In vitro effects of commercial mouthwashes on several virulence traits of Candida albicans, viridans streptococci and Enterococcus faecalis colonizing the oral cavity

Oral microbiota consists of hundreds of different species of bacteria, fungi, protozoa and archaea, important for oral health. Oral mycoses, mostly affecting mucosae, are mainly caused by the opportunistic pathogen Candida albicans. They become relevant in denture-wearers elderly people, in diabetic patients, and in immunocompromised individuals. Differently, bacteria are responsible for other pathologies, such as dental caries, gingivitis and periodontitis, which affect even immune-competent individuals. An appropriate oral hygiene can avoid (or at least ameliorate) such pathologies: the regular and correct use of toothbrush, toothpaste and mouthwash helps prevent oral infections. Interestingly, little or no information is available on the effects (if any) of mouthwashes on the composition of oral microbiota in healthy individuals. Therefore, by means of in vitro models, we assessed the effects of alcohol-free commercial mouthwashes, with different composition (4 with chlorhexidine digluconate, 1 with fluoride, 1 with essential oils, 1 with cetylpyridinium chloride and 1 with triclosan), on several virulence traits of C. albicans, and a group of viridans streptococci, commonly colonizing the oral cavity. For the study here described, a reference strain of C. albicans and of streptococci isolates from pharyngeal swabs were used. Chlorhexidine digluconate- and cetylpyridinium chloride-containing mouthwashes were the most effective in impairing C. albicans capacity to adhere to both abiotic and biotic surfaces, to elicit proinflammatory cytokine secretion by oral epithelial cells and to escape intracellular killing by phagocytes. In addition, these same mouthwashes were effective in impairing biofilm formation by a group of viridans streptococci that, notoriously, cooperate with the cariogenic S. mutans, facilitating the establishment of biofilm by the latter. Differently, these mouthwashes were ineffective against other viridans streptococci that are natural competitors of S. mutans. Finally, by an in vitro model of mixed biofilm, we showed that mouthwashes-treated S. salivarius overall failed to impair C. albicans capacity to form a biofilm. In conclusion, the results described here suggest that chlorhexidine- and cetylpyridinium-containing mouthwashes may be effective in regulating microbial homeostasis of the oral cavity, by providing a positive balance for oral health. On the other side, chlorhexidine has several side effects that must be considered when prescribing mouthwashes containing this molecule.

Introduction mitis, anginosus, mutans, salivarius, and bovis [18]. With the exception of the Streptococcus mutans species (the main etiological agent of dental caries), oral streptococci are mainly considered avirulent or even beneficial organisms. In some cases, certain oral streptococci can even hinder the development of a cariogenic S. mutans biofilm [19]. Recently, attention has been focused on several components of the mitis group (S. gordonii, S. oralis, S. mitis, S. parasanguinis, S. sanguinis), which have been implicated in supporting a pathogenic multispecies biofilm, given their ability to co-aggregate with other bacterial species and fungi as well. Not only such co-aggregation has been shown to confer a mutual advantage in biofilm formation, but also interactions among these microorganisms have been hypothesized to play a role during both commensal colonization of oral surfaces and in the onset and development of opportunistic infections [20].
E. faecalis, which is another microbial component of the oral microbiota, is also important for oral health, since it is one of the species mainly involved in failures of endodontic treatments [21,22]. E. faecalis localizes in dentinal canaliculi of devitalized teeth, resists to common disinfectants and endodontic irrigants [23,24], and can survive for long periods even under starving conditions, until changes of root canal microenvironment will allow its re-growth [25,26]. For bacteria, as well as for Candida, the biofilm production is an important virulence trait, especially in an ecological setting like the oral cavity, where many microbial species share the same ecological niche, compete for spaces and nutrients and therefore influence each other. In particular, reciprocal interaction among different species of streptococci can be of importance for the onset of oral diseases, such as carious lesions.
Not only some non-mutans streptococci, but also fungi may play a relevant role in the onset of carious lesions. C. albicans, in particular, has been hypothesized to significantly contribute to caries pathogenesis, especially in children, adolescents and young adults, because of its acidogenicity, capability to form hyphae and to secrete dentine-degrading enzymes [27,28]. In addition, both in vitro [29] and in vivo [30,31] experimental evidence suggests that, within the oral cavity, Candida-Streptococcal interactions do occur. Such interactions result in the formation of multi-species biofilm communities, where microorganisms use quorum sensing mechanisms to communicate with each other, to adjust their population density and to modulate gene expression patterns, in order to better adapt to the microenvironment where they dwell [20].
Following the observations recently described [16], this study was aimed to expand our knowledge about MoWs effects on several components of the oral microbiota. In particular, by means of in vitro models, we investigated the behaviour of MoWs-treated C. albicans in terms of: a) adhesion to both abiotic and biotic surfaces, b) capacity to elicit pro-inflammatory response by epithelial cells and c) susceptibility to phagocytosis and intracellular killing, as assessed by phagolysosome acidification. In parallel studies, we tested the capacity of these same MoWs of impairing biofilm formation by several species of viridans streptococci and/or by a mixed bacterial-fungal cell population.

Candida albicans strains
The reference strain SC5314 (ATCC MYA-2876) of C. albicans was employed for the present study. Fungal cultures were maintained by passages onto Sabouraud Dextrose Agar (SDA) plates (OXOID, Milano, Italy) performed at least twice a week. Moreover, the previously described [32] C. albicans CA1398, carrying the bioluminescence ACT1p-gLUC59 fusion product (BLI C. albicans) was used. The latter was maintained by biweekly passages in Yeast Peptone Dextrose Agar (YPD). The day before each experiment, fresh Candida cultures were seeded onto SDA or YPD plates and incubated at 37˚C. After overnight incubation, fungal cells were harvested by a sterile inoculating loop, suspended in phosphate-buffered saline (PBS, EuroClone, Whethereby, UK), washed twice by centrifugation at 3,500 rpm for 10 minutes, counted by Burker's chamber and suspended at 1 x 10 6 yeast cells/ml in RPMI 1640 medium supplemented with 10% heat-inactivated fetal bovine serum (hiFBS) (Defined Hyclone, Logan, UT, USA), gentamicin (50 mg/ml; Bio Whittaker, Verviers, Belgium), Ciproxin (2 mg/ml; ICN) and L-glutamine (2 mM; EuroClone, Milan, Italy), from here indicated as "cRPMI". This fungal cell suspension, composed of yeast-like forms, was used throughout the study to set up all the experiments.

TR146 oral epithelial cell line
The previously described human oral epithelial cell line TR146 [33] was maintained in Dulbecco's Modified Eagle's Medium/Ham's Nutrient Mixture F12 (Sigma, St Louis, MO, USA) supplemented with 15% hiFBS, gentamicin (50 mg/ml), Ciproxin (2 mg/ml) and L-glutamine (2 mM), hereafter referred to as "cF12". For maintenance, medium was replaced thrice a week. Cells were detached 8 days before the adhesion experiment by adding Trypsin/EDTA solution (2 ml), incubating for 15 min at 37˚C with 5% CO 2 and scraping to detach cells. Then fresh cultures were started in cF12, at a concentration of 10 6 /ml. TR146 cells were used for adhesion assays, cytokines and chemokines determination and in mixed C. albicans/S. salivarius biofilm production.

BV2 microglial cell line
The previously established murine microglial cell line BV2 [34] was maintained in cRPMI. Cells were detached biweekly by vigorous shaking, and fresh cultures were started at a concentration of 5 x 10 5 /ml the day before each experiment.
BV2 cells were employed for the phagocytosis assay.

Bacterial strains
Bacteria were collected from pharyngeal swabs, seeded on COS plates (Biomerieux, Marcy-L'Etoile, France) and allowed to grow for 24 hours at 35˚C. Alpha-haemolytic streptococci colonies were isolated and identification, at species level, was achieved by MALDI-TOF (Biomerieux). The system allowed to identify isolates belonging to the species S. salivarius, S. sanguinis, S. parasanguinis, S. vestibularis, and Enterococcus faecalis, but it could not discriminate between the species S. mitis and S. oralis; therefore, the 3 isolates used in the present study were indicated as S. mitis/oralis. Seventeen isolates were employed: 3 isolates each of the species S. salivarius, S. mitis/oralis, S. sanguinis, S. parasanguinis and E. faecalis and 2 isolates of the species S. vestibularis. For long-term storage, bacteria were maintained as frozen stocks at -80˚C, in Tryptic Soy Broth (TSB, Difco Laboratories, Detroit, MI, USA) with glycerol 5% (v/v).
The day before each experiment, 0.2 ml of bacteria from frozen stocks were seeded in 4 ml of TSB and incubated at 37˚C. After overnight incubation, 0.1 ml of bacterial suspensions were placed in 96-well plates and counted by measuring optical density (OD) at 595 nm wavelength, by means of a plate reader (Tecan Sunrise, Austria). The bacterial concentration was calculated by interpolating OD values to a McFarland standard curve [35]. Bacteria were then centrifuged and suspended in Brain Heart Infusion (BHI) broth (Liofilchem S.r.l., Italy) to a concentration of 1.5 x 10 8 CFU/ml.

Mouthwashes
Six commercial Mouthwashes (MoWs), operationally indicated with Arabic numbers, were employed. These mouthwashes are the same already employed in a previous study conducted by our group [16]. MoW 1 (CURASEPT 0.20, Curadent Healthcare S.p.A., Saronno (VA), Italy), MoW 2 (DENTOSAN COLLUTORIO, Recordati S.p.A., Milano, Italy) and MoW 3 (MERIDOL COLLUTORIO, Gaba-Colgate-Palmolive, Świdnica, Poland) contained 0.2% chlorhexidine digluconate (CHX). MoW 7 (PARODONTAX, GlaxoSmithKline, Brentford, UK) contained 0.06% CHX and 250 ppm F-sodium fluoride. MoW 4 (ELMEX SENSITIVE PROFESSIONAL, Gaba, Therwil, Switzerland) and MoW 5 (LISTERINE TOTAL CARE ZERO, Johnson&Johnson, Maidenhead, UK) included in their formulation fluorine-containing molecules and essential oils respectively. Two additional commercial mouthwashes, with different formulations, were employed in focused experiments: MoW 8 (ORAL B, Procter & Gamble, U.K.), containing 0.05% cetylpyridinium chloride (CPC) and 0.05% sodium fluoride and MoW 9 (PlaKKontrol Protezione Totale, IDECO, Bolzano, Italy), containing Triclosan and sodium fluoride. Details on MoWs composition are available in S1 Table. All the MoWs selected for the present study were ethanol-free. Operationally, the contact times of 5 and 15 minutes between Candida and the different MoWs have been chosen in order to mimic the everyday life employment of MoWs. Indeed, manufacturers recommend to avoid eating and drinking for at least 30 minutes after rinsing with MoWs, since the active molecules bind to the teeth and mucosal surfaces and are released gradually over time. Such property, known as substantivity, has been demonstrated for CHX, a molecule widely employed in MoWs formulations. This molecule, besides its direct antibacterial and antifungal action, can bind to surrounding tissues and then be released slowly, over extended periods of time [10,36]. Our previous study has shown that at the Candida:MoW 1:2 dilution (i.e., the same employed for the present study) there are no differences on biofilm formation and Candida metabolic activity between 1 and 5 minutes contact time with the MoWs; similarly, no differences have been observed between 15 and 30 minutes contact time [16,18]. Therefore, the 5 minutes contact time has been chosen to describe the immediate MoWs effect on Candida, whereas the 15 minutes contact time has been chosen to provide information on MoWs substantivity.

In vitro adhesion assays
Adhesion of Candida yeast cells was assessed both on abiotic surfaces and to oral epithelial cells.
Adhesion to abiotic surfaces: BLI C. albicans cells were harvested by a sterile inoculating loop, suspended in PBS, washed twice by centrifugation at 3,500 rpm for 10 minutes, counted by Burker's chamber and suspended at 2 x 10 6 yeast cells/ml in PBS. Then, 0.5 ml of Candida suspension were added to Eppendorf tubes containing 0.5 ml of PBS or 0.5 ml of the different MoWs, and incubated at 37˚C with 5% CO 2 for 5 or 15 minutes. MoWs were then removed by spinning twice all the tubes at 4,500 rpm for 8 minutes, and by replacing the supernatants every time with PBS. After the last wash, each pellet was suspended in cF12 in order to have a final concentration of 1 x 10 6 fungal cells/ml. One hundred microliters of these fungal suspensions were added (each condition in triplicate) to black 96-well microtiter plates with a transparent bottom (Perkin Elmer Life Sciences) and incubated for 1 hour at 37˚C. Then, each well was gently washed (3 times with 200 μl of PBS) to remove the non-adhered fungal cells. Finally, according to a previously described procedure [32], 2 μM coelenterazine (SynChem, Ohm, Germany) in luciferase assay buffer (LA buffer) were added to each well; the bioluminescence was read immediately with a luminometer (Victor X Light, Perkin Elmer Life Sciences) and the signal was expressed as relative luminescence units (RLU). According to the data by Enjalbert and coworkers [32], bioluminescence sensitivity corresponds to 1,000 gLUC59-expressing C. albicans cells. After reading, the adhered fungal cells were also evaluated by colony forming units counts (CFU). As detailed elsewhere [37], the LA buffer was removed from each well, 100 μl of Trypsin/EDTA solution were added to each well, then plates were incubated for 15 min at 37˚C plus 5% CO 2 to detach cells. Hence, fungal suspensions were seeded onto SDA plates, which were placed at 37˚C for 24 hours. The CFU were then evaluated.
Adhesion to oral epithelial cells: Two-hundred microliters of TR146 human epithelial cells (1 x 10 5 cells/ml) were seeded into black 96-well microtiter plates with a transparent bottom (Perkin Elmer Life Sciences). The plate was incubated for 3 days at 37˚C in order to allow the formation of a monolayer of epithelial cells homogeneously distributed on the bottom of each well. After 3 days, the BLI C. albicans cells were harvested and the experiment was carried out as detailed above. After reading, the adhered fungal cells were evaluated also by colony forming units counts (CFU), by the same procedure described above for the adhesion to abiotic surfaces.
Candida viability/growth assay BLI C. albicans cells (4 x 10 5 /ml) were pre-treated with MoWs or with PBS for 5 and 15 minutes. Then, fungal suspensions were washed twice with PBS to remove MoWs and suspended in 1 ml of PBS. Then, 100 μl of each fungal suspension were seeded onto SDA plates. CFU counts were evaluated after 24 hours of incubation at 37˚C.

Cytokines and chemokines
Two-hundred microliters of TR146 human epithelial cells (1 x 10 5 cells/ml) were seeded into 96-well microtiter plates (Costar 3595, Corning, NY, USA) and incubated for 3 days at 37˚C in order to allow the formation of a homogeneously distributed monolayer of epithelial cells on the bottom of each well. After 3 days, 0.5 ml of Candida suspension (2 x 10 6 yeast cells/ml in PBS) were added to Eppendorf tubes containing 0.5 ml of PBS or 0.5 ml of the different MoWs, and incubated at 37˚C with 5% CO 2 for 5 or 15 minutes. MoWs were then removed by spinning twice all the tubes at 4,500 rpm for 8 minutes, and by replacing the supernatants every time with PBS. After the last wash, each pellet was suspended in cF12 in order to have a final concentration of 1 x 10 6 fungal cells/ml. One hundred microliters of these suspensions were added to the TR146-containing wells of the 96-well microtiter plate and incubated for 24 hours at 37˚C. As positive control, TR-146 cells were stimulated with 1 μg/ml of LPS. As negative controls, wells containing TR-146 cells, which did not undergo infection with Candida or stimulation with LPS, were included. Each condition was tested in triplicate. Then, supernatants were collected and cytokines/chemokines levels were assessed by means of Quantibody Human Inflammation Array 1 (Ray Biotech Inc., Norcross, GA, USA), an antibody microarray system already employed for cytokines/chemokines determinations in several biological fluids [38,39]. With this array the following cytokines/chemokines could be quantitatively detected: IL-1α, IL-1β, Interleukin-4 (IL-4), IL-6, IL-8, Interleukin-10 (IL-10), Interleukin-13 (IL-13), Monocyte Chemoattractant Protein-1 (MCP-1), Interferon-γ (IFN-γ), Tumor Necrosis Factor-α (TNF-α). The supernatants were used undiluted and the assay was conducted according to the Manufacturer's instructions. The fluorescent signal from the slides was read by using a ScanArray Gx scanner (Perkin-Elmer, Cambridge, UK). Images generated were saved as TIFF files and quantified with the ScanArrayExpress software, provided by Perkin-Elmer. Since no specific indications on cut-offs were provided, all the cytokine/chemokine levels which, interpolated to their respective calibration curves, resulted below 0 pg/ml, were considered as negative.

In vitro phagocytosis assay
C. albicans SC5314 cells were harvested by a sterile inoculating loop, suspended in PBS, washed twice by centrifugation at 3,500 rpm for 10 minutes, counted by Burker's chamber and resuspended at 2 x 10 6 yeast cells/ml in PBS. Then, 0.5 ml of Candida suspension were added to Eppendorf tubes containing 0.5 ml of PBS or 0.5 ml of the different MoWs, and then incubated at 37˚C with 5% CO 2 for 5 or 15 minutes. MoWs were then removed by spinning all the tubes twice at 4,500 rpm for 8 minutes, and by replacing the supernatant every time with PBS. After the last wash, each pellet was suspended in 100 μl of FITC fluorescent dye (Sigma), previously diluted 1:10 with PBS, and incubated for 20 minutes. FITC was then removed by spinning all the tubes three times at 4,500 rpm for 8 minutes and by replacing the supernatant every time with PBS. After the last washing, each pellet was suspended in 240 μl of cRPMI in order to have working strength suspensions of 4 x 10 6 fungal cells/ml.
In the meantime, BV2 cells were seeded into Lab-Tek II chamber slides wells (Nalge Nunc International, Naperville, IL, USA). In order to strengthen adhesion of BV2 cells to the glass bottom of the wells, the latter were pretreated with poly-L-lysine (Sigma-Aldrich; 10 μg/well), for 30 minutes, and washed twice with PBS. BV2 cells (2 x 10 6 cells/ml, 100 μl/well) were then seeded and incubated for 30 minutes at 37˚C with 5% CO 2 . BV2 cells were thus infected by adding 100 μl of Candida suspension, the effector-target (E:T) ratio being 1:2. Then, 10 μl/well of LysoTracker Red DND-99 fluorescent dye (Molecular Probes, Life Technologies, Eugene, Oregon, USA), at 5 mM working strength concentration, were added. The chamber slides were then incubated for 1.5 hours at 37˚C with 5% CO 2 . Thirty minutes before the end of incubation, an additional volume of 10 μl of LysoTracker Red DND-99 fluorescent dye were added to each well. Fifteen minutes before the end of incubation, 40 μl of Uvitex 2B fluorescent dye (Polysciences, Inc, PA, USA) were added to each well of the chamber slide.
At the end of the incubation, all the wells were washed twice with warm cRPMI (5 minutes each) and fixed for 30 minutes with 4% paraformaldehyde (PFA) (Sigma-Aldrich) in PBS at +4˚C, protected from light. PFA was then removed by means of a double wash with cold PBS (10 minutes each). The chamber slide was then unset, by removing the plastic walls of the wells and the slide surface was treated with ProLong Gold Antifade Reagent (Molecular Probes, Invitrogen, St. Louis, Mo, USA) in order to suppress the photobleaching effect and to preserve the signals of fluorescently labeled target molecules. The fluorescence of phagocytosed and non-phagocytosed fungi was visualized by means of epifluorescence microscopy Nikon Eclypse 90i (Nikon Instruments, Tokyo, Japan) equipped with different filters for reading different fluorescent signals. For clarity, FITC fluorescent dye stained in green all Candida cells and its signal was read by FITC filter. LysoTracker Red DND-99 fluorescent dye stained in red the Candida cells inside acidic phagolysosomes and its signal was read by TRITC filter. Uvitex 2B fluorescent dye stained in blue all the Candida cells that had not been phagocytosed and its signal was read by DAPI filter. The phagocytosis percentage was then calculated by counting the cells containing one or more Candida on a total of 200 phagocytic cells scored. The phagocytosis index was calculated by counting the total number of phagocytosed Candida (including those inside acidic phagolysosomes) on a total of at least 200 Candida-containing phagocytic cells. The acidic phagolysosomes percentage was calculated on a total of at least 200 Candidacontaining phagocytic cells by counting Candida cells inside acidic phagolysosomes on the total number of phagocytosed Candida cells.

Biofilm production by oral streptococci and E. faecalis
Bacterial biofilm production was assessed for each individual Streptococcus isolate. Biofilm was allowed to form on flat-bottom 96-well plates (Corning Incorporated, NY, USA), according to the protocol described by Stepanović and coworkers, with minor modifications [40]. Briefly, 1.8 ml of bacteria cell suspensions (1.2 x 10 7 CFU/ml) in BHI broth were added to 1.8 ml of each MoW or PBS and incubated at 37˚C for 1 minute. Then, MoWs were removed by spinning at 4,500 rpm for 8 minutes and by suspending the pellets in PBS. After the last washing step, pellets were suspended in BHI broth to a working strength concentration of 1.5 x 10 7 CFU/ml, seeded in the wells of a 96-well plate (each condition in triplicate) and incubated at 37˚C with 5% CO 2 for 48 hours. Finally, to measure biofilm formation, Crystal Violet (CV) assay was performed. The OD was read at 570 nm wavelength, by means of a plate reader (Tecan Sunrise, Austria). The capacity to form BF of MoWs-treated bacteria was expressed as OD percentage, as compared to the OD of the PBS-treated counterpart, which was considered as 100%.

C. albicans biofilm production upon co-culture with MoWs-treated S. salivarius
An isolate belonging to the species S. salivarius (considered a "pro-cariogenic" species) was selected to study the influence (if any) of MoWs-pretreated streptococci on the capacity to form biofilm by C. albicans. This particular isolate was chosen because of its susceptibility to most of the MoWs employed (see below).
Two-hundred microliters of TR146 human epithelial cells (1 x 10 5 cells/ml) were seeded into black 96-well microtiter plates and incubated for 3 days at 37˚C, in order to allow the formation of a monolayer of epithelial cells homogeneously distributed on the bottom of each well. After 3 days, 600 microliters of S. salivarius isolate 12 cell suspensions (1.5 x 10 7 CFU/ml) in BHI broth were added to 600 μl of each MoW or PBS and incubated at 37˚C for 1 minute. Then, MoWs were removed by spinning at 4,500 rpm for 8 minutes and by suspending the pellets in PBS. After the last washing step, pellets were suspended in cF12 medium, seeded (each condition in triplicate) together with BLI C. albicans (1 x 10 6 /ml, 100 μl/well) in the TR146-containing wells of the 96-well plate described above and incubated at 37˚C with 5% CO 2 for 24 hours. After removal of the suspensions containing the non-adhered fungal cells, the wells were washed 3 times with 200 μl of PBS. Finally, 2 μM coelenterazine in LA buffer was added to each well and the bioluminescent signal was read immediately with a luminometer (Victor X Light, Perkin Elmer Life Sciences). After reading, the supernatants from each well were thrown, substituted with the same volume of fresh cF12 medium and incubated for further 24 hours. After this time, all the wells were washed 3 times (as described above), 2 μM coelenterazione in LA buffer was added to each well and the bioluminescent signal was read immediately with the luminometer.

Statistical analysis
All the data contained in the graphs are expressed as average values ± standard errors of at least 3 independent experiments. Analysis of variance (ANOVA) and Bonferroni post-hoc test were carried out by IBM SPSS Statistics 23 to assess overall differences amongst MoWs-treated groups in relation to control groups. Statistical significance was set at p<0.05.

Adhesion of MoWs-treated C. albicans to abiotic surface and epithelial cells
In order to evaluate the effects of the MoWs on C. albicans capacity to adhere to abiotic (plastic) and biotic (TR146 epithelial cells) surfaces, BLI C. albicans was employed. In particular, fungal cells treated either with MoWs or PBS (for 5 or 15 minutes) were seeded into a 96-well plate and were allowed to adhere for 1 hour at 37˚C, before washing and bioluminescence measurement. As expected, Candida cells treated with PBS returned the highest bioluminescent signals, indicating the capacity of the fungus per se to adhere to the plastic of the microwells. When Candida was treated with MoWs 1, 2, 3, 5, 7 and 8, little or no bioluminescent signals were detectable, indicating a failure of the fungus to adhere to the plastic. These results were consistent, irrespective of the MoWs incubation times; statistical analysis showed that bioluminescent readings from the wells containing MoWs-treated Candida were significantly lower than readings from PBS-treated Candida, used as negative control. MoW 4-and MoW 9-treated Candida returned a bioluminescent signal higher than those from the other MoWs. Yet, after 15 minutes of incubation such signal was significantly lower than that recorded from PBS-treated Candida, indicating that, tough less effective than the others, MoWs 4 and 9 were still capable to significantly impair Candida adhesion to the wells (Fig 1A).
Adhesion of MoWs-treated C. albicans to TR146 cells was significantly lower than that of PBS-treated Candida, for all but MoWs 4 and 9. Candida cells treated with MoWs 1, 2, 3, 5, 7 and 8 returned little or no bioluminescent signals, indicating an almost complete failure of the fungus to adhere to the oral epithelial cells, after MoWs treatment. On the contrary, MoW 4and MoW 9-treated Candida returned a bioluminescent signal, comparable to that collected from PBS-treated Candida, indicating that adhesion to the epithelial cells remained substantially not affected by these MoWs (Fig 1B). In order to verify if adhesion impairment by MoWs was due also to a decrease in fungal cells viability, we evaluated the CFU counts of fungal cells pre-incubated for 5 or 15 minutes with MoWs. Our results showed that growth of Candida pre-treated with MoWs 2, 3, 5, 7 and 8 (irrespective of the contact times) was almost completely inhibited (< 30 CFU). Differently, from undiluted supernatants of MoW 4-and PBS-treated Candida, more than 300 CFU could be counted, regardless of the contact times. From fungal suspensions pre-treated with MoW 1 and MoW 9, 115 ± 5.6 and 112 ± 4.2 and 245 ± 7.4 and 230 ± 2.5 CFU respectively could be counted after 5 or 15 minutes contact time.

Cytokines and chemokines secretion by TR146 oral epithelial cells incubated with MoWs-treated C. albicans
C. albicans, pretreated with MoWs or PBS, was used to infect monolayers of TR146 oral epithelial cells; 24 h later, the capacity of Candida to induce a secretory response was evaluated. By using an antibody microarray assay, the supernatants were screened for levels of IL-1α, IL-1β, IL-4, IL-6, IL-8, IL-10, IL-13, MCP-1, IFNγ and TNF-α. Candida cells treated with PBS were capable to induce the secretion of both IL-1α and IL-1β (but not of the other cytokines and chemokines) by TR146 oral epithelial cells. When TR146 cells were infected with MoWspretreated Candida, a significant reduction in the secretion of both IL-1α (Fig 2A) and IL-1β ( Fig 2B) was observed with respect to controls, namely TR146 cells infected with PBS-pretreated Candida. The most pronounced reduction was observed upon Candida pretreatment with MoWs 1, 2, 3, 5 and 7, irrespective of the 5 or 15 minutes contact time. Also, Candida pretreatment with MoW 4 significantly impaired fungal capacity to elicit IL-1α and IL-1β, but to a lesser extent (Fig 2A and 2B). None of the other cytokines/chemokines assessed by protein array were detectable at appreciable levels in the same culture supernatants. As expected, LPS stimulation induced secretion of IL-1α, IL-1β and TNF-α. The values of pg/ml obtained after LPS stimulation, were 1004.7 pg/ml ± 35.8 for IL-1α and 53 pg/ml ± 5.2 for IL-1β.

Susceptibility to phagocytosis of MoWs-treated C. albicans
Susceptibility to phagocytosis of MoWs-treated Candida cells was assessed by means of an in vitro assay, which uses a previously established murine microglial cell line BV2, known to efficiently ingest and intracellularly kill Candida [41]. Thus, PBS-or MoWs-pretreated C. albicans Effects of mouthwashes on oral microbiota cells (5 or 15 minutes) were exposed to BV2 cells for 1.5 hours; then, phagocytosis percentage, phagocytosis index and acidic phagolysosomes percentage were calculated. By ANOVA test, the differences in phagocytosis percentage amongst the MoWs-treated samples and the PBStreated controls never reached the level of significance (always p>0.05), irrespective of the contact time (Fig 3A).
Similarly, the differences in phagocytosis index amongst the MoWs treated samples and the PBS-treated controls never reached the level of significance (always p>0.05) (Fig 3B). Differently, when the acidic phagolysosomes percentage was assessed, two trends were detected. Specifically, Candida cells treated with MoWs 1 and 4 returned acidic phagolysosomes percentages not statistically different from those of the PBS-pretreated Candida, as assessed by ANOVA test (p>0.05). On the contrary, Candida cells treated with MoWs 2, 3, 5 and 7 returned an acidic phagolysosome percentage much higher than the controls'. By ANOVA test, such values were significantly higher than those of the control samples (p<0.05). Such results were independent from the 5 or 15 minutes contact times (Fig 3C).
Morphological analysis by microphotographs revealed that Candida cells, exposed to MoW 1, MoW 4 or PBS could develop hyphae, (Fig 4A) and, as shown by the TRITC filter which reads the red fluorescence of the Lyso-Tracker dye, only a few red spots were detectable ( Fig  4B-white arrows). On the contrary, in samples treated with MoWs 2, 3, 5 and 7, no hyphal forms were observed (Fig 4C) and, as assessed by TRITC filter, red spots were abundantly observed (Fig 4D).

Effects of MoWs on biofilm production by streptococci and E. faecalis
The effects (if any) of the MoWs on the capacity of oral streptococci to produce biofilm were investigated. Isolates from 5 species of oral streptococci and Enterococcus faecalis were treated for 1 minute with the same MoWs used for Candida and then allowed to form a biofilm. Biofilm formation was assessed after 48 hours incubation. Control groups, treated for 1 minute with PBS, showed that all the isolates were capable to produce biofilm, under the experimental conditions detailed in Material and Methods. However, when pre-treated with the MoWs, differences in the capacity to form biofilm could be observed, amongst species and according to the MoWs employed. Specifically, for S. parasanguinis, E. faecalis, S. vestibularis and S. salivarius the Effects of mouthwashes on oral microbiota Effects of mouthwashes on oral microbiota biofilm production was significantly reduced (even though with slight differences amongst isolates) by the treatment with all the MoWs, but MoW 4. The bacteria incubated with MoW 4 were able to produce a biofilm mass very similar to the controls (Fig 5A to 5D).
Differently, the S. mitis/oralis and S. sanguinis isolates were poorly affected by the treatment with any of the MoWs, with the exception of S. mitis/oralis isolate 34 (that was inhibited by MoWs 2, 3 and 7) and S. sanguinis isolate 10 (that was inhibited by MoWs 1, 2, 3 and 5) (Fig  5E and 5F).
Moreover, being S. salivarius isolate 12 particularly susceptible to most of the MoWs, it was chosen to test CPC-containing MoW 8 and triclosan-containing MoW 9. We found that the two MoWs had opposite effects on biofilm formation. As shown in Fig 5, panel D, MoW 8 completely inhibited biofilm formation, whereas the contact with MoW 9 did not have any effect.

Effects of MoWs on mixed biofilm
As detailed above, amongst the investigated streptococci, S. salivarius isolates were the most affected by MoWs treatment. Therefore, in order to assess if the MoWs-mediated impairment of S. salivarius biofilm formation could affect also C. albicans biofilm formation, S. salivarius isolate 12 was treated with the different MoWs or PBS for 1 minute and then added, together Effects of mouthwashes on oral microbiota with untreated bioluminescent C. albicans, to wells containing 3 day old monolayers of TR146 oral epithelial cells. Plates were then incubated at 37˚C and the BLI C. albicans biofilm formation was assessed, by bioluminescence, after 24h (early biofilm) and 48 hours (mature biofilm). The results, after 24 hours incubation, showed that the treatment of S. salivarius with MoWs 1 and 5 significantly (p<0.05) impaired C. albicans biofilm formation at early stages, with respect to Candida incubated with PBS-pretreated S. salivarius. Differently, no effects on C. albicans early biofilm were observed when S. salivarius had been pretreated with the other MoWs or with PBS. Bioluminescence measurement after 48 hours showed no effects on C. albicans biofilm production, irrespective of the fact that S. salivarius had or had not been pretreated with any of the MoWs (Fig 6).

Discussion
In a recent study [16], we have shown that several commercial MoWs exert anti-Candida activity. The observed differences in yeast cell growth, hyphal formation and biofilm production and persistence have been shown to depend on the MoWs composition and the fungal morphological/functional status. Here, we expand such results by investigating the effects of the same commercial MoWs on other Candida virulence factors, such as adhesion to abiotic and biotic surfaces, capacity to elicit pro-inflammatory responses by epithelial cells and susceptibility to phagocytosis and intracellular killing.
C. albicans capacity to adhere to both abiotic and biotic surfaces is the first stage of fungal invasion [9] and biofilm formation [42]. Here, we show that the treatment with the MoWs significantly impairs Candida capacity to adhere to plastic, irrespective of the contact times. Such impairment has been observed with all the MoWs, even though to a lesser extent when using MoWs 4 and 9. Therefore, the MoWs employed in the present study may help prevent oral fungal infections via the impairment of C. albicans adhesion to abiotic surfaces, frequently present in several types of patients. In addition, MoWs 1, 2, 3, 5, 7 and 8 have shown to impair Candida adhesion also to oral epithelial cells. Interestingly, MoWs 4 and 9 only impair (without blocking completely) Candida adhesion to plastic, leaving the fungus the ability to adhere to oral epithelial cells. These data suggest that such two MoWs are likely effective only in reducing non-specific forces, such as cell surface hydrophobicity [6][7][8][9][10], while the binding between fungal adhesins and epithelial receptors remains unaffected. Noteworthy, the observed impairment of adhesion mirrors a decrement of viable fungal cells by the direct contact with MoWs, as assessed by bioluminescence sensitivity.
Pretreatment of the fungus with the MoWs alters epithelial cell secretory response to some extent. In particular, IL-1α and IL-1β, secreted at high levels upon infection, significantly decrease in epithelia infected with MoWs-pretreated Candida. To our opinion, this result suggests that MoWs may play a relevant role in modulating Candida virulence. Epithelia, which cover all the mucosal surfaces of human body, must be able to discriminate between the invasive (pathogenic) and the commensal (harmless) forms of Candida, in order to prevent fungal invasion [12]. The reduced secretion of the pro-inflammatory cytokines IL-1α and IL-1β when Candida is pre-exposed to MoWs, indicates that under these conditions epithelial cells do not sense Candida as a threat, suggesting that fungal virulence potential has been attenuated. Differently, significant production of IL-1α and IL-1β occurs when epithelial cells are infected with PBS-treated fungi (control group); thus, under these conditions, the fungus retains its pathogenic potential, maintaining its capacity to adhere and to invade host tissues, as it may often be perceived at the mucosal level.
By a previously established in vitro model, which makes use of the murine microglial cell line BV2, as a prototype of tissue macrophages endowed with potent phagocytic activity and intra-cellular killing against Candida [34,41], here we show that the exposure to MoWs affects Candida susceptibility to such host-mediated first line defense. We show that, irrespective of the PBS or MoWs pretreatment, approximately 50% of the BV2 cells are able to ingest Candida cells, with a phagocytosis percentage and a phagocytosis index that remain substantially unchanged among groups. These data imply that MoWs do not alter the way phagocytes perceive the fungus in their surrounding environment, since their phagocytic activity remains almost unaffected. Differently, assessment of acidic phagolysosomes percentage highlights significant differences, depending upon the MoWs used. When Candida cells are pretreated with MoWs 1 and 4, the acidic phagolysosomes percentages remain substantially unchanged and similar to those of PBS-pretreated Candida. Morphological observations clearly show that microglial cells that have ingested Candida pretreated with MoWs 2, 3, 5 and 7 return acidic phagolysosomes percentages significantly higher than those observed with Candida pretreated with PBS. Interestingly, this phenomenon is to be ascribed to the inhibition of fungal dimorphic transition. Indeed, fungi treated with such MoWs remain in a yeast form and they are rapidly associated with acidic phagolysosomes It is worth noting that the MoWs capable to impair Candida dimorphic transition and, in turn, intracellular compartment acidification contain either CHX or essential oils in their formulation. These results are in line with the literature reporting that alcohol-and/or CHXcontaining MoWs have the highest antibacterial effects in vitro [43,44] and in the oral cavity [45]. Initial studies employing Candida spp. [46] have shown that CHX-containing MoWs are also effective against fungal cells, both in planktonic [47] and sessile forms [16,48]. Furthermore, in agreement with our present data, recent reports have shown that also MoWs containing essential oils have antifungal effects on several species of Candida [49]. Differently, the pretreatment with MoWs 1 and 4 does not impair Candida dimorphic transition and the fungus maintains its capacity to form hyphae as well as control group, irrespective of its intracellular and extracellular localization. Consequently, once phagocytosed, Candida can avoid phagolysosome acidification and in turn intracellular killing, probably via disturbance of the vacuolar membranes by the growing hyphae. We favour the idea that the presence of an anti-discoloration system (ADS) in the formulation of MoW 1 (which nevertheless contains 0.20% CHX, as well as MoWs 2 and 3), may partially interfere with CHX-mediated anti-fungal effects. Interestingly, upon incubation with MoW 1, only Candida capacity to form biofilm [16] and its susceptibility to phagocytes remain unchanged, while other virulence traits, such as adhesion and elicitation of proinflammatory cytokines, are affected. Once again, the results of fungal cell viability are in line with these data.
On selected experiments, MoWs containing either CPC or Triclosan have been used. CPCcontaining MoWs have been reported to be more effective on the planktonic, rather than the sessile form of C. albicans [48]. Here, we strengthen and expand such data, by showing that only CPC-containing MoW 8 indeed impairs adhesion of the yeast form of C. albicans to both biotic and abiotic surfaces.
As for the streptococci employed in the present study, we have chosen several species acting as early colonizers in multispecies microbial biofilm formation on dental surface [20]. All these species may play a role, albeit an indirect one, in the onset of several pathologies affecting the oral cavity, including caries. When using CHX-containing MoWs, a significant impairment in bacterial biofilm production has been observed for 4 out of 6 species investigated, i.e., S. salivarius, S. vestibularis, S. parasanguinis and E. faecalis. A slightly lower effect has been observed upon treatment with MoW 5. Based on the fact that CHX-containing MoWs (1, 2, 3 and 7) and, although to a lesser extent, the essential oils-containing MoW 5 impair biofilm formation by the so-called "pro-cariogenic" bacterial species [50][51][52], we speculate that such MoWs may indirectly provide protection against carious lesions onset. CPC, which is included in the formulation of several toothpastes and MoWs, is known to have antiplaque and anti-gingivitis effects [53]. In our hands, CPC-containing MoW 8 happens to impair biofilm formation by S. salivarius even more effectively than CHX-containing MoWs. In contrast, Triclosan-containing MoW 9 does not seem to have any effect, in line with previous data reporting lack of antimicrobial effects and the induction of bacterial resistance mechanisms by Triclosan-containing cosmetic products [54]. Differently, the treatment of S. sanguinis and S. mitis/oralis isolates with the MoWs does not affect their biofilm formation. Noteworthy, S. sanguinis acts as a natural competitor of S. mutans, by establishing spatial and temporal antagonistic relationships with several mechanisms [55]. In addition, S. sanguinis, S. mitis and S. oralis produce small molecules with antibiotic-like activity, capable to inhibit the growth of other potentially pathogenic microorganisms, such as C. albicans and Staphylococcus aureus [20,56]. As a direct evidence of their beneficial role, S. sanguinis, S. mitis and S. oralis have been found to be predominant on oral mucosae of healthy subjects [57]. Taking together all these data, we may speculate that the MoWs here employed are effective in contrasting caries onset since not only they reduce the colonization of bacterial species that (albeit indirectly) have a potential pathogenic effect, but also they do not damage bacterial species which are competitors of the cariogenic S. mutans.
Finally, in order to assess if MoWs may impact on Candida-streptococci interplay, we have co-infected TR146 oral epithelial cells with untreated C. albicans and MoWs-treated S. salivarius. In particular, S. salivarius isolate number 12 has been chosen, being its biofilm formation the most impaired by MoWs. Our results show that once exposed to MoW 1 and MoW 5, S. salivarius partially inhibits Candida biofilm formation. In particular, early biofilm (24 hours from infection) but not mature Candida biofilm (48 hours from infection) is slightly affected, the latter irrespective of the MoW employed. By these data, we conclude that S. salivarius treated with MoWs 1 and 5 delays Candida biofilm formation, but it fails to stop it completely.
MoWs affect several virulence traits of the oral microbiota, acting on both fungal and bacterial cells. According to our past [16] and present results, CHX-and essential oils-containing MoWs exhibit the most powerful anti-Candida effects impairing biofilm formation, adhesion, elicitation of proinflammatory responses and enhancing intracellular killing via acidic phagolysosomes. Also, viridans streptococci capacity to form biofilm is deeply affected. Notwithstanding its effectiveness, notoriously CHX has several side-effects (staining of teeth, loss of taste, numb feeling [58]) that have to be considered when employing MoWs containing this molecule. On the other side, CHX-free MoWs, which include fluorine-containing molecules in their formulation, provide some help in the process of enamel remineralization (and consequently in caries prevention) [59], but they may not ensure a direct protection against pathogenic microorganisms. CPC-containing MoW seems a promising alternative to CHXcontaining MoWs, even though more in-depth comparative studies are required to confirm these data. Finally, the lack of effect of Triclosan-containing MoW 9 does not warrant further studies. In addition, Triclosan has recently been banned from some cosmetic products by the FDA in the USA, because of the concerns on its side effects on human health [54].
In conclusion, because of the complexity of the oral environment and of the possible side effects of the active molecules of their formulations, special attention should be used when choosing MoWs for prevention and/or treatment of oral pathologies, also according to the etiological agent likely involved.
Supporting information S1