The Herbivore-Induced Plant Volatiles Methyl Salicylate and Menthol Positively affect Growth and Pathogenicity of Entomopathogenic Fungi

Some herbivore-induced-plant volatiles (HIPVs) compounds are vital for the functioning of an ecosystem, by triggering multi-trophic interactions for natural enemies, plants and herbivores. However, the effect of these chemicals, which play a crucial role in regulating the multi-trophic interactions between plant-herbivore-entomopathogenic fungi, is still unknown. To fill this scientific gap, we therefore investigated how these chemicals influence the entomopathogenic fungi growth and efficacy. In this study, Lipaphis erysimi induced Arabidopsis thaliana HIPVs were collected using headspace system and detected with GC-MS, and then analyzed the effects of these HIPVs chemicals on Lecanicillium lecanii strain V3450. We found that the HIPVs menthol and methyl salicylate at 1 and 10 nmol·ml−1 improved many performance aspects of the fungus, such as germination, sporulation, appressorial formation as well as its pathogenicity and virulence. These findings are not only important for understanding the multi-trophic interactions in an ecosystem, but also would contribute for developing new and easier procedures for conidial mass production as well as improve the pathogenicity and virulence of entomopathogenic fungi in biological pest management strategies.


Results
Headspace volatiles collected and detected by GC-MS. Nine HIPVs compounds notably limonene, 2-methyl-6-heptene, menthol, methyl salicylate, 1-octen-3-ol benzaldehyde, phenylacetaldehyde, decan-3-ol and terpineols were in total collected and identified from aphid-induced A. thaliana. Limonene and 2-methyl-6-heptene were detected in all the treatments (I, II, III, IV and V) i.e. independent of the different densities of L. erysimi apterous adults (1, 2, 5, 10 and 20) that infested A. thaliana plants, and with no significant differences (F 4, 10 = 1.79, P = 0.23; F 2, 4 = 3.99, P = 0.31) between treatments as regards to the quantities emitted (Table 1). Similarly, Menthol and 1-octen-3-ol were also collected from all treatments, but the quantities of Menthol produced in treatments III-V were significantly higher than those emitted in treatments I -II (F 4, 10 = 50.89, P = 0.04); whereas the quantity of 1-octen-3-ol in treatment III was significantly higher than all the rest of the treatments (F 4, 10 = 5.38, P = 0.009) ( Table 1). Methyl salicylate was absent in treatment I but detected in all the other treatments II-V, and with the lowest quantity collected in treatment I compared to treatments II -V (F 3, 8 = 30.74, P = 0.01) ( Table 1). Likewise, benzaldehyde, phenylacetaldehyde and decan-3-ol were only detected in treatment III-V and completely absent in I and II. However, the benzaldehyde quantities collected from the three treatments were not significantly different (F 2, 6 = 28.27, P = 0.1). The phenylacetaldehyde quantity was much higher in treatment V, compared to the other two treatments (F 2, 6 = 84.81, P = 0.001), while the quantity of decan-3-ol emitted was significantly lower in treatment IV compared to treatments III and V (F 2, 6 = 64.08, P = 0.03). In addition, Terpineols were only detected in treatment V ( Table 1).
Influence of aphid density on conidial germination and appressorial formation rate. The conidial germination and appressorial formation rate after exposure of L. lecanii strain V3450 to different concentrations of HIPVs induced by different densities' number of aphids varied greatly after 12 h exposure (Table 2). For instance, the germination rate of conidia in treatments IV and V were significantly high (F 4, 4 = 6.06; P < 0.05) compared to treatments I (control) ( In addition, the rates have significantly increased (P < 0.0001) in treatment III and V compared with control (Table 2).

Correlation analysis for HIPVs chemical compounds and conidial performance. The correlation
between HIPVs chemicals and conidial performance of L. lecanii showed that the germination and appressorial formation rates were influenced by multiple HIPVs compounds as summarized in Table 3. For conidial germination, the correlation coefficient for menthol was higher than 0.5, indicating a significant correlation trend. Similarly, for appressorial formation, the correlation coefficients for menthol, methyl salicylate (MeSA), benzaldehyde and phenylacetaldehyde were all higher than 0.5, indicating a significantly strong correlation between these HIPVs compounds and the conidial performance (Table 3).
Hyphal growth. Impacts of synthetic chemicals on colony extension of L. lecanii after 15 days are summarized (Table 4). Growth speed of the fungal colony was faster when treated with 1 and 10 nmol·ml −1 menthol, and 1 nmol methyl salicylate (0.89, 1.09 and 0.92 cm·day −1 , respectively), as compared to the control treatment (0 nmol·ml −1 treatment, 0.74 cm·day −1 ). Similarly, the diameter of the colony growth was also observed to be significantly larger after exposure to 1 and 10 nmol·ml −1 menthol (F 4, 20 = 162.33, P = 0.0002), as well as 1 nmol·ml −1 methyl salicylate (F 4, 20 = 34.83, P = 0.012) compared to the control, respectively (Table 4). However, the other chemical treatments (decan-3-ol, benzaldehyde, and phenylacetaldehyde) not only had a negative impact on the diameter of the fungal colony but also on the fungal growth speed (Table 4).
Pathogenicity of Lecanicillium lecanii. The pathogenicity of L. lecanii to the adult aphids after being pre-treated with the chemicals was assessed (Table 5). Lecanicillium lecanii fungus exposed to menthol had a lowest LT 50 (4.57 d) at the concentration of 10 nmol·ml −1 after 15 days post-exposure and with percent mortality significantly higher after 7 days compared to other chemical treatments as well as the control (F 4, 20 = 19.36, P = 0.0021) ( Table 5). A similar trend was observed for 1 nmol·ml −1 methyl salicylate treated fungus (F 4, 20 = 24.34, P = 0.0001). For the other chemicals tested, the lethal time of the fungus as well as percent mortality was similar compared to the control (Table 5).

Discussion
From the volatiles trapping, nine different compounds were emitted and identified as HIPVs when A. thaliana was infested with L. erysimi. Limonene, 2-methyl-6-heptene, menthol and 1-octen-3-ol were detected in all the aphid-induced A. thaliana plants. However, equal quantities of Limonene and 2-methyl-6-heptene were emitted independent of the aphid densities, while high quantities of menthol and 1-octen-3-ol were produced where aphid densities were increased (from 5 to 20 aphids). These results indicate that the quantities of some HIPVs (aphid-induced A. thaliana) compounds production depend on the densities of the herbivores. In addtition, our results indicated that, benzaldehyde, phenylacetaldehyde and decan-3-ol were produced only if the densities of aphids were between 5 and 20 adult aphids, while Terpineols were produced only when de density reached 20 aphids. This shows that some HIPVs could only be emitted if the pest population is high on the host plants as observed for Terpineols in aphid-induced A. thaliana. Furthermore, decan-3-ol, benzaldehyde, phenylacetaldehyde and salicylate acid were recorded in the headspace when A. thaliana was damaged by L. erysimi, and the quantity of menthol emitted from infested plants was significantly higher than control and the other HIPVs. Since insect herbivores are the key factor that induces HIPVs, it could be deduced from these results that the quantity and the composition of the emitted HIPVs might be influenced by the density of the pest population present at a particular time and stage of the host plants. Similar results were reported by Rodriguez-Saona, et al. 13 through a gradual perennial shrub Vaccinium corymbosum infestation by gypsy moth caterpillars Lymantria dispar over time. The authors indicated that, the plant VOC emission rate also increased concomitantly as regards to the level of infestations. Our results also showed similar effects where the composition in terms of the number of HIPVs emitted from aphid-induced A. thaliana and their quantities increased with the subsequent increase in the number of aphids on the A. thaliana plant. In our previous finding and in this present study, HIPVs emitted from aphids-induced Arabidopsis increased performance of L. lecanii, and the rates of conidial germination as well as appressorial formation varied depending on the exposed aphid densities 12 . The results of this study showed that, decan-3-ol, benzaldehyde, phenylacetaldehyde, salicylate acid and menthol were considered as key synergetic compounds that enhanced conidial performance and consequently, improved the pathogenicity and virulence of L. lecanii strain V3450. Similar results were also reported by Lin, et al. 12 using the same host plant and different aphid densities. In addition, we collected volatiles from 20 adult aphids in the glass jar of diagrammatic sketch of headspace collecting system, and analyzed the volatiles profile with GC-MS in a supplementary experiment (Supplementary experiment 3). The results showed that there were not menthol, methyl salicylate, decan-3-ol, benzaldehyde and phenylacetaldehyde in the headspace volatiles from aphids ( Supplementary Fig. 1). These results suggested that the above mentioned chemical compounds (benzaldehyde, menthol, phenylacetaldehyde, decan-3-ol and methyl salicylate) found in the headspace volatiles were not emitted by L. erysimi but rather produced by the L. erysimi-induced Arabidopsis host plants.
When applied to L. lecanii, the five identified compounds produced different influences as regards to the level of fungal performance vis-a-vis aphid densities. Methyl salicylate and menthol did not provoke significant  changes to the germination of the fungus, while decan-3-ol, benzaldehyde, phenylacetaldehyde inhibited the conidial germination. But when focusing on the appressorium formation, methyl salicylate (1 nmol·ml −1 ) and menthol (1 and 10 nmol·ml −1 ) showed their inducing function on L. lecanii performance, while the other compounds showed their inhibition function vis-a-vis to the fungus. This finding is significant because the formation of the appressoria is a precursor to the infection process of pathogenic fungi, which are swollen, dome-shaped cells 12,14,15 . These two compounds, menthol and methyl salicylate might therefore contribute to enhancing the pathogenicity and virulence of the entomopathogenic fungus L. lecanii with a synergetic effect. These results are similar to the previous research, which reported that salicylate acid indirectly promoted arbuscular mycorrhizal fungus in clover roots 16 . Spence, et al. 17 also found that appressorial formation in Magnaporthe oryzae infecting rice was negatively impacted by hydrogen cyanide, a volatile produced by rhizospheric bacteria, had suppressed the rice blast infections.
In addition, when also applying the five identified compounds to the culture of L. lecanii, we found that methyl salicylate (10 nmol·ml −1 ) and menthol (1 and 1 nmol·ml −1 ) promoted hyphal growth and the toxicity of the entomopathogenic fungus increased significantly, while other compounds inhibited significantly the growth speed (sporulation) and consequently the pathogenicity of the fungus. Similarly, it was also found that methyl salicylate (10 nmol·ml −1 ) and menthol (10 nmol·ml −1 ) promoted sporulation significantly, while others inhibited significantly the fungal sporulation. These results indicated that methyl salicylate and menthol may increase or enhance the toxicity by up regulating the appressorial formation, hyphal growth and sporlulation of L. lecanii.
Entomopathogenic fungi (EPFs) are satisfactory candidates which have the potential to replace synthetic broad spectrum insecticides without having a major impact on the environment 18,19 . EPFs showed their potential and prospects of controlling insect vectors [20][21][22] , pests [23][24][25][26][27] under laboratory and field conditions through inundative and autodissemination devices applications. Some of them, such as Metarhizium anisopliae and Beauveria bassiana had been used in the field with significant effects on their target pests 28,29 . EPFs infect insect pests and are not harmful to other participants of the particular agroecosystem. However, these EPFs encountered low pathogenicity, conidia viability and instability due to the effect of UVB irradiance, and as consequence, there are only few EPF products being used in the field 30,31 . In this regards, to overcome these challenges, many methods of molecular biology 32,33 , chemical biology [34][35][36] , and biophysics 37,38 have been investigated to increase the use of EPFs under adverse field conditions and some studies look promising 39 . Therefore, our findings in this study are important for understanding the multi-trophic interactions in ecosystems as well as for developing new and easier means for increasing conidial production and improving the pathogenicity of EPFs. Further molecular studies are warranted to determine, which gene locus or loci is responsible for the production of these chemicals. Once identified, it might possibly be overexpressed as the gene segment to increase the efficacy of L. lecanii or promote its use in other EPFs. Headspace collection and volatiles analysis. The aphids-induced-Aradopsis plant volatiles were collected in the first 12 h using a headspace collecting system as shown in Fig. 3. According to the supplementary experiment 1, different densities of L. erysimi apterous adults (1, 2, 5, 10 and 20 which corresponded to five different defined treatments I, II, III, IV and V respectively) were released on A. thaliana for infestation for one hour. And then, each four infested plants with the same density of L. erysimi were transferred into a clean glass jar (12 cm diameter, 30 cm high). The air was blown into the jar at the rate of 300 ml·min −1 and flowed out of the headspace through a glass tube filled with 200 mg Tenax TA (60/80 mesh; Grace-Alltech, Deerfield, USA). After collection, the HIPVs were eluted into 1 ml methenyl trichloride and stored at − 20 °C immediately. Four clean plants without aphid infestation were used as a control, and the experiment was repeated 3 times.

Methods
HIPVs were analyzed with a gas chromatograph (Agilent Technologies 7890B GC System)-mass spectrometer (Agilent Technologies 5977 A MSD) (GC-MS) via a HP-5 column (30 mm × 0.25 mm/1.0 μ m film thickness, Agilent). The GC oven temperature was programmed from 40 °C (5-min hold), which was increased at a rate of 10 °C·min −1 regularly until it reached a temperature of 280 °C. The column effluent was ionized by electron impact ionization at 70 eV. Mass spectra were acquired by scanning from 35 to 350 m/z with the scanning rate of 5.38 scans·sec −1 .
Compounds were identified by using the deconvolution software AMDIS (version 2.64, NIST, USA) in combination with NIST 05 and Wiley 7th edition spectral libraries and by comparing their retention indices with those Scientific RepoRts | 7:40494 | DOI: 10.1038/srep40494 from the literature. Quantities of headspace compounds were also compared with a standard sample quantity and by the detected peak area from GC-MS results.
Influence of aphid density on conidial germination and appressorial formation. HIPVs were collected as described above based on the various aphid densities on the host plant. The air from the headspace was then allowed to flow into another jar (incubator jar) instead of absorbent column. There was 10 ml sterile water in the incubator jar to keep humidity for conidial germination. A concave slide was plastered onto the top of the incubator jar. Ten microliter of L. lecanii conidial suspension with a concentration of 1 × 10 8 conidia·ml −1 was printed on the wall of the concave slide. Both conidial germination and appressorial formation of L. lecanii were determined after 12 h under a compound light microscope at 400X magnification for all the aphid density levels. The criterion used to assess conidial germination was that the length of germ tube > 50% the length of the conidia 12 . The treatment of 0 or no L. lecanii adults was used as control. Each test was conducted four times with 5 replicates.

Efficiency of HIPVs compounds on fungal germination and appressorial formation. The results
of GC-MS test and bioassay for conidial performance were analyzed by SPSS software 21.0 version to determine the correlation between single HIPVs chemical and germination rate or appressorial formation rate. Major compounds menthol, methyl salicylate, decan-3-ol, benzaldehyde, and phenylacetaldehyde identified in HIPVs of A. thaliana induced plant by the aphids revealed significantly high correlation with appressorial formation rate when different aphids' densities were used as treatments (Table 3). Prior to the experiment, synthetic of similar identified compounds were dissolved in triethyl citrate (TEC) to achieve slow volatilization as described in Uefune, et al. 43 . A supplementary experiment was conducted to determine the experimental quantities of synthetic chemical (Supplementary experiment 2). All the synthetic chemicals previously listed above were purchased from Macklin Biochemical Co. Ltd. (Shanghai, China).
To assess germination and appressorial formation of conidia, a 2.5 μ l conidial suspension (1 × 10 8 conidia·ml −1 ) was prepared in Czapek's liquid medium and placed onto the wall of a concave glass slide, and allowed to dry for 5 min 44 . A sterilized cotton roll was put into the jar of headspace collecting system (Fig. 3). One milliliter of each synthetic compound solution was applied on the cotton ball. The concentrations of menthol and methyl salicylate (MeSA) were 1, 10, 100 and 1000 nmol·ml −1 TEC, respectively; while the ones of decan-3-ol, benzaldehyde, and phenylacetaldehyde were 0.2, 2, 20 and 200 nmol·ml −1 TEC, respectively (Supplementary Table 2). The air from the headspace was then allowed to flow into another jar (incubator jar) instead of absorbent column as described above in the aphid density section. There was 10 ml sterile water in the incubator jar to keep humidity for conidial germination. A concave slides were plastered onto the top of the incubator jar. Ten microliter of L. lecanii conidial suspension with the concentration of 1 × 10 8 conidia·ml −1 was printed on the wall of the concave slide. Both conidial germination and appressorial formation of L. lecanii were determined after 12 h under a compound light microscope at 400X magnification for all the defined concentrations. The criterion used to assess conidial germination was the same as describes above where the length of germ tube > 50% the length of the conidia 12 . Each test was conducted four times with 5 replicates per synthetic compound.
Activity of synthetic compounds on fungal sporulation and growth. The fungus L. lecanii V3450 was cultured on Czapek's solid medium and incubated at 25 °C, 75% RH and a photoperiod of 12:12 L:D for 14 days. Conidia were harvested by scraping the surface of sporulating cultures with an inoculating loop. Conidial suspension adjusted at the concentration of 1 × 10 8 conidia·ml −1 was prepared with sterile water containing 0.05% Triton X-80 45 . Ten microliters of the fungal suspension were transferred to the center of round cellophane membranes (80 mm diameter) which covered the surface of the Czapek's medium in a Petri dish (90 mm diameter). Sterilized tiny cotton wool ball was placed on the bottom of the Petri dish which was inoculated with L. lecanii V3450, and then amended with 1 ml of serial dilutions of each synthetic compound solution (1, 10, 100 and 1000 nmol·ml −1 TEC for menthol and MeSA; 0.2, 2, 20 and 200 nmol·ml −1 TEC for decan-3-ol, benzaldehyde, and phenylacetaldehyde). All the dishes were sealed tightly with Parafilm and kept in the incubator at 25 °C, at 75% RH, and with 12:12 L:D photoperiod. The range of dilutions used for the bioassay was based on the pre-experimental outputs mentioned above (Supplementary Table 2). Control petri dishes were treated with 0.1% Tween X-80, and each dilution was repeated 5 times for each synthetic compounds.
To access the growth rate, diameter of colony in each Petri dish was determined by vernier caliper every 24 h, and the diameters were measured for 14 days 46 .
To quantify conidial production, colonies and cellophane membranes were transferred into a 50 ml-centrifuge tube containing 10 ml sterile solution of 0.1% Tween-80 and filtered by three layers of sterile gauze to make conidial suspensions after 14-days incubation 47 . The concentration of suspensions was determined by vortexing for about 5 min to produce homogenous conidial suspensions. Conidial counts were then made using an improved Neubauer Hemacytometer to determine the right concentration of conidia.
In vivo pathogenicity and virulence determination. The pathogenicity and virulence of L. lecanii to the adult aphids were assessed after the fungus was pre-treated with the HIPVs synthetic chemicals. Conidia from multi treated by synthetic chemicals was collected from previous experiment and made into conidial suspensions with distilled water respectively. Conidial suspensions were diluted to 5 × 10 7 conidia·ml −1 with sterilized solution of 0.1% Tween-80 for infection bioassays. Fifty apterous adult L. erysimi per treatment were immersed in conidial suspension for 30 sec in a 50 ml beaker to inoculate them with fungal conidia. After inoculation, all L. erysimi were air dried for 5 min and transferred to cauliflower leaves (5 × 5 cm) with wetted cotton to keep them fresh 48 .
Each treatment in this experiment was repeated three times, while the control were treated with sterilized solution of 0.1% Tween X-80. The lethal time (LT 50 ) and mortality were assessed where LT 50 was analyzed by linear regression. The whole experiment was conducted in an illumination incubator under constant conditions of 25 °C, 75% RH, and 12:12 L:D photoperiod.
Statistical analysis. Mortality of L. erysimi in treated groups was corrected by that observed in controlled conditions following the method described by Jones, et al. 49 . The median lethal time (LT 50 ) was estimated using probit analysis in the statistical software SPSS 20.0. The means of compound quantities, fungal colony growth, conidial production, corrected mortality, germination rate and appressorial formation rate were compared by Tukey's HSD test (α = 0.05) following an analysis of variance using the SPSS 20.0.