Inhibition of Satellite RNA Associated Cucumber Mosaic Virus Infection by Essential Oil of Micromeria croatica (Pers.) Schott

The present results dealing with the antiphytoviral activity of essential oil indicate that these plant metabolites can trigger a response to viral infection. The essential oil from Micromeria croatica and the main oil components β-caryophyllene and caryophyllene oxide were tested for antiphytoviral activity on plants infected with satellite RNA associated cucumber mosaic virus. Simultaneous inoculation of virus with essential oil or with the dominant components of oil, and the treatment of plants prior to virus inoculation, resulted in a reduction of virus infection in the local and systemic host plants. Treatment with essential oil changed the level of alternative oxidase gene expression in infected Arabidopsis plants indicating a connection between the essential oil treatment, aox gene expression and the development of viral infection.


Introduction
Essential oils (EOs) have become a subject of interest due to their numerous biological effects, attributed primarily to prooxidant effects at the cellular level [1,2]. Significant quantitative and qualitative variations of essential oil components have been observed within and between species, but also within populations and even within individuals, with strong implications for plant-herbivore and plant-pathogen interactions. The defensive action of essential oils can be direct, through toxicity, while some volatile organic compounds released upon chewing act as specific cues to attract parasitoids [3][4][5]. While many individual compounds are toxic to herbivores, others act as antifungal, antibacterial or allelopathic agents, or for priming of systemic defences in both the host and neighbouring plants [6][7][8][9]. A recent area of research is the antiphytoviral activity of EOs. Although limited, the results to date indicate that these plant metabolites can trigger a response to viral infection [10][11][12][13][14][15][16]. Qualitative differences in EO composition of different plant species do not exhibit a uniform response that would explain the mechanism of antiphytoviral activity. Previous studies on the EOs of different plant species have led to the conclusion that sesquiterpene-rich oils can reduce viral infection in local host plants [10,12,13,16]. Synthesis of certain Arabidopsis thaliana sesquiterpenes, including β-caryophyllene, are induced by the phytohormones gibberellin (GA) and jasmonate (JA), indicating that the production of these sesquiterpenes is integrated within the GA and JA signalling pathways [17]. Sesquiterpenes are the main oil components of the EO of Micromeria croatica (Pers.) Schott described by Kremer et al. (2012) [18], and this study aimed to determine its antiphytoviral potential for the prevention of diseases caused by plant viruses. As endemic Illyric-Balcanic species with small isolated populations, M. croatica (Lamiaceae) has been explored insufficiently to date. The EO composition of certain species can display annual variations, and variations in composition between populations. Genetic factors, evolution, geographic variation, environmental conditions (i.e., harvest date, planting time), physiological variations (i.e., organ and leaf position), and developmental stage are known to affect the biosynthesis of EOs [19][20][21]. Thus, the first objective of this study was to analyse the chemical composition of the EO of M. croatica collected at three localities on Mt Velebit (Croatia), and to establish annual variations in oil composition, comparing the present results with those of a previous study [18]. The second objective was to analyse the uniformity of oil composition between M. croatica populations and to statistically analyse the results to clarify variations in the EO profile. The effect of these variations in the EO profile on the antiphytoviral activity of oil of M. croatica was examined to determine whether this activity is independent of oil composition. Cucumber mosaic virus (CMV) was selected for determining the antiphytoviral activity of EO due to its wide host range of some one thousand plant species, including many of the most important crops. Some strains of CMV are associated with satellite RNA (satRNA), which depends on their helper virus for replication and movement [22] and can have pronounced effects on host plant symptoms [23].
With the exception of our study dealing with the antiphytoviral activity of the EO of Micromeria graeca [16], no other literature reports on the antiphytoviral activity of EOs within the genus Micromeria were available. In this sense, the aim of this study was to provide answers to the following questions: (v) does the prolongation of treatment with EO/DC prior to inoculation reflect on the antiviral activity in local host plants; (vi) do the above treatments affect virus concentrations in systemic host plants; (vii) is there a correlation between the antiviral activity of the EO and the antiviral activity of the DC of oil; and finally, (viii) is the antiviral activity constant due to differences in oil composition?
The next step was to establish the antiviral activity of EOs by testing the expression of one of the genes important in plant defence against pathogens. Alternative oxidase (AOX) is the terminal oxidase of the cyanide-resistant alternative respiratory pathway in plants and has been implicated in resistance to viruses [24]. While AOX abundance and alternative pathway activity are low in unstressed plants, both increase when plants are subjected to suboptimal conditions, such as chilling [25] or pathogen attack [26]. The AOX limits the formation of mitochondrial reactive oxygen species (ROS) in all plant species [27], and in stress conditions, AOX reduces ROS formation and has an active role in the adaptation of Arabidopsis to cold conditions [28]. Different species under similar stress conditions have a different response to the synthesis of AOX, though several Arabidopsis studies have described Aox1a as the most stress responsive Aox gene [29]. Based on the above, we proposed that at least one part of defence response mediated by EO involves a change of expression of the Aox gene. Since CMV is able to infect Arabidopsis thaliana, we took advantage of this model system to investigate the correlation between essential oil treatment, Aox gene expression and the development of viral infection.
New EO/virus/host combinations and methodological approaches in this field enabled us to correlate EO composition/method of application with antiviral activity-gene expression. Although no single answer is expected to explain the antiphytoviral activity of EOs, answers to the above questions will improve the existing knowledge about the antiphytoviral activity of EOs, and our understanding of the role of EOs in plant-virus interactions.

Essential Oil Composition
The EO of the endemic Micromeria croatica from Mt. Velebit, Croatia (localities Bojinac (Boj), Bačić kuk (Bk), and Stupačinovo (St)) was subjected to detailed GC and GC-MS analysis to determine chemical composition and the similarity and variability in oil composition depending on locality. The chemical composition of EOs, the percentage of identified components and their retention indices are given in Table 1, where all the components are arranged in order of their elution from the VF-5ms capillary column [30]. The overall composition of EO of M. croatica is characterized by a high percentage of sesquiterpenes and monoterpenes (Table 1). In the oil from Boj, sesquiterpene β-caryophyllene (25.2%) is the dominant component, followed by oxygenated sesquiterpene caryophyllene oxide (10.1%). In the oil from Bk, the cumulative amount of monoterpenes is substantially higher than sesquiterpenes, although the oxygenated sesquiterpene caryophyllene oxide (21.1%) is a major component of this oil. In the oil from St, caryophyllene oxide (20.2%) is the major compound, followed by β-caryophyllene (10.2%) ( Table 1). Minor qualitative and quantitative differences in the annual oil composition at the localities Boj and Bk were observed in comparison to previous results presented by Kremer et al. (2012), though the major components in oil composition remained unchanged [18]. Namely, β-caryophyllene and caryophyllene oxide were also identified as the dominant components of oil at the described localities [18]. Due to their high content in the composition of oil, β-caryophyllene and caryophyllene oxide were also subjected to n antiphytoviral assay to investigate their role in antiviral activity of the oil. The composition of oil at the St locality (Table 1) presents new data and consequently was not comparable to previous results at the annual level. Principal component analysis (PCA) was employed to provide an overview of the capacity to distinguish essential oil components based on GC-MS data ( Figure 1). The common components of EO composition at the localities Boj, Bk and St clustered in the middle of the diagram. The EO components grouped at the periphery of the diagram show qualitative discrepancies in oil composition among the localities (Figure 1). Based on Figure 1, the EO composition in the studied populations of M. croatica can be considered to be quite uniform. A somewhat more pronounced degree of specificity was observed in the composition of oil from St ( Figure 1). Cluster analysis classifies the number of samples studied into groups according to the chemical composition of essential oil by 'magnifying' their similarities. Multivariate analysis of the samples differentiated two well-defined clusters of M. croatica (Figure 2), in which one cluster included the localities Boj and Bk, and the locality St was isolated in its own cluster, indicating the specificity of its EO composition. Retention indices (RI) were determined relative to a series of n-alkanes (C 8 -C 40 ) on capillary columns VF5-ms (RI a ) and CP Wax 52 (RI b ); c average value from two columns, where possible; * identification method: RI, comparison of RIs with those listed in a homemade library, reported in the literature [30], and/or authentic samples; comparison of mass spectra with those in mass spectral libraries NIST02 and Wiley 9; ** co-injection with reference compounds; SD, standard deviation.

Antiphytoviral Activity
In the field of plant virology, there is a limited amount of information on the effects of EOs on viruses or viral infections in plant systems [10][11][12][13][14][15][16]. The first objective of this part of the study was to test the hypothesis on the antiviral potential of the EO of M. croatica. Thus, the EO/DC of oil was

Antiphytoviral Activity
In the field of plant virology, there is a limited amount of information on the effects of EOs on viruses or viral infections in plant systems [10][11][12][13][14][15][16]. The first objective of this part of the study was to

Antiphytoviral Activity
In the field of plant virology, there is a limited amount of information on the effects of EOs on viruses or viral infections in plant systems [10][11][12][13][14][15][16]. The first objective of this part of the study was to

Antiphytoviral Activity
In the field of plant virology, there is a limited amount of information on the effects of EOs on viruses or viral infections in plant systems [10][11][12][13][14][15][16]. The first objective of this part of the study was to

Antiphytoviral Activity
In the field of plant virology, there is a limited amount of information on the effects of EOs on viruses or viral infections in plant systems [10][11][12][13][14][15][16]. The first objective of this part of the study was to

Antiphytoviral Activity
In the field of plant virology, there is a limited amount of information on the effects of EOs on viruses or viral infections in plant systems [10][11][12][13][14][15][16]. The first objective of this part of the study was to

Antiphytoviral Activity
In the field of plant virology, there is a limited amount of information on the effects of EOs on viruses or viral infections in plant systems [10][11][12][13][14][15][16]. The first objective of this part of the study was to test the hypothesis on the antiviral potential of the EO of M. croatica. Thus, the EO/DC of oil was rubbed onto leaves of local and systemic host plants simultaneously with virus inocula, or the EO/DC of oil were introduced into host plants prior to virus inoculation. The aim was to gain insight into the antiphytoviral activity of the oil by providing data on EO-plant-virus interactions using a variety of inoculation methods of local and systemic host plants. In simultaneous inoculation on the leaves of local host plants, the virus was incubated with EO/DC of oil for 0, 1, 2 and 3 h prior to inoculation. When inoculation was carried out after a short incubation period (0 h), EO Boj and β-caryophyllene were most effective in reducing the number of local lesions compared to the control (33.9% and 38.3%, respectively) ( Figure 3). EO St and caryophyllene oxide were most effective during the experiment with two-hour preincubation with the virus in comparison to the control (34.5% and 33.4%, respectively), while EO Bk was the most effective after one-hour preincubation (33.6%) (Figure 3). Three-hour preincubation was statistically ineffective for reduction of local lesions in all experimental groups, with the exception of β-caryophyllene, which inhibited lesions by 25.3% (Figure 3). Based on these results, Pearson's correlation coefficient established a positive correlation of the antiviral activity of EO Boj with β-caryophyllene. Antiviral activity of EO Bk and EO St positively correlated with the antiviral activity of caryophyllene oxide.
of oil were introduced into host plants prior to virus inoculation. The aim was to gain insight into the antiphytoviral activity of the oil by providing data on EO-plant-virus interactions using a variety of inoculation methods of local and systemic host plants. In simultaneous inoculation on the leaves of local host plants, the virus was incubated with EO/DC of oil for 0, 1, 2 and 3 h prior to inoculation. When inoculation was carried out after a short incubation period (0 h), EO Boj and β-caryophyllene were most effective in reducing the number of local lesions compared to the control (33.9% and 38.3%, respectively) ( Figure 3). EO St and caryophyllene oxide were most effective during the experiment with two-hour preincubation with the virus in comparison to the control (34.5% and 33.4%, respectively), while EO Bk was the most effective after one-hour preincubation (33.6%) (Figure 3). Three-hour preincubation was statistically ineffective for reduction of local lesions in all experimental groups, with the exception of β-caryophyllene, which inhibited lesions by 25.3% (Figure 3). Based on these results, Pearson's correlation coefficient established a positive correlation of the antiviral activity of EO Boj with β-caryophyllene. Antiviral activity of EO Bk and EO St positively correlated with the antiviral activity of caryophyllene oxide. The next aim of this study was to determine whether different methods of EO/DC application affect antiviral activity in local host plants and whether EO/DC has the ability to activate the defensive response of local host plants against viruses. Thus, local host plants were treated with EO/DC for one, two and three consecutive days prior to virus inoculation. Although all pretreatments significantly reduced the number of local lesions, the strongest antiviral effect was manifested after three days of treatment in all experimental groups. Reduction of local lesion numbers ranged between 66.8% and 71.4% (Figure 4), confirming the hypothesis of the antiviral activity of EO of M. croatica. Pearson's correlation coefficient showed a positive correlation of the antiviral effect of EOs with both dominant components of oil. This part of the study led to the conclusion that simultaneous inoculation of virus inocula and EO/DC of oil is less effective for the reduction of local infection than treatment prior to inoculation ( Figure 3, Figure 4). Furthermore, preincubation of viral inoculum with EO/DC that lasted more than two hours did not enhance the antiphytoviral effect ( Figure 3). Due to the volatility of EOs, the concentration of active compounds was likely reduced. On the contrary, longer pretreatment of host plants with EO/DC of oil prior to inoculation enhanced the antiphytoviral effect ( Figure 4). Based on the above, it can be concluded that induction of the defence response of local host plants is the mode of antiviral activity of EO of M. croatica, as opposed to direct "in vitro" inactivation of the viral particles. The next aim of this study was to determine whether different methods of EO/DC application affect antiviral activity in local host plants and whether EO/DC has the ability to activate the defensive response of local host plants against viruses. Thus, local host plants were treated with EO/DC for one, two and three consecutive days prior to virus inoculation. Although all pretreatments significantly reduced the number of local lesions, the strongest antiviral effect was manifested after three days of treatment in all experimental groups. Reduction of local lesion numbers ranged between 66.8% and 71.4% (Figure 4), confirming the hypothesis of the antiviral activity of EO of M. croatica. Pearson's correlation coefficient showed a positive correlation of the antiviral effect of EOs with both dominant components of oil. This part of the study led to the conclusion that simultaneous inoculation of virus inocula and EO/DC of oil is less effective for the reduction of local infection than treatment prior to inoculation (Figure 3, Figure 4). Furthermore, preincubation of viral inoculum with EO/DC that lasted more than two hours did not enhance the antiphytoviral effect ( Figure 3). Due to the volatility of EOs, the concentration of active compounds was likely reduced. On the contrary, longer pretreatment of host plants with EO/DC of oil prior to inoculation enhanced the antiphytoviral effect ( Figure 4). Based on the above, it can be concluded that induction of the defence response of local host plants is the mode of antiviral activity of EO of M. croatica, as opposed to direct "in vitro" inactivation of the viral particles.  (Figure 1, Figure 2), it is possible that this slightly different oil composition affects its antiviral activity. It is possible that aside from the main components, other components in this oil or their synergistic effects contributed to the antiviral activity of oil from St. A similar conclusion was reported in previous studies [10,16] and is supported here by new evidence involving multiple treatments and different hosts (Figures 3-5). In addition, the effect of these treatments on virus concentrations in systemically infected plants supports the conclusion about the defence response as the mode of antiviral activity of M. croatica EO. The results of the inhibition of Since both the PCA and cluster analysis showed that composition of the oil from locality St was specific (Figure 1, Figure 2), it is possible that this slightly different oil composition affects its antiviral activity. It is possible that aside from the main components, other components in this oil or their synergistic effects contributed to the antiviral activity of oil from St. A similar conclusion was reported in previous studies [10,16] and is supported here by new evidence involving multiple treatments and different hosts (Figures 3-5). In addition, the effect of these treatments on virus concentrations in systemically infected plants supports the conclusion about the defence response as the mode of antiviral activity of M. croatica EO. The results of the inhibition of CMVsat (satellite associated Cucumber mosaic virus) infection in local and systemic host plants by EO/DC of oil (Figures 3-5) leads to the conclusion that the antiphytoviral effect is more affected by the method of EO application and the EO-virus-plant host combinations studied than by the described variations in the oil composition.  (Figures 3-5) leads to the conclusion that the antiphytoviral effect is more affected by the method of EO application and the EO-virus-plant host combinations studied than by the described variations in the oil composition.

Aox Gene Expression
In order to better elucidate the plant defence pathways against viruses at the molecular level, Arabidopsis thaliana was used as a plant that is a well-known model system susceptible to CMV infection. The aim was to study the effects of EO/DC of M. croatica oil on the expression of one of the important genes in plant defence against pathogens. Changes in the expression of the Aox gene in different biotic and/or abiotic stress conditions, such as infection by a pathogen, suggests that AOX belongs to the proteins involved in defence against biotic and abiotic stress [31,32]. The enzyme AOX reduces the formation of new reactive oxygen species (ROS) in the mitochondria and, under stress conditions, can prevent an increased accumulation of ROS molecules [28]. Our experimental results showed that virus infected plants increased Aox expression compared to healthy, untreated plants (Figure 6a). Literature reports show that plants exposed to stress conditions, biotic or abiotic, synthesise ROS molecules, and the virus defence response directs the flow of electrons to the enzyme AOX that acts as negative regulator of ROS synthesis [33]. Another study found that increased expression of Aox in the Arabidopsis plant is accompanied by an enhanced spread of the virus CMVsat throughout the plant and decreased expression with a slowing of the systemic spread of the virus [34]. We found that plants treated with EO Boj/EO St/β-caryophyllene prior to virus inoculation decreased Aox gene expression compared to untreated infected plants (Figure 6b). In accordance with the literature data, it can be assumed that the EO treated plant, rather than the virus, silences the expression of the Aox gene, thereby protecting the plant by slowing the spread of CMVsat. Lower expression of the Aox gene and increased concentration of mitochondrial ROS molecules protects the plant through the appearance of induced resistance to CMV and a deceleration of the spread of the virus [35]. On the contrary, plants treated with the both essential oil from the locality Bk and caryophyllene oxide showed increased Aox gene expression compared to untreated infected plants (Figure 6b). Another study found that AOX induction also occurs in response to other pathogen infections [26,36], and hence may be a general consequence of pathogen infection rather than a specific resistance response to the virus. These results lead to the conclusion that

Aox Gene Expression
In order to better elucidate the plant defence pathways against viruses at the molecular level, Arabidopsis thaliana was used as a plant that is a well-known model system susceptible to CMV infection. The aim was to study the effects of EO/DC of M. croatica oil on the expression of one of the important genes in plant defence against pathogens. Changes in the expression of the Aox gene in different biotic and/or abiotic stress conditions, such as infection by a pathogen, suggests that AOX belongs to the proteins involved in defence against biotic and abiotic stress [31,32]. The enzyme AOX reduces the formation of new reactive oxygen species (ROS) in the mitochondria and, under stress conditions, can prevent an increased accumulation of ROS molecules [28]. Our experimental results showed that virus infected plants increased Aox expression compared to healthy, untreated plants (Figure 6a). Literature reports show that plants exposed to stress conditions, biotic or abiotic, synthesise ROS molecules, and the virus defence response directs the flow of electrons to the enzyme AOX that acts as negative regulator of ROS synthesis [33]. Another study found that increased expression of Aox in the Arabidopsis plant is accompanied by an enhanced spread of the virus CMVsat throughout the plant and decreased expression with a slowing of the systemic spread of the virus [34]. We found that plants treated with EO Boj/EO St/β-caryophyllene prior to virus inoculation decreased Aox gene expression compared to untreated infected plants (Figure 6b). In accordance with the literature data, it can be assumed that the EO treated plant, rather than the virus, silences the expression of the Aox gene, thereby protecting the plant by slowing the spread of CMVsat. Lower expression of the Aox gene and increased concentration of mitochondrial ROS molecules protects the plant through the appearance of induced resistance to CMV and a deceleration of the spread of the virus [35]. On the contrary, plants treated with the both essential oil from the locality Bk and caryophyllene oxide showed increased Aox gene expression compared to untreated infected plants (Figure 6b). Another study found that AOX induction also occurs in response to other pathogen infections [26,36], and hence may be a general consequence of pathogen infection rather than a specific resistance response to the virus. These results lead to the conclusion that one part of defence response mediated by EO involves a change of expression of the Aox gene. This is consistent with Murphy et al. (2004), who suggested that the enzyme AOX could be a regulator of the ROS molecule mediated signalling in plant defence response to viruses [24]. The involvement of EO constituents in plant defence was Molecules 2019, 24, 1342 9 of 15 also described by Hong et al. (2012), who examined the emission of volatile terpenes by A. thaliana flowers, which may function in plant-insect interactions [17]. The same authors state that Arabidopsis MYC2, a basic helix-loop-helix transcription factor, directly binds to promoters of the sesquiterpene synthase genes, TPS21 and TPS11, and activates their expression. Expression of TPS21 and TPS11 can be induced by the phytohormones GA and JA, and both inductions require MYC2. The induction of TPS21 and TPS11 results in increased emission of sesquiterpene, especially (E)-β-caryophyllene. These results indicate that the production of these sesquiterpenes is integrated with the GA and JA signalling pathways involved in plant defence [17]. Accordingly, it can be concluded that exogenously applied EOs and sesquiterpene constituents of the oil of M. croatica are involved in plant defence against CMVsat infection.
Molecules 2019, 24, x 9 of 15 one part of defence response mediated by EO involves a change of expression of the Aox gene. This is consistent with Murphy et al. (2004), who suggested that the enzyme AOX could be a regulator of the ROS molecule mediated signalling in plant defence response to viruses [24]. The involvement of EO constituents in plant defence was also described by Hong et al. (2012), who examined the emission of volatile terpenes by A. thaliana flowers, which may function in plant-insect interactions [17]. The same authors state that Arabidopsis MYC2, a basic helix-loop-helix transcription factor, directly binds to promoters of the sesquiterpene synthase genes, TPS21 and TPS11, and activates their expression. Expression of TPS21 and TPS11 can be induced by the phytohormones GA and JA, and both inductions require MYC2. The induction of TPS21 and TPS11 results in increased emission of sesquiterpene, especially (E)-β-caryophyllene. These results indicate that the production of these sesquiterpenes is integrated with the GA and JA signalling pathways involved in plant defence [17]. Accordingly, it can be concluded that exogenously applied EOs and sesquiterpene constituents of the oil of M. croatica are involved in plant defence against CMVsat infection. The results presented here provide new insight into the antiphytoviral activity of essential oils. Understanding the defence response mechanisms at the molecular level will improve our understanding of the plant defence responses and enhance our knowledge about the antiphytovial potential of essential oils and their use in the prevention of viral plant diseases. The above presents several directions for future research that should aim to clarify the mechanism of antiviral activity of essential oils.

Essential Oil Isolation
The aboveground parts of several dozen randomly selected plants were harvested from plants on a dry day and mixed to obtain a randomly selected sample. Samples were air-dried for ten days in a well-ventilated room at 60% relative air humidity and room temperature (22 • C), single-layered and protected from direct sunlight. Dried aerial parts (100 g) from each locality were subjected to hydro-distillation for 3 h in a Clevenger type apparatus. Essential oils were dried over anhydrous sodium sulphate and stored at −20 • C for further experiments. The essential oils (EOs) were named according to abbreviation of the locality of collection (EO Boj, EO Bk, EO St).

Virus and Plant Hosts
Cucumber mosaic virus with associated satellite RNA (CMVsat), previously described by Škorić et al. (1996) [37], was propagated in a systemic host, Nicotiana megalosiphon Van Heurck & Müll. Arg. A voucher specimen of this plant virus is deposited at the Faculty of Science, University of Zagreb. Systemically infected leaves were homogenised in phosphate buffer (0.06 mol L −1 , pH 7.0), briefly centrifuged at 5.000× g for 5 min at 4 • C and the supernatant was used for inoculation. Prepared inoculum was diluted with the same buffer to give a suitable number    St. Louis, MO, USA) were added separately to virus inocula at a final concentration adjusted to 500 ppm. Prepared inocula were incubated at room temperature for a short time (0 h), 1, 2 or 3 h. Inoculation on the half-leaves of C. quinoa using a sterile glass spatula (0.1 mL amount per leaf half) followed. The corresponding opposite (control) half was rubbed with the control virus inocula to which distilled water was added instead of essential oil or oil compounds. Inoculated leaves were washed thoroughly with distilled water 1 min after inoculation. Ten to fifteen experimental plants (four leaves per plant) were inoculated in each experiment.

Quantitative Reverse Transcription-Polymerase Chain Reaction
Total RNA was purified from the collected plant material of Arabidopsis thaliana using TRI Reagent Solution (Ambion, Austin, TX, USA) according to the manufacturer's instructions. For Dnase treatment, 10 µg purified total RNA from each sample was used and processed with 2U TURBO Dnase (Ambion, Austin, TX, USA) according to manufacturer's instructions. RNA concentration was determined using NanoDrop 2000c (Thermo Fisher Scientific, Waltham, MA, USA). 10 µg total RNA treated with DNase was used for cDNA synthesis and was mixed with oligo d(T)16 and random hexamer primers, both in the final concentration of 0.1 µmol L −1 . To that mixture, dNTPs (2.5 mmol l −1 ) were added, incubated for 10 min at room temperature and subsequently cooled on ice. For reverse transcription, a master mix (12 µL, Applied Biosystems, Foster City, CA, USA) containing 5 µL of 25 mmol l −1 MgCl 2 , 2.5 µL 10× PCR buffer II, Rnase inhibitor in a final concentration of 1 U µL −1 and MuLV reverse transcriptase in a final concentration of 2.5 U µL −1 was prepared and mixed with total RNA, primers and dNTPs. Each sample was incubated for 45 min at 42 • C and then heated to 99 • C for 5 min.
The cDNA was then used for quantitative Real-Time PCR (qRT-PCR). Expression levels of alternative oxidase 1A-(Aox1a) (Assay ID: At02278012_g1) was measured using TaqMan Gene Expression Assays (Applied Biosystems, Foster City, CA, USA) in an optical 96-well plate using a 7300 Real-Time PCR system (Applied Biosystems, Foster City, CA, USA). The procedure was performed according to the manufacturer's instructions. The average CT values calculated from triplicate samples were used to determine the fold expression relative to the controls. Data were normalized to the expression level of F-box (Assay ID: At02200102_s1) and EF 1α (Assay ID: At02337969_g1), constitutively expressed reference genes, and relative expression of investigated genes was calculated according to Pfaffl (2004) [38].

Statistical Analysis
All analyses were carried out in triplicate and the data were presented as means ± standard deviations. Statistical significance of the differences between the investigated groups was evaluated using the Student t-test (GraphPad InStat software, San Diego, CA, USA). p values < 0.05 were considered statistically significant. Virus concentration is presented as the mean value of two repeated experiments.
Principal component analysis (PCA) and the Pearson correlation coefficient were carried out using the statistical software Statistica 8 (StatSoft, Tulsa, OK, USA). The cluster analysis was performed with the software PRIMER 5.0 PREMIER (Biosoft International, Corina Way, Palo Alto, CA, USA).

Conclusions
β-caryophyllene and caryophyllene oxide were the main oil components in investigated samples of the essential oil of Micromeria croatica from the localities Bojinac, Bačić kuk and Stupačinovo (Mt. Velebit, Croatia). An antiphytoviral assay showed that simultaneous inoculation of CMVsat with essential oil or the dominant components of the oil, or treatment of plants with essential oil or the dominant components of the oil prior to virus inoculation resulted in a reduction of the number of lesions on the leaves of local host plants, and a reduction of the virus concentration in systemically infected plants. Antiviral activity of the essential oil was correlated with the antiviral activity of the main oil component, demonstrating the role of β-caryophyllene and caryophyllene oxide in the activity of the essential oil in plant defence against viruses. At least one part of the defence response mediated by essential oil involves a change in the level of expression of the alternative oxidase (Aox) gene.

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