A conventional PCR technique to detect Septoria tritici in wheat seeds

A PCR assay was developed for detection of wheat seed naturally contaminated with Septoria tritici. S. tritici specific primers were derived from strict alignment of ITS and α-tubulin sequences of the pathogen. The specificity of four sets of synthesised oligonucleotide pairs (A, B, C and D) were tested using isolates from S. tritici, other selected fungi and wheat seeds. A single DNA fragment was amplified from S. tritici isolates with all primer pairs, whereas no product was generated from other DNA sources. S. tritici was also detected in wheat seed lots collected from plants with variable pycnidial coverage on the upper two leaves. PCR detection of as little as 0.5 pg of S. tritici genomic DNAwas possible. This is the first report on the detection of S. tritici DNA in naturally infested wheat seeds. This PCR based assay is simple, rapid, specific, sensitive and suitable for routine detection of the wheat pathogen in infested wheat seeds.


Introduction
Mycosphaerella graminicola (anamorph Septoria tritici) is an economically important pathogen causing leaf blotch of bread and durum wheat.The pathogen is distributed worldwide and causes significant losses in many wheat-growing areas (Van Ginkel and Rajaram 1993).The disease occurs particularly in countries with a humid, temperate to cool climate (Wiese 1987) and the symptoms are characterised by necrotic blotches that contain varying densities of pycnidia, the asexual fructifications (Kema et al. 1996).
Wheat seedlings may become infested soon after emergence and under favourable weather conditions, e.g.frequent rains together with temperatures of 12À25 C. The disease progresses vertically to the upper plant parts.Severe yield losses occur if the fungus infests the upper two leaves (Ziv and Eyal 1974;Robert et al. 2005) that are the main source of carbohydrates for grain filling (Zilberstein et al. 1985).
Breeding for leaf blotch resistance is the most economical approach to controlling the disease (Eyal et al. 1985).Alternatively, foliar fungicides or protection of seedlings by seed treatments with systemic fungicides (e.g.benzimidazoles) are used as chemical control (Rohel et al. 1998).The no-tillage system harbours more inoculum of S. tritici than the conventional system and fungicidal seed treatment alone may not be effective for disease control (Eyal et al. 1987).
Mycosphaerella graminicola is thought to have originated in the Middle East and been globally dispersed on wheat seed as a result of the expansion of wheat cultivation (McDonald et al. 1999;Banke et al. 2004).It has been proposed that movements of infested wheat seeds are likely to have influenced the population genetics of S. tritici (Banke and McDonald 2005).
Although several foliar wheat diseases (tan spot by Drechslera tritici-repentis, spot blotch by Bipolaris sorokiniana, Septoria glume blotch by Stagonospora nodorum and Alternaria infectoria complex) are associated with seed transmission (Mathur and Cunfer 1993), it is unknown how S. tritici grain infestation occurs.Experiments using an agar plate technique to determine the presence of the pathogen inside the seeds were first reported by Brokenshire (1975).This method is not practical for routine seed testing because it is time consuming and S. tritici is easily overgrown by fast-growing saprophytes of wheat.
Advances in molecular biology such as the polymerase chain reaction (PCR) have provided exciting opportunities for development of specific and sensitive assays that have been extensively used for identification and detection of several fungal plant pathogens (Vandemark et al. 2000;Konstantinova et al. 2002;Lee et al. 2002;Kulik et al. 2004;Fountaine et al. 2007;Gayoso et al. 2007).In this sense, molecular approaches have gained special attention as they are useful for detecting seed infestations that are of low infestation levels or asymptomatic (Lee et al. 2001;Tegli et al. 2002;Chadha and Gopalakrishna 2006).
Several PCR based assays for S. tritici detection have been reported so far.Beck and Ligon (1995) reported specific primers to distinguish M. graminicola from S. nodorum and Fraaije et al. (1999Fraaije et al. ( , 2001) ) developed two specific primer pairs for quantitative detection of the pathogen.However, Guo et al. (2006) found that the above-mentioned primers were not specific, yielding multiple bands from some other wheat pathogens.So they developed four new specific primer pairs and developed a sensitive real-time PCR assay for monitoring M. graminicola development in inoculated and naturally infested leaves of wheat plants.Abramova et al. (2008) described a PCR system using the fluorescent amplification-based specific hybridisation (FLASH) for detection of S. tritici and S. nodorum in infested soil, leaf, and seed samples.However, both techniques have a high cost and are not available in many laboratories.
We hypothesised that, as is common in other foliar wheat pathogens, the transport of S. tritici into the wheat seed could be one of the mechanisms to complete the disease cycle, and therefore, there are two possibilities for S. tritici transmission through the seed: (a) the pathogen is transported as vegetative mycelium inside internal tissues of the seed, e.g. S. nodorum, D. tritici repentis, Alternaria triticina and B. sorokiniana, or (b) the pathogen is latent in the embryo, e.g.Ustilago spp.(Mathur and Cunfer 1993).For this reason, it is desirable to develop a reliable and sensitive method for detecting seedborne infestation by S. tritici.
Ribosomal genes and the spacers between them provide attractive targets for detection and phylogenetic studies of microorganisms.They possess conserved and variable sequences that allow their easy amplification and sequencing.Genes for 5.8 S, 18 S and 28 S fungal nuclear rRNA genes are usually organised in identical conserved repeats with 60-200 copies per haploid genome (Bruns et al. 1991;Stackebrandt et al. 1992).Considerably greater sequence variation is found in the internal transcribed spacer (ITS) region between the rRNA genes within a rRNA repeat unit, and this region has been widely used for developing species-specific primers and probes for detection of fungal plant pathogens (Johanson and Jeger 1993;Beck and Ligon 1995;Goodwin et al. 1995;Lee et al. 2001;Zur et al. 2002;Balmas et al. 2005).
Other specifically conserved sequences among fungi are ideal candidates as targets for detection assays.Microtubules in eukaryotes contain three related tubulin proteins (a, b and g) that are conserved, although they have shown small differences.Tubulins from a wide variety of fungi have been characterised.In particular, a-tubulin sequence analysis was used to group S. tritici isolates (Rohel et al. 1998).
The availability of improved diagnostic techniques for the rapid and accurate identification of S. tritici would be useful to growers and field pathologists.We report here a conventional PCR-based assay approach using ITS and a-tubulin-derived primers that can be used for specific detection of S. tritici in naturally infested wheat seeds.

Fungal isolates and wheat seed lots
The S. tritici isolates FALP 92067 and FALP  Wheat seeds from plants with variable pycnidial coverage in the upper two leaves were collected from five commercial varieties and three experimental lines (Table 1).They were collected from naturally infested fields from an endemic region of the Great Wheat Argentine Zone.

DNA extraction
Plugs of all fungal isolates grown in solid culture medium were transferred to liquid minimal medium (sucrose 10 g, yeast extract 10 g, distilled water 1 L) and grown for 1 week at room temperature with shaking on a rotary shaker, collected by filtration, lyophilised and ground into a powder with a mortar and pestle.DNA was extracted using a modified version of the CTAB (cetyltrimethylammonium bromide) method as described below (Murray and Thompson 1980).Seven hundred and fifty microlitres of extraction buffer (100 mM Tris-HCl pH 8, After centrifugation the supernatant was transferred to a new tube, purified with 500 mL of phenol : chloroform : isoamyl alcohol (25 : 24 : 1) and chloroform : isoamyl alcohol (24 : 1) until the interphase became transparent.Nucleic acids were precipitated by adding 750 mL of cold isopropanol followed by incubation at 4 C for 30 min.After centrifugation, the pellet was rinsed twice with 500 mL of 70% ethanol, air-dried and dissolved in 100 mL of Tris-EDTA buffer containing 10 mM Tris-HCl, (pH 8) and 1 mM EDTA.DNA from infested wheat seeds was isolated as follows: a minimum of 100 seeds of each wheat cultivar described in Table 1 and healthy plants of the Pincén cultivar were surface sterilised by rinsing once with HgCl 2 0.01% in 1% ethanol for 2 min and twice with sterile deionised water.To favour S. tritici mycelium proliferation within the seeds, they were incubated in a moist chamber for 48 h then immediately frozen at À20 C, lyophilised and ground into a powder with a pestle.For each extraction of total genomic DNA, 0.1 g (20 seeds) of lyophilised seeds was used.DNA was extracted using the DNeasy PLANT miniKit (Qiagen, Hilden, Germany) according to the manufacturer's instructions.DNA quality and concentration were determined with a spectrophotometer and electrophoresis on 1% agarose gels.

Primer design
Forward and reverse S. tritici specific primers were designed by aligning and comparing the nucleotide sequence of S. tritici a-tubulin (EMBL accession No Y14509) and ITS sequences of S. tritici (EMBL accession No AY152603) with those of other fungi (Alternaria triticina, A. alternata, A. nidulans, B. sorokiniana, F. graminearum, Mycosphaerella fijiensis, M. musicola, N. crassa, Penicillium sp., R. secalis, S. cerevisiae, S. pombe, Septoria glycines, S. nodorum and U. maydis) commonly associated with wheat and wheat relatives, and with those of the plant species Triticum aestivum, Avena sativa and Hordeum vulgare.The CLUSTAL X method of the MegAlign program of the DNAStar software package (Jeanmougin et al. 1998) was used for the alignments.Four oligonucleotides within the a-tubulin region and two oligonucleotides within the ITS region were designed based on sequence differences between S. tritici and the other fungi and plants, with an average G + C content ranging from 45 to 72% (Table 2).The OligoTech program version 1.0 was used to check the properties of the designed oligonucleotides.
Primer selection and specificity to S. tritici All designed primers were tested for the ability to amplify fragments of the expected size (Table 2) and for specificity to S. tritici at different annealing temperatures using purified DNA preparations from the pathogen and other fungi.PCR amplifications were performed in a total volume of 20 mL containing 5 ng of purified fungal DNA, 1.5 to 2.0 mM MgCl 2, each deoxynucleoside triphosphate at a concentration of 0.25 mM, 0.5 mM of each primer and 1 U of Go-Taq DNA polymerase (Promega Corporation, Madison, USA) in the corresponding reaction buffer.Amplifications were performed with a thermocycler Mastercycler gradient (Eppendorf AG, Hamburg, Germany).The number of cycles and annealing temperature for every PCR reaction was adjusted for each primer combination.PCR products were analysed by electrophoresis in 1% (w/v) agarose gels in 1Â TAE buffer containing 40 mM Tris/NaAcO (pH 8) and 1 mM EDTA, at 100 V for 40 min, stained with ethidium bromide and photographed with a FOTO/ Analyst Express system (Fotodyne Inc., Hartland, WI, USA).

Detection of S. tritici on wheat seeds
Primers were also tested for the detection of S. tritici from wheat seeds.Twenty seeds per cultivar were ground to a fine powder and DNA was extracted as described above.Amplifications were performed using each primer set listed in Table 2 and PCR conditions were adjusted for each primer pair.There were five replicates per experiment.

Cloning and sequencing of PCR products
PCR amplified products were ligated to the cloning vector pGEM-T Easy (Promega, Madison, Wisconsin), which was used for transforming E. coli DH5a by standard protocols

Sensitivity and specificity assay
To determine the sensivity of the a-tubulin and ITS primers, a serial dilution of S. tritici DNA was prepared in Tris-EDTA buffer to concentrations ranging from 5 ng/mL to 1 fg mL.The serial dilution was repeated twice.

Results
Evaluation and specificity of a-tubulin and ITS primers Amplification of S. tritici DNA with ITS and a-tubulin primer sets using the indicated annealing temperatures resulted in fragments of expected sizes (Table 2).The identity of the amplified DNA fragments was confirmed after cloning and sequencing.

Optimisation of conditions and evaluation of plant infestation level to detect S. tritici on seeds
In the initial PCR assays with the four primer sets using DNA from S. tritici-contaminated wheat seeds, the target DNA sequence was not amplified.However, the amplification was successful when we used DNA from disinfested seeds incubated in humid chamber for 48 h.DNA fragments of expected sizes were obtained with the four primer pairs (Fig. 1, lanes 12-16, and not shown).
We also investigated the relationship between the plant leaf infestation level and the transmission of the pathogen to the seeds.S. tritici target DNA sequence was detected in seeds from plants with a minimum of 40% of pycnidial coverage in the upper two leaves.

Limit of detection (sensitivity) of the S. tritici assay
In sensitivity tests, the amplification reactions were performed with decreasing amounts of S. tritici genomic DNA to determine the minimum amount of input DNA required to yield a detectable product with the four primer sets.The threshold of S. tritici detection by PCR differed depending on the primer pair used.Set A successfully amplified the target DNA sequence from as little as 0.5 pg.Sets B, C, and D amplified the target DNA from 50, 50 and 25 pg, respectively (Fig. 2, and not shown).

Discussion
In our preliminary studies conducted on several lots of S. tritici infested wheat seeds, fungal mycelium was not detected by applying seed washing technique, plating seeds on PDA medium or microscope examination of the intact or cut infested seeds, as was demonstrated by Brokenshire (1975).These results prompted us to investigate molecular-based detection techniques.
PCR-based assays are very important tools that have facilitated the identification of many plant pathogens (Ward et al. 2004).In this study, a conventional PCR based test was developed to detect S. tritici in wheat seed using pathogenspecific primers derived from the ITS region of rDNA and a-tubulin gene of the fungus.This approach is much more rapid than conventional methods for S. tritici detection such as the cultural method, which takes ~10 days for incubation

Detection of Septoria tritici in wheat seeds
Australasian Plant Pathology (Brokenshire 1975).An additional advantage of the PCR-based assay is that it does not rely on visual identification of S. tritici, which is complicated by the frequent occurrence of other fastgrowing fungi.Nuclear rDNA sequences have been an alternative source for designing species-specific primers, and particularly for detecting S. tritici from DNA extracted from infested leaves.A set of specific S. tritici ITS primers was designed by Beck and Ligon (1995).However, Fraaije et al. (1999) demonstrated that when using those primers at an annealing temperature of 60 C, two amplification products were obtained: a fragment specific for S. tritici and a second product which was also amplified from healthy wheat leaf samples.Guo et al. (2006) designed four specific primers for S. tritici detection based on rRNA and actin gene sequences of the pathogen, or random amplified polymorphic DNA (RAPD) for developing a real-time assay for monitoring fungal development on wheat plants.We have used a different strategy for primer design to specifically detect S. tritici on naturally infested wheat seeds, being mindful that other genetically similar fungi are also present in the wheat environment and on taxonomically-related wheat species.Our results show that the four set of primers designed based on ITS rDNA or a-tubulin sequences are specific for S. tritici detection.Since no other fragments were detected from healthy plants or other microorganisms living in the wheat seed or soil environment, our assay could be used for S. tritici tests on field samples.
As reported, the detection of phytopathogenic fungi in plant material and seeds can be difficult, especially at low infestation levels (Konstantinova et al. 2002).Particularly, DNA extracts prepared from seeds usually contain low levels of the target DNA sequence.This can be easily avoided by including an enrichment step to increase the amount of target fungus in the seed sample.This approach has been adopted in other PCR-based seed assays such as the detection of Leptosphaeria spp.from rapeseed (Taylor 1993) and Alternaria radicina from carrot seed (Pryor and Gilbertson 2001).To overcome the low pathogen levels in our work, a pre-assay incubation step was used to increase the fungal biomass of S. tritici in wheat seeds.This approach was not applied in experiments carried out for S. tritici detection on wheat seeds by Abramova et al. (2008), which could not detect S. tritici using a fluorescent method.
Results presented here demonstrate that the PCR-based method developed in this study provides an efficient strategy for the detection of S. tritici in wheat seeds, even when they look healthy.This type of test has the potential to be used for early identification and routine testing of the leaf blotch pathogen on wheat seeds and could contribute to more effective disease management practices.In practical terms, this research forms the basis of the development of an ideal quantitative methodology using a real-time PCR procedure.Future research will be directed towards understanding the infestation pathway and localisation of S. tritici in the seed.

A
Primer set derived from the internal ITS region of S. tritici.B Primer sets derived from the a-tubulin gene of S. tritici.d, direct primer; r, reverse primer.C Position starting from the first base of the ITS sequence (NO AY152603) and from the start codon sequence of a-tubulin gene (No Y14509) of S. tritici.

Table 1 . Wheat seeds lots tested in this study for the presence of S. tritici
, 250 mM NaCl, 2% CTAB) and 15 mL of 2-mercaptoethanol were added to the ground tissue and incubated at 65 C for 30 min.Cellular proteins were precipitated with 300 mL of potassium acetate 300 mM, pH 4.8.

Table 2 . Description of primer sets and PCR conditions used for amplification of S. tritici DNA
(Sambrook and Russell 2001).DNA sequencing was carried out by MACROGEN services (Korea).BLASTX of the National Center for Biotechnology Information (http:// www.ncbi.nlm.nih.gov,version 2.2.6) was used for DNA sequence analysis.