Ethics statement

All procedures in this study were approved by the Ethics Committee for Animal Experimentation (CETEA) of Federal University of Minas Gerais (UFMG), Brazil (Protocol # 04/2012, Protocol # 054/2012, and Protocol # 187/2014). The animals were maintained in accordance with the recommendations of the Brazilian College of Animal Experimentation guidelines (COBEA).

Animals

Wild type (WT) BALB/c and WT C57BL/6 mice were obtained from the Central Animal Facility of the Federal University of Minas Gerais, Brazil. GATA1^-/- mice were provided by the Laboratory of Immunopharmacology, Institute of Biological Sciences, Federal University of Minas Gerais, Belo Horizonte, Minas Gerais, Brazil. TLR2^-/- and TLR4^-/- mice were provided by the Laboratory of Cellular and Molecular Immunology, René Rachou Institute, Oswaldo Cruz Foundation-FIOCRUZ, Belo Horizonte, Minas Gerais, Brazil.

Experimental design

The following four independent experimental designs were used in this study depending on the objective: 1) To evaluate the kinetics of the immunological and parasitological parameters, WT C57BL/6 and BALB/c mice were divided into seven groups—a control group = non-infected (NI) and six single-infected groups (SI). 2) To evaluate the role of eosinophils in TLR-deficient mice, immunological and parasitological parameters were observed at 8 dpi. The animals were divided into two groups—a control group = non-infected (NI) and a single-infected group (SI). 3) To evaluate the effect of probiotics on immunological and parasitological parameters, animals were divided into four groups—a non-infected group treated with placebo (NI-M17), a non-infected group treated with probiotics (NI-NCDO), a single-infected group treated with placebo (SI-M17), and a single-infected group treated with probiotic (SI-NCDO). 4) To evaluate the role of eosinophils in single and multiple infections, GATA1^-/- and WT BALB/c mice were divided into three groups—a non-infected group (NI), single-infected group (SI), and re-infected group (RE). Mice from the NI groups were administered phosphate buffered saline (PBS) only. Mice from the SI group were treated with two doses of PBS followed by one dose of suspension containing 2500 fully embryonated eggs of A. suum (single-infection), maintaining an interval of 14 days between doses. Mice from the RE group were administered three doses of a suspension containing 2500 fully embryonated eggs of A. suum, maintaining an interval of 14 days between doses (re-infection). 5) To evaluate the participation of eosinophils in probiotic-mediated SIgA modulation, GATA1^-/- and WT BALB/c mice were divided into four groups—a treated with placebo and non-infected (PBS NI), treated with placebo and infected with A. suum (PBS SI), treated with NCDO (2118) and non-infected, and, treated with NCDO (2118) and infected with A. suum (NCDO SI). Tissue harvesting for parasitological and immunological evaluation was performed as previously described [9] at all time points of kinetics experiments or following the peak migration of larvae in the liver, lungs, and intestine, which were observed on days 4, 8, and 12 of infection, respectively [8]. The experimental design and results are outlined in the figures.

Bacterial growth conditions for probiotic diet

A gram-positive and non-invasive bacterial strain from Lactococcus lactis subsp. lactis (NCDO 2118) was used to assess the impact of a probiotic diet on ascariasis. Bacteria were grown as previously described [22], with some modifications. First, for experimental design 3, bacteria were cultured for 18 to 21 hours, at 30°C in M-17 broth medium (Sigma) supplemented with 0.5% glucose and 10 μg/mL of chloramphenicol (Sigma) without agitation. Each animal was given a volume of 5 mL containing either M17 medium (group M17) or freshly cultured bacteria in the stationary growth phase (NCDO group) daily. An estimated dose of 1 × 10⁹ CFU/mL bacteria per day for 22 days was ingested by each mouse [22].

For experimental design 5, the bacterial culture was prepared similarly, but without the use of antibiotics. Upon reaching the stationary phase, the culture was centrifuged at 4000 g at 4°C for 10 minutes. The supernatant was discarded, and the pellet was resuspended in PBS. This solution was administered in mice intragastrically, with the aid of a gavage needle, at a concentration of 1 × 10⁹ CFU/100μL bacteria per day for 22 days [23].

During the experimental period, the mice had unlimited access to food and water.

Parasites

Adult A. suum worms were harvested from pigs from a Brazilian slaughterhouse (Belo Horizonte, Minas Gerais, Brazil). Eggs were isolated from the uteri of female worms by gentle mechanical maceration and purified using cell strainers (70 μm). The isolated eggs were incubated with 0.2 M H₂SO₄ for embryonation, as described by Boes et al. [24]. After the 100^th day of culture, which corresponds to the peak of larval infectivity [8], fully embryonated eggs were used in experimental infections and some were allowed to hatch for harvesting larvae to obtain antigens.

Experimental infection and parasitological analysis

The mice were administered either 200 μL of PBS or 2500 embryonated eggs suspended in 200 μL PBS using a gavage needle, according to the group, time point, and experimental design. To assess the parasite burden, tissues were collected, sliced using scissors, and placed in a Baermann apparatus for 4 hours in PBS at 37°C. The recovered larvae were then fixed with 1% formaldehyde in PBS and counted under an optical microscope. Total parasite burden in the kinetic experiment was assessed as the total number of larvae recovered from the liver, lungs, and intestines at all six time points. In all other experimental designs, the parasite burden was assessed as the number of larvae recovered from the liver, lungs (parenchyma + airways), or intestine, separately.

Attainment and analysis of bronchoalveolar lavage (BAL)

Bronchoalveolar lavage (BAL) was obtained and used for the analyses of cellularity, blood loss, protein leakage, SIgA levels in the BAL, and the number of larvae in the airways. For obtaining BAL, the mice were administered a lethal dose of anesthetic (xylazine 45 mg/kg and ketamine 120 mg/kg). The animals were then tracheostomized and a 1.7 mm catheter was inserted into the trachea to obtain the BAL through two infusions and aspirations using 1 mL of PBS. The BAL obtained was filtered through 40 μm cell filters (BD, USA) to recover A. suum larvae, which were used to assess the parasite burden in the airways [9]. Filtered BAL was centrifuged at 3000 g for 10 minutes and the pellet was used to determine the total number of leukocytes and their subpopulations (macrophages, lymphocytes, eosinophils, and neutrophils) using optical microscopy. The supernatant was used to quantify blood loss, protein leakage, and SIgA levels by ELISA.

Assessment of respiratory mechanics

For assessing respiratory mechanics, mice were anesthetized using subcutaneous ketamine and xylazine injection (8.5 mg/kg xylazine and 130 mg/kg ketamine) to maintain spontaneous breathing under anesthesia, following which they were tracheostomized, placed in a body plethysmograph, and connected to a computer-controlled ventilator (Forced Pulmonary Maneuver System, Buxco Research Systems, Wilmington, North Carolina USA), as previously described by Nogueira and colleagues (2016) [9]. An average breathing frequency of 160 breaths/min was imposed on the anesthetized animal by pressure-controlled ventilation until a regular breathing pattern and complete expiration at each breathing cycle was obtained. Under mechanical respiration, the dynamic compliance (Cdyn) and lung resistance (Rl) were determined using resistance and compliance RC tests. To measure the forced vital capacity (FVC), a quasi-static pressure-volume maneuver was performed, allowing lung inflation to a standard pressure of +30 cm H₂O and then slowly exhalation until a negative pressure of -30 cm H₂O. Quasi-static chord compliance (from 0- to +10 cm H₂O) was calculated with this maneuver considering the volume/pressure of the expiration. With the fast flow volume maneuver, the lungs were first inflated to +30 cm H₂O and immediately connected to a highly negative pressure to enforce expiration until -30 cm H₂O. Forced expiratory volume (forced expiratory volume at 100 milliseconds, FEV100) was recorded during this maneuver. Suboptimal maneuvers were rejected, and for each test in every mouse, at least three acceptable maneuvers were conducted to obtain a reliable mean for all numeric parameters.

Histopathological analysis

For histopathological analysis after single-infection, mice were euthanized and the left lobe of the lung was dissected, fixed in 10% buffered formalin (pH 7.2), and embedded in paraffin. Tissue sections (5 μm thick) were mounted on glass slides and stained with hematoxylin and eosin (H&E) for inflammation analysis, Masson’s trichrome or Picrosirius Red for quantification of collagen neoformation as well as determination of the type of collagen fiber under polarized light illumination. The tissues were then imaged using a standard light microscope (Leica DM5000). For studying the morphometric parameters of inflammation, the images were analyzed from 15 randomly selected fields (total area 1.12 × 10⁶ μm²) on a single slide per animal. The segmentation allowed us to select the nuclei of all cell types present in the lung and to exclude the cell cytoplasm and other structures present in the histological sections. Using this process, a binary image was created, and the nuclei of all cell types present in the lung were counted. The difference between the number of cell nuclei present in infected mice and that in non-infected animals was considered as the number of inflammatory cells. For studying the morphometric of collagen neoformation, the images were analyzed from 20 randomly selected fields (total area 1.5 × 10⁶ μm²) on a single slide per animal. Collagen neoformation in the lung was quantified by measuring the collagenized areas present in the sections, and the difference between the collagenized areas in the infected and non-infected mice was considered as fibrosis process. All sections were viewed using a 40X objective and the images were digitized using a Leica DFC340FX microcamera associated with Leica DM5000B microscopy and analyzed using the image processing and analysis software Leica Qwin V3 (Leica Microsystems, Wetzlar, Germany).

For histopathological analyses after re-infection experiments, mice were euthanized on the 8^th day after infection, and their lungs were removed. The left lobe was fixed in a 10% solution of formaldehyde (Synth, Brazil) in PBS for 72 hours. After processing in alcohol and xylol, tissue fragments were embedded in paraffin and 4 μm thick sections were obtained and stained with H&E for morphometric quantification of inflammatory and Gomori’s trichrome for morphometric quantification of collagen deposition and fibrosis. Images from the tissue sections were collected using a 20X objective of the Axiolab-Carl Zeiss microscope (Oberkochen, Germany) and a Samsung SDC-415 micro-camera (Seoul, South Korea). Forty random images were captured with a 20X magnification lens, covering an area of 3.2 × 10⁶ μm² of lung analyzed. The KS400 software was used to select all pixels of the lung tissue in the real image for creating a binary image, digital processing, and calculation of the area in μm² of the interalveolar septum and fibrosis [25].

Tissue IgA immunostaining

To further evaluate the expression of lung IgA, we performed immunohistochemistry as previously described [26]. Briefly, tissue sections were deparaffinized, hydrated, and washed in PBS (pH 7.2). Endogenous peroxidase activity was eliminated by treatment with H₂O₂ 40 v/v solution in 0.2% PBS and non-specific binding was blocked by incubation with goat serum diluted at 1:40. The sections were then incubated with polyclonal anti-IgA (Southern Biotechnology Associates, Birmingham, USA) diluted at 1:60 for 18 hours, and then incubated with biotinylated rabbit anti-goat IgG at 1:150 (Bethyl Laboratories Inc., Montgomery, USA). The reaction was then revealed using DAB 0.05% in H₂O₂ 40% v/v in 0.2% PBS. PBS-treated sections were used as negative controls (S1 Fig). All sections were stained with H&E, desihydrated, and diaphanized for entellan (Meck Millipore) assembly.

Adult worm and infective larvae antigen production

A. suum adult worm and larval antigen production was performed as described previously [27]. Briefly, larvae antigen was used to induce the hatching of fully embryonated A. suum eggs, a standard protocol for Toxocara canis was used with some modifications [28]. Embryonated A. suum eggs were incubated in RPMI-1640 medium (SIGMA, USA; pH 7.2) supplemented with 4% penicillin/streptomycin (Invitrogen, USA), and placed in 24-well plates for larval hatching. To produce L3 larvae antigen, purified larvae were collected and transferred to a 50 mL graduated tube, centrifuged at 800 g for 10 minutes at room temperature. The supernatant was discarded and the pellet was resuspended in 5 mL of PBS, and then sonicated (Cole Parmer Ultrasonic Homogenizer Power Supply 4710 Series, USA) at 60 W for 1 minute, with a 30-second interval for each cycle, totaling 10 cycles. After sonication, the extract was centrifuged at 800 g for 15 minutes at 4°C. The supernatant was then collected and stored at -80°C until use. The amount of protein in all antigenic extracts was measured using a commercial BCA protein assay kit (Thermo Fisher Scientific, USA), according to the manufacturer’s instructions.

Adult worm antigens were initially obtained by mechanical maceration of the parasites in PBS and then using a sonicator (Cole Parmer Ultrasonic Homogenizer Power Supply 4710 Series, USA). The macerated extract was kept under ice-cooling and then sonicated at 60 W for 1 minute, with a 30-second interval for each cycle, totaling five cycles. Finally, the protocol was performed as previously described for larval antigens.

Cytokine profile

To determine the cytokine profile in the lungs, the lungs were homogenized using a tissue homogenizer (Power Gen 125 –Fisher Scientific Pennsylvania, USA) in 100 mL of PBS supplemented with protease inhibitors (0.1 mM phenylmethylsulfonyl fluoride (PMSF), 0.1 mM benzethonium chloride, 10 mM EDTA and 20 KI aprotinin A) and 0.05% Tween 20. The lung homogenates were then centrifuged at 8000 g for 10 minutes at 4°C and the supernatant was used to determine cytokine production. Production of IL-2, IL-4, IL-6, IL-10, IL-17A, IFN-γ, and TNF-α was assessed by flow cytometry (Th1/Th2/Th17 Cytometric Bead Array, BD Biosciences, USA) using a FACScan (BD Biosciences, USA) according to the manufacturer’s recommendations. IL-5, TGF-β, and IL-13 levels were measured using a sandwich ELISA kit (R&D Systems, USA) according to the manufacturer’s instructions. Finally, the intensity of the reaction was determined at a wavelength of 492 nm using an automated reader (VersaMax Tunable Microplate Reader, Molecular Devices), and the cytokine concentration (pg/mL) for each sample was calculated by interpolation using a standard curve fitted using the five- parameter logistic (5-PL).

Eosinophil peroxidase and neutrophil myeloperoxidase assays

Eosinophil peroxidase (EPO) and neutrophil myeloperoxidase (MPO) activity in lung homogenates were measured according to the method described by Strath and modified by Silveira [29,30]. The tissues were homogenized (Power Gen 125-Fisher Scientific Pennsylvania, USA), the homogenate was centrifuged at 8000 g for 10 minutes at 4°C, and the pellet obtained was used to determine EPO and MPO activity. For the EPO assay, the pellet was homogenized in 950 μL PBS and 0.5% hexadecyltrimethylammonium bromide (Sigma Chemical Co, St. Louis, MO, USA), and then frozen-thawed three times using liquid nitrogen. The lysate was then centrifuged (1500 g, 4°C, 10 minutes) and the supernatant was distributed in a 96-well microplate (75 μL/well) (Corning, USA), followed by the addition of substrate (1.5 mM OPD and 6.6 mM H₂O₂ in 0.05 M Tris-HCl, pH 8.0) (75 μL/well), and incubation for 30 minutes at room temperature. The reaction was stopped by the addition of 50 μL of 1 M H₂SO₄ and absorbance was determined at 492 nm.

For the MPO assay, the pellet obtained after centrifugation was homogenized in 200 μL of buffer 1 solution (0.1 M NaCl, 0.02 M Na₃PO₄, 0.015 M Na₂EDTA, pH 4.7), then centrifuged (1500 g, 4°C, 10 minutes). Then, 800 μL of buffer 2 solution (0.05 M NaPO₄, 0.5% hexadecyltrimethylammonium bromide) was added to the pellet and this mixture was homogenized and then frozen-thawed three times using liquid nitrogen. The lysate was centrifuged (1500 g, 4°C, 10 minutes) and the supernatant was used for the enzymatic assay. To each well of a 96-well microplate, 25 μL of supernatant was distributed (Corning, USA) followed by the addition of 25 μL of substrate TMB (3.3’-5.5—tetramethylbenzidine + 1.6 mM dimethylsulfoxide), and 100 μL of 0.5 M H₂O₂ and incubated for 5 minutes at room temperature. The reaction was stopped by adding 100 μL of sulfuric acid (1 M H₂SO₄). Finally, the intensity of the reaction was determined at a wavelength of 450 nm using an automated reader (VersaMax Tunable Microplate Reader, Molecular Devices).

Immunoenzymatic assays for evaluation of immunoglobulins

To determine Total SIgA levels, polystyrene microplates (NUNC, Roskilde, Denmark) were coated with 0.5 μg/mL Goat Anti-Mouse Ig, Human ads-UNLB (catalog no. 1040–01, Southern Biotechnology Associate Inc., USA). To evaluate Ascaris-specific immunoglobulin levels, the microplates were coated with 1 μg/mL of excretory/secretory (ES) products or adult worm antigens and incubated overnight at 4°C. Subsequently, they were washed with PBS containing 0.05% Tween 20 (Sigma Chemical Co. USA). Blocking was performed at room temperature for 2 hours with PBS + 3% BSA (Fitzgerald Industries, USA). Samples (BAL, intestinal lavage, or serum) were then added to the plates and incubated for at least 24 hours at 4°C. The plates were then washed 5 times using washing buffer (PBS-0,05% Tween20), and incubated at 37°C for 1 hour with biotinylated anti-IgA (1:500 for specific levels; catalog no. 1040–05), or (1:10000 for total levels: catalog no. 1040–08), or IgG1-3 (1:1000; catalog no. 1070–05, 1080–05, 1090–05, and 1100–05, respectively) and HRP-conjugated IgE (1:500; catalog no. 1110–05) (Southern Biotechnology Associate, Inc.), or HRP-conjugated IgM (1:1000; catalog no. 550588) (BD Pharmingen, Inc.). The plates were then washed and for distinguishing total and Ascaris-specific IgA, an additional incubation was performed using an HRP-streptavidin-conjugated solution for 2 hours (Horseradish Peroxidase) (Streptavidin-HRP, R&D Systems, USA). The plates were then incubated with substrate O-phenylenediamine (OPD, SIGMA Chemical Co., + 30% H₂O₂). Finally, the signal intensity was determined at a wavelength of 492 nm using an automated reader (VersaMax Tunable Microplate Reader, Molecular Devices).

For quality control, positive and negative controls were included on all plates. Serial dilutions of IgA standards (Mouse IgA standard, Southern Biotechnology) were used in all complete total SIgA evaluation plates. Concentrations of total SIgA (μg/mL) were determined by interpolation of the sample optical density (OD) in a standard curve fitted using the five-parameter logistic (5-PL). Levels of Ascaris-specific Igs are expressed as OD.

Statistical analyses

Statistical analyses were performed using the GraphPad Prism 8 software (GraphPad Inc., USA). The Grubb’s test was used to detect possible outliers in the groups. The Mann-Whitney test was used for comparing parasite burden and the areas of lesions in the liver and lungs. One-way ANOVA followed by Tukey´s multiple comparisons test was used to evaluate differences between groups in kinetics experiments. Pearson’s Linear Regression was used to evaluate the correlation. Data from EPO and MPO assays and pulmonary mechanics, hemoglobin, protein levels, and BAL cellularity were analyzed using the Kruskal-Wallis test followed by Dunn’s test. Finally, a two-way ANOVA with multiple comparisons was performed to assess differences between the groups as a function of time. All values were considered significant when at p ≤ 0.05.

Mucosal SIgA production correlates with eosinophilia and controls parasite burden during larval ascariasis in mice

The relationship between eosinophils and SIgA in A. suum infection model was initially evaluated by comparing immune response at different timeframes during larval migration kinetics in WT BALB/c and WT C57BL/6 mice. Immune responses associated with larval migration were assessed based on leukocyte influx in the lungs and SIgA production. After A. suum infection, there was a progressive increase in the total number of leukocytes, macrophages, lymphocytes, and eosinophils from the 8^th dpi to 12^th dpi as larvae migrate through the lungs. Whereas the number of neutrophils increased on the 8^th dpi but decreased on the 10^th and 12^th dpi (Fig 1A and S1 Table). Interestingly, we found similar kinetics of eosinophil influx (Fig 1B) and total SIgA levels (Fig 1C) in the lung mucosa, with a progressive increase after the 6^th dpi. This correlation suggests that eosinophils contribute to control of the parasite burden and is relevant to SIgA production during larval ascariasis in mice. In mouse airways, increased SIgA levels on the 8^th dpi was positively correlated with eosinophil influx (Fig 1D) and negatively correlated with the number of larvae (Fig 1E). Fig 1F shows the experimental design. WT C57BL/6 and WT BALB/c mice were divided into non-infected (n = 20) and single-infected (n = 84) groups. The data in Fig 1 represent the mean of the results of the two strains, as there were no significant differences in the responses observed between the two strains.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Fig 1. Kinetics of immune response and parasite burden during experimental ascariasis in mice.

Number of leukocytes and their cell subpopulations in bronchoalveolar lavage fluid (BAL) (n = 7) (A). Coincidence curve showing inverse relationship observed between total parasite burden and the number of eosinophils (n = 7) (B) and total SIgA levels (n = 7) (C) in BAL. Scatterplot graphics showing a positive correlation between total SIgA levels and the number of eosinophils from BAL (n = 7) (D) and a negative correlation between total parasite burden and total SIgA levels in BAL (n = 7) (E) on the8^th dpi. Schematic representation of experimental design (F). One-way ANOVA followed by Tukey´s multiple comparisons test was used to evaluate differences between groups in (A), and Pearson’s Linear Regression was used to evaluate the correlation in (D) and (E). The letters in (B) and (C) indicate statistically significant difference between mean values, where [a < b < c < d] and [a’ < b’ < c’]. Significant differences between groups (p < 0.05) are represented by the p values in A, D, and E graphs. Data are presented as mean ± SEM.

https://doi.org/10.1371/journal.ppat.1010067.g001

Apart from our objective, we also verified that there was a significant increase in the production of IgM in the BAL and serum of WT C57BL/6 mice on the 10^th dpi. However, there was no significant difference between WT BALB/c and WT C57BL/6 strains (S2 Fig).

TLR2 and TLR4 mediates mucosal SIgA production and eosinophilia that controls parasite burden during larval ascariasis in mice

After demonstrating the relationship between SIgA levels and eosinophils controlling larval migration, and after verifying the increase in the expression of TLR2 and TLR4 in single-infected mice (S3 Fig), we assessed the potential role of TLRs in triggering SIgA production and eosinophilia. Thus, we investigated the immune response and parasite burden during A. suum infection in mice deficient for Toll-like receptor 2 or Toll-like receptor 4 (TLR2^-/- or TLR4^-/-) (Fig 2A). We found that both TLR2 and TLR4 receptors are required to induce a protective immune response, as a significant reduction in eosinophil numbers in the airways was observed in TLR2^-/- and TLR4^-/- mice (Fig 2B) compared to WT mice. Furthermore, we also found lower levels of total SIgA in BAL and intestine lavage from TLR2^-/- and TLR4^-/- deficient mice, compared to that of WT mice (Fig 2C and 2D). This reduction in immune response was due to a significant increase in the number of Ascaris larvae recovered from lung tissue and airways from TLR2^-/- and TLR4^-/- mice. The most marked burden occurred in the airways of TLR4^-/- mice (Fig 2E), suggesting that TLR deficiency also affects the number of larvae with the potential to complete their life cycle in the intestine.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Fig 2. TLR2 and TLR4 receptors trigger eosinophilia and mucosal SIgA production that control larval ascariasis.

Schematic representation of the experimental design using TLR2^-/-, TLR4^-/- and WT C57BL/6 mice (A). Number of eosinophils recovered from BAL (n = 14) (B), total SIgA levels from BAL (n = 14) (C) and intestine lavage (n = 14) (D) on the 8^th dpi. Total parasite burden, that is, the number of larvae recovered in lung parenchyma and airways of WT C57BL/6, TLR2^-/- and TLR4^-/- mice on the 8^th dpi (n = 14) (E). Effect of probiotic Lactococcus lactis (NCDO 2118) on immunological and parasitological parameters during the acute phase of experimental ascariasis in mice, namely, number of eosinophils in BAL (n = 7), total SIgA (n = 7), and coincident reduction in parasite burden (n = 7) (F). Two-way ANOVA followed by Sidak´s multiple comparisons test was used to evaluate differences between groups (B-E) and the Mann-Whitney test was used to assess differences between the groups (F). Differences were considered significant at p ≤ 0.05 and are represented by symbols in the graphs. * represents differences between knockout and WT C57BL/6 mice groups and # represents differences between non-infected and single-infected groups from same strain (B, C, and D); and, the * represents differences between the single-infected and non-infected group and & represents differences between the NCDO and M17 groups (F). [1 symbol = p < 0.05], [2 symbols = p < 0.01], [3 symbols = p < 0.001]. Data are presented as mean ± SEM.

https://doi.org/10.1371/journal.ppat.1010067.g002

Considering the relevant role of TLRs in A. suum infection, we investigated the use of probiotics in inducing a response mediated by TLR receptors in the course of infection. The bacterium Lactococcus lactis (NCDO 2118) is a gram-positive bacterium that has lipoteichoic acid (LTA) on its surface. These PRRs are conserved molecular structures that can be recognized by TLR2 [31]. We found an increase in SIgA production and the number of eosinophils, as well as a concomitant reduction in parasite burden at pulmonary sites in mice treated with the probiotic bacteria during infection with A. suum when compared to mice that received the placebo (M17) (Fig 2F). This finding suggests that treatment with probiotics may stimulate receptors responsible for modulating eosinophilia and SIgA levels, which have a role in parasite control in single-infected mice. We also assessed, Ascaris’ potential to modulate the expression of these receptors after multiple exposures (S3 Fig).

Eosinophils are required for mucosal SIgA production and parasite burden control during larval ascariasis in mice

To further investigate the role of eosinophils in SIgA production during larval ascariasis, we evaluated the immune response to Ascaris infection in non-eosinophilic mice (GATA1^-/-) and compared it to that of wild-type mice (WT BALB/c) (Fig 3A). The absence of eosinophils leads to the loss of parasite burden control in GATA1^-/- mice (Fig 3B). The role of eosinophils in regulating the levels of total and A. suum antigen-specific SIgA was established by the increase in eosinophils in WT BALB/c mice BAL, which was not observed in GATA1^-/- mice (Fig 3C). BAL from WT BALB/c and GATA1^-/- mice showed a significant increase in total SIgA levels on the 8^th dpi, however, this increase was A. suum antigen-specific in only WT BALB/c mice. This demonstrates the importance of eosinophils in the control of parasites (Fig 3B) and reinforces the important role of eosinophils in the modulation of the humoral immune response during A. suum infection in mice.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Fig 3. Eosinophils are required for SIgA production and parasite burden control in mice.

Schematic representation of the experimental design of a single-infection model using WT BALB/c and GATA1^-/- mice (eosinophil deficient) (n = 5) (A). Number of larvae recovered from the liver, lung, and intestine (B) of WT BALB/c and GATA1^-/- mice on the 4^th, 8^th, 10^th, and 12^th dpi. Levels of total and A. suum antigen-specific SIgA from BAL on the 8^th dpi (C). T-test was used to assess differences between the groups in (B); one-way ANOVA followed by Tukey´s multiple comparisons test was used to evaluate differences between groups in (C). Differences were considered significant at p ≤ 0.05 and are represented by symbols in the graphs. * represents differences between the single- and non-infected group’s mice of the same strains, and & represents differences between the WT and GATA1^-/- mice with the same treatment [1 symbol = p < 0.05], [2 symbols = p < 0.01]. Data are presented as mean ± SEM.

https://doi.org/10.1371/journal.ppat.1010067.g003

Eosinophils are essential in modulating SIgA production during Ascaris suum infection

To evaluate the role of eosinophils in SIgA modulation after treatment with probiotic L. lactis (NCDO 2118), GATA1^-/- and WT BALB/c mice were treated for 22 days with probiotic and infected with A. suum (Fig 4A). We found that GATA1^-/- NI mice treated with placebo showed high levels of total SIgA in the BAL when compared to WT BALB/c NI and GATA1^-/- SI group mice treated with placebo. Interestingly, GATA1^-/- SI mice treated with placebo showed lower total SIgA levels (Fig 4B and S2 Table).

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Fig 4. Eosinophils are important in probiotic modulation after infection.

Schematic representation of single-infection model using WT BALB/c and GATA1^-/- mice after treatment with Lactococcus lactis NCDO (2118) (n = 6) (A). Total SIgA levels in BAL on the 22^nd day after treatment with probiotic and 8 days after infection (n = 6) (B). Number of larvae recovered from WT BALB/c and GATA1^-/- mice lungs (C). Differences between groups were statistically evaluated by the following tests: Two-Way ANOVA followed by Sidak´s multiple comparisons test was used to evaluate differences between groups in (B); One-Way ANOVA followed by Tukey´s multiple comparisons test was used to evaluate differences between groups in (C). Differences were considered significant at p ≤ 0.05 and are represented by symbols in the graph (C) where the * = p < 0.05. Data are presented as mean ± SEM.

https://doi.org/10.1371/journal.ppat.1010067.g004

We found an increase in total SIgA levels in the WT BALB/c NI group treated with NCDO than in the WT BALB/c NI group treated with placebo. This indicates the efficiency of L. lactis in inducing the production of total SIgA. Finally, our findings showed that GATA1^-/- SI mice treated with NCDO had higher levels of total SIgA than GATA1^-/- SI mice treated with placebo, suggesting that probiotic treatment contributes to the maintenance of high levels of total SIgA in the BAL in the absence of eosinophils (Fig 4B and S2 Table).

By analyzing the parasite burden on the 22^nd day after treatment with probiotics and 8 dpi (Fig 4C), we found that GATA1^-/- SI mice treated with placebo showed higher parasite burden than in WT BALB/c SI mice treated with placebo. Whereas GATA1^-/- mice treated with NCDO showed reduced parasite burden than GATA1^-/- mice treated with placebo (Fig 4C). Interestingly, no significant differences were observed between wild-type mice.

Together, these results confirm that in the absence of eosinophils, the ability of mice to control parasite burden as well as post-infection SIgA levels reduces. However, the absence of eosinophils did not affect SIgA production under basal conditions or infection, suggesting that eosinophils play different roles under different conditions.

Eosinophils are associated with inflammation, tissue remodeling, and pulmonary dysfunction caused by larval migration during A. suum single-infection

In addition to contributing to the control of parasite burden, eosinophils are important in pulmonary pathophysiology during larval ascariasis, and consequently, fibrosis modulation [10]. Therefore, the role of eosinophils in the modulation of pulmonary inflammation following Ascaris infection was assessed through the number of innate cells, cytokine profile, and eosinophil and neutrophil activity in lung tissue between non-infected and single-infected WT BALB/c and GATA1^-/- mice. Qualitative analyses of leukocyte influx into the lungs during larval migration in WT BALB/c mice revealed neutrophils to the most abundant at the peak of larval migration (8^th dpi). However, eventually (12^th dpi), the inflammatory infiltrate was more pronounced, mostly composed of mononuclear cells and eosinophils (Fig 5A). Similarly, quantitative morphometric analyses confirmed that Ascaris larval migration induced significant pulmonary cell infiltration on the 8^th and 12^th dpi (243 ± 48 vs. 484 ± 95cells/74588.6 μm² of tissue) (Fig 5B), with progressive accumulation of eosinophils up to the 12^th dpi (Fig 5C). In contrast, GATA1^-/- mice showed a significant reduction in the number of leukocytes in the airways (Fig 5D) and a reduction in inflammatory cell accumulation in the lung tissues at 12^th dpi than in WT BALB/c mice (330 ± 20 vs. 484 ± 95cells/74588.6 μm² of tissue).

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Fig 5. Eosinophils contribute to pulmonary inflammation during A. suum single-infection in mice.

Histopathological analysis of the lesion caused by larval ascariasis and lung inflammation in WT BALB/c and GATA1^-/- mice on the 8^th and 12^th dpi (n = 6). Representative microphotography of lung tissue stained with H&E (Bar scale = 50 μm) (A-D). Number of total leukocytes recovered from BAL on the 8^th dpi (n = 6) (E). Number of eosinophils in situ in the lung at 0, 4, 8 and 12 dpi (n = 5) (F). Optical density representing Eosinophil peroxidase (EPO) and neutrophil myeloperoxidase (MPO) activity in the lung at 0, 4, 8, and 12 dpi (n = 12) (G and H). Levels of cytokines in the lung at the 12^th dpi (n = 6) (I—Q). Data are represented as mean ± SEM. One-way ANOVA followed by Tukey´s multiple comparisons test was used to evaluate differences between groups (G, I, J, K, L, N, O, and P). Kruskal-Wallis test followed by Dunn´s multiple comparisons test was used to evaluate differences between groups (H, M, and Q). The p values are represented by symbols in the graphs wherein * represents differences between the non-infected group, * represents differences between same groups in the 8 dpi, # represents differences between groups from different strains that received the same treatment. [1 symbol = p < 0.05], [2 symbols = p < 0.01], [3 symbols = p < 0.001] and [4 symbols = p < .0001].

https://doi.org/10.1371/journal.ppat.1010067.g005

The results revealed that larval migration in lungs induced a marked increase in cytokine and granule production in the pulmonary parenchyma on the 12^th dpi in WT BALB/c mice than in non-infected mice (Fig 5E–5Q). However, GATA1^-/- mice showed a marked ablation of EPO (Fig 5G) and reduced MPO levels (Fig 5F), mainly on the 12^th dpi than in single-infected WT BALB/c mice. In addition, at the 12^th dpi, there was a significant increase in IL-4, IL-10, IFN-γ, and TNF-α levels (Fig 5L, 5N, 5P, and 5Q) in single-infected WT BALB/c group but not in single-infected GATA1^-/- mice. Whereas an increase in IL-5 and IL-13 was observed mainly in GATA1^-/- SI mice (Fig 5K and 5M). Collectively, our data support that eosinophils contribute to lung inflammation during larval ascariasis in mice.

Despite the increased levels of mediators, such as IL-4 and IL-13, we further investigated tissue fibrosis and pulmonary mechanical dysfunction. Histological analysis performed at the 8^th dpi showed normal collagen levels in the lung tissue of A. suum single-infected GATA1^-/- and WT BALB/c mice, while tissue fibrosis was observed at the 12^th dpi in both groups. (Fig 6A–6D). Tissue remodeling-induced larval migration lead to pulmonary dysfunction, which was more pronounced in GATA1^-/- mice than in WT BALB/c mice. We observed that larval migration through the lungs induced a significant reduction in pulmonary volumes, as depicted by (FVC), inspiratory capacity (IC), and forced expiratory volume (FEV100) at 8^th dpi with partial recovery at 12^th dpi in WT BALB/c mice (Fig 6E, 6H and 6I); however, GATA1^-/- mice showed a marked decline in pulmonary volumes at 12th dpi (Fig 6E and 6I). Similarly, increase in lung resistance (Rl) and dynamic compliance (Cdyn) decline revealed that GATA1^-/- infected mice¸ compared to WT infected mice, showed a progressive loss in pulmonary elasticity (Fig 6F and 6G), suggesting that excessive larval migration could induce increased tissue damage and trigger excessive fibrosis.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Fig 6. Pulmonary mechanical dysfunction and fibrosis induced by larval ascariasis in the lungs.

Representative photomicrographs of single-infected WT BALB/c and GATA1^-/- mice (n = 5): Fibrosis analysis using Masson’s trichrome technique (stains collagen fibers blue) and Picrosirius Red (stains collagen fibers red) performed at the 8^th dpi (A, J and B, K) showing normal histological appearance with few collagen fibers and at 12^th dpi (C, L, and D, M) showing increase in collagen deposition in single-infected WT BALB/c and GATA1^-/- mice. Histological analysis at the 12^th dpi using Picrosirius Red with polarized light showing increased deposition of type I collagen fibers (stained red, P and Q), distinct from type III collagen observed at the 8^th dpi (stained green and yellow, N and O). Assessment of invasive pulmonary spirometry of Ascaris single-infected WT BALB/c and GATA1^-/- mice through the following parameters: Forced Vital Capacity, FVC (E), Lung Resistance, Rl (F), Dynamic Compliance, Cdyn (G), Inspiratory Capacity, IC (H), Forced Expiratory Volume at 100 msec, FEV100 (I). Data are represented as mean ± SEM. Significant differences between GATA1^-/- and WT BALB/c groups at the 12^th dpi are represented in the graph by p values. P values are represented by symbols in the graphs wherein * represents differences between non-infected groups from the same strain, * represents differences between the single-infected groups from the same strain, # represents differences between groups from different strains that received the same treatment. One-way ANOVA followed by Tukey´s multiple comparisons test was used to evaluate differences between groups. [1 symbol = p < 0.05], [2 symbols = p < 0.01], [3 symbols = p < 0.001] and [4 symbols = p < 0.0001]. Bar scale = 50 μm.

https://doi.org/10.1371/journal.ppat.1010067.g006

To further characterize the predominance of collagen types in lung tissue, we used Masson’s trichrome staining and Picrosirius Red, with or without polarized light. We observed a normal histological appearance with few collagen fibers in both WT BALB/c and GATA1^-/- mice groups at the 8^th dpi (Fig 6J and 6K). Interestingly, an increase in collagen deposition was observed in single-infected WT BALB/c mice at the 12^th dpi (Fig 6L) and was even more pronounced in GATA1^-/- mice at the 12^th dpi (Fig 6M). However, analysis with polarized light on Picrosirius Red showed increased deposition of type I collagen at the 12^th dpi (stained red, Fig 6P and 6Q) instead of type III collagen observed at the 8^th dpi (stained green, Fig 6N and 6O).

Eosinophils contribute to parasite burden control and SIgA production during larval ascariasis after multiple exposures

Having highlighted the importance of SIgA in controlling parasite burden as well as its correlation with the number of eosinophils, we used a re-infection model previously established by our group [9], which provided evidence of protection after re-infection. Thus, in the experimental design, we compared single-infected (SI) WT BALB/c and GATA1^-/- groups to re-infected (RE) WT BALB/c and GATA1^-/- groups (Fig 7A) at the 8^th dpi, which correspond to the pulmonary phase of Ascaris larval migration. Initially, we observe decreased parasite burden in WT BALB/c re-infected mice. We found that eosinophil deficiency in GATA1^-/- mice in both single- and re-infected groups reduced parasite burden control when compared to the WT BALB/c groups (Fig 7B). Furthermore, parasite numbers in different lung compartments (airways and tissue), showed that more larvae transmigrated into the airways of GATA1^-/- than in the WT BALB/c re-infected group (Fig 7B), suggesting that eosinophils may control the parasite cycle in mice.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Fig 7. Eosinophils contribute to parasite burden control and SIgA production during larval ascariasis after multiple exposures.

Experimental design of A. suum multiple infections (A). Number of larvae in WT BALB/c and GATA1^-/- mice lung parenchyma and bronchoalveolar lavage (BAL) (n = 14) (B). Number of total leukocytes and their subpopulations, namely, lymphocytes, macrophages, neutrophils, eosinophils, and total protein and hemoglobin levels in BAL (n = 6) (C). All immunological and parasitological parameters were evaluated at the 8^th dpi. Data are represented as mean ± SEM. Significant differences (p < 0.05) between are represented in the graph by p symbols. * represents differences between non-infected groups from the same strain, * represents differences between the single-infected groups from the same strain, # represents differences between groups from different strains that received the same treatment. [1 symbol = p < 0.05], [2 symbols = p < 0.01], [3 symbols = p < 0.001] and [4 symbols = p < .0001].

https://doi.org/10.1371/journal.ppat.1010067.g007

We found no eosinophils in the BAL in GATA1^-/- mice, whereas the WT BALB/c re-infected group showed a significant increase in eosinophils (Fig 7C). In addition, re-infected WT BALB/c and GATA1^-/- mice showed a significant increase in the total number of leukocytes, macrophages, neutrophils, and lymphocytes than NI and SI BALB/c and GATA1^-/- mice. However, the increase in lymphocyte numbers was more accentuated in the GATA1^-/- re-infected group (Fig 7C). BAL evaluation also showed increased levels of total protein and hemoglobin in the airways of single- and re-infected WT BALB/c and GATA1^-/- mice but was more pronounced in the GATA1^-/- single-infected group (Fig 7C). There was a positive correlation between total protein and hemoglobin levels in the airways (r² = 0.666, p = 0.0001), probably due to hemorrhage caused by larval migration.

The relevance of eosinophils and SIgA levels in the lung mucosa and their ability to control parasite burden was more evident in the re-infection model, as in contrast to high levels of total and A. suum antigen-specific SIgA levels (against antigens from adult worms and larvae of A. suum) in WT infected mice, the total and specific SIgA levels remained low in eosinophil-deficient mice (GATA1^-/-) even after re-infection (Fig 8A). To reinforce this relationship, we assessed lung tissue of mice for IgA-positive cells and found A. suum induced high IgA immunostaining, predominantly in the peribronchial spaces and in the bronchial epithelium to be more prominent in re-infected WT BALB/c mice (Fig 8F) than in single-infected WT BALB/c mice (Fig 8D). The single-infected (Fig 8E) and re-infected GATA1^-/- mice (Fig 8G) showed little to no staining.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Fig 8. Effect of eosinophil deficiency in IgA production in lungs after larval ascariasis.

Total SIgA and Ascaris-specifics SIgA levels in bronchoalveolar lavage (BAL) (n = 5) (A). Representative microphotography of IgA immunohistochemistry in lung tissues (B-G) showing immunostained cells predominantly in the peribronchial spaces and the bronchial epithelium. Non-infected WT BALB/c (B), non-infected GATA1^-/- (C), single-infected WT BALB/c (D), single infected GATA1^-/- (E), re-infected WT BALB/c (F), and re-infected GATA1^-/- (G) (n = 6). IgA is marked in brown. Arrows indicate positive markings on the bronchial epithelium. Bar scale = 40 μm).

https://doi.org/10.1371/journal.ppat.1010067.g008

Furthermore, we measured specific antibodies against Ascaris-antigens IgM, IgG1, IgG2A, IgG2B, and IgG3 and found that reinfected WT BALB/c and GATA1^-/- mice showed a significant increase in the serum antibodies, indicating that the absence of eosinophils did not compromise specific antibody production (S4 Fig). These data suggest that despite increased production of IgM, IgG1, IgG2A, IgG2B, and IgG3 antibodies, the absence of eosinophils and reduction in SIgA compromised the infection control. Specific IgE levels were not detected.

Eosinophils are critical for chronic pulmonary inflammation and dysfunction recovery after multiple parasite exposure in mice

Evaluation of tissue injury caused by larval migration and lung inflammation revealed that all infected animals (subjected to single and multiple infections) from both mouse strains presented microscopic lesions in the parenchyma characterized by septal thickening, hemorrhage, vascular hyperemia, perivascular edema, and inflammation of the pulmonary parenchyma (Fig 9A–9H). However, these findings were more prominent in re-infected groups regardless of mouse strain, corroborating the results of morphometric evaluations of the lesion area. The lesion area did not differ between groups that received the same treatment, however, lesion characteristics differed between groups. Hemorrhage was more pronounced in single-infected groups from both mouse strains than in the GATA1^-/- mice re-infected group, reinforcing that the injuries caused by larval migration are responsible for increased hemorrhage in the lungs. The inflammatory infiltrate was more prominent in the WT BALB/c and GATA1^-/- re-infected groups. However, re-infected WT BALB/c mice showed mixed inflammatory infiltrate (with neutrophil and eosinophil predominance). These findings corroborate the finding of EPO and MPO assays that showed a significant increase in eosinophil and neutrophil activity in WT BALB/c RE mice when compared to the other groups (Fig 9I). Single- and re-infected GATA1^-/- mice had blunted EPO levels and reduced MPO levels from tissue neutrophils and discrete bronchus-associated lymphoid tissues (BALT) formation.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Fig 9. Effect of eosinophil deficiency on chronic pulmonary inflammation induced by multiple A. suum infections.

Representative microphotography of lungs tissues stained with H&E and observed on the 8^th dpi at low magnification (Bar scale = 100 μm) (A-D) and at high magnification (Bar scale = 50 μm) (E-H), showing histopathological lesions in the single-infected group of WT BALB/c (A and E) and GATA-/- (B and F) (n = 6), and re-infected WT BALB/c (C and G) and GATA1^-/- groups (D and H) (n = 6). Thickened septa are indicated with arrowheads, inflammatory infiltrates are indicated with ‘#’, hemorrhage is indicated with ‘*’. Optical density represents eosinophil peroxidase (EPO) and neutrophil myeloperoxidase (MPO) activity in the lung tissue at the 8^th dpi (n = 10) (I). Tissue cytokine profile in the 8^th dpi (n = 10) (J). Differences between groups were statistically evaluated using one-way ANOVA followed by Tukey´s multiple comparisons test for the data represented in (I) and the Kruskal-Wallis followed by Dunn’s multiple comparisons test in the data represented in (J). Data are represented as mean ± SEM. P values are represented by symbols in the graphs wherein * represents differences between non-infected groups, * represents differences between single-infected groups, # represents differences between groups from different strains that received the same treatment. [1 symbol = p < 0.05], [2 symbols = 0.01 > p > 0.05], [3 symbols = 0.001 > p > 0.01] and [4 symbols = p < 0.0001].

https://doi.org/10.1371/journal.ppat.1010067.g009

Th1/Th2/Th17 type cytokines were measured to further characterize lung inflammation (Fig 9J). We found a significant increase in IL-2, IL-4, IL-10, and IFN-γ in the WT BALB/c single- and re-infected groups when compared to non-infected mice, but not in the GATA1^-/- single- and re-infected groups. Inflammatory cytokines, such as IL-6, IL-17A, and TNF, were increased in WT BALB/c and GATA1^-/- single- and re-infected groups, when compared to the respective non-infected groups. IL-5 levels were significantly increased in the GATA1^-/- and WT BALB/c single-infected mice, but more pronounced in GATA1^-/- single-infected mice. No differences in IL-13 and TGF-β1 levels were observed between WT and GATA1^-/- single- and re-infected mice (Fig 9J).

Next, we evaluated pulmonary fibrosis in WT BALB/c and GATA1^-/- mice in the single and re-infected groups (Fig 10). Histopathological evaluation of lung tissue sections stained with Gomori’s trichrome was performed to determine the chronic tissue remodeling and fibrosis induced by A. suum larvae migration. Fibrosis was evident in some areas of alveolar septa, bronchi, bronchioles, and pulmonary blood vessels, which contributed to the thickening of the alveolar septum in single-infected and re-infected mice from both strains (Fig 10) but was more pronounced in re-infected WT BALB/c mice (Fig 10C and 10F). Re-infected GATA1^-/- mice showed larger injury areas than GATA1^-/- single-infected mice (Fig 10B, 10D, and 10E). Interestingly, corroborating the findings of morphometric analyses, re-infected WT BALB/c mice exhibited large areas of consolidated fibrosis, causing considerable thickening of the alveolar septa, and compromising a larger number of bronchi, bronchioles, and vessels than that observed in GATA1^-/- mice (Fig 10C–10F). Invasive spirometry confirmed pulmonary lesions caused by larval migration, inflammation, and fibrosis, which were reflected in pulmonary mechanical dysfunction (Fig 10G). Re-infected WT BALB/c mice did not show worsening lung resistance and dynamic compliance (Fig 10C, 10F and 10G), however, re-infected GATA1^-/- mice showed an increase in lung resistance and decreased dynamic compliance resulting from chronic pulmonary inflammation (Fig 10D–10G).

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Fig 10. Effect of eosinophil deficiency in lung injury, fibrosis, and dysfunction induced by multiple larval ascariasis.

Representative microphotography of fibroplasia/fibrosis histopathological analysis from lung tissues stained with Gomori’s trichrome (A-D) showing lesions and inflammation caused by larval ascariasis in single-infected (A-B) compared to re-infected (C-D) WT BALB/c (A and C) and GATA1^-/- (B and D) mice (n = 6). Fibrosis is marked in bluish green. Bar scale = 50 μm. The area of fibrosis is represented by ‘*’. Bar graphs from morphometric analysis of the area of the lesion (E) and fibrosis (F) caused by larval migration and lung inflammation (n = 6). Assessment of lung mechanics in mice after single and multiple infections with A. suum (G) (n = 6). Data are represented as mean ± SEM. One-way ANOVA followed by Tukey´s multiple comparisons test was used to evaluate differences between groups in (E) and (F). Kruskal-Wallis test followed by Dunn´s multiple comparisons test was used to evaluate differences among groups in (G). Statistical differences are represented by symbols in the graphs, where * represents differences observed between non-infected groups, * represents differences between single-infected groups, # represents differences between groups of different strains that received the same treatment. [1 symbol = p < 0.05], [2 symbols = 0.01 > p > 0.05], [3 symbols = 0.001 > p > 0.01] and [4 symbols = p < 0.0001]. All experiments were analyzed at the 8^th dpi.

https://doi.org/10.1371/journal.ppat.1010067.g010