Potato plants, Solanum tuberosum cv. Désirée, were grown in a glasshouse at 25 ± 2°C under a 16L: 8D h photoperiod. In all experiments, potato plants with 30–45 cm height (nearly 3-week-old) were used. For the insect model, we used the potato aphid (Macrosiphum euphorbiae), which was collected from the field at Rothamsted Research (51.8093° N, 0.3548° W). Afterwards, colonies were maintained from parthenogenetic individuals on the above-mentioned potato variety in a controlled environment room (20 ± 1°C, 60 ± 10% RH, 16L: 8D h photoperiod), which ensured continuous asexual reproduction. Growth assay and volatiles entrainment experiments were carried out using apterae (wingless) adults.

Plants were treated with an aqueous emulsion of cis-jasmone (CJ; 90%; Avocado, Lancaster, UK) as described in Sobhy et al. (2017) using a hydraulic nozzle (Brown 015-F110) mounted on a variable speed spray track at 1 ms⁻¹. Plants were randomly allocated to one of the following treatments: (a) untreated plants (Control); (b) surfactant formulation (SUR): non-ionic surfactant Ethylan BV (100 μl) in deionized water (100 ml); and (c) CJ formulation (CJ): CJ (25 μl) and Ethylan BV (100 μl) in deionized water (100 ml). Ethylan BV is often used in this formulation to emulsify CJ in water and to lower the interfacial tension between the spray and the plant's epidermis to increase CJ uptake (Hazen, 2000). Given its ability to penetrate minute cavities such as stomata (Wood et al., 1997), it is highly possible that this surfactant can affect stomatal conductance and therefore may cause a change in the volatile emission. To rule out surfactant effects, a SUR treatment was also included in our study as a positive control. Prior to experiments, plants were kept for 24 h in separate glasshouse compartments to minimize unwanted induction of plant defense or “priming” effects by plant–plant volatile interactions between treatments. Plants were thereafter moved, with particular care to avoid any damage, from the glasshouse to laboratory where the experiments took place.

The performance of M. euphorbiae on Control, SUR and CJ potato plants was assessed. Previous studies have shown that most of the inducible changes occur after 48 h following CJ treatment (Dewhirst et al., 2012; Sobhy et al., 2017). Therefore, experiments were carried out 48 h after the plant treatment. Six alate M. euphorbiae were placed in one clip cage (5 cm diameter, N = 10), which were attached to the lower surface of potato leaves and left overnight for nymphs viviparity as shown in Figure 1. Subsequently, adults were removed from the clip cages the following morning and neonate nymphs were then collected and weighed together in batches of five in Eppendorf tubes (1.5 mL) on a microbalance (Cahn C33; Scientific and Medical Products Ltd, Manchester, UK), and the average birth weight was calculated. Nymphs were transferred back to fresh potato plants of the same treatment on which they started and then confined to clip cages and their survival and reproduction were monitored over 7 days (Figure 1). The plants were kept in a controlled environment room (25°C, 40% RH, 16L: 8D h photoperiod) with supplementary overhead lighting. After one week, survived aphid individuals were reweighed, and the mean relative growth rate (MRGR) of aphid was calculated (Bruce et al., 2003; Dewhirst et al., 2012), which indicates the growth per unit weight per unit time.

To further study the intrinsic rate of population increase (r_m), these obtained nymphs were then allowed to grow on fresh potato plants of the same treatment, which had been treated 48 h earlier. The time taken to reach adulthood and produce their first nymph was recorded, after which the number of neonate nymphs produced per day was noted. To prevent overcrowding, the nymphs were removed soon after recording. The r_m value was calculated using the time taken from birth to produce the first nymph (D) and the number of nymphs produced over a period equivalent to time D (FD) starting at the production of the first nymph. We used in the calculations a constant obtained from the mean pre-reproductive times for numerous aphid species (Wyatt and White, 1977), using the following equation:

Data were recorded in the form of number of reproduced nymphs, number of survived nymphs, weights of aphids at birth and after 7 days, the time taken from birth to produce the first nymph and the number of nymphs produced.

Dynamic headspace collection was carried out following standard procedures as described in Sobhy et al. (2017). For VOC collection, as shown in Figure 1, a compound leaf with 5–7 leaflets from a potato plant was enclosed in a glass vessel (22 cm high × 10 cm internal diameter). It should be noted that we used here identical compound leaves of potato plants to those used earlier in the aphid performance assay in terms of both height and position on potato stem. The bottom of the used glass vessel, which has two ports at the top (one for inlet of air and the other for outlet), was open. To seal such vessels, the bottom was “closed” without pressure around the plant stem using two semicircular aluminum plates with a hole in the center to accommodate the stem, which was wrapped with PTFE tape (Gibbs and Dandy, Luton, UK) at the connection point. To ensure no damage has occurred to plants, plates were clipped to the glass vessel base without constricting the plant. Air, purified by passing through an activated charcoal filter (BDH, 10–14 mesh, 50 g), was pushed into the vessel through the inlet port at 700 ml min⁻¹ (flow rate controlled by needle valve and measured by a flow meter). Air was pulled out at 500 ml min⁻¹ through Porapak Q 50/80 (50 mg, Supelco, Bellefonte, PA, USA) held by two plugs of silanised glass wool in a 5-mm diameter glass tube (Alltech Associates, Carnforth, Lancashire, UK). All connections were made with polytetrafluoroethylene (PTFE) tubing (Alltech Associates) with brass ferrules and fittings (North London Valve, London, UK) and sealed with PTFE tape. Upon each experiment, glassware, metal plates and other equipment were washed with Teepol detergent (Teepol, Kent, UK) in an aqueous solution, acetone and distilled water, and then baked overnight at 180°C. Porapak Q filters were conditioned before use by washing with redistilled diethyl ether (4 ml) and heated to 132°C under a stream of purified nitrogen and kept for 2 h. Diethyl ether was purchased from Sigma Aldrich and was distilled prior to use.

VOC extracts required for GC–MS analyses were collected over 5 days in 24 h periods (n = 3) (Hegde et al., 2012; Sobhy et al., 2017). Although the trapped VOCs in Porapak Q tubes were eluted after each collection period with freshly redistilled diethyl ether (750 μL), we analyzed here only the ones that were collected at 0–24, 48–72, and 96–120 h to be thus consonant with the growth assay. The samples were concentrated under a stream of nitrogen to ~50 μL and stored in small vials (Supelco, Amber Vial, 7 mL with solid cap w/PTFE liner) at −20°C until required for analysis.

VOC extracts were analyzed by high-resolution gas chromatography (GC) using an Agilent 6890 GC equipped with a cool on-column injector, flame ionization detector (FID) and a nonpolar HP-bonded phase fused silica capillary column (50 m × 0.32 mm inner diameter, film thickness 0.5 μm; J & W Scientific). The GC oven temperature was maintained at 30°C for 1 min after sample injection and then raised by 5°C min⁻¹ to 150°C, thereafter by 10°C min⁻¹ to 230°C. The carrier gas was hydrogen. 4 μl of each eluted sample were manually injected into the injector port of the GC instrument.

Coupled gas chromatography–mass spectrometry (GC–MS) analysis of VOCs from treated potato plants was performed using a non-polar column (HP-1, 50 m × 0.32 mm inner diameter, film thickness 0.5 μm), attached to a cool on–column injector, which was directly coupled to a magnetic sector mass spectrometer (Autospec Ultima, Fisons Instruments, Manchester, UK). Ionization was by electron impact at 70 eV, 250°C (source temperature). Helium was the carrier gas. The GC oven temperature was maintained at 30°C for 5 min, and then programmed at 5°C min⁻¹ to 250°C.

Tentative identifications were made by comparison of spectra with mass spectral databases (NIST, 2011). Peak enhancement by co-injection with authentic standards confirmed tentative identifications (Pickett, 1990). Excluding (E)-β-farnesene, (E)-4,8-dimethyl-1,3,7-nonatriene (DMNT), and (E,E)-4,8,12-trimethyl-1,3,7,11-tridecatetraene (TMTT) that were synthesized as described in Sobhy et al. (2017), chemicals (>95% pure) were obtained from Sigma-Aldrich® (Gillingham, Kent, UK) and Botanix Ltd® (Paddock Wood, Kent, UK). An external standard (100 ng/μl non-anal) method was used to quantify the amount of identified chemical components present in the air entrainment samples. The amounts of compounds present in VOC extracts were determined in accordance with the weight of sampled plant material, and the duration of entrainment period. Data were analyzed using HP Chemstation software.

Visualization together with hierarchical clustering of VOCs data was done using the comprehensive online tool suite MetaboAnalyst 4.0 (Chong et al., 2018). Data matrix was first mean-centered, cube-root transformed prior to analysis. Average linkage hierarchical clustering based on Ward clustering algorithm of the Euclidean distance measure for the differentially emitted VOCs detected by ANOVA (Tukey's post-hoc tests, P < 0.05) was used to construct a heatmap. Additionally, principal component analysis (PCA) was used to visualize overall differences among VOC blends emitted from plants subjected to each treatment (i.e., Control, CJ, and SUR) at each collection point (i.e., 24, 72, and 120 h). In this analysis, we used two outputs from the data; a matrix of “scores” which provides the location of each compound along each principal component (PC) and a matrix of “loadings” which indicates the strength of correlation between individual VOCs and each PC.

We used multivariate analysis of variance (two-way MANOVA) to analyze the effects of treatment, sampling time, and their interaction on volatile emission, with the “treatment” and “time” as fixed factors and the volatile compounds as dependent variables (IBM SPSS Statistics version 22.0, Armonk, NY, USA). Given that these VOCs share common precursors and their variation patterns within groups of compounds can be attributed to variation in the quantity of their common precursor (Hare, 2011), VOCs were grouped into alcohols, aldehydes, benzenoids, esters, homoterpene, ketones, monoterpenes, and sesquiterpenes based on their functional or biosynthetic characteristics. Groups containing two compounds (or more) were considered for two-way MANOVA. Subsequently, two-way MANOVA was followed by two-way ANOVA for the different VOC groups on the sum of all compounds under each to determine the effects of treatment, sampling time and their interaction, or by one-way ANOVA to only determine the effects of treatment only, followed by Tukey post-hoc comparisons (P < 0.05). Two-way ANOVA was also performed on individual compounds not included in MANOVA, and on total emitted volatiles (i.e., the sum of all VOCs).

For the aphid performance experiments, natural log (ln) transformed data for birth, 7 days and adult weights, the MRGR, D, FD, and r_m were compared by one-way ANOVA using SigmaPlot 12.3 (SPSS Inc., Chicago, IL, USA). A natural log (ln) transformation was needed in order to normalize the residuals of the weight data. Prior to analysis, data were examined for a Normal distribution using the Shapiro-Wilk test. Comparisons among means were performed using Tukey post-hoc test (P < 0.05).

The growth rate parameters of M. euphorbiae were significantly affected by CJ treatment. Females of M. euphorbiae produced significantly less neonate nymphs on treated potato plants compared to control or SUR plants [F_{(2, 27)} = 27.94; P = <0.001; Figure 2A]. In addition, the survival of M. euphorbiae nymphs significantly decreased on CJ-treated than on control plants [F_{(2, 27)} = 40.05; P = <0.001; Figure 2B]. Further, the MRGR of M. euphorbiae nymphs that fed on CJ treated plants was significantly lower compared to those reared on either control or SUR plants [F_{(2, 27)} = 6.203; P = 0.006; Figure 2C]. The mean intrinsic rate of population increase (r_m) of M. euphorbiae was significantly reduced on CJ-treated potato compared to control or SUR plants [F_{(2, 27)} = 6.716; P = 0.004; Figure 2D].

GC-MS Analysis of headspace collections from potato plants revealed the presence of 22 detectable VOCs following eight various classes based on their functional or biosynthetic characteristics (i.e., alcohols, aldehydes, benzenoids, esters, homoterpenes, ketones, monoterpenes, and sesquiterpenes). Overall, two-way MANOVA revealed that the emission of these VOC classes from potato plants was affected significantly by treatments, sampling time, and their interaction (Table 1). In addition, most volatile individuals under the above-mentioned VOC groups were significantly affected by potato treatment with CJ, particularly during 48–72 and 96–120 h after volatile collection was initiated (Table 2). More specifically, the total amount of emitted VOCs was significantly higher from CJ-treated plants compared to the control or SUR plants [0–24 h: F_{(2, 6)} = 165.93, P < 0.001; 48–72 h: F_{(2, 6)} = 154,07, P < 0.001; 96–120 h: F_{(2, 6)} = 232.89, P < 0.001], particularly due to the high emission of several VOCs from CJ-treated plants at the three collection points (i.e., 0–24, 48–72, 96–120 h), including methyl salicylate (MeSA), indole, DMNT, TMTT, CJ, MHO, and (E)-β-farnesene (Table 2). VOCs that were significantly increased but for which identifications remained tentative included (Z)-3-hexen-1-ol, a longifolene isomer, (E)-(1R,9S)-caryophyllene, and a germacrene D isomer (Table 2). Similarly, two-way ANOVA revealed that the emission of all assigned VOC groups, the individual compounds [i.e., (Z)-3-hexen-1-ol and (Z)-3-hexen-1-yl butyrate] and total emitted VOCs were significantly affected by both factors (i.e., treatments and sampling time) and their interactions with one exception for (Z)-3-hexen-1-yl butyrate (Supplementary Table S1).

The resulting heatmap showed differential magnitude of VOC emission between treatments with a greater increase from CJ-treated plants (Figure 3).

Furthermore, principal component analysis (PCA) of the VOCs showed that the first two principal components accounted for around 47.8, 46.1, and 43.6% of the total variation between treatment groups at 0–24, 48–72, and 96–120 h VOC collection points, respectively (Figure 4). Hence, these two PCs illustrated most of the variation in the data of likely biological relevance. Overall, PCA revealed a clear separation in the obtained VOC profiles, which was mainly driven by cis-jasmone treatment. Specifically, the score plot visualized three neatly separated groups of control plants, SUR plants and CJ-primed plants at 0–24, 48–72, and 96–120 h VOC collection points. The greatest loadings of PC1_0–24h, in descending order, were for TMTT (0.31), α-pinene (0.29), and DMNT (0.27), whereas the greatest loadings of PC2_0–24h were for longifolene isomer (0.31), bisabolene isomer (0.28), and cis-jasmone (0.25). In addition, the main loadings of PC1_48–72h were for α-copaene (0.37), bisabolene isomer (0.30), and MHO (0.23), whereas the main loadings of PC2_48–72h were for (E)-ocimene (0.42), cis-jasmone (0.39), and indole (0.27). Finally, the major loadings of PC1_96–120h were for MeBA (0.36), MHO (0.34), and α-humulene (0.31), whereas the major loadings of PC2_48–72h were for (Z)-3-hexen-1-ol (0.33), DMNT (0.32), and (E)-ocimene (0.31). This suggests that these VOCs are contributing to both PC1 and PC2 at each collection point and may subsequently impact aphid overall performance on potato. Interestingly, as shown in Figure 4C, the vectors of several VOCs were more associated with CJ treatment, indicating enhanced VOC emissions from CJ-plants.

Growth assays revealed that adults of M. euphoribae produced fewer neonates on CJ plants compared to control ones. Similarly, the numbers of grain aphid (Sitobion avenae Fabricius) were reduced in plots of winter wheat treated with CJ (Birkett et al., 2000; Bruce et al., 2003). Using our study model, Brunissen et al. (2010) reported that potato treatment with MeJA, which is JA volatile derivative, reduced the fecundity of M. euphoribae and further prolonged its pre-reproductive period. Likewise, induced tomato plants drastically reduced the reproductive rate of M. euphorbiae (Guerrieri et al., 2004).

The suppressed performance of aphids on CJ-primed plants has been reported on many different crop plants. For instance, cereal aphids were negatively affected by CJ treatment and fewer aphids were observed on CJ plants than on untreated ones (Bayram and Tonğa, 2018). Furthermore, the developmental rate and MRGR, but not intrinsic rate, of the glasshouse potato aphid (Aulacorthum solani Kalt.) were significantly reduced on sweet pepper plants treated with CJ (Dewhirst et al., 2012). Congruously, Bruce et al. (2003) reported a reduced intrinsic rate of population increase and fewer numbers of grain aphid on CJ-treated wheat. This negative influence of CJ application can be also extended to include the foraging activity of aphids (Paprocka et al., 2018). Interestingly, more deleterious insects are negatively influenced by CJ application as such as thrips (Egger and Koschier, 2014; Egger et al., 2016) and stink bugs (da Graça et al., 2016). Further, not only puncturing and/or sucking insects are adversely affected by CJ application, but also chewing lepidopteran pests such as the noctuid, Spodoptera exigua (Hübner), which laid fewer egg masses when reared on CJ-treated tomato plants (Disi et al., 2017). Such broader negative impact on diverse herbivore species raises the question whether CJ may be a common defense-priming signal in nature.

VOC analysis disclosed a drastic increase in VOC emission from CJ primed plants which clearly illustrated by heatmap analysis. Most promising was our finding that such increased VOC emission was detected at the three collection points (i.e., 0–24, 48–72, and 96–120 h), suggesting long-lasting induction for nearly a week following plant treatment with CJ.

(Z)-3-hexen-1-ol was significantly emitted in higher amounts from CJ plants. (Z)-3-Hexen-1-ol is a key VOC emitted from plants upon aphid herbivory (Webster et al., 2008), as demonstrated by being electrophysiologically and behaviorally active to aphids (Webster et al., 2010) and highly attractive to their natural enemies (Du et al., 1998).

Benzenoid VOCs (i.e., MeSA and indole) were released in considerably higher amounts from CJ-primed plants at all VOC sampling points. Concordant with this, cotton priming with CJ induced the emission of several VOCs, including MeSA (Hegde et al., 2012). For the latter benzenoid, previous work has shown an increased emission of indole from CJ-primed plants (Delaney et al., 2013), which has been shown to be an electrophysiologically active VOC to M. euphorbiae (Sobhy et al., 2017), suggesting its potential role in potato/aphid interactions.

Production of both homoterpenes, DMNT and TMTT, was robustly elicited upon plant treatment with CJ. Similarly, the release of TMTT and/or DMNT was substantially increased following bean (Birkett et al., 2000), soybean (Moraes et al., 2009), maize (Oluwafemi et al., 2013), and potato (Sobhy et al., 2017) treatment with CJ. Boosting the long-lasting induction of CJ treatment, induced release of DMNT and TMTT was observed in CJ-primed cotton over a period of 5 days (Hegde et al., 2012). Interestingly, plant application with other priming agents such as BTH, laminarin (β-1,3 glucan) and JA also increased the emission of both homoterpenes (Smart et al., 2013; Sobhy et al., 2015a, 2018).

CJ treated plants released higher amounts of the ketones MHO and CJ, which have been shown to possess biological activity against aphids (Sobhy et al., 2017). In particular, a higher emission of MHO was reported from various crop plants treated with jasmonate related compounds, such as CJ (Pickett et al., 2007a; Dewhirst et al., 2012; Sobhy et al., 2017) and MeJA (Yu et al., 2018). It should be noted that CJ itself is a stress-related VOC that is emitted in elevated concentrations from plants treated with different priming agents (Sobhy et al., 2015a, 2017).

Of all detected sesquiterpenes, the emission of (E)-β-farnesene and the tentatively identified longifolene isomer, (E)-(1R,9S)-caryophyllene, and (R) or (S)-germacrene D was notably induced upon CJ treatment. Consistent with our results, Oluwafemi et al. (2013) found that primed maize released higher quantities of germacrene D when exposed to CJ. Moreover, it has been also shown that significant emission of germacrene D was released from plants induced by either JA (Obara et al., 2002) or MeJA (Yu et al., 2018). A growing body of evidence suggest that plant priming with CJ elicits the emission of (E)-(1R,9S)-caryophyllene and (E)-β-farnesene (Delaney et al., 2013; Oluwafemi et al., 2013). It should be noted that majority of the above-mentioned VOCs that were induced upon CJ priming are indeed key semiochemicals in plant/aphid interactions.

One mechanistic interpretation for the reduced aphid performance observed on CJ-primed plants would be attributed to the higher emission of aphid defense-related HIPVs following CJ application. Aphids may therefore perceive such induced or altered HIPV profiles from CJ-primed plants as signals of a greater risk of competition from conspecifics or as a greater risk of predation from natural enemies (Bruce and Pickett, 2011). Consistent with our findings, the coordinated defense responses expressed in primed plants, when exposed to contact-induced volatiles, reduced aphid settling, making these plants a less suitable host for aphids (Markovic et al., 2019). Furthermore, exposing potato to VOCs from undamaged onion plants altered its volatile profile and this had a deterrent effect against host-seeking M. persicae, which was further confirmed in the field where migration of aphids into potato was significantly reduced by intercropping with onion (Ninkovic et al., 2013). Similarly, lower MRGR of the bird cherry oat aphid (Rhopalosiphum padi L) was reported on barley plants exposed to the volatiles of the weed Chenopodium album (Ninkovic et al., 2009). Remarkably, such reduced aphid performance on primed plants can be also obtained by exposing these aphid-infested plants to a single stress-related volatile. For instance, the fertility of M. euphoribae was significantly reduced when individuals fed on tomato plants exposed to MeSA (Digilio et al., 2012). In addition, exposing Chinese cabbage to β-ocimene negatively influenced the feeding behavior of M. persicae (Kang et al., 2018). Similarly, using M. euphoribae, significant lower numbers of newborn nymphs and lower number of settled individuals, with lower weight, were recorded on tomato plants when primed by β-ocimene (Cascone et al., 2015). More recently, Maurya et al. (2019) reported a reduced fecundity of pea aphid on barrel medic plants upon their exposure to indole.

Multivariate analysis (PCA) of VOC data further indicated which volatiles may be involved in aphid reduced performance. In particular, TMTT, α-pinene, and DMNT had the greatest loadings for PC1, which separated the CJ from the Control and SUR treatments, from the 0–24 h air entrainment. For both the 48–72 and 96–120 h air entrainment periods, PC2 separated the CJ from the Control and SUR treatments. (E)-ocimene, cis-jasmone and indole had the greatest loadings for PC2_48–72h, whereas (Z)-3-hexen-1-ol, DMNT, and (E)-ocimene had the greatest loadings for PC2_96–120h and thus were much likely dominating in the emitted VOC blends. This suggests that these volatiles correlate most strongly with aphid reduced performance. Indeed, these VOCs are key semiochemicals in aphid trophic interactions and are directly associated with aphid repellency and/or attraction of aphid natural enemies. Previous work has shown an increased emission of (Z)-3-hexen-1-ol (Sasso et al., 2007; Webster et al., 2008), α-pinene (Sasso et al., 2007), (E)-β-ocimene (Sasso et al., 2007; Staudt et al., 2010), DMNT, and TMTT (Hegde et al., 2011), CJ (Birkett et al., 2000), and indole (Sobhy et al., 2017) in response to aphid herbivory. It should be noted that these aphid-stress signals are significantly repellent to aphids (Bruce et al., 2005; Beale et al., 2006; Hegde et al., 2011). Further, we have shown recently that alate M. euphorbiae showed electrophysiological activity to most of these VOCs that are emitted from CJ treated plants (Sobhy et al., 2017). Alternatively, (Z)-3-hexen-1-ol (Du et al., 1998; Sasso et al., 2009), α-pinene (Corrado et al., 2007), (E)-β-ocimene (Birkett et al., 2000), DMNT and TMTT (Hegde et al., 2011), CJ (Birkett et al., 2000), and indole (Han and Chen, 2002) have shown robust potential in attracting several natural enemies of aphids. This points to possible association of the emission of these VOCs as danger signals and the likely risk of being attacked by their parasitoids and/or predators.

The reduced performance of other herbivores following exposure to HIPVs has been reported previously. For example, exposure to HIPVs decreased the growth rate of Spodoptera littoralis (Boisduval) caterpillars at early larval stages (von Mérey et al., 2013). In addition, leaf beetle, Plagiodera versicolora (Laicharting), larvae showed decreased performance on uninfested host plants when exposed to HIPVs from plants infested by conspecific larvae (Yoneya et al., 2014). Interestingly, such decreased herbivore performance following exposure to HIPVs can be also achieved by treating the plants with priming agents. In this respect, using CJ but with a chewing herbivore, females of S. exigua laid fewer egg masses when subjected to filter papers containing VOCs of CJ-primed plants (Disi et al., 2017). Furthermore, plant treatment with the priming agent prohydrojasmone [propyl (1RS, 2RS)-(3-oxo-2-pentylcyclopentyl) acetate] negatively affected the oviposition of Tetranychus urticae Koch (Uefune et al., 2013), due to several quantitative and qualitative changes in HIPVs in prohydrojasmone-treated plants.

Additional indirect evidence on the potential of bioactive volatiles as priming agents that render exposed plants less appealing to herbivores and possibly more attractive to natural enemies comes from cotton growers in developing countries. More specifically, farmers used to remove the growing buds of cotton plants just before flowering, which is known as cotton topping (Renou et al., 2011). This in turn, in addition to branching and flowers increase, significantly reduced the attack of many lepidopteran pests in cotton. Remarkably, this not only occurred in topped plants, but also was extended to encompass neighboring plants that were not topped, suggesting that the released VOCs from topped plants can protect their neighbor plants from an impending attack by repelling herbivores and/or attracting beneficial insects (Llandres et al., 2019).

Another interpretation of the reduced performance of M. euphorbiae when fed on CJ-plants is that treating plants with CJ may induce the accumulation of certain anti-herbivore secondary metabolites that renders CJ-plants more inappropriate for aphid survival. In this respect, significantly higher levels of 2,4-dihydroxy-7-methoxy-2H-1,4-benzoxazin-3(4H)-one (DIMBOA), 2-hydroxy-7-methoxy-(2H)-1,4-benzoxazin-3(4H)-one (HBOA), and phenolic acids were detected in aerial parts and roots of wheat, following CJ treatment (Moraes et al., 2008). Similarly, increased levels of the defensive isoflavonoids (i.e., daidzein and genistein) were reported by da Graça et al. (2016) upon CJ treatment, which negatively impacted the weight gain of stink bug (Euschistus heros).

Yet, many questions remain unknown and require more studies to answer, especially concerning the underlying induced changes upon plant priming and how induced responses precisely reduce the herbivore performance. To this end and to obtain further empirical evidence, it is advised that similar experimental work should be performed but using knock-out mutant plants in which the pathways producing specific classes of VOCs are silenced (Matthes et al., 2010; Christensen et al., 2013).

Potato is currently one of the five major crops worldwide that contributes with 2.2% of the global human calorific intake (King and Slavin, 2013). While the increase in consumption requires drastic increase in potato yield (FAO, 2009), insect infestation is still a major constraint reducing potato production worldwide, especially those vector plant viruses and have tremendous indirect damage such as aphids (van Emden and Harrington, 2017). Our results show clear evidence that CJ application decreases aphid growth, reproduction, intrinsic rate and survival. These life history traits are indeed vital factors determining the success of aphid population (Leather and Dixon, 1984). Thus, negatively affecting these traits would likely reduce the overall aphid fitness. Given the current thinking that effective control does not necessarily have to rely on aphid mortality, we believe that priming potato defense against aphids using CJ has the potential to provide new opportunities for sustainable crop protection, especially if being used in the context of IPM, through the induction of VOC emissions (Turlings and Erb, 2018; Brilli et al., 2019). This could minimize the need for applying conventional toxic agrochemicals in potato fields. Bearing in mind that plant genotype, application method and other environmental conditions strongly influence the magnitude of any enhancement in plant defense acquired by these priming agents, more field research is needed before upscaling their application under field conditions.

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.