Heat-stress induced sesquiterpenes of Chrysanthemum nankingense attract herbivores but repel herbivore feeding

Plants are frequently exposed to heat stress as a result of global warming. Heat stress leads to a series of physiological responses including stress volatile elicitation, but how heat stress-induced volatile cues affect the behavior of herbivores is poorly understood. In this study, the polyphagous herbivore Spodoptera litura (tobacco cutworm, TCW) and Chrysanthemum nankingense were selected as the model to elucidate the interactions between herbivore behavior and heat stress-induced plant physiological changes. Photosynthetic characteristics and volatile emissions were measured in C. nankingense control plants (25 °C for 3 h), in C. nankingense exposed to moderate (35 °C for 3 h), and severe (45 °C for 3 h) heat stresses. Net photosynthetic rate (An) decreased by more than two-fold after exposure to 45 °C due to non-stomatal inhibition of photosynthesis. 45 °C treatment induced emissions of the camphor and (E)-β-caryophyllene. Exposure to 35 °C had minor effects on photosynthetic characteristics and did not induce terpene emissions. Using dual-choice olfactometer bioassays, we found that 45 °C treatment enhanced the attractiveness of the plants to TCW. Moreover, the leaf concentrations of nine sesquiterpenes were increased and the feeding of TCW was strongly inhibited after 45 °C treatment compared with control plants. Taken together, our study highlights the impact of heat stress on the behavior of the herbivore mediated by the accumulation and emission of sesquiterpenes and suggests altered pest-host interactions under future warmer climates. Modulation of terpenoid emissions and contents should be considered in developing future ecological pest control strategies in agricultural fields.


Introduction
Global environmental change is associated with unprecedently rapid increases in global mean temperature and enhanced frequency and duration of extreme heat stress episodes (Parmesan & Yohe 2003;Harvey et al. 2020;Jagadish et al. 2021). Thus, heat stress is becoming one of the most frequent and harmful stressors among different abiotic factors in nature (Perkins et al. 2012), and its impact on vegetation is expected to gradually increase in the future as global warming proceeds (IPCC 2013).
The response of plants to heat stress typically involves several stages from initial response to recovery and acclimation for moderately severe stress to necrosis and death for extremely severe stress (Morimoto 1998;Hamerlynck et al. 2000;Sung et al. 2003;Baniwal et al. 2004;Wang et al. 2018;Turan et al. 2019). Heat stress acclimation is associated with a series of physiological and biochemical reactions, including changes in plant morphology, photosynthetic characteristics, respiration, hormone production, volatile organic compound (VOC) production and antioxidant capacity (Larkindale & Knight 2002;Peñuelas & Llusià 2003;Niinemets 2010;Asseng et al. 2011;Hüve et al. 2012;Casal & Balasubramanian 2019;Ma et al. 2021). Photosynthesis is typically the first cellular function to be impaired by heat stress (Al-Khatib & Paulsen 1984;Schrader et al. 2004;Kepova et al. 2005;Hüve et al. 2011Hüve et al. , 2019Zou et al. 2017). Inhibition of foliage photosynthetic activity by mild heat stress is typically readily reversible, while severe heat stress can lead to sustained inhibition or even continuous reduction in photosynthesis after return to the lower temperature (Hüve et al. 2011;Turan et al. 2019).
Several studies have reported that the irreversible reduction of photosynthetic activity is related to the release of stress-induced VOC, including rapid release of shortchained oxygenated compounds such as methanol, acetone, acetaldehyde, and various lipoxygenase pathway-derived compounds (volatile LOX products, also called green leaf volatiles) (Loreto et al. 2006;Copolovici et al. 2012;Kask et al. 2016;Turan et al. 2019). Apart from stress-induced LOX-emissions, heat stress can also result in changes in the emissions of terpenoids that are stored or de novo synthesized (Monson & Fall 1989;Llusià & Peñuelas 1999;Copolovici et al. 2012;Turan et al. 2019). In addition, in storage emitters, severe stress can also enhance terpene emissions by breaking the trichomes or increasing the permeability of glandular trichome outer surfaces (Guenther et al. 1993;Grote et al. 2013).
Terpenoids represent the largest and most diverse group of plant specialized metabolites (Aharoni et al. 2005) and play key roles in the interaction of plants with their environment. Thus, the stress-induced terpene emissions have been demonstrated with multiple biological implications. In particular, volatile terpenoids emitted from plants can affect herbivore behavior and serve as an indirect defense mechanism by attracting the natural enemies of the herbivores to protect the plant from further damage (Dobson 1994;Köllner et al. 2008;Abbas et al. 2017;Eberl et al. 2018). In addition, synthesis of volatiles stored in plants can be enhanced by different abiotic and biotic stresses, and these stored compounds can play a vital role as direct defenses deterring herbivore feeding (Agliassa & Maffei 2018;Mahajan et al. 2020;Wang et al. 2021). Stress-induced onset of the release of de novo synthesized terpenoids needs the expression of corresponding terpene synthases. Hence, the stress-induced terpene emissions usually occur with a time delay (Pazouki et al. 2016). The notice of the delayed response of terpenoids to the stresses will help select the time scale when monitoring the dynamics and be important to understand the role of the induced terpenoids.
In past reports, drought, waterlogging, and high salinity are some of the abiotic stresses that have been studied most and can indirectly affect the performance, diversity and abundance of feeding insect through changes in plant physiology (Wearing 1967;Mattson & Haack 1987;Huberty & Denno 2004;Han et al. 2014;Hoang et al. 2016). However, there are few studies on the impact of volatile cues on orientation and feeding behavior of herbivores after heat stress (Harvey 2015).
Polyploid species in the genus Chrysanthemum are commercially important ornamental plants that have been cultivated in China for more than 3000 years (Cheng et al. 2010). Temperature is one of most critical factors in the field production of Chrysanthemum spp. Generally, the growth of Chrysanthemum spp. will be disturbed above 32 °C and be terminated at 40°Ct (Kong et al. 2019). The expected increase of the occurrence of extremely high temperatures above 40 °C due to global warming is predicted to have a major negative influence on field production of Chrysanthemum spp., and this could be further exacerbated by potential enhancement of pest attacks (Lemoine et al. 2014;Sagheer 2019;Hamann et al. 2020). The diploid species Chrysanthemum nankingense is a wild relative of C. morifolium (Cheng et al. 2010). Because most of the ornamental chrysanthemums are tetraploid or hexaploid and have a complicated genetic background, C. nankingense can be used as a simple model to study the response of plants from Compositae to abiotic stress (Cheng et al. 2010). Besides, chrysanthemum plants are susceptible to a broad array of herbivores, such as leaf miners, beet armyworms and aphids, which affect the vegetative organs and also decrease the quality of flowers during their growth and development (Visser et al. 2007;Xia et al. 2014). It has been observed that the larvae of the polyphagous moth Spodoptera litura Fabr. (tobacco cutworm, TCW) occurred on about 29 species of Compositae (Lim et al. 2012). TCW is a major pest that feeds on plant leaves in tropical and subtropical areas of Asia (CABI 2021). Given the severe damage of TCW to Chrysanthemum spp., TCW can be regarded as an ideal model herbivore for the volatile-mediated biological assay.
In this study, we first address the impact of heat stress of different severity (mild vs. severe) on photosynthetic traits and volatiles emissions of C. nankingense leaves. We further explored the influence of heat stress-mediated changes in volatile emissions and volatile contents on the orientation and feeding behavior of TCW in C. nankingense. We hypothesized that foliage photosynthetic activity decreases and volatile emissions increase in heat stress severitydependent manner. We also hypothesized that heat stressinduced volatiles reduce the attractiveness of C. nankingense leaves to TCW, and heat stress-dependent changes in secondary metabolite content reduce the feeding of TCW. Our results could be used to develop new strategies for ecological control of pests in the agriculture fields.

Plant materials
Chrysanthemum nankingense plants were obtained from the greenhouse at the Chrysanthemum Germplasm Resource Preserving Centre, Nanjing Agricultural University, China (118° 98′ N, 32° 07′ E). Healthy shoot cuttings of similar size were grown under controlled conditions in a growth chamber for 35 days (Ningbo Southeast Instrument Co., Ltd.). Light intensity at plant level was 12.6-19.8 μmol m −2 s −1 , photoperiod length 16 h, air temperature 25 °C (day)/18 °C (night), relative humidity 68-75%. Morphologically uniform seedlings with 10-15 leaves were used for all treatments.

Heat stress treatment
Three treatments with 25 °C (control), 35 °C (moderate heat stress), and 45 °C (severe heat stress) were applied to the seedlings for 3 h in a growth chamber. When the growth chamber (light intensity 19.8 μmol m −2 s −1 , relative humidity 68-75%) reaches the set temperature, the plants were put into it at 8:00 am. A total of 23 plants were used for 25 °C and 45 °C treatment respectively (five plants for photosynthetic traits measurements; five plants for headspace volatile collection; three plants for organic extraction; five plants for Y-tube olfactometer bioassay; five plants for herbivore feeding). A total of 10 plants were used for 35 eatreatment (five plants for photosynthetic traits measurements; five plants for headspace volatile collection). A total of 33 plants were used for all the treatments.

Photosynthetic trait measurements
After heat stress treatment, from each plant, a fully-expanded fifth leaf from the top was selected to measure leaf photosynthetic traits with a portable photosynthesis system (LI-6800, LI-COR Biosciences, USA). The leaf was clipped in the leaf chamber and the following environmental conditions were established and maintained during the measurements: chamber CO 2 concentration, 400 μmol mol −1 ; air flow rate, 600 μmol s −1 ; leaf temperature, 25 °C; relative humidity, 50-75%; light intensity, 1000 μmol m −2 s −1 (referring to the LI-6800 operating manual). The leaf was kept as the standard conditions for 30 min., and net assimilation rate (A n ), stomatal conductance to water vapor (G s ), and intercellular CO 2 concentration (C i ) were recorded in 5 min intervals for 30 min.

Organic extraction of leaf terpenoids
The third leaf from the top collected right after heat stress treatment from 45 °C treated and control plants were powdered in liquid nitrogen. Ethyl acetate (Macklin Technology, Shanghai, China) was added to the powder in a 5:1 (volume to mass) ratio, with 0.002% nonyl acetate (CAS:143-13-5, ≥ 98%, Sigma Aldrich, St Louis, MO, USA) included as an internal standard (0.2 g dry starting material: 1 ml extraction solvent) (Chen et al. 2009). After shaking at 200 rpm at room temperature for two hours and centrifugation at 5000 rpm for five minutes, the organic phase was collected for subsequent GC-MS analysis.

Volatile collection and identification by GC-MS
Volatiles were immediately collected from the heat stresstreated and control plants after 3 h treatment with an open headspace sampling system (Analytical Research Systems, Gainesville, FL, USA) as previously reported (Yuan et al. 2008). The volatiles were collected for 4 h by sucking air through a volatile collection trap filled with Super Q adsorbent (Alltech Assoc., USA) and were eluted with 100 μl of CH 2 Cl 2 containing 0.001% nonyl acetate as the internal standard (Jiang et al. 2018). 1 μl of the eluent was injected into a gas chromatography-mass spectrometry (GC-MS) system (Agilent Intuvo 9000 GC system coupled with an Agilent 7000D Triple Quadrupole mass detector) for separation and identification of terpenes. The separation was performed on an Agilent HP 5 MS capillary column (30 m length × 0.25 mm inner diameter). Helium with a flow rate of 5 ml min −1 was used as the carrier gas. Splitless injection (injection temperature 250 °C) with a temperature gradient of 6 °C min to 300 °C was applied. The temperature of the injection port was 260 °C, with a split mode (5:1), and a linear temperature gradient was used for compound separation. The column initial temperature was 40 °C and the temperature was increased to 250 °C at a rate of 5 °C min −1 . The MS was operated in the electron impact mode with ion source temperature set at 230 °C, ionization energy was 70 eV, and mass scan range 40-500 amu. Terpenoid products were identified using the National Institute of Standards and Technology mass spectral database (NIST 17.0) and by comparison of retention times and mass spectra with available authentic standard compounds. Quantification was performed based on peak areas of mass chromatograms ).

Y-tube olfactometer bioassay
The third instar larvae of S. litura (tobacco cutworm, TCW) were used as model herbivores to infest plants. Larvae of TCW were purchased from Henan Jiyuan Baiyun Industry Co., Ltd. We used a Y-tube olfactometer equipped with a Y-shaped glass tube (one 20 cm arm and two 15 cm branched arms, 1 cm diameter) to explore the olfactory orientation behavior of the TCW larvae. The branched arms were connected to two glass bottles as odor sources. One arm was connected to a bottle containing a control plant, another to a bottle with a plant exposed to 45 °C exposed to an empty bottle. Each glass bottle was supplied by air with a low-pressure air pump through a charcoal filter at a rate of 100 ml min −1 . The third instar TCW larvae were collected from rearing cages in a separate insectary room and starved for 20 h before each trial, then a group with 10 larvae was released into the base of the tube and each larva entered the Y-tube one at a time. If a larva crossed the bifurcation point of 3 cm within one arm, a positive or a negative response was recorded. If the larva did not make a clear choice after 5 min, it was considered as unresponsive. Besides, the time spent by larvae in the different arms of the olfactometer was recorded. After passage of five larvae, the olfactometer was rotated 180° to exclude the position interference. Each experiment was conducted with plants of control and 45 °C heat treatment in five replicates (Sun et al. 2015).

Evaluation of damage severity by herbivore feeding
Third instar TCW larvae were collected from rearing cages in an insectary room and starved for 20 h before each trial. With another batch of plants treated by 45 °C for 3 h in five replicates, seven larvae were evenly put on the leaves (one larva per leaf) of each plants at 5:00 pm. Based on the nighttime feeding habit of S. litura (Kawasaki 1986), the leaves fed by the larvae in each plant were collected after the larval infestation for overnight (16 h), and the infested area was calculated using image J (1.52v, National Institutes of Health, USA).

Statistical analysis
One-way ANOVA was used to evaluate the differences in photosynthetic traits, and t-test was used to evaluate the differences in terpenoid emission, terpenoid concentration, Y-tube olfactometer bioassay and feeding area (SPSS version 19.0).

Effects of heat stress on photosynthetic traits of C. nankingense leaves
Compared with the control treatment (25 °C), 3 h exposure to 35 °C did not affect net assimilation rate (A n ), but A n was reduced by about 70% in the plants exposed to 45 °C (Fig. 1a). Stomatal conductance to water vapor (G s ) was initially reduced in plants exposed to 35 °C, but it reached similar levels as G s in control plants during 30 min measurement period (Fig. 1b). G s was not affected by exposure to 45 °C (Fig. 1b). Similarly to G s , the intercellular CO 2 concentration (C i ) was initially lower for plants exposed to 35 °C than for plants in the control treatment, and reached a similar level as the measurements continued (Fig. 1c). At the end of the measurements, plants exposed to 45 °C had a higher C i than controls plants and plants exposed to 35 °C (Fig. 1c).

Heat stress treatment induced the terpene emission in C. nankingense
There is no modification of the terpene profile with the exposure to 25 °C (Control) and 35 °C (moderate heat stress). However, heat treatment at 45 °C (severe heat  Fig. 1 Effect of moderate (exposure to 35 °C for 3 h) and severe (45 °C for 3 h) heat stress on average ± SE net assimilation rate (A n ) (a), stomatal conductance to water vapor (G s ) (b), and the intercellular CO 2 concentration (C i ) (c) of the perennial herb Chrysanthemum nankingense. The control plants were kept at 25 °C for 3 h under otherwise identical conditions. Different letters indicate statistically significant differences among the means according to ANOVA analysis (n = 5, P < 0.05) stress) enhanced both emissions of mono-and sesquiterpenes (Fig. 2a). The monoterpene camphor and sesquiterpene (E)-β-caryophyllene were the key induced terpenes emitted after the 45 °C treatment with the emission rates of 5.45 ± 0.59 ng h −1 g −1 and 9.56 ± 2.16 ng h −1 g −1 , respectively (Fig. 2b).

Heat stress enhanced TCW visitation of C. nankingense leaves
Y-tube olfactometer assays showed that the control plants were not attractive to the larvae of TCW (no choice difference between control plants and clean; Fig. 3). The plants exposed to 45 °C were more attractive to TCW (62.00 ± 3.37%; P < 0.01) than control plants (Fig. 3). These results clearly revealed the determinant role of volatile emissions from heat stress treated plants in TCW attraction. Chromatogram profiles (total ion chromatograms, TIC) of terpenoid compounds emitted from leaves of C: nankingense control plants (25 °C) and plants exposed to moderate (35 °C for 3 h) and severe (45 °C for 3 h) heat stress (a). The volatiles were collected from the headspace and analyzed with GC-MS. IS denotes the inter-nal standard (nonyl acetate). Unlabeled peaks are not terpenoids. 1, camphor; 2, (E)-β-caryophyllene. Average ± SE (n = 5) of emission rates of total terpenoids, and the monoterpene camphor, and the sesquiterpene (E)-β-caryophyllene at 4 h after the treatments (b). nd no detection

Heat stress-treated leaves repel the TCW-infestation
TCW preferred the control plants (Fig. 5a) and after 16 h feeding, TCW consumed a higher total leaf area in the control plants than in the plants under the 45 °C treatment (Fig. 5b). The percentage of total leaf area consumed (4.28% ± 0.77% for heat-stressed plants vs. 61.94 ± 2.61% for control plants) was also greater in control plants (P < 0.01).

Correlations between the contents of different terpenoids and the leaf area consumed by TCW
Pearson correlation analysis showed that the total area consumed by TCW scaled negatively with total sesquiterpene concentration (r = −0.988 for total area, r = −0.991 for percentage of total area, P < 0.01). In addition, the concentrations of 8 sesquiterpenes were negatively correlated with the area consumed by TCW larvae. Concentrations of β-copaene, γ-elemene, γ-muurolene, isogermacrene D, and germacrene D were most strongly correlated with feeding area (r ranging between −0.967 to −0.987, P < 0.01). Concentrations of three other sesquiterpenes ((E)-βcaryophyllene, (E)-β-farnesene, germacrene D-4-ol) were also negatively correlated with the feeding area (r ranging from 0.877 to −0.895, P < 0.05), whereas the correlation of α-patchoulene and consumed area was positive (r = 0.852, P < 0.05).

Discussion
Photosynthesis is one of the most sensitive physiological processes to heat stress (Allakhverdiev et al. 2008;Ashraf & Harris 2013;Hüve et al. 2019;Okereke et al. 2022). Several studies have demonstrated that heat stress reduces leaf photosynthetic activity due to inhibition of multiple ratelimiting processes and damage to key components of photosynthetic machinery (Salvucci & Crafts-Brandner 2004;Schrader et al. 2004;Sharkey 2005;Hüve et al. 2011;Ashraf & Harris 2013;Zhu et al. 2018). Typically, moderate heat stress results in readily reversible changes in photosynthetic activity, primarily due to the reductions in stomatal conductance (G s ) to water vapor (Hüve et al. 2019;Okereke et al. 2022). Moreover, severe heat stress leads to non-stomatal slowly reversible or non-reversible inhibition of photosynthesis due to damage of the photosynthetic machinery (Salvucci & Crafts-Brandner 2004;Schrader et al. 2004;Sharkey 2005;Hüve et al. 2011;Zhu et al. 2018). In our study, net assimilation rate (A n ), was not affected by plant exposure to 35 °C, although there was evidence of increasing G s after enclosure to leaf chamber (Fig. 1a, b). Such delayed responses of photosynthetic characteristics have been observed after heat exposure (Hüve et al. 2019), and are expected to reduce leaf carbon gain in fluctuating environments. Leaf exposure to 45 °C treatment resulted in a major reduction of A n without concomitant reductions in G s , indicating non-stomatal reduction of foliage photosynthetic activity (Fig. 1a, b) as was also confirmed by increased intercellular CO 2 concentration (Fig. 1c). Our results are in accordance with previous reports suggesting that severe heat stress leads to non-stomatal inhibition of photosynthesis, and In several species, the increasing temperature has been demonstrated to strongly promote the emission of various terpenes (Staudt & Bertin 1998;Loreto & Schnitzler 2010;Grote et al. 2013;Maja et al. 2016). To understand the impact of temperature on terpene emissions, it is important to consider that only some species are strong constitutive terpene emitters, and terpenes might be emitted right after their synthesis (de novo emissions) or from specialized (h), cis-α-bergamotene (i), α-farnesene (j), germacrene D-4-ol (k), and α-patchoulene (l) from control and 45 °C exposed C. nankingense leaves. The asterisks indicate significant differences: *P < 0.05, **P < 0.01 storage (storage emissions) (Grote et al. 2013;. Moderately high temperatures can enhance constitutively synthesized de novo terpene emissions primarily by reversibly enhancing the activity of terpene synthases and substrate availability (Niinemets et al. 2002;Grote et al. 2013). Both induction of emissions as well as enhancement of storage emissions have been observed in heat-shocked Solanum lycopersicum and Nicotiana tabacum (Copolovici et al. 2012;Pazouki et al. 2016;Turan et al. 2019). In our study, the volatile emission form C. nankingense leaves might come from both de novo synthesis and specialized storage due to the existence of small amount of glandular and non-glandular trichomes (Guo et al. 2020). no terpene emissions were observed in control plants at 25 °C, and the release of mono-and sesquiterpenes was initiated only after 3 h under 45 °C treatment (Fig. 2). Given that the dominant monoterpene induced by heat stress, camphor, was not found in leaf extracts (Fig. 4), we conclude that the release of monoterpenes reflects de novo induction of monoterpene synthesis in heat-stressed leaves. Given that sesquiterpenes were also observed in leaf extracts (Fig. 4), the heat stress-induced emissions of the sesquiterpene (E)-βcaryophyllene could be related to the breakage of glandular trichomes or increases permeability of storage cells. However, leaf extracts demonstrated the presence of 10 other sesquiterpenes, but only (E)-β-caryophyllene was observed in leaf emissions (Fig. 2). This suggests that the emitted (E)β-caryophyllene might also come from de novo synthesis as has been observed in response to different stresses (Hansen & Seufert 2003;Jiang et al. 2017;Kanagendran et al. 2018).
Analysis of leaf extracts further demonstrated that total sesquiterpene concentration, and the concentrations of most individual sesquiterpenes detected increased after plant exposure to 45 °C (Fig. 4). There is limited information of heat stress impact on terpene contents, and the results are controversial. In conifers Pinus sylvestris and Picea abies, foliage terpene concentrations were enhanced when growing at higher temperature (Sallas et al. 2003), but in the annual herb Artemisia annua terpenoid content was not affected by growth temperature (Daussy & Staudt 2020). In the conifer Pseudotsuga menziesii, growth under high temperature reduced foliage terpene concentrations in one provenance and did not affect the concentrations in another provenance (Duan et al. 2019). In the perennial herb Valeriana jatamansi grown in a Free Air Temperature Increase Experiment, the concentration of the sesquiterpene globulol was increased by high temperature, but not the concentration of other compounds (Kuandal et al. 2018). Clearly more work is needed to gain an insight into the impact of heat stress of foliage terpenoid contents.
Our study showed that the volatile terpenoids emitted from C. nankingense after 3 h exposure 45 °C increased plant attractiveness for TCW. Enhanced attractiveness of the plants to TCW could be mediated by the induced emissions of (E)-β-caryophyllene. However, (E)-β-caryophyllene is typically associated with attraction of pest and herbivore enemies rather than attraction of herbivores themselves. For example, (E)-β-caryophyllene emitted from maize (Zea mays) roots strongly attracted an entomopathogenic nematode of the pest Diabrotica virgifera virgifera (Rasmann et al. 2005;Köllner et al. 2008). In addition, Z. mays leaves infested by TCW emitted (E)-β-caryophyllene that strongly attracted a larval parasitoid Cotesia marginiventris ). On the other hand, it has been also observed that (E)-β-caryophyllene could attract the pest D. virgifera virgifera larvae to locate the maize plant (Robert et al. 2012). Moreover, (E)-β-caryophyllene could function as a ** 25 ℃ 45 ℃ a b Different temperature Fig. 5 Effect of heat stress on the feeding area on C. nankingense after 16 h of infestation by Spodoptera litura. a Representative images of infested leaves from a control and 45 °C-treated plant. b Average ± SE (n = 5) of leaf area consumed in control and 45 °C-treated plants. **P < 0.01 host location signal for the rice (Oryza sativa) pest Sogatella furcifera (Wang et al. 2015), and it also attracted the whitefly (Bemisia tabaci) to the leaves of rosemary (Rosmarinus officinalis) (Sadeh et al. 2017). Our results are consistent with the observations that (E)-β-caryophyllene could serve as host location signal for TCW.
The comparison of feeding area between control plants and 45 °C-treated plants suggested that elevated concentrations of sesquiterpenes could inhibit the infestation of TCW (Fig. 5). In particular, total sesquiterpene concentration and concentrations of γ-elemene, β-copaene, isogermacrene D, γ-muurolene, germacrene D, (E)-β-caryophyllene, (E)-βfarnesene, and germacrene D-4-ol were negatively correlated with the total leaf area consumed and the percentage of leaf area removed by TCW. Our result implies that the chemical response to heat stress in C. nankingense leaves leads to higher resistance to insect infestation. Anti-feedant activity and toxicity of sesquiterpenes to related herbivore Spodoptera littoralis larvae has been observed in multiple studies, including 38 sesquiterpenes extracted from leaves of different Celastraceae species (Gonalez et al. 1997), two eremophilane sesquiterpenes isolated from the forb Senecio adenotrichus (Ruiz-Vasquez et al. 2017), and terpenoids from the forb Origanum vulgaris (Agliassa & Maffei 2018). It has been further observed that among eight tropical woody species, the essential oil extracted from leaves of Piper pseudolanceifolium and Ocimum campechianum most strongly repelled the food grain pest Tribolium castaneum; in these species, germacrene D and (E)-β-caryophyllene were the dominant components of the essential oil (Caballero- Gallardo et al. 2014). In our study, the concentrations of these two sesquiterpenes increased after heat stress in C. nankingense leaves, and this was associated with reduced TCW feeding area, underscoring the high repellence of germacrene D and (E)-β-caryophyllene.
Our study indicates that (E)-β-caryophyllene can serve as host location cue for TCW. However, together with other sesquiterpenes, it also serves as feeding deterrent for TCW. However, sesquiterpene synthesis is elicited by multiple stresses, some of which can lead to a systemic defense response in all leaves, and some to localized response in only impacted leaves. In the case of a localized stress, such as herbivory on single leaves, attraction to (E)-β-caryophyllene could be advantageous as the herbivores might feed on as yet undamaged leaves. In the case of a systemic stress such as whole plant heat stress, however, all leaves are impacted and the (E)-β-caryophyllene-emitting plant will be a lower quality food source than a non-emitting plant.
In summary, our findings could provide new opportunities for pest management. Heat stress exposed C. nankingense will have elevated emissions of (E)-β-caryophyllene, and this could effectively attract TCW. This could indirectly protect the adjacent crops that cannot produce (E)-β-caryophyllene by reducing their herbivore infestation. Further studies should focus on the mechanisms underlying the terpene metabolism in C. nankingense by functional identification of the terpene synthase genes responding to high temperature. It is also necessary to explore the ecological function of individual terpenoids on the behavior of herbivores. In particular, electroantennogram (EAG) analyses would be very useful to understand the relationships between olfactory stimuli of individual terpenes released after heat stress and after herbivory and resultant behavioral responses of TCW.