Selection of Protective Epitopes for Brucella melitensis by DNA Vaccination

INFECTION AND IMMUNITY, Nov. 2005, p. 7297 7303 Vol. 73, No. 11 0019-9567/05/$08.00 0 doi:10.1128/iai.73.11.7297 7303.2005 Copyright 2005, American Society for Microbiology. All Rights Reserved. Selection of Protective Epitopes for Brucella melitensis by DNA Vaccination Xinghong Yang, 1 Mary Hudson, 2 Nancy Walters, 1 Robert F. Bargatze, 2 and David W. Pascual 1 * Veterinary Molecular Biology, Montana State University, Bozeman, Montana 59717-3610, 1 and LigoCyte Pharmaceuticals, Inc., 2155 Analysis Drive, Bozeman, Montana 59718-6831 2 Received 15 April 2005/Returned for modification 20 May 2005/Accepted 18 July 2005 The Brucella melitensis 16M genome was examined for proteins in excess of 100 amino acids and for immunogenicity-associated genes. One subset of 32 annotated genes or open reading frames was identified, and each of these were cloned into the eukaryotic vector pcdna3.1. Purified recombinant plasmids were used to intramuscularly (i.m.) immunize BALB/c mice. After challenge with B. melitensis 16M strain, two protective antigens were found: the periplasmic protein, bp26, and the chaperone protein, trigger factor (TF). Protective efficacy was confirmed with DNA vaccines for these two B. melitensis proteins and, when combined, protection against wild-type challenge was significantly enhanced. Both proteins were found to be immunogenic since elevated serum immunoglobulin G (IgG) antibodies without a specific IgG subclass bias were induced subsequent to i.m. DNA immunization. Antigen-restimulation assays revealed that bp26 and TF stimulated gamma interferon and only bp26 induced interleukin-4 (IL-4), IL-5, and IL-6 cytokines as measured by cytokine enzyme-linked immunospot assay. These collective results suggest that both bp26 and TF are excellent candidates for use in future vaccination studies against brucellosis. Brucella spp., facultative intracellular pathogens, are the etiological agents of brucellosis, a disease that affects livestock and humans (9). The attenuated strains such as Brucella melitensis Rev1 and B. abortus S19 and RB51 are used to control brucellosis in domesticated animals. However, these are less than ideal because of their limited efficacy and potential to cause disease in humans. Moreover, both B. abortus S19 and B. melitensis Rev1 strains induce antibodies to their lipopolysaccharide (LPS), making it difficult to differentiate vaccinated animals from those naturally infected (3, 17, 20). Recently, Brucella spp. have also been recognized as a bioterror threat by the Centers for Disease Control (16). Therefore, a subunit vaccine that is protective against B. melitensis is desirable. DNA vaccines offer a promising approach because they can stimulate both cellular and humoral immunity (13, 26). Furthermore, DNA vaccines have many advantages over traditional protein-based vaccines, including ease of development, induction of long-lived immunity, and minimal preparation costs. With regard to effectiveness, previous studies have already shown that DNA vaccination with sodc (22), lumazine synthase gene (27), and P39 (2) can elicit partial protection against Brucella challenge. Furthermore, in contrast to live attenuated vaccines, there are no concerns of induced disease, and the DNA vaccines are stable. With the completion of sequencing the Brucella genome, identification of novel protective antigens is feasible. In the present study, we applied a search strategy to screen the B. melitensis 16M genome for potential immunogenic antigens. By cloning these potential antigen candidates into the * Corresponding author. Mailing address: Veterinary Molecular Biology, Montana State University, Bozeman, MT 59717-3610. Phone: (406) 994-6244. Fax: (406) 994-4303. E-mail: dpascual@montana.edu. pcdna3.1 vector and testing their efficacy in BALB/c mice, two protective antigens were identified. MATERIALS AND METHODS Animals. Specific-pathogen-free female BALB/c mice (National Cancer Institute, Frederick Cancer Research Facility, Frederick, MD) or gamma interferondeficient (IFN- / ) on a BALB/c background (29) at 6 to 9 weeks of age were used throughout the present study. All mice were maintained at Montana State University Animal Resource Center under pathogen-free conditions in individually ventilated cages under HEPA-filtered barrier conditions and were fed sterile food and water ad libitum. In conducting research with animals, the investigator(s) adhered to the Guide for the Care and Use of Laboratory Animals, prepared by the Committee on Care and Use of Laboratory Animals of the Institute of Laboratory Animal Resources, National Research Council (NIH publication no. 86-23, revised 1985). For challenge studies with B. melitensis strain 16M, mice were maintained under similar isolation conditions, but in the ABSL-3 facilities. All animal care and procedures were in accordance with institutional policies for animal health and well-being. Bacterial strains and growth conditions. B. melitensis virulent strain 16M was obtained from the National Veterinary Services Laboratory, U.S. Department of Agriculture, Ames, IA. Bacteria were grown under aerobic conditions in potato infusion agar for 72 h or in brucella broth (Difco Laboratories, Detroit, MI) overnight at 37 C and 5% CO 2. For inoculation, the bacterial suspensions were adjusted spectrophotometrically to an optical density at 600 nm corresponding to 10 4 CFU/200 l. All experiments with live brucellae were performed in biosafety level 3 facilities. Escherichia coli strain DH5 (Life Technology, Gaithersburg, MD) was used for producing the necessary plasmid constructs. The E. coli cultures were routinely grown at 37 C in Luria-Bertani (LB) broth or agar supplemented, when required, with 50 g of ampicillin and 10 g of kanamycin per ml. Construction of DNA vaccine candidates. DNA fragments of the 32 target genes from B. melitensis 16M were amplified by a PCR in which the EcoRI and XbaI or BamHI and XhoI sites were integrated into upstream and downstream primers. The Kozak sequence was introduced within the upstream DNA primer (18). These fragments were then cloned under the cytomegalovirus (CMV) promoter in the eukaryotic vector pcdna3.1( ) (Invitrogen Corp., San Diego, CA). The resulting plasmids (using the prefix pcmv_) were cultured in LB broth containing ampicillin (50 g/ml) and kanamycin (10 g/ml). Large-scale plasmid DNA isolation was performed using a Plasmid Giga Kit (QIAGEN, Valencia, 7297

7298 YANG ET AL. INFECT. IMMUN. CA) according to the manufacturer s instructions. Plasmids were finally resuspended in phosphate-buffered saline at a concentration of 1,000 g/ml. DNA concentration and purity were determined by measuring the optical density, and the A 260 /A 280 ratio was typically greater than 1.8. The recombinant plasmid construct was verified both by restriction enzyme digest and by sequencing the complete insert. In conducting work involving the use of recombinant DNA, the investigator(s) adhered to the Guidelines for Research Involving Recombinant DNA Molecules (notice, Federal Register, 5 July 1994, vol. 59, no. 127). Production of recombinant bp26 and TF proteins. To enable the assessment of immunity to bp26 and TF, bp26 and TF gene fragments from pcdna3.1 vector were amplified by PCR. EcoRI and ApaI sites were integrated to the 5 and 3 ends to ensure the fragments were in frame with the C-terminal myc and His tag epitopes of Pichia pastoris vector ppicza (Invitrogen Corp.). These two plasmids were designated pic-bp26 and pic-tf, respectively. The plasmids were digested by PmeI. After precipitation and washing in 70% ethanol, the pellets were dried, resuspended in double-distilled water, and transformed into Pichia pastoris X-33 cells (Invitrogen Corp.). Cells were resuspended in 1 ml of 1 M sorbitol and incubated for 1 h at 30 C. They were then resuspended in 0.5 ml of medium (yeast extract, 10 g/liter; peptone, 20 g/liter; glucose, 20 g/liter; sorbitol, 1 M [final concentration]; agar [optional], 15 g/liter) and incubated for 2 h at 30 C. Transformants were applied to plates containing 100, 500, and 1,000 U of zeocin/ml, respectively. After 3 days, 9 colonies were selected from the 1,000- U/ml zeocin plate. Cells were lysed using glass beads (Sigma-Aldrich Chemical Co., St. Louis, MO), and their expression was examined by Western blot analysis. Proteins were transferred from the sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) on a 12% (wt/vol) polyacrylamide gel to 0.2- mpore-size nitrocellulose membranes (Bio-Rad Laboratories, Hercules, CA). Membranes were probed, first with the rabbit polyclonal His tag antiserum and then with a goat anti-rabbit IgG conjugated to horseradish peroxidase (Southern Biotechnology Associates, Inc. [SBA], Birmingham, AL). Detection of His tag fusion protein bands was achieved upon development with the substrate 4-chloro-1-naphthol chromogen and H 2 O 2 (Sigma-Aldrich). The clones providing the best expression were selected and named picbp26 and pictf, respectively. Large-scale expression and purification were performed as follows: X-33/ picbp26 and X-33/pICTF were grown in four flasks, each containing 500 ml of buffered minimal glycerol medium (100 mm potassium phosphate buffer [ph 6.0], 1.34% yeast nitrogen base, 4 10 5 % biotin, 1% glycerol), in 2-liter shaker flasks for 48 h at 30 C and 200 rpm, with glycerol replenished after 24 h. Biomass was collected through centrifugation. Pellets were resuspended at a 1:2 ratio in 2-liter baffled flasks containing buffered minimal methanol media (100 mm potassium phosphate buffer [ph 6.0], 1.34% yeast nitrogen base, 4 10 5 % biotin, 0.5% methanol) and shaken for 24 h at 30 C 200 rpm. Pellets were collected by centrifugation, and cells were lysed using glass beads (425 to 600 nm) and acid washed (Sigma-Aldrich) in lysis buffer (300 mm NaCl, 50 mm Na 2 H 2 PO 4, and 10 mm imidazole [ph 7.2] containing 0.1% Triton X-100). Lysate was cleared through centrifugation and filtered through a 0.45- m-poresize bottle top filter (Nalge-Nunc International, Rochester, NY). Purification was accomplished by batch/gravity flow over Talon resin (BD BioSciences Clontech, Franklin Lakes, NJ). The 20-ml final bed volume was equilibrated with Talon wash-equilibration buffer (ph 7.2; 300 mm NaCl, 50 mm Na 2 H 2 PO 4,10mM imidazole) and then incubated with cell lysate for 1 h at 4 C with rotation. A 35-ml glass column was packed with lysate-resin suspension using gravity flow and washed with 10 bed volumes of Talon wash-equilibration buffer (ph 7.2). Protein was diluted with Talon elution buffer (ph 7.2; 300 mm NaCl, 50 mm Na 2 H 2 PO 4, 150 mm imidazole), and the pooled fractions containing the protein of interest were stored at 80 C. Protein purity was checked by SDS-PAGE/ Coomassie staining and Western blot analysis with the anti-myc antibody (Invitrogen Corp.). SDS-PAGE and Western blot analysis. SDS-PAGE and immunoblot analysis of purified bp26 and TF proteins were performed using standard procedures. For Western blot analysis, proteins were transferred from the SDS-PAGE (12% [wt/vol] polyacrylamide) gel to 0.2- m-pore-size nitrocellulose membranes (Bio- Rad). Membranes were probed, first with a monoclonal mouse anti-myc antibody (Invitrogen Corp.) and then with a goat anti-mouse IgG conjugated to horseradish peroxidase (SBA). Detection of bp26 and TF antigens was achieved upon development with the substrate 4-chloro-1-naphthol chromogen and H 2 O 2 (Sigma-Aldrich). Immunization and colony counts. To screen the 32 constructs made in pcdna3.1, BALB/c mice were anesthetized with inhaled isoflurane, and 300 g of a single plasmid DNA construct plus 50 g of CpG was injected at the anterior tibial tuberosity to ensure the vaccines were delivered to the muscle, as previously described (10). The sequence of CpG used was as follows: TCCATGAC GTTCCTGACGTT (30). Vaccination was performed at weeks 0, 2, 4, and 6. For negative controls, mice were immunized with an equal amount of empty pcdna3.1 vector plus 50 g of CpG. At 4 weeks after the final immunization, mice were challenged intraperitoneally with 2.4 10 4 CFU B. melitensis 16M in 200 l of PBS buffer. At 4 weeks postchallenge, mice were euthanized, and the spleens were removed for CFU determinations. Spleens were Dounce homogenized in 1 ml of sterile water to completely lyse cells. Serial 10-fold dilutions in duplicate of homogenates in sterile water were grown on potato infusion agar. After incubation for 3 to 5 days at 37 C with 5% CO 2, Brucella colonies were enumerated, and the number of CFU per spleen was calculated from the dilutions. An arbitrary criterion of a reduction in splenic colonization by 0.6 log may identify potential candidates. For the repeat experiments, mice (five animals/group) were intramuscularly (i.m.) immunized with 50 g of CpG plus 300 g of vector, pcmvbp26, pcmvtf, or 150 g of pcmvbp26 and 150 g of pcmvtf; for the third experiment, mice (five animals/group) were i.m. immunized with 50 g of CpG plus 80 g of vector or 40 g of pcmvbp26 and pcmvtf (total 80 g of DNA). ELISA. To determine induced antibodies to bp26 and TF, an enzyme-linked immunosorbent assay (ELISA) was used to measure immune serum IgG, IgG1, IgG2a, and IgG2b levels. Purified bp26 and TF proteins (5 g/ml) were used to coat Maxisorp microtiter plates (Nunc, Roskilde, Denmark) at 100 l/well overnight at 4 C. Plates were washed four times in a wash buffer (Tris-buffered saline [ph 7.4] with 0.05% Tween 20) and blocked with 2% milk in Tris-buffered saline for2hat37 C and then incubated with serial dilutions of the sera from mice for 3 h at room temperature, and washed three times. Horseradish peroxidaseconjugated goat anti-mouse IgG, IgG1, IgG2a, or IgG2b antibodies (SBA) were used for detection. After 90 min of incubation at 37 C and a washing step, the specific reactivity was determined by the addition of an enzyme substrate, ABTS [2,2 azinobis(3-ethylbenzthiazolinesulfonic acid)] diammonium (Moss, Inc., Pasadena, CA) at 100 l/well. The absorbance was measured at 415 nm on a Kinetics Reader model EL312 (Bio-Tek Instruments, Winooski, VT). Endpoint titers were defined as the highest reciprocal of dilution of sample giving an optical density at 415 nm of 0.100 U above negative controls after 1 h of incubation at 25 C. Cytokine ELISPOT. Groups of BALB/c or IFN- / mice (five to ten/group) were euthanized 4 weeks after the last immunization to collect spleens. Splenic lymphocytes were isolated by conventional methods using Dounce homogenization (28), yielding 95% viability using trypan blue exclusion. Total splenic mononuclear cells (5 10 6 /ml) were resuspended in complete medium (CM; RPMI 1640 [Gibco-BRL/Life Technologies, Grand Island, NY], 10% fetal bovine serum [Atlanta Biologicals, Atlanta, GA], 10 mm HEPES buffer, 10 mm nonessential amino acids, 10 mm sodium pyruvate, 100 U of penicillin/ml, 100 g of streptomycin/ml) and restimulated with 20 g of recombinant bp26 or TF/ml in the presence of 10 U of human IL-2 (PeproTech, Inc., Rocky Hill, NJ)/ml for 2 days at 37 C. Cells were washed and resuspended in CM. Stimulated lymphocytes were then evaluated by IFN- -, IL-4-, IL-5-, IL-6-, and IL-10-specific enzyme-linked immunospot (ELISPOT) assays as previously described (28). To control for nonspecific reactivity against bp26 and TF, naive BALB/c lymphocytes were cultured as described above with 20 g of bp26 or TF/ml in the presence of 10 U of human IL-2/ml for 2 days. Statistical analysis. An analysis of variance followed by Tukey s method was used to evaluate differences between variations in splenic colonization, antibody titers, and cytokine production levels. The P value for statistical differences between plasmid vector and vaccines and the performance of these vaccines stimulating B and T lymphocyte responses was discerned at the 95% confidence interval. RESULTS Evaluation of B. melitensis 16M genome to obtain vaccine targets for cloning into eukaryotic expression plasmid. Since the B. melitensis 16M genome sequence was published in 2001 (12), the entire 16M genome was blasted against the National Center for Biotechnology Information database for proteins with a sequence of 100 amino acids and designated immunogenic. If the annotation results matched these selection criteria, the gene coding for this protein was subsequently cloned into the pcdna3.1( ) plasmid as a possible DNA vaccine candidate. A total of 32 genes or open reading frames (ORFs) were

VOL. 73, 2005 SUBUNIT VACCINES FOR B. MELITENSIS 7299 TABLE 1. B. melitensis genes or ORFs found to be potential antigens constructed in pcdna3.1( ) ORF Function and characteristic(s) Accession no. bp26 Diagnostic antigen BMEI0536 exor Exopolysaccharide regulatory protein BMEI1094 nlpd Lipoprotein, membrane protein, 43-kDa antigen BMEI1079 tolc Outer membrane protein, type I secretion system BMEI1029 lipa 17-kDa antigen BMEI1335 omp17 Common antigen BMEI1515 nlpd Outer membrane protein, antigenic lipoprotein BMEI1079 orf1* Integrase/resolvase recombinase, surface antigen BMEI0147 tf Trigger factor/immunophilin BMEI1069 perm Acid membrane antigen A, permease BMEII0806 rfe UDP-N-acetylglucosamine transferase, common antigen BMEII31 oppa1 Oligopeptide ABC uptake transporter, surface antigen BMEII0734 oppa2 Oligopeptide ABC uptake transporter, surface antigen BMEII0735 ompa Transmembrane protein BMEI0786 omp28 plpb, outer membrane lipoprotein BMEII0338 orf* Heat-resistant agglutinin, surface antigen BMEII0376 chve Sugar binding protein, 46-kDa surface antigen BMEII0360 flgc Flagellar basal-body rod, IgA1 protease BMEII1088 oprf Outer membrane porin F BMEII0036 bmpa Basic membrane lipoprotein A BMEII0083 omp10 Outer membrane lipoprotein BMEII0017 cp24 Cytosoluble immunogenic BMEI0826 orf* O-antigen transporter BMEII1003 orf* Outer membrane protein, protective surface antigen BMEI1895 sucb Dihydrolipoamide succinyltransferase BMEI0141 orf* Tetratricopeptide repeat family protein BMEI0328 yaec ABC transporter substrate-binding protein BMEI1954 prec Periplasmic oligopeptide protein BMEI1934 orf* 19-kDa periplasmic protein, pathogen-specific membrane antigen BMEII0885 orf* Similar to immunogenic 14-kDa BA14K, transmembrane protein BMEII0679 orf* ABC transporter, periplasmic binding protein BMEII0702 acvb Virulence protein BMEII0681 112P L7/L12 ribosomal protein BMEI0748 a, Unnamed ORF. Downloaded from http://iai.asm.org/ termed as being antigenic or immunogenic as indicated by the BLAST search. These genes were amplified by PCR and cloned into pcdna3.1 plasmid. The ribosomal gene L7/L12, previously known to be a protective antigen for Brucella abortus (19), was also inserted into pcdna3.1( ) to use as a positive control (Table 1). The empty pcdna3.1 vector was used as a negative control. Evaluation of protective efficacy by 32 DNA vaccines. Individual BALB/c mice were vaccinated i.m. four times with a single DNA construct, and two mice were immunized with vector only as negative controls. After 4 weeks, mice were challenged with 2.4 10 4 CFU of B. melitensis 16M, and 4 weeks later, splenic CFU levels were enumerated (Table 2). Compared to negative controls, TF provided the greatest protection, with no detectable ( 100 CFU) bacteria isolated from the spleen. Bp26 also showed some efficacy as evidenced by a 1-log reduction in colonization. Other genes of potential interest were the genes for omp17, nlpd, integrase, dihydrolipoamide succinyltransferase, prec, and 19-kDa periplasmic protein since these showed greater reduction in colonization than the remaining 24 vaccine candidates. Confirmation of protective efficacy of bp26 and TF. To validate the protection data from Table 2, subsequent studies were performed. BALB/c mice were immunized i.m. with either bp26, TF, or both DNA vaccines as previously described. It was evident that the TF and bp26 DNA vaccines were each significantly protective (P 0.011 and P 0.036) against B. melitensis 16M challenge (Fig. 1A), whereas coimmunization with both vaccines afforded greater protection (P 0.016) by 20-fold reduction in splenic colonization. To test whether a lesser dose of the combined DNA vaccines would be efficacious, a 40- g dose of each vaccine was used according to the dosing regimen described above; the vaccines by themselves were not tested. Again, a reduction in splenic colonization by B. melitensis 16M was observed (P 0.029) showing that the combined vaccines afforded the best protection (Fig. 1B). To ascertain potential efficacy of the remaining vaccine candidates, one group of mice was dosed with the remaining six vaccines plus bp26; this combination failed to show any efficacy so these vaccines were not pursued any further (data not shown). Immunization with bp26 and TF DNA vaccines induces elevated IgG to these B. melitensis proteins. In order to investigate host responses to bp26 and TF subsequent to DNA vaccination, bp26 and TF cdnas were cloned into a yeast (Pichia pastoris) expression system, and recombinant proteins were analyzed by SDS-PAGE (Fig. 2A and B, lane 1) and by Western immunoblotting (Fig. 2A and B, lane 2). Both the SDS- PAGE and the Western blot results demonstrated that the purified recombinant proteins had the expected molecular sizes: bp26 at 29.0 kda (26.6 kda plus C-terminal myc tag 2.5 on June 13, 2018 by guest

7300 YANG ET AL. INFECT. IMMUN. TABLE 2. Spleen CFU of the 35 BALB/c mice challenged with B. melitensis 16M Plasmid a CFU/spleen (log 10 ) Protection ( CFU) bp26 3.08 1.16 exor 5.95 1.71 nlpd 4.48 0.24 tolc 3.89 0.35 lipa 5.08 0.84 omp17 3.56 0.68 nlpd 3.56 0.68 orf1* 3.57 0.67 tf 2.0 2.24 perm 3.73 0.51 rfe 5.42 1.18 oppa1 4.00 0.24 oppa2 3.92 0.32 ompa 4.80 0.56 omp28 4.68 0.44 orf* 3.62 0.62 L7/L12 4.86 0.62 chve 3.86 0.38 flgc 3.73 0.51 oprf 3.62 0.62 bmpa 3.68 0.56 omp10 3.78 0.46 cp24 3.89 0.35 orf* 3.78 0.46 orf* 4.40 0.16 sucb 3.56 0.68 orf* 4.58 0.34 yaec 4.73 0.49 prec 3.56 0.68 orf* 3.56 0.68 orf* 4.12 0.12 orf* 4.16 0.08 acvb 4.57 0.33 3.1( ) 4.16/4.31 a Each plasmid contains the ORF indicated in Table 1. kda) and TF at 57.0 kda (54.5 kda plus C-terminal myc tag 2.5 kda). To determine the type of antibody responses elicited by these two protective antigens, BALB/c mice were i.m. immunized with pcmvbp26 and pcmvtf with CpG, and the control group was immunized with pcdna3.1. plus CpG. Vaccination was performed on days 0, 7, and 14. At 6, 8, and 10 weeks after the final immunization, serum samples were obtained to assess antigen-specific endpoint titers (Fig. 3). The results showed the pcmvbp26-vaccinated mice elicited a peak IgG titer of 2 16 ; mice vaccinated with pcmvtf showed a lower IgG titer of 2 13.8, which peaked early at week 6. No detectable titers were observed for the vector immunized mice. Evaluation of anti-tf and anti-bp26 IgG subclass endpoint titers showed no significant differences between IgG1, IgG2a, and IgG2b subclass responses (Fig. 3C and D). Immunization with bp26 and TF DNA vaccines induces a mixed T helper (Th) cell response. Since the IgG subclass profile is a reflection of the types of Th cells simulated, our results suggested that i.m. DNA vaccination induced a mixed Th1 and Th2 cell immunity. To confirm what the supportive Th cells induced, a cytokine-specific ELISPOT assay was conducted. BALB/c or IFN- -deficient (IFN- / ) mice (on a BALB/c background) were i.m. immunized with either FIG. 1. BALB/c mice vaccinated with bp26 or TF DNA vaccines showed reduced splenic colonization following challenge with 2.4 10 4 CFU of B. melitensis 16M. (A) Groups of mice (five/group) were immunized on days 0, 14, 28, and 42 with 300 g of pcdna3.1 vector, pcmvtf, or pcmvbp26 or with 150 g of pcmvtf and pcmvbp26, and each group received a 50- g CpG/immunization. At 4 weeks after the last immunization, mice were challenged intraperitoneally with B. melitensis 16M. Compared to the group immunized with pcdna3.1 vector, the pcmvtf (P 0.011) and pcmvbp26 (P 0.036) immunized groups had significantly less splenic colonization, and the combination of pcmvtf and pcmvbp26 (P 0.016) had the least colonization. (B) To ascertain whether a lesser dose of pcmvtf and pcmvbp26 would still be efficacious (five mice/group), 40 g of each vaccine was given i.m. on days 0, 14, 28, and 42 and compared to mice (five/group) dosed with 80 g of pcdna3.1 vector. As above, each group received 50 g of CpG/immunization. Compared to the vector-immunized group, the combination of pcmvtf and pcmvbp26 (P 0.029) showed reduced colonization. These results show that a subunit vaccine for B. melitensis is possible. pcmvbp26 or pcmvtf as described above and, 4 weeks after the last immunization, spleens from individual mice were harvested. Whole splenic cells were cultured with either bp26 or TF for 2 days and then evaluated for IFN-, IL-4, IL-5, IL-6, and IL-10 secretion by the ELISPOT method (Fig. 4). In BALB/c mice, both bp26 and TF were supported by IFN- (Fig. 4) compared to unstimulated cultures (P 0.001 and P FIG. 2. bp26 and TF expression analysis. picbp26 and pictf transformed Pichia pastoris in buffered minimal glycerol media were induced by methanol (0.5%) for 24 h to express the recombinant proteins. (A and B) Expression of bp26 (A) and TF (B) was verified by SDS-PAGE (lane 1) and Western blotting with an anti-myc monoclonal antibody (lane 2) showing that bp26 had the expected molecular mass of 29 kda (26.6 kda plus myc tag [2.5 kda]) and TF had the expected molecular mass of 57 kda (54.5 kda plus myc tag [2.5 kda]).

VOL. 73, 2005 SUBUNIT VACCINES FOR B. MELITENSIS 7301 DISCUSSION FIG. 3. Immunization with DNA vaccines for bp26 and TF elicits modest serum IgG antibody titers. BALB/c mice (five/group) were i.m. immunized with both pcmvbp26 and pcmvtf plus CpG on days 0, 7, and 14, and serum IgG anti-bp26 and TF titers were measured by standard ELISA methods on weeks 6, 8, and 10. (A) The IgG antibp26 titers peaked between weeks 6 and 8 after primary immunization, and (B) IgG anti-tf titers peaked at week 6 after primary immunization, but with less intensity. Control mice immunized with the control vector, pcdna3.1 plus CpG, failed to elicit antibodies to bp26 or TF. IgG1, IgG2a, and IgG2b anti-bp26 (C) and anti-tf IgG (D) subclass responses were not significantly different. IgG subclass responses were determined using serum samples from week 6. 0.009, respectively). Although TF did not stimulate any other Th2-type cytokines, bp26 produced significant increases in IL-4 (P 0.020), IL-5 (P 0.034), and IL-6 (P 0.001), but not IL-10 (Fig. 4). Since IFN- / mice were unable to produce IFN-, only increases in the measured Th2-type cytokines were observed. Significant increases were observed in IL-4 (P 0.001 and P 0.002), IL-5 (P 0.001), and IL-6 (P 0.001) after restimulation with bp26 or TF, respectively, compared to unstimulated cultures (Fig. 4). A slight, but significant increase (P 0.021) in IL-10 was obtained upon restimulation with only TF. The recombinant bp26 and TF proteins failed to stimulate naive lymphocytes (Fig. 4). Differences in the magnitude of the cytokine responses were also observed between IFN- / and IFN- / mice (Fig. 4). Aside from the differences in IFN- in response to three proteins (P 0.001), IL-4 was enhanced in IFN- / mice against bp26 (P 0.003) and TF (P 0.001); IL-5, bp26 (P 0.002) and TF (P 0.021); and IL-6, TF (P 0.015). Thus, these studies confirm that a mixed Th-type response was obtained to bp26 and TF. The B. melitensis vaccine, Rev1, is currently the best vaccine for caprine and ovine brucellosis (5, 14), yet it bears the caveat of inducing anti-lps responses, making it difficult to distinguish vaccinated from infected animals (17). Thus, the development of a subunit vaccine without the B. melitensis LPS would have significant benefits (5, 15). A subunit vaccine would also eliminate potential infection of humans who administer this live vaccine. To enable this effort, significant attempts have been made to discover whether a single protein can confer protection (2, 4, 21, 22, 27). To date, no such protein has been described. Not to disallow the significance of this prior work, we hypothesized that it may take more than a single B. melitensis protein to induce protection. Bearing such a possibility, the B. melitensis genome was scanned for potential vaccine candidates, focusing on mostly periplasmic proteins since these may be responsible for survival in the host. As such, we identified 32 candidates and conducted preliminary immunization studies using DNA vaccines as a rapid approach to circumvent concerns associated with conventional recombinant protein expression systems. Moreover, since we hypothesized that multiple B. melitensis proteins will be required for protection, we then set the arbitrary criterion that a reduction in splenic colonization by 0.6 log may identify potential candidates. We also hypothesized that, in the presence of an appropriate adjuvant, this level of colonization could be significantly reduced. As a result, the list of 32 candidates was reduced to six. When retested, using a larger number of mice, two DNA vaccine candidates showed reproducible protection. These vaccines encoded two proteins, bp26 and TF, and were each found to reduce splenic colonization by 7-fold, and this level of protection was significantly enhanced to 20-fold when both DNA vaccines were coadministered. Furthermore, while the use of a lesser dose of the combined vaccines (40 g of each) did not achieve the level of efficacy with the 150- g dose, a significant reduction by 6.5-fold in splenic colonization was obtained. Thus, these studies demonstrated the dose dependency of two potential vaccine candidates for B. melitensis and was successfully accomplished without using a live vaccine. The success of the other vaccine candidates showing 0.6-log reduction in splenic colonization may not be viable candidates or may require greater doses of vaccine to show their potential. Studies are currently investigating this possibility. Alternatively, these may perform better in combination with a molecular adjuvant. Studies are currently pursuing the adaptation of the described vaccine candidates coexpressing such molecular adjuvants. It was previously shown that the Brucella periplasmic protein, bp26, could be used to distinguish vaccinated livestock from naturally infected animals (11, 23, 24). bp26 is highly conserved among B. abortus, B. suis, B. ovis, and B. melitensis (25). Thus, work with this protein has mostly focused on its diagnostic properties in infected livestock and humans (8, 11, 23, 24), but its protective efficacy has yet to be evaluated. In the same vein, to further its diagnostic potential, the bp26 gene of B. abortus S19 vaccine strain was deleted and showed no improved efficacy against wild-type B. abortus challenge in mice (6), suggesting the relevance of other proteins being responsible for stimulating protection against B. abortus. To develop a

7302 YANG ET AL. INFECT. IMMUN. FIG. 4. Serum antibody responses were supported by mixed Th1 and Th2 cells for bp26 and Th1 cells for TF. Whole splenic cells isolated from BALB/c or IFN- / mice immunized with both bp26 and TF were cultured with or without 20 g of recombinant bp26 or TF for 2 days, and then cells were analyzed for production of IFN-, IL-4, IL-5, IL-6, and IL-10 by the ELISPOT method. In IFN- / mice, IFN-, IL-4, IL-5, and IL-6 were induced against bp26, whereas only IFN- was induced against TF. In the absence of IFN-, the Th2 cytokines compensated, as evidenced by increased production of IL-4, IL-5, and IL-6 upon restimulation with bp26 or TF, and a slight induction of IL-10 was observed against TF only. As a control, naive BALB/c lymphocytes remained unstimulated when cocultured with recombinant bp26 or TF (depicted as dashed line). Thus, IFN- supports antibody responses to both bp26 and TF. The data depict results from five individual mice the standard error of the mean, and significant differences were determined between bp26- or TF-restimulated cultures and unstimulated cells:, P 0.001;, P 0.009;, P 0.05. better test to distinguish vaccinated from naturally infected sheep, a double mutant of B. melitensis Rev1 vaccine was recently developed (7). It was shown that this bp26 omp31 B. melitensis Rev1 vaccine had the same vaccine efficacy in mice as the parental Rev1 vaccine. Our study is the first to evaluate bp26 as a vaccine candidate as opposed to past studies that mostly concentrated on bp26 as a diagnostic determination of infection. As evidenced from our studies, bp26 does have vaccine potential when combined with TF and possibly other, yet to be defined, vaccine candidates. Our evidence also suggests that bp26 may contribute to protective efficacy against other Brucella serovars. Based upon sequence homology in E. coli, we identified the Brucella homology for TF. Regarding the function of TF, studies have focused mostly on the E. coli gene. In the E. coli cytosol, a fraction of newly synthesized proteins requires the activity of molecular chaperones for folding to the native state. The major chaperones implicated in this folding process are the ribosome-associated TF and DnaK and GroEL chaperones, with their respective cochaperones (1). TF is an ATPindependent chaperone and displays chaperone and peptidylprolyl-cis-trans-isomerase activities in vitro. The positioning of TF at the peptide exit channel, together with its ability to interact with nascent chains as short as 57 residues, renders TF a prime candidate for being the first chaperone that binds to the nascent polypeptide chains. In the present study, TF was revealed for the first time as a protective antigen. As with bp26, the potential protective efficacy in other Brucella spp. species is currently being evaluated. Interestingly, no preferential bias appeared to exist by the DNA vaccines toward a Th1- or Th2-type response. It was evident here that both IgG1 and IgG2a subclass responses were induced in response to both bp26 and TF. The anti-tf antibody responses were supported by elevated numbers of IFN- -producing cells but not significantly by elevated levels of IL-4, IL-5, IL-6, or IL-10. However, this does not preclude other Th2-type cytokines, possibly IL-13, which we are currently investigating. Obviously, in the absence of IFN-, anti-tf antibodies were induced and supported by IL-4, IL-5, IL-6, and possibly other cytokines. In contrast, bp26 induced a mixed Th cell response in which IFN-, IL-4, IL-5, and IL-6 were induced. It will be interesting to learn whether protection conferred by these Brucella proteins is dependent upon cellmediated immunity, humoral immunity, or both. Studies are currently pursuing these avenues of research. From the present study, we learned that the combination of bp26 and TF showed a 20-fold reduction in B. melitensis colonization of BALB/c spleens, resulting in the first evidence of

VOL. 73, 2005 SUBUNIT VACCINES FOR B. MELITENSIS 7303 a DNA vaccine capable of rendering this magnitude of protection in the mouse model for B. melitensis. Although we do not suggest that these two vaccine candidates confer complete protection in livestock, their combination with other viable vaccine candidates could potentially enhance the observed protection. Although such additional candidates remain to be discovered, this evidence clearly suggests the potential for developing a subunit vaccine that can circumvent live brucellae usage and thereby minimize infection of livestock and livestock handlers. ACKNOWLEDGMENTS This study was supported by a U.S. Army Medical Research and Material Command under contract DAMD17-01-C-0040 (LigoCyte Pharmaceuticals) and in part by Montana Agricultural Station, USDA Formula Funds, and NIH/NCRR COBRE-supported (P20 RR-020185) BSL-3 facilities. The views, opinions, and/or findings contained in this report are those of the authors and should not be construed as an official Department of the Army position, policy, or decision unless so designated by other documentation. We thank Gary A. Splitter, Department of Animal Health and Biomedical Sciences, University of Wisconsin, for providing us with the cdna for the Brucella L7/L12 ribosomal protein as a fusion construct. We thank Massimo Maddaloni for his advice on constructing and expression of bp26 and TF proteins in Pichia pastoris and Deanne Stookey and Amy Robison for technical assistance. We also thank Nancy Kommers for assistance in preparing the manuscript. REFERENCES 1. Agashe, V. R., S. Guha, H. C. Chang, P. Genevaux, M. Hayer-Hartl, M. Stemp, C. Georgopoulos, F. U. Hartl, and J. M. Aarral. 2004. Function of trigger factor and DnaK in multidomain protein folding: increase in yield at the expense of folding speed. Cell 117:199 209. 2. Al-Mariri, A., A. Tibor, P. Mertens, X. De Bolle, P. Michel, J. Godfroid, K. Walravens, and J. J. Letesson. 2001. Induction of immune response in BALB/c mice with a DNA vaccine encoding bacterioferritin or P39 of Brucella spp. Infect. Immun. 69:6264 6270. 3. Baldi, P. C., G. H. Giambartolomei, F. A. Goldbaum, L. F. Abd, C. A. Velikovsky, R. Kittelberger, and C. A. Fossati. 1996. Humoral immune response against lipopolysaccharide and cytoplasmic proteins of Brucella abortus in cattle vaccinated with B. abortus S19 or experimentally infected with Yersinia enterocolitica serotype O:9. Clin. Diagn. Lab. Immunol. 3:472 476. 4. Baloglu, S., T. E. Toth, G. G. Schurig, N. Sriranganathan, and S. M. Boyle, S.M. 2000. Humoral immune response of BALB/c mice to a vaccinia virus recombinant expressing Brucella abortus GroEL does not correlate with protection against a B. abortus challenge. Vet. Microbiol. 76:193 199. 5. Blasco, J. M. 1997. A review of the use of B. melitensis Rev. 1 vaccine in adult sheep and goats. Prev. Vet. Med. 31:275 283. 6. Boschiroli, M. L., S. L. Cravero, A. I. Arese, E. Campos, and O. L. Rossetti. 1997. Protection against infection in mice vaccinated with Brucella abortus mutant. Infect. Immun. 65:798 800. 7. Cloeckaert, A., I. Jacques, M. J. Grilló, C. M. Marín, M. Grayon, J. M. Blasco, and J. M. Verger. 2004. Development and evaluation as vaccines in mice of Brucella melitensis Rev. 1 single and double deletion mutants of the bp26 and omp31 genes coding for antigens of diagnostic significance in ovine brucellosis. Vaccine 22:2827 2835. 8. Cloeckaert, A., S. Baucheron, N. Vizcaino, and M. S. Zygmunt. 2001. Use of recombinant BP26 protein in serological diagnosis of Brucella melitensis infection in sheep. Clin. Diagn. Lab. Immunol. 8:772 775. 9. Corbel, M. J. 1997. Brucellosis: an overview. Emerg. Infect. Dis. 3:213 221. 10. Davis, D. L. 2000. Intramuscular and intradermal injection of DNA vaccines in mice and primates, p. 71 78. In D. B. Lowrie and R. C. Whalen (ed.), DNA vaccines: methods and protocols. Humana Press, Totowa, N.J. 11. Debbarh, H. S., M. S. Zygmunt, G. Dubray, and A. Cloeckaert. 1996. Competitive enzyme-linked immunosorbent assay using monoclonal antibodies to the Brucella melitensis BP26 protein to evaluate antibody responses in infected and B. melitensis Rev1-vaccinated sheep. Vet. Microbiol. 53:325 337. 12. DelVecchio, V. G., V. Kapatral, R. J. Redkar, G. Patra, C. Mujer, T. Los, N. Ivanova, I. Anderson, A. Bhattacharyya, A. Lykidis, G. Reznik, L. Jablonski, N. Larsen, M. D Souza, A. Bernal, M. Mazur, E. Goltsman, E. Selkov, P. H. Elzer, S. Hagius, D. O Callaghan, J. J. Letesson, R. Haselkorn, N. Kyrpides, and R. Overbeek. 2002. The genome sequence of the facultative intracellular pathogen Brucella melitensis. Proc. Natl. Acad. Sci. USA 99:443 448. 13. Donnelly, J. J., J. B. Ulmer, J. W. Shiver, and M. A. Liu. 1997. DNA vaccines. Annu. Rev. Immunol. 15:617 648. 14. Garin-Bastuji, B., J. M. Blasco, M., Grayon, and J. M. Verger. 1998. Brucella melitensis infection in sheep: present and future. Vet. Res. 29:255 274. 15. Golding, B., D. E. Scott, O. Scharf, L. Huang, M. Zaitseva, C. Lapham, N. Eller, and H. Golding. 2001. Immunity and protection against Brucella abortus. Microbes Infect. 3:43 48. 16. Izadjoo. M. J., A. K. Bhattacharjee, C. M. Paranavitana, T. L. Hadfield, and D. L. Hoover. 2004. Oral vaccination with Brucella melitensis WR201 protects mice against intranasal challenge with virulent Brucella melitensis 16M. Infect. Immun. 72:4031 4039. 17. Jiménez de Bagüés, M. P., C. M. Marin, J. M. Blasco, I. Moriyon, and C. Gamazo. 1992. An ELISA with Brucella lipopolysaccharide antigen for the diagnosis of B. melitensis infection in sheep and for the evaluation of serological responses following subcutaneous or conjunctival B. melitensis strain Rev. 1 vaccination. Vet. Microbiol. 30:233 241. 18. Kozak, M. 1986. Point mutations define a sequence flanking the AUG initiator codon that modulates translation by eukaryotic ribosomes. Cell 44:283 292. 19. Oliveira, S. C., and G. A. Splitter. 1996. Immunization of mice with recombinant L7/L12 ribosomal protein confers protection against Brucella abortus infection. Vaccine 14:959 962. 20. Olsen, S. C., D. Evans, S. G. Hennager, N. F. Cheville, and M. G. Stevens. 1996. Serologic responses of Brucella abortus strain 19 calfhood-vaccinated cattle following adult vaccination with strain RB51. J. Vet. Diagn. Investig. 8:451 454. 21. Oñate, A. A., R. Vemulapalli, E. Andrews, G. G. Schurig, S. Boyle, and H. Folch. 1999. Vaccination with live Escherichia coli expressing Brucella abortus Cu/Zn superoxide dismutase protects mice against virulent B. abortus. Infect. Immun. 67:986 988. 22. Oñate, A. A., S. Cespedes, A. Cabrera, R. Rivers, A. Gonzalez, C. Munoz, H. Folch, and E. Andrews. 2003. A DNA vaccine encoding Cu, Zn superoxide dismutase of Brucella abortus induces protective immunity in BALB/c mice. Infect. Immun. 71:4857 4861. 23. Rossetti, O..I., A. I. Arese, M. L. Boschiroli, and S. L. Cravero. 1996. Cloning of Brucella abortus gene and characterization of expressed 26-kilodalton periplasmic protein: potential use for diagnosis. J. Clin. Microbiol. 34:165 169. 24. Salih-Alj, D. H., A. Cloeckaert, G. Bezard, G. Dubray. And M. S. Zygmunt. 1996. Enzyme-linked immunosorbent assay with partially purified cytosoluble 28-kilodalton protein for serological differentiation between Brucella melitensis-infected and B. melitensis Rev. 1-vaccinated sheep. Clin. Diagn. Lab. Immunol. 3:305 308. 25. Seco-Mediavilla, P., J. M. Verger, M. Grayon, A. Cloeckaert, C. M. Marin, M. S. Zygmunt, L. Fernandez-Lago, and N. Vizcaino. 2003. Epitope mapping of the Brucella melitensis bp26 immunogenic protein: usefulness for diagnosis of sheep brucellosis. Clin. Diagn. Lab. Immunol. 10:647 651. 26. Srivastava, I. K., and M. A. Liu. 2003. Gene vaccines. Ann. Intern. Med. 138:550 559. 27. Velikovsky, C. A., J. Cassataro, G. H. Giambartolomei, F. A. Goldbaum, S. Estein, R. A. Bowden, L. Bruno, C. A. Fossati, and M. Spitz. 2002. A DNA vaccine encoding lumazine synthase from Brucella abortus induces protective immunity in BALB/c mice. Infect. Immun. 70:2507 2511. 28. Walters, N., Sura, M., Trunkle, T., and Pascual, D. W. 2005. Enhanced immunoglobulin A response and protection against Salmonella enterica serovar Typhimurium in the absence of the substance P receptor. Infect. Immun. 73:317 324. 29. Wang, X., Hone, D. M., Haddad, A., Shata, M. T., and Pascual, D. W. 2003. M cell DNA vaccination for CTL immunity to HIV. J. Immunol. 171:4717 4725. 30. Xie, H., R. B. Raybourne, U. S. Babu, H. S. Lillehoj, and R. A. Heckert. 2003. CpG-induced immunomodulation and intracellular bacterial killing in a chicken macrophage cell line. Dev. Comp. Immunol. 27:823 834. Editor: J. D. Clements