,
Phaeosphaeria species are important pathogens of cereals with global distribution. The origin of Phaeosphaeria species is in the Fertile Crescent coinciding with their hosts (McDonald et al., 2012). Phaeosphaeria nodorum (Muller) Hedjar (anamorph Parastagonospora nodorum (Berk.) Quaedvl., Verkley & Crous) is the causal agent of Septoria nodorum leaf and glume blotch (SNB) on wheat, a widespread and yield-reducing disease in many of wheat-growing regions of the world (Shipton et al., 1971; Sommerhalder et al., 2006). P. nodorum is also pathogenic on barley and a wide range of wild grasses (Solomon et al., 2006). Phaeosphaeria avenaria f. sp. avenaria (Paa), (Weber) Eriksson (anamorph Stagonospora avenae f. sp. avenae Frank) is a major leaf pathogen of oat and other cereals. Phaeosphaeria avenaria f. sp. tritici (Pat) described by Shaw (1957), is morphologically similar to Paa but not pathogenic on oat while pathogenic on wheat and other cereals (McDonald et al., 2012). Ueng and Chen (1994) and Ueng et al. (1998) studies on genetic differences between biotypes, split Pat into three groups, Pat1, Pat2, and Pat3. Later, McDonald et al. (2012) included over 300 Phaeosphaeria isolates collected from wild grasses on different continents in a three-gene phylogeny of internal transcribed spacer (ITS), β-tubulin and β-xylosidase and mating type loci to determine the relationships among Phaeosphaeria spp. complex and introduced three new groups; Pat4, Pat5, and Pat6.
Different genes have been used to analyze Phaeosphaeria species complex including mating type loci, β-tubulin, β-glucosidase, RNA polymerase II; histidine synthase (Bennett et al., 2003; Malkus et al., 2005, 2006; Reszka et al., 2005; Ueng et al., 2003; Wang et al., 2007). In a recent study, whole-genome sequencing data have been used to explore the phylogenetic relationships among Phaeosphaeria species (Croll et al., 2021).
Phaeosphaeria species are heterothallic fungi. Sexual reproduction in these species requires the presence of two isolates carrying opposite forms of mating type idiomorphs, called MAT1-1 and MAT1-2, at the same geographic location (Solomon et al., 2004; Sommerhalder et al., 2006). The extent of sexual reproduction and the contribution of airborne ascospores as the source of primary inoculum is important in epidemiology and management of pathogens. Recombination resulting from sexual mating have the potential to give rise to fitter genotypes that are more virulent and fungicide resistant. In asexual reproduction, the main source of the inoculum is pycnidiospores that have a limited increase in genetic diversity comparing to ascospores (Sommerhalder et al., 2006). The presence of both mating types and the mating type ratios have been studied to obtain information about the sexual reproduction by heterothallic fungi (Cowger and Silva-Rojas, 2006; Notteghem and Silué, 1992).
There are arguments about the degree of sexual reproduction in populations of Phaeosphaeria species. Studies on P. nodorum isolates from North Africa, North America, Australia, Europe, and Near East showed that MAT1-1 and MAT1-2 were not evenly distributed (Halama, 2002). Skewed mating type ratios among P. nodorum isolates are reported among populations from Central Asia (Vergnes et al., 2006). On the other hand, random mating within populations of P. nodorum from Texas, Oregon, and Switzerland have been proved (Keller et al., 1997a, 1997b 1994; Sommerhalder et al., 2006). McDonald et al. (2012) studied isolates of Pat from five continents and observed that all Pat1 isolates carried only the MAT1-1 allele, Pat3 and Pat5 isolates had both MAT1-1 and MAT1-2 alleles, and Pat4 and Pat6 isolates were all MAT1-2.
The objectives of this study were (1) to collect Phaeosphaeria isolates from wheat, barley and wild grasses in Bushehr, Kohgiluyeh va Boyer-Ahmad and Khuzestan Provinces in Iran and identify the species using morphological and molecular data, (2) to analyze mating type loci of the collected isolates in addition to isolates collected from wheat, barley, and wild grasses in Iran from our previous studies and use them in a phylogeny to determine the relationships among 273 isolates within the Phaeosphaeria spp. complex; and (3) to investigate mating type distribution among Phaeosphaeria species in Iran.
Materials and Methods
Sampling, fungal isolation, and morphological characterization
Symptomatic leaves and ears of wheat, barley, and wild grasses were collected from Bushehr, Kohgiluyeh va Boyer-Ahmad and Khuzestan Provinces in Iran, and taken to laboratory. The diseased leaves and ears showing typical symptoms of SNB were cut into segments of 5-7 mm, sterilized for 2 min in 1% sodium hypochlorite, rinsed in sterile water, placed in glass slides with tape, and kept in high humidity until the pycnidia produced cirri containing pycnidiospores. Purification was carried out using single-spore method on 2% water agar medium in plastic Petri dishes with a flame-sterilized needle. After 2-3 days of incubation, germinated spore was transferred to yeast sucrose agar (YSA, 10 g/l yeast extract, 10 g/l sucrose, 1.2% agar). Pure cultures of each isolate were stored on lyophilized filter-paper strips at −80°C (Adhikari et al., 2008). Only one single-spore strain of Phaeosphaeria sp. was isolated from each infected plant and morphological characteristics i.e., colony color, conidia and conidiomata morphology, pigmentation and colony growth rate were used for species identifications (Quaedvlieg et al., 2013).
DNA extraction
For molecular identifications, mycelium plugs from isolates grown on YSA (10 g/l yeast extract, 10 g/l sucrose, 16 g/l agar) for 5 days were transferred to flasks containing 50 ml yeast sucrose broth medium (YSB, 10 g/l yeast extract, 10 g/l sucrose) and incubated on an orbital shaker for 7 days at 120 rpm at 18°C. Harvested mycelia were freeze-dried and stored at −20°C until further use. Lyophilized mycelium was ground into powder and total DNA was extracted using CTAB method according to Murray and Thompson (1980). Genomic DNA was visualized on a 1.2% agarose gel (1.2% agarose, 0.5× TAE) using UV light (GelDoc, Bio-Rad Laboratories, Hercules, CA, USA). Obtained sequences were deposited in Gen-Bank (Tables 1 and 2).
Mating type identification and fertility
Mating type idiomorphs, MAT1-1 and MAT1-2, were amplified and sequenced for 67 Phaeosphaeria isolates that were obtained in this study and 201 isolates from our collection, which were isolated in Ghaderi et al. (2017, 2020) (Tables 1 and 2).
The amplifications were carried out using a multiplex polymerase chain reaction (PCR) with primers (Table 3) designed by Bennett et al. (2003). PCR amplifications were performed in 25 μl reactions containing 2 μm primers, 0.4 mM dNTPs (Fermentas Inc., Waltham, MA, USA), 8 pg DNA, 0.05 U Taq DNA polymerase (MBI Fermentas), 3 mM MgCl2 and corresponding reaction Dream Taq buffer (MBI Fermentas) (Sommerhalder et al., 2006). The PCR condition was set up in 2 min initial denaturation at 96°C, 35 cycles of 30 s at 96°C, annealing at 55°C for 30 s, extension at 72°C for 1 min and a final 5 min extension at 72°C (Sommerhalder et al., 2006). The PCR products were visualized on 1.5% agarose gel (1.5% agarose, 0.5× TBE). Sequencing was performed by Macrogen (Seoul, Korea). The ratio of MAT1-1 to MAT1-2 alleles was evaluated. Deviations from a 1:1 ratio of the two mating types within fields was tested using chi-square statistics.
To confirm the development of sexual phase, genetic crosses were carried out between opposite mating types for 67 Phaeosphaeria isolates collected in this study following the procedure of Halama and Lacoste (1992). Opposite mating types were grown on 2% water agar medium encompassing sterilized wheat straws and incubated at 10°C with a 12-h photoperiod, near ultraviolet light (300-400 nm) and intensities of 400 and 600 pW/cm2 for 50 days. Isolates were paired with themselves as control.
Genetic data analyses
Mat1-1 and Mat1-2 sequences were generated for 273 isolates. The obtained sequences of 273 isolates of this study and some sequences from McDonald et al. (2012) which were obtained from GenBank, were used in phylogenetic analyses to determine the taxonomic status of Phaeosphaeria species and identifications (Tables 1 and 2).
The phylogenetic analyses of Mat1-1 and Mat1-2 alleles were performed separately. Sequences were edited manually and aligned by Geneious version 7 (Biomatters Ltd., Auckland, New Zealand). Phylogenetic analyses were performed using heuristic searches in PAUP v. 4.0a133 (Swofford, 2002) for parsimony, neighbour-joining and maximum likelihood analyses. For maximum likelihood analyses, models of sequence evolution were evaluated for both datasets by JModeltest v.2.1.4 (Posada, 2008) using the Akaike information criterion and model parameter estimates were implemented in PAUP v. 4.0a133. The resulted trees were midpoint rooted. The resulted trees were observed and edited in FigTree v1.4.0. Mating Type polymorphism within each species assessed using DnaSP v5 (Librado and Rozas, 2009).
PCR-restriction fragment length polymorphism technique
In order to confirm identification of P. avenaria f. sp. tritici 1 (Pat1) isolates, we used PCR-restriction fragment length polymorphism (PCR-RFLP) assay (McDonald et al., 2012) to differentiate between P. nodorum and Pat1. Isolates of Phaeosphaeria sp. were surveyed using this method.
An isolate obtained from earlier studies of McDonald et al. (2012) was used as positive control. Partial sequence of β-xylosidase gene (962 bp) was amplified using specific primers (Table 3) and was used as template DNA. Species-specific restriction enzyme recognition sites were distinguishing by NEB Cutter v2.0. Digestion of β-xylosidase gene PCR amplicons was carried out with 2 units of the restriction enzyme ScaI (MBI Fermentas) at 37°C for 90 min. A 15-min treatment at 65°C was applied for inactivation. Digested PCR products were visualized on 2% agarose gels with ethidium bromide staining.
Results
Fungal isolation and identification
In total, 67 Phaeosphaeria isolates were obtained from symptomatic leaves and ears of wheat, barley and wild grasses from Bushehr, Khuzestan and Kohgiluyeh va Boyer-Ahmad Provinces. Based on morphological characterization and molecular analysis, isolates were identified as P. nodorum, Paa, Pat1, Pat3, and Pat5. Paa, Pat1, Pat3, and Pat5 were identified based on molecular phylogeny of MAT genes.
In Bushehr province, 48 isolates were collected from wheat and wild grasses (Tables 1 and 2). We obtained 30 isolates of P. nodorum from wheat ears, two from Aegilops tauschii ears and two from Avena sativa leaves. A. tauschii and A. sativa are new hosts for P. nodorum to the world. Four isolates from wheat ears were identified as Pat3. Pat3 was first isolated from Iran in this study. Ten pat1 isolates were obtained from wheat and barley. Barley is a new host for pat1 to the world.
In Kohgiluyeh va Boyer-Ahmad Province, 16 isolates were collected from leaves and ears of wild grasses (Tables 1 and 2). Three Paa isolates, one from P. arundinacea and two from Convolvulus arvensis leaves were obtained. Convolvulus arvensis is a new host for Paa to the world. Five isolates of Pat5 were identified from P. arundinacea and A. tauschii. Two P. nodorum isolates were identified from Dactylis glomerata ears, which is a new host to the world. Six Pat3 isolates were obtained from barley ears. Barley is a new host for Pat3 to the world. Pat3 is isolated for the first time from Iran.
In Khuzestan Province, three isolate were collected. One Paa was obtained from ear of A. tauschii, which is a new host for Paa to the world and two Pat5 were isolated from A. tauschii and P. arundinacea.
Phylogenetic analyses
Amplification of MAT1-1 and MAT1-2 gene fragments of 510 bp and 360 bp from all isolates (Tables 1 and 2) was conducted successfully. The obtained sequences were deposited in GenBank and accession numbers were obtained (Tables 1 and 2). The aligned data sets of MAT1-1 and MAT1-2 gene consisted of 269 and 396 characters of which 52 and 55 characters were parsimony informative, respectively. The three phylogenetic analysis methods parsimony, neighbor-joining and maximum likelihood generated trees with similar topologies amongst species. The topology and branch lengths of the parsimony phylogenetic trees are shown in Figs. 1 and 2.
To elucidate phylogenetic relationships among 273 Phaeosphaeria species from wheat, Barley and wild grasses and accurate species identifications, separate parsimony trees were created using MAT1-1 and MAT1-2 gene sequences (Figs. 1 and 2). Fig. 1 shows the phylogenetic position of Phaeosphaeria isolates using MAT1-1 sequences. The phylogenetic reconstruction revealed four highly supported clades corresponding to P. nodorum (containing 6 nucleotide haplotypes, 4 with synonymous mutations, 2 with non-synonymous mutations) (Table 4) and three formae speciales of P. avenaria including Pat5 (containing 2 nucleotide haplotypes, 1 with non-synonymous mutations), Pat3 (containing 2 nucleotide haplotypes, 1 with non-synonymous mutations) and Pat1. Fig. 2 shows the phylogenetic position of Phaeosphaeria isolates using MAT1-2 sequences. The tree contains six highly supported clades corresponding to P. nodorum (containing 4 nucleotide haplotypes, 4 with synonymous mutations, 1 with non-synonymous mutations) and five P. avenaria formae specials including Pat1, Pat3 (containing 2 nucleotide haplotypes, 1 with non-synonymous mutations), Pat5 (with 1 nucleotide haplotype), P. avenaria f. sp. tritici 6 (Pat6), and Paa (with 01 nucleotide haplotype).
PCR-RFLP technique
We used a PCR-RFLP technique to distinguish P. nodorum isolates from Pat1 isolates based on fixed species polymorphisms. The 962-bp PCR products were amplified from genomic DNA with β-xylosidase gene-specific primers and digested using ScaI enzyme. PCR products from Pat1 isolates had a specific restriction site and 695-bp and 267-bp fragments were produced. There was no restriction site in PCR products of 221 P. nodorum isolates, which produced 962-bp amplicons (Fig. 3).
Mating type distribution and fertility
The ratio of MAT1-1 to MAT1-2 alleles and the results of chi-square statistics for testing deviations from a 1:1 ratio of the two mating types within fields are presented in Table 5. The mating type ratio was not significantly different from 1:1 ratio for Kohgiluyeh va Boyer-Ahmad and Bushehr.
The mating ability of Phaeosphaeria species obtained in this study was examined. None of isolates formed ascocarp (pseudothecia) when grown alone. The pseudothecia were obtained after 50 days of incubation on sterilized wheat straws. Pseudothecia were recognized from pycnidia by the absence of cirrhi. We could not observe mature pseudothecia containing asci and ascospores in any of the crosses in laboratory conditions (Fig. 4).
Discussion
In this study, 67 Phaeosphaeria spp. were obtained from symptomatic leaves and ears of wheat, barley and wild grasses collected in Bushehr, Kohgiluyeh va Boyer-Ahmad and Khuzestan Provinces in Iran. Based on morphological characteristics and molecular data, P. nodorum, Paa, Pat5, Pat3, and Pat1 were identified. These included seven new host records including A. tauschii, A. sativa and D. glomerata for P. nodorum, barley for Pat1 and Pat3, C. arvensis and A. tauschii for Paa. Phaeosphaeria species complex have reported to have the ability to infect several grass hosts (McDonald et al., 2012; Solomon et al., 2006). However, the host range for Phaeosphaeria species is yet unknown. Pat3 is reported in this study for the first time from Iran.
P. nodorum isolates were wildly distributed in the sampled areas. McDonald et al. (2012) observed the same trend in the distribution of Phaeosphaeria species in global scale. Ghaderi et al. (2010) studied the diversity of Phaeosphaeria species associated with poaceous plants in Iran and identified P. nodorum, Paa, and Pat5. In another study, Ghaderi and Razavi (2018) reported Phaeosphaeria dactylidis from wild grasses. Species richness of Phaeosphaeria in Iran is consistent with the hypothesis of the origin in the Fertile Crescent.
Since morphological characters between Phaeosphaeria species often overlap and cultural characteristics are in many cases variable (Bennett et al., 2003; Cunfer, 2000; Shoemaker and Babcock, 1989), species relationships and taxonomy need further molecular characterization. We used molecular phylogeny of mating type genes to elucidate relationships among the Phaeosphaeria species. Mat1-1 and Mat1-2 sequences of 67 isolates collected in this study and 206 isolates which were collected in our previous studies (Ghaderi et al., 2017, 2020) were generated (Tables 1 and 2). The obtained sequences were combined with some sequences of Mat1-1 and Mat1-2 published previously by McDonald et al. (2012), and were used in the phylogenetic analyses to infer relationships between Phaeosphaeria species. In the resulting trees, all clades were separated with high bootstrap support. Isolates were grouped as P. nodorum clade, and four clades corresponding to different formae speciales including Paa, Pat5, Pat3, and Pat1. According to Turgeon (1998) within species variations is low for MAT genes while between-species variation is high making them a useful region to test the biological and phylogenetic species concepts for outcrossing fungi. Turgeon (1998) suggested MAT sequences are more useful in phylogenetic resolution than ITS rDNA and GPD sequence regions. Ueng et al. (2003) studied the potential use of mating type genes in phylogeny and molecular classification of Phaeosphaeria species. They observed that phylogenetic relationships in cereal Phaeosphaeria isolates based on mating type gene sequences were consistent with those based on RFLP fingerprints and rDNA ITS sequences. Bennett et al. (2003) observed between-species MAT variations in the genus Phaeosphaeria. They suggested MAT genes as a reliable diagnostic procedure to elucidate species relationships in the Phaeosphaeria species pathogenic to cereal crops. In addition to MAT phylogeny, we successfully used a PCR-RFLP technique developed by McDonald et al. (2012) to distinguish P. nodorum isolates from Pat1 isolates based on fixed species polymorphisms and 221 P. nodorum isolates were identified.
Both mating types were present in all sampling regions from Iran. Mating type ratio for Khuzestan was not calculated because sample size was small. Mating type ratio of the sampled areas in Bushehr was not significantly different from 1:1. In Kohgiluyeh va Boyer-Ahmad, the ratio of mating types for 16 isolates obtained from wild grasses in this study was 1:1. However, when we added the data of 62 isolates which were previously obtained from wheat in Ghaderi et al. (2020) (45 MAT1-1 vs. 17 MAT1-2), the overall mating types ratio showed a significantly skewed distribution in this area. A possible explanation for this skewed distribution is that part of our sampling have been done within a pycnidial clone. However, obtaining a robust estimate of the mating type distribution requires large number of isolates and samplings that are more extensive (Solomon et al., 2004). The ratio of mating types was not deviated significantly from 1:1 when considering our entire dataset of Iranian Phaeosphaeria population in Tables 1 and 2 (The chi-square statistic = 0.3091 and P = 0.578213). However, MAT1-1 isolates were predominant. The same results were obtained by Solomon et al. (2004) in populations of P. nodorum in Western Australia. They observed that the ratio of mating type alleles was not significantly different from equal proportions while MAT1-1 isolates were predominant. They tested different hypotheses regarding the predominance of MAT1-1 alleles including greater virulence and higher asexual fitness of MAT1-1 strains. None of these hypotheses could explain the prevalence of MAT1-1 strains.
We did not observe sexual structures in the sampled areas. One possible explanation would be the dry springs in the sampling years, which have substantially decreased the frequency of ascocarps. Another explanation would be that we have collected plant materials at wrong timepoint in the disease cycle. Mutations in other genes and sex barriers such as female sterility have also been proposed as possible explanations for inability to find the teleomorphs (Bennet et al., 2003; Sommerhalder et al., 2006).
Mating type ratios have been explored for Phaeosphaeria species to study these pathogens biology in order to reach evidence on the extent of sexual reproduction in populations. Halama (2002) observed that MAT1-1 alleles predominated in all of the populations sampled from different parts of the world. Bennett et al. (2003) observed skewed distribution of mating types in one population of P. nodorum and equal distribution in another population from a different field in New York, USA. Sommerhalder et al. (2006) tested a comprehensive collection of P. nodorum isolates from six countries on five continents and reported that this pathogen has even distribution of both mating types among all field populations. Vergnes et al. (2006) examined Central Asia populations of P. nodorum and reported the presence of both mating types in Kazakh and Russian origins while no MAT1-2 isolates were found in Tajikistan population. Mating type ratios data would be used to infer interesting information about population genetics, epidemiology, and control strategies of Phaeosphaeria species.
Our observations of the presence of both mating types and a 1:1 mating type ratio for our entire data set indicate that the Iranian Phaeosphaeria population have the opportunity to interact and undergo regular sexual reproduction resulting high genetic diversity. This hypothesis is consistent with McDonald et al. (2012) and Ghaderi et al. (2020) who showed that Iranian populations of Phaeosphaeria species had high levels of genetic diversity. It is likely that the main primary inoculum of Phaeosphaeria diseases is airborne ascospores in years with favorable climatic conditions. In the years with high sexual reproduction, clean seed, and crop rotation techniques are not preventive enough. The presence of a mixed reproductive system in these pathogens should be considered in plant breeding and fungicide screening programs. Quantitative resistance via polygenic control would be useful to overcome the possible break up of co-adapted gene complexes in the sexual reproducing periods.
PDF Links
PubReader
ePub Link
Full text via DOI
Full text via PMC
Download Citation
Print
