Skip main navigation
Previous
Next
RESEARCHFree Access icon

Rapid and Sensitive Detection of Meloidogyne graminicola in Soil Using Conventional PCR, Loop-Mediated Isothermal Amplification, and Real-Time PCR Methods

    Affiliations
    Authors and Affiliations
    • Qingcong He1 2
    • Dongwei Wang1
    • Bei Tang1 2
    • Jian Wang3
    • Deyong Zhang1 2
    • Yong Liu1 2 †
    • Feixue Cheng1 2 †
    1. 1Key Laboratory of Integrated Management of the Pests and Diseases on Horticultural Crops in Hunan Province, Institute of Plant Protection, Hunan Academy of Agriculture Sciences, Changsha 410125, China
    2. 2Long Ping Branch, Graduate School of Hunan University, Changsha 410125, China
    3. 3Agricultural Economy and Regional Planning Research Institute, Hunan Academy of Agricultural Sciences, Changsha 410125, China
    Published Online:16 Dec 2020https://doi.org/10.1094/PDIS-06-20-1291-RE

    Abstract

    Meloidogyne graminicola is one of the major plant-parasitic nematodes (PPNs) that affect rice agriculture. Rapid identification and quantification of M. graminicola in soil is crucial for early diagnosis so that measures can be taken to reduce the impact of PPN diseases and ensure food security. In this study, M. graminicola species-specific primers for conventional PCR, loop-mediated isothermal amplification (LAMP), and real-time PCR were designed based on the sequence-characterized amplified region. The primers were highly specific and sensitive, and only samples containing M. graminicola DNA showed positive results. The sensitivity of LAMP and real-time PCR (two second-stage juvenile [J2] M. graminicola in 100 g of soil) was higher than that of conventional PCR (200 J2s in 100 g of soil). A standard curve (correlation coefficient R2 = 0.970, P < 0.001) was generated by amplifying DNA extracted from 0.5 g of soil, and a significant correlation was observed between the number of M. graminicola determined by microscopic examination and that predicted from the standard curve (R2 = 0.477, P = 0.0160). In quantification analyses of M. graminicola isolated from 31 naturally infested soils, the sensitivity of LAMP and real-time PCR (22 M. graminicola in 100 g of soil) was higher than that of conventional PCR (211 M. graminicola in 100 g of soil). The conventional PCR, LAMP, and real-time PCR methods have the potential to provide a useful platform for rapid species identification according to the experimental conditions. The real-time PCR assay and standard curve can be used for quantification of M. graminicola. These newly developed assays will help to facilitate the control of these economically important PPNs.

    Plant-parasitic nematodes (PPNs) are the most important plant pathogens in many countries and they contribute greatly to estimated total annual agricultural losses of approximately $80 billion USD (Mantelin et al. 2017; Nicol et al. 2011). One of the most economically important genera is the root-knot nematodes (RKNs; Nematoda: Secernentea: Tylenchida: Heteroderidae: Meloidogyne spp.). RKNs are sedentary endoparasites of the underground parts of plants; they have a wide plant host range including more than 3,000 plant species (Blok et al. 2008; Jones et al. 2013; Trudgill and Blok 2001). M. graminicola, commonly termed the rice RKN, is one of the most important PPNs and is a major threat to rice agriculture, causing an estimated 20 to 87% of economic losses in rice production (Haegeman et al. 2013; Kyndt et al. 2012; Mantelin et al. 2017). China is one of the main rice-growing regions in the world, responsible for about 35% of global rice production (Ji 2016; Luo et al. 2018). M. graminicola was first described and identified in 2001 on onions in Sanya City in Hainan Province (Zhao et al. 2001) and also occurs in the Guangdong, Guangxi, Fujian, Hunan, Hubei, Jiangxi, Jiangsu, and Sichuan Provinces in China (Tian et al. 2017). M. graminicola has a short life cycle and a high reproductive output; therefore, even a small number of M. graminicola in soil could increase to a high population density. The identification of M. graminicola is complex and is usually based on symptoms of hook-like galls produced on rice roots; however, once rice has been infected and shows symptoms of hook-like galls, it is difficult to effectively control these symptoms and the rice growth process. Therefore, establishment of rapid and sensitive identification of M. graminicola in soil is crucial for early diagnosis and effective nematode management strategies.

    Current control methods for RKNs include agricultural practice, physical control, chemical prevention, and planting disease-resistant cultivars, although application of chemical nematicides is still the major method in use. To achieve optimum results in the control of nematodes in the field, it is generally accepted that detection and quantification of RKNs in soil is very important infacilitating selection of the optimal disease management and RKN control strategies. The main criteria for identification of RKNs are based on the morphological characteristics of females, males, and second-stage juveniles (J2s) (Wesemael et al. 2011); morphological identification is time consuming and requires skilled technicians and high-powered and expensive microscopes. So, there is an urgent need for new technology to identify M. graminicola quickly and precisely, and this detection depends largely on the sensitivity and specificity of the detection method.

    With the rapid development of molecular biology technologies, many species, including nematodes, can now be identified using molecular methodologies. PCR-based DNA amplification and sequence analysis has been applied to the identification of nematodes (Powers 2004; Umehara et al. 2006). The molecular identification methods most commonly used for RKNs include the sequence-characterized amplified region (SCAR) (Bellafiore et al. 2015; Feng et al. 2017; Randig et al. 2002; Tigano et al. 2010; Zijlstra 2000; Zijlstra et al. 2000), restriction fragment length polymorphism (Brown et al. 1996; Garcia et al. 1996; Han et al. 2004; Kamran et al. 2019; Klein-Lankhorst et al. 1991; Messeguer et al. 1991), random amplified polymorphic DNA (Adam et al. 2007; Yi et al. 1998; Zhu et al. 2006), real-time PCR (Berry et al. 2008; Painter and Lambert 2003), loop-mediated isothermal amplification (LAMP) (He et al. 2013; Niu et al. 2011, 2012; Peng et al. 2017; Wei et al. 2016; Zhang and Gleason 2019), and lateral flow dipstick (LFD) testing based on LAMP products (LAMP-LFD) (Cai et al. 2016). To date, only the rDNA internal transcribed spacer region (ITS) and real-time PCR have been used to identify the rice RKN, M. graminicola (Htay et al. 2016; Katsuta et al. 2016). Although these methods are highly specific and sensitive, they are not optimized for field sample studies. In addition, there has been no report of M. graminicola detection using the total DNA of soil. Therefore, the development of new, reliable, and easy-to-use methods for the detection of M. graminicola is critical for the effective control of M. graminicola in rice production. In this study, we tested and compared three different methods (i.e., conventional PCR, LAMP, and real-time PCR) for detection of M. graminicola in soil samples. These three methods should allow rapid and accurate detection of M. graminicola infestation in soil samples, and they will permit selection of the most reasonable, convenient, and effective identification method depending on the test conditions.

    Materials and Methods

    Nematodes and DNA extraction.

    M. graminicola were isolated from roots of rice plants grown in Hunan, China. The other nematode species, M. incognita, M. javanica, M. arenaria, M. hapla, M. enterolobii, Hirschmanniella oryzae, Heterodera elachista, Pratylenchus coffeae, and Caenorhabditis elegans, were collected, separated, identified, cultured, and preserved in the Key Laboratory of Integrated Management of the Pests and Diseases on Horticultural Crops in Hunan Province at the Hunan Academy of Agricultural Science in Changsha, China. M. graminicola were cultured on rice (Oryza sativa) in a greenhouse at 28°C under a 16-h/8-h light/dark cycle. J2s were collected from an in vivo culture on rice plants as described previously (Cheng et al. 2017). Total DNA was extracted from samples using two different methods. First, total DNA was extracted from soil samples artificially infested with nematode or from soil samples collected from infested fields using the FastDNA SPIN Kit for Soil (MP Biomedicals, Solon, Ohio) according to the manufacturer’s instructions, and each DNA sample was dissolved in 100 µl of sterile double-distilled water (ddH2O). Second, DNA was extracted from single cultured nematodes as described previously (Song et al. 2017). The resulting DNA samples were stored at −20°C for further use.

    Primer design.

    M. graminicola forward and reverse primers (MgFW and MgRev) described previously (Bellafiore et al. 2015) were first used to amplify the M. graminicola SCAR sequence (639 bp). The resulting PCR product was gel purified and cloned into the pMD18T vector (Takara, Dalian, Liaoning, China). After propagation in Escherichia coli DH5α competent cells, the plasmid DNA was sequenced and aligned with the M. graminicola SCAR marker sequence (GenBank KF499563.1) deposited in the NCBI database as described (Wang et al. 2019a). The confirmed M. graminicola SCAR marker sequence was used to design new primers for further PCR, LAMP, and/or real-time PCR assays. The LAMP primers were designed based on the conserved regions of the SCAR marker sequence using Primer Explorer software version 5 (http://primerexplorer.jp/e/). Four primers and a probe were designed to identify the target gene regions: a forward internal primer (Mg-FIP), a reverse internal primer (Mg-BIP), a forward external primer (Mg-F3), a reverse external primer (Mg-B3), and a probe (Mg-HP-FITC) (Fig. 1). The Mg-FIP and Mg-HP-FITC primers were 5′ labeled with biotin and fluorescein isothiocyanate (FITC), respectively. Primers Mg-F3 and Mg-B3 were also used for conventional PCR. The real-time PCR primers, Mg-F and Mg-B, were also designed according to the M. graminicola SCAR marker sequence with Primer Premier 5. All primers were synthesized by Sangon Biotech Co., Ltd. (Shanghai, China) and are listed in Table 1.

    Fig. 1.

    Fig. 1. Sequence-characterized amplified region (SCAR) sequence (GenBank KF499563.1) of Meloidogyne graminicola (Mg). The following primers are indicated by underlined nucleotides and arrows: forward external primer Mg-F3, reverse external primer Mg-B3, forward internal primer Mg-FIP (F1c+F2), and reverse internal primer Mg-BIP (B1c+B2). The Mg-HP-FITC-labeled probe sequence is shown by underlined nucleotides. The real-time PCR primers are denoted by shaded arrow boxes. FITC, fluorescein isothiocyanate.

    Download as PowerPoint

    Table 1. Primers used in this study

    Conventional PCR reaction.

    The PCR reaction was performed in 20 µl of reaction mixture containing 10 µl of 2.5 U EasyTaq DNA polymerase (TransGen Biotech, Beijing, China), 1 µl each of the Mg-F3 and Mg-B3 primers (10 µM), 1 µl of template DNA, and 7 µl of ddH2O. PCR was performed using a DNA thermal cycler for 35 cycles of denaturation at 94°C for 1 min, annealing at 55°C for 15 s, and extension at 72°C for 30 s, followed by a final extension at 72°C for 10 min. The PCR products were visualized by 1.0% agarose gel electrophoresis and photographed on an ultraviolet transilluminator. A nontemplate control was included with ddH2O instead of template DNA. The experiment was replicated three times and each sample had three technical replicates.

    LAMP reaction.

    The LAMP reaction was conducted in a 25-µl reaction mixture containing 2.5 µl of 10× ThermoPol buffer, 3 µl of dNTPs (10 mM), 1 µl of MgSO4 (100 mM), 0.5 µl of Mg-F3/Mg-B3 (10 µM), 1 µl of Mg-FIP/Mg-BIP (40 µM), 1 µl of Bst 2.0 WarmStart DNA polymerase (8 U/µl) (New England Biolabs, Ipswich, MA), 1 µl of DNA, and 13.5 µl of ddH2O. The reaction mixture was incubated at 63°C for 60 min and terminated at 82°C for 5 min. The LAMP products were visualized by electrophoresis or with SYBR Green I fluorescent dye (LAMP-Green) (Invitrogen, Shanghai, China) or an LFD (LAMP-LFD) (Milenia Biotec GmbH, Gießen, Germany). The positive LAMP sample appeared as ladder-like bands in the gel. The SYBR Green I dye detection method was based on visual observation where the positive reaction was green in color, while the negative reaction remained orange. In addition, 5 µl of Mg-HP-FITC (20 pM) was added to the LAMP products and hybridization was performed at 65°C for 5 min. Subsequently, 8 µl of hybridized product was added to 100 µl of assay buffer in a new tube and mixed, then the LFD was dipped into the mixture for 5 min and observed for the presence of the test line. A nontemplate control was included with ddH2O instead of template DNA. The experiment was replicated three times and each sample had three technical replicates.

    Real-time PCR reaction.

    Real-time PCR was performed on a CFX-96 real-time PCR machine (Bio-Rad, Hercules, CA) and each reaction (20 µl) contained 1 µl of template DNA, 0.4 µl of Mg-F (10 µM), 0.4 µl of Mg-R (10 µM), 10 µl of 2× SYBR Premix Ex Taq (Takara), and 8.2 µl of ddH2O. Conditions of real-time PCR were 95°C for 30 s, followed by 39 cycles of 95°C for 5 s and 64°C for 30 s. The real-time PCR primer pair was evaluated based on amplification efficiency, endpoint fluorescence, and melting curve profile. Melting curve profiles were obtained between 60 and 95°C at 0.3°C per 15 s. The soil sample collected from a rice field and confirmed to have no M. graminicola by light microscopy and PCR was used as the reference soil. Different numbers of M. graminicola (0, 1, 5, 10, 50, 100, 1,000, and 10,000 J2s) were added to 0.5 g of reference soil, and the DNA was extracted as described above. A standard curve was created for the resulting series of M. graminicola DNA pools. Real-time PCR amplification efficiency (E) values were obtained using the following formula as described:

    E=10[1/alope]−1

    (Wang et al. 2019b). The experiment was replicated three times and each sample had three technical replicates.

    Test Specificity.

    To determine the specificity of the conventional PCR, LAMP, and real-time PCR assays, DNA was extracted from soil samples infected with M. graminicola collected from rice fields in seven different districts (namely Yiyang, Changde, Zhuzhou, Changsha, Yueyang, Xiangtan, and Hengyang) in Hunan Province. DNA was also extracted from eight other PPNs (M. incognita, M. javanica, M. arenaria, M. hapla, M. enterolobii, Hirschmanniella oryzae, Heterodera elachista, and P. coffeae) and from the nonplant-parasitic nematode, C. elegans.

    Test sensitivity.

    To determine the sensitivity of the conventional PCR, LAMP, and real-time PCR assays, DNA was extracted from 0.5 g of reference soil containing zero (control), one, five, and 10 J2s. The DNA sample extracted from 0.5 g of reference soil containing one J2 was further diluted 10-, 100-, 1,000-, and 10,000-fold in sterile ddH2O. These DNA samples were then used for conventional PCR, LAMP, and real-time PCR. The experiment was replicated three times and each sample had three technical replicates.

    Evaluation of the detection methods using natural field soil samples.

    To validate the usefulness of the newly developed conventional PCR, LAMP, and real-time PCR assays for detection of M. graminicola in field samples, soil samples (approximately 200 g) were collected from 31 sites in rice fields at five locations in Hunan Province (Table 2). These field soils were naturally infested with M. graminicola, other PPNs, and free-living nematodes. The sampled fields varied in soil type and moisture status depending on the location. Soil samples were placed in sealable plastic bags and then immediately transported in cooler boxes to our laboratory. Samples were stored at 4°C prior to analysis. Each sample was mixed thoroughly and divided into two 100-g subsamples. Tests were subsequently performed as follows. First, nematode populations were extracted from a subsample of 100 g of soil using the Baermann funnel method (Whitehead and Hemming 1965), and the number of M. graminicola was calculated by identification under an inverted microscope. Second, 0.5 g of soil was taken from the other 100-g soil subsample in triplicate and the DNA extraction of nematode samples in soil was conducted as described above. The DNA within the tested samples was amplified using conventional PCR, LAMP, and real-time PCR. For real-time PCR, the cycle number that showed the detection of M. graminicola in a DNA sample isolated from 0.5 g of naturally infested soil was converted to the number of M. graminicola against the established standard curve. The number of M. graminicola in a 100-g soil sample indicated by real-time PCR was predicted using the following formula:

    200×10(29.708−cycle number)/3.146

    Comprehensive comparison of the M. graminicola detection methods, including traditional separation using the Baermann funnel method and the three methods described in this study, was conducted to evaluate the accuracy and applicability of these newly developed detection methods.

    Table 2. Population density of second-stage juveniles (J2s) of Meloidogyne graminicola in naturally infested soil detected with the Baermann funnel method, conventional PCR, loop-mediated isothermal amplification (LAMP), and real-time PCR from naturally infested soil

    Statistical analysis.

    Statistical analyses were performed using analysis of variance and the SPSS 13.0 statistical software package (SPSS Inc., Chicago, IL). The correlation coefficient (R2) between the number of M. graminicola determined by the Baermann funnel method and that determined by the real-time PCR estimation was evaluated using the bivariate correlations analysis software in the SPSS 13.0 package.

    Results

    Primer design and detection.

    A 639-bp fragment was PCR amplified from M. graminicola DNA using the primers MgFW and MgRev (Table 1). Sequence alignment results confirmed that this fragment was from the M. graminicola SCAR marker sequence region reported previously (Bellafiore et al. 2015). The DNA of pure culture M. graminicola was amplified using primers designed for conventional PCR (Mg-F3 and Mg-B3), LAMP (Mg-F3, Mg-B3, Mg-FIP, Mg-BIP, and Mg-HP-FITC), and real-time PCR (Mg-F and Mg-R). The results showed that the conventional PCR produced a single DNA band of about 250 bp in the gel (Fig. 2A). LAMP products were detected by electrophoresis, SYBR Green I dye, and LFD. The LAMP characteristic ladder-like bands could be observed in the positive sample containing M. graminicola template DNA (Fig. 2B), the positive reaction turned green with visual fluorescence detection after SYBR Green I dye was added (Fig. 2C), and two black lines (control and test) appeared (Fig. 2D) after the test strip was added to the LAMP product in the LAMP-LFD detection. The control sample showed no band, color variation, or test line. Real-time PCR results showed specific amplification (87 bp) from the M. graminicola DNA sample (Fig. 2E) and melting curve analyses showed a single peak at 77.4 ± 0.3°C (Fig. 2F), while the control did not produce specific amplification and no peak was observed in the melting curve analyses.

    Fig. 2.

    Fig. 2. Detection of Meloidogyne graminicola using conventional PCR, loop-mediated isothermal amplification (LAMP), and real-time PCR assays. Total DNA was extracted from a soil sample with M. graminicola (sample 1) or one without M. graminicola (sample 2, control). The extracted DNA samples were used in these assays. A, Conventional PCR detection. B, LAMP-electrophoresis detection. C, LAMP-Green detection with SYBR Green I fluorescent dye. D, LAMP-lateral flow dipstick detection. E, Amplification of real-time PCR detection. F, The melt peak of real-time PCR detection. M, DNA ladder; RFU, relative fluorescence unit.

    Download as PowerPoint

    Assay specificity.

    Results of the conventional PCR, LAMP, and real-time PCR assays demonstrated that in the DNA samples containing M. graminicola, the conventional PCR assay showed a single specific DNA band (Fig. 3A); the LAMP assay showed characteristic ladder-like bands (Fig. 3B), green fluorescence (Fig. 3C), and two black lines (control and test) (Fig. 3D); and the real-time PCR assay showed specific amplification (Fig. 3E). There were no DNA bands, green fluorescence, test lines, or specific amplification in the other nine nematodes and the control (Fig. 3). The results indicate that the conventional PCR, LAMP, and real-time PCR assays could specifically detect M. graminicola, including the seven isolates from different districts, and could distinguish M. graminicola from the other nematodes.

    Fig. 3.

    Fig. 3. Specificity of conventional PCR, loop-mediated isothermal amplification (LAMP), and real-time PCR assays for Meloidogyne graminicola. Total DNA was extracted from soil samples artificially infested with M. graminicola (samples 1 to 7), M. incognita (sample 8), M. javanica (sample 9), M. arenaria (sample 10), M. enterolobii (sample 11), Hirschmanniella oryzae (sample 12), Heterodera elachista (sample 13), Aphelenchoides besseyi (sample 14), Pratylenchus coffeae (sample 15), and Caenorhabditis elegans (sample 16). Total DNA extracted from the reference soil sample without nematodes was used as a negative control (sample 17). A, The results of conventional PCR detection. The correct size of the PCR band is indicated by a white arrow. B, The results of LAMP-electrophoresis detection. Only samples 1 to 7 produced typical LAMP ladder-like bands. C, The results of LAMP-Green detection with SYBR Green I fluorescent dye. Only samples 1 to 7 produced a positive reaction (green). D, The results of LAMP-lateral flow dipstick detection. Only samples 1 to 7 gave a positive test line. E, The results of real-time PCR detection. Only samples 1 to 7 gave positive melting curves. M, DNA ladder; RFU, relative fluorescence unit.

    Download as PowerPoint

    Assay sensitivity.

    The sensitivity of the conventional PCR, LAMP, and real-time PCR assays was evaluated with the soil DNA templates containing M. graminicola. The results showed that conventional PCR can be used to detect M. graminicola DNA in 0.5 g of reference soil with 10 J2s, five J2s, and a single J2 but not in its diluted sample (Fig. 4A). The LAMP assay produced specific DNA bands (Fig. 4B), green fluorescence (Fig. 4C), and two black lines (control and test) (Fig. 4D). The real-time PCR assay produced specific amplification (Fig. 4E) using the DNA samples not only from the soil sample with 10 J2s, five J2s, and a single M. graminicola J2, but also in a 100× diluted sample. The results indicated that the sensitivity of conventional PCR permitted detection with 200 J2s in 100 g of soil, while the sensitivity of the LAMP and real-time PCR assays permitted detection of two single J2s in 100 g of soil.

    Fig. 4.

    Fig. 4. Sensitivity of conventional PCR, loop-mediated isothermal amplification (LAMP), and real-time PCR assays. Total DNA was extracted from soil samples with 10, five, or one second-stage juveniles (J2s) of Meloidogyne graminicola (samples 1 to 3). The total DNA from the soil sample with one J2 of M. graminicola was further diluted 10-, 100-, 1,000-, and 10,000-fold in double-distilled water (samples 4 to 7). These DNA samples were used in these assays. Total DNA from a soil sample without nematodes was used as a negative control (sample 8). A, Conventional PCR detection results. Only samples 1, 2, and 3 gave positive PCR bands. B, LAMP-electrophoresis detection results. Samples 1, 2, 3, 4, and 5 gave typical LAMP ladder-like bands. C, LAMP-Green detection results with SYBR Green I fluorescent dye. Samples 1, 2, 3, 4, and 5 gave positive reactions (green). D, LAMP-lateral flow dipstick detection results. Samples 1, 2, 3, 4, and 5 gave positive test lines. E, Results of real-time PCR detection. Samples 1, 2, 3, 4, and 5 gave positive melting curves. M, DNA ladder; RFU, relative fluorescence unit.

    Download as PowerPoint

    Generation of the standard curve from M. graminicola in artificially infested soil.

    The standard curve was generated by plotting the log number of M. graminicola in 0.5 g of soil (x) against the real-time PCR threshold cycle numbers (y). An equation was obtained:

    y=−3.146x+29.708 (P<0.001)

    The real-time PCR amplification E and R2 values were 1.079 and 0.970, respectively. In this study, the control soil sample, without M. graminicola, showed no amplification.

    Quantification of M. graminicola isolated from naturally infested soil.

    The initial microscopic analysis showed that the 31 field-collected soil samples contained 0 to 5,000 M. graminicola in 100 g of soil. Total DNA was extracted from these soil samples (0.5 g each) and used to evaluate the sensitivity of the conventional PCR, LAMP, and real-time PCR assays. Conventional PCR results showed that M. graminicola DNA could be detected in a soil sample (100 g) with 211 M. graminicola J2s (Table 2). When LAMP was used to detect M. graminicola DNA in soil samples, the soil sample with 22 M. graminicola J2s gave a positive reaction (Fig. 5; Table 2); when real-time PCR was used to detect M. graminicola DNA in soil samples, the soil sample with 22 M. graminicola J2s also gave a positive reaction. When the threshold cycle numbers were plotted individually against the standard curve in real-time PCR, these numbers were found to scatter around the standard curve, with a few exceptions (Fig. 5). Therefore, we conclude that the nematode enumeration obtained by real-time PCR is positively correlated with the standard curve. The R2 value between the numbers of M. graminicola obtained through the microscopic examination and the standard curve was 0.477 (P = 0.0160), indicating that the results from these two methods are significantly correlated (Supplementary Fig. S1).

    Fig. 5.

    Fig. 5. Establishment of a standard curve for real-time PCR detection of Meloidogyne graminicola. The standard curve was established using the threshold real-time PCR cycle number and the log of M. graminicola numbers (1, 5, 10, 50, 100, 1,000, and 10,000) in 0.5 g of soil. Each black triangle indicates a threshold real-time PCR cycle number to the log of the M. graminicola number found in a 0.5-g soil sample. Each gray cross indicates the threshold real-time PCR cycle number to the log of M. graminicola number found in a 0.5-g field-collected infested soil sample.

    Download as PowerPoint

    Discussion

    Rapid and accurate detection of M. graminicola infestation in soil samples is essential for the diagnosis and effective management of rice RKN disease. Isolation of nematodes from soil samples or plant roots, followed by microscopic examinations of the morphological features of J2s and adult females and males, is a common practice for Meloidogyne spp. (Eisenback et al. 1981). This method, however, has several limitations for routine applications; it is time consuming and less reliable and requires skilled personnel and expensive high-resolution microscopes. In addition, adult nematodes, especially adult males, are difficult to obtain and adult females are often malformed. To overcome these difficulties, many laboratories have developed several biochemical and molecular tools for nematode detections and identifications. For example, several DNA-based methods, including PCR, have been reported for large-scale detection of Meloidogyne spp. especially. ITS-based PCR is a popular method for detection of Meloidogyne spp. and utilizes the ITS region of rDNA as a marker for identification. Htay et al. (2016) showed that a PCR method using primers specific for M. graminicola ITS can be used to detect single J2 nematodes. The species-specific SCAR marker is also a new method developed for the detection of specific RKNs. Using this method, researchers can now accurately identify four major RKNs (M. incognita, M. javanica, M. enterolobii, and M. arenaria) in mixed nematode samples and soil samples (Fourie et al. 2001; Tigano et al. 2010; Zijlstra 2000; Zijlstra et al. 2000). The M. graminicola SCAR marker sequence is published and has been used to detect M. graminicola in a fast and reliable manner (Bellafiore et al. 2015). In this study, species-specific primers for conventional PCR, LAMP, and real-time PCR assays to detect M. graminicola in soil samples were designed based on the M. graminicola SCAR marker sequences and used to rapidly detect the target nematode DNA. Our results showed that these primers are highly specific for M. graminicola. In this study, all assays using these primers and DNA extracted from soil samples containing other PPNs or a soil sample without PPN yielded negative results. Our results also showed that the sensitivity of LAMP and real-time PCR was two J2s in 100 g of soil and that of conventional PCR was 200 J2s in 100 g of soil, respectively. Therefore, the conventional PCR, LAMP, and real-time PCR assays could directly detect M. graminicola, but the sensitivity of conventional PCR was inferior to that of LAMP and real-time PCR.

    Quantification of M. graminicola isolated from naturally infested soil showed a positive sample detection rate of 17 of 31 (54.84%) by conventional PCR and 25 of 31 (80.65%) by the LAMP and real-time PCR assays. In addition, all positive samples identified by conventional PCR tested positive on the LAMP and real-time PCR assays. The conventional PCR detection limit for M. graminicola was 211 J2s/100 g of soil and the detection limits of the LAMP and real-time PCR assays were 22 J2s per 100 g of soil (Table 2) based on the Baermann funnel and microscopic counting methods. In this study, the soil samples with 18 or fewer J2s tested negative, indicating that further improvement is needed to provide more accurate results. We speculate that one possible cause of the false-negative results was the uneven distribution of nematodes in the soil samples. We recommend that for fields with a low number of M. graminicola, 100 g of soil from multiple locations in the same field is needed. Test sensitivity using field-collected soil samples can also be improved by increasing the amount of soil used for DNA extraction. It is also noteworthy that the sensitivity of LAMP and real-time PCR is above the recommended damaging threshold level for M. graminicola in rice (i.e., 1,000 J2s/kg of soil) (Jaiswal et al. 2011), indicating that the LAMP and real-time PCR assays will be useful for disease forecasting and management. In addition, the number of M. graminicola in some samples estimated by real-time PCR was greater than the number observed by microscopic examination with the Baermann funnel method. We believe that the Baermann funnel method might only detect mobile nematodes, and not the nonactive forms (e.g., egg masses and eggs).

    A study by Katsuta et al. (2016) showed that M. graminicola can be detected and quantified in samples extracted from soils through SYBR Green I-based real-time PCR using primers specific for the ITS region. The amplification efficiency and detection limit of that assay were reported at E values of 0.45 and 1.57 J2s/20 g of soil, respectively (Katsuta et al. 2016). In this study, real-time PCR amplification efficiency reached an E value of 1.079, which allowed us to detect 0.01 J2s in a 0.5-g soil sample. The detection limit of the real-time PCR described by us was calculated to be about two J2 individuals of M. graminicola in a 100-g soil sample. The standard curve generated in this study was based on DNA samples extracted from 0.5 g of soil with different numbers of M. graminicola. Our validation assay showed that the real-time PCR cycle numbers of the positive soil samples were scattered around the standard curve. Therefore, we consider that this standard curve can be used to quantify M. graminicola in field-collected soil samples or to validate the results from other assays as previously suggested (Berry et al. 2008; Yan et al. 2013). The R2 value between the number of M. graminicola determined by microscopic examination and the standard curve was 0.477 (P < 0.05). We believe that the differences between the number of M. graminicola in soil samples counted in microscopic examinations and that determined by real-time PCR were caused by the fact that M. graminicola egg masses and eggs could not be observed in the microscopic examination. In addition, the distribution of nematodes in soil is uneven, which leads to sampling errors. This problem can, however, be solved by collecting more soil from multiple locations in the same field.

    The LAMP assay was considered one of the most promising methods for detection of nematodes because it is simpler, faster, and more sensitive compared with the traditional morphological method and conventional PCR, in which the products could be detected by SYBR Green I dye (LAMP-Green) or LFD (LAMP-LFD). The green florescence observed in the LAMP-Green assay or the positive test obtained in the LAMP-LFD assay can be distinguished by untrained persons. Although LAMP has been used successfully to detect several PPN species, including Bursaphelenchus xylophilus (Kikuchi et al. 2009), Meloidogyne spp. (Niu et al. 2011), Radopholus similis (Peng et al. 2012), and Tylenchulus semipenetrans (Lin et al. 2016; Song et al. 2017), to our knowledge, the LAMP assay for M. graminicola detection has not been reported previously. The LAMP assay developed in this study is very useful for the detection of M. graminicola in soil samples. It is simpler to operate, is rapid and sensitive, and does not require expensive equipment and skilled personnel. The assay can be done using a thermostatic water bath or other containers with a heating source. In addition, the assay is cost effective and environmentally friendly. We believe that this assay is suitable for large-scale field surveys. Because the LAMP assay showed the same detection sensitivity and specificity as real-time PCR and does not require an expensive real-time PCR thermal cycler or electrophoresis equipment, we recommend the LAMP assay for future routine soil surveys for M. graminicola infestation. When needed, the results of the LAMP assay can be further validated through PCR and DNA sequencing using representative samples.

    In conclusion, we established and compared three methods for detection of M. graminicola in soil: conventional PCR, LAMP, and real-time PCR. All three methods exhibited high specificity and sensitivity, although the sensitivity of the LAMP and real-time PCR methods was higher than that of conventional PCR. The LAMP and real-time PCR methods could be utilized as rapid and efficient diagnostic tools for the early detection of M. graminicola in naturally infested field soil depending on the test conditions.

    The author(s) declare no conflict of interest.

    Literature Cited

    Q. C. He and D. W. Wang contributed equally to this manuscript.

    The author(s) declare no conflict of interest.

    Funding: This work was supported by grants from the National Key Research and Development Program of China (2017YFD0201606), the National Natural Science Foundation of China (31871941), and the Hunan Agricultural Science and Technology Innovation Fund Project (2019LS01).