Five S. aureus isolates from the strains library at the Applied Microbiology and Food Safety Laboratory, School of Veterinary Medicine, Texas Tech University, were used. These strains were selected after testing a larger group of isolates based on their ability to form biofilm at similar rates. A drug-resistant and biofilm-forming Staphylococcus aureus subsp. aureus Rosenbach ATCC 43300 was included as a reference strain/positive control. The isolates were cryopreserved at −80°C with 20% glycerol, and prior to the experiments, they were streaked onto Tryptic Soy Agar (TSA) (Remel, Lenexa, KS, United States) and incubated at 37°C for 24 h. After incubation time, one colony was transferred to a test tube containing 9 mL Brain Heart Infusion broth (BHI) (Millipore Sigma, Burlington, MA, United States) and incubated in motion at 37°C for 24 h at 200 rpm. Upon incubation, a bacterial suspension was created by transferring 0.1 mL of the overnight BHI culture into a 9.9 mL BHI tube and mixing thoroughly to obtain a starting optical density (OD) of 0.003 ± 0.001 at OD 600 nm (approximately 10⁷ CFU/mL), which was measured by transferring 100 μL of the bacterial suspension into a 96-well plate and using a microplate reader. Strains were grown and tested individually.

To allow biofilm formation, we used the microtiter dish biofilm formation assay described elsewhere (O'Toole, 2011). In brief, from the bacterial suspension prepared for each strain, 200 μL were deposited in triplicate into a 96-flat bottom well plate (Costar®, Washington, D.C., United States) and incubated at 37°C for 48 h in static conditions to permit biofilm formation. Upon incubation, the broth was removed from each well, and the wells were rinsed three times with 300 μL of sterile distilled water. To aid the biofilms being fixed in the wells, 200 μL of 90% ethanol were added and left for 15 min. Finally, the ethanol was removed, and the plates were allowed to air dry for 1 h. The biofilm was stained with 200 μL of 1% crystal violet (CV) (Carolina®, Burlington, NC, United States), which remained in the wells for 45 min. The CV was removed, and the 96-well plate was washed by submerging it in distilled water. The latter was repeated three times. Lastly, upside-down plates were allowed to air dry overnight at room temperature, and biofilms were formed at the bottom of each well. Biofilm density was quantified further by measuring optical density. A 200 μL aliquot of 70% ethanol was added to each well to destain the CV and allowed to set for 3 min. Following this, 100 μL of the last suspension was transferred to a new 96-well plate. Subsequently, the plate was placed on a Microplate Reader LX Multi-Mode Microplate Reader (Synergy LX Multi-Mode, BioTek, Winooski, VT, United States) to measure absorbance at optical density (OD) of 580 nm.

Antibiotics were selected based on mode of action, concentration, and exposure time. The antibiotics tested were grouped based on the key aspect of their effectiveness: concentration-dependent antibiotics (amikacin, tobramycin, gentamicin, levofloxacin, and ciprofloxacin), and time-dependent antibiotics (imipenem, meropenem, and doripenem). The inclusion of both types of antibiotic activities for the experimental design allows for a thorough evaluation of biofilm formation under different antibiotic exposure dynamics. Diflunisal was chosen due to its documented inhibitory effects on the disruption of S. aureus Agr system. All chemical compounds were provided by the Department of Pharmaceutical Sciences, School of Pharmacy, Texas Tech University Health Sciences Center, in Amarillo, Texas. Based on their solubility, antibiotics and diflunisal were reconstituted in autoclaved distilled water (DW) or dimethyl sulfoxide (DMSO) (Sigma-Aldrich, St. Louis, MO, United States) (Table 1). After reconstitution, the stock solutions of all compounds were stored at −20°C. For each experiment, stock solutions were thawed in a water bath at room temperature and diluted in autoclaved DW to obtain the desired concentrations. The antibiotics were tested at different concentrations, and the concentrations chosen were based on the quality control range of acceptable minimum inhibitory concentration (MIC) for S. aureus provided by the M100 Performance Standards for Susceptibility Testing document by Clinical and Laboratory Standards Institute (2021). Diflunisal was tested only at one concentration (25 μg/mL), and this was chosen based on literature in which it is reported that at this dose, it inhibits S. aureus quorum sensing (Bernabè et al., 2021). Table 1 describes the details and specific concentrations of each compound tested.

For this set of experiments, a total of eight antibiotics were tested with and without diflunisal. Two different treatment scenarios were investigated. The first scenario was the individual treatment, which consisted of testing diflunisal individually (no antibiotic) and each antibiotic independently without diflunisal on one S. aureus strain at the time by adding into 100 μL of the target drug into a well of a 96-well plate, followed by 100 μL of bacterial suspension. The second scenario consisted of a co-treatment, which involved the evaluation of the antibiotics in combination with diflunisal on each S. aureus strain; in this case, 50 μL of diflunisal solution, 50 μL of the antibiotic solution, and 100 μL of the bacterial suspension were added into a well. In both cases, experiments were conducted in triplicate; plates were incubated at 37°C for 48 h under static conditions to allow biofilm formation. Upon incubation, biofilm density (OD 580 nm) was quantified using crystal violet as described in section 2.2. Specific concentrations of the antibiotics and diflunisal are listed in Table 1.

For these experiments, only amikacin, imipenem, doripenem, levofloxacin, and ciprofloxacin were tested. The selection of these five antibiotics was based on results from the previous experiments, as they exhibited the highest effect in controlling S. aureus biofilm formation. To evaluate the effect of antibiotics in combination with diflunisal over time, each isolate was subjected to biofilm growth individually and treated with the five antibiotics, both in the presence and absence of diflunisal. Biofilm density was measured every 4 h at an OD of 580 nm, during 48 h. The selection of time points was considered based on the different stages of growth and biofilm formation.

To evaluate the effect of diflunisal in combination with antibiotics over time on S. aureus planktonic growth, growth curves for the isolates used were created by incubating each isolate independently in BHI broth at 37°C with each of the five antibiotics previously chosen, using the same concentrations tested for biofilm formation as described in Table 1. Experiments were conducted with and without diflunisal. The treatments of antibiotics without diflunisal were conducted to evaluate if there was a statistically significant difference between the effect of antibiotics alone and combined with diflunisal on planktonic growth, while untreated S. aureus served as the control to establish a baseline for comparison. Bacterial concentration was measured over time every 4 h for 48 h based on absorbance at an OD of 600 nm. To analyze the data, the absorbance (OD 600 nm) values obtained for each isolate (including S. aureus ATCC 43300) were averaged.

Antibiotics were quantified using a combination of previously reported methods with modifications. Briefly, 10 mg/mL stock solutions of the test drugs were prepared in either water or DMSO. After serial dilutions, 25 μg/mL solution of diflunisal (in BHI) was mixed with 25 μg/mL solution of each antibiotic (in BHI). The standard curve for diflunisal was also prepared in the absence of antibiotics to accurately quantify the concentration of diflunisal without interference from antibiotics. From each sample, test sample, or standard curve sample, new samples were prepared composed of 50 μL of diflunisal-antibiotic sample or standard curve sample + 50 μL of internal standard (prepared in ACN: H2O, 70: 30) + 150 μL of ACN: H2O, 70: 30. The samples were quantified using LCMS/MS via the following method: For analysis of samples, AB SCIEX QTRAP 5500 triple quadrupole mass spectrometer, attached to a Nexera UPLC system (Shimadzu Corporation), was used. The UPLC system contained an autosampler (Sil-30 AC), pumps (LC-30 AD), a controller (CBM-20A), a degasser (DGA-20A5), and a column oven (CTO-30A). Analyst software was used for data acquisition and quantification. Detailed information for the chromatographic separation of test compounds is summarized below. Column: Gemini-C18 (2 mm × 50 mm, 3 μm; Phenomenex, Torrance, CA, United States); mobile phases: (A): 2 mM Ammonium Formate in water, (B): ACN; flow rate: 0.4 mL/min; gradient: 0–0:10 min; 30% B, 0:10–1 min; to 90% B, 1:0–2:0 min; 90% B, 2:0–2:10 min; to 30% B, 2:10–3:0 min; 30% B.

A microscopy live/dead assay of S. aureus cells within the biofilms subjected to the treatments was performed. Biofilms were grown and treated according to the protocols described in sections 2.2 and 2.4, with the assay scaled to a 24-well plate (Corning™ Costar™, Corning, NY) format. Since no statistical differences were observed in biofilm densities across the isolates used, a bacterial mix was prepared by combining equal volumes of the six isolates. From this bacterial cocktail, a suspension was created by diluting 0.1 mL of the cocktail in 9.9 mL of BHI. To treat the biofilms with antibiotics individually, 500 μL of the drug was added to each well, followed by 500 μL of the bacterial suspension, resulting in a final volume of 1 mL. For treatments combining antibiotics with diflunisal, 250 μL of the NSAID, 250 μL of the antibiotic, and 500 μL of the bacterial suspension were added to each well. The plate was incubated as specified above. After incubation, the remaining media was removed, and the wells were washed three times with sterile distilled water. Biofilms were prepared for microscopy using the Live/Dead BacLight Bacterial Viability Kit (Molecular Probes, Inc., Eugene, Oregon). A staining solution was prepared by mixing 1 μL of Propidium iodide (for dead cells) and Syto 9 (for live cells) with 148 μL of sterile distilled water per well. The solution was added to each well, and the plate was incubated at room temperature for 15 min, protected from light. Subsequently, confocal laser scanning microscopy (CLSM) was performed using the Leica Stellaris 8 Falcon STED Confocal Microscope (Leica Microsystems, Wetzlar, Germany) to visualize the cells in the biofilms.

Data management was done using a Microsoft Excel spreadsheet (2016, Microsoft Corp., Redmond, WA), and it was imported into R software for analysis. For the evaluation of the effect of antibiotics and diflunisal on S. aureus biofilms after 48 h of treatment, the focus of the analysis was to explore if there was any statistical difference in S. aureus biofilm density (OD 580 nm) between each treatment type (i.e., antibiotics versus antibiotics combined with diflunisal, at the different antibiotics concentrations). Initial descriptive statistics were carried out to summarize S. aureus biofilm density after the treatments and were presented graphically as a mean plot with respective standard error of the mean. For data analysis, a two-way ANOVA was used. Considering the eight antibiotics (amikacin, tobramycin, gentamicin, imipenem, meropenem, doripenem, levofloxacin, and ciprofloxacin) with different concentrations; eight different two-way ANOVA were conducted. In each model, standard parametric model assumptions for two-way ANOVA were explored. The Shapiro–Wilk test and quantile plot were used to explore the normality test statistically and graphically, respectively. The assumption of homoscedasticity was employed using Levene’s test. Whenever there was a violation of parametric assumption, robust two-way ANOVA was conducted. When a variable was statistically significant, a post-hoc comparison of the treatment type and/or antimicrobial concentrations was carried out, adjusting for multiple comparisons using Bonferroni correction.

On the other hand, to analyze the effect of the five antibiotics chosen (amikacin, imipenem, doripenem, levofloxacin, and ciprofloxacin) with or without diflunisal in biofilm density and planktonic growth over time (at 4 h intervals for 48 h), we focused our analysis in exploring the difference in S. aureus biofilm and planktonic growth densities over time with the different antibiotic concentrations, individually or in combination with diflunisal. Initial descriptive statistics were carried out to summarize outcomes of interest over time and were presented graphically as mean plots with respective standard error of mean. For data analysis, a repeated measured ANOVA was used to account for possible correlation in S. aureus biofilm and planktonic growth densities over time. For each antibiotic, a repeated measured ANOVA was performed for all the S. aureus biofilm and planktonic growth densities as outcome variables. In each model, standard parametric model assumptions for repeated measure ANOVA were explored. Shapiro–Wilk test and quantile plot were used to explore the normality test statistically and graphically, respectively. Possible outlying observations were also investigated. The assumption of sphericity was explored using Mauchly’s test; whenever there was a violation of the sphericity assumption, Greenhouse–Geisser sphericity correction was applied to factors violating this assumption. Whenever there is a statistically significant interaction effect between the antibiotic concentration and time factors in the model, a statistical comparison of the concentration at each time point was done, adjusting for multiple comparisons, using Bonferroni correction. The interaction effects were presented graphically to explain the effects of these factors on the S. aureus biofilm and planktonic densities.

A statistically significant difference between the factors was considered at p < 0.05. All analyses were done using the computing environment of R software (v 4.2.2) using the appropriate packages.

Staphylococcus aureus subsp. aureus Rosenbach ATCC 43300 (known biofilm producer) was used as a reference strain for biofilm formation. After 48 h of incubation, its biofilm density average was 0.466 ± 0.04 (OD 580 nm) (Figure 1), and this value was used to determine biofilm production by S. aureus strains selected for this study. As depicted in Figure 1, all isolates evaluated were biofilm producers. The OD values for biofilm density were 0.49 ± 0.06, 0.67 ± 0.07, 0.53 ± 0.08, 0.48 ± 0.05, and 0.47 ± 0.05, for isolates 1 through 5, respectively. No statistical difference between the strains was observed (p > 0.05). The OD average for all isolates was 0.529 ± 0.02, and this value was used as the point of reference or baseline for the data analysis pertaining to the antibiotics and diflunisal effect on biofilm formation.

Changes in biofilm density were monitored in the presence of antibiotics with and without diflunisal. Considering that the chemical structures of the utilized antibiotics contain amine groups and given the inherently acidic nature of diflunisal (pKa 2.9), we evaluated if alterations in pH caused by mixing of various antibiotics with diflunisal could change the solubility of compounds in the media. Results showed no noticeable alteration in the solubility of the antibiotics when mixed with diflunisal vs. individual samples (Figure 2), ensuring the accurate delivery of proposed drug concentrations in the following experiments.

Biofilm density was tested after the antibiotic treatments with and without diflunisal. After the growth period of 48 h, statistical analysis comparing density before and after treatment revealed a reduction in biofilm formation, which was influenced by the antibiotic dose (p < 0.01) irrespective of diflunisal. However, when we compared the level of reductions achieved between treatments with diflunisal and without diflunisal, no significant differences (p > 0.05) were observed; in other words, the presence of diflunisal in combination with the antibiotics did not influence biofilm density. Table 2 describes the antibiotics tested, the concentrations, the percentage of biofilm reduction per each treatment, and their corresponding p value.

The effect of diflunisal alone (in the absence of any antibiotics) on S. aureus biofilm was tested using a concentration of 25 μg/mL. Findings show at the end of the growth period (48 h), a significant effect (p < 0.0001) in reducing biofilm density by 53.12%.

Biofilm density was clearly affected by exposure time. In the case of amikacin treatments, a prolonged time of exposure led to an increase in biofilm density. Conversely, fluctuations in biofilm formation over time were observed at specific intervals with imipenem, doripenem, levofloxacin, and ciprofloxacin. In general, at the lower doses of the antibiotics and in the absence of diflunisal, longer exposure demonstrated higher control over biofilm growth. Additionally, the time-dependent effect of antibiotic concentration on biofilm formation was evident in all treatments.

The effects of amikacin concentration (with and without diflunisal) and time on biofilm density were significant (p < 0.0001). As shown in Figure 3A, at 1 μg/mL and 4 μg/mL of amikacin without diflunisal, a biofilm density increase was observed from 0 to 24 h, but compared to the control group, the biofilm density produced was less. However, when diflunisal was added, a decreased efficacy of the treatment was observed between 28 and 48 h. No statistical difference in biofilm density was observed between treatments. Nonetheless, significant differences were observed among the different concentrations.

Regarding imipenem, significant effects of the antibiotic concentration and time on biofilm density (p < 0.0001) were observed, indicating that the impact of the antibiotic concentration was dependent upon the duration of exposure. Biofilm density of S. aureus increased over time with both antibiotic concentrations used individually and with diflunisal (Figure 3B). At 0.06 μg/mL of imipenem alone, the antibiotic caused a bacteriostatic effect from the 28-h time point. However, at both 0.016 μg/mL and 0.06 μg/mL concentrations of the antibiotic, the presence of diflunisal resulted in a diminished effect on the treatment over time. Nonetheless, there were no significant differences between the treatments in S. aureus biofilm density at 0.016 μg/mL and 0.06 μg/mL (except at 4 and 8 h time points, p < 0.036). Significant differences were observed among the control group, 0.016 μg/mL, and 0.06 μg/mL imipenem concentrations, regardless of diflunisal presence.

In the case of doripenem (Figure 3C), the ANOVA demonstrated that both doripenem concentrations and time had statistically significant effects on biofilm density (p < 0.0001). Biofilm density increased over time with 0.016 μg/mL and 0.06 μg/mL of the antibiotic alone and in combination with diflunisal. Additionally, at both concentrations of the antibiotic, the presence of diflunisal promoted biofilm growth, compared to the effect of the antibiotic alone. However, there were no statistically significant differences in S. aureus biofilm density between the treatments, except with 0.016 μg/mL at 44 h (p = 0.018), and 0.06 μg/mL at 4, 8, 12, and 36 h (p < 0.007). In addition, significant differences in S. aureus biofilm density were observed among the control group, 0.016 μg/mL, and 0.06 μg/mL concentrations of doripenem, regardless of diflunisal presence.

Levofloxacin effect on S. aureus biofilm formation is represented in Figure 3D. The repeated measure analysis of variance (ANOVA) indicated that both levofloxacin concentrations and time had statistically significant effects on biofilm density (p < 0.0001). Biofilm density increased over time in presence of both concentrations of levofloxacin with and without diflunisal. However, with 0.5 μg/mL of levofloxacin in the absence of diflunisal, the biofilm growth was lower compared to the 0.06 μg/mL treatments. Additionally, at both concentrations of the antibiotic, the presence of diflunisal caused a decrease in the treatment efficacy on the growth control of S. aureus biofilm over time. However, there were no significant differences in biofilm density with both concentrations of levofloxacin alone and with diflunisal at each time point, except with 0.5 μg/mL at the 20 and 40 h time points, where the biofilm density is higher compared to the treatment without diflunisal. However, there were statistically significant differences in S. aureus biofilm density among the control group, 0.06 μg/mL, and 0.5 μg/mL treatments, regardless of the presence of the NSAID.

Regarding ciprofloxacin, a statistically significant effects of the antibiotic concentration and time on biofilm density (p < 0.0001) was observed. As shown in Figure 3E, S. aureus biofilm density increased over time irrespective of ciprofloxacin concentrations and the presence of diflunisal. However, the treatment that exerted the most substantial control over biofilm growth was 0.5 μg/mL of ciprofloxacin in the absence of diflunisal between 20 and 36 h. With the three concentrations (0.12 μg/mL, 0.5 μg/mL, and 2 μg/mL), the presence of diflunisal decreased the efficacy of the treatments in controlling biofilm formation over time. However, there were no statistically significant differences in biofilm density observed between the 0.12 μg/mL treatments, the 0.5 μg/mL treatments (except at 28 h, where p = 0.041), and the 2 μg/mL treatments at each of the time points. Although, there were some statistically significant differences in S. aureus biofilm density between the control group and the treatments.

Staphylococcus aureus biofilm was treated with diflunisal alone. In the presence of 25 μg/mL of the NSAID, biofilm formation increased between 0 and 16 h, was static between 16 and 32 h, and increased again after 32 h. From the repeated measure ANOVA, the main effects of diflunisal concentration and time were statistically significant (p < 0.001), and the effect of diflunisal concentration on the biofilm density was time-dependent. No significant differences (p < 0.05) were observed between control and treatment group during the first 4 h of treatment, however, changes became statistically different comparing control and treatment group after 4 h of diflunisal exposure (Figure 4).

Growth curve assays for S. aureus permitted the examination of the impact of five specific antibiotics (amikacin, imipenem, doripenem, levofloxacin, and ciprofloxacin). The average density of untreated S. aureus increased over time with an observed lag phase from 0 to 4 h, an exponential phase from 4 to 12 h, and a stationary phase between 12 and 48 h of incubation (Figure 5). Overall, for all treatments, no bacterial reduction or death phase was observed, suggesting that all tested antibiotics produced a bacteriostatic effect.

In the case of amikacin (Figure 5A), with 1 μg/mL, a lag phase was observed from 0 to 4 h, the exponential growth of planktonic cells was observed between 4 and 12 h, and from 12 to 48 h the antibiotic caused a bacteriostatic effect. However, when diflunisal was added, caused a decrease in effectiveness to control bacterial growth control compared to amikacin alone during the 12–48-h period. At 4 μg/mL of amikacin alone and with diflunisal, the treatments caused a bacteriostatic effect, maintained over time. There was a notable difference between the use of 1 μg/mL and 4 μg/mL of amikacin, as the higher concentration suppressed planktonic growth from 4 to 48 h, which explains that the effect of the antibiotic on planktonic cell growth was dose-dependent. Additionally, a two-way interaction was observed between the concentration of amikacin and the exposure time of the bacteria to the antibiotic, indicating that the impact of amikacin concentration on S. aureus growth was contingent on time. Except for the 28 and 32-h time points, there was no significant difference in S. aureus growth between the use of 1 μg/mL of amikacin alone and in combination with diflunisal, or between the use of 4 μg/mL of amikacin with or without diflunisal. Nonetheless, statistically significant differences in S. aureus density were found between the control group (no treatment) and 1 μg/mL or 4 μg/mL of amikacin with or without diflunisal.

Regarding imipenem treatments (Figure 5B), with both concentrations of the antibiotic irrespective of diflunisal presence, exponential growth was observed from 0 to 12 h, and then from 12 to 48 h, the treatments showed a bacteriostatic effect. The statistical analysis indicated that both imipenem concentration (with and without diflunisal) and time, resulted in significant effects on microbial growth (p < 0.0001). Moreover, a two-way interaction between imipenem concentration and time was observed, implying that the impact of imipenem concentrations on S. aureus growth was dependent upon the duration of exposure to the antibiotic. Except for the 24 h time point, there were no significant differences in S. aureus growth between the use of 0.016 μg/mL of imipenem with and without diflunisal at each time point. Likewise, no significant effect was found between the treatment with and without diflunisal for the 0.06 μg/mL imipenem concentration at the different time points. Nevertheless, there were several statistically significant distinctions in S. aureus density among the control group and 0.016 μg/mL or 0.06 μg/mL of imipenem with or without diflunisal.

With doripenem treatments (Figure 5C), a bacteriostatic effect maintained over time was observed when using 0.06 μg/mL of the antibiotic individually and in combination with diflunisal, in comparison to 0.016 μg/mL of doripenem with and without the NSAID. At 0.016 μg/mL of doripenem alone, the exponential growth of planktonic cells was observed between 8 and 24 h of treatment, and then from 24 to 48 h the antibiotic controlled bacterial growth. However, the control over bacterial growth was more pronounced when utilizing 0.016 μg/mL of doripenem in combination with diflunisal. Furthermore, the ANOVA revealed that the effects of doripenem concentration (individually and in combination with diflunisal) were statistically significant (p < 0.0001), and a two-way interaction was observed between doripenem concentration and time concerning the impact on S. aureus growth. Additionally, there were no significant differences in bacterial growth between the use of 0.016 μg/mL of the antibiotic alone and in combination with diflunisal at each time point, except for the 24 h time point. Similarly, there were no statistically significant differences in the effect on S. aureus growth between the use of 0.06 μg/mL of doripenem with and without diflunisal at the various time points. Nonetheless, several statistically significant differences in S. aureus growth were observed among the control group, as well as the 0.016 μg/mL and 0.06 μg/mL doripenem concentrations with and without the NSAID.

Regarding levofloxacin (Figure 5D), at a concentration of 0.06 μg/mL of the antibiotic alone, S. aureus planktonic cell density increased between 0 and 12 h time points, then from 12 to 48 h the antibiotic exerted control over bacterial growth. When using 0.06 μg/mL of levofloxacin in combination with diflunisal, bacterial density increased between 4 and 28 h, and from 28 h to 48 h the mix controlled bacterial growth, maintaining it static. Additionally, when using 0.5 μg/mL of the antibiotic irrespective of diflunisal status, a bacteriostatic effect of the treatments was observed. The ANOVA indicated that both levofloxacin concentrations (individually and in combination with diflunisal) and time caused statistically significant effects on S. aureus growth (p < 0.0001). Additionally, a two-way interaction was observed between levofloxacin concentration and time in relation to the impact on S. aureus growth. There were no statistically significant differences in bacterial growth between the use of 0.06 μg/mL of levofloxacin alone and in combination with diflunisal, as well as between the use of 0.5 μg/mL of the antibiotic with and without the NSAID, at each of the time points. However, there were significant differences in S. aureus growth among the control group, the 0.06 μg/mL, and the 0.5 μg/mL levofloxacin concentrations, both individually and with diflunisal.

With respect to ciprofloxacin treatments (Figure 5E) on S. aureus growth, the treatments exerted control over bacterial growth in a dose-dependent manner. A bacteriostatic effect of ciprofloxacin was observed from 4 to 48 h at concentrations of 0.5 μg/mL and 2 μg/mL, regardless of the presence of diflunisal, compared to the 0.12 μg/mL treatments. At the 0.12 μg/mL concentration of ciprofloxacin in combination with diflunisal, a decrease in the bacteriostatic effect over S. aureus planktonic cells between 12 and 16 h, whereas the growth control was greater between 24 and 48 h compared to the effect of the antibiotic alone. The statistical analysis (ANOVA) revealed significant effects of ciprofloxacin concentration (with and without the NSAID) and time on S. aureus growth (p < 0.0001), indicating that the impact of antibiotic concentration on bacterial growth was dependent on the duration of exposure. Moreover, there were no statistically significant differences in S. aureus planktonic cell growth between the use of 0.12 μg/mL of ciprofloxacin alone and in combination with diflunisal at each time point, as well as with the 0.5 μg/mL and 2 μg/mL treatments. However, there were statistically significant differences in S. aureus growth among the control group, the 0.5 μg/mL, and the 2 μg/mL ciprofloxacin concentrations, both individually and in combination with diflunisal.

The effect of time using diflunisal was evaluated individually against S. aureus growth and compared to bacterial growth without any treatment (control group). As shown in Figure 6, diflunisal exerted a bacteriostatic effect over S. aureus planktonic cells from 12 to 48 h. A two-way interaction was observed between diflunisal presence and time. In addition, there was a statistical difference regarding control over bacterial growth between the control group (no treatment) and diflunisal treatment from 12 to 20 h and 28 to 36 h of incubation.

Cell images captured using a live/dead assay with CLSM were able to show the effect of the treatments and cell viability within the biofilm matrix (Figure 7). The untreated biofilm (control) exhibited a high prevalence of live cells (green), indicating large viability in the absence of treatment. At 1 μg/mL of amikacin the image shows a mix of live and dead cells within the biofilm, suggesting a partial bactericidal effect at this concentration. When combined with diflunisal, the red staining increased (dead cells), which could indicate an enhanced killing activity. At 4 μg/mL, amikacin alone shows greater cell death, and this effect is maintained when combined with diflunisal.

For levofloxacin, the cells within the biofilm are mostly dead when treated with 0.06 μg/mL and when combined with diflunisal, the effect is sustained. When levofloxacin was used individually at 0.5 μg/mL there was a smaller number of aggregated cells, and most of them showed to be dead. However, when levofloxacin was combined with diflunisal, S. aureus viability increased, as well as cell aggregation.

Biofilms treated with 0.016 μg/mL of imipenem alone, showed few cells aggregation with a predominance of live cells. The addition of diflunisal showed an increase in dead and live cells, but also greater cell aggregation. At 0.06 μg/mL, imipenem alone had partial bactericidal effect, but when combined with diflunisal, there is a substantial increase in cell death.

Doripenem at 0.016 μg/mL alone and in combination with diflunisal showed limited effectiveness, with mostly live cells. At 0.06 μg/mL, doripenem became more effective in terms of bactericidal activity. However, when combined with diflunisal, bacterial cell viability in the biofilm increased.

With ciprofloxacin at 0.12 μg/mL, there was high cell aggregation and viability of live cells, suggesting poor effectiveness. However, when combined with diflunisal, the amount of dead cells increased. The biofilm treated with 0.5 μg/mL of ciprofloxacin alone showed moderate cell viability, but when combined with diflunisal, there was an increase in dead cells. At 2 μg/mL of ciprofloxacin used as an individual treatment, less cell aggregation and high cell death was observed. However, the combination of ciprofloxacin with diflunisal resulted in an increase in cell viability and density.

Regarding diflunisal individual treatment (at 25 μg/mL), it showed a greater cell aggregation when compared to antibiotics treatments. Nonetheless, there is a high prevalence of dead cells.