Diclofenac May Induce PIA-Independent Biofilm Formation in Staphylococcus aureus Strains

Staphylococcus aureus is a pathogen commonly resistant to antibiotics. Bioﬁlm formation is one of the important factors related to its virulence. Non-antibiotics drugs, such as nonsteroidal anti-inﬂammatory agents (NSAIDs), have been studied as an alternative for treating infections by multiresistant pathogens and bioﬁlm-associated infections. In this study, the eﬀects of NSAID sodium diclofenac on growth inhibition and bioﬁlm formation of S. aureus were evaluated. The minimum inhibitory concentration (MIC) of diclofenac for ﬁfty isolates ranged from 200 to 400 μ g/mL. Diclofenac sub-MICs induced bioﬁlm in 32.3% of bioﬁlm-negative strains in tryptic soy broth. All bioﬁlms induced by the drug showed a PIA- (polysaccharide intercellular adhesion-) independent composition, and the scanning electron microscopy showed that the induced bioﬁlm presented a very discrete matrix. The combination of diclofenac with rifampicin sub-MICs induced strong production of PIA-dependent bioﬁlm in three of four strains, while combination of NSAID with NaCl induced the formation of partially polysaccharide bioﬁlm in two strains and PIA-independent bioﬁlm in another strain. The combination of NSAID with glucose resulted in PIA-independent bioﬁlms in all four strains tested. The results showed that diclofenac can commonly induce bioﬁlm production by a PIA-independent pathway. However, when this NSAID is combined with other types of inducing agents, the composition of the bioﬁlm produced may vary.


Introduction
Several drugs have a greater or lesser degree of antibacterial activity, although this effect is not their primary therapeutic goal. Compounds with these properties are included in medicine classes such antihistamines, statins, antipsychotics, local anesthetics, nonsteroidal anti-inflammatory drugs (NSAIDs) [1], anticonvulsants, and sympatholytic agents [2] and have been called non-antibiotics [3].
Among non-antibiotics, many studies have focused on NSAIDs. e NSAIDs are compounds that, in addition to anti-inflammatory action, have analgesic and antipyretic properties. ese drugs are among the most widely drugs used in the world. eir main mechanism of action involves the inhibition of cyclooxygenase, which leads to a decrease in the synthesis of prostaglandins [4]. e most prescribed NSAIDs in clinical practice are drugs such as diclofenac, ibuprofen, indomethacin, and acetylsalicylic acid. eir antimicrobial action includes bacterial isolates of different species, as well as fungi [5]. Diclofenac (2-[2-[(2,6-dichlorophenyl)amino]phenyl]acetic acid) is a NSAID with an antimicrobial activity in vitro against various bacterial pathogens [6][7][8][9]. Studies have shown that the broad-spectrum antibacterial effect of this drug can also be obtained in vivo [10][11][12]. e antibacterial mechanism of action of diclofenac seems to reside mainly in the inhibition of DNA synthesis [13]. Besides that, the increase in the uptake of ethidium bromide by Staphylococcus aureus cells exposed to diclofenac sodium has provided evidence that NSAIDs may also act by compromising cellular membrane integrity [14]. is drug has also induced extensive alterations in the transcriptome of methicillin-resistant S. aureus (MRSA) by changing the expression of hundreds of genes, including those associated with antimicrobial resistance [15].
In addition to investigating their potential for antibacterial activity, researchers have analyzed the effects of diclofenac on biofilms. Biofilms are complex multicellular communities of microorganisms that proliferate in an amorphous polymeric matrix. ey play an important role in microbial pathogenicity, protecting against host immune defense mechanisms and conferring survival capacity to antimicrobial agents [16].
In S. aureus, an important nosocomial and community pathogen, biofilm formation is mediated by icaADBCdependent and -independent pathways [17]. e genes of ica operon encode enzymes involved in the production of the polysaccharide intercellular adhesin (PIA), which mediates the intercellular adherence of bacteria and the accumulation of multilayered biofilm. However, specific proteins can substitute PIA in cell-cell adhesion in the PIA-independent biofilm, and sometimes, extracellular DNA (eDNA) can be the main component of the matrix [18,19].
Many studies have indicated that diclofenac has antibiofilm properties on different bacterial pathogens [9,14,20]. However, it has not been investigated whether, by contrast, subinhibitory concentrations of this NSAID can induce biofilm production. erefore, the aim of this study is to investigate the antimicrobial effect of diclofenac against S. aureus strains and their action on biofilm induction, alone or combined with other potential inducing agents.

Bacterial Isolates.
Twenty-five clinical isolates of S. aureus (strains A) were obtained from different infected patients admitted to a hospital in Rio de Janeiro (HSE/RJ), and twenty-five nonclinical strains were obtained from the nasal swabs of healthy individuals (strains B). Only one bacterial isolate was considered for each individual. e isolates were submitted to Gram staining, catalase, and tube coagulase tests. e identification and susceptibility to antimicrobial agents were performed using the MicroScan WalkAway-96 System (Dade Behring Inc.). In addition, the isolates were tested for resistance to oxacillin, confirmed by the cefoxitin disk screen test. Species confirmation was made by PCR assay [21]. Bacterial stocks were kept at −80°C in tryptone soy broth (TSB, HiMedia) containing glycerol (50% v/v). e Ethics Committee on Research with Humans of the HSE/RJ approved this study under reference number 000.41.

Determination of Minimum Inhibitory Concentration (MIC) and Minimum Bactericidal Concentration (MBC) of Diclofenac.
e MICs of the diclofenac were determined by the broth dilution method according to the Clinical and Laboratory Standards Institute [22] using 96-well flatbottomed polystyrene microtiter plates (Kasvi). Briefly, a diclofenac (Sigma) working solution was prepared in methanol, sterilized by filtration, and diluted in TSB. Drug concentrations ranging from 1.600 to 1.5 μg/mL were obtained by two-fold serial dilutions in TSB, and one hundred microliters of each drug dilution were added to the wells. e bacterial suspension was prepared in a saline solution, and turbidity was adjusted to 0.5 McFarland standard, equivalent to 1.5 × 10 8 colony forming units (cfu)/mL. e bacterial inoculum consisted of five microliters of the diluted culture to obtain about 5 × 10 5 cfu/well. TSB without the drug and TSB only with the bacterial inoculum were used as controls. e plates were incubated at 35°C for 24 h and visually read to establish the presence or absence of turbidity.
e MIC was defined as the minimum concentration of the drug that inhibited visible bacterial growth. Additionally, optical density (OD 620nm ) was measured using a plate reader ( ermo Plate RT6000). e MBC was determined by the subculture, in Triptyc soy agar (TSA, HiMedia), of 10 μl of medium taken from wells that did not show visible growth. e plates were incubated at 37°C for 24 h, and MBC was defined as the lowest drug concentration that resulted in the absence of bacterial growth or in the presence of only one colony.

Effect of Subinhibitory Concentrations (Sub-MICs) of Sodium Diclofenac on Biofilm Formation.
e effect of sub-MICs of diclofenac on biofilm formation was studied using the microtiter plate (MTP) assay [23], with some modifications. Twenty clinical strains and twenty carrier strains were tested. Briefly, two-fold serial dilutions of diclofenac solution were prepared in TSB and added (1 : 100) to overnight cultures of isolates and reference strains (final concentrations equivalent to 1/2 MIC to 1/16 MIC). Strains with biofilm production induced by diclofenac concentration equivalent to 1/16 MIC were tested for lower drug concentrations. Controls were prepared in TSB without the drug. S. aureus ATCC 43300 and S. epidermidis ATCC 12228, positive and negative for biofilm production, respectively, were used as reference strains.
After homogenization, the bacterial suspensions were seeded in 96-well flat-bottomed polystyrene microtiter plates (200 μL/well) and incubated for 24 h at 35°C. en, the cultures were removed, the wells were washed three times with distilled water, and the attached bacteria were fixed with 200 μL of methanol for 15 minutes. e plates were emptied, air-dried, and stained for 15 min with 200 μL of 2% Hucker's crystal violet solution. After removal of the dye, the plates were washed under running distilled water and air dried. e dye bound to the adherent cells was extracted with 200 μL of 95% ethanol for 30 min, and the optical density (OD 570nm ) of biofilm extracts was measured.
All experiments were performed in triplicate and repeated at least three times. In each plate, four wells were used as blanks containing only TSB medium. e cutoff OD value (ODc) used to differentiate biofilm producer and nonproducer isolates was defined as three standard deviations (SD) above the mean OD of the negative control: ODc � OD average of negative control + (3 × SD of negative control) [24]. Isolates with OD ≤ ODc were considered as nonproducers. e result of the OD average of biofilm extract was used to classify the isolates as weak producer (ODc < OD ≤ 2xODc), moderate producer (2xODc < OD ≤ 4xODc), and strong producer of the biofilm (4xODc < OD) [25].

Biofilm Detachment Assay.
e PIA-independent or PIA-dependent nature of the biofilm matrix induced by diclofenac was evaluated using a degradation test with sodium metaperiodate [26]. e methodology was based on the MTP assay, as described by Izano et al. [27], with some modifications. e isolates were grown overnight in TSB at 35°C and diluted 1 : 100 in TSB without and with the drug (1/ 2 MIC to 1/8 MIC). en, the microtiter plates were seeded with 200 μL per well of bacterial suspensions and incubated at 35°C for 24 hours.
Subsequently, the cultures were removed, the wells were washed with distilled water, and 200 μL of 50 mM sodium metaperiodate solution (Sigma) in distilled water was added to triplicate wells. An equal volume of distilled water without the degrading agent was added as a control. e plates were incubated for two hours at 35°C, and then, the wells were washed twice with distilled water. e next steps followed those described in the MTP assay for biofilm. e PIA-independent or PIA-dependent composition of the matrix was determined based on biofilm degradation levels by metaperiodate and comparing the OD values in the presence of the degrading agent to the untreated control. In this study, biofilm with a PIA-dependent composition was considered that which suffered dissolution by the sodium metaperiodate greater than 50%. A similar procedure to that described for metaperiodate was used for proteinase K (Sigma) (1 mg/mL in 0.1 M PBS-pH 7.0); trypsin (Inlab) (1 mg/mL em Tris 20 mM, pH 7.5) [28], and deoxyribonuclease I type II (Sigma) (100 μg/ml in 150 mM NaCl and 1 mM CaCl 2 ) [27].

Composition of the Biofilm Matrix in Strains of S. aureus Grown in Medium with Diclofenac Combined with Other
Potential Biofilm Inducers. Four positive strains in which biofilm formation was induced by diclofenac were tested for other substances with the potential to induce biofilm production. Previously, the presence of the ica operon genes A-D was detected by multiplex polymerase chain reaction (PCR), according to the procedure reported by de Lima e Silva et al. [29]. e substances chosen for the tests were rifampicin, at sub-MICs concentration, glucose (1%) and NaCl (3%). e sub-MICs were determined after previous tests at concentrations equivalent to 1/2 and 1/4 of the MIC (0.004 μg/mL and 0.002 μg/mL, respectively). Rifampicin sub-MICs were determined after previous tests at concentrations equivalent to 1/2 and 1/4 of the MIC (0.004 μg/mL and 0.002 μg/mL, respectively). Antimicrobial susceptibility tests previously performed showed that these strains were rifampicin sensitive. e strains grew in TSB containing each of these agents, alone and in combination with diclofenac (50 or 100 μg/mL). In addition, the strains grew in TSB without the agents studied and TSB only with diclofenac. e tests for researching the production and composition of the biofilm were carried out by MTP assay, as described above.

Quantitative Real-Time PCR (qRT-PCR).
One clinical strain of S. aureus (SA139) was selected to evaluate the relative expression levels of the sarA, sigB, icaA, and icaR genes. Previously, the presence of the icaADBC operon was investigated and detected in this strain, according to the technique described by Ninin et al. [30]. e gene sequences were obtained from the GenBank website (http://www.ncbi. nlm.nih.gov/genbank/). e primers of the genes used in this study were designed by the Primer Express software (Applied Biosystems © ).
e culture was grown in TSB medium until the exponential phase before the addition of the diclofenac solution (50 and 80 μg/ml, final concentration) or an equal volume of sterile methanol (control) and incubated by 20 min. Total RNA extraction was performed based on the phenol-chloroform method at 70°C [31]. en, the extracted RNA was resuspended in 15 μl DEPC water, and the tube was vortexed for three minutes. RNA purity was evaluated spectrophotometrically on the NanoDrop2000 system ( ermo Fisher Scientific), and the RNA integrity was verified by loading 5 μl of the total RNA into 1.4% agarose gel containing 1.0 × TBE buffer. In addition, RNA concentration was measured using the Qubit® RNA BR Assay Kit ( ermo Fisher Scientific) with the reading of reaction using the Qubit® 2.0 Fluorometer (Life Technologies, ermo Fisher Scientific).
Reverse transcription of the extracted RNA was performed using a LifePro ™ gradient thermal cycler (Bioer Technology Co.) and the kit TaqMan® Reverse Transcription Reagents (Applied Biosystems), according to the manufacturer's protocols. e qRT-PCR was prepared in 96-well RT-PCR plates containing the GoTaq® qPCR Master Mix (Promega), according to the manufacturer's instruction. cDNA was diluted (1 : 4), and then, 10 μL of the GoTaq® qPCR Master Mix, 4 μL of the oligonucleotide primer mix (0.8 μM), and 2 μL of DNAse-free water were added. e reaction was performed in triplicate, and negative controls were prepared for each primer oligonucleotide using 5 μL of Milli-Q H 2 O with DEPC instead of the first cDNA strand. In all experiments, the gyrB gene was used as an endogenous control of reactions. Analysis of data generated in the qRT-PCR reaction was performed using the Applied Biosystems 7500 Fast Real-Time PCR System (Life Technologies Corporation) and the method of relative quantification of gene expression normalized by endogenous control levels (2 −ΔΔCT ).
International Journal of Microbiology 3

Scanning Electron Microscopy (SEM).
ree diclofenacinduced biofilm-producing strains were selected to be cultured on coverslips in TSB without diclofenac (control) and with diclofenac (50 μg/mL). Briefly, a 1 : 100 dilution of overnight culture was inoculated into 24-well polystyrene microtiter plates (2 mL/well). Subsequently, a sterile ermanox plastic coverslip (13 mm) (Nalgene-USA) was placed on the bottom of each well. After 24 hours of incubation at 35°C, the coverslips were removed from wells with sterile forceps, rinsed several times with PBS, and then fixed with 2% glutaraldehyde in 0.1 M cacodylate buffer, pH 7.2, for one hour at 4°C. After additional washing, the samples were dehydrated with a series of increasing concentrations of ethanol, 30%, 50%, 70%, 80%, and 95% for 30 min each, and pure ethanol for one hour. Finally, the specimens were coated with gold (Baltec SCD 050) and examined using SEM equipment (Zeiss EVO 40) operated at 20 kV.

Antibacterial Activity and Effects of Diclofenac Sub-MICs on Biofilm Formation.
e antibacterial activity of diclofenac was tested against 50 strains of S. aureus. Among the 25 clinical strains, seven were characterized as MRSA, while among carriers strains, only one strain was identified as community-associated MRSA (CA-MRSA). e diclofenac MIC ranged from 200 to 400 μg/mL. For the concentration of 800 μg/mL, the action of the drug was bacteriostatic, while the MBC was 1600 μg/mL. Forty isolates were tested for biofilm production. Among 34 negative strains for biofilm production in TSB medium without the drug, supplementation with diclofenac promoted an induction for biofilm production in 11 of them (six clinical strains and five carrier isolates). All these 11 isolates were induced in the sub-MIC of 50 μg/mL, and 9 also were induced in 100 μg/mL. ese concentrations were equivalent to 1/2 MIC or 1/4 MIC or 1/8 MIC depending on the strain. Figure 1 shows the levels of production of biofilm induced by different concentrations of diclofenac in three clinical and three carrier strains. For the A150 strain, diclofenac induced the production of biofilm at concentrations as low as 3.1 and 1.5 μg/mL (Figure 2(a)). ese concentrations were equivalent to 1/64 MIC and 1/128 MIC, respectively. Six strains were characterized as biofilm producers in TSB without the drug, and one strain (A116), with the addition of diclofenac to TSB, enhanced the biofilm formation even at a concentration of only 3.1 μg/mL of the drug. At this concentration, the increase was about 32% (Figure 2(b)).

Diclofenac-Induced Biofilm Type.
e type of biofilm induced by diclofenac was characterized based on the degree of degradation after treatment with metaperiodate, trypsin, proteinase K, and DNAse. Metaperiodate is an oxidizing agent that cleaves bonds of glucosamine polymers such as PIA, and it is then effective in dispersing polysaccharide biofilms, unlike nonpolysaccharide biofilms, which are resistant to such treatment [32]. Trypsin and proteinase K are enzymes that degrade biofilms with a matrix of protein composition, while DNAse disperses biofilms consisting mainly of extracellular DNA [27,28].
In both groups of S. aureus strains isolated from carriers and those of clinical origin, all biofilms induced by sub-MICs of NSAID were nonpolysaccharides (PIA-independent) in composition: protein (6), protein-DNA (3), and DNA (1). In one isolate, the induced biofilm was not dispersed by any of the tested agents. In the A116 strain, which produced biofilm in TSB without diclofenac and increased production when grown in the presence of the drug, the metaperiodate dispersed the biofilm produced. us, the biofilm of this strain has a polysaccharide composition (PIA-dependent) in both growing conditions.

Composition of Biofilm Induced by Diclofenac When
Combined with Other Inducing Agents. Biofilms produced by four strains (A112, A139, A150, and B103) were evaluated after the growth in diclofenac-supplemented media combined with rifampicin sub-MICs, NaCl (3%), and glucose (1%). ese strains showed sensitivity to rifampicin in the antimicrobial susceptibility test and harbored the icaA and icaD genes.
Rifampicin sub-MICs did not induce biofilm production in the isolate A139, while it observed high induction of PIAdependent biofilm (polysaccharide) in the other three isolates. e mean (OD 570 nm ) of biofilm extract for 1/2 MIC of rifampicin ranged from 1.263 to 1.659 and from 0.368 to 0.515 for diclofenac. e polysaccharide composition of the biofilm induced by rifampicin was maintained in the culture with rifampicin + diclofenac, while the biofilm of the control isolate grown only with diclofenac kept its original PIA-independent composition (Table 1). For rifampicin in 1/4 MIC + diclofenac, the results were similar as those of rifampicin in 1/2 MIC + diclofenac, except for the B103 strain, for which the combination of drugs resulted in the production of a partially polysaccharide biofilm (degradation by metaperiodate only about 45%). e biofilm induced by both rifampicin and the combination rifampicin + diclofenac was classified as strong, in contrast to the weak biofilm induced only by diclofenac.
NaCl did not induce biofilm in only one of the four isolates (SA112). For this strain, the biofilm originally induced by diclofenac was inhibited by NaCl + diclofenac. In two strains (A150 and B103), NaCl induced PIA-dependent biofilm formation. However, in the growth with NaCl + diclofenac, the induced biofilm revealed a partially polysaccharide composition, with a biofilm dispersion of about 30% and 36%, respectively, after treatment with metaperiodate ( Table 2). Contrary to what we observed for B103 and A150 isolates, NaCl alone and combined with diclofenac induced only PIA-independent biofilm production in the A139 strain.
As for the effects of glucose, there was induction to produce PIA-independent biofilm in all four isolates studied.
is result was maintained when the strains were grown in the medium containing glucose + diclofenac (Table 3). 4 International Journal of Microbiology

Quantitative Real-Time PCR (qRT-PCR).
e relative expression levels of the sarA, sigB, icaA, and icaR genes of the strain A139 were evaluated after treatment with diclofenac ( Figure 3). e expression of icaA was suppressed in 86% and 54% at concentrations of 50 and 80 μg/mL, respectively, compared to the expressions in the control. In contrast, icaR activity increased about five-fold and three-fold, respectively. Diclofenac also induced an increase in the activity of the two global regulators studied (sarA and sigB). Concentrations of 50 and 80 μg/mL increased the activity of sarA by three times and two times, respectively, in relation to the control, while the increases were 5.6 and about 3 times, respectively, for sigB.       International Journal of Microbiology

Scanning Electron Microscopy (SEM).
e B103, A139, and A150 strains were chosen for morphological evaluation (SEM) of diclofenac-induced biofilm (Figure 4). In strains grown on the surface of ermanox plastic membranes in TSB medium without diclofenac (control), the images revealed very few adhered cells. Conversely, after cultivation in medium with 50 μg/mL of NSAID, there were large amounts of cell aggregates. Magnified images of the cultures showed cell aggregates with a small amount of extracellular matrix material, mainly in the B03 strain. No significant changes were observed in the morphology of bacterial cells grown in the medium with NSAID.

Discussion
Bacterial multidrug resistance is a growing phenomenon worldwide. New resistance mechanisms have been described for different bacterial pathogens. e difficulty in effectively treating common infectious diseases with conventional antibiotics available has stimulated the study of the antimicrobial action of drugs known as "non-antibiotics" [1,3,5,33].
is group of drugs includes diclofenac, an NSAID that has been shown to have a broad-spectrum antimicrobial activity in vitro and in vivo [6-9, 13, 34]. On the other hand, it has not been investigated whether sub-MICs of this drug can induce biofilm formation, as in the case for certain antibiotics [29,35].
In the present study, the MIC of the diclofenac ranged from 200 to 400 μg/mL, with slight differences regarding the origin of isolates (clinical or carrier). ese concentrations are well above the therapeutic plasma levels normally attained [36]. However, it should be considered that there are topical formulations of this NSAID that can locally determine the presence of concentrations much higher than those achieved in plasma with the systemic use of the drug [37]. erefore, the possibility of effects under these conditions on the microorganism in the interaction with the drug cannot be excluded. e literature reports the MIC of diclofenac for S. aureus with values ranging from 50 to >1000 μg/mL, which may be due to intrinsic physiological characteristics of isolates studied, as well as to the methodological procedures, such as the way for drug solubilization, culture medium, and the technique [8]. e MBC was detected only at 1600 μg/mL, since plating in TSA of aliquots of wells containing ≤800 μg/ mL resulted in spots with bacterial growth. is result is in contrast with earlier findings, which pointed to a bactericidal action of diclofenac in concentrations lower than 1600 μg/ mL [10,13,38,39].
Positive and negative isolates for biofilm production in TSB medium were tested to evaluate the effects of diclofenac sub-MICs on biofilm production. Diclofenac induced biofilm production in 11 of 34 biofilm-negative isolates grown in TSB, and all induced biofilms showed a nonpolysaccharide composition. ese results add new data on drugs with the potential to induce production of nonpolysaccharide biofilm in S. aureus, since this phenomenon has not yet been reported for diclofenac. Although diclofenac at a concentration equal to that achieved for human serum has limited the formation of strong biofilms by clinical isolates of S. aureus grown on the surface of polypropylene mesh, Reśliński et al. [40] showed that the strains producing weak biofilm increased from 1.4% (control) to 10.0% (diclofenac). However, the chemical nature of this biofilm was not characterized by the authors. e concentrations that largely determined this effect were equivalent to 1/2 MIC, 1/4 MIC, and 1/8 MIC, according to the strain. In a nonbiofilm producing strain (SA150), induction of biofilm production by diclofenac was noted at a concentration as low as 1.5 μg/mL. Besides that, in a producing strain (SA116), an increase in biofilm production at concentrations from 3.1 μg/mL of the drug was observed. It is important to note that these concentrations are within the plasma levels that the drug can reach [36]. e nonpolysaccharide composition of the biofilm matrix induced by diclofenac, indirectly determined in this study by the degree of degradation after treatment with sodium metaperiodate solution, suggests that the drug activates PIAindependent pathways to produce biofilm. In fact, complementary tests using trypsin and proteinase K, in addition to DNAse, showed that the induced biofilm had a protein composition in most isolates, while in others, the matrix composition was eDNA-protein, eDNA, or not affected by the dispersing agents used. On the other hand, in a strain (SA116) that naturally produces biofilm in TSB without the drug, diclofenac determined an increase in the production of polysaccharide biofilm. It suggests that depending on the strain, the NSAID could stimulate ica-(PIA-) dependent pathway in order to increase biofilm production.
Biofilms with a PIA-independent extracellular matrix have been associated with MRSA strains [41,42]. Among the clinical isolates tested, seven were MRSA, but diclofenac did not induce biofilm production in any of these strains. ere are no other studies on the chemical nature of diclofenacinduced biofilms; however, it is important to note that Dotto et al. [43] reported that NSAID salicylic acid, the main biometabolite of acetylsalicylic acid, promoted the formation of biofilm in S. aureus in a PIA-dependent way.
It is not yet fully known how the biofilm matrix can vary in response to different inducing agents. A better understanding of the chemical composition of biofilms induced by different conditions could be useful in strategies for treating infections associated with biofilm, since, for example, cells in biofilms with the protein matrix have a different organization of growing compared with those in PIA-dependent biofilm. Four isolates that were positive for biofilm induction by diclofenac were evaluated for the effects of combining NSAID with other potential biofilm-inducing substances (rifampicin, glucose, and NaCl). e aim of these tests was to assess whether the combination of diclofenac with these substances could induce changes in the composition of the biofilm matrix and which pathway predominates, whether PIA-dependent or PIA-independent. e results showed that rifampicin stimulated high polysaccharide biofilm production, both alone and in combination with diclofenac, suggesting that the activation of the ica-dependent pathway promoted by this antibiotic was strong enough to overlap the ica-independent stimulus implemented by diclofenac. It is important to note that this effect can be dose-dependent, since in one of the strains studied, the combination of a lower concentration of rifampicin (1/4 MIC) with diclofenac resulted in the formation of a biofilm that was only partially polysaccharide.
Unlike rifampicin, both glucose alone and the combination of glucose with diclofenac induced the formation of a nonpolysaccharide biofilm.
is result is not surprising, since this sugar is considered a common activator of biofilm production through a PIA-independent pathway [41,44,45]. However, it is interesting to note that the combination of diclofenac + glucose did not result in an additive stimulatory effect of increasing biofilm production, compared to the effect of these substances in isolation.
As for NaCl, this salt has been considered an activator of the ica operon in non-MRSA strains [41], which is in agreement with the results obtained for two of the 4 isolates tested. However, the biofilm produced by these two isolates in the medium supplemented with NaCl + diclofenac was poorly dispersed by metaperiodate (only 29% and 36%), suggesting that, in this case, a greater stimulation of NSAID in the activation of an ica-independent pathway has predominated, while the ica-dependent pathway stimulated by NaCl was not completely inhibited, resulting in low PIA production. On the other hand, the third isolate (A139) produced a PIA-independent biofilm both in the presence of NaCl and in the medium containing this salt and diclofenac, although this strain previously showed the presence of ica operon genes (icaA and icaD). is result is suggestive of the complexity of the regulation of biofilm production in S. aureus, since it can also involve characteristics specific to each strain. e results obtained with qrt-PCR to some extent corroborate the phenotypic results found for the production of diclofenac-induced PIA-independent biofilm in strain SA139. e relative expression of icaA was downregulated by diclofenac treatment, while the activity of icaR increased considerably. icaR is a negative regulator of icaADBC because it binds to the icaA promoter region [46]. It was also observed that diclofenac increased the activity of the global regulators sigB and sarA.
sigB, an alternative sigma factor of RNA polymerase that is activated in stress response, has been shown to regulate numerous genes involved in virulence, including genes involved in biofilm formation [47]. Although the effects of this regulator on ica transcription may vary depending on the study, Lauderdale et al. [48] demonstrated that it was essential for ica-independent biofilm formation in S. aureus.
SarA, on the other hand, is a global regulatory protein affecting the expression of many genes in S. aureus, including a diversity of genes involved in pathogenesis. is regulator seems to affect both ica-dependent and ica-independent biofilm formations [19,49]. A mechanism of action of SarA in the production of ica-independent biofilm seems to be associated mainly with its capacity to repress bacterial proteases, allowing the formation of a proteindependent biofilm [50].
Magnified SEM images confirmed the induction of PIAindependent biofilm production in the SA139 strain by diclofenac. e presence of little extracellular material was noted. According to Vergara-Irigaray et al. [51], the morphology of cells visualized by SEM in a PIA-rich biofilm matrix is different from that of proteinaceous biofilm. e authors showed that in the case of PIA-dependent biofilm, cells are embedded in an abundant extracellular matrix mesh that interconnected the bacteria. However, cells growing in a proteinaceous biofilm formed by FnBP showed dense bacterial aggregates with close contact between cells, without an appreciable extracellular amorphous matrix.

Conclusion
Diclofenac showed relatively high MIC values (200-400 μg/ mL) for the strains tested, and the bactericidal activity occurred only at concentrations greater than 800 μg/mL. In addition, this study is the first to show that diclofenac sub-MICs can induce the production of nonpolysaccharide biofilm in strains of S. aureus negative for biofilm production in TSB, suggesting that this drug activates a PIAindependent pathway in S. aureus. However, in strains naturally producing PIA-dependent biofilm in TSB without additives other than diclofenac, the drug can stimulate an increased production of this type of biofilm.
Despite the limited number of strains studied here, the results suggest that if diclofenac is combined with other potential biofilm inducers, such as rifampicin sub-MICs or NaCl, the composition of the biofilm matrix may show variations probably depending on the type of inducing stimulus or characteristics of the bacterial strain. Such stimulus may result in the formation of a PIA-dependent or only partially polysaccharide biofilm. In contrast, when diclofenac is combined with glucose, the resulting biofilm maintains the PIA-independent composition observed for diclofenac, perhaps because both substances activate a single pathway to biofilm formation.

Data Availability
e data used to support the findings of this study are available from the corresponding author upon request.

Conflicts of Interest
e authors declare that they have no conflicts of interest.
International Journal of Microbiology 9