Construction of a high-density genetic linkage map in pear (Pyrus communis × Pyrus pyrifolia nakai) using SSRs and SNPs developed by SLAF-seq

Highlights

• The high-density map is the densest map constructed by SNP and SSR markers based on European pear and Asian pear.

• The genetic map provide worthy references for identifying QTLs, comparing genetic maps and molecular breeding.

• The pear genetic map will facilitate assembling scaffolds of the ‘Bartlett’ genome.

Abstract

High-density genetic maps are valuable and effective tools for exploring complex quantitative traits, including mapping quantitative trait loci and developing linkage markers. Here, we developed 4797 single nucleotide polymorphic (SNP) markers using specific length amplified fragment sequencing. As a result, 4664 SNP markers, combined with 201 simple sequence repeat markers, were used to build a high-density linkage map of the F₁ population ‘Red Clapp’s Favorite’ (Pyrus communis L.) × ‘Mansoo’ (Pyrus pyrifolia Nakai). The integrated map contained 17 linkage groups, spanning 2,703.61 cM, with an average distance of 0.56 cM between adjacent markers. Each linkage groups anchored from 6 to 18 SSR markers. Additionally, detailed information on the female and male maps was determined. We also identified the SNP markers’ physical map locations in the P. communis L. ‘Bartlett’ genome. The results will be useful for identifying candidate genes associated with economically important characteristics.

Introduction

Pear (Pyrus spp.), belonging to the subfamily Maloideae in Rosaceae, have been cultivated in Europe, East Asia, and North America for up to 3000 years and is commercially grown in about 50 temperate-climate countries(Bell et al., 1996). The pear’ cultivation area and yield worldwide were 1.77 million hectares and 25.2 million tonnes, respectively, in 2013 (FAOSTAT 2013, http://faostat.fao.org). However, the pear genome is highly heterozygous because of self-incompatibility. This characteristic, along with its long juvenile period, make it difficult for breeders to directly determine the phenotypes associated with genotypes. Thus, the construction of high-density genetic linkage maps and analysis of quantitative trait loci (QTLs) to obtain molecular markers that are tightly linked to interested traits for marker-assisted selection is of significant value, because it can accelerate the breeding process, and reduce the population size and the cost of fostering progenies to maturity(Luby and Shaw, 2001).

From 2001–2013, several genetic linkage maps were constructed for pear using random amplified polymorphic DNA (RAPD), restriction fragment length polymorphism (RFLP), amplified fragment length polymorphism (AFLP), simple sequence repeats (SSRs) and other markers, excluding single nucleotide polymorphisms (SNPs) (Dondini et al., 2004, Iketani et al., 2001, Lu et al., 2010, Pierantoni et al., 2004, Terakami et al., 2009, Yamamoto et al., 2002a, Yamamoto et al., 2007, Zhang et al., 2012). These maps contained only a few hundred loci because of the limited number of markers and their being not conducive to automated detection. With the recent release of the de novo Chinese pear genome(Wu et al., 2013), 1341 SSR primers were designed and used for pear genetic mapping, and a 734 loci-containing genetic map of pear was constructed using the new and common SSRs (Chen et al., 2014). To date, this map contained the maximum SSR markers in pear. With the development of plant genome sequencing, SNPs have been commonly used as markers in the construction of genetic maps and marker-assisted breeding. Compared with SSR markers, SNP markers are far more abundant and can easily be detected using computer software. Therefore, SNP markers are more useful in constructing a high-density genetic map for pear. In 2014, a high-density molecular map constructed using 3143 SNP markers and 98 SSRs was reported (Wu et al., 2014), which is the maximum number of markers reported to date.

Recently, the complexity of genomes has been reduced by the development of reduced representation libraries to reduce the costs of sequencing (Van Tassell et al., 2008). To separate and purify the restriction fragments in a fixed size range is one of the simplest methods. For example, the high-density genetic maps of grape and pear have been constructed by restriction site-associated DNA sequencing (RAD-seq)(Wang et al., 2012, Wu et al., 2014). Specific length amplified fragment sequencing (SLAF-seq) is a recently developed strategy that, like RAD-seq, is used for the large-scale de novo discovery and genotyping of SNPs (Sun et al., 2013). Several high-density genetic maps for different species, such as Agropyron Gaertn., Prunus mume, soybean and sesame(Li et al., 2014, Zhang et al., 2015a, Zhang et al., 2015b, Zhang et al., 2013), have been successfully constructed based on SLAF-seq in the past few years.

Here, we obtained a segregating pear F₁ population derived from an interspecific cross between the European pear ‘Red Clapp’s Favorite’ (Pyrus communis L.) and the Asian pear ‘Mansoo’ (Pyrus pyrifolia Nakai) that was used to construct the genetic map. The fruits of ‘Red Clapp’s Favorite’ pear cultivar were full red skin with average fruits weight 180 g, and the maturation of fruits was at early July. The ‘Mansoo’ pear had brown skin colour, 400 g average fruits weight, and the fruits maturation was at middle September. SLAF-seq was used for the large-scale discovery and genotyping of the SNPs. These SNP markers, together with SSR markers, were used to construct a high-density genetic map for pear. This map should serve as a platform for the future identification and genetic dissection of many complex and important pear traits.

An F₁ pear population of 161 individuals derived from ‘Red Clapp’s Favorite’ × ‘Mansoo’ was used for mapping. ‘Red Clapp’s Favorite’ is a bud mutation of the European pear ‘Clapp’s favorite’ (P. communis L.) and the Asian pear ‘Mansoo’ (P. pyrifolia Nakai) that was bred in Korea (Fig. 1). The population was hybridized in the year 2010, and the plants were grown in the Zhengzhou Fruit Research Institute, Chinese Academy of Agricultural Sciences in Henan Province, China. Genomic DNA was extracted from young leaves of each individual, as well as from the parents, following the CTAB protocol (Iketani et al., 1998).

In this experiment, the SLAF-seq strategy of high-throughput sequencing was used with some modifications in the library construction as reported by Sun et al. (2013). Briefly, the reference genome of P. communis (http://www.rosaceae.org/species/pyrus/pyrus_communis/genome_v1.0) was used as the reference when performing the pre-experiment in silico simulation of the number of markers generated by different enzyme combinations. Next, the SLAF library was constructed in accordance using the predesigned scheme. For the F₁ population the RsaI and HaeIII (New England Biolabs, USA) enzyme combination was applied to digest the genomic DNA. The digested fragments were added a single nucleotide (A) using Klenow Fragment (3′ → 5′ exo−) (NEB) and dATP at 37 °C. Then the A-tailed fragments were ligated to duplex tag-labeled sequencing adapters (PAGE-purified, Life Technologies, USA) using T4 DNA ligase. Polymerase chain reaction (PCR) was performed using diluted restriction-ligation DNA samples, dNTP, Q5^® High-Fidelity DNA Polymerase and PCR primers (Forward primer: 5′-AATGATACGGCGACCACCGA-3′, reverse primer: 5′-CAAGCAGAAGACGGCATACG-3′)(PAGE-purified, Life Technologies). PCR products were then purified using Agencourt AMPure XP beads (Beckman Coulter, High Wycombe, UK) and pooled. Pooled samples were separated by 2% agarose gel electrophoresis. Then, the SLAF library was constructed and eventually the RsaI and HaeIII (New England Biolabs, USA) enzyme combination was applied to digest the genomic DNA. Finally, the DNA fragments, which ranged from 264 to 314 base pairs (bp) (with indexes and adaptors) in size, were excised and purified using a QIAquick Gel Extraction kit (Qiagen, Hilden, Germany). Gel-purified products were then diluted. Paired-end sequencing (each end being 125 bp) was performed on an Illumina HiSeq 2500 system (Illumina Inc., San Diego, CA, USA) according to the manufacturer’s recommendations.

SLAF markers were identified and genotyped using the procedures described by Sun et al. (2013). After filtering out the low-quality reads (quality score < 30), the remaining reads were assigned to each progeny based on the duplex barcode sequences. Clean reads generated from high-quality reads with their terminal 5 bp positions trimmed off were mapped onto the pear (http://www.rosaceae.org/species/pyrus/pyrus_communis/genome_v1.0) genomic sequence using SOAP software (Li et al., 2008). Sequences that mapped to the same position were defined as a single SLAF locus (Zhang et al., 2015a). SNP loci at each SLAF locus were then detected between parents using the software GATK (https://www.broadinstitute.org/gatk/guide/best-practices?bpm=DNAseq#variant-disco very-ovw), and SLAFs with more than three SNPs were eliminated. Alleles were defined in each SLAF using the minor allele frequency evaluation. Because pear is a diploid species, one SLAF locus contains at most four genotypes, so SLAF loci with more than four alleles were defined as repetitive SLAFs and discarded. The alleles of each SLAF locus were then defined according to parental reads having a sequence depth greater than 10-fold. Only SLAFs with two to four alleles were identified as polymorphic and considered potential markers. The polymorphic SLAF loci were genotyped for consistency in the parental and offspring SNP loci. The average sequence depths of SLAF markers were greater than 33-fold in parents and greater than 5-fold in progeny.

All of the polymorphic SLAF loci were genotyped using the strategy described by Sun et al. (2013) and Zhang et al. (2015a). Prior to genetic map construction, the SLAF markers were filtered using a criteria detailed by Zhang et al. (2015a), And those with less than 99% integrity in the progeny were eliminated.

Sequence information on SSR primers were obtained from several published papers (Chen et al., 2014, FernÁNdez FernÁNdez et al., 2006, Inoue et al., 2006, Liebhard et al., 2002, Silfverberg Dilworth et al., 2006, Wang et al., 2016, Yamamoto et al., 2002b, Yamamoto et al., 2002a, Yue et al., 2014). The segregation data for SSR markers in the same population were collected, and 207 were used for constructing an integrated map. Microsatellites were PCR amplified in a Biometra TProfessional Thermocycler (Biometra Corp., Germany) using the following conditions: the reaction mixture was 20 μL, containing 2.0 μL of 10 × reaction buffer (100 mM Tris–HCl, pH 8.3), 1.6 μL 2.5 mmol L⁻¹ dNTP, 0.6 μL 10 mmol L⁻¹ of each primer, 0.5 U Taq polymerase (TaKaRa Bio Inc., Japan), and 50 ng genomic DNA. The PCR reactions were performed using the following conditions: 94 °C for 5 min, then 32 cycles of annealing at 55 °C for 30 s and extension at 72 °C for 45 s, followed by a final extension at 72 °C for 10 min. The PCR amplification products were analyzed by 8% polyacrylamide gel electrophoresis.

The marker segregation distortions were analyzed by Jonimap4.0 software. After data had been imported, marker segregation ratios were calculated using the chi-square test at different levels (P < 0.05, P < 0.01, P < 0.005). Then the genetic linkage map was constructed, based on the pear genome database, using the protocol detailed by Liu et al. (2014) and Zhang et al. (2015a). A HighMap strategy was used for ordering the SLAF and correcting genotyping errors within the linkage groups (LGs)(Liu et al., 2014). SMOOTH was also applied to remove genotyping errors based on the parental contribution to the genotypes of the offspring(Van Os et al., 2005), and a k-nearest neighbor algorithm was used to determine the missing genotypes(Huang et al., 2012). Skewed markers were then added to this map by applying a multipoint method of maximum likelihood. Map distances were estimated using the Kosambi mapping function(Kosambi, 1943).