Molecular Breeding

, Volume 24, Issue 3, pp 293–303

A genetic linkage map of the cultivated strawberry (Fragaria × ananassa) and its comparison to the diploid Fragaria reference map

  • D. J. Sargent
  • F. Fernandéz-Fernandéz
  • J. J. Ruiz-Roja
  • B. G. Sutherland
  • A. Passey
  • A. B. Whitehouse
  • D. W. Simpson
Article

DOI: 10.1007/s11032-009-9292-9

Cite this article as:
Sargent, D.J., Fernandéz-Fernandéz, F., Ruiz-Roja, J.J. et al. Mol Breeding (2009) 24: 293. doi:10.1007/s11032-009-9292-9

Abstract

The cultivated strawberry, Fragaria × ananassa, is the most economically-important soft-fruit species, but few practical molecular tools for the purpose of marker assisted selection currently exist. As a precursor to the development of such tools, a genetic linkage map was developed from a F1 population comprising 174 seedlings derived from a cross between two F. × ananassa cultivars, ‘Redgauntlet’ × ‘Hapil’. The resultant map is composed of 315 molecular markers—218 microsatellites, 11 gene-specific markers and 86 AFLP and RAPD markers—and spans 3,116 cM. In total, 69 linkage group fragments were recovered, more than the 56 linkage groups expected for the cultivated strawberry, however, all fragments contained a transferable marker that could be associated with one of 56 linkage group scaffolds. The female (Redgauntlet) and male (Hapil) linkage maps are composed, respectively of 170 loci in 32 linkage groups covering 1,675.3 cM and 182 loci in 37 linkage groups covering 1,440.7 cM, with 37 markers common to both maps. The maximum number of markers in one linkage group was 15, the minimum was two. All linkage groups resolved contained at least one transferable marker (SSR or gene-specific) that had been mapped on the diploid Fragaria reference map (FV × FB), and therefore all linkage groups could be identified as homologous to one of the seven diploid Fragaria linkage groups. When marker order was compared to the diploid Fragaria reference map, effectively complete colinearity was observed. However, the occurrence of duplicated loci on homologues of linkage groups FG1 and FG6 provided evidence of a putative chromosomal duplication or translocation event in Fragaria. The development of this linkage map will facilitate the study and dissection of QTL associated with traits of economic importance such as disease resistance and fruit quality, and provides a foundation for the development of markers for the purpose of marker assisted breeding and selection in the cultivated strawberry, F. × ananassa.

Keywords

Linkage mapping Synteny Comparative genomics Polyploidy Rosaceae 

Introduction

The cultivated strawberry, Fragaria × ananassa, is the most economically important soft fruit species and belongs to the Rosaceae, a large and diverse family that also includes Malus (apples), Prunus (the stone fruits and almonds), Rubus (the blackberries and raspberries), Pyrus (pears) and many ornamental species including the genus Rosa (the roses). F. × ananassa is an allo-octoploid species (2n = 8x = 56) derived from artificial hybridisation between two progenitor species F. chiloensis and F. virginiana (Darrow 1966). Breeding for many traits in the cultivated strawberry could be enhanced through the use of marker assisted selection (MAS); however, the octoploid nature of the genome provides a challenge to the development of molecular breeding tools. To circumvent the genetic complexity of the cultivated strawberry, a comprehensive reference map has been developed from an interspecific F2 cross between two closely-related diploid species: F. vesca (2n = 2x = 14) and F. bucharica (2n = 2x = 14; Sargent et al. 2006). The F. bucharica parent was formerly described as F. nubicola but a correction was subsequently published for the erroneous identification of this accession (Staudt 2006). The mapping population has been denoted FV × FN but for clarity will be referred to as FV × FB to acknowledge the change in progenitor species nomenclature. Currently, the FV × FB map spans 568.8 cM over seven linkage groups, and is composed of some 348 published transferable genetic markers, including microsatellite (SSR) and gene-specific markers (173 SSRs, 31 gene-specific and STS markers, 40 RFLPs and 1 SCAR) scored in the full progeny of 76 seedlings (Vilanova et al. 2008). Additionally, 103 markers (99 SSRs and four gene-specific markers) have been assigned to the map using a bin mapping strategy (Sargent et al. 2008).

Until recently, very few linkage maps of the cultivated strawberry had been reported, and those that had been published were either partial maps, reports of markers linked to single traits or were constructed using non-transferrable markers such as AFLPs and RAPDs (Lerceteau-Kohler et al. 2003, 2005; Sugimoto et al. 2005). Due to the type of markers employed in the construction of these maps, they had only limited utility for the purposes of marker assisted breeding before being converted to SCAR markers (Haymes et al. 1997). In addition, the study of Lerceteau-Kohler et al. (2003) suggested that the cultivated strawberry genome might display partial polysomy, which would have further hampered attempts to develop useful molecular markers for breeding through a map-based approach. However, with the advent of the diploid reference map, the cultivated strawberry genome has begun to be further characterised using transferable molecular markers such as SSRs.

Rousseau-Gueutin et al. (2008) reported the development of an F1 linkage map for F. × ananassa, using AFLP and SSR markers, and they were able to define almost all the expected 56 linkage groups associated with an F1 map of the species. The linkage groups they defined were constructed from markers displaying only disomic inheritance, suggesting, as has been reported previously using inheritance patterns of isoenzymes markers (Arulsekar et al. 1981), that the genome of F. × ananassa behaves as a functional diploid. Using SSR markers as anchor loci, Rousseau-Gueutin et al. (2008) were able to compare marker order between the diploid and octoploid maps and, as well as defining many of the homologous linkage groups of their map, they observed almost complete macrosynteny between the genomes of the diploid and octoploid species.

Spigler et al. (2008), studying the genetics of sex determination in the sexually dimorphic octoploid species F. virginiana, reported an F1 molecular linkage map for that species that was composed almost exclusively of SSR markers. Their map covered 42 putative linkage groups, more than the expected 28 for an integrated map of both parental lines. However, a number of the linkage groups they defined were composed of markers segregating in only one parent, and most of the mapped loci they reported were from novel SSRs that had yet to be mapped in the diploid reference map. It is therefore probable that, once further markers with known reference map positions are mapped in their progeny, they will be able to resolve which of the 42 linkage groups are sections of the same Fragaria chromosomes.

In this investigation, we have developed a molecular linkage map for the cultivated strawberry composed predominantly of SSR and gene-specific markers. The progeny used to develop the map is from the cross ‘Redgauntlet’ × ’Hapil’ a heterozygous cross that segregates for a number of economically-important traits, including fruit quality, disease resistance and post harvest characteristics. The majority of the markers used in this investigation were previously assigned to linkage groups on the diploid Fragaria reference map (Sargent et al. 2006, 2007, 2008). As such, we were able to identify the homologous diploid linkage groups of all linkage groups recovered, as well as to identify the homologous groups within the octoploid genome. The map presented will thus serve as a useful tool for comparative genomics within Fragaria and will serve as the basis of future investigations to elucidate QTL associated with a range of agronomically important traits, and ultimately for the development of molecular markers for marker assisted breeding and selection.

Materials and methods

Plant material and DNA extraction

‘Redgauntlet’ × ‘Hapil’ F. × ananassa population

An F1 mapping population was raised, comprising 188 seedlings, from the cross between two cultivated strawberry (F. × ananassa) cultivars ‘Redgauntlet’ × ‘Hapil’. For clarity, the population is referred to as RG × H. Crosses were performed and seedlings were germinated and grown according to the methods described by Sargent et al. (2004). DNA was isolated from young leaf tissue of ‘Redgauntlet’ and ‘Hapil’ and the 188 seedlings of RG × H using the DNeasy plant miniprep kit (Qiagen) according to the manufacturer’s protocol, and diluted 1:50 (~1–10 ng/μl) for use in PCR and AFLP analysis.

FV × FB diploid Fragaria reference population

Twelve microsatellites previously described by Spigler et al. (2008), developed from the octoploid strawberry species F. virginiana and subsequently mapped here in the RG × H progeny, were scored as described below in the six seedlings of the diploid Fragaria reference population (FV × FB) bin set. Map positions were determined following the procedures described in Sargent et al. (2008).

PCR conditions and product visualization

Microsatellite markers

Microsatellite markers that had been previously mapped on the diploid Fragaria reference map (FV × FB), and were distributed evenly throughout the seven Fragaria linkage groups, were selected to be tested for polymorphism in the parents of the RG × H mapping population. In addition, a set of SSR primer pairs that had not previously been mapped in the FV × FB population, but that generated markers that mapped to more than one linkage group on the map of Spigler et al. (2008), were selected to test for polymorphism in the parents of both the FV × FB and RG × H maps. Markers were initially amplified individually from the parents of the mapping population following the touchdown protocol described by Sargent et al. (2003) using primers labelled on the forward primer with one of four fluorescent dyes: 6-FAM, NED, VIC or PET (Applied Biosystems). PCR products were then diluted 1:10 and fractionated by capillary electrophoresis through a 3100 genetic analyser (Applied Biosystems). The data generated were collected and analysed using the GENESCAN and GENOTYPER (Applied Biosystems) software.

Polymorphic markers were combined by product size and fluorescent dye colour into multiplexes of up to eight primer pairs and PCR was performed using the ‘Type-it’ PCR mastermix (Qiagen) following the manufacturer’s recommendations, except that PCRs were performed in a final volume of 12.5 μl. Reactions were performed using the following PCR cycles: an initial denaturation step of 95°C for 5 min was followed by 28 cycles of 95°C for 30 s, an annealing temperature of 55°C decreasing by 0.5°C per cycle until 50°C for 90 s and 72°C for 30 s, followed by a 30 min final extension step at 60°C. Samples were then diluted 1:10 and analysed by capillary electrophoresis as described above.

AFLP markers

Template DNA of the parents and seedlings was prepared for generation of AFLP fragments using AFLP analysis system II (Invitrogen) following the procedure described by Sutherland et al. (2008). Forward primers E37, E40, E41 and E48, and those with the two base extensions E-AA, E-AC and E-TC were labelled using either 6-FAM or NED fluorescent dyes (Applied Biosystems) and used in combination with reverse primers M-CAA, M-CAC, M-CAG, M-CTA, and M-CTC. In total, ten different primer combinations were found to generate scoreable polymorphic bands and were used to genotype the RG × H progeny. Markers were named according to the primer combination used and the size of the allele generated, i.e. E37MCAG121 was generated with primers E-37 and M-CAG and the allele generated was 112 nt in length.

RAPD markers

The RAPD primers used in this investigation were selected from the Operon RAPD series A-L (Operon Biotechnologies GmbH, Koln, Germany). PCR was performed following the methods described in Sutherland et al. (2008). The segregating RAPD markers were scored by eye and scores were verified by at least two researchers. Segregating amplicons were named after the RAPD primer with which they were generated, followed by a letter signifying the amplicon scored, A being the smallest fragment and successively larger fragments coded B, C, D etc.

Marker nomenclature

The prefix a or b in the names of scored markers indicates whether the marker segregated in the female ‘Redgauntlet’ (a) or male ‘Hapil’ (b) parent. Those markers without a prefix segregated in both parents.

Data analysis and map construction

All single dose polymorphisms generated by PCR with segregation patterns approximating to either 1:1 or 3:1 Mendelian ratios were initially analysed for co-segregation with other fragments generated with the same primer pair to identify allelic pairs. This stage also served to identify any alleles that had been scored erroneously, and to identify any contaminants in the population of 188 seedlings. Markers segregating in each parent were then analysed separately using JOINMAP 4.0 (Kyasma, NL) applying the Kosambi mapping function. Marker placement was determined using a minimum LOD score threshold of 3.0, a recombination fraction threshold of 0.35, ripple value of 1.0, jump threshold of 3.0 and a triplet threshold of 5.0. The maps presented were constructed using MAPCHART 2.2 for Windows (Voorrips 2002).

Results

Microsatellite mapping in FV × FB

In total, seven of the 12 SSRs previously mapped in the F. virginiana population (hereby denoted Y3 × PA) by Spigler et al. (2008) and scored in the RG × H progeny produced length polymorphisms between the parents of the FV × FB mapping population. These were scored and mapped in the bin set described for that population (Sargent et al. 2008) and all seven markers mapped to previously described bins. The seedling genotype scores and the bins to which those scores correspond are given in Table 1.
Table 1

Genotype scores for the seven SSRs previously reported by Spigler et al. (2008) that were scored in the FV × FB bin set of Sargent et al. (2008), detailing the marker name and bin name to which each marker was located, along with the mapping interval of each bin and the number of markers each contains

Marker

32

46

48

59

62

83

Bin name

Interval (cM)

Number of markers in bin

CO380002

b

h

h

b

b

a

I:41

0.7

7

CO661626b

b

h

b

a

h

h

VII:21

20.7

25

CO817431

h

a

h

b

a

h

II:56

0

4

CO817853

b

b

h

b

b

a

I:44

1.3

3

CO818147

b

h

b

a

h

h

VII:21

20.7

25

CX661492

b

h

h

b

b

a

I:41

0.7

7

CX661792

a

b

b

h

h

h

V:46

0

5

Single dose marker amplification and polymorphism in RG × H

In total, 71 SSR and five gene-specific marker primer pairs, as well as 10 AFLP primer combinations and 11 RAPDs, produced 457 polymorphic amplicons in the parents of the RG × H progeny that could be scored in the population. The SSR markers generated 319 segregating amplicons, the gene-specific markers 18, the AFLPs 79 and the RAPDs 41 segregating amplicons. In total, 105 amplicons segregated 3:1 in both parents (an × an), 165 segregated 1:1 in ‘Redgauntlet’ (an × nn) and 187 segregated 1:1 in ‘Hapil’ (nn × an).

Cosegregation analysis and mapping in RG × H

After initial co-segregation analyses of segregating amplicons generated from the same primer pairs, 14 rogues were identified in the population of 188 individuals and removed from subsequent analyses. The final cosegregation analyses were therefore performed on data from 174 seedlings. The initial cosegregation analyses identified 74 loci composed of more than one allele. Fifty-three of these loci segregated in one parent approximating to the ratio 1:1, whilst the remaining 21 segregated in both parents approximating to either a 1:2:1 or 1:1:1:1 ratio.

After a subsequent co-segregation analysis, 170 loci coalesced into 32 linkage groups covering 1,675.3 cM of the ‘Redgauntlet’ genome whilst 182 loci assimilated into 37 linkage groups covering 1,440.7 cM of the ‘Hapil’ genome. Thirty-seven markers were common to both maps. The maximum number of markers in one linkage group was 15, the minimum was just two. Despite both maps covering more than the expected 28 linkage groups, all linkage groups that were resolved contained at least one transferable marker (SSR or gene-specific) that had been mapped on the diploid Fragaria reference map (FV × FB), either previously, or in this investigation by bin mapping, and thus all linkage groups recovered were identified as homologous to one of the seven diploid Fragaria linkage groups and assigned to one of 56 linkage group scaffolds. We estimate that the map covers between 74 and 78% of the octoploid genome, assuming an average linkage group length of between 71 and 75 cM.

The genetic map comprises in total, 315 loci (over two parental maps, 37 loci shared between the two maps) and covers 3,116 cM over 69 linkage group fragments. Figure 1 show the six linkage groups recovered containing markers mapped to FG1 of the FV × FB map, and their positions on the four homeologous ‘Hapil’ linkage groups in relation to the reference map. Supplemental Fig. 1 shows the seven linkage groups of FV × FB and the 69 linkage groups resolved in this investigation assigned names according to the FV × FB linkage group nomenclature. The ‘Redgauntlet’ and ‘Hapil’ homeologues are presented in each figure with their corresponding FV × FB homologue. Supplemental Table 1 lists the 317 markers mapped in the RG × H population (some loci contained allelic pairs that could not be combined as a single locus prior to linkage analysis) along with their monogenic marker segregation ratios, their associated chi-squared values, and the FV × FB linkage group homologue to which the marker mapped. Of the 315 loci that mapped to the 69 linkage groups of the RG × H map, 89 (28.1%) displayed segregation distortion; 27 (8.5%) at P < 0.05, 4 (1.3%) at P < 0.01 and 58 (18.3%) at P ≤ 0.001. Markers with distorted segregations were evenly distributed throughout the linkage groups recovered and no single linkage group contained a significantly higher number of loci displaying distorted ratios than any other.
Fig. 1

A comparison of linkage group FG1 of the diploid Fragaria reference map FV × FB with the six linkage groups identified as homologues of FG1 on the male ‘Hapil’ map of the RG × H mapping population (HA1–HA4) showing complete macrosynteny between diploid and octoploid genomes. Microsatellite markers mapped in the full FV × FB progeny are given in bold, those mapped using the FV × FB bin set are given in red, and markers in blue (in the online version of the figure) are SSRs that did not segregate in the FV × FB progeny. AFLP and RAPD markers are shown in grey. Common markers are joined by red lines between the linkage groups. Mapping distances are given in cM

Gene-specific markers

Five gene-specific primer pairs produced segregating markers in the RG × H progeny. ACO mapped to one locus that segregated in both parents of the progeny and mapped to an FG6 homologue; ACP mapped to one locus on an FG6 homologue of the ‘Hapil’ map; DFR mapped to one locus on an FG2 homologue of the ‘Redgauntlet’ map; EXP mapped to one locus on an FG3 homologue of the ‘Hapil’ map, one locus on an FG4 homologue of the ‘Redgauntlet’ map, two loci on FG5 homologues, one on each of the ‘Redgauntlet’ and ‘Hapil’ maps and two loci on FG7 homologues of the ‘Hapil’ map; LOX mapped to two loci, one on each of the ‘Redgauntlet’ and ‘Hapil’ maps. In total therefore, 11 gene-specific loci were mapped using primer pairs for the five genes.

Duplicated loci

A small number of markers that were generated with the same primer pairs and that segregated in the RG × H progeny, mapped to loci on more than one linkage group of the RG × H map, indicating duplicated loci in the octoploid genome. Marker CFVCT032 produced loci that mapped to FG4 and FG7 homologues on the RG × H map, whilst it mapped to FG3 on the reference map. EMFn181 produced loci that mapped to FG3, FG4, FG5 and FG6 homologues on the RG × H map, and to FG5 on the reference map. UDF065 produced a single segregating locus in RG × H that mapped to a homologue of FG6, but this marker was located to a mapping bin on FG3 by Sargent et al. (2008). However, despite only a single locus being mapped on the FV × FB map, for markers EMFn181 and UDF065, their primer pairs generated multiple amplicons in the parents of the reference map, indicating duplicated loci in diploid Fragaria. EXP produced loci that mapped to FG3, FG4, FG5 and FG7 on the RG × H map, but this marker produced multiple loci that mapped to linkage groups FG3, FG4 and FG7 homologues of the linkage map of Nier et al. (2006), as well as other non-segregating amplicons.

Figure 2 shows a comparison between markers mapped to FG1 and FG6 of the diploid Fragaria reference map and their map positions on linkage groups of the RG × H map, the Y3 × PA map of Spigler et al. (2008), and the T × E Prunus reference map. Markers positioned on PG6 of the Prunus reference map were located to FG1 and FG6 of the diploid Fragaria reference map (Vilanova et al. 2008). Microsatellite loci that mapped to FG1 and FG6 of the diploid reference map were located on the same linkage group (LG13 and LG27) in the investigation of Spigler et al. (2008) and in this investigation (HA1-B and RG1-B), indicating a putative chromosomal duplication event in Fragaria of an ancestral chromosome that resembled PG6 in Prunus but which has since been lost in the diploids F. bucharica and F. vesca.
Fig. 2

A comparison between markers mapped to FG1 and FG6 of the diploid Fragaria reference map FV × FB and their map positions on linkage groups of the RG × H map, the map of Spigler et al. (2008), and PG6 of the T × E Prunus reference map. Markers positioned on PG6 of the Prunus reference map were located to FG1 and FG6 of the diploid Fragaria reference map by Vilanova et al. (2008). Loci associated with FG1 are given in blue, whilst those associated with FG6 are given in red (in the online version of the figure)

Marker order and comparison with the diploid Fragaria reference map

Figure 1 shows FG1 of the FV × FB diploid Fragaria reference map and the six homeologous linkage group fragments from the ‘Hapil’ linkage map. The red lines between linkage groups connect homeologous loci on the ‘Hapil’ map to their homologues on the FV × FB map. In general, the marker order of the linkage groups on the RG × H corresponded exactly to that of the FV × FB reference map. With the exception of the identification of a putative ancient chromosomal duplication event leading to the formation of FG1 and FG6, no evidence of any major chromosomal rearrangements or translocations between the diploid and octoploid genomes could be found between the diploid and octoploid maps.

On a few of linkage groups on the RG × H map, there were a number of minor discrepancies in marker order in relation to the FV × FB reference map. These were the order of markers BFACT002, EMFn214 and CFVCT015 on RG2-B, BFACT002 and CFVCT015 on HA2-A, the position of CFVCT032 on RG3-A, the order of CFVCT032 and BFACT045 on HA3-A, the position of EMFvi166 on HA3-A, the position of EMFn111 on HA4-A, the order of Fvi20 and EMFv104 on RG6-B, and the position of Fvi20 on HA6-B (Supplemental Fig. 1). With the exception of these small discrepancies, the RG × H map showed essentially complete colinearity with the diploid reference map, and in general, map distances between markers on the two maps, and between homologous groups within the RG × H map were comparable.

Discussion

We have produced a genetic linkage map of an F1F. × ananassa population consisting of 174 seedlings, derived from the cross ‘Redgauntlet’×’Hapil’ (RG × H). The two parental maps are composed of a total of 69 linkage groups, made up of 315 loci (37 of which are shared between the two maps) covering a total genetic distance of 3,116 cM. As the majority of markers used to create the map are transferable SSR or gene-specific STS markers, all 69 linkage groups have been identified according to their homologous groups on the diploid Fragaria reference map (Sargent et al. 2006, 2008; Vilanova et al. 2008) and the homeologous groups within the RG × H map have been identified through the mapping of marker loci derived from the same primer pair to different linkage groups on each of the parental maps. Thus all the markers could be assigned to one of 56 linkage group scaffolds on the RG × H map. Despite essentially complete colinearity between the diploid and octoploid maps, the duplication of markers on the homologues of linkage groups FG1 and FG6, which were not apparent in the diploid reference map, suggests a putative ancient chromosomal duplication event in Fragaria.

Comparison with other octoploid maps

The linkage map of Rousseau-Gueutin et al. (2008), derived from the F1 cross ‘Capitola’×’CF1116’ (CA × CF) covers a genetic distance of 5,017 cM, which is longer than the total map distance of 3,116 cM presented here for the RG × H map. However, they mapped a total of 162 transferable loci across the male and female maps, 23 of which were shared between the two maps and used for map integration. In contrast, the map presented here contains 230 transferable markers, 28 of which are shared between the two parental maps. Thus, whilst the CA × CF map covers a greater genetic distance, the RG × H map presented here is composed of a greater number of potentially transferable genetic loci.

The Y3 × PA map presented by Spigler et al. (2008) contains a total of 212 mapped transferable marker loci, 39 of which segregate in both the male and female maps and were used for map integration. Their integrated map covers a total genetic distance of 2,373 cM, which, considering that some of their linkage groups contain only markers segregating in one parent, is comparable to the RG × H map presented here. Many of the primer pairs used to generate markers by Spigler et al. (2008) produced alleles that mapped to more than one locus on the same linkage group, and, as suggested by those authors, the separation of alleles derived from the same primer pair on a linkage group may have been due to changes within and between genomes of the highly heterozygous polyploid genome of F. virginiana and may thus represent the same genetic locus. In contrast to that study, and in agreement with the study of Rousseau-Gueutin et al. (2008), we found no evidence of such differentiation within the genomes of the F. × ananassa cultivars used to create the RG × H mapping population, suggesting that the genome of F. × ananassa is much less differentiated than that of F. virginiana. The 212 transferable marker loci mapped by Spigler et al. (2008) is comparable to the 230 transferable loci mapped in this investigation. However, if marker loci from the study of Spigler et al. (2008) that were produced from the same primer pair and mapped to the same linkage group are taken to be alleles from the same locus segregating in a highly differentiated genetic background, then 161 discrete loci were mapped in their investigation. Thus, the RG × H linkage map presented in this investigation represents the most comprehensive genetic linkage map of transferable markers presented to date for an octoploid Fragaria species.

Extending the RG × H map using marker data from other maps

Many of the markers mapped by Spigler et al. (2008) were also mapped in this investigation. As the Y3 × PA map was composed of very few markers that had been previously mapped in Fragaria, we have been able to identify the homologues of many of their linkage groups on the diploid reference map by cross-referencing common markers mapped in RG × H and thus define them in relation to the seven linkage groups of the FV × FB map. Identifying homologous linkage groups between the Y3 × PA map and FV × FB will enable us to select further markers from the study of Spigler et al. (2008) to enrich the RG × H map in areas not currently covered by SSR markers.

Evidence of an ancient chromosomal duplication event in Rosaceae

One of the loci duplicated on the RG × H map, Fvi6b, that mapped to the proximal end of linkage group FG6 of the diploid Fragaria reference map, was also duplicated on the Y3 × PA map of Spigler et al. (2008) on the linkage group they denoted LG13. On both maps, the locus was duplicated close to marker ARSFL-010, which maps to the end of FV × FB linkage group FG1 and its octoploid homologues. Additionally, in the study of Spigler et al. (2008), marker ARSFL013, which mapped to FG1 on the diploid reference map, was linked to marker ARSFL022, which mapped to the distal end of FG6 of the diploid reference map. Vilanova et al. (2008) compared the genomes of diploid Fragaria and diploid Prunus using a map based approach and found a high degree of synteny of genome structure between the two genera. They showed that Prunus linkage group PG6 contained two markers (EPpCU1830 and Pgl1) that mapped to linkage group FG6 of the diploid reference map, and two RFLP markers (PC21 and AC8) that mapped to FG1 of the Fragaria reference map. In other species such as Arabidopsis, comparisons with distantly-related plant species have identified duplication events that were undetectable in the host species (Vandepoele et al. 2002). Comparisons between Prunus and the diploid and octoploid Fragaria have revealed the occurrence of linkage groups in the octoploid maps sharing markers from two linkage groups of the diploid Fragaria reference map and thus more closely resembling a Prunus linkage group than the linkage groups of diploid Fragaria. This indicates the occurrence of a putative chromosomal duplication event in Fragaria that led to the evolution of FG1 and FG6 and their homologues in the F. × ananassa genome.

It has been shown that, following duplication within a genome, large-scale losses of up to 90% of the redundant gene copies can occur (Seoighe and Wolf 1998; Seoighe and Gehring 2004, Wang et al. 2005). Duplicated loci on homologues of linkage groups FG1 and FG6 have been preserved in the octoploid genome, but are not evident in the genomes of the diploid Fragaria species (F. bucharica and F. vesca) that were used to create the reference mapping population. The discovery of this putative duplication event raises questions about the relative age of the polyploid Fragaria and which species contributed to the allo-octoploid genome of the original octoploid Fragaria. However, it is important to note that such hypotheses are highly speculative as they are currently based on molecular evidence from only a small number of loci in two mapping populations. Indeed, the evidence presented here could be the result of a translocation event leading to a partial chromosomal duplication that occurred after polyploidisation in the ancestors of the cultivated strawberry, or in one of the progenitors of the octoploid Fragaria, such as F. iinumae (Rousseau-Gueutin et al. 2009), which has not been extensively investigated at the molecular level but that is clearly differentiated from the other diploid Fragaria both morphologically (Sargent et al. 2004), genetically (Rousseau-Gueutin et al. 2009) and cytologically (Iwatsubo and Naruhashi 1989).

It will be useful to screen additional markers from linkage groups FG1 and FG6 in both octoploid and diploid Fragaria mapping populations to elucidate the extent to which markers have been duplicated in the genus and to assess the relative age of the duplication or translocation event in the history of the genus Fragaria.

Polysomy and gene-specific markers

It has been reported previously that allelic segregation patterns of markers scored in the octoploid Fragaria species display disomic Mendelian inheritance (Ashley et al. 2003; Rousseau-Gueutin et al. 2008; Spigler et al. 2008). In this investigation, we found no evidence of polysomic segregation, and thus our findings support previous studies that suggest that the genomes of octoploid Fragaria are functionally diploidised. The molecular evidence for diploidisation of the F. × ananassa genome is further supported by the phenotypic segregation data of Ahmadi et al. (1990) and Sugimoto et al. (2005), who showed that the everbearing (perpetual flowering) trait in the cultivated strawberry is controlled by a single dominant gene (Ev), which segregates according to expected Mendelian ratios in progenies raised from parents heterozygous for the trait.

Comparison with the FV × FB reference map

The map presented in this investigation represents the most comprehensive comparative data set available to date for the determination of synteny between the diploid and octoploid Fragaria. As in the previous investigation of Rousseau-Gueutin et al. (2008), we found almost complete conservation of marker order between the RG × H and FV × FB maps (Supplemental Fig. 1). Except for the apparent duplication of locus Fvi6b on homologues of FG1 and FG6, which may indicate an ancient chromosomal duplication or translocation event in Fragaria, we found no evidence of any chromosomal rearrangements between the diploid and octoploid maps. Indeed, only a few minor differences in marker order were observed between the maps, and these were represented by conflicts in the order of single loci, which are more likely due to conflicts between marker data for tightly-linked loci in both the RG × H and FV × FB maps, rather than physical differences in chromosome structure. Thus, the RG × H map displays extremely high levels of colinearity and essentially complete macrosynteny with the diploid Fragaria reference map.

Concluding remarks

The use of the diploid Fragaria reference map has permitted an informed choice to be made about the selection of markers used in the construction of the RG × H maps. It enabled us to select markers distributed throughout the seven linkage groups of the diploid Fragaria reference map and thus the RG × H map provides good coverage of the F. × ananassa genome. The development of this genetic linkage map for the RG × H mapping progeny, composed predominantly of transferable SSR and gene-specific markers, and precise allele sizing using an automated sequencing platform, will facilitate the future study and dissection of QTL associated with disease resistance and with fruit quality traits in this and other cultivated strawberry mapping populations. Thus, the maps presented here will provide a foundation for the future identification of markers for the purpose of marker assisted breeding and selection in the cultivated strawberry.

Acknowledgments

Molecular genetics in Fragaria at East Malling Research is funded by the BBSRC.

Supplementary material

11032_2009_9292_MOESM1_ESM.pdf (1.4 mb)
Supplemental Fig. 1. A comparison of linkage groups associated with the 56 homeologous linkage groups of the RG × H linkage map and homologous linkage groups on the diploid Fragaria reference map FV × FB. Microsatellite markers mapped in the full FV × FB progeny are given in bold, those mapped using the FV × FB bin set are given in red, and markers in blue are SSRs that did not segregate in the FV × FB progeny. AFLP and RAPD markers are shown in grey. Mapping distances are given in centi-Morgans (cM) (PDF 1401 kb)
11032_2009_9292_MOESM2_ESM.doc (427 kb)
Supplementary material 2 (DOC 427 kb)

Copyright information

© Springer Science+Business Media B.V. 2009

Authors and Affiliations

  • D. J. Sargent
    • 1
  • F. Fernandéz-Fernandéz
    • 1
  • J. J. Ruiz-Roja
    • 2
  • B. G. Sutherland
    • 1
  • A. Passey
    • 1
  • A. B. Whitehouse
    • 1
  • D. W. Simpson
    • 1
  1. 1.East Malling Research (EMR)East Malling, KentUK
  2. 2.Department of HorticultureVirginia Polytechnic Institute and State UniversityBlacksburgUSA

Personalised recommendations