Rapid and sensitive on‐site genetic diagnostics of pest fruit flies using CRISPR‐Cas12a

Abstract BACKGROUND Bactrocera zonata, a major fruit pest species, is gradually spreading west from its native habitat in East Asia. In recent years it has become a significant threat to the Mediterranean area, with the potential of invading Europe, the Americas, and Australia. To prevent it spreading, monitoring efforts in cultivation sites and border controls are carried out. Despite these efforts, and due to morphological similarities between B. zonata and other pests in relevant developmental stages, the monitoring process is challenging, time‐consuming, and requires external assistance from professional laboratories. CRISPR‐Cas12a genetic diagnostics has been rapidly developing in recent years and provides an efficient tool for the genetic identification of pathogens, viruses, and other genetic targets. Here we design a CRISPR‐Cas12a detection assay that differentially detects two major pest species, B. zonata and Ceratitis capitata. RESULTS We demonstrate the specificity and high sensitivity of this method. Identification of target pests was done using specific and universal primers on pooled samples, enabling differentiation of pests with high certainty. We also demonstrate reaction stability over time for future on‐site applications. DISCUSSION Our easy‐to‐use and affordable assay employs a simple DNA extraction technique together with isothermal amplification and Cas12a‐based detection. This method is highly modular, and the presented target design method can be applied to a wide array of pests. This approach can be easily adapted to fit local threats and requires minimal training of operators in border controls and other relevant locations, reshaping pest control and making state‐of‐the‐art technologies available worldwide, including in developing countries. © 2022 The Authors. Pest Management Science published by John Wiley & Sons Ltd on behalf of Society of Chemical Industry.


INTRODUCTION
In recent years, the peach fruit fly Bactrocera zonata (Saunders, Diptera: Tephritidae) has become a major invasive species in Africa and the Arab peninsula. Globally, it is responsible for annual losses of hundreds of millions of USD. 1,2 B. zonata is an aggressive, highly adaptable invasive species with a broad range of host plants, covering more than 50 commercial and wild plant species, mainly fleshy fruits and vegetables. 3 This fly is native to East and South-East Asia and hence well accustomed to tropical and subtropical climates. Nonetheless, it was shown to establish in colder climates reaching freezing point, enabling its proliferation in the Mediterranean climate. 4 Combined with global warming, this makes B. zonata a serious threat to West Asia, several European countries as well as parts of Australia and the Americas. 5,6 B. zonata is listed as an A1 pest in the European and Mediterranean Plant Protection Organization (EPPO) and is regulated by many EPPO member countries. 3 The ability to accurately detect different insect pests is a vital first step in the combat against the global spreading of invasive pest species. This effort typically takes place in border controls and involves the investigation of eggs and larvae found in fruits and vegetables. Morphological identification in these early life stages is a major tool used to distinguish between different species, but can be problematic due to high similarities between species, such as in Israel, with B. zonata and the much more abundant species Ceratitis capitata ( Fig. 1(a)). These similarities require specimens to be reared to adulthood for identification. Such practices can be hazardous as not all countries possess adequate quarantines. Moreover, rearing eggs is timely and often unfruitful. 7 Aside from the risks involved in rearing, during this process the entire cargo is halted and could go to waste, leading to considerable losses both to farmers and customers.
Extensive efforts have been made to facilitate faster, more accurate diagnostic techniques. Molecular methods for the identification of different Bactrocera species have been developed to aid in the battle against the spread of these harmful pests. Species-specific markers based on the mitochondrial cytochrome oxidase COII gene, 8 restriction fragment length polymorphism (RFLP) detected in a polymerase chain reaction (PCR), amplified ribosomal DNA (rDNA), 7 and high-resolution melt (HRM) real-time PCR assays 9 have been developed and are currently being used to distinguish between different pests. These methods have considerably reduced the quarantine time but still depend on the availability of laboratory space, technicians, expensive instruments, and reagents. Such methods are usually performed off-site, requiring specimen transportation to specialized facilities, thus increasing the handling time and raising the probability of spreading events.
Recently, CRISPR-Cas-based diagnostic approaches have provided a new set of accurate detection tools at the molecular level. Specifically, a type V CRISPR-Cas enzyme, Lachnospiraceae bacterium ND2006 Cas12a (LbCas12a), presents great promise in the field of in situ molecular detection of viruses, pathogens, or any DNA sequence of interest. [10][11][12][13] LbCas12a is an RNA-guided enzyme capable of recognizing double-stranded DNA (dsDNA) targets and performing a protospacer adjacent motif (PAM)-distal dsDNA break with staggered ends. 14 On detection of a dsDNA target, the enzyme is activated, and in addition to its target-specific cleavage activity it initiates indiscriminate single-stranded DNA (ssDNA) cleavage. 15,16 This attribute was found to be very effective for DNA detection. 17,18 LbCas12a-based DNA detection requires a modified ssDNA reporter, usually containing a fluorophore and a quencher to be added to the detection reaction. Once activated, the LbCas12 cleaves the ssDNA reporter, releasing the fluorophore from the quencher, leading to an easily detectable fluorescence signal ( Fig. 1(b)). The detection reaction only requires a short incubation at 37°C, and its components can be lyophilized to increase stability during storage. With minor variations, detection can be held using lateral flow strips, thus eliminating the need for a fluorescence-reading device. 13,19 Most DNA detection methods require a DNA amplification step, which might be vital if DNA material is scarce. Different methods are used to amplify the target sequence prior to detection. PCR amplification is the gold-standard method for DNA amplification and has been demonstrated to work well with the Cas12a detection system. 10,20 Two leading isothermal methods are commonly used for in situ applications: loop-mediated isothermal amplification (LAMP) and recombinase polymerase amplification (RPA). LAMP is based on a strand-displacing polymerase with four to six primers, and enables the amplification of short DNA segments (80-250 bp) in ∼30 min at 60-65°C. 21 RPA is based on an ssDNA binding protein (SSB), a recombinase that leads the SSB-primer complex to its destination, and a strand-displacing polymerase that initiates an isothermal amplification reaction. RPA requires only two primers, performs better than PCR with longer-sized primers, and offers fewer constraints on primer design, allowing higher GC content. 22 RPA reactions are 10-20 min long, are performed at 37-42°C, and offer high sensitivity even with minuscule amounts of starting material. 23 Amplification of a pool of several specimens is of great benefit since co-infestation of fruit by different Tephritidae and Drosophilidae families occurs regularly. [24][25][26] Therefore, an amplification method that allows the pooling of different specimens as well as multiplexed detection of the different targets in the pool can prove beneficial for simple in situ detection of pests.
In this work, we present a rapid and simple method to accurately differentiate between two highly hazardous pest species: B. zonata and C. capitata. We have designed an on-site assay that utilizes a simple DNA purification technique together with RPA amplification and LbCas12a detection. Our assay requires only a hot plate, a bench-top centrifuge, and a hand-held fluorometer, yet it can distinguish between the two pests with high confidence in a little over an hour. We believe this assay can aid in the battle against invasive species global spreading, avoiding quarantines and losses for farmers and consumers. Our assay is modular and can be easily expanded to other species, making it relevant for pests of concern worldwide. Notably, the affordability and simplicity of this method make it uniquely suited for pest control in developing countries.

Sample collection
All experiments were performed on samples received from the Plant Protection and Inspection Services, Israel.
2.2 gRNA design gRNAs were designed using CRISPOR. 27 Both B. zonata and C. capitata mitochondrial sequences were obtained from NCBI (accessions NC_027725.1 and NC_000857.1, respectively). Sequences were aligned using the MAFFT version 7 online tool. 28 To find target hotspots within the mitochondrial genome sequences of both insects, we looked for variable regions of up to 500 bp flanked by conserved sequences of at least 30 bp to enable universal RPA amplification. To create unique gRNAs for each insect, both variable regions were concatenated into a single long sequence, and this sequence was used in the CRISPOR gRNA search ( Fig. 2(a)). Subsequently, gRNAs were ordered from Integrated DNA Technologies (IDT, gRNA sequences provided in Table S1).

DNA extraction for gRNA testing
For purification of total DNA from insects, the Qiagen Blood & Tissue kit with the supplementary insect protocol was used (https://www.qiagen.com/kr/resources/resourcedetail?id= cabd47a4-cb5a-4327-b10d-d90b8542421e&lang=en). Specimens were homogenized and then lysed with Proteinase K for 10 min at 56°C. Next, ethanol was added and the homogenate was loaded onto DNeasy mini columns, washed, and eluted.

Chelex 100 genomic extraction
Samples were softened in 20 μL of ultrapure water (UPW) and ground using a pestle. Next, 100 μL of 1×PBS 1% saponin (Sigma-Aldrich, Cat # 8047-15-2) was added, and the samples were vortexed and incubated at room temperature for 20 min. Lysates were centrifuged for 2 min at 20 000 × g, the supernatant was removed, and 100 μL 1×PBS was added and centrifuged for 2 min at 20 000 × g. The supernatant was removed and 100 μL of 5% Chelex 100 (Bio-Rad, Cat # 1422822) was added. Lysates were vortexed for 5 s and boiled for 10 min. Next, samples were centrifuged for 1 min at 20 000 × g. The DNA remained in the supernatant, which was subsequently used for RPA amplification.

RPA amplification
RPA was performed using the TwistAmp Basic kit (Cat # TABAS03-KIT) according to the manufacturer's protocol. A mix of primers, UPW, MgOAc, and 2 μL from the Chelex reaction were prepared following the protocol concentrations. Next, a lyophilized RPA reaction was suspended with the primer-free rehydration buffer, and the primer mix was added to the reaction, followed immediately by a 20-min incubation at 37°C. Then, 10-μL reactions were similarly performed by keeping the original protocol ratios. Primer sequences can be found in Table S1.

LbCas12a cleavage assays
All reactions were prepared on ice. gRNA-LbCas12a (NEB, Cat. # M0653T) complexes were prepared by mixing 62.5 nM gRNA with 50 nM LbCas12a in 1×NEBuffer 2.1 to a final volume of 20 μL and incubated in 37°C for 30 min. Next, 1 μM FAM reporter (Table S1) and 2 μL of the RPA reaction template were added to the complexes together with 80 μL of 1× NEBuffer 2.1 and incubated for 10 min at 37°C.
Samples were transferred to a black 96-well plate and fluorescence was measured using a Tecan Spark plate-reader with an excitation wavelength of 485 nm and emission was measured at 535 nm. The gain was calibrated to 90.
The detection in the specific primer amplification experiments was performed with a lower concentration of LbCas12a (1 μM, NEB, Cat. # M0653S). Interestingly, we noticed a slight improvement in reaction stability using this product. It has previously been reported that glycerol addition can improve RPA-Cas12a reactions, which might explain our observation. 29

Protocol outline
Our protocol consists of three main steps: (a) sample collection and DNA extraction using the Chelex 100 resin, (b) an RPA amplification of specific targets, and (c) Cas12a detection. As illustrated in Fig. 1(b), the entire protocol takes roughly 1.5 h to complete, most of which consists of incubation rather than hands-on time.
The protocol requires only a benchtop centrifuge and a hot-plate. In this work, the readout was obtained using a plate reader for throughput purposes, but alternative methods such as lateralflow-based assays or handheld field fluorometers are also applicable. 13,30 www.soci.org DM Alon et al. wileyonlinelibrary.com/journal/ps

Pest-specific identification
Specific target selection for B. zonata was performed through pairwise sequence alignment of the mitochondrial genome sequences of B. zonata and C. capitata. 28,31,32 We used C. capitata as a reference genome for a nontarget fly since its larvae are morphologically very similar and cannot be easily differentiated ( Fig. 1(a)). We sought variable regions within the mitochondrial genomes of both insects and used one such region (VR1; Fig. 2(a) and Table S1) to design primers and guide RNAs (gRNAs) specific to B. zonata. First, we designed several gRNAs using the CRISPOR online platform. 27 To ensure the gRNAs are specific to our target B. zonata only, we concatenated both variable regions from B. zonata and C. capitata, and used the merged sequence as a template for CRIS-POR. We also added a linker sequence between the variable regions to avoid the selection of gRNAs derived from the stitch between the two regions. We selected the best hits and doublechecked the sequences to verify no cross-reactivity is anticipated. Based on the gRNAs position within VR1, we designed specific primers for PCR and RPA amplification (Table S1). Next, we used a commercial DNA extraction protocol (see Methods section) on B. zonata and C. capitata adult flies, amplified our target with PCR, and tested three different gRNAs (Table S1 and Fig. S1). Only one of the gRNA (Bz1) displayed a specific signal in the presence of B. zonata's genome, and it was used for assay development.
Next, we tested the full protocol on samples of both fly species. DNA was extracted from the samples using Chelex 100 resin. Chelex is a chelating resin with a high affinity for polyvalent ion metals, enabling simple and cheap DNA extraction. It has been shown to facilitate DNA extraction from cells at 100°C by chelating metal ions that act as DNase catalysts, thus preventing DNA degradation. 33 With slight modification to the protocol, Chelex 100 can be used to efficiently extract DNA from insects with as little as a single egg for starting material. 34 Following extraction, DNA was isothermally amplified using RPA and tested with Cas12a-gRNA complexes. We began by examining the ability to detect a positive and specific signal from all developmental stages of B. zonata (Fig. 2(b)). The Chelex 100 DNA extraction method proved to be highly efficient throughout all developmental stages of the flies, including single eggs, as demonstrated previously. The RPA efficiently amplified the Chelex-originated DNA material in only 20 min at 37°C. The complexes consisting of the Cas12 and the gRNAs we designed properly detected their targets and were inactive on the nontargeted sequences.
Next, we tested the sensitivity of our assay. We combined larvae from both B. zonata and C. capitata in ratios ranging from 1:1 up to 1:50 larvae, respectively (Fig. 2(c)). A robust signal was observed in all ratios, and no signal was observed in the absence of B. zonata DNA. These results suggest that pooling larvae from  (Table S1). As a negative control, C. capitata larvae were used with Bz1 gRNA (green). (c) Sensitivity of Cas12a-Bz gRNA detection in varying ratios of pooled larvae. Samples containing a single larva of B. zonata with increasing amounts of C. capitata larvae were prepared. DNA was then extracted and diagnosed as in (b), using B. zonata-specific primers and Cas12a-Bz gRNA complexes. As a negative control, C. capitata DNA was used. All experiments were performed with three biological repeats and three technical repeats. several fruits is feasible without loss of specificity, allowing the required time, costs, and labor to be reduced. We further tested for cross-reactivity between B. zonata and Drosophila melanogaster as a representative fermentation fruit fly. This test is relevant to real-life scenarios since opportunistic species such as D. melanogaster may lay eggs in damaged fruits after picking and shipment, leading to larvae in the pooled sample tube. We used the same ratios (1:1 up to 1:50) and no cross-reactivity was detected (Fig. S2).

Universal amplification assay
After validating the method using species-specific primers, we expanded the assay to include C. capitata identification using a universal RPA amplification reaction. By designing universal primers that can amplify variable regions in different fly species, we were able to produce more relevant data from a single pooled tube, while reducing the hands-on time and simplifying the protocol. We designed universal primers that match both B. zonata and C. capitata using the VR1 flanking regions (Figs 2(a) and S3). Next, we picked the highest-ranking gRNAs for C. capitata from the CRISPOR predictions, and after sequence verification of all gRNA hits we decided to proceed with only one gRNA (Cc2), as other hits were not specific to C. capitata.
To assess whether our gRNA sequence and universal amplification method can be specific to B. zonata, we performed a multiple sequence alignment with other commercially relevant fruit fly species (Figs S3 and S4). Our in-silico analysis demonstrated that Bz1 VR gRNA targets a unique sequence that differs from all other selected pests: B. dorsalis lacks the full PAM recognition site by a single thymine nucleotide, and the rest are different by at least two bases with one substitution within the seed region of the target (which is crucial for recognition and defined as up to 10 base pairs proximal to the PAM sequence). These results suggest that this gRNA would not recognize the other pests analyzed, but experimental testing of these organisms is required to validate this analysis.
The Chelex-extracted DNA from the samples described above was amplified, this time using the universal primers we designed and tested for the new gRNAs. First, we tested the ability to detect the targeted organisms in all developmental stages, as described above (Fig. 3(a)). These results established that the universal primers are indeed capable of amplifying DNA from both species. Next, the sensitivity of the universal detection assay was Figure 3. Universal amplification assay. (a) Cas12a detection of B. zonata (orange) and C. capitata (green) from different developmental stages. DNA was obtained using Chelex 100 from fresh samples (Methods). Amplification was performed using RPA with universal primers (Table S1) and detection was achieved using Bz1 gRNA for B. zonata and Cc2 for C. capitata (Table S1). As a negative control (NC), C. capitata and B. zonata larvae were used with noncorresponding gRNAs. (b) Detection sensitivity of Cas12a-Bz1 gRNA in varying ratios of pooled larvae. Samples containing a single larva of B. zonata with increasing amounts of C. capitata larvae were prepared. DNA was then extracted using Chelex 100, RPA amplified (using VR forward/reverse primers; Table S1) and diagnosed using Cas12a-Bz1 gRNA complexes. As a negative control, C. capitata DNA was used. (c) Detection sensitivity of Cas12a-Cc2 gRNA for C. capitata in varying ratios of pooled larvae. Samples containing a single larva of C. capitata with increasing amounts of B. zonata larvae were prepared. DNA was extracted and amplified as in (b) and diagnosed using Cas12a-Cc2 gRNA complexes. As a negative control, B. zonata DNA was used. All experiments were performed with three biological repeats and three technical repeats. www.soci.org DM Alon et al.
wileyonlinelibrary.com/journal/ps examined, and again we observed positive detection for both species with high confidence even at very low ratios of 1:50 ( Fig. 3(b),(c)). We also cross-tested the gRNAs for the opposing RPA amplification products and no cross-reactivity was detected. We did observe, however, a slight drop in the fluorescence signal in both reactions in the higher ratios (1:25 and 1:50) compared to the specific primers presented in Fig. 2(c). It is possible the lower signal is related to the higher amount of amplified DNA, as both insect species were amplified during RPA universal amplification. Lastly, we were also interested in assessing the stability of the detection reactions in storage. This aspect is important as long storage enables the preparation of bulk amounts of detection reactions in advance, reducing the amount of processing time per sample. We initially tested several conditions, including room temperature, 4°C, and −20°C for a period of over 160 h (Fig. 4). In this experiment, we prepared a large number of detection complexes and placed them in several 96-well plates (the starting point of the experiment). Each set of plates was placed at a different temperature (three plates for each temperature), and at each time point a single plate from each temperature was placed at room temperature for 30 min for thawing. Amplified DNA from a single adult fly was added to the detection reaction (the same amplification product was used for all detection reactions) and fluorescence was measured after 1-h incubation at 37°C. Room temperature samples lost activity after 24 h probably due to evaporation (plates were covered with lids and placed in a 25°C incubator, which might have contributed to the evaporation rate), which might account for the loss of activity. Samples kept at 4°C lost activity between 48 and 168 h, and samples kept at −20°C showed no significant loss of activity within the period tested. We did not test the stability of RPA reactions as they are reported by the manufacturer to be stable for up to 12 months at ambient temperature.

DISCUSSION
In this work, we demonstrate a fast, reliable, straightforward, and affordable genetic detection assay for pest species, enabling accurate point-of-care pest control. Our assay requires only a bench-top centrifuge, a hot-plate, and a hand-held fluorometer, and can further be adapted for paper-based detection. 13 The current price estimate per reaction is about 1.6 USD, and reagent prices may considerably decrease when purchased in bulk (Table S2).
We have demonstrated that the assay can be tailored to detect a specific fruit fly species or used for two species with the potential for universal amplification, including species that are highly similar, both morphologically and genetically. Universal reactions reduce hands-on time and enable the pooling of specimens in a single reaction. While we aimed at differentiating between two fruit fly genera, working with additional flies will require testing of potential gRNA cross-reactivity. The Chelex method for DNA extraction has proven to be efficient in all developmental stages of the target flies and together with the robust RPA method, it allowed the detection of even minute amounts of sample. The ability to distinguish a single target larva from a pooled tube containing up to 50× nontargets allows this assay to meet and even surpass current gold-standard methods while also dramatically reducing labor and cost. 35 There are several limitations to our current setup. First, RPA reagents are currently sold by a single company, creating a potential bottleneck in reagent supply. Second, as in all genetic-based methods, the genomic data of target pests is crucial for gRNA design and target acquisition. Finally, the use of a plate reader in our work limits the ability to use this assay in the field and requires expensive equipment. At the moment, there is no full genomic sequence of B. zonata, as well as other pests that might have high sequence similarity in the Bactrocera family. Although mitochondrial sequences are sufficient for gRNA acquisition for a single species, designing an assay for universal amplification and multispecies detection or differentiation might require a broader genetic database, specifically in cases of high genetic resemblance. The specificity of Cas12a detection relies on genomic data availability, which can contribute to the gRNA design process. This problem also exists with other pests, many of which have no available genomic data or sequence data restricted to the COI gene, which is sometimes not sufficient for unique gRNA acquisition. 36 Aside from using lateral-flow-based assays or handheld fluorometers as suggested previously, other modifications have been described that can dismiss the use of a plate reader. Such modifications include replacing the original fluorescence ssDNA reporter (a short ssDNA oligo with a 5' FAM modification and a 3 0 black-hole quencher modification) with more stable and robust  (Table S1) as a negative control. Blank (gray) are detection reactions without DNA.
reporters such as the coupling of the reporter to gold nanoparticles, gRNA modifications 37,38 or colorimetric reporters. 39,40 Amplification-free Cas12a methods that utilize complex chemistries to accomplish detection have also been described and may replace the need for RPA or LAMP amplification steps. 41,42 Several insect sequencing efforts are underway and hopefully will increase the available genomic data for agriculturally important insect pests. With this increase in sequence availability we hope our assay will be widely adopted and extended to other pests for one-pot amplification and detection assays in the field and in border controls, enabling rapid, onsite, affordable, and highly specific identification.