Scaphoideus titanus fitness on grapevine varieties with different susceptibility to Flavescence dorée phytoplasma

Scaphoideus titanus (Ball) is a Nearctic leafhopper (Cicadellidae: Deltocephalinae), monophagous on Vitis spp., naturalized in European vineyards since the 20th century. Following its introduction and establishment in Europe, S. titanus became the main vector of Flavescence dorée phytoplasma (FDp) to grapevine. FDp causes heavy economic losses to viticulture in Europe. The control of the disease mainly relies on insecticide applications against the vector and on the removal of infected plants. The exploitation of plant resistance or tolerance against pathogens and pests can represent a valuable tool for a more sustainable viticulture. The first step in identifying resistance traits in grapevine varieties towards S. titanus is the evaluation of insect fitness parameters on different grapevine genotypes. Therefore, in this work, nymph mortality and developmental time, adult survival and prolificacy of S. titanus have been studied on three grapevine varieties, characterized by different susceptibility to FD. Scaphoideus titanus showed highest fitness when reared on Barbera, whereas the worst performances were recorded on Moscato, with a significant reduction in nymph and adult survivals, a slowed nymph development, and a decreased number of available eggs. On Brachetto, S. titanus showed an intermediate level of fitness parameters. Consistently with previous studies on feeding behaviour, Barbera is the most suitable host for the FD‐vector. The high suitability of Barbera for S. titanus may partly explain the high susceptibility to FD of this variety. On the contrary, the low suitability of Moscato, poorly susceptible to FD, may be due to antibiosis and antixenosis defence mechanisms that act against S. titanus.

& Mori, 2013). There is an urgent need for developing new, innovative and environmentally friendly control strategies, as the current measures are expensive, have side effects on non-target insects and human health, and have no definitive effect, as FD phytoplasma (FDp) is still spreading (Jarausch et al., 2021).
The best sustainable strategy to minimize damages due to pathogens or parasites is the exploitation of plant resistance or tolerance.
For arthropod-borne plant pathogens, plant resistance can exploit its activity against the pathogens or against insect vectors. Resistance against insects occurs when plant structural or chemical traits impair herbivore feeding and thus minimize the amount of herbivore damage experienced by the plant, while tolerance occurs when plant traits reduce the negative effects of herbivore damage on crop yield (Mitchell et al., 2016). Resistance against pathogens is the host ability to limit pathogen multiplication, while tolerance is the host ability to reduce the effect of infection on its fitness regardless of the level of pathogen multiplication (Pagán & García-Arenal, 2018). These definitions of the same terms against two different targets largely overlap. Indeed, resistance deters (insects) or limits (pathogens) the presence of the non-self being, while tolerance is the ability of the plant to live with it.
In the present work, vitellogenin expression was also included to describe possible differences attributable to the grapevine varieties. In fact, the vitellogenin support during embryo development is well-known, as its abundance in oocytes of most insect species (Sappington, 1998) and its involvement in immunity regulation (Amdam et al., 2004). Recently, more functions were highlighted for this protein, such as in planta immunity suppressor effector (Ji et al., 2021), target for viruses internalization (He et al., 2021), transmission (Huo et al., 2018) and transovarial transmission (Huo et al., 2014), transovarial carrier of immune priming signals (Salmela et al., 2015) and bacterial symbionts (Mao et al., 2020).
With the aim of identifying sources of resistance or tolerance to FD phytoplasmas within the grapevine germplasm, two works have been conducted in France and Italy (Eveillard et al., 2016;Ripamonti et al., 2021). However, only very few information on the resistance/ tolerance of the tested grapevine genotypes towards the insect vector S. titanus is available. Eveillard et al. (2016)  a FD-tolerant variety. These preliminary hints suggest that the impact of grapevine genotype on vector fitness is worthy of investigation with the aim of understanding if the mechanism underlying the reduced susceptibility to FD acts against the phytoplasma or its vector. Therefore, to gain information on the resistance/tolerance mechanism and to test the hypothesis that different susceptibilities to FDp of some cultivars might be vector-mediated, a study on S. titanus fitness on the selected varieties was conducted here. Three Vitis vinifera varieties were chosen among the extremes of the FD tolerance range (Ripamonti et al., 2021), considering both their tolerance to FDp and the impact on S. titanus short-term survival. In particular, Barbera was chosen because it is highly susceptible to FD and the leafhopper showed high survival on this variety. Brachetto was picked as a tolerant variety for FD with none/little effects on S. titanus short-term longevity. Moscato was selected as tolerant to FD with possible negative effects on insect survival (Ripamonti et al., 2021). Some key fitness parameters, such as development time, survival and fecundity, can be regarded as markers of host plant acceptability by the insect. A description of longevity and fecundity of S. titanus on Kober 5BB, a hybrid of Vitis riparia that is considered its most preferred natural plant host (Bocca et al., 2020) Here, we describe four key fitness parameters of S. titanus: nymphal developmental time and survival, adult longevity, and female prolificacy, together with the expression of vitellogenin mRNA, in insects grown on three grapevine varieties and we discuss the possible implications of these data on grapevine tolerance to FD.

| Scaphoideus titanus collection and plant rearing
Scaphoideus titanus colony was reared in greenhouse condition, starting from eggs, as described in Ripamonti et al. (2021)

| Developmental time and nymph survival
The experiments were repeated five times: twice in 2019, twice in 2020 and once in 2021. In 2019, two groups (60 and 99) of S. titanus first instar nymphs were collected from the main rearing, one at the beginning of July and one at the beginning of September. In 2020, two groups of 210 first instar nymphs were collected, one at the end of May and one in mid-August. In 2021, a group of 450 first instar S. titanus nymphs were collected at the end of April. Nymphs were randomly subdivided and equally assigned to each cultivar treatment. Nymphs were left growing undisturbed and checked every day for the presence of newly emerged adults. As soon as adults emerged, they were collected, and sex and day of emergence were recorded. The total number of nymphs exposed to each cultivar treatment was the same (343 nymphs). The non-emerged nymphs were counted as dead during the development and included in the nymph survival analysis.

| Adult survival
Two batches (33 and 70) of fourth/fifth instar nymphs S. titanus were collected from the main rearing, one in summer 2018 and one in summer 2019, and maintained in a separate cage on broadbean until adult emergence. Newly emerged adults were collected twice, the day of the first emergences (day 0) and 2 days later (day 2), subdivided per sex, randomly assigned to one cultivar-treatment; the same ratio of males/females was caged on the three cultivars.
Survival status was recorded every day, from the beginning of the test up to the death of the last insect. Dead insects were removed from the cage and discarded.
Three more replicates were conducted, two in 2020 and one in 2021, with 67, 66 and 219 newly emerged adults, respectively. The newly emerged adults derived from nymphs developed on the three previously assigned cultivars.
The overall amount of newly emerged adults per cultivar was 215 for Barbera, 150 for Brachetto and 90 for Moscato, due to the differences in survival of the nymphs on the different cultivars during the development.

| Prolificacy
The main focus of the experiment was the estimate of S. titanus female prolificacy, as measured by counting the number of mature eggs and quantifying vitellogenin gene expression. Additional data were also acquired related to nymphal mortality, nymphal developmental time and adult survival.
In 2020, two groups (made of 210 insects each) of first instar S. titanus nymphs were collected from the main rearing, one at the end of May and one at mid-August. In 2021, a group of 450 first instar S. titanus nymphs were collected at the end of April. The nymphs were randomly subdivided and equally assigned to one cultivar treatment. They were left developing undisturbed until they reached the adult stage. Adults emerged from the same cultivar were grouped per day of emergence on the same cultivar, on a different branch, using a net cage (30 × 10Ø cm). Sex ratio was maintained at 1:1 or with an excess of males in every net cage. In case of absence of males due to protandry (Chuche & Thiéry, 2012) for the 'cultivar-day of emergence' combination, adult males were taken from the main rearing. Insects' survival status was recorded twice per week, during both nymphal and adult stages. At 14, 25 or 35 day post-emergence, females were sampled from every cultivar, their abdomen dissected, and eggs counted. Adults were left undisturbed until the scheduled day of dissection. As already mentioned, different sets of data were obtained in this experiment on the three cultivars, besides the one on female prolificacy: nymphal mortality, developmental time of nymphs and adult survival. Females were taken from the rearing of the adult survival for abdomen dissection and egg counting; they were defined as 'censored' and so considered in the analyses. Males remained alone in the net cage were eventually moved to a different branch, if needed, thus considered censored too. Dissections were conducted under a stereomicroscope (Leica S9E, Deutschland), females were CO 2 anaesthetized and then the abdomen dissected with two entomological needles in a 50 μl drop of PBS 1×. Only mature eggs were counted. Eggs were considered mature when elongated and with a curved tapering apex, as explained in Bocca et al. (2020). After egg counting, the single dissected female was collected, transferred in a 1.5 ml Eppendorf tube with the same buffer and stored at −80°C until RNA extraction.

| RNA extraction and gene expression
Total RNAs were extracted from single S. titanus females following dissection and egg count, with Direct-zol RNA Mini Prep kit (Zymo Research), following manufacturer's protocol and including the optional DNAse treatment step. Concentration, purity and quality of extracted RNA samples were analysed in a Nanodrop ND-1000 spectrophotometer (Thermo Fisher Scientific).
Quantitative RT-PCR (qRT-PCR) was used to quantify the possible effect of the cultivar on the expression of female vitellogenin mRNA (Table 1), in order to correlate the vitellogenin expression level with egg count. The vitellogenin sequences were retrieved from S. titanus transcriptome (Abbà et al., 2022). For each sample, cDNA was synthesized from total RNA (250 ng) with random hexamers using a High-Capacity cDNA Reverse Transcription Kit (Applied Biosystems). The resulting cDNA was used as a template (1 μl) for qPCR in a 10 μl volume mix, containing 1× iTaq Universal Sybr Green Supermix (Bio-Rad) and 300 nM of each primer. All the primer pairs used for qRT-PCR are listed in Table 1.
Samples were run in triplicate in a CFX Connect Real-Time PCR Detection System (Bio-Rad). Cycling conditions were as follows: 95°C for 3 min and 40 cycles at 95°C for 30 s and 60°C for 60 s of annealing/extension step. The specificity of the PCR products was verified by a post-amplification melting curve analysis for all samples.
No-template controls (water devoid of cDNA) were included in the plates. Primers targeting glutathione S-transferase and elongation factor-1α were used as housekeeping genes to normalize the cDNA among samples (Table 1). Normalized expression levels (ΔΔCq) of the target gene for each sample was calculated by CFXMaestro™ Software (Bio-Rad). The stability of the expression of reference genes was validated in a multiplate gene study using the M-value (Vandesompele et al., 2002) prov by the above-mentioned software (Supporting Information S1).

| Statistical analysis
All the statistical analyses were conducted on R software v 4.  Bates et al., 2015).
A Linear Mixed Model (LMM) was applied on number of mature eggs (package nlme, Pinheiro et al., 2022), adding cultivar and day of dissection as fixed effects, and experimental replicate as random effect (Supporting Information S5). LMM performances were analysed with the 'performance' package, without raising any concern on the TA B L E 2 Total number of dead nymphs and emerged adults in the experiments   R packages used for analyses and production of figures were survival (Therneau, 2021), survminer (Kassambara et al., 2020), ggplot2 (Wickham, 2016) and patchwork (Pedersen, 2019).
The complete R code will be made publicly available on GitHub (https://github.com/matte o-rpm/papers), while the original datasets are available on OSF (Ripamonti, 2022).

| Nymph survival
The total number of emerged and dead nymphs is reported in

| Developmental time
Our results highlighted substantial differences between S. titanus developmental times measured on the three Vitis varieties, and also between males and females (Table 3). In particular, as evidenced from the Kaplan-Meier curves ( Figure 1) and the following pairwise comparisons (Table 3, Supporting Information S3), the well-known S. titanus protandry (Bocca et al., 2020;Chuche & Thiéry, 2012) was confirmed. Moreover, a considerable cultivarrelated effect was present ( Figure 1, Table 3). Leafhoppers developing on Barbera emerged significantly earlier than those on Brachetto and Moscato. A notable effect of the cultivar-related delay in the development was appreciable even between females emerged on Barbera and males on Brachetto and Moscato. The cultivar effect was higher than the impact of protandry, since Barbera females emerged significantly earlier than Brachetto and Moscato ones (Figure 1, Table 3).

| Adult survival
Summary data of adult survival are reported in Table 4. Adult longevity showed considerable differences in the survival probability for S. titanus reared on the three cultivars ( Figure 2,  Note: The areas were calculated as approximated integrals of every function (describing the interaction Cultivar × Sex) using trapezoidal rule integration. Moscato was considered ( Table 5). Focusing on the Barbera, a significant difference in the number of mature eggs per female was found. As expected, the number F I G U R E 4 Scaphoideus titanus female prolificacy on three grapevine varieties. Number of mature eggs found in ovaries after dissection (upper row, light grey boxplots), and normalized expression of vitellogenin gene (lower row, dark grey boxplots), according to the day postemergence at which females were sampled (columns). The x-axis reports the cultivar in which females were reared. The total number of dissected females (and prepared for RNA extraction) at 14, 25 and 35-day post-emergence is: 20, 16 and 18 for Barbera; 11, 11 and 9 for Brachetto; 3, 4 and 1 for Moscato. Post hoc comparisons, only for the number of mature eggs (upper row) were conducted with least-square means method and Tukey's method for p-value adjustment, at significance level as 0.05 and 95% confidence intervals, and represented by letters for every specific group (Supporting Information S5) The total number of eggs at every sampling time in Barbera females was significantly higher than in Moscato females. Brachetto-and

| Prolificacy and vitellogenin gene expression
Moscato females showed similar amounts of mature eggs at all dpe.
Interestingly, a delay in egg maturation was evident in Brachettoand Moscato-reared females, compared with Barbera ones.
The vitellogenin expression was similar among females from the three groups (Supporting Information S6). The relation between the number of mature eggs and vitellogenin relative transcription of the same female was investigated (Supporting Information S7).
Maximum vitellogenin expression was reached when the female beared ~10 eggs.

| DISCUSS ION
Vitis vinifera cv Barbera is highly susceptible to FD, while Brachetto, Merlot and Moscato show some degree of resistance (Eveillard et al., 2016;Ripamonti et al., 2021), and preliminary observations on Moscato have suggested that this behaviour may result from cultivar-specific effects on vector fitness (Ripamonti et al., 2021).
In this work, we showed that the ampelophagous Scaphoideus titanus leafhopper performs better on Barbera variety compared with Brachetto and Moscato, both at nymphal and adult stages. Actually, all the fitness parameters selected for the study, nymphal developmental time, nymphal mortality, adult longevity and female prolificacy, point out better performances on Barbera.
In particular, egg-to-adult developmental time on Moscato was delayed compared with Barbera, suggesting that this latter variety is more acceptable for the leafhoppers at the nymphal stage. Indeed, delayed development is, in general, an index of negatively impacted fitness Munyaneza & Upton, 2005), and, for example, Lobesia botrana reared on non-preferred grapevine varieties shows delayed development (Moreau et al., 2006 Incidentally, males developed faster than females, thus proving that protandry occurs in this species, as already noticed by Chuche and Thiéry (2012) and Bocca et al. (2020).
Adult leafhoppers also differed in their survival rate on the three cultivars, and S. titanus reared on Moscato lived significantly less than on the other two varieties.  (Bocca et al., 2020). A delayed production of egg, coupled with reduced longevity, results in a lower population rate of increase (Birch, 1948  , clearly indicating that Barbera, very susceptible host of FDp, is also a much-liked host for the vector, while, among the FD-tolerant varieties, Moscato is definitively a non-preferred host. Resistance/tolerance to FDp may result not only from a direct plant response against the phytoplasma, but also against the vector, as low numbers of visiting insect vectors, with less efficient feeding in the phloem, may explain the low incidence of this phloem-limited pathogen in some grapevine varieties. Identification of the genetic traits underlining antibiosis, as shown in this work, and antixenosis ('modification of herbivore behavior by plant factors, which results in the inability of a plant to serve as a host' (Kogan & Ortman, 1978;Kordan et al., 2019), as shown in Ripamonti et al. (2022), should be

CO N FLI C T O F I NTE R E S T
The authors have no competing interests to declare that are relevant to the content of this article.

DATA AVA I L A B I L I T Y S TAT E M E N T
The data that support the findings of this study are available on