• chromosome evolution;
  • domestication;
  • Cucumis sativus;
  • genetic/physical mapping;
  • NGS de novo assembly


  1. Top of page
  2. Summary
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental Procedures
  7. Acknowledgments
  8. References
  9. Supporting Information

Cucumber, Cucumis sativus L. is the only taxon with 2n = 2= 14 chromosomes in the genus Cucumis. It consists of two cross-compatible botanical varieties: the cultivated C. sativus var. sativus and the wild C. sativus var. hardwickii. There is no consensus on the evolutionary relationship between the two taxa. Whole-genome sequencing of the cucumber genome provides a new opportunity to advance our understanding of chromosome evolution and the domestication history of cucumber. In this study, a high-density genetic map for cultivated cucumber was developed that contained 735 marker loci in seven linkage groups spanning 707.8 cM. Integration of genetic and physical maps resulted in a chromosome-level draft genome assembly comprising 193 Mbp, or 53% of the 367 Mbp cucumber genome. Strategically selected markers from the genetic map and draft genome assembly were employed to screen for fosmid clones for use as probes in comparative fluorescence in situ hybridization analysis of pachytene chromosomes to investigate genetic differentiation between wild and cultivated cucumbers. Significant differences in the amount and distribution of heterochromatins, as well as chromosomal rearrangements, were uncovered between the two taxa. In particular, six inversions, five paracentric and one pericentric, were revealed in chromosomes 4, 5 and 7. Comparison of the order of fosmid loci on chromosome 7 of cultivated and wild cucumbers, and the syntenic melon chromosome I suggested that the paracentric inversion in this chromosome occurred during domestication of cucumber. The results support the sub-species status of these two cucumber taxa, and suggest that C. sativus var. hardwickii is the progenitor of cultivated cucumber.


  1. Top of page
  2. Summary
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental Procedures
  7. Acknowledgments
  8. References
  9. Supporting Information

Cucumber, Cucumis sativus L. (2n = 2x = 14) is an economically important crop and an ideal system for studying several important biological processes such as sex determination (Tanurdzic and Banks, 2004; Wang et al., 2010), gene transfer among nuclear and organellar genomes (Havey, 1997; Alverson et al., 2011), and phloem physiology (Lough and Lucas, 2006; Zhang et al., 2010a). Cucumber is native to southern Asia (Candolle, 1959; Sebastian et al., 2010), and has been cultivated in India for at least 3000 years (Whitaker and Davis, 1962). Cucumber spread eastward to China, which is considered the secondary diversity center (Sebastian et al., 2010), approximately 2000 years ago (Keng, 1974), and westward to Europe around 700–1500 years ago (Paris et al., 2011).

The primary gene pool of cucumber consists of two cross-compatible botanical varieties: the cultivated C. sativus var. sativus L. (hereafter CSS) and the wild C. sativus var. hardwickii Alef. (hereafter CSH). There is disagreement whether CSH is a feral form or progenitor of CSS. CSH is widely distributed throughout India, southwest China, Myanmar, Nepal and Thailand (Whitaker and Davis, 1962; Bisht et al., 2004; de Wilde and Duyfjes, 2010). Morphological differences between the two botanical varieties are obscured by the large genetic variation in cultivated cucumber (Whitaker and Davis, 1962; Kirkbride, 1993; de Wilde and Duyfjes, 2010). However, there are clear differences at the DNA level: for example, RAPD (randomly amplified polymorphic DNA) or isozyme markers (Meglic et al., 1996; Staub et al., 1997, 1999; Mliki et al., 2003) clearly separate CSS from CSH. Genetic mapping using recombinant inbred lines derived from a cross between CSS inbred line Gy14 and CSH accession PI 183967 (Ren et al., 2009) revealed clustering of markers in chromosomes 4, 5 and 7, possibly due to structure rearrangements between CSS and CSH, and fluorescence in situ hybridization (FISH) revealed an inversion in the short arm of chromosome 5 (Ren et al., 2009). Although these results clearly indicate a very different genetic structure between CSS and CSH, a genome-wide, systematic evaluation of genetic differentiation between the two taxa is lacking.

In the genus Cucumis, CSS and CSH are the only two taxa with 2n = 2= 14 chromosomes; all other species have either 2n = 2= 24 chromosomes [for example, melon (C. melo)] or multiples of = 12 (Kirkbride, 1993). Current evidence suggests that the seven cucumber chromosomes evolved from 12 chromosomes of a species that shared a common ancestor with melon (Huang et al., 2009; Li et al., 2011a), but the details are not known. Elucidation of chromosomal differentiation between CSH and CSS may help understand cucumber domestication and chromosome evolution in Cucumis. This knowledge will also be helpful for efficient use of genetic resources in wild cucumber or its close relatives for cucumber improvement.

The cultivated cucumber has a very narrow genetic base (Knerr et al., 1989; Kennard et al., 1994; Dijkhuizen et al., 1996), making it difficult to develop high-density genetic maps. However, this has changed due to application of next-generation sequencing (NGS) technologies. Scaffold assemblies of three cucumber lines (9930, Gy14 and B10) are now available (Huang et al., 2009; Cavagnaro et al., 2010; Woycicki et al., 2011). Thousands of simple sequence repeat (SSR) markers have been developed from these genome sequences (Ren et al., 2009; Cavagnaro et al., 2010), greatly facilitating genetic mapping, molecular tagging and gene cloning in cucumber (Ren et al., 2009; Weng et al., 2010; Zhang et al., 2010b, 2011; Li et al., 2011b; Miao et al., 2011).

Each of the three released NGS cucumber genome assemblies comprises approximately 4000 de novo assembled scaffolds accounting for approximately 55–65% of the 367 Mbp cucumber genome, but the quality of these assemblies has not been addressed. Due to limitations of NGS technologies in genome assembly (Alkan et al., 2011), scaffold mis-assemblies are evident (e.g. Li et al., 2011b; this study). Chromosome-level assemblies are either not available or the quality of the assembly has not been validated. A dense, high-quality genetic map is necessary to verify the accuracy of the assembled draft genome.

FISH is a powerful cytological tool for integration of genetic and physical maps (Jiang and Gill, 2006). Due to its high resolution, FISH analysis of meiotic pachytene chromosomes is particularly useful in estimating the amount and distribution of heterochromatin, integrating genetic and phyical maps, and detecting chromsome rearrangemetns and syntenic blocks, as well as filling the gaps in whole-genome assemblies (Cheng et al., 2001; Iovene et al., 2008, 2011; Szinay et al., 2008; Gao et al., 2009). In cucumber, FISH has been used successfully to karyotype its relatively small chromosomes, in comparative mapping with melon, and for development of physical maps (Koo et al., 2005, 2010; Han et al., 2008, 2009). In this study, integrated genetic, cytogenetic and NGS approaches were employed, with the aim of elucidating the domestication history of cucumber. We first developed a high-density linkage map for cultivated cucumber. Draft genome scaffolds were anchored to this genetic map to develop a chromosome-level cucumber draft genome assembly. Information from the genetic map and genome assembly was used to identify fosmid clones for comparative pachytene FISH to reveal chromosome differentiation between CSS and CSH cucumbers.


  1. Top of page
  2. Summary
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental Procedures
  7. Acknowledgments
  8. References
  9. Supporting Information

Development of a high-density genetic map for cultivated cucumber

Nearly 5000 markers were tested for polymorphisms between Gy14 and 9930. Among 1266 SSRs derived from the 9930 genome sequences, 835 had in silico PCR products from both Gy14 and 9930 draft genome templates, and 731 (57.7% of 1266) were polymorphic between Gy14 and 9930. However, when the 1266 SSRs were experimentally tested using 9% polyacrylamide gel electrophoresis (PAGE), only 312 (25%) were polymorphic. Therefore, all subsequent polymorphism screening was performed using PAGE, and 490 polymorphic markers (21%) were identified from additional 2340 cucumber SSRs, and 60 (5%) from 1178 melon SSRs. In total, 862 polymorphic markers were identified, of which 767 were applied to the mapping population. Due to incomplete data or poor data quality, 32 markers were eliminated from linkage analysis. Thus, 735 markers (704 from cucumber and 31 from melon) were finally mapped. The names, sources and types of these 735 mapped markers are presented in Table S1.

At an LOD threshold of 8.0, the 735 markers were grouped into seven linkage groups spanning 707.8 cM, with a mean distance of 0.96 cM between adjacent markers. The numbers of markers mapped in chromosomes 1–7 were 151, 70, 157, 104, 75, 112 and 66, respectively. Statistical data for this high-density genetic map are presented in Table 1, and a graphical presentation is shown in Figure S1.

Table 1.   Summary of the high-density genetic map of cultivated cucumber anchored with Gy14 and 9930 draft genome scaffolds
ChromosomeNumber of loci mappedMap length (cM)Marker intervalGy149930Gy14 assembly (version 1.0) (Mb)
Number of scaffolds anchoredPhysical length (Mb)Number of scaffolds anchoredPhysical length (Mb)
1151103.10.685528.1 4329.128.4
2 70101.51.452421.32823.223.5
7 6667.41.02 2816.61918.919.3

Integration of genetic and physical maps and assembly of the Gy14 draft genome

In silico PCR and BLASTn sequence alignment assigned a majority of the 735 mapped markers to the Gy14 and 9930 draft genome scaffolds. For many large scaffolds, more than one marker was assigned to the same scaffold. For example, 25, 22 and 10 mapped markers were from the three longest scaffolds of the Gy14 genome assembly: scaffold02229 (6 896 804 bp), scaffold03356 (5 534 371 bp) and scaffold03611 (3 882 770 bp), respectively (Table S1). The order of markers on the genetic map and in the scaffolds provided an indication of the quality of both the genetic map and the assembly of the genome scaffolds. With few exceptions (see below for examples), markers from the same scaffold were placed in the same linkage block (Table S1). Although inconsistencies were found at fine scale where the order of markers in a linkage block was not collinear with their positions in the scaffold, this generally occurred within 5 cM genetic distance and the markers were associated with relatively short scaffolds, which may be due to either mapping errors or mis-assembly of relevant scaffolds.

In six cases, molecular markers from the same scaffold were mapped in different chromosomal locations, including one from the Gy14 draft assembly (scaffold02047) and five from the 9930 draft assembly (scaffold000022, scaffold000031, scaffold000036, scaffold000052, and scaffold000081). For example, among the eight SSRs derived from Gy14 scaffold02047 (total length 2614 kb), five (located between 1406 and 2281 kb) were mapped on chromosome 3 and three (located between 435 and 857 kb) on chromosome 7. In the 9930 draft genome scaffold000052, four SSRs mapped on chromosomes 4 and 6, respectively (two each) (Table S1). Their positions were verified by FISH. Clearly, these scaffolds were mis-assembled.

In total, 244 Gy14 draft genome scaffolds (173.1 Mbp; 86% of the 203.5 Mbp assembled sequences) and 237 9930 draft genome scaffolds (193.3 Mbp, 79% of the 243.5 Mbp assembled sequences) were anchored onto the genetic map. Among the 100 largest scaffolds of the Gy14 assembly, all but one (scaffold02140, 1 096 332 bp in length) were anchored to the genetic map; among the 100 largest scaffolds of the 9930 assembly, three (scaffold000072 with 1 001 594 bp, scaffold000083 with 660 293 bp and scaffold000146 with 550 576 bp) were not anchored. We designed over 60 SSR primer pairs from these four scaffolds, but none was polymorphic between Gy14 and 9930. Additional markers may be required to anchor these scaffolds to the genetic map.

Based on the integrated genetic/physical map, a Gy14 draft genome assembly (version 1.0) with seven pseudochromosomes was developed. This assembly had 192.6 Mbp sequences accounting for 53% of the 367 Mbp cucumber genome, of which 19.5 Mbp were taken from the 9930 draft genome to bridge the gaps. Statistical data for this assembly are summarized in Table 1, and the assembly is publicly accessible and searchable at The resulting integrated genetic/physical map is shown schematically in Figure 1.


Figure 1.  An integrated genetic/physical map of cultivated cucumber.For each chromosome, the light blue vertical bar (left) represents a linkage map with selected mapped loci and their map locations (in cM); the purple bar (right) indicates physical map of the pseudo-chromosome of the Gy14 draft genome assembly (version 1.0). Numbers to the right of each physical map indicate the cumulative physical length (Mbp) of the chromosome. Light color horizontal bars between chromosome blocks on the physical map indicate gaps in the assembly. Lines connecting the genetic and physical maps indicate colinearity of makers on the genetic and physical maps.

Download figure to PowerPoint

This integrated genetic/physical map allowed a global view of the relationship of genetic and physical distances along each cucumber chromosome. No suppression of genetic recombination was observed on this map. Comparison of the genetic and physical distances of markers present in the same anchored scaffold made it possible to compare the recombination frequency in different chromosome regions. A plot depicting such relationships is presented in Figure S2. The mean recombination frequency was 3.5 cM/Mb, or 286 kb/cM. However, it was clear that this relationship was highly variable in different chromosomal regions. For instance, the highest recombination frequency was 12.0 cM/Mb (83.3 kb/cM) at the telomeric end of chromosome 2, and the lowest was 0.6 cM/Mb (1666.7 kb/cM) at the end of chromosome 4.

Cytological assessment of physical coverage of the cucumber genetic map

BLAST alignment of the 1046 fosmid clone end sequences against the Gy14 and 9930 draft genomes identified 56 fosmid clones carrying putatively single or low-copy sequences in the cucumber genome. Among them, 13 showed repetitive FISH signals in mitotic metaphase chromosomes and were discarded; the remaining 43 with unique hybridization signals were used in subsequent studies. Fosmid clones were also selected through PCR-based library screening. Among the 185 SSR markers selected from the genetic map for PCR screening, 89 detected positive clones, and 33 were used in this study. Information on all 76 fosmid clones from both sources is given in Table S2.

The somatic chromosomes of cucumber are short (approximately 2–4 μm long) and morphologically similar, and thus are difficult to differentiate from each other (Figure 2a). For accurate chromosome identification, a FISH-based karyotype for mitotic metaphase chromosomes was established using four DNA repeats (Figure 2b). After simultaneous application of the four repeated sequence probes (Figure 2c), all seven chromosomes showed distinctive labeling patterns, allowing easy recognition of individual chromosomes (Figure 2d).


Figure 2.  Cytological validation of physical coverage of the high-density genetic map of cultivated cucumber. Cucumber mitotic metaphase chromosomes were small and similar in morphology (a), but could be differentiated by unique FISH patterns of four sub-telomeric and pericentromeric repetitive DNA probes (b), allowing establishment of a FISH-based karyotype (d). The arrows in (b) indicate two homologous chromosome 3 detected by pericentromeric probe CENT3. The FISH signal patterns in seven chromosomes (C1–C7) from simultaneous probing with 28 fosmid clones (c) suggested complete physical coverage of the genetic map (e). Scale bar = 5 μm.

Download figure to PowerPoint

Four fosmid clones were identified from each chromosome to determine physical ends and the putative centromere location of the chromosome. The FISH signals of 28 probes and corresponding map locations are shown in Figure 2(e) and Figure S3. Corresponding markers at these mapped loci are listed in Table S2. In Figure 2(e), some chromosomes show only a single signal instead of two signals, probably due to the limitations of somatic FISH (for example, reduced hybridization efficiency due to inadequate and non-homogeneous enzyme digestion). Nevertheless, these results demonstrated that genetically linked molecular markers were indeed co-localized physically, and that the physical order of the anchored fosmids along all chromosomes correlates well with their positions on the genetic map. In particular, all the fosmid clones genetically anchored to the distal ends of seven linkage groups were indeed physically located at the telomeric ends of each chromosome, suggesting that our high-density genetic map covered almost the entire physical length of cucumber chromosomes.

Chromosome rearrangements between cultivated and wild cucumbers

Seventy-six genetically anchored fosmid clones were used as probes (Table S2) in pachytene FISH to examine genome-wide differentiation between cultivated (CSS) and wild (CSH) cucumber chromosomes. The physical order of adjacent fosmid clones in each chromosome was determined by two-color FISH. During this process, to avoid mis-identification, the same pachytene chromosome preparation was used up to eight times for repeated probing (Cheng et al., 2001). A multi-fosmid cocktail with probes from the same chromosome was then hybridized to CSS and CSH pachytene chromosomes. Representative pachytene FISH images are shown in Figure S4. Comparative FISH patterns for all seven chromosomes of the two botanical varieties are shown in Figure 3.


Figure 3.  Major chromosomal rearrangements between cultivated cucumber C. sativus var. sativus (C1–C7) and wild cucumber C. sativus var. hardwickii (H1–H7) revealed by comparative pachytene FISH. Numbers to the left of each cultivated cucumber chromosome (C1–C7) indicate fosmid clones. Dashed lines connect FISH signals detected by the same probe between each pair of cultivated and wild cucumber chromosomes. The centromere locations in the pachytene chromosomes are indicated by diamonds. The approximate region involved in each inversion is indicated by a vertical dashed line along the chromosome. The name of a particular inversion is given along the dashed line. The assignment and orientation of each chromosome are based on molecular markers mapped in each chromosome (Ren et al., 2009).

Download figure to PowerPoint

In the pachytene chromosomes, the centromeres may be recognized as FISH signals from the type III repeat probe. The heterochromatin blocks are visualized as light, DAPI-stained regions (DAPI binds preferentially to AT-rich regions). Heterochromatin distribution patterns between CSS and CSH chromosomes were quite different. In CSS (chromosomes C1–C7), telomeric heterochromatic knobs were detected in all chromosome arms except the long arm of chromosome 6 (C6L), but were detected only in H2S, H4S, H5L and H7S in the wild cucumber. In addition, the large heterochromatic knobs in H1L and H4L were absent in the cultivated cucumber, whereas the large heterochromatin block in C4L was absent from H4L of wild cucumber (Figure 3).

All fosmid probes from CSS generated strong FISH signals in CSH. FISH signal patterns suggested inversions in chromosomes 4, 5 and 7 between CSS and CSH (Figure 3). In chromosome 4, two paracentric inversions, one in the short arm (Inv4.1, defined by fosmid clones 41.1, 41.2 and 41.3) and one in the long arm (Inv4.2, defined by fosmid clones 41.4, 41.5 and 41.6), were evident. In chromosome 5, there were two paracentric inversions in the short arm (Inv5.1, defined by fosmid clones 51.1, 51.2 and 51.3, and Inv5.2, defined by fosmid clones 51.4 and 51.5) and one pericentric inversion (Inv5.3, from the centromere to fosmid clone 51.7). The large paracentric inversion in the short arm of chromosome 7, Inv7.1, occurred between fosmid loci 71.1 and 71.4 (Figure 3).

The approximate locations of the six inversions on the cucumber genetic map are shown in both Figure S1 and Table S1. The genetic distances covered by the five paracentric inversions on the genetic map were 29.5, 52.2, 19.4, 23.6 and 23.7 cM for Inv4.1, Inv4.2, Inv5.1, Inv5.2 and Inv7.1, respectively (mean 29.7 cM). To validate and understand the scope of these inversions, pachytene FISH analysis was conducted in a PI 249561 (CSS) × PI 183967 (CSH) F1 hybrid. The result for chromosomes C7 and H7 is shown in Figure 4, in which the inverted fosmid clone signals between homologous chromosome arms C7S and H7S were confirmed (Figure 4a). In some pollen mother cells, an inversion loop between C7 and H7 was observed due to the heterozygous nature of the inversion region in the CSS × CSH F1 hybrid (Figure 4b). To further validate this inversion loop, the pachytene chromosome spread of Figure 4(b) was re-probed with type I and type IV satellite repeat probes. While the type IV repeat was present in sub-telomeric regions of CSS chromosome C7 (Figure 2d), it showed no FISH signal in CSH chromosome H7 (Figure S5). The signal patterns of type I and IV repeats in CSS × CSH F1 provided additional evidence for the inversion loop between C7 and H7 (Figure 4c). Similar features were also found for the paracentric inversions in chromosomes 4 and 5, suggesting that the chromosomal regions involved in these inversions were probably large.


Figure 4.  Cytological evidence and origin of the paracentric inversion in the short arm of chromosome 7 during domestication of cucumber. C7 and H7 are chromosome 7 of cultivated and wild cucumbers, respectively. In the F1 hybrid between cultivated (PI 249561) × wild (PI 183967) cucumbers, the FISH signals of two fosmid probes (red 71.1 and green 71.4) in C7 and H7 (a) are inverted. In some cells, an inversion loop involved in the two chromosome arms was observed as evidenced from the signal patterns of fosmid probes 71.1 (red) and 71.4 (green) (b). Re-probing of the pachytene chromosomes in (b) using sub-telomeric type I (red) and type IV (green) repeats provided further evidence of the presence of an inversion loop (c). H7 is syntenic and collinear with melon chromosome I (MI) at the fosmid loci (d), suggesting that the paracentric inversion between C7 and H7 occurred during cucumber domestication.

Download figure to PowerPoint

Previous studies revealed high-level synteny along cultivated cucumber chromosome 7 (C7) and melon chromosome I (MI) (Li et al., 2011a). To examine whether the paracentric inversion (Inv7.1) between C7S and H7S occurred only in the cucumber lineage, comparative FISH analysis was performed in CSS accession PI 249561, CSH accession PI 183967 and the melon inbred line ‘Top Mark’. Nine fosmid probes from C7 (Table S2) were employed for pachytene FISH. Representative FISH images of C7, H7 and MI pachytene chromosome are shown in Figure S6, and the comparative FISH result is shown in Figure 4(d). All nine fosmid probes detected strong signals in both C7 and H7. All except probe 71.6 detected signals in melon chromosome MI. Although the paracentric inversion between C7S and H7S was evident, the mapped fosmid loci in H7 and MI were completely collinear, suggesting that this paracentric inversion, Inv7.1, may have occurred during domestication of cultivated cucumber.

In a previous study, we established syntenic relationships between cucumber and melon chromosomes through comparative mapping (Li et al., 2011a). Syntenic blocks of CSS chromosomes 4, 5 and 7 with the melon genome in terms of the Gy14 draft genome assembly (this study) and melon consensus genetic map locations (Li et al., 2011a) were compared and are shown in Table S3. The colinearity of fosmid loci in the Inv7.1 region shown in Figure 4(d) was consistent with the comparative mapping data. Similarly, marker orders in the inversion regions of Inv4.1 and Inv5.1 were also collinear between the Gy14 draft assembly and the melon consensus genetic map. Clearly, the three inversions, Inv4.1, Inv5.1 and Inv7.1, were all specific to the cucumber lineage. Further work is needed to ascertain the origins of the other three inversions (Inv4.2, Inv5.2 and Inv5.3) because the involved region was either relatively small (Inv5.2) or spanned two melon chromosomes (Inv4.2 and Inv 5.3) (Table S3).


  1. Top of page
  2. Summary
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental Procedures
  7. Acknowledgments
  8. References
  9. Supporting Information

The genetic map of cultivated cucumber and Gy14 assembly

In this study, we developed a cucumber genetic map with 735 loci markers in seven linkage groups spanning 707.8 cM. Analysis of chromosome pairing in meiosis of pollen mother cells of cucumber suggested a mean of 14.8 (Ramachandran and Seshadri, 1986) to 17.8 (Fanourakis, 1984) chiasmata per meiocyte. As one chiasma corresponds to 50 map units of the genome (e.g. King et al., 2002), this implies that the total map length for cucumber is between 750 and 900 cM. Given that the early estimates were from male meiosis, and crossover rates in female meiosis may be lower (Giraut et al., 2011), the actual expected total map length of the cucumber genome may be in the lower limit of the estimated 750–900 cM range. Indeed, two previous recombinant inbred line-based intra-varietal cucumber maps by Miao et al. (2011) and Zhang et al. (2011) covered 711.9 and 749.2 cM, respectively. The high-density linkage map built herein had a length of 707.8 cM, and appears to cover the whole physical length of the cucumber genome (Figure 3). The slightly lower-than-expected map length of the present map may be due to the relatively small size of the mapping population (92 F2 plants), and thus fewer recombination events.

Draft genome sequences have been generated in a number of species by de novo assembly of sequences obtained with NGS technologies. However, the large amount of repetitive DNA sequences in most plant genomes pose significant technical challenges in both sequencing quality and assembly accuracy (Schatz et al., 2010; Treangen and Salzberg, 2011; Ye et al., 2011). Mis-assemblies of NGS sequences are probably common (Meader et al., 2010; Alkan et al., 2011). In this study, we identified six mis-assembled scaffolds, including one from the Gy14 draft genome and five from the 9930 draft genome, three of which were validated with FISH. Clearly more mis-assembled scaffolds may exist in the draft genomes. These observations point out the importance of a high-quality genetic map of densely spaced markers for verifying the accuracy of whole-genome assemblies and affirming the correct placement of scaffolds in each chromosome.

The relationships of cultivated and wild cucumbers

Possible structural rearrangements between the cultivated CSS and wild CSH cucumbers were inferred from several earlier mapping studies (Ren et al., 2009; Weng et al., 2010; Miao et al., 2011). The present study provided a global picture on chromosomal differentiation between the two botanical varieties, including the amount and distribution of heterochromatin and chromosome rearrangements (Figure 3).

Ten clusters of marker loci (> 10 loci per cluster) in five chromosomes (3, 4, 5, 6 and 7) were observed on the inter-sub-species cucumber genetic map by Ren et al. (2009). The six inversions identified herein probably correspond to the largest clusters located in chromosomes 4 (Inv4.1 and Inv4.2), 5 (Inv5.1, Inv 5.2 and Inv5.3) and 7 (Inv7.1). Except for pericentric inversion Inv5.3, all other inversions covered relatively large genetic regions (mean 29.7 cM; Table S1), as evidenced by inversion loops present in pollen mother cells of the PI 249561 (CSS) × PI 183967 (CSH) F1 hybrid (Figure 4b,c). No evidence of large inversions was found in chromosomes 3 and 6 between PI 183967 and PI 249561. However, it is possible that smaller inversions or other structural changes beyond the resolving power of pachytene FISH exist between chromosomes of the two taxa, as has been shown in other species such as tomato (Anderson et al., 2010).

In plants, large inversions (> 10 cM genetic distance) between different populations (genotypes, ecotypes, races or sub-species) within a species or between very closely related species have been reported for some Liliaceae species (Garrido- Ramos et al., 1998), wheat (Triticum spp.) (Badaeva et al., 2007), rice (Oryza sativa) (Jiang et al., 2007), the yellow monkeyflower (Mimulus guttatus) (Lowry and Willis, 2010) and soybean (Glycine spp.) (Singh and Hymowitz, 1988; Findley et al., 2010). Lineage- or population-specific inversions are believed to play important roles in evolution (for example, local adaptation and speciation) (Hoffmann and Rieseberg, 2008; Kirkpatrick, 2010; Lowry and Willis, 2010). Comparison of the order of marker loci on genetic maps developed within cultivated cucumber populations (Weng et al., 2010; Miao et al., 2011; Zhang et al., 2011; this study) suggested that cultivated cucumbers share the six inversions relative to CSH accession PI 183967. Using pachytene FISH, we examined the FISH signal patterns of fosmid loci in the F1 hybrid between two CSH accessions (PI 183967 and PI 215589) and found that both CSH accessions shared these inversions (Figure S7). According to the US Department of Agriculture National Plant Germplasm System (, these two wild cucumber accessions were collected from two geographically different areas of India. Therefore, these six inversions are probably common between cultivated and wild cucumbers.

The cultivated (CSS) and wild (CSH) cucumber are the only two taxa in Cucumis with 2n = 2x = 14 chromosomes (Kirkbride, 1993), and their relationship has long been a contentious issue. Naudin (1859) postulated that the origin of the cultivated cucumber, C. sativus, is C. hardwickii, which was described as a species by Royle (1835). Duthie (1903) regarded C. hardwickii as the wild form of the cultivated cucumber, as both have all essential characteristics in common, and this was echoed by Whitaker and Davis (1962), who suggested C. hardwickii as a feral form of C. sativus, rather than the putative ancestor [see de Wilde and Duyfjes (2010) for more discussions]. Kirkbride (1993) and Jeffrey (2001) noted that typical members of CSH are easily identified but that no morphological characteristics clearly separate it from CSS. In both greenhouses and their natural habitats, the wild cucumber hybridizes readily with the cultivated one, producing fertile F1 hybrids (Deakin et al., 1971; Bisht et al., 2004). The discovery of feral forms of cultivated cucumber (de Wilde and Duyfjes, 2010) further obscures the morphological differences between the two botanical varieties. For these reasons, de Wilde and Duyfjes (2010) suggested lowering the rank of the wild form to forma, C.s. forma hardwickii, at the same time relegating all other materials to C.s. forma sativus.

Plant domestication is an evolutionary process operating under the influence of human activities (Harlan, 1992). Over time, artificial selection causes cultivated populations to diverge morphologically and genetically from their wild progenitors (Clement, 1999; Pickersgill, 2007). The wild progenitors are commonly sympatric with their domestic forms. They usually differ strikingly in phenotype and adaptation, but remain sufficiently related genetically to produce fertile hybrids (Zeven and de Wet, 1982). Hybridization is common and genes are exchanged. Hybrids between cultivated and wild races, and derivatives of such introgression, survive in ‘intermediate’ habitats as weeds, and provide a bridge between them for occasional gene exchange (Zeven and de Wet, 1982). Thus, plant domestication is a dynamic evolutionary process that produces a continuum of plant populations, ranging from exploited wild plants, via incipient domesticates, to cultivated populations that cannot survive without human intervention (Clement, 1999; Pickersgill, 2007). These characteristics of progenitor-crop features are clearly applicable to the relationships between CSH and CSS cucumbers. As the only two taxa with 2n = 14 chromosomes in Cucumis, previous studies and data presented herein strongly support that CSH is a progenitor from which CSS was domesticated, and the two taxa are differentiated enough to be treated as two separate sub-species of Cucumis sativus. Lines of evidence are summarized below.

Traits with domestication features

Domestication leads to morphological and physiological changes, collectively called the ‘domestication syndrome’, that distinguishdomesticated taxa from their wild ancestors (Harlan, 1992; Hancock, 2005). The CSH accession PI 183967 has typical and characteristic phenotypes of the wild cucumber (Kirkbride, 1993; Jeffrey, 2001). Under greenhouse conditions in Madison, Wisconsin, PI 183967 usually requires two weeks of short-day treatment for flowering, and the seeds require time to break dormancy and germinate. PI 183967 flowers late and has a sequential fruit setting habit, producing many small, round, bitter fruits that fall easily from the vine when mature. These traits are in sharp contrast with cultivated cucumbers, which are usually day-length neutral and show no seed dormancy. Cultivated cucumbers generally set relatively few large fruits per plant with fewer but larger seeds, which may well be the result of selection during domestication.

Genetic diversity  Most crops contain less genetic variation than their wild ancestors, which is probably a product of a small initial population, coupled with intense selection for desirable traits (Tanksley and McCouch, 1997). These domestication bottleneck effects may have been intensified in cucumber and other cucurbit crops because cucurbit plants usually grow large and require a lot of space, and a few plants are likely to satisfy the requirements of any one grower. Cucurbits may consequently have existed in small colonies both in nature and under cultivation, and this restriction in population size increased inbreeding despite monoecious flowering that favors out-crossing (Allard, 1960). Indeed, cultivated cucumber has a much narrower genetic base than its wild relative CSH, with polymorphism levels ranging from 3–12% for isozyme, RFLP and RAPD markers (Knerr et al., 1989; Kennard et al., 1994; Dijkhuizen et al., 1996) or 10-20% for microsatellite markers (Li et al., 2011b; this study). CSH populations generally have much higher genetic diversity than CSS ones (nearly twice) (Ren et al., 2009; estimated from data of Meglic et al., 1996; Bisht et al., 2004).

Genetic differentiation between cultivated and wild cucumbers

Although the level of genetic diversity in the two gene pools is clearly different, significant differentiations between CSS and CSH at the chromosome level were revealed in the present study (Figure 3) that further supported the sub-species status of the two taxa in C. sativus. Nevertheless, additional work is needed to examine chromosome polymorphisms in wild cucumber populations from different regions, and their distribution in the wild.

Gene flow

Segregating populations of natural crosses between cultivated and wild cucumbers were observed in all areas where the wild cucumber was collected, indicating the existence of free gene flow between CSS and CSH (Bisht et al., 2004). However, fruits on hybrid derived plants from natural crosses collected from the wild showed a high proportion of non-viable seeds (Bisht et al., 2004), which is probably due at least in part to the reproductive barrier caused by chromosome rearrangements between the two taxa. The relatively larger and oblong fruits from the hybrid derivative plants are used by local people to feed animals (Bisht et al., 2004). It is clear that dynamic gene flow exists in nature, mingled with human intervention, a characteristic process of domestication in action, which may explain the continuum of morphological traits in both taxa observed in nature. However, this should not be viewed as evidence against separation of the two taxa; rather, it provides additional evidence to support CSH as a progenitor of CSS.

Experimental Procedures

  1. Top of page
  2. Summary
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental Procedures
  7. Acknowledgments
  8. References
  9. Supporting Information

Plant materials

An F2 mapping population was developed for linkage analysis and map construction from a cross between two cultivated (CSS) cucumber inbred lines Gy14 and 9930 whose genomes have been sequenced. A single F1 plant from Gy14 × 9930 was self-pollinated to produce 92 F2 progenies used for genetic mapping. The CSS accession PI 249561, wild cucumber (CSH) accessions PI 183967 and PI 215589, and a melon inbred line ‘Top Mark’ were used in cytological analyses. The F1 hybrids of PI 249561 ×  PI 183967 and PI 183967 ×  PI 215589 were used to validate paracentric inversions.

Molecular markers for linkage mapping

Nearly 5000 molecular markers were screened for polymorphisms between Gy14 and 9930 from three major sources: 1266 SSRs (prefixed SSR in marker names) originally developed from the 9930 draft genome (Ren et al., 2009), 2014 SSRs (prefixed UW in marker names) developed from the Gy14 scaffold assmblies (Cavagnaro et al., 2010), and 1178 melon SSRs previously mapped in melon (Danin-Poleg et al., 2000; Gonzalo et al., 2005). Additional markers included 211 SSRs or SCARs (sequence-characterized amplified regions) from cucumber (Fazio et al., 2003; Robbins et al., 2008).

For all markers, in silico PCR was implemented using the Gy14 and 9930 genome scaffold assemblies as templates to assign markers to scaffolds, reveal polymorphisms between the two parental genomes, and estimate copy numbers of expected PCR products. This was performed with a custom Perl script that used the NCBI BLASTN program as a search engine (Cavagnaro et al., 2010).

Molecular marker analysis and map construction

Unexpanded young leaves from plants were collected into 2.0 mL microcentrifuge tubes, lyophilized in a freeze dryer, and ground into fine powder. Genomic DNA was extracted using the CTAB method (Murry and Thompson, 1980) and purified with phenol/chloroform.

Each PCR contained 25 ng template DNA, 0.5 μM of each primer, 0.2 mM dNTP mix, 0.5 unit of Taq DNA polymerase and 1 × PCR buffer (Fermentas, Glen Burnie, MD) in a total volume of 10 μL. Touchdown PCR was employed for all primer sets (Weng et al., 2005). PCR products were size-fractionated in 9% polyacrylamide gels and visualized with silver staining, and images were captured with a digital camera.

For each marker, a χ2 test for goodness-of-fit was performed against the expected 1:2:1 segregation ratio. Linkage analysis was carried out using JoinMap 3.0 software (van Ooijen and Voorrips, 2001). Linkage groups were determined with a minimum LOD score of 8.0 and a recombination fraction of 0.3. Genetic distance was calculated with Kosambi mapping function.

Genetic/physical map integration and development of Gy14 draft genome assembly

For most large scaffolds, multiple markers were mapped that helped to orientate scaffolds on the genetic map. Cross-validation was also carried out between the Gy14 and 9930 draft genome assemblies. When conflicts arose with regard to marker orders between the genetic and physical maps, which occurred more often in markers within a short genetic distance (< 5 cM), the marker order in the scaffold was assumed to be correct. In some cases, it was possible to bridge gaps between adjacent scaffolds in the Gy14 draft genome using the 9930 draft genome scaffolds. Thus, a chromosomal-level assembly of the Gy14 draft genome was performed according to their order and orientation on the integrated genetic/physical map.

Cytological investigation with fluorescence in situ hybridization (FISH)

A fosmid library from the cucumber cultivar ‘Straight 8’ was developed previously (Meyer et al., 2008), which contained 99 840 clones with a mean insert size of 38 kb, corresponding to 4.3-fold coverage of the cucumber genome. Two approaches were used to identify fosmid clones for FISH analysis. First, 1046 fosmid end sequences were downloaded from GenBank (accession numbers ET024028ET025073), and BLASTed against the Gy14 and 9930 draft genome assemblies. Clones with end sequences that had fewer than four BLAST hits in the draft genome sequences were selected as candidate FISH probes. The putative map locations of these selected fosmid clones were inferred from the scaffolds with which they were associated.

In the second method, 15 360 fosmid clones from 40 384-well (24 × 16) plates were randomly chosen to construct two fosmid pools: the column pool and the super-pool. The column pool was made by mixing fosmid clones from two columns of one 384-well plate into a single well of a 96-well plate (12 × 8), resulting in five 96-well column pools (32 clones in each column pool). Next, one 96-well plate of super-pools was constructed by pooling all column clones from the five column-pool plates into a single column of a 96-well plate. Thus, there were 96 super-pools (160 clones each). Growth of Escherichia coli cultures and extraction of fosmid DNA was performed as described by Meyer et al. (2008). PCR-based screening of the fosmid library with selected SSR primer pairs was performed using three rounds of PCR amplifications to identify positive column pools, plates with positive clones and individual clones, respectively.

For identification of the seven cucumber chromosomes, a FISH-based karyotype was established using four repetitive DNA probes, including the sub-telomeric type IV repeat (Ganal and Hemleben, 1988; Koo et al., 2005; Han et al., 2008), the pericentromeric type III repeat (Ganal et al., 1986; Han et al., 2008) and the BAC-E38 repeat (Koo et al., 2010), as well as the chromosome 3-specific pericentromeric CENT3 repeat (this study). The CENT3 probe (fosmid clone 3M24) was identified during library screening with microsatellite marker UW085172. Although UW085172 was mapped on chromosome 6 (Table S1), fosmid probe 3M24 hybridized strongly to the pericentromeric region of chromosome 3 (Figure 2b,d). Further fiber-FISH analysis indicated that this clone contained tandemly repeated DNA spanning approximately 150–200 kb in chromosome 3, and was named CENT3.

To estimate the physical coverage of the genetic map, four cytological landmarks, one from each telomeric end, and two flanking the putative centromere location, were established for each chromosome by fosmid library screening with molecular markers mapped on the genetic map. Briefly, two fosmid probes were first hybridized to determine their relative positions in the chromosome. Then, for each chromosome, a four-fosmid probe cocktail along with the pericentromeric type III repeat (Figure S3) was used in FISH to validate their expected locations. Finally, a cocktail containing all 28 probes (four per chromosome) was hybridized simultaneously to the somatic metaphase chromosomes (Figure 2c). The pre-defined two color (red and green) FISH labeling patterns of the four probes allowed correct identification of individual chromosomes (Figure 2e).

To examine and validate chromosome rearrangements between cultivated and wild cucumbers, 7 to17 fosmid probes from each chromosome, as well as the type I and IV repeats (Ganal et al., 1986; Ganal and Hemleben, 1988) were used in comparative FISH mapping of meiotic pachytene chromosomes prepared from pollen mother cells of CSS inbred line PI 249561, CSH accession PI 183967 and their F1 hybrid.

The FISH procedure was performed as described by Koo et al. (2010). Biotin- and digoxigenin-labeled probes were detected with Alexa Fluor 488 streptavidin antibody (Invitrogen, Carlsbad, CA) and rhodamine-conjugated anti-digoxigenin antibody (Roche Diagnostics USA, Indianapolis, IN), respectively. Chromosomes were counterstained by 4′,6-diamidino-2-phenylindole (DAPI) in ‘Vector Shield’ antifade solution (Vector Laboratories, Burlingame, CA). FISH signals were captured using a CCD camera. The images were processed using Meta Imaging Series 7.5 software (Molecular Devices, Downingtown, PA, USA). Chromosome straightening was performed using the ‘straighten-curved-objects’ plug-in of ImageJ (Kocsis et al., 1991). Chromosomes were ordered as described by Ren et al. (2009).


  1. Top of page
  2. Summary
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental Procedures
  7. Acknowledgments
  8. References
  9. Supporting Information

The authors thank Linda Crubaugh for technical help and Doug Senalik for help with bioinformatic analysis of data. We also thank Dr David Spooner for critical reading of the manuscript, and three anonymous reviewers for valuable suggestions to improve the previous version of the manuscript. This research was supported by a grant from the United States Department of Agriculture - Specialty Crop Research Initiative (SCRI) (project number 2011-51181-30661), and research grant number IS-4341-10 from BARD (The United States–Israel Bi-national Agricultural Research and Development Fund) to Y.W. Work by F.L and Y.L. was partly supported by the ‘948 Project’ of the Agriculture Ministry of China (2011-S17) and the National Science Foundation of China (NSFC) grant (project number 31171955), respectively.


  1. Top of page
  2. Summary
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental Procedures
  7. Acknowledgments
  8. References
  9. Supporting Information
  • Alkan, C., Sajjadian, S. and Eichler, E.E. (2011) Limitations of next-generation genome sequence assembly. Nat. Methods, 8, 6165.
  • Allard, R.W. (1960) Principles of Plant Breeding. New York: John Wiley & Sons Inc.
  • Alverson, A.J., Rice, D.W., Dickinson, S., Barry, K. and Palmer, J.D. (2011) Origins and recombination of the bacterial-sized multichromosomal mitochondrial genome of cucumber. Plant Cell, 23, 24992513.
  • Anderson, L.K., Covey, P.A., Larsen, L.R., Bedinger, P. and Stack, S.M. (2010) Structural differences in chromosomes distinguish species in the tomato clade. Cytogenet Genome Res. 129, 2434.
  • Badaeva, E.D., Dedkova, O.S., Gay, G., Pukhalskyi, V.A., Zelenin, A.V., Bernard, S. and Bernard, M. (2007) Chromosomal rearrangements in wheat: their types and distribution. Genome, 50, 907926.
  • Bisht, I.S., Bhat, K.V., Tanwar, S.P.S., Bhandari, D.C., Joshi, K. and Sharma, A.K. (2004) Distribution and genetic diversity of Cucumis sativus var. hardwickii (Royle) Alef in India. J. Hortic. Sci. Biotechnol. 79, 783791.
  • Candolle, A.D. (1959) Origin of Cultivated Plants. New York: Hafner Publishing Co.
  • Cavagnaro, P.F., Senalik, D.A., Yang, L., Simon, P.W., Harkins, T.T., Kodira, C.D., Huang, S. and Weng, Y. (2010) Genome-wide characterization of simple sequence repeats in cucumber (Cucumis sativus L.). BMC Genomics, 11, 569.
  • Cheng, Z., Presting, G.G., Buell, C.R., Wing, R.A. and Jiang, J. (2001) High-resolution pachytene chromosome mapping of bacterial artificial chromosomes anchored by genetic markers reveals the centromere location and the distribution of genetic recombination along chromosome 10 of rice. Genetics, 157, 17491757.
  • Clement, C.R. (1999) 1492 and the loss of Amazonian crop genetic resources. I. The relation between domestication and human population decline. Econ. Bot. 53, 188202.
  • Danin-Poleg, Y., Reis, N., Baudracco-Arnas, S., Pitrat, M., Staub, J.E., Oliver, M., Arus, P., deVicente, C.M. and Katzir, N. (2000) Simple sequence repeats in Cucumis mapping and map merging. Genome, 43, 96374.
  • Deakin, J.R., Bohn, G.W. and Whitaker, T.W. (1971) Interspecific hybridization in Cucumis. Econ. Bot. 25, 195201.
  • Dijkhuizen, A., Kennard, W.C., Havey, M.J. and Staub, J.E. (1996) RFLP variation and genetic relationships in cultivated cucumber. Euphytica, 90, 7987.
  • Duthie, J.F. (1903) Flora of the Upper Gangetic Plain, and of the Adjacent Siwalik and Sub-Himalayan Tracts. Calcutta, India: Superintendent of Government Printing Publication.
  • Fanourakis, N.E. (1984) Inheritance and linkage studies of the fruit epidermis structure, and investigation of linkage relationships of several traits and of meiosis in cucumber. PhD Thesis, University of Wisconsin, Madison, WI.
  • Fazio, G., Staub, J.E. and Stevens, M.R. (2003) Genetic mapping and QTL analysis of horticultural traits in cucumber (Cucumis sativus L.) using recombinant inbred lines. Theor. Appl. Genet. 107, 864874.
  • Findley, S.D., Cannon, S., Varala, K., Du, J.C., Ma, J.X., Hudson, M.E., Birchler, J.A. and Stacey, G. (2010) A fluorescence in situ hybridization system for karyotyping soybean. Genetics, 185, 727744.
  • Ganal, M. and Hemleben, V. (1988) Insertion and amplification of a DNA sequence in satellite DNA of Cucumis sativus L. (cucumber). Theor. Appl. Genet. 75, 357361.
  • Ganal, M., Riede, I. and Hemleben, V. (1986) Organization and sequence analysis of two related satellite DNAs in cucumber (Cucumis sativus L.). J. Mol. Evol. 23, 2330.
  • Gao, D.Y., Gill, N., Kim, H.R. et al. (2009) A lineage-specific centromere retrotransposon in Oryza brachyantha. Plant J. 60, 820831.
  • Garrido- Ramos, M.A., Jamilena, M., De la Herran, R., Rejon, R., Camacho, J.M. and Rejon, M.R. (1998) Inheritance and fitness effects of a pericentric inversion and a supernumerary chromosome segment in Muscari comosum (Liliaceae). Heredity, 80, 724731.
  • Giraut, L., Falque, M., Drouaud, J., Pereira, L., Martin, O.C. and Mezard, C. (2011) Genome-wide crossover distribution in Arabidopsis thaliana meiosis reveals sex-specific patterns along chromosomes. PLoS Genet. 7, e1002354.
  • Gonzalo, M.J., Oliver, M., Garcia-Mas, J., Monfort, A., Dolcet-Sanjuan, R., Katzir, N., Arus, P. and Monforte, A.J. (2005) Simple-sequence repeat markers used in merging linkage maps of melon (Cucumis melo L.). Theor. Appl. Genet. 110, 802811.
  • Han, Y.H., Zhang, Z.H., Liu, J.H., Lu, J.Y., Huang, S.W. and Jin, W.W. (2008) Distribution of the tandem repeat sequences and karyotyping in cucumber (Cucumis sativus L.) by fluorescence in situ hybridization. Cytogenet. Genome Res., 122, 8088.
  • Han, Y., Zhang, Z., Liu, C., Liu, J., Huang, S., Jiang, J. and Jin, W. (2009) Centromere repositioning in cucurbit species: implication of the genomic impact from centromere activation and inactivation. Proc. Natl Acad. Sci. USA, 106, 1493741.
  • Hancock, J.F. (2005) Contributions of domesticated plant studies to our understanding of plant evolution. Ann. Bot. 96, 953963.
  • Harlan, J.R. (1992) Crops & Man, 2nd edn. Madison, WI: American Society of Agronomy/Crop Science Society of America.
  • Havey, M.J. (1997) Predominant paternal transmission of the mitochondrial genome in cucumber. J. Hered. 88, 232235.
  • Hoffmann, A.A. and Rieseberg, L.H. (2008) Revisiting the impact of inversions in evolution: from population genetic markers to drivers of adaptive shifts and speciation? Annu. Rev. Ecol. Evol. Syst., 39, 2142.
  • Huang, S., Li, R., Zhang, Z. et al. (2009) The genome of the cucumber, Cucumis sativus L. Nat. Genet. 41, 12751281.
  • Iovene, M., Wielgus, S.M., Simon, P.W., Buell, C.R. and Jiang, J.M. (2008) Chromatin structure and physical mapping of chromosome 6 of potato and comparative analyses with tomato. Genetics, 180, 13071317.
  • Iovene, M., Cavagnaro, P.F., Senalik, D., Buell, C.R., Jiang, J.M. and Simon, P.W. (2011) Comparative FISH mapping of Daucus species (Apiaceae family). Chromosome Res. 19, 493506.
  • Jeffrey, C. (2001) Cucurbitaceae. In Mansfeld’s Encyclopedia of Agricultural and Horticultural Crops (Hanelt, P., ed). Berlin: Springer-Verlag, pp. 15101557.
  • Jiang, J. and Gill, B.S. (2006) Current status and the future of fluorescence in situ hybridization (FISH) in plant genome research. genome, 49, 10571068.
  • Jiang, L., Zhang, W.L., Xia, Z.H., Jiang, G.H., Qian, Q., Li, A.L., Cheng, Z.K., Zhu, L.H., Mao, L. and Zhai, W.X. (2007) A paracentric inversion suppresses genetic recombination at the FON3 locus with breakpoints corresponding to sequence gaps on rice chromosome 11L. Mol. Genet. Genomics, 277, 263272.
  • Keng, H. (1974) Economic plants of ancient North-China as mentioned in Shih-Ching (Book of Poetry). Econ. Bot. 28, 391410.
  • Kennard, W.C., Poetter, K., Dijkhuizen, A., Meglic, V., Staub, J.E. and Havey, M.J. (1994) Linkages among RFLP, RAPD, isozyme, disease resistance, and morphological markers in narrow and wide crosses of cucumber. Theor. Appl. Genet. 89, 4248.
  • King, J., Roberts, L.A., Kearsey, M.J., Thomas, H.M., Jones, R.N., Huang, L., Armstead, I.P., Morgan, W.G. and King, I.P. (2002) A demonstration of a 1 to 1 correspondence between chiasma frequency and recombination using a Lolium perenne/Festuca pratensis substitution. Genetics, 161, 307314.
  • Kirkbride, J.H. (1993) Biosystematic Monograph of the Genus Cucumis (Cucurbitaceae): Botanical Identification of Cucumbers and Melons. Boone, NC: Parkway Publishers.
  • Kirkpatrick, M. (2010) How and why chromosome inversions evolve. PLoS Biol. 8, e1000501.
  • Knerr, L.D., Staub, J.E., Holder, D.J. and May, B.P. (1989) Genetic diversity in Cucumis sativus L. assessed by variation at 18 allozyme coding loci. Theor. Appl. Genet. 78, 119128.
  • Kocsis, E., Trus, B.L., Steer, C.J., Bisher, M.E. and Steven, A.C. (1991) Image averaging of flexible fibrous macromolecules: the clathrin triskelion has an elastic proximal segment. J. Struct. Biol. 107, 614.
  • Koo, D.H., Choi, H.W., Cho, J., Hur, Y. and Bang, J.W. (2005) A high-resolution karyotype of cucumber (Cucumis sativus L ‘Winter Long’) revealed by C-banding, pachytene analysis, and RAPD-aided fluorescence in situ hybridization. Genome, 48, 534540.
  • Koo, D.H., Nam, Y.W., Choi, D., Bang, J.W., de Jong, H. and Hur, Y. (2010) Molecular cytogenetic mapping of Cucumis sativus and C. melo using highly repetitive DNA sequences. Chromosome Res. 18, 325336.
  • Li, D., Cuevas, H.E., Yang, L. et al. (2011a) Syntenic relationships between cucumber (Cucumis sativus L.) and melon (C. melo L.) chromosomes as revealed by comparative genetic mapping. BMC Genomics, 12, 396.
  • Li, Y., Yang, L., Pathak, M., Li, D., He, X. and Weng, Y. (2011b) Fine genetic mapping of cp, a recessive gene for compact (dwarf) plant architecture in cucumber, Cucumis sativus L. Theor. Appl. Genet. 123, 973983.
  • Lough, T.J. and Lucas, W.J. (2006) Integrative plant biology: role of phloem long-distance macromolecular trafficking. Annu. Rev. Plant Biol. 57, 203232.
  • Lowry, D.B. and Willis, J.H. (2010) A widespread chromosomal inversion polymorphism contributes to a major life-history transition, local adaptation, and reproductive isolation. PLoS Biol. 8, e1000500.
  • Meader, S., Hillier, L.W., Locke, D., Ponting, C.P. and Lunter, G. (2010) Genome assembly quality, assessment and improvement using the neutral indel model. Genome Res., 20, 675684.
  • Meglic, V., Serquen, F. and Staub, J.E. (1996) Genetic diversity in cucumber. 1. A reevaluation of the US germplasm collection. Genet. Resour. Crop Evol. 43, 533546.
  • Meyer, J.D., Deleu, W., Garcia-Mas, J. and Havey, M.J. (2008) Construction of a fosmid library of cucumber (Cucumis sativus) and comparative analyses of the eIF4E and eIF(iso)4E regions from cucumber and melon (Cucumis melo). Mol. Genet. Genomics, 279, 473480.
  • Miao, H., Zhang, S.P., Wang, X.W. et al. (2011) A linkage map of cultivated cucumber (Cucumis sativus L.) with 248 microsatellite marker loci and seven genes for horticulturally important traits. Euphytica, 182, 167176.
  • Mliki, A., Staub, J.E., Sun, Z.Y. and Ghorbel, A. (2003) Genetic diversity in African cucumber (Cucumis sativus L.) provides potential for germplasm enhancement. Genet. Resour. Crop Evol. 50, 461468.
  • Murry, M.G. and Thompson, M.F. (1980) Rapid isolation of high molecular weight DNA. Nucleic Acids Res. 8, 43214325.
  • Naudin, C. (1859) Espèces et des variétés du genre Cucumis. Ann. Sci. Nat. Bot., 6, 587.
  • van Ooijen, J.W. and Voorrips, R.E. (2001) Joinmap 3.0: Software for the Calculation of Genetic Linkage Maps. Wageningen, The Netherlands: Plant Research International.
  • Paris, H.S., Daunay, M.C. and Janick, J. (2012) Occidental diffusion of cucumber (Cucumis sativus) 500-1300 CE,two routes to Europe. Ann. Bot. 109, 117126.
  • Pickersgill, B. (2007) Domestication of plants in the Americas: insights from Mendelian and molecular genetics. Ann. Bot. 100, 925940.
  • Ramachandran, C. and Seshadri, V.S. (1986) Cytological analysis of the genome of cucumber (Cucumis sativus L) and muskmelon (Cucumis melo L). Plant Breed., 96, 2538.
  • Ren, Y., Zhang, Z., Liu, J. et al. (2009) An integrated genetic and cytogenetic map of the cucumber genome. PLoS ONE, 4, e5795.
  • Robbins, M.D., Casler, M.D. and Staub, J.E. (2008) Pyramiding QTL for multiple lateral branching in cucumber using inbred backcross lines. Mol Breed. 22, 131139.
  • Royle, J.F. (1835) Illustrations of the Botany of the Himalayan Mountains. London: Wm. H. Alland & Co.
  • Schatz, M.C., Delcher, A.L. and Salzberg, S.L. (2010) Assembly of large genomes using second-generation sequencing. Genome Res. 20, 11651173.
  • Sebastian, P., Schaefer, H., Telford, I.R.H. and Renner, S.S. (2010) Cucumber (Cucumis sativus) and melon (C. melo) have numerous wild relatives in Asia and Australia, and the sister species of melon is from Australia. Proc. Natl Acad. Sci. USA, 107, 1426914273.
  • Singh, R.J. and Hymowitz, T. (1988) The genomic relationship between Glycine max (L) Merr. and G. soja Sieb. and Zucc. as revealed by pachytene chromosome analysis. Theor. Appl. Genet. 76, 705711.
  • Staub, J.E., Serquen, F.C. and McCreight, J.D. (1997) Genetic diversity in cucumber (Cucumis sativus L). 3. An evaluation of Indian germplasm. Genet. Resour. Crop Evol. 44, 315326.
  • Staub, J.E., Serquen, F.C., Horejsi, T. and Chen, J.F. (1999) Genetic diversity in cucumber (Cucumis sativus L.), IV. An evaluation of Chinese germplasm. Genet. Resour. Crop Evol. 46, 297310.
  • Szinay, D., Chang, S.B., Khrustaleva, L., Peters, S., Schijlen, E., Bai, Y.L., Stiekema, W.J., van Ham, R.C.H.J., de Jong, H. and Lankhorst, R.M.K. (2008) High-resolution chromosome mapping of BACs using multi-colour FISH and pooled-BAC FISH as a backbone for sequencing tomato chromosome 6. Plant J. 56, 627637.
  • Tanksley, S.D. and McCouch, S.R. (1997) Seed banks and molecular maps: unlocking genetic potential from the wild. Science, 277, 10631066.
  • Tanurdzic, M. and Banks, J.A. (2004) Sex-determining mechanisms in land plants. Plant Cell, 16(Suppl.), S61S71.
  • Treangen, T.J. and Salzberg, S.L. (2011) Repetitive DNA and next-generation sequencing: computational challenges and solutions. Nat. Rev. Genet. 13, 3646.
  • Wang, D.H., Li, F., Duan, Q.H., Han, T., Xu, Z.H. and Bai, S.N. (2010) Ethylene perception is involved in female cucumber flower development. Plant J. 61, 862872.
  • Weng, Y., Li, W., Devkota, R.N. and Rudd, J.C. (2005) Microsatellite markers associated with two Aegilops tauschii-derived greenbug resistance loci in wheat. Theor. Appl. Genet. 110, 462469.
  • Weng, Y., Johnson, S., Staub, J.E. and Huang, S.W. (2010) An extended intervarietal microsatellite linkage map of cucumber, Cucumis sativus L. HortSci. 45, 882886.
  • Whitaker, T.W. and Davis, G.N. (1962) Cucurbits. London/New York: Interscience Publishers.
  • de Wilde, W. J. J. and Duyfjes, B.E.E. (2010) Cucumis sativus L. forma hardwickii (Royle) W.J. de Wilde & Duyfjes and feral forma sativus. Thai. For. Bull. (Bot.) 38, 98107.
  • Woycicki, R., Witkowicz, J., Gawronski, P. et al. (2011) The genome sequence of the North-European cucumber (Cucumis sativus L.) unravels evolutionary adaptation mechanisms in plants. PLoS ONE, 6, e22728.
  • Ye, L., Hillier, L.W., Minx, P. et al. (2011) A vertebrate case study of the quality of assemblies derived from next-generation sequences. Genome Biol. 12, R31.
  • Zeven, A.C. and de Wet, J.M.J. (1982) Dictionary of Cultivated Plants and their Regions of Diversity, excluding most Ornamentals, Forest Trees and Lower Plants, 2nd edn. Wageningen, The Netherlands: Centre for Agricultural Publishing and Documentation.
  • Zhang, B., Tolstikov, V., Turnbull, C., Hicks, L.M. and Fiehn, O. (2010a) Divergent metabolome and proteome suggest functional independence of dual phloem transport systems in cucurbits. Proc. Natl Acad. Sci. USA, 107, 1353213537.
  • Zhang, S.P., Miao, H., Gu, X.F. et al. (2010b) Genetic mapping of the scab resistance gene in cucumber. J. Am. Soc. Hortic. Sci. 135, 5358.
  • Zhang, W.W., Pan, J.S., He, H.L. et al. (2011) Construction of a high density integrated genetic map for cucumber (Cucumis sativus L.). Theor. Appl. Genet. 124, 249259.

Supporting Information

  1. Top of page
  2. Summary
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental Procedures
  7. Acknowledgments
  8. References
  9. Supporting Information

Table S1. Information of 735 cucumber and melon markers placed on the high-resolution cultivated cucumber genetic map. Marker loci are arranged by increasing order of map locations in each linkage group (LG). The physical location of each marker in the new Gy14 draft genome assembly (Gy14_Chr_V1.0), their positions in the original 9930 and Gy14 scaffolds are also shown.

Table S2. Information of fosmid clones used in fluorescence in situ hybridization (FISH). Anchored markers are either mapped SSRs used for fosmid library screening or fosmid clone end sequences deposited in GenBank that were used in BLAST sequence alignment. In the later case, no primer information is provided. Map locations of fosmid end sequences were inferred from their scaffold position on the genetic map.

Table S3. Cucumber and melon chromosome synteny inferred from the melon consensus genetic map (Li et al. 2011a) and the Gy14 draft genome assembly (V1.0).

As a service to our authors and readers, this journal provides supporting information supplied by the authors. Such materials are peer-reviewed and may be re-organized for online delivery, but are not copy-edited or typeset. Technical support issues arising from supporting information (other than missing files) should be addressed to the authors.

TPJ_5017_sm_FigS1.pdf129KSupporting info item
TPJ_5017_sm_FigS2.pdf164KSupporting info item
TPJ_5017_sm_FigS3.tif3981KSupporting info item
TPJ_5017_sm_FigS4.tif4885KSupporting info item
TPJ_5017_sm_FigS5.tif2488KSupporting info item
TPJ_5017_sm_FigS6.tif965KSupporting info item
TPJ_5017_sm_FigS7.tif1744KSupporting info item
TPJ_5017_sm_supporting-information-Legends.doc29KSupporting info item
TPJ_5017_sm_TableS1-S3-Rev.xlsx140KSupporting info item

Please note: Wiley Blackwell is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.