Olfactory responses to volatile organic compounds and movement parameters of Philaenus spumarius and Cicadella viridis

Xylella fastidiosa, the causal agent of several diseases in crops of economic interest, could be rapidly transmitted and spread throughout the agroecosystem landscape by xylem sap‐feeding insects. Chemical signals in the environment drive the behaviour of insects. Attractive or repulsive responses to signals may affect insects’ fitness, survival and reproduction, and elicit different movement patterns. Specific movement patterns derived from olfactory cues towards the selection of plants for feeding may trigger the transmission of the pathogen by the insect. Thus, understanding vectors’ olfactory response and movement parameters are of utmost importance. This work aimed to assess the olfactory response of Philaenus spumarius and Cicadella viridis adults to two Volatile Organic Compounds (VOCs) (cis‐3‐hexenyl acetate and cis‐3‐hexen‐1‐ol) present in almond, olive and vine leaves. Insects’ behavioural responses were conducted in a four‐arm olfactometer, and the two aforementioned VOCs were tested at different concentrations (5, 10, 20 and 30 µg/µl). At the lowest concentration, females of P. spumarius were significantly attracted by the two VOCs. At the highest concentrations, no significant differences were detected among treatments. Cicadella viridis individuals showed no significant differences in their choice at any concentration. Additionally, the walking movement of these insects was also studied in the absence of food. Females of P. spumarius and C. viridis can walk significantly more at a significantly higher velocity than males. Knowledge of the olfactory response of X. fastidiosa vectors and their movement parameters can be essential to develop new tools to limit the spread of this phytopathogen.


| INTRODUC TI ON
Insect-borne plant pathogens are an increasing concern in global agriculture (Huang et al., 2020). They are causal agents of devastating diseases that pose imminent threats to the global economy, diversity conservation and public health (Anderson et al., 2004;Huang et al., 2020;Schneider et al., 2020;Tumber et al., 2014). Xylella fastidiosa (Wells et al., 1987) (Xanthomonadales: Xanthomonadaceae) is one of the most important and severe insect-borne plant pathogens described in Europe (Schneider et al., 2020). This phytopathogenic bacteria is a xylem-limited Gram-negative gammaproteobacterium, and it is the causal agent of insidious diseases in important economic tial spread and implementing measures to limit their damage to the ecosystem.
The behaviour of the insects is mainly mediated by naturally occurring semiochemical cues (Tumlinson, 2014). These cues are detected by olfactory sensilla located on the insect antennae (Bruce & Pickett, 2011). The stimuli may derive a positive behaviour towards a food source or a suitable oviposition site, or repulsive behaviours against predators, or toxic substances (Depetris-Chauvin et al., 2015). Volatile Organic Compounds (hereafter VOCs), emitted by plants, provide important cues driving the behaviour of many leafhoppers and froghoppers in their search for hosts (Ganassi et al., 2020;Germinara et al., 2017;Mazzoni et al., 2009;Riolo et al., 2012). Plants can produce VOCs in response to the attack of herbivorous insects, conferring protection to the plant by acting as signalling compounds at different trophic levels (Arimura et al., 2009;Dicke & Baldwin, 2010;Holopainen & Blande, 2013;Ye et al., 2018). For example, the cis-3-hexen-1-ol is responsible for the green aroma in the leaves and is one of the most important VOCs in plants (Cofer et al., 2018). It can directly affect the physiology and behaviour of herbivores through its attractive or repelling properties (Wei & Kang, 2011). The cis-3-hexen-1-ol can be esterified by alcohol acyltransferases yielding the ester cis-3-hexenyl acetate (Akacha & Gargouri, 2009;Ameye et al., 2018;Ozawa et al., 2013). Cis-3hexenyl acetate is a sweet, apple and banana-tasting compound that might act as an insect's infochemical and play an important role in triggering plant communication.
Identifying the VOCs emitted by plants, and disentangling their role on the response of insect vectors, can contribute to the implementation of approaches to manipulate the behaviour of these insects and implement sustainable control strategies such as the push-pull (Cook et al., 2007), lure-and-kill (El-Sayed et al., 2009) or attract-and-kill (Gregg et al., 2018).
These approaches are based on the use of specific VOCs to aggregate the target pest population in a specific location. Then, the individuals are subsequently removed or repelled from the main crop. Therefore, they also can lure the natural enemies of the pest, enhancing biological control. Indeed, several studies related that cis-3-hexenyl acetate mediated plant communication and can attract parasitoids when emitted from infested plants (Ameye et al., 2018;Cofer et al., 2018;Ozawa et al., 2013). Altogether, these suggest that these VOCs might play an essential role in host location and selection by inducing attraction or repulsion behaviour.
Movement is also fundamental to animal behaviour, it is responsible for governing how animals use habitats, social interaction, avoid predators, obtain food and even adapt to human-modified landscapes (Wilson et al., 2015). Therefore, insect movement is considered an element of survival (Dickinson et al., 2000) and the main way to spread insect-borne plant pathogens throughout the landscape (Finke, 2012).
The most common modes of locomotion of P. spumarius and C. viridis are flying, jumping and walking (Burrows, 2003(Burrows, ,2007. The movement parameters related to flying and jumping have been deeply studied (Burrows, 2003(Burrows, ,2007Clemente et al., 2017;Lago et al., 2021;Weaver & King, 1954); however, those related to walking were largely neglected.
Spittlebug adults are poor fliers, opting to walk or jump more often instead of flying . During horizontal movements, P. spumarius walk with the hind legs usually dragged (Burrows, 2003(Burrows, ,2006. On the contrary, C. viridis uses all the legs to walk (Burrows, 2007). Weaver and King (1954) observed that adults of P. spumarius travelled more than 30 m in a single flight and moved as much as 100 m within 1 day, whereas Bodino et al. (2020) found that individuals were able to disperse up to 400 m during the population peak in Italy. Nevertheless, Lago et al. (2021), using a flight mill, concluded that spittlebugs were able to move up to 500 m in a 30 min flight. This insect can also jump up to 70 cm above the ground with an acceleration of 400 m/s 2 (Burrows, 2003). Regarding C. viridis, adults move around the landscape, in response to visual or olfactory stimuli, by jumping or by performing short flights (Beok, 1972).
On average, C. viridis can perform jumps with a take-off velocity of 0.88 m/s and a constant acceleration of 152 m/s 2 (Bonsignori et al., 2013).
Philaenus spumarius and C. viridis are polyphagous insects that feed on a wide range of plant species (Bodino et al., 2020;Dmitriev & Nickel, 2006) due to the low nutritional value of their food (Ranieri et al., 2016). Once the insect acquires the bacteria, the transmission of X. fastidiosa to other plants is a fast process, and the inoculation can occur in 2 to 7 min after the first probe (Cornara et al., 2020). Therefore, understanding the olfactory response of both vectors to VOCs and the extent to which the pathogen can be spread throughout the landscape according to their spatial dynamics (e.g. velocity and distance moved) will help limit the spread or even prevent the diseases.
The main goal of this work was to characterize the volatile profile of the three most important crops in the Mediterranean region (olive, almond and vine) and to study the olfactory response of adults P. spumarius and C. viridis to two VOCs common to the three crops (cis-3-hexenyl acetate and cis-3-hexen-1-ol). Additionally, the movement parameters (1) distance moved, (2) mean velocity and (3) total time moving of both species' insects were also assessed.

| HS-SPME
For each plant species, approximately 4 g of healthy leaves were placed in 50 ml individual vials, with the leaf petiole protected with aluminium foil, to avoid the potential registration of volatiles produced after cutting the petiole. The vials were sealed with a polypropylene cap with a silicon septum. After sealing, 5 μl of 4-metil-2-pentanol (0.127 mg/ml) (Sigma Aldrich, USA) were added with a syringe to the vials as an internal standard. In a water bath, the volatiles were released at 40°C for 5 min. Then, a fibre coated with Divinylbenzene/Carboxen/Polydimethylsiloxane (DVB/CAR/ PDMS 50/30 µm) (Supelco, Bellefonte, USA), was exposed to the headspace for volatile adsorption. Vine leaves have a larger leaf area than olive and almond leaves. Therefore, there is a greater release of volatiles which could lead to sensor saturation. After some trial tests, the headspace of 3 min was the most adequate for the vine leaves and the headspace of 15 min for olive and almond leaves.
For each plant species, five replicates of HS-SPME analysis were performed.

| GC-MS conditions
Chromatographic analysis was performed using a Shimadzu GC-2010 Plus. The volatile compounds were eluted from the fibre by thermal desorption for 1 min in the injection port of the chromatography system (220°C). The fibre was maintained for another 10 min in the injector port for cleaning and conditioning for further analyses. The gas chromatographer used was a Shimadzu

| Multiple-choice olfactory response assays
A four-arm olfactometer ( Figure S1) was used to test the behavioural responses of P. spumarius and C. viridis adults to the VOCs cis-3-hexenyl acetate (Sigma Aldrich, Switzerland) and cis-3-hexen-1-ol selected (Sigma Aldrich, USA) (see section 3.1.). To avoid the rapid evaporation of these volatiles, they were dissolved in sunflower oil (VitaD'or, Portugal) and stored at −20°C until utilization.
Four different concentrations (5, 10, 20 and 30 µg/µl) of each volatile compound were prepared following Ganassi et al. (2020). For the test, each arm of the olfactometer arena was connected to a polypropylene tube (3 cm long and 1 cm in diameter) that worked as an odour source container. Each odour source container was connected to a gas washing bottle (250 ml) with activated charcoal (AppliChem, Panreac ITW©) dissolved in 100 ml of distilled water to purify and humidify the airflow supplied by pumps with an airflow of 12 cm 3 /min. Each volatile compound was applied (10 µg) separately to strips of filter paper (4 mm ×25 mm) placed in the odour source containers of two opposite olfactometer arms.
The remaining olfactometer arms operated as controls. A strip of filter paper soaked in 10 µg of sunflower oil was placed in one of them, whereas the opposite arm contained a blank strip of filter paper. The air streams containing the odours were directed towards a four-entry arena (25 cm × 25 cm). The arena was built with four 4 mm high overlapping acrylic layers. The bottom layer was a white opaline plate with a 25 × 25 cm, and the following two layers were white opaline triangles (12 cm × 12 cm × 17 cm) positioned in each corner of the bottom layer to create the arena for the insect.
The top layer consisted of a transparent plate (25 cm × 25 cm) with a central hole (4 mm in diameter) to allow insect insertion and airflow. Before the olfactometer assay, insects were sexed using a binocular stereoscopic microscope and kept individually in 2 ml Eppendorf tubes to be starved for 4 hr. After starvation, one insect each time was released into the central hole of the arena, and its behaviour was recorded for 20 min with Computar ® lens (H2Z0414C-MP, f = 4-8 mm, F 1.4, ½", CCTV lens) mounted on a Basler ® GigE HD Camera (acA1300-60gc with e2v EV76C560 CMOS sensor) (Noldus, 1991). The recording tool used was the Media Recorder 2.5 software (Noldus Media Recorder, 2013). The olfactometer was illuminated from below using 20 white cold LED lamps that provide uniform illumination (7500 ± 10 lux). A choice was considered valid if the individual accessed a designed area of 1.4 cm 2 around each air entry. Choice was codified through two behavioural parameters, (1) the frequency of visits (i.e. the number of times the insect accessed an area) and (2) total length of stay in each area (i.e. sum of the time in each area). For each VOC concentration tested, 30 females and 30 males of P. spumarius and C. viridis were used respectively. The assays were conducted in a climate chamber at 21°C and 70%RH. This temperature was chosen to increase the activity of the insects (formerly maintained at 18°C) and to avoid the rapid volatilization of VOCs. The olfactometer was rotated 180° after 15 tests to avoid directional bias.

| Estimation of movement parameters
To estimate the movement parameters of P. spumarius and C. viridis, a total of 30 males and 30 females, respectively, were recorded individually in an arena (27 cm × 35 cm × 43 cm) for 10 min with the same camera and recording tool described in section 2.3.
The arena was made of acrylic and illuminated from below (negative contrast) using a computer screen placed horizontally with a full white Microsoft PowerPoint slide loaded. Before each trial, the insects were maintained in the rearing plants and immediately transferred to the arena for the assay. The assays were conducted in a climate chamber at 21°C 70%RH and in the absence of olfactory stimuli.

| Volatile characterization
Analysis of variance (ANOVA) was performed using the software PAST v.4.03 (Hammer et al., 2001) to compare the volatile profile between the crops. Subsequent multiple-comparison post hoc tests were applied using the Tukey's test (α < 0.05). Additionally, a Principal Component Analysis (PCA) was conducted in R software v.3.5.1 (R Core Team, 2020) using the function pca from the 'FactoMineR' package (Le et al., 2008). The correlation biplot of the two first PCs was drawn using the fviz_pca_biplot function from the 'factoextra' package (Kassambara & Mundt, 2020).

| Multiple-choice olfactory response assays
For each species, the behaviour was analysed with the Noldus Observer XT 11.5 software (Noldus et al., 2001). Generalized Estimating Equations (α = 0.05) with Poisson distribution were used to compare the frequency of visits and the total length of stay of the individuals in each delimited area. Sex and choice (air, oil, cis-3-hexenyl acetate, and cis-3-hexen-1-ol) selected were considered response variables, and the interaction between the two terms was also used as explanatory variable. The frequency of visits and the total length of stay in each area was assessed and compared between the choices using post hoc Tukey's tests (α = 0.05) for males and females separately. Since all the insects exhibited activity, no individuals were excluded from the analysis. The model was developed in R using the geeglm function from packages 'geepack'.

| Estimation of movement parameters
The total distance moved (m), mean velocity (cm/s) and the total time moving (s) were estimated using the Noldus Ethovision XT 11.5 software (Noldus et al., 2001). Each movement parameter was compared between females and males of P. spumarius and C. viridis using a Student's t-test (α = 0.05) in PAST.

| Volatile characterization
In total, 83 compounds were identified from the three plant species (Table S1). The olive leaves presented the highest number of VOCs with 54 identified compounds, followed by almond and vine leaves with 40 and 16 identified compounds respectively.
According to the PCA and ANOVA analyses, the volatile profiles significantly differed between plant species ( Figure S2 and Table S1).
PC1 and PC2 explained 82.1% of the variation.
Therefore, these two compounds were selected for the olfactory assay due to their frequency and quantity.

| Multiple-choice olfactory response assays
At the lowest VOCs concentration (5 µg/µl), the females of P. spumarius were significantly attracted by the two volatile compounds (Table 1 and Figure 1). Females showed a significantly higher frequency of visits and a longer permanence in the areas with cis-3-hexenyl acetate and cis-3-hexen-1-ol only at the lowest concentration (Figure 2), whereas males of P. spumarius showed significantly higher frequency of visits and longer stay only in the area with cis-3-hexen-1-ol compared to oil.
The frequency of visits significantly differed between the sexes except when the individuals were exposed to the volatile compounds at 10 µg/µl (Table 1). For this concentration, the stream of purified air was significantly more visited by females, whereas no significant choice was observed in the case of males (Figure 1). Regarding the total length of stay in each area, it was not significantly influenced by treatment or sex for the highest concentrations (20 and 30 µg/µl) (Table 2). At these concentrations (20 and 30 µg/µl), no significant differences were detected in the choice of P. spumarius.
In the case of Cicadella viridis, neither the frequency of visits nor the total length of stay significantly differed between treatments for any concentration (Table 1 and Table 2). However, the frequency of visits to each treatment varied significantly between sex for all the concentrations (Table 1).

| Estimation of movement parameters
Females of P. spumarius were able to walk a total distance of 2.42 ± 0.20 m in 10 min, with a mean velocity of 0.43 ± 0.04 cm/s. Both movement parameters (total distance and mean velocity) were significantly higher compared with the males (p < .01) (Figure 3a1 and a2). Regarding the total time of moving, there were no significant differences between males and females (p = .81) (Figure 3a3).

Females of C. viridis walked with mean velocities significantly
higher (p < .01) than the males (Figure 3b1 and b2). In 10 min, females of C. viridis walked 3.13 ± 0.33 m with a mean velocity of 0.63 ± 0.07 cm/s. Females also were moving significantly more time than males (Figure 3b3).

| DISCUSS ION
This work evaluated the olfactory response of adults of P. spumarius and C. viridis to cis-3-hexen-1-ol, and cis-3-hexenyl acetate, the main volatiles emitted by the leaves of the almonds, olives and vines.
Also, movement parameters by walking of these individuals, without food stimulus, were analysed. The development of innovative tools and approaches, as an alternative to synthetic pesticides, for the environmentally friendly control of P. spumarius and other potential vectors is essential to limit the dissemination of X. fastidiosa.
Regardless of where the bacteria was acquired, once the individuals are infected, the transmission of X. fastidiosa to the crop plants may then occur (Cornara et al., 2017).
Crops of economic importance, such as almond orchards, vineyards and olive orchards, are key crops in the landscape across the Mediterranean basin (FAOSTAT, 2021). Our results showed that these crops have a very distinct volatile profile. In our work, the leaves of the olive tree, the main crop affected by X. fastidiosa in Europe , showed the highest number of VOCs compared to the other two plants. These results agree with Malheiro et al. (2016), which found cis-3-hexen-1-ol and cis-3-hexenyl acetate the main components in the olive tree leaves. We also found these compounds to be abundant in the almond and vine leaves.
Cis-3-hexen-1-ol and cis-3-hexenyl acetate are generated through the oxylipin pathway from C18-polyunsaturated fatty acids (FAs; αlinolenic acid (ALA) and linoleic acid (Kost & Heil, 2006). Both are naturally occurring VOCs in the plants and their occurrence at a high percentage is related to tissue damage (Ameye et al., 2018). Since we used intact and healthy leaves from field plants, the high level of cis-3-hexen-1-ol and cis-3-hexenyl acetate could be due to abiotic stress such as high temperature or plant water deficit. In fact, according to Sofo et al. (2004), the synthesis of these compounds is more actively observed in stressed plants under drought conditions. Also, Malheiro et al. (2016) suggested that higher temperatures can mediate the emission of the two VOCs in the absence of tissue damage.
Spittlebugs interact socially by vibrational signals (Avosani et al., 2020), and pheromones are not common. Recent studies described that spittlebugs have a wide range of responses to VOCs, despite having a low number of antennal sensory structures (Ranieri et al., 2016). This fact indicates that spittlebugs use this type of stimuli to choose and locate host plants. Germinara et al. (2017) suggested that both sexes of spittlebugs have a general similarity in antennal sensitivity; However, we found a positive attraction to cis-3- Several studies on the movement of spittlebugs and leafhoppers focused on jumping or flight performance (e.g. Beok, 1972;Bodino et al., 2021;Bonsignori et al., 2013;Burrows, 2003Burrows, , 2007Clemente et al., 2017;Goetzke et al., 2019;Lago et al., 2021). The body of P. spumarius and C. viridis is designed essentially for jumping, and they have long hind legs to increase leverage for jumping, wedge-shaped heads and rigid front wings that form a continuous smooth structure to reduce drag when jumping (Burrows, 2003(Burrows, , 2007. However, they can also move by walking. The larger body size and longer legs of C. viridis (Burrows, 2007;Cornara et al., 2018) may allow it walking faster and longer distances than P. spumarius, although further studies are needed to compare the performance between both species. Beok (1972) suggested that C. viridis movements are trivial and confined within the habitat and individuals have been observed to stay on the same spot for days. This agrees with the low activity we found for the species in this work. On the contrary, Philaenus spumarius has a migratory behaviour. Several authors described that P. spumarius begins his migratory journey at early summer, coinciding with the death of the vegetation cover, until September, returning to the vegetation cover to lay eggs (Antonatos et al., 2020;Cruaud et al., 2018;Morente et al., 2018).
We found the females of both species being significantly faster and travelling longer than males. Although we did not study the flight, our results in terms of velocity and distance travelled, agree with Lago et al. (2021), which found better female performance in terms of distance travelled and flight duration.

| CON CLUS IONS
In the present study, it was possible to verify that females of P. spumarius are attracted by the two VOCs under study, but only when they are in low concentration. Since the individuals of C. viridis did not present any significant choice, further tests on the sensilla and the electroantennographic responses may help better understand the role of olfactory cues in the selection of host plants by the species. The two VOCs studied occur naturally in olive, almond and vine, crops that can host X. fastidiosa. Females of P. spumarius can play a key role in disseminating X. fastidiosa due to their ability to walk longer distances with higher speeds than males. Further research on the olfactory response of the vectors of X. fastidiosa and their movement behaviour is essential to design new techniques to limit the spread of this pathogen throughout Europe.

ACK N OWLED G EM ENTS
The authors are grateful to the Foundation for Science and

CO N FLI C T O F I NTE R E S T
The authors declare no conflict of interest.