This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrowReprints and Permissions
Right arrow Copyright Information
Right arrow Books from ASM Press
Right arrow MicrobeWorld
Citing Articles
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Liautard, J.
Right arrow Articles by Liautard, J.-P.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Liautard, J.
Right arrow Articles by Liautard, J.-P.

 Previous Article  |  Next Article 

Infection and Immunity, November 2007, p. 5167-5174, Vol. 75, No. 11
0019-9567/07/$08.00+0     doi:10.1128/IAI.00690-07
Copyright © 2007, American Society for Microbiology. All Rights Reserved.

Identification and Isolation of Brucella suis Virulence Genes Involved in Resistance to the Human Innate Immune System{triangledown}

Janny Liautard, Safia Ouahrani-Bettache, Véronique Jubier-Maurin, Virginie Lafont, Stephan Köhler, and Jean-Pierre Liautard*

CNRS-UMR 5236 and Université Montpellier 2, place Eugène Bataillon, 34000 Montpellier, France

Received 22 May 2007/ Returned for modification 21 June 2007/ Accepted 9 August 2007


arrow
ABSTRACT
 
Brucella strains are facultative intracellular pathogens that induce chronic diseases in humans and animals. This observation implies that Brucella subverts innate and specific immune responses of the host to develop its full virulence. Deciphering the genes involved in the subversion of the immune system is of primary importance for understanding the virulence of the bacteria, for understanding the pathogenic consequences of infection, and for designing an efficient vaccine. We have developed an in vitro system involving human macrophages infected by Brucella suis and activated syngeneic {gamma}9{delta}2 T lymphocytes. Under these conditions, multiplication of B. suis inside macrophages is only slightly reduced. To identify the genes responsible for this reduced sensitivity, we screened a library of 2,000 clones of transposon-mutated B. suis. For rapid and quantitative analysis of the multiplication of the bacteria, we describe a simple method based on Alamar blue reduction, which is compatible with screening a large library. By comparing multiplication inside macrophages alone and multiplication inside macrophages with activated {gamma}9{delta}2 T cells, we identified four genes of B. suis that were necessary to resist to the action of the {gamma}9{delta}2 T cells. The putative functions of these genes are discussed in order to propose possible explanations for understanding their exact role in the subversion of innate immunity.


arrow
INTRODUCTION
 
Bacteria are a major cause of human diseases and a formidable foe of the mammalian immune system. Bacteria that cause chronic infections are among the most immune system-demanding human pathogens; they avoid or subvert immune defenses, and, as a consequence, vaccination is often inefficient. Yet the immune response is a very complex process in which there are numerous possibilities for countering pathogen actions, and a successful pathogen needs to use many different targets in order to efficiently subvert the host response (26).

Virulence of a microorganism is defined as "the relative capacity of a pathogen to overcome the body defenses." Four steps can be identified in the process by which an intracellular bacterium causes a persistent intracellular infection. The first step is invasion of the host organism, the second step is establishment and multiplication inside host cells (epithelial or macrophages), the third step is resistance to the innate immunity, and the fourth step is overcoming the specific immunity. Most of the virulence genes uncovered by systematic screening are involved in resistance to the first two steps (i.e., invasion of the host and multiplication inside the host cell) because intracellular multiplication has been used very often as a test for screening methods. On the other hand, animal models have been developed with methods such as signature-tagged mutagenesis (21). However, when animals are employed for virulence screening of intracellular bacteria, most of the attenuated clones identified are impaired in the first steps of the infection because of the greater sensitivity of bacterial multiplication to these steps. Furthermore, due to the species specificity of most of intracellular bacteria, the genes involved in subverting the human immune system cannot be analyzed with animal models. Thus, specifically designed screening methods have to be developed to uncover bacterial genes involved in subverting the human immune response.

Brucella spp. cause chronic infections in humans and animals. We took advantage of a particular trait of human infection by these intracellular bacteria to design a specific approach for uncovering virulence genes involved in immune system subversion. Brucella strains that multiply inside host cells rapidly induce a dramatic increase in the number of a specific type of circulating lymphocytes, the V{gamma}9V{delta}2 T lymphocytes ({gamma}9{delta}2 T lymphocytes) (4). This phenomenon is not restricted to Brucella as it is the hallmark of most intracellular infections, suggesting that these immune cells have a specific role in controlling pathogen invasion. We thus decided to investigate the bacterial mechanisms that overcome this specific innate immune response.

We have developed an in vitro coculture system using infected human macrophages with syngeneic {gamma}9{delta}2 T lymphocytes. These lymphocytes can be isolated from human buffy coat and activated in vitro by a nonpeptidic antigen such as 4-hydroxy-3-methyl-but-2-enyl pyrophosphate (HMBPP), which is a common metabolite of bacteria (38). We observed that these lymphocytes reduced the intramacrophagic multiplication of B. suis, but with poor efficiency (35). We reasoned that the wild-type strains of B. suis probably either impaired the recognition of the infected macrophages by lymphocytes or induced resistance to their action, resulting in escape from this immune response. Thus, it should be possible to identify genes involved in this breakout by screening a library of transposon mutants.

In previous work, we identified the majority of the virulence genes involved in intramacrophagic multiplication by screening a library of more than 10,000 clones of B. suis (27). We reasoned that the genes involved in {gamma}9{delta}2 T-cell subversion should be among the genes that do not affect the multiplication in macrophages but that mutation of these genes results in increased sensitivity to the action of these lymphocytes. As a consequence, multiplication of B. suis in infected macrophages should be impaired by syngeneic {gamma}9{delta}2T lymphocytes previously activated by HMBPP. However, counting the bacteria in a well is time-consuming and is almost impossible when more than 1,000 clones have to be analyzed. To overcome this difficulty, we developed a new method to rapidly estimate the number of living bacteria inside macrophages after the action of lymphocytes. We took advantage of the fact that Alamar blue changes color when it is metabolized by bacteria. We show here that it is very easy to rapidly discriminate between bacteria whose multiplication is affected by more than 1 log compared to the wild type. This 1-log difference in multiplication inside macrophages is considered the hallmark of attenuation, thus identifying virulence genes. In the present paper, we report the results obtained when we screened 2,000 clones of B. suis. The importance of genes identified was individually analyzed, the role of these genes in virulence was confirmed, and the relationship between putative function and subversion of {gamma}9{delta}2 T lymphocytes is discussed below.


arrow
MATERIALS AND METHODS
 
Bacterial strains and media. B. suis 1330 (= ATCC 23444) and all derived mini-Tn5 mutant strains were grown in tryptic soy broth (TS broth) at 37°C. The Escherichia coli donor strain used for conjugation, SM10 {Delta}pir containing a tagged mini-Tn5 Km2 transposon in pUT (21, 27), was cultivated in LB broth. Kanamycin was used at a concentration of 50 µg/ml. Alamar blue was purchased from CliniScience, France.

Library design. Transposon mutagenesis is a well-established method to target genes, provided that enough individual clones are analyzed. B. suis was the acceptor strain for the mini-Tn5 Km2 transposon (29). The library used was the library described previously (27).

Cell preparations. Peripheral blood mononuclear cells (PBMC) from healthy donors were prepared by density centrifugation on Ficoll-Paque (Eurobio, Les Ulis, France). Monocytes were purified from PBMC based on their adherence properties. This preparation method provided blood monocytes with a purity of 95%. Autologous {gamma}9{delta}2 T cells were purified from nonadherent PBMC using anti-{gamma}9 monoclonal antibodies and goat anti-mouse immunoglobulin G-coated Dynal magnetic beads (Dynal, Compiègne, France). After one night, the coated cells separated spontaneously from the magnetic beads and then were stimulated with 10 nM HMBPP in a 24-well culture plate in the presence of autologous monocytes (2 x 106 cells/ml) and recombinant human interleukin-2 (rhIL-2) (150 U/ml). After 3 weeks of expansion culture in complete medium (RPMI 1640) containing 5% fetal calf serum, 5% human AB serum, and rhIL-2 (150 U/ml), the expanded g9d2 T cells were stimulated again at a concentration of 2 x 106 to 3 x 106 cells/ml in complete culture medium without rhIL-2.

Infection. Purified monocytes were seeded in 48-well plastic plates at a density of 5 x 105 cells/ml in complete culture medium supplemented with 10–7 M 1,25-dihydroxyvitamin D3 (a generous gift from Hoffman La Roche, Basel, Switzerland) for 5 days. 1,25-Dihydroxyvitamin D3-differentiated monocytes display macrophagic cell properties and can be readily infected with Brucella (6). These cells are referred to below as macrophages. The culture medium was then removed, and the adherent cells were washed twice and infected with the B. suis wild type or Tn5 mutants at a multiplicity of infection of 30 in 100 µl complete medium without gentamicin. After a 1-h infection, the cells were washed with phosphate-buffered saline (PBS) containing gentamicin and incubated in 500 µl of complete medium alone or complete medium containing 2 x 106 to 3 x 106 {gamma}9{delta}2 T lymphocytes/ml.

Infection of activated macrophages. Human macrophages were purified as described above. A total of 5 x 105 macrophages were seeded into 48-well plates in 500 µl RPMI with 10% fetal calf serum and 10–7 M vitamin D3. Gamma interferon (IFN-{gamma}) (250 U/ml) was added for 16 h before infection. The macrophages were infected as described above, except that the medium contained 500 pg/ml tumor necrosis factor alpha (TNF-{alpha}) and 250 U/ml IFN-{gamma}.

Assessment of the number of intracellular living bacteria. We designed a new protocol to rapidly assess the number of intracellular living bacteria in a well of a 96-well plate (Fig. 1). We took advantage of the change in color of Alamar blue induced by living bacteria. Supernatants (200 µl) in the wells containing the cultures were removed with a multichannel pipette and then the wells were washed carefully with 180 µl of PBS. Most of the lymphocytes were removed, while macrophages adhered to the bottom of the wells. Twenty microliters of Triton X-100 (0.1%) in water was added to each well and rapidly diluted with 170 µl of TS broth and 10 µl of Alamar blue (final concentration, 5%). The plates were incubated at 37°C for at least 15 h in a microbiologic incubator. The change in color was monitored each hour until the wild-type control turned completely pink. Photos were taken with a simple Nikon Coolpix 4300 digital camera (see Fig. 4).


Figure 1
View larger version (28K):
[in this window]
[in a new window]

  		FIG. 1. Schematic diagram of the screening system developed. M.O.I., multiplicity of infection.


Figure 4
View larger version (106K):
[in this window]
[in a new window]

  		FIG. 4. Example of a color plate. The plate was seeded with B. suis wild type and mutants and incubated as described in Materials and Methods. The plate was divided into two parts: a side with only infected macrophages (left) and a side with infected macrophages incubated with activated {gamma}9{delta}2 T lymphocytes (right). The picture was taken when all the wild-type wells had completely changed color. The mutants attenuated in macrophages alone are indicated by red circles (a, 3H10; b, 6E12; c, rsh control mutant; d, 36A6; e, recA control mutant), and the mutants attenuated only with {gamma}9{delta}2 T lymphocytes are indicated by green circles (1, 25G5; 2, 14F3; 3, 23C11; 4, 31B1). Two false positives in the first round are indicated by blue circles.

The time necessary to obtain a full color change (i.e., pink color) was proportional to the quantity of living bacteria, and a 1-log increase resulted in a reduction of at least 1 h for the change in color (see above).

Precise counts of bacteria were obtained by CFU plating after cell lysis with 0.2% Triton X-100 as previously described (34). The quantity of Brucella was assessed using a standard curve. The action of {gamma}9{delta}2 T cells (36) on infected macrophages resulted in a 1-log reduction in wild-type Brucella multiplication at 48 h. We considered the 1-log reduction a threshold for detecting a mutation involved in protection. Serial 10-fold dilutions of lysates were plated on TS agar, and CFU were counted after 48 h of incubation at 37°C.

Analysis of transposon insertion sites. Chromosomal DNA was isolated from the attenuated mutants as described previously (27), and the sequences of the regions flanking the transposon insertion sites were obtained by a reverse PCR method (22). Thirty micrograms of chromosomal DNA was digested by NcoI (0.75 U/µl) for 2 h at 37°C. After denaturation of the restriction enzyme, the digest was ligated using 1 U/µl ofT4 DNA ligase overnight at 16°C. DNA was precipitated with 2 volumes of ethanol in the presence of sodium acetate. Inverse PCR was performed with the following primers: PS2 (5'-TGC GCT GCC CGG ATT ACAG-3') and PA2 (5'-ATG AAT GTT CCG TTG CGG TG-3'). Thirty cycles of 94°C for 5 min, 94°C for 30 s, 63°C for 1 min, 72°C for 8 min, and 72°C for 10 min were performed with GoldStar DNA polymerase (5 U/µl). A nested PCR was performed using primers PS1 (5'-CGA CCT GCA GGC ATG CAA G-3') and PA1 (5'-GTA CCG AGC TCG AAT TCG GC-3') and 29 cycles identical to those used for the inverse PCR. Automated sequence analysis (Genome Express, Toulouse, France) was performed using primers P6 and P7 (21). Sequence database searches were performed by using the BLASTN and BLASTX algorithms and by comparison with the annotation for the genome of B. suis at the TIGR database (http:www.tigrblast.tigr.org).

Analysis of lipopolysaccharide (LPS) by Western blotting. Fifty microliters of bacteria grown in rich medium with appropriate antibiotics was centrifuged, and the pellet was washed with PBS. Whole-cell lysates of the different clones were prepared by boiling preparations in sample buffer (2% sodium dodecyl sulfate, 62.5 mM Tris-HCl [pH 6.8], 2.5% mercaptoethanol, 10% glycerol) for sodium dodecyl sulfate-polyacrylamide gel electrophoresis. After electrophoresis, the gels were transferred onto nylon membranes and were probed with an anti-O-chain mouse monoclonal antibody (04/F09), a generous gift from A. Clockaert (5), using anti-mouse peroxidase-labeled antibodies for detection.

Sensitivity to polymyxin B. Bacteria grown overnight in TS medium were centrifuged and suspended in TpB buffer (0.133 M NaCl, 0.1 M NaH2PO4; pH 4.5) at a concentration of 6 x 103 bacteria/ml. Polymyxin B (Sigma) was diluted in the same buffer to obtain the desired concentration. Mixtures consisting of 100 µl of bacteria and 100-µl portions of polymyxin B at different concentrations were incubated for 1 h at 37°C. The number of viable bacteria was determined by counting on TS agar plates plus antibiotics.

Statistical analysis. The mean of triplicate samples is shown below for each data point along with the standard deviation and is representative of a minimum of three experiments performed with separate human blood donors. The P value was calculated by using an paired Student t test.


arrow
RESULTS
 
Intramacrophagic multiplication of wild-type B. suis is only slightly reduced by activated {gamma}9{delta}2 T lymphocytes. In humans, the {gamma}9{delta}2 T-lymphocyte population increases dramatically during infection by Brucella (4). We have demonstrated that Brucella produces nonpeptidic, small molecules that cause proliferation of these cells (36). The same type of results has been obtained for many intracellular bacteria (38). We have designed an in vitro system comprising Brucella-infected macrophages and syngeneic activated {gamma}9{delta}2 T lymphocytes (35) to analyze the effect of these cells on the intracellular multiplication of Brucella. Although {gamma}9{delta}2 T lymphocytes interacted with infected macrophages (Fig. 2A), only a slight decrease in the intracellular multiplication of the bacteria was observed (Fig. 2B). The number of intracellular B. suis bacteria at 48 h postinfection decreased (by approximately 1 log), but the bacteria were still capable of multiplication, implying that most of them were able to overcome the activation of macrophages by {gamma}9{delta}2 T cells and resist the bactericidal effect of these lymphocytes. This means that the {gamma}9{delta}2 T cells, although activated, could not impair the invasion. We reasoned that the capacity of Brucella to overcome elimination by {gamma}9{delta}2 T cells is probably associated with bacterial genes specifically designed for this purpose, and thus in bacteria that do not multiply in macrophages surrounded by {gamma}9{delta}2 T cells genes involved in subverting the innate immune system are affected. We thus designed a screening method to identify these genes. A diagram of this method is shown in Fig. 1.


Figure 2
View larger version (16K):
[in this window]
[in a new window]

  		FIG. 2. Human {gamma}9{delta}2 T lymphocytes recognize infected macrophages but are not able to eliminate wild-type B. suis. Infected human macrophages were incubated with activated syngeneic {gamma}9{delta}2 T lymphocytes as described in Materials and Methods. (A) Microscopic image showing that {gamma}9{delta}2 T lymphocytes interact with macrophages. (B) Determination of the number of living B. suis bacteria inside macrophages showed that multiplication was hardly affected by incubation with activated {gamma}9{delta}2T lymphocytes. •, control without addition of {gamma}9{delta}2 T lymphocytes; {triangleup}, multiplication of wild-type B. suis in the presence of activated {gamma}9{delta}2 T lymphocytes. WT, wild type.

Semiquantitative analysis of the number of bacteria using Alamar blue. To assess bacteria whose multiplication in vitro is defective, intracellular multiplication has to be quantified. This task is generally performed by counting CFU after cell lysis with 0.2% Triton X-100. However, when applied to a massive screening, this protocol is unwieldy and time-consuming. We have designed a new protocol to rapidly obtain a rough estimate of the number of living bacteria inside a well (Fig. 1). We took advantage of the change in color of Alamar blue induced by living bacteria. The time necessary to obtain full transformation and thus color change is proportional to the quantity of living bacteria at the beginning of incubation (Fig. 3). It should be noted that a 1-log difference in the number of bacteria in a well is perfectly shown by the change in color. This is the threshold used in this work to define an attenuated mutant. The results are reproducible and comparable when experiments are performed in 96-well plates.


Figure 3
View larger version (8K):
[in this window]
[in a new window]

  		FIG. 3. Color change of Alamar blue is proportional to the input number of bacteria. A known number of B. suis bacteria was incubated in TS medium containing Alamar blue as described in Materials and Methods. The color was visually monitored as a function of time by taking pictures and comparing them. A linear relationship was clearly demonstrated. {circ}, beginning of the color change; •, complete color change. A 1-log difference in the inoculum resulted in clearly distinguishable colors.

Screening of the library. Macrophages obtained from healthy donors were infected by individual B. suis mutants (27) and the acceptor strain as a control at a multiplicity of infection of 30 in 96-well plates (11). The macrophages were incubated for 45 min; the excess bacteria were washed off, and activated syngeneic {gamma}9{delta}2 T cells were added in the presence of 30 µg/ml gentamicin. After 24 h of incubation at 37°C, the cells were washed, lysed by Triton X-100, and incubated with Alamar blue. The change in color was recorded each hour until the wild-type controls became completely pink. Wells containing attenuated mutants remained blue, allowing rapid visual identification (Fig. 4). The clones that appeared to be attenuated were subjected to another round of screening using the same conditions. Under the conditions used, approximately one-half of the clones were found not to be positive in the second round. It is possible that in some wells, the washing procedure was too drastic. Alternatively, in some wells, mainly on the side of the plate, macrophages did not adhere correctly, and the number of these cells was reduced during washing. Starting with 2,000 individual clones, which corresponded to more than one-third of the possible genes, we identified six mutants that were retained for further analysis.

Individual confirmation of attenuation. The clones found to be positive after the second round were tested individually. A kinetic analysis of the multiplication of the bacteria was performed. An example of the results is presented in Fig. 5. The clone studied multiplied normally in macrophages alone, compared with wild-type B. suis. The difference observed at 48 h is not statistically significant. The multiplication of the wild type was reduced by 1 log by coculture with activated {gamma}9{delta}2 T lymphocytes, as generally observed (35); on the other hand, the level of the mutant was reduced by more than 2 logs by the action of {gamma}9{delta}2 T lymphocytes. This result confirmed the greater sensitivity of this mutant to the action of {gamma}9{delta}2 T lymphocytes. All the mutants identified were analyzed in the same way; three clones (14F3, 16E7, and 25G5) were retained, and they are described in Table 1. Three clones (19A4, 6E12, and 3H10) were found to be partially attenuated in macrophages alone and thus were not included. These clones were not detected by screening performed previously with THP-1 cells (27).


Figure 5
View larger version (14K):
[in this window]
[in a new window]

  		FIG. 5. Time dependence of the intramacrophagic multiplication of B. suis in the presence or absence of activated {gamma}9{delta}2 T lymphocytes. Infection with the wild type (WT) and the 23C11 mutant was performed as described in Materials and Methods.


View this table:
[in this window]
[in a new window]

  		TABLE 1. B. suis genes involved in resistance to {gamma}9{delta}9 T cells

Sequence analysis of the clones obtained. In order to better characterize the clones obtained, we performed a sequence analysis as detailed in Materials and Methods. The genes disrupted by the mini-Tn5 transposon were identified by BLAST analysis using the B. suis sequence published by Essenberg et al. (12). All the sequences obtained allowed attribution to a gene with close to 100% identity and thus no ambiguity. The results are summarized in Table 1. The designations of the genes were those proposed in the B. suis database annotated by TIGR (http://www.tigr.org/). Genes with very different putative functions were involved in resistance to human {gamma}9{delta}2 T lymphocytes. In order to confirm that genes disrupted by the transposon were involved in this phenomenon, we tested clone 23C11 of the library, in which norD was disrupted (31). Clone 23C11 was as sensitive to the action of {gamma}9{delta}2 T lymphocytes as 14F3 (Table 1).

Phenotypic characterization of the mutants (polymyxin and LPS). In order to confirm that the resistance observed did not result from a nonspecific change in LPS composition, we performed a phenotypic analysis of the mutants. It has been observed previously that certain rough mutants of Brucella spp. are not attenuated in macrophages but exhibit a drastic reduction in multiplication in mice (17). One of the different hypotheses that can be put forward to explain this observation is sensitivity to the immune system. This could explain the results that we obtained. We therefore analyzed by Western blotting the structure of the LPS. The results are presented Fig. 6. They clearly demonstrated that all the clones selected have a complete O chain and thus are smooth, eliminating the possibility that rough mutants could explain the results obtained. However, LPS of pathogenic bacteria and particularly of Brucella spp. are very complex structures that are not totally revealed by gel electrophoresis. Wild-type Brucella spp. strains are resistant to polymyxin, but some modifications of the LPS are associated with sensitivity to this compound (15). Thus, we decided to test the resistance to polymyxin of the mutants. The results are presented in Fig. 7. Clearly, 25G5 is very sensitive to low concentrations of polymyxin. Sensitivity to polymyxin could result either from alteration of the LPS structure or from destabilization of the membrane due to modification of a membrane protein. We have tested two mutants with deletions of the genes encoding outer membrane proteins Omp25 and Omp31 (described previously [25]). Only the Omp31 mutant (clone 31B1) was sensitive to polymyxin (Fig. 7). This mutant was also sensitive to {gamma}9{delta}2 T-lymphocyte action (Table 1).


Figure 6
View larger version (40K):
[in this window]
[in a new window]

  		FIG. 6. All the mutants selected have an LPS with a complete O chain. The different mutants and the {Delta}manB and wild-type (WT) controls were grown in TS medium for 24 h. The LPS was prepared and analyzed by electrophoresis as described in Materials and Methods. Lane MW contained molecular weight markers.


Figure 7
View larger version (15K):
[in this window]
[in a new window]

  		FIG. 7. Sensitivity of the B. suis mutants to polymyxin B. Bacteria were incubated for 1 h with different concentrations of polymyxin B and counted as described in Materials and Methods. Ten measurements were obtained for each point. Student t tests were performed with the raw results before normalization in order to avoid possible bias due to deviation to the normal distribution caused by normalization. Three asterisks indicate that the P value is <0.001. WT, wild type.

Individual sensitivity to macrophage activation. We previously showed that {gamma}9{delta}2 T cells produced large quantities of cytokines when they were activated by nonpeptidic antigen or by direct contact with bacteria (35, 36). Thus, these cytokines should activate macrophages, and it is known that IFN-{gamma} activates macrophages to limit Brucella multiplication (33); this has also been shown for TNF-{alpha} (7). We decided to examine if the effect observed was due to increased sensitivity of the bacteria to the activation of macrophages by IFN-{gamma} and TNF-{alpha}. The macrophages were activated by these two cytokines as described in Materials and Methods, and the impact on intracellular multiplication was monitored. One mutant, 25G5 (pgm), appeared to be sensitive to the activation (Table 1). On the other hand, three other clones, 14F3 (norD), 23C11 (norD), and 16E7 (atpG), resisted the macrophage activation.


arrow
DISCUSSION
 
In the present work we describe identification of four genes of B. suis that are necessary to resist specifically the action of {gamma}9{delta}2 T cells. The results confirm the hypothesis that bacteria, such as B. suis, that induce chronic diseases in humans have developed specific strategies to subvert the immune system at the level of the innate secondary response (i.e., the innate response mediated by lymphocytes that respond to pathogen-associated molecular patterns). Because the immune system and bacterial pathogens have evolved together, a complex mechanism of action and reaction takes place during infection. Many bacterial genes are probably involved in this kind of cross talk; identification of four genes suggests that 10 to 30 genes are probably necessary to avoid elimination by {gamma}9{delta}2 T lymphocytes. Subversion of the immune system by bacteria was generally examined at the level of macrophages (9) or complement (30). The roles of lymphocytes involved in innate immunity, such as {gamma}9{delta}2 T lymphocytes, NKT cells, or NK cells, have never been studied in detail. We describe here the first screening to uncover bacterial genes involved in subversion of one of these types of lymphocytes by pathogenic bacteria. It would be interesting to know if these genes are also involved in the resistance to NK or NKT cells or if there is some specificity.

Physiological relevance of the gene functions identified. The genes identified by this screening method encode very different functions, and each of them seems to play a specific role in the metabolism of the bacteria. The reason why mutants are attenuated is not straightforward, and no genes with unknown functions have been sorted out. Thus, a more in-depth analysis should be performed to understand why the genes identified are involved in subversion of the innate immune system. Such experimental work is under way. However, theoretical analysis has revealed some particular properties of the genes retained that could be helpful for conducting the research.

Three different hypotheses may explain the hypersensitivity of the mutants to the action of {gamma}9{delta}2 T cells. The first hypothesis involves the hypersensitivity of the mutants to the bactericidal activity of the lymphocytes (for instance, the action of antimicrobial peptides) (10). This possibility was tested individually with each mutant by assessing the mutant's sensitivity to polymyxin, an antibiotic that is inefficient against wild-type B. suis (a selective medium to identify Brucella spp. includes polymyxin) and whose activity reveals membrane alteration. A correlation with sensitivity to antibacterial peptides is generally observed (14). Furthermore, a screening procedure using sensitivity to polymyxin has revealed that a mutant with a mutation in the gene coding for the BvrR/S two-component system regulating the modification of membrane structure during infection is attenuated due to sensitivity to macrophage bactericidal activity (20). Two of the mutants exhibited notable sensitivity to polymyxin, suggesting that there was increased sensitivity either to the activated macrophage or to the direct action of the {gamma}9{delta}2 T lymphocytes. In one of the mutants (25G5), the transposon is inserted in a putative pgm gene (BRA 0348) that is probably in an operon with a downstream putative pmm (phosphomanomutase) gene and probably results in knockout of these two genes. However, B. suis possesses two genuine genes to perform these functions, a phosphoglucomutase gene (BR0058; http://helab.bioinformatics.med.umich.edu/) and a phosphomanomutase gene (BR0537). Inactivation of one of these two genes results in rough mutants (1, 13, 40), which is compatible with the phenotype that has been observed. These specific rough mutants are not attenuated in macrophages but are very sensitive in a mouse model of infection (1, 39). On the other hand, the 25G5 clone has a complete O chain and a smooth phenotype. Like many of the rough mutants, it is not attenuated in macrophages but is sensitive to the action of {gamma}9{delta}2 T cells; this phenomenon could be due to greater sensitivity to bactericidal peptides produced by activated macrophages or by the lymphocytes (34). This proposition is reinforced by the finding that there is greater sensitivity to macrophages activated by IFN-{gamma} and TNF-{alpha} (Table 1). Why is this mutant with a complete O chain more sensitive than the wild type to polymyxin? Two hypotheses can be proposed to explain this phenomenon. First, the pgm gene (BRA0348) is used by the bacteria only to modify the LPS, perhaps at the lipid A level, and it has been shown that modification of lipid A correlates with virulence (32). The second hypothesis takes into account the ß(1-2)-glucans synthesized by Brucella spp., which have been shown to be necessary for virulence (2). ß(1-2)-Glucan is synthesized from UDP-glucan, which requires an active pgm gene (12). It can be hypothesized that ß(1-2)-glucan is synthesized from one pgm gene (BRA0348), while the LPS depends on the other pgm gene (BR0058). This hypothesis is supported by the observation that BLAST analysis of organisms outside the genus Brucella revealed that the highest level of homology is with noeK from Rhizobium etli, a gene necessary for nodulation that is involved in the synthesis of carbohydrate morphogens (16). Only a complete study of this mutant will give a definitive answer.

The second mutant that exhibits polymyxin sensitivity is 31B1, which has an impaired BR1622 gene that codes for membrane protein Omp31. It is well known that this protein is tightly to bound to the LPS structure, and this characteristic probably results in stabilization of the outer membrane of the bacteria (8). Sensitivity to antimicrobial peptides has also been observed in BvrR/S mutants in which certain outer membrane proteins are depleted (20). Mutation of an outer membrane protein also results in specific sensitivity to polymyxin. Analysis of the mechanisms involved in the increased sensitivity of Omp31 mutants is under way.

The second hypothesis to explain the increased sensitivity to {gamma}9{delta}2 T-cell action involves the possible necessity to adapt to intracellular conditions that change after specific activation of macrophages by lymphocytes. Adaptation to intracellular conditions has been shown to be important for intracellular multiplication and thus for expression of virulence (28). It is noteworthy that activation of macrophages results in reduced oxygen tension inside the cell (23). It has been shown that Brucella must adapt to a microaerobic environment to be able to multiply actively by inducing high-oxygen-affinity cytochrome oxidase (24) and/or by switching to nitrate respiration (31). Mutation of the norD gene impairs nitrate respiration and results in attenuation in mice but not in macrophages growing with a normal oxygen concentration (31). This suggests that activation of the macrophages by {gamma}9{delta}2 T cells could result in a reduction in the oxygen tension. However, another hypothesis involves the production of NO by the infected macrophage-{gamma}9{delta}2 T-lymphocyte couple. Indeed, NorD is involved in detoxification of NO in Brucella, and it has been shown that murine macrophages produce NO to impair the growth of Brucella (19, 31) just as well as the human macrophages engineered to artificially produce NO do (18). However, it has never been proved that human macrophages or {gamma}9{delta}2 T cells are able to produce NO under infection conditions, although it was recently shown that {gamma}9{delta}2 T lymphocytes can produce NO when they are activated by stress proteins (Hsp60 and Hsp70) which could be produced by the infected macrophages (3). Thus, there are two hypotheses, (i) impairment of nitrate respiration under low oxygen tension and (ii) detoxification of NO produced by {gamma}9{delta}2 T lymphocytes and macrophages. Only a detailed analysis of the phenomenon will provide an answer to this question.

The third hypothesis supposes that there is stronger activation of macrophages by lymphocytes with Brucella mutants than with the wild type. This could mean that either an inhibitor of macrophage activation or an inhibitor of {gamma}9{delta}2 T lymphocytes is not present in the mutant. Alternatively, it could mean that activation of the macrophages by lymphocytes is impaired with the wild type but not with the mutant. The 16E7 mutant (BR1800), with a transposon inserted inside F1-ATP synthase, could be this type of mutant. Indeed, it has been recently shown that human {gamma}9{delta}2 T lymphocytes can be activated by a human F1-ATPase-related structure present on the surface of tumor cells (37). This molecule, involved in ATP synthesis and generally found inside the mitochondria, was detected at the cell surface. This observation has been interpreted as a possible explanation for the ability of {gamma}9{delta}2 T lymphocytes to recognize and kill tumor cells. It should be noted that mitochondria are derived from bacteria. It is thus possible that Brucella has diverted this molecule to impair {gamma}9{delta}2 T-cell activation.

Although all the genes seem to be very different in nature, a plausible role can be put forward for each of them to explain the reduced rate of multiplication of the mutants. This strengthens the validity of the method employed to identify the genes involved in subversion of the immune system.

Generality of the screening method. Virulence is a complex cross talk between the parasite and the host immune system that results from the necessity to diversify the steps in order to avoid definitive defeat. In this process innate immunity is a central component that determines the outcome of the infection. Indeed, this step can result in elimination of or at least a significant reduction in the parasite burden. To achieve this goal, alternative or cooperative mechanisms have been developed by mammals. Innate immunity can be divided into two main steps: (i) direct activation of phagocytic cells through the Toll-like receptor and (ii) recognition and management of infected cells by nonpolymorphic lymphocytes. {gamma}9{delta}2 T lymphocytes are specially designed to achieve this goal. They sense the presence of bacteria by recognition of specific prokaryote metabolites and activate the macrophages or directly attack the parasites. For intracellular bacteria, activation of the host cells and direct attack by antibacterial peptides are important processes affecting the success of the infection. It is thus very important to understand how pathogens can avoid or overcome this obstacle.

Studies can be performed either with entire animals or with fractionated reconstituted systems, provided that syngeneic cells can be used. Coculture of infected cells (in most cases macrophages) with purified lymphocytes is a time-consuming task. Thus, designing a screening method to approximate the dynamics of bacterial multiplication is necessary to tackle this difficult problem. We demonstrate in the present work that, in spite of some weaknesses of the method, such screening can be achieved.

Some questions arise from the screening procedure described here. The first question concerns false positives and false negatives that may partially reduce the quality of the results. In order to eliminate the false positives, every clone selected was tested individually. More than 50% of the clones selected after two rounds of screening were found to be positive when they were tested individually; false positives generally result from slight attenuation inside macrophages (Table 1). In order to improve the quality of the screening, a kinetic analysis of the intramacrophagic multiplication was performed with each clone. An example is shown in Fig. 5. We considered attenuated only clones for which multiplication was reduced by at least 1 log after 48 h of intracellular life compared to wild-type bacteria.

The case of false negatives is more difficult to unravel; they can be rescued only by a second round of screening. We applied our approach to the recent determination of the intramacrophagic virulome of Brucella (27), and very few new mutants were identified by a new screening, confirming that most of the mutants were detected in the first round. However, the development of a rapid screening method, such as the one described in this work, would help workers perform a second round of mutant detection.

Another question concerns the reproducibility of the results obtained with cells originating from more than one individual. In our experience, reproducibility from one patient to another is very good. Although we cannot completely eliminate the possibility of some genetic bias, the general susceptibility of the human population to Brucella infection and our laboratory experience support the hypothesis that there is a common mechanism for subversion of the innate immune system. Furthermore, it should be stressed that each positive clone was analyzed using different cell preparations that were obtained from different patients, hence eliminating specific bias.

In the work presented here, we used macrophages because they are the major hosts of Brucella. Macrophages adhere tightly to the bottom of wells. Thus, it is easy to wash away the supernatant that contains dead bacteria and the antibiotic gentamicin. The washing step allows removal of the antibiotic to determine the number of bacteria inside the macrophages. Many intracellular bacteria multiply inside macrophages (Mycobacterium, Legionella, Francisella, etc.), and the screening method described here can be used with them. However, bacteria that multiply inside epithelial cells can also be analyzed in this way, provided that the cells can attach on the bottom of the wells. This can be done using HeLa cells that attach when Ca2+ is added to the medium. Thus, the method has a wide range of possible applications.


arrow
ACKNOWLEDGMENTS
 
This work was supported in part by grant QLK2-CT-1999-00014 from the European Union.


arrow
FOOTNOTES
 
* Corresponding author. Mailing address: Université Montpellier II, CC100, place Eugène Bataillon, 34000 Montpellier, France. Phone: 33 4 67 14 32 09. Fax: 33 4 67 14 33 38. E-mail: liautard{at}univ-montp2.fr Back

{triangledown} Published ahead of print on 20 August 2007. Back

Editor: D. L. Burns


arrow
REFERENCES
 
    1
  1. Allen, C. A., L. G. Adams, and T. A. Ficht. 1998. Transposon-derived Brucella abortus rough mutants are attenuated and exhibit reduced intracellular survival. Infect. Immun. 66:1008-1016.[Abstract/Free Full Text]
    2. 2
  2. Arellano-Reynoso, B., N. Lapaque, S. Salcedo, G. Briones, A. E. Ciocchini, R. Ugalde, E. Moreno, I. Moriyon, and J. P. Gorvel. 2005. Cyclic beta-1,2-glucan is a Brucella virulence factor required for intracellular survival. Nat. Immunol. 6:618-625.[CrossRef][Medline]
    3. 3
  3. Atre, N., L. Thomas, R. Mistry, K. Pathak, and S. Chiplunkar. 2006. Role of nitric oxide in heat shock protein induced apoptosis of {gamma}{delta}T cells. Int. J. Cancer 119:1368-1376.[CrossRef][Medline]
    4. 4
  4. Bertotto, A., R. Gerli, F. Spinozzi, C. Muscat, F. Scalise, G. Castellucci, M. Sposito, F. Candio, and R. Vaccaro. 1993. Lymphocytes bearing the {gamma}{delta} T cell receptor in acute Brucella melitensis infection. Eur. J. Immunol. 23:1177-1180.[Medline]
    5. 5
  5. Bowden, R. A., S. M. Estein, M. S. Zygmunt, G. Dubray, and A. Cloeckaert. 2000. Identification of protective outer membrane antigens of Brucella ovis by passive immunization of mice with monoclonal antibodies. Microbes Infect. 2:481-488.[CrossRef][Medline]
    6. 6
  6. Caron, E., J. P. Liautard, and S. Kohler. 1994. Differentiated U937 cells exhibit increased bactericidal activity upon LPS activation and discriminate between virulent and avirulent Listeria and Brucella species. J. Leukoc. Biol. 56:174-181.[Abstract]
    7. 7
  7. Caron, E., T. Peyrard, S. Kohler, S. Cabane, J. P. Liautard, and J. Dornand. 1994. Live Brucella spp. fail to induce tumor necrosis factor alpha excretion upon infection of U937-derived phagocytes. Infect. Immun. 62:5267-5274.[Abstract/Free Full Text]
    8. 8
  8. Cloeckaert, A., N. Vizcaino, J. Y. Paquet, R. A. Bowden, and P. H. Elzer. 2002. Major outer membrane proteins of Brucella spp.: past, present and future. Vet. Microbiol. 90:229-247.[CrossRef][Medline]
    9. 9
  9. Duclos, S., and M. Desjardins. 2000. Subversion of a young phagosome: the survival strategies of intracellular pathogens. Cell. Microbiol. 2:365-377.[CrossRef][Medline]
    10. 10
  10. Dudal, S., C. Turriere, S. Bessoles, P. Fontes, F. Sanchez, J. Liautard, J. P. Liautard, and V. Lafont. 2006. Release of LL-37 by activated human V{gamma}9V{delta}2 T cells: a microbicidal weapon against Brucella suis. J. Immunol. 177:5533-5539.[Abstract/Free Full Text]
    11. 11
  11. Ekaza, E., L. Guilloteau, J. Teyssier, J. P. Liautard, and S. Kohler. 2000. Functional analysis of the ClpATPase ClpA of Brucella suis, and persistence of a knockout mutant in BALB/c mice. Microbiology 146:1605-1616.[Abstract/Free Full Text]
    12. 12
  12. Essenberg, R. C., R. Seshadri, K. Nelson, and I. Paulsen. 2002. Sugar metabolism by brucellae. Vet. Microbiol. 90:249-261.[CrossRef][Medline]
    13. 13
  13. Foulongne, V., G. Bourg, C. Cazevieille, S. Michaux-Charachon, and D. O'Callaghan. 2000. Identification of Brucella suis genes affecting intracellular survival in an in vitro human macrophage infection model by signature-tagged transposon mutagenesis. Infect. Immun. 68:1297-1303.[Abstract/Free Full Text]
    14. 14
  14. Freer, E., E. Moreno, I. Moriyon, J. Pizarro-Cerda, A. Weintraub, and J. P. Gorvel. 1996. Brucella-Salmonella lipopolysaccharide chimeras are less permeable to hydrophobic probes and more sensitive to cationic peptides and EDTA than are their native Brucella sp. counterparts. J. Bacteriol. 178:5867-5876.[Abstract/Free Full Text]
    15. 15
  15. Freer, E., J. Pizarro-Cerda, A. Weintraub, J. A. Bengoechea, I. Moriyon, K. Hultenby, J. P. Gorvel, and E. Moreno. 1999. The outer membrane of Brucella ovis shows increased permeability to hydrophobic probes and is more susceptible to cationic peptides than are the outer membranes of mutant rough Brucella abortus strains. Infect. Immun. 67:6181-6186.[Abstract/Free Full Text]
    16. 16
  16. Freiberg, C., R. Fellay, A. Bairoch, W. J. Broughton, A. Rosenthal, and X. Perret. 1997. Molecular basis of symbiosis between Rhizobium and legumes. Nature 387:394-401.[CrossRef][Medline]
    17. 17
  17. Godfroid, F., B. Taminiau, I. Danese, P. Denoel, A. Tibor, V. Weynants, A. Cloeckaert, J. Godfroid, and J. J. Letesson. 1998. Identification of the perosamine synthetase gene of Brucella melitensis 16M and involvement of lipopolysaccharide O side chain in Brucella survival in mice and in macrophages. Infect. Immun. 66:5485-5493.[Abstract/Free Full Text]
    18. 18
  18. Gross, A., S. Bertholet, J. Mauel, and J. Dornand. 2004. Impairment of Brucella growth in human macrophagic cells that produce nitric oxide. Microb. Pathog. 36:75-82.[CrossRef][Medline]
    19. 19
  19. Gross, A., S. Spiesser, A. Terraza, B. Rouot, E. Caron, and J. Dornand. 1998. Expression and bactericidal activity of nitric oxide synthase in Brucella suis-infected murine macrophages. Infect. Immun. 66:1309-1316.[Abstract/Free Full Text]
    20. 20
  20. Guzman-Verri, C., L. Manterola, A. Sola-Landa, A. Parra, A. Cloeckaert, J. Garin, J. P. Gorvel, I. Moriyon, E. Moreno, and I. Lopez-Goni. 2002. The two-component system BvrR/BvrS essential for Brucella abortus virulence regulates the expression of outer membrane proteins with counterparts in members of the Rhizobiaceae. Proc. Natl. Acad. Sci. USA 99:12375-12380.[Abstract/Free Full Text]
    21. 21
  21. Hensel, M., J. E. Shea, C. Gleeson, M. D. Jones, E. Dalton, and D. W. Holden. 1995. Simultaneous identification of bacterial virulence genes by negative selection. Science 269:400-403.[Abstract/Free Full Text]
    22. 22
  22. Huang, G., L. Zhang, and R. G. Birch. 2000. Rapid amplification and cloning of Tn5 flanking fragments by inverse PCR. Lett. Appl. Microbiol. 31:149-153.[CrossRef][Medline]
    23. 23
  23. James, P. E., O. Y. Grinberg, G. Michaels, and H. M. Swartz. 1995. Intraphagosomal oxygen in stimulated macrophages. J. Cell. Physiol. 163:241-247.[CrossRef][Medline]
    24. 24
  24. Jimenez de Bagues, M. P., S. Loisel-Meyer, J. P. Liautard, and V. Jubier-Maurin. 2007. Different roles of the two high-oxygen-affinity terminal oxidases of Brucella suis: cytochrome c oxidase, but not ubiquinol oxidase, is required for persistence in mice. Infect. Immun. 75:531-535.[Abstract/Free Full Text]
    25. 25
  25. Jubier-Maurin, V., R. A. Boigegrain, A. Cloeckaert, A. Gross, M. T. Alvarez-Martinez, A. Terraza, J. Liautard, S. Kohler, B. Rouot, J. Dornand, and J. P. Liautard. 2001. Major outer membrane protein Omp25 of Brucella suis is involved in inhibition of tumor necrosis factor alpha production during infection of human macrophages. Infect. Immun. 69:4823-4830.[Abstract/Free Full Text]
    26. 26
  26. Kaufmann, S., A. Sher, and R. Ahmaed. 2002. Immunology of infectious diseases. ASM Press, Washington, DC.
    27. 27
  27. Kohler, S., V. Foulongne, S. Ouahrani-Bettache, G. Bourg, J. Teyssier, M. Ramuz, and J. P. Liautard. 2002. The analysis of the intramacrophagic virulome of Brucella suis deciphers the environment encountered by the pathogen inside the macrophage host cell. Proc. Natl. Acad. Sci. USA 99:15711-15716.[Abstract/Free Full Text]
    28. 28
  28. Kohler, S., S. Michaux-Charachon, F. Porte, M. Ramuz, and J. P. Liautard. 2003. What is the nature of the replicative niche of a stealthy bug named Brucella? Trends Microbiol. 11:215-219.[Medline]
    29. 29
  29. Kohler, S., S. Ouahrani-Bettache, M. Layssac, J. Teyssier, and J. P. Liautard. 1999. Constitutive and inducible expression of green fluorescent protein in Brucella suis. Infect. Immun. 67:6695-6697.[Abstract/Free Full Text]
    30. 30
  30. Lachmann, P. J. 2002. Microbial subversion of the immune response. Proc. Natl. Acad. Sci. USA 99:8461-8462.[Free Full Text]
    31. 31
  31. Loisel-Meyer, S., M. P. Jimenez de Bagues, E. Basseres, J. Dornand, S. Kohler, J. P. Liautard, and V. Jubier-Maurin. 2006. Requirement of norD for Brucella suis virulence in a murine model of in vitro and in vivo infection. Infect. Immun. 74:1973-1976.[Abstract/Free Full Text]
    32. 32
  32. Manterola, L., I. Moriyon, E. Moreno, A. Sola-Landa, D. S. Weiss, M. H. Koch, J. Howe, K. Brandenburg, and I. Lopez-Goni. 2005. The lipopolysaccharide of Brucella abortus BvrS/BvrR mutants contains lipid A modifications and has higher affinity for bactericidal cationic peptides. J. Bacteriol. 187:5631-5639.[Abstract/Free Full Text]
    33. 33
  33. Murphy, E. A., J. Sathiyaseelan, M. A. Parent, B. Zou, and C. L. Baldwin. 2001. Interferon-gamma is crucial for surviving a Brucella abortus infection in both resistant C57BL/6 and susceptible BALB/c mice. Immunology 103:511-518.[CrossRef][Medline]
    34. 34
  34. Oliaro, J., S. Dudal, J. Liautard, J. B. Andrault, J. P. Liautard, and V. Lafont. 2005. V{gamma}9V{delta}2 T cells use a combination of mechanisms to limit the spread of the pathogenic bacteria Brucella. J. Leukoc. Biol. 77:652-660.[Abstract/Free Full Text]
    35. 35
  35. Ottones, F., J. Dornand, A. Naroeni, J. P. Liautard, and J. Favero. 2000. V{gamma}9V{delta}2 T cells impair intracellular multiplication of Brucella suis in autologous monocytes through soluble factor release and contact-dependent cytotoxic effect. J. Immunol. 165:7133-7139.[Abstract/Free Full Text]
    36. 36
  36. Ottones, F., J. Liautard, A. Gross, F. Rabenoelina, J. P. Liautard, and J. Favero. 2000. Activation of human V{gamma}9V{delta}2 T cells by a Brucella suis non-peptidic fraction impairs bacterial intracellular multiplication in monocytic infected cells. Immunology 100:252-258.[CrossRef][Medline]
    37. 37
  37. Scotet, E., L. O. Martinez, E. Grant, R. Barbaras, P. Jeno, M. Guiraud, B. Monsarrat, X. Saulquin, S. Maillet, J. P. Esteve, F. Lopez, B. Perret, X. Collet, M. Bonneville, and E. Champagne. 2005. Tumor recognition following V{gamma}9V{delta}2 T cell receptor interactions with a surface F1-ATPase-related structure and apolipoprotein A-I. Immunity 22:71-80.[CrossRef][Medline]
    38. 38
  38. Sicard, H., and J. J. Fournie. 2000. Metabolic routes as targets for immunological discrimination of host and parasite. Infect. Immun. 68:4375-4377.[Free Full Text]
    39. 39
  39. Ugalde, J. E., D. J. Comerci, M. S. Leguizamon, and R. A. Ugalde. 2003. Evaluation of Brucella abortus phosphoglucomutase (pgm) mutant as a new live rough-phenotype vaccine. Infect. Immun. 71:6264-6269.[Abstract/Free Full Text]
    40. 40
  40. Ugalde, J. E., C. Czibener, M. F. Feldman, and R. A. Ugalde. 2000. Identification and characterization of the Brucella abortus phosphoglucomutase gene: role of lipopolysaccharide in virulence and intracellular multiplication. Infect. Immun. 68:5716-5723.[Abstract/Free Full Text]

Infection and Immunity, November 2007, p. 5167-5174, Vol. 75, No. 11
0019-9567/07/$08.00+0     doi:10.1128/IAI.00690-07
Copyright © 2007, American Society for Microbiology. All Rights Reserved.




This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrowReprints and Permissions
Right arrow Copyright Information
Right arrow Books from ASM Press
Right arrow MicrobeWorld
Citing Articles
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Liautard, J.
Right arrow Articles by Liautard, J.-P.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Liautard, J.
Right arrow Articles by Liautard, J.-P.

Copyright © 2009 by the American Society for Microbiology.   For an alternate route to Journals.ASM.org, visit: http://intl-journals.asm.org | More Info»