Biofilm-Forming Potential of Ocular Fluid Staphylococcus aureus and Staphylococcus epidermidis on Ex Vivo Human Corneas from Attachment to Dispersal Phase

The biofilm-forming potential of Staphylococcus aureus and Staphylococcus epidermidis, isolated from patients with Endophthalmitis, was monitored using glass cover slips and cadaveric corneas as substrata. Both the ocular fluid isolates exhibited biofilm-forming potential by the Congo red agar, Crystal violet and 2,3-bis (2-methoxy-4-nitro-5-sulfophenyl)-5-(phenylamino) carbonyl-2H-tetra-zolium hydroxide (XTT) methods. Confocal microscopy demonstrated that the thickness of the biofilm increased from 4–120 h of biofilm formation. Scanning electron microscopic studies indicated that the biofilms grown on cover slips and ex vivo corneas of both the isolates go through an adhesion phase at 4 h followed by multilayer clumping of cells with intercellular connections and copious amounts of extracellular polymeric substance. Clumps subsequently formed columns and eventually single cells were visible indicative of dispersal phase. Biofilm formation was more rapid when the cornea was used as a substratum. In the biofilms grown on corneas, clumping of cells, formation of 3D structures and final appearance of single cells indicative of dispersal phase occurred by 48 h compared to 96–120 h when biofilms were grown on cover slips. In the biofilm phase, both were several-fold more resistant to antibiotics compared to planktonic cells. This is the first study on biofilm forming potential of ocular fluid S. aureus and S. epidermidis on cadaveric cornea, from attachment to dispersal phase of biofilm formation.


Introduction
The eye has a number of defense mechanisms including several components in the tears (lysozyme, immunoglobulins, lactoferrin, lipocalin, β-lysin, etc.) which act as the first line of defense against bacterial infection [1][2][3]. Blinking the eyelids also plays an important role in the spread of tears across the ocular surface and thus acts as a barrier to the microbial-colonization of the ocular surface [4]. The cornea also has an immune surveillance system [5,6] comprised of innate defenses contributed by numerous cellular (corneal epithelial cells, corneal nerves, keratocytes, polymorphonuclear cells, neutrophils, eosinophils, macrophages, NK cells, Langerhans cells, etc.) and molecular elements (components of complement, interferons, interleukins, etc.) to eliminate pathogens [5][6][7]. Despite these defense mechanisms, many microorganisms do survive on the ocular surface and recent studies either based on 16S ribosomal RNA (rRNA) gene amplification, cloning and sequencing or by using NGS (next-generation sequencing) based on 16S rRNA gene amplification and analysis (16S rRNA meta-barcoding) revealed a greater degree of diversity and abundance in the bacterial microbiome of the ocular surface [8][9][10][11][12][13]. Keilty [14] was the first to cultivate hemolytic Staphylococcus from the conjunctival swabs of normal subjects. Subsequently, it was observed that several Gram-positive bacteria, including coagulasenegative staphylococci (CoNS) and other staphylococcal species such as Staphylococcus aureus, S. epidermidis, and a species of Streptococcus, S. viridans cause acute postoperative cadaveric cornea as a substratum is important since bacteria colonize the cornea, allow prolonged survival of microorganisms and are the cause of active inflammation and infection [59]. Further, antibiotic susceptibility was monitored both in the planktonic and biofilm phases. The results confirmed, based on qualitative (Congo red agar (CR) and Scanning Electron Micrscopy (SEM)) and quantitative methods (Crystal Violet Method (CV), [2,3-bis (2-methoxy-4-nitro-5-sulfophenyl)-5-(phenylamino) carbonyl-2H-tetra-zolium hydroxide)] (XTT) and Confocal Laser Scanning Microscopy (CLSM)), that the two ocular isolates possess the potential to form biofilm. More importantly, SEM analysis of the biofilms indicated that ocular fluid S. aureus and S. epidermidis produced biofilms both on a synthetic substratum and on cadaveric cornea. However, biofilm formation on the cadaveric cornea was more rapid and cornea was probably the preferred substratum. This is the first study on biofilm forming potential of ocular fluid S. aureus and S. epidermidis on cadaveric cornea, from attachment to the dispersal phase of biofilm formation.

Bacterial Cultures and Characterisation
In the present study, two vitreous samples from patients with Endophthalmitis were received from the L V Prasad Eye Institute, Hyderabad, India, and cultured on 5% sheep blood agar medium plates [23] and two single colonies were purified by repeated streaking and subjected to basic microbiological tests. Both the isolates were Gram positive, coccoid, occurred in groups and were positive for catalase. Isolate L-1058-2019 (2) produced pink color colonies on MSA agar and white opaque color colonies on non-hemolytic blood agar, and was negative for coagulase and oxidation-fermentation test, suggestive of Staphylococcus epidermidis. In contrast, isolate L-1054-2019 (2) produces yellow color colonies on MSA agar and cream white opaque color colonies on β-hemolytic on blood agar and is positive for coagulase and oxidation-fermentation test suggestive of Staphylococcus aureus. The identity of the two isolates as Staphylococcus aureus and S. epidermidis was also confirmed using Vitek 2 Compact System (BioMérieux, Marcy l'Etoile, France), an automated system for the identification of bacterial isolates up to species level. Vitek 2 has been reported to identify 95% of Staphylococcal isolates correctly [60]. The two isolates were preserved in tryptone soya broth (TSB) [61] with 30% glycerol at −80 • C. All the preserved isolates were revived on 5% sheep blood agar media plates and incubated overnight at 37 • C. Henceforth, the above two isolates would be referred to as ocular fluid isolates.

Congo Red Agar Method
The CRA method is a qualitative assay for monitoring the potential of a microorganism to form a biofilm when grown on a solid CRA plate [64]. Congo red is known to bind to amyloid-like proteins which are a component of the extra-cellular polymeric substance of the biofilm [65,66]. In this method a single colony of S. aureus and S. epidermidis were cultured on a CRA plate [containing (37 g/L) Brain Heart Infusion (BHI), (50 g/L) sucrose, (10 g/L) agar and (8 g/L) Congo Red indicator (Himedia, Secunderabad, India)] at 37 • C for 24 h. Black colored colonies were indicative of biofilm positive isolates [64][65][66].

Crystal Violetmethod
The CV method is a quantitative method which monitors biofilm mass. CV is a positively charged molecule which binds to negatively charged bacteria and polysaccharides of EPS which could be quantitated [67][68][69]. In this method an overnight culture was diluted 10,000 times (v/v) and then 100 µL of this suspension was added to a 96 well polystyrene plate (Nunclon™, Thermo scientific, Roskilde, Denmark) containing 100 µL of BHI medium. Cultures were incubated at 37 • C for 4, 24, 48, 72, 96 and 120 h, after which the broth was decanted, the wells washed twice using 200 µL of phosphate-buffered saline, pH 7.4 (1X PBS contains 137 mM NaCl, 2.7 mM KCl, 10 mM Na 2 HPO 4 and 1.8 mM KH 2 PO 4 ) (PBS) (Sigma Chemical Co., St. Louis, MO, USA), plates air dried at room temperature (RT) and bacterial cells that had adhered to the wells were stained using 0.1% CV (Sigma Chemical Co., St. Louis, MO, USA). Excess crystal violet was discarded and each well was washed twice with 200 µL of PBS and dried at RT. CV associated with the bacteria was extracted with 200 µL of absolute ethanol and quantified using a Spectrophotometer [SpectraMax M3, with a cuvette adaptor (Molecular Devices, San Jose, CA, USA)] set at 595 nm [67][68][69]. Wells without cells served as the control (OD was < 0.1 at 595 nm) and the OD value was deducted from the biofilm positive (OD > 0.3 at 595 nm) and biofilm negative strains (OD < 0.3 at 595 nm) [62,63]. The experiment was performed with three replicates. XTT was also used for the detection of biofilm [63,70]. The assay is based on the cleavage of the tetrazolium salt XTT to water-soluble formazan (orange color) due to the metabolic activity of the organism. Formazan formed is quantitated using a scanning multi-well spectrophotometer. The measured absorbance directly correlates to the number of viable cells. In this method cultures were diluted 10,000 times and incubated as in the CV method. Subsequently, media was decanted, each well washed twice using 200 µL of autoclaved milliQ water and allowed to air dry for 30 m. was added to each well and incubated in the dark at 37 • C for 3 h. From each well, 100 µL was then transferred to a new 96 well plate and biofilm formation was quantified using a was quantified at 490 nm using a SpectraMax M3, microplate reader (Molecular Devices, CA, USA). A well without the inoculums served as a blank and E. coli ATCC 25922 and S. aureus ATCC 25923 were used as negative and positive control, respectively, for biofilm production. Experiment was performed with three replicates.

Monitoring the Thickness of the Biofilm and Visualisation of Extracellular Polymeric Substance (EPS) by Confocal Laser Scanning Microscopy
EPS and thickness of the biofilm (on cover slips) was monitored by CLSM using dual staining [63]. An overnight culture was diluted as in the CV and XTT methods and the bacterial suspension was added to a glass cover slip (Blue star, Mumbai, India) placed in the well of a 12 well polystyrene plate (Nunclon™, Thermo Scientific, Roskilde, Denmark) and incubated at 37 • C for 4, 24, 48, 72, 96, and 120 h. After the incubation period, the cover slips were washed with autoclaved distilled water and fixed with 250 µL of formaldehyde (4%) for 3 h. Fixed biofilms were then washed twice as above and stained for 30 min with 200 µL of 1.67 µM Syto ® 9 (Invitrogen, Carlsbad, CA, USA), a nuclear fluorescent dye, that stains DNA of viable cells and emits green color. After staining with Syto 9, biofilms were stained in the dark with 0.025% Calcofluor white M2R (Sigma Chemical Co., St. Louis, MO, USA) for 30 min. This dye binds to β-linked polysaccharides and fluoresces under long-wave UV light and biofilm could be visualized (blue) using confocal microscopy. Calcofluor white was excited at 363-nm using a 455/30 band-pass filter [71] and emits blue color. The thickness of the biofilm at each time point was measured across the entire biofilm and the values are reported as (Z axis, Average ± standard deviation in µm). Calcofluor white has been used to study exopolysaccharides (EPSs) involved in biofilm formation in a variety of organisms [72].

Visualisation of Biofilm on Cover Slips by Scanning Electron Microscopy
Cultures were processed for biofilm formation as above on glass cover slips (Blue star, Mumbai, India) and were then transferred to a new 12 well polysytrene plate (Nunclon™, Thermo scientific, Roskilde, Denmark), washed thrice with autoclaved distilled water and fixed for 3 h with 250 µL of glutaraldehyde (2.5%). After fixation, the glass cover slip was washed thrice with autoclaved distilled water, dehydrated for 20 m through graded ethanol (10,25,50,70,90 and 100%) and finally air dried overnight. Biofilms on the cover slips were sputtered with gold for 60 s using a High Vacuum Evaporator (SC7620 PALARON Sputter Coater, Quorum Technologies Ltd., East Sussex, UK) and visualized using a scanning electron microcope (SEM) (Carl Zeiss-Model EVO 18, Carl Zeiss, Germany). The Voltage used for acquiring the SEM images ranged between 5-20 kV.

Visualisation of Biofilm on Human Cadaveric Cornea by Scanning Electron Microscopy
Human cadaveric cornea, which do not meet the stringent quality required for transplantation were obtained from The Ramayamma International Eye Bank (RIEB), LVPEI, Hyderabad, India. All corneas were obtained following procedures approved by the institutional review board for the protection of human subjects. Corneas were received in MK medium containing gentamicin [73]. Therefore, they were thoroughly washed with PBS and biofilm formation was set up as described earlier [74]. The cadaveric corneo-scleral button (cornea + 2 mm of peripheral sclera) was placed on a 35 mm Petri-dish with the endothelial layer facing up. In this orientation, the cornea appears as a shallow cup. Into this cup, 500 µL of a semisolid Dulbecco's modified Eagle's medium (DMEM) (© 2021 Merck KGaA, Darmstadt, Germany) with agarose (0.5% w/v) was transferred and allowed to solidify. The corneas were then inverted so that the epithelial side was now facing upward. The cornea was then immersed in an antibiotic free DMEM culture medium containing 10% fetal calf serum, 5 µg/mL insulin and 10 ng/mL epidermal growth factor and incubated for 24 h at 37 • C in a 5% CO 2 incubator, to remove the residual antibiotics. Corneas from the antibiotic free DMEM medium were washed with PBS and a sterile steel scalpel was used to create three vertical and horizontal cuts [74]. Subsequently, the bacterial inoculum from an overnight culture grown in BHI media was diluted 10,000 times with BHI broth and centrifuged at 12,000 rpm (Eppendorf USA, Framingham, MA, USA, model no: 5430) for 5 min at room temperature (25 • C) and the pellet washed with 200 µL of autoclaved distilled water and centrifuged. The final pellet was suspended in 100 µL of DMEM (without fetal calf serum and antibiotics) and was gently transferred onto the surface of the corneas and incubated for 4, 24, 48, 72, 96, 120 h at 37 • C in a CO 2 incubator (5% CO 2 in air). After the incubation period, the cornea were processed for SEM to visualize biofilms on the cornea.

Antibiotic Susceptibility in Planktonic and Biofilm Phase
Several antibiotics were evaluated for their antimicrobial activity as per Clinical and Laboratory Standards Institute guidelines [75]. For this purpose, the overnight bacterial suspension in BHI was diluted 10,000 times and 100 µL of the suspension was added to each well of the 96 well polystyrene plate (Nunclon™, Thermo scientific, Roskilde, Denmark) containing 100 µL of an antibiotic of a known concentration [62,63]. Minimum inhibitory concentration (MIC) for each antibiotic was determined, as outlined in the CLSI-M07-A10 guidelines (CLSI, 2012).
For monitoring the inhibitory effects of the antibiotics, in the biofilm phase, cultures were allowed to form biofilms in the 96 well plate for 96 h, washed twice with PBS to remove planktonic bacteria and then known concentrations of the antibiotics were added and further incubated for 24 h. After incubation, the wells were gently washed using 200 µL of PBS to remove the free cells and the plates were then processed for monitoring the effect of the compound on the biofilm by the XTT method [62,63]. Inoculums without the addition of the compound served as a negative control. All experiments were performed in triplicate.

Biofilm Formation in Ocular Fluid Staphylococcus aureus and S. epidermidis
Ocular fluid S. aureus and S. epidermidis grew as black colonies on CRA plates indicative of biofilm formation ( Figure 1A,B). Further, the Crystal Violet and XTT methods confirmed the biofilm formation potential of the two isolates ( Figure 1C,D). In the CV method both S. aureus and S. epidermidis showed increase in biofilm formation from 4-48 h and reached a peak at 72 h (p ≤ 0.05) after which it stabilized (96 and 120 h). This was also in accordance with S. aureus ATCC 25923 the positive control for biofilm formation which also showed a similar trend in increase in biofilm formation. The negative control E. coli ATCC 25922 did not show any biofilm formation ( Figure 1C). Interestingly, when biofilm formation was monitored by the XTT method it was observed that both S. aureus and S. epidermidis were probably more efficient in biofilm formation potential and by 48-72 h reached peak biofilm formation which was sustained till 120 h ( Figure 1D). Experiments were performed in triplicates.
CLSM studies indicated that both in S. aureus and S. epidermidis the biofilm on cover slips stained positively between 4-120 h and EPS, which appeared blue in color, was clearly visible between 48-120 h (Figure 2A) in S. aureus but in S. epidermidis EPS was observed by 4 h and was visible even at 120 h ( Figure 2B). CLSM studies also indicated that the thickness of biofilm of S. aureus on cover slips increased from 2 ± 0.25 to 11.96 ± 0.90 µm between 4 to 72 h of biofilm growth, after which between 72-120 h the thickness sustained between 11.96 ± 0.90 µm and 12.99 ± 0.46 µm at 120 h ( Figure 2A and Table 1). In S. epidermidis also the thickness of the biofilm increased with time and continued to increase in thickness from 4-96 h after which there was slight decrease in thickness ( Figure 2B and Table 1). Statistical analysis indicated that the thickness of the biofilm both in S. aureus and S. epidermidis was significantly increased after 48 h compared to the thickness at 4 h ( Table 1).

Monitoring Biofilm Formation in Ocular Fluid Staphylococcus aureus and S. epidermidis by Scanning Electron Microscopy Using Cover Slips as a Substratum
Ocular S. aureus at 4 h were attached to the cover slips as dispersed cells in groups of 2 or more or and mostly as small mono-layer of cells (Figure 3). At 24 h, a microcolony of multilayer clumping of cells was visible (>4 layers of cells) which also gets transformed into column-like structures between 24-48 h ( Figure 3). Copious amount of EPS was formed between 72-120 h of biofilm formation (E in Figure 3). Intercellular connections were seen at 4, 24, 48 and 96 h (single arrows in Figure 3) but not clearly visible at 72 h, may be due to excessive of EPS at this stage. By 96 to 120 h, the clumps showed a tendency to disperse and several single cells were visible. The morphology of the cells was discernible between 4-120 h of biofilm formation. (Figure 3).
In ocular fluid S. epidermidis also, attachment of cells to the substratum occurred at 4 h and small multilayer clumps which by 24 h were very prominent (Figure 4). Column-like structures were seen between 48-120 h ( Figure 4) and EPS were clearly seen at 24-120 h of biofilm formation. Intercellular connections were visible at all stages of biofilm formation from 4-120 h (arrow) and were prominent at 96 h (Figure 4). Two adjacent prominent columns were visible at 120 h and the morphology of the cells in part of the column was obliterated (Figure 4). film formation from 4-120 h (arrow) and were prominent at 96 h (Figure 4). Two adjacent prominent columns were visible at 120 h and the morphology of the cells in part of the column was obliterated (Figure 4).    For each sample, 10 randomly selected microscopic fields were chosen and the thickness was measured and the average and standard deviation for each time point was calculated. Significant change in thickness was calculated against the thickness at 4 h based on t-test and p-value calculation. Indicates significant increase in biofilm formation compared to 4 h (p ≤ 0.05).

Figure 3. Scanning Electron Microscopy monitoring of biofilm formation in ocular surface Staphylococcus aureus L-1054/2020(2) between 4 to 120 h of biofilm growth on a cover slip. Single arrows represent intercellular connections (4-120 h), E represents EPS (72-120 h), C represents a column (24 and 48 h), M represents micro-colony (24 h) and W represents a water channel (72 h). Inset at 4 and 96 h represent intercellular connections and inset at 72 h represents EPS.
Microorganisms 2021, 9, x FOR PEER REVIEW 10 of 19

Monitoring Biofilm Formation in Ocular Fluid Staphylococcus aureus and S. epidermidis by SEM Using Cornea as a Substratum
Biofilm formation by ocular S. aureus on cadaveric cornea was different from that observed on cover slips. By 4 h, cells adhered to the cornea, micro-colonies of multilayer cells and a small column were observed ( Figure 5). Between 24-96 h, column-like struc-

Monitoring Biofilm Formation in Ocular Fluid Staphylococcus aureus and S. epidermidis by SEM Using Cornea as a Substratum
Biofilm formation by ocular S. aureus on cadaveric cornea was different from that observed on cover slips. By 4 h, cells adhered to the cornea, micro-colonies of multilayer cells and a small column were observed ( Figure 5). Between 24-96 h, column-like structures were visible ( Figure 5). By 48 h the column-like structures were less prominent and by 72 h, these structures were enclosed in EPS and not visible. By 96 h the columns reduced in size. A few single cells were visible by 48 h and the number of single cells increased by 120 h. The appearance of the single cells implied that the biofilm had entered the dispersal phase ( Figure 5). In contrast to the biofilm on cover slip S. aureus also showed prominent water channels at 24 and 96 h of biofilm growth. Further, intercellular connections were seen between 4-120 h of biofilm formation ( Figure 5).
In ocular fluid S. epidermidis also attachment of cells and multilayer clumps of cells visible at 4 h and by 24 h column-like structures developed which were also visible at 72 h of biofilm formation ( Figure 6). EPS was present in copious amounts between 24-120 h and intercellular connections were seen between 4 and 48 h of biofilm formation ( Figure 6). The appearance of single cells between 48 to 120 h was indicative of the dispersal phase of the biofilm (Figure 6).

Antibiotic Susceptibility in Ocular Staphylococcus aureus and S. epidermidis in the Planktonic and Biofilm Phase
A total of 22 different antibiotics were evaluated for their MIC on the growth of ocular fluid Staphylococcus aureus and S. epidermidis. In the biofilm phase, the MIC increased several fold compared to the bacteria in the planktonic phase. In the biofilm phase Staphylococcus aureus showed 2.6 (monocycline) to 51.2 (gatifloxacin) fold increase in MIC whereas S. epidermidis showed 3.2 (vancomycin) to 85.3 (amikacin) fold increase in MIC depending on the antibiotic used (Table 2)

Discussion
This study confirms earlier observations demonstrating that ocular Staphylococcus aureus and S. epidermidis collected from the cornea, conjunctiva, eyelid margin, intraorbital foreign body, intraocular lenses, vitreous and aqueous humors of corneal ulcer patients and from patients with other ocular diseases including endophthalmitis exhibit biofilmforming capacity as determined by CRA [54], CV [54,76], and SEM methods [55,77,78]. However, temporal dynamics of the biofilm formation by ocular Staphylococcal spp. from attachment to dispersal phase were rarely monitored [53,54,58]. Hou et al. [54] using SEM, observed that ocular biofilm-positive staphylococcal strains go through a phase of adhesion and accumulate as small monolayer sheets which then get transformed into multilayer sheets enclosed in the self-secreted EPS. By 24-72 h, biofilm structures (vertical columns or mushroom-like assemblies) were formed, and bacterial clusters were enclosed in the self-secreted EPS. Water channels and thread-like appendages between cells were also distinctly observed [54]. The present study on ocular fluid Staphylococcus aureus and S. epidermidis when grown on cover slips or cadaveric cornea demonstrated the potential of the two isolates to form biofilm on both abiotic and biotic substrata. When cover slips was used as the substratum both S. aureus and S. epidermidis go through an adhesion phase by 4 h. This phase is a key stage in the formation of biofilms and surface structures such as pili, fimbriae and flagella are very important in studies related to the dynamics of biofilm formation [48]. These motility organelles such as fimbriae [79], pili and flagella [80] and other surface proteins such as autolysin [81,82], exo-polysaccharides [83], extracellular DNA [84] and bacterial microbial surface components [85] interact with matrix proteins such as fibrinogen [86] and fibronectin [87] and facilitate adhesion of the cells to the substratum [88]. Castonguay et al. [89] demonstrated that a particular strain of E. coli PHL565 was unable to attach to solid surfaces and form a biofilm, but the strain of E. coli PHL565, in mixed cultures with Pseudomonas putida MT2 resulted in co-adhesion and in the formation of a mixed E. coli and P. putida biofilm, on glass surfaces. In contrast, E. coli with Staphylococcus epidermidis did not form a biofilm. It was suggested that mixed biofilms might represent an important mechanism, and a possible alternative strategy to form a biofilm when one of the partners does produce adhesion determinants. Such a strategy may help to increase the virulence of bacteria with low biofilm-forming potential. After the adhesion phase, in the proliferation phase, multi-layered colonies were formed between 24-48 h in S. aureus ( Figure 3) and in S. epidermidis by 24 h (Figure 4). The maturation phase when the biofilm appears as vertical columns or mushroom-like assemblies enclosed in EPS was not very prominent in S. aureus (Figure 3, 24 h) but very well developed in S. epidermidis (Figure 4, 24-120 h). A feature that discriminated biofilm formation in ocular fluid S. aureus and S. epidermidis is that the former exhibited single cells between 96-120 h indicative of dispersion phase but in the same time frame S. epidermidis was still in the pen-ultimate maturation phase exhibiting a prominent column (Figures 3 and 4). In the dispersal phase individual cells and/or multicellular aggregates are dispersed from the mature biofilms to seed new biofilms [20,41,42]. In contrast when biofilm formation of ocular fluid S. aureus and S. epidermidis were monitored on cadaveric cornea the dynamics of biofilm formation differed from that observed on cover slips. Both the isolates went through the 4 phases. Adhesion and microcolony formation were observed by 4 h, column-like structures were visible in both between 24 h to 96 h and the dispersion phase indicated by the appearance of single cells was visible by 48 h and beyond (Figures 5 and 6). Thus biofilm formation appears to be more rapid on cadaveric cornea compared to when cover slip was used as the substratum. These studies imply that the temporal dynamics of biofilm formation was dependent on the substratum on which the biofilm was grown.
Earlier studies on biofilm formation in ocular isolates of S. aureus and S. epidermidis were done using synthetic material such as silicone, polymethyl-acrylate, hydrophilic acrylic, hydrophobic acrylic, etc. [25,62,63,90], to understand their biofilm formation ability on indwelling devices including contact lenses, sutures, scleral buckles, valvular tubes and keratoprostheses [20,[24][25][26]52,77,[91][92][93]. In our opinion, monitoring biofilm formation of ocular fluid S. aureus and S. epidermidis on cadaveric cornea is equally important because they are a major source of hospital-acquired infections and more importantly, corneal biofilms have been reported following experimental keratitis in mice [77], in patients with infectious crystalline keratopathy [94][95][96] or pterygium scleritis [97] and also in the absence of prosthetic material and in the absence of active corneal inflammation or infection [59]. Comparison of the biofilms formed on cover slips with that of cadaveric cornea indicated that EPS secretion was more copious on cornea (Figures 5 and 6) than on cover slip (Figures 3 and 4). Unknown host factors associated with the cadaveric cornea may facilitate copious EPS secretion. Generally bacterial cells bind to the host binding molecules, including fibrinogen, host extracellular matrix proteins, fibronectin and components of the blood plasma and such proteins have been detected as mediators of adhesion in clinical Staphylococcal isolates [98,99] which facilitate the binding of bacterial adhesins [100] to the surface.
A biofilm is also defined based on its architecture [29,30,[32][33][34] which includes, in addition to the microbial cells and EPS, also some structural attributes such as the thickness (varying from monolayer of cells to structures such as columns and mushrooms) water channels [35] and the presence of voids. In this study it was observed that EPS was prominent in the biofilms of both the ocular fluid S. aureus and S. epidermidis and more so when the two were grown on cadaveric cornea. This observation is not surprising since EPS is an integral part of the biofilm architecture and is also known to maintain the structural integrity of the biofilm, anchor the biofilm to a substratum [44], facilitate cell to cell communication and the viability of cells by modulating substrate absorption, oxygen diffusion and transport of molecules within the biofilm [45,46]. In this study, thickness of the biofilm, yet another architectural feature was monitored by confocal microscopy as reported earlier [56,57]. The thickness of the biofilm ofocular fluid S. aureus and S. epidermidis increased with time and this is in accordance with the CV and XTT results which indicated increase in biofilm formation with time. However, based on the SEM results one should have observed a decrease in thickness by 120 h when the biofilm attains the dispersal phase. The observed discrepancy could be attributed to the fact that in the CV and XTT methods the substratum used was a polystyrene plate (Nunclon™, Thermo scientific, Roskilde, Denmark) which is different from the glass cover slip and definitely different from the cadaveric cornea. Earlier studies have indicated that biofilms release and disperse cells into the environment to colonize new sites [101]. This dispersal phase is a complex process and involves numerous environmental signals, signal transduction pathways and effectors [102]. That decrease in biomass is due to dispersal and not death of cells was elegantly demonstrated by Barraud et al. [103] who assessed biofilm dispersal as a concomitant decrease in biofilm biomass and an increase in planktonic biomass [103]. Other biofilm architectural features such as columns were also seen when ocular fluid S. aureus and S. epidermidis were grown either on cover slips or cadaveric cornea (Figures 3-6) and water channels could also be identified only when grown on cadaveric cornea. Voids were not detected in the biofilms since voids are normally detected by time lapse monitoring of the biofilm after every 15 min [36][37][38].
Our studies also confirm that cells in the biofilm phase are several fold more resistant to antibiotics, a phenomenon associated with biofilm formation in bacteria [20,[26][27][28]62,63]. Earlier studies had indicated that several ocular bacteria including S. epidermidis, S. aureus and Streptococcus spp. form biofilms [104] and majority of them were resistant to antibiotics. Further, our results were in accordance with these earlier studies which had demonstrated that the MIC of the antibiotic in the biofilm phase was significantly greater than that required for killing the cells in the planktonic phase [91,105]. This increase in MIC in the biofilm phase could be attributed to: inefficient penetration of the drug into the biofilm [106], inability of the drug to exerts its effect within the biofilm [107], transformation of the microorganisms in the biofilm into viable-but-nonculturable state [108], emergence of persister cells which are resistant to drugs [109], ability to survive under nutrient and oxygen limitation conditions and up-regulation of drug resistance-associated genes [105], ability of EPS to limit diffusion of aminoglycosides [110], ability of EPS to inactivate antibiotics [111], acquiring resistance to phagocytosis and induction of LPS modification genes [112]. A few of the strategies that have been demonstrated in ocular isolates as responsible for AMR include biofilm formation as indicated above and a high concordance between the presence of AMR genes and antibiotic resistance in 10 ocular E. coli strains and the presence of several virulent genes (fimB to fimI, papB to papX, etc.) and prophages (Enterobacteria phage HK97, Enterobacteria phage P1, Escherichia phage D108, etc.) which were unique to ocular E. coli [62,63,113]. To the best of our knowledge, this is probably the first study on biofilm forming potential of ocular S. aureus and S. epidermidis on cadaveric cornea from the attachment to the dispersal phase of biofilm formation.