Handling of cows and blood sampling were conducted in accordance with Spanish and EU legislation (Law 32/2007, concerning animals, their exploitation, transportation, experimentation and sacrifice; Royal Decree 53/2013 for the protection of animals employed in research and teaching; Directive 2010/63/UE, related to the protection of animals used for scientific goals). Protocols were approved by the Ethical Committee of the Council of Agriculture, Farming, and Autoctonous Resources of the Principality of Asturias, Spain (permit number PROAE 25/2016) and the Animal Welfare Committee of the Community of Madrid, Spain (permit number PROEX 236/17).

BoMØs were generated as previously described (26, 27) from monocytes isolated from peripheral blood collected from six adult dairy cows testing negative for N. caninum, infectious bovine rhinotracheitis virus (IBRV), and bovine viral diarrhea virus (BVDV), both of which are known to induce immunosuppression. Motility, transmigration, cytoskeletal morphology, and gelatin degradation assays were performed using monocytes isolated from bovine peripheral blood purchased from a commercial supplier (Håtunalab AB, Bro, Sweden).

Independent of the source of blood, peripheral blood mononuclear cells (PBMCs) were separated by gradient density centrifugation on Histopaque 1077 (Sigma-Aldrich, USA), and monocytes were isolated by positive selection using mouse anti-human CD14-coupled microbeads (Miltenyi Biotec Ltd., USA) according to the manufacturer's instructions. The identity and purity of monocytes (≥95%) was determined by flow cytometry using a mouse anti-bovine CD14 FITC-labeled antibody (clone CC-G33, Bio-rad Laboratories, USA). Monocytes (CD14⁺ cells) were seeded in 6-well culture plates at a density of 10⁶ cells ml⁻¹ in 3 ml of RPMI 1,640 medium (Invitrogen, Life Technologies, UK) supplemented with 10% heat-inactivated fetal calf serum (FCS), 100 IU ml⁻¹ of penicillin, 100 μg ml⁻¹ of streptomycin, and 50 μM β-mercaptoethanol (Merck Millipore, USA), referred to as complete medium (CM). Cells were incubated at 37°C in the presence of 5% CO₂ with 100 ng ml⁻¹ recombinant bovine (rbo) GM-CSF (Kingfisher Biotech, USA). At day 3, 1 ml of medium from each well was replaced with 1 ml of fresh CM with 100 ng ml⁻¹ rboGM-CSF. After 5 days of culture, further evidence of in vitro generation of MØs was assessed by light microscopy and flow cytometry based on increased size, increased adherence, cytoplasmic granularity, presence of cell appendages, and expression of surface antigens using antibodies specific for the bovine molecules CD14, MHC Class II, CD80, CD86, CD172a, CD11b, and CD1b) (Supplementary Figure 1).

Prior to parasite infection, MØs were harvested using ice-cold PBS with 2 mM EDTA and soft scraping, reseeded in culture plates at the density of viable cells indicated for each assay and incubated for 24 h to minimize possible cellular stress due to the harvesting procedure. The viability of the cells was checked with a trypan blue exclusion test, and was normally above 95%.

N. caninum isolates of high and low virulence (Nc-Spain7 and Nc-Spain1H, respectively) were obtained from healthy calves infected congenitally (14, 28) and characterized in vitro and in vivo using murine and bovine models. Virulence differences were based on infection and proliferation rates in vitro and percentage of abortion and vertical transmission in vivo (14–17). Tachyzoites were routinely maintained in a monolayer culture of the African green monkey cell line (MA-104 clone 8, provided by Hipra Laboratories, SA) as described previously (28). On each passage, the cultures were scraped, passed by 25 G needle, and inoculated onto a new monolayer cell culture. To reduce potential changes in virulence due to prolonged maintenance in vitro (29), parasites were used at a passage lower than 15 for all the experiments. Tachyzoites used for boMØ infection were collected from 3 to 3.5 day-growth cultures, when the majority of the parasites (at least 80%) were still in parasitophorous vacuoles, and purified using PD-10 Desalting Columns (G.E. Healthcare, Buckinghamshire, UK) as previously described (13). Viable tachyzoites were counted in a Neubauer chamber via trypan blue exclusion and were inoculated within 1 h of parasite collection.

For cell infection rate (cIR) determination, multiplicities of infection (MOI)s of 3–5 were chosen based on previous in vitro studies (30, 31). A MOI of 3 was used in most of the assays to obtain an elevated cIR with a low percentage of multi-infection. For transmigration assay protocol, which requires an increased handling of the MØs, a MOI of 2 was used to ensure cell survival.

The T. gondii RFP-expressing Prugniaud line (PRU-RFP, type II) (32) was used as reference line in motility, transmigration, morphology and matrix degradation assays. Tachyzoites were cultured as described for N. caninum and collected from 2.5 to 3 day-growth cultures. As a control of phagocytosis and cell response, MØs were inoculated with heat-inactivated (HI) N. caninum tachyzoites from Nc-Spain7 and Nc-Spain1H isolates mixed at a ratio 1:1. Purified tachyzoites were killed by incubation at 56°C for 30 min as described previously (33) Lack of viability of tachyzoites was confirmed by trypan blue exclusion and by real-time PCR that measured the expression of NcTUBα as previously described (34) in cDNA samples of MA-104 cultures infected with HI tachyzoites for 1 week.

To study the differential ability of the N. caninum isolates to infect and survive in boMØs, cIR defined as the percentage of cells infected with one or multiple tachyzoites using different parasite doses, was determined and compared at 8 and 36 h post-infection (hpi). MØs were cultured in 24-well plates at a density of 2.5 × 10⁵ cells/well and inoculated with live Nc-Spain7 or Nc-Spain1H at a MOI of 3, 4, and 5. In addition, MØs were inoculated with HI tachyzoites at the same MOIs as a control for phagocytosis. Cultures were fixed with 0.05% glutaraldehyde (GA) and 3% paraformaldehyde (PF) and stained using double immunofluorescence staining as described previously (30). Overall number of cells, number of cells containing at least one tachyzoite, and number of cells containing more than one tachyzoite (multi-infected cells) were counted in 10 arbitrarily selected fields using an inverted fluorescence microscope (Nikon Eclipse TE 200, Japan) at a magnification of 200 ×. Counting of events was carried out on images taken with different filters for visualization of nuclei and intracellular and extracellular tachyzoites using a Nikon DSL1 camera; the images were overlaid using Photoshop software (Adobe Systems Incorporated, USA). A mean value of 50 cells was counted in each field by a single operator in order to avoid possible differences due to subjectivity in individual appreciations, and only tachyzoites that retained an unaltered morphology were considered in the count.

To assess the ability of Nc-Spain7 and Nc-Spain 1H to actively invade the host cell, phagocytosis activity of MØs was inhibited by treatment with 10 μM Cytochalasin D (Sigma-Aldrich, Spain) for 30 min, following by washing cells three times with PBS. MØs were cultured, infected, fixed and stained as described above. A MOI of 3 was used for inoculation with Nc-Spain7, Nc-Spain1H, and HI tachyzoites. Parasite invasion was determined in parallel in Cytochalasin D-treated and untreated MØs at 8 hpi.

To identify intracellular localization of tachyzoites and their ability to evade degradation, lysosomes were labeled by the addition of LysoTracker Red DND-99 (Thermo Fisher Scientific, USA) into the culture media at a concentration of 75 nM 30 min prior the fixing. Cells were cultured and infected as indicated above using a MOI of 3. At 24 hpi cultures were fixed with 0.05% glutaraldehyde (GA) and 3% paraformaldehyde (PF). Immunofluorescence staining of the parasites is described below.

Proliferation kinetics of Nc-Spain7 and Nc-Spain1H isolates in MØs were determined by quantifying the number of tachyzoites at specific times (8, 24, 36, 48, 60, and 72 hpi) using quantitative real-time PCR (qPCR). Cells were cultured and infected as indicated above using an MOI of 3. Samples were collected by adding 200 μl of PBS, 180 μl of lysis buffer, and 20 μl of proteinase K (Qiagen, Germany) to each well and were stored at −80°C until DNA extraction. DNA extraction and qPCR were carried out as specified previously (30). Briefly, genomic DNA was extracted using a DNeasy Blood & Tissue Kit (Qiagen, Germany) according to the manufacturer's instructions. Quantification of N. caninum DNA was performed via qPCR using a 7,300 Real-Time PCR System (Applied Biosystems, USA). A standard curve of 10⁻¹ to 10⁴ tachyzoites was used for the quantification (35). A bovine β-actin standard curve was constructed (from 64 to 0.2 ng DNA/μl) in order to normalize the quantification of the parasites in each sample. The results are expressed as the relationship between the amounts of parasite DNA and cell DNA (R² ≥ 0.99; slope values varied from −3.47 to −3.11).

The doubling time (Td), defined as the period of time required for a tachyzoite to duplicate during the exponential multiplication phase, was calculated as previously described (13) by applying non-linear regression analysis and an exponential growth equation using GraphPad Prism (GraphPad Software, USA). The Td for each isolate is presented as the average value obtained from all determinations that revealed a linear regression, R² ≥ 0.95 (13). The tachyzoite yield (TY_48h) was defined as the average number of tachyzoites quantified by qPCR at 48 hpi for each isolate.

In parallel, cells were seeded on coverslips, infected and labeled using a double-immunostaining protocol as described previously (30) to study the proliferation kinetics of both isolates in MØs via microscopy, as a complementary technique to the quantification by qPCR. Three coverslips were photographed for each condition using an inverted fluorescence microscope (Nikon Eclipse TE 200).

MØs were seeded on poly-L-lysine (Sigma Aldrich, USA)-coated coverslips at a density of 4 × 10⁴ cells/coverslip and inoculated with Nc-Spain7, Nc-Spain1H, HI tachyzoites, or T. gondii (PRU-RFP) tachyzoites (MOI 2). Non-infected MØs were used as the negative control. At 8 hpi, cells were fixed with 0.05% GA and 3% PF and stained as indicated below. The morphology of MØs (100 cells/condition) was analyzed using a Leica DMRB epifluorescence microscope with a 100 × objective and scored on a 0–5 scale according to the following criteria (with non-infected MØs as the reference): (i) Podosome structures: present (score 0) vs. reduced (score 1) or absent (score 2); (ii) Cell shape: elongated (score 0) vs. rounded (score 1); (iii) Filopodia-like extensions: present (score 0) vs. absent (score 1); (iv) Presence of membrane veils and/or ruffles: present (score 0) vs. absent (score 1).

The in vitro gelatinolytic activity of MØs, as a marker for their ability to degrade collagen extracellular matrix, was analyzed by gelatinolysis of Oregon green 488 (OG 488)-conjugated porcine gelatin (Molecular probes, Thermo Fisher Scientific, USA) as described previously (24). Briefly, 2.5 × 10⁴ MØs were inoculated with N. caninum Nc-Spain7, Nc-Spain1H tachyzoites or with T. gondii (PRU-RFP) tachyzoites (MOI as indicated), deposited on OG-488 gelatin-coated Lab-tek chambers (VWR) and incubated for 24 h in CM. Cells were subsequently fixed with 4% PF and stained with DAPI (Thermo Fisher Scientific, USA). N. caninum tachyzoites were stained as indicated below. After fixation and staining, images were generated (160 FOV/ condition, 1 FOV = 0.617 mm²) using the 10 × objective of the Zeiss Observer Z1. For each chamber, fluorescence (channels, Ex/Em: DAPI, 360/475; OG-488, 450/525; RFP, 555/585) and phase contrast images covering 0.98/1.6 cm² (61.25% of the total well area) were recorded. Image analysis was automated with the open source software package Cellprofiler (v2.1.1, rev: 9969f42). All images were run through a pipeline designed to define gelatin degradation, non-infected cells and infected cells. Gelatin degradation was defined as loss of signal (gelatin, Oregon green-488). Cells were defined as the Euclidian center of a signal (nuclei) + 15 μm radius. Infected cells were defined as the presence of signal (N. caninum: Alexa 594 or T. gondii: RFP) overlapping with a Euclidian center of signal (nuclei) + 15 μm radius. Degradation was then ascribed to each cell population by overlaying non-infected and infected cells with degradation. Co-ascribed degradation was omitted.

The motility assay was conducted as previously described (23). Briefly, MØs were seeded in a 96-well plate at a density of 2 × 10⁴ cells/well and inoculated with N. caninum Nc-Spain7, Nc-Spain1H, or HI tachyzoites (MOI 3). In addition, MØs inoculated with T. gondii (PRU-RFP, MOI 3) tachyzoites were used for motility comparisons, and non-infected MØs served as the negative control. Prior to infection, N. caninum tachyzoites were incubated for 15 min with CMTMR dye (2.5 μM, Thermo Fisher Scientific, USA). Bovine collagen I (0.75 mg ml⁻¹, Thermo Fisher Scientific, USA) was added at 4 hpi, and cells were imaged every min for 60 min with a 10 × objective (Zeiss Observer Z1, Zen 2 Blue v.4.0.3). The infection rate was determined, and the absence of differences between groups of inoculated MØs was confirmed (data not shown). Motility data were obtained from 50 cells/condition using ImageJ (Manual Tracking and Chemotaxis and Migration tool plugins).

Transmigration assays were performed as previously described (20), with minimal modifications. Briefly, MØs were seeded at a density of 1 × 10⁶ cells/well and inoculated with Nc-Spain7, Nc-Spain1H, or HI tachyzoites at a MOI of 2. T. gondii-infected MØs (PRU-RFP) were also included in the study as above, and non-infected MØs were used as the negative control. At 6 hpi, MØs were recovered by soft scraping, and 3 × 10⁵ cells/condition were transferred to transwell filters (8 μm pore size; BD biosciences, San José, CA, USA) and incubated for 16 h at 37°C with 5% CO_2. In addition, 2 × 10⁴ cells/condition were seeded on coverslips, fixed and stained as indicated below, and the infection rate was determined, confirming the absence of differences between groups of inoculated MØs (data not shown). Migrated MØs were recovered by trypsinization and quantified using a Neubauer chamber.

Analysis of intracellular ROS generation by infected MØs was performed via flow cytometry. MØs were seeded in 6-well plates at a final density of 1.5 × 10⁶ cells/well and incubated in CM for 1 or 24 h post inoculation with Nc-Spain7, Nc-Spain1H or HI tachyzoites (MOI 3). MØs inoculated with H₂O₂ (2 μM) were used as the positive control and non-infected MØs as the negative control. Cells were then harvested and stained with 5 μM CellROX Green Reagent (Thermo Fisher Scientific, USA) for 30 min at 37°C, and the fluorescence of viable cells, gated for propidium iodide (5 μg ml⁻¹) staining, was assessed in a FACS Calibur cell Analyzer (BD Bioscience, USA).

The mRNA expression levels of IL-10 and IL-12p40 were determined by quantitative reverse-transcription real-time PCR (RT-qPCR). MØs were seeded at a density of 10⁶ cells ml⁻¹ in 6-well culture plates and inoculated with Nc-Spain7, Nc-Spain1H, and HI tachyzoites at MOI 3 or with LPS (100 ng ml⁻¹) as the control for MØ activation for 8 h and then resuspended in 300 μl of RNAlater (Qiagen, Germany). Non-infected MØs were used as the negative control. Samples were recovered 8 h later by scraping and centrifugation at 1,350 × g for 15 min at 4°C. The obtained pellets were resuspended in 300 μl of RNAlater (Qiagen, Germany) and stored at −80°C until RNA extraction. Analysis was performed on three replicates obtained from three independent experiments. RNA was extracted using the commercial Maxwell 16 LEV simply RNA Purification Kit (Promega, USA) following the manufacturer's recommendations. Integrity was checked by 1% agarose gel and RNA concentrations were determined using a spectrophotometer (Nanophotomer, Implen GmbH, Germany). cDNA was obtained by reverse transcription using a master mix SuperScript VILO cDNA Synthesis Kit (Invitrogen, UK), and IL-10 and IL-12p40 mRNA expression was determined using the primers and conditions previously described (17). Genes were considered to be differentially expressed when they presented a fold change (FC) ≥2 and p < 0.05.

Surface expression of CD14, MHC Class II, CD80, CD86, CD172a, CD11b, and CD1b by boMØs after exposure to N. caninum was determined by flow cytometry. To do so, MØs were seeded in 24-well plates at a density of 3 × 10⁵ cells/well and inoculated with Nc-Spain7, Nc-Spain1H, and HI tachyzoites (MOI 3) or LPS (100 ng ml⁻¹) for 4 h as the control for MØ activation. After 4 h, cells were detached from the culture plates by incubation with cold PBS with 2 mM EDTA at 4°C for 10 min and soft scraping. Cultures were recovered in 30 ml centrifuge tubes and centrifuged at 1,200 rpm for 5 min at 4°C. After centrifugation, cells were resuspended in cold PBS at a density of 2 × 10⁵ cell/100 μl. A volume of 100 μl of cells per well was pelleted in a V-bottom 96-well plate and centrifuged at 1,300 rpm 4°C for 3 min. PBS was discarded and cells were incubated with 50 μl of diluted antibody (1:100) or 50 μl of PBS for 30 min on ice and protected from light. After the incubation, samples were washed with PBS before adding the fixative BD Cellfix (BD Bioscience, USA).

The negative fraction (CD14⁻ cells) obtained during positive selection of monocytes (CD14⁺ cells) and used for determination of IFN-γ secretion assay was characterized following the protocol described for surface marker analysis of MØs. The percentage of CD4 T cells (CD4⁺), CD8 T cells (CD8⁺), Natural Killer (NKp46, CD335⁺), gamma delta T cells (WC1⁺), and B cells (CD21⁺) in the population was determined (Supplementary Figure 2).

The percentage of positive cells and mean fluorescence intensity (MFI) was measured for each marker using a Becton Dickinson FACSCalibur cytometer (BD Bioscience, USA). The data were analyzed using FlowJo software (FlowJo, LLC, USA). The list of antibodies used to analyze subsets is given in Supplementary Table 1.

IFN-γ secretion was measured in supernatants of lymphocytes incubated with MØs. Lymphocytes used in the assay (CD14⁻ cells) were obtained as the flow through after magnetic isolation of CD14⁺ cells. This cell population, characterized by flow cytometry as indicated below, contained CD4 T cells (26.53% ± 1.45), CD8 T cells (34.05% ± 1.07), B cells (18.06% ± 0.77) Natural Killer (8.46% ± 0.23), and γδ T cells (18.26% ± 1.14). MØs were seeded in 24-well plates at a density of 2.5 × 10⁵ cells/well and inoculated with Nc-Spain7, Nc-Spain1H or HI tachyzoites at MOI 3 or with LPS (100 ng ml⁻¹). Twenty-four hours later, lymphocytes were added at a density of 12.5 × 10⁵ cells/well. To verify that IFN-γ was secreted in our samples exclusively by lymphocytes upon induction by infected MØs, MØs inoculated at the conditions previously described without lymphocyte addition and lymphocytes without MØs were included in the experiment as controls. Supernatants were recovered 48 and 72 hpi and bovine IFN-γ concentrations were determined using a bovine IFN-γ ELISA development kit (Mabtech AB, Sweden), following the manufacturer's recommendations. The color reaction was developed by the addition of 3,3′,5,5′-tetramethylbenzidine substrate (TMB, Sigma-Aldrich, Spain) and incubated for 5–10 min in the dark. Reactions were stopped by adding 2N H₂SO₄. Then, plates were read at 450 nm. The cytokine concentrations were calculated by interpolation from a standard curve generated with recombinant cytokines provided by the kit.

For the gelatin degradation assay, a single immunofluorescence staining of tachyzoites was carried out as described previously (30). After permeabilization with Triton X-100 (Thermo Fisher Scientific, USA), parasites were stained using hyperimmune rabbit antiserum directed against N. caninum tachyzoites (1:1,000) as the primary antibody (36) and goat anti-rabbit IgG conjugated to Alexa Fluor-594 (Thermo Fisher Scientific, USA) (1:1,000) as the secondary antibody.

For cIR determination, intracellular proliferation and cytoskeletal morphology assays, a double immunofluorescence staining of tachyzoites was carried out as described previously (30). Parasites were stained using hyperimmune rabbit antiserum directed against N. caninum tachyzoites (1:1,000) as the primary antibody. For the double immunostaining, when F-actin filaments were stained (intracellular proliferation assay) with Alexa Fluor-594 Phalloidin (Thermo Fisher Scientific, USA) or non-stained (cIR determination), a 1:1,000 dilution of goat anti-rabbit IgG conjugated to Alexa Fluor-488 (Thermo Fisher Scientific, USA) was added before permeabilization with Triton X-100 to stain extracellular tachyzoites. After permeabilization, a 1:1,000 dilution of goat anti-rabbit IgG conjugated to Alexa Fluor 594 was used to stain both extra- and intracellular tachyzoites. When Alexa Fluor-488 Phalloidin was used for the staining of F-actin filaments (morphology assay), donkey anti-rabbit Alexa Fluor 350 (Thermo Fisher Scientific, USA) was added to stain extracellular tachyzoites, and goat anti-rabbit IgG conjugated to Alexa Fluor 594 (Thermo Fisher Scientific, USA) was used to stain both extra- and intracellular tachyzoites. The nuclei were stained by washing the cells with a solution of 1:5,000 DAPI (Thermo Fisher Scientific, USA) in PBS, and the coverslips were embedded in Fluoroprep (BioMerieux, France).

For the study of the lysosomal activity of infected MØs, cells were permeabilized with 0.5% saponin (Fluka BioChemika, Spain). Parasites were stained using hyperimmune rabbit antiserum directed against N. caninum tachyzoites (1:1,000) as the primary antibody and goat anti-rabbit IgG conjugated to Alexa Fluor-488 (Thermo Fisher Scientific, USA) (1:1,000) as the secondary antibody.

After testing the samples for normal distribution, differences in cIR, proliferation, and cytokine expression were analyzed using a non-parametric Kruskal-Wallis test followed by Dunn's post-hoc test for comparisons between groups. Mann–Whitney U-test was used for pairwise comparisons. ROS generation, IFN-γ secretion, surface marker detection, motility, transmigration, cytoskeletal morphology, and gelatin degradation assay data were analyzed via parametric one-way ANOVA, followed by Tukey's post-hoc test for multiple comparisons and Student's t-test for pairwise comparisons. Dunnett's test was used to compare all the groups to the non-infected group. Statistical significance was established at p < 0.05.GraphPad Prism 7 v.7.04 (San Diego, CA, USA) software was used to perform all statistical analyses and graphical illustrations.

In an initial experiment, we compared the ability of N. caninum isolates of different virulence to complete their life cycle in boMØs. First, we evaluated the capacity of the parasite to infect and survive in boMØs. At 8 hpi, MØs internalized tachyzoites with the same efficiency for all assayed conditions. The cIR and the percentage of multi-infected cells did not increase significantly with increasing the MOI, and non-significant differences were found between isolates or between live vs. HI tachyzoites (Figure 1A, p > 0.05). To study the ability of both isolates to survive in MØs, cIR and the percentage of multi-infected cells was also evaluated at 36 hpi. The time point of 36 hpi was chosen to permit degradation of dead tachyzoites in order to count only live tachyzoites. Discerning between live and dead tachyzoites was also facilitated by the higher multiplication of the parasites at that time-point. Tachyzoites with unaltered morphology (regarding size, shape, and keeping a smooth surface in the absence of visible defects on its membrane) were counted in MØs infected with Nc-Spain7 and Nc-Spain1H isolates. However, only degraded tachyzoites were observed in MØs inoculated with HI tachyzoites at 36 hpi. A statistically significant reduction in cIR (from 61.34 ± 12.4 to 34.94 ± 14.35; p < 0.001) and the percentage of multi-infected cells (from 35.91 ± 15.08 to 12.50 ± 8.74; p < 0.001) was shown for both isolates at 36 hpi compared with 8 hpi at all MOIs assayed (Figure 1A). Additionally, when comparing both isolates at 36 hpi, the highly virulent isolate Nc-Spain7 showed a higher multi-infected cell number (p < 0.05 for MOI 4 and 5) than the low virulence isolate Nc-Spain1H.

As the presence of intracellular tachyzoites can be either based on active invasion or actin-filament dependent phagocytosis, we next assessed these possibilities by disrupting the actin skeleton using Cytochalasin D. After inhibition of phagocytosis, Nc-Spain7 showed a higher cIR and percentage of multi-infected cells (34.96 ± 5.89 and 13.22 ± 4.89, respectively; p < 0.001) compared to Nc-Spain1H (25.45 ± 6.80 and 7.09 ± 3.33) at 8 hpi (MOI 3). Cytochalasin D-treated MØs inoculated with HI tachyzoites showed a cIR of 2.24 ± 2.68, with a percentage of multi-infected cells of 0.44 ± 0.75. No differences were observed between cIR in Cytochalasin D-treated MØs at 8 hpi (Figure 1B) vs. cIR at 36 hpi (Figure 1A).

Having established that tachyzoites of both isolates are able to actively invade boMØs, we next assessed whether these would thus also evade lysosomal degradation at 24 hpi. As expected based on the results obtained with Cytochalasin D, MØs exposed to HI tachyzoites showed clear co-localization of LysoTracker, and degraded tachyzoites. In contrast, this was absent for tachyzoites of both strains in replication, indicating no fusion of the parasitophorous vacuole with lysosomal compartments (Figure 1C). These results show that those tachyzoites internalized by phagocytosis suffer the macrophage driven degradation effects.

As a second marker to identify potential differences, we assayed the ability of both N. caninum isolates to proliferate in boMØs in vitro. Microscopic examination of cultures infected at a MOI of 3 and fixed at different times post-infection showed that Nc-Spain7 and Nc-Spain1H tachyzoites had already begun to multiply at 24 hpi. Parasitophorous vacuoles of Nc-Spain7 were bigger than Nc-Spain1H vacuoles from 48 hpi onwards. For Nc-Spain7, rupture of the host cells and egression of the tachyzoites began to be visualized from 48 hpi and was complete before 72 hpi, at which time point, invasion of new cells by the egressed tachyzoites was observed. For Nc-Spain1H egression did not begin until 60 hpi. At 72 hpi egression is not completed, since intact vacuoles were still visualized (Figure 2A). Proliferation kinetics over time assessed by qPCR are presented in Figure 2B. From 24 hpi onwards, the average number of tachyzoites was higher for Nc-Spain7 for each time point assayed (p < 0.001). The Td value was analyzed to determine the length of the cell cycle for both isolates. Nc-Spain7 showed an exponential growth with an average Td value of 13.15 ± 3.64. Nc-Spain1H failed to produce an exponential pattern growth, and thus, their Td value could not be calculated. The TY_48h was also assessed to determine the number of tachyzoites produced during the same intracellular period after invasion, prior to complete tachyzoite egress from cell cultures (Figure 2B). The TY_48h value was four times higher in Nc-Spain7-infected cultures (1,663 ± 185) than in Nc-Spain1H-infected cultures (442 ± 99; p < 0.001).

Altogether, these findings demonstrated that N. caninum was able to infect, survive, and complete the parasite lytic cycle in boMØs. Remarkably, a higher active invasion rate and proliferation over time was observed for the highly virulent isolate Nc-Spain7. We therefore analyzed in the next steps whether this increased rate was due to a more severe impact of this isolate in immune function of MØs.

Upon infection of human or murine DCs, T. gondii rapidly induces cytoskeletal changes which have been linked to the migratory activation of the parasitized cell (23, 37). To analyze the cytoskeletal morphology of boMØs challenged with N. caninum and T. gondii, F-actin filaments were stained and MØs were scored according to set criteria, detailed under Materials and Methods. A dramatic impact on cytoskeletal parameters was observed upon challenge with parasites (Figure 3A), with significant differences in total scores between T. gondii- or N. caninum-infected MØs (median score = 3) and unchallenged MØs (median score = 1) (p < 0.001). BoMØs challenged with HI tachyzoites and by-stander boMØs (non-infected cells in challenged cell cultures) exhibited a median score of 2 (Figure 3B). Significant score differences were observed between groups when the morphology parameters were studied separately (Figure 3C). Infection with T. gondii resulted in cytoskeletal remodeling characterized by an increase in the percentage of cells that exhibited a rounded shape (p < 0.05), total absence of podosomes (p < 0.001), maintenance of filopodia-like extensions and a higher presence of ruffles, and/or veils (p < 0.001). For Nc-Spain7, the most frequently observed phenotype was partial loss of podosomes (p < 0.001), loss of filopodia-like extensions (p < 0.01) and a higher presence of ruffles and/or veils (p < 0.001). Significant differences were also observed in the percentage of cells that exhibited a rounded shape (p < 0.01). The morphology of Nc-Spain1H-infected MØs was chiefly characterized by total absence of podosomes (p < 0.001), an elongated or dysmorphic shape and a higher frequency of ruffles and/or veils (p < 0.001). HI and by-stander cells exhibited an elongated shape, a partial or null loss of podosomes, presence of filopodia, and an increment of ruffles/veils (p < 0.001).

Overall, we conclude that the intracellular presence of live N. caninum induced cytoskeletal remodeling in boMØs consistent with cellular activation. The morphological impact was similar to that observed for T. gondii but significantly different from the morphological changes induced by HI N. caninum or those observed in by-stander boMØs.

Toxoplasma gondii modulates the interaction of hypermigratory parasitized DCs with extracellular matrix (24, 25). To address if N. caninum infection had an impact on matrix degradation by boMØs, cells were challenged with N. caninum isolates, and the pericellular degradation of gelatin (denatured collagen) was assessed (Figure 4A). T. gondii PRU-RFP-challenged MØs were included in the assays as the positive control. The proteolytic activity of non-infected by-stander boMØs was also determined. A similar significant reduction in gelatin degradation was observed in boMØs infected with Nc-Spain7, Nc-Spain1H, and T. gondii PRU-RFP compared with unchallenged boMØs (p > 0.05). A dose-dependent (MOI) reduction in pericellular proteolysis was observed in by-stander MØs. In contrast, abrogation of the matrix proteolysis in infected boMØs was MOI-independent (Figures 4B–D).

We conclude that live intracellular N. caninum reduces or abrogates the matrix degradation capability of infected boMØs.

In murine T. gondii infections, induced hypermigration of parasitized DCs has been associated with enhanced parasite dissemination to peripheral organs (20, 22). To investigate whether N. caninum is able to exploit the migratory properties of boMØs, the motility and transmigration of boMØs challenged with Nc-Spain7 and Nc-Spain1H isolates were assessed. BoMØs challenged with the T. gondii PRU-RFP line were also included in the study. A significantly enhanced velocity was observed in boMØs challenged with both N. caninum and T. gondii parasites (p < 0.001). When comparing both N. caninum isolates, significantly higher elevated velocity values were recorded for the highly virulent isolate Nc-Spain7 (p < 0.01), which were similar to those for T. gondii (p > 0.05). In contrast, challenge of boMØs with N. caninum HI tachyzoites yielded velocities significantly lower than the baseline velocities of unchallenged boMØs (p < 0.05) (Figures 5A,B).

In transmigration assays, significantly higher transmigration frequencies were also recorded for boMØs infected with both N. caninum and T. gondii tachyzoites (p < 0.001) than for unchallenged boMØs. When comparing both N. caninum isolates, non-significant differences in transmigration frequency were found. T. gondii induced the highest enhanced transmigration, with statistically significant differences compared with Nc-Spain1H (p < 0.001) but not with Nc-Spain7 (p > 0.05). Challenge of boMØs with HI tachyzoites resulted in non-significantly altered transmigration frequency (p > 0.05) (Figure 5C).

Taken together, these results show for the first time that N. caninum induces a hypermigratory phenotype (hypermotility and enhanced transmigration) in bo MØs. A significantly higher hypermotility was observed for the highly virulent isolate Nc-Spain7 compared with Nc-Spain1H.

Having established the impact of different N. caninum isolates on boMØ morphology and cytoskeletal functionality, we next evaluated whether the infection also impact on innate immune parameters in a virulence-specific manner. To assess the ability of boMØ to kill invading N. caninum by production of ROS, differential ROS production upon N. caninum inoculation with Nc-Spain7, Nc-Spain1H, and HI tachyzoites was evaluated by flow cytometry at 1 and 24 hpi (Figure 6). At 1 hpi, a significant increase in ROS production was found in MØs inoculated with Nc-Spain1H (p < 0.05) and HI tachyzoites (p < 0.001) vs. non-infected cells but not in MØs inoculated with Nc-Spain7 (p > 0.05). However, at this time-point no significant differences were found between isolates, which both induced lower ROS generation by MØs compared with HI tachyzoites (p < 0.05). At 24 hpi, the production of ROS by Nc-Spain1H-infected MØs was enhanced (p < 0.001) compared to that of non-infected MØs as well as MØs incubated with HI tachyzoites. Interestingly however, no ROS production above either control or MØs incubated with HI tachyzoites was seen in MØs inoculated with the highly virulent isolate Nc-Spain7.

We conclude that live intracellular N. caninum reduces the ROS response of infected MØs at early infection, but only the highly virulent isolate Nc-Spain7 maintains the abrogated ROS production over time.

MØ have the ability to subsequently prime the adaptive immune response through secretion of cytokines, specifically the Th1-supporting IL-12, and the regulatory IL-10. Thus, in a next step, we investigated whether the differences in strain-specific ROS production would also be mirrored on the cytokine level. As expected, untreated MØ produced hardly any mRNA for either cytokine, whereas MØ incubated with LPS responded with a significantly enhanced mRNA expression for the pro-inflammatory cytokine IL-12 (148.7-fold; p < 0.001), and only mildly increased IL-10 mRNA production (3.6-fold; p < 0.01). Inoculation of MØ with HI tachyzoites did not increase IL-12p40 mRNA expression levels (0.99-fold; p > 0.05), and only marginally impacted on IL-10 mRNA expression (2.22-fold; p < 0.01) (Figure 7). In contrast, incubation of MØ with both types of live N. caninum tachyzoites induced a strain-specific increase in IL-12p40 mRNA and IL-10 mRNA expression compared to non-infected MØ, with the low virulent isolate Nc-Spain1H inducing more IL-12p40 (7.7-fold; p < 0.001) and IL-10 (19.6-fold; p < 0.001) than the highly virulent Nc-Spain7 isolate (4.8-fold; p < 0.01 and 14.5-fold; p < 0.001, respectively), although the differences between isolates were not statistically significant (p > 0.05).

In summary, live tachyzoites of both isolates induced IL-12p40 and IL-10 expression in boMØs. However, expression levels of both cytokines were higher for cells infected with the low virulent isolate Nc-Spain1H.

As the production of cytokines can also impact on surface antigen expression levels, we next analyzed the effects of the different N. caninum isolates on MØ activation by assessing surface antigen expression. Flow cytometry was used to quantify the expression of molecules involved in antigen presentation (CD1b, MHC Class II), T-cell activation and maturation (CD80, CD86), cell adhesion (CD11b), and phagocytosis (CD172a). As expected, MØ upregulated CD86 in response to LPS (p < 0.05). Infection of MØs with Nc-Spain7 or Nc-Spain1H resulted in a significant reduction in the percentage of MHC Class II (p < 0.05), CD86 (p < 0.05), and CD1b (p < 0.0001) expressing cells, whereas inoculation with HI tachyzoites resulted only in a significant reduction of MHC Class II (p < 0.05) and CD1b (p < 0.0001) (Table 1). The mean MFI of CD11b and CD1b diminished in MØs inoculated with Nc-Spain7, Nc-Spain1H and HI tachyzoites vs. non-infected MØs (P < 0.01–0.001; Figure 8). Significant differences between isolates were not found in the expression of any surface antigen studied.

As the previous data clearly indicated an isolate specific impact on the innate immune response created by boMØ to N. caninum isolates, we lastly assessed whether these differences would also affect the adaptive immune response. To do so, IFN-γ release by lymphocytes induced by N. caninum-infected MØs was assessed in the supernatant of co-cultures (Figure 9). As expected, the highest concentrations of IFN-γ were produced by lymphocytes when MØs were inoculated with LPS (p < 0.0001). MØs infected with Nc-Spain1H also stimulated IFN-γ production by lymphocytes, higher production than that by MØs infected with Nc-Spain7, at 48 hpi (p < 0.05) and 72 hpi (p < 0.0001). Exposure to Nc-Spain7 or HI tachyzoites did not result in significant IFN-γ variations compared with the negative control (co-culture of non-infected MØs and lymphocytes) at any time point. Samples of lymphocyte cultures reached average values of IFN-γ lower than 1,000 pg ml⁻¹, and IFN-γ was not detected in MØ cultures without lymphocytes (MØs, MØs inoculated with Nc-Spain7, Nc-Spain1H, HI tachyzoites or LPS).

Altogether, our results showed that the IFN-γ response was impaired by Nc-Spain7 infection.

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.