LAMP assay for the detection of the Asian citrus psyllid, Diaphorina citri Kuwayama (Hemiptera: Psylloidea: Psyllidae)

Diaphorina citri Kuwayama, also known as the Asian citrus psyllid (ACP), can vector the bacterium Candidatus Liberibacter asiaticus (CLas), agent of Huanglongbing (HLB): an incurable disease affecting citrus trees worldwide. In citrus growing regions where ACP and HLB are absent, such as Australia, the risk of an incursion and consequent economic damage to citrus industries make this psyllid one of the top-priority pests. Due to ACP’s small dimensions and the generally poorly studied native psylloid fauna worldwide, morphological identification of this insect to distinguish it from harmless species is challenging, especially in the field, and with immature, partial or damaged specimens. To allow rapid and efficient detection of ACP in the field, we designed and optimised a new Loop-mediated isothermal amplification (LAMP) assay for the detection of D. citri based on the mitochondrial 16S locus. The optimised ACP 16S LAMP assay produced amplification from D. citri samples within 13.3 ± 3.6 min, with an anneal derivative of ~ 78.5 °C. A synthetic gBlock gene fragment was also developed to be used as positive control for the new LAMP assay with a different anneal derivative of ~ 83 °C. An existing commercially available LAMP assay for detection of the bacterium CLas was also tested in this study on ACP DNA. The ACP 16S LAMP assay we developed and tested here provides a valuable new in-field compatible tool that can allow early detections of ACP, enabling a quick biosecurity response, and could potentially be adopted by a wide range of users, from farmers to agronomists and from researchers to industry.


Scientific Reports
| (2023) 13:10895 | https://doi.org/10.1038/s41598-023-37721-w www.nature.com/scientificreports/ very similar to the markings on the wings of species of the genus Acizzia (Hemiptera: Psyllidae), a species-rich genus in Australia, where it comprises at least forty described species 2,[11][12][13] , but is also present worldwide 1,2 . Molecular assays may offer an alternative method for an efficient identification of ACP. For example, standard DNA barcoding targeting the subunit I of the Cytochrome oxidase gene (COI; 14 ) has been successfully applied to psyllids [15][16][17] . However, from collection to sequencing, this process may take several days to achieve a diagnosis, highlighting a lack of rapid, in-field compatible molecular tests for detection of ACP.
LAMP, Loop-mediated isothermal amplification 18 , is a highly target-specific, rapid method for amplification of DNA which can be conducted in the laboratory or field in portable devices producing identification results in less than one hour. Several LAMP assays have recently been developed for diagnosing priority plant pests such as Queensland fruit fly Bactrocera tryoni (Froggatt) 19 , grape phylloxera Daktulosphaira vitifoliae (Fitch) 20 , Khapra beetle Trogoderma granarium Everts 21 , and fall armyworm Spodoptera frugiperda (J. E. Smith) 22 , proving this technique can be an extremely valuable tool for in-field detection of pests. Like qPCR 23 LAMP assays can also employ gBlocks Gene Fragment (gBlock) as targeted synthetic oligonucleotides as controls to monitor assay performance 22 .
While there are at least four different LAMP assays which have been developed for the detection of CLas [24][25][26][27] , no LAMP assay has been developed for the detection of ACP, the potential insect vector, to date. Here we developed the first molecular diagnostic LAMP assay to detect the Asian citrus psyllid which has been fully optimised and tested for laboratory and in-field use.
In this study we: (1) developed a new ACP LAMP assay based on the mitochondrial 16S locus, (2) tested a broad panel of non-target psyllid species to assess the specificity of the new LAMP assay, (3) tested in-field non-destructive DNA extraction methods, (4) designed a gBlock to use as a synthetic positive control, as well as (5)  Specimens examined. All specimens examined in the current study were initially identified morphologically, based on the latest taxonomical classification 1 , and confirmed through DNA barcoding (Table 2). COI sequences of 48 specimens were generated in this study and submitted to GenBank, with accession numbers ON970381-ON970418, OP185136-OP185142, and OP804505-OP804507 (Table 2). COI sequences from 41 different species (in addition to D. citri), represent four families within the Psylloidea (Aphalaridae, Carsidaridae, Psyllidae and Triozidae) (Fig. 2). Of these, at least 29 sequences (~ 70%) belong to species that are either undescribed or previously molecularly uncharacterised.
Evaluation of ACP gBlock Gene fragment. The detection sensitivity of the 207 bp ACP gBlock was found to be sensitive, amplifying as low as ~ 1 × 10 3 copies/µL of ACP gBlock within 30 min with an anneal derivative of ~ 83 °C. One million copies (1 × 10 6 ) of gBlock amplified in 8 min, which is earlier than the amplification time of two D. citri samples which amplified in 16 and 22 min (Fig. 3a). Based on this amplification time, 1 × 10 6 copies/µL of ACP gBlock was found to be suitable for use as synthetic positive in ACP 16S LAMP assay. The anneal derivative of LAMP amplicons produced two distinct peaks (Fig. 3b), ~ 78.5 °C for D. citri DNA (see Table 1. Asian citrus psyllid (ACP) 16S LAMP primer and amplicon (gBlock) sequences and parameters. The F2 and B2 primer regions of the FIP and BIP primers are marked in bold and underlined. Lowercase letters in the gBlock indicate extra c's added between LAMP primer sites to increase the overall Tm of the amplicon. ACP 16S LAMP assay specificity results. All D. citri specimens (n = 6) tested ( Table 2, Fig. 2) produced positive amplification for the ACP 16S LAMP assay (Fig. 4a), while none of the non-target psyllid samples (n = 42), belonging to at least 18 genera and four families, amplified DNA. D. citri DNA produced positive amplification in 13.3 ± 3.6 min, with amplification within 20 min considered as positive. All six D. citri samples produced an anneal profile at ~ 78.5 °C (Fig. 4b, Table 2). None of the negative non-template controls amplified in any of the LAMP runs tested for optimising this assay confirming the absence of primer dimer and reagent contamination (Fig. 4). The non-destructive QuickExtract DNA extraction method was demonstrated in the field, using three nontarget psyllid specimens of Mycopsylla fici (Carsidaridae), and a specimen each of Agelaeopsylla sp. and Cardiaspina fiscella (both Aphalaridae) ( Table 2). None of the non-target samples and negative non-template control amplified, except for the gBlock 1 × 10 6 copies/µL which amplified in 8.25 min with an anneal derivative temperature of 82.7 °C.
Candidatus Liberibacter asiaticus LAMP assay results. None of the D. citri DNA samples (n = 6) tested for the presence of the bacterium CLas produced positive amplification (Table 3, Fig. 5a). The recommended DNA template used in the kit for the LAMP assay was 5 µL of positive DNA which amplified in 9.75 min. We found that using a lower volume of 2 µL of positive DNA increased the amplification time by only approx. 5 min at 14.25 min. The anneal derivative temperature was 85.4 °C ( Table 3, Fig. 5b). The negative non-template control did not amplify confirming that there was no contamination in the reagents and the primers performed as expected (Table 3, Fig. 5).

Discussion
Reliable identification of psyllids in the field is of paramount importance for early detection and management of pest species, such as ACP 9 . However, identification of these insects can be extremely challenging without taxonomic expertise, especially when dealing with immature stages; with many species being morphologically uncharacterised 1 or appearing very similar to one another 1,2 . Likewise, adult specimens collected in the field often have incomplete morphology, being damaged by commonly used trapping methods, such as sticky traps. Furthermore, psyllids can easily be windblown to plants that are not their host,which can lead to psyllids collected from citrus plants being incorrectly suspected of being exotic pest species 29 . Molecular techniques for the identification of psyllids can be used as valuable tools for diagnostics and surveillance. DNA-based diagnostic techniques can be used by non-taxonomists, are generally less time-consuming than morphological examination, and can be applied to partial, immature or damaged specimens.
In designing and optimising this LAMP assay for ACP, we also generated and deposited on public database DNA sequences of psyllid taxa mostly native to Australia, that can now be used for diagnostics using the DNA barcoding technique 14 . Here, we generated 48 partial COI sequences across four families within the Psylloidea. These have proven to be a valuable tool to assess the genetic diversity of the psyllid species tested in this study, clearly separating the target species, ACP, from all others. Furthermore, these COI sequences provide additional genetic information on the diversity of the Australian Psylloidea. COI sequences from 41 different species were generated in this study (in addition to ACP), contributing to the number of works that have used COI as a marker for species delimitation and biodiversity assessments for this insect group [15][16][17] . Of these, 29 sequences (~ 70%)  www.nature.com/scientificreports/ belong to species that are either undescribed or previously uncharacterised, therefore providing important genetic information for this poorly studied group 17 .
The main aim of this work, however, was to design and optimise a new LAMP assay for rapid detection of Asian citrus psyllid for both laboratory and in-field use. This test will be a useful tool in regions of the world where ACP is currently absent and diagnostic tests in the field are required within surveillance programs, but also in those areas where ACP is present, to monitor seasonal fluctuations in the populations and range expansions across different properties. While the ACP 16S LAMP assay we have developed has been optimised for use on a The scale bar is a genetic distance of 5%. All COI sequences were generated in this study, GenBank accession numbers are indicated together with the VAITC database identification number.     21,22 .
In order to make this LAMP assay functional and ready-to-use specifically in those countries or regions where ACP is not present, such as Australia, we designed and optimised a synthetic DNA positive control (gBlock) for use in ACP 16S LAMP assays (following [20][21][22] ). This synthetic DNA is beneficial in: (i) providing a consistent supply of positive control DNA as ACP is an exotic psyllid and often hard to get hold of positive samples for optimising the LAMP assay, (ii) providing a consistent control to allow tracking of the performance of LAMP assays across runs, (iii) providing a positive control that is easily differentiated from LAMP amplification of ACP insect DNA, possessing a different annealing temperature.
To further aid the use of this LAMP assay in the field, the QuickExtract non-destructive DNA extraction method can be performed using the Genie III machine. Considering the small size of ACP and different life stages of the insect, using 25 µL of QuickExtract buffer would yield higher DNA amount producing faster amplification time in the LAMP assay as shown in previous studies 20 .
In the current study, laboratory testing of the ACP 16S LAMP assay was performed only on DNA extracted from adult ACP specimens. This LAMP assay could not be tested on samples from other life stages, including eggs or immatures, as these were not available. Indeed, it is often difficult to source specimens of exotic species to use for testing new molecular assays (e.g. 28) . However, LAMP assays developed for other insects have been shown to work on all life stages 19,20 . Given the sensitivity of our LAMP assay, i.e., down to ~ 1 × 10 3 copies/µL of ACP gBlock, it is anticipated that this LAMP assay would be able to accurately identify these early life stages of ACP.
Our validation of a currently available CLas assay showed that this test performed well, and could be used with low amounts of template, demonstrating the potential sensitivity of this assay and potential cost savings associated with using lower quantities of commercially available positive control DNA. The new ACP 16S LAMP assay we developed here together with the commercial CLas LAMP test now provide complementary tools for rapid laboratory or in-field detection and management of this potentially devastating citrus pest and pathogen, using highly specific rapid molecular LAMP assays.

Materials and methods
Psyllid samples and DNA extractions. DNA extracts of D. citri (n = 6) were sourced from the Victorian Agricultural Insect Tissue Collection (VAITC) and used for the development and testing of the new LAMP assay. Psyllid species (n = 41) belonging to at least 18 genera (Table 2), mostly native to Australia, as well as some key native and exotic pest species, such as Trioza melaleucae 29 , Trioza erytreae and Bactericera cockerelli, were sourced from the Victorian Agricultural Insect Collection (VAIC). These were used to assess the genetic distances between ACP and other psyllids tested, and to test the LAMP assay's specificity. These samples were chosen to include a diverse set of species and genera, representative of the Australian psyllid fauna. DNA extractions of these psyllids were performed using DNeasy Blood and Tissue kit (Qiagen, Germany), as previously described 30 for individual insects, using a non-destructive approach (except VAITC7222d, destructive approach).
Non-destructive DNA extraction is a pre-requisite for conducting LAMP assay in the field. The LAMP assay is more sensitive and less impacted by impurities present in the DNA extract, compared to real-time PCR, and can amplify from crude DNA extracts efficiently. DNA of additional individual psyllids (n = 3), from an additional two species, was extracted in loco (Table 2), from single adult psyllids using QuickExtract DNA extraction solution 1.0 (Epicentre, USA) following the protocol presented elsewhere 19 . Marker selection, isolation and amplification. DNA barcode COI reference sequences. Based on numerous studies commonly utilising the COI barcode region for psyllid identification (e.g. [15][16][17], and the known suitability of this marker for LAMP assays (e.g. 19,20 , DNA sequences of the COI locus were obtained from ACP  Development of a 16S LAMP assay for ACP. While the COI sequences (above) were utilised to estimate genetic relatedness and are extremely useful as barcoding marker for psyllids, this region appeared to be unsuitable for development of a LAMP assay. Our initial attempt to develop a COI LAMP assay using this region (data not shown) proved unsuccessful due to the presence of extensive primer dimer interactions between LAMP primers. A second attempt to design the ACP LAMP assay was made using an alternative mitochondrial locus, the 16S gene, based on DNA sequences obtained from GenBank, outlined in 34 (Supp Fig. S1). Two ACP individuals, representing different 16S haplotypes, and the closest genetic relative Diaphorina communis were included in a DNA alignment, along with the most similar reference DNA sequences from GenBank (Fig. 1). LAMP primers ( Table 1) were developed to target a 340 bp region of the ACP 16S locus, which appears to be conserved within ACP, but shows a small number of differences from other psyllids (Fig. 1). Six new LAMP primers were manually designed by eye to target eight DNA regions from the 16S reference alignment (Fig. 1). The complete set of LAMP primers (Table 1) was analysed together to detect potential primer-dimer interactions using the Thermo Fisher Multiple Primer Analyzer tool (www. thermofisher.com). Primers were synthesized by Sigma (Australia). Each LAMP reaction mix was made by adding 10 µL of primer master mix to 14 µL of Isothermal Master Mix (ISO-004, OptiGene, UK) and 1 µL of template DNA into each well of the Genie strip (25 µL total reaction volume). Each run included a positive control (i.e., 1 × 10 6 gBlock), a no-template negative control, and six test samples. All LAMP assays were run in the Genie III at 65 °C for 25 min followed by an annealing curve analysis from 98 °C to 73 °C with ramping at 0.05 °C/s. The total run time being approximately 35 min. The run files were transferred and analysed using a PC version of the software Genie Explorer version 2.0.7.11.

ACP
Evaluation of a gBlock Gene Fragment for ACP 16S LAMP assay. A gBlock dsDNA fragment (Integrated DNA Technologies, Iowa, USA) was designed for use as synthetic DNA positive control for the ACP 16S LAMP assay. This synthetic fragment consisted solely of concatenated LAMP primers separated by runs of "ccc", to increase the overall Tm of the gBlock (Table 1). To evaluate detection sensitivity, the copy number and a tenfold serial dilution (1:10) of the gBlock was prepared as outlined in 20 . Sensitivity of the LAMP assay was tested using the serially diluted (1 × 10 8 to 1 × 10 copies/µL) of gBlock in the Genie III, following the same ACP 16S LAMP assay conditions as mentioned above, with the run time increased from 25 to 30 min, to allow detection of the gBlock present at low concentrations. Following this another LAMP run was conducted using two D. citri DNA samples to compare amplification time with a standard amount of one million copies (1 × 10 6 copies/µL) of gBlock.
Test for commercial LAMP assay kit for the Candidatus Liberibacter asiaticus (CLas) optimisation. A Commercially available CLas LAMP assay kit (Cat. No. PK-C.lib_asiaticus050W) was sourced for testing in the current study following the manufacturer's recommendations (OptiGene, UK). Each LAMP reaction mix was made by adding 5 µL of primer master mix to 15 µL of Isothermal Master Mix (ISO-004, OptiGene, UK) and 2 µL of template DNA (recommended 5 µL of DNA) into each well of the Genie strip. Both 5 µL and 2 µL of template DNA were tested in two separate Genie runs. Each run included a CLas synthetic positive control (i.e., Cat. No. CD-CLas_050), a no-template negative control, and six D. citri DNA test samples. All LAMP assays were run in the Genie III at 65 °C for 30 min followed by an annealing curve analysis from 98 °C to 73 °C with ramping at 0.05 °C/s and results analysed on the blue channel.