Salmonella enterica Serovar Typhimurium Strains with Regulated ...

INFECTION AND IMMUNITY, Mar. 2009, p. 1071–1082 0019-9567/09/$08.00⫹0 doi:10.1128/IAI.00693-08 Copyright © 2009, American Society for Microbiology. All Rights Reserved.

Vol. 77, No. 3

Salmonella enterica Serovar Typhimurium Strains with Regulated Delayed Attenuation In Vivo䌤 Roy Curtiss III,1* Soo-Young Wanda,1 Bronwyn M. Gunn,1† Xin Zhang,2‡ Steven A. Tinge,3 Vidya Ananthnarayan,1 Hua Mo,1 Shifeng Wang,1 and Wei Kong1 Center for Infectious Diseases and Vaccinology, Biodesign Institute and School of Life Sciences, Arizona State University, Tempe, Arizona 85287-54011; Department of Biology, Washington University, St. Louis, Missouri 631302; and Avant Immunotherapeutics, Inc., 8620 Pennell Drive, Overland, Missouri 631143 Received 2 June 2008/Returned for modification 31 July 2008/Accepted 8 December 2008

Recombinant bacterial vaccines must be fully attenuated for animal or human hosts to avoid inducing disease symptoms while exhibiting a high degree of immunogenicity. Unfortunately, many well-studied means for attenuating Salmonella render strains more susceptible to host defense stresses encountered following oral vaccination than wild-type virulent strains and/or impair their ability to effectively colonize the gut-associated and internal lymphoid tissues. This thus impairs the ability of recombinant vaccines to serve as factories to produce recombinant antigens to induce the desired protective immunity. To address these problems, we designed strains that display features of wild-type virulent strains of Salmonella at the time of immunization to enable strains first to effectively colonize lymphoid tissues and then to exhibit a regulated delayed attenuation in vivo to preclude inducing disease symptoms. We recently described one means to achieve this based on a reversible smooth-rough synthesis of lipopolysaccharide O antigen. We report here a second means to achieve regulated delayed attenuation in vivo that is based on the substitution of a tightly regulated araC PBAD cassette for the promoters of the fur, crp, phoPQ, and rpoS genes such that expression of these genes is dependent on arabinose provided during growth. Thus, following colonization of lymphoid tissues, the Fur, Crp, PhoPQ, and/or RpoS proteins cease to be synthesized due to the absence of arabinose such that attenuation is gradually manifest in vivo to preclude induction of diseases symptoms. Means for achieving regulated delayed attenuation can be combined with other mutations, which together may yield safe efficacious recombinant attenuated Salmonella vaccines. Attenuation of Salmonella vaccine vectors should decrease, if not eliminate, induction of undesirable disease symptoms while the vaccine retains immunogenicity. The attenuated vaccine should be sufficiently invasive and persistent to stimulate both strong primary and lasting memory immune responses and should be designed to minimize consequential adverse events. As even attenuated vaccines may sometimes cause disease (72), the vaccine should be susceptible to clinically useful antibiotics. Achieving a balance between adequate attenuation and safety and maximal immunogenicity in vaccine construction is difficult. Many means to attenuate Salmonella vaccines make them less able to tolerate stresses encountered in the gastrointestinal tract after oral administration, including exposure to acid, bile, increasing osmolarity and iron, and decreasing O2, and/or reduce invasion of the gut-associated lymphoid tissue (GALT). The doses for recombinant Salmonella vaccines to elicit maximal immune responses in mice are lower for intranasal immunization than they are for oral immunization (37, 55, 58). This may be due, in part, to killing of orally administered vaccines by the acid stress of the stomach (24, 30)

quickly followed by exposure to bile in the duodenum. We have determined that these two stresses in succession are more effective in causing bacterial cell death than the sum of killing by each stress alone (M. R. Wilmes-Riesenberg and R. Curtiss, unpublished data). Salmonella possesses a large constellation of genes that confer acid tolerance and resistance to acid stress (1, 17, 20, 21, 51), and inactivation of these genes or their inability to be expressed by induction reduces virulence (76). In this regard, the regulatory proteins RpoS (44), Fur (32), PhoPQ (6, 7), and OmpR (3, 4) are all necessary to confer resistance to acid stress and/or shock in Salmonella enterica serovar Typhimurium. Similarly, many genes are turned on in response to exposure to bile, and some of these gene products transiently repress invasion while bacteria reside in the intestinal lumen (29, 60, 73, 75). The exceedingly low dose of Shigella needed for oral infectivity correlates well with the innate expression of high resistance to acid stresses (74, 75) and the presumed unimportance of bile stress. However, complete lipopolysaccharide (LPS) is of considerable importance as rough mutants of Salmonella lacking LPS O-antigen side chains or portions of the core are avirulent, fail to colonize the intestinal tract, and are deficient in invading cells of the intestinal mucosa (69, 70). This could be due to increased sensitivity to bile or complement and/or an inability to penetrate mucin to enable adherence to intestinal cells prior to invasion. As Salmonella traverses the intestinal tract, there is an increase in osmolarity and a decrease in available oxygen; both of these environmental signals induce the expression of the Salmonella pathogenicity island 1 genes necessary for cell invasion (18, 23,

* Corresponding author. Mailing address: Biodesign Institute, Arizona State University, P.O. Box 875401, Tempe, AZ 85287-5401. Phone: (480) 727-0445. Fax: (480) 727-0466. E-mail: [email protected]. † Present address: Department of Microbiology, University of North Carolina, Chapel Hill, NC. ‡ Present address: Department of Pathology, Washington University, St. Louis, MO 63108. 䌤 Published ahead of print on 22 December 2008. 1071

1072

CURTISS ET AL.

42), as does the succession of low-pH passage through the stomach followed by the neutral pH of the ileal contents (2). There are also likely stresses to ions, defensins, and other metabolites that might impair the ability of bacterial vaccine vectors, depending on the means of attenuation, to persist in the intestinal tract for sufficient time to enable cell attachment and invasion. In this regard, genes regulated by PhoPQ (25, 26, 61, 73) and PmrAB (77) very much contribute to resistance to bile stress, defensins, and iron stress. Serovar Typhimurium mutants with ⌬phoP, ⌬phoQ, or ⌬phoPQ mutations are all totally avirulent for mice and highly immunogenic in inducing protective immunity to challenge with virulent wild-type strains. This is surprising in that such mutants, although colonizing the GALT to reasonable levels in spite of their increased sensitivity to acid stress, defensins, and bile (61, 73), are found in the mesenteric lymph nodes and spleens of orally immunized mice at much reduced levels (22) compared to titers in numbers of CFU observed after oral administration of either ⌬aro or ⌬cya ⌬crp attenuated strains (14, 36). These collective results demonstrate that ⌬phoPQ mutants are totally avirulent and highly immunogenic but imply that some of the attenuation is due to a reduced ability to colonize lymphoid tissues. RpoS controls expression of the serovar Typhimurium virulence plasmid spv genes (19, 57). The spvRABCD gene cluster controls the growth rate of Salmonella in deep organs and is required for systemic infection and bacteremia in animals and humans (see reference 28 for a review). As expected, Salmonella rpoS mutants have a severely impaired capacity to colonize spleens of infected mice, resulting in avirulence in mice (10, 11, 40). In addition, rpoS mutations reduce the ability of serovar Typhimurium to colonize Peyer’s patches of infected mice (11, 56). Based on the above observations and thoughts, we reasoned that it might be important to have mutations contributing to attenuation or other beneficial vaccine attributes that do not impair the abilities of the vaccine to adjust to and/or withstand a diversity of stresses encountered at any location within the gastrointestinal tract if the vaccine is administered orally or in the respiratory tract if it is administered intranasally. Likewise, there may be a benefit to having a vaccine strain that expresses wild-type abilities not compromised by direct mutations to penetrate through mucin, to attach to cells in the mucosal epithelium, and to be invasive into those cells. To achieve these objectives, we have developed six means using three strategies to achieve regulated delayed attenuation of Salmonella in vivo such that strains at the time of immunization exhibit almost the same abilities as fully virulent wild-type strains to contend with stresses and successfully reach effector lymphoid tissues before displaying attenuation, which precludes onset of any disease symptoms. The first strategy (15) involves a smooth-to-rough phenotypic change in LPS in vivo and makes use of pmi mutants that lack the phosphomannose isomerase needed to interconvert fructose-6-phosphate and mannose-6-phosphate (49). Strains with the ⌬pmi mutation grown in the presence of mannose synthesize a complete LPS O antigen but lose LPS O-antigen side chains after about seven generations of growth in medium devoid of mannose or in tissues since nonphosphorylated mannose, required for uptake to synthesize O antigen, is unavailable. We report here our second strategy based on regulated delayed expression in vivo of virulence genes. We

INFECT. IMMUN.

thus describe four means to be used alone or in combination to provide a regulated delayed attenuation phenotype so that vaccine strains with these mutations have nearly the ability of wild-type Salmonella to colonize lymphoid tissues before exhibiting an attenuated phenotype. Each means confers significant attenuation and improved immunogenicity compared to selected attenuated strains made by direct mutation in virulence genes. Our third strategy (39) uses a system for regulated delayed lysis in vivo to provide both attenuation and biological containment.

MATERIALS AND METHODS Bacterial strains, media, and bacterial growth. All strains for testing in mice were derived from the highly virulent S. enterica serovar Typhimurium strain UK-1 (13). All bacterial strains are listed in Table 1. LB broth and agar (8) were used as complex media for propagation and plating of bacteria. Nutrient broth and agar (Difco), which are devoid of arabinose and mannose, and minimal salts medium and agar (12) were also used. Some studies were done with bacterial strains grown in tissue culture medium to simulate environments to be encountered in vivo. MacConkey agar with 0.5% lactose, 0.2 or 0.5% arabinose, or 0.5% maltose was used to indicate fermentation of sugars and enumerate bacteria from mice. Chrome azurol S (CAS) plates, which were used to determine siderophore production, were made by the addition of an indicator solution of CAS mixed with Fe⫹3 and hexadecyltrimethyl ammonium bromide to morpholinepropanesulfonic acid basal agar (68). To detect phosphatase activity, X-P plates to detect phosphatase activity were made by addition of 5-bromo-4-chloro-3-indolyl-phosphate to nutrient agar at a final concentration of 50 ␮g/ml. Kornberg agar was prepared as a glycogen indicator agar (33, 41, 64). Selenite broth, with or without supplements, was used for enrichment of Salmonella from tissues although later results demonstrated that enrichment with tetrathionate broth gave better results when vaccine strains had multiple mutations. Bacterial growth was monitored spectrophotometrically and by plating for colony counts. Molecular and genetic procedures. Methods for DNA isolation, restriction enzyme digestion, DNA cloning, and use of PCR for construction and verification of vectors were standard (65). DNA sequence analysis was performed in the DNA Sequence Laboratory in the School of Life Sciences at Arizona State University. All oligonucleotide and/or gene segment syntheses were done commercially. PCR amplification with primers designed for specific modifications was used to alter promoter, ribosome binding/Shine-Dalgarno (SD), and start codon sequences. Conjugational transfer of suicide vectors for generation of unmarked deletion and deletion-insertion mutations was performed by standard methods (52, 63) using the suicide vector donor strain ␹7213 (Table 1). Since live vaccine strains cannot display resistance to antibiotics, we used means to generate defined deletion mutations using suicide vector technologies that did not use drug resistance markers or leave molecular scars. Subsequently, these unmarked defined deletion mutations with and without specific insertions were introduced into strains using P22HTint (66, 67) transduction of suicide vectors integrated into the deletion or deletion-insertion mutation, followed by selection for sucrose resistance as described previously (38). Whenever insertion of a regulatory sequence might adversely affect expression of an adjoining gene, we included a transcription terminator (TT) to prevent such consequences. We generally used strong TTs from bacteriophages. Plasmid constructs were evaluated by DNA sequencing, the ability to complement various serovar Typhimurium mutant strains (Table 1), and the ability to specify synthesis of proteins using gel electrophoresis and Western blot analyses. His- or glutathione S transferase-tagged proteins were produced and used to obtain anti-protein rabbit antiserum for Western blot analyses. Strain characterizations. We took exquisite care in strain construction and did complete biochemical and genetic characterizations after every step in strain construction. This included running an LPS gel (34, 71) to make sure that we did not select rough variants. We conducted comparative growth analyses since our objective was to have single and multiply mutant strains grow at almost the same rate and to the same density as the wild-type parental strains when strains were grown under permissive conditions. We also evaluated vaccine strain stability in respect to possible recombinational and/or mutational events as described in Results. Strains were also evaluated for biochemical and metabolic attributes, sensitivity to antibiotics and drugs, serological properties, and resistance compared to wild-type parental strains to stresses associated with exposure to acid (7) and bile (29).

VOL. 77, 2009

REGULATED DELAYED ATTENUATION IN VIVO

1073

TABLE 1. Bacterial strains Genotypea

Strain

E. coli K-12 strain ␹7213

F⫺ supE42 ␭⫺ T3r thi-1 thr-1 leuB6 supE44 tonA21 lacY1 recA1 RP4–2-Tc::Mu ␭pir ⌬asdA4 ⌬zhf-2::Tn10

S. enterica serovar Typhimurium UK-1 strains ␹3761 ␹8848 ␹9201 ␹9273 ␹9346 ␹9269 ␹9347 ␹8956 ␹9200 ␹8918 ␹9199 ␹9021 ␹9202 ␹9371 ␹9382 ␹9372 ␹9383 ␹9107 ␹9550 ␹9576 ␹9578 ␹9108 ␹9064

Wild-type serovar Typhimurium UK-1 ⌬Pfur33::TT araC PBAD fur ⌬Pfur33::TT araC PBAD fur ⌬araBAD23 ⌬Pfur77::TT araC PBAD fur ⌬Pfur77::TT araC PBAD fur ⌬araBAD23 ⌬Pfur81::TT araC PBAD fur ⌬Pfur81::TT araC PBAD fur ⌬araBAD23 ⌬PrpoS183::TT araC PBAD rpoS ⌬PrpoS183::TT araCPBAD rpoS ⌬araBAD23 ⌬PphoPQ107::TT araC PBAD phoPQ ⌬PphoPQ107::TT araC PBAD phoPQ ⌬araBAD23 ⌬Pcrp527::TT araC PBAD crp ⌬Pcrp527::TT araC PBAD crp ⌬araBAD23 ⌬PphoPQ173::TT araC PBAD phoPQ ⌬PphoPQ173::TT araC PBAD phoPQ ⌬araBAD23 ⌬PphoPQ177::TT araC PBAD phoPQ ⌬PphoPQ177::TT araC PBAD phoPQ ⌬araBAD23 ⌬Pfur33::TT araC PBAD fur ⌬Pcrp527::TT araC PBAD crp ⌬Pfur77::TT araC PBAD fur ⌬Pcrp527::TT araC PBAD crp ⌬Pfur81::TT araC PBAD fur ⌬Pcrp527::TT araC PBAD crp ⌬Pcrp527::TT araC PBAD crp ⌬Pfur81::TT araC PBAD fur ⌬PphoPQ107::TT araC PBAD phoPQ ⌬Pcrp527::TT araC PBAD crp ⌬PrpoS183::TT araC PBAD rpoS ⌬Pcrp527::TT araC PBAD crp

Reference or derivation

63

␹3761 ␹8848 ␹3761 ␹9273 ␹3761 ␹9269 ␹3761 ␹8956 ␹3761 ␹8918 ␹3761 ␹9021 ␹3761 ␹9371 ␹3761 ␹3761 ␹8848 ␹9273 ␹9269 ␹9021 ␹8918 ␹8956

a In the descriptions of the genotype, TT is transcription terminator, P stands for promoter, and the subscripted number refers to a composite deletion and insertion of the indicated gene.

Cell biology. The ability of various constructed Salmonella strains to attach to, invade into, and survive in various murine and human epithelial and/or macrophage cell lines was quantitated by well-established methods (16, 22) that we have used routinely. Animal experimentation. BALB/c and C57BL/6 female mice, 6 to 8 weeks of age, were used for most experiments. Mice were held in quarantine for 1 week before use in experiments. They were deprived of food and water 6 h before oral immunization. No bicarbonate was administered. Food and water were returned 30 min after immunization. Candidate vaccine strains were quantitatively enumerated in various tissues as a function of time after inoculation (14, 27). The inoculation procedures were the same as in the immunization studies. All animals were housed in biosafety level 2 containment with filter bonnet-covered cages. If high immunogenicity was observed in initial tests after primary immunization, subsequent studies were done to determine the lowest level of vaccine inoculum to induce a significant protective immune response to oral or intraperitoneal challenge with the wild-type serovar Typhimurium UK-1 parental strain ␹3761. All animal protocols were approved by the Arizona State University IACUC and complied with all standards and policies of the American Association for Accreditation of Laboratory Animal Care.

RESULTS Construction of deletion-insertion mutations to achieve regulated delayed attenuation. We describe four means to permit a regulated delayed attenuation phenotype so that strains at the time of oral inoculation exhibit nearly wild-type attributes for survival and colonization of lymphoid tissues and become avirulent after 5 to 10 cell divisions. The means to achieve regulated delayed attenuation rely on using an araC PBAD activator-promoter that is more tightly regulated by arabinose (39) than the original sequence from Escherichia coli B/r strain (31). We deleted the promoter, including all se-

quences that interact with activator or repressor proteins, for the fur, phoPQ, rpoS, and crp genes, and substituted the improved araC PBAD cassette (39) to yield Salmonella strains with the ⌬Pfur33::TT araC PBAD fur, ⌬PphoPQ107::TT araC PBAD phoPQ, ⌬PrpoS183::TT araC PBAD rpoS, and ⌬Pcrp527::TT araC PBAD crp deletion-insertion mutations (where P stands for promoter and the subscripted number refers to a composite deletion and insertion of the indicated gene). The suicide vectors used to generate these four deletion-insertion mutations depicted in Fig. 1a to d are listed in Table 2. We have included a strong phage-derived TT at the C-terminal end of the araC gene in all these constructions since its transcription in the presence of arabinose could often lead to altered overexpression of downstream adjacent genes with the same transcriptional orientation as the araC gene or to diminished expression when the downstream adjacent gene is in opposite orientation, resulting in synthesis of antisense mRNA from ParaC. Phenotypic characterization of mutant strains. Growth of these mutant strains in the presence of arabinose leads to transcription of the fur, phoPQ, rpoS, and/or crp genes, but gene expression ceases in the absence of arabinose. These activities can be readily observed by appropriate tests. Thus, ␹9021 with the ⌬Pcrp527::TT araC PBAD crp deletion-insertion mutation can ferment maltose only when grown in the presence of arabinose and not in the absence of arabinose, as revealed by streaking cultures on MacConkey maltose agar without and with 0.2% arabinose (Fig. 2a). Similarly, ␹8848 with the ⌬Pfur33::TT araC PBAD fur and ␹9107 with ⌬Pfur33::TT

1074

CURTISS ET AL.

INFECT. IMMUN.

fur

phoPQ

FIG. 1. Deletion-insertion mutations resulting in arabinose-regulated virulence traits and deletion mutations altering arabinose metabolism and uptake. Strains are as identified on the panels.

araC PBAD fur and ⌬Pcrp527::TT araC PBAD crp mutations reveal siderophore production when streaked on CAS plates without arabinose and no siderophore production when grown in the presence of arabinose (Fig. 2b). ␹8918 with the

⌬PphoPQ107::TT araC PBAD phoPQ and ␹9108 with the ⌬PphoPQ107::TT araC PBAD phoPQ and ⌬Pcrp527::TT araC PBAD crp mutations when streaked on X-P plates without and with 0.2% arabinose reveal acid phosphatase activity due to

VOL. 77, 2009

REGULATED DELAYED ATTENUATION IN VIVO TABLE 2. Plasmids

Name

Genotype and descriptiona

Derivation or source

pRE112 pMEG-375 pYA3485 pYA3599 pYA3722 pYA3832 pYA3735 pYA3723 pYA4180 pYA4181 pYA4062 pYA4109 pYA4345 pYA4344 pYA4343

sacB mobRP4 R6K ori Cmr sacRB mobRP4 R6K ori Cmr Apr ⌬araE25 ⌬araBAD23 ⌬Pfur33::TT araC PBAD fur ⌬Pcrp527::TT araC PBAD crp ⌬PrpoS183::TT araC PBAD rpoS ⌬PphoPQ107::TT araC PBAD phoPQ ⌬Pfur77::TT araC PBAD fur ⌬Pfur81::TT araC PBAD fur ⌬PphoPQ173::TT araC PBAD phoPQ ⌬PphoPQ177::TT araC PBAD phoPQ ⌬PphoPQ174::TT araC PBAD phoPQ ⌬PphoPQ175::TT araC PBAD phoPQ ⌬PphoPQ176::TT araC PBAD phoPQ

Megan Health Megan Health pMEG375 pMEG375 pMEG375 pRE112 pRE112 pRE112 pRE112 pRE112 pRE112 pRE112 pRE112 pRE112 pRE112

expression of the PhoP-activated phoN gene only when grown in the presence of arabinose (Fig. 2c). ␹8956 with the ⌬PrpoS183::TT araC PBAD rpoS and ␹9064 with the ⌬PrpoS183::TT araC PBAD rpoS and ⌬Pcrp527::TT araC PBAD crp mutations reveal glycogen accumulation when streaked on glycogen indicator agar with 0.2% arabinose and sprayed with iodine indicator solution (Fig. 2d). The presence or absence of RpoS in these strains can also be revealed by adding hydrogen peroxide to cultures to detect the activity of the RpoS-dependent catalase, KatE (9, 47, 53), when arabinose is present during strain growth. Since Crp positively enhances transcription from PBAD such that transcription is reduced 10-fold in the absence of Crp (45), the inclusion of the ⌬Pcrp527::TT araC

1075

PBAD crp mutation with other araC PBAD-regulated genes causes a tighter cessation of transcription in the absence of arabinose. This is seen by close examination of the photographs in Fig. 2. Thus, ␹9108 produces less acid phosphatase than ␹8918, and ␹9064 accumulates less glycogen than ␹8956. For this reason, the ⌬Pcrp527::TT araC PBAD crp mutation is included in all vaccine strains when araC PBAD is used to regulate virulence genes. In regard to differential plating on LB agar with and without 0.5% bile salts, there were no significant differences among all the strains with the ⌬Pfur33::TT araC PBAD fur, ⌬Pfur71::TT araC PBAD fur, ⌬Pfur81::TT araC PBAD fur, ⌬PrpoS183::TT araC PBAD rpoS, ⌬Pcrp527::TT araC PBAD crp, ⌬PphoPQ107::TT araC PBAD phoPQ, ⌬PphoPQ173::TT araC PBAD phoPQ, or ⌬PphoPQ177::TT araC PBAD phoPQ mutation in combination with the ⌬araBAD23 mutation. This was true whether strains were grown in LB broth or nutrient broth with 0.0, 0.05, or 0.2% arabinose. The wild-type strain ␹3761 was used as the control. In regard to acid stress, ⌬araBAD23 strains with the ⌬Pfur33:: TT araC PBAD fur, ⌬Pfur71::TT araC PBAD fur, ⌬Pfur81::TT araC PBAD fur, ⌬PrpoS183::TT araC PBAD rpoS, and ⌬PphoPQ173::TT araC PBAD phoPQ mutations were sensitive to an acid stress of pH 4.5 when grown in nutrient broth with no arabinose but were resistant to immediate exposure to pH 4.5 when cultivated in medium with either 0.05 or 0.2% arabinose. Strains with other araC PBAD-regulated genes were as resistant or more resistant to the pH 4.5 stress than the wild-type strain ␹3761. Studies on attachment and invasion of mutant strains depending on growth medium using Int-407 cells were somewhat variable, with no clear pattern dependent on presence or ab-

FIG. 2. Phenotypes of strains with deletion-insertion mutations to enable arabinose-dependent expression of virulence traits. (a) ␹9021 with the ⌬Pcrp527::TT araC PBAD crp mutation streaked on MacConkey maltose agar without and with 0.2% arabinose. (b) ␹8848 with the ⌬Pfur33::TT araC PBAD fur and ␹9107 with the ⌬Pfur33::TT araC PBAD fur and ⌬Pcrp527::TT araC PBAD crp mutations spotted on CAS agar plates without and with 0.2% arabinose to visualize siderophore production. (c) ␹8918 with the ⌬PphoPQ107::TT araC PBAD phoPQ and ␹9108 with the ⌬PphoPQ107::TT araC PBAD phoPQ and ⌬Pcrp527::TT araC PBAD crp mutations streaked on X-P plates without and with 0.2% arabinose to reveal acid phosphatase activity. (d) ␹8956 with the ⌬PrpoS183::TT araC PBAD rpoS and ␹9064 with the ⌬PrpoS183::TT araC PBAD rpoS and ⌬Pcrp527::TT araC PBAD crp mutations streaked on glycogen indicator agar without and with 0.2% arabinose and sprayed with iodine indicator solution.

1076

CURTISS ET AL.

FIG. 3. Stability of Crp, Fur, RpoS, and PhoP proteins during incubation of cultures induced for expression of these proteins prior to addition of 50 ␮g/ml chloramphenicol of culture. Rabbit antibodies raised against His-tagged Crp, Fur, RpoS, and PhoP were used for Western blot analyses. ␹9021 (⌬Pcrp527), ␹8848 (⌬Pfur33), ␹8956 (⌬PrpoS183), and ␹8918 (⌬PphoPQ107) were grown in LB broth with 0.2% arabinose for these studies.

sence of arabinose. In view of the primary objective to discern differences in immunogenicity dependent on genotype and growth conditions, we focused most attention on animal studies, as described below, that have provided a clear indication of differences due to genotype and the growth conditions used to prepare strains for inoculation into mice. Stability of Crp, Fur, RpoS, and PhoP proteins and their decline during growth in the absence of arabinose. Growth of strains with araC PBAD-regulated genes in the presence of arabinose results in acid production that can cause cessation of growth. We have therefore included the ⌬araBAD23 mutation (Fig. 1i) that prevents use of arabinose (5, 62). Inclusion of this mutation also prevents breakdown of arabinose retained in the cell cytoplasm at the time of oral immunization, and inclusion of the ⌬araE25 mutation (Fig. 1j) that enhances retention of arabinose (35, 48) further delays cessation in expression of araC PBAD-regulated genes for an additional cell division or so. The suicide vectors for introducing the ⌬araBAD23 and ⌬araE25 mutations are listed in Table 2. The stability of virulence gene products in strains with each of the araC PBAD-regulated virulence genes was determined by growing cultures to an optical density at 600 nm (OD600) of 0.8 in LB broth with 0.2% arabinose and then adding 50 ␮g chloramphenicol/ml (43) for Crp, Fur, and PhoP and 200 ␮g chloramphenicol/ml for RpoS (43) to arrest further protein synthesis. As can be seen by the results presented in Fig. 3, the Crp, Fur, and PhoP proteins are very stable and not subject to breakdown, whereas the RpoS protein displays no stability in the log phase (59). However, the RpoS protein seemed to be stable when 50 ␮g/ml chloramphenicol was added to saturated overnight stationery phase cultures (data not shown). The mutant strains were also grown in nutrient broth with 0.2% arabinose to an OD600 of 0.8 and then diluted 1:4 into nutrient broth with no added arabinose; these 1:4 dilutions were continued after each culture until cultures again reached an OD600 of 0.8. We observed no significant reductions in the amounts of Crp, Fur, and PhoP proteins until a final dilution of 1:16, with an arabinose concentration of 0.0125%, or until a final dilution of 1:64, with an arabinose concentration of 0.003125%. There-

INFECT. IMMUN.

FIG. 4. Decrease in amounts of Crp, Fur, RpoS, and PhoP proteins as a consequence of growth of ␹9021 (⌬Pcrp527), ␹8848 (⌬Pfur33), ␹8956 (⌬PrpoS183), and ␹8918 (⌬PphoPQ107) in the absence of arabinose. The same bacterial strains as used for the results shown in Fig. 3 were grown in nutrient broth with 0.2% arabinose, and at the commencement of sampling to measure the amounts of proteins, the cultures were diluted 1:4 into prewarmed nutrient broth lacking arabinose. Rabbit antibodies raised against His-tagged Crp, Fur, RpoS, and PhoP were used for Western blot analyses. Some synthesis of three of the four proteins continued until after the third 1:4 dilution when the arabinose concentration was 0.003%.

after, the amounts of the proteins decreased by a factor of 4 for each subsequent 1:4 dilution of the culture. In the case of RpoS protein, we observed a significant amount of reduction after a dilution of 1:4 with an arabinose concentration of 0.05% (Fig. 4). Such a rapid decline in RpoS was not observed when cultures of ␹8956 were grown to an OD600 of 2.0 prior to making the successive 1:4 dilutions (data not shown). The decline in the amounts of these proteins in vivo would be expected to be more accelerated since there is no arabinose present in tissues upon invasion of Salmonella into the GALT (39). In other experiments, strains grown in nutrient broth with 0.2% arabinose were sedimented by centrifugation and resuspended at a density one-fourth of the original culture. In this case after growth to the original density, the amounts of each of the four virulence gene proteins was three to four times less than in the culture grown with arabinose (data not shown). In other experiments, we determined that the levels of Fur, PhoP, RpoS, and Crp synthesis were nearly the same when mutant cultures were grown in LB broth with either 0.05% or 0.2% arabinose (data not shown). Attenuation of mutant strains in orally immunized female BALB/c mice. We evaluated levels of attenuation in serovar Typhimurium UK-1 strains with different araC PBAD-regulated virulence genes by oral inoculation of female BALB/c mice with doses approximating 107, 108, and 109 CFU from cultures grown in LB broth with 0.0, 0.05, and 0.2% arabinose. It should be noted that LB broth contains arabinose in the yeast extract at a concentration equivalent to 0.003% based on mass spectrometry analysis. The collective results presented in Table 3 indicate that the strains with the ⌬PphoPQ107::TT araC PBAD phoPQ, ⌬PrpoS183::TT araC PBAD rpoS, and ⌬Pcrp527::TT araC PBAD crp deletion-insertion mutations were highly attenuated, whereas the strain with the ⌬Pfur33::TT araC PBAD fur mutation was less attenuated. In this regard, we noted a higher level of attenuation when ␹8848 was grown in LB broth with no added arabinose and a greater virulence when ␹8848 was

VOL. 77, 2009


TABLE 3. Attenuation of mutant strains in orally immunized female BALB/c micea Strain

Genotype

Dose range (CFU)

No. of survivors/ total no. of mice

% Survivors

␹8848 ␹8918 ␹8956 ␹9021

⌬Pfur33 ⌬PphoPQ107 ⌬PrpoS183 ⌬Pcrp527

9.0 ⫻ 106–2.2 ⫻ 109 9.0 ⫻ 106–1.2 ⫻ 109 9.4 ⫻ 106–1.5 ⫻ 109 9.5 ⫻ 106–1.5 ⫻ 109

138/189 182/185 179/184 163/164

73.0 98.4 97.3 99.4

a Mice were 7 to 8 weeks of age. Bacterial strains were grown in LB broth with 0, 0.05, or 0.2% arabinose that did not have a significant effect on levels of attenuation on strains with the ⌬PphoPQ107::TT araC PBAD phoPQ, ⌬PrpoS183::TT araC PBAD rpoS, and ⌬Pcrp527::TT araC PBAD crp deletion-insertion mutations but did affect the results for ␹8848 with the ⌬Pfur33::TT araC PBAD fur mutation (see Results for an explanation of the notation).

grown in LB broth with 0.2% arabinose. We address the basis for this observation later in this report. It is evident, however, from the collective results (Table 3) that attenuation develops as the products of the fur, phoPQ, rpoS, and/or crp genes are diluted at each cell division in vivo. Abilities of orally administered strains with araC PBADregulated virulence genes to induce protective immunity to oral challenge with wild-type serovar Typhimurium UK-1. Strains with each of the araC PBAD-regulated virulence genes were next evaluated for induction of protective immunity to challenge with the highly virulent serovar Typhimurium UK-1 strain ␹3761 (oral 50% lethal dose of 1.2 ⫻ 104 CFU). The results shown in Table 4 reveal that ␹8848 with the ⌬Pfur33::TT araC PBAD fur mutation displayed some virulence even at low doses when the strain was grown in LB broth with 0.2% arabinose. However, for immunizing doses of 107 CFU and higher, 100% of the survivors developed protective immunity to challenges with 108 and 109 CFU doses of ␹3761. Thus, the ⌬Pfur33::TT araC PBAD fur mutation, while displaying moderate attenuation, is highly immunogenic. This is a very important attribute of an attenuating mutation to include in a vaccine strain. We had previously reported (15) that ␹8848 with the ⌬Pfur33::TT araC PBAD fur mutation was completely attenuated even at high (109 CFU) doses when grown in LB broth

1077

with no added arabinose. This observation implies that production of too much Fur protein may diminish attenuation. The results shown in Table 5 reveal that ␹8918 with the ⌬PphoPQ107::TT araC PBAD phoPQ deletion-insertion mutation is very attenuated but displays more moderate immunogenicity in regard to inducing protection against challenge with ␹3761. These results suggest that some of the attenuation may be due to a reduced ability of ␹8918 to effectively colonize lymphoid tissues, quite possibly due to the overexpression of the phoPQ genes when ␹8918 is grown in LB broth with 0.2% arabinose. In accord with this expectation, ␹8918 is better able to colonize Peyer’s patches, mesenteric lymph nodes, and spleens in orally immunized mice when the strain is grown in LB broth without added arabinose than when grown in LB broth with 0.2% arabinose (data not shown). Nevertheless, ␹8918 is still less capable of colonizing these lymphoid tissues than ␹9021 with the ⌬Pcrp527::TT araC PBAD crp deletion-insertion mutation, which colonizes equally well independent of the arabinose concentration in the LB broth. This undesirable attribute of the ⌬PphoPQ107::TT araC PBAD phoPQ mutation will be addressed later in the text. The results shown in Table 6 confirm the oral avirulence of ␹8956 with the ⌬PrpoS183::TT araC PBAD rpoS deletion-insertion mutation. However, the two experiments gave very different results on the ability of this strain to induce protective immunity to oral challenge with wild-type serovar Typhimurium. We therefore repeated the experiment; with oral doses of ␹8956 (⌬PrpoS183) of 1.4 ⫻107, 1.4 ⫻ 108, and 1.4 ⫻ 109 CFU, there were 15 survivors at each dose. After a subsequent challenge with 3.1 ⫻ 109 CFU of ␹3761, we observed 13, 13, and 14 survivors, respectively, out of 15 mice challenged. It thus appears that the data in the second experiment shown in Table 6 are more indicative of the correct attenuating and immunogenic phenotypes. We have no objective basis to discard the data from the first experiment as all three experiments were done by the same individual many months apart. We also observed no differences in results when ␹8956 (⌬PrpoS183) was grown in LB broth with or without arabinose. The results shown in Table 7 indicate that ␹9021 with

TABLE 4. Oral immunization of mice with ␹8848 (⌬Pfur33) and with survivors challenged orally with wild-type ␹3761 30 days latera Immunization data Dose (CFU) Expt 1

Expt 2

Expt 1

Expt 1

6/10

4/10

1.1 ⫻ 108

7/10

7/10

9.0 ⫻ 106

1.1 ⫻ 107

7/10

5/10

9.0 ⫻ 105

1.1 ⫻ 106

5/10

8/10

9.0 ⫻ 104

1.1 ⫻ 105

10/10

7/10

1.0 ⫻ 10 1.0 ⫻ 108 1.0 ⫻ 109 1.0 ⫻ 108 1.0 ⫻ 109 1.0 ⫻ 108 1.0 ⫻ 109 1.0 ⫻ 108 1.0 ⫻ 109 1.0 ⫻ 108

9.0 ⫻ 10

1.1 ⫻ 10

9.0 ⫻ 107

Total (all doses) Total (107–109 CFU doses)

No. of survivors/total no. of mice (%)

Dose (CFU)

Expt 2

9

8

a

Challenge data No. of survivors/total no. of mice

Expt 2 9

66/100 36/60

Female BALB/c mice were 6 to 8 weeks of age. ␹8848 was grown in LB broth with 0.2% arabinose.

1.5 ⫻ 10 1.5 ⫻ 108 1.5 ⫻ 109 1.5 ⫻ 108 1.5 ⫻ 109 1.5 ⫻ 108 1.5 ⫻ 109 1.5 ⫻ 108 1.5 ⫻ 109 1.5 ⫻ 108 9

Expt 1

Expt 2

3/3 3/3 2/2 5/5 4/4 3/3 1/2 0/3 0/5 0/5

2/2 2/2 4/4 3/3 2/2 3/3 1/4 2/4 2/4 3/3 45/66 (68.2) 36/36 (100)

1078

CURTISS ET AL.

INFECT. IMMUN.

TABLE 5. Oral immunization of mice with ␹8918 (⌬PphoPQ107) and with survivors challenged orally with wild-type ␹3761 30 days latera Immunization data No. of survivors/total no. of mice

Dose (CFU) Expt 1

Challenge data

Expt 2

Expt 1

Expt 2

Expt 1

10/10

9/10

1.2 ⫻ 108

10/10

10/10

9.0 ⫻ 10

7

1.2 ⫻ 10

10/10

9/10

9.0 ⫻ 105

1.2 ⫻ 106

10/10

10/10

9.0 ⫻ 10

1.2 ⫻ 10

10/10

10/10

1.0 ⫻ 10 1.0 ⫻ 108 1.0 ⫻ 109 1.0 ⫻ 108 1.0 ⫻ 109 1.0 ⫻ 108 1.0 ⫻ 109 1.0 ⫻ 108 1.0 ⫻ 109 1.0 ⫻ 108

9.0 ⫻ 10

9

1.2 ⫻ 10

9.0 ⫻ 107

8

6

4

5

Total (all doses) Total (107–109 CFU doses) a


Dose (CFU) Expt 2 9

1.5 ⫻ 10 1.5 ⫻ 108 1.5 ⫻ 109 1.5 ⫻ 108 1.5 ⫻ 109 1.5 ⫻ 108 1.5 ⫻ 109 1.5 ⫻ 108 1.5 ⫻ 109 1.5 ⫻ 108 9

Expt 1

Expt 2

4/5 4/5 3/5 3/5 2/5 2/5 3/5 0/5 0/5 0/5

5/5 4/4 4/5 5/5 1/4 2/5 0/5 3/5 0/5 0/5

98/100 58/60

45/98 (45.9) 39/58 (67.2)


the ⌬Pcrp527::TT araC PBAD crp deletion-insertion mutation is both highly attenuated and also very immunogenic. Neither of these attributes was altered when the strain was grown in LB broth with or without arabinose. Alterations in strains with the ⌬Pfur::TT araC PBAD fur and ⌬PphoPQ::TT araC PBAD phoPQ deletion-insertion mutations to increase the attenuation of the former and increase the immunogenicity of the latter. As noted above, ␹8848 with the ⌬Pfur33::TT araC PBAD fur mutation was more attenuated when grown in LB broth without arabinose and more virulent when grown in LB broth with 0.2% arabinose prior to oral inoculation of mice. This implied that overproduction of Fur, which would require more cell divisions in vivo to dilute out, reduced attenuation without adversely altering immunogenicity in mice surviving immunization. We therefore constructed two derivatives in which the ATG start codon for the fur gene was changed to GTG, and in one of these we also changed the SD sequence from AGGA to AAGG. The structure

of these two mutations, ⌬Pfur77::TT araC PBAD fur and ⌬Pfur81::TT araC PBAD fur, are diagrammed in Fig. 1e and f. ␹9273 with the ⌬Pfur77::TT araC PBAD fur mutation and ␹9269 with the ⌬Pfur81::TT araC PBAD fur mutation both synthesize much less Fur, as revealed by Western blot analysis, when grown in LB broth with 0.2% arabinose than does ␹8848 with the ⌬Pfur33::TT araC PBAD fur mutation (data not shown). It was also noted above that the immunogenicity of ␹8918 with the ⌬PphoPQ107::TT araC PBAD phoPQ mutation was decreased when the strain was grown in LB broth with 0.2% arabinose although its attenuation was independent of the arabinose concentration in LB broth. This implied that overproduction of PhoP and/or PhoQ decreased induction of immunity to challenge. This inference was also supported by studies that demonstrated that ␹8918 was less able to colonize Peyer’s patches, mesenteric lymph nodes, and spleen when the strain was grown in LB broth with 0.2% arabinose than when grown with no added arabinose. We therefore constructed two

TABLE 6. Oral immunization of mice with ␹8956 (⌬PrpoS183) and with survivors challenged orally with wild-type ␹3761 30 days latera Immunization data

Challenge data No. of survivors/total no. of mice

Dose (CFU)


Dose (CFU)

Expt 1

Expt 2

Expt 1

Expt 2

Expt 1

Expt 2

Expt 1

Expt 2b

9.4 ⫻ 108

1.5 ⫻ 109

9/10

9/10

9.4 ⫻ 107

1.5 ⫻ 108

10/10

10/10

9.4 ⫻ 106

1.5 ⫻ 107

10/10

9/10

9.4 ⫻ 105

1.5 ⫻ 106

10/10

10/10

9.4 ⫻ 104

1.5 ⫻ 105

10/10

10/10

1.0 ⫻ 109 1.0 ⫻ 108 1.0 ⫻ 109 1.0 ⫻ 108 1.0 ⫻ 109 1.0 ⫻ 108 1.0 ⫻ 109 1.0 ⫻ 108 1.0 ⫻ 109 1.0 ⫻ 108

1.5 ⫻ 109 1.5 ⫻ 108 1.5 ⫻ 109 1.5 ⫻ 108 1.5 ⫻ 109 1.5 ⫻ 108 1.5 ⫻ 109 1.5 ⫻ 108 1.5 ⫻ 109 1.5 ⫻ 108

0/4 1/5 0/5 0/5 2/5 0/5 0/5 0/5 0/5 0/5

4/4 5/5 5/5 3/5 5/5 2/4 4/5 2/5 0/5 0/5

Total (all doses) Total (107–109 CFU doses)

97/100 57/60

33/97 (34.0) 27/57 (47.4)

Female BALB/c mice were 6 to 8 weeks of age. ␹8956 was grown in LB broth with 0.2% arabinose. The result of experiment 2 was replicated in a third experiment described in the text. It is thus likely that the results in experiment 1 are a chance occurrence to be expected once out of 20 experimental repeats. a b

VOL. 77, 2009


1079

TABLE 7. Oral immunization of mice with ␹9021 (⌬Pcrp527) and with survivors challenged orally with wild-type ␹3761 30 days latera Immunization data No. of survivors/total no. of mice

Dose (CFU) Expt 1

Challenge data

Expt 2

Expt 1

Expt 2

Expt 1

10/10

10/10

1.6 ⫻ 108

10/10

10/10

9.5 ⫻ 10

7

1.6 ⫻ 10

10/10

10/10

9.5 ⫻ 105

1.6 ⫻ 106

10/10

9/10

9.5 ⫻ 10

1.6 ⫻ 10

10/10

10/10

1.0 ⫻ 10 1.0 ⫻ 108 1.0 ⫻ 109 1.0 ⫻ 108 1.0 ⫻ 109 1.0 ⫻ 108 1.0 ⫻ 109 1.0 ⫻ 108 1.0 ⫻ 109 1.0 ⫻ 108

9.5 ⫻ 10

9

1.6 ⫻ 10

9.5 ⫻ 107

8

6

4

5

Total (all doses) Total (107–109 CFU doses) a


Dose (CFU) Expt 2 9

99/100 60/60

1.5 ⫻ 10 1.5 ⫻ 108 1.5 ⫻ 109 1.5 ⫻ 108 1.5 ⫻ 109 1.5 ⫻ 108 1.5 ⫻ 109 1.5 ⫻ 108 9

Expt 1

Expt 2

5/5 5/5 5/5 5/5 4/5 5/5 5/5 3/5 4/5 3/5

5/5 5/5 5/5 5/5 5/5 5/5 3/5 2/4

78/89 (87.6) 59/60 (98.3)


derivatives in which the ATG start codon for the phoP gene was changed to GTG, and in one of these we also changed the SD sequence from AGGA to AAGG. The structure of these two mutations, ⌬PphoPQ173::TT araC PBAD phoPQ and ⌬PphoPQ177::TT araC PBAD phoPQ, are diagrammed in Fig. 1g and h. ␹9382 with the ⌬PphoPQ173::TT araC PBAD phoPQ mutation and ␹9383 with the ⌬PphoPQ177::TT araC PBAD phoPQ mutation both synthesize much less PhoP, as revealed by Western blot analysis, when grown in LB broth with 0.2% arabinose than does ␹8918 with the ⌬PphoPQ107::TT araC PBAD phoPQ mutation (data not shown). Table 8 contains results that demonstrate the high immunogenicity of ␹9273 with the ⌬Pfur77::TT araC PBAD fur mutation and ␹9269 with the ⌬Pfur81::TT araC PBAD fur mutation, the latter of which demonstrates much better attenuation when

TABLE 8. Oral immunization of mice with strains with modified ⌬Pfur and ⌬PphoPQ mutations and with survivors challenged orally with wild-type ␹3761 30 days latera Immunization data Strainb

Genotype

␹9273

⌬Pfur77

␹9269

⌬Pfur81

␹9382

⌬PphoPQ173

␹9383

⌬PphoPQ177

Challenge data

Dose (CFU)


Dose (CFU)


1.5 ⫻ 109 1.7 ⫻ 109 1.0 ⫻ 109 1.0 ⫻ 108 1.8 ⫻ 109 1.7 ⫻ 109 1.0 ⫻ 109 1.0 ⫻ 108 1.0 ⫻ 107 1.1 ⫻ 109 1.1 ⫻ 108 1.1 ⫻ 107

6/10 6/10 11/15 15/15 10/10 19/20 15/15 15/15 15/15 15/15 15/15 15/15

1.7 ⫻ 109 1.6 ⫻ 109 8.7 ⫻ 108 8.7 ⫻ 108 1.3 ⫻ 109 1.6 ⫻ 109 1.8 ⫻ 109 1.8 ⫻ 109 1.8 ⫻ 109 1.8 ⫻ 109 1.8 ⫻ 109 1.8 ⫻ 109

6/6 6/6 11/11 15/15 10/10 19/19 11/15 11/15 12/15 10/15 11/15 13/15

a Female BALB/c mice were six to eight weeks of age. Strains were grown in LB broth with no added arabinose or with 0.05% or 0.2% arabinose with no significant differences noted. b ␹9382 and ␹9383 have in addition to the ⌬PphoPQ insertion-deletion mutations the ⌬araBAD23 deletion (Table 1).

grown in LB broth with 0.2% arabinose. The data in Table 8 also indicate that both ␹9382 with the ⌬PphoPQ173::TT araC PBAD phoPQ mutation and ␹9383 with the ⌬PphoPQ177::TT araC PBAD phoPQ mutation are completely attenuated when grown in LB broth with 0.2% arabinose and display essentially the same immunogenicity, which is much improved over that exhibited by ␹8918 with the ⌬PphoPQ107::TT araC PBAD phoPQ mutation when it is grown in LB broth with 0.2% arabinose. Abilities of intraperitoneally administered strains with araC PBAD-regulated virulence genes to induce protective immunity to oral challenge with wild-type serovar Typhimurium UK-1. Although we designed our vaccines for oral administration, we deemed it worthwhile to determine if strains with these mutations when administered intraperitoneally would also display attenuation and induce immunity to challenge with orally administered wild-type ␹3761. The serovar Typhimurium UK-1 strain ␹3761 has an 50% lethal dose by the intraperitoneal route of less than 10 CFU. Table 9 demonstrates that strains with ⌬Pfur::TT araC PBAD fur mutations retain considerable virulence by this route of administration although ␹9269 with the ⌬Pfur81::TT araC PBAD fur mutation displays the highest attenuation of the three strains evaluated and yet induces complete protective immunity to all survivors when they are challenged with about 109 CFU of ␹3761. ␹8918 with the ⌬PphoPQ107::TT araC PBAD phoPQ mutation displays fairly good attenuation by this route and moderate immunogenicity. On the other hand, ␹8956 with the ⌬PrpoS183::TT araC PBAD rpoS mutation and ␹9021 with the ⌬Pcrp527::TT araC PBAD crp mutation are the most attenuated and induce a very high level of protective immunity when delivered at i.p doses in the 102 to 104 CFU range (Table 9). Enhanced control over araC PBAD-regulated virulence genes in vivo by inclusion of the ⌬Pcrp527::TT araC PBAD crp mutation. Maximum levels of transcription of genes regulated by the araC PBAD system require not only arabinose to interact with the AraC protein but also the Crp protein (46, 50). We thus will always include the ⌬Pcrp527::TT araC PBAD crp mutation in vaccine strains whenever other araC PBAD-regulated genes are included. The benefit of this addition is readily observed by the

1080

CURTISS ET AL.

INFECT. IMMUN.

TABLE 9. Intraperitoneal immunization of mice with strains with various deletion-insertion mutations conferring regulated delayed oral attenuation and with survivors orally challenged with wild-type ␹3761 30 days latera Immunization data Strainb

Genotype

␹8848

⌬Pfur33

␹9273

⌬Pfur77

␹9269

⌬Pfur81

␹8918

⌬PphoPQ107

␹8956

⌬PrpoS183

␹9021

⌬Pcrp527

a b

Dose (CFU)


1.2 ⫻ 104 1.2 ⫻ 103 1.2 ⫻ 102 1.2 ⫻ 101 1.6 ⫻ 104 1.6 ⫻ 103 1.6 ⫻ 102 1.6 ⫻ 101 1.7 ⫻ 104 1.7 ⫻ 103 1.7 ⫻ 102 1.7 ⫻ 101 1.2 ⫻ 105 9.6 ⫻ 104 1.2 ⫻ 104 9.6 ⫻ 103 1.3 ⫻ 103 9.6 ⫻ 102 1.2 ⫻ 102 1.5 ⫻ 104 1.5 ⫻ 103 1.5 ⫻ 102 1.5 ⫻ 101 1.4 ⫻ 105 1.4 ⫻ 104 1.4 ⫻ 103 1.4 ⫻ 102 1.4 ⫻ 101

0/5 0/5 0/5 0/5 0/5 0/5 0/5 1/5 1/10 7/10 4/10 6/10 0/5 6/10 5/5 8/10 5/5 5/5 5/5 5/5 5/5 5/5 5/5 0/5 4/10 10/10 10/10 9/10

Challenge data Dose (CFU)


1.6 ⫻ 109 1.6 ⫻ 109 1.6 ⫻ 109 1.6 ⫻ 109 1.6 ⫻ 109

1/1 1/1 7/7 4/4 6/6

1.5 ⫻ 109 9.6 ⫻ 108 1.5 ⫻ 109 9.6 ⫻ 108 1.5 ⫻ 109 9.6 ⫻ 108 1.0 ⫻ 109 1.0 ⫻ 109 1.0 ⫻ 109 1.0 ⫻ 109

4/6 5/5 6/8 2/5 2/5 3/5 5/5 5/5 3/5 3/5

1.5 ⫻ 109 1.5 ⫻ 109 1.5 ⫻ 109 1.5 ⫻ 109

4/4 10/10 10/10 8/9

Female BALB/c mice were 6 to 8 weeks of age. All strains were grown in LB broth with 0.2% arabinose.

results previously presented in Fig. 2 that demonstrate this tighter regulation in the absence of arabinose in strains that also have the ⌬Pcrp527::TT araC PBAD crp mutation. This also acts as a backup and should enhance the safety and efficacy of vaccine strains. Means for delay in the in vivo timing of onset of regulated delayed attenuation. As shown by Guzman et al. (31), the inclusion of mutations that abolish utilization of arabinose prolong expression of genes under the control of the araC PBAD system. We therefore can delay onset of attenuation by including ⌬araBAD23, which prevents use of arabinose retained in the cell cytoplasm at the time of oral immunization, and/or ⌬araE25, which enhances retention of arabinose. These mutations are diagrammed in Fig. 1i and j. DISCUSSION We have described four different means to achieve regulated delayed attenuation of serovar Typhimurium strains using araC PBAD regulation of virulence genes such that vaccine strains possessing these deletion-insertion mutations will be better able to withstand the host defense-imposed stresses following oral inoculation. We have also further modified some of these constructs by changing the SD and start codon se-

quences to optimize attenuation and improve immunogenicity. Strains with the optimized ⌬Pfur::TT araC PBAD fur and ⌬PphoPQ::TT araC PBAD phoPQ mutations as well as strains with the ⌬Pcrp527::TT araC PBAD crp mutation and ⌬PrpoS183::TT araC PBAD rpoS mutation were very attenuated and highly immunogenic when grown with the optimal concentration of arabinose and inoculated orally. However, vaccine strains with the ⌬Pfur::TT araC PBAD fur mutations and one of the ⌬PphoPQ::TT araC PBAD phoPQ mutations were less attenuated and significantly less immunogenic when delivered intraperitoneally. This was not the case for strains with either the ⌬Pcrp527::TT araC PBAD crp or ⌬PrpoS183::TT araC PBAD rpoS mutation, which exhibited very good attenuation and high immunogenicity when administered intraperitoneally. Presumably, the combination of two of these means of attenuation would yield strains with enhanced safety and immunogenicity if they were administered intraperitoneally. Although comparative studies with vaccine strains having defined deletion mutations in the fur, phoPQ, rpoS, and crp genes with strains having araC PBAD regulation of the same genes might resolve doubt about the enhanced efficacy of the regulated delayed attenuation strategy, such comparative studies become difficult to justify based on animal use in studies using challenge to the wild-type virulent serovar Typhimurium parent strain. However, such comparisons, to be reported separately, are being made with recombinant vaccine strains that deliver a protective antigen to induce protective immunity to Streptococcus pneumoniae challenge. In addition, we are now including these mutations in strains with multiple attenuating mutations both to investigate their tolerance to acid stress and bile stress, their success in colonizing effector lymphoid tissues, and their ability to induce maximal immune responses to expressed protective antigens encoded on plasmids using the balanced-lethal vector-host systems (23, 54) and to ensure safety in administering the strains to newborn mice. The results of these studies will be reported separately. We are also including some of these mutations in S. enterica serovar Typhi strains to be evaluated soon in human clinical trials. ACKNOWLEDGMENTS We thank our many colleagues for input and suggestions throughout our research enterprise. Their questions and critiques have stimulated us to improve our constructions. We are hopeful that our endeavors will enhance the success of the research undertakings. Our research has been supported by USDA grants 2001-02994 and 03-35204-13748; NIH grants DE06669, AI24533, and AI056289; and grant 37863 from the Bill and Melinda Gates Foundation. REFERENCES 1. Audia, J. P., C. C. Webb, and J. W. Foster. 2001. Breaking through the acid barrier: an orchestrated response to proton stress by enteric bacteria. Int. J. Med. Microbiol. 291:97–106. 2. Bajaj, V., R. L. Lucas, C. Hwang, and C. A. Lee. 1996. Co-ordinate regulation of Salmonella typhimurium invasion genes by environmental and regulatory factors is mediated by control of hilA expression. Mol. Microbiol. 22:703– 714. 3. Bang, I. S., J. P. Audia, Y. K. Park, and J. W. Foster. 2002. Autoinduction of the ompR response regulator by acid shock and control of the Salmonella enterica acid tolerance response. Mol. Microbiol. 44:1235–1250. 4. Bang, I. S., B. H. Kim, J. W. Foster, and Y. K. Park. 2000. OmpR regulates the stationary-phase acid tolerance response of Salmonella enterica serovar typhimurium. J. Bacteriol. 182:2245–2252. 5. Bass, R., L. Heffernan, K. Sweadner, and E. Englesberg. 1976. The site for catabolite deactivation in the L-arabinose BAD operon in Escherichia coli B/r. Arch. Microbiol. 110:135–143.

VOL. 77, 2009 6. Bearson, B. L., L. Wilson, and J. W. Foster. 1998. A low pH-inducible, PhoPQ-dependent acid tolerance response protects Salmonella typhimurium against inorganic acid stress. J. Bacteriol. 180:2409–2417. 7. Bearson, S., B. Bearson, and J. W. Foster. 1997. Acid stress responses in enterobacteria. FEMS Microbiol. Lett. 147:173–180. 8. Bertani, G. 1951. Studies on lysogenesis. I. The mode of phage liberation by lysogenic Escherichia coli. J. Bacteriol. 62:293–300. 9. Buchmeier, N. A., S. J. Libby, Y. Xu, P. C. Loewen, J. Switala, D. G. Guiney, and F. C. Fang. 1995. DNA repair is more important than catalase for Salmonella virulence in mice. J. Clin. Investig. 95:1047–1053. 10. Chen, C. Y., N. A. Buchmeier, S. Libby, F. C. Fang, M. Krause, and D. G. Guiney. 1995. Central regulatory role for the RpoS sigma factor in expression of Salmonella dublin plasmid virulence genes. J. Bacteriol. 177:5303– 5309. 11. Coynault, C., V. Robbe-Saule, and F. Norel. 1996. Virulence and vaccine potential of Salmonella typhimurium mutants deficient in the expression of the RpoS (sigma S) regulon. Mol. Microbiol. 22:149–160. 12. Curtiss, R., III. 1965. Chromosomal aberrations associated with mutations to bacteriophage resistance in Escherichia coli. J. Bacteriol. 89:28–40. 13. Curtiss, R., III, and J. O. Hassan. 1996. Nonrecombinant and recombinant avirulent Salmonella vaccines for poultry. Vet. Immunol. Immunopathol. 54:365–372. 14. Curtiss, R., III, and S. M. Kelly. 1987. Salmonella typhimurium deletion mutants lacking adenylate cyclase and cyclic AMP receptor protein are avirulent and immunogenic. Infect. Immun. 55:3035–3043. 15. Curtiss, R., III, X. Zhang, S. Y. Wanda, H. Y. Kang, V. Konjufca, Y. Li, B. Gunn, S. Wang, G. Scarpellini, and I. S. Lee. 2007. Induction of host immune responses using Salmonella-vectored vaccines, p. 297–313. In K. A. Brogden, F. C. Minion, N. Cornick, T. B. Stanton, Q. Zhang, L. K. Nolan, and M. J. Wannemuehler (ed.), Virulence mechanisms of bacterial pathogens, 4th ed. ASM Press, Washington, DC. 16. Daigle, F., J. E. Graham, and R. Curtiss III. 2001. Identification of Salmonella typhi genes expressed within macrophages by selective capture of transcribed sequences (SCOTS). Mol. Microbiol. 41:1211–1222. 17. de Jonge, R., W. S. Ritmeester, and F. M. van Leusden. 2003. Adaptive responses of Salmonella enterica serovar Typhimurium DT104 and other S. Typhimurium strains and Escherichia coli O157 to low pH environments. J. Appl. Microbiol. 94:625–632. 18. Ernst, R. K., D. M. Dombroski, and J. M. Merrick. 1990. Anaerobiosis, type 1 fimbriae, and growth phase are factors that affect invasion of HEp-2 cells by Salmonella typhimurium. Infect. Immun. 58:2014–2016. 19. Fang, F. C., S. J. Libby, N. A. Buchmeier, P. C. Loewen, J. Switala, J. Harwood, and D. G. Guiney. 1992. The alternative sigma factor katF (rpoS) regulates Salmonella virulence. Proc. Natl. Acad. Sci. USA 89:11978–11982. 20. Foster, J. W. 1999. When protons attack: microbial strategies of acid tolerance. Curr. Opin. Microbiol. 2:170–174. 21. Foster, J. W., and M. P. Spector. 1995. How Salmonella survive against the odds. Annu. Rev. Microbiol. 49:145–174. 22. Galan, J. E., and R. Curtiss III. 1989. Cloning and molecular characterization of genes whose products allow Salmonella typhimurium to penetrate tissue culture cells. Proc. Natl. Acad. Sci. USA 86:6383–6387. 23. Ga ´lan, J. E., K. Nakayama, and R. Curtiss III. 1990. Cloning and characterization of the asd gene of Salmonella typhimurium: use in stable maintenance of recombinant plasmids in Salmonella vaccine strains. Gene 94:29–35. 24. Giannella, R. A., S. A. Broitman, and N. Zamcheck. 1973. Gastric acidity and cholera. Ann. Intern. Med. 78:780. 25. Groisman, E. A. 2001. The pleiotropic two-component regulatory system PhoP-PhoQ. J. Bacteriol. 183:1835–1842. 26. Groisman, E. A., C. Parra-Lopez, M. Salcedo, C. J. Lipps, and F. Heffron. 1992. Resistance to host antimicrobial peptides is necessary for Salmonella virulence. Proc. Natl. Acad. Sci. USA 89:11939–11943. 27. Gulig, P. A., and R. Curtiss III. 1987. Plasmid-associated virulence of Salmonella typhimurium. Infect. Immun. 55:2891–2901. 28. Gulig, P. A., H. Danbara, D. G. Guiney, A. J. Lax, F. Norel, and M. Rhen. 1993. Molecular analysis of spv virulence genes of the Salmonella virulence plasmids. Mol. Microbiol. 7:825–830. 29. Gunn, J. S. 2000. Mechanisms of bacterial resistance and response to bile. Microbes Infect. 2:907–913. 30. Guzman, C. A., S. Borsutzkya, M. Griot-Wenkb, I. C. Metcalfeb, J. Pearmanb, A. Collioudb, D. Favreb, and G. Dietrich. 2006. Vaccines against typhoid fever. Vaccine 24:3804–3811. 31. Guzman, L. M., D. Belin, M. J. Carson, and J. Beckwith. 1995. Tight regulation, modulation, and high-level expression by vectors containing the arabinose PBAD promoter. J. Bacteriol. 177:4121–4130. 32. Hall, H. K., and J. W. Foster. 1996. The role of fur in the acid tolerance response of Salmonella typhimurium is physiologically and genetically separable from its role in iron acquisition. J. Bacteriol. 178:5683–5691. 33. Hengge-Aronis, R., and D. Fischer. 1992. Identification and molecular analysis of glgS, a novel growth-phase-regulated and rpoS-dependent gene involved in glycogen synthesis in Escherichia coli. Mol. Microbiol. 6:1877–1886. 34. Hitchcock, P. J. 1983. Aberrant migration of lipopolysaccharide in sodium


35. 36. 37.

38. 39.

40. 41. 42. 43. 44.

45.

46. 47. 48. 49. 50.

51. 52.

53. 54.

55.

56. 57.

58.

59.

1081

dodecyl sulfate/polyacrylamide gel electrophoresis. Eur. J. Biochem. 133: 685–688. Hogg, R. W. 1977. L-Arabinose transport and the L-arabinose binding protein of Escherichia coli. J. Supramol. Struct. 6:411–417. Hoiseth, S. K., and B. A. Stocker. 1981. Aromatic-dependent Salmonella typhimurium are non-virulent and effective as live vaccines. Nature 291:238– 239. Hopkins, S., J. P. Kraehenbuhl, F. Schodel, A. Potts, D. Peterson, P. de Grandi, and D. Nardelli-Haefliger. 1995. A recombinant Salmonella typhimurium vaccine induces local immunity by four different routes of immunization. Infect. Immun. 63:3279–3286. Kang, H. Y., C. M. Dozois, S. A. Tinge, T. H. Lee, and R. Curtiss III. 2002. Transduction-mediated transfer of unmarked deletion and point mutations through use of counterselectable suicide vectors. J. Bacteriol. 184:307–312. Kong, W., S. Y. Wanda, X. Zhang, W. Bollen, S. A. Tinge, K. L. Roland, and R. Curtiss III. 2008. Regulated programmed lysis of recombinant Salmonella in host tissues to release protective antigens and confer biological containment. Proc. Natl. Acad. Sci. USA 105:9361–9366. Kowarz, L., C. Coynault, V. Robbe-Saule, and F. Norel. 1994. The Salmonella typhimurium katF (rpoS) gene: cloning, nucleotide sequence, and regulation of spvR and spvABCD virulence plasmid genes. J. Bacteriol. 176:6852–6860. Lange, R., and R. Hengge-Aronis. 1991. Identification of a central regulator of stationary-phase gene expression in Escherichia coli. Mol. Microbiol. 5: 49–59. Lee, C. A., B. D. Jones, and S. Falkow. 1992. Identification of a Salmonella typhimurium invasion locus by selection for hyperinvasive mutants. Proc. Natl. Acad. Sci. USA 89:1847–1851. Lee, H. C., and H. D. Bernstein. 2002. Trigger factor retards protein export in Escherichia coli. J. Biol. Chem. 277:43527–43535. Lee, I. S., J. Lin, H. K. Hall, B. Bearson, and J. W. Foster. 1995. The stationary-phase sigma factor ␴S (RpoS) is required for a sustained acid tolerance response in virulent Salmonella typhimurium. Mol. Microbiol. 17: 155–167. Lee, N. L., W. O. Gielow, and R. G. Wallace. 1981. Mechanism of araC autoregulation and the domains of two overlapping promoters, Pc and PBAD, in the L-arabinose regulatory region of Escherichia coli. Proc. Natl. Acad. Sci. USA 78:752–756. Lobell, R. B., and R. F. Schleif. 1991. AraC-DNA looping: orientation and distance-dependent loop breaking by the cyclic AMP receptor protein. J. Mol. Biol. 218:45–54. Loewen, P. C., and B. L. Triggs. 1984. Genetic mapping of katF, a locus that with katE affects the synthesis of a second catalase species in Escherichia coli. J. Bacteriol. 160:668–675. MacPherson, A. J., M. C. Jones-Mortimer, and P. J. Henderson. 1981. Identification of the AraE transport protein of Escherichia coli. Biochem. J. 196:269–283. Makela, P. H., and B. A. Stocker. 1969. How genes determine the structure of the Salmonella lipopolysaccharide. J. Gen. Microbiol. 57:vi. Martin, K., L. Huo, and R. F. Schleif. 1986. The DNA loop model for ara repression: AraC protein occupies the proposed loop sites in vivo and repression-negative mutations lie in these same sites. Proc. Natl. Acad. Sci. USA 83:3654–3658. Merrell, D. S., and A. Camilli. 2002. Acid tolerance of gastrointestinal pathogens. Curr. Opin. Microbiol. 5:51–55. Miller, V. L., and J. J. Mekalanos. 1988. A novel suicide vector and its use in construction of insertion mutations: osmoregulation of outer membrane proteins and virulence determinants in Vibrio cholerae requires toxR. J. Bacteriol. 170:2575–2583. Mulvey, M. R., J. Switala, A. Borys, and P. C. Loewen. 1990. Regulation of transcription of katE and katF in Escherichia coli. J. Bacteriol. 172:6713– 6720. Nakayama, K., S. M. Kelly, and R. Curtiss III. 1988. Construction of an Asd⫹ expression-cloning vector: stable maintenance and high level expression of cloned genes in a Salmonella vaccine strain. Bio/Technology 6:693– 697. Nardelli-Haefliger, D., R. B. Roden, J. Benyacoub, R. Sahli, J. P. Kraehenbuhl, J. T. Schiller, P. Lachat, A. Potts, and P. De Grandi. 1997. Human papillomavirus type 16 virus-like particles expressed in attenuated Salmonella typhimurium elicit mucosal and systemic neutralizing antibodies in mice. Infect. Immun. 65:3328–3336. Nickerson, C. A., and R. Curtiss III. 1997. Role of sigma factor RpoS in initial stages of Salmonella typhimurium infection. Infect. Immun. 65:1814– 1823. Norel, F., V. Robbe-Saule, M. Y. Popoff, and C. Coynault. 1992. The putative sigma factor KatF (RpoS) is required for the transcription of the Salmonella typhimurium virulence gene spvB in Escherichia coli. FEMS Microbiol. Lett. 78:271–276. Pasetti, M. F., T. E. Pickett, M. M. Levine, and M. B. Sztein. 2000. A comparison of immunogenicity and in vivo distribution of Salmonella enterica serovar Typhi and Typhimurium live vector vaccines delivered by mucosal routes in the murine model. Vaccine 18:3208–3213. Peterson, C. N., N. Ruiz, and T. J. Silhavy. 2004. RpoS proteolysis is regu-

1082

60. 61. 62. 63.

64.

65. 66. 67. 68.

CURTISS ET AL.

lated by a mechanism that does not require the SprE (RssB) response regulator phosphorylation site. J. Bacteriol. 186:7403–7410. Prouty, A. M., and J. S. Gunn. 2000. Salmonella enterica serovar Typhimurium invasion is repressed in the presence of bile. Infect. Immun. 68: 6763–6769. Prouty, A. M., J. C. Van Velkinburgh, and J. S. Gunn. 2002. Salmonella enterica serovar Typhimurium resistance to bile: identification and characterization of the tolQRA Cluster. J. Bacteriol. 184:1270–1276. Pueyo, C., and J. Lopez-Barea. 1979. The L-arabinose-resistance test with Salmonella typhimurium strain SV3 selects forward mutations at several ara genes. Mutat. Res. 64:249–258. Roland, K., R. Curtiss III, and D. Sizemore. 1999. Construction and evaluation of a ⌬cya ⌬crp Salmonella typhimurium strain expressing avian pathogenic Escherichia coli O78 LPS as a vaccine to prevent airsacculitis in chickens. Avian Dis. 43:429–441. Romeo, T., and J. Preiss. 1989. Genetic regulation of glycogen biosynthesis in Escherichia coli: in vitro effects of cyclic AMP and guanosine 5⬘-diphosphate 3⬘-diphosphate and analysis of in vivo transcripts. J. Bacteriol. 171: 2773–2782. Sambrook, J., E. F. Fritsch, and T. Maniatis. 1989. Molecular cloning: a laboratory manual, 2nd ed. Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NY. Schmieger, H. 1972. Phage P22-mutants with increased or decreased transduction abilities. Mol. Gen. Genet. 119:75–88. Schmieger, H., and H. Backhaus. 1976. Altered cotransduction frequencies exhibited by HT-mutants of Salmonella-phage P22. Mol. Gen. Genet. 143: 307–309. Schwyn, B., and J. B. Neilands. 1987. Universal chemical assay for the detection and determination of siderophores. Anal. Biochem. 160:47–56.

Editor: R. P. Morrison

INFECT. IMMUN. 69. Stocker, B. A., and P. H. Makela. 1986. Genetic determination of bacterial virulence, with special reference to Salmonella. Curr. Top. Microbiol. Immunol. 124:149–172. 70. Stocker, B. A., and P. H. Makela. 1978. Genetics of the (gram-negative) bacterial surface. Proc. R. Soc. Lond. B 202:5–30. 71. Tsai, C. M., and C. E. Frasch. 1982. A sensitive silver stain for detecting lipopolysaccharides in polyacrylamide gels. Anal. Biochem. 119:115–119. 72. VanCott, J. L., S. N. Chatfield, M. Roberts, D. M. Hone, E. L. Hohmann, D. W. Pascual, M. Yamamoto, H. Kiyono, and J. R. McGhee. 1998. Regulation of host immune responses by modification of Salmonella virulence genes. Nat. Med. 4:1247–1252. 73. van Velkinburgh, J. C., and J. S. Gunn. 1999. PhoP-PhoQ-regulated loci are required for enhanced bile resistance in Salmonella spp. Infect. Immun. 67:1614–1622. 74. Waterman, S. R., and P. L. Small. 1998. Acid-sensitive enteric pathogens are protected from killing under extremely acidic conditions of pH 2.5 when they are inoculated onto certain solid food sources. Appl. Environ. Microbiol. 64:3882–3886. 75. Waterman, S. R., and P. L. Small. 1996. Identification of sigma S-dependent genes associated with the stationary-phase acid-resistance phenotype of Shigella flexneri. Mol. Microbiol. 21:925–940. 76. Wilmes-Riesenberg, M. R., B. Bearson, J. W. Foster, and R. Curtiss III. 1996. Role of acid tolerance response in virulence of Salmonella typhimurium. Infect. Immun. 64:1085–1092. 77. Wosten, M. M., L. F. Kox, S. Chamnongpol, F. C. Soncini, and E. A. Groisman. 2000. A signal transduction system that responds to extracellular iron. Cell 103:113–125.