific mating pairformation gene trbN followed by a truncated version oftrbO. The arrangement of these genes is exactly the same asdescribed for the reference plasmid pGNB1 (Schluter et al.,2007). Downstream of the mating pair formation modulegenes on plasmid pKV11 there are four additional genes

encoding a putative transcriptional regulator protein (Orf14),a conserved hypothetical protein (Orf13), a triphenylme-thane reductase (Tmr) and a hypothetical protein (Orf11)(Fig. 2). The arrangement and sequences of these four genesare identical to those of the corresponding genes on plasmidpGNB1. The described module is flanked by an insertion of atruncated ISGNB1-2 element. Differences between plasmids

Fig. 1. Genetic map of the IncP-1b CV-resistance plasmid pKV29. Coding regions are shown by arrows, indicating thedirection of transcription. Coding sequences representing a specific plasmid module are differentiated by different colours. Theorigin of vegetative replication (oriV) is symbolized by a black circle, inverted repeats by black bars. GC content (orange/blue)and GC skew (red/green) are plotted on the inner circles. The first accessory plasmid region (red) located downstream of thereplication module (ssb–trbA) is flanked by a truncated and a complete transposable element (tnpA) and contains five furthergenes similar to corresponding genes of the reference CV-resistance plasmid pGNB1 (Schluter et al., 2007). The secondaccessory region of the plasmid (light green) located downstream of the tra module includes the triphenylmethane reductasegene tmr, which is involved in decolorization of triphenylmethane dyes.

Crystal violet resistance IncP-1b plasmids

http://mic.sgmjournals.org 2065

Downloaded from www.microbiologyresearch.org by

IP: 54.160.113.5

On: Sun, 29 May 2016 21:00:06

pGNB1 and pKV11 were found beyond the ISGNB1-2insertion. Plasmid pKV11 possesses three genes in thisregion which have no counterparts on pGNB1. These genesencode a protein of the glucose-methanol-choline (GMC)oxidoreductase family (Cavener, 1992), a conserved hypo-thetical protein and a putative long-chain fatty acid-CoAligase. The module described above is flanked by a 39-truncated transposase gene of ISGNB1-2. Since plasmidpKV11 contains two copies of ISGNB1-2, homologousrecombination via these elements could lead to deletion ofthe GMC oxidoreductase and the long-chain fatty acid-CoAligase genes, which is reminiscent of the gene arrangement inthe accessory region L2 on plasmid pGNB1 (Schluteret al., 2007). This plasmid contains only a single copy ofISGNB1-2. Accordingly, plasmid pGNB1 can be regarded asa deletion derivative of plasmid pKV11, at least as far as theaccessory region L2 is considered. The pKV11 segmentconsisting of IS1071, 9lpdA (truncated dihydrolipoamidedehydrogenase gene) and ISGNB1-2 is completely identicalto the corresponding region on the reference plasmidpGNB1. Moreover, plasmids pKV11 and pGNB1, and plas-mid 1 of B. ambifaria AMMD (accession no. CP000443),share the same insertion site of the insertion sequenceIS1071.

In contrast, the accessory CV-resistance region of plasmidpKV36 (Fig. 2) is more related to the corresponding regionof plasmid pKV29. The tmr module and the IS1071–9lpdAgene region are conserved on pKV36. In addition, thisplasmid also contains a cobA module composed of sixgenes that are also present on plasmid pKV29 (Fig. 2).However, plasmid pKV36 possesses accessory genes that

are not present in any sequenced accessory region of theother plasmids. These pKV36 genes encode a short-chaindehydrogenase/reductase (sdr) and a hypothetical proteinwith highest similarity (79 %) to the Comamonas testoster-oni CNB-2 protein TctC (Ma et al., 2009). TctC is thoughtto be the extra-cytoplasmic receptor component of thetripartite tricarboxylate transporter family (pfam03401).The gene products of two further coding sequences in thisregion are hypothetical. Moreover, plasmid pKV36 har-bours a second truncated module consisting of an incom-plete IS1071 element, 9lpdA and a partial copy of ISGNB1-2. Homologous recombination via two identical insertionsequence elements can potentially convert plasmid pKV36into a pKV29-like or a pGNB1-like plasmid, respectively.In summary, the accessory CV-resistance regions of plas-mids pKV11, pGNB1, pKV29 and pKV36 are related toeach other and are most probably derived from a commonancestor.

The plasmid-encoded triphenylmethanereductase (Tmr) is responsible for decolorizationof CV

The triphenylmethane reductase (Tmr) encoded on each ofthe pKV plasmids was thought to be the key enzyme fordecolorization of CV. To analyse whether other genesidentified on the pKV plasmids are involved in degradationof CV, decolorization assays were carried out. As a control,plasmid pIK405, containing a cloned copy of the completetmr gene originating from plasmid pKV29, was includedin the assays. E. coli DH5a strains harbouring plasmids

Table 3. Comparison of pKV29 backbone modules with corresponding modules of other IncP-1b catabolic, antibiotic-resistance andmercury-resistance plasmids

pKV29 backbone

module*

Included genes Identity with:

AMMD Plasmid 1D pB8d pB10§ pTP6|| pADP-1

tra traC–traO 99.98 % 99.28 % 99.89 % 99.98 % 99.99 %

trb trbA–trbO 99.99 % 99.91 % 99.90 % 99.96 % 99.77 %

trfA ssb, trfA 100 % 99.87 % 99.81 % 99.87 % 99.77 %

klu kluA, kluB 100 % 99.66 % 2# 99.66 % 100 %

klc klcA, klcB, klcC 99.95 % 99.84 % 85.53 % 99.90 % 98.46 %

kor korA, incC, incC2, korB 99.91 % 99.73 % 96.44 % 99.82 % 99.86 %

kle kleA, kleB, kleE, kleF 100 % 99.93 % 89.36 % 99.93 % 100 %

kfr kfrA, kfrB, kfrC 100 % 99.96 % 99.78 % 100 % 100 %

*Plasmid backbone modules have the following functions: tra, conjugative transfer; trb, mating pair formation; trfA, replication initiation; klu,

putative post-segregational killing system; klc, antirestriction and plasmid control; kor, stable inheritance and plasmid control; kle, stable

inheritance; kfr, plasmid stability.

DB. ambifaria AMMD IncP-1b plasmid 1, NC_008385 (unpublished results).

dUncultured bacterium, IncP-1b multiresistance plasmid pB8, NC_007502 (Schluter et al., 2005).

§Uncultured bacterium IncP-1b multiresistance plasmid pB10, NC_004840 (Schluter et al., 2003).

||Uncultured bacterium IncP-1b mercury-resistance plasmid pTP6, NC-007680 (Smalla et al., 2006).

Pseudomonas sp. ADP IncP-1b atrazine catabolic plasmid pADP-1, NC_004956 (Martinez et al., 2001).#This module is not present on the plasmid.

Y. Stolze and others

2066 Microbiology 158

Downloaded from www.microbiologyresearch.org by

IP: 54.160.113.5

On: Sun, 29 May 2016 21:00:06

pGNB1, pKV11, pKV29, pKV36 or pIK405 were cultivatedin M9 minimal medium supplemented with CV to a finalconcentration of 10 mg ml21. Samples (1 ml) were takenevery 2 h and centrifuged, and the supernatant was mea-sured by spectrophotometry at 200–700 nm. CV featurescharacteristic absorption peaks at 212, 248, 303 and590 nm (Fig. 3a). Plasmid-containing E. coli DH5a strainswere able to decolorize the medium within 10 h (Fig. 3b).The decolorization capacity was comparable for all E. coliderivatives tested, irrespective of the plasmid harbouredby the cells. Also, plasmid pIK405, which contains thetmr gene alone, conferred decolorization ability upon E.coli DH5a. This observation indicates that the triphenyl-methane reductase gene tmr is necessary and sufficient fordecolorization of CV in E. coli. The plasmid-free E. coliDH5a parental strain was unable to grow in the mediumsupplemented with CV and did not decolorize the dye. Itappeared that none of the accessory genes identified on

plasmids pKV11, pKV29, pKV36 and pGNB1 affected thedecolorization efficiency of the corresponding E. coli DH5a

derivatives. Accordingly, the triphenylmethane reductase(Tmr) is the key enzyme for decolorization of CV or atleast catalyses the first step in this pathway. These obser-vations are in line with earlier results obtained by Jang et al.(2005). These authors showed that a purified triphenyl-methane reductase from a Citrobacter sp. strain was able tocatalyse the NADH-dependent reduction of triphenyl-methane dyes.

To compare the decolorization rate of E. coli DH5a (pKV29)with those of the original isolates Comamonas sp. KV11,Delftia sp. KV29 and Comamonas sp. KV39, decolorizationassays were also done for these strains. All isolates were ableto decolorize the medium (LB) containing CV (10 mg ml21)within 6 h. The maximum degradation rates varied between0.957 and 1.258 mg 121 h21 (see Table 4), indicating that

Fig. 2. Comparison of the accessory regions that include the triphenylmethane reductase gene tmr of plasmids pKV11,pGNB1, pKV29 and pKV36. Annotated coding sequences are displayed as arrows and inverted repeats of insertion sequencesas green rectangles. Coding sequences are coloured based on their modular assignment: lilac, triphenylmethane reductasegene tmr; green, transposase gene tnpA of the insertion sequence ISGNB1-2; yellow, truncated dihydrolipoamidedehydrogenase gene 9lpdA; blue, transposase gene tnpA of the insertion sequence IS1071; orange, module comprising aGMC oxidoreductase gene gmc and putative long-chain fatty acid-CoA ligase gene; red, module containing a putativeuroporphyrinogen-III C-methyltransferase gene, cobA; light blue, module consisting of five coding sequences of unknownfunction. Coding sequences located upstream and downstream of the tmr gene are shown in white. Regions highlighted in greyindicate homologous gene clusters that were determined by calculating the largest common homologous regions applying theBLASTN algorithm.

Crystal violet resistance IncP-1b plasmids

http://mic.sgmjournals.org 2067

Downloaded from www.microbiologyresearch.org by

IP: 54.160.113.5

On: Sun, 29 May 2016 21:00:06

the performance of all strains regarding CV degradation wasin the same range.

Identification of CV degradation metabolites

Metabolites of the CV degradation process were analysedby UHPLC-ESI-Q-TOF-MS. Leuco crystal violet (leuco-CV), a reduced and colourless form of CV, was previouslyidentified as the first product of CV degradation. Theenzyme triphenylmethane reductase (Tmr) catalyses theconversion of CV to leuco-CV (Jang et al., 2005). Furtherpossible degradation metabolites of CV are MK as wellas 4-dimethylaminobenzaldehyde (Yatome et al., 1992).Accordingly, these compounds were used as referencesubstances in liquid chromatography-MS (LC-MS) mea-surements (Fig. 4). Supernatants of decolorization assayswith E. coli DH5a derivatives carrying plasmids pGNB1,

pKV11, pKV29, pKV36 or pIK405 were analysed byUHPLC-ESI-Q-TOF-MS to elucidate the plasmid-encodedCV degradation pathway. The CV degradation metaboliteprofiles are shown as base peak chromatograms (BPCs)and as extracted ion chromatograms (EICs), also includingextracted masses for reference substances. The EICs of allplasmid-mediated degradation reactions revealed the samecomposition and quantity of degradation products in thesupernatants extracted. An example of a metabolite pro-file is shown for E. coli DH5a carrying plasmid pIK405,which only encodes the triphenylmethane reductase genetmr (Fig. 4). Leuco-CV, MK and 4-dimethylaminobenzal-dehyde were found in the culture supernatants afterdegradation of CV. The CV peak itself was no longerdetectable after 10 h of incubation, indicating quantitativedegradation of CV. The doubly protonated form of leuco-CV had a low peak intensity and was found to elute at2.3 min (Fig. 4d, 1). Peak 2 had a similar accurate mass andthe same elemental compositon (C9H12N1O1) as peak 3.The latter had a retention time of 3.17 min and corres-ponds to p-dimethylaminobenzaldehyde, as demonstratedin comparison with the purified standard (Fig. 4h). TheMK standard eluted at 4.55 min (Fig. 4g). Dimethyla-minobenzaldehyde and MK peak intensities were lower forthe E. coli (pIK405) supernatant (Fig. 4d) as compared withthe leuco-CV peak. This observation suggests that leuco-CV was degraded into further compounds. Due to itsrelatively high peak intensity, it might be suggested thatMK is one of the end-products of this reaction cascade inE. coli, whereas 4-dimethylaminobenzaldehyde may befurther degraded into compounds that so far cannot beidentified (Fig. 5). Since the metabolite profile of the E. coli

Fig. 3. Plasmid-mediated decolorization of CV in E. coli DH5a. (a) Absorption spectrum of CV (5 mg ml”1) in M9 minimalmedium. (b) Decolorization assay of an E. coli DH5a (pIK405) culture in M9 medium containing CV (5 mg ml”1) incubated at37 6C for 10 h. Aliquots of the culture (1 ml) were taken every 2 h, centrifuged and measured by UV spectrophotometry (200–700 nm).

Table 4. CV degradation rates of the original isolatesComamonas sp. KV11, Delftia sp. KV29 and Comamonas

sp. KV36 in comparison with E. coli DH5a (pKV29)

Strain Maximum CV degradation

rate (mg l”1 h”1)

Comamonas sp. KV11 1.258

Delftia sp. KV29 0.957

Comamonas sp. KV36 0.974

E. coli DH5a (pKV29) 1.063

Y. Stolze and others

2068 Microbiology 158

Downloaded from www.microbiologyresearch.org by

IP: 54.160.113.5

On: Sun, 29 May 2016 21:00:06

DH5a derivative harbouring plasmid pIK405, which onlycontains the triphenylmethane reductase gene tmr, is indis-tinguishable from those of the other strains, it has to beconcluded that enzymes encoded in the E. coli chro-mosome are involved in the leuco-CV degradation cascade.The NADH-dependent reduction of CV to leuco-CV by theenzyme triphenylmethane reductase (Tmr) has been de-monstrated unambiguously in vitro (Jang et al., 2005).Currently, there are no indications that triphenylmethanereductase possesses further enzymic activity except for re-duction of triphenylmethane dyes. However, E. coli is knownfor its capacity to degrade certain aromatic compounds (Dıazet al., 2001), although candidate enzymes involved in leuco-CV decomposition have not been identified so far.

Diversity and evolution of WWTP plasmidsconferring degradation of CV

Specific autochthonous bacteria residing in the activatedsludge community of a municipal WWTP are able todecolorize and thereby at least to some extent detoxifythe triphenylmethane dye CV. For the bacterial isolatesdescribed in this study, resistance to and decolorization ofthe dye are associated with the presence of the plasmid-borne triphenylmethane reductase gene tmr. This gene hasprobably been captured by the plasmids only recently, sinceits G+C content differs considerably from the rest of theplasmid. Interestingly, tmr is the only gene on the plasmidsanalysed that has a proven function in triphenylmethane

Fig. 4. Analysis of CV degradation products byUHPLC-ESI-Q-TOF-MS. BPC (a) and EIC[(b), with selected molecular masses of ref-erence compounds] of supernatants origin-ating from a culture of the plasmid-free E. coli

strain DH5a in M9 medium after 10 h ofincubation at 37 6C. BPC (c) and EIC (d) withselected molecular masses of reference com-pounds of supernatants originating from aculture of the E. coli strain DH5a (pIK405) inM9 medium containing CV (80 mg ml”1) after10 h of incubation at 37 6C. (e–h) BPCs ofthe reference compounds CV [(e), peak A],leuco-CV [(f), peak B], MK [(g), peak C] and4-dimethylaminobenzaldehyde [(h), peak D]

diluted in 0.01 M acetonitrile.

Fig. 5. Putative CV degradation pathway of E. coli DH5a harbouring plasmid pKV11, pKV29, pKV36 or pIK405. The first step inCV degradation is catalysed by the plasmid-encoded enzyme triphenylmethane reductase (Tmr), leading to the formation ofleuco-CV. Further degradation steps are thought to be accomplished by chromosomally encoded E. coli enzymes involved inutilization of aromatic compounds. It is possible that MK is an end product of the degradation pathway in E. coli, whereas4-dimethylaminobenzaldehyde may be further degraded to so-far-unidentified compounds.

Crystal violet resistance IncP-1b plasmids

http://mic.sgmjournals.org 2069

Downloaded from www.microbiologyresearch.org by

IP: 54.160.113.5

On: Sun, 29 May 2016 21:00:06

dye degradation. Obviously, the corresponding gene pro-duct triphenylmethane reductase renders the triphenyl-methane molecule accessible for further degradation byenzymes that are not encoded on the plasmids. Mostprobably, chromosomally encoded enzymes functioning inaromatic compound degradation accomplish the furtherdecomposition of the leuco form of CV in E. coli DH5acarrying the CV-resistance plasmids pKV11, pKV29 orpKV36. Currently, the original function of these plasmidsremains unknown. In addition to the tmr gene, all the pla-smids possess further accessory genes. However, althoughfor some of these genes functional predictions wereobtained, it was not possible to deduce a general functionalcontext or metabolic pathway in which corresponding geneproducts were involved. It cannot be excluded that enzymicfunctions encoded on the plasmids complement chromo-somally encoded metabolic pathways of the host bac-terium. Contrary to this assumption is the fact that IncP-1plasmids usually possess a very broad host range. For theIncP-1b CV-resistance plasmid pGNB1 it has been shownthat it is self-transmissible and can be conjugated to a- andc-proteobacteria, and therefore is a broad-host-range plas-mid (Schluter et al., 2007). It is unlikely that plasmid-encoded traits are compatible with different chromosomalbackgrounds. However, the CV-resistance and decolori-zation phenotype mediated by plasmid pGNB1 is alsoexpressed in the a-proteobacterium Sinorhizobium meliloti(Schluter et al., 2007), indicating that the promoter of thetriphenylmethane reductase gene is functional in differenthost bacteria. The activity of the triphenylmethane reduc-tase seems to be very specific, since it accepts malachitegreen, CV and basic fuchsin as substrates, but it does notcatalyse the reduction of brilliant green, bromophenol blueand azo dyes (Jang et al., 2005). However, it is conceiva-ble that triphenylmethane reductase is also able to reducenaturally occurring compounds featuring a chemical struc-ture that is similar to that of triphenylmethane dyes. Insummary, reduction of triphenylmethane dyes seems to bea secondary function of the plasmids analysed in this study.The primary function of these plasmids remains unknownso far.

Comparative analysis of the accessory regions of theplasmids pKV11, pKV29, pKV36 and pGNB1 helped inunderstanding the evolution of these plasmids. Obviously,insertion sequences played a role in the acquisition of newplasmid modules, since recognizable units are flanked byeither ISGNB1-2 or IS1071. Recombination events viaidentical insertion sequence elements can lead to deletionor inversion of certain accessory modules, depending onthe orientation of flanking insertion sequences. Since alltransposase genes (tnpA) of the insertion sequence ele-ments located on the plasmids analysed here are tran-scribed in the same direction, homologous recombinationvia identical insertion sequence copies will cause deletionof modules. Accordingly, it is assumed that the plasmidspKV11, pKV29, pKV36 and pGNB1 originate from a com-mon ancestor which step by step acquired additional

modules, to develop the composition that has beenachieved in plasmid pKV36, for example. Thus, insertionsequence elements in accessory plasmid modules ensureflexibility and facilitate rapid adaptive responses tochanging environmental conditions. However, the drivingforce(s) in the evolution of the pKV plasmids is not yetclear. In any case, acquisition of the triphenylmethanereductase gene (tmr) seems to be an important evolu-tionary event, since all plasmids analysed in this studycontain this gene, suggesting that it represents an adaptiveadvantage.

In conclusion, WWTP bacteria harbour diverse mobilegenetic elements specifying degradation capabilities thatmight be exploitable in biotechnological processes designedfor the bioremediation of dye-contaminated industrialwastes.

ACKNOWLEDGEMENTS

F. E. and D. W. acknowledge the receipt of a scholarship from the

CLIB Graduate Cluster ‘Industrial Biotechnology’ co-financed by theMinistry of Innovation of North Rhine-Westphalia. The bioinfor-

matics support of the Bioinformatics Resource Facility (BRF) at the

Center for Biotechnology (CeBiTec, Bielefeld University) is gratefully

acknowledged. A. S. gratefully acknowledges the METAEXPLORE

grant of the European Commission (KBBE-222625). This work was

also in part funded by the National Science Foundation grant EF-

0627988 to E. M. T. We are grateful to Kerrie Barry, Brian Foster andAlla Lapidus at the US Department of Energy (DOE) Joint Genome

Institute (JGI) for providing draft genome sequences, supported by

the DOE Office of Science under contract no. DE-AC02-05CH11231.

REFERENCES

Altschul, S. F., Madden, T. L., Schaffer, A. A., Zhang, J., Zhang, Z.,Miller, W. & Lipman, D. J. (1997). Gapped BLAST and PSI-BLAST: a new

generation of protein database search programs. Nucleic Acids Res 25,

3389–3402.

Au, W., Pathak, S., Collie, C. J. & Hsu, T. C. (1978). Cytogenetic

toxicity of gentian violet and crystal violet on mammalian cells in

vitro. Mutat Res 58, 269–276.

Azmi, W., Sani, R. K. & Banerjee, U. C. (1998). Biodegradation of

triphenylmethane dyes. Enzyme Microb Technol 22, 185–191.

Bumpus, J. A. & Brock, B. J. (1988). Biodegradation of crystal violet

by the white rot fungus Phanerochaete chrysosporium. Appl Environ

Microbiol 54, 1143–1150.

Cavener, D. R. (1992). GMC oxidoreductases. A newly defined family

of homologous proteins with diverse catalytic activities. J Mol Biol

223, 811–814.

Chapman, J. A., Kirkness, E. F., Simakov, O., Hampson, S. E., Mitros,T., Weinmaier, T., Rattei, T., Balasubramanian, P. G., Borman, J. &other authors (2010). The dynamic genome of Hydra. Nature 464,

592–596.

Chen, C. C., Liao, H. J., Cheng, C. Y., Yen, C. Y. & Chung, Y. C. (2007).Biodegradation of crystal violet by Pseudomonas putida. Biotechnol

Lett 29, 391–396.

Chen, C. H., Chang, C. F., Ho, C. H., Tsai, T. L. & Liu, S. M. (2008).Biodegradation of crystal violet by a Shewanella sp. NTOU1. Chemos-

phere 72, 1712–1720.

Y. Stolze and others

2070 Microbiology 158

Downloaded from www.microbiologyresearch.org by

IP: 54.160.113.5

On: Sun, 29 May 2016 21:00:06

Dıaz, E., Ferrandez, A., Prieto, M. A. & Garcıa, J. L. (2001).Biodegradation of aromatic compounds by Escherichia coli. Micro-

biol Mol Biol Rev 65, 523–569.

Docampo, R. & Moreno, S. N. (1990). The metabolism and mode of

action of gentian violet. Drug Metab Rev 22, 161–178.

Eikmeyer, F., Hadiati, A., Szczepanowski, R., Wibberg, D.,Schneiker-Bekel, S., Rogers, L. M., Brown, C. J., Top, E. M.,Puhler, A. & Schluter, A. (2012). The complete genome sequences

of four new IncN plasmids from wastewater treatment plant effluent

provide new insights into IncN plasmid diversity and evolution.

Plasmid 68, 13–24.

Gordon, D., Abajian, C. & Green, P. (1998). Consed: a graphical tool

for sequence finishing. Genome Res 8, 195–202.

Gordon, D., Desmarais, C. & Green, P. (2001). Automated finishing

with Autofinish. Genome Res 11, 614–625.

Gotz, A., Pukall, R., Smit, E., Tietze, E., Prager, R., Tschape, H., vanElsas, J. D. & Smalla, K. (1996). Detection and characterization of

broad-host-range plasmids in environmental bacteria by PCR. Appl

Environ Microbiol 62, 2621–2628.

Grant, S. G., Jessee, J., Bloom, F. R. & Hanahan, D. (1990).Differential plasmid rescue from transgenic mouse DNAs into

Escherichia coli methylation-restriction mutants. Proc Natl Acad Sci

U S A 87, 4645–4649.

Heinl, S., Wibberg, D., Eikmeyer, F., Szczepanowski, R., Blom, J.,Linke, B., Goesmann, A., Grabherr, R., Schwab, H. & other authors(2012). Insights into the completely annotated genome of Lacto-

bacillus buchneri CD034, a strain isolated from stable grass silage.

J Biotechnol. (Epub ahead of print).

Hynes, M. F., Simon, R. & Puhler, A. (1985). The development of

plasmid-free strains of Agrobacterium tumefaciens by using incom-

patibility with a Rhizobium meliloti plasmid to eliminate pAtC58.

Plasmid 13, 99–105.

Jang, M. S., Lee, Y. M., Kim, C. H., Lee, J. H., Kang, D. W., Kim, S. J. &Lee, Y. C. (2005). Triphenylmethane reductase from Citrobacter sp.

strain KCTC 18061P: purification, characterization, gene cloning, and

overexpression of a functional protein in Escherichia coli. Appl Environ

Microbiol 71, 7955–7960.

Jones, J. J. & Falkinham, J. O., III (2003). Decolorization of malachite

green and crystal violet by waterborne pathogenic mycobacteria.

Antimicrob Agents Chemother 47, 2323–2326.

Kwasniewska, K. (1985). Biodegradation of crystal violet (hexam-

ethyl-p-rosaniline chloride) by oxidative red yeasts. Bull Environ

Contam Toxicol 34, 323–330.

Littlefield, N. A., Blackwell, B. N., Hewitt, C. C. & Gaylor, D. W. (1985).Chronic toxicity and carcinogenicity studies of gentian violet in mice.

Fundam Appl Toxicol 5, 902–912.

Ma, Y. F., Zhang, Y., Zhang, J. Y., Chen, D. W., Zhu, Y., Zheng, H.,Wang, S. Y., Jiang, C. Y., Zhao, G. P. & Liu, S. J. (2009). The

complete genome of Comamonas testosteroni reveals its genetic

adaptations to changing environments. Appl Environ Microbiol 75,

6812–6819.

Margulies, M., Egholm, M., Altman, W. E., Attiya, S., Bader, J. S.,Bemben, L. A., Berka, J., Braverman, M. S., Chen, Y. J. & otherauthors (2005). Genome sequencing in microfabricated high-density

picolitre reactors. Nature 437, 376–380.

Martinez, B., Tomkins, J., Wackett, L. P., Wing, R. & Sadowsky, M. J.(2001). Complete nucleotide sequence and organization of the

atrazine catabolic plasmid pADP-1 from Pseudomonas sp. strain

ADP. J Bacteriol 183, 5684–5697.

Meyer, F., Goesmann, A., McHardy, A. C., Bartels, D., Bekel, T.,Clausen, J., Kalinowski, J., Linke, B., Rupp, O. & other authors

(2003). GenDB—an open source genome annotation system for

prokaryote genomes. Nucleic Acids Res 31, 2187–2195.

Mittal, A., Mittal, J., Malviya, A., Kaur, D. & Gupta, V. K. (2010).Adsorption of hazardous dye crystal violet from wastewater by waste

materials. J Colloid Interface Sci 343, 463–473.

Morita, Y., Kodama, K., Shiota, S., Mine, T., Kataoka, A., Mizushima,T. & Tsuchiya, T. (1998). NorM, a putative multidrug efflux protein,

of Vibrio parahaemolyticus and its homolog in Escherichia coli.

Antimicrob Agents Chemother 42, 1778–1782.

Ren, S., Guo, J., Zeng, G. & Sun, G. (2006). Decolorization of

triphenylmethane, azo, and anthraquinone dyes by a newly isolated

Aeromonas hydrophila strain. Appl Microbiol Biotechnol 72, 1316–

1321.

Sambrook, J., Fritsch, E. F. & Maniatis, T. (1989). Molecular Cloning:

a Laboratory Manual, 2nd edn. Cold Spring Harbor, NY: Cold Spring

Harbor Laboratory.

Sani, R. K. & Banerjee, U. C. (1999). Decolorization of triphenyl-

methane dyes and textile and dye-stuff effluent by Kurthia sp. Enzyme

Microb Technol 24, 433–437.

Schluter, A., Heuer, H., Szczepanowski, R., Forney, L. J., Thomas,C. M., Puhler, A. & Top, E. M. (2003). The 64 508 bp IncP-1b antibiotic

multiresistance plasmid pB10 isolated from a waste-water treat-

ment plant provides evidence for recombination between members of

different branches of the IncP-1b group. Microbiology 149, 3139–

3153.

Schluter, A., Heuer, H., Szczepanowski, R., Poler, S. M., Schneiker,S., Puhler, A. & Top, E. M. (2005). Plasmid pB8 is closely related to the

prototype IncP-1b plasmid R751 but transfers poorly to Escherichia

coli and carries a new transposon encoding a small multidrug resis-

tance efflux protein. Plasmid 54, 135–148.

Schluter, A., Krahn, I., Kollin, F., Bonemann, G., Stiens, M.,Szczepanowski, R., Schneiker, S. & Puhler, A. (2007). IncP-1b

plasmid pGNB1 isolated from a bacterial community from a

wastewater treatment plant mediates decolorization of triphenyl-

methane dyes. Appl Environ Microbiol 73, 6345–6350.

Smalla, K., Haines, A. S., Jones, K., Krogerrecklenfort, E., Heuer, H.,Schloter, M. & Thomas, C. M. (2006). Increased abundance of IncP-

1b plasmids and mercury resistance genes in mercury-polluted river

sediments: first discovery of IncP-1b plasmids with a complex mer

transposon as the sole accessory element. Appl Environ Microbiol 72,

7253–7259.

Sota, M., Kawasaki, H. & Tsuda, M. (2003). Structure of

haloacetate-catabolic IncP-1b plasmid pUO1 and genetic mobility

of its residing haloacetate-catabolic transposon. J Bacteriol 185,

6741–6745.

Szczepanowski, R., Krahn, I., Bohn, N., Puhler, A. & Schluter, A.

(2007). Novel macrolide resistance module carried by the IncP-1b

resistance plasmid pRSB111, isolated from a wastewater treatment

plant. Antimicrob Agents Chemother 51, 673–678.

Szczepanowski, R., Eikmeyer, F., Harfmann, J., Blom, J., Rogers, L. M.,Top, E. M. & Schluter, A. (2011). Sequencing and comparative analysis

of IncP-1a antibiotic resistance plasmids reveal a highly conserved

backbone and differences within accessory regions. J Biotechnol 155,

95–103.

Trefault, N., De la Iglesia, R., Molina, A. M., Manzano, M., Ledger,T., Perez-Pantoja, D., Sanchez, M. A., Stuardo, M. & Gonzalez, B.(2004). Genetic organization of the catabolic plasmid pJP4 from

Ralstonia eutropha JMP134 (pJP4) reveals mechanisms of adapta-

tion to chloroaromatic pollutants and evolution of specialized

chloroaromatic degradation pathways. Environ Microbiol 6, 655–

668.

Crystal violet resistance IncP-1b plasmids

http://mic.sgmjournals.org 2071

Downloaded from www.microbiologyresearch.org by

IP: 54.160.113.5

On: Sun, 29 May 2016 21:00:06

Wang, Q., Garrity, G. M., Tiedje, J. M. & Cole, J. R. (2007). NaiveBayesian classifier for rapid assignment of rRNA sequences into thenew bacterial taxonomy. Appl Environ Microbiol 73, 5261–5267.

Yanisch-Perron, C., Vieira, J. & Messing, J. (1985). Improved M13phage cloning vectors and host strains: nucleotide sequences of theM13mp18 and pUC19 vectors. Gene 33, 103–119.

Yatome, C., Ogawa, T. & Matsui, M. (1991). Degradation of crystalviolet by Bacillus subtilis. J Environ Sci Health, Part A 26, 75–87.

Yatome, C., Fujimi, K. & Ogawa, T. (1992). Effect of inhibitors on azo-

reduction of 49-dimethylaminoazobenzene-2-carboxylic acid by

Clostridium diaphorase. Bull Chem Soc Jpn 65, 2349–2351.

Yatome, C., Yamada, S., Ogawa, T. & Matsui, M. (1993). Degradation

of crystal violet by Nocardia corallina. Appl Microbiol Biotechnol 38,

565–569.

Edited by: H. Daims

Y. Stolze and others

2072 Microbiology 158