Fungal Genetics and Biology 40 (2003) 115–125 www.elsevier.com/locate/yfgbi Further evidence for sexual reproduction in Rhynchosporium secalis based on distribution and frequency of mating-type alleles Celeste C. Linde,* Marcello Zala, Sara Ceccarelli, and Bruce A. McDonald Institute of Plant Sciences, Plant Pathology Group, Federal Institute of Technology, ETH-Zentrum, LFW, CH-8092 Z€urich, Switzerland Received 19 March 2003; accepted 25 July 2003 Abstract Rhynchosporium secalis, the causal agent of scald on barley, is thought to be exclusively asexual because no teleomorph has been found. Partial sequences of the HMG-box and a-domain of Rhynchosporium secalis isolates were identified and used to develop a PCR assay for the mating-type locus. PCR amplification of only one of these two domains was possible in each strain, suggesting that R. secalis has a MAT organization that is similar to other known heterothallic fungi. A multiplex PCR with primers amplifying either a MAT1-1- or MAT1-2-specific amplicon was used to determine the distribution of mating types in several R. secalis populations. In total, 1101 isolates from Australia, Switzerland, Ethiopia, Scandinavia, California, and South Africa were included in the analysis. Mating types occurred in equal frequencies for most of these populations, suggesting frequency-dependent selection consistent with sexual reproduction. In addition, both mating types were frequently found occupying the same lesion or leaf, providing opportunities for isolates of opposite mating type to interact and reproduce sexually. We propose that R. secalis should be considered a sexual pathogen, although the sexual cycle may occur infrequently in some populations. Ó 2003 Elsevier Inc. All rights reserved. Keywords: Rhynchosporium secalis; Mating-type gene; a-Domain; HMG-domain; MAT locus; Mating type frequency; Sexual reproduction 1. Introduction Rhynchosporium secalis is a haploid imperfect fungus (deuteromycete) that causes scald on barley and various other Gramineae hosts (Caldwell, 1937; Shipton et al., 1974). The disease is present in all barley-growing areas and is particularly damaging in temperate regions with cool, moist winters. In regions of intensive barley production where environmental conditions are conducive for disease, severe attacks of scald have caused yield losses as high as 35–40% (James et al., 1968; Shipton et al., 1974), although losses of 1–10% are more common (Shipton et al., 1974). The seed-borne nature of the pathogen (Shipton et al., 1974) allows it to spread to new locations via man-mediated gene flow. In a field, short distance dispersal (up to a few meters) is achieved by splash dispersal of conidia (McDonald et al., 1999; Shipton et al., 1974). Although a teleomorph has never * Corresponding author. Fax: +41-1-632-15-72. E-mail address: celeste.linde@ipw.agrl.ethz.ch (C.C. Linde). 1087-1845/$ - see front matter Ó 2003 Elsevier Inc. All rights reserved. doi:10.1016/S1087-1845(03)00110-5 been identified, high pathogenic variation (e.g., McDermott et al., 1989) and high levels of genetic variation for neutral markers (McDermott et al., 1989; McDonald et al., 1999; Salamati et al., 2000) are consistent with the presence of a teleomorph, most likely to be a discomycete (Goodwin, 2002). The closest known relative was shown to be Tapesia yallundae (Goodwin, 2002) which is heterothallic. An important question regarding R. secalis is whether it has the potential for sexual recombination as suggested by allele frequencies and lack of multilocus associations observed with RFLP and isozyme markers (Goodwin et al., 1993; McDonald et al., 1999; Salamati et al., 2000). The MAT genes of several heterothallic ascomycetes have been cloned and identified recently (reviewed in P€ oggeler, 2001). Heterothallic ascomycetes require contributions from two genetically distinct parents to reproduce sexually. This process is controlled by a single locus referred to as mating type (MAT) (Coppin et al., 1997; Turgeon, 1998), which has two different idiomorphs. Although the identification of MAT loci 116 C.C. Linde et al. / Fungal Genetics and Biology 40 (2003) 115–125 has become nearly routine for ascomycetes, it was only recently that mating types could also be identified in asexual fungi using PCR-based approaches (Arie et al., 1997). So far, no teleomorph has been identified for R. secalis, therefore determination of mating type using traditional mating studies with tester strains was not possible. Instead, we took the approach of identifying the HMG- and a-boxes present in R. secalis using a PCR-based approach based on primers designed from a close relative, T. yallundae. The presence of MAT genes is not sufficient to prove the existence of a asexual stage or to confirm random mating. However, if frequency-dependent selection is operating on MAT genes (May et al., 1999), then mating types should occur in equal frequencies (Milgroom, 1996). Many studies have attempted to determine mating type frequencies for various ascomycetes in the past, with findings consistent with random mating based on equal mating type frequencies in Tapesia acuformis and T. yallundae (Douhan et al., 2002), Mycosphaerella graminicola (Zhan et al., 2002), Ascochyta rabiei (Armstrong et al., 2001), some Cryphonectria parasitica populations (Liu et al., 1996), and Cochliobolus carbonum (Welz and Leonard, 1995). Skewed mating type ratios consistent with lack of sexual reproduction or differences in fertility between mating types were found for Magnaporthe grisea populations (Dayakar et al., 2000; Mekwatanakarn et al., 1999; Notteghem and Silue, 1992; Viji and Uddin, 2002) and Ascochyta fabae f. sp. lentis in Canada (Ahmed et al., 1996). Asexual (mitosporic) fungi also have MAT genes as identified for Bipolaris sacchari (Sharon et al., 1996b), as well as Fusarium oxysporum and Alternaria alternata (Arie et al., 2000; Sharon et al., 1996b). Organization of mating-type idiomorphs in asexual fungi is similar to that of sexually reproducing ascomycetes (Arie et al., 2000). But the strains of asexual fungi do not interact to reproduce sexually. It has been shown that asexuality in these fungi results from a loss of function in genes other than the MAT loci, as MAT1-2 of B. sacchari (Sharon et al., 1996a) and MAT genes of A. alternata (Arie et al., 2000) were confirmed to be functional in a close sexual relative, Cochliobolus heterostrophus, by heterologous expression. The MAT-idiomorphs lack significant sequence similarity (Turgeon, 1998), except for the flanking regions (Coppin et al., 1997). In ascomycetes, MAT idiomorphs encode for proteins with confirmed or putative DNAbinding motifs, i.e., MAT1-2 encodes a high mobility group (HMG) protein and MAT1-1 isolates encode an a-domain protein (Kronstad and Staben, 1997; Turgeon, 1998). In homothallic species, each isolate carries both MAT1-1 and MAT1-2 genes, usually closely linked or fused (Yun et al., 2000). In this report, strains carrying the idiomorph which encodes a protein with an a-box motif are designated MAT1-1, and strains carry- ing the idiomorph that contains the HMG-motif are designated MAT1-2, following nomenclature suggested by Turgeon and Yoder (Turgeon and Yoder, 2000). The objectives of this research were (i) to partly identify the MAT loci in R. secalis. This was to enable us to design primers for rapid identification of R. secalis mating types, and (ii) to determine the mating type frequencies of several R. secalis populations across spatial scales ranging from field plots to continents. Although equal mating type frequencies can be obtained by chance, equal frequencies across many spatial scales indicates the occurrence of sexual reproduction due to frequency-dependent selection. Equal mating type frequencies would provide evidence additional to the index of association tests conducted previously (Salamati et al., 2000) showing that RFLP alleles in these populations are in gametic equilibrium, consistent with the occurrence of sexual reproduction in R. secalis. 2. Materials and methods 2.1. Fungal isolation and DNA extraction Isolation of small mycelial tufts from lesion borders was as described previously (McDonald et al., 1999). For the Californian, Scandinavian, and Australian populations, DNA was extracted as described previously (McDonald et al., 1999), but for populations from Switzerland, Ethiopia, and South Africa, DNA was extracted with the DNeasy Plant Mini DNA extraction kit (Qiagen GmbH, Hilden, Germany) according to the specifications of the manufacturer. 2.2. Population descriptions For clone-corrected data sets, barley leaves were collected from 21 geographic locations in six countries representing four continents. Leaves were collected from naturally infected fields at a single point in time. The biggest collection was made from Australia (Aus) in 1996 where seven field populations were sampled (McDonald et al., 1999). The infected leaves of collection 1 were sampled along four parallel transects from cultivar Clipper planted at the Wagga Wagga Agricultural Institute, New South Wales. One leaf was collected at 30 cm intervals along 6 m transects which were separated by 18 cm. Collection 2 was made from the same plot as collection 1, but 30 days later near the end of the growing season. The total geographic area covered in this field was
4.3 m2 for collections 1 and 2. The remaining Australian collections, except collections 4 and 5, were taken from farmerÕs fields using a larger scale six-site hierarchical sampling strategy as described previously (McDonald et al., 1999). Hierarchical collections covered a total geographic area of 200 m2 , although only C.C. Linde et al. / Fungal Genetics and Biology 40 (2003) 115–125
6 m2 was represented by the collection of leaves taken from each field. Collection 3 was made from cultivar OÕConner grown at Methul, New South Wales. Collection 4 was from barleygrass bordering the OÕConnor field of collection 3 covering an area of
15 m2 . This collection was unordered because very few infected leaves were found. Collection 5 was from a
200 m transect of bromegrass and barleygrass alongside a field of cultivar Schooner at Rannock, New South Wales, covering an area of
200 m2 , though only
10 m2 was represented by the collection of leaves. Collection 6 was from cultivar Arapolis planted at Horsham, Victoria. Collection 7 was from cultivar Stirling planted at Quindanning near Perth, Western Australia. This was the only collection from Australia where seed was treated with triadimenol. In Switzerland (CH), infected leaves were collected from six fields. Collection 1 was from an unknown barley variety near Leuggern, Canton Aargau, collection 2 from a two-component mixture of unknown varieties of two-row barley, near Lucens, Canton Vaud, and collection 3 was from an unknown commercial two-row variety planted near Cugy, Canton Vaud. These collections were eight-site hierarchical collections covering a total geographic area of
300 m2 for each collection, although only
8 m2 was represented by the collection of leaves. Collection 4 originated from experimental barley plots near the Reckenholz experimental station, Canton Z€ urich. This collection was an eight-site unordered hierarchy, with each site representing a different field plot measuring 1.5 m  3 m, and 15 leaves taken equally spaced from each of these plots. The total geographic area covered by this collection was
800 m2 , although only
36 m2 was represented by the collection of leaves. Collection 5 was from cultivar Baretta planted at Eschikon, Canton Z€ urich, along two parallel transects separated by
2 m, covering a total geographic area of 100 m2 . Collection 6 was from an unknown variety collected from a bio-farm near Dietikon, Canton Z€ urich. Collection 6 was based on two parallel transects separated by
3 m, covering a total geographic area of
120 m2 . Distances between collections from Switzerland ranged from 20 to 200 km. The collections from Ethiopia (ET) were all collected in 2002 from the highlands of Ethiopia. Collections 1 and 2 were from Edagahamus and Atsbi in the southern part of the Tigrai region, and collections 3 and 4 from Adineba and Zata in the northern Tigrai region. Distances between these fields ranged from 20 to 200 km. Collections were made from unknown local varieties, in an eight-site hierarchy, each covering a total geographic area of
300 m2 , although only
8 m2 was represented by the collection of leaves. Collections from Norway were both from cultivar Tyra, collection 1 from southeastern Norway planted near Buskerud, and collection 2 from central Norway near Størdal. The collection from 117 Finland was from the cultivar Kymppi near Joikioinen. All Scandinavian collections were hierarchical, covering total geographic areas of
400,
18, and
400 m2 , respectively (Salamati et al., 2000). The collection from California was a representative sub-sample of a large, single field collection made in 1988 and characterized previously (McDermott et al., 1989). This collection was from Composite Cross II and CCV in Davis, California, covering an area of
200 m2 . However, the crop area represented in this population was much less (
50 m2 ), as only a sub-sample of the original collection was analyzed in this study. Leaves were also collected from two locations in South Africa. SA1 was an unordered collection from an unknown cultivar near Bredasdorp from a 15 m2 area. SA2 was collected along two 20 m transects separated by 1 m from different F2 crosses at the Langewens experimental station near Malmesbury. 2.3. Amplification of the HMG-box and a-domain by PCR PCR primers designed from the HMG-box (TyHMG 3131F and TyHMG 3668R) in MAT1-2 isolates, and the a-domain (TyMat1 348F and TyMat1 1400R, primer sequences in Table 1) in MAT1-1 isolates of T. yallundae (idiomorph sequences provided by Dr. P. Dyer, Nottingham University, UK) were used in attempts to amplify the respective regions in R. secalis on a subset of 24 isolates of R. secalis from California, Scandinavia, Australia, and Switzerland. Annealing temperatures for all possible primer combinations were determined with a gradient cycler (Biometra). Consequently, annealing temperatures of 52 °C were used to amplify the highmobility group (HMG) region present in all ascomycete MAT loci (Turgeon, 1998). Each PCR mixture contained 5–10 ng of genomic DNA in 20 ll reaction volumes containing 10 pmol of each primer, 0.1 mM dNTPs (dATP, dCTP, dTTP, and dGTP), 2 ll of 10 reaction buffer, and 0.5 U Taq polymerase (Amersham Pharmacia Biotech). For analyses of PCR products, 10 ll of each reaction was subjected to agarose gel (1%) electrophoresis in 0.5 TBE buffer, and stained with ethidium bromide. 2.4. Amplification of HMG-box flanks by thermal asymmetric interlaced PCR Thermal asymmetric interlaced PCR (TAIL-PCR) with degenerate primers AD1, AD2, AD3, and AD4 (Liu and Whittier, 1995) and nested primers designed within the HMG-box of R. secalis (Table 1), was performed to obtain upstream and/or downstream sequences of the HMG-box. TAIL-PCR was used to extend the already identified sequence of the HMG-box (
400 bp) obtained with the T. yallundae MAT1-2 118 C.C. Linde et al. / Fungal Genetics and Biology 40 (2003) 115–125 Table 1 Primer sequences used to identify the HMG-box and a-domain of Rhynchosporium secalis Primer Sequence Notes TyHMG 3131F TyHMG 3668R RsHMG 356R RsHMG 212R RsHMG 336F RsHMG 275F RsHMG 297R TyMAT1 348F TyMAT1 1400R RsMAT1F RsMAT1R RsMAT2F RsMAT2R TTCTCTGCTCCCGAGGATGC CAGGTTGTGGTTGACGATGTC GTATTGCTCTCTGATCTCGG GTGGTCGATGTTGGAATAGATG CCGAGATCAGAGAGCAATAC GTACAACGTCGATGATGCCTCCG CGGAGGCATCATCGACGTTGTAC GAGCGAGAAATCCATCCTG GTTACAGCGATGACTCCAGCG CCAACGATCCGCCTAGCCACA CATCGACAAGTAGCCACAGCT CATCTATTCCAACATCGACCAC CTCCTTGTATTGCTCTCTGATC Initial amplification of HMG-box Initial amplification of HMG-box Primary TAIL-PCR Tertiary TAIL-PCR Tertiary TAIL-PCR Secondary TAIL-PCR Secondary TAIL-PCR Identification of a-domain Identification of a-domain Multiplex Multiplex Primary TAIL-PCR, Multiplex Multiplex HMG-box primers, as described by Arie et al. (2000). PCR conditions for TAIL-PCR were as described by Liu and Whittier (1995), and primer sequences used for the primary, secondary and tertiary PCRs are given in Table 1. 2.5. Sequencing of PCR products PCR products were extracted from gel slices with the QIAquick Gel Extraction Kit (Qiagen, GmbH, Hilden, Germany), or sequenced directly if only one amplicon was present. Sequencing was performed on a ABI 3100 cycler using Taq-Cycle automated sequencing with DyeDeoxy Terminators (BigDye Terminator v3.0 Cycle Sequencing Ready Reaction; Applied Biosystems) with both primers (except degenerate primers) to ensure reliability of the sequence data. Sequences were assembled and aligned with Sequencher 4.1 (Gene Codes Corporation, Ann Arbor, Michigan). Sequence data from the extreme primer ends of the products were omitted due to possible errors resulting from misincorporation of basepairs. BLAST searches (Altschul et al., 1997) and BLAST2 (Tatusova and Madden, 1999) were done against the NCBI/GenBank databases. 2.6. Multiplex PCR Primers were designed from the sequences of the HMG-box and a-domain of R. secalis. Primer pair RsMAT1F and RsMAT1R amplified a
590 bp fragment in MAT1-1 isolates. Primer pair RsMAT2F and RsMAT2R amplified a
360 bp fragment in MAT1-2 isolates. To confirm that these primers amplified the mating-type idiomorphs of R. secalis, 60 randomly selected isolates from California, Scandinavia, Australia, and Switzerland were amplified with the matingtype-specific primers. In all cases, only products resembling either the MAT1-1- or MAT1-2-specific bands were obtained. These primer combinations were also tested on two isolates of T. yallundae (DNA obtained from T.D. Murray, Washington State University, USA). A mixture of all four primers was used in a multiplex PCR to generate either the
590 or
360 bp fragment in all isolates. All PCRs were conducted in 20 ll reaction volumes containing 10 pmol of each primer, 5–20 ng genomic DNA as template, 0.1 mM dNTPs (dATP, dCTP, dTTP, and dGTP), 2 ll of 10 reaction buffer, and 0.5 U Taq polymerase (Amersham Pharmacia Biotech). PCR was always performed on a Biometra thermocycler with the following conditions: an initial hold at 96 °C for 2 min, 40 cycles of 96 °C for 1 min, 58 °C for 1 min, 72 °C for 1 min; and a final hold at 72 °C for 5 min. Products (7 ll of reaction mixture) were separated in 1% agarose gels, 0.5 TBE, and stained with ethidium bromide. All gel images were captured digitally. 2.7. Microgeographical distribution of mating types (within lesions and leaves) Spatial distribution of R. secalis mating types within leaves and lesions were determined on heavily infected barley leaves from the South African collections SA1 and SA2. One strain was isolated from each of three to five lesions per leaf from SA1. Similarly, one strain was isolated from each of two to seven lesions per leaf from SA2. Two strains were isolated from one of the lesions from each of the six leaves. In addition, multiple isolations from 12 lesions were made for the Australian population, and two to three strains were isolated from each of 50 leaves. 2.8. Macrogeographical distribution of mating types (within fields and regions) The mating type frequencies of a much larger collection of isolates (N ¼ 1101) representing four continents (Africa, Europe, North America, and Australia) and 23 field populations (seven from Australia, six from Switzerland, four from Ethiopia, three from Scandinavia, C.C. Linde et al. / Fungal Genetics and Biology 40 (2003) 115–125 one from California, and two from South Africa), was determined using a multiplex PCR. 2.9. Clone correction The fungal data set was clone-corrected for each population using RFLPs. Clones were identified with seven single-locus probes and one multicopy probe, pRS 26 (McDonald et al., 1999). In the case of the Swiss and Ethiopian populations, the hybridization pattern obtained with the pRS 26 DNA fingerprint probe was too simple to be used as a fingerprint (data not shown). Instead, an Operon RAPD primer, OPL11 (Qiagen Operon Technologies) was used with PCR conditions as described in Liebhard et al. (2002) to obtain a fingerprint pattern that was combined with the data from the individual RFLP loci. R. secalis isolates having the same multilocus haplotype and DNA fingerprint were considered individual members of the same genotype (clone) resulting from asexual reproduction. Only one isolate from each clone was included in the analysis of mating type frequency. Isolates from South Africa were not analyzed using RFLPs or RAPDs. Instead, clones were distinguished using rep-PCR with primers ERIC2F and BOXA1R (Versalovic et al., 1991, 1994). PCR mixtures for rep-PCR were similar to that of the multiplex reactions except with using the rep-PCR primers as stated above. PCR conditions were; an initial hold at 96 °C for 2 min, then 35 cycles of 94 °C for 30 s, 52 °C for 1 min, 65 °C for 5 min; and then a final hold at 65 °C for 8 min. Isolates having the same multilocus rep-PCR patterns were considered clones. 2.10. Data analyses Isolates were pooled across different spatial scales in a hierarchy to determine mating type distribution within lesions, within leaves, within field plots, within field populations, and within country or region. To avoid bias due to the repeated sampling of the same clone, analyses were performed on a clone-corrected data set only. We tested whether the proportion of MAT1-1 and MAT1-2 isolates were equal in the populations studied. A v2 test (Everitt, 1977) was used to determine whether departures from a 1:1 frequency were significant. Contingency v2 analyses were performed on subpopulations within a country (Everitt, 1977). This analysis determines whether individual subpopulations (different fields in this case) within a larger population (country or region in this case) differ in mating type frequency. In addition, an exact binomial test for goodness-of-fit was performed to determine whether observed mating type frequencies within populations fit an empirical 1:1 ratio (null hypothesis). When sample sizes are small, the exact binomial test is more accurate than a v2 test (Sokal and Rohlf, 1995). 119 3. Results 3.1. Amplification of the HMG-box and a-domain by PCR The primer pair TyHMG 3131F and TyHMG 3668R did not amplify any fragment of the expected size (
540 bp) in any of the 24 isolates used. In three of the isolates a
850 bp product, and in another three isolates a
300 bp product was amplified. These products were sequenced for isolate RS99CH5D6a (
850 bp product) and RS99CH5A7b (
400 bp product). The sequence of the small fragment did not show any similarity with known HMG-boxes with a BLAST search. The larger fragment (only 400 bp of reliable sequence was obtained) showed high similarity with the MAT1-2 HMG-domain of T. yallundae (97%) and M. graminicola (92%). TAILPCR was then used to extend this sequence. The tertiary sequenced TAIL-PCR product of RsHMG 212R and AD4 of
600 bp aligned to the existing
400 bp HMGdomain and extended this sequence to 679 bp (Accession No. AY257473). The sequences between basepairs 337– 659 and 89–202 of R. secalis showed 80% similarity (E value ¼ 7  1063 ) and 79% similarity (7  108 ), respectively to basepair regions 3230–3551 and 2972–3087 of the MAT1-2 HMG-domain of T. yallundae. To identify MAT1-1 isolates, primer pair TyMAT1 348F and TyMAT1 1400R amplified a
1020 bp product of expected size in seven of the 24 isolates. None of these seven isolates also gave an amplification product with the TyHMG primers. The
1050 bp fragment of isolate R14 was sequenced, and a 858 bp reliable sequence was obtained (Accession No. AY257472). The sequence between bases 1–155 showed 80% similarity (5  1024 ) to the sequence between bases 432 and 590 of the T. yallundae MAT1-1 a-domain. Also bases 207–857 showed 81% similarity (1  10155 ) to bases 645–1298 of the a-domain of T. yallundae. 3.2. Multiplex PCR Mating types of R. secalis were readily identified in a multiplex PCR following amplification of a
590 bp MAT1-1-specific product (part of the a-domain) and a
360 bp MAT1-2-specific product (part of the HMGbox) (Fig. 1). A total of 1101 isolates were assayed for mating type. Of these, 516 produced a MAT1-1-specific fragment, and 496 isolates produced a MAT1-2-specific fragment. One isolate produced a
590 bp MAT1-1specific product and a
300 bp product. No amplification product could be obtained for 64 isolates (5.8%) (Table 2). It appears that DNA concentrations were too low to perform PCR as amplification was also not successful with rep-PCR primers for a tested subset of 20 of these isolates. Twenty-four isolates (2.2%) gave two equally strong bands representing MAT1-1 and 120 C.C. Linde et al. / Fungal Genetics and Biology 40 (2003) 115–125 different genotypes. Mixed cultures can be explained by the way we isolate R. secalis from the leaves. As we collect small tufts of mycelium from lesion edges, there is a possibility of isolating two different genotypes occupying the same area on the lesion. This and previous studies showed that it is common to find more than one genotype within a scald lesion (Salamati et al., 2000). RFLP analyses showed that
5% of isolations made using mycelial tufts were mixed cultures (Salamati et al., 2000). A faint MAT1-2 product was obtained in one of the two T. yallundae isolates amplified with MAT1-2 specific primers (RsMAT2F/RsMAT2R), but no product was amplified with the MAT1-1 specific primer pair (RsMAT1F/RsMAT1R), or when all four primers were used together in a multiplex (Fig. 1). Fig. 1. PCR amplification of two isolates each of Tapesia yallundae (lanes 1 and 2) and Rhynchosporium secalis (lanes 5 and 6) with primers RsMAT1F/R, and Tapesia yallundae (lanes 3 and 4) and Rhynchosporium secalis (lanes 7 and 8) with MAT1-2 specific primers, RsMAT2F/R. Lanes 9–12 show a multiplex PCR with primers RsMAT1F/R and RsMAT2F/R on Tapesia yallundae (lanes 9 and 10) and Rhynchosporium secalis (lanes 11 and 12). First and last lanes are a 100 bp size ladder. 3.3. Microgeographical distribution of mating types (within leaves, lesions, and field plots) Table 2 Summary of mating type frequencies in a global collection of Rhynchosporium secalis CR products Number of isolatesa Percentage MAT1-1 MAT1-2 MAT1-1 plus a
300 bp product Both MAT1-1 and MAT1-2 products No product Total 516 496 1 24 64 1101 46.9 45.1 0.0 2.2 5.8 a A fine scale investigation of R. secalis on barley leaves from South Africa revealed that both mating types could be isolated from three of six lesions where two strains per lesion were tested. In the Australian populations we isolated different mating types in three out of 12 lesions tested. Six of the 12 lesions were occupied by different genotypes as differentiated by RFLP markers. In 30% (15 out of 50) of the cases tested, both mating types were isolated from the same leaf in the Australian populations. Sixty-six percent (33 out of 50) of the leaves were occupied by different genotypes. Both mating types were isolated from three of the six leaves analyzed from the South African populations. In field plots with at least eight distinct genotypes, spatial scales ranging from 1 to 4.5 m2 were tested for mating type frequencies (Table 3). In only two cases Number of isolates before clone-correction. MAT1-2-specific bands (Table 2). These isolates were thought to be mixed cultures, thus not representing single clones. Single spores were isolated for 10 of these isolates having both MAT alleles, and subjected to a new matingtype multiplex PCR. In all 10 cases, the single-spore isolates could be identified as either MAT1-1 or MAT1-2, confirming that these isolates had been mixed cultures of Table 3 Mating type distribution of Rhynchosporium secalis on a small geographic scale within individual plots Population Aus1 Aus2 Aus3 Aus3 Aus3 Aus3 Aus6 CH3 CH3 CH4 ET3 Central Norway Finland y 2 Plot number Plot size (m2 ) 1 2 3 4 1 1 2 1 1 1 1
4.3
4.3 1.0 1.0 1.0 1.0 1.0 1.0 1.0 4.5 1.0 1.0 1.0 No. of isolates MAT1-1 MAT1-2 26 13 4 4 3 7 6 3 5 8 2 3 6 23 27 4 7 11 2 2 11 3 6 7 6 3 v2 y Pz 0.184 4.900 ... ... ... ... ... ... ... ... ... ... ... 0.775 0.039 ... ... ... ... ... ... ... ... ... ... ... v value based on a 1:1 ratio with 1 degree of freedom. A probability value as obtained with an exact binomial analysis to test whether mating type frequencies deviate significantly from a 1:1 ratio. yz . . . indicates populations where sample sizes were too small to perform a v2 test.  indicates mating type frequencies which are significantly different at P < 0:05. z C.C. Linde et al. / Fungal Genetics and Biology 40 (2003) 115–125 were sample sizes large enough to perform v2 and exact binomial tests, and one of them (Aus2) deviated significantly (P < 0:05) from a 1:1 ratio with the v2 test as well as the exact binomial test (P ¼ 0:039) (Table 3). 3.4. Macrogeographical distribution of mating types (within fields and regions) Different fields from the same geographical region e.g. Australia, Switzerland, Ethiopia, Norway, and South Africa, were analyzed for mating type frequencies and tested for deviations from a 1:1 ratio as determined with a v2 analyses and an exact binomial test (Table 4). In Australia, two of the seven populations deviated significantly (v2 test, P < 0:05 and P < 0:01; exact binomial test, P ¼ 0:039 and P ¼ 0:007, respectively) from a 1:1 mating type ratio, and one out of six populations in Switzerland showed significant (v2 test, P < 0:01; exact binomial test, P ¼ 0:001) differences in mating type frequencies. All four populations from Ethiopia, and both populations from Norway and South Africa did not deviate significantly from a 1:1 mating-type ratio (Table 4). Only one population each was analyzed from Finland and California. In both of these populations mating type frequencies deviated significantly (P < 0:05) from a 1:1 ratio (Table 4). Results with the v2 analyses were confirmed in all cases with the exact binomial test (Table 4). 3.5. Macrogeographical distribution of mating types (within and among continents) When subpopulations were combined to represent larger geographic areas (country), mating type ratios did not deviate significantly from a 1:1 ratio as determined with a v2 analysis and one degree of freedom (Table 4). To take into account the differences in sample sizes among subpopulations, a contingency v2 analyses was performed on subpopulations. This revealed that mating type frequencies among Australian populations differed significantly (v2 ¼ 16:230; P < 0:05; and df ¼ 6), as was the case among Swiss populations (v2 ¼ 14:909; P < 0:05; and df ¼ 5), and the total collection (v2 ¼ 50:561; P < 0:01; and df ¼ 22). Mating type frequencies did not differ significantly among Australian populations when the Aus7 population was removed from the analyses (v2 ¼ 7:172 and df ¼ 5). Similarly, after removal of the CH6 population from the Swiss populations, mating type frequencies did not differ significantly among populations (v2 ¼ 5:000 and df ¼ 4). Upon removal of both Aus7 and CH6 from the contingency v2 analyses among all populations, mating type frequencies did not differ significantly (v2 ¼ 31:158 and df ¼ 20). No significant differences in mating type frequencies could be detected among subpopulations from Norway and Ethiopia (Table 4). Mating type frequencies of combined geographic regions (regions or countries) did not 121 differ significantly as tested with the exact binomial test (Table 4). The contingency v2 analyses tests whether fields within a country or region differ in mating type frequencies. The exact binomial test determines whether the combined mating type frequency for each specific country or region deviates significantly from a 1:1 ratio. In total 401 isolates could be identified as MAT1-1 and 381 as MAT1-2 (Table 4). Comparison of mating type frequencies in the clonecorrected and non-clone-corrected data sets with a contingency v2 analyses (Table 4), showed no significant differences in mating type frequencies. A v2 analyses on the non-clone-corrected data set showed that mating type frequencies were significantly different for the same populations, whether the data set was clone-corrected or not (data not shown). An exception was the combined Swiss population (df ¼ 1) where mating type frequencies were significantly different in the non-clone-corrected data set and non-significant in the clone-corrected data set. We also found that the significance level increased in the non-clone-corrected data set from P < 0:05 to P < 0:01 in California, Aus2, and Finland populations (data not shown). 4. Discussion We partially sequenced the HMG- and a-domains in two isolates of R. secalis. The domains were present in different idiomorphs of the MAT gene, indicating that R. secalis is heterothallic. Primers were designed within the HMG- and a-domain to identify R. secalis mating types in a multiplex PCR. Mating type frequencies were determined in a large collection (N ¼ 1101) of R. secalis from populations representing some of the major barley-growing regions of the world. Contrary to other mating type distribution studies (e.g., Douhan et al., 2002), our collections were clone-corrected to remove bias due to clonal increase of specific mating types. The only populations which were not clone-corrected were those from South Africa. However, high levels of genetic diversity were reported previously for South African R. secalis populations (Robbertse and Crous, 2000), suggesting that clonal reproduction is not common in these populations. Mating type frequencies were equal for most populations studied, consistent with random mating. On a microgeographical scale, both mating types were often found to occupy the same leaf and lesions. 4.1. R. secalis resembles other heterothallic fungi All heterothallic ascomycetes studied so far shared the same organization of the MAT1-2 locus, having a HMGbox as DNA binding motif (Arie et al., 2000; Dyer et al., 2001; Kronstad and Staben, 1997; Waalwijk et al., 2002; 122 Table 4 Mating type frequencies of Rhynchosporium secalis in populations from Australia, Switzerland, Ethiopia, Scandinavia, California, and South Africa Area (m2 )a Aus1 Aus2 Aus3 Aus4 Aus5 Aus6 Aus7 Australia 1996 1996 1996 1996 1996 1996 1996 1996 CH1 CH2 CH3 CH4 CH5 CH6 Switzerland N No. of isolates MAT1-1b MAT1-2b
4.3 (
4.3)
4.3 (
4.3)
200 (
6)
15 (
15)
200 (
10)
200 (
6)
200 (
6)
823.6 (
51.6) 26 (34) 13 (14) 25 (25) 16 (16) 15 (16) 16 (22) 17 (35) 128 23 (41) 27 (33) 30 (30) 9 (9) 18 (26) 20 (24) 4 (13) 131 1999 1999 1999 2000 2001 2001 1999–2001
300 (
8)
300 (
8)
300 (
8)
800 (
36)
100 (
10)
120 (
8)
1920 (
78) 12 (20) 8 (8) 14 (14) 22 (25) 14 (14) 25 (31) 95 ET1 ET2 ET3 ET4 Ethiopia 2002 2002 2002 2002 2002
300 (
8)
300 (
8)
300 (
8)
300 (
8)
1200 (
32) SE. Norway C. Norway Norway Finland California 1996 1996 1996 1996 1988 SA1 SA2 South Africa Total collection Mating type frequency v2 c v2 d 0.184 4.900 0.455 1.960 0.273 0.444 8.048 0.035 Contingency v2 e Pf MAT1-1 MAT1-2 49 40 55 25 33 36 21 259 53.1 32.5 45.5 64.0 45.5 44.4 81.0 49.4 46.9 67.5 54.5 36.0 54.5 55.6 19.0 50.6 0.709 0.074 0.000 0.413 0.413 0.093 0.508 12 (12) 17 (22) 15 (15) 15 (15) 11 (11) 6 (6) 76 24 25 29 37 25 31 171 50.0 32.0 48.3 59.5 56.0 80.6 55.0 50.0 68.0 51.7 40.5 44.0 19.4 45.0 0.004 0.189 0.000 0.120 0.000 0.209 11 (12) 5 (5) 13 (15) 16 (18) 45 16 (17) 6 (7) 17 (19) 8 (8) 47 27 11 30 24 92 40.7 45.5 43.3 66.7 48.9 59.3 54.5 56.7 33.3 51.1 0.002 0.033 0.004 0.403
400 (
10)
18 (
6)
418 (
16)
400 (
10)
200 (
50) 12 (13) 24 (65) 36 33 (50) 6 (6) 20 24 44 18 18 32 48 80 51 24 37.5 50.0 45.0 64.7 25.0 62.5 50.0 55.0 35.3 75.0 0.004 0.056 2002 2002 2002
15 (
15)
40 (
2)
55 (
17) 31 27 58 28 19 47 59 46 105 52.5 58.7 55.2 47.5 41.3 44.8 0.153 1.391 1.152 0.396 (1) 0.795 0.302 0.329 1988–2002
4386.6 (
254.6) 401 381 782 51.3 48.7 0.512 50.561 (22) 0.497 (21) (60) (21) (40) 0.446 2.812 0.000 3.240 0.034 1.324 0.360 11.645 2.111 0.926 0.091 0.533 2.667 0.043 2.000 0.000 0.800 4.412 6.000 16.230 (6) 0.775 0.039 0.590 0.230 0.728 0.618 0.007 0.901 14.909 (5) 1.000 0.108 1.000 0.324 0.690 0.001 0.169 4.353 (3) 0.442 1.000 0.585 0.152 0.917 1.212 (1) 0.215 1.000 0.434 0.490 0.023 a Numbers indicate the total area of the field covered by transect sampling or the sites of a hierarchical sampling (McDonald et al., 1999). Numbers in parentheses indicate the actual area from which leaves were collected. b Values in parenthesis indicate number of MAT1-1 and MAT1-2 isolates in the data set before it was clone-corrected. c 2 v value based on a contingency v2 between clone-corrected and non-clone-corrected mating type ratios with 1 degree of freedom. d 2 v value based on a 1:1 ratio with 1 degree of freedom for the clone-corrected data set;  indicates mating type frequencies are significantly different at P < 0:05;  indicates significantly different at P < 0:01. e 2 v value based on a contingency v2 analyses among populations. Numbers in parentheses indicate degrees of freedom;  indicates mating type frequencies among populations are significantly different at P < 0:05;  indicates significantly different at P < 0:01. f A probability value as obtained with an exact binomial analysis to test whether clone-corrected mating type frequencies deviate significantly from a 1:1 ratio. C.C. Linde et al. / Fungal Genetics and Biology 40 (2003) 115–125 Year collected Population C.C. Linde et al. / Fungal Genetics and Biology 40 (2003) 115–125 Yun et al., 2000). In the MAT1-1 locus on the other hand, one can find either only an a-domain such as in the loculascomycetes (P€ oggeler, 2001), or both an a- and a HMG-domain as found in the pyrenomycetes and discomycetes (P€ oggeler, 2001; Singh et al., 1999; Turgeon et al., 1993). Since we could amplify these domains in different idiomorphs of R. secalis, it should be heterothallic and we presume its mating-type idiomorphs share the same organization as that of its closest known relative, T. yallundae (discomycete) (Goodwin, 2002). Further sequencing of both MAT idiomorphs is, however, necessary to confirm its MAT locus organization. Partial identification of the HMG- and a-domains of R. secalis enabled us to design primers to identify R. secalis mating types in a multiplex PCR. Because these primers were designed within regions which are similar among genera and even fungal classes (Picard et al., 1991; Singh et al., 1999), it is not surprising that the MAT1-2 primers gave a faint amplification product in one of the T. yallundae strains. However, neither the MAT1-1 primers, nor the multiplex reaction resulted in any amplification products in T. yallundae. The high level of similarity between HMG-box sequences of T. yallundae and R. secalis confirms the close genetic relationship between these two fungi (Goodwin, 2002) and provides further opportunities to study their phylogenetic relationship based on MAT loci. 4.2. Mating type distribution at microgeographic scales– diversity within and among lesions on a leaf and within field plots We found both mating types at all spatial scales tested. Both mating types could be detected in the same lesions in barley leaves collected from South Africa (three out of six lesions) and Australia (three out of 12 lesions). This is not unique to R. secalis, as both mating types were also found within the same lesions of e.g. M. graminicola (Linde et al., 2002), a fungus known to undergo regular sexual recombination (Chen and McDonald, 1996). Similarly, among lesion diversity on the same leaf was high for both the Australian and South African populations studied. Given the now known heterothallic nature of R. secalis, and a possibility for strains of opposite mating type to interact, co-occurrence of isolates having different mating types within the same lesion would theoretically allow mating interaction. Mating types also occurred in equal frequencies at a spatial scale of
4.3 m2 for one of the two plots analyzed. 4.3. Mating type distribution at macrogeographic scales—diversity within a region The null hypothesis in this analysis was that the mating type frequencies would be equal across all spatial scales if, as a result of frequency-dependent selection, 123 sexual reproduction is common in R. secalis. Mating types occurred at equal frequencies for most field populations analyzed. For instance, the Aus7 population was one of only two Australian populations with unequal proportions of mating types. This was also the only population from Western Australia and was therefore geographically separated from the other Australian populations. It was shown in an earlier RFLP analysis that this population (Quindanning) was genetically distinct from other populations sampled in Australia and possibly represents a distinct founder event (McDonald et al., 1999), which could explain the unequal mating type frequencies. Unequal mating type frequencies were also observed in four other populations. Sexual reproduction may be less frequent in these populations, or selection may have favored one mating type over the other by chance. In either case, it is possible that the populations have not reached equilibrium. For populations where mating types occurred at significantly different frequencies, it was either the MAT1-1 or MAT1-2 isolates that occurred in excess and not always just the one mating type. This finding, in addition to the lack of differences in mating type frequencies of clone-corrected vs. non-clone-corrected data sets, indicates that one mating type does not, on average, have an advantage over the other mating type under clonal reproduction. This suggests that there is no selection bias favoring one of the mating types in R. secalis. 4.4. Mating type distribution at macrogeographic scales—diversity within and among continents Finland was the only population in Scandinavia with unequal mating type frequencies. This could represent a founder event of R. secalis, although a previous study indicated high genetic diversity in this population (Salamati et al., 2000) which is not consistent with a founder effect. Mating type frequencies differed significantly among populations in Australia (v2 ¼ 16:230 and df ¼ 6) and Switzerland (v2 ¼ 14:909 and df ¼ 5). This was due mainly to one population that was not in equilibrium. When these populations were removed from the Australian or Swiss populations, mating type frequencies no longer differed significantly among populations. Some populations with significantly different mating type frequencies were also found for T. yallundae and T. acuformis populations in the US (Douhan et al., 2002). As fungal plant pathogen populations in an agricultural field are often founded by a small fraction of the total population, it is perhaps not surprising that e.g. environmental conditions within each field are not always favorable for sexual reproduction, and therefore not all populations will undergo the frequent sexual cycles needed to maintain a 1:1 ratio of mating types. The equal frequencies of mating types across all spatial scales in most field populations of R. secalis is 124 C.C. Linde et al. / Fungal Genetics and Biology 40 (2003) 115–125 consistent with frequency-dependent selection (May et al., 1999), and therefore, random mating. This finding supports previous work where it was suggested that R. secalis must reproduce sexually to account for high genotypic diversity and random associations among RFLP loci (McDonald et al., 1999; Salamati et al., 2000). Identification of mating-type genes in R. secalis opens up many possibilities to study the genetics of this fungus as attempted sexual crossings can now be made with known opposite mating types. Acknowledgments We thank the many collaborators who contributed collections of leaf samples used in this study. They are David Moody (Australia), Robert Loughman (Australia), Barbara Read (Australia), Marja Jalli (Finland), Buskerud Extension Office (Buskerud forsøksring, Norway), Kiros Meles (Ethiopia), Lucienne Mansvelt (South Africa), Denise Liebenberg (South Africa), and Vincent Michel (RAC, Switzerland). We also recognize Pascal Zaffarano and Elena Turco for technical assistance, Paul Dyer (University of Nottingham, UK) for providing the T. yallundae MAT sequences, Tim Murray (Washington State University, USA) for providing T. yallundae DNA, and Simon Foster (Rothamsted Research Institute, UK) for useful discussions. References Ahmed, S., Morrall, R.A.A., Kaiser, W.J., 1996. Distribution of mating types of Ascochyta fabae f. sp. lentis. Can. J. Plant Pathol. 18, 347–353. Altschul, S.F., Madden, T.L., Schaffer, A.A., Zhang, J.H., Zhang, Z., Miller, W., Lipman, D.J., 1997. Gapped BLAST and PSI-BLAST: a new generation of protein database search programs. Nucleic Acids Res. 25, 3389–3402. Arie, T., Christiansen, S.K., Yoder, O.C., Turgeon, B.G., 1997. Efficient cloning of ascomycete mating type genes by PCR amplification of the conserved MAT HMG box. Fungal Genet. Biol. 21, 118–130. Arie, T., Kaneko, I., Yoshida, T., Noguchi, M., Nomura, Y., Yamaguchi, I., 2000. Mating-type genes from asexual phytopathogenic ascomycetes Fusarium oxysporum and Alternaria alternata. Mol. Plant Microbe Interact. 13, 1330– 1339. Armstrong, C.L., Chongo, G., Gossen, B.D., Duczek, L.J., 2001. Mating type distribution and incidence of the teleomorph of Ascochyta rabiei (Didymella rabiei) in Canada. Can. J. Plant Pathol. 23, 110–113. Caldwell, R.M., 1937. Rhynchosporium secalis of barley, rye, and other grasses. J. Agric. Res. 55, 175–198. Chen, R.S., McDonald, B.A., 1996. Sexual reproduction plays a major role in the genetic structure of populations of the fungus Mycosphaerella graminicola. Genetics 142, 1119–1127. Coppin, E., Debuchy, R., Arnaise, S., Picard, M., 1997. Mating types and sexual development in filamentous ascomycetes. Microbiol. Mol. Biol. Rev. 61, 411–428. Dayakar, B.V., Narayanan, N.N., Gnanamanickam, S.S., 2000. Crosscompatibility and distribution of mating type alleles of the rice blast fungus Magnaporthe grisea in India. Plant Dis. 84, 700–704. Douhan, G.W., Murray, T.D., Dyer, P.S., 2002. Species and matingtype distribution of Tapesia yallundae and T. acuformis and occurrence of apothecia in the US Pacific Northwest. Phytopathology 92, 703–709. Dyer, P.S., Furneaux, P.A., Douhan, G., Murray, T.D., 2001. A multiplex PCR test for determination of mating type applied to the plant pathogens Tapesia yallundae and Tapesia acuformis. Fungal Genet. Biol. 33, 173–180. Everitt, B.S., 1977. The Analysis of Contingency Tables. Wiley, New York. Goodwin, S.B., 2002. The barley scald pathogen Rhynchosporium secalis is closely related to the discomycetes Tapesia and Pyrenopeziza. Mycol. Res. 106, 645–654. Goodwin, S.B., Maroof, M.A.S., Allard, R.W., Webster, R.K., 1993. Isozyme variation within and among populations of Rhynchosporium secalis in Europe, Australia and the United States. Mycol. Res. 97, 49–58. James, W.C., Jenkins, J.E., Jemmet, J.L., 1968. The relationship between leaf blotch caused by Rhynchosporium secalis and losses in grain yield of spring barley. Ann. Appl. Biol. 62, 273–288. Kronstad, J.W., Staben, C., 1997. Mating type in filamentous fungi. Annu. Rev. Genet. 31, 245–276. Liebhard, R., Gianfranceschi, L., Koller, B., Ryder, C.D., Tarchini, R., Van de Weg, E., Gessler, C., 2002. Development and characterisation of 140 new microsatellites in apple (Malus  domestica Borkh.). Mol. Breed. 10, 217–241. Linde, C.C., Zhan, J., McDonald, B.A., 2002. Population structure of Mycosphaerella graminicola: from lesions to continents. Phytopathology 92, 946–955. Liu, Y.C., Cortesi, P., Double, M.L., MacDonald, W.L., Milgroom, M.G., 1996. Diversity and multilocus genetic structure in populations of Cryphonectria parasitica. Phytopathology 86, 1344–1351. Liu, Y.G., Whittier, R.F., 1995. Thermal asymmetric interlaced PCR: automatable amplification and sequencing of insert end fragments from P1 and YAC clones for chromosome walking. Genomics 25, 674–681. May, G., Shaw, F., Badrane, H., Vekemans, X., 1999. The signature of balancing selection: fungal mating compatibility gene evolution. Proc. Natl. Acad. Sci. USA 96, 9172–9177. McDermott, J.M., McDonald, B.A., Allard, R.W., Webster, R.K., 1989. Genetic variability for pathogenicity, isozyme, ribosomal DNA and colony color variants in populations of Rhynchosporium secalis. Genetics 122, 561–565. McDonald, B.A., Zhan, J., Burdon, J.J., 1999. Genetic structure of Rhynchosporium secalis in Australia. Phytopathology 89, 639–645. Mekwatanakarn, P., Kositratana, W., Phromraksa, T., Zeigler, R.S., 1999. Sexually fertile Magnaporthe grisea rice pathogens in Thailand. Plant Dis. 83, 939–943. Milgroom, M.G., 1996. Recombination and the multilocus structure of fungal populations. Annu. Rev. Phytopathol. 34, 457–477. Notteghem, J.L., Silue, D., 1992. Distribution of the mating type alleles in Magnaporthe grisea populations pathogenic on rice. Phytopathology 82, 421–424. Picard, M., Debuchy, R., Coppin, E., 1991. Cloning the mating types of the heterothallic fungus Podospora anserina : developmental features of haploid transformants carrying both mating types. Genetics 128, 539–547. P€ oggeler, S., 2001. Mating-type genes for classical strain improvements of ascomycetes. Appl. Microbiol. Biotechnol. 56, 589–601. Robbertse, B., Crous, P.W., 2000. Genotypic variation in Rhynchosporium secalis pathotypes collected in the Western Cape Province of South Africa. S. Afr. J. Sci. 96, 391–395. Salamati, S., Zhan, J., Burdon, J.J., McDonald, B.A., 2000. The genetic structure of field populations of Rhynchosporium secalis C.C. Linde et al. / Fungal Genetics and Biology 40 (2003) 115–125 from three continents suggests moderate gene flow and regular recombination. Phytopathology 90, 901–908. Sharon, A., Yamaguchi, K., Christiansen, S., Horwitz, B.A., Yoder, O.C., Turgeon, B.G., 1996a. An asexual fungus has the potential for sexual development. Mol. Gen. Genet. 251, 60–68. Sharon, A., Yamaguchi, K., Christiansen, S., Horwitz, B.A., Yoder, O.C., Turgeon, B.G., 1996b. An asexual fungus has the potential for sexual development. Mol. Gen. Genet. 251, 60–68. Shipton, W.A., Boyd, J.R., Ali, S.M., 1974. Scald of barley. Rev. Plant Pathol. 53, 839–861. Singh, G., Dyer, P.S., Ashby, A.M., 1999. Intra-specific and interspecific conservation of mating-type genes from the discomycete plant-pathogenic fungi Pyrenopeziza brassicae and Tapesia yallundae. Curr. Genet. 36, 290–300. Sokal, R.R., Rohlf, F.J., 1995. Biometry: the principles and practice of statistics in biological research. Freeman, New York. Tatusova, T.A., Madden, T.L., 1999. BLAST 2 SEQUENCES, a new tool for comparing protein and nucleotide sequences. FEMS Microbiol. Lett. 174, 247–250. Turgeon, B.G., 1998. Application of mating type gene technology to problems in fungal biology. Annu. Rev. Phytopathol. 36, 115–137. Turgeon, B.G., Bohlmann, H., Ciuffetti, L.M., Christiansen, S.K., Yang, G., Schafer, W., Yoder, O.C., 1993. Cloning and analysis of the mating-type genes from Cochliobolus heterostrophus. Mol. Gen. Genet. 238, 270–284. 125 Turgeon, G., Yoder, O.C., 2000. Proposed nomenclature for mating type genes of filamentous ascomycetes. Fungal Genet. Biol. 31, 1–5. Versalovic, J., Koeuth, T., Lupski, J.R., 1991. Distribution of repetitive DNA-sequences in Eubacteria and application to fingerprinting of bacterial genomes. Nucleic Acids Res. 19, 6823–6831. Versalovic, J., Schneider, M., de Bruijn, F.J., Lupski, J.R., 1994. Genomic fingerprinting of bacteria using repetitive sequence-based polymerase chain reaction. Methods Mol. Cell Biol. 5, 25–40. Viji, G., Uddin, W., 2002. Distribution of mating type alleles and fertility status of Magnaporthe grisea causing gray leaf spot of perennial ryegrass and St. Augustinegrass turf. Plant Dis. 86, 827–832. Waalwijk, C., Mendes, O., Verstappen, E.C.P., de Waard, M.A., Kema, G.H.J., 2002. Isolation and characterization of the matingtype idiomorphs from the wheat septoria leaf blotch fungus Mycosphaerella graminicola. Fungal Genet. Biol. 35, 277–286. Welz, H.G., Leonard, K.J., 1995. Gametic phase disequilibria in populations of race-2 and race-3 of Cochliobolus carbonum. Eur. J. Plant Pathol. 101, 301–310. Yun, S.H., Arie, T., Kaneko, I., Yoder, O.C., Turgeon, B.G., 2000. Molecular organization of mating type loci in heterothallic, homothallic, and asexual Gibberella/Fusarium species. Fungal Genet. Biol. 31, 7–20. Zhan, J., Kema, G.H.J., Waalwijk, C., McDonald, B.A., 2002. Distribution of mating type alleles in the wheat pathogen Mycosphaerella graminicola over spatial scales from lesions to continents. Fungal Genet. Biol. 36, 128–136.