HeJ mice

Vaccine 22 (2004) 3797–3808

Mucosal immunization with purified flagellin from Salmonella induces systemic and mucosal immune responses in C3H/HeJ mice Lena Strindelius, Malin Filler, Ingvar Sjöholm∗ Department of Pharmacy, Biomedical Centre, Uppsala University, P.O. Box 580, SE-751 23 Uppsala, Sweden Received 2 July 2003; accepted 15 December 2003 Available online 26 April 2004

Abstract This study investigated the immune response elicited in C3H/HeJ mice after oral, parenteral and nasal immunization with purified flagellin from Salmonella enterica serovar Enteritidis alone or conjugated to starch microparticles as adjuvant or together with the uptake-enhancer recombinant cholera toxin B-subunit (rCTB). Systemic (IgM–IgG, IgA, IgG2a, IgG2b, IgG1) and local (s-IgA) humoral immune responses in the mice were analyzed using enzyme-linked immunosorbent assays (ELISA). Primed splenocytes were also stimulated in vitro with flagellin and the supernatants analyzed for cytokine production. Finally, immunized mice were challenged orally with live Salmonella. A high flagellin-specific IgM–IgG response was seen in all groups, especially in mice immunized nasally with flagellin plus rCTB or subcutaneously, but a strong systemic antibody response was also induced when free antigen was given orally. Intranasal or subcutaneous immunization of mice with flagellin plus rCTB or oral immunization with flagellin plus microparticles resulted in a significantly greater mucosal response (higher s-IgA titers in feces) than seen in the control group (P < 0.05). The mucosal IgA responses were significantly correlated with the serum IgA titers. The subclass profile in serum revealed a mixed Th1/Th2-type response, with a predominance of Th1-type, as indicated by the subclass ratio (IgG1/IgG2a + IgG2b). The splenocytes stimulated in vitro produced interferon (IFN)-␥, at levels, which increased with time. The group immunized with flagellin plus rCTB subcutaneously had a relatively higher IFN-␥ response than the other groups. Interleukin (IL)-2 was also produced, especially in mice immunized nasally or subcutaneously with flagellin conjugated to microparticles. However, neither IL-4 nor IL-5 was produced in any of the groups. After oral challenge with live serovar Enteritidis, the groups immunized orally or nasally with free flagellin had significantly lower degree of infection than the control group (P < 0.05). © 2004 Elsevier Ltd. All rights reserved. Keywords: Salmonella; Flagellin; Starch microparticles; rCTB; Mucosal; Vaccine

1. Introduction The Gram-negative, facultative, intracellular bacteria of the genus Salmonella are the cause of a wide range of infectious syndromes, from life-threatening diseases via transitory diarrhea to subclinical infections. The capacity of the different members of the genus to accommodate to the varying conditions of their hosts during the evolution of the species, their intriguing pathogenicity, and their complex interactions with both the innate and adaptive immune systems have been attracting increasing interest among the scientific community. A recent Forum in Immunology (summarized by Kaufmann et al.) [1] and a review by Hughes and Galán [2] describe the current level of understanding of the microbiology and immunology of Salmonella. The ∗ Corresponding author. Tel.: +46-18-471-44-67; fax: +46-18-471-42-23. E-mail address: [email protected] (I. Sjöholm).

0264-410X/$ – see front matter © 2004 Elsevier Ltd. All rights reserved. doi:10.1016/j.vaccine.2003.12.035

different Salmonella serovars utilize various ways of invading the hosts in order to adapt themselves to the immune defense and to proliferate in vivo. These properties of the Salmonella bacteria provide a challenge for those searching for a suitable effective vaccine formulation. It is understood that Salmonella species adhere to and pass through intestinal epithelial cells, primarily the M-cells of the follicle-associated epithelium (FAE) [3,4], and that they also pass through other epithelial cells using the so-called type III secretion system, regulated by the genes forming the Salmonella pathogenicity island-I (SPI-I) (reviewed by Lostroh and Lee) [5]. The initial immune response to Salmonella is controlled by the innate immune system, in which macrophages and polymorphonuclear neutrophils play a major role in controlling the progression of the infection. However, an acquired immune response (comprising both cellular and specific humoral responses) is essential in order to effectively eliminate the bacteria.

3798

L. Strindelius et al. / Vaccine 22 (2004) 3797–3808

We have previously shown that oral immunization of BALB/c mice with secreted antigens from Salmonella enterica serovar Enteritidis conjugated to starch microparticles or administered in free form induces strong local and systemic immune responses [6]. The antigen responsible for most of the antibody response was shown by Western blot analysis to be flagellin, which has recently gained particular attention with regard to its role in the host’s innate and acquired immune response towards Salmonella infections. Toll-like receptors (TLRs), situated on epithelial cells, dendritic cells (DC) and macrophages, have recently been shown to play a central role in inducing the innate immune response by recognizing pathogen-associated molecular patterns and also by sending out alert signals to the adaptive immune system (for review see Kaisho and Akira) [7]. Flagellin binds to TLR5 [8] and activates this receptor on the basolateral side of the epithelium in a human colonic cell line, T84 [9]. Furthermore, in several studies flagellin activated proinflammatory cytokine (e.g. tumour necrosis factor [TNF]-␣, interleukin [IL]-1␤) secretion in human monocytes [10–13], probably through its affiliation for TLR5 [14,15]. In addition, flagellin attracts dendritic cells in human intestinal epithelia cell lines by inducing production of the chemokine CCL20 [16]. Flagellin-specific CD4+ T cell responses develop after oral immunization with attenuated Salmonella in mice [17], and protection against an oral challenge with Salmonella is also provided by subcutaneous immunization of mice with flagellin [18]. Flagellin has also been used as a carrier for heterologous antigens; a significant antibody response against the inserted epitopes was elicited after parenteral administration [19–21]. However, no studies utilizing mucosal administration of purified flagellin have been performed until now. The C3H/HeJ mice used in this study carry a mutation in the TLR4 gene, making them unable to respond to lipopolysaccharide (LPS) [22]. These mice have also been shown to be highly susceptible to infection by Gram-negative bacteria [23]. The aim of this study was to compare the mucosal and systemic immune responses elicited after immunization with flagellin administered by the oral, nasal and parenteral routes, in free form, given with or conjugated to polyacryl starch microparticles or given with recombinant cholera toxin B-subunit (rCTB), which is believed to increase the uptake of antigen across the mucosa [24]. The study also investigated whether immunization with flagellin would induce a protective immune response against an oral challenge with live serovar Enteritidis.

2. Materials and methods 2.1. Preparation of polyacryl starch microparticles Microparticles were prepared by radical polymerization of acryloylated starch (maltodextrin, Stadex AB, Malmö,

Sweden) in a water-in-oil emulsion [25,26]. Briefly, 3 g of acryloylated starch was dissolved in 30 mL of a 0.2 M phosphate buffer (pH 7.5) containing 50 mM EDTA, mixed with ammonium peroxodisulphate to a final concentration of 0.8 M, and homogenized in an organic phase containing toluene/chloroform (4:1) and 0.08 g of Pluronic® F68 as an emulsifier. N,N,N ,N -tetramethylethylenediamine (TEMED) (Sigma, St. Louis, MO.) was added to initiate the polymerisation. After washing several times with saline, the microparticles were filtered through an 11 ␮m pore-size nylon filter. The particle size was determined using a laser diffractometer (LS 230; Coulter Electronics Ltd., England); the mean particle size was 1.85 ± 0.68 ␮m (based on the number distribution). 2.2. Purification of Salmonella flagellin Flagellin was isolated from wild-type S. enterica serovar Enteritidis (O-9, 12) phage type 1 with a monophasic distribution of flagella (human isolate strain 225 from the Swedish Institute for Infectious Disease Control, Stockholm, Sweden) using the method of Ibrahim et al. [27]. Briefly, isolates from Salmonella were inoculated into Luria broth-medium (1% tryptone, 0.5% yeast agar, 1% NaCl from Difco, Detroit, MI) and incubated at 37 ± 0.2 ◦ C in an orbital shaker incubator at 150 rpm overnight. Cells were harvested by centrifugation at 3000 × g for 30 min. The bacterial pellets were resuspended in saline, and the pH was lowered to 2. After further centrifugation steps, including ultracentrifugation, the supernatant was filtered through a 0.22 ␮m filter (Millipore, Bedford, MA). Proteins were precipitated, at pH 7.2, with 2.67 M ammonium sulphate and centrifuged at 17,400×g. The pellet was resuspended in distilled water and dialyzed in a dialysis tube, with a cut-off at 50 kDa (Spectrapore, Spectrum Laboratories Inc., CA) for 48 h at 4 ± 1 ◦ C against phosphate-buffered saline (PBS). The protein was purified further using gel filtration on a Sephacryl-300 (Pharmacia, Uppsala, Sweden) column. Purity was verified using sodium dodecyl sulphate-polyacrylamide gel electrophoresis (SDS-PAGE) (12%) followed by Coomassie brilliant blue staining. The protein concentration was determined by amino acid analysis. Endotoxins were analyzed by the gel-clot method A of the European Pharmacopoeia, which is based on a Limulus amoebocyte lysate. Escherichia coli 055:B5 was used as a standard. 2.3. Conjugation of flagellin to polyacryl starch microparticles The microparticles were activated with carbonyldiimidazole (CDI) (Merck, Darmstadt, Germany) in dry dimethylformamide (DMF) end over end at room temperature for 1 h [28]. After several washings with DMF to remove unreacted CDI, the microparticles were transferred to a coupling buffer containing 0.25 M boric acid with saline at pH 8.6 and flagellin, and incubated end over end at 4 ± 1 ◦ C for 72 h. The

L. Strindelius et al. / Vaccine 22 (2004) 3797–3808

3799

Table 1 Immunization schedule Group/treatment

Immunization route

Flagellin dosage (␮g per day) [duration (days)]

Free flagellin Flagellin-conjugated microparticles Free flagellin + microparticles Free flagellin + rCTB Flagellin-conjugated microparticles Free flagellin + rCTB

p.o. p.o. p.o. s.c. s.c. i.n.

9 9 9 3 3 3

or or or or or or

42 42 42 14 14 18

[3] [3] [3] [1] [1] [3]

Adjuvant dosage [duration (days)]

3 mg per day [3] 3 mg per day [3] 10 ␮g per day [1] 1 mg per day [1] 10 ␮g per day [3]

The lower doses (3 ␮g flagellin/mg microparticle) were given to 3 mice per group, and the higher doses (14 ␮g flagellin/mg microparticle) to 7 mice per group. i.n.: intranasally; p.o.: orally; s.c.: subcutaneously; rCTB: recombinant cholera toxin subunit B.

flagellin-conjugated microparticles were washed with saline and filtered through an 11 ␮m pore-size filter. The amount of flagellin conjugated to the microparticles was determined by amino acid analysis. A coupling yield of 39% was obtained. 2.4. Immunization of mice Female C3H/HeJ mice (Charles River, Uppsala, Sweden), 8–12 weeks old, were immunized orally with free flagellin, flagellin-conjugated microparticles or flagellin plus unconjugated microparticles, subcutanously with flagellin plus recombinant cholera toxin subunit B (rCTB) (SBL Vaccine AB, Solna, Sweden) or flagellin-conjugated microparticles, or nasally with flagellin plus rCTB, as described in Table 1. Each group comprised 10 mice. Three mice in each group received a low dose of flagellin (3 or 9 ␮g depending on route of administration), while seven mice per group received a higher dose (14, 18 or 42 ␮g). Oral and intranasal doses were given on three consecutive days, the oral group by gastric intubation and the nasal group by pipetting 25 ␮l of the formulation into the nostrils under light anesthesia with isofluran. After three weeks, a booster dose was given to all mice. The mice receiving the low flagellin dose were given a second booster of the high dose three weeks after the first booster. Negative controls were provided by a group of non-immunized mice, and positive controls by a group that was injected intraperitoneally with free antigen in Freund’s complete adjuvant followed by a booster in Freund’s incomplete adjuvant (Difco). 2.5. Sampling Blood samples were taken from the tail artery on days 0, 20 and 35 and centrifuged to obtain the serum. Fecal samples were collected on days 0, 26, 27 and 28 and freeze-dried. The feces were vortexed in PBS containing 5% non-fat dry milk, 0.1 mg of soybean trypsin inhibitor/ml and 2 mM phenylmethylsulphonyl fluoride (Sigma), 20 ␮l/mg dry feces, and centrifuged at 18,000 × g for 10 min at 4 ± 1 ◦ C. The clear supernatants were then frozen until analysis.

2.6. Enzyme-linked immunosorbent assays of flagellin-specific antibodies (IgM–IgG, IgA, IgG1, IgG2a and IgG2b) in serum, and IgA in feces Nunc Immunoplate Maxisorb F96 plates (Nalge Nunc International, Rochester, NY) were coated with flagellin (10 ␮g/ml) in 0.05 M sodium bicarbonate buffer at pH 9.6 (100 ␮l per well) and incubated overnight in a moist chamber at 4 ± 1 ◦ C. The plates were blocked with 1% OVA in PBS for 2 h at room temperature. The plates were washed five times in PBS with 0.05% Tween 20 (PBS-T) and all subsequent washes were performed in the same manner. The antisera or extracts of feces were diluted to appropriate concentrations in PBS-T (0.2%) and added to the plates in series of two-fold dilutions. The plates were incubated for 2 h and washed. The secondary antibodies, alkaline phosphatase-conjugated goat-anti-mouse IgM–IgG (Biosource, Camarillo, CA) (1:1000) or goat-anti-mouse IgA (Sigma) (1:250), rat-anti-mouse IgG2a, rat-anti-mouse IgG2b or rat-anti-mouse-IgG1 (Pharmingen, San Diego, CA) were added and incubated for 2 h at room temperature. 4-Nitrophenylphosphate (Merck) 1 mg/ml in 10% diethanolamine buffer, pH 9.6, with 0.5 mM MgCl2 was used as substrate. The absorbance was measured at 405 nm in a microtiter-plate spectrophotometer (Titertek Multiscan MC, Flow Laboratories). A sandwich-type ELISA was developed to determine IgG1, IgG2a and IgG2b in sera against relevant subclass standard curves. Nunc Immunoplate Maxisorb F96 plates were coated with rat-anti-mouse IgG1, IgG2a or IgG2b (Pharmingen) and incubated overnight in a moist chamber at 4 ◦ C. The plates were blocked with 1% OVA and the standard mouse IgG1, IgG2a or IgG2b was diluted to appropriate concentrations in PBS with 3% OVA to create a standard curve. The subsequent procedures were performed as described above. In each sample, the final concentration of antigen-specific IgG1, IgG2a or IgG2b was determined from pooled standard curves by calculating a mean concentration value from three different dilutions of a sample. When measuring IgM–IgG and IgA antibodies, pooled negative serum was added to each plate as a negative control. The negative serum absorbance values collected from all the plates were below 0.1. Thus, 0.1 was set as the limiting value

3800

L. Strindelius et al. / Vaccine 22 (2004) 3797–3808

for a positive result. Titers are given as −log2 (dilution × 10). A positive sample (from hyperimmunized mice) was included as a standard on all plates and treated in the same way as the antisera to ensure consistency between the plates. 2.7. ELISA of total IgA in fecal samples A sandwich-type ELISA was developed to determine the total amount of IgA in fecal samples. It was performed as described above for antigen-specific IgA with the plates coated with goat-anti-mouse IgA (0.5 ␮g per well) (Sigma). The standard IgA was diluted to appropriate concentrations in PBS with 1% OVA to create a standard curve and the extracts from the fecal samples were diluted in PBS-T (0.2%) and added to the plates in series of two-fold dilutions. The concentrations of specific and total IgA were determined from the standard curve by calculating the mean concentration from three different dilutions of the sample. 2.8. ELISA of rCTB-specific antibodies (IgM–IgG) in serum DYNEX 96 plates (DYNEX Laboratories Inc., Chiantilly, VA) were coated with monosialoganglioside-GM1 (Sigma) (0.45 ␮g/ml) in PBS (100 ␮l per well) and incubated overnight at room temperature. The plates were washed five times in PBS with 0.05% Tween 20 (PBS-T), blocked with 0.1% OVA (250 ␮l per well) in PBS for 0.5 h at 37 ± 0.2 ◦ C, and emptied. rCTB (0.5 ␮g per ml) (SBL Vaccine AB, Solna, Sweden) in 0.1% OVA-PBS was then added to the plates (100 ␮l per well) and incubated for 1 h at room temperature. After washing, the antisera were diluted to appropriate concentrations in 0.1% OVA-PBS and added to the plates in series of two-fold dilutions. The plates were incubated for 1 h at room temperature and washed. The subsequent steps used were identical to those described above. 2.9. ELISA of LPS-specific antibodies (IgM–IgG) in serum EB-plates (Thermo Labsystems Oy, Helsinki, Finland) were coated with LPS (O-1, 9, 12) from serovar Enteritidis (L-2012; Sigma) (4 ␮g/ml) in PBS buffer (100 ␮l per well) and incubated overnight in a moist chamber at 4 ± 1 ◦ C. The plates were treated in the same manner as the flagellin plates analysed for IgM–IgG in serum described above in Section 2.6. 2.10. Splenocyte cultures Single-cell suspensions were prepared by puncturing the spleen with a thin syringe and squeezing the cells out. The single-cell suspensions were washed in sterile PBS and then incubated on ice for 10 min in TRIS buffer with ammonium chloride. After two washes in RPMI 1640 medium,

the cells were resuspended in culture medium (RPMI 1640 including 10% fetal calf serum (FCS), 100 U/ml penicillin, 100 ␮g/ml streptomycin, 50 ␮M 2-mercaptoethanol and 2 mM l-glutamine (Gibco BRL, Gaithersburg, MD)) and counted. They were then plated (3 × 106 /well, with a total volume of 1.5 ml per well) into sterile 24-well microplates (Costar, Corning Inc., NY) and stimulated either with flagellin (50 ␮g per well), concanavalin A (Con A, 4 ␮g per well) (Sigma) or, as a negative control, sterile RPMI-medium. Cultures were incubated for 48 and 72 h at 37 ± 0.2 ◦ C in 5% CO2 . The cells were harvested by centrifugation at 5000 × g for 10 min at 4 ± 1 ◦ C and the supernatants collected and frozen at −80 ± 1 ◦ C until analyzed for cytokine production. 2.11. Cytokine assay and proliferation The supernatants from the splenocyte cultures were collected on days 2 and 3 and analyzed for IL-2, IL-4, IL-5 and interferon (IFN)-␥. A sandwich-type ELISA was developed to determine levels of IFN-␥, IL-2, IL-4 and IL-5 produced by the splenocytes during antigen activation. Antibody pairs and recombinant standards were purchased from Pharmingen. Nunc Immunoplate Maxisorb F96 plates were coated in 0.1 M sodium bicarbonate buffer at pH 9.5 with rat-anti-mouse-IFN-␥, -IL-2, -IL-4 or -IL-5, and incubated overnight in a moist chamber at 4 ± 1 ◦ C. The plates were blocked with 1% OVA. The standard (IFN-␥, IL-2, IL-4 or IL-5) was diluted to appropriate concentrations in PBS with 1% OVA to create a standard curve. The supernatants from cell media were added in duplicate to the plates in series of two-fold dilutions. Biotin-conjugated rat-anti-mouse IFN-␥, -IL-2, -IL-4 or -IL-5, diluted in PBS with 1% OVA, was added to the plates, and avidin-peroxidase and later 2,2 -azinobis-3-ethylbenzthiazoline-sulfonic acid (ABTS, Sigma) with 0.3% H2 O2 were added. The reaction was stopped with 0.7 M SDS in DMF and purified water, 1:1 (Sigma). The absorbance was measured at 405 nm. In each sample, the final concentration of IFN-␥, IL-2, IL-4 or IL-5 was determined from pooled standard curves by calculating a mean concentration value from three different dilutions. For proliferation studies, the cells were plated (200 ␮l and 5 × 105 cells per well) into sterile 96-well microplates (TC Microwell 96 F, Nunc A/S, Roskilde, Denmark) and cultured either together with flagellin (50 ␮g per well) or Con A (1 ␮g per well). Cultures were incubated for 48 h at 37 ± 0.2 ◦ C in 5% CO2 , and pulsed with 1 ␮Ci [methyl-3 H] thymidine (TRK-120, 1 mCi/ml, Amersham International plc, Amersham, England) for 16 h. Cells were harvested on glass fibre filters and cell-bound radioactivity determined by scintillation counting of the filters. 2.12. SDS-PAGE and immunoblotting SDS-PAGE was performed according to Laemmli [29]. The Salmonella protein samples (secreted antigens obtained

L. Strindelius et al. / Vaccine 22 (2004) 3797–3808

3801

after cultivating the bacteria for a short time, as described by Strindelius et al. [6]) or purified flagellin were spun down in a Speed Vac Plus centrifuge, resuspended in loading buffer, heated for 5 min at 100 ◦ C just before the run, and separated on an SDS-PAGE gel (12%) (Bio-Rad Laboratories, CA). Electrophoresis was carried out at 100 V for 3 h and the proteins were transferred onto a Hybond ECL nitrocellulose membrane (Amersham Pharmacia Biotech, Uppsala, Sweden) at 50 V in a Trans-blot system (Bio-Rad) with blotting buffer (25 mM Tris–HCl, 0.2 M glycine in 20% methanol) for 4 h. After blocking with 5% non-fat milk in PBS, the immune sera were diluted to 1:100,000 in blocking solution and incubated with the membranes overnight at 4 ± 1 ◦ C. A negative control from non-immunized mice (dilution 1:50) was run in parallel. After washing in PBS with Tween 0.05%, the membranes were incubated with the secondary antibody, anti-mouse-IgG-HRP, and developed using an enhanced chemiluminescence system (ECL) according to the manufacturer’s instructions (Amersham Pharmacia Biotech, Uppsala, Sweden). Biotinylated molecular size markers (Bio-Rad) ranging from 14 to 200 kDa were used. No band was detectable with the negative control samples.

for the IgA quota and the IgM–IgG and IgA titers, and these were analyzed with ANOVA repeated measures in combination with Fisher’s multiple comparison test. The limit of significance was set to P < 0.05.

2.13. Challenge

A strong systemic IgM–IgG response was obtained in all groups after immunization with the high dose of flagellin (42 or 14 ␮g, see Table 1), as shown in Fig. 2A. The groups immunized intranasally and subcutaneously had the highest antibody response, but even free flagellin given orally elicited a high response. Furthermore, the two groups immunized with free flagellin mixed with rCTB also developed titers against rCTB that were in the same high range as for the specific antibody response against flagellin (data not shown). The lower dose of flagellin (9 or 3 ␮g) also

The virulence of the strain used was first estimated by following the proliferation of the bacteria in the spleens of na¨ıve mice. Groups of 8–12-week-old female C3H/HeJ mice were inoculated orally with different concentrations of bacteria. The mice were sacrificed on day seven and the spleens were removed and homogenized. Suitable dilutions of the homogenates were plated to determine the number of bacteria present. The concentration of bacteria giving a clear infection was chosen for the subsequent challenge. Immunized mice were challenged with wild-type S. enterica serovar Enteritidis (O-9, 12) phage type 1 with a monophasic distribution of flagella (human isolate strain 225 from the Swedish Institute for Infectious Disease Control, Stockholm, Sweden) (2 × 109 CFU per mouse) 3 weeks after the booster. The mice were deprived of food and water 16 h before challenge. They were then given 50 ␮l of an oral bacterial suspension in PBS with 1% NaHCO3 (4 × 1010 CFU/ml). The spleens were taken out 6 days after challenge and homogenized in 5 ml of sterile PBS using a Stomacher lab-blender (Seward Medical, London, UK). The homogenates, diluted in sterile PBS to appropriate concentrations, were plated onto LB-agar plates in duplicate and incubated at 37 ± 0.2 ◦ C overnight, after which the number of colonies was counted.

3. Results 3.1. Purification of Salmonella flagellin SDS-PAGE of the purified flagellin revealed one dominating band at about 56 kDa (Fig. 1), which corresponds to the molecular mass of flagellin (approximately 55 kDa). The LPS contamination, as tested with the LAL-test, corresponded to 7 × 103 EU per ␮g protein. As found previously [6], only a minor fraction is bound, when flagellin is conjugated to the microparticles, containing 52 EU per ␮g protein. 3.2. A strong systemic humoral response is induced in all immunized groups

2.14. Statistics Data from the challenge studies were logarithmically transformed and factorial analysis of variance (ANOVA) in combination with Fisher’s multiple comparison test was used in the statistical analysis. Arithmetic data were used

Fig. 1. Electrophoretic analysis of purified flagellin from Salmonella enteritidis using SDS-PAGE (12%) with Coomassie brilliant blue staining. Lane 1: molecular mass markers; lane 2: purified flagellin showing one major band at 50–60 kDa.

3802

L. Strindelius et al. / Vaccine 22 (2004) 3797–3808 Free flagellin p.o.

16

Antibody titre

14

Flagellin-conjugated microparticles p.o.

12 Free flagellin + microparticles p.o.

10 8

Free flagellin + rCTB s.c.

6 4

Flagellin-conjugated microparticles s.c.

2

Free flagellin + rCTB i.n.

0 0

5

10

15

20

25

30

35

Day

(A) 14

Free flagellin p.o.

Antibody titre

12

Flagellin-conjugated microparticles p.o.

10

Free flagellin + microparticles p.o.

8

Free flagellin + rCTB s.c.

6

Flagellin-conjugated microparticles s.c.

4 2

Free flagellin + rCTB i.n.

0 0

(B)

5

10

15

20

25

30

35

Day

Fig. 2. Serum IgM–IgG antibody response after administration of (A) the high dose (n = 7), and (B) the low dose (n = 3) of flagellin to mice (for dosages see Table 1). A booster dose was given on day 21. Antibody titers are given as –log2 (dilution × 10) ± S.D. The groups that received flagellin intranasally or subcutaneously (both high and low doses) had significantly higher antibody titers than the oral groups (P < 0.05). No specific IgM–IgG response was detected in the control groups. Single symbols on day 35 indicate the respective specific anti-LPS antibody responses.

induced a relatively high IgM–IgG response, especially after intranasal or subcutaneous immunizations (Fig. 2B). The contribution of LPS-specific antibodies to the total IgM–IgG responses at day 35 was low and are shown in Fig. 2A. None of the groups treated orally with free flagellin or with flagellin-conjugated microparticles showed a significant response (titres <1), while the other groups had titers between 4 and 5, i.e. about 1/256 of the total IgM–IgG titres. The subclass ratio, calculated as IgG1/(IgG2a + IgG2b), was very low after immunization with both the high (Table 2) and the low (data not shown) doses, indicating a Th1-type dominating response. The highest ratio was seen in the group receiving flagellin-conjugated microparticles orally with a ratio of 2.12 at day 20, which decreased slightly to 0.90 after the booster. We also analyzed the IgA response in serum (Fig. 3). In this case, free flagellin given orally elicited the lowest titer of IgA.

Table 2 Subclass ratio Group/treatment

Free flagellin Flagellin-conjugated microparticles Free flagellin + microparticles Free flagellin + rCTB Flagellin-conjugated microparticles Free flagellin + rCTB

Immunization route

Subclass ratio Day 20

Day 35

p.o. p.o.

0.49 ± 0.21 2.12 ± 2.76

0.35 ± 0.40 0.90 ± 0.65

p.o.

0.50 ± 0.25

0.55 ± 0.16

s.c. s.c.

0.12 ± 0.05 0.18 ± 0.12

0.20 ± 0.08 0.18 ± 0.15

i.n.

0.03 ± 0.02

0.05 ± 0.01

The subclass ratio was calculated as IgG1/(IgG2a + IgG2b) and shown as means per group from the mice receiving the higher dose of flagellin, ±S.D. (n = 7).

L. Strindelius et al. / Vaccine 22 (2004) 3797–3808 9

3803

Free flagellin p.o.

8 Flagellin-conjugated microparticles p.o.

7

Antibody titre

6

Free flagellin + microparticles p.o.

5

Free flagellin + rCTB s.c.

4 3

Flagellin-conjugated microparticles s.c.

2

Free flagellin + rCTB i.n.

1 0 0

5

10

15

20

25

30

35

Day Fig. 3. Serum IgA antibody response after administration of the high dose of flagellin (see Table 1) to mice (n = 7). A booster dose was given on day 21. Antibody titers are given as −log2 (dilution × 10) ± S.D. All immunized groups had significantly higher antibody titers than the group that received free flagellin orally (P < 0.05). No specific IgA response was detected in the control group.

Immunoblotting analysis of sera obtained from mice immunized with flagellin against secreted proteins from Salmonella, showed two bands between 50 and 60 kDa (Fig. 4). These bands were similar to those detected in mice immunized with secreted Salmonella antigens in a previous study [6], which were found, using electrospray mass-spectroscopy, to originate from flagellin. 3.3. A strong mucosal response against flagellin is generated Mice immunized intranasally or subcutaneously with free flagellin plus rCTB or orally with free flagellin plus microparticles showed significantly higher (P < 0.05) antigen-specific IgA responses than the control group after immunization with the high dose of flagellin (Fig. 5A). When the low dose of flagellin was given, only mice im-

munized intranasally showed a significantly higher IgA response than the control group. Antigen-specific IgA antibodies in the fecal extracts of other groups were hardly detectable (Fig. 5B). The IgA quota obtained from the individual mice were used in a regression analysis against the individual IgA antibody titers in serum and the degree of correlation was shown with an r 2 = 0.54 (P < 0.002, n = 14) for the subcutaneous groups and an r2 = 0.37 (P < 0.01, n = 21) for the oral groups. The P-values indicate that the correlation was significantly different from a random distribution. 3.4. Flagellin-stimulated primed spleen cells proliferate and produce IFN-γ and IL-2 The mice receiving the low flagellin dose were given an additional booster of the high dose. These mice were sacrificed 2 weeks after the second booster and the pooled splenocytes were stimulated in vitro with 50 ␮g flagellin per well (3 × 106 cells) or 4 ␮g ConA. The supernatants were collected on days 2 and 3 and analyzed for IL-4, IL-5, IL-2 and IFN-␥ (Table 3). The supernatants contained no IL-4 or IL-5; however, IFN-␥ levels increased over time in most of the groups. IL-2 was only detected on day 2, with quite low values. The highest values were observed in the groups immunized nasally with flagellin plus rCTB and subcutaneously with flagellin-conjugated microparticles. 3.5. Challenge

Fig. 4. Western blot analysis of sera from mice immunized with the high dose of flagellin, 2 weeks after booster, against extracellular antigens from serovar Enteritidis. Lane 1: molecular mass markers. Lane 2: sera from the group receiving flagellin intranasally, diluted 1:100,000; this sample is representative of all immunized groups.

Of the mice sacrificed after live bacterial challenge, only the groups immunized orally with flagellin in free form and the group receiving flagellin intranasally had a significantly lower degree of infection (CFU/ml spleen) than the control

3804

L. Strindelius et al. / Vaccine 22 (2004) 3797–3808

Fig. 5. (A) Amount of IgA in feces, expressed as the mean IgA quota (specific IgA/total IgA) + S.D. (n = 7) after administration of the high dose of flagellin (see Table 1), 5–7 days after the booster. The groups immunized intranasally or subcutaneously with flagellin plus rCTB or orally with flagellin plus microparticles had significantly higher specific IgA amounts than the control group (P < 0.05). (B) IgA quota (specific IgA/total IgA) + S.D. (n = 3) after administration of the low dose of flagellin, 5–7 days after the first booster. The group immunized intranasally had significantly higher specific IgA amounts than the control group (P < 0.05). No specific IgA response was detected in the control groups. Table 3 IFN-␥ and IL-2 in supernatants from activated primed splenocytes (ng/ml). Group/treatment

Immunization route

Day 2 (IFN-␥)

Flagellin-conjugated microparticles Free flagellin Microparticles + free flagellin Flagellin-conjugated microparticles Free flagellin + rCTB Free flagellin + rCTB Control

p.o. p.o. p.o. s.c. s.c. i.n.

1.03 2.5 0.78 0.52 23 4.2 0.66

± ± ± ± ± ± ±

0.21 0.05 0.26 0.04 0.9 0.5 0.07

Day 3 (IFN-␥) 2.2 5.5 3.7 0.84 40 4.8 1.4

± ± ± ± ± ± ±

0.5 0.6 0.6 0.23 3.4 0.1 0.4

Day 2 (IL-2) 0.11 0.17 0.14 0.39 0.22 0.52 0.03

± ± ± ± ± ± ±

0.01 0.01 0.01 0.02 0.02 0.05 0.001

Two mice from each group in the low dose study were sacrificed 2 weeks after a second booster with a high dose and in vitro stimulation of the pooled splenocytes was performed with 50 ␮g flagellin/well (3 × 106 cells). Cytokine assays were performed on duplicate supernatants collected on days 2 and 3. The IFN-␥ and the IL-2 responses are shown as arithmetic means from duplicate samples ± S.D. (n = 2).

L. Strindelius et al. / Vaccine 22 (2004) 3797–3808

3805

Fig. 6. CFU in spleen 6 days after oral challenge with serovar Enteritidis, in mice immunized with the high dose of flagellin (see Table 1). Values are given as geometric means + S.D. (n = 5–7). The groups that received free flagellin orally or free flagellin plus rCTB intranasally had significantly lower degree of infection than the control group (P < 0.05), indicated by the asterisk (*). †: one dead mouse during challenge.

group (Fig. 6). However, one or two mice died during the challenge (<6 days) in the subcutaneous and control groups. In order to be able to include these mice in the analyses, they were assigned 105 CFU (the highest infection rate among those that survived).

4. Discussion Our results support those of the previous study in BALB/c mice vaccinated with secreted antigens from S. enterica serovar Enteritidis, which indicated that the anti-flagellin response was an important component of the host’s defence against Salmonella infections [6]. In this study, which used purified flagellin in C3H/HeJ mice, a remarkably high systemic humoral immune response was induced in all the immunized groups. In addition, flagellin induced a mucosal immune response, expressed as the ratio between specific IgA and total IgA in feces, which varied in strength according to the route of administration and formulation. The mice immunized intranasally with flagellin plus rCTB or subcutaneously with flagellin conjugated to microparticles or with rCTB produced the highest antibody titers in serum, but even free antigen given orally without any adjuvant at all induced very high antibody titers. The mice that had received a low dose of flagellin orally also had relatively high antibody titers. It is thus apparent that flagellin is a strong immunogen that can induce an immune response even when given orally in its free form. Effective, mucosal vaccination with purified flagellin has not previously been reported. Flagellin has been shown to induce a strong specific CD4+ T-cell response in mice after oral immunization with attenuated Salmonella [17]. Also, McSorley et al. [18,30,31] have characterised the CD4+ T-cell response after Salmonella typhimurium infection and they identified

flagellin as a strong adjuvant for CD4+ T-cell activation in vivo. We have analyzed the subclass ratio of the systemic humoral response and detected a relatively low ratio of IgG1/(IgG2a + IgG2b), which indicates a dominating Th1-type response. The Th1-type response is known to be an important component of the protective defence against Salmonella [32]. The production of IFN-␥ and IL-2 induced after activation of primed spleen cells, and the lack of IL-4 or IL-5 in the supernatant, also supports the conclusion that the immune response is polarized towards a Th1-type response. The same strong Th1-type influence of the cellular response was also seen in our previous study using secreted proteins from Salmonella [6]. In addition, proliferation occurred in all groups on stimulation with flagellin in vitro in the present study (data not shown). A preliminary flow cytometry analysis was performed on the same splenocyte population that showed production of IFN-␥ upon stimulation with flagellin. These preliminary results showed that only a small fraction of CD4+ T-cells from the spleen produced IFN-␥. However, it might be that cells other than CD4+ T-cells (e.g. CD8+ T cells, NK cells or maybe antigen-presenting cells) produced the IFN-␥ observed in the supernatants in our study. A recent study showed that flagellin-specific CD4+ T-cells are primarily located in the gut-associated lymphoid tissue after an oral infection with Salmonella and that they do not migrate to the spleen or liver [30]. It would therefore be of great interest to also investigate the flagellin-specific CD4+ T-cell response at other locations, such as Peyer’s patches and lymph nodes, after vaccination through different routes of administration. In our study, the flagellin-specific mucosal IgA antibodies were detected in feces, irrespective of the route of administration used. The group immunized subcutaneously with flagellin and rCTB demonstrated a rather high specific s-IgA level in feces compared to the group that received flagellin

3806

L. Strindelius et al. / Vaccine 22 (2004) 3797–3808

conjugated to the microparticles. It is generally accepted that mucosal immunization gives rise to a local immune response while parenteral immunization does not, at least not to the same extent. In our previous study, an IgA-ELISPOT on the intestine showed specific IgA-producing cells even in the group that received free antigen intramuscularly. It thus seems possible to induce a mucosal response with parenteral immunization if a strongly immunogenic antigen and adjuvant are used, in this case flagellin and rCTB. Other studies have also shown that a mucosal immune response can be induced using a parenteral route of administration [33,34]. In addition, a weak correlation between the serum IgA titers and the mucosal IgA was observed when studying the subcutaneous groups and the oral groups individually. This observation suggests a possible contribution of IgA from serum to the mucosa via the hepatobiliary pathway. CTB has been extensively used as a mucosal carrier molecule [35–39]. The mechanisms behind the induction by CTB of an immune response at mucosal sites are not yet clear but are believed to be associated with the binding of CTB to the GM1 receptor, which is abundant on most cell types [40,41]. A recent study showed that antigen conjugated to CTB enhanced the antigen-presenting capacity of dendritic cells, B-cells and even macrophages by indirect upregulation of co-stimulatory molecules on the antigen-presenting cell APC [24]. However, this effect was not observed when rCTB was just mixed with the antigen. In our study, flagellin was mixed with rCTB; the mucosal anti-flagellin responses in both groups receiving rCTB (intranasal and subcutaneous) could well have been increased because of the immunomodulating effect of rCTB. Flagellin induced a good immune response irrespective of the formulation used (given with rCTB, conjugated covalently to starch microparticles or given in free form), and despite the fact that it was taken up and handled differently by the immune system depending on the route of administration and formulation. The flagellin-conjugated microparticles, which were most likely taken up via M-cells in the Peyer’s patches after oral administration, would have been processed with the flagellin in the APCs (dendritic cells or macrophages) in the submucosal tissues. When given with rCTB by the nasal or subcutaneous route, the uptake and processing of flagellin may have been facilitated in some way by the presence of rCTB, as discussed above. The purified flagellin given orally in free form was most likely in the monomeric state in the PBS solution at the time of administration, as suggested by Furukawa et al. [42]. Nonetheless, the flagellin might have polymerized; aggregated proteins have a higher immunogenicity than soluble proteins [43]. However, when it passes through the gastrointestinal tract it may well be fragmented before it reaches the region of uptake. Parish [44], for example, has shown that the protecting epitopes are assembled within a conserved part of the flagellin. Others have also indicated that the proinflammatory and immunogenic sites of flagellin are concentrated in conserved parts of the molecule [13,18]. We have not specifi-

cally studied the active epitopes of flagellin in this work, but we can surmise that they were preserved and able to induce a strong immune response, when flagellin was given either in free form or conjugated to starch microparticles. The strong immune response induced by flagellin may partly be explained by recent studies, which demonstrated that the Salmonella bacterium translocates flagellin across the intestinal epithelia and subsequently induces the release of IL-8, which is a known neutrophil chemoattractant [9]. Translocation of the purified flagellin used in this study, however, is unlikely. While flagellin is known to bind to TLR5 on the basolateral side of an intestinal epithelia model based on a human colonic cell line [45], it is not known if this also occurs in vivo in mice. However, TLR5 is also situated on dendritic cells, which can translocate their dendrites into the lumen and sample its contents [46]. Furthermore, dendritic cells can migrate to the intestine upon the flagellin-induced release of CCL20 [16]. Taken together, the evidence to date suggests that flagellin or fragments thereof would be a promising and potent component of a mucosal vaccine against enteric Salmonella or as an adjuvant for other antigens. There are indications that previous exposure to flagellin induces self-tolerance upon restimulation in some cell-lines [47]. However, the impact of the development of tolerance may be more complex in an in vivo situation and requires further study. No signs of tolerance against flagellin were noted in this study. One of the main foci of our research was to find information which could be used for the development of a vaccine, i.e. to test if the induced immune response was protective against a challenge with live bacteria. Challenge studies in animals are limited by the animal protection guidelines, which state that the survival rate should not be a parameter for evaluation of protection. In this study, after assessing the virulence of the bacterium strain, the number of CFUs after a high oral dose (2 × 109 /mouse) was used as a quantitative parameter to evaluate protection. Interestingly, there was no direct correlation between the challenge results and the immunochemical data obtained. The best protection was obtained in the groups immunized orally with free flagellin and intranasally with free flagellin plus rCTB. However, only the groups immunized nasally or subcutaneously with flagellin and rCTB, or orally with flagellin plus microparticles had significant specific s-IgA response in feces compared to the control group over time. The presence of specific s-IgA in feces is beneficial, as it indicates that the bacterium would be prevented from entering the mucosal epithelium. However, the oral bacterial challenge load was probably too high for the limited levels of s-IgA present in the intestine. It could be that free flagellin given mucosally gives rise to a more efficient cellular immune response in the mucosa, initiated by the induction of proinflammatory cytokines/chemokines, which then triggers a cascade of immune reactions in the FAE and submucosal tissues. It should be stressed that the implications of the inflammatory responses induced by flagellin should also be taken into consideration in any

L. Strindelius et al. / Vaccine 22 (2004) 3797–3808

potential development of a vaccine based on flagellin. Even though C3H/HeJ mice are known to be unable to respond to the small amount of LPS contaminating the formulations with proinflammatory responses via TLR-4, antibodies directed towards LPS can however be generated [48]. In our study, these LPS-antibodies were only found in the groups treated subcutaneously and nasally (which may result in a systemic response via the lungs, too). The orally treated groups did not have any significant LPS-antibodies (titers <1), which suggests that the antibodies directed towards LPS are of minor importance. Thus, this study allows the conclusion to be drawn that purified flagellin has the capacity to effectively trigger an immune response after nasal or oral immunization. Recombinantly produced flagellin would therefore appear to be a promising component of a mucosal subunit vaccine against Salmonella, by itself or in combination with other bacterial components, and may even have potential as an adjuvant in vaccine formulations.

Acknowledgements We thank Brittmarie Andersson and Linda Stertman for skillful technical assistance; Kelly Hughes, Joyce Karlinsey, Phillip Aldridge, Heather Bonifield at the University of Washington, Seattle, USA, for valuable information regarding the flagellin; Anna Björkland for discussions regarding the cytokine analyses. References [1] Kaufmann SH, Raupach B, Finlay BB. Introduction: microbiology and immunology: lessons learned from Salmonella. Microbes Infect 2001;3:1177–81. [2] Hughes EA, Galan JE. Immune response to Salmonella: location, location location? Immunity 2002;16:325–8. [3] Clark MA, Jepson MA, Simmons NL, Hirst BH. Preferential interaction of Salmonella typhimurium with mouse Peyer’s patch M cells. Res Microbiol 1994;145:543–52. [4] Jones BD, Ghori N, Falkow S. Salmonella typhimurium initiates murine infection by penetrating and destroying the specialized epithelial M cells of the Peyer’s patches. J Exp Med 1994;180:15–23. [5] Lostroh CP, Lee CA. The Salmonella pathogenicity island-1 type III secretion system. Microbes Infect 2001;3:1281–91. [6] Strindelius L, Degling Wikingsson L, Sjöholm I. Extracellular antigens from Salmonella enteritidis induce effective immune response in mice after oral vaccination. Infect Immun 2002;70:1434– 42. [7] Kaisho T, Akira S. Critical roles of Toll-like receptors in host defense. Crit Rev Immunol 2000;20:393–405. [8] Hayashi F, Smith KD, Ozinsky A, Hawn TR, Yi EC, Goodlett DR, et al. The innate immune response to bacterial flagellin is mediated by Toll-like receptor 5. Nature 2001;410:1099–103. [9] Gewirtz AT, Simon Jr PO, Schmitt CK, Taylor LJ, Hagedorn CH, O’Brien AD, et al. Salmonella typhimurium translocates flagellin across intestinal epithelia, inducing a proinflammatory response. J Clin Invest 2001;107:99–109. [10] Ciacci-Woolwine F, Blomfield IC, Richardson SH, Mizel SB. Salmonella flagellin induces tumor necrosis factor alpha in a human promonocytic cell line. Infect Immun 1998;66:1127–34.

3807

[11] Wyant TL, Tanner MK, Sztein MB. Salmonella typhi flagella are potent inducers of proinflammatory cytokine secretion by human monocytes. Infect Immun 1999;67:3619–24. [12] McDermott PF, Ciacci-Woolwine F, Snipes JA, Mizel SB. High-affinity interaction between gram-negative flagellin and a cell surface polypeptide results in human monocyte activation. Infect Immun 2000;68:5525–9. [13] Eaves-Pyles T, Murthy K, Liaudet L, Virag L, Ross G, Soriano FG, et al. Flagellin, a novel mediator of Salmonella-induced epithelial activation and systemic inflammation: I kappa B alpha degradation, induction of nitric oxide synthase, induction of proinflammatory mediators, and cardiovascular dysfunction. J Immunol 2001;166:1248–60. [14] Moors MA, Li L, Mizel SB. Activation of interleukin-1 receptor-associated kinase by Gram-negative flagellin. Infect Immun 2001;69:4424–9. [15] Liaudet L, Murthy KG, Mabley JG, Pacher P, Soriano FG, Salzman AL, et al. Comparison of inflammation, organ damage, and oxidant stress induced by Salmonella enterica serovar Muenchen flagellin and serovar Enteritidis lipopolysaccharide. Infect Immun 2002;70:192–8. [16] Sierro F, Dubois B, Coste A, Kaiserlian D, Kraehenbuhl JP, Sirard JC. Flagellin stimulation of intestinal epithelial cells triggers CCL20-mediated migration of dendritic cells. Proc Natl Acad Sci U S A 2001;98:13722–7. [17] Cookson BT, Bevan MJ. Identification of a natural T cell epitope presented by Salmonella-infected macrophages and recognized by T cells from orally immunized mice. J Immunol 1997;158:4310–9. [18] McSorley SJ, Cookson BT, Jenkins MK. Characterization of CD4+ T cell responses during natural infection with Salmonella typhimurium. J Immunol 2000;164:986–93. [19] Newton SM, Jacob CO, Stocker BA. Immune response to cholera toxin epitope inserted in Salmonella flagellin. Science 1989;244:70– 2. [20] Stocker BA, Newton SM. Immune responses to epitopes inserted in Salmonella flagellin. Int Rev Immunol 1994;11:167–78. [21] Luna MG, Martins MM, Newton SM, Costa SO, Almeida DF, Ferreira LC. Cloning and expression of colonization factor antigen I (CFA/I) epitopes of enterotoxigenic Escherichia coli (ETEC) in Salmonella flagellin. Res Microbiol 1997;148:217–28. [22] Poltorak A, He X, Smirnova I, Liu MY, Huffel CV, Du X, et al. Defective LPS signaling in C3H/HeJ and C57BL/10ScCr mice: mutations in Tlr4 gene. Science 1998;282:2085–8. [23] O’Brien AD, Rosenstreich DL, Taylor BA. Control of natural resistance to Salmonella typhimurium and Leishmania donovani in mice by closely linked but distinct genetic loci. Nature 1980;287:440– 2. [24] George-Chandy A, Eriksson K, Lebens M, Nordström I, Schon E, Holmgren J. Cholera toxin B subunit as a carrier molecule promotes antigen presentation and increases CD40 and CD86 expression on antigen-presenting cells. Infect Immun 2001;69:5716–25. [25] Artursson P, Edman P, Laakso T, Sjöholm I. Characterization of polyacryl starch microparticles as carriers for proteins and drugs. J Pharm Sci 1984;73:1507–13. [26] Laakso T, Artursson P, Sjöholm I. Biodegradable microspheres. IV. Factors affecting the distribution and degradation of polyacryl starch microparticles. J Pharm Sci 1986;75:962–7. [27] Ibrahim GF, Fleet GH, Lyons MJ, Walker RA. Method for the isolation of highly purified Salmonella flagellins. J Clin Microbiol 1985;22:1040–4. [28] Bethell GS, Ayers JS, Hancock WS, Hearn MT. A novel method of activation of cross-linked agaroses with 1,1 -carbonyldiimidazole which gives a matrix for affinity chromatography devoid of additional charged groups. J Biol Chem 1979;254:2572–4. [29] Laemmli UK. Cleavage of structural proteins during the assembly of the head of bacteriophage T4. Nature 1970;227:680–5. [30] McSorley SJ, Asch S, Costalonga M, Reinhardt RL, Jenkins MK. Tracking Salmonella-specific CD4 T cells in vivo reveals

3808

[31]

[32] [33]

[34]

[35]

[36]

[37]

[38]

[39]

L. Strindelius et al. / Vaccine 22 (2004) 3797–3808 a local mucosal response to a disseminated infection. Immunity 2002;16:365–77. McSorley SJ, Ehst BD, Yu Y, Gewirtz AT. Bacterial flagellin is an effective adjuvant for CD4+ T cells in vivo. J Immunol 2002;169:3914–9. Mittrucker HW, Kaufmann SH. Immune response to infection with Salmonella typhimurium in mice. J Leukoc Biol 2000;67:457–63. Decroix N, Hocini H, Quan CP, Bellon B, Kazatchkine MD, Bouvet JP. Induction in mucosa of IgG and IgA antibodies against parenterally administered soluble immunogens. Scand J Immunol 2001;53:401–9. Decroix N, Quan CP, Pamonsinlapatham P, Bouvet JP. Mucosal immunity induced by intramuscular administration of free peptides in-line with PADRE: IgA antibodies to the ELDKWA epitope of HIV gp41. Scand J Immunol 2002;56:59–65. McKenzie SJ, Halsey JF. Cholera toxin B subunit as a carrier protein to stimulate a mucosal immune response. J Immunol 1984;133:1818– 24. Czerkinsky C, Russell MW, Lycke N, Lindblad M, Holmgren J. Oral administration of a streptococcal antigen coupled to cholera toxin B subunit evokes strong antibody responses in salivary glands and extramucosal tissues. Infect Immun 1989;57:1072–7. Lipscombe M, Charles IG, Roberts M, Dougan G, Tite J, Fairweather NF. Intranasal immunization using the B subunit of the Escherichia coli heat-labile toxin fused to an epitope of the Bordetella pertussis P.69 antigen. Mol Microbiol 1991;5:1385–92. Shen X, Lagergard T, Yang Y, Lindblad M, Fredriksson M, Holmgren J. Preparation and preclinical evaluation of experimental group B streptococcus type III polysaccharide-cholera toxin B subunit conjugate vaccine for intranasal immunization. Vaccine 2000;19:850– 61. Shen X, Lagergard T, Yang Y, Lindblad M, Fredriksson M, Holmgren J. Group B Streptococcus capsular polysaccharide-cholera toxin B subunit conjugate vaccines prepared by different methods for intranasal immunization. Infect Immun 2001;69:297–306.

[40] Holmgren J, Czerkinsky J, Sun B, Svennerholm, A-M. Oral vaccination, mucosal immunity and oral tolerance with special reference to cholera toxin. In: Kaufmann SHE, editor. Concepts in vaccine development. Berlin, Germany: Walter de Gruyter; 1996. p. 437–58. [41] Nashar TO, Webb HM, Eaglestone S, Williams NA, Hirst TR. Potent immunogenicity of the B subunits of Escherichia coli heat-labile enterotoxin: receptor binding is essential and induces differential modulation of lymphocyte subsets. Proc Natl Acad Sci U S A 1996; 93:226–30. [42] Furukawa Y, Imada K, Vonderviszt F, Matsunami H, Sano K, Kutsukake K, et al. Interactions between bacterial flagellar axial proteins in their monomeric state in solution. J Mol Biol 2002; 318:889–900. [43] St. Clair N, Shenoy B, Jacob LD, Margolin AL. Cross-linked protein crystals for vaccine delivery. Proc Natl Acad Sci U S A 1999;96:9469–74. [44] Parish CR. Immune deviation: a historical perspective. Immunol Cell Biol 1996;74:449–56. [45] Gewirtz AT, Navas TA, Lyons S, Godowski PJ, Madara JL. Cutting edge: bacterial flagellin activates basolaterally expressed TLR5 to induce epithelial proinflammatory gene expression. J Immunol 2001;167:1882–5. [46] Rescigno M, Urbano M, Valzasina B, Francolini M, Rotta G, Bonasio R, et al. Dendritic cells express tight junction proteins and penetrate gut epithelial monolayers to sample bacteria. Nat Immunol 2001;2:361–7. [47] Mizel SB, Snipes JA. Gram-negative flagellin-induced self-tolerance is associated with a block in interleukin-1 receptor-associated kinase release from toll-like receptor 5. J Biol Chem 2002;277:22414– 20. [48] Colwell DE, Michalek SM, Briles DE, Jirillo E, McGhee JR. Monoclonal antibodies to Salmonella lipopolysaccharide: Anti-Opolysaccharide antibodies protect C3H mice against challenge with virulent Salmonella typhimurium. J Immunol 1984;133:950–7.