- Email: [email protected]
A bioluminescent derivative of Pseudomonas putida KT2440 for deliberate release into the environment
FEMS Microbiology Ecology 34 (2000) 91^102
www.fems-microbiology.org
MiniReview
A bioluminescent derivative of Pseudomonas putida KT2440 for deliberate release into the environment Cayo Ramos b
a;
*, La¨zaro Molina
b;1
, Lars MÖlbak
a;2
, Juan L. Ramos b , SÖren Molin
a
a Technical University of Denmark, Department of Microbiology, Building 301, Lyngby 2800, Denmark CSIC, Estacio¨n Experimental del Zaid|¨n, Department of Biochemistry, Prof. Albareda, 1, E-18008 Granada, Spain
Received 8 May 2000; received in revised form 3 August 2000; accepted 5 September 2000
Abstract Recombinant derivatives of Pseudomonas putida strain KT2440 are of potential interest as microbial inoculants to be deliberately released for agricultural applications. To facilitate tracking of this strain and its derivatives after introduction into the environment, a mini-Tn5PluxAB transposon was introduced into the chromosome of P. putida KT2440, yielding strain P. putida S1B1. Sequencing of the DNA region located upstream of the PluxAB genes and similarity search with the P. putida KT2440 genome sequence, localized the transposon within a 3021-bp open reading frame (ORF), whose translated sequence showed significant similarity with the hypothetical YdiJ proteins from Escherichia coli and Haemophilus influenzae. A second ORF adjacent to and divergent from the ydiJ sequence was also found and showed significant homology with various LysR-type transcriptional activator proteins from several bacteria. Disruption of the ydiJ locus in P. putida S1B1 did not affect the survival of the strain in unvegetated or vegetated soils. Bioluminescent detection of P. putida S1B1 cells enriched in selective media directly from soil allowed detection of culturable cells in soil samples over a period of at least 8 months. The addition of the luxAB biomarker facilitates tracking in the root system of several plant species grown under sterile and non-sterile conditions. The correlation of the bioluminescent phenotype with the growth activity of P. putida S1B1 cells colonizing the root system of barley and corn plants was estimated by monitoring ribosomal contents using quantitative hybridization with fluorescence-labeled ribosomal RNA probes. A correlation between inoculum density, light output, and ribosomal contents was found for P. putida cells colonizing the root system of barley seedlings grown under sterile conditions. Although ribosomal contents, and therefore growth activity, of P. putida S1B1 cells extracted from the rhizosphere of corn plants grown in non-sterile soil were similar to those found in starved cells, the luminescent system permitted non-destructive in situ detection of the strain in the upper root system. ß 2000 Federation of European Microbiological Societies. Published by Elsevier Science B.V. All rights reserved. Keywords : luxAB; Rhizosphere; Genetically engineered microorganism; Deliberate release; Pseudomonas putida KT2440
1. Introduction The so-called £uorescent Pseudomonas group includes strains whose biochemical, physiological and genetic properties have been well characterized. A number of genetic tools ^ broad host range vectors, transposons and minitransposons, reverse genetics etc. ^ have made it possible
* Corresponding author. Tel.: +45 (45) 252770; Fax: +45 (45) 932809; E-mail : [email protected] 1
Present address: CNRS-LCB, 31 Chemin Joseph Aiguier, P.O. Box 71, Marseille 13492 Cedex 20, France. 2 Present address: NERI, Department of Microbiology and Biotechnology, Frederiksborgvej 399, Roskilde 4000, Denmark.
to design recombinant derivatives of this group of bacteria for use as microbial inoculants in agriculture for the biological control of soil-borne pests [1^3], for promoting plant growth [4,5], or for plant-assisted microbial elimination of pollutants [6,7]. Pseudomonas putida KT2440, a strain recognized as non-pathogenic by the National Institute of Health of the USA, derives from the soil bacterium P. putida mt-2, the natural host for the archetypal TOL plasmid pWW0 [8]. This strain has been shown to be highly versatile in the acquisition of recombinant DNA for expansion of its catabolic potential to remove pollutants [9], and in the expression of heterologous genes of interest in biological containment [10^13]. Furthermore, P. putida KT2440 and its derivatives survive well in unvegetated soils [14,15] and in the rhizosphere of plants grown both under greenhouse [16] and environmental conditions [12,16]. In
0168-6496 / 00 / $20.00 ß 2000 Federation of European Microbiological Societies. Published by Elsevier Science B.V. All rights reserved. PII: S 0 1 6 8 - 6 4 9 6 ( 0 0 ) 0 0 0 8 9 - 1
FEMSEC 1182 23-11-00
Cyaan Magenta Geel Zwart
92
C. Ramos et al. / FEMS Microbiology Ecology 34 (2000) 91^102
addition, the 6.1-Mb genome of P. putida KT2440 has been mapped [17], and its sequence, currently under assembly of contigs, is already available from The Institute for Genomic Research (TIGR), Rockville, MD, USA (http://www.tigr.org/cgi-bin/BlastSearch/blast.cgi?). All these characteristics make P. putida KT2440 particularly relevant for the development of microbial inoculants to be used in agricultural applications. Indeed, since the publication of the EU Council Directive on the deliberate release into the environment of genetically modi¢ed organisms (GMO) (90/220/EEC), 4 out of the 16 approved releases of Pseudomonas sp. have dealt with P. putida KT2440 and its derivatives. Initial approval of the deliberate release of a genetically engineered microorganism (GEM) requires not only detailed information on the genetic modi¢cation introduced to the parental strain, but also a documented description on its ecology and environmental impact. Assessment of environmental impact and of risks associated with environmental releases of GEMs requires knowledge of microbial survival, growth, activity and dispersal within the environment, and of the persistence of recombinant DNA and its transfer to the indigenous micro£ora. Surveillance and colonization of target environments by bacteria require adequate detection methods to monitor the released microorganism and to assess its activity. Although the green £uorescent protein (GFP) is one of the markers becoming more popular, mainly because unlike other biomarkers it does not require any substrate or additional cofactor to £uoresce, the GFP £uorescent phenotype does not indicate the metabolic status of the cells. One of the more promising markers for determination of cellular metabolic activity is bacterial luciferase, encoded by the luxAB genes. These genes are available in a wide variety of vectors and have been successfully used to study the colonization of plant roots by the potential biocontrol agents Pseudomonas £uorescens, Xanthomonas campestris and Enterobacter cloacae, and by several strains of Rhizobium (for a review see [18]). We have previously reported the construction of a bioluminescent derivative of P. putida KT2440, called P. putida S1B1, which was obtained after random mutagenesis with a mini-Tn5-luxAB-Km transposon. Bioluminescence production by this strain, expressed from a chromosomal promoter, did not a¡ect the growth characteristics of the strain under laboratory conditions [16]. Using this strain we showed that P. putida KT2440 colonizes and survives in the rhizosphere of a number of plant species, i.e. corn and broad bean, and that the introduction of this strain had no signi¢cant e¡ect on the survival of a variety of soil bacteria [16]. However, the above studies did not provide detailed information required for further applications and/or approval of derivatives of P. putida S1B1 as GEMs to be deliberately released into the environment, i.e., the locus interrupted by the reporter cassette into the chromosome of the parental strain, the e¡ect of such a modi¢cation on the competitive-
FEMSEC 1182 23-11-00
ness of the strain, and the spatial distribution and physiological activity of P. putida S1B1 cells colonizing the rhizosphere environment. In this study, we have addressed and described all these issues. 2. Materials and methods 2.1. Bacterial strains, plasmids and growth media P. putida S1B1 [16] is a bioluminescent derivative of the RifR P. putida strain KT2442 [19] containing in the chromosome a mini-Tn5-PluxAB-KmR . P. putida KT2442-5 is a SmR derivative of P. putida KT2442 (this study). Bacteria were grown at 30³C in LB medium or M9 minimal medium [20] with 28 mM glucose or 10 mM benzoate as the only carbon source. Antibiotics were added at the following concentrations (Wg ml31 ): kanamycin 50, rifampicin 20, and streptomycin 50. Bacterial inocula for seed coating or soil inoculation were prepared as follows : A colony of the corresponding P. putida strain was inoculated into 10 ml of LB medium and the culture was incubated at 30³C until the stationary phase was reached. Cells were washed twice in M8 bu¡er (50 mM phosphate, 100 mM NaCl) prior to their use. 2.2. DNA techniques Plasmid DNA was isolated by the alkaline lysis method using the QIAprep spin plasmid minipreps kit (Qiagen cat. no. 27104). Total DNA was isolated as described before [21], except for the 30-min incubation step at 55³C which was omitted. DNA digestions with restriction enzymes, ligations, and transformations, were performed by using standard procedures [22]. The insertion of the mini-Tn5/luxAB-KmR into the chromosome of P. putida S1B1 was located by sequencing the DNA region located upstream of the PluxAB genes using plasmid pS1KPN (see Section 3) as a template and oligonucleotide O (5P-CTGACTCTTATACACAAGT-3P), annealing to the `O' mini-Tn5 end sequence, as a primer. DNA walking was performed using the following speci¢c primers designed upon the available sequence: AGPD1, 5P-CATCAGCACCTTGGAGTCGAC-3P; CP2, 5P-ATGCCGCCGACCGTGGCGC-3P and; S1KPNPR, 5P-GCCTCTTGATATTGGCTAG-3P. DNA was sequenced with the dideoxy sequencing method, using ABI Prism dRhodamine terminators kit (Perkin-Elmer ref. 403042). 2.3. DNA and protein sequence analysis Similarity search with the incomplete P. putida KT2440 genome and preliminary DNA sequence data of the P. putida open reading frame (ORF)1 and ORF2 (see Section 3) were obtained from the TIGR website at http://www. tigr.org. using `TIGR BLAST' (Search Engine for Un¢n-
Cyaan Magenta Geel Zwart
C. Ramos et al. / FEMS Microbiology Ecology 34 (2000) 91^102
ished Microbial Genomes). ORFs were identi¢ed using `ORF Finder' from the Basic Local Alignment Search Tool (BLAST) at http://www.ncbi.nlm.nih.gov/gorf/ gorf.html. Promoter search was performed using `NNPP/ Prokaryotic' from the Human Genome Sequencing Center at http://dot.imgen.bcm.tmc.edu:9331/seq-search/genesearch.html. Translation of DNA sequences into protein sequences was obtained using the `Translate Tool' from the Expert Protein Analysis System (EXPASY) proteomics server of the Swiss Institute of Bioinformatics (SIB) at http://www.expasy.ch. Protein similarity search was performed using `Basic BLAST' at http://www.ncbi.nlm.nih.gov/blast. Helixturn-helix DNA-binding motifs and multiple protein sequence alignment were obtained using the programs `HTH Motif Prediction' [23] and `MULTIALIN Network Protein Sequence @nalysis', respectively, from Poªle Bio-informatique Lyonnais at http://pbil.ibcp.fr/cgi-bin/ npsa_automat.pl?page = /NPSA/npsa_server.html. 2.4. Unvegetated soil microcosm assays Two sandy-loam soils, a £uvisol soil (2.30% (w/w) organic matter, 6.0% (w/w) CaCO3 ) and a cambisol soil (0.63% (w/w) organic matter, 20.4% (w/w) CaCO3 ), were used in this study; before the soil was used it was sifted through a 4-mm mesh metal sieve. One milliliter of bacterial inoculum, prepared as described above, was added to the pots containing soil and evenly distributed to the indicated initial cell density in each assay. Cultivable cells (CFU per gram of soil) were estimated as described previously [14]. 2.5. Seed sterilization and coating with bacteria Corn (Zea mays), broad bean (Vicia faba var. Aguadulce) and tomato (Lycopersum esculentum) seeds were surface-sterilized for 20 min by soaking in 1 g l31 HgCl2 supplemented with 1 ml l31 Tween 80, and washed four times for 5 min in sterile water. Barley seeds (Hordeum vulgaris var. Alexis) were surface-sterilized by immersion in 70% (v/v) ethanol followed by treatment with 3% (v/v) hypochlorite as described by Kragelund and Nybroe [24]. Seed sterilization was veri¢ed by incubating 10^20 seeds on LB-agar at 25³C for several days without any contamination appearing. Sterile seeds were pre-germinated on moist ¢lter paper for 24 h prior to inoculation. Surface-sterilized seeds were coated with bacteria by soaking them for 30 min at 20³C in a 10-ml bacterial inoculum (see above) turbidimetrically adjusted to the appropriate cell density. The non-adhering liquid was decanted, and the seeds were air-dried for 10 min in a petri dish. Bacteria were washed from three seeds into 3 ml of 0.9% (w/v) NaCl, and serially diluted onto selective medium to determine the number of bacteria applied to the seeds.
FEMSEC 1182 23-11-00
93
2.6. Rhizosphere colonization experiments Glass tubes (15 cm highU3 cm diameter) were ¢lled with 200 ml of vermiculite and 75 ml water or 200 g of a £uvisol soil (2.3% (w/w) organic matter, 6.0% (w/w) CaCO3 ). Soil moisture was adjusted to 15% (w/w) with sterile water. Sterile corn seeds coated with bacteria were sown at about 1 cm below the soil or vermiculite surface, and the plants were germinated and grown in a greenhouse at 25³C with natural daily light period (approximately 11 h light, 13 h dark). Rhizosphere colonization of corn plants grown in vermiculite was assessed by photon counting imaging as described below. When soil was used, plants were removed from the pots after di¡erent periods of time and the number of CFU g31 of rhizosphere soil (the soil loosely attached to the roots) and the number of CFU g31 of bulk soil (the soil remaining into the pots) were estimated as described previously [14,15]. In co-inoculation experiments bacterial suspensions were mixed together before inoculation. Germination and growth of barley seedlings under sterile conditions was performed in M8 minimal medium plates (M9 minimal medium [20] containing no carbon or nitrogen sources) containing 1% (w/v) agar. The plates were incubated in the dark at 20³C for 3 days. Extraction of bacteria from the root system of agar-grown seedlings was performed by placing the roots into 2 ml of 0.9% (w/v) cold sodium chloride and mixing by vortexing during 3 min. Bacteria were counted on LB plates containing the appropriate antibiotics for the selection of the corresponding strain. As sterility control, bacteria were also counted on LB plates and no di¡erence in colony numbers was observed between the two media for any of the tested strains. 2.7. Whole-cell hybridization Cells were extracted from the rhizosphere as previously described [25]. Extracted cells as well as cells from batch cultures, were ¢xed in 3% (v/v) paraformaldehyde as described by Poulsen et al. [26]. Fixed cells were stored at 320³C in a storage bu¡er (50% ethanol, 10 mM Tris pH 7.5, 0.1% Nonidet P-40) until use. Hybridization of ¢xed cells was performed as described by Poulsen et al. [26] using 16S rRNA probe PP986, an oligonucleotide probe speci¢c for the P. putida subgroup A labeled with the incarbocyanine £uorescent dye CY3 [27]. 2.8. Epi£uorescence microscopy An Axioplan epi£uorescence microscope (Carl Zeiss) was used to visualize the hybridizations. The microscope was equipped with a 100-W mercury lamp, and ¢lter set XF40 (Omega Optical, Brattlebror, VT, USA) was used to visualize CY3. A slow-scan charge-coupled device (CCD) camera CH250 (Photometrics, Tucson, AZ, USA)
Cyaan Magenta Geel Zwart
94
C. Ramos et al. / FEMS Microbiology Ecology 34 (2000) 91^102
equipped with a KAF 1400 chip (pixel size, 6.8 by 6.8 Wm), was used for capturing digitilized images (PMIS). The camera was operated at 340³C, and the chip was read out in 12 bits (4096 intensity levels) at a rate of 200 kHz. For quanti¢cation of £uorescence, the PMIS images were analyzed using the Cellstat program [28]. A DOSbased 486 computer was used as a controller for the CCD camera, and a SUN IPX computer was used to run Cellstat. 2.9. Luciferase activity of intact cells in liquid cultures of P. putida S1B1 Light emission was determined as described by Rattray et al. [29]. To 0.5 ml of a cell suspension with a turbidity of 0.01 at 660 nm, we added 0.4 ml of 0.01% (v/v) n-decyl aldehyde and recorded the time course of light emission immediately thereafter for 1 min in a LKB1250 luminometer. The activity is given as the peak height, in relative light units (RLU), per turbidity unit. Light output per unit turbidity in cultures of P. putida S1B1 growing on LB or minimal medium with benzoate as the only C-source was relatively constant during the exponential growth phase; thereafter light output decreased. On LB medium light output during the exponential growth phase, around 100 RLU per OD660 , was clearly lower than when cells were growing on M9 minimal medium supplemented with benzoate (around 1000 RLU per OD660 ) (data not shown). When the number of CFU g31 of soil was below our limit of detection by colony plating (50 CFU g31 of soil), bioluminescence detection of P. putida S1B1 cells enriched in LB directly from soil was performed as described by Shaw et al. [30]. One gram of the soil sample was placed in 9 ml of LB medium supplemented with the appropriate antibiotics, and containing 40 Wg ml31 of cycloheximide to inhibit fungal growth. After 24 h of incubation at 30³C, aliquots (0.5 ml) were removed and used to determine bioluminescence activity as described above. 2.10. Luminescence detection by photon counting imaging A Hamamatsu C-2400 photon counting intensi¢ed CCD camera and an Argus-50 image processor were used to obtain images of transmitted photons. The camera was enclosed in a light-tight box with a removable door to reduce entry of external light during dark-¢eld exposures. Bright- and dark-¢eld images were superimposed with the aid of Argus 10 software. Images were then processed with Adobe Photoshop software. Plants were removed from the soil, vermiculite or agar systems and the roots were placed on ¢lter paper soaked with 0.01% (w/v) n-decyl aldehyde. For luminescence visualization with the CCD camera, samples were exposed for 3 s in the light to obtain a bright-¢eld image. The dark-¢eld exposure time was 10^90 s.
FEMSEC 1182 23-11-00
For microscopic examination of bioluminescence in root samples the photon counting intensi¢ed CCD was connected to an Axioplan microscope (Carl Zeiss). A Plan Neo£uar 2.5U/0.075 objective and an Epiplan 50U/0.50 were used. Samples were exposed for 10 s in the light to obtain a bright-¢eld picture. The dark-¢eld exposure time was 10 min. 3. Results and discussion 3.1. Determination of the insertion point of the mini-Tn5-PluxAB-Km transposon in the chromosome of P. putida S1B1 The integration of reporter systems into the chromosome of bacterial strains by transposon mutagenesis is an accepted routine practice, as long as the chromosomal insertions do not have phenotypic consequences in relation to the investigations carried out. P. putida strain S1B1 carries a mini-Tn5-PluxAB transposon inserted into the chromosome making the cells luminescent after exposure to n-decyl aldehyde [16]. In this transposon, the luxAB genes are transcribed only if the insertion occurs downstream of an active promoter. To identify the integration point of the mini-Tn5 transposon in the chromosome of P. putida S1B1, a DNA fragment located upstream of the mini-Tn5 `O' end was cloned and sequenced. Chromosomal DNA from P. putida S1B1 was digested with KpnI, which does not cut within the luxAB-km resistant determinant genes and the resulting DNA fragments were cloned into pUC19 [31]. The ligation mix was transformed into Escherichia coli DH5K and a bioluminescent transformant was selected on LB plates containing Km. A plasmid containing a 6.5-kb KpnI insert, called pS1KPN, was isolated from this transformant. Restriction enzyme mapping of this plasmid showed that the 6.5-kb fragment contained indeed the mini-Tn5/luxAB-Km (data not shown). The DNA region located upstream of the PluxAB genes was sequenced using plasmid pS1KPN as a template. Oligonucleotide O, annealing to the `O' mini-Tn5 end sequence, was used as primer to obtain the immediate adjacent DNA sequence. A total of 2290 bp were sequenced by DNA walking using speci¢c primers designed from the available sequences (see Section 2). Similarity search using TIGR BLAST (Search Engine for Un¢nished Microbial Genomes) yielded 98% identity with the 3P and 5P ends of contigs number 356 (8726 bp) and 340 (4645 bp), respectively, of the P. putida KT2440 un¢nished genome sequence. The nucleotide sequence of these two contigs overlapped 564 bp (Fig. 1). BLAST search of the corresponding 2290-bp sequence obtained from the P. putida KT2440 genome sequence identi¢ed two incomplete ORFs transcribed divergently and separated by 129 bp (Fig. 1). The mini-Tn5/luxAB-Km was located within ORF1 downstream codon number 227. The complete nu-
Cyaan Magenta Geel Zwart
C. Ramos et al. / FEMS Microbiology Ecology 34 (2000) 91^102
Fig. 1. Location of the mini-Tn5/luxAB/KmR into the chromosome of P. putida S1B1. A schematic representation of ORF1 and ORF2 into contigs numbers 340 and 356 from the P. putida KT2440 genome sequence, the direction of the transcription is indicated by the arrows. The triangle indicates the location of the mini-Tn5 within ORF1. The location of the 2290-bp sequence obtained from plasmid pS1KPN within ORF1 and ORF2 at the nucleotide sequence of the putative promoter located upstream of ORF1 are indicated.
cleotide sequence of P. putida KT2440 ORF1 and ORF2, accession numbers AF289509 and AF289508, respectively, was obtained from the sequence of contigs 356 and 340. We have not determined the transcription initiation point of ORF1, but a putative promoter sequence was identi¢ed 74 bp upstream the ATG initiation codon of this locus using program NNPP/Prokaryotic (Fig. 1). The translated sequence obtained from ORF1 (1006 residues) had a 67 and 65% similarity with the hypothetical proteins translated from the ydiJ locus from E. coli (D90811) and Haemophilus in£uenzae (U32796), respectively (Fig. 2). The 580 N-terminal residues of the P. putida KT2440 YdiJ protein produced various degrees of similarity with glycolate oxidase subunit D (GlcD) from Bacillus subtilis (Z99118, 44%), Helicobacter pylori (AE000565, 42%), Synechocystis sp. (D64001, 41%), E. coli (AE000380, 40%), Deinoccocus radiodurans (AE002015, 40%) and Archaeoglobus fulgidus (AE001049, 39%) (see Fig. 2 for homology with GlcD from E. coli and Synechocystis sp.). Various degrees of similarity were also found at the N-terminal end of the P. putida YdiJ protein with D-lactate dehydrogenase from Aquifex aeolicus (AE000755, 42%), A. fulgidus (AE001077, 42%), H. pylori (AE001542, 42%) and yeast (Z74031, 39%). Interestingly, the 426 residues located at the C-terminal end of the P. putida YdiJ protein had similarity with another subunit of glycolate oxidase (the iron-sulfur subunit GlcF) from E. coli (L43490, 37%) and Synechocystis sp. (D90903, 33%) (Fig. 2), however, the function of this hypothetical protein is completely unknown. The translated 296 amino acid sequence from ORF2 had signi¢cant similarity with several transcriptional regulators belonging to the LysR family: the regulator of the glycine cleavage system (GcvA) from E. coli (X73413, 54%), the trpBA operon transcriptional activator (TrpI) from Pseudomonas syringae (M95710, 52%), the AmpR protein from several bacteria, i.e. E. cloacae (AB003148, 50%) and Pseudomonas aeruginosa (X67095, 47%), the LysR-type L-lactamase transcriptional regulator PenR
FEMSEC 1182 23-11-00
95
from Burkholderia cepacia (U85041, 46%) and the transcriptional regulator FonR-1 from Serratia fonticola (AJ251239, 49%). A helix-turn-helix DNA-binding motif was detected using program HTH [23] between residues 25 and 46 of the P. putida ORF2 amino acid sequence (ESYTRAAAELSLTQSAVSRQVQ). This region was found to align exactly with the DNA-binding motifs detected for all the other transcriptional activator proteins showing similarity to the P. putida ORF2-translated sequence (data not shown). The possible function of the ORF2 protein as a regulator of the expression of the ydiJ locus was not studied here. 3.2. Competitiveness of P. putida strain S1B1 in unvegetated soil and in the soil rhizosphere The ecological performance of rhizosphere colonizing bacteria is a complex phenotype likely to be a¡ected by many di¡erent traits. Gene inactivation by transposon mutagenesis has been successfully used to identify bacterial traits involved in rhizosphere competence, e.g. motility, synthesis of the O-antigen lipopolysaccharide, thiamine production, amino acid synthesis, biotin production and a site-speci¢c recombinase (for a review see [32]). We have previously reported that the insertion of the mini-Tn5 cassette into the chromosome of P. putida S1B1 does not a¡ect the growth characteristics of the strain under laboratory conditions [16]. Since the function and ecological role of the P. putida ydiJ locus is completely unknown, it was, however, necessary to further test whether interruption of this locus a¡ected the survival of the strain in soil and the plant rhizosphere in comparison with its parental strain P. putida KT2442. The behavior of P. putida S1B1 and its parental strain KT2442 in non-sterile soil was similar. P. putida strains KT2442 or S1B1 were introduced at a cell load of 105 CFU g31 of soil in pots containing 90 g of either a cambisol soil, poor in organic matter, or a £uvisol soil relatively rich in organic matter. The pots were incubated at 17^22³C as described previously [14]. In the £uvisol soil, both strains maintained relatively constant cell numbers for at least 3 months (data not shown). In contrast, in the cambisol soil the number of CFU g31 of soil for both bacteria decreased steadily with time, so that after 56 days of incubation none of the strains could be detected after plating in selective culture plates (data not shown). Detection with antibiotic markers has been successfully used to show survival in soil of P. putida KT2440 and its derivatives [14,15]. Due to the decline of the bacterial population size after introduction into natural soil, most of the studies have focused on the behavior of these microbes during the initial period after introduction into the microcosms, usually 4 and 8 weeks [9]. The bioluminescent phenotype of P. putida S1B1 cells allowed a study of the longterm survival of P. putida S1B1 in the non-sterile soil microcosms described above. At every sampling time, P.
Cyaan Magenta Geel Zwart
96
C. Ramos et al. / FEMS Microbiology Ecology 34 (2000) 91^102
FEMSEC 1182 23-11-00
Cyaan Magenta Geel Zwart
C. Ramos et al. / FEMS Microbiology Ecology 34 (2000) 91^102
97
Fig. 2. Sequence alignment of YdiJ proteins and glycolate oxidase subunits D and F. The sources of the sequences were: P. putida YdiJ (this study, AF289509), E. coli YdiJ [41], H. in£uenzae YdiJ [42], E. coli GlcD and GlcF [43] and Synechocystis sp. GlcD and GlcF [44]. The MULTIALIN program (NPSA, Lyon, France) was used. Residues that are identical in at least four aligned proteins are highlighted in black, residues that are identical in the three YdiJ proteins are highlighted in gray.
putida S1B1 cells were enriched from 1 g of soil on selective LB-medium (see Section 2) and after a period of 20 h, luminescence emission was measured. Soil inocula produced detectable bioluminescence at every sampling time over a period of at least 8 months (data not shown), demonstrating that although the number of cells present in the soil was below the detection limit by plating in selective plates (50 CFU g31 of soil), metabolically active P. putida S1B1 cells were present in the soil until the end of the
FEMSEC 1182 23-11-00
assay. We previously established that suspensions of 1 g of soil containing as few as 5 CFU of P. putida S1B1 in 10 ml LB medium resulted in detectable luminescence after about 20 h of growth at 30³C. The insertion of the mini-Tn5 cassette into the chromosome of P. putida S1B1 did not a¡ect the ability of the strain to colonize the rhizosphere of corn, barley or bean plants in comparison with its parental strain P. putida KT2442. Corn seeds were coated with either P. putida
Cyaan Magenta Geel Zwart
98
C. Ramos et al. / FEMS Microbiology Ecology 34 (2000) 91^102
3.3. Distribution of P. putida S1B1 cells in the rhizosphere
Fig. 3. Survival of P. putida strains KT2442 (open symbols) and S1B1 (closed symbols) in the rhizosphere of corn. Corn seeds were coated with 108 CFU of the corresponding strain, sown in £uvisol-soil and incubated in a greenhouse. Bacteria in the rhizosphere (circles) and in the bulk soil (squares) were counted on LB plates containing the appropriate antibiotics. Con¢rmation that the colonies selected on media containing Rif and Km indeed represented the bioluminescent P. putida S1B1 strain was obtained by the fact that all the colonies become luminescent after exposure to n-decyl aldehyde. Each point represents an average based on three independent plants, the error bars indicate standard deviations from the mean.
KT2442 or P. putida S1B1 to about 104 , 106 and 108 CFU per seed. After 7 days of plant growth in non-sterile £uvisol soil we detected about 107 ^108 CFU g31 of rhizosphere soil regardless of the initial inoculum size and of the bacterial strain used. Thereafter the number of bacteria in the rhizosphere declined to about 106 ^107 CFU g31 of rhizosphere soil after 21 days and then remained relatively constant at this cell density during at least 42 days. However, at any given sampling time the number of CFU g31 of soil was generally two to three orders of magnitude higher in the rhizosphere than in the bulk soil (see Fig. 3 for results obtained when the seeds were coated with 108 CFU). Similar results were obtained when bean and barley plants were used instead of corn (data not shown). The behavior of the strain in direct competition with a streptomycin-labeled P. putida KT2442-5 derivative was also investigated. Corn seeds were coated with 107 CFU of each strain and sown in non-sterile £uvisol soil. The cells of each strain in the rhizosphere of corn were counted, and similar numbers, on the order of 106 ^107 CFU g31 of soil, were found in assays lasting 42 days (data not shown). All these results together show that interruption of the ydiJ locus does not have an e¡ect on the ¢tness of P. putida S1B1 either in unvegetated soil or in the soil rhizosphere.
FEMSEC 1182 23-11-00
Bioluminescence production by P. putida S1B1 is so intense that it can be detected at the single cell level using a photon counting camera (C. Sternberg, The Technical University of Denmark, personal communication). We therefore used luminescence-based detection of lux-labeled cells to examine in situ the distribution of this strain in the rhizosphere of di¡erent plant species. Sterile corn seeds coated with about 106 CFU per seed of P. putida strain S1B1 were germinated in vermiculite or non-sterile soil. After di¡erent periods of time, bioluminescent emission was assessed through photon imaging. Detection of luminescence from P. putida S1B1 cells in the rhizosphere of vermiculite-grown corn plants 7^21-day-old showed that luminescence was closest to the seed part of the root, although other zones of intense light emission were also observed throughout the entire root system (not shown), indicating a non-uniform distribution of bacteria in the rhizosphere. This is consistent with data on root colonization by plant growth-promoting lux [33,30] and lacZmarked [34] Pseudomonas bacteria. Dark-¢eld images of P. putida S1B1 cells in the root systems of broad bean, tomato and barley showed a similar pattern of colonization (data non-shown). Microscopic examination (magni¢cationU25) of a bright spot located in the upper root system showed two parallel bioluminescent lines along the root segment (Fig. 4A,B). A higher magni¢cation (500U) showed that bioluminescent lines were composed of bright spots of di¡erent sizes and intensities (Fig. 4C,D). This pattern of bioluminescence probably re£ects the localization of the bacterial cells (single cells and microcolonies) at the border of adjacent plant root cells, in agreement with data reported for other Pseudomonas strains [35,36,25]. 3.4. Activity of P. putida S1B1 cells in the rhizosphere Luciferase activity has been shown to be proportional to biomass in growing bacterial populations. However, under nutrient-limited conditions, such as those of the soil and rhizosphere environment, the concentration of FMNH2 required for the luciferase reaction becomes limiting, and the bioluminescence of the population decreases substantially compared to that of the biomass [37^39]. In this study, we combined luminescence-based detection of luxlabeled cells with quantitative hybridization with £uorescence labeled ribosomal RNA probes to ascertain the dependence of the luminescent phenotype on the physiological status of P. putida S1B1 cells colonizing the rhizosphere environment. Previously, we found that the ribosomal contents, and therefore growth activity, of P. putida cells (P. putida JB156, [40]) colonizing the root system of barley seedlings grown under sterile conditions was proportional to the increment in bacterial biomass [25]. Sterile barley seeds were inoculated with P. putida strain
Cyaan Magenta Geel Zwart
C. Ramos et al. / FEMS Microbiology Ecology 34 (2000) 91^102
99
Fig. 4. In situ visualization of bioluminescence from of P. putida S1B1 cells in the root system of corn. Sterile corn seeds were coated with approximately 106 CFU per seed of P. putida strain S1B1, sown in vermiculite and incubated in a greenhouse. Bright-¢eld exposure (10 s) (A and C) and dark-¢eld exposure (10 min) (B and D) of a 7-day-old root after the addition of n-decyl aldehyde. A and B, magni¢cationU25; C and D, magni¢cationU500. Bright- and dark-¢eld images were processed with Adobe Photoshop software. Relative intensity of light emission is indicated by the color scale.
S1B1 at four di¡erent cell loads, from approximately 103 to about 107 CFU per seed, and germinated at 20³C in the dark in agar plates containing no carbon or nitrogen sources. At the time of inoculation, only the seeds inoculated with a high cell density (between 107 and 105 CFU per seed) showed detectable bioluminescence (data not shown). During the ¢rst day of plant growth and regardless of the initial inoculum size, cell density increased to approximately the same level in all the roots. However, light intensity in the roots, and therefore the metabolic activity of the cells, increased with decreasing inoculum cell density (Fig. 5). Light intensity also correlated with the ribosomal contents of the cells extracted from the roots of the 1-day-old barley seedlings. The ribosomal content of the extracted cells, normalized to 1 for the starved cells used as inoculum, varied between 1.53 (initial
FEMSEC 1182 23-11-00
cell density of approximately 103 CFU per seed) and 0.60 (initial cell density of approximately 107 CFU per seed). During the following 2 days, the number of CFU increased in a similar fashion in all the roots (Fig. 5), whereas the light output (not-shown) and ribosomal contents of the cells extracted from the roots decreased in a similar fashion for all the plants (Fig. 5). We also assessed the correlation of the luminescent phenotype with the physiological status of P. putida S1B1 cells colonizing the rhizosphere in non-sterile £uvisol soil. A stationary culture of P. putida strain S1B1 was used to inoculate sterile corn seeds at a cell load of approximately 105 CFU per seed. Coated seeds were planted in the soil systems and the pots were incubated at 25³C. Six days after planting, luminescence was detected only close to the seed part of the root (not shown). The lack of detect-
Cyaan Magenta Geel Zwart
100
C. Ramos et al. / FEMS Microbiology Ecology 34 (2000) 91^102
Fig. 5. Cell density, light emission and ribosomal contents of P. putida S1B1 cells colonizing the root system of barley seedlings as a function of the inoculum size. Sterile barley seeds were coated at the indicated cell density of P. putida strain S1B1 and germinated in agar plates at 20³C for 3 days. Bioluminescence in the root system of the developing barley seedlings was detected by photon count with a CCD camera. Bright-¢eld exposures (3 s) and dark-¢eld exposures (30 s) were processed with Adobe Photoshop software. Relative intensity of light emission is indicated by the color scale. Ribosomal contents were estimated by quantitative hybridization with £uorescently labeled rRNA. The £uorescence intensity was normalized to one relative £uorescence unit (r.f.u.) for the starved cells used as inoculum. The values are mean þ standard deviations based on the values obtained from three different roots.
able luminescence in the lower root segments does not necessarily indicate absence of cells; luminescence emission could be below the detection limit by photon-counting imaging due to low cell number and/or low physiological activity of the cells. Other authors [39] have reported that luciferase activity of luxAB-tagged Pseudomonas strains decreases rapidly after bacterial introduction into soil due to a decrease in cellular metabolic activity, however, the light emitted by cells extracted from soil can be quanti¢ed by luminometry in cell suspensions containing a relatively high number of viable cells (104 ^105 CFU ml31 ). Although the number of CFU per cm of root was found to be approximately two orders of magnitude higher in the upper (luminescent, 105 CFU per cm) than in lower (nonluminescent, 103 CFU per cm) root segment, the average ribosomal content and cell volume of the cells extracted from the upper root segment (1 cm) was about 3.2 and 1.4
FEMSEC 1182 23-11-00
times lower, respectively, than that of the starved cells used as inoculum. We have previously reported that a signi¢cant decrease in the ribosomal content and cell volume of P. putida JB156 cells occurs immediately after bacteria introduction in the rhizosphere of barley plants grown in non-sterile soil [25]. To our knowledge, this is the ¢rst study correlating the bioluminescent phenotype with the ribosomal contents, and therefore growth activity, of luxAB-tagged bacterial cells introduced into environmental samples. In agreement with data reported by other authors [37^39], our results show that the bioluminescent phenotype of P. putida S1B1 is strongly dependent on the growth activity of the cells. However, even at the low growth rates supported by root exudate, the reporter system allows in situ detection of high density of cells colonizing the root surface of di¡erent plant species.
Cyaan Magenta Geel Zwart
C. Ramos et al. / FEMS Microbiology Ecology 34 (2000) 91^102
4. Concluding remarks The addition of a luxAB marker into the chromosomal ydiJ locus of the non-aggressive root colonizer P. putida KT2440, P. putida strain S1B1, does not a¡ect the competitiveness of the strain in unvegetated soil or the rhizosphere of several plant species. The bioluminescent phenotype allows in situ tracking of this strain after introduction into soil and rhizosphere samples, facilitating assessment of the environmental impact and of risk associated with the environmental release of this strain and its derivatives. Acknowledgements
[12]
[13]
[14]
[15]
[16]
This work was partly supported by a grant to S.M. from the Danish Biotechnology Programme and by Grant BIO97-0641 to J.L.R. from CICYT, Spain. C.R. was partially supported by EMBO fellowship ASTF8056 and by The Carlsberg Foundation (Copenhagen, Denmark). Preliminary sequences from the P. putida KT2440 genome were obtained from The Institute of Genomic Research.
[17]
[18]
[19]
References [20] [1] O'Sullivan, D.J. and O'Gara, F. (1992) Traits of £uorescent Pseudomonas spp. involved in suppression of plant root pathogens. Microbiol. Rev. 56, 662^676. [2] Dowling, D.N. and O'Gara, F. (1994) Metabolites of Pseudomonas involved in the biocontrol of plant disease. Trends Biotechnol. 12, 133^141. [3] Thomashow, L.S. (1996) Biological control of plant root pathogens. Curr. Opin. Biotechnol. 7, 343^347. [4] Lugtenberg, B.J.J., deWeger, L.A. and Bennett, J.W. (1991) Microbial stimulation of plant growth and protection from disease. Curr. Opin. Biotechnol. 2, 457^464. [5] Kloepper, J.W., Li¡shitz, R. and Zablotowicz, R.M. (1989) Free living bacterial inocula for enhancing crop productivity. Trends Biotechnol. 7, 39^44. [6] Crowley, D.E., Alvey, S. and E.S. Gilbert (1997) Rhizosphere ecology of xenobiotic-degrading microorganisms. In: Phytoremediation of Soil and Water Contaminants (Kruger, E.L. Anderson, T.A. and Coats, J.R., Eds.), ACS Symposium Series 664, pp. 19^34. American Chemical Society, Washington, DC. [7] Brazil, G.M., Kene¢ck, L., Callanan, M., Haro, A., deLorenzo, V., Dowling, D.N. and O'Gara, F. (1995) Construction of a rhizosphere Pseudomonad with potential to degrade polychlorinated biphenyls and detection of bph gene expression in the rhizosphere. Appl. Environ. Microbiol. 61, 1946^1952. [8] Worsey, M.J. and Williams, P.A. (1974) Metabolism of toluene and xylenes by Pseudomonas putida (arvilla) mt-2: evidence for a new function of the TOL plasmid. J. Bacteriol. 124, 7^13. [9] Ramos, J.L., D|¨az, E., Dowling, D., de Lorenzo, V., Molin, S., O'Gara, F., Ramos, C. and Timmis, K.N. (1994) The behavior of bacteria designed for biodegradation. BioTechnology 12, 1349^1356. [10] Ronchel, M.C., Ramos, C., Jensen, L.B., Molin, S. and Ramos, J.L. (1995) Construction and behaviour of biologically contained bacteria for environmental applications in bioremediation. Appl. Environ. Microbiol. 61, 2990^2994. [11] Ramos, J.L., Andersson, P., Jensen, L.B., Ramos, C., Ronchel, M.C.,
FEMSEC 1182 23-11-00
[21]
[22]
[23]
[24]
[25]
[26]
[27]
[28]
[29]
101
D|¨az, E., Timmis, K.N. and Molin, S. (1995) Suicide microbes on the loose. BioTechnology 13, 35^37. Molina, L., Ramos, C., Ronchel, M.C., Molin, S. and Ramos, J.L. (1998) Construction of an e¤cient biologically contained Pseudomonas putida strain and its survival in outdoor assays. Appl. Environ. Microbiol. 64, 2072^2078. Ronchel, M.C., Molina, L., Witte, A., Lutbiz, W., Molin, S., Ramos, J.L. and Ramos, C. (1998) Characterization of cell lysis in Pseudomonas putida upon expression of heterologous killing genes. Appl. Environ. Microbiol. 64, 4904^4911. Ramos, J.L., Duque, E. and Ramos-Gonza¨lez, M.I. (1991) Survival in soils of an herbicide-resistant Pseudomonas putida strain bearing a recombinant TOL plasmid. Appl. Environ. Microbiol. 57, 260^266. Duque, E., Marque¨s, S. and Ramos, J.L. (1993) Mineralization of p-methyl-14 C-benzoate in soils by Pseudomonas putida (pWW0). Microb. Releases 2, 175^177. Molina, L., Ramos, C., Duque, E., Ronchel, M.C., Garc|¨a, J.M., Wyke, L. and Ramos, J.L. (2000) Survival of Pseudomonas putida KT2440 in soil and the rhizosphere of plants under greenhouse and environmental conditions. Soil Biol. Biochem. 32, 315^321. Ramos-D|¨az, M.A. and Ramos, J.L. (1998) Combined physical and genetic map of the Pseudomonas putida KT2440 chromosome. J. Bacteriol. 180, 6352^6363. Prosser, J.I. (1994) Molecular marker systems for detection of genetically engineered micro-organisms in the environment. Microbiology 140, 5^17. Franklin, F.C.H., Bagdasarian, M.M., Bagdasarian, M. and Timmis, K.N. (1981) Molecular and functional analysis of the TOL plasmid pWW0 from Pseudomonas putida and cloning of the genes for the entire regulated aromatic ring meta cleavage pathway. Proc. Natl. Acad. Sci. USA 78, 7458^7462. Abril, M.A., Micha¨n, C., Timmis, K.N. and Ramos, J.L. (1989) Regulator and enzyme speci¢cities of the TOL plasmid-encoded upper pathway for degradation of aromatic hydrocarbons and expansion of the substrate range of the pathway. J. Bacteriol. 171, 6782^ 6790. Ramos-Gonza¨lez, M.I., Ru|¨z-Cabello, F., Brettar, I., Garrido, F. and Ramos, J.L. (1992) Tracking genetically engineered bacteria : monoclonal antibodies against surface determinants of the soil bacterium Pseudomonas putida 2440. J. Bacteriol. 174, 2978^2985. Ausubel, F.M., Brent, R., Kingston, R.E., Moore, D.D., Seidman, J.G., Smith, J.A. and Struhl, K. (Eds.), (1991) Current Protocols in Molecular Biology, John Wiley and Sons, New York. Dodd, I.B. and Egan, J.B. (1990) Improved detection of helix-turnhelix DNA-binding motifs in protein sequences. Nucleic Acids Res. 18, 5019^5026. Kragelund, L. and Nybroe, O. (1996) Competition between Pseudomonas £uorescens Ag1 and Alcaligenes eutrophus JMP134 (pJP4) during colonization of barley roots. FEMS Microbiol. Ecol. 20, 41^51. Ramos, C., MÖlbak, L. and Molin, S. (2000) Bacterial activity in the rhizosphere analyzed at the single-cell level by ribosomal contents and synthesis rates. Appl. Environ. Microbiol. 65, 801^809. Poulsen, L.K., Ballard, G. and Stahl, D.A. (1993) Use of rRNA £uorescence in situ hybridization for measuring the activity of single cells in young and established bio¢lms. Appl. Environ. Microbiol. 59, 1354^1360. MÖller, S., Sternberg, C., Andersen, J.B., Christensen, B.B., Ramos, J.L., Givskov, M. and Molin, S. (1998) In situ gene expression in mixed-culture bio¢lms : evidence of metabolic interactions between community members. Appl. Environ Microbiol. 64, 721^732. MÖller, S., Kristensen, C.S., Poulsen, L.K., Carstensen, J.M. and Molin, S. (1995) Bacterial growth on surfaces : automated image analysis for quanti¢cation of growth rate-related parameters. Appl. Environ. Microbiol. 61, 741^748. Rattray, E.A.S., Prosser, J.I., Killham, K. and Glover, L.A. (1990) Luminescence-based nonextractive technique for in situ detection of Escherichia coli in soil. Appl. Environ. Microbiol. 56, 3368^3374.
Cyaan Magenta Geel Zwart
102
C. Ramos et al. / FEMS Microbiology Ecology 34 (2000) 91^102
[30] Shaw, J.J., Dane, F., Geiger, D. and Kloepper, J.W. (1992) Use of bioluminescence for detection of genetically engineered microorganisms released into the environment. Appl. Environ. Microbiol. 58, 267^273. [31] Sambrook, J., Fritsch, E.F. and Maniatis, T. (1989) Molecular Cloning: a Laboratory Manual, 2nd edn., Cold Spring Harbor, NY. [32] Lugtenberg, B.J.J. and Dekkers, L.C. (1999) What makes Pseudomonas bacteria rhizosphere competent? Environ. Microbiol. 1, 9^13. [33] de Weger, L.A., Dunbar, P., Mahafee, W.F., Lugtenberg, B.J.J. and Sayler, G. (1991) Use of bioluminescence markers to detect Pseudomonas spp. in the rhizosphere. Appl. Environ. Microbiol. 57, 3641^ 3644. [34] Simons, M., van der Bij, A.J., Brand, I., de Weger, L.A., Wij¡elman, C.A. and Lugtenberg, B.J.J. (1996) Gnotobiotic system for studying rhizosphere colonization by plant growth-promoting Pseudomonas bacteria. Mol. Plant Microbe Interact. 9, 600^607. [35] Bloemberg, G.V., O'Toole, G.A., Lugtenberg, B.J.J. and Kolter, R. (1997) Green £uorescent protein as a marker for Pseudomonas spp.. Appl. Environ. Microbiol. 63, 4543^4551. [36] Hansen, M., Kragelund, L., Nybroe, O. and SÖrensen, J. (1997) Early colonization of barley roots by Pseudomonas £uorescens studied by immuno£uorescence technique and confocal laser scanning microscopy. FEMS Microbiol. Ecol. 23, 353^360. [37] Duncan, S.L.A., Killham, K. and Prosser, J.I. (1994) Luminescencebased detection of activity of starved and viable but nonculturable bacteria. Appl. Environ. Microbiol. 60, 1308^1316. [38] Meikle, A., Glover, L.A., Killham, K. and Prosser, J.I. (1994) Potential luminescence as an indicator of activation of genetically modi¢ed Pseudomonas £uorescens in liquid culture and in soil. Soil Biol. Biochem. 26, 747^755. [39] Unge, A., Tombolini, R., MÖlbak, L. and Jansson, J. (1999) Simultaneous monitoring of cell number and metabolic activity of speci¢c bacterial populations with a dual gfp-luxAB marker system. Appl. Environ Microbiol. 65, 813^821.
FEMSEC 1182 23-11-00
[40] Christensen, B.B., Sternberg, C., Andersen, J.B., Eberl, L., Moller, S., Givskov, M. and Molin, S. (1998) Establishment of new genetic traits in a microbial bio¢lm community. Appl. Environ. Microbiol. 64, 2247^2255. [41] Blattner, F.R., Plunkett III, G., Bloch, C.A., Perna, N.T., Burland, V., Riley, M., Collado-Vides, J., Glasner, F.D., Rode, C.K., Mayhew, G.F., Gregor, J., Davis, N.W., Kirkpatrick, H.A., Goeden, M.A., Rose, D.J., Mau, B. and Shao, Y. (1997) The complete genome sequence of Escherichia coli K-12. Science 277, 1453^1474. [42] Fleischmann, R.D., Adams, M.D., White, O., Clayton, R.A., Kirkness, E.F., Kerlavage, A.R., Bult, C.J., Tomb, J.-F., Dougherty, B.A., Merrick, J.M., McKenney, K., Sutton, G., Fitzhugh, W., Fields, C.A., Gocayne, J.D., Scott, J.D., Shirley, R., Liu, L.-I., Glodek, A., Kelley, J.M., Weidman, J.F., Phillips, C.A., Spriggs, T., Hedblome, E., Cotton, M.D., Utterback, T.R., Hanna, M.C., Nguyen, D.T., Saudek, D.M., Brandon, R.C., Fine, L.D., Fritchman, J.L., Fuhrmann, J.L., Geoghagen, N.S.M., Gnehm, C.L., McDonald, L.A., Small, K.V., Fraser, C.M., Smith, H.O. and Venter, J.C. (1995) Whole-genome random sequencing and assembly of Haemophilus in£uenzae Rd. Science 269, 496^512. [43] Pellicer, M.T., Badia, J., Aguilar, J. and Baldoma, L. (1996) glc locus of Escherichia coli: characterization of genes encoding the subunits of glycolate oxidase and the glc regulator protein. J. Bacteriol. 178, 2051^2059. [44] Kaneko, T., Sato, S., Kotani, H., Tanaka, A., Asamizu, E., Nakamura, Y., Miyajima, N., Hirosawa, M., Sugiura, M., Sasamoto, S., Kimura, T., Hosouchi, T., Matsuno, A., Muraki, A., Nakazaki, N., Naruo, K., Okumura, S., Shimpo, S., Takeuchi, C., Wada, T., Watanabe, A., Yamada, M., Yasuda, M. and Tabata, S. (1996) Sequence analysis of the genome of the unicellular cyanobacterium Synechocystis sp. strain PCC6803. II. Sequence determination of the entire genome and assignment of potential protein-coding regions. DNA Res. 3, 109^136.
Cyaan Magenta Geel Zwart
www.fems-microbiology.org
MiniReview
A bioluminescent derivative of Pseudomonas putida KT2440 for deliberate release into the environment Cayo Ramos b
a;
*, La¨zaro Molina
b;1
, Lars MÖlbak
a;2
, Juan L. Ramos b , SÖren Molin
a
a Technical University of Denmark, Department of Microbiology, Building 301, Lyngby 2800, Denmark CSIC, Estacio¨n Experimental del Zaid|¨n, Department of Biochemistry, Prof. Albareda, 1, E-18008 Granada, Spain
Received 8 May 2000; received in revised form 3 August 2000; accepted 5 September 2000
Abstract Recombinant derivatives of Pseudomonas putida strain KT2440 are of potential interest as microbial inoculants to be deliberately released for agricultural applications. To facilitate tracking of this strain and its derivatives after introduction into the environment, a mini-Tn5PluxAB transposon was introduced into the chromosome of P. putida KT2440, yielding strain P. putida S1B1. Sequencing of the DNA region located upstream of the PluxAB genes and similarity search with the P. putida KT2440 genome sequence, localized the transposon within a 3021-bp open reading frame (ORF), whose translated sequence showed significant similarity with the hypothetical YdiJ proteins from Escherichia coli and Haemophilus influenzae. A second ORF adjacent to and divergent from the ydiJ sequence was also found and showed significant homology with various LysR-type transcriptional activator proteins from several bacteria. Disruption of the ydiJ locus in P. putida S1B1 did not affect the survival of the strain in unvegetated or vegetated soils. Bioluminescent detection of P. putida S1B1 cells enriched in selective media directly from soil allowed detection of culturable cells in soil samples over a period of at least 8 months. The addition of the luxAB biomarker facilitates tracking in the root system of several plant species grown under sterile and non-sterile conditions. The correlation of the bioluminescent phenotype with the growth activity of P. putida S1B1 cells colonizing the root system of barley and corn plants was estimated by monitoring ribosomal contents using quantitative hybridization with fluorescence-labeled ribosomal RNA probes. A correlation between inoculum density, light output, and ribosomal contents was found for P. putida cells colonizing the root system of barley seedlings grown under sterile conditions. Although ribosomal contents, and therefore growth activity, of P. putida S1B1 cells extracted from the rhizosphere of corn plants grown in non-sterile soil were similar to those found in starved cells, the luminescent system permitted non-destructive in situ detection of the strain in the upper root system. ß 2000 Federation of European Microbiological Societies. Published by Elsevier Science B.V. All rights reserved. Keywords : luxAB; Rhizosphere; Genetically engineered microorganism; Deliberate release; Pseudomonas putida KT2440
1. Introduction The so-called £uorescent Pseudomonas group includes strains whose biochemical, physiological and genetic properties have been well characterized. A number of genetic tools ^ broad host range vectors, transposons and minitransposons, reverse genetics etc. ^ have made it possible
* Corresponding author. Tel.: +45 (45) 252770; Fax: +45 (45) 932809; E-mail : [email protected] 1
Present address: CNRS-LCB, 31 Chemin Joseph Aiguier, P.O. Box 71, Marseille 13492 Cedex 20, France. 2 Present address: NERI, Department of Microbiology and Biotechnology, Frederiksborgvej 399, Roskilde 4000, Denmark.
to design recombinant derivatives of this group of bacteria for use as microbial inoculants in agriculture for the biological control of soil-borne pests [1^3], for promoting plant growth [4,5], or for plant-assisted microbial elimination of pollutants [6,7]. Pseudomonas putida KT2440, a strain recognized as non-pathogenic by the National Institute of Health of the USA, derives from the soil bacterium P. putida mt-2, the natural host for the archetypal TOL plasmid pWW0 [8]. This strain has been shown to be highly versatile in the acquisition of recombinant DNA for expansion of its catabolic potential to remove pollutants [9], and in the expression of heterologous genes of interest in biological containment [10^13]. Furthermore, P. putida KT2440 and its derivatives survive well in unvegetated soils [14,15] and in the rhizosphere of plants grown both under greenhouse [16] and environmental conditions [12,16]. In
0168-6496 / 00 / $20.00 ß 2000 Federation of European Microbiological Societies. Published by Elsevier Science B.V. All rights reserved. PII: S 0 1 6 8 - 6 4 9 6 ( 0 0 ) 0 0 0 8 9 - 1
FEMSEC 1182 23-11-00
Cyaan Magenta Geel Zwart
92
C. Ramos et al. / FEMS Microbiology Ecology 34 (2000) 91^102
addition, the 6.1-Mb genome of P. putida KT2440 has been mapped [17], and its sequence, currently under assembly of contigs, is already available from The Institute for Genomic Research (TIGR), Rockville, MD, USA (http://www.tigr.org/cgi-bin/BlastSearch/blast.cgi?). All these characteristics make P. putida KT2440 particularly relevant for the development of microbial inoculants to be used in agricultural applications. Indeed, since the publication of the EU Council Directive on the deliberate release into the environment of genetically modi¢ed organisms (GMO) (90/220/EEC), 4 out of the 16 approved releases of Pseudomonas sp. have dealt with P. putida KT2440 and its derivatives. Initial approval of the deliberate release of a genetically engineered microorganism (GEM) requires not only detailed information on the genetic modi¢cation introduced to the parental strain, but also a documented description on its ecology and environmental impact. Assessment of environmental impact and of risks associated with environmental releases of GEMs requires knowledge of microbial survival, growth, activity and dispersal within the environment, and of the persistence of recombinant DNA and its transfer to the indigenous micro£ora. Surveillance and colonization of target environments by bacteria require adequate detection methods to monitor the released microorganism and to assess its activity. Although the green £uorescent protein (GFP) is one of the markers becoming more popular, mainly because unlike other biomarkers it does not require any substrate or additional cofactor to £uoresce, the GFP £uorescent phenotype does not indicate the metabolic status of the cells. One of the more promising markers for determination of cellular metabolic activity is bacterial luciferase, encoded by the luxAB genes. These genes are available in a wide variety of vectors and have been successfully used to study the colonization of plant roots by the potential biocontrol agents Pseudomonas £uorescens, Xanthomonas campestris and Enterobacter cloacae, and by several strains of Rhizobium (for a review see [18]). We have previously reported the construction of a bioluminescent derivative of P. putida KT2440, called P. putida S1B1, which was obtained after random mutagenesis with a mini-Tn5-luxAB-Km transposon. Bioluminescence production by this strain, expressed from a chromosomal promoter, did not a¡ect the growth characteristics of the strain under laboratory conditions [16]. Using this strain we showed that P. putida KT2440 colonizes and survives in the rhizosphere of a number of plant species, i.e. corn and broad bean, and that the introduction of this strain had no signi¢cant e¡ect on the survival of a variety of soil bacteria [16]. However, the above studies did not provide detailed information required for further applications and/or approval of derivatives of P. putida S1B1 as GEMs to be deliberately released into the environment, i.e., the locus interrupted by the reporter cassette into the chromosome of the parental strain, the e¡ect of such a modi¢cation on the competitive-
FEMSEC 1182 23-11-00
ness of the strain, and the spatial distribution and physiological activity of P. putida S1B1 cells colonizing the rhizosphere environment. In this study, we have addressed and described all these issues. 2. Materials and methods 2.1. Bacterial strains, plasmids and growth media P. putida S1B1 [16] is a bioluminescent derivative of the RifR P. putida strain KT2442 [19] containing in the chromosome a mini-Tn5-PluxAB-KmR . P. putida KT2442-5 is a SmR derivative of P. putida KT2442 (this study). Bacteria were grown at 30³C in LB medium or M9 minimal medium [20] with 28 mM glucose or 10 mM benzoate as the only carbon source. Antibiotics were added at the following concentrations (Wg ml31 ): kanamycin 50, rifampicin 20, and streptomycin 50. Bacterial inocula for seed coating or soil inoculation were prepared as follows : A colony of the corresponding P. putida strain was inoculated into 10 ml of LB medium and the culture was incubated at 30³C until the stationary phase was reached. Cells were washed twice in M8 bu¡er (50 mM phosphate, 100 mM NaCl) prior to their use. 2.2. DNA techniques Plasmid DNA was isolated by the alkaline lysis method using the QIAprep spin plasmid minipreps kit (Qiagen cat. no. 27104). Total DNA was isolated as described before [21], except for the 30-min incubation step at 55³C which was omitted. DNA digestions with restriction enzymes, ligations, and transformations, were performed by using standard procedures [22]. The insertion of the mini-Tn5/luxAB-KmR into the chromosome of P. putida S1B1 was located by sequencing the DNA region located upstream of the PluxAB genes using plasmid pS1KPN (see Section 3) as a template and oligonucleotide O (5P-CTGACTCTTATACACAAGT-3P), annealing to the `O' mini-Tn5 end sequence, as a primer. DNA walking was performed using the following speci¢c primers designed upon the available sequence: AGPD1, 5P-CATCAGCACCTTGGAGTCGAC-3P; CP2, 5P-ATGCCGCCGACCGTGGCGC-3P and; S1KPNPR, 5P-GCCTCTTGATATTGGCTAG-3P. DNA was sequenced with the dideoxy sequencing method, using ABI Prism dRhodamine terminators kit (Perkin-Elmer ref. 403042). 2.3. DNA and protein sequence analysis Similarity search with the incomplete P. putida KT2440 genome and preliminary DNA sequence data of the P. putida open reading frame (ORF)1 and ORF2 (see Section 3) were obtained from the TIGR website at http://www. tigr.org. using `TIGR BLAST' (Search Engine for Un¢n-
Cyaan Magenta Geel Zwart
C. Ramos et al. / FEMS Microbiology Ecology 34 (2000) 91^102
ished Microbial Genomes). ORFs were identi¢ed using `ORF Finder' from the Basic Local Alignment Search Tool (BLAST) at http://www.ncbi.nlm.nih.gov/gorf/ gorf.html. Promoter search was performed using `NNPP/ Prokaryotic' from the Human Genome Sequencing Center at http://dot.imgen.bcm.tmc.edu:9331/seq-search/genesearch.html. Translation of DNA sequences into protein sequences was obtained using the `Translate Tool' from the Expert Protein Analysis System (EXPASY) proteomics server of the Swiss Institute of Bioinformatics (SIB) at http://www.expasy.ch. Protein similarity search was performed using `Basic BLAST' at http://www.ncbi.nlm.nih.gov/blast. Helixturn-helix DNA-binding motifs and multiple protein sequence alignment were obtained using the programs `HTH Motif Prediction' [23] and `MULTIALIN Network Protein Sequence @nalysis', respectively, from Poªle Bio-informatique Lyonnais at http://pbil.ibcp.fr/cgi-bin/ npsa_automat.pl?page = /NPSA/npsa_server.html. 2.4. Unvegetated soil microcosm assays Two sandy-loam soils, a £uvisol soil (2.30% (w/w) organic matter, 6.0% (w/w) CaCO3 ) and a cambisol soil (0.63% (w/w) organic matter, 20.4% (w/w) CaCO3 ), were used in this study; before the soil was used it was sifted through a 4-mm mesh metal sieve. One milliliter of bacterial inoculum, prepared as described above, was added to the pots containing soil and evenly distributed to the indicated initial cell density in each assay. Cultivable cells (CFU per gram of soil) were estimated as described previously [14]. 2.5. Seed sterilization and coating with bacteria Corn (Zea mays), broad bean (Vicia faba var. Aguadulce) and tomato (Lycopersum esculentum) seeds were surface-sterilized for 20 min by soaking in 1 g l31 HgCl2 supplemented with 1 ml l31 Tween 80, and washed four times for 5 min in sterile water. Barley seeds (Hordeum vulgaris var. Alexis) were surface-sterilized by immersion in 70% (v/v) ethanol followed by treatment with 3% (v/v) hypochlorite as described by Kragelund and Nybroe [24]. Seed sterilization was veri¢ed by incubating 10^20 seeds on LB-agar at 25³C for several days without any contamination appearing. Sterile seeds were pre-germinated on moist ¢lter paper for 24 h prior to inoculation. Surface-sterilized seeds were coated with bacteria by soaking them for 30 min at 20³C in a 10-ml bacterial inoculum (see above) turbidimetrically adjusted to the appropriate cell density. The non-adhering liquid was decanted, and the seeds were air-dried for 10 min in a petri dish. Bacteria were washed from three seeds into 3 ml of 0.9% (w/v) NaCl, and serially diluted onto selective medium to determine the number of bacteria applied to the seeds.
FEMSEC 1182 23-11-00
93
2.6. Rhizosphere colonization experiments Glass tubes (15 cm highU3 cm diameter) were ¢lled with 200 ml of vermiculite and 75 ml water or 200 g of a £uvisol soil (2.3% (w/w) organic matter, 6.0% (w/w) CaCO3 ). Soil moisture was adjusted to 15% (w/w) with sterile water. Sterile corn seeds coated with bacteria were sown at about 1 cm below the soil or vermiculite surface, and the plants were germinated and grown in a greenhouse at 25³C with natural daily light period (approximately 11 h light, 13 h dark). Rhizosphere colonization of corn plants grown in vermiculite was assessed by photon counting imaging as described below. When soil was used, plants were removed from the pots after di¡erent periods of time and the number of CFU g31 of rhizosphere soil (the soil loosely attached to the roots) and the number of CFU g31 of bulk soil (the soil remaining into the pots) were estimated as described previously [14,15]. In co-inoculation experiments bacterial suspensions were mixed together before inoculation. Germination and growth of barley seedlings under sterile conditions was performed in M8 minimal medium plates (M9 minimal medium [20] containing no carbon or nitrogen sources) containing 1% (w/v) agar. The plates were incubated in the dark at 20³C for 3 days. Extraction of bacteria from the root system of agar-grown seedlings was performed by placing the roots into 2 ml of 0.9% (w/v) cold sodium chloride and mixing by vortexing during 3 min. Bacteria were counted on LB plates containing the appropriate antibiotics for the selection of the corresponding strain. As sterility control, bacteria were also counted on LB plates and no di¡erence in colony numbers was observed between the two media for any of the tested strains. 2.7. Whole-cell hybridization Cells were extracted from the rhizosphere as previously described [25]. Extracted cells as well as cells from batch cultures, were ¢xed in 3% (v/v) paraformaldehyde as described by Poulsen et al. [26]. Fixed cells were stored at 320³C in a storage bu¡er (50% ethanol, 10 mM Tris pH 7.5, 0.1% Nonidet P-40) until use. Hybridization of ¢xed cells was performed as described by Poulsen et al. [26] using 16S rRNA probe PP986, an oligonucleotide probe speci¢c for the P. putida subgroup A labeled with the incarbocyanine £uorescent dye CY3 [27]. 2.8. Epi£uorescence microscopy An Axioplan epi£uorescence microscope (Carl Zeiss) was used to visualize the hybridizations. The microscope was equipped with a 100-W mercury lamp, and ¢lter set XF40 (Omega Optical, Brattlebror, VT, USA) was used to visualize CY3. A slow-scan charge-coupled device (CCD) camera CH250 (Photometrics, Tucson, AZ, USA)
Cyaan Magenta Geel Zwart
94
C. Ramos et al. / FEMS Microbiology Ecology 34 (2000) 91^102
equipped with a KAF 1400 chip (pixel size, 6.8 by 6.8 Wm), was used for capturing digitilized images (PMIS). The camera was operated at 340³C, and the chip was read out in 12 bits (4096 intensity levels) at a rate of 200 kHz. For quanti¢cation of £uorescence, the PMIS images were analyzed using the Cellstat program [28]. A DOSbased 486 computer was used as a controller for the CCD camera, and a SUN IPX computer was used to run Cellstat. 2.9. Luciferase activity of intact cells in liquid cultures of P. putida S1B1 Light emission was determined as described by Rattray et al. [29]. To 0.5 ml of a cell suspension with a turbidity of 0.01 at 660 nm, we added 0.4 ml of 0.01% (v/v) n-decyl aldehyde and recorded the time course of light emission immediately thereafter for 1 min in a LKB1250 luminometer. The activity is given as the peak height, in relative light units (RLU), per turbidity unit. Light output per unit turbidity in cultures of P. putida S1B1 growing on LB or minimal medium with benzoate as the only C-source was relatively constant during the exponential growth phase; thereafter light output decreased. On LB medium light output during the exponential growth phase, around 100 RLU per OD660 , was clearly lower than when cells were growing on M9 minimal medium supplemented with benzoate (around 1000 RLU per OD660 ) (data not shown). When the number of CFU g31 of soil was below our limit of detection by colony plating (50 CFU g31 of soil), bioluminescence detection of P. putida S1B1 cells enriched in LB directly from soil was performed as described by Shaw et al. [30]. One gram of the soil sample was placed in 9 ml of LB medium supplemented with the appropriate antibiotics, and containing 40 Wg ml31 of cycloheximide to inhibit fungal growth. After 24 h of incubation at 30³C, aliquots (0.5 ml) were removed and used to determine bioluminescence activity as described above. 2.10. Luminescence detection by photon counting imaging A Hamamatsu C-2400 photon counting intensi¢ed CCD camera and an Argus-50 image processor were used to obtain images of transmitted photons. The camera was enclosed in a light-tight box with a removable door to reduce entry of external light during dark-¢eld exposures. Bright- and dark-¢eld images were superimposed with the aid of Argus 10 software. Images were then processed with Adobe Photoshop software. Plants were removed from the soil, vermiculite or agar systems and the roots were placed on ¢lter paper soaked with 0.01% (w/v) n-decyl aldehyde. For luminescence visualization with the CCD camera, samples were exposed for 3 s in the light to obtain a bright-¢eld image. The dark-¢eld exposure time was 10^90 s.
FEMSEC 1182 23-11-00
For microscopic examination of bioluminescence in root samples the photon counting intensi¢ed CCD was connected to an Axioplan microscope (Carl Zeiss). A Plan Neo£uar 2.5U/0.075 objective and an Epiplan 50U/0.50 were used. Samples were exposed for 10 s in the light to obtain a bright-¢eld picture. The dark-¢eld exposure time was 10 min. 3. Results and discussion 3.1. Determination of the insertion point of the mini-Tn5-PluxAB-Km transposon in the chromosome of P. putida S1B1 The integration of reporter systems into the chromosome of bacterial strains by transposon mutagenesis is an accepted routine practice, as long as the chromosomal insertions do not have phenotypic consequences in relation to the investigations carried out. P. putida strain S1B1 carries a mini-Tn5-PluxAB transposon inserted into the chromosome making the cells luminescent after exposure to n-decyl aldehyde [16]. In this transposon, the luxAB genes are transcribed only if the insertion occurs downstream of an active promoter. To identify the integration point of the mini-Tn5 transposon in the chromosome of P. putida S1B1, a DNA fragment located upstream of the mini-Tn5 `O' end was cloned and sequenced. Chromosomal DNA from P. putida S1B1 was digested with KpnI, which does not cut within the luxAB-km resistant determinant genes and the resulting DNA fragments were cloned into pUC19 [31]. The ligation mix was transformed into Escherichia coli DH5K and a bioluminescent transformant was selected on LB plates containing Km. A plasmid containing a 6.5-kb KpnI insert, called pS1KPN, was isolated from this transformant. Restriction enzyme mapping of this plasmid showed that the 6.5-kb fragment contained indeed the mini-Tn5/luxAB-Km (data not shown). The DNA region located upstream of the PluxAB genes was sequenced using plasmid pS1KPN as a template. Oligonucleotide O, annealing to the `O' mini-Tn5 end sequence, was used as primer to obtain the immediate adjacent DNA sequence. A total of 2290 bp were sequenced by DNA walking using speci¢c primers designed from the available sequences (see Section 2). Similarity search using TIGR BLAST (Search Engine for Un¢nished Microbial Genomes) yielded 98% identity with the 3P and 5P ends of contigs number 356 (8726 bp) and 340 (4645 bp), respectively, of the P. putida KT2440 un¢nished genome sequence. The nucleotide sequence of these two contigs overlapped 564 bp (Fig. 1). BLAST search of the corresponding 2290-bp sequence obtained from the P. putida KT2440 genome sequence identi¢ed two incomplete ORFs transcribed divergently and separated by 129 bp (Fig. 1). The mini-Tn5/luxAB-Km was located within ORF1 downstream codon number 227. The complete nu-
Cyaan Magenta Geel Zwart
C. Ramos et al. / FEMS Microbiology Ecology 34 (2000) 91^102
Fig. 1. Location of the mini-Tn5/luxAB/KmR into the chromosome of P. putida S1B1. A schematic representation of ORF1 and ORF2 into contigs numbers 340 and 356 from the P. putida KT2440 genome sequence, the direction of the transcription is indicated by the arrows. The triangle indicates the location of the mini-Tn5 within ORF1. The location of the 2290-bp sequence obtained from plasmid pS1KPN within ORF1 and ORF2 at the nucleotide sequence of the putative promoter located upstream of ORF1 are indicated.
cleotide sequence of P. putida KT2440 ORF1 and ORF2, accession numbers AF289509 and AF289508, respectively, was obtained from the sequence of contigs 356 and 340. We have not determined the transcription initiation point of ORF1, but a putative promoter sequence was identi¢ed 74 bp upstream the ATG initiation codon of this locus using program NNPP/Prokaryotic (Fig. 1). The translated sequence obtained from ORF1 (1006 residues) had a 67 and 65% similarity with the hypothetical proteins translated from the ydiJ locus from E. coli (D90811) and Haemophilus in£uenzae (U32796), respectively (Fig. 2). The 580 N-terminal residues of the P. putida KT2440 YdiJ protein produced various degrees of similarity with glycolate oxidase subunit D (GlcD) from Bacillus subtilis (Z99118, 44%), Helicobacter pylori (AE000565, 42%), Synechocystis sp. (D64001, 41%), E. coli (AE000380, 40%), Deinoccocus radiodurans (AE002015, 40%) and Archaeoglobus fulgidus (AE001049, 39%) (see Fig. 2 for homology with GlcD from E. coli and Synechocystis sp.). Various degrees of similarity were also found at the N-terminal end of the P. putida YdiJ protein with D-lactate dehydrogenase from Aquifex aeolicus (AE000755, 42%), A. fulgidus (AE001077, 42%), H. pylori (AE001542, 42%) and yeast (Z74031, 39%). Interestingly, the 426 residues located at the C-terminal end of the P. putida YdiJ protein had similarity with another subunit of glycolate oxidase (the iron-sulfur subunit GlcF) from E. coli (L43490, 37%) and Synechocystis sp. (D90903, 33%) (Fig. 2), however, the function of this hypothetical protein is completely unknown. The translated 296 amino acid sequence from ORF2 had signi¢cant similarity with several transcriptional regulators belonging to the LysR family: the regulator of the glycine cleavage system (GcvA) from E. coli (X73413, 54%), the trpBA operon transcriptional activator (TrpI) from Pseudomonas syringae (M95710, 52%), the AmpR protein from several bacteria, i.e. E. cloacae (AB003148, 50%) and Pseudomonas aeruginosa (X67095, 47%), the LysR-type L-lactamase transcriptional regulator PenR
FEMSEC 1182 23-11-00
95
from Burkholderia cepacia (U85041, 46%) and the transcriptional regulator FonR-1 from Serratia fonticola (AJ251239, 49%). A helix-turn-helix DNA-binding motif was detected using program HTH [23] between residues 25 and 46 of the P. putida ORF2 amino acid sequence (ESYTRAAAELSLTQSAVSRQVQ). This region was found to align exactly with the DNA-binding motifs detected for all the other transcriptional activator proteins showing similarity to the P. putida ORF2-translated sequence (data not shown). The possible function of the ORF2 protein as a regulator of the expression of the ydiJ locus was not studied here. 3.2. Competitiveness of P. putida strain S1B1 in unvegetated soil and in the soil rhizosphere The ecological performance of rhizosphere colonizing bacteria is a complex phenotype likely to be a¡ected by many di¡erent traits. Gene inactivation by transposon mutagenesis has been successfully used to identify bacterial traits involved in rhizosphere competence, e.g. motility, synthesis of the O-antigen lipopolysaccharide, thiamine production, amino acid synthesis, biotin production and a site-speci¢c recombinase (for a review see [32]). We have previously reported that the insertion of the mini-Tn5 cassette into the chromosome of P. putida S1B1 does not a¡ect the growth characteristics of the strain under laboratory conditions [16]. Since the function and ecological role of the P. putida ydiJ locus is completely unknown, it was, however, necessary to further test whether interruption of this locus a¡ected the survival of the strain in soil and the plant rhizosphere in comparison with its parental strain P. putida KT2442. The behavior of P. putida S1B1 and its parental strain KT2442 in non-sterile soil was similar. P. putida strains KT2442 or S1B1 were introduced at a cell load of 105 CFU g31 of soil in pots containing 90 g of either a cambisol soil, poor in organic matter, or a £uvisol soil relatively rich in organic matter. The pots were incubated at 17^22³C as described previously [14]. In the £uvisol soil, both strains maintained relatively constant cell numbers for at least 3 months (data not shown). In contrast, in the cambisol soil the number of CFU g31 of soil for both bacteria decreased steadily with time, so that after 56 days of incubation none of the strains could be detected after plating in selective culture plates (data not shown). Detection with antibiotic markers has been successfully used to show survival in soil of P. putida KT2440 and its derivatives [14,15]. Due to the decline of the bacterial population size after introduction into natural soil, most of the studies have focused on the behavior of these microbes during the initial period after introduction into the microcosms, usually 4 and 8 weeks [9]. The bioluminescent phenotype of P. putida S1B1 cells allowed a study of the longterm survival of P. putida S1B1 in the non-sterile soil microcosms described above. At every sampling time, P.
Cyaan Magenta Geel Zwart
96
C. Ramos et al. / FEMS Microbiology Ecology 34 (2000) 91^102
FEMSEC 1182 23-11-00
Cyaan Magenta Geel Zwart
C. Ramos et al. / FEMS Microbiology Ecology 34 (2000) 91^102
97
Fig. 2. Sequence alignment of YdiJ proteins and glycolate oxidase subunits D and F. The sources of the sequences were: P. putida YdiJ (this study, AF289509), E. coli YdiJ [41], H. in£uenzae YdiJ [42], E. coli GlcD and GlcF [43] and Synechocystis sp. GlcD and GlcF [44]. The MULTIALIN program (NPSA, Lyon, France) was used. Residues that are identical in at least four aligned proteins are highlighted in black, residues that are identical in the three YdiJ proteins are highlighted in gray.
putida S1B1 cells were enriched from 1 g of soil on selective LB-medium (see Section 2) and after a period of 20 h, luminescence emission was measured. Soil inocula produced detectable bioluminescence at every sampling time over a period of at least 8 months (data not shown), demonstrating that although the number of cells present in the soil was below the detection limit by plating in selective plates (50 CFU g31 of soil), metabolically active P. putida S1B1 cells were present in the soil until the end of the
FEMSEC 1182 23-11-00
assay. We previously established that suspensions of 1 g of soil containing as few as 5 CFU of P. putida S1B1 in 10 ml LB medium resulted in detectable luminescence after about 20 h of growth at 30³C. The insertion of the mini-Tn5 cassette into the chromosome of P. putida S1B1 did not a¡ect the ability of the strain to colonize the rhizosphere of corn, barley or bean plants in comparison with its parental strain P. putida KT2442. Corn seeds were coated with either P. putida
Cyaan Magenta Geel Zwart
98
C. Ramos et al. / FEMS Microbiology Ecology 34 (2000) 91^102
3.3. Distribution of P. putida S1B1 cells in the rhizosphere
Fig. 3. Survival of P. putida strains KT2442 (open symbols) and S1B1 (closed symbols) in the rhizosphere of corn. Corn seeds were coated with 108 CFU of the corresponding strain, sown in £uvisol-soil and incubated in a greenhouse. Bacteria in the rhizosphere (circles) and in the bulk soil (squares) were counted on LB plates containing the appropriate antibiotics. Con¢rmation that the colonies selected on media containing Rif and Km indeed represented the bioluminescent P. putida S1B1 strain was obtained by the fact that all the colonies become luminescent after exposure to n-decyl aldehyde. Each point represents an average based on three independent plants, the error bars indicate standard deviations from the mean.
KT2442 or P. putida S1B1 to about 104 , 106 and 108 CFU per seed. After 7 days of plant growth in non-sterile £uvisol soil we detected about 107 ^108 CFU g31 of rhizosphere soil regardless of the initial inoculum size and of the bacterial strain used. Thereafter the number of bacteria in the rhizosphere declined to about 106 ^107 CFU g31 of rhizosphere soil after 21 days and then remained relatively constant at this cell density during at least 42 days. However, at any given sampling time the number of CFU g31 of soil was generally two to three orders of magnitude higher in the rhizosphere than in the bulk soil (see Fig. 3 for results obtained when the seeds were coated with 108 CFU). Similar results were obtained when bean and barley plants were used instead of corn (data not shown). The behavior of the strain in direct competition with a streptomycin-labeled P. putida KT2442-5 derivative was also investigated. Corn seeds were coated with 107 CFU of each strain and sown in non-sterile £uvisol soil. The cells of each strain in the rhizosphere of corn were counted, and similar numbers, on the order of 106 ^107 CFU g31 of soil, were found in assays lasting 42 days (data not shown). All these results together show that interruption of the ydiJ locus does not have an e¡ect on the ¢tness of P. putida S1B1 either in unvegetated soil or in the soil rhizosphere.
FEMSEC 1182 23-11-00
Bioluminescence production by P. putida S1B1 is so intense that it can be detected at the single cell level using a photon counting camera (C. Sternberg, The Technical University of Denmark, personal communication). We therefore used luminescence-based detection of lux-labeled cells to examine in situ the distribution of this strain in the rhizosphere of di¡erent plant species. Sterile corn seeds coated with about 106 CFU per seed of P. putida strain S1B1 were germinated in vermiculite or non-sterile soil. After di¡erent periods of time, bioluminescent emission was assessed through photon imaging. Detection of luminescence from P. putida S1B1 cells in the rhizosphere of vermiculite-grown corn plants 7^21-day-old showed that luminescence was closest to the seed part of the root, although other zones of intense light emission were also observed throughout the entire root system (not shown), indicating a non-uniform distribution of bacteria in the rhizosphere. This is consistent with data on root colonization by plant growth-promoting lux [33,30] and lacZmarked [34] Pseudomonas bacteria. Dark-¢eld images of P. putida S1B1 cells in the root systems of broad bean, tomato and barley showed a similar pattern of colonization (data non-shown). Microscopic examination (magni¢cationU25) of a bright spot located in the upper root system showed two parallel bioluminescent lines along the root segment (Fig. 4A,B). A higher magni¢cation (500U) showed that bioluminescent lines were composed of bright spots of di¡erent sizes and intensities (Fig. 4C,D). This pattern of bioluminescence probably re£ects the localization of the bacterial cells (single cells and microcolonies) at the border of adjacent plant root cells, in agreement with data reported for other Pseudomonas strains [35,36,25]. 3.4. Activity of P. putida S1B1 cells in the rhizosphere Luciferase activity has been shown to be proportional to biomass in growing bacterial populations. However, under nutrient-limited conditions, such as those of the soil and rhizosphere environment, the concentration of FMNH2 required for the luciferase reaction becomes limiting, and the bioluminescence of the population decreases substantially compared to that of the biomass [37^39]. In this study, we combined luminescence-based detection of luxlabeled cells with quantitative hybridization with £uorescence labeled ribosomal RNA probes to ascertain the dependence of the luminescent phenotype on the physiological status of P. putida S1B1 cells colonizing the rhizosphere environment. Previously, we found that the ribosomal contents, and therefore growth activity, of P. putida cells (P. putida JB156, [40]) colonizing the root system of barley seedlings grown under sterile conditions was proportional to the increment in bacterial biomass [25]. Sterile barley seeds were inoculated with P. putida strain
Cyaan Magenta Geel Zwart
C. Ramos et al. / FEMS Microbiology Ecology 34 (2000) 91^102
99
Fig. 4. In situ visualization of bioluminescence from of P. putida S1B1 cells in the root system of corn. Sterile corn seeds were coated with approximately 106 CFU per seed of P. putida strain S1B1, sown in vermiculite and incubated in a greenhouse. Bright-¢eld exposure (10 s) (A and C) and dark-¢eld exposure (10 min) (B and D) of a 7-day-old root after the addition of n-decyl aldehyde. A and B, magni¢cationU25; C and D, magni¢cationU500. Bright- and dark-¢eld images were processed with Adobe Photoshop software. Relative intensity of light emission is indicated by the color scale.
S1B1 at four di¡erent cell loads, from approximately 103 to about 107 CFU per seed, and germinated at 20³C in the dark in agar plates containing no carbon or nitrogen sources. At the time of inoculation, only the seeds inoculated with a high cell density (between 107 and 105 CFU per seed) showed detectable bioluminescence (data not shown). During the ¢rst day of plant growth and regardless of the initial inoculum size, cell density increased to approximately the same level in all the roots. However, light intensity in the roots, and therefore the metabolic activity of the cells, increased with decreasing inoculum cell density (Fig. 5). Light intensity also correlated with the ribosomal contents of the cells extracted from the roots of the 1-day-old barley seedlings. The ribosomal content of the extracted cells, normalized to 1 for the starved cells used as inoculum, varied between 1.53 (initial
FEMSEC 1182 23-11-00
cell density of approximately 103 CFU per seed) and 0.60 (initial cell density of approximately 107 CFU per seed). During the following 2 days, the number of CFU increased in a similar fashion in all the roots (Fig. 5), whereas the light output (not-shown) and ribosomal contents of the cells extracted from the roots decreased in a similar fashion for all the plants (Fig. 5). We also assessed the correlation of the luminescent phenotype with the physiological status of P. putida S1B1 cells colonizing the rhizosphere in non-sterile £uvisol soil. A stationary culture of P. putida strain S1B1 was used to inoculate sterile corn seeds at a cell load of approximately 105 CFU per seed. Coated seeds were planted in the soil systems and the pots were incubated at 25³C. Six days after planting, luminescence was detected only close to the seed part of the root (not shown). The lack of detect-
Cyaan Magenta Geel Zwart
100
C. Ramos et al. / FEMS Microbiology Ecology 34 (2000) 91^102
Fig. 5. Cell density, light emission and ribosomal contents of P. putida S1B1 cells colonizing the root system of barley seedlings as a function of the inoculum size. Sterile barley seeds were coated at the indicated cell density of P. putida strain S1B1 and germinated in agar plates at 20³C for 3 days. Bioluminescence in the root system of the developing barley seedlings was detected by photon count with a CCD camera. Bright-¢eld exposures (3 s) and dark-¢eld exposures (30 s) were processed with Adobe Photoshop software. Relative intensity of light emission is indicated by the color scale. Ribosomal contents were estimated by quantitative hybridization with £uorescently labeled rRNA. The £uorescence intensity was normalized to one relative £uorescence unit (r.f.u.) for the starved cells used as inoculum. The values are mean þ standard deviations based on the values obtained from three different roots.
able luminescence in the lower root segments does not necessarily indicate absence of cells; luminescence emission could be below the detection limit by photon-counting imaging due to low cell number and/or low physiological activity of the cells. Other authors [39] have reported that luciferase activity of luxAB-tagged Pseudomonas strains decreases rapidly after bacterial introduction into soil due to a decrease in cellular metabolic activity, however, the light emitted by cells extracted from soil can be quanti¢ed by luminometry in cell suspensions containing a relatively high number of viable cells (104 ^105 CFU ml31 ). Although the number of CFU per cm of root was found to be approximately two orders of magnitude higher in the upper (luminescent, 105 CFU per cm) than in lower (nonluminescent, 103 CFU per cm) root segment, the average ribosomal content and cell volume of the cells extracted from the upper root segment (1 cm) was about 3.2 and 1.4
FEMSEC 1182 23-11-00
times lower, respectively, than that of the starved cells used as inoculum. We have previously reported that a signi¢cant decrease in the ribosomal content and cell volume of P. putida JB156 cells occurs immediately after bacteria introduction in the rhizosphere of barley plants grown in non-sterile soil [25]. To our knowledge, this is the ¢rst study correlating the bioluminescent phenotype with the ribosomal contents, and therefore growth activity, of luxAB-tagged bacterial cells introduced into environmental samples. In agreement with data reported by other authors [37^39], our results show that the bioluminescent phenotype of P. putida S1B1 is strongly dependent on the growth activity of the cells. However, even at the low growth rates supported by root exudate, the reporter system allows in situ detection of high density of cells colonizing the root surface of di¡erent plant species.
Cyaan Magenta Geel Zwart
C. Ramos et al. / FEMS Microbiology Ecology 34 (2000) 91^102
4. Concluding remarks The addition of a luxAB marker into the chromosomal ydiJ locus of the non-aggressive root colonizer P. putida KT2440, P. putida strain S1B1, does not a¡ect the competitiveness of the strain in unvegetated soil or the rhizosphere of several plant species. The bioluminescent phenotype allows in situ tracking of this strain after introduction into soil and rhizosphere samples, facilitating assessment of the environmental impact and of risk associated with the environmental release of this strain and its derivatives. Acknowledgements
[12]
[13]
[14]
[15]
[16]
This work was partly supported by a grant to S.M. from the Danish Biotechnology Programme and by Grant BIO97-0641 to J.L.R. from CICYT, Spain. C.R. was partially supported by EMBO fellowship ASTF8056 and by The Carlsberg Foundation (Copenhagen, Denmark). Preliminary sequences from the P. putida KT2440 genome were obtained from The Institute of Genomic Research.
[17]
[18]
[19]
References [20] [1] O'Sullivan, D.J. and O'Gara, F. (1992) Traits of £uorescent Pseudomonas spp. involved in suppression of plant root pathogens. Microbiol. Rev. 56, 662^676. [2] Dowling, D.N. and O'Gara, F. (1994) Metabolites of Pseudomonas involved in the biocontrol of plant disease. Trends Biotechnol. 12, 133^141. [3] Thomashow, L.S. (1996) Biological control of plant root pathogens. Curr. Opin. Biotechnol. 7, 343^347. [4] Lugtenberg, B.J.J., deWeger, L.A. and Bennett, J.W. (1991) Microbial stimulation of plant growth and protection from disease. Curr. Opin. Biotechnol. 2, 457^464. [5] Kloepper, J.W., Li¡shitz, R. and Zablotowicz, R.M. (1989) Free living bacterial inocula for enhancing crop productivity. Trends Biotechnol. 7, 39^44. [6] Crowley, D.E., Alvey, S. and E.S. Gilbert (1997) Rhizosphere ecology of xenobiotic-degrading microorganisms. In: Phytoremediation of Soil and Water Contaminants (Kruger, E.L. Anderson, T.A. and Coats, J.R., Eds.), ACS Symposium Series 664, pp. 19^34. American Chemical Society, Washington, DC. [7] Brazil, G.M., Kene¢ck, L., Callanan, M., Haro, A., deLorenzo, V., Dowling, D.N. and O'Gara, F. (1995) Construction of a rhizosphere Pseudomonad with potential to degrade polychlorinated biphenyls and detection of bph gene expression in the rhizosphere. Appl. Environ. Microbiol. 61, 1946^1952. [8] Worsey, M.J. and Williams, P.A. (1974) Metabolism of toluene and xylenes by Pseudomonas putida (arvilla) mt-2: evidence for a new function of the TOL plasmid. J. Bacteriol. 124, 7^13. [9] Ramos, J.L., D|¨az, E., Dowling, D., de Lorenzo, V., Molin, S., O'Gara, F., Ramos, C. and Timmis, K.N. (1994) The behavior of bacteria designed for biodegradation. BioTechnology 12, 1349^1356. [10] Ronchel, M.C., Ramos, C., Jensen, L.B., Molin, S. and Ramos, J.L. (1995) Construction and behaviour of biologically contained bacteria for environmental applications in bioremediation. Appl. Environ. Microbiol. 61, 2990^2994. [11] Ramos, J.L., Andersson, P., Jensen, L.B., Ramos, C., Ronchel, M.C.,
FEMSEC 1182 23-11-00
[21]
[22]
[23]
[24]
[25]
[26]
[27]
[28]
[29]
101
D|¨az, E., Timmis, K.N. and Molin, S. (1995) Suicide microbes on the loose. BioTechnology 13, 35^37. Molina, L., Ramos, C., Ronchel, M.C., Molin, S. and Ramos, J.L. (1998) Construction of an e¤cient biologically contained Pseudomonas putida strain and its survival in outdoor assays. Appl. Environ. Microbiol. 64, 2072^2078. Ronchel, M.C., Molina, L., Witte, A., Lutbiz, W., Molin, S., Ramos, J.L. and Ramos, C. (1998) Characterization of cell lysis in Pseudomonas putida upon expression of heterologous killing genes. Appl. Environ. Microbiol. 64, 4904^4911. Ramos, J.L., Duque, E. and Ramos-Gonza¨lez, M.I. (1991) Survival in soils of an herbicide-resistant Pseudomonas putida strain bearing a recombinant TOL plasmid. Appl. Environ. Microbiol. 57, 260^266. Duque, E., Marque¨s, S. and Ramos, J.L. (1993) Mineralization of p-methyl-14 C-benzoate in soils by Pseudomonas putida (pWW0). Microb. Releases 2, 175^177. Molina, L., Ramos, C., Duque, E., Ronchel, M.C., Garc|¨a, J.M., Wyke, L. and Ramos, J.L. (2000) Survival of Pseudomonas putida KT2440 in soil and the rhizosphere of plants under greenhouse and environmental conditions. Soil Biol. Biochem. 32, 315^321. Ramos-D|¨az, M.A. and Ramos, J.L. (1998) Combined physical and genetic map of the Pseudomonas putida KT2440 chromosome. J. Bacteriol. 180, 6352^6363. Prosser, J.I. (1994) Molecular marker systems for detection of genetically engineered micro-organisms in the environment. Microbiology 140, 5^17. Franklin, F.C.H., Bagdasarian, M.M., Bagdasarian, M. and Timmis, K.N. (1981) Molecular and functional analysis of the TOL plasmid pWW0 from Pseudomonas putida and cloning of the genes for the entire regulated aromatic ring meta cleavage pathway. Proc. Natl. Acad. Sci. USA 78, 7458^7462. Abril, M.A., Micha¨n, C., Timmis, K.N. and Ramos, J.L. (1989) Regulator and enzyme speci¢cities of the TOL plasmid-encoded upper pathway for degradation of aromatic hydrocarbons and expansion of the substrate range of the pathway. J. Bacteriol. 171, 6782^ 6790. Ramos-Gonza¨lez, M.I., Ru|¨z-Cabello, F., Brettar, I., Garrido, F. and Ramos, J.L. (1992) Tracking genetically engineered bacteria : monoclonal antibodies against surface determinants of the soil bacterium Pseudomonas putida 2440. J. Bacteriol. 174, 2978^2985. Ausubel, F.M., Brent, R., Kingston, R.E., Moore, D.D., Seidman, J.G., Smith, J.A. and Struhl, K. (Eds.), (1991) Current Protocols in Molecular Biology, John Wiley and Sons, New York. Dodd, I.B. and Egan, J.B. (1990) Improved detection of helix-turnhelix DNA-binding motifs in protein sequences. Nucleic Acids Res. 18, 5019^5026. Kragelund, L. and Nybroe, O. (1996) Competition between Pseudomonas £uorescens Ag1 and Alcaligenes eutrophus JMP134 (pJP4) during colonization of barley roots. FEMS Microbiol. Ecol. 20, 41^51. Ramos, C., MÖlbak, L. and Molin, S. (2000) Bacterial activity in the rhizosphere analyzed at the single-cell level by ribosomal contents and synthesis rates. Appl. Environ. Microbiol. 65, 801^809. Poulsen, L.K., Ballard, G. and Stahl, D.A. (1993) Use of rRNA £uorescence in situ hybridization for measuring the activity of single cells in young and established bio¢lms. Appl. Environ. Microbiol. 59, 1354^1360. MÖller, S., Sternberg, C., Andersen, J.B., Christensen, B.B., Ramos, J.L., Givskov, M. and Molin, S. (1998) In situ gene expression in mixed-culture bio¢lms : evidence of metabolic interactions between community members. Appl. Environ Microbiol. 64, 721^732. MÖller, S., Kristensen, C.S., Poulsen, L.K., Carstensen, J.M. and Molin, S. (1995) Bacterial growth on surfaces : automated image analysis for quanti¢cation of growth rate-related parameters. Appl. Environ. Microbiol. 61, 741^748. Rattray, E.A.S., Prosser, J.I., Killham, K. and Glover, L.A. (1990) Luminescence-based nonextractive technique for in situ detection of Escherichia coli in soil. Appl. Environ. Microbiol. 56, 3368^3374.
Cyaan Magenta Geel Zwart
102
C. Ramos et al. / FEMS Microbiology Ecology 34 (2000) 91^102
[30] Shaw, J.J., Dane, F., Geiger, D. and Kloepper, J.W. (1992) Use of bioluminescence for detection of genetically engineered microorganisms released into the environment. Appl. Environ. Microbiol. 58, 267^273. [31] Sambrook, J., Fritsch, E.F. and Maniatis, T. (1989) Molecular Cloning: a Laboratory Manual, 2nd edn., Cold Spring Harbor, NY. [32] Lugtenberg, B.J.J. and Dekkers, L.C. (1999) What makes Pseudomonas bacteria rhizosphere competent? Environ. Microbiol. 1, 9^13. [33] de Weger, L.A., Dunbar, P., Mahafee, W.F., Lugtenberg, B.J.J. and Sayler, G. (1991) Use of bioluminescence markers to detect Pseudomonas spp. in the rhizosphere. Appl. Environ. Microbiol. 57, 3641^ 3644. [34] Simons, M., van der Bij, A.J., Brand, I., de Weger, L.A., Wij¡elman, C.A. and Lugtenberg, B.J.J. (1996) Gnotobiotic system for studying rhizosphere colonization by plant growth-promoting Pseudomonas bacteria. Mol. Plant Microbe Interact. 9, 600^607. [35] Bloemberg, G.V., O'Toole, G.A., Lugtenberg, B.J.J. and Kolter, R. (1997) Green £uorescent protein as a marker for Pseudomonas spp.. Appl. Environ. Microbiol. 63, 4543^4551. [36] Hansen, M., Kragelund, L., Nybroe, O. and SÖrensen, J. (1997) Early colonization of barley roots by Pseudomonas £uorescens studied by immuno£uorescence technique and confocal laser scanning microscopy. FEMS Microbiol. Ecol. 23, 353^360. [37] Duncan, S.L.A., Killham, K. and Prosser, J.I. (1994) Luminescencebased detection of activity of starved and viable but nonculturable bacteria. Appl. Environ. Microbiol. 60, 1308^1316. [38] Meikle, A., Glover, L.A., Killham, K. and Prosser, J.I. (1994) Potential luminescence as an indicator of activation of genetically modi¢ed Pseudomonas £uorescens in liquid culture and in soil. Soil Biol. Biochem. 26, 747^755. [39] Unge, A., Tombolini, R., MÖlbak, L. and Jansson, J. (1999) Simultaneous monitoring of cell number and metabolic activity of speci¢c bacterial populations with a dual gfp-luxAB marker system. Appl. Environ Microbiol. 65, 813^821.
FEMSEC 1182 23-11-00
[40] Christensen, B.B., Sternberg, C., Andersen, J.B., Eberl, L., Moller, S., Givskov, M. and Molin, S. (1998) Establishment of new genetic traits in a microbial bio¢lm community. Appl. Environ. Microbiol. 64, 2247^2255. [41] Blattner, F.R., Plunkett III, G., Bloch, C.A., Perna, N.T., Burland, V., Riley, M., Collado-Vides, J., Glasner, F.D., Rode, C.K., Mayhew, G.F., Gregor, J., Davis, N.W., Kirkpatrick, H.A., Goeden, M.A., Rose, D.J., Mau, B. and Shao, Y. (1997) The complete genome sequence of Escherichia coli K-12. Science 277, 1453^1474. [42] Fleischmann, R.D., Adams, M.D., White, O., Clayton, R.A., Kirkness, E.F., Kerlavage, A.R., Bult, C.J., Tomb, J.-F., Dougherty, B.A., Merrick, J.M., McKenney, K., Sutton, G., Fitzhugh, W., Fields, C.A., Gocayne, J.D., Scott, J.D., Shirley, R., Liu, L.-I., Glodek, A., Kelley, J.M., Weidman, J.F., Phillips, C.A., Spriggs, T., Hedblome, E., Cotton, M.D., Utterback, T.R., Hanna, M.C., Nguyen, D.T., Saudek, D.M., Brandon, R.C., Fine, L.D., Fritchman, J.L., Fuhrmann, J.L., Geoghagen, N.S.M., Gnehm, C.L., McDonald, L.A., Small, K.V., Fraser, C.M., Smith, H.O. and Venter, J.C. (1995) Whole-genome random sequencing and assembly of Haemophilus in£uenzae Rd. Science 269, 496^512. [43] Pellicer, M.T., Badia, J., Aguilar, J. and Baldoma, L. (1996) glc locus of Escherichia coli: characterization of genes encoding the subunits of glycolate oxidase and the glc regulator protein. J. Bacteriol. 178, 2051^2059. [44] Kaneko, T., Sato, S., Kotani, H., Tanaka, A., Asamizu, E., Nakamura, Y., Miyajima, N., Hirosawa, M., Sugiura, M., Sasamoto, S., Kimura, T., Hosouchi, T., Matsuno, A., Muraki, A., Nakazaki, N., Naruo, K., Okumura, S., Shimpo, S., Takeuchi, C., Wada, T., Watanabe, A., Yamada, M., Yasuda, M. and Tabata, S. (1996) Sequence analysis of the genome of the unicellular cyanobacterium Synechocystis sp. strain PCC6803. II. Sequence determination of the entire genome and assignment of potential protein-coding regions. DNA Res. 3, 109^136.
Cyaan Magenta Geel Zwart