Antibacterial Activities and Synergistic Interaction of Citrus Essential Oils and Limonene with Gentamicin against Clinically Isolated Methicillin-Resistant Staphylococcus aureus

Citrus reticulata Blanco and Citrus aurantifolia are the edible plants which contain several biological properties including antibacterial activity. The aims of the present study were to determine the chemical compositions and evaluate antibacterial activities of citrus essential oils extracted from the fruit peels of C. reticulata (CREO) and C. aurantifolia (CAEO), alone and in combination with gentamicin, against a panel of clinically isolated methicillin-resistant S. aureus (MRSA) (n = 40) and methicillin-susceptible S. aureus (MSSA) (n = 45). Gas chromatography-mass spectrometry analysis revealed that 12 and 25 compounds were identified in CREO and CAEO with the most predominant compound of limonene (62.9–72.5%). The antibacterial activities were determined by agar disk diffusion and resazurin-based microdilution methods. The results found that almost all MRSA isolates were resistant to ciprofloxacin, erythromycin, and clindamycin, and some isolates were resistant to gentamicin. CREO and CAEO exhibited inhibitory effects toward clinical isolates (MIC: 1.0–32.0 and 8.0–32.0 mg/mL, respectively), with a similar trend to limonene (MIC: 1.0–32.0 mg/mL). However, the higher antibacterial effects were found in CREO and limonene when compared to CAEO (p < 0.01). In combination effect, the results showed the synergistic interaction of gentamicin with CREO and limonene on the MRSA and MSSA isolates (FIC indexes: 0.012–0.258 and 0.012–0.375), but that interaction of gentamicin with CAEO was observed only on MRSA (FIC index: 0.012–0.016). These findings demonstrated the potential of these citrus essential oils as natural antibacterial agents that may contribute to reduce the emerging of antimicrobial-resistant bacteria.


Introduction
Staphylococcus aureus, a Gram-positive commensal bacterium, is mainly colonized in the nasal carriage as well as in the skin, axillae, perineum, and pharynx [1]. It can cause a wide variety of infectious diseases, ranging from mild skin and soft tissue infections to life-threatening such as endocarditis, osteomyelitis, pneumonia, and bacteremia. Due to overuse of antibiotics, the emergence of the antibiotic-resistant phenotype has been reported worldwide. Methicillinresistant S. aureus (MRSA), caused by the acquisition of hypotension, and hypersensitivity reactions are commonly presented; thus, drug monitoring is required [4]. In this context, the searching for a new candidate of alternative anti-MRSA agent with a lesser toxicity is required, and one of the possible ways to support this goal is the use of plantderived agents. e herbal medicinal products have been considered as a natural source for alternative treatment for bacterial infections. Citrus reticulata Blanco (commonly known as mandarin orange and tangerine orange) and Citrus aurantifolia (commonly known as key lime, common lime, and lime) are edible fruits belonging to the family Rutaceae. ey are cultivated in tropical and subtropical regions worldwide. Generally, they can be applied as food and drink flavoring ingredients as they have signature citrus flavoring and scents. C. aurantifolia is traditionally used to promote the digestion process and for antidiabetic and antihypercholesterolemic purposes. e oil extracted from the C. aurantifolia fruits can be used for relieving cold, asthma, and arthritis [5].
Additionally, antibacterial activities of crude extracts and essential oils (EOs) from different parts (leaves, stems, roots, flowers, and peels) of C. reticulata and C. aurantifolia have demonstrated broad-spectrum antibacterial activities toward clinically important pathogens with a major activity against Gram-positive bacteria, especially S. aureus [7,[9][10][11][14][15][16][17][18][19][20]. Previously, some studies have investigated the antistaphylococcal effect of these citrus-derived products, mostly against the reference strains of MRSA and methicillin-susceptible S. aureus (MSSA). Bektaš et al. reported the inhibitory effect of orange juices on MSSA ATCC 25923, MRSA NCTC 1249, and clinical isolates of MRSA [21]. Vong et al. reported the inhibitory effect of the fruit juice extract of C. aurantifolia from Malaysia against MRSA ATCC 33591 [22]. Chao et al. reported the inhibitory effects of several commercial citrus EOs including C. aurantifolia EO which inhibited the growth of MRSA ATCC 700699 [23]. Up to now, there are limited data on anti-MRSA effects of the citrus extracts and EOs toward the clinical isolates, in which more variation in the antimicrobial susceptibility pattern is observed. erefore, this study aimed to determine the antibacterial activities of the citrus EOs and their major compounds against clinically isolated MRSA and MSSA. In addition, the synergistic effects of citrus EOs and their major compounds to improve the efficacy of the antibacterial agent, gentamicin, against clinically isolated MRSA and MSSA were also investigated.

Plant Materials.
e fresh fruits of C. reticulata Blanco (mandarin orange) and C. aurantifolia (Christm.) Swingle (lime) were collected from Chiang Rai Province, the northernmost of ailand, in December 2019. e plant samples were identified, and voucher specimens (BCU no. 015859 and BCU no. 015858) were housed at the herbarium of the Department of Botany, Faculty of Science, Chulalongkorn University, ailand. e fruit peels were ground, suspended in distilled water, and then processed through hydrodistillation for 3 h. Essential oils were separated from the aqueous layer using a micropipette. e oils were dried over anhydrous sodium sulfate, filtered, and stored at 4°C [24]. In this study, the percentage yields of extracted C. reticulata essential oil (CREO) and C. aurantifolia essential oil (CAEO) were 0.48% and 0.30%, respectively. e extracted EOs with a density of 0.8 g/mL were stored at 4°C and protected from light. A stock solution was prepared at the concentration of 400 mg/mL in dimethyl sulfoxide (DMSO) before use.

Bacterial Organisms.
e tested bacterial organisms contained 2 American Type Culture Collection (ATCC) bacterial strains including Staphylococcus aureus ATCC 43300 (methicillin-resistant S. aureus, MRSA) and S. aureus ATCC 25923 (methicillin-susceptible S. aureus, MSSA) as well as clinically isolated MRSA (n � 40) and MSSA (n � 45). e clinical isolates were obtained from the Division of Microbiology, Department of Central Laboratory and Blood Bank, Faculty of Medicine, Vajira Hospital, Navamindradhiraj University, ailand, and identified by matrixassisted laser desorption/ionization time-of-flight mass spectrometry (MALDI-TOF MS).
ese clinical isolates were originally collected from skin, bloodstream, respiratory, and urinary tracts (Table 1). e bacteria were maintained at −70°C and subcultured on blood agar at 37°C overnight prior the assay.

GC-MS Analysis.
e separation and identification of volatile components of CREO and CAEO were carried out by gas chromatography-mass spectrometry (GC-MS) (GC 7890A/MS 5975C-MSD; Agilent Technologies, CA, USA). e capillary column Mega-5MS (30 m × 0.25 mm × 0.25 μm) was used. e GC conditions were programmed as the injection temperature 250°C, with oven temperature initially set at 50°C for 1 min and then gradually increasing at the rate of 3°C/min up to 250°C and held for 5 min. Helium was used as the carrier gas with a constant flow rate of 1.0 mL/min. e volume of injection was 1 μL of ethanol solution in a split mode (1 : 10). e MS transfer line temperature was set at 250°C with the electron ionization (EI) mode at 70 eV ionization potential. e mass-to-charge (m/z) range was from 40 to 650 m/z. Compounds were further identified by matching their mass spectra fragmentation pattern and retention time with standard reference compounds, compared their MS results with the NIST 2011 library, and stored in the GC/MS database for confirmation.

Agar Disk Diffusion.
Agar disk diffusion was performed to screen the in vitro antibacterial activities of the EOs as previously described [26]. Sterilized disks (6 mm in diameter) impregnated with 10 μL of each EO or pure limonene (Lot no. MKCD9298; Sigma-Aldrich, St. Louis, MO, USA) were placed on the surface of the Mueller-Hinton agar (MHA; Oxoid, England) plate after tested bacteria (0.5 McFarland unit) were inoculated. e disk containing 10 μL of 4% DMSO and commercial gentamicin disk (10 μg) (Oxoid, England) were used as negative and positive controls, respectively. After incubation at 37°C for 18−24 h, the IZD of the EOs was measured and interpreted following the criteria: no activity, IZD � 6 mm; weak activity, 6 mm < IZD ≤ 12 mm; moderate activity, 12 mm < IZD < 20 mm; and strong activity, IZD > 20 mm [27].

Determination of MIC.
e minimum inhibitory concentrations (MICs) of the EOs were evaluated by the resazurinbased 96-well plate microdilution method as previously described with some modifications [28]. Fifty microliters of various concentrations of EOs and limonene were prepared by a serial two-fold dilution with cation-adjusted Mueller-Hinton broth (CAMHB) in a sterile 96-well microplate to obtain the final concentrations ranging from 0.1 to 32.0 mg/mL. In addition, gentamicin supplement was included in the experiment with the final concentration of 0.1 to 256.0 μg/mL. Afterward, 50 μL of tested bacteria was added into each well to obtain a final concentration of 5 × 10 5 CFU/mL. Only EO dissolved in CAMHB (oil control), only bacterial suspension in CAMHB (bacterial control), and 4% DMSO with bacterial suspension (diluent control) were also included. After incubation at 37°C for 24 h, 5 μL of 0.015% resazurin (Sigma-Aldrich, St. Louis, MO, USA) was added into each well and further incubated for 2 h in the dark. e bacterial growth was visually observed by the change of resazurin natural color (blue-purple) into the reduced form (red-colorless). e MIC was defined by the lowest concentration that completely inhibits the growth of bacteria (no color change). Consequently, one loop of the MIC suspension that showed no color change was cultivated on the MHA plate and further incubated at 37°C for 18−24 h. e minimum bactericidal concentration (MBC) was defined by the lowest concentration that completely kills bacteria on the agar plate. e MIC index (MBC/MIC ratio) was calculated to classify the type of antimicrobial substances and interpreted using the following criteria: bactericidal, MIC index ≤ 4; bacteriostatic, MIC index > 4; and resistance, MIC index ≥ 32 [29].

Checkerboard Titration Assay.
e checkerboard titration assay was performed to evaluate the synergistic interaction among EOs and limonene combined with gentamicin against 7 clinical isolates. is method was based on the broth microdilution assay with the final volume of 100 μL. In brief, 25 μL of various concentrations of EOs or limonene ranging from 0.3 to 32.0 mg/mL was prepared by serial 2-fold dilution in the 96-well microplate. In the meanwhile, various concentrations of gentamicin were prepared ranging from 0.001 to 128.0 μg/mL. en, 25 μL of each concentration of gentamicin was added into each concentration of EO or limonene to perform checkerboard testing. Fifty microliters of tested bacteria were added into each well to obtain a final concentration of 5 × 10 5 CFU/mL and incubated at 37°C for 18−24 h. Afterward, 5 μL of 0.015% resazurin was added and further incubated for 2 h in the dark. e bacterial growth was visually observed by the color change of resazurin. e combination effect of either EOs or limonene with gentamicin was determined by using the fractional inhibitory concentration index (FICI) value using the following formula: e interaction was interpreted by using the following criteria: FICI ≤ 0.5, synergy; 0.5 < FICI ≤ 1, additive; 1 < FICI ≤ 4, indifference; and FICI > 4, antagonism [30].

Statistical Analysis.
All experiments were performed in triplicate. e data were analyzed with the descriptive statistics, Kruskal-Wallis test, and Mann-Whitney U test using the IBM Statistical Package for Social Services (SPSS) version 21.0 (IBM, Armonk, NY, USA). A p value <0.05 was considered statistically significant.

Antibiotic Susceptibility Pattern.
e susceptibility patterns of tested bacteria in this study are shown in Table 4. e susceptibility to cefoxitin (10 μg) was used to classify between MRSA and MSSA. e results showed that S. aureus ATCC 43300 and 40 clinical isolates of S. aureus were  It also showed that MSSA ATCC 25923 was susceptible to all other tested antibiotics including gentamicin (IZD: 26.7 ± 1.    For other antibiotics, MRSA isolates showed high rates of resistance to ciprofloxacin, erythromycin, and clindamycin (76.9-95.0%) when compared to MSSA isolates (6.7-16.7%) with the IZDs of 8.5 ± 7.2 mm, 9.7 ± 7.9 mm, and 9.7 ± 8.1 mm for MRSA and the IZDs of 25.2 ± 6.3 mm, 23.0 ± 8.6 mm, and 23.3 ± 7.7 mm for MSSA (p < 0.01). In this study, all MRSA (40/40, 100%) were susceptible to vancomycin, which is a last-resort antibiotic for multidrug-resistant (MDR) bacteria at MIC 0.5 ± 0.2 μg/ mL. As the expected results, MRSA isolates were less susceptible to β-lactam antibiotics, ampicillin and amoxicillin, when compared to MSSA isolates with the IZDs of 9.7 ± 5.8 mm vs. 21.9 ± 8.6 mm for ampicillin and 10.4 ± 6.6 mm vs. 25.6 ± 6.9 mm for amoxicillin (p < 0.01).
All these together, it demonstrated that MRSA isolates seem to be resistant to all tested antibiotics except for vancomycin.
In addition, MDR S. aureus, which is classified by the resistant pattern to at least one agent among at least three antibiotic classes, was found at high rate among MRSA isolates (

Antibacterial Activity of Citrus EOs and Limonene.
e values of IZD, MIC, and MBC of the citrus EOs and pure limonene against tested bacteria are shown in Table 5. In this study, sterile disk containing 4% DMSO had no inhibitory effect, while commercial gentamicin disk (10 μg) had the effect toward all S. aureus isolates including laboratory strains. By agar disk diffusion, the results of antibacterial activity of CREO, CAEO, and limonene revealed that these agents had an inhibitory effect against almost all tested bacteria including MRSA and MSSA. CREO exhibited antibacterial activity against both MRSA ATCC 43300 and MSSA ATCC 25923 with the IZDs of 11.3 ± 1.5 mm and 11.7 ± 1.5 mm, respectively. Limonene exhibited antibacterial activity against both MRSA ATCC 43300 and MSSA ATCC 25923 with the IZDs of 12.3 ± 1.5 mm and 13.0 ± 1.7 mm, respectively. On the contrary, CAEO exhibited antibacterial activity against MSSA ATCC 25923 (IZD: 10.3 ± 3.8 mm) but not MRSA ATCC 43300 (IZD: 6.0 ± 0.0 mm).
For clinical isolates, the IZDs of CREO against MRSA and MSSA were 11.7 ± 3.1 mm and 12.0 ± 2.5 mm, respectively.

Synergistic Activities of Gentamicin Combined with Citrus EOs and Limonene.
e combination interaction of gentamicin, which is a standard antibiotic against S. aureus, combined with citrus EOs or limonene is presented in Table 6. In this study, the synergistic effect was evaluated against 7 clinical isolates containing 5 isolates of MRSA and 2 isolates of MSSA using the checkerboard titration assay. e results demonstrated that gentamicin in combination with CREO showed synergistic interaction (FICI: 0.012-0.258) among the most MRSA (3/5, 60.0%) and MSSA isolates (1/2, 50.0%). On the contrary, gentamicin in combination with CAEO showed synergistic interaction (FICI: 0.012-0.016) in only 2 isolates of MRSA (2/5, 40.0%) and none in MSSA isolates (0/2, 0.0%). Likewise, gentamicin in combination with limonene showed synergistic interaction (FICI: 0.012-0.375) in almost all MRSA isolates (4/5, 80.0%), but none in MSSA isolates (0/2, 0.0%). However, no antagonistic effect was observed in the combination of gentamicin with CREO, CAEO, or limonene. ese findings revealed the synergistic effect of gentamicin with the citrus EOs or limonene on clinical isolates of S. aureus, especially MRSA.

Discussion
e fruit peels of Citrus spp. are byproducts remained in manufacturing processes of several products such as orange juice and lime juice. e development of these byproducts should be approved to increase their worth. One advantage of the fruit peels is that some can be developed as natural antibiotics. Antibacterial activity of the Citrus spp. on a panel of clinical isolates of the resistant and susceptible strains of S. aureus has not been reported. is study was therefore interested to apply EOs extracted from the fruit peels of C. reticulata (CREO) and C. aurantifolia (CAEO) to fight against the clinical strains of resistant and susceptible S. aureus, with our expectation to increase the economic value of their byproducts and reduce the emerging of antibiotic-resistant S. aureus. Additionally, this study determined the antibacterial activities of these two citrus essential oils, CREO and CAEO, and their major components in both single effect and combination interaction with gentamicin. e extraction yields of CREO (0.48%) and CAEO (0.30%) were in similar ranges of amount with the previous studies (0.22-0.57%) [13,14]. However, this is difficult to make a comparison since there are many variation factors influenced to the yields of EOs such as climate, geographic distribution, genetics of the plant, the part of the plant used, the degree of freshness, the drying period, and the extraction method [31].
e present study showed that CREO and CAEO are rich in monoterpenes with the major component being D-limonene (72.53% in CREO and 62.95% in CAEO).
is is similar to the previous study review that mentioned the contents of limonene in CREO and CAEO of 67.0 to 86.0% [7,11,14,20,32]. However, the contents of limonene found in CREO and CAEO in this study were higher than previous studies that reported 29.3 to 58.9% of limonene [16,17,33]. In fact, the chemical compositions of EOs vary depending on origin, genetic background, season, climate, age, ripening stage, and method of extraction [10].
is study used a panel of clinical Gram-positive strains with different susceptibility profiles to evaluate antibacterial activities of citrus EOs. e chosen bacterial strain was S. aureus, including MSSA, MRSA, and MDR phenotypes. It showed that 77.5% of MRSA was MDR. In addition, almost all MRSA were less susceptible than MSSA in various antibiotics such as ampicillin (β-lactam), amoxicillin (β-lactam), ciprofloxacin (fluoroquinolone), erythromycin (macrolide), and clindamycin (lincosamide). erefore, these antibiotics could not be used for the treatment of MRSA infections. Fortunately, some MRSA were still susceptible to gentamicin (25/40), and vancomycin-resistant isolates were not observed in this study. However, the emergence of vancomycin-resistant S. aureus (VRSA) has been reported. e prevalence of VRSA was 1.2% in Asia, 1.1% in Europe, 3.6% in America, and 2.5% in Africa [34]. e prevalence might be increased in the future. To reduce the usage of vancomycin, the combinations of EOs and limonene with gentamicin were evaluated.
In this study, it appeared that CREO, which contained 72.53% of limonene, had higher effectiveness against MRSA e Scientific World Journal and MSSA strains than that observed in CAEO, which contained 62.95% of limonene; in addition, purified limonene (97%) also showed a promising effect. In addition, a higher concentration of limonene in CREO leads to a better synergistic effect than that of CAEO. Based on these observations, it indicated that limonene had an influence on the antibacterial potential of EOs. is is in agreement to previous reports [16,20,[35][36][37] but in contrast to some other studies [14,31,38]. e antibacterial mechanism of citrus EOs has been described by a previous study. A possible mechanism was cell wall disruption by citrus EOs on MRSA [39]. is study firstly demonstrated the synergistic effects of CREO, CAEO, and limonene with gentamicin on clinical isolates of MRSA and MSSA. is study proposed that the antibacterial activity of the citrus EOs could attribute to their lipophilicity property and the synergy between their major monoterpene hydrocarbons, limonene, and other minor components such as linalool which has been considered to be an antibacterial agent [37]. e mode of action of EOs could be due to the diffusion and accumulation of the oil in the bacterial cell membrane and then increasing cell membrane permeability, leading to cell lysis and leakage of intracellular components. In addition, the disturbance of the cell membrane may disturb vital processes such as energy conversion, nutrient processing, the synthesis of structural macromolecules, and the secretion of growth regulators [16,20]. In addition, limonene is believed to accumulate in the bacterial cell membrane and cause the loss of membrane integrity, dissipation of the proton motive force, inhibition of respiration, and ion transport processes [7,37,40]. In fact, there are several gentamicin-resistant mechanisms. One resistant mechanism is reducing uptake or decreased cell permeability. e proposed mechanism of synergism is bacterial membrane disruption by the EOs, leading to easier diffusion of gentamicin, an aminoglycoside, across the bacterial membrane. Consequently, this aminoglycoside is able to inhibit bacterial protein synthesis by binding to the 30S ribosomal subunit.
Both CREO and CAEO had an inhibitory effect towards MRSA and MSSA isolated from various types of clinical specimens including skin, bloodstream, respiratory, and urinary tracts (data not shown). It implied that these EOs can be applied as antibacterial agents in several products, such as handwashes and nasal or oral sprays, as well as in several routes of exposure via topical, inhalation, or oral routes. e citrus EOs have been classified as Generally Recognized As Safe (GRAS) [41]. Aumeeruddy-Elalf et al. reported that hydrodistilled essential oils from Citrus spp. fruit peels (C. reticulata) have no cytotoxicity to human cells [12]. Lime and mandarin essential oils up to 100% had no irritating and sensitizing effects on humans. Acute dermal LD 50 of lime EO in rabbits and that oral LD 50 in rats were equal to >5 g/kg [10]. By this strategy, it can minimize the adverse effects of gentamicin in view of reducing the treatment dosage to the resistant bacteria, including reducing treatment costs and providing a therapeutic option with greater antimicrobial potential. Further studies on the precise mode of action, therapeutic dosage, tolerability, and safety of the EOs are necessary to provide therapeutic usage of EOs and in combination with antibiotics.

Conclusions
is study demonstrated that EOs extracted from C. reticulata and C. aurantifolia exhibited antibacterial activities against clinical isolates of S. aureus, both MRSA and MSSA. e synergistic effects of EOs with gentamicin toward the clinical isolates of MRSA were also revealed. e usage of these EOs would directly inhibit both susceptible and resistant bacteria and indirectly delay the emergence of bacterial resistance, hence the potential of plant-derived antibacterial agents to be used as a complementary therapy with the established antibiotics that would allow for dose reduction of the antibiotic, thereby delaying and reducing the emergence of antibiotic-resistant strains as well as minimizing the possible side effects.

Data Availability
All data are included within the manuscript as tables and figures.

Ethical Approval
is study, including the collection of clinical bacterial isolates, was approved by the Research Ethics Committee of the Faculty of Medicine, Vajira Hospital, Navamindradhiraj University, ailand (no. COE011/2019), and the Research Ethics Committee of Rangsit University, ailand (no. RSUERB2019-027).

Conflicts of Interest
e authors declare that they have no conflicts of interest.