Simple and rapid detection of Tilletia horrida causing rice kernel smut in rice seeds

A simple and rapid method for the detection of Tilletia horrida, the causal agent of rice kernel smut, in rice seeds is developed based on specific polymerase chain reaction (PCR). To design the specific primers for the detection of T. horrida, partial sequences of internal transcribed spacer (ITS) DNA region of T. horrida, T. controversa, T. walkeri, T. ehrhartae, T. indica and T. caries were analyzed and compared. A 503-bp fragment was amplified with the designed primers from the T. horrida genomic DNA. However, no PCR product was obtained from the DNA of other five Tilletia species and 22 fungal plant pathogens tested in the present work indicating the specificity of the primers for the detection of T. horrida. The PCR was performed by directly using the spores, isolated from the 21 different rice seed samples, as template DNA. The T. horrida was detected in 6 of the samples, indicating that 28.6% of the rice samples were contaminated with the kernel smut pathogen. This simple PCR based diagnostic assay can be applied for the direct and rapid detection and identification of T. horrida to screen large numbers of rice seed samples.

The current diagnostic protocol for RKS of rice involves the morphological identification of spores, followed by germination to confirm the identity of the pathogen. But these protocols are very time consuming, labor intensive and require highly skilled personnel to distinguish the morphological similar species 14 . Therefore, there is a dire need to develop a simple and rapid method for the detection and identification of T. horrida in rice seeds from the other common contaminant fungal species. Traditional PCR based detection methods are simple, rapid, highly specific and sensitive for the target species and can be used to discriminate allelic homologues genomic fragments with minor nucleotide differences. Some of such differences exist in the internal transcribed spacer (ITS) sequences of eukaryotes. ITS region between the nuclear small and large subunit ribosomal DNA has proved useful to distinguish the closely related species of fungi 14 .
This work describes the rapid and accurate PCR based method for the quarantine detection and identification of T. horrida in the rice seed samples. The partial ITS region sequences were targeted in the present study to design species specific primers. According to our knowledge this is the first report on the use of conventional PCR based identification and detection of T. horrida by using it partial ITS sequences.

Results
Primer design and PCR. To design the primer and develop the PCR assay for the detection of T. horrida, ITS rDNA sequence of T. horrida CN1 (Accession No. DQ827699.1) was used. The sequence was analyzed for homology with the target Tilletia species including T. controversa, T. walker, T. ehrhartae, T. indica and T. caries (Fig. 1). The forward and reverse primers were designed for the amplification of the target region only from T. horrida. Specificity of the designed primers. The specificity of the T. horrida primers was first evaluated using a collection of genomic DNA samples from T. horrida, T. controversa, T. walkeri, T. ehrhartae, T. indica and T. caries as template for PCR assay. The analyzing of amplicons revealed that the designed primers only amplified a 503-bp fragment from the DNA of T. horrida but not from the other five tested Tilletia species (Fig. 2). Furthermore, when the PCR was performed using genomic DNA of the 22 other fungi from the genera other than Tilletia (Table 1), a fragment of the same size could not be amplified (Fig. 3), suggesting that there was no cross species specificity for the PCR assay developed for the detection of T. horrida.
Detection of T. horrida in rice seeds. The optimized PCR assay was performed to evaluate RKS in 21 rice seed samples, out of all these samples, 6 (28.6%) were found to be PCR positive, indicating these were contaminated with the pathogen T. horrida, while remaining 15 seed samples were free of the causative pathogen (Fig. 4). These results were also confirmed by the morphological studies of the spores isolated from all the samples under investigation. The specificity of this test was further confirmed by performing PCR using genomic DNA from 11 different strains of T. horrida (Table 1) and spores isolated from healthy and infected rice seeds. The diseased status of the seeds was confirmed prior to the experiment from AEEIQB, China. All the 11 strains of T. horrida, and infected seeds showed the positive results after agarose gel electrophoresis (Fig. 5). Whereas, no PCR product was obtained in the case of rice seeds with confirmed healthy status. Sensitivity of the test. Firstly, the sensitivity of the present method was tested by varying the number of spores isolated from the T. horrida contaminated seeds in the PCR reaction mixture. The results depicted that the lower limit for the detection of target pathogen was 30 spores per 25 μ l of PCR reaction mixture (Fig. 6). No amplified product was observed in UV illuminator when less than 30 spores were used in the reaction mixture.
Secondly, in order to know the sensitivity of this method for the detection of T. horrida in rice seeds, artificially 10, 15, 20, 25, 30, 50 and 100 spores were added into 100 g healthy rice seeds per sample. The results indicated that the lower limit for this method was 25 spores/100 g rice seeds and no PCR product was obtained when less than 20 spores were added in the rice seeds (Fig. 7).
Thirdly, to test the sensitivity of this method at the DNA level, the purified DNA (100 ng, 10 ng, 1 ng, 100 pg, 10 pg, 1 pg and 100 fg, respectively) was used as a template for the PCR and the results indicated that this method could detect ≥ 100 pg genomic DNA of T. horrida (Fig. 8).

Discussion
The emerging epidemic trends throughout the world has placed a greater stress to manage rice kernel smut in the future 9 . Thus, rapid detection and identification of T. horrida in rice is crucial for the implementation of surveillance and quarantine regulation in the international rice trade. Detection assays using classical PCR techniques have been developed for numerous plant pathogens, including bacteria, viruses, and fungi 15 . These tests are attractive for several reasons. First, the assays are extremely sensitive and highly specific for the pathogen in question. Second, PCR tests require minimal amounts of sample material, and commercial kits are available for extracting high quality genomic DNA from a wide variety of organisms. Finally, PCR reactions are relatively simple to set up and perform, and results can be obtained quickly, usually within a few hours. In this study, a simple method for the rapid detection and identification of T. horrida in rice seeds was developed by the analysis of the ITS region of rDNA from the six Tilletia species. The ITS region is long tandem DNA repeat array that is located between ITS1 and ITS2 rRNA genes in the (rDNA) unit in eukaryotes 16 . This region has a high degree of variation even between the closely related species and is widely used for the molecular phylogeny and taxonomic studies. According to a previous study, the ITS region has the highest probability of successful identification for the broadest range of fungi, with the most clearly defined barcode gap between inter-and intraspecific variation among the regions of the ribosomal cistron. ITS will be formally proposed for adoption as the primary fungal barcode marker to the Consortium for the Barcode of Life, with the possibility that supplementary barcodes may be developed for particular narrowly circumscribed taxonomic groups 17 . Other than belonging to the same genus, T. horrida is not closely related to the other species included in the study, they possess considerable genomic differences to differentiate them from each other (Fig. 1). Therefore, to distinguish T. horrida from other five Tilletia species, PCR method was developed by designing a specific pair of primers to amplify a 503 bp fragment of T. horrida ITS1 + 5.8 S + ITS2 rDNA region. The designed primers did not amplify any product from the genomic DNA of any of the tested fungi, except T. horrida implicating the specificity of the developed PCR protocol. Many other workers have also reported the use of ITS regions to resolve Tilletia species comprising closely related T. horrida, T. walker, and T. indica, T. contraversa, T. laevis, T. caries, T. bromi and T. fusca 14 .
In the last two decades, isozyme analysis was used to distinguish the isolates of T. indica from T. horrida by using proteins extracted from germinated teliospores. However, considerable experience with the interpretation of complex isozyme polymorphisms associated with Tilletia species is required, and this is not considered as a practical approach for the routine identification [18][19][20] . Several studies have been reported for the identification and differentiation of T. indica (a quarantine pathogen causing Karnal bunt of wheat) from T. walkeri and T. horrida [20][21][22] by using conventional PCR with. In 2006, a molecular protocol, using quenched fluorescence resonance energy transfer (FRET) probes, was developed for the detection and differentiation of T. indica from T. walkeri, T. horrida and T. tritici 23 . A one-tube fluorescent assay for the quarantine detection and identification of T. indica has also been developed 14 . However, all these reports just focused on the detection of T. indica and not to distinguish T. horrida from the other Tilletia species. Moreover, despite of the high sensitivity and accuracy, these methods cannot be commonly used because of the high-priced apparatus and reagent kits. Similarly, other methods such as enzyme-linked immunosorbent assay (ELISA) 24 , sequence-characterized amplified region (SCAR) 25 inter simple sequence repeat (ISSR) marker 26 have also been developed for the diagnosis of T. caries and T. controversa, respectively. However, these methods are either expensive or time-consuming.
Compared to these methods, the method for detection and differentiation of T. horrida from other Tilletia species in our study possesses numerous advantages. It depends on common techniques of molecular biology and needs merely basic PCR reagents, electrophoresis apparatus and thermocycler. This equipment is relatively cheap, inexpensive to maintain, and simple to manipulate. In our method we report the direct use of teliospores for the diagnosis of T. horrida without the culturing and genomic DNA isolation. Thus eliminating the lengthy culturing and difficult morphological identification procedures 27,28 . Furthermore, sensitivity test results showed that the present method can be employed using only ~30 spores in the reaction mixture. The traditional morphological identification methods usually require considerable expertise and a significant of ~50 spores for statistical determination 14,29 . The potential benefits of this technology can be especially recognized in the regulatory field, where both the timeliness and accuracy of identifications are crucial 20,30 .
In our study, the time required for this detection is less than three hours, therefore, this developed technique can be used to screen out a large number of samples in a short time. Hence, it will be extremely useful to resolve the disputes regarding contamination of rice with smut teliospores. In this study, we have also confirmed the identity of the pathogen of the kernel smut of rice using the specifically designed primers. This information would be useful in establishing quarantine areas and preventing the contamination of clean rice shipments with teliospores from infested grains.       (Table 1) used in the present study were provided by the Anhui Entry-Exit Inspection and Quarantine Bureau (AEEIQB), China. The fungal cultures were maintained on potato dextrose agar (PDA) medium and stored at 4 °C.

Isolation of T. horrida spores.
The rice seed samples used for the detection and isolation of T. horrida spores were also supplied by the courtesy of AEEIQB, China that were collected from the different geographic locations ( Table 1). The rice seeds were surface-sterilized in 0.1% sodium hypochlorite for 5 min, rinsed in sterile distilled water for 30 sec twice, striped and the teliospores scraped into a clear tube standby. The isolated spores were identified and confirmed by AEEIQB, China using Chinese National Standard 31 for the identification of T. horrida. The spore counting was performed using hemocytometer.
Extraction of genomic DNA. Genomic DNA of T. horrida was extracted directly from the isolated spores using the Fungal DNA Kit (Omega Bio-Tek) according to the manufacturer's protocol. Whereas, the genomic DNAs of the other tested pathogens (Table 1) were extracted from 50 mg mycelia using the same kit. The spores and mycelia were treated with a minibeadbeater (607EUR, Biospec, USA) before isolation of the DNA. The DNA concentration for each sample was determined by Nanodrop (NanoVue Plus, GE Healthcare Life Sciences).

Primer designing and PCR amplification.
To design the primers, Internal Transcribed Spacer (ITS) sequence of ribosomal DNA of T. horrida, Accession No. DQ827699.1 32 was retrieved from the NCBI database. The sequence was compared and analyzed with the other Tilletia species sequences from the NCBI database using online tools BLASTN 33 and Clustal Omega 34 . The forward primer (HF-F): 5′ -GAGAGTCAACTTATGTTCA-3′ and reverse primer (HF-R): 5′ -GATGAAAGTTACTCTCAT-3′ were designed by using the Bioedit software (v7.0.5) to amplify 503 bp fragment of 5.8 S ribosomal DNA (rDNA) of T. horrida. The PCR conditions were optimized using T. horrida CN1 as template DNA. The DNA amplification was carried out in DNA Engine System PT-200 (Bio-Rad) using 25 μ l reaction mixture containing 1 U ExTaq polymerase, 2.5 μ l of 10× ExTaq buffer, 2 μ l Mg 2+ (25 mM), 200 μ m of dNTPs, 1 μ m of each primer and DNA template (50 ng). The thermal cycler was programed for one cycle of initial denaturation of 5 min at 94 °C; followed by 30 cycles of 1 min at 94 °C, 30 sec at 50 °C and 30 sec at 72 °C with a final extension for 5 min at 72 °C; and a holding temperature of 4 °C. A negative control, replacing the DNA template with sterile distilled water and DNA ladder DL2000 were also run in parallel. The PCR products were electrophoresed using 1% agarose gel in 1xTAE buffer for 45 min at 90 V to analyze the results.

Specificity of the test.
To determine the specificity of the PCR and primers to T. horria, the PCR was carried out with different species of Tilletia and other common contaminant fungal strains (Table 1) using the above mentioned optimized PCR conditions. Detection of T. horrida in rice seeds. A total of 21 rice seed samples, provided by AEEIQB, China, were used for the detection of T. horrida. For each sample, 10 g of the seeds were washed with sterile distilled water for 30 sec twice and saturated in sterile distilled water for 5 min. The saturated water was centrifuged at 12000 rpm for 10 min, then discarded the supernatant and re-suspended the pellet in 50 μ l water. 10 μ l of this spore suspension was directly used as template DNA for the PCR.
Sensitivity of the test. Firstly, the sensitivity test for the detection T. horrida contaminated seeds was performed by varying the number of spores per 25 μ l of PCR reaction mixture. PCR was performed using individual reaction mixtures containing 1, 10, 20, 30, 40 & 50 spores of T. horrida, respectively, followed by agarose gel electrophoresis analysis.
Secondly, artificially 10, 15, 20, 25, 30, 50 and 100 spores were added into the individual 100 g healthy rice seeds samples and then, the spores were collected by seed washing and centrifugation. The PCR was performed using the individual spore suspensions.
Thirdly, in order to know the sensitivity of this PCR method for DNA limit, the purified DNA (100 ng, 10 ng, 1 ng, 100 pg, 10 pg, 1 pg and 100 fg, respectively) was used as a template in the indicidual PCR assays followed by the same cycling conditions as described above.