Relationships among the A Genomes of Triticum L. Species as Evidenced by SSR Markers, in Iran

The relationships among 55 wheat accessions (47 accessions collected from Iran and eight accessions provided by the Institute of Plant Biology of the University of Zurich, Switzerland) belonging to eight species carrying A genome (Triticum monococcum L., T. boeoticum Boiss., T. urartu Tumanian ex Gandilyan, T. durum Desf., T. turgidum L., T. dicoccum Schrank ex Schübler, T. dicoccoides (Körn. ex Asch. & Graebner) Schweinf. and T. aestivum L.) were evaluated using 31 A genome specific microsatellite markers. A high level of polymorphism was observed among the accessions studied (PIC = 0.77). The highest gene diversity was revealed among T. durum genotypes, while the lowest genetic variation was found in T. dicoccoides accessions. The analysis of molecular variance (AMOVA) showed a significant genetic variance (75.56%) among these accessions, representing a high intra-specific genetic diversity within Triticum taxa in Iran. However, such a variance was not observed among their ploidy levels. Based on the genetic similarity analysis, the accessions collected from Iran were divided into two main groups: diploids and polyploids. The genetic similarity among the diploid and polyploid species was 0.85 and 0.89 respectively. There were no significant differences in A genome diversity from different geographic regions. Based on the genetic diversity analyses, we consider there is value in a greater sampling of each species in Iran to discover useful genes for breeding purposes.

This study was aimed to use SSR markers to estimate the level of A genome polymorphism and to identify the relationships among the species carrying A genome of the genus Triticum native to Iran.

Results and Discussion
All 31 A genome specific SSR primers yielded 410 bands (alleles) from genomic DNA of all 55 accessions of eight A genome containing Triticum species from which 316 (0.77) were polymorphic (Table 1). Table 1. Amplification of the homologous microsatellites in 55 accessions of the genus Triticum using 31 primer sets originally designed for the microsatellites of A genome (for the primer sequence see Röder et al. 1998 [36] The number of alleles per microsatellite ranged from 5 (Xgwm512 and Xgwm179) to 22 (Xgwm666) with an average of 12.8 alleles per locus (Table 1). Major allele frequency ranged from 0.13 to 0.46 averaging 0.29 (Table 1). The mean value for polymorphism information content (PIC) for all microsatellites was 0.77. The microsatellite Xgwm427 with 20 alleles had the highest (0.92) and the microsatellite Xgwm136 with 6 alleles had the lowest (0.63) PIC value (Table 1).

Genetic Similarity Analysis
The results distinguished all the 55 accessions (Figure 1), from which 46 were divided into two major groups designated as A and B in Figure 1 with 100% bootstrap support (data not shown).
These two groups, with several subgroups, were heterogeneous. The accessions of diploid species were grouped with considerable genetic similarities (except T.ura-84). Four accessions of tetraploid cultivated wheat T. durum were grouped with diploid accessions (group A, Figure 1). The group B included 14 tetraploid, 11 hexaploid and one diploid accession. The remaining eight accessions (provided by the Institute of Plant Biology, University of Zurich, Switzerland) were not grouped with the above main groups, and were clearly separated from the Iranian ones (group C, Figure 1).
At the species level (Table 2), the highest genetic similarity (0.89) was found between T. aestivum and T. durum; although T. aestivum and T. turgidum with a genetic similarity of 0.86 appeared relatively close too. The two species T. dicoccum and T. dicoccoides with 0.64 and 0.67 genetic similarity respectively, were grouped well away from the other species, indicating that the A genome in tetraploids was distant from the genome in the diploid and polyploid species.  Table 2. The analysis of genetic similarity between A genomes of diploid and diploid, diploid and tetraploid, diploid and hexaploid, tetraploid and tetraploid, and tetraploid and hexaploid pair species of 55 accessions belonging to eight Triticum L. species as revealed by SSR markers.

Analysis of Molecular Variance (AMOVA)
The main portion of genetic variance (75.56%) was attributed to the variation among populations within species. A significant genetic variation (17.44%) was calculated between different species. There was no significant difference between A genome of species with different ploidy levels ( Table 3). The accessions were mainly collected from Iran as it is considered to be a part of centre of origin of cultivated wheat [15]. The germplasms presented in the centre of origin of a taxon are considered to be more diverse than those growing at the margins of its geographic distribution. The higher genetic diversity observed in this study concur with previous reports [37,41,42,69], indicating Iran as a likely part of the centre of diversity of this genus. The differences can also be attributed to the number of accessions studied, their genetic background, and the number of markers used. A total of 410 polymorphic bands (alleles) detected in this study seem to be enough to assess genetic variation among accessions. Zhang et al. (2002) [70] discussed that the presence of 350-400 alleles is enough for objective assessment of genetic relationship between wheat accessions. From the geographic point of view, the A genome SSR differentiations were not correlated with geographic distribution ( Figure 2); however, some groupings related to both taxa and geographic origin (e.g., of T. durum accessions collected from NW and T. boeoticum subsp. boeoticum accessions collected from the West) were evident. The data were able to group geographically closely related collections ( Figure 2). There were no significant differences between diversity measures calculated for different regions.
The recognition of the tetraploid species T. durum as a subspecies of T. turgidum by Kihara (1994) [1] and Mc Fadden and Sears (1966) [71] has been followed up by some other botanists. In spite of high genetic similarity between the two taxa showed in this study, they were enough apart to be considered as a distinct species; this is supported by morphological studies [72]. Analyses at the accessional level ( Figure 1) indicated higher genetic distances among the accessions of these two species. The origin of the A genome encountered in the hexaploid wheat has always been under discussion, and its two closely related species, i.e., T. durum and T. turgidum, have been generally known as the putative A genome donors to T. aestivum [71,73]. When one considers the very close genetic similarities between the above tetra-and the latter hexaploid Triticum species provided by this study (0.89 and 0.86 respectively, see Table 2), this notion is strengthened. Considering the topology of the UPGMA dendrogram (Figure 3), the tetraploid species were divided into two groups: (1) T. turgidum and T. durum, and (2) T. dicoccoides and T. dicoccum. This can be interpreted either as different post hybridization A genome modifications among the tetraploid species or involving two different origins as the A genome donor to them or preferential gene flow occurring between pair species within each group. The SSR analysis showed a close relationship between the diploid Triticum species. Based on the calculated genetic similarities (Table 2), the A genomes occurring among the polyploids appeared to be more similar to that of the diploid species T. monococcum than the other diploids. This observation is partly in accordance with Johnson (1975) [74] and Tsunewaki (1999) [75] who pointed out that the A genome donor to T. aestivum, T. turgidum and T. durum, is T. monococcum, and to T. dicoccoides and T. dicoccum is T. urartu.

Experimental Section
A total of 47 accessions particularly collected for this study and eight accessions provided by the Institute of Plant Biology, University of Zurich (Table 4) were examined.   The materials were taxonomically identified based on Rahiminejad and Kharazyan [72]. DNA was isolated from fresh leaves of twenty individuals of each accession using CTAB DNA extraction method [76]. In order to assess genetic relationships and diversity of the species carrying A genome in the genus Triticum, 31 A genome specific SSR primers [36] were used. Markers name and other details regarding the SSR markers are presented in Table 1.
Polymerase chain reactions (PCRs) were performed based on Jakson and Matthews (2000) [77]. Briefly, the PCR was carried out in 10 µL containing 2.5 µL of the 13 ng/µL genomic DNA sample, 1 µL of 10 × reaction buffer, 0.3 µL of 30 mM MgCl 2 , 1 U Taq polymerase, 0.5 µL of 2.5 mM dNTPs and 0.3 pmol each of IRD 800 dye and IRD 700 dye labeled arbitrary primers. The amplification program consisted of the following cycles: 95 °C for 4 min, 95 °C for 30 sec, 50 °C to 60 °C (depending on the primer set) for 30 sec, 72 °C for 1.5 min and a final extension at 72 °C for 10 min. After 4 min at 95 °C , 35 cycles were performed for 30 sec at 95 °C , 30 sec at annealing temperature (50-60 °C , see Table 1), 1.5 min at 72 °C , followed by a final extension step of 10 min at 72 °C . Upon completing the PCR cycles, 0.8 µL of PCR products of each sample was loaded onto a 6% polyacrylamide sequencing gel in a Li-COR Global DNA Sequencer, and electrophoresed at 2000 V for 1.5 h.

Statistical Analyses
In order to perform a simple phenetic analysis, the presence (1) and absence (0) of each band with particular mobility was scored. Genetic similarities were calculated using Powermarker [78] and NTSYS-PC softwares from the data tables. A genetic similarity based dendrogram (based on Jaccard's similarity coefficient) [56] was constructed to show relationships between populations and species using the UPGMA clustering method implemented in NTSYS-PC [79].
For the genetic diversity analysis, allele number per locus, major allele frequencies, polymorphism information content ( alleles of a given marker, respectively [80]) and gene diversity (D i ) were calculated as genetic parameters of polymorphism for each marker in all accessions, and for all markers in the accessions collected from the same geographical regions (see Figure 2). For all the parameters, the overall estimates were calculated as the averages across all loci. The analysis of molecular variance (AMOVA; implemented in ARLEQUIN software [81]) was carried out to estimate the variance components of fingerprinting patterns, and to partition the variation among ploidy levels (groups), among the species with a same ploidy level (within groups) and among the accessions within the species.

Conclusion
Based on the results of this study, and those previously reported [82,83], it can be concluded that a hypothetical process of hybridization, polyploidization and genomic differentiation can result in the A genome of bread wheat as below:  [84] suggested that the genome of T. urartu (A u ) diverged about 0.5-3 Mya from the genome of T. monococcum (A m ). Therefore, it can be proposed that T. monococcum is the parental A genome species in the genus Triticum and the A genome of di-, tetraand hexaploid species carrying A genome are directly or indirectly originated from this species. The populations of T. monococcum grow in restricted areas and the data (not shown) indicated low seed germination ability (about 10%) of this species. Therefore, this can be interpreted that T. monococcum is facing the risk of extinction in this area and we suggest that carrying out a wider sampling with more depth across its geographic range of distribution would be vital for gene pool conservation proposes.
As Figure 1 shows the materials provided by the Institute of Plant Biology, University of Zurich are clearly separated from the Iranian accessions. These originated from Turkey, Syria, Lebanon, Greece and Iran. In dendrogram, these accessions were mainly divided according to their country of origin. The T. dicoccoides accessions originating from Iran and Turkey were grouped together; however, because of less materials being studied no interpretation can be made about their genetic relationship. Since there were no materials belonging to T. dicoccum and T. dicoccoides among the 47 accessions specifically collected from Iran for this study, no comparison was made with their Iranian genepool (Table 4).
AMOVA analyses showed that the main portion of diversity is attributed to the variation among the accessions within species, suggesting high genetic diversity within the Triticum species in Iran. We would expect that a great sampling of each species in Iran and analysis of genetic diversity would be worthwhile to reveal the genetic structure of their gene pools and to discover new useful alleles for breeding proposes.