Introduction

Root-lesion nematodes of the genus Pratylenchus are migratory endoparasites that have a wide host range and are distributed worldwide (Castillo and Vovlas 2007). They feed and migrate within root cortical tissue, causing roots necrosis and a reduction in lateral branching of roots, which are responsible for root lesion disease (Vanstone et al. 1998; Castillo and Vovlas 2007). Pratylenchus are considered the third most important economically damaging plant-parasitic nematodes after root-knot nematodes and cyst nematodes (Jones et al. 2013).

To manage root lesion disease, the causative factor must first be identified. Presently, sampling, manual extraction and identification of Pratylenchus species is the only method available to correctly diagnose root lesion disease caused by root-lesion nematodes. Currently, over 70 Pratylenchus species have been reported (Castillo and Vovlas 2007; Mizukubo et al. 2007; Bairwa et al. 2008; Troccoli et al. 2008; Múnera et al. 2009; Palomares-Rius et al. 2010; De Luca et al. 2010, 2012; Palomares-Rius et al. 2014; Wang et al. 2015). Among them, P. zeae is one of the most important species, which can attack various crops, such as potato, corn, cereal and especially sugarcane, causing serious losses (Castillo and Vovlas 2007). For unskilled observers, it is very difficult to correctly identify P. zeae, because there is high intraspecific variability of some morphological characters and some species have very similar morphologies. More recently, P. parazeae was found to parasitize sugarcane in China, with P. parazeae and P. zeae being present in mixed populations (Wang et al. 2015). Polymerase chain reaction (PCR) has been used widely to identify major Pratylenchus spp., including P. zeae. For example, P. zeae from sugarcane can be identified and detected by using SYBR-Green qPCR (Berry et al. 2008); Four Pratylenchus species, including P. zeae, can be discriminated using rDNA-internal transcribed spacer region (ITS) restriction fragment length polymorphism (RFLP) (Li and Zheng 2013); P. zeae can be distinguished from P. parazeae using duplex PCR (Wang et al. 2015). These studies showed that Pratylenchus spp., including P. zeae, can be identified and detected using molecular techniques. However, an expensive thermal cycler is needed for these conventional PCRs, limiting their routine use.

The loop-mediated isothermal amplification (LAMP) assay, a relatively new DNA amplification technique, was developed in 2000 (Notomi et al. 2000). The LAMP reaction is usually performed at an isothermal temperature of 60 to 65 °C, which negates the need for expensive thermal cyclers. All LAMP steps can be done in a single tube within 0.5 to 1h, and the amplification product can be detected by the naked eye using intercalating dyes, such as SYBR Green, propidium iodide or ethidium bromide. LAMP requires a polymerase with a high strand displacement activity and two or three sets of primers that identify six or eight distinct regions on the target gene, which increases the specificity of the reaction (Notomi et al. 2000; Nagamine et al. 2002; Maeda et al. 2005; Duan et al. 2014; Lin et al. 2016). Thus, LAMP has the advantages over conventional PCR in sensitivity, rapidity, specificity and cost.

Recently, LAMP assays have been developed to identify and detect plant-parasitic nematodes, such as several economically important plant-parasitic nematodes, including Bursaphelenchus xylophilus, Meloidogyne spp., Radopholus similis and Tylenchulus semipenetrans (Kikuchi et al. 2009; Niu et al. 2011, 2012; Peng et al. 2012; Lin et al. 2016). However, LAMP has not yet been developed to detect Pratylenchus species. The aim of the current study was to develop a LAMP assay to detect P. zeae based on the mitochondrial DNA (mtDNA) Cytochrome Oxidase I (COI) sequence, and to evaluate its efficiency.

Materials and methods

Nematode populations and DNA extraction

Eight populations of P. zeae from distant geographical areas and different hosts were used. Seventeen other plant-parasitic nematode species collected from various hosts, including eight Pratylenchus species (P. parazeae, P. brachyurus, P. coffeae, P. loosi, P. vulus, P. neglectus, P. scribneri and P. thornei), Pratylenchoides ritteri, Radopholus similis, Hirschmanniella sp., Meloidogyne incognita, M. javanica, Tylenchorhynchus annulatus, Helicotylenchus dihystera, Filenchus sp. and Criconemoides sp., were also used to verify the specificity of the primers (Table 1). All the nematodes were identified using morphological characteristics, together with molecular data. DNA was extracted from individual nematode specimens as described by Subbotin et al. (2008). For Meloidogyne spp., the mtDNA COII/rrnL region was amplified. For the other nematodes, two rDNA fragments the D2D3 expansion domains of large subunit (LSU D2D3) or the internal transcribed spacer (ITS) were amplified. Primers for mtDNA COII/rrnL amplification were C2F3 (5′-GGTCAATGTTCAGAAATTTGTGG-3′) and 1108 (5′-ACCTTTGACCAATCACGCT-3′) (Powers and Harris 1993). Primers for ITS amplification were TW81 (5′-GTTTCCGTAGGTGAACCTGC-3′) and AB28 (5′-ATATGCTTAAGTTCAGCGGGT-3′) (Subbotin et al. 2000). Primers for LSU D2D3 amplification were D2A (5′-ACAAGTACCGTGGGGAAAGTTG-3′) and D3B (5′-TCGGAAGGAACCAGCTACTA-3′) Subbotin et al. (2006). Detailed protocols of the PCR amplification were described in Tanha Maafi et al. (2003). DNA sequencing was performed as described by Zhuo et al. (2010).

Table 1 Pratylenchus species and other nematodes used in this study
Full size table

PCR amplification of the partial mtDNA COI gene

The partial mtDNA COI sequences of all Pratylenchus species, Pratylenchoides ritteri, Radopholus similis and Hirschmanniella sp. were amplified. The primers used were COI F (5′-GATTTTTTGGKCATCCWGARG-3′) and COI R (5′-CWACATAATAAGTATCATG-3′) (He et al. 2005). Detailed protocols of the PCR amplification were described in He et al. (2005). The newly obtained sequences were deposited in the GenBank database under the accession numbers listed in Table 1. The sequences were compared in GenBank using the BLASTn program to analyze intraspecific and interspecific differences among the specimens.

Phylogenetic analyses

At the time of analysis, there were nineteen mtDNA COI sequences of root-lesion nematodes available in GenBank. Additionally, some mtDNA genome-, SSU- and LSU-based phylogenetic studies have indicated that Meloidogyne is closely related to the Pratylenchidae (Subbotin et al. 2006; Holterman et al. 2009; Rybarczyk-Mydłowska et al. 2013; Humphreys-Pereira and Elling 2014; Sun et al. 2014). Thus, to confirm the identity of the newly acquired COI sequences, the new sequences, the known COI sequences of Pratylenchus and several COI sequences of Meloidogyne from GenBank were used in phylogenetic analyses. Outgroup taxa were chosen according to previous molecular phylogenetic analyses of plant-parasitic nematodes (Subbotin et al. 2006; Holterman et al. 2009; Humphreys-Pereira and Elling 2014; Sun et al. 2014). DNA sequences were aligned using MEGA4.0 (Tamura et al. 2003), with default parameters and the alignment quality was examined by eye. The model of base substitution was evaluated using MODELTEST (Posada and Crandall 1998; Huelsenbeck and Ronquist 2001). The Akaike information criterion indicated that we should use the TVM + I + G substitution model to construct the Bayesian consensus tree (lnL = 5479.3257; freqA =0.2849; freqC =0.0975; freqG =0.2188; freqT =0.3988; R(a) = 0.0339; R(b) = 6.1549; R(c) = 3.6032; R(d) = 6.9599; R(e) = 6.1549; R(f ) = 1; Pinva =0.2618; Shape =0.9910). The Bayesian analysis was performed using MrBayes 3.1.0 (Huelsenbeck and Ronquist 2001), running the chain for 1 × 106 generations and setting the ‘burn-in’ at 2500. The Markov Chain Monte Carlo method was used within a Bayesian framework to estimate the posterior probabilities of the phylogenetic trees (Larget and Simon 1999) using the 50% majority rule.

Primer design

Sequences of the partial mtDNA COΙ region in all Pratylenchus species, Pratylenchoides ritteri, Radopholus similis and Hirschmanniella sp. obtained from this study (Table 1) and in Meloidogyne, including M. incognita (KM887154) and M. javanica (KM887156) obtained from GenBank, were chosen as the candidate targets for primer design. A multiple sequence alignment was performed to compare these sequences, using MegAlign program in the Lasergene software (DNASTAR Inc., Madison, WI, USA). The primers PZF3 (5′-GGTTTAGATCTTGATTCTCGG-3′), PZB3 (5′-GTAAAAATAAATCCAAAGTGGCA-3′), PZFIP (F1c + BamHI + F2: 5′-ACCAGGTAAAAACCTTAATTCCGGggatccGCTTATTTTAGGGGGGCT-3′) and PZBIP (B1c + aaaaaa + B2: 5′-TATTCGTTTTGGTCCGGTTTTTTTaaaaaaCAAAATAACCCCAGAGCAAC-3′) were then designed using PrimerExplorer V4 software (Eiken Chemicals, Tokyo, Japan) (Fig. 1). All primers were synthesized by Invitrogen Biotech (Shanghai, China).

Fig. 1
figure 1

Primers used for loop-mediated isothermal amplification to detect Pratylenchus zeae. a schematic representation of the double-stranded target DNA and loop-mediated isothermal amplification inner (FIP and BIP) and outer (F3 and B3) primers. The FIP primer consists of F1c, Bam HI and F2; the BIP primer consists of B1c and B2. b Multiple sequence alignment of partial COI sequences of eight P. zeae isolates and other Pratylenchus species. Arrows and shadows indicate direction and location of the primers

Full size image

LAMP protocol optimization

According to the user manual (New England Biolabs Ltd., London, UK), the LAMP reaction was performed in a 25 μl volume comprising 2.5 μl of the 10 × isothermal amplification buffer, 3.5 μl of 10 mM dNTPS, 1.5 μl of 100 mM MgSO4, 2.5 μl of the 10 × primers (1.6 μM each for the PZFIP and PZBIP primers, and 0.2 μM each for the PZF3 and PZB3 primers), 1 μl of Bst 2.0-DNA polymerase (New England Biolabs Ltd., London, UK), 4 μl of 5 M betaine (Sigma Aldrich Co., St. Louis, MO, USA), and 1 μl of the template DNA.

To optimize the reaction temperature of the LAMP assay, the reactions were performed in a water bath (YiHeng Technical Co. Ltd., Shanghai, China) at 62, 64, 66 and 68 °C for 90 min. Then, 10 μl of the products were separated on standard 2% agarose gels stained with Goldview (Toyobo Co., Shanghai, China). Trans2K DNA markers (Transgene, Beijing, China) were used to determine the molecular size of the bands, which were photographed under UV light using an Alphalmager system (Alpha Innotech Co., San Leandro, CA, USA). The optimal temperature was determined according to the band intensity obtained at different temperatures. The experiment was performed three times.

To optimize the reaction time, the LAMP assay was carried out using the TP860 system (Takara, Dalian, China). Firstly, DNA was isolated from 1 or 100 individuals of P. zeae. Subsequently, a 10-fold dilution of the DNA from one P. zeae, and 1.25-fold, 2.5-fold and 10-fold dilutions of the DNA from the 100 P. zeae were obtained. All these DNA samples were subjected to real-time PCR. The real-time LAMP reaction was performed in a 25-μL volume comprising 2.5 μL of 10 × isothermal amplification buffer, 3.5 μL of 10 mM dNTPs, 1.5 μL of 100 mM MgSO4, 2.5 μL of the 10 × primers, 1 μL of Bst 2.0-DNA polymerase, 1 μL of template DNA, 4 μL of 5 M betaine and 2 μL EVA green (Mayene, Guangzhou, China). Thirty-five reaction cycles were performed at 64 °C, and each cycle lasted 2 min: the signal after each cycle was determined.

LAMP specificity analysis

To determine the specificity of the LAMP reaction, the LAMP products were digested with Bam HI restriction enzyme (Thermo Fisher Scientific, Shanghai, China) and analyzed by electrophoresis. The restriction enzyme site was chosen based on the appropriate sequence information (shown in the inner primer PZFIP). The restriction digestion system comprised 2 μl of the 10 × FastDigest green buffer, 2 μl of LAMP products, 1 μl of Bam HI and 15 μl of double-distilled water. The reaction was conducted at 37 °C for 30 min. In addition, the PCR product generated using the outer primers PZF3/PZB3 was sequenced to determine the specificity of the outer primers. The PCR reaction was performed in a 25 μl volume comprising 12.5 μl of the 2 × Taq Green Mix (Dingguo, Guangzhou, China), 0.5 μl each of the outer primers and 1 μl of the template DNA. The amplification program was as follows: 95 °C for 3 min; followed by 35 cycles at 95 °C for 30 s, 40 °C for 30 s and 72 °C for 40 s; and an extension at 72 °C for 7 min.

To determine whether the primers were specific to P. zeae, DNA was isolated from 25 nematode populations comprising 18 different nematode species (Table 1). Each sample was loaded in triplicate and a negative control sample was also prepared, using double-distilled water instead of the DNA template. LAMP was then performed as described above. The LAMP results could be inspected visually with the naked eye in daylight by the addition of 1 μl of 1000 × SYBR Green I solution (Dingguo, Guangzhou, China) to the amplification products.

LAMP sensitivity analysis

DNA was firstly extracted from one and 100 individuals of P. zeae. Then, a 10-fold dilution of the DNA from one P. zeae or 100 P. zeae was obtained. Subsequently, the four concentrations of DNA were subjected to LAMP and conventional PCR separately. The LAMP reaction was performed as described above. The conventional PCR was conducted using the primers COI F and COI R, as described above. A negative control sample was also prepared, using double-distilled water instead of DNA template. Each sample was run in triplicate and the experiment was performed three times.

Results

Analysis of the partial sequences of mtDNA COI

PCR amplifications of the partial mtDNA COI regions indicated that all Pratylenchus species, Pratylenchoides ritteri, Radopholus similis and Hirschmanniella sp. could generate a single fragment of approximately 400 bp. The sequencing results showed that these COI amplified fragments varied from 414 to 417 bp. All these sequences were deposited in GenBank and the accession numbers are listed in Table 1. When using R. similis as the outgroup taxon, the Bayesian consensus tree (Fig. 2) demonstrated that all the newly acquired COI sequences of Pratylenchus and some known COI sequences of Pratylenchus were clustered within the same clade. All the newly acquired COI sequence of the eight P. zeae populations and the six known COI sequences of P. zeae formed a strongly supported clade with 100% support and short branch differences, which was sister to the P. parazeae clade, with 100% support. Thus, the Bayesian consensus tree confirmed the identities of these newly acquired COI from Pratylenchus.

Fig. 2
figure 2

Bayesian consensus tree inferred from partial COI sequences among different species of Pratylenchus under the TVM + I + G model. Posterior probability values exceeding 50% are provided for the appropriate clades

Full size image

The genetic variation among Pratylenchus species was examined by aligning the mtDNA COΙ sequences. The results indicated that intraspecific variations among the mtDNA COI sequences of the eight P. zeae populations were 0.2–2.4% (1–7 bp) without gaps, and the interspecific variations between P. zeae and other Pratylenchus species were 25.4–35.1% (103–130 bp) with 2–19 gaps.

Optimization of LAMP reaction conditions

When LAMP was performed based on P. zeae DNA at 62–68 °C for 90 min, electrophoresis showed the typical ladder-like bands produced at 62–66 °C. The bands were most intense at 64 °C, and no ladder-like bands appeared in the negative control (Fig. 3). The fluorescence intensity of DNA extracted from 0.1 to 100 nematodes started to increase after 25 min of reaction time, and reached a plateau at 40 min in the DNA sample of 0.1 nematode (Fig. 4). Therefore, the most suitable conditions for the LAMP assay were determined as 64 °C for 40 min.

Fig. 3
figure 3

Temperature optimization for the loop-mediated isothermal amplification reaction. The reaction was performed for 90 min at temperatures ranging from 62 °C to 68 °C. Lane M: Trans 2 K DNA Markers; CK: negative control (distilled water)

Full size image
Fig. 4
figure 4

Optimization of the duration of the loop-mediated isothermal amplification reaction. Real-time PCR was used to optimize the reaction time. Numbers on the x-axis represent the reaction time. Lines labeled 0.1 and 1 represent 10-fold and 1-fold dilutions of the DNA template from one nematode, respectively. Lines labeled 10, 40, 80 and 100 represent 10-fold, 2.5-fold, 1.25-fold, and 1-fold dilutions of the DNA template from 100 nematodes, respectively. CK: negative control (distilled water)

Full size image

Specificity of the LAMP assay

Electrophoresis indicated that the products of the LAMP assay were digested completely using the restriction enzyme Bam HI (Fig. 5). In addition, the sequence of the 220-bp product amplified using the outer primers PZF3/PZB3 was determined. The sequence revealed a 100% match with the mtDNA COI sequence of original P. zeae population (data not shown). Based on the visual evaluation with SYBR Green I using DNA isolated from P. zeae and other nematodes, the P. zeae samples produced positive reactions, i.e. the SYBR Green I turned yellow-green, while the other nematode samples and the control double-distilled water sample produced negative reactions, i.e. the SYBR Green I remained brown or orange (Fig. 6). These results demonstrated that the LAMP assay was highly specific for P. zeae.

Fig. 5
figure 5

Restriction enzyme analysis of the loop-mediated isothermal amplification products. Lanes 1 and 2 represent DNA templates from two different populations of Pratylenchus zeae. M: Trans DNA Marker II; CK: negative control (distilled water)

Full size image
Fig. 6
figure 6

Specificity of the loop-mediated isothermal amplification (LAMP) assay for Pratylenchus zeae and other nematodes. SYBR Green I was added to the LAMP products from: (tubes 1–8) eight different populations of Pratylenchus zeae; (tubes 9–25) populations of P. parazeae, P. brachyurus, P. coffeae, P. loosi, P. neglectus, P. scribneri, P. thornei, P. vulnus, Hirschmanniella sp., Pratylenchoides ritteri, Tylenchorhynchus annulatus, Filenchus sp., Criconemoides sp., Helicotylenchus dihystera, Meloidogyne incongnita, M. javanica and Radopholus similis; (tube 26) negative control (distilled water)

Full size image

Sensitivity of the LAMP assay

The LAMP and conventional PCR assays, performed on different serial dilutions of the DNA extracted from P. zeae, demonstrated the increased sensitivity of the LAMP assay. The conventional PCR could only detect the DNA equivalent to one individual nematode; however, the LAMP assay could detect approximately a tenth of the DNA extracted from a single nematode, i.e. 10-fold dilution of the template DNA from one nematode (Fig. 7). These demonstrated that the LAMP assay is 10 times more sensitive than conventional PCR.

Fig. 7
figure 7

Sensitivity of the loop-mediated isothermal amplification (LAMP) and convention PCR methods. a conventional PCR and b LAMP products. c LAMP products visualized with SYBR green I using natural daylight. Lane 1 or tube 1 represent the DNA template from 100 Pratylenchus zeae; lane 2 or tube 2 represent 10-fold dilution of the DNA template from 100 P. zeae; lane 3 or tube 3 represent the DNA template from one P. zeae; and lane 4 or tube 4 represent 10-fold dilution of the DNA template from one P. zeae; M: Trans 2 K DNA Markers; CK: negative control (distilled water)

Full size image

Discussion

It is challenging to identify Pratylenchus species because of the high intraspecific variability and the interspecific overlap of some morphological characteristics. Therefore, different PCR techniques, such as PCR amplification using species-specific primers, PCR-RFLP and qPCR based assay, have been used to identify and detect Pratylenchus spp. (Castillo and Vovlas 2007; Berry et al. 2008; Li and Zheng 2013; Wang et al. 2015). These conventional PCR techniques require an expensive thermal cycler to achieve amplification at different temperatures, which limits their routine use, particularly in certain developing countries and in the field (Lin et al. 2016). About ten years ago, a fast and inexpensive technology for sequence-specific DNA amplification, namely LAMP, was invented. LAMP does not require changes in temperature during the reaction, meaning that LAMP does not require expensive thermal cyclers. LAMP has been applied successfully to detect pathogenic microorganisms, including plant-parasitic nematodes (Fu et al. 2011; Duan et al. 2014; Lin et al. 2016). Several important plant-parasitic nematodes species, including B. xylophilus, M. incognita, M. enterolobii, R. similis and T. semipenetrans, have been detected using LAMP (Kikuchi et al. 2009; Niu et al. 2011, 2012; Peng et al. 2012; Lin et al. 2016). In this study, a new LAMP method, based on the partial mtDNA COI region, was developed successfully to detect P. zeae. This is the first application of LAMP to identify a root-lesion nematode. To develop the LAMP method, eight P. zeae populations, one from the USA and seven from China, were included. The seven populations from China were collected from remote locations and isolated from several different hosts. It has been reported that P. zeae exhibits variation in the shape of its stylet knobs, from more or less rounded to markedly cup-shaped anteriorly; and in its tail shape, including almost pointed, narrowly rounded to subacute (Roman and Hirschmann 1969; Castillo and Vovlas 2007). All of these variations were observed in our populations. More recently, the intraspecific variation for mtDNA COI sequences of P. zeae from distant localities, including the USA, South Africa, Suriname and Kenya (Troccoli et al. 2016), was reported as 0–1.1%, which was lower than the 0.2–2.4% in our populations. Thus we believe that the P. zeae populations used in this study have enough diversity.

Some studies have shown that ITS-rDNA is an important marker to identify and differentiate nematode species (Gasser and Newton 2000; Powers et al. 1997; Subbotin et al. 2001). However, ITS sequences within a species or an individual species of some Pratylenchus might have extensive polymorphisms (Waeyenberge et al. 2009; De Luca et al. 2011), making it difficult to design species-specific primers. Wang et al. (2015) reported that intraspecific variations in the ITS sequences of P. zeae could reach 9.5%. In this study, the intraspecific variations in the ITS sequences of eight P. zeae populations were 3.0–9.1% (data not shown), while intraspecific variations for mtDNA COI sequences were only 0.2–2.4%. In addition, interspecific variations of the COI region between P. zeae and other Pratylenchus species ranged from 25.4–35.1%. Therefore, we decided to use the COI sequences to design LAMP primers to detect P. zeae. The results showed that the LAMP assay could detect P. zeae specifically and there was no cross-reaction with other Pratylenchus species. The mtDNA COI gene is one of the most popular markers for identifying animals (Derycke et al. 2010); however, only a few COI sequences have been obtained from Pratylenchus species (Sultana et al. 2013; Palomares-Rius et al. 2014; Troccoli et al. 2016). In the present study, sixteen COI sequences from eight P. zeae populations and eight other Pratylenchus species were obtained. Among these sequences, the COI sequences of P. scribneri, P. loosi and P. brachyurus were amplified for the first time. Intraspecific comparisons within P. zeae showed that the COI sequences varied by <0.03, while the interspecific comparisons between the nine Pratylenchus species gave values >0.2. Moreover, the LAMP assay using the primers that targeted the COI sequence of P. zeae was highly specific and could distinguish P. zeae from other Pratylenchus species. These results suggested that the COI gene might provide good resolution at the Pratylenchus species level. More Pratylenchus samples could be included to confirm the utility of COI to identify Pratylenchus species.

In the present study, the LAMP assay was 10 times more sensitive than conventional PCR, which was similar to the results for the detection of Tylenchulus semipenetrans in soil (Lin et al. 2016). In addition, the results of the LAMP reaction could be observed visually by the addition of SYBR Green I staining to the products, which is not possible with conventional PCR. The LAMP is also rapid, detecting P. zeae in 40 min. In addition, LAMP only requires a constant-temperature bath of 64 °C. In conclusion, the developed LAMP method provides a good alternative to conventional PCR for the rapid, specific and simple detection of P. zeae. Hence, it will be useful for the rapid diagnosis of root lesion disease caused by P. zeae.