Carvacrol exhibits rapid bactericidal activity against Streptococcus pyogenes through cell membrane damage

Streptococcus pyogenes is an important human pathogen worldwide. The identification of natural antibacterial phytochemicals has renewed interest due to the current scarcity of antibiotic development. Carvacrol is a monoterpenoid found in herbs. We evaluated carvacrol alone and combined with selected antibiotics against four strains of S. pyogenes in vitro. The minimum inhibitory concentration (MIC) and minimum bactericidal concentration (MBC) of carvacrol against S. pyogenes were 125 µg/mL (0.53 mM) and 250 µg/mL (1.05 mM), respectively. Kill curve results showed that carvacrol exhibits instantaneous bactericidal activity against S. pyogenes. We also demonstrated the potential mechanism of action of carvacrol through compromising the cell membrane integrity. Carvacrol induced membrane integrity changes leading to leakage of cytoplasmic content such as lactate dehydrogenase enzymes and nucleic acids. We further confirmed dose-dependent rupturing of cells and cell deaths using transmission electron microscopy. The chequerboard assay results showed that carvacrol possesses an additive-synergistic effect with clindamycin or penicillin. Carvacrol alone, combined with clindamycin or penicillin, can be used as a safe and efficacious natural health product for managing streptococcal pharyngitis.

The activity of carvacrol with conventional antibiotics and among antibiotic combinations. The antibacterial activity of carvacrol in combinations with antibiotics was analyzed by the checkerboard assay (Fig. S1). The fractional inhibitory concentration index (FICI) values for different combinations of carvacrol, penicillin G salt, penicillin Vk, clindamycin, and erythromycin are summarized in Table 2. Carvacrol displayed marginal additive synergism with clindamycin, with a FICI value of 1.0 against S. pyogenes, whereas the combinations with other antibiotics showed no significant synergistic effects.
Carvacrol instantaneously kills S. pyogenes. The bactericidal effect of carvacrol on the planktonic growth of S. pyogenes (ATCC 19615, clinical isolate, and Spy 1558) was investigated, and the minimum bactericidal concentration (MBC) values are shown in Table 1 and Fig. 1C. The effect of carvacrol against S. pyogenes was investigated by studying time-kill kinetics (Fig. 2), and carvacrol showed concentration-and time-dependent bacterial killing ability. The complete killing of S. pyogenes was observed immediately after exposure to the carvacrol at the concentration of 250 μg/mL (2 MIC = MBC), and such killing was observed for all three strains. A potent killing was observed with 125 μg/mL (MIC) of carvacrol throughout the incubation; however, complete bacterial killing was not persisted until 24 h. Furthermore, a significant growth inhibition activity was observed at 1/2 × MIC of carvacrol. However, reducing the bacterial count at the end of the 24-h incubation period did not go beyond 3 to 4 log units for three strains. Both methods (viable cell counts method ( Fig. 2A) and spectrophotometric method (Fig. 2B)) gave similar results.

Carvacrol induces morphological changes in S. pyogenes. Transmission Electron Microscopy
(TEM) was employed to observe the morphological and ultrastructural alterations induced in three S. pyogenes strains upon exposure to 1/8 × MIC, ¼ × MIC, and ½ × MIC of carvacrol treatment compared to the control. Both untreated control (Fig. 3Aa,a′,a′′) and vehicle control (Fig. 3Ab,b′,b′′) cells remained intact with a complete cell wall, a visible cell membrane, and a homogeneous cytoplasm. In contrast, significant cell deaths (indicated in white arrows) such as ruptured or completely broken cell wall, detached cytoplasmic membrane from the cell wall, dispersion of the intracellular contents, and noticeable cytoplasmic clear zones were observed in ½ × MIC-carvacrol treated ATCC 19615, clinical isolate and Spy 1558 (Fig. 3Ac,c′,c′′) cells. The percentage of ruptured and dead cells relative to the total cells was calculated using images of TEM (Fig. 3B). Cells with the distorted shape compared to the smooth spherical shape of control were also observed. Cellular debris and broken cell wall parts were also abundant.
Morphologies of dead and disintegrated cells including, membrane fusing, clumping, ruptured, the disintegration of the cell wall and/or membrane, cytoplasmic disruptions of cells following exposure to ¼ × MIC carvacrol were noticeable in S. pyogenes strains (Fig. 3Ad,d′,d′′). The round shape of most of the intact cells turned into a clinical isolate from a pharyngeal patient and an erythromycin-resistant Spy 1558) by carvacrol at the given concentrations was measured using micro-broth dilution assay in BHI broth after 24 h of incubation at 37 °C. Bacterial turbidity was measured at OD = 600 nm, and percentage growth inhibition is expressed relevant to the vehicle (0.25% DMSO) as the mean ± SE of three independent experiments. Statistical analysis was performed using one-way ANOVA, and the differences among means were compared using Tukey's test; *P < 0.05, **P < 0.01, and ***P < 0.001. (C) Bactericidal concentrations of carvacrol for S. pyogenes strains were determined by inoculating on agar plate from each replicates well in broth dilution plate that shows a complete absence of growth and was incubated at 37  www.nature.com/scientificreports/ abnormally elongated shapes with ruptured or broken cell walls. The ruptured and dead cell-morphologies were observed in all three bacterial strains; however, % ruptured and dead cells was significantly higher in Spy 1558 than in other strains (Fig. 3B). Carvacrol at or greater than MIC was highly bacteriocidal; therefore, it did not generate enough bacterial cell pellets for TEM analysis. TEM images also confirmed that cell density was significantly reduced in all treated samples compared to vehicle control in a concentration-dependent manner.
Carvacrol induces cytoplasmic content leakage. The membrane leakage assays were carried out to assess the integrity of the cell membranes due to cytoplasmic membrane damage of S. pyogenes. The release of RNA, double and single-strand DNA was detected only in the supernatants of 125 μg/mL (MIC) of carvacrol treatment. The concentrations of RNA, dsDNA, and ssDNA released by ATCC 19615 were estimated to be 87 ± 0.0 ng/µL, 156 ± 19.3 ng/µL, and 82 ± 3.7 ng/µL, respectively, while that by the clinical isolate were 85 ± 2.4 ng/ µL, 71 ± 8.2 ng/µL, and 69 ± 2.5 ng/µL, respectively. Since some 260-280 nm absorbing proteins in cell supernatants can interrupt the absorbance measurement of nucleic acids, to confirm the carvacrol-induced cytoplasmic nucleic acid release, ethanol precipitation was carried out and then was visualized, followed by agarose gel electrophoresis. When ATCC 19615 and the clinical isolate were grown in the presence of 62.5 μg/mL and 125 μg/mL concentrations (1/2 × MIC and MIC, respectively) of carvacrol, cells released visualizable nucleic acids after 24 h (Fig. 4Aa). When high-density bacterial pellets were treated with carvacrol for 1 h, the concentration-dependent release of nucleic acid was observed (Fig. 4Ab).
Leakage of a common cytosolic enzyme, L-lactate dehydrogenase (LDH), into the cell medium was measured. The LDH release by S. pyogenes treated with 125 μg/mL of carvacrol was estimated to be 47-71% relative to the total cellular LDH, which was significantly higher than the vehicle control (Fig. 4B). Furthermore, the amount of LDH released by carvacrol concentration of 250 μg/mL or more was not significantly different with LDH released by Triton X-100 treated. These results suggest the concentration-dependent disturbance to membrane integrity compromise by carvacrol. In contrast, carvacrol did not cause the release of LDH from cultured human tonsil epithelial cells (Fig. 4C).
In vitro cell cytotoxicity of carvacrol. The results showed that carvacrol was not cytotoxic for normal human tonsil epithelium (HTonEpiCs) cells at concentrations below or equal to 250 μg/mL (Fig. 5A). The viability of HTonEpiCs cells was determined to be 89%. The rest of the carvacrol concentrations tested exhibited safety, as > 95% of the cells were viable following carvacrol treatments (Fig. 5). The cell morphology agreed with the Table 1. Minimum inhibitory concentration (MIC) and minimum bactericidal concentration (MBC) (μg/ mL) of carvacrol against four strains of Streptococcus pyogenes. Data are representative of three independent experiments. a Minimum inhibitory concentration (MIC) was determined as the lowest concentration of tested compound that inhibited bacterial growth. b Minimum bactericidal concentration (MBC) was determined as the lowest concentration of tested compound that killed at least 99.9% of the initial inoculums.  . The bacteria and carvacrol solutions were prepared in BHI broth and were assessed for a period of 24 h at a 37 °C incubation period. *Cell growth/killing at 0.08, 1, 2, 3, 4, 5, 6, and 24 h incubation was measured by performing viable cell counts by dilution of cultures in saline solution (0.85% NaCl) and enumeration on BHI agar plates in duplicate. **The bacterial turbidity at OD = 600 nm was measured using a spectrophotometer in 1 h intervals over 24 h incubation period with different carvacrol concentrations or diluent in triplicate. Both methods were performed in three independent experiments. "0" in the scale represents "below the limit of detection".    www.nature.com/scientificreports/ above observations (Fig. 5B,C). Results demonstrated that carvacrol within the tested range of concentration is not toxic to mammalian cells (Fig. 4C) but rather specific toward bacterial cells.

Discussion
For thousands of years, natural compounds have been used to treat infectious diseases worldwide 36,37 . Carvacrol is a monoterpenoid phenolic derivative primarily found in essential oils of herbal plants, including thyme and oregano. Extensive studies have been conducted investigating the biological properties of carvacrol for its potential use in clinical applications 23,25,26 . In this study, we investigated the antimicrobial activity of carvacrol against S. pyogenes and attempted to understand the potential mechanisms of action. The MIC and MBC values (125 and 250 μg/mL, respectively) for four strains coincide with the results of a previous study 38 that reported MICs of carvacrol ranged from 64 to 256 μg/mL against clinical isolates of S. pyogenes isolated from children with pharyngotonsillitis in Italy. Our previous study with oregano essential oils that consist of more than 90% of carvacrol exhibited a similar bactericidal activity 22 . Similar to many other essential oil compounds, carvacrol's antimicrobial activity is related to its phenol structure and the presence of delocalized electrons 39 .
The synergistic antibacterial activity of various plant extracts and phytochemicals in the presence of conventional antibiotics has been reported. The mechanism of synergistic actions of plant extracts and phytochemicals is postulated as a modification of active sites on bacterial cells, inhibition of enzymes that catalyze modification or degradation of antibiotics, inhibition of efflux pumps, or increase of membrane permeability [40][41][42] . The ability of essential oils containing carvacrol, cinnamaldehyde, cinnamic acid, eugenol, and thymol to exert synergistic effects combined with several antibiotics, including macrolides, has also been reported 43 . However, in contrast to the evidence of synergism between several different combinations of erythromycin and carvacrol reported in Magi et al. 38 we have not observed a synergistic effect of tested erythromycin and carvacrol combinations over all the strains. However, among all the combinations, only additive synergy, were seen in some concentrations of carvacrol and penicillin Vk as well as carvacrol and clindamycin. The effective carvacrol-antibiotics combinations of the current study can be further investigated using killing time, cell targets, and mechanisms of action, considering them as a potential therapeutic strategy for common drug-resistant bacteria.
The time-kill assay is a robust and appropriate tool for collecting information about the dynamic interaction between the antibacterial agent and the bacterial strain. Therefore, bacterial killing kinetics of a range of carvacrol concentrations at different times were assessed. Carvacrol-treated S. pyogenes was killed within 5 min exposure at concentrations over 250 μg/mL (1.05 mM), whereas low concentrations (≤ 125 μg/mL) of carvacrol inhibits the growth gradually via a sublethal phase. Magi et al. 38 reported that several dead cells were detected as early as 1 h after incubation with carvacrol at the MIC using live/dead assay. We had previously investigated that time taken to accomplish complete S. pyogenes killing by penicillin G. When the MIC concentration of penicillin G (2 × MIC = MBC = 0.016 μg/mL) was used in the kill curve against S. pyogenes, 24 h were taken for the complete bacterial killing 22 . The observed instantaneous bactericidal action of carvacrol in this study suggests that carvacrol may have affected the bacterial membrane integrity. On the other hand, the observed time-dependent killing and growth inhibition by carvacrol indicate that additional mechanisms can be involved in cellular processes, such as inhibition of protein, nucleic acids, and lipid synthesis.
Ultrastructural morphological abnormalities induced by carvacrol were observed by electron microscopy. Comparisons of TEM images over their respective untreated controls revealed that both cell wall and membrane destruction by carvacrol and leakage of bacterial cytoplasmic contents when bacterial cells were exposed to carvacrol. The changes were evident with an increase in the concentration, which was consistent with the timekill study results and cell leakage assays. These observations are parallel to the scanning electron microscopy observations of a previous study with carvacrol and thymol, an isomer of carvacrol, which was also able to disturb the S. mutans membrane and caused the release of cellular contents 44 . Moreover, a similar membrane destructive activity of carvacrol against another Gram-positive bacteria, Staphylococcus aureus ATCC 43300, was reported previously 45 . The scanning electron micrograph-observations showed that carvacrol treated cells became rough and wrinkled with depression appearing on their surfaces at the concentration of 1.03, 2.06, and 4.12 mM for 4-h treatment 45 . However, our TEM observation shows more detailed sectional views of the interior of an intact and damaged cell structure instead of the cell surface changes.
Leakage of nucleic acids and LDH enzymes are indicators that confirm the effect of carvacrol on cytoplasmic leakage of S. pyogenes cells. Carvacrol caused complete or severe damage to the membrane at higher concentrations and increased the leaching out of cytosolic proteins, enzymes, nutrients, and genetic materials. Therefore, correlations between leakage-inducing concentrations of carvacrol and MIC value suggest that membrane damage is an important mechanism of action. Similarly, monoterpenes cooperate with the lipid bilayer of cytoplasmic membranes affecting membrane leakage of cellular contents such as ATP, ions, and nucleic acid [45][46][47] . Wang et al. 45 discussed the carvacrol-induced intracellular components leakage using β-galactosidase as an indicator. These observations showed that carvacrol treatment affected the integrity of S. aureus cell membranes, which likely resulted in a decrease in cell viability. Their findings were in line with our present results of concentrationdependent lactate dehydrogenase leakage in carvacrol treated S. pyogenes cells. Therefore, we could suggest that carvacrol's cell membrane damage mechanism would not differ between these two species. Besides increasing the permeability of the bacterial cell membrane, it was suggested that carvacrol directly binds to genomic DNA as the second key mechanism of action 45 . Future studies need to be aimed at understanding DNA damage in S. pyogenes cells by carvacrol. The ability of carvacrol to cause leakage of cellular contents suggests that its action may cause pores in the bacterial membranes. It is widely believed that the antimicrobial action of many small molecules results in the formation of pores in the bacterial membranes and cause leakage of cellular contents 48 .
Carvacrol has also been reported to be safe and exert minimal toxicity on human cells 49  www.nature.com/scientificreports/ bacterial selectivity might be attributed to different physicochemical properties of membrane components, especially phospholipids found in bacteria versus mammalian cells. Cholesterol is the major lipid component of the eukaryotic membrane but not of the bacterial cell membrane and may lead to discrimination between bacterial and host cell membranes 50 . As in all Gram-positive bacteria, the cell wall of S. pyogenes is composed of thick peptidoglycan (PGN) covered with proteins, teichoic acid, and lipoteichoic acid (LTA) 51,52 . Hence, carvacrol must cross the bacterial cell wall before interacting with the cytoplasmic membrane; however, the role of the cell wall in interacting with carvacrol to be revealed. We can postulate that cell wall PGN and teichoic acid may allow penetration of carvacrol to the cytoplasmic membrane. Therefore, future studies can be targeted to understand the effect of carvacrol on inhibition of PGN biosynthesis and induction of PGN degradation. The LTA present in most Grampositive bacterial species, including Streptococcus 53 . Fatty acid chains in the LTA are anchored in the membrane, whereas the remaining part of LTA (glycerol phosphate or ribitol-phosphate chain) hangs out through the cell wall. Lipophilic ends of these LTA are found on the surface of the cell wall might facilitate the easy penetration of small monoterpene hydrophobic compounds such as carvacrol 54 to the cell membrane, and accumulation of carvacrol in the membrane is expected. Therefore, these accumulated higher carvacrol concentrations explain the membrane disruption and instant bactericidal effect. Bacteria would certainly need either modification of the charge of cytoplasmic membrane lipids or its composition to become resistant to those membrane-active antibacterial agents 55 . However, our results indicated that bacteria are susceptible to carvacrol. Carvacrol exerts its bactericidal and inhibitory activities against S. pyogenes cells by disturbing cytoplasmic membrane integrity. Due to the potential bacterial membrane-targeted biocidal action, carvacrol can be assessed as a novel approach to treat drug-resistant pathogens such as erythromycin-resistant S. pyogenes.
Our results show that the efficacy of carvacrol on disrupting the cell membrane integrity of S. pyogenes. However, to validate potential carvacrol applications as a new antibacterial agent, other potential mechanisms such as inhibition of macromolecular synthesis of S. pyogenes, inhibition of biofilm formation, and inhibition of quorum sensing, and preventing adhesion needs to be further investigated. The effects of carvacrol on bacterial membrane proteins, nucleic acid, and enzymes that are involved in membrane lipid biosynthesis and degradation need to be explored. Based on the observations, we suggest carvacrol as a potential antibacterial agent to treat or manage infections caused by S. pyogenes. Since carvacrol is listed as a Generally Recognized as Safe (GRAS) food additive by the United States Food and Drug Administration 56 , a significant potential exists for carvacrol as a safe molecule for broader therapeutic applications.

Conclusion
This study investigated the antibacterial effect of carvacrol against S. pyogenes, particularly potential growth inhibition and rapid bactericidal mechanisms. Carvacrol at 250 μg/mL (1.05 mM) exhibited instantaneous bactericidal activity against three tested strains of S. pyogenes. Our study revealed that carvacrol kills S. pyogenes primarily by compromising the cell membrane integrity, leading to cytoplasmic content leakage and, ultimately, bacterial cell death. These findings suggest that carvacrol has the potential to develop as a novel natural health product in the forms of throat vapor, lozenge, or mouthwash to manage the discomfort associated with streptococcal pharyngitis. Furthermore, carvacrol can be further explored as a promising antibacterial agent with higher cell selectivity for potential clinical applications against drug-resistant pathogens.
Micro-dilution method. The micro-broth dilution assay was conducted the same as the above method but in 96-well plates as per the guidelines of CLSI. Carvacrol (0.125 to 2000 µg/mL) was incubated along with bacteria in 200 µL of total volume. Absorbance was measured at OD 600 after 24 h incubation at 37 °C.
Bactericidal activity. The   www.nature.com/scientificreports/ Cell cytotoxicity assay. Cell cytotoxicity was determined using tonsil epithelial cells (HTonEpiCs) (Sci-enCell Research Laboratory, San Diego, CA, USA) 59 . Briefly, TonEpiC cells (6000 cells/100 μL) were seeded in poly-L-lysine-coated 96-well plates, and media was discarded after 24 h-incubation (5% CO 2 at 37 °C) without disturbing to the adhered cell layer. A 100 μL of fresh media containing carvacrol was added. A mixture of LTA and PGN (5 mg/mL from each) was used as bacterial antigen controls. After 24 h-incubation, 10 μL of MTS: PMS (20:1) was added. The absorbance was measured at 490 nm after a 2.5 h-incubation, and results were expressed as percent cell viability compared with untreated cells.
Release of cytoplasmic LDH for bacteria and tonsil cells. The LDH activity in the cell supernatant was measured using the CytoTox 96 Non-Radioactive Cytotoxicity kit (Promega Corporation, Fitchburg, WI, USA) according to the manufacturer's protocol. The 24-h seeded cells (6000 cells/ 100 µL density) were washed and replaced with 100 µL of fresh growth medium. Both cells and bacterial suspensions were treated with DMSO or carvacrol (1/2 × MIC, MIC, 2 × MIC) at 37 °C. After 4 h incubation, cells/bacteria were centrifuged, and 100 µL of supernatant was mixed with 100 µL of CytoTox96 reagent. After 30 min at room temperature, acetic acid (1 M) was added to stop the reaction, and absorbance was measured at 490 nm. Released LDH was calculated by comparing cellular/bacterial LDH obtained by carvacrol treated supernatant with lysis buffer (9% Triton X-100).
Statistical analysis. Statistical analyses were performed using MINITAB statistical software (Version 17.0; Inc., Chicago, IL, USA) and GraphPad Software 5.0 (La Jolla, San Diego, CA, USA). All the experiments were performed three times with triplicates. The data were presented as means ± standard errors. The mean separations were analyzed using Student's t-test and one-way analysis of variance using Tukey's test; Differences were considered statistically significant at *P < 0.05, **P < 0.01, and ***P < 0.001.