Flavobacterium psychrophilum is a Gram-negative fish pathogen that causes important economic losses in aquaculture worldwide. Although the genome of this bacterium has been determined, the function and relative importance of genes in relation to virulence remain to be established. To investigate their respective contribution to the bacterial pathogenesis, effective tools for gene inactivation are required. In the present study, a markerless gene deletion system has been successfully developed for the first time in this bacterium. Using this method, the F. psychrophilum fcpB gene, encoding a predicted cysteine protease homologous to Streptococcus pyogenes streptopain, was deleted. The developed system involved the construction of a conjugative plasmid that harbors the flanking sequences of the fcpB gene and an I-SceI meganuclease restriction site. Once this plasmid was integrated in the genome by homologous recombination, the merodiploid was resolved by the introduction of a plasmid expressing I-SceI under the control of the fpp2 F. psychrophilum inducible promoter. The resulting deleted fcpB mutant presented a decrease in extracellular proteolytic activity compared to the parental strain. However, there were not significant differences between their LD50 in an intramuscularly challenged rainbow trout infection model. The mutagenesis approach developed in this work represents an improvement over the gene inactivation tools existing hitherto for this “fastidious” bacterium. Unlike transposon mutagenesis and gene disruption, gene markerless deletion has less potential for polar effects and allows the mutation of virtually any non-essential gene or gene clusters.
Citation: Gómez E, Álvarez B, Duchaud E, Guijarro JA (2015) Development of a Markerless Deletion System for the Fish-Pathogenic Bacterium Flavobacterium psychrophilum. PLoS ONE 10(2): e0117969. doi:10.1371/journal.pone.0117969
Academic Editor: Ulrike Gertrud Munderloh, University of Minnesota, UNITED STATES
Received: June 25, 2014; Accepted: December 8, 2014; Published: February 18, 2015
Copyright: © 2015 Gómez et al. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited
Data Availability: The data are available in GenBank (http://www.ncbi.nlm.nih.gov/genbank/). The accession number of fcpB gene is KJ605411.
Funding: Funding provided by grant AGL2009-07003 from the Spanish Ministerio de Economía y Competitividad. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.
Competing interests: The authors have declared that no competing interests exist.
Flavobacterium psychrophilum is the etiological agent of the bacterial cold-water disease (BCWD) and rainbow trout fry syndrome (RTFS), which particularly affects juvenile rainbow trout (Oncorhynchus mykiss), causing important economic losses in salmonid aquaculture worldwide. The disease mainly appears when water temperatures range between 10°C and 14°C  and, as there is no commercial vaccine, its control requires the massive use of antibiotics.
It is considered a “fastidious” bacterium because it is difficult to isolate and manipulate . Therefore, the knowledge regarding its virulence factors is still fairly limited. In this respect, some factors associated with pathogenesis have been described. Among these, an iron acquisition system , adhesion ability , hemolytic activity [5–7] and a thiol:disulfide oxidoreductase  were reported to play role in virulence. Predicted virulence genes have also been identified by genomic analysis  and extracellular proteases have been proposed to play major function in the virulence of the bacterium [10, 11]. However, two of them (Fpp1 and Fpp2) were suggested to have a nutritional role .
The study of virulence factors of F. psychrophilum has been drastically hampered by the difficulty to genetically manipulate this organism. Transformation and conjugation frequencies are very low and strain specific , growth is slow  and recovery of cells on solid media comes up against the presence of viable but non-culturable cells . In spite of this, it is worth to highlight the development, in the last few years, of a transposon mutagenesis technique  and a GFP-based promoter probe vector . Furthermore, mutagenesis via single-crossover homologous recombination was successfully used for the mutation of the fpp1 and fpp2 genes  as well as other genes (Gomez et al., unpublished results). Nevertheless, this technique presents some potential limitations such as the generation of polar mutations, obtaining partially-inactivated proteins and reversibility of mutations. Therefore, additional tools are still needed in order to improve the genetic manipulation in this microorganism. The principal objective of this work was to develop a deletion mutagenesis method for F. psychrophilum, based on the one designed by Pósfai et al. (1999)  for Escherichia coli, which has been also used in Bacteroides fragilis , a species phylogenetically more closely related to F. psychrophilum and other bacteria [19–21]. The system is based on the I-SceI meganuclease, a restriction enzyme recognizing an 18 bp sequence that serves as a stimulator of the recombination process . Briefly, a suicide plasmid, which carries the flanking sequences of the gene to be deleted and the recognition site of I-SceI, is inserted into the genome by homologous recombination between the flanking sequences of the gene to be deleted and those matching sequences in the chromosome of the bacterium. Resolution of this cointegrate via intramolecular recombination, which is stimulated by the action of the I-SceI meganuclease, generates either mutant or wild type chromosome. The I-SceI coding sequence is present in a replicative plasmid that is introduced into the merodiploid clones and its expression is regulated by the calcium-temperature inducible promoter of the gen fpp2 .
The fcpB gene of F. psychrophilum THCO2/90 was selected for being deleted. This locus is specifically present in the genome of this virulent strain and codes for a protein homologous to the cysteine protease “streptopain”, a major virulence determinant in Streptococcus pyogenes [22–26]. ScpB is a critical virulence factor for invasive disease episodes [22, 23]. It also destroys most of signaling and antibacterial properties of chemokines expressed by an inflamed epithelium and increases capillary permeability and histamine release from mast cells . All of these properties, amongst others, make ScpB a major virulence determinant of S. pyogenes. Therefore, we hypothesize that the FcpB protein, which is homologous to ScpB, could have some relevance in the pathogenic process of F. psychrophilum THC02/90. The additional objective was then to know the implication of this gene in the pathogenic process of F. psychrophilum in a rainbow trout infection model.
Materials and Methods
Bacterial strains, plasmids, growth conditions and proteolytic activity
Escherichia coli strain S17–1λpir  was grown at 37°C in 2xTY medium (10 g tryptone, 10 g yeast extract, 5 g NaCl per liter) with 20 g/l agar added for solid medium. This strain was used to transfer DNA into F. psychrophilum. The F. psychrophilum THC02/90 strain was grown at 12°C or 18°C in nutrient broth (NB; Pronadisa; 5 g gelatin peptone, 3 g beef extract per liter) or nutrient broth containing 10 mM CaCl2 (NBF) . Nutrient agar (NA; NB containing 15 g/l agar) or nutrient agar charcoal [NAC; NA supplemented with activated charcoal (Sigma)] were used for solid cultures, as previously described . For selective growth of E. coli S17–1 λpir, 50 μg/ml streptomycin was used and transformants were selected with 100 μg/ml ampicillin. Selection of F. psychrophilum transconjugants was carried out with 10 μg/ml erythromycin or 10 μg/ml tetracycline. The plasmids and primers used are listed in Table 1. Colony spreading was analyzed according to Pérez-Pascual et al. (2010) . Briefly, plates containing one-sixth NA were inoculated in their center with 8 μl liquid culture in middle exponential growth phase. The plates were then incubated at 20°C, and spreading and biomass production was quantified at 120 h by measuring the colony diameter (mm) and OD525 of 1 ml aqueous suspensions of colonies, respectively. These experiments were performed in triplicate. Azocasein assays using supernatants of liquid cultures were carried out according to Secades et al. (2001) . One unit of enzyme activity (enzymatic units; EU) was defined as the amount of enzyme which yielded an increase in the A420 of 0.01 in 2 h at 30°C. The protease activity assays were performed in triplicate (supernatants were taken from 3 different cultures of each strain) and the protease activity was measured in triplicate for each sample.
Plasmid DNA was isolated using the “Gen Elute Plasmid miniprep” commercial kit and genomic DNA was purified with the “Gen Elute Bacterial Genomic DNA” kit, both from Sigma Aldrich Co., Switzerland. Routine DNA procedures such as DNA digestion with restriction enzymes, DNA ligations, and gel electrophoresis were performed essentially as described by Sambrook et al. (2001) . Restriction enzymes were from Takara Bio Co., Japan. All PCR products were amplified using Pfu polymerase (New England Biolabs, MA) according to the manufacturer recommendations. The oligonucleotides used in this study are described in Table 1. T4 DNA ligase was from Roche Diagnostics GmbH, Germany.
Generation of the fcpB- deletion mutant
Flanking sequences of the fcpB gene of approximately 1.5 kb each were amplified by PCR. Flanking regions were cloned consecutively into the multiple cloning site of pLYL03 plasmid (Table 1). The 3′ fcpB flanking sequence was amplified with the oligonucleotides FcpBdel3’_F and FcpBdel3’_R (Table 1) and cloned into the BamHI-XbaI sites whereas the 5′ fcpB flanking sequence was amplified with the primers FcpBdel5’_F and FcpBdel5’_R and cloned into the XbaI-PstI sites. The FcpBdel5’_ F primer contained an I-SceI restriction site. The resulting plasmid, named pLYL03–3S5, was transferred to F. psychrophilum THC02/90 from E. coli S17–1λpir by conjugation as described by Álvarez et al. (2004)  and transconjugants were selected in NAC with 10 μg/ml erythromycin. One hundred microliters of NB with erythromycin were inoculated with each clone in microtiter plates and those that showed superior growth were selected to verify, by PCR and Southern Blot, if they had the plasmid integrated into the chromosome. PCRs were performed with RP and FcpBdel5’ primers (Table 1), using genomic DNA from the different strains as template. Genomic DNA from F. psychrophilum THC02/90 and pLYL03–3S5 plasmid were used as negative and positive controls, respectively. To perform Southern Blot, HindIII digested genomic DNA was hybridized with a 970 bp DIG-labelled PCR probe (DIG DNA labelling mix, Roche Diagnostics GmbH, Germany) using the oligonucleotides FcpBdel5’_F and FcpBdel5’_R and genomic DNA from the parental strain as template. Strains with the plasmid pLYL03–3S5 integrated into the chromosome were named FcpB5+, if the recombination has occurred in the 5’ flanking sequence, and FcpB3+, if it has occurred in the 3’ flanking sequence (Table 1).
A replicative plasmid, derived from pCP23 (Table 1), was constructed to facilitate the next over-crossing event. The calcium and temperature-inducible promoter Pfpp2 from the fpp2 gene  was fused to the I-SceI coding region by cross-over PCR. The promoter sequence was amplified with the primers promfpp2_F and pfpp2_Sce_R (Table 1) and I-SceI coding sequence was amplified with Fus_Sce_F and Sce_R (Table 1), using the plasmids pCP23-Gfpp2 and pACBSR (Table 1) as templates, respectively. Both fragments were used as templates in a cross-over PCR with the primers promfpp2_F and Sce_R (Table 1), to obtain the fused fragment of 1.3 kb containing the Pfpp2 promoter in front of the I-SceI coding region (Pfpp2-SceI), This fragment was cloned into the plasmid pCP23-Gfpp2, previously digested with BamHI and SphI enzymes to replace the Pfpp2-gfpmut3 region for the fused Pfpp2-SceI fragment. The resulting plasmid was named pCP23-TFS. The pCP23-TFS plasmid was transferred from E. coli S17 1 λpir to F. psychrophilum FcpB5+ or FcpB3+ by conjugation as described by Álvarez et al. (2004) . Transconjugants were selected in NAC with tetracycline and 10 mM CaCl2 and incubated at 12°C. One hundred microliters of NB with tetracycline were inoculated with each clone in microtiter plates and those that showed superior growth (because they carried the pCP23-TFS plasmid) were selected to confirm, by PCR and Southern Blot, if the deletion of the fcpB gene had ocurred. PCR was performed, using the primers FcpBint_F and FcpBint_R (Table 1), and genomic DNA from F. psychrophilum THC02/90, FcpB5+ and FcpB3+ as controls. To confirm the fcpB gene deletion by Southern Blot, genomic DNA from F. psychrophilum THC02/90 and fcpB- mutant was digested with HindIII and hybridized with a 970 bp DIG-labelled PCR probe using the oligonucleotides FcpBint_F and FcpBint_R and genomic DNA from the parental strain as template.
Effect of pCP23-TFS in F. psychrophilum growth and I-SceI RT-PCR
The absence of I-SceI restriction sequences in the F. psychrophilum THC02/90 genome was assessed by in silico analysis. However, in order to discard any potential negative effect of I-SceI expression in F. psychrophilum, the plasmid pCP23-TFS was transferred from E. coli S17 1–λpir to F. psychrophilum THC02/90 by conjugation. Two hundred and fifty microliters of the resulting strain, named THC02/90-TFS (Table 1), was then inoculated to 25 ml of NB in 250 ml flasks and incubated at 12°C with shaking. The parental strain was used as a control. The optical density at 525 nm of both cultures was measured with a Hitachi U2900 Spectrophotometer over time. RT-PCR. Total RNA was obtained from 2 ml of early stationary phase cultures of F. psychrophilum THC02/90–23 (Table 1) and THC02/90-TFS strain, which were grown in NBF supplemented with 5 μg/ml tetracycline, at 12°C. RNA was isolated using a High Pure RNA Isolation kit (Roche) and treated with DNase I (RNase-free) (Ambion) to eliminate traces of DNA. Reverse transcription (RT-PCR) was performed using Superscript One-Step with Platinum Taq System (Invitrogen Life Technologies), using 20 ng of RNA in each reaction. In order to determine whether RNA was free of contaminant DNA, reactions omitting the reverse-transcription step were included in each run as negative controls. The primers used for RT-PCR, which target an intragenic region of I-SceI coding sequence, were RTSceI_F (nucleotides 54–77) and RTSceI_R (nucleotides 674–694). RT-PCR program was set following manufacturer’s indications, with an annealing temperature of 62°C and an elongation time of 50 s.
Animal experiments were performed in accordance with the European legislation governing animal welfare, and they were authorized and supervised by the Animal Experimentation Ethics Committee of Universidad de Oviedo (see S1 Information). Rainbow trout (O. mykiss) fingerlings weighing between 5 and 7 g used for the animal experiments were obtained from a commercial fish farm. Fish were acclimatized to experimental conditions and randomly selected fish were analysed to discard the presence of bacteria in spleen, gut and liver. Fish were kept in 60 l tanks at 12±1°C in continually flowing dechlorinated water. For LD50 determinations, F. psychrophilum cultures were grown to exponential phase, harvested by centrifugation and washed with PBS. Cells were resuspended in PBS and serial dilutions were prepared. Groups of 10 fish were challenged by intramuscular injection with 50 μl of dilutions containing 103–105 CFU and they were monitored twice daily. LD50 was calculated according to the PROBIT method using IBM SPSS Statistics 19.0 (Armonk, New York, USA), establishing a 95% confidence limit. LD50 experiments were performed in duplicate.
In silico analysis
FcpB (Flavobacterium Cysteine protease B) gene sequence was obtained from the ongoing genome project of F. psychrophilum THC02/90 (Duchaud E., unpublished data). The National Center for Biotechnology Information (NCBI) Basic Local Alignment Search Tool (BLAST) was used to compare protein sequences and Simple Modular Architecture Research Tools (SMART) to detect conserved domains. MotifScan software from MyHits was used to identify the motifs present in each sequence. The ProtParam program (ExPASy) was used for molecular mass computation and SignalP3.0 (Center for Biological Sequence Analysis; CBS) to predict the location of a signal peptide cleavage site. The MEROPS peptidase database was used to classify the protease. Accession number of fcpB gene is GenBank KJ605411.
Analysis of the fcpB gene of F. psychrophilum THC02/90
The fcpB gene (locus THC0290–0491) is encompassed in a large genomic island, specific of the THC02/90 strain (Duchaud, unpublished results). This region has an unusually high proportion of genes of unknown function and also contains an important number of transposases-encoding genes belonging to different families (Duchaud, unpublished results). The fcpB gene is flanked by the loci 0492 and 0490 encoding a putative transmembrane protein of unknown function and a hypothetical lipoprotein precursor, respectively. Putative promoter and terminator sequences were also identified (Fig. 1A).
(A) Gene organization of the fcpB locus in the F. psychrophilum genome. The position of each gene and the direction of the transcription are shown by arrows. The putative fcpB promoters (P), ribosomal binding site (RBS), and rho-independent transcriptional terminator (hairpin loop), together with the corresponding sequences, are showed. The flanking genes 0492 (encoding a putative transmembrane protein) and 0490 (encoding a putative lipoprotein precursor) are indicated. (B) Structural organization and predicted domains of the FcpB cystein peptidase. Domain regions are dark grey and the position of the initial and final residues of each domain are indicated below. The amino acids corresponding to the active site are indicated in bold.
The fcpB gene is predicted to encode a 394 amino-acid protein with significant homology with bacterial cysteine endopeptidases such as streptopains and others C10 family peptidases from different bacterial species including Dyadobacter fermentans (identity 32%; E-value 4e-45), Spirosoma linguale (identity 37%; E-value 1e-44), Bacteroides intestinalis (identity 34%, E-value 2e-32), S. pyogenes (identity 28%, E-value 4e-13) and Flavobacterium branchiophilum (identity 26%, E-value 2e-15). FcpB seems to be an extracellular protein since it contains a predicted signal peptide of 27 amino acids (Fig. 1B). It also has a peptidase C10 family domain located from amino acid 191 to 336 (Fig. 1B). In addition, FcpB as other bacterial cysteine endopeptidases, seems to be synthesized as a proenzyme, having a predicted pro-peptide that blocks the active site of the enzyme corresponding to the C217 amino acids position to H327 (Fig. 1B).
Generation of a gene deletion in F. psychrophilum
The genome of F. psychrophilum THC02/90 does not have any I-SceI restriction site as revealed by in silico analysis. The procedure used to obtain a markerless fcpB deletion in F. psychrophilum THC02/90 was based on the method of Pósfai et al. (1999) . The conjugative and integrative plasmid pLYL03–3S5 harboring 1.5 kb flanking sequences of fcpB and an I-SceI restriction site (Fig. 2A.1) was transferred to F. psychrophilum THC02/90 from the strain E. coli S17–1λpir by conjugation. Transconjugants were selected on NAC plates with erythromycin and the 3% of the transconjugants (12 out of 400 transconjugants grown on NB with erythromycin in microtiter plates) harbored the plasmid integrated into the chromosome (Fig. 2A.1). This was determined by PCR (data not shown), and Southern Blot, which showed the two expected band patterns corresponding to the two possible integrations depending on the flanking region where the recombination had occurred (Fig. 2A.2). The two merodiploid strains obtained were named FcpB3+ and FcpB5+ (Fig. 2A.1). To resolve the merodiploids, the conjugative pCP23-TFS plasmid that expresses the I-SceI meganuclease under the control of the F. psychrophilum fpp2 gene promoter  was constructed. This fpp2 gene promoter is induced at 12°C in the presence of CaCl2 . In order to determine whether or not the presence of this plasmid could modify the bacterial physiology it was introduced into the F. psychrophilum THC02/90 strain. The growth curve of THC02/90-TFS strain was similar to that of the parental (data not shown). Additionally, transcription of the I-SceI gene under the Pfpp2 promoter in the induction conditions was demonstrated by RT-PCR analysis (Fig. 3).
(A1) A suicide plasmid (pLY03–3S5), containing an I-SceI recognition site and two fcpB flanking sequences of 1.5 kb (del5’ and del3’), was introduced in F. psychrophilum THC02/90 by conjugation. RK2oriT: conjugal transfer origin; ermF: erythromycin resistance gene in F. psychrophilum; bla: ampicillin resistance gene in E. coli; ori: replication origin in E. coli; lacI: lacI gene. Two possible integrations, depending on the flanking region where the recombination had occurred (FcpB5+ and FcpB3+ strains), are represented following continuous and dashed lines. HindIII restriction sites are indicated with HIII and the size of Southern Blot expected bands, using del5´ as a probe (grey line), are indicated above each scheme. (A2) Southern blot hybridization using a PCR generated labeled del5´ DNA fragment as a probe. Genomic DNA, from the parental strain F. psychrophilum THCO2/90 (lane 2) and the transconjugants which harbor the two possible generated cointegrates: lanes 3, FcpB3+ and lane 4, FcpB5+, was digested with HindIII and hybridized with del5´ fragment labeled probe. Molecular mass ladder, phage λ DNA digested with PstI (lane 1). The two top hybridization bands corresponding to lane 3 (5.27 and 3.85 kb) and lane 4 (4.91 and 4.22 kb) are tenuous because the length of their homologous sequences with the probe is very short. (B1) The FcpB5+ and FcpB3+ merodiploid strains were conjugated (only FcpB5+ is represented) with an E. coli S17–1 λ pir strain harboring a plasmid (pCP23-TFS) expressing I-SceI (I-SceI) under the control of the F. psychrophilum fpp2 calcium-temperature inducible promoter (Pfpp2). T: transcription terminator; pCP1ori: replication origin in F. psychrophilum; tetQ: tetracycline resistance gene in F. psychrophilum; bla: ampicillin resistance gene in E. coli; oriT: conjugal transfer origin; pUCori: replication origin in E. coli. Transconjugants were selected at 12°C in a medium containing tetracycline and CaCl2 leading to the resolution of the merodiploid by generation of a wild-type genotype (left side of the figure) or in a markerless fcpB gene deleted strain (right side of the figure). I-SceI mediated double strain break is indicated by an arrow head symbol. (B2) Southern Blot hybridization using genomic DNA from the wild-type and the fcpB deleted mutant strain, digested with HindIII and hybridized with an internal fcpB fragment. Lane 2, wild-type; lane 3 fcpB deleted mutant strain; Lane 1, molecular mass ladder (phage λ DNA digested with PstI).
Each PCR was performed with 20 ng of F. psychrophilum RNA. Oligonucleotides homologous to I-SceI gene, giving an internal RT-PCR fragment of 640 nt, were used. RT-PCRs were performed with RNA from F. psychrophilum THC02/90 strain carrying pCP23 plasmid (lane2) and pCP23-TFS plasmid (lane 3), which were incubated at 12°C and CaCl2 until early stationary phase. As controls for DNA contamination, PCRs were also carried out with the primers used for RT-PCR analysis by omitting the reverse transcription step, using RNA samples from F. psychrophilum THC02/90 strain carrying pCP23 plasmid (lane 4), pCP23-TFS plasmid (lane 5) and DEPC water as templates. Lane 1: 100 bp molecular marker (Biotools).
The pCP23-TFS plasmid was then introduced by conjugation into the merodiploids FcpB3+ and FcpB5+, and the resulting transconjugants were incubated on a medium containing tetracycline, to maintain the plasmid, and 12°C in the presence of CaCl2, to induce the I-SceI expression (Fig. 2B.1). I-SceI activity should mediate a unique double-strand break at its recognition site and homologous recombination repair should allow resolution of the merodiploids and vector loss (Fig. 2B.1). Transconjugants that showed a better growth in NB with tetracycline, in microtiter plates, were screened by PCR for the absence of fcpB. One out of 50 transconjugants resistant to tetracycline was PCR negative (2%). Southern Blot using a 970 bp labeled internal fragment of the fcpB gene as a probe showed the absence of a 4.9 kb hybridization band, indicating the complete deletion of fcpB (Fig. 2B.2).
Phenotypic characterization and LD50 determinations
To characterize the F. psychrophilum fcpB- strain, some of its phenotypic traits were studied. The growth of fcpB- mutant in NBF at 12°C was similar to that of the parental strain (data not shown). However, the determination of the extracellular proteolytic activity during growth indicated that fcpB gene contributed to the caseinolytic activity of the bacterium, reaching a maximum difference of about 30% less activity in the mutant fcpB- in comparison with the parental strain after 140 h of incubation (THC02/90: 121.20 ± 0.18 UE ml-1; fcpB-: 83,94 ± 0,66 UE ml-1). Analysis of colony spreading on diluted NA medium showed that fcpB- presented a colony diameter similar to that of the parental strain and both colonies had practically the same biomass production (data not shown).
In order to evaluate the effect of the fcpB gene deletion on the virulence of the bacterium, LD50 experiments were carried out on rainbow trout. The medium LD50 10 days post-injection was calculated to be 1.48 × 105 CFU, with a lower bound of 9.08 × 104 CFU and an upper bound of 2.51 × 105 CFU, for the parental strain. Under the same conditions, fcpB- showed a LD50 value of 1.70 × 105 CFU, with a lower bound of 1.01 × 104 CFU and an upper bound of 2.86 × 105 CFU (Table A in S1 Information). Furthermore, macroscopically examination of skin lesions usually produced around the injection site, revealed that they were similar in both parental and fcpB mutant strain (data not shown).
F. psychrophilum is a member of the Cytophaga-Flavobacterium-Bacteroides (CFB) group and is very distantly related to organisms with well-developed genetic systems, such as members of the proteobacteria. In general, plasmids, selectable markers, and transposons that function in proteobacteria fail to work in members of the CFB group. Techniques to genetically manipulate these bacteria have been developed [32–34]. In particular, Tn4351-based transposon mutagenesis  and a site-specific mutagenesis method by homologous recombination  have been useful in F. psychrophilum in order to study virulence determinants. All these studies have decisively contributed to increase the knowledge of the genetic and physiology of this “fastidious” microorganism. Taking into account the limitations of these systems for some studies, a deletion mutagenesis method has been developed in the present work. Two major determinants were important in this procedure: i) large sequences of about 1.5 kb flanking the fcpB gene were used to maximize the probability that the first cross-over recombination event took place, ii) the expression of the I-SceI meganuclease gene under the inducible fpp2 promoter  to facilitate the second recombination event to obtain the fcpB deletion mutant. It has been described that double-strand breaks mediated by I-SceI enhances the resolution of the merodiploid generating the two possible genotypes: mutant or wild-type . A similar method was used by Patrick et al. (2009) , for deletion of two genes involved in LPS and capsular polysaccharide biosynthesis of B. fragilis. Although the efficiency of obtaining double recombination mutants in F. psychrophilum was very low, this mutagenesis strategy was proved to be feasible in this bacterium. Both steps of this markerless deletion system had similar conjugation frequencies than that observed in other previous single-crossover recombination experiments (fpp1 and fpp2, ), but having the advantage of reducing or even avoiding the possibility to obtain polar effects. As it occurs with site-directed and transposon mutagenesis systems, conjugation and recombination frequencies are the principal bottleneck to succeed in all these procedures [12, 13]. Moreover, it should be pointed out that, in this bacterium, it is usual to find transconjugants lacking antibiotic resistance determinants. In conclusion, factors that influence conjugation and recombination efficiencies are still not completely known in F. psychrophilum and might be improved in the future. So far, this can only be solved by increasing the number of conjugation experiments.
The fcpB gene from F. psychrophilum THC02/90 strain was chosen to be deleted with the designed method because this gene is specific of this virulent strain and its product has homology with streptopain-like peptidases that are described to be virulence factors such as the streptococcal pyrogenic exotoxin (ScpB). ScpB is a critical virulence factor for invasive disease episodes caused by S. pyogenes [22–26]. Therefore, we hypothesize that the FcpB protein, which is homologous to ScpB, could have some relevance in the pathogenic process of F. psychrophilum THC02/90. Interestingly, although growth of the fcpB- strain was not affected, a significant decrease in extracellular proteolytic activity when azocasein was used as substrate was found in comparison with the parental strain. ScpB also possesses caseinolytic activity . Therefore, the reduction of this activity in fcpB- strain seems to indicate that FcpB could act as an extracellular protease. However, deletion of the fcpB gene in F. psychrophilum THC02/90 had no effect on virulence since DL50 assays did not showed significant differences between mutant and parental strain after intramuscular injection challenges in a rainbow trout model. Similar results were obtained in the studies carried out with the fpp1 and fpp2 genes encoding extracellular metalloproteases since fpp1- and fpp2- mutants showed similar virulence levels. At this point it should be noted that intramuscular injection is the only way to get reproducible values in LD50 experiments with the F. psychrophilum, but it does not reproduce a natural infection process. Additionally, it should be considered that the genome of F. psychrophilum JIP02/86 encodes for, at least, 14 other extracellular proteases . Therefore, this redundancy that, according to preliminary genome sequencing data, could be extrapolated to F. psychrophilum THC02/90 may enable some compensation, although their role on virulence should be studied in each case.
The mutagenesis method described in this work should allow relevant progress in the study of F. psychrophilum gene function. This strategy could be used for deleting any non-essential gene, creating mutants with multiple deletions, removing large DNA fragment containing cluster of genes, introducing point mutation in genes of interest or inserting foreign DNA fragment at the desired location in the F. psychrophilum genome.
S1 Information. Animal experiment protocols and mortality data (Table A: Daily dead fish injected with different doses of bacterial strains).
Esther Gómez was the recipient of grant from the Spanish Ministerio de Ciencia e Innovación (MICINN). We thank Dr. Luis Angel Fernández for the generous gift of plasmid pACBSR.
Conceived and designed the experiments: BA EG JAG. Performed the experiments: EG BA. Analyzed the data: EG BA ED JAG. Contributed reagents/materials/analysis tools: EG BA ED JAG. Wrote the paper: EG BA ED JAG.
- 1. Borg AF (1960) Studies on myxobacteria associated with diseases in salmonid fishes., vol. 8. American Association for the Advancement of Science, Wildlife Disease, Washington, DC.
- 2. Cipriano RC, Holt RA (2005). Flavobacterium psychrophilum, cause of bacterial cold-water disease and rainbow trout fry syndrome. US Department of the Interior, US Geological Survey, National Fish Health Research Laboratory.
- 3. Alvarez B, Alvarez J, Menendez A, Guijarro JA (2008) A mutant in one of two exbD loci of a TonB system in Flavobacterium psychrophilum shows attenuated virulence and confers protection against cold water disease. Microbiology 154:1144–1151. doi: 10.1099/mic.0.2007/010900-0. pmid:18375806
- 4. Nematollahi A, Decostere A, Pasmans F, Ducatelle R, Haesebrouck F (2003) Adhesion of high and low virulence Flavobacterium psychrophilum strains to isolated gill arches of rainbow trout Oncorhynchus mykiss. Dis Aquat Organ 55:101–107. doi: 10.3354/dao055101. pmid:12911057
- 5. Lorenzen E, Dalsgaard I, Bernardet J-F (1997) Characterization of isolates of Flavobacterium psychrophilum associated with coldwater disease or rainbow trout fry syndrome I: phenotypic and genomic studies. Dis Aquat Organ 31:197–208.
- 6. Moller JD, Larsen JL, Madsen L, Dalsgaard I (2003) Involvement of a sialic acid-binding lectin with hemagglutination and hydrophobicity of Flavobacterium psychrophilum. Appl Environ Microbiol 69:5275–5280. doi: 10.1128/AEM.69.9.5275-5280.2003. pmid:12957914
- 7. Lorenzen E, Olesen NJ (1997) Characterization of isolates of Flavobacterium psychrophilum associated with coldwater disease or rainbow trout fry syndrome I: serological studies. Dis Aquat Organ 31:209–220. doi: 10.3354/dao037081.
- 8. Alvarez B, Secades P, Prieto M, McBride MJ, Guijarro JA (2006) A mutation in Flavobacterium psychrophilum tlpB inhibits gliding motility and induces biofilm formation. Appl Environ Microbiol 72:4044–53. doi: 10.1128/AEM.00128-06. pmid:16751514
- 9. Duchaud E, Boussaha M, Loux V, Bernardet J-F, Michel C, et al. (2007) Complete genome sequence of the fish pathogen Flavobacterium psychrophilum. Nat Biotechnol 25:763–769. doi: 10.1038/nbt1313. pmid:17592475
- 10. Bertolini J, Wakabayashi H, Watral V, Whipple M, Rohovec J (1994) Electrophoretic detection of proteases from selected strains of Flexibacter psychrophilus and assessment of their variability. J Aquat Anim Health 6:224–233.
- 11. Ostland V, Byrne P, Hoover G, Ferguson H (2000) Necrotic myositis of rainbow trout, Oncorhynchus mykiss (Walbaum): proteolytic characteristics of a crude extracellular preparation from Flavobacterium psychrophilum. J Fish Dis 23:329–336.
- 12. Perez-Pascual D, Gomez E, Alvarez B, Mendez J, Reimundo P, et al. (2011) Comparative analysis and mutation effects of fpp2-fpp1 tandem genes encoding proteolytic extracellular enzymes of Flavobacterium psychrophilum. Microbiology 157:1196–204. doi: 10.1099/mic.0.046938-0. pmid:21292745
- 13. Alvarez B, Secades P, McBride MJ, Guijarro JA (2004) Development of genetic techniques for the psychrotrophic fish pathogen Flavobacterium psychrophilum. Appl Environ Microbiol 70:581–587. doi: 10.1128/AEM.70.1.581-587.2004. pmid:14711690
- 14. Holt RA (1987) Cytophaga psychrophila, the causative agent of bacterial cold-water disease in salmonid fish. PhD Thesis, Oregon State University, Corvallis.
- 15. Michel C, Antonio D, Hedrick RP (1999) Production of viable cultures of Flavobacterium psychrophilum: approach and control. Res Microbiol 150:351–358. doi: 10.1016/S0923-2508(99)80061-8. pmid:10422696
- 16. Gomez E, Perez-Pascual D, Fernandez L, Mendez J, Reimundo P, et al. (2012) Construction and validation of a GFP-based vector for promoter expression analysis in the fish pathogen Flavobacterium psychrophilum. Gene 497:263–8. doi: 10.1016/j.gene.2012.01.069. pmid:22327027
- 17. Posfai G, Kolisnychenko V, Bereczki Z, Blattner FR (1999) Markerless gene replacement in Escherichia coli stimulated by a double-strand break in the chromosome. Nucleics Acids Res 27:4409–4415. doi: 10.1093/nar/27.22.4409. pmid:10536150
- 18. Patrick S, Houston S, Thacker Z, Blakely GW (2009) Mutational analysis of genes implicated in LPS and capsular polysaccharide biosynthesis in the opportunistic pathogen Bacteroides fragilis. Microbiology 155:1039–1049. doi: 10.1099/mic.0.025361-0. pmid:19332806
- 19. López CM, Rholl DA, Trunck LA, Schweizer HP (2009) Versatile dual-technology system for markerless allele replacement in Burkholderia pseudomallei. Appl Environ Microbiol 75:6496–6503. doi: 10.1128/AEM.01669-09. pmid:19700544
- 20. Horzempa J, Shanks RM, Brown MJ, Russo BC, O’Dee DM, et al. (2010) Utilization of an unstable plasmid and the I-SceI endonuclease to generate routine markerless deletion mutants in Francisella tularensis. J. Microbiol Methods 80:106–108. doi: 10.1016/j.mimet.2009.10.013. pmid:19879904
- 21. Wang H, Liu X, Liu S, Yu Y, Lin J, et al. (2012) Development of a markerless gene replacement system for Acidithiobacillus ferrooxidans and construction of a pfkB mutant. Appl Environ Microbiol 78:1826–1835. doi: 10.1128/AEM.07230-11. pmid:22210219
- 22. Tsai P-J, Kuo C-F, Lin K-Y, Lin Y-S, Lei H-Y, et al. (1998) Effect of group A streptococcal cysteine protease on invasion of epithelial cells. Infect Immun 66:1460–1466. pmid:9529068
- 23. Stockbauer KE, Magoun L, Liu M, Burns EH, Gubba S Jr, et al. (1999) A natural variant of the cysteine protease virulence factor of group A Streptococcus with an arginine-glycine-aspartic acid (RGD) motif preferentially binds human integrins alphavbeta3 and alphaIIbbeta3. Proc Natl Acad Sci USA 96:242–247. doi: 10.1073/pnas.96.1.242. pmid:9874803
- 24. Ohkuni H, Todome Y, Watanabe Y, Ishikaw T, Takahashi H, et al. (2004) Studies of recombinant streptococcal pyrogenic exotoxin B/cysteine protease (rSPE B/SCP) in the skin of guinea pigs & the release of histamine from cultured mast cells & basophilic leukocytes. Indian J Med Res 119:33–36. pmid:15232159
- 25. Sumitomo T, Nakata M, Higashino M, Jin Y, Terao Y, et al. (2011) Streptolysin S contributes to group A streptococcal translocation across an epithelial barrier. J Biol Chem 286: 2750–2761. doi: 10.1074/jbc.M110.171504. pmid:21084306
- 26. Hung C-H, Tsao N, Zeng Y-F, Lu S-L, Chiang-Ni C, et al. (2012) Synergistic effects of streptolysin S and streptococcal pyrogenic exotoxin B on the mouse model of group A streptococcal infection. Med Microbiol Immunol 201: 357–369. doi: 10.1007/s00430-012-0241-6. pmid:22610375
- 27. Simon R, Priefer U, Pühler A (1983) A broad host range mobilization system for in vivo genetic engineering: transposon mutagenesis in gram negative bacteria. Nat Biotechnol 1:784–791.
- 28. Secades P, Alvarez B, Guijarro JA (2001) Purification and characterization of a psychrophilic, calcium-Induced, growth-phase-dependent metalloprotease from the fish pathogen Flavobacterium psychrophilum. Appl Environ Microbiol 67:2436–2444. pmid:11375148
- 29. Álvarez B, Guijarro JA (2007) Recovery of Flavobacterium psychrophilum viable cells using a charcoal based solid medium. Lett Appl Microbiol 44:569–572. doi: 10.1111/j.1472-765X.2007.02126.x. pmid:17451527
- 30. Perez-Pascual D, Menendez A, Fernandez L, Mendez J, Reimundo P, et al. (2010) Spreading versus biomass production by colonies of the fish pathogen Flavobacterium psychrophilum: role of the nutrient concentration. Int Microbiol 12:207–214. doi: 10.2436/20.1501.01.100 PubMed: 20112225.
- 31. Sambrook J, Russell DW (2001) Molecular cloning: a laboratory manual., 3rd edition ed. Cold Spring Harbor Laboratory Press, New York.
- 32. Li L-Y, Shoemaker NB, Salyers AA (1995) Location and characteristics of the transfer region of a Bacteroides conjugative transposon and regulation of transfer genes. J Bacteriol 177:4992–4999. pmid:7665476
- 33. McBride M, Kempf M (1996) Development of techniques for the genetic manipulation of the gliding bacterium Cytophaga johnsonae. J Bacteriol 178:583–590. pmid:8550486
- 34. McBride MJ, Baker SA (1996). Development of techniques to genetically manipulate members of the genera Cytophaga, Flavobacterium, Flexibacter, and Sporocytophaga. Appl Environ Microbiol 62: 3017–3022. pmid:8702294
- 35. Ohara-Nemoto Y, Sasaki M, Kaneko M, Nemoto T, Ota M (1994) Cysteine protease activity of streptococcal pyrogenic exotoxin B. Can J Microbiol 40:930–936. pmid:7804905
- 36. Chakroun C, Grimont F, Urdaci MC, Bernardet JF (1998) Fingerprinting of Flavobacterium psychrophilum isolates by ribotyping and plasmid profiling. Dis Aquat Organ 33:167–177. pmid:9841121
- 37. Agarwal S, Hunnicutt DW, McBride MJ (1997) Cloning and characterization of the Flavobacterium johnsoniae (Cytophaga johnsonae) gliding motility gene, gldA. Proc Natl Acad Sci USA 94: 12139–12144. pmid:9342376
- 38. Herring CD, Glasner JD, Blattner FR (2003) Gene replacement without selection: regulated suppression of amber mutations in Escherichia coli. Gene 311:153–163. doi: 10.1016/S0378-1119(03)00585-7. pmid:12853150