Prevalence of mating type idiomorphs in Pyrenophora teres f. teres and P. teres f. maculata populations from the Canadian prairies

Abstract

The fungus Pyrenophora teres Drechs. occurs as two morphologically similar but genetically distinct forms, P. teres f. teres (Ptt) and P. teres f. maculata (Ptm), which cause the net form and spot form of net blotch of barley, respectively. A collection of 220 isolates from the Canadian prairie provinces (Alberta, Saskatchewan and Manitoba) was evaluated for mating type (MAT) idiomorph distribution and frequency. Fungal isolates were classified as Ptt or Ptm using form-specific polymerase chain reaction (PCR) primers. PCR analysis with MAT-specific primers indicated that the MAT1 and MAT2 idiomorphs of Ptt and Ptm could be identified within the same field, on the same plant, and on the same leaf. There was no significant departure from the expected 1:1 MAT1/MAT2 ratio for both forms in all three provinces or in the Canadian prairies population as a whole. Polymorphic simple sequence repeat primers were used to detect evidence of possible recombination between the two forms. Cluster analysis revealed that all P. teres isolates, including 30 isolates causing intermediate symptoms, clustered in two distinct groups conforming to either Ptt or Ptm. Therefore, hybridization was not detectable from the 220 isolates collected in western Canada. Pyrenophora teres f. teres is still the dominant form (58%) of the net blotch pathogen, and the data suggest both Ptt and Ptm go through regular cycles of sexual reproduction in the Canadian prairies.

Résumé

Le champignon Pyrenophora teres Drechs. exixte sous deux formes morphologiquement similaires, mais génétiquement différentes, P. teres f. teres (Ptt) et P. teres f. maculata (Ptm), qui causent, respectivement, la forme réticulée et la forme localisée de la tache réticulée de l’orge. Une collection de 220 isolats provenant des Prairies canadiennes (Alberta, Saskatchewan et Manitoba) a été utilisée pour analyser la distribution et de la fréquence des types sexuels (MAT) idiomorphes. Les isolats fongiques ont été classés en tant que Ptt ou Ptm à l’aide d’amorces spécifiques pour la réaction en chaîne de la polymérase (PCR). L’analyse par PCR avec amorces spécifiques du MAT a indiqué que les idiomorphes MAT1 et MAT2 de Ptt et de Ptm pouvaient être décelés dans un même champ, sur une même plante et sur une même feuille. Il n’y avait pas de déviation significative du ratio attendu de 1:1 des MAT1 et MAT2 quant aux deux formes, et ce, dans les trois provinces ou dans l’ensemble de la population des Prairies canadiennes. Des amorces spécifiques des microsatellites ont été utilisées pour détecter une possible recombinaison entre les deux formes. L’analyse typologique a révélé que tous les isolats de P. teres, y compris 30 isolats causant des symptômes atypiques, étaient regroupés en deux différentes grappes conformes à Ptt ou à Ptm. Par conséquent, il a été impossible de détecter de l’hybridation chez les 220 isolats collectés dans l’Ouest canadien. Pyrenophora teres f. teres demeure la forme dominante (58 %) de l’agent pathogène causant la tache réticulée, et les données suggèrent Ptt et Ptm amorcent régulièrement des cycles de reproduction sexuée sur les Prairies canadiennes.

Keywords:

• barley
• mating type
• molecular detection
• net blotch
• Pyrenophora teres

Mots-clés:

• détection moléculaire
• orge
• Pyrenophora teres
• tache réticulée
• type sexuel

Introduction

The heterothallic fungus Pyrenophora teres Drechs. (anamorph: Drechslera teres [Sacc.] Shoem.) has two morphologically similar but genetically distinct forms: P. teres f. teres (Ptt) and P. teres f. maculata (Ptm), which cause the net form of net blotch (NFNB) and spot form of net blotch (SFNB), respectively, on barley (Hordeum vulgare L.). Both NFNB and SFNB are economically important foliar diseases throughout the major barley growing regions of the world, including western Canada (Mcdonald 1963; Smedegard-Petersen 1978; Tekauz 1990; Rau et al. 2005; Liu et al. 2011). Both forms of net blotch are stubble-borne diseases, producing asexual conidia and sexual pseudothecia, which produce ascospores on overwintered infected crop debris (Van Den Berg & Rossnagel 1991; Liu et al. 2011). Yield losses of 10–40% were reported as typical in severe cases of NFNB, but the pathogen has the potential to cause total yield loss (Mathre 1997; Murray & Brennan 2010). Similarly, yield losses of up to 44% were reported for SFNB (Jayasena et al. 2007). Crop rotation, fungicide application and the use of resistant cultivars are all effective in managing net blotch and are components of an integrated management approach in barley (Tekauz 1990; Turkington et al. 2011).

The last major study of variation in the populations of P. teres from western Canada was conducted in the 1980s, and revealed that Ptt was the dominant form of the fungus, representing 82% of the isolates in a collection that came mainly from the prairie provinces (Alberta, Saskatchewan and Manitoba) (Tekauz 1990). Nevertheless, Ptm was shown to be important locally in some areas of Saskatchewan, with SFNB being the most prevalent foliar disease of spring barley and more important economically than NFNB in those areas (Weller & Rossnagel 1988; Tekauz 1990; Van Den Berg & Rossnagel 1991).

Pyrenophora teres can reproduce sexually and asexually; therefore, the genetic structure of the pathogen population is dependent on the relative importance of these two types of reproduction in the fungal life cycle (Liu et al. 2011). Some studies have shown that sexual reproduction is important in P. teres populations (Peever & Milgroom 1994; Jonsson et al. 2000; Rau et al. 2003). In contrast, it also has been reported that reproduction within some P. teres populations is mainly asexual (Campbell et al. 2002; Lehmensiek et al. 2010). Information regarding the extent of asexual versus sexual reproduction in the net blotch pathogen is required to understand its evolutionary potential, which in turn will help to assess the durability of the resistance present in existing and future cultivars (Sommerhalder et al. 2006).

In Canada, Piening (1961) noted the general occurrence of mature ascocarps of P. teres on barley straw from many fields in the Calgary and Edmonton regions of Alberta. Similarly, Duczek et al. (1999) found pseudothecia of P. teres in two fields near Dafoe and Churchbridge, Saskatchewan. The P. teres teleomorph also was identified in Quebec in 1940 (Crowell 1941 cited in Piening 1961). In another study, Piening (1968) indicated that ascospores of P. teres were responsible for almost 50% of all net blotch lesions examined on volunteer barley plants in a field at the Lacombe Research Station in Alberta.

Polymerase chain reaction (PCR)-based mating-type studies can be useful to evaluate the potential of P. teres for sexual recombination (Rau et al. 2005). In all heterothallic ascomycetes including P. teres, sexual compatibility and recombination are controlled by a single regulatory mating-type (MAT) locus (Kronstad & Staben 1997; Turgeon 1998). The two alleles present in the mating type locus occupy the same chromosomal position and are referred to as idiomorphs, since they consist of two different sequences and encode dissimilar transcripts (Metzenberg & Glass 1990; Kronstad & Staben 1997; Rau et al. 2005). The existence of two fungal strains of different idiomorphs in close proximity to each other is a prerequisite for the development of the teleomorph stage, with each mating type detecting the other through pheromones produced by the opposite type (Kronstad & Staben 1997; Turgeon 1998; Rau et al. 2005; Sommerhalder et al. 2006; Vail & Banniza 2009).

A 1:1 mating type ratio is assumed when regular random mating occurs within populations (Milgroom 1996). The hypothesis that regional populations of ascomycete pathogens proceed to the teleomorph stage could be examined by studying the occurrence, distribution and frequencies of the two pathogen mating types (Serenius et al. 2005; Sommerhalder et al. 2006). If the net blotch pathogen can produce pseudothecia, then ascospores could initiate disease in barley fields as the primary source of inoculum. For ascospores to be considered the major cause of primary infections, however, the two mating types must occur in statistically equal frequencies (Rau et al. 2005; Sommerhalder et al. 2006; Bogacki et al. 2010). Departures from statistically equal frequencies of the two mating types would be associated with a likely predominance of asexual reproduction via conidia (Sommerhalder et al. 2006). Several studies of P. teres have successfully used this approach to assess the relative importance of sexual versus asexual reproduction in fungal populations (Rau et al. 2005; Bogacki et al. 2010; Mclean et al. 2010; Liu et al. 2012).

Williams et al. (2001) described a PCR-based test that can differentiate the two forms of P. teres. For the identification of P. teres mating types without considering forms, two specific primer pairs were developed: MAT1 forward and reverse primers that generate an approximately 1300 bp product, and MAT2 forward and reverse primers that generate an approximately 1150 bp product (Rau et al. 2005). More recently, MAT-specific single nucleotide polymorphism (SNP) primers were developed for a PCR-based analysis, which can also discriminate the two forms by the amplification of distinct PCR products. These primers include PttMAT1F/R (1143 bp) and PttMAT2F/R (1421 bp) for NFNB MAT1 and MAT2 isolates, and PtmMAT1F/R (194 bp) and PtmMAT2F/R (939 bp) for SFNB MAT1 and MAT2 isolates, respectively (Lu et al. 2010).

Inconsistent results have been reported with respect to the frequencies of P. teres mating types in pathogen populations from different parts of the world (Liu et al. 2011). Rau et al. (2005) found that the two mating type genes occurred in equal frequencies in P. teres populations of both forms collected from the island of Sardinia, Italy, and concluded that sexual reproduction was the main source of primary inoculum. In contrast, Lehmensiek et al. (2010) observed a high level of genetic relatedness within Ptt and Ptm populations collected from the south-western Cape in South Africa and from across Australia, and suggested that asexual reproduction is predominant for both forms of the pathogen in those regions. The occurrence and frequencies of mating types of P. teres have not been studied in Canada. Therefore, this study was conducted to test the hypothesis that net blotch pathogen populations from the Canadian prairies are reproducing mainly by sexual reproduction.

Materials and methods

Isolate collection

A total of 124 barley fields were sampled in Alberta, Saskatchewan and Manitoba from 2009 to 2011. Leaves with symptoms of NFNB and/or SFNB were collected, placed in paper envelopes, air-dried at room temperature, and stored at 4°C. Leaf sections (about 10 mm × 5 mm) were surface-sterilized in 50% ethanol for 15 s, and in 2% sodium hypochlorite for 30 s, then rinsed with sterile water and placed on moistened filter paper in 9 cm-diameter plastic Petri dishes (Tekauz 1990). Dishes were put into an incubator at 20 ± 0.5°C with a 12 h photoperiod under fluorescent and near-ultraviolet light at 368 nm to induce pathogen sporulation. After 3–5 days, single conidia of P. teres were transferred onto 10% V-8 juice agar (V-8 juice, 100 mL; CaCO₃ 3 g; Difco agar 20 g; distilled water 900 mL) supplemented with 50 mg L⁻¹ kanamycin, and incubated as above to produce new colonies (Tekauz 1990). Single-spore isolations were made from at least one leaf showing symptoms of NFNB and/or SFNB from each of the 124 fields. For collections in 2011, two isolations were made from each infected plant, with an isolate obtained from an upper leaf (either the flag, penultimate or antepenultimate leaf) and another from the lower canopy. In some cases, single-spore isolations were made from different lesions on the same leaf or from different areas on a single lesion. The resulting isolates were initially identified as P. teres based on the morphology of the conidia and conidiophores (Mclean et al. 2009; Liu et al. 2011). Spore suspensions of single-spore isolates were prepared in 25% sterile glycerol, and placed in liquid nitrogen or at −80°C in an ultra-low temperature freezer for storage.

Genomic DNA extraction

Isolates chosen for genomic DNA extraction were removed from liquid nitrogen or a −80°C freezer, and grown on fresh 10% V-8 juice agar. Mycelia were then transferred from 14-day-old cultures into fresh potato dextrose broth containing 50 mg L⁻¹ kanamycin and 10 mg L⁻¹ streptomycin. The cultures were incubated for 10 days on a shaker rotating at 100 rpm at room temperature. Mycelia were harvested by transferring the cultures to plastic centrifuge tubes (50 mL), followed by centrifugation at 4000 rpm (Heraeus Megafuge 40 R, Thermo Scientific Inc.) for 5 min at 4°C. The supernatant was discarded and mycelia were washed three times by filling the tubes with distilled water, shaking thoroughly and centrifuging at 4000 rpm for 3 min at 4°C. The mycelial pellets were lyophilized and kept at −20°C (or −80°C for long-term storage) (Serenius et al. 2005). DNA was extracted from 20–30 mg of the lyophilized mycelium with a Wizard^® Genomic DNA Extraction Kit (Promega Corp, Madison, WI) following the manufacturer’s instructions. This was followed by two extractions with phenol:chloroform:isoamyl alcohol (25:24:1) (Abboukhaddour et al. 2011). To verify the quality and quantity of the extracted DNA, all samples were analysed with a NanoDrop® ND-1000 spectrophotometer (Thermo Scientific, Wilmington, DE). The extracted DNA was then diluted to a final concentration of 10 ng μL⁻¹ prior to PCR analysis.

Species-, form-, mating type- and mating type form-specific PCR analysis

Species-specific PCR analysis was performed in a reaction mixture containing 20 mM Tris-HCl (pH 8.4), 50 mM KCl, 0.25 mM of each dNTP, 2.5 mM MgCl_2, 1 U Taq polymerase, 10 pmol of each of the OPF01F900 and OPF01R900 primers (Table 1), and 20 ng template DNA (Williams et al. 2001). Reaction conditions consisted of an initial denaturation step at 94°C for 60 s, followed by 35 cycles of denaturation at 94°C, annealing at 57°C and extension at 72°C each for 30 s, and a final extension at 72°C for 7 min. DNA extracted from an isolate of Pyrenophora tritici-repentis was included as a negative control. Amplicons were resolved on a 1.5% agarose gel and visualized with SYBR Safe (Invitrogen, Carlsbad, CA) stain.

Table 1. Sequences of species-, form-, mating type- and mating type locus-specific primers used to characterize Pyrenophora teres f. teres and P. teres f. maculata populations from western Canada.

Primer name	Primer sequence	Amplicon size	Reference
OPF01F900 OPF01R900	ACGGATCCTGGGTCCAGA CGGATCCTGAGCAGAAAATG	~ 900 bp	Williams et al. (2001)
PTT-F PTT-R	CTCTGGCGAACCGTTC ATGATGGAAAAGTAATTTGTA	~ 380 bp	Williams et al. (2001)
PTM-F PTM-R	TGCTGAAGCGTAAGTTTC ATGATGGAAAAGTAATTTGTG	~ 410 bp	Williams et al. (2001)
MAT-1 forward MAT-1 reverse	AACAGACTCCTCTTGACAACCCG TGACGATGCATAGTTTGTAAGGGTC	~ 1300 bp	Rau et al. (2005)
MAT-2 forward MAT-2 reverse	CAACTTTTCTCTACCACACGTATCCC TGTGGCGAT GCATAGTTCGTAC	~ 1150 bp	Rau et al. (2005)
PttMAT1F PttMAT1R	ATGAGACGCTAGTTCAGAGTCT GATGCCCAGCCAAGGACAA	1143 bp	Lu et al. (2010)
PttMAT2F PttMAT2R	TACGTTGATGCAGCTTTCTCAAT AACACCGTCCAAAGCACCT	1421 bp	Lu et al. (2010)
PtmMAT1F PtmMAT1R	TGTTAGAGACCCCACCAGCGT CAGCTTTCTTGGCCTTCTGAA	194 bp	Lu et al. (2010)
PtmMAT2F PtmMAT2R	ACGCAAGGTACTCTGTACGCA GACGTCGAGGGAGTCCATTT	939 bp	Lu et al. (2010)

PCR amplification with form-specific primers to differentiate the two forms of P. teres as either Ptt or Ptm was optimized as a duplex PCR protocol. The reaction mixture was the same as for the species-specific PCR analysis, except that 10 pmol of each of the form-specific PTT-F, PTT-R, PTM-F and PTM-R primers (Table 1) was included (Williams et al. 2001). Reaction conditions consisted of an initial denaturation step at 94°C for 60 s, followed by 35 cycles at 94°C for 30 s, 53°C for 30 s and 72°C for 30 s, with a final extension of 72°C for 7 min. PCR products were resolved and visualized as above.

Mating type-specific PCR analysis was optimized and conducted for all P. teres isolates as a duplex reaction, as per Rau et al. (2005), to determine mating type (MAT-1 or MAT-2) regardless of form. The reaction mixture was as described for the species- and form-specific PCR assays, except that the mating type-specific MAT-1 forward, MAT-1 reverse, MAT-2 forward and MAT-2 reverse primers (Table 1) were substituted for the other primers. Reaction conditions consisted of an initial denaturation step at 94°C for 60 s, followed by 35 cycles at 94°C for 30 s, 55°C for 30 s, 72°C for 30 s and a final extension at 72°C for 7 min.

A mating type form-specific PCR assay was used to confirm form/type classification of the P. teres isolates (Lu et al. 2010). This assay can also discriminate between the two forms of the pathogen by the amplification of distinct PCR products. The primers PttMAT1F/R, PttMAT2F/R, PtmMAT1F/R and PtmMAT2F/R (Table 1) were included in the analysis to confirm the identification of Ptt MAT1, Ptt MAT2, Ptm MAT1 and Ptm MAT2 isolates, respectively. PCR assays were performed in a reaction mixture having the same composition as described for the species-specific PCR analysis, with thermal cycling conditions identical to those described by Lu et al. (2010). Amplification products were resolved on 2.25% agarose gels and visualized using SYBR Safe (Invitrogen).

Clone correction

Simple sequence repeat (SSR) polymorphisms were detected with 13 polymorphic SSR primers developed from the genome assembly of P. teres (Ellwood et al. 2010; Liu et al. 2012), in order to identify and eliminate repeated genotypes and uncover clonality (James et al. 2009) in the isolate collection. The PCR analysis was conducted as described by Liu et al. (2012), and the SSR-PCR products were separated by capillary electrophoresis on a 3730 DNA analyser (Applied Biosystems, Foster City, CA). For each locus, microsatellite allele sizes were determined by comparing the amplicons with a LIZ 500 internal size standard (Applied Biosystems) using GeneMapper software v3.7 (Applied Biosystems) (Bogacki et al. 2010).

Data analysis

Chi-square tests were conducted to determine if the observed mating type ratio for each of the populations of P. teres from each province and the cross province pool for western Canada departed significantly from the null hypothesis of a 1:1 MAT1:MAT2 ratio (Bogacki et al. 2010). Mating type frequencies were calculated using original and clone corrected data sets in which only one representative haplotype per clone was selected for further analysis (Sommerhalder et al. 2006; James et al. 2009). Chi-square values were calculated according to the formula: x² = ∑[(o-e)²/e], where: o is the observed value of the mating type and e is the expected value (Moore & Novak Frazer 2002; Serenius et al. 2005; Sommerhalder et al. 2006). A 0.05 Type I error rate was applied to accept or reject the null hypothesis of a statistically equal mating type ratio of 1:1 (Vail & Banniza 2009). Chi-square tests were not conducted on province-year populations with sample sizes of less than five because of a lack of statistical power (Sommerhalder et al. 2006). Alternatively, the pooled provincial populations were used to provide suitable sample sizes and the associated statistical power. For clone correction, similarity matrices representing all possible pairwise comparisons of the tested isolates were constructed based on the presence/absence data for each marker type. The similarity matrix was then used to perform cluster analysis by the unweighted pair-group method using arithmetic means (UPGMA) procedure. The software NTSYSpc version 2.2 (Exeter Software, New York, NY) was used to construct the similarity matrix and perform the UPGMA analysis (McLean et al. 2010).

Results

Pyrenophora teres f. teres was readily isolated from barley leaf lesions showing typical symptoms of the net form of net blotch. In contrast, both Cochliobolus sativus (Ito and Kuribayashi) Drechs. ex Dastur (anamorph Bipolaris sorokiniana (Sacc.) Shoemaker), the causal agent of spot blotch, and P. teres f. maculata were isolated from dark necrotic spots on green and senesced barley leaf tissue placed on wet filter paper. A total of 190 P. teres isolates were isolated from leaves exhibiting typical symptoms of NFNB or SFNB. Thirty isolates also were collected and identified from plants exhibiting intermediate symptoms that were not clearly distinguishable as NFNB or SFNB, but were identified as P. teres based on the morphological characteristics of the conidiophores and conidia. Thus, a total of 220 single-spore isolates were derived and used in the analyses.

Species- and form-specific analyses

Species-specific primers amplified a 900bp band in each of the 220 isolates and confirmed them to be P. teres. No amplicon was produced from DNA of P. tritici-repentis, which was included as a negative control. There were a total of 128 isolates identified as Ptt and 92 as Ptm, based on form-specific markers (Table 2). The Ptt form-specific primers PTT-F and PTT-R amplified a ~ 380 bp fragment, while the Ptm form-specific primers PTM-F and PTM-R generated a ~ 410 bp fragment from isolates morphologically designated as Ptt and Ptm, respectively. It was common to find both forms in the same field, the same plant, and on the same leaf in very close proximity. Of the 30 P. teres isolates recovered from lesions that were not clearly distinguishable as NFNB or SFNB, 19 were Ptt and 11 were Ptm.

Table 2. Form and mating type frequencies in western Canadian populations of Pyrenophora teres f. teres (Ptt) and P. teres f. maculata (Ptm) (original data set).

Province	Year	Total number of fields	Ptt1	Ptt2	Ptt χ^2 (1:1) types	P	Ptm1	Ptm2	Ptm χ^2 (1:1) types	P
Alberta	2010	20	19	10	2.793	0.095	11	12	0.043	0.835
Alberta	2011	11	13	13	0.000	1.000	3	1	–	–
Manitoba	2009	10	4	0	–	–	1	5	2.667	0.102
Manitoba	2010	10	0	2	–	–	6	5	0.091	0.763
Manitoba	2011	9	6	3	1.000	0.317	0	2	–	–
Saskatchewan	2010	22	5	5	0.000	1.000	7	5	0.333	0.564
Saskatchewan	2011	42	26	22	0.333	0.564	11	23	4.235	0.040*
Alberta	pooled^†	31	32	23	1.473	0.225	14	13	0.037	0.847
Manitoba	pooled^†	29	10	5	1.667	0.197	7	12	1.316	0.251
Saskatchewan	pooled^†	64	31	27	0.276	0.599	18	28	2.174	0.140
W Canada	pooled^†	124	73	55	2.531	0.112	39	53	2.130	0.144

* Denotes mating-type frequencies that are significantly different from a 1:1 ratio at P = 0.05.

– Indicates that χ² analysis was not performed due to small sample size.

^† Pooled data for each province are for the combined years; for the western Canada (W Canada) entry, pooled data include all years and provinces.

In all cases, morphological-based species identification of the isolates was confirmed by PCR analysis. A separate analysis was done to identify repeated genotypes among the isolates. This analysis revealed a clonal fraction of approximately 10% in the entire population. There were 198 distinct genotypes consisting of 115 Ptt and 83 Ptm isolates (Table 3). These included 49 Ptt and 43 Ptm isolates from Saskatchewan, 51 Ptt and 21 Ptm isolates from Alberta, and 15 Ptt and 19 Ptm isolates from Manitoba, respectively.

Table 3. Form and mating type frequencies in western Canadian populations of Pyrenophora teres f. teres (Ptt) and P. teres f. maculata (Ptm) (clone corrected data set).

Province	Year	Total number of fields	Ptt1	Ptt2	Ptt χ^2 (1:1) types	P	Ptm1	Ptm2	Ptm χ^2 (1:1) types	P
Alberta	2010	20	16	10	1.385	0.239	9	8	0.059	0.808
Alberta	2011	11	12	13	0.040	0.841	3	1	–	–
Manitoba	2009	10	4	0	–	–	1	5	2.667	0.102
Manitoba	2010	10	0	2	–	–	6	5	0.091	0.763
Manitoba	2011	9	6	3	1.000	0.317	0	2	–	–
Saskatchewan	2010	22	5	3	0.500	0.480	7	5	0.333	0.564
Saskatchewan	2011	42	25	16	1.976	0.1599	11	20	2.613	0.106
Alberta	pooled^†	31	28	23	0.490	0.484	12	9	0.429	0.513
Manitoba	pooled^†	29	10	5	1.667	0.197	7	12	1.316	0.251
Saskatchewan	pooled^†	64	30	19	2.469	0.1161	18	25	1.140	0.286
W Canada	pooled^†	124	68	47	3.835	0.0502	37	46	0.976	0.323

* Denotes form and mating-type frequencies that are significantly different from a 1:1 ratio at P = 0.05.

– Indicates that χ² analysis was not performed due to small sample size.

^† Pooled data for each province are for the combined years; for the western Canada (W Canada) entry, pooled data include all years and provinces.

Cluster analysis of the SSR data was done using the UPGMA procedure and Jaccard’s similarity coefficient. These results indicated that all isolates, including the 30 recovered from the indistinct lesions, clustered in two distinct divergent groups conforming to either Ptt or Ptm. There was no intermediate clade detected between the two forms.

Mating type-specific PCR analysis

Among the 128 Ptt isolates, 73 produced a single 1300 bp amplicon corresponding to MAT-1, while 55 isolates produced a single 1150 bp amplicon corresponding to the MAT-2 idiomorph (Table 2). Among the 92 Ptm isolates, 39 produced the amplicon corresponding to MAT-1, while 53 produced the amplicon corresponding to the MAT-2 idiomorph (Table 2). Using the clone correction step, 30 Ptt isolates from Saskatchewan were identified as MAT1 and 19 as MAT2, 28 Ptt isolates from Alberta were identified as MAT1 and 23 as MAT2, and 10 isolates from Manitoba were identified as MAT1 and five were MAT2 (Table 3). Similarly, 18 Ptm isolates from Saskatchewan were identified as MAT1 and 25 as MAT2. Twelve Ptm isolates from Alberta were identified as MAT1 and nine as MAT2, and seven isolates from Manitoba were identified as MAT1 and 12 as MAT 2 (Table 3). MAT1 and MAT2 idiomorphs of both forms were identified within the same crop, on the same plant, and on the same leaf. In three instances, both mating types were identified within a single lesion caused by Ptt; however, only one mating type was found within each single lesion caused by Ptm in all cases.

Chi-square analysis was conducted to test the null hypothesis of a 1:1 ratio between the two mating types for both Ptt and Ptm. With the exception of the 2011 Ptm population from Saskatchewan, no significant departure from a 1:1 mating type ratio was observed in any of the populations of Ptt or Ptm in the original data set (Table 2). Chi-square analysis of the data set following the clone correction step revealed no statistical differences in mating type frequency for either Ptt or Ptm in all populations (Table 3).

Form and mating type classifications using species-, form- and mating type-specific PCR assays were confirmed by analysis with MAT-specific SNP primers. In isolates originally classified as Ptt and MAT1, the primer set PttMAT1 F/R amplified a 1143 bp fragment, while for those classified as Ptt and MAT2, the primer set PttMAT2 F/R amplified a 1421 bp fragment. Similarly, in isolates originally classified as Ptm and MAT1, the primer set PtmMAT1 F/R amplified a 194 bp band, while for those classified as Ptm and MAT2, the primer set PtmMAT2 F/R amplified a 939 bp band. These results confirm the form and mating type designations using the species-, form- and mating type-specific PCR markers for all isolates.

Discussion

This is the first study of mating type frequencies in western Canadian P. teres f. teres and P. teres f. maculata populations. Given the 1:1 ratio of the two mating types for both forms of the pathogen, and the small number of clones identified in the isolate collections, it appears that Ptt and Ptm go through regular cycles of sexual reproduction in the Canadian prairie provinces. Overall, these results concur with the hypothesis that primary infection of barley fields is likely caused by ascospores discharged from pseudothecia on stubble.

The occurrence and frequency of mating types was assessed in a collection of P. teres isolates from 124 barley fields across Alberta, Saskatchewan and Manitoba. Analysis of both the original and clone corrected data sets revealed no significant departure, with the exception of Saskatchewan in 2011, from the null hypothesis of a 1:1 ratio between the mating types for either Ptt or Ptm in any of the three Prairie provinces or across the entire prairie region. This finding implies that there is no selection for mating types in either the Ptt or Ptm populations, supporting the likelihood of sexual hybridization within each form. Serenius et al. (2005) and Liu et al. (2012) also demonstrated the existence of both mating types in a 1:1 ratio in Finland and North Dakota, respectively, and concluded that sexual recombination is common in those regions. A ratio of almost 1:1 between mating types also has been reported in Ptm and Ptt populations collected from barley fields in Australia (Bogacki et al. 2010; Mclean et al. 2010), indicating that sexual recombination is possible.

A two-step process was conducted to ensure the accuracy of the mating type data presented in this report. The mating type of individual isolates was first assessed by PCR analysis with the mating type-specific primers of Rau et al. (2005), and then confirmed with a mating type locus-specific PCR assay developed by Lu et al. (2010). The data also were clone corrected to avoid over-representation of clonal isolates (Sommerhalder et al. 2006; James et al. 2009). The identification of both mating types in both forms of P. teres in a statistically equal 1:1 ratio in all tested populations suggests that both Ptt and Ptm are capable of sexually reproducing. Therefore, it can be concluded that ascospore dispersal is likely to have a key role in the initiation of both NFNB and SFNB in each of the prairie provinces, likely followed by several cycles of conidia production, with the conidia serving as secondary inoculum during the growing season. Results from this study are in agreement with the earlier findings of Piening (1961, 1968), and Duczek et al. (1999), who identified the P. teres teleomorph in Alberta and Saskatchewan, respectively.

The two mating types of both Ptt and Ptm were occasionally recovered from the same leaf, different leaves of the same plant, and different plants in the same barley field, indicating that the mating types occurred in close proximity to each other. However, while both mating types could occasionally be recovered from single lesions caused by Ptt, single lesions caused by Ptm typically consisted of only one mating type. This may reflect the fact that what may appear to be single net form lesions are not necessarily caused by single-spores of Ptt. Instead, two or more spores of different mating types may initially cause small lesions in close proximity, which would then coalesce to form larger lesions with no distinct boundary. Such lesions may be considered a single lesion at the time of single-spore isolation. In contrast, spot form lesions can be easily distinguished from one another under a dissecting microscope.

All isolates derived from leaves with symptoms that were not clearly distinguishable as NFNB or SFNB were confirmed as either Ptt or Ptm by PCR analysis. Cluster analysis using the UPGMA procedure revealed that all isolates, including those from indistinct lesions, clustered in two distinct Ptt and Ptm groups with no intermediate clade. Rau et al. (2003, 2007) also demonstrated that Ptt and Ptm isolates cluster in two genetically distinct clades and did not find any genetically intermediate isolates in an analysis of amplified fragment length polymorphisms, suggesting that Ptt and Ptm are genetically isolated. Furthermore, Serenius et al. (2005) showed that meiosis did not occur properly in crosses between isolates of Ptm and Ptt, resulting in the production of ascocarps that lacked asci or contained abnormal, non-culturable ascospores. This finding led Serenius et al. (2005) to conclude that successful sexual reproduction between the two forms of P. teres is very unlikely in nature. In contrast, Campbell et al. (1999) reported an efficient crossing of the two forms under laboratory conditions and demonstrated that most of the sexual offspring caused intermediate symptoms on barley leaves. The resulting sexual offspring were later shown to be genetically stable (Campbell & Crous 2003). Moreover, Campbell et al. (2002) concluded, based on a random amplified polymorphic DNA analysis, that sexual recombination between Ptt and Ptm isolates might be occurring under field conditions. Similarly, Leisova et al. (2005) suggested that hybridization between the two forms was possible based on the presence of an intermediate clade consisting of haplotypes with shared markers. These apparently contradictory results might reflect differences in the isolates studied, population structure and (or) environmental conditions in different geographical regions and cropping systems. Nonetheless, the results from the current study suggest that no case of successful hybridization between Ptt and Ptm was detected among the 220 studied isolates collected from western Canada.

Acknowledgements

The authors would like to thank Mr Meconnen Beyene, Ms Noryne Rauhala, and Ms Jackie Busaan (Agriculture and Agri-Food Canada) for assistance with the collection of isolates, and Ms Ileana S. Strelkov, Ms Kelley Dunfield, Dr Reem Aboukhaddour and Dr Tiesen Cao (University of Alberta) for technical advice and insight. The constructive suggestions of two anonymous reviewers also are gratefully acknowledged.

Additional information

Funding

The authors thank the Western Grains Research Foundation, Alberta Barley Commission, Rahr Malting Inc., Canadian Wheat Board, Agriculture and Agri-Food Canada and the University of Alberta for financial support of this research.