Grasspea (Lathyrus sativus L., 2n = 14), a member of the family Leguminosae, holds great agronomic potential as grain and forage legume crop in the arid areas for its superb resilience to abiotic stresses such as drought, flood and salinity. The crop could not make much progress through conventional breeding in the past, and there are hardly any detailed molecular biology studies due to paucity of reliable molecular markers representative of the entire genome.
Using the 454 FLX Titanium pyrosequencing technique, 651,827 simple sequence repeat (SSR) loci were identified and 50,144 nonredundant primer pairs were successfully designed, of which 288 were randomly selected for validation among 23 L. sativus and one L. cicera accessions of diverse provenance. 74 were polymorphic, 70 monomorphic, and 144 with no PCR product. The number of observed alleles ranged from two to five, the observed heterozygosity from 0 to 0.9545, and Shannon’s information index ranged from 0.1013 to 1.0980, respectively. The dendrogram constructed by using unweighted pair group method with arithmetic mean (UPGMA) based on Nei's genetic distance, showed obvious distinctions and understandable relationships among the 24 accessions.
The large number of SSR primer pairs developed in this study would make a significant contribution to genomics enabled improvement of grasspea.
Grasspea (Lathyrus sativus L.) is an excellent candidate crop to provide protein and starch for human diets and animal feeds in the arid areas . It is one of the hardiest crops for adaptation to climate change because of its ability to survive drought, flood and salinity . It also plays a vital role in many low input farming systems . However, undesirable features such as prostrate plant habit, indeterminate growth, pod shattering, later maturity and presence of neurotoxin, β-N-oxalyl-L-α,β-diaminopropionic acid (β-ODAP), limit its cultivation under various agro-ecological conditions [4–6].
To date, less than 205 microsatellite (SSR) markers have been published for grasspea, and only 61 of them were characterized for size polymorphism [7–9]. Lioi et al.,  searched for the presence of SSRs with the European Molecular Biology Laboratory (EMBL) nucleotide sequence database. Ten out of 20 SSR primers were successfully amplified, and only six of them exhibited size polymorphism. In addition, Ponnaiah et al.,  searched for EST-SSRs in the National Center for Biotechnology Information (NCBI) database. Seven of the 19 Lathyrus EST-SSRs and four of the 24 Medicago EST-SSRs revealed polymorphism when screening L. sativus accessions . Sun et al.,  analyzed a total of 8,880 Lathyrus genus ESTs from the NCBI database (up to March 2011), identified 300 EST–SSR and designed primers to characterize for size polymorphism among 24 grasspea accessions. Among them 44 SSR markers were polymorphic, 117 markers monomorphic and 139 markers with no bands . Lioi sequenced 400 randomly selected clones and get 119 retrieving SSR containing sequences. 7 primer pairs produced clearly distinguishable DNA banding patterns in 10 randomly selected SSRs, The transferability of SSR markers was high among three related species of Lathyrus, namely Lathyrus cicera, Lathyrus ochrus and Lathyrus tingitanus, and the legume crop, Pisum sativum .
Next generation sequencing (NGS) technologies has become popular on its success of sequencing DNA at unprecedented speed thereby enabling impressive scientific achievements and novel biological applications [11, 12]. Next generation RNA sequencing (RNA-Seq) is rapidly replacing microarrays as the technology of choice for whole-transcriptome studies . RNA-Seq also provides a far more precise measurement of levels of transcripts and their isoforms than other methods . However, few studies solely focused on high-throughput novel microsatellite markers discovery of orphan crops via next generation sequencing [15–19].
Recently, we applied next generation sequencing to obtain high-quality putative SSR loci and flanking primer sequences inexpensively and efficiently. The novel SSR sequences were characterized and validated through successful amplification of randomly selected primer pairs across a selection of 23 grasspea accessions and one accession of its direct ancestor red pea (Lathyrus cicera) as an outgroup.
Eight grasspea (L. sativus) accessions consisted of two Chinese, two Asian, one African and three European accessions were used for the 454 sequencing.
A set of 23 grasspea (L. sativus) accessions and one red pea (L. cicera) accession were used in SSR marker testing and genetic diversity analysis. These genetic resources contained six accessions from China, seven each from Asia (including one L. cicera accession) and Europe, and four from Africa.
The seed samples were obtained from the National Genebank of China at Institute of Crop Science, Chinese Academy of Agricultural Sciences, Beijing, China. Details information is given in Additional file 1: Table S1.
DNA isolation, library preparation and 454 sequencing
The sprouts from each of the eight genotypes were collected and total genomic DNA was isolated using the CTAB method from the seven-day old seedlings grown under dark condition at 18°C. A selective hybridization with streptavidin coated bead method was used to construct SSR-enriched genomic libraries. The following eight probes were used: p(AC)10, p(GA)10, p(AAC)8, p(AAG)8, p(AAT)8, p(ATGT)6, p(GATA)6 and p(AAAT)6. Libraries quality control was conducted by randomly selecting and sequencing 186 clones. The DNA fragments were inserted into pGEM-T EASY vector, and insert fragments were validated by Sanger sequencing. If the libraries had high ratio of insert fragments and most fragments length were from 500 to 800, they were considered as high quality.
The eight SSR-enriched DNA libraries were equally pooled for pyrosequencing using the 454 Genome Sequencer FLX Titanium System at Beijing Autolab Biotechnology Co. Ltd (China). Finally, the 454 System collected the data and generated standard flow gram file (.sff) which contained raw data for all the reads. Then, grasspea.sff file was submitted to the sequence read archive (SRA) at the National Center for Biotechnology Information (NCBI) with the accession number SRX272771.
All high quality reads were processed to remove adaptor-ligated regions using the Vectorstrip program in EMBOSS software package . Moreover, in-house developed program such as: SeqTools.pl, ACGT.pl, ave_length.pl, and max.pl programs were used to analyze the total number of nucleotide A, T, C, G in all reads, the average length of all read sequences, and the maximum length read in our study.
Before SSRs searching, “clean reads” were filtered redundant at 98% sequence identity, using CD-HIT program (http://weizhong-lab.ucsd.edu/cd-hit/). A high-throughput SSR search was performed using MISA (Microsatellite identification) tool (http://pgrc.ipk-gatersleben.de/misa/). The parameters were as following: minimum SSR motif length of 10 bp and repeat length of mono-10, di-6, tri-5, tetra-5, penta-5, and hexa-5. The maximum size of interruption allowed between two different SSR in a compound sequence was 100 bp.
The MISA file was used to analyse the number of sequences containing SSRs, the number of SSRs detected, the number of SSRs starting within 200 bp of read sequences, the dominant types of SSR motifs within mono-, di-, tri-, tetra-, penta- and hexa- repeats, and the ratio of single, perfect compound and interrupted compound SSRs. These characterizations were obtained by statistical analysis from the MISA files  by a small Perl program and plotted by R language , and OpenOffice.org Calc.
Primer pairs designing
Primer pairs were designed by Primer 3.0 interface modules containing p3_in.pl Primer 3.0  and p3_out.pl files (http://pgrc.ipk-gatersleben.de/misa/primer3.html). These Perl scripts were used to normalize the format in order to design primers flanking the microsatellite locus. Amplification product sizes ranged from 100 to 300 bp. Then, the in-house developed script primer_random_pick.pl was used to gain the non-redundant primers.
Polymerase chain reactions (PCR) amplification
For each of primer pair, PCRs were performed twice, each time with a different Taq enzyme and reaction buffer. All the primer pairs were amplified in the first round experiment with 20 μl reaction volumes containing 0.5 U of TaKaRa Taq polymerase (Code No.: R001A, TaKaRa, Dalian, China), 2 μl of 10 × PCR Buffer (Mg2+ plus), 0.2 μl of dNTP (2.5 mM each), 0.4 μM primer, and 50 ng of genomic DNA. Then the no bands or weak bands primers were used in the second round PCR reaction using TAKaRa LA Taq polymerase with GC buffer (Code No.: RR02AG, TaKaRa, Dalian, China) according to the manufacturer’s instructions. SSRs were amplified on Heijingang Thermal Cycler (Eastwin, Beijing, China). Under the following conditions: 5 min initial denaturation at 95°C; 35 cycles of 30 s at 95°C, 30 s at the optimized annealing temperature (Table 1), 45 s of elongation at 72°C, and a final extension at 72°C for 10 min. PCR products were tested for polymorphism using 6% denaturing polyacrylamide gels and visualized by silver nitrate staining.
Evaluation of polymorphic primers in different accessions
288 SSR markers were randomly selected for validation feasibility and size polymorphism among 23 grasspea (L. sativus) genotypes from diverse geographical locations and one red pea (L. cicera) genotype. POPGEN1.32  software was used to calculate the observed number of alleles (Na), the level of observed heterozygosity (Ho) and the Shannon’s information index (I).
Genetic diversity analysis
Cluster analysis was conducted based on Nei’s  unbiased genetic distance, by using POPGEN1.32  software with the unweighted pair group method on arithmetic averages (UPGMA) algorithm. The resulting clusters were expressed as a dendrogram drawn by MEGA4 .
Quality control during library construction
The quality of SSR enriched grasspea library was inspected by sequencing 186 randomly selected clones. The resulting data verified that, the recombination rate was 95%, and 29 sequences contained 89 SSR motifs within the cloned sequences.
454 sequencing and characterization reads
A total of 493,364 reads were generated from the Roche 454 GS FLX Titanium platform. After adaptor removing, 370,079 read sequences were used for further analysis. The most common nucleotide was thymidine, according for 27.7% of total nucleotides, followed by adenosine (27.2%), guanine (22.2%) and cytosine (22.1%). The mean GC content was 44.3%. The average length of read sequence was 453 bp, with a maximum length of 1,162 bp (Figure 1).
Mining for SSRs (simple sequence repeats)
Firstly, we employed the program CD-HIT (http://weizhong-lab.ucsd.edu/cd-hit/) to produce a set of 280,791 non-redundant representative sequences. Then, Microsatellite identification tool (MISA) (http://pgrc.ipk-gatersleben.de/misa/) was used for microsatellite mining. As a result, 651,827 SSRs were identified in 129,886 read sequences. Among them, 115,172 read sequences contained more than one SSR. The number of SSRs presenting in compound formation was 464,271 (Table 2), which meant high proportion of SSR loci (71.2%) was located within compound repeats. The majority of identified SSRs (65.4%) were located within 200 bp from the 5′-terminus, and few of SSRs fell into the 3′-terminus (Figure 2).
SSR motifs characterizing
The identified SSRs included 995 (0.2%) mononucleotide repeat motifs, 385,385 (59.1%) dinucleotide repeat motifs, 238,752 (36.6%) trinucleotide repeat motifs, 21,200 (3.3%) tetranucleotide repeat motifs, 2,911 (0.4%) pentanucleotide repeat motifs, and 2,584 (0.4%) hexanucleotide repeat motifs (Figure 3). Thus over 95% of the motifs were di- and tri-nucleotides. The most abundant repeat motif type was (AC/GT)n, followed by (AAC/GTT)n, (AG/CT)n, (ACG/CTG)n and (ACGT/ATGC)n, respectively (Additional file 2: Figure S1, Additional file 3: Figure S2, Additional file 4: Figure S3, Additional file 5: Figure S4, Additional file 6: Figure S5, Additional file 7: FigureS6).
Compound SSR analysis
In our study, perfect SSRs (i.e., (CA)8 which were named as P2 type) were relatively less frequent (29.4%) than the compound SSRs (70.6%). In addition, there were two kinds of compound SSRs: those with interruption between two motifs (i.e., (CT)8cacacg(CA)9 which were named as C type); and those without interruption between two motifs (i.e., (GT)6(GTC)6 which were named as C* type). There were 123,444C type (93.2%) and 8,989C* type (6.8%) compound SSRs detected, which suggested the complexity of the grasspea genome.
Primer pairs designing
A total of 62,342 primer pairs flanking the SSRs were successfully designed using the public shareware Primer 3.0 (http://www-genome.wi.mit.edu/genome_software/other/primer3.html.), based on criteria of melting temperature, GC content and the lack of secondary structure. Furthermore, 50,144 non-redundant primers were achieved by in house developed programs (Additional file 8: Table S2).
Validation of SSR markers
To validate the SSR sequences, 288 SSR primer pairs were randomly selected for PCR amplification for size polymorphism among 23 grasspea (L. sativus) genotypes from diverse geographical locations and one red pea (L. cicera) genotype. After two rounds of PCR amplifications, 74 primer pairs were confirmed of being able to amplify polymorphic based across the 24 genotypes (Table 1), 70 primer pairs were confirmed to amplify only monomorphic fragments, and 144 primer pairs produced no products. The number of observed alleles (Na) ranged from two to five, the observed heterozygosity (Ho) from 0 to 0.9545, and Shannon’s information index (I) ranged from 0.1013 to 1.0980 (Table 3). These results indicate the broad utility of the SSR markers obtained from next-generation sequencing for future studies of grasspea genetics.
Genetic diversity study
To assess the efficiency of microsatellites for differentiation of L. sativus from other Lathyrus species, we chose one L. cicera accession (ELS 0246, Syria) as outgroup in the genetic diversity study. Cluster analysis based on Nei’s  genetic distance indicated good separation between L. sativus and L. cicera. Furthermore, the UPGMA procedure grouped most Chinese accessions into one cluster; come from the center of origin, Mediterranean accessions discovered the major genetic diversity in cultivated grasspea species as they spread allover, except Chinese cluster (Figure 4). These results absolutely validated the accuracy and effectiveness of our approach for developing SSR markers in grasspea with the NGS technology.
Grasspea as a potential vital crop in arid areas
Frequent drought and water shortage are worldwide problems, especially for agricultural production. Dryland agriculture plays an important role in national economy and food security. For example, in China, 55% of the total arable land, and 43% of the total food supplies are related to dryland agriculture. Grasspea is popular among the resource poor farmers in marginal areas due to the ease with which it can be grown successfully under adverse agro-climatic conditions without much production inputs. Presently at global scale, it is grown on 1.5 million ha area with 1.2 million tonnes production . In recent years, efforts are underway in many countries including China, Australia, Spain, Italy, and Canada to expand its cultivation as a break crop between cereals and as a bonus crop in fallow land because of its ability to fix large amount of atmospheric nitrogen in association with Rhizobium bacteria . However, the presence of a neurotoxin, β-N-Oxalyl-L-α,β-diaminopropionic acid (β-ODAP), renders this crop neglected and underutilized. Despite the undesirable features such as high neurotoxin, grasspea has potential as an important crop in western China and other arid areas in the world.
Mining genomic SSR loci using 454 pyrosequencing technology
The traditional methods of microsatellite development used a library-based approach for targeted SSR repeat motifs, which was time consuming, expensive, with low-throughput. Hunting in silico for EST-SSRs from public database method is an alternative way, which was cost effective and easy to access. However, the total number of ESTs from grasspea and related species was very limited since grasspea has received less attention for molecular studies.
The identification of SSRs from genomic DNA using the 454 pyrosequencing technology was relatively new and two strategies were published. These were shotgun sequencing [16–18] and SSR-enriched sequencing [15, 19]. In the present study, we used SSR-enriched sequencing technology and generated 370,079 high quality grasspea genomic reads, with an average length of 453 bp. Theoretically, the longer reads would increase our chances of successfully designing primer pairs while making it possible to identify long SSR repeats comparable to the size obtained using traditional library-based approach [18, 27]. According to the MISA analysis, 651,827 SSRs were identified from 129,886 reads. This was a very positive result, as the high ratio of SSR-containing reads and the large number of putative SSRs we obtained. Among them, di- and tri-nucleotide repeat motifs dominated the grasspea genomic sequences, similar to findings in other crops . (AC/GT)n was not only the predominant di-nucleotide repeat motif, but also the most frequent motif in the entire genome, accounting for 55.2% of the total SSRs, followed by (AAC/GTT)n, (AG/CT)n, (ACG/CTG)n, while, (AT/TA)n, (CG/GC)n, (CCG/CGG)n were rarely detected in this study. The pattern was moderately similar to that previously observed in faba bean . Furthermore, isolated and identified low proportion of unwanted repeat motifs such as (AT/TA)n, (CG/GC)n, (CCG/CGG)n would enhance the success ratio in designing primers.
Utilization of new SSR resources for ‘orphan crop’ grasspea research
Conventional breeding and phenotype research achieved great progress in improving agricultural crops in the last few years. However, grasspea was left as ‘orphan crop’ due to the lack of available genetic and genomic resources . The use of SSR markers as a conventional tool has played an important role in the study of genetic diversity, genetic linkage map, QTL mapping and association mapping, and paved the way to the integration of genomics for crop breeding.
Due to the scarcity of user-friendly, highly polymorphic molecular markers in grasspea and other Lathyrus species, high-density genetic maps were not available. In the present study, we validated 288 non-redundant SSR primer pairs and 144 (50.0%) SSR primer pairs produced amplified bands, with 74 being polymorphic, and 70 monomorphic. This very large set of potential genomic-SSR markers will facilitate the construction of high-resolution maps for positional cloning and QTL mapping.
The genus Lathyrus L. (Fabaceae) is consisted of about 160 species  distributed throughout the temperate regions of the northern hemisphere and extends into tropical East Africa and South America [31, 32]. This study, we used 74 new SSR primer pairs to clearly separate the 23 L. sativus accessions from one L. cicera accession, which is in agreement with the reported phylogenic studies of Lathyrus L. (Fabaceae) based on morphological and molecular markers [7, 31].
This study provides an extensive characterization of the SSRs in grasspea genome. For the first time, large-scale SSR-enriched sequence data was generated for the identification of SSRs and development of SSR markers to accelerate basic and applied genomics research in grasspea.
Campbell CG, Mehra RB, Agrawal SK, Chen YZ, Abd El Moneim AM, Khawaja HIT, Yadov CR, Tay JU, Araya WA: Current status and future strategy in breeding grasspea (Lathyrus sativus). Euphytica. 1993, 73 (1): 167-175.
Patto M, Skiba B, Pang E, Ochatt S, Lambein F, Rubiales D: Lathyrus improvement for resistance against biotic and abiotic stresses: from classical breeding to marker assisted selection. Euphytica. 2006, 147 (1): 133-147.
Enneking D: The nutritive value of grasspea (Lathyrus sativus) and allied species, their toxicity to animals and the role of malnutrition in neurolathyrism. Food Chem Toxicol. 2011, 49 (3): 694-709. 10.1016/j.fct.2010.11.029.
Ponnaiah M, Shiferaw E, Pe ME, Porceddu E: Development and application of EST-SSRs for diversity analysis in Ethiopian grass pea. Plant Genetic Resources. 2011, 9 (2): 276-280. 10.1017/S1479262111000426.
Sun X-L, Yang T, Guan J-P, Ma Y, Jiang J-Y, Cao R, Burlyaeva M, Vishnyakova M, Semenova E, Bulyntsev S, Zong X-X: Development of 161 novel EST-SSR markers from Lathyrus sativus (Fabaceae). Am J Bot. 2012, 99 (10): e379-e390. 10.3732/ajb.1100346.
Yang T, Bao S, Ford R, Jia T, Guan J, He Y, Sun X, Jiang J, Hao J, Zhang X, Zong X: High-throughput novel microsatellite marker of faba bean via next generation sequencing. BMC Genomics. 2012, 13 (1): 602-10.1186/1471-2164-13-602.
Thiel T, Michalek W, Varshney R, Graner A: Exploiting EST databases for the development and characterization of gene derived SSR markers in barley (Hordeum vulgare L.). Theor Appl Genet. 2003, 106: 411-422.
Zalapa JE, Brunet J, Guries RP: Isolation and characterization of microsatellite markers for red elm (Ulmus rubra Muhl.) and cross-species amplification with Siberian elm (Ulmus pumila L.). Molecular Ecology Resources. 2008, 8 (1): 109-112. 10.1111/j.1471-8286.2007.01805.x.
Shi J, Huang S, Fu D, Yu J, Wang X, Hua W, Liu S, Liu G, Wang H: Evolutionary dynamics of microsatellite distribution in plants: insight from the comparison of sequenced Brassica, Arabidopsis and other Angiosperm. PLoS One. 2013, 8 (3): e59988-10.1371/journal.pone.0059988.
We acknowledge the financial support from the Ministry of Agriculture of China, under the China Agriculture Research System (CARS-09) Program, the international cooperation projects (2010DFR30620 and 2010DFB33340), national research program (2013BAD01B03) from the Ministry of Science and Technology of China and also supported by The Agricultural Science and Technology Innovation Program (ASTIP) in CAAS. We are grateful to Dr. Dahai Wang, Liping Sun and Dr. Qi Liu (Beijing Autolab Biotechnology Co., Ltd) for their special contribution to this work. We also thank Dr. Xianfu Yin (Institute of Health Sciences, Shanghai Institutes for Biological Sciences, Chinese Academy of Sciences & Shanghai Jiao Tong University School of Medicine) for his SeqTools.pl script.
Authors and Affiliations
The National Key Facility for Crop Gene Resources and Genetic Improvement/Institute of Crop Science, Chinese Academy of Agricultural Sciences, Beijing, 100081, China
Department of Leguminous Crops Genetic Resources, N. I. Vavilov Research Institute of Plant Industry, St. Petersburg, 190000, Russia
USDA-ARS Western Regional Plant Introduction Station (WRPIS), Pullman, WA, 99164, USA
Jinguo Hu & Clarice J Coyne
International Center for Agricultural Research in the Dry Areas (ICARDA), Aleppo, 5466, Syria
Australian Temperate Field Crops Collection, Grains Innovation Park, The Department of Primary Industries, Private Bag 260, Horsham, Victoria, 3401, Australia
The Key Laboratory of Crop Gene Resources and Germplasm Enhancement on Loess Plateau of Ministry of Agriculture/Institute of Crop Germplasm Resources, Shanxi Academy of Agricultural Sciences, Taiyuan, 030031, China
The authors declare that they have no competing interests.
TY performed bioinformatic analysis, primer design and drafted the manuscript. JYJ created the SSR enriched DNA library and tested SSR markers. MB provided L. sativus accessions. JGH, CJC, SKA and RR assisted in designing experiment and preparing the manuscript. XLS and FW participated in 454 sequencing. JWC and XPH participated in quality inspection of the DNA library. JPG prepeared all the seed of L. sativus. XXZ designed and coordinated the study, and assisted in preparing the manuscript. All authors read and approved the final manuscript.
Tao Yang, Junye Jiang contributed equally to this work.
This article is published under license to BioMed Central Ltd. This is an Open Access article is distributed under the terms of the Creative Commons Attribution License (
), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.
Yang, T., Jiang, J., Burlyaeva, M. et al. Large-scale microsatellite development in grasspea (Lathyrus sativus L.), an orphan legume of the arid areas.
BMC Plant Biol14, 65 (2014). https://doi.org/10.1186/1471-2229-14-65