Male C57Bl/6 (WT), MyD88^-/-, TRIF^-/-, TLR4^-/-, TLR2^-/-, IFN-γ-/-, and NOD-scid IL2Rγnull (NSG) mice, aged 6 to 8 weeks and kept in standard pathogen-free conditions, were acquired from the Center for Development of Experimental Models (CEDEME) and the Institute of Pharmacology and Molecular Biology (INFAR) at the Federal University of São Paulo (UNIFESP). The Animal Experimentation Ethics Committee (CEUA) at UNIFESP approved all conducted experiments (Protocol numbers 1882220815 and 5635020419).

The cell line B16F10 was isolated from a spontaneous murine melanoma in C57Bl/6 mice (35) and obtained from Ludwig Cancer Research in São Paulo, Brazil. The B16F10-Nex2 subline was established at the Experimental Oncology Unit (UNONEX, at UNIFESP) and maintains the same metastatic and aggressive features as the original line (36). The cells were cultured in RPMI 1640 medium (Gibco) supplemented with 10% heat-inactivated fetal calf serum (FCS, Life Technologies), 10 mM HEPES (4-(2-hydroxyethyl)-1-piperazineethanesulfonic acid; Sigma Aldrich), 24 mM NaHCO₃ (Sigma Aldrich), and 40 mg/mL gentamicin (Hipolabor Farmacêutica, MG, Brazil). Cultures were incubated at 37 °C with 5% CO₂. IFN-γ was purchased from Peprotech, while lipopolysaccharides (LPS) and polymyxin B were obtained from Sigma Aldrich. The TLR-4 inhibitors (CLI-095), MyD88 inhibitor (Pepinh-MyD88), TRIF inhibitor (Pepinh-TRIF), and peptide control (Pepinh-Control), as well as SP600125 (JNK inhibitor), SB203580 (p38 inhibitor), and PD98059 (MAP/ERK inhibitor), were sourced from Invivogen, CA, USA. Peptides PAR1-AP (RLLFT-NH2) and PAR2-AP (LRGILS-NH2) were synthesized as previously described (37). The antibiotics (kanamycin and chloramphenicol) and IPTG (isopropyl β-D-thiogalactoside) were obtained from Sigma-Aldrich.

The expression vector pET-28 OpdA was constructed by inserting the OpdA gene into the pET28 plasmid, as previously described (32). Briefly, the OpdA gene was cloned via PCR from the total DNA of the E. coli strain DH5α using NcoI and XhoI restriction enzymes, resulting in a final construct with a poly-histidine tag at the C-terminus. Bacterial DNA was extracted using the PureLink Genomic DNA Purification Kit (Invitrogen). The PCR mixture contained 200 µM deoxynucleoside triphosphates, 2 mM MgCl₂, 50 mM KCl, and 20 mM Tris-HCl (pH 8.4), 1 U pfu DNA polymerase, and 50 pmol of the primers (5’-GATAATCCATGGGCTCTAAAATTCTCCCGGAACATGTCG-3’ and 5’-ACCACCCTCGAGGCCCTTAATGCCGTAATGCTCCAGC-3’). The PCR reaction started with one cycle at 95 °C for 3 minutes, followed by 35 cycles at 95 °C for 1 minute, 72 °C for 2.5 minutes, and ended with 72 °C for 5 minutes. The PCR product was analyzed on a 1% agarose gel containing 1.0 µM ethidium bromide in TAE buffer (40 mM Tris-acetate and 1 mM EDTA, pH 8.0), purified using the Wizard SV Gel kit (Promega), and cloned into the pGEM-T Easy vector (Promega, Madison, USA). The cloned OpdA gene was excised from the plasmid by digestion with NcoI and XhoI (Invitrogen) and inserted into the EcoRI site of the pET28 vector, then sequenced.

For expression, E. coli BL21 (DE3) pLysS was transformed via heat shock with the expression vector pET-28 OpdA and grown under agitation at 37°C for 16 hours in Luria-Bertani medium supplemented with 50 µg/mL kanamycin and 50 µg/mL chloramphenicol. The bacteria pLysS/pET28-OpdA were re-inoculated in fresh medium containing the same antibiotic selection and grown until the culture density reached an A600 of 0.6. Subsequently, the expression of the OpdA recombinant was induced with 1 mM IPTG for 12 hours at 20°C. The bacterial cultures were centrifuged at 8,000 rpm for 20 minutes at 4°C, re-suspended in 20 mM Na₂HPO₄, 300 mM NaCl, and 20 mM imidazole, pH 8.0, and lysed by pressure in a French Pressure Cell Press (model FA-078, Thermo Spectronic). The suspension was incubated for 20 minutes under agitation with 20 mM MgSO₄ and barnase (2.4 units/mL). The solution was centrifuged to remove the bacterial debris for 15 minutes at 14,000 rpm at 4°C, and the supernatant was filtered through a 0.45 µm filter. The solution was loaded into a Ni-Sepharose column (HisTrapTM HP Columns – GE Healthcare) coupled with the ÄKTA purifier System (GE Healthcare). The column was washed with a buffer containing 20 mM Na₂HPO₄ and 300 mM NaCl, and then OpdA was eluted using segmented step elution with 50 mM and 100 mM imidazole. The recombinant protein was desalted using a PD-10 column (Amersham Pharmacia Biotech) and analyzed by SDS-PAGE electrophoresis (10%), stained with silver, and further assessed by western blotting ( Supplementary Figure 1A ).

The hydrolysis of the fluorogenic peptidyl substrate Abz-GFSIFRQ-EDDna was measured using a HITACHI F-2000 spectrofluorometer with excitation and emission wavelengths of 320 nm and 420 nm, respectively, at 37°C in a 50 µM Tris-HCl buffer, pH 7.4 (32). The substrate solution was kept in a thermostatically controlled chamber for 5 minutes before adding 20 µM recombinant OpdA. The slope was converted to moles of hydrolyzed substrate per minute based on the fluorescence curves of standard peptidase solutions before and after complete hydrolysis. The data were used to plot the kinetic curve of OpdA enzymatic activity ( Supplementary Figure 1E ). OpdA was heated for 45 minutes at 65 °C to achieve heat inactivation. The structural integrity of the heat-inactivated protease was confirmed by SDS-PAGE electrophoresis (10%), stained with silver ( Supplementary Figure 1B ). The loss of proteolytic activity was demonstrated by the absence of activity on the fluorogenic substrate Abz–GFSIFRQ–EDDnp under the same conditions described ( Supplementary Figure 1D ).

OpdA was separated using SDS-PAGE and then electro-transferred onto a nitrocellulose membrane. The membrane was incubated for 16 hours at 4°C with in-house prepared murine anti-OpdA primary antibodies. After this, it was treated with a peroxidase-conjugated anti-mouse IgG secondary antibody for 90 minutes at 20°C. Finally, the blot was incubated with SuperSignal Western Blot Enhancer (Thermo Scientific) and visualized using the UVITEC System (UVITEC Cambridge) ( Supplementary Figure 1A ).

OpdA (100 µg/mL) was digested with 50 µg/mL proteinase K (Sigma Aldrich) and 2.5 mM CaCl₂ in RPMI 1640 medium. The sample was incubated for 45 minutes at 37°C with agitation, and the digestion was confirmed by SDS-PAGE 10% with silver staining ( Supplementary Figure 1C ).

The endotoxin contaminants in OpdA samples were measured using the Endpoint Chromogenic LAL (Limulus amebocyte lysate) assay kit (Lonza) following the manufacturer’s instructions.

Cell proliferation was assessed by measuring mitochondrial dehydrogenase enzyme activity using the MTT (3-(4,5–dimethyl–2–thiazolyl)-2,5-diphenyl-2H-tetrazolium bromide) cell assay. B16F10-Nex2 cells (4x10³) were grown in 96-well plates, and after 24 hours, they were incubated for an additional 24 or 48 hours with various concentrations of OpdA. Next, the cells were incubated with 10µL of MTT (5 mg/mL) for 3 hours at 37°C in 5% CO₂, followed by the addition of 100 µL SDS in 0.01 M HCl. The plates were analyzed using Spectramax M2 equipment (Molecular Devices) at 570 nm with a 650 nm reference filter ( Supplementary Figure 3 ).

To analyze the effect of OpdA on B16F10-Nex2 melanoma development in vivo, male C57BL/6, NSG (Nod Scid gamma null), and MyD88^-/- mice were inoculated via the caudal vein with 5x10⁵ B16F10-Nex2 cells in 100µL of RPMI medium without FCS. One day after inoculation, these animals received intraperitoneal injections of either active or heat-inactivated OpdA (50µg) or PBS (control) on alternate days for two weeks. Afterwards, the melanotic lung nodules were visually quantified using an SMZ 745T stereoscopic microscope (Nikon), and images were captured with a DS-U3 digital camera (Nikon) attached to the system.

The adaptive immune response induced by OpdA treatment was assessed in male C57BL/6 and IFN-γ KO mice, divided into three groups (control and treated with active or inactive OpdA), each consisting of three animals. Mice were challenged intravenously with 1.5 x 10⁵ B16F10-Nex2 cells inoculated via the tail vein and underwent the same treatment protocol described.

After this period, the animals were anesthetized intraperitoneally with ketamine/xylazine (100 mg/kg and 10 mg/kg, respectively) for blood collection via cardiac puncture, then euthanized with 300 mg/kg and 30 mg/kg of ketamine/xylazine, respectively. The lungs, whole blood, draining lymph nodes, and spleens were collected from each animal for later analysis of effector cells and cytokines.

Spleens and draining lymph nodes (cervical, axillary, and inguinal) were macerated and filtered through a cell strainer (Corning) for cell isolation. Erythrocytes were removed using a hemolytic buffer (8.3 g/L NH₄Cl, 1 g/L NaHCO₃, and 0.02 g/L disodium EDTA) for one minute at 4°C. To neutralize the hemolytic buffer, three washes with 10% FCS-RPMI medium were performed. Following centrifugation, the supernatant was discarded, and the cells were resuspended in 10% FCS-RPMI medium. The viable cells were counted with trypan blue dye. Cells were cultured in 24-well plates for 96 hours, either with or without tumor lysate, to analyze and quantify tumor-specific CD4⁺ and CD8⁺ T lymphocytes that produce IFN-γ and IL-10. Ex vivo re-stimulation was done at a 10:1 ratio of immune cells to tumor cells per well, using tumor lysate equivalent to 10⁵ B16F10-Nex2 cells. The tumor lysate was prepared by lysing 5×10⁶ B16F10-Nex2 cells (in 2.5 mL of RPMI) through 8 cycles of freezing in liquid nitrogen and heating in a 37°C water bath, followed by centrifugation at 3,000 rpm, discarding the pellet, and applying 50 µL per well.

Splenocytes (5 x 10⁶) were restimulated ex vivo in 96-well plates with tumor lysate equivalent to 5 x 10⁵ tumor cells (ratio 10 splenocytes to 1 tumor cell) in 200 µL of RPMI/10% FCS medium. The splenocytes were stimulated for 96 hours, and during the last 5 hours, 10 µg/mL of Brefeldin A (Sigma Aldrich) was added. The cells were collected, centrifuged, and their Fc receptors blocked with a pool of sera from normal C57Bl/6 mice, inactivated at 56°C and diluted 1:30 in PBS containing 1% BSA (Sigma Aldrich) for 1 hour at 4°C. They were then washed and labeled with 50 µL of monoclonal antibodies conjugated to fluorescent molecules: anti-CD3ϵ-eFluor450 (clone 145-2C11, eBioscience), anti-CD4-FITC (clone RM4-5, eBioscience), and anti-CD8α-PerCPCy5.5 (clone 53-6.7, eBioscience), diluted 1:50 in PBS containing 1% BSA (Sigma Aldrich), for 1 hour at 4°C in the dark. Next, they were washed and fixed with 2% paraformaldehyde for 15 minutes. Cells were permeabilized with a buffer composed of 0.5% saponin (Sigma Aldrich) in PBS for 10 minutes at room temperature. The reaction was then blocked as above. Anti-IFN-γ-biotin (clone XMG1.2, BD Pharmingen) and anti-IL-10-PE (clone JES5-16E3, BD Pharmingen) monoclonal antibodies were diluted 1:50 in permeabilization buffer and incubated with the cells for 1 hour at 4°C. After washing twice with permeabilization buffer, the cells were incubated with streptavidin-FITC (BD Pharmingen) for 1 hour at 4°C, diluted in the same buffer. All samples underwent two additional washes with permeabilization buffer and PBS before being fixed with 500 µL of 2% paraformaldehyde in PBS. The samples were acquired on the FACSCanto II flow cytometer (BD Biosciences) and analyzed using FlowJo software (Tree Star). Positive controls for IFN-γ production were stimulated for 5 hours with PMA at 50 ng/mL and ionomycin at 1 µg/mL (both Sigma-Aldrich). Controls for IL-10 production were stimulated for 96 hours with anti-CD3 and anti-CD28 (both at 1 µg/mL, BD Biosciences), and in the last 48 hours, IL-2 (20 ng/mL, PeproTech) and IL-4 (50 ng/mL, PeproTech) were added.

C57Bl/6 mice received intraperitoneal treatments of two doses (3 days before and 7 days after tumor cell challenge) of 0.5 mg of anti-CD4 monoclonal antibody (clone GK1.5), or 1 mg of anti-CD8 (clone YTS), both prepared and purified in-house by Luiz S. Silva. Three days after the initial antibody treatment, the animals were inoculated intravenously with 1.5x10⁵ B16F10-Nex2 cells. On the day following the challenge, they were treated with 50 µg of active or inactive OpdA, administered every other day for 14 days. On the 15th day after the challenge, the lungs were harvested for nodule counting.

Bone marrow-derived DCs (BMDCs) were differentiated from C57Bl6, TLR4^-/-, TLR2^-/-, TRIF^-/-, or MyD88^-/- mice as described previously (38). BMDCs were cultured in Petri dishes (100 mm, Corning) in RPMI/10% FCS, supplemented with non-essential amino acids (1X), 50 µM 2-mercaptoethanol (all from Gibco), and 30 ng/mL murine rGM-CSF (PrepoTech), 10 mL per femur. On the 4th day, complete medium was added to the cultures, and after 7 days, numerous non-adherent, typical DCs were obtained.

BMDCs were cultured in 6-well plates (2x10⁶ cells per well) and stimulated with 50 µg/mL of either active or heat-inactivated OpdA for 48 hours. The cells were washed twice with PBS containing 1% BSA, and Fc receptors were blocked as described above. BMDCs were then incubated with antibodies diluted 1:60 in PBS with 1% BSA for one hour at 4 °C, washed twice with PBS, and then fixed with 2% paraformaldehyde. BMDC phenotypes were assessed using APC-labeled anti-CD11c, PE-labeled anti-CD80, PE-labeled anti-CD86, and FITC-labeled anti I-Ab, all produced by eBioscience/Thermo Fisher Scientific. Data acquisition was performed with a FACS Canto II (BD Biosciences), and population analyses were conducted using FlowJo Software (Tree Star, CA, USA).

BMDCs were cultured in 96-well plates (1x10⁵ cells per well) with varying concentrations of active or inactive OpdA, 200 U/mL IFN-γ, 200 ng/mL LPS, and, in some wells, with 20 µg/mL polymyxin B or OpdA digested with 50 µg/mL proteinase K for 48 hours. Afterwards, the culture supernatant was collected, and nitric oxide (NO) levels were measured using Griess reagent (39). Tumor necrosis factor-alpha (TNF-α) and IL-12p40 were quantified by ELISA, and a TMB substrate solution (both from BD Bioscience) was used for detection. Both ELISA and NO measurements were read with a Spectramax M2 plate reader (Molecular Devices), with NO at 540 nm and cytokines at 450 nm, applying correction at 570 nm.

Cytokines were also measured in the culture supernatant of splenocytes and lymph node cells, as well as in sera and lung homogenates, all collected on the 15th day after tumor cell inoculation, from animals treated or not with OpdA. Kits for the cytokines IFN-γ and IL-10 (BD Bioscience) were used according to the manufacturer’s instructions. At the end of the protocols, plates were washed six times with Wash buffer, and 50 μL of ELISA chromogen substrate containing hydrogen peroxide (BD Pharmigen^® TMB Substrate BD 555214) was added. Plates were kept in the dark for an additional 30 minutes, and after this period, 50 μL of Stop Solution (4N sulfuric acid) was added. The reaction was read at a wavelength of 450 nm (reference 630 nm) using a SpectraMax M2e (Molecular Devices). The cytokine concentration was calculated from the standard curve of concentration versus absorbance graph, using Microsoft Excel.

Statistical analysis was performed using GraphPad Prism software (GraphPad Software, La Jolla, CA, USA) as described in each figure. In all studies, a p-value <0.05 was considered statistically significant.

The recombinant OpdA was produced using a highly efficient process, resulting in a high-purity product that was identified by an in-house prepared anti-OpdA polyclonal antibody ( Supplementary Figure 1A ). This metalloprotease demonstrated proteolytic activity on a specific substrate that exhibited intramolecular fluorescence suppression ( Supplementary Figures 1D, E ). The heat-inactivated metalloprotease maintained the structural integrity of the active enzyme ( Supplementary Figure 1B ) and showed no proteolytic activity on the same substrate ( Supplementary Figure 1D ).

To investigate the in vivo antitumor activity of OpdA, male C57Bl/6 mice were intravenously inoculated with B16F10-Nex2 cells and treated with either active or inactive OpdA. Mice receiving inactive OpdA showed fewer metastatic lung nodules than the control group (PBS-treated mice) ( Figure 1A , Supplementary Figure 2 ), suggesting a potential indirect and non-specific effect of OpdA on tumor growth, which was further confirmed by the MTT cell viability assay ( Supplementary Figure 3 ).

To determine whether the antitumor response triggered by inactive OpdA depends on a functional adaptive immune system, immunodeficient mice (NSG) were injected with B16F10-Nex2 cells and treated with inactive OpdA. The antitumor effect of inactive OpdA was lost in these mice; in fact, the control and treated groups had a similar number of lung nodules ( Figure 1B ) and comparable lung weights ( Figure 1C ). This shows that the antitumor response from inactive OpdA relies on an active adaptive immune response.

To assess how key effectors in the adaptive immune system contribute to the protective anti-tumor response triggered by OpdA, we used genetically modified animals lacking IFN-γ and performed in vivo depletion of CD4⁺ and CD8⁺ T lymphocytes using specific monoclonal antibodies. Notably, it was observed that the absence of the IFN-γ gene caused both active and inactive OpdA-treated animals to develop similar numbers of pulmonary nodules as the PBS-treated control group ( Figure 1D ). A similar outcome was seen in CD4- and CD8-depleted animals ( Figures 1E and F , respectively), suggesting that these immune components are essential for the protective anti-tumor response mediated by either active or inactive OpdA.

We also measured the levels of IFN-γ and IL-10 in the lung homogenates and serum of mice treated with OpdA. In the lung homogenates, there was a notable increase in IFN-γ production in both the active and inactive OpdA-treated groups compared to the control group ( Figure 2A ), while IL-10 levels significantly increased only in the inactive recombinant-treated group ( Figure 2B ). The ratio of IFN-γ to IL-10, which indicates a proinflammatory antitumor protective response, was significantly higher in the active OpdA group. However, the inactive OpdA group also showed a tendency toward an increase ( Figure 2C ).

Serum from each group was pooled and analyzed for IFN-γ and IL-10 levels. The group treated with active OpdA showed a significant increase in IFN-γ levels ( Figure 2D ), while IL-10 production did not differ significantly across the groups ( Figure 2E ). The ratio of IFN-γ to IL-10 was elevated only in the group receiving active OpdA ( Figure 2F ).

We also examined the presence of tumor-specific CD4⁺ and CD8⁺ T lymphocytes that produce IFN-γ and IL-10 in the spleens and lymph nodes of OpdA-treated animals.

In mice treated with active OpdA, there was a significant increase in CD4⁺/IFN-γ, CD4⁺/IL-10, and CD8⁺/IFN-γ lymphocytes compared to those treated with PBS ( Figures 3A, B ). Conversely, mice receiving inactive OpdA only showed a statistically significant decrease in CD8⁺/IL-10 lymphocytes compared to the control group ( Figure 3B ).

The evaluation of IFN-γ and IL-10 levels in the culture supernatant, following ex vivo stimulation of cells from tumor-draining lymph nodes with tumor antigens, indicated that animals given active OpdA produced significantly higher levels of IFN-γ and IL-10 even without tumor stimuli compared to the control group. This suggests that these cells were likely pre-activated at that site. When stimulated with tumor antigens, a notable increase in IL-10 secretion was observed ( Figure 3C ). Cells from animals treated with inactive OpdA showed a significant increase in IL-10 secretion upon stimulation with tumor antigens ( Figure 3C ). Similar results were seen in splenocytes, where, in the absence of tumor antigen stimulation, their behavior mirrored that of the cells from the lymph nodes ( Figure 3D ). All groups showed a significant rise in IL-10 production upon re-stimulation with tumor antigens, especially in cells from animals treated with active OpdA. However, whether in lymph node cells or splenocytes, animals treated with active OpdA had a higher IFN-γ/IL-10 ratio compared to the other groups ( Figures 3C, D ).

The findings indicate that the antitumor protective effect of OpdA relies on a functioning adaptive immune system, as well as the presence of IFN-γ, CD4⁺, and CD8⁺ T cells.

To gain a deeper understanding of how OpdA’s immunomodulatory mechanism induces a protective antitumor immune response, we examined the effect of the metalloprotease on BMDCs.

BMDCs exposed to either active or inactive OpdA produced NO, especially when IFN-γ was present, although this was less effective than LPS stimulation ( Figure 4A ). There was a significant increase in IL-12p40 production, similar to levels seen with LPS stimulation, and IFN-γ did not affect this process ( Figure 4B ). BMDCs treated with active or inactive OpdA released higher amounts of TNF-α, whereas LPS was less effective ( Figure 4C ). Incubation with OpdA in the presence of IFN-γ led to increased expression of CD80 and CD86, yielding results comparable to LPS ( Figure 4D ). These findings show that OpdA activates BMDCs to promote an antitumor immune response.

The recombinant OpdA is expressed in E. coli, suggesting that the purified metalloprotease may contain endotoxin contaminants, mainly LPS, a potent TLR-4 ligand with immunomodulatory activity in antigen-presenting cells. To rule out possible BMDC activation caused by contaminant LPS, we treated in vitro cultures with polymyxin B, and stimulated BMDC cultures with proteinase K-treated OpdA. Polymyxin B is a cationic cyclic lipopeptide that binds to the lipid A portion of LPS and blocks its interaction with TLR-4 in BMDCs (40). Proteinase K is a serine protease that fully cleaves OpdA (41, Supplementary Figure 1C ). The presence of polymyxin B (20 μ/mL) in the assay significantly inhibited the stimulatory effect of LPS as expected, and it is evident that, although there was a partial reduction in the production of NO and TNF-α (but not IL-12p40), the activation induced by OpdA in the BMDCs remains largely intact. Conversely, when BMDCs were stimulated with OpdA treated with proteinase K, the production of cytokines and NO was completely abolished, indicating that the stimulation of BMDCs depended on the presence of OpdA in the recombinant preparation, since LPS is resistant to proteinase K but the protease fully degraded OpdA ( Supplementary Figure 1C ).

Next, we evaluated how endotoxin residues in the recombinant OpdA preparation might influence the immunomodulatory properties of the protease. In a BMDC activation assay, we measured endotoxin levels in wells containing LPS, OpdA, or both, using a Limulus amebocyte Lysate (LAL) assay. The results showed that adding 0.1 ng of LPS to BMDCs equals 0.5 EU (endotoxin units), while adding 25 µg of OpdA contains 0.68 EU ( Figure 5 ).

Figure 5 shows that BMDCs incubated with 25 µg/mL of either active or heat-inactivated OpdA in the presence of IFN-γ produced significant levels of NO, IL-12p40, and TNF-α. In contrast, 0.1 ng/mL of LPS (0.5 EU/mL) did not induce NO, IL-12p40, and TNF-α as effectively as 25 µg of OpdA (0.68 EU). Additionally, the higher concentration of LPS tested, 5 ng/mL (25 EU/mL), also failed to activate BMDCs with the same efficiency as 25 µg/mL OpdA (0.68 EU/mL).

Another potential scenario involves a synergistic effect between LPS and OpdA. To evaluate this, BMDCs were stimulated with 25 μg/mL of either active or heat-inactivated OpdA along with increasing concentrations of LPS (from 0.1 to 5 EU/mL). Results showed that none of the tested LPS concentrations increased the production of NO, IL-12p40, or TNF-α, thereby ruling out the possibility of a synergistic interaction between LPS and OpdA ( Figure 5 ).

These results support our previous findings, indicating that although the recombinant OpdA preparation contains low levels of contaminating endotoxin residues, these are not responsible for the immunomodulatory effect of OpdA on antigen-presenting cells.

For OpdA to effectively activate antigen-presenting cells, recognition is essential, so we assessed the role of TLRs. The results showed that BMDCs from TLR4^-/- mice significantly reduced NO and TNF-α secretion in response to OpdA or LPS stimulation. Additionally, a decrease in IL-12p40 secretion was observed, especially with heat-inactivated OpdA ( Figure 6A ). In contrast, BMDCs from TLR2^-/- mice did not affect cytokine and NO production ( Supplementary Figure 4A ).

Furthermore, BMDCs from C57BL/6 mice were pre-treated for one hour with CLI-095, a cyclohexane derivative that inhibits the TLR4 intracellular domain, and then exposed to either active or heat-inactivated OpdA. With this inhibitor present, the production of NO, IL-12p40, and TNF-α was completely suppressed ( Figure 4B ). Additional experiments showed that protease-activated receptors 1 and 2 (PAR-1 and PAR-2) did not participate in the activation of BMDCs by OpdA ( Supplementary Figure 4B ).

In conclusion, the findings suggest that TLR4 is involved in activating BMDCs through OpdA.

To investigate the signaling pathway responsible for the downstream activation of TLR4 by OpdA, the MyD88 and TRIF proteins were analyzed. In BMDCs from MyD88^-/- mice stimulated with LPS or OpdA, the production of NO, IL-12p40, and TNF-α was completely suppressed. In BMDCs from TRIF^-/- mice, NO production was entirely inhibited, IL-12p40 levels remained unchanged, and TNF-α showed a partial reduction ( Figure 7A ).

Additionally, BMDCs derived from C57Bl/6 mice underwent a 1-hour pre-incubation with the MyD88 inhibitory peptide (Pepinh-MyD), which disrupts MyD88 homodimerization, and the TRIF inhibitory peptide (Pepinh-TRIF), which hinders TLR-TRIF interaction, before being stimulated with heat-inactivated OpdA. As expected, the levels of NO, IL-12p40, and TNF-α were reduced due to the presence of the inhibitory peptides ( Figure 7B ).

The antitumor effect of OpdA was not observed in vivo in MyD88^-/- mice. Both treated and untreated groups had the same number of metastatic lung nodules and comparable lung weights ( Figure 7C ), confirming the in vitro findings.

In conclusion, the results show that MyD88 and TRIF molecules are essential for activating BMDCs by OpdA.

The MAPK (mitogen-activated protein kinase) signaling pathways are essential for the development and function of dendritic cells. This pathway includes several proteins, such as p38, JNK (c-Jun NH2-terminal kinases), and ERK (extracellular signal-regulated protein kinases) (42, 43). To evaluate how MAPKs influence the activation of BMDCs by OpdA, the cells were pre-incubated for one hour with SP600125, SB203580, and PD98059, which serve as pharmacological inhibitors of JNK, p38, and MAP/ERK, respectively. Following this, they were stimulated with either active or heat-inactivated OpdA or LPS (used as a positive control). The production of NO stimulated by OpdA or LPS was significantly reduced by the JNK and MAP/ERK inhibitors, while the p38 inhibitor had a less pronounced impact. TNF-α production showed a partial decrease in the presence of the inhibitors, but IL-12p40 secretion remained unaffected ( Figure 8 ).

These results suggest that proteins associated with the MAPK pathway contribute to the activation of BMDCs by OpdA.