During the past years, the development in genetic studies and decrease of genotyping costs, have resulted in the rapid growth of the use of molecular markers (Kantartzi, 2013). Different genetic marker systems have been developed including restriction fragment length polymorphism (RFLP), randomly amplified polymorphic DNA (RAPD), amplified fragment length polymorphism (AFLP), sequence-related amplified polymorphism (SRAP), Diversity Arrays Technology (DArT), restriction-site associated DNA sequencing (RADseq), single-nucleotide polymorphism (SNP), specific-locus amplified fragment sequencing (SLAFseq), and random selective amplification of microsatellite polymorphic loci (RSAMPL). However, simple sequence repeats (SSR) also known as microsatellite has become the molecular marker of choice because of its versatility, operational flexibility, and low-cost. This has provided the foundation for its successful application in a wide range of fundamental and applicable fields, such as, genetic diversity, linkage/association mapping of gene/QTL, marker-assisted selection (MAS), variety identification, and evolution analysis (Jiao et al., 2012; Zhang Q. et al., 2012; Li et al., 2014; Shi et al., 2014; Dossa et al., 2016c).

SSRs are relatively short tandem repeats (STRs) of DNA that are widely distributed throughout whole genomic sequences (Sharma, 2007). They are present in coding regions but are more abundant in non-coding regions (Hancock, 1995). They are characterized by a high co-dominant inheritance, reproducibility, and multi-allelic variation (Morgante and Olivieri, 1993; Kalia et al., 2011). In addition, SSRs have been demonstrated to have several important biological functions including the regulation of chromatin organization, DNA metabolic processes, gene activity, and RNA structure (Li et al., 2002, 2004).

Sesame (Sesamum indicum L.) is an emerging oil crop in the world with one of the highest oil content (up to 64%) and quality (Dossa et al., 2017) among major oilseed crops. It is mainly grown in developing countries, as such, its improvement through modern molecular breeding techniques has lagged behind other oilseed crops. Up to now, different types of molecular markers have been developed and applied to sesame genotyping and breeding efforts, such as RAPD (Bhat et al., 1999; Ercan et al., 2004), inter-SSR (ISSR) (Kim et al., 2002), AFLP (Laurentin and Karlovsky, 2006), but SSR has been the preferential marker (Zhang H. et al., 2012; Zhang Y. et al., 2012; Yepuri et al., 2013; Wei et al., 2014; Dossa et al., 2016c). Although their importance in gene mapping and MAS, only few SSR markers are available for sesame research and the available ones fail to adequately represent the entire genome (Dossa, 2016). More importantly, there is no database to search for sesame SSR information at the whole genome level and to perform functional analyses, as developed in other crops such as chickpea (CicArMiSatDB: Doddamani et al., 2014), (CMsDB: Parida et al., 2015), Cucumis melo (CmMDb: Bhawna et al., 2015), tomato (TomSatDB: Iquebal et al., 2013), sugar beet (SBMDb: Iquebal et al., 2015), brassicas (Shi et al., 2014), etc.

The completion of the full nuclear genome sequence (Wang et al., 2014a) recently updated (Wang et al., 2016) and the newly sequenced landraces (Wei et al., 2015, 2016) provide a cardinal framework to identify highly informative SSRs at the whole genome level. In this study, we took advantage of these three genome sequence resources and provided not only a large amount of genome-wide informative SSR markers for large-scale genotyping and breeding research in sesame, but also a user-friendly online database for convenient search and functional analyses of SSRs.

While the integration of molecular marker technologies have significantly improved the speed and precision of modern plant breeding, the molecular research in sesame has lagged behind other model crops mainly because sesame is a minor crop often grown by smallholders in developing countries. Hence, highly informative molecular marker systems with the advantage of easy and low-cost detection are capital for sesame breeding research. Microsatellite markers constitute undoubtedly the best candidate and in this study, we identified 138,194 SSRs at the whole genome level, along with their primer pairs and genome location.

The number of SSRs identified and the SSR density were higher than previous reports in sesame, mainly, because the genomic sequences examined in this study are more important (Wei et al., 2014; Uncu et al., 2015; Dossa, 2016). Furthermore, by exploiting the latest version of the reference genome, we are able to provide the accurate position of SSRs in the sesame genome compared with previous reports. This would be helpful for gene fine-mapping and association analysis in sesame. Mono-nucleotide and di-nucleotide repeats accounted for 92.5% of the whole genome SSRs in sesame. Our results are in agreement with conclusions of Cardle et al. (2000) and Sonah et al. (2011), who identified mono-nucleotide and di-nucleotide repeats as the predominant repeat types in several plant genomes including Arabidopsis thaliana, Brachypodium distachyon, Sorghum bicolor, Oryza sativa, Medicago truncatula, and Populus trichocarpa. Similarly to previous reports of Wei et al. (2014), Uncu et al. (2015), and Dossa (2016), the distribution of A/T rich motif as the major motif is highly in accordance with the AT (0.68%) vs GC (0.32%) content in the sesame genome (Wang et al., 2014a). The same findings were also observed in Brassica rapa (Xu et al., 2010; Shi et al., 2014), Brassica napus (Cheng et al., 2009), Brassica oleracea (Li et al., 2011), cucumber (Cavagnaro et al., 2010). The high correlation of SSR number and pseudo-chromosome length suggested that this type of DNA considerably increase the length of the sesame pseudo-chromosomes.

In sesame, SSRs were more concentrated in the intergenic regions compared to genic regions which is consistent with findings in Sativa japonica (Zhang et al., 2007), maize (Xu et al., 2013), and other crops (Hancock, 1995). The landraces “Baizhima” and “Mishuozhima” exhibited similar polymorphic rates with the genome of “Zhongzhi13.” This suggested that the two landraces are much closer to each other than the elite variety “Zhongzhi13.” Our findings are in agreement with the conclusions of Wei et al. (2016) who found that the two landraces clustered together and were more closely related in the phylogenetic tree compared to “Zhongzhi13.” We further discovered that the majority of genic SSRs in the sesame genome have been found within the conserved markers between the three genotypes. This result is understandable given that SSRs within genic regions are associated with genes which constitute the genome component more conserved within species (Xiao et al., 2016). On the other hand, this implies that the conserved set of SSRs might be related to important genes which were retained during improvement from landraces to elite cultivar, as demonstrated in soybean (Zhou et al., 2015). Therefore, we infer that these genic informative microsatellites may be linked to some important biological functions and could be potential tools for sesame breeding (Lata et al., 2014; Dossa et al., 2016a,b).

In our knowledge, there are no specific molecular markers developed for other related species in the Sesamum genus. It has been demonstrated that SSR markers have a good transferability between species of the same genus or even in the same taxa (Fan et al., 2013; Buso et al., 2016; Huang et al., 2016; Thakur et al., 2017). In sesame, Uncu et al. (2015) uncovered a high rate of SSR marker transferability between the cultivated species S. indicum and the proposed wild ancestor species S. malabaricum. In addition, different sets of SSR markers developed in the cultivated sesame also yielded good amplicons in the wild-related species including Sesamum radiatum, S. angustifolium, S. latifolium, S. angolense (Zhang H. et al., 2012; Nyongesa et al., 2013; Wu et al., 2014). Based on these reports, we speculate that our developed informative SSR markers might be relevant for other wild-related species of the Sesamum genus. This will be significant for the genetic improvement of the cultivated form by exploiting the potential of the wild-related species (Dossa et al., 2017). Such transferable SSR markers between Sesamum-related species could be used for conducting macro-synteny studies, genetic mapping, and molecular breeding. Therefore, in future studies, we will employ several wild-related species of the Sesamum genus as well as a diverse panel of the cultivated sesame to evaluate the cross-species transferability of our developed SSR markers and initiate genetic researches in the wild-related species of the Sesamum genus.

Although some SSR sets have been previously identified in the sesame genome, transcriptome, etc. (Spandana et al., 2012; Zhang H. et al., 2012; Wei et al., 2014; Uncu et al., 2015; Dossa, 2016), information regarding their amplification efficiency and polymorphic potential is limited. In the present study, we took advantage of the three available sequenced genomes to screen for amplification efficiency and polymorphism potential of our developed SSR markers. This led to the identification of 79,957 informative SSR markers of which 23 selected SSRs successfully discriminated 48 genotypes from Africa based on their geographical origins. This result suggested that e-PCR is a useful strategy for a rapid screening and an effective identification of informative markers (Wang and Wang, 2016; Xiao et al., 2016). In the works of Dossa et al. (2016c), 33 polymorphic SSRs were employed to assess the genetic diversity of 96 sesame accessions from Africa and Asia which resulted in a high genetic diversity within the African germplasm. The 23 selected SSRs used in the present study to scan the diversity of 48 African accessions were all polymorphic and yielded comparable alleles number (123 vs 137) although fewer genotypes were examined here. Similarly, a high genetic diversity was also observed in the studied germplasm proving that the global 79,957 informative SSR markers could be effectively considered as the reference SSR for large-scale genotyping and molecular breeding research in sesame (Billot et al., 2012).

All SSR data were integrated into SisatBase which also supplied useful and user-friendly tools to assist users to extract more information related to SSR markers in the sesame genome. The database will be continuously updated with new versions of the sesame genome. Moreover, with the aim of extending the utility of SisatBase over other species of the Sesamum genus, new information about the cross-species transferable SSR markers as well as novel and specific SSRs for each species will be supplied in the future.

In conclusion, based on the latest version of the sesame reference genome and the two newly released genome sequences, we identified 138,194 SSRs of which 79,957 are proposed as the reference SSR for future genetics/genomics and breeding studies in sesame. All microsatellite data reported in this study are integrated into a user-friendly online database (SisatBase) for a convenient exploitation and further functional analyses. These tools will undoubtedly help to speed-up sesame molecular breeding especially in the developing countries.

KD and JY produced the sesame SSR data, developed the online database, and drafted the manuscript. KD performed the experiments. BL, NC, and XZ designed the project, supervised the works, and revised the draft manuscript. All authors have read and approved the final manuscript.

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.