RPMI culture medium Dutch modified (Invitrogen) supplemented with 50 μg/mL gentamicin, 2 mM Glutamax (Gibco), and 1 mM pyruvate (Gibco) was used for human in vitro experiments. RPMI 1640 medium (Lonza) supplemented (cRPMI) with 10 % heat‐inactivated fetal bovine serum (Gibco), 100 μg/mL normocin (In vivo Gen), 50 μg/mL penicillin-streptomycin, 1% nonessential amino acids, 1% MEM vitamins and 1 mM sodium pyruvate (all from Life Technologies) was used for animal experiments. V132 composed of whole heat-inactivated C. albicans, MV140 (Uromune) composed of whole heat-inactivated bacteria (25% E. coli, 25% P. vulgaris, 25% K. pneumoniae and 25% E. faecalis), MV140/V132 (combination of MV140 and V132) and control excipients (glycerol, saline, pineapple flavor), identical for MV140 or V132, were provided by Inmunotek S.L. Lipopolysaccharide (LPS) from E. coli O55:B5 (Sigma-Aldrich) was used.

Buffy coats from healthy donors were obtained after written informed consent (Sanquin Blood Bank, Nijmegen, the Netherlands). Peripheral blood mononuclear cells (PBMC) isolation was performed by using Ficoll-Paque density gradient media (GE Healthcare). Cells were washed twice in PBS, re-suspended in RPMI culture medium and counted. Monocyte isolation was performed by hyper-osmotic Percoll (Sigma) density gradient centrifugation (580g, 15 min). The interphase layer was collected, and cells were washed with cold PBS, re-suspended in RPMI culture medium Dutch modified and counted. An extra purification step was added by adhering Percoll-isolated monocytes or Ficoll-isolated PBMC to a flat-bottom 96-well plates for 1 h at 37°C. Subsequently, monocytes were washed with warm PBS to yield maximal purity.

Trained immunity experiments were performed as previously described (42). 100 μL monocytes at 1 x 10⁶ cells/mL or PBMC at 5 x 10⁶ cells/mL density were added to a flat-bottom 96-well plates. Cells were incubated with control excipients (containing all excipients except the microbial components), V132 (3 FTU, Formazine Turbidity Units, per mL, as turbidity measurement for the monitoring of fungal suspensions), MV140 (10⁷ bact. per mL) or MV140/V132 (MV140, 10⁷ bact. per mL, and V132, 3 FTU per mL, as turbidity measurement for the monitoring of fungal suspensions) for 24 h. Supernatants from monocytes were collected 24 h after stimulation. Cells were washed once with 200 μL of warm PBS and incubated for 6 days in RPMI culture medium Dutch modified with 10% human pooled serum. On day 6, supernatants were collected for lactate measurement. Cells were re-stimulated for 24 h with 10 ng/mL LPS or 10⁷ bact. per mL MV140 to assess the responsiveness of the cells to heterologous secondary stimulation, after which supernatants were collected.

Cytokine production was quantified in supernatants using commercial ELISA kits for human IL-6, TNFα, IL-1β, IL-10, IL-1RA (R&D Systems); mouse IL-6, IFNγ, (BD Biosciences); TNFα, IL-10 (Invitrogen); and IL-17A (R&D Systems), following manufacturer’s instructions.

Lactate concentration was measured from supernatants using a Lactate Fluorometric Assay Kit. Briefly, lactate was oxidized resulting in H₂O₂ which was coupled to the conversion of Amplex Red reagent to fluorescent resorufin by horseradish peroxidase (HRP).

Cell viability was analyzed using CytoTox 96 Non-Radioactive Cytotoxicity Assay (Promega), following manufacturer’s recommendations. Briefly, released lactate dehydrogenase (LDH) upon cell lysis is measured with a 30-minute coupled enzymatic assay, which results in conversion of a tetrazolium salt (INT) into a red formazan product. The amount of color formed is proportional to the number of lysed cells.

Adherent monocytes were obtained by incubating 5 x 10⁷ PBMC in 10 cm Petri dishes for 1 h. The adherent monocytes were treated with 10 mL of RPMI culture medium Dutch modified containing control excipients, V132, MV140 or MV140/V132 for 24 h, washed once with warm PBS and incubated in RPMI culture medium Dutch modified with 10% human pooled serum at 37°C, 5% CO₂. Following 6 days in culture, cells were detached with Versene solution (ThermoFisher Scientific) and 5 x 10⁵ cells were plated to overnight-calibrated cartridges in DMEM medium supplemented with 1 mM glutamine (pH adjusted to 7.4) and incubated for 1 h in a non-CO₂-corrected incubator at 37°C. Extracellular acidification rate (ECAR) and oxygen consumption rate (OCR) in response to 10 mM glucose, 1 μM oligomycin, 1 μM carbonyl cyanide 4-(trifluoromethoxy) phenylhydrazone (FCCP), and 1.25 μM rotenone and 1.5 μM antimycin A injections were measured using a Seahorse XFe96 Analyzer (Agilent). A complete ECAR study was performed in two consecutive stages: basal extracellular acidification (after glucose injection), and mitochondrial complex V inhibition (oligomycin). A complete OCR study was performed in four consecutive stages: basal respiration (without drugs), mitochondrial complex V inhibition (oligomycin), maximal respiration induction (FCCP), and electron transportation chain inhibition (rotenone and antimycin A).

Adherent monocytes were obtained by incubating 5 x 10⁷ PBMC in 10 cm Petri dishes for 1 h. The adherent monocytes were treated with 10 mL of RPMI culture medium Dutch modified containing control excipients, V132, MV140 or MV140/V132 for 24 h, washed once with warm PBS and incubated in RPMI culture medium Dutch modified with 10% human pooled serum at 37°C, 5% CO₂. Following 6 days in culture, cells were detached with Versene solution (ThermoFisher Scientific) and fixed in 1% methanol-free formaldehyde. Afterward, cells were sonicated using a Diagenode Bioruptor UCD-300 for 10 min (30 seconds on; 30 seconds off). 33 μL of the solution containing chromatin was incubated with 255 μL of dilution buffer, 12 μL of protease inhibitor cocktail and 1 μg of H3K27ac or H3K9me3 antibodies (Diagenode), overnight at 4°C with rotation. Protein A/G magnetic beads were washed in dilution buffer with 0.15% SDS and 0.1% BSA, added to the chromatin/antibody mix and rotated for 60 min at 4°C. Beads were washed with 500 μL buffer for 5 min at 4°C with five rounds of washes. After washing chromatin was eluted using elution buffer for 20 min. Supernatant was collected, 8 μL 5 M NaCl and 1 μL proteinase K were added, and samples were de-crosslinked at 64°C for 4 h. DNA was isolated with the MinElute PCR Purification Kit (QIAGEN), following manufacturer’s protocol. Real-time quantitative PCR was performed using the SYBR Green method. Samples were analyzed by a comparative Ct method according to the manufacturer’s instructions. The sequences of the used pair primers were: IL6-1 (forward, TCGTGCATGACTTCAGCTTT; reverse GCGCTAAGAAGCAGAACCAC), IL6-2 (forward, AGGGAGAGCCAGAACACAGA; reverse GAGTTTCCTCTGACTCCATCG), TNFA1 (forward, AGAGGACCAGCTAAGAGGGA; reverse AGCTTGTCAGGGGATGTGG), and TNFA2 (forward, GTGCTTGTTCCTCAGCCTCT; reverse ATCACTCCAAAGTGCAGCAG).

All mice procedures included in this study were reviewed and ethically approved by Universidad Complutense de Madrid (UCM) and Comunidad Autónoma de Madrid (CAM) within the context of project SAF-2017-84978-R, (CAM: ref.10/250312.9/18). BALB/c mice (female, 6 weeks old, Charles River) were intraperitoneally immunized with 20 μL of control excipients or V132 (300 FTU per mL) at day 0 and 3 and sacrificed 4 days after the last immunization. In another animal model, BALB/c mice (6 weeks old) were sublingually immunized for 5 consecutive days during 2 weeks with 20 μL of control excipients or V132 (300 FTU per mL) and sacrificed 3 days after last immunization. Moreover, some immunized mice were sublingually administered with 20 μL of MV140 (10⁹ bact. per mL) 3 days after last sensitization and sacrificed after 8 days of the MV140 administration. Peritoneal cells and spleens were collected.

Peritoneal cells were collected with PBS/EDTA 2mM and resuspended in cRPMI. Duplicates of 0.25 x 10⁶ peritoneal cells were cultured in vitro with control excipients or MV140 (10⁶ bact. per mL) in flat-bottom 96-well plates for 24 h at 37°C and 5% CO₂. Cell-free supernatants of duplicates were pooled and used to quantify IL-6 and TNFα.

Spleens were minced and filtered through 40 μm nylon cell strainers to obtain a single-cell suspension. Then, red blood cells were lysed with ACK lysis buffer before being resuspended in cRPMI. Splenocyte viability was assessed using Trypan Blue exclusion. Triplicates of 0.8 x 10⁶ splenocytes were cultured in vitro with control excipients or MV140 (10⁵-10⁷ bact. per mL) in flat-bottom 96-well plates. After 24 or 72 h of culture at 37°C and 5% CO₂, the triplicates were pooled and cell-free supernatants were used to quantify IL-6, TNFα, IL-17, IFNγ and IL-10 by ELISA.

All the data are expressed as means ± s.e.m. of the indicated parameters. One-way ANOVA, Wilcoxon test or Unpaired Student t test for statistical analysis were performed using GraphPad Prism software, version 6.0. Significance was defined as *P < 0.05, **P < 0.01, ***P < 0.001, and ****P < 0.0001.

To assess whether priming of human peripheral blood mononuclear cells (PBMC) with V132, MV140, or a simultaneous combination of V132 and MV140 would increase PBMC’s responses upon second stimulation, an in vitro model for the induction of trained immunity was employed (43, 44). As shown in Figure 1A , PBMC were initially treated with control excipients (ctrl), V132 alone, MV140 alone, and the combination of MV140 and V132 (MV140/V132) and the cytokine signature after 24 h of stimulation quantified. Next, cells were washed, left resting for 6 days with medium, and then re-stimulated with LPS to assess cytokine production after 24 h ( Figure 1A ). PBMC stimulation with V132 alone induced significant IL-6 and anti-inflammatory IL-1RA production ( Figure 1B ). Exposure to MV140 and MV140/V132 significantly increased the production of IL-6, TNFα and IL-1β compared to control excipients or V132 ( Figure 1B ), without affecting cell viability ( Supplementary Figure 1A ). PBMC stimulated with MV140 alone produced significantly increased IL-1β compared to the MV140/V132 combination ( Figure 1B ). Although overall IL-10 production was higher after stimulation with MV140 than with control, V132 or MV140/V132, significant differences were not detected ( Figure 1B ). MV140 or MV140/V132-stimulated PBMC also significantly produced higher IL-1RA levels after 24 h, with higher levels observed after MV140 or MV140/V132 than V132 stimulation ( Figure 1B ). Consequently, the IL-1β/IL-1RA ratio was significantly higher when PBMC were stimulated with MV140 or MV140/V132 than control excipients or V132, and slight but higher with MV140 than with MV140/V132 ( Figure 1C ). PBMC conditioned with V132 displayed a trained response on day 7 after re-stimulation with LPS resulting in a significant IL-6 and TNFα production than control excipients ( Figure 1D ). This was not the case when PBMC were stimulated with MV140, and remarkably not with V132 in the presence of MV140 (MV140/V132) ( Figure 1D ). No IL-10 production was observed on day 7 in these cell cultures (data not shown). Collectively, these results suggest that in the assayed conditions, V132 but not MV140 or MV140/V132 induced trained immunity in human PBMC ( Figure 1D ).

Trained cells undergo strong metabolic and epigenetic reprogramming, two major molecular mechanisms underlying trained immunity (26). Therefore, we first studied the metabolic changes imprinted by V132 in PBMCs compared to those induced by control excipients, MV140 or MV140/V132. PBMC treated with V132, MV140 or MV140/V132 produced significantly higher lactate concentrations in cell culture supernatants after 6 days of resting (prior to re-stimulation) compared to control cultures with excipients, reflecting an increase in the activity of glycolysis ( Figure 2A ). Remarkably, the levels of lactate produced by V132-trained PBMC were significantly higher than MV140- or MV140/V132-treated cells ( Figure 2A ). To further verify these results, we performed functional metabolic experiments using a Seahorse bioanalyzer to monitor real-time extracellular acidification rate (ECAR) and mitochondrial oxygen consumption rate (OCR) after 6 days of resting of the initially exposed PBMC. V132-trained PBMC showed an increased basal and maximal ECAR, as well as glycolytic reserve compared to PBMC stimulated with MV140, MV140/V132 or control excipients ( Figure 2B ). Only a slight but significant increase in basal and maximal ECAR was observed after the treatment with MV140/V132 compared to MV140 or control excipients ( Figure 2B ). In addition, basal and maximal respiration, ATP production coupled to respiration, and spare respiratory capacity were significantly higher after V132 training than the other assayed stimuli, also supporting the induction of mitochondrial oxidative phosphorylation in V132-trained PBMC ( Figure 2C ).

To determine whether the enhanced production of pro-inflammatory cytokines after the second re-stimulation with LPS might be linked to V132-induced epigenetic changes, we studied chromatin status near the IL6 and TNFA genes by quantifying H3K27ac and H3K9me3 histone marks by chromatin immunoprecipitation (ChIP) analysis. H3K27ac is previously described as the histone mark associated with active chromatin and increased gene transcription in trained immunity (45). In contrast, H3K9me3 is considered a repressor mark (46). V132-trained PBMC display an increase of H3K27ac and a decrease of H3K9me3 at the promoters of IL6 and TNFA ( Figure 2D, E , respectively). Collectively, these results demonstrate that the induction of trained immunity with V132 in PBMC is mediated by metabolic and epigenetic reprogramming.

To confirm that V132 induces trained in purified human monocytes, the same in vitro trained immunity protocol described above was performed using human monocytes isolated from healthy donors ( Figure 3A ). Initial exposure of monocytes to V132 alone did not induce pro-inflammatory cytokines, in contrast to the IL-6, IL-1β and TNFα production when stimulated with MV140 or MV140/V132 ( Figure 3B ). Cell viability was not affected in these cultures ( Supplementary Figure 1B ). Interestingly, monocytes simultaneously activated with MV140/V132 displayed significantly higher levels of TNFα and lower IL-1β than MV140 alone ( Figure 3B ). Stimulation of monocytes with the indicated conditions increased anti-inflammatory cytokine IL-1RA production, which was significantly higher upon V132 treatment than MV140 or MV140/V132 ( Figure 3B ), without IL-10 production (data not shown). The IL-1β/IL-1RA ratio was significantly higher when monocytes were treated with MV140 or MV140/V132 than with V132 ( Figure 3C ). MV140-treated monocytes showed a higher IL-1β/IL-1RA ratio than cells treated with MV140/V132 ( Figure 3C ). Monocytes treated with V132, MV140 or MV140/V132 significantly increased the production of lactate after 6 days of resting, reflecting an increased glycolysis metabolism ( Figure 3D ). Supporting our previous data with PBMC, monocytes trained with V132 produced higher amounts of lactate compared to the other assayed stimuli, without significant differences observed in this case ( Figure 3D ). V132-trained monocytes produced significantly higher amounts of IL-6 and TNFα upon re-stimulation 7 days after first priming and resting with LPS, without IL-10 induction ( Figure 3E ). Collectively, these data indicate that in the assayed conditions V132 but not MV140 or MV140/V132 induced trained immunity also in purified human monocytes.

Next, we wanted to assess whether V132-trained monocytes could display an enhanced response to MV140 used as secondary stimulus ( Figure 4A ). V132-trained monocytes treated with MV140 produced significantly higher concentration of IL-6 and TNFα compared to control monocytes ( Figure 4B ). The production of pro-inflammatory cytokines by V132-trained monocytes was higher upon secondary stimulation with MV140 than LPS ( Figure 4B ). Collectively, these data indicate that exposure of monocytes to V132 significantly enhances their capacity to mount potent pro-inflammatory responses upon exposure to MV140.

To assess the in vivo relevance of V132-induced trained immunity, BALB/c mice were immunized by intraperitoneally administration of control excipients or V132 as shown in Figure 5A . Intraperitoneal immunization with V132 significantly increased the accumulation of infiltrating cells into the peritoneal cavity compared to controls ( Figure 5B ). Peritoneal cells and splenocytes from both V132- and control mice did respond to MV140 in vitro by producing IL-6 and TNFα ( Figure 5C, D , respectively). However, cells derived from V132-immunized mice produced significantly higher levels of IL-6 and TNFα than those from control mice ( Figure 5C, D ). To further confirm the results when V132 was administered through the sublingual route, BALB/c mice were immunized as shown in Figure 5E . Splenocytes from both V132-immunized and control mice produced IL-6 and TNFα in vitro upon MV140 stimulation ( Figure 5F ). Interestingly, the production of TNFα, but not of IL-6, were higher in mice immunized sublingually with V132 ( Figure 5F ).

To assess potential effects of V132-induced training in vivo on adaptive immune responses, BALB/c mice immunized sublingually with control excipients or V132 were treated in vivo with MV140 as shown in Figure 6A . Sublingual immunization with V132 and subsequent sublingual treatment with MV140 significantly increased the cell infiltration into the peritoneal cavity ( Figure 6B ). Peritoneal cells from both mice groups activated in vitro with MV140 produced significantly higher concentrations of IL-6 and TNFα than those without MV140 stimulation ( Figure 6C ). The levels of TNFα, but not of IL-6, produced by peritoneal cells from V132-sublingually immunized mice were slightly and significantly higher than those induced by peritoneal cells from control mice after in vitro stimulation with MV140 ( Figure 6C ). To assess whether V132 enhances specific T cell responses to MV140, splenocytes from immunized mice ( Figure 6A ) were stimulated in vitro with or without MV140 ( Figure 6D ). As shown, mice both immunized sublingually with V132 or control mice produced a significant IL-17A, IFNγ and IL-10 in vitro response to MV140 stimulation ( Figure 6D ). Remarkably, the levels of IL-17A, but not IFNγ or IL-10, produced by splenocytes from V132-immunized mice after in vitro MV140 stimulation were significantly higher than those induced by splenocytes from control mice ( Figure 6D ), indicating that sublingual V132 vaccination potentiates Th17 adaptive immune responses in mice against MV140.