OENO One Metagenomic ecosystem monitoring of soft scale insects and mealybug communities 1 Australian Wine Research Institute, Urrbrae, SA 5064, Australia 2 School of Agriculture, Food and Wine, The University of Adelaide, SA, 5064. Australia Ward ChrisM. 1 Onetto CristobalA. 1 Heuvel StevenVanDen 1 Borneman RobynDixon1andAnthonyR. 1 2 * * corresponding author: anthony.borneman@awri.com.au Abstract

Soft scale insects and mealybugs are phloem-feeding Hemipterans that are considered major pests in agriculture and horticulture throughout the world. However, correct taxonomic identification in the field can be difficult, making it hard for growers to implement control strategies. In viticulture, soft scale insects are a major issue due to their ability to secrete honeydew, which facilitates the development of sooty mould, and their propensity for being transmission vectors of several viral diseases of grapevine. To facilitate the rapid identification and quantification of vineyard-associated insects a metagenomic-based bioinformatic pipeline (MitoMonitor) was developed for generalised ecosystem monitoring, which automated the assembly and classification of insect mitochondrial genomes from shotgun sequencing data using the Barcode of Life Database API. The proof-of-concept application of MitoMonitor on metagenomic data obtained from eight samples from South Australian vineyards led to the identification of Parthenolecanium corni (European fruit scale)—which was thought to be absent in Australian vineyards—as the dominant coccoid species across the samples, with less frequent, and also lower abundance of Pseudococcus viburni (obscure mealybug) and Pseudo. longispinus (long-tailed mealybug). In addition, parisitoidism by Coccophagus scutellaris (Aphelinidae) wasps was also detected. The discovery of Parth. corni as a member of scale communities in these samples has significant implications for the development of effective control strategies for this important group of pests in affected areas.

Keywords: grapevine, scale, mealybug, metagenomics, DNA barcoding

 Introduction

Worldwide, the family Coccidae (Coccomorpha) contains more than 1100 species classified as 163 genera, although only 80 species have been described as occurring in Australia (García Morales et al., 2016). Likewise, of the 2143 described Pseudococcidae (Coccomorpha), only 210 have been detected as contemporary Australian populations (García Morales et al., 2016; Moir, 2021).

Several coccids have been shown to affect the grapevine Vitisvinifera , including various species within the genera Ceroplastes, Coccus, Cryptinglisia, Eulecanium, Mesolecanium, Neopulvinaria, Neolecanium, Parasaissetia, Parthenolecanium, Pseudokermes, Pulvinaria, Saissetia and Trijuba (Ben-Dov, 1977; Ben-Dov and Hodgson, 1997; Rakimov et al., 2013; Rakimov et al., 2015). Among these, Parth. persicae (European peach scale), Parth. pruinosum (frosted scale), and Coccus hesperidum (brown soft scale) are frequently encountered in Australian vineyards, while Co. longulus (long brown scale), Parasaissetia nigra (black scale), and Saissetia species have also been observed at a lower frequency (Rakimov et al., 2013).

Recent surveys conducted in Australia to assess disease pressure and the distribution of Coccids indicate that soft scale insects are rapidly becoming a significant pest of grapevines (Essling, 2018; Venus, 2017). The distribution of soft scale insects and mealybugs appears to be uneven across different grape-growing regions, with Western Australia and Queensland showing less frequent infestations than other states (Rakimov et al., 2013).

While severe infestations of soft scale insects and mealybugs can have direct detrimental effects on vine growth through the depletion of nutrients (Schulze-Sylvester et al., 2021), the main negative impacts of infestation manifest themselves through secondary effects. Firstly, the secretion of sugar-rich honeydew as a by-product of soft scale insect and mealybug feeding facilitates the growth of sooty mould throughout the vine, which can lead to quality reductions and/or total loss of the grape crop (Ben-Dov and Hodgson, 1997).

In addition to promoting sooty mould, soft scale insects and mealybugs also represent known or suspected vectors of grapevine ampelo- and viti-viruses, including grapevine virus A (GVA) (Hommay et al., 2008) and grapevine leafroll-associated viruses (GLRaV) 1 and 3 (Bahder et al., 2013; Engelbrecht and Kasdorf, 1990; Sforza et al., 2003). Infection by these viruses can have negative effects on yield and fruit quality, including reduced accumulation of sugars, increased acidity and irregular flavour profiles (Goheen and Cook, 1959; Naidu et al., 2014). The mobile nymph stages of both types of insects are thought to transmit these viruses between adjacent vines, although recent studies indicate that passive aerial dispersal of the nymphs can promote long-distance dispersal throughout a vineyard (Barrass et al., 1994; Hommay et al., 2019).

Studies have demonstrated the existence of several natural enemies of coccids in Australia, including parasitoid wasps, beetles, moth larvae, and lacewings (Rakimov et al., 2013; Rakimov et al., 2015). Therefore, chemical control strategies for coccids are only recommended in severe infestation cases to avoid harming the beneficial parasitoids that inhabit the vineyards (Frank, 2012; Rakimov et al., 2015b).

Ecosystem monitoring utilising molecular and/or next generation sequencing methodologies have been used to successfully characterise beetle species richness (Crampton-Platt et al., 2015) and airborne fungal diversity (Banchi et al., 2018), and to track pollinator populations (Harper et al., 2023). However, successful species classification generally requires reliable and well populated genetic databases to first be developed to carry out high-throughput monitoring (Crampton-Platt et al., 2015). Targeted ecosystem monitoring was previously carried out by Rakimov et al. (2013), utilising a COX1 amplicon to identify individual soft scale insects that infested grapevines across several Australian vineyards. The results of this work led to the conclusion that Parth. persicae was the dominant soft scale insect present, with secondary infestations commonly observed by Parth. pruinosum.

In order to facilitate the rapid identification and quantification of vineyard-associated insects, an untargeted ecosystem monitoring protocol was developed and its utility was demonstrated by determining the composition of insect populations associated with soft scale insect infestation. Shotgun metagenomic sequencing was employed on population scrapings from vines in four wine regions of South Australia subject to soft scale insect infestation. To enable the high-throughput analysis of the large datasets produced by this approach, a rapid pipeline for assembling and classifying the insect mitochondrial genomes present was established, taking advantage of the Barcode of Life Database (BOLD) API (Ratnasingham and Hebert, 2007).

 Materials and methods 1. Collection and DNA extraction and sequencing

In 2022, samples were collected from eight different vineyards in South Australia (Table 1) by taking scrapings containing multiple individuals of soft scale insects from infested vines. The choice of vines for sampling was based on visible soft scale insect infestation; a row within each vineyard was randomly selected and scrapings were taken every 2–3 bays alternating between the left and right sides of the row. The scrapings were sampled from several parts of the vine (trunk, shoots, backs of leaves and bunches) to ensure different life stages could be captured. All the scales from one vine were put in the same tube in ethanol to preserve tissue. DNA isolation was carried out by grinding scrapings taken from one plant per vineyard using a Geno/Grinder 2010 (SPEX), followed by extraction using the DNEasy Plant Kit (Qiagen).

Sequencing libraries were prepared using the Illumina DNA library kit and sequenced on an Illumina NovaSeq 6000 using 2 x 150bp chemistry (Ramaciotti Centre for Genomics, University of New South Wales, Sydney, Australia). Quality control was carried out on short read data using fastqc (

https://www.bioinformatics.babraham.ac.uk/projects/fastqc/

) and ngsReports (Ward et al., 2020b).

All sequencing data can be obtained from NCBI under the BioProject accession PRJNA986325.

Table 1. Assembly statistics for each metagenomic sample.
titre du tableau
Sample Location Contigs Total length (kb) Longest contig N50 Full mitochondrial genomes Unique insect taxa
7361 Langhorne Creek 452 425.6 44283 955 4 6
7372 McLaren Vale 305 212.5 10753 747 0 3
7373 McLaren Vale 963 787.4 12427 1023 1 2
7374 Langhorne Creek 910 735.4 15563 1078 1 1
7375 Adelaide Hills 873 717.9 15198 1124 1 2
7376 Adelaide Hills 1015 786.6 15297 900 1 1
7377 Barossa Valley 940 689.8 6523 861 1 2
7378 Barossa Valley 981 763.3 15429 911 1 2

2. Genome assembly and analysis

Mitochondrial genome assembly was first carried out by mapping raw sequence reads to all available and complete insect mitochondrial genomes (n = 12880) in the NCBI GenBank using BWA-mem (Li, 2013) and sorted using SAMtools (Li et al., 2009); all unmapped reads were removed. Read IDs of mitochondrial mapped reads were used to extract putative mitochondrial reads from the original fastq files using seqtk subseq (Li, 2012). Putative mitochondrial reads were then assembled using megahit under default settings. Contigs were remapped back to the RefSeq insect mitochondrial genome database using minimap2 (Li, 2018) with the ‘splice' alignment preset. Contigs that were successfully mapped to the insect mitochondrial database were extracted using seqkit subseq (Shen et al., 2016). Each mitochondrial contig was then annotated independently using MITOS (Bernt et al., 2013) with the insect mitochondrial genetic code and the MITOS reference database.

COX1 sequences were extracted from each of the contig annotations and provided as a seed sequence to NOVOplasty (Dierckxsens et al., 2017) to carry out circularised assembly of the mitochondrial genome using the raw short reads. Mitochondrial genes, rRNAs, and tRNAs were then annotated within each circularised genome using MITOS (Bernt et al., 2013) with the insect mitochondrial genetic code and MITOS reference database.

 3. Species assignment and phylogenetics

Taxa identification was carried out using two methods. First, COX1 sequences were extracted from all putative mitochondrial contigs and queried against the ‘COX1' BOLD database in a highly parallelised manner using the BOLD.R (Mudalige, 2021) and futures parallelisation (Bengtsson, 2020) R packages to pass sequences to the BOLD API (Ratnasingham and Hebert, 2007). Second, mitochondrial contigs were queried against the RefSeq Insect mitochondrial genome database from above using BLASTn (Camacho et al., 2009).

COX1 sequences identified as Parthenolecaniumcorni using the BOLD database were extracted using geaR (Ward et al., 2020a) and aligned to the two COX1 amplicon loci available for this genus using MAFFT (Katoh and Standley, 2013) and maximum likelihood phylogenies were constructed using the best fit model argument in IQ-TREE (Nguyen et al., 2015).

To provide further evidence that Parthenolecanium sp. from the Langhorne Creek scraping were Parth. corni, theindividuals were visually assessed for taxonomic characteristics, including dorsal tubular duct density and tubular duct presence on the dorsum of the anal cleft, which according to Gill (1988), differentiates Parth. pruinosum from Parth. corni.

 4. MitoMonitor pipeline

The MitoMonitor pipeline was written utilising the snakemake workflow framework and can be accessed at github.com/AWRI/MitoMonitor.

 Results and Discussion 1. Highly parallelised, automated assembly of insect mitochondrial genomes from metagenomic samples using MitoMonitor

Ecosystem monitoring using high throughput genomic methods is a powerful tool to deconvolute the composition of mixed population samples (Banchi et al., 2018; Crampton-Platt et al., 2015; Harper et al., 2023). The transformation of shotgun sequencing data into estimated species and abundance requires a multi-stage procedure of genome assembly, annotation and species assignment. To facilitate the automated and reproducible implementation of these bioinformatic workflows, the MitoMonitor pipeline was developed. MitoMonitor is a snakemake implementation that extracts putative mitochondrial reads from short read data and assembles them utilising a metagenomic assembly method in megahit. Putative mitochondrial contigs are then compared to a database of all the insect mitochondrial genomes available on NCBI to remove nuclear and non-insect mitochondrial fragments. Insect mitochondrial contigs are automatically annotated using the invertebrate genetic code, circularised and classified to taxonomic origin utilising information present in databases from both NCBI and BOLD (Figure 1).

Figure 1. Workflow schematic of the MitoMonitor pipeline implemented in snakemake.

Figure 1. Workflow schematic of the MitoMonitor pipeline implemented in snakemake.

2. Application of MitoMonitor to investigate insect communities associated with soft scale infestation of grapevine

To investigate the composition of insect populations associated with soft scale insect infestation, shotgun metagenomic sequencing was employed on population scrapings from areas of soft scale insect infestation from vines across four wine regions within South Australia, which was then passed through the MitoMonitor pipeline. Individual grapevines were surveyed for soft scale insect infestation across eight vineyards in Langhorne Creek, Adelaide Hills, Barossa Valley and McLaren Vale (Table 1), with an average of 28.9 ± 14 Gb of sequencing data produced across the samples.

MitoMonitor confidently classified an average of 2.5 ± 0.78 M reads as having a mitochondrial origin, which were retained for subsequent assembly and analysis. Assembled putative mitochondrial contigs ranged in size from 0.22 to 44.2 kb, and an average of 804 ± 270 mitochondrial contigs were recovered per metagenomic assembly (Table 1). Mitochondrial annotation identified a COX1 coding sequence present in 4.9 % of all mitochondrial contigs, suggesting many contigs are likely fragments derived from the same species (Table 1).

Classification of COX1 against BOLD identified seven Insect taxa across all samples, confidently (≥ 0.95 probability) assigning six to species level. COX1 sequences classified to any insect lineages were then utilised as seed sequences for circularised mitochondrial genome assembly, resulting in circularised complete assemblies for four unique taxa (Table 1). Across the eight vineyard samples, complete mitochondrial genomes, containing 12 protein-coding genes and 2 rRNA arrays, were recovered for Parth. corni (15.5 kb), Pseudo. viburni (15.2 kb), Anatrachyntis badia (Lepidoptera)(15.4 kb) and Corticarina gibbosa (Coleoptera)(16.1 kb). Whereas partial mitochondrial genomes were recovered for Coccophagus scutellaris (Hymenoptera) (6.5 kb), and Pseudo. longispinus (0.88 kb).

When considering individual samples, multiple Hemipteran pest species were identified within the same scraping, suggesting that soft scale insect and mealybugs can co-occur in close proximity to each other (Figure 2). Only DNA from a single insect species, predicted to be Parth. corni, was identified across all 8 locations (Figure 2), with all 8 samples sharing 100 % DNA identity across the mitochondrial COX1 locus. Three of the sampled sites were also predicted to contain DNA from Pseudococcus species, with two samples being predicted to contain Pseudo. viburni (7361, Langhorne Creek; 7372, McLaren Vale),while the third was predicted to contain mitochondrial DNA from Pseudo. longispinus (7373, McLaren Vale) (Figure 2).

Previous studies have highlighted the presence of several natural enemies of coccids in Australia, including parasitoid wasps, beetles, moth larvae, and lacewings (Rakimov et al., 2013; Rakimov et al., 2015). Mitochondrial genome fragments were recovered for two distinct Hymenopteran taxa from three different locations. The parasitoid wasp Coccophagus scutellaris was identified in the samples from the Barossa Valley, both of which contained Parth. corni as the only other predicted insect taxon. A single mitochondrial fragment with best BLAST hits against multiple Chalcidoid wasp mitochondrial genomes was also identified in the 7361 Langhorne Creek sample, although the exact species associated with this sequence remains unknown due to gaps in the available databases that link taxonomic assignments to appropriate genetic information.

The MitoMonitor ecosystem monitoring process is therefore capable of determining the abundance of specific soft scale insect and mealybug species, while detecting insect species that can act as potential biocontrol agents for these pests. As such, this pipeline could be used as an important biodiversity monitoring tool for assessing the impact of various biocontrol strategies, although a complete description of all potential species is limited by the completeness of reference databases. Therefore, collaboration between taxonomists and molecular biologists is required to increase the power of databases and broaden the analytical power of the methodology.

In addition to mitochondrial genomes from mealybugs, soft scale insects and their potential parasitoids, complete mitochondrial genomes were recovered for the insect species Anatrachyntis badia (Florida Pink Scavenger)and Corticarina gibbosa (a species of Minute Brown Scavenger Beetle) in sample 7361 from Langhorne Creek, as well as a partial mitochondrial genome from an unknown species of Liposcelis (bark louse)in scrapings taken from the Adelaide Hills (7375), Langhorne Creek (7361) and McLaren Vale (7372). This would suggest that these species may also be living within or be associated with colonies of soft scale insect and mealybug; however, due to the limited number of samples, the full scope of ecological diversity across the vineyards is likely to have been underestimated.

Figure 2. Abundance estimates of insect species present in Australian vineyard samples.

Insect abundance estimates based on shotgun sequencing reads mapped against assembled mitochondrial fragments that were classified as insect species, either using the BOLD API or by BLASTn against insect NCBI GenBank mitochondrial genomes. Species abundance is presented as the log2 of total read depth per million reads in each sample. A full table of these values can be found in Table S1.

Figure 2. Abundance estimates of insect species present in Australian vineyard samples.

Insect abundance estimates based on shotgun sequencing reads mapped against assembled mitochondrial fragments that were classified as insect species, either using the BOLD API or by BLASTn against insect NCBI GenBank mitochondrial genomes. Species abundance is presented as the log2 of total read depth per million reads in each sample. A full table of these values can be found in Table S1.

3. Evidence for Parthenolecanium corni in South Australian vineyards

While Parth. corni was described in the Australian taxonomic record from a single set of samples in 1976 (Belbin et al., 2021), since then its presence has not been recorded and it was not previously considered to be an active pest species in Australia. Instead, all subsequent taxonomic studies on the distribution of soft scale insects in Australian vineyards have identified Parth. persicae and Parth. pruinosum as the only species of Parthenolecanium observed (Rakimov et al., 2013). Given the ubiquitous detection of Parth. corni in the metagenomic samples, additional phylogenetic analysis was carried out on the COX1 sequence against datasets derived from BOLD and NCBI to confirm the MitoMonitor classification.

Phylogenetic reconstruction utilising all available sequences for the Parthenolecanium genus across the COX1-5P region revealed that the Parth. corni COX1 7361 haplotype was monophyletic, with all other publicly available COX1 sequences for Parth. corni and a maximum genetic distance of 0.9 % across the accessions (Figure 3). Furthermore, the Parth. corni 7361 haplotype generated by MitoMonitor was identical to sequences obtained from Australian (GBMNA27795-19), Chinese (GBMNC41505-2), and Chilean (GBMIN67815-17, GBMIN67813-17, GBMHH24788-19, and GBMHH24786-19) populations of Parth. corni (haplotype H2, Table 2). The two Parth. persicae haplotypes present in BOLD were separated by a single base difference. However, both haplotypes shared a genetic distance of 9.1 % from the Parth. corni 7361 COX1 haplotype (Table 2). In comparison, the single publicly available Parth. pruinosum COX1-5P, which was also identical to several sequences from isolates of Parth. corni, differed by 0.18 % from the 7361 COX1 haplotype.

Table 2. Genetic distance (%) between the Parthenolecanium COX1 haplotypes.
titre du tableau
Parth. corni 7361 Parth. corni H1 Parth. corni H2 Parth. corni/ pruinosum Parth. persicae H1 Parth. persicae H2 Parth. persicae Parth. pruinosum Parth. corni/ pruinosum*
COX1-5P 0.94 0 0.18 9.1 9.1
COX1-central 9.1 1.8 0

*Haplotype from “unusual” Parth. pruinosum isolated from Gumeracha South Australia.

While the COX1-5P fragment is the defacto standard for molecular-based phylogenetic reconstruction in insects, technical issues necessitatedthe use of an alternate marker from within the central region of the COX1 mitochondrial gene (COX1-central) in an earlier study by (Rakimov et al., 2013). It was therefore not possible to compare the COX1-central data with the curated sequences within BOLD (Ratnasingham and Hebert, 2007).

As MitoMonitor provided complete mitochondrial genomes for the predicted Parth. corni samples in this study, it was possible to construct a second phylogeny that encompassed sequences form the COX1-central locus produced in Rakimov et al. (2013), in addition to a single, independent COX1-centralsequence from Parth. corni (NCBI:AB439534.1) (Figure 4). As seen for COX1-5P, the Parth. corni 7361 COX1 haplotype was highly divergent (9.1 %) from the sequences associated with Parth. persicae (Rakimov et al., 2013), clearly distinguishing the two species at the molecular level.

Figure 3. Phylogenetic reconstruction of the COX1-5P locus.

Phylogenetic reconstruction of the Parth. corni 7361 haplotype against publicly available Parthenolecanium datasets across the COX1-5P universal locus sourced from Barcode of Life Database (BOLD). Publicly available data are denoted by their BOLD accessions.

Figure 3. Phylogenetic reconstruction of the COX1-5P locus.

Phylogenetic reconstruction of the Parth. corni 7361 haplotype against publicly available Parthenolecanium datasets across the COX1-5P universal locus sourced from Barcode of Life Database (BOLD). Publicly available data are denoted by their BOLD accessions.

However, unlike the COX1-5P region, which could not differentiate Parth. corni and Parth. pruinosum, the 7361 haplotype was shown to be identical to the Parth. corni AB439534.1 accession, but distinct (1.8 % difference) from the majority of Parth. pruinosum samples (Table 2). The exception to this clear species delineation was the presence of sequences from two isolates from Gumeracha, South Australia (KC784922.1 and KC789423.1), which has previously been flagged as representing taxonomically-divergent Parth. pruinosum (Rakimov et al., 2013).

Figure 4. Phylogenetic reconstruction of the COX1-central locus.

Phylogenetic reconstruction of the Parth. corni 7361 haplotype against publicly available Parthenolecanium datasets utilising the COX1-central locus (Rakimov et al., 2013). Samples of Parth. pruinosum that displayed unusual taxonomic and lifecycle traits from Rakimov et al. (2013) are highlighted in pink. Publicly available data are denoted by their NCBI GenBank accessions.

Figure 4. Phylogenetic reconstruction of the COX1-central locus.

Phylogenetic reconstruction of the Parth. corni 7361 haplotype against publicly available Parthenolecanium datasets utilising the COX1-central locus (Rakimov et al., 2013). Samples of Parth. pruinosum that displayed unusual taxonomic and lifecycle traits from Rakimov et al. (2013) are highlighted in pink. Publicly available data are denoted by their NCBI GenBank accessions.

In addition to providing information from both COX1 loci (5P and central), the availability of a complete mitochondrial genome for the Parth. sp. identified in this studyallowed the two phylogenies to be directly compared. By combining the associated metadata information from both topologies it is likely that the Parth. corni species designation given to the 7361 haplotype by the BOLD classification (which includes vouchered museum specimens and sequences from the Centre for Biodiversity Genomics) is likely to be correct. Furthermore, it also indicates that both Gumeracha sequences are likely to represent Parth. corni,rather than “an unusual” representation of Parth. pruinosum (Rakimov et al., 2013).

While the failure to observe Parth. corni as a historical pest within Australian vineyards may be due to a recent biological incursion, it may also be a case of misidentification due to complications within the Parthenolecanium taxonomic record, as several studies question the validity of current species limits (Danzig, 1997; Gill, 1988; Nakahara, 1981). Specifically, Gill (1988) stated that Parth. pruinosum was indistinguishable in the field from Parth. corni if the powdery wax was absent, which is generally the case when dealing with dead and/or dried individuals. However, a lack of tubular ducts on the dorsum, including on the anal cleft, can be used as a taxonomic character to delineate Parth. corni and Parth. pruinosum. Accordingly, the distinct lack of tubular ducts along the dorsum of individual soft scale insects isolated from the South Australian vineyard scrapings (Figure S1), further supports the notion that the sequence data obtained from the metagenomic datasets were derived from Parth. corni.

The discovery of Parth. corni as a common component of soft scale communities within these South Australian vineyards is consistent with recent observations of multivoltine reproduction within soft scale populations in many Australian viticultural regions (Venus, 2017). First described in California (Coquillett) Parth. pruinosum is generally described as univoltine (García Morales et al., 2016; Gill, 1988; Ueda et al., 2008), as is Parth. persicae,the other major soft scale insect species previously reported in Australian vineyards (García Morales et al., 2016; Gill, 1988). In contrast, Parth. corni is frequently described as multivoltine (García Morales et al., 2016), especially in warmer climates (Canard, 1960; Gill, 1988).

Whether the presence of Parth. corni is due to misidentification, incursion, or a combination of both, it has significant implications for the control of soft scale insects within Australian viticulture: current agrochemical applications, which target soft scale insect nymphs, are based on the control of strictly univoltine species (Parth. persicae and Parth. pruinosum), in which nymphs are present at a single defined time each year. The multivoltine nature of Parth. corni dictates that a spread of life stages exists throughout the year, likely rendering univoltine-based control strategies far less effective, and which could potentially explain the increasing prevalence of serious soft scale insect infestations in Australian vineyards.

 Conclusion

Soft scale insects are a major pest in global viticulture due to their ability to facilitate the development of sooty mould and their propensity for being transmission vectors of grapevine viral diseases. The development of the metagenomic-based ecosystem monitoring approach described in this study has resulted in a proof-of-concept to detect and quantify major pest species, together with key parasitoids and other associated insect species. The discovery of Parth. corni as a significant member of infesting soft scale insects in South Australia corroborates empirical observations of soft scale insects with multivoltine lifestyles; it thus has significant implications for the development of effective control strategies for this important group of pests in these areas.

 Acknowledgements

This work was supported by the AWRI and by Australian grapegrowers and winemakers through their investment body Wine Australia, with matching funds from the Australian Government. The AWRI is a member of the Wine Innovation Cluster (WIC) in Adelaide. Special thanks to Marcel Essling for valuable discussions regarding control measures for soft scale insects in the Australian Viticultural industry, as well as to the grapegrowers that provided access to their vineyards and/or assisted with sample collection.

Bahder B. , Poojari S. , Alabi O. , Naidu R. , Walsh D. , 2013. Pseudococcus maritimus (Hemiptera: Pseudococcidae) and Parthenolecanium corni (Hemiptera: Coccidae) are capable of transmitting grapevine leafroll-associated virus 3 between Vitis x labruscana and Vitis vinifera. Environmental entomology, 42(6), 1292-1298. https://doi.org/10.1603/EN13060. Banchi E. , Ametrano C. G. , Stanković D. , Verardo P. , Moretti O. , Gabrielli F. , Lazzarin S. , Borney M. F. , Tassan F. , Tretiach M. , 2018. DNA metabarcoding uncovers fungal diversity of mixed airborne samples in Italy. PloS one, 13(3), e0194489. https://doi.org/10.1371/journal.pone.0194489. Barrass I. , Jerie P. , Ward S. , 1994. Aerial dispersal of first-and second-instar longtailed mealybug, Pseudococcus longispinus (Targioni Tozzetti) (Pseudococcidae: Hemiptera). Australian Journal of Experimental Agriculture, 34(8), 1205-1208. https://doi.org/10.1071/EA9941205. Belbin L. , Wallis E. , Hobern D. , Zerger A. , 2021. The Atlas of Living Australia: History, current state and future directions. Biodiversity Data Journal, 9. https://doi.org/10.3897/BDJ.9.e65023. Ben-Dov Y. , 1977. Taxonomy of the long brown scale Coccus longulus (Douglas), stat. n.(Homoptera: Coccidae). Bulletin of Entomological Research, 67(1), 89-95. https://doi.org/10.1017/S0007485300010919. Ben-Dov Y. , Hodgson C. J. , 1997. Soft Scale Insects. Elsevier. Bengtsson H. , 2020. A unifying framework for parallel and distributed processing in R using futures. arXiv preprint arXiv:2008.00553. https://doi.org/10.32614/RJ-2021-048. Bernt M. , Donath A. , Jühling F. , Externbrink F. , Florentz C. , Fritzsch G. , Pütz J. , Middendorf M. , Stadler P. F. , 2013. MITOS: improved de novo metazoan mitochondrial genome annotation. Molecular phylogenetics and evolution, 69(2), 313-319. https://doi.org/10.1016/j.ympev.2012.08.023. Camacho C. , Coulouris G. , Avagyan V. , Ma N. , Papadopoulos J. , Bealer K. , Madden T. L. , 2009. BLAST+: architecture and applications. BMC bioinformatics, 10, 1-9. https://doi.org/10.1186/1471-2105-10-421. Canard M. , 1960. Recherches sur la morphologie et la biologie de la cochenille eulecanium corni bouché (Homoptères-Coccoideoe). . Crampton-Platt A. , Timmermans M. J. , Gimmel M. L. , Kutty S. N. , Cockerill T. D. , Vun Khen C. , Vogler A. P. , 2015. Soup to tree: the phylogeny of beetles inferred by mitochondrial metagenomics of a Bornean rainforest sample. Molecular biology and evolution, 32(9), 2302-2316. https://doi.org/10.1093/molbev/msv111. Danzig E. M. , 1997. 1.1. 3.5 Intraspecific variation of taxonomic characters. In World Crop Pests (Vol. 7, pp. 203-212). Elsevier. https://doi.org/10.1016/S1572-4379(97)80051-6. Dierckxsens N. , Mardulyn P. , Smits G. , 2017. NOVOPlasty: de novo assembly of organelle genomes from whole genome data. Nucleic acids research, 45(4), e18-e18. https://doi.org/10.1093/nar/gkw955. Engelbrecht D. , Kasdorf G. , 1990. Transmission of grapevine leafroll disease and associated closteroviruses by the vine meatybug, Planococcus ficus. Phytophylactica, 22(3), 341-346. Essling M. , 2018. Understanding factors leading to sooty mould. Wine Australia. Retrieved 16/06/2023 from. Frank S. D. , 2012. Reduced risk insecticides to control scale insects and protect natural enemies in the production and maintenance of urban landscape plants. Environmental entomology, 41(2), 377-386. https://doi.org/10.1603/EN11230. García Morales M. , Denno B. D. , Miller D. R. , Miller G. L. , Ben-Dov Y. , Hardy N. B. , 2016. ScaleNet: a literature-based model of scale insect biology and systematics. Database, 2016. Gill R. , 1988. The scale insects of California: Part 2. The Soft Scales (Homoptera: Coccoidea: Coccidae). California Dept. of Food & Agriculture, Sacramento, CA, 132. Goheen A. , Cook J. A. , 1959. Leafroll (red-leaf or rougeau) and its effects on vine growth, fruit quality, and yields. American Journal of Enology and Viticulture, 10(4), 173-181. https://doi.org/10.5344/ajev.1959.10.4.173. Harper L. R. , Niemiller M. L. , Benito J. B. , Paddock L. E. , Knittle E. , Molano‐Flores B. , Davis M. A. , 2023. BeeDNA: Microfluidic environmental DNA metabarcoding as a tool for connecting plant and pollinator communities. Environmental DNA, 5(1), 191-211. https://doi.org/10.1002/edn3.370. Hommay G. , Komar V. , Lemaire O. , Herrbach E. , 2008. Grapevine virus A transmission by larvae of Parthenolecanium corni. European Journal of Plant Pathology, 121, 185-188. https://doi.org/10.1007/s10658-007-9244-3. Hommay G. , Wiss L. , Chadoeuf J. , Le Maguet J. , Beuve M. , Herrbach E. , 2019. Gone with the wind: Aerial dispersal of Parthenolecanium corni crawlers in a newly planted grapevine plot. Annals of Applied Biology, 174(3), 372-387. https://doi.org/10.1111/aab.12505. Katoh K. , Standley D. M. , 2013. MAFFT multiple sequence alignment software version 7: improvements in performance and usability. Molecular biology and evolution, 30(4), 772-780. https://doi.org/10.1093/molbev/mst010. Li H. , 2012. seqtk Toolkit for processing sequences in FASTA/Q formats. GitHub, 767, 69. Li H. , 2013. Aligning sequence reads, clone sequences and assembly contigs with BWA-MEM. arXiv preprint arXiv:1303.3997. Li H. , 2018. Minimap2: pairwise alignment for nucleotide sequences. Bioinformatics, 34(18), 3094-3100. https://doi.org/10.1093/bioinformatics/bty191. Li H. , Handsaker B. , Wysoker A. , Fennell T. , Ruan J. , Homer N. , Marth G. , Abecasis G. , Durbin R. , Subgroup G. P. D. P. , 2009. The sequence alignment/map format and SAMtools. Bioinformatics, 25(16), 2078-2079. https://doi.org/10.1093/bioinformatics/btp352. Moir M. L. , 2021. Coextinction of Pseudococcus markharveyi (Hemiptera: Pseudococcidae): A case study in the modern insect extinction crisis. Austral Entomology, 60(1), 89-97. https://doi.org/10.1111/aen.12506. Mudalige N. , 2021. BOLD. R: A Software Package to Interface with BOLD Through R. Recent Developments in Mathematical, Statistical and Computational Sciences: The V AMMCS International Conference, Waterloo, Canada, August 18–23, 2019. https://doi.org/10.1007/978-3-030-63591-6_18. Naidu R. , Rowhani A. , Fuchs M. , Golino D. , Martelli G. P. , 2014. Grapevine leafroll: A complex viral disease affecting a high-value fruit crop. Plant disease, 98(9), 1172-1185. https://doi.org/10.1094/PDIS-08-13-0880-FE. Nakahara S. , 1981. The proper placements of Nearctic soft scale species assigned to the genus Lecanium Burmeister (Homoptera: Coccidae). Proceedings of the Entomological Society of Washington, 83(2), 282-286. Nguyen L.-T. , Schmidt H. A. , Von Haeseler A. , Minh B. Q. , 2015. IQ-TREE: a fast and effective stochastic algorithm for estimating maximum-likelihood phylogenies. Molecular biology and evolution, 32(1), 268-274. https://doi.org/10.1093/molbev/msu300. Rakimov A. , Ben‐Dov Y. , White V. , Hoffmann A. A. , 2013. Soft scale insects (H emiptera: Coccoidea: Coccidae) on grapevines in A ustralia. Australian Journal of Entomology, 52(4), 371-378. https://doi.org/10.1111/aen.12039. Rakimov A. , Hoffmann A. , Malipatil M. , 2015. Natural enemies of soft scale insects (H emiptera: Coccoidea: Coccidae) in A ustralian vineyards. Australian Journal of Grape and Wine Research, 21(2), 302-310. https://doi.org/10.1111/ajgw.12134. Ratnasingham S. , Hebert P. D. , 2007. BOLD: The Barcode of Life Data System (http://www. barcodinglife. org). Molecular ecology notes, 7(3), 355-364. https://doi.org/10.1111/j.1471-8286.2007.01678.x. Schulze-Sylvester M. , Corronca J. A. , Paris C. I. , 2021. Vine mealybugs disrupt biomass allocation in grapevine. OENO One, 55(1), 93-103. https://doi.org/10.20870/oeno-one.2021.55.1.4458. Sforza R. , Boudon-Padieu E. , Greif C. , 2003. New mealybug species vectoring Grapevine leafroll-associated viruses-1 and-3 (GLRaV-1 and-3). European Journal of Plant Pathology, 109(9), 975-981. https://doi.org/10.1023/B:EJPP.0000003750.34458.71. Shen W. , Le S. , Li Y. , Hu F. , 2016. SeqKit: a cross-platform and ultrafast toolkit for FASTA/Q file manipulation. PloS one, 11(10), e0163962. https://doi.org/10.1371/journal.pone.0163962. Ueda S. , Quek S.-P. , Itioka T. , Inamori K. , Sato Y. , Murase K. , Itino T. , 2008. An ancient tripartite symbiosis of plants, ants and scale insects. Proceedings of the Royal Society B: Biological Sciences, 275(1649), 2319-2326. https://doi.org/10.1098/rspb.2008.0573. Venus J. , 2017. Scale in Vineyards – Identification and Control. Wine Australia. Retrieved 18/06/2023 from https://www.wineaustralia.com/getmedia/ac2d8790-432e-4f3d-9824-d509b9e903a0/SA-Central-2016_17-AOP-Activity-2-Scale-in-Vineyards-Final-REPORT. Ward C. M. , Ludington A. J. , Breen J. , Baxter S. W. , 2020a. Genomic evolutionary analysis in R with geaR. bioRxiv, 2020.2008. 2006.240754. https://doi.org/10.1101/2020.08.06.240754. Ward C. M. , To T.-H. , Pederson S. M. , 2020b. ngsReports: a Bioconductor package for managing FastQC reports and other NGS related log files. Bioinformatics, 36(8), 2587-2588. https://doi.org/10.1093/bioinformatics/btz937.