Figures
Abstract
Background
Microsatellite markers or Simple Sequence Repeats (SSRs) are the most popular markers in population/conservation genetics. However, the development of novel microsatellite markers has been impeded by high costs, a lack of available sequence data and technical difficulties. New species-specific microsatellite markers were required to investigate the evolutionary history of the Euphratica tree, Populus euphratica, the only tree species found in the desert regions of Western China and adjacent Central Asian countries.
Methodology/Principal Findings
A total of 94,090 non-redundant Expressed Sequence Tags (ESTs) from P. euphratica comprising around 63 Mb of sequence data were searched for SSRs. 4,202 SSRs were found in 3,839 ESTs, with 311 ESTs containing multiple SSRs. The most common motif types were trinucleotides (37%) and hexanucleotides (33%) repeats. We developed primer pairs for all of the identified EST-SSRs (eSSRs) and selected 673 of these pairs at random for further validation. 575 pairs (85%) gave successful amplification, of which, 464 (80.7%) were polymorphic in six to 24 individuals from natural populations across Northern China. We also tested the transferability of the polymorphic eSSRs to nine other Populus species. In addition, to facilitate the use of these new eSSR markers by other researchers, we mapped them onto Populus trichocarpa scaffolds in silico and compiled our data into a web-based database (http://202.205.131.253:8080/poplar/resources/static_page/index.html).
Conclusions
The large set of validated eSSRs identified in this work will have many potential applications in studies on P. euphratica and other poplar species, in fields such as population genetics, comparative genomics, linkage mapping, QTL, and marker-assisted breeding. Their use will be facilitated by their incorporation into a user-friendly web-based database.
Citation: Du FK, Xu F, Qu H, Feng S, Tang J, Wu R (2013) Exploiting the Transcriptome of Euphrates Poplar, Populus euphratica (Salicaceae) to Develop and Characterize New EST-SSR Markers and Construct an EST-SSR Database. PLoS ONE 8(4): e61337. https://doi.org/10.1371/journal.pone.0061337
Editor: Jean-Marc Lacape, CIRAD, France
Received: November 13, 2012; Accepted: March 8, 2013; Published: April 11, 2013
This is an open-access article, free of all copyright, and may be freely reproduced, distributed, transmitted, modified, built upon, or otherwise used by anyone for any lawful purpose. The work is made available under the Creative Commons CC0 public domain dedication.
Funding: This work is financially supported by Fundamental Research Funds for the Central Universities (TD 2012-01), the Research Fund for the Doctoral Programme of Higher Education of China to FKD; Changjiang Scholars Award and “Thousand-person Plan” Award to RLW. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.
Competing interests: Co-author Rongling Wu is a PLOS ONE Editorial Board member; this does not alter the authors' adherence to all the PLOS ONE policies on sharing data and materials.
Introduction
Populus euphratica Oliv. (Salicaceae) inhabits semi-arid areas, with a natural distribution ranging from Western China and the Middle East to Spain and Western Morocco. As a desert tree, it is considered to be salinity and drought tolerant [1]. Moreover, it plays an important role in maintaining the ecological equilibrium of desert environments by counteracting the effects of dust storms, stabilizing sand, and retaining water [2]. Changes in its habitats and distribution have caused natural stands of P. euphratica to diminish over the last few decades, and some populations have disappeared [3]. P. euphratica is a useful model organism for studying salt and drought resistance mechanisms in trees because it can tolerate NaCl concentrations of up to 450 mM [4], [5] and mannitol concentrations up to 400 mM [6] in controlled experiments. Much of the work conducted on this species to date has focused on the physiological mechanisms responsible for its resistance to abiotic factors (e.g. [7] and references therein). More recent studies using Amplified Fragment Length Polymorphism (AFLP) or microsatellite analysis based on polyacrylamide gel electrophoresis (PAGE) with silver staining have provided useful information on its genetics and evolutionary history [8–0]. However, partly because of the scarcity of suitable molecular markers, much remains to be learned about the genetic factors responsible for the ability of P. euphratica to cope with various adverse environmental conditions.
SSRs (simple sequence repeats) [11] are tandem repeated DNA sequences with a repeating motif of one to six base pairs. They are also known as STRs (short tandem repeats) [12], SSLPs (simple sequence length polymorphisms) [13], and microsatellites [14]. SSRs are common and widely distributed in prokaryotic and eukaryotic genomes [15]. Because of their high rates of mutation, they tend to be highly polymorphic [16]. This makes it possible to distinguish between alleles based on their length using either classical PAGE, or by DNA sequencing when working on a larger scale or when higher accuracy is required [17]. Because of these useful properties, SSRs are the most popular markers in population genetics [17] and have been used in many applications such as gene tagging, Quantitative Trait Locus (QTL) mapping, marker assisted selection (MAS), parentage analysis, fingerprinting, and phylogenetic and taxonomic studies [18]. Two categories of SSRs can be defined based on their locations in the genome: Expressed Sequence Tag-SSRs (eSSRs) are embedded in transcribed sequences [19], while genomic SSRs (gSSRs) are embedded in both transcribed and non-transcribed sequences. Until recently, the lack of the detailed genomic information for many plants has meant that SSR marker development was primarily conducted using selective hybridization strategies, which require fragmented genomic DNA that is bound to either a nylon membrane using repeat-containing probes or to biotinylated probes that are subsequently captured on streptavidin-coated beads [20]–[23]. This is complicated by the difficulty of preparing uniformly digested genome DNA, the cumbersome nature of the molecular cloning, the limited scope for obtaining sufficiently long sequences, and the high cost of the overall process. However, thousands of ESTs can be obtained from a single cDNA library comprising fragments with lengths of 300–500 bp, from which plenty of eSSRs can be designed. Because ESTs are conserved across phylogenetically related species, eSSRs developed for one species are often highly transferable and this characteristic has made them the markers of choice in comparative mapping and QTL analysis [24]. In recent years, traditional Sanger sequencing and second generation sequencing methods have produced huge numbers of EST sequences that have been archived in public databases. These databases offer a great opportunity for identifying SSRs through data mining, especially for non-model organisms for which few genomic resources and reference sequences are available. ESTs have been used to design SSR markers for many plant taxa, including crops (rice [25], maize [26], wheat [26], [27], cotton [28], etc) and trees (coffee [29], conifers [24], [30], rubber tree [31], oak [24] and poplar [32], etc.).
Despite its ecological importance, there have been comparatively few genetic studies focusing on P. euphratica. In particular, there is a lack of genomic studies using methods such as QTL analysis, for which a relatively large battery of markers is required. To our knowledge, only 12 gSSRs have been reported for P. euphratica to date [33]. The development of SSR markers in this species has been hindered by a lack of genomic data as well as the expensive and laborious nature of the methods used to develop SSRs from genomic libraries.
The study reported herein was conducted to address these issues by: (1) screening P. euphratica ESTs to identify new SSRs, (2) developing new eSSR markers and mapping them onto P. trichocarpa scaffolds in silico, (3) screening the newly-identified polymorphic eSSR markers, (4) validating the transferability of the new polymorphic eSSRs to other related species, and (5) compiling all of the new results in a freely-accessible web-based database.
Results
Frequency and distribution of eSSRs
Our ESTs originated from two resources: the National Center for Biotechnology Information (NCBI) database (http://www.ncbi.nlm.nih.gov/nucest/) and a published paper reporting P. euphratica transcriptome data [34]. We initially obtained 13,979 ESTs from the NCBI, most of which were originally identified by studying a cDNA library derived from salt-stressed leaves of two-year-old seedlings. We also included 94,196 UniGenes from a publication describing the first deep transcriptomic analysis of living tissues from desert-grown trees and from two types of calluses (salt-stressed and unstressed) using next generation sequencing and transcriptome analysis [34].
This starting dataset of 108,175 ESTs was reduced by removing 266 sequences: 109 that were less than 100 bp in length, 148 that were similar to chloroplast genes, 3 that were similar to mitochondrial genes, and 6 that were similar to vector sequences. We then clustered 107,909 valid ESTs using the CD-HIT [35] with a 90% similarity's cut-off, which left a total of 94,090 unique ESTs with a combined length of 62,650,653 bp sequences identified.
The perl script MIcroSAtellite identification tool (MISA) was used to search for SSRs within the set of 94,090 ESTs [36] (http://pgrc.ipk-gatersleben.de/misa). A total of 4,202 SSRs were identified in 3,839 ESTs (4% of the entire EST data set), with 3,528 ESTs containing only one SSR and 311 containing multiple SSRs (Table S1). On average, every 15 kb of the P. euphratica transcripts contained one SSR. The most abundant SSR type among the 4,202 identified were trinucleotides (37%), followed by hexanucleotides (33%), dinucleotides (19%), pentanucleotides (6%), and tetranucleotides (4%; Figure S1 and Table S2). 175 compound SSRs were identified, representing around 4% of the 4,202 SSR loci. Overall, the average number of SSR repeats across all motifs was six. The dominant motif among dinucleotides SSRs was AG/CT (14% of all SSR motifs), followed by AC/GT (3%) and AT/TA (2%). The most common trinucleotides motif was AAG/CTT (10%), followed by AGG/CCT (7%) and AGC/GCT (5%). AAAN, AAAAN and AAAAAN were most common tetra-, penta-, and hexanucleotides SSR motifs respectively (Figure 1).
We also investigated the eSSRs' locations by mapping the eSSR regions on the genome of P. trichocarpa [33], the first tree to have its whole genome sequenced. In total, 3,152 (75%) of the identified P. euphratica eSSRs could be located in the P. trichocarpa genome. Most (72%) of them were located in coding regions, with only 18% being located in 5′-untranslated regions (5′-UTR) and 10% in 3′-untranslated regions (3′-UTR) (Figure 2A). Of the SSRs identified in coding regions, 47% were trinucleotides and 44% were hexanucleotides. Of the eSSRs identified in non-coding regions, 35% were trinucleotides and 30% were hexanucleotides, with most of them being embedded in 5′ UTRs (63%) (Figure 2B).
(A). Abundance of eSSRs with different motif sizes and their distributions in non-coding and coding region (B).
Marker development
We were able to design primers for 3,069 (73%) of the 4,202 eSSRs that were identified. In most cases where it was not possible to design primers for a given eSSR, the failure was because its flanking sequences were too short. Primers were designed using a two-step process. First, for each eSSR, we selected the three primer pairs that achieved the best scores according to the design criteria of the software used. Data on the primers, including the sequences of the forward/reverse primers as well as their melting temperatures (Tm), expected amplicon size, corresponding EST ID, type of SSR, amplification status, and the degree of polymorphism of the target locus are provided in the supplemental data (Table S3). Second, we selected 673 primer pairs at random (one pair for each eSSR) for 413 di-, 154 tri-, one tetra-, 15 penta- and 90 hexanucleotides SSRs to perform further validation. Amplicons were successfully generated using 575 (85%) of the tested primer pairs.
Polymorphism and the genetic diversity of eSSRs in P. euphratica
To evaluate the polymorphism of the identified eSSR markers, we performed experiments using the 575 primer pairs that successfully generated amplicons. The 575 SSR loci targeted by these primers comprise 315 dinucleotides, 154 trinucleotides, one tetranucleotides, 15 pentanucleotides, and 90 hexanucleotides repeats. We used these primer pairs to amplify six, 12 or 24 DNA samples of P. euphratica individuals that were randomly selected from a natural population in Xinjiang Province, Northwestern China. 464 (80.7%) of the amplified eSSRs showed some degree of polymorphism, with Polymorphism Information Content (PIC) [37] values ranging from 0.062 to 0.895, allele numbers (NA) ranging from two to 13, and expected heterozygosity (HE) values ranging from 0.067 to 0.942 (Table S4). Of the polymorphic eSSRs, 260 (56%) were dinucleotides eSSRs, 130 (28%) were trinucleotides eSSRs, 59 (13%) were hexanucleotides eSSRs, 14 (3%) were pentanucleotides eSSRs and one was a tetranucleotides.
Cross species transferability of eSSRs in Populus
In order to generate transferable eSSR markers for genetic studies on other Populus species, a total of 302 polymorphic primer pairs developed for P. euphratica were used to amplify DNA samples from nine other Populus species belonging to five sections (Aigeiros, Leucoides, Populus, Tacamahaca, Turanga) of the genus. The tested eSSRs exhibited high levels of transferability in P. pruinosa (>90%), P. ussuriensis (78%), P. trichocarpa (77%) and P. tomentosa (73%). Moderate levels of transferability were observed in P. simonii and P. wilsonii (62% and 51%, respectively; Figure 3 and Table S5). However, the levels of transferability for P. pseudoglauca, P. nigra var. thevestina and P. nigra var. italica were rather low (37%, 33% and 31% respectively).
The uppermost figure shows the transferability of the 305 polymorphic eSSRs over nine species of Populus and P. euphratica. The lower figure shows a Neighbor-Joining tree for the tested species based on eSSR transferability. The box in the upper right-hand side of the figure shows the symbols used to indicate species belonging to different sections of the genus Populus.
To determine whether these transferability values might reflect the phylogenetic relationships between species in the Populus genus, we constructed a Neighbor joining tree using transferability data for 302 primers (Figure 3). Inspection of the tree revealed two groups of phylogenetically closely related species: P. pruinosa and P. euphratica in section Turanga Bge, and P. nigra var. thevestina and P. nigra var. italica in section Aigeiros Duby. However, no relationships involving species in the other three sections of the genus were identified.
In silico mapping of eSSRs onto the P. trichocarpa genome and their functional annotation
The selection of evenly spaced markers on linkage groups/chromosomes of interest is typically a crucial issue in molecular breeding, and especially in linkage group construction. Previous genomic and genetic studies have revealed high levels of collinearity for markers (including SSR markers) between Populus species [32], [38]. P. trichocarpa was the first perennial plant to have its whole genome sequenced. We therefore mapped the newly-developed P. euphratica eSSRs onto the scaffolds of P. trichocarpa. Using sequence homology and digital mapping, we were able to putatively assign 3,152 eSSRs onto 19 pseudo-chromosomes (major scaffolds) and other un-anchored scaffolds of P. trichocarpa (Figure 4 and Table S6). We observed that the eSSRs were unevenly distributed both among and within chromosomes. In addition, eSSRs deserts were also observed on several chromosomes, especially at the end of each scaffold. These phenomenons, which may be due to the differential selection between paralogous chromosomes, are consistent with observations made in previous eSSR-based studies on Populus [39].
Gray columns represent 19 pseudo-chromosomes of P. trichocarpa. P. euphratica eSSRs are arranged on these pseudo-chromosomes, with green lines representing experimentally validated EST-SSRs and red lines indicating EST-SSRs that are not validated.
The 3,839 SSR-containing ESTs were annotated against the GO protein database using Blast2GO [40]. GO assignments were obtaine for 2,202 SSR-containing ESTs. Based on their GO annotations, most of the SSR-containing ESTs were involved in cellular processes (∼21%), metabolism (∼18%), responses to stimuli (∼18%) and biological regulation (17%) (Figure S2 A). More than half of the cellular component GO annotations indicated that the corresponding ESTs were expressed in the cell (Figure S2B), with macromolecular complexes accounting for ∼6% and the membrane-enclosed lumen for ∼2%. The major molecular functions identified from the annotations were related to binding (∼53%) and catalytic activity (∼33%) (Figure S2C).
To determine whether any GO categories were overrepresented among SSR-containing ESTs relative to the case for all ESTs, we performed Fisher's Exact Test (FDR cutoff of p<0.05) using Blast2GO (Figure 5). Some categories were indeed substantially overrepresented among the SSR-containing ESTs, including transcription, regulation of transcription, transcription complexes, RNA biosynthesis and macromolecular biosynthesis (Figure 5). The most heavily overrepresented GO term in the SSR-containing ESTs was sequence-specific DNA binding transcription factor activity (Figure 5), which was typically associated with genes involved in the transcription process. Similar results were obtained in previous studies on Quercus mongolica [41] and Cryptomeria japonica [30]. However, no GO category was overrepresented among SSR-containing ESTs by more than a factor of 2.5 relative to its representation among all ESTs.
The Y axis represents significantly enriched GO terms and the X axis represents the relative frequency of the GO term.
A web database for P. euphratica eSSR markers
To facilitate wider uses of the new P. euphratica markers, we constructed a web-based, downloadable and searchable database for P. euphratica eSSR markers that also contains transferability data for some of the validated polymorphic markers in nine other Populus species. The database can be accessed freely via a web interface (http://202.205.131.253:8080/poplar/resources/static_page/index.html). In addition, we have created a graphical tool that is accessible on the homepage and displays 19 P. trichocarpa pseudo-chromosomes (Figure 4) showing the positions of the eSSR markers. This tool enables users to select locations of interest and directly obtain information on the relevant eSSRs. It is possible to search for EST sequences, primer information, and specific annotations in several ways, including searches based on ID, location (CDS or UTR), SSR motif, scaffold, etc.
Discussion
Frequency, distribution and location of eSSRs of P. euphratica
Due to their co-dominance and high level of polymorphism, SSR markers are the most popular markers in studies on population genetics, QTL mapping, and related areas. EST sequences are the main resources from which new SSR markers are developed [42]. The increasing availability of transcriptome data produced using next generation sequencing methods has greatly enhanced the scope for developing markers in this way. However, both short reads and redundancy are common in next generation data, which can be problematic because only non-redundant ESTs can be used to generate unique SSR markers.
In this study, we identified 4,202 eSSRs from 94,090 non-redundant EST sequences. It thus seems that eSSRs account for around 4% of the transcriptome in P. euphratica. To compare the SSR frequency in P. euphratica with that in other plants, we used the same software (SSRLocator [43]) and same cut-off as were previously used by Victoria et al. [44], focusing on SSR lengths of >20 bp. In our dataset, 2,445 SSRs satisfied these criteria, corresponding to a frequency of around 2.6%. This is comparable to the situation in Chlamydomonas reinhardtii (about 2.4%) and Syntrichia ruralis (about 2.7%). To compare the SSR density of P. euphratica with that in other plant species, we used the same criteria as described in Cardle et al. [45] with thresholds of 7, 5, 4 and 3 repeats for di-, tri-, tetra-, and pentanucleotides SSRs, respectively. We identified 6,573 SSRs in total with one SSR for roughly every 9.5 kb (1/9.5 kb) of EST sequence data in P. euphratica. This density is lower than that in rice (1/3.4 kb), soybean (1/7.4 kb), and maize (1/8.1 kb). However, it is higher than those previously reported for tomato (1/11.1 kb), Arabidopsis (1/13.8 kb), poplar (1/14.0 kb) and cotton (1/20.0 kb). Although the SSR density in P. euphratica determined in this work was greater than that reported for poplar by Cardle et al., the two results are similar in magnitude. The difference may be due to the relatively small number of EST sequences (1,880 kb) examined by Cardle et al. compared to the dataset used in this work (62,650 kb).
Previous studies have identified differences in the abundance of different motifs in eSSRs [46]. In general, trinucleotides repeats are the most common, followed by either di- or tetranucleotides repeats [42]. As in other higher plants, trinucleotides repeats are the most common SSR motifs in P. euphratica, accounting for 37% of all its SSRs. This is consistent with the fact that changes in the number of trinucleotides repeats in a coding region will not cause a frame shift [47]. As expected, an analysis of the eSSR distribution showed that trinucleotides repeats were most commonly found in coding regions. We also found that hexanucleotides repeats were frequently found in coding regions, with a similar ratio to trinucleotides repeats. This is further evidence for the positive correlation between the incidence of SSRs in a coding region and the likelihood that a change in the SSR's repeat number will cause a frame shift. In contrast to trinucleotides repeats, our results showed that dinucleotides repeats are preferentially located in non-coding regions, especially the 5′UTR. Interestingly, we found that most of the eSSRs in non-coding regions were located in 5′UTR. This has also been observed in oaks [24] and rice [48], and was proposed to be due to the regulatory role of the 5′UTR in gene expression.
To further explore eSSR motif bias in P. euphratica, we analyzed the abundance of each motif type in detail. AG/CT was the most common motif identified, accounting for nearly 10% of the total pool of eSSRs pool. Similar findings have been reported for several other plants, including Arabidopsis, maize, soybean, wheat and rice [46], [49]. This seems to be common in the plant kingdom, suggesting that the AG/CT type of motif may have been under positive selective pressure during eSSR evolution. One possible reason for this phenomenon was suggested by Kantety et al. [26] who noted that the codons that incorporate AG/CT i.e. GAG, AGA, UCU and CUC correspond to Arg, Glu, Ala and Leu respectively. Of these, Ala and Leu are relatively common in proteins, occurring at frequencies of 8% and 10%. Although AG/CT was the most common single motif, trinucleotides repeats were the most abundant class of eSSRs among the studied ESTs, and AAG/CTT was the most frequent type of trinucleotides repeat. This is consistent with results for dicots such as oak [24], castor bean [50] and Arabidopsis [44]. However, in monocots such as rice, CCG/CGG is the most common trinucleotides repeat [46].This difference has been attributed to the high GC content of grass genomes in several studies [49]. In this work, we did not find any GC-containing dinucleotides repeats and only 76 of the 1,573 trinucleotides eSSRs contained GC. This probably reflects the relatively low GC content of dicot genes [49].
New eSSR markers
There have been several publications describing the development of SSR markers in the genus Populus [32], [51], [52]. However, to our knowledge, no more than 12 gSSR markers have been developed for P. euphratica. As such, there is a great need to increase the number of available markers for this unique desert tree. Based on the high transferability of SSRs in Populus (for example, the transferability rate of markers from P. trichocarpa to P. euphratica is around 70% [32]), we initially tested over 499 primers that were randomly selected from the OAK RIDGE National Laboratory (http://www.ornl.gov/sci/ipgc/ssr_resource.htm) database. However, less than 20% of the selected primers could be successfully amplified. This could be because EST-derived SSRs, which derived from the transcribed DNA regions, are expected to be more highly conserved and should thus have a higher transferability rate across species than gSSR markers.
Validating SSR primers is important in the early stages of the development process [17]. We have found that the genotyping of SSRs is often laborious and time consuming when using traditional PAGE analysis and so we instead used M13 technology with fluorescent tagging [53]. In addition, pseudomultiplex genotyping based on the use of mixtures of PCR products with different fluorescent labels was also used and found to be an efficient and economical method for identifying SSRs. More than 600 primers were designed and validated in less than one month. 676 of the newly-developed primer pairs were selected at random for amplification to test the reliability of the corresponding eSSR markers in P. euphratica, and amplicons were obtained for 575 of them (86%). The instance of unsuccessful amplification may have been due to any of the following factors: 1) The presence of introns in the corresponding ESTs, 2) The presence of SNPs or INDELs in the primers, 3) The tendency of M13 tails to decrease PCR efficiency, leading to a need for additional PCR cycles, 4) Assembly errors in the EST sequences.
Polymorphism of eSSRs for P. euphratica
Only polymorphic markers are useful for population genetic studies at the genomic level. Of the 575 eSSR markers with validated primer pairs, 302 were polymorphic, giving a polymorphism rate of 52.5%. For comparative purposes, it has been reported that 38% of the primer pairs corresponding to a set of eSSR markers successfully amplified a single polymorphic locus each when applied to a reference full-sib pedigree of Quercus robur [24]. The polymorphism ratio for P. euphratica eSSRs is thus relatively high compared to the values observed in other large-scale polymorphic eSSRs screening studies.
Previous studies have revealed a positive correlation between SSR length and polymorphism rate in plants and animals [54]. We confirmed this observation by comparing the lengths of each type of polymorphic SSR to monomorphic SSRs. Based on studies of dinucleotides, trinucleotides and hexanucleotides repeats, polymorphic eSSRs all have higher number of repeats than monomorphic eSSRs (Figure S3). For all SSR types considered, the repeat number had a significant effect on the rate of polymorphism: the estimated correlation coefficients for this relationship in di-, tri-, and hexanucleotides repeats were 0.091, 0.269 and 0.421, respectively, with corresponding p values of p<0.001, p<0.001and p<0.05. This relationship persisted even when the data for all three SSR types were combined, giving an estimated correlation coefficient of 0.149 (p<0.001).
A web-based database of eSSRs resource for P. euphratica
There are quite few plant SSR databases. The Triticeae eSSR coordination (http://wheat.pw.usda.gov/ITMI/EST-SSR/) provides a collection of 2,463 nonredundant "Uni-EST-SSR" contigs from wheat, barley, rice, maize and sorghum, together with their consensus sequences [55]. The Simple Sequence Repeat Database (http://intranet.icrisat.org/gt1/ssr/ssrdatabase.html) provides eSSRs from sorghum, soybean, medicago, lotus, rice and maize and can be queried based on motif, minimum length, maximum length, and match type [56]. This database also provides cross-species eSSR clusters. The tomato SBM & Marker Database (http://marker.kazusa.or.jp/Tomato/) provides eSSRs and gSSRs for the tomato, but only with the corresponding primer sequences [57]. The CMD (http://www.cottonmarker.org/) incorporates all publicly available SSRs and related information [28]. The P. trichocarpa genome has been released and its SSRs have been developed and deposited in the archives of The International Populus Genome Consortium (http://www.ornl.gov/sci/ipgc/ssr_resource.htm). However, these data can only be obtained in an EXCEL file that provides limited information, such as primer sequences, amplification status, motif type etc. [32], [51], [52] Our Euphrates Poplar EST-SSR DB integrates both general information regarding its eSSRs (such as the associated EST sequences, primers that have been designed against them, motif type information, polymorphic content, and transferability) and has several other useful features. Notably, it provides the location of each SSR (CDS or UTR) via a graphical interface that shows their relative position on the 19 P. trichocarpa pseudo-chromosomes and provides links to more detailed information on SSRs of interest. It also displays annotations for SSR-containing ESTs and makes all of the available information on each eSSR searchable and interlinked.
In conclusion, we have developed and characterized the first large-scale set of eSSRs for P. euphratica by analyzing public transcriptome. More than 300 polymorphic eSSR markers have been screened and validated, and the resulting data have been integrated into a user-friendly web-based database. This information will have many potential applications in studies on Populus in fields such as population genetics, comparative genomics, linkage mapping, QTL, and marker-assisted breeding.
Materials and Methods
Plant material and DNA isolation
Three populations were sampled from a 4×104 hectare poplar forest (Latitude 41.0325°–41.0825°, Longitude 86.1181°-86.4693° and Altitude: 884 m) located along the Tarim River in western China (Yuli County, Xinjiang) in Xinjiang province, China. Six, 12 or 24 individuals were selected at random from each of the three populations. Leaves from each individual were collected and dried over silica gel. The dried leaves were placed in a 2 mL tube, two φ3 mm steel beads were added, and the leaves were ground in a shaker (Mini-Beadbeater, Biospec USA). DNA was extracted using the CTAB method with some modifications [58]. DNA concentrations were measured using a Nano Drop 2000 spectrophotometer (Thermo, USA) and the extracted DNA was transferred to 96 well plates at an adjusted concentration of 10-15 ng/ µL.
EST sources and eSSR analysis
ESTs from P. euphratica were obtained from the NCBI EST database and the transcriptome data reported by Qiu et al. [34]. PolyA/T sequences at the 5′ and 3′ ends of the ESTs were removed using EST-trimmer (http://pgrc.ipk-gatersleben.de/misa/download/est_trimmer.pl). ESTs that were below 100 bp in length were not included in the analyses. Vector, chloroplast and mitochondrial sequences were removed from the data set using SeqClean (http://compbio.dfci.harvard.edu/tgi/software/). Redundant ESTs were cleared with CD-HIT [35] using a 90% sequence similarity threshold. Non-redundant ESTs were then searched for SSRs using MIcroSAtellite identification tool (MISA) [36]. We searched for di- (repeat count > = 10), tri- (> = 6), tetra- (> = 5), penta- (> = 4), hexa- (> = 3) nucleotides, with an interruption (max_difference_for_2_SSRs) of 100 bp.
Mapping and annotation of eSSRs
We used a series of in-house perl scripts to map P. euphratica eSSRs onto the P. trichocarpa genome. First, the BLASTN program (http://blast.ncbi.nlm.nih.gov/) was used to locate the ESTs on the P. trichocarpa genome. Second, BLAST results and information on the locations of the SSRs were combined to infer SSR locations in the P. trichocarpa genome. At this stage, only blast results with an identity score of >0.8 were used. Third, we aligned the SSRs against the corresponding sequences in the P. trichocarpa genome using MUSCLE (http://www.ebi.ac.uk/Tools/msa/muscle/) with the default criteria. Finally, based on the SSRs' locations in the P. trichocarpa genome and gene annotation data for the P. trichocarpa genome, the location of each SSR with respect to the nearest gene (UTR or CDS) was evaluated and the mapping results were recorded in an EXCEL document (Table S6).
Functional annotations were assigned to ESTs and eSSRs using Blast2GO [40]. In brief, BLAST was used to find sequences that were similar to each EST by searching against the NCBI non-redundant protein database with an e-value cutoff of 1e-3. Then GO terms associated with each blast hit were extracted and the resulting GO annotations were returned. The associated cellular component, biological process, and molecular function were identified for each EST. Fisher's Exact Test was performed in Blast2GO to identify overrepresented GO categories in SSR–containing ESTs compared with total unigenes using an FDR cutoff of <0.05.
Design of PCR primers and SSR genotyping
Primers for eSSR amplification were designed in batches using the PRIMER3 program [59] in conjunction with MISA [36] and in-house perl scripts. Information on each primer pair, including their sequences, positions, and Tm values, are presented in Table S3.
We selected 676 primer pairs at random from the primer database to assess their validity. For primary primer screening, PCR reactions were performed with a 5′ M13-tailed forward primer and reverse primer pair using DNA from four randomly selected P. euphratica individuals sampled in Xinjiang Province, China. Each 25 µL polymerase chain reaction (PCR) mixture contained 1x Taq buffer, 0.2 mM dNTPs, 10-20 ng template DNA, 1.6 pmol of the 5′ M13-tailed forward primers and the reverse primer each, and 1 U Taq polymerase (BioMed). The PCR conditions were as follows: 94°C for 5 min; then 30 cycles of 94°C for 30 s, annealing at 56°C for 45 s, elongating at 72°C for 45 s, and a final extension at 72°C for 10 min. 5 µL of the PCR products were separated on 2% agarose gels and stained with ethidium bromide to check for successful amplification. After primary primer screening, SSR genotyping was then performed with primer pairs that yielded amplicons. At this stage, DNA from six to 12 or 24 individuals from three to four locations across the distribution of P. euphratica was used to investigate the polymorphism of the new eSSR markers using a rapid and inexpensive method based on the M13-tail technique [53]. The detailed M13-tail technique used in this study was as follows. Three primers were synthesized for each genotyping run: a 5′ M13-tailed forward primer, a reverse primer, and a fluorescently-labeled M13 primer. The M13 primers were labeled with FAM, ROX, HEX, and TAMRA (Sangon) respectively. Each 10 µL polymerase chain reaction (PCR) contained 1x Taq buffer, 0.2 mM dNTPs, 10-20 ng template DNA, 1.6 pmol of the reverse and fluorescently-labeled M13 primers each, 0.4 pmol of the forward primer, and 1 U Taq polymerase (BioMed). The PCR conditions were as follows: 94°C for 5 min; then 30 cycles of 94°C for 30 s, annealing at 56°C for 45 s, elongating at 72°C for 45 s, followed by 8 cycles of 94°C for 30 s, annealing at 53°C for 45 s, elongating at 72°C for 45 s, and a final extension at 72°C for 10 min. Subsequently, 0.5 µL of each PCR product with the four different fluorescent dyes were mixed and added to 10 µL formamide and 0.5 µL LIZ standard (ABI) and run on an ABI 3730 Prism genetic Analyzer. Raw data were analyzed using GeneMarker Version 1.75 (Softgenetics, USA).
Evaluation of eSSR polymorphism and genetic diversity analysis
Genetic diversity was calculated at each locus for allelic PIC using the PIC calculator (http://www.liv.ac.uk/~kempsj/pic.html) based on allelic frequencies among all the genotypes analyzed. PIC values for each SSR were estimated by determining the frequency of alleles per locus using the formula: PIC = 1-Σ(Pi)2, where Pi is the relative frequency of the ith allele of the SSR loci. The expected heterozygosity (HE) and number of alleles (NA) were estimated for six to 12 or 24 individuals using GENALEX 6 [60]. The effect of repeat number on the polymorphism was tested using a logistic regression model created with v 2.6.2 of the R software package (R development core team 2008).
Transferability of polymorphic eSSRs to other Populus species
The transferability of the new polymorphic eSSRs was evaluated in nine other species (three endemic in China) belonging to five sections of the Populus genus. Three species were selected from the Tacamahaca Spach section, two each from the Aigeiros Duby and Leucoides Spach sections, and one each from the Leucoides Duby and Turanga Bge sections. PCR reactions were performed using primer pairs corresponding to the polymorphic eSSRs in one to four individuals of each Populus species. If a primer pair yielded successful amplification, the corresponding eSSR was scored as being transferable.
Phylogenetic tree construction
To determine whether there was any correlation between the observed rates of transferability and the phylogeny of the Populus genus, we constructed a Neighbour joining tree by assigning successful amplifications a value of ‘1’ and amplification failures a value of ‘0’. The final Neighbour joining tree for the 10 species was constructed using over 300 information sites by MEGA5 [61] with 1000 bootstraps.
Construction of a web-based P. euphratica eSSR database
In order to share our data and analytical results on P. euphratica eSSRs, including ESTs, SSRs, the relative positions of each SSR on the 19 P. trichocarpa pseudo-chromosomes, primer sequences, polymorphisms, and transferability, we integrated the data from each step of our investigation into a web-based P. euphratica eSSR database based on open-source software (Apache, PHP, and MySQL). Our P. euphratica eSSR database has a user-friendly interface and is freely available at the following website: http://202.205.131.253:8080/poplar/resources/static_page/index.html.
Supporting Information
Figure S1.
Frequency distribution of different size of motifs in EST-SSRs of Populus euphratica.
https://doi.org/10.1371/journal.pone.0061337.s001
(TIF)
Figure S2.
GO ontology of SSR-containing ESTs was categorized by biological process (A), cell component (B) and molecular function (C).
https://doi.org/10.1371/journal.pone.0061337.s002
(TIF)
Figure S3.
Correlation between PIC and motif size of EST-SSRs of Populus euphratica.
https://doi.org/10.1371/journal.pone.0061337.s003
(TIF)
Table S1.
Occurrence of 4202 SSRs identified in a set of 94,090 non-redundant Populus euphratica ESTs.
https://doi.org/10.1371/journal.pone.0061337.s004
(XLS)
Table S2.
Frequency of different motif types.
https://doi.org/10.1371/journal.pone.0061337.s005
(XLS)
Table S3.
Primer information for all the 3,069 EST-SSRs.
https://doi.org/10.1371/journal.pone.0061337.s006
(XLS)
Table S4.
Characteristics of the 673 EST-SSRs tested in six, twelve or 24 individuals of Populus euphratica.
https://doi.org/10.1371/journal.pone.0061337.s007
(XLS)
Table S5.
Transferability of all the polymorphic EST-SSRs in other nine Poplar species.
https://doi.org/10.1371/journal.pone.0061337.s008
(XLS)
Table S6.
Location of the Populus euphratica EST-SSRs on the genome of Populus trichocarpa.
https://doi.org/10.1371/journal.pone.0061337.s009
(XLS)
Acknowledgments
The authors thank Dongrui Jia, and Wenchun Luo from Lanzhou University for data analysis, Weiliang Shuai for the database construction, Shu Zhao and Xiaolin Wei for the assistance of the lab work, Wenhao Bo and Xiyang Zhao for sampling, Prof. Deqiang Zhang provide leaf materials of one species. Authors would like to thanks the two anonymous reviewers, which greatly improved the earlier version of this paper. Jijun Tang conducted this work when he visited the Center for Computational Biology, Beijing Forestry University, as a visiting professor.
Author Contributions
Conceived and designed the experiments: FKD RW. Performed the experiments: FKD FX SF. Analyzed the data: FKD. Contributed reagents/materials/analysis tools: HQ JT. Wrote the paper: FKD.
References
- 1. Sharma A, Dwivedi BN, Singh B, Kumar K (1999) Introduction of Populus euphratica in Indian semi-arid trans-Gangetic Plains. Ann For Sci 7: 1–8.
- 2.
Hou H (1985) Floodplain meadows, gallery forests, and oases. In: Goodin JR & Northington DK (eds) Plant Resource of Arid and Semi Arid Lands - A global perspective Academic Press, Orlando. pp. 246–249.
- 3. Bruelheide H, Jandt U, Gries D, Thomas FM, Foetzki A, et al. (2003) Vegetation changes in a river oasis on the southern rim of the Taklamakan Desert in China between 1956 and 2000. Phytocoenologia 33: 801–818.
- 4. Gu R, Fonseca S, Puskás LG, Hackler L, Zvara Á, et al. (2004) Transcript identification and profiling during salt stress and recovery of Populus euphratica. Tree Physiol 24: 265–276.
- 5. Kang JM, Kojima K, Ide Y, Sasaki S (1996) Growth response to the stress of low osmotic potential, salinity and high pH in cultured shoot of Chinese poplars. J For Res 1: 27–29.
- 6. Watanabe S, Kojima K, Ide Y, Sasaki S (2000) Effects of saline and osmotic stress on proline and sugar accumulation in Populus euphratica in vitro. Plant Cell Tissue Organ Cult 63: 199–206.
- 7. Zeng F, Yan H, Arndt SK (2009) Leaf and whole tree adaptations to mild salinity in field grown Populus euphratica.. Tree Physiol 29: 1237–1246.
- 8. Fay MF, Lledó MD, Kornblum MM, Crespo MB (1999) From the waters of Babylon? Populus euphratica in Spain is clonal and probably introduced. Biodivers Conserv 8: 769–778.
- 9. Vonlanthen B, Zhang X, Bruelheide H (2010) Clonal structure and genetic diversity of three desert phreatophytes. Am J Bot 97: 234–242.
- 10. Wang J, Wu Y, Ren G, Guo Q, Liu J, et al. (2011) Genetic differentiation and delimitation between ecologically diverged Populus euphratica and P. pruinosa. PLoS ONE 6: e26530. Available: http://www.plosone.org/article/info%3Adoi%2F10.1371%2Fjournal.pone.0026530.Accessed 6 March 2013.
- 11. Jacob JJ, Lindpaintner K, Lincoln SE, Kusumi K, Bunker RK, et al. (1991) Genetic mapping of a gene causing hypertension in the stroke-prone spontaneously hypertensive rat. Cell 67: 213–224.
- 12. Edwards AL, Civitello A, Hammond HA, Caskey CT (1991) DNA typing and genetic mapping with trimeric and tetrameric tandem repeats. Am J Hum Genet 49: 746–756.
- 13. Tautz D (1989) Hypervariability of simple sequences as a general source for polymorphic DNA markers. Nucleic Acids Res 17: 6463–6471.
- 14. Litt M, Luty JA (1989) A hypervariable microsatellite revealed by in vitro amplification of a dinucleotide repeat within the cardiac muscle actin gene. Am J Hum Genet 44: 397–401.
- 15. Zane L, Bargelloni L, Patarnello T (2002) Strategies for microsatellite isolation: a review. Mol Ecol 11: 1–16.
- 16. Kelkar YD, Strubczewski N, Hile SE, Chiaromonte F, Eckert KA, et al. (2010) What is a microsatellite: A computational and experimental definition based upon repeat mutational behavior at A/T and GT/AC repeats. Genome Biol Evol 2: 620–635.
- 17. Guichoux E, Lagache L, Wagner S, Chaumeil P, Léger P, et al. (2011) Current trends in microsatellite genotyping. Mol Ecol Res 11: 591–611.
- 18. Kalia RK, Rai MK, Kalia S, Singh R, Dhawan AK (2011) Microsatellite markers: an overview of the recent progress in plants. Euphytica 177: 309–334.
- 19. Scott KD, Eggler P, Seaton G, et al. (2000) Analysis of SSRs derived from grape ESTs. Theor Appl Genet 100: 723–726.
- 20. Karagyozov L, Kalcheva ID, Chapman VM (1993) Construction of random small-insert genomic libraries highly enriched for simple sequence repeats. Nucleic Acids Res 21: 3911–3912.
- 21. Armour JAL, Neumann R, Gobert S, Jeffreys AJ (1994) Isolation of human simple repeat loci by hybridization selection. Hum Mol Genet 3: 599–605.
- 22. Kandpal RP, Kandpal G, Weissman SM (1994) Construction of libraries enriched for sequence repeats and jumping clones, and hybridization selection for region-specific markers. Proc Natl Acad Sci U S A 91: 88–92.
- 23. Kijas JM, Fowler JC, Garbett CA, Thomas MR (1994) Enrichment of microsatellites from the citrus genome using biotinylated oligonucleotide sequences bound to streptavidin-coated magnetic particles. Biotechniques 16: 656–660.
- 24.
Durand J, Bodenes C, Chancerel E, Frigerio J-M, Vendramin G, et al.. (2010) A fast and cost-effective approach to develop and map EST-SSR markers: oak as a case study. BMC Genomics 11: : 570. Available: http://www.biomedcentral.com/1471-2164/11/570.Accessed 6 March 2013.
- 25. McCouch SR, Teytelman L, Xu Y, Lobos KB, Clare K, et al. (2002) Development and mapping of 2240 new SSR markers for rice (Oryza sativa L.). DNA research 9: 199–207.
- 26. Kantety RV, La Rota M, Matthews DE, Sorrells ME (2002) Data mining for simple sequence repeats in expressed sequence tags from barley, maize, rice, sorghum and wheat. Plant Mol Biol 48: 501–510.
- 27. Gupta PK, Rustgi S, Sharma S, Singh R, Kumar N, et al. (2003) Transferable EST-SSR markers for the study of polymorphism and genetic diversity in bread wheat. Mol Genet Genomics 270: 315–323.
- 28.
Blenda A, Scheffler J, Scheffler B, Palmer M, Lacape JM, et al.. (2006) CMD: a Cotton Microsatellite Database resource for Gossypium genomics. BMC Genomics 7: : 132. Available: http://www.biomedcentral.com/1471-2164/7/132/.Accessed 6 March 2013.
- 29. Poncet V, Rondeau M, Tranchant C, Cayrel A, Hamon S, et al. (2006) SSR mining in coffee tree EST databases: potential use of EST-SSRs as markers for the Coffea genus. Mol Genet Genomics 276: 436–449.
- 30.
Ueno S, Moriguchi Y, Uchiyama K, Ujino-Ihara T, Futamura N, et al.. (2012) A second generation framework for the analysis of microsatellites in expressed sequence tags and the development of EST-SSR markers for a conifer, Cryptomeria japonica. BMC Genomics 13: : 136. Available: http://www.biomedcentral.com/1471-2164/13/136/.Accessed 6 March 2013.
- 31. Feng SP, Li WG, Huang HS, Wang JY, Wu YT (2009) Development, characterization and cross-species/genera transferability of EST-SSR markers for rubber tree (Hevea brasiliensis). Mol Breed 23: 85–97.
- 32. Tuskan GA, Gunter LE, Yang ZK, Yin TM, Sewell MM, et al. (2004) Characterization of microsatellites revealed by genomic sequencing of Populus trichocarpa. Can J For Res 34: 85–93.
- 33. Tuskan GA, Difazio S, Jansson S, Bohlmann J, Grigoriev I, et al. (2006) The genome of black cottonwood, Populus trichocarpa (Torr. & Gray). Science 313: 1596–1604.
- 34. Qiu Q, Ma T, Hu Q, Liu B, Wu Y, et al. (2011) Genome-scale transcriptome analysis of the desert poplar, Populus euphratica. Tree Physiol 31: 452–461.
- 35. Huang Y, Niu B, Gao Y, Fu L, Li W (2010) CD-HIT Suite: a web server for clustering and comparing biological sequences. Bioinformatics 26: 680–682.
- 36. Thiel T, Michalek W, Varshney R, Graner A (2003) Exploiting EST databases for the development and characterization of gene-derived SSR-markers in barley (Hordeum vulgare L.). Theor Appl Genet 106: 411–422.
- 37. Botstein D, White RL, Skolnick M, Davis RW (1980) Construction of a genetic linkage map in man using restriction fragment length polymorphisms. Am J Hum Genet 32: 314–331.
- 38.
Yin TM, Zhang XY, Gunter LE, Li SX, Wullschleger SD, et al.. (2009) Microsatellite primer resource for Populus developed from the mapped sequence scaffolds of the Nisqually-1 genome. New Phytol 181: : 498– 503.
- 39. Li S, Yin T, Wang M, Tuskan GA (2011) Characterization of microsatellites in the coding regions of the Populus genome. Mol Breed 27: 59–66.
- 40. Conesa A, Götz S, García-Gómez JM, Terol J, Talón M, et al. (2005) Blast2GO: a universal tool for annotation, visualization and analysis in functional genomics research. Bioinformatics 21: 3674–3676.
- 41. Ueno S, Tsumura Y (2008) Development of ten microsatellite markers for Quercus mongolica var. crispula by database mining. Conserv Genet 9: 1083–1085.
- 42. Varshney RK, Graner A, Sorrells ME (2005) Genic microsatellite markers in plants: features and applications. Trends Biotechnol 23: 48–55.
- 43.
Maia LC, Palmieri DA, Souza VQ, Kopp MM, Carvalho FIF, et al.. (2008) SSR Locator: Tool for simple sequence repeat discovery integrated with primer design and PCR simulation. Int J Plant Genomics Article ID 412696. Available: http:// http://www.hindawi.com/journals/ijpg/2008/412696/.Accessed 6 March 2013.
- 44.
Victoria FC, da Maia LC, de Oliveira AC (2011) In silico comparative analysis of SSR markers in plants.BMC Plant Biol 11: : 15. Available: http://www.biomedcentral.com/1471-2229/11/15.Accessed 6 March 2013.
- 45. Cardle L, Ramsay L, Milbourne D, Macaulay M, Marshall D, et al. (2000) Computational and experimental characterization of physically clustered simple sequence repeats in plants. Genetics 156: 847–854.
- 46. Varshney RK, Thiel T, Stein N, Langridge P, Graner A (2002) In silico analysis on frequency and distribution of microsatellites in ESTs of some cereal species. Cell Mol Biol Lett 7: 537–546.
- 47. Metzgar D, Bytof J, Wills C (2000) Selection against frameshift mutations limits microsatellite expansion in coding DNA. Genome Res 10: 72–80.
- 48. Grover A, Aishwarya V, Sharma PC (2007) Biased distribution of microsatellite motifs in the rice genome. Mol Genet Genomics 277: 469–480.
- 49. Morgante M, Hanafey M, Powell W (2002) Microsatellites are preferentially associated with nonrepetitive DNA in plant genomes. Nat Genet 30: 194–200.
- 50.
Qiu L, Yang C, Tian B, Yang JB, Liu A (2010) Exploiting EST databases for the development and characterization of EST-SSR markers in castor bean (Ricinus communis L.). BMC Plant Biol 10: : 278. Available: http://www.biomedcentral.com/1471-2229/10/278.Accessed 6 March 2013.
- 51. van der Schoot J, Pospíšková M, Vosman B, Smulders MJM (2000) Development and characterization of microsatellite markers in black poplar (Populus nigra L.). Theor Appl Genet 101: 317–322.
- 52. Smulders MJM, Van Der Schoot J, Arens P, Vosman B (2001) Trinucleotide repeat microsatellite markers for black poplar (Populus nigra L.). Mol Ecol Notes 1: 188–190.
- 53. Schuelke M (2000) An economic method for the fluorescent labeling of PCR fragments. Nat Biotechnol 18: 233–234.
- 54. Brandström M, Ellegren H (2008) Genome-wide analysis of microsatellite polymorphism in chicken circumventing the ascertainment bias. Genome Res 18: 881–887.
- 55. Yu JK, La Rota M, Kantety RV, Sorrells ME (2004) EST derived SSR markers for comparative mapping in wheat and rice. Mol Genet Genomics 271: 742–751.
- 56. Jayashree B, Punna R, Prasad P, Bantte K, Hash CT, et al. (2006) A database of simple sequence repeats from cereal and legume expressed sequence tags mined in silico: survey and evaluation. In Silico Biol 6: 607–620.
- 57. Shirasawa K, Asamizu E, Fukuoka H, Ohyama A, Sato S, et al. (2010) An interspecific linkage map of SSR and intronic polymorphism markers in tomato. Theor Appl Genet 121: 731–739.
- 58. Xu F, Feng SS, Wu RL, Du KF (2013) Two highly validated SSR multiplexes (8-plex) for Euphrates' poplar, Populus euphratica (Salicaceae). Mol Ecol Res 13: 144–153.
- 59. Rozen S, Skaletsky H (2000) Primer3 on the WWW for general users and for biologist programmers. Methods Mol Biol 132: 365–386.
- 60. Peakall ROD, Smouse PE (2006) GENALEX 6: genetic analysis in Excel. Population genetic software for teaching and research. Mol Ecol Notes 6: 288–295.
- 61. Tamura K, Peterson D, Peterson N, Stecher G, Nei M, et al. (2011) MEGA5: Molecular Evolutionary Genetics Analysis Using Maximum Likelihood, Evolutionary Distance, and Maximum Parsimony Methods. Mol Biol Evol 28: 2731–2739.