Pea (Pisum sativum L.) is one of the most important legumes grown and consumed worldwide. White mold caused by the fungal pathogen Sclerotinia sclerotiorum (Lib.) de Bary is a significant yield-limiting disease of pea in most areas that pea is cultivated. Despite the agricultural importance of pea, pea breeding is constrained by a large genome size (∼4300 Mb), lack of genomic resources, and rich repetitive DNA (estimated at 75–97% of the pea genome) (Macas et al., 2007). Molecular markers have great potential to speed up the process of developing improved cultivars. Although several hundred simple sequence repeat (SSR) markers have been identified (Burstin et al., 2001; Loridon et al., 2005; Gong et al., 2010), additional SSR markers with polymorphism are needed, particularly for the development of linkage maps for use in white mold—resistance mapping studies.

With the development of next-generation sequencing technologies, large amounts of expressed sequence tags (ESTs) have been generated for model species as well as economically important nonmodel plants. These ESTs offer an opportunity to discover novel genes and have also provided a resource to develop markers (Davey et al., 2011). Recently, we sequenced the transcriptome of pea infected by S. sclerotiorum using next-generation sequencing to understand this host—pathogen interaction. The transcriptome sequences from pea contain abundant SSRs, which we have used in this study to develop SSR markers. The SSR markers were screened against 23 pea cultivars and plant introductions (PIs), including parents of four recombinant inbred line (RIL) populations (Lifter and PI240515; Medora and PI169603; Bohatyr and Shawnee; Melrose and Radley) for white mold—resistance mapping studies. These new markers will be very useful for linkage mapping studies.

METHODS AND RESULTS

LIFTER, a cultivar susceptible to S. sclerotiorum (McPhee and Muehlbauer, 2002), was inoculated with S. sclerotiorum isolate WMA-1 (≡ATCC MYA-4521) on the stem between the fourth and fifth detectable nodes. Seventy-two hours after inoculation, total RNA was extracted from 18 infected plants by cutting a 1 cm piece of pea stem containing the advancing lesion front toward the base of the plant using the TRIzol Plus RNA Purification Kit (Invitrogen, Carlsbad, California, USA). Messenger RNA was purified from the total RNA with the Oligotex mRNA Mini Kit using the mRNA Spin-Column Protocol (QIAGEN, Valencia, California, USA) and converted into a normalized cDNA pool with the services of Evrogen ( http://www.evrogen.com). Transcriptome sequencing of pea infected by S. sclerotiorum was conducted on a full plate of the Roche 454 GS FLX sequencer (454 Life Sciences, Branford, Connecticut, USA) at Washington State University. In total, 128720 high-quality reads with an average length of 215 nucleotides were obtained and assembled into 10 158 contiguous sequences (contigs) with the program ABySS (Simpson et al., 2009). Pea and S. sclerotiorum contigs were parsed with a tBLASTx method (Zhuang et al., 2012) against publicly available, closely related plant and fungal genome databases. The fungal genome database consisted of S. sclerotiorum (strain 1980) and six closely related fungal (Ascomycete) species (Botrytis cinerea Pers., Chaetomium globosum Kunze, Fusarium graminearum Schwabe, Magnaporthe grisea (T. T. Hebert) M. E. Barr, Neurospora crassa Shear & B. O. Dodge, and Verticillium dahlia Kleb.), and the plant genome database consisted of three sequenced legume (Fabaceae) genomes (Glycine max (L.) Merr., Lotus japonicus (Regel) K. Larsen, and Medicago truncatula Gaertn.). After parsing, 10 158 contigs were separated into 6299 pea ESTs, 2780 S. sclerotiorum ESTs, and 1079 unassigned ESTs. Among the pea ESTs, 118 potential SSRs, with more than five repeat units or a minimum repeat size of 20 nucleotides, were identified in 112 contigs of pea with the program SSRIT (Temnykh et al., 2001;  Appendix S1 (APPS_12-00249_AppendixS1.fasta)). Of these 118 SSRs, trinucleotide repeats represented the largest fraction (50%) followed by dinucleotide (39.8%) SSRs. Two tetranucleotide, three pentanucleotide, and seven hexanucleotide SSRs were also identified in this pool. It was possible to design primers to the SSR flanking regions of 46 of the 118 SSRs using Primer3 (Rozen and Skaletsky, 2000; Table 1) with default parameters.

TABLE 1.

Characteristics of 11 polymorphic and 26 monomorphic Pisum sativum EST-SSR markers.

Continued.

The SSR markers were tested against 23 individual pea cultivars (Appendix 1), including parents of four pea RIL mapping populations (Lifter and PI240515; Medora and PI169603; Bohatyr and Shawnee; Melrose and Radley), which are being used to map quantitative trait loci for resistance to white mold. Genomic DNA from each individual was extracted from leaves using the DNeasy Plant Mini Kit (QIAGEN). PCR contained 4 µL of 5× GoTaq PCR Buffer (Promega Corporation, Madison, Wisconsin, USA), 200 µM each dNTP, 2.5 µM each primer, 0.4 U of GoTaq polymerase, and ∼50 ng of DNA template in a final volume of 20 µL. PCR were held at 94°C for 2 min; followed by 35 cycles of 94°C for 30 s, 55–60^°C for 30 s, and 72°C for 1 min; with a final extension at 72°C for 10 min. The PCR products were separated in 10% Polyacrylamide gels run in a Mega-Gel high-throughput electrophoresis system for 2.5 h at 300 V (C.B.S. Scientific, San Diego, California, USA). SSR bands were visualized with ethidium bromide, which was added to the running buffer. SSR band size was calculated by comparison with a 25-bp DNA ladder (Invitrogen). PCR products with expected sizes were successfully amplified for 37 primer sets, among which 11 showed clear polymorphisms with two to four alleles (Table 2). Observed heterozygosity and expected heterozygosity were calculated using POPGENE (version 1.32; Yeh and Boyle, 1997), and ranged from 0 to 0.43 and from 0.31 to 0.83, respectively. Ten of 11 markers (except Psat7598) were polymorphic in parents of at least one RIL population for white mold—resistance mapping studies. To determine if there was any redundancy between the SSRs described in this study and those previously published, all 37 ESTs were executed with BLASTn against P. sativum EST databases in the National Center for Biotechnology Information (taxid: 3888) with a cutoff parameter of 1e^-20. BLASTn results show that only one EST (Psat4741) matched to a previously described but unpublished SSR marker; all other 36 ESTs including all 11 polymorphic SSRs were found to be unique to this study.

TABLE 2.

Results of initial primer screening in 23 Pisum sativum individuals.

CONCLUSIONS

In this study we demonstrate that next-generation sequencing is an effective tool to rapidly develop EST-derived SSR markers. We identified 37 P. sativum EST-SSRs, with 11 being polymorphic in 23 P. sativum individuals. These novel EST-SSR markers will be valuable tools for marker-assisted breeding, development of pea linkage maps, and comparative mapping of pea.