Evaluation of a Multilocus Indel DNA Region for the Detection of the Wheat Tan Spot Pathogen Pyrenophora tritici-repentis

One of the most devastating and widespread diseases affecting the production of wheat worldwide is tan spot or yellow (leaf) spot, caused by the plant-pathogenic fungus Pyrenophora tritici-repentis. P. tritici-repentis is a necrotroph that produces multiple host-specific effectors which cause distinct necrosis and chlorosis on host plants (Lamari et al. 2003). In addition to causing foliar damage on wheat, P. tritici-repentis also infects the heads, causing red smudge and affecting grain quality (Fernandez et al. 1998). Tan spot is the single most economically damaging crop pathogen in Australia, with losses exceeding AUD$200 million per annum (Murray and Brennan 2009).

The incidence of tan spot has been reported in countries of relatively diverse geographic locations such as Australia (Oliver et al. 2008), the United States (Ali et al. 2010; Andrie et al. 2007), Canada (Aboukhaddour et al. 2013), South America (Gamba et al. 2012), and the Middle East (Lamari et al. 2003). P. tritici-repentis isolates are classified into eight races based on their virulence on specific differential wheat lines (Ali and Francl 2002; Lamari et al. 2003, 1995). However, the classical race structure of P. tritici-repentis has been challenged with the emergence of studies that have identified isolates that do not conform to any of the eight races (Ali et al. 2010; Andrie et al. 2007), suggestive of a broader range of pathotypes and the presence of as yet uncharacterized effectors. The economic importance of wheat and the global distribution of tan spot disease, in conjunction with the complex race structure of the pathogen, highlight the importance of developing detection tools to study the epidemiology and biology of P. tritici-repentis.

Detection of P. tritici-repentis using the conventional cultured-based diagnosis is time consuming and prone to misdiagnosis if diseased leaf samples are contaminated with saprophytes. Imaging techniques such as chlorophyll fluorescence imaging have been evaluated for the detection of tan spot (Kenny et al. 2011). Using fluorescence data, photosynthetic activity was correlated to the resistance or susceptibility of the host, and infection was detected as early as 3 days postinfection (dpi). Although imaging technology offers promises in improving the assessment of plant disease phenotypes, imaging approaches are not an effective diagnostic tool in distinguishing diseases (Mutka and Bart 2015). Since the discovery of polymerase chain reaction (PCR), faster and more robust detection methods employing real-time quantitative PCR (qPCR) have been adopted that combine specificity at taxonomic levels and enable quantification of species DNA in complex matrices such as soil and in planta (Amatulli et al. 2012; Jiménez-Fernández et al. 2011; Sanzani et al. 2014). For in planta analysis, nuclear ribosomal DNA (rDNA) has been extensively used in qualitative PCR and qPCR assays for the detection of fungi and bacteria (Arif et al. 2014; Dorn-In et al. 2015). In fungi, the internal transcribed spacer (ITS) regions are commonly used as probes for taxonomic and phylogenetic identification (Monard et al. 2013; O’Brien et al. 2005). In addition to the general ITS primers that were developed in the early 1990s (Gardes and Bruns 1993; White et al. 1990), numerous primers targeting the ITS region have been utilized as DNA probes in fungi (Ihrmark et al. 2012; Martin and Rygiewicz 2005). The broad application of ITS regions is due to their highly conserved repetitive rDNA sequences that occur as a process of concerted evolution (Elder and Turner 1995), and the extensive depository of ITS sequences available in the public sequence database, providing a large reference resource for species identification (Begerow et al. 2010). ITS regions have a high probability of identifying a broad taxonomic group of fungi (most notably, species of Ascomycota and Basidiomycota) and are regarded as a universal DNA barcode (Schoch et al. 2012). In comparison with microsatellite loci, the copy number of ribosomal small subunit DNA loci has been shown to be 1,000-fold higher compared with simple-sequence repeat (Dreaden et al. 2014), distinguishing rDNA as a popular detection probe.

Although nuclear rDNA sequences are widely used as universal barcodes for fungi, there are limitations in terms of their applications. Tandem repeats of rDNA that vary significantly within species have been reported in various fungal species such as Aspergillus fumigatus (Herrera et al. 2009), Coccidioides spp. (Johnson et al. 2015), and Batrachochytrium dendrobatidis (Longo et al. 2013). Although the multicopy nature of the rDNA region increases the detection sensitivity of small DNA quantity, such variation in the probe copy number is strain dependent and, thus, may lead to over- or underestimation of fungal load in quantitative and comparative studies. Furthermore, although some degree of intragenomic variations are observed (Lindner and Banik 2011; Lindner et al. 2013), the ITS sequences are insufficiently variable, which makes it less desirable when species-level resolution is required. In P. tritici-repentis, sequence analysis of 20 geographically diverse strains revealed only 2 strains with sequence variation of a single base pair (Lepoint et al. 2010).

Comparative genomic analysis of the P. tritici-repentis effector ToxA genetic locus with the colinear locus in Parastagonospora nodorum (a closely related species from the pleosporalean taxa) revealed an indel region located upstream of the transcription site (Friesen et al. 2006). It has been postulated that, in the 1940s, a horizontal gene transfer event occurred in which the ToxA gene was transferred interspecifically from P. nodorum to P. tritici-repentis, thus contributing to the success of P. tritici-repentis as a major pathogen in wheat. The 11-kb sequence which contains the ToxA loci is relatively conserved in both species, with a similarity of 80 to 100% varying at the peripheral and central region. However, a region of 197 bp located upstream from the ToxA gene is only present in the P. tritici-repentis genomic sequence and absent in P. nodorum. It was present as multiple copies in the genome sequence (Manning et al. 2013), indicating that it existed in the P. tritici-repentis genome prior to ToxA invasion. This unique asymmetric region adjacent to the ToxA gene in P. tritici-repentis genome, designated as PtrMulti, allows the development of a species-specific PCR assay for identification and quantification of this pathogen. A preliminary assay using the PtrMulti_F and PtrMulti_R primer pair that targets the indel region was successfully used to differentiate spontaneous yellowing of ‘Mace’ wheat from tan spot disease (Moffat et al. 2015).

In this study, the PtrMulti loci region is evaluated for suitability as a qualitative PCR and qPCR probe, and for application as a diagnostic tool in the detection and quantification of P. tritici-repentis in wheat field samples. The advantages and limitations of the PtrMulti as a qPCR diagnostic tool are discussed.

Materials and Methods

Plant and fungal materials.

Seed of the Australian wheat ‘Yitpi’ (Triticum aestivum L.), rated as susceptible to very susceptible (DAFWA 2015), were sown in two independent 2 × 1 m plots in June 2013. The plots were overlaid with wheat stubble as the primary source of inoculum collected from a field with high tan spot disease pressure. Whole wheat plant samples were collected at a monthly intervals (July, August, September, and October) at the growth stages of stem elongation (GS30), flag leaf emergence or booting (GS39 to GS40), flowering (GS60), and grain fill period (GS70). To overcome sample variability resulting from natural infection, eight plants were randomly collected from the plot at each time point. Tan spot disease was visually assessed and leaves photographed. Plant material was stored at −80°C prior to DNA extraction.

To represent the genetic diversity of the P. tritici-repentis population, pure isolates from diverse geographic locations in Australia as well as two American isolates were used in the primer specificity assay (Table 1). Race designation of the Australian P. tritici-repentis isolates was determined by PCR amplification of DNA using primers ToxAscreeningF/R and TB10f/TB12r to detect the presence or absence of ToxA and ToxB genes, respectively (Antoni et al. 2010), as well as visual assessment of disease phenotype on wheat differential lines (Lamari et al. 2003). The race of isolates DW5 and DW7 was based on published data (Friesen and Faris 2004). Pure fungal cultures of Alternaria tritici, A. infectoria, and Fusarium lateritium were obtained from the field and the fungal species were confirmed by sequencing the ITS regions, as previously described by Moffat et al. (2015).

Table 1. Pyrenophora tritici-repentis isolates used in this study

Isolate	Year	Geographic origin	Race
M4	2009	Western Australia, Australia	1
WAC11137	2001	Western Australia, Australia	1
239	2003	Victoria, Australia	1
134	2006	Queensland, Australia	1
WAC5216	1987	New South Wales, Australia	1
DW5	1998	North Dakota, United States	5
DW7	1998	North Dakota, United States	5

View as image HTML

DNA extraction of plant and fungal materials.

For individual wheat plants, each leaf was excised, freeze-dried, and ground into a fine powder in liquid nitrogen using a mortar and pestle. A fine balance was used to measure out 10 mg of ground leaf material, which was placed in a 1.5-ml microtube. Plant DNA was extracted using the Biosprint 15 DNA kit (Qiagen) according to the manufacturer’s protocol.

Fungal mycelia of P. tritici-repentis, P. teres f. maculata, P. teres f. teres, A. tritici, A. infectoria, and F. lateritium were grown on V8 potato dextrose agar media plates and harvested after 10 days (Moffat et al. 2014). DNA was extracted from the mycelium using Biosprint 15 DNA kit (Qiagen) according to the manufacturer’s protocol.

DNA concentrations were determined using a NanoDrop 2000 UV-Vis spectrophotometer (Thermo Scientific). DNA from pure culture of P. nodorum, Ascochyta rabiei, A. lentil, and Botrytis cinerea were kindly provided by other researchers.

Sequence analysis and primer design.

A 1.4-kb region containing the ToxA gene from P. tritici-repentis was identified in the genomic scaffold_00011 (isolate M4) and aligned to the orthologous DNA sequence of P. nodorum. The sequence of P. nodorum containing the ToxA locus (SNOG_16571) and the flanking regions was downloaded from Broad Institute of MIT (http://www.broadinstitute.org). Nuclear ribosomal RNA sequences of genus Pyrenophora were retrieved from fungal barcoding database (http://www.fungalbarcoding.org) (Schoch et al. 2012) and the P. tritici-repentis rDNA sequences used in this study were obtained from P. tritici-repentis genome database (http://www.broadinstitute.org).

To determine the repeat regions of PtrMulti, the 235-bp region containing the PtrMulti sequence was identified in the genomic scaffolds of isolate M4. For the comparison of PtrMulti sequences in P. teres species, the region was subjected to blast searches of P. teres f. maculata and P. teres f. teres genomes. Sequence analysis and multiple sequence alignment were constructed using Geneious version 6.1.8. Primers were designed using Primer3.

Primer specificity.

Conventional PCR was used to evaluate the specificity of PtrMulti_F (5′ GTAAGCCCGAGCAGAAGGAC 3′) and PtrMulti_R (5′ CCATAGGCGACCGAGTAGAG 3′) via amplifications with fungal strains, including P. tritici-repentis, P. nodorum, Alternaria spp., Ascochyta spp., B. cinerea, and F. lateritium. The specificity of the single-copy locus PtrUnique region has been described previously (Antoni et al. 2010) and was not tested in this study. Each 20 μl of PCR contained 1× MyTaq buffer, 250 nM each forward and reverse primer, 1 U of MyTaq DNA polymerase (Bioline), and 50 ng of DNA template. Amplification was performed with the following cycling conditions: 2 min of denaturation, 35 cycles of 30 s of denaturation at 95°C and 30 s of annealing at 59°C, and 1 min of extension at 72°C. Primer pair NL5mun (Egger 1995) and NLC2 (Martin and Rygiewicz 2005) that target the large subunit (LSU) of the ribosomal operon was used as a PCR amplification control.

For the barley pathogens P. teres f. maculata and P. teres f. teres, the primer specificity of PtrMulti_F/R, Ptt_F (5′ ATAAGCCCGAGCAGAAGAAC 3′) and Ptt_R (5′CCATTAGAAAACCGAGTAGAG 3′) was tested on the DNA of two different strains of each P. teres form using DyNAzyme Ext Taq DNA polymerase (Thermo Fisher Scientific). Each 20 μl of PCR contained 1× DyNAzyme Ext Taq buffer, 250 nM each forward and reverse primer, 1 U of DyNAzyme Ext Taq DNA polymerase, and 50 ng of DNA template. Amplification was carried out using the following cycling conditions: 2 min of denaturation, 35 cycles of 30 s of denaturation at 95°C and 30 s of annealing at 65°C, and 30 s of extension at 72°C. PCR products were electrophoresed on 1.5% agarose gel prestained with SYBR safe (Life Technologies).

Determination of PtrMulti copy number.

Copy number determination of PtrMulti was performed using the qPCR standard curve method, as previously described by Whelan et al. (2003). Briefly, primer pairs PtrMulti_F/R plus Act1F4 (5′CGAGACCTTCAACGCTCCCGC 3′) and Act1R4 (5′ GCGTGGGGAAGAGCGAAACCC 3′) were used to amplify 150 bp of the PtrMulti region and the single-copy actin gene (Act1), respectively, in 20 μl of PCR containing 1× MyTaq buffer, 250 nM each forward and reverse primer, 1 U of MyTaq DNA polymerase, and 50 ng of M4 DNA. Amplification was performed with the following cycling conditions: 2 min of denaturation, 35 cycles of 30 s of denaturation at 95°C and 30 s of annealing at 55°C, and 1 min of extension at 72°C. PCR products were analyzed in gel electrophoresis and gel purified using the QIAquick Gel Extraction Kit (Qiagen) according to the manufacturer’s instruction. Purified PCR amplicons were quantified using a NanoDrop 2000 UV-Vis spectrophotometer (Thermo Scientific). Based on the DNA concentration and sequence length, the copy numbers of the PtrMulti and Act1 fragments were calculated using the following formula (Whelan et al. 2003): copy number = [6.02 × 10²³ (copy/mol) × DNA concentration (g)]/[DNA length (bp) × 660 (g/mol/bp)].

A 10-fold serial dilution of 10¹ to 10⁸ copies for each fragment was then prepared as template DNA for real-time PCR. To generate the standard curves, real-time qPCR detection was carried out in a 20-μl reaction consisting of template DNA, 10 μl of QuantiTect SYBR Green PCR mix (Qiagen), and 300 nM each forward and reverse primer. Real-time PCR was performed in a CFX96 Real-time PCR Detection System (Bio-Rad) with the following cycling conditions: 95°C for 15 min of denaturation, 35 cycles of 15 s of denaturation at 94°C and 30 s of annealing at 65°C, and 30 s of extension at 72°C, with the acquisition of fluorescence data during each step of annealing cycle. All the reactions were performed with three replicates. The standard curves were generated by plotting the threshold cycle (Ct) value against the log copy number. The PCR efficiencies of the reactions were calculated from the standard curve slope using the following equation: E = 10^(−1/slope) − 1. To determine the genome copy number of PtrMulti, qPCR (20 μl) assays were prepared as described above, except that the DNA template consisted of 50 ng of genomic DNA of P. tritici-repentis isolates (Table 1). Absolute quantification was then performed using the standard curves to calculate the copy number of PtrMulti and Act1 corresponding to the Ct values obtained from the real-time PCR analysis of the genomic DNA. The PtrMulti copy number was then normalized to Act1 copy number. Because Act1 is single copy, the copy numbers obtained for PtrMulti represent the genome copy number.

Sensitivity of PtrMulti primers.

To assess the performance of the multiloci primer pair PtrMulti_F/R, a comparison between PtrMulti_F/R and the single-locus primer pair PtrUniqueF1 (5′ CCGGCTGATATTCCCCCACCG 3′) and PtrUniqueR1 (5′ GGCCTCCGCTGGCGAGTAAA 3′) was assessed in a real-time PCR assay with a 10-fold serial dilution of P. tritici-repentis DNA (from the range of 0.001 to 100 ng). To evaluate the performance of these primer pairs in infected wheat materials, a series of diluted wheat DNA (1, 10, and 100 ng) was spiked with P. tritici-repentis DNA at concentration of 0.1, 0.01, and 0.001 ng to simulate a DNA mixed matrix of pathogen-host DNA at ratios of 1:10, 1:100, and 1:1,000. The qPCR (20 μl) consisted of template DNA, 10 μl of QuantiTect SYBR Green PCR mix (Qiagen), and 300 nM forward and reverse primers. Real-time PCR was performed as described above. At the end of the qPCR, a default dissociation curve was performed for melt curve analysis. Standard curves were generated by plotting the Ct values against the log concentration of the 10-fold serial diluted DNA. Concentrations of P. tritici-repentis DNA were then inferred from the standard curves using the absolute quantification method. All the reactions were performed with three replicates.

Detection of P. tritici-repentis DNA in planta using PtrMulti qPCR assay.

qPCR was performed using primer PtrMulti_F and PtrMulti_R to detect relative amounts of P. tritici-repentis DNA in the leaf material. The components for the qPCR were as follows: 50 ng of template DNA, 10 μl of QuantiTect SYBR Green PCR mix (Qiagen), and 300 nM PtrMulti_F and PtrMulti_R. A standard dilution series of P. tritici-repentis DNA at 100, 50, 10, 1, 0.1, and 0.01 ng was included in the qPCR analysis to generate the standard curve for absolute quantification. Real-time PCR was performed as described above. For each time point, at minimum, seven biological repeats were analyzed to overcome biological variation. Standard curves were generated by plotting the Ct values against the log concentration of DNA. The resulting data were assessed statistically using one-way factorial analysis of variance. Mean comparisons were performed using an all-pairs Tukey’s test. All statistically analyses were performed using JMP version 10 software.

Photosynthetic pigment measurements.

Freeze-dried plant materials were ground to fine powder using a mortar and pestle with liquid nitrogen. A homogenized leaf sample (10 mg) was transferred to a microtube and extracted with 120 μl of 80% (vol/vol) acetone. The tubes were wrapped in foil and mixed for 30 min on a rocking platform. Samples were then centrifuged at 3,000 rpm for 15 min at 4°C. The crude acetone extract (50 μl) was then transferred to a flat-bottomed 96-well plate. Chlorophylls (a and b) and carotenoid were measured spectrophotometrically at wavelengths of 663, 646, and 470 nm using a Synergy HT multimode microplate reader (BioTek). The concentrations of the pigments chlorophyll a, chlorophyll b, and carotenoid were calculated according to Lichtenthaler and Wellburn (1983).

Results

Sequence analysis and primer design of PtrMulti region.

Alignment of the 1.4-kb locus region of the ToxA gene from P. tritici-repentis and P. nodorum showed the indel polymorphic region of 282 bp located 669 bp upstream from the gene (Fig. 1). In order to evaluate the indel sequence within the P. tritici-repentis genome, this region was identified in the genome of isolate M4. A DNA region of 235 bp containing the indel sequence not only was found upstream from the ToxA promoter but also was conserved within the genome, as shown in the alignment in Figure 2. Initially, two sets of primer pairs (F1/R1 and F2/R2) that generate a short PCR fragment (150 bp) suitable for qPCR were designed targeting the 235-bp multilocus region. Primer specificity was tested on various plant fungal pathogens that included closely related Pleosporales species and saprobes that are commonly found in wheat. From the two sets of primer pairs, F2/R2 generated the expected amplicon of 150 bp from P. tritici-repentis DNA, whereas primer set F1/R1 had unspecific amplification bands on most of the DNA tested (data not shown). Thereafter, primer pair F2/R2, which was named PtrMulti_F and PtrMulti_R, was used for the subsequent PCR assays. PtrMulti_F/R was species specific and no amplification was observed for the DNA of other pathogens examined (Fig. 3A). A blast search of the 235-bp region against the National Center for Biotechnology Information GenBank database revealed no significant homology to other fungal species of Pyrenophora except for the P. teres f. teres genome, which revealed a region that has a 62% identity to the PtrMulti region (Fig. 4). To examine these regions of homology between P. tritici-repentis and P. teres, two sets of distinct primers (PtrMulti_F/R and Ptt_F/R) were designed and tested for their specificity against P. tritici-repentis and P. teres isolates using conventional PCR. Primer optimization was essential in order to obtain specific amplicons that distinguish P. tritici-repentis and P. teres f. teres due to their close homology. Upon optimization of the PCR, a single amplification size of 150 bp that corresponds to the genomic DNA sequences of P. tritici-repentis and P. teres f. teres using primer pairs PtrMulti_F/R and Ptt_F/R, respectively, was achieved (Fig. 3B). The conventional PCR assay using primer pair Ptt_F/R showed an amplicon of 150 bp in size for the two P. teres f. teres DNA isolates tested in the assay and, as expected, no amplification was observed for P. tritici-repentis DNA. However, DNA from the two P. teres f. maculata isolates also yielded no amplification, indicating the presence of the PtrMulti-like region in P. teres f. teres but not in P. teres f. maculata.

Download as PowerPoint

Download as PowerPoint

Download as PowerPoint

Download as PowerPoint

Detection of PtrMulti copy number in P. tritici-repentis.

The copy number of the PtrMulti region was determined by normalizing the target region to the actin Act1 gene, which was confirmed by whole-genome sequence to be a suitable single-copy reference gene (Ellwood et al. 2012). To perform the normalization strategy, first, a 10-fold serial dilution of the DNA amplicon ranging from 10¹ to 10⁸ copies was employed to generate the standard curves for PtrMulti and Act1, respectively. The standard curves of the log DNA versus Ct for both PtrMulti and Act1 were near linear (coefficient correlation value R² of 0.9989 and 0.9969, respectively), with slope values of −3.5455x (PtrMulti) and −3.5217x (Act1). Calculated amplification efficiencies were comparable between PtrMulti and Act1 at 91.4 and 92.3%, respectively. The copy numbers of PtrMulti and the Act1 gene (single-copy control) of the local and two overseas P. tritici-repentis isolates listed in Table 1 were calculated using the standard curves obtained. Because Act1 is single copy, the relative copy number of PtrMulti to actin represents the genome copy number. The qPCR data showed that the isolates tested have near equal copy number of PtrMulti, ranging from 63 ± 20 to 85 ± 22 copies per genome (Fig. 5).

Download as PowerPoint

Specificity and sensitivity of PtrMulti qPCR.

In order to use PtrMulti as an in planta probe, the amplification specificity and sensitivity of PtrMulti were assessed in a simulated mixed matrix of fungal and plant DNA. The PtrMulti_F/R primers were tested by performing qPCR using wheat plant DNA spiked with P. tritici-repentis DNA at the ratio of 1:10, 1:100, and 1:1,000. The qPCR analysis showed that the amplification of PtrMulti was specific, as indicated by a single melt curve (Supplementary Fig. S1). No amplification curves were detected in wheat DNA-only samples. Absolute quantification analysis using PtrMulti as the probe detected the P. tritici-repentis DNA at the expected concentration levels of 1, 10, and 100 pg (Fig. 6A). In comparison with the single-locus PtrUnique, qPCR amplification detected P. tritici-repentis DNA at the levels of 10 and 100 pg. However, amplification curves were below the threshold for DNA level of 1 pg. The lower sensitivity of PtrUnique F1/R1 is shown in the regression curve generated from the serial diluted DNA with corresponding Ct value of 34 for DNA at 0.01 ng, whereas the equivalent DNA level of PtrMulti_F/R has a Ct value of 27 (Fig. 6B). This result confirms that PtrMulti amplification is not only species-specific but also has a higher sensitivity in comparison with a single-locus probe, which makes the PtrMulti_F/R SYBR qPCR assay a more sensitive and specific assay to detect the presence of P. tritici-repentis DNA in planta.

Download as PowerPoint

Detection of tan spot disease in naturally infected wheat.

The suitability of the PtrMulti_F/R primers and qPCR assay were evaluated on tan-spot-infected Yitpi wheat. Growth stages of stem elongation (GS30), flag leaf emergence or booting (GS39 to GS40), flowering (GS60), and grain fill (GS70) were chosen for qPCR analysis to represent a cross section of wheat development phases. Specificity and sensitivity of the assay was assessed by the detection of P. tritici-repentis DNA on the emerged leaves of individual wheat plants from the oldest (L1) to the youngest (L6) leaves collected for a duration of 4 months, representing the different growth stages.

From visual inspection, this cultivar developed typical symptoms of tan spot during the growing period, with distinctive lesion of tan with chlorotic halos clearly visible on older leaves of samples collected from August onward (Fig. 7B, C, and D), whereas less symptoms were observed and no distinctive lesions were visible on the youngest leaf (L3) of the wheat samples for the first samples collected (Fig. 7A). In the qPCR assay, in general, DNA of P. tritici-repentis was detected in all the wheat leaf samples collected from July to October but differed in the levels of DNA concentrations (Fig. 8). In the stem elongation stage, emergence of tan spot symptoms was seen on the older leaves (L1 and L2; Fig. 7A). Although no distinct tan spot symptoms were observed on the youngest leaf (L3), qPCR analysis detected P. tritici-repentis DNA in all the leaves (Fig. 8). A concentration of 0.0111 ± 0.0050 ng was detected in L1, while a very low amount of DNA, at a concentration of 0.0006 ± 0.0003 and 0.0001 ± 5E-05 ng, was detected in L2 and L3 leaves, respectively. Although the amount of DNA detected in the older leaves was higher (100-fold) than the younger leaves, Tukey’s test analysis showed no significant difference. For the next growth stage, a very high level of P. tritici-repentis DNA at a concentration of 4.4662 ± 0.6882 ng was observed in the oldest leaves (L1), which were predominantly senesced leaves. A sequential decrease in the level of DNA was observed from older L2 leaf to younger L4 leaf, with concentrations at 0.8076 ± 0.3926, 0.1281 ± 0.0954, and 0.0097 ± 0.0047 ng, respectively. The fungal DNA for the following leaves (L5 and L6) remained low, at concentrations of 0.0002 ± 0.0001 and 0.0009 ± 0.0003 ng, respectively. At the flowering stage, the symptoms were less discerning in L1 and L2 leaves as the lower leaves were going through senescence (Fig. 7C). The concentration of P. tritici-repentis DNA was higher in the older L1 leaf in comparison with the other leaves at that stage. A half-fold of DNA level was detected in the L2 leaf, although this was not statistically significantly different from L1. The remaining L3 and L4 leaves had lower DNA levels of 0.0591 ± 0.0247and 0.0015 ± 0.0009 ng, respectively. For the fourth samples collected in October, the P. tritici-repentis DNA levels for all the leaves were consistently high within the range of a nanogram, which was reflected by the more progressed disease symptoms observed on the leaves (Fig. 7D). Leaf L1 had a DNA level of 1.1951 ± 0.4759 ng; however, fungal DNA detected in the younger leaves (L2 and L3) was significantly higher, with a four- to sixfold increase in the levels of DNA.

Download as PowerPoint

Download as PowerPoint

Determination of photosynthetic pigments in infected wheat leaves.

In the stem elongation stage, the leaves (L1, L2, and L3) had similar range of chlorophyll a and b content at 5.60 ± 0.78, 6.18 ± 1.96, and 3.51 ± 0.54 μg ml⁻¹, respectively (Fig. 9). The carotenoid content in the leaves was more than 10-fold lower than the chlorophyll contents, ranging from 0.40 ± 0.14 to 0.63 ± 0.12 μg ml⁻¹. Similar to the chlorophyll contents, no significant difference in the concentration of carotenoid between the leaves was found. For the August samples, the chlorophyll content of the oldest leaf (L1) was significantly lower at 0.49 ± 0.05 μg ml⁻¹ in comparison with the younger leaves (L3, L4, and L5). These leaves had more than 10-fold higher measurement of chlorophyll a and b than L1. The other leaves (L2 and L6) had chlorophyll contents of 3.78 ± 1.24 and 4.48 ± 1.24 μg ml⁻¹, respectively, which was not statistically significantly different from L1. The distributions of the carotenoid concentrations between the leaves of different ages were similar to the chlorophyll concentrations, with a significantly lower level detected in the oldest leaf (L1) and increased carotenoid levels in the younger leaves. A very low level of carotenoid was detected in leaf L1 at concentration of 0.17 ± 0.09 μg ml⁻¹. The rest of the younger leaves had higher carotenoid concentrations, ranging from 0.70 ± 0.23 to 1.20 ± 0.30 μg ml⁻¹, with no significant difference observed between the leaves. The leaves from the September sample collection had a concentration of chlorophyll a and b that ranged from of 1.33 ± 0.75 to 5.73 ± 2.52 μg ml⁻¹. Data showed a gradual increment in the concentrations of chlorophyll a and b corresponding to the age of the leaves, with the highest concentration detected in L4. The carotenoid levels were not significantly different between the leaves (L1 to L4). The chlorophyll content of the leaf samples of different ages collected in October were not statistically different, with the concentrations ranging from 0.81 ± 0.76 to 2.9 ± 2.02 μg ml⁻¹. The carotenoid content of L1 was 0.21 ± 0.25 μg ml⁻¹; however, no carotenoid was detected in L2 and L3.

Download as PowerPoint

Discussion

In the past, timely diagnosis of tan spot disease in wheat using conventional methods was challenging due to the reliance of agar plate fungal culturing and microscopic observation for identification. However, certain Pyrenophora spp. are not distinctly different in morphology (Zhang and Berbee 2001). For this reason, the tan spot diagnosis method requires a combination of conventional PCR using species-specific primers or sequencing of the ITS region in order to confirm at the species level. With the technology of fluorescent detection and the availability of genome information, the development of plant disease diagnostics has been geared toward a more robust, rapid, and specific assay for pathogen detection, with a certain degree of quantitation. A multicopy-locus probe, PtrMulti, was developed to facilitate the diagnosis of tan spot disease, and the probe proved useful in discerning tan spot symptoms from spontaneous yellowing of wheat leaves in Mace (Moffat et al. 2015). In this study, PtrMulti was shown as a multicopy locus with near-equal copies per genome for the different geographical isolates examined. In contrast to other multilocus probes that have more than twofold differences between the isolates (Herrera et al. 2009; Johnson et al. 2015; Longo et al. 2013), the lack of variation between the PtrMulti copy numbers of the different isolates makes the PtrMulti probe useful for comparative analysis of P. tritici-repentis loads between the diverse pathogen populations.

In order to employ PtrMulti as a routine diagnostic probe, a wider application of PtrMulti was evaluated in this study. PtrMulti not only was used as a diagnostic probe but also was assessed as a quantitation probe to determine the levels of tan spot disease on naturally infected wheat. Amplification of PtrMulti_F/R is clearly specific, with a distinct band observed only in P. tritici-repentis and not in other closely related Pleosprales members such as P. nodorum and Alternaria spp., which have been shown to coexist on infected tissues as complexes or opportunistic saprophytes (Bhathal et al. 2003; Moffat et al. 2015). Even though a locus in P. teres f. teres genome showed some homology to the PtrMulti probe region, further optimization of the PCR conditions (i.e., higher annealing temperature and proofreading DNA polymerase) for the PtrMulti_F/R primer pair could discern the two different pathogens. Furthermore, P. tritici-repentis and P. teres f. teres are two major pathogens on different crops and they produce very different foliar disease symptoms. If situations of mixed hosts of wheat and barley are routinely encountered, it is recommended to incorporate fluorescence-labeled probes in the middle of the amplicon such as Taqman because this may well increase the specificity to the target multiloci PtrMulti region. Interestingly, in this study, preliminary quantitative analysis of qPCR primer set Ptt_F and Ptt_R produced the amplicon from genomic DNA of P. teres f. teres but not P. teres f. maculata. More analyses of P. teres isolates are needed to evaluate the specificity of Ptt_F/R primers because, in this study, only two isolates of each P. teres form were examined. It is worth further investigating this locus in P. teres because, in this study, the primer pair Ptt_F/R appears to be P. teres f. teres species specific.

The detection of the multilocus PtrMulti has a higher sensitivity in comparison with the single-locus PtrUnique. Quantification of P. tritici-repentis DNA using PtrMulti was feasible over three orders of magnitude (from 1 to 100 pg) whereas PtrUnique has a detection limit of >1 pg. The higher sensitivity of PtrMulti amplification is seen in the standard curve, with 10 pg detected at a Ct value of 27. Based on the regression curve, the PtrMulti detection limit would be 80 fg. The detection sensitivity was demonstrated in wheat field samples with the quantification of P. tritici-repentis DNA in the range of fentograms mainly in the younger leaves, with less discerning tan spot symptoms leaves at the stem elongation stage. This highlighted the potential of PtrMulti to be used for the early detection of P. tritici-repentis in wheat leaves without distinct tan spot symptoms. The sensitivity of PtrMulti detection is consistent with other multicopy probes used in qPCR detection of plant diseases such as those that employ rDNA ITS multicopy loci with the detection of pathogen DNA at fentogram levels (Arif et al. 2014; Molnar et al. 2013). A comparable detection study of the Phomopsis scleorotiodies pathogen using rDNA primers reported a lower limit detection level of 0.4 fg for both SYBR green and Taqman probe-based real-time PCR assays, which was shown to be 100 times more sensitive than a standard PCR assay (Shishido et al. 2013). Meanwhile, the detection limit of a single-copy probe such as the elongation factor gene usually falls in the range of ≥1 pg (Amatulli et al. 2012; Haegi et al. 2013), as shown in the detection limit of the single-locus PtrUnique in this study.

A concern with most qPCR detection methods is the inability of the assay to discern whether the quantified DNA is from viable fungal materials or dead cells. Although, in this study, the viability of the fungal load present in the infected leaves was not assayed, the levels of P. tritici-repentis DNA detected corresponded with the visual assessment of the foliar disease. Numerous studies have investigated the correlation of positive quantification of PCR with the detection of fungal propagules in the infected tissues, and have found significant correlations. Significant correlation has been demonstrated between quantification of pathogen DNA using a qPCR method with conventional agar plate culture in the studies of Sclerotinia sclerotiorum infection of oilseed red petals (Almquist and Wallenhammar 2015) and P. teres seed infection (Bates et al. 2001). Intriguingly, a qPCR assay supported the observation of different degrees of F. oxysporum f. sp. melonis colonization on melon stem between compatible and incompatible interactions, with higher fungal DNA detected in compatible interactions (Haegi et al. 2013). These studies demonstrated that levels of DNA correlate with the degree of pathogenicity, thus supporting the application of PtrMulti as a probe for quantitative diagnostic.

The qPCR data in this study generally showed that younger leaves have less P. tritici-repentis DNA, and a higher incidence of infection was observed in older leaves, indicated by a higher concentration of the DNA. A higher incidence of pathogens in lower leaves was also reported for S. sclerotiorium (Almquist and Wallenhammar 2015). There are several reasons that have been suggested for this occurrence; lower leaves are exposed longer to airborne spores, are closer to the ground, have a larger surface area, and are more senescent. Despite varying degrees of infection observed at the stem elongation and ear emergence stages, at the last sampling date, all the flag leaves tested had high levels of P. tritici-repentis DNA. This high incidence of tan spot disease indicates a continual progression of disease development throughout the various growth stages examined in this study. This observation is of a particular disease progression of tan spot infection, with infection mostly occurring in older leaves (Summerell and Burgess 1988). This study shows that, even though P. tritici-repentis DNA was detected at a lower concentration at the earlier growth stage and on younger leaves, at the later stage, the tan spot disease can progress and continue to intensify, producing large chlorotic and necrotic lesions and, consequently, decreasing the photosynthetic pigments of the leaves. This is perhaps influenced by conditions such as rainfall during the growing season that promote the development of disease. The lack of disease suppression in this study is likely caused by the susceptibility of the cultivar that was selected. At the maturation stage, the foliar structure also goes through senescence, with a reduction of photosynthetic pigment content in the leaf samples of the fourth sampling stage. Tan spot disease continues to progress as the pathogen enters the saprophytic phase of development and colonizes the senescent leaf and leaf sheath. Higher concentrations of P. tritici-repentis DNA appear to occur on leaf samples with lower contents of photosynthetic pigments, supporting the visual observation by Summerell and Burgess (1988) that leaf senescence stimulates P. tritici-repentis host colonization.

To our knowledge, this is the first report of using molecular diagnostics for tan spot disease in the field. The multicopy-locus PtrMulti_F/R SYBR green qPCR assay was reliable in the detection and quantification of P. tritici-repentis on wheat leaves of naturally infected field samples. The sensitivity of the diagnostic assay has an advantage in the early detection of tan spot disease and as a useful tool for managing application of fungicide. Such application also has practical implications as a quantitative tool to study the epidemiology of the tan spot pathogen in the field and aid in the determination of high-risk areas. Furthermore, the PtrMulti SYBR green PCR assay is relatively inexpensive in comparison with other fluorescence technology for large-scale, routine monitoring of tan spot disease or rapid screening of germplasm for enhanced resistance.

Acknowledgments

This work was supported by grants from the Australian Grains Research and Development Corporation. We thank T. Friesen (United States Department of Agriculture–Agricultural Research Service) for the U.S. P. tritici-repentis isolates; the research staff at the Centre for Crop and Disease Management, Curtin University, for providing DNA of P. nodorum, Ascochyta spp., and B. cinerea; the Department of Agriculture and Food Western Australia and Queensland Department of Agriculture and Fisheries for providing the Australian isolates; and S. Ellwood (Curtin University) for his technical assistance.