- Email: [email protected]
A method for allelic replacement in Francisella tularensis
FEMS Microbiology Letters 222 (2003) 273^280
www.fems-microbiology.org
A method for allelic replacement in Francisella tularensis Igor Golovliov a , Anders Sjo«stedt a
a;
, Alexander Mokrievich b , Vitaly Pavlov
b;
Department of Clinical Microbiology, Clinical Bacteriology, Umeafi University, 901 85 Umeafi, Sweden b State Research Center for applied Microbiology, Obolensk, Moscow region, Russia Received 27 February 2003; received in revised form 3 April 2003 ; accepted 7 April 2003 First published online 29 April 2003
Abstract A vector for mutagenesis of Francisella tularensis was constructed based on the pUC19 plasmid. By inserting the sacB gene of Bacillus subtilis, oriT of plasmid RP4, and a chloramphenicol resistance gene of Shigella flexneri, a vector, pPV, was obtained that allowed specific mutagenesis. A protocol was developed that allowed introduction of the vector into the live vaccine strain, LVS, of F. tularensis by conjugation. As a proof of principle, we aimed to develop a specific mutant defective in expression of a 23-kDa protein (iglC) that we previously have shown to be prominently upregulated during intracellular growth of F. tularensis. A plasmid designated pPV-viglC was developed that contained only the regions flanking the encoding gene, iglC. By a double crossover event, the chromosomal iglC gene was deleted. However, the resulting strain, denoted viglC1, still had an intact iglC gene. Southern blot analysis verified that LVS harbors two copies for the iglC gene. The mutagenesis was therefore repeated and a mutant defective in both iglC alleles, designated viglC1+2, was obtained. The viglC1+2 strain, in contrast to viglC1, was shown to display impaired intracellular macrophage growth and to be attenuated for virulence in mice. The developed genetic system has the potential to provide a tool to elucidate virulence mechanisms of F. tularensis and the specific F. tularensis mutant illustrates the critical role of the 23-kDa protein, iglC, for the virulence of F. tularensis LVS. 8 2003 Federation of European Microbiological Societies. Published by Elsevier Science B.V. All rights reserved. Keywords : Francisella tularensis ; Allelic replacement; 23-kDa protein; Intracellular survival
1. Introduction Tularemia is a zoonotic disease caused by the highly virulent, facultative intracellular bacterium Francisella tularensis. The disease has been described in many mammalian species and the most important vectors are hares, rabbits, and rodents [1]. In humans, the clinical presentations and outcome of infection vary considerably depending on the etiological agent. The clinically important subspecies tularensis (alternative designation F. tularensis type A) and holarctica (type B) are both highly infectious but show pronounced di¡erences in virulence [2]. If not properly treated, infections caused by isolates of subspecies tularensis may cause considerable mortality whereas strains of subspecies holarctica almost never cause any mortality. The live vaccine strain, LVS, of F. tularensis
* Corresponding author. Tel. : +46 (90) 7851120; Fax : +46 (90) 7852225. E-mail addresses : [email protected] (A. Sjo«stedt), [email protected] (V. Pavlov).
has been a widely used strain for studies on F. tularensis in experimental models. Originally it was derived from a Russian strain of subspecies holarctica [3]. It is a human live vaccine strain that provides good protection against both naturally and laboratory-acquired tularemia [4]. In mice, the strain is as virulent as wild-type holarctica isolates by most routes of inoculation and therefore considered to be a relevant model organism for studies of the pathogenicity of F. tularensis in vivo and in vitro [5]. A hallmark of tularemia is the ability of the bacterium to replicate intracellularly into high numbers before the onset of a protective cell-mediated immune response [6]. F. tularensis survives within many types of host cells but there is a relative paucity of information about the factors essential for its intracellular growth [2]. The work has been hampered by the lack of suitable genetic tools for the manipulation of subspecies holarctica and tularensis. Exogenous DNA can be delivered to F. tularensis by electroporation [7,8], cryotransformation [9], or conjugation [10] but no studies have systematically analyzed whether these methods can be used to develop methods for allelic replacement. The most widely used cloning vectors do not
0378-1097 / 03 / $22.00 8 2003 Federation of European Microbiological Societies. Published by Elsevier Science B.V. All rights reserved. doi:10.1016/S0378-1097(03)00313-6
FEMSLE 10987 19-5-03
274
I. Golovliov et al. / FEMS Microbiology Letters 222 (2003) 273^280
replicate in F. tularensis. Therefore, they may be used to introduce mutations and allelic replacement via homologous recombination. Creation of mutants by allelic exchange using chemical transformation or electroporation of transposon-containing suicide vectors has been shown to work in subspecies novicida but has not been demonstrated to be successful in other subspecies [11^13]. Since there are distinct genetic di¡erences between the subspecies, it cannot be assumed that loci shown to be essential for the intracellular survival of novicida have the same functions in the more virulent subspecies so there is a de¢nite need to develop useful systems for the creation of mutants of subspecies holarctica and tularensis. Preliminary studies have indicated that the frequency of recombination in F. tularensis LVS is low and that only single homologous recombination occurred (unpublished). To this end, we aimed to develop a strategy that allowed allelic exchange via double crossover events. Such strategies are often based on introduction of genes that confer sensitivity of bacteria to certain extracellular compound, e.g. sucrose, streptomycin, or fusaric acid. The sacB gene encodes the secreted enzyme levansucrase, the expression of which is toxic for Gram-negative bacteria in the presence of sucrose [14,15]. Here we describe the construction of a vector and procedures designed for gene replacement in F. tularensis based on the sacB gene of Bacillus subtilis. We aimed to create a mutant defective in expression of a 23-kDa protein of F. tularensis LVS since we in a previous publication showed that this protein was one of only a few proteins that were signi¢cantly upregulated during intracellular infection or after exposure of F. tularensis to the bactericidal agent hydrogen peroxide [16]. Nothing is known about the function of this protein but in a recent publication it was identi¢ed as a component necessary for the ability of F. tularensis subspecies novicida to replicate
intracellularly [11]. The operon, igl (intracellular growth locus), containing the gene encoding the 23-kDa protein, henceforth designated iglC, the third gene of the operon, comprises four genes. We here demonstrate that the iglCcontaining operon exists in two copies in F. tularensis LVS and that the strategy developed allowed us to generate a mutant with deletions in both iglC alleles. Moreover, we observed that the mutant showed impaired multiplication in a macrophage cell line and was attenuated in mice.
2. Materials and methods 2.1. Bacterial strains and plasmids Bacterial strains and plasmids used in this study are described in Table 1. 2.2. Cell cultures and macrophage infection One day before the start of an experiment, J774A.1 cells were detached, resuspended in culture medium, Dulbecco’s modi¢ed Eagle medium (DMEM, Gibco BRL, Grand Island, NY, USA) with 10% (v/v) fetal calf serum and added to wells of a 24-well tissue culture plate. Peritoneal exudate cells (PEC) were obtained from Balb/cJ mice 3 days after intraperitoneal injection of 2 ml of 3% thioglycolate. They were washed with DMEM and suspended in the same medium as described for J774A.1 cells. After incubation overnight at 37‡C, wells were washed and reconstituted with fresh culture medium. To each well, a suspension of bacteria was added and bacterial uptake was allowed to occur for a 2-h period at 37‡C. After bacterial uptake, the monolayer was washed twice and incubated for indicated periods of time in culture medium with 5 Wg ml31 of gentamicin. For determination of numbers of intracellular bacteria, J774 cells were washed once in phos-
Table 1 Description of plasmids and strains used in the study Bacteria or plasmid
Characteristics
E. coli DH5K E. coli S17-1 F. tularensis LVS
F (B80dlacZvM15) recA1 endA1 gyrA96 thi-1 v(lacZYA-argF) U169 thi thr leu tonA lacY supE recA : :RP4-2-Tc: :Mu, Kn: :Tn7 Pmr , vaccine strai
E. coli S17-1/pHV33mob pUC19 pBlue-ScriptKS+
Ampr Ampr
PSa pDM4 pHV33 pHV33mob pHV33-sacB pPV pPVviglC
Cmr Ampr , Cmr , sacB, mob Ampr , Cmr , Tcr Ampr , Cmr , mob sacB, Cmr , Tcr Ampr , Cmr , sacB, mob Ampr , Cmr , sacB, mob
3
Reference and source hsdR17(r3 k
FEMSLE 10987 19-5-03
mþ k)
supE44 relA1 deoR
[29] [30] USAMRIID, Frederick, MD This study [31] Stratagene, La Jolla, CA [19] [20] [32] This study This study This study This study
I. Golovliov et al. / FEMS Microbiology Letters 222 (2003) 273^280
phate-bu¡ered saline (PBS) and lysed with 0.2 ml of 0.1% sodium deoxycholate. After addition of 1.8 ml of PBS, 100-Wl portions of each lysate, serially diluted in PBS, were plated on modi¢ed Thayer^Martin agar plates for determination of viable counts. 2.3. Detection of F. tularensis-mediated cytopathogenicity Cytopathogenicity was assessed by measuring the release of cytosolic lactate dehydrogenase (LDH) into the supernatant, which re£ects loss of plasma membrane integrity in infected cells. LDH levels were measured using the colorimetric Cytotox 96 Kit (Promega, Madison, WI, USA) according to the manufacturer’s instructions. The percentage of cytopathogenicity was calculated as 100U [(experimental release-spontaneous release)/(total releaseWspontaneous release)], where spontaneous release is the amount of LDH activity in the supernatant of uninfected cells and total release is the activity in cell lysates. 2.4. DNA manipulations All DNA manipulations, including puri¢cation, digestion with restriction endonucleases, ligations, and gel electrophoresis, were performed according to methods described by Sambrook et al. [17]. Puri¢cation of DNA fragments was done with QIAquick Gel Extraction Kit (Qiagen, Hilden, Germany) and chromosomal DNA with DNeasy Tissue Kit (Qiagen). Southern blot analysis was carried out on nylon membranes (Hybond-Nþ , Amersham Pharmacia, Uppsala, Sweden) as instructed by the manufacturer. Single-stranded DNA probe was end-labeled with [Q-32 P]dATP (Amersham Pharmacia) by use of T4 polynucleotide kinase (Promega). Hybridization was run in ExpressHyb hybridization solution (Clontech, Palo Alto, CA, USA) in accordance with the manufacturer’s protocol. 2.5. Gel electrophoresis and Western blotting Samples were applied on a uniform 12% acrylamide gel for electrophoresis as described by Laemmli [18]. For Western blot analysis, proteins in acrylamide gel were transferred to nitrocellulose ¢lters and probed with rabbit polyclonal anti-iglC sera. A horseradish peroxidase-conjugated secondary antibody system was used and ¢lters were developed by using the ECL Kit (Amersham Pharmacia). 2.6. Construction of vectors A suicide vector designed to be used for allelic replacement in F. tularensis contained the following elements: (i) an antibiotic marker, the chloramphenicol resistance gene (crg) from the Shigella £exneri plasmid pSa [19], which confers chloramphenicol resistance in F. tularensis [10] ; (ii) an origin of transfer (mob) and (iii) the SacB gene of
275
B. subtilis for excision of the vector. A PstI fragment from the B. subtilis plasmid pDM4 [20] containing the SacB gene including the promoter was cloned to pKK214 [21] or pHV33 [22]. Both are low-copy number plasmids able to replicate in F. tularensis. Spontaneous sucrose resistance occurred in F. tularensis LVS at a frequency of 1036 (data not shown), similar to what has been reported previously for Escherichia coli and Yersinia enterocolitica [23,24]. A BamHI fragment form pDM4 containing the origin of transfer, oriT, of plasmid RP4 [25] was cloned to BamHI of pHV33 and the resulting plasmid designated pHV33mob. The presence of oriT on plasmid usually results in e⁄cient conjugal transfer to recipient strains from donor strain expressing RP4 conjugative functions [26]. PHV33mob was e⁄ciently transferred from E. coli S17-1, containing the RP4 conjugative transfer apparatus, to F. tularensis LVS by mating and was used to optimize conditions of conjugation. A vector for site-directed mutagenesis was constructed using the pUC19 plasmid which is unable to replicate in F. tularensis. First, the BamHI fragment (1.6 kb) from pDM4 containing oriT was ligated to BamHI-digested pUC19 resulting in a pUC-mob plasmid. The SacB gene was excised from pDM4 as a 2.6-kb PstI fragment and ligated to PstI-digested pUC-mob. The resulting plasmid was partially digested with HindIII and a HindIII fragment (3.2 kb) from pSa containing the crg gene was inserted. Resistant transformants were obtained by selection on plates containing 10 Wg ml31 of chloramphenicol and one of clones containing the crg gene inserted in the HindIII site of the polylinker cloning site of pUC19 was selected. It was named pPV, and it has unique SalI and XbaI sites for cloning (Fig. 1A). 2.7. Construction of an iglC mutagenesis plasmid A region approximately 1500 bp upstream (primers C and E) and a region downstream (primers D and F) of the iglC gene were ampli¢ed by polymerase chain reaction (PCR). The 5P-primers contained SalI restriction sites (upstream region, primer C: 5P-ACG CGT CGA CTA CAA TGC TAG AAA CCT T-3P, and downstream region, primer D: 5P-GCG TGT CGA CTT AGC CGT GCC AAT TAC CA-3P) and the 3P-primers a BamHI site (upstream region, primer E: 5P-TCG GGA TCC TAT ACA TGC AGT AGG A-3P) or a PstI site (downstream region, primer F: 5P-AAA CTG CAG CTG CAT AGT AAG ATC GGA-3P). Each DNA fragment had a size of approximately 1.5 kb and included the ¢rst 80 nucleotides or the last 15 nucleotides, respectively, of the iglC gene. They were ligated to SalI/BamHI or SalI/PstI-digested plasmid pBlue-ScriptKS+ (Stratagene, La Jolla, CA, USA). From the recombinant plasmids, the cloned DNA fragments were excised with SalI and BamHI and both fragments ligated simultaneously to SalI-digested pPV. The resulting vector was designated pPV-viglC.
FEMSLE 10987 19-5-03
276
I. Golovliov et al. / FEMS Microbiology Letters 222 (2003) 273^280
Fig. 1. A: Description of the recombinant genes of pPV inserted in the multicloning site of pUC19. oriT and SacB genes originated from pDM4. The chloramphenicol resistance gene, crg, originated from pSA. B: Description of the igl operon of F. tularensis LVS. The operon contains four open reading frames, designated A^D. Localization of primers used for ampli¢cation of the various regions of the operon is indicated.
2.8. Conjugal transfer of plasmids
plating 10-fold serial dilutions. Colonies were counted after 3 days of incubation at 37‡C.
Early exponential cultures of 107 CFU ml31 of E. coli S17-1 and 109 CFU ml31 of F. tularensis LVS were concentrated by centrifugation and resuspended in 50 Wl of culture medium, mixed, and plated on either Luria agar (LA) or McLeod [16] plates. For optimization of conjugation, plates were incubated at various temperatures and after di¡erent time intervals. After incubation, cells were resuspended in PBS and plated on modi¢ed Gc agar base plates [BBL Gc agar base (Becton Dickinson, Sunnyvale, CA, USA) with the addition of 1% (w/v) of hemoglobin (Oxoid, Stockholm, Sweden) and 1% (v/v) of BBL IsoVitalex (Becton Dickinson)] containing 100 Wg ml31 of polymyxin B for counterselection of the donor E. coli strain [10] and 2.5 Wg ml31 of chloramphenicol. Frequencies of mating were expressed as the ratio of recipient cells that harbored the plasmid antibiotic markers versus the total number of recipients.
3. Results 3.1. Optimization of mating between E. coli and F. tularensis LVS F. tularensis LVS was conjugated with E. coli S17 carrying the pHV33-mob plasmid at various temperatures as described in Section 2. The results are presented in Table 2. Optimal conditions were obtained when the mixture of bacteria were incubated on LA plates at 25‡C for 18 h. Total numbers of transformants on LA were approximately 2U107 /reaction and the estimated frequency of transfer was approximately 2U1032 . These conditions were used in the work described henceforth. If McLeod plates were used instead of LA, the number of transformants was at least 2 log10 lower (data not shown).
2.9. Inoculation and enumeration of bacteria 3.2. Creation of a viglC protein mutant An area of approximately 2 cm2 of the skin of upper thorax was shaved 2 days before inoculation. Mice were challenged with an intradermal inoculation of 5U107 of F. tularensis LVS and killed by decapitation after 4 weeks of infection. Spleen and liver samples were homogenized and the number of F. tularensis LVS was calculated by
F. tularensis LVS was conjugated with E. coli S17-1 carrying the pPV-viglC plasmid and after incubation on LA plates overnight, plated on McLeod plates containing 100 Wg ml31 of polymyxin B and 2.5 Wg ml31 of chloramphenicol. Transformants were observed after 5^6 days of
Table 2 In£uence of temperature on conjugation frequency (‡C)
15 25 30 37
Bacterial number before mating
Bacterial number after mating
F. tularensis
E. coli
F. tularensis
E. coli
1U109
1U107
1U109 2U109 2U109 1.5U109
4U107 3U109 1U109 1U109
FEMSLE 10987 19-5-03
Number of conjugants
Frequency of conjugation
2U105 2U107 3U106 5U105
2U1034 1U1032 1.5U1033 3.3U1034
I. Golovliov et al. / FEMS Microbiology Letters 222 (2003) 273^280
Fig. 2. Southern blot hybridization of DNA from mutant and wild-type strains. Samples of DNA were digested with ClaI prior to separation by agarose gel electrophoresis and Southern blotting and hybridization. The blot was probed with a labelled oligonucleotide complementary to a part of iglC lacking in viglC. The sizes were estimated from the mobility of a 1-kb DNA standard. Lane 1: F. tularensis viglC1+2 DNA; lane 2: F. tularensis viglC1 DNA; lane 3: F. tularensis LVS DNA.
incubation at 37‡C in 5% CO2 . In total, 20 transformants were obtained and none in control reactions where S17-1 was mixed with F. tularensis LVS. No replicating plasmids were recovered from any of the transformants. Since pPVviglC cannot replicate in F. tularensis, either illegitimate recombination should have occurred or the transformants should contain the plasmid inserted into the chromosome after recombination of the wild-type, chromosomal iglC and the truncated variant of pPV-viglC. To analyze the recombination event, we used primers complementary to the upstream (A) or downstream (B) region of viglC of the pPV vector and primers C and D, described in Section 2, located upstream and downstream, respectively, of iglC (Fig. 1B). PCR was then performed using primer pairs A+D or B+C. Using the two primer combinations, it was possible to amplify DNA from all transformants, indicating that legitimate recombination had occurred in all cases. Recombination had occurred with an equal frequency in the upstream and downstream £anking regions. Twenty recombinant clones were isolated and incubated on McLeod plates with 10 Wg ml31 of chloramphenicol. To obtain a clone lacking the wild-type gene, we selected for a second recombination event by plating on medium containing 5% sucrose. Of the 15 analyzed sucrose-resistant colonies, all were sensitive to chloramphenicol, indicating that plasmid exclusion had occurred. This was con¢rmed since no ampli¢cation resulted from PCR with primers speci¢c to the plasmid. Of the chloramphenicol-sensitive clones, 70% had a wild-type phenotype of the
277
iglC gene, presumably as a result of abortive allelic exchange through the same sequence as during the ¢rst crossover event. All other colonies had a mixed phenotype retaining both wild-type and the truncated iglC gene. This indicated the presence of two copies of the iglC gene in the LVS genome. One of the clones, designated viglC1, was selected for the second round of conjugation and all procedures were repeated as described above. After mating, 40 chloramphenicol-resistant clones were obtained. A PCR using primers A+B was performed. Only if the ¢rst crossover of the wild-type gene had occurred, these primers should amplify the truncated form of the 23-kDa protein gene. Of the 40 clones tested, only one showed the absence of the wild-type gene. In all other clones, recombination between the truncated chromosomal and plasmid genes had occurred (data not shown). After selection on plates with 5% sucrose, more than 90% of the clones tested were chloramphenicol sensitive and harbored the truncated form of the iglC gene only. Those clones were considered as real iglC mutants, henceforth designated viglC1+2. To further analyze these mutants, a Southern blot with ClaI-digested chromosomal DNA from F. tularensis LVS was performed using a 32 P-labeled oligonucleotide complementary to a region of the deleted region as a probe. Preliminary results showed that cleavage with ClaI allowed discrimination between the two copies of iglC (data not shown). As expected, the probe hybridized to two bands of wild-type LVS DNA (approximately 8 and 15 kb) but only one in viglC1 and no hybridization was observed with viglC1+2 mutant (Fig. 2). A PCR analysis
Fig. 3. Expression of iglC. Pellets after centrifugation of bacterial cultures were lysed in Laemmli bu¡er, separated by electrophoresis in a 12% polyacrylamide gel and transferred to a nitrocellulose ¢lter. The ¢lter was probed with a polyclonal rabbit antiserum speci¢c to the iglC protein at a dilution of 1:200. Lane 1: F. tularensis LVS; lane 2: F. tularensis viglC1; lane 3: F. tularensis viglC1+2.
FEMSLE 10987 19-5-03
278
I. Golovliov et al. / FEMS Microbiology Letters 222 (2003) 273^280
Fig. 4. Growth of F. tularensis in J774.A1 (A) or PEC (B) cells. Data from a representative experiment are shown and the results are expressed as the CFU (mean log10 U S.D.)/well of F. tularensis, based on triplicate wells.
(primers A+B) veri¢ed the presence of a mutated copy of the iglC gene and a wild-type gene in viglC1 and no wildtype copy in viglC1+2 mutant (data not shown). 3.3. Protein expression in viglC To verify that the 23-kDa protein was absent in viglC1+2, bacteria were grown on a McLeod plate and subjected to Western blot analysis. After sodium dodecyl sulfate^polyacrylamide gel electrophoresis and transfer to nitrocellulose ¢lter, the extracts were probed with a rabbit polyclonal serum against iglC. No immunoreactive band was present in the mutant strain (Fig. 3). 3.4. Growth of F. tularensis mutant strains in J774 cells and peritoneal macrophages The 23-kDa mutant strains and wild-type F. tularensis LVS grew equally well in Chamberlain medium (data not
shown). There was no signi¢cant di¡erence in the uptake in J774A.1 cells of the two bacterial strains. However, in contrast to F. tularensis LVS or viglC1, viglC1+2 showed no growth in J774 cells and, in fact, bacterial numbers decreased over time. The former two strains increased 2.5 log10 during the ¢rst 24 h whereas the numbers of viglC1+2 increased only 0.5 log10 (Fig. 4). F. tularensis LVS and viglC1 gave a pronounced cytopathogenic e¡ect and after 24 h, release of LDH was more than 65% of the maximum release, whereas viglC1+2 exerted no cytopathogenic e¡ect (Fig. 5A). Similar results were obtained in peritoneal macrophages. The wild-type and viglC1 grew approximately 1 log10 in the cells whereas viglC1+2 showed no growth (Fig. 4B). Again, the former two strains exerted a strong cytopathogenic e¡ect but no such e¡ect was seen in cells infected with the mutant strain (Fig. 5B). 3.5. Growth of F. tularensis viglC1+2 in mice Mice were challenged with 5U107 CFU of F. tularensis LVS or F. tularensis viglC1+2. This dose represents approximately 100 LD50 of F. tularensis LVS. Accordingly, all of the 10 mice died within 6 days after challenge with F. tularensis LVS whereas all of the 10 mice challenged with the mutant strains survived for 4 weeks and when killed, had eradicated the bacteria.
4. Discussion
Fig. 5. Cytopathogenicity of mutant and wild-type F. tularensis strains in J774.A1 (A) or PEC (B) cells after 24 h of infection. Cytopathogenicity was assayed by release of LDH from infected cells as described in Section 2.
Little is known regarding virulence mechanisms of F. tularensis. A major drawback has been a lack of methods to generate mutants of isolates of the clinically important subspecies, tularensis and holarctica. This is the ¢rst demonstration of a genetic system that allowed the gener-
FEMSLE 10987 19-5-03
I. Golovliov et al. / FEMS Microbiology Letters 222 (2003) 273^280
ation of de¢ned mutants in F. tularensis. The constructed vector allows the creation of mutants with point mutations, deletions, or insertions. Previously published methods on Francisella have been validated using F. tularensis subspecies novicida, a subspecies more amenable to genetic manipulation than the other subspecies of F. tularensis, and all have been based on random insertions of minitransposons carrying antibiotic markers [7]. A disadvantage of the latter strategy is that it will give deletion mutants only and may also have polar e¡ects. An advantage of the present strategy is the possibility to use conjugation for plasmid delivery. It appears as if conjugation is an e⁄cient method whereas no mutants were obtained when pPV-viglC was delivered by electroporation (data not shown). Another important feature of the system is that the recombinants do not express any antibiotic resistance markers and, in fact, no foreign DNA remains after selection on sucrose. This feature may prove advantageous if the method is going to be used for the generation of live attenuated vaccine strains. Previously, we identi¢ed the 23-kDa protein of F. tularensis as markedly upregulated during intracellular macrophage infection [16]. The previous study demonstrated that remarkably few proteins were induced during the intracellular growth. This is in contrast to ¢ndings on other facultative intracellular bacteria such as S. enterica serovar typhimurium which responded to the intracellular environment by up- or downregulating a large number of proteins [27]. Previously, we were not able to assign any speci¢c role to the 23-kDa protein since it showed no homology to other bacterial proteins. The creation of a speci¢c mutant will therefore be of tremendous help in understanding the role of the protein in the virulence of F. tularensis. In fact, when screening the mutant for a phenotype, we discovered that the F. tularensis iglC1+2 mutant lacked the property of the wild-type F. tularensis LVS to inhibit Toll-likemediated signaling via MAPK and NF-UB pathways and secretion of TNF-K and IL-1L of murine macrophages [28]. Although we have not identi¢ed the underlying molecular mechanism, the development of the genetic system will allow means to study this. Our demonstration of impaired intracellular growth and marked attenuation in vivo of the mutant strain provides direct evidence for an important role of the 23-kDa protein in the virulence of F. tularensis. Our ¢ndings also corroborate those of Nano et al. [11], who recently demonstrated that a transposon insertion in F. tularensis subspecies novicida rendered it incapable of intracellular proliferation. In this study, insertions in both iglA and iglC gave an attenuated phenotype. However, it could not be excluded that the attenuation was the result of polar effects on downstream genes of the operon. We have also been successful in developing other speci¢c mutants in F. tularensis LVS and also in virulent strains of subspecies holarctica (unpublished data). The development of genetic tools together with the rap-
279
idly increasing genomic information on F. tularensis will result in signi¢cant progress in our understanding of the hitherto ill-de¢ned virulence mechanisms of this potent intracellular pathogen.
Acknowledgements Grant support was obtained from the the Swedish Medical Research Council, Samverkansna«mnden Norra Sjukvafirdsregionen, Umeafi, and the Medical Faculty, Umeafi University, Umeafi, Sweden.
References [1] Sjo«stedt, A. (2003) Family XVII. Francisellaceae, Genus I. Francisella. In: Bergey’s manual of systematic bacteriology (Brenner, D.J., Ed.). Springer, New York. [2] Sjo«stedt, A. (2003) Virulence determinants and protective antigens of Francisella tularensis. Curr. Opin. Microbiol. 6, 1^6. [3] Eigelsbach, H.T. and Down, C.M. (1961) Prophylactic e¡ectiveness of live and killed tularemia vaccines. I. Production of vaccine and evaluation in the white mouse and guinea pig. J. Immunol. 87, 415^ 425. [4] Burke, D.S. (1977) Immunization against tularemia: analysis of the e¡ectiveness of live Francisella tularensis vaccine in prevention of laboratory-acquired tularemia. J. Infect. Dis. 135, 55^60. [5] Fortier, A.H., Slayter, M.V., Ziemba, R., Meltzer, M.S. and Nacy, C.A. (1991) Live vaccine strain of Francisella tularensis: infection and immunity in mice. Infect. Immun. 59, 2922^2928. [6] Ta«rnvik, A. (1989) Nature of protective immunity to Francisella tularensis. Rev. Infect. Dis. 11, 440^451. [7] Anthony, L.S., Gu, M.Z., Cowley, S.C., Leung, W.W. and Nano, F.E. (1991) Transformation and allelic replacement in Francisella spp. J. Gen. Microbiol. 137, 2697^2703. [8] Baron, G.S., Myltseva, S.V. and Nano, F.E. (1995) Electroporation of Francisella tularensis. Methods Mol. Biol. 47, 149^154. [9] Pavlov, V.M., Mokreivich, A.N. and Volkovoy, K. (1996) Cryptic plasmid pFNL10 from Francisella novicida-like F6168 : The base of plasmid vectors for Francisella tularensis. FEMS Immunol. Med. Microbiol. 13, 253^256. [10] Pomerantsev, A.P., Domaradskii, I.V. and Shishkova, N.A. (1991) Hybrid plasmid pSKEFT5 assures the transfer of plasmids Sa from Escherichia coli cells into cells of the tularemia microbe vaccine strain cells. Mol. Gen. Mikrobiol. Virusol. 8, 13^16. [11] Gray, C.G., Cowley, S.C., Cheung, K.K. and Nano, F.E. (2002) The identi¢cation of ¢ve genetic loci of Francisella novicida associated with intracellular growth. FEMS Microbiol. Lett. 215, 53^56. [12] Cowley, S.C., Gray, C.J. and Nano, F.E. (2000) Isolation and characterization of Francisella novicida mutants defective in lipopolysaccharide biosynthesis. FEMS Microbiol. Lett. 182, 63^67. [13] Baron, G.S. and Nano, F.E. (1999) An erythromycin resistance cassette and mini-transposon for constructing transcriptional fusions to cat. Gene 229, 59^65. [14] Gay, P., Le Coq, D., Steinmetz, M., Berkelman, T. and Kado, C.I. (1985) Positive selection procedure for entrapment of insertion sequence elements in gram-negative bacteria. J. Bacteriol. 164, 918^ 921. [15] Reyrat, J.M., Pelicic, V., Gicquel, B. and Rappuoli, R. (1998) Counterselectable markers: untapped tools for bacterial genetics and pathogenesis. Infect. Immun. 66, 4011^4017. [16] Golovliov, I., Ericsson, M., Sandstro«m, G., Ta«rnvik, A. and Sjo«stedt, A. (1997) Identi¢cation of proteins of Francisella tularensis induced
FEMSLE 10987 19-5-03
280
[17]
[18] [19]
[20]
[21]
[22] [23]
[24]
I. Golovliov et al. / FEMS Microbiology Letters 222 (2003) 273^280 during growth in macrophages and cloning of the gene encoding a prominently induced 23-kilodalton protein. Infect. Immun. 65, 2183^ 2189. Sambrook, J., Fritch, E.F. and Maniatis, T. (1989) Molecular Cloning: A Laboratory Manual. Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NY. Laemmli, U.K. (1970) Cleavage of structural proteins during the assembly of the head of bacteriophage T4. Nature 227, 680^685. Watanabe, T., Furuse, C. and Sakaizumi, S. (1968) Transduction of various R factors by phage P1 in Escherichia coli and by phage P22 in Salmonella typhimurium. J. Bacteriol. 96, 1791^1795. Milton, D.L., O’Toole, R., Horstedt, P. and Wolf-Watz, H. (1996) Flagellin A is essential for the virulence of Vibrio anguillarum. J. Bacteriol. 178, 1310^1319. Kuoppa, K., Forsberg, A. and Norqvist, A. (2001) Construction of a reporter plasmid for screening in vivo promoter activity in Francisella tularensis. FEMS Microbiol. Lett. 205, 77^81. Ehrlich, S.D. (1978) DNA cloning in Bacillus subtilis. Proc. Natl. Acad. Sci. USA 75, 1433^1436. Kaniga, K., Delor, I. and Cornelis, G.R. (1991) A wide-host-range suicide vector for improving reverse genetics in gram-negative bacteria: inactivation of the blaA gene of Yersinia enterocolitica. Gene 109, 137^141. Blom¢eld, I.C., Vaughn, V., Rest, R.F. and Eisenstein, B.I. (1991) Allelic exchange in Escherichia coli using the Bacillus subtilis sacB gene and a temperature-sensitive pSC101 replicon. Mol. Microbiol. 5, 1447^1457.
[25] Priefer, U.B., Simon, R. and Puhler, A. (1985) Extension of the host range of Escherichia coli vectors by incorporation of RSF1010 replication and mobilization functions. J. Bacteriol. 163, 324^330. [26] Miller, V.L. and Mekalanos, J.J. (1988) A novel suicide vector and its use in construction of insertion mutations : osmoregulation of outer membrane proteins and virulence determinants in Vibrio cholerae requires toxR. J. Bacteriol. 170, 2575^2583. [27] Abshire, K.Z. and Neidhardt, F.C. (1993) Analysis of proteins synthesized by Salmonella typhimurium during growth within a host macrophage. J. Bacteriol. 175, 3734^3743. [28] Telepnev, M., Golovliov, I., Grundstrom, T., Tarnvik, A. and Sjostedt, A. (2003) Francisella tularensis inhibits Toll-like receptor-mediated activation of intracellular signalling and secretion of TNF-alpha and IL-1 from murine macrophages. Cell. Microbiol. 5, 41^51. [29] Woodcock, D.M., Crowther, P.J., Doherty, J., Je¡erson, S., DeCruz, E., Noyer-Weidner, M., Smith, S.S., Michael, M.Z. and Graham, M.W. (1989) Quantitative evaluation of Escherichia coli host strains for tolerance to cytosine methylation in plasmid and phage recombinants. Nucleic Acids Res. 17, 3469^3478. [30] Simon, R., Priefer, U. and Puhler, A. (1983) A broad host range mobilization system for in vivo genetic engineering : transposon mutagenesis in Gram-negative bacteria. Biotechnology 1, 784^791. [31] Yanisch-Perron, C., Vieira, J. and Messing, J. (1985) Improved M13 phage cloning vectors and host strains : nucleotide sequences of the M13mp18 and pUC19 vectors. Gene 33, 103^119. [32] Ehrlich, S.D. (1978) DNA cloning in Bacillus subtilis. Proc. Natl. Acad. Sci. USA 75, 1433^1436.
FEMSLE 10987 19-5-03
www.fems-microbiology.org
A method for allelic replacement in Francisella tularensis Igor Golovliov a , Anders Sjo«stedt a
a;
, Alexander Mokrievich b , Vitaly Pavlov
b;
Department of Clinical Microbiology, Clinical Bacteriology, Umeafi University, 901 85 Umeafi, Sweden b State Research Center for applied Microbiology, Obolensk, Moscow region, Russia Received 27 February 2003; received in revised form 3 April 2003 ; accepted 7 April 2003 First published online 29 April 2003
Abstract A vector for mutagenesis of Francisella tularensis was constructed based on the pUC19 plasmid. By inserting the sacB gene of Bacillus subtilis, oriT of plasmid RP4, and a chloramphenicol resistance gene of Shigella flexneri, a vector, pPV, was obtained that allowed specific mutagenesis. A protocol was developed that allowed introduction of the vector into the live vaccine strain, LVS, of F. tularensis by conjugation. As a proof of principle, we aimed to develop a specific mutant defective in expression of a 23-kDa protein (iglC) that we previously have shown to be prominently upregulated during intracellular growth of F. tularensis. A plasmid designated pPV-viglC was developed that contained only the regions flanking the encoding gene, iglC. By a double crossover event, the chromosomal iglC gene was deleted. However, the resulting strain, denoted viglC1, still had an intact iglC gene. Southern blot analysis verified that LVS harbors two copies for the iglC gene. The mutagenesis was therefore repeated and a mutant defective in both iglC alleles, designated viglC1+2, was obtained. The viglC1+2 strain, in contrast to viglC1, was shown to display impaired intracellular macrophage growth and to be attenuated for virulence in mice. The developed genetic system has the potential to provide a tool to elucidate virulence mechanisms of F. tularensis and the specific F. tularensis mutant illustrates the critical role of the 23-kDa protein, iglC, for the virulence of F. tularensis LVS. 8 2003 Federation of European Microbiological Societies. Published by Elsevier Science B.V. All rights reserved. Keywords : Francisella tularensis ; Allelic replacement; 23-kDa protein; Intracellular survival
1. Introduction Tularemia is a zoonotic disease caused by the highly virulent, facultative intracellular bacterium Francisella tularensis. The disease has been described in many mammalian species and the most important vectors are hares, rabbits, and rodents [1]. In humans, the clinical presentations and outcome of infection vary considerably depending on the etiological agent. The clinically important subspecies tularensis (alternative designation F. tularensis type A) and holarctica (type B) are both highly infectious but show pronounced di¡erences in virulence [2]. If not properly treated, infections caused by isolates of subspecies tularensis may cause considerable mortality whereas strains of subspecies holarctica almost never cause any mortality. The live vaccine strain, LVS, of F. tularensis
* Corresponding author. Tel. : +46 (90) 7851120; Fax : +46 (90) 7852225. E-mail addresses : [email protected] (A. Sjo«stedt), [email protected] (V. Pavlov).
has been a widely used strain for studies on F. tularensis in experimental models. Originally it was derived from a Russian strain of subspecies holarctica [3]. It is a human live vaccine strain that provides good protection against both naturally and laboratory-acquired tularemia [4]. In mice, the strain is as virulent as wild-type holarctica isolates by most routes of inoculation and therefore considered to be a relevant model organism for studies of the pathogenicity of F. tularensis in vivo and in vitro [5]. A hallmark of tularemia is the ability of the bacterium to replicate intracellularly into high numbers before the onset of a protective cell-mediated immune response [6]. F. tularensis survives within many types of host cells but there is a relative paucity of information about the factors essential for its intracellular growth [2]. The work has been hampered by the lack of suitable genetic tools for the manipulation of subspecies holarctica and tularensis. Exogenous DNA can be delivered to F. tularensis by electroporation [7,8], cryotransformation [9], or conjugation [10] but no studies have systematically analyzed whether these methods can be used to develop methods for allelic replacement. The most widely used cloning vectors do not
0378-1097 / 03 / $22.00 8 2003 Federation of European Microbiological Societies. Published by Elsevier Science B.V. All rights reserved. doi:10.1016/S0378-1097(03)00313-6
FEMSLE 10987 19-5-03
274
I. Golovliov et al. / FEMS Microbiology Letters 222 (2003) 273^280
replicate in F. tularensis. Therefore, they may be used to introduce mutations and allelic replacement via homologous recombination. Creation of mutants by allelic exchange using chemical transformation or electroporation of transposon-containing suicide vectors has been shown to work in subspecies novicida but has not been demonstrated to be successful in other subspecies [11^13]. Since there are distinct genetic di¡erences between the subspecies, it cannot be assumed that loci shown to be essential for the intracellular survival of novicida have the same functions in the more virulent subspecies so there is a de¢nite need to develop useful systems for the creation of mutants of subspecies holarctica and tularensis. Preliminary studies have indicated that the frequency of recombination in F. tularensis LVS is low and that only single homologous recombination occurred (unpublished). To this end, we aimed to develop a strategy that allowed allelic exchange via double crossover events. Such strategies are often based on introduction of genes that confer sensitivity of bacteria to certain extracellular compound, e.g. sucrose, streptomycin, or fusaric acid. The sacB gene encodes the secreted enzyme levansucrase, the expression of which is toxic for Gram-negative bacteria in the presence of sucrose [14,15]. Here we describe the construction of a vector and procedures designed for gene replacement in F. tularensis based on the sacB gene of Bacillus subtilis. We aimed to create a mutant defective in expression of a 23-kDa protein of F. tularensis LVS since we in a previous publication showed that this protein was one of only a few proteins that were signi¢cantly upregulated during intracellular infection or after exposure of F. tularensis to the bactericidal agent hydrogen peroxide [16]. Nothing is known about the function of this protein but in a recent publication it was identi¢ed as a component necessary for the ability of F. tularensis subspecies novicida to replicate
intracellularly [11]. The operon, igl (intracellular growth locus), containing the gene encoding the 23-kDa protein, henceforth designated iglC, the third gene of the operon, comprises four genes. We here demonstrate that the iglCcontaining operon exists in two copies in F. tularensis LVS and that the strategy developed allowed us to generate a mutant with deletions in both iglC alleles. Moreover, we observed that the mutant showed impaired multiplication in a macrophage cell line and was attenuated in mice.
2. Materials and methods 2.1. Bacterial strains and plasmids Bacterial strains and plasmids used in this study are described in Table 1. 2.2. Cell cultures and macrophage infection One day before the start of an experiment, J774A.1 cells were detached, resuspended in culture medium, Dulbecco’s modi¢ed Eagle medium (DMEM, Gibco BRL, Grand Island, NY, USA) with 10% (v/v) fetal calf serum and added to wells of a 24-well tissue culture plate. Peritoneal exudate cells (PEC) were obtained from Balb/cJ mice 3 days after intraperitoneal injection of 2 ml of 3% thioglycolate. They were washed with DMEM and suspended in the same medium as described for J774A.1 cells. After incubation overnight at 37‡C, wells were washed and reconstituted with fresh culture medium. To each well, a suspension of bacteria was added and bacterial uptake was allowed to occur for a 2-h period at 37‡C. After bacterial uptake, the monolayer was washed twice and incubated for indicated periods of time in culture medium with 5 Wg ml31 of gentamicin. For determination of numbers of intracellular bacteria, J774 cells were washed once in phos-
Table 1 Description of plasmids and strains used in the study Bacteria or plasmid
Characteristics
E. coli DH5K E. coli S17-1 F. tularensis LVS
F (B80dlacZvM15) recA1 endA1 gyrA96 thi-1 v(lacZYA-argF) U169 thi thr leu tonA lacY supE recA : :RP4-2-Tc: :Mu, Kn: :Tn7 Pmr , vaccine strai
E. coli S17-1/pHV33mob pUC19 pBlue-ScriptKS+
Ampr Ampr
PSa pDM4 pHV33 pHV33mob pHV33-sacB pPV pPVviglC
Cmr Ampr , Cmr , sacB, mob Ampr , Cmr , Tcr Ampr , Cmr , mob sacB, Cmr , Tcr Ampr , Cmr , sacB, mob Ampr , Cmr , sacB, mob
3
Reference and source hsdR17(r3 k
FEMSLE 10987 19-5-03
mþ k)
supE44 relA1 deoR
[29] [30] USAMRIID, Frederick, MD This study [31] Stratagene, La Jolla, CA [19] [20] [32] This study This study This study This study
I. Golovliov et al. / FEMS Microbiology Letters 222 (2003) 273^280
phate-bu¡ered saline (PBS) and lysed with 0.2 ml of 0.1% sodium deoxycholate. After addition of 1.8 ml of PBS, 100-Wl portions of each lysate, serially diluted in PBS, were plated on modi¢ed Thayer^Martin agar plates for determination of viable counts. 2.3. Detection of F. tularensis-mediated cytopathogenicity Cytopathogenicity was assessed by measuring the release of cytosolic lactate dehydrogenase (LDH) into the supernatant, which re£ects loss of plasma membrane integrity in infected cells. LDH levels were measured using the colorimetric Cytotox 96 Kit (Promega, Madison, WI, USA) according to the manufacturer’s instructions. The percentage of cytopathogenicity was calculated as 100U [(experimental release-spontaneous release)/(total releaseWspontaneous release)], where spontaneous release is the amount of LDH activity in the supernatant of uninfected cells and total release is the activity in cell lysates. 2.4. DNA manipulations All DNA manipulations, including puri¢cation, digestion with restriction endonucleases, ligations, and gel electrophoresis, were performed according to methods described by Sambrook et al. [17]. Puri¢cation of DNA fragments was done with QIAquick Gel Extraction Kit (Qiagen, Hilden, Germany) and chromosomal DNA with DNeasy Tissue Kit (Qiagen). Southern blot analysis was carried out on nylon membranes (Hybond-Nþ , Amersham Pharmacia, Uppsala, Sweden) as instructed by the manufacturer. Single-stranded DNA probe was end-labeled with [Q-32 P]dATP (Amersham Pharmacia) by use of T4 polynucleotide kinase (Promega). Hybridization was run in ExpressHyb hybridization solution (Clontech, Palo Alto, CA, USA) in accordance with the manufacturer’s protocol. 2.5. Gel electrophoresis and Western blotting Samples were applied on a uniform 12% acrylamide gel for electrophoresis as described by Laemmli [18]. For Western blot analysis, proteins in acrylamide gel were transferred to nitrocellulose ¢lters and probed with rabbit polyclonal anti-iglC sera. A horseradish peroxidase-conjugated secondary antibody system was used and ¢lters were developed by using the ECL Kit (Amersham Pharmacia). 2.6. Construction of vectors A suicide vector designed to be used for allelic replacement in F. tularensis contained the following elements: (i) an antibiotic marker, the chloramphenicol resistance gene (crg) from the Shigella £exneri plasmid pSa [19], which confers chloramphenicol resistance in F. tularensis [10] ; (ii) an origin of transfer (mob) and (iii) the SacB gene of
275
B. subtilis for excision of the vector. A PstI fragment from the B. subtilis plasmid pDM4 [20] containing the SacB gene including the promoter was cloned to pKK214 [21] or pHV33 [22]. Both are low-copy number plasmids able to replicate in F. tularensis. Spontaneous sucrose resistance occurred in F. tularensis LVS at a frequency of 1036 (data not shown), similar to what has been reported previously for Escherichia coli and Yersinia enterocolitica [23,24]. A BamHI fragment form pDM4 containing the origin of transfer, oriT, of plasmid RP4 [25] was cloned to BamHI of pHV33 and the resulting plasmid designated pHV33mob. The presence of oriT on plasmid usually results in e⁄cient conjugal transfer to recipient strains from donor strain expressing RP4 conjugative functions [26]. PHV33mob was e⁄ciently transferred from E. coli S17-1, containing the RP4 conjugative transfer apparatus, to F. tularensis LVS by mating and was used to optimize conditions of conjugation. A vector for site-directed mutagenesis was constructed using the pUC19 plasmid which is unable to replicate in F. tularensis. First, the BamHI fragment (1.6 kb) from pDM4 containing oriT was ligated to BamHI-digested pUC19 resulting in a pUC-mob plasmid. The SacB gene was excised from pDM4 as a 2.6-kb PstI fragment and ligated to PstI-digested pUC-mob. The resulting plasmid was partially digested with HindIII and a HindIII fragment (3.2 kb) from pSa containing the crg gene was inserted. Resistant transformants were obtained by selection on plates containing 10 Wg ml31 of chloramphenicol and one of clones containing the crg gene inserted in the HindIII site of the polylinker cloning site of pUC19 was selected. It was named pPV, and it has unique SalI and XbaI sites for cloning (Fig. 1A). 2.7. Construction of an iglC mutagenesis plasmid A region approximately 1500 bp upstream (primers C and E) and a region downstream (primers D and F) of the iglC gene were ampli¢ed by polymerase chain reaction (PCR). The 5P-primers contained SalI restriction sites (upstream region, primer C: 5P-ACG CGT CGA CTA CAA TGC TAG AAA CCT T-3P, and downstream region, primer D: 5P-GCG TGT CGA CTT AGC CGT GCC AAT TAC CA-3P) and the 3P-primers a BamHI site (upstream region, primer E: 5P-TCG GGA TCC TAT ACA TGC AGT AGG A-3P) or a PstI site (downstream region, primer F: 5P-AAA CTG CAG CTG CAT AGT AAG ATC GGA-3P). Each DNA fragment had a size of approximately 1.5 kb and included the ¢rst 80 nucleotides or the last 15 nucleotides, respectively, of the iglC gene. They were ligated to SalI/BamHI or SalI/PstI-digested plasmid pBlue-ScriptKS+ (Stratagene, La Jolla, CA, USA). From the recombinant plasmids, the cloned DNA fragments were excised with SalI and BamHI and both fragments ligated simultaneously to SalI-digested pPV. The resulting vector was designated pPV-viglC.
FEMSLE 10987 19-5-03
276
I. Golovliov et al. / FEMS Microbiology Letters 222 (2003) 273^280
Fig. 1. A: Description of the recombinant genes of pPV inserted in the multicloning site of pUC19. oriT and SacB genes originated from pDM4. The chloramphenicol resistance gene, crg, originated from pSA. B: Description of the igl operon of F. tularensis LVS. The operon contains four open reading frames, designated A^D. Localization of primers used for ampli¢cation of the various regions of the operon is indicated.
2.8. Conjugal transfer of plasmids
plating 10-fold serial dilutions. Colonies were counted after 3 days of incubation at 37‡C.
Early exponential cultures of 107 CFU ml31 of E. coli S17-1 and 109 CFU ml31 of F. tularensis LVS were concentrated by centrifugation and resuspended in 50 Wl of culture medium, mixed, and plated on either Luria agar (LA) or McLeod [16] plates. For optimization of conjugation, plates were incubated at various temperatures and after di¡erent time intervals. After incubation, cells were resuspended in PBS and plated on modi¢ed Gc agar base plates [BBL Gc agar base (Becton Dickinson, Sunnyvale, CA, USA) with the addition of 1% (w/v) of hemoglobin (Oxoid, Stockholm, Sweden) and 1% (v/v) of BBL IsoVitalex (Becton Dickinson)] containing 100 Wg ml31 of polymyxin B for counterselection of the donor E. coli strain [10] and 2.5 Wg ml31 of chloramphenicol. Frequencies of mating were expressed as the ratio of recipient cells that harbored the plasmid antibiotic markers versus the total number of recipients.
3. Results 3.1. Optimization of mating between E. coli and F. tularensis LVS F. tularensis LVS was conjugated with E. coli S17 carrying the pHV33-mob plasmid at various temperatures as described in Section 2. The results are presented in Table 2. Optimal conditions were obtained when the mixture of bacteria were incubated on LA plates at 25‡C for 18 h. Total numbers of transformants on LA were approximately 2U107 /reaction and the estimated frequency of transfer was approximately 2U1032 . These conditions were used in the work described henceforth. If McLeod plates were used instead of LA, the number of transformants was at least 2 log10 lower (data not shown).
2.9. Inoculation and enumeration of bacteria 3.2. Creation of a viglC protein mutant An area of approximately 2 cm2 of the skin of upper thorax was shaved 2 days before inoculation. Mice were challenged with an intradermal inoculation of 5U107 of F. tularensis LVS and killed by decapitation after 4 weeks of infection. Spleen and liver samples were homogenized and the number of F. tularensis LVS was calculated by
F. tularensis LVS was conjugated with E. coli S17-1 carrying the pPV-viglC plasmid and after incubation on LA plates overnight, plated on McLeod plates containing 100 Wg ml31 of polymyxin B and 2.5 Wg ml31 of chloramphenicol. Transformants were observed after 5^6 days of
Table 2 In£uence of temperature on conjugation frequency (‡C)
15 25 30 37
Bacterial number before mating
Bacterial number after mating
F. tularensis
E. coli
F. tularensis
E. coli
1U109
1U107
1U109 2U109 2U109 1.5U109
4U107 3U109 1U109 1U109
FEMSLE 10987 19-5-03
Number of conjugants
Frequency of conjugation
2U105 2U107 3U106 5U105
2U1034 1U1032 1.5U1033 3.3U1034
I. Golovliov et al. / FEMS Microbiology Letters 222 (2003) 273^280
Fig. 2. Southern blot hybridization of DNA from mutant and wild-type strains. Samples of DNA were digested with ClaI prior to separation by agarose gel electrophoresis and Southern blotting and hybridization. The blot was probed with a labelled oligonucleotide complementary to a part of iglC lacking in viglC. The sizes were estimated from the mobility of a 1-kb DNA standard. Lane 1: F. tularensis viglC1+2 DNA; lane 2: F. tularensis viglC1 DNA; lane 3: F. tularensis LVS DNA.
incubation at 37‡C in 5% CO2 . In total, 20 transformants were obtained and none in control reactions where S17-1 was mixed with F. tularensis LVS. No replicating plasmids were recovered from any of the transformants. Since pPVviglC cannot replicate in F. tularensis, either illegitimate recombination should have occurred or the transformants should contain the plasmid inserted into the chromosome after recombination of the wild-type, chromosomal iglC and the truncated variant of pPV-viglC. To analyze the recombination event, we used primers complementary to the upstream (A) or downstream (B) region of viglC of the pPV vector and primers C and D, described in Section 2, located upstream and downstream, respectively, of iglC (Fig. 1B). PCR was then performed using primer pairs A+D or B+C. Using the two primer combinations, it was possible to amplify DNA from all transformants, indicating that legitimate recombination had occurred in all cases. Recombination had occurred with an equal frequency in the upstream and downstream £anking regions. Twenty recombinant clones were isolated and incubated on McLeod plates with 10 Wg ml31 of chloramphenicol. To obtain a clone lacking the wild-type gene, we selected for a second recombination event by plating on medium containing 5% sucrose. Of the 15 analyzed sucrose-resistant colonies, all were sensitive to chloramphenicol, indicating that plasmid exclusion had occurred. This was con¢rmed since no ampli¢cation resulted from PCR with primers speci¢c to the plasmid. Of the chloramphenicol-sensitive clones, 70% had a wild-type phenotype of the
277
iglC gene, presumably as a result of abortive allelic exchange through the same sequence as during the ¢rst crossover event. All other colonies had a mixed phenotype retaining both wild-type and the truncated iglC gene. This indicated the presence of two copies of the iglC gene in the LVS genome. One of the clones, designated viglC1, was selected for the second round of conjugation and all procedures were repeated as described above. After mating, 40 chloramphenicol-resistant clones were obtained. A PCR using primers A+B was performed. Only if the ¢rst crossover of the wild-type gene had occurred, these primers should amplify the truncated form of the 23-kDa protein gene. Of the 40 clones tested, only one showed the absence of the wild-type gene. In all other clones, recombination between the truncated chromosomal and plasmid genes had occurred (data not shown). After selection on plates with 5% sucrose, more than 90% of the clones tested were chloramphenicol sensitive and harbored the truncated form of the iglC gene only. Those clones were considered as real iglC mutants, henceforth designated viglC1+2. To further analyze these mutants, a Southern blot with ClaI-digested chromosomal DNA from F. tularensis LVS was performed using a 32 P-labeled oligonucleotide complementary to a region of the deleted region as a probe. Preliminary results showed that cleavage with ClaI allowed discrimination between the two copies of iglC (data not shown). As expected, the probe hybridized to two bands of wild-type LVS DNA (approximately 8 and 15 kb) but only one in viglC1 and no hybridization was observed with viglC1+2 mutant (Fig. 2). A PCR analysis
Fig. 3. Expression of iglC. Pellets after centrifugation of bacterial cultures were lysed in Laemmli bu¡er, separated by electrophoresis in a 12% polyacrylamide gel and transferred to a nitrocellulose ¢lter. The ¢lter was probed with a polyclonal rabbit antiserum speci¢c to the iglC protein at a dilution of 1:200. Lane 1: F. tularensis LVS; lane 2: F. tularensis viglC1; lane 3: F. tularensis viglC1+2.
FEMSLE 10987 19-5-03
278
I. Golovliov et al. / FEMS Microbiology Letters 222 (2003) 273^280
Fig. 4. Growth of F. tularensis in J774.A1 (A) or PEC (B) cells. Data from a representative experiment are shown and the results are expressed as the CFU (mean log10 U S.D.)/well of F. tularensis, based on triplicate wells.
(primers A+B) veri¢ed the presence of a mutated copy of the iglC gene and a wild-type gene in viglC1 and no wildtype copy in viglC1+2 mutant (data not shown). 3.3. Protein expression in viglC To verify that the 23-kDa protein was absent in viglC1+2, bacteria were grown on a McLeod plate and subjected to Western blot analysis. After sodium dodecyl sulfate^polyacrylamide gel electrophoresis and transfer to nitrocellulose ¢lter, the extracts were probed with a rabbit polyclonal serum against iglC. No immunoreactive band was present in the mutant strain (Fig. 3). 3.4. Growth of F. tularensis mutant strains in J774 cells and peritoneal macrophages The 23-kDa mutant strains and wild-type F. tularensis LVS grew equally well in Chamberlain medium (data not
shown). There was no signi¢cant di¡erence in the uptake in J774A.1 cells of the two bacterial strains. However, in contrast to F. tularensis LVS or viglC1, viglC1+2 showed no growth in J774 cells and, in fact, bacterial numbers decreased over time. The former two strains increased 2.5 log10 during the ¢rst 24 h whereas the numbers of viglC1+2 increased only 0.5 log10 (Fig. 4). F. tularensis LVS and viglC1 gave a pronounced cytopathogenic e¡ect and after 24 h, release of LDH was more than 65% of the maximum release, whereas viglC1+2 exerted no cytopathogenic e¡ect (Fig. 5A). Similar results were obtained in peritoneal macrophages. The wild-type and viglC1 grew approximately 1 log10 in the cells whereas viglC1+2 showed no growth (Fig. 4B). Again, the former two strains exerted a strong cytopathogenic e¡ect but no such e¡ect was seen in cells infected with the mutant strain (Fig. 5B). 3.5. Growth of F. tularensis viglC1+2 in mice Mice were challenged with 5U107 CFU of F. tularensis LVS or F. tularensis viglC1+2. This dose represents approximately 100 LD50 of F. tularensis LVS. Accordingly, all of the 10 mice died within 6 days after challenge with F. tularensis LVS whereas all of the 10 mice challenged with the mutant strains survived for 4 weeks and when killed, had eradicated the bacteria.
4. Discussion
Fig. 5. Cytopathogenicity of mutant and wild-type F. tularensis strains in J774.A1 (A) or PEC (B) cells after 24 h of infection. Cytopathogenicity was assayed by release of LDH from infected cells as described in Section 2.
Little is known regarding virulence mechanisms of F. tularensis. A major drawback has been a lack of methods to generate mutants of isolates of the clinically important subspecies, tularensis and holarctica. This is the ¢rst demonstration of a genetic system that allowed the gener-
FEMSLE 10987 19-5-03
I. Golovliov et al. / FEMS Microbiology Letters 222 (2003) 273^280
ation of de¢ned mutants in F. tularensis. The constructed vector allows the creation of mutants with point mutations, deletions, or insertions. Previously published methods on Francisella have been validated using F. tularensis subspecies novicida, a subspecies more amenable to genetic manipulation than the other subspecies of F. tularensis, and all have been based on random insertions of minitransposons carrying antibiotic markers [7]. A disadvantage of the latter strategy is that it will give deletion mutants only and may also have polar e¡ects. An advantage of the present strategy is the possibility to use conjugation for plasmid delivery. It appears as if conjugation is an e⁄cient method whereas no mutants were obtained when pPV-viglC was delivered by electroporation (data not shown). Another important feature of the system is that the recombinants do not express any antibiotic resistance markers and, in fact, no foreign DNA remains after selection on sucrose. This feature may prove advantageous if the method is going to be used for the generation of live attenuated vaccine strains. Previously, we identi¢ed the 23-kDa protein of F. tularensis as markedly upregulated during intracellular macrophage infection [16]. The previous study demonstrated that remarkably few proteins were induced during the intracellular growth. This is in contrast to ¢ndings on other facultative intracellular bacteria such as S. enterica serovar typhimurium which responded to the intracellular environment by up- or downregulating a large number of proteins [27]. Previously, we were not able to assign any speci¢c role to the 23-kDa protein since it showed no homology to other bacterial proteins. The creation of a speci¢c mutant will therefore be of tremendous help in understanding the role of the protein in the virulence of F. tularensis. In fact, when screening the mutant for a phenotype, we discovered that the F. tularensis iglC1+2 mutant lacked the property of the wild-type F. tularensis LVS to inhibit Toll-likemediated signaling via MAPK and NF-UB pathways and secretion of TNF-K and IL-1L of murine macrophages [28]. Although we have not identi¢ed the underlying molecular mechanism, the development of the genetic system will allow means to study this. Our demonstration of impaired intracellular growth and marked attenuation in vivo of the mutant strain provides direct evidence for an important role of the 23-kDa protein in the virulence of F. tularensis. Our ¢ndings also corroborate those of Nano et al. [11], who recently demonstrated that a transposon insertion in F. tularensis subspecies novicida rendered it incapable of intracellular proliferation. In this study, insertions in both iglA and iglC gave an attenuated phenotype. However, it could not be excluded that the attenuation was the result of polar effects on downstream genes of the operon. We have also been successful in developing other speci¢c mutants in F. tularensis LVS and also in virulent strains of subspecies holarctica (unpublished data). The development of genetic tools together with the rap-
279
idly increasing genomic information on F. tularensis will result in signi¢cant progress in our understanding of the hitherto ill-de¢ned virulence mechanisms of this potent intracellular pathogen.
Acknowledgements Grant support was obtained from the the Swedish Medical Research Council, Samverkansna«mnden Norra Sjukvafirdsregionen, Umeafi, and the Medical Faculty, Umeafi University, Umeafi, Sweden.
References [1] Sjo«stedt, A. (2003) Family XVII. Francisellaceae, Genus I. Francisella. In: Bergey’s manual of systematic bacteriology (Brenner, D.J., Ed.). Springer, New York. [2] Sjo«stedt, A. (2003) Virulence determinants and protective antigens of Francisella tularensis. Curr. Opin. Microbiol. 6, 1^6. [3] Eigelsbach, H.T. and Down, C.M. (1961) Prophylactic e¡ectiveness of live and killed tularemia vaccines. I. Production of vaccine and evaluation in the white mouse and guinea pig. J. Immunol. 87, 415^ 425. [4] Burke, D.S. (1977) Immunization against tularemia: analysis of the e¡ectiveness of live Francisella tularensis vaccine in prevention of laboratory-acquired tularemia. J. Infect. Dis. 135, 55^60. [5] Fortier, A.H., Slayter, M.V., Ziemba, R., Meltzer, M.S. and Nacy, C.A. (1991) Live vaccine strain of Francisella tularensis: infection and immunity in mice. Infect. Immun. 59, 2922^2928. [6] Ta«rnvik, A. (1989) Nature of protective immunity to Francisella tularensis. Rev. Infect. Dis. 11, 440^451. [7] Anthony, L.S., Gu, M.Z., Cowley, S.C., Leung, W.W. and Nano, F.E. (1991) Transformation and allelic replacement in Francisella spp. J. Gen. Microbiol. 137, 2697^2703. [8] Baron, G.S., Myltseva, S.V. and Nano, F.E. (1995) Electroporation of Francisella tularensis. Methods Mol. Biol. 47, 149^154. [9] Pavlov, V.M., Mokreivich, A.N. and Volkovoy, K. (1996) Cryptic plasmid pFNL10 from Francisella novicida-like F6168 : The base of plasmid vectors for Francisella tularensis. FEMS Immunol. Med. Microbiol. 13, 253^256. [10] Pomerantsev, A.P., Domaradskii, I.V. and Shishkova, N.A. (1991) Hybrid plasmid pSKEFT5 assures the transfer of plasmids Sa from Escherichia coli cells into cells of the tularemia microbe vaccine strain cells. Mol. Gen. Mikrobiol. Virusol. 8, 13^16. [11] Gray, C.G., Cowley, S.C., Cheung, K.K. and Nano, F.E. (2002) The identi¢cation of ¢ve genetic loci of Francisella novicida associated with intracellular growth. FEMS Microbiol. Lett. 215, 53^56. [12] Cowley, S.C., Gray, C.J. and Nano, F.E. (2000) Isolation and characterization of Francisella novicida mutants defective in lipopolysaccharide biosynthesis. FEMS Microbiol. Lett. 182, 63^67. [13] Baron, G.S. and Nano, F.E. (1999) An erythromycin resistance cassette and mini-transposon for constructing transcriptional fusions to cat. Gene 229, 59^65. [14] Gay, P., Le Coq, D., Steinmetz, M., Berkelman, T. and Kado, C.I. (1985) Positive selection procedure for entrapment of insertion sequence elements in gram-negative bacteria. J. Bacteriol. 164, 918^ 921. [15] Reyrat, J.M., Pelicic, V., Gicquel, B. and Rappuoli, R. (1998) Counterselectable markers: untapped tools for bacterial genetics and pathogenesis. Infect. Immun. 66, 4011^4017. [16] Golovliov, I., Ericsson, M., Sandstro«m, G., Ta«rnvik, A. and Sjo«stedt, A. (1997) Identi¢cation of proteins of Francisella tularensis induced
FEMSLE 10987 19-5-03
280
[17]
[18] [19]
[20]
[21]
[22] [23]
[24]
I. Golovliov et al. / FEMS Microbiology Letters 222 (2003) 273^280 during growth in macrophages and cloning of the gene encoding a prominently induced 23-kilodalton protein. Infect. Immun. 65, 2183^ 2189. Sambrook, J., Fritch, E.F. and Maniatis, T. (1989) Molecular Cloning: A Laboratory Manual. Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NY. Laemmli, U.K. (1970) Cleavage of structural proteins during the assembly of the head of bacteriophage T4. Nature 227, 680^685. Watanabe, T., Furuse, C. and Sakaizumi, S. (1968) Transduction of various R factors by phage P1 in Escherichia coli and by phage P22 in Salmonella typhimurium. J. Bacteriol. 96, 1791^1795. Milton, D.L., O’Toole, R., Horstedt, P. and Wolf-Watz, H. (1996) Flagellin A is essential for the virulence of Vibrio anguillarum. J. Bacteriol. 178, 1310^1319. Kuoppa, K., Forsberg, A. and Norqvist, A. (2001) Construction of a reporter plasmid for screening in vivo promoter activity in Francisella tularensis. FEMS Microbiol. Lett. 205, 77^81. Ehrlich, S.D. (1978) DNA cloning in Bacillus subtilis. Proc. Natl. Acad. Sci. USA 75, 1433^1436. Kaniga, K., Delor, I. and Cornelis, G.R. (1991) A wide-host-range suicide vector for improving reverse genetics in gram-negative bacteria: inactivation of the blaA gene of Yersinia enterocolitica. Gene 109, 137^141. Blom¢eld, I.C., Vaughn, V., Rest, R.F. and Eisenstein, B.I. (1991) Allelic exchange in Escherichia coli using the Bacillus subtilis sacB gene and a temperature-sensitive pSC101 replicon. Mol. Microbiol. 5, 1447^1457.
[25] Priefer, U.B., Simon, R. and Puhler, A. (1985) Extension of the host range of Escherichia coli vectors by incorporation of RSF1010 replication and mobilization functions. J. Bacteriol. 163, 324^330. [26] Miller, V.L. and Mekalanos, J.J. (1988) A novel suicide vector and its use in construction of insertion mutations : osmoregulation of outer membrane proteins and virulence determinants in Vibrio cholerae requires toxR. J. Bacteriol. 170, 2575^2583. [27] Abshire, K.Z. and Neidhardt, F.C. (1993) Analysis of proteins synthesized by Salmonella typhimurium during growth within a host macrophage. J. Bacteriol. 175, 3734^3743. [28] Telepnev, M., Golovliov, I., Grundstrom, T., Tarnvik, A. and Sjostedt, A. (2003) Francisella tularensis inhibits Toll-like receptor-mediated activation of intracellular signalling and secretion of TNF-alpha and IL-1 from murine macrophages. Cell. Microbiol. 5, 41^51. [29] Woodcock, D.M., Crowther, P.J., Doherty, J., Je¡erson, S., DeCruz, E., Noyer-Weidner, M., Smith, S.S., Michael, M.Z. and Graham, M.W. (1989) Quantitative evaluation of Escherichia coli host strains for tolerance to cytosine methylation in plasmid and phage recombinants. Nucleic Acids Res. 17, 3469^3478. [30] Simon, R., Priefer, U. and Puhler, A. (1983) A broad host range mobilization system for in vivo genetic engineering : transposon mutagenesis in Gram-negative bacteria. Biotechnology 1, 784^791. [31] Yanisch-Perron, C., Vieira, J. and Messing, J. (1985) Improved M13 phage cloning vectors and host strains : nucleotide sequences of the M13mp18 and pUC19 vectors. Gene 33, 103^119. [32] Ehrlich, S.D. (1978) DNA cloning in Bacillus subtilis. Proc. Natl. Acad. Sci. USA 75, 1433^1436.
FEMSLE 10987 19-5-03