Although Phytopythium is a relatively newly defined genus different from the Pythium and Phytophthora genera, organisms in all three genera have similar micromorphology and growth patterns. Therefore, in this study, 20 representative isolates of Phytopythium, Pythium, and Phytophthora, as well as of selected Fusarium species that are also common causes of brown root rot in plants (Table 1) were selected for this study. The 20 isolates representing 19 brown root rot species were collected from different provinces in China and were maintained at the Institute of Bast Fiber Crops, Chinese Academy of Agricultural Sciences (Changsha, China). All isolates were grown on potato dextrose agar (PDA) medium at 25°C for 3–5 days before extracting DNA. Mycelia were harvested, collected, and stored at −20°C. Mycelial DNA was extracted by a Rapid Extraction Kit for Fungi Genomic DNA (Aidlab, Beijing, China) following the user’s manual. Conventional PCR amplification was carried out with the specific primers of P. vexans (PvF1/PvR1) (Spies et al., 2011b). The ITS region (ITS1, 5.8S, and ITS2) of all 20 isolates was amplified and sequenced using primers V9G (5′-TTACGTCCCTGCCCTTTGTA-3′) and LS266 (5′-GCATTCCCAAACAACTCGACTC-3′) (van den Ende and de Hoog, 1999).

In this study, the ITS sequence was selected as the target gene for LAMP assay. To identify the portions of ITS sequence that are unique to P. vexans, we performed multiple sequence alignment including sequences of tested isolates as well as sequences of type strains of P. vexans (GenBank accession number: HQ643400.2), P. irregulare (AY598702.2), Pythium ultimum (AY598657.2), Pythium spinosum (AY598701.2), Pythium aphanidermatum (AY598622.2), Pythium myriotylum (AY598678.2), P. litorale (NNIBRFG9359), Pythium sylvaticum (AY598645), P. helicoides (AB108026), Pythium recalcitrans (KJ716861), Pythium oligandrum (AY598618), Pythium heterothallicum (AY598654), Pythium carolinianum (HQ643484.1), Pythium guiyangense (AY987037), and Phytophthora capsici (FN257939) that were retrieved from the NCBI databases. The program ClustalW2¹ (Larkin et al., 2007) was used to perform multiple sequence alignment to find the regions within ITS which were specific for P. vexans. Based on the aligned sequences, a set of primers was designed. The LAMP primers were designed using PrimerExplore V5,² all the parameters were set by default, and primers were synthesized by TSINGKE Biotechnology Co. Ltd. Nine primers sets were tested (Supplementary Table 1), however, the best primers (Pv8 primers set) were screened by specificity and sensitivity test. In addition, a phylogenetic tree was constructed with the maximum likelihood method implemented in MEGA 7.0 (Kumar et al., 2016). A Poisson correction was used for multiple substitution models and a pairwise deletion was used for handling missing data. Statistical support for individual branches of the phylogenetic tree showing taxa relationships was assessed with 1,000 bootstrap replicates.

The LAMP reaction was carried out in a total volume of 25 μL containing 1.6 μM of each FIP and BIP primer, 0.2 μM of each F3 and B3 primer, 20 mM Tris–HCL (pH 8.8), 10 mM KCl, 0.1% Tween20, 0.8 M betaine (Shanghai Yuanye Bio-Technology, Shanghai, China), 8 mM MgSO₄, 10 mM (NH4)₂SO₄, 1.4 mM dNTPs, and 8U Bst DNA polymerase large fragment (New England Biolabs Japan, Tokyo, Japan), 180 μM HNB (Macklin, Shanghai, China), and 2 μL template DNA (Feng W. et al., 2015). The mixture was incubated at 64°C for 60 min to perform the LAMP reaction. The genomic DNA template of P. vexans was diluted in a 10-fold series after concentration determination, so that the DNA concentration gradient was 100 ng/μL, 10 ng/μL, 1 ng/μL, 100 pg/μL, 10 pg/μL, 1 pg/μL, 100 fg/μL, 10 fg/μL, and 1 fg/μL. Among them, 1 pg DNA contained 23.4 copies of the P. vexans genome.

Based on the above LAMP reaction system, the reaction time and temperature, the concentration of MgSO₄, betaine, dNTPs were optimized using a gradient of values. Specifically, the reaction temperatures were set at 60, 62, 64, 66, and 68°C; the reaction times were set at 10, 20, 30, 40, 50, 60, 70, and 80 min; the concentrations of betaine were 0, 0.2, 0.4, 0.6, 0.8, 1.0, and 1.2 M; of MgSO₄ were 0,2, 4, 6, 8, and 10 mM; and of dNTP were 0, 0.2, 0.4, 0.6, 0.8, 1.0, 1.2, 1.4, and 1.6 mM. In all optimizations, a negative control was set up with sterilized distilled water instead of DNA template. The LAMP products were visualized by adding HNB before amplification. Samples that turned blue were considered positive, while samples that remained violet were negative. In addition, the LAMP products were detected through electrophoresis in 1.5% agarose gel by staining with GelRed (TSINGKE Biological Technology, Beijing, China), and positive reactions resulted in a ladder-like banding pattern. The reaction was performed with three biological repeats.

Using the selected primers, the genomic DNA of 20 strains were used as templates, and sterilized distilled water instead of DNA template was used as negative control. The optimized LAMP reaction system was used for amplification. After the reaction, a color change in the products was assessed with the naked eyes of at least two independent investigators.

To evaluate the application of the developed LAMP protocol as a tool for the diagnosis of P. vexans in infected plants, ramie cultivar “Zhongzhu No. 1” roots were inoculated as described below. Two-week-old seedlings (approximately 20–30 cm tall) were prepared via the cutting propagation method (Yu et al., 2015). To inoculate ramie plants, mycelial disks (8 mm in diameter) obtained from 3-day-old culture of each of the 20 isolates of Phytopythium spp., Pythium spp., and Phytophthora spp. on PDA were placed upside-down on the roots. Plants inoculated with agar disks without mycelia were used as controls. Both the inoculated and control plants were grown in the greenhouse at 26 ± 1°C with a 12 h photoperiod. After 10 days, tissues surrounding the infected region was harvested for DNA extraction by a Rapid Extraction Kit for Fungi Genomic DNA (Aidlab, Beijing, China). These DNA samples were evaluated using the developed LAMP technique.

In total, two P. vexans isolates, two isolates of two other Phytopythium species, 11 isolates of 11 Pythium species, one P. capsici isolate, and four isolates of four Fusarium species obtained from different provinces in China (Table 1) were used to test the specificity of the LAMP primers and reaction conditions. All tested strains were first analyzed with the above-mentioned P. vexans diagnostic primers PvF1 and PvR1, a reported specific primer of P. vexans. The results showed that lanes 1 and 2 are strains of P. vexans, and the other tested strains were not P. vexans (Figure 1). For all 20 isolates, their species identities were confirmed based on their ITS sequences. Based on the ITS sequences of type strains of these species, we constructed a phylogeny among the tested species (Supplementary Figure 1). The phylogenetic relationships among these strains and species are overall consistent with what has been described previously for these organisms. Specifically, the fungal genus Fusarium is distinct from the oomycetes. Among the three oomycete genera, the genus Phytophthora was diverged first, followed by Phytopythium and Pythium (Supplementary Figure 1).

To identify P. vexans -specific LAMP primers at the ITS region, multiple sequence alignment of ITS sequences between P. vexans and other 18 isolates was carried out. The variable regions of ITS between P. vexans and other isolates were found (Figure 2). Based on the sequence variability patterns, nine primer sets were designed using the online tool PrimerExplore V5 (Supplementary Table 1). Then, according to the LAMP amplification system and procedure, reactions were set up for all the 20 isolates. The specificity and sensitivity of various primer combinations were compared. Finally, among these nine sets of primer combinations, the Pv8 set consists of four core primers Pv8F3 (5′-CGTGTAGTCGTCGGTTGTT-3′), Pv8B3 (5′-CGCAAATCGAGCAATCCACT-3′), Pv8FIP (5′-CGCGTCCGACTTTAAAGGGACTTGCAGATGTGAGGTTG TCTC-3′), and Pv8BIP (5′-GTTTTGTGCTTGATGGGGTG CGGCCATCGCCAAAGGTCAC-3′) performed the best and was the most consistent for detecting P. vexans (Figure 2). Using the Pv8 primer combinations, positive amplifications were obtained only with the two P. vexans isolates in Table 1 while no amplification product (Figure 3A) nor color change (Figure 3B) was observed for the remaining 18 isolates.

A 25 μL LAMP reaction system was established with P. vexans DNA as template for optimizing the reaction temperature and time, and the concentrations of dNTPs, MgSO₄ and betaine in the system. The evaluated reaction temperatures were set at 60, 62, 64, 66, and 68°C; the reaction times were set at 10, 20, 30, 40, 50, 60, 70, and 80 min; the concentrations of betaine were 0, 0.2, 0.4, 0.6, 0.8, 1.0, and 1.2 M; of MgSO₄ were 0, 2, 4, 6, 8, and 10 mM; and of dNTP were 0, 0.2, 0.4, 0.6, 0.8, 1.0, 1.2, 1.4, and 1.6 mM. In each series of amplifications, we included a negative control where pure sterilized distilled water without any DNA template was used. The results showed that under the recommended LAMP condition, the amplification products could be detected between reaction time 20–80 min. However, the brightest amplification products and the amplification efficiency was reached at a reaction time of 60 min (Figure 4A). Thus, we chose 60 min as the reaction time to observe the LAMP amplified products, and with coloration intensity from HNB staining and/or DNA band brightness through agarose gel electrophoresis as indicators of amplification efficiency in each reaction. Among the tested temperature range 60–68°C, we found little difference among the five temperature treatments in both color change and in the amounts of amplified products as shown on the electrophoretic gel. The temperature experiments indicated that the LAMP reaction condition was robust in the 60–68°C temperature range; therefore, for the following experiments, we chose the optimum temperature of Bst DNA polymerase, i.e., at 64°C (Figure 4B). Among the concentrations of dNTP from 0.2 to 1.6 mM, the amplification products could all be detected, but the highest amplification product was observed at the dNTP concentration of 1.4 mM (Figure 4C). Similar to the limited effect of dNTP concentration change, the change of concentration had relatively little effect on amplification efficiency, but the highest amplification efficiency was observed at the concentration of 6.0 mM MgSO₄ (Figure 4D). In contrast to MgSO₄, where no MgSO₄ resulted in no amplification product, betaine was found to be not essential in our LAMP protocol. However, because betaine has the effect of stabilizing enzyme activity and promoting melting (Rees et al., 1993; Notomi et al., 2000), to improve the stability of the reaction system, especially for field applications, 0.8 M betaine where the color change was the most obvious was added to the system (Figure 4E). In conclusion, the optimal LAMP reaction system was determined as follows: 1.6 μM FIP/BIP, 0.2 μM F3/B3, and 8U Bst DNA polymerase, 1× isothermal amplification buffer [20 mM Tris–HCL (pH 8.8), 10 mM (NH₄)₂SO₄, 50 mM KCl, 0.1% Tween20], 6 mM MgSO₄, 1.4 mM dNTPs, 0.8 M betaine, 180 μM HNB, 2 μL DNA template in a total volume of 25 μL with the final volume adjusted with sterilized distilled water, with the reaction temperature of 64°C for 60 min.

To determine the sensitivity of primer set Pv8, ten-fold serial dilutions of P. vexans genomic DNA were tested. The DNA template concentrations ranged from 10 ng/μL to 1 fg/μL. The detection limit of the conventional PCR assay using the specific primers Pv8F3 and Pv8R3 was 100 pg/μL DNA (∼2,350 copy of the P. vexans genome) (Figure 5A). In comparison, as indicated by a color change and DNA amplification of the reaction products, the lowest limit of P. vexans detection was 1 pg/μL DNA (∼24 copy of the P. vexans genome) for the LAMP assay (Figures 5B,C). This indicates that the LAMP assay has a wider dynamic range, with a sensitivity at least 100 times greater than that of conventional PCR for the detection of P. vexans.

All 15 Phytopythium and Pythium isolates were inoculated onto the “Zhongzhu No. 1” cultivar of ramie plants. After 10 days at 25°C, the fibrous roots of all inoculated plants turned red-brown surrounding the inoculation site. However, no symptom was observed on the roots of negative control plants (Supplementary Figure 2). Genomic DNA was extracted from the infected region and subjected to the LAMP reaction. The results of electrophoresis showed that, except for P. vexans, no DNA bands were produced in the DNA system of ramie roots infected by other 13 tested pathogens (Figure 6A). The results were confirmed based on color change as shown in Figure 6B. Both assay systems indicate that LAMP amplification reaction using primer set Pv8 can accurately detect P. vexans in ramie roots.