INTRODUCTION

Top Abstract Introduction Materials and Methods Results Discussion References

Many pathogenic bacterial species are composed of multiple strains (serotypes) that can cause disease in animal hosts vaccinated against only a single pathogenic strain. Thus, it is desirable to develop bacterial vaccines that can stimulate cross-protective host immune responses to several pathogenic strains. Much of the work regarding the construction of live bacterial vaccines has been performed with Salmonella spp. since they establish an infection by direct interaction with the gut-associated lymphoid tissue, resulting in strong mucosal responses. Salmonella spp. also invade and proliferate within host cells and thus are capable of eliciting strong cell-mediated immune responses (9, 20, 23, 39).

Conceptually, cross-protective immunity could be elicited by live vaccines that express multiple antigens. The rationale is that although different serotypes possess different antigen repertoires, some of the protective antigens may be shared among heterologous serotypes and that expression of these shared antigens may lead to cross-protective immunity. We have recently shown that Salmonella DNA adenine methylase (Dam) mutants ectopically express multiple genes that are normally induced during infection (18). These Dam mutants are markedly attenuated but highly effective as live vaccines against Salmonella infection of mice (12, 18) and chickens (E. L. Dueger, J. K. House, D. M. Heithoff, and M. J. Mahan, submitted for publication). Similarly, Dam mutants are attenuated for virulence in Vibrio cholerae and Yersinia pseudotuberculosis and elicit protective immune responses against Yersinia bacteremia (26; S. M. Julio, D. M. Heithoff, D. Provenzano, K. E. Klose, R. L. Sinsheimer, D. A. Low, and M. J. Mahan, submitted for publication). Here we show that Dam and Dam-overproducing (Dam^OP) strains were sensitive to mediators of innate immunity and conferred cross-protective immunity to heterologous Salmonella serotypes, which are designated principally by variations in lipopolysaccharide O antigen structure.

	MATERIALS AND METHODS

Top Abstract Introduction Materials and Methods Results Discussion References

Bacterial strains and phage. Salmonella enterica serovar Typhimurium strains used in this study (Table 1) were derived from strain ATCC 14028 (CDC 6516-60). Strains used in infection studies were grown overnight in Luria broth (LB) at 37°C with shaking. The dam-102::Mud-Cm allele was obtained from John Roth and transduced into virulent S. enterica serovar Typhimurium strain 14028 and S. enterica serovar Enteritidis O1,9,12; CDC SSU7998, obtained from the Salmonella Genetic Stock Center, SARB, #16 (3, 36), resulting in the Dam strain, MT2223. S. enterica serovar Dublin Lane was obtained from Don Guiney (6). The construction of S. enterica serovar Typhimurium dam232 (MT2188) was described previously (18). Salmonella Dam overproducer strains were constructed by introduction of the E. coli Dam overproducer plasmid, pTP166 (29). The high-frequency generalized transducing bacteriophage P22 mutant HT105/1, int-201, was used for all transductional crosses (37), and phage-free, phage-sensitive transductants were isolated as previously described (5).

Media and chemicals. Luria broth (7) was the laboratory medium used in these studies. Final concentrations of antibiotics (Sigma) were as follows: ampicillin, 50 µg/ml; tetracycline, 20 µg/ml; and chloramphenicol, 20 µg/ml.

Virulence and protection assays. A 50% lethal dose (LD₅₀) assay was used to determine the lethal dose required to kill 50% of the animals; this virulence assay was performed as described previously (18). Briefly, mutant and wild-type cells were grown overnight in Luria broth with shaking. BALB/c mice, 6 to 8 weeks old, were perorally infected or perorally immunized by gastrointubation 0.2 M sodium phosphate buffer (pH 8.0). The protective capacity of Dam derivatives was determined by challenging immunized mice with the virulent strain. Mice were examined daily following challenge for morbidity and mortality. To determine the number of bacteria in host tissues, moribund mice were sacrificed and bacteria were recovered from host tissues and plated for determination of colony counts. Host tissues assayed include Peyer's patches (the four Peyer's patches proximal to the ileal-cecal junction), mesenteric lymph nodes, liver, and spleen.

Two-dimensional protein gel electrophoresis. Two-dimensional protein gel electrophoresis was performed by the method of O'Farrell (31) on whole-cell protein extracts of log-phase cells grown in Luria broth. Isoelectric focusing using pH 5 to 7 ampholines (Bio-Rad Laboratories, Hercules, Calif.) was carried out at 800 V for 17 h. The second dimension consisted of 12.5% polyacrylamide slab gels which were run for 5.5 h at 175 V. Proteins were visualized by silver staining (30).

	RESULTS

Top Abstract Introduction Materials and Methods Results Discussion References

Salmonella Dam mutant strains ectopically produce proteins in vitro. Recent work has shown that Salmonella Dam mutants ectopically express multiple genes (18) that are normally induced only during infection (17, 27). Therefore, we determined whether Salmonella Dam mutant strains ectopically expressed proteins in vitro since such inappropriate protein expression could lead to reduced virulence and the elicitation of protective immune responses. To this end, we analyzed protein expression in Dam, Dam⁺, and Dam^OP S. enterica serovar Typhimurium strains by two-dimensional gel electrophoresis of crude whole-cell protein extracts. Dam and Dam^OP strains expressed specific proteins, different from each other and different from those produced by wild-type Salmonella grown in vitro (Fig. 1A and C). In addition, at least one protein was preferentially expressed in wild-type Salmonella compared to the two Dam mutant strains (Fig. 1B). This latter expression pattern is similar to that of the Dam-regulated uropathogenic Escherichia coli pyelonephritis-associated pilus (pap) operon, in which under- and overexpression of Dam blocks Pap pilus production (2, 4, 41). Taken together, these data indicate that Salmonella strains that lack or overproduce Dam ectopically produce proteins in vitro. Such ectopic protein expression may contribute to the virulence attenuation of Dam mutant strains and the elicitation of protective immunity via an expanded source of potential antigens in vaccinated hosts.

View larger version (49K): [in this window] [in a new window]   FIG. 1.   Ectopic protein expression by Dam and DamOP Salmonella strains. Two-dimensional protein gel electrophoresis was performed on whole-cell protein extracts of Dam (MT2188) (A), Dam+ (ATCC 14028) (B), and DamOP (MT2128) (C) S. enterica serovar Typhimurium strains (30, 31). Arrows indicate representative examples of proteins that are preferentially expressed in the strains indicated.

Salmonella Dam-based vaccines confer cross-protective immune responses to heterologous serotypes. Since Dam is a global regulator of gene expression (18), we questioned whether immunization of mice with Salmonella Dam strains might elicit cross-protective immunity to heterologous Salmonella serotypes. Vaccination with a single oral dose of Dam S. enterica serovar Enteritidis significantly protected mice (P < 0.05) against challenge with either serovar Dublin or Typhimurium at 10,000 times the LD₅₀ (Table 2). Reciprocally, vaccination with Dam serovar Typhimurium significantly protected mice (P < 0.05) against challenge with either serovar Dublin at 10,000 times the LD₅₀ or serovar Enteritidis at 1,000 times the LD₅₀ (Table 2). Since Dam^OP strains of serovar Typhimurium are also attenuated (18), they were also evaluated for the elicitation of protective immune responses to homologous and heterologous Salmonella serotypes. We found that 75% of mice immunized with Dam^OP serovar Typhimurium survived a challenge at 1,000 times the LD₅₀ with either serovar Dublin or serovar Typhimurium (P < 0.05) (Table 2). Live attenuated Salmonella strains have previously been shown to elicit cross-protective immunity that is often short-lived and dependent on nonspecific immune responses due to persistent infection with the vaccine strain (21, 22). However, in this study, mice were challenged with virulent Salmonella 6 weeks after they had cleared the Dam or Dam^OP vaccine strains. That is, mice infected with an oral dose of 10⁹ Dam Salmonella had cleared the vaccine strain from the Peyer's patches, mesenteric lymph nodes, liver, and spleen 5 weeks postimmunization (data not shown). Accordingly, these data suggest that the cross-protection reported here was not mediated through nonspecific immune responses conferred by vaccine persistence in host tissues.

Mice immunized with Dam Salmonella exhibit enhanced protection compared to mice that recovered from a sublethal infection with the Dam⁺ virulent strain. It is generally thought that individuals who have recovered from a natural infection exhibit the strongest state of immunity to reinfection. Thus, the protection conferred by immunization with Dam Salmonella was compared to that elicited following recovery from a sublethal infection (at the LD₅₀) with the virulent wild-type strain. At an immunizing dose of 10⁵ bacteria, animals vaccinated with the Dam S. enterica serovar Typhimurium were significantly more protected (P < 0.05) against a lethal challenge with virulent organisms than were animals that recovered from a sublethal infection with Dam⁺ Typhimurium (Fig. 2). Additionally, mice immunized at this dose (10⁵) with Dam Typhimurium were protected similarly over a range of virulent challenge doses (10⁷ to 10⁹ bacteria). Previous work showed that mice immunized with 10⁹ Dam organisms are fully protected against a challenge with 10⁹ virulent organisms (18). Therefore, an immunizing dose of 10⁵ Dam bacteria appears to be a suboptimal number of organisms to elicit protection against high challenge doses.

View larger version (29K): [in this window] [in a new window]   FIG. 2.   Mice immunized with Dam Salmonella exhibit high levels of protection compared to mice recovering from a sublethal infection with the virulent parental strain. The protective immunity elicited in mice immunized with 105 Dam S. enterica serovar Typhimurium organisms (MT2188) was compared to either that elicited in mice that survived an infection of 105 bacteria (LD50) of the virulent parental Dam+ strain (14028) or to nonimmunized mice. Twenty immunized or nonimmune mice were challenged 5 weeks postinfection with doses of 107, 108, or 109 virulent serovar Typhimurium organisms (ATCC 14028). The enhanced protection conferred by Dam Salmonella was significant at all challenge doses according to the two-tailed Fisher exact test (P < 0.05).

Mice immunized with Dam Salmonella show hindered proliferation of virulent Dam⁺ Salmonella in mucosal and systemic tissues. Since mice immunized with Dam-based vaccines showed no overt signs of disease after challenge with virulent bacteria, we determined the fates of virulent S. enterica serovar Typhimurium in vaccinated versus nonvaccinated mice. Following a challenge dose with 10,000-fold the LD₅₀, nonvaccinated mice showed a rapid increase in bacterial number in the Peyer's patches, mesenteric lymph nodes, liver, and spleen and succumbed to the infection on day 6 (Fig. 3). In contrast, although mice vaccinated with Dam serovar Typhimurium carried high loads (10³ to 10⁴ CFU/g) of virulent bacteria in both mucosal and systemic tissues (days 1 to 5), these immunized mice hindered the further proliferation of, and in some cases eliminated, wild-type bacteria from both mucosal and systemic tissues (on day 28, no bacteria were detected in two of four immunized mice). These results indicate that for the high challenge dose tested, the Dam vaccine protects mice from the lethal effects of Salmonella by blocking the proliferation of Salmonella at mucosal and systemic sites.

View larger version (31K): [in this window] [in a new window]   FIG. 3.   Mice immunized with Dam Salmonella block the proliferation of virulent Salmonella in mucosal and systemic tissues. Virulent S. enterica serovar Typhimurium (109 14028 organisms) was perorally administered to nonvaccinated mice (open boxes) or to mice perorally vaccinated with 109 Dam serovar Typhimurium organisms (MT2188) (solid boxes). Vaccinated mice were challenged 5 weeks postimmunization. The fate of the virulent organisms was determined in host tissues at the times indicated after challenge. All nonvaccinated mice were dead by day 6. PP, Peyer's patches; MLN, mesenteric lymph nodes.

Salmonella Dam strains are not immunosuppressive in infected mice. To begin to understand the immunological basis of protection, we examined the contribution of innate immune responses to infection by Salmonella Dam mutant strains. One of the host responses that follows wild-type Salmonella infection is the activation of macrophages and the concomitant release of nitric oxide (NO), which has been shown (via peroxynitrite) to have antibacterial activity (8). In addition to being involved in innate immune functions, Salmonella-induced NO has immunosuppressive effects on the adaptive immune response via lymphocyte inactivation (10), resulting in a condition wherein animals experience a transient state of nonspecific immunosuppression (9). Therefore, we compared both the levels of induced NO and the extent of immunosuppression induced by Dam S. enterica serovar Typhimurium to those induced by wild-type infection or by administration of a serovar Typhimurium live attenuated vaccine deficient in aromatic amino acid biosynthesis (AroA) (24). The results show that NO levels measured in splenocytes derived from mice infected with Dam Salmonella are well below those observed after infection with virulent Dam⁺ Salmonella or after an AroA immunization (Table 3). The inability to induce high NO levels suggests that Dam Salmonella strains are not immunosuppressive in mice.

View this table: [in this window] [in a new window]   TABLE 3.   Dam Salmonella strains do not exhibit elevated NO levelsa

View this table: [in this window] [in a new window]   TABLE 4.   Dam Salmonella strains are not immunosuppressive in infected micea

Salmonella Dam mutant strains are sensitive to components of innate immunity. To assess the contribution of innate immune responses to infection by Dam mutant Salmonella strains, we tested whether these strains were sensitive to components of innate immunity, including antimicrobial peptides (defensins NP-1 and bactinecin [15]), detergents (bile salts [13]), and mediators of oxidative damage (H₂O₂ [1]). Our results show that Dam S. enterica serovar Typhimurium strains are more sensitive to these agents than are wild-type Dam⁺ bacteria (Table 5). Analysis of S. enterica Typhimurium Dam^OP strains showed that they are as sensitive to bile salts and hydrogen peroxide as are Dam strains but, unlike Dam Salmonella strains, are relatively resistant to the antimicrobial peptides evaluated, which may reflect differences in virulence gene expression between these strains. These data suggest that an enhanced sensitivity to innate immune functions contributes to the attenuated virulence of Salmonella Dam mutant strains.

View this table: [in this window] [in a new window]   TABLE 5.   Dam and DamOP Salmonella strains are sensitive to antimicrobial peptides, bile salts, and hydrogen peroxidea

	DISCUSSION

Top Abstract Introduction Materials and Methods Results Discussion References

One of the problematic aspects of vaccine design is that there are often many different pathogenic isolates of a given species that contribute to disease. Thus, vaccination against one strain may not elicit protection against another strain or even against a variant of the parental strain. This is a principal reason why protective immunity against some microbes may require annual vaccinations with different strains, why vaccine efficacy may depend on the specific pathogenic isolates endemic to a given geographical region, and why mutant variants can cause disease in populations that are immune to infection with the parental strain. Some of these problems may be circumvented by the use of vaccine strains that ectopically express multiple antigens that are shared among different pathogenic strains. Here we show that Salmonella Dam and Dam^OP live attenuated vaccines elicited cross-protective immunity to heterologous Salmonella serotypes.

Previous reports have shown that the cross-protective response to S. enterica serovar Typhimurium live attenuated vaccines is highly dependent on nonspecific immune responses attributed to the persistence of the vaccine strain within host tissues (reviewed in references 16, 21, and 22). In this study, the cross-protective immunity elicited was not attributed to the persistence of the vaccine strain since mice were protected against heterologous challenge 6 weeks after the Dam vaccine strain was eliminated from immunized animals. Notably, Salmonella Dam vaccines were not immunosuppressive; unlike wild-type infection or aroA vaccination, they failed to induce increased NO levels and permitted a subsequent robust humoral response to DT antigen in infected mice.

Insights into the possible mechanisms by which Dam regulates gene expression come from regulatory analysis of the E. coli pap operon (2, 4, 41), which codes for pili that are essential for virulence in two animal models of pyelonephritis (32, 35). Dam target sites in the pap promoter are protected from methylation by the binding of regulatory proteins at or near these sites, forming specific DNA methylation patterns similar to those observed in eukaryotes (4, 14, 19, 34, 40). These DNA methylation patterns regulate gene expression by modulating the binding of regulatory proteins to Dam target sites. Notably, DNA methylation conveys additional information to DNA without altering the sequence, and such an epigenetic methylation signal is transmitted to future generations. This provides a cellular memory mechanism in which the behavior of daughter cells can be influenced by the environment that their parent cells experienced. This methylation-directed cellular memory system may be important for the infection cycle, which can be viewed analogously to a developmental program (25). Supporting this notion, the cell cycle-regulated methyltransferase, CcrM, is thought to play an important role in coordinating gene expression with the cell cycle in Caulobacter, which undergoes a morphogenetic alteration between a motile swarmer cell and a sessile stalked cell (33).

The role of Dam in virulence and in the elicitation of protective immune responses may rely on its capacity as a global regulator of gene expression (18, 24a, 26, 28). One possible consequence of Dam dysregulation is an expanded repertoire of antigens that contribute to the heightened immunity observed in vaccinated hosts. Additionally, the nonimmunosuppressive nature and low-grade persistence of Dam mutant vaccines in host tissues (12) may provide a stable source of antigens over the time needed to transition to the development of strong adaptive immune responses. Since DNA adenine methylases are highly conserved in a wide variety of virulent bacteria (24a, 26), dysregulation of Dam activity is potentially a general strategy for the generation of vaccines against bacterial pathogens. In addition, Salmonella Dam mutants may serve as a platform to express passenger bacterial and viral antigens that elicit protective immune responses to the cognate pathogen.