Phenazine-1-carboxamide functionalized mesoporous silica nanoparticles as antimicrobial coatings on silicone urethral catheters

Microbial infections due to biofilms on medical implants can be prevented by antimicrobial coatings on biomaterial surfaces. Mesoporous silica nanoparticles (MSNPs) were synthesized via base-catalyzed sol-gel process at room temperature, functionalized with phenazine-1-carboxamide (PCN) and characterized by UV-visible, FT-IR, DLS, XRD spectroscopic techniques, SEM, TEM, TGA and BET analysis. Native MSNPs, PCN and PCN-MSNPs were evaluated for anti-Candida minimum inhibitory concentration (MIC), minimum fungicidal concentration (MFC), Candida albicans (C. albicans) biofilms and C. albicans-Staphylococcus aureus (S. aureus) polymicrobial biofilm inhibition. PCN-MSNPs were four-fold effective (MIC 3.9 µg mL−1; 17.47 µM) and MFC (7.8 µg mL−1; 34.94 µM) as compared to pure PCN (MIC 15.6 µg mL−1; 69.88 µM) and MFC (31.2 µg mL−1; 139.76 µM). PCN-MSNPs inhibited in vitro C. albicans MTCC 227-S. aureus MTCC 96 biofilms at very low concentration (10 µg mL−1; 44.79 µM) as compared to pure PCN (40 µg mL−1; 179.18 µM). Mechanistic studies revealed that PCN induced intracellular ROS accumulation in C. albicans MTCC 227, S. aureus MTCC 96 and S. aureus MLS-16 MTCC 2940, reduction in total ergosterol content, membrane permeability, disruption of ionic homeostasis followed by Na+, K+ and Ca2+ leakage leading to cell death in C. albicans MTCC 227 as confirmed by confocal laser scanning micrographs. The silicone urethral catheters coated with PCN-MSNPs (500 µg mL−1; 2.23 mM) exhibited no formation of C. albicans MTCC 227 - S. aureus MTCC 96 and C. albicans MTCC 227 - S. aureus MLS -16 MTCC 2940 biofilms. This is the first report on PCN-MSNPs for use as antimicrobial coatings against microbial adhesion and biofilm formation on silicone urethral catheters.


Functionalization of MSNPs with PCN. The synthesized MSNPs
were sonicated in Milli Q water to form aqueous suspension of MSNPs, which was mixed vigorously with the freshly purified PCN (5 mg mL −1 ). This mixture was stirred for 6 h to facilitate the loading and binding of PCN to MSNPs at room temperature. The unbound PCN was removed by centrifugation at 8000 × g followed by discarding the supernatant. The binding of PCN to MSNPs was confirmed by shift in the two characteristic peaks in the UV spectrum of phenazines 32,33 . In vitro release of PCN. The in vitro release of PCN from the PCN-MSNPs was analyzed at two different pH conditions (pH 5.2 and pH 7.4), which was monitored by scanning in the wavelength range from 190 to 700 nm on a UV-visible double beam spectrophotometer (Lambda 25, PerkinElmer, Shelton, CT, USA), at an interval of 4 h for a total period of 40 h. To this, 3 mg mL −1 of PCN-MSNPs was suspended in PBS buffer and maintained at pH 5.2 and pH 7.4 separately at 37 °C under shaking conditions. The recorded absorbance values were compared with PCN standard curve to derive the percentage of PCN released from PCN-MSNPs at every time interval. Finally a plot was constructed showing percentage of the PCN release at both pH 5.2 and pH 7.4 against various time intervals. All these release studies were performed in triplicates and standard deviation is represented for each data point.
PCN standard curve. The purified PCN was dissolved in PBS to attain different concentrations ranging from 0.4-2 mg mL −1 and scanned in the wavelength range from 190 to 700 nm on a Lambda 25 UV-visible double beam spectrophotometer (PerkinElmer). A standard curve of PCN was constructed by plotting various concentrations of PCN against corresponding absorbance values. All the absorbance readings were measured in triplicates and the average values were used to plot the standard curve.
Antifungal activity assay. The antifungal activity of the purified PCN and PCN-MSNPs was performed by agar well diffusion method 34  Minimum fungicidal concentration (MFC) assay. The viability of various Candida strains upon treatment with PCN and PCN-MSNPs was studied and their minimum fungicidal concentration (MFC) was determined based on the fungicidal assay with slight modifications 35 . The MFC assay was performed on those Candida strains that were susceptible in the antifungal assay. MFC determination was carried out in sterile 2.0 mL microfuge tubes with MHB. PCN and PCN-MSNPs were added at a dose range of 62.5 to 0.9 µg mL −1 (279.98 to 4.03 µM) to each Candida strain and incubated at 30 °C for 24 h. MFC is the lowest concentration of PCN or www.nature.com/scientificreports www.nature.com/scientificreports/ PCN-MSNPs required to kill a particular Candida strain. The native MSNPs (250 to 0 µg mL −1 ) and MZ standard (250 to 0 µg mL −1 ; 600.78 to 0 µM) served as negative and positive controls, respectively. All the experiments were repeated thrice and their mean values were represented.
Polymicrobial inhibition assay. The effect of PCN and PCN-MSNPs against a mixture of C. albicans MTCC 227 and the bacterial strains, S. aureus MTCC 96 or S. aureus MLS-16 MTCC 2940 was assessed by modified agar well diffusion method 34 . These pathogenic yeast and bacterial strains were cultured in MHB separately and diluted to optical density (OD) of 0.1. Equal volumes of C. albicans MTCC 227 + S. aureus MTCC 96 and C. albicans MTCC 227 + S. aureus MLS-16 MTCC 2940 were mixed and spread over MHA plates. Different concentrations of PCN and PCN-MSNPs were loaded into the wells at concentrations of 250 to 0 µg mL −1 (1.11 to 0 mM) and incubated at 37 °C for 24 h. The native MSNPs (250 to 0 µg mL −1 ) and MZ standard (250 to 0 µg mL −1 ; 600.78 to 0 µM) served as controls. The experiments were carried out in triplicates and the diameters of zone of inhibition versus concentration of PCN and PCN-MSNPs against various Candida strains were plotted as an antibiogram.
Candida biofilm inhibition assay. The PCN and PCN-MSNPs were evaluated for their Candida biofilm inhibition property in microtiter plates 36  Polymicrobial biofilm inhibition assay. The effect of PCN and PCN-MSNPs on mixed biofilms was studied by a previously described method 12 with minor modifications. The overnight cultures of C. albicans MTCC 227, S. aureus MTCC 96, S. aureus MLS-16 MTCC 2940 were diluted with MHB medium to OD of 0.1 at 600 nm, mixed and loaded into each well of 96 well microtitre plate and incubated at 37 °C for 24 h. The medium was discarded and the formed biofilms were stained with 0.1% crystal violet solution for 45 min. Solubilization and elution of the stained biofilms was performed with 95% ethanol and then the plates were air-dried to remove excess water, if any. The absorbance values of the solubilized biofilms were recorded at 540 nm on an Infinite M200Pro microtitre plate reader (Tecan). All the mixed biofilm studies were performed in triplicates and average values were considered for determining the standard deviations.
Ergosterol quantification assay. A panel of Candida strains susceptible in antifungal activity, were used for observing the effect of PCN on their total sterol content. The sterols from the test Candida strains were isolated by using total sterol extraction method 37 with minor modifications. Initially, MHB (50 mL) spiked with various concentrations of PCN (0, 4, 8, 16 and 32 µg mL −1 ; 0, 17.91, 35.83, 71.67, 143.3 µM) was inoculated with single colonies of test Candida strains. Untreated cells (0 µg mL −1 ) were used a control. The treated culture medium was incubated at 37 °C for 20 h with agitation at 150 rpm. The cells were collected at stationary phase by centrifugation at 8200 × g for 6 min followed by washing the pellet twice with sterile distilled water. The net wet weight of each Candida pellet was weighed and recorded. To each Candida pellet, 5 mL of 25% alcoholic potassium hydroxide solution was added and mixed. Each of these Candida suspensions were collected in sterile glass vials and incubated at 85 °C for 60 min in a water bath. After cooling to room temperature, mixture of sterile distilled water and n-heptane (1:3) was added to each Candida suspension and vortexed vigorously for 5 min for the extraction of intracellular sterols into the n-heptane layer. From each Candida strain bilayer, the heptane layer with sterols was collected carefully and stored at −20 °C for 24 h. The sterol suspension (30 µL) of each Candida strain was diluted with 100% ethanol in 1:5 ratio and scanned from 230-310 nm wavelength using UV-visible spectrophotometer (PerkinElmer) to observe the four characteristic peaks corresponding to ergosterol and dehydroergosterol. The following equations were used to calculate the ergosterol content of each test Candida strain as percentage of wet weight of the cell pellet. www.nature.com/scientificreports www.nature.com/scientificreports/ = E 518 value (percent per centimeter) for crystalline dehydroergosterol All the experiments were performed in triplicates and mean values were considered for calculating the standard deviations. The statistically significant differences in ergosterol content in various treatment groups at 95% confidence level were determined using ANOVA in GraphPad PRISM software ver. 8.0.2 (GraphPad Software, Inc, La Jolla, CA, USA).

ROS quantification assay in C. albicans MTCC 227. Several antifungals cause ROS induced cell death
in fungi. Considering this fact, the intracellular ROS accumulation was measured in C. albicans MTCC 227 by 2′,7′-dichlorofluorescein diacetate (DCFH-DA)-dependant fluorometric assay 38 . Initially, the C. albicans MTCC 227 biofilm development in 96-well microtiter plate was achieved by culturing the strain at 37 °C for 24 h. Later, these biofilms were incubated for 24 h with PCN, PCN-MSNPs and MZ at a concentration of their corresponding MFCs (15.6, 7.8 and 7.8 µg mL −1 ; 69.88, 34.94 and 18.74 µM), respectively. As the catheters were coated with 500 µg mL −1 of PCN-MSNPs, the effect of MSNPs alone on the ROS accumulation was measured, to know if there was any effect of MSNPs at a dose of 500 µg mL −1 , where untreated cells served as control. All the groups were treated with 10 µM of DCFH-DA and incubated in dark for 24 h to allow the reaction of PCN induced cells with DCFH-DA for determining the intracellular ROS accumulation. The resultant fluorescent product was measured at excitation and emission wavelengths of 485 nm and 535 nm on Infinite M200Pro microtiter plate reader (Tecan). The resulting fluorescence intensities were plotted against concentration of PCN and MZ standard. All the measurements were performed in triplicates and the mean values were considered for the calculation of standard deviation. Statistically significant differences in data variances (p < 0.05) in the amount of ROS accumulated in different treatment groups were determined, using one-way ANOVA, in GraphPad PRISM software ver. 8.0.2 (GraphPad Software, Inc, La Jolla, CA, USA). Groups with significant differences were analyzed by Turkey's multiple comparisons test. As the catheters were coated with 500 µg mL −1 of PCN-MSNPs, the effect of MSNPs alone on the ROS accumulation was measured, to know if there is any effect of MSNPs at a dose of 500 µg mL −1 , where untreated cells served as control. Initially, the NBT was yellowish in color and on reaction with PCN induced ROS formed a blue formazan product that was spectrophometrically read at 570 nm. All the experiments were performed in triplicates and the mean values were considered for the calculation of standard deviation. The absorbance which corresponded to the percentage of intracellular ROS accumulated versus concentration of PCN was plotted. Statistically significant differences in data variances (p < 0.05) in the ROS accumulation due to different treatment groups were determined, using one-way ANOVA, in GraphPad PRISM software ver. 8.0.2 (GraphPad Software, Inc, La Jolla, CA, USA). Groups with significant differences were analyzed by Turkey's multiple comparisons test.

ROS quantification assay in
Confocal laser scanning microscopy. The effect of PCN-MSNPs on cell membrane integrity and viability of C. albicans MTCC 227 cells was assessed by staining with LIVE/DEAD ® BacLight ™ kit (Molecular Probes, Eugene, OR, USA) and further visualized by confocal laser scanning microscopy (CLSM). Initially, C. albicans MTCC 227 cells (OD 600 = 0.01) were cultured in MHB in two CLSM dishes (SPL Scientific, Gyeonggi-do, South Korea) at 37 °C for 18 h. Later, one group of C. albicans MTCC 227 was treated with PCN (15.6 µg mL −1 ; 69.88 µM) for 24 h and another group was untreated. To these groups, 1 µL each of SYTO 9 and propidium iodide were added in dark, mixed and incubated for 40 min at room temperature. The stained C. albicans MTCC 227 cells were fixed with 0.1% formaldehyde for 20 min. The fixed cells were then visualized by Eclipse Ti confocal laser scanning microscope (Nikon Corporation, Tokyo, Japan) interfaced with an argon ion laser at 480-490 nm wavelength for excitation and 500-635 nm band pass filter wavelength for emission. CLSM images were processed by Ti Control software ver. 4.4.4 and the scale bar for each image was 10 μm.

Elemental analysis using Inductively Coupled Plasma-Optical Emission Spectrometry (ICP-OES).
C. albicans MTCC 227 was cultured in two groups in MHB. One group was untreated and used as control, while the another group was treated with 15.6 µg mL −1 (69.88 µM) of purified PCN for 24 h. Later, both the groups were centrifuged at 8000 × g and the supernatants were filtered through 0.2 µm syringe filters. The amount of Na + , K + and Ca 2+ ions released into both the supernatants due to cell lysis was quantified using iCAP TM 6500 ICP-OES DUO Analyzer (Thermo Fisher Scientific, Cambridge, UK). Standards of Na + , K + and Ca 2+ ions (1 and 5 ppm) were prepared by serial dilution in sterile Milli Q water. The quantification of Na + , K + and Ca 2+ ions were carried out at 588.9, 766.4 and 317.9 nm, respectively. This analysis was performed thrice and the concentration of various ions were calculated using linear regression method.

Evaluation of anti-biofilm activity of PCN-MSNPs coating on silicone urethral catheters.
Silicone urethral catheters were coated with PCN-MSNPs using simple dispersion method 29 . A silicone urethral catheter (RÜSCH Sterile, Teleflex Medical Sdn, Perak, Malaysia) was cut into 4 cm long pieces, which were incubated in PCN-MSNPs solution (500 µg ml −1 ; 2.23 mM) at 37 °C for 24 h for effective coating. The coated catheter pieces were dried well to form the antimicrobial coating. The performance of this antimicrobial coating was assessed in a time-dependant manner at periodic interval of 24 h for up to 72 h against C. albicans MTCC www.nature.com/scientificreports www.nature.com/scientificreports/ 227-S. aureus MTCC 96 and C. albicans MTCC 227-S. aureus MLS-16 MTCC 2940 mixed biofilms. Coated and uncoated catheter pieces were incubated for biofilm formation in overnight grown mixed cultures of C. albicans MTCC 227-S. aureus MTCC 96 and C. albicans MTCC 227-S. aureus MLS-16 MTCC 2940 for 24, 48 and 72 h, at 37 °C with agitation at 150 rpm. After corresponding incubation period, each catheter piece were washed with distilled water twice and stained with crystal violet (0.1%) solution for 15 min. The dye from each PCN-MSNP coated catheter piece was extracted with the solubilization reagent (0.04 N HCl -isopropanol and 0.25% SDS). The collected dye at different time intervals was loaded in wells of 96 well microtitre plate and read at 590 nm employing microtitre plate reader (Tecan), where the absorbance was directly proportional to biofilm formation. The anti-biofilm effect of PCN-MSNPs coating was expressed as percentage of biofilm formation as a function of time. All the coating studies were performed in triplicates and mean values are represented. Statistically significant differences in data variances (p < 0.05) of biofilm growth percentage at different time points were determined using one-way ANOVA, in GraphPad PRISM software ver. 8.0.2 (GraphPad Software, Inc, La Jolla, CA, USA). Groups with significant differences were analyzed by Turkey's multiple comparisons test.
In vitro cytotoxicity assay. The NIH-3T3 cell line derived from mouse embryonic fibroblasts (ATCC No. CRL-1658) were employed for 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT) based cytotoxicity assay 40 . The cells were cultured in DMEM supplemented with 10% FBS, 2 mM L-glutamine, penicillin and streptomycin (0.05%) and maintained in a humidified incubator with 5% CO 2 at 37 °C. The cells were seeded in microtiter plates and were allowed to form a monolayer. PCN-MSNPs concentrations (0.1 to 100 µM) were added to the cells in triplicates and incubated for 24 h. MTT was added to the treated cells for 3 h, followed by addition of 100 µL of DMSO. Along with the PCN-MSNPs treatment, Mitomycin-C and untreated cells were run in parallel as positive and negative controls, respectively. The absorbance was recorded at 570 nm. Average values of triplicates were considered to plot a graph with number of cells versus absorbance.

Statistical analysis.
The results were presented as mean values ± standard error of mean. The experimental data were analyzed by one-way ANOVA using GraphPad PRISM version 6.0 (GraphPad Software, Inc, La Jolla, CA, USA). The values of P < 0.05, were accepted as level of significance.

Results and Discussion
The mesoporosity of the synthesized MSNPs was confirmed by BET surface area analysis with the following results: Specific surface area = 83.7 m 2 g −1 ; pore volume = 0.13 cm 3 g −1 and pore diameter = 6.43 nm. The binding of PCN to the MSNPs was confirmed by BET surface area analysis with the following results: Specific surface area = 29.7 m 2 g −1 ; pore volume = 0.043 cm 3 g −1 and pore diameter = 5.86 nm 41 . Characterization studies. To determine the size and shape, the MSNPs were examined by FE-SEM and TEM and the corresponding micrographs ( Fig. 1) showed that the formed MSNPs and PCN-MSNPs were uniform, monodispersed and spherical in shape. The TEM micrographs revealed that the pores of MSNPs were also filled with PCN. The low angle powder XRD data showed characteristic Bragg diffraction peak for MSNPs with a 2θ value of 2.1° corresponding to (100) facet peak of hexagonally ordered pore channeling ( Supplementary  Fig. S1).
The MSNPs were functionalized with freshly purified PCN to record the FT-IR spectrum. FT-IR spectrum of synthesized MSNPs (Fig. 2) showed characteristic signal transmittance of silica nanoparticles 40  In our previous study 31 , the FT-IR spectrum of the pure PCN showed characteristic peaks in the range of 2900-3000, especially at wavenumber of 2922 cm −1 , which is closer to the peaks resulting from CTAB improper washing. The wavenumber peak at 2922 cm −1 attributed to aromatic CH stretching. While in the FT-IR spectrum (Fig. 2), PCN-MSNPs showed characteristic peaks of PCN in the wavenumber range of 2900-3000 cm −1 along with that of native MSNPs. However, in the FT-IR spectra of native MSNPs, the peaks corresponding to CTAB residues were also observed as reported earlier 42 . We presume that PCN was successfully functionalized as evident from the characteristic peak observed at a wavenumber of 2922 cm −1 in the revised spectrum of PCN-MSNP. This aspect was further confirmed through thermogravimetric analysis (TGA) of MSNPs and PCN-MSNPs.
The average size of synthesized MSNPs and PCN-MSNPS based on DLS was 193 nm and 234 nm, suggesting the shape of nanoparticles to be uniform and monodispersed (Supplementary Fig. S2). The zeta potential values of MSNPs and PCN-MSNPs were 2.5 mV and 14 mV, respectively. The nitrogen adsorption-desorption isotherms of MSNPs and PCN-MSNPs shown in Fig. 3A depicts the characteristic type IV hysteresis loops of mesoporous silica nanoparticles 41 . According to BET analysis (Fig. 3), the volumes of MSNPs and PCN-MSNPs were 0.13 and 0.043 cm 3 g −1 , respectively. Based on the comparative analysis of the pore volumes, it could be deduced that an amount of PCN corresponding to a volume of 0.087 cm 3 g −1 of each nanoparticle was loaded with PCN with a drug loading capacity of 53%. The functionalization of PCN to MSNPs resulted in reduced pore size indicating the successful loading of PCN into the mesopores. The TGA of MSNPs showed the characteristic curve corresponding to the temperature vs. weight percentage plots, with a slight decrease in the weight percentage www.nature.com/scientificreports www.nature.com/scientificreports/ owing to the removal of CTAB at temperature above 250 °C (Supplementary Fig. S3). The curve representing the PCN-MSNPs showed a considerable decrease in weight percentage that could possibly be attributed to the melting of both PCN and residual CTAB. The melting of PCN started after 220 °C and the curve observed confirmed the efficient loading of PCN into the synthesized MSNPs, as evident from the characteristic melting and degradation patterns of PCN (Supplementary Fig. S3).  www.nature.com/scientificreports www.nature.com/scientificreports/ UV-visible spectrum of the purified PCN showed characteristic absorbance at 294 and 431 nm ( Supplementary  Fig. S4). After functionalization of PCN to MSNPs, a shift in both the characteristic absorbance peaks of PCN was observed. Only PCN was being released into the medium which was observed by UV-visible spectrum, while there was no characteristic absorbance peaks observed for MSNPs. DLS studies showed that MSNPs were stable up to 4 months without any disintegration to form smaller and soluble particles, which is in correlation with our previous study 43 . The successful coating of phenazine-1-carboxamide on mesoporous silica nanoparticles (PCN-MSNPs) was confirmed by the change in the color of catheter surface.
In vitro release of PCN. The in vitro PCN release profile from PCN-MSNPs was monitored for 40 h at regular intervals of 4 h each ( Supplementary Fig. S5). PCN release (85%) was higher at pH 5.2 as compared to pH 7.4 where the release was only 46%, which was in a slow, stable and controlled pattern, thus increasing the bioavailability and lowering the effective dosage by two to three folds, which was in a similar trend to that of our previous study that showed slow and controlled drug release from MSNPs in acidic conditions 43 . Only PCN was being released into the medium which was observed by UV absorbance spectra. Especially, a slow release pattern of PCN at pH 7.4 was observed suggesting that it remained bound to MSNPs under physiological conditions thus exhibiting promising lower effects on the normal cells and tissues. At pH 5.2, which was acidic in nature mimicked the infected tissues, the release profile of PCN was very stable and controlled. This could plausibly be due to a simple change in redox potential of amino groups of PCN, which enabled the release of PCN. Overall, the in vitro release profile indicated that functionalization of PCN to MSNPs could possibly attribute to lowered therapeutic dosage and improved efficacy. This release profile corroborated with the results of fungicidal and polymicrobial biofilm inhibition assays. The synthesized MSNPs were stable up to 4 months at physiological pH as evident from the DLS analysis (data not shown).
Antifungal activity. Candiduria is prevalent in ~10% of intensive care catheterized patients, where C. albicans was present in 69.1% of candiduria cases, forming a prevalent cause for catheter-associated candiduria 10 . In the current study, PCN and PCN-MSNPs were evaluated for their antifungal activity and their MICs were determined and depicted in Table 1. Miconazole (MZ) was used as standard control. The purified PCN exhibited good antifungal activity against test Candida strains at concentrations ranging between 7.8 µg mL −1 (34.94 µM) to 15.6 µg mL −1 (69.88 µM). PCN-MSNPs exhibited an improved antifungal activity at lower concentrations ranging between 1.9 to 3.9 µg mL −1 (8.51 to 17.47 µM), showed enhanced and persistent antifungal activity for extended time period of 124 h as compared to pure PCN which was potent up to 72 h only. This improved activity may be attributed to increased bioavailability of PCN due to increased surface area distribution and sustained release of PCN from PCN functionalized MSNPs. Moreover, PCN-MSNPs exhibited promising antifungal activity similar to that of standard MZ.  74 to 69.88 µM), which was two-fold lower as compared to that of PCN. In addition, the activity of PCN-MSNPs was similar to the standard MZ. Based on the results, it was evident that PCN-MSNPs were promising as compared to PCN alone and equipotent as that of MZ with regard to antifungal activity. The results to this regard are depicted in Table 1.  Table S1). This improved activity of PCN-MSNPs could be due to the stable and controlled release of PCN under acidic growth conditions of C. albicans which was similar to the in vitro release profile at pH 5.2. These results suggest that PCN-MSNPs could effectively inhibit mixed populations of C. albicans MTCC 227 and the two S. aureus strains under in vitro conditions. The antibiogram results are depicted in Supplementary Fig. S6.  www.nature.com/scientificreports www.nature.com/scientificreports/ Inhibition of Candida biofilms. Biofilms formed by various Candida species contribute to almost 15% of blood stream-and medical device-associated infections [44][45][46] . In the present study, the effect of PCN on biofilms formed by various Candida strains was studied and the results showed that PCN effectively inhibited these biofilms at a lower dose. Both the purified PCN and PCN-MSNPs were evaluated for their biofilm inhibition activity against C. albicans MTCC 227, C. albicans MTCC 1637, C. albicans MTCC 3017, C. albicans MTCC 3018, C. albicans MTCC 4748, C. albicans MTCC 7315, C. parapsilosis MTCC 1744, C. glabrata MTCC 3019, C. krusei MTCC 3020 and I. hanoiensis MTCC 4755, with MZ drug run in parallel as a standard control. The results of Candida biofilm inhibition assay are shown in Table 2. The purified PCN exhibited good inhibition of various Candida biofilms with inhibitory concentration ranging from 80-95 µM. Whereas, PCN-MSNPs exhibited promising inhibition of the test Candida biofilms at a lower dose ranging between 34 to 44 µM as compared to that of purified PCN. The above results suggest that PCN-MSNPs effectively inhibited biofilms formed in vitro by the test Candida strains.

Polymicrobial biofilm inhibition.
Under clinical situations, mixed populations of C. albicans and various bacteria coexist as biofilms in patients, which make the treatment very difficult and contributes to an increase in financial burden on hospitals and patients 6 (Table 3). This enhanced performance of PCN-MSNPs could be due to improved bioavailability and controlled release of PCN at acidic conditions of C. albicans growth as indicated by in vitro release studies.
Reduction in ergosterol content. Ergosterol is mainly essential to maintain fungal cell membrane integrity and fluidity. Many of antifungals like amphotericin B interacts with ergosterol, altering the selective permeability of the cell membrane 47 . In the current study, it was noticed that ergosterol content was 100% in the untreated Candida strains, namely C. albicans MTCC 227, C. albicans MTCC 4748, C. albicans MTCC 1637, while a significant dose-dependent decrease in the ergosterol content was observed in case of both PCN and MZ-treated strains. As the dose of the PCN increased from 0 to 20 µg mL −1 (89.59 µM), resulted in decreased ergosterol content, which is a key component of the fungal cell membrane as indicated by the decline in peak heights. The UV-visible scans showed a flat line at the effective dose of PCN against C. albicans MTCC 227, C. albicans MTCC 4748 and C. albicans MTCC 1637 strains suggesting no detectable levels of sterols. The results to this regard are shown in Fig. 4. Our findings suggest that PCN alters the sterol profile and could be possibly exerting this antifungal activity by reducing the total ergosterol content, cell membrane disruption and subsequent cell death. www.nature.com/scientificreports www.nature.com/scientificreports/ Intracellular ROS generation in C. albicans MTCC 227. Reactive oxygen species (ROS) is an important inducer of apoptosis in yeast cells 48,49 . Moreover, the oxidative stress in different Candida species results in lipid peroxidation, cell wall damage, reduced antioxidant capacity, protein carbonylation and catalase activities that are biologically significant 50 . ROS induced oxidative damage in Candida cells is one of the crucial mechanism of many antifungal agents. In order to understand any role played by PCN in oxidative damage of C. albicans MTCC 227, the intracellular ROS accumulation within the MSNPs treated C. albicans MTCC 227 sessile cells of mature biofilms and biofilms treated with PCN, MSNPs, PCN-MSNPs and MZ (MIC) was measured using fluorimetric 2′,7′-dichlorofluorescein diacetate (DCF-DA) assay. In this assay, the dye conversion is dependent on the metabolic and redox state of cells in the biofilm. The PCN-MSNPs treated biofilms exhibited approximately three folds higher to that of ROS accumulation as compared to that exhibited by the MZ treated counterparts. While the ROS accumulation exhibited by PCN-MSNPs treated were nearly two folds higher to that of ROS accumulation as compared to that exhibited by the PCN treated counterparts. The MSNPs treated biofilms showed negligible fluorescence, whereas PCN and MZ treated counterparts exhibited high fluorescence intensities which corresponded to increased levels of intracellular ROS accumulation. The fluorescence intensities of PCN and MZ-treated biofilms were almost similar indicating the efficient intracellular ROS accumulation caused by PCN (Fig. 5A). On the other hand, native MSNPs (500 µg/mL) treated and untreated C. albicans MTCC 227 biofilms showed a negligible amount of ROS accumulation (Fig. 5B).  www.nature.com/scientificreports www.nature.com/scientificreports/ absorbance was directly proportional to the amount of intracellular ROS accumulation by PCN. The amount of ROS accumulated by PCN was seven-fold higher than the control (MZ). On the other hand, native MSNPs (500 µg mL −1 ) treated and untreated S. aureus MTCC 96 and S. aureus MLS-16 MTCC 2940 biofilms showed a negligible amount of ROS accumulation (Fig. 5D). Also this accumulated ROS could possibly damage the cell membrane which could be the result of high redox potential of PCN 52 . Moreover, the amount of intracellular ROS accumulated with PCN-MSNPs treatment was almost fifteen fold more than that observed with MZ, and this value was higher as compared to the previously reported antifungal mechanisms of nanoparticle conjugated antifungal compounds, such as that of MZ-conjugated AgNPs 49 and N-acetyl cysteine conjugated AgNPs 53  Confocal scanning laser microscopy. Cell membrane in yeast cells plays an important role in structural and functional processes, the damage of which leads to lysis and cell death 54 . In the present study, PCN www.nature.com/scientificreports www.nature.com/scientificreports/ reduced the total content of ergosterol, that is crucial for the cell membrane integrity of C. albicans MTCC 227 which was assessed by LIVE/DEAD ® BacLight ™ kit and visualized by CSLM. The LIVE/DEAD BacLight kit was used to monitor the viability of C. albicans MTCC 227 cells as a function of cell membrane integrity of the cells which uses appropriate mixtures of SYTO 9, a green fluorescent nucleic acid dye and propidium iodide, a red fluorescent nucleic acid dye. SYTO 9 is permeable to the live C. albicans MTCC 227 cells having intact cell membranes and thus stains them green, while the propidium iodide stains red the C. albicans MTCC 227 cells having damaged cell membranes that were considered to be dead or in dying state. Under CSLM, the micrographs were thus observed as green and red fluorescent images for respective live and dead C. albicans MTCC 227 cells (Fig. 6). PCN-MSNPs significantly reduced the percentage of live C. albicans MTCC 227 cells at a concentration of 15.6 µg mL −1 (69.88 µM) as compared to the untreated cells. CSLM images suggest that PCN-MSNPs damaged the structural integrity of C. albicans MTCC 227 cells and caused yeast cell death which correlates well with the decrease in ergosterol content.

ROS assay in
Elemental analysis. ICP-OES was employed to analyze the levels of metal ions released during ion leakage (Supplementary Table S2). From the results, it was noticed that the concentration of metal ions such as Na + , K + and Ca 2+ released were in higher amounts (463.5, 293.1 and 380.4 ppm, respectively) in the PCN-treated C. albicans MTCC 227 cells as compared to the untreated control (300.2, 47.47 and 18.40 ppm, respectively), which indicates that PCN caused membrane damage resulting in Na + , K + and Ca 2+ leakage and subsequent cell death.
Anti-biofilm activity of PCN-MSNPs coating on silicone urethral catheter. Among the medical devices, urinary catheters form a very suitable surface for rapid microbial colonization, adhesion and biofilm growth making their removal as a sole choice for preventing the further spread of infection 55 . S. aureus is one of the frequent causative agent for the urinary catheter associated infections in patients 56 . Among fungi, C. albicans is a major causative agent for urinary catheter-associated infections in intensive care patients 10 . With the emergence of resistant C. albicans and S. aureus strains, the treatment and management of Candida-Staphylococcus mixed biofilms has become more challenging 56 . Considering the facts, the novel compounds that can specifically target and inhibit the biofilm formation would be interesting as compared to the rational use of antibiotics. The PCN-MSNPs anti-biofilm coating was effective in inhibiting preformed mixed biofilms of C. albicans MTCC 227 and S. aureus MTCC 96. It was reported earlier that surface charge and hydrophobicity properties plays a key functional role which influenced the surface chemistry of the coating material and prevents the bacterial attachment and biofilm formation 57 . Various clinical manifestations like oral and ear infections, diabetic wound infections, pulmonary and urinary tract infections as well as medical implant associated infections are largely caused by biofilms formed by multi-species microbial communities that exhibit high resistance to antimicrobial agents which is significant from a clinical perspective 8,58 .
In the present study, PCN-MSNPs were coated on silicone urethral catheters and were further evaluated for their performance as anti-biofilm coating agents. The coating of PCN-MSNPs over silicone catheters could be the www.nature.com/scientificreports www.nature.com/scientificreports/ result of surface modification by functionalization of PCN moieties which prevents the attachment of mixed biofilms. The successful coating was confirmed by change of color on the catheter surface ( Supplementary Fig. S7). The performance of anti-biofilm coatings of PCN-MSNPs revealed that mixed biofilms of C. albicans MTCC 227-S. aureus MTCC 96 and C. albicans MTCC 227-S. aureus MLS-16 MTCC 2940 were effectively inhibited as a function of time as illustrated in Fig. 7A,B. The anti-biofilm assay showed that mixed biofilm formation of C. albicans MTCC 227-S. aureus MTCC 96 on coated catheters was reduced by 89.1%, 88.2% and 88.1% at 24, 48 and 72 h of catheter incubation, respectively, while the mixed biofilm formation of C. albicans MTCC 227-S. aureus MLS-16 MTCC 2940 was reduced by 93.8%, 92% and 91% at 24, 48 and 72 h of catheter incubation, respectively (Fig. 7C).
In vitro cytotoxicity. The in vitro cytotoxic effect of PCN-MSNPs was evaluated in NIH-3T3 cell line based on MTT assay, which indicated that PCN-MSNPs were non-toxic up to IC 50 value of 100 µM ( Supplementary  Fig. S8). While the antifungal activity and polymicrobial biofilm inhibition by PCN-MSNPs were exhibited at a three-fold lower concentrations than the IC 50 value in MTT assay. All the above results suggest that PCN-MSNP coatings could serve as promising antimicrobial agents to combat CAUTI. Further, in vivo studies could possibly transform the PCN-MSNP application under healthcare situations. This is the first report on PCN-MSNPs for use as antimicrobial coatings against microbial adhesion and biofilm formation on silicone urethral catheters.
In conclusion, we hypothesize that PCN induced ROS accumulation and reduction in ergosterol content could damage the cell membrane resulting in the disregulation of intracellular concentrations of key ions such as Na + , K + and Ca 2+ , thus breaking the osmotic balance of the cell. Leakage of these ions may cause cell lysis and death in view of their crucial roles in activating various biologically significant enzymes involved in respiration and biosynthetic pathways 59 . Moreover, the membrane in Candida albicans is composed of various sterols and lipids that act as barriers for various components and ions apart from the forming matrix for various significant proteins 60,61 . The disturbance in any of these lipids or sterols results in efflux of ions 62 . This hypothesis was proved based on confocal micrographs that clearly reveal the cell membrane damage and lysis leading to leakage of intracellular cell contents. For understanding the status of ionic homeostasis, the PCN treated and untreated C. albicans MTCC 227 cell supernatants were subjected to elemental analysis. The results of elemental analysis indicated that PCN treated cells showed significant leakage of Na + , K + and Ca 2+ ions as compared to control groups.