Introduction

Siberian wildrye (Elymus sibiricus L.) is the typical species of the genus Elymus, which is the largest genus in the Triticeae family and there are approximately 150 species of this genus distributed in most temperate regions of the world1. E. sibiricus is a perennial, self-fertilizing grass and an allotetraploid with the StStHH genome constitution (2n = 28)2. Moreover, E. sibiricus has the characteristics of good palatability, high yield, rich in nutrients and high digestibility, which are conducive to the growth and reproduction of domestic animals. It has been widely used as an important forage grass in cultivated pastures and natural grassland, particularly in the Qinghai-Tibet plateau, due to its excellent cold and drought tolerance, high forage quality, good adaptability to the local environment and important role in animal husbandry and environmental sustenance3. However, recent research has suggested that climate warming and excessive grazing threaten the productivity and growth of E. sibiricus4. Therefore, it is important to study its conservation and exploitation for germplasm and knowledge of the genetic diversity and population genetic structure of a species is a prerequisite for the successful management of conservation programs5. However, the transcriptomic and genomic information of E. sibiricus is very limited, hindering its use in genetic and breeding studies.

Simple sequence repeats (SSRs) or microsatellite markers are tandem repeated sequences comprising mono-, di-, tri-, tetra-, penta- or hexa-nucleotide units that possess high information content, co-dominance and locus specificity and are easy to detect compared with other molecular markers6. SSRs have been used as a powerful tool in studies of genetic variation, genetic mapping and molecular breeding7,8,9,10,11. Compared with genomic-SSRs, EST-SSRs have a higher level of transferability across related species because EST-SSRs originate from the transcribed regions in genomes and possess conserved sequences among homologous genes12.

To date, many SSRs have been developed in many plants through Illumina sequencing; these plants include alfalfa13,14, Vicia sativa15 and Indian sesame16. However, in the genus Elymus, the applications of SSRs have only been presented in a few species, including E. alaskanus5, E. caninus17, E. trachycaulus18 and E. sibiricus19. Most of these reported SSR primers were derived from barley and wheat microsatellite markers until a recent study reported that 53 genomic-SSRs were developed in E. sibiricus4. These novel SSR markers are the first characterized in E. sibiricus and will be useful for investigating genetic diversity and molecular-assisted breeding. However, these SSRs are still insufficient for genetic applications compared with those in some other plants and there are 1,281 polymorphic EST-SSRs in peanut20.

Transcriptome sequencing is an efficient method to generate genomic-level data, large EST sequences and molecular markers21. In recent years, next-generation sequencing technology has emerged as a cutting-edge approach for high-throughput sequence determination22. Additionally, it not only allows rapid and comprehensive analyses of the plant genome but also offers a cost-effective means of analysing gene transcripts23. Next-generation sequencing has been successfully and increasingly used in most plants, such as rice24, alfalfa13,25, V. sativa26 and barley27, but it has not yet been applied to research on E. sibiricus or even other species belonging to the Elymus genus.

The present study involves the first transcriptome sequencing of 11 E. sibiricus tissues using the Illumina Hiseq2000 sequencing platform. The objective of this study was to achieve a valuable sequence resource and develop some high polymorphism EST-SSR markers that would allow a better understanding of the genetic diversity in both E. sibiricus and the Elymus genus, which may be useful in modern E. sibiricus breeding programs.

Results

Sequencing and de novo assembly

The constructed cDNA library from 11 distinct tissues (Fig. 1) was sequenced and generated 84,905,976 raw reads, which contained the adapter-primer sequences, low-quality sequences and empty reads (Table 1). After a rigorous quality check and data filtering, a total of 76,686,804 high-quality clean reads with 97.97% Q20 bases were obtained. The clean reads had a total nucleotide number of 6,901,812,360 nt and the N and GC percentages for the clean reads were 0 and 54.70%, respectively. Additionally, the high-quality reads were deposited into the U.S. National Center for Biotechnology Information (NCBI) sequence read archive (SRA) database (SRX574376).

Table 1 Summary of the analysis of de novo assembled EST-SSRs for Elymus sibiricus L.
Full size table
Figure 1
figure 1

Representative tissues and samples used in this study.

(a) Callus cells (induced by young inflorescences). (b) Radicles (seven days after seed germination). (c) Young inflorescences (10 days before fertilization). (d) Tufted leaves in the tillering stage. (e) Flag leaves in the heading stage. (f) Inflorescences (five days before fertilization). (g) Old inflorescences (five days after fertilization). (h) Stems (less lignified stems, moderately lignified stems and highly lignified stems). (i) Whole seedlings (three weeks after seed germination).

Full size image

As a result, 246,164 contigs with a mean length of 268 bp and an N50 length of 356 bp were obtained after de novo assembly. The total number of unigenes with paired-end reads was 94,458, including 42,058 distinct clusters and 52,400 distinct singletons and the total length of the unigenes was 60,972,579 bp, with an average length of 645 bp and an N50 value of 942 bp. Among the 94,458 unigenes, the length of 76,500 unigenes (80.99%) ranged from 200 to 1,000 bp, the length of 17,385 unigenes (18.41%) ranged from 1,000 to 3,000 bp and 573 unigenes (0.61%) were more than 3,000 bp in length. The length distributions of the unigenes are shown in Fig. S1.

Of the 94,458 unigenes, 75,384 (79.81%) unigenes were successfully annotated in the Nr, Nt, Swiss-Prot, KEGG, COG and GO databases (Table 2) and 21,406 (22.66%) unigenes were assigned to the COG classifications (Fig. S2). After searching all 94,458 unigenes against the Nr database, 41,711 unigenes were assigned to one or more GO terms based on 62,046 Nr annotations and these terms could be grouped into the following three main categories: biological process, cellular component and molecular function (Fig. S3).

Table 2 Functional annotation of the E. sibiricus transcriptome.
Full size table

Frequency and distribution of SSRs

A total of 8,769 potential EST-SSRs were identified from 7,732 unigenes (Table 1) and 1,078 primer pairs were successfully designed. Of these unigenes, 902 unigenes contained more than one EST-SSR. An average of one EST-SSR was found every 6.95 kb and the frequency of SSRs was 8.19%. The type and distribution of the 8,769 potential EST-SSRs were then investigated. The most abundant type of repeat was tri-nucleotide repeats (5,319, 60.66%), followed by di-nucleotide (2,086, 23.79%), mono-nucleotide (444, 5.06%), penta-nucleotide (426, 4.86%), quad-nucleotide (303, 3.46%) and hexa-nucleotide (191, 2.18%) repeats (Fig. S4). As shown in Table 3, EST-SSRs with five tandem repeats (43.43%) were the most common and these were followed by six tandem repeats (25.54%), seven tandem repeats (10.38%) and four tandem repeats (6.52%), whereas the remaining tandem repeats each accounted for less than 5% of the EST-SSRs. The EST-SSR length ranged from 12 to 24 bp and 15 bp was the most frequently observed length (36.53%). Furthermore, a total of 317 motif sequence types were identified, including 25, 24, 40, 45, 90 and 93 types of mon-, di-, tri-, tetra-, penta- and hexa-nucleotide repeats, respectively. The most dominant dinucleotide repeat was AG/CT (1,264, 60.59%) and CCG/CGG was the most abundant trinucleotide repeat motif, accounting for 33.84% of these repeats and the most common repeat motif in all EST-SSRs (1,800, 20.53%) (Fig. S4, Table S1). Additionally, the GO enrichment of these 7,732 SSR-containing unigenes was executed using agriGO (http://bioinfo.cau.edu.cn/agriGO/)28 with the annotations of the assembled 94,458 unigenes as a reference. As a result, the proportion of “transcription” (GO: 0006350)-related unigenes was significantly increased, indicating that “transcription”-related unigenes were significantly enriched (Fig. S5).

Table 3 Length distribution of the EST-SSRs based on the number of nucleotide repeat units.
Full size table

Development of EST-SSR markers

Based on the SSR-containing sequences, 500 of 1,078 primer pairs were randomly selected and synthesized to investigate whether the potential EST-SSR loci that were mined were true-to-type ones for use in population genetics. Among the 500 primer pairs selected, 438 successfully amplified the genomic DNA of E. sibiricus (Table S2) and the remaining 62 pair primers failed to amplify the PCR products at various annealing temperatures. Of the 438 successful primer pairs, 369 were able to yield amplification products of the expected size and the other 69 primer pairs generated PCR products that were larger or smaller than expected. Using 45 E. sibiricus individuals from 15 accessions as PCR templates (Table S3), 112 of the 369 primer pairs were found to be polymorphic (Fig. 2, Table S4), whereas 257 were identified as monomorphic.

Figure 2
figure 2

>EST-SSR marker variations at the ES-7, ES-43, ES-50 and ES-282 loci of 15 E. sibiricus accessions.

Each accession includes three individual plants; the letter ‘M’ denotes the molecular markers, which are 200 bp, 150 bp and 100 bp (top to bottom) in ES-7, ES-43 and ES-282 and 150 bp and 100 bp in ES-50 (top to bottom).

Full size image

In total, 553 alleles were detected at the 112 polymorphic loci in the 45 genotyped individuals and the number of alleles ranged from three to nine with an average of 4.94. Estimates of the observed heterozygosity (Ho), expected heterozygosity (He) and polymorphism information content (PIC) ranged from 0 to 1, 0.51 to 0.83 and 0.39 to 0.81, with mean values of 0.49, 0.59 and 0.50, respectively. Detailed information for the 112 polymorphic primer pairs is shown in Table S4. Furthermore, PCR amplicons of three EST-SSRs from different individuals were sequenced to check the authenticity of the SSR locus. In all of the cases, the sequenced alleles from the different individuals were homologous to the original locus from which the marker was designed (Fig. 3). Using the unweighted pair-group method with arithmetic mean (UPGMA) and FreeTree program29, 15 germplasms of E. sibiricus were clustered into clusters A and B supported by bootstrap values of 0.97 and 1.00, respectively (Fig. S6). Cluster A contained 14 germplasms, whereas cluster B comprised only one germplasm, Pop3, which originated from Luqv, Gansu Province, China. Moreover, cluster A was divided into two groups, clusters C and D. The accessions in cluster C originated from Lintan, Zhuoni, Xiahe and Maqv, Gansu Province and Ruoergai, Sichuan Province, whereas those in cluster D originated from Hezuo and Xiahe, Gansu Province, indicating that there is no clear relationship between the clustering pattern and geographical distance.

Figure 3
figure 3

Comparative electropherogram analysis of three EST-SSR loci (ES-7, ES-21 and ES-192) among different populations of E. sibiricus.

Full size image

Transferability of the newly developed EST-SSR markers

As a result, 55 out of the 112 primer pairs successfully amplified all of the accessions and displayed high polymorphism (Fig. S7, Table S4). A total of 327 alleles were discovered at the 55 polymorphic loci in the 41 genotyped individuals. To evaluate the polymorphic information of 55 loci within a species, the number of alleles, Ho, He and PIC were calculated within each 13 different Elymus species (Table S5). The number of alleles ranged from 1.62 to 2.91 with an average of 2.16 and the mean values of Ho, He and PIC were 0.56, 0.40 and 0.34, respectively. In addition, the PCR amplicons of two developed EST-SSRs from different species were sequenced to assess cross-species conservation and transferability. These sequence files were analysed and the results unequivocally confirmed cross-species conservation and transferability but did not reflect significant differences (Fig. S8). The UPGMA tree revealed that 13 species were grouped into four clusters supported by bootstrap values ranging from 0.55 to 1.00 (Fig. S9). Cluster A comprised nine species, namely E. abolinii, E. gmelinii, E. antiquus, E. ciliaris, E. tschimganicus, E. burchan-buddae, E. semicostatus, E. barbicallus and E. macrochaetus. In contrast, cluster B comprised E. caninus and E. nevskii, whereas clusters C and D comprised only one species, namely E. longearistatus and E. panormitanus, respectively.

Discussion

Traditional Sanger sequencing technology can not meet the developmental needs emerging due to the progress of large-scale genomics30, whereas next-generation sequencing overcame the current limitations of Sanger sequencing with respect to throughput and costs. Next-generation sequencing has been widely used to analyse transcriptome sequencing and assembly in many plants because of its high efficiency, speed, accuracy and low cost. However, next-generation sequencing has not been applied to research on E. sibiricus. In the present study, we used the Illumina HiSeqTM 2000 platform to profile the E. sibiricus transcriptome from 11 distinct tissues and a total of 76.69 million clean reads with a length of 6,901,812,360 bp were generated. In addition, 97.97% of the clean reads had Phred quality scores at the Q20 level and an N percentage (percentage of ambiguous “N” bases) of 0, which ensure the quality of the sequencing and is consistent with the results reported in Dysosma versipellis31. Next, 94,458 unigenes were assembled from the E. sibiricus transcriptome with a mean unigene length of 645 bp. This length was longer than that reported in other studies, involving, for example, tea (402 bp)32 and sweet potato (581 bp)33, possibly because the paired-end reads (100 bp) obtained in this study were longer than those obtained in previous studies (75 bp)21. However, the length was shorter than that documented in other reports, such as those describing studies in alfalfa (803 bp)13 and seashore paspalum (970 bp)21. This result may be due to the fact that the percentage of long sequences (more than 1,000 bp) in the E. sibiricus transcriptome (19.01%) was smaller than that calculated in alfalfa (26.97%) and seashore paspalum (35.48%). Moreover, it may be related to the difference in the assembler and the parameters as well as the nature of the species. For example, a longer mean length of unigenes in alfalfa (Medicago sativa) can be explained by the well-assembled reference genome of M. truncatula.

The assembled unigenes were subjected to BLAST analysis against the known databases and a total of 75,384 (79.81%) unigenes were annotated. Additionally, 65.69% of the unigenes were identified by searching with BLASTX against the Nr database and this percentage is higher than that obtained for other plants, such as orchid (49.25%)34, sesame (53.91%)35, wax gourd (55.4%)36 and litchi (59.65%)37. Furthermore, limited genomic and transcriptomic information is currently available for E. sibiricus, influencing the annotation efficiency and some unigenes without BLAST hits may function as specific E. sibiricus genes.

In the present research, a total of 8,769 potential EST-SSRs were identified in 7,732 unigenes and the frequency of the occurrence of EST-SSRs was one SSR in every 6.59 kb, which is much higher than those obtained for tree peony (1/9.24 kb)12, alfalfa (1/12.06 kb)13, pineapple (1/13 kb)38 and lotus (1/13.04 kb)39. However, this frequency is lower than those obtained in Levant cotton (1/2.4 kb)40, castor bean (1/1.77 kb)41, radish (1/3.45 kb)42 and gerbera (1/5.6 kb)43. It has been speculated that the frequency of SSRs strongly depends on the size of the databases, SSR search criteria and mining tools used44,45. In this study, trinucleotide repeats were the most abundant type, which is consistent with the results obtained in presented studies on alfalfa13, tea32 and radish42. As shown in Fig. S4, the most dominant trinucleotide repeat motif was CCG/CGG and the same result was found in seashore paspalum21, but AAG/CTT was the most abundant type in rubber tree46 and sesame35, indicating that the EST-SSR abundance usually differs between species. Among the dinucleotide repeats, AG/CT was the most frequent motif in our dataset, which is similar to that found in sesame35, radish47 and sweet potato33. One possible explanation is that CT motifs frequently occur in 5’ UTRs and may play an important role in gene regulation21,35. Furthermore, the results of the GO enrichment analysis showed that unigenes related to the category “transcription” were significantly enriched. Similar GO analyses of the SSR-containing unigenes in our published data for alfalfa13 and V. sativa26 were performed and the results revealed that “transcription”-related unigenes were also significantly enriched, indicating that “transcription”-related unigenes may be more likely to contain SSR repeats than other unigenes48.

Of the 500 pair primers that were randomly selected for PCR validation, 438 (87.60%) produced clear bands. This PCR success rate was higher than the rates reported for alfalfa (30%)49, tree peony (47.30%)12 and rubber tree (59.8%)50. Among the successful primer pairs, 369 amplified PCR products were of the expected sizes and 69 primer pairs resulted in larger or smaller PCR products than expected. These deviations may be attributed to the presence of introns, large insertions or repeat number variations, a lack of specificity, or assembly errors33,35. Nonetheless, 112 of those 369 primer pairs were polymorphic among 45 individuals of E. sibiricus; thus, the percentage of polymorphic loci in the tested species was 30.35%, which is higher than the results reported by Lei et al. (16.06%)4 but lower than that obtained in some of previous studies46,51,52. The decreased levels of polymorphism may be due to the smaller number or close geographic origin of the materials used in the study12. In the present study, 112 EST-SSR variations were found in the coding regions, whereas five were found in genes not associated with known proteins, which is similar to the location of EST-SSR markers in common vetch15.

In addition, the number of alleles for polymorphic markers ranged from three to nine with a mean of 4.94 and these values are higher than those reported by Lei et al.4 which ranged from two to five with an average of 3.09. These results indicate that the EST-SSR markers developed in this study had a higher level of polymorphism compared with the genomic SSR markers reported by Lei et al.4. Furthermore, a series of achievements on the genetic diversity of wild E. sibiricus germplasm and populations were reported3,19,53,54 and these showed a clear demarcation between accessions from different regions. However, the dendrogram of 15 E. sibiricus accessions obtained in the present study did not show any clear geographical patterns, which may be due to the lack of adequate accession numbers and the fact that these E. sibiricus accessions were sampled from adjacent areas, where the frequent exchange of E. sibiricus germplasm may obscure an existing pattern following the geographical origin of the accessions. Therefore, the use of a higher number of accessions from close geographical locations and more individual plants per accession will be essential for verifying the genetic diversity of E. sibiricus in future studies14.

In this research, we applied 112 newly developed EST-SSR markers to 13 species of the Elymus genus to evaluate the transferability of these EST-SSR markers as well as to offer some polymorphic EST-SSR markers to 13 other species. In total, 55 of the 112 primer pairs successfully amplified all of the species and obtained moderate transferability (49.11%), which is similar to that reported by Lei et al.4 and higher than that obtained in bottle gourd (4 to 41%)55 and Cucumis (12.7%)56. This moderate transferability of EST-SSRs in E. sibiricus was partly due to the moderate conservation of the sequences flanking the SSR among these 13 related species. The average of PIC ranged from 0.21 for E. caninus to 0.47 for E. nevskii at the 55 newly developed loci. Although the average of PIC for E. nevskii (0.47) and E. panormitanus (0.46) were similar to E. sibiricus (0.48), the values of other 11 species were less than E. sibiricus (Fig. S5). The reason could be related with 45 individual plants were selected in E. sibiricus while only three individual plants were selected in other 13 Elymus species. The EST-SSR markers that were developed from E. sibiricus offer a feasible solution for both correlational research of other related species that lack molecular markers and the study of comparative genomics in the Elymus genus. As shown in the dendrogram, E. abolinii, E. ciliaris, E. tschimganicus and E. burchan-buddae were clustered into the same groups or subgroups, which is consistent with the findings obtained in previous studies57,58. However, part of our clustering results differ from those reported in previous studies of E. abolinii and E. nevskii57,59, suggesting that the use of a greater number of EST-SSR loci and a greater number of individuals per species would be essential to verify the relationship among Elymus species in future studies.

Conclusions

To the best of our knowledge, this study describes the first assembly and characterization of the transcriptome of E. sibiricus using the Illumina paired-end sequencing method. This work presents a de novo transcriptome sequencing analysis of mixed RNAs from 11 different tissues. A total of 94,458 unigenes were generated and 8,769 EST-SSRs were identified, providing a solid foundation for molecular marker development in E. sibiricus. Of these EST-SSRs, 1,078 primer pairs were successfully designed and 500 were randomly selected for further validation. A total of 112 polymorphic primer pairs successfully amplified fragments, revealing abundant polymorphisms between 15 E. sibiricus accessions. Additionally, of these 112 polymorphic primer pairs, 55 were transferable among 13 other Elymus species, indicating that these 55 newly developed primer pairs can be used with confidence in future population genetic studies of the 13 related species. This study provides a valuable sequence resource for novel gene discovery and analysis of the genetic diversity in both E. sibiricus and the Elymus genus.

Methods

Tissue material and RNA isolation

In this study, a total of 11 tissue samples from E. sibiricus were collected, including callus cells (induced by young inflorescences), radicles (seven days after seed germination), whole seedlings (three weeks after seed germination), tufted leaves in the tillering stage, flag leaves in the heading stage, less lignified stems, moderately lignified stems, highly lignified stems, young inflorescences (10 days before fertilization), inflorescences (five days before fertilization) and old inflorescences (five days after fertilization) (Fig. 1). The callus cells were induced from young spikes on solid MS medium containing 2,4-dichlorophenoxyacetic acid (3.0 mg/L) at 25 °C for 30 d under 16-h-light/8-h-dark cycles. In addition, radicles and whole seedlings were obtained through seed germination and from different individual plants, but other tissues were collected from the same plant that grew for two years. The plants of E. sibiricus were grown in a greenhouse under a 16-h-light/8-h-dark cycle at 22 °C at Lanzhou University, Lanzhou, China. All of the sampled tissues were immediately placed in liquid nitrogen and stored at −80 °C until RNA extraction. The total RNA from 11 samples was isolated using the RNeasy Plant Mini Kit (Qiagen, Cat. #74904) according to the manufacturer’s instructions. The concentration of each sample must be greater than 600 ng/μl for transcriptome sequencing, as was determined using a NanoDrop ND1000 spectrophotometer (Thermo Scientific, USA).

cDNA library construction and sequencing

To cover more tissue-specific transcripts in E. sibiricus, every sample was adjusted to the same concentration (400 ng/μl) and a total of 20 μg of RNA was pooled equally from the 11 tissues for preparation of the cDNA library. The cDNA library construction was conducted via the mRNA-Seq Sample Preparation Kit (Illumina Inc.) according to the manufacturer’s instructions. Briefly, the poly (A) mRNA was isolated by magnetic oligo (dT) beads and first-strand cDNA was detected using random hexamer primers and reverse transcriptase (Invitrogen). The short cDNA fragments were then purified using a MinElute PCR Purification Kit (Qiagen) and resolved with EB buffer (Qiagen) for end reparation and adding poly (A). Finally, sequencing adapters were ligated to the fragments. The libraries were sequenced using the Illumina HiSeq2000 sequencing platform at the BGI TECH Company (Shenzhen, China). In addition, the processing of the fluorescent images for sequence base-calling and calculation of quality values was performed using the Illumina data processing pipeline, which yielded 100 bp paired-end reads.

Sequence assembly and annotation

All of the raw reads were filtered before assembly and this filtering included the removal of poly (A/T), low-quality sequences and empty reads or reads with more than 10% of bases having Q < 20. The de novo transcriptome assembly of these clean reads was performed using the short read assembling program Trinity21. Contigs are longer fragments lacking N that were obtained by combined reads with a certain degree of overlap. Paired-end reads were used to obtain reads that were mapped back to contigs, which allows the detection of contigs from the same transcript as well as the distances between these contigs. Scaffolds were then produced via N, which represents unknown sequences between each set of two contigs that connect these contigs. Gap filling of the scaffolds was performed using paired-end reads and the obtained sequences with the lowest numbers of Ns, until the process could not be extended on either end; the resulting sequences were called unigenes.

To gain protein function annotation information for the unigenes, BLASTX alignment (e-value < 10−5) was first conducted between unigenes and protein databases, such as Nr, Swiss-Prot, KEGG and COG and the unigene sequences were then searched against the Nt database using BLASTN. According to the Nr annotation information, the GO annotation information for the unigenes was obtained using the Blast2GO program. GO functional classification of the unigenes was performed using the WEGO software after obtaining GO annotation information for all of the unigenes.

Detection of the EST-SSR markers and primer design

SSRs were detected in the assembled unigenes using the Simple Sequence Repeat Identification Tool program (MicroSatellite, http://www.gramene.org/db/markers/ssrtool)51; the SSRs were considered to contain mono-, di-, tri-, tetra-, penta- and hexa-nucleotides with minimum repeat numbers of 12, six, five, five, four and four, respectively. The EST-SSR primers were designed using BatchPrimer3 (http://probes.pw.usda.gov/cgi-bin/batchprimer3/batchprimer3.cgi)13,51 and the designed EST-SSR primers were synthesised by Shanghai Sangon Biological Engineering Technology (Shanghai, China).

EST-SSR amplification and diversity analysis

A total of 15 accessions of E. sibiricus (Table S3), which were obtained the southern Gansu Province and northwestern plateau of Sichuan Province, were selected for polymorphism analyses with the EST-SSRs and each accession contains three individual plants. Genomic DNA was separately isolated from the young leaves of three individual plants in each accession using the modified cetyltrimethylammonium bromide (CTAB) method60. The quantity and quality of the DNA samples used for PCR amplification were determined using a NanoDrop ND1000 spectrophotometer (Thermo Scientific, USA) and the concentration of each sample was adjusted to 50 ng/μl. PCR amplifications were performed in a final volume of 10 μL containing 40 ng of template DNA, 1 × PCR buffer, 2.0 mM MgCl2, 2.5 mM dNTPs, primers (4 μM each) and 0.8 U of Taq polymerase (TaKaRa, Kyoto, Japan)14. The PCR amplification conditions were as follows: initial denaturation at 94 °C for three min followed by 35 cycles of 30 s at 94 °C, 30 s at the annealing temperature (Tm) and 20 s at 72 °C and a final extension of seven min at 72 °C. The PCR products were subjected to electrophoresis on 8.0% non-denaturing polyacrylamide gels and stained using nucleic acid dye (Lot# I20826, GelStain, China). In addition, the DL500 DNA marker (TaKaRa, Kyoto, Japan) was used to determine the sizes of the PCR products. The number of alleles and the Ho, He and PIC values were calculated as previously described61. Cluster analysis was performed to generate a dendrogram using UPGMA and Nei’s unbiased genetic distance with the FreeTree program and the TreeView software package29. Bootstrap values were obtained by 1000 replicate resamplings of replacements over the loci.

Cross-species amplification

Thirteen species of the genus Elymus were chosen to evaluate the transferability of these newly developed EST-SSR markers to other related species (Table S6) and each species contains three individual plants. These species were provided by the U.S. National Plant Germplasm System (NPGS). In addition, Pop3 and Pop12, which appear in Table S3, were selected as controls. The extraction of genomic DNA, PCR amplification and diversity analysis were performed as described above.

Additional Information

How to cite this article: Zhou, Q. et al. Development and cross-species transferability of EST-SSR markers in Siberian wildrye (Elymus sibiricus L.) using Illumina sequencing. Sci. Rep. 6, 20549; doi: 10.1038/srep20549 (2016).