This study was carried in strict accordance with the Brazilian laws 6638 and 9605 in Animal Experimentation. The protocol was approved by the Committee on the Ethics of Animal Experiments of the Federal University of Alfenas (CEUA 16/2020).

The strain C57BL/6 mice aged 6–8 weeks were purchased from the Federal University of Minas Gerais animal facility (UFMG, Belo Horizonte, Brazil). Bone marrow cells were obtained from femora and tibia of mice and they were grown in bone marrow-derived macrophages (BMDMs) as previously described by our group (33). B. abortus virulent strain 2308 was obtained from our own laboratory collection. They were grown in Brucella broth medium (BD Pharmingen, San Diego, CA, USA) for 3 days at 37°C.

BMDMs were infected with virulent B. abortus strain 2308 at a multiplicity of infection of 100:1. Bacteria were centrifuged onto macrophages at 400 × g for 10 min at 4°C and then cells were incubated for 30 min at 37°C under 5% CO₂. Macrophages were extensively washed with HBSS to remove extracellular bacteria and incubated for an additional 90 min in medium supplemented with 100 μg/mL gentamycin to kill extracellular bacteria. Thereafter, the antibiotic concentration was decreased to 10 μg/mL. Thirty minutes after infection, BMDMs were washed three times with HBSS before processing following homogenization with 100μl of LS TRIzol^® reagent Invitrogen (Waltham, Massachusetts, EUA) for total RNA isolation.

The construction and sequencing of a strand-specific small RNA (15–50 nt) library was conducted by FASTERIS SA (Plan-les-Ouates, Switzerland), based on the Illumina^® TruSeq^® Small RNA Library Prep Kit for Illumina HiSeq 2000 sequencing (Illumina Inc., San Diego, CA, USA). To remove adapter sequences from small RNA raw reads of sequencing, the Cutadapt tool was used (34) and sequencing quality was analyzed using the Trimmomatic V0.32 tool (35). Reads were mapped to the B. melitensis biovar Abortus (strain 2308) genome using the Bowtie program (36) to report the best alignment for each read allowing a maximum of one replacement per alignment. Using BedTool, reads mapped at alignment were analyzed to determine the depth of coverage across the genome determining the hotspots areas of small RNAs (37). Coverage with more than 50 reads and areas with a distance of less than 50bp were joined as the same hotspot area.

To assess the putative targets of those small RNAs highly expressed by B. abortus during infection, one hundred coding genes were obtained using the intersect BedTool to indicate the overlap between small RNAs and coding genes (37). As the overlaps showed the possibility of affecting more than 50% of the B. abortus coding genes, which target genes contain the hottest hotspots (i.e., with the highest average coverage), by Watson-Crick base pairing complementarity was evaluated. The proteins from the putative target coding genes were filtered by assigning selection criteria according to biological characteristics: (1) subcellular location (SCL) using Psortb v3.0.2 program (38), (2) the presence of a signal peptide using SignalP 5.0 Server tool (39), and (3) the presence of transmembrane helices and exposed regions using three tools: TMPRED, SOSUI 1.11, and TMHMM 2.0 (40–42). The obtained proteins were then analyzed for biological function using an online tool UniProt Knowledgebase (UniProtKB) and only one of the 7 was selected (43). The selected protein apolipoprotein N-acyltransferase (WP_002965220.1) was used to identify epitopes composed of fifteen amino acids present in the extracellular portion of Int. The extracellular portion of the surface-associated target protein was subjected to the sequential mapping of epitopes predicted to bind tightly to major histocompatibility complex (MCH) class II, using the tools NetMHCII 2.2, SYFPEITHI, and RANKPEP (44–46). To obtain the best candidates of putative immunogenic epitopes from Int, using Multalin 5.4, the obtained epitopes were aligned, and the linear amino acid sequence including approximately 15 amino acids that presented a repeats sequence was selected (47). For the epitope antigenicity test, the VaxiJen 2.0 server was used with the 0.5 and “probable antigen” cutoff (48). For the allergenicity test, the server “AlgPred: Prediction of Allergenic Proteins and Mapping of IgE Epitopes” was used with the hybrid method that consists of using the following five tools available on the server: SVMc, IgE, epitope, ARPs BLAST, and MAST (49). For the epitope similarity analysis, the Protein BLAST tool available on the NCBI platform was used (50). For the physical and chemical properties test, the ExPASy server ProtParam tool was used with a cutoff of 40 for structure stability values (51). The pipeline of bioinformatic analysis is depicted and summarized in Figure 1 .

Thirty-minute B. abortus-infected BMDMs or the exponential growth of B. abortus in Brucella broth medium total RNA were extracted using TRIzol reagent (Invitrogen). Reverse-transcription of 100 ng from total RNA was performed using random primers according to the Illustra™ Ready-To-Go RT-PCR Beads kit (GE Healthcare, Buckinghamshire, UK). Real-time RT-PCR was conducted in a final volume of 10 μL containing the following: SYBR^® Green PCR Master Mix (Applied Biosystems, Foster City, CA, USA), with cDNA as the PCR template and primers to amplify specific fragments corresponding to specific gene targets: IntF: 5`-CTGATGTGATTGTCTGGCCG-3`, IntR: 5`-CCTGAGGTGTCGATTCCAGT-3`, Brucella 16S F: 5`- TCTCACGACACGAGCTGACG -3`, Brucella 16S R: 5`- CCTGAGGTGTCGATTCCAGT -3`. The PCR reaction was performed using the ABI 7500 Real-Time PCR System (Applied Biosystems, Foster City, CA), with the following cycling parameters: 60°C for 10 min, 95°C for 10 min, 40 cycles of 95°C for 15 sec and 60°C for 1 min, and a dissociation stage of 95°C for 15 sec, 60°C for 1 min, 95°C for 15 sec, and 60°C for 15 sec. All data are presented as relative expression units after normalization to the Brucella 16S gene. PCR measurements were conducted in triplicate. The differences in the relative expression were analyzed by Student’s t test with a two-tailed distribution (p < 0.05 indicates statistical significance).

The Int protein from Brucella was retrieved via the NCBI database, followed by molecular modeling by homology via WebServer Phyre², which uses the Markov model for the best possible global alignments to generate the most accurate protein in its main function (52). After modeling, the epitope under investigation from the protein was obtained, and the preparation was performed using the MGL tool, in which missing atoms were corrected and water molecules were removed (53). The same preparation was performed for MHCII molecules found on the PDB server (Protein Data Bank) (54). For docking execution, both files were converted to the PDBQT format required by AutoDock Vina (55), using the Openbabel tool (56). The gridbox was generated around the active sites of the recovered MHCII proteins. To visualize the interactions between epitope-MHCII in 2D and 3D diagram format, the Ligplot+, and Pymol tools were used, respectively (57, 58).

To assess the ability of vaccinal peptide to induce immune responses, 6–8-week-old C57BL/6 mice were contained in cages on a 12:12 light/dark cycle and fed ad libitum with standard rodent diet and no water restrictions. Mice were randomly separated into four treatment groups (n = 5): (I) vaccinated and infected with B. abortus, (II) vaccinated and non-infected with B. abortus, (III) non-vaccinated and infected with B. abortus, (IV) non-vaccinated and non-infected with B. abortus and (V) Adjuvant and infected with B. abortus. Mice were immunized intraperitoneally (IP) with a prime and two boosts (0, 7, 14 days) of vaccine formulation containing 10 μg of peptide in phosphate-buffered saline (PBS), combined with complete Freund’s adjuvant (Sigma-Aldrich, St. Louis, MO, USA) at day 0 and incomplete Freund’s adjuvant at days 7 and 14. Group V (adjuvant and infected) followed the same protocol receiving IP phosphate-buffered saline (PBS) combined with Freund’s adjuvant (complete and incomplete). On day 21, mice were infected intraperitoneally with B. abortus S2308 virulent strain at a dose of 1x10⁶ CFU/animal in 100 μL PBS. Bacterial loads in the spleen, liver, and axillary lymph node from individual animals were homogenized in PBS, serially diluted 10-fold, and plated on Brucella broth agar (Difco, BD-Pharmingen, San Diego, CA). Plates were incubated at 37°C, and the CFU were counted after 3 days as previously described (59). The experimental design is represented in Figure 2 .

To detect the immunogenicity of apolipoprotein N-acyltransferase epitope, antigen-specific IgG, IgG1 and IgG2c in the immunized mice were tested by ELISA. Briefly, the immunoassay plates (Maxisorp, Nunc, Denmark) were coated with apolipoprotein N-acyltransferase epitope (1 μg/mL) diluted in 0.1 M bicarbonate buffer (pH 9.0) and incubated at 4°C overnight. The wells were washed three times with phosphate-buffered saline-Tween20 (PBS-T) and then blocked with PBS containing 5% skimmed milk at 37°C for 1h. Vaccinated and non-vaccinated mice serum (1:10) was added to the well and incubated at room temperature for 2h. After adequate washing with PBS-T, plates were incubated with peroxidase-conjugated anti-mouse IgG (1:2500), IgG1 (1:5000), and IgG2c (1:20000) (Sigma Chemical Co., USA) for 1h at 37°C. After washing with PBS-T, substrate solution (0,5 mM TMB, 20 mM H2O2 no buffer PBS, pH 5) was added to each well, and plates were incubated in the dark at room temperature. Color development was stopped by adding 50 μL of 2 M H₂SO₄ after 10 min. The absorbance value was measured using a microplate reader (Bio-Tropsch Tek Instruments, Winooski, Vt., USA) at a wavelength of 450 nm. To confirm antigen-specificity, the serum from vaccinated and non-vaccinated mice was also used in immunoassay plates coated with the unrelated antigen lysozyme from hen egg white (1 μg/mL) (Sigma Aldrich) following the same parameters described above.

The medial lobes of the mice liver were collected, fixed in 10% buffered formaldehyde solution, dehydrated, diaphanized, and embedded in paraffin. Four-micrometer-thick tissue sections were stained with hematoxylin and eosin (H&E). Digital images were captured and digitized with the AxioVision LE software (Carl Zeiss, Oberkochen, Germany); all of the sample fields were photographed for histopathological evaluation. The histopathological changes were analyzed by Image Pro-PlusR 4.5 software (Media Cybernetics Inc., Silver Spring, MD, USA). The total number and size of granulomas present in histological liver sections was determined using an Axiophot microscope (Carl Zeiss, Oberkochen, Germany) with a 40x objective lens. Immunohistochemistry was performed as previously described (60). Briefly, liver sections were hydrated and incubated with 10% hydrogen peroxide in PBS for 30 min. After being washed with PBS, slides were transferred to a humid chamber at room temperature, incubated with 25 mg/ml of skim milk for 45 min, and then incubated with a primary antibody for 30 min. For immunolabeling, diluted (1:5,000) serum from a rabbit experimentally inoculated with B. abortus S19 strain was used as polyclonal anti-B. abortus antibody. Then, tissue sections were washed with PBS, incubated with secondary antibody for 20 min, washed again with PBS, and incubated for 20 min with streptavidin-peroxidase from a commercial kit (LSAB + kit; Dako Corporation, Carpinteria, CA). The reaction was revealed using 0.024% diaminobenzidine (DAB; Sigma), and sections were counterstained with Mayer’s hematoxylin.

Spleens cells from C57BL/6 mice under treatment obtained after maceration were treated with ACK buffer (0.15 M NH4Cl, 1.0 mM KHCO3, 0.1 mM Na2EDTA, pH 7.2) to lyse red blood cells. After that, the cells were washed with saline (NaCl 0.8%, wt/vol) and suspended in RPMI 1640 (Gibco, Carlsbad, Calif) supplemented with 2 mM L-Glutamine, 25 mM HEPES, 10% (vol/vol) heat-inactivated FBS (Gibco, Carlsbad, Calif), penicillin G sodium (100 U/mL), and streptomycin sulfate (100 μg/mL). Spleen cells (1x10⁶) were cultured in 200µL culture medium and incubated at 37°C with 5% CO₂. The supernatant of splenocyte cultures was collected after 24h and nitric oxide (NO) measurement was performed according to the Griess method (61). The levels of IFN-γ secreted were quantified from the supernatant of splenocytes cultures by antigen-capture ELISA. For this, 1x10⁶ splenocytes (per well) were cultured in 200 μL of culture medium in the absence and presence of 1ug of the peptide as a stimulus for 72h at 37°C with 5% CO₂. Aliquots from each well were then taken and IFN-γ levels were measured using the Murine IFN-γ Mini ABTS ELISA Development Kit (PeproTech Inc) following the manufacturer’s instructions. Final cytokine concentrations were calculated using the standard curve for IFN-γ. The final reaction was measured using a microplate reader (Bio-Tropsch Tek Instruments, Winooski, Vt., USA) and read at 405 nm with wavelength correction set at 650 nm.

BMDMs (1x10⁶ cells per well) were plated on imaging slides (µ-Slide 12-well, glass bottom, Ibidi GmbH, Munich, Germany), followed by stimulation with splenocytes supernatant for 48h. The cells were then washed three times with PBS and incubated with the anti-CD16/32 antibody (BD Biosciences, San Jose, CA) for 2 hours to block nonspecific bonds. The cells were then incubated with anti-CD86 and anti-CD11b, followed by staining with FITC-conjugated and PE-conjugated (BD Biosciences), respectively, overnight at 4°C. The slides were washed with PBS and the nuclei were stained with 150 ng/mL 40,6-diamino-2-phenylindole (DAPI; Thermo Scientific) for 1 hour. All images were captured using a Nikon Eclipse 80i fluorescence microscope (Melville, New York, U.S.A). Image J software was used to analyze the markings obtained for the nucleus (blue fluorescence), CD11b+ cells (green fluorescence), and CD86+ cells (red fluorescence).

Liver and spleen macerate as well as BMDMs stimulated with splenocytes supernatant from the four experimental groups were homogenized with TRIzol reagent (Invitrogen) to isolate total RNA. Reverse-transcription of 1 μg total RNA was performed using Illustra™ Ready-To-Go RT-PCR Beads (GE Healthcare, Buckinghamshire, UK). Real-time RT-PCR was conducted in a final volume of 10 μL containing the following: SYBR^® Green PCR Master Mix (Applied Biosystems, Foster City, CA, USA), with cDNA as the PCR template and primers to amplify specific fragments corresponding to specific gene targets: TNF-α F: 5’-CATCTTCTCAAAATTCGAGTGACA-3’, TNF-α R: 5’-TGGGAGTAGACAAGGTACAACCC-3’; IFN-γ F: 5’-TCTGGAGGAACTGGCAAAG-3’, IFN-γ R: 5’-TTCAGACTTCAAAGAGTCTGAGG-3’; IL-6 F: 5’-CCAGGTAGCTATGGTACTCCAGAA-3’, IL-6 R: 5’-GATGGATGCTACCAAACTGGA-3’; IL-10 F: 5’-GGTTGCCAAGCCTTATCGGA-3’, IL-10 R: 5’-ACCTGCTCCACTGCCTTGCT-3’; TGF-β F: 5’-TGACGTCACTGGAGTTGTACGG-3’, TGF-β R: 5’-GGTTCATGTCATGGATGGTGC-3’; iNOS F: 5’-CAGCTGGGCTGTACAAACCTT-3’, iNOS R: 5’-CATTGGAAGTGAAGCGTTTCG-3’. The PCR reaction was performed with ABI 7500 Real-Time PCR System (Applied Biosystems, Foster City, CA), using the following cycling parameters: 60°C for 10 min, 95°C for 10 min, 40 cycles of 95°C for 15 sec and 60°C for 1 min, and a dissociation stage of 95°C for 15 sec, 60°C for 1 min, 95°C for 15 sec, 60°C for 15 sec. All data are presented as relative expression units after normalization to the β-actin gene (F: 5’-AGGTGTGCACTTTTTATTGGTCTCAA-3’, R: 5’-TGTATGAAGGTTTGGTCTCCCT-3’). PCR measurements were conducted in triplicate. The differences in the relative expression were analyzed by analysis of variance (ANOVA) followed by Tukey’s test (p < 0.05 denotes statistical significance).

Graphs were created and data analysis was performed using GraphPad Prism 8 software (San Diego, CA, USA), using one-way ANOVA or two-way ANOVA (Bonferroni post hoc test). Values <0.05 were considered statistically significant.

In the small RNA libraries from infected macrophages, we observed that 7.26% of all mapped sequences belonged to the B. abortus genome ( Table 1 ). Knowing this, using the Bowtie software, it was seen that the small RNAs were found to be distributed in both chromosomes of the bacteria. By analyzing the depth of coverage of the sequences, we identified a total of 3954 regions of broad mapping of small RNAs in the genome of B. abortus, 2694 in chromosome I and 1260 in chromosome II. However, in local data of the bacterial genome, the presence of three peaks (hotspots) was observed, with two in chromosome I and one in chromosome II, constituted by small RNAs being mapped in the same position of the genome and large quantities ( Supplementary Figure 1 ). With that, these three hotspots were selected, and we next concentrated our analysis on selecting the small RNAs that were mapped multiple times in these hotspots ( Supplementary Figure 2 ). In total, we identified one hundred small RNAs with this characteristic which were then submitted to further analysis to detect the target mRNA in NCBI.

After the sequences were obtained from NCBI, the surface-associated proteins were selected by our SCL prediction tools. In this phase, 10 proteins were included, making up the final list of surface-associated proteins. The exposure of extracellular structures is an attraction for the immune system in the recognition of antigens and, therefore, the topology predictive tools selected seven proteins composed of extracellular domain amino acids. All selected extracellular portions were composed of at least 35 amino acids. To gain more insight into the biological functions of these 7 proteins, an analysis was performed to identify biological patterns associated with antigens. Of the 7 proteins analyzed, all were inferred by homology in UniProtKB, and a few are those reviewed regarding their biological function ( Table 2 ). The information obtained by the Gene Ontology (GO) project showed that only two of these proteins would be involved in the biosynthesis process of the bacterial structure, yet one of them had a smaller number of exposed amino acids. The other, in addition to exposing 277 amino acids in the extracellular portion, also participates in the bacterial lipopolysaccharide biosynthesis process, becoming the target protein for further analyses. The other proteins analyzed had a transmembrane transport function, in addition to mediating cellular vesicle fusion processes.

Therefore, apolipoprotein N-acyltransferase (BAB1_2158) was selected for the further analysis of differential expression during BMDM infection. Knowing that small RNAs were expressed in large quantities by the bacterium during intracellular infection by Brucella, we investigated the levels of mRNA expression of Int in samples of intracellular and extracellular growth in a model of BMDM infection with B. abortus. The result of differential expression analysis showed that during BMDM infection, the bacteria decreased the Int gene expression when compared to an exponential extracellular growth model ( Figure 3A ). Taking into account the results of the in silico analysis, the negative expression of the Int coding gene may be related to an initial post-transcriptional gene regulation process used by the bacteria to repress the expression of this protein during intracellular infection, thus establishing its replicative niche in the host cell. After reaching the replicative niche, B. abortus may increase the Int coding gene since it plays a crucial role in bacterial LPS biosynthesis during replication.

Since the results indicate a possible mechanism of gene expression control by the bacterium in stressful intracellular environments, we were searching for exposed immunogenic epitopes in the extracellular amino acid sequence of Int. The results obtained indicated approximately 60 epitopes that showed a strong binding affinity for HLA molecules (HLA-DRB1 0101, HLA-DRB1 0301, HLA-DRB1 0401, HLA-DRB1 0701, HLA-DRB1 1101, and HLA-DRB1 1501). In this stage, the predictor tools selected epitopes composed of exactly 15 amino acids. When subjected to alignment with the extracellular portion of the protein, the most promiscuous epitopes formed a conserved region; from this region, the epitope “AIPYILESTPQALAH” was selected. To identify the position of the selected putative immunogenic epitope in the Int, we performed molecular modeling, showing the selected epitope highlighted in red with transmembrane helices indicated in the model in green, as indicated in Figure 3B . When submitted to in silico screening tests, the epitope did not show the desired antigenicity (0.2800), however showed 100% similarity with sequences present only in the apolipoprotein N-acyltransferase of B. abortus. To increase the confidence of using the selected epitope in future in vitro and in vivo evaluations, allergenicity was tested using the AlgPred software, characterizing it as non-allergenic. Although when analyzing the physicochemical properties, the epitope was shown to be unstable (40.27), since it did not reach the minimum cut-off value for stability, it had a relatively good half-life in mammalian cells (Half-life reticulocytes: 4.4h; Half-life yeast: >20h; Half-life E. coli: >10h), that is, despite its instability in the medium, it is suggestive that when it binds to the MHCII cleft, it can present the desired stability.

Although not all structures of MHC-II were found in the database (IEDB), we used the structure of 3 available proteins: HLA-DRB1:0101 (PDB: 5NI9), HLA-DRB1:0401 (PDB: 5V4M) and HLA-DRB1:1501 (PDB: 5V4N). When performing the molecular docking of the “AIPYILESTPQALAH” epitope, the results showed great interaction energy of all alleles based on the AutoDock Vina software ( Table 3 ). To represent the molecular docking between the epitope and the MHC-II allele, we chose the interaction with the highest interaction power pointed out by the software. The epitope in question interacted very well with the MHCII of the HLA-DRB1:0101 allele, having the best interaction values, with an energy of -8.1 Kcal.mol^-1 and presenting a total of 9 hydrogen bonds between the epitope and the MHC, with two bonds involving the amino acid HIS259 of the MHC with the amino acids LEU303 at a distance of 3.09Å and with the amino acid ALA304 at a distance of 3.26Å from the epitope. One link of amino acid ASN260 of MHC with amino acid HIS305 was 3.21Å from the epitope, one link of amino acid GLN242 of MHC with amino acid PRO293 showed a distance of 3.09Å from the epitope, two linkages involving amino acid TYR238 of MHC with the amino acid ILE292 was 2.91Å away and the link with the amino acid ALA291 was at a distance of 3.00Å from the epitope; also, a link of the amino acid LYS249 of the MHC with the amino acid GLU297 was at a distance of 3.06Å and finally two bonds involving amino acids TYR208 and HIS191 of the MHC, showed distances of 2.90Å and 3.24Å, respectively, with amino acid SER298 of the epitope ( Figures 4A, B ). In summary, the peptide was shown to have a great ability to interact with MHC-II slits, especially with the HLA-DRB1:0101 allele, showing excellent interaction with 9 of the 15 amino acids of the epitope. Although minor binding strengths have been identified, this does not preclude the possibility of interaction between peptide-alleles.

After immunization, vaccinated and unvaccinated mice were challenged by intraperitoneal infection with B. abortus; after seven days, organs were collected to determine bacterial loads. In the spleen, liver and axillary lymph node of mice vaccinated with the peptide, fewer viable bacteria were recovered when compared to the control group ( Figure 5A ).

To evaluate the potential use of the epitope-based vaccine, the protection level induced in mice against virulent challenge infection was assessed. The degree of vaccine efficacy in C57BL/6 mice was determined by subtracting the mean CFU/organ recovered from mice after vaccination and challenged from the mean CFU/organ recovered from non-vaccinated but challenged control mice. At this time, it was seen that the presence of vaccinal peptide in the animal organism triggered a higher degree of protection against infection, approximately 1.20/0.80/0.84-log in spleen, liver, and axillary lymph node, respectively ( Table 4 ). Finally, aiming to exclude the bias of adjuvant composing the vaccine solution, the group that received only adjuvant was evaluated. The result showed 6.12 ± 0.36, 5.73 ± 0.09, and 5.73 ± 0.12 for spleen, liver, and lymph node, respectively, where there was no statistically significant difference compared to the unvaccinated group. Ours results showed that the RV-selected epitope provided significant protection to C57BL/6 mice against B. abortus.

To verify the peptide potential to induce a specific humoral immune response in immunized mice, the levels of specific IgG, IgG1, and IgG2c antibodies were measured by ELISA in mice serum. The immunoglobulins in immunized mice with peptide vaccine were significantly higher than those of the non-vaccinated group ( Figure 5B ). It was not observed reaction to unrelated antigen lysozyme from hen egg white. In addition, the amount of IFN-γ showed significantly increased in splenocytes from vaccinated mice stimulated with peptide when compared to splenocytes from non-vaccinated mice, or when compared to non-stimulated cells ( Figure 5C ), indicating the specificity of the immune response of immunized mice not only in humoral but also in cellular response.

B. abortus infection is associated with the formation of focal granulomatous lesions in the spleen, liver, and lymphoid tissues of both humans and rodents, starting 1–2 weeks post-infection (62). To determine the characteristics of liver pathology upon vaccination with peptide during B. abortus infection, we performed the histopathological analysis of liver tissue from vaccinated and unvaccinated C57BL/6 mice and, challenged with B. abortus. Infection with B. abortus resulted in the formation of hepatic granulomas in both groups ( Figures 6A, C ). In morphometric analysis, a greater number of granulomas was observed in the tissue of animals that were not vaccinated with the peptide and challenged by bacteria compared to the vaccinated group ( Figure 6G ). The same result was seen when analyzing the area, which was greater in granulomas from unvaccinated animals ( Figure 6H ). Histopathological lesions during Brucella infection usually are associated with the bacterial load. To determine the relationship between granuloma formation and the bacteria present in the granuloma, we performed immunohistochemistry to immunolabel B. abortus. Figures 6E, F showed the presence of B. abortus in the granulomatous lesions presented in the liver of vaccinated and infected mice and non-vaccinated and infected mice, respectively. The detection of intralesional bacteria confirms that the inflammatory lesions described in this study are due to systemic B. abortus infection. No observable lesions were found in tissues from uninfected mice ( Figures 6B, D ).

The liver is the most commonly affected organ in patients with active brucellosis (63), which is why liver macerates from the four experimental groups were collected for evaluation of the expression of the anti-inflammatory genes IL-10 and TGF-β. In the differential expression analysis, an increase in these cytokines was observed in vaccinated and infected C57BL/6 mice when compared to the other groups ( Figures 7A, B ). Therefore, the reduction in liver pathology can be attributed to a decrease in the number of viable bacteria and an increase in anti-inflammatory cytokines, resulting in a consequent reduction in liver damage. Concomitantly, the splenic tissue was evaluated and the results obtained showed that all groups, except for the control, showed upregulation of the proinflammatory cytokine-coding genes INF-γ, TNF-α, and IL-6, which are characteristic of inflammation ( Figures 7C–E ), but the vaccinated and infected group stood out due to the increased expression when compared to other groups. Likewise, it was seen that IL-10 ( Figure 7E ) is up-regulated in this group, suggesting that vaccination induces an attempt to control the inflammatory process generated by systemic infection with B. abortus.

In vivo analysis results indicate the activation of an adaptive immune response. Knowing this, we tried to understand the mechanisms by which a more efficient immune response activation process against the bacteria occurs. BMDMs were evaluated for the expression of co-stimulatory molecules after being stimulated with the splenocyte supernatant. The results obtained by fluorescence microscopy show that BMDMs stimulated with the supernatant from the spleen of vaccinated and infected animals were more activated, due to intense CD86 labeling ( Figure 8A ), and that they even expressed a higher level of iNOS expression, the gene that stimulates NO production ( Figure 8B ). To confirm this result, the measurement of NO was performed in the in the supernatant of these macrophages. The results obtained showed the greater production of NO by BMDMs that were more activated ( Figure 8C ), which may reveal an increased phagocytic and microbicide capacity to eliminate the bacteria.