Entomologia Generalis, Vol. 40 (2020), Issue 1, 1–13 Article Published online 27 December 2019

Analysis of vector behavior as a tool to predict Xylella fastidiosa patterns of spread

Anna Markheiser1, Daniele Cornara2, Alberto Fereres2, and Michael Maixner1

1 Institute for Plant Protection in Fruit Crops and Viticulture, Julius Kühn-Institut (JKI) – Federal Research Centre for Cultivated Plants, Geilweilerhof, 76833 Siebeldingen, Germany. 2 Instituto de Ciencias Agrarias, Consejo Superior de Investigaciones Cientificas (ICA-CSIC), Calle Serrano 115 b, 28006 Madrid, Spain. * Corresponding author: [email protected]

With 5 figures and 2 tables

Abstract: The most likely scenarios for Xylella fastidiosa introduction in Central Europe is through infected ornamental plants, with a successive spillover from gardens and parks to cultivated orchards. Given its polyphagy and wide distribu- tion, the meadow spittlebug Philaenus spumarius, the only ascertained vector of X. fastidiosa in Europe so far, might play an important role in such a scenario. Here, we combined and analyzed spittlebug’s behavioral data obtained through Electrical Penetration Graph (EPG), preference and survival tests as well as field surveys, in order to infer possible bacte- rium patterns of spread. For our case study, we selected oleander and rosemary as potential introductory hosts and grape- vine and cherry as economically important threatened plants. Philaenus spumarius was collected in field near all the four plant species, although choice and no-choice tests indicated that the spittlebug rather prefers to settle on cherry and grape- vine than on rosemary and oleander. Considering the results of the EPG, the duration of xylem sap ingestion was longer in cherry, grapevine and rosemary than in oleander. However, P. spumarius spent on rosemary most of the time in resting activities, this implying a lower duration of xylem sap ingestion compared to grapevine and cherry. Overall, our data sug- gest that cultivated plants as grapevine and cherry could be more relevant than oleander and rosemary as X. fastidiosa source plants; therefore, P. spumarius might acquire the bacterium from cultivated plants, then first spread it within culti- vated orchards, and successively to ornamental plants during its dispersal.

Keywords: Philaenus spumarius, spittlebugs, EPG, Electrical Penetration Graph, host plants, feeding behavior, choice, preference

1 Introduction Aphrophoridae) in Europe. This species has recently raised from its condition of almost a naturalist’s curiosity to being New trade routes among previously disconnected coun- considered one of the most relevant threatening pest in Europe tries as well as novel transportation technologies have due to its role in the transmission of Xylella fastidiosa Wells increased the risk of new disease outbreaks worldwide (1987) (Proteobacteria: Xanthomonadaceae) (Cornara et al. (Hulme 2009). In addition, extensive land use and climate 2017a; Saponari et al. 2017; Cornara et al. 2018a; Cornara change facilitate the spread of invasive species from local et al. 2019). Bacterium introduction and establishment in to continental level (Benning et al. 2002; Dukes & Mooney Europe was first confirmed in 2013, when it was found to 1999). Once an -borne pathogen is introduced into a be the causal agent of a disease called “Olive Quick Decline new environment, its establishment and spread rely on the Syndrome” (OQDS), which leads to severe economic losses combination of at least two factors: suitable climatic con- in olive production in Southern Italy (Martelli et al. 2016; ditions and presence of an efficient vector (Fereres 2015; Saponari et al. 2013). Studies conducted in the affected area Sicard et al. 2018). Hemipterans are by far the most impor- showed that the local strain of X. fastidiosa (subsp. pauca, tant vectors of plant-pathogens (Nault 1997). In some cases, strain ST53 also named “CoDiRO” strain) is able to infect their role as agricultural pests depends only on the ability almond, cherry, oleander and other ornamentals beside to transmit plant pathogens, as in the case of the Meadow olive and various wild plant species (Cariddi et al. 2014; Spittlebug, Philaenus spumarius L. (1758) (Cercopoidea: Cornara et al. 2017b; Saponari et al. 2013; 2014). Since this

© 2019 E. Schweizerbart’sche Verlagsbuchhandlung, 70176 Stuttgart, Germany www.schweizerbart.de DOI: 10.1127/entomologia/2019/0841 2 Anna Markheiser et al. first report of X. fastidiosa, the bacterium was subsequently Nevertheless, once X. fastidiosa is introduced into a new detected in France (2015: Mainland, Corsica), Spain (2016: area, its eradication is considered barely feasible (Almeida Mainland; 2017: Balearic Island) and Germany (2016: et al. 2005). Consequently, proactive research on essential Saxony) (EPPO 2018). epidemiological aspects such as vector biology, ecology and Regarding the German detection, a potted oleander behavior is of paramount importance to understand possible plant held by a private person in a greenhouse of a nurs- pathogen patterns of spread and to develop methods to coun- ery for hibernation was infected with X. fastidiosa subsp. teract pathogen spread in case of new introductions (Rathé fastidiosa ST1 (JKI 2016). Within the demarcated area, X. et al. 2012; Blua & Morgan 2003; Purcell et al. 1994). Vector fastidiosa was also detected in single plants of Rosmarinus behavior, often disregarded in pathogen transmission stud- sp., Streptocarpus hybrid and Erysimum hybrid (JKI 2018). ies, is crucial for disease epidemiology. Indeed, pathogen Strains belonging to the subspecies fastidiosa are known to spread depends among other things on vector selection of cause Pierce’s Disease (PD) of grapevine (Vitis vinifera) and the host plant, preferred plant parts for settlement, probing thus pose a high risk for local viticulture. Within Europe, and feeding behavior, fitness and dispersal (Daugherty et al. X. fastidiosa was detected in grapevine for the first time in 2011; Del Cid et al. 2018; Fereres & Moreno 2009; Lopes 2017 (Mallorca, Balearic Island, Spain) (Olmo et al., 2017). et al. 2010). In this study, the behavior of a possible vector as Previous findings, which misleadingly confirmed the intro- P. spumarius pivotal for X. fastidiosa transmission was ana- duction of the grapevine-associated subspecies into Europe lyzed, i.e. probing and feeding behavior, host plant selection (from Kosovo and France), were based on visual PD symp- and survival on the host plant. Such analyses were conducted toms (Janse & Obradovic 2010). Worldwide, 19 other plant on plants that were considered important for bacterium intro- species were found to be naturally infected with strains duction and subsequent spread in Germany, namely olean- matching ST1, among them Prunus avium (cherry), Prunus der (Nerium oleander L.), rosemary (Rosmarinus officinalis dulcis (almond) and Medicago sativa (alfalfa) (EFSA 2018; L.), grapevine (Vitis vinifera L.) and cherry (Prunus avium Olmo et al. 2017). Due to its susceptibility to cold temper- L.). Oleander and rosemary are X. fastidiosa host plants, atures (Burbank & Stenger 2016; Feil & Purcell 2001), it irrespective of the subspecies (EFSA 2018) and considered has been assumed that X. fastidiosa is not able to survive as potential introductory hosts in this study. Cherry, host of winter under Central European conditions in herbaceous three subspecies of X. fastidiosa ( fastidiosa, multiplex and plants or shrubs (EFSA 2015). However, it is not unlikely pauca), and grapevine, host of subsp. fastidiosa (EC 2018; that the bacterium could overcome cold periods in the trunks EFSA 2018), were assumed as highly endangered in case of or roots of woody plants, since infection has been reported a further spread of the bacterium within Europe. from trees of cooler climate regions like Canada and New Philaenus spumarius probing and feeding behavior on Jersey (EFSA 2015; Goodwin & Zhang 1997). Furthermore, these four different host plants of X. fastidiosa was analyzed climate change might expand the geographic range suitable using the Electrical Penetration Graph (EPG) technique and for the fastidious bacterium, with possible unexpected out- obtained data were combined with observations on host pro- breaks (Godefroid et al. 2018). Given these premises, the pensity as well as survival on different host plants. The final most likely scenario for the establishment of X. fastidosa in objective of this study is to infer, through the analysis of the the Central European environment is introduction through behavioral data collected, possible P. spumarius-mediated infected ornamentals or propagation material (EFSA 2015) patterns of spread of X. fastidiosa in case of bacterium intro- and subsequent transmission to woody plants with massive duction and establishment in Germany, and in general in trunks as non-transient hosts, such as fruit and forest trees of Central Europe. susceptible genera. Philaenus spumarius is polyphagous and widespread in Europe; currently it is assumed to play a role in all the 2 Material and methods European outbreaks detected so far (Cornara et al. 2018a; Cornara et al. 2019; Cruaud et al. 2018; Morente et al. 2.1 Plant material 2018). Furthermore, the spittlebug is present and abundant in Potted plants, propagated at Julius Kühn-Institut (JKI), Germany either on indigenous and ornamental plants or on Institute for Plant Protection in Fruit Crops and Viticulture economically important crops such as grapevine and cherry, (Siebeldingen, Germany), were used for the behavioral stud- also threatened by the bacterium (Markheiser & Maixner ies. Plants were cultivated in 1 L pods in a glasshouse at 2017). 25±5°C, 20-40% relative humidity and a photoperiod of The ability of P. spumarius to transmit X. fastidiosa to 16:8 h light/dark with a minimum of 2.500 lux light intensity grapevine was demonstrated under laboratory conditions and supplementing artificial light if necessary. Cherry plants (Cornara et al. 2016; Severin 1950). However, neither poten- were, due to their extended root growth, cultured in 2 L pots. tial vector species nor further plants were positive for X. Substrate was a mixture of garden soil (Fruhstorfer Erde – fastidiosa after the first detection in Germany, and the bac- Typ ‘Tray Substrat + Perlite’, Hawita Gruppe GmbH, Vechta, terium is presently considered as “eradicated” (JKI 2018). Germany) and sand (4:1). Plants were irrigated three times a Vector behavior to predict Xylella fastidiosa patterns of spread 3 week via an ebb-flow-system, including a 0.2% solution of lux light intensity, 23:18°C and 70% relative humidity. The fertilizer twice a week (Hakaphos Soft Elite 24+6+12(+2), mortality of P. spumarius on the respective plants was deter- Compo Expert, Krefeld, Germany). mined over a period of 40 days at intervals of 2-3 days; the Vitis vinifera cv. Cabernet sauvignon were propagated as plants were watered three times a week. The soil in the pots dormant cuttings from a grapevine assortment located at JKI, was covered by rubber foam to enable an easier detection Institute for grapevine breeding (Siebeldingen, Germany). of dead individuals, which were removed from the cage to Developing inflorescences were removed to promote shoot assess the sex. growth. Prunus avium plants were prepared by grafting a bud of scion cv. Hedelfinger on an annual rootstock cv. 2.4 Choice experiment Gisela 5. Rosmarinus officinalis and N. oleander plants were The preference of P. spumarius for a specific host plant spe- propagated from green cuttings collected in a private garden, cies was studied in a four-choice experiment using landing the variety is unknown. All plants were potted 12-16 weeks and settlement on a plant as proxies for plant choice. The prior to the beginning of the experiments and not treated different plant species were introduced into a ventilated with pesticides to prevent potential repellent side effects of cage (90 × 60 × 60 cm, Caterpillar Castle, Live Monarch, the plant protection products on P. spumarius. Blairsville, Georgia, USA) and placed near the center, 20 cm apart from each other. Four cages were provided within one 2.2 Insect rearing walk-in climatic chamber. The plants were offered in a rotat- For EPG studies, nymphs of P. spumarius were collected ing system between the cages to avoid a possible exposition- from a X. fastidiosa-free area (Sierra de Aracena, Huelva, effect on the preference evaluation. The experiment was set Spain) on Sonchus sp. L., Cirsium sp., Borago officinalis L., up at the same climatic conditions described for the no-choice Calendula sp. L., Scolymus hispanicus L., and transferred experiment above to avoid any interference of external fac- to the laboratory to potted Sonchus oleraceus plants (age: tors. Single females and males, which were tested separately, 4 weeks). Individuals were kept under controlled climatic were transferred into a 2 ml vial and stored for 1 hour before conditions (14:10 h light/dark photoperiod, 23:18°C and introducing them into the cage. They were placed on the bot- 60% relative humidity) at Instituto de Ciencias Agrarias- tom in the center of the cage to ensure a uniform distance to Consejo Superior de Investigaciones Cientificas (ICA-CSIC, each plant. 24 repetitions were carried out for each sex. The Madrid, Spain) until adulthood. Potted plants were enclosed position of P. spumarius within the cage, the selected host in groups of five into a ventilated cage and humidified with a plant as well as the host plant organ was recorded every 15 nebulizer once per day. Newly molted adults (age: 1–3 days minutes during the first hour and hourly for the seven fol- after adult molt) were separated and kept in groups of 10 on lowing hours. new S. oleraceus plants, which were covered by a ventilated, cylindrical cage. 2.5 Electrical penetration graph (EPG) recording Further behavioral studies described below were per- The feeding behavior of adult P. spumarius (age: 2-4 days formed with adult P. spumarius collected by sweep net in a after adult molt) on uninfected X. fastidiosa host plants nutrition-poor grassland with a high diversity of herbaceous (P. avium, V. vinifera, R. officinalis and N. oleander) was plants (Bernkastel-Kues, Germany) Specimens were accli- explored by EPG, following the procedure described in matized in a 33 × 33 × 60 cm gauze cage (Live Monarch, Cornara et al. (2018b). Four-channel DC-EPG amplifiers Blairsville, Georgia, USA) on potted Vicia faba plants (age: (model Giga-4d, EPG-Systems, Wageningen, Netherlands) 2 weeks) for one day before the experiments onset. were used to register the stylet position within the plant tis- sue as well as the feeding activity in the xylem vessel. Data 2.3 No-choice experiment acquisition was performed with ‘Stylet+d’ software (EPG- To evaluate the suitability of the selected host plant for P. Systems, Wageningen, Netherlands). To avoid any plant spumarius nutrition, their survival under controlled con- position effects on the behavior of P. spumarius within the ditions was examined. Twenty adults (10 males and 10 setup, the position of the four plants (one plant per spe- females) were collected in the field in the mid of June (2-4 cies) among the four channels of the EPG device was daily weeks after adult mold) and introduced in a ventilated switched. The experiment was setup in a Faraday cage at cage (33 × 33 × 60 cm, Caterpillar Castle, Live Monarch, semi-controlled conditions of 25±2°C, 20% relative humid- Blairsville, Georgia, USA). Each cage contained one potted ity and artificial light. The substrate of the potted plants was plant of either P. avium, V. vinifera, R. officinalis or N. ole- slightly moistened prior to the beginning of the recording. ander. Four replicate cages per plant species were provided, The were anesthetized with carbon dioxide for 5s, totaling 40 females and 40 males of P. spumarius observed tethered and connected to the EPG following the protocol per plant species. The cages were placed randomly in a walk- by (Cornara et al. 2018b). Philaenus spumarius probing and in climatic chamber ‘Fitotron type SGR233’ (Weiss Technik feeding behavior was recorded for six hours, with two EPG UK Ltd, Loughborough, United Kingdom). Climatic condi- recordings per day (two spittlebugs per day, two per plant, tions were adjusted to 16:8 h light/dark photoperiod at 1.500 both males and females randomly distributed). For each of 4 Anna Markheiser et al. the four different host plants, the insects were placed on the ear mixed models (GLMMs) using the function ‘glmer’ of plant part, where P. spumarius was observed to preferen- package ‘lme4’ (Bates et al. 2015) with binomial family, tially settle if given a free choice (see results from choice whereas gender and host plant were set as fixed factors and experiment below). For the characterization of P. spumar- observed hour as random effect. No-choice experiments, ius probing and feeding behavior on the tested host plants, based on censored data, were analyzed by Cox proportional we referred to the waveforms definition by Cornara et al. hazard models, included in the package ‘survival’ (Therneau (2018b): np (non-probing), C (pathway; stylet tip in plant tis- & Grambsch 2000). A hazard ratio (HR) > 1 indicates an sue), Xc (xylem contact), Xi (xylem ingestion), N (non-path- increased risk, whereas an HR < 1 indicates a decreased way interruption; stylet tip remains in xylem) and R (resting; risk of mortality in comparison to a fixed reference. EPG stylet tip remains in xylem). This study was focused on the waveforms were manually marked in ‘Stylet+a’ software non-sequential EPG variables NWEI (number of waveform (EPG-Systems, Wageningen, Netherlands) and the behav- events per insect) and WDI (waveform duration per insect). ioral variables were subsequently calculated using an Excel To evaluate the host plant acceptance by P. spumarius, the macro analogous to the one used in (Cornara et al. 2018b). Potential Xylem ingestion Index IPXi = WDIXi / (PXi – FXi) Differences in EPG variables were evaluated by generalized × 100, derived from the formula by Girma et al. (1994) for linear models (GLMs) with Akaike information criterions phloem-feeding species, was calculated (WDIXi = waveform (AICs). Time-measurements were modeled using tweedie duration of xylem ingestion per insect, PXi = potential dura- family (package: ‘tweedie’, Dunn 2017) with waveform, sex tion of xylem ingestion (total recording time; 360 min.) and and host plant as fixed factors. Square root transformation FXi = duration to perform the first xylem ingestion >10 min.). was carried out in GLMs for counts and time-measurements The higher the index IPXi (ranges from 0 to 1), the higher the due to the observed overdispersion. The significance of sin- host plant acceptance for the insect species. gle terms within the models was tested using F-test (Chi-test for binomial family) and the function ‘drop1’. The models 2.6 Sampling of P. spumarius in gardens were simplified by removing non-significant (p>0.05) inter- and parks actions and factors. Post hoc comparisons between host The spread of X. fastidiosa from introductory to potential plants, waveforms and sexes were obtained from estimated establishment hosts in gardens and parks requires the pres- marginal means (EMMs) with Tukey adjustment using the ence of vectors in these habitats were ornamentals and function ‘emmeans’ (package: ‘emmeans’; Russell 2018). woody plants grow intermixed. The presence of P. spumarius Significance level was set at p<0.05. All figures were cre- and other potential vector species was recorded by installing ated using the package ‘ggplot2’ (Wickham 2016). Survival yellow sticky traps at four different locations in Germany: curves, based on Kaplan-Meier estimates, were created with Siebeldingen (49°13′06.2″N 8°02′48.3″E), Neustadt the package ‘survminer’ (Kassambara & Kosinski 2018). (49°22′25.8″N 8°08′55.0″E), Weinheim (49°32′50.8″N 8°40′11.0″E) and Bernkastel Kues (49°54′40.4″N 7°05′55.9″E). Each location contained at least one of the 3 Results four plants: P. avium, R. officinalis, V. vinifera and/or N. oleander. Ten (16 in Bernkastel Kues) yellow sticky traps 3.1 Survival on host plants (10 × 25 cm, Aeroxon Insect Control GmbH, Waiblingen, The survival probability of P. spumarius females and males Germany) were installed per location, on a previously speci- over a period of six weeks on the four hosts is visualized in fied position. The traps were attached at a plant height of Fig. 1. The sex of P. spumarius had no significant effect (Cox 100-150 cm from the ground and preferably placed on trees proportional hazard model, p=0.07) on the survival which and shrubs, which were reported to be hosts of X. fastidiosa was rather triggered by the plant species. Cumulative data according to EFSA (2018). Traps were replaced biweekly of females and males showed that, compared to P. avium, over a period of three months between June and September the mortality rate on V. vinifera was not significantly higher 2018 to cover the expected population peak of P. spumarius (HR: 1.3; Cox proportional hazard model, p=0.29). In con- adults in the field. All xylem-feeding species were identi- trast, Rosmarinus officinalis as well as N. oleander were fied, according to morphological characteristics, using keys significantly less suitable (Cox proportional hazard model, of Biedermann and Niedringhaus (2007) and Kunz et al. p<0.01) for the survival of P. spumarius than P. avium (HR: (2011), and the number of individuals per trap was counted. 5.1 and 12.8, respectively). The mortality on N. oleander The plants in the different locations were neither pruned nor was significantly higher (HR: 2.5; Cox proportional hazard treated with pesticides over the observed period. model, p<0.01) than on R. officinalis.

2.7 Data analysis 3.2 Host plant preferences Statistical analysis was performed using the software R – Females and males of P. spumarius were individually version 3.6.0 – ‘Planting of a Tree’ (R Core Team 2019). allowed to choose between the four plant species (Fig. 2). Choice experiments were evaluated by generalized lin- The settlement position was influenced by the plant species Vector behavior to predict Xylella fastidiosa patterns of spread 5

Fig. 1. Kaplan Meier survival curve of Philaenus spumarius representing the mean (± SE) survival probability of females (n=40) and males (n=40) on Prunus avium, Vitis vinifera, Rosmarinus officinalis and Nerium oleander.

Fig. 2. Host plant settlement (mean ± SE) of Philaenus spumarius females (n=24) and males (n=24) in a four-choice test with Prunus avium, Vitis vinifera, Rosmarinus officinalis and Nerium oleander (none refers to cage) over an access period of 6 h.

2 (GLMM, χ 4,1608=180.11, p<0.01), and not by the sex of P. after the first headed. Among them, 15% shifted from N. 2 spumarius (GLMM, χ 1,1608=31.99, p=0.99). V. vinifera (z5= oleander to V. vinifera, R. officinalis or P. avium, whereas 6.48, p<0.01), P. avium (z5= 5.40, p<0.01) and R. officinalis 6% switched from R. officinalis to P. avium or V. vinifera. (z5= 3.34, p<0.01) were significantly preferred over N. ole- No specimens switched from P. avium or V. vinifera to ander for settlement. Specimens settled in the same extent another host. Finally, 29% of the specimens were located on V. vinifera and P. avium (z5= -2.19, p=0.18), whereas at the basal or middle part of the plant, whereas 71% of P. P. spumarius was observed to significantly prefer both (P. spumarius settled on apical parts of the shoot. In the case of avium: z5= 3.28, p<0.01; V. vinifera: z5= 5.21, p<0.01) over V. vinifera, 77 % of the spittlebugs were located on the peti- R. officinalis. Until the end of the observation period of six ole and 23% were distributed on the leaf-veins. The leaf- hours, 90% (43 out of 48) of all specimens made a choice veins were also headed on P. avium (71%) beside the petiole for a specific host plant. An extension of the experiment up (29%). On R. officinalis, P. spumarius settled on leaf mid- to eight hours did not result in any further changes of the vein (58%) or main-stem (42%) of the plant. No specimens settling preference (data not shown). Specimens did not finally settled on N. oleander, although P. spumarius was switch frequently between the host plants. A proportion of observed to use this plant as stopover during the experiment 21% of all P. spumarius switched to another plant species only on leave mid-veins. 6 Anna Markheiser et al.

Table 1. Summary (mean ± SE) of EPG non-sequential parameters (WDI: waveform duration per insect; NWEI: number of waveform events per insect) to study probing of Philaenus spumarius females and males on four different host plants during a 6 h access period to the young shoot (np: non-probing; C: pathway; Xc: xylem contact; Xi: xylem ingestion; N: non-pathway interruption; R: resting). Prunus avium Vitis vinifera Rosmarinus officinalis Nerium oleander wave- sex WDI [min.] NWEI [n] WDI [min.] NWEI [n] WDI [min.] NWEI [n] WDI [min.] NWEI [n] form n n n n (mean ± SE) (mean ± SE) (mean ± SE ) (mean ± SE) (mean ± SE ) (mean ± SE) (mean ± SE ) (mean ± SE) np 52.9 ± 24.0 3.1 ± 0.5 31.3 ± 13.4 3.5 ± 0.6 63.0 ± 18.1 5.6 ± 1.3 152.9 ± 29.6 12.2 ± 2.0 C 11.2 ± 2.1 5.1 ± 0.9 12.1 ± 2.3 5.3 ± 1.1 9.3 ± 2.3 7.5 ± 1.7 26.4 ± 5.0 16 ± 2.8 Xc 2.5 ± 0.6 2.9 ± 0.5 2.2 ± 1.1 2.9 ± 0.5 2.4 ± 0.8 3.0 ± 0.6 2.1 ± 0.5 4.5 ± 0.8 female 15 15 13 13 Xi 219.7 ± 24.0 2.5 ± 0.3 257.8 ± 23.5 2.7 ± 0.4 144.7 ± 16.4 2.7 ± 0.4 102.5 ± 18.9 4.1 ± 0.7 N 5.7 ± 1.0 22.9 ± 4.9 4.8 ± 0.8 20.1 ± 3.4 4.2 ± 1.0 11.4 ± 2.8 5.0 ± 1.4 11.2 ± 3.0 R 73.6 ± 21.1 13.5 ± 4.7 56.6 ± 15.6 8.3 ± 1.9 139.9 ± 21.7 20.5 ± 4.0 76.1 ± 20.3 17.6 ± 4.6 np 57.4 ± 16.2 3.6 ± 0.9 51.6 ± 17.7 4.1 ± 0.6 45.8 ± 13.3 8.2 ± 1.1 231.7 ± 32.8 10.7 ± 2.0 C 11.1 ± 3.1 5.6 ± 1.4 10.8 ± 1.5 6.1 ± 1.0 12.8 ± 1.6 11.2 ± 1.6 19.5 ± 3.7 12.2 ± 2.3 Xc 3.8 ± 1.3 3.4 ± 0.7 1.3 ± 0.5 2.7 ± 0.4 2.0 ± 0.4 3.8 ± 0.5 0.8 ± 0.2 2.2 ± 0.5 male 14 15 12 13 Xi 204.4 ± 24.5 2.9 ± 0.6 205.4 ± 27.3 2.5 ± 0.4 174.2 ± 24.1 3.2 ± 0.4 51.6 ± 13.2 2.0 ± 0.5 N 3.3 ± 0.8 13.5 ± 3.5 2.6 ± 0.4 11.1 ± 2.2 5.8 ± 1.9 16.3 ± 4.2 4.1 ± 1.3 6.3 ± 2.6 R 82.2 ± 17.1 12.8 ± 3.2 90.9 ± 22.5 13.4 ± 2.0 123.8 ± 26.2 15.5 ± 3.3 56.5 ± 24.3 8.9 ± 3.2

3.3 Probing and feeding behavior ber of interruptions in Xi, which are described by waveform The interaction between the waveforms and the host plants N, showed a trend comparable to the duration of Xi, which was a significant factor explaining the variations in was significantly higher in P. avium (z24=4.57, p<0.01), V. WDI (GLM: F15,635=101.74, p<0.01) and NWEI (GLM: vinifera (z24=3.93, p<0.01) and R. officinalis (z24=2.90, F15,632=77.01, p<0.01) observed among the host plants. p=0.02) than in N. oleander. Although evident by analyzing the rough dataset IPXi was calculated for females and males separately (Table 1), sex-related differences were neither confirmed (Table 2). The factor sex could not explain differences for WDI (GLM: F1,635=0.99, p=0.32) nor NWEI (GLM: observed among IPXi indices (GLM: F1,88=3.87, p=0.06), F1,632=2.14, p=0.14). Therefore, for the analysis of differ- thus data of both sexes were combined. The index was trig- ences in P. spumarius probing and feeding behavior among gered by the plant species (GLM: F3,88=3.87, p<0.01) and the tested plants, observations on males and females were was significantly higher in V. vinifera (IPXi=0.77; z4=3.27, pooled in a single dataset. This result was corroborated by p<0.01) and P. avium (IPXi=0.77; z4=3.23, p<0.01) than in N. the visualization of sequential changes in the probing behav- oleander (IPXi=0.45). In addition, R. officinalis (IPXi=0.57) ior of females and males (WDI calculated for each hour) dur- was neither classified higher than N. oleander (z4=–1.36, ing the 6 h EPG (Fig. 3). p=0.53) nor lower than V. vinifera (z4=2.09, p=0.16) or P. In the case of N. oleander, P. spumarius spent significantly avium (z4=2.04, p=0.17). more time in np than measured for P. avium (z24=–5.16, p<0.01), R. officinalis (z24=–4.99, p<0.01) and V. vinifera 3.4 Xylem feeder diversity in gardens and parks (z24=–6.34, p<0.01) (Fig. 4A). Simultaneously, the C phase A total of 1103 specimens of xylem feeders were caught lasted longer in N. oleander as in V. vinifera (z24=–4.99, on yellow sticky traps exposed in four different gar- p<0.01) and R. officinalis (z24=–2.67, p=0.04). The signifi- den and park habitats between June and September 2018. cantly lowest duration of Xi was observed for N. oleander Philaenus spumarius and another four potential vector spe- if compared to V. vinifera (z24=4.63, p=<0.01), P. avium cies could be identified. Three of them belong to the family (z24=4.22, p=<0.01) or R. officinalis (z24=2.89, p=0.02), Cidadellidae: ( fennahi Young whereas the number of Xi phases, which is analogous to the (1977), viridis Müller (1764) and Evacanthus number of successful probes, and the number of Xc phases interruptus L. (1758)) and one to the family Aphrophoridae did not differ between the plant species (p>0.05; statistics (Aphrophora alni Fallen (1805). The most common species included in Fig. 4B). In contrast, the number of C phases is captured by sticky traps was the ‘ ’ more than two-times higher in N. oleander than in P. avium G. fennahi (1027 individuals), followed by P. spumarius (29 (z24=–4.47, p<0.01) and V. vinifera (z24=–4.20, p<0.01). individuals), the ‘green leafhopper’ C. viridis (25 individu- Recordings of R phases on R. officinalis lasted significantly als), the ‘European alder spittle bug’ A. alni (18 individuals) longer (z24=2.73, p=0.03) and occurred in higher frequency and E. interruptus (4 individuals). The vector P. spumarius (z24=2.65, p=0.04) in comparison to N. oleander. The num- was present in all plots and could be found on shrubs and Vector behavior to predict Xylella fastidiosa patterns of spread 7

Fig. 3. Mean hour-sequential changes in Philaenus spumarius probing of females and males on four host plants (Prunus avium, Vitis vinifera, Rosmarinus officinalis and Nerium oleander) during 6 h EPG recording, expressed by the percentage of time dedicated to np (non-probing), C (pathway), Xc (xylem contact), Xi (xylem ingestion), N (non-pathway interruption) and R (resting).

trees from mid of July until the end of August (Fig. 5A). It mission efficiency and the rate of pathogen spread influenced was trapped specifically near all the four plant species tested among others by feeding preferences (Daugherty et al. 2011; in this study (Fig. 5B). Philaenus spumarius was addition- Sisterson 2008). The EPG technique enables distinguishing ally caught on traps exposed on Carpinus betulus, Corylus between probing and feeding activities of piercing-sucking avellana, Lavandula sp., Ligustrum vulgare, Pinus strobus, insects and has already been applied as a tool to assess the Prunus domestica, Prunus laurocerasus and Salix sp. feeding behavior of introduced vectors on economically important crops in New Zealand by Sandanayaka et al. (2013). Assuming that host plant acceptance in xylem-feed- 4 Discussion ing insects is positively correlated with an extensive feed- ing activity in the xylem vessels, we focused primarily on Under Central European conditions, it is assumed that the EPG variables that assess the duration (WDI) and frequency most likely pattern of X. fastidiosa introduction is through (NWEI) of single probing and feeding activities. Insects on infected ornamental plants (EFSA 2018). Gardens and parks suitable host plants typically show shorter pathway prob- are habitats were ornamentals, forest and fruit trees are ing, shorter non-probing periods and longer feeding activity closely associated. An initial infection of a perennial host than on non-suitable host plants (Sandanayaka et al. 2017). in such an environment could subsequently lead to a spread Considering these assumptions, plants such as grapevine and of the pathogen to susceptible cultures such as orchards or cherry were more suitable hosts for P. spumarius than rose- vineyards, giving that competent vectors are widely spread. mary and oleander. The results of the EPG were corrobo- Vectors of X. fastidiosa play an active role in the inoculation rated by the corresponding results of the survival studies, of bacterial cells into plants (Almeida et al. 2005), with trans- with P. spumarius surviving significantly longer on cherry 8 Anna Markheiser et al.

Fig. 4. (A) Mean (± SE) duration of waveform per insect (WDI) and (B) Mean (± SE) number of waveform events per insect (NWEI) of Philaenus spumarius females and males on Prunus avium (n=29), Vitis vinifera (n=30), Rosmarinus officinalis (n=25) and Nerium oleander (n=26). Small letters indicate statistically significant differences between host plants according to GLM and pairwise com- parisons of EMMs (α=0.05).

Table 2. Summary (mean ± SE) of EPG xylem ingestion parameters of Philaenus spumarius females and males on four different host plants (PXi: potential duration of xylem ingestion; FXi: duration to perform the first xylem ingestion (>10 min.); WDIXi: waveform dura- tion per insect of xylem ingestion; IPXi: potential xylem ingestion index; n: Number of insects; nP: Number of insects (performed Xi>10 min.)). Prunus avium Vitis vinifera Rosmarinus officinalis Nerium oleander sex parameter time [min] time [min] time [min] time [min] n nP n nP n nP n nP (mean ± SE) (mean ± SE) (mean ± SE) (mean ± SE)

Pxi 360.0 360.0 360.0 360.0

FXi 72.6 ± 13.9 48.1 ± 8.4 39.8 ± 12.4 104.7 ± 25.2 female 15 14 15 15 13 11 13 9 WDIXi 233.5 ± 21.0 257.8 ± 23.5 160.5 ± 14.7 134.4 ± 18.6

IPXi 0.86 ± 0.11 0.84 ± 0.08 0.52 ± 0.06 0.59 ± 0.11

Pxi 360.0 360.0 360.0 360.0 F 40.6 ± 12.0 36.4 ± 6.9 54.0 ± 10.7 40.5 ± 10.4 male Xi 14 14 14 13 12 11 12 6 WDIXi 204.4 ± 24.5 222.8 ± 28.5 185.7 ± 23.2 77.9 ± 20.3

IPXi 0.67 ± 0.10 0.69 ± 0.09 0.62 ± 0.09 0.25 ± 0.07 Vector behavior to predict Xylella fastidiosa patterns of spread 9

Fig. 5. Seasonal traps counts (mean ± SE) for Philaenus spumarius between June and September 2018 observed by yellow sticky traps (A) in two private gardens (Bernkastel Kues (n=16), Neustadt (n=10)) and two public parks (Siebeldingen (n=10), Weinheim (n=10)) and (B) specific on Prunus avium (n=3), Vitis vinifera (n=3), Rosmarinus officinalis (n=3) and Nerium oleander (n=5).

and grapevine than on rosemary and oleander. Rosemary is mary and oleander at least for P. spumarius. Nevertheless, it the only plant on which males showed longer ingestion com- should be noticed that our study was performed on uninfected pared to females, which may explain their higher survival asymptomatic plants; infections with X. fastidiosa and symp- probability on this plant. A subsequent calculation of the toms expression might alter vector behavior with respect to potential xylem ingestion index (IPXi), which ranged from the data reported here (Marucci et al. 2005; Miranda et al. 0.77 (for cherry and grapevine) over 0.57 (for rosemary) to 2013; Zeilinger & Daugherty 2014). Regarding X. fastidi- 0.45 (for oleander), enabled to clearly classify host plants for osa inoculation, the efficiency seems to depend more on the suitability. number of probes per vector on the recipient plants rather According to Almeida & Purcell (2003), Purcell & than on the time spent by the single vector feeding on the Finlay (1979) and Cornara et al. (2016), transmission effi- host (Jackson et al. 2008; Daugherty et al. 2009). Obviously, ciency is positively correlated to the time the insect is given suitable host plants such as grapevine and cherry, being more access to the X. fastidiosa source plant. Xylella fastidiosa attractive for both settlement and feeding than rosemary is unevenly distributed in the plant (Newman et al. 2003); and oleander, would be more exposed to the risk of infec- furthermore, most of the bacterial cells are likely ingested tion by X. fastidiosa. However, the inoculation period could rather than attached to the vector foregut (Retchless et al. be a matter of minutes after infectious P. spumarius settle 2014). Therefore, a long time spent on an infected plant, and probe on any of the plant species tested in this study. direct reflection of vector preference for settlement and feed- Therefore, we can infer that after acquisition on grapevine ing would make the plant a suitable source of the bacterium. and/or cherry, and bacterium secondary spread among the Thus, according to our data, grapevine and cherry, economi- cultivated plants, X. fastidiosa could be transmitted to orna- cally important plants threatened by X. fastidiosa in Central mentals as rosemary and oleander during spittlebugs disper- Europe, would be better sources of the bacterium than rose- sal outside the cultivated orchards. 10 Anna Markheiser et al.

The population peak of adult P. spumarius in gardens and essential role for the establishment and further dissemination parks was observed at the beginning of August, although first of X. fastidiosa in the Central European situation. adults of this species on the tree or shrub canopy occur at We conclude that, considering data on P. spumarius the beginning of June under our environmental conditions behaviors conducive to transmission, economically impor- (Markheiser & Maixner 2017). Adults possibly entered this tant cultivated plants threatened by X. fastidiosa in Central area from adjacent neglected fields. A comparable effect of Europe as grapevine and cherry could be better sources migration was noticed by Morente et al. (2018) and Bodino of the bacterium for P. spumarius than rosemary and ole- (2018), who observed a movement of P. spumarius from ander, the latter considered as potential introductory hosts. the olive orchard, where juvenile stages developed, to However, our data do not rule out the possibility of an occa- neighbored conifers. Philaenus spumarius was found in all sional acquisition even from rosemary and oleander, given observed areas and on several plant species that can host X. that we recorded xylem ingestions from all the hosts tested. fastidiosa, including P. avium, V. vinifera, Rosmarinus sp. Upon acquisition, the spittlebug might mediate bacterium and N. oleander. secondary spread within the cultivated orchards and succes- In this study P. spumarius settled on oleander at least tem- sively transmit X. fastidiosa to surrounding plants including porarily despite of the low attractiveness of this plant and is rosemary and oleanders during insect’s dispersal to other also able to ingest xylem sap from it. Interestingly, Cornara habitats. Such possible pattern of spread should be carefully et al. (2017a) observed even higher transmission rates by P. considered when devising phytosanitary measured aimed at spumarius to oleander than to olive. They highlighted that reducing X. fastidiosa risk of spread in case of introduction transmission efficiency may depend on factors other than the in Central Europe. Furthermore, our experimental approach time a vector spends on a host. Jackson et al. (2008) found could be adopted for other vectors and candidate vectors of the number of probes to be positively correlated with trans- X. fastidiosa in other scenarios. mission efficiency. In our study, P. spumarius had a higher number of total probes on oleander than on grapevine and cherry, whereas the number of successful probes leading to Acknowledgements: We are deeply grateful for the daily assis- xylem ingestion was similar over the four plant species. In tance of all laboratory members at CSIC-ICA, in particular Marina the case of N. oleander, P. spumarius exhibited more fre- Morente for insect sampling and Jaime Jiménez for technical sup- quently (in 21 of 25 plants) than on other plants (P. avium: port. AM and DC have been financially supported by research 5/29 plants; V. vinifera: 7/29 plants and R. officinalis: 5/25 grants in the frame of European Union Horizon 2020 research and plants) patterns similar to spikelet burst B1s, described by innovation program under grant agreements no. 727987 Backus et al. (2005) and Joost et al. (2006) for sharpshoot- XF-ACTORS (Xylella Fastidiosa Active Containment Through a multidisciplinary-Oriented Research Strategy). ers, which represent precibarial valve movement for tasting. This pattern occurred during Xi or R phase. In this study, these patterns were marked as Xi or R, although further work is needed to better characterize its possible biological Author contribution: AM and DC designed and conducted the meaning. experiments and analyzed EPG data. AM performed statistical Other xylem-feeder species such as G. fennahi may analysis and wrote the original draft of the manuscript. DC, AF and MM revised the draft, whereas AF and MM supervised the work also play a role in the bacterium’s spread within gardens and provided funding. and parks from transient hosts, where it is introduced, to permanent plant hosts, where it may overcome cold peri- ods in winter. This species is taxonomically related to the References ‘blue-green sharpshooter’ G. atropunctata Signoret (1854), a native vector in the American PD-outbreaks (Hill & Purcell Almeida, R. P. P., Blua, M. J., Lopes, J. R. S., & Purcell, A. H. 1995). Observation of G. fennahi by yellow sticky traps (2005). Vector transmission of Xylella fastidiosa: Applying fun- resulted in a higher numbers of trapped specimens than for damental knowledge to generate disease management strate- P. spumarius but this method is assumed to be less efficient gies. Annals of the Entomological Society of America, 98(6), for P. spumarius than sweep net samplings as discussed by 775–786. https://doi.org/10.1603/0013-8746(2005)098[0775: Morente et al. (2018). Observation on G. fennahi in field VTOXFA]2.0.CO;2 revealed presence of adults near a high diversity of plants in Almeida, R. P. P., & Purcell, A. H. (2003). Transmission of Xylella contrast to immature stages (Markheiser & Maixner 2017), fastidiosa to grapevines by Homalodisca coagulata (: Cicadellidae). Journal of Economic Entomology, 96(2), 264– which are assumed to be monophagous on Rhododendron 271. https://doi.org/10.1603/0022-0493-96.2.264 (Sergel 1987). Studies of the feeding behavior of this species Backus, E. A., Habibi, J., Yan, F. M., & Ellersieck, M. (2005). on X. fastidiosa susceptible hosts are ongoing. Stylet penetration by adult Homalodisca coagulata on grape: Up to now, a potential establishment of the bacterium in Electrical penetration graph waveform characterization, tissue countries of Central Europe, like Germany, cannot be fore- correlation, and possible implications for transmission of Xylella seen. The results indicate that P. spumarius could play an fastidiosa. Annals of the Entomological Society of America, Vector behavior to predict Xylella fastidiosa patterns of spread 11

98(6), 787–813. https://doi.org/10.1603/0013-8746(2005)098 worldwide vectors of Xylella fastidiosa. Entomologia Generalis. [0787:SPBAHC]2.0.CO;2 https://doi.org/10.1127/entomologia/2019/0811 Bates, D., Mächler, M., Bolker, B., & Walker, S. (2015). Fitting Cruaud, A., Gonzalez, A. A., Godefroid, M., Nidelet, S., Streito, J. Linear Mixed-Effects Models Using lme4.Journal of Statistical C., Thuillier, J. M., … Rasplus, J.-Y. (2018). Using insects to Software, 67(1), 1–48. https://doi.org/10.18637/jss.v067.i01 detect, monitor and predict the distribution of Xylella fastidiosa: Benning, T. L., LaPointe, D., Atkinson, C. T., & Vitousek, P. M. A case study in Corsica. Scientific Reports 8,15628 . https://doi. (2002). Interactions of climate change with biological invasions org/10.1038/s41598-018-33957-z and land use in the Hawaiian Islands: Modeling the fate of Daugherty, M. P., & Almeida, R. P. P. (2009). Estimating Xylella endemic birds using a geographic information system. fastidiosa transmission parameters: Decoupling sharpshooter Proceedings of the National Academy of Sciences of the United number and feeding period. Entomologia Experimentalis et States of America, 99(22), 14246–14249. https://doi.org/ Applicata, 132(1), 84–92. https://doi.org/10.1111/j.1570-7458. 10.1073/pnas.162372399 2009.00868.x Biedermann, R., & Niedringhaus, R. (2007). Die Zikaden Daugherty, M. P., Rashed, A., Almeida, R. P. P., & Perring, T. M. Deutschlands – Bestimmungstafel für alle Arten (2. Auflage). (2011). Vector preference for hosts differing in infection status: Wissenschaftlich-Akademischer Buchvertrieb (WABV). Fründ, Sharpshooter movement and Xylella fastidiosa transmission. Scheeßel, Germany. Ecological Entomology, 36(5), 654–662. https://doi.org/ Blua, M. J., & Morgan, D. J. W. (2003). Dispersion of Homalodisca 10.1111/j.1365-2311.2011.01309.x coagulata (Hemiptera: Cicadellidae), a vector of Xylella fastidi- Del Cid, C., Krugner, R., Zeilinger, A. R., Daugherty, M. P., & osa, into vineyards in southern California. Journal of Economic Almeida, R. P. P. (2018). Plant water stress and vector feeding Entomology, 96(5), 1369–1374. https://doi.org/10.1603/0022- preference mediate transmission efficiency of a plant pathogen. 0493-96.5.1369 Environmental Entomology, 47, 1471–1478. https://doi. Bodino, N., Cavalieri, V., Saladini, M., Ciniero, A., Dongiovanni, org/10.1093/ee/nvy136 C., Altamura, G., … (2018). Preliminary results on the transmis- Dunn, P. K. (2017). Tweedie: Evaluation of Tweedie exponential sion characteristics of Xylella fastidiosa subsp. pauca (ST53) by family models. R package version 2.3. https://cran.r-project.org/ Philaenus spumarius and Cicadella viridis. European Research web/packages/tweedie/index.html. Accessed on 04 December of Emerging Plant Diseases. 2nd Joint Annual Meeting, Valencia 2018. October 2018, p. 69. Dukes, J. S., & Mooney, H. A. (1999). Does global change increase Burbank, L. P., & Stenger, D. C. (2016). A Temperature-Independent the success of biological invaders? Trends in Ecology & Cold-Shock Protein Homolog Acts as a Virulence Factor in Evolution, 14(4), 135–139. https://doi.org/10.1016/S0169- Xylella fastidiosa. Molecular Plant-Microbe Interactions, 5347(98)01554-7 29(5), 335–344. https://doi.org/10.1094/MPMI-11-15-0260-R EC (2018). Commission database of host plants found to be suscep- Cariddi, C., Saponari, M., Boscia, D., De Stradis, A., Loconsole, tible to Xylella fastidiosa in the Union territory – Update 11. G., Nigro, F., … Martelli, G. (2014). Isolation of a Xylella fas- https://ec.europa.eu/food/sites/food/files/plant/docs/ph_biosec_ tidiosa strain infecting olive and oleander in Apulia, Italy. legis_emergency_db-host-plants_update11.pdf. Accessed on 19 Journal of Plant Pathology, 96, 425–429. September 2018. Cornara, D., Sicard, A., Zeilinger, A. R., Porcelli, F., Purcell, A. H., EFSA. (2015). Scientific Opinion on the risks to plant health posed & Almeida, R. P. P. (2016). Transmission of Xylella fastidiosa to by Xylella fastidiosa in the EU territory, with the identification grapevine by the meadow spittlebug. Phytopathology, 106(11), and evaluation of risk reduction options. EFSA Journal, 13, 1285–1290. https://doi.org/10.1094/PHYTO-05-16-0202-R 1–262. https://doi.org/10.2903/j.efsa.2015.3989 Cornara, D., Cavalieri, V., Dongiovanni, C., Altamura, G., EFSA. (2018). Update of the Xylella spp. host plant database. EFSA Palmisano, F., Bosco, D., … Saponari, M. (2017a). Transmission Journal, 16, 1–9. https://doi.org/10.2903/j.efsa.2018.5408 of Xylella fastidiosa by naturally infected Philaenus spumarius EPPO (2018). Xylella fastidiosa (XYLEFA): Distribution. https:// (Hemiptera, Aphrophoridae) to different host plants. Journal of gd.eppo.int/taxon/XYLEFA/distribution. Accessed on 20 Applied Entomology, 141(1-2), 80–87. https://doi.org/10.1111/ February 2018. jen.12365 Feil, H., & Purcell, A. H. (2001). Temperature-dependent growth Cornara, D., Saponari, M., Zeilinger, A. R., de Stradis, A., Boscia, and survival of Xylella fastidiosa in vitro and in potted grape- D., Loconsole, G., … Porcelli, F. (2017b). Spittlebugs as vectors vines. Plant Disease, 85(12), 1230–1234. https://doi.org/ of Xylella fastidiosa in olive orchards in Italy. Journal of Pest 10.1094/PDIS.2001.85.12.1230 Science, 90(2), 521–530. https://doi.org/10.1007/s10340-016- Fereres, A. (2015). Perspective Insect vectors as drivers of plant 0793-0 virus emergence. Current Opinion in Virology, 10, 42–46. Cornara, D., Bosco, D., & Fereres, A. (2018a). Philaenus spumar- https://doi.org/10.1016/j.coviro.2014.12.008 ius: When an old acquaintance becomes a new threat to European Fereres, A., & Moreno, A. (2009). Behavioural aspects influencing agriculture. Journal of Pest Science, 91(3), 957–972. https://doi. plant virus transmission by homopteran insects. Virus Research, org/10.1007/s10340-018-0966-0 141(2), 158–168. https://doi.org/10.1016/j.virusres.2008.10.020 Cornara, D., Garzo, E., Morente, M., Moreno, A., Alba-Tercedor, J., Girma, M., Wilde, G. E., Reese, J. C., & Backus, E. A. (1994). & Fereres, A. (2018b). EPG combined with micro-CT and video Committed phloem ingestion of aphids (Homoptera, recording reveals new insights on the feeding behavior of Aphididae) – Difficulties in its application to host-plant resis- Philaenus spumarius. PLoS One, 13(7), e0199154. https://doi. tance studies. Environmental Entomology, 23(5), 1061–1065. org/10.1371/journal.pone.0199154 https://doi.org/10.1093/ee/23.5.1061 Cornara, D., Morente, M., Markheiser, A., Bodino, N., Tsai, C. W., Godefroid, M., Cruaud, A., Streito, J.-C., Rasplus, J.-Y., & Rossi, Fereres, A., … Lopes, J. R. S. (2019). An overview on the J.-P. (2018). Climate change and the potential distribution of 12 Anna Markheiser et al.

Xylella fastidiosa in Europe. bioRxiv. PrePrint. https://doi. Miranda, M. P. de, Villada, E. S., Lopes, S. A., Fereres, A., & Lopes, org/10.1101/289876 J. R. S. (2013). Influence of citrus plants infected with Xylella Goodwin, P. H., & Zhang, S. (1997). Distribution of Xylella fastidi- fastidiosa on stylet penetration activities of Bucephalogonia osa in southern Ontario as determined by the polymerase chain xanthophis (Hemiptera: Cicadellidae). Annals of the reaction. Canadian Journal of Plant Pathology, 19(1), 13–18. Entomological Society of America, 106(5), 610–618. https://doi. https://doi.org/10.1080/07060669709500564 org/10.1603/AN12148 Hill, B. L., & Purcell, A. H. (1995). Acquisition and retention of Morente, M., Cornara, D., Plaza, M., Duran, J. M., Capiscol, C., Xylella-fastidiosa by an efficient vector, Graphocephala- Trillo, R., … Fereres, A. (2018). Distribution and relative abun- atropunctata. Phytopathology, 85(2), 209–212. https://doi.org/ dance of insect vectors of Xylella fastidiosa in olive groves of 10.1094/Phyto-85-209 the Iberian Peninsula. Insects, 9(4), 1–18. https://doi.org/ Hulme, P. E. (2009). Trade, transport and trouble: Managing inva- 10.3390/insects9040175 sive species pathways in an era of globalization. Journal of Nault, L. R. (1997). Transmission of Plant Viruses: A Applied Ecology, 46(1), 10–18. https://doi.org/10.1111/j.1365- New Synthesis. Annals of the Entomological Society of America, 2664.2008.01600.x 90(5), 521–541. https://doi.org/10.1093/aesa/90.5.521 Jackson, B. C., Blua, M. J., & Bextine, B. (2008). Impact of dura- Newman, K. L., Almeida, R. P. P., Purcell, A. H., & Lindow, S. E. tion versus frequency of probing by Homalodisca vitripennis (2003). Use of a green fluorescent strain for analysis of Xylella (Hemiptera : Cicadellidae) on inoculation of Xylella fastidiosa. fastidiosa colonization of Vitis vinifera. Applied and Journal of Economic Entomology, 101(4), 1122–1126. https:// Environmental Microbiology, 69(12), 7319–7327. https://doi. doi.org/10.1603/0022-0493(2008)101[1122:IODVFO]2.0.CO;2 org/10.1128/AEM.69.12.7319-7327.2003 Janse, J. D., & Obradovic, A. (2010). Xylella fastidiosa: Its Biology, Olmo, D., Nieto, A., Adrover, F., Urbano, A., Beidas, O., Juan, diagnosis, control and risks. Journal of Plant Pathology, 92, A., … Landa, B. B. (2017). First detection of Xylella fastidiosa 35–48. infecting cherry (Prunus avium) and Polygala myrtifolia plants JKI (2016). First finding of Xylella fastidiosa ssp. fastidiosa in in Mallorca Island, Spain. Plant Disease, 101, 1820. https://doi. Germany. https://pflanzengesundheit.julius-kuehn.de/dokumente/ org/10.1094/PDIS-04-17-0590-PDN upload/3a817_xylella-fastidiosa_pest-report.pdf. Accessed on Purcell, A. H., & Finlay, A. H. (1979). Acquisition and transmission 20 February 2018. of bacteria through artificial membranes by leafhopper vectors JKI (2018). Notification of the eradication of a harmful organism of Pierces Disease. Entomologia Experimentalis et Applicata, (update - closing note). https://pflanzengesundheit.julius-kuehn. 25(2), 188–195. https://doi.org/10.1111/j.1570-7458.1979. de/dokumente/upload/10190_xylefa-eradication2018sn.pdf. tb02870.x Accessed on 04 April 2018. Purcell, A. H., Suslow, K. G., & Klein, M. (1994). Transmission via Joost, P. H., Backus, E. A., Morgan, D., & Yan, F. M. (2006). plants of an insect pathogenic bacterium that does not multiply Correlation of stylet activities by the glassy-winged sharp- or move in plants. Microbial Ecology, 27(1), 19–26. https://doi. shooter, Homalodisca coagulata (Say), with electrical penetra- org/10.1007/BF00170111 tion graph (EPG) waveforms. Journal of Insect Physiology, Rathé, A. A., Pilkington, L. J., Gurr, G. M., Hoddle, M. S., 52(3), 327–337. https://doi.org/10.1016/j.jinsphys.2005.11.012 Daugherty, M. P., Constable, F. E., … Edwards, O. R. (2012). Kassambara, A., & Kosinski, M. (2018). survminer: Drawing Incursion preparedness: anticipating the arrival of an economi- Survival Curves using ‘ggplot2’. R package version 0.4.3. cally important plant pathogen Xylella fastidiosa Wells https://CRAN.R-project.org/package=survminer. Accessed on (Proteobacteria: Xanthomonadaceae) and the insect vector 06 January 2018. Homalodisca vitripennis (Germar) (Hemiptera: Cicadellidae) in Kunz, G., Nickel, H., & Niedringhaus, R. (2011). Photographic Australia. Australian Journal of Entomology, 51(3), 209–220. Atlas of the Planthoppers and of Germany https://doi.org/10.1111/j.1440-6055.2011.00856.x (1. Auflage). Wissenschaftlich-Akademischer Buchvertrieb R Core Team (2019). R: A language and environment for statistical (WABV) – Fründ, Scheeßel, Germany. computing. R Foundation for Statistical Computing, Vienna, Lopes, J. R. S., Daugherty, M. P., & Almeida, R. P. P. (2010). Strain Austria. URL: https://www.R-project.org/ origin drives virulence and persistence of Xylella fastidiosa in Retchless, A. C., Labroussaa, F., Shapiro, L., Stenger, D. C., alfalfa. Plant Pathology, 59(5), 963–971. https://doi.org/ Lindow, S. E., & Almeida, R. P. (2014). Genomic insights into 10.1111/j.1365-3059.2010.02325.x Xylella fastidiosa interactions with plant and insect hosts. In Markheiser, A., & Maixner, M. (2017). Significance of xylem feed- S. B. Heidelberg (Ed.), Genomics of Plant-Associated Bacteria ing Auchenorrhyncha in orchards and vineyards in Germany. (pp. 177–202). https://doi.org/10.1007/978-3-642-55378-3_8 European conference on Xylella fastidiosa: finding answers to a Russell, L. (2018). emmeans: Estimated Marginal Means, aka global problem. Palma de Mallorca, November 2017, p 21. Least-Squares Means. R package version 1.3.0. https:// Martelli, G. P., Boscia, D., Porcelli, F., & Saponari, M. (2016). The CRAN.R-project.org/package=emmeans. Accessed on 06 olive quick decline syndrome in south-east Italy: A threatening January 2019. phytosanitary emergency. European Journal of Plant Pathology, Sandanayaka, M., Jia, Y., & Charles, J. G. (2013). EPG technique as 144(2), 235–243. https://doi.org/10.1007/s10658-015-0784-7 a tool to reveal host plant acceptance by xylem sap-feeding Marucci, R. C., Lopes, J. R. S., Vendramim, J. D., & Corrente, J. E. insects. Journal of Applied Entomology, 137(7), 519–529. (2005). Influence ofXylella fastidiosa infection of citrus on host https://doi.org/10.1111/jen.12025 selection by leafhopper vectors. Entomologia Experimentalis et Sandanayaka, M., Nielsen, M., Davis, V., & Butler, R. (2017). Do Applicata, 117(2), 95–103. https://doi.org/10.1111/j.1570-7458. spittlebugs feed on grape? Assessing transmission potential for 2005.00336.x Xylella fastidiosa. New Zealand Plant Protection, 70, 31–37. https://doi.org/10.30843/nzpp.2017.70.23 Vector behavior to predict Xylella fastidiosa patterns of spread 13

Saponari, M., Boscia, D., Nigro, F., & Martelli, G. P. (2013). osa: Insights into an Emerging Plant Pathogen. Annual Review Identification of DNA sequences related to Xylella fastidiosa in of Phytopathology, 56(1), 181–202. https://doi.org/10.1146/ oleander, almond and olive trees exhibiting leaf scorch symp- annurev-phyto-080417-045849 toms in Apulia (Southern Italy). Journal of Plant Pathology, 95, Sisterson, M. (2008). Effects of insect-vector preference for healthy 668. or infected plants on pathogen spread: Insights from a model. Saponari, M., Boscia, D., Loconsole, G., Palmisano, F., Savino, V., Journal of Economic Entomology, 101(1), 1–8. https://doi. Potere, O., & Martelli, G. P. (2014). New hosts of Xylella fas- org/10.1093/jee/101.1.1 tidiosa strain CoDiRO in Apulia. Journal of Plant Pathology, Therneau, T. M., & Grambsch, P. M. (2000). Modeling Survival 96, 611. Data: Extending the Cox Model. New York: Springer. https:// Saponari, M., Boscia, D., Altamura, G., Loconsole, G., Zicca, S., doi.org/10.1007/978-1-4757-3294-8 D’Attoma, G., … Martelli, G. P. (2017). Isolation and pathoge- Wickham, H. (2016). ggplot2: Elegant Graphics for Data Analysis. nicity of Xylella fastidiosa associated to the olive quick decline Springer-Verlag, New York. http://ggplot2.org. Accessed on 06 syndrome in southern Italy. Scientific Reports, 7(1), 17723. January 2019. https://doi.org/10.1038/s41598-017-17957-z Zeilinger, A. R., & Daugherty, M. P. (2014). Vector preference and Sergel, R. (1987). On the Occurrence and Ecology of the host defense against infection interact to determine disease Rhododendron-Leafhopper, Graphocephala-Fennahi Young dynamics. Oikos, 123(5), 613–622. https://doi.org/10.1111/ 1977, in the Western Palearctic Region (Homoptera, j.1600-0706.2013.01074.x Cicadellidae). Anzeiger für Schädlingskunde Pflanzenschutz Umweltschutz, 60(7), 134–136. https://doi.org/10.1007/ BF01907495 Severin, H. H. P. (1950). Spittle-insect vectors of Pierce’s disease virus. II. Life history and virus transmission. Hilgardia, 19(11), Manuscript received: 5 March 2019 357–382. https://doi.org/10.3733/hilg.v19n11p357 Revisions requested: 1 May 2019 Sicard, A., Zeilinger, A. R., Vanhove, M., Schartel, T. E., Beal, D. Modified version received: 27 May 2019 J., Daugherty, M. P., & Almeida, R. P. P. (2018). Xylella fastidi- Accepted: 18 July 2019