Synergistic effects of essential oils and phenolic extracts on antimicrobial activities using blends of Artemisia campestris, Artemisia herba alba, and Citrus aurantium

Plant materials

The aerial parts (stems and leaves) of AC and AHA were gathered in July 2019 from two locations in Laghouat, called Milok and Aflou, respectively. The ripe fruits of CA were collected in January 2019 from trees located in the downtown Laghouat. The identity of the plant samples was confirmed by Professors Mohamed Yousfi and Mohamed Ouinten from the Laboratory of Fundamental Sciences at Amar Telidji University of Laghouat, Algeria. The samples were air-dried in shade at room temperature for 2 weeks. A voucher specimen with registration numbers (AC-AP07/19, AHA-AP07/19, and CA-P01/19) was deposited at the herbarium of the Fundamental Sciences Research Laboratory, Laghouat University, Algeria.

EO extraction and composition analysis

Extraction of EOs from the peels of fresh fruits of CA and the dried aerial part of Artemisia species (AC and AHA) was performed by a hydro-distillation method using a Clevenger-type apparatus. The obtained EOs were dried over anhydrous sodium sulfate (Na₂SO₄) and kept in the dark at 4°C until further processing. The yield of EOs was calculated upon dried sample’s weight and was expressed as “% v/m.” The identification of EOs’ composition was previously reported in recent work [20].

Phenolics solvent extraction

After drying, the plant material (peel of CA and aerial parts of Artemisia species) was finely powdered using a coffee grinder (Cyber Electric Coffee Grinder 200 W, CYCG-72, Cyber RC, Jiangsu, China). The phenolics were extracted using a modified version of the solid–liquid extraction method described by Djeridane et al. [21] and Ghiaba et al. [22]. The sample powder (5 g) was mixed with a solvent system (100 mL) composed of methanol and distilled water at 80% (v/v) and allowed to stand at room temperature for 24 h. Afterward, the extracts were passed through a Whatman No.4 paper, and the residue was re-extracted twice with 50 mL of the same solvent for 24 h and filtered. The three filtrates were combined. After removing the methanol at a reduced pressure using a rotary evaporator (Heidolph Instruments GmbH & Co. KG, Germany) (40°C), the obtained water fraction was treated with petroleum ether to eliminate lipids and pigments. The extraction was continued by adding ethyl acetate thrice (1:1, v/v) to obtain organic phases. The three organic phases were gathered, dried with anhydrous sodium sulfate, and then evaporated using a rotary evaporator (Heidolph Instruments GmbH & Co. KG, Germany). Finally, the crude extracts were weighed and stored at 4°C for further investigation.

Microbial strains

Five microorganisms were used in this study to evaluate the antimicrobial activities of extracts and their combinations. The target strains are accounted for the most commonly involved pathogenic microbes in human infections, including Gram-positive bacteria (Staphylococcus aureus ATCC 25923 and Enterococcus faecalis ATCC 29212), Gram-negative bacteria (E. coli ATCC 25922 and P. aeruginosa ATCC 27853), and a yeast of the Ascomycetous phylum (Candida albicans ATCC 10231). These microorganisms are part of the American Type Culture Collection (ATCC). All the strains were maintained on selective agar slants.

Antimicrobial activity

Evaluation of the synergistic effects between extracts

There is currently no standardized methodology available for quantifying antioxidant synergistic interactions. Different methods have been used by researchers, including the combination index (CI), isobolographic method, comparison between experimental and theoretical values of combined ingredients, and comparison between combinatorial and individual effects [24].

In this study, we employed the CI to analyze possible synergistic effects. The CI is a practical model designed for the analysis of synergistic interactions between two or more components. The isobologram, on the other hand, is a graphical method that is only feasible for binary combinations as it is not possible to form multi-dimensional isobole lines. Additionally, comparing the combinatory effect to the individual effect is a common mistake in synergy analysis because it cannot distinguish between an additive and a synergistic effect. Comparing the experimental value to the theoretical value (additive effect) calculated by summing the single effects is also a mistake in synergy quantification because, in this case, the single ingredient does not affect the other (no interactions) [24].

To determine the type of interaction within extracts (EOs and PEs), different doses of extracts from AC and AHA were mixed together with varying doses of extract from CA, and the antimicrobial activities of the mixtures were determined.

The antimicrobial effects of interactions between extracts (EOs or PEs) against bacterial strains were determined using the agar diffusion method (measured by using DIZ), and the antimicrobial MIC and MBC were determined using the broth microdilution method.

The CI (also called interaction index), as described by Chou [25] and Zhou et al. [26], was used to evaluate the interactions of binary and ternary combinations of extracts (EOs or PEs). The CI for a combination of extracts is defined as follows:

where CE is the combined effect of the mixture from extracts (EOs or PEs), namely DIZ, MIC, and MBC. x n is the fractional volume composition for extract (EOs or PEs) in the mixture. IE n is the individual effect of the extract (EOs or PEs) alone, namely DIZ, MIC, and MBC.

For broth microdilution method, higher antimicrobial activity is related to lower MIC and MBC values. Inversely, higher antimicrobial activity for the agar diffusion method is related to higher DIZ. Hence, the interpretations of synergism effects for the agar diffusion method are also dependent on the inverse of the calculated CI (1/CI)

where CE is the combined effect of mixture from extracts (EOs or PEs) including I% and DIZ. x n is the fractional volume composition for extract (EOs or PEs) in the mixture. IE n is the individual effect of the extract (EOs or PEs) alone including I% and DIZ.

The interpretations of synergism effects are dependent on the CI values as follows: Strong synergism: 0.1–0.3; synergism: 0.3–0.7; moderate synergism: 0.7–0.85; slight synergism: 0.85–0.9; nearly additive: 0.9–1.1; slight antagonism: 1.1–1.2; moderate antagonism: 1.2–1.45; antagonism: 1.45–3.3; strong antagonism: 3.3–10 [25].

Modeling of antimicrobial responses using simplex lattice mixture design (SLMD) method

An SLMD method was used to determine the effect of interactions of different combinatory plant extracts (EOs and PEs), namely AC (A), AHA (B), and CA (C), on the performance of antimicrobial response. Constituent ratios were presented as a portion of the combination with a total value of 1. Response variables consisted of antimicrobial properties measured as DIZ (agar diffusion assay). A total of 16 combinations of each extract (EOs and PEs) were adopted. The computational works, mathematical modeling, and preparation of the ternary contour graphical presentations of the models were performed using Design Expert statistical software (Version 10.0.3, SAS Institute Inc., Cary, NC, USA).

Statistical analysis

All experiments in this work were repeated three times; each data point in the results is the mean of three parallel measurements, presented as mean values ± SD (standard deviation) using Microsoft Excel software. Experimental results were analyzed through analysis of variance (ANOVA) combined with Tukey’s post hoc (HSD) multiple range tests as a post-test procedure. Values with p < 0.05 were considered significant. Data modeling by the SLMD was used to determine the effects of different combinatory plant extracts, namely AC (A), AHA (B), and CA (C), on the performance of antimicrobial response using different methods. The computational works, mathematical modeling, and preparation of the ternary contour graphical presentations of the models were performed using Design Expert (Stat-Ease Inc., Minneapolis, USA) statistical software, version 12.

Antimicrobial activity in single extracts

The screening of antimicrobial activity of EOs and PEs from AC, AHA, and CA used individually against different microorganisms was first evaluated by the agar disc-diffusion assay. The DIZ of microbial growth exerted by different extracts is reported in Table 1. Amoxicillin, tested as a positive control, showed antimicrobial efficacy against all microbial strains except C. albicans, while no effects were seen for the negative control (DMSO 10%, v/v) (Table 1, Exp. 33 and 34).

In most cases, the tested EOs exhibited an appreciable antimicrobial effect against different microorganisms (Table 1, Exp. 1–3). The best antimicrobial activity was achieved by AC EO with DIZ ranging from 7.83 ± 0.15 to 19.03 ± 2.27 mm, followed by AHA EO displaying DIZ ranging from 8 ± 0.46 to 11.07 ± 0.76 mm. Except for E. faecalis (DIZ value of 13.17 ± 2.15 mm), CA EO showed generally lower antimicrobial activity with DIZ values ranging from 0 to 8.80 ± 0.95 mm. The PEs (Table 1, Exp. 17–19) showed a marked activity against all the tested strains, except for CA PE against E. coli (7.17 ± 0.55 mm). The highest activity was also observed with AC PE (DIZ ranging from 9.53 ± 0.75 to 24.63 ± 0.55 mm). The maximum DIZ of AC PE was found against E. faecalis (24.63 ± 0.55 mm) and S. aureus (17.73 ± 0.97 mm), with DIZ closer to that of Amoxicillin. An intermediate effect was seen for AHA PE with DIZ ranging from 11.30 ± 1.30 to 19.53 ± 0.55 mm. CA PE exhibited the lowest effect (7.17 ± 0.55 ≥ DIZ ≥ 13.67 ± 1.25 mm).

Based on the MIC and MBC results (Table 2), the lowest MIC values of 1.25 and 2.5 mg/mL (corresponding to higher activity) were found in AC PE against E. coli and S. aureus, respectively. AHA PE showed intermediate activity toward both tested strains (MICs were 2.5 and 5 mg/mL, respectively), followed by CA PE with MIC values of 5 and 10 mg/mL, respectively. The MBC values were higher or equal to their MIC values; the lowest MBC value (5 mg/mL) was exerted by AC PE, followed by AHA PE (10 mg/mL), and CA PE displayed an MBC value of 10 mg/mL (S. aureus) and 20 mg/mL (E. coli). Both test results confirm that the antimicrobial activity of AC PE is strong, that of AHA PE is moderate, and that of CA PE is weak against Gram-positive bacteria (E. faecalis and S. aureus).

The Gram-positive bacteria (S. aureus and E. faecalis) showed the best susceptibility to the action of both EOs and PEs, mainly to the AC EO and AC PE. In one specific case, S. aureus was more resistant to the effect of CA EO (DIZ = 8.17 ± 0.35 mm) (Table 1). Among all the tested strains, only C. albicans was not sensitive to the effect of all tested EOs (DIZ ≤ 8 mm). However, it was susceptible to PEs (10.77 ± 1.42 mm > DIZ > 11.60 ± 0.53 mm). Negligible differences were observed between EOs and PEs’ response against E. coli. Except for AC EO, we have to note that P. aeruginosa exhibited strong resistance to the effect of EOs, and it was sensitive to the action of PEs, mainly to CA PE (DIZ = 13.67 ± 1.25 mm). Except for some specific cases, we have to note that Gram-negative strains (E. coli and P. aeruginosa) were more resistant to the effects of EOs and PEs than the Gram-positive bacteria (S. aureus and E. faecalis) and yeast (C. albicans). These findings were in line with other literature referring to the antimicrobial activity of EOs [6] and phenolic extracts [27] of other plants. However, some authors showed that EOs manifest no difference between the behaviors toward Gram-positive and Gram-negative bacteria [28]. The difference in the behaviors of EOs, as well as PEs, can be attributed to the fact that Gram-negative bacteria have an outer membrane containing lipopolysaccharide that surrounds an inner peptidoglycan layer, making it more complex and thereby limiting the entry of antibacterial agents. On the other hand, the outer membrane is absent in Gram-positive bacteria, and they are surrounded by a thick peptidoglycan layer. This structure, not dense enough to resist small molecules, may ease the diffusion of antibacterial agents through it to the cell membrane [2,29,30].

The observed variation in antibacterial effects among EOs and extracts is due to at least two factors: (i) the main compounds of EOs or extracts and (ii) the type of strain tested [6,31,32,33,34,35]. Based on gas chromatography–mass spectrometry analysis of three EOs performed in our previous study [20], a strong relationship was observed between the number of compounds and major compounds (>2%) in the EOs and their antimicrobial activity, especially against Gram-positive bacteria. Specifically, AC EO was found to have 59 constituents, with 10 represented by the main compounds (β-pinene, α-pinene, limonene/sylvestrene, p-cymene, β-myrcene, spathulenol, γ-terpinene, cis-β-ocimene, terpinen-4-ol, and α-terpineol). AHA EO had 33 constituents, with 5 compounds (camphor, α-thujone, β-thujone, camphene, and borneol) as the main components. CA EO possessed only 19 constituents, with limonene and β-myrcene as the major components. Previous reports have shown that many of the above-mentioned main components exhibit individual antimicrobial effects against different microorganisms [36,37,38,39,40]. However, the effect of minor compounds should not be ignored, as confirmed by Miladinović et al. [41], who emphasized the importance of minor components on the antibacterial activity of EOs via chemometrics. The mode of action of EOs involves multiple targets in the cell due to the high number of bioactive compounds [42]. Some common modes of action include targeting the cell wall, cell membrane, and cytoplasmic material, inhibiting the production of bacterial toxins, and in some cases, causing cell lysis and death [30,43].

Polyphenols have been reported to have considerable antibacterial capacity against foodborne bacteria [44]. The higher inhibition action of polyphenols on microbial strains, especially Gram-positive bacteria, might be due to their higher phenolics/flavonoids content which have been quantified in our previous study [20]. The current study showed a strong correlation between the phenolics/flavonoids content and antimicrobial efficacy in the case of Gram-positive bacteria (S. aureus and E. faecalis), but no correlation was observed in the case of Gram-negative bacteria and yeast. Similarly, no correlation was detected between the quantity of phenolics and antibacterial efficacy against Neisseria gonorrhoeae, a Gram-negative bacteria [12]. Moreover, Sateriale et al. [5] reported that the polyphenolic concentration of extracts seemed to be related to their antimicrobial activity when tested against cariogenic bacteria, including Streptococcus mutans and Rothia dentocariosa (Gram-positive strains). Additionally, the aqueous methanol extracts from AC and AHA were found by a report of us to be rich in total phenolics/flavonoids content [20]. According to Bakchiche et al. [45], the most abundant phenolic compounds in the aqueous methanol extracts of the two plants are 3-O-methylquercetin, eupatilin, acacetin, rutin, and chlorogenic acid. These extracts have been evaluated against different microorganisms, showing antibacterial activity mainly against Gram-positive bacteria, especially S. epidermidis.

The different modes of action in which polyphenols interfere with the physiology of bacteria may include disrupting membrane functions or inhibiting some virulence factors, such as enzymes, toxins, signal receptors, and bacterial biofilm formation [46].

Antimicrobial activity in extract combinations (synergistic effects)

The results of CI values and antimicrobial activity (DIZ) of different EO mixtures are gathered in Table 1 (Exp. 4–16). Interestingly, all tested mixtures exhibited synergism or moderate synergism against Gram-positive bacteria, E. faecalis (1/CI values ranging from 0.50 to 0.83) and S. aureus (1/CI values ranging from 0.60 to 0.85). Concerning Gram-negative bacteria, two moderate synergisms were detected with AC–AHA EO (2/3–1/3) and ternary combination (1/3–1/3–1/3), and one slight synergism was found with AC–AHA EO (1/3–2/3) against E. coli strain, indicating 1/CI values of 0.81, 0.78, and 0.86, respectively. However, no synergistic or additive effects were observed against P. aeruginosa strain in all tested combinations. For C. albicans, except AC–AHA EO, which showed slight synergism and nearly additive effect. No synergistic effect was observed in all other binary and ternary mixtures.

Similar to the results of single EOs, the EO mixtures are more effective against S. aureus and E. faecalis, confirming that Gram-positive species are more sensitive than Gram-negative bacteria and yeast to EOs. When comparing the antimicrobial activity of combinations with their single EOs, we can observe that all EO combinations caused significantly higher inhibition zones than CA and AHA EOs alone against S. aureus and E. faecalis (Gram-positive bacteria). Instead, many EO mixtures exhibit lower antimicrobial activity than AC EO alone against these strains. Moreover, AC EO significantly contributes to the antimicrobial activity of AC–AHA and AC–CA EO mixtures. AHA EO contributes only to the activity of AHA-CA EO combination. For ternary combinations, AC EO was found to contribute to high degree, followed by AHA EO. The increases of AC EO content, as well as AHA EO, in mixtures, cause a significant increase in the antimicrobial activity of Gram-positive strains. This suggests that the antimicrobial activity of the combinations is mainly related to the content of Artemisia EOs, especially AC EO. A strong positive correlation was observed between the number of compounds/main compounds in the EO and their antimicrobial activity against Gram-positive strain (as mentioned in the antimicrobial activity of single extracts section). We believe that increasing the proportion of EO, with a high number of components and major components, in a mixture would enhance the multiplicity and the quantity of active compounds present in the mixture, which could enhance their antimicrobial efficacy. Notably, the mechanism of action of EO depends on their chemical profile and the proportion of their bioactive compounds [30]. Otherwise, the effect of EO mixtures against E. coli, P. aeruginosa, and C. albicans is not always higher than the weak EO of CA and can, in many cases, lead to a decrease or almost no change in the inhibition zone. These findings elucidated that mixing between AC, AHA, and CA EOs improves the total antimicrobial effect, at least against Gram-positive strains.

Regarding phenolic extracts, the CI values and antimicrobial activities (DIZ) were evaluated for each experimental combination, as summarized in Table 1 (Exp. 20–32). Notably, the binary mixtures of AC PE (or AHA PE) with CA PE at all ratios exhibited moderate synergism against S. aureus (Gram-positive bacteria), with 1/CI values ranging from 0.71 to 0.83. Similarly, these mixtures showed synergism to moderate synergism against E. faecalis (Gram-positive bacteria) and demonstrated synergism at all proportions, with 1/CI values ranging from 0.62 to 0.80. Additionally, the combination of AC PE and CA PE at all ratios demonstrated synergism against Gram-negative E. coli (1/CI values of 0.55, 0.61, and 0.63), while the combination of AHA–CA PE at all ratios showed moderate to slight synergism against the same bacteria. Moderate synergism was observed only with the combinations of AC–CA PE (1/2–1/2) and AHA–CA PE (2/3–1/3) against Gram-negative P. aeruginosa. As for C. albicans, all binary combinations displayed slight synergism to nearly additive effects (1/CI ranging from 0.87 to 1.09). Moreover, the combinations prepared by mixing AC and AHA PE at different proportions showed moderate synergism to nearly additive effects against tested strains, but moderate antagonism was also observed at a ratio (2/3–1/3) against C. albicans strain (1/CI = 1.18). With the exception of C. albicans, the ternary mixture (AC–AHA–CA PE) at different proportions mostly exhibited synergism and moderate synergism (1/CI ranging from 0.55 to 0.85). In one specific case, the ratio (2/3–1/6–1/6) demonstrated nearly additive effects against P. aeruginosa strain (1/CI = 0.91).

The MIC and MBC values for the phenolic extract combinations are presented in Table 2. The MIC results showed that only two combinations (ternary combinations at ratios 1/6–1/6–2/3 and 2/3–1/6–1/6) exhibited moderate synergism against S. aureus strains, with CI values of 0.83 and 0.79, respectively. Furthermore, only the combinations of AHA with CA PEs at three different ratios showed synergistic effects (one synergism and two moderate synergisms) against E. coli strain. Notably, the calculated CI values from the MBC results indicated even more synergistic interactions than the MIC results, with many synergistic and additive effects observed for different mixtures against the tested strains. The ternary combination at the ratio 2/3–1/6–1/6 showed maximum synergistic effects against S. aureus (CI = 0.42), and the binary combination of AHA–CA PE (1/2–1/2) displayed synergism against E. coli with CI = 0.38. These findings suggest that the mixtures exert a synergistic bactericidal effect against the tested bacteria.

In terms of antimicrobial activity, all combinations showed strong activity compared to CA and AHA PE alone when tested against Gram-positive strains, which explains the observed synergistic effects. The ratio of the single extracts influenced the antimicrobial ability of the combinations. Interestingly, increasing the ratio of AC PE (active extract) in the mixtures led to an increase in the antimicrobial ability. This was demonstrated by the activity contribution of AC PE to the mixtures that include it. It was also observed that in the AHA–CA PE combination, an increase in the AHA PE from 1/3–2/3 to 2/3–1/3 resulted in an increase in activity, indicating that AHA PE contributes to the activity of this combination. As mentioned previously, for single extracts, a positive correlation was detected between the phenolics/flavonoids content and antimicrobial efficacy against Gram-positive bacteria. Therefore, it was expected that increasing the ratio of an extract with high phenolics/flavonoids content in a combination would enhance their phenolics/flavonoids content, which could enhance the antimicrobial ability of the mixture. One of the most likely hypotheses is that the synergistic antibacterial effects of phenolic extracts are due to differences in polyphenolic content and relative modes of action [5]. On the other hand, these observations were less pronounced toward Gram-negative strains and C. albicans. The increase in the ratio of active extract in the combinations led to an increase/decrease or almost no variation in the antimicrobial ability against these strains. Our findings demonstrated that the combination of AC, AHA, and CA PEs significantly enhanced the overall antimicrobial capacity of the combinations, especially against Gram-positive strains, which, in many cases, was more potent compared to the single extracts.

Some mechanisms of antimicrobial interaction that produce synergistic effects have been accepted. It has been demonstrated that the mixture of two or more EOs produces synergism due to the combined activities of several bioactive components of EOs, which have multiple sites of action. This makes it difficult for microbial organisms to develop resistance to multiple components of two or more EOs. EOs mainly target several sites in microbial organisms [47]. Moreover, many investigations have reported that the synergistic effects of EOs may be due to the disintegration of the cell wall by some components, which then facilitate the entry of other active components into the cytoplasm. Once inside, these components can interact with different intracellular targets [13,48,49].

Furthermore, it has been suggested that the synergistic effect of EOs can be attributed to the increase of one of three factors: hydrophobic properties, the potency of their functional groups, and water solubility. These factors determine the antimicrobial ability of monoterpenes [49]. Additionally, the amphipathic properties of flavonoids play an essential role in their antimicrobial abilities. Flavonoids are chemical compounds that have both hydrophilic and hydrophobic moieties. Hydrophobic substituents, such as prenyl groups, alkylamino chains, alkyl chains, and nitrogen or oxygen-containing heterocyclic moieties, usually enhance the antimicrobial activity of all flavonoids [50].

Due to the high amount of peel produced, Citrus byproduct processing could be a significant source of EO, polyphenols, pectin, dietary fibers, and carotenoids [51]. Worldwide, each year, over 30% of CA production is used in juice processing. Consequently, a considerable volume of byproduct wastes, such as peels, is generated yearly, about 0.5 kg per 1 kg of raw fruit [52,53]. In addition, a small amount of this byproduct waste is used as molasses for animal feeds, while the largest part is disposed of without taking advantage of its pharmacological properties [52]. Therefore, it is very interesting to utilize the peel extracts (EO and phenolic compounds) of CA as food additives, preservatives, or antimicrobial agents by integrating them into plant mixtures with a high proportion (2/3). Fruit peel EOs are considered generally recognized as safe (GRAS) and can be used to improve food safety due to their unique antimicrobial properties [30].

On the other hand, the obtained antimicrobial activities, as well as the synergistic effects of the combinations, were mainly provided by Artemisia species extracts, especially AC even when represented in low proportions (1/3) of combination. Considering the abundance of CA, the mixtures AC–CA and AHA–CA, at a ratio 1/3–2/3, and AC–AHA–CA, at a ratio of 1/6–1/6–2/3, can be used as effective and appropriate natural antimicrobials in food preservation, at least against Gram-positive microbial strains. In addition to the synergistic/additive effects and the considerable antimicrobial activity compared to single extracts (EOs and phenolic compounds), these mixtures can serve as food preservatives for other benefits: increasing the content of abundant plants, increasing the dose of high-yielding extract (EOs), decreasing the extract dose with high toxicity, and reducing the side effects, as well as decreasing the required doses of the combined extracts. As a result, their impacts on the texture and organoleptic quality of foods would be minimized.

Antimicrobial responses using simplex lattice design

In order to determine the area of interest for the ternary combinations of the three EOs that provide the economical optimal antimicrobial activities while considering the use of CA EO as a dilution strategy, the SLMD method of general response surface methodology (RSM) was chosen. This method allows for the identification of potential synergistic effects and the prediction of the optimal area of interest with the highest antimicrobial activity using the low-abundant and low-cost CA EO. This strategy could have significant practical implications in the food industry for food preservation and conservation. Additionally, SLMD can be used to create mixtures of different EOs, which can enhance organoleptic properties such as smell and taste while also potentiating antioxidant and antimicrobial activities.

The antimicrobial activity data obtained from the mixture design analysis (responses) resulted in the development of prediction equations (equations (1)–(6)) for the tested strains. These equations allow for the prediction of antimicrobial activity in the ternary mixtures. All equations were expressed in coded variables. The antimicrobial activity, expressed as the DIZ, is also visually represented in Figures 1 and 2 as surface responses and contours plots (these illustrations were presented in 2D and in 3D).

Figure 1

Ternary contour plots in 2D and 3D showing the antimicrobial responses of the ternary mixture of the three EOs. A: AC, B: AHA, C: CA. (a) S. aureus and (b) E. faecalis.

Figure 2

Ternary contour plots in 2D and 3D showing the antimicrobial responses of the ternary mixture of the three phenolic extracts. A: AC, B: AHA, C: CA. (a) S. aureus, (b), E. faecalis, (c) E. coli, and (d) P. aeruginosa.