Pine wilt disease (PWD) is a devastating forest disease caused by the pine wood nematode (PWN) (Mota and Vieira, 2008). The spread of PWNs is mainly caused by increased global trade. As PWD is regarded as an “incurable disease,” rapid and accurate detection, quarantine, and monitoring can manage disease spread (Futai, 2013; Yazaki et al., 2018; Kim et al., 2020). Therefore, the primary task in PWD control is to accurately detect PWN.

Various Bursaphelenchus xylophilus diagnostic methods have been developed, including morphological methods using microscopic observations and molecular diagnostic methods using diverse molecular markers (Inácio et al., 2015; Lee et al., 2021). However, due to the similar morphological characteristics of B. xylophilus and its closely related species, Bursaphelenchus mucronatus, accurately identifying B. xylophilus requires specialized knowledge and professional skills (Gu et al., 2011; Gu and Wang, 2011; Wang et al., 2012; Gu, 2014; Li et al., 2021). Furthermore, molecular methods require sophisticated equipment and are not suitable for implementation in the field, causing delayed detection and response to PWD pandemics (Sultana et al., 2013). Therefore, a specific, sensitive, and field-applicable diagnostic method is needed to improve the efficiency of B. xylophilus detection and quarantine.

Alternatively, isothermal DNA amplification techniques that do not require the use of thermal cycling apparatus have been applied to detect B. xylophilus. These include loop-mediated isothermal amplification (LAMP) (Kikuchi et al., 2009; Meng et al., 2018; Ahuja and Somvanshi, 2021), denaturation bubble-mediated strand exchange amplification (SEA) (Liu C. et al., 2019), and recombinase polymerase amplification (RPA) (Cha et al., 2019, 2020; Fang et al., 2021). RPA, an isothermal nucleic acid amplification method that has attracted much attention because it is sensitive, rapid, works at isothermal temperature (Piepenburg et al., 2006; Glökler et al., 2021). In addition, RPA assays have been developed to detect some plant-parasitic and animal-parasitic nematodes, including B. xylophilus (Cha et al., 2019, 2020), Meloidogyne enterolobii (Subbotin, 2019), Meloidogyne spp. (M. incognita, M. arenaria, M. javanica, and M. enterolobii) (Ju et al., 2019), Meloidogyne javanica (Chi et al., 2020), Meloidogyne hapla (Song et al., 2020; Subbotin and Burbridge, 2021), Angiostrongylus cantonensis (Jarvi et al., 2021), and Trichinella spp. (Li et al., 2019).

Recombinase polymerase amplification amplicons can be visualized using lateral flow dipstick (LFD), with results could be directly interpreted by the naked eye within a few minutes (Daher et al., 2016). LFDs are products based on lateral flow technology using gold nanoparticles and can promptly detect amplification products making results easy to interpret (Qi et al., 2018).

The LFD assay requires a special nfo probe with a fluorescein amidites (FAM) (carboxyfluorescein) antigen label at the 5′-end, a tetrahydrofuran (THF) spacer (abasic site) in the middle, a C3 spacer (amplification blocker) at the 3′-end, and reverse primer labeled with biotin at the 5′-end. RPA amplicons are amplified using two primers and a probe, which contains both biotin and FAM, and the mouse anti-FAM antibody is enveloped with AuNPs. After reaction mixture is added to the sample well, they are driven by capillary force to move across the conjugate pad and bind to the anti-FAM AuNPs. The test line, which is enveloped with streptavidin, captures molecules with a biotin label when the amplification products go through. The control line, which is used to validate LFD detection, only captures the anti-FAM antibody enveloped with AuNPs because the anti-FAM antibody is from a mouse (Figure 1; Dai et al., 2019; Miao et al., 2019).

It should be noted that false-positive signals from the self-linked primer–probe complex should be considered a vital inherent defect of the RPA-LFD assay (Wu et al., 2020). Without the temperature cycles in PCR-based methods, primer combination in RPA is performed at ambient temperature, which does not provide an opportunity for dissociation if mispairing occurs and may generate a false-positive signal. In addition, LFD could not differentiate the size of the amplification product that releases signal. Furthermore, the RPA-LFD assay is overly sensitive; thus, interference from false-positive signals due to primer-dependent artifacts must be avoided (Poritz and Ririe, 2014; Wang et al., 2020).

In this study, we developed a simple, rapid, specific, and sensitive RPA-LFD assay to detect B. xylophilus. This method eliminated false-positive signals from non-specific primer–probe complexes due to an elaborate design and strict screening of primers and probe, as well as by bringing in base substitutions on the reverse primer and probe. The specificity and sensitivity of the assay were also investigated. The practicability was also analyzed by detecting B. xylophilus in spiked pine wood samples.

Bursaphelenchus xylophilus induced PWD has destroyed numerous pine trees and has spread to 18 provinces and 693 counties in China over the past 40 years, causing considerable economic losses of hundreds of billions of yuan [Announcement of the State Forestry and Grassland Administration (No. 6 of 2021), 2021]. Therefore, a primary procedure for disease prevention and quarantine comprising an accurate, rapid, sensitive, and convenient method for detecting B. xylophilus in pine wood is urgently needed.

Polymerase chain reaction-based amplification is considered the gold standard of molecular diagnostics; however, it requires trained personnel and specialized, expensive equipment generally unavailable outside laboratory settings (Patel et al., 2014). As an emerging isothermal amplification technique, RPA has displayed many strengths, such as short detection time, high sensitivity, easy operation, and minimum instrument needed. Besides, the RPA amplification results could be interpreted using LFD within a few mins. All these characteristics make RPA-LFD very suitable for on-site detection (James and Macdonald, 2015; Li et al., 2019).

The choice of detection target is crucial for specific diagnostics. Several genes have been selected as biomarkers for PWN detection, including 5S rRNA, hsp70, satellite DNA, ITS rRNA, topoisomerase I, syg-2, and pel-3 (Kang et al., 2004; Castagnone et al., 2005; Leal et al., 2007; Huang et al., 2010; Zhuo et al., 2011; Kang et al., 2015; Meng et al., 2018). Among these, the syg-2 gene has been reported to have superior performance in the LAMP method and was thus selected as the target in this study. To identify regions specific to B. xylophilus, the syg-2 gene sequence of B. xylophilus was amplified using the PCR primers designed by Gou (2014) and then compared with two closely related species, after which the diverged region specific to B. xylophilus was selected (Supplementary Figure 1). The diverged region was considered a possible target area for designing primer and probe in the assay. Since the protocadherin DNA family has been shown to consist of tandemly arranged repeats, the amplification of a ladder of multimers of the amplicon-sized monomer was observed in basic-RPA experiments using the designed primers (Figure 2; Hirano et al., 2003; Washbourne et al., 2004).

On the LFD, the control line is enveloped with anti-mouse antibody, and the test line was enveloped with streptavidin. The test line first captures products labeled with biotin when the reaction mixture passes through. If the products also had a FAM label, it will combine with the AuNPs, which are enveloped with anti-FAM antibody, led to the enrichment of AuNPs at the test line, showing a red line (Wang et al., 2020). The control line is used to validate LFD detection, only captures the anti-FAM antibody enveloped with AuNPs, because the anti-FAM antibody is from a mouse (Figure 1).

Nonetheless, the RPA-LFD method has a non-ignorable inherent defect of primer-dependent artifacts. In the RPA-basic-based LFD reaction, forward and reverse primers were labeled with FAM and biotin at 5′-end, respectively. Amplicons are labeled with both biotin and FAM, if primer dimers are formed, it also produces a positive signal (Figures 8A,B). No matter how cautious the screening of primer is, it is nearly impossible to prevent primer dimers in the DNA amplification process (Meagher et al., 2018). Thermal cycling strategies could prevent primer-dimer formation in PCR but are unsuitable for RPA because RPA works at an isothermal temperature of about 38°C, and mispaired primers are difficult to dissociate (Brownie et al., 1997). Furthermore, the LFD method cannot differentiate the size of the molecule emitting signal and treats each molecule as a positive signal (Safenkova et al., 2020). Considering the RPA-LFD method is highly sensitive (Miao et al., 2019), any interference from primer-dependent artifacts could result in false-positive detection.

Research has shown that introducing a probe in RPA would enhance specificity and decrease non-specific amplification products (Figures 8C,D; Piepenburg et al., 2006; Ivanov et al., 2021). FAM was labeled at the 5′-end of the probe, while the C3-spacer was labeled at the 3′-end of the probe that block extension. The THF site was placed on the probe to be cut by nfo enzyme. In this way, amplification led by primer pair would result in a product that only has biotin label, which would not generate a signal on the LFD (Figure 8C). The probe fully pairing amplified DNA strand can be cut by nfo at THF site (nfo was activated if bases flanking THF site paired with complementary bases), freeing the 3′-end for the extension to produce a positive signal on the LFD (Figure 8D). Using this probe-guided amplification, primer-dimers did not produce a signal (Figure 8E). The partly matched primer–probe complex could not be amplified (Figure 8G). Only in the rare case, the primer and probe paired several bases flanking the THF site (Figure 8F), and this would generate false-positive signal (Wu et al., 2020).

Therefore, elimination of the false-positive signal could not only count on using a probe, elaborate design and strict screening are still essential to prevent false-positive scenarios (Figure 8). Based on previous research, we inferred that the false-positive signal occurred in this study should be caused by the probe-primer complex since the labels were on the probe and reverse primer (Wu et al., 2020). By setting stringent parameters, probes were designed to have the lowest theoretical chance of primer pairing. However, cross-dimer analysis displayed many possible primer–probe dimer scenarios between the nfo probe and the reverse primer, and some cross-dimers fit the special case of false-positive signals. That is, the reverse primer formed mismatches with bases flanking the THF site on the probe, which could elucidate why the original primer–probe set (S1-nfo-P/S1R-nfo) displayed false-positive signals.

The probe needed to be in the middle of two primers. However, obtaining an ideal probe without any continuous bases pairing to the reverse primer was scarcely possible. Nonetheless, we could exploit the tolerance of RPA to certain base mispairings on primers against the template, and attempt to bring in base substitutions in RPA-LFD method (Daher et al., 2015; Liu X. et al., 2019). By imitating a previously described method of bringing in three mismatches (Wu et al., 2020) in which two are in the probe and one in the reverse primer, the probe-primer dimer ligation was artificially interrupted, and false-positive was avoided. In addition, the RPA-LFD efficacy was not noticeably influenced when compared the color difference between the S1-nfo-P/S1R-nfo line and mS1-nfo-P/mS1R-nfo line, as shown in Figure 3.

Primer-dependent artifacts were successfully eliminated, resulting in the establishment of the RPA-LFD method. The analytical specificity test showed that the assay displayed high specificity toward B. xylophilus and avoided cross-reactions with the other two Bursaphelenchus species tested. A previous study reported that the detection limit of the RPA assay is approximately 1.6 fg (1.6 × 10^–6 ng) of gDNA from B. xylophilus (using pure gDNA) (Cha et al., 2020). However, in our RPA-LFD assay, a sensitivity test using B. xylophilus gDNA showed that the detection limit was 1 pg. This is probably because the RPA probe-based approach had a lower sensitivity than the approach with labeled primers (Ivanov et al., 2021).

The whole procedure for the RPA-LFD assay could be finished under 38°C within about 30 min, including 15 min for nematode gDNA extraction and master mix preparation, 15 min for the RPA-LFD assay (10 min for RPA reaction, 5 min for visual detection on the LFDs) (Figure 9). Compared with other diagnostic methods for B. xylophilus detection, the time needed for RPA-LFD detection was the shortest. Furthermore, RPA displayed tolerance to interference from pinewood (Figure 7), presenting an opportunity for further investigation into the prevalence of B. xylophilus, especially in new areas of occurrence.

However, applying RPA-LFD detection to PWN diagnostics still faces some issues, with the expense being a major one. Reagent costs for RPA-LFD assays are approximately 50 RMB per reaction at present, even when using the RPA nfo kit manufactured locally, which are higher than those for polymerase chain reaction. However, when considering instrumentation costs, the RPA-LFD method becomes relatively more affordable. Even though PCR is currently considered the gold standard for nematode diagnostics, thermocycling instruments are expensive. Local forestry bureaus or quarantine departments are not commonly equipped with these, results usually take days to weeks to obtain due to the shortage of instruments. The minimal equipment needed for conducting RPA-LFD assays would allow reactions to be completed timely in the field or at quarantine departments for point-of-care diagnostics within minutes rather than days.

The relatively low cost of RPA-LFD assay can be used by more people to support extensive surveillance across the country, whether in areas known to have high levels of PWD infection, such as Zhejiang Province, or in areas with relatively low levels of infection regions, supporting decision-making. In addition, detection capability in remote areas with limited resources could drastically increase with the emergency deployment of RPA-LFD assay kits. Thus, the RPA-LFD assay developed here has potential applications in the field and areas with limited resources.

In conclusion, a rapid, on-site RPA-LFD assay for specific and sensitive detection of B. xylophilus was developed. The RPA-LFD assay prevents the risk of false-positive results from primer-dependent artifacts by using a probe and bringing base substitutions in the primer and probe. Furthermore, the assay successfully detected B. xylophilus with high specificity and sensitivity in less than 30 min at an isothermal temperature of 38°C. The developed RPA-LFD assay can provide a novel alternative for PWN point-of-care testing that has a simple read-out system and can be performed under field conditions without any special instrumentation or in areas with minimal laboratory infrastructure.

The original contributions presented in the study are included in the article/Supplementary Material, further inquiries can be directed to the corresponding author.

QZ conceptualized and designed the research, analyzed the data, interpreted the results, and wrote the manuscript. QZ and YL performed the experiments. QL, YL, ZW, and HW participated in the discussion for experimental design. XZ helped funding acquisition. QL revised the manuscript and directed the project. All authors contributed to the article and approved the submitted version.

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.