Introduction
⌅Auchenorrhyncha (Hemiptera) from the families Cicadellidae, Aphrophoridae, and Cercopidae transmit Xylella fastidiosa (Wells et al. (Xanthomonadales: Xanthomonadaceae) (Redak et al., 2004Redak RA, Purcell AH, Lopes JR, Blua MJ, Mizell III, RF, Andersen PC, 2004. The biology of xylem fluid-feeding insect vectors of Xylella fastidiosa and their relation to disease epidemiology. Annu Rev Entomol49: 243-270. 10.1146/annurev.ento.49.061802.123403). This bacterium has previously been linked to citrus variegated chlorosis and Pierce’s disease of grapevine in the Americas (Saponari et al., 2013Saponari M, Boscia D, Nigro F, Martelli GP, 2013. Identification of DNA sequences related to Xylella fastidiosa in oleander, almond and olive trees exhibiting leaf scorch symptoms in Apulia (southern Italy). J Plant Pathol95: 659-668.). Since 2013, X. fastidiosa has been reported in olive, oleander, and almond trees in Italy (Saponari et al., 2013Saponari M, Boscia D, Nigro F, Martelli GP, 2013. Identification of DNA sequences related to Xylella fastidiosa in oleander, almond and olive trees exhibiting leaf scorch symptoms in Apulia (southern Italy). J Plant Pathol95: 659-668.). In fact, in the Lecce Area of Apulia (Italy), X. fastidiosa subsp. pauca cause “Olive quick decline syndrome,” affecting over 8000 km2 of infected territory (Serio et al., 2024Serio F, Imbriani G, Girelli CR, Miglietta PP, Scortichini M, Fanizzi FP, 2024. A Decade after the Outbreak of Xylella fastidiosa subsp. pauca in Apulia (Southern Italy): Methodical Literature Analysis of Research Strategies. Plants13(11): 1433. 10.3390/plants13111433). The known vectors in Europa are Philaenus spumarius L., Neophilaenus campestris Fallén, and Philaenus italosignus Drosopolous and Remane (Saponari et al., 2014Saponari M, Loconsole G, Cornara D, Yokomi RH, De Stradis A, Boscia D, Bosco D, Martelli GP, Krugner R, Porcelli F, 2014. Infectivity and transmission of Xylella fastidiosa Salento strain by Philaenus spumarius L., (Hemiptera: Aphrophoridae) in Puglia, Italy. J Econ Entomol107: 1316-1319. 10.1603/EC14142; Cavalieri et al., 2019Cavalieri V, Altamura G, Fumarola G, di Carolo M, Saponari M, Cornara D, Bosco D, Dongiovanni C, 2019. Transmission of Xylella fastidiosa subspecies pauca sequence type 53 by different insect species. Insects10(10): 324. 10.3390/insects10100324).
Vector control is one of the most important measures for preventing disease spread. In this regard, previous research has shown that control strategies based on chemical insecticides and physical methods can reduce vector populations, particularly those of P. spumarius (Morelli et al., 2021Morelli M, García-Madero JM, Jos A, Saldarelli P, Dongiovanni C, Kovacova M, Saponari M, Baños Arjona A, Hackl E, Webb E, Compant S, 2021. Xylella fastidiosa in olive: A review of control attempts and current management. Microorganisms9(8): 1771. 10.3390/microorganisms9081771). However, to reduce the negative impact of the synthetic insecticides, while preserving agricultural ground cover, safer products and methods should be developed.
There is currently very little knowledge on the natural enemies affecting X. fastidiosa’s vector populations in Europe (Val Nolan, 1956Val Nolan, Jr, 1956. Spittle Insects as Food of Prairie Warblers. The Auk, 73(4): 557. 10.2307/4081955; Phillipson, 1960Phillipson J, 1960. A contribution to the feeding biology of Mitopus morio (F) (Phalangida). J. Anim. Ecol.29: 35-43. 10.2307/2269; Whittaker, 1970Whittaker JB,1970. Cercopid spittle as microhabitat. Oikos21: 59-64. 10.2307/3543839; Harper & Whittaker, 1976Harper G, Whittaker J, 1976. The role of natural enemies in the colour polymorphism of Philaenus spumarius (L.). J. Anim. Ecol.45,91-104. 10.2307/3769). Recently, some authors have analysed the control potential of certain predator groups coexisting in both space and time within particular agroecosystems, mainly in olive groves. For example, the predator hemipteran Zelus renardii Kolenati (Reduviidae) was recommended by Liccardo et al. (2020Liccardo A, Fierro A, Garganese F, Picciotti U, Porcelli F, 2020. A biological control model to manage the vector and the infection of Xylella fastidiosa on olive trees. PLoS One15(4): e0232363. 10.1371/journal.pone.0232363) as a potential candidate for controlling P. spumarius using an inundation method. Additionally, in the laboratory, Benhadi-Marín et al. (2020Benhadi-Marín J, Villa M, Pereira LF, Rodrigues I, Morente M, Baptista P, Pereira JA, 2020. A guild-based protocol to target potential natural enemies of Philaenus spumarius (Hemiptera: Aphrophoridae), a vector of Xylella fastidiosa (Xanthomonadaceae): A case study with spiders in the olive grove. Insects. 3; 11(2): 100. 10.3390/insects11020100) investigated the differences in prospective predatory effectiveness of the spiders Araniella cucurbitina Clerck and Synaema globosum Fabricius with P. spumarius as the prey. They found differences in the functional response of the two spider species, with S. globosum exhibiting higher predation efficiency in locations with elevated prey densities.
Spiders (Araneae) are among the most important predators of other arthropods in agroecosystems, and several studies have demonstrated its key effect in the natural control of phytophagous insects (Marc et al., 1999Marc P, Canard A, Ysnel F, 1999. Spiders (Araneae) useful for pest limitation and bioindication. Agric Ecosyst Environ. 74: 229-273. 10.1016/B978-0-444-50019-9.50015-7). In addition to their high consumption rates, spiders induce other effects, such as dislodging or promoting behavioural changes, that can reduce pest populations (Michalko et al., 2019Michalko R, Pekár, S, Entling MH, 2019. An updated perspective on spiders as generalist predators in biological control. Oecologia189: 21-36. 10.1007/s00442-018-4313-1). Although predation of spiders on X. fastidiosa vectors (mainly P. spumarius) in the field has been previously reported, the data available is limited and is often based on punctual observations (Phillipson, 1960Phillipson J, 1960. A contribution to the feeding biology of Mitopus morio (F) (Phalangida). J. Anim. Ecol.29: 35-43. 10.2307/2269; Harper & Whittaker, 1976Harper G, Whittaker J, 1976. The role of natural enemies in the colour polymorphism of Philaenus spumarius (L.). J. Anim. Ecol.45,91-104. 10.2307/3769; Rodrigues et al., 2022Rodrigues I, Ramos V, Benhadi-Marín J, Moreno A, Fereres A, Pereira JA, Baptista P, 2022. A novel molecular diagnostic method for the gut content analysis of Philaenus DNA. Sci Rep.12(1): 492. 10.1038/s41598-021-04422-1).
Recently, Rodrigues et al. (2022Rodrigues I, Ramos V, Benhadi-Marín J, Moreno A, Fereres A, Pereira JA, Baptista P, 2022. A novel molecular diagnostic method for the gut content analysis of Philaenus DNA. Sci Rep.12(1): 492. 10.1038/s41598-021-04422-1) reported the predation of a lynx spider of the genus Oxyopes on P. spumarius, using a PCR-based diagnostic method developed to detect generalist predators. Molecular techniques and PCR-based diagnostic methods have grown in prominence in the analysis of trophic interactions in recent years (Liu et al., 2020Liu M, Clarke LJ, Baker SC, Jordan GJ, Burridge CP, 2020. A practical guide to DNA metabarcoding for entomological ecologists. Ecological Entomology (2020), 45: 373-385. 10.1111/een.12831; Uiterwaal & DeLong, 2020Uiterwaal SF, DeLong JP, 2020. Using patterns in prey DNA digestion rates to quantify predator diets. Mol Ecol Resour20(6): 1723-1732. 10.1111/1755-0998.13231). Additionally, trophic interactions have been revealed using metabarcoding, a method that combines high-throughput sequencing (HTS) with DNA barcoding to identify food remnants in faeces or gut contents (Deagle et al., 2019Deagle BE, Thomas AC, McInnes JC, Clarke LJ, Vesterinen EJ, Clare EL, Kartzinel TR, Eveson JP, 2019. Counting with DNA in metabarcoding studies: How should we convert sequence reads to dietary data?Mol Ecol.28(2): 391-406. 10.1111/mec.14734). DNA metabarcoding of gut content has proven to be a valuable tool for characterising trophic interactions, estimating the relative importance of intraguild prey or pest species and identifying prey diets in spiders, thus revealing new insights into biological control by this group of predators (Hambäck et al., 2021Hambäck PA, Cirtwill AR, García D, Grudzinska-Sterno M, Miñarro M, Tasin M, Yang X, Samnegård, U, 2021. More intraguild prey than pest species in arachnid diets may compromise biological control in apple orchards. Basic Appl Ecol57: 1-13. 10.1016/j.baae.2021.09.006; Saqib et al., 2021Saqib, HSA, Liang, P, You, M, Gurr, GM, 2021. Molecular gut content analysis indicates the inter- and intra-guild predation patterns of spiders in conventionally managed vegetable fields. Ecol. Evol.11: 9543-9552. 10.1002/ece3.7772; Ortiz et al., 2022)
In the present study, we determined which species and specific guilds of spiders prey on vectors of X. fastidiosa in agroecosystems in southwestern Spain using a gut content metabarcoding approach.
Material and methods
⌅Sampling and selection of spiders
⌅Spider sampling was conducted in six localities across southwestern Spain (Sevilla, Huelva, and Córdoba Provinces; Table S1), focusing on the ground cover vegetation of olive, citrus, and vineyard fields. These locations were chosen because they have been sites of ongoing monitoring of aphrophorids for the past three years. Samplings were performed from 19 to 27 April 2022, coinciding with the development of spittlebug nymphs. For each locality, spiders were vacuumed for 5 min (InsectaZooKa, Bioquip Products, Rancho Dominguez, California, USA) in a randomly selected 0.25 m2 area of ground cover vegetation and stored individually in 100% ethanol. Once in the laboratory, the spiders were frozen at −20 °C until DNA extraction. Taxonomic determination to genus/species level was performed using identification keys developed by the Iberian Arachnology Group (Barrientos, 2006Barrientos, JA, 2006. Claves de los arácnidos ibéricos (documento de trabajo). Jornadas sobre taxonomía de arácnidos ibéricos. III Curso Práctico de Aracnología, Grupo Ibérico de Aracnología, Córdoba 20 al 24 de julio, 198 pp.) and a binocular stereomicroscope. Table 1 shows the spiders selected for the analysis of gut content.
Gut metabarcoding
⌅DNA isolation and metabarcoding analyses were carried out by AllGenetics & Biology SL (www.allgenetics.eu). DNA was isolated from each sample using an E.Z.N.A Tissue DNA kit (Omega Bio-tek Inc., Norcross, GA, USA). The abdomens were washed several times with 96% ethanol and dried with filter paper, to avoid external contamination. Subsequently, the DNA was isolated following the manufacturer’s instructions, including a negative extraction control (Bex1 and Bex2) in each round of DNA extractions, and treating them as regular samples to check for contamination. The DNA was then resuspended in a final volume of 50 μL and quantified using the Qubit High Sensitivity dsDNA Assay (Thermo Fisher Scientific, Alcobendas, Madrid, Spain).
For DNA metabarcoding library preparation and sequencing, a primer pair was designed to specifically amplify DNA from the following vectors and potential vectors of X. fastidiosa belonging to the order Hemiptera: P. spumarius, N. campestris, N. lineatus L., Lepyronia coleoptrata L., Aphrophora sp., Cicadella viridis L., and Cercopis sp. Geneious 11.1.5 (Biomatters Ltd, Boston, MA, USA) was used to test the primers in silico against the COI mitochondrial reference sequences (GenBank, https://www.ncbi.nlm.nih.gov/genbank/) of the prey species and the families of spiders represented in the samples. The primers were designed to minimise the amplification of spider DNA. The resulting primers amplified a fragment of around 212 bp: forward - MetArachnida_fwd (5’ GGVTCATTTATTGGRGATGATC 3’) designed by AllGenetics; reverse - MetArachnida_rev (5’ GGNGGATAWACYGTTCAWCCNGTHCC 3’), modified by AllGenetics from Leray et al. (2013Leray M, Yang JY, Meyer CP, Mills SC, Agudelo N, Ranwez V, Boehm JT, Machida RJ, 2013. A new versatile primer set targeting a short fragment of the mitochondrial COI region for metabarcoding metazoan diversity: Application for characterizing coral reef fish gut contents. Front Zool10: 1-14. 10.1186/1742-9994-10-34). These primers also included Illumina (Illumina Inc., San Diego, CA, USA) primer sequences attached to their 5’ ends. A negative control that contained no DNA (BPCR) was included in each round PCR to check for contamination during library preparation. Library size was verified through electrophoresis. The libraries were then purified using Mag-Bind RXNPure Plus magnetic beads (Omega Biotek), following the manufacturer’s instructions. The final libraries were pooled in equimolar amounts according to the results of a Qubit dsDNA HS Assay (Thermo Fisher Scientific) quantification. The pool was sequenced in a fraction (4/16) of a MiSeq PE300 flow cell (Illumina).
Prior to taxonomic assignment, a length filter for the Amplicon Sequence Variant (ASV) sequences was applied using Cutadapt to select a range of lengths (190––234 bp), based on the expected amplicon length. We conducted taxonomic assignment using the MetaCOXI reference database (Balech et al., 2022Balech B, Sandioniggi A, Marzano M, Pesole G, Santamaria M, 2022. MetaCOXI: An integrated collection of metazoan mitochondrial cytochrome oxidase subunit-i DNA sequences. Database11, 1918. 10.1093/database/baab084). The ASV sequences were compared using the feature-classifier approach classify-consensus-vsearch implemented in QIIME 2 (Bokulich et al., 2018Bokulich NA, Kaehler BD, Rideout JR, Dillon M, Bolyen E, Knight R, Huttley GA, Gregory Caporaso J, 2018. Optimizing taxonomic classification of marker-gene amplicon sequences with QIIME 2's q2-feature-classifier plugin. Microbiome6: 90. 10.1186/s40168-018-0470-z), with a sequence similarity threshold of 90%. Additionally, two more taxonomic assignments were performed: one against the GenBank database using BLASTN 2.11.9+ (Camacho et al., 2009Camacho C, Coulouris G, Avagyan V, Ma N, Papadopoulos J, Bealer K, Madden TL, 2009. BLAST+: architecture and applications. BMC Bioinformatics10: 421. 10.1186/1471-2105-10-421) with the following parameters: minimum percent identity of 90%, e-value of 1e−05, and minimum hit coverage of 80%; and another against the BOLD database using its own identification tool (BOLD Identification Engine). For all taxonomic assignments, sequences were accepted with a minimum similarity of 90%, allowing for up to a 10% mismatch. Any discrepancies were manually curated by aligning the sequence of each ASV with the best-matching reference sequences using Geneious 8.1.9. In DNA metabarcoding studies, it has been observed that a low percentage of the reads of a given library might be erroneously assigned to another library. This phenomenon, is referred to as mistagging (also tag jumping, index hopping, or index jumping). To correct for this bias, ASVs occurring at a frequency below 0.01% in each sample were removed (Illumina, 2017Illumina, 2017. Effects of Index Misassignment on Multiplexing and Downstream Analysis. 4 pp. https://www.illumina.com/content/dam/illumina-marketing/documents/products/whitepapers/index-hopping-white-paper-770-2017-004.pdf.). Despite using different and taxon-specific reference databases, some of the ASVs remained unidentified. Therefore, we removed these sequences. Finally, based on the rarefaction curves, we decided to exclude samples that contained less than 300 sequences from the final ASV table.
Statistical procedures
⌅All analyses were performed in R 4.2.1 (R Core Team, 2022R Core Team, 2022. R: A language and environment for statistical computing. R Foundation for Statistical Computing, Vienna, Austria. URL https://www.R-project.org/). Generalised linear models (GLM) with binomial error structure and a log link function were used to compare the mean number of prey taxa detected per spider specimen among crops and spider guilds. Where differences were detected using GLM, multiple comparisons and, post-hoc Tukey honestly significant difference (HSD) tests (p < 0.05) were performed using the ‘glht’ function in the ‘multcomp’ package (Hothorn et al., 2008). The DHARMa package (Hartig, 2022Hartig F, 2022. DHARMa: Residual diagnostics for hierarchical (Multi-Level / Mixed) regression models. R package version0.4.6. https://CRAN.R-project.org/package=DHARMa [11 July 2023].) was used to check models for overdispersion and residual distribution.
Results and discussion
⌅Among the 29 spiders analysed, 26 (89.7%) produced fruitful amplification. For eight specimens (Cheiracanthium sp., Cheiracanthium erraticum Walckenaer, Runcinia grammica C. L. Koch, Zelotes tenuis L. Koch, Zelotes sp., Monaeses paradoxus Lucas, Gibbaranea bituberculata Walckenaer, Pelecopsis mengei Simon), no prey were detected.
A total of 54 ASVs and 7.9×105 reads were obtained corresponding to 7 prey orders, 19 families and 30 species (Table 2). Neophilaenus campestris was detected in two spider species, namely Enoplognatha caricis Fickert and Thanatus vulgaris Simon, with 3.4 and 6.9% of the captured specimens, respectively. In addition, L. coleoptrata was preyed on by Haplodrassus macellinus Thorell, with 3.4% of the specimens. Philaenus spumarius were not detected in the samples. As for the most frequent prey, 43.3% of the total prey detected were spider species, with 23.3% hemipterans (6.7% of which were aphrophorids), and 13.3% dipterans (including two parasitoid species; Table 2). Enoplognatha species are known for their mobility, frequently move within inflorescences and can drop on a silken thread and move out of the plant either to the ground or to another plant (Greco & Kevan, 1999Greco C, Kevan P, 1999. Polyethism in foraging in a polymorphic predator, Enoplognatha ovata (Araneae: Theridiidae): a case for balance. Can Entomol131(2): 259-268. 10.4039/Ent131259-2), increasing their chances of encountering the vector nymphs, a feature rarely found in other web-building spiders. The other two spiders preying on aphrophorids (T. vulgaris and H. macellinus included in the “other-hunters” and “ground-hunters” guilds, respectively) do not build webs. Instead, they actively move across the foliage and ground in search of prey (Uetz, 1992Uetz, GW, 1992. Foraging strategies of spiders. Trends Ecol Evol7: 155-159. 10.1016/0169-5347(92)90209-T). These hunting strategies are more effective for less mobile prey, such as vector nymphs. Limited data have been previously reported regarding natural predators on European X. fastidiosa vectors. Punctual observations of birds (Val Nolan, 1956Val Nolan, Jr, 1956. Spittle Insects as Food of Prairie Warblers. The Auk, 73(4): 557. 10.2307/4081955; Whitaker, 1970Whittaker JB,1970. Cercopid spittle as microhabitat. Oikos21: 59-64. 10.2307/3543839), carabids, phalangids and spiders (Harper & Whitaker, 1976Harper G, Whittaker J, 1976. The role of natural enemies in the colour polymorphism of Philaenus spumarius (L.). J. Anim. Ecol.45,91-104. 10.2307/3769; Rodrigues et al., 2022Rodrigues I, Ramos V, Benhadi-Marín J, Moreno A, Fereres A, Pereira JA, Baptista P, 2022. A novel molecular diagnostic method for the gut content analysis of Philaenus DNA. Sci Rep.12(1): 492. 10.1038/s41598-021-04422-1) preying on aphrophorids (mainly P. spumarius) have been previously reported. Among spiders, the Araneidae, Lycosidae and Oxyopidae families have been previously identified as predators of P. spumarius (Harper & Whitaker, 1976Harper G, Whittaker J, 1976. The role of natural enemies in the colour polymorphism of Philaenus spumarius (L.). J. Anim. Ecol.45,91-104. 10.2307/3769; Rodrigues et al., 2022Rodrigues I, Ramos V, Benhadi-Marín J, Moreno A, Fereres A, Pereira JA, Baptista P, 2022. A novel molecular diagnostic method for the gut content analysis of Philaenus DNA. Sci Rep.12(1): 492. 10.1038/s41598-021-04422-1). This is the first report of predation by spiders from Philodromidae and Gnaphosidae, expanding the range of spider species and families capable of exerting control of X. fastidiosa vectors.
In this study, N. campestris was only detected in vineyards, and L. coleoptrata was detected only in citrus orchards. No aphrophorids were detected in olive orchards. “Other hunters”, “ground hunters” and “space-web” spiders preyed on aphrophorids while vectors were not detected in the guts of “ambush hunters” or “orb-web” guilds (Figure 1). The number of taxa detected in the gut content per spider varied significantly among crops (LRT χ2=17.9; p<0.001) and spider guilds (LRT χ2=25.9; p<0.001). In this context, the mean number of prey taxa per spider was significantly higher in vineyards (4.2±2.7) and citrus orchards (2.9±0.7) than in olive orchards (1.2±0.3; z=4.03 and −2.8; p<0.05). Among spider guilds, the mean number of prey taxa was lowest for “ambush hunters” (0.4±0.2) with significant differences with respect to highest obtained for “other hunters” group (3.5±1.2; z=2.9; p<0.02). Finally, the other three guilds presented no significant differences with mean number of prey taxa ranging from 1.0±0.6 and 1.0±0.7 for “orb-web” and “ground hunter” spiders, respectively, to 3.0±1.5 for “sheet-web” spiders.
Rosas-Ramos et al. (2018Rosas-Ramos N, Baños-Picón L, Tobajas E, de Paz V, Tormos J, Asís JD, 2018. Value of ecological infrastructure diversity in the maintenance of spider assemblages: a case study of mediterranean vineyard agroecosystems. Agric Ecosyst Env265: 244-253. 10.1016/j.agee.2018.06.026) documented the predominance of Philodromidae spiders in Spanish vineyards. Our results suggest that this spider family (represented by T. vulgaris) is involved in the natural control of aphrophorids in vineyards, consistent with their observed abundance. Previous research has shown that Philodromidae, Salticidae and Araneae are among the most abundant spider families in olive orchards in the south of the Iberian Peninsula (Morris et al., 1999Morris TI, Symondson W, Kid N, Campos M, 1999. Las arañas y su incidencia sobre Prays oleae en el olivar. Boletín Sanidad Vegetal Plagas25: 475-489.; Cárdenas et al., 2012Cárdenas M, Castro J, Campos M, 2012. Short-term response of soil spiders to cover-crop removal in an organic olive orchard in a Mediterranean setting. J Insect Sci.12: 61. 10.1673/031.012.6101). In addition, Rodrigues et al. (2022Rodrigues I, Ramos V, Benhadi-Marín J, Moreno A, Fereres A, Pereira JA, Baptista P, 2022. A novel molecular diagnostic method for the gut content analysis of Philaenus DNA. Sci Rep.12(1): 492. 10.1038/s41598-021-04422-1) detected P. spumarius in the gut of linx spider Oxyopes sp. in a survey carried out in olive orchards in northeastern Portugal. Our findings align with these studies, reinforcing the role of these spider families as natural predators of aphrophorids. However, our study did not detect aphrophorids in the gut content of spiders in the olive orchards of southwest Spain. This absence may be due to regional differences in prey availability or spider behaviour. To better understand these dynamics, more representative and intensive sampling-based studies are required.