Prevalence of the endoparasitoid Verrallia aucta in Central Scotland: A potential biocontrol for Philaenus spumarius (Hemiptera, Aphrophoridae), an important vector of Xylella fastidiosa

Xylella fastidiosa is a bacterial plant pathogen transmitted by xylem‐feeding insect vectors such as the meadow spittlebug, Philaenus spumarius (Hemiptera, suborder Auchenorrhyncha, family Aphrophoridae). Although X. fastidiosa is not currently found in the UK, methods for reducing vector populations will play an important role in controlling the spread of disease in the event X. fastidiosa is introduced. The endoparasitoid big‐headed fly, Verrallia aucta Fallén 1817 (Diptera, Pipunculidae), is a parasite specific to Aphrophoridae and has the potential to be used as a biological control agent; however, there is limited information available on its current presence and prevalence in Scotland. This study used Sanger sequencing and morphology to identify V. aucta within Central Scotland. Additionally, four species of adult Auchenorrhyncha collected during Summer 2021 from sites across Central Scotland were molecularly screened for presence of V. aucta parasitism. Sex and habitat prevalence in V. aucta parasitism in P. spumarius was also examined. Findings from this study demonstrate V. aucta is present in Scotland and specifically parasitizes Aphrophoridae. Parasitism rates of P. spumarius ranged between 17% and 57%, with both male and female adult P. spumarius parasitized equally and rates differing according to habitat. Together, this study improved our knowledge and understanding of the parasitism rate, distribution and nature of V. aucta parasitism in Scotland. Further work is required to strengthen our understanding of the host–parasite relationship and take the first steps towards using V. aucta as a natural biocontrol agent to protect plants against transmission of X. fastidiosa.


| INTRODUC TI ON
Xylella fastidiosa is a bacterial plant pathogen, transmitted via xylemfeeding insect vectors, that invades xylem vessels of plant hosts.
Once inside the xylem vessel, the bacteria create a biofilm to adhere to the vessel walls and produce cell wall-degrading enzymes to invade the pit membranes of the vessel (Castro et al., 2021;Ingel et al., 2019).Bacterial invasion and colonization also induce the plant to overproduce tylose outgrowths in the vessel, causing further xylem blockage and inhibiting transportation of water and soluble nutrients in the plant resulting in leaf scorching, stunting and reduction in size and quality of fruit (Sun et al., 2013).A range of plant species are affected by the bacteria, including crop plants such as grapevines, citrus, olive, oleander, almond and coffee, and trees such as oak, elm and maple (EFSA et al., 2021;Godefroid et al., 2019).The bacterium was first linked to Pierce's disease in California's grapevines in the 1970s and has since been identified as the cause of several other severe diseases, such as Citrus Variegated Chlorosis and Coffee Leaf Scorch (Purcell, 2013;Rapicavoli et al., 2017).Most recently, the bacterium was introduced to Europe (Trkulja et al., 2022) and there have been detections within Southern France including Corsica (Chauvel et al., 2015), mainland Spain (Moralejo et al., 2019), the Balearic Islands (Olmo et al., 2021) and, most notably, the bacterium was proven responsible for Olive Quick Decline Syndrome (OQDS) in Italy (Saponari et al., 2013(Saponari et al., , 2019)).The economic consequences of OQDS are estimated to be in the range of 5.2 billion euros over the next 50 years, not to mention the associated environmental and societal impacts (Schneider et al., 2020).Trading of infected plant species like olive and oleander to the UK poses the biggest risk of introducing X. fastidiosa to the UK (Bickle, 2019).The bacterium may lie dormant in infected plant tissues under cooler temperatures and climate change could increase minimum winter temperatures that would facilitate its spread (Bragard et al., 2019).Therefore, although X. fastidiosa has not been detected in the UK and the current climate may limit its potential spread and establishment (Giménez-Romero et al., 2022), an outbreak would have serious impacts on many plant-related activities, businesses and the wider environment (Broadmeadow et al., 2019).
Spittlebugs, or froghoppers (order Hemiptera, suborder Auchenorrhyncha, superfamily Cercopoidea, family Aphrophoridae), are xylem fluid-feeding insects which have been identified as important vectors transmitting X. fastidiosa.In particular, the meadow spittlebug, Philaenus spumarius L., has contributed to the spread of X. fastidiosa in Europe due to its wide geographical distribution and highly polyphagous feeding behaviour (Cornara et al., 2016(Cornara et al., , 2017;;EFSA, 2015;Saponari et al., 2014).Due to the devastating impacts of P. spumarius facilitating X. fastidiosa outbreaks in Southern Europe and the potential severity of an introduction or outbreak of X. fastidiosa in the UK, methods for controlling or reducing the number of vectors will play an important role in controlling the spread of disease throughout Europe.The need for control measures that safeguard crops and human health, alongside protecting biodiversity and the environment, mean the use of targeted biocontrol measures are more desirable than the use of insecticides.Ideally, the biocontrol agent would be a natural enemy, specific to the target pest or vector species and is native to the area being deployed.The endoparasitoid big-headed fly, Verrallia aucta Fallén 1817 (Diptera, Pipunculidae), has recently been studied as a strong candidate to control P. spumarius populations in Northern Italy (Molinatto et al., 2020).Indeed, Whittaker (1973) speculated that the pest status of P. spumarius in the USA on strawberries was partially due to the large populations that developed in the absence of the parasitoid.It is not the first Pipunculidae to be considered as a biocontrol.Research into the control of the potato leafhopper, Empoasca fabae (Harris), a major pest of alfalfa in North America, included rearing of the big-headed fly species Chalarus for potential release to reduce populations (Jervis, 1992).
Pipunculidae, or big-headed flies, are exclusive parasites of leafhoppers and planthoppers (order Hemiptera, suborder Auchenorrhyncha) with greatest diversity and numbers found in forest openings and along forest edges (Kirk-Spriggs & Sinclair, 2021).
Each Pipunculidae species generally parasitizes specific groups or families and V. aucta is no exception having only been observed targeting Aphrophoridae (Coe, 1966;Whittaker, 1969).Similar to its primary host P. spumarius, the parasitoid has a wide geographical distribution and has been recorded throughout Europe (de Meyer, 1992).More detailed studies on V. aucta populations in the UK were carried out by Whittaker; but were limited to two geographic sites located in Northern (Pennines) and Southern (Berkshire) England (Whittaker, 1969(Whittaker, , 1971)).As a result, there are few records available on the abundance of V. aucta in different habitats other than general observations such as: 'among brambles edging woodland' and 'swarms at Chippenham Fen' (Coe, 1966;Kehlmaier, 2006).
The studies by Whittaker (1969) and Molinatto et al. (2020) did, however, reveal that V. aucta life cycle is synchronous to the life cycle of P. spumarius and Neophilaenus lineatus, and spittlebug adults are parasitized shortly after emergence.In the UK, adult V. aucta were observed to be present from June onwards coinciding with the emergence of adult Aphrophoridae (Whittaker, 1973).Waloff (1975) and de Meyer and de Bruyn (1984) also demonstrated that the first occurrence of adult Pipunculidae coincided with emergence of adult Aphrophoridae from spittle.Studies agree that V. aucta parasitized both sexes but there is variation in the observed ratios of parasitized males to females (Molinatto et al., 2020;Whittaker, 1973).Female adult P. spumarius are sterile when parasitized with V. aucta but are not directly killed by the parasitoid until near the end of their life cycle (Molinatto et al., 2020).Therefore, any reduction of populations of P. spumarius will not be seen until the next year.
Together, these previous findings suggest V. aucta has potential as a natural control agent for reducing X. fastidiosa vector populations.Firstly, to determine if there was any variation in life cycle and emergence of adult V. aucta in Scotland due to climatic differences, we collected adult big-headed flies from Edinburgh, UK, through the Summer of 2021 and identified the species both morphologically and by molecular barcoding.Secondly, to our knowledge, no study has assessed the specificity of V. aucta parasitism in Auchenorrhyncha species using molecular techniques.To confirm specificity of V. aucta parasitism in relation to other potential X. fastidiosa vectors, adults from four species of the suborder Auchenorrhyncha (three species of Aphrophoridae and one Cicadellidae) were tested for the parasitoid.Finally, although V. aucta is distributed over most of Europe, records in the UK are sporadic and specimens are generally identified as part of general fauna surveys rather than targeted studies, meaning many records are unverified (CABI, 2019;de Meyer, 1992;NBN Atlas, 2023).
Moreover, there are substantially fewer verified V. aucta records in Scotland compared to the rest of mainland UK, with only three confirmed specimens recorded in the last 10 years (NBN Atlas, 2023).
Therefore, adult Aphrophoridae collected during July and August 2021 from several sites and different habitats in Central Scotland were screened for presence of V. aucta parasitism using the assays developed by Molinatto et al. (2020).Sampling was restricted to the central belt of Scotland as it has been identified as the highest risk area for introduction of the X. fastidiosa due to warmer accumulated temperatures in Southern Scotland (Broadmeadow et al., 2019).
Furthermore, as there is limited information available on whether V. aucta parasitism in P. spumarius is linked to habitat type, the effects of host sex and habitat on the prevalence of V. aucta parasitism were compared in P. spumarius (it being the most important X. fastidiosa vector and the most abundant Aphrophoridae species collected across all habitats).This study improved our knowledge and understanding of the parasitism rate and prevalence of V. aucta in spittlebugs and serves as a starting point to investigate use of the parasitoid as a biocontrol to protect against transmission and spread of X. fastidiosa.Cicadella viridis, a species of leafhopper within the Cicadellidae family of the suborder Auchenorrhyncha and another potential vector of X. fastidiosa (Bodino et al., 2022;Bonsignori et al., 2013), were collected with sweep nets between July and August 2021 from 11 other sites across Scotland (Figure 1).Specimens were sampled from broadleaf woodland, semi-natural grassland and heathland habitats at each of the 11 sites (see Table 2 for exceptions of two sites) (Land Cover Map, 2021 (10 m classified pixels, GB) -EIDC (ceh.ac.uk); Table S1).Each habitat hosted a variety of flora: broadleaf woodland habitats consisted of native trees with the understorey mostly covered with grasses, bramble, ferns and saplings; heathland habitats consisted mainly of Calluna vulgaris heather and Vaccinium myrtillus bilberry; grassland habitats contained grasses, fescues and associated common herbaceous plants (Table S1).One hundred and twenty sweeps were conducted per habitat.For grassland and heathland habitats, 120 sweeps were taken over a 100 metre transect.For woodland habitats, sweep netting was split evenly between the understorey (60 sweeps along a 50 metre transect) and the lower canopies of trees (four sweeps per tree for 15 individual trees).Samples were preserved in absolute ethanol for molecular analysis.

| Primers
The primers used in this study are listed in Table 1.Primers specific to the V. aucta mitochondrial partial cytochrome c oxidase subunit 1 (COI) gene and nuclear partial non-coding internal transcribed spacer 2 (ITS2) (Kehlmaier & Assmann, 2010) were designed by Molinatto et al. ( 2020) for conventional and real-time PCR, respectively.These primers were tested on V. aucta parasitized and non-parasitized P. spumarius to verify their specificity Molinatto et al. (2020).Primers targeting 18S rRNA (Mq) were used as an internal real-time PCR control, designed by Marzachi and Bosco (2005).For species identification using Sanger sequencing, universal primers targeting the COI 'Folmer' region (forward: LCO1490, reverse: HCO2198; Table 1) were used.

| Morphological identification of Pipunculidae adults
Potential adult Pipunculidae specimens were removed from the malaise trap pots for identification.Prior to DNA extraction, each specimen was examined under the microscope using the guide by Skevington and Yeates (2001) to be identified and photographed.Of the specimens examined, 12 individuals of a small size (2.0-11.5 mm) with a semi-globose to globose head composed of large compound eyes, characteristic of Pipunculidae (Kirk-Spriggs & Sinclair, 2021), were selected for sequencing.The main characteristic looked for was the M 2 vein on the wings, as this feature is a key identifier of the Verrallia genus (Skevington and Yeates, 2001).

| DNA extraction
Total genomic DNA was isolated using the Qiagen DNeasy Blood & Tissue kit (Qiagen).Each insect was pierced once in the thorax and once in the abdomen with a pin (stainless steel A2 pin (0.0056″ (14 mm) X 12.5 mm) E6872, Watkins & Doncaster) and incubated overnight in 180 μL ATL buffer and 20 μL proteinase K at 56°C.DNA was then purified following the manufacturer's instructions.DNA concentration and purity were assessed with a NanoDrop® 1000 spectrophotometer (ThermoFisher Scientific).
Nanodrop spectrophotometry showed DNA concentrations were below or approximately 40 ng/μL, therefore, dilution of DNA was not required.

| Conventional PCR for sequencing COI region
The 710 bp region of the mitochondrial cytochrome c oxidase subunit I (COI) gene (Folmer et al., 1994) was amplified using the primers LCO1490 and HCO2198 (Table 1).Each reaction comprised 10 μL of MyFi™ Mix (Meridian Bioscience); 0.2 μM of each primer and 2 μL DNA (40 ng/μL) made up to 20 μL with molecular grade water.
Thermocycling parameters used were: 3 min at 94°C, 5 cycles of 30 s at 94°C + 30 s at 45°C + 1 min at 72°C, 35 cycles of 30 s at 94°C + 1 min at 51°C + 1 min at 72°C, finishing with 10 min at 72°C.All PCR products were separated by electrophoresis in a 1% agarose gel to check the quality of the product.

| Sanger sequencing reaction
Prior to sequencing, the PCR product was purified using Exonuclease I (10 units/μL) (Applied Biosystems) and Shrimp Alkaline Phosphatase (SAP) (1 unit/μL) (Applied Biosystems).Separate forward and reverse sequencing reactions using the purified PCR product were then carried out using the BigDye Terminator v. 3.1 Cycle Sequencing Kit (Applied Biosystems).Sequencing products were prepared using ethanol precipitation, with products resuspended in Hi-Di Formamide (Applied Biosystems) and analysis carried out on a Sanger sequencing platform (3500xL Genetic Analyser, Applied Biosystems).

| Detection of Verrallia aucta-parasitized Auchenorrhyncha -Real-time and conventional PCR
To examine V. aucta parasitism specificity, real-time PCR was used to detect presence of V. aucta in three species of Aphrophoridae -P.spumarius, N. lineatus and A. alni -and in another leafhopper species, C. viridis (family Cicadellidae).To examine potential P. spumarius sex-specificity and differences in habitat prevalence of V. aucta parasitism, real-time PCR was used to detect presence of V. aucta in adult male and female P. spumarius from different habitats across the sample sites.Overall, 1148 adult P. spumarius (574 of each sex) from across habitats (Table 2), 238 adult N. lineatus, 22 adult A. alni and 30 C. viridis were analysed by real-time PCR.

| DNA extraction
Adult male and female P. spumarius were separated under a microscope.Each specimen was homogenized individually, and DNA was extracted using BioSprint® 96 DNA Blood Kits and a BioSprint® 96 workstation (Qiagen).DNA concentration and purity was determined using a NanoDrop® 1000 spectrophotometer (ThermoFisher Scientific) and diluted to approximately 40 ng/μL.The same procedure was carried out with N. lineatus, A. alni and C. viridis samples without sexing.

| Real-time PCR
Specific primers targeting V. aucta nuclear partial non-coding internal transcribed spacer 2 (ITS2), designed by Molinatto et al. (2020) for real-time PCR, were used in this study to detect parasitized vectors.18S rRNA (Mq) was used as a reference gene using a previously published primer sequence (Marzachi and Bosco, 2005) (Table 2) At the melting curve specificity stage, the temperature was lowered to 60°C for 1 min, then increased by 0.03°C every second until 95°C where it was held for 15 s.The temperature at which the DNA denatures verifies whether real-time PCR amplified one specific product.A sample tested positive for V. aucta parasitism if the average Ct value was within the range 12-33 and showed a specific melting peak (Tm) at approximately 78°C, as determined by Molinatto et al. (2020) (Figure S1).

| Real-time and conventional PCR
Statistical analyses were performed in Microsoft Excel and in R version 3.6.3(R Core Team, 2020).Figures were plotted using excel or the 'ggplot2' package within R (Wickham, 2016).The percentage number of positive samples from each of the four species of Auchenorrhyncha were calculated and compared.The 'lmerTest' R package (Bates et al., 2015;Kuznetsova et al., 2017) was used to run a generalized linear mixed model with a binomial distribution to determine whether V. aucta parasitism is sex-specific in P. spumarius by binding the total number of positive values to the total number of P. spumarius samples, with sex fitted as a categorical variable and sample site fitted as a random factorial effect.A separate generalized linear mixed model with a binomial distribution and Tukey's post-hoc test ('lsmeans' package in R; Lenth, 2016) were used to examine differences in V. aucta parasitism across habitats, with habitat fitted as a categorical variable and sample site fitted as a random factorial effect.The 'DHARMa' R package (Hartig, 2022) was used to assess model fit and deviation from normality for both models through visualizing the QQ residual plots.For plotting sex and habitat data, the percentage number of positive P. spumarius was calculated.

| Identification of adult Verrallia aucta -Morphology and Sanger sequencing
Out of the 12 Pipunculidae specimens directly captured in malaise traps and sampled for sequencing, one sample captured 14 July 21 was identified as Verrallia aucta, based on the NCBI GenBank results from the COI consensus Sanger sequence (Table 3).Nine specimens were identified as belonging to the Pipunculus genus, while the other two fly specimens were identified as belonging to the Chalarus and Eudorylas genus, respectively.Microscopic examination for the M 2 vein on the wings of the voucher specimen confirmed the molecular identification of specimen C (Figure 2), using the Skevington and Yeates (2001) taxonomic key.

| Detection of Verrallia aucta-parasitized Auchenorrhyncha -Real-time and conventional PCR
Out of the total number of P. spumarius, N. lineatus, A. alni and C. viridis collected from Central Scotland and molecularly screened for V. aucta using real-time and conventional PCR, 28.57% of P. spumarius, 24.37% of N. lineatus and 4.54% of A. alni tested positive for V. aucta parasitism while no C. viridis samples tested positive for V. aucta (Figure 3).In addition, the overall average percentage of Note: For specimens A, D, F, G and J, only a genus level identification was obtained.
For information on the number of V. aucta positive P. spumarius over the total number of P. spumarius samples within each sex, collected in each habitat from each site, see Table S2.

| DISCUSS ION
With the introduction of X. fastidiosa to Europe and the devastating impact the disease has had on Olive groves in Apulia, Southern Italy, research into methods to control its primary vector P. spumarius has increased.Although insecticides are an option to control insect pest and vectors, the use of integrated pest management, with less reliance on broad acting chemicals to control pests, is favoured.
Taking advantage of natural enemies such as predators and parasitic insects to manage pests or vectors is becoming increasingly important.This is especially useful if host-specific controls can be identified to further reduce the number of non-target organisms that could be affected.Weaver and King (1954)  and were able to successfully rear V. aucta from infected P. spumarius from larvae to adult in Northern Italy.Our results further support findings of life cycle synchrony and high level of parasitoid specificity towards X. fastidiosa host vectors of the family Aphrophoridae.
Pipunculidae, or big-headed flies, although common and widespread across Europe, are rather inconspicuous and overlooked.The family is well known for parasitizing leafhoppers and spittlebugs, each species generally demonstrating a preference for particular groups or families.V. aucta specifically targets Aphrophoridae over other families and therefore is well suited to controlling vector populations.It has a synchronous life cycle with its host, spending its larval stage in the adult spittlebug, overwintering in the soil in its pupa stage and emerging as an adult at the same time as the host adult (Molinatto et al., 2020).From our trapping we found that adult Pipunculidae including the V. aucta specimen were only present in July -which correlates with findings that adult parasites emerge as the adult spittlebugs or leafhoppers emerge in the Summer (Molinatto et al., 2020).In Scotland, Aphrophoridae adults begin to emerge July to early August (Chapman et al., 2022).As adults emerge from the spittle, they are parasitized by V. aucta but are not killed until the end of their life cycle.However, they do render the host sterile, thus have an indirect effect on populations the following year.In the UK, Whittaker (1973) proposed an interaction between P. spumarius, N. lineatus and V. aucta, where each species of spittlebug acts as a reservoir for V. aucta in years where one host is less abundant due to high parasitism rates in the previous year.
The specific morphological characteristics that make V. aucta distinguishable from other pipunculid fly species have already   been identified.For instance, the distribution and length of hairs on the thorax is species-specific within Pipunculus (Skevington and Marshall, 1998).However, the characteristic that best distinguishes the Verrallia genus from Pipunculus genus is the presence of the M 2 vein on the wings of Verrallia species -which was found to be present only on the V. aucta specimen in this study (Skevington and Marshall, 1998;Skevington and Yeates, 2001; Figure 2).Further confirmation was obtained by sequencing of the COI gene region (Table 3).However, the M 2 vein allows easy identification of adult V. aucta specimens from other Pipunculidae and is useful for quick identification of specimens from the Pipunculus genus for future studies.
Using the assay developed by  (Whittaker, 1969) are higher than those recorded in Northern Italy (Molinatto et al., 2020), possibly due to differences in field sampling across studies or climatic differences discussed below.Nevertheless, it suggests that this parasitoid is a major biological control measure against X. fastidiosa vectors profoundly in the UK.
This finding is similar to Molinatto et al. (2020), who also found no difference in parasitism in male and female P. spumarius, but contrasts Whittaker (1969) who reported slightly higher parasitism rates for males (46%) than females (31%).Previous studies have shown that parasitized adult female P. spumarius have degenerate ovarioles which render them sterile (Molinatto et al., 2020).It is unknown whether parasitism has detrimental effects in male P. spumarius, for instance, on male reproductive organs.However, findings from this study indicate that V. aucta parasitizes both sexes, and future studies should examine the effects of parasitism on male P. spumarius.This will provide a greater understanding of how these parasitoids influence both vector fitness and reproductive success.
There was a significant difference in habitat prevalence of V. aucta parasitism in P. spumarius, with the greatest prevalence in grassland habitats and lowest prevalence in heathland habitats (Figure 5).Molinatto et al. (2020) suggested that abundance of V. aucta is primarily associated with the presence and distribution of its host, as V. aucta has a wide geographical distribution that varies in temperature and elevation, therefore specific factors like temperature are not severe limitations (de Meyer, 1992).Present findings may contradict this association -in a parallel seasonal variation study conducted in Loch Leven (a site used in the present study), Chapman et al. (2022) found that the heathland habitat had a larger population size (per m 2 ) of both nymph and adult P. spumarius compared to grassland and woodland habitats.Yet, in the present study, heathland had the lowest prevalence of V. aucta parasitism.As this study focussed on natural areas whereas the survey conducted in Italy was in agroecosystems with a higher amount of soil disturbance (Molinatto et al., 2020), the results are not directly comparable.Secondly, it may be that the habitat and feeding behaviours of adult V. aucta and P. spumarius do not fully overlap.Nymphs and adult P. spumarius are highly polyphagous and found in a wide range of habitats but preferentially feed on actively growing, amino acid-rich plants (Thompson, 1994).For instance, previous studies in heathlands have shown that P. spumarius are more abundant on fertilized Calluna vulgaris plants, that is, nitrogen-rich plants, than on unfertilized heather (Hartley and Gardner, 1995).Meanwhile, adult V. aucta are commonly found in wet meadows and along woodland edges and are known to feed on sugar-rich honeydew secretions (Skevington and Marshall, 1998).The apparent preference for meadow (or grassland) and woodland edges in V. aucta could therefore explain the higher prevalence in parasitism in these habitats.The varying P. spumarius population levels across habitats indicates that V. aucta parasitism may be playing a role in the differences in adult P. spumarius population levels observed in different habitats.However, it is also worth considering that P. spumarius may have a strong affinity to heathland host plants and are therefore simply more abundant in these habitats.Chartois et al. (2023) found a strong association between nymph and adult P. spumarius and the rockrose shrub Cistus monspeliensis in Corsica, suggesting that retained chemical signals in generations of P. spumarius may be influencing its preference to this host, which could also provide a possible explanation to high abundance in heathland habitats in Scotland.
Several bioecology studies investigating P. spumarius associations with host plants in Mediterranean climates have commonly found nymphs feeding on ground cover plants within the Asteraceae, Fabaceae and Apiaceae families (e.g., Antonatos et al., 2021;Bodino et al., 2020;Dongiovanni et al., 2019;Villa et al., 2020).In Scotland, P. spumarius nymphs have also been recorded to feed on similar plant hosts with an observed preference to Fabaceae (trefoils and vetches), Asteraceae (thistles), and Ranunculaceae (buttercup).In

Field
work was conducted in Summer of 2021.Malaise traps were set up near Edinburgh for collection of Pipunculidae adults in June through to August 2021, and only three pots were found to contain big-headed flies in July.Malaise trap 1 was erected on 7 July 21 and specimens A-G collected 14 July 21, malaise trap 2 erected on 14 July 21 and specimens H-J collected 21 July 21 and malaise trap 3 erected on 21 July 21 and specimens K and L collected 28 July 21.Adults of three species of spittlebug, P. spumarius, N. lineatus and Aphrophora alni, (family Aphrophoridae) and of Map of Scotland showing locations of the 11 sites (blue dots with site names) sampled for Philaenus spumarius.(b) Satellite image of Forest Wood sample site.Habitat locations sampled from Forest Wood indicated in orange diamond, from top to bottom: grassland, woodland and heathland.Images created using QGIS.org and Fores terGIS.com.[Colour figure can be viewed at wileyonlinelibrary.com] . SsoAdvanced Universal SYBR supermix (Bio-Rad) was used to detect amplification of V. aucta ITS2 in the DNA samples during realtime PCR.A reaction mastermix was made using 10 μL SYBR, 1 μL of 10 μM ITS2 forward primer, 1 μL of 10 μM ITS2 reverse primer, 6 μL nuclease-free water and 2 μL diluted DNA for a total volume of 20 μL.Real-time PCRs were also run using 18S reference gene to exclude the presence of PCR inhibitors and therefore a false negative diagnosis.A DNA extraction control (devoid of a spittlebug individual), nuclease-free water and diluted DNA of a Eudorylas pipunculid fly were used as negative controls.A positive control was initially represented by diluted plasmid containing V. aucta ITS2; however, upon detection of V. aucta ITS2 in a P. spumarius DNA sample, dilutions of this sample were thereon used as the positive control for V. aucta ITS2.The first three real-time PCR runs were performed in technical triplicate, while the remaining runs were performed in technical duplicate.The quantification stage involved denaturing DNA at 95°C for 10 minutes, followed by 40 cycles of: (1) 15 s at 95°C, (2) gradual decrease by 1.6°C per second to 58°C and (3) annealing of primers at 58°C for 1 min.The number of cycles until amplification, detected by SYBR Green fluorescence, determined the cycle thresholds (Ct).
PCRTo confirm positivity, only Auchenorrhyncha DNA samples that were below the 12th cycle (outside the lower range determined byMolinatto et al. (2020)) or displayed a Ct value that was higher than average within each qPCR plate, were tested by conventional PCR using the specific primers against V. aucta mitochondrial partial cytochrome c oxidase subunit 1 (COI) gene, designed byMolinatto et al. (2020).The same MyFi™ reaction mix (Meridian Bioscience) and thermocycling conditions described inMolinatto et al. (2020) (95°C for 5 min, 35 cycles of 30 s at 95°C + 30 s at 58°C + 1 min at 72°C, finishing with 10 min at 72°C) were used.Detection of DNA fragments of approximately 300 bp from gel electrophoresis indicated positive samples and these samples were also included in the final dataset for analysis(Molinatto et al., 2020).2.5 | Statistical analysis2.5.1 | Sanger sequencingCOI sequences obtained from the pipunculid flies collected from malaise traps in Edinburgh were imported onto Geneious Prime (Geneious Prime® 2022.0.2,Biomatters).The chromatograms were combined to create a single consensus file.Using the BLAST function in Geneious, taxonomic ID, bit score and percentage identity were obtained from NCBI GenBank and a species, genus or family assigned to each specimen.A species or genus identification requires a minimum 'percentage pairwise identity' of 96% or above, otherwise taxonomic identification can only be confirmed to family taxonomic level (EPPO, 2021).
first reported parasitism of P. spumarius by Ooctonus vulgatus Haliday, 1833 (Hymenoptera, Mymaridae) an egg parasitoid wasp, in North America.In addition, Mesmin et al. (2020) are so far the only researchers to have assessed its biology as a parasitoid to P. spumarius eggs in Europe and modelled the wasp's distribution throughout Europe, indicating its potential use in controlling P. spumarius across Europe, particularly in areas currently experiencing X. fastidiosa outbreaks (for instance Corsica) and therefore urgently require X. fastidiosa vector control strategies.However, the degree of host specificity of O. vulgatus and the ability to rear them in controlled conditions (in preparation for potential future introductions to X. fastidiosa affected areas) was not assessed.Other studies have focussed on introducing predators, such as the assassin bug Zelus renardii (Hemiptera, Reduviidae) (Liccardo et al., 2020); however, the release of this generalist predator could have indirect consequences on the local biodiversity.Meanwhile, Molinatto et al. (2020) demonstrated high levels of parasitism specificity towards P. spumarius with respect to its synchronous life cycle F I G U R E 2 Difference in wing morphology of (a) Verrallia aucta, specimen C and (b) fly specimen belonging to the Pipunculus genus, specimen K.In Figure 2a, the M 2 vein is present on the wing; in contrast, the M 2 vein is absent in Figure 2b.Both microscope images were taken during this study.[Colour figure can be viewed at wileyonlinelibrary.com]F I G U R E 3 Total number of Verrallia aucta positive Philaenus spumarius, Neophilaenus lineatus, Aphrophora alni and Cicadella viridis in comparison to the sum total of samples per species.[Colour figure can be viewed at wileyonlinelibrary.com] Total number of P. spumarius, total number of V. aucta positive P. spumarius and percentage positive P. spumarius in each sex and habitat.F I G U R E 4 Percent Verrallia aucta positive Philaenus spumarius males (blue, left) and females (pink, right; n = 62).Each box represents the median, lower 25th and upper 75th percentile, with whiskers showing the values within 1.5 times the interquartile ranges.Outside values are >1.5 times and <3 times beyond the interquartile ranges.[Colour figure can be viewed at wileyonlinelibrary.com]TA B L E 5 Mixed model explaining differences in percentage Verrallia aucta positive Philaenus spumarius in each habitat (baseline = Grassland, n = 31).

F
Percent Verrallia aucta positive Philaenus spumarius in grassland (green, left), heathland (purple, middle) and woodland (orange, right) habitats (n = 31).Different letters indicate significant differences between habitats (p < 0.05) (Tukey's post-hoc test): V. aucta parasitism of P. spumarius in grassland is significantly different to V. aucta parasitism in heathland (a vs. b).V. aucta parasitism of P. spumarius in grassland is significantly different to V. aucta parasitism in woodland (a vs. c).V. aucta parasitism of P. spumarius in heathland is significantly different to V. aucta parasitism in woodland (b vs. c).Each box represents the median, lower 25th and upper 75th percentile, with whiskers showing the values within 1.5 times the interquartile ranges.Outside values are >1.5 times and <3 times beyond the interquartile ranges.[Colour figure can be viewed at wileyonlinelibrary.com] contrast to Mediterranean climates, very few adult P. spumarius are observed migrating into tree canopies in Scotland, likely due to there being sufficient ground flora all year round.Understanding different X. fastidiosa vector plant host ecology within differing climates and landscapes could further help our understanding of its parasitoid, V. aucta, and whether it prefers or indeed can develop more efficiently within certain climatic conditions or habitats.The study byMolinatto et al. (2020) found that V. aucta parasitism rates in vectors collected in semi-natural environments were not greater than those collected from agroecosystems (olive groves and vineyards).The higher parasitism rates found in England and Scotland could suggest this parasitoid performs better within cooler climates, but further research is needed to confirm this.As suggested byMolinatto et al. (2020), other regions in Europe where X. fastidiosa is already established should survey for the presence and abundance of V. aucta.

TA B L E 1
Primers used in this study.Number of male and female Philaenus spumarius sampled from each habitat across sample sites to test for presence of Verrallia aucta parasitism.
^Conventional PCR.†Sanger sequencing.TA B L E 2Note: At two sites, Gargunnock Estate and Glen Vale, no specimens were collected from broadleaf woodland for testing.
Identification of 12 Pipunculidae specimens using Sanger sequencing of the COI gene region.
TA B L E 3