A. castellanii ATCC 30234 (American Type Culture Collection, Manassas, VA) was cultured in Peptone- Yeast Extract- Glucose (PYG) () medium (pH 6.5) containing 20 g/L of peptone, 1 g/L yeast extract and 100 mM glucose, supplemented with 0.4 mM CaCl₂, 0.4 mM MgSO₄, 2.5 mM Na₂HPO₄, 2.5 mM KH₂PO₄, 1 g/L sodium citrate, 0.05 mM Fe(NH4)₂(SO4)₂) and 1% penicillin/streptomycin (ThermoFisher Scientific, MA, USA) at 28°C (Gonçalves et al., 2019). RAW 264.7 murine macrophages were cultured in Dulbecco Modified Eagle’s Medium (DMEM, pH 7.0) supplemented with 10% heat-inactivated fetal bovine serum, 10% NCTC-109 medium (Sigma-Aldrich, MO, USA), 1% non-essential amino acids (Sigma-Aldrich) and 1% penicillin/streptomycin (ThermoFisher) at 37°C in a 5% CO₂ incubator. Both cells were cultured until the formation of monolayers for about 48 h and subsequently used in all experiments (Taciak et al., 2018).

Fungi were grown according to standard protocols: Ca SC5314 (ATCC MYA 2876) was cultured in Sabouraud medium (pH 5.6) containing 10 g/L of peptone and 40 g/L dextrose (Gácser et al., 2007), Cn H99 serotype A (ATCC 208821) in a minimum medium (pH 5.5) composed of glucose (15 mM), MgSO₄ (10 mM), KH₂PO₄ (29.4 mM), glycine (13 mM), and thiamine-HCl (3 μM) (Guimarães et al., 2010) and Hc G217B (ATCC 26032), in Ham’s F-12 medium (pH 7.2) supplemented with glucose (18.2 g/L), glutamic acid (1 g/L), HEPES (6 g/L) and cysteine (8.4 mg/L) (Guimarães et al., 2019). All fungi were cultured for 48 h at 37°C with 150 rpm shaking.

A. castellanii trophozoites and RAW 264.7 macrophages were grown for 48 h and the percentage of viability was determined by 0.4% Trypan Blue dye exclusion, as the number of viable (white) × 100, over total cells (white + blue), with values ~100% for both cells (Strober, 2001). The phagocytes were detached from the culture flasks and washed three times with PBS (Phosphate Buffered Saline pH 8.0;137 mM NaCl, 2.7 mM KCl, 10 mM Na₂HPO₄ and 1.8 mM KH₂PO₄) at 4°C, with centrifugation at 250xg for 10 min for amoebas and 400xg for 5 min for macrophages. The cell concentration was adjusted to 25x10⁶cells/mL and surfaces biotinylated by incubating cells with 2 mM Sulfo-NHS-LC-Biotin (ThermoFisher) at 4°C for 30 min (Azim-Zadeh et al., 2007; Gonçalves et al., 2019). The excess of biotin was inactivated by four washes with 100 mM glycine in cold PBS (pH 8.0) and 10 min centrifugation.

The cell pellets were resuspended in a cold extraction buffer (0.5% CHAPS; 2 mM β-mercaptoethanol; 25 mM Tris-HCl; 100 mM NaCl; 20 mM CaCl₂; 1 mM phenylmethylsulfonyl fluoride), with addition of protease inhibitors (Complete Mini Tablets, Roche), and disrupted with a dounce homogenizer and zirconium beads, for 20 cycles in an ice bath, to obtain the total cell extract. The preparations were centrifuged at 18,000xg for 30 min at 4°C to obtain the clarified total extracts. The protein concentrations in both extracts were determined by the Pierce™ Bicinchonic Acid (BCA) Protein Assay Kit, ThermoFisher, following manufacturer’s instruction. To analyze the protein pattern of the extracts, 10 μg of proteins were resolved by 10% SDS-PAGE, as described elsewhere (Gonçalves et al., 2019). The efficiency of surface protein biotinylation was analyzed by Western blot using an alkaline phosphatase-streptavidin conjugate (SouthernBiotech, AL, USA) and developed with an NBT/BCIP substrate (ThermoFisher) as described (Gonçalves et al., 2019).

Clarified surface protein extracts of the phagocytes were purified by affinity chromatography assay, using a mannose column, according to manufacturer’s instruction (D-Mannose Agarose, Sigma-Aldrich) (Garate et al., 2004). After washing to remove the preservation solution, the resin was equilibrated with Dulbecco’s Phosphate Buffered Saline (DPBS) with MgCl₂ and CaCl₂ (pH 7.4, 0.49 mM MgCl₂ and 0.90 mM CaCl₂), with washes at 640xg for 5 min. Then, clarified extracts were incubated with the resins, in a proportion of 2.5 mL resin for every 5 mL of clarified extract (2 mg/mL), for 1 h at room temperature. Subsequently, resins were packed, and fractions of unbound proteins (F1) were collected. Then, the columns were washed with 10 bed volume of DPBS, at a 1 mL/min flow to remove non-specific binding proteins (F2). Thereafter, column-bound proteins were eluted with PBS/Ca⁺²/Mg⁺², plus 2 mM EDTA and 150 mM soluble mannose, and the eluent (F3) was collected in 0.5 mL fractions. The collected fractions F1, F2, and F3 had absorbances monitored at 280 nm, along with quantification by the BCA method, and were evaluated by SDS PAGE and Western blot as described.

The mannose-purification of clarified surface protein extracts of Ac and RAW 264.7 was scaled up to 96-well microplates previously sensitized with 50 µL of mannose solution (10 µg/mL) for 1 h at 37°C, and subsequently overnight at 4°C. After, the plates were blocked with blocking buffer (SuperBlock™ (PBS) Blocking Buffer, ThermoFisher), for 1 h at 37°C. Thereafter, plates were incubated with 200 µg/mL (50 µL/well) of phagocytes’ clarified extracts in DPBS, for 1 h at 37°C. Posteriorly, plates were washed three times with PBS to remove mannose unbound proteins. Then, mannose-attached proteins were recovered with stripping buffer (50 mM Tris-HCl, 50 mM DTT, and 2% SDS; pH 7.0) for 1 h at 70°C. Next, the content of each well was collected and dialyzed in dialysis cassettes (Slide-A-Lyzer™ 10K MWCO, ThermoFisher) against PBS/Ca⁺²/Mg⁺² at 4°C, under 200 rpm shaking for three days, with daily buffer exchanges. Then, the mannose-purified surface proteins (MPPs) were collected, identified, and analyzed by Western blot. The protein banding pattern of the MPPs were compared between the distinct phagocytes.

Ac and RAW MPPs were processed for analysis by mass spectrometry. We sought to identify and compare the sequences of mannose-binding proteins from both phagocytes to establish any possible degrees of similarity regarding mannose affinity. For isolation of biotinylated surface proteins, the MPPs were purified using the Dynabeads^® M-280 Streptavidin Kit with 100 μg of magnetic beads. The beads were washed with PBS + Tween 20 (0.01%) at pH 7.4, separated using a magnet for 5 min and incubated with biotinylated proteins for 30 min at room temperature under orbital shaking. Beads were washed five times with PBS and dehydrated in a speed vac. Beads were resuspended in Urea (7 M) and Thiourea (2 M), added of 1 M HEPES (to a final concentration of 100 mM), followed by 100 mM dithiothreitol (to 10 mM), and incubated for 1 h at 30°C. Iodoacetamide (400 mM) was added to a final concentration 40 mM, with subsequent incubation for 30 min in the dark. Then, 175 μL of ultrapure H₂O TEDIA and trypsin solution in 0.1 M acetic acid (0.1 μg/μL were added to a final concentration of 16 ng/μL and incubated overnight at 37°C. After this incubation, 2 μL of 10% TFA trifluoroacetic acid (TFA) (0.1% final concentration) were added. Then, peptides were purified in an UltraMicroSpin™columns (Havard Apparatus) and dried in a speed vac. At the end of the process, tryptic peptides were solubilized in 10 μL of 0.1% TFA and about 4 μL of the mixture wwere loaded onto an in‐house packed column (15 cm × 75 μm) filled with 3 μm ReproSil C₁₈ resin (Dr. Maisch GmbH) using the Nano LC-Ultra nano liquid chromatography system (Eksigent Technologies) (Gonçalves et al., 2018).

An amount of 0.8 μg of peptides was injected in a nano-LC–MS/MS system consisting of an EASY II-nano LC system (Proxeon Biosystem, Denmark), for LC-based separation, coupled to a nano-ESI LTQ-Orbitrap Velos mass spectrometer (Thermo Fisher). Peptide separation was performed in NanoAccquity system equipped with C-18 ReproSil 5 μm resin (Dr. Maisch) and a New Objective PicoFrit^® analytical column (75 μm × 20 cm) packed with Reprosil-pur C18-AQ 3 μm resin (Dr. Maisch). For each sample, peptides elution used a linear gradient of solvent solution ACN 95%, 0.1% TFA from 5% to 20% for 85 min, 20%-40% for 22 min, 40%-95% for 5 min and 95% for 8 min, using a flow rate of 250 nL/min. Mass spectra were acquired in a positive mode top 10 DDA (Data-Dependent Acquisition) method. MS1 scan was acquired in an Orbitrap analyzer set for a 350-1800 m/z range, 60000 resolution (at m/z 400) with a minimal signal required of 10000, isolation width of 2.0. The 10 most intense ions were fragmented by collision-induced dissociation (CID) at 30 NCE (Normalized Collision Energy) with a dynamic exclusion of 30s. The acquired spectra were processed using the peptides search engine BSI PEAKS^® X+ 10.5 (Bioinformatics Solution Inc.) for peptide identification. The considered parent mass error tolerance and fragment mass error tolerance were both set to 0.6 Da and 0.5 Da, with precursor mass search type monoisotopic. The maximum missed tryptic cleavages was set to two, and one non‐specific cleavage was allowed. Carbamidomethylating was set as fixed modification (MW = 57.02 Da) and as variable modifications, methionine oxidation (MW = 15.99 Da) and lysine Sulfo-NHS-LC-Biotin conjugation (MW = 339.16 Da) were allowed. The A. castellanii trophozoites (14,944 proteins) and Mus musculus (86,586 proteins) databases were downloaded at www.uniprot.org as for April 2021. The considered false discovery rate was set at 1%. The mass spectrometry proteomics data have been deposited to the ProteomeXchange Consortium via the PRIDE (Perez-Riverol et al., 2019) partner repository with the dataset identifier PXD031121.

An output of de novo peptides was obtained from the Peaks^®X+ software and analyzed in the DeNovo Sequencing (PepExplorer) of the PatternLab for Proteomics (http://www.patternlabforproteomics.org). The PepExplorer was set up against the same target-decoy sequence database of Ac and RAW as described above, according to instructions informed elsewhere (da Veiga Leprevost et al., 2015), with a set of parameters that were adjusted to optimize the analysis process and most stringent comparison parameters, including the ReverseDecoy Label insertion, minimal identity of 0.7, a minimum peptide size of 5 amino acids (for Ac), minimal identity of 0.8, a minimum peptide size of 5 amino acids (for M. musculus), and a denovo cut off score of 90, for both. After analysis, the dynamic report showed a list of identified proteins and the corresponding peptides based on sequence alignments that were compared between samples for the determination of proteins with affinity to mannosylated residues. Proteins abundances were calculated according to the number of spectra of all peptides identified and validatedby de novo sequencing a specific protein divided by the sum of all counted spectra for each of the candidate proteins (Ishihama et al., 2005; Guimarães et al., 2011b). Further proteins annotation and gene ontology molecular function determination were performed using the online DAVID database for annotation (https://david.ncifcrf.gov). To detect similarities between A. castellanii and M. musculus MPPs sequences identified, we performed a BLASTp search (Johnson et al., 2008) with the proteins identified in Supplementary Table 1 as queries. The results were filtered by selecting the macrophage identified MPPs ( Supplementary Table 2 ) with a percentage of identity greater than 30% and suitable maximum score (>30).

An indirect ELISA was used to evaluate the binding of clarified surface protein extracts of phagocytic cells to mannose and mannan. Initially, 96-well plates were coated with mannose or mannan as described and blocked with blocking solution (1% BSA in TBS-T) for 1 h at 37°C. After three washes, the biotinylated clarified extracts were serially diluted (200 to 0.09 µg/mL) in blocking solution and added to wells for 1 h at 37°C. Non-biotinylated clarified extracts of both phagocytes were used as negative controls. Thereafter, the plates were washed three times and incubated with an alkaline phosphatase-streptavidin conjugate (SouthernBiotech) diluted at 0.5 µg/mL in blocking solution, for 1 h at 37°C. The wells were washed three times with TBS-T and reactions developed with p-nitrophenyl phosphate (pNPP tablets, Sigma-Aldrich), at 37°C for 15 to 30 min. The absorbances were read at 405 nm on a microplate reader (Molecular Devices, CA, USA) (Guimarães et al., 2009).

Likewise, MPPs binding performances were evaluated under similar indirect ELISA, as previously described, and reactions compared between both phagocytes and to clarified total extracts.

Several techniques were used to characterize the binding of the MPPs from both phagocytes to fungi, including indirect ELISA, flow cytometry, and fluorescence microscopy Guimarães et al., 2009; (Guimarães et al., 2010; Cordero et al., 2016; Gonçalves et al., 2019). For indirect ELISAs, yeasts were washed three times with PBS at 1,100 x g for 5 min, fixed in 4% paraformaldehyde for 30 min at room temperature, resuspended in PBS and enumerated. After, plates were coated (10⁶ yeasts/well) and blocked with blocking solution for 1 h at 37°C. Subsequently, plates were incubated with the serially diluted MPPs of both phagocytes, alkaline phosphatase-streptavidin conjugate, and upon incubation with substrate, absorbances were measured as previously described. Non-biotinylated clarified extracts were used as negative controls. For the flow cytometry and fluorescence microscopy assays, yeasts were washed three times and blocked with 2% BSA in PBS for 2 h at room temperature with shaking. Upon washes, 10⁷ yeasts were incubated with clarified or MPPs (100 µg) diluted in 1 mL PBS/Ca⁺²/Mg⁺² plus 2% BSA and incubated for 2 h at room temperature. After three washes with PBS/Ca⁺²/Mg⁺², yeasts were incubated with the streptavidin-Alexa 488 conjugate (ThermoFisher) at 5 μg/mL blocking solution for 1 h at room temperature, in the dark, and subsequently washed three times. As a positive control for the binding to mannosylated components on the fungal cell wall, yeasts were incubated with ConA-TRITC (at 5 μg/mL; Thermo Fisher) for 1 h at room temperature, in the dark. After washing, yeasts were incubated with 0.5 µg/mL of Uvitex 2B (Polyscience, USA) for 30 min at room temperature in the dark, with three further washes with PBS/Ca⁺²/Mg⁺². The yeasts were examined using an Axio Imager microscope (Carl Zeiss MicroImaging Inc., USA), using a 495 nm filter, appropriate for the fluorochrome used, at 100x. In addition, the preparations were analyzed in a FACSCalibur (BD Biosciences, USA) and the intensity of FL1⁺ cells (for binding to streptavidin-Alexa 488) and FL2⁺ cells (for binding to ConA-TRITC) were recorded, Yeasts were also incubated with either only non-biotinylated clarified extracts or fluorophore conjugates, as negative controls (Gonçalves et al., 2019) and the average fluorescence intensity compared among groups.

To confirm the similarity of recognition to fungal mannosylated residues by phagocytic cells, their crossed-inhibiting effect on fungal interaction was evaluated. Ac and RAW were plated at a concentration of 4x10⁵ cells/well in 24-well plates and incubated overnight under specific conditions. Yeasts were washed with PBS and labeled with 0.04 mg/ml NHS-Rhodamine (Thermo Fisher) diluted in PBS, for 1 h at room temperature. After five washes, the yeasts were pre-incubated with 50 µg/mL of MPPs from either phagocyte for 1 h at room temperature. Following three washes with PBS, yeast concentrations were adjusted and added to phagocytes at a multiplicity of infection (MOI) of 5 fungi: 1 phagocytic cell in a crossed-way: (i) yeasts pre-incubated with MPPs of RAW 264.7 interacted with amoeba and (ii) yeasts pre-incubated with MPPs of Ac interacted with macrophages, for 30 min and 2 h, under appropriate conditions for each phagocytic cell. The stipulated interaction times were based on information from live-microscopy experiments, for early adhesion and late interaction (Gonçalves et al., 2019). After, phagocytes were washed three times, detached from the plate by pipetting up and down and analyzed in a FACSCalibur (BD Biosciences). The interaction ratio (i.e., FL2⁺ phagocytes interacting with fungi, over the total number of cells analyzed) was calculated:

Inhibition experiments were performed to confirm the importance of MBPs from each specific phagocyte model during phagocytosis. Phagocytes were plated (4x10⁵ cells/well) in 24-well plates under the appropriate growth conditions. Yeasts were washed, NHS-Rhodamine stained as described, and diluted in the required media for each phagocyte (PYG or DMEM) with or without 100 mM soluble mannose, at a MOI of 5 fungi: 1 phagocytic cell. Plates were incubated for 30 min and 2 h (Gonçalves et al., 2019) under best condition for each phagocyte and washed three times. The intracellular yeasts that remained adhered to the phagocytes were stained with 0.5 µg/ml of Uvitex 2B. Therefore, it was possible to distinguish the internalized yeasts, displaying a red fluorescence, whereas yeasts adhered to the phagocyte surface, displayed simultaneously red and blue fluorescence. Phagocytes were detached from the plates, fixed, and analyzed by flow cytometry. Additionally, the cells were examined under a fluorescence microscope at 63x objective magnification. Through the records and cell counts of each experimental group, several parameters were calculated (Gácser et al., 2007; Guimarães et al., 2019; Cseresnyes et al., 2020), including:

The Interaction ratio (Φh) describes the ratio of host cells that has at least one interacting pathogen. The adhesion (Φadh_ratio) and internalization (Φint_ratio) ratios describe respectively the proportion of adhered and internalized fungi relative to the total number of interacting fungi. Symmetrized adhesion (Φadh_sym) and internalization (Φint_sym) indices reveal the overall probability of a fungus to be respectively adhered or internalized by a host cell. All parameters were compared between the mannose treated versus untreated and between both phagocytes for early adhesion and late interaction events.

Fungal viability was assessed by determining colony forming units (CFUs) upon interaction inhibition with mannose. Phagocytic cells were plated (4x10⁵ cells/well), incubated overnight, and yeast added at a MOI of five fungi: one phagocyte in the presence or absence of 100 mM soluble mannose. Initially, interactions occurred for 2 h to allow phagocytosis of the majority of yeasts, as described in previous procedures (Gonçalves et al., 2019). Upon incubations, systems were washed three times with PBS, the appropriate culture media for each phagocyte was added and the plates returned to the appropriate conditions of either phagocytic cells, remaining for additional 24 h. Subsequently, the phagocytes were lysed with cold sterile H₂O for 1 h, followed by 15 to 20 passages through a 26 G needle and recovered yeasts plated on specific media and kept at the appropriate temperature conditions for either yeasts (Cordero et al., 2016; Gonçalves et al., 2019; Guimarães et al., 2019).

In addition, fungal viability upon interactions was also assessed by fluorescence microscopy. Yeasts were incubated with two-color fluorescent viability probe FUN™ 1 (10 mM, ThermoFisher), diluted in 1 mL of glucose buffer (2% glucose, 10 mM HEPES; pH 7.2), for 30 min at 37°C, in the dark (Gantner et al., 2005; Gácser et al., 2007; Slesiona et al., 2012). After washes with glucose buffer, yeasts were incubated with phagocytes, in the absence and presence of 100 mM mannose, as previously described. Yeasts that remained adhered to the phagocyte surface were labeled with Uvitex 2B. Therefore, internalized yeasts displayed a red (viable) or yellow-green (non-viable) fluorescence, whereas yeasts adhered to the phagocyte surface displayed a red (viable) or yellow-green (non-viable) and a blue fluorescence. Analyses were carried out by examining the preparations in an Axio Imager microscope at 63x magnification. Cell counts of each experimental group, in different fields, were performed and the following parameters were determined (Cseresnyes et al., 2020).

Statistical analyses were performed using GraphPad Prism version 8.00 for Windows (GraphPad Software, La Jolla California USA, www.graphpad.com). Comparisons between three or more groups were made by two-way ANOVA, with subsequent Tukey’s or Dunnett’s multiple comparisons test. Comparisons between two groups were made using the Student t-test. Values of p ≤ 0.05 were considered statistically significant. Each experiment was repeated at least three times.

The clarified extracts of both phagocytes were purified using a mannose-coupled resin, as described elsewhere (Garate et al., 2004), and both chromatograms display the purification efficiency and isolation of proteins with affinity to mannose ( Figures 1A, B ). For the purification of amoeboid extracts, the eluate (F3 fraction) comprised the fractions 2-10 and displayed concentrations ranging from 0.32 to 0.16 mg/mL, with the amount of total proteins that bound to mannose consisting of 12.9% of the total extract ( Figure 1A ). Eluted fractions (F3) from the purification of macrophage extracts localized at the fractions 4 and 5, with concentrations of 0.9 and 0.5 mg/ml respectively, with higher values when compared to ameboid extracts, and recovered mannose-binding proteins totalized 17.3% of the total protein mass ( Figure 1B ). Altogether, differences in protein concentration and elution time of MPPs between the phagocytes suggest higher relative content or greater affinity of the mannose-binding lectins in macrophage extracts relative to those in the ameboid extracts.

Upon a large-scale purification/enrichment of MPPs, Western blot of purified extracts of Ac revealed four mannose-affinity proteins of different molecular weights - approximately 55 KDa; 75 KDa; 110 KDa and above 150 KDa, whereas in macrophages, nine bands were identified, ranging from 37 KDa to 150 KDa ( Figure 1C ). Differences observed in the complexity of protein patterns between phagocytes point to greater diversity and variability of mannose-binding lectins on the surface of macrophages. However, upon purification, the presence of bands around 55 KDa, 75 KDa, and 100 KDa in similar ways was also evidenced in the two phagocytes.

The indirect ELISA assays revealed a dose-dependent binding to mannose with concentrations above 0.78 μg/mL for clarified biotinylated surface extracts of amoeba, in contrast to concentrations above 0.20 μg/mL for macrophages ( Figures 2A, B ). Nevertheless, the clarified RAW macrophages’ extracts ( Figure 2B ) overall displayed higher absorbances than Ac extracts ( Figure 2A ), configuring higher affinity.

Upon purification, the evaluation of MPPs’ affinity to mannose ( Figure 2C ) displayed detectable binding above 12.5 μg/mL for proteins of amoeba origin, in comparison to 3.1 μg/mL for macrophages’ proteins; however, MPPs preparations of both phagocytes displayed saturation at 50 μg/mL, despite the similar K_d values for both extracts (K_damoeba= 59.15 and K_dmacrophage= 51.44). The maximum binding capacity was higher for macrophages (Bmax_macrophage= 1.55) versus Ac MPPs (Bmax_amoeba= 0.88), confirming the higher affinity.

Amoeba and macrophage MPPs displayed a dose-dependent binding to all three fungi of distinct cell wall structure tested ( Figure 3 , p<0.05). Corroborating with the aforementioned results, macrophage MPPs exhibited greater binding intensity when compared to the ameboid MPPs to Ca ( Figure 3A ), Cn ( Figure 3B ) and Hc ( Figure 3C ), pointing to greater affinity of MPPs present on the macrophage surface.

The binding intensity and qualitative pattern of the MPPs of Ac and RAW macrophages to fungi were analyzed by flow cytometry and fluorescence microscopy and compared with ConA-TRITC mannose-binding lectin ( Figure 4 and Supplementary Figure S1 ). The means of fluorescence intensity of yeasts incubated with MPPs of either origin differed significantly from their respective negative controls ( Figure 4A , p<0.0001; Figure 4B , p<0.01; Figure 4C , p<0.0001), and demonstrated a similar profile to the positive ConA control, which bound significantly more to the surface of Ca and Hc ( Figure 4D , p<0.0001; Figure 4F , p<0.0001), and weakly to Cn ( Figure 4E ). Phagocytes’ comparison, therefore, demonstrated that macrophage MPPs bound more efficiently than amoeba MPPs ( Figures 4A–C ). Confirming the binding to the same target, addition of ConA inhibited the binding of both MPPs in all instances (light curves, Figures 4A–C ); however, addition of either MPPs had a statistically significant impact on ConA binding only in the case of Hc ( Figure 4F ).

Besides, it is possible to observe similarities in the patterns of binding of MPPs as a punctuated way to along the external layer of the cell walls of Ca ( Figure 5A ), Cn ( Figure 5B ) and Hc ( Figure 5C ), which co-localized, at the cell wall level with mannosylated regions marked by ConA and chitin by Uvitex2B.

Upon the confirmation of binding to mannose/mannan or yeasts´ surface by the proteins present in MPPs of Ac and RAW, cross-inhibitions were performed to evaluate whether, in fact, MPPs from either Ac and RAW bind to the same target on the fungal cell wall. The pre-incubation of yeasts with MPPs of either phagocyte inhibited the fungus-second phagocytic cell interaction, at early (30 min) and late (2 h) events ( Figure 6 ). Pre-incubations with macrophage MPPs inhibited the early association (at 30 min) of Ac with Ca (63.2% inhibition, p<0.0001; Figure 6A ) and Hc (58.5% inhibition, p<0.0001; Figure 6C ). Similarly, pre-incubations with Ac MPPs were able to impact the association of macrophages with Ca (60.8% inhibition; p<0.0001; Figure 6A ) and Hc (25.8% inhibition; p<0.001; Figure 6C ). For 2 h of interaction, pre-incubations with macrophage MPPs inhibited the Ac engulfment of Ca (74.8%, p<0.0001; Figure 6A ) and Hc (38.5%, p<0.0001; Figure 6C ) by amoeba, whereas pre-treatment with Ac MPPs inhibited the interaction between macrophages and Ca (43.9%, p<0.0001; Figure 6A ) and Cn (30.4%, p<0.01; Figure 6B ). The occurrence of inhibition in the two phagocytic models suggests binding of MPPs of both phagocytes to the same targets of the fungal cell wall.

To clearly determine the impact of MBPs on the phagocyte-fungi interaction, we further performed the interaction assays for 30 min and 2 h using 100 mM soluble mannose as an inhibitor. The results were obtained through analysis by flow cytometry and fluorescence microscopy to establish within the interaction the probability of adhesion and internalization events ( Figures 7 , 8 ). The presence of mannose inhibited the interaction between phagocytic cells and yeasts, at all times analyzed. At 30 min of interaction, inhibition of association with Ac was observed for Ca (29.1%; p<0.0001, Figure 7A ), Cn (65.5%; p<0.05, Figure 7B ) and Hc (52.1%; p<0.0001, Figure 7C ). For late events (2 h), mannose also had an impact on the interaction of Ac with Ca, Cn, and Hc (27.1%, 73.6%, and 20.0% inhibition, respectively; p<0.0001). Regarding macrophages, within the first 30 min, soluble mannose affected RAW-fungi association, to Ca and Hc (21.5% and 19.3% inhibition, respectively; p<0.001). At 2 h, mannose inhibition was less impacting, but it still impaired the phagocytosis of Ca and Hc (13.4%, Figure 7A and 11.9% inhibition, Figure 7C , respectively; p<0.01). Mannose was not able to inhibit the interactions of Cn and RAW macrophages at the two time-points evaluated ( Figure 7B ). These observations suggest the participation of mannose-binding lectins on the phagocytes-yeast interaction, and blocking these receptors affects both early fungal adhesion and phagocytosis. Overall, we can conjecture a greater inhibitory effect by mannose in the interaction between fungi and Ac than macrophage, and possibly, MBPs are essential and exclusive pathways for the interaction of amoeba with pathogens, whereas in the context of interaction with macrophages, mannose receptors have less expressive action or there is a compensation, with the participation of other pathways for fungal recognition.

To dissect the impact of the mannose dependency on the events of fungal adherence and internalization to phagocytes, the samples were examined under a fluorescence microscope and symmetrized indices of adherence and internalization were determined, i.e., the overall probability of any fungi to be adhered or internalized by a phagocytic cell (Gácser et al., 2007; Guimarães et al., 2019) ( Figure 8 ). For the Ac-fungi interaction upon 30 min, the presence of soluble mannose seemed to directly impair the adhesion and initial internalization of Ca; at 2 h, despite similar average of adhered fungi, the mannose greatly inhibited the overall internalization of Ca ( Figure 8A ). For Cn, a higher number of adhered yeasts along with less internalization were observed under mannose inhibition at 30 min; however, at 2 h, mannose inhibition seems to impair both adherence and internalization process of Cn by Ac ( Figure 8B ). For Hc at 30 min of interaction, soluble mannose impaired initial internalization, resulting in higher number of adhered yeasts; therefore at 2 h, even with similar average number of attached yeasts per Ac, the internalization process is greatly impaired ( Figure 8C ). Overall, inhibition with mannose also resulted in a disbalance, with greater adhesion and lower fungal internalization by macrophages at 30 min; within 2 h, mannose inhibition greatly impacted the internalization of Ca ( Figure 8D ), Cn ( Figure 8E ) and Hc ( Figure 8F ) by RAW macrophages, with more adhered yeast as a result. Altogether, these findings demonstrate that mannose blocks the early fungal entry through high-affinity receptors, which constitute faster pathways of interaction and for the initial adhesion/encompassing phase. Furthermore, mannose inhibition continues to impact the late interaction mechanisms and the action of lower affinity receptors in the establishment of posterior connections for the phagocytic process.

As demonstrated in competition trials with soluble mannose, the mannose binding proteins play an essential role in the adhesion and internalization of fungi by phagocytes. However, we also assessed the impact of the mannose recognition on the viability of internalized fungi by monitoring the viability and mortality ratios (Gantner et al., 2005; Gácser et al., 2007; Slesiona et al., 2012) ( Figure 9 ). Upon 2 h, co-incubations with mannose reduced the viability ratios for Ca and Hc in comparison to the control groups in the Ac-fungi interaction ( Figures 9A, C , respectively). For these two fungi, similar results upon interactions with RAW macrophages were observed. Instead, for Cn, the inhibition of the mannose receptor had no impact on the proportion of alive/dead yeasts in the interactions with Ac ( Figure 9B ) in contrast to RAW macrophages, in which blockade of mannose receptors seemed to reduce fungal viability. Altogether, the inhibition assays highlight the role of the mannose receptor on fungal entry and survival.

Given the results exposed and the likelihood of mannose inhibition to impact on fungal survival in the intracellular environment of phagocytes, we assessed the fungicidal capacity of phagocytes by determining the colony-forming units (CFUs), at 24 h upon internalization. The absolute CFU numbers obtained were normalized by the symmetrized internalization (phagocytic) index, to estimate relatively the number of viable pathogens that were internalized by phagocytes. For Ca ( Figure 10A ) and Hc ( Figure 10C ) mannose inhibition reduced viability about 2-logs in comparison to control groups for both phagocytes. Similar behavior was observed for mannose inhibition during Ca and Hc interactions with RAW macrophages. However, for Cn, despite the lower absolute CFU numbers in the presence of mannose, the normalization by the symmetrized internalization index ( Figure 10B ) demonstrated overall similar viability upon interactions with amoeba ( Figure 10B ), whereas a reduction was observed for Cn interaction with macrophages.

After the isolation of MPPs of Ac and RAW, the samples were processed for identification by mass spectrometry. In the Ac MPPs pool, 43 proteins were identified and relatively quantified ( Supplementary Table 1 ), whose molecular function annotation classified the vast majority of proteins as carbohydrate binding (GO:0030246) and seven other main groups ( Figure 11A ). For macrophages instead, 120 MPPs were identified ( Supplementary Table 2 ) and fell into 15 distinct molecular functions ( Figure 11B ), with the vast majority demonstrating the carbohydrate binding function (GO:0030246).

Regarding specifically the literature recognized mannose binding lectins of Ac (Gonçalves et al., 2019), the transmembrane protein MBP1 (Q6J288) was the most abundant (2.6%), followed by the MBP (L8GXW7, 0.7%). In RAW macrophages, the mannose receptor Mrc2 (Q64449) was the most abundant (0.8%), followed by the Mrc1 (Q61830; 0.7%), both featuring transmembrane helical domain. Alignment of the sequence of these knows mannose receptors from both organisms demonstrated an overlapping of the DUF4114 domains of Ac proteins and the C-type lectin-like domains of RAW receptors and denoted a perfect alignment of the regions equivalent to the secondary structure of ConA-like lectins/glucanases superfamily in all proteins ( Figure 11C ).

Protein sequences of identified Ac MPPs ( Supplementary Table 1 ) were used as query to find the highest similarity among macrophage identified MPPs ( Supplementary Table 2 ), and in this subset specifically, six amoeba proteins, including the MBPs Q6J288 and five other lectins, had mimetic macrophages proteins with high similarity ( Figure 11D ). Overall, amoebas seem to possess a simpler repertory for mannose binding and lower number of analog genes, whereas MPPs from macrophages represented a more diverse group, which could be the result of paralogical genetic evolution from ancestor genes.