| Collection Mémoires et thèses électroniques | |
| AccueilÀ proposNous joindre |
| Appendice 1 | ||
|---|---|---|
 |  | |
Table des matières
Dario Lehoux, François Sanschagrin, Irena Kukavica-Ibrulj, Eric Potvin and Roger C. Levesque*
Centre de Recherche sur la Fonction Structure et Ingénierie des Protéines, Pavillon Charles-Eugène Marchand et Faculté de Médecine, Université Laval, Ste-Foy, Québec, Canada. G1k 7P4
*Correspondence: E-mail: rclevesq@rsvs.ulaval.ca; Tel: (418)656-3070; Fax: (418)-656-7176
Introduction
Microbial pathogens possess a repertoire of virulence determinants that make unique contributions to bacterial fitness during an infection. Understanding the expression of these genes involved in infection remains the holy grail of research in bacterial virulence. The simultaneous development of molecular biological technologies including recent progress in microbial genomics and bioinformatics coupled to molecular genetic bacterial approaches has facilitated the dissection of a plethora of regulatory systems implicated in virulence (1). However, the complexities of bacterial gene expression in vivo during mammalian infection cannot be addressed solely by in vitro experiments coupled to in vivo analysis of a single gene. The infected host presumably represents a complex and dynamic environment, which is modified in a time and dose-dependent manner during the infection process. One may assume then that the bacterial pathogen must turn on or off genes specific for the conditions encountered in vivo via a variety of stimuli. In addition, several discrete cellular steps presumably permit the controlled expression of genes essential for initiation of the infection process and its maintenance. The pattern of bacterial genes expressed solely in vivo and genes essential for infection remains to be investigated if one wants to have a global vision of the infection versus the pathogen genomic content. This will be a formidable task requiring several complementary techniques such as in vivo expression technology (IVET), signature-tagged mutagenesis (STM), elegant bioinformatics analysis and functional genomics analysis via proteomics using 2D gel electrophoresis differential display, DNA and protein arrays and chip technology (1). There is also an urgent need for new technologies for analysis of genes expressed in vivo as the current tools have significant limitations. In this chapter, we focus on the recent progress and adaptations of signature-tagged mutagenesis (STM).
STM can now be used in any typical laboratory; the screening can now be done rapidly by automated high throughput robotics using polymerase chain reaction (PCR) instead of hybridization. The design of tags has been simplified and several different miniTn5s with a unique phenotypic selection for each can now be utilized. These modifications of STM developed in our laboratory have been applied to Pseudomonas aeruginosa PAO1 whose 6.3 Mb genome has been completely sequenced (2). The PCR-based STM is described in detail including the strategies and protocols using as prototype the P. aeruginosa STM project on-going in our laboratory.
Description of signature-tagged mutagenesis
STM is an elegant mutation-based screening method using a population of bacterial mutants for the simultaneous identification of virulence genes of microbial pathogens by negative selection (3). The technique depends upon an in vivo selection phenotypewhereas genes inactivated will have an effect on diminishing bacterial virulence tested in a model host of infection. This elegant concept developed by Holden and colleagues was to allow a relatively rapid unbiased search for virulence genes using an animal host to select against strains carrying mutations in genes affecting virulence, among a mixed population of mutants. To avoid the typical labour-intensive screening of individual mutants in the first steps, each mutant is tagged with a different and unique DNA signature; the power of the STM technique allows large numbers of different strains to be screened at the same time in the same animal host.
In the classical STM protocol developed by Holden (3), a comparative hybridization technique is used and employs as mutagen a collection of transposons, each modified by the incorporation of a different DNA tag sequence. The concept here is that when the tagged mutagens integrate into the bacterial chromosome, each individual mutant can be distinguished from every single other mutant based on the unique and different tags carried by the strains. The first tag collection were originally designed as short DNA segments containing 40 bps variable regions flanked by invariant arms that facilitated the co-amplification and radioactive labelling of the central portions by PCR subsequently used as a hybridization probe (3).
Since STM is an en masse screening technique, mutagenized strains are usually kept individually in arrays using microtiter plates. Colony blots or DNA dot blots are prepared from this in vitro pool and compared by hybridization to the same pool of strains obtained, but after passage in the animal host. Technically, PCR is used to prepare labelled probes representing the tags present in the inoculum (input) and recovered from the host (output). Hybridization of the tags from the input and output pools to the colony or dot blot permits the identification of mutants that fail to grow in vivo, because these tags will not be present in the output pools. These strains can then be identified and recovered from the original arrays and the nucleotide sequence of DNA flanking the mutation site can be determined. Hence, STM has and will benefit immensely from the sequence data of all bacterial genomics projects.
In the original method, tags were incorporated into a miniTn5 transposon and their suitability was checked prior to use by hybridization of amplified labelled tags to DNA colony blots used to generate the probes. Mutants whose tags failed to yield clear signals on autoradiograms were discarded; those that gave good signals were assembled into new pools for screening into animals (3). This careful analysis was done prior to STM so as to diminish the inherent problems of hybridization caused by problematic tags. The STM method was subsequently modified to avoid the pre-screening process where a series of Tn917 transposons were selected prior to mutagenesis on the basis of efficient tag amplification and labelling, and lack of cross-hybridization to other tags (4). Since in this case the identity of each tag is known, each modified transposon can then be used to generate an infinite number of mutants, it can also be screened using plasmid or tag DNA rather than colony blots and increases the sensitivity and specificity of the assay. An additional modification of the technique was done where each of 96 tags were introduced at a disrupted URA3 locus of Candida glabrata (5). For each tagged strain, different mutants were created by insertional mutagenesis using a plasmid that simultaneously complemented the URA3 mutation. The DNA flanking the mutations could be cloned easily by plasmid rescue in Escherichia coli. From the original STM, several variations have been applied to many different bacterial pathogens on the basis of the particularities and limitations of each living system studied with the data recently summarized (6). Indeed, the bacterial genetic tools of transposons are not easily amenable to genetic analysis in some species of bacteria. For example, Tn5 transposons and their derivatives do not transpose well in some Gram negative bacteria such as Pasteurellacea, while not at all in Gram positive bacteria. However, the recent development of new transposons as tools for genetic analysis in a wide range of organisms and development of a complete in vitro transposition assay should permit the application of STM to a large variety of life forms (7). We present here a general approach of PCR-based STM but that can be adapted to any organism of interest.
General Strategy for the PCR-based STM
As in the original STM, PCR-based STM is divided into two major steps, an in vitro step and the in vivo selection step (Fig1 A and Fig 1 B). The first step involves the construction of specific and defined DNA tags and the preparation of the library of tagged mutants for the organism to be studied. In this first phase, a strategy needs to be devised and careful planning implies: a) the design of DNA tags as oligonucleotides; b) the cloning of these tags into a transposable element that gives reliable mutations in the organism to be studied by insertional mutagenesis; c) the transfer of the transposon containing specific tags to the organism to be analyzed and a reliable method to select the recipients; d) the assembly of an array of tagged mutants. The second major step necessitates an animal or presumably a cell model for in vivo screening of the library. In this second phase, a strategy needs to be defined so as to consider the power and the limitations of the animal models used and decide the number of rounds of STM screening and the number of animals to be utilized. Also, STM involves a systematic characterization of mutant strains selected in vivo. The first crucial data is: 1) to obtain as much DNA sequence as possible around the site of the insertion mutation by cloning the transposon marker or by RT-PCR; 2) to confirm that the tagged mutation is the cause of virulence attenuation. The final phase in STM is to classify the genes identified as playing a significant role in virulence, define their function and their role in pathogenesis. Indeed this is the ultimate goal of STM. Here, we describe the rationale and step by step protocols essential for PCR-based STM and demonstrate typical results using P. aeruginosa as a model organism to be studied.
Methods
The design of oligonucleotides as signature DNA tags
A collection of complementary oligomers are first synthesized, rendered double stranded DNA, and are cloned into a mobile element for insertional mutagenesis, in this case derivatives of transposon mini-Tn5s. The PCR-based STM method that we developed involves designing pairs (24 in this case, but in theory 48 and 96 could be utilized) of 21-mers (Table 1) synthesized as complementary DNA strands for cloning into the mini-Tn5Km2, miniTn5Tc and miniTn5TcGFP plasmid vectors. The rationale here was to limit the tag complexity and used the power of bacterial genetics phenotypic selection to obtain a collection of 72 mutants per pool (selected using the miniTn5markers Km, Tc, Tc or GFP (Fig 3). There is a precedent for limiting the number of tags used. When Salmonella typhimurium was inoculated into the peritoneal cavities of mice, pools of 96 different mutants gave reproducible hybridization signals after 3 days of infection, whereas 192 did not (3). Problems with the complexity of the pool tags has been described also for Vibrio cholerae where it was necessary to reduce the complexity of orally inoculated pool to 48 strains to give reproducible results (8).
The sets of 24 tags are repeatedly used toconstruct 24 libraries (Fig. 1 A)and used for specific DNA amplifications as signature tags easily detectable by multiplex PCR (Fig. 4). Tagged products from arrayed bacterial clones can be compared as DNA products of a specific length separated by agarose gel electrophoresis.
Protocol 1: Synthesis and in vitro construction of tagged plasmids.
Twenty-four pairs of 21-mers were used. The first twelve pairs corresponded to ones described in Lehoux et al., 1999 (9) and the other twelve pairs were designed following the same three basic rules: (i) similar Tm of 64°C to simplify tag comparisons by using one step of PCR; (ii) invariable 5’-ends with higher G than at the 3’-end to optimize PCR amplification reactions; (iii) a variable 3’-end for an optimized yield of specific amplification product from each tag.
Annealing of complementary DNA:
1. A collection of twenty four defined 21-mers oligonucleotides were synthesized along with their complementary DNA strands as tags and are listed in Table 1). The synthesis was supplied by MGW Biotech. Inc. (High Point, NC, USA).
2. Annealing reactions contained 50 pmoles of both complementary oligonucleotides in 100 µl of medium salt buffer.
3. This oligonucleotide mixture is heated 5 min. at 95°C, left to cool slowly at room temperature in the block heater, and kept on ice.
Cloning of tags into a transposable element
The cloning if tags into an appropriate transposable will depend upon the organism to be studied by insertional mutagenesis. A plethora of mobile elements are currently available (7). The number of transposons that can be used in STM is limited by several factors including easy manipulation in vitro for DNA constructions, high frequency of transposition and random insertion in the genome of the host and controlled frequency of insertions as in miniTn5s. Obviously, STM cannot provide any information about genes that are essential which are totally inactivated by insertional mutagenesis. In contrast, an insertion within a non-essential gene crucial in virulence would not be expected to affect the growth rate of cells (10).
In our case for P. aeruginosa, each of the 3 pUTmini-Tn5 plasmid DNAs (pUTmini-Tn5Km2; pUTmini-Tn5Tet and pUTmini-Tn5TetGFP) was used (12, 13, 14). This collection of Tn5-derived minitransposons has been constructed that simplifies substantially the generation of insertion mutants, in vivo fusions with reporter genes, and the introduction of foreign DNA fragments into the chromosome of a variety of Gram negative bacteria. The miniTn5 consists of genes specifying resistance to kanamycin, tetracycline and the green fluorescent protein (Fig. 2) with unique cloning sites for tag insertion flanked by 19-base-pair terminal repeats, the I and the O ends. The transposons are located on a R6K-based suicide delivery plasmid pUT where the Pi protein is furnished by the donor cell; the pUT plasmid provides the IS50R transposase tnp gene in cis but external to the mobile element and whose conjugal transfer to recipients is mediated by RP4 mobilization functions in the donor (15).
Plasmid DNA was prepared, ligated with double stranded DNA tags in 24 separate reactions (16). The pUTmini-Tn5 Km2 was digested with KpnI (New England Biolabs, Mississauga, Ontario, Ca.) and recombinant molecules constructed in vitro by blunt-end fill-in with T4 DNA polymerase (GIBCO BRL Products, Gaithersburg, MD, USA). The pUTmini-Tn5 Tc and Gfp were digested with NotI (New England Biolabs, Mississauga, Ontario, Ca.) and recombinant molecules constructed in vitro by blunt-end fill-in with Klenow (New England Biolabs). The ligation reactions were doneusing T4 DNA ligase, following the manufacturer’s recommendations.Plasmids were transformed into E. coli S17-1 λpir (15) by electroporation and transformants were selected on TSA (Difco, Detroit, MI, USA) supplemented with 50µg/ml of ampicillin (Ap) (Sigma Chemical, St. Louis, MO, USA) and kanamycin (Km) (Sigma Chemical) for pUTmini-Tn5 Km2, tetracycline (5 µg/ml) for pUTmini-Tn5 Tc and Gfp.Technically, we provide below a simple protocol for ligation, cloning and electroporation.
Protocol 2: Ligation of tags, electroporation and screening of mini-Tn5s
1. 0.04 pmoles of plasmid are ligated to 1 pmole of double stranded DNA tags in a final volume of 10 µl of T4 DNA ligase 1X buffer containing 400 units of T4 DNA ligase.
2. Ligated products are purified using microcon PCR (Millipore) as described by the manufacturer's instructions and resuspended in 5 µl of H2O.
3. All the 5 µl containing ligated products are transformed into E. coli S17-λpir by electroporation using a Bio-Rad apparatus at 2.5 KV, 200 Ohms, 25 µF in a 2 mm electroporation gap cuvette. After electroporation, 0.8 ml of SOC is added to cells which are transferred in culture tubes to incubated 1 hour at 37°C.
4. Transformed bacteria containing tagged plasmids are selected on TSB supplemented with 50 µg/ml of ampicillin and 50 µg/ml of kanamycin by plating 100 µl of transformed cells.
5. Single colonies are selected, purified and screened by colony PCR in 50 µl reaction volumes containing: 10 µl of boiled bacterial colonies in 100 µl of TE PCR (10 mM Tris-HCl pH 7.4; EDTA 0.1 mM); 5 µl of 10X Taq polymerase (GibcoBRL) reaction buffer; 1,5 mM MgCl2; 200 µM of each dNTPs; 10 pmoles of one of the oligonucleotide tag (9) used to construct the DNA tags as a 5-’ primer and 10 pmoles of the pUTKanaR1 (5’-GCGGCCTCGAGCAAGACGTTT-‘3) as the 3-’ primer in the kanamycin resistance gene; 2,5 units of Taq polymerase (GibcoBRL). Thermal cycling conditions were (touchdown PCR): a hot start for 7 min. at 95°C, 2 cycles at 95°C for 1 min., 70 to 60°C for 1 min., and at 72°C for 1 min., then followed by 10 cycles at 95°C for 1 min., 60°C for 1 min., 72°C for 1 min. in a DNA Thermal Cycler (Perkin Elmer Cetus). Ten microliters of DNA amplified products were analyzed by electrophoresis in a 1% agarose gel, 1X Tris-borate EDTA buffer and stained for 10 min in 0.5 µg/ml ethidium bromide solution (16). The amplified product will have a size of 500 base pairs.
Protocol 3: Growth of bacterial strains
Escherichia coli strains were grown in tryptic soy broth (TSB). P. aeruginosa strains (PAO1 and PAO909) were grown in brain heart infusion (Difco). When needed, these media were supplemented with 1.5% of bacto-agar, ampicillin (50 µg/ml), chloramphenicol (5 µg/ml), kanamycin (50 µg/ml for E. coli and 500 µg/ml for P. aeruginosa in media with Bacto agar, and 300 µg/ml in liquid media), tetracycline (5 µg/ml for E. coli and 15 µg/ml for P. aeruginosa))
Transfer of the transposon containing specific tags to the organism to be analyzed
The signature tags inserted into transposons are replicated into E. coli and will need to be inserted into the chromosome of the host to be studied. With bacteria, this is usually achieved by conjugation at a high frequency. Generalities on the methods of transfer that can be used in any system of choice will depend upon the simplicity and capability of transferring the tags to the organism to be studied. We present below the protocol used with P. aeruginosa where transfer by mating is at a high level.
Protocol 4: Construction of 72 mini-Tn5 P. aeruginosa mutant libraries
1. E. coli S17-λpir containing the pUTminiTn5 plasmids with tags is used as a donor for conjugal transfer into the recipient strain, in our case P. aeruginosa PAO1. It is important to establish the ratio of donor: recipient to obtain the maximum of exconjugants (17). For P. aeruginosa, we used 1 donor for 10 recipient cells. Bacterial cells are mixed and spotted as a 50 µl drop on a sterile nylon membrane placed on a non-selective BHIA plate. Plates are incubated at 30 °C for 8h.
2. Filters are washed with 10 ml of sterile phosphate buffered saline to recover bacteria.
3. Five 100 µl aliquots of the PBS solution containing exconjugants are plated on 5 BHIA plates supplemented with the appropriate antibiotics to select for the strain. Kanamycin is used to select exconjugants with the mini-Tn5Km2 inserted into their chromosomes and plates are incubated overnight at 37°C.
4. Selected colonies are picked on BHIA supplemented with ampicillin to exclude bacterial colonies having the suicide donor plasmid pUTmini-Tn5 Km2 inserted into the chromosome by homologous recombination. Exconjugants were selected on BHIA supplemented with Cm (5µg/ml) (Sigma Chemical) and Km (500 µg/ml) for the mini-Tn5 Km2 or Tc (15 µg) for the mini-Tn5 Tc and mini-Tn5gfp.
5. Kanamycin resistant and ampicillin sensitive exconjugants are arrayed as libraries of 96 clones in 2 ml 96-wells plate in 1.5 ml of BHI supplemented with kanamycin and appropriate antibiotic.
6. As an STM working scheme, one mutant from each library is picked to form 96 pools of 72 unique tagged mutants in wells of 2 ml microtiter plates. The 2 ml 96-wells plates are incubated from 18-22h at 37°C.
Assembly of an array of tagged mutants
The exconjugants were arrayed as libraries of 96 clones in 2 ml 96 wells plates. In a defined library, each mutant had the same tag but assumed to be inserted at a different location in the bacterial chromosome. As an STM working scheme, one mutant from each library was picked to form 96 pools of 72 unique tagged mutants.
Screening of STM mutants in animal models
STM usually implies the use of an animal or simple cell model to mimic an insect, a plant, an animal or a human disease. As expected, these models cannot necessarily reproduce faithfully all the conditions of the infection. Several parameters need to be considered: 1) the number of different mutant strains to be used in a pool; 2) the route of administration, the dose for the inoculum and the incubation period; 3) an additional consideration is the use of different animal hosts for screening the same pool of STM mutants.
The inoculum size necessary to initiate an infection will determine the complexity of mutants pooled. In fact, each mutant in a defined input pool has to be in a sufficient cell number to initiate infection. The inoculum size must not be too high, resulting in the growth of mutants which would otherwise have not been detected. At higher doses, the immune system may be overwhelmed and the animals die of shock (6). Other important parameters in STM include the route of inoculation and the time-course of a particular infection. Also, certain gene products important directly or indirectly for initiation or maintenance of the infection may be niche-dependent or expressed specifically in certain animal or plant tissues only. If the duration of the infection in STM in vivo selection is short, genes important for establishment of the infection will be found, and if the duration is long, genes important for maintenance of infection will be identified (6). As STM is used and parameters are better defined in different models of infection, several routes of inoculation and different animal models can be used for the same organism studied by STM. We present below the protocol that we have used with P. aeruginosa.
Protocol 5: Chronic infection in the rat lung
For P. aeruginosa, we have used the chronic lung infection in a rat model which was adapted for this work (18). Female Sprague-Dawley rats of 140 to 160 g in weight were used.Isofluorane anesthetized rats were inoculated into the left lobe of lungs with 100 µl of a suspension of agar beads containing 106 bacterial cells (the in vitro pool). After 7 days, rats were sacrificed and lungs were removed and homogenized tissues were plated on BHIA supplemented with chloramphenicol. A concentration of 104colonies was recovered after in vivo selection and was used for colony PCR (the in vivo pool) as described previously (6). Ninety-six pools of 72 mutants forming a collection of 6912 mutants were maintained in the rat animal model causing a chronic lung infection. We used an infecting dose of 106 bacteria per animal to ensure an initial inoculum of 105. After 7 days of infection, the lungs were isolated and an average of 106 bacteria was recovered from the organ of each animal. To identify mutants not recovered after the in vivo passage, screening was done by PCR using bacterial colonies. Mutants which gave no amplification products by multiplex PCR after the in vivo selection were retested by single PCR. Colony PCR amplification products obtained from the in vitro pool was compared to the in vivo pool. Mutants which gave positive results from the in vitro pool and absent from the in vivo pool were kept for further analysis (Fig. 3). From 6912 P. aeruginosa mutants tested, we identified 214 attenuated mutants whose tag did not give a PCR amplification product from the in vivo pool. Attenuated mutants were assessed for growth on minimal media (M9 plates); 5 auxotrophic mutants were identified.
Protocol 6: Screening of tagged mutants by multiplex PCR
Detection of mutants was done by doing multiplex colony PCR in 50µl reaction volumes containing:
1.10 µl of boiled bacterial colony in 100 µl of TE PCR (10 mM Tris pH 7,4; 0,1 mM EDTA); 5 µl of 10X HotStart Taq polymerase (Qiagen) reaction buffer containing 15 mM MgCl2; 200 µM of each dNTPs; 10 pmoles of one of 21-mers.
2. The 21-mers numbered 1 to 24 in Table 1 were used as a first primer in combination with 10 pmoles of pUTKana2, 10 pmoles of tetR1 primer and 10 pmoles of pUTgfpR2 primer. HotStart Taq polymerase 2.5 U (Qiagen, Mississauga, Ont., Canada) was used in each PCR.
3. Amplification conditions were: hot start 15 min. at 95°C, 2 cycles at 95°C for 1 min., 65 to 55°C for 1 min., and at 72°C for 1 min. followed by 10 cycles at 95°C for 1 min., 55°C for 1 min., 72°C for 1 min. (touchdown PCR) in a ICycler (BioRad).
4. Amplified products were analyzed in a 1% agarose gel, 1X Tris-borate EDTA buffer and stained in 0.5µg/ml ethidium bromide solution.
5. PCR amplification products of tags absent in the in vivo pool are compared with amplified products of tags present in the in vitro pool (Fig. 3).
6. Mutants that give PCR amplification products from in vitro pool and not from in vivo pool are purified and kept for further analysis.
7. To confirm STM mutants, the PCR reaction is repeated for each individual STM mutant giving a negative amplification product in the multiplex PCR screening step. These STM mutants are carefully identified and used for a second round of in vivo screening.
Second round of in vivo screening
We arrayed 14 new groups with the 214 mutants from the first screening and 29 P. aeruginosa STM mutantspreviously screened once to confirm the STM attenuated phenotype (19, 20). When necessary, we have completed a group with the wild-type strain PAO1 so as to always maintain 72 clones in each group. The in vivo selection and detection of mutants were done as described above. From the 214mutants initially identified, we retained 42 highly attenuated mutants whose tags did not give any PCR amplification product from the in vivo pool. These results showed that to identify and obtain the most significant and highly attenuated mutants, a second round of in vivo screening is a pre-requisite. The genomic DNA from these P. aruginosa STM mutants was isolated, digested with PstI and cloned into pTZ18R (Fig. 4)
Protocol 7: Cloning and analysis of disrupted STM genes mutants selected.
1. Chromosomal DNA from the STM mutants selected is prepared using the QIAGEN genomic DNA extraction kit as described in the manufacturer’s protocol.
2. Chromosomal DNA (1µg) is digested with a restriction endonuclease giving a large range of fragment sizes; in our case we utilized PstI with P. aeruginosa and cloned DNA fragments ranging in size from 1 Kb to 6 Kb.
3. Digested chromosomal DNAs are cloned into pTZ18R (Amersham Pharmacia Biotech) predigested with the corresponding restriction enzyme. Ligation reactions are done as follows: 1 µg of digested chromosomal DNA is mixed with 50 ng of digested pTZ18r in 20 µl of 1X T4 DNA ligase buffer with 40 units of T4 DNA ligase. Incubate overnight at 16°C.
4. DNA ligation products are purified using the microcon PCR (Millipore) as described by the manufacturer's instructions and the DNA is resuspended in 5 µl of H2O.
5. The 5 µl recombinant plasmid solution is used for electroporation into E. coli DH5α.
6. Bacterial clones are kept and purified for plasmid analysis.
7. Plasmid DNA is prepared with QIAGEN midi preparation kit as described by the manufacturer.
Characterization of mutant strains
When a potential STM mutant has been obtained, it is essential to rapidly confirm that the tagged mutant is the cause of attenuation, even after two rounds of in vivo screening. Our recent approach is to do this first, even before the actual DNA sequence flanking the insertion point is known. The degree of virulence attenuation is a pre-requisite and we use the more sensitive and increasingly popular competitive index (CI) tests (21). The analysis is done with an STM mutant strain combined as single or mixed infections with the wild-type parent strain to describe the kinetics of growth, and in certain cases to identify the time and body site where the virulence defect is apparent. In a mixed infection, the ability of the strains to initiate or colonize the host provides a measure of their relative virulence. The CI can then be defined as the output ratio of mutant to wild-type bacteria divided by the input ratio of mutant to wild-type bacteria. The CI is thus a quantitative value for the degree of attenuation of a mutant strain, with the CI of a wild-type strain versus a fully virulent derivative being approximately 1.0 (21).
Protocol 8: CI of P. aeruginosa STM mutants
The in vitro growth rates of the PAO1 and mutant strains were measured in the chronic rat lung model. Here, we present a variation of the CI, the growth index.
1. A collection of bacterial strains (PAO1 wild-type, PAO909, a Pur- mutant and STM mutants isolated in vivo) were pre-cultured overnight in 10 ml of BHI adding kanamycin in the case of the mutants.
2. These cultures were diluted 1:100 in fresh medium. At different time points, aliquots of cultures were diluted and plated on BHIA to determine colony-forming units (CFU) per ml. The growth indices are given by the ratio of growth rate during exponential growth of the mutant on the ratio of the wild-type strain.
3. For the in vivo growth rates, both strains were pre-cultured overnight in 10 ml of BHI with kanamycin in the case of the mutants. From the pre-culture, 1 ml was washed 3 times in PBS, and 100 µl of washed pre-cultures were used for preparation and inoculation of agar beads as described above.
4. Bacteria were recovered after one and sevendays post-infection from 2 animals for each point by direct plating of lung tissue homogenates on BHIA plates supplemented with chloramphenicol (5 µg/ml). Animals used as controls were injected with PAO909 strain (a purine auxotroph of PAO1) and PAO1.
5. Nine mutants were found to be attenuated and are listed in Table 2. Among the nine in vivo attenuated mutants, levels of growth rate varied from low to high as reflected by index values the selected mutants obtained (Table 2). The less attenuated mutant has an index of 0.6, taking into consideration that this mutant is slightly attenuated in vivo. Mutants that were considered to be clearly attenuated had a growth rate index of 0.001.
An additional confirmation that we use in combination with the CI is to show that the tagged mutation is the cause of attenuation of virulence is by complementation with a functional allele. In this case, the wild-type gene of the STM mutant identified is cloned as a PCR product and introduced in the STM mutant. In a first step, we use RT-PCR to demonstrate in vitro expression and proceed with a CI analysis of the wild-type, the STM mutant and the mutant strain expressing the complementing allele. The obvious result expected will show that the mutation and the virulence phenotype are linked. Gene analysis can indicate if the inactivated gene is part of an operon and that STM attenuation of virulence may be due to a polar effect. In this case, inactivation of the downstream gene(s) is essential to confirm their implication in the decrease of virulence by further testing with the CI tests.
To demonstrate allelic complementation of STM mutants, we will use the G18T12 mutant described in Tables 1and 2 (20). Mutant G18T2 corresponds to an insertion in an ORF (PA0158) encoding for a protein of 1016 residues and having 32% identity with a putative ORF VC1673 of Vibrio cholerae, 31% identity with AcrD of Rickettsia prowazekii and CzrA of P. aeruginosa and 45 % identity with a membrane protein which is a putative cation efflux protein (HP0969) of Helicobacter pylori also identified as Hef. The protein PA0158 has been classified as a probable multidrug efflux system component encoding for a probable RND efflux transporter. In H. pylori, hefF encodes an RND pump protein homolog. The role and function of RND multiple-drug efflux systems in intrinsic antibiotic resistance has been demonstrated in several human pathogens, including E. coli, Haemophilus influenzae, Neisseria gonorrhoeae, and P. aeruginosa.The corresponding mutant has been shown to be mildly attenuated with a GI value of 0.294 in Table 2. Several pseudomonads possess an intrinsic resistance to many front-line antibiotics, due mainly to its low outer membrane permeability and to the mechanism of active efflux of antibiotics. We have found by STM that this efflux system may be implicated in vivo; this system could presumably be active against the host defense system by pumping out toxic molecules.
As proof of concept, we have cloned, and expressed the modA gene, as determined by RT-PCR (Fig. 5) and shown complementation of ModA in vitro and in vivo using the mutant G13T12 (PA1863). The complemented strain G13T12/pMON-PA1863 was used and compared to the G13T12 strain. We observed that the complemented strain was able to grow 3.4 times faster in vivo than the strain containing the modA insertional mutation (GI of 0.24 compare to 0.07, respectively). These results showed that the complementation with modA gave a significant recovery to the wild-type phenotype. This indicated that the disruption of modA in the mutant G13T12 was responsible for the attenuated phenotype obtained by STM, and confirmed by the in vivo GI experiments. Indeed, insertional inactivation of modA gave a significant attenuation of virulence in vivo but attempts to restore to the wild-type phenotype the mutant G13T12 was not complete and partly successful (GI of 0.6 instead of 1.0).
Protocol 9: preparation of RNA for RT-PCR
Total RNA from P. aeruginosa wild type strain, the modA mutant and the complemented mutant were obtained by RNeasy Midi Kit (Qiagen) flowing recommended procedure by Qiagen. All preparations were done using Aerosol-Resistant Tips and glove under a fume hood.
Protocol 10: DNA removal from RNA preparation
The total RNA preparation was treated with DNaseI prior to RT-PCR analysis. Careful removal of any remaining DNA in the RNA preparation is crucial and appropriate controls should be done to confirm that no DNA is present.
1. The DNase procedure was done using RNase-Free DNase Set (Qiagen) prepared folllowing the recommended procedures by Qiagen. Solid DNase I (1500 Kunitz units) in 550 µl of the RNase-free water was prepared. The reagent is kept in aliquots of 10 µl in microtubes at -20°C, this stock is stable up to 9 months.
2. DNase digestion:
100 µl RNA (80 µg)
10.3 µl Buffer RDD
2.9 µl DNase I stock solution (7.8 Kunitz)
3. Incubation 1h at 37°C.
4. Heat inactivation of DNase I 15 min. at 75°C. Repeat this DNase I digestion if necessary (if the PCR control is still positive with DNA).
Protocol 11: PCR confirmation for absence of DNA in the RNA preparations
The absence of DNA in each RNA preparation was confirmed by PCR using the following protocol. PCR reactions were done with the primer that will be used in RT-PCR to amplify the gene modA (PA1863) used for complementation (RT1863FOR 5’- CCG ATC CAG GCC ATC GCC AAG -3’, RT1863REV 5’- CGA CAG AGC GAC GAA GCC CAG -3’) and primers used for controls from the ftsZ gene (ftsZ3 5’- CAT CGC ACA AAC GCG CGT CAT -3’, ftsZ4 5’- ACG CAG GAA CGC CGG GAT ATC -3’).
1. Mix 10 µl of RNA preparation (1.4 µg), 5 µl of 10x HotStartTaq PCR buffer (Qiagen) (1.5 mM MgCl2 ), 8 µl dNTP (200 µM), 2 µl of each primers (0.1 µM) (RT1863FOR-RT1863REV pairs or ftsZ3-ftsZ4 pairs), 0.5 µl HotStartTaq DNA polymerase (2.5 U) (Qiagen) and complete the final volume up to 50 µl with nanopure water.
2. Reactions were performed in 200 µl thin-wall microtube in the Icycler thermocycler (BioRad).
3. HotStartTaq DNA polymerase requires an activation step of 15 min. at 95°C.
4. Amplification was done using TouchDown cycles: 1 min. at 94°C denaturation, 1 min. at 70°C annealing, 1 min. at 72°C elongation, every 2 cycles annealing temperature decreased by 1°C, 22 cycles were done until annealing temperature reach 60°C.
5. Amplification continued for 10 cycles: 1 min. at 94°C denaturation, 1 min. at 60°C annealing, 1 min. at 72°C elongation. A final step of 7 min. at 72°C was done to complete elongation of PCR products.
6. DNA amplification products were visualized by agarose gel electrophoresis and positive reactions indicates the presence of DNA contamination. DNase I digestion is repeated and the preparation is tested by PCR.
Protocol 12: RT-PCR reactions
The RT-PCR was done on treated RNA from P. aeruginosa wild type strain, the mutant and the complemented mutant using internal primers from the modA sequence (RT1863FOR 5’- CCG ATC CAG GCC ATC GCC AAG -3’, RT1863REV 5’- CGA CAG AGC GAC GAA GCC CAG -3’). The expected result is to confirm transcription of modA for positive complementation as visualized in an agarose gel (Fig.5).
1. Mix 10 µl of the RNA preparation (1.4 µg), 10 µl of 5x OneStep RT-PCR buffer (Qiagen) (1.25 mM MgCl2 ), 2 µl dNTP (400 µM), 6 µl of each primers (0.6 µM) (RT1863FOR-RT1863REV), 2 µl OneStep RT-PCR Enzyme Mix (Qiagen) (containing Omniscript Reverse Transcriptase, Sensiscript Transcriptase and HotStartTaq DNA polymerase) and raise final volume up to 50 µl with RNase-free water water.
2. Reactions are performed in 200 µl thin-wall microtube in Icycler thermocycler (BioRad).
3. Reverse Transcription is done in the thermocycler 30 min. at 50°C.
4. HotStartTaq DNA polymerase requires an activation step of 15 min. at 95°C, Omniscript and Sensiscript Reverse Transcriptases were inactived at the same time.
5. Amplification was done using TouchDown cycles: 1 min. at 94°C denaturation, 1 min. at 70°C annealing, 1 min. at 72°C elongation, every 2 cycles annealing temperature decrease by 1°C, 22 cycles were done until annealing temperature reach 60°C.
6. Amplification was continued for 10 cycles at: 1 min. at 94°C denaturation, 1 min. at 60°C annealing, 1 min. at 72°C elongation. A final step 7 min. at 72°C was done to complete PCR product.
7. Amplifications were visualized on agarose gel electrophoresis and positive reaction indicated RNA transcription of modA.
Acknowledgements
We express our gratitude to R. W. Hancock, UBC, Vancouver, B.C. for his constant encouragement, support and suggestions. Work on STM in R.C.L.’s laboratory is funded by the Canadian Institute for Health Research as part of the CIHR Genomics Program, and by the Canadian Bacterial Diseases Network via the Canadian Centers of Excellence. R.C.L. is a Scholar of Exceptional Merit from Le Fonds de la Recherche en Santé du Québec.
References
1. M. Handfield and R.C. Levesque, FEMS Microbiol. Rev. 23, 69, 1999.
2. C. K. Stover, X. Q. Pham, A. L. Erwin, S. D. Mizoguchi, P. Warrener, M. J. Hickey, F. S. Brinkman, W. O. Hufnagle, D. J. Kowalik, M. Lagrou, R. L. Garber, L. Goltry, E. Tolentino, S. Westbrock-Wadman, Y. Yuan, L. L. Brody, S. N. Coulter, K. R. Folger, A. Kas, K. Larbig, R. Lim, K. Smith, D. Spencer, G. K. Wong, Z. Wu, and I. T. Paulsen.Nature 406, 959 (2000).
3. M. Hensel, J.E. Shea, C. Gleeson, M.D. Jones, E. Dalton and D.W. Holden, Science 269, 400 (1995).
4. J.-M. Mei, F. Nourbakhsh, C.W. Ford, and D. Holden, Mol. Microbiol. 26, 399(1997)
5. B. P. Cormack, N. Ghori, and S. Falkow, Science 285, 578 (1999).
M. Hensel, J. E. Shea, A. J. Baumler, C. Gleeson, F. Blattner, and D. W. Holden, J. Bacteriol.179,1105 (1997).
6. D.E. Lehoux, and R.C. Levesque, Curr. Op. Biotech. 11,434(2000).
7. N.L. Craig, R. Craigie, M. Gellert and A.M. Lambowitz, Mobile DNA II, ASM Press, Washington, D.C., 2002.
8. S.L. Chiang, and J.J. Mekalanos, Mol. Microbiol., 27 , 797 (1998).
9. D.E. Lehoux, F. Sanschagrin and R.C. Levesque, Biotechniques 26 , 473 (1999)
10. D.E. Lehoux, F.Sanschagrin, and R.C. Levesque, Curr.Op.Microbiol.4,515(2001).
11. V. De Lorenzo, M. Herrero, U. Jakubzik, and K.N. Timmis , J. Bacteriol. 172, 6568 (1990).
12. M. Herrero, V. de Lorenzo, and K.N. Timmis, J. Bacteriol. 172,6557 (1990).
13. A.G. Matthysse, S. Stretton, C. dandie, N. C. McClure, and A. Goodman, FEMS Microbiol. Lett. 145,87, (1996).
14. R. Fellay, J. Frey, and H. Krish. Gene 52, 147, (1987).
15. R. Simon, U. Priefer and A. Pühler, Bio/Technology 1, 784 (1983).
16. J. Sambrook, and D. Russel, Molecular Cloning: A Laboratory Manual, Cold Spring Harbor Laboratory, Cold Spring Harbor, New York, 2001.
17. F. J. DeBruijn, and S. Rossbach. Transposon mutagenesis, In: Methods for General and Molecular Bacteriolgy, (P. Gerhardt, R.G.E. Murray, and N. Krieg, editors), p. 387, ASM Press, Washigton, D.C. (1994).
18. H.A.Cash, D. E. Woods, B. McCullough, W. G. Johanson, Jr., and J. A. Bass.Am. Rev. Respir. Dis. 119, 453 (1979)
19. D.E. Lehoux, F. Sanschagrin, and R.C. Levesque, FEMS Microbiol. Lett. 210,73 (2002).
20. A.J. Darwin, and V.L. Miller, Mol. Microbiol., 32, 51 (1999).
21. K.E. Unsworth, and D. Holden, Phil. Trans. R. Soc. Lond. B 355 , 613 (2000).
Each 21-mers has a Tm of 64°C and permits PCR amplification in one step when the 3 primer combinations are used for multiplex screening. Two sets of consensus 5’-ends comprising the first 13 nucleotides have higher Gs for optimizing PCR. Twelve variable 3’-ends define tag specificity and allow amplification of specific DNA fragments. The set of 24 21-mers representing the complementary DNA strand in each tag are not represented and can be deduced from the sequences present.
The disrupted open reading frames (ORF) in each STM mutant were: PAO090 (G20T2); PA1863 (G13T12); Pa5441 (G19T12); PAO082 (G10T7); PA4491 (G30T12); PA4115 (G56T2); PAO0158(G18T2); PA0073 (G94T2); PA1596(G38T4). Identity values were obtained by comparing the amino acid sequences deduced from the complete ORF.
a The in vitro growth index (GI) is the ratio of the growth rate of the mutant compared to that of the wild-type strain PAO1 in rich BHI broth.
b The relative growth index (GI) is the in vivo growth index divided by the in vitro growth index. PAO909 is a purine auxotrophic strain used as a negative control.
Figure Legends
Fig.1A.General strategy for the construction of the arrayed libraries tagged with miniTn5Km2, miniTn5Tc and miniTn5Tcgfp. In a defined library, each mutant has the same tag but inserted at different locations in the bacterial chromosome. One mutant from each library is picked to form 96 pools of 24 mutants with unique tag for each miniTn5.
Fig.1B. Comparative analysis between the in vitro and in vivo pools using multiplex PCR. An aliquot is kept as the in vitro pool, and a second aliquot from the same preparation is used for passage into an animal model for in vivo negative selection. After determined time points of infection, bacteria are recovered from animal organs and constitute the in vivo pool. The in vitro and in vivo pools of bacteria are used to prepare DNA in 24 PCR reactions using the 21-mers 1 to 24 in table 1 and the Km, Tc and gfp primers.
Fig.2. Physical and genetic maps of the miniTn5Km2, miniTn5Tet and miniTn5Tetgfp transposons used. The elements are represented by thick black lines, inverted repeats are indicated as vertical boxes and genes are indicated by arrows. Abbreviations: I and O inverted repeat ends; Km, kanamycin; Tc, tetracycline; gfp, green fluorescent protein.
Fig.3. Agarose gel electrophoresis separations of multiplex PCR DNA amplified products obtained from the in vivo pool of 72 tagged-transposon mutants. Lane 1 to 24, 10 µl of a multiplex PCR reaction using pUTKana2, tetR1 and pUTgfpR2 primers with Tag 1 to 24 respectively. The lane indicated as – is the negative control. The multiplex PCR conditions are described in the text. The amplified PCR products from tagged mini-Tn5 Km2, gfp and Tc are 220, 820 and 980 bps respectively. Note the absence of amplification in lanes 5,7,15, 24 where one or two DNA bands are not visible. Rigorous analysis by PCR-based STM implies that the weak signals in lanes 2, 9,10,11,15, would be considered as positive amplification and would not be tested further. The clones in lanes 5, 7, 15 and 24 would be further tested by individual PCR and an additional round of in vivo STM screening. In this specific example, no STM miniTn5Km mutants were obtained.
Fig.4. General scheme for cloning genomic DNA encoding the inactivated gene presumably implicated in attenuation of virulence via selection of the antibiotic resistance genes inserted in STM mutants. The genomic DNA of P. aeruginosa is digested completely with PstI and cloned DNA fragments are selected on the basis of encoding either Tc or Km resistance. The cloned genomic DNA is sequenced using primers annealing in the I end of each transposon DNA fragment (I-End sequencing primer: 5’- AGA TCT GAT CAA GAG ACA G-3’).
Fig. 5 Agarose gel electrophoresis of RT-PCR products confirming expression of modA from the complementing allele. Lanes: 1, wild-type; 2, modA STM mutant complemented; 3, STM mutant. The intense 500 bps band is due to high level expression from the lacZ constitutive promoter in pTZ18R.
© Éric Potvin, 2007
| | | |
| Bibliographie |  | Appendice 2 |
