Development and application of a rapid and visual loop-mediated isothermal amplification for the detection of Sporisorium scitamineum in sugarcane

Smut is a fungal disease with widespread prevalence in sugarcane planting areas. Early detection and proper identification of Sporisorium scitamineum are essential in smut management practices. In the present study, four specific primers targeting the core effector Pep1 gene of S. scitamineum were designed. Optimal concentrations of Mg2+, primer and Bst DNA polymerase, the three important components of the loop-mediated isothermal amplification (LAMP) reaction system, were screened using a single factor experiment method and the L16(45) orthogonal experimental design. Hence, a LAMP system suitable for detection of S. scitamineum was established. High specificity of the LAMP method was confirmed by the assay of S. scitamineum, Fusarium moniliforme, Pestalotia ginkgo, Helminthospcrium sacchari, Fusarium oxysporum and endophytes of Yacheng05-179 and ROC22. The sensitivity of the LAMP method was equal to that of the conventional PCR targeting Pep1 gene and was 100 times higher than that of the conventional PCR assay targeting bE gene in S. scitamineum. The results suggest that this novel LAMP system has strong specificity and high sensitivity. This method not only provides technological support for the epidemic monitoring of sugarcane smut, but also provides a good case for development of similar detection technology for other plant pathogens.

fields. This method is more reliant on visual assessment, but sometimes it is difficult to confirm infection by phenotypic symptom observation due to a latent infection of pathogens. Additionally, prevalence investigation and data collection must be carried out through the full disease epidemic period during crop season, which can last between 8 to 10 months. All these methods are costly and remain vulnerable to environmental influence 1 . Furthermore, it typically takes one week to separate and culture pathogens and conduct morphological identification 9 . Electron microscopy, on the other hand, requires expensive equipment, and an assay of serological diagnosis needs high quality antibodies 10 . In 1996, Albert and Schenck designed the amplification primer bE4/bE8 that targets the conserved sequence of the bE gene in S. scitamineum for the conventional PCR test 11 . Using the above described methods it was not possible to make timely and accurate detections of the target pathogens due to the long test period, high costs, low sensitivity and vulnerability to the influence of external environmental conditions. Our previous work designed the specific primer bEQ-F/bEQ-R and the TaqMan probe targeting the bE gene, and we have developed a method to detect S. scitamineum based on real-time quantitative fluorescence PCR 12 . By using this assay, the content of S. scitamineum in resistant/susceptible sugarcane genotypes could be detected, but it required expensive quantitative PCR instrument and suffered from the high cost of reagents and other consumables 12 .
Loop-mediated isothermal amplification (LAMP) is an isothermal amplification technology invented by Notomi et al. in 2000 which can complete automatic looping, strand displacement and DNA synthesis using Bst DNA polymerase and four specific primers (two inner and two outer primers) 13 . The two inner primers, which are called the forward inner primer (FIP) and the backward inner primer (BIP), contain two distinct sense and antisense sequences of the target DNA, respectively. After initiation by one of the inner primers, the following strand displacement DNA synthesis in the LAMP reaction is primed by an outer primer (either F3 or B3). A released single-stranded DNA serves as a template for DNA synthesis facilitated by the second inner and outer primers that hybridize to the other end of the target. This results in a DNA sequence which can form a stem-loop structure at both ends. This stem-loop DNA is the starting material for the subsequent LAMP auto-cycling amplification. An accumulation of 10 9 copies of target DNA are observed in less than 1 h. The amplification products are stem-loop DNAs with several inverted repeats of the target and cauliflower-like structures with multiple loops 13 . The LAMP assay has high amplification rates, and is time efficient, easy to manipulate and results in visible products 13 . It does not rely on expensive instruments or temperature gradients. These advantages make LAMP a good onsite rapid detection method and it has been widely used to detect viruses, bacteria and fungi [14][15][16][17] . Nie designed four primers specific to the coat protein gene in Potato virus Y (PVY) and established a LAMP system, a one-step RT-LAMP and a two-step RT-LAMP system for the detection of PVY 14 . The results showed that 234 of 240 samples were accurately determined 14 . Nagdev et al. designed primers according to the sequence of the IS6110 gene of Mycobacterium tuberculosis, and found that the sensitivity and specificity of the optimized LAMP systems were 88.23% and 80%, respectively, which was higher than that of nested PCR (52.9% sensitivity) 15 . There have also been some reports on the application of LAMP technology in the detection of sugarcane pathogenic bacteria and transgenic lines 18 Colletotrichum falcatum in sugarcane, with sensitivity 10 times higher than that of conventional PCR 21 . Zhou et al. optimized a LAMP system to detect cry1Ac transgenic sugarcane, with a sensitivity of 10-100 times higher than that of conventional PCR 22 . However, there is still no report on using LAMP to detect S. scitamineum in sugarcane.
Doehlemann et al. indicated that the protein essential during penetration 1 (Pep1) of Ustilago maydis was a secreted effector required for its successful invasion into plant cells, concluding that the conservative feature of Pep1 has a conserved function essential for establishing compatibility that was not restricted to the maize-U. maydis interaction 23 . Hemetsberger et al. showed that Pep1 could suppress plant immunity by inhibiting plant peroxidase, which served as an effective inhibitor for early defense reactions in plants 24 . Hemetsberger et al. also revealed that Pep1 was a core fungal effector in monocot and dicot smut diseases, with a highly conserved sequence 25 . Hence, in this study, specific primers were designed to target the Pep1 gene of S. scitamineum (gene ID: SmutADNA4_GLEAN_10005329_Smut) identified by homology in the S. scitamineum whole genome sequence obtained in our previous research 26 . The single factor experiment combined with orthogonal experimental design was conducted to optimize the three impact factors of the LAMP reaction system (Mg 2+ concentration, primer ratio and Bst DNA polymerase content) for S. scitamineum detection. The sensitivity and specificity of the developed method were also verified. In addition, the accuracy was verified by determining S. scitamineum in artificially inoculated sugarcane. This study aims to establish a LAMP system for rapid detection of S. scitamineum in sugarcane before symptoms of smut disease appear, laying a foundation for epidemiological studies on sugarcane, disease control and exit-entry quarantine in the exchange of sugarcane germplasms.

Results
Single factor experiment of LAMP. In order to obtain a LAMP system suitable to detect S. scitamineum, a single factor optimization experiment was conducted on a Mg 2+ concentration (5.75 mmol/L, 6.00 mmol/L, 6.25 mmol/L and 6.50 mmol/L), an inner vs. outer primer ratio (2:1, 4:1, 6:1 and 8:1) and the Bst DNA polymerase content (2.0 U, 4.0 U, 6.0 U and 8.0 U) in the LAMP system (Figs 1-3). Appropriate concentration ranges were screened. Combining tests of SYBR Green I method and product electrophoresis, we found that with the optimal conditions of Mg 2+ (5.75 mmol/L, 6.00 mmol/L, 6.25 mmol/L and 6.50 mmol/L) (Fig. 1), ratios of inner vs. outer primer (4:1, 6:1 and 8:1) (Fig. 2) and Bst DNA polymerase (4.0 U, 6.0 U and 8.0 U) (Fig. 3), flavogreen fluorescent color and amplification bands were displayed. However, in the negative and blank controls, the tubes maintained the orange color in the amplification products of the SYBR Green I test and no band was detected in electrophoresis.    (Table 1). Color changes were observed in the amplification when concentrations of Mg 2+ , inner and outer primers and Bst DNA polymerase were altered. As shown in Fig. 4A, combinations 1, 2, 5, 6, 11 and 16 exhibited no or extremely weak flavogreen fluorescent signals in the SYBR Green I test. Additionally, there were few or no amplification band(s) displayed after electrophoretic analysis (Fig. 4B). LAMP products of combination 7 showed high flavogreen fluorescence, and displayed bright typical ladder-like bands after electrophoresis, apparent in distinction from the negative and blank controls, whereas combinations 4, 8 and 10 showed intensive flavogreen fluorescent signals but weaker band brightness than combination 7 in electrophoresis. For other combinations, results were not as good as combination 7. Thus, the LAMP reaction conditions in combination 7 were optimal for detecting S. scitamineum in sugarcane, which consisted of Mg 2+ at a concentration of 6.00 mmol/L, a ratio of inner vs.  Specificity of LAMP. To know whether the optimized LAMP system was specific, the common sugarcane pathogenic DNA was tested. Only target DNA of S. scitamineum and pMD19-T-Pep1 plasmid exhibited positive reactions, while no positive reaction was observed in the other six commonly encountered sugarcane pathogens, i.e. Fusarium moniliforme, Pestalotia ginkgo, Helminthospcrium sacchari, Fusarium oxysporum, and two endophytes from Yacheng05-179 and ROC22 (Fig. 5), indicating the newly developed method has good specificity for detecting S. scitamineum.
Comparison of the sensitivity of LAMP and conventional PCR assay. The LAMP assay showed excellent results with regard to sensitivity, compared to conventional PCR by conducting amplification using ten-fold diluted samples of S. scitamineum genomic DNA (10 6 copies/μL-1 copies/μL) as a template. As indicated by the color transition of LAMP reaction products (Fig. 6A), the lowest limit of S. scitamineum detection was 10 4 copies/μL for LAMP, whereas in conventional PCR for Pep1 (Fig. 6B) and bE tests (Fig. 6C), the detection limits were 10 4 copies/μL and 10 6 copies/μL, respectively. This implied that the LAMP assay had a wider dynamic range and was nearly 100 times more sensitive than conventional PCR for detection of bE of S. scitamineum 11 .  Detection of S. scitamineum in sugarcane inoculated with smut pathogen. In order to test the reliability of the LAMP reaction system optimized in this study, we inoculated buds of ROC22 with a spore suspension of S. scitamineum. At 0 d, 1 d, 2 d, 5 d, 7 d, 10 d and 14 d, genomic DNA of the inoculated buds was isolated and tested by the developed LAMP assay and a conventional PCR. A total of six sugarcane samples after inoculation with S. scitamineum were detected by the LAMP method developed in this study, and the conventional PCR assay targeting bE gene and Pep1 gene. No amplification was observed in the mock (0 h), negative and blank controls (Fig. 7). Positive reactions were displayed in all the infected samples, as was the case in the positive controls by LAMP assay (Fig. 7A) and conventional PCR assay for the Pep1 gene in S. scitamineum (Fig. 7B), indicating a positive detection rate of 100% for genomic DNA of cane buds incubated for 1 d to 14 d after inoculation.
In the conventional PCR reaction for the bE gene (Fig. 7C), no visible band could be observed at 5 d, suggesting a positive detection rate of 83.33%.

Discussion
Sugarcane smut is one of the most prevalent diseases in sugarcane industry worldwide. High sensitivity and efficiency in the detection of S. scitamineum infection is desirable in the process of sugarcane smut resistance breeding. Currently, the common methods for S. scitamineum detection involve field disease symptom investigation, electron microscopic observation, and serologic diagnosis, which are complicated to manipulate. In addition, accuracy and sensitivity vary greatly among different approaches 1,10 . Albert and Schenck designed the primers bE4 and bE8 targeting the bE gene for detection of S. scitamineum by conventional PCR 11 . In our previous study 12 , a TaqMan real-time PCR assay for detection and quantification of S. scitamineum, which allowed detection of the target pathogen in the infected sugarcane plants, was established. However, this method requires special equipment, including a PCR device and a real-time quantitative fluorescence PCR device, which raises the instrumental and time costs of pathogen detection. By LAMP, the reaction can be carried out in a PCR tube incubated at one certain temperature for nearly 1 h 13 . In addition, the operation of opening the tube cover is not required because inspection of the detection results due to a visual color change makes it possible to complete the target gene amplification and product detection in one step, which reduces the danger of cross contamination 22 . For efficient amplification, smart design and appropriate dosage of primers in a LAMP system can ensure the loop hairpin structure in the initial and the continuous amplifications in the reaction 13,27 . It could also promote the specific amplification that is mediated by the chain replacement enzyme (Bst DNA polymerase) under constant temperature. Additionally, primer quality and inner vs. outer primer ratio play key roles in the LAMP reaction system as these factors have more significant influence on its effectiveness, repeatability and sensitivity 13,28 . As reported, the concentration of inner primers (FIP and BIP) was often higher (4-10 times) than that of outer primers (F3 and B3) 14,29 . In this study, four primers (Pep1F3/Pep1B3 and Pep1FIP/Pep1BIP) corresponding to six independent segments targeting the Pep1 gene in S. scitamineum were designed. An optimal dosage ratio (6:1) of inner vs. outer primer (Figs 2 and 4) distinguished the negative and positive controls.
As Bst DNA polymerase is a Mg 2+ dependent enzyme, the Mg 2+ concentration was assumed to be the most important component in the LAMP assay 30 . Mg 2+ also binds to dNTPs, primers and templates, affecting specificity of the LAMP reaction, amplification of product yield and the formation of primer dimers 14  Although the MgCl 2 reagent was also used in some investigations 14,29 , the MgSO 4 reagent was recommended for the 10× ThermoPol Reaction Buffer supplied in the Bst DNA polymerase kit mentioned in our experiments. Here, when the Mg 2+ concentration ranged from 5.75 mmol/L to 6.50 mmol/L (Fig. 1), LAMP results showed a distinction between positive and negative controls. As suggested by electrophoresis, a Mg 2+ concentration of 6.25 mmol/L was optimal (Fig. 1B). This was consistent with the results reported by Nie and Liu et al., which suggested the existence of an optimum concentration of Mg 2+ during the LAMP reaction 14,29 . Fluoroscopic and electrophoretic signals were extremely weak at lower dosages of the Bst DNA polymerase (2.0 U) (Fig. 3). When the concentration of the Bst DNA polymerase was 8.0 U, negative and positive controls could clearly be distinguished. This was similar to the Nie's findings that showed that the product concentration was reduced due to a higher consumption rate of the Bst DNA polymerase 14 .
Normally, a single factor experiment requires firstly to define initial reaction conditions according to experiences, then one factor is altered to screen approximately optimal concentration ranges under the condition that all the other factors remain unchanged. This method requires more lab work and higher costs. Meanwhile, it is difficult to truly achieve the optimal experimental results using this method due to possible interactions between the factors not considered in the reaction system 12 . Orthogonal experimental design considers the interaction effects between the reaction factors, with the advantages of scattered balance and comprehensive comparisons 31 . Therefore, the optimal combination can be rapidly established. However, in order to know the appropriate dosage ranges of impact factors before employing orthogonal experimental design, researchers usually adjust concentration levels of factors to obtain the optimal reaction system 32 . In the present study, a double optimized LAMP reaction system was rapidly developed that was suitable to detect S. scitamineum by employing a single factor experiment followed by an orthogonal experimental design.
Several studies found that the specificity and sensitivity of the LAMP technology were higher than those of conventional PCR and RT-PCR 21,33,34 . Guo et al. found that the sensitivity of LAMP was 10 times higher than that of the conventional PCR for detecting Mycoplasma hyopneumoinae 33 . Similarly, Wang et al. indicated that the LAMP method was 10 times more sensitive than that of conventional PCR for the detection of the cry1Ac gene in Bt transgenic Oryza sativa 34 . Chandra et al. demonstrated that the sensitivity of the LAMP assay was 5 times greater than that of conventional PCR for detection of gDNA from C. falcatum-infected sugarcane tissues 21 . In this study, except for the positive control, LAMP had no cross reaction with other pathogens, which suggests its high specificity. Sensitivity analysis showed that the detection limit of the LAMP assay was 100 times higher than that of conventional PCR targeting the bE gene for S. scitamineum. In addition, the LAMP reaction was positive for all tested cane buds that were artificially inoculated with S. scitamineum.
In conclusion, the present work established a LAMP method with the specific primers of Pep1F3/Pep1B3 and Pep1FIP/Pep1BIP for S. scitamineum in sugarcane. Double optimal methods, the single factor experiment as well as L 16 (4 5 ) orthogonal experimental designs, were employed to optimize the LAMP reaction system. This novel LAMP assay was sensitive and rapid in the detection of S. scitamineum. It might be also suitable to detect imported or exported sugarcane seeds or seed stems, and provide technical support for realizing smut-free sugarcane supervision and administration.

Methods
Plant materials and treatments. Both the S. scitamineum spores and sugarcane genotype ROC22 were from the Key Laboratory of Sugarcane Biology and Genetic Breeding, Ministry of Agriculture (Fuzhou, China). Smut whips in diseased plants of ROC22 collected in the fields were sealed and stored in 4 °C refrigerator after dried under sunshine or in an incubator at 35 °C. Spore suspension was prepared by diluting smut spores with ddH 2 O. Then, 100 μL of diluent was applied on a potato dextrose agar (PDA) plate containing 75 μg/mL of streptomycin and cultured at 28 °C in darkness. After 4 d, the single spore was re-cultured on a new PDA plate at 28 °C in darkness for 5 d. Finally, it was used to extract hypha DNA by the SDS (Sodium Dodecyl Sulfonate) method 35 . Hypha DNA was treated by 100 μg/mL RNase A (Promega company, USA) and its purity was determined by electrophoresis and spectrophotometer, and then stored at −20 °C.
Healthy and mature sugarcane stalks of ROC22 with uniform growth rates were selected for the test. They were cut into single-bud segments and soaked in flowing water for 24 h, then placed in incubators at 32 °C with 65% humidity and 24 h lighting to accelerate germination. When the cane buds were 2 cm long, a 0.5 μL microliter syringe was used to conduct inoculation on cane buds with a 5 × 10 6 /mL of S. scitamineum spore suspension. Key compositions of the LAMP system were optimized using a single factor experiment and an orthogonal experiment to obtain an optimal LAMP system which is stable and sensitive enough for the detection of S. scitamineum.
Single factor experiment. During the first optimization test, a single factor experiment with three factors and four levels was conducted to screen the dosage of the major factors (Mg 2+ concentration, inner vs. outer primer ratio and Bst DNA polymerase content) influencing the LAMP reaction. Mg 2+ concentrations were set to 5.75 mmol/L, 6.00 mmol/L, 6.25 mmol/L and 6.50 mmol/L, respectively. For the fixed outer primer concentration of 0.2 μmol/L, the inner vs. outer primer ratios were set to 2:1, 4:1, 6:1 and 8:1, respectively. Four different Bst DNA polymerase concentrations, 2.0 U, 4.0 U, 6.0 U and 8.0 U, were tested. The amplification procedure followed 65 °C incubation for 1 h, and a 80 °C hot bath for 5 min to terminate the reaction. The products were detected by color transition and 2.0% agarose gel electrophoresis. 1.5 μL 1 000× SYBR Green I fluorescent dyes (Biotek Co., Ltd., Beijing, China) were added and dropped inside the tube cap before reaction, which allowed the positive product to be indicated by color transition from orange to flavogreen after incubation; those showing intense ladder-like electrophoretic bands under UV light were defined as positive. All tests were repeated twice.
Orthogonal experiment. Due to interactions between factors of LAMP reaction, L 16 (4 5 ) orthogonal experimental design was selected for this study 31 . It was carried out on the basis of the single factor experiment to conduct orthogonal screening with three factors and four levels ( Table 1). Each combination was repeated twice.

Specificity of LAMP.
Specificity of primers was verified by the optimized LAMP system using 100 ng of the following common sugarcane pathogenic DNA as template, including F. moniliforme, P. ginkgo, H. sacchari, F. oxysporum and S. scitamineum genome DNA. In addition, cane exudates from smut resistant and susceptible sugarcane genotypes (Yacheng05-179 and ROC22) were cultivated at PDA liquid medium (containing 75 μg/mL streptomycin) at 28 °C for 1 d to procure the endophytes. Genome DNA of these pathogens and endophytes was extracted following SDS protocol established by Que et al. 35 . The amplification products were detected by color transition and 2.0% agarose gel electrophoresis. The experiment was repeated twice. Analytical sensitivity of LAMP. According to the formulas of MW = genome length × 660 dalton/bp and copies/mL = 6.02 × 10 23 × (concentration g/mL)/(MW g/mol) 36 , the copy number of S. scitamineum (19.8 Mb) 26 was calculated. A ten-fold gradient dilution was conducted on S. scitamineum genome DNA (217 ng/μL, amount to 10 7 copies/μL) to obtain a total of seven samples which ranged from 10 6 copies/μL-1 copies/μL. The optimized LAMP amplification was conducted and repeated twice. Additionally, conventional PCR amplification was conducted using S. scitamineum testing primers of bE4 (5′ -CGCTCTGGTTCATCAACG-3′ ) and bE8 (5′ -TGCTGTCGATGGAAGGTGT-3′ ) for the bE gene 11 , as well as another pair of primers Pep1QF (5′ -TAGCCTTCGTGCTCACCTTTACC-3′ ) and Pep1QR (5′ -TGCTTTCCGCCATCCTCCT-3′ ) which were designed by ourselves based on the Pep1 gene sequences. The conventional PCR reaction system containing 2.0 μL dNTP (2.5 mmol/L), 2.5 μL 10× Ex Taq Buffer (Mg 2+ Plus), 0.125 μL Ex Taq enzyme (5 U/μL), 1.0 μL forward primer (10 μmol/L) and 1.0 μL reward primer (10 μmol/L), 1.0 μL template DNA, ddH 2 O was complemented to a volume of 25 μL. The PCR amplification was performed under the following conditions: 94 °C initial denaturation for 4 min; 94 °C denaturation for 30 s, 55 °C (bE gene) or 60 °C (Pep1 gene) annealed for 30 s, 72 °C extension for 30 s, cycled for 35 times; 72 °C re-extension for 10 min 11 . The 2.0% agarose gel electrophoresis was used to test the PCR amplified product.
Evaluation of LAMP using S. scitamineum inoculated samples. Two-bud sets of ROC22 were inoculated with 0.5 μL of the S. scitamineum suspension containing 5 × 10 6 spores/mL in 0.01% (v/v) Tween-20. All the treated samples were planted in sterile soil and cultivated at 28 °C in 16 h of light and 8 h of darkness. At each time point of 0 d, 1 d, 2 d, 5 d, 7 d, 10 d and 14 d, five inoculated buds were collected for DNA extraction using the CTAB method. These DNA samples (1 μg) were subjected to the optimized LAMP reaction and conventional PCR (bE gene and Pep1 gene detection) reaction systems. The 2.0% agarose gel electrophoresis was used to detect the amplified product of the conventional PCR.