Immunogenicity and protective efficacy offered by a ribosomal-based vaccine from Shigella flexneri 2a

Vaccine 25 (2007) 4828–4836

Immunogenicity and protective efficacy offered by a ribosomal-based vaccine from Shigella flexneri 2a Doo-Hee Shim a , Sun-Young Chang a , Sung-Moo Park a , Hyun Jang b , Rodney Carbis b , Cecil Czerkinsky c , Satoshi Uematsu d , Shizuo Akira d , Mi-Na Kweon a,∗ a

d

Mucosal Immunology Section, International Vaccine Institute, Seoul National University Research Park, Kwanak-Gu, Seoul 151-818, Republic of Korea b Vaccine Development Section, International Vaccine Institute, Seoul 151-818, Republic of Korea c Laboratory Science Division, International Vaccine Institute, Seoul 151-818, Republic of Korea Department of Host Defense, Research Institute for Microbial Diseases, Osaka University, Osaka 565-0871, Japan Received 19 December 2006; received in revised form 7 March 2007; accepted 31 March 2007 Available online 20 April 2007

Abstract Shigellosis is a major form of bacillary dysentery caused by Shigella infection. Shigella ribosome-based vaccines (SRV), considered among the potent vaccine candidates, are composed of O-antigen and ribosome isolated from S. flexneri 2a. To investigate the immunogenicity and protective efficacy of SRV, mice were vaccinated with SRV via the intranasal (i.n.) route. Interestingly, robust levels of Shigella-derived LPSspecific IgG and IgA Abs and antibody-forming cells were elicited in systemic and mucosal compartments following two i.n. administrations of SRV. Groups of mice receiving i.n. SRV developed milder pulmonary pneumonia upon challenge with virulent S. flexneri 2a than did those receiving parenteral SRV. We further found that the MyD88-dependent TLR2 signal partially mediates SRV-induced mucosal immunity, with the exception of TLR4- and TLR5-governed innate immunity. Most importantly, polymeric immunoglobulin receptor knockout (pIgR−/− ) mice, which lack secretory IgA Ab, were afforded less protective efficacy than were wild-type mice. It can be concluded then that SRV is immunogenic and provides protective efficacy in mice. It can also be surmised that a mucosal SRV vaccine would be particularly relevant in targeting shigellosis, which provokes inflammation in the human colon. © 2007 Elsevier Ltd. All rights reserved. Keywords: S. flexneri 2a; Ribosomal vaccine; Mucosal immunity; Secretory IgA

1. Introduction Annually, 164.7 million Shigella episodes occur worldwide, 1.1 million of them leading to death and 163.2 million occurring in developing countries. The Shigella spp., a genus of gram-negative, nonmotile, facultative anaerobic bacilli comprised of non-spore-forming rods, can be divided into four serogroups: S. dysenteriae, S. flexneri, S. boydii, and S. Abbreviations: SRV, Shigella ribosomal vaccine; S-IgA, secretory immunoglobulin A; AFCs, antibody-forming cells; pIgR, polymeric immunoglobulin receptor knockout; SMG, sub-mandibular gland ∗ Corresponding author. Tel.: +82 2 881 1153; fax: +82 2 881 1211. E-mail address: [email protected] (M.-N. Kweon). 0264-410X/$ – see front matter © 2007 Elsevier Ltd. All rights reserved. doi:10.1016/j.vaccine.2007.03.050

sonnei (among those serogroups, S. flexneri is the most pandemic in developing countries). In fact, it has been recorded that 69% of episodes and 61% of all Shigella-related deaths involved children under 5 years of age. In developed countries, Shigella infection may result from people being exposed to unhygienic conditions. Indeed, Shigella is the decisive agent causing traveler’s diarrhea in individuals from industrialized countries visiting developing countries [1]. Shigellosis, also known as bacillary dysentery, is an acute recto-colitis caused by Shigella spp. in humans that can cause such symptoms as fever, watery diarrhea, vomiting, dehydration, dysentery, bloody stools, and tenesmus [1]. Recent studies have attempted to elucidate how Shigella translocates into the colonic epithelium, where it provokes the infectious

D.-H. Shim et al. / Vaccine 25 (2007) 4828–4836

processes characteristic of bacillary dysentery. When the Shigella bacilli reach the colon and rectum, they translocate through the epithelial barrier via M cells that overlie colonic patches and isolated lymphoid follicles. Once they reach the sub-lumen region, Shigella encounter and are phagocytosed by resident macrophages. The phagocytosed bacteria then escape from the phagosome into the cytoplasm, eventually resulting in extensive inflammation and tissue destruction of the colon and rectum [2]. A unique feature of immune responses at mucosal surfaces is the production of secretory immunoglobulin A (S-IgA) antibodies (Abs) and their transport across the intestinal epithelium [3]. Importantly, mucosal vaccines can induce both systemic and mucosal immunity, resulting in two layers of host protection, and so are considered as potent tools for eradicating pathogens invading via the mucosa [4]. As the inflammatory symptoms of shigellosis are restricted to mucosal tissues (i.e., colon), a mucosally administered vaccine may be required for effective protective immunity. The World Health Organization has made the development of a safe and effective vaccine against Shigella a high priority, because Shigella is showing increasing antibiotic resistance, even to the newest antibiotics [5]. Several strategies to develop vaccines targeting Shigella have been explored; however, no licensed vaccines are available outside of China. Conjugate vaccines, in which Shigella O-specific polysaccharide (O-Ag) is conjugated with protein from other strains, were found to be both safe and immunogenic in children and young adults [6]. Live attenuated vaccines have also been developed, but with such vaccines it is always difficult to properly calibrate attenuation so that immunogenicity is achieved with the minimum of toxicity [7,8]. In addition to live attenuated vaccines, a hybrid vaccine, in which attenuated Shigella bacteria are used as vectors for expressing enterotoxigenic Escherichia coli (ETEC) antigen [9] and a microbial ribosomal vaccine have been proposed as new-generation vaccines [10]. Shigella ribosomal subunit vaccines (SRV) are composed of O-Ag and ribosome, the former being an essential virulent factor on the bacterial surface and a protective antigen of Shigella, the latter acting as an adjuvant to amplify the immune response to the carbohydrate. Previous studies have shown that parenterally administered SRV elicits immunity and protection against Shigella infection in guinea pigs and monkeys [11]. Among all of the proposed vaccine candidates, SRV vaccines seem to be most promising, as they are inexpensive, safe, and immunogenic, making them practical for use in developing countries. However, the exact mechanism underlying SRV-elicited inductive and protective immunity remains elusive. Here, we have tried to purify a safe and highly pure ribosomal vaccine from S. flexneri 2a and have adopted an animal model, pulmonary pneumonia of mice, to investigate its immunogenicity and protective efficacy. Interestingly, mucosal but not parenteral administration with SRV resulted

4829

in efficacious protection in mouse pulmonary infection models. Further, the protection relied on the presence of the secretory form of IgA Ab in mucosal compartments. Taken together, our current results strongly suggest that mucosally delivered SRV could be a promising vaccine strategy for humans.

2. Materials and methods 2.1. Animals Female BALB/c mice were purchased from Charles River Co. (Seoul, Korea), and polymeric immunoglobulin receptor knockout (pIgR−/− ) mice were kindly provided by Dr. Masanobu Nanno (Yakult Central Institute, Tokyo, Japan). The animals were maintained in the animal care facilities of the International Vaccine Institute (Seoul, Korea) under specific pathogen-free conditions and received sterilized food (certified diet MF; Orient Co.) and water ad libitum. 2.2. Isolation of ribosome of S. flexneri 2a A virulent S. flexneri 2a 2457T strains, kindly provided by Dr. Philippe Sansonettii (Pasteur Institute, Paris), were incubated on a 0.1% congo red soybean-casein digest broth (Difco, BD, Sparks, MD) overnight at 37 ◦ C, and then the red colony was selected and cultured with aeration at 37 ◦ C and 160 rpm. Next, a high-pressure homogenizer, modified from a previous study, was used to break down the bacterial ribosome of S. flexneri 2a 2457T, which was then purified by centrifugation and ultra-filtration methods [12]. The O-antigen concentration, measured using the competition ELISA method, accounted for approximately 5% of the preparation. Endotoxin levels were measured using a Limulus amebocyte lysate (LAL) test and LPS were maintained at less than 50 EU/ml. 2.3. Vaccination and challenge The mice were subcutaneously (s.c.) or intranasally (i.n.) vaccinated on days 0 and 14 with a 2.5 ␮g dose of Oantigen under anesthetic conditions. For control experiments, the attenuated S. flexneri 2a SC602 strain (5 × 106 cfu) was administered via the i.n. route. One week after the second vaccination, the mice were challenged i.n. with virulent S. flexneri 2a (1 × 107 or 5 × 107 cfu) to induce pulmonary pneumonia. 2.4. Sample collection One week after the second vaccination, sera were collected from animals anesthetized with ketamine and xylazine hydrochloride by orbital bleeding. To collect saliva, pilocarpine (5 mg/kg of BW; Sigma–Aldrich, St. Louis, MO) was intraperitoneally injected, as described previously [13].

4830

D.-H. Shim et al. / Vaccine 25 (2007) 4828–4836

To obtain lung lavage, mice lungs were cannulated with an intravenous polyethylene catheter (BD Bioscience; San Jose, CA) and washed twice with 0.6 ml of sterile PBS. Nasal and vaginal washes were collected by gently flushing the nasal passage or vaginal canal twice with 50 ␮l of sterile PBS after sacrifice. Fecal extracts were obtained by adding weighed pellets to PBS containing 0.2% sodium azide (1 ml/100 mg fecal sample). The pellet was vortex mixed and centrifuged, and the supernatants were collected for assay as described previously [14]. 2.5. ELISA assay for detection of antigen-specific antibodies ELISA (Enzyme-linked immunosorbent assay) plates (Falcon, Franklin Lakes, NJ) were coated with S. flexneri 2a-derived LPS (5 ␮g/ml) in 50 mM sodium bicarbonate (pH 9.4) and incubated overnight at 4 ◦ C. The plates were incubated with 1% BSA in PBS at 37 ◦ C for 1 h. Two-fold serially diluted samples (starting from 1:32 for serum; 1:2 for fecal extracts, saliva, nasal wash, and lung lavage) were applied to plates and incubated for 2 h at 37 ◦ C. HRP-conjugated goat anti-mouse IgG Ab [Southern Biotechnology Associates (SBA), Birmingham, AL] or goat anti-mouse IgA Ab (SBA) was added to each well and incubated for 4 h at room temperature (RT). For color development, 100 ␮l of tetramethylbenzidin (TMB; Moss, Pasadena, MD) solution was added as a substrate to each well and incubated for 15 min at RT. A 50 ␮l stopping solution (0.5 N HCl) was added, and then plates were measured by ELISA reader (Molecular Devices, Sunnyvale, CA). Endpoint titers of LPS-specific Abs were expressed as the reciprocal log2 of the last dilution giving an optical density at 450 nm of 0.1 greater than background. 2.6. ELISPOT assay for detection of numbers of antibody-forming cells (AFCs) Mononuclear cells (MNCs) were obtained from the spleen, nasal passage (NP), lung, and sub-mandibular gland (SMG) of vaccinated mice, as described previously [15,16]. We coated 96-well nitrocellulose microplates (Millipore, Billerica, MA) with S. flexneri 2a-derived LPS (5 ␮g/ml) and incubated them overnight at 4 ◦ C. Plates were blocked with complete RPMI media (2 mM l-glutamin, 50 ␮M 2-mercaptoethanol, 100 ␮g/ml penicillin and streptomycin sulfate) by incubating at 37 ◦ C for 1 h. Serially diluted MNCs in complete media were applied to the plates, which were then incubated for 4 h in 5% CO2 at 37 ◦ C. Then, HRP-conjugated goat anti-mouse IgG or IgA Abs (SBA) were added to each well and incubated overnight at 4 ◦ C. To develop color, substrate reagents (Moss) were added and color developed at RT until spots could be visualized. Spots were counted under a stereomicroscope (SZ2-ILST; Olympus, Tokyo, Japan).

2.7. Histopathological analysis of tissue samples Twenty-four hours after bacterial infection, the lung was perfused with PBS through the abdominal aorta, removed and fixed in 4% formaldehyde for 1 h at 4 ◦ C. The tissues were dehydrated by gradually soaking in alcohol and xylene and then embedded in paraffin as described previously [17]. The paraffin-embedded specimens were cut into 5-␮m sections and stained with hematoxylin and eosin (H&E) in accordance with the manufacturer’s instructions. The sections were viewed using a digital light microscope (Olympus). 2.8. Induction of pulmonary pneumonia with S. flexneri 2a 2457T Virulent S. flexneri 2a 2457T strains were grown in soybean broth media overnight and harvested using centrifugation for 10 min at 4 ◦ C, 10,000 rpm before being suspended in PBS. To develop pulmonary pneumonia, anesthetized mice were i.n. challenged with 1 × 107 or 5 × 107 cfu in 20 ␮l of PBS. 2.9. Disease score Using a blinded test slightly modified from previous studies [18], we arrived at disease scores by assessing the level of lung tissue destruction, of epithelial cell layer damage, of polymorphonuclear cell infiltration into the inflammation site, and of alveolitis. 2.10. Data and statistical analysis The Ab levels are expressed as mean ± S.D., calculated using log2 reciprocal titers, and compared by the t-test (Sigmaplot program). To determine the statistical significance of the survival rate, the Kaplan–Meier method was used. We ran SPSS 12.0 K for Windows and performed the Log-Rank test (Mantel-Cox) to analyze the statistical significance of survival. All experiments were repeated three times, with five mice per experimental group used each time.

3. Results 3.1. Enhancement of LPS-specific IgG and IgA Abs in mice following vaccination with SRV via the intranasal route To ascertain the immunogenicity of SRV and to determine its ability to elicit protective immunity against Shigella infection, groups of mice were vaccinated with SRV by the intranasal (i.n.) or subcutaneous (s.c.) routes twice at a 2week interval. One week after the second vaccination, sera and mucosal secretions were collected and assessed for antiS. flexneri 2a-derived LPS-specific IgG and IgA Ab levels. As a positive control, an avirulent attenuated S. flexneri 2a SC602

D.-H. Shim et al. / Vaccine 25 (2007) 4828–4836

4831

Fig. 1. Enhancement of systemic and mucosal immunity in mice by intranasal (i.n.) administration with SRV. Each group of mice was vaccinated with PBS or SRV or attenuated S. flexneri 2a SC602 by the subcutaneous (s.c.) or i.n. routes. Following the second vaccination, samples were collected and S. flexneri 2a-derived LPS-specific Ig antibodies (Abs) or Ab-forming cells (AFCs) were measured by ELISA or ELISPOT assay, respectively. *p < 0.05; **p < 0.01; ***p < 0.001; compared to the PBS-vaccinated group. SMG: sub-mandibular gland.

strain was administered via the i.n. route. Both s.c. and i.n. vaccinations with SRV induced high levels of Ag-specific IgG Ab in sera (Fig. 1). As expected, i.n. administration with SRV elicited robust levels of LPS-specific IgA Ab in the mucosal secretions, i.e., nasal washes, saliva, vaginal washes, and fecal extracts. The increased levels were identical to those achieved by i.n. vaccination with the SC602 strain. Using mononuclear cells isolated from spleen (SP), nasal passage (NP), lung, and sub-mandibular gland (SMG) tissues from each of the vaccinated groups, we next determined the number of Ag-specific Ab-forming cells (AFCs) in an effort to ascertain whether the SRV-induced IgA Abs in mucosal compartments had originated in local tissues (Fig. 1). Intranasal administration of SRV elicited significantly higher numbers of LPS-specific IgA AFCs in the NP, lung, and SMG of mice than did s.c. administration. Much higher numbers of IgA AFCs were induced by i.n. vaccination with SRV than with the SC602 strain. These data indicate i.n. vaccination with SRV induces effective systemic and mucosal immune responses. 3.2. Efficacious protection provided by i.n. vaccination with SRV In the murine pulmonary pneumonia model, Shigella can be used as a challenge to trigger inflammation [18]. We have adopted this murine pneumonia infection model to investigate whether i.n. vaccination elicits strong protective efficacy in mice. One week after the second immunization, groups of mice were challenged by instilling two different dosages (5 × 107 and 1 × 107 CFU) of virulent wild-type S. flexneri 2a 2457T into their nostrils. The lungs of some animals underwent H&E staining to determine the severity of the

pneumonia at 24 h after infection, and the remaining challenged mice were monitored daily to assess their survival rate. Disease score was assigned as described in Section 2. Interestingly, groups of mice intranasally vaccinated with SRV or attenuated S. flexneri 2a SC602 demonstrated less severe pneumonia than did those receiving i.n. vaccinations of PBS or s.c. vaccinations of SRV (Fig. 2A). Thus, SRV administration via the parenteral route did not effectively protect against challenge with S. flexneri 2a; in fact, disease scores in animals receiving parenteral SRV were almost identical to levels recorded for the PBS-vaccinated group. Most interestingly, almost 65% of mice receiving i.n. SRV survived the two different challenge doses, in contrast to only approximately 20% of those receiving s.c. vaccinations (Fig. 2B). These data clearly suggest that a higher degree of protective immunity is afforded against shigellosis by SRV when it is administered via the i.n. rather than the parenteral route. 3.3. Intranasal vaccination with SRV elicited long-term B cell memory responses To be considered seriously for clinical application, any newly developed vaccine must be able to induce long-term memory protection. To investigate whether SRV could induce long-term memory responses, we checked LPS-specific IgG and IgA Ab responses at 15 and 28 weeks after the second i.n. vaccination. Interestingly, i.n. SRV-vaccinated mice showed stronger IgG Ab responses in the sera and lung washes and stronger IgA Ab responses in the sera, lung washes, and mucosal secretions (Fig. 3A). In addition, robust numbers of AFCs were detected in the SP, NP, lung, and SMG at week 15 following i.n. administration with SRV

4832

D.-H. Shim et al. / Vaccine 25 (2007) 4828–4836

Fig. 2. Vaccination with SRV by the intranasal (i.n.) route elicited efficacious protection in the murine pulmonary pneumonia model against challenge with wild-type virulent S. flexnri 2a. Panel A shows histological changes of lung and disease scores. (a) Normal lung isolated from non-infected na¨ıve mice, (b) PBS-vaccinated group, (c) s.c. SRV-vaccinated group, (d) i.n. S. flexneri 2a SC602-vaccinated group after infection, and (e) i.n. SRV-vaccinated mice. The bar indicates 100 ␮m. Panel B shows the survival rate 14 days after challenge with two different dosages of virulent S. flexneri 2a. *p < 0.05; **p < 0.01; compared to the PBS-vaccinated group.

Fig. 3. Long-term memory B cell responses were observed at 15 and 28 weeks following intranasal (i.n.) vaccination with SRV. At 15 or 28 weeks following i.n. vaccination with SRV, samples were collected and LPS-specific Ig Abs levels or Ab-forming cells (AFCs) were measured by ELISA (A) or ELISPOT (B) assay, respectively. Panel C shows their survival rate 14 days after challenge with virulent S. flexneri 2a. *p < 0.05; **p < 0.01; ***p < 0.001 compared to the PBS-vaccinated group. SMG: sub-mandibular gland.

D.-H. Shim et al. / Vaccine 25 (2007) 4828–4836

4833

Fig. 4. Induction of systemic and mucosal immunity in TLR-knockout mice by i.n. vaccination with SRV. Wild-type BALB/c, MyD88−/− , TLR2−/− , TLR4−/− , and TLR5−/− mice were twice vaccinated with SRV. One week after the second vaccination, S. flexneri 2a-derived LPS-specific Ig Abs or Ab-forming cells (AFCs) were measured by ELISA or ELISPOT assay, respectively. *p < 0.05; **p < 0.01; ***p < 0.001; compared to wild-type mice. SMG: sub-mandibular gland.

(Fig. 3B). The high levels of LPS-specific IgG and IgA Ab as well as of AFCs induced by i.n. administration with SRV were sustained for 28 weeks. Interestingly, 100% of mice receiving i.n. vaccination with SRV were still living at week 28 after challenge with virulent S. flexneri 2a (Fig. 3C). Taken together, these findings suggest that mucosal vaccination with SRV is an effective regimen for inducing both systemic and mucosal immune responses in mice and for building a two-layered defense system against Shigella infection. 3.4. The immunogenicity by SRV is not mediated by LPS We assumed that in the amplification by SRV of immune responses, O-Ag (free of lipid A) acted as the antigen and the ribosome as an adjuvant. It is possible, however, that the specific molecular patterns present in microbial components [i.e., peptidoglycan (PG), LPS, and flagellin] also play a role in that amplification. To investigate the role of these microbial components in the induction of systemic and mucosal immune responses, we adopted the mice-depleted signals

MyD88, TLR2, TLR4, and TLR5, which recognize specific molecular patterns of PG, LPS, and flagellin [19]. Groups of knockout mice were immunized with SRV twice via the i.n. route and then Shigella-derived LPS-specific Abs were measured. Interestingly, i.n. vaccination with SRV resulted in significant levels of IgG and IgA Abs in the sera and of IgA Ab in the nasal washes of all knockout mice (Fig. 4). Indeed, levels were comparable to those observed in wildtype mice. In contrast, levels of IgA Ab were significantly lower in the lung, saliva, and fecal extracts of MyD88−/− and TLR2−/− mice than in those of wild-type, TLR4−/− and TLR5−/− mice. The number of LPS-specific IgA AFCs in the NP and SMG from TLR4−/− or TLR5−/− mice was identical to those from wild-type mice; however, fewer IgA AFC were elicited in the MyD88−/− and TLR2−/− mice. These results suggest that SRV-induced immunogenicity relies at least in part on the MyD88-dependent TLR2 signal. Most interestingly, neither LPS nor flagellin, contaminants removed during the purification process of SRV, affected the induction of systemic and mucosal immunity following SRV vaccination via the i.n. route.

4834

D.-H. Shim et al. / Vaccine 25 (2007) 4828–4836

Fig. 5. Secretory IgA (S-IgA) Abs are important for the protection induced by intranasal (i.n.) vaccination with SRV. pIgR−/− mice were vaccinated with SRV by the i.n. route, and then LPS-specific Ig Ab levels were measured by ELISA (A). Panel B indicates the survival rate of vaccinated wild-type and pIgR−/− mice 14 days after challenge with virulent S. flexneri 2a. *p < 0.05; **p < 0.01, ***p < 0.001; compared to the PBS-vaccinated group.

3.5. Secretory IgA Ab is important in protecting against Shigella infection Inasmuch as our results clearly demonstrated that mucosal vaccination with SRV is more effective than parenteral vaccination, we next analyzed the role of secretory IgA (SIgA) Ab using polymeric immunoglobulin receptor knockout (pIgR−/− ) mice which are unable to deliver dimeric IgA and pentameric IgM into the mucosa from submucosal regions [20]. As expected, pIgR−/− mice elicited significant levels of Ag-specific IgG and IgA Ab in serum and lung lavage but not in the saliva and fecal extracts following two i.n. administrations of SRV (Fig. 5A). We assumed that the high levels of IgA Ab in the nasal washes of pIgR−/− mice might be derived from serum, since our methods for obtaining nasal washes can cause contamination by serum components. To further clarify the role played by S-IgA Abs in protecting against Shigella infection, pIgR−/− mice were challenged with virulent S. flexneri 2a after receiving two i.n. vaccinations with SRV. Most interestingly, only approximately 20% of pIgR−/− mice survived the challenge, in comparison to 90% of wild-type mice (Fig. 5B). Indeed, the survival rate for pIgR−/− mice vaccinated with i.n. SRV was identical to that for PBS-vaccinated mice. Overall, it can be surmised that S-IgA Abs induced by i.n. vaccination play a decisive role in inducing efficient protection against Shigella infection.

4. Discussion In this study, we have demonstrated that mucosal (i.e., i.n.) but not parenteral (i.e., s.c.) administration of Shigella ribosomal-based vaccine (SRV) resulted in robust levels of LPS-specific Abs in both systemic and mucosal compartments. In mice receiving an initial i.n. vaccination with SRV followed by a second i.n. administration 2 weeks later, resistance was enhanced against local Shigella infection in the lung. Further, the induction of systemic and mucosal immu-

nity by SRV was not due to contamination by LPS and flagellin even though these microbial components are well known to possess strong adjuvanticity. These results indicate that SRV vaccination via mucosal compartments could be a promising shigellosis-targeted vaccine regimen. Existing ribosomal-based vaccines developed from organisms such as bacteria, fungi, and protozoa have been shown to possess strong immunogenicity and protective efficacy [10]. Such vaccines have been adopted as useful tools in preventing several infectious diseases such as Pseudomonas aeruginosa, Streptococcus sp., Klebsiella pneumoniae, and Heamophilus influenzae [21,22]. Berry et al. suggested that the ribosomal particle serves as an efficient delivery system for weak antigen [9]. The first ribosomes to be isolated from avirulent S. sonnei and S. flexneri were obtained using sonic disruption followed by differential ultracentrifugation [12]. Levenson et al. predicted that the formation of immunogenic O-Ag and ribosome complexes would be of considerable importance because these complexes can increase the efficiency of antigen presentation [23]. Another study demonstrated that parenteral vaccination with SRV elicited a significant level of O-Ag-specific IgG Ab in serum and of IgA Ab in the tears of guinea pigs, as well as in the saliva and bile of monkeys [11]. Another attractive feature of SRV is that it delivers effective protection at a low cost. Using Levenson’s method with modifications, we have attempted to purify ribosome from avirulent S. flexneri 2a that could be used for clinical trials in developing countries. Our newly purified SRV are both safe and effective at enhancing immunogenicity and protection, qualities that suggest the potential magnitude of their contribution to shigellosis vaccine development in humans. Although previous studies suggested that parenteral SRV vaccination was capable of eliciting immune responses [23], our current study strongly suggests that vaccination via the mucosal route is essential for the induction of effective immunity and protection against Shigella infection of mucosal tissues. An advantage of mucosal vaccination is that it can induce both mucosal and systemic immune responses, result-

D.-H. Shim et al. / Vaccine 25 (2007) 4828–4836

ing in two layers of host protection against infectious diseases [3]. Also important in immune responses at mucosal surfaces is the production of S-IgA and its transport across the epithelium [24]. S-IgA is one of the cardinal components of the mucosal immune system, and its production upon mucosal immunization leads to strong protective responses against bacterial and viral antigens [25]. In the development of effective mucosal vaccines, it is crucial to select the appropriate immunization route. Most current means of mucosal vaccine delivery are intended to mimic natural encounters of mucosal inductive sites with environmental antigens and pathogens [4]. For instance, administration of conjugate vaccine with proteosome and Shigella-derived LPS by the i.n. or intragastric routes induced strong protection against pneumonia in mice [18]. In addition, a previous study demonstrated that S-IgA plays a key role in protecting against bacterial invasion and in sustaining Salmonella typhimurium in the large intestines of pIgR−/− mice [20]. Thus, one of the criteria in screening Shigella vaccine candidates should be their ability to induce S-IgA in the mucosal compartments. Shigella preferentially invade the colon, provoking severe inflammation in humans but not in mice, guinea pigs, and rabbits, which are generally resistant to the invasion of this bacteria into the intestinal epithelium [26]. Indeed, a serious obstacle to the development of a Shigella vaccine is posed by the lack of an animal shigellosis model and the resulting dearth of reliable animal studies screening potential vaccine candidates. Many approaches to using mice, guinea pigs, and macaques as possible animal models of bacillary dysentery have been explored [11,26–28]. Among them, we chose the murine pneumonia model to screen the vaccine efficacy of SRV because it manifested inflammation of the mucosal epithelium closely mirroring that seen in human shigellosis [18]. Our current results clearly demonstrate that i.n. but not s.c. vaccination of mice with SRV elicited strong protective immunity to pulmonary pneumonia as well as high levels of systemic and mucosal immunity. Mucosal administration with SRV would then seem to be the privileged form of administration for achieving the highest levels of protective immunity. Toll-like receptor (TLR) stimulation by various pathogenassociated molecular patterns leads to activation of innate and subsequent adoptive immune responses [19]. TLR have received special attention as potent adjuvant receptors. In this regard, recent studies have demonstrated that TLR agonists such as CpG DNA, microbial lipoprotein, LPS, and flagellin can trigger dendritic cells to mature [29,30], raising the possibility that the strong immunogenicity demonstrated by ribosomal vaccines may be due to contamination with such cell membrane materials. Our current results rule out that possibility, demonstrating that i.n. SRV-induced immunogenicity is not affected by potential contamination by LPS or flagellin but relies in part on the MyD88dependent TLR2 signal for induction of IgA Abs in the lung, saliva, and fecal extracts. However, it remains a subject of debate whether TLRs are necessary for response to

4835

vaccines [31,32]. Innate immune signals mediated by TLRs have been thought to make a considerable contribution to acquired immune responses [31]. However, a recent study revealed that mice deficient in all TLR signals nonetheless mounted robust levels of Ab responses to T cell-dependent antigen with several well-known adjuvants [32]. Our current results demonstrate that i.n. vaccination with SRV elicited normal levels of antigen-specific acquired immunity in TLRdeficient mice, even though mucosal IgA Ab responses are thought to rely in part on the MyD88-dependent TLR2 signal (Fig. 4). Taken together, these results support the notion that ribosome in the SRV might induce antigen-specific immunity in both systemic and mucosal compartments in the absence of TLR signaling. Further studies will be needed to elucidate these TLR-independent mechanisms for immune induction. The adjuvanticity of ribosome remains a somewhat controversial subject. A previous study proposed that ribosome acts as an adjuvant and a carrier for O-antigen and other cell wall antigens [33]. Venemman and Bigley demonstrated that treatment with RNase, Pronase or trypsin did not reduce the immunogenicity of ribosomal vaccines [34]. In contrast, others postulated that RNase treatment could reduce the immunogenicity of ribosomal vaccines and that the RNA-rich part of the ribosome could induce immunity but not protective antigens in the absence of O-Ag [35]. Yumans demonstrated that the immunogenicity of SRV vaccines depended on the purity of the RNA and so underlined the importance of the RNA preparation process [36]. Although immune specificity may depend on the given type of O-Ag, it can be surmised that ribosome likely plays an important role in inducing immune responses. However, the mechanism by which it does so – whether as an effective carrier or as a delivery system for various antigens – remains elusive. By demonstrating that i.n. administration with SRV induces strong immunogenicity, long-term memory cells, and protective efficacy in mice, the findings of this study support the candidacy of SRV as a promising and effective mucosal vaccine targeting shigellosis in human beings.

Conflict of interest The authors have no conflicting financial interests.

Acknowledgments This work was supported by the governments of the Republic of Korea, Sweden, and Kuwait and the Korean Ministry of Science and Technology and Grant No. RT104-01-01 from the Regional Technology Innovation Program of the Ministry of Commerce, Industry and Energy (MOCIE). We thank Dr. Hyun-Jeong Ko of Seoul National University for statistical analysis and Dr. Kimberly McGhee of the Medical College of Georgia for editing of the manuscript.

4836

D.-H. Shim et al. / Vaccine 25 (2007) 4828–4836

References [1] Niyogi SK. Shigellosis. J Microbiol 2005;43(2):133–43. [2] Sansonetti PJ. War and peace at mucosal surface. Nat Rev Immunol 2004;4(12):953–64. [3] McGhee JR, Kiyono H. The mucosal immune system. In: Paul WE, editor. Fundamental Immunology. 4th ed. Lippincott-Raven Publishers; 1999. p. 909. [4] Czerkinsky C, Holmgren J. Mucosal immunity and vaccines. Nature Med 2005;11(4):S45–53. [5] Kotloff KL, Winickoff JP, Ivanoff B, Clemens JD, Swerdlow DL, Sansonetti PJ, et al. Global burden of Shigella infections: implications for vaccine development and implementation of control strategies. Bull World Health Organ 1999;77(8):651–66. [6] Passwell JH, Harlev E, Ashkenazi S, Chu C, Miron D, Ramon R, et al. Safety and immunogenicity of improved Shigella O-specific polysaccharide-protein conjugate vaccines in adults in Israel. Infect Immun 2001;69(3):1351–7. [7] Coster TS, Hoge CW, VanDeVerg LL, Hartman AB, Oaks EV, Vankatesan MM, et al. Vaccination against shigellosis with attenuated Shigella flexneri 2a strain SC602. Infect Immun 1999;67(7):3437–43. [8] Katz DE, Coster TS, Wolf MK, Trespalacios FC, Cohen D, Robins G, et al. Two studies evaluating the safety and immunogenicity of a live, attenuated Shigella flexneri 2a vaccine (SC602) and excretion of vaccine organisms in North American volunteers. Infect Immun 2004;72(2):923–30. [9] Berry EM, Wang J, Wu T, Davis T, Levine MM. Immunogenicity of multivalent Shigella-ETEC candidate vaccine strains in a guinea pig model. Vaccine 2006;24(18):3728–34. [10] Gregory RL. Microbial ribosomal vaccines. Rev Infect Dis 1986;8(2): 208–17. [11] Levenson VI, Chernokhvostova EV, Lyubinskaya MM, Salamatova SA, Dzhikidze EK, Stasilevitch ZK. Parenteral immunization with Shigella ribosomal vaccine elicits local IgA response and primes for mucosal memory. Int Arch Aller Appl Immun 1988;87(1):25–31. [12] Levenson VI, Subbotina IuL, Gofman IL, Iarovaia LM. Ribosomal dysentery vaccine. I. Method of production, physical and chemical properties. Zh Mikrobiol Epidemiol Immunobiol 1978;(3): 60–5. [13] Kweon MN, Yamamoto M, Watanabe F, Tamura S, Van Ginkel FW, Kiyono H. A nontoxic chimeric enterotoxin adjuvant induces protective immunity in both mucosal and systemic compartments with reduced IgE antibodies. J Infect Dis 2002;186(9):1261–9. [14] Yamamoto S, Kiyono H, Yamamoto M, Imaoka K, Fujihashi K, Van Ginkel FW, et al. A nontoxic mutant of cholera toxin elicits Th2-type responses for enhanced mucosal immunity. Proc Natl Acad Sci USA 1997;94(10):5267–72. [15] Hagiwara Y, McGhee JR, Fujihashi K, Kobayashi R, Yoshino N, Kataoka K, et al. Protective mucosal immunity in aging is associated with functional CD4+ T cells in nasopharyngeal-associated lymphoreticular tissue. J Immunol 2003;170(4):1754–62. [16] Yamamoto M, Kweon MN, Rennert PD, Hiroi T, Fujihashi K, McGhee JR, et al. Role of gut-associated lymphoreticular tissues in antigenspecific intestinal IgA immunity. J Immunol 2004;173(2):762–9. [17] Kweon MN, Yamamoto M, Rennert PD, Park EJ, Lee AY, Chang SY, et al. Prenatal blockage of lymphotoxin ␤ receptor and TNF receptor p55 signaling cascade resulted in the acceleration of tissue genesis for isolated lymphoid follicles in the large intestine. J Immunol 2005;174(7):4365–72. [18] Mallett CP, Hale TL, Kaminski RW, Larsen T, Orr N, Cohen D, et al. Intransal or intragastric immunization with proteosome-Shigella

[19] [20]

[21]

[22]

[23]

[24] [25]

[26]

[27]

[28]

[29]

[30]

[31] [32]

[33]

[34]

[35]

[36]

lipopolysaccharide vaccines protects against lethal pneumonia in a murine model of Shigella infection. Infect Immun 1995;63(6):2382–6. Akira S, Takeda K. Toll-like receptor signalling. Nat Rev Immunol 2004;4(7):499–511. Wijburg OL, Uren TK, Simpfendorfer K, Johansen FE, Brandtzaeg P, Strugnell RA. Innate secretory antibodies protect against natural Salmonella typhimurium infection. J Exp Med 2006;203(1):21–6. Lieberman MM, Wright GL, Wolcott KM, McKissock-Desoto DC. Polyvalent antisera to Pseudomonas ribosomal vaccines: protection of mice against clinically isolated strains. Infect Immun 1980;29(2): 489–93. Serrano E, Demanez ZP, Morqon A, Chastang C, Van Cauwenberge P. Effectiveness of ribosomal fractions of Klebsiella pneumoniae, Streptococcus pneumoniae, Streptococcus pyogenes, Haemophilus influenzae and the membrane fraction of Kp (Ribomunyl) in the prevention of clinical recurrences of infectious rhinitis. Results of a multicenter double-blind placebo-controlled study. Eur Arch Otorhinolaryngol 1997;254(8):372–5. Levenson VI, Egorova TP. Polysaccharide nature of O antigen in protective ribosomal preparations from Shigella: experimental evidence and implications for the ribosomal vaccine concept. Res Microbiol 1990;141(6):707–20. Brandtzaeg P, Pabst R. Let’s go mucosal: communication on slippery ground. Trends Immunol 2004;25(11):570–7. Kurohane K, Kobayashi C, Imai Y. Facilitated production of secretory IgA against Shiga toxin B subunits by intranasal application of antigen-coated polystyrene microspheres. Microbiol Immunol 2005;49(2):149–54. Shim DH, Suzuki T, Chang SY, Park SM, Sansonetti PJ, Sasakawa C, et al. New animal model of shigellosis in the Guinea pig: its usefulness for protective efficacy studies. J Immunol 2007;178(4):2476–82. Singer M, Sansonetti PJ. IL-8 is a key chemokine regulating neutrophil recruitment in a new mouse model of Shigella-induced colitis. J Immunol 2004;173(6):4197–206. Zhang Z, Jin L, Champion G, Seydel KB, Stanley Jr SL. Shigella infection in a SCID mouse-human intestinal xenograft model: role for neutrophils in containing bacterial dissemination in human intestine. Infect Immun 2001;69(5):3240–7. Hertz CJ, Kiertscher SM, Godowski PJ, Bouis DA, Norquard M, Roth MD, et al. Microbial lipopeptides stimulate dendritic cell maturation via Toll-like receptor 2. J Immunol 2001;166(4):2444–50. Lee SE, Kim SY, Jeong BC, Kim YR, Bae SJ, Ahn OS, et al. A bacterial flagellin, Vibrio vulnificus FlaB, has a strong mucosal adjuvant activity to induce protective immunity. Infect Immun 2006;74(1):694–702. Pasare C, Medzhitov R. Control of B-cell responses by Toll-like receptors. Nature 2005;438(7066):364–8. Gavin AL, Hoebe K, Duong B, Ota T, Martin C, Beutler B, et al. Adjuvant-enhanced antibody responses in the absence of toll-like receptor signaling. Science 2006;314(5807):1936–8. Normier G, Pinel AM, Dussourd d’Hinterland L, Wiqzell H, Binz H. Ribosomes as carriers for antigenic determinants of the surface of micro-orgamisms. Dev Biol Stand 1992;77:79–85. Venneman MR, Bigley NJ. Isolation and partial characterization of an immunogenic moiety obtained from Salmonella typhimurium. J Bacteriol 1969;100(1):140–8. Kita E, Oku D, Nishikawa F, Emoto M, Yasui K, Kashiba S. Requirement of the conformational stability of a Salmonella ribosomal vaccine for its mouse protection. FEMS Microbiol Immunol 1991;3(4):229–39. Youmans AS, Youmans GP. Factors affecting immunogenic activity of mycobacterial ribosomal and ribonucleic acid preparations. J Bacteriol 1969;99(1):42–50.