Six weeks old C57BL/6 female mice, specific-pathogen-free from CEMIB/UNICAMP were housed in ABSL-2 racks fitted with HEPA-filtered air intake and exhaust system. The animals were maintained in controled temperature (20–24°C) and relative humidity (40–70%), and a 12-h light/dark cycle, according to recommendations of the Brazilian National Council for Animal Experimentation (Conselho Nacional de Controle de Experimentação Animal- CONCEA). The protocols used in this study were approved by the Ethics Committee on the Use of Animals of the Federal University of Goiás (Protocol number: 027/14).

The mc² and mc²-CMX vaccines used in this work were produced according to Junqueira-Kipnis et al. (25). The vaccines were diluted in PBS 0.05% Tween 80 (10⁸ CFU/mL), and animals were subcutaneously vaccinated twice with 100 μL/animal in a 15 days interval. BCG Moreau vaccine (10⁷ CFU/mL; 100 μL/animal) was administered once subcutaneously as the control.

Mice were vaccinated subcutaneously with mc² (10⁷ CFU/mouse), mc²-CMX (10⁷ CFU/mouse), BCG (10⁶ CFU/mouse) or saline. Two, five, and seven days after the last immunization, spleen, axillary lymph nodes and the skin and accessory tissues at the immunization site were collected for histological and cytometry analyses.

Cell preparation and cytometry analyses were done as described by Junqueira-Kipnis et al. (25) and da Costa et al. (28). Briefly, mice were euthanized and the lymph nodes, spleen and tissue at the site of vaccine injection were collected. Spleens and lymph nodes were prepared as single-cell suspensions using 70-μm cell strainers (BD Biosciences), and the cells were resuspended with RPMI medium. Erythrocytes were lysed with lysis solution (0.15 M NH₄Cl, 10 mM KHCO₃), and the cells were washed and resuspended with RPMI supplemented with 10% fetal calf serum, 0.15% sodium bicarbonate, 1% L-glutamine (200 mM; Sigma-Aldrich, Brazil), and 1% non-essential amino acids (Sigma-Aldrich). Cells were counted in a Neubauer chamber, and the concentration were adjusted to 1 × 10⁶ cells/mL. The in-situ tissue was digested with DNAse IV (30 μg/mL; Sigma-Aldrich) and collagenase III (0.7 mg/mL; Sigma-Aldrich Brazil) for 30 min at 37°C. The digested tissue was prepared as single-cell suspensions using 70-μm cell strainers and subjected to erythrocyte lysis. The cells were washed and resuspended with RPMI, and the concentrations were adjusted to 1 × 10⁶ cells/mL.

Briefly, 10⁶ cells obtained as described previously, were distributed in a 96-well plate, labeled with the antibodies: FITC-anti-CD3 (clone 145-2C11); PE-anti-CD8 (clone 53-6.7); PE-anti-CD27 (clone LG.7F9); PERCP-anti-CD11b (clone M1 / 70); APC-anti-GR1 (clone RB6-8C5), and incubated for 30 min. Afterward, the cells were washed with PBS containing 0.1% sodium azide and fixed with Perm Fix (BD PharMingen). Subsequently, cell acquisition of 50,000 events per sample was performed in a BD FACSVerse™ flow cytometer (Universidade Federal de Goiás, Faculdade de Veterinária e Zootecnia), and the acquired data were analyzed using FlowJo 9.0 software.

Briefly, the cells were incubated with 3 mM monensin (eBioscience) for 4 h at 37°C in a 5% CO₂ incubator. Subsequently, the cells were labeled with FITC-anti-CD3 (clone 145-2C11) and APC-anti-NK1.1 (clone PK136), APC-anti-GR1 (clone RB6-8C5) for 30 min. After that, cell suspensions were washed with PBS containing 0.1% sodium azide, and fixed and permeabilized with Perm Fix/Perm wash and further incubated with PE-anti-IFN-γ (clone XMG1.2) and PERCP-anti-TNF-α (clone MP6-XT22) for 30 min. After washing with Perm wash (BD PharMingen), the cell suspensions were fixed with Perm Fix. Cell acquisition of at least 50,000 events per sample with minimum 10,000 events in the NK cell gate was performed in a BD FACSVerse™ flow cytometer (Universidade Federal de Goiás, Faculdade de Veterinária e Zootecnia), and the acquired data were analyzed using FlowJo 9.0 software.

C57BL/6 mice were treated every 5 for 30 days by intraperitoneal route with 100 μg of anti-asialo-GM1 (WAKO). One day after the first depletion treatment with anti-asialo-GM1 (WAKO), mice were immunized subcutaneously with mc²-CMX (10⁷ CFU/animal), mc² (10⁷ CFU/animal), BCG (10⁶ CFU/animal) or saline. Fifteen days after the first immunization mice which received mc²-CMX or mc² were boosted with the respective vaccine. Fifteen days after the last immunization, the spleen, axillary lymph nodes and lungs were collected for evaluation of specific immune response induced by vaccination.

For evaluation of DCs and neutrophils, spleen, axillary lymph node, and lung cells were collected and processed as described above. Cells suspensions were labeled with FITC-anti-CD11b (clone M1/70), PE-anti-CD11c (clone N418), PerCP-anti-MHCII (clone M5/114.15.2), and APC-anti-GR1 (clone RB6-8C5) for 30 min all from Ebioscience™. The cells were then washed with PBS containing 0.1% sodium azide, then fixed with Perm Fix (BD PharMingen). 50,000 events were acquired in the BD FACSVerse™ flow cytometer and the data were analyzed with FlowJo 9.0 software. After excluding doublets, cells were gated using the size and granularity of monocytes (FSC-A vs. SSC-A), then DC cell populations were analyzed for specific markers.

10⁶ cells from the spleen, axillary lymph nodes, and lungs were plated in 24 well culture plates, stimulated with ConA (1 μg/mL), CMX (10 μg/mL) or without stimulus (medium) and incubated for 4 h in a 5% CO₂ chamber at 37°C. Then, the plates were incubated for another 6 h with 3 mM monensin (eBioscience). After that, the cells were labeled with FITC-anti-CD4 (GK1.5 clone) for 30 min. Subsequently, the labeled cells were fixed with Perm Fix (BD Cytofix/Cytoperm Kit) for 20 min, and their membranes were permeabilized with Perm wash (BD Cytofix/Cytoperm Kit) for 20 min. After this step, the cells were labeled with the intracellular antibodies: APC-anti-IFN-γ (clone XMG1.2) incubated for 30 min and fixed with Perm Fix. After labeling the cells, 50,000 events were acquired in the Biosciences BD FACSVerseTM flow cytometer and the data were analyzed with FlowJo 9.0 software.

Frozen Mtb aliquots (H37Rv) were diluted in PBS 0.05% Tween 80 to a final concentration of 10⁶ CFU/mL. Mice were challenged 30 days after the last vaccination with 10⁵ CFU/animal, via the intravenous route (lateral tail vein).

The infecting dose was determined by plating the lung homogenate of one mouse per group, 1 day after infection with Mtb. The homogenate was diluted in PBS 0.05% Tween 80 and overlaid on plates containing 7H11 medium supplemented with OADC.

Thirty days after Mtb challenge, the right and medial right lung lobes were collected aseptically, processed, and distributed on plates containing 7H11 medium supplemented with OADC. After distributing the dilutions, the plates were incubated for 21 days at 37°C, and then CFU quantified.

C57BL/6 mice received 100 μg of anti-asialo-GM1 (WAKO) at day −1, by intraperitoneal route; or 200 μg of anti-mouse Ly6G antibodies (BioXcell; clone 1A8) at day −1 and day 2, by intraperitoneal route. At day 0 mice were immunized subcutaneously with mc²-CMX (10⁷ CFU/animal), BCG (10⁶ CFU/animal) or saline. Five days after the immunization the spleen, axillary lymph nodes and lungs were collected for evaluation of neutrophils or NK cells responses induced by vaccination.

The results were tabulated in Excel (version 14.3.4, 2011) and Prism 4 software (Graphpad Software 4.0). The difference between groups was assessed by ANOVA followed by Mann-Whitney test. Differences with p < 0.05 were considered as statistically significant.

Although we have previously observed the presence of neutrophils at the site of vaccination with mc²-CMX and demonstrated their importance for the induction of specific Th1 and Th17 immune responses (27), the demonstration whether those neutrophils produced any cytokines that could influence the general inflammatory outcome was not investigated. Thus, the presence of TNF-α⁺ neutrophils at the site of vaccination with mc²-CMX was evaluated after 2, 5, and 7 days. It can be seen in Figures 1A,B that the percentage of neutrophils, following vaccination with mc²-CMX, presented a huge increase starting as soon as 2 days post infection. BCG vaccination presented significant neutrophils migration only at 7 days post infection when compared to the saline injected skin. In addition, the percentage of neutrophils (GR-1) that were TNF-α⁺ peaked 2 days after vaccinations with mc²-CMX and BCG while mc²-CMX vaccination remained with increased levels at 7 days post infection when compared to saline and BCG (Figures 1C,D). These results indicate that neutrophils migrating to both the mc²-CMX or BCG vaccination sites were able to produce TNF-α. However, the mc²-CMX vaccine induced significantly higher neutrophil migration and activation at the immunization site.

Since mc²-CMX vaccine induced an increase in the number neutrophils at the site of vaccination, we questioned whether those neutrophils could be drained to the secondary lymphoid organs (draining lymph nodes and spleen). Although we are not sure that the neutrophils found in those organs were from the mc²-CMX or BCG vaccination site, similar increase of neutrophils was observed in both organs (Figures 2A,B). However, when verifying the percentage of TNF-α⁺ neutrophils, only the mc²-CMX vaccine induced an increase of these cells in the draining lymph nodes (Figure 2C). Both the mc²-CMX and BCG vaccines induced increase of TNF-α⁺ neutrophils in the spleen (Figure 2D).

Since neutrophils TNF-α⁺ were increased at the site of vaccination and secondary lymphoid organs, and it has already been shown that these cells may favor NK cell survival and activation (14), we evaluated whether NK cells were present at the site of vaccination and whether these cells were also increased or activated in the draining lymphoid organs (Figure 3). Vaccination with both mc²-CMX and BCG induced increase of NK cells at the site of immunization (Figure 3B) as well as in the draining lymph nodes (Figure 3C) and spleen (Figure 3D).

As mc²-CMX vaccine induced increase of NK cells in the lymph nodes, we investigated the phenotype (subpopulations) of these cells (Figure 4A). It was observed that during vaccination with mc²-CMX and BCG there was a predominant increase in cytotoxic NK cell subpopulations (NKc: CD3⁻NK1.1⁺ CD11b⁺ CD27⁻) that accounted for 37.7 and 40.3% of total NK cells, respectively, at day 7. However, twice the percentage of proinflammatory NK cells (NKi; CD3⁻NK1.1⁺ CD11b⁺ CD27⁺) was observed when the animals were vaccinated with mc²-CMX (NKi = 13.8%) when compared to BCG group (NKi = 5.1%) at the same time point. Vaccination with mc²-CMX also induced greater differentiation of NK cells (NKc + NKi = 54.1%) than vaccination with BCG (NKc + NKi = 42.9%) (Figure 4B). The same NK phenotype changes were observed in the spleen (data not shown).

Next, we questioned if the increase of proinflammatory NK subpopulation after mc²-CMX vaccination could have aided the induction of Th1 specific immune response by expressing IFN-γ, thus NK⁺IFN-γ⁺ cells (CD3⁻NK1.1⁺IFN-γ⁺) were evaluated in the different vaccinated groups. Mice vaccinated with mc²-CMX showed higher percentage of NK cells producing IFN-γ than mice from control group, at the vaccination site (Figure 5A), in the draining lymph nodes (Figure 5B) and in the spleen (Figure 5C). Vaccination with mc²-CMX induced higher percentage of NK⁺IFN-γ⁺ compared to BCG vaccinated groups in the lymph nodes at 5 days post infection (^#p < 0.05; Figure 5B) and at the site of injection at days 2 and 5 (Figure 5A).

Since mc²-CMX vaccination induced a greater number of NK cells with a proinflammatory phenotype, it was investigated whether those cells influenced neutrophil or DC migration. For that, C57BL/6 mice were treated with anti- ganglioside (ganglion-N-tetraoxyl ceramide) antibody (anti-asialo antibody; Figure 6A) which resulted in about 80% of the NK cell population depletion, as shown in Figure 6B. NK cell depletion did not affect any change on the number of neutrophils induced by any vaccine treatment 31 days after initiation of anti-asialo treatment (Figure 6). Thus, mc²-CMX or BCG vaccination continued to induce an increase in neutrophil numbers in the draining lymph nodes (Figure 6C), spleen (Figure 6D), and lung (Figure 6E) regardless of the presence of NK cells.

As NK cells can aid DCs in the generation of a specific immune response, it was evaluated whether the absence of these cells could alter the proportion of DCs generated by the vaccines. Figure 7 shows that, in the absence of NK cells, the number of DCs did not change, and vaccination with mc²-CMX or BCG induced an increase of these cells in lymph nodes (Figure 7A), spleen (Figure 7B), and lungs (Figure 7C), when compared with saline. mc²-CMX vaccination presented at this time point, higher levels of DC in all evaluated organs. Although, in the absence of NK, the number of DC did not change, their activation status could have been modified. That was indeed the case, as shown in Figure 8 for all evaluated organs, activated DC were reduced upon NK cell depletion among mc²-CMX vaccinated animals. NK depletion mainly affected DC activation in the spleen (Figure 8B). Thus, NK cells could be important for the activation of dendritic cells, which may be crucial in inducing a specific immune response to vaccines.

Because the absence of NK cells reduced the number of activated DCs after vaccination with mc²-CMX, generation of the specific Th1 immune response (CD4⁺IFN-γ⁺ T cells) could also be altered. In Figure 9, it is observed that in the absence of NK cells, after stimulation with recombinant CMX protein, the number of Th1 lymphocytes was reduced after vaccination with either mc²-CMX or BCG. However, it was noted that despite the reduction of specific Th1 cells, vaccination with mc²-CMX still induced greater numbers of CD4⁺IFN-γ⁺ T cells in the lymph nodes (Figure 9A) and in the lungs (Figure 9D) when compared with BCG.

The absence of NK cells reduced the number of specific CD4⁺IFN-γ⁺ T cells induced by vaccination with mc²-CMX, thus it was enquired whether this could influence the protection generated by the vaccination. The bacillary load was evaluated 30 days after infection with M. tuberculosis and while NK cells depletion did not influence the bacterial load in naïve mice (Figures 10A,D) as our group have previously shown (7) the protection generated by either BCG or mc²-CMX vaccines was significantly diminished in the absence of NK cells (Figures 10B–D). These results suggest that NK cells are important for vaccine Th1 cells generation and required for the protection induced by mc²-CMX and BCG vaccines.

To evaluate the influence of neutrophils on NK cells migration/presence at the site of vaccine injection, or vice versa, mice were vaccinated during each cell population depletion, and the presence of neutrophils or NK cells were investigated at the site of vaccination, lymph nodes and spleens. When NK cells were depleted, the number of neutrophils at the site of vaccination increased even further than in non-depleted group (Figures 11A,B) and maintained with higher levels in the lymph node and spleen (Figures 11C,D). Depletion of neutrophils resulted in reduction on the number of NK cells at the vaccination site and spleens to the same level of the non-vaccinated saline group (Figure 12).