Specific and sensitive loop-mediated isothermal amplification (LAMP) method for Madurella strains, eumycetoma filamentous fungi causative agent

Background Filamentous fungi of the genus Madurella are the primary causative agents of mycetoma, a disease observed in tropical and subtropical regions. Since early diagnostics based on a morphological approach are difficult and have many shortcomings, a molecular diagnostic method suitable for rural settings is required. In this study, we developed the loop-mediated isothermal amplification (LAMP) method to present a foundational technique of the diagnosis of Madurella spp. (M. mycetomatis, M. pseudomycetomatis, M. tropicana, and M. fahalii), the common causative organisms of eumycetoma. Principal findings We successfully designed a primer pair targeting the rDNAs of three Madurella spp. excluding M. fahalii, and detected up to 100 fg of genomic DNA extracted from isolates of M. mycetomatis and 1 pg of M. pseudomycetomatis and M. tropicana, within one hour. Second, a primer pair specific to M. mycetomatis, the most common causative species, or M. fahalii, a drug-resistant species, was constructed, and the detection limit of both primer pairs was 1 pg. The designed primers accurately distinguished 16 strains of the genus Madurella from various fungal species known to cause mycetomas. Conclusion In summary, we established the first model of a LAMP detection method that rapidly and sensitively detects and identifies Madurella isolates for clinical diagnostics. Moreover, the combined designed primer sets could identify mycetoma-causing strains simultaneously.


Introduction
Mycetoma is a chronic granulomatous inflammatory disease of the subcutaneous tissue classified as bacterial (actinomycetoma) or fungal (eumycetoma) according to the causative agents.It is endemic to tropical and subtropical areas (the mycetoma belt) [1,2].Especially, eumycetoma has been frequently reported in African countries [2,3].Since mycetoma in the late stage leads to massive deformities and disabilities and sometimes becomes fatal, a rapid and proper diagnosis is essential [4,5].Madurella mycetomatis and related species are common causative agents of eumycetoma; thus, they are important targets of diagnosis [2,3,6,7].Morphological identification of Madurella spp. is difficult because culture media lacks sporulation and has a slow growth rate [8].These circumstances have become a severe diagnostic problem [6,9].
Molecular diagnostic methods have been developed for the rapid and accurate detection of Madurella spp. in mycetoma specimens [9].In 1999, Ahmed et al. designed PCR primer pairs that specifically target the rDNA-ITS region of Madurella spp.[10,11].Recently, it was confirmed that not only M. mycetomatis but also other Madurella species, including M. fahalii, an itraconazole-resistant agent [7,12,13], can cause eumycetoma [13,14].Therefore, species-specific detection of causative Madurella spp. is essential.Species-specific PCR methods have improved the detection of M. mycetomatis [11] and M. pseudomycetomatis [15].Furthermore, a real-time PCR method, which could distinguish M. mycetomatis and M. fahalii from other strains, such as M. pseudomycetomatis and M. tropicana, was developed to exclude false-positive amplification based on the melting points of products and Ct values [16].
A community-based diagnostic test is required for the early detection and management of mycetoma in rural settings and its epidemiologic study.[4,5,17].From this viewpoint, the isothermal DNA amplification technique is preferred because it does not require expensive instruments or skilled laboratory personnel compared with the PCR method [18,19].Rolling circle amplification (RCA), recombinase polymerase amplification (RPA), and loop-mediated isothermal amplification (LAMP) have been developed to detect M. mycetomatis [18,20].LAMP is the best choice among the isothermal amplification techniques due to its sensitivity and specificity [21,22].Moreover, this is a cost-effective, well-developed, and versatile detection method [18,19].
In this study, we developed a LAMP method that targets four known Madurella spp.(M.mycetomatis, M. pseudomycetomatis, M. tropicana, and M. fahalii), which are the most frequent causative organisms for rapid, accurate, and versatile diagnostic methods for mycetoma.

Fungal and bacterial strains
A total of 34 fungal-and 5 bacterial strains were used in this study (Table 1).The details of fungal strains are as follows: 16 belong to Madurella, three to other members of Chaetomiaceae, four to Aspergillus, and 11 to other taxa recognized as causative agents of mycetoma [3,11].The species of five bacterial strains used in this study are also known to cause actinomycetoma [2].Briefly, 14 clinical strains were preserved at Medical Mycology Research Center (MMRC; Chiba University, Japan) as IFM collection through the National Bio-Resource Project, Japan, and 13 mycetoma strains preserved at the Mycetoma Research Center (MRC) Khartoum University, Khartoum, Sudan.In addition, 12 strains were obtained from the culture collections of the Westerdijk Fungal Biodiversity Institute (CBS; Netherlands), NITE Biological Resource Center (NBRC; Japan), American Type Culture Collection (ATCC; United States) and Japan Collection of Microorganisms (JCM; Japan).The MRC strains were brought to Chiba University as slant-cultured samples, and their genomic DNAs were extracted after subculturing, as described below.Several DNA samples were purchased from CBS (two samples) and brought from MRC (five samples), and tested in Japan (Table 1).

Preparation of genomic DNA
All strains were grown on a PDA medium at 25˚C for 1-3 weeks.Bacterial colonies or mycelia were picked by inoculating the needle and suspended in 500 μL of 100 mM Tris-HCl buffer (pH 8.5) containing 40 mM EDTA.The bacterial cell or fungus body in the suspension was disrupted using MagNA Lyser and MagNA Lyser Green Beads (Roche Diagnostics, Indianapolis, IN) at 5000 rpm for 20 s.Genomic DNAs were extracted using the benzyl chloride method [23] with some modifications.Finally, the concentration of DNA solution was determined by absorbance at 260 nm using microvolume spectrophotometer DS-11 (Denovix, Wilmington, DE) and the solution was stored at 4˚C in Tris-EDTA buffer.

LAMP amplification of fungal DNA
The LAMP reaction was performed in a 25 μL reaction volume by dissolving the dried LAMP reagent in the primer solution and DNA solution and diluting it with 10 mM Tris-HCl buffer (pH 8.0).An aliquot (15 μL) of the primer solution containing 40 pmol FIP and BIP, 20 pmol

MRC 59v7
The extracted DNA was brought to Japan from Sudan

MRC 68v7
The extracted DNA was brought to Japan from Sudan

MRC 73v7
The extracted DNA was brought to Japan from Sudan

MRC 88v7
The extracted DNA was brought to Japan from Sudan

Design of the primer pairs detecting Madurella spp
We constructed a primer set targeting the four strains of Madurella based on their rDNA sequences.A negative control strain was searched using NCBI BLAST with the rDNA sequence of M. mycetomatis as the query, and Chaetomium rectangulare (accession no.NR_144817) showed the highest similarity among the strains preserved in the MRC.Therefore, C. rectangulare CBS 126778 T was selected as the negative control to validate specificity, though there is no report that it causes mycetoma.Nucleotide alignment of partial rDNA sequences derived from the four Madurella species and C. rectangulare was conducted to design primers, as shown in Fig 1 .However, the target region is conserved among these species  PLOS NEGLECTED TROPICAL DISEASES

Validation of LAMP primers
The optimal reaction temperature for each primer was determined.For PV and PM, 100 pg of genomic DNA derived from M. mycetomatis IFM 46458 was used.In contrast, for PF, 100 pg of genomic DNA of M. fahalii CBS 129176 T was used.As shown in S1A Fig, PV showed the comparable amplification rates at the reaction temperature between 62˚C and 65˚C, with decreased efficiency at 61˚C or 66˚C.Similarly, the amplification rates using PM and PF reached a maximum at 63˚C, as shown in S2 and S3 Figs.Therefore, the optimal reaction temperature for all three primer sets was determined as 63˚C.
To assess the specificity of these primers for Madurella spp., we performed a LAMP reaction at 63˚C using a sufficient amount ( 10

PLOS NEGLECTED TROPICAL DISEASES
shown in Fig 3B and 3C, respectively.On the other hands, we also designed the primer pair (primer set PV2) based on the ITS-2 region, which has potential to amplify all four Madurella spp.due to its highly conserved sequence, as shown in S4 Fig.As expected, primer PV2 could amplify the genomic DNAs of four Madurella spp., but it also showed the reactivity to C. rectangulare, at an optimal reaction temperature (65˚C).Thus, we performed further experiments using the primers PV, PM and PF as basal primer sets.
The sensitivity of the primer sets was determined by amplifying the genome in different amounts, ranging from 10 fg to 10 ng.PV could detect up to 100 fg genomic DNA of M. mycetomatis and M. pseudomycetomatis but failed to do for M. tropicana, as shown in

LAMP amplification using known causative species
To verify the availability of the designed primers, we performed a LAMP reaction using 10 ng DNA from mycetoma clinical isolates and species known as the mycetoma causative agents including bacteria [3,11,16].The PV amplification test detected only ten strains of M. mycetomatis, M. pseudomycetomatis, and M. tropicana (Table 3).Furthermore, PM and PF completely distinguished their targets from those of the other Madurella species.

Discussion
Mycetoma is a severe deep mycosis and is widely recognized as one of the most neglected tropical disease [1].Globally, Madurella spp.are the most common causative agents of eumycetomas.Due to the challenges in isolating and morphologically identifying Madurella in culture media [4,9], many studies have reported using molecular techniques for diagnosing mycetoma [11,16,18,24].Since isothermal DNA amplification techniques are regarded as a promising tool in community-based medical treatment, this study aimed to develop a LAMP method for Madurella spp.. Three primer sets were designed to target three Madurella spp.: one for each  PLOS NEGLECTED TROPICAL DISEASES M. mycetomatis and M. fahalii.Finally, these primers enabled a LAMP method which specifically amplified a small quantity of genomic DNAs derived from Madurella spp.using a constant temperature.
Combining the designed primers simultaneously and specifically identified four common eumycetoma-causing organisms; M. mycetomatis, M. fahalii, M. pseudomycetomatis, and M. tropicana at the same reaction temperature, i.e. in one batch reaction, as shown in Fig 3 and Table 3.Additionally, a double check based on the specific detection of M. mycetomatis is also important for rapid downstream treatment since M. mycetomatis is the primary fungus causing eumycetoma [2,6], and isolates with reduced susceptibility to azoles have emerged [25].Similarly, the double check for M. fahalii will become significant because it is known to exhibit azole-resistance.To overcome this problem, the primer set targeting all Madurella spp. is needed.Unfortunately, such primer sets were not obtained as far as the rDNA ITS regions were used as target, as shown in Figs 4A and S4.On the other hand, M. pseudomycetomatis and M. tropicana have only been tested in single strains each.Thus, a larger number of strains belonging to these species should be tested in the future to verify the specificity of these primers.
This study reported that LAMP system could detect up to 1 pg genomic DNA of Madurella spp.within 60 min, as shown in Fig 4 .Based on the estimated molecular number of genomic DNA of M. mycetomatis, which is calculated to be 26.6 copies/pg, assuming a genome size of 36.7 Mbp [26].The system detection limits are comparable not only to those targeting other fungi, such as Aspergillus [27] and Fusarium [28] but also to those of real-time PCR methods for Madurella (up to 3 pg) [16].Furthermore, the sensitivity of our LAMP system is approximately 200 times higher than that reported previously [18].This may be because the loop primer drastically enhances reaction efficiency [22].We improved the M. mycetomatis-specific detection method in this study and extended the target species described above.
Recently, real-time PCR has been reported to be a reliable diagnostic tool for mycetoma because the products' Ct value (cycle to amplify) and melting point show clear false positives and species identification criteria by monitoring a fluorescence thermal cycler [16].On the other hand, the data obtained from this study suggest that the LAMP method can also identify Madurella spp.using small amount of DNA (>1 pg) at the endpoint in one batch.Thus, the LAMP method might be an alternative to PCR as a diagnostic method for mycetoma in community-based medicine.
It has been demonstrated that the LAMP reaction can be used with the crude fungal DNA extracted from clinical specimens [29][30][31], infected plants, or a few conidia with a simple treatment [28,32] due to its outstanding sensitivity and specificity.In addition, the preparation of crude DNA derived from mycetoma specimens for PCR diagnosis has been reported [13,16].These findings support the idea that our LAMP system will be applicable to clinical samples.However, in this study, specimens of mycetoma were not used and the LAMP reaction was performed only in the laboratory.Thus, the further study in clinical sites is required for practical use of this technique.
In conclusion, the present study showed that the isothermal DNA amplification by LAMP method could identify Maderalla spp. as the major causative agent of eumycetoma worldwide.This method is simple, quick, reliable, affordable, cost-effective, and field-friendly.It can replace PCR-based diagnostic tests in low-resource communities and community-based mycetoma epidemiological surveys.

Fig 2 .
Fig 2. The primers' design specific to M. mycetomatis (primer PM) or M. fahalii (primer PF).The rDNA sequence of M. mycetomatis (A) or M. fahalii (B) was aligned with that of M. pseudomycetomatis (accession no.EU815933.1),and arrows represented the primer positions.The matched nucleotide position was represented using dot signs, while the gap and mismatch positions using corresponding substitution ones.The species name and the number of first nucleotide were represented at each line using the following abbreviations: MM, M. mycetomatis; MP, M. pseudomycetomatis; MF, M. fahalii.The nucleotide alignment and illustration were performed using Genetyx Ver. 14 (Genetyx Co., Tokyo, Japan) with its default parameters.https://doi.org/10.1371/journal.pntd.0011644.g002

Fig 3 .
Fig 3.The specificity of designed LAMP primers, (A) PV, (B) PM and (C) PF. 10 ng of the extracted genomic DNAs derived from Madurella spp.were used as templates.LAMP reaction was performed at 63˚C for 60 min, and the turbidity of the reaction solution was monitored as an indicator of amplification.As for primer PV, the genomic DNA of Chaetomium rectangulare was also used.The sample legends of the plots were represented on the bottom right panel.Abbreviations: MM, M. mycetomatis IFM 46458; MP, M. pseudomycetomatis IFM 46460; MT, M. tropicana CBS201.38 T ; MF, M. fahalii CBS129176 T ; CR, Chaetomium rectangluare CBS 126778 T ; B, Blank sample with a buffer used for DNA dilution.https://doi.org/10.1371/journal.pntd.0011644.g003 Fig 4.However, the 1 pg genome of these three species was successfully amplified, as shown in S4B Fig.In addition, PM and PF recognized up to 1 pg of genomic DNAs derived from M. mycetomatis and M. fahalii, respectively, as shown in Fig 4B and 4C.Finally, the specificity of the LAMP reaction was verified using agarose gel electrophoresis.As shown in S5 Fig, the resultant solution in the positive samples showed the typical ladder bands while none of amplicons were detected in the reaction using non-target sample.These results suggest that the expected amplification occurred.

Fig 4 .
Fig 4. The detection limits of designed LAMP primers, (A) PV, (B) PM and (C) PFThe genomic DNAs of M. mycetomatis IFM 46458 (for PV and PM) and M. fahalii CBS 126778 T (for PF) were serially diluted by 10-fold, and 10 ng-10 fg of DNAs were used as a template.LAMP reaction was performed at 63˚C for 60 min, and the turbidity of the reaction solution was monitored as an indicator of amplification.The sample legends of the plots were represented on the bottom right panel, and a plot labelled with B corresponds to the data of a blank sample with a buffer used for DNA dilution.https://doi.org/10.1371/journal.pntd.0011644.g004 /doi.org/10.1371/journal.pntd.0011644.t003

Table 1 .
Eiken Chemical Co., Ltd.) and mixed with 10 μL template DNA solution using a centrifuge machine equipped with a shaking function (MS-16; Eiken Chemical Co., Ltd.) to dissolve the dried LAMP reagent in the tube.The resultant mixture was immediately transferred to a LAMP instrument (LoopampEXIA (Eiken Chemical)), and its turbidity, reflecting DNA amplification, was measured in real-time for 60 min at a constant temperature.Unless otherwise stated, the reaction was performed at 63˚C.To verify the specificity and versatile reactivity to Madurella strains, 10 μL of 1 ng/μL genomic DNA solution (10 ng of genomic DNA) was used per reaction tube.In contrast, to determine the detection limit, 10 μL of 1 ng/L-1 fg/L (10 ng-10 fg DNA) solution was tested.(Continued)