C. albicans CA1398 carrying the ACT1p-gLUC59 fusion (gLUC59) was used (Enjalbert et al., 2009). The gLUC59 luciferase reporter has previously been described (Enjalbert et al., 2009). C. albicans gLUC59 (BLI-Candida) was cultured in yeast peptone dextrose (YPD) as described by Solis and Filler (2012). The C. albicans CA 1398 carrying the ACT1p-gLUC59 fusion and its parental strain CA 1938 were equally pathogenic as previously demonstrated (Enjalbert et al., 2009).

The products studied were provided by Gnosis by Lesaffre (Marcq-en-Baroeul, France). S. cerevisiae live yeast (referenced as GI) is a proprietary, well-characterized strain of Lesaffre, registered in the French National Collection of Cultures of Microorganisms (CNCM) under the number I-3856. The S. cerevisiae species is characterized by using phenotypic (API®ID32C, Biomerieux SAS) and genotypic referenced methods (genetic amplification and sequencing of 26S DNA; Kurtzman and Robnett, 1997, 1998). Moreover, the strain CNCM I-3856 has been characterized by polymerase chain reaction (PCR) Interdelta typing techniques (European Committee for Standardization, 2009), and its genome has been sequenced. The inactivated yeast S. cerevisiae CNCM I-3856 (referenced as IY) is a primary grown dried whole yeast, obtained by drum drying of S. cerevisiae CNCM I-3856 and inactivated through the process of drying. Furthermore, the strain of L. rhamnosus GG (referenced as G 250) is registered in the American Type Culture Collection (ATCC) under the number 53103. The CFU counts of the probiotics used in these experiments are at least of 5 × 10⁹ CFU/g for GI and 2.5 × 10¹¹ CFU/g for G 250. All products were used at 100 mg/ml.

Female C57BL/6J mice from Charles River (Calco, Italy) were used at 6–8 weeks of age. Mice were treated subcutaneously with 225 mg/kg cortisone acetate (Sigma-Aldrich) every 2 days starting from 1 day before infection until the end of experiment. After anesthesia with a subcutaneous injection of a mixture of Tiletamine/Zolazepam-Xylazine (50–5 mg/kg) (Mosci et al., 2013) mice were infected with 1 × 10⁶ CFU/ml BLI-Candida suspension as previously described (Solis and Filler, 2012).

Then mice were treated sublingually with 10 μl of saline, FLZ (4 mg/ml), GI, IY, or G 250 (all 100 mg/ml) on days +1, +2, +3, and +6 post-infection. The oral cavity was swabbed just before the infection and streaked on YPD agar plus chloramphenicol (50 μg/ml; both from Sigma-Aldrich) to verify the absence of Candida spp.

Starting on day 1 after challenge and at each selected day, 10 μl (0.5 mg/ml in 1:10 methanol:H₂O) of coelenterazine (Synchem, OHM) was added sublingually. Mice were then imaged in the IVIS Lumina XRMS Imaging system (Perkin Elmer) under subcutaneous anesthesia. The total photon emission was quantified as previously described (Mosci et al., 2013). In the selected experiments, an ex vivo analysis of esophagus, stomach, kidneys, liver, and feces from OPC mice was performed at day +8 post-infection as previously described (Mosci et al., 2014). The C. albicans burden of the tongue, esophagus, stomach, and duodenum was evaluated at days +3, +6 and +8 post-infection as previously described (Mosci et al., 2014).

The animals infected and treated with saline, FLZ, IY, GI, or G 250 were sacrificed at day +8 post-infection to analyze gross and histopathologic lesions, and tongues were excised. Histological examination was, also, performed on tongues of uninfected mice treated with saline, FLZ, GI, IY, or G 250 to evaluate the integrity of the tissue and neutrophil recruitment. The tissues were fixed immediately in 10% formalin and then embedded in paraffin. The tongues were sectioned longitudinally to evaluate the extension of the lesions. The 3–5 μm thick sections were stained using the periodic acid-Schiff (PAS) procedure to detect fungi and examined by light microscopy (Leica DM2500). The scale bars are in micrometer.

At day +6 post-infection tongue homogenates from mice infected and treated with saline, FLZ, GI, or IY were centrifuged at 3,000 rpm for 5 min. Then cellular fractions were lysed using Trizol reagent (Life Technology). Total RNA was extracted and retro-transcribed by using the Moloney murine leukemia virus reverse transcriptase reaction (M-MLV RT) as described in the manufacturer’s instructions. cDNA concentration was determined using a spectrophotometer. C. albicans ACT1, SAP2, SAP6, and ALS3 gene transcription was detected by using primers reported in the literature (Naglik et al., 2008; Roudbarmohammadi et al., 2016). Real-time quantitative PCR (RT-qPCR) was performed in 96-well PCR SYBR green plates (all from BioRad) using 200 ng of cDNA for each sample. In this way, the amount of genes expressed by C. albicans gLUC59 during infection was independent by fungal burden. All samples were measured in triplicate. The relative level of Candida gene expression was reported as 2-∆∆Ct relative to transcripts of C. albicans inoculum (basal gene expression level of Candida at day 0). Briefly, 3 days before infection, a colony of the C. albicans gLUC59 was added into 10 ml of YPD broth and incubate in a 30°C shaker. The next day, 100 μl of the overnight culture were transferred to 10 ml fresh YPD broth and incubate in a 30°C shaker overnight. This step was repeated one more time. The next day, C. albicans gLUC59 was recovered, counted, and diluted at desired concentration (Solis and Filler, 2012). Amplification conditions used were the same for ACT1, ALS3, SAP2, and SAP6 genes: 3 min at 95°C, 40 cycles of 10 s at 95°C and 30 s at primer specific annealing temperature. The experiments were performed using the Eppendorf Master cycler.

At day +6 post-infection, peritoneal neutrophils of mice orally infected and treated as described in section “Materials and Methods” with saline, FLZ (4 mg/ml), GI, or IY (both at 100 mg/ml) were collected 18 h after the intraperitoneal injection of 0.5 ml endotoxin-free 10% thioglycolate solution (Difco) The percentage of neutrophils was >90% (Mosci et al., 2014). The oxidative burst of neutrophils was detected by labeling cells (4 × 10⁶/ml) with 1 μM of 2′,7′-dichlorofluoresceindiacetate (DCFH-DA) for 30 min at room temperature. Cells were then incubated with C. albicans (CA-6) (2 × 10⁶ CFU/ml) into a black 96-well plate (Nunc), and the emission of fluorescence was measured as previously described (Mosci et al., 2014; Ricci et al., 2019). Quantification of reactive oxygen species (ROS) production was determined by calculation of the area under curve (AUC) (Ermert et al., 2009; Kenno et al., 2016). Killing activity of neutrophils was determined by CFU assays. Briefly 0.1 ml/well of neutrophils (10⁵ cells) was immediately incubated in flat-bottom 96-well microtiter tissue culture plate with 0.1 ml/well (10⁴ cells) of C. albicans (CA-6) in RPMI-1640 plus 5% FCS for 2 h at 37°C plus 5% CO₂. After incubation, plates were vigorously shaken, and cells were lysed by adding Triton X-100 (0.1% in distilled water; final concentration in the well was 0.01%). Serial dilutions were prepared in distilled water for each well. The samples were then spread on Sabouraud dextrose agar plus chloramphenicol (50 μg/ml) in triplicate, and CFU counts were determined after 24 h of incubation at 37°C. Control cultures consisted of C. albicans (CA-6) incubated in RPMI-1640 plus 5% FCS without effector cells. Killing activity was expressed as the percentage of CFU reduction with respect to neutrophils from uninfected mice and according to the following formula: % killing activity = 100 – (CFU experimental/CFU control) × 100.

GraphPad Prism 7.0 software was used for all statistical analysis presented. Data are reported as boxplot graphics with median and the 25th and 75th percentile or mean ± SEM from the experiments indicated in each figure legend. Statistical analysis was performed using Mann-Whitney U test. p < 0.05 was considered as significant.

To analyze whether inactivated S. cerevisiae (IY), live S. cerevisiae (GI), or live L. rhamnosus GG (G 250) were able to affect the course of oropharyngeal candidiasis, C57BL/6J mice were treated subcutaneously with cortisone acetate every 2 days starting from 1 day before infection until the end of the experiment. Briefly, the mice were infected with bioluminescent (BLI) C. albicans (1× 10⁶ CFU/ml; 10 μl/mouse) as previously described by Solis and Filler (2012). The animals were then treated sublingually with saline, FLZ (4 mg/ml; 10 μl/mouse), IY, GI, or G 250 (all 100 mg/ml; 10 μl/mouse) on days +1, +2, +3, and +6 post-infection. Saline-treated and FLZ-treated infected mice were used as negative and positive controls, respectively. The effect of the various treatments on oral candidiasis was evaluated at days +1, +3, +6, and +8 post-infection by measurement of photon emission from the oral cavity as well as by determination of CFU counts on the tongue. As shown in Figure 1, both IY and GI significantly decreased the BLI-C. albicans load in the oral cavity at each time point tested, both as visually evident (Figure 1A) and as measured by total photon emission (Figure 1B). Of note, the anti-Candida effect of IY and GI was comparable to the effect of positive control FLZ at each time point tested. By contrast, no effect was observed for G 250 (Figure 1B). The reduction of the Candida load in the oral cavity was also confirmed through a significant decrease in CFU counts on the tongues of IY- and GI-treated infected animals (Figure 2A). In order to characterize the lesions associated with the OPC, we performed histological analysis of the tongues at day +8 post-infection. The fungal invasion of the tongue of saline-treated infected mice was limited to a portion on the surface of the tongue due to the formation of pseudomembranes. The pathogenic fungi and the inflammatory cells appeared to be confined to the keratinized layer of the tongue where micro abscesses formation could be observed. No Candida cells and Candida-induced lesions were found in the tongues of FLZ, IY, and GI-treated infected mice (Figure 2B), while Candida-induced lesions were observed in the tongue of G 250-treated infected mice (Figure 2B). Histological examination of the tongues of uninfected mice treated with saline, FLZ, IY, GI, or G 250 showed complete tissue integrity and an absence of neutrophils recruitment (Figure 2B).

The progression of the infection to esophagus, stomach, and intestine was evaluated by CFU counts at days +3, +6, and +8 post-infection. The results show that both IY and GI were able to reduce the esophageal Candida load only at day +8 post-infection (Figure 3A). In the stomach, the beneficial effect of GI was evident at both days +6 and +8 post-infection, while the effect of IY was appreciable only at day +8 post-infection (Figure 3B). In the duodenum, only GI induced a significant inhibition of the C. albicans load at days +6 and +8 post-infection (Figure 3C). By contrast, G 250 had no significant effect (Figure 3). Notably and as expected, FLZ was active (Figure 3). Given that the treatment with the Lactobacillus probiotic did not influence the course of oropharyngeal infection, only S. cerevisiae products were evaluated in the subsequent experiments. First, the effect of IY and GI was evaluated by ex vivo bioluminescence emission of explanted gastric tracts (pharynx, esophagus, and stomach) at day +8 post-infection. The results show that treatment with IY and GI strongly reduced the BLI signal from the explanted organs as compared to organs from saline-treated infected mice (Figure 4). Next, the BLI signal from feces was tested to examine whether IY and GI were able to accelerate the passage of C. albicans from the stomach to intestine. The results show that BLI-C. albicans was evident only in the feces of saline-treated infected mice (Figure 4). Fungus dissemination to kidneys and liver was also evaluated. No fungal load was detected in these organs in any experimental group at day +8 post-infection (Figure 4). Further observations show that, by day +8 post-infection, infected mice did not show relevant clinical signs and manifested varying extents of weight loss. G 250- and saline-treated infected mice both showed similar extent of weight loss, while the weight loss in IY- and GY- treated groups was much less pronounced (data not shown).

In complementary experiments, we wondered if the inhibition of the pathogen load in the oral cavity may be related to the modulation of some important C. albicans virulence factors involved in the adhesion and invasion of epithelial cells (Naglik et al., 2011). To this end, the expression of ALS3 and two aspartyl proteases (SAP2 and 6) were analyzed. As shown in Figure 5A, the expression of the three virulence genes was significantly downregulated by the treatment with GI, but not with IY.

Neutrophils represent a major class of immune effector cells in fungal killing (Gazendam et al., 2016), of which reactive oxygen species production is the major killing mechanism (Moyes and Naglik, 2011), and they play a central role in protecting oral epithelium from C. albicans injury (Weindl et al., 2007). Given the difficulty to recover these cells from mucosal tissue, we examined whether the neutrophils recruited in the peritoneal cavity had candidacidal activity. The results reported in Figure 5B show that neutrophils from infected mice presented a reduction in killing activity as compared to those from uninfected mice and that the in vivo treatment with GI resulted in a greater increase in candidacidal activity of neutrophils compared to that of the two other treated groups. Moreover, the oxidative burst of neutrophils was also increased following the GI treatment. As expected, FLZ was able to increase antimicrobial capacity of neutrophils (Brummer and Stevens, 1996; Figure 5B).

NB and AD were employed by the company Lesaffre International, Lesaffre Group, Marcq-en-Baroeul, France.

The remaining 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.