Characterization of Novel Di-, Tri-, and Tetranucleotide Microsatellite Primers Suitable for Genotyping Various Plant Pathogenic Fungi with Special Emphasis on Fusaria and Mycospherella graminicola

The goals of this investigation were to identify and evaluate the use of polymorphic microsatellite marker (PMM) analysis for molecular typing of seventeen plant pathogenic fungi. Primers for di-, tri-, and tetranucleotide loci were designed directly from the recently published genomic sequence of Mycospherlla graminicola and Fusarium graminearum. A total of 20 new microsatellite primers as easy-to-score markers were developed. Microsatellite primer PCR (MP-PCR) yielded highly reproducible and complex genomic fingerprints, with several bands ranging in size from 200 to 3000 bp. Of the 20 primers tested, only (TAGG)4, (TCC)5 and (CA)7T produced a high number of polymorphic bands from either F. graminearum or F. culmorum. (ATG)5 led to successful amplifications in M. graminicola isolates collected from Germany. Percentage of polymorphic bands among Fusarium species ranged from 9 to 100%. Cluster analysis of banding patterns of the isolates corresponded well to the established species delineations based on morphology and other methods of phylogenetic analysis. The current research demonstrates that the newly designed microsatellite primers are reliable, sensitive and technically simple tools for assaying genetic variability in plant pathogenic fungi.

complete fungal genomes. Their achievements offer a possibility to design new microsatellite primers suitable for multiple fungi based on the relative abundance of SSRs in genomes. In this study, we designed 20 new microsatellite primers based on SSRs of several fungal species reported by Karaoglu et al. [11] and tested their availability using 16 common fungal species.

Culture of Fungal Species and DNA Extraction
Nineteen isolates of Mycosphaerella graminicola and 16 isolates representing other fungal species originating from different geographic locations (Table 1) were used to test the suitability of the tested primers. Pyrenophora teres was kindly provided by Dr. F. Maier in Biocenter Klein Flottbek, University of Hamburg and P. tritici-repentis by Prof. Dr. Holger B. Deising, Institut für Pflanzenzüchtung und Pflanzenschutz, University of Halle, Germany. The other isolates were obtained from Institute of Phytopathology, University of Kiel, Germany. M. graminicola was cultured in a malt-yeast-extract broth and the other species in potato dextrose broth on a shaker at 22 °C, 120 rpm. M. graminicola was harvested after 5 days by centrifugation at 8000 g for 2 min and washed once with sterile distilled water. The other isolates were harvested after 5-10 days by filtration and the water residue was removed with sterile filter paper. Spores or mycelia were ground to a fine powder in liquid nitrogen using a mortar and pistil and DNA was extracted using the method described by Cenis et al. [31] with minor modifications.

Primer Design
Seventeen repeated motifs were selected to design new primers based on their abundance in eight fungal genomes by analyzing in the SSR database generated by Karaoglu et al. [11]. Genome sequence data of F. graminearum (36.2 Mbp size) available from Broad Institute (www.broadinstitute.org) and the genome sequence of Mycosphaerella graminicola (39.7 Mbp) available at Mycosphaerella graminicola v2.0, (http://genome.jgi-psf.org/Mycgr3) were used to search for mono-, di-, tri-, and tetranuclotide motifs. The aim of this search was to explore repetitive sequences that were expected to have a very high degree of polymorphism. Altogether the twenty primers listed in Table 2 were designed. The suitability of primers was tested with 19 isolates of Mycosphaerella graminicola and 16 isolates representing other fungal species (Table 1).

Visual Analysis of Banding Patterns and Gel Documentation
All PCR-amplified products were separated by electrophoresis on 1.5% agarose gels for 1.5 h at 7.0 V/cm 2 . After that, the gel was photographed using the GelSystem Flexi gel documentation system (Biostep, Jahnsdorf, Germany). The banding patterns obtained were compared for polymorphism by visual observation. Visible bands among isolates with the same migration distance were considered the same. Each distinct band was considered as polymorphic band and was scored for the presence (1) or the absence (0) among the isolates. Only reproducible bands in repeated PCR amplification were considered for analyses. A dendogram was computed (Gene Tools by SynGene) using the DICE similarity coefficient and Unweighted Pair Group Method with Arithmetic Mean (UPGMA).

Results
Based on the abundance of SSR in eight fungal genomes in the SSR database Karaoglu et al., [11], 20 repeat motifs were selected to design new primers. Altogether 20 microsatellite primers were designed (Table 2). To get the optimal annealing temperature of each primer, a PCR was performed with 12 different annealing temperatures ranging from 40-55 °C ( Table 2). The results showed that annealing temperature had different influence on these primers. For (GTCA)4, (TTTC)4, no clear difference were observed between the banding patterns produced at different annealing temperatures. (TGAC)4 amplified clearer banding pattern at higher temperatures, whereas (GTA)5, (TAC)5, (TGC)5, (TGT)5 and (GGTT)5 gave the best banding patterns at 40 °C. For (CT)7G, (AG)7C and (TAGG)4, the optimum temperature was 48 °C, at which clearer fingerprints with a higher number of bands was produced as compared with those at other annealing temperatures.
The suitability of these primers for typing fungi was tested with 16 common fungal species in comparison with five previously established primers. All the primers tested produced banding patterns for at least one fungal species. The sizes of fragments amplified ranged mostly from 300-2000 bp ( Table 2) Almost all the primers that led to clear multiple banding patterns also showed a high polymorphism among different fungal species. We compared the production of polymorphic bands by these primers between two genetically closely related Fusarium species (Table 3). Of the 20 primers tested, only (TGG)5, (GTA)5 and (GGTT)4 produced no polymorphic bands from either F. graminearum or  (Table 4). Using the Gene Tools software package and the Dice coefficient, trinucleotide microsatellite primer (ATG)5 amplified genomic DNA fragments from all sixteen isolates and clear distinction of the isolates was possible (Figure 2). Genetic similarity of the isolates was determined using the Dice's coefficient and the UPGMA clustering method. Two major groups were observed in the resulting dendrogram, which was divided into three subgroups. Between M. graminicola isolates, similarities ranged from 74 to 100% for intraspecific comparisons (Figure 3). The results showed that isolates 3 and 4 had the highest (100%) similarity suggesting that they might originate from the same clonal lineage.

Discussion
Microsatellite markers distinguish themselves as codominant, multiallelic, highly polymorphic genetic markers, requiring small amounts of DNA for straightforward PCR and gel electrophoresis analysis [32] .
In the current research, di, tri, tetra nucleotide repeat primers were used to obtain multilocus profiles using a MP-PCR assay for the evaluation of relative abundance, nature and polymorphism among 16 different common phytopathogenic fungal species. Also, multilocus profiles obtained in F. graminearum, F. culmorum, and Mycosphaerella graminicola as models for plant pathogens were evaluated. (ATG)5 amplification provided better discrimination of intraspecific diversity with M. graminicola strains, but we assumed that these may arise through primer competition, differential efficacy of primer extension at the chosen annealing temperature, in contrast to what could happen with the MP-PCR technique. Our results revealed a high degree of genetic diversity in the M. graminicola population in Germany.
The polymorphism found using primer (ATG)5 distinguished different patterns among the analyzed M. graminicola isolates, even though in some cases based in some faint bands. Thus, their application would be more inclined to diagnosis than to population genetics studies. The usual method to develop microsatellites involved hybridization-based identification of genomic clones containing possible nucleotide repeats, sequencing of the clones and designing primers that flank the repeat region. This strategy can be effective, but is costly in resources and time. Its use in the genus Mycosphaerella resulted in the identification of only nine microsatellites in M. graminicola [33]. Highly polymorphic microsatellite markers that can be multiplexed for high-throughput genetic analyses of M. graminicola and related species [34].
The highest number of polymorphic DNA fragments were produced using ISSR primers (ATC) 7 and (GTG) 5 , which detected bands in 38 M. graminicola isolates [35]. Nine pairs of single-locus microsatellite primers were used to analyze the genomic DNA of 90 isolates of M. graminicola that were collected using a hierarchical sampling procedure from different locations, leaves, and lesions within a wheat field near Saskatoon. Allelic series at eight different loci were detected [36]. Eight single-locus microsatellite markers were used to study genetic variability in a German population of M. graminicola. The results showed that there was a high level of genetic variability within the population. The genetic diversity detected within the M. graminicola population and the pattern of the distribution of genetic variability suggests that sexual reproduction occurs in the population. Chen and McDonald [37] reported that sexual reproduction played a major role in the genetic structure of populations of M. graminicola in the United States. This hypothesis is supported by the recent discovery of the sexual stage of this pathogen in Manitoba [38]. Of the 20 primers tested, only (TGG)5, (GTA)5 and (GGTT)4 were found to produce non polymorphic bands from either F. graminearum or F. culmorum while (AG)7C, (TTTC)4 and (TGT)5 amplified 100% of polymorphic bands. Microsatellite primer PCR (MP-PCR) yielded highly reproducible and complex genomic fingerprints, with several bands ranging in size from 200 to 3000 bp. Of the 20 primers tested, only (TAGG)4, (TCC)5 and (CA)7T produced a high number of polymorphic bands from either F. graminearum or F. culmorum. Based on the specific PCR fingerprints and the high interspecies variation of these banding patterns, a clear distinction between all species was possible. Among trinucleotide repeats, 53 different types of repeat motifs were identified and the CTT repeat motif was predominant in the F. graminearum genome [39]. Microsatellite-primed polymerase chain reaction using the dinucleotide and tetranucleotide primers showed clear polymorphisms among the different Fusarium spp. isolates. Microsatellite-primed PCR fingerprinting with (GACA) 4 primer discriminated between C. naterciae, C. radicalis, and C. parasitica [40].
Both primers gave similar results in phenetic analysis of genetic similarity between populations [24]. Of all the primers tested, only the microsatellite repeats (CA)8 and (GACA)4 were unable to generate visible DNA fingerprints. It is possible that these repeats are not present in the genomes of these fungi. Another possibility is that the primer annealing sites are at such a distance that amplification by Taq polymerase is impossible. These microsatellites will be useful in population genetic studies of these fungi. The results presented herein indicated that a microsatellite technique provides an efficient tool for the identification of polymorphic loci that can be used to monitor the genetic differences between phytopathogenic fungi. Upcoming research is warranted to develop more microsatellite primers with a wider array of Fusarium spp.

Conclusions
Twenty two microsatellite primers presented here will be valuable for future investigations of the population structure of plant pathogenic fungi. Considering a considerable diagnostic value of markers developed in the current study, they might be a useful tool for screening large fungal collections and generating fingerprints patterns convenient to cluster isolates into specific groups. It could be applied for differentiating individual isolates from a clonal lineage. It will also be useful to characterize undefined collections in order to group isolates into potential species groups.