And Oral Acquired Brucellosis Self-Adjuvanting Vaccine Against

The Protein Moiety of Brucella abortus Outer Membrane Protein 16 Is a New Bacterial Pathogen-Associated Molecular Pattern That Activates Dendritic Cells In Vivo, Induces a This information is current as Th1 Immune Response, and Is a Promising of September 28, 2021. Self-Adjuvanting Vaccine against Systemic and Oral Acquired Brucellosis Karina A. Pasquevich, Clara García Samartino, Lorena M. Coria, Silvia M. Estein, Astrid Zwerdling, Andrés E. Ibañez, Downloaded from Paula Barrionuevo, Fernanda Souza de Oliveira, Natalia Barbosa Carvalho, Julia Borkowski, Sergio Costa Oliveira, Heribert Warzecha, Guillermo H. Giambartolomei and Juliana Cassataro

J Immunol 2010; 184:5200-5212; Prepublished online 29 http://www.jimmunol.org/ March 2010; doi: 10.4049/jimmunol.0902209 http://www.jimmunol.org/content/184/9/5200

Supplementary http://www.jimmunol.org/content/suppl/2010/03/26/jimmunol.090220 Material 9.DC1 by guest on September 28, 2021 References This article cites 74 articles, 31 of which you can access for free at: http://www.jimmunol.org/content/184/9/5200.full#ref-list-1

Why The JI? Submit online.

• Rapid Reviews! 30 days* from submission to initial decision

• No Triage! Every submission reviewed by practicing scientists

• Fast Publication! 4 weeks from acceptance to publication

*average

Subscription Information about subscribing to The Journal of Immunology is online at: http://jimmunol.org/subscription Permissions Submit copyright permission requests at: http://www.aai.org/About/Publications/JI/copyright.html

The Journal of Immunology is published twice each month by The American Association of Immunologists, Inc., 1451 Rockville Pike, Suite 650, Rockville, MD 20852 Copyright © 2010 by The American Association of Immunologists, Inc. All rights reserved. Print ISSN: 0022-1767 Online ISSN: 1550-6606. Email Alerts Receive free email-alerts when new articles cite this article. Sign up at: http://jimmunol.org/alerts Downloaded from http://www.jimmunol.org/ by guest on September 28, 2021

The Protein Moiety of Brucella abortus Outer Membrane Protein 16 Is a New Bacterial Pathogen-Associated Molecular Pattern That Activates Dendritic Cells In Vivo, Induces a Th1 Immune Response, and Is a Promising Self-Adjuvanting Vaccine against Systemic and Oral Acquired Brucellosis

Karina A. Pasquevich,*,† Clara Garcı´a Samartino,*,† Lorena M. Coria,*,† Silvia M. Estein,‡ Astrid Zwerdling,*,† Andre´sE.Iban˜ez,*,† Paula Barrionuevo,*,† Fernanda Souza de Oliveira,x Natalia Barbosa Carvalho,x Julia Borkowski,{ Sergio Costa Oliveira,x Heribert Warzecha,{ Guillermo H. Giambartolomei,*,† and Juliana Cassataro*,† Downloaded from

Knowing the inherent stimulatory properties of the lipid moiety of bacterial lipoproteins, we first hypothesized that Brucella abortus outer membrane protein (Omp)16 lipoprotein would be able to elicit a protective immune response without the need of external adjuvants. In this study, we demonstrate that Omp16 administered by the i.p. route confers significant protection against B. abortus infection and that the protective response evoked is independent of the protein lipidation. To date, Omp16 is the first Brucella protein that without the requirement of external adjuvants is able to induce similar protection levels to the control live vaccine S19. Moreover, http://www.jimmunol.org/ the protein portion of Omp16 (unlipidated Omp16 [U-Omp16]) elicits a protective response when administered by the oral route. Either systemic or oral immunization with U-Omp16 elicits a Th1-specific response. These abilities of U-Omp16 indicate that it is endowed with self-adjuvanting properties. The adjuvanticity of U-Omp16 could be explained, at least in part, by its capacity to activate dendritic cells in vivo. U-Omp16 is also able to stimulate dendritic cells and macrophages in vitro. The latter property and its ability to induce a protective Th1 immune response against B. abortus infection have been found to be TLR4 dependent. The facts that U-Omp16 is an oral protective Ag and possesses a mucosal self-adjuvanting property led us to develop a plant-made vaccine expressing U-Omp16. Our results indicate that plant-expressed recombinant U-Omp16 is able to confer protective immunity, when given orally, indicating that a plant-based oral vaccine expressing U-Omp16 could be a valuable approach to controlling this disease. The Journal by guest on September 28, 2021 of Immunology, 2010, 184: 5200–5212.

rucella abortus is a zoonotic Gram-negative pathogen that the protective immune response are closely related to this property causes abortion and infertility in ruminants. In humans, it (2). Although both Ab- and cell-mediated immune responses can B causes undulant fever, characterized by malaise, aches, inﬂuence the course of infection with Brucella, the latter are es- and fevers. Human brucellosis can be contracted by accidental sential for clearance of intracellular bacteria. In this respect, IFN-g contamination from infected animals, handling infected tissues, or plays a central role in acquired resistance against Brucella, upreg- consuming undercooked meat or unpasteurized dairy products ulating macrophage microbial killing (3, 4). from infected animals (1). Brucella is a facultative intracellular Because brucellosis has serious medical and economic con- bacterial parasite; the pathogenesis of brucellosis and the nature of sequences, prevention of animal infection by vaccination is key.

*Laboratory of Immunogenetics, Clinical Hospital Jose´ de San Martıń, School of C.G.S., and L.M.C. are recipients of a fellowship from Consejo Nacional de Investigaciones Medicine and †Institute of Studies on Humoral Immunity (National Council of Re- Cientificas y Tecnicas (Argentina). A.E.I. is a recipient of a fellowship from Agencia Na- search), School of Pharmacy and Biochemistry, University of Buenos Aires, Buenos cional de PromociońCientı´fica y Tecnolo´gica (Argentina). S.M.E., P.B., G.H.G., and J.C. are Aires; ‡Laboratory of Immunology, Department of Animal Healthcare and Preventive members of the Research Career of Consejo Nacional de Investigaciones Cientificas Medicine, School of Veterinary Sciences, National University of the Center of the y Tecnicas. Buenos Aires Province, Tandil, Argentina; xDepartment of Biochemistry and Immu- Address correspondence and reprint requests to Dr. Juliana Cassataro, Laboratorio de nology, Institute of Biological Sciences, Federal University of Minas Gerais, Belo { Inmunogene´tica, Hospital de Clıńicas Jose´ de San Martıń, Facultad de Medicina, Horizonte-Minas Gerais, Brazil; and Technische Universita¨t Darmstadt, Institute of University of Buenos Aires, Co´rdoba 2351 3er Piso Sala 4 1120, Buenos Aires, Botany, Darmstadt, Germany Argentina. E-mail address: [email protected] Received for publication July 13, 2009. Accepted for publication February 25, 2010. The online version of this article contains supplemental material. This work was supported by grants from the Agencia Nacional de PromociońCientı´fica y Abbreviations used in this paper: BM, bone marrow; DC, dendritic cell; DTH, de- Tecnolo´gica: PICT 2006 No. 1670 and No. 1335 (to J.C. and G.H.G.); from Consejo Nacional layed-type hypersensitivity; HS, B. ovis REO 198 hot saline extract; i.g., intragastri- de Investigaciones Cientificas y Tecnicas: PIP 5239 and 5213 (to J.C. and G.H.G.); from cally; L-, lipidated; M, marker lane; Omp, outer membrane protein; Pam Cys, N- Universidad de Buenos Aires: M408 and B819 (to J.C. and G.H.G.); from Conselho Nacional 3 palmitoyl-S-(2RS)-2,3-bis-(palmitoyloxy)propyl-cysteine; PAMP, pathogen-associ- de Desenvolvimento Cientı´fico e Tecnolo´gico/Prosul No. 490485/2007-3, Conselho Nacio- ated molecular pattern; PB, polymyxin B; U-, unlipidated; wt, wild-type. nal de Desenvolvimento Cientı´fico e Tecnolo´gico/Agencia Nacional de PromociońCientı´fica ´ ˆ yTecnologica No. 490528/2008-2, and Instituto Nacional de Ciencia e Tecnologia de Vacinas Ó (to S.C.O); and from Agencia Nacional de PromociońCientı´fica y Tecnolo´gica/Conselho Copyright 2010 by The American Association of Immunologists, Inc. 0022-1767/10/$16.00 Nacional de Desenvolvimento Cientı´fico e Tecnolo´gico PICT 2008 No. 18 (to J.C.). K.A.P., www.jimmunol.org/cgi/doi/10.4049/jimmunol.0902209 The Journal of Immunology 5201

Thus, efforts have been made to prevent the infection through the use Materials and Methods of vaccines (3, 5). All commercially available brucellosis vaccines Mice are based on live, attenuated strains of Brucella. Although effective, Female 6- to 8-wk-old specific-pathogen-free BALB/c mice were purchased these vaccines have disadvantages: they can be infectious for hu- from the Universidad Nacional de La Plata (La Plata, Argentina). The wild- mans; they can interfere with diagnosis; they may result in abor- type (wt) strain C57BL/6 mice were provided by Federal University of tions when administered to pregnant animals; and the vaccine strain Minas Gerais (Belo Horizonte, Brazil), and genetically deficient TLR22/2, 2/2 2/2 can spread in the region (3, 6). Currently, no vaccine against human TLR4 , and TLR6 C57BL/6 mice were provided by S. Akira (Osaka brucellosis is available (7). Therefore, improved vaccines that University, Osaka, Japan). Mice were housed in appropriate conventional animal care facilities and handled according to international guidelines combine safety and efficacy to all species at risk need to be de- required for animal experiments. Animals were used in the experiments 8– signed. Within the past few years, we and others have made efforts 10 wk postbirth. to develop a vaccine without these drawbacks, a vaccine that would be more effective and safer than those used at present (8–11). Bacteria Subunit vaccines, like recombinant proteins, are promising The virulent B. abortus 544 and B. abortus 2308 (used in the protection vaccine candidates, because they can be produced at high yield and experiments), B. ovis REO 198 (used for production of the antigenic purity and can be manipulated to maximize desirable activities and preparation B. ovis REO 198 hot saline extract [HS]), and the vaccine strain B. abortus S19 and B. abortus RB51 were obtained from our own minimize undesirable ones. Moreover, they are safer for manip- laboratory collection (8, 33). Bacterial growth and inocula preparation ulators, well defined, not infectious, and cannot revert to a virulent were performed as previously described (33, 34). All live Brucella ma- strain. However, despite these advantages, they tend to be poorly nipulations were conducted in biosafety level 3 facilities. immunogenic in vivo, and require the coadministration of adjuvants Downloaded from that indirectly enhance the immune response against recombinant Ags proteins. Therefore, recombinant vaccine success is usually de- The recombinant lipidated (L-) and unlipidated (U-) Omp16 proteins were pendent on the use of substances endowed with immunomodulatory manufactured as previously described (8, 32). Briefly, recombinant properties that instruct and control the selective induction of the L-Omp16 was isolated from bacterial membranes, and U-Omp16 was isolated from bacterial cytoplasm and then purified by affinity chroma- appropriate type of Ag-specific immune response (12–14). tography with an Ni-NTA resin (Qiagen, Dorking, U.K.). Expression and

A promising immune stimulator is the lipid moiety N-palmitoyl- purification of the recombinant proteins were checked by SDS-PAGE, http://www.jimmunol.org/ S-(2RS)-2,3-bis-(palmitoyloxy)propyl-cysteine (Pam3Cys). This followed by silver staining. To confirm the identity of the Omp16, Western moiety is found at the N terminus of bacterial lipoproteins and is blot was performed and developed with anti–Omp16-specific mAb (data not shown). Protein concentration was determined by the bicinchoninic ubiquitous in and unique to bacteria. Synthetic peptides that are not acid assay with BSA as a standard (Pierce, Rockford, IL). LPS contami- immunogenic themselves induce a strong Ab response when co- nation was adsorbed with Sepharose-polymyxin B (PB) (Sigma-Aldrich, valently coupled to Pam3Cys (15, 16). In vitro and in vivo evidence St. Louis, MO). Endotoxin determination was performed with Limulus indicates that bacterial lipoproteins and synthetic lipoprotein ana- amoebocyte assay (Associates of Cape Cod, Woods Hole, MA). All protein , logs (lipopeptides) are potent activators of innate immune cells and preparations contained 0.25 endotoxin U/mg protein. HS was obtained as previously described (34). In some experiments, as that these pathogen-associated molecular patterns (PAMPs) trigger control, a U-Omp16 enzymatically digested preparation was used. U- cellular activation by binding to pattern recognition receptors on Omp16 was treated with proteinase K-agarose from Tricirachium album by guest on September 28, 2021 the surfaces of monocytes/macrophages and dendritic cells (DCs) (Sigma-Aldrich) for 2 h at 37˚C, following manufacturer indications. The (17, 18). Numerous studies also report different proteins with im- enzyme immobilized in agarose was then centrifuged out (2000 3 g,5 munoadjuvant properties, such as outer membrane protein (Omp) min), and the supernatants were incubated for 1 h at 60˚C to inactivate any fraction of soluble enzyme. The complete digestion of the proteins was A from Klebsiella pneumoniae, Neisseria meningitidis porins, and checked by SDS-PAGE, followed by Coomassie blue staining. the Tat (transcriptional transactivator) protein of HIV-1. These proteins are able to induce strong immune responses in the absence Immunizations and experimental design of external adjuvants (19–22). Groups of 5–8 mice were anesthetized with methoxyflurane (Mallinckrodt, If we take into account that in brucellosis the infection route Phillipsburg, NJ) and then immunized. usually involves the entry of the pathogen through the mucosal Some groups of mice were immunized on days 0 and 15 by the i.p. route surfaces (23), another key point is the induction of immune re- with 30 mg of 1) L-Omp16, 2) U-Omp16 in PBS, 3) U-Omp16 emulsified sponses on the mucosal surfaces. Therefore, an oral vaccine could in IFA, or 4) PBS-immunized mice as control. As positive control in the protection experiments, a group of mice received a single dose i.p. of 1 3 be a promising candidate to control the disease or to enhance the 104 CFU live B. abortus strain 19. immune protection provided by currently available vaccines. The Other groups of mice were intragastrically (i.g.) immunized with three protective capacity of several vaccine preparations has already consecutive weekly doses of U-Omp16 (100 mg) or PBS in 200 ml 0.1 M been evaluated by the oral route, including live attenuated strains bicarbonate buffer (pH 8), as previously described (8, 28). As positive control, a group of mice received a single dose of live B. abortus RB51 (24, 25), live recombinant vectors (26, 27), and a few recombinant (0.5 3 109 UFC) by the i.g. route. proteins with mucosal adjuvants (8, 28). At 30 d after the last immunization, mice were challenged with virulent Omp16 of Brucella spp. is a lipoprotein and is expressed in all B. abortus for protection experiments or were sacrificed to perform cellular biovars of B. abortus, B. melitensis, B. suis, B. canis, B. ovis, and in vitro experiments; other groups of immunized mice were injected in B. neotomae (29–31). We have previously reported that B. abortus their footpads to perform the delayed-type hypersensitivity (DTH) test. Sera were obtained from blood samples on days 0, 15, 30, and 45 after the Omp16 confers protection against a challenge with virulent B. first immunization. abortus when administered i.p. with systemic adjuvants (IFA or aluminum hydroxide) or orally with a mucosal adjuvant (cholera Brucella challenge and protective response evaluation toxin) (8). Omp16 is also able to activate monocytes, inducing the At 1 mo after the last immunization, mice were challenged i.p. with 4 3 104 production of proinflammatory cytokines, a phenomenon medi- CFU live B. abortus 544 or 2308 or i.g. with 3 3 108 CFU B. abortus ated by its lipid moiety (32). 2308. At 1 mo postchallenge, their spleens where aseptically removed, We speculate that the N-terminal lipid moiety of Omp16 could homogenized in sterile PBS, diluted, plated, and incubated as described (8, 33), and the number of B. abortus 544 or 2308 CFU was counted. Results confer enough adjuvanticity to evoke a protective immune response were represented as the mean log10 CFU 6 SD per group. Units of pro- against Brucella, abolishing the need of external adjuvants. In the tection were calculated as the difference between the mean log10 CFU present work, we evaluate this hypothesis. from the PBS group and the mean log10 CFU from the experimental group. 5202 B. abortus U-Omp16 IS A SELF-ADJUVANTING VACCINE

In vitro cellular responses Differentiated BM-derived DCs or macrophages were stimulated for 24 h with complete medium alone or complete medium with: U-Omp16 (1, 5, 10, Single spleen cell suspensions from immunized and control mice were or 50 mg/ml), E. coli LPS (1 mg/ml), or Pam Cys (1 mg/ml) (InvivoGen, 3 6 3 cultured in duplicate at 4 10 cells/ml in RPMI 1640 (Life Technologies San Diego, CA). A plant-made U-Omp16 was used as the stimulant protein BRL, Grand Island, NY) supplemented with 10% FCS (Life Technolo- in some assays. For control groups, in some experiments, cells were gies), 1 mM sodium pyruvate, 2 mM L-glutamine, 100 U penicillin/ml, stimulated either with proteinase K-digested U-Omp16 (50 mg/ml) or with and 100 mg streptomycin/ml (complete medium) with stimuli. The dif- U-Omp16 (50 mg/ml) plus PB (15 mg/ml). After stimulation, cell-free ferent stimuli were HS (20 mg/ml), U-Omp16 (50 mg/ml), or complete supernatants were collected and assayed by sandwich ELISA for TNF-a medium alone. When HS was used as the stimulus, the complete medium and IL-12 (p40) production, using paired cytokine-specific mAbs accord- also contained PB (15 mg/ml; Sigma-Aldrich). After 72 h incubation at ing to the manufacturer’s instructions (R&D Systems, Minneapolis, MN). 37˚C in a humidified atmosphere (5% CO2 and 95% air), cell culture supernatants were collected. IFN-g, IL-2, IL-4, IL-5, and IL-10 production was analyzed by sandwich ELISA, using paired cytokine-specific mAbs TransientexpressionofU-Omp16inNicotiana benthamina plants according to the manufacturer’s instructions (Pharmingen, San Diego, + + The sequence of omp16 lacking the segment coding for the N-terminal CA). In some experiments, splenocytes were depleted of CD4 or CD8 signal peptide was amplified by PCR, using primers PO16-102 (59-TTTG- T cells, using mouse CD4 (L3T4) or mouse CD8 (Lyt2) Dynabeads ac- GTCTCAAGGTATGAACCTTCCGAATAATGCCGGTGATCTGGGAC- cording to the manufacturer’s instructions (Dynal Biotech, Oslo, Norway). TCG-39) and PO16-201 (59-GCTCTAGATCAGTGGTGGTGGTGGTGG- Depletion was performed before antigenic stimulation. Each T cell pop- TGCTC-39). Primer PO16-201 also included the nucleotides encoding for ulation was depleted with an efficacy greater than 99%, as determined by a63 histidine extension that was already present in the templatevector [pET- flow cytometry analysis (not shown). After being depleted, cells were U-Omp16 (32)]. The PCR product was cloned into the vector pCRblunt suspended in the original volume of complete medium and used in the (Invitrogen, Karlsruhe, Germany), and its identity was confirmed by se- in vitro stimulation assay. quencing. For subcloning, a BsaI–XbaI fragment from this vector was in-

serted into the vector pICH10990 (38); the resulting vector was termed Downloaded from DTH test PO16-6121 and mobilized into agrobacteria by chemical transformation (39). Infiltration of N. benthamina plants was carried out as described by At 4 wk after the last i.g. immunization, mice were injected intradermally in Marillonnet et al. (38). one footpad with 30 mg U-Omp16 and in the contralateral footpad with an equal volume of saline, as negative control. The footpad thickness was measured 24, 48, and 72 h later by using a digital caliper with a precision of Plant-expressed protein extraction, purification, and 0.01 mm, and the mean increase in footpad thickness (expressed in mm) was immunoblot analysis calculated according to the following formula: (footpad thickness)U-Omp16 2 http://www.jimmunol.org/ Forrapidprotein detection,100–200mg frozenplant material was groundand (footpad thickness)saline. 200 ml sample buffer (0.125 M Tris-Cl, pH 6.8; 20% [v/v] glycerin; 4% [v/v] In vivo analysis of DC activation SDS; and 0.02% [w/v] bromophenol blue) was added. Samples were heated at 95˚C for 10 min and subjected to SDS-PAGE (15%; see below). In vivo induction of DC activation was evaluated by measuring the ex- For extraction of total plant protein, 100–200 mg frozen plant material pression of various surface markers by flow cytometry. BALB/c mice were was thoroughly ground, mixed with 200 ml extraction buffer (23 mM injected i.v. with 100 mg U-Omp16, either untreated or digested with Na2HPO4; 17 mM NaH2PO4; and 100 mM NaCl, pH 7.4), and incubated proteinase K; with 50 mg Escherichia coli LPS (Sigma-Aldrich); or with on ice for 20 min. After centrifugation at 4˚C and maximum speed for PBS alone. At 20 h postinjection, mice were sacrificed, and spleens from 20 min, the supernatant was collected and centrifuged for a second time mice were removed and treated for 45 min at 37˚C with 400 U/ml colla- under the same conditions. genase type IV and 50 mg/ml DNase I (Boehringer Mannheim, Indian- Recombinant protein was purified from plant material, using the Protino by guest on September 28, 2021 apolis, IN) in RPMI 1640. After inhibition of collagenase with 6 mM Protein Purification SystemTM (Ni-TED 2000 packed columns; Macherey- EDTA and 0.5% FCS (Invitrogen Life Technologies, Carlsbad, CA), Nagel, Du¨ren, Germany) according to the instructions. Purified protein was a single spleen cell suspension was prepared. Spleen cells were incubated desalted and concentrated, using Amicon Ultra centrifugal filter devices for 20–45 min at 4˚C in 5% mouse serum in the presence of primary Abs: (Millipore, Carrigtwohill, Ireland) according to the supplier’s protocol. FITC-, PE-, CyChrome- or PE-Cy5–conjugated Abs specific for CD8a, Total soluble protein and purified protein concentration were determined CD11c, CD40, CD80, CD86, or isotype control. mAbs were purchased with the BCA Protein Assay Kit (Pierce). from eBioscience (San Diego, CA)and BD Biosciences (Franklin Lakes, Defined amounts of total plant protein (0.1 mg, 0.5 mg, 1 mg, 5 mg, and NJ). After staining, cells were fixed and analyzed by flow cytometry, using 15 mg) and of purified Omp16 (0.05 mg, 0.1 mg, and 0.25 mg) in a volume of a FACSAria II (BD Biosciences). Data were analyzed using FlowJo soft- 20 ml were mixed with 5 ml53 SDS sample buffer (250 mM Tris-Cl, pH 6.8; ware (Tree Star, Ashland, OR). 10% [v/v] SDS; 0.05% [w/v] bromophenol blue; and 50% [v/v] glycerin). Samples were heated at 95˚C for 10 min and subjected to SDS-PAGE (15%). In vitro activation of bone marrow-derived DCs and Electrophoresis was performed in 1 3 SDS running buffer (25 mM Tris, pH macrophages 8.3; 192 mM glycine; and 0.1% [v/v] SDS) with 40 V for 1 h and 100 V for the remaining run time. After electrophoresis, the gels were equilibrated in DCs and macrophages were generated from bone marrow (BM) mono- transfer buffer (25 mM Tris; 192 mM glycine; 20% [v/v] methanol) for 2/2 2/2 2/2 nuclear cells from wt, TLR2 , TLR4 , or TLR6 C57BL/6 mice, in 10 min, and proteins were transferred to a polyvinylidenedifluoride mem- medium containing rGM-CSF, as previously described (35, 36). Briefly, brane (Hybond-P; Amersham Biosciences, Freiburg, Germany), using Pe- femora and tibiae were collected from 4- to 12-wk-old mice. After removal qulab tank-blot equipment (Pequlab, Erlangen, Germany) at 100 V for 1 h. of adjacent muscles, bones were flushed throughout their interior with 5 ml The membrane was blocked in 5% (w/v) dried milk powder in 1 3 PBS-T HBSS to extract marrow cells. The marrow suspension was filtered through (5.6 mM Na2HPO4 3 12 H2O; 2.9 mM K2HPO4; 2.9 mM NaCl; 0.05% [v/v] 70-mm cell strainers (Falcon; BD Biosciences) to remove unwanted Tween 20; and pH 7.2–7.4). Subsequently, the membrane was incubated with fragments. To obtain BM-derived DCs, BM cells were centrifuged and the primary Ab (murine anti-Omp16 mAb) at a dilution 1:2000 in 1 3 PBS-T 7 plated on 100-mm petri dishes (1 3 10 cells/10 ml) in DC culture medium for 1 h at room temperature. After washing three times in 1 3 PBS-T for (complete medium, 50 mM 2-ME, 200 U GM-CSF). On days 2 and 4, 10 min, the membrane was incubated with a HRP-conjugated secondary Ab fresh DC culture medium was added. On day 5, cells were plated on 6-well (goat anti-mouse IgG-HRP; Santa Cruz Biotechnology, Heidelberg, Ger- 6 plates (3 3 10 cells/well) in 3 ml fresh DC culture medium. After many; dilution 1:10000 in 1 3 PBS-T) for 1 h at room temperature. Washing overnight culture, on day 6, cells demonstrated differentiated morphology was repeated as described, and immunoblots were developed using the + and expressed DC markers (CD11c , MHC class II (MHCII)low, and Chemiluminescence Detection Kit (AppliChem, Darmstadt, Germany) ac- CD40low; data not shown). cording to the supplier’s instructions. BM-derived macrophages were obtained as previously described (37). Briefly, BM cell suspension was centrifuged, and cells were resuspended in Sequence analysis DMEM (Life Technologies) containing 10% FCS and 10% L929 cell- conditioned medium as a source of M-CSF. Cells were plated in 24-well The amino acid sequence of U-Omp16 was used to perform searches of plates and incubated at 37˚C in a 5% CO2 atmosphere. Three days after nonredundant protein databases. The BLAST program available at www. seeding, another 0.1 ml L929 cell-conditioned medium was added. On day ncbi.nlm.nih.gov/BLAST/ and also the Phylogeny.fr BLAST platform at 7, the medium was renewed. On day 10 of culture, cells were completely www.phylogeny.fr/version2_cgi/index.cgi were run (using MUSCLE for differentiated into macrophages. multiple alignment and PhyML for phylogeny) (40, 41). The Journal of Immunology 5203

Statistical analysis Table II. Protection against B. abortus 544 in BALB/c mice immunized with L-Omp16 or U-Omp16 without adjuvant Statistical analysis and plotting were performed using GraphPad Prism 4 software (GraphPad, San Diego, CA). Protection results were evaluated a using one-way ANOVA and the Dunnett multiple-comparison posttest. The Log10 CFU of B. abortus Units of n 6 data from cytokine production, DTH response, and ﬂow cytometry were Vaccine ( = 5) Adjuvant 544 at Spleen (mean SD) Protection analyzed using one-way ANOVA with Bonferroni’s posttest. When data L-Omp16 None 4.68 6 0.55b,d 1.59 were not normally distributed, a logarithmic transformation was applied U-Omp16 None 4.59 6 0.42b 1.68 prior to the analysis; then parameters showed normal distribution. A p B. abortus strain 19 None 3.98 6 0.26b 2.29 value ,0.05 was taken as the level of signiﬁcance. PBS None 6.27 6 0.14c 0

a The content of bacteria in spleens is represented as the mean log10 CFU 6 SD Results per group. bSignificantly different from PBS-immunized mice; p , 0.01 estimated by Dun- Omp16 confers protection against B. abortus challenge nett´s test. without the need of external adjuvants or its lipid moiety cSignificantly different from B. abortus strain 19 immunized mice; p , 0.01 estimated by Dunnett´s test. As stated before, it was already described that the lipid moiety of dSignificantly different from B. abortus strain 19 immunized mice; p , 0.05 lipoproteins has adjuvant activity per se (42). To address whether estimated by Dunnett´s test. this was the case for the B. abortus Omp16 lipoprotein, a group of mice was immunized i.p. with the lipoprotein alone, whereas other groups of animals were immunized with PBS (negative control) or of Omp16. Besides, the addition of an external adjuvant, like IFA, did not improve the U-Omp16–elicited protective responses (U- the vaccine strain B. abortus S19 (positive control). Immunized Downloaded from . mice were challenged with virulent B. abortus 544, and the po- Omp16 + IFA: 1.46 versus U-Omp16: 1.33 U of protection; p tential of the L-Omp16 per se to impart protection without ad- 0.05) (Table III). juvants was evaluated on the basis of its ability to eliminate the Altogether, these results suggest that the protein portion of the B. bacterial burden (B. abortus 544) in spleens of immunized mice at abortus Omp16 has an inherent adjuvant activity that allows it to day 30 postchallenge. Immunization with L-Omp16 without ad- induce a protective immune response against an in vivo Brucella challenge. juvants induced a significant level of protection (1.49 U of pro- http://www.jimmunol.org/ , tection, p 0.01) when compared with the control PBS- Immunization with U-Omp16 induces a specific Th1 cellular immunized mice (Table I). This result indicates that L-Omp16 is immune response endowed with an intrinsic adjuvant activity that allows it to confer protection against a challenge with virulent B. abortus, without the Given the intracellular nature of Brucella spp., it is well known that need for external adjuvants. cellular immune responses (in particular, Th1) are crucial for To corroborate the finding that the intrinsic adjuvanticity of conferring protection against this pathogen (43). Thus, we decided L-Omp16 was due to its lipid moiety, another experiment has been to test whether immunization with U-Omp16 without adjuvants is conducted in which mice were immunized with L-Omp16 or with able to induce a specific cellular immune response. Splenocytes

U-Omp16, in both cases without adjuvants. As before, L-Omp16 from U-Omp16–immunized mice were stimulated in vitro with by guest on September 28, 2021 conferred protection (1.59 U of protection, p , 0.01versus PBS), but a known Brucella membrane extract that contains native Omp16 unexpectedly U-Omp16 also induced significant protection levels (HS). After stimulation, splenocytes from U-Omp16–immunized against an experimental B. abortus infection (1.68 U of protection, p mice produced large amounts of IFN-g in comparison with the same , , 0.01 versus PBS). Notably, the protection levels elicited by U- splenocytes incubated with complete medium alone (p 0.01). Omp16 were similar to those elicited by its respective lipidated This effect was specific, because spleen cells from U-Omp16– version (p . 0.05), indicating that the molecular region responsible immunized mice produced higher levels of IFN-g than did cells , for the observed self-adjuvanticity is the protein portion of this li- from PBS-immunized mice in response to the same stimulus (p poprotein. It is worth mentioning that the protection levels elicited 0.01) (Fig. 1A). The splenocytes from both groups of mice did not by U-Omp16 were similar to those elicited by the attenuated vac- produce IL-2, IL-4, IL-5, or IL-10 in response to HS (data not . shown). To evaluate which cells were responsible for the specific cine strain B. abortus S19 (p 0.05) (Table II). + + It is important to bear in mind that the purification yield of IFN-g production, we selectively depleted CD4 T cells or CD8 T from whole splenocytes. When splenocytes from U-Omp16– L-Omp16 is much lower than that obtained when producing + U-Omp16. Moreover, it is necessary to make extra LPS depletion immunized mice were depleted of the CD4 T cell population, the specific IFN-g production in response to HS was abrogated (p , rounds to get the recombinant lipoprotein LPS-free (8). Those as- + pects and its self-adjuvanticity led us to focus on the protein region 0.01 versus not depleted) (Fig. 1B). On the contrary, CD8 T cell population depletion did not affect the specific IFN-g production (Fig. 1B). This result indicates that U-Omp16 immunization

Table I. Protection against B. abortus 544 in BALB/c mice immunized with L-Omp16 without adjuvant Table III. Protection against B. abortus 544 in BALB/c mice immunized with U-Omp16 without adjuvant or in IFA Log10a CFU of B. abortus Units of Vaccine (n = 6) Adjuvant 544 at Spleen (mean 6 SD) Protection Log a CFU of B. abortus Units of b,c 10 L-Omp16 None 4.25 6 0.44 1.49 Vaccine (n = 5) Adjuvant 544 at Spleen (mean 6 SD) Protection B. abortus strain 19 None 3.71 6 0.17b 2.03 b PBS None 5.74 6 0.15c 0 U-Omp16 None 4.37 6 0.14 1.33 U-Omp16 IFA 4.24 6 0.18b 1.46 a The content of bacteria in spleens is represented as the mean log CFU 6 SD per PBS None 5.70 6 0.20 0 group. b a Significantly different from PBS-immunized mice; p , 0.01 estimated by Dun- The content of bacteria in spleens is represented as the mean log10 CFU 6 SD nett´s test. per group. cSignificantly different from B. abortus strain 19 immunized mice; p , 0.01 bSignificantly different from PBS-immunized mice; p , 0.01 estimated by Dun- estimated by Dunnett´s test. nett´s test. 5204 B. abortus U-Omp16 IS A SELF-ADJUVANTING VACCINE

Table IV. Protection against an oral challenge of B. abortus 2308 in BALB/c mice orally immunized with U-Omp16 expressed in E. coli or in leaves of the tobacco species Nicotiana benthamiana

a Log10 CFU of B. abortus 2308 Vaccine Expressed at Spleen Units of (n =5) in Adjuvant (mean 6 SD) Protection U-Omp16 E. coli None 4.61 6 0.17b,c 0.94 U-Omp16 Tobacco None 4.29 6 0.08b,d 1.26 B. abortus – None 3.74 6 0.44b 1.81 RB51 PBS – None 5.55 6 0.05c 0

a The content of bacteria in spleens is represented as the mean log10 CFU 6 SD per group. bSignificantly different from PBS-immunized mice; p , 0.01 estimated by Dun- nett´s test. cSignificantly different from B. abortus RB51 immunized mice; p , 0.01 estimated by Dunnett´s test. dSignificantly different from B. abortus RB51 immunized mice; p , 0.05 estimated by Dunnett´s test. Downloaded from munized mice did not produce IL-2, IL-4, IL-5, or IL-10 in response to HS (data not shown). FIGURE 1. U-Omp16 induces a specific cellular immune response when All in all, these results indicate that the protein region of the administered i.p. without adjuvants. A, Spleen cells of PBS- or U-Omp16– Omp16 lipoprotein, when delivered by the oral route without immunized mice (4 3 106/ml) were stimulated in vitro with an extract of Omps of Brucella (HS) or complete medium alone (–). Cell-free culture adjuvants, has the ability to induce a Th1 immune response in vitro supernatants were collected at 72 h poststimulation for the measurement of as well as in vivo while mediating immunoprotection against an http://www.jimmunol.org/ IFN-g (pg/ml) by ELISA. Results are shown as mean 6 SEM for each oral Brucella challenge. group and are representative of three independent experiments. ppSignificantly different from the same stimulus in PBS-immunized mice (p , 0.01). B, CD4+ or CD8+ T cell IFN-g responses in U-Omp16– immunized mice. Splenocytes from PBS- or U-Omp16–immunized mice were depleted of CD4+ cells (2CD4+) or CD8+ T cells (2CD8+), using mouse CD4 (L3T4) Dynabeads or mouse CD8 (Lyt2) Dynabeads or were not depleted (T). These cells were stimulated, and IFN-g was measured in pp

culture supernatants as described in A. Significantly different from by guest on September 28, 2021 analogous treated cells from PBS-immunized mice (p , 0.01). ##Signif- icantly different from nondepleted cells (T) from the same group stimulated with the same stimulus (p , 0.01). without adjuvants induces specific CD4+ T cells that are the major IFN-g producers in response to HS. Incontrast,immunizationwithU-Omp16without adjuvantsdidnot elicit a specific humoral immune response, inducing nearly un- detectableanti-Omp16immunoglobulintitersinsera(datanotshown). Taken together, these results indicate that the protein region of the Omp16 lipoprotein has the ability to induce a Th1 immune response when delivered i.p. without adjuvants.

Oral delivery of U-Omp16 confers protection against B. abortus challenge and elicits a Th1 immune response without the need of external adjuvants

Oral infection is one of the principal ways in which the disease is FIGURE 2. U-Omp16 induces a specific cellular immune response when acquired, in both humans and animals (23, 28). For that reason, we administered orally without adjuvants. A, To study DTH reactions in mice sought to evaluate the protection afforded when mice were im- orally immunized with U-Omp16 or PBS without adjuvants, mice were munized i.g. with U-Omp16 without adjuvants. As negative con- injected in one footpad with U-Omp16 and in the contralateral footpad trol, a group was immunized similarly with PBS, and as positive with an equal volume of saline. At 24, 48, and 72 h later, the footpad control another group received the already reported oral protective thickness was measured and its increment was calculated by: (footpad 2 vaccine strain B. abortus RB51 (24, 25). Oral immunization with thickness)U-Omp16 (footpad thickness)saline. Significant differences from p , U-Omp16 without adjuvants induced a significant protective re- the same stimulus in PBS-immunized mice are denoted on the graph. p 0.05; ppp , 0.01. B, Production of IFN-g by spleen cells of PBS-immu- sponse against an oral challenge with B. abortus 2308 (Table IV). nized or U-Omp16 orally immunized mice stimulated in vitro with HS or This immunization elicited a specific DTH response at 48 and 72 h complete medium alone (–). Cell-free culture supernatants were collected post–U-Omp16 intradermal injection (Fig. 2A). Furthermore, at 72 h poststimulation, and IFN-g (pg/ml) was measured by ELISA. splenocytes from mice orally immunized with U-Omp16 secreted Results are shown as mean 6 SEM for each group and are representative IFN-g after an in vitro stimulation with HS (p , 0.05 versus PBS- of three independent experiments. pSignificantly different from the same immunized mice) (Fig. 2B). In contrast, splenocytes from all im- stimulus in PBS-immunized mice (p , 0.05). The Journal of Immunology 5205

U-Omp16 activates DCs in vitro and in vivo recombinant protein (U-Omp19) was unable to stimulate DCs to DCs are initiators and modulators of immune responses with the produce cytokines in vitro (data not shown). ability to activate naive T cells. Therefore, we evaluated the ability The latter results indicate that U-Omp16 is able to activate DCs of U-Omp16 to induce the activation of DCs in vitro. BM-derived in vitro, but did not ensure that the same would occur in vivo; DCs of C57BL/6 mice were incubated in vitro with U-Omp16, and therefore, we decided to evaluate the U-Omp16 ability to activate DCs in vivo. To achieve this, BALB/c mice were injected i.v. with the TNF-a production was measured. As controls, BM-derived U-Omp16 (Fig. 3A). As controls, other groups of mice were in- DCs were incubated with U-Omp16 plus PB or with U-Omp16 jected with either U-Omp16 completely digested with proteinase completely digested with proteinase K (Fig. 3A); at the same time, K (Fig. 3A), E. coli LPS (positive control), or PBS alone (negative complete medium and E. coli LPS were used as negative and control). At 20 h postinjection, splenic CD11c+ DCs were ana- positive controls, respectively. After being stimulated with U- lyzed directly ex vivo by flow cytometry for the expression of the , Omp16, DCs produced significant amounts of TNF-a (p 0.001 surface markers CD40, CD80, and CD86. DCs from U-Omp16– versus complete medium-stimulated cells). The ability of inoculated mice showed an increased expression of the cos- U-Omp16 to activate DCs was not modified by the addition of timulatory molecules CD40, CD80, and CD86, compared with the PB in the medium (Fig. 3B), indicating that the result is not an expression on DCs from control mice receiving PBS (Fig. 3C). outcome of LPS contamination. In the same way, when U-Omp16 Similarly, E. coli LPS induced the upregulation of the three cos- digested with proteinase K was used to stimulate DCs, its acti- timulatory molecules on splenic DCs from injected mice (Fig. vating ability was lost (p . 0.05 versus complete medium-stim-

3C). The ability of U-Omp16 to induce DC maturation in vivo was Downloaded from ulated cells) (Fig. 3B), indicating that the measured activity totally abrogated by protein digestion with proteinase K, whereas resides in the protein. This result was also speciﬁc for U-Omp16 the same treatment did not inﬂuence the E. coli LPS effect on because in the same stimulation conditions, another B. abortus these maturation DC markers (Fig. 3C; data not shown).

FIGURE 3. U-Omp16 induces in vitro and in vivo activation of DCs. A, Coomassie blue http://www.jimmunol.org/ stained 12.5% SDS PAGE of U-Omp16 untreated or digested with proteinase K (U- Omp16 + PK) that was used for the in vivo DC activation assay. B, In vitro stimulation of BM- derived DCs. C57BL/6 BM-derived DCs were incubated in vitro with complete medium (–), U-Omp16 (50 mg/ml) (NT U-Omp16), U- Omp16 (50 mg/ml) with PB (15 mg/ml) (Pol B U-Omp16), proteinase K digested U-Omp16 (50 mg/ml) (PK U-Omp16), or E. coli LPS by guest on September 28, 2021 (1 mg/ml) (LPS). At 24 h later, the TNF-a (pg/ ml) concentration was measured by ELISA in supernatants. Results are shown as mean 6 SEM for each stimulation condition. pppSig- nificantly different from the TNF-a amounts in complete medium stimulated cells (p , 0.001). C, In vivo stimulation of splenic DCs. BALB/c mice were injected i.v. with 100 mgof U-Omp16 either untreated or digested with proteinase Kor with E. coli LPS or PBS alone as controls. At 24 h postinjection, splenic CD11c+ DCs were analyzed for their activation status by assessing the surface expression of CD40, CD80, and CD86 molecules by flow cytometry. This experiment was conducted three times with similar results. Histograms display results from one representative experiment. D, CD40 upregulation on the different DC subpopulations. BALB/c mice were injected i.v., as in C, and the surface expression of CD40 was analyzed on total splenic CD11c+ DCs, myeloid CD11c+CD8a+ DCs, or lymphoid CD11c+ CD8a2 DCs. Histograms display results from one representative experiment out of three. E, Mean fluorescence intensities for the CD40 expression on the indicated subpopulations from three different mice per group are shown in the bar graph. Significant differences between U-Omp16– or LPS-injected mice and PBS-injected mice are denoted on the graph: pp , 0.05; ppp , 0.01. Results are representative of two independent experiments. 5206 B. abortus U-Omp16 IS A SELF-ADJUVANTING VACCINE

Splenic conventional CD11c+ DCs can be subdivided pheno- U-Omp16 in vitro. The same pattern of response to U-Omp16 stimuli typically into three different subsets based on surface marker in vitro was elicited in BM-derived DCs and macrophages from expression: CD8a+CD42 DCs (myeloid DCs), CD8a2CD4+ DCs, TLR22/2 or TLR62/2 C57BL/6 mice, indicating that neither TLR2 and CD8a2CD42 DCs; the last two subsets are often referred to nor TLR6 is involved in the activation of both cell types by U-Omp16 as CD8a2 DCs (lymphoid DCs) (44, 45). The effect of U-Omp16 in vitro. In contrast, U-Omp16 was unable to stimulate BM-derived on CD8a+ DC and CD8a2 DC subsets in vivo is shown in Fig. DCs or macrophages from TLR42/2 C57BL/6micetosecretethese 3D. Total splenic DCs of mice treated with U-Omp16 upregulated cytokines, suggesting that the stimulating activity of U-Omp16 in vitro the expression of the costimulatory molecule CD40 (Fig. 3C,3D), would be mediated by this receptor. E. coli LPS and Pam3Cys stimuli this upregulation involved both splenic DC subpopulations: my- were included as controls, and, as expected, they induced the activation eloid DCs (CD11c+CD8a+, Fig. 3D) and lymphoid DCs (CD11c+ of all cell types excluding those derived from TLR42/2 mice stimu- 2 2/2 CD8a , Fig. 3D), the effect being more evident in the myeloid lated with E. coli LPS or TLR2 mice stimulated with Pam3Cys. subpopulation (Fig. 3E). Then we sought to determine if in vivo TLR4 plays a role in the Thus, U-Omp16 self-adjuvanticity could be mediated, at least in generation of the adaptive response elicited by U-Omp16. For that part, by the stimulation of DC maturation in vivo, in particular, purpose, C57BL/6 wt or TLR42/2 mice were vaccinated with U- activation of myeloid DCs. Omp16 without adjuvants, and the elicited cellular and protective responses were evaluated. Spleen cells of immunized mice were TLR-4 is involved in U-Omp16 self-adjuvanticity cultured in vitro in the presence of U-Omp16 or culture medium Signaling through TLRs has been shown to be immunostimulatory, alone. The antigenic stimulus induced the production of signiﬁcant inducing host cells to proliferate, secrete cytokines and chemokines, levels of IFN-g from the splenocytes of U-Omp16–vaccinated Downloaded from and upregulate costimulatory molecules. The latter characteristic mice, compared with the production in spleen cells of the control renders TLR ligands a potent addition to the vaccine adjuvant rep- mice (p , 0.05) (Fig. 6). In agreement with our previous in vitro ertoire by possessing the capability to link the innate and adaptive results, splenocytes from mice lacking TLR4 did not respond to this immunesystems,thusinducingnot onlyaninﬂammatoryresponsebut stimulus in vitro, indicating that U-Omp16 requires TLR4 in vivo also activation of the adaptive arm of the immune system (46). Among for eliciting a cellular Th1 immune response. Similarly, immuni-

the TLRs, TLR2 and TLR4 mediate the response to the most diverse zation with U-Omp16 did not elicit signiﬁcant protection against B. http://www.jimmunol.org/ set of molecular structures, including some bacterial proteins (47), abortus infection in TLR42/2 mice (0.51 U of protection, p . 0.05 and it has been reported that some Omps interact with heterodimers of versus PBS). As expected, U-Omp16 delivery in the absence of TLR2 and TLR6 (48). To determine if TLR2, TLR4, or TLR6 is in- external adjuvants elicited 1.15 U of protection (p , 0.01 versus volved in the immunostimulant activity of U-Omp16, BM-derived PBS) in C57BL/6 wt mice (Table V). DCs or macrophages from TLR22/2, TLR42/2,TLR62/2,orwt Overall, these results indicate that U-Omp16 requires the C57BL/6 mice were stimulated in vitro with different doses of presence of TLR4 for displaying its self-adjuvanting properties, U-Omp16 (1, 5, 10, and 50 mg/ml), complete medium alone (un- activating macrophages and DCs while eliciting a protective Th1 stimulatedcontrol), E. coli LPS, or Pam3Cys. After 24 h of stimulation, immune response against Brucella. the TNF-a and IL-12 p40 concentrations in supernatants were mea- by guest on September 28, 2021 sured. Both BM-derived DCs and macrophages from C57BL/6 wt U-Omp16 is a new bacterial PAMP mice produced signiﬁcant levels of both cytokines after stimulation The ability of U-Omp16 to interact with the innate immune system with U-Omp16 in a dose-dependent fashion (Figs. 4, 5, respectively) through TLR4 suggested that it would be a PAMP. As PAMPs are when compared with the same cells treated with culture media alone generally constituted by conserved structures that are widespread (p , 0.01), indicating that both cell types were stimulated by on different pathogens (49), we decided to investigate whether

FIGURE 4. The production of cytokines induced by U-Omp16 stimulation of DCs in vitro is TLR4 dependent. BM-derived DCs from wt, TLR22/2, TLR42/2, or TLR62/2 C57BL/6 mice were treated in vitro for 24 h with complete medium alone (–) or complete medium containing either U-Omp16 (1, 5,

10, 50 mg/ml), E. coli LPS (1 mg/ml), or Pam3Cys (1 mg/ml). Cell-free supernatants were collected and the concentration of TNF-a (A) or IL-12 p40 (B) was determined by ELISA. Results are represented as mean 6 SEM of triplicate measurements. These data are representative of at least three independent experiments. pp , 0.01, signiﬁcant differences between stimulated and complete medium alone treated cells; #p , 0.01, signiﬁcant differences between the indicated group and cells from wt mice exposed to the same stimulus. The Journal of Immunology 5207

FIGURE 5. The production of cytokines induced by U-Omp16 stimulation of macrophages in vitro is TLR4 dependent. BM-derived macrophages from wt, TLR22/2, TLR42/2, and TLR62/2 C57BL/6 mice were treated in vitro for 24 h with complete medium alone (–) or complete medium containing either U- Omp16 (1, 5, 10, or 50 mg/ml), E. coli

LPS (1 mg/ml), or Pam3Cys (1 mg/ml). Cell-free supernatants were collected, and the concentration of TNF-a (A) or IL-12 p40 (B) was determined by ELISA. Re- sults are represented as mean 6 SEM of triplicate measurements. These data are representative of at least three independent experiments. pp , 0.01, signiﬁcant dif- Downloaded from ferences between stimulated and complete medium alone treated cells; #p , 0.01, signiﬁcant differences between the indicated group and cells from wt mice exposed to the same stimulus. http://www.jimmunol.org/

homologous proteins are present among different organisms. the C-terminal region of many Gram-negative bacterial Omps— BLAST analysis of the protein portion of Omp16 (U-Omp16) re- for example, porin-like integral membrane proteins (such as vealed high homology—near to identity (identities, 100–76%; OmpA), small lipid-anchored proteins (such as pal), and MotB scores, 297–215 bits)—with other Omps of the a Proteobacteria proton channels. In terms of the innate immune system, pub- class, order Rhizobiales, particularly in the Brucellaceae, Rhizo- lished studies have indicated that K. pneumoniae and E. coli biaceae, Aurantimonadaceae, and Phyllobacteriaceae families OmpA induce DC activation (19, 50); therefore, we conducted by guest on September 28, 2021 (identical proteins are present in 27 different organisms among a multiple sequence alignment comparing B. abortus U-Omp16 them) (Supplemental Fig. 1). Moreover, U-Omp16 is highly con- with E. coli OmpA and K. pneumoniae OmpA. Although amino served (identities, 67–49%; scores, 189–142) among the a Proteo- acids are different, U-Omp16 from Brucella has significant ho- bacteria class, homologous proteins being present in 99 organisms mology with K. pneumoniae OmpA (24% identity; score, 45.4 from Caulobacterales, Rhizobiales, Rhodobacterales, Rhodospir- bits) and from E. coli OmpA (25% identity; score, 48.5) (Sup- illales, and Sphingomonadales, as well as in the d ε subdivision, plemental Fig. 3). such as in Helicobacter bizzozeronii (Supplemental Fig. 2). Taken together, these results indicate that Omp16 would con- It is noteworthy that BLASTand pfam sequence analyses stated stitute a highly conserved protein among a Proteobacteria. By that Omp16 belongs to a protein family with a peptidoglycan recognizing this protein, the innate immune system could also binding domain similar to the C-terminal domain of OmpA from sense the presence of these particular types of bacteria. E. coli. This domain has a b/a/b/a-b(2) structure and is found in A plant-based U-Omp16 vaccine confers protection against B. abortus The notion that U-Omp16 evokes a protective and cellular Th1 immune response when given orally without adjuvants prompted us to develop a plant-made vaccine based on U-Omp16. One major prerequisite for vaccine production in plants is the development of a reliable system that allows the expression of the desired Ag. As a first approach, we studied the transient expression of U-Omp16 in N. benthamiana plants. The protein portion of Omp16 was generated by PCR and cloned into the 39-provector module pICH10990 for transient nuclear expression. The resulting vector termed pO16 6122 was mobilized FIGURE 6. The elicited U-Omp16–specific cellular Th1 immune re- into agrobacteria and infiltration performed with the 59-module sponse is TLR4 dependent. Spleen cells of PBS- or U-Omp16–immunized 2 2 pICH15879 and recombinase module pICH10881, as described C57BL/6 wt or TLR4 / mice were stimulated in vitro with complete medium alone (–) or complete medium with U-Omp16. After 72 h, cell- (38). Expression of the recombinant U-Omp16 was monitored by free culture supernatants were collected for the measurement of IFN-g (pg/ SDS-PAGE and immunoblot. As illustrated in Fig. 7A, U-Omp16 ml) by ELISA. Results are shown as mean 6 SEM for each group and are accumulated to significant amounts in infiltrated N. benthamiana representative of three independent experiments. pp , 0.05, significantly leaves. Determination of Omp16 content by immunoblot revealed different from the same stimulus in cells from PBS-immunized mice. that a fraction of 2% of the total soluble protein consists of the 5208 B. abortus U-Omp16 IS A SELF-ADJUVANTING VACCINE

Table V. Protection against B. abortus 2308 in C57BL/6 wt or TLR42/2 mice immunized with U- Omp16 without adjuvants

C57BL/6 wt C57BL/6 TLR42/2

a a Log10 CFU of B. abortus Log10 CFU of B. abortus Vaccine 2308 at Spleen Units of 2308 at Spleen Units of (n = 5) Adjuvant (mean 6 SD) Protection (mean 6 SD) Protection U-Omp16 None 4.60 6 0.34b 1.15 5.64 6 0.31 0.51 PBS None 5.83 6 0.06 0.00 6.15 6 0.54 0.00

a The content of bacteria in spleens is represented as the mean log10 CFU 6 SD per group. bSignificantly different from PBS-immunized mice; p , 0.01 estimated by Dunnett´s test. recombinant bacterial protein. Total expression level was rather To assess the protective capacity of plant-made U-Omp16, moderate for this protein; however, sufficient plant material could groups of mice were orally immunized in parallel either with to- be provided for purification of U-Omp16 for subsequent immu- bacco-expressed purified U-Omp16 or with E. coli-expressed nization studies. Purification of the recombinant protein was purified U-Omp16, and the elicited protective response was achieved owing to the C-terminal poly-histidine extension (63His evaluated. Plant-made U-Omp16 was able to elicit a significant tag) by immobilized metal affinity chromatography. Up to 20 mg level of protection (1.26 U of protection, p , 0.01 versus PBS) of U-Omp16 could be purified from 1 g of fresh leaf material. The against oral B. abortus infection. The elicited protection was Downloaded from U-Omp16 preparation from plants showed a high purity, as ex- similar to that induced by U-Omp16 produced in E. coli (p . 0.05 emplified by SDS-PAGE and subsequent Coomassie staining (Fig. tobacco versus E. coli) (Table V). 7B). All visible protein bands also gave a positive signal on Altogether, these results indicate that plant-made U-Omp16 can Western blot with the appropriate anti-Omp16 Ab (Fig. 7A); be correctly expressed in a plant system and conserve its intrinsic therefore, we conclude that they constitute different manifes- immunogenicity: it activates the innate immune system through its

tations of Omp16. interaction with TLR4, allowing it to induce a protective response http://www.jimmunol.org/ To evaluate whether the plant-made U-Omp16 contained the against an oral Brucella challenge. Therefore, a plant-made vac- immunostimulating properties from its E. coli-made pair, an in vitro cine could be a successful approach to control of brucellosis. BM-derived DC or macrophage stimulation assay was performed. When wt-derived cells were incubated in vitro with plant-made U- Discussion Omp16, both cell types produced large amounts of TNF-a and IL- Effective vaccines have three key components: 1) an Ag against 12 p40 in response to the stimulus (p , 0.001 versus nonstimulated which adaptive immune responses are generated, 2) an immune 2/2 cells), whereas TLR4 cells did not produce any from the mea- stimulus or adjuvant to signal the innate immune system to po- sured cytokines (Fig. 8). This result indicates that, as occurs with E. tentiate the Ag-specific response, and 3) a delivery system to ensure coli U-Omp16, plant-made U-Omp16 stimulates the innate immune that the Ag and adjuvant are delivered together at the right time and by guest on September 28, 2021 system by a TLR4-mediated pathway. location (51). Available brucellosis vaccines are live attenuated strains that have all these key components but also have several disadvantages; thus, new improved vaccines need to be developed. Highly purified Ags offer potential advantages over traditional vaccines, including a high degree of safety and the capacity of eliciting highly specific immune responses, but in general they need to be coadministered with additional immunostimulant substances (adjuvants) because they are poorly immunogenic (12–14). Knowing the inherent stimulatory properties of the lipid moiety of bacterial lipoproteins (32, 52), we hypothesized that B. abortus Omp16 lipoprotein would be able to elicit a protective immune response without the need of external adjuvants. Our results confirm that the lipoprotein Omp16 of B. abortus as a recombinant Ag is able to induce a protective immune response against a challenge with virulent B. abortus when given to mice without external adjuvants, indicating that Omp16 lipoprotein has an intrinsic adjuvanticity. To our surprise, however, U-Omp16 administered without adjuvants elicited similar levels of protection, suggesting that the protein portion of Omp16 has enough intrinsic adjuvant activity. This property is not unique to Omp16, because several bacterial FIGURE 7. Characterization of recombinant U-Omp16 expressed in proteins have intrinsic adjuvant activity that allows them to induce plants. A, To estimate the level of recombinant U-Omp16 in N. ben- specific and effective immune responses without the help of ex- thamiana leaves, crude protein extracts of various amounts of total protein ternal adjuvants, such as the K. pneumoniae OmpA or the Omp were subjected to SDS-PAGE and subsequent Western blot, and the in- complex of N. meningitidis serogroup B. The self-adjuvanting tensity of the signals was compared with a series of Omp16 preparations of defined concentration. B, Recombinant U-Omp16 from leaves was purified properties of these proteins were attributed to their ability to ac- via immobilized metal affinity chromatography; 5 mg of the protein was tivate different cellular types, mainly DCs (19–21, 53). Recently, separated on SDS-PAGE and visualized with Coomassie staining (the ar- other investigators have identified a Borrelia burgdorferi lipo- row points to U-Omp16). Apparent sizes are given on the left side in kDa. protein (BmpA) whose protein portion stimulates the secretion of M, marker lane. proinflammatory cytokines in human synovial cells (54). The Journal of Immunology 5209

FIGURE 8. Plant-made U-Omp16 stimulates DCs and macrophages in vitro in a TLR4-dependent fashion. BM-derived DCs (A, B) and macrophages (C, D) from wt and TLR42/2 C57BL/6 mice were treated in vitro for 24 h with complete medium alone (–) or complete medium containing U-Omp16 (50 mg/ml for DCs or 1, 5, and 10 mg/ml for macrophages) or E. coli LPS (1 mg/ml). Cell-free supernatants were collected, and the concentration of TNF-a (A, C) or IL-12p40 (B, D) was determined by ELISA. Results are represented as mean 6 SEM of triplicate measurements. pp , 0.001, significant differences between stimulated and complete medium alone treated cells. Downloaded from The ability of U-Omp16 to induce a protective immune response live vaccines. They are not infectious, the spread of strains to the without the addition of adjuvants is not a frequently occurring environment is avoided, and they are easier to transport. Currently, characteristic in the brucellosis field. Indeed, other Brucella proteins three recombinant proteins of Brucella have been evaluated as oral have been evaluated for their ability to induce protective responses vaccine candidates: choloylglycine hydrolase (28), Omp16, and when administered without adjuvants, giving unsatisfactory results. Omp19 (8). In all cases, proteins were administered with the None of the Brucella p39, BLS, or L7/L12 proteins elicited a sig- mucosal adjuvant cholera toxin. In this report, U-Omp16 was also http://www.jimmunol.org/ nificant protective response when administered without adjuvants able to induce a protective response without the need of adjuvants (33, 55, 56), whereas immunization with the DnaK protein or by the oral route, indicating that this protein also has mucosal self- a fusion protein of Omp16 and L7/L12 without adjuvants induced adjuvanting properties. This is an outstanding property because moderate protection levels, but lower than those induced by the the mucosal adjuvant repertoire is scant at present. control attenuated vaccine strain B. abortus S19 or the same pro- In contrast, evaluation of the cellular immune responses in orally teins with adjuvants (10, 11). In the case of the fusion protein or i.p. immunized mice indicates that U-Omp16 immunization containing Omp16, the lower protection levels elicited by this induces a specific Th1 response in vitro as well as in vivo when preparation compared with those elicited by U-Omp16 in this re- administered without adjuvants. This cellular Th1-specific report could be due to the lower doses used in immunization or to sponse could be responsible for the elicited protection against by guest on September 28, 2021 differences between the immunization routes used. Brucella infection, because IFN-g plays a central role in pro- The idea that protection elicited by Omp16 without adjuvants tective responses against Brucella (61). was independent of protein acylation represents an advantage in Induction of adaptive specific immune responses involves the vaccine development, particularly in the scale-up for bulk manu- activation of specific T cells by APCs. Among them, DCs are the facture and in the final cost of the product, the production of key mediators of adaptive immunity. The quality of signals re- L-Omp16 being a more time-consuming and expensive process ceived by DCs in response to PAMPs influences the nature of the than the production of U-Omp16 (8). Moreover, the protection evoked adaptive response. Following stimulation with PAMPs, DC levels induced by U-Omp16 in the absence of external adjuvants maturation occurs, a process in which DCs undergo phenotypic were statistically similar to those elicited by the vaccine strain B. changes resulting in an improved ability to promote T cell res- abortus S19 and were not improved by the addition of an adjuvant, ponses. Many immunopotentiators mediate their effect by acti- such as IFA, indicating that U-Omp16 is a worthy vaccine can- vating DCs (62, 63). Our results show that U-Omp16 is able to didate for brucellosis control. stimulate DC activation in vitro and in vivo. This stimulation To our knowledge, Omp16 is the first Brucella protein able to could not be due to LPS contamination in the preparations, be- induce protection levels similar to those induced by the control cause the protein preparations were exhaustively depleted of LPS live vaccine strain without the requirement of external adjuvants. with Sepharose-PB, as assessed by Limulus amoebocyte assay. This unique quality represents an exceptional benefit because Moreover, U-Omp16 lost its DC-activating capacity when com- external adjuvants might sometimes present risk, inducing adverse pletely digested with proteinase K, indicating that the in vivo reactions, such as local inflammation at the injection site with the elicited DC activation was due to the protein rather than to another induction of granuloma or sterile abscess formation (57). nonprotein contaminant. The same treatment did not affect the Given that most pathogens gain entry into their hosts via mucosal stimulating properties of E. coli LPS. surfaces, the induction of a robust mucosal immune response by The DC subpopulation (CD8a+) is able to produce large vaccination may be an effective means of preventing infectious amounts of IL-12 when activated with several immunostimulants, diseases. Oral delivery of vaccines is particularly appealing, as it being involved in the induction of Th1 adaptive immune responses would avoid the use of needles and allow self-administration. (44, 45, 63–65). In the current study, we have demonstrated that Because one of the main pathways of acquiring brucellosis is the the DC stimulating activity of U-Omp16 was more prominent in oral route, several efforts have been reported to develop an oral the CD8a+ DC subpopulation. Therefore, our results suggest that vaccine. Most studies involved the use of attenuated Brucella the intrinsic adjuvant activity of the protein portion of Omp16 strains (24, 58, 59) or live vectors expressing Brucella Ags (26, could be mediated, at least in part, by the in vivo activation of 60) or carrying DNA plasmids for DNA vaccine delivery at the DCs, leading to the priming of specific naive T cells and evoking mucosal gut (27). Subunit vaccines have several advantages over a protective Th1 immune response against Brucella infection. 5210 B. abortus U-Omp16 IS A SELF-ADJUVANTING VACCINE

It is now known that signaling of one or more receptors on together. The latter notion, together with our findings showing that immune cells (for example, TLRs) results in a rapid inflammatory oral U-Omp16 delivery without adjuvants induced protection response, leading to enhanced presentation of Ags to the immune against Brucella, led us to develop a plant-made vaccine ex- system (47, 66, 67). In the current study, we have demonstrated pressing U-Omp16. As a first approach, we studied the transient that the ability of U-Omp16 to stimulate innate immune cells expression of U-Omp16 in N. benthamiana plants. The fact that (DCs and macrophages) as well as to induce an adaptive specific U-Omp16 expressed in N. benthamiana leaves demonstrates immune response in the absence of adjuvants required the pres- a TLR4-mediated capacity to activate in vitro DCs and macro- ence of TLR4. Although U-Omp16 immunization did not induce phages indicates that it retains its immunostimulating activity. The a statistically significant protection against Brucella in TLR42/2 latter result, in conjunction with the demonstration that E. coli- mice, the protection was not totally abrogated, suggesting that made U-Omp16 retains its stimulating activity when it is co- other mechanisms (additive or redundant) would account for the incubated with PB and loses its activity when completely digested elicited protection against Brucella. The best characterized TLR4 with proteinase K, contributes to the interpretation that the mea- agonist is E. coli LPS. However, there are several agonists for this sured immunostimulating activity is not an outcome of contami- receptor that are not related to LPS, including different proteins. nating LPS but rather a specific effect of the U-Omp16 protein. Among the protein TLR4 agonists are endogen molecules, such as Moreover, tobacco-made U-Omp16 induced the same protective the extradomain A of fibronectin (68, 69) and gp96 (70), and efficacy as E. coli-expressed U-Omp16, indicating that the plant- pathogen-derived proteins, such as the peptidyl-propil cys-trans made U-Omp16 retains its self-adjuvanticity and immunogenicity. isomerase from Helicobacter pylori (71), the DnaK protein from These findings suggest that future attempts to obtain an edible

Francisella tularensis (72), Hsp60 from Chlamydia pneumoniae vaccine for cattle in another plant system (for example, alfalfa or Downloaded from (73), and the Brucella lumazine synthase BLS from B. abortus barley) hold much promise. (74). Altogether, our results indicate that U-Omp16 could con- Finally, this report presents U-Omp16 as a new bacterial PAMP stitute a new PAMP recognized by TLR4. that signals through TLR4, is able to activate in vivo DCs, induces A previous report by our laboratory indicates that B. abortus a Th1 immune response, and is a very promising self-adjuvanting L-Omp16 possesses proinﬂammatory activity on human mono- vaccine against systemic as well as orally acquired brucellosis. This

cytes in a TLR2-dependent fashion. This activity was dependent study also set the precedent for development of a plant-made http://www.jimmunol.org/ on protein lipidation (32). In that report, assays were designed to vaccine against brucellosis. imitate the lipoprotein concentration present on the bacterial surface at the site of infection (0.01–1 mg/ml). The current study is different because the protein concentrations used (5–50 mg/ml) Acknowledgments attempted to resemble those present at the site of injection after We thank Icon Genetics (Halle, Germany) for providing vectors for transient plant expression. vaccine administration (150 mg/ml). Therefore, the absence of response by human monocytes following U-Omp16 stimulus reported in our previous paper could be due to the lower concen- Disclosures trations used in such assays. The authors have no financial conflicts of interest. by guest on September 28, 2021 PAMPs are generally well conserved structures among different pathogens (49). The U-Omp16 sequence is highly conserved among a Proteobacteria, and BLAST and pfam sequence analyses References revealed that U-Omp16 belongs to a family of proteins with 1. Boschiroli, M. L., V. Foulongne, and D. O’Callaghan. 2001. Brucellosis: a worldwide zoonosis. Curr. Opin. Microbiol. 4: 58–64. a peptidoglycan binding domain similar to the C-terminal domain 2. Cheers, C. 1984. Pathogenesis and cellular immunity in experimental murine of OmpA. Well-studied members of this family include the E. coli brucellosis. Dev. Biol. Stand. 56: 237–246. 3. Schurig, G. G., N. Sriranganathan, and M. J. Corbel. 2002. Brucellosis vaccines: OmpA, the E. coli lipoprotein PAL, and N. meningitidis RmpM, past, present and future. Vet. Microbiol. 90: 479–496. which interact with the outer membrane, as well as the E. coli 4. Zhan, Y., and C. Cheers. 1993. Endogenous gamma interferon mediates re- motor protein MotB and the Vibrio flagellar motor proteins PomB sistance to Brucella abortus infection. Infect. Immun. 61: 4899–4901. 5. Nicoletti, P. 1990. Vaccination against Brucella. Adv. Biotechnol. Processes 13: and MotY, which interact with the inner membrane. From the 147–168. standpoint of the innate immune system, it has been stated that K. 6. Ashford, D. A., J. di Pietra, J. Lingappa, C. Woods, H. Noll, B. Neville, R. Weyant, pneumoniae OmpA induces DC maturation by binding TLR2 (19). S. L. Bragg, R. A. Spiegel, J. Tappero, and B. A. Perkins. 2004. Adverse events in humans associated with accidental exposure to the livestock brucellosis vaccine In addition, E. coli OmpA induces DC maturation independently RB51. Vaccine 22: 3435–3439. of TLR4 (50). Although U-Omp16 from Brucella has significant 7. Thakur, S. D., R. Kumar, and D. C. Thapliyal. 2002. Human brucellosis: review of an under-diagnosed animal transmitted disease. J. Commun. Dis. 34: 287–301. homology with K. pneumoniae OmpA and E. coli OmpA, there 8. Pasquevich, K. A., S. M. Estein, C. Garcıá Samartino, C. G. Samartino,A.Zwerdling, are some differences in their amino acid sequences that may ex- L. M. Coria, P. Barrionuevo, C. A. Fossati, G. H. Giambartolomei, and J. Cassataro. plain the differences in TLR binding. A similar situation has al- 2009. Immunization with recombinant Brucella species outer membrane protein Omp16 or Omp19 in adjuvant induces specific CD4+ and CD8+ T cells as well as ready been described in the case of flagellin from H. pylori, which systemic and oral protection against Brucella abortus infection. Infect. Immun. 77: no longer recognizes TLR5 as the flagellin from E. coli because of 436–445. changes in its amino acid sequence (75). At present, it is not 9. Cassataro, J., K. A. Pasquevich, S. M. Estein, D. A. Laplagne, A. Zwerdling, S. de la Barrera, R. Bowden, C. A. Fossati, G. H. Giambartolomei, and F. A. Goldbaum. known which domain of K. pneumoniae OmpA binds TLR2, but it 2007. A DNA vaccine coding for the chimera BLSOmp31 induced a better degree is important to have in mind that the homology with B. abortus of protection against B. ovis and a similar degree of protection against B. melitensis U-Omp16 is predominantly at the C-terminal portion. than Rev.1 vaccination. Vaccine 25: 5958–5967. 10. Delpino, M. V., S. M. Estein, C. A. Fossati, P. C. Baldi, and J. Cassataro. 2007. Vaccine production in plants is attractive in terms of safety and Vaccination with Brucella recombinant DnaK and SurA proteins induces pro- cost effectiveness. From a safety perspective, the plant expression tection against Brucella abortus infection in BALB/c mice. Vaccine 25: 6721– 6729. system would produce a vaccine free of animal pathogens and 11. Luo, D., B. Ni, P. Li, W. Shi, S. Zhang, Y.Han, L. Mao, Y.He, Y.Wu, and X. Wang. animal proteins. The production of plant-derived vaccines is, in 2006. Protective immunity elicited by a divalent DNAvaccine encoding both the L7/ principle, almost limitless and may require little or no downstream L12 and Omp16 genes of Brucella abortus in BALB/c mice. Infect. Immun. 74: 2734–2741. processing. An edible vaccine could be very practical to administer 12. Schijns, V. E. 2000. Immunological concepts of vaccine adjuvant activity. Curr. in cattle, but in this situation Ag and adjuvant must be expressed Opin. Immunol. 12: 456–463. The Journal of Immunology 5211

13. Lima, K. M., S. A. dos Santos, J. M. Rodrigues, Jr., and C. L. Silva. 2004. combination of DNA modules delivered by Agrobacterium. Proc. Natl. Acad. Vaccine adjuvant: it makes the difference. Vaccine 22: 2374–2379. Sci. USA 101: 6852–6857. 14. Reed, S. G., S. Bertholet, R. N. Coler, and M. Friede. 2009. New horizons in 39. Chen, H., R. S. Nelson, and J. L. Sherwood. 1994. Enhanced recovery of adjuvants for vaccine development. Trends Immunol. 30: 23–32. transformants of Agrobacterium tumefaciens after freeze-thaw transformation 15. Bessler, W. G., and G. Jung. 1992. Synthetic lipopeptides as novel adjuvants. and drug selection. Biotechniques 16: 664–668, 670. Res. Immunol. 143: 548–553, discussion 579–580. 40. Dereeper, A., V. Guignon, G. Blanc, S. Audic, S. Buffet, F. Chevenet, 16. Babu, M. M., M. L. Priya, A. T. Selvan, M. Madera, J. Gough, L. Aravind, and J. F. Dufayard, S. Guindon, V. Lefort, M. Lescot, et al. 2008. Phylogeny.fr: K. Sankaran. 2006. A database of bacterial lipoproteins (DOLOP) with func- robust phylogenetic analysis for the non-specialist. Nucleic Acids Res. 36(Web tional assignments to predicted lipoproteins. J. Bacteriol. 188: 2761–2773. Server issue): W465–469. 17. Salazar, J. C., C. D. Pope, M. W. Moore, J. Pope, T. G. Kiely, and J. D. Radolf. 41. Altschul, S. F., T. L. Madden, A. A. Scha¨ffer, J. Zhang, Z. Zhang, W. Miller, and 2005. Lipoprotein-dependent and -independent immune responses to spirochetal D. J. Lipman. 1997. Gapped BLAST and PSI-BLAST: a new generation of infection. Clin. Diagn. Lab. Immunol. 12: 949–958. protein database search programs. Nucleic Acids Res. 25: 3389–3402. 18. Vasselon, T., P. A. Detmers, D. Charron, and A. Haziot. 2004. TLR2 recognizes 42. BenMohamed, L., S. L. Wechsler, and A. B. Nesburn. 2002. Lipopeptide vac- a bacterial lipopeptide through direct binding. J. Immunol. 173: 7401–7405. cines—yesterday, today, and tomorrow. Lancet Infect. Dis. 2: 425–431. 19. Jeannin, P.,G. Magistrelli, L. Goetsch, J. F. Haeuw, N. Thieblemont, J. Y.Bonnefoy, 43. Baldwin, C. L., and R. Goenka. 2006. Host immune responses to the intracellular and Y. Delneste. 2002. Outer membrane protein A (OmpA): a new pathogen- bacteria Brucella: does the bacteria instruct the host to facilitate chronic in- associated molecular pattern that interacts with antigen presenting cells-impact on fection? Crit. Rev. Immunol. 26: 407–442. vaccine strategies. Vaccine 20(Suppl 4): A23–A27. 44. Neuenhahn, M., and D. H. Busch. 2007. Unique functions of splenic CD8alpha+ 20. Jeannin, P.,T. Renno, L. Goetsch, I. Miconnet, J. P.Aubry, Y.Delneste, N. Herbault, dendritic cells during infection with intracellular pathogens. Immunol. Lett. 114: T. Baussant, G. Magistrelli, C. Soulas, et al. 2000. OmpA targets dendritic cells, 66–72. induces their maturation and delivers antigen into the MHC class I presentation 45. Lo´pez-Bravo, M., and C. Ardavı´n. 2008. In vivo induction of immune responses pathway. Nat. Immunol. 1: 502–509. to pathogens by conventional dendritic cells. Immunity 29: 343–351. 21. Singleton, T. E., P. Massari, and L. M. Wetzler. 2005. Neisserial porin-induced 46. Pulendran, B., and R. Ahmed. 2006. Translating innate immunity into immu- dendritic cell activation is MyD88 and TLR2 dependent. J. Immunol. 174: 3545– nological memory: implications for vaccine development. Cell 124: 849–863. 47. Akira, S., and K. Takeda. 2004. Toll-like receptor signalling. Nat. Rev. Immunol.

3550. Downloaded from 22. Wu, Y., C. Przysiecki, E. Flanagan, S. N. Bello-Irizarry, R. Ionescu, O. Muratova, 4: 499–511. G. Dobrescu, L. Lambert, D. Keister, Y. Rippeon, et al. 2006. Sustained high-titer 48. Biswas, A., P. Banerjee, G. Mukherjee, and T. Biswas. 2007. Porin of Shigella antibody responses induced by conjugating a malarial vaccine candidate to outer- dysenteriae activates mouse peritoneal macrophage through Toll-like receptors 2 membrane protein complex. Proc. Natl. Acad. Sci. USA 103: 18243–18248. and 6 to induce polarized type I response. Mol. Immunol. 44: 812–820. 23. Nicoetti, P. 1989. Relationship between animal and human disease. In Brucel- 49. Medzhitov, R., and C. A. Janeway, Jr. 1997. Innate immunity: impact on the losis: clinical and laboratory aspects. E. J. Young, and M. J. Corbel, eds. CRC adaptive immune response. Curr. Opin. Immunol. 9: 4–9. Press, Boca Raton, Fla., p. 41. 50. Torres, A. G., Y. Li, C. B. Tutt, L. Xin, T. Eaves-Pyles, and L. Soong. 2006. 24. Pasquali, P., A. Rosanna, C. Pistoia, P. Petrucci, and F. Ciuchini. 2003. Brucella Outer membrane protein A of Escherichia coli O157:H7 stimulates dendritic cell abortus RB51 induces protection in mice orally infected with the virulent strain activation. Infect. Immun. 74: 2676–2685. http://www.jimmunol.org/ B. abortus 2308. Infect. Immun. 71: 2326–2330. 51. Ulmer, J. B., U. Valley, and R. Rappuoli. 2006. Vaccine manufacturing: chal- 25. Stevens, M. G., S. C. Olsen, M. V. Palmer, and G. W. Pugh, Jr. 1996. Immune lenges and solutions. Nat. Biotechnol. 24: 1377–1383. responses and resistance to brucellosis in mice vaccinated orally with Brucella 52. Mu¨ller, M. R., K. H. Wiesmu¨ller, G. Jung, T. Loop, M. Humar, S. D. Pfannes, abortus RB51. Infect. Immun. 64: 4534–4541. W. G. Bessler, and K. Mittenbu¨hler. 2002. Lipopeptide adjuvants: monitoring 26. Pontes, D. S., F. A. Dorella, L. A. Ribeiro, A. Miyoshi, Y. Le Loir, A. Gruss, and comparison of P3CSK4- and LPS-induced gene transcription. Int. Im- munopharmacol. S. C. Oliveira, P. Langella, and V. Azevedo. 2003. Induction of partial protection 2: 1065–1077. 53. Jeannin, P., G. Magistrelli, N. Herbault, L. Goetsch, S. Godefroy, P. Charbonnier, in mice after oral administration of Lactococcus lactis producing Brucella A. Gonzalez, and Y. Delneste. 2003. Outer membrane protein A renders den- abortus L7/L12 antigen. J. Drug Target. 11: 489–493. dritic cells and macrophages responsive to CCL21 and triggers dendritic cell 27. Al-Mariri, A., A. Tibor, P. Lestrate, P. Mertens, X. De Bolle, and J. J. Letesson. migration to secondary lymphoid organs. Eur. J. Immunol. 33: 326–333. 2002. Yersinia enterocolitica as a vehicle for a naked DNA vaccine encoding 54. Yang, X., H. Izadi, A. S. Coleman, P. Wang, Y. Ma, E. Fikrig, J. Anguita, and

Brucella abortus bacterioferritin or P39 antigen. Infect. Immun. 70: 1915–1923. by guest on September 28, 2021 U. Pal. 2008. Borrelia burgdorferi lipoprotein BmpA activates pro-inflammatory 28. Delpino, M. V., M. I. Marchesini, S. M. Estein, D. J. Comerci, J. Cassataro, responses in human synovial cells through a protein moiety. Microbes Infect. 10: C. A. Fossati, and P. C. Baldi. 2007. A bile salt hydrolase of Brucella abortus 1300–1308. contributes to the establishment of a successful infection through the oral route 55. Al-Mariri, A., A. Tibor, P. Mertens, X. De Bolle, P. Michel, J. Godefroid, in mice. Infect. Immun. 75: 299–305. K. Walravens, and J. J. Letesson. 2001. Protection of BALB/c mice against 29. Tibor, A., V. Weynants, P. Denoel, B. Lichtfouse, X. De Bolle, E. Saman, Brucella abortus 544 challenge by vaccination with bacterioferritin or P39 re- J. N. Limet, and J. J. Letesson. 1994. Molecular cloning, nucleotide sequence, combinant proteins with CpG oligodeoxynucleotides as adjuvant. Infect. Immun. and occurrence of a 16.5-kilodalton outer membrane protein of Brucella abortus 69: 4816–4822. with similarity to pal lipoproteins. Infect. Immun. 62: 3633–3639. 56. Mallick, A. I., H. Singha, P. Chaudhuri, A. Nadeem, S. A. Khan, K. A. Dar, and 30. Tibor, A., B. Decelle, and J. J. Letesson. 1999. Outer membrane proteins Omp10, M. Owais. 2007. Liposomised recombinant ribosomal L7/L12 protein protects Omp16, and Omp19 of Brucella spp. are lipoproteins. Infect. Immun. 67: 4960– BALB/c mice against Brucella abortus 544 infection. Vaccine 25: 3692–3704. 4962. 57. Vogel, F. R. 2000. Improving vaccine performance with adjuvants. Clin. Infect. 31. Tibor, A., E. Saman, P. de Wergifosse, A. Cloeckaert, J. N. Limet, and J. J. Letesson. Dis. 30(Suppl 3): S266–S270. 1996. Molecular characterization, occurrence, and immunogenicity in infected 58. Stevens, M. G., S. C. Olsen, G. W. Pugh, Jr., and D. Brees. 1995. Comparison of sheep and cattle of two minor outer membrane proteins of Brucella abortus. Infect. immune responses and resistance to brucellosis in mice vaccinated with Brucella Immun. 64: 100–107. abortus 19 or RB51. Infect. Immun. 63: 264–270. 32. Giambartolomei, G. H., A. Zwerdling, J. Cassataro, L. Bruno, C. A. Fossati, and 59. Izadjoo, M. J., A. K. Bhattacharjee, C. M. Paranavitana, T. L. Hadfield, and M. T. Philipp. 2004. Lipoproteins, not lipopolysaccharide, are the key mediators D. L. Hoover. 2004. Oral vaccination with Brucella melitensis WR201 protects of the proinflammatory response elicited by heat-killed Brucella abortus. mice against intranasal challenge with virulent Brucella melitensis 16M. Infect. J. Immunol. 173: 4635–4642. Immun. 72: 4031–4039. 33. Velikovsky, C. A., F. A. Goldbaum, J. Cassataro, S. Estein, R. A. Bowden, 60. Stabel, T. J., J. E. Mayfield, L. B. Tabatabai, and M. J. Wannemuehler. 1990. L. Bruno, C. A. Fossati, and G. H. Giambartolomei. 2003. Brucella lumazine Oral immunization of mice with attenuated Salmonella typhimurium containing synthase elicits a mixed Th1-Th2 immune response and reduces infection in a recombinant plasmid which codes for production of a 31-kilodalton protein of mice challenged with Brucella abortus 544 independently of the adjuvant for- Brucella abortus. Infect. Immun. 58: 2048–2055. mulation used. Infect. Immun. 71: 5750–5755. 61. Golding, B., D. E. Scott, O. Scharf, L. Y.Huang, M. Zaitseva, C. Lapham, N. Eller, 34. JimeńezdeBaguë´s, M. P., P. H. Elzer, J. M. Blasco, C. M. Marıń, C. Gamazo, and H. Golding. 2001. Immunity and protection against Brucella abortus. Mi- and A. J. Winter. 1994. Protective immunity to Brucella ovis in BALB/c mice crobes Infect. 3: 43–48. following recovery from primary infection or immunization with subcellular 62. Steinman, R. M., and H. Hemmi. 2006. Dendritic cells: translating innate to vaccines. Infect. Immun. 62: 632–638. adaptive immunity. Curr. Top. Microbiol. Immunol. 311: 17–58. 35. Lutz, M. B., N. Kukutsch, A. L. Ogilvie, S. Ro¨ssner, F. Koch, N. Romani, and 63. Pulendran, B. 2004. Modulating vaccine responses with dendritic cells and Toll- G. Schuler. 1999. An advanced culture method for generating large quantities of like receptors. Immunol. Rev. 199: 227–250. highly pure dendritic cells from mouse bone marrow. J. Immunol. Methods 223: 64. Maldonado-Lo´pez, R., T. De Smedt, P. Michel, J. Godfroid, B. Pajak, 77–92. C. Heirman, K. Thielemans, O. Leo, J. Urbain, and M. Moser. 1999. CD8alpha+ 36. Macedo, G. C., D. M. Magnani, N. B. Carvalho, O. Bruna-Romero, R. T. Gazzinelli, and CD8alpha- subclasses of dendritic cells direct the development of distinct and S. C. Oliveira. 2008. Central role of MyD88-dependent dendritic cellmaturation T helper cells in vivo. J. Exp. Med. 189: 587–592. and proinflammatory cytokine production to control Brucella abortus infection. 65. Pulendran, B., J. L. Smith, G. Caspary, K. Brasel, D. Pettit, E. Maraskovsky, and J. Immunol. 180: 1080–1087. C. R. Maliszewski. 1999. Distinct dendritic cell subsets differentially regulate 37. Gomes, M. S., and R. Appelberg. 2002. NRAMP1- or cytokine-induced bacte- the class of immune response in vivo. Proc. Natl. Acad. Sci. USA 96: 1036–1041. riostasis of Mycobacterium avium by mouse macrophages is independent of the 66. Iwasaki, A., and R. Medzhitov. 2004. Toll-like receptor control of the adaptive respiratory burst. Microbiology 148: 3155–3160. immune responses. Nat. Immunol. 5: 987–995. 38. Marillonnet, S., A. Giritch, M. Gils, R. Kandzia, V. Klimyuk, and Y. Gleba. 67. Lahiri, A., P. Das, and D. Chakravortty. 2008. Engagement of TLR signaling as 2004. In planta engineering of viral RNA replicons: efficient assembly by re- adjuvant: towards smarter vaccine and beyond. Vaccine 26: 6777–6783. 5212 B. abortus U-Omp16 IS A SELF-ADJUVANTING VACCINE

68. Okamura, Y., M. Watari, E. S. Jerud, D. W. Young, S. T. Ishizaka, J. Rose, pylori induces apoptosis of gastric epithelial cells in a TLR4- and apoptosis J. C. Chow, and J. F. Strauss, 3rd. 2001. The extra domain A of fibronectin signal-regulating kinase 1-dependent manner. J. Immunol. 174: 5672–5680. activates Toll-like receptor 4. J. Biol. Chem. 276: 10229–10233. 72. Ashtekar, A. R., P. Zhang, J. Katz, C. C. Deivanayagam, P. Rallabhandi, 69. Lasarte, J. J., N. Casares, M. Gorraiz, S. Herva´s-Stubbs, L. Arribillaga, S. N. Vogel, and S. M. Michalek. 2008. TLR4-mediated activation of dendritic C. Mansilla, M. Durantez, D. Llopiz, P. Sarobe, F. Borra´s-Cuesta, et al. 2007. cells by the heat shock protein DnaK from Francisella tularensis. J. Leukoc. The extra domain A from fibronectin targets antigens to TLR4-expressing cells Biol. 84: 1434–1446. and induces cytotoxic T cell responses in vivo. J. Immunol. 178: 748–756. 73. Maguire, M., S. Poole, A. R. Coates, P.Tormay, C. Wheeler-Jones, andB. Henderson. 70. Vabulas, R. M., S. Braedel, N. Hilf, H. Singh-Jasuja, S. Herter, P. Ahmad-Nejad, 2005. Comparative cell signalling activity of ultrapure recombinant chaperonin 60 C. J. Kirschning, C. Da Costa, H. G. Rammensee, H. Wagner, and H. Schild. proteins from prokaryotes and eukaryotes. Immunology 115: 231–238. 2002. The endoplasmic reticulum-resident heat shock protein Gp96 activates 74. Berguer, P. M., J. Mundinãno, I. Piazzon, and F. A. Goldbaum. 2006. A poly- dendritic cells via the Toll-like receptor 2/4 pathway. J. Biol. Chem. 277: 20847– meric bacterial protein activates dendritic cells via TLR4. J. Immunol. 176: 20853. 2366–2372. 71. Basak, C., S. K. Pathak, A. Bhattacharyya, S. Pathak, J. Basu, and M. Kundu. 75. Roy, C. R., and E. S. Mocarski. 2007. Pathogen subversion of cell-intrinsic in- 2005. The secreted peptidyl prolyl cis,trans-isomerase HP0175 of Helicobacter nate immunity. Nat. Immunol. 8: 1179–1187.