**Use of Essential Oils and Volatile Compounds as Biological Control Agents**

Editors

**Marie-Laure Fauconnier Ha¨ıssam Jijakli Caroline De Clerck**

MDPI • Basel • Beijing • Wuhan • Barcelona • Belgrade • Manchester • Tokyo • Cluj • Tianjin

*Editors* Marie-Laure Fauconnier University of Liege ` Belgium

Ha¨ıssam Jijakli University of Liege ` Belgium

Caroline De Clerck University of Liege ` Belgium

*Editorial Office* MDPI St. Alban-Anlage 66 4052 Basel, Switzerland

This is a reprint of articles from the Special Issue published online in the open access journal *Foods* (ISSN 2304-8158) (available at: https://www.mdpi.com/journal/foods/special issues/Use Essential Oils Volatiles Compounds Biocontrol Agents).

For citation purposes, cite each article independently as indicated on the article page online and as indicated below:

LastName, A.A.; LastName, B.B.; LastName, C.C. Article Title. *Journal Name* **Year**, *Volume Number*, Page Range.

**ISBN 978-3-0365-4127-3 (Hbk) ISBN 978-3-0365-4128-0 (PDF)**

© 2022 by the authors. Articles in this book are Open Access and distributed under the Creative Commons Attribution (CC BY) license, which allows users to download, copy and build upon published articles, as long as the author and publisher are properly credited, which ensures maximum dissemination and a wider impact of our publications.

The book as a whole is distributed by MDPI under the terms and conditions of the Creative Commons license CC BY-NC-ND.

## **Contents**



## **About the Editors**

## **Marie-Laure Fauconnier**

Marie-Laure Fauconnier, Full Professor, Head of the Laboratory of Chemistry of Natural Molecules. She is specialized in extraction, chemical characterization and use of plant secondary metabolites including essential oils for applications in agronomy.

## **Ha¨ıssam Jijakli**

Ha¨ıssam Jijakli, Full Professor, Professor in Urban agriculture and Plant Pathology, Director of Research Center in Urban Agriculture, Director of Integrated and Urban Plant Pathology Laboratory, co-funder of APEO, he is developing and implementing biocontrol methods against plant diseases.

## **Caroline De Clerck**

Caroline De Clerck, Assistant Professor, in Charge of the AgricultureIsLife Platform, she is specialized in plant bio-assays in controlled and field conditions, in the implementation of new agricultural practices and new biocontrol methods against insects and plant diseases.

## *Editorial* **Use of Essential Oils and Volatile Compounds as Biological Control Agents**

**Caroline De Clerck 1,†, Manon Genva 2,†, M. Haissam Jijakli <sup>3</sup> and Marie-Laure Fauconnier 2,\***


Plants containing essential oils have been used for centuries as spices, remedies or for their pleasant odor. In the Middle Ages, the development of distillation techniques made it possible to obtain essential oils, which have continued to be used in their historical applications in food, medicine or cosmetics [1]. However, over the last few decades, the essential oil sector has entered a new dimension, as its fields of application are constantly increasing, largely due to the biocidal properties of its constituents.

The emergence of the resistance of targeted populations, ecological concern and impact on human health paved the way to the development of more sustainable alternatives to synthetic conventional biocides. Essential oils that combine highly biocidal properties with a specific or broad spectrum of action as well as a high volatility, thus limiting residues in foodstuff or the environment, are perfect candidates for a new generation of biocides. Used in plant protection as bactericides, fungicides or insecticides in both pre- and postharvest treatments; as food ingredients to increase shelf-life; or incorporated in innovative packaging, research in the field of essential oils has a bright future ahead of it.

Three major subjects were discussed in the present Special Issue entitled "Use of Essential Oils and Volatile Compounds as Biological Control Agents": stored product insecticides, plant protection and food additives-food packaging.

Six research articles were published on the first topic, focusing on the insecticidal properties of essential oils, with the challenging perspective of replacing chemical insecticides that are widely used during crop cultivation and post-harvest treatments and therefore reducing the quantities of residues in foods. Oftadeh et al. first highlighted the high level of interest in essential oil from flowers of *Artemisia annua* L. in the control of *Glyphodes pyloalis* Walker, which damages mulberry leaves and induces the transmission of plant pathogenic agents [2]. In the second paper, the authors described the interesting contact toxicity of *Satureja intermedia* C.A.Mey essential oil against *Aphis nerii* Boyer de Fonscolombe, which is an insect pest in many ornamental plant cultures causing direct plant damage and transmitting pathogenic viruses. Interestingly, *Coccinella septempunctata* L., which is a predator of *A. nerii* and is used as biocontrol agent, was less susceptible to the essential oil. Moreover, the authors also described the elevated fumigant toxicity of *S. intermedia* essential oil against *Trogoderma granarium* Everts, *Rhyzopertha dominica* Fabricius, *Tribolium castaneum* Herbst, and *Oryzaephilus surinamensis* L., which are all common insect pests in stored products [3]. Loss during food storage due to insect infestation is a huge problem, both in developing and in developed countries. Contact chemical insecticides are therefore traditionally used to reduce food losses, with the problems of resistance appearance and the persistence of chemical residues in food. Essential oils, along with their complex composition, their low mammal toxicity and their high volatility, have emerged as promising

**Citation:** De Clerck, C.; Genva, M.; Jijakli, M.H.; Fauconnier, M.-L. Use of Essential Oils and Volatile Compounds as Biological Control Agents. *Foods* **2021**, *10*, 1062. https:// doi.org/10.3390/foods10051062

Received: 8 May 2021 Accepted: 10 May 2021 Published: 12 May 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/).

alternatives to chemical insecticides in stored products. In the next article, Demeter et al. studied the insecticidal activity of 25 essential oils against *Sitophilus granarius* L., which is one of the main insect pests during grain storage. The authors showed a high potential in different essential oils, such as those from *Allium sativum* L., *Mentha arvensis* L. and *Eucalyptus dives* Schauer for the control of *S. granarius* in stored products [4]. Tanoh et al. also showed the toxicity of the newly described essential oils from *Zanthoxylum leprieurii* Guill. & Perr. against the same insect [5]. Thereafter, Owolabi et al. described the high toxicity of essential oils from a Nigerian plant, *Launaea taraxacifolia* (Willd.) Amin ex C. Jeffrey, against *Sitophilus oryzae* L., the rice weevil that causes high food losses during grain storage [6]. Finally, Liang et al. showed the high insecticidal properties of essential oil from *Elsholtzia ciliata* (Thunb.) Hyl. and of its major components, carvone and limonene, in the control of *Tribolium castaneum* Herbst, a common beetle affecting many stored products, such as cereals and flours [7].

In the second topic of the present Special Issue, two articles described the high interest of essential oils in crop plant protection. De Clerck et al. firstly screened 90 commercially available essential oils for their in vitro antifungal and antibacterial activities against 10 phytopathogens that particularly attack plant crops and decrease food production yields. The authors highlighted that several essential oils, such as that from *Allium sativum* L., are active on diverse pathogens and thus have a "generalist" effect, while other essential oils such as that from *Citrus sinensis* (L.) Osbeck have an action on one to three pathogens, and thus a more "specific" effect [8]. In the review from Werrie et al., the authors described the high interest of essential oils for the development of biopesticides, but they also underlined the different restrictions on their use, as some of them display phytotoxicity on untargeted crops. The authors mentioned the different parameters that need to be taken into account to limit that risk, such as the mode of application, the phenological state and the product formulation [9].

In the last topic of this Special Issue, different authors studied the potential of essential oils as food additives or for their incorporation into food packaging. Siroli et al. firstly showed that the incorporation of essential oils into the marinade increased the sensorial perception of the marinated pork loin [10]. In the next article, Licon et al. showed that the incorporation of essential oils from *Thymus vulgaris* L. in milk used for the production of pressed ewes' cheese had an interesting antimicrobial effect, with a decrease in the growth of exogenous detrimental microorganisms without affecting the cheese natural flora [11]. Ben-Fadhel et al. then highlighted the antimicrobial interest of essential oils for the treatment of ready-to-eat carrots. Indeed, their incorporation into emulsions that were applied to the carrot surface allowed the lengthening of the carrot shelf-life by two days [12]. Ruengvisesh et al. studied the antimicrobial activities of micelles formed from sodium dodecyl sulfate. The authors showed that eugenol-loaded micelles were particularly effective in inhibiting *Escherichia coli* and *Salmonella enterica* when applied on fresh spinach surfaces [13]. Essential oils also emerged as interesting bioactive additives for their incorporation into active packaging. In their article, Díaz-Galindo et al. showed that the incorporation of cinnamon essential oil emulsions into thermoplastic starch leads to a decrease in the growth rate of *Botrytis cinerea* without affecting the thermal stability of the packaging [14]. As essential oil volatility may limit their applications when the release is too fast, Maes et al. studied the potential of biosourced dendrimers to encapsulate essential oils. Their results show that stirring time and stirring rate are crucial parameters that need to be optimized for an efficient encapsulation, which paves the way to numerous essential oil applications when a slower release is needed [15]. Bleoancă et al. studied two different treatments for the formation of edible films containing thyme extracts. Both highpressure-thermally treated films and thermally treated ones display different structures with different abilities to retain volatile compounds [16]. Finally, Kostoglou et al. showed the promising potential of three plant terpenoids—carvacrol, thymol and eugenol—as antibiofilms agents, as they showed significant anti-biofilm activities against *Staphylococcus*

*aureus* and *Staphylococcus epidermidis*. Those two microorganisms are notably the cause of foodborne diseases and nosocomial infections [17].

The success of this Special Issue demonstrates clearly the scientific interest around the use of volatile compounds, especially essential oils, as biological control agents in food products. In addition, with controversial products being removed from the market, alternative products such as essential oils are expected to rise.

While this topic seems to have a bright future, some questions and difficulties remain. One of the first challenges encountered in the development of biopesticides using volatile molecules is their short persistence (volatility, degradation, etc.) in comparison to synthetics. This can be positive in terms of environmental impacts and in terms of food residues, but the release kinetic of the compounds and their molecular dynamics have to be known and controlled to ensure the product's efficacy. The formulation thus plays an important role, and technology is evolving, as highlighted in several papers of this Special Issue, with the development of nano-emulsions and encapsulation, among others. These formulations are also important to avoid the apparition of any adverse tastes or odors on stored food products. The authors also pointed out the need for an upscaling of the tests, which will help to assess the practical applicability of the treatments. A number of compounds have proven their efficacy in vitro and seem promising. However, in vitro tests will always need to be confirmed in vivo.

Essential oils and volatile compound activities are often attributed to mixtures of compounds. While this could be an advantage to prevent the development of resistances if they present different modes of action, as has been shown in [18], with two constituents of essential oils with distinct chemical structure interacting differentially with plant plasma membrane, this complex composition presents challenges to regulatory standards, where regulations are generally designed for synthetic substances that contain a single, highly concentrated and persistent molecule. This is leading to difficulties regarding market approval by the different regulatory agencies throughout the world, as well as economic considerations. Even if procedures are sometimes available for plant-based products, few active substances have been registered so far, especially in the pre- and post-harvest fields. Uses as ingredients in food products are less problematic, as only a few essential oils have restricted regulation concerns (e.g., mint essential oils).

More investigations need to be performed to decipher the mechanism of action of these volatile compounds, including the role of minor components and the synergic or additive effect among them. This will be crucial to evaluate the risks on the environment (plants, beneficial organisms (insects, worms ... ), soil microbiota, etc.), and human health, as well as to secure their industrial use.

To conclude, the use of volatile compounds and essential oils in particular for sustainable agricultural practices or as food ingredients seems promising, and extensive research will probably clarify or deny their relevance in diverse applications. They can be an efficient alternative to synthetic plant protection products when properly formulated and integrated with other pest management strategies; they can also be valuable food ingredients or innovative packaging constituents

The works collected in this Special Issue will certainly contribute to the field by increasing the knowledge on volatile compounds used as biological control agents, their efficiency and formulation in a large panel of situations related to the food sector.

**Author Contributions:** Conceptualization, C.D.C., M.G., M.H.J. and M.-L.F.; writing—original draft preparation, C.D.C., M.G., M.H.J. and M.-L.F.; writing—review and editing, C.D.C., M.G., M.H.J. and M.-L.F. All authors have read and agreed to the published version of the manuscript.

**Funding:** This research was funded by the Education, Audiovisual and Culture Executive Agency (EACEA) trough EOHUB project 600873-EPP-1-2018-1ES-EPPKA2-KA.

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

## **References**


## *Article* **Mulberry Protection through Flowering-Stage Essential Oil of** *Artemisia annua* **against the Lesser Mulberry Pyralid,** *Glyphodes pyloalis* **Walker**

**Marziyeh Oftadeh 1, Jalal Jalali Sendi 1,2,\*, Asgar Ebadollahi 3,\*, William N. Setzer 4,5 and Patcharin Krutmuang 6,7,\***


**Abstract:** In the present study, the toxicity and physiological disorders of the essential oil isolated from *Artemisia annua* flowers were assessed against one of the main insect pests of mulberry, *Glyphodes pyloalis* Walker, announcing one of the safe and effective alternatives to synthetic pesticides. The LC50 (lethal concentration to kill 50% of tested insects) values of the oral and fumigant bioassays of *A. annua* essential oil were 1.204 % W/V and 3.343 μL/L air, respectively. The *A. annua* essential oil*,* rich in camphor, artemisia ketone, β-selinene, pinocarvone, 1,8-cineole, and α-pinene, caused a significant reduction in digestive and detoxifying enzyme activity of *G. pyloalis* larvae. The contents of protein, glucose, and triglyceride were also reduced in the treated larvae by oral and fumigant treatments. The immune system in treated larvae was weakened after both oral and fumigation applications compared to the control groups. Histological studies on the midgut and ovaries showed that *A. annua* essential oil caused an obvious change in the distribution of the principal cells of tissues and reduction in yolk spheres in oocytes. Therefore, it is suggested that the essential oil from *A. annua* flowers, with wide-range bio-effects on *G. pyloalis*, be used as an available, safe, effective insecticide in the protection of mulberry.

**Keywords:** essential oil; sweet wormwood; mulberry pyralid; mulberry; immunity; reproductive system; digestive system

## **1. Introduction**

The mulberry (*Morus* sp. (Rosales: Moraceae)) leaves are used for rearing silkworm (*Bombyx mori* L. (Lepidoptera: Bombycidae)). The importance of lesser mulberry pyralid *Glyphodes pyloalis* Walker (Lepidoptera: Pyralidae)) is from the larvae damaging mulberry leaves and the transmission of plant pathogenic agents [1]. The extensive use of synthetic chemical pesticides has led to many concerns about the safety of humans, beneficial insects, and the environment [2,3]. Thus, management of insect pest through eco-friendly and biodegradable agents is critical in sericulture.

The essential oils obtained from several parts of plants, including leaves, flowers, fruits, twigs, bark, seeds, wood, rhizomes, and roots, are made as secondary metabolites in

**Citation:** Oftadeh, M.; Sendi, J.J.; Ebadollahi, A.; Setzer, W.N.; Krutmuang, P. Mulberry Protection through Flowering-Stage Essential Oil of *Artemisia annua* against the Lesser Mulberry Pyralid, *Glyphodes pyloalis* Walker. *Foods* **2021**, *10*, 210. https://doi.org/10.3390/foods 10020210

Academic Editors: Marie-Laure Fauconnier, Haïssam Jijakli and Caroline De Clerck Received: 26 November 2020 Accepted: 19 January 2021 Published: 20 January 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/).

the plant and possess diverse chemical compositions [4]. The effectiveness of essential oils as a more sustainable pest management tool has been noted previously [5–7]. It can easily be inferred from their biodegradable nature and safety compared to many of the synthetic insecticides. Since they have multiple target sites in insects, their application is less likely to result in resistance in comparison with synthetic insecticides [8]. It was indicated that plantderived essential oils may have several effects, including ovicidal, ovipositional deterrents, feeding deterrents, growth retardants, and inhibition in detoxification enzymes [9–11].

The annual wormwood*, Artemisia annua* L. (Asterales: Asteraceae), native to temperate Asia, has been naturalized in many countries [12]. The *A. annua* has traditionally been used to treat certain diseases of humans, including asthma, fever, malaria, skin diseases, jaundice, circulatory disorders, and hemorrhoids [13]. Although our previous findings of the essential oil or extracts in the vegetative stage of *A. annua* showed the high potential of this medicinal plant species on insect pest control [14–18], the insecticidal effects of its floral essential oil were evaluated against *G. pyloalis* in the present study.

The evaluation of lethal (acute) and sublethal (chronic) effects of essential oil extracted from *A. annua* flowers on *G. pyloalis* was the main objective of the current study, recommending a biorational and available agent as a possible replacement for synthetic insecticides. Fumigant toxicity is considered to be a non-residual treatment in which no residue will commonly remain for future contaminants. In oral toxicity, the pest is eliminated by swallowing infested food, and it is a suitable method for controlling leaf-eating pests. Therefore, fumigant and oral toxicity and the effect on some key enzymes and biochemical compounds, immunology, digestive system in the larvae, and the ovary of emerged adults of insects, along with the chemical analysis of the essential oil, were evaluated.

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

#### *2.1. Insects' Rearing*

The larvae of *G. pyloalis* were handpicked from a mulberry orchard within the University of Guilan campus, Rasht (37.2682◦ N, 49.5891◦ E), Iran. The larvae were maintained on fresh leaves of 'Shin Ichinoise' mulberry variety in disposable transparent containers (high-density polyethylene plastic containers, 10 × 20 × 5 cm) in a rearing room set at 25 ± 1 ◦C, 75 ± 5% RH (Relative Humidity), and 16:8 L:D (Light:Dark). The emerging adults were reserved in glass jars (18 × 7 × 5 cm), in which fresh leaves were positioned for egg laying, and 10% honey-soaked cotton wool was provided for feeding.

## *2.2. Essential Oil*

#### 2.2.1. Extraction of the Essential Oil

The mature and immature flowers of *A. annua* (autumn 2018) were collected on the University of Guilan campus. Samples were dried on a table out of direct sunlight for about a week until sufficiently dry to form a powder when ground. The dried flowers were made into a fine powder by a grinder (354, Moulinex, Normandy, France), and a solution was made with distilled water (50 g/750 mL). The solution was let to stand in the dark at laboratory room temperature for 24 h to maximum essential oil extraction. The mixture was distilled to extract the essential oil using a Clevenger apparatus (J3230, Sina glass, Tehran, Iran). The distillation process was run for two hours and the obtained essential oil was dried over anhydrous sodium sulfate. The obtained essential oil was stored in dark glass vials at 4 ◦C in a refrigerator until used.

#### 2.2.2. Determination of Essential Oil Composition

The essential oil was analyzed through gas chromatography (Agilent Technologies 7890B) coupled with a mass spectrometer (Agilent Technologies 5977A), which was armed with an HP-5MS ((5%-phenyl)-methylpolysiloxane) capillary column with a 30-m length, 0.25-mm width, and an internal thickness of 0.25 μm. Helium gas at a 1 mL/min flow rate was used, while the column temperature started from 50 and reached to 280 ◦C at a rate of 5 ◦C/min. A 10% *A. annua* essential oil solution in methanol (*v*/*v*) was prepared, and

1 μL of solution was injected. Spectra were obtained in the electron impact mode with 70 eV of ionization energy. The scan range was between 30–600 *m*/*z*. The identification of components was performed by comparing mass spectral fragmentation patterns and retention indices with those described in the databases [19,20].

### *2.3. Insecticidal Activity*

## 2.3.1. Oral Toxicity

Initial tests were conducted to assist in selecting the appropriate range of concentrations. Bioassays were carried out on 4th instar larvae, which were deprived of nutrition for 4 h before the onset of experiments. The essential oil concentrations of 0.5, 0.7, 1, 1.4 and 2% (*W*/*V*) in acetone as solvent (Merck, Darmstadt, Germany) were selected. For bioassays, mulberry leaf disks (8 cm in diameter) were immersed in desired concentrations for 10 s and then air-dried at room temperature for 30 min. Ten 4th instar *G. pyloalis* were placed on each disk. The mortality was documented after 24 h. Control groups were placed on disks treated with acetone. The control and treated groups were replicated four times.

## 2.3.2. Fumigant Activity

In order to carry out fumigation bioassays, two transparent polyethylene plastic containers (Pharman polymer company, Rasht, Iran) were used. A 250-mL container was used to place 10 4th instar larvae of mulberry pyralid. They were provided with fresh mulberry leaf disks, and the container top was covered with fine cotton fabric for aeration. The container was then placed inside a 1000-mL container. The desired amount of pure essential oil was poured onto filter papers (Whatman No. 1) cut to 2 cm in diameter using a micro applicator. It was then placed in the corner of the larger container, and its lid tightly sealed using Parafilm. The concentrations of 2, 3, 4, 5 and 6 μL/L air were used for this bioassay based on the initial tests. The controls were treated in the same way without any treatments of the filter papers. All tests were replicated four times.

#### *2.4. Digestive Enzymes' Assays*

In order to evaluate digestive enzymes activity, the larvae that were treated with LC50, LC30, and LC10 (Lethal Concentration to kill 50, 30, and 10% of insects, respectively) dosages of essential oil obtained from oral and fumigant bioassays and the controls were dissected in ringer's solution (9% *v*/*v* NaCl and isotonic) 24 h after treatment and their digestive systems (only midguts) were dissected out. Five midguts for each treatment and control were first homogenized in 500 μL of universal buffer (50 mM sodium phosphate-borate at pH 7.1) in a tissue homogenizer (DWK885300-0001-1EA, Merk, Darmstadt, Germany). The supernatant was then kept at −20 ◦C until analyzed.

## 2.4.1. The α-Amylase Activity

The reagent dinitrosalicylic acid (DNS, Sigma, St. Louis, MI, USA) in 1% soluble starch was used to estimate α-amylase activity according to the method of Bernfeld (1955) [21]. Briefly, 20 μL of the enzyme was poured into 40 μL of soluble starch and 100 μL of universal buffer (pH 7). The mixture was incubated for 30 min at 35 ◦C, and DNS (100 μL) was then added to stop the reaction. The absorbance was read at 540 nm in an ELISA reader (Awareness, Temecula, CA, USA).

#### 2.4.2. Protease Assay

The protease activity was assessed by addition of 200 μL of casein solution casein (1%) to 100 μL of enzyme and 100 μL universal buffer (pH 7). Then, the obtained mixture was incubated at 37 ◦C for 60 min [22]. The mixture was centrifuged at 8000× *g* within 15 min and the absorbance was read at 440 nm.

## 2.4.3. Lipase Estimation

The method of Tsujita et al. (1989) [23] was adopted to estimate lipase. Concisely, 10 μL enzyme, 18 μL p-nitrophenyl butyrate (50 mM), and 172 μL universal buffer (pH 7) were mixed and incubated at 37 ◦C for 30 min. The absorbance was recorded at 405 nm in the ELISA reader.

#### 2.4.4. The α- and β-Glucosidase Estimation

Here, we used Triton X-100 in order to hydrolyze glucosidases (α- and β-) for 20 h at 40 ◦C in a ratio of 10 mg of Triton X-100/mg protein. Then, we incubated 75 mL p-nitrophenylα-D-glucopyranoside (pNaG, 5 mM), p-nitrophenyl-β-D-glucopyranoside (pNbG, 5 mM), 125 mL universal buffer (made of 2%Mol MES (2-(N-morpholino)ethanesulfonic acid), glycine, and succinate, 100 mM, pH 5.0), and 50 mL enzyme solution. In order to stop the reaction, 2 mL of sodium carbonate (1 M) was used and the absorbance was read at 450 nm [24].

#### *2.5. Detoxifying Enzymes' Assays*

Quantitative analyses of biochemical constituents were carried out on insects remaining after treatments with LC10, LC30, and LC50 and controls. To quantify the whole body protein, the method of Bradford (1976) [25], using the kit (GDA01A, Biochem Co., Tehran, Iran), was incorporated, while glucose and triglyceride were measured by Siegert (1987) [26] method and the triglyceride diagnostic kit, respectively (Pars Azmoon Co., Tehran, Iran). Key enzymes including esterase (general esterases with α- and β-naphthyl acetate substrates), glutathione S-transferase (GST), and phenol oxidase (PO) were assessed by the method described by van Asperen (1962) [27], Habing et al. (1974) [28], and Parkinson and Weaver (1999) [29], respectively.

## *2.6. Hematological Study*

The amount of various circulating blood cells in mm−<sup>3</sup> of larval lesser mulberry pyralid treated with sublethal doses of *A. annua* oil and in controls were assessed. The hemolymph was drawn from one of the larval prolegs, cutting by a fine scissor, using a capillary glass tube (10 μL for each treatment). Then, the blood was diluted five times with a solution of anticoagulant (0.017 M EDTA, 0.186 M NaCl, 0.098 M NaOH, and 0.041 M citric acid at pH 4.5). An improved Neubauer hemocytometer (mlabs, HBG, Giessen, Germany) [30] was used to assess the total cells using the formula of Jones (1962) [31]. A drop of hemolymph was collected from cut proleg of treated and control larvae. A smear was formed and stained with diluted Giemsa (Merck, Darmstadt, Germany) in distilled water (1:9) for 25 min, then just dipped in a saturated solution of lithium carbonate, and, finally, washed with distilled water. Permanent slides were prepared in Canada balsam (Merck Darmstadt, Germany). The percentage profile of different cells was done after identification and counting of 200 cells per slide [32].

## Immunity Responses

Initially the treated or control larvae were made immobile by keeping them on ice cubes for five minutes. Then, they were surface sterilized and injected with <sup>1</sup> × 104 spores/mL in 0.01% Tween-80 of *Beauveria bassiana* (IRAN403C isolate) or latex beads (1:10 dilution for each suspension and Tween-80, respectively) on the second abdominal sternum using a 10-μL Hamilton syringe. The treated larvae were then transferred to glass jars and were given fresh leaves of mulberry. The control larvae were injected with 1 μL of distilled water comprising 0.01% of Tween-80 only. The hemolymph was collected 24 h post-injection from each larva, and the number of nodules formed was scored in a hemocytometer [33]. The counting was repeated four times for each group.

#### *2.7. Histological Studies of Larvae Midgut and Adults' Ovary*

The larvae midguts were separated from the whole dissected gut in insect ringer and were immediately fixed in aqueous Buine solution for 24 h [10]. Also, the ovary of adults (2 days old), emerging from either treated or control larvae, were separated and fixed. The tissues were processed for embedding in paraffin after being dehydrated in grades of ethanol alcohol and also cleaned by xylene. The fixed tissues were then cut by 5-μM thickness through a rotary microtome (Model 2030; Leica, Wetzlar, Germany). The hematoxylin and eosin were used for staining and then permanent slides were thus prepared, observed, and photographed under a light microscope (M1000 light microscope; Leica, Wetzlar, Germany) armed with an EOS 600D digital camera (Canon, Tokyo, Japan).

#### *2.8. Statistical Analysis*

LC values were determined using the Polo-Plus software (2002) [34]. All the data were analyzed by ANOVA (SAS Institute, Cary, Cary, NC, USA, 1997) [35], and the comparison of means was performed using Tukey's multiple comparison test (*p* < 0.05).

#### **3. Results**

## *3.1. A. annua Essential Oil Analysis*

The chemical composition of extracted *A. annua* essential oil is presented in Table 1. We identified 55 compounds in flowers of this plant, which represent 93.0% of the total composition. Camphor (13.1%), artemisia ketone (11.8%), β-selinene (10.7%), pinocarvone (7.4%), 1,8-cineole (6.8%), and α-pinene (5.9%) were considered as the major compounds detected, all of which are terpenes. However, other groups such as ester and phenylpropene were also recognized (Table 1).


RIcalc = retention index determined with respect to a homologous series of n-alkanes on a HP-5 ms column; RIdb = retention index from the databases [19,20]; MH = monoterpene hydrocarbone; OM = oxygenated monoterpene; SH = sesquiterpene hydrocarbone; OS = oxygenated sesquiterpene; DT = diterpene; PP = phenylpropene; E = ester; OC = other components.

## *3.2. Insecticidal Activity*

Based on oral and fumigant bioassays, *A. annua* essential oil was toxic to 4th instar larva of *G. pyloalis* 24 h post treatments. Probit analysis revealed that the LC50 values were 1.204 % W/V and 3.343 μL/L air for oral and fumigant toxicity, respectively. The mortality of tested larvae was augmented with increasing concentration (Table 2). Besides LC50, the LC10 and LC30 values were used to evaluate sublethal bio-activities, including effects on energy reserves, digestive and detoxifying enzymes activity, and hematological and immunity responses and histological study of midgut and ovary of larvae (Table 2).

**Table 2.** Probit analysis of the oral and fumigant toxicity of *Artemisia annua* floral essential oil on 4th instar larva of *Glyphodes pyloalis*.


LC: lethal concentration (% W/V for oral toxicity and μL/L for fumigant toxicity), CL: confidence limits, X2: Chi-square value, and df: degrees of freedom. According to Chi-square values, no heterogeneity factor was used in the calculation of confidence limits. Concentration rates were 0.5–2% (W/V) and 2–6 μL/L air for oral and fumigant toxicity, respectively.

### *3.3. Energy Reserves*

The essential oil of *A. annua* flowers on the energy reserves of *G. pyloalis* larvae is shown in Table 3. As can be seen, for all macromolecules, increasing dose of essential oil decreased the concentrations of protein, glucose, and triglycerides. For example, doubling the essential oil concentration (LC10 to LC50) reduced glucose by 29% in oral tests, while a 1.7-fold increase in fumigant concentration resulted in a 32% drop in glucose levels. The protein was also affected but the decrease in protein with increasing essential oil levels was insufficient to detect given background variability.



In each separate column, means followed by different letters designate significant differences at *p* < 0.05 according to Tukey's test.

#### *3.4. Digestive and Detoxifying Enzymes*

The effects of *A. annua* floral essential oil on digestive enzymes' activity of *G. pyloalis* larvae was manifested by a decrease in protease, α-glucosidase, β-glucosidase, α-amylase, and lipase contents. The difference was significant between the LC50 versus the control in both oral and fumigant applications while other concentrations of the essential oil produced intermediate responses (Table 4).

The effect of essential oil of *A. annua* flowers on the activity of esterase and glutathione S-transferase (GST) of *G. pyloalis* larvae is shown in the Table 5. Glutathione S-transferase and esterase contents were reduced significantly when LC50 was applied in both oral and fumigation methods compared to the controls (Table 5).

### *3.5. Hematological Study and Immunity Responses*

The essential oil affected the immune system, which included cellular quantity and quality, phenol oxidase activity, and the immune responses after *B. bassiana* and latex beads' injection (Figures 1–4). Total hemocyte counts (THC), plasmatocytes and granular cells, nodule formation, and phenol oxidase activity was recorded the lowest in LC50 both in oral and fumigation assays, respectively*.*

**Figure 1.** The effect of *Artemisia annua* floral essential oil on total hemocyte counts (THC) of *Glyphodes pyloalis* larvae treated with oral (**A**) and fumigant (**B**) assays. Bars with different letters above them indicate significant differences between means at *<sup>p</sup>* < 0.05, Tukey's test. Number of hemocytes <sup>×</sup>104.

**Figure 2.** The effect of *Artemisia annua* floral essential oil on the plasmatocytes and granular cells of *Glyphodes pyloalis* larvae treated with oral (**A**) and fumigant (**B**) assays. Bars with different letters indicate significant differences among means of each hemocyte at *<sup>p</sup>* < 0.05, Tukey's test. The number of hemocytes <sup>×</sup>104.




**Table 5.** Effect of the different concentrations of *Artemisia annua* flowers' essential oil on the activity of glutathione S-transferase (GST) and esterase in 4th instar larvae of *Glyphodes pyloalis*.

In each separate column, means followed by different letters indicate significant differences at *p* < 0.05 according to Tukey's test.

**Figure 3.** Effects of *Artemisia annua* floral essential oil on the nodule formation of *Glyphodes pyloalis* larvae treated with oral (**A**) and fumigant assays (**B**) and inoculated with *Beauveria bassiana* spores or latex beads. Bars with different letters indicate significant differences between means at *<sup>p</sup>* < 0.05. Tukey's test. The number of hemocytes <sup>×</sup>104.

**Figure 4.** The effect of *Artemisia annua* floral essential oil on phenol oxidase (PO) activity of *Glyphodes pyloalis* larvae treated with oral (**A**) and fumigant (**B**) assays. Bars with different letters above them indicate significant differences between means at *<sup>p</sup>* < 0.05, Tukey's test. The number of hemocytes <sup>×</sup>104.

## *3.6. Histological Studies*

The histological texture of larval midgut upon treatment with *A. annua* essential oil revealed significant differences with the controls, the most significant of which was the elongation and separation of epithelial cells losing the compactness (Figure 5). The most significant changes in ovarian structure was thinning of epithelial cells around each follicle compared with that of control. Also, the significant reduction in cytoplasm was seen after vacuolization in yolk spheres of the oocytes (Figure 6).

**Figure 5.** Light microscopy of the larval midgut of *Glyphodes pyloalis* in control (**a**) and after oral treatment with *Artemisia annua* floral essential oil (**b**). Normal texture of all cell types (**a**) was contrasted to changes in size and texture in treated larvae (**b**). In the midgut of insects treated with essential oil from *A. annua* the cohesion of the columnar epithelial layer was damaged. (**BM**) basement membrane, (**CC**) columnar cell, (**GC**) goblet cell, and (**PM**) peritrophic membrane.

**Figure 6.** Histology of ovaries in adults of *Glyphodes pyloalis* emerging from untreated (**a**) and treated larvae by *Artemisia annua* floral essential oil (**b**). In treatments of the ovarian sheath significant changes and yolk granules were reduced under the influence of vacuolization in cytoplasm compared to the control. (**FE**) follicular epithelium, (**V**) vacuole, and (**Y**) yolk granules.

#### **4. Discussion**

The chemical composition of *A. annua* essential oil in the vegetative stage was investigated in the previous studies [15,36–39], in which terpenes such as 1,8-cineole, camphor, and artemisia ketone were introduced as major constituents. Although 1,8-cineole (6.8%), camphor (13.1%), and artemisia ketone (11.8%) were also identified as main compounds in the essential oil extracted from *A. annua* flowers, some other terpenes such as β-selinene (10.7%), pinocarvone (7.4%), and α-pinene (5.9%) had high amounts. However, a range of minor constituents, including compounds from ester and phenylpropene groups, were also recognized. Such differences can be caused by exogenous and endogenous factors, including geographic location, harvesting time, and the growth stage of plants [40]. The chemical composition of each essential oil has a significant impact on its insecticidal activity. For example, the promising insecticidal effects of terpenes like camphor and 1,8-cineole identified and extracted from essential oils were reported [41,42].

Our study clearly showed decreased enzymatic activity in *G. pyloalis* larvae related to ingestion of *A. annua* essential oil-treated mulberry leaves. Our findings support earlier findings where disruption in insects' physiology and their inability to digest food was reported [43,44]. Reduction in α-amylase, protease, and α- and β-glucosidase, and disruptions on immunology and digestive system in the larvae and the ovary of emerged adults of *G. pyloalis* were described in our results. Such activities are common for botanical

insecticides against several insect pests [45–47]. Also, there were further supports for the interference or even deformation of midgut cells, which were responsible for the production of key enzymes in insects [15,48].

Protein plays a key role in digestion, metabolism, and also energy conversion. Klowden (2007) [49] believes that reduction in the insect's protein content after applying biopesticides may stem from the reduction of growth hormone level. We observed a reduction in protein content and also retardation in growth; however, growth hormone level was not worked out. Lipids are other important macromolecules that help the insect reserve energy from feeding. They play a key role in insects' intermediary metabolism and, therefore, they are essential in insect physiology [49]. Significant reduction in the triglyceride content of *G. pyloalis* larvae treated with *A. annua* essential oil was observed in the present study. There are several reasons for reducing insect lipid content after treatments by toxins, alteration in lipid synthesis patterns, and hormonal dysfunction to control its metabolism [49]. Glucose as a key carbohydrate (monosaccharide) was also decreased following treatment with *A. annua* essential oil. This reduction could be related to reduced feeding following treatment, since the essential oil acts as a deterrent [2]. Any disruption causing reducing resources at larval stages could affect insects' survival and reproduction in their later generations. A reduction in protein, lipid, and glucose contents may have adverse effects on the reproductive parameters such as egg production, fertility, and fecundity [50].

Detoxifying enzymes, including esterases and glutathione *S*-transferases, are involved in reducing the impacts of exogenous compounds [51]. In the current study, the activity of detoxifying enzymes, including esterases and glutathione *S*-transferases, was reduced by essential oil of *A. annua* flowers. Certainly, the reduced activity of these enzymes is related to their production halt somewhere in the process of production [15].

Insect cellular immunity is considered as the main system challenging natural enemies entering the insect body [52]. The immunocytes provide the insect ability to combat invading organisms by several means including phagocytosis, nodulation, and encapsulation [53]. So, the reduced immunocytes, as shown for *G. pyloalis* larvae treated with *A. annua* essential oil in the present study, could cause larvae to become susceptible to any invasion [54,55]. The reduced number of hemocytes is mostly due to cytotoxic effect of the botanicals used [56]. We do believe this toxic effect of botanicals to be more reliable as a reasoning for the reduction of immunocytes [57–59].

Phenol oxidase system is considered as the key component in the immune system of insect and a bridge in the gap between cellular and humeral insect immunity. Its action is critically required in the last stage of cellular defense in order to form melanization, a process that terminates the action and kills the pathogenic agent. Phenol oxidase inhibition, documented for *G. pyloalis* larvae treated with *A. annua* essential oil in the present study, probably helps to make the insects susceptible to pathogenic agents if they have not received the toxic concentration [45,58,60].

The insect midgut principal cells are the main cells taking the role of producing the enzymes needed for digestion and then absorbing the nutrients. Therefore, any damages to these cells will lower the activities in digestive enzymes already reported by other researchers [15,31,61]. The elongation and separation of midgut epithelial cells of *G. pyloalis* larvae treated by *A. annua* essential oil were observed in the present study.

Inhibiting insect reproduction has long been the subject of many studies. In lepidopterans, obtaining all nutrients at larval stages is necessary for reproductive development [62]. So, if larval nutrition is disrupted by any means, it will be reflected in adult reproductive function. Our previous findings and the current study display the changes in morphology and histology of emerging adults [15,31]. Our study showed the essential oil of *A. annua* brought about subtle changes in ovarian tissue, such as disruption of follicular cells. As the insect tries to compromise to reduce nutrients in detoxification processes, follicles' cells deplete its content into the oocytes, which then disrupts the cell texture [63].

## **5. Conclusions**

Plant-derived allelochemicals are beneficial agents in controlling pests. As we know, the plant kingdom mainly depends on secondary metabolites to defend against herbivores. With this knowledge in mind, scientists exploit the use of secondary plant chemicals for pest control. One of the main reasons for this increased demand is that the plant-originated chemicals are comparatively safer for humans and the environment. Our study's results clearly document that the essential oil of *A. annua* flowers is toxic to larval mulberry pyralid and disrupt its various physiological systems in a way that the insect can hardly get resistance to it. Consequently, this wild-growing plant in Iran can be considered an efficient natural source capable of controlling insect pests. To apply the research results, it is recommended to evaluate the possible side effects of essential oil on mulberry and the biological control agents in future research. Regarding the insect pest's resistance, identifying specific modes of action of essential oil active components and their overlapping with other insecticides should also be assessed.

**Author Contributions:** Conceptualization, M.O., J.J.S., and A.E.; methodology, M.O. and J.J.S.; formal analysis, M.O., J.J.S., A.E., and W.N.S.; investigation, M.O.; writing—original draft preparation, M.O., J.J.S., A.E. and P.K.; writing—review and editing, M.O., J.J.S., A.E., P.K. and W.N.S.; supervision, J.J.S. and A.E.; funding acquisition, P.K. All authors have read and agreed to the published version of the manuscript.

**Funding:** This research received no external funding.

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

**Informed Consent Statement:** Not applicable.

**Data Availability Statement:** The data that support the findings of this study are available upon request from the authors.

**Acknowledgments:** This research was financially supported by the University of Guilan, Rasht, Iran, and was partially supported by Chiang Mai University, Thailand, which is greatly appreciated. W.N.S. participated in this work as part of the activities of the Aromatic Plant Research Center (APRC, https://aromaticplant.org/).

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

#### **References**


## **Use of New Glycerol-Based Dendrimers for Essential Oils Encapsulation: Optimization of Stirring Time and Rate Using a Plackett—Burman Design and a Surface Response Methodology**

**Chloë Maes 1,2,\*, Yves Brostaux 3, Sandrine Bouquillon 1,† and Marie-Laure Fauconnier 2,†**


**Abstract:** Essential oils are used in an increasing number of applications including biopesticides. Their volatility minimizes the risk of residue but can also be a constraint if the release is rapid and uncontrolled. Solutions allowing the encapsulation of essential oils are therefore strongly researched. In this study, essential oils encapsulation was carried out within dendrimers to control their volatility. Indeed, a spontaneous complexation occurs in a solution of dendrimers with essential oils which maintains it longer. Six parameters (temperature, stirring rate, relative concentration, solvent volume, stirring time, and pH) of this reaction has been optimized by two steps: first a screening of the parameters that influence the encapsulation with a Plackett–Burmann design the most followed by an optimization of those ones by a surface response methodology. In this study, two essential oils with herbicide properties were used: the essential oils of *Cinnamomum zeylanicum* Blume and *Cymbopogon winterianus* Jowitt; and four biosourced dendrimers: glycerodendrimers derived from polypropylenimine and polyamidoamine, a glyceroclikdendrimer, and a glyceroladendrimer. Metaanalysis of all Plackett–Burman assays determined that rate and stirring time were effective on the retention rate thereby these parameters were used for the surface response methodology part. Each combination gives a different optimum depending on the structure of these molecules.

**Keywords:** essential oil; encapsulation; controlled release; biosourced; surface response methodology

## **1. Introduction**

For the last 70 years, industrial countries intensively used chemical pesticides in order to increase agricultural yields to feed a constantly growing population. Unfortunately, with time passing, controversies and the knowledge about their harmful effects on human health and environment have blown up quickly [1]. In this context, biopesticides are priceless candidates to create new weeds- and crops-managing strategies. Among natural compounds from plant origin, essential oils (EOs) are increasingly used for their various biological properties [2,3].

Essential oils are natural mixtures of volatile compounds frequently used in cosmetics, perfume, and sanitary products for both their fragrance and biological activities [4–7]. Another principal characteristic of EOs is their volatility, which limits residues after treatment. Unfortunately this can be a constraint for their utilization as biopesticide because their spread is not controlled [8]. To counter this, scientists developed several different encapsulation techniques. Depending on their properties, emulsion, coacervation, spray drying,

**Citation:** Maes, C.; Brostaux, Y.; Bouquillon, S.; Fauconnier, M.-L. Use of New Glycerol-Based Dendrimers for Essential Oils Encapsulation: Optimization of Stirring Time and Rate Using a Plackett—Burman Design and a Surface Response Methodology. *Foods* **2021**, *10*, 207. https://doi.org/10.3390/foods10020207

Academic Editor: Qin Wang Received: 18 December 2020 Accepted: 18 January 2021 Published: 20 January 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/).

complexation, ionic gelation, and nanoprecipitation help maintain a controlled release of EOs, either quick or slow [9]. EOs encapsulation may appear useless to enhance herbicidal activities on plants, because shoot death occurs after 1 h to 1 day of application [10]. However, an actual interest exists for the improvement of the seed's germination inhibition effects because this one occurs for longer periods (up to 30 days) thus EO encapsulation with controlled release allows to use a lower concentration. Lethal dose depends on the target plant/seed [11].

Cinnamon and Java citronella essential oils are of particular interest for herbicidal applications in a context where the replacement of conventional herbicides is increasingly wanted [12–14]. In a previous study [12], we determined that the major constituents in cinnamon essential oil are trans-cinnamaldehyde (70%) followed by eugenol, caryophyllene, cinnamyl acetate, and linalool in decreasing concentration order. Java citronella EO is constituted of 57 different molecules; among them citronellal (40%), geraniol (20%), citronellol (15%), limonene (5%), and eugenol (2%) are the main representatives [12,15,16]. The modes of action of the main constituents of these EOs as herbicides are not fully characterized but their interaction with respectively the lipid and protein fraction of the plant plasma membrane might be involved [12].

In the present research, glycerol-based dendrimers (GDs) are proposed as new and original matrix to encapsulate EOs. GDs are macromolecules synthesized from glycerol carbonate (a side product from biofuel production) which already showed good encapsulation ability of contrast agent for medical sectors, metals (nanoparticles), and organic pollutants of used water. Indeed, their tree structure allows intern cavities (Figure 1), from various sizes depending of the dendrimer generation, to retain molecules [17–20]. Glyceroclikdendrimer (GAD) and glyceroladendrimer (GCD) have been recently developed and described in two patents with specific encapsulation abilities toward organic pollutants and metallic salts [21,22]. Beyond the agronomic field, EOs encapsulation within dendrimers can be used in a wide range of applications, including food industry (active packaging) and pharmaceutical (drug delivery system) through their bactericidal, viricidal, and fungicidal activities [23,24].

**Figure 1.** Structures of dendrimers: (**A**) Glycerodendrimers polypropylenimine 3rd generation (GD-PPI-3). (**B**) Glycerodendrimers polyamidoamine 2nd generation (GD-PAMAM). (**C**) Glyceroclikdendrimers 2nd generation (GCD-2). (**D**) Glyceroladendrimers 1st generation (GAD).

The goal of this study is to optimize the encapsulation reaction of two essential oils by four selected dendrimers by maximizing the retention of two GDs, a GCD and a GAD using a Plackett–Burman design (PBD) and response surface methodology (RSM) in order to eventually create an effective biosourced herbicide or for other applications where a slow release of EOs is required. PBDs are a screening design that takes into account a large number of factors with a minimal number of trials, while RSMs are an experimental design intended to optimize factors and their combinations [25]. Obviously, since this study highlights the statistical optimization of the encapsulation, these results can be applied in other fields cited before such as food preservatives creations [26].

## **2. Materials and Methods**

## *2.1. Chemicals and Reagents*

The essential oils of *Cinnamomum zeylanicum* Blume bark (Cinnamon, CAN) and *Cymbopogon winterianus* Jowitt leaves (citronella, CIT) were purchased from Pranarom (Belgium).

Glycerodendrimers-polypropilenImine (GD-PPI) and glycerodendrimers-polyamidoa mine (GD-PAMAM) were synthesized according the previously described work related to the decoration of dendrimers [17,18].

GlycerolADendrimers (GAD) and GlyceroClickDendrimers (GCD) were synthesized following the procedures described in two patents [21,22].

## *2.2. Essential Oils Encapsulation*

Essential oils encapsulation take place by a spontaneous complexation; the dendrimers were dissolved in H2O (8 mL) and EOs were dissolved in ethanol (various concentrations). EOs solutions or pure ethanol was added to dendrimers solution (3/1 *v*/*v*) in a 22 mL glass vial which was directly hermetically sealed with a Teflon cap and covered with an aluminum foil to avoid light interference. Solutions were then stirred for at least 10 min at 100 rpm. According on the stirring settings, an emulsion of EOs occurs in the dendrimer solution, which provides a liquid phase EOs retention. This retention leads to a change in dynamic balance between solution and headspace compared to free EO solution (control), which is quantified by the following analysis.

#### *2.3. Dynamic-Headspace Gas Chromatography–Mass Spectrometry (DHS-GC–MS) Analysis*

The percentage of retention (r) of EOs by GDs was determined by dynamic head- space sampling (DHS, Gerstel, Germany) coupled to a thermal desorption unit (TDU, Gerstel, Germany), a gas chromatograph (Agilent Technologies 7890A), and a mass spectrometer (MS, Agilent Technologies 5975C). During treatment in the DHS unit, the vials were conditioned at 25 ◦C for 30 min with stirring (500 rpm). The head-space sampling was performed on Gerstel TDU desorption tubes (OD 6.00 mm, filled with 60 mg of Tenax TA, Gerstel, Germany), on 200 mL at 20 mL/min, followed by 200 mL at 60 mL/min of drying phase. Desorption then occurred for 10 min at 300 ◦C and coupled to a cooled injection system (CIS, Gerstel, Germany) set at −80 ◦C. EOs were then transferred to the GC column (VF-WAXms, Agilent technologies USA; 30 m length, 0.250 mm I.D, 0.25 l m film thickness) for separation with temperature program as follow: Java citronella—from 70 ◦C (5 min) to 100 ◦C at a rate of 8 ◦C/min, then 2 ◦C/min to 160 ◦C, and then 20 ◦C/min to 260 ◦C (10 min); Cinnamon—from 40 ◦C (4 min) to 80 ◦C at a rate 3.5 ◦C/min, then 5 ◦C/min to 160 ◦C, and then 20 ◦C/min to 220 ◦C (10 min) with helium as carrier gas at a flow rate of 1.5 mL/min. The MS were recorded in electron ionization mode at 70 eV (scanned mass range: 35 to 300 m/z); source and quadrupole temperature at 230 ◦C and 150 ◦C respectively. The component identification was performed by comparison of the recorded spectra with two data libraries (Pal 600K® and Wiley275) and injection of pure commercial standards in the same chromatographic conditions.

The percentage of retention (r) of EOs by GDs was calculated by the equation [27]:

$$r(\%) = \left(1 - \frac{\sum A\_D}{\sum A\_0}\right) \times 100\tag{1}$$

∑*AD*: sum of peak areas of EO component in the presence of dendrimers, ∑*A*0: sum of peak areas of EO component in free EO solution (control).

#### *2.4. Screening of Six Encapsulation Parameters with Plackett–Burman Design*

Plackett–Burman design was used to select the significant parameters for essential oils encapsulation. This design was applied to four combinations of dendrimers and EOs previously selected owing to their noticeable essential oil retention capacity (preliminary assays, data not shown but published soon). The combinations are: GD-PPI-3/CAN EO, GD-PAMAM-2/CIT EO, GAD-1/CAN EO, GCD-1/CIT EO. The independent parameters were set on the basis of those preliminary analyses, which considered the properties of the dendrimers for relative concentration and pH, the technical feasibility for rate of stirring, the solvent volume, and stirring time and the temperature which can be found in realistic agronomical conditions.

For each combination, a 12-run PBD was applied to evaluate six factors. Each variable was examined at two levels: –1 for the low level and +1 for the high level. Table 1 illustrates these parameters and the corresponding levels used. The values of two levels were set according to our previous preliminary experimental results. In Table 2, representing PBD and experimental results, data listed indicate the variations in retention rate between each combination of dendrimers-Eos, depending on the treatment. Negative values indicated that the opposite effect is observed: presence of dendrimers increase the volatility of EOs.

**Table 1.** Factors and their levels selected for the Plackett–Burman design.


**Table 2.** Experimental setting (12 runs) generated by Minitab® 19 and retention rate for the fourth combinations of dendrimers and essential oils (Eos) (%, experimental).


#### *2.5. Optimization of Two Encapsulation Parameters by Response Surface Methodology*

Based on the results of the PDB design, only the most influential parameters on the encapsulation reaction have been selected for further optimization through response surface methodology. Experiments were performed according to a design with two parameters and three levels for each parameter [25]. Two blocks have been used to cover the potential

heterogeneity during the course of the experiment. The selected independent variables were stirring rate (R) and stirring duration (D). The experimental design in the actual levels is shown in Table 3. As for PBD, variations in retention rate between each couple dendrimers-EOs were recorded. In RSM experimental results (Table 4), negative percentage of retention notifies an increase in EOs volatility in presence of dendrimers. Maximums were represented with contour plots.

**Table 3.** Factors and their levels selected for the Box–Behnken design (response surface methodology).


**Table 4.** Experimental setting (28 runs) generated by Minitab® 19 and retention rate for the fourth combinations of dendrimers and EOs (%, experimental).


## *2.6. Data Analysis*

PBD and RSM were designed and processed using Minitab® 19 software [25].

#### **3. Results and Discussion**

*3.1. Volatiles Profiles and Major Components of EOs*

Chromatograms obtained by DHS-GS-MS for encapsulation optimizations show the volatile profiles of both EOs in Figures 2 and 3. Major compounds have been identified as it was previously mentioned [12]. On these figures, chromatograms of control and encapsulation solutions are overlaid which show that the only difference found is in the height (and peak area) of all compounds. Therefore, profiles were similar in the presence and absence of dendrimers. A thorough examination of the retention rate of each compound in Table 5 allows to observe that chemical structures and volumes of the major components of cinnamon EOs (volumes from 210 to 377 Å3) are more variable than in citronella EOs (volumes from 270 to 303 Å3), which seems to affect somewhat the profile (12% retention rate variations between eugenol and β-caryophyllene)

**Figure 3.** Overlaying of chromatographic analysis of free Java citronella EO (control) and Java citronella EO encapsulated within GD-PAMAM-2 under optimized conditions—(1) ethanol (sample solvent), (2) limonene, (3) citronellal, (4) linalool, (5) β-citronellol, (6) geraniol.


**Table 5.** Chemical structures and calculated molecular volumes of the major compounds of cinnamon and Java citronella EOs; and their individual retention rate in the optimized encapsulation within dendrimers.

\* V = M/dNA with M: molecular weight; d: density; NA: Avogadro's number [27].

## *3.2. Influence of Parameters with PBD*

In the present study, the dendrimer/EOs complexes were successfully prepared by a simple solubilization and stirring in controlled conditions. To minimize the experimental runs and times for the screening of the encapsulation parameters, the PBD was applied on the basis of two coded levels of the six independent variables, resulting in twelve experiments (Table 2).

Analysis of PBD has been done for each couple dendrimer/EO (Table 6) which showed that almost no one had a variable influencing significantly the encapsulation rate (*p* < 0.05). However, the meta-analysis of all results and a particular attention at the ranking of variables show that time and rate of stirring appeared important in the encapsulation process. Considering that, it seems the lack of significance of these results reveals that the influence had been attenuated by the variability among repetitions in the manipulations. Both parameters (duration and rate of stirring) were selected for further optimization both with RSM.

## *3.3. Rate and Duration Stirring Optimization with Response Surface Methodology* 3.3.1. GD-PPI-3/CAN

For the first studied combination of dendrimer/EO, initially settled parameters were not optimal to find a maximum (Figure 4A) so new ones were defined in Table 7. Figure 5A shows that the model with those parameters was significant, with F-value equal to 10.34 and *p*-value < 0.001. Despite a slight rejection of the lack-of-fit test (*p* = 0.022) the applied model presented a good fitting to the encapsulation efficiency response (Figure 5B).


**Table 6.** Analyses of variance (ANOVA) of Plackett–Burman screening design experiments.

**Figure 4.** Contour plots showing the crossed effect of duration (D) and rate of stirring (R) on the retention rate (r) of cinnamon essential oil by GD-PPI-3 with the first sets of parameters (**A**) and the second one (**B**).


**Table 7.** Factors and their levels selected for the second assay of Box–Behnken design (response surface methodology) for the GD-PPI-3/CAN EO encapsulation.

**Figure 5.** Analysis of variance (ANOVA) for the response surface methodology (RSM) (**A**) and normal probability plot of the residuals of GD-PPI−3/CAN EO (2) (**B**).

As the model is trustworthy, we can focus on the influence and optimization of factors. Linear and square of each parameter were significant (*p*-value < 0.05), so they were both influencing the encapsulation rate following the curves independently because their interaction (D\*R) was not significant (*p*-value = 0.245). The regression equation describing these mathematical relationships is:

$$f(\mathbf{r}) = 22.6 + 6.30 \,\mathrm{D} + 4.12 \,\mathrm{R} - 7.08 \,\mathrm{D}^2 - 5.49 \,\mathrm{R}^2 + 2.23 \,\mathrm{D} \times \mathrm{R} \tag{2}$$

Contour plot present in Figure 4B illustrates the level of parameters that allowed to reach the maximum of retention (>20%) which can be found with a stirring time between 240 and 420 min at a rate between 1500 and 2000 rpm.

## 3.3.2. GD-PAMAM-2/CIT

Second studied combination of dendrimer/EO showed that the model was significant with an F-value of 6.07 and *p*-value is 0.001 (Figure 6A). In addition, Figure 6B revealed a good correspondence between the linear regression model of RSM and the experimental data despite a slight rejection of the lack-of-lit test (*p*-value = 0.011). As for the first combination, linear and square of each parameter were significant but not their respective interaction. The regression equation describing these mathematical relationships is:

$$\mathbf{r}(\mathbf{r}) = 13.03 - 3.54 \,\mathrm{D} - 6.43 \,\mathrm{R} - 3.69 \,\mathrm{D}^2 - 3.45 \,\mathrm{R}^2 + 3.40 \,\mathrm{D} \times \mathrm{R} \tag{3}$$

**Figure 6.** Analysis of variance (ANOVA) for the RSM (**A**) and normal probability plot of the residuals of GD-PAMAM-2/CIT EO (**B**).

Contour plot present in Figure 7. illustrates that a stirring during between 10 and 60 min at a rate between 150 and 1000 rpm allowed to reach the maximum of retention (>15%).

**Figure 7.** Contour plots showing the crossed effect of duration (D) and rate of stirring (R) on the retention rate (r) of citronella essential oil by GD-PAMAM-2.

## 3.3.3. GAD-1/CAN

Third studied combination of dendrimer/EO showed that the model is significant with an F-value of 7.06 and *p*-value < 0.001 (Figure 8A) and the lack-of-lit is non-significant (*p*-value = 0.645). In addition, Figure 8B reveals a good correspondence between the linear regression model of RSM and the experimental data. Linear and square of only the duration of stirring are significant (*p*-value of R is 0.175 and R2 is 0.258) and influence the encapsulation rate following the curves. The regression equation describing these mathematical relationships is:

$$\mathbf{r}(\mathbf{r}) = \begin{bmatrix} 0.93 \ -7.98 \ \mathbf{D} \ + \ 1.92 \ \mathbf{R} \ -5.56 \ \mathbf{D}^2 \ -1.66 \ \mathbf{R}^2 \ -1.79 \ \mathbf{D} \times \mathbf{R} \end{bmatrix} \tag{4}$$

**Figure 8.** Analysis of variance (ANOVA) for the RSM (**A**) and normal probability plot of the residuals of GAD-1/CAN EO (**B**).

Contour plot present in Figure 9 illustrates the level of parameters that allow to reach the maximum of retention even if this one is very low (>5%). The best results can be found with a stirring time between 10 and 30 min at a rate between 1500 and 2000 rpm.

**Figure 9.** Contour plots showing the crossed effect of duration (D) and rate of stirring (R) on the retention rate (r) of cinnamon essential oil by GAD-1.

### 3.3.4. GCD-2/CIT

The last studied combination of dendrimer/EO showed that the model is significant with an F-value of 4.17 and *p*-value = 0.005 (Figure 10A) however, lack-of-lit is rejected with a *p*-value equal to 0.003 so results have to be discussed. Nevertheless, Figure 10B reveals a good correspondence between the linear regression model of RSM and experimental data which confirms the global correctness of the model. Only the linear effect rate of stirring was significant (*p*-value = 0.240) and the square effect of both parameters were significant. The regression equation describing these mathematical relationships is:

$$\mathbf{r}(\mathbf{r}) = \mathbf{8.62 + 3.55D + 7.41R - 11.65D^2 - 6.77R^2 - 2.04D \times R} \tag{5}$$

**Figure 10.** Analysis of variance (ANOVA) for the RSM (**A**) and normal probability plot of the residuals of GCD-2/CIT EO (**B**).

Contour plot present in Figure 11 illustrates the level of parameters that allowed to reach the maximum of retention (>10%) which was found with a stirring during around 60 min at a rate of 1500 rpm.

**Figure 11.** Contour plots showing the crossed effect of duration (D) and rate of stirring (R) on the retention rate (r) of citronella essential oil by GCD-2.

## **4. Conclusions**

For the first time, essential oils encapsulation by bio-sourced dendrimers was successfully carried out, and this reaction was optimized using PBD and RSM. The first part proved that only the rate and the time of stirring influenced the retention rate among the six factors analyzed. The second part optimizes both factors for each couple dendrimer/EO and resulted in very different results. This is quite understandable considering the apolar nature of the EOs' constituents and the differences of structure between dendrimers. Indeed, we can see in Figure 1 that even if all dendrimers contain glycerol or glycerol derivatives in the intern structure or on the periphery of the dendrimer, and a polar surface, the properties of the cores are different. On one point, the core of GD-PAMAM-2 is more polar than the GD-PPI-3 s one; on another point, some have strong steric hindrance and important electronic charge (GCD-2) while others are less energy-intensive (GAD-1). Previous study about encapsulation by dendrimers showed that the hydrodynamic radius of GD-PPI and GD-PAMAM influenced the encapsulation and that one occurred at the core level of dendrimers rather than at its periphery. Metal complexes were successfully encapsulated in the fourth and fifth generation of GD-PPI (around 25% of encapsulation rate), but not in the third probably because this one had a smaller hydrodynamic radius (2.81 nm) [20]. Organic compounds as β-estradiol, atrazine, diclofenac salt, or diuron have been also encapsulated in GD-PPI-4 and GD-PAMAM-3 up to 95% [18]. As the transcinnamaldehyde (Table 5), one of the major compounds responsible of herbicidal activity, is a smaller molecule than the previous encapsulated ones, it seems obvious that smaller dendrimer generations give here the best results for its encapsulation. Furthermore, this α,β-unsaturated aldehyde presents an important electronic density as the previous organic compounds used. It must be pointed out that chromatographic profiles were similar, for EOs encapsulated in dendrimers or not (control) which suggests that all compounds of each EOs were encapsulated in the same way (Figures 2 and 3). It can be concluded that first the size of molecules encapsulated in comparison with size of intern cavities of dendrimers, and secondly the amount of free electron in the EOs (aromatic circle and double bonds promotes electrostatic interactions) appear to be principal factors influencing the EOs encapsulation within dendrimers [28].

In the optimized conditions, the best encapsulation rates varied from 5 to 40% depending on the dendrimer-EO combination (Table 8). The combination of GD-PPI-3 with cinnamon EO leads to the most promising results with an r = 40% when the stirring is long (6 h) and strong (1735 rpm). As there is no other study on encapsulation of EOs within GDs yet, comparing these results with previous results is not possible. However a comparison with other encapsulations techniques can be done: for example, dendrimers have a better encapsulation rate than the powder optimized by Huynh T. V. et al. who obtained 18% as optimum EO concentration [29]. On the opposite, the rate of encapsulation is quite lower than encapsulation by coacervation in gelatin optimized by Sutaphanit P. and Chitprasert P. (66.5 to 98.4%) but the release from these capsules is almost impossible (stable for 18 months storage) [30]. In another field of application, optimized encapsulation of gallic acid in calcium alginate microbeads was of the same order (42.8%) [31].

**Table 8.** Optimized values of stirring rate and time for all combinations obtained using RSMs.


In the context of the use of dendrimer-EOs formulations as biopesticide, it is essential to go further in the study of the encapsulation rate with a dynamic study of the release of EOs by the dendrimer. It is also worthwhile to determine the stability and biological effects of the new biosourced herbicide formulation. In addition, it would be relevant to study with a more fundamental point of view the encapsulation of the selected pure compounds from EO like trans-cinnamaldehyde within GD-PPI-3 to better understand the interactions between EO constituent and dendrimer particularly through NMR studies. This work is in progress.

This article shows for the first time that it is possible to effectively encapsulate essential oils in dendrimers. Given the numerous biocidal properties of essential oils, this technique opens the road to numerous applications in agronomy but also in other sectors where a slow release of essential oils is being researched, such as in pharmaceuticals or in the food industry with the design of innovative packaging.

**Author Contributions:** Conceptualization, M.-L.F. and S.B.; methodology, C.M., M.-L.F., S.B., and Y.B.; formal analysis, C.M.; data curation, C.M.; writing—original draft preparation, C.M.; writing review and editing, C.M., M.-L.F., S.B., and Y.B.; supervision, Y.B., M.-L.F., and S.B. All authors have read and agreed to the published version of the manuscript.

**Funding:** We are grateful to the Universities of Reims Champagne Ardenne (France) and Liège (Belgium) for material funds and the doctoral position to Chloë Maes. This research was supported by the Education, Audiovisual and Culture Executive Agency (EACEA), through EOHUB project 600873EPP-1-2018-1ES-EPPKA2-KA. Published with the assistance of the University Foundation of Belgium.

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

**Informed Consent Statement:** Not applicable.

**Data Availability Statement:** The data presented in this study are available on request from the corresponding author.

**Acknowledgments:** We thank Franck Michiels and Pierre Jacquet for their technical help, and Chun Yu Yang for his proofreading.

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

#### **References**


## *Article* **Insecticidal Activity of 25 Essential Oils on the Stored Product Pest,** *Sitophilus granarius*

**Sébastien Demeter 1,\*, Olivier Lebbe 1, Florence Hecq 1, Stamatios C. Nicolis 2, Tierry Kenne Kemene 3, Henri Martin 3, Marie-Laure Fauconnier <sup>3</sup> and Thierry Hance <sup>1</sup>**


2 Passage des Déportés, 5030 Gembloux, Belgium; kenne@gmx.com (T.K.K.); henri.martin@uliege.be (H.M.); marie-laure.fauconnier@uliege.be (M.-L.F.)

**\*** Correspondence: sebastien.demeter@uclouvain.be

**Abstract:** The granary weevil *Sitophilus granarius* is a stored product pest found worldwide. Environmental damages, human health issues and the emergence of resistance are driving scientists to seeks alternatives to synthetic insecticides for its control. With low mammal toxicity and low persistence, essential oils are more and more being considered a potential alternative. In this study, we compare the toxicity of 25 essential oils, representing a large array of chemical compositions, on adult granary weevils. Bioassays indicated that *Allium sativum* was the most toxic essential oil, with the lowest calculated lethal concentration 90 (LC90) both after 24 h and 7 days. *Gaultheria procumbens*, *Mentha arvensis* and *Eucalyptus dives* oils appeared to have a good potential in terms of toxicity/cost ratio for further development of a plant-derived biocide. Low influence of exposure time was observed for most of essential oils. The methodology developed here offers the possibility to test a large array of essential oils in the same experimental bioassay and in a standardized way. It is a first step to the development of new biocide for alternative management strategies of stored product pests.

**Keywords:** essential oil; insecticide; eco-friendly; stored product pest; *Sitophilus granarius*; *Allium sativum*; *Gaultheria procumbens*; *Mentha arvensis*; *Eucalyptus dives*

## **1. Introduction**

Loss of food during storage by pest infestation is a major problem in our societies in both developed and developing countries, causing significant financial losses [1–4]. Stored cereals are, indeed, a source of food for many insects, mites and fungi which degrade the product quality and can cause from 9 to 20% of net losses [5]. Around 1660 insect species worldwide are known to affect the quality of stored food products [6]. Despite this worrying situation, few research funds are allocated to offset these losses [7].

Since 1960, stored product pests have been mainly controlled by synthetic contact pesticides [8,9]. The utilization of those pesticides is being criticized more and more. Appearance of resistance in addition to the increased risks of residues dangerous to the environment and human health have led to an increasingly restricted use of those compounds [9,10]. These environmental concerns and demand for food safety have underlined the need for alternative research [10,11]. In the last decades, plant essential oils have been reported to be a potential alternative for many applications such as anti-microbial, antifungal or herbicide uses [12]. More particularly, essential oils also have interesting properties to replace synthetic insecticides [13,14]. Isman and Grieneisen [15] showed that from 1980 to 2012 the proportion of papers on botanicals among all papers on insecticides raised from

**Citation:** Demeter, S.; Lebbe, O.; Hecq, F.; Nicolis, S.C.; Kenne Kemene, T.; Martin, H.; Fauconnier, M.-L.; Hance, T. Insecticidal Activity of 25 Essential Oils on the Stored Product Pest, *Sitophilus granarius*. *Foods* **2021**, *10*, 200. https://doi.org/10.3390/ foods10020200

Academic Editor: Francesco Visioli Received: 4 December 2020 Accepted: 15 January 2021 Published: 20 January 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/).

1.43% to 21.38%. Increasing interest in essential oils as an alternative to synthetic pesticides comes from their characteristics [16]. Due to their high volatility, temperature and UV light degradation sensitivity, essential oils are less persistent in the environment than traditional pesticides [17]. In addition, most essential oils have low mammalian toxicity in comparison with synthetic insecticides and are considered as eco-friendly [18]. For instance, Stroh et al. [19] showed that eugenol was 1500 times less toxic than pyrethrum and 15,000 times less toxic than the organophosphate azinphosmethyl for juvenile rainbow trout based on 96 h-LC50 values.

In temperate regions, the granary weevil is considered as one of the major pests of stored grain [9,20–22]. Many authors [23–29] have investigated the use of essential oils as alternative insecticides against *S. granarius.* Yildirim et al. [30] demonstrated the high fumigation toxicity of *Satureja hortensis* among eleven essential oils from Lamiaceae family on *S. granarius*. Others have highlighted the contact toxicity by topical application of essential oils, such as Conti et al. [28] with *Hyptis* genera plants. A few less have worked on treated grains taking into account contact, fumigation and ingestion intoxication paths together [31]. The repellency potential of essential oils was also analysed [32,33] for *S. granarius*. In addition, essential oils were reported as a good food deterrent, as in the case of *H. spicigera* essential oils against *Sitophilus zeamais,* preventing grain degradation [34].

Nevertheless, few actual applications have emerged for the protection of stored foodstuffs and we still lack a systematic screening of potently active oils under conditions mimicking storage reality and with a standardized strain of insects. The aim of our study was precisely to test and rank 25 essential oils commonly used and available on the market against *S. granarius.* Special care has been taken for the selection of essential oil based on a large array of chemical composition (different major compounds or groups of major compounds, Table 1). In order to remain under realistic conditions for industrial large-scale application, data as price, availability on the market or health implications has been taken into account in our discussion. To allow comparison, a standardized strain of *S. granarius* was used for all the test performed under the same experimental conditions. Determination of essential oils toxicity has been done by treating the wheat grains directly, considering that the presence of wheat may influence results [35] and mostly because, in practice, it is the grain itself that will be treated in storage facilities.


**Table 1.** List of the essential oils tested and their composition for compounds (main compounds representing more than 10% of the total composition on peak area basis).


**Table 1.** *Cont.*

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

## *2.1. Biological Material*

The granary weevil, *S. granarius*, was collected in Belgium from infested wheat grain stock in 2016. They were reared at the Biodiversity Section of the Earth and Life Institute, under controlled conditions in a climatic chamber (28 ◦C ± 1, 75 ± 5% RH, in the dark) on organic wheat (*Triticum aestivum*).

## *2.2. Selected Essential Oils (EO) and Their Composition*

EOs were selected based on their availability on the market and their composition. Selected essential oils have all a distinctive major compound or a combination of major compounds to make sure to test a large range of composition.

Essential oils have been mainly obtained from Pranarom S.A. (7822—Ghislenghien, Belgium) as well as their composition. Only *Ocimum sanctum* essential oil has been purchased from "Herb and tradition" S.A company (CP59560—Comines, France) and was analyzed by GC-MS. List of the essential oils tested and their composition is indicated in Table 1. The GC-MS used for EOs characterization was a Hewlett Packard system (HP Inc., Palo Alto, CA, USA) in splitless injection mode system, with a HP INNOWAX column of 60 m, 0.25 mm of diameter and 0.5 μm of film thickness. The initial temperature of 50 ◦C was maintained for 6 min before a progressive warming of 2 ◦C per minute up to 250 ◦C. Once the temperature peak of 250 ◦C was reached, it has been maintained for 20 min. The injector and interface temperature were 250 ◦C whereas that of the source was 230 ◦C. The gas vector was helium at a pressure of 23 psi and the total ion chromatogram was recorded by using an electron-impact source at 70 eV of ion kinetic energy. The compound identification was made by comparison of the spectra to National Institute of Standards

and Technology (NIST, Gaithersburg, MD, USA) spectral library and pure commercial standards injection in the same chromatographic conditions.

#### *2.3. Toxicity Test in Treated Grain*

To be as close as possible to realistic application conditions, we have chosen to treat the grains directly with a standardized quantity of oils. A determined quantity of insects of the same age group was then directly placed on the grains. Consequently, the observed mortality is a result of contact with the treated grain, attempts at nutrition and a fumigation effect.

Toxicity tests were performed in 15 mL plastic Falcon tubes containing 8 g of treated wheat. One mL of essential oil diluted in acetone at concentrations of respectively 1; 2; 3; 4 and 5% (*v/v*) were applied on the wheat except for *Gaultheria procumbens* for which concentrations of 5; 3; 2; 1 and 0.75% were used. Moreover, because of its efficiency, the same tests of mortality have been realized for *Allium sativum* at lower concentrations of 0.75%; 0.5%, 0.25% and 0.125%.

After EO application, samples were mixed by a vortex for 1 min to homogenize the treatment. The control treatment consisted of five Falcons with 8 g of wheat treated with 1 mL of acetone only. Treated wheat dried for 15 min under hood to eliminate the acetone. Then, twenty insects per falcon were added to the wheat and Falcon were closed by a tulle to allow air circulation. Tubes were placed under controlled condition (28 ◦C ± 1 ◦C; 75 ± 5% RH). Temperature and humidity were chosen as the optimum for *S. granarius* [36] and to be representative of the conditions at the harvest period. Five repetitions were performed for each concentration.

The mortality was recorded after 24 h and 7 days of exposure. Light is repellent to *S. granarius* [37]. This particularity was used to identify dead individuals by placing a cold lamp of 100 watt in front of eyes of insects for 5 s. Individuals unable to move were considered dead.

## *2.4. Data Analysis*

In the control treatment, in one case the average mortality reached 5 percent and consequently, the Abbott formula [38] has been used to correct mortality.

The relationship between the mortality rate and the concentrations of the different oils tested was fitted with a Hill function using Scipy module of Python v.3.8.2 (Beaverton, OR, USA). This allowed us to estimate the LC50 (lethal concentration that produces 50% of mortality) and LC90 (lethal concentration that produces 90% of mortality). The Hill function is frequently used in different disciplines, from biochemistry and cellular biology to Physics [39] with the following Equation (1):

$$M = \frac{\mathbb{C}^n}{\mathbb{C}^n + K^n} \tag{1}$$

where *M* is the mortality proportion; *C* is the concentration of oil used; *K* a threshold concentration value beyond which the mortality exceeds 50% (which corresponds to the LC50) and *n* a cooperativity exponent. A value of *n* that is larger than 1 signals the presence of cooperative processes between the concentration level and the propagation of the mortality inside the population. In order to calculate the resulting LC for an arbitrary proportion of the population by rearranging the previous Equation (1):

$$\mathcal{L}\_x = \left(\frac{M\_\mathcal{X}}{1 - M\_\mathcal{X}}\right)^{1/n} \mathcal{K} \tag{2}$$

which as in Equation (3) gives for the *LC*90

$$\mathcal{L}\_{\theta0} = \mathcal{L}\mathcal{C}\theta = \mathcal{G}^{1/n}\mathcal{K} \tag{3}$$

LC90 has been used to compare essential oils' toxicity. Toxicities are considered significantly different if its standard deviation does not overlap.

#### **3. Results**

## *Mortality Analyses*

Mortality levels clearly varied among oils. When tested at the highest concentration of 5%, nine out of 25 essential oils provoked a mortality of less than 60% of the individuals after 24 h (ranging from 0 to 59%). We considered that this threshold must be exceeded to give sufficient efficiency in practice. In consequence, for these oils lower concentrations were not further tested. Looking at the results, it appeared that EOs listed in Table 2 are not effective at this concentration on *S. granarius.*



For the 16 remaining EOs, a positive relation was observed between mortality and concentration. Most of them showed a zero or almost zero mortality at a concentration of 1% except *A. sativum* which still provoked 75% of mortality after 24 h at that concentration and represents therefore the most toxic oil tested. Among the remaining oil, *G. procumbens, O. sanctum* and *Eucalyptus* dives reached respectively 81%, 68% and 51% of mortality (24 h) for a 2% concentration (Table 3).

For most of EOs tested, time of exposure did not have a significant effect on percentage of mortality, indicating that a knock down effect is rapidly observed (Table 4). However, this observation does not hold for three EOs after 24 h and 7 days, *Thymus vulgaris* CT geraniol, *Myristica fragrans* and *O. sanctum,* indicating a cumulative toxic effect probably linked to physiological or neurological disorders.

With a LC90 of 1.04% after 7 days of exposure, *A. sativum* is the most toxic essential oil tested. It is followed by *G. procumbens* and *O. sanctum* that showed similar results with LC90 of 2.10 and 2.11% (7 days). The third position in the list of the most toxic essential oils is shared by *Mentha arvensis*, *T. vulgaris* CT geraniol and *E. dives* which present respectively a LC90 of 3.08; 3.08 and 3.11% after 7 days of exposure.

Calculation of mortality curves was realized for 24 h and 7 days treatment (Figure 1). Table 4 indicates the LC90 after 24 h and 7 days for these 16 essential oils tested.


**Table 3.** Summary of mortality percentages after 24 h hours of exposure for the concentrations tested (*n* = 5).

**Figure 1.** Mortality curves of EOs tested on *S. granarius* 24 h after treatment (blue) and 7 days after treatment (orange).

**Table 4.** Summary of mortality data presented at the Figure 1 for the 16 essential oils tested. Lethal concentrations are expressed in percent.


### **4. Discussion**

## *4.1. Insecticidal Potential*

This study compares the toxicity of 25 essential oils on the granary weevil. Sixteen of these were found to have an interesting insecticidal activity on *S. granarius*. Our results show that *A. sativum*, *G. procumbens*, *O. sanctum*, *M. arvensis, T. vulgaris* (geraniol) and *E. dives* present a potential to control *S. granarius* population directly in the grain.

Garlic essential oil has been identified as the most toxic oil with a LC90 two to four times lower than other EOs, probably because of its content in sulfur compounds. Its toxicity on other insect pests of stored products like *Tenebrio molitor* [40], *Sitotroga cerealella* [41], *Tribolium castaneum* and *Sitophilus zeamais* [42,43] has already been described. The efficiency of garlic essential oils and his constituents may vary with the target species, the stage of life and the exposure mode (fumigation or contact). For example, Ho et al. [42] calculated a KD50 (knock down) of 1.32 mg/cm<sup>2</sup> and 7.65 mg/cm2 of garlic essential oil against *T. castaneum* and *S. zeamais* respectively. In addition, Plata-Rueda et al. [40] have

identified diallyl disulfide as the most toxic compounds present in the garlic essential oil explaining its efficiency on *Tenebrio molitor.* Contact and fumigation toxicities of diallyl trisulfide has been highlighted by Huang et al. [43] on *T. castaneum* and *S. zeamais*. Contrary to most other essential oils, these molecules are not present in the garlic clove itself, but arise from the conversion of thiosulfinates (water-soluble) to sulfides (oil-soluble) during the hydrodistillation process [44]. In short, the main sulfur compounds in the whole garlic clove are cysteine sulfoxides like allylcysteine sulfoxide (alliine) and methylcysteine sulfoxide (methiine) which are located in the clove mesophyll storage cells. After crushing the clove, those compounds come in contact with the enzyme *alliinase* that is normally localized in the vascular bundle sheath cells. The vast majority of cysteine sulfoxides are then converted in sulfenic acids which self-condense to thiosulfinates like allicin which is the most abundant compound (60–90% of total thiosulfinate). Allicin is quite unstable depending on the medium and temperature. Upon hydrodistillation, thiosulfinates are transformed into diallyl trisulfide, diallyl disulfide and allyl methyl trisulfide as major products [44].

Essential oils toxicity of *M. avensis* [45], *G. procumbens* [46] and *E. dives* [47] as well as geraniol (main compound of *T. vulgaris* essential oil) [48] was also been highlighted for their activities against various stored product pests. In addition, Yazdgerdian et al. [46] identified *G. procumbens* as the most toxic oil, both by fumigation (6.8 μL/L air) and contact on treated wheat (0.235μL/g), among five essential oils tested on *S. oryzae*. These results confirm the toxicity of *G. procumbens* observed in our study. However, although many studies highlighted toxicities of essential oils, lack of a common protocol or of major compounds description often prevent from reliable and univocal comparison. For example, in the study of Teke et al. [49] the fennel essential oils applied on *S. granarius* contains 71.64% of estragol, which closely resembles the composition of the basil oil in our study (73.43% estragol). However, in their case they realized topical application without grain presence which is quite different that in our case.

At the opposite, Zohry et al. [50] tested toxicity of 10 essential oils on *S. granarius* by exposure to treated wheat in a protocol close to that of this study. Garlic oil was identified as the most toxic one with a concentration of oil per grams of grain similar to ours. However, no precision on composition of EOs are available in their publication, which do not allow a deeper comparison. Further studies on the evaluation of the industrial potential of essential oils need to be based on a common protocol taking into account the influence of the media [35] and a full description of the composition of essential oils.

Despite numerous studies on the toxicity of essential oil on stored product pests, little data is available on the mechanism of action of the insecticidal effect of these essential oils as a mixture of molecules. However, some studies highlight some mechanisms. For instance, Jankowska et al. [51] showed that menthol acts on octopamine receptors and trigger protein kinase A phosphorylation pathway on cockroach DUM neurons. Hong et al. [52] indicate a potential interference of methyl salicylate and eugenol with octopamynergic system as well. Action on octopamine receptors is an advantage in the elaboration of an insecticide due to absence of key role in vertebrates involving a relative security for human health. However, methyl salicylate is known to have a LD50 oral (rat) of 887 mg/kg indicating that it should have another mechanism of action on mammals. Therefore, the mere fact that octopamynergic system is targeted by an essential oil cannot guarantee safety for human health. β-caryophyllene was identified as an inhibitor of the activities of acetylcholine esterase, polyphenol oxidase and carboxylesterase on *Aphis gossypii* [53]. *α*-phellandrene is believed to have a neurotoxic effect on *Lucinia cuprina* [54]. Diallyl disulfide is known to impact digestion of *Ephestia kuehniella* by decreasing activity of digestive enzymes [55]. Diallyl trisulfide, another major compound of garlic EO, has been recently described as a regulator of the expression of the chitin synthase A gene which generates alteration of the morphology and inhibition of the oviposition of *Sitotroga cerealella* [56]. Finally, essential oils are complex mixture of molecules, possibly interacting and entering in synergy for their mechanism of action. Therefore, it is important to analyse their impact on insect as a whole. For example, a recent study shows that *M. arvensis* EO is associated with a systemic mode of action on *S. granarius* since it is capable of altering the nervous and muscular systems, cellular respiration processes and the cuticle, the first protective barrier of insects [57].

#### *4.2. Human Health Risk*

Toxicity on the target pests is a first step for any kind of new pesticide elaboration. However, in the perspective of a potential utilization of essential oils in an industrial context, it is also essential to focus on some other aspects, such as the price, the wheat deterioration or the mammal toxicity to determine their actual industrial potential. Concerning mammal toxicity, the WHO classification ranked compounds from "extremely hazardous" to "unlikely to present acute hazard" based on the concentration in mg/kg that provoke 50% of mortality in rat (WHO, 2009). Concerning *A. sativum,* diallyl trisulfide is ranked as "unlikely to present acute hazard" while diallyl disulfide is considered as moderately hazardous with an oral LD50 (rat) of 260 mg/kg. Even if this toxicity is two to four times lower than deltamethrin currently used in granaries, it remains to be carefully considered in the case of a conception of healthy and ecofriendly alternatives to insecticide.

*Gaultheria procumbens* which showed the second highest acute toxicity to *S. granarius* is constituted at 99% of methyl salicylate, a molecule classified as moderately hazardous for human health. Because of this specific composition, this essential oil should be use in association to avoid a rapid development of resistance. Further analyses have also to be done on the persistence of methyl salicylate, on its environmental and mammal toxicity to estimate the potential of this EO as a stand-alone or mix product. Two molecules of *O. sanctum* (eugenol and methyl eugenol) as well are classified as moderately hazardous to mammals and need to be considered with the same caution.

For the three other oils identified, major compounds are all classified as "slightly hazardous" to "unlikely to present acute hazard" and their use should not be a problem to treat food product.

## *4.3. Prices*

If we considered prices (Table 5), essential oils are quite expensive, particularly garlic oil probably because its low availability and its use mainly as an aroma in food industry. Moreover, sulfides are also well known for their unpleasant odor complicating its practical application. These two points explained its low practical applications. *O. sanctum* also seems too expensive to be used at an industrial scale.

**Table 5.** Price of the most lethal oils tested and the mammal's toxicity of their major compounds.


Data obtained from safety data sheet from: \* Cayman (Ann Arbor, MI, USA); \*\* Fisher Science education (Rochester, NY, USA); \*\*\* Echemi (Qingdao, China); \*\*\*\* Carl Roth (D-76185 Karlsruhe, Germany); \*\*\*\*\* CDH Fine Chemicals (New Delhi, India). Prices have been obtained from Ultra Internationnal B.V. (Spijkenisse, The Netherlands). WHO Classification: II: Moderately hazardous; III: Slightly hazardous; U: Unlikely to present acute hazard.

> *Gaultheria procumbens, M. arvensis* and *E. dives* are among the less costly essential oils on the market. Moreover, these three oils are easily available on the market. Based on our results, their toxicity and their price, these three essential oils could represent good

opportunity to develop a botanical insecticide to control insect pest in stored product. We did not obtain a commercial price for *T. vulgaris* CT geraniol at an industrial scale.

#### *4.4. Duration of Exposure*

Only three essential oils (*M. fragrans*, *O. sanctum* and *T. vulgaris* CT geraniol) showed an increase in mortality 7 days after the treatment (Figure 1). This could be the consequence of a cumulative contamination during all the period, including by feeding. It is also possible that physiological disorders took times and was linked to an arrest of feeding and water losses.

For the other essential oils, little differences of mortality were observed after 1 and 7 days of exposure. Several hypotheses could explain that observation. First as mortality arise soon after the insect introduction, we may expect a strong selection effect on susceptible individuals, leaving alive after one day only more resistant individuals. Secondly, the absorption of essential oils by the grains (by fumigation or contact) could reduce the biodisponibility of the active compounds and thus the lack of efficiency on long terms period. Indeed, Lee et al. [35] put into light that fumigation toxicity of certain essential oils is three to nine times lesser in presence of wheat due to the absorption phenomena.

Thirdly, our experiment has been conducted at 28 ◦C. The evaporation rate of essential oils is rapid at this temperature and a substantial part of the essential oil may have vanished after 24 h. Heydarzade et al. [58] highlighted the low persistence of essential oils of *Teucrium polium* and *Foeniculum vulgare.* Treated filter paper induced a 99% mortality at time zero and 0% 30 h after application on *Callosobruchus maculatus* adults. This downgrade of activity is supposed to be caused by high volatility and/or quick degradation of active compounds.

Studies must be carried out on the combined influence of evaporation and absorption by grains of essential oils in order to demonstrate their toxicity persistence over time. In further studies, it is a priority to include GC-MS analyses of treated wheat that allowed scientists to determine the behavior of essential oils and its remanence at the surface and inside the treated wheat until the end of experiment. This factor is essential to control insect pests that lay eggs into the grain, which causes a delay between treatment and the potential contact with the insecticide product by emerging individuals.

Finally, we cannot exclude that the low difference between mortalities for both exposure times could be explained by the absence of accumulation of toxic compounds in the insect and its capacity to metabolize them. The few cases where a difference was identified between both exposure times could be explain by a more physiologic mode of action inducing drying or no feeding effect which induces slower death pattern.

#### **5. Perspectives**

Moreover, to precise if these essential oils could be a viable alternative to pesticide in an industrial point of view, further studies has to be conduct on the comparison of their efficiency with the one of actual synthetic insecticides and/or natural substances well known for their insecticidal properties in a protocol mimicking the actual mode of treatment. To answer eventually the question: "Are these essential oils actually a good alternative to the current standards", future studies should include a positive control with a treatment protocol based on pulverization.

Experiments should also be carried out at a larger scale, such as experimental granaries, with the purpose of estimating the quantity of oil per ton of wheat needed and thus the practical applicability of these treatments. Indeed, under mass storage conditions, the application of essential oils during the grain filing process in the silo is based on nano-drop pulverization which could greatly increase the evaporation of the product. Moreover, the formulation of the essential oil is also of tremendous importance as discussed by Maes et al. [59]. In our cases, dilutions were made using acetone which is also quite different from actual industrial application. These points should be further analyzed in details.

## **6. Conclusions**

Considering insecticidal effects, prices, availability and mammal toxicity of essential oils tested, *M. arvensis, E. dives* and *G. procumbens* can be considered as good potential alternatives to the synthetic pesticides presently used to control grain weevils. As essential oils are products of very variable composition, studies must be performed to clearly identify the compound(s) responsible of the insecticidal toxicity of these three essential oils to avoid variable responses to future treatments. More investigations need to be done on the mechanism of action of these oils, including the role of minor components, both on insects and mammals, to secure their industrial use.

**Author Contributions:** Conceptualization, S.D.; M.-L.F. and T.H.; writing—original draft, S.D.; writing—review & editing, S.D.; M.-L.F.; S.C.N. and T.H.; data curation, S.D., O.L., F.H., S.C.N., T.K.K., H.M.; formal analysis: S.D.; S.C.N.; supervision: M.-L.F.; T.H.; funding acquisition: T.H. and M.-L.F. All authors have read and agreed to the published version of the manuscript.

**Funding:** This study is part of the Walinnov project OILPROTECT (1610128) granted by Wallonia via the "SPF-Economie Emploi Recherche", Win2Wal program. This research was funded by the Education, Audio-visual and Culture Executive Agency (EACEA) through the EOHUB project 600873-EPP1-2018-1ES-EPPKA2-KA.

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

**Informed Consent Statement:** Not applicable.

**Data Availability Statement:** The data presented in this study are openly available in 10.6084/m9. figshare.13603526.

**Acknowledgments:** This paper is publication BRC275 of the Biodiversity Research Center, Université catholique de Louvain.

**Conflicts of Interest:** The authors declare no conflict of interest. The funders had no role in the design of the study; in the collection, analyses, or interpretation of data; in the writing of the manuscript, or in the decision to publish the results.

#### **References**


## **Screening of Antifungal and Antibacterial Activity of 90 Commercial Essential Oils against 10 Pathogens of Agronomical Importance**

**Caroline De Clerck 1,\*,**†**, Simon Dal Maso 1,**†**, Olivier Parisi 1, Frédéric Dresen 1, Abdesselam Zhiri <sup>2</sup> and M. Haissam Jijakli 1,\***


Received: 2 September 2020; Accepted: 3 October 2020; Published: 7 October 2020

**Abstract:** Nowadays, the demand for a reduction of chemical pesticides use is growing. In parallel, the development of alternative methods to protect crops from pathogens and pests is also increasing. Essential oil (EO) properties against plant pathogens are well known, and they are recognized as having an interesting potential as alternative plant protection products. In this study, 90 commercially available essential oils have been screened in vitro for antifungal and antibacterial activity against 10 plant pathogens of agronomical importance. EOs have been tested at 500 and 1000 ppm, and measures have been made at three time points for fungi (24, 72 and 120 h of contact) and every two hours for 12 h for bacteria, using Elisa microplates. Among the EOs tested, the ones from *Allium sativum*, *Corydothymus capitatus*, *Cinnamomum cassia*, *Cinnamomum zeylanicum*, *Cymbopogon citratus*, *Cymbopogon flexuosus*, *Eugenia caryophyllus*, and *Litsea citrata* were particularly efficient and showed activity on a large panel of pathogens. Among the pathogens tested, *Botrytis cinerea*, *Fusarium culmorum*, and *Fusarium graminearum* were the most sensitive, while *Colletotrichum lindemuthianum* and *Phytophthora infestans* were the less sensitive. Some EOs, such as the ones from *A. sativum*, *C. capitatus*, *C. cassia*, *C. zeylanicum*, *C. citratus*, *C. flexuosus*, *E. caryophyllus*, and *L. citrata*, have a generalist effect, and are active on several pathogens (7 to 10). These oils are rich in phenols, phenylpropanoids, organosulfur compounds, and/or aldehydes. Others, such as EOs from *Citrus sinensis*, *Melaleuca cajputii*, and *Vanilla fragrans*, seem more specific, and are only active on one to three pathogens. These oils are rich in terpenes and aldehydes.

**Keywords:** essential oil; biocontrol; antifungal; antibacterial; biopesticide

## **1. Introduction**

Fruits, vegetables, and cereals are important components of the human diet at every age [1]. The increased demand for these commodities exert significant pressure on the environment, leading to intensive agriculture and the use of chemical pesticides. However, the use of these chemicals, and the resulting presence of their residues in food and water, are leading to several health safety breakdowns. Moreover, the use of chemical pesticides affects the environment and the biodiversity. The constant (and sometimes inadequate) use of pesticides is also responsible of the development of pathogen resistances leading to possible food safety issues [2].

Today, the demand for a reduction of chemical pesticides, and for the development of alternative ways to protect crops from pathogens and pests, is growing [3]. In response, research and development in the field of biopesticides has grown exponentially in the last 20 years.

Among the natural alternatives to chemical pesticides, products based on plant extracts and/or plant essential oils (EOs) have received increasing attention because of their generally recognized as safe (GRAS) compounds, due to their very low human toxicity, high volatility, and rapid degradation [4].

Essential oils possess a strong odor and are produced by aromatic plants as secondary metabolites [5]. They are usually obtained from several plant parts by steam hydrodistillation [6]. They are made of a mixture of volatile compounds (between 20 and 100), even if they are, in most cases, characterized by two or three main compounds, representing the major part of the EO (20–70%). As an example, EO of *Citrus limon* is composed, in majority, of limonene and β-pinene [7,8]. Two kinds of molecules can enter in the composition of essential oils: terpenes and terpenoids (e.g., limonene, linalool); and aromatic and aliphatic molecules (e.g., cinnamaldehyde, safrole) [9]. All of these components are characterized by a low molecular weight [10].

Essential oils were known, for a long time, for their antimicrobial and medicinal properties. The latter have, among others, led to the development of aromatherapy, where they are used as bactericide (e.g., tea tree and cinnamon EOs), fungicide (*Lavandula spica* EO), or virucid (*Cinnamomum camphora*) [5,11].

In the last 20 years, the antibacterial and antifungal properties of essential oils have been assessed against a large variety of plant pathogens in order to determine their potential as alternative plant protection products [6,12]. The complex composition of essential oils is interesting, as they could act as multisite chemicals, lowering the risk of resistance [13].

Furthermore, essential oils are composed by low molecular weight molecules and are highly volatile. This property is of great interest, particularly when used on fresh products or during postharvest applications. However, this advantage, in terms of residue reduction, is also a major inconvenience for crop application, which has to be overcome by a formulation allowing to maintain the efficacy of the product [14].

In this study, the in vitro efficacy of 90 commercially available essential oils against 10 plant pathogens of agronomical importance has been assessed. This is, to our knowledge, the largest screening for antifungal and antibacterial activity of EOs made so far.

## **2. Materials and Methods**

## *2.1. Essential Oils*

The 90 essential oils (EOs) tested in our study were supplied by Pranarom International (Ghislenghien, Belgium) (Table 1).


**Table 1.** List of essential oils tested in this study.


**Table 1.** *Cont.*

## *2.2. Fungal and Bacterial Strains*

The 10 host–pathogen combinations used in this study are listed in Table 2. All of the cultures were carried out at a 16D:8N photoperiod on the most appropriate solid media (see Table 2). The Potato dextrose agar (PDA) (Merck) medium was prepared according to the manufacturer's instructions (39 g of powder in 1 L of water). The Luria-Bertani-agar (LB-agar) medium was composed of 10 g/L of peptone 5g/L of yeast extract, 10g/L of NaCl, and 15 g/L of agar. The V8 medium was made with 100 mL/L of V8 juice, 200 mg/L of CaCO3, and 20 g of agar. For in vitro screening procedures in liquid medium, pathogens have been cultured in the same media without the addition of agar. All of the media were autoclaved during 20 min at 120 ◦C.


**Table 2.** List of the pathogens tested in this study and their culture conditions.

PDA (Potato dextrose agar); LB-Agar (Luria-Bertani-agar).

## *2.3. Making of a Stable EO Emulsion*

EOs are not water soluble. In order to get homogenous and stable emulsions, a formulation was developed to get a final EO concentration of 1000 ppm (maximum dose tested in the in vitro screening). The EOs were first diluted in ethanol in a ratio of 16.7:83.3%. Half a milliliter of this solution was then mixed with 555 μL of Tween 20 and 26.71 mL of distilled water, in order to get an EO concentration of 0.3%. For the in vitro screening procedure, this emulsion was diluted to reach the desired final EO concentration (see Section 2.4.2).

## *2.4. In Vitro Screening Procedure*

## 2.4.1. Determination of the Pathogens Kinetic Growth

The aim of this step was to determine the optimal growth conditions for each of the pathogens tested (exponential growth phase between 0 h and 48 h, followed by a growth plateau).

The kinetic growth of each pathogen in liquid media was determined using 96 wells ELISA microplates, following the method developed and validated by [15]. Three dilutions (3x, 30x, and 300x) of the medium and three concentrations (104, 105, and 106 spores/mL) of spores' suspensions were tested for each fungus (except for *P. infestans*, for which suspensions of 104, 105, and 0.3 106 spores/mL were tested). For bacteria, three dilutions of the medium (3x, 30x, 300x) and three bacterial suspensions (106, 107, and 108 bacteria/mL) were tested.

Each well was filled with one volume of culture medium, one volume of the pathogen suspension in culture medium, and one volume of water containing 2% of tween 20. The plates were then incubated in the dark at 23 ◦C. Pathogen growth was assessed by measuring the optic density at 630 nm with a spectrophotometer (Thermo, LabSystems Multiskan RC 351, Chantilly, VA, USA) every 24 h for 144 h. Sixteen replicates (wells) were made for each growing condition (medium and pathogen concentrations). Conditions giving the best pathogen growth are listed in Table 3, and will be the growth conditions selected to go further in the EO screening tests.

**Table 3.** Pathogen growth conditions selected for the screening tests.


#### 2.4.2. Screening

The in vitro screening method in liquid medium is similar to the method used to determine pathogen kinetic growth (see Section 2.4.1). In 96-well ELISA plates—each well was filled with one volume of the selected medium at the optimal concentration (see Table 3), one volume of the pathogen at the optimal suspension (see Table 3), and one volume of the selected EO emulsion (see Section 2.3) at 500 and 1000 ppm (final concentration), except for *P. infestans*, for which EOs have only be tested at 1000 ppm. The plates were incubated in the dark at 23 ◦C. Growth was assessed by measuring the optic density at 630 nm with a spectrophotometer (Thermo, LabSystems Multiskan RC 351, Chantilly, VA, USA) after 24, 72, and 120 h (for fungi) or every 2 h during 12 h (for bacteria).

Figure 1 shows the wells repartition on the plate for an optimal screening procedure, minimizing the contaminations, following [16].



**Figure 1.** Objects repartition on the ELISA plate for an optimal screening procedure. O1 to O10 represent the tested objects (essential oils (Eos)), T1 to T10 represent negative controls (without pathogen), T' is the culture medium only and X' is the growth control (medium and pathogen). Eight replicates (wells) were made by object.

The efficacy of each EO against each pathogen was calculated using the following formula (1):

$$\text{Efficiency of treatment n } (\%) \, = \frac{(X \text{ } \acute{'} X\_0) - (X\_n \acute{'} X\_{n0})}{(X \acute{'} X\_0)} \times 100\tag{1}$$

where X' is the optical density of the non-treated growth control at time "t", X0 is the optical density of the non-treated growth control at time "0", Xn is the optical density of treatment "n" at time "t" and Xn0 is the optical density of treatment "n" at time "0". The values of the negative control Tn (negative control for treatment n: EO and medium only) and T' (medium only) are also checked to be sure that no contaminations occurred. Heat maps were created using the "ggplot2" package of the R software using the mean of the eight replicates for each couple "EO x Pathogen x Time".

## **3. Results**

*3.1. Evaluation of the E*ff*ect of the 90 EOs on the 10 Pathogens*

## 3.1.1. *P. expansum*

At 500 ppm, 20 compounds have shown an interesting effect on *P. expansum* growth (efficacy comprised between 67 and 100%) lasting at least 24 h (see Figure 2). In general, the efficacy of the EOs at 500 ppm do not last very long (around 24 h), with some exceptions, for which the activity lasts more than 120 h: *A. sativum*, *C. cassia*, *C. zeylanicum*, and *E. caryophyllus*.

**Figure 2.** Heat map showing the efficiency of the 90 EOs at 500 and 1000 ppm on the growth of eight plant fungal pathogens after 24 to 120 h of contact in liquid medium in vitro. Red squares represent efficiencies below 50% of growth reduction. Yellow squares represent reduction of growth comprised between 50 and 66%. Efficiencies between 67 and 99% are represented by green squares, while blue squares show a complete inhibition of the organism.

At 1000 ppm, 20 compounds have shown an efficacy comprised between 67 and 100% lasting at least 24 h. In this case, there is also an increasing number of compounds keeping high efficiencies upon time: *A. sativum*, *C. cassia*, *C. zeylanicum*, *C. citratus*, *C. flexuosus*, *Leptospermum petersonii*, *L. citrata*, *C. capitatus*, *Origanum heracleoticum*, *Origanum compactum*, and *E. caryophyllus*.

In particular, EOs of *Monarda fistulosa* at 500 ppm and *O. heracleoticum* at 1000 ppm completely inhibited *P. expansum* during the first 24 h.

## 3.1.2. *B. cinerea*

At 500 ppm, 35 EOs have shown high activities (efficacies comprised between 67 and 100%) against *B. cinerea*, lasting at least 24 h. However, the growth inhibition was observed with a delay of at least 48 h for most of them (23/35). Moreover, EOs of *C. cassia*, *C. zeylanicum*, *C. citratus*, *C. flexuosus*, and *Pimpinella anisum* completely inhibited the pathogen growth from 72 h of contact, while EOs of *Myristica fragrans* and *Thymus vulgaris* ct. thymol showed 100% efficacies from 120 h of contact with the oil.

At 1000 ppm, the majority of the tested EOs (54) have shown efficacies between 67 and 100%. Among these, 34 showed efficiencies higher than 67%, lasting at least 72 h. EOs of *A. sativum*, *Cuminum cyminum*, *Eucalyptus dives*, *Lavendula angustifolia*, *Lavendula x burnetii*, and *Mentha pulegium* completely inhibited the growth of the pathogen the first 24 h and EO of *Copaifera o*ffi*cinalis* showed 100% of efficacy the first 72 h. In addition, oil from *Satureja hortensis* and *T. vulgaris* ct. thymol showed 100% efficacies from, respectively, 72 h and 120 h of contact with the EOs.

The pathogen was completely inhibited by EO of *C. capitatus* at 500 as well as 1000 ppm.

## 3.1.3. *C. beticola*

At 500 ppm, 14 EOs have shown efficacies between 67 and 100% against *C. beticola*, lasting at least 72 h. In particular, EOs of *A. sativum*, *C. cassia*, *C. zeylanicum*, *Canarium luzonicum*, *C. capitatus*, *C. flexuosus*, and *E. Caryophyllus* have shown activities lasting more than 120 h.

At 1000 ppm, 22 EOs have been highly efficient in reducing the pathogen growth (100% inhibition during the first 24 h). Moreover, 26 more have shown inhibition between 67 and 100%, lasting at least 24 h. However, only three EOs kept a high efficacy during the whole period of screening: *L. petersonii*, *Vetiveria zizanioides*, *E. caryophyllus*.

This is also the only pathogen for which EOs of *C. cassia* and *C. zeylanicum* have efficacies lower than 50% at 1000 ppm.

## 3.1.4. *F. culmorum*

At 500 ppm, 20 EOs showed maximal activities (67–100%) against the pathogen, lasting at least 24 h. In particular, EOs of *A. sativum*, *C. cassia*, *C. zeylanicum, C. citratus*, and *E. caryophyllus* completely inhibited the growth of *F. culmorum* up to 120 h of culture. EOs of *A. sativum*, *C. cassia*, and *C. citratus* completely inhibited the growth of *F. culmorum* for 120 h at this concentration, while EOs of *C. capitatus*, *C. zeylanicum*, *L. citrata*, and *O. heracleoticum* inhibited it completely during the first 24 h, and oil of *C. flexuosus* during the first 72 h.

At 1000 ppm, 61 EOs had efficacies comprised between 67 and 100% lasting at least 24 h. For 18 of these EOs the effect lasted for at least 120 h. Moreover, EOs of *A. sativum*, *C. cassia*, *C. flexuosus*, and *L. citrata* completely inhibited the growth of the pathogen during the 120 h of the test. In addition, 26 other EOs showed efficacies of 100% lasting at least 24 h.

## 3.1.5. *F. graminearum*

At 500 ppm, 75 of the 90 EOs tested had efficacies comprised between 67 and 100%, lasting at least 24 h. In addition, 21 EOs provided 100% of inhibition lasting at least 24 h, including *A. sativum*, *C. cassia*, and *C. zeylanicum*.

At 1000 ppm, almost all of the EOs (78) showed efficacies superior to 67%, lasting at least 24 h. Moreover, 29 EOs provided a complete inhibition of the pathogen, lasting at least 24 h, including EOs of *C. cassia* and *C. capitatus*.

Interestingly, EOs of *E. caryophyllus* at 500 ppm and of *C. capitatus*, at 1000 ppm, completely inhibited the growth of the pathogen during 120 h.

Some EOs (*C. sinensis* and *V. fragrans* auct., among others) have shown high activities (more than 67) during the first 24 h at 500 ppm, while their maximal efficacy at 1000 ppm never exceeded 50%.

## 3.1.6. *P. ultimum*

At 500 ppm, 37 EOs have shown efficacies between 67 and 100%, lasting at least 24 h. Interestingly, it can observed that EOs of *A. sativum* and *E. caryophyllus* completely inhibit the pathogen for at least 120 h.

At 1000 ppm, 61 EOs have efficacies greater than 67%, lasting at least 24 h, among which 12 have an activity lasting 120 h. EOs of *C. capitatus*, *C. citratus*, and *O. heracleoticum* completely inhibited the pathogen growth for at least 120 h.

## 3.1.7. *C. lindemuthianum*

At 500, only eight EOs have shown activities greater than 67%, lasting at least 24 h. EOs of *A. sativum*, *C. cassia*, *C. zeylanicum*, and *E. caryophyllus* showed the best results over time.

At 1000 ppm, three EOs have shown activities greater than 67%, lasting at least 24 h. EOs of *A. sativum*, *C. citratus*, and *L. citrata* are the most efficient EOs in this case.

None of the oils tested provided a total inhibition of the pathogen.

## 3.1.8. *P. infestans*

At 1000 ppm, 10 EOs showed efficacies higher than 67% lasting at least 24 h. Among these, only five EOs showed efficacies greater than 67% during 120 h. EOs of *C. cassia*, *C. flexuosus*, *C. zeylanicum*, and *M. pulegium* completely inhibited the pathogen for at least 120 h.

## 3.1.9. *P. carotovorum* (PCC)

At 500 ppm, four EOs are causing 100% inhibition, lasting at least 12 h: *A. sativum*, *C. capitatus*, *C. cassia*, and *C. citratus* (See Figure 3).

**Figure 3.** Heat map showing the efficiency of the 90 EOs at 500 and 1000 ppm on the growth of two plant bacterial pathogens after 2 to 12 h of contact in liquid medium in vitro. Red squares represent efficiencies below 50% of growth reduction. Yellow squares represent reduction of growth comprised between 50 and 66%. Efficiencies between 67 and 99% are represented by green squares, while blue squares show a complete inhibition of the organism.

At 1000 ppm, the same four EOs caused a complete inhibition of the pathogen, in addition to the one of *O. heracleoticum*.

#### 3.1.10. *P. atrosepticum* (PCA)

Two EOs completely inhibited the bacterium at the two concentrations tested: *C. cassia* and *E. caryophyllus*.

At 1000 ppm, nine additional EOs caused a total inhibition: A. sativum, C. capitatus, C. citratus, C. flexuosus, Cymbopogon martini, C. zeylanicum, L. citrata, L. petersonii, and O. heracleoticum.

## **4. Discussion**

In this study, the efficacy of 90 commercial essential oils against 10 plant pathogens of agronomical importance was studied.

Similar to the majority of the papers about antifungal and antibacterial effects of EOs [17,18], a dose dependent response was observed for almost all of the EOs tested in this study, the effects being stronger at 1000 ppm than at 500 ppm.

However, they were some exceptions. This is, for example, the case of *C. lindemuthianum* and *C. beticola*, for which most of the EOs showing activities were more effective at 500 ppm than at 1000 ppm. While this is not commonly found in the literature, there are some studies showing similar results [19,20]. Possible explanations are that diluted EOs could diffuse easier in aqueous environments, or that a higher rate of polymerization in concentrated EOs may reduce their antimicrobial activity [20,21].

In most of the cases, the comparison between screenings at 500 and 1000 ppm tend to show that the EO concentrations influence the time of their effectiveness on pathogens, with more concentrated formulations giving longer protection. This fact is certainly due to the high volatility of EOs.

Some EOs, such as the ones from *A. sativum*, *C. capitatus*, *C. cassia*, *C. zeylanicum*, *C. citratus*, *C. flexuosus*, *E. caryophyllus*, and *L. citrata*, have a generalist effect, and are active on several pathogens (between 7 and 10). These oils are rich in phenols, phenylpropanoids, organosulfur compounds, and/or aldehydes, known in the literature to have antifungal effects (thymol and carvacrol for *C. capitatus* [22]; neral and geranial for *C. citratus*, *C. flexuosus*, and *L. citrata* [23]; eugenol for *E. caryophyllus* and *C. zeylanicum* [24]; cinnamaldehyde for *C. cassia* and *C. zeylanicum* [25]; and diallyl di and tri-sulfide for *A. sativum* [26]).

Others, such as EOs from *C. sinensis*, *M. cajputii*, and *V. fragrans*, seem more specific, and are only active on one to three pathogens. These oils are rich in terpenes (limonene, myrcene, and pinenes for C. sinensis [27]; elemene, caryophyllene, terpinolene, humulene for *M. cajputii* [28]), and aldehydes (vanillin for *V. fragrans*) [29].

Some pathogens are more sensitive to the EOs tested, such as *B. cinerea* and the two *Fusarium* species. Some studies have already reported that fact [12,30].

Pathogens, such as *C. lindemuthianum* and *P. infestans*, seem less sensitive. Studies showing efficacies of EOs against *C. lindemuthianum* exist in the literature, but are indeed scarce: Khaledi and al [31] showed that EO of *Bunium persicum* was effective, while [32] showed effects for peppermint EO and winter green oil.

The moderate sensitivity of *P. infestans* to EOs was already reported in the literature [33] and could be explained by the fact that it is an oomycete, differing from fungi in cell wall composition and lifecycle, among others [34]. *P. ultimum*, another oomycete tested in our study, was affected by more EOs than *P. infestans*, but the observed effects were limited in time (lasting for only the first 24 h). All of the EOs having an effect on *P. infestans* also showed an activity on *P. ultimum*, except for *C. cyminum*.

For some pathogens, such as *F. graminearum*, *C. beticola*, and *P. ultimum*, the inhibition effect is very high the first 24 h, then it decreases or disappears. This result could indicate that these pathogens are more sensitive to EOs in the form of spores.

The opposite situation was observed with *B. cinerea*, where most of the efficient EOs only become active after at least 24 h of contact with the pathogen. This delayed efficacy could indicate that, in the case of this pathogen, the EOs are more efficient on the mycelium rather than on spores.

For bacteria, we observed that EOs are more efficient at 1000 ppm. PCC seem more sensitive. In general, after 10 h of contact, EOs showing an effect on PCA, which is less sensitive, are also acting on PCC. The most efficient EOs for bacteria are the same as the ones showing high activities for fungi (*C. cassia*, *E. caryophyllus*, *C. capitatus*, *A. sativum*, etc.). EOs rich in carvacrol, like the one of *C. capitatus*, were already found to be effective against PCC [35]. No oil showed specific activity against bacteria.

#### **5. Conclusions**

The number of studies available in the literature about fungicidal and fungistatic effects of essential oils, as well as their mechanism of action, is growing, and it is now commonly accepted that EOs have great potential in the development of new biopesticides [6,12].

In our study, 90 EOs were tested on eight fungal pathogens and two bacterial pathogens of agronomical importance. This is, to our knowledge, the largest in vitro screening of EOs made so far. This study allowed us to have a global vision of a large panel of EO efficacies, and to identify several interesting candidates, acting on a large range of pathogens: EOs of *A. sativum*, *C. capitatus*, *C. cassia*, *C. zeylanicum*, *C. citratus*, *C. flexuosus*, *E. caryophyllus*, and *L. citrata*. These oils could be promising candidates in the development of new biopesticides.

However, we have to be careful, as all of our tests have been made in vitro. The promising effects that we have observed need to be confirmed in vivo and, in particular, phytotoxic activities, which are often reported for Eos, will have to be studied [36]. We agree with [6], stating that more studies about the mode of action of EOs, the synergic effect among them or their components, and the identification of their more active components are required. More knowledge is also needed about the effect of these EO applications on the environment (beneficial organisms, soil microbiota, etc.), and on human health, even if the high volatility of EOs should minimize these effects.

**Author Contributions:** Conceptualization, O.P. and M.H.J.; methodology, O.P. and F.D.; formal analysis, O.P., C.D.C., and S.D.M.; investigation, C.D.C.; writing—original draft preparation, C.D.C. and S.D.M.; writing—review and editing, C.D.C., S.D.M., O.P., and M.H.J.; supervision, M.H.J.; funding acquisition, A.Z., M.H.J. All authors have read and agreed to the published version of the manuscript.

**Funding:** This research was funded by Pranarom International and Wallonia DGO6 (grant n◦6282).

**Conflicts of Interest:** The authors declare no conflict of interest. The funders had no role in the design of the study; in the collection, analyses, or interpretation of data; in the writing of the manuscript, or in the decision to publish the results.

## **References**


© 2020 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 (http://creativecommons.org/licenses/by/4.0/).

## **Use of Essential Oils to Increase the Safety and the Quality of Marinated Pork Loin**

**Lorenzo Siroli 1,**†**, Giulia Baldi 1,**†**, Francesca Soglia 1, Danka Bukvicki 2, Francesca Patrignani 1,3, Massimiliano Petracci 1,3 and Rosalba Lanciotti 1,3,\***


Received: 8 July 2020; Accepted: 21 July 2020; Published: 24 July 2020

**Abstract:** This study aimed at evaluating the effects of the addition of an oil/beer/lemon marinade solution with or without the inclusion of oregano, rosemary and juniper essential oils on the quality, the technological properties as well as the shelf-life and safety of vacuum-packed pork loin meat. The results obtained suggested that, aside from the addition of essential oils, the marination process allowed to reduce meat pH, thus improving its water holding capacity. Instrumental and sensorial tests showed that the marination also enhanced the tenderness of meat samples, with those marinated with essential oils being the most positively perceived by the panelists. In addition, microbiological data indicated that the marinated samples showed a lower microbial load of the main spoiling microorganisms compared to the control samples, from the 6th to the 13th day of storage, regardless of the addition of essential oils. Marination also allowed to inhibit the pathogens *Salmonella enteritidis*, *Listeria monocytogenes* and *Staphylococcus aureus,* thus increasing the microbiological safety of the product. Overall outcomes suggest that the oil/beer/lemon marinade solution added with essential oils might represent a promising strategy to improve both qualitative and sensory characteristics as well as the safety of meat products.

**Keywords:** essential oil; marinating solution; pork loin; quality; safety

## **1. Introduction**

In the past decade, global consumer demand for marinated meat products has significantly increased [1,2]. The reasons behind this scenario are mainly related to the nutritional characteristics, the extended shelf-life as well as the improvement of sensorial and textural traits of this kind of commodity [2,3]. In addition, marination technology allows to diversify meat products and, conferring them peculiar sensorial traits, to offer a broader choice to the consumers [4]. Marination is a widely used process in the meat industry consisting in the injection or immersion of meat cuts into aqueous solutions containing a wide range of ingredients such as water, salt, vinegar, lemon juice, wine, soy sauce, brine, essential oils, tenderizers, herbs, spices and organic acids [5,6]. Depending on the selected ingredients, a huge variety of marinade solutions, either alkaline or acid, exists. The firsts contain phosphates, while the seconds are usually prepared with the addition of organic acids or their salts [7,8]. Another type of marinade solution are the water/oil emulsions. Overall, the addition of marinade solutions to a meat cut is usually performed to improve the production yields (i.e., by increasing the

moisture content of the product), improve the organoleptic characteristics of the final product and, eventually, limit (or at least retard) the occurrence of oxidative reactions [9–11]. In addition, recent studies have reported that marinade solutions including "natural" ingredients (e.g., spices, herbs, essential oils, etc.) can exert an antimicrobial effect against pathogenic and spoilage microorganisms in poultry, beef and pork meat [5,12,13]. Aside from their ability to improve the safety and the shelf-life of marinated meat [14], the utilization of ingredients such as essential oils may also enhance consumers' willingness to buy, in light of the recent increasing attitude towards the consumption of clean-label products [15].

The use of essential oils or of their components (extracted from flowers, fruits, roots, buds and leaves through distillation processes) is widespread in the food industry, precisely in light their organoleptic, antimicrobial and antioxidant properties [16–18]. Within this context, a remarkable antimicrobial effect of several essential oils (included during processing) has been recently highlighted. To cite some examples, the use of rosemary essential oil (0.05%) on beef and chicken meat was found to be able to inhibit the growth of *Listeria monocytogenes*, *Escherichia coli* and *Staphylococcus aureus* [19,20]. On the other hand, the inclusion of thyme essential oil (0.08%) allowed to inhibit the growth of both spoiling microorganisms such as *Pseudomonas* spp. and pathogens such as *Staphylococcus aureus* [16]. Oregano essential oil has been found to exert antimicrobial effects against various pathogenic microorganisms such as *Escherichia coli*, *Listeria monocytogenes* and *Salmonella enteritidis* on both beef and pork meat [16]. However, it is noteworthy to mention that, as essential oils have low sensory thresholds [17], their sensory compatibility as well as their impact on the sensory profile of the final product should be carefully considered [21,22].

In addition, the flavor innovation represents a marketing strategy aimed at keeping up with the continually changing food trends [23]. Within this context, creating appealing alternatives for the consumers represents an important challenge for the meat industry. As a matter of fact, the possibility to set-up a marinade solution with typical ingredients of the Mediterranean area could certainly offer an added value to the final product and differentiate it from the alternatives currently existing on the market. In this framework, the purpose of this research was to evaluate the effect of the addition of a marinade solution composed by olive oil, beer and lemon (i.e., typical ingredients from Mediterranean area) with or without the inclusion of a mixture of essential oils on the shelf-life, the safety as well as the sensory and quality traits of vacuum-packed pork loin slices.

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

## *2.1. Preliminary Tests: Selection of the Marinade Solution's Composition and Essential Oils Mixture*

Preliminary tests were performed on pork loin slices (weighing about 60 g) in order to set the best combination and concentration of ingredients in the marinade solution in terms of either organoleptic traits (taste, smell, tenderness) and technological properties (absorption of the marinade solution, tenderness, color, etc.). In detail, 8 ingredients (i.e., water, lemon juice, olive oil, balsamic vinegar, red wine, white wine, beer and mustard) have been tested through different combinations and ratios as well as percentage of marinade solution (*w*/*w*) added to the meat slices, as reported in Table 1.


**Table 1.** Marinade solutions tested in the preliminary trials.


**Table 1.** *Cont.*

With the aim to obtain homogenous solutions, the ingredients of each combination were mixed with an Ultraturrax (IKA–WERKE, Labortechnik, Staufen, Germany) (13,000 rpm, 30 s, in ice). To each slice of pork loin, 1% NaCl (*w*/*w*) was added in the marinated product. The samples were placed in heat-resistant plastic bags, in which the marinating solution was directly added. The slices were then vacuum packaged (99.9%) and placed in a small-scale tumbler (model MHG-20, VakonaQualitat, Lienen, Germany) under vacuum conditions (−0.95 bar) and at a temperature of 2–4 ◦C. Tumbling was performed in 60 min at a speed of 20 rpm including two working cycles (25 min per cycle) and a 10 min pause cycle.

Subsequently, in order to select the combination allowing to obtain the best organoleptic properties of the product without altering its flavor, the addition of essential oils to the marinade solutions was tested. The essential oils considered during the preliminary tests were thyme, rosemary, oregano, and juniper, in different combinations and concentrations (0.02, 0.04, 0.06 and 0.08% on the final product). The evaluation was done by an untrained panel of 20 panellist taking into consideration the sensory parameters such as color, odour, overall accettability before and after cooking.

On the basis of preliminary results (data not shown), the marinade solution selected for the main experiment was composed by olive oil/beer/lemon juice (1:2:1, 10% *w*/*w*) with a mixture of oregano (0.02%), rosemary (0.03%) and juniper (0.03%) essential oils.

## *2.2. Ingredients and Microorganisms Used*

The pork loin slices used in this work were obtained from a local retailer the same day of the trial and kept at refrigerated temperatures (4 ± 1 ◦C) until the analyses. The marinade solution was composed as follows: the bock style beer Moretti la rossa (7.2% ABV) (Heineken Italia S.p.A., Pollein, AO, Italy), extra virgin olive oil (Monini, Spoleto, PG, Italy) and concentrated lemon juice (LIMMI, Perugia, PG, Italy). The essential oils used in this experimentation were oregano, rosemary, and juniper (Flora, Pisa, PI, Italy).

The strains used in the challenge test trial, *Listeria monocytogenes* Scott A, *Salmonella enteritidis* E5 and *Staphylococcus aureus* SR41 belonged to the Department of Agricultural and Food Sciences of Bologna University. The strains were maintained at −80 ◦C before experiments and before inoculation they were cultured twice in Brain Heart Infusion broth (BHI, Oxoid Ltd. Basingstoke, UK) at 37 ◦C for 24 h.

## *2.3. Preparation of the Samples and Shelf-Life Trials*

The experiment was carried out on a total of 81 slices of pork loin (having an average weight of 60 g), divided into 3 groups (27 slices/group) as follows:


As previously mentioned, the marinade solution was realized by mixing bock style beer, concentrated lemon juice and extra virgin olive oil at a 2:1:1 ratio using an Ultraturrax (IKA–WERKE, Labortechnik, Staufen, Germany) (13,000 rpm, 30 s, in ice). Part of this solution was used for samples belonging to the experimental group M, while the remaining was added of a mixture of essential oils (0.08% of the final weight) consisting of oregano (0.02%), rosemary (0.03%) and juniper (0.03%) and included in the samples M + E. Each pork loin slice (about 60 g), was added of 1% NaCl, calculated on the final weight of the marinated product. Subsequently, the samples were placed in heat-resistant plastic bags, in which the marinating solution was directly added, with the only exception of the samples belonging to the control group to which only 1% NaCl was included. The amount of marinade solution added to the samples corresponded to 10% (*w*/*w*) of the final product. The slices were then vacuum packaged (99.9%) and placed in a small-scale tumbler (model MHG-20, VakonaQualitat, Lienen, Germany) under vacuum conditions (−0.95 bar) and at a temperature of 2–4 ◦C. Tumbling was performed in 60 min at a speed of 20 rpm including two working cycles (25 min per cycle) and a 10 min pause cycle. The vacuum-tumbled loin slices were then stored at 4 ◦C and used for analytical determinations after 3, 9 and 15 days of storage.

## 2.3.1. pH

The pH of the samples was determined by taking an aliquot of meat (avoiding fat and connective tissue) according to Jeacocke [24]. About 2.5 g of finely chopped meat were homogenized for 30 s by Ultraturrax in 25 mL of a solution 5 mM of sodium iodoacetate and 150 mM of KCl at pH 7.0. The pH was determined by pH meter (mod. Jenway 3510; Electrode 924001, Cole-Parmer, Stone, UK) previously calibrated. The pH determination was performed after 3, 6 and 15 days of refrigerated storage on raw meat samples.

## 2.3.2. Color

Color was assessed by a Minolta® CR-400 colorimeter (Milan, Italy), previously calibrated using a standard white ceramic tile, in standardized illuminant (C) and observation angle (0◦ with respect to an area of 8 mm in diameter) conditions. The CIELAB system [25] was utilized and the parameter of lightness (L\*), redness (a\*) and yellowness (b\*) were used to objectively define color. The color determination was performed, for each group, after 3, 9 and 15 days of refrigerated storage on raw meat samples.

## 2.3.3. Marinade Uptake

Marinade uptake (i.e., the ability of meat to bind the saline solution added) was calculated by the difference in weight of the samples before and after the marination process. The amount of marinade solution absorbed was calculated as a percentage of the initial weight of the meat sample, according to the formula:

Marinade uptake (%) = [(Weight after marination − Initial weight)/Initial weight] × 100

## 2.3.4. Cooking Loss

After 3, 9 and 15 days of storage, samples were cooked in a in a stone grill (model GL-33, Fimar, Rimini, Italy) in standardized conditions (200 ◦C, 190 s) and the cooking loss (amount of liquid lost after cooking) was calculated as a percentage of the initial weight of the sample according to the formula:

$$\text{Cooling loss (\%)} = \text{[(Raw weight - Gooded weight)/Raw weight]} \times 100$$

## 2.3.5. Shear Force

Shear force was assessed by a texture analyzer TA-HDi 500 (Stable Micro System, Godalming, Surrey, UK) equipped with a 5-kg load cell and a Warner-Bratzler shear probe. From each cooked sample, sub-samples (having the dimension of 4 × 1 × 0.5 cm) were excised and placed inside the load cell. The resulting shear force was expressed as kg/cm2.

## 2.3.6. Sensory Analysis

Panel tests were performed after 3, 9 and 15 days of refrigerated storage on cooked samples in order to test their visual appearance, olfactory acceptability and taste. The analysis was carried out by 20 untrained panelists who evaluated on a 1 to 5 scale the following parameters: meat odor intensity, spicy odor intensity, color intensity, flavor intensity, tenderness, overall assessment and finally favorite sample.

## 2.3.7. Microbiological Analysis

During storage at 4 ◦C, the cell count over time of lactic acid bacteria, yeasts, total aerobic mesophilic bacteria, total aerobic psychrotrophic bacteria, *Pseudomonas* spp. and *Brochotrix thermosphacta* was evaluated by plate counting in specific agar media. Aerobic mesophilic and psychotrophic bacteria were detected on Plate Count Agar (PCA, Oxoid Ltd., Basingstoke, UK), lactic acid bacteria on de Man Rogosa and Sharpe Agar (MRS, Oxoid Ltd. Basingstoke, UK) with added 0.05% cycloheximide (Sigma-Aldrich, St. Louis, US), yeasts on Sabouraud Dextrose Agar (SAB, Oxoid Ltd. Basingstoke, UK), added to 0.02% chloramphenicol (Sigma-Aldrich, St. Louis, US), *Pseudomonas* spp. on Pseudomonas Agar Base (PAB, Oxoid Ltd. Basingstoke, UK) supplemented with Pseudomonas CFC selective agar supplement (Oxoid Ltd. Basingstoke, UK) and *Brochotrix thermosphacta* on STAA Agar base (Oxoid Ltd. Basingstoke, UK) supplemented with STAA selective supplement (Oxoid Ltd. Basingstoke, UK). To perform microbiological analyses, 10 g of meat sample were diluted into 90 mL of physiological solution (0.9% (*w*/*v*) NaCl), homogenized by a BagMixer 400 P (Interscience, St Nom la Bretèche, France), followed by serial dilution in physiological solution. The MRS agar plates were incubated 24 h at 37 ◦C, the PCA plates for the detection of psychrotrophic bacteria were incubated at 10 ◦C for 7 days, all the other agar media were incubated at 30 ◦C for 24–48 h.

## *2.4. Challenge-Test Trials*

The preparation of marinated pork loin was done similarly to what reported in paragraph 2.3. The experiment was carried out on a total of 60 slices of pork loin (having an average weight of 60 g), divided into 3 groups (20 slices/group). Three groups of samples were obtained:


*Listeria monocytogenes* Scott A, *Salmonella enteritidis* E5 and *Staphylococcus aureus* SR231, used in the challenge test belongs to the Department of Agricultural and Food Sciences (DISTAL, University of Bologna) collection. The bacterial strains were cultured overnight two times in Brain Heart Infusion (Oxoid Ltd., Basigstone, UK) at 37 ◦C. The pathogens were directly inoculated on the loin slices through 0.5 mL of physiological solution for the control group, while for the groups M + P and M + E + P were added to the marinating solution before the addition to the product. The inoculum was done in order to have an initial cell load of the pathogens, on the product, of approximately 4.0 log CFU/g. After the addition of the marinating solution the product was packaged and churned as reported in paragraph 2.3. The samples were stored at 4 ◦C and used for microbiological analyses immediately after the marinating and after 3, 6, 9, 13 and 15 days.

## *2.5. Microbiological Analysis*

During the storage microbiological analyses were performed in order to detect the cell loads of the inoculated *L. monocytogenes*, *S. enteritidis* and *S. aureus*. Specifically, the entire slice of loin (about 60 g) was placed in sterile bags and added with sterile physiological solution in a 1:2 (*w*/*w*) ratio and then homogenized for 2 min by a BagMixer 400 P (Interscience, St Nom la Bretèche, France) followed by serial dilution in physiological solution. *L. monocytogenes*, *S. enteritidis* and *S. aureus* were detected in specific selective agar media. Listeria Selective Agar (LSA, Oxoid Ltd., Basingstoke, UK) supplemented with Listeria selective supplement (SR0140, Oxoid Ltd., Basingstoke, UK) for the enumeration of *L. monocytogenes*; Bismuth Sulphite Agar (BSA, Oxoid Ltd., Basingstoke, UK) for the detection of *S. enteritidis*, while Baird-Parker Agar base (BPA, Oxoid Ltd., Basingstoke, UK) added with Egg Yolk Tellurite Emulsion (SR0054, Oxoid Ltd., Basingstoke, UK) for the enumeration of *S. aureus*. The agar plates were then incubated at 37 ◦C for 24 h.

## *2.6. Statistical Analysis*

Data were analyzed using the one-way ANOVA option of Statistica software (version 8.0; StatSoft., Tulsa, Oklahoma, USA) in order to test the effect of the addition of a marinade solution (with or without the inclusion of essential oils) at each sampling time (3, 9 and 15 days). Following, mean values were separated through Tukey honest significant difference (HSD) test, by considering a significance level of *p* < 0.05.

#### **3. Results**

## *3.1. Shelf-Life Trials*

## 3.1.1. pH and Color

As reported in Figure 1, at each sampling time, control samples showed significantly higher pH than the marinated ones (*p* < 0.05) which, in their turn, exhibited similar values. A slight decrease in pH following refrigerated storage was observed for all the experimental groups, with C samples showing the greatest pH decline. In more detail, control samples exhibited an average pH decrease of 0.32 units, while M and M + E decreased of 0.19 and 0.17, respectively.

**Figure 1.** Average pH values of non-marinated (C), marinated (M) and marinated with essential oils (M + E) pork loin slices at 3, 9 and 15 days of refrigerated storage. Data represent means ± SD. a, b = average values lacking a common letter significantly differ among the same sampling time.

Results concerning the evolution of color parameters (lightness—L\*, redness—a\*, yellowness—b\*) during the refrigerated storage are reported in Figure 2. Overall, regardless the storage time, no significant differences were found either in L\* or a\* values among the experimental groups. Although these differences were not statistically significant, non-marinated samples showed noticeably higher a\* values at both 9 and 15 days of storage. On the contrary, marination treatment exploited a remarkable effect on yellowness (b\*): at each storage time, both M and M + E exhibited significantly higher b\* values if compared to the control (*p* < 0.05).

**Figure 2.** Average lightness (L\*), redness (a\*) and yellowness (b\*) values of non-marinated (C), marinated (M) and marinated with essential oils (M + E) pork loin slices at 3, 9 and 15 days of refrigerated storage. Data represent means ± SD. a, b = average values lacking a common letter significantly differ among the same sampling time. At the same storage time, ns indicates no significant differences among the samples.

#### 3.1.2. Marinade Uptake and Cooking Loss

Data concerning the marinade uptake during the refrigerated storage are shown in Figure 3. Albeit any difference has been detected among the experimental groups at 9 and 15 days of storage, at day 3, a significantly (*p* < 0.05) higher marinade uptake has been observed in M + E samples in comparison to M (7.8 vs. 7.3%, respectively).

**Figure 3.** Average marinade uptake (%) values of marinated (M) and marinated with essential oils (M + E) pork loin slices at 3, 9 and 15 days of refrigerated storage. Data represent means ± SD. \*\*\* = *p* < 0.001. At the same storage time, ns indicates no significant differences among the samples.

Results concerning the cooking losses at different storage times are shown in Figure 4. After 3 days of refrigerated storage, C (non-marinated samples) showed significantly higher liquid losses if compared to M + E samples (*p* < 0.001), while M group exhibited intermediate values. However, different results were observed at day 9 with the C group showing significantly lower values if compared to M, whereas no significant differences were found at day 15.

**Figure 4.** Average cooking loss values (%) of non-marinated (C), marinated (M) and marinated with essential oils (M + E) pork loin slices at 3, 9 and 15 days of refrigerated storage. Data represent means ± SD. a, b = average values lacking a common letter significantly differ among the same sampling time. At the same storage time, ns indicates no significant differences among the samples.

#### 3.1.3. Shear Force

Results concerning the shear force of cooked pork loin samples after 3, 9 and 15 days of refrigerated storage are displayed in Figure 5. After 3 days of refrigerated storage, non-marinated samples showed significantly higher shear forces than the marinated ones (M and M + E) (*p* < 0.05), with M + E group exhibiting the lowest values. Albeit no statistical difference has been detected at 9 and 15 days likely due to the high variability of data, M + E samples showed the lowest shear force values, thus suggesting that the effect of essential oils on improving meat tenderness is considerable in particular in the first days of storage.

**Figure 5.** Average shear force values (kg/cm2) of non-marinated (C), marinated (M) and marinated with essential oils (M + E) pork loin slices at 3, 9 and 15 days of refrigerated storage. Data represent means ± SD. a, b = average values lacking a common letter significantly differ among the same sampling time. At the same storage time, ns indicates no significant differences among the samples.

## 3.1.4. Sensory Analysis

Panel tests were performed on pork loin samples after 3, 9 and 15 days of storage with the aim of determining the acceptability of the product by the consumers. The results of the panel tests are shown in Figure 6a–c.

**Figure 6.** *Cont.*

(**c**)

**Figure 6.** Sensory data of pork loin slices after 3 days (**a**), 9 days (**b**) and 15 days (**c**) of storage in relation to the sample (Control (C), Marinated (M), marinated with essential oils (M + E)). Data represent means ± SD. a, b, c = average values of each sensorial parameter lacking a common letter significantly differ among the same sensory parameter.

The results showed that, regardless the sampling time, the marinated meat, and especially that with essential oils (M + E), exhibited better scores compared to the non-marinated one, with the only exception of meat flavor intensity parameter. In addition, the marinated samples showed a greater intensity of flavor and taste, positively perceived by the panelists. In particular, marinated meat slices showed higher tenderness, color, flavor and taste intensities for the whole storage period, resulting in an overall improved acceptability compared to the controls. Considering the effect of essential oils, no differences between M and M + E samples were observed after 3 days of storage. However, starting from the second panel test (day 9), M + E samples showed higher scores for spicy flavor and taste intensity compared to M samples. The differences among M and M + E samples intensified at the end of storage (day 15), when the M + E group showed the highest scores for overall acceptability, thus being the preferred sample for over 60% of panelists.

#### 3.1.5. Microbiological Analysis

The microbiological analyses were aimed to detect various microbiological groups frequently associated with the spoilage of processed meat products. In particular, during the refrigerated storage of the samples, the cell loads of total aerobic mesophilic and psychotropic bacteria, lactic acid bacteria, yeasts, *Pseudomonas* spp., total coliforms and *Brochotrix thermosphacta* were detected.

In Figure 7, the cell loads of mesophilic aerobic bacteria, lactic acid bacteria, yeasts, *Pseudomonas* spp., total coliforms and *Brochotrix thermosphacta* are reported.

**Figure 7.** Cell load (log CFU/g ± SD), during the refrigerated storage, of total aerobic mesophilic bacteria, yeast and lactic acid bacteria (**a**) *Brochothrix thermosphacta*, *Pseudomonas* spp. and total coliforms (**b**) in different pork loin slices: Control (C), Marinated (M), marinated with essential oils (M + O). Data represent means ± SD. a-b-c = for each microorganism, at the same time of storage, average values lacking a common letter significantly differ among the same sampling time (*p* < 0.05).

The data obtained indicated a satisfactory microbiological quality of the raw meat. In fact, the initial cell load of the main spoiling microorganisms was below 3.0 log CFU/g, independently on the use of marinade solution or the addition of essential oils. As expected, the mesophilic bacteria represented the main microbial spoiling group. In fact, a fast increase of the cell load of this group was observed in all the samples starting from the sixth day of refrigerated storage. However, from day 6 of storage, samples M and M + E showed significant lower cell loads compared to C, while no differences were observed between M and M + E samples. The C samples were the only ones found to exceed 8.0 log CFU/g after 15 d of storage. The same trend was observed for psychotropic aerobic bacteria.

A similar tendency was observed for *Pseudomonas* spp. Starting from day 3 of storage C samples showed significant higher cell loads compared to M and M + E samples. No significant differences regarding the cell load of *Pseudomonas* spp. were observed between M and M + E samples, with the only exception of day 3. At the end of the storage *Pseudomonas* spp. resulted 6.67, 5.61 and 5.88 log CFU/g respectively in C, M and M + E samples. Total coliforms resulted significantly lower in M and M + E samples compared to C ones, excepted at day 3 of storage. The greatest differences were observed at day 15 when coliforms were 5.25, 4.18 and 4.22 log CFU/g respectively in samples C, M and M + E. In general, the highest inhibition due to marination and the addition of essential oils was observed against the Gram-negative bacteria *Pseudomonas* spp. and total coliforms. Otherwise, minor differences were observed considering *B. thermosphacta* since no significant differences were observed starting from day 9 of storage. However, depending on the sample, this microorganism reached a cell load ranging between 4.4 and 4.8 log CFU/g.

A different trend was observed for yeasts and lactic acid bacteria. In fact, starting from day 6 of storage, yeasts resulted significantly higher in samples M and M + E compared to the control. However, yeasts never exceed 5.0 log CFU/g for the whole period of storage. In case of lactic acid bacteria, no significant differences were detected at the end of the storage among the samples.

## *3.2. Challenge Test*

In order to evaluate the effects of the marinade solution with or without essential oils on the safety of vacuum packed pork loin slices, a challenge test inoculating *Listeria monocytogenes* Scott A, *Salmonella enteritidis* E5 and *Staphylococcus aures* SR31 was performed. Figure 8a–c shows the cell loads of the pathogen microorganisms during the refrigerated storage.

#### (**a**)

**Figure 8.** *Cont.*

$$\left(\mathsf{b}\right)$$

**Figure 8.** Cell load (log CFU/g ± SD), during refrigerated storage, of *Listeria monocytogenes* (**a**), *Salmonella enteritidis* (**b**), and *Staphylococcus aureus* (**c**). in different pork loin samples: Control (C), Marinated (M), marinated with essential oils (M + E). Data represent means ± SD. a, b, c = for each microorganism, at the same time of storage, average values lacking a common letter significantly differ among the same sampling time (*p* < 0.05).

It is noteworthy to mention that marination allowed a significant (*p* < 0.05) reduction of the initial microbial cell load of all the pathogens, regardless of the presence or absence of essential oils. The highest initial cell load reduction, compared to control samples, was observed for *S. aureus*, and ranged between 0.7 and 1.0 log CFU/g, followed by *S. enteritidis* (0.7–0.8 log CFU/g) and *L. monocytogenes* (0.5–0.6 log CFU/g). In all cases, the differences in the pathogen levels between marinated and not marinated samples increased during the storage period. At the end of the storage, M and M + E samples showed cell loads lower than 2.0 logarithmic cycles for *L. monocytogenes* and *S. aureus* and lower than 1.5 logarithmic cycles in the case of *S. enteritidis*. On the contrary, an increase of the level of all the pathogens in C samples, greater in the case of *L. monocytogenes*, was observed during storage. Contrarily, a decrease of the pathogen loads in the marinated products was observed during the storage but without allowing their complete inactivation. Considering the effect of the addition of essential oils,

no significant differences were found between the samples M and M + E for *S. enteritis* and *S. aureus* while in the case of *L. monocytogenes* the samples M + E showed a significantly lower cell load with respect to samples M starting from day 13 of storage. The greatest antimicrobial effect from marinating was observed against *S. aureus*. In fact, a reduction of more than 3.5 log CFU/g at the end of storage compared to the initial load of C samples was observed.

## **4. Discussion**

The marinade solution prepared with extra virgin olive oil, beer, concentrated lemon juice and a mixture of essential oils used within this study was selected based on the findings of preliminary trials. Considering that offering a marinated product including typical ingredients and flavors belonging to the Mediterranean diet may represent an added value to product itself, all the marinade ingredients and essential oils chosen in this work derive from plants commonly used in the traditional recipes of this area. The selected marinade solution was then tested with the aim of exploring its effect on the shelf-life, safety and quality traits of pork loin slices during refrigerated storage.

Aside from the inclusion of essential oils, the addition of the marinade solution significantly reduced the pH of vacuum-packed pork loin. These outcomes might be ascribed to the addition of an acid marinade solution in which the inclusion of beer (pH = 3.96) and concentrated lemon juice (pH = 2.26) results in a remarkable reduction in pH. This might be desirable for several reasons. First, meat pH exerts a direct effect on its water holding capacity (WHC), since it is generally held that the ability of meat to retain water progressively improves above and below pH values corresponding to the isoelectric point of meat proteins (i.e., 5.5 in the case of pork meat) [26]. Furthermore, processed meat products with a low pH are less likely to develop pathogen microbial growth and off-odors, thus having an improved safety and shelf-life [27,28]. Lastly, reduced pH values might also be advantageous to facilitate the action of collagenases and other proteolytic enzymes responsible for meat tenderization during the refrigerated storage [29].

The addition of the marinade solution, regardless of the use of essential oils, also exerted a significant effect on the yellowness (b\*) of meat samples, while lightness (L\*) and redness (a\*) were not affected. The higher b\* values detected for marinated samples might be likely due to the presence of coloring compounds in the solution itself (i.e., extra virgin olive oil, beer and concentrated lemon juice) which might have increased the yellowness of samples. However, the increase in b\* values did not negatively affect the sensory evaluation by panelists who associated to the marinated samples in general, and to those including essential oils in particular, a better color retention if compared to the control.

Beside all, the marinating process is a widely used procedure at industrial level implemented with the aim to improve not only the sensory and eating qualities of meat products but also their technological properties, with a special reference to WHC [30,31]. Accordingly, satisfactory marinade uptakes (of more than 7%) were observed for both marinated pork loin groups after 3 days of storage. Albeit little literature is available concerning the effects of essential oils to improve the technological properties of meat, the remarkable improvement in marinade uptakes might be ascribed to the acid pH of the marinade solution. Indeed, as lemon juice contains citric acid, this ingredient is often included within the marinade solution to improve meat WHC by lowering its pH [32]. These outcomes are in agreement with those reported by other authors that observed a marinade uptake ranging between 4.6 and 9.7% in acidic marinated *Longissimus dorsi* muscles [33]. However, it is noteworthy to remember that the marinade uptake is strongly related to the meat type, marination technique as well as the duration of the process [34].

The marination process allowed to remarkably reduce the cooking losses compared to control samples after 3 days of refrigerated storage. This trend is in agreement with what reported by Gao et al. [35] who assessed the effect of marination on the main quality aspects of vacuum-packed pork loin meat. However, after both 9 and 15 days of storage, marinated meat (either M or M + E) exhibited slightly higher cooking losses if compared to the control group. This trend might be likely due to the greater marinade uptake measured during the storage period, which might have resulted in a higher loss of fluids during cooking. Therefore, it is reasonable that raw meat, added with salt without the inclusion of marinade solution, presented reduced cooking losses after a week of refrigerated storage.

Several authors have reported an increase in tenderness of marinated poultry, pork and beef [11,32,35]. Accordingly, the addition of marinade solution with or without essential oils allowed to reduce the shear forces of pork loin meat of about 40% and 22.8%, respectively, just after 3 days of refrigerated storage. These outcomes suggest the effectiveness of an acidic marinade solution to improve the tenderness of meat samples, as previously reported by Miller [36]. Accordingly, several studies have reported that acidic substances in the marinating solution (including lemon juice) can play a crucial role in the tenderization of marinated meat, leading to meat fibers swelling and enhancing proteolysis [37,38].

The sensory analysis data, according to the available literature, suggested that the marinated samples, and in particular those in which essential oils were added to the marinade, were tender and characterized by better color, flavor and taste intensity compared to the control samples. On the other hand, the positive effect of acidic marinade solutions on tenderness and other quality characteristics of different types of meat is widely reported in the literature [2,39]. The addition of essential oils strongly increased the overall acceptability of the samples, especially at the end of the storage, resulting in the preference of the consumers. Recently, many studies have reported an improvement of the sensory qualities and an extended shelf life of meat and meat products supplemented with different essential oils including, rosemary, thyme, oregano, basil, coriander, ginger, garlic, clove, juniper and fennel, used alone or in combination [40,41]. In addition, essential oils are widely reported as characterized by a strong antioxidant activity [42,43]. A wide literature reports a reduction of the lipid oxidation of meat and meat products added with essential oils during storage [40,44,45]. A better sensory quality and a longer shelf-life is normally associated to the reduction of lipid oxidation [45,46].

The predominant spoiling bacteria associated to refrigerated pork and beef, are *Pseudomonas* spp. during storage in aerobic conditions and lactic acid bacteria belonging to the genus *Lactobacillus* spp., *Leuconostoc* spp. and *Carnobacterium* spp. but also *Brochothrix thermosphacta*, *Enterobacteriaceae* and psychrophilic *Clostridium* spp. in case of anaerobic conditions [47,48]. Meat defects due to off-odors and off-flavors normally linked to a discoloration, gas production and acidification are generally associated to the growth of these microorganisms [49–51]. Our results indicate a satisfactory initial microbiological quality of the pork loin used in the present study. In fact, for all the main microbiological spoilage agents considered, the cell load was lower than 3.0 log CFU/g. During storage, an increase of the total viable mesophilic and psychotropic bacteria, *Pseudomonas* spp. lactic acid bacteria and *B. thermosphacta* was observed. The enumeration of total viable mesophilic and psychotropic microorganisms represents one of the most widely used and recognized criteria for evaluating the microbiological quality of meat [52]. Generally, the product is considered acceptable when the cell load of these microorganisms is lower than 7.0 log CFU/g [53] and this level is generally taken at industrial level as the upper threshold to determine the product expiry date. Our results showed that marinated samples overcome this limit only after 15 days of storage while control samples exceeded the limit after 9 days of refrigerated storage. The marination, regardless the addition of essential oils, showed the highest inhibition against the Gram-negative bacteria *Pseudomonas* spp. and total coliforms. Several literature data showed that species belonging to *Pseudomonas* and other psychotropic microorganisms are the predominant cause of alteration of fresh packaged meat [54]. Several *Pseudomonas* spp. are responsible for the formation of superficial patinas and off-flavor when their concentration reaches levels between 7–8 log CFU/g in chilled meat products [55].

Currently, foodborne outbreaks caused by foodborne pathogens transmitted from meat product still represent a significant public health challenge [56]. Considering the last 10–15 years the most important foodborne bacterial pathogens associated to meat belong to *Salmonella* spp., *Escherichia coli*, *Campylobacter jejuni* and *Staphylococcus aureus* [57–59]. Our results showed a clear inhibitory effect of the tested marinades on the growth kinetic of *Listeria monocytogenes*, *Salmonella enteritidis* and *Staphylococcus*

*aureus*resulting in an increased safety of the product. In particular, the tested marinating solution proved an immediate inhibitory effect against all the pathogens. In addition, an increase of pathogens cell load during storage was observed in control samples, while the marinated products induced a more or less marked decrease of the pathogens load without allowing their complete deactivation. Regarding the addition of essential oils, a significant additional antimicrobial effect, compared to marinated samples, was observed only against *Listeria monocytogenes*. The antimicrobial activities of essential oils and their bioactive components are well known and reviewed in a wide literature even if strongly affected by microbial species, strains, and physico-chemical and process variables [17,18,21,60,61]. Although strain dependent and affected by application conditions, the greatest resistance of Gram-negative bacteria, due to the presence of the outer membrane acting as a barrier to hydrophobic molecules, to many essential oils is well known [62]. Among the Gram-positive bacteria, the very high resistance of *Staphylococcus aureus* to many stress factors and antimicrobials including essential oils and their components is well documented [62,63]. Also the action mechanisms of several essential oil components against many microorganisms, including the target microorganisms taken into consideration in the present research, have been clarified by molecular tools [64–67]. The limited antimicrobial effects of the essential oils in the present work is probably due to the masking effect of ethanol and its synergistic effects with low pH values and NaCl of marinade. In fact, as shown by Lanciotti et al. [68] studying the boundary between the growth and no growth of *Salmonella enteritidis*, *Bacillus cereus* and *Staphylococcus aureus* in the presence of different growth controlling factors through probabilistic models, the effects of ethanol on the limitation of growth of the considered species was significant also at concentration of about 1% and not merely additive with temperature and NaCl concentration. Also, the presence of organic acids and the pH reduction by marinade contribute to mask the effects of the essential oils on the target microorganisms considered [69].

Several authors have reported the antimicrobial effect of marinating solution components [10,12]. In particular the antimicrobial effect of some acidic marinade solutions containing alcoholic drinks is associated to the presence of ethanol but also to phenolic derivatives and organic acids, contributing the last to the reduction of the pH of the product [10,70,71]. In addition, the combination of organic acids, ethanol and sodium chloride can strongly inhibit several microorganisms including pathogens like *Salmonella*, *Listeria monocytogenes*, *Escherichia coli* and *Staphylococcus aureus* [72,73].

## **5. Conclusions**

The results of the present study highlighted that the marination of pork loin slices using a solution (formulated with typical ingredients from Mediterranean area) with a mix of extra virgin olive oil, beer and lemon juice (in the presence/absence of essential oils) allows to obtain an overall improvement of the technological and sensory properties of meat. In particular, panel test results suggest a clear preference for marinated products with the addition of essential oils. Furthermore, the tested marinade solution exerted a remarkable meat pH reduction and significant antimicrobial activity both towards the common spoiling microflora normally present on the product and on pathogenic microorganisms deliberately inoculated, improving product safety and shelf-life. The use of marinade allowed the extension of the shelf-life of six days. In addition, offering a marinated product formulated with typical ingredients and flavors belonging to the Mediterranean diet may represent an added value to product itself. However, the addition of essential oils did not lead to a further increase of the antimicrobial activity exerted by the marinade solution. Though, the results obtained in this study suggest that an optimization of the concentration and type of essential oils used for the marination of pork loin could further increase its antimicrobial activity.

**Author Contributions:** Conceptualization, M.P., R.L. and F.P.; methodology, L.S., M.P., R.L., F.P. and F.S.; software, G.B., L.S. and F.S.; validation, G.B., L.S. and F.S.; formal analysis, G.B. and L.S.; investigation, G.B., D.B., L.S. and F.S.; resources, M.P. and R.L.; data curation, G.B., L.S., D.B. and F.S.; writing—original draft preparation, G.B., L.S. and F.S.; writing—review and editing, L.S., F.P., M.P., D.B. and R.L.; visualization, G.B. and D.B.; supervision, F.P., M.P. and R.L.; project administration, M.P. and R.L. All authors have read and agreed to the published version of the manuscript.

**Funding:** This research received no external funding.

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

## **References**


© 2020 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 (http://creativecommons.org/licenses/by/4.0/).

## **Composition of the Essential Oil and Insecticidal Activity of** *Launaea taraxacifolia* **(Willd.) Amin ex C. Je**ff**rey Growing in Nigeria**

## **Moses S. Owolabi 1,\*, Akintayo L. Ogundajo 1, Azeezat O. Alafia 2, Kafayat O. Ajelara <sup>2</sup> and William N. Setzer 3,4,\***


Received: 12 June 2020; Accepted: 9 July 2020; Published: 11 July 2020

**Abstract:** The rice weevil (*Sitophilus oryzae*) is a pest of stored grain products such as rice, wheat, and corn. Essential oils represent a green environmentally-friendly alternative to synthetic pesticides for controlling stored-product insect pests. *Launaea taraxacifolia* is a leafy vegetable plant found in several parts of Nigeria. The leaves are eaten either fresh as a salad or cooked as a sauce. The essential oil obtained from fresh leaves of *L. taraxacifolia* was obtained by hydrodistillation and analyzed by gas chromatography/mass spectrometry (GC-MS). Twenty-nine compounds were identified, accounting for 100% of the oil composition. The major component classes were monoterpene hydrocarbons (78.1%), followed by oxygenated monoterpenoids (16.2%), sesquiterpene hydrocarbons (2.1%), oxygenated sesquiterpenoids (0.3%), and non-terpenoid derivatives (3.3%). The leaf essential oil was dominated by monoterpene hydrocarbons including limonene (48.8%), sabinene (18.8%), and (*E*)-β-ocimene (4.6%), along with the monoterpenoid aldehyde citronellal (11.0%). The contact insecticidal activity of *L. taraxacifolia* essential oil against *Sitophilus oryzae* was carried out; median lethal concentration (LC50) values of topical exposure of *L. taraxacifolia* essential oil were assessed over a 120-h period. The LC50 values ranged from 54.38 μL/mL (24 h) to 10.10 μL/mL (120 h). The insecticidal activity of the *L. taraxacifolia* essential oil can be attributed to major components limonene (48.8%), sabinene (18.8%), and citronellal (11.0%), as well as potential synergistic action of the essential oil components. This result showed *L. taraxacifolia* essential oil may be considered as a useful alternative to synthetic insecticides.

**Keywords:** essential oil composition; limonene; sabinene; citronellal; *Sitophilus oryzae*

## **1. Introduction**

Insects such as *Callosobruchus maculatus* (Fabr.) (bruchid beetle), *Sitophilus granarius* (L.) (wheat weevil), *S. oryzae* (L.) (rice weevil), *S. zeamais* (Motsch.) (maize weevil), and *Tribolium castaneum* (Herbst) (red flour beetle), are important pests that attack stored grains, causing widespread economic losses [1–3]. The long-term use of synthetic insecticides to control these pests has become problematic, however. Compounds such as chlorinated hydrocarbons, organophosphates, carbamates, etc., tend to be toxic to non-target organisms such as mammals, birds, and fish [4–6], they are persistent in the environment [7–10], and many stored-grain insect pests have developed insecticide resistance [11–13]. Essential oils have emerged as viable alternatives to synthetic pesticides for control of stored-grain

insect pests; they are generally non-toxic to mammals, birds, fish, or humans, have limited persistence, are readily biodegradable, and are renewable resources [14–17].

*Launaea taraxacifolia* (Willd.) Amin ex. C. Jeffrey (syn. *Lactuca taraxacifolia* (Willd.) Schumach, wild lettuce) is a leafy vegetable plant belonging to the Asteraceae (Compositae). The family consists of roughly 1100 genera, and 20,000 species distributed across several countries including Mexico, West Indies, Central and South America, Europe, North Africa, and tropical West African countries like Ghana, Senegal, Benin, and Nigeria [18]. *L. taraxacifolia* is a wild erect perennial herb that grows up to 1–3 m in height with 3−5 pinnately lobed leaves at the base of the stem in a rosette form. The plant is found singly or in clusters of rocky soil, but it is also cultivated in small open gardens near homes for family consumption. The leaves are eaten fresh as a salad or cooked as sauces [18–24]. The plant is known as 'efo yanrin' among the Yorubas of the southwestern part of Nigeria, 'ugu' among the Ibos of the eastern part of Nigeria, and 'nonon barya' among the Hausas of the northern part of Nigeria. Minerals, proteins, flavonoids, fatty acids, and vitamins have been reported to be found in the leaves of *L. taraxacifolia* [25,26]. The nutritional aspects of *L. taraxacifolia* have been reviewed [27,28]. The antioxidant and antiviral activities as well as the use of *L. taraxacifolia* leaves in treatment and control of blood cholesterol levels, blood pressure, and diabetes have been reported [29,30]. Phytochemical studies of *L. taraxacifolia* revealed that the plant possesses chemical classes such as phenolic glycosides, flavonoids, saponins and triterpenoids, which are known to have phytotherapeutic value for humans [25,31–34]. To the best of our knowledge, there is little or no information on the composition of the essential oil or the insecticidal activity of *L. taraxacifolia*. Therefore, the present research was undertaken with the aim of investigating the essential oil composition and evaluating the insecticidal potential of *L. taraxacifolia* leaves from southwestern Nigeria.

## **2. Materials and Methods**

## *2.1. Plant Materials*

The leaves of *L. taraxacifolia* were collected from Ipara, Badagry (6◦4 54.07 N and 2◦52 52.75 E) Lagos state, Nigeria. Botanical identification was done at the Herbarium, University of Lagos, Nigeria, where a voucher specimen (LUH: 7959) was deposited. Fresh leaves of *L. taraxacifolia* were cut into pieces, air dried, and pulverized in a blender to increase the surface area. A 450-g sample of blended *L. taraxacifolia* was hydrodistilled for 4 h in an all-glass modified Clevenger-type apparatus according to British Pharmacopoeia [35]. The obtained essential oil was stored in a sealed glass bottle with a screw lid cover under refrigeration at 4 ◦C until ready for use. Oil yield was calculated on a dry weight basis.

#### *2.2. Gas Chromatographic–Mass Spectral Analysis*

The chemical composition of *L. taraxacifolia* essential oil was determined by gas chromatography–mass spectrometry (GC-MS) using a Shimadzu GCMS-QP2010 Ultra operated in the electron impact (EI) mode (electron energy = 70 eV), scan range = 40–400 atomic mass units, with a scan rate of 3.0 scans per s, with GC-MS solution software. The GC column was a ZB-5 fused silica capillary column (30 m length × 0.25 mm inner diameter) with a 5% phenyl-polymethylsiloxane stationary phase and a film thickness of 0.25 μm. Helium gas was used as a carrier gas with column head pressure of 552 kPa at a flow rate of 1.37 mL/min. The injector temperature was 250 ◦C and the ion source temperature was 200 ◦C. The oven temperature of 50 ◦C was initially programmed for the GC and gradually increased at 2 ◦C/min to 260 ◦C. The sample (5% w/v) was dissolved in dichloromethane and 0.1 μL of the solution was injected using a split injection technique (30:1). Identification of the essential oil components was achieved by comparing the retention indices determined with respect to a homologous series of *n*-alkanes, and by comparison of the mass spectral fragmentation patterns with those stored in the MS databases [36–39].

## *2.3. Insecticidal Activity Screening*

The essential oil was screened for insecticidal activity based on the method of Ilboudo and co-workers [40] with modifications. *Sitophilus oryzae* (L.) (rice weevil) were reared on whole rice (10:1 w/w). Adult insects, 1–7 days old, were used for contact toxicity tests. The insects were cultured in a dark growth chamber at a temperature of 27 ± 1 ◦C with relative humidity of 65 ± 5%. The insecticidal activity of *L. taraxacifolia* oil against *S. oryzae* (rice weevil) was evaluated by treatment of Whatman No. 1 filter paper discs with the essential oil diluted in ethanol. The required quantities of oil (0.10, 0.20, 0.30, and 0.40 μL) were diluted to 1 mL with ethanol and applied to filter paper discs, respectively. Permethrin (0.6% w/w) and ethanol were used as positive and negative controls, respectively. The solvent was allowed to evaporate from the filter paper, which was then placed into polyethylene cups (80 mm diameter). Ten well-fed mixed sex adult *S. oryzae* were introduced into the polyethylene cups, containing 20 g uninfected rice grains, and covered with a muslin cloth, held in place with rubber bands. Each treatment was replicated four times. Control experiments were set up as described as above without the essential oil. The experiment was arranged in a complete randomized design on a laboratory bench. The insect was considered dead when the legs or antennae were observed to be immobile. Insect mortalities were investigated by observing the recovery of immobilized insects after 24 h intervals for 120 h and the percentage of insect mortality was corrected using the Abbott formula [41]. Probit analysis [42] using XLSTAT version 2018.1.1.60987 (Addinsoft™, Paris, France) was used to estimate median lethal concentration (LC50) values and insect toxicity data were analyzed using one-way ANOVA Tukey's honestly significant difference test.

## **3. Results and Discussion**

#### *3.1. Essential Oil Composition*

The essential oil from *L. taraxacifolia* was obtained by hydrodistillation with a yield of 1.68% as a pale-yellow essential oil, which was analyzed by GC-MS. The chemical composition of the leaf volatile oil of *L. taraxacifolia* is listed in Table 1. A total of 29 compounds were identified, accounting for 100% of the essential oil composition. The major chemical classes were monoterpene hydrocarbons (78%) and oxygenated monoterpenoids (16.2%), followed by sesquiterpene hydrocarbons (2.1%), oxygenated sesquiterpenoids (0.3%), and non-terpenoid derivatives (3.3%). The leaf essential oil was dominated by monoterpene hydrocarbons including limonene (48.8%), sabinene (18.8%), and (*E*)-β-ocimene (4.6%), along with the monoterpenoid aldehyde citronellal (11.0%). The chemical constituents of *L. taraxaciflora* essential oil have not been previously reported to the best of our knowledge. However, a phytochemical study and antioxidant and bacterial screening of the leaf extract of *L. taraxacifolia* have been reported [43].


**Table 1.** The chemical constituents of *Launaea taraxacifolia* leaf essential oil.


**Table 1.** *Cont*.

<sup>1</sup> RIcalc <sup>=</sup> Kovats retention index determined with respect to a homologous series of *<sup>n</sup>*-alkanes on a ZB-5 column. <sup>2</sup> RIdb <sup>=</sup> Retention index from the databases [36–39].

#### *3.2. Insecticidal Activity*

The contact toxicity of *L. taraxacifolia* against *S. oryzae* revealed considerable differences in insect mortality rate to the essential oil with different concentrations and different exposure times. Table 2 shows that at a dose of 10.00 μL/mL, the volatile oil produced 25.00% mortality after 48 h (not significantly different than the negative EtOH control) and 52.50% after 120 h (significantly higher toxicity than the EtOH control). The essential oil produced 30.00%, 47.50%, 60.00%, and 75.00% mortality after 48, 72, 96, and 120 h at a dose of 20.00 μL/mL, respectively, while a dose of 30.00 μL/mL yielded a mortality rate of 42.50%, 57.50%, 75.00%, and 75.00%, respectively, over the same period of time. With longer contact times (≥48 h), 20 μL/mL and 30 μL/mL concentrations of *L. taraxacifolia* essential oil was significantly more toxic than the EtOH control, but less toxic than the permethrin positive control. The highest concentration of 40.00 μL/mL produced a mortality of 97.50%, and 100.00% after 96 and 120 h, respectively, which is significantly comparable to the permethrin positive control. Permethrin (0.6% w/w) against *S. oryzae* caused 40.0% mortality with 24 h of exposure and 100.0% mortality after 48 h. The negative control showed no appreciable activity against *S. oryzae* until after 120 h.

**Table 2.** Contact insecticidal effects of *Launaea taraxacifolia* essential oil on adult mortality of *Sitophilus oryzae* reared on rice grains 120 h after treatment.


<sup>1</sup> Mean followed by different letters in a column is significantly different at (*p* < 0.05). Insect toxicity data were analyzed using one-way ANOVA followed by Tukey's test. <sup>2</sup> Degrees of freedom.

Median lethal concentration (LC50) values at 95% confidence limits over exposure of *L. taraxacifolia* essential oil were assessed and are shown in Table 3. After 120 h of exposure with an increase in concentration at regular intervals of 24 h, the LC50 values were 54.38, 31.64, 21.48, 16.38, and 10.10 μL/mL, respectively. In this study, the essential oil of *L. taraxacifolia* demonstrated contact toxicity to *S. oryzae*, since it had higher insecticidal activity with increasing essential oil concentration and exposure time. This result showed *L. taraxacifolia* essential oil to have promising insecticidal activity against *S. oryzae* and therefore may be considered as a useful, environmentally benign alternative to synthetic insecticides.

**Table 3.** Median lethal concentrations (LC50, μL/mL, and 95% confidence limits) of *Launaea taraxacifolia* essential oil against *Sitophilus oryzae*.


To best of our knowledge, there have been no previous literature reports on the insecticidal activity of *L. taraxacifolia* essential oil against *S. oryzae* insect pest. However, contact toxicity of both limonene and sabinene, the major chemical components in this present study, have shown insecticidal activity against *S. oryzae* [44]. Limonene has been previously reported to have a moderate contact effect against *S. zeamais* (LD50 values of 198.66 μg/cm2) and *S. oryzae* (with LD50 of 260.18 μg/cm2) [45] as well as fumigant toxicity against *S. oryzae* (24-h LC50 61.5 μL/L) [46]. Garcia et al. reported that limonene showed contact toxicity against *T. castaneum* [47]. Sabinene, on the other hand, demonstrated weaker insecticidal activity against *S. oryzae* (24-h LC50 463 μL/L) [44]. Interestingly, the *S. oryzae* fumigant insecticidal activities of limonene and sabinene parallel the acetylcholinesterase (AChE) inhibitory activities; AChE IC50 = 9.57 μL/mL and 85.03 μL/mL, respectively, for limonene and sabinene [48]. Furthermore, the binary combination of limonene + sabinene showed synergistic AChE inhibition [48]. The insecticidal activity of the *L. taraxacifolia* essential oil could be attributed to those known major components and the resulting synergistic action of the monoterpene hydrocarbons limonene (48.8%) and sabinene (18.8%).

The major aldehyde essential oil component, citronellal (11.0%), has also shown contact insecticidal activity against *Musca domestica* [49] and *S. oryzae* [50] and fumigant insecticidal activity against *T. castaneum* [51] and *S. zeamais* [52]. (–)-Citronellal has also shown AChE inhibitory activity with IC50 of 18.4 mM [50]. The contact toxicities of bornyl acetate, (+)-limonene, myrcene, α-phellandrene, α-pinene, sabinene, and terpinolene, essential oil constituents obtained from leaves of *Chamaecyparis obtusa*, against *Callosobruchus chinensis* (L.) and *Sitophilus oryzae* (L.) have been reported [44]. The insecticidal activity of the essential oil components 1,8-cineole, *p*-cymene, α-pinene, and limonene has been previously reported with the order of activity 1,8-cineole > *p*-cymene > α-pinene > limonene [46]. Abdelgaleil et al. reported a comparative study of eleven monoterpenes contact and fumigant toxicity: camphene, (+)-camphor, (−)-carvone, 1-8-cineole, cuminaldehyde, (L)-fenchone, geraniol, (−)-limonene, (−)-linalool, (−)-menthol, and myrcene, against two important stored products insects, *S. oryzae,* and *T. castaneum*, and discovered that the toxicity varied according to insect pest with *S. oryzae* more susceptible to most of the components than *T. castaneum* [53].

#### **4. Conclusions**

This study investigated the essential oil composition and evaluated the insecticidal potential of *L. taraxacifolia* leaves for the first time as a potential substitute to synthetic insecticides. *L. taraxacifolia* offers an advantage in Nigeria due to its accessibility and renewability. Despite many advantages of medicinal plants, especially the essential oils, further studies need to be conducted to ascertain the safety of this essential oil before its practical use as an insecticide for controlling stored product insect pests. In addition, while the insecticidal properties of *L. taraxacifolia* essential oil are promising, this work is preliminary and future investigations extrapolating the use of the essential oil under grain-storage conditions should be pursued. In addition, studies on the controlled-release formulations of the essential oil could be examined to curb some of the challenges of essential oil treatments such as rapid degradation, volatility, and low bioavailability of the essential oils.

**Author Contributions:** Conceptualization, M.S.O.; methodology, M.S.O, K.O.A., and W.N.S.; validation, W.N.S., formal analysis, K.O.A. and W.N.S.; investigation, M.S.O., A.L.O., A.O.A., K.O.A., and W.N.S.; data curation, M.S.O.; writing—original draft preparation, M.S.O.; writing—review and editing, M.S.O. and W.N.S.; supervision, M.S.O.; project administration, M.S.O. All authors have read and agreed to the published version of the manuscript.

**Funding:** This research received no external funding.

**Acknowledgments:** W.N.S. participated in this work as part of the activities of the Aromatic Plant Research Center (APRC, https://aromaticplant.org/).

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

## **References**


© 2020 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 (http://creativecommons.org/licenses/by/4.0/).

## *Article* **Thyme Antimicrobial E**ff**ect in Edible Films with High Pressure Thermally Treated Whey Protein Concentrate**

## **Iulia Bleoancă, Elena Enachi and Daniela Borda \***

Faculty of Food Science and Engineering, Dunarea de Jos University of Galati, 800201 Galati, Romania; Iulia.Bleoanca@ugal.ro (I.B.); Elena.Ionita@ugal.ro (E.E.)

**\*** Correspondence: Daniela.Borda@ugal.ro; Tel.: +40-336-130-177

Received: 5 June 2020; Accepted: 26 June 2020; Published: 30 June 2020

**Abstract:** Application of high pressure-thermal treatment (600 MPa and 70 ◦C, 20 min) for obtaining edible films functionalized with thyme extracts have been studied in order to evaluate the antimicrobial capacity of films structure to retain and release the bioactive compounds. The high pressure-thermally treated films (HPT) were compared with the thermally treated (TT) ones (80 ± 0.5 ◦C, 35 min). The film structures were analyzed and the sorption isotherms, water vapor permeability, antimicrobial activity and the volatile fingerprints by GC/MS were performed. The HPT film presented more binding sites for water chemi-sorption than TT films and displayed significantly lower WVP than TT films (*p* < 0.05). TT films displayed slightly, but significant higher, antimicrobial activity (*p* < 0.05) against *Geotrichum candidum* in the first day and against *Bacillus subtilis* in the 10th day of storage. The HPT film structure had ~1.5-fold higher capacity to retain volatiles after drying compared to TT films. From the HPT films higher amount of p-cymene and α-terpinene was volatilized during 10 days of storage at 25 ◦C, 50% RH while from the TT films higher amount of caryophyllene and carvacrol were released. During storage HPT films had a 2-fold lower capacity to retain monoterpenes compared to TT films.

**Keywords:** thyme; essential oil; edible films; high pressure thermal treatment; ultrasonication; antimicrobial; thymol; carvacrol; food safety

## **1. Introduction**

Consumers' increasing demand for minimally processed food products led to increased researchers' attention towards new ways to valorize the potential of plant-based extracts as preservatives for extending food shelf-life and insuring food safety. Essential oils (EOs), used conventionally as flavorings by the food industry are considered for new applications as antimicrobials and antioxidants and are generally recognized as safe (GRAS) by the United States Food and Drug Administration [1]. In the EU, Regulation 1334/2008 sets the maximum levels of certain substances present as flavorings in or on foods, EOs included.

The biological properties of EOs are determined by its components, which are typically low molecular weight terpenes and terpenoids, nonetheless other aromatic and aliphatic molecules could be present. From the aromatic plants volatile profile, terpenes (C10) are representing 90% of the EOs but sesquiterpenes (C15) are also frequently present [2]. Even though EOs have antimicrobial effect against a wide range of food related spoilage and pathogenic microorganisms, the required concentration is often too high and their intense odor may negatively interfere with food quality and consumers' acceptance. One solution to reduce the negative effect of EOs on food flavor is the inclusion of EOs into edible packaging, such as films and coatings.

Edible films (EF) are thin layers of edible materials (polysaccharides, proteins and lipids, and the combination of two or more of the above), which once formed can be placed on or between food components [3].

EF functionalized with EOs act as antimicrobial and antioxidant carriers, enabling their release at the interface between packaging and food product while maintaining the antimicrobial effect [4] and preserving food quality for longer periods of time.

Recently it has been demonstrated that lemongrass EOs have succeeded to limit the extent of depolymerization in chia mucilage emulsion and prevented autooxidation [5]. To overcome the EOs inherent photo-, thermal-sensitivity coupled with their high volatility micro and nanoencapsulation methods have been employed [6–9].

In the same time, edible packagings are an environment- friendly solution as their constituents are fully biodegradable and in some cases they valorize industrial waste, as is the case of whey protein recovered from the cheese- making process. Nonetheless, EOs incorporation into EF change their most relevant properties, such as the continuity of polymer matrix, weakening the film structure, reducing its transparency, while improving water barrier properties [3]. In this regard it is necessary to investigate the specific interactions between the polymer matrix and the EOs composition in order to determine the effectiveness of EOs as active ingredients. Application of whey proteins with EOs in EF has been investigated by several researchers showing the EOs' antimicrobial effect, the excellent oxygen barrier properties, transparency but the relatively low water vapor permeability of the films [10–12]. To favor film formations, the whey proteins should undergo thermal denaturation, above 70 ◦C. Further, the unfolded globular whey proteins, expose the buried SH groups and hydrophobic groups that can react forming inter- and intra-molecular bonding during film drying [13]. Besides thermal treatment, ultrasound and addition of transglutaminase have been tested for protein denaturation prior EF drying [13]. These alternative methods could also dictate the capacity of film structure to retain and gradually release the volatiles compounds, but also influence the mechanical properties and water vapor permeability of films.

High pressure processing (HPP) is an alternative to thermal treatment that can induce structural changes in macromolecules which are distinct from those of conventional thermal treatment [14]. However, to favor the protein film formation, a combination of high pressure with thermal treatment is required. Due to the different mechanism involved in protein denaturation, high pressure thermal processing (HPT) could result in the formation of a protein-based network with different properties compared to thermal treatment (TT).

In this study, combined high pressure at 600 MPa with thermal treatment (70 ◦C) was employed as original alternative to thermal treatment alone for whey protein aggregation, promoting intermolecular interactions between film forming substances, which are crucial for film forming step [15]. For obtaining a homogenous film forming emulsion, ultrasound treatment was used here [16].

The objective of this research was to obtain a homogenous, flexible, resistant film formulae made of whey proteins and functionalized with thyme EO (TEO) as antimicrobial agent. The films obtained by casting were further characterized to assess their potential for food packaging applications, in terms of mechanical, physico-chemical and antimicrobial properties. The capacity of HPT and TT films to retain and release the EOs trapped in the films structure was assessed over time in relation with their antimicrobial activity.

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

#### *2.1. Materials*

Whey protein concentrate, ProMilk 852FB1 was kindly offered by KUK-Romania (composition on dry-weight basis: 86% protein, 1% total fat, 11% lactose, 2.9% total ash, 5% moisture). Anhydrous glycerol (98% purity) purchased from Redox SRL (Bucharest, Romania). Tween 20, was purchased from Sigma- Aldrich (Bucharest, Romania). Thyme (*Thymus vulgaris*) EO, kindly provided by SC Hofigal SRL (Bucharest, Romania).

#### *2.2. Film Preparation*

The film was prepared by dispersing 7.6% (*w*/*w*) WPC powder, into distilled water under continuous magnetic stirring (180 rpm, 15 min) following a method previously optimized by Bleoanca et al. [17]. The pH was adjusted to 7.0 using 2 N NaOH [18]. In order to transform the protein solution in a flexible film either thermal crosslinking (80 ± 0.5 ◦C, for 35 min) or combined HPT denaturation (600 MPa, 70 ◦C, for 20 min) were applied.

#### 2.2.1. Thermal Treatment

The thermal inactivation was applied in a thermostatic water bath at 80 ± 0.5 ◦C for 35 min. Timing was started after the temperature measured inside the sample has reached 80 ◦C, as measured by type K thermocouple in one of the glass vials. Immediately after finishing the thermal treatment the samples were cooled in iced-water to stop the thermal effect.

#### 2.2.2. Combined Mild-Thermal High Pressure Treatment

Combined pressure- temperature treatments were conducted in a multivessel (4 vessels of 100 mL) high-pressure equipment (Resato, Roden, The Netherlands). As a pressure transmitting fluid, a mixture of water and propylene glycol (TR15, Resato) was used. The sample, approximately 30 mL, was first heated at 65 ◦C, and then filled without air into Teflon cylinders and placed into the HPP vessels to avoid temperature gradients. Compression started when the temperature was equal to target temperature, 70 ◦C, up to 600 MPa, and 20 min holding times. The compression rate was of approximately 10 MPa/s, until the preset pressure was reached, whereupon the valves of the individual vessels were closed and the central circuit was decompressed. An additional one-minute equilibration period was taken into account to ensure constant temperatures. Temperature inside the samples was monitored during the treatment with a thermocouple placed in the upper part of the Teflon cylinders. Decompression of the vessels was almost instantaneously (~5 s). After the pressure-temperature treatment, the samples were immediately transferred into iced water.

After forced cooling on ice, into the resulting film forming mixture obtained either by thermal or combined high pressure- thermal denaturation, anhydrous glycerol was added at a concentration of 8.0% (*w*/*w*) as plasticizer to reduce the brittleness of the WPC films, thus improving its mechanical properties. As surfactant for reducing the surface tension, tween 20 was used in a concentration of 0.9% (*w*/*w*). Then, thyme (*Thymus vulgaris*) EO, was added in the mixture as antimicrobial compound in a concentration of 2.5% (*w*/*w*). This plant EO was chosen due to its high content of carvacrol, thymol and p-cymene, all known to be efficient antimicrobials [19] and considering the results of previous tests performed by our research group [20].

Further the mix was homogenized by ultrasonication with equipment Sonoplus HD3100 Bandelin, Germany equipped with a sonication probe of 8 mm diameter, at 35% amplitudes, for 3 min. The sonication probe was immersed 1 cm below the liquid surface and the temperature of the film forming emulsion was kept at 23 ± 2 ◦C during sonication by placing the tube in an iced water bath [16].

The film forming emulsion was then poured onto silicone trays (diameter 5 cm). To control film thickness, the same amount (11 mL) of film forming mixture was poured. The spread solutions were allowed to dry at room temperature, approximately 22 ◦C, for 48 h at 50% RH [21,22], then easily peeled off.

Considering the hydrophilic nature of the protein film, therefore its susceptibility to absorb humidity from the environment, a standardization of the films was necessary to ensure that the mechanical properties of the film are not impaired. For this reason, prior to all investigations, the films were preconditioned by storing them in a controlled temperature- humidity environment, at 50 ± 3% RH and 25 ± 1 ◦C, for at least 72 h [23].All the experiments were performed in triplicate.

## *2.3. Film Characterization*

## 2.3.1. Film Thickness

A digital micrometer (Digimatic Micrometer, Mitutoyo, Japan) was used to measure film thickness to the nearest 0.0001 mm. The mean thickness was calculated from five measurements taken randomly at different locations on each film.

## 2.3.2. Moisture Content

The moisture content (MC) of the whey protein films was determined after oven drying at 105 ± 1 ◦C for 24 h until a constant weight was attained. After adequate conditioning, 3.4 cm diameter discs were cut from the edible film and weighed in order to be compared to the ones after drying. The moisture content values were determined as percentage of initial film weight loss during drying [24].

$$\text{MC} = \frac{w\_1 - w\_2}{w\_1 - w\_0} \times 100 \,\text{[\,\%]} \tag{1}$$

*w*<sup>0</sup> is the weight of empty and dry weighing glass bottle, (g); *w*<sup>1</sup> is the weight of weighing glass bottle with film, before drying, (g); *w*<sup>2</sup> is the weight of weighing glass bottle with film, after drying, (g).

#### 2.3.3. Water Activity

The water activity (*a*w) of preconditioned edible films was measured with a (Fast lab water activity meter; GBX, Loire, France), using discs of films (4 ± 0.1 cm diameter).

#### 2.3.4. Moisture Sorption Isotherms

Moisture sorption isotherms were determined by static gravimetric method [25]. Dried film samples were first conditioned for 5–10 days into a controlled humidity environment at a constant temperature until equilibrium has been reached. Samples discs of 49.58 ± 0.31 mm were placed into desiccators, each containing one saturated salt solution giving various RH at 25 ◦C: LiCl for an aw of 0.114, MgCl2 giving a 0.331 *a*w, KI giving an *a*<sup>w</sup> of 0.700, NaCl for an *a*<sup>w</sup> of 0.755, KCl giving an *a*<sup>w</sup> of 0.851 and KNO3 for an *a*<sup>w</sup> of 0.935. Film samples were equilibrated at each environment for 5–10 days at 25 ± 0.5 ◦C; following removal from desiccators they were immediately weighed, the *a*<sup>w</sup> was determined and moisture content was measured gravimetrically as described above. The Guggenheim-Anderson-de-Boer and Halsey models [26] as indicated by Tudose et al. [27] were applied by nonlinear regression analysis (SAS, 2009):

$$M = \frac{M\_0 \times \mathbb{C} \times K \times a\_w}{(1 - K \times a\_w) \times (1 - K \times a\_w + \mathbb{C} \times K \times a\_w)}\tag{2}$$

where *M* is the equilibrium moisture content (% dry basis); *M*<sup>0</sup> is the monolayer moisture content (% dry basis); *C*—Guggenheim constant; *K*—corrective constant; *a*w is the water activity (dimensionless);

The Halsey equation is:

$$a\_{\text{lv}} = \exp\left(-\frac{k}{M^n}\right) \tag{3}$$

where *k* and *n* are model constants.

#### 2.3.5. Water Vapor Permeability

Water vapor permeability (WVP) was estimated gravimetrically according to ASTM E96 [28], adapted for edible films. Film discs of 49.58 ± 0.31 mm diameter equilibrated at 25 ◦C, 50% RH for 48 h with saturate salt solution (Mg(NO3)2) were cut and mounted on glass cups filled with distilled water to 10 mm below the film underside. The glass cups had 46 mm diameter and 150 mm depth. The steady-state films water- vapor flow was measured at certain intervals for 48 h by digital-balance

nearest to 0.0001 g. Films permeability was calculated according to the method described by Zinoviadou et al. [11]. The weight loss was monitored and expressed by the slopes calculated using linear regressions equations where *R*<sup>2</sup> > 0.99. At least five replicates were tested for WVP estimation.

$$NVP = \frac{Slope \times x}{A \times \Delta p} \left(\text{g-mm} / \text{m}^2 \cdot \text{s-Pa}\right) \tag{4}$$

where *slope* is the weight loss of the cup per second, (g/s); *x* is the average film thickness, (mm); *A* is the area of exposed film, (m2); Δ*p* is the difference in vapor pressure across the test film (Pa).

## 2.3.6. Microstructural Analysis of The Film Forming Mixtures

A scanning electron microscope (Quanta 250, Thermo Fisher Scientific) (Waltham, MA 02451, USA) was used to determine the microstructure of thermal treated (TT) and combination of high pressure with temperature treatment (HPT) whey protein film samples with an accelerating voltage of 12.5 kV in a low vacuum environment. A magnification of 400×–1400× was used to scan each film sample.

#### *2.4. Antimicrobial Assay*

The antimicrobial effect of edible films was tested against three target microorganisms, *Bacillus subtilis*, *Geotrichum candidum* and *Torulopsis stellata,* all part of MIUG collection from Dunarea de Jos University of Galati- Romania. The antimicrobial effectiveness of the edible films was tested 10 days after the films were obtained, by vapor phase test [29]. This specific indirect contact assay for testing the antimicrobial activities was chosen to assess the protection provided by the thyme antimicrobial volatiles under no direct contact between the food product and the packaging. To perform vapor-phase diffusion tests, edible films of approx. 50 mm diameter discs were placed on the lids of Petri dishes, with previously spread 10<sup>6</sup> cfu/mL microbial inoculum. The inoculated agar plate was inverted with discs on the top of each lid containing antimicrobial film. Parafilm was used to tightly seal the edge of each Petri dish. Sealed and inverted Petri dishes were incubated at 27 ◦C for evaluation of anti-*Torulopsis* and anti- *Geotrichum* activity and at 37 ◦C for anti-*Bacillus* activity. Growth of each test microorganism was evaluated after two days of incubation. The inhibition radius (absence of growth) on each Petri dish was measured with a digital caliper and the inhibition area was calculated and expressed as mm2. The negative control, represented by whey protein EF without TEO, were also tested under the same conditions. The vapor phase inhibition test was performed in duplicate, in two separate experimental runs.

## *2.5. Solid-Phase Micro-Extraction (SPME)*

Before analysis the HPT and TT films were placed in desiccators of 6 L capacity with Mg(NO3)2 salt at 25 ◦C and 50% RH and stored for maximum 10 days. Each film had a 19.65 cm2 surface exposed to the environment and 3 discs were present in each desiccator for all the duration of the experiments. From each film discs with 34 mm diameter were cut, weight, introduced in sealed vials and maintained at 40 ◦C for 10 min for equilibration before concentration by SPME on a CAR/PDMS fiber. The extraction of the volatiles under isothermal conditions at 40 ◦C was made over 30 min followed by 5 min of desorption into the GC injection port.

## *2.6. Gas Chromatography-Mass Spectrometric (GC-MS) Analysis*

The volatiles fingerprints of the edible film samples were analyzed using a Trace GC-MS Ultra equipment with ionic trap- ITQ 900 from Thermo Scientific (USA). The GC column was a TG-WAX capillary column (60 m × 0.25 mm, i.d. 0.25 μm). The carrier gas was helium (99.996% purity, Messer S.A., Bucharest, Romania) that ran at a flow rate of 1 mL/min. The temperature ramp selected for the analysis was: 40 ◦C isothermal treatment for 4 min followed by an increase to 50 ◦C at 5 ◦C/min and to 100 ◦C with 7 ◦C/min, to 150 ◦C at 10 ◦C/min and finally to 230 ◦C at 12 ◦C/min, when temperature was kept constant for 2 min. The temperature of the transfer line in MS was set to 270 ◦C. Mass spectra were obtained from the full scan of the positive ions resulted with a scanning in the 35 to 450 *m*/*z* range and operated with an electron impact (EI)-mode of 200 eV. The compounds were identified in comparison with the mass spectra from Wiley and Nist 08 library database available with Xcalibur 2.1 software. The retention indices (RI) of each compound were calculated by using n-alkane series from C8-C40 (Sigma Aldrich Chemie GmbH, Steinheim, Germany) under the same conditions. Each analysis was performed in triplicate, in the first and the 10th day of storage.

The volatile organic compounds (VOCs) were estimated semi-quantitatively using n-octanol as internal standard (IS) and Equation (5) [20,30]:

$$\text{VOC}\_{\text{conv}} = \text{IS}\_{\text{conv}} \times \left( \text{VOC}\_{\text{peak area}} \Big| \text{IS}\_{\text{peak area}} \right) \tag{5}$$

where *VOCpeak area* is the area of the integrated individual peak, *ISpeak* area is the area of 2-octanol in the spiked samples and *ISconc* is the concentration of internal standard (2-octanol).

## *2.7. Statistical Analysis*

Data were expressed as mean ± standard deviation (SD). The statistical analysis was carried out using analysis of variance (ANOVA) and Tuckey' s post-hoc test was applied to evaluate significant differences among groups (*p* < 0.05).

The quality of the sorption isotherms models' fit applied was evaluated by the regression coefficient (*R*<sup>2</sup> adj) and the mean relative percentage deviation (%*E*):

$$E = \frac{100}{N} \sum\_{i=1}^{N} \frac{\left| m\_i - m\_{pi} \right|}{m\_i} \tag{6}$$

where *m*<sup>i</sup> and *m*pi are the experimental and predicted values, respectively, and *N* is the population of the experimental data.

$$R\_{adj}^2 = 1 - \left(\frac{n\_t - 1}{n\_t - n\_p}\right) \cdot \frac{SSE}{SSTO} \tag{7}$$

Principal component analysis (PCA) was performed using the Unscrambler software (Version 9.7; CAMO, Norway). PCA was performed with the peak list resulting from SPME GC/MS analysis for all the volatile compounds. The data matrix was formed by *n* = 6 cases and 25 variables defined as the VOCs peak areas obtained for each individual component. Data were transformed by unit vector normalization prior to statistical analysis.

#### **3. Results and Discussion**

## *3.1. Film Appearance*

Appearance of the two sides of the WPC film was similar for HPT and TT films. The film side facing the casting plate was shiny, while the other was dull; this is likely an indication of some phase separation occurring in the mixture during drying. HPT and TT types of film were easily separated from the casting plates. During the TT the three dimensional structure of proteins was unfolded and the internal sulfhydrilic groups were exposed, later forming intermolecular disulfide bonds while hydrophobic groups interactions also might have occurred during film drying [18,31]. Combination of HPP and TT resulted in both denaturation via above referred mechanism and by forcing the water molecules inside the protein matrix, that exposed the hydrophobic core, followed by protein unfolding [32,33].

Films manufactured from WPC with 7.6%(*w*/*w*) protein showed a thickness of 0.133–0.193 mm, close to those reported by other researchers [34–36]. Neither one of the HPT and TT WPC-based films

functionalized with thyme essential oils (TEO) did not exhibit any statistically significant differences either (*p* < 0.05) (Table 1).


**Table 1.** Thickness and WVP of TT and HPT films #.
