Innate Immunity Trained in the Protective Response of Vaccine Candidates Against Intracellular Pathogens

TL;DR

This study shows that trained innate immunity can boost the effectiveness of inactivated vaccines against certain bacterial infections but may reduce the effectiveness of live vaccines.

Contribution

The study reveals that β-glucan enhances inactivated vaccine efficacy against Brucella ovis and Listeria monocytogenes but impairs live attenuated vaccine performance.

Findings

β-glucan reduced bacterial colonization in mice vaccinated with gamma-irradiated Brucella ovis.

50% of trained and vaccinated mice showed no detectable Listeria monocytogenes after challenge.

Overstimulation with β-glucan doses impaired infection control, suggesting a negative effect on trained immunity.

Abstract

Background/Objectives: Trained innate immunity refers to the enhanced responsiveness of innate immune cells, particularly macrophages, following exposure to stimuli such as β-glucan or zymosan, enabling improved defense against unrelated pathogens. This phenomenon has been widely investigated to better understand host–pathogen interactions and to support the development of improved infection control strategies. This study evaluated whether these training stimuli could enhance the protective efficacy of attenuated or inactivated vaccine models against Brucella ovis and Listeria monocytogenes infection. Methods: Trained innate immunity was induced in vivo using β-glucan or zymosan, and seven days later mice were vaccinated with attenuated or gamma-irradiated formulations and subsequently challenged with B. ovis or L. monocytogenes. Vaccine-induced protection and immune responses were assessed through multiple experimental approaches. Results: β-glucan significantly reduced bacterial infection in vitro in bone-marrow-derived macrophages and in vivo in target organs compared with zymosan. Although β-glucan did not enhance the efficacy of the attenuated B. ovis ΔabcBA vaccine, it markedly reduced bacterial colonization in mice vaccinated with gamma-irradiated B. ovis. β-glucan also did not improve the efficacy of the gamma-irradiated L. monocytogenes vaccine; however, 50% of the trained and vaccinated mice showed no detectable bacterial recovery. Increasing the number of β-glucan doses negatively affected infection control, suggesting that overstimulation may impair trained immunity. Conclusion: Trained innate immunity enhances the protective effect of inactivated experimental vaccines against B. ovis and L. monocytogenes, while exerting a detrimental influence on the efficacy of a live attenuated B. ovis vaccine model.

Figures

Full text

1. Introduction

Trained innate immunity is characterized by cells of the innate immune system, especially macrophages, that become more effective at controlling infectious agents after being sensitized by substances unrelated to the agents themselves [1,2].

Trained innate immunity was demonstrated by vaccination against tuberculosis using the Bacillus Calmette-Guérin (BCG) vaccine strain [2], which induces trained innate immunity that protects against numerous infectious agents, even though the host organism has never been exposed before [1,2]. Other microorganisms, such as Candida albicans [3,4], and substances such as polysaccharides, zymosan [5], and β-glucan [6], can also induce immune training in macrophages. These components are recognized by receptors that trigger related epigenetic and metabolic modifications in cells [7,8] that result in greater capacity to eliminate infectious agents and greater induction of pro-inflammatory cytokine synthesis [3,6,7,8]. Therefore, activation of innate immune cells by training can greatly contribute to protection against numerous infectious agents, especially intracellular pathogens [1,5].

Bacteria of the genera Brucella and Listeria are intracellular pathogens that cause significant diseases in humans and animals. Brucella spp. can infect various species of domestic animals, causing a chronic disease primarily affecting the reproductive system, resulting in abortion, stillbirth, placentitis, epididymitis, and orchitis [9,10]. Infection with L. monocytogenes has various clinical manifestations such as encephalitis, abortion, and septicemia in neonates [11,12]. One of the most effective ways to control and eradicate infectious diseases is through vaccination [13]. There has been a major research effort on the development of novel vaccines. Vaccinology in brucellosis has been intense over the past decades [14], and there are several studies seeking to develop a vaccine against listeriosis [15,16].

Although some studies have evaluated the ability of trained innate immunity to control Brucella or Listeria infection in vitro or in vivo [5,17], the role of trained immunity in favoring an effective vaccine response against these infections remains unexplored. Therefore, our hypothesis is that vaccine efficacy may be improved by training innate immunity prior to vaccination. In this study, we examined the impact of trained immunity on two distinct vaccine models, one based on an attenuated strain and the other on inactivated whole antigen, against B. ovis [18] and L. monocytogenes [14] infections in a murine model, respectively.

2. Materials and Methods

2.1. Animals

This study was carried out based on vaccine preparations developed by our research group, which have known protective characteristics [15,19,20].

BALB/c and C57BL/6 female mice, aged six to eight weeks, were obtained from the Universidade Federal de Minas Gerais (UFMG) Central Animal Facility. The mice were housed in a suitable animal experimentation facility with a controlled environment throughout the experiment. The facility had alternating 12 h light/dark cycles, 70% humidity, and a temperature of 25 °C. The mice were fed sterilized commercial feed and given water ad libitum.

All procedures involving cell cultures and bacterial recovery were carried out in a laminar flow hood in the Biosafety Level 2 Molecular Pathology Laboratory at the UFMG Veterinary School.

2.2. Bacterial Strains and Culture Conditions

This study used the wild-type bacterial strains B. ovis ATCC 25840, L. monocytogenes serotype 4b ATCC 19115 and B. ovis ∆abcBA, a mutant strain generated from the deletion of the open reading frame (orf) abcA and orf abcB of the abcEDCBA locus located on chromosome II of B. ovis which encodes an ABC-type transporter system [21].

Brucella ovis strains were grown on tryptone soy agar medium (TSA, Invitrogen, Carlsbad, CA, USA) containing 1% hemoglobin (Becton-Dickinson, Franklin Lakes, NJ, USA) for three days at a constant temperature of 37 °C in a humidified incubator with a 5% CO_2_ atmosphere, and 100 µg/mL kanamycin (Gibco, Thermo Fisher Scientific, Waltham, MA, USA) was added to the plates of the mutant strain.

The bacterial colonies were resuspended in a sterile solution of phosphate-buffered saline (PBS/pH 7.4, Gibco, Thermo Fisher Scientific, USA) and the bacterial concentration in the suspension was measured by the optical density at 600 nm (OD_600_) in a spectrophotometer (Bio-Rad, Hercules, CA, USA). It was assumed that at OD_600_ of 0.250 the estimated concentration of bacteria is approximately 3 × 10^9^ and 1 × 10^9^ CFU/mL for the wild-type and mutant strains respectively. The bacterial cultures were prepared at the desired working concentration. To confirm the concentration of the culture, serial dilution and plating were carried out in TSA with 1% hemoglobin [19].

For L. monocytogenes culture, a started culture was prepared and grown in Brain Heart Infusion broth medium (BHI, Kasvi, Pinhais, Brazil) maintained under agitation at 200 RPM and at 37 °C. for 15 h. After this period, the concentration of this started culture was adjusted using a spectrophotometer (Bio-Rad, USA) considering that at OD_600_ of 1is 1 × 10^9^ CFU/mL, and aliquots of 500 µL containing 1 × 10^8^ CFU/mL were prepared and stored in a freezer at −80 °C in a solution containing 60% glycerol. To prepare the inoculum, 500 µL aliquots containing 1 × 10^8^ CFU/mL were added to Falcon tubes containing 10 mL BHI broth, shaken at 200 RPM and at a temperature of 37 °C for approximately 4 h until this bacterial suspension reached a concentration of 1 × 10^9^ CFU/mL. The concentration of bacterial suspension was confirmed after serial dilution in sterile PBS and plating on BHI agar (Kasvi, Pinhais, Brazil) [15].

2.3. Inactivation of L. monocytogenes and B. ovis by Gamma Irradiation

A volume of 100 mL of sterile PBS containing L. monocytogenes or B. ovis at a concentration of 1 × 10^9^ CFU/mL was subjected to gamma radiation at a dose of 5 kilogray at the Nuclear Technology Development Center (CDTN—UFMG, Belo Horizonte, Brazil). After this process, an aliquot of this suspension was plated on BHI agar (L. monocytogenes) or TSA with 1% hemoglobin (B. ovis), where no colony growth was observed, thus confirming bacterial inactivation [15,20].

2.4. Preparation of Vaccines Encapsulated in Alginate and Chitosan

The vaccine strains were encapsulated with alginate (Sigma-Aldrich, Saint Louis, MO, USA) and chitosan (Sigma-Aldrich USA), based on a bacterial suspension containing a concentration of 1 × 10^9^ CFU/mL of B. ovis ∆abcBA [16], 1 × 10^9^ CFU/mL of gamma-irradiated B. ovis (BO-γ-AC) [20] and 1 × 10^8^ CFU/mL of gamma-irradiated L. monocytogenes (KLM-γ-AC) [15]. For this procedure, the bacterial suspension was prepared in sterile PBS, centrifuged at 200× g for 10 min at 4 °C; then the supernatant was discarded, and the bacterial pellet was resuspended in 1% alginate solution. Using an insulin syringe (1 mL) attached to a 33G needle, this solution was dripped into a polymerizing solution (CaCl 0.5 mM) and left to slowly homogenize for 15 min. After the capsules were formed, they were washed twice, under slow homogenization for 5 min, with a solution of MOPS (10 mM, Sigma-Aldrich, USA) with 0.85% NaCl at pH 7.4. After washing, the alginate capsules were immersed in a solution composed of chitosan containing 1% acetic acid and sodium acetate at pH 5.0, for 30 min under homogenization [19].

2.5. Preparation of the Substances Used to Train Innate Immunity

The β-glucan (purified insoluble β-glucan) was isolated from Saccharomyces cerevisiae extract (GoldCell, Biorigin, Lençóis Paulista, Brazil), containing 71.9% total glucans, according to the procedure described by Stone (2009) [22], with modifications. Briefly, 10 µL of extract was washed with 90 mL of acetone and centrifuged at 8000 rpm for 10 min. The resulting pellet was collected and oven-dried at 35 °C. An aliquot of the supernatant was retained for dry matter determination. Subsequently, the pellet was treated for 1 h with 90 mL of a 0.25 M solution at 70 °C, followed by centrifugation at 8000 rpm for 10 min. The pellet was again collected and oven-dried at 35 °C, and an aliquot of the supernatant was used for dry matter determination. The final concentration of insoluble β-glucan was estimated at 82.12% by the difference between the initial and final dry matter contents. The methodology used to prepare β-glucan suspension was similar to that described by Cheng et al. (2014) [8]. A volume of 50 mL of sterile PBS containing a concentration of 1 mg of β-glucan per 100 µL (for in vivo use) was autoclaved and checked for sterility.

The zymosan suspension was prepared similarly to the method described by Ciarlo et al. (2020) [5]. Zymosan (Z4250, Sigma-Aldrich, Merck KGaA, Germany) was diluted in sterile PBS to a final concentration of 0.1 mg/mL (for in vivo use). After dilution, the suspension was left for 30 min in a laminar flow hood under ultraviolet light and then checked for sterility.

2.6. Isolation of Bone-Marrow-Derived Macrophages from Mice

To obtain macrophage precursor cells from the bone marrow, the methodology used was similar to that described by Zhang et al. (2008) [23]. During the necropsy, the pelvic limbs were aseptically disarticulated, followed by dissection of the muscles. After this procedure, the limbs were submerged in 70% alcohol for a maximum of 15 min, and then the bone marrow from the femur and tibia was removed.

The epiphyses of both bones were removed and with a 26 G needle coupled to a syringe containing sterile PBS solution with 100 µL/mL streptomycin/penicillin (10,000 U penicillin, 10 mg streptomycin/mL) (Thermo Fisher Scientific, USA); the marrow cavity was washed and the contents dispensed into a 50 mL Falcon tube. After the bone marrow had been completely removed, the contents were centrifuged at 6500 RPM for 10 min at 4 °C. After this stage, the supernatant was discarded and the cell sediment was resuspended in 10 mL of Roswell Park Memorial Institute medium (RPMI, Gibco, Thermo Fisher Scientific, USA) containing 10% fetal bovine serum (BFS, Gibco, Thermo Fisher Scientific, USA), 20% supernatant from the L929 cell culture and 100 µL/mL streptomycin/penicillin (10,000 U penicillin, 10 mg streptomycin/mL). Then 1 mL of this cell suspension was added to Petri dishes and a volume of 9 mL of complete RPMI medium with 20% supernatant from L929 was added to each one. These plates were incubated at 37 °C for a period of 7 days, and on the third day 10 mL of complete RPMI medium with 20% L929 supernatant was added.

After 7 days, macrophages were removed from the culture plates, and the cell concentration was checked. The supernatant was discarded and 3 mL of Trypsin (Gibco, Thermo Fisher Scientific, USA) was added to each plate for 10 min at 37 °C. The cell suspension was removed to a 50 mL Falcon tube, and the plate was washed with 10 mL of sterile PBS. After this procedure, cell suspensions were centrifuged at 4000 RPM for 10 min at 4 °C. The supernatant content was then discarded, and the cells resuspended in 1 mL of RPMI medium containing 10% BFS. The cell number was counted in a Neubauer chamber using the trypan blue exclusion method at a concentration of 0.3%. Macrophages were plated to 96-well culture plates at 1 × 10^5^ cells per well and cultured for a period of 24 h before in vitro infection.

2.7. In Vitro Evaluation of the Training on the Microbicidal Activity of Bone-Marrow-Derived Macrophages

An in vitro experiment was conducted to assess the effects of stimulators on the training of innate immune cells. The experiment used bone-marrow-derived macrophages from mice that had been treated with zymosan or β-glucan, intraperitoneally (i.p.) [5]. Twelve animals were used and divided into three groups (n = 4): the zymosan group, the β-glucan group, and the PBS (untreated) group. Cultured macrophages were inoculated, and intracellular survival curves of B. ovis and L. monocytogenes were determined.

2.8. In Vitro Infection of Bone Marrow Macrophages

Macrophages were inoculated with B. ovis (multiplicity of infection [MOI] of 100) and L. monocytogenes (MOI of 10). Intracellular bacterial recovery from B. ovis-inoculated macrophages was carried out at 0, 24 and 48 h post-infection (hpi). For L. monocytogenes, intracellular bacterial recovery was performed at 0, 4 and 8 hpi. Bacteria were added to the wells containing macrophages and centrifuged at 6500 RPM for 5 min at 15 °C. Plates were then incubated at 37 °C and 5% CO_2_ for 30 min. After this time, the wells were washed with sterile PBS and RPMI medium supplemented with 10% FBS and containing 50 µg/mL of gentamicin was added for one hour at 37 °C to kill any extracellular bacteria. Culture media in the subsequent time points were changed to RPMI medium supplemented with 10% FBS and containing 25 µg/mL gentamicin. At each point, plates were washed three times with PBS, and cells were lysed with sterile distilled water. The lysates were then serially diluted and intracellular bacteria was counted by plating on TSA plates containing 1% hemoglobin for B. ovis and BHI agar plates for L. monocytogenes. It should be noted that experiments using this methodology involved four or six mice, all in triplicate. The supernatant from uninfected or infected macrophages at each time point was collected and stored at −20 °C.

2.9. In Vivo Evaluation of the Training on Vaccine Protection

Several experiments were carried out to evaluate the effect of training on vaccine response. All experiments used six female mice between six and eight weeks of age per group (n = 6). In one experiment Balb/c mice were used to assess protection against L. monocytogenes infection, whereas in the other experiments C57BL/6 mice were used. These two different strain mice were used considering different susceptibility to intracellular pathogens infection. The C57BL/6 mice are more resistant to Listeria infection [24]. Mice were divided into four groups: those stimulated only with training substances (zymosan or β-glucan); those that received only the vaccine strain (KLM-γ-AC, BO-γ-AC or ∆abcBA-AC, according to the infectious agent with which they would be challenged); those that were previously trained and then immunized with the vaccine strain; and those that did not receive any treatment (mock infected). All groups were subsequently challenged with their respective virulent strains, i.e., L. monocytogenes or B. ovis.

Mice from the stimulated groups received zymosan or β-glucan i.p. at a dose of 1 mg/100 µL [8], and non-stimulated mice received sterile PBS under the same conditions at seven days before vaccination (day 0). After seven days (day 7), mice from the non-immunized group were inoculated with sterile PBS, while those from the vaccinated groups were immunized subcutaneously (s.c.). Animals immunized with KLM-γ-AC or BO-γ-AC received two vaccine doses, one on day seven and the booster at 15 days after the first dose, at a concentration of 1 × 10^7^ colony-forming units (CFU)/animal s.c. Mice immunized with B. ovis ∆abcBA-AC received a single 1 × 10^8^ CFU/animal dose, s.c.

Four weeks post the first immunization, the mice were challenged. Animals immunized with KLM-γ-AC were inoculated i.p. with 100 µL of 1 × 10^5^ CFU/animal of the wild type of strain of L. monocytogenes and euthanized four days post-infection. The animals immunized with B. ovis ∆abcBA or BO-γ-AC were inoculated i.p. with a dose of 1 × 10^6^ CFU/animal of the wild type of strain of B. ovis in a volume of 100 µL and then euthanized two weeks post-infection. All mice were challenged, except the animals from the uninfected mock group.

For euthanasia, an anesthetic overdose of xylazine hydrochloride (2%, 30 mg/kg, Syntec, Santana de Parnaíba, Brazil) and ketamine hydrochloride (1%, 210 mg/kg, Syntec, Brazil) were injected i.p. in a volume of 100 µL. During necropsy, spleen was collected to assess bacterial recovery and determine the protection index of each vaccine in the spleen. The protection index is determined by the difference in mean of CFU in the spleen of non-immunized mice and treated or immunized mice.

2.10. Ex Vivo Evaluation of Training on Vaccine Protection

To evaluate the immune response induced by training and vaccination in mice, the animals were trained with β-glucan seven days prior to being vaccinated with KLM-γ-AC. Female C57BL/6 mice aged six to eight weeks were divided into four groups of six animals (n = 6). The groups were as follows: group stimulated with β-glucan and immunized with KLM-γ-AC; group stimulated with β-glucan; group immunized with KLM-γ-AC; and group stimulated with PBS (mock). Fifteen days after the final vaccination, the mice were euthanized, and cells were collected from their spleens for ex vivo analysis of cell viability (item 2.11) and Interferon gamma (IFNγ) and from their bone marrow for differentiation into macrophages and in vitro evaluation of bacterial intracellular survival (item 2.8) and Interleukin-6 (IL-6).

2.11. In Vitro Stimulation of Murine Splenocytes

After the training and vaccination protocols described above, mice were euthanized four weeks after the first dose of the vaccine. At necropsy, spleen was removed aseptically to cell isolation. Spleens were macerated using a syringe plunger in a sterile Petri dish with 3 mL of RPMI medium enriched with 10% BFS at a temperature of approximately 4 °C. Cell suspensions were centrifuged at 6500 RPM for 10 min at 4 °C. After this procedure, the supernatant was discarded, and the cell pellet resuspended in 5 mL of red blood cell lysis buffer (Tris-ammonium chloride, Sigma-Aldrich) for 5 min. Cell suspension was centrifuged again at 6500 RPM for a period of 10 min at 4 °C, and after discarding the supernatant, the sediment was washed three times with 10 mL of PBS at 4 °C. The cells were then resuspended in 1 mL of RPMI medium with 10% BFS and cell concentration was measured by excluding 0.3% Trypan blue in a hemocytometer chamber.

The cells were then added to 96-well plates (Corning, Corning, NY, USA) at 5 × 10^5^ cells/well in 100 µL, all in triplicate. The cells were then stimulated with 100 µL of RPMI (negative control), LPS from E. coli 2 µg/mL, concanavalin A 5 µg/mL or 1 × 10^6^ CFU/mL L. monocytogenes gamma irradiated. The plates were then incubated at 37 °C with a 5% CO_2_ atmosphere for 72 h. After this stage, the plates were centrifuged at 6500 RPM for 10 min at room temperature, 100 µL culture supernatant was collected and stored in microtubes at −20 °C for verification of cytokine levels and the cells were treated for cell viability assay.

2.12. Cell Viability Assay

After collecting the supernatant, 20 μL of 5 mg/mL 3-(4,5-dimethyl-2-thiazolyl) -2,5-diphenyl-2H-tetrazolium bromide (MTT, Invitrogen, USA) was added to each well with the splenocytes suspensions. The plates were protected from light and incubated at 5% CO_2_ at 37 °C for 2 h, with 95% humidity. Next, 30 μL of SDS (sodium dodecyl sulfate) was added to the wells, followed by overnight incubation in a 5% CO_2_ at 37 °C. So, the plates were read at a wavelength of 595 nm on an ELISA MR-96A apparatus (Thermo Fisher Scientific, USA).

2.13. Measurement of Cytokines from the Culture of Splenocytes and Macrophages

The supernatant acquired during the cultivation of macrophages was evaluated for cytokine levels. IL-6 concentrations were measured in supernatants from macrophage cultures infected with B. ovis and L. monocytogenes in 96-well plates using ELISA kits (DuoSet, R&D Systems, Minneapolis, MN, USA) according to the manufacturer’s instructions. Reading was carried out at a wavelength of 490 nm using the ELISA MR-96A apparatus (Thermo Fisher Scientific, USA). Cytokine concentrations (pg/mL) were determined by comparison with an assay standard curve.

2.14. Determination of Bacterial Load in Organs

To determine the bacterial recovery of both L. monocytogenes and B. ovis in the target organs (spleen and liver) of infected mice, fragments of these organs were collected aseptically in 15 mL Falcon tubes containing 2 mL of sterile PBS, weighed and macerated using a tissue homogenizer (80 II, Eikonal, São Paulo, Brazil). The tissue macerates were serially diluted and plated on TSA with 1% hemoglobin for bacterial recovery of B. ovis and on BHI agar for bacterial recovery of L. monocytogenes. The seeded plates were kept in a humidity-controlled incubator at a temperature of 37 °C and at a constant atmosphere 5% CO_2_ and the colonies were counted after a period of 18 h for L. monocytogenes and 3 to 5 days for B. ovis.

2.15. Histopathology

Samples of the liver and spleen collected during necropsy were fixed in 10% buffered formalin solution for 24 h. Tissue samples were then dehydrated with alcohol in increasing concentrations (70%, 80%, 90%, and 100%) and then diaphanized in xylene. Subsequently, the samples were immersed in paraffin to make microtome-sectioned blocks (3 µm thick) adhered to glass slides and stained with hematoxylin and eosin (HE). Histopathological evaluation was carried out for the inflammation and necrosis lesions observed in the spleen and liver and a score was assigned, classified according to the intensity of the inflammatory lesions: 0 (zero) absence, 1 (one) mild, 2 (two) moderate and 3 (three) intense lesion and for the absence 0 (zero) or the presence of necrosis 1 (one), making a maximum total value of 4 (four) [19]. All slides were assessed blindly by two veterinary pathologists.

2.16. Immunofluorescence for Acetylated Histones

The training of innate immunity was checked by bone marrow cells using immunofluorescence of acetylated histones to identify the marking for acetylation of lysine 9 and 27 on histone 3 (H3K9ac and H3K27ac). The increased acetylation of these histones is related to the training of innate immunity induced by β-glucan [6,7].

For this purpose, female C57BL/mice were distributed into three groups containing three animals (n = 3) as follows: β-glucan, 1 dose (1 mg/100 µL i.p.); β-glucan, 2 doses (1 mg/100 µL i.p.), one dose every 7 days and mock (100 µL of PBS i.p.). Seven days after the last stimulus in each group, the mice were euthanized to obtain bone marrow cells. After euthanasia and cell collection, the cells were added to 50 mL Falcon tubes, washed with 5 mL of PBS and centrifuged at 6500 RPM for 5 min. The supernatant was then discarded, and 5 mL of lysis solution was added for 5 min, with the samples placed on ice. The samples were centrifuged again at 6500 RPM for 5 min; then the supernatant was discarded and fixation was carried out, adding 2 mL of 4% paraformaldehyde solution for 20 min. The cells were centrifuged again and the supernatant was discarded, followed by the addition of 70% alcohol and kept at −20 °C until immunofluorescence was performed.

Briefly, bone marrow cells were rehydrated and 200 µL of the cell suspension was added to glass coverslip in culture plates and permeabilized with a 0.2% solution of TritonX-100 for 1 h at 37 °C in a humid chamber. Then, the coverslips were washed in PBS with 1% polyvinylpyrrolidone (PVP), followed by the addition of 1 mL of blocking solution (0.05% TritonX-100 and 2% BSA diluted in PBS with 1% PVP) in a humid chamber under agitation for 2 h at room temperature. Next step was the addition of 1 mL per coverslip of the primary antibodies for H3K9ac (rabbit anti-H3K9ac, GeneTex, Irvine, CA, USA) and H3K27ac (rabbit anti-H3K27ac, GeneTex, USA), both at a concentration of 1:1000. Both antibodies were polyclonal, produced in rabbits with cross reactivity to several species including mice, humans, and cattle. The antibodies were incubated overnight under agitation in a humid chamber. Next, the primary antibody was washed with blocking solution and the secondary antibody (polyclonal donkey anti-rabbit 488, Thermo Scientific, USA), was added at a concentration of 1:500 for 2 h in a humid chamber under agitation; then, the coverslips were washed in PBST (1 mL per plate) and mounted together with ordinary glass slides. The images were taken on a Leica DMI6000 fluorescence microscope (Leica, Wetzlar, Germany) located at the UFABC Multiuser Experimental Center. At least 20 images were taken at different magnifications: 5 images under a 10× objective, 5 images under a 20× objective and 10 images under a 40× objective from each sample.

The images generated with 40X objective were used for analysis. Fluorescence intensity was assessed using ImageJ 1.54g. software (National Institutes of Health, Bethesda, MD, USA). A total of 100 cells were assessed per animal, with the fluorescence intensity value obtained from each area clearly identified as a nucleus (excluding overlapping areas or those that did not appear completely in the image), followed by subtraction of the fluorescence intensity of the background of the same image [25].

2.17. Statistical Analysis

Statistical analysis of the CFU count data was normalized by logarithmic transformation and submitted to analysis of variance (ANOVA). The means were then compared using the Tukey test. The non-parametric Kruskal–Wallis test was used to assess the histopathologic score. The immunofluorescence results were assessed for outliers using the Grubbs test (alpha = 0.05) and then the D’Agostino and Pearson normality tests. The means of the groups were evaluated using the non-parametric Kruskal–Wallis test. All these analyses were carried out using GraphPad Prism software version 8.0.1 (GraphPad Prism software, San Diego, CA, USA). Values were considered statistically different when p < 0.05.

3. Results

3.1. Bone-Marrow-Derived Macrophages from β-Glucan Trained Mice Control In Vitro Infection of B. ovis and L. monocytogenes

Cultured macrophages from mice stimulated with zymosan or β-glucan, known to act as “trainers” of the innate immune cells, had an enhanced ability to control intracellular growth of L. monocytogenes (Figure 1A) and B. ovis (Figure 1B). Macrophages from these mice had lower bacterial recovery of L. monocytogenes at 4 and 8 h hpi than macrophages from unstimulated mice (mock-inoculated with PBS) (p < 0.01). Reductions of 0.84 log_10_ (zymosan) and 1.15 log_10_ (β-glucan) were observed at 8 hpi.

In the case of B. ovis infection, only the macrophages of mice trained with β-glucan had a significant 2 log_10_ reduction in bacterial count compared to the mock group at 48 hpi (p < 0.001). Macrophages of mice trained with zymosan did not have a significant reduction in bacterial count. Animals that had been trained with β-glucan were evidently more effective at controlling the infection, especially when challenged with B. ovis.

These results demonstrated the ability of β-glucan to improve the ability of macrophages derived from myeloid cells to eliminate L. monocytogenes and B.ovis, even without prior contact with the agent, suggesting the induction of trained immunity in these stimulated macrophages.

3.2. Mice Stimulated with β-Glucan Show Induction of Trained Immunity Evidenced by High Acetylation in Histone 3 in Splenic Cells and Myeloid Cells

To assess innate immune cells of bone marrow of mice trained with β-glucan (a single dose of 1 mg/mouse), histone 3 acetylation profile was evaluated using immunofluorescence. Epigenetic alterations are key events in transition from the naïve to the trained state. Acetylation of lysines in histone 3 is observed in cells of the innate immune system trained with various training substances, including β-glucan [6].

Higher fluorescence intensity was observed in bone marrow cells from the trained mice, where both histone marks (Figure 2) exhibited a 7.00-fold increase for H3K9ac (Figure 2B) and a 1.66-fold increase for H3K27ac (Figure 2F). Based on these findings, we were able to confirm that training of bone marrow cells was indeed taking place in mice stimulated with a dose of β-glucan.

3.3. Trained Innate Immunity Contributed to a Partial Control of L. monocytogenes Infection in Mice Vaccinated with the Inactivated, Gamma-Irradiated L. monocytogenes Vaccine (KLM-γ-AC)

Considering the results of the in vitro experiments with L. monocytogenes, we investigated whether innate immunity training with zymosan or β-glucan enhances vaccine-induced protection against challenge with a wild-type of strain of L. monocytogenes. Experiments were performed using a vaccine model established by our research group in BALB/c mice. We have previously demonstrated that L. monocytogenes inactivated by gamma irradiation and encapsulated with alginate and chitosan (KLM-γ-AC) induces significant protection in mice against challenge with L. monocytogenes [15].

Innate immunity training, in isolation, did not influence the ability of BALB/c mice to control infection by L. monocytogenes. Previously trained mice exhibited no significant difference in CFU counts post-challenge compared to those in the infected mock group (Figure 3A,B). Although training did not statistically improve vaccine efficacy in the trained and vaccinated groups compared to the vaccinated-only group, 3/6 of mice in the trained groups exhibited bacterial recovery levels below the detection limit in spleen and liver (Figure 3A,B).

Interestingly, even though there was no statistical difference in bacterial loads between the trained and immunized and only immunized groups, trained mice had a reduction of 3.72 log_10_ of CFU, while in untrained and vaccinated mice, the reduction was only 1.99 log_10_.

In view of this result, we wondered whether the strain of mice could influence this effect, so we decided to evaluate the same experimental protocol in C57BL/6 mice. This strain is more resistant to L. monocytogenes [26]. Interestingly, when C57BL/6 mice were only β-glucan trained, they exhibited a greater ability to control the infection in spleen and liver compared to mock mice, which was not observed in Balb/c mice (Figure 3C,D). This demonstrates that β-glucan training alone can confer protection in C57BL/6 mice. Furthermore, this control occurred even after an extended time interval of more than 30 days, in contrast to in vitro training where the interval between training and infection was one week. These results reaffirm that training with these substances, especially β-glucan, improves the innate immune response’s ability to control pathogens non-specifically.

Although training did not improve the efficacy of the vaccine in the trained and vaccinated groups compared to the vaccinated-only group, at least 4/6 and 3/6 of C57BL/6 mice in the trained groups had bacterial loads below the detection limit in the spleen and liver, respectively (Figure 3C,D), similar to that observed in Balb/c mice (Figure 3A,B). These results confirm the dichotomous effect of training in conjunction with vaccination, providing effective protection for 50% of the animals in these groups. Furthermore, mice that received training, primarily with β-glucan, prior to vaccination exhibited a reduction of over 4.5 log_10_ CFU and mice that were solely vaccinated demonstrated a reduction of 3 log_10_ CFU.

3.4. Training Innate Immunity with β-Glucan Partially Altered the Profile of the Immune Response Induced by the KLM-γ-AC Vaccine

Initially, we demonstrated that β-glucan-trained immunity improves the ability of bone-marrow-derived macrophages to control L. monocytogenes infection in vitro (Figure 1A). Therefore, we decided to evaluate whether the microbicidal capacity of macrophages was improved by association of training and vaccination.

C57BL/6 mice were assigned to the KLM-γ-AC, βG+KLM-γ-AC, β-glucan, or mock groups. Two-week post immunization, bone marrow was collected for macrophage differentiation. The ability to control L. monocytogenes infection in vitro was assessed in macrophages (Figure 4A). The level of the cytokine IL-6 in the culture supernatant was measured (Figure 4B).

We observed that infection with L. monocytogenes in macrophages from mice that were trained or vaccinated, or both, were able to significantly reduce the intracellular survival of the bacteria at 4 and 8 hpi compared to naïve (mock) macrophages (p < 0.001). Reductions of approximately 2 log_10_ were observed at 4 and 8 hpi. However, training did not result in a greater reduction in intracellular bacterial load in trained–vaccinated mice compared to those that were only vaccinated (Figure 4A). These results suggest that combining training with vaccination does not improve the microbicidal capacity of bone-marrow-derived macrophages in vitro.

We demonstrated level of IL-6 in the in vitro pooled macrophage culture supernatant. This cytokine is considered a marker of trained immunity in mice stimulated with β-glucan [27,28]. Macrophages from mock mice produced 7 pg/mL, from vaccinated mice produced 112 pg/mL, and from mice trained with β-glucan produced 206 pg/mL, while those from trained and vaccinated mice produced 509 pg/mL (Figure 4B). When evaluating IL-6 production by L. monocytogenes 8 hpi, the three groups exhibited comparable cytokine levels reaching approximately 700 pg/mL, the assay’s maximum detection (Figure 4B). Interestingly, these three groups exhibited similar control of infection (Figure 4A).

Recent studies have demonstrated the protective effect and immune response profile of the KLM-γ-AC vaccine candidate, as evidenced by splenocyte proliferation after in vitro stimulation and IFNγ production [15]. We therefore evaluated the immune response profile of splenocytes from mice that were first given a dose of β-glucan and then immunized with KLM-γ-AC. C57BL/6 mice were assigned to KLM-γ-AC, βG+KLM-γ-AC, β-glucan, or mock groups. Two-week post immunization, splenocytes were obtained for the cell viability assay and IFNγ was measured in the culture supernatant (Figure 4C,D).

Stimulation with KLM-γ revealed that only splenocytes from the vaccinated groups (KLM-γ-AC or βG+KLM-γ-AC) exhibited greater viability compared to their non-stimulated counterparts. Additionally, splenocytes viability was significantly higher in the vaccinated groups than in the mock group (p < 0.05), suggesting that vaccination favors cell viability or induces splenocytes proliferation. However, there was no difference in viability rates between the KLM-γ-AC and βG+KLM-γ-AC groups, suggesting that β-glucan training did not improve vaccine-induced immune response (Figure 4C). The βG+KLM-γ-AC group showed higher IFNγ level when compared to other groups (p < 0.01, Figure 4D). This result indicated the better control of infection by βG+KLM-γ-AC may be related to IFNγ production.

3.5. Training Innate Immunity Did Not Improve Control of B. ovis Infection in Mice Vaccinated with the Attenuated B. ovis ∆abcBA-AC

We demonstrated that bone-marrow-derived macrophages from mice that had been trained with β-glucan could control B. ovis infection (Figure 1B). Considering that trained immunity by β-glucan could partially enhance the protective effects of the inactivated L. monocytogenes vaccine (Figure 4), we investigated if trained immunity could enhance the protective effects of an attenuated vaccine model (B. ovis ∆abcBA-AC) [18,19,29]. In this experiment, a significant reduction in bacterial recovery was observed in the spleen (Figure 5A) and liver (Figure 5B) of the group only vaccinated compared to the infected mock group (p < 0.001). These results confirmed the efficacy of this vaccine, which has been previously demonstrated [18,19,29]. In groups trained with β-glucan or zymosan, no difference was observed in bacterial loads in the spleen or liver compared to the infected mock group (Figure 5). This suggests that in vivo training under these conditions does not favor control of B. ovis infection. In groups that were vaccinated after training, a reduction in bacterial load was observed in the spleen and liver only in the group stimulated with β-glucan (p < 0.001), with bacterial recovery like that observed in the only vaccinated group. However, with zymosan training, no reduction in bacterial load was observed in the group that was vaccinated and trained with zymosan (Figure 5). These results suggest that stimulation with β-glucan does not affect the efficacy of the vaccine, whereas stimulation with zymosan impairs the performance of the attenuated B. ovis vaccine.

3.6. Training with β-Glucan Associated with Immunization with Gamma-Irradiated Inactivated B. ovis (BO-γ-AC) Protects Against B. ovis

Since induction of trained immunity impairs the protective effects of vaccination with the attenuated B. ovis ∆abcBA strain, we hypothesized that a trained innate immune system could more efficiently eliminate the vaccine strain more quickly, thereby preventing a vaccine-induced response that requires multiplication of the vaccine strain in the host. We therefore investigated whether vaccination with inactivated B. ovis, which is not protective in the absence of trained innate immunity [20], could provide protection in our vaccine model.

Vaccination with inactivated B. ovis (BO-γ-AC) did not reduce bacterial loads in the spleen and liver, confirming that it does not protect against B. ovis infection in a murine model, as previously demonstrated by Silva et al. (2024) [20]. Interestingly, mice trained with β-glucan and vaccinated with BO-γ-AC had significantly lower bacterial loads in their spleens than infected mock animals and mice that were only vaccinated (p < 0.01, Figure 6). Importantly, the dichotomous profile of bacterial recovery in the spleen of mice that were trained with β-glucan and vaccinated was like the experiment with the inactivated KLM-γ-AC vaccine (Figure 3), with at least 33% of the mice showing bacterial recovery below the detection limit (Figure 6). Mice trained and vaccinated with BO-γ-AC had an index protection of 2.91 log_10_ CFU.

Liver from the mock infected mice had microscopic multifocal areas of intense inflammatory infiltrate composed of lymphocytes, macrophages and neutrophils, associated with random multifocal distribution necrosis (Figure 7A–D). Mice from β-glucan, BO-γ-AC groups had similar liver lesions to infected mock group. Only mice in the βG+BO-γ-AC group had a significant reduction in the intensity of the lesions present in the liver (Figure 7E) (p < 0.05).

4. Discussion

This work demonstrates for the first time that the use of innate immunity trainers, especially β-glucan, when combined with inactivated vaccines, but not with an attenuated experimental vaccine, enhances protection against L. monocytogenes or B. ovis in a murine model. It became evident that several factors influence the induction of trained immunity and its contribution to vaccine efficacy against intracellular pathogens. These include the host immune background (e.g., mouse strain), the training protocol (type of substance), and the vaccine platform employed (live attenuated or inactivated).

Protection induced by inactivated Brucella spp. strains remains questionable [20,30], since they elicit humoral responses capable of driving immunity toward a Th1 profile yet fail to provide effective protection [30]. However, when innate immune training was combined with inactivated vaccine preparations to control B. ovis infection, significant protection was achieved. β-glucan-mediated training associated with the inactivated B. ovis vaccine resulted in marked post-infection protection. In listeriosis model, β-glucan co-administration with inactivated L. monocytogenes yielded ~50% improvement in protection. The findings presented here support the hypothesis that immunomodulatory agents capable of inducing trained immunity can enhance adaptive responses to inactivated vaccines. However, they appear to provide no comparable benefit when the immunogen is a live vaccine. A plausible explanation is that trained macrophages may eliminate live vaccine strains more rapidly, reducing their ability to replicate and stimulate protective immunity. Macrophage training is known to increase basal inflammatory cytokines such as IL-1β, TNF-α, and IL-6; to enhance phagocytic and microbicidal activity; and to upregulate activation markers like CD80, CD86, and MHC-II [1,2,31].

A single administration of β-glucan was sufficient to induce trained innate immunity, as demonstrated by epigenetic modifications in histone 3 (H3K9ac and H3K27ac) and increased IL-6 production by bone-marrow-derived macrophages. Epigenetic alterations are considered a hallmark of trained immunity [8,31]. Changes in H3K9ac and H3K27ac have been reported following exposure to β-glucan [6], zymosan [5], C. albicans [4], and BCG [17]. These alterations were confirmed in bone marrow cells [32]. Protective effect of trained immunity, observed both in vitro and in vivo infections, has been linked to epigenetic remodeling [3,5,8,33] and metabolic reprogramming, including increased anaerobic glycolysis, enhanced reactive oxygen species production, and elevated synthesis of pro-inflammatory cytokines [34,35].

Zymosan represents the soluble, low-molecular-weight β-glucan fraction from Saccharomyces cerevisiae, whereas insoluble β-glucans have higher molecular weight and contain both β-1,3 and β-1,6 linkages, features that make β-glucan a more potent training agent due to its greater β-1,6 density and prolonged receptor engagement after phagocytosis [36,37]. Acting as pathogen-associated molecular patterns (PAMPs), β-glucans activate innate immune responses through pattern recognition receptors, and high-molecular-weight forms provide sustained stimulation as they degrade gradually. β-glucans containing both β-(1,3) and β-(1,6) linkages exhibit higher affinity for dectin-1, triggering downstream signaling that enhances immune activation, and they also interact with TLR-2, TLR-4, CR3, scavenger receptors, and lactosylceramide, contributing to their broad immunostimulatory effects [8,33,38,39,40,41].

In this study, the combination of an innate immunity trainer β-glucan with an inactivated vaccine improved control of intracellular pathogens. The OVA vaccine model shows that β-glucan-induced training enhances antigen presentation by macrophages and promotes stronger activation of antigen-specific T cells. Similarly, β-glucan-trained mice vaccinated with influenza hemagglutinin A exhibit increased long-term IgG production [42], supporting the idea that trained immunity can potentiate vaccine efficacy.

β-glucan-induced training did not enhance splenocyte viability and proliferation. Increase in viability occurred only as a direct vaccine-driven adaptive response [43]. Zymosan has been shown to influence splenocyte proliferation and viability, but only at low concentrations (1 µg/mL), with higher doses being detrimental [27]. Interferon production was significantly higher in trained-and-vaccinated animals, indicating that innate immune training can modulate adaptive responses. In the listeriosis model, IFN-γ produced by NK and T cells is essential for infection control [44]. β-glucan-trained macrophages increase IFN-γ expression when co-cultured with T cells, promoting a Th1 response [44]. IFN-γ also plays a central role in antitumoral trained immunity [45].

C57BL/6 mice trained with zymosan or β-glucan exhibited superior control of Listeria infection compared to Balb/c mice. Resistance in C57BL/6 mice relies on robust IFN-γ production by NK and T cells [24,46]. Zymosan-induced protection in C57BL/6 mice has been associated with enhanced hematopoietic proliferation and increased IL-1β levels [46]. Recent findings by Isachesku et al. (2025) show that individuals with higher baseline IFN-γ production mount stronger trained immunity responses after BCG vaccination, suggesting that IFN-γ not only shapes Th1 responses but also influences the epigenetic reprogramming of innate cells [47].

A noteworthy aspect of the present work is the dichotomous effect observed when training was combined with the inactivated vaccine, resulting in effective protection in only 50% of the animals. Sex-based differences in trained immunity have been reported, with females exhibiting stronger responses due to higher estradiol levels. Estradiol enhances trained immunity by activating the mTOR pathway in macrophages and improving their metabolic and phagocytic functions and promotes vaccine-induced immunoglobulin production [48]. All animals used here were female, but estrous cycle phase was not determined, and its influence on trained immunity remains unknown. These findings highlight the complexity of inducing protective trained immunity and suggest that optimal conditions must be defined to enhance vaccine responses against intracellular pathogens.

5. Conclusions

In conclusion, trained innate immunity can enhance the protection provided by inactivated experimental vaccines against B. ovis and L. monocytogenes, whereas it may impair the efficacy of live attenuated vaccines. The evidence presented here opens new possibilities for applying similar approaches to other inactivated vaccines, particularly in natural hosts.