Skip Navigation

“Gently Rough”: The Vaccine Potential of a Salmonella enterica Regulatory Lipopolysaccharide Mutant

  1. Gábor Nagy1,
  2. Tamás Palkovics1,
  3. Andreas Otto2,
  4. Harald Kusch2,
  5. Béla Kocsis1,
  6. Ulrich Dobrindt3,
  7. Susanne Engelmann2,
  8. Michael Hecker2,
  9. Levente Emödy1,
  10. Tibor Pál1 and
  11. Jörg Hacker3
  1. 1Department of Medical Microbiology and Immunology University of Pécs, Pécs, Hungary
  2. 2Institute for Microbiology, Ernst-Moritz-Arndt University, Greifswald, Germany
  3. 3Institute for Molecular Biology of Infectious Diseases, University of Würzburg, Würzburg, Germany
  1. Reprints or correspondence: Gábor Nagy, Dept. of Medical Microbiology and Immunology, University of Pécs, Szigeti út 12, 7624 Pécs, Hungary (gabor.nagy{at}aok.pte.hu).

  1. Presented in part: 108th General Meeting of the American Society for Microbiology, Toronto, 21–25 May 2007 (poster D-093); 162nd Meeting of the Society for General Microbiology, Edinburgh, 31 March–3 April 2008 (poster MI 43).

 
Next Section

Abstract

Background. An alternative to multivalent vaccines could be to construct strains capable of conferring broad protection through shared antigens. Down-regulation of immunodominant major antigens has been proposed to enhance the immunogenicity of conserved antigens.

Methods. The protection provided by an aroA as well as structural and regulatory lipopolysaccharide (LPS) mutants of Salmonella enterica serovar Typhimurium against homologous and heterologous challenges was assessed in the murine model of typhoid. The reactivity and cross-reactivity of the immune sera raised was tested by enzyme-linked immunospot assay and immunoblots. Conserved outer membrane proteins were identified by mass spectrometry.

Results. Unlike any structural LPS mutants, the regulatory mutant lacking RfaH was finely balanced between safety and immunogenicity, and its vaccine potential was comparable to that of the well-characterized ΔaroA mutant. Loss of the transcriptional antiterminator RfaH resulted in a heterogeneous length of LPS chains, designated here as the “gently rough” phenotype. Our study also provides evidence that the rough phenotype enhances the immunogenicity of minor antigens, which may improve cross-protection against heterologous bacteria. A panel of conserved antigens shared by members of the Enterobacteriaceae family was identified as abundant porins and lipoprotein antigens.

Conclusions. Fine-tuned down-regulation of immunodominant epitopes can create live vaccine strains that are not only desirably attenuated but that also exhibit an improved cross-protective potential.

Members of the Enterobacteriaceae family are among the most common human pathogens that cause significant morbidity and mortality [1]. Moreover, some species are prone to developing resistance to numerous antibiotics and, hence, are able to cause severe nosocomial infections or even outbreaks difficult to treat [2]. Vaccines preventing some of these infections are anticipated to have a positive impact on fighting enterobacterial infections and epidemics [1]. The development of vaccines against enteric bacteria, however, is hindered by the huge number of serotypes. The species Salmonella enterica alone has >2500 different serovars, distinguished according to the extremely variable and highly immunogenic surface antigens (flagellar and lipopolysaccharide [LPS] O antigens) expressed. On the other hand, these antigenically highly variable structural elements cover and shield a rather “conserved” panel of surface antigens. An attractive approach would be to construct vaccine strains lacking the multiform external cover and, hence, expose the shared antigens, which otherwise bear minor immunogenic potential. The highly variable outermost surface structures, however, appear to be indispensable for the virulent phenotype. Consequently, structural mutants lacking these virulence factors tend to be overattenuated—with limited in vivo fitness—and, hence, inappropriate as live vaccine candidates [3].

The regulatory protein RfaH is a transcriptional antiterminator that reduces the polarity of long operons encoding secreted and surface-associated cell components of enteric bacteria [4]. It affects the synthesis of LPS through the regulation of the expression of both core oligosaccharide [5, 6] and O side chain synthesis [7]. Additionally, several other gene clusters acquired horizontally seem to have evolved to coutilize this regulatory protein. The global regulatory role of RfaH in the virulence of uropathogenic Escherichia coli has been demonstrated [8]. On the other hand, genomewide transcriptome analysis in S. Typhimurium revealed that only the sii operon encoded on Salmonella pathogenicity island (SPI)-4 is directly coregulated with LPS synthesis by RfaH [9]. Two recent studies showed that SPI-4 has only a moderate impact on the virulence of Salmonella pathogens [10, 11]; therefore, attenuation of ΔrfaH mutants can be mainly attributed to the rough LPS phenotype [9].

Here, the vaccine potential of the S. Typhimurium ΔrfaH mutant is compared with that of the well-characterized ΔaroA mutant as well as with that of various structural mutants exhibiting truncation at different levels of the LPS molecule.

Previous SectionNext Section

Methods

Strains and culture conditions. Sequenced strains (http://www.sanger.ac.uk) of Salmonella enterica serovars Typhimurium (SL1344) and Enteritidis (NCTC13349) were used. The regulatory (ΔrfaH) and structural (ΔwaaG and ΔwaaL) LPS mutants have been described elsewhere [9, 12]. The SL1344ΔaroA mutant was constructed by P22 transduction from donor strain MvP473 (S. Typhimurium 14028ΔaroA::aph), which was provided by M. Hensel (Universität Nürnberg-Erlangen, Germany). The panel of enterobacterial strains used in ELISAs (table 1), protein preparations, and immunoblotting originated from the strain collection of the University of Pécs Department of Medical Microbiology and Immunology. Mutants of these strains were generated by the Red recombinase method [15]. Bacteria were routinely grown in Luria-Bertani (LB) broth or on agar plates at 37°C, with the exception of the Erwinia carotovora strain (30°C).

Figure 1
View larger version:
Figure 1

A, Lipopolysaccharide (LPS) phenotypes of SL1344 (lane 1) and its isogenic ΔrfaH (lane 2), ΔwaaG (lane 3), and ΔwaaL (lane 4) deletion mutants. Purified LPS was silver stained after separation by SDS-PAGE on 15% gel. B, Schematic representation of the Salmonella Typhimurium lipopolysaccharide molecule. Arrows indicate the level of truncation in the structural LPS mutants. AA, 4-aminoarabinose; EtNP, phophoethanolamine; P, phosphate.

Figure 2
View larger version:
Figure 2

Survival of immunized mice after homologous (Salmonella Typhimurium) (A) or heterologous (Salmonella Enteritidis) (B) infection. Groups of 8 mice or, in the case of the heterologous challenge, 16 mice (except for the double mutant) were orally vaccinated with the indicated SL1344 mutants, as described in Methods. Animals were challenged orally 5 weeks after the last booster with 5 × 106 cfu of either SL1344 (2000-fold LD50) or NCTC13349 (10-fold LD50).

Figure 3
View larger version:
Figure 3

Reactivity of the ΔrfaH immune serum to homologous and heterologous bacteria (see table 1 for details on the target bacteria used). Pooled immune sera were tested to smooth, rough, and deep-rough mutants of Salmonella Typhimurium SL1344, Salmonella Enteritidis NCTC13349, Shigella flexneri 2457T, and Klebsiella pneumoniae 3091 by indirect ELISA. Reactivity is expressed relative to that of the ΔaroA immune serum at the same dilution. Means of values from 3 independent experiments are shown.

Figure 4
View larger version:
Figure 4

Conserved protein antigens identified by the ΔrfaH immune serum. Outer membrane proteins of different enterobacterial pathogens (lanes 1–9) and Pseudomonas aeruginosa (lane 10) were separated by SDS-PAGE (A). The proteins were blotted onto nitrocellulose membranes and reacted with the ΔrfaH immune serum (B). The shared cross-reactive antigens are marked with arrows.

Table 1
View larger version:
Table 1

Phenotypes and genotypes of bacterial strains used in the ELISAs.

LPS work. LPS was purified, separated on 15% gels (SDS-PAGE), and stained as described elsewhere [9]. P22 transduction was performed following standard protocols. The phage used for transduction was grown on a kanamycin-resistant strain (14028ΔaroA::aph). SL1344 and its isogenic mutants were infected with equal volumes of the obtained lysate. The number of transductants was determined after overnight incubation on LB agar supplemented with 100 µg/mL kanamycin.

Immunization experiments. Animal experiments were conducted in a laboratory authorized by the Hungarian government (decree number XXVII, 1998) and by the subsequent regulation (government order number 243/1998). Groups of eight 6–8-week-old female BALB/c mice were immunized via the orogastric route, using a sterile gavage without any prior neutralization of gastric acid. Priming was performed with 2 × 107 cfu (∼1 × 104-fold LD50 of the wild-type strain) of an attenuated mutant of SL1344 (either ΔwaaL, ΔwaaG, ΔrfaH, or ΔaroA deletion mutants or the ΔaroAΔrfaH double mutant), which was followed by 2 boosters containing 2 × 108 and 2 × 109 cfu at 2-week intervals. Vaccinated mice were challenged 5 weeks after the last booster with 5 × 106 cfu of either SL1344 (S. Typhimurium) or NCTC13349 (S. Enteritidis). Groups of 7 mice were either vaccinated 3 times with 1 × 109 cfu of NCTC13349ΔrfaH [11] or mock vaccinated with saline. Four weeks after the last booster, mice were challenged with 2 × 106 cfu of SL1344. In all cases, animals were observed for 28 days after infection, and survival curves were statistically analyzed by the log rank test.

ELISAs. Descriptions of phenotypes and genotypes of the target bacterial strains used are provided in table 1. Assays were conducted as described elsewhere [8]. The immunoreactivity of the different sera was expressed in relation to the reactivity of the ΔaroA immune sera at the same dilution (1:800). Means were calculated from 3 independent assays.

Western blots. Outer membrane protein (OMP) preparation was principally done as described by others [16]. OMPs were separated by SDS-PAGE using 12.5% gels and blotted onto nitrocellulose membranes using a Mini Trans-Blot cell (Bio-Rad). The blocked membranes were treated with the diluted (1:200) immune or control serum and secondary antibody (anti-mouse IgG-horseradish peroxidase conjugate; Dako) in Tris-buffered saline with Tween containing 2% skimmed milk. The blot was developed using enhanced chemiluminescence detection reagents (NEN Life Science) followed by luminography.

Proteomics methods. Purified OMPs of various enterobacterial species were depleted using 2 pools of naive mouse serum. Subsequently, the remaining proteins were reacted with the ΔrfaH immune serum, and the complexes were collected on protein G beads. Following elution, the proteins were separated by SDS-PAGE (10% gels) and stained with SYPRO Ruby (Molecular Probes).

For protein identification, nano liquid chromatography (LC) tandem mass spectrometry (MS/MS) was performed as described in detail in appendix A, which appears only in the electronic edition of the Journal. In brief, peptides were separated via C18 reverse-phase chromatography before MS analysis. The resulting MS/MS data (table 2) were used to run the Mascot search engine (Matrix Science) with organism-specific sequence decoy databases. Proteins were regarded as having been positively identified when at least 2 peptides were identified at a false-positive rate of 0% and when the calculated mass matched the mass that was expected on the basis of the SDS-PAGE.

Table 2
View larger version:
Table 2

Outer membrane proteins identified by mass spectrometry (MS).

Previous SectionNext Section

Results

LPS phenotypes. The phenotypes of the LPS mutants were determined by silver staining (figure 1A). Loss of WaaL (O antigen ligase; figure 1B) resulted in an intact core with no O antigens attached to it [17]. The ΔwaaG mutant exhibited an LPS phenotype (figure 1), where all molecules are truncated at the level of the inner core [18]. Inactivation of RfaH resulted in a truncated core with a molecular mass clearly between those of the ΔwaaL and ΔwaaG mutants (figure 1A). Truncation of the LPS molecules in the ΔrfaH mutant, however, did not appear to happen at a discrete site within the outer core (figure 1A). Considering the nature of RfaH, we investigated the hypothesis that truncation of the LPS molecules in the ΔrfaH regulatory mutant might have been partial. To determine whether the O side chains were indeed present—albeit at a level not detectable by silver staining—infection with the O antigen-specific P22 phage was done (i.e., using the S. Typhimurium O antigen as receptor). The ΔwaaL and ΔwaaG mutants were resistant to transduction by P22, confirming the complete lack of the O antigens in these clones (table 3). In contrast, transduction of the ΔrfaH mutant resulted in >200 kanamycin-resistant colonies, proving the presence of some intact O side chains on the bacterial surface. The number of transductants obtained, however, was ∼18-fold less than that in the case of the wild-type strain, confirming a major reduction in the quantity of O antigens.

Table 3
View larger version:
Table 3

Ability of Salmonella Typhimurium mutants to be transduced by phage P22.

Balance between safety and immunogenicity. To obtain a clear picture of the vaccine potential of the ΔrfaH mutant, we compared it to that of an ΔaroA mutant as well as to those of structural LPS mutants exhibiting discrete truncation of LPS molecules. Groups of mice were vaccinated with various mutants of the S. Typhimurium strain and were subsequently challenged with the homologous wild-type strain. The ΔwaaL mutant was not sufficiently attenuated, given that 5 (63%) of the 8 mice immunized died as a result of an overwhelming systemic infection. However, the surviving animals were fully protected (figure 2A). On the other hand, the ΔwaaG mutant conferred a low level of protection (13%), although all mice survived the immunization procedure. In agreement with former reports, both the ΔrfaH [12] and the ΔaroA [19] mutant provided significant protection (86% and 88%, respectively), with only 1 mouse succumbing to the challenge in each immunized group. The virulence of the ΔrfaH strain was minimal (1 of the 8 mice died), whereas the ΔaroA mutant was completely avirulent in this vaccination scheme. On the other hand, the ΔaroAΔrfaH double mutant appeared to be highly overattenuated. Although all of the mice survived vaccination, none were protected against the challenge.

Cross-protection. To test whether cross-immunity among different serovars of S. enterica is triggered on vaccination, mice were immunized as described above and were challenged by the mouse-virulent S. Enteritidis strain NCTC13349. The trend toward lethality after vaccination was similar as before: virulence of the vaccine strain was highest in the case of the ΔwaaL mutant, resulting in a death rate of 31% (5/16) among immunized mice. On the other hand, no lethality was observed on vaccination with the ΔwaaG mutant. One (6%) of the 16 immunized animals died from both the ΔrfaH as well as the ΔaroA mutant. After challenge with 10-fold LD50 of S. Enteritidis, all vaccinated groups showed statistically significantly different survival curves (figure 2B) in comparison to mock-vaccinated mice. The degree of protection provided by the various mutants, however, was different. The best protection was provided by the ΔwaaL mutant strain (91% survival; 10/11). The ΔaroA and ΔrfaH mutants elicited 60% (9/15) and 40% (6/15) survival, respectively. The ΔaroAΔrfaH double mutant as well as the ΔwaaG mutant protected 38% (3/8 and 6/16, respectively) of the immunized animals.

To confirm that the serovar-independent protection provided by the ΔrfaH mutant was reproducible, part of the opposite experiment was also conducted: animals were vaccinated with the ΔrfaH mutant of the S. Enteritidis strain NCTC13347 and challenged with 1000-fold LD50 of the wild-type S. Typhimurium strain SL1344. This protected 43% (3/7) of the infected animals, whereas all mock-vaccinated mice (7/7) died after receipt of this challenge dose (data not shown).

Increased immunogenicity of conserved antigens. We [9] and others [20] proposed earlier that abolished or down-regulated expression of dominant surface antigens that are highly variable (such as LPS) should increase the immunogenic potential of conserved minor epitopes. Accordingly, one would anticipate a stronger immune reactivity to heterologous bacteria of the sera raised against LPS mutants than against smooth cells. To test this hypothesis, the reactivity of pooled immune sera (n = 14) originating from vaccination with the S. Typhimurium ΔrfaH mutant was compared with that of ΔaroA immune serum (pool of 16 samples) in ELISA against a panel of homologous and heterologous wild-type strains, as well as their various LPS mutants (table 3). As expected, the control (nonimmune) serum exhibited weak reactivity (<50% of that of both immune sera) to either of the target cells (data not shown). Immunoreactivity of the ΔrfaH immune serum to the respective targets was expressed relative to that of the ΔaroA immune serum (figure 3). Antibodies raised against the ΔrfaH mutant showed superior reactivity to heterologous bacteria, such as Shigella flexneri and Klebsiella pneumoniae, suggesting that the net immunogenicity of conserved antigens expressed by this strain is higher compared with that of the smooth ΔaroA vaccine strain. This trend was shown for both smooth and rough heterologous test strains, indicating that smooth LPS does not mask the shared reactive epitopes. On the other hand, the immunoreactivity of the ΔaroA immune serum to the homologous smooth strain (SL1344) was greater, implying that the majority of antibodies raised against this vaccine strain were LPS specific. That S. Typhimurium and S. Enteritidis share epitopes of their LPS (the core as well as factors 1 and 12 of the O antigen) was clearly shown by the results: the smooth and rough variants of both serovars were more reactive with the immune sera raised against the ΔaroA mutant. However, when using target cells lacking the majority of the LPS epitopes (i.e., deep-rough bacteria), the ΔrfaH immune serum pool was more reactive. All these data suggest that loosing the bulk of the immunodominant LPS antigen (as in the ΔrfaH mutant) increases the immune response-inducing potential of the LPS-independent antigens conserved among different enterobacterial pathogens.

Identification of shared antigens. To identify cross-reactive antigens, OMPs from different enteric bacteria were prepared and probed on Western blots with pooled (n = 9) ΔrfaH immune serum [12]. Figure 4 shows that several proteins from extracts of various representatives of Enterobacteriaceae were labeled, whereas no reactivity to proteins of Pseudomonas aeruginosa, a nonenterobacterial Gram negative rod (figure 4, lane 10), was seen. No proteins were detected by pooled naive mouse serum, either (data not shown). Two different pools of immune sera (n = 8 and n = 12, respectively) from independent vaccinations detected identical panels of protein antigens, confirming the reproducibility of the immune response (data not shown). Immunoreactive proteins were collected by protein G beads and subjected to MS analysis. Proteins whose orthologues were identified in at least 3 different enterobacterial species are listed in table 4. The antigens identified included abundant (OmpA and OmpC) and minor (e.g., NmpC and OmpX) protein antigens as well as the most copious lipoprotein (murein lipoprotein) in the cell wall.

Table 4
View larger version:
Table 4

Cross-reactive outer membrane proteins shared by enterobacterial species.

Previous SectionNext Section

Discussion

Live attenuated enteric pathogens are thought to exhibit significant vaccine potential [1, 21]. Application of such strains is, however, hindered by the difficulties of developing rationally attenuated strains [22]. Overattenuation hampers fitness and, hence, results in an inadequate immune response. On the other hand, insufficiently attenuated strains may induce diseases resembling those caused by natural infection. The ideal vaccine candidate balances on the thin edge between safety (i.e., appropriate attenuation) and the retained ability to elicit long-lasting protective immunity [23].

LPS is a major virulence factor in pathogenesis of Salmonella infection and contributes to several stages of the infectious process, such as swarming motility [24], intestinal colonization [25], serum resistance [26], invasion [27], and intracellular replication [9]. Therefore, structural rough mutants have been considered to be inappropriate live vaccine candidates [3]. On the other hand, conditional down-regulation of LPS is the key factor of attenuation in the licensed typhoid vaccine strain Ty21a [23] as well as in some recently developed Salmonella vector strains [28]. Instead of transient expression of LPS, our vaccine candidate strain relies on rational down-regulation of LPS expression. On the basis of their increased susceptibility to detergents and to some antibiotics, S. Typhimurium rfaH mutants were shown to exhibit some characteristics of the so-called deep-rough phenotype and, hence, are sufficiently attenuated [12]. At the same time, however, oral vaccination with an ΔrfaH mutant was still efficient in inducing protection against a subsequent challenge by wild-type Salmonella strains, suggesting that the truncation of LPS molecules might not have been complete. The lack of any detectable O ladder and the position of the core on the silver-stained gel suggested that the majority of LPS molecules in the ΔrfaH mutant must have been truncated at the level of the outer core. However, that the ΔrfaH mutant could be successfully transduced by phage P22 proved that at least some long-chained LPS molecules were, in fact, retained. These results are in agreement with the nature of regulatory protein RfaH; loss of this transcriptional antiterminator elicits polarity in the affected (LPS-encoding) operons but does not fully abolish synthesis of full-length transcripts and, hence, does not entirely eliminate synthesis of any of the enzymes encoded by these loci [9]. Therefore, few LPS molecules with intact core regions are to be expected, with some even being capped by shorter or longer polymers of O antigen-repeating units. Apparently, loss of RfaH results in a heterogeneous population of LPS chains that differ in their length. We propose that this phenotype be designated “gently rough,” to reflect the heterogeneous nature of the LPS in ΔrfaH mutants.

According to our results, as far as the safety and efficacy in BALB/c mice are concerned, the ΔrfaH mutant has a vaccine potential comparable to that of the highly efficient and well-characterized ΔaroA mutant [19]. Truncation of the majority of LPS molecules at the level of the LPS core sufficiently attenuates virulence, whereas the retained long-chained LPS molecules allow some of the invasiveness required to elicit protective immunity. Indeed, the virulence of the ΔrfaH mutant was much lower than that of the isogenic ΔwaaL mutant (devoid of the O side chain only) while still providing a high level of protection against a homologous challenge. On the other hand, the immunogenicity of the ΔrfaH mutant was superior to that of the structural deep-rough mutant (ΔwaaG), which exhibited complete truncation at the level of the inner core.

Because O antigen is the most immunodominant antigen on the surface of enterobacterial species, down-regulation of LPS synthesis may result in enhanced immunogenicity of minor (“immunosubdominant”) antigens. This concept has also been proposed in strategies for vaccines against viruses that exhibit a huge number of different serotypes [29]. Furthermore, a recent study by Russo et al. [20] showed that vaccination with killed cells of a rough and nonencapsulated mutant of a uropathogenic E. coli strain elicited an increased net immune response to conserved minor antigens. Here, we could also show that partial loss of LPS antigens in the ΔrfaH mutant allowed higher immunogenic potential for conserved antigens. Whether the improved immunogenicity of minor antigens on LPS mutants originate from their increased surface exposure or from the lack of antigenic competition imposed by the highly immunogenic LPS epitopes still needs to be elucidated. Nevertheless, microarray analysis did not reveal differing expression of the identified proteins in the ΔrfaH mutant [9]. On the other hand, the ectopic expression of shared antigens was shown to generate cross-protective immunity in another Salmonella mutant lacking the global regulatory protein Dam (DNA adenine methylase) [30–32]. Similarly, an altered cell wall structure of Salmonella Δaro mutants [33] may be responsible for the cross-protection provided by the ΔaroA vaccine strain.

Recent sequencing projects have revealed that various representatives of Enterobacteriaceae share a huge number of surface-exposed antigens that retain a high degree of sequence similarity. This information corroborates former observations showing that enteric bacteria share a panel of OMPs with immunological relatedness [34]. Here, we also identified several well-conserved surface antigens by immunobloting using pooled immune sera obtained from mice vaccinated with the gently rough Salmonella strain. In light of these data, it is tempting to speculate on the potential of using purified forms of these proteins as subunit vaccines for the generation of a broadly protective immune response. Indeed, it has been shown previously by Isibasi et al. [35] that purified OMPs were able to mediate serotype-independent protection against Salmonella organisms in mice. Tabaraie et al. [36] provided further evidence that porin-immunized mice were protected against both homologous and heterologous challenges. Moreover, that work as well as others [37, 38] proved that OMPs derived from rough mutants were also capable of inducing a cross-protective immune response. This observation is especially relevant in light of studies showing that the outer membrane composition of the cell wall of rough mutants could be substantially different from that of smooth strains [39–41].

The murein (Braun) lipoprotein was identified by the ΔrfaH immune serum in most enterobacterial species. Along with LPS, this lipoprotein is a major constituent of the gram-negative bacterial cell wall. Besides being a critical molecule for virulence [42, 43], it has been shown to cross-react with the immune sera raised against whole cells of different enterobacterial pathogens [44]. Furthermore, Braun et al. [44] showed that the highest anti-lipoprotein titers were elicited by deep-rough mutants, which is in agreement with our results showing a higher immunogenic potential of surface antigens in the (partial) absence of full-length LPS molecules. Furthermore, monoclonal antibodies against the Braun lipoprotein have been shown to cross-react with an array of smooth enterobacteria [45], rendering this molecule an important target for strategies for broadly protective immunization.

Besides these abundant molecules, other minor OMPs were also recognized by the anti-rough Salmonella immune serum. Proteins belonging to the OmpX family were identified in several enterobacterial species investigated. Ail is required for adhesion and invasion [46] as well as for serum resistance [47] of Yersinia enterocolitica. Although speculations on the role played by OmpX in the virulence of E. coli were based only on structure analysis [48], its immunogenic potential was proven recently by Maisnier-Patin et al. [49]. Additionally, several other OMPs originating from various genera (e.g., NmpC in salmonellae and SfpA in Y. enterocolitica) reacted with the immune serum. Some of these antigens have been proposed to be involved in virulence as well [50]; hence, their neutralization through an efficient immune response might contribute to the feasibility of a broad-spectrum vaccination approach.

In summary, the rational down-regulation of the expression of LPS, as shown here for the ΔrfaH mutant, can establish an optimally attenuated live vaccine strain with potential similar to that of the well-characterized smooth ΔaroA mutant. Furthermore, we have shown that the immune response-inducing capacity of several conserved minor antigens was increased in the ΔrfaH mutant. Recently, it has been suggested that several of these epitopes have virulence-related functions and, hence, protective capacities. Consequently, the gently rough ΔrfaH mutant may have the potential to serve as a vaccine that generates cross-protective immunity among enterobacterial pathogens.

Previous SectionNext Section

Acknowledgments

We are grateful to Michael Hensel (Universität Nürnberg-Erlangen, Germany) for providing Salmonella Typhimurium strain MvP473 and to Zoltán Péterfi (University of Pécs, Hungary) for some of the outer membrane protein preparations used in this work.

Previous SectionNext Section

Footnotes

  • Potential conflicts of interest: none reported.

  • Financial support: This study was supported by the Hungarian Scientific Research Fund (grants K62092 and F48526 to the Pécs group, grant SFB479 TP A1 to the Würzburg group, and grant PTJ-Bio/0313807A to the Greifswald group). The work was performed within the framework of the European Community Network of Excellence EuroPathoGenomics program. G.N. was supported by Bolyai and Humboldt fellowships.

  • Received April 17, 2008.
  • Accepted July 8, 2008.
Previous Section
 

References

  1. ↵
    1. Girard MP,
    2. Steele D,
    3. Chaignat CL,
    4. Kieny MP
    . A review of vaccine research and development: human enteric infections. Vaccine 2006;24:2732-50.
    CrossRefMedlineWeb of Science
  2. ↵
    1. Paterson DL
    . Resistance in gram-negative bacteria: Enterobacteriaceae. Am J Infect Control 2006;34:S20-8.
    CrossRefMedlineWeb of Science
  3. ↵
    1. Nagy G,
    2. Pal T
    . Lipopolysaccharide: a tool and target in enterobacterial vaccine development. Biol Chem 2008;389:513-20.
    CrossRefMedlineWeb of Science
  4. ↵
    1. Bailey MJ,
    2. Hughes C,
    3. Koronakis V
    . RfaH and the ops element, components of a novel system controlling bacterial transcription elongation. Mol Microbiol 1997;26:845-51.
    CrossRefMedlineWeb of Science
  5. ↵
    1. Lindberg AA,
    2. Hellerqvist CG
    . Rough mutants of Salmonella typhimurium: immunochemical and structural analysis of lipopolysaccharides from rfaH mutants. J Gen Microbiol 1980;116:25-32.
    Abstract/FREE Full Text
  6. ↵
    1. Creeger ES,
    2. Schulte T,
    3. Rothfield LI
    . Regulation of membrane glycosyltransferases by the sfrB and rfaH genes of Escherichia coli and Salmonella typhimurium. J Biol Chem 1984;259:3064-9.
    Abstract/FREE Full Text
  7. ↵
    1. Wang L,
    2. Jensen S,
    3. Hallman R,
    4. Reeves PR
    . Expression of the O antigen gene cluster is regulated by RfaH through the JUMPstart sequence. FEMS Microbiol Lett 1998;165:201-6.
    CrossRefMedlineWeb of Science
  8. ↵
    1. Nagy G,
    2. Dobrindt U,
    3. Schneider G,
    4. Khan AS,
    5. Hacker J,
    6. Emody L
    . Loss of regulatory protein RfaH attenuates virulence of uropathogenic Escherichia coli. Infect Immun 2002;70:4406-13.
    Abstract/FREE Full Text
  9. ↵
    1. Nagy G,
    2. Danino V,
    3. Dobrindt U,
    4. et al
    . Down-regulation of key virulence factors makes the Salmonella enterica serovar Typhimurium rfaH mutant a promising live-attenuated vaccine candidate. Infect Immun 2006;74:5914-25.
    Abstract/FREE Full Text
  10. ↵
    1. Gerlach RG,
    2. Jackel D,
    3. Stecher B,
    4. et al
    . Salmonella pathogenicity island 4 encodes a giant non-fimbrial adhesin and the cognate type 1 secretion system. Cell Microbiol 2007;9:1834-50.
    CrossRefMedline
  11. ↵
    1. Kiss T,
    2. Morgan E,
    3. Nagy G
    . Contribution of SPI-4 genes to the virulence of Salmonella enterica. FEMS Microbiol Lett 2007;275:153-9.
    CrossRefMedlineWeb of Science
  12. ↵
    1. Nagy G,
    2. Dobrindt U,
    3. Hacker J,
    4. Emody L
    . Oral immunization with an rfaH mutant elicits protection against salmonellosis in mice. Infect Immun 2004;72:4297-301.
    Abstract/FREE Full Text
    1. Wei J,
    2. Goldberg MB,
    3. Burland V,
    4. et al
    . Complete genome sequence and comparative genomics of Shigella flexneri serotype 2a strain 2457T. Infect Immun 2003;71:2775-86.
    Abstract/FREE Full Text
    1. Oelschlaeger TA,
    2. Tall BD
    . Invasion of cultured human epithelial cells by Klebsiella pneumoniae isolated from the urinary tract. Infect Immun 1997;65:2950-8.
    Abstract/FREE Full Text
  15. ↵
    1. Datsenko KA,
    2. Wanner BL
    . One-step inactivation of chromosomal genes in Escherichia coli K-12 using PCR products. Proc Natl Acad Sci USA 2000;97:6640-5.
    Abstract/FREE Full Text
  16. ↵
    1. Osborn MJ,
    2. Munson R
    . Separation of the inner (cytoplasmic) and outer membranes of gram-negative bacteria. Methods Enzymol 1974;31:642-53.
    Medline
  17. ↵
    1. Mulford CA,
    2. Osborn MJ
    . An intermediate step in translocation of lipopolysaccharide to the outer membrane of Salmonella typhimurium. Proc Natl Acad Sci USA 1983;80:1159-63.
    Abstract/FREE Full Text
  18. ↵
    1. Yethon JA,
    2. Vinogradov E,
    3. Perry MB,
    4. Whitfield C
    . Mutation of the lipopolysaccharide core glycosyltransferase encoded by waaG destabilizes the outer membrane of Escherichia coli by interfering with core phosphorylation. J Bacteriol 2000;182:5620-3.
    Abstract/FREE Full Text
  19. ↵
    1. Hoiseth SK,
    2. Stocker BA
    . Aromatic-dependent Salmonella typhimurium are non-virulent and effective as live vaccines. Nature 1981;291:238-9.
    CrossRefMedline
  20. ↵
    1. Russo TA,
    2. Beanan JM,
    3. Olson R,
    4. et al
    . A killed, genetically engineered derivative of a wild-type extraintestinal pathogenic E. coli strain is a vaccine candidate. Vaccine 2007;25:3859-70.
    CrossRefMedlineWeb of Science
  21. ↵
    1. Dietrich G,
    2. Griot-Wenk M,
    3. Metcalfe IC,
    4. Lang AB,
    5. Viret JF
    . Experience with registered mucosal vaccines. Vaccine 2003;21:678-83.
    CrossRefMedlineWeb of Science
  22. ↵
    1. Raupach B,
    2. Kaufmann SH
    . Bacterial virulence, proinflammatory cytokines and host immunity: how to choose the appropriate Salmonella vaccine strain? Microbes Infect 2001;3:1261-9.
    CrossRefMedlineWeb of Science
  23. ↵
    1. Guzman CA,
    2. Borsutzky S,
    3. Griot-Wenk M,
    4. et al
    . Vaccines against typhoid fever. Vaccine 2006;24:3804-11.
    CrossRefMedlineWeb of Science
  24. ↵
    1. Toguchi A,
    2. Siano M,
    3. Burkart M,
    4. Harshey RM
    . Genetics of swarming motility in Salmonella enterica serovar typhimurium: critical role for lipopolysaccharide. J Bacteriol 2000;182:6308-21.
    Abstract/FREE Full Text
  25. ↵
    1. Nevola JJ,
    2. Laux DC,
    3. Cohen PS
    . In vivo colonization of the mouse large intestine and in vitro penetration of intestinal mucus by an avirulent smooth strain of Salmonella typhimurium and its lipopolysaccharide-deficient mutant. Infect Immun 1987;55:2884-90.
    Abstract/FREE Full Text
  26. ↵
    1. Shaio MF,
    2. Rowland H
    . Bactericidal and opsonizing effects of normal serum on mutant strains of Salmonella typhimurium. Infect Immun 1985;49:647-53.
    Abstract/FREE Full Text
  27. ↵
    1. Hoare A,
    2. Bittner M,
    3. Carter J,
    4. et al
    . The outer core lipopolysaccharide of Salmonella enterica serovar Typhi is required for bacterial entry into epithelial cells. Infect Immun 2006;74:1555-64.
    Abstract/FREE Full Text
  28. ↵
    1. Brogden KA,
    2. Minion FC,
    3. Cornick N,
    4. et al.
    1. Curtiss R III.,
    2. Zhang X,
    3. Wanda SY,
    4. et al.
    ; Brogden KA, Minion FC, Cornick N, et al., editors. Induction of host immune responses using Salmonella-vectored vaccines. Virulence mechanisms of bacterial pathogens Washington, DC: American Society for Microbiology Press; (4th ed.) 2007:297-313.
  29. ↵
    1. Spier RE
    . On the need for, and the delivery of, cross-protective vaccines. Vaccine 2005;23:2027-9.
    CrossRefMedlineWeb of Science
  30. ↵
    1. Heithoff DM,
    2. Enioutina EY,
    3. Daynes RA,
    4. Sinsheimer RL,
    5. Low DA,
    6. Mahan MJ
    . Salmonella DNA adenine methylase mutants confer cross-protective immunity. Infect Immun 2001;69:6725-30.
    Abstract/FREE Full Text
    1. Mohler VL,
    2. Heithoff DM,
    3. Mahan MJ,
    4. et al
    . Cross-protective immunity conferred by a DNA adenine methylase deficient Salmonella enterica serovar Typhimurium vaccine in calves challenged with Salmonella serovar Newport. Vaccine 2008;26:1751-8.
    CrossRefMedlineWeb of Science
  32. ↵
    1. Mohler VL,
    2. Heithoff DM,
    3. Mahan MJ,
    4. et al
    . Cross-protective immunity in calves conferred by a DNA adenine methylase deficient Salmonella enterica serovar Typhimurium vaccine. Vaccine 2006;24:1339-45.
    CrossRefMedlineWeb of Science
  33. ↵
    1. Sebkova A,
    2. Karasova D,
    3. Crhanova M,
    4. Budinska E,
    5. Rychlik I
    . aro mutations in Salmonella enterica cause defects in cell wall and outer membrane integrity. J Bacteriol 2008;190:3155-60.
    Abstract/FREE Full Text
  34. ↵
    1. Hofstra H,
    2. Van Tol JD,
    3. Dankert J
    . Cross-reactivity of major outer membrane proteins of Enterobacteriaceae, studied by crossed immunoelectrophoresis. J Bacteriol 1980;143:328-37.
    Abstract/FREE Full Text
  35. ↵
    1. Isibasi A,
    2. Ortiz V,
    3. Vargas M,
    4. et al
    . Protection against Salmonella typhi infection in mice after immunization with outer membrane proteins isolated from Salmonella typhi 9,12,d,Vi. Infect Immun 1988;56:2953-9.
    Abstract/FREE Full Text
  36. ↵
    1. Tabaraie B,
    2. Sharma BK,
    3. Sharma PR,
    4. Sehgal R,
    5. Ganguly NK
    . Evaluation of Salmonella porins as a broad spectrum vaccine candidate. Microbiol Immunol 1994;38:553-9.
    MedlineWeb of Science
  37. ↵
    1. Kuusi N,
    2. Nurminen M,
    3. Saxen H,
    4. Valtonen M,
    5. Makela PH
    . Immunization with major outer membrane proteins in experimental salmonellosis of mice. Infect Immun 1979;25:857-62.
    Abstract/FREE Full Text
  38. ↵
    1. Udhayakumar V,
    2. Muthukkaruppan VR
    . Protective immunity induced by outer membrane proteins of Salmonella typhimurium in mice. Infect Immun 1987;55:816-21.
    Abstract/FREE Full Text
  39. ↵
    1. van Alphen W,
    2. Lugtenberg B,
    3. Berendsen W
    . Heptose-deficient mutants of Escherichia coli K12 deficient in up to three major outer membrane proteins. Mol Gen Genet 1976;147:263-9.
    CrossRefMedlineWeb of Science
    1. Ried G,
    2. Hindennach I,
    3. Henning U
    . Role of lipopolysaccharide in assembly of Escherichia coli outer membrane proteins OmpA, OmpC, and OmpF. J Bacteriol 1990;172:6048-53.
    Abstract/FREE Full Text
  41. ↵
    1. Rivera M,
    2. Bertasso A,
    3. McCaffrey C,
    4. Georgopapadakou NH
    . Porins and lipopolysaccharide of Escherichia coli ATCC 25922 and isogenic rough mutants. FEMS Microbiol Lett 1993;108:183-7.
    CrossRefMedlineWeb of Science
  42. ↵
    1. Sha J,
    2. Fadl AA,
    3. Klimpel GR,
    4. Niesel DW,
    5. Popov VL,
    6. Chopra AK
    . The two murein lipoproteins of Salmonella enterica serovar Typhimurium contribute to the virulence of the organism. Infect Immun 2004;72:3987-4003.
    Abstract/FREE Full Text
  43. ↵
    1. Sha J,
    2. Agar SL,
    3. Baze WB,
    4. et al
    . Braun lipoprotein (Lpp) contributes to virulence of yersiniae: potential role of Lpp in inducing bubonic and pneumonic plague. Infect Immun 2008;76:1390-409.
    Abstract/FREE Full Text
  44. ↵
    1. Braun V,
    2. Bosch V,
    3. Klumpp ER,
    4. Neff I,
    5. Mayer H,
    6. Schlecht S
    . Antigenic determinants of murein lipoprotein and its exposure at the surface of Enterobacteriaceae. Eur J Biochem 1976;62:555-66.
    MedlineWeb of Science
  45. ↵
    1. Henriksen AZ,
    2. Maeland JA,
    3. Brakstad OG
    . Monoclonal antibodies against three different enterobacterial outer membrane proteins: characterization, cross-reactivity, and binding to bacteria. APMIS 1989;97:559-68.
    CrossRefMedlineWeb of Science
  46. ↵
    1. Miller VL,
    2. Falkow S
    . Evidence for two genetic loci in Yersinia enterocolitica that can promote invasion of epithelial cells. Infect Immun 1988;56:1242-8.
    Abstract/FREE Full Text
  47. ↵
    1. Bliska JB,
    2. Falkow S
    . Bacterial resistance to complement killing mediated by the Ail protein of Yersinia enterocolitica. Proc Natl Acad Sci USA 1992;89:3561-5.
    Abstract/FREE Full Text
  48. ↵
    1. Vogt J,
    2. Schulz GE
    . The structure of the outer membrane protein OmpX from Escherichia coli reveals possible mechanisms of virulence. Structure 1999;7:1301-9.
    Medline
  49. ↵
    1. Maisnier-Patin K,
    2. Malissard M,
    3. Jeannin P,
    4. et al
    . The outer membrane protein X from Escherichia coli exhibits immune properties. Vaccine 2003;21:3765-74.
    CrossRefMedlineWeb of Science
  50. ↵
    1. Mildiner-Earley S,
    2. Miller VL
    . Characterization of a novel porin involved in systemic Yersinia enterocolitica infection. Infect Immun 2006;74:4361-5.
    Abstract/FREE Full Text
| Table of Contents

This Article

  1. J Infect Dis. (2008) 198 (11): 1699-1706. doi: 10.1086/593069
  1. AbstractFree
  2. » Full Text (HTML)Free
  3. Full Text (PDF)Free

Classifications

Share

  1. Email this article

Navigate This Article

Current Issue

  1. March 1, 2012 205 (5)
  1. Journal of Infectious Diseases
  1. Alert me to new issues

Most