Tree Genetics & Genomes

, Volume 4, Issue 4, pp 897–910

Construction of an intra-specific sweet cherry (Prunus avium L.) genetic linkage map and synteny analysis with the Prunus reference map

Authors

  • James W. Olmstead
    • Michigan State University
    • Washington State University
  • Audrey M. Sebolt
    • Michigan State University
  • Antonio Cabrera
    • Ohio Agriculture Research and Development CenterThe Ohio State University
  • Suneth S. Sooriyapathirana
    • Michigan State University
  • Sue Hammar
    • Michigan State University
  • Gloria Iriarte
    • Ohio Agriculture Research and Development CenterThe Ohio State University
  • Dechun Wang
    • Michigan State University
  • Charles Y. Chen
    • Michigan State University
    • USDA/ARS, National Peanut Research Laboratory
  • Esther van der Knaap
    • Ohio Agriculture Research and Development CenterThe Ohio State University
    • Michigan State University
    • Department of HorticultureMichigan State University
Original Paper

DOI: 10.1007/s11295-008-0161-1

Cite this article as:
Olmstead, J.W., Sebolt, A.M., Cabrera, A. et al. Tree Genetics & Genomes (2008) 4: 897. doi:10.1007/s11295-008-0161-1

Abstract

Linkage maps of the sweet cherry cultivar ‘Emperor Francis’ (EF) and the wild forest cherry ‘New York 54’ (NY) were constructed using primarily simple sequence repeat (SSR) markers and gene-derived markers with known positions on the Prunus reference map. The success rate for identifying SSR markers that could be placed on either the EF or NY maps was only 26% due to two factors: a reduced transferability of other Prunus-species-derived markers and a low level of polymorphism in the mapping parents. To increase marker density, we developed four cleaved amplified polymorphic sequence markers (CAPS), 19 derived CAPS markers, and four insertion–deletion markers for cherry based on 101 Prunus expressed sequence tags. In addition, four gene-derived markers representing orthologs of a tomato vacuolar invertase and fruit size gene and two sour cherry sorbitol transporters were developed. To complete the linkage analysis, 61 amplified fragment length polymorphism and seven sequence-related amplified polymorphism markers were also used for map construction. This analysis resulted in the expected eight linkage groups for both parents. The EF and NY maps were 711.1 cM and 565.8 cM, respectively, with the average distance between markers of 4.94 cM and 6.22 cM. A total of 82 shared markers between the EF and NY maps and the Prunus reference map showed that the majority of the marker orders were the same with the Prunus reference map suggesting that the cherry genome is colinear with that of the other diploid Prunus species.

Keywords

PrunusGenetic linkage mapSynteny analysisSweet cherry

Introduction

The genus Prunus contains many economically important tree fruit and nut crops including peach [Prunus persica (L.) Batsch], apricot (Prunus armeniaca L.), plum (Prunus salicina Lindl., Prunus domestica L.), cherry (Prunus avium L., Prunus cerasus, L.), and almond [Prunus dulcis (Miller) D.A. Webb]. The construction of genetic linkage maps for these species has and continues to be an important research goal to facilitate quantitative trait locus (QTL) analysis and gene tagging for use in marker-assisted selection. Prunus linkage maps are especially well developed in peach (Chaparro et al. 1994; Rajapakse et al. 1995; Dirlewanger et al. 1998; Abbott et al. 1998; Lu et al. 1998; Yamamoto et al. 2001; Dirlewanger et al. 2006), almond (Viruel et al. 1995; Joobeur et al. 2000), and apricot (Hurtado et al. 2002; Vilanova et al. 2003; Lambert et al. 2004; Dondini et al. 2007). The most detailed map is the Prunus reference map that was constructed from an inter-specific almond cv. ‘Texas’ × peach cv. ‘Earlygold’ (abbr. T × E) F2 mapping population (Joobeur et al. 1998; Aranzana et al. 2003). This reference map consists of 562 markers, 264 of which are simple sequence repeat (SSR) markers arranged in the expected eight linkage groups (G1–G8), covering 519 cM, with an average marker density of 0.92 cM (Dirlewanger et al. 2004a; Howad et al. 2005).

Comprehensive alignments of available Prunus linkage maps in the diploid peach, almond, and apricot (2n = 2x = 16) reveal a strong colinearity among the genomes (Dirlewanger et al. 2004a, b; Lambert et al. 2004; Dondini et al. 2007). In contrast, the synteny of a sweet cherry (2n = 2x = 16) linkage map with the T × E reference map remains to be thoroughly examined as comparative mapping with cherry has lagged behind other Prunus species due to the low number of mapped markers in cherry that are also on other Prunus linkage maps (Stockinger et al. 1996; Bošković and Tobutt 1998; Wang et al. 1998). More recently, partial linkage maps were constructed of two elite sweet cherry cultivars from 133 F1 progeny from a ‘Regina’ × ‘Lapins’ population in the Institut National de la Recherche Agronomique (INRA) of Bordeaux, France (Dirlewanger et al. 2004a). The ‘Regina’ and ‘Lapins’ maps have 30 and 28 SSRs in common with the T × E map and only one SSR was detected in a nonhomologous linkage group. These results are in agreement with the high level of synteny reported in Prunus (Dirlewanger et al. 2004a, b; Lambert et al. 2004; Dondini et al. 2007). However, more complete genome coverage in cherry is necessary to rigorously examine synteny across all eight linkage groups.

Reciprocal F1 intra-specific sweet cherry mapping populations were generated at Michigan State University to complement the other international efforts and provide a population suitable for QTL analysis of fruit traits that would have been improved through domestication, namely, fruit size and quality. Reciprocal crosses were made between the domesticated parent ‘Emperor Francis’ (EF) and the wild forest tree cherry ‘NY 54’ (NY). EF is an old landrace variety originating from Northern Europe that has fruit of approximately 6–8 g and has a yellow flesh with pink blush on the skin. Our primary reason for choosing EF as the domesticated parent was because it is a founding clone that is in the pedigree of almost every newly released sweet cherry cultivar (Choi and Kappel 2004). For example, EF is a great-grandparent of ‘Lapins,’ a parent of the INRA sweet cherry mapping population. Therefore, QTL validation could be greatly facilitated when using pedigree genotyping or association mapping approaches. NY is a wild “mazzard” clone originally obtained from a German forest and introduced at the New York State Agricultural Experiment Station, Cornell University (R.L. Andersen, personal communication). NY is typical of wild forest cherries, having small fruit (∼2 g) with black skin, purple flesh, and unpalatable taste. This particular clone was selected because its self-incompatibility alleles are S2S6, and it is therefore fully compatible with EF (S3S4) providing the capability to conduct a genetic analysis of those traits linked to the S-locus (Ikeda et al. 2005). S-allele genotyping of 511 progeny from the EF and NY populations previously determined that all four S-locus progeny classes (S2S3, S2S4, S3S6, S4S6) are equally represented. This is in contrast to many other Prunus mapping populations where only one pollen S-allele type is compatible in the style resulting in just two S-locus progeny classes (Foolad et al. 1995; Joobeur et al. 1998; Bliss et al. 2002; Foulongne et al. 2003; Vilanova et al. 2003; Lambert et al. 2004). However, it is more difficult to map other markers using our sweet cherry mapping population compared to the T × E mapping population as: (1) our cherry population is from an intra-specific cross and the T × E population is from an inter-specific cross and (2) our cherry population is an F1 and therefore a marker must be heterozygous in at least one parent to be mapped, while the T × E population is an F2 and therefore markers monomorphic in the peach and/or almond parents can also be placed on the linkage map.

SSR markers derived from either cDNA or genomic sequences have thus far been the markers of choice for Prunus linkage mapping. However, with the increase in available Prunus expressed sequence tag (EST) sequences, many of which have been assigned locations on the peach transcript map (Horn et al. 2005), we sought to increase the marker density of our cherry maps with the addition of gene-derived markers. Markers that are based on expressed gene sequences are particularly useful as they represent nonrepetitive regions of the genome and are sufficiently conserved across the Rosaceae family to be useful in synteny analysis. Moreover, gene-derived markers also permit future association studies that relate gene function to phenotypes of interest. The majority of the markers designed were cleaved amplified polymorphic sequence markers (CAPS; Konieczny and Ausubel 1993) and derived CAPS (dCAPS) markers that can detect single-nucleotide polymorphisms (SNP) that differentially affect the presence or absence of a restriction enzyme cut site (Michaels and Amasino 1998; Neff et al. 1998). Contrary to SSR markers that require high-resolution acrylamide-based systems to separate the fragments, CAPS and dCAPS can be visualized on lower-resolution agarose gel systems which are available in many laboratories.

The objectives of this work were (1) to construct two new sweet cherry genetic linkage maps from progenies generated from reciprocal crosses that would be useful for future QTL detection, (2) develop gene-derived markers that could be visualized on agarose gels to further increase marker density, and (3) compare marker orders and positions between the sweet cherry maps generated and the Prunus T × E reference map.

Materials and methods

Plant material

In 2001, pollen was collected from NY and EF trees in the National Research Support Project 5 (NRSP5) planting in Prosser, WA, USA. NY was used as a maternal parent in Washington State, and pollen was transported to Michigan for use in reciprocal crosses with EF as the maternal parent. From the crosses, 617 F1 individuals were planted at Michigan State University’s Clarksville Horticultural Experiment Station in Clarksville, MI, USA in the spring of 2002. The seedlings were planted at 1.5 and 6.1 m within and between row spacing, respectively. From the total population, a linkage mapping subset of 190 individuals was selected. This subset consisted of 86 individuals from the NY × EF cross, 103 individuals from the EF × NY cross, and one individual with no reciprocal cross information. Approximately equal numbers of progeny from each of the four S-allele groups (48, S2S3; 49, S2S4; 47, S3S6; 46, S4S6) were included in the mapping population. These four S-allele groups were shown previously to segregate according to the expected 1:1:1:1 Mendelian ratio (Ikeda et al. 2005).

DNA isolation

For DNA extraction, young unfolded leaves from the parents and each progeny individual were collected, placed immediately on dry ice, transported to the laboratory, and placed at −80°C for at least 24 h. Leaves were lyophilized for 48 h and stored long term at −20°C after which DNA was isolated using the cetyl trimethylammonium bromide method described by Stockinger et al. (1996).

DNA markers

A total of 433 SSR markers used for linkage map construction were derived from cDNA and genomic libraries of six Prunus species: P. armeniaca (apricot), P. dulcis (almond), P. persica (peach), P. salicina (plum), P. avium (sweet cherry), and P. cerasus (sour cherry; Table 1). SSR markers were screened using the two parents and six progeny individuals. For the majority of the SSR markers, the PCR conditions were constant except that the recommended annealing temperature for each individual primer pair was used: 94°C for 5 min, 35 cycles of 94°C (45 s), X°C (45 s), 72°C (90 s), and a final extension step of 72°C for 5 min, where X = the published optimum annealing temperature for each primer. For the EMPA and EMPaS primers, a touchdown polymerase chain reaction (PCR) temperature profile was used as described by Clarke and Tobutt (2003). The reaction mixture contained 1× PCR buffer, 2.5 mM MgCl2, 120 µM of each deoxynucleotide triphosphate, 2.5 pmol of each primer, 50 ng of genomic DNA, and 0.3 U Taq polymerase (Invitrogen Corporation, Carlsbad, CA, USA) in a 12.5-µl reaction.
Table 1

Origins and proportion of amplified, heterozygous, and mapped simple sequence repeat markers used in the development of the ‘NY 54’ × ‘Emperor Francis’ sweet cherry genetic linkage maps

SSR name

Prunus species

Number of SSRs

Reference

Origin

Tested

Amplifieda

Heterozygousb

Number Mapped (%)c

AMPA

P. armeniaca

Genomic

4

4

2

2 (50)

Hagen et al. 2004

BPPCT

P. persica

Genomic

35

31

14

13 (37)

Dirlewanger et al. 2002

CPDCT

P. dulcis

Genomic

22

16

5

3 (14)

Mnejja et al. 2005

CPPCT

P. persica

Genomic

27

25

8

8 (30)

Aranzana et al. 2002

CPSCT

P. salicina

Genomic

24

18

5

5 (21)

Mnejja et al. 2004

EMPA

P. avium

Genomic

11

11

5

4 (36)

Clarke and Tobutt 2003

EMPaS

P. avium

Genomic

7

7

4

4 (57)

Vaughan and Russell 2004

EPDCU

P. dulcis

cDNA

14

13

7

7 (50)

P. Arús (personal communication)

EPPB

P. persica

cDNA

16

16

5

5 (31)

E. Dirlewanger (personal communication); Dirlewanger et al. 2006

EPPCU

P. persica

cDNA

81

74

21

18 (22)

GDR database (http://www.bioinfo.wsu.edu/gdr/)

M

P. persica

cDNA

9

8

0

0 (0)

Yamamoto et al. 2002

MA

P. persica

cDNA

44

35

16

12 (27)

Yamamoto et al. 2002

MD

P. persica

cDNA

5

4

1

1 (20)

Yamamoto et al. 2005

PacC25

P. armeniaca

cDNA

1

1

0

0 (0)

Decroocq et al. 2003

PaCITA

P. armeniaca

Genomic

6

6

0

0 (0)

Lopes et al. 2002

Pce

P. cerasus

Genomic

7

7

3

3 (43)

Cantini et al. 2001; Struss et al. 2002

Pchcms

P. persica

cDNA

3

3

2

1 (33)

Sosinski et al. 2000

Pchgms

P. persica

Genomic

12

10

3

2 (17)

Sosinski et al. 2000

PMS

P. avium

Genomic

8

8

4

4 (50)

Cantini et al. 2001; Struss et al. 2002

PS

P. cerasus

Genomic

5

5

3

2 (40)

Cantini et al. 2001; Sosinski et al. 2000

UCD-CH

P. avium

Genomic

5

4

2

2 (40)

Struss et al. 2003

UDA

P. dulcis

Genomic

29

25

5

5 (17)

Testolin et al. 2004

UDAp

P. armeniaca

Genomic

33

29

9

4 (12)

Messina et al. 2004

UDP

P. persica

Genomic

25

20

7

6 (24)

Cipriani et al. 1999; Testolin et al. 2000

  

Total

433

380 (88%)

131 (30%)

111 (26)

 

aSSR markers that did not amplify a product with EF or NY DNA are listed in Supplemental Table 1.

bSSR markers that were monomorphic in EF and/or NY are listed in Supplemental Table 2.

cPercentage of markers that were mapped of the total markers tested.

Amplified fragment length polymorphism (AFLP) analysis consisting of genomic DNA digestion with EcoRI and MseI restriction enzymes, adapter ligation, preamplification, and selective amplification using EcoRI plus two and MseI plus three selective nucleotide primers were similar to those from Vos et al. (1995) with modifications described by Hazen et al. (2002).

Two sequence-related amplified polymorphism (SRAP) forward and reverse primer combinations me1–em1 and me1–em2 were used as reported by Li and Quiros (2001); me1 = 5′-TGAGTCCAAACCGGATA, em1 = 5′-GACTGCGTACGAATTAAT, em2 = 5′-GACTGCGTACGAATTTGC. DNA amplification was performed as in Li and Quiros (2001).

Gene-derived markers

The S-locus was scored using the S-RNase primer pair Pru-C2 and PCE-R (Tao et al. 1999; Yamane et al. 2001) under previously described conditions (Ikeda et al. 2005). The S2-RNase-specific PCR primer pair, PaS2-F and PaS2-R, was used for confirmation of S2 presence (Sonneveld et al. 2001). Primer sequences for PrpFT, a peach homolog to a Malus × domestica T-like protein, were also utilized (Silva et al. 2005). In addition, the following three candidate genes were developed into markers: INV5, corresponding to a vacuolar invertase (Fridman et al. 2000) and SorT1 and SorT2, corresponding to sorbitol transporters (Gao et al. 2003).

EST sequences were selected from the bin map as well as the Prunus physical map which were both anchored to the genetic T × E map (http://www.bioinfo.wsu.edu/gdr/). Intron regions in the genes were predicted by comparing each sequence to the Populus trichocarpa (poplar) genome available in the Joint Genome Institute database using the P. trichocarpa v1.1 browser (http://genome.jgi-psf.org/cgi-bin/runAlignment?db=Poptr1_&advanced=1). ESTs showing high similarity to poplar genomic sequences (Basic Local Alignment Search Tool nucleotide vs nucleotide at the cutoff E-value of 1e−5) were selected to design primer flanking the putative introns using Primer3 v 0.3.0 (http://frodo.wi.mit.edu/). For the genes that showed lower similarity to poplar, primers were designed irrespective of putative intron position. Products were amplified from genomic DNA of each parent using the primers and resulting fragments were directly sequenced at the Molecular and Cellular Imaging Center, in Wooster, OH, USA. Sequences were aligned to each other and compared to the original EST using Sequencher software v 4.2 (Gene Codes Corporation) to detect SNPs and/or insertions–deletions (InDels). The presence of double peaks in an otherwise high-quality chromatogram revealed the presence of a potential SNP. The sudden decay of high-quality chromatogram revealed a potential InDel. CAPS and dCAPS markers were developed following standard protocols (Konieczny and Ausubel 1993; Michaels and Amasino 1998).

A marker was also derived from a cherry ortholog of the tomato FW2.2 (Frary et al. 2000) using the following strategy. Two translated poplar scaffold sequences (scaffolds 2121 and 678) for which the translated sequence showed homology to tomato FW2.2 were aligned to find conserved protein sequences using the Block Maker server (http://blocks.fhcrc.org/blocks/make_blocks.html). Highly conserved regions were used to design the degenerate primers PR8 (ACTTACTGGTGCCCATGCATHACNTTYGG) and PR9 (CGAATCCTCTATG CTTCAGCTCTCKRTAYTCYTG) using the Consensus-Degenerate Hybrid Oligonucleotide Primers server (http://blocks.fhcrc.org/blocks/codehop.html). A test PCR using the gradient function on the MJ Research thermocycler was conducted to find the appropriate annealing temperature for amplification. PCR reactions were performed using Phusion HF DNA Polymerase (New England, USA, BioLabs) under conditions recommended by the manufacturer. PCR products were phosphorylated and ligated into the HincII site of pUC118. The ligation products were transformed into Escherichia coli DH5α electrocompetent cells. Five clones from each parent harboring the expected size of the PCR products were selected; the plasmids were isolated and inserts were sequenced. Sequences were aligned using Sequencher software v 4.2 in order to develop locus-specific primers (PR26F and PR26R). These primers were used to amplify the alleles from each parent and the resulting product was cloned into the HincII site of pUC118. Eighteen clones from each parent were sequenced for detection of the SNP that distinguished the haplotypes.

Genotyping platforms

AFLP and SRAP fragments were run on a 6% denaturing polyacrylamide gel in a 50-cm Sequi-Gen GT vertical sequencing apparatus (Bio-Rad Laboratories, Hercules, CA, USA) for 2.5 h at 70 W with 1× Tris–borate–ethylene diamine tetraacetic acid (TBE) buffer. After electrophoresis, the bands were visualized by silver staining (Promega Corporation, Madison, WI, USA). SSR and InDel markers were size-separated by electrophoresis on a 6% denaturing polyacrylamide gel using two different platforms. Fragments separated on the Sequi-Gen GT vertical sequencing apparatus were visualized by silver staining. Fragments separated on the LI-COR IR2 DNA Analyzer system (LI-COR. Lincoln, NE, USA) were detected after excitation of the fluorescently labeled fragments. These fluorescent fragments were generated using either the IRDye800 or IRDye700 universally labeled M13 (−29) primer in the PCR reaction following the M13-tailed PCR protocol (Schuelke 2000). The CAPS and dCAPS markers were separated on a 3–4% agarose gel in 0.5× TBE buffer.

Chi square analysis and linkage map construction

All fragments were scored as present or absent. Segregating fragments present in one parent and absent in the other parent were tested for fit to a 1:1 ratio, while segregating fragments present in both parents were tested for fit to a 3:1 ratio. Chi square goodness-of-fit tests were performed using functions in Excel 2002 (Microsoft Corp., Redmond, WA, USA).

Linkage analysis was performed with JoinMap 3.0 (Van Ooijen and Voorrips 2001) using a minimum logarithm of odds score of 3.0, a maximum recombination fraction of 0.4, and centimorgan distances calculated by the Kosambi (1944) function. Linkage groups were drawn using MapChart 2.1 (Voorrips 2002) with distances presented in centimorgan.

Results and discussion

Transferability and polymorphism of SSR markers

All but 32 of the SSR markers placed on the T × E genetic map and for which primer sequences were publicly available were tested on EF, NY, and a subset of six progeny to determine their suitability for linkage map construction. To fill gaps in the maps, additional SSR markers with known map locations from other Prunus maps were also tested to determine if they were heterozygous within EF and/or NY. In total, 433 SSR markers that were derived from six Prunus species and 24 genomic and cDNA libraries were screened (Table 1). Of the 43 SSR primers derived from cherry sequence (P. avium and P. cerasus), only one failed to amplify a fragment from NY and EF. However, the amplification success rate was considerably less with the other Prunus species primers. In total, 12% of the screened SSR markers failed to amplify products (see Supplemental Table 1). Primer pairs derived from genomic or cDNA sequences performed similarly, as 87% and 89% of the primer pairs tested, respectively, amplified a product from cherry DNA.

In addition to the reduced amplification rates, the number of SSR markers suitable for comparative mapping was further reduced due to the low level of heterozygosity in the parents (Table 1). Of the amplified SSR products, only 35% and 34% of the primer pairs designed from genomic or cDNA sequences, respectively, were heterozygous in EF and/or NY. Therefore, of the 433 SSR primer pairs tested, 307 either did not amplify a product or were homozygous. Some of the monomorphic SSR primer pairs could be suitable in other cherry germplasm; therefore, a list of the monomorphic SSR markers is provided in the supplemental materials (see Supplemental Table 2). In addition, 15 primer pairs that identified 18 loci on the T × E map were not used as they provided complex patterns that could not be reliably scored (see Supplemental Table 3 for a list of these markers).

Taken together, the success rate for identifying SSR markers from the T × E map and other Prunus maps that could be placed on either the EF or NY map was only 26%. Several factors are likely contributing to this low success rate: the limited number of mapped cherry SSR markers, the reduced transferability of SSR markers from other Prunus species to cherry, and, lastly, the low level of heterozygosity in the mapping parents. As reported previously, amplification of cherry DNA with primers that were derived from other Prunus species resulted in the lowest level of success within the genus (Dirlewanger et al. 2006; Dondini et al. 2007). For example, the success of SSR amplification in peach relative to the source DNA was as follows: 95.6% for apricot, 95% for almond, 90% for Japanese plum, and 81.5% for sweet cherry (Dirlewanger et al. 2006). In another study, only 19% of the SSRs derived from cherry were suitable for mapping in apricot (Dondini et al. 2007). Cherry is in the subgenus Cerasus which is phylogenetically distant from the other Prunus species that are placed within the subgenera Prunophora (= Prunus; plum and apricot) and Amygdalus (peach and almond). An analysis of chloroplast DNA polymorphisms placed the subgenera Prunuophora and Amygdalus in a separate clade, referred to as the Pruno-Amygdaloid (peach-almond–apricot–plum) clade, that is more distantly related to the CerasusLaurocerasusPadus (cherry–laurel-cherry) clade (Shaw and Small 2004).

Our difficulty in finding heterozygous SSR markers for these parents is in agreement with that for ‘Lapins,’ a self-fertile cultivar and one of the parents in the INRA mapping population (Dirlewanger et al. 2004a). Of 25 polymorphic single-locus SSR markers derived from almond, five (20%) either did not amplify or produced complex patterns unsuitable for genetic analysis and, of the remaining 20 SSR markers, ‘Lapins’ was heterozygous for only five (25%) of them (Mnejja et al. 2005). By comparison, the observed heterozygosity in sweet cherry has been reported to range from 0.49 to 0.90 (Dirlewanger et al. 2002; Wünsch and Hormaza 2002; Schueler et al. 2003, 2006; Pedersen 2006; Stoeckel et al. 2006; Marchese et al. 2007). However, several of these studies only reported the heterozygosity for SSR markers selected based on prior knowledge of allelic polymorphism. Therefore, when the original numbers of nonamplifying or monomorphic SSR markers used in these studies were reported and therefore could be added to the calculation of heterozygosity, the observed sweet cherry heterozygosity ranged from only 0.22 to 0.32 (Dirlewanger et al. 2002; Wünsch and Hormaza 2002; Schueler et al. 2003). This suggests that the low level of SSR polymorphism in sweet cherry is not unique to EF and NY. However, in natural sweet cherry populations, heterozygote excess has been documented and has been suggested to be due to asexual reproduction and the ability to maintain heterozygosity over clonal generations (Stoeckel et al. 2006). NY was selected as a parent because it is fully compatible with EF and as a representative of wild germplasm. However, of the eight markers used by Stoeckel et al. (2006) to survey wild sweet cherry, six were homozygous in NY.

Gene-derived markers

To increase marker coverage, ESTs were selected that were predicted to map to low marker density regions due to their previous placement on the T × E genetic map or the peach physical map. ESTs have the advantage that they correspond to genes as opposed to the random genomic sequences that are represented by the SSR and AFLP fragments. Moreover, gene-derived markers will aid future studies involving gene function and the observed phenotypes. To increase the chance of finding polymorphisms in genic regions, we attempted to design primers that flank introns based on predictions made from the poplar genomic sequences. From the 101 expressed genes for which we obtained amplification products, a total of 23 CAPS and/or dCAPS and four InDels were developed that were mapped in the EF and NY population (Table 2). Thus, the marker discovery rate on a per gene basis was 23%. Nine of the resulting EST-derived markers (PR33, PR41, PR51, PR56, PR69, PR70, PR72, PR74, and PR84) were designed from ESTs that had been previously placed on the Prunus genetic map, while 14 of the EST-derived markers (PR85, PR86, PR90, PR93, PR96, PR98, PR101, PR103, PR110, PR117, PR121, PR122, PR126, and PR127) were designed from ESTs that had previously been placed on the peach physical map. All together, these 23 newly designed “PR” markers represent successfully converted peach markers that can be utilized in cherry for linkage and comparative mapping. Eleven of the PR EST-derived markers identified two alleles in NY and EF while one PR marker identified two alleles in both parents (Table 3).
Table 2

Primer sequences and diagnostic restriction enzymes for two InDel markers and 21dCAPS markers developed from EST sequences

Marker name

Accession no.

Marker type

Enzyme

Sequence of forward primera

Sequence of reverse primer

PR33

BU042407

Indel

CGTTACAGATTGGTTGACCTGTGA

ATCAGGCTATGCACACTCTT

PR41

AJ854221

dCAPS

Hinf I

CATTGCAAAATCTCACAGAA

CAATTATGATCTAAATAGAGGACT

PR51

AJ854216

dCAPS

Hinf I

GGTGTCATAAGGAATTCCCCGACT

TGCAAGATATTCTTCCCAGT

PR56

BU044092

CAPS

Tru1 I

TTGGTCTCGTGGTGGATAGA

TTGGCTGTGACACCCTTCTT

PR69

BU046536

CAPS

Dde I

AGCCTTTGCGTGAACAACTT

CGATGCTGGAAAGAAGAAGC

PR70

BU039366

dCAPS

Hinf I

CTGGGACCGAAGCACTTCTA

GATGCTCAGGGCTGAAAAAG

PR72

BU043855

dCAPS

Hinc II

AATGAGCAGAAACTGTTACGTTGA

GCAAGACATCTCCCAATTGAT

PR74

BU039994

CAPS

Rsa I

GGTGGTTTGTTGGGTTGAAA

TCGAGGATGGGATTCAAAAG

PR84

BU040863

dCAPS

Cac8 I

GGATGCATTTGAAAGGGATTT

ATTGGGTACTCTTGAGGGGCTGG

PR85

BU042636

dCAPS

Dde I

AGGATTCCATTTGGTGCTCA

TCTGTGTACTTCCACTATCTCCTT

PR86

BU048031

dCAPS

BamH I

AGTACAATTAGTTGTGGGCGGATC

TGCTTTGCTTCAGCATGTTC

PR90

BU041499

dCAPS

Hinc II

GCTTTTATGCGTTTCGGTTG

ATCACAACAAATACAGTACGTCAA

PR93

BU042551

dCAPS

BsaJ I

TAGAACTTGTAGGGGTTGTCCCTG

CACCCAGCCTACCTCTCGTA

PR96

BU039798

dCAPS

Tai I

GGTTGCATGTTGATTGATGG

TCATGCTAATATTAACAGATAACG

PR98

BU039221

CAPS

Rsa I

GCACTGGGCTTCTTGTTGTA

CGACCACATTGCGAAGTAGA

PR101

BU043277

Indel

CACGACGTTGTAAAACGACTGCGATGTGTTTGGTTGTCT

GTCCAGATGAGCCTTCAAGC

PR103

BU039816

dCAPS

Bcc I

CCCAGGCTTTAAGTGTGCTC

GAGTGAGGAGGAGGCTACCCTCCA

PR110

BU041732

dCAPS

BsaJ I

GGAAGTGGGGCAAAAACTGT

GGTACAATGCAGATATTAACCAGG

PR117

BU042018

dCAPS

BsaJ I

TCAAAGTTGTCACGTACTAACCTA

TGGTCCACGTCCATTGATTA

PR121

BU044081

dCAPS

Apo I

CCCTGACGTTCTGGATGATT

TGCAAGTTAGTTCTCTAGGCAAAATT

PR122

BU048663

dCAPS

Bcc I

TGTTTTCCCATTTGGTTATCACC

AAAGTGCAGACATCCTTGGAG

PR126

BU042394

dCAPS

Hpa II

TAGATTTTGATTTCCATAAAACCG

CCTAAACAGATATGACCGTTGC

PR127

BU045325

dCAPS

Taq I

GCTCAATGATGCCTTATGC

GACCAACATAATACTTCTCAAATC

PR25/22

AF482011

Indel

CACGACGTTGTAAAACGACATAAAAATGGCCGTAGGTG

TCCATATCATCTCATCTCCAC

PR26

NA

dCAPS

Bgl I

CACGACGTTGTAAAACGACTATTCTTGACAAATTGCCCAACTGG

TTGATGCATGGCAATGTTAAG

PR27

AY100638

dCAPS

Bgl II

CACGACGTTGTAAAACGACATGATCGCTCCGGTCTACACCGCCGAGAT

CCAAAACGAAAATTCCCAAGT

Inv 5

AAL05427.2

InDel

CACGACGTTGTAAAACGACAACCAGAACGTAATTGGATG

TGGAGACATATCTAACCTACCA

aPrimer sequences (5′ to 3′) read left to right. For those primer sequences that are longer than 25 nucleotides, the primer sequence is continued on a second line.

Table 3

Locations of the gene-derived markers on the T × E bin or physical maps compared to their locations on the ‘Emperor Francis’ (EF) and ‘NY54’ (NY) linkage groups

Marker name

Reference on T × E bin or physical map

Heterozygous parent(s)

EF and/or NY G and cM location(s)

T × E bin or location in physical map

Reference

PR33

EPPCU2407

EF

EF G1, 56.4

1:28a (26.5 to 27.9)b

Howad et al. 2005

PR41

EPPB4221

EF

EF G3, 43

3:37 (24.8 to 36.4)

E. Dirlewanger, GDR

PR51

EPPB4216

EF & NY

NY G5, 19.6

5:49 (49.1)

E. Dirlewanger, GDR

   

EF G5, 67.8

  

PR56

EPPCU4092

NY

NY G6, 80.8

6:80 (78.8 to 79.6)

Howad et al. 2005

PR69

EPPCU6536

EF

EF G2, 8.7

2:08 (0 to 8.1)

Howad et al. 2005

PR70

EPPCU9366

NY

NY G2, 2.8

2:08 (0 to 8.1)

Howad et al. 2005

PR72

EPPCU3855

NY

NY G6, 4.2

6:25 (0 to 24.9)

Howad et al. 2005

PR74

EPPCU9994

NY

NY G3, 30.7

6:25 (0 to 24.9)

Howad et al. 2005

PR84

EPPCU0863

EF

EF G5, 46.9

5:41 (21.7 to 40.7)

Howad et al. 2005

PR85

PP_LEa0013G08f

NY

NY G6, 0

6: 53.6

GDR

PR86

PP_LEa0033P09f

NY

NY G6, 60.8

6: 56.4

GDR

PR90

PP_LEa0009I13f

NY

NY G5, 1.0

6: 63.4

GDR

PR93

PP_LEa0013A14f

NY

NY G6, 80.8

1: 40.5, 6: 39.3

GDR

PR96

PP_LEa0003O11f

EF

EF G2, 21.3

2: 19.2

GDR

PR98

PP_LEa0001K03f

NY

NY G1, 137.8

2: 19.2

GDR

PR101

PP_LEa0015I11f

NY

NY G1, 73.6

2: 19.2, 5: 13.4

GDR

PR103

PP_LEa0003P11f

NY

NY G7, 25.5

2: 19.2

GDR

PR110

PP_LEa0010E22f

EF

EF G3, 49.2

2: 25

GDR

PR117

PP_LEa0011D19f

EF

EF G8c

2: 19.2

GDR

PR121

PP_LEa0018B24f

EF

EF G6, 75.9

2: 7.9

GDR

PR122

PP_LEa0036J22f

EF

EF G1, 140.9

2: 7.9

GDR

PR126

PP_LEa0012H09f

EF

EF G7, 49.0

2: 27.8

GDR

PR127

PP_LEa0022D10f

EF

EF G6, 26.4

1: 72.9, 2: 38

GDR

PR25/22

NA

EF

EF G8, 28.6

NA

GDR

PR26

NA

EF

EF G5, 69.3

NA

GDR

PR27

NA

NY

NY G8, 35

NA

GDR

Inv 5

NA

EF

EF G2c

NA

GDR

GDR Genome Database for Rosaceae (http://www.bioinfo.wsu.edu/gdr), NA not applicable (markers not present in the T × E reference map)

aBin feature name represented by chromosome number: end of the bin (cM).

bStart and end of the bin on the corresponding chromosome (cM).

cThese markers grouped in the linkage groups listed but could not be placed with confidence on the linkage map due to high chi square values.

Two allelic variants were identified for four candidate genes using newly designed primer pairs based on cherry sequence (Table 2). The SorT1 (PR25/22) and Inv5 allelic variants were based on InDels, while the SorT2 (PR27) and FW2.2 (PR26) variants were based on SNPs.

AFLP and SRAP markers

To further increase marker density, AFLP and SRAP markers were employed. By using eight different EcoRI and MseI selective primer combinations, 72 polymorphic fragments were identified (Table 4). Depending upon the selective nucleotides used, the number of polymorphic fragments ranged from four to 18 with an average of nine polymorphic fragments per primer combination. This was higher than the average of 6.8 reported in peach by Lu et al. (1998) who reported that the use of EcoRI and MseI with two and three selective nucleotides, respectively, offered the best compromise between the number of polymorphic fragments produced and the ease of scoring.
Table 4

Enzymes used for digestion, selective nucleotide combinations used as primers, number of polymorphic fragments, and number of mapped fragments generated by amplified fragment length polymorphism analysis in the development of the ‘NY 54’ × ‘Emperor Francis’ sweet cherry genetic linkage maps

Selective nucleotides

Number of polymorphic fragments

Number of mapped fragments

EcoRI

MseI

AA

CTT

6

5

AA

CAC

9

9

AA

CCC

8

7

AA

CCT

6

4

AA

CAA

4

2

AT

CTC

18

15

AT

CCC

9

7

AC

CTA

12

12

 

Total

72

61 (84.7%)

Two SRAP primer combinations identified seven polymorphic fragments. Six of these seven fragments were identified using the reverse primer em2 which had the selective nucleotides TGC at the 3′ end.

Map construction

The EF and NY linkage maps consisted of the expected eight linkage groups and totaled 711.1 and 565.8 cM, respectively (Table 5, Fig. 1). This compares well to a recent intra-specific map of peach which totaled 621.2 cM (Dirlewanger et al. 2006). In addition, the centimorgan size for the EF linkage map is consistent with the predicted genome size of sweet cherry which is slightly larger than peach (6.6 × 108 vs 5.3 × 108) (Dickson et al. 1992). The nomenclature and orientation of the linkage groups were according to that assigned to the T × E map (Joobeur et al. 1998). The average distances between markers were 4.94 and 6.22 cM for the EF and NY linkage maps, respectively. The largest gaps between markers were present on G1, the longest Prunus linkage group, and were 29.5 and 33.5 cM for EF and NY, respectively (Table 5). The maps consisted of a total of 197 markers, of which 102 were SSRs, 61 were AFLPs, seven were SRAP markers, and 27 were gene-derived markers. Of the 27 gene-derived markers, 25 were mapped (Fig. 1). Inv5 and PR117 could not be mapped with high accuracy due to skewed segregation of these markers.
https://static-content.springer.com/image/art%3A10.1007%2Fs11295-008-0161-1/MediaObjects/11295_2008_161_Fig1a_HTML.gifhttps://static-content.springer.com/image/art%3A10.1007%2Fs11295-008-0161-1/MediaObjects/11295_2008_161_Fig1b_HTML.gif
Fig. 1

Alignment of ‘Emperor Francis’ (EF) and ‘New York 54’ (NY) sweet cherry parental maps with the Prunus reference and bin map [‘Texas’ almond × ‘Earlygold’ peach (T×E)]. Only common markers present on EF and NY maps are presented on the Prunus reference map. Shaded areas on the reference map linkage groups indicate Prunus bin locations. Solid lines indicate homology between mapped markers; dashed lines indicate markers present in Prunus bins. Boxed markers indicate anchor points between the EF and NY parental maps. Markers followed by an asterisk indicate significant deviation (P < 0.05) from the expected chi-square segregation value

Table 5

Number and type of marker, length, marker density and marker gap length for the ‘Emperor Francis’ (EF) and ‘NY 54’ (NY) maps and their individual linkage groups

 

Parental maps

Linkage groups

Marker type

Total

EF

NY

EF1

NY1

EF2

NY2

EF3

NY3

EF4

NY4

EF5

NY5

EF6

NY6

EF7

NY7

EF8

NY8

SSR

102

84

51

13

7

15

8

13

5

8

5

11

4

11

10

8

8

5

4

AFLP

61

41

23

5

4

4

4

6

2

3

0

4

0

6

4

2

3

11

6

SRAP

7

4

3

0

0

0

0

0

0

0

0

1

0

1

0

2

1

0

2

Gene-derived

27

15

14

2

2

2

1

2

1

0

0

3

2

4

6

1

1

1

1

Total

197

144

91

20

13

21

13

21

8

11

5

19

6

22

20

13

13

17

13

Map statistics

Length (cM)

 

711.11

565.8

143.0

147.9

96.0

91.5

82.4

30.7

84.5

30.9

83.5

20.8

75.7

84.1

83.6

73.2

62.4

86.7

Marker density (markers/cM)

 

0.2

0.16

0.14

0.09

0.22

0.14

0.25

0.26

0.13

0.16

0.23

0.29

0.29

0.24

0.16

0.18

0.27

0.15

Average marker distance (cM)

 

4.94

6.22

7.2

11.4

4.6

7.0

3.9

3.8

7.7

6.2

4.4

3.5

3.4

4.2

6.4

5.6

3.7

6.7

Largest gap (cM)

 

29.5

33.5

29.5

33.5

18.3

25.0

16.3

10.2

13.2

18.1

11.1

11.4

7.1

21.4

16.3

11.1

9.2

16.7

Gaps > 15 cM (n)

 

18

11

2

5

1

1

1

0

0

1

0

0

0

1

2

0

0

1

Average cM/linkage group

 

88.9

70.7

                

A total of 49 (34%) and 16 (18%) skewed loci were placed on the EF and NY linkage groups, respectively. Linked markers that significantly deviated from their chi-square expectations identified regions of the linkage groups that were skewed towards favoring a particular homolog. This was especially apparent for EF G2, G6, and G8 and NY G1. The percentage of skewed loci in EF was higher than that reported from other intra-specific Prunus crosses: 15% to 18% in peach (Lu et al. 1998; Dettori et al. 2001), 10% in almond (Joobeur et al. 2000), and 11% to 14% in apricot (Hurtado et al. 2002; Vilanova et al. 2003). The region flanking the self-incompatibility locus (Si) on G6 for both EF and NY did not contain skewed markers unlike the homologous region on other Prunus linkage maps (Foolad et al. 1995; Joobeur et al. 1998; Bliss et al. 2002; Foulongne et al. 2003; Vilanova et al. 2003; Lambert et al. 2004). In addition, our ability to map Si for both EF and NY is in contrast to the partially incompatible ‘Regina’ × ‘Lapins’ population (S1S3 × S1S4’) where the S-locus would only be placed on the ‘Regina’ map as all the progeny would have the S4’ allele from ‘Lapins.’

Because the NY and EF populations consisted of reciprocal crosses, we were able to examine the influence of gamete sources from each parent on the observed marker distortion. A 30-cM span from the top of EF G2 to the marker UDAp-461 was the only region where linked markers were distorted in both reciprocal crosses. In all other cases, segregation distortion was only present from the pollen parents (data not presented), in that one of the two alleles in the pollen parent was present in more than 50% of the offspring. This observation is similar to that described by Foulongne et al. (2003) in which gametophytic selection causing distorted segregation in a peach × P. davidiana F2 population was assumed to occur only among male gametes.

For both EF and NY, the marker profile on G8 was unique in that SSR and gene-derived marker polymorphisms were low resulting in only six and seven markers on the EF and NY linkage groups, respectively. However, G8 was supplemented with 11 and six AFLP markers in EF and NY, respectively, resulting in a well-built linkage group. In contrast, in a peach F2 population (Ferjalou Jalousia × Fantasia), G8 was not identified at all as all of the 181 molecular markers mapped to the first seven linkage groups (Dirlewanger et al. 2006). Similarly, G8 could not be resolved from a BC2 population from P. persica × P. davidiana as only one G8 marker was polymorphic (Quilot et al. 2004). Interestingly, two of our candidate gene-derived markers representing sorbitol transporters SorT1 (PR25/22) and SorT2 (PR27) mapped to EF G8 and NY G8, respectively (Table 4, Fig. 1). The cherry ortholog of FW2.2 (PR26) was placed on EF G5.

Despite screening many publicly available SSR markers that were placed on the sweet cherry cv. ‘Regina’ and ‘Lapins’ partial linkage maps, only 16 SSR markers were in common with those placed on the EF and NY maps (Dirlewanger et al. 2004a). The maximum number of shared markers per linkage group was four for EF G2 and only three other linkage groups shared three common markers (EF G1, NY G2, and EF G5). In most cases, marker order was in agreement. Interestingly, two markers, BPPCT034 and BPPCT002, were in a similar order in NY G2 and ‘Lapins’ G2 but a reversed order in EF and ‘Regina.’

The analysis of the shared markers between the NY and EF parents showed a conservation of marker and gene order for the two genomes. However, the NY parent exhibited several linkage groups smaller than those of EF. In particular, the NY linkage groups 3, 4, and 5 were less than half the centimorgan lengths of the respective EF linkage groups. Based on an analysis of the shared markers, this can be explained by the low overall level of heterozygosity of the NY parent. This analysis allowed us to determine that the partial G3 and G4 identified for NY represented the top segments of these linkage groups, whereas the partial NY G5 represents the bottom segment of G5. Our inability to resolve these genomic regions for NY suggests that NY may be homozygous for substantial regions of these three linkage groups. This was unexpected in sweet cherry as it is an obligate outcrossing species with an active gametophytic self-incompatibility system. This outcrossing mechanism is controlled by alleles at the S-locus on G6; therefore, this region of G6 must by definition be heterozygous. Interestingly, NY G6 is longer than EF G6 (84.1 cM compared to 75.7 cM), suggesting that the self-incompatibility system has maintained heterozygosity for the S-locus and linked regions while other regions of the sweet cherry genome may have become homozygous.

Alignment with the T × E Prunus reference map and other Prunus maps

A total of 82 markers were shared between the EF and NY linkage maps and the T × E linkage and bin maps (Fig. 1). The number of shared markers for the eight linkage groups were as follows: G1, 12; G2, 14; G3, 9; G4, 7; G5, 10; G6, 12; G7, 10; and G8, 8. Nine of these 82 shared markers represented previously mapped peach ESTs, with the prefix EPPCU or EPPB (Table 1) for which cherry InDel, CAPS, or dCAPs markers had been designed. Of these nine markers, only one, PR74 (EPPCU9994), mapped to an alternate linkage group (Table 6) and PR41 (EPPB4221) mapped to an additional group in NY (Table 4) compared to its location on the T × E map. The other previously mapped gene-derived markers, PrpFT and the Si S-RNase, also mapped to their predicted locations on G6 (Fig. 1; Silva et al. 2005).
Table 6

List of markers that were placed on different ‘Emperor Francis’ (EF) and ‘NY 54’ (NY) linkage groups (G) compared to other Prunus maps with the exception of the cherry markers developed from physically mapped peach ESTs

Marker

G in EF and/or NY

G in other Prunus maps (reference)

UDAp-426

G1: NY

G7: Bin Mapa, b; LG1: L, Bc

BPPCT006

G2: EF; G6: NY

G8: T × Ea; G1: P2175d; G3: Bc

EPPCU8702

G2: EF

G3: Bin Mapa, b

UDA-059

G2: EF

G1: Bin Mapa, b

UDP98–416

G3: EF

G6: GNd

EPPCU9994 (PR74)

G3: NY

G6: Bin Mapa, b

UDA-037

G4: EF

G2: Bin Mapa, b

BPPCT013

G4: EF

G2: T × Ea; G2: GN & P2175d

UDP96–019

G5: EF

G8: T × E, Bin Mapa, b; G5: P2175d

CPPCT029

G6: EF

G1: T × Ea; G2: GNd

UDAp-471

G6: NY

G1: Bin Mapa, b; G7: L & Bc

All the markers are SSRs derived from genomic sequences except for the two EPPCU markers that are derived from EST sequences

aT × E Bin Map (Genome Database for Rosaceae, http://www.bioinfo.wsu.edu/gdr)

bT × E Bin Map (Howad et al. 2005)

cL and B are maps derived from the apricot cultivars ‘Lito’ and ‘BO81604311,’ respectively (Dondini et al. 2007)

dP2175: map constructed from myrobalan plum, Prunus cerasifera, clone number P2175; GN map constructed from the almond × peach F1 hybrid between Garfi × Nemared (Dirlewanger et al. 2004b)

Of the 102 SSR markers located on the EF and NY linkage maps, only ten were placed on different linkage groups compared to the T × E map, bin map, apricot map (Dondini et al. 2007), or the interspecific myrobalan plum × (peach × almond) maps (Dirlewanger et al. 2004b; Table 6). BPPCT006 has previously been shown to be a multilocus marker and therefore could be expected to map elsewhere. UDAp-426 and UDAp-471 were placed in different linkage groups in apricot compared to T × E and BPPCT013, UDP96–019 and CPPCT029 were placed in different linkage groups compared to plum and/or a peach × almond hybrid. It is possible that we have found additional loci in the cherry genome which would be one of the explanations for the discrepancy in marker position. The fact that, of these ten markers, all but two (EPPCU8702 and EPPCU9994) were derived from genomic DNA also suggests that these markers could represent repetitive regions within the Prunus genome.

Within each linkage group, an analysis of the 82 SSR and gene-derived markers shared between the EF and NY maps and the T × E linkage and bin maps revealed extensive conservation of marker order (Fig. 1). However, notable exceptions were observed. On EF G4, the marker BPPCT035 was placed at position 0.0 cM, but it is at position 46.4 cM on the T × E map. As the segregating fragment size of BPPCT035 from EF was large (∼1,000 bp), it is possible that this marker identified an additional locus in EF. Another exception was MA014a that was placed at 22.9 cM on EF G6 and bin location 6.74 (G6 72 cM to 74.3 cM) on the T × E map. However, by far, the majority of the markers that were derived from other Prunus genetic maps mapped in the expected order in sweet cherry. Thus, our results demonstrate significant colinearity of the cherry and other Prunus species genomes.

The colinearity exhibited in sweet cherry using markers placed on Prunus linkage and bin maps was in stark contrast to our results using 14 EST-derived markers from the Prunus physical map (Table 4, PR85, PR86, PR90, PR93, PR96, PR98, PR101, PR103, PR110, PR117, PR121, PR122, PR126, and PR127). The map positions of these 14 ESTs were based on hybridization to bacterial artificial chromosome (BAC) clones that carry peach genomic DNA inserts (Zhebentyayeva et al. 2006). These BAC clones were then anchored to the Prunus reference genetic map. However, most of the ESTs that were anchored to the physical map did not correspond to the expected location (Table 4). Only two (15%) of these ESTs mapped to the expected position (PR86 and PR96), whereas another two ESTs mapped to the predicted linkage group but not at the expected position (PR85 and PR93). As the comparison of the EF and NY linkage groups with the T × E linkage and bin maps identified significant regions of colinearity, the basis for the lack of correspondence for marker position for the physically mapped peach ESTs is not likely due to lack of synteny. Instead, our observations may be due to gene duplication that could have resulted in physical mapping of two paralogs in different positions depending on the mapping strategy that was used. However, only three of the 14 physically mapped peach ESTs that we utilized (PP_LEa0013A14f, PP_LEa0015I11f, and PP_LEa002D10f) have been shown to represent duplicate genes as they mapped to two locations on the peach physical map (Table 3). Nevertheless, our results demonstrate that developing markers for targeted regions in Prunus using ESTs placed on the peach physical map is highly inefficient.

Conclusions

In this study, two new intraspecific sweet cherry molecular linkage maps were constructed by leveraging all the marker resources developed for Prunus species. The transferability of SSR markers developed from other Prunus species to cherry was evaluated and showed a reduced level of amplification from cherry genomic DNA that likely represents the phylogenetic distance between cherry and the other Prunus species from which the SSRs were derived. In addition, linkage map construction was further hampered by the low level of polymorphism within EF and NY that was consistent with results from previous cherry linkage mapping studies. Therefore, additional markers were developed by converting peach ESTs into cherry InDel, dCAP, or CAP markers, plus the addition of four new candidate genes. All these markers, combined with AFLP and SRAP markers, permitted the construction of the sweet cherry linkage maps. However, further map development and identification of markers that are tightly linked to traits of importance would need much higher marker density than can be achieved using the available resources. The high level of colinearity observed between the sweet cherry and other Prunus genetic maps, especially with the genetically mapped EST markers, offers a future opportunity to continue to develop gene-derived markers that can be utilized across Prunus species and ultimately across genera within the Rosaceae.

Acknowledgement

This project was supported by the National Research Initiative Competitive Grant 2005-35300-15454 from the US Department of Agriculture Cooperative State Research, Education, and Extension Service.

Supplementary material

11295_2008_161_MOESM1_ESM.doc (29 kb)
Supplemental Table 1Markers that did not amplify EF or NY DNA arranged by linkage groupa. Markers that identify multiple loci are in bold (DOC 29 kb)
11295_2008_161_MOESM2_ESM.doc (40 kb)
Supplemental Table 2Markers that were monomorphic for EF and NY arranged by linkage group position. Markers that identify multiple loci are identified in bolda (DOC 40 kb)
11295_2008_161_MOESM3_ESM.doc (26 kb)
Supplemental Table 3Markers that were “complex” or “difficult.” Bold identifies those SSRs that identified multiple loci (DOC 25.5 kb)

Copyright information

© Springer-Verlag 2008