Rapid Identification of Black Grain Eumycetoma Causative Agents Using Rolling Circle Amplification

Accurate identification of mycetoma causative agent is a priority for treatment. However, current identification tools are far from being satisfactory for both reliable diagnosis and epidemiological investigations. A rapid, simple, and highly efficient molecular based method for identification of agents of black grain eumycetoma is introduced, aiming to improve diagnostic in endemic areas. Rolling Circle Amplification (RCA) uses species-specific padlock probes and isothermal DNA amplification. The tests were based on ITS sequences and developed for Falciformispora senegalensis, F. tompkinsii, Madurella fahalii, M. mycetomatis, M. pseudomycetomatis, M. tropicana, Medicopsis romeroi, and Trematosphaeria grisea. With the isothermal RCA assay, 62 isolates were successfully identified with 100% specificity and no cross reactivity or false results. The main advantage of this technique is the low-cost, high specificity, and simplicity. In addition, it is highly reproducible and can be performed within a single day.


Introduction
Black grain eumycetoma represents the most common fungal mycetoma worldwide. This chronic, erosive infection of subcutaneous tissues particularly affects the lower extremities and leads to severe disability [1]. The disease is considered a major health problem in tropical areas and is prevalent among people of low socio-economic status [2].
Mycetoma presents as a subcutaneous mass with multiple sinuses that discharge pus, serous fluid and grains, i.e. the characteristic compact grains of the causative agent formed inside the lesion [3].
A wide range of microorganisms has been reported to cause mycetoma. For treatment, not only differentiation between (fungal) eumycetoma and (bacterial) actinomycetoma is important, but also the identity of the causative agent, since species differ in their response to antimicrobial drugs [4]. In endemic countries, clinical diagnosis may be the only diagnostic method. A fully developed mycetoma lesion is easily identified clinically, whereas in early stages with the absence of grains, the infection may be confused with phaeomycosis or soft tissue tumors [1]. In such cases fine needle aspiration cytology or deep surgical biopsy for histological examination are useful [1,5]. Some fungal and bacterial grains have a characteristic histological appearance which helps in provisional identification, but recognition of the causative species remains impossible [6]. Isolation of the pathogen from discharged grains or from biopsies allows identification of agents that sporulate, but most of the species lack phenotypic characteristics [3]. Molecular techniques have been introduced to facilitate the identification of nondescript organisms [7,8,9], but are of high cost and time-consuming. Thus, there is a need for a fast, simple and reliable method for identification.
Rolling circle amplification (RCA) is a powerful diagnostic method based on detection of specific nucleic-acid sequences and enzymatic amplification of circularized oligonucleotide probes under isothermal conditions [10]. The probes are linear oligonucleotides that contain two target-complementary sequences at their ends joined by linkers [11]. The ends of the probe hybridize to the complementary target in juxtaposition and then ligate which allows the circularization of the probe [11]. The circular structured molecule then amplifies with DNA polymerase that has strand displacement and progressive DNA synthesis activity resulting in series of repeats of the original circular template [11,12]. The technique has been proven to be rapid, specific and low-cost for molecular identification of viruses, bacteria, and fungi [13,14,15,16]. It has been applied for identification of a rare black grain mycetoma species Exophiala jeanselmei [17]. In addition, RCA has been used successfully for identification of white grain mycetoma species Scedosporium boydii [18]. The aim of the present study is to develop RCA-based diagnostics for the most common agents of black-grain eumycetoma.

DNA extraction and target amplification
DNA was extracted using cetyltrimethyl ammonium bromide (CTAB) method as described by Möller et al. [21]. Amplification of the ITS region was performed using primers V9G and LS266 [22] in a 25 mL reaction mixture containing: 10 ng of template DNA, 0.1 mM each dNTP, 0.6 U Taq polymerase (GC Biotech, Alphen aan den Rijn, The Netherlands), 1 mL of each primer (10 pmol) and 2.5 mL reaction buffer (0.1 M Tris-HCl, 0.5 M KCl, 25 mM MgCl 2 , 0.1% gelatin, 1% Triton X-100). PCR reactions consisted of a 5 min predenaturation step at 95uC, followed by 30 cycles of 95uC for 30 s, 52uC for 30 s and 72uC for 1 min, with final post elongation step at 72uC for 7 min. PCR products were detected by electrophoresis using 1% agarose gels.

Padlock probe design
Sequences of the ITS region were used to design 8 probes specific for each species used in this study. Two alignments were generated since the analyzed species were known to belong to two different fungal orders [19,20]. ITS derived from Madurella (Sordariales) were aligned with 200 isolates of Chaetomiaceae including Chaetomium, Thielavia, and Achaetomium. For the remaining species (Pleosporales) an alignment was constructed to include representative isolates of the family Trematosphaeriaceae and of coelomycetes in the suborder Pleosporineae. Sequences were aligned using BioNumerics v4.61 (Applied Maths, Sint-Martens-Latem, Belgium). Probes were designed with minimum secondary structure and were checked using PrimerSelect (DNASTAR Lasergene, WI, U.S.A.). To insure specificity of the probes, target-specific sites of each padlock probe was submitted to BLAST in NCBI sequence database for homologous sequences. The melting temperature of the 59 end of the probe binding arm was designed to be.63uC while for the 39 end binding arm it was designed to be at least 15uC below the annealing temperature. The probes were phosphorylated at the 59 end and are listed in Table 2. Probe linkers were taken from Zhou et al. [23].

Exonucleolysis
Prior to RCA amplification reaction and in order to reduce the ligation-independent amplification, ligation products were treated by addition of 10 U exonucleases I and 10 U exonucleases III (New England Biolabs, Hitchin, U.K.) with a final volume of 20 ml. The mixture was then incubated for 30 min at 37uC, followed by 3 min at 94uC to deactivate the endonuclease enzymes.

Rolling circle amplification (RCA)
RCA amplification reaction was performed in a 50 ml mixture containing; 2 ml ligation product, 8 U Bst DNA polymerase (New England Biolabs), 10 pmol of each RCA primer (Table 2), and 400 mM dNTP mix. The mixture was incubated at 65uC for 60 min and cooled at 10uC. Electrophoresis on a 1% agarose gel was used to visualize RCA products. A positive reaction is indicated by the presence of ladder-like pattern. The reaction was also visualized by adding 1.0 ml of a 10-fold diluted SYBR Green I (Cambrex BioScience, Workingham, U.K.) to 10 ml of the amplification product. Accumulated double stranded DNA was detected with UV transilluminator (Vilber Lourmat, Marne-la-Vallée, France).

Determination of analytical specificity and sensitivity
The specificity of the 8 RCA probes was tested using strains of black-grain mycetoma causative species listed in table 1. Analytical sensitivity was determined using 10-fold serial dilution of M. mycetomatis (CBS 109801) and M. fahalii (CBS 129176) DNA and the test was performed as mentioned above. In addition, RCA was performed directly using DNA samples without amplification of the target gene. To evaluate the detection limit from direct DNA samples two-fold serial dilutions of target DNA were tested.
The sensitivity of the RCA probes was also determined by 10fold serial dilution of MYC and MFAH probes tested with amplified ITS of M. mycetomatis and M. fahalii respectively.

Results
RCA was used to identify 62 strains belonging to eight species causing human eumycetoma. Since black grain eumycetoma species are known to be phylogenetically distant, it is easy to find unique sites for their identification. The ribosomal ITS region was sufficient for identification of all species and showed no intraspecific variability within a set of 100 M. mycetomatis strains in our collection. For M. mycetomatis, M. tropicana, M. pseudomycetomatis, and F. senegalensis the ITS1 region was

Author Summary
Treatment of eumycetoma largely depends on the causative pathogen. Identification of mycetoma agent with phenotypic features is too limited, and physiological and biochemical techniques are laborious, time-consuming and nonspecific, whereas the currently available molecular methods based on DNA sequencing are specific but extremely expensive. We describe rolling circle amplification method for identification of black grain eumycetoma using species-specific padlock probes. Eight probes were designed and successfully used for species identification and the results were easily visualized in 1% agarose gel. RCA provides a simple, reproducible, and costeffective method for rapid identification of mycetoma agent that can be used in low-resource clinical settings. RCA results for the tested strains were easily visualized in 1% agarose gel. Positive reactions demonstrated ladder like patterns while negative reactions resulted in a clear background (Fig. 1). With SYBR green, positive results showed green fluorescence when exposed to UV light, while negatives did not. When exonucleolysis was performed some inhibition was observed with low RCA positive signals on gel or with fluorescence. Faint nonspecific bands were observed when this step was omitted. RCA reactions were performed successfully without digestion with exonucleases, as the non-specific bands did not interfere with RCA results.
All M. mycetomatis strains were correctly identified with RCA, irrespective of their geographical origin (Sudan, India, Mali) (Fig. 2). For the other agents, each individual species-specific probes yielded positive results with their corresponding species and with 100% agreement with ITS sequencing (Fig. 2, Table 3). No cross reactivity or false positive and negative results were observed. The sensitivity of RCA when using amplified product of the target gene was less than 32610 23 ng of DNA. A higher concentration of 100 ng is needed when the test is carried out directly from the DNA samples without amplification of the ITS. The probes were very sensitive and a concentration of 6.6610 25 ng was successfully ligated and then amplified with RCA.
The turnaround time required for conducting the entire experiment including PCR amplification of target DNA, RCA processing and analysis was found to be 6 hours. DNA sequencing of the ITS region took more than 8 hours to be performed (Fig. 3).

Discussion
Mycetoma is a unique tropical disease, endemic in many tropical and subtropical regions that has been recently added to the WHO list of neglected tropical diseases. [24]. It is mainly prevalent in what is known as ''mycetoma belt'' which includes Mexico, Senegal, Sudan, India and other countries between tropic of cancer [1]. In 2014, a mycetoma consortium of scientists and physicians published research gaps on mycetoma which need to be addressed in the coming years [2]. One of the research priorities identified was the need to develop a reliable and cost-effective method for species identification to improve diagnosis [2].
Mycetoma agents have been extensively studied in recent years [8,9,20]. The large phylogenetic distance between a number of these agents provides the possibility to use a moderately variable marker like rDNA ITS for species identity. Ahmed et al. [25] developed PCR-restriction fragment length polymorphism (RFLP) for identification of M. mycetomatis targeting the ITS region. However, with the description of the molecular siblings M. fahalii, M. pseudomycetomatis, and M. tropicana [26] the method might be insufficiently accurate. Moreover, there is a need for identification these siblings species; Madurella grisea appeared to be distantly related and was re-named as T. grisea [20].
In the present study we developed a simple, fast and highly specific molecular method for the identification of agents of black grain mycetoma. In this method, the ITS region is easily amplified using one set of primers, which simplifies the use. In a second, isothermal amplification reaction padlock probes are used to identify the species by RCA. The only equipment necessary is a thermocycler for the PCR reaction and a water bath or heating block for the RCA reaction. This relative simplicity enhances possible use in routine laboratories in endemic areas. Due to its  robustness, high potential, and reproducibility, RCA is increasingly used as a diagnostic tool in pathogenic fungi, e.g. agents of chromoblastomycosis, dermatophytes, Aspergillus, Candida, and Talaromyces marneffei [16,23,27,28]. The method does not require DNA sequencing and is therefore considered as a rapid and cost-effective. Applications are being expanded to nano-and biotechnology [29].
In the present study eight species-specific probes were designed and used for identification of 62 isolates. For the RCA reaction species probe hybridization to the 39 and 59 ends of target DNA and joining of adjacent ends by DNA ligase when both show perfect complementarity. The ligation appears to be highly specific and thus the method can detect single nucleotide polymorphism [30]. The amplification reaction is driven by an isothermal DNA polymerase to amplify the circularized probes with high efficiency and an estimated capacity to synsthesize more than 70,000 bp per hour [31]. RCA products can be detected with different methods including gel electrophoresis, radiolabeling, UV absorbance, fluorescence, and single molecule detection [32]. It was known that the positive signals can be detected within 15 min after starting the RCA reaction by real time PCR [23]. In the present study the RCA positive signal was easily visualized using both gel electrophoresis and fluorescent dye. The duration of our RCA protocol was 2 h, but additional time is required for DNA extraction and ITS amplification. Compared to the DNA sequencing the turnaround time for RCA is 2 hours less than sequencing and this even more if there is no in-house sequencer available.
Our results with eight padlock probes showed that RCA accurately identified all species with no cross reactivity (Fig. 1). It may be concluded that RCA is extremely useful for specific identification of agents of mycetoma. Performance and rapid turnaround time features make the RCA suitable for quick and reliable diagnosis, which is an enormous improvement compared to the current phenotypic identification of mostly non-sporulating cultures. Future application of RCA could be the detection of agents DNA directly from clinical samples without requirement of culturing.