*Review* **Essential Oils and Their Major Components: An Updated Review on Antimicrobial Activities, Mechanism of Action and Their Potential Application in the Food Industry**

**Manasweeta Angane 1,2,3, Simon Swift 2, Kang Huang 1, Christine A. Butts <sup>3</sup> and Siew Young Quek 1,4,\***


**Abstract:** A novel alternative to synthetic preservatives is the use of natural products such as essential oil (EO) as a natural food-grade preservative. EOs are Generally Recognized as Safe (GRAS), so they could be considered an alternative way to increase the shelf-life of highly perishable food products by impeding the proliferation of food-borne pathogens. The mounting interest within the food industry and consumer preference for "natural" and "safe" products means that scientific evidence on plantderived essential oils (EOs) needs to be examined in-depth, including the underlying mechanisms of action. Understanding the mechanism of action that individual components of EO exert on the cell is imperative to design strategies to eradicate food-borne pathogens. Results from published works showed that most EOs are more active against Gram-positive bacteria than Gram-negative bacteria due to the difference in the cell wall structure. In addition, the application of EOs at a commercial scale has been minimal, as their flavour and odour could be imparted to food. This review provides a comprehensive summary of the research carried out on EOs, emphasizing the antibacterial activity of fruit peel EOs, and the antibacterial mechanism of action of the individual components of EOs. A brief outline of recent contributions of EOs in the food matrix is highlighted. The findings from the literature have been encouraging, and further research is recommended to develop strategies for the application of EO at an industrial scale.

**Keywords:** essential oil; peel; antibacterial; antimicrobial; mechanism of action; preservation

#### **1. Introduction**

Antimicrobial agents used to kill or inhibit the growth of pathogenic or food spoilage bacteria can exist in natural or synthetic forms. The use of synthetic antimicrobial compounds as food preservatives has raised consumers' concerns, since they present numerous toxicological difficulties and may not be safe for human consumption [1]. Hence, over the last two decades, natural antimicrobial agents such as essential oils (EOs) have received renewed interest from the scientific community, owing to their unique physicochemical properties and diverse biological activities [2]. In the definition coined by Rios [3], EOs are aromatic, oil-like volatile substances present in plant materials such as fruits, bark, seeds, pulp, peel, root and whole plant. These substances form in the cytoplasm, and generally exist as tiny droplets sandwiched between the cells. In recent years, increasing awareness about the "green, safe and clean" environment and a growing appeal for "green consumerism" have prompted the production of foods free of synthetic preservatives [4,5].

**Citation:** Angane, M.; Swift, S.; Huang, K.; Butts, C.A.; Quek, S.Y. Essential Oils and Their Major Components: An Updated Review on Antimicrobial Activities, Mechanism of Action and Their Potential Application in the Food Industry. *Foods* **2022**, *11*, 464. https://doi.org/ 10.3390/foods11030464

Academic Editor: Yiannis Kourkoutas

Received: 8 January 2022 Accepted: 3 February 2022 Published: 4 February 2022

**Publisher's Note:** MDPI stays neutral with regard to jurisdictional claims in published maps and institutional affiliations.

**Copyright:** © 2022 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https:// creativecommons.org/licenses/by/ 4.0/).

EOs have been used for medicinal purposes and as therapeutic agents since ancient times [6]. Although food industries utilize EOs as a flavoring agent, their potential as a natural food grade preservative has not been fully explored. EOs present a valuable tool for food preservation due to their natural antimicrobial properties [7]. However, a detailed understanding regarding individual components of EOs, their antibacterial properties, mechanism of action and target organisms is required to support the implementation of EOs as food preservatives. Calo et al. [8] reported that EOs comprise numerous compounds such as aromatic hydrocarbons, terpene (monoterpenes and sesquiterpenes), terpenoids, esters, alcohols, acids, aldehydes and ketones, and their antibacterial activity is not solely contributed by any one compound. Recognizing the most potent antibacterial compounds from EOs is often tricky due to their chemistry complexity. To date, most studies have focused on studying the antimicrobial activity of EOs [5,8], with little discussion on the antibacterial activity of individual components in the EO or their mechanism of action. The antibacterial activity of EOs is not reliant on one specific mode of action; instead, EOs can attack several targets in a cell to inactivate the bacterium [7]. Evaluating EO's antibacterial properties and mechanism of action of their components may provide new insights into their applications in the food industry. This approach may reveal the concealed antibacterial properties of individual EO components, otherwise masked when EOs are studied as one single substance.

Several reviews [2,9,10] have outlined the antimicrobial activity of EOs extracted from various plant sources such as stem, bark, leaf, fruit, and seeds, but did not discuss the waste parts such as peel. The amount of waste produced by fruit processing industries is diverse [11]. Fruit peels generated by food industries are treated as agro-waste and are discarded in landfills, composted or fed to livestock [12]. Fruit waste produced in enormous quantities during commercial processing could present severe environmental threats [13]. Ayala-Zavala et al. [14] proposed using fruit by-products as an antimicrobial food additive, reporting that mandarins, papayas, pineapple, and mangoes accounted for 16.05%, 8.47%, 13.48% 11% of peel waste, respectively. On the other hand, fruit peel is a rich source of EOs and contains promising novel components of potential pharmacological, pharmaceutical and economic significance [13]. Moreover, fruit peel EOs are classified as GRAS (generally recognized as safe) and can be used to improve food safety due to their unique antimicrobial properties [15].

Studies on EOs extracted from various plant sources are well represented in the literature, and it is widely recognized that EOs possess a range of biological activities. For instance, EOs extracted from thyme [16], oregano, lavender [4,17], cinnamon, clove [18] and turmeric [19] have antibacterial, antifungal, algicidal, antioxidant, anticancer and anti-inflammatory activities. Chemical compositions and biological properties of plant EOs, in general, have been discussed in detail in reviews by Bakkali et al. [20], Burt [5] and Ju et al. [21]. A substantial amount of work has been carried out to evaluate the antimicrobial properties of EOs extracted from fruit peels; however, none of the reviews in the compiled data have exclusively discussed peel EOs. In light of these factors, this review aims to summarize the most significant findings of the antimicrobial properties of fruit peel EOs and their major components that contribute to microbial inactivation, with a focus on the mode of action of EO/EOs components. Finally, the application of various plant-derived EOs in the food industry is discussed, and future research directions and applications are presented.

#### **2. Chemical Composition of Fruit Peel Essential Oils**

Plants produce a variety of chemical compounds with antimicrobial properties. Some of these compounds are always present, while others are secreted in response to stress, such as infection, damage, predators, and weather variations. The chemical constituents in EOs are prone to variations depending on the time of harvest, cultivar, and the extraction method. Hydro distillation and steam distillation are frequently used to produce EOs at a commercial scale [5]. Identifying the most active compounds from EO can be a

cumbersome process. Gas chromatography (GC), gas chromatography-mass spectrometry (GC-MS) [22–24], high-performance liquid chromatography (HPLC) [25–27] and liquid chromatography coupled to mass spectrometry (LC-MS) [28] are the most widely used methods to study the chemical composition of EOs. The primary chemical components of EOs are terpenes and polyphenols. Figure 1 shows the structural formula of some of the major components of EOs. These chemical compounds have been reported to have antimicrobial properties and their mechanisms of action are discussed later (Section 4).

**Figure 1.** Chemical composition of essential oils (EOs).

Terpenes can be defined as a framework of numerous isoprene units (C5H8) merging to form a hydrocarbon molecule. They are derived from mevalonate and mevalonateindependent pathways [29]. Terpenes usually exist in EOs in the form of monoterpenes (C10H16) or sesquiterpenes (C15H24). However, other long-chain molecules such as diterpenes (C20H32), triterpenes (C30H48), tetraterpenes (C40 H64) are found in EOs in minor quantities [30]. Examples of terpene compounds include β-caryophyllene, *p*-cymene, αpinene, β-pinene, limonene, sabinene, γ-terpinene, α-terpinene, β-myrcene, cinnamyl alcohol, and δ-3-carene. Additionally present are terpenoids, identified as an oxygenated derivative of terpene compounds with an additional oxygen molecule, or their methyl group being moved or eliminated. Terpenoids are further categorized into esters, aldehydes, ketones, alcohols, ethers, and epoxides, with examples including menthol, geraniol, eugenol, thymol, carvacrol, geraniol, linalyl acetate, linalool, citronellal, citronellol and terpineol [7,31].

Polyphenols are secondary metabolites widely distributed in nature, usually derived from the phenylpropanoid pathway [32]. Polyphenols can be categorized into phenylpropenes and flavonoids, based on the number of phenol rings [33]. Phenylpropenes have derived their name from the six-carbon aromatic phenol group, and the three-carbon propene tail of cinnamic acid formed during the first step of phenylpropanoid biosynthesis [34]. Flavonoids are a group of phenolic compounds with a carbon framework (C6-C3-C6). The basic skeletal structure of flavonoids comprises a 2-phenyl-benzo-γ-pyrone consisting of two benzene rings (ring A and ring B) cross-linked to a heterocyclic pyrone (ring C) [35]. Based on the degree of oxidation, flavonoids are further classified into flavones, flavonols, flavanones and others [36].

A detailed analysis of the EOs of orange peel identified an abundant amount of limonene, ranging between 73.9–97.6%, while other monoterpenic alcohols, namely geraniol, linalool, nerol and α-terpineol, were present in minor quantities at concentrations of 2.1%, 4.1%, 1.5%, 2.4%, respectively [24]. This finding was in agreement with Ambrosio et al. [22] and Guo et al. [37], who reported similar compounds in orange peel EOs. However, some compounds such as cis-*p*-mentha and trans-*p*-mentha [22,37] were not reported previously [24]. These differences could be attributed to the different cultivars or growing conditions of the fruit analyzed in these studies. Moreover, a close resemblance was noted in the limonene content of grapefruit peel EO, which was present at a concentration of 93.3% [23], 91.5% [38] and 91.8% [39]. Other monoterpene compounds such as β-myrcene, α-pinene, sabinene, linalool and thujene were also reported [23,38,39]. In pummelo peel EO, limonene contributed up to 55.7% of the total EO composition, followed by β-pinene (14.7%), linalool (6.2%), β-citral (4.1%), germacrene-D (2.7%), α-pinene (2.3%), α-terpineol (2.0%), geraniol (1.6%), sabinene (1.3%) [39]. Tao et al. [40] reported similar compounds but at a much lower concentration, ranging from 0.08% to 0.63%. The difference in the extraction method, such as using a rotary evaporator at 40 ◦C [38], could have contributed to the significant loss of highly volatile compounds from the EO. Furthermore, Hosni et al. [41] and Hou et al. [42] found limonene to be the main component in mandarin peel EO, but other secondary compounds such as lauric acid, 1-methyl-1,4-cyclohexadiene, methyl linoleate, myristic acid, palmitic acid and β-myrcene were reported only by Hou et al. [42]. More recent evidence [43] highlights that out of 158 compounds found in feijoa peel EO, 89 compounds identified were novel; these compounds include esters, sesquiterpenes, monoterpenes, aromatic hydrocarbons, alcohols, aldehydes, ketones, hydrocarbon, acids and ethers.

Limonene is the predominant component in the EOs of orange [22,24,37,41], grapefruit [23,39], mandarin and pummelo [39,40] peels, and is thought to contribute to most of the antimicrobial activity of the fruit peels reviewed above [44]. However, Ambrosio et al. [22] argued that limonene is present in different concentrations in different fruit peels; thus, the antimicrobial activity of EOs cannot be ascribed solely to limonene. Additionally, studies have reported low antimicrobial activity of limonene when the pure compound was tested [45]. Hence, in citrus fruits, other minor compounds such as α-pinene, sabinene, linalool, β-citral, and germacrene-D could contribute to the antimicrobial activity.

#### **3. Antimicrobial Properties of Fruit Peel Essential Oils**

The antimicrobial activity of EOs can be seen as the inhibition of cell growth or by cell-killing. However, it is not easy to differentiate between these modes of action. The antimicrobial efficacy of EOs is dependent on their chemical composition, environmental conditions and the structures of the target bacteria (either Gram-positive or Gram-negative bacteria) [46]. Numerous in vitro techniques [47], such as the determination of minimum inhibitory concentration (MIC) and minimum bactericidal concentration (MBC) by broth macro dilution/microdilution or agar disk/well diffusion are applied to determine the efficacy of an antimicrobial compound. Agar disk/well diffusion and broth macro dilution/microdilution are widely used methods in clinical microbiology laboratories [48] and have recently been recognized as useful tools to determine the antimicrobial activity of EOs [49,50].

Many studies have illustrated the antimicrobial effect of fruit peel EOs against drugresistant, pathogenic and food spoilage bacterial strains. Some studies have found that EOs extracted from the fruit peels of banana [13], pomegranate [1] and citrus fruits such as sweet orange, grapefruit, lime, sweet lemon, mandarin, tangerine and pummelo [22,40,51–53] exhibited inhibitory activity against Gram-positive and Gram-negative bacteria. These studies indicate that fruit peels are a potentially valuable anti-microbial resource [42]. A wide range of foodborne pathogens could be inhibited by fruit peel EOs, including *Escherichia coli*, *Enterobacter cloacae*, *Klebsiella pneumoniae*, *Pseudomonas aeruginosa*, *Salmonella enterica* serovar Typhimurium, *Salmonella enteritidis*, *Bacillus subtilis*, *Bacillus cereus*, *Streptococcus faecalis*, *Listeria monocytogenes*, *Proteus vulgaris, Staphylococcus aureus* and others (Table 1). An overview of the antimicrobial activity of various fruit peel EOs and detection methods over the last 15 years is presented in Table 1.

#### *3.1. Citrus Essential Oils*

Abd-Elwahab et al. [51] reported the efficacy of EOs extracted from citrus peels, i.e., orange, lime, mandarin, and grapefruit, as having moderate to high antibacterial activity against *S. aureus*, *B. subtilis*, *E. faecalis*, *E. coli*, *Neisseria gonorrhoeae* and *P. aeruginosa*. Among those citrus EOs, lime peel EO was the most effective at inhibiting all six strains of pathogenic bacteria. The presence of coumarine and tetrazene in lemon peel [13] and citral, limonene and linalool in other citrus peel EO [51] may have accounted for their antimicrobial activity against these bacteria. On the contrary, Javed et al. [15] reported that amongst all tested citrus peel EOs (mandarin, tangerine, sweet orange, lime, grapefruit) mandarin peel EO possessed the highest antimicrobial activity. The inhibition zone for *Salmonella enterica* serovar Typhi, *E. coli*, *Streptococcus sp.* and *P. fluorescence* ranged from 20 to 30 mm for 10 μL and 9–16 mm for 5μL treatments of mandarin peel EO. The differing concentrations of the citrus peel EOs between the studies might explain these contradictory results.

#### *3.2. Orange Essential Oils*

Over the past decade, several studies [24,37,53–55] have examined the antibacterial properties of sweet orange (*Citrus sinensis*) EO. A broad-spectrum antibacterial activity was observed against a range of foodborne pathogens, confirming its potential to be a natural antimicrobial agent for food preservation. In a study conducted by Guo et al. [37], the antimicrobial activity of cold-pressed and light phase EO extracted from orange peel was compared using *E. coli*, *S. aureus*, and *B. subtilis*. It was reported that light phase EO showed a better antimicrobial activity compared to the cold-pressed EO. The higher antimicrobial activity can be attributed to a higher quantity of carvone and limonene in the light phase EO. Nwachukwu et al. [56] tested the efficacy of orange peel EO extracted using water and ethanol (hot and cold) against *E. coli*, *S. aureus*, and *Bacillus* sp. It was noted that hot ethanol extracted EO was more effective than the water extracted EO at inhibiting the three bacteria strains. Hot ethanol might have facilitated the better release of volatile compounds present in orange peel EO. These findings are similar to those of Ali et al. [55], Bendaha et al. [52], and Kirbaslar et al. [57], who reported similar antimicrobial activity of orange (*Citrus aurantim)* peel EO against *L. monocytogenes*, *S. aureus*, *E. coli*, *E. faecalis*, *B. cereus*, *K. pneumoniae* and *P. aeruginosa*. One of the significant drawbacks of these studies [15,37,51–53,55–57] was that they fail to consider the MIC and MBC values, thus providing no foundation for EO application in food. However, Geraci et al. [24] and Tao et al. [54] had reported the MIC values of orange peel EO, and as anticipated Gram-positive (*B. cereus*, *B. subtilis*, *S. aureus*) bacteria were reported to be more susceptible to the orange peel EO compared to the Gram-negative (*E. coli* and *P. aeruginosa*) bacteria.

#### *3.3. Grapefruit Essential Oils*

The antimicrobial activity of grapefruit (*Citrus paradisi*) peel EO against *B. subtilis*, *E. coli*, *S. aureus*, *S. enterica* serovar Typhimurium and *P. aeruginosa* was reported by Deng

et al. [23]. It was noted that Gram positive *B. subtilis* was the most sensitive amongst all strains investigated, while Gram negative *P. aeruginosa* was the least sensitive organism. This antibacterial activity may be attributed to the presence of abundant limonene in the grapefruit peel EO [44]. Similarly, pummelo (*Citrus grandis*) peel EO showed good inhibitory activity against Gram-positive bacteria (MIC- 9.38 μL/mL) and moderate activity against Gram-negative bacteria (MIC- 37.50 μL/mL) [40]. Terpene alcohols such as linalool are known for their inhibitory activity against Gram-negative bacteria [58]. Although a substantial amount of linalool was found in the pummelo peel EO, it did not inhibit *E. coli* [40]. This microorganism was only susceptible to pure linalool, but not to EO with linalool as one of the components in a mixture of compounds [59]. The use of EO instead of linalool alone might have contributed towards a higher MIC value of pummelo peel EO against *E. coli*.

#### *3.4. Essential Oils from Other Fruit Peels*

Several researchers have examined the antibacterial activity of various other fruit peels such as tamarillo [60], bergamot (*Citrus bergamia*) [57,61], sweet lemon (*C. limetta*) [53,62], *C. deliciosa* [63], kumquat (*C. japonica)* [64] and feijoa (*Acca sellowiana*) [65]. Surprisingly, Diep et al. [60] and Mandalari et al. [61] revealed that the tamarillo and bergamot peel flavonoids, respectively, exhibited strong antibacterial activity against Gram-negative bacteria such as *E. coli*, *Pseudomonas putida*, *S. enterica* serovar Typhimurium and *P. aeruginosa*, while Gram-positive bacteria (*B. subtilis*, *L. innocua*, *S. aureus*) were resistant. Similarly, El-Hawary et al. [63] found that *C. deliciosa* EO extracted from its leaves and peel was more effective against Gram-negative bacteria than the Gram-positive bacteria. In contrast, the inhibitions zones for bergamot peel EO (11mm to 16mm) with no clear distinction between Gram-positive and Gram-negative bacteria [57], and sweet lemon EO, demonstrated good antibacterial activity against both Gram-positive and Gram-negative bacteria with inhibition zones measuring between 10 to 35 mm [62].

Due to the difference in their cell wall structure [34], Gram-positive bacteria are more susceptible to EOs than Gram-negative bacteria [23,40,54,66]. However, published data have shown no clear differentiation between Gram-positive and Gram-negative bacteria [60,63]. The reason for this contradictory result is discussed in Section 4. It is somewhat surprising that many studies have assessed the antimicrobial activity by using only the agar disk/well diffusion method [15,22,39,50–53,55,56,60,63,65,67–72]. Agar disk/well diffusion is a quick typing tool used to determine the sensitivity of the bacterial strain. However, this quick typing tool cannot differentiate between bacteriostatic and bactericidal effects. The agar disk/well diffusion is a preliminary method that is not suitable to determine MIC or MBC, since it becomes quite challenging to measure the amount of EO diffused in the medium. Moreover, the hydrophobic nature of EO might pose an added challenge with regard to its ability to diffuse through the media, potentially resulting in uneven distribution. On the other hand, though tedious and time-consuming, broth macro dilution or microdilution methods allow quantifying the exact antimicrobial agent concentration that is effective against the pathogen and visibly distinguishes between bacteriostatic and bactericidal effects [49]. Most of the studies reviewed so far tend to overlook the importance of the broth dilution method for the determination of MIC and MBC of EOs, which is vital for determining the exact concentration required to kill bacteria, a prerequisite for assessing their potential application in food preservation.



**1.**Overviewofantimicrobialactivitiesoffruitessentialoils(EOs)and

10





*Lactococcus lactis*

showed no effect.

#### **4. Effect of Chemical Components of Essential Oils on Food Spoilage and Pathogenic Microbes**

In the literature, various modes of antimicrobial activity of EOs against a range of bacteria have been discussed [5,7,78,79]. However, before investigating the effect of fruit peel EO on microbes, we should have a closer look at the cell-wall structure of Gramnegative and Gram-positive bacteria (Figure 2).

**Figure 2.** Schematic representation of Gram-positive and Gram-negative bacterial cell wall.

The hypothesis that Gram-positive bacteria are more susceptible to the effect of hydrophobic compounds such as EOs was first proposed by Plesiat et al. [80] followed by Nazzaro et al. [34], Chouhan et al. [79] and Raut et al. [81]. The difference between the susceptibility is attributable to the fact that Gram-positive bacteria have a thick layer of peptidoglycan linked to other hydrophobic molecules such as proteins and teichoic acid. This hydrophobic layer surrounding the Gram-positive bacterial cell may facilitate easy entry of hydrophobic molecules. On the other hand, Gram-negative bacteria have a more complex cell envelope comprising an outer membrane linked to the inner peptidoglycan layer via lipoproteins. The outer membrane contains proteins and lipopolysaccharides (lipid A), making it more resistant to the hydrophobic molecules in EO [82].

Other researchers investigating the antimicrobial activity of EOs showed no notable difference between the MIC values of Gram-positive and Gram-negative bacteria [13,39,60,61,64]. Although it has been hypothesized that the outer membrane is almost impermeable to the hydrophobic compounds, Plesiat et al. [80] argued that some hydrophobic compounds might cross the outer membrane via porin channels. Similarly, Van de Vel et al. [58] believe that some EO molecules are more active against Gram-positive bacteria, while others are active against Gram-negative bacteria, but the mechanisms remain unknown. Most studies on the antimicrobial activity of EOs have used *E. coli* and *S. aureus* as model microorganisms to represent Gram-negative and Gram-positive bacteria, respectively [65,83,84]. This could lead to a generalization of results, as not all Gram-negative and Gram-positive bacteria would follow a similar trend as observed in *E. coli* and *S. aureus*. Furthermore, the mode of action of EO depends on its chemical profile and the ratio of its active components [85]. The possible mechanisms wherein EOs interfere with bacterial proliferation may involve the following: (1) the disintegration of the bacterial outer membrane or phospholipid bilayer, (2) alteration of the fatty acid composition, (3) increase in membrane fluidity resulting in leakage of potassium ions and protons; (4) interference with glucose uptake, and (5) inhibition of enzyme activity or cell lysis (Figure 3) [5,86].

**Figure 3.** Antibacterial mechanism of essential oils (EOs).

In general, fruit peel EOs may comprise more than a hundred compounds [43]. Major compounds can contribute around 85–95% of the total EO composition, while other minor compounds can be present in trace amounts. While these compounds may have specific antimicrobial effects, Cho et al.'s [86] review draws attention to the synergistic and additive effect minor compounds might have in combination with the other components. Terpenes and terpenoids are primary components of essential oil followed by polyphenols [32]. Here, we discuss the antimicrobial activity and mode of action of EOs and their components on the bacterial cell.

#### *4.1. Terpenes and Terpenoids*

Terpenes and terpenoids constitute a significant class of compounds in EOs known to have antimicrobial activity. The potential antimicrobial activity of thymol and carvacrol has been extensively discussed in previous reviews [7,34,79]; hence we exclude them from our discussion to focus on other EO compounds. Thymol and carvacrol are the major components of thyme and oregano oil, respectively, and are structurally analogous differing in the location of hydroxyl groups on the phenol ring [7].

It is well recognized that terpenes can disrupt the lipid assembly of the bacterial cell wall, leading to disintegration of the cell membrane, denaturation of cell proteins, leakage of cytoplasmic material, which ultimately causes cell lysis and cell death [47]. Kim et al. [87] were amongst the first to show the antimicrobial potential of EO components including citral, limonene, perillaldehyde, geraniol, linalool, α-terpineol, carvacrol, citronellal, eugenol, β-ionone and nerolidol against *E. coli*, *S. enterica* serovar Typhimurium, *L. monocytogenes* and *Vibrio vulnificus*. It was suggested that terpenes and terpenoids might interfere with oxidative phosphorylation or oxygen uptake in microbial cells, thereby inhibiting microbial growth [88]. Later, this hypothesis was supported by Zengin and Baysal's

study [89], wherein terpene compounds such as linalool, α-terpineol and eucalyptol were reported to damage the cell membrane and alter the morphological structure of *S. aureus*, *S. enterica* serovar Typhimurium and *E. coli* O157:H7. The plausible explanation for this observation was that these terpene compounds interacted with the membrane proteins and phospholipids, leading to cellular respiratory chain inhibition, interruption in oxidative phosphorylation, disruption of nucleic acid synthesis, and loss of metabolites [90].

Two studies conducted by Togashi et al. [90,91] examined the effect of geranylgeraniol, geraniol, nerolidol, linalool and farnesol on *S. aureus*. All these terpene alcohols were reported to have antibacterial activity, with farnesol and nerolidol demonstrating the most potent antibacterial activity as determined by the broth dilution technique. They also explored the mechanism of these terpene alcohols on the bacterial cell membrane by measuring the leakage of K<sup>+</sup> ions from the bacterial cell, anticipating that distortion of the bacterial cell membrane leads to leakage of K<sup>+</sup> ions, thus indicating the presence of membrane disrupting compounds. In support of this, Akiyama et al. [92] reported the strong inhibitory effect of farnesol against *S. aureus*. Farnesol has also exhibited notable antibacterial activity against biofilms of *S. aureus* and *S. epidermidis* [93,94]. Akiyama et al. [92] attributed these inhibitory effects of farnesol to its hydrophobic nature, which accumulates in the cell membrane, thus disrupting the cell membrane as illustrated by scanning electron microscopy (SEM). Furthermore, an ester compound of geranyl acetate makes it a more potent antimicrobial compound than its parent moiety (geraniol), purportedly due to its hydrophobicity [95]. However, past studies [31,96] have demonstrated the antimicrobial activity and mechanism of geraniol, rather than geranyl acetate. For instance, geraniol was noted to inhibit *E. coli* and *S. aureus* [97], and multidrug-resistant *Enterobacter aerogenes* by acting as an efflux pump inhibitor [96,98]. Similar, to farnesol, it is thought that the antimicrobial potential of geraniol was due to its hydrophobic nature.

Han et al. [44] and Liu et al. [99] examined the antibacterial mechanism of limonene on *L. monocytogenes* and the antibacterial mechanism of linalool on *P. aeruginosa*, respectively. In their analysis, Han et al. [44] and Liu et al. [99] demonstrated that the compounds distorted the cell wall structure of bacteria and led to leakage of intracellular molecules such as nucleic acids and proteins, which also affected the functionality of the respiratory chain complexes and hampered the process of adenosine triphosphate (ATP) synthesis. Moreover, Gao et al. [100] elaborated the anti-listeria activities of linalool against its planktonic cells and biofilms using RNA-sequence analysis. Other articles have discussed the antimicrobial efficacy of limonene [101] and linalool [102] against various strains of microorganisms. The antimicrobial activity of limonene is due to the presence of alkenyl substituent and a double bond in the molecular structure that enhances its antimicrobial activity [95]. Other authors proposed that the cell membrane may be an important site for linalool to inactivate the cell [100]. The interaction causes thickening of the Gram-positive cell wall, eventually leading to cell disruption [103]. The *S* (+) enantiomer of linalool enables it to interact with the negatively charged outer membrane of the Gram-negative cell, thus facilitating the easy entry of the compound into the intracellular space, leading to disruption [104].

Dorman et al. [105] tested 14 EO compounds against 25 strains of bacteria and reported that monoterpenoid and sesquiterpene demonstrate potent antimicrobial activity against most strains tested. In the same way, Trombetta et al. highlighted the antimicrobial potential of monoterpenes (linalyl acetate, thymol and menthol) against *E. coli* and *S. aureus* [106]. The hydroxyl group present in the compound may have contributed to its antimicrobial activity. Guimaraes et al. [31] evaluated 33 terpene compounds commonly isolated from EOs for their antimicrobial efficacy, of which only 16 compounds were reported to possess antibacterial activity. Scanning electron microscopy results revealed that individual components of EOs such as geraniol, citronellol, carveol, and terpineol altered the cellular morphology and destroyed the cell membrane. This is supported by two previous studies where similar compounds were found to be potent [105,106]. Lopez-Romero et al. [107] conducted a similar study wherein the antibacterial effect and mechanism of action of essential oil components such as carveol, carvone, citronellol, and citronellal were evaluated against

*E. coli* and *S. aureus*. Citronellol was found to be the most effective, which led to a change in the cell membrane integrity, the surface charge followed by leakage of K+ ions. In another study, two pentacyclic triterpenes, namely α-amyrin and ursolic acid, were also reported to have a disorganizing effect on the *E. coli* cell membrane [108]. Additionally, Garcia et al. [66] listed five monoterpene compounds (citronellal, citral, α-pinene, isopullegol and L-carvone) which possessed antifungal properties against three fungal strains and suggested their potential use in tropical fruit preservation. Other researchers [109,110] have investigated the antimicrobial potential of a bicyclic sesquiterpene, i.e., β-caryophyllene, against a range of microorganisms. However, they were unable to explain for the antibacterial mechanism with their study.

#### *4.2. Polyphenols*

Studies on polyphenols extracted from various fruit sources are well represented in the literature, and it is acknowledged that polyphenols possess a range of antimicrobial activities against pathogenic microbes. For example, the polyphenols in the skin extracts of Italian red grape, plum and elderberries demonstrated strong inhibitory properties against *S. aureus*, *B. cereus*, *E. coli*, *L. monocytogenes* while showing a growth-promoting effect on beneficial microbes such as *Lactobacillus rhamnosus*, *L paracasei* and *Lactobacillus plantarum* [111].

#### 4.2.1. Phenylpropenes

Although phenylpropenes account for a smaller proportion of total volatiles than terpenes and terpenoids, they have been noted to have a significant contribution to the antimicrobial activity of EOs [112]. Phenylpropenes are not only found in some fruit varieties such as apple peel [113], lemon peel [114] and grapefruit peel [115], but are also found in a wide variety of spices and herbs such as clove, star anise, sweet basil and fennel [116].

The antimicrobial potential of eugenol has been extensively investigated [117–120]. Eugenol is thought to alter the permeability of the cell membrane, followed by leakage of intracellular ATP and macromolecules such as protein and nucleic acids, ultimately leading to cell death [119]. This theory was supported by Cui et al.'s [118] study wherein eugenol permeabilized the cell membrane leading to leakage of intracellular macromolecules and enzymes such as β-galactosidase, ATP and alkaline phosphatase (AKP). Furthermore, Qian et al. [117] noted that eugenol demonstrates cell membrane permeability properties and presents potent inhibition against the biofilm formation of *K. pneumoniae* cells. Likewise, Ashrafudoulla et al. [119] reported antibiofilm activity against *Vibrio parahaemolyticus* and cell membrane damaging properties, which led to leakage of cell contents. Research by Nazzaro et al. found that isoeugenol worked in a similar way to eugenol [34]. Hyldgaard et al. [121] explained that isoeugenol formed hydrogen bonds with the lipid headgroup, thus disturbing the lipid structure and destabilizing the membrane. This mechanism of action is known as a "non-disruptive detergent-like mechanism", and the free hydroxyl group and the molecule's hydrophobic nature were considered accountable for their antimicrobial activity [122]. However, Gharib et al. [112] argued that hydrophobicity might not be the only factor contributing to the molecule's antimicrobial activity, since in his study, eugenol and isoeugenol demonstrated a fluidizing effect on the bacterial cell wall. Furthermore, Auezova et al. [123] and Gharib et al. [112] examined the mechanism of allylic (eugenol and isoeugenol) and propenylic (estragole and anethole) phenylpropenes on the cell wall of *E. coli* and *Staphylococcus epidermidis*. They demonstrated the distinctive ability of estragole and anethole to penetrate the outer membrane of *E. coli*. The antimicrobial potency is conferred by the higher lipophilic nature of estragole and anethole (log *P* values of 3.5 and 3.4, respectively) in comparison to eugenol and isoeugenol (log *P* values of 2.5 and 3.0, respectively).

Cinnamaldehyde has also demonstrated anti-biofilm activities against *S. epidermidis* [124]. Other researchers have studied the antibacterial mechanism of cinnamaldehyde against *E. coli*, *S. aureus*[125] and *Aeromonas hydrophila* [126], reporting that it caused cell membrane distortion and leakage, in addition to condensation and polarization of the cytoplasmic content. The

antibacterial activity of vanillin was studied against *Mycobacterium smegmatis,* and it was able to enhance the cell membrane permeability and alter cell membrane integrity [127].

#### 4.2.2. Flavonoids

Flavonoids are polyphenolic compounds with a benzo-γ-pyrone group and are ubiquitously found in plant cells [36]. Few examples of flavonoids are flavanones, flavan-3,4-diols, chalcones, flavan-3-ols, flavonols, flavones, isoflavones, catechins, quercetin, anthocyanidins and proanthocyanidins [128]. Recent evidence suggests that flavonoids possess antibacterial activities against plant pathogens and human pathogens. Their antimicrobial mechanism is similar to traditional drugs [33], and hence could be of importance for use as natural antimicrobial agents.

A study on catechins showed that the compounds caused oxidative damage in *E. coli* and *B. subtilis* cells, thus altering cell membrane permeability and damaging the cell membrane [129]. Moreover, Cushnie et al. [130] also reported that catechins were responsible for potassium ion leakage in methicillin-resistant *S. aureus* (MRSA), which is the primary signal of membrane damage, and Tsuchiya et al. [131] reported that sophoraflavanone G significantly affected the membrane fluidity of the bacterial cells.

#### **5. Application of Essential Oils in Food Products**

#### *Preservation*

Traditional food preservation methods include chilling, frozen storage, drying, salting, smoking and fermentation [132]. However, consumers have questioned techniques such as fermentation, brining, and salting, due to the increasing demand for reduced-salt foods [133]. The meat industries utilize chemical preservatives such as nitrate salt, sulfites, chlorides to inhibit the growth of foodborne pathogens. These compounds have been associated with carcinogenic effects and other health complications [133]. Hence, the options available to substitute chemical preservatives with natural compounds have attracted increased interest in recent years. Lucera et al. [134], in her review, outlined some natural preservatives of animal origin, (lactoferrin, lysozyme); bacteriocin from microbes (natamycin, nisin); natural polymers (chitosan); organic acids (citric and propionic acid); EOs and extracts derived from plants. In this context, EOs are attracting considerable attention due to their application as a natural bio-preservative and inhibitor in food matrices or food products. At present, the investigations have focused primarily on EOs from herbs and spices. There is limited research on fruit peel EOs. So, the discussion is widened here to cover the food applications of all plant-derived EOs. Some publications have investigated the potential contributions EOs/extracts to extend the shelf-life and to inhibit the growth of pathogens in fresh-cut vegetable mixtures [135], lettuce, purslane [136], fruit juices [137], ready to eat meat [138], chicken nuggets [76] and breast [139], minced beef [140,141] and turkey [142]. A literature review [141] published in 2018 included 2473 publications since 1990 on the antimicrobial activity of EOs. Many of these publications investigated the application of EO's on food products, including 657 papers on fruits, 403 on vegetables, 415 on fish products, 410 on meat products, 216 on milk and dairy products, and 97 on bread and baked foods [143]. Other recent reviews have discussed the application of rosemary extract in meat [144], the synergistic effect of EO in seafood preservation [145], application of EO in active packaging [146] and as a food preservative [147]. The following section includes the recent history of EO by restricting the citations to the last 5–6 years to provide the readers with an update on EOs and their application in the food matrix (Table 2).

As consumers have gained greater awareness on issues related to health, processing and food additives, demand for natural and minimally processed food has soared. However, maintaining the freshness of fruits and fresh-cut vegetables for extended periods has been challenging. Spraying, dipping, coating, and impregnation are ways EOs can be applied to fruits and vegetables for maintaining shelf-life [134]. Some recent examples of these approaches are discussed here. He et al. [148] evaluated the effects of dipping cherry tomatoes in thyme EO nanoemulsion (TEON) against *E. coli* O157:H7 and the effect of

TEON in combination with ultrasound treatment. Their study showed that TEON alone could effectively inhibit the growth of *E. coli* O157:H7 on the surface of cherry tomatoes, and there was a substantial synergistic effect of the combined treatment. Kang et al. [149] found that freshly cut red mustard leaves, when washed with cinnamon leaf EO nanoemulsion, reduced the count of *E. coli*, *L. monocytogenes*, *S. enterica* serovar Typhimurium by more than one log. Another study conducted by the same author showed that washing with cinnamon leaf EO nanoemulsion improved physical detachment and inhibited both *L. monocytogenes* and *E. coli* O157:H7 on kale leaves [150]. Both studies did not show any adverse changes in the quality attributes of mustard [149] and kale leaves [150]. The lettuce leaves examined during 7-day storage periods showed a reduction in *E. coli* O157:H7 population when rinsed with a combination of carvacrol/eugenol and thymol/eugenol when compared to the control (water rinse). However, the treatments had adverse effects on the sensory analyses [151]. In contrast, a combination of Spanish origanum oil and Spanish marjoram oil successfully inhibited *L. monocytogenes* from a mixture of fresh-cut vegetables without showing any adverse sensory attributes [135]. A recent study elucidated that *Litsea cubeba* EO added to bitter gourd, cucumber, carrot and spinach juices at MIC concentration decreased the counts of *E. coli* O157:H7 by 99.1%, 99.92%, 99.94%, 99.96%, respectively [152]. Krogsgård Nielsen et al. [153] tested the inhibitory potential of isoeugenol and encapsulated isoeugenol against *L. monocytogenes*, *S. aureus*, *Leuconostoc mesenteroides*, *P. fluorescens* in carrot juice. Contrary to expectations, their study did not find a significant difference in the inhibitory activity of encapsulated and non-encapsulated isoeugenol.

Besides fruits and vegetables, much work on the antimicrobial potential of EO was studied in meat products especially beef and beef products [154–158]. Pistachio EO [155] and *Melaleuca alternifolia* (tea tree) EO [157] reduced the total viable and total *L. monocytogenes* counts in ground beef. The efficiency of 5% and 10% clove EO on the inactivation of *L. monocytogenes* in ground beef at refrigeration (8 ◦C), chilling (0 ◦C) and freezing (18 ◦C) temperatures was investigated by Khaleque et al. [158]. They observed that 10% clove EO was a lethal concentration to inactivate *L. monocytogenes* irrespective of temperature conditions, but 5% clove EO was ineffective at inactivating the pathogen [158]. Similarly, Yoo et al. [154] found that 0.5%, 1.0% and 1.5% clove EO did not significantly reduce the count of *E. coli* O157:H7 and *S. aureus* in beef jerkies. However, their study took an additional step and demonstrated that the combined effect of clove EO with encapsulated atmospheric pressure plasma had a bactericidal effect on both pathogens. Likewise, a study conducted by Lin et al. [156] pointed out the synergistic effect of chrysanthemum EO incorporated into chitosan nanofibers which inhibited *L. monocytogenes* in beef at a rate of 99.9%.

A triple combination of thyme/cinnamon/clove EO in the food matrix was first applied experimentally by Chaichi et al. [159]. The triple combination at FIC of 0.3, 0.39.0.43 had a bacteriostatic effect on *P. fluorescens* inoculated in chicken breast meat, while a triple combination at higher concentration (200 mg/kg) had an instant bactericidal effect. Thyme EO effectively inhibited *P. aeruginosa*, *E. coli* and *S. enterica* serovar Typhimurium in ground beef [160]. A recent study by Kazemeini et al. [161] prepared edible coatings of alginate containing *Trachyspermum ammi* EO (TAEO) as nanoemulsion to control the growth of *L. monocytogenes* in turkey fillets. The turkey fillets were coated with the emulsion and stored at 4 ◦C for 12 days. They observed the highest reduction of *L. monocytogenes* numbers in turkey fillets treated with 3% alginate containing 0.5% and 1% TAEO compared to noncoated samples. Other research articles have reported that EO nano emulsions effectively inhibited pathogens in rainbow trout fillet [162] and chicken breast fillets [163]. Apart from fruits, vegetables and meat products, the application of EO has also been evaluated on bakery [164,165] and dairy products [166].

Although several authors [152,157,165,166] have claimed successful testing for the application of EOs in different food systems, their approach has not escaped criticism. Santos et al. [167] emphasized the use of MBC concentration rather than MIC concentration in the food matrix to ensure a complete inhibition. These authors [167] questioned the usefulness of EOs in food systems because various factors such as environmental condition, age and

cultivar of the plant, time harvested, extract composition and extraction method may impact the antimicrobial activity of the EO. All the above factors might challenge the rationale of applying EOs at a commercial level. Moreover, it is known that fat and protein present in food can solubilize or bind to phenolic compounds in EO, thus reducing its antimicrobial efficacy [157]. This view was supported by Khaleque et al. [158], who analyzed the effect of cinnamon EO at a higher concentration (2.5 and 5.0%) against *L. monocytogenes* in ground beef and found that cinnamon EO was unsuccessful in inactivating *L. monocytogenes* in ground beef. They also reported adverse organoleptic impacts upon using higher concentrations of EOs. In a study by Lages et al. [168], thyme EO combined with beet juice powder failed to give a desirable effect in reducing coagulase-positive *Staphylococcus* in meat sausage. It was recommended that combining half of the suggested dosage of chemical preservatives such as nitrites with EO could be feasible. Despite the question regarding the suitability of EO in minimally processed food products [167], only a few studies did not show effective inhibition by EOs of foodborne pathogens. In contrast, many studies have demonstrated the successful replacement of synthetic preservatives with EO in different food systems [165,166,169]. Since Santos et al. [167] did not use EOs in minimally processed food products, their assumptions need further validation. Their paper would have been more convincing if the authors had used food matrices to prove their hypothesis. There is evidence that EOs exhibit antimicrobial properties, therefore, their ability to be used as a natural preservative on an industrial scale needs further rigorous evaluation.

**Table 2.** Overview of recent studies on antimicrobial activity of different essential oils (EOs) in the food matrix.


#### **6. Food Regulations on Applications of Essential Oils**

The European Commission has documented a variety of EO compounds as approved flavour additives in different types of food products. In 2008, the European Commission released a list of approved compounds which is updated regularly. Some of the registered flavoring compounds that pose no risk to human health are limonene, linalool, β-caryophyllene, pinene, thymol, carvacrol, carvone, eugenol, isoeugenol, vanillin, citral, citronellal, cinnamaldehyde, menthol and lavandulol [181]. Moreover, the Food and Drug Administration (FDA) of the United States also recognizes these compounds as GRAS. Crude EOs such as mustard, oregano, clove, cinnamon, nutmeg, thyme, basil, rosemary and lavender are recognized as GRAS. The regulatory limits on acceptable daily intake on EO compounds and EOs are in place to govern their use in food products [7]. Despite the regulatory limits, EOs might cause allergic reactions and ingesting high doses of EOs or topical applications of EOs for a long period have been associated with severe health problems, such as oral toxicity and dermatitis [182]. Therefore, it is crucial to attain a fine balance between toxicity and effective dose in food products.

#### **7. Conclusions and Future Prospects**

Evidence from in vitro and in situ studies suggests that EOs possess good antibacterial activity against a wide range of foodborne pathogens. This review has evaluated studies on EOs that have the potential to act as natural preservatives in food products, due to their antioxidant and antimicrobial properties [183,184]. The potential of all plant-derived EOs, not just fruit peel EOs, has been evaluated for use as a preservative in foods. However, their application in food products have been restricted at an industrial scale as high doses are required to attain good antimicrobial activity, and the quantity, source and active composition profile of the EO to be used in food has not been optimized. In addition, components of the foods, such as fat [185], starch [186] and protein [187], may bind to the active compounds in EOs and reduce their efficacy. The volatile compounds in EOs may also produce undesirable chemical compounds by interacting with other food components such as proteins. To validate the use of EOs at an industrial level, the evaluation of these aspects is of paramount importance.

Firstly, high concentrations of EO in food have shown unappealing sensory attributes. However, this problem may be addressed by evaluating an effective synergistic/additive combination of EOs or a combination of EOs with other food preservation techniques such as temperature, irradiation, and pulse-electric field to reduce the required dosage of EO for the inhibition of pathogens. Another plausible solution for minimizing the interaction of EO compounds with food components such as fat, starch and proteins is by encapsulating the EO in an appropriate biodegradable material (e.g., chitosan), which might ensure controlled release without altering its biological activity. Secondly, a detailed understanding of how EOs work (the mechanism of action) will provide insights into the application of EO in the food industry to combat the proliferation of food-borne pathogens. To further study the mechanism of action, proteomic and transcriptomic analyses are needed to understand the pathways targeted by the EO compounds. The transition of in vitro experiments to in vivo trials to evaluate the efficacy of EOs has always posed an added challenge. Another future opportunity lies in the potential effects of EOs on immunity and gut health. Recent research reported that a combination of oregano extract with peppermint and thyme EO supported the growth of probiotic bacteria and positively affected the gut's microbial composition [188]. Further research regarding the role of EO on the gut microbiome would be worth exploring.

**Author Contributions:** Conceptualization, M.A., S.S. and S.Y.Q.; writing—original draft preparation, M.A.; writing—review and editing, M.A., S.S., K.H., C.A.B., S.Y.Q.; supervision, S.S., K.H., S.Y.Q.; project administration, S.Y.Q.; funding acquisition, S.S., S.Y.Q. All authors have read and agreed to the published version of the manuscript.

**Funding:** This research is partially funded by The University of Auckland (Press Account Number-9448-UOA-MANG207) and Food and Health Programme Seed Grant (4200-UOA-48422-A8AN).

**Informed Consent Statement:** Not applicable.

**Acknowledgments:** The authors would like to thank The University of Auckland for the Doctoral Scholarship awarded to the first author and Food and Health Programme Seed Grant.

**Conflicts of Interest:** The authors declare no conflict of interest.

#### **References**


## *Article* **In Vitro Potential of Clary Sage and Coriander Essential Oils as Crop Protection and Post-Harvest Decay Control Products**

**Robin Raveau 1, Joël Fontaine 1, Abir Soltani 2, Jouda Mediouni Ben Jemâa 2, Frédéric Laruelle <sup>1</sup> and Anissa Lounès-Hadj Sahraoui 1,\***


**Abstract:** Owing to their various application fields and biological properties, natural products and essential oils (EO) in particular are nowadays attracting more attention as alternative methods to control plant pathogens and pests, weeds, and for post-harvest applications. Additionally, to overcome EO stability issues and low persistence of effects, EO encapsulation in *β*-cyclodextrin (*β*-CD) could represent a promising avenue. Thus, in this work, the EO distilled from two aromatic plants (*Salvia sclarea* L. and *Coriandrum sativum* L.) have been evaluated in vitro for their antifungal, herbicidal and insecticidal activities, against major plant pathogens and pests of agronomical importance. Both plants were grown on unpolluted and trace-element-polluted soils, so as to investigate the effect of the soil pollution on the EO compositions and biological effects. These EO are rich in oxygenated monoterpenes (clary sage and coriander seeds EO), or aliphatic aldehydes (coriander aerial parts EO), and were unaltered by the soil pollution. The tested EO successfully inhibited the growth of two phytopathogenic fungi, *Zymoseptoria tritici* and *Fusarium culmorum*, displaying IC50 ranging from 0.46 to 2.08 g L−1, while also exerting anti-germinative, herbicidal, repellent and fumigant effects. However, no improvement of the EO biological effects was observed in the presence of *β*-CD, under these in vitro experimental conditions. Among the tested EO, the one from aerial parts of coriander displayed the most significant antifungal and herbicidal effects, while the three of them exerted valuable broad-range insecticidal effects. As a whole, these findings suggest that EO produced on polluted areas can be of great interest to the agricultural area, given their faithful chemical compositions and valuable biological effects.

**Keywords:** essential oils; aromatic plants; antifungal; anti-germinative; herbicidal; insecticidal

#### **1. Introduction**

Historically used in traditional medicine, essential oils (EO) are these days raising great interest, owing to their diverse application fields [1–3]. Made of a mixture of volatile compounds, up to 100, and synthesized by all aromatic plant parts as secondary metabolites [4,5], EO were recently outlined for their interest in the preservation of food quality and flavor [6]. They also have received increasing attention as potential alternatives to commercial pesticides in crop protection, given their promising biological properties against plant pathogens, pests and weeds [1,5].

Among fungal phytopathogens, *Fusarium culmorum* and *Zymoseptoria tritici* are of major importance, responsible for *Septoria tritici* blotch and *Fusarium* head blight on cereals, respectively, whose damage on host plant are considerable, and may cause yield losses up to 50% [7,8]. *Fusarium* spp. are also known to produce a wide range of mycotoxins, secondary metabolites that may be highly toxic to human and animal health [8,9]. The

**Citation:** Raveau, R.; Fontaine, J.; Soltani, A.; Mediouni Ben Jemâa, J.; Laruelle, F.; Lounès-Hadj Sahraoui, A. In Vitro Potential of Clary Sage and Coriander Essential Oils as Crop Protection and Post-Harvest Decay Control Products. *Foods* **2022**, *11*, 312. https://doi.org/10.3390/ foods11030312

Academic Editors: Evaristo Ballesteros, Lisa Pilkington and Siew-Young Quek

Received: 14 December 2021 Accepted: 20 January 2022 Published: 24 January 2022

**Publisher's Note:** MDPI stays neutral with regard to jurisdictional claims in published maps and institutional affiliations.

**Copyright:** © 2022 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https:// creativecommons.org/licenses/by/ 4.0/).

control of both *F. culmorum* and *Z. tritici* is mostly achieved through the use of triazole fungicides [8]. However, resistance levels to triazoles have significantly increased since their marketing authorization, and hence compromise their reliability [7,8]. Similarly, insect pests cause significant losses in terms of quantity and quality of the products, in field or during postharvest storage. This is the case for the silverleaf whitefly, *Bemisia tabaci* Genn. (Hemiptera: Aleyrodidae), the lesser grain borer, *Rhyzopertha dominica* F. (Coleoptera: Bostrychidae), or the Mediterranean flour moth, *Ephestia kuehniella* Zeller (Lepidoptera: Pyralidae), who are recognized worldwide as some of the most destructive pests on several economically important crops [10–12]. Their control is also mainly achieved through the use of chemical insecticides, displaying resistance phenomena in pest populations [10,12]. Notably, *B. tabaci* has been identified as resistant to a wide number of systemic insecticides, such as organophosphates, synthetic pyrethroids or neonicotinoids [12,13]. Moreover, their use may lead to detrimental effects on beneficial insects, as well [10,14]. From a wider perspective, the excessive and inappropriate use of pesticides is controversial, because of their noxious impact on both environmental and human health [5,15]. As part of an integrated pest management system, the use of natural products, including EO, considered as biocontrol tools, is then greatly encouraged [16,17], especially in the current context, where regulatory restrictions lead to the withdrawal of several commercial products [15,18]. Owing to their relatively low toxicity for humans and animals, EO are moreover registered as "Generally Recognized As Safe (GRAS)" products by the Food and Drug Administration, and regarded as less harmful to both environmental and human health, in comparison with commercial pesticides [19,20]. They also receive increasing public support, considered as eco-friendly products [4,20].

Nevertheless, EO are highly volatile, which could be of great interest for the reduction of the residues, as well as for postharvest applications [4]. Yet, this appears prohibitive for field applications, due to stability issues, and a short persistency of the biological effects over time [5]. To tackle these problems, the use of appropriate EO formulations could offer a promising tool [4,5,21]. In particular, the use of *β*-cyclodextrins (*β*-CD), cyclic oligosaccharides able to encapsulate hydrophobic compounds in aqueous solutions, and cited in the Food and Drug Administration's list of Inactive Pharmaceutical Ingredients, may be of great interest to avoid degradation, while maintaining the products' efficiency [19,22].

Within the plant kingdom, Lamiaceae are one of the biggest flowering plant families, comprising a wide number of valuable aromatic species [23,24]. Among them, clary sage (*Salvia sclarea* L.), a biennial aromatic plant species grown all around the world for its highvalue EO, has been long known for its use in the perfumery and cosmetic sectors, but also for its applications in medicine [20,25,26]. Coriander (*Coriandrum sativum* L.), another aromatic plant belonging to the Apiaceae family, is an annual herbaceous plant grown all over the world, for the consumption of its green leaves and its seeds, as a spice, or for EO production [27,28]. Both clary sage and coriander EO have drawn attention given their biological effects, in particular antifungal properties against the phytopathogenic fungi *Phoma* spp., *Alternaria alternata* or yeasts, such as *Candida* spp. [26,28,29], and insecticidal effects against the whitefly *Trialeurodes vaporariorum* [26], or several coleoptera species [30–32].

In addition to the production of EO, aromatic plants may also appear as valuable choices within the framework of phytomanagement approaches [33,34]. There is in fact an urgent need to address the issue bound to the presence in ecosystems of inorganic pollutants, such as trace elements (TE), whose pollution extent may exceed 5 million sites worldwide [35]. They are posing serious threats to environmental and human health, as they are non-degradable, tend to accumulate in living organisms, and exhibit toxic effects when their concentration exceeds a certain bearable threshold [33,36]. With the emergence of phytotechnologies as rising tools to mitigate TE-polluted spaces, the capacity of some aromatic plants to tolerate elevated concentrations of TE could be particularly valuable [33,34,37]. Their cultivation on marginal lands, unsuitable for food production, tends to minimize the risk of food-chain contamination, while avoiding competition with feeding agriculture [33,34]. Moreover, one of the major drawbacks to phytotechnologies lies

in most cases in the lack of economic profitability [38]. In that regard, the use of aromatic plants grown on polluted soils for the production of EO, which are biosourced products bearing a high added-value, and displaying a content free of TE, could help towards the obtention of an economic profit [34,39,40].

Nonetheless, EO composition is strongly influenced by environmental factors, namely geographical location, sunlight, climatic conditions and soil properties, including the presence of pollutants [26,41]. Notably, the presence of TE in soil could result in modified EO yield, and altered composition and quality [42–44].

Thus, the aim of this work was first to investigate the potential influence of the soil pollution by TE on the chemical composition of the EO distilled from clary sage inflorescences, and from both aerial parts and seeds of coriander, and then on their biological properties in the presence or in the absence of *β*-CD. The antifungal, anti-germinative and herbicidal potential, as well as the insecticidal activity of the EO, were evaluated against two major phytopathogenic fungi, namely, *F. culmorum* and *Z. tritici*, two plant species commonly used for chemicals' herbicidal assessments, namely *Lactuca sativa* L. and *Lolium perenne* L., and adults of three insect species, namely, *B. tabaci*, *R. dominica* and *E. kuehniella*, so as to explore their potential use as crop protection products.

#### **2. Materials and Methods**

#### *2.1. Essential Oils*

The EO tested in this study were acquired by steam distillation of coriander (*Coriandrum sativum* L.), and clary sage (*Salvia sclarea* L.), two aromatic plant species grown in situ on two experimental sites: a TE-polluted one, displaying elevated amounts of TE (7, 394 and 443 ppm of Cd, Pb, and Zn, respectively), and an unpolluted one. Their full description, as well as plant physiological data, are available in [34,45]. The distillation of the harvested aromatic plant biomass was realized in collaboration with a private EOdistiller. The steam distillation (14 m<sup>3</sup> distillation unit—saturated water steam, 0.3 bar) was carried out over a three-hour cycle, until no more EO was recovered, under the previously described experimental conditions [20]. Aerial parts of coriander and seeds were harvested at full blossoming, or at seed maturity, respectively, for their distillation. In the same way, clary sage distillation was performed using harvested inflorescences at full blossoming, during its second year of cultivation, when the highest yields are expected [34,46]. EO were stored at 4 ◦C, in tightly closed brown glass vials, and under modified nitrogen protective atmosphere, until their use.

#### *2.2. Determination of the EO Chemical Composition*

EO samples were first diluted in ethyl acetate (ratio 1:200 (*v*/*v*)), and then analyzed by electron ionization gas chromatography–mass spectrometry (Shimadzu QP 2010 Ultra), according to the method previously described [20]. Briefly, volatile EO components were separated on a ZB-5MS (Phenomenex—5%-phenyl-arylene/95% dimethylpolysiloxane— 10.0 m × 0.10 mm × 0.10 μm) capillary column. The EO solution was then injected in a split mode (0.2 μL; split ratio 1:10). Helium was used as a carrier gas to operate the system, at a constant linear velocity (60 cm s−1). The column temperature was held for 2 min at 60 ◦C, then programmed to linearly increase to 280 ◦C, at a constant rate of 40 ◦C min−1, and remained at 280 ◦C for 1 min.

Mass spectra were recorded within a mass range of 35.0 to 350 (*m*/*z*), at an interface temperature of 280 ◦C, and an ionisation energy of 70 eV. The EO components were identified by comparison of their retention indices relative to (C8–C30) n-alkanes (Kovats indices), and their obtained mass spectra, with those listed in the NIST (National Institute of Standards and Technology, Gaithersburg, MD, USA), and Wiley 275 computer libraries, as well as those found in the literature [26,27,41,47,48]. Relative percentages of oil constituents were measured from the GC peak areas.

#### *2.3. Biological Activities of the EO* 2.3.1. Antifungal Activity Phytopathogenic Fungal Strains

Antifungal activities of the different EO were evaluated against two major phytopathogenic fungi—*Fusarium culmorum* and *Zymoseptoria tritici*—by using in vitro assays. *F. culmorum* strain was maintained on PDA (Potato-Dextrose-Agar, Condalab, Spain) medium. The hemibiotrophic fungus *Z. tritici* (strain T02596) was conserved at −80 ◦C in cryopreservation tubes. Five days prior to the assays, the fungus was cultivated on a PDA medium, in order to produce spores [49].

#### Determination of In Vitro Antifungal Activity against *F. culmorum*

Essential oils' antifungal activity against *F. culmorum* was evaluated by using an in vitro direct contact assay. It was evaluated according to the method previously described [19,50], with slight modifications. A PDA medium (39 g L−1) was first prepared, complemented with 1% (*v*/*v*) DMSO (Thermo Fisher Scientific, Illkirch, France), in which EO were mixed at 50 ◦C, so as to obtain a final scale of 5 EO concentrations, ranging from 0.005 to 1.0% (*v*/*v*) of EO in the medium. Discs of *F. culmorum* (0.9 cm) were then cut out from the periphery of a 7-day-old fungal colony, and placed at the center of a 9 cm Petri dish containing the PDA medium complemented with EO. The assay was carried out in the absence, and the presence of *β*-CD in the medium, at 10 mM. Mycelium radial growth was measured after a seven-day incubation (20 ± 1 ◦C). The inhibition rate was calculated following Equation (1):

$$\text{Inhibition rate } (\%) = \frac{\chi\_0 - \chi\_i}{\chi\_0} \times 100 \tag{1}$$

where X0 and Xi stand for the average diameter of the fungal colony in control and in treatment, respectively.

Aqueous solutions of DMSO 1% (*v*/*v*) or *β*-CD (10 mM), as well as a marketed fungicide, Aviator XPro (prothioconazole—150 g L−<sup>1</sup> and bixafen—75 g L<sup>−</sup>1, Stolz, Wailly-Beaucamp, France), were tested as negative and positive controls, respectively. The positive control was evaluated with concentrations ranging from 5 × <sup>10</sup>−<sup>5</sup> to 0.5% (*v*/*v*). Analyses were led in triplicates for each condition.

#### Determination of In Vitro Antifungal Activity against *Z. tritici*

Essential oils' antifungal activity against *Z. trici* was evaluated by using an in vitro microplate assay, adapted from Fungicide Resistance Committee methods, and similar to the one developed by [4], with a range of eight concentrations (from 0 to 0.8% of EO in the medium) for each and all EO. Briefly, spores of *Z. trici* were collected and placed in a glucose-peptone suspension. The microplates were then inoculated with 60 μL of the calibrated pathogen suspension (2 × 105 spores mL−1). The microplates were then incubated under agitation at 110 rpm (20 ± 1 ◦C), in darkness, with an incubation time of 6 days, determined according to the pathogen's optimal growth time. The evaluation of its growth was carried out using a spectrometer (620 nm). For each EO concentration, eight replicate wells were used. Additionally, each assay was carried out in triplicate to compare the products. Controls include four non-inoculated wells per EO concentration. Additionally, EO's effects were assessed in the presence and in the absence of *β*-CD in the culture medium (10 mM). These natural products were compared to a homologated and marketed product: Aviator XPro (prothioconazole—150 g L−<sup>1</sup> and bixafen—75 g L<sup>−</sup>1), within the same range of concentrations.

#### Determination of In Vitro Antifungal Properties of the EO

The half-maximal inhibitory concentration (IC50) of EO (expressed in g L<sup>−</sup>1), required to obtain a fungal pathogens' growth inhibition of 50%, was calculated for all in vitro assays. A graphical interpolation, complemented with a statistical analysis based on a nonlinear regression, were used to calculate the IC50 value [51]. The IC50 of each of the tested EO was

also classified as either fungicidal or fungistatic, considering its effects. The fungistatic or fungicidal nature of EO was tested by observing growth revival of the inhibited mycelial disc, following its transfer on EO-free PDA medium: no mycelial return to growth defined fungicidal effect, whereas fungistatic effect was characterized by a fungal regrowth capacity on the EO-free medium.

#### 2.3.2. Anti-Germinative and Herbicidal Activities

Inhibitory effects on seedlings' emergence and growth of the different EO, were assessed against two plant species, *Lolium perenne* L. (monocotyledon) and *Lactuca sativa* L. (dicotyledon), commonly used for chemicals' herbicidal assessments [52], and listed in the OECD guidelines (2003). An in vitro method was adapted [53–55]. EO aqueous solutions (DMSO 1% (*v*/*v*)) were prepared, in the presence or in the absence of *β*-CD (10 mM), and mixed in an agar non-complemented medium (50 ◦C), then poured into square Petri dishes (120 × 120 mm). EO concentrations used ranged from 5 × <sup>10</sup>−<sup>4</sup> up to 0.5% (*v*/*v*). Seeds were then placed on the solidified agar medium, in sealed Petri dishes, and incubated for 8 days on a day/night cycle, with a 16 h photoperiod (20 ± 1 ◦C), and an obscurity period of 8 h (16 ± <sup>1</sup> ◦C). Glyphosate (isopropylamine salt—360 g L−1) was used as a positive control within the same concentrations' range as the tested EO, while aqueous solutions of DMSO 1% (*v*/*v*) or *β*-CD (10 mM) were tested as negative ones.

After the incubation period, germination rates were evaluated by counting germinated seeds, while root elongation was assessed through an imaging software (ImageJ), by measuring root length [55,56]. The analyses were led in triplicates. Graphical interpolation, complemented with a statistical analysis [51], was used to calculate the IC50 values, regarding both germination and root elongation parameters.

#### 2.3.3. Insecticidal Activities

#### Insect Individuals

*Bemisia tabaci* adults were collected from a tomato (*Solanum lycopesicum* L.) greenhouse. Adults of *Rhyzoperta dominica* were kept on whole wheat, while *Ephestia kuehniella* adults were reared on wheat flour. Insects were maintained at 25 ± 1 ◦C and 65 ± 5% relative humidity. Both female and male adult insects were used for bioassays.

#### Repellency Bioassay

Repellency bioassays for the different tested EO were carried out according to the experimental methods previously described [57], at 25 ± 1 ◦C and 65 ± 5% relative humidity. For that, Whatman filter papers (8 cm diameter) were cut in half. Test solutions were prepared by dissolving 0.4, 1 and 2.5 μL of EO in 1 mL acetone. Each solution was applied to half of the filter paper discs, using a micropipette. The other half of the filter paper was treated with acetone only, as a control. The treated and control half discs were then air-dried under a fan, in order to evaporate the totality of the solvent. Treated and untreated halves were attached to their opposites, using adhesive tape, and placed in Petri dishes. Twenty male and female adult insects were then released at the center of each filter paper disc. Parafilm was used to seal the dishes. Three replicates were used for each concentration, and for each EO. The number of insects in the treated and untreated halves was recorded after 1, 3, 5 and 24 h. Three trials were made for each concentration, and tested by applying the χ<sup>2</sup> test for homogeneity ratio (1:1). Numbers of *R. dominica*, *E. kuehniela* and *B. tabaci* adults present on both treated and untreated portions of the experimental paper halves were recorded at different times of exposure. Percentage Repellency (PR) was calculated according to the following formula [58]:

$$\text{Percentage repellerncy } (\%) = [\text{Nc} - \text{Nt } (\text{Nc} + \text{Nt})] \times 100\tag{2}$$

where Nc and Nt stand for the number of insects on the untreated area and on the treated area, respectively, after various exposure times.

The data were also expressed as RC50 values, corresponding to the concentration that repelled 50% of the exposed insects. Three replicates were observed for each EO, at the different exposure times. Replicates were also used for each EO concentration. Comparison was made between the mean number of treated and untreated insects.

#### Fumigation Bioassay

The toxicity of the three EO by fumigation was tested in Plexiglas bottles of 38 mL, in which 10 *R. dominica*, *E. kuehniella* or *B. tabaci* adults were released. Filter paper was cut into 2 cm in diameter pieces, and impregnated with the different EO concentrations 9.09, 22.72 and 56.81 μL L−1. Caps were tightly screwed on the vials. Mortality was recorded after 2, 4, 6, 24, 36, 48, 72, 96, 120 and 144 h from the start of exposure. Three replicates were done for each EO, and for each concentration. The control did not show any mortality. Results were expressed as median lethal time (LT50), time after which half of a sample population has died, and median lethal concentration (LC50), the chemical concentration that results in the death of 50% of a sample population.

#### *2.4. Statistical Analyses*

Statistical analyses were performed using XLSTAT 2018.1.1 (Adinsoft, Paris, France) software and R 3.6.1 [59]. Before any statistical analysis, Shapiro–Wilk and Bartlett tests were performed to verify normality and homoscedasticity assumptions, respectively. When necessary, non-normal data were "square-root" or "log10" transformed.

Regarding antifungal and herbicidal properties, IC50 values resulted from non-linear regression analyses from triplicate assays, and were expressed as mean values and standard deviation (mean ± SD). The comparison of IC50 values was carried out using two-way analysis of variance (ANOVA), complemented with a *post-hoc* Tukey-HSD (Honestly Significant Difference) test.

For the insecticidal activity, the statistical analysis was performed using SPSS statistical software, version 20.0. When necessary, data were transformed by common logarithm or exponential, to meet the normality assumptions. All obtained values were the mean of three replications, and were expressed as means ± standard error. For the repellent activity, differences between each EO were tested by one-way ANOVA, followed by Duncan test. From the bioassays data, a Probit analysis was further conducted to estimate RC50 on one side, and LC50 and LT50 values on the other.

#### **3. Results**

#### *3.1. Determination of the EO Chemical Composition*

The GC-MS profiles of the three tested EO, from aerial parts or seeds of coriander, and from clary sage inflorescences, are listed in Table 1. In the EO distilled from aerial parts of coriander, 15 compounds were identified, most of which are aliphatic aldehydes, along with some oxygenated monoterpenes. In contrast, coriander seeds and clary sage EO, in which 11 and 22 compounds were identified, respectively, were particularly rich in terpene compounds (Table 1).

The chromatographic profile showed that linalool was the only compound present in the three EO, from different plants, and plant parts. It was particularly abundant in the EO distilled from coriander seeds, with relative proportions ranging between 76.2 and 80.6%, for the EO distilled from the biomass grown on unpolluted and polluted sites, respectively (Table 1). *γ*-terpinene also represented a significant proportion of coriander seeds EO (from 7.8 to 8.7%).

*Foods* **2022**, *11*, 312

**Table 1.** Chemical composition of the EO from aerial parts or seeds of coriander, and from sage inflorescences, grown on unpolluted or TE-polluted sites. Data are relative percentages of EO compounds, expressed as means ± SD (n = 3). For a same plant part, means followed by an asterisk "\*" are significantly different, between polluted and unpolluted conditions, by one-way ANOVA test (α = 0.05).


35

In the EO distilled from coriander aerial parts, the other compounds identified with a proportion higher than 5% were decanal (7.5%) and (Z)-2-decenal (44 to 49.1%), whereas linalyl acetate and germacrene D were the other major compounds in clary sage EO (varying between 52.2 and 62.7%, and between 7.1 and 15.6%, respectively).

It should also be noted that, even though the overall composition was highly similar for a same plant species and part, the balance between several compounds varied slightly, between EO distilled from aromatic plants grown on unpolluted and TE-polluted sites. It is notably the case for linalool, linalyl acetate, and germacrene-D in EO from sage inflorescences, for undecanal, 2-dodecenal, or 2-tridecenal in the EO distilled from aerial parts of coriander, and for *α*-pinene, linalool or *β*-farnesene in the EO from coriander seeds (Table 1).

#### *3.2. EO Antifungal Activity*

#### 3.2.1. In Vitro Antifungal Activity against *F. culmorum*

Our results have shown that the EO from both coriander and sage presented antifungal properties against the phytopathogenic fungus, *F. culmorum*. The results obtained with coriander aerial parts' EO range of concentrations are provided in Figure S1. It was characterized as fungistatic, since fungal regrowth was observed when the discs containing the fungus were transferred on an EO-free medium. IC50 obtained for the direct contact assay ranged from 0.46 to 2.08 g L−1, with no significant difference observed between the EO from a same plant, either aerial parts of coriander, seeds of coriander or clary sage, but cultivated under the different experimental conditions (polluted or unpolluted site—Figure 1). Additionally, EO distilled from either aerial parts or seeds of coriander demonstrated a higher efficiency (displaying lower IC50) than those from sage, with IC50 ranging from 0.46 to 0.53 g L−1, and from 1.47 to 2.08 g L−1, for coriander and sage, respectively. On another note, no significant improvement was observed in the presence of *β*-CD, displaying either similar or higher IC50, in comparison with the *β*-CD-free condition. In comparison with the positive control, all the obtained IC50 were significantly higher, up to 104 times (Figure 1).

**Figure 1.** EO's IC50 values (g L<sup>−</sup>1) arising from the antifungal in vitro direct contact bioassay against *F. culmorum*. Values are means ± SD (n = 3). Means followed by the same lowercase letter are not significantly different, by two-way ANOVA comparison (α = 0.05). The positive control (Aviator XPro) value is represented by the black dotted line. All conditions are different from the positive control. IC50: half-maximum inhibitory concentration; CD: cyclodextrins.

#### 3.2.2. In Vitro Antifungal Activity against *Z. tritici*

Our results from the microplate assay against *Z. tritici* have shown that the EO from coriander aerial parts or seeds, and sage, presented fungistatic properties against this phytopathogenic fungus. The results obtained for the in vitro microplate bioassay with the three tested EO are provided in Figure S2. IC50 obtained for this assay ranged from 0.001 to 0.08 g L<sup>−</sup>1, with no significant difference observed, whatever the EO and the plant part it is distilled from (aerial parts or seeds of coriander, sage inflorescences) and the experimental cultivation conditions (polluted or unpolluted sites—Figure 2). No significant improvement was observed in the presence of *β*-CD, displaying either similar or higher IC50 in comparison with the *β*-CD-free condition. In comparison with the positive control, the IC50 obtained for both coriander EO (aerials parts or seeds) and sage were similar in our experimental conditions.

**Figure 2.** EO's IC50 values (g L−1) arising from the antifungal in vitro microplate bioassay against *Z. tritici*. Values are means ± SD (n = 3). Means followed by the same lowercase letter are not significantly different, by two-way ANOVA comparison (α = 0.05). The positive control (Aviator XPro) value is represented by the black dotted line. All conditions different from the positive control are displayed with an asterisk "\*". IC50: half-maximum inhibitory concentration; CD: cyclodextrins.

#### *3.3. EO Anti-Germinative and Herbicidal Activities*

#### 3.3.1. Seedlings' Emergence Inhibition Bioassay

Our results have shown that all tested EO exerted a significant anti-germinative effect on both lettuce and rye-grass. The obtained IC50 ranged from 0.05 to 6.22 g L<sup>−</sup>1, and from 0.15 to 9.9 g L−1, for EO tested on lettuce, and rye-grass, respectively (Table 2). EO from aerial parts and seeds of coriander have demonstrated a higher efficiency, on both lettuce and rye-grass, than those from sage. Additionally, no difference was found between the EO distilled from a same coriander part, but under the different experimental conditions (polluted or unpolluted soil).

In the presence of *β*-CD, no significant improvement was demonstrated in our experimental conditions, with even significantly higher effects in the absence of *β*-CD for the EO of coriander's aerial parts, and of sage, on both tested plants (Table 2). Due to the retention of EO by *β*-CD, it sometimes resulted in a very limited efficiency at the tested concentrations. Thus, IC50 have not been calculated (NC) for several conditions, as it would have resulted in particularly high and inaccurate values. On another note, in comparison with the positive control, only the EO from aerial parts of coriander are in the same range of efficiency on lettuce, whereas all the tested EO are at least as efficient as the control on rye-grass.

#### 3.3.2. Seedlings' Growth Inhibition Bioassay

IC50 obtained regarding radicle growth inhibition varied from 0.017 to 1.17 g L<sup>−</sup>1, and from 0.050 to 0.66 g L−1, for lettuce and rye-grass bioassays, respectively (Table 3). On both lettuce and rye-grass, EO from coriander aerial parts displayed the highest efficiency (Figure S3), in comparison with those from coriander seeds, and from sage. In addition, EO from coriander seeds displayed higher efficiency than those from sage on lettuce. On another note, no difference was visible between the EO distilled from a same plant, but under the different experimental conditions (polluted or unpolluted soil), against both lettuce and rye-grass (Table 3). Additionally, in the presence of *β*-CD, no significant improvement was obtained in our experimental conditions, with even significantly lower effects in the presence of *β*-CD (negative effect), in most cases. In comparison with the positive control, the tested EO displayed IC50 more than 100 times higher in the case of both lettuce and rye-grass assays.

#### *3.4. EO Insecticidal Activities*

The mortality rates of *E. kuehniella*, *B. tabaci* and *R. dominica*, exposed for 24 h to different concentrations of clary sage and coriander EO, are presented in Figure 3.

Data related to the effect of the three tested EO against *E. kuehniella* showed that the EO obtained from clary sage and coriander seeds were toxic at the lowest tested concentration (9.09 μL L−<sup>1</sup> air), displaying mortality percentages ranging from 3.33 to 16.7% (Figure 3A). The mortality percentage significantly increased with higher EO concentrations, to attain 50% at the highest evaluated concentration (56.81 μL L−<sup>1</sup> air), for both EO. Conversely, the EO from coriander aerial parts did not exert toxic effect at the lowest concentration, while displaying mortality percentages up to 13.3 and 10%, for the unpolluted and polluted conditions, respectively (Figure 3A). According to the mortality percentages, EO toxicity increased in the following order: coriander seeds EO ≤ clary sage EO ≤ coriander aerial parts EO.

These results also demonstrated that the three tested essential oils had toxic effects against *B. tabaci*, even at the lowest tested concentration (9.09 μL L−<sup>1</sup> air—Figure 3B). The mortality percentages ranged between 26.7 and 50%, at the lowest and highest tested concentrations, respectively. Similar values were obtained between the three tested EO (*p* > 0.05), whatever the condition (polluted or unpolluted site).

Regarding *R. dominica* mortality, the results indicated that no mortality was caused by clary sage, and coriander aerial parts EO, at the lowest concentration (9.09 μL L−<sup>1</sup> air). Both EO displayed a similar toxicity pattern, with mortality percentages significantly increasing with higher EO concentrations, to attain 20.0 and 6.7% at the highest concentration (56.81 μL L−<sup>1</sup> air), for the unpolluted and polluted conditions, respectively (Figure 3C). Moreover, lower mortality percentages were obtained with the EO obtained from the plants grown in polluted conditions (F = 72.64; *p* ≤ 0.001). Additionally, in comparison with these 2 EO, significantly higher toxic effects were obtained with the EO from coriander seeds, displaying lethal effects at the lowest tested concentration (ranging from 6.7 to 16.7%), increasing to up to 100% at the highest concentration (56.81 μL L−<sup>1</sup> air—Figure 3C).

The statistical analysis showed highly significant differences between EO, especially between clary sage or coriander aerial parts EO on one side, and coriander seeds EO on the other, for *R. dominica* (F = 4194.41; *p* ≤ 0.001), and *B. tabaci* (F = 435; *p* ≤ 0.001), or between coriander aerial parts on one side, and clary sage and coriander seeds EO on the other, for *E. kuehniella* (F = 583.65; *p* ≤ 0.001). Moreover, whatever the EO, the concentration had a significant effect on the mortality percentages of the three tested insects, and especially *E. kuehniella* and *R. dominica* (F = 3693.01; *p* ≤ 0.001).


**Table 2.** EO's IC50 (g L−1) resulting from the seedlings' emergence inhibition bioassay against lettuce and rye-grass. Values are means ± SD (n = 3).


**Table 3.** EO's IC50(g L−1) resulting from the growth inhibition bioassay against lettuce and rye-grass. Values are means ± SD (n = 3).


obtained in free *β*-CD condition, and in the presence of *β*-CD. Means followed by the same

significantly differ.

letter—without

 and with apostrophe for lettuce and rye-grass assays,

respectively—do

 not

**Figure 3.** Mortality rates (%) of *E. kuehniella* (**A**), *B. tabaci* (**B**) and *R. dominica* (**C**) adults, exposed during 24 h to three EO (clary sage, coriander aerial parts and coriander seeds) at different concentrations (n = 3).

#### 3.4.1. Repellent Activity

The results in terms of EO repellent activity against *E. kuehniella*, *B. tabaci* and *R*. *dominica*, have been evaluated using impregnated filter paper test, and are illustrated in Table 4. Whatever the condition (unpolluted or polluted sites), and the target insect, the three tested EO exhibited an important repellent activity (F = 9.01; *p* ≤ 0.001), ranging from 13.3 to 66.7%, and dependent upon EO, and concentration (*p* < 0.05). In this test, the strongest repellent activity against *E. kuehniella* was caused by the EO distilled from the aerial parts of coriander, at the highest tested concentration (0.1 μL cm<sup>−</sup>2—66.7%), while clary sage EO showed the highest repellent activity against *R. dominica*, up to 63.3% (Table 4). Against *B. tabaci*, the three tested EO revealed similar efficiencies (F = 1.01; *p* = 0.36).

Furthermore, these results showed that the origin of the EO (polluted or unpolluted sites) had no effect on EO repellent potential (F = 1.42; *p* = 0.24), and that the effects of increasing EO concentrations are particularly marked against *E. kuehniella* (F = 65.31; *p* ≤ 0.001) and *B. tabaci* (F = 12.61; *p* ≤ 0.001).

Regarding the median repellent concentrations (RC50), the three tested EO showed good repellent activity against the three target insects, displaying RC50 values ranging from 2.61 to 3.80 μL cm−2, from 2.61 to 3.77 μL cm−2, and from 0.07 to 0.16 μL cm−2, against *E. kuehniella*, *B. tabaci* and *R. dominica*, respectively (Table S1). It should be noted that *R. dominica* displayed the highest sensitivity to the EO treatments, whatever the EO, and the condition (polluted or unpolluted sites). Furthermore, whatever the insect, the tested EO displayed similar ranges of efficiency, after 24 h of exposure, regardless of the condition (Table S1).


**Table 4.** Percentage repellency of the three different EO (sage and aerial parts and seeds of coriander), from the two experimental plots—after 24 h of exposure—against *E. kuehniella*, *B. tabaci* and *R. dominica* adults. Values are means ± SE (n = 3).

#### 3.4.2. Fumigant Toxicity

The results in terms of fumigant activity were expressed as both median lethal time (LT50), and median lethal concentration (LC50) values.

Regarding *E. kuehniella*, the three tested EO exerted a significant activity by fumigation. LC50 values ranged between 3.0 and 5.2 μL L<sup>−</sup>1, with mean median lethal times estimated between 87 and 141 h (Table 5). Moreover, no significant difference was observed between LC50 values, whatever the EO (*p* > 0.05). However, LC50 values were significantly different between the EO distilled from aerial parts of coriander, originating from the unpolluted (3.5 μL L<sup>−</sup>1), and polluted (5.2 μL L<sup>−</sup>1) sites.

**Table 5.** LT50 (h) and LC50 (μL L<sup>−</sup>1) values resulting from the fumigation bioassay against *E. kuehniella* adults, for the three different tested EO, originating from the two experimental plots (n = 3).


LT50: median lethal time; LC50: median lethal concentration; SE: standard error.

Against *B. tabaci*, the data related to LT50 values ranged from 25.5 to 37.6 h. All the three tested EO have demonstrated a fumigant lethal potential, at low concentrations ranging from 2.7 to 3.7 μL L−<sup>1</sup> of EO (Table 6). No significant difference was observed between either LT50 or LC50 values, whatever the EO, and the plant growing conditions (unpolluted or polluted sites).

**Table 6.** LT50 (h) and LC50 (μL L−1) values resulting from the fumigation bioassay against *B. tabaci* adults, for the three different tested EO, originating from the two experimental plots (n = 3).


LT50: median lethal time; LC50: median lethal concentration; SE: standard error.

Concerning *R. dominica*, LC50 values ranged between 2.2 and 4.1 μL L−1, while LT50 values were measured between 19.8 and 123.1 h (Table 7). No significant difference was observed regarding LC50 values, between EO from clary sage inflorescences and aerial parts of coriander (*p* > 0.05), and whatever the plant growing conditions (unpolluted or polluted sites—*p* > 0.05). Nonetheless, the EO distilled from seeds of coriander was more toxic against *R. dominica*, displaying lower LC50, ranging between 2.2 and 2.9 μL L<sup>−</sup>1.

**Table 7.** LT50 (h) and LC50 (μL L<sup>−</sup>1) values resulting from the fumigation bioassay against *R. dominica* adults, for the three different tested EO, originating from the two experimental plots (n = 3).


LT50: median lethal time; LC50: median lethal concentration; SE: standard error.

As a whole, it should be noted that the three target insects displayed similar ranges of susceptibility to the different EO, tested by fumigation.

#### **4. Discussion**

Natural compounds from plants, including EO, may be efficient alternatives to the conventional pesticides, especially in integrated approaches. First of all, the possible influence of the soil pollution by TE, on the EO chemical compositions, was assessed.

Then, the potential of EO obtained from clary sage (inflorescences) and coriander (aerial parts and seeds), regarding their antifungal, anti-germinative, herbicidal, and insecticidal activities, was investigated in vitro.

#### *4.1. Effect of Soil Pollution on the Chemical Compositions of EO Distilled from Coriander and Clary Sage*

In this study, the chemical composition of the EO distilled from coriander aerial parts was characterized by significant proportions of 2-decenal (between 44 and 49%), linalool (up to 35%), decanal, 2-dodecenal, and 2-tridecenal, among 15 different detected aromatic compounds. It is hence mostly composed of aliphatic aldehydes, which is consistent with previous investigations [28,60,61]. Some of these previous reports have shown that the EO chemical composition, and that of coriander EO in particular, was dependent upon the plant part that was used for the distillation [28,60,62]. Thus, it is not surprising that the EO distilled from seeds of coriander displayed a significantly different composition from the one distilled from its aerial parts, and was mostly constituted of monoterpenes, such as α-pinene or *γ*-terpinene, and especially of linalool, up to 81%. These results are consistent with previously published data [60,61]. Clary sage EO mostly consisted of oxygenated monoterpenes, up to 85% of the EO composition, such as linalool, *β*-myrcene, *α*-copaene, or *β*-caryophyllene. Linalyl acetate and linalool, both monoterpenes, are in fact the EO major compounds, as previously reported [20,26,41,48]. Moreover, the rather elevated amount of germacrene-D obtained in our experimental conditions corresponds to a previously described chemotype, rich in that specific compound [26,48,63].

However, even though the EO chemical composition for a same plant and plant part was highly similar between the tested experimental conditions (unpolluted or polluted soils), the relative abundances of several chemical compounds, such as linalool, linalyl acetate, or several aliphatic aldehydes, were found modified. Attention should be drawn to the influence of the environmental parameters on EO composition, such as geographic location, climate, soil conditions, along with cultivation practices [11,48,60]. Notably, the presence of elevated amounts of TE in soil has been shown to result in lower EO yields [43,44,64], or in altered EO chemical compositions, in response to the TE-induced stresses [37,39,65]. It is suspected that in response to TE exposure, inhibition or an activation of several key enzymes—involved in the biosynthesis pathways—could result in a modification of the plant secondary metabolism, and hence lead to either a reduction, or to an increase of specific secondary metabolites, respectively [43,64,65], which could explain the obtained differences. However, the variability among the experimental conditions was rather low, and the quality of the three different EO was faithful according to the chemotypes reported in the literature body [26,28,48,60,61], while the EO yields were in a related publication found unaffected by the soil pollution [34]. Furthermore, as previously highlighted [42,44], the response to TE exposure seems to vary greatly among aromatic plant species. In that regard, the chemical composition from clary sage inflorescence seemed to be less affected than the EO from coriander by the environmental conditions, and in particular by the presence of TE in soil. Aromatic plants from the genus *Salvia* (Lamiaceae) in particular, were in fact described as being able to tolerate elevated TE amounts, and to consistently grow in such conditions, displaying unaffected EO compositions [40,42], corroborating the obtained results. Finally, from a wider perspective, the variability recorded in terms of EO composition could be attributed not only to the presence of TE in soil, but also to the geographical location, and to the soil conditions [48,60]. In previously published

data [34,45], it was indeed highlighted that the soil physico-chemical parameters were slightly different between the two experimental sites, which could explain that the two aromatic plants grown in situ displayed slightly different EO chemical compositions, even though the plant maturity stages at harvest were identical, and that TE in soil did not hinder plant growth [34].

#### *4.2. EO Biological Activities towards a Potential Application in Crop Protection*

EO from both coriander and clary sage were previously investigated for their antifungal activity against a wide spectrum of fungal pathogens, but reports targeting plant pathogens in particular are scarce. Until now, positive results have been reported regarding the antifungal effects of EO from coriander seeds, and from Lamiaceae species, on the development of *Fusarium* spp. or various other fungal phytopathogens [4,60]. As highlighted in the majority of the previous studies, EO biological effects are often dependent upon the EO concentration [4]. In the same way, our data regarding antifungal activity, against both *F. culmorum* and *Z. tritici*, have shown that all the EO that were evaluated in this study inhibited fungal growth. Notably, the observed antifungal effect was defined as fungistatic rather than fungicidal, depicted by the revival of hyphae and mycelial growth, after transfer on a medium exempted from EO. This feature could be valuable in preventive applications as a means to control pre- and postharvest fungal diseases. The efficiency of the EO increased in the following order: clary sage EO < coriander seeds EO ≤ coriander aerial parts EO. Although the concentrations of the EO were up to 1000 times higher than those of the chemical marketed fungicides, which is commonly observed [66], the tested EO still displayed consistent antifungal activity. Furthermore, EO are known to exert lower harmful effects on non-target organisms, and on environment and human health [4,11]. Notably, they are known to possess a low persistence in soils, owing to their volatility [4,67–69], while the occurrence of resistance phenomena bound to the use of EO has not been reported so far. This feature could be bound to their action as multisite chemicals [4]. It should also be noted that, in the case of *Z. tritici*, all the tested EO were in the same range of efficiency as the positive control consisting in a marketed fungicidal product. This feature could be particularly interesting. Indeed, by displaying a substantial biocidal activity, combined with a limited toxicity towards non-target organisms, and a high volatility hence limiting environmental risks, the tested EO appear as promising candidates, when compared to conventional pesticides or even other biocontrol products.

In addition to antifungal properties, the tested EO revealed a significant anti-germinative effect, and herbicidal activity, on both lettuce and rye-grass. Our results depicted a promising activity of the EO, especially the one distilled from aerial parts of coriander, which displayed lower IC50 values than those from coriander seeds and clary sage, whatever the bioassay, and the target plant. In comparison with glyphosate, which is a systemic herbicide, and was evaluated as a positive control in this work, the tested EO displayed a consistent herbicidal activity. In previous studies, reported glyphosate IC50 varied from 15.3 mg L−<sup>1</sup> [70] to 23 and 46.2 mg L−<sup>1</sup> [71] regarding the inhibition of ryegrass growth, while the results reported on lettuce ranged from 8.9 mg L−<sup>1</sup> [72] to 20 mg L−<sup>1</sup> [73], which are comparable to those obtained in this study. In addition, the EO from aerial parts of coriander, reported as the most efficient in terms of in vitro herbicidal activity, exerted effects similar to those of glyphosate, and even higher on rye-grass. Since glyphosate is not homologated as an anti-germinative product, the use of EO to fulfil this purpose could be promising.

#### *4.3. EO Potential Applications as Post-Harvest Pests Control Products—Insectidical Properties*

Secondary metabolites from plants are also recognized to play a role in plant–insect interactions, and as such have been widely investigated for their insecticidal properties [26,74]. Their quick degradation could also favor their use as fumigants [11,75].

In this study, the potent repellent and fumigant activities were examined against the adults of *E. kuehniella*, *B. tabaci* and *R. dominica*.

In response to EO exposure, *E. kuehniella* and *B. tabaci* displayed mortality percentages up to 50%, whatever the EO, while coriander seeds EO displayed a mortality rate up to 100% against *R. dominica*. Moreover, whatever the EO concentration, the three tested EO displayed a similar range of efficiency against *B. tabaci*—the EO, and in particular those from coriander, resulted in a substantial insect mortality, even at low concentrations. Regarding repellence, the three tested EO displayed similar efficiencies against *B. tabaci*, while the EO from clary sage and from the aerial parts of coriander displayed the highest repellence percentages against *R. dominica* and *E. kuehniella*, respectively. Moreover, the obtained RC50 values ranged between 2.61 and 3.80 μL cm−2, and between 2.61 and 3.77 μL cm<sup>−</sup>2, against *E. kuehniella*, and *B. tabaci*, respectively, while those obtained against R. *dominica* were significantly lower, varying from 0.07 to 0.16 μL cm<sup>−</sup>2. Finally, in the current fumigant bioassays, the three tested EO showed similar ranges of toxicity against the three target insects, ranging from 2.2 to 5.2 μL L<sup>−</sup>1.

The insecticidal activity of several plant extracts and EO has previously been reported in several studies [26,66,67,74,75]. Little work has however been done using coriander or clary sage EO against the insects that are targeted in the present study.

Against *B. tabaci*, diverse EO were previously evaluated, such as those from *Citrus aurantium peels*, *Citrus sinensis*, *Allium sativum*, *Agastache rugosa*, *Illicium verum*, *Chenopodium ambrosioides*, *Schizonepeta tenuifolia*, *Curcuma aeruginosa*, *Syzygium aromaticum* or *Valeriana officinalis* [12,76,77]. Among all the tested EO, the strongest fumigant activities were obtained with the EO from *A. sativum*, *C. aurantium* and *A. rugosa*, with respective LC50 values of 0.11 μg L<sup>−</sup>1, 3.97 and 5.8 μL L−<sup>1</sup> and 7.08 μg L−<sup>1</sup> [12,76,77]. In contrast, some EO did not result in any fumigant toxicity at the tested concentrations, such as those from *S. tenuifolia*, *C. aeruginosa* or *V. officinalis* [76]. Whatever the bioassay, the three EO tested in the current study displayed a fumigant activity, and similar efficiencies against *B. tabaci*. Moreover, in comparison with the body of literature, the obtained LC50 values would put them among the most efficient EO reported so far against *B. tabaci*.

Concerning *E. kuehniella*, EO from *Ocimum basilicum*, *Mentha pulegium* or *Ruta graveolens* previously displayed LC50 values ranging from 0.3 to 1.02 μL L−<sup>1</sup> [11], while the one from *Pistacia lentiscus* was about 40.2 μL L−<sup>1</sup> [11,78]. Essential oils from *Eucalyptus astringens*, *Eucalyptus leucoxylon*, *Eucalyptus lehmannii*, *Eucalyptus rudis*, *Eucalyptus camaldulensis*, and *Laurus nobilis* were also effective against *E. kuehniella*, since the related LC50 values ranged between 20.5 and 33.8 μL L−<sup>1</sup> [11,79,80]. Thus, the results obtained during the present investigation suggest good potential for the three tested EO to be used as both fumigant and repellent products.

Coriander seeds EO, as well as its isolated major compounds, were previously evaluated against *R. dominica* [81]. A high mortality rate after 24 h of exposure (up to 100%) was observed, using a dose of 1 μL/15 mL of EO, which corroborates the high mortality percentages obtained with the EO distilled from seeds of coriander in the current experimental conditions. The EO from seeds of coriander has also been evaluated against several stored products pests, such as *Tribolium castaneum*, *Lasioderma serricorne* and *Sitophilus oryzae* [82], while the one distilled from coriander aerial parts was investigated for its effects against *T. castaneum* [83]. High inhibition of *T. castaneum* early development stages was observed [83], along with a significant fumigant toxicity reflected in LC50 values of 276.3, 5.3, and 145.5 μL L−<sup>1</sup> of air, against *T. castaneum*, *L. serricorne* and *S. oryzae*, respectively [82]. The LC50 values acquired for the EO of clary sage, and coriander seeds and aerial parts, are this way within the same range of efficiency, and among the most efficient ones.

Against *R. dominica*, formulated aqueous extracts of clary sage were also previously reported for their toxic effects, with mortality rates above 95%, at the highest tested concentration [84]. These mortality rates, higher than those obtained in the present study, highlight the importance of an adequate EO formulation, so as to improve the EO biological effects as well as their persistence in time, often brought forward as limited [84–86].

#### *4.4. Essential Oil Encapsulation in β-CD*

To address this issue, EO encapsulation in cyclic oligosaccharides, such as *β*-CD, could help preventing EO oxidation, thermal degradation and quick evaporation, and allow a controlled-release of the EO and of their major compounds [87–90]. Interestingly, the EO studied in this work were previously demonstrated to be efficiently complexed with *β*-CD, since retention percentages ranged from 63 to 80% [20]. These are within the same range as those commonly described for some other EO [20,86]. It hence suggests that CD can efficiently retain EO and further reduce their volatility. However, in our experimental conditions, EO complexation with *β*-CD did not result in a significant improvement of the investigated biological properties. In some specific cases, notably in the antifungal and herbicidal assays, it even resulted in a lower efficiency of the EO (negative effect), owing to their complexation with *β*-CD and consequently their reduced volatility and availability. From an agricultural perspective, encapsulation could nonetheless significantly raise the persistence of the EO's effects, given their efficient retention by *β*-CD, allowing a controlled release [89,91]. It could particularly be valuable towards a lengthening of the fungistatic effects in time, which could then be of great interest to legitimize their use as natural alternatives.

Overall, our results suggest that the presence of TE in soil did not alter the EO biological effects, whatever the assessed property. Whether they were evaluated for applications in crop protection or as post-harvest treatments, EO originating from the biomass cultivated on the polluted site mostly displayed similar efficacies as the one distilled from the unpolluted one.

#### *4.5. Insights on the Relationships between EO Composition and Their Biological Effects*

Mono- and sesquiterpenoids are commonly described as responsible for the EO biological activities, whether they are antimicrobial, herbicidal or insecticidal [12,28,81,92,93]. As such, linalool, camphor and geranyl acetate were highlighted as the active compounds of the EO distilled from seeds of coriander, in terms of fumigant toxicity against *R. dominica* [82]. Similarly, clary sage EO insecticidal activity could be bound to its high amount of linalyl acetate in particular, since the exclusion of that compound from the EO mixture resulted in a substantial decrease in terms of repellence (halved) against a mite species, *Tetranychus urticae* [31]. From a wider perspective, linalool which is present in all the three tested EO, but in different proportions, is often highlighted as one of the main factors responsible for the EO bioactivity [81,94]. However, the variation observed between the different tested EO, whatever the biological property, cannot be explained by the action of their major components only.

In fact, it has been repeatedly emphasized that EO's biological effects were rather the result of a synergism between their compounds, since the evaluation of the latter isolated or of the mixture purified from one of its compounds, resulted in lower activities [12,28,95–97]. Since EO could act as multisite chemicals, lowering the risk of resistance phenomena [4], a deeper knowledge of their action mechanisms, and of some of their compounds, alone or in combination, would be of great interest. Even though the biological properties of a wide number of EO against various pathogenic microorganisms and pests have been covered, the investigation of the action mechanisms remains indeed limited. Several main features have nonetheless been highlighted regarding antifungal activity, such as the inhibition of the fungal cell wall formation, the disruption of the cell membrane (through the inhibition of the ergosterol synthesis), the inhibition of the mitochondrial electron transport, the inhibition of the cellular division, the interference with RNA, DNA synthesis and/or protein synthesis, and the inhibition of efflux pumps [5,98,99]. In that regard, coriander EO was demonstrated as efficient against *Candida albicans*, by increasing membrane permeability through a binding interaction with a membrane ergosterol [28,100]. Insecticidal activity of the EO, which has also been thoroughly investigated, points towards a site of action in the insect nervous system [66]. Plants' EO, and especially terpenoids compounds in it, seem to exhibit their toxicity through an interaction with different putative receptors, namely

acetylcholinesterase, nicotinic acetylcholine receptor, octopamine receptor, or gammaaminobutyric acid receptor ion channel [66,101]. They could moreover target multiple sites simultaneously [66], and act as insect repellents [81]. Regarding EO phytotoxic effects, resulting in visible symptoms, they can notably be the result of mitosis inhibition, a decrease of cellular respiration, ion leakage, membrane depolarization, decrease of the chlorophyll content, oxidative damages or removal of the cuticular waxy layer [5,102–104]. In the case of cinnamon and Java citronella EO, or of their main compounds, which could act as efficient herbicides, it was, for instance, demonstrated that the plant plasma membrane could be one of the EO's cellular targets, owing to the amphiphilic nature of several compounds [102]. The authors concluded that the mentioned EO or compounds were susceptible to affect lipid organization and/or domain formation, especially in the case of monoterpenes, while phenylpropanoids are likely to interact with membrane receptors [102]. However, no comprehensive study has so far been carried out on the detailed herbicidal mechanisms [102], which could be a valuable addition to the field.

#### **5. Conclusions**

The growing number of studies related to EO biological effects tends to demonstrate their suitability for the development of natural products-based biopesticides [4,17], provided that EO stability issues are solved. Our results demonstrate that growing aromatic plant on TE-polluted surfaces—and distilling EO from the grown biomass—could be a relevant tool to engage the reclaiming of these marginal lands.

As a whole, the obtained results indicate that the three evaluated EO, from coriander (aerial parts and seeds) and clary sage (inflorescences), displayed faithful chemical compositions, despite the soil pollution by TE. They also were able to inhibit the growth of two major fungal phytopathogens, while also exerting anti-germinative and herbicidal effects, against both mono- and dicotyledon species. Notably, the EO distilled from aerial parts of coriander possessed a higher efficiency, whatever the tested biological activity. Interestingly, significant repellent and fumigant activities were also demonstrated against three major post-harvest pests, whatever the EO. As a result, these EO could be promising candidates for the development of new biopesticides. Nonetheless, if such in vitro assays may indicate the EO's potential towards applications in crop protection or as post-harvest decay control products, these effects need to be confirmed by further *in planta* or in vivo assays, so as to legitimate their use. Moreover, even though the encapsulation of the tested EO in *β*-CD did not result in any improvement of the biological properties, further assessments should be conducted to confirm the efficiency of the controlled release of EO in glasshouse or field conditions.

Furthermore, these EO could be tested in combination with conventional marketed products, as well as with other EO or biocontrol products, so as to reduce the amounts used, or investigate potential synergistic effects.

**Supplementary Materials:** The following supporting information can be downloaded at: https:// www.mdpi.com/article/10.3390/foods11030312/s1, Figure S1: Inhibitory effect of increasing EO concentrations from aerial parts of coriander, on the mycelial growth of *F. culmorum*, incubated for seven days, Figure S2: Antifungal activity of the three tested EO (aerial parts of coriander, a; seeds of coriander, b; and clary sage, c) against *Z. tritici*. Results for the in vitro microplate assay are displayed as optical densities—means from 4 values per well, Figure S3: Herbicidal activity of increasing EO concentrations from aerial parts of coriander, against *L. perenne* (a) and *L. sativa* (b), Table S1: RC50 values (μL cm−2) for the three different tested EO, from the two experimental plots—after 24 h of exposure—against *E. kuehniella*, *B. tabaci* and *R. dominica* adults (n = 3).

**Author Contributions:** Conceptualization, methodology, validation, R.R., J.F., F.L. and A.L.-H.S.; writing—original draft preparation, R.R.; writing—review and editing, R.R., J.F., F.L., J.M.B.J., A.S. and A.L.-H.S.; supervision A.L.-H.S.; project administration, A.L.-H.S.; funding acquisition, A.L.-H.S. and J.F. All authors have read and agreed to the published version of the manuscript.

**Funding:** This work was supported by l'Agence De l'Environnement et de la Maîtrise de l'Energie (ADEME, Angers, France) in the framework of PhytEO and DEPHYTOP projects. This work has also been carried out in the framework of the Alibiotech project which is financed by the European Union, the French State and the French Region of Hauts-de-France.

**Institutional Review Board Statement:** Not applicable.

**Informed Consent Statement:** Not applicable.

**Data Availability Statement:** The original contributions presented in the study are publicly available.

**Acknowledgments:** The authors wish to thank the "Université du Littoral Côte d'Opale" and the "Pôle Métropolitain de la Côte d'Opale" for providing financial support for R. Raveau's Ph.D thesis. The authors also wish to acknowledge P. Ferrant from Ferrant PHE for his technical help, and EO distillation, as well as N. Facon for her technical help.

**Conflicts of Interest:** The authors declare no conflict of interest.

#### **References**


## *Article* **Global Proteomic Analysis of** *Listeria monocytogenes'* **Response to Linalool**

**Zhipeng Gao 1,\*, Weiming Zhong 1, Ting Liu 2, Tianyu Zhao <sup>1</sup> and Jiajing Guo 2,3,\***


**Abstract:** *Listeria monocytogenes* (LM) is one of the most serious foodborne pathogens. Listeriosis, the disease caused by LM infection, has drawn attention worldwide because of its high hospitalization and mortality rates. Linalool is a vital constituent found in many essential oils; our previous studies have proved that linalool exhibits strong anti-*Listeria* activity. In this study, iTRAQ-based quantitative proteomics analysis was performed to explore the response of LM exposed to linalool, and to unravel the mode of action and drug targets of linalool against LM. A total of 445 differentially expressed proteins (DEPs) were screened out, including 211 up-regulated and 234 down-regulated proteins which participated in different biological functions and pathways. Thirty-one significantly enriched gene ontology (GO) functional categories were obtained, including 12 categories in "Biological Process", 10 categories in "Cell Component", and 9 categories in "Molecular Function". Sixty significantly enriched biological pathways were classified, including 6 pathways in "Cell Process", 6 pathways in "Environmental Information Processing", 3 pathways in "Human Disease", 40 pathways in "Metabolism", and 2 pathways in "Organic System". GO and Kyoto Encyclopedia of Genes (KEGG) enrichment analysis together with flow cytometry data implied that cell membranes, cell walls, nucleoids, and ribosomes might be the targets of linalool against LM. Our study provides good evidence for the proteomic analysis of bacteria, especially LM, exposed to antibacterial agents. Further, those drug targets discovered by proteomic analysis can provide theoretical support for the development of new drugs against LM.

**Keywords:** linalool; *Listeria monocytogenes*; antimicrobial; proteomics

#### **1. Introduction**

The prevention and control of foodborne pathogens is always an urgent need for food safety and human health worldwide [1]. *Listeria monocytogenes* is listed as one of the most serious foodborne pathogens by the World Health Organization (WHO) [2]. It is widely distributed in nature and can survive in many extreme environments such as high salt, low temperature, low pH, and so on [3,4]. Listeriosis caused by LM infection is a serious food-borne zoonotic disease with high hospitalization and mortality rates [5–8]. The clinical manifestations mainly include meningitis, septicemia, and endocarditis. Pregnant women, newborns, the elderly, and people with weakened immunity are susceptible groups. Among them, pregnant women are more likely to suffer from listeriosis, and severe cases can even cause premature birth, stillbirth, and neurological diseases in the offspring [3,9]. Thus, the control and prevention of LM has become a crucial issue all over the world.

Chemical antimicrobial agents are usually used for the prevention of LM in the food industry. However, today more and more consumers are likely to pursue "green and

**Citation:** Gao, Z.; Zhong, W.; Liu, T.; Zhao, T.; Guo, J. Global Proteomic Analysis of *Listeria monocytogenes'* Response to Linalool. *Foods* **2021**, *10*, 2449. https://doi.org/10.3390/ foods10102449

Academic Editors: Lisa Pilkington and Siew-Young Quek

Received: 30 August 2021 Accepted: 13 October 2021 Published: 14 October 2021

**Publisher's Note:** MDPI stays neutral with regard to jurisdictional claims in published maps and institutional affiliations.

**Copyright:** © 2021 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https:// creativecommons.org/licenses/by/ 4.0/).

natural" foods with few or no chemicals. Yet natural antimicrobial agents, especially essential oils or their ingredients, have become a popular research area [10–13]. Linalool (3,7-dimethyl-1,6-octadien-3-ol) is a vital constituent found in many essential oils with good antibacterial activity against different kinds of microorganisms. It is generally recognized as a safe (GRAS) food additive [14]. In our previous study, we demonstrated that linalool was a major component in citrus essential oils and exhibited strong antibacterial activity against LM—the ZOI, MIC, and MBC values were 39.58 ± 0.74 mm, 0.5% (*v*/*v*), and 1% (*v*/*v*), respectively—and it exhibited significant anti-biofilm activity by the dispersal and killing of cells in biofilm [15–17], but little is known about its mode of action against LM.

Recently, with the rapid development of omics-technology, many omics, such as genomics, transcriptomics, and proteomics, have been used as effective research tools for microbiology study [18–20]. Among them, proteomics is often used to investigate bacterial behavior under different environmental conditions in protein levels [13,21]. Research focused on the theme of "changes of microbial proteomics after drug treatment" has become a hot topic. Especially in "antibiotic treatment" fields, some research groups have used proteomics technology on the following research objects: erythromycin against *Streptococcus suis* [22], vancomycin against *Streptomyces coelicolor* [23], daptomycin against *Staphylococcus aureus* [24], bostrycin against *Mycobacterium tuberculosis* [25], emodin against *Haemophilus parasuis* [26], oxytetracycline against *Edwardsiella tarda* [27], etc.

Thus, in this study, we performed a global protein analysis by using iTRAQ (isobaric tags for relative and absolute quantitation)-based quantitative proteomic technology [28,29] to explore how LM responds to the treatment of linalool, and to unravel the mode of action and the drug targets of linalool against LM. We hope this continuous research will provide more of a theoretical basis for the prevention and control of LM.

#### **2. Materials and Methods**

#### *2.1. Bacterial Strains*

The LM (ATCC 19115) strain was obtained from Guangdong Microbiology Culture Center (GMCC, Guangdong, China) and stored at −80 ◦C.

#### *2.2. Linalool*

Linalool solution (95%) was purchased from Sigma-Aldrich (Sigma-Aldrich, Burlington, MA, USA). The antimicrobial activity of linalool was tested in our previous study, which showed the MIC value was 0.5% (*v*/*v*) [16].

#### *2.3. Treatment of LM by Linalool*

LM was grown in Brain Heart Infusion broth (BHI, Guangdong Huankai Microbial, Guangdong, China) at 37 ◦C with shaking overnight and transferred to fresh BHI broth at a ratio of 1:50. When the growth state of LM reached the logarithmic phase, linalool at a concentration of 4 × MIC was added to the bacterial solution and cultured at 37 ◦C with shaking for 8 h. After that, the cells were centrifuged (4000× *g*, 10 min) and washed three times with sterile PBS. Finally, the cells were collected for both flow cytometry and proteomics assay. For the proteomics assay, cells were frozen in liquid nitrogen for 3 h and then stored at −80 ◦C before protein extraction.

#### *2.4. iTRAQ-Based Quantitative Proteomics Analysis*

#### 2.4.1. Protein Extraction, Digestion, and Labeling with iTRAQ Reagents

The treatment of LM cells was mentioned in Section 2.3. According to the manufacturer's protocol ( from the Majorbio company, Shanghai, China), LM cells were resuspended with a lysis buffer (cocktail of 1% SDS and 8 M urea) in the ratio of 1:8, sonicated (Fielda-650D, Jiangsu TRON Intelligent Technology Co., Ltd., Jiangsu, China) for 4 min, and incubated on ice for 30 min. The lysates were centrifuged at 12,000× *g* at 4 ◦C for 15 min and the supernatants were collected. The concentration of protein was determined by bicinchoninic acid (BCA) assay. Protein digestion was carried out according to the

standard procedure and the resulting peptide mixture was labeled by using 8-plex iTRAQ reagents according to the instructions (Applied Biosystems, MA, USA). For 8-plex labeling, each iTRAQ reagent was dissolved in 50 μL of isopropanol, added to the peptide mixture, and incubated at room temperature for 2 h. The samples were labeled as (HN12-1)-115, (HN12-2)-116, (HN12-3)-117, (PG45-1)-118, (PG45-2)-119, and (PG45-3)-121. Finally, all the samples were mixed together and vacuum dried before LC-MS/MS Analysis.

#### 2.4.2. Chromatographic Separation and LC-MS/MS Analysis

Samples were re-suspended by loading buffer (ammonium hydroxide solution containing 2% acetonitrile, pH 10) and separated by high-pH reversed phase liquid chromatography (RPLC, Waters, Milford, MA, USA). The gradient elution was performed on a high pH RPLC column at 400 μL/min with the gradient increased for 66 min. Twenty fractions were collected from each sample. LC-MS/MS analysis was performed by a Q Exactive mass spectrometer (Thermo Fisher Scientific, Waltham, MA, USA) coupled with an Easy-nLC 1200 (Thermo Fisher Scientific, Waltham, MA, USA) in the data-dependent mode. Survey full-scan MS spectra were acquired at a mass resolution of 70 K, followed by 20 sequential high energy collisional dissociation LC-MS/MS scans at a resolution of 17.5 K. One micro-scan was recorded by using a dynamic exclusion of 18 s.

#### 2.4.3. Proteomic Analysis

All the LC-MS/MS spectra were searched by using the Protein Discoverer Software (ProteomeDiscovererTM Software 2.4, Thermo Fisher Scientific, Waltham, MA, USA) against the Mycoplasma database. The highest score for a given peptide mass was used to identify parent proteins. The parameters for protein searching were as follows: tryptic digestion with up to two missed cleavages, carbamidomethylation of cysteine as the fixed modification, and oxidation of methionine and protein N-terminal acetylation as variable modifications. Peptide spectral matches were validated based on q-values at a 1% false discovery rate.

#### *2.5. Flow Cytometry Analysis*

Flow cytometry analysis was carried out to investigate the effects of linalool on LM. After treatment as mentioned in Section 2.3, the bacterial cells were collected by centrifugation and adjusted to the concentration of 1 × 106 cfu/mL; then, different staining procedures were proceeded as follows [30,31].

#### 2.5.1. Membrane Permeability

Thiazole orange (TO, Sigma-Aldrich, Burlington, MA, USA) and propidium iodide (PI, Sigma-Aldrich, Burlington, MA, USA) were used to evaluate the membrane permeability of the cells. For TO staining, 1 μL of TO solution was added to 1 mL bacterial suspensions (final concentration of TO: 10 μg/mL in DMSO), and then incubated at room temperature for 15 min. For PI staining, 1 μg PI was added to 1 mL bacterial suspensions (final concentration of PI: 1 μg/mL in PBS) and then incubated at 37 ◦C for 15 min.

#### 2.5.2. Membrane Potential

Bis-1,3-dibutylbarbutiric acid (BOX, Sigma-Aldrich, MA, USA) and PI were used to evaluate the membrane potential of cells. For BOX staining, 2.5 μg of BOX was added to 1 mL bacterial suspensions (final concentration of BOX: 2.5 μg/mL in PBS with 4 mM EDTA), and then incubated at 37 ◦C for 15 min. For PI staining, the procedure was the same as mentioned in Section 2.5.1.

#### 2.5.3. Efflux Activity

Ethidium bromide (EB, Sigma-Aldrich, Burlington, MA, USA) was used to evaluate the efflux activity of cells. For EB staining, 10 μg of EB was added to 1 mL bacterial suspensions (final concentration of EB: 10 μg/mL in DMSO), and then incubated at 37 ◦C for 15 min.

#### 2.5.4. Respiratory Activity

5-cyano-2,3-ditolyl tetrazolium chloride (CTC, Sigma-Aldrich, MA, USA) was used to evaluate the respiratory activity of cells. For CTC staining, 5 mM of CTC was added to 1 mL bacterial suspensions (final concentration of CTC: 5 mM in PBS with 1% (*w*/*v*) glucose), and then incubated at 37 ◦C with shaking (250 rpm) for 30 min.

After finishing these staining procedures, samples were washed with PBS three times, the concentration of the bacterial suspension was adjusted to about OD600 = 0.1, and the samples were placed on ice for flow cytometry analysis by a flow cytometer (BD Accuri C6 plus, BD, Franklin Lakes, NJ, USA).

#### *2.6. Statistical Analysis*

All the experiments were performed in triplicate. Statistical analysis was performed by GraphPad Prism 7.0 for *t*-tests. All asterisks indicate significant differences (*p* < 0.05).

#### **3. Results**

#### *3.1. Quality Assessment of Proteomics Sequencing*

As shown in Figure 1, a total of 26,880 peptides and 2102 proteins were identified by proteomics sequencing (Figure 1C). The lengths of the peptides were mainly between 6–20 amino acids, among which 7 was the most common (Figure 1A). The number of peptides that make up proteins was mainly concentrated between 1–21, among which 1–3 was the most common (Figure 1B). Among those identified proteins, the molecular weight was mainly concentrated between 11–60 kDa, among which 21–30 kDa was the most common (Figure 1D).

**Figure 1.** The quality assessments of proteomics sequencing. (**A**) Histogram of peptide length distribution. The abscissa represents the range of peptide length, and the ordinate represents the number of peptides of the corresponding length. (**B**) Histogram of peptide quantity distribution. The abscissa represents the range of the number of peptides covering the proteins, and the ordinate represents the number of the proteins. (**C**) Statistical histogram of different types of identified proteins.

(**D**) Histogram of protein molecular weight distribution. The abscissa represents the distribution range of protein molecular weight, and the ordinate represents the number of proteins corresponding to the molecular weight.

#### *3.2. Functional Annotation and Analysis of Proteins*

After quality assessment, GO and KEGG databases were used to annotate and analyze the functions of the identified proteins to explore their biological pathways and functions. As shown in Figure 2A, Biological Processes, Cell Components, and Molecular Functions (within which all the identified proteins were included) were obtained by GO classification annotation. Among "Molecular Functions", the top five were the catalytic activity, binding, transporter activity, transcription regulator activity, and structural molecule activity, and the number of proteins involved were 1192, 860, 164, 77, and 60, respectively. Among "Cell Components", cell part, membrane part, membrane, protein-containing complex, and organelle were the top five, and the number of proteins involved were 548, 424, 128, 102, and 56, respectively. Among "Biological Processes", metabolic process, cellular process, localization, biological regulation, and response to stimulus were the top five, and the number of proteins involved were 847, 795, 166, 161, and 79, respectively.

**Figure 2.** The annotation and analysis of all the identified proteins. (**A**) Annotation based on GO function classification. Each column represents a secondary classification. The ordinate represents

the secondary classification term of GO, and the capital letters in front of the term represent the following categories: BP, biological process; CC, cellular component; MF, molecular function. (**B**) The top 20 KEGG pathways with the largest number of proteins. The abscissa represents the name of pathways, and the ordinate represents the number of proteins within each pathway.

As shown in Figure 2B, the top 20 KEGG pathways with the largest number of proteins involved were obtained by KEGG classification annotations. Among them, the top 10 pathways were biosynthesis of antibiotics, biosynthesis of amino acids, carbon metabolism, ABC transporter, phosphotransferase system (PTS), ribosome, starch and sucrose metabolism, quorum sensing, glycolysis/gluconeogenesis, and purine metabolism.

#### *3.3. Analysis of DEPs*

Through the differential expression analysis of all the identified proteins between the linalool treatment group and control group, a total of 445 DEPs were screened out, including 211 up-regulated and 234 down-regulated proteins. As shown in Figure 3, the data of the DEPs were converted to volcano plots and heat maps for more intuitionistic comparative analysis.

**Figure 3.** DEPs of LM cells analyzed between the treated group (treated with linalool) and the control group (untreated with linalool). (**A**) Volcano plot of DEPs. The abscissa represents the fold change value of the difference between control and treated samples. The difference value is obtained by dividing the expression level of control sample by treated sample, and this value is logarithmized. The ordinate represents *p*-value (by the analysis of statistical *t*-test) of the difference of protein expressions. The smaller the *p*-value, the more significant the difference in protein expression. Each point represents a specific protein: the yellow point (significantly up-regulated, *p* < 0.05), the red point (significantly up-regulated, *p* < 0.01), the light blue point (significantly down-regulated, *p* < 0.05), the blue point (significantly down-regulated, *p* < 0.01), and the black dots (non-significantly different proteins, *p* > 0.05). (**B**) Heat map of DEPs. The left (CK1, control group) and right columns (TR1, treated group) represent control and treated groups, respectively. Each row represents a protein. Red and green colors represent the high and low expression levels of the protein, respectively. On the left is the dendrogram of protein clustering, and on the right is the name of the protein.

#### *3.4. The GO Enrichment Analysis*

The enrichment analysis method is usually used to analyze whether a group of proteins has appeared on a certain functional node in a certain pathway; the aim is to make the annotation analysis from a single protein to a protein set. Enrichment analysis improves the reliability of research and can screen out the biological processes most relevant to biological phenomena. Therefore, GO enrichment analysis was performed to analyze the functional enrichment of differential proteins and clarify the differences between treated and control groups at the functional level.

As shown in Figure 4, a total of 31 significantly enriched functional categories were obtained. Among them, there were 12 categories in "Biological Process", including biological adhesion (GO:0022610), biological regulation (GO:0065007), carbon utilization (GO:0015976), cellular component organization or biogenesis, (GO:0071840), cellular process (GO:0009987), developmental process (GO:0032502), localization (GO:0051179), locomotion (GO:0040011), metabolic process (GO:0008152), multi-organism process (GO:0051704), reproductive process (GO:0022414), and response to stimulus (GO: 0050896). There were 10 categories in "Cell Component", including cell (GO:0005623), cell part (GO:0044464), extracellular region (GO:0005576), extracellular region part (GO:0044421), membrane (GO:0016020), membrane part (GO:0044425), nucleoid (GO:0009295), organelle (GO:0043226), organelle part (GO:0044422), and protein-containing complex (GO:0032991). There were nine categories in "Molecular Function", including antioxidant activity (GO:0016209), binding (GO:0005488), catalytic activity (GO:0003824), molecular carrier activity (GO:0140104), molecular function regulator (GO:0098772), structural molecule activity (GO:0005198), transcription regulator activity (GO:0140110), translation regulator activity (GO:0045182), and transporter activity (GO:0005215).

**Figure 4.** The GO enrichment analyzed between the treated group (treated with linalool) and the control group (untreated with linalool). (**A**) Histogram of GO enrichment of DEPs. Each column represents a GO term, and the abscissa represents the name and category of GO. The ordinate represents the enrichment rate. The color represents the significance of enrichment. *p* < 0.05, *p* < 0.01, *p* < 0.001 are marked as \*, \*\*, and \*\*\*, respectively. Subfigures (**B**–**D**) are chord diagrams of the GO enrichment of DEPs, respectively showing the (**B**) different proteins participating in specific functions in the three GO categories of biological process, (**C**) cell composition, (**D**) and molecular function.

#### *3.5. The KEGG Enrichment Analysis*

In addition to the GO enrichment analysis, KEGG pathway enrichment analysis was also performed to analyze the biological pathways involved in DEPs in this study. As shown in Figure 5, through KEGG enrichment analysis, a total of 60 significantly enriched biological pathways were identified. Among them, there were 6 pathways in the "Cell

Process" category, 6 pathways in the "Environmental Information Processing" category, 3 pathways in the "Human Disease" category, 40 pathways in the "Metabolism" category,

**Figure 5.** The enrichment of the DEPs in KEGG pathways analyzed between the treated group (treated with linalool) and the control group (untreated with linalool). Each column represents a pathway, and the abscissa represents the name and classification of the pathway: CP (cellular process), EIP (environmental information processing), GIP (genetic information processing), HD (human diseases), M (metabolism), and OS (organismal systems). The height of the column or the ordinate represents the enrichment rate. The color represents the significance of enrichment. *p* < 0.05, *p* < 0.01, and *p* < 0.001 are marked as \*, \*\*, and \*\*\*, respectively.

#### *3.6. Flow Cytometry Analysis*

Five fluorescent dyes (TO, PI, BOX, EB, and CTC) were used to evaluate several vital biological functions in LM cells by flow cytometry analysis. Membrane integrity was evaluated by double staining of TO and PI as shown in Figure 6A. In the control group, 90.4% of the cells were located in plot Q1 (TO+ and PI−), which represented cells with an intact cell membrane. By contrast, in the treated groups, 98.8% of the cells were located in plot Q4 (TO− and PI−), which represented cells with damaged DNA or RNA.

Membrane potential was evaluated by double staining of BOX and PI as shown in Figure 6B. In the control group, 70.8% of the cells were located in plot Q4 (BOX− and PI−), which represented cells with a polarized membrane. By contrast, in the treated groups, 76.1% and 17.6% of the cells were located in plot Q1 (BOX+ and PI−) and plot Q1 (BOX+ and PI+), which represented cells with depolarized nonpermeabilized and permeabilized membranes, respectively. In total, after the treatment of linalool, 93.7% cells became depolarized.

The efflux activity was evaluated by EB staining as shown in Figure 6C. EB− represents the efflux pump functioning properly, while EB+ represents the malfunction of the efflux pump. The percentages of EB+ cells in the control and treated groups were 2% and 85.5%, respectively, and the percentages of EB− cells in these two groups were 96% and 6.97% respectively. These data show that the efflux pump of 85.5% of the cells became malfunctioning after the treatment of linalool.

Respiratory activity was evaluated by CTC staining as shown in Figure 6D. CTC+ and CTC− represent respiratory active and inactive cells, respectively. The percentage of CTC+ cells in the control and treated groups were 99% and 8.53%, and the percentage of CTC− cells in these two groups were 0.96% and 84.6%.

**Figure 6.** Fluorescence density plots of LM treated (treated) and untreated (control) with linalool stained with (**A**) PI and TO, (**B**) PI and BOX, (**C**) EB, and (**D**) CTC. For subfigures (**A**) and (**B**), the vertical and horizontal axis indicate the fluorescence intensity. The percentages of the cell population in each gate are demonstrated in the four corners of each plot.

#### **4. Discussion**

#### *4.1. Proteomic Technology Used in the Antimicrobial Research Field*

In this study, iTRAQ-based quantitative proteomics analysis was performed to identify proteins differentially expressed between the linalool treated group and the untreated group. Finally, a total of 445 DEPs were screened out, including 211 up-regulated and 234 downregulated proteins which participated in different biological functions and pathways.

Except for our study, there have been few other studies focused on the theme of "changes of microbial proteomics after essential oils treatment"; some of these studies are summarized as follows. Xu et al. [32] studied the proteomic changes of *Botrytis cinerea* after tea tree oil treatment, finding a total of 718 DEPs, of which 17 were up-regulated and 701 were down-regulated. These proteins were annotated to 30 GO categories and 133 KEGG pathways, including glycolysis, tricarboxylic acid cycle, and purine metabolism pathways. Hu et al. [33] screened out a total of 745 DEPs of LM after thyme essential oil treatment, of which 246, 45, and 309 proteins were involved in biological processes, cellular components, and molecular functions, respectively. Meanwhile, these proteins participated in 86 KEGG pathways, such as flagella assembly and chemotaxis. Yang et al. [13] demonstrated the proteomic changes of *Klebsiella pneumoniae* after treatment by lavender essential oil. A total of 135 DEPs were found, of which 57 were up-regulated and 78 were down-regulated; they were annotated to 30 GO categories and 133 KEGG pathways. Moreover, 35.78%, 34.01%, 8.44%, and 8.84% of these proteins were involved in cellular processes, metabolic processes, cellular component structure, and stress response, respectively; 58.27%, 17.53%, 13.33%, and 6.17% were involved in the cytoplasm, cell membrane, protein complex, and ribosomal proteins, respectively; and 44.40% and 42.28% were involved in binding and catalytic activity.

Although there are many differences between the above studies and ours, there are also some similarities, such as changes in glycolysis and other metabolic pathways. These differences might be caused by different proteomic technologies, types of essential oils, and the different microbial strains used. Proteomic technology has been proved to be an efficient, fast, and useful approach to identify differentially expressed proteins in various types of microorganisms.

#### *4.2. Further Analysis of Important Functions in GO Enrichment Analysis*

Several categories relevant to "cell component" were enriched as shown in Figure 4A. Firstly, the enrichment of "cell (GO:0005623)" and "cell part (GO:0044464)" indicated that the structure and component of bacterial cells were altered significantly after linalool treatment. Secondly, the enrichment of "membrane (GO:0016020)" and "membrane part (GO:0044425)" was in accordance with the results of SEM and TEM in our previous study, which showed obvious damage of the cell membrane [15]. Thirdly, the enrichment of "nucleoid (GO:0009295), organelle (GO:0043226), and organelle part (GO:0044422)" echoed the data of the flow cytometry analysis, which proved that the DNA or RNA (main part of the nucleoid) might be destroyed by linalool treatment; meanwhile, "organelle part" also echoed the enrichment of the "ribosome pathway (ko03010)" in KEGG analysis. To sum up, all the above data implied that the cell membrane, nucleoid, and ribosome may be the potential targets of linalool.

As shown in Figure 4A, some categories related to "molecular function" were also enriched. Initially, the enrichment of "structure molecule activity (GO:0005198)" meant that the action of molecules which contributes to the structural integrity or assembly of a complex significantly changed. Furthermore, the enrichment of both "transcription regulator activity (GO:0140110)" and "translation regulator activity (GO:0045182)" indicated the gene expression and polypeptide synthesis were significantly influenced by linalool, which was further evidence showing that the nucleoid and ribosome might be the potential targets.

Some categories relevant to "biological process" were also enriched as shown in Figure 4A. Primarily, the enrichment of "response to stimulus (GO:0050896)" indicated that cells may produce many adaptive responses to meet the challenge of the stimulus (linalool). Further, the enrichment of "biological adhesion (GO:0022610)" might be related to two proteins (InlA and InlB) as shown in the "bacterial invasion of epithelial cells (ko05100)" pathway in KEGG analysis. These two proteins are surface invasions, which could promote the uptake of LM by host cells [34] and enhance the adherence of LM to a glass surface [35]. Moreover, the enrichment of "reproductive process (GO:0022414)" suggested that linalool treatment might influence the reproduction of cells, which was also shown in our previous study [15].

#### *4.3. Further Analysis of Important Pathways in KEGG Enrichment Analysis*

First of all, as shown in the "peptidoglycan biosynthesis (ko00550)" pathway in Figure S1, several proteins related to peptidoglycan biosynthesis were down-regulated, which indicated that the biosynthesis of peptidoglycan was reduced by linalool treatment. Two kinds of penicillin-binding proteins (PBPs), PBP 2 and PBP 1a, were down-regulated. PBPs are a major class of enzymes related to peptidoglycan synthesis, which were identified as targets of β-lactam antibiotics (such as penicillin); they inactivate the crosslinking domains of peptidoglycan synthesis covalently [36]. Peptidoglycan is a vital component of cell walls, maintaining the morphology and viability of bacterial cells [37]. It has been proven that peptidoglycan is the target of many antimicrobial drugs, such as penicillin [36] and cephalosporin [38]. Thus, our results imply that peptidoglycan and cell walls might be another important drug target of linalool.

Next, two peptidoglycan enzymes, AmiA and AmiC, were down-regulated as shown in Figure S2. AmiA and AmiC are N-acetylmuramyl-l-alanine amidases that move side chains away from bacterial peptidoglycan through cleaving the amide bond between peptides and N-acetylmuramic acids [39], which are crucial for septal splitting and the separation of daughter cells [40]. Thus, the down-regulation of these two proteins indicated that the cell division was inhibited by linalool treatment, and this phenomenon might also prove the changing of the "reproductive process" in the GO enrichment analysis.

Moreover, many proteins relevant to the structure and function of ribosome were significantly up- or down-regulated as shown in the "ribosome (ko03010)" pathway in Figure S3, which indicates that the function of ribosome was obviously affected by linalool treatment. These were consistent with our results in GO enrichment analysis, which again suggested ribosomes were the target of linalool.

To sum up, based on our results from the GO and KEGG enrichment analysis, cell membranes, cell walls, nucleoids, and ribosomes might be the main targets of linalool against LM. These results also proved the multi-target effects of linalool, which is one of its most important advantages compared to traditional antibiotics.

#### **5. Conclusions**

In this study, iTRAQ-based quantitative proteomics sequencing was carried out to unravel the mode of action of linalool against LM at the protein level. A total of 445 DEPs (including 211 up-regulated and 234 down-regulated proteins) were screened out between the linalool treatment group and the control group. GO and KEGG enrichment analysis implied that cell membranes, cell walls, nucleoids, and ribosomes might be the targets of linalool against LM, which was also supported by the results of the flow cytometry analysis. In the future, we will focus on each of these targets and investigate the specific mechanisms. Meanwhile, the application of linalool for the prevention of LM in foods or food facilities should also be explored.

**Supplementary Materials:** The following are available online at https://www.mdpi.com/article/10 .3390/foods10102449/s1, Figure S1: Peptidoglycan biosynthesis pathway from the KEGG analysis. Figure S2: AmiA and AmiC annotation from the KEGG analysis. Figure S3: Ribosome pathway from the KEGG annotation analysis.

**Author Contributions:** Methodology, Z.G.; validation, J.G.; investigation, W.Z. and T.Z.; writing original draft preparation, J.G.; writing—review and editing, Z.G.; visualization, T.L.; supervision, Z.G. and J.G.; project administration, Z.G.; funding acquisition, Z.G. All authors have read and agreed to the published version of the manuscript.

**Funding:** This research was funded by the National Natural Science Foundation of China (32073020), the Key Projects of Hunan Education Department (20A238), the Changsha Municipal Natural Science Foundation (kq2014070), the Key Laboratory of Agro-Products Processing, and the Ministry of Agriculture and Rural Affairs of P. R. China (S2021KFKT-22).

**Conflicts of Interest:** The authors declare no conflict of interest.

#### **References**

