Plant Pathology (2018) Doi: 10.1111/ppa.12886 Low genetic diversity of Rhynchosporium commune in Iran, a secondary centre of barley origin E. Seifollahiab* , B. Sharifnabia, M. Javan-Nikkhahc and C. C. Lindeb a Department of Plant Protection, College of Agriculture, Isfahan University of Technology, Isfahan 84156-83111, Iran; bEcology and Evolution, Research School of Biology, College of Science, Australian National University, Canberra, ACT 2601, Australia; and cDepartment of Plant Protection, College of Agriculture and Natural Resources, University of Tehran, Karaj 31587–77871, Iran Rhynchosporium commune is a destructive pathogen of barley, causing leaf scald. Previous microsatellite studies used Syria as a representative of cultivated barley’s centre of origin, the Fertile Crescent. These suggested that R. commune and Hordeum vulgare (cultivated barley) did not co-evolve in the host’s centre of origin. The present study compares R. commune populations from Syria with those from Iran, which represents a secondary centre of origin for barley at the eastern edge of the Iranian Plateau. Results from this study also suggest that R. commune and barley did not co-evolve in the centre of origin of cultivated barley. This was evidenced by the low pathogen genetic diversity in Iran, which was even lower than in Syria, indicating that the pathogen may have been introduced recently into Iran, perhaps through infected barley seed. Hierarchical analyses of molecular variance revealed that most genetic diversity in Iran and Syria is distributed within populations, with only 14% among populations. Analyses of multilocus association, genotype diversity and mating type frequency suggest that Iranian populations reproduce predominantly asexually. The presence of both mating types on barley and uncultivated grasses suggest a potential for sexual reproduction. Rhynchosporium commune was also found on Hordeum murinum subsp. glaucum, H. vulgare subsp. spontaneum, Lolium multiflorum and, for the first time, on Avena sativa. The variety of wild grasses that can be infected with R. commune in Iran raises concerns of these grasses acting as evolutionary breeding grounds and sources of inoculum. Keywords: co-evolution, Hordeum vulgare, microsatellite, population structure, Rhynchosporium commune, Syria Introduction One of the most destructive diseases of barley, especially in areas with a cool temperate climate, is scald caused by Rhynchosporium commune. Typically, yield losses up to 45% are experienced (Brown, 1985). Furthermore, reduced grain quality has a negative effect on malting (Shipton et al., 1974). The disease is widespread in most barley-growing areas of the world and has also been reported from Iran, with yield losses in barley landraces estimated as high as 35% (Beigi et al., 2013). Barley was domesticated in the Fertile Crescent, but subsequent studies suggested a multiregional nature of cereal domestication based on DNA sequence data and phylogeographic studies (Morrell & Clegg, 2007; Saisho & Purugganan, 2007). One of the regions suggested as additional centres of origin lies at the eastern edge of the Iranian Plateau (1500–3000 km east of the Fertile Crescent; Morrell & Clegg, 2007). North African and European barley originated from the Fertile Crescent and much of Asian barley originated from the eastern edge of the Iranian Plateau (Saisho & Purugganan, 2007). However, phylogeographic analyses on the origin of *E-mail: e.seifollahi@ag.iut.ac.ir ª 2018 British Society for Plant Pathology responsive (Ppd-H1) and especially nonresponsive (ppdH1) flowering time genes show additional domestication of barley and subsequent spread to Europe from the region east of the Fertile Crescent (Jones et al., 2008). Therefore, as the Iranian Plateau represents another centre of diversity for barley, it may also be a previously overlooked centre of diversity for R. commune. Previous hierarchical gene diversity studies considering populations from Europe, the Middle East (Fertile Crescent), Australia, New Zealand, Africa and the USA suggested that the centre of origin for R. commune is in northern Europe rather than sharing an origin with barley in the Fertile Crescent (Zaffarano et al., 2006, 2009; Brunner et al., 2007; Linde et al., 2009; Kiros-Meles et al., 2011). Also, phylogeographic sequence analyses support northern Europe as the source area and it has been suggested that the pathogen recently spread to barley-growing areas of the world, now yielding founder populations in those areas (Zaffarano et al., 2009). In contrast to a source population, founder populations typically have lower genetic diversity (Linde et al., 2009). Populations of R. commune from the Fertile Crescent were found to have the lowest genetic diversity and so it was concluded that the pathogen did not track the host in the Fertile Crescent (Zaffarano et al., 2006). Also, subsequent research indicated low levels of genetic 1 2 E. Seifollahi et al. diversity on cultivated and wild barley from Syria and Jordan in the Middle East, suggesting the recent introduction of R. commune into the Middle East (KirosMeles et al., 2011). Whereas all previous studies on R. commune diversity were based on either microsatellite or sequencing markers, a study using RAPDs on Iranian isolates of R. commune found high levels of genetic diversity, possibly suggesting host tracking in Iran (Beigi et al., 2013). Unfortunately, the Iranian study did not include isolates from other regions such as the Fertile Crescent, and although it is tempting to speculate that Iran is a secondary centre of origin for R. commune, comparative studies using the same markers are required. High genotypic diversity, gametic equilibrium and equal proportions of mating types usually suggest random mating (Milgroom, 1996). As yet, the sexual form of R. commune has not been observed (McDonald, 2015), but studies using RFLP markers showed gametic equilibrium in most populations (Salamati et al., 2000). Similarly gametic equilibrium was observed in most populations in a global investigation of R. commune using sequencing markers (Zaffarano et al., 2006). In contrast R. commune populations from cultivated and wild barley displayed significant allelic association and linkage disequilibrium using microsatellite markers (Kiros-Meles et al., 2011). Significant linkage disequilibrium was also shown in R. commune populations on barley and barley grass in Australia using microsatellite markers (Linde et al., 2016). Most populations of R. commune analysed to date showed mating type equilibrium (Linde et al., 2003). Similarly, mating type frequencies in the Fertile Crescent populations from Syria and Jordan did not significantly deviate from a 1:1 ratio of mating types (Zaffarano et al., 2006). However, a survey of R. commune mating type frequency in Iran indicated that, although both mating types were detected at equal frequencies in one population, only one mating type occurred in two other populations, suggesting asexual reproduction (Arzanlou et al., 2016). The role of uncultivated hosts as evolutionary breeding grounds, alternative hosts and sources of inoculum has been shown in various fungal pathogens (Kastelein et al., 2001; Mourelos et al., 2014). Thus, uncultivated hosts play an important role in the evolution of the pathogen and the intensity of the disease. Earlier research on R. commune has shown gene flow from uncultivated barley to cultivated barley and higher virulence of isolates on uncultivated barley (Kiros-Meles et al., 2011; Linde et al., 2016). Thus, it is important to understand the role of uncultivated hosts on the occurrence of disease and distribution of mating type in Iran. In this study, the hypothesis that R. commune and cultivated barley co-evolved in barley’s centres of origin was tested further by expanding the number of R. commune populations analysed with microsatellites to include those from the suggested secondary centre of origin for barley, i.e. the Iranian Plateau. Specifically, it was investigated whether the genetic diversity (gene and genotypic diversity) of R. commune populations at the eastern edge of the Iranian Plateau was similar to genetic diversity of those in the Fertile Crescent of Iran and Syria. In addition, evidence for sexual reproduction of R. commune in Iran was explored by determining linkage disequilibrium among SSR alleles and microspatial distribution of mating types. Lastly, the possibility that uncultivated grasses next to barley fields in Iran were possible sources of R. commune inoculum was investigated. Materials and methods Fungal isolation and DNA extraction Infected barley leaves were collected in the spring of 2014 and 2015 at four locations (Eyvan, Baghmalek, Gorgan and Miandoab) in the Iranian provinces of Ilam, Khuzestan, Golestan and West Azerbaijan, respectively. The locations were selected to represent the climatic diversity of Iran. Gorgan, Eyvan and Miandoab have a warm and temperate climate with an average annual temperature of 17.8, 15.8 and 12.2 °C and 515, 433 and 405 mm annual precipitation, respectively. In contrast, Baghmalek has a local steppe climate with an average annual temperature of 22.3 °C and 377 mm of precipitation per year. In addition, three of the locations (Baghmalek, Eyvan and Miandoab) are close to or in the Fertile Crescent, whereas the Gorgan site is close to the eastern edge of the Iranian Plateau (Fig. 1). Scald-infected leaves from Gorgan, Eyvan and Baghmalek were collected following a hierarchical method as described previously (McDonald et al., 1999). Hierarchical collections encompassed a total geographic area of 200 m2 per population. Infected leaves at Miandoab were collected from six sites that were spaced at 10 m intervals along a transect. Each site comprised 1 m2. This collection encompassed a total area of 50 m2. Some samples were also collected from uncultivated grasses next to barley fields. Uncultivated grasses in Baghmalek and Gorgan grew in close proximity to barley (10–20 m) whereas the distance between barley fields and uncultivated grasses in Eyvan was 5–10 km. Infected leaf pieces were surface-sterilized for 40 s in 70% ethanol and 1% sodium hypochlorite mixed in a 1:1 ratio. Thereafter, leaves were washed twice in sterile water for 1 min and then transferred to Petri dishes containing 1% water agar (Brown, 1985). Plates were incubated in the dark at 17 °C for 10 days. Although sporulation on water agar is not profuse, it was sufficient to allow all isolates to be single-spored. In total, 204 isolates were collected and maintained on 1% water agar slants at 4 °C in the fungal collection of Isfahan University of Technology. For DNA extraction, isolates were grown on lima bean agar (LBA) plates. To obtain a mixture of spores and mycelia, 1 mL sterile distilled water was added to the plate and the mycelia and spores were scraped off with a sterile scalpel. Flasks (100 mL) containing 50 mL potato dextrose broth (PDB) and three drops of Chlobiotic 0.5% solution (0.5 g chloramphenicol per 100 mL) were inoculated with the spore and mycelial suspension and incubated on a shaker at 17 °C and continuous dark for 2 weeks. Subsequently, small mycelial tufts were harvested by filtration thorough filter paper in a Buchner funnel, frozen in liquid nitrogen and ground to a fine powder in a mortar and pestle. DNA was extracted using a CTAB method (Murray & Thompson, 1980). Microsatellite analyses In total, 160 isolates of R. commune from barley and 17 isolates from uncultivated grasses in Iran were analysed using simple sequence repeat (SSR) loci. To improve comparability between Plant Pathology (2018) R. commune populations in barley origins 3 Figure 1 Locations of four field populations of Rhynchosporium commune collected in Iran. The distances between Gorgan–Miandoab, Gorgan–Eyvan, Gorgan–Baghmalek, Miandoab–Eyvan, Miandoab–Baghmalek and Baghmalek–Eyvan are 1039, 1047, 1181, 563, 979 and 551 km respectively. Mountain ranges are coloured in brown and large water bodies in blue. studies, populations were assessed with the same 14 microsatellite loci used in previous studies (Linde et al., 2005, 2009, 2016). In addition, 28 isolates from Syria were also included as reference standards in order to compare allele sizes of Iranian isolates with previously published Syrian isolates (Linde et al., 2009). For each microsatellite marker, the 30 lL PCR mix contained 1 lL genomic DNA (3–30 ng), 3 lL 109 PCR buffer, 0.9 lL of 50 mM MgCl2, 0.3 lL of 10 mM dNTP, 0.4 lL of 10 lM fluorescently labelled and 10 lM unlabelled primer mix, 0.2 lL of 5 U lL 1 Taq DNA polymerase, 24.2 lL molecular grade water. Fluorescent labelling (with VIC, NED, FAM and PET) and PCR conditions followed those used by Linde et al. (2005). After amplification, 1–8.6 lL of PCR amplicons of each set of 1 (microsatellite loci Rhyncho_1 to 7) or 2 (microsatellite loci Rhyncho_8 to 14) were mixed and purified using 3 lL Ampure XP beads and 17 lL of binding buffer (20% PEG 8000 + 2.5 M NaCl, filter sterilized), washed three times with 180 lL 70% ethanol and resuspended in 50 lL 19 TE buffer. Subsequently, 6 lL of purified amplicons of each set were mixed with 3 lL GeneScan500LIZ size standard and 7 lL Hi-Di formamide. All samples were denatured for 3 min at 96 °C and then fragments were separated on an ABI 3130xl sequencer. Allele sizes were determined using PEAKSCANNER v. 1.0 (Applied Biosystems). Genotype richness and diversity Isolates that had the same allelic combination at 14 SSR loci were considered as clones or the same multilocus genotype (MLG). The number of MLGs and linkage disequilibrium were Plant Pathology (2018) computed using the package POPPR (Kamvar et al., 2014) in R (R Development Core Team, 2014). Genotype richness and diversity was assessed by comparing the occurrence and frequency of MLGs among populations, the number of MLGs in individual populations of each country and for all populations of each country, the number of expected MLGs at the smallest sample size based on rarefaction (eMLG). In addition, an MLG evenness index (E5; Gr€ unwald et al., 2003), Shannon–Wiener index of MLG diversity (H; Shannon & Weaver, 1949), Simpson’s index (k; Simpson, 1949) and corrected Simpson’s index (corrected k) were calculated to further compare genotype diversities among populations. Evenness is a measure of distribution of genotypes in the population. With equally abundant genotypes, evenness (E5) is equal to 1 and in a population dominated by a single genotype the evenness value is closer to zero. Distribution of MLGs was defined based on the occurrence and frequency of recurrent MLGs within populations and among regions. Allelic diversity within population Parameters of neutral genetic diversity for populations in Iran and Syria were calculated using GENALEX v. 6.501 (Peakall & Smouse, 2006). Parameters included the number of private alleles for 14 loci, the number of different alleles (Na), the effective number of alleles (Ne; Kimura & Crow, 1964), Shannon’s information index (I; Brown & Weir, 1983) and Nei’s unbiased gene diversity (uh; Nei, 1978). An online GRAPHPADQUICKCALCS software (https://www.graphpad.com/quickcalcs/ttest1/?Format=SEM) 4 E. Seifollahi et al. was used for t-tests to compare the means of Na, Ne, I and uh between pairwise comparisons of populations. The following analyses were conducted in GENALEX v. 6.501 (Peakall & Smouse, 2006) to assess population structure. A principal coordinate analysis (PCoA) based on Nei’s unbiased genetic distance (Nei, 1978) between all pairwise populations was calculated to visualize genetic distances among populations and investigate whether populations belong to divergent genetic pools. Population differentiation (ΦPT) was calculated among all pairwise populations. An analysis of molecular variance (AMOVA) was performed to estimate the distribution of genetic diversity between populations of Iran and Syria. Significance of AMOVA and population differentiation between pairwise comparisons of populations was evaluated with permutation tests of 999 permutations. Isolation by distance between populations was assessed with a Mantel test (Mantel, 1967) comparing geographic and genetic distances in pairwise-population comparisons from Iran. The degree of association among alleles (linkage disequilibrium) was assessed by calculating the index of association (IA) and rd using 999 permutations (Brown et al., 1980). The index rd is less dependent than IA on the number of loci (Agapow & Burt, 2001). Mating type identification and distribution Mating type primers designed by both Linde et al. (2003) and King et al. (2015) were used to assess mating type distribution among Iranian R. commune isolates. However, because of an ambiguity between primer dimers and the 149 bp amplicon for the MAT1-2 isolates using primers from King et al. (2015), further testing was done using mating type primers of Linde et al. (2003). Each 20 lL PCR contained 4.5 lL Taq DNA polymerase red (Amplicon), 0.5 lL of each of four primers (20 pM concentration of each primer), 12 lL sterile deionized water and 3 lL of genomic DNA (5–20 ng). PCR was always performed in a MJ Mini (Bio-Rad) thermocycler with temperature cycles as described by Linde et al. (2003). Determination of mating type idiomorphs of uncultivated grass isolates Seventeen isolates of R. commune were obtained from uncultivated grasses adjacent to barley fields and assayed for mating type idiomorphs as previously described. Microspatial distribution of mating type (within lesions, leaves, plants and plots) To survey spatial distribution of mating types within lesions, leaves and plants, heavily infected leaves and plants were collected. Three isolates were obtained from one lesion on Avena sativa from Eyvan, and one isolate each from two lesions on a leaf of Lolium multiflorum from Gorgan. Single heavily infected barley leaves were collected from Eyvan, Baghmalek and Gorgan and one isolate from each of six lesions per leaf was isolated. In Miandoab, one isolate from each of nine lesions on three barley leaves was isolated. Macrogeographical distribution of mating type (within field and regions) Mating type frequencies of the four field populations in Iran were assessed using a multiplex PCR mating type assay (Linde et al., 2003). Data analysis The distribution of mating type at different levels in the hierarchy, including lesions, leaves, plants, plots and field populations, was investigated. Significant deviation from a 1:1 mating type ratio was assessed using a v2 test (Everitt, 1992). Contingency v2 (Everitt, 1992) was calculated for populations of Iran to assess whether individual populations within a larger area (Iran) differed in mating type frequency. Results Microsatellite analyses: genotype richness and diversity The structure of R. commune populations from Iran was assessed using 14 SSR loci. In total, 112 MLGs were identified among 177 Iranian isolates from barley and uncultivated grasses. Most of the MLGs were represented by only one isolate. No MLG was shared between uncultivated grasses and barley. Among populations from barley, only three MLGs were shared, one each between Baghmalek and Eyvan, Baghmalek and Miandoab, and Eyvan and Miandoab (Table S1). The Eyvan population had the highest genotypic richness (eMLG = 28.73) and the Miandoab population the lowest (eMLG = 19.65; P ≤ 0.0139). The Miandoab population also showed the most uneven distribution of MLGs (E5 = 0.51). All of Eyvan, Baghmalek and Gorgan populations had a more even distribution of MLGs, a higher Shannon’s index and higher corrected Simpson’s index than Miandoab (Table 1). The lower MLG diversity and less even MLG distribution in Miandoab might be attributed to an MLG that was represented by 14 isolates. MLG diversity indices comparing Iranian and Syrian populations showed that Syrian populations had a higher MLG diversity (eMLGSyria = 137.53, eMLGIran = 98, P < 0.0001), higher Shannon’s index (HSyria = 5.08, HIran = 4.26), higher corrected Simpson’s index (corrected kSyria = 1.00, higher corrected kIran = 0.98), and a more even MLG distribution (E5Syria = 0.84, E5Iran = 0.57) than Iranian populations (Table 1). Allelic diversity within populations The Miandoab population had significantly (P ≤ 0.012) lower Nei’s unbiased gene diversity (uh = 0.38), number of alleles (Na = 3), effective number of alleles (Ne = 1.69) and Shannon’s information index (I = 0.65) than other populations from Iran (Table 2). All other population parameters did not differ significantly among Gorgan, Baghmalek and Eyvan populations, except for Ne where Gorgan had a significantly (P = 0.0274) lower effective number of alleles than the Baghmalek and Eyvan populations and the most private alleles (Table 2). Comparing the mean parameters of the Iranian and Syrian populations, the number of alleles (Na), effective number of alleles (Ne), Shannon’s information index and Nei’s unbiased gene diversity did not differ significantly (P ≥ 0.4640; Table 2). Plant Pathology (2018) R. commune populations in barley origins 5 Table 1 Microsatellite diversity, multilocus genotypes and indices of linkage disequilibrium for Rhynchosporium commune populations from Iran and Syria. Country Population N MLG eMLG SE H k Corrected k E5 IA* IAa* r d* r da* Iran Baghmalek Eyvan Gorgan Miandoab Total Sy1 Sy2 Sy3 Sy4 Sy6 Total 39 41 40 40 160 50 42 51 44 28 215 160 215 375 27 30 24 20 98 47 39 51 40 27 179 98 179 277 27.00 28.73 23.58 19.65 31.50 27.07 26.68 28.00 26.40 27.00 27.05 98.00 137.53 133.25 0.00 0.67 0.49 0.48 2.20 0.78 0.82 0.00 0.92 0.00 0.96 0.00 2.60 3.87 3.15 3.23 2.96 2.50 4.26 3.83 3.64 3.93 3.66 3.28 5.08 4.26 5.08 5.41 0.95 0.95 0.94 0.85 0.98 0.98 0.97 0.98 0.97 0.96 0.99 0.98 0.99 0.99 0.97 0.97 0.96 0.87 0.98 1.00 1.00 1.00 1.00 1.00 1.00 0.98 1.00 1.00 0.82 0.80 0.78 0.51 0.57 0.97 0.97 1.00 0.96 0.98 0.84 0.57 0.84 0.64 3.39 3.90 3.30 4.57 2.73 1.68 1.69 2.08 3.19 2.20 2.22 2.73 2.22 2.20 3.02 2.49 2.31 2.89 1.89 1.65 1.61 1.89 2.96 2.10 1.97 1.89 1.97 1.97 0.29 0.30 0.26 0.36 0.21 0.14 0.14 0.18 0.27 0.17 0.18 0.21 0.18 0.17 0.26 0.19 0.18 0.23 0.15 0.13 0.14 0.16 0.25 0.17 0.16 0.15 0.16 0.16 Syria Iran Syria Total N, number of individuals analysed; MLG, number of multilocus genotypes (MLG) observed; eMLG, the number of expected MLGs at the smallest sample size based on rarefaction; SE, standard error of eMLG; H, Shannon–Wiener index of MLG diversity (Shannon & Weaver, 1949); k, Simpson’s €nwald et al., 2003); IA, index of association (Brown et al., 1980); r d, stanindex (Simpson, 1949), Corrected k = (N/(N 1)) 9 k; E5, evenness (Gru dardized index of association (Agapow & Burt, 2001). *All estimates significant at P = 0.001. a Based on clone-corrected data set. Principal coordinate analysis clearly separated the Gorgan population from the other Iranian populations, with 70.86% of total genetic distance explained by axis 1 (Fig. 2). Pairwise ΦPT values were all significant (P = 0.001) but lowest between Baghmalek and Miandoab populations (ΦPT = 0.10) and highest between Gorgan and Miandoab (ΦPT = 0.31) populations (Table 3). Analysis of molecular variance (AMOVA) was implemented in order to estimate molecular variance among and within populations and between regions (Iran and Syria). Most genetic diversity (65%, ΦPT = 0.35) was distributed within populations, only 14% of genetic diversity among populations (ΦPR = 0.17) and 21% was distributed between the two countries (ΦRT = 0.21; Table 4). The Mantel test showed that the relationship between geographic and genetic distance was not significant (r = 0.169; P = 0.349; Fig. 3). Analyses of linkage disequilibrium on SSR data showed that all populations (clone-corrected and nonclone-corrected) were in significant (P = 0.001) linkage disequilibrium (Table 1). Mating type identification and distribution Iranian R. commune populations were evaluated for mating type frequencies and the deviation from a 1:1 mating Table 2 Estimates of microsatellite diversity in Rhynchosporium commune populations from Iran and Syria. Country Population Iran Baghmalek Eyvan Gorgan Miandoab Mean Total Sy1 Sy2 Sy3 Sy4 Sy6 Mean Total Syria No. private allelesa Na SE Ne SE I SE uh SE 39 41 40 40 13 28 30 5 160 50 42 51 44 28 72 13 6 11 6 19 215 76 5.00 6.07 5.36 3.00 4.86 10.43 5.14 4.50 6.71 5.57 6.00 5.59 10.71 0.66 0.69 0.46 0.42 0.32 1.10 0.75 0.58 0.93 0.80 0.60 0.34 1.29 3.16 3.86 2.72 1.69 2.86 3.73 2.30 2.19 3.61 3.38 3.92 3.08 3.26 0.51 0.42 0.28 0.12 0.21 0.52 0.29 0.25 0.53 0.44 0.54 0.20 0.47 1.15 1.40 1.14 0.65 1.09 1.50 0.92 0.89 1.33 1.24 1.40 1.16 1.34 0.15 0.13 0.11 0.09 0.07 0.15 0.15 0.13 0.16 0.17 0.14 0.07 0.17 0.59 0.70 0.59 0.38 0.57 0.66 0.48 0.47 0.64 0.62 0.70 0.58 0.60 0.06 0.05 0.05 0.05 0.03 0.05 0.07 0.06 0.06 0.07 0.06 0.03 0.06 N Na, number of different alleles; Ne, effective number of alleles (Kimura & Crow, 1964); I, Shannon’s information index (Brown & Weir, 1983); uh, unbiased diversity (Nei, 1978); SE, standard error. a Total number of private alleles for 14 loci. Plant Pathology (2018) E. Seifollahi et al. 6 4500 Eyvan Gorgan Miandoab Baghmalek 3500 Genetic distance Coord. 2 (29.14%) 4000 3000 2500 2000 y = 1.506x + 2338.6 R² = 0.028 1500 1000 Coord. 1(70.86%) 500 Figure 2 Scatter plot of the first two components of the principal coordinate analysis of pairwise population genetic distance of Rhynchosporium commune sampled from barley in Iran. Table 3 Pairwise population genetic differentiations (ΦPT) of Iranian Rhynchosporium commune populations (above diagonal), based on 999 permutations, and Nei’s unbiased genetic distance (Nei, 1978) among populations (below diagonal). Baghmalek Eyvan Gorgan Miandoab Baghmalek Eyvan Gorgan Miandoab – 0.19 0.23 0.12 0.14* – 0.52 0.35 0.20* 0.27* – 0.36 0.10* 0.26* 0.31* – *Significantly different (P = 0.001). type ratio by calculating v2. Using non-clone-corrected data, Baghmalek, Eyvan and Miandoab populations deviated significantly (P < 0.01) from a 1:1 mating type ratio, but the Gorgan population showed equal mating type ratios. Similar results were obtained for the clonecorrected data set where equal frequencies of mating types were observed for both Gorgan and Miandoab populations (Table 5). For combined populations representing Iran, mating types in both clone-corrected and non-clone-corrected data sets differed significantly (P < 0.01) from a 1:1 ratio. This was also found with a contingency v2 analysis to consider the differences in sample sizes among populations (Table 5). Microspatial distribution of mating types showed that in most cases, both mating types were recovered from a single leaf or plant (Table 6). Field plots on scales of 1 m2 with at least four isolates were assessed for mating 0 0 100 200 300 400 Geographic distance (km) 500 Figure 3 Mantel test to assess relationships between genetic and geographic distance of Iranian Rhynchosporium commune isolates. type frequencies (Table 5). Both mating types were found within 66.7% of the plots while only one mating type (MAT1-1) was found in the remaining 33.3% of the plots (Table 5). Mating type analysis of R. commune in uncultivated grasses showed that both mating types were observed on Hordeum murinum subsp. glaucum and L. multiflorum while isolates from A. sativa and isolates from Hordeum vulgare subsp. spontaneum possessed exclusively either the MAT1-1 or MAT1-2 idiomorph (Table 7). However, only two or three isolates were analysed in the latter cases. Discussion Previous studies on the origin of R. commune (Zaffarano et al., 2006; Linde et al., 2009) only included Syrian populations from Asia. Given that the eastern edge of the Iranian Plateau is suggested as a secondary centre of origin for barley (Morrell & Clegg, 2007; Saisho & Purugganan, 2007), this study included Iranian populations to extend the test of whether R. commune and barley co-evolved in barley’s centres of origin (Zaffarano et al., 2006, 2009; Brunner et al., 2007; Linde et al., 2009; Kiros-Meles et al., 2011). Furthermore, the occurrence of sexual reproduction of R. commune on both cultivated barley and uncultivated grasses in Iran was assessed. Uncultivated hosts may enhance the evolution Table 4 Hierarchical analyses of molecular variance (AMOVA) of Rhynchosporium commune populations from Iran and Syria based on simple sequence repeat (SSR) data. Source d.f. SS Between regions Among populations Within populations Total 1 7 366 374 88 86 458 633 185.51 172.14 878.87 236.52 MS Estimated variance % variance AMOVA statistic P 88 185.51 12 310.31 1 253.77 412.87 266.51 1253.77 1933.15 21 14 65 100 ΦRT = 0.21 ΦPR = 0.17 ΦPT = 0.35 0.001 0.001 0.001 P-value estimates are based on 999 permutations. d.f., degrees of freedom; SS, sum of squares; MS, mean squared deviations. Plant Pathology (2018) Plant Pathology (2018) Table 5 Distribution and frequencies of mating types in Rhynchosporium commune isolates from Iran (within plots, within fields and among fields). No. of isolates 2 a Plot number Plot size (m ) Year collected MAT1-1 MAT1-2 Baghmalek B1 B2 B3 B4 B5 B6 Total E1 E2 E3 E4 E5 E6 Total G1 G2 G3 G4 G5 G6 Total M1 M2 M3 M4 M5 M6 Total 1 1 1 1 1 1 6 1 1 1 1 1 1 6 1 1 1 1 1 1 6 1 1 1 1 1 1 6 2015 24 2014 2014 2014 2014 2014 2014 2014 2014 2014 2014 2014 2014 2014 2014 2015 2015 2015 2015 2015 2015 2015 2015 2015 2015 2015 2015 2015 31 (12) 2014–2015 6 6 7 4 4 8 35 (23) 5 7 6 4 8 8 38 (26) 3 3 2 5 1 1 15 (10) 5 6 6 4 6 4 9 (6) 119 (71) 0 2 0 2 0 0 4 (4) 1 1 1 0 0 0 3 (3) 4 4 4 3 4 6 25 (14) 1 1 1 3 0 3 40 (18) 41 (27) Eyvan Gorgan Miandoab Grand total N c 6 8 7 6 4 8 39 (27) 6 8 7 4 8 8 41 (29) 7 7 6 8 5 7 40 (24) 6 7 7 7 6 7 77.5 (66.7) 160 (9) MAT1-1d MAT1-2e v2f 100.0 75.0 100.0 66.7 100.0 100.0 89.7 83.3 87.5 85.7 100.0 100.0 100.0 92.7 42.9 42.9 33.3 62.5 20.0 14.3 37.5 83.3 85.7 85.7 57.1 100.0 57.1 22.5 74.4 0.0 25.0 0.0 33.3 0.0 0.0 10.3 (14.8) 16.7 12.5 14.3 0.0 0.0 0.0 7.3 (10.3) 57.1 57.1 66.7 37.5 80.0 85.7 62.5 (58.3) 16.7 14.3 14.3 42.9 0.0 42.9 12.10** 25.6 (27.6) – – – – – – 24.64** – – – – – – 29.88** – – – – – – 2.50 – – – – – – 2.00 38.03** (85.2) (89.7) (41.7) (33.3) (72.4) v2g Contingency v2h Contingency v2i 40.79 (3)** 18.19 (3)** 13.37** R. commune populations in barley origins Population Mating type frequency b 18.24** 0.67 19.76** a Values indicate number of MAT1-1 isolates before clone-correction and values in parentheses indicate number of MAT1-1 isolates after clone-correction. Values indicate number of MAT1-2 isolates before clone-correction and values in parentheses indicate number of MAT1-2 isolates after clone-correction. c Values indicate the number of isolates before clone-correction and values in parentheses indicate number of isolates after clone-correction. d Values indicate frequency of MAT1-1 isolates for non-clone-corrected data set and values in parentheses indicate frequency of MAT1-1 isolates for clone-corrected data set. e Values indicate frequency of MAT1-2 isolates for non-clone-corrected data set and values in parentheses indicate frequency of MAT1-2 isolates for clone-corrected data set. f 2 v value based on 1:1 ratio with 1 degree of freedom for non-clone-corrected data set. *shows significantly different at P < 0.01. g 2 v value based on 1:1 ratio with 1 degree of freedom for clone-corrected data set. *shows mating type frequencies are significantly different at P < 0.01. h 2 v value based on a contingency v2 analysis, among populations for non-clone-corrected data set. The number in parenthesis shows the degree of freedom. *shows mating type frequencies among populations are significantly different at P < 0.01. i 2 v value based on a contingency v2 analysis, among populations for clone-corrected data set. The number in parenthesis shows the degree of freedom. *shows mating type frequencies among populations are significantly different at P < 0.01. b 7 E. Seifollahi et al. 8 Table 6 Distribution and frequencies of mating types of Rhynchosporium commune isolates sampled within leaves, lesions and plants. Geographical region Plant Leaf No. of lesions No. of isolates MAT1-1 MAT1-2 Baghmalek Eyvan 1 1 2 1 2 1 1 1 1 1 1 All leaves 6 6 1 6 2 9 6 6 3 6 2 9 5 5 3 2 – 3 1 1 – 4 2 6 Gorgan Miandoab Table 7 Mating type distribution of Rhynchosporium commune isolates within uncultivated grasses. Geographical region Year collected Host No. of isolates MLGa MAT1-1 MAT1-2 Eyvan Eyvan Eyvan Baghmalek Gorgan 2014–2015 2014–2015 2014 2014 2015 Hordeum murinum subsp. glaucum Hordeum vulgare subsp. spontaneum Avena sativa H. murinum subsp. glaucum Lolium multiflorum 5 2 3 4 3 5 1 1 4 3 2 0 3 2 1 3 2 0 2 2 a Number of multilocus genotypes. of R. commune (Ali & Boyd, 1974; Linde et al., 2016) so threatening barley cultivation further. Miandoab, Eyvan and Baghmalek populations in western Iran are located near or inside the Fertile Crescent while the Gorgan population is near the eastern edge of the Iranian Plateau. Although the Gorgan population had a lower genotypic (eMLG) and MLG diversity (H) than the Baghmalek and Eyvan populations, it did not differ significantly from populations of Baghmalek and Eyvan in terms of genetic diversity, the number of alleles and gene diversity (I, Na and uh). Therefore, this study suggests that the location of R. commune populations in regard to the centre of origin for barley in Iran did not significantly affect the population genetic diversity of R. commune populations. Together, the observations in the Gorgan population of the most private alleles, a lack of shared MLGs with other populations, high genetic differentiation and high genetic distance in the PCoA analysis, suggest that this population originated from a separate genetic pool and that gene flow (dispersal) from other populations is limited. The Gorgan population is geographically separated from the other Iranian populations by geographic barriers in the form of the Alborz and Zagros Mountains. This has most likely contributed to limited gene flow leading to high genetic differentiation. The same geographic barriers have also previously been implicated in high numbers of private alleles in populations of Bipolaris orzyzae from Baghmalek compared to populations from northern Iran (Ahmadpour et al., 2017). The Miandoab population is highly clonal, with a prevalent MLG dominating the population. In contrast to the other studied populations in Iran, the prevalence and uniformity of scald distribution in Miandoab fields is low (authors’ personal observations). Of the three populations from temperate climates studied here, this population has the lowest rainfall, which may explain the low prevalence of R. commune in this region. Low genetic diversity and dominance of an MLG in Miandoab suggest a small founder population, with high levels of genetic drift due to suboptimal conditions for the disease. A nonsignificant relationship between geographic and genetic distance implies that Iranian R. commune populations are not in migration–mutation equilibrium. This may be because the populations were established quite recently from genetically diverse sources (e.g. the genetically diverse Gorgan population). Alternatively, the migration–mutation disequilibrium may be because there has been recent gene flow between some populations. The latter is suggested by shared MLGs among most populations. This recent gene flow is most probably due to human-mediated movement of infected seed. Comparison of Iranian and Syrian populations showed that Syrian populations have a higher MLG diversity and a more even MLG distribution than those from Iran, but do not differ significantly in terms of gene diversity and allelic richness. In previous studies, the lowest gene and genotype diversity were determined for the Syrian populations (Zaffarano et al., 2006; Linde et al., 2009). The even lower genotype diversity of Iranian compared to Syrian populations further disproves the hypothesis that R. commune co-evolved with barley in the host’s centres of diversity. Furthermore, the low genotype diversity in Iran suggests founder populations that were recently introduced, probably with infected seeds (Zaffarano et al., 2006; Linde et al., 2009; Kiros-Meles et al., 2011). The detection of equal mating type frequency and nonsignificant linkage disequilibrium in some populations Plant Pathology (2018) R. commune populations in barley origins using SSRs and RFLPs (Salamati et al., 2000; Linde et al., 2003, 2009; Zaffarano et al., 2006; Kiros-Meles et al., 2011), suggested R. commune is reproducing sexually, although the sexual state of the fungus has not been observed. In this study both mating types were found in all populations from Iran, even in a single leaf or plant. However, two of the four populations deviated significantly from a 1:1 mating type ratio, indicating those two populations were reproducing predominantly clonally. The equal mating type ratios in two of the populations may not necessarily be because of frequency-dependent selection; the mating types could occur in equal frequencies simply by chance. This is likely if isolates of different mating types do not differ in fitness or are not differentially selected. Asexual reproduction is further suggested for all Iranian populations because of significant linkage disequilibrium as well as low MLG and allele diversity. It is possible that the environmental factors required for sexual reproduction are suboptimal in countries with R. commune founder populations. Even though R. commune is reported as specific on Hordeum spp. and Bromus diandrus (Zaffarano et al., 2011), infection on L. multiflorum has been observed (King et al., 2013). In the present study, R. commune was isolated from H. vulgare, H. vulgare subsp. spontaneum, H. murinum subsp. glaucum, L. multiflorum and, for the first time, from A. sativa in Iran. The isolates from A. sativa and L. multiflorum could not be distinguished from R. commune by DNA sequencing and phylogenetic analysis according to the method of King et al. (2013), PCR for mating type idiomorphs or SSR analysis (data not shown). It was suggested that the sexual stage of R. commune might be formed on weedy hosts (Linde et al., 2003; McDonald, 2015). Therefore, knowledge of the pathogen on these hosts is important. The occurrence of both mating types on uncultivated grasses in barley fields provides potential for sexual reproduction on these weeds. However, these results are based on 17 isolates only and this needs to be confirmed with larger sample sizes. Alternate hosts (or weeds) as a source of primary inoculum and a host for sexual reproduction have been documented in other pathogens such as Puccinia striiformis (Jin et al., 2010), Pyrenophora tritici-repentis (Kastelein et al., 2001) and Fusarium graminearum (Mourelos et al., 2014). In addition, pathogenicity of isolates from wild grasses on cultivated barley confirmed the former as a source of scald inoculum (Linde et al., 2016). Although, in the present study, MLGs were not shared among isolates of barley and uncultivated grasses, strong evidence for pathogen migration from barley grass to barley in Australia (Linde et al., 2016) and sharing of MLGs between H. vulgare and H. spontaneum in Syria (Kiros-Meles et al., 2011) has been found, suggesting genetically undifferentiated populations on cultivated and noncultivated hosts and thus transmission of virulent MLGs between hosts (Kiros-Meles et al., 2011; Linde et al., 2016). Thus, uncultivated grasses may play an important role in R. commune evolution and epidemiology. Plant Pathology (2018) 9 This study disproved the hypothesis of a long-standing co-evolution of R. commune with barley in the Fertile Crescent, or at the eastern edge of the Iranian Plateau. All populations from Iran are considered as founder populations and show the lowest genetic diversity detected so far for R. commune. Continued quarantine regulations and restricted movement of infected barley seed within Iran will help to restrict the genetic expansion of the pathogen and assist disease management strategies by reducing disease incidence and hence evolution of the fungus. Acknowledgements E.S. thanks Isfahan University of Technology and the Iranian Ministry of Science, Research and Technology (MSRT) and the Australian National University for providing financial support. The authors also wish to thank S. Beygi, J. Shokri and A. Ahmadpour for their contribution during sampling and L. Smith (ANU) for guidance in the laboratory. The authors would like to thank the anonymous reviewers and editor for their valuable comments and suggestions to improve the quality of the paper. References Agapow PM, Burt A, 2001. Indices of multilocus linkage disequilibrium. Molecular Ecology Notes 1, 101–2. Ahmadpour A, Castell-Miller C, Javan-Nikkhah M et al., 2017. Population structure, genetic diversity and sexual state of the rice brown spot pathogen Bipolaris oryzae from three Asian countries. Plant Pathology 67, 181–92. Ali SM, Boyd WJR, 1974. Host range and physiologic specialization in Rhynchosporium secalis. Crop and Pasture Science 25, 21–31. Arzanlou M, Karimi K, Mirabi F, 2016. Some evidence for skewed mating type distribution in Iranian populations of Rhynchosporium commune, the cause of barley scald disease. Journal of Plant Protection Research 56, 237–43. Beigi S, Zamanizadeh H, Razavi M, Zare R, 2013. Genetic diversity of Iranian isolates of barley scald pathogen (Rhynchosporium secalis) making use of molecular markers. Journal of Agricultural Science and Technology 15, 843–54. Brown JS, 1985. Pathogenic variation among isolates of Rhynchosporium secalis from cultivated barley growing in Victoria, Australia. Euphytica 34, 129–33. Brown AHD, Weir BS, 1983. Measuring genetic variability in plant populations. In: Tanksley SD, Orton TJ, eds. Isozymes in Plant Genetics and Breeding, Part A. Amsterdam, Netherlands: Elsevier, 219–39. Brown AHD, Feldman MW, Nevo E, 1980. Multilocus structure of natural populations of Hordeum spontaneum. Genetics 96, 523–36. Brunner PC, Sch€ urch S, McDonald BA, 2007. The origin and colonization history of the barley scald pathogen Rhynchosporium secalis. Journal of Evolutionary Biology 20, 1311–21. Everitt BS, 1992. The Analysis of Contingency Tables. 2nd edn. Boca Raton, FL, USA: CRC Press. Gr€ unwald NJ, Goodwin SB, Milgroom MG, Fry WE, 2003. Analysis of genotypic diversity data for populations of microorganisms. Phytopathology 93, 738–46. Jin Y, Szabo LJ, Carson M, 2010. Century-old mystery of Puccinia striiformis life history solved with the identification of berberis as an alternate host. Phytopathology 100, 432–5. Jones H, Leigh FJ, Mackay I et al., 2008. Population-based resequencing reveals that the flowering time adaptation of cultivated barley 10 E. Seifollahi et al. originated east of the Fertile Crescent. Molecular Biology and Evolution 25, 2211–9. Kamvar ZN, Tabima JF, Gr€ unwald NJ, 2014. POPPR: an R package for genetic analysis of populations with clonal, partially clonal, and/or sexual reproduction. PeerJ 2, e281. Kastelein P, K€ ohl J, Gerlagh M, de Geijn HM, 2001. Inoculum sources of the tan spot fungus Pyrenophora tritici-repentis in the Netherlands. Mededelingen (Rijksuniversiteit te Gent. Fakulteit van de Landbouwkundige en Toegepaste Biologische Wetenschappen) 67, 257–67. Kimura M, Crow JF, 1964. The number of alleles that can be maintained in a finite population. Genetics 49, 725–38. King KM, West JS, Brunner PC, Dyer PS, Fitt BDL, 2013. Evolutionary relationships between Rhynchosporium lolii sp. nov. and other Rhynchosporium species on grasses. PLoS ONE 8, e72536. King KM, West JS, Fitt BDL, Dyer PS, 2015. Differences in MAT gene distribution and expression between Rhynchosporium species on grasses. Plant Pathology 64, 344–54. Kiros-Meles A, Gomez D, McDonald BA, Yahyaoui A, Linde CC, 2011. Invasion of Rhynchosporium commune onto wild barley in the Middle East. Biological Invasions 13, 321–30. Linde CC, Zala M, Ceccarelli S, McDonald BA, 2003. Further evidence for sexual reproduction in Rhynchosporium secalis based on distribution and frequency of mating-type alleles. Fungal Genetics and Biology 40, 115–25. Linde CC, Zala M, McDonald BA, 2005. Isolation and characterization of microsatellite loci from the barley scald pathogen, Rhynchosporium secalis. Molecular Ecology Notes 5, 546–8. Linde CC, Zala M, McDonald BA, 2009. Molecular evidence for recent founder populations and human-mediated migration in the barley scald pathogen Rhynchosporium secalis. Molecular Phylogenetics and Evolution 51, 454–64. Linde CC, Smith LM, Peakall R, 2016. Weeds, as ancillary hosts, pose disproportionate risk for virulent pathogen transfer to crops. BMC Evolutionary Biology 16, 101. Mantel N, 1967. The detection of disease clustering and a generalized regression approach. Cancer Research 27, 209–20. McDonald BA, 2015. How can research on pathogen population biology suggest disease management strategies? The example of barley scald (Rhynchosporium commune). Plant Pathology 64, 1005–13. McDonald BA, Zhan J, Burdon JJ, 1999. Genetic structure of Rhynchosporium secalis in Australia. Phytopathology 89, 639–45. Milgroom MG, 1996. Recombination and the multilocus structure of fungal populations. Annual Review of Phytopathology 34, 457–77. Morrell PL, Clegg MT, 2007. Genetic evidence for a second domestication of barley (Hordeum vulgare) east of the Fertile Crescent. Proceedings of the National Academy of Sciences of the United States of America 104, 3289–94. Mourelos CA, Malbran I, Balatti PA, Ghiringhelli PD, Lori GA, 2014. Gramineous and non-gramineous weed species as alternative hosts of Fusarium graminearum, causal agent of Fusarium head blight of wheat, in Argentina. Crop Protection 65, 100–4. Murray MG, Thompson WF, 1980. Rapid isolation of high molecular weight plant DNA. Nucleic Acids Research 8, 4321–6. Nei M, 1978. Estimation of average heterozygosity and genetic distance from a small number of individuals. Genetics 89, 583–90. Peakall R, Smouse PE, 2006. GENALEX 6: genetic analysis in EXCEL. Population genetic software for teaching and research. Molecular Ecology Notes 6, 288–95. R Development Core Team, 2014. R: A Language and Environment for Statistical Computing. Vienna, Austria: R Foundation for Statistical Computing. [http://www.R-project.org]. Accessed 18 May 2018. Saisho D, Purugganan MD, 2007. Molecular phylogeography of domesticated barley traces expansion of agriculture in the old world. Genetics 177, 1765–76. Salamati S, Zhan J, Burdon JJ, McDonald BA, 2000. The genetic structure of field populations of Rhynchosporium secalis from three continents suggests moderate gene flow and regular recombination. Phytopathology 90, 901–8. Shannon CE, Weaver W, 1949. The Mathematical Theory of Communication. Urbana, IL, USA: University of Illinois Press. Shipton WA, Boyd WJR, Ali SM, 1974. Scald of barley. Review of Plant Pathology 53, 839–61. Simpson EH, 1949. Measurement of diversity. Nature 163, 688. Zaffarano PL, McDonald BA, Zala M, Linde CC, 2006. Global hierarchical gene diversity analysis suggests the Fertile Crescent is not the center of origin of the barley scald pathogen Rhynchosporium secalis. Phytopathology 96, 941–50. Zaffarano PL, McDonald BA, Linde CC, 2009. Phylogeographical analyses reveal global migration patterns of the barley scald pathogen Rhynchosporium secalis. Molecular Ecology 18, 279–93. Zaffarano PL, McDonald BA, Linde CC, 2011. Two new species of Rhynchosporium. Mycologia 103, 195–202. Supporting Information Additional Supporting Information may be found in the online version of this article at the publisher’s web-site. Table S1. Multilocus genotypes occurring more than once among populations of Rhynchosporium commune from barley, based on analysis with 14 microsatellite loci. Plant Pathology (2018)