Citral modulates virulence factors in methicillin-resistant Staphylococcus aureus

Methicillin-resistant Staphylococcus aureus (MRSA) is responsible for high morbidity and mortality rates. Citral has been studied in the pharmaceutical industry and has shown antimicrobial activity. This study aimed to analyze the antimicrobial activity of citral in inhibiting biofilm formation and modulating virulence genes, with the ultimate goal of finding a strategy for treating infections caused by MRSA strains. Citral showed antimicrobial activity against MRSA isolates with minimum inhibitory concentration (MIC) values between 5 mg/mL (0.5%) and 40 mg/mL (4%), and minimum bactericidal concentration (MBC) values between 10 mg/mL (1%) and 40 mg/mL (4%). The sub-inhibitory dose was 2.5 mg/mL (0.25%). Citral, in an antibiogram, modulated synergistically, antagonistically, or indifferent to the different antibiotics tested. Prior to evaluating the antibiofilm effects of citral, we classified the bacteria according to their biofilm production capacity. Citral showed greater efficacy in the initial stage, and there was a significant reduction in biofilm formation compared to the mature biofilm. qPCR was used to assess the modulation of virulence factor genes, and icaA underexpression was observed in isolates 20 and 48. For icaD, seg, and sei, an increase was observed in the expression of ATCC 33,591. No significant differences were found for eta and etb. Citral could be used as a supplement to conventional antibiotics for MRSA infections.


Results
The minimum inhibitory concentration (MIC) and minimum bactericidal concentration (MBC) data are listed in Table 1. The lowest concentration that inhibited bacterial growth was 5 mg/mL (0.5%), whereas the highest concentration was 40 mg/mL (4%). The bactericidal concentrations varied between 10 mg/mL (1%) and 40 mg/ mL (4%). Of the nine MRSA strains tested, five showed the same concentration in both tests. Using the growth curve, it was possible to determine the concentration that would decrease bacterial growth but would not kill the isolates. After defining MIC and MBC, the subinhibitory concentration of citral was determined to verify the modulation of virulence factors without affecting the viability of microorganisms. Therefore, tests were performed with the concentrations 40 mg/mL (4%), 20 mg/mL (2%), 10 mg/mL (1%), 5 mg/mL (0.5%), 2.5 mg/ mL (0.25%), 1.2 mg/mL (0.12%), and 0.6 mg/mL (0.06%), with a concentration of 2.5 mg/mL being determined for achieving the curve ( Table 1).
The analysis of modulating activity was compared with the Clinical & Laboratory Standards Institute (CLSI) manual (2020), with the following resistance halos: ≤ 28 for ampicillin, ≤ 15 for ciprofloxacin, ≤ 14 for clindamycin, ≤ 13 for erythromycin, ≤ 21 for oxacillin, and ≤ 14 for tetracycline. The inhibition zone (mm) and fractional inhibitory concentration were measured after 24 h of incubation. Our results showed that 5 mg/mL (0.5%) citral was able to modulate the antimicrobial activity of different antibiotics through synergistic, antagonistic, or indifferent actions. In the inhibition zone (mm), the results were statistically significant for strain 80 with ciprofloxacin, for strains 52 and 80 with the antibiotic erythromycin, strains 80 and ATCC 33591 with oxacillin, and for strain 52 with the antibiotic tetracycline (P < 0.05). Citral acted antagonistically with the antibiotic ampicillin for strains 52 and 80, with the use of ciprofloxacin for strain 52, oxacillin in isolate 20, and tetracycline in isolate 18. All tested strains remained resistant to the antibiotics tested (Table 2). In the analysis of the fractional inhibitory concentration (μg/mL), the results were statistically significant for strain 80 for oxacillin (Table 2). Comparing the modulating effect of citral on antibiotic activity when measured via the disk diffusion and broth microdilution methods, different results were obtained. We observed that only strain 80 maintained its results www.nature.com/scientificreports/ for the antibiotic oxacillin. Based on our experiment, the checkerboard method is more sensitive for detecting synergy and disk-diffusion may serve as an initial screening method for the detection of potential synergies.      (18,20,27,33,48,52,80,137) and ATCC 33591 strains. The yellow segments represent greater modulation of 0.5% citral on antibiotics, and the blue segments represent less modulation. www.nature.com/scientificreports/ Antibiograms were also analyzed using the heatmapper platform (http:// www. heatm apper. ca/ expre ssion/), and differentiated modulating activity patterns induced by different strains of MRSA were observed. The heatmap separated the isolates into three distinct clusters (cluster 1: ATCC 33591 and strain 80; cluster 2: strains 48, 27, 137, 18, 33, and 20; cluster 3: strain 52) and, together with the qPCR data, showed different expression patterns induced by different strains of MRSA (Fig. 1).
Biofilm formation capacity of the MRSA were represented in a quantitative way in MRSA isolates using citral in the initial phase of biofilm formation from 0 to 24 h, along with the addition of citral in the mature phase of the biofilm, corresponding to 24 h after inoculation. The concentration of 25 mg/mL (2.5%) of citral was more effective in isolates with citral added in the initial phase of biofilm formation than in the mature phase. In the initial phase, there was a significant reduction in biofilm formation, as isolate 20 (93.6% reduction), and thus isolates 27 (62.5% reduction) and 33 (55.71% reduction). In the mature phase, isolate 137 (51.2% reduction) after 24 h of growth was the only isolate showing a positive result. Thus, it was observed that using citral in the initial phase of biofilm formation presented better results (Fig. 2).
To confirm the results of biofilm inhibition, we visualized its structural organization using confocal microscopy. Citral treatment substantially reduced S. aureus biomass, with a decrease in living and dead cells (Fig. 3), consistent with what was observed in the plate biofilm.
The expression of the virulence genes of the enterotoxins seg, sei, and sej of exfoliative toxins eta and etb and of genes related to the formation of the biofilm, icaA and icaD, were evaluated, considering the characterization of the isolates regarding the virulence genes. Evaluation of the expression of the enterotoxin genes seg and sei (Fig. 4A (18,20,27,33,48,52,80,137) and ATCC 33591 control treated with 25 mg/mL (2.5%) citral for 24 h and with the addition of 2.5% citral after 24 h of the formed biofilm represented in a quantitative way. The experiment was carried out in quadruplicate with two independent experiments. Significant difference at P ≤ 0.05 (*).  (Fig. 4C,D), eta and etb, no significant differences were found between the control group (P < 0.05). However, significant differences were found for bacteria and treatment with citral. Regarding the icaA gene (Fig. 4E), the effect of citral was observed with the underexpression of two strains, 20 (79.5% reduction), 48 (83.9% reduction), 52 (100% reduction), 137 (100% reduction) and ATCC (70.3% reduction). Isolates 20 (87.2% reduction) and 48 (76.2% reduction) treated with citral also showed a reduction in their control in the expression of the icaD gene (

Discussion
S. aureus, MRSA, is one of the most important infectious agents because it can cause nosocomial infections, often with prolonged hospitalization, in addition to increased morbidity, mortality, and hospital costs, making bacterial resistance a major public health problem 8,22 . The use of natural products as microorganism control agents has been attracting interest in food and pharmaceutical sectors, as well as other fields, as pathogens associated with infectious diseases are developing resistance to commonly used drugs. Essential oils are secondary metabolites that contain hydrophobic compounds with the ability to easily diffuse through the cell wall of microorganisms, causing damage to the membrane, especially in fluidity and permeability, leading to the loss of intracellular substances and, consequently, their death 23,24 .
In the present study, our findings show that citral is capable of interfering with bacterial development, presenting bacteriostatic and/or bactericidal activity depending on the concentration, but the most strains and antibiotics did not show a change in antibiotic activity when citral was added. Other studies have evaluated the activity of essential oils against MRSA, including lemongrass, which has citral as one of its major components, and obtained an MIC of 0.78% and an MBC of 3.12% 25 . Another study analyzed the action of citral, linalool, decanal, and valencene on S. aureus, diluted with Tween 80, and observed that only citral and linalool were able to inhibit growth, with MIC values of 0.02% and 0.12%, respectively 18 . While other authors observed lower MIC (0.03%) and MBC (0.06%) values with essential oils and their main components, including citral, against S. aureus 26 , compared to those found in this study.
Owing to the emergence of resistance to penicillin, methicillin, and vancomycin, the options and effectiveness of antimicrobials have been reduced against S. aureus 27 . Thus, the use of natural products as microorganism control agents is relevant, since pathogens are becoming increasingly resistant to drugs used in clinical practices 28 . The mechanism of action of these essential oils depends on their chemical composition, and their antimicrobial activity is not the result of just one mechanism, but of various reactions that involve the entire bacterial cell 29 . It was observed that citral showed synergism, antagonism, or did not interfere with the antibiotic action, but the isolates remained resistant. Another study observed that the essential oil from the leaves of chemotype II  30 . In addition, synergistic modulating activity was also observed for this essential oil with the antibiotics amikacin and clarithromycin, while it was indifferent with oxacillin, ciprofloxacin, and vancomycin or antagonist (clindamycin) when treating S. aureus ATCC 6538P 31 . The effect of biofilm modulation is important because S. aureus can develop intractable infections by forming biofilms, mainly by colonizing devices such as heart valve prostheses, cardiac stimulators, contact lenses, and intravascular lines 32 . In general, the reduction in MRSA biofilm formation showed greater efficacy when the strains were treated with citral in the initial period of formation. The reduction of the fatty acid composition of the cell membrane and hydrophobicity are possible features of antibiofilm activity caused by terpenes, one of the components of this oil 33 . These components primarily target the cell wall and cytoplasmic membrane or www.nature.com/scientificreports/ membrane proteins, thus leading to cell death and, consequently, loss of attachment to surfaces 34 . Other authors have observed that sub-inhibitory concentrations of lemongrass essential oil and its major components, such as citral and geraniol, were able to significantly inhibit the biofilm formation of S. aureus strains isolated from subclinical mastitis, in addition to preformed biofilms 35 . In another study, citral inhibited biofilm formation by MRSA without affecting growth at 100 μg/mL, while greatly hampering surface adherence 36 . As in this study, the evaluation of clove and cinnamon essential oil activity under the strain of S. aureus ATCC 6538, classified as strong producers of biofilm, also showed no activity on the mature biofilm 36 .
Furthremore, another study treating mature biofilms of S. aureus ATCC 29740, with citral at concentrations of 0.02, 0.04, and 0.08% did not report significant differences in their responses 18 . The induction of biofilm formation can occur in potentially toxic conditions for bacterial cells, such as high levels of osmolarity, oxidative stress, and in the presence of sub-inhibitory concentrations of possible treatments, among others 37 . Citral possibly has a greater effect on planktonic cells, as they are more susceptible to the action of the compound since they are not embedded in a matrix. The reduced activity of citral on the strains with mature biofilms may be related to a concentration of 25 mg/mL. Furthermore, the mass production of biofilm is directly related to the incubation time 38 .
Studies have shown the use of essential oils against susceptible and resistant methicillin strains of S. aureus and their action against virulence factors 38,39 . The virulence potential of different S. aureus isolates is determined by the presence or absence of virulence genes; each bacterial strain has a unique combination of surface proteins and regulators, and they carry a variety of mobile genetic elements coding for bacterial resistance and virulence genes, indicating frequent horizontal transfer 40 . The icaA and icaD genes, responsible for biofilm formation, were underexpressed during citral exposure in strains 20 and 48. In strain ATCC 33591, expression of icaD (relative to the housekeeping control gene gyrA), increased on exposure to citral. These results show the effect of citral on the two isolates in the deregulation of these genes, possibly contributing to reducing pathogen virulence. The icaADBC operon encodes enzymes involved in the biosynthesis of polysaccharide intercellular adhesin (PIA) or polymeric N-acetylglucosamine (PNAG), which play an important role in biofilm formation. It is known that the decrease in these enzymes leads to negative regulation of icaA, and consequently a reduction in biofilm 41 . Contrary to the results found in this study, other authors did not observe deregulation of the icaA and icaD genes in S. aureus with citral activity at concentrations of 0.02% and 0.04%, despite being effective for other genes 18 . Studies report the activity of essential oils through the quorum-sensing system, being a possible alternative for divergence and bacterial susceptibility 42 . It is known that in the presence of antibiotics and under certain conditions, microorganisms can present an unexpected phenotype, such as an increase in gene expression 43 . In the presence of citral, some isolates showed a significant difference compared to the positive control, increasing gene expression, such as in strains 18 and 48 for the seg gene, and strains 18,20,48,52,80,137 and ATCC 33591 for the sei gene. These data suggest that the sub-inhibitory concentration of essential oil for some strains was not sufficient to act on the dysregulation of enterotoxin genes. Some studies have shown that at concentrations below the MIC, antibiotics can alter the gene expression of microorganisms causing infections, triggering a contrary response 44 . Another study using the essential oils of Cinnamomum zeylanicum and Ocimum basilicum observed a reduction in the expression of sea, sec, and enterotoxin genes 45 . Comparing the results obtained for biofilm  (18,20,27,33,48,52,80,137) and ATCC 33591 control and with 2.5% citral treatment. The green segments in the plot represent greater differential expression and the red segments represent less modulation, comparing different genes and treatment with citral. www.nature.com/scientificreports/ structure and gene expression in MRSA 20, we noticed that after incubation with citral for 24 h, a significant reduction in biofilm formation. This finding is consistent with the results observed in the gene expression of icaA and icaD also reduced when treated with citral (Fig. 4).

Conclusion
Our results showed that citral is also capable of modifying the biofilm produced by MRSA in the initial stage, and also exerts regulatory activity against virulence genes. It is important to emphasize the possibility of using essential oils and their major components as prophylactics or for supporting the treatment of MRSA infections, which are important for public health issues. Finally, there is a need for further studies to evaluate the activity of citral in combination with conventional antibiotics or prophylactic use, since there are few studies on the interaction between citral and MRSA.

Determination of minimum inhibitory concentrations (MIC) and minimum bactericidal concentrations (MBC).
Bacterial suspension was prepared using a 0.9% sodium chloride solution, and the inoculum (1 mL) was adjusted using a spectrophotometer at 660 nm (1-5 × 10 8 CFU/mL). MIC tests were performed by microdilution using 96-well microplates with 190 μL of the inoculum and 10 μL of citral at concentrations of 40 mg/mL (4%), 20 mg/mL (2%), 10 mg/mL (1%), 5 mg/mL (0.5%), 2.5 mg/mL (0.25%), 1.2 mg/ mL (0.12%), and 0.6% (0.06%), which were used as negative controls for brain heart infusion. The plates were incubated at 37 °C for 24 h. The MIC was considered the lowest concentration of citral, which showed no visible bacterial growth. To determine the MBC, suspensions used in the MIC test were inoculated onto BHI agar plates. All experiments were performed in triplicate with three independent repetitions 50 .
Determination of subinhibitory dose. To verify the modulation of virulence factors without influencing the viability of microorganisms, after the definition of MIC and MBC, the effect of a subinhibitory dose on bacterial growth was evaluated. The growth curves of MRSA alone and MRSA treated with citral at concentrations of 40 mg/mL (4%), 20 mg/mL (2%), 10 mg/mL (1%), 5 mg/mL (0.5%), 2.5 mg/mL (0.25%), 1.2 mg/mL (0.12%), and 0.6 mg/mL (0.06%) was performed in a closed system, with an inoculum of 1-5 × 10 8 CFU/mL, at 37 °C. The microbial growth curve was determined every hour for 24 h by spectrophotometry with absorbance at 660 nm (Bel Photonics Spectrophotometer UV-VIS M51), and the viable cell count was determined by serial dilutions and seeding on BHI every hour 50 .

Evaluation of modulating activity: citral interference in antibiotics.
To evaluate citral as a modulator of antibiotic action, the evaluation was performed with six antibiotics, including ampicillin (10 µg), ciprofloxacin (5 μg), clindamycin (10 μg), erythromycin (15 μg), oxacillin (1 μg), and tetracycline (30 μg) associated with citral (5 mg/mL, 0.5%), and the lowest MIC found in the tested strains. Antibiotic susceptibility testing was performed using the disc diffusion method according to the CLSI 51 . The MIC value was determined according to CLSI 51 . The experiments were performed in triplicate with three independent repetitions.
Inhibition of biofilm formation. The ability of MRSA to form biofilms following treatment with 25 mg/ mL (2.5%) citral was analyzed according to previously proposed methods 52 . Biofilm assays were performed in 96-well polystyrene microplates, using trypticase soy broth (TSB/Difco) with 1% (w/v) glucose (TSB-1% Glc). Cultures of MRSA strains (5 mL) were incubated in a shaker at 250 rpm at 37 °C for 18 h, after which they were diluted 1:100 in TSB-1% Glc and 200 μL were inoculated into each well and incubated at 37 °C for 24 h. Supernatants were removed from each well and biofilms were washed twice with PBS, dried, fixed at 65 °C for 1 h, stained with crystal violet 1%, and gently washed twice with PBS. MRSA strains were tested using citral at a concentration 10 times greater than that obtained in the subinhibitory dose 53 . Two phases of biofilm formation were evaluated: early phase 0 to 24 h after inoculation (the period of microcolony formation) and mature phase 24 h after inoculation (the period after thin biofilm formation) 54  www.nature.com/scientificreports/ microscopy (CLSM) was used to obtain structural images. Here, the biofilm assays were performed in the same way, but after being fixed, the bacterial cells were stained with 25 nM SYTO9 and propidium iodide (Live/Dead Bacteria, Invitrogen, Brazil) for 15 min in the dark. The stain was gently removed, and biofilms were observed using a CLSM (Carl Zeiss LSM 510, Germany) equipped with an argon laser (488 nm, and 2 helium/neon 543 nm wavelengths) to visualize the luminescence of fluorochromes 55 .
Virulence factor expression modulation. To determine a possible modulating effect of citral on different virulence genes, a sub-inhibitory concentration of 2.5 mg/mL (0.25%) was used and incubated in a shaker at 37 °C for 4 h (when all strains reached their exponential growth). Total mRNA was extracted using the Pure-Link™ RNA Mini kit (Life Technologies) and quantified by spectrophotometry (NanoDrop 2000/2000c Spectrophotometer, Thermo Fisher Scientific, Waltham, MA, US). cDNA was obtained using the SuperScript III Reverse Transcriptase Kit (Invitrogen). To detect genes related to virulence sec, sei, sej 56 , icaD, icaA 57 , eta and etb 58 , qPCR was used, using primers and conditions previously described. The gyrB gene 59 was used as an endogenous control. This protocol included, in a final reaction volume of 27 μL, which included 12.5 μL of SYBR® Green Master Mix (Applied Biosystems), 1 μL (13 μM) of each primer, 10.5 μL of H 2 O RNAse free and 2 μL of cDNA. Gene expression data were analyzed using the 2 -ΔΔCT method and analyzed in triplicate.
Statistical analysis. GraphPad Prism 6.0 (GraphPad Software, San Diego, CA, USA) was used for analysis.
The following nonparametric tests were used: Kruskal-Wallis with Dunn's test (when evaluating more than two groups) and a Mann-Whitney one-tailed test (when evaluating two groups). Statistical differences were considered significant at P < 0.05, using a 95% confidence interval. Citral interference in antibiotics and modulation of virulence factor expression were analyzed using the heatmapper platform (http:// www. heatm apper. ca/ expre ssion/) and represented as a heatmap. Unsupervised hierarchical grouping was performed using the average distance and Euclidean distance as metrics.

Data availability
The datasets used and/or analyzed during the current study are available from the corresponding author upon reasonable request.