Development of genic SSR markers from transcriptome sequencing of pear buds

Abstract

A total of 8375 genic simple sequence repeat (SSR) loci were discovered from a unigene set assembled from 116 282 transcriptomic unigenes in this study. Dinucleotide repeat motifs were the most common with a frequency of 65.11%, followed by trinucleotide (32.81%). A total of 4100 primer pairs were designed from the SSR loci. Of these, 343 primer pairs (repeat length ≥15 bp) were synthesized with an M13 tail and tested for stable amplification and polymorphism in four Pyrus accessions. After the preliminary test, 104 polymorphic genic SSR markers were developed; dinucleotide and trinucleotide repeats represented 97.11% (101) of these. Twenty-eight polymorphic genic SSR markers were selected randomly to further validate genetic diversity among 28 Pyrus accessions. These markers displayed a high level of polymorphism. The number of alleles at these SSR loci ranged from 2 to 17, with a mean of 9.43 alleles per locus, and the polymorphism information content (PIC) values ranged from 0.26 to 0.91. The UPGMA (unweighted pair-group method with arithmetic average) cluster analysis grouped the 28 Pyrus accessions into two groups: Oriental pears and Occidental pears, which are congruent to the traditional taxonomy, demonstrating their effectiveness in analyzing Pyrus phylogenetic relationships, enriching rare Pyrus EST-SSR resources, and confirming the potential value of a pear transcriptome database for the development of new SSR markers.

1. Introduction

The genus Pyrus is classified in subtribe Pyrinae within tribe Pyreae of the family Rosaceae (Potter et al., 2007), and includes approximately 22 well-recognized primary species as well as several natural or artificial interspecific hybrids (Bell and Zwet, 1998). It is a functionally diploid genus (2n=34) of allopolyploid origin and is believed to have arisen about 55–65 million years ago in the mountainous regions of western and southwestern China (Rubtsov, 1944). However, because pears are considered to be typically self-incompatible and interspecies compatible, phylogenetic relationships within the genus Pyrus are complicated and difficult to be established because of high heterozygosity from widespread crossing and low morphological diversity or poor differentiating characters (Teng et al., 2002; Bao et al., 2007; Yao et al., 2010; Zheng et al., 2011). Hence, more DNA-based genetic markers need to be exploited to resolve these problems.

Simple sequence repeats (SSRs) or microsatellites are one of the most efficient genetic markers. By virtue of their reproducibility, multi-allelic nature, co-dominant inheritance, relative abundance, and good genome coverage (Powell et al., 1996), SSR markers have been largely applied to genetic diversity studies (Gupta and Varshney, 2000; Chen et al., 2011; Zhang et al., 2013) and used in a variety of applications related to pears, demonstrating their adaptability for Pyrus (Yamamoto et al., 2002; 2004; Bao et al., 2007; Katayama et al., 2007; Yao et al., 2010; Cao et al., 2012). SSR loci consist of randomly repeated DNA regions with motif lengths of one to six base pairs (bp) and are spread throughout the genome (Tóth et al., 2000). According to their locations in the genome, SSR markers are generally divided into genomic SSRs and genic SSRs (or expressed sequence tag (EST)-SSRs); in comparison with genomic SSR markers, genic SSR markers are derived from coding regions and are believed to have some intrinsic advantages, such as inexpensiveness and relative ease of obtaining, applicability to assaying functional diversity in natural populations or germ-plasm collections, high transferability to related species, and usefulness as anchor markers for comparative mapping or evolutionary studies (Varshney et al., 2005).

However, the development of genic SSRs can be difficult because of the laborious and costly collection of ESTs (Bouck and Vision, 2006), so few pear ESTs (about 4413, before June 7, 2013) have been submitted to GenBank and few studies on the development of EST-SSR markers from Pyrus have been reported (Nishitani et al., 2009). Most of the EST-SSRs used in pear genetic diversity were derived from Malus (Yamamoto et al., 2001; Yao et al., 2010). With the development of the next-generation sequencing technology, creating transcriptome-level sequence collections has become much quicker and cheaper (Mardis, 2008). As a result, a wealth of gene-based SSRs and other genetic markers depending on such resources have been identified and developed in various plant species (Vendramin et al., 2006; Blanca et al., 2011; Dutta et al., 2011; Kaur et al., 2012; Lesser et al., 2012), but have not been reported in Pyrus.

Although a pear genome sequence was obtained with a genome size of 512 Mbp recently (Wu et al., 2013), which provides an excellent platform for future genetic marker development and phylogenetic relationship analysis in Pyrus, SSR markers designed from ESTs or unigenes possess higher amplification rates and cross-species transferability (Barbará et al., 2007). Therefore, a transcriptome database offers an attractive alternative to complement existing SSR collections. In this study, we reported the first development of genic SSR markers in Pyrus, which adds to the available resources and provides new opportunities for assessing molecular phylogeny and genetic diversity in the Pyrus species.

2. Materials and methods

2.1. Plant materials and DNA extraction

A total of 29 accessions from Pyrus were used in the development of the genic SSR markers in this study (Table 1). Genomic DNA was extracted from young leaves of these Pyrus accessions using a modified version of a cetyltrimethylammonium bromide (CTAB) protocol (Doyle and Doyle, 1987). DNA quality and quantity were checked in 1% agarose gels and Eppendorf BioSpectrometer (Eppendorf, Germany), respectively.

Table 1.

List of 29 Pyrus accessions used in the polymorphism analysis

No.	Accession	Species	Type	Origin	Leaf source
1	‘Huiyangqingli’	P. pyrifolia	Cultivar	Guangdong, China	CPGR
2	‘Huiyangsuanli’	P. pyrifolia	Cultivar	Guangdong, China	CPGR
3	‘Nijisseiki’	P. pyrifolia	Cultivar	Japan	TU
4	‘Imamuraaki’	P. pyrifolia	Cultivar	Japan	TU
5	‘Jinchuanxueli’	P. pyrifolia	Cultivar	Japan	CPGR
6	‘Manyuanxiang’	P. ussuriensis	Cultivar	Liaoning, China	CPGR
7	‘Nanguoli’	P. ussuriensis	Cultivar	Liaoning, China	CPGR
8	P. ussuriensis	P. ussuriensis	Wild material	Liaoning, China	CPGR
9	‘Cili’	P. pyrifolia white pear group	Cultivar	Shandong, China	CPGR
10	‘Yali’	P. pyrifolia white pear group	Cultivar	Hebei, China	CPGR
11	P. pashia 1	P. pashia	Wild material	Guizhou, China	Guizhou
12	P. pashia 2	P. pashia	Wild material	Guizhou, China	Guizhou
13	P. pashia 3	P. pashia	Wild material	Nepal	TU
14	P. dimorphophylla	P. dimorphophylla	Wild material	Japan	TU
15	P. fauriei	P. fauriei	Wild material	Korean	TU
16	P. betulaefolia 3	P. betulaefolia	Wild material	Japan	TU
17	P. betulaefolia 1	P. betulaefolia	Wild material	Lanzhou, China	CPGR
18	P. betulaefolia 2	P. betulaefolia	Wild material	Linwu, China	CPGR
19	P. serrulata	P. serrulata	Wild material	Hebei, China	CPGR
20	P. hopeiensis	P. hopeiensis	Wild material	Hebei, China	Hebei
21	P. phaeocarpa	P. phaeocarpa	Wild material	North of China	CPGR
22	P. xerophila	P. xerophila	Wild material	Lanzhou, China	GPI
23	P. hondoensis	P. hondoensis	Wild material	Center of Japan	TU
24	P. pyrifolia	P. pyrifolia	Wild material	Japan	TU
25	P. nivalis	P. nivalis	Wild material	Europe	TU
26	P. longipes	P. longipes	Wild material	Europe	TU
27	P. amygdaliformis	P. amygdaliformis	Wild material	Mediterranean, South Europe	TU
28	P. elaegrifolia	P. elaegrifolia	Wild material	Turkey, South Europe, Crimean Peninsula	TU
29	P. communis	P. communis	Wild material	Europe	TU

CPGR: China Pear Germplasm Repository; TU: Tottori University; GPI: Gansu Pomology Institute

2.2. Identification of SSR loci from the pear bud transcriptome

A total of 116 282 unigenes were obtained from the ‘Suli’ pear (Pyrus pyrifolia white pear group) bud transcriptome in a previous study (Liu et al., 2012). Genic SSR loci in this pear bud transcriptome were identified from the unigene set by MIcroSAtellite identification tool (MISA) (Thiel et al., 2003), which is based on the Perl language. The SSR loci containing repeat units of 2–6 nucleotides were identified, and the minimum SSR length criteria were defined as six iterations for dinucleotide repeats, and five iterations for other repeat units.

2.3. Primer design and polymorphism testing of SSR loci

BatchPrimer3 interface modules (You et al., 2008) were used to construct polymerase chain reaction (PCR) primers in the flanking regions of the SSR loci. The screening strategies for primer pairs were set as follows: 18–22 bp with an optimum size of 20 bp for the length of the primers, 60 °C for annealing temperature, 100–280 bp for the size of PCR products, and 40%–60% with an optimum of 50% for the guanine-cytosine (GC) content. Other parameters were at the default setting of the software. Three primer pairs were designed for each SSR locus, and the primer pair with the highest score was chosen as the best one.

Primers for compound SSR types were excluded. According to the length variation analysis of microsatellites in rice, all primers with a repeat length ≥15 bp were selected, a large proportion of which are the type I SSR markers (repeat length ≥20 bp), deemed to have a high chance of showing polymorphism (Singh et al., 2010). A subset of 343 primers without more than four continuous single base repeats in the primer pairs was synthesized and tested for amplification. An economic method for fluorescent labeling of PCR fragments was adopted to test their polymorphism (Schuelke, 2000). A tail (M13 universal sequence, TGTAAAACGACGGCCAGT) was added to the 5' end of each of the 343 forward primers. The primers were synthesized by Invitrogen Trading (Shanghai) Co., Ltd. Four Pyrus accessions, ‘Huiyangqingli’, ‘Huiyangsuanli’, P. ussuriensis, and P. communis, were used to test the 343 SSR markers to confirm their amplification and polymorphism. The tail primers were labeled with the following four dyes: FAM (blue), HEX (green), NED (yellow), and PET (red). FAM-tail and HEX-tail were synthesized by Invitrogen Trading (Shanghai) Co., Ltd.; NED-tail and PET-tail by Applied Biosystems (Foster City, CA, USA).

2.4. PCR and polymorphic marker validation

PCR was carried out in a final reaction volume of 20 μl, containing 20–30 ng total genomic DNA, 10× PCR buffer, 1.0 mmol/L Mg²⁺, 0.2 mmol/L of each dNTP, 5 pmol of each reverse primer, 4 pmol of the tail primer, 1 pmol of the forward primer, and 0.5 U Taq DNA polymerase (Sangon Biotech, Shanghai, China). Every primer’s amplification program contained two stages: after 5 min initial denaturation at 94 °C, 32 cycles were run for 30 s at 94 °C, 30 s at 58 °C and 30 s at 72 °C, followed by 8 cycles for 30 s at 94 °C, 30 s at 53 °C and 30 s at 72 °C. PCR products were electrophoresed on 3% agarose gels at 5 V/cm and stained with ethidium bromide to check their quality. Two independent PCRs were performed to confirm stable amplification.

The primers were divided into four subsets based on their product sizes, and then amplified with the four fluorescent tail primers using the same reaction volume. The PCR products with different fluorescent tails were diluted, then mixed with the internal size standard LIZ500, and loaded onto an ABI 3130 genetic analyzer (Applied Biosystems, Foster City, CA, USA). The sizes of fragments and alleles were calculated using GeneMapper version 4.0 (Applied Biosystems, Foster City, CA, USA).

Twenty-eight SSR markers were randomly selected from the polymorphic markers to validate genetic diversity in 28 Pyrus accessions, using the same PCR program and scoring approach. Several primers were separately amplified twice to confirm the veracity of scoring. The occurrences of null alleles at each locus were checked using MICROCHECKER 2.2.3 software (van Oosterhout et al., 2004). POPGENE version 1.32 (Yeh and Boyle, 1997) was used to calculate the frequency of the major genotypes (A), the number of genotypes (N _g), the number of alleles (N _a), observed heterozygosity (H _o), and expected heterozygosity (H _e). PowerMarker version 3.25 (Liu and Muse, 2005) was used to test for polymorphism information content (PIC). Freetree.0.9.1.50 (Pavlicek et al., 1999) was used to construct a dendrogram using UPGMA (unweighted pair-group method with arithmetic average) cluster analysis based on Nei’s genetic identity (Nei, 1973), and the confidence of the branch support was evaluated by bootstrap analysis with 1000 replicates.

3. Results and discussion

3.1. Frequency and distribution of genic SSRs in the pear bud transcriptome

A total of 8375 SSR loci containing 2–6 bp repeat motifs were discovered from the ‘Suli’ pear bud transcriptome, representing 7.2% of the total unigenes. This frequency was a little lower than that for the EST-SSRs reported in Malus (Yao et al., 2010), but similar to the 7.0% and 7.6% for transcriptome genic SSRs from the Capsicum annuum (Ashrafi et al., 2012) and Cajanus cajan (Dutta et al., 2011), respectively. The differences between these results could be due to different identification criteria for the SSR loci. The SSR loci were distributed in 6521 (5.61%) unigenes, of which 5055 (76.03%) possessed a single SSR locus, 1466 (22.48%) contained more than one SSR, and 858 had compound SSRs (13.16%).

Repeat motifs among these SSR loci consisted of 64 types (considering sequence complementarity), and the number of iterations of a given repeat unit varied from 5 to 18. Incidences of different repeat types and frequencies for each motif were evaluated based on the repeat unit number (Table 2). The most abundant motif was dinucleotide (65.11%), consistent with our observations of Pyrus EST-SSRs (Nishitani et al., 2009), followed by trinucleotide (32.81%). The most common class was n=6 (27.20%), containing mostly dinucleotide repeats. AG/CT comprised 86.94% of all dinucleotide motifs and was the most common type (Fig. 1a), similar to our analysis of EST-SSRs (Kantety et al., 2002; Yao et al., 2010). This is probably because the AG/CT dinucleotide motif can represent multiple codons resting within the reading frame that can be translated into different amino acids, and AG/CT could be present in the codons of Ala and Leu, which have the highest frequency in proteins (Yao et al., 2010). The predominant trinucleotide repeat motifs were AAG/CTT and AGG/CCT, which accounted for 25.29% and 20.74%, respectively (Fig. 1b).

Table 2.

Frequencies of different SSR repeat motif types observed in the ‘Suli’ pear bud transcriptome

SSR motif	Repeat number									Percentage (%)
	5	6	7	8	9	10	11	>11	Total
Dinucleotide	0	1532	994	888	1073	782	173	11	5453	65.11
Trinucleotide	1637	709	344	38	15	0	2	3	2748	32.81
Tetranucleotide	106	33	0	3	0	0	0	1	143	1.71
Pentanucleotide	18	1	0	0	0	0	0	0	19	0.22
Hexanucleotide	7	3	1	1	0	0	0	0	12	0.14
Total	1768	2278	1339	930	1088	782	175	15	8375	100.00
Percentage (%)	21.11	27.20	15.99	11.10	12.99	9.34	2.09	0.17	100.00

Fig. 1.

Percentages of different motifs among dinucleotide (a) and trinucleotide (b) repeats in the ‘Suli’ pear bud transcriptome

3.2. Development of genic SSR markers and polymorphism detection

Of the 8375 SSR loci, primer pairs were successfully designed for 4100 loci. A total of 343 primer pairs, including 184 (53.64%) for dinucleotide repeats, 147 (42.86%) for trinucleotide repeats, 6 (1.75%) for tetranucleotide repeats, 5 (1.46%) for pentanucleotide repeats, and 1 (0.29%) for a hexanucleotide repeat, were synthesized to amplify SSRs in four Pyrus accessions. Stable amplification of products of expected size was confirmed for 152 SSR primer pairs (44.31% of tested SSR primer pairs) in the preliminary test, a success rate comparable to 48.5% for Saccharum (Cordeiro et al., 2001), but lower than 92.2% for Linum usitatissimum (Cloutier et al., 2009) and 80.0% for Cajanus cajan (Dutta et al., 2011). The lack of amplification could be because some flanking primer pairs were designed across a splice site or chimeric cDNA sequence (Varshney et al., 2006). Although the primer pairs were designed with an annealing temperature of 60 °C, the majority were amplified at 58 °C to maximize the availability of genic markers for Pyrus.

Polymorphism was confirmed for 104 SSR markers (Table S1) with PCR products ranging from 110 to 270 bp. The polymorphic ratio was 68.42% (104/152), which was higher than 51.09% for Pinus (Lesser et al., 2012) and 29.6% for Vicia faba (Kaur et al., 2012), probably because the type I genic SSRs showed a high level of polymorphism or because of high heterozygosity in the Pyrus (Wu et al., 2013). Dinucleotide and trinucleotide motifs represented 97.11% (101/104) of the polymorphic SSRs, which was in agreement with higher-order repeat motifs being less polymorphic than lower-order repeats (Dreisigacker et al., 2004). In addition, no AT/TA was found among these polymorphic markers; AT/TA is not usually used to develop markers because its self-complementary nature leads to the formation of dimers (Wang et al., 2011).

3.3. Functional annotation for SSR-containing unigene sequences

All SSR-containing unigene sequences were used for searching homology proteins in the National Central for Biotechnology Information (NCBI) database. Only 47.9% of the unigene sequences containing SSR loci were BLAST-annotated in the previous transcriptome analysis. This may be because the transcriptome unigenes were sequenced directly instead of after cloning, which makes new rare transcript discovery easier (Wang et al., 2009). Among these polymorphic SSR loci, 47.12% of the SSR-containing unigene sequences showed significant hits in the NCBI non-redundant protein database, and their putative functions were mostly correlated with specific biological activities (Table S1). For example, the unigene sequence containing the TXY86 SSR locus was similar to the gene of farnesyl pyrophosphate synthase (FPPS), a key branch-point enzyme in the mevalonate pathway, which is involved in the syntheses of cytokinins, abscisic acid, and gibberellins required in essential biological processes (Reilly et al., 2002). These markers with particular significance can be further applied in molecular breeding to facilitate genetic improvement of the pear.

3.4. Assessment of the genetic relationship among 28 Pyrus accessions by genic SSR markers

The applicability of the polymorphic genic SSR markers in Pyrus was validated in 28 accessions (Table 3). The PCR product sizes of the 28 genic SSR markers were 102–268 bp, and corresponded to the expected sizes for each primer. The number of alleles detected at these 28 loci ranged from 2 to 17 with an average of 9.42 alleles per locus. The observed heterozygosity at the different loci ranged from 0.14 (TXY16) to 0.89 (TXY104) with a mean of 0.49, which was lower than 0.63 reported in genomic SSR marker analysis in Pyrus (Kimura et al., 2002; Bao et al., 2007), consistent with the expectation that gene-based SSRs developed from transcribed regions are less polymorphic than genome-derived SSRs. However, this was similar to values of 0.44 for Malus EST-SSRs in European pears (Wünsch and Hormaza, 2007) and 0.48 for Pyrus EST-SSRs in identification of European and Asian pears (Bassil and Postman, 2010). Another important index of the level of polymorphism, PIC values, displayed high polymorphism at each locus, ranging from 0.26 to 0.91 with an average of 0.75.

Table 3.

Characteristics of the 28 polymorphic SSR markers validated in 28 Pyrus accessions

Locus	Repeat motif	Primer sequence (5'–3')	Size range (bp)	A	N g	N a	H o	H e	PIC
XY121	(CT)8	F: <FAM><Tail>ACTGCTTGTAACACCCAACG; R: CGGAGTCAGTTGGATATCGG	102–158	0.3333	25	16	0.6071	0.8714	0.8460
TXY86	(AG)8	F: <FAM><Tail>TTGGGTCTTTAAATGCCAGC; R: CCAGACGTGAGTTGTTGCC	114–156	0.1667	28	17	0.8214	0.9318	0.9091
TXY170	(CAT)9	F: <FAM><Tail>TGCCTTTCCATTCTCAGGTC; R: AAGCGGCTGTCTAAATGAACA	104–144	0.3611	17	8	0.6071	0.7571	0.7095
TXY88	(AG)8	F: <FAM><Tail>GAGGGAGAGAGGAAGCCATT; R: CACAAGCTTTGTATTCGGCA	110–150	0.3472	18	11	0.4286	0.8123	0.7730
TXY5	(TG)8	F: <FAM><Tail>GGAGCAATGTGTGTTGTCACT; R: CCTTGCGATCGATAATTTCC	118–146	0.2222	22	9	0.4643	0.8766	0.8459
TXY276	(CAGCT)5	F: <FAM><Tail>CCCTACAGAGTCATGCATCC; R: TTGATGCTGGAGACGAGAAA	119–144	0.5714	12	6	0.4074	0.6408	0.5882
TXY11	(TC)8	F: <FAM><Tail>CAGAATTCAACATTCACTCTCTCTC; R: GAGTAGGGATGTGTCGGCTC	120–166	0.3235	13	9	0.2692	0.8228	0.7814
TXY15	(TC)8	F: <HEX><Tail>CAGGCTAGGGTTCAGGGTTT; R: CCTTTGAAGCCAAAGACTCG	146–184	0.2500	23	14	0.6429	0.8565	0.8250
TXY102	(AC)8	F: <HEX><Tail>CTCTTAACCCTCTCTCCCCC; R: CCCGATGACTCTTGAATGGT	159–183	0.2639	21	12	0.5000	0.8532	0.8204
TXY16	(TC)8	F: <HEX><Tail>TTGCTGAAGCTTCTCTTCTTCTC; R: GAGCCCACAAGGGTTCAATA	160–210	0.2500	19	8	0.1429	0.7766	0.7362
TXY267	(TC)9	F: <HEX><Tail>GTCTTAGCTTCTGCTCCCGA; R: AACCTGATGAGAATGACGGG	149–195	0.1944	21	12	0.4286	0.8909	0.8625
TXY63	(CT)8	F: <HEX><Tail>ACCTGTCATCACTTCTACTGGTTA; R: AGCCTACAAACATGCATCCC	169–189	0.1944	24	11	0.4643	0.8903	0.8623
TXY94	(AG)8	F: <HEX><Tail>ACACGCGCACACAGAGATAC; R: CCGTCACGTCGCTTTCTT	172–206	0.2083	26	14	0.5000	0.8961	0.8689
TXY95	(AG)8	F: <HEX><Tail>TACGGACAAGGACACCATGA; R: CTTGACGGTGTCAAAGGGAT	177–205	0.4028	14	10	0.5000	0.7721	0.7240
TXY236	(TCC)6	F: <NED><Tail>TCCACTCCAAACCCAGAAAC; R: GCTTCAGGGAAGCTGAAATG	183–189	0.7571	4	2	0.2222	0.3075	0.2562
TXY185	(CAG)7	F: <NED><Tail>GGAGGACCAACAGCAACATT; R: AGCAAGATCAAGCCAGGTGT	181–196	0.3889	11	6	0.4643	0.7682	0.7134
TXY2	(AC)9	F: <NED><Tail>ACGCTTCAGGTTTGGACTTC; R: TCAACCTGGACCATACATTCA	172–206	0.2083	22	10	0.5714	0.8656	0.8325
TXY280	(AAGG)5	F: <NED><Tail>CGTCCAAACTCCTCAGCTTC; R: CTCTGTCTGTCTGGTCGGGT	187–215	0.2778	17	8	0.6786	0.8253	0.7845
TXY150	(TC)9	F: <NED><Tail>GCCTTGGGCTTCTTCTTCTT; R: TCGTCGTGTGATTCGACTTC	190–216	0.4583	14	8	0.3929	0.7370	0.6898
TXY219	(GTG)6	F: <NED><Tail>CTAAGAAGCCGTTCCGATGA; R: GTCTCCCATCCTCCTCTTCC	194–218	0.5556	11	7	0.3929	0.6792	0.6304
TXY45	(GA)8	F: <NED><Tail>CAGACCCCAATTCAACGTCT; R: TCTTCTGGACATGGAGGACC	184–230	0.4167	14	7	0.5000	0.7409	0.6956
TXY214	(ACC)6	F: <PET><Tail>AACCTAAACGCATTGCAACC; R: AAGATGTGAAACCGAAACCG	202–235	0.2361	16	9	0.8214	0.8474	0.8109
TXY25	(TC)8	F: <PET><Tail>TCGATTCGATCAGGTGCTAA; R: CCGTCTCTCTTTCTCCGTTG	145–235	0.3194	18	10	0.3929	0.7773	0.7338
TXY138	(GCC)6	F: <PET><Tail>AGCATTTCCGCTGTCAGAAT; R: ATGAAGTCGTCCAGAGCGTT	218–230	0.3750	11	5	0.5357	0.7565	0.7014
TXY48	(GA)8	F: <PET><Tail>ACTAGGCAAACCAATCCAGC; R: TTGGGATCTTGTGAAGAGGG	221–245	0.4032	13	8	0.2500	0.7385	0.6923
TXY144	(GAG)11	F: <PET><Tail>ACGGAGAAGAAGCAGCAGAG; R: ACAATCTCATCCTCAACGCC	216–228	0.5161	8	5	0.1667	0.7048	0.6490
TXY104	(AC)8	F: <PET><Tail>CCCTTGATACTCAAACCCTCC; R: GATCGTTCCTTCACCAGCAT	238–270	0.3286	17	13	0.8889	0.8519	0.8204
TXY164	(TGC)5	F: <PET><Tail>CACACGAATAGCTCCTGCAA; R: AGCATCAGCAACAGCAACAG	241–268	0.2917	16	9	0.6071	0.8279	0.7903
Mean				0.3437	16.96	9.429	0.4882	0.7884	0.7483

<FAM>, <HEX>, <NED>, and <PET> represent four different fluorescent labels; <Tail> represents the M13 universal primer (TGTAAAACG ACGGCCAGT) added to the 5' end of forward primers. Shown for each primer pair are the repeat motif, primer sequence, size range (bp), major allele frequency (A), number of genotypes (N _g), number of alleles (N _a), observed heterozygosity (H _o), expected heterozygosity (H _e), and polymorphism information content (PIC)

The UPGMA cluster analysis divided the 28 Pyrus accessions into two groups, Oriental pears and Occidental pears, with a 100% confidence coefficient (Fig. 2), which was similar to many previous related studies (Teng and Tanabe, 2004; Bao et al., 2007; Yao et al., 2010), demonstrating the applicability of genic SSRs from transcriptomic unigenes in Pyrus.

Fig. 2.

Dendrogram for 28 Pyrus accessions derived from UPGMA analysis of 28 highly polymorphic SSR markers

The numbers are bootstrap values based on 1000 iterations; only values larger than 50 are included

In addition, the Oriental pear cluster contained two subgroups: A1, consisting of three P. betulifolia accessions, and A2, which clustered the other cultivars and species together. The position of P. betulifolia at the first independent branch in the Oriental pears might be related to its older origin (Zheng et al., 2011), which was confirmed by recent phylogenetic research from our research group (unpublished data). The species and cultivars in A2 displayed a complex relationship with a lower confidence coefficient, probably due to wide hybridization within Pyrus (Aldasoro et al., 1996). Some species of unknown or controversial origin were also included in this validation experiment, such as P. hopeiensis and P. phaeocarpa. P. hopeiensis was grouped together with P. pashia, and P. phaeocarpa clustered with P. serrulata, indicating their hybrid origins (Zheng et al., 2011). In addition, P. ussuriensis and ‘Yali’ formed a branch in agreement with Yao (2010). ‘Cili’, another cultivar from the P. pyrifolia white pear group, clustered with P. pyrifolia from Japan, which supports the viewpoint that Chinese white pears are a variety or ecotype of P. pyrifolia (P. pyrifolia white pear group) (Teng and Tanabe, 2004; Bao et al., 2007; Zheng et al., 2011).

These polymorphic gene-based SSR markers can be used to support germ-plasm enhancement, population structure, and genetic diversity to understand the genetic architecture of important agronomic traits in Pyrus, or transferred to other species in the subfamily for further research. In addition, these polymorphic SSR loci located in SSR-containing unigene sequences with putative special functions can be combined with phenotypic data to extend their applications in germ-plasm evaluation and examining the functional diversity at a certain locus.

4. Conclusions

This is the first report of genic SSR marker development in Pyrus from transcriptomic unigenes by next generation deep sequencing technology. A total of 8375 genic SSR loci were discovered from transcriptomic unigenes and a set of 104 (30.3% in 343) genic SSR markers were developed to enrich rare Pyrus EST-SSRs and provide additional tools for genetic analysis in the genus Pyrus. Twenty-eight polymorphic loci were further validated in 28 accessions and showed high polymorphism, demonstrating their effectiveness in analyzing Pyrus genetic relationships and confirming the potential value of a pear transcriptome database for the development of new SSR markers.

List of electronic supplementary materials

Table S1

Detailed characteristics of 104 polymorphic SSR markers in four Pyrus accessions