Markers Linked to Agronomic Traits of
Cucumber (Cucumis sativus L.)

Hidemasa Mizusawa, Nobuo Hirama, and Seiji Matsuura

Kiyohara Breeding Station, Tohoku Seed Co., Utsunomiya, Tochigi 321 3232, Japan

Additional index words. Cucumis sativus, DNA marker, agronomic trait, little leaf (ll )locus

Abstract. Two DNA markers linked to the little-leaf character, controlled by little-leaf (ll) locus, and other agronomic traits of cucumber were found. RFLP marker (C-396 ) and ll locus are linked with a recombination value of 34.4% ± 0.6%. RAPD marker (B-31) is linked with the ll locus from the other side with a recombination value of 18.1% ± 0.03%. Furthermore, there is a linkage relationship between B-31 and a QTL controlling petiole length and internode length, and a linkage relationship between C-396 and a QTL controlling monoecious sex phenotype.


Marker-assisted selection (MAS) promises to greatly increase the efficiency of selecting individuals possessing a desirable phenotype from a segregating population in a cross-breeding program (Edwards and Johnson et al., 1994). Most agronomic traits (for example, leaf size, yielding ability, fruit length, and disease resistance) in commercial varieties of Japanese cucumbers are quantitative, and so it is difficult to introgress other desirable traits into an existing variety. Although isozyme, RFLP and RAPD markers have been used for construction of a genetic map of cucumber (Kennard et al., 1994; Knerr and Staub, 1992; Meglic and Staub, 1996), there has not been much progress in identifying isozyme and molecular markers linked to agronomic traits of cucumber.

We have studied the possibility of breeding a Japanese little leaf cucumber for intensive greenhouse cultivation as a means of easing the work load for greenhouse workers. This is necessitated due to an overall decrease in greenhouse workers and an increase in the number of older greenhouse workers in Japan. Initial results indicated that there are no cucumber varieties possessing the little leaf and high yielding ability for greenhouse cultivation in Japan (Matsuura and Fujita, 1995). There was no relationship between leaf size and yielding ability in segregating cucumber populations (Mizusawa et al., 1996). Here we report on RFLP and RAPD markers linkage with agronomic traits for use in

a MAS cucumber breeding program for development of a Japanese little leaf cucumber.

Materials and methods

An F2 population (containing 130 individuals) derived from a cross between 'Kyuuraku No. 2' and 'Morioka No. 1', the parents and the F1 hybrids were used. 'Morioka No. 1' is a little leaf intermediate parental line containing ll locus (Ito et al., 1989). 'Kyuuraku No. 2' is a Japanese local variety of a South Chinese type with a large leaf and high yielding ability.

All plant materials were planted in a greenhouse isolated from pollinating insects in Utsunomiya, Japan in Summer 1995. Plants were trained on vertical netting and uniformly pruned by pinching the main stems at the 17th node. Lateral branches at the first three nodes were removed, and 14 first lateral branches were pruned at the second node (all secondary laterals were removed as soon as they appeared). All pistillate flowers on the main stem were removed before anthesis. The number of pistillate flowers allowed to appear on the first lateral branch of each plant was limited to 12 by removing unnecessary pistillate flowers before flowering.

Fruit were harvested when they were 100 g. The number of aborted pistillate flowers on each individual were recorded. True leaf width and length, a petiole length, internode length of the lateral stem at the 10th node, and internode length of the main stem were measured. The number of

Cucurbitaceae '98

pistillate flowers bearing on a main stem were also counted. These data were used for linkage analyses.

In the summer of 1997, 120 individuals of a little leaf phenotype in a F2 population derived from a cross between 'NK-1' and 'KM-25' were planted in a greenhouse for linkage analysis of the RAPD marker B-31 and ll locus, which were in coupling phase. Because the gene pair of B-31 marker and ll locus in a F2 population of 'Kyuuraku No. 2' and 'Morioka No. 1' was in repulsion phase, it is not possible to accurately estimate the recombinant value. 'NK-1' is a greenhouse type parental line of a F1 commercial variety of Tohoku Seed Co. 'KM-25' is a selected line from a cross between 'Kyuuraku No. 2' and 'Morioka No. 1', and possesses ll genotype and high-yielding ability (Mizusawa et al., 1997). Plants were pinched uniformly as in 1995. True leaf width and leaf length at the 10th node were measured 30 days after planting.

Total DNA was isolated from the shoot apex of the lateral stem by CTAB method (Murrey and Thompson, 1980). Total DNA of both parents was independently digested with five restriction enzymes (Eco RI, Eco RV, Hind III, Bam HI and Pst I); 3 µg were electrophoresed through 1.2% agarose gel, and then transferred onto nylon membrane (Amersham). Nineteen DNA clones selected from cucumber genomic and cDNA libraries constructed by Matsuura and Fujita (1995a, 1995b) were used as probes for Southern blot analyses after being labeled with ECL gene detection kit (Amersham).

Membranes were hybridized, overnight at 42 oC in a hybridization buffer that was in the ECL kit. They were washed twice in a primary wash buffer that contained 0.5¥ SSC, 0.4% SDS and 6 m Urea at 42 oC for 20 min, and then twice in a secondary

wash buffer that contained 1.0x SSC at room temperature for 10 min. The membranes were exposed to Fuji medical X-ray film for 1 h after being soaked with a detection solution.

One-hundred and twenty primers (12-mer) for RAPD analyses were obtained from WAKO (DNA Oligomer 12 Set A and B; 60 primers per set). Reactions were performed in 25-µL volumes in 0.5-mL polypropylene tubes containing 50 ng of DNA, 0.1 mm of primer, 4 units of Tth polymerase (Boehringer Mannheim), 10 mm each dNTP and ion-exchanged water. PCR reactions times were 15s at 94 oC, 15s at 47 oC, and 2 min at 72 oC for 30 cycles (Boimetra). PCR products were electrophoresed through 2.0% agarose gel with 0.5 g·mL­1 of ethidium bromide in 1x TAE (Sambrook et al., 1989) at 50 V for 90 min. DNA clones and primers detecting polymorphisms between both parents were used for segregation analyses in F2 populations.

Segregations in F2 populations were tested for chi-square goodness of fit. Significant differences among mean values of the measured traits in each genotype of DNA marker in F2 population were tested by t test. Means separations were performed using lsd. Linkage analyses were performed using the maximum likelihood method of Allard (1956).

Results and discussion

Mean values of agronomic traits of 'Kyuuraku No. 2', 'Morioka No. 1' and their F1 hybrids are shown on Table 1. Distributions of agronomic traits in the F2 population were continuous with the exception of the true leaf width and length, variables that segregated in an expected 3 (long):1 (short) ratio (Wehner et al., 1987). Seven DNA markers were polymorphic between 'Kyuuraku No. 2' and 'Morioka No. 1'. Four of these markers

Table 1. Mean values of agronomic traits of 'Kyuuraku No.2' (K), 'Morioka No.1' (M), and their F1 hybrids.

True leaf (cm) Petiole Internode length (cm) Pistillate Fruit set

Entry Width Length length (cm) Main stemz Lateral stem flowers (no.) (no.)

K 32.7 ± 0.6 26.1 ± 0.4 25.2 ± 0.7 157.6 ± 2.5 17.1 ± 1.9 2.9 ± 0.4 11.5 ± 0.5

M 15.9 ± 0.2 12.4 ± 0.2 18.1 ± 0.9 75.1 ± 1.6 6.8 ± 0.5 11.4 ± 0.5 4.3 ± 0.8

F1 (K ¥ M) 28.6 ± 0.3 20.6 ± 0.9 27.4 ± 0.9 147.0 ± 2.6 17.8 ± 2.3 3.8 ± 0.5 7.0 ± 0.8

F1 (M ¥ K) 28.4 ± 0.3 21.1 ± 0.7 26.9 ± 0.7 158.0 ± 2.7 13.8 ± 2.7 4.0 ± 0.0 4.8 ± 1.3

zTotal length of 17 nodes.

Cucurbitaceae '98

Table 2. Mean values of agronomic traits in each genotype of C-396 RFLP and B-31 RAPD markers in an F2 population derived from a cross between 'Kyuuraku No. 2' and 'Morioka No. 1'.

No. True leaf (cm) Petiole Internode length (cm) Pistillate Fruit set

Marker Genotypez tested Widthx Length length (cm) Main stemy Lateral stem flowers (no.) (no.)

C-396 AAK 33 28.5 ax 22.5 a 27.3 144.9 a 17.1 a 5.8 a 8.4

A_ 58 26.3 ab 20.0 b 25.2 138.4 ab 15.2 a 5.2 a 7.6

aaM 31 24.0 18.0 c 24.6 129.5 b 12.7 b 7.8 b 8.3

B-31 A_M 96 24.6** 18.9** 24.8** 133.4** 14.2* 6.3 7.8

aaK 28 31.6 23.8 28.6 151.6 17.4 4.7 8.0

zK = 'Kyuuraku No. 2' genotype; M = 'Morioka No. 1' genotype.

yTotal length of 17 nodes.

xMeans followed by the same letter do not differ at 5% significance level by Fisher's lsd-method.

*,**Significant at the 5% and 1% level, respectively.

Table 3. Linkage analysis between the ll locus and C-396 RFLP marker, and between B-31 RAPD and C-396 markers in an F2 population derived from a cross between 'Kyuuraku No. 2' and 'Morioka No. 1'.

Gene pair Segregation Goodness of fit (3:6:3:1:2:1)

A B A_BB A_Bb A_bb aaBB aaBb aabb Total c2 P (5 df) rcv (%)

ll x C-396 Obs. 27 39 16 6 19 15 122 12.2 0.05­0.03 34.4 ± 0.06

B-31 x C-396 Obs. 23 41 31 9 14 4 122 6.3 0.5­0.25 ---

Exp. 22.9 45.7 22.9 7.6 15.3 7.6 122.0


Table 4. Linkage analysis between the ll locus and B-31 RAPD marker of little leaf phenotype individuals in an F2 population derived from a cross between 'NK-1' and 'KM-25'.

Gene pair Segregation Goodness of fit (12:3:1)

A B AB, Ab aB aabb Total c2 P (2 df) rcv (%)

ll x B-31 Obs. (367) 38 77 115 103.5 <0.001 18.1 ± 0.03

Exp. 361.5 90.4 30.1 482.0

were RFLPs (C-143, C-396, C-566 and P-023). Three markers were RAPD markers (B-27, B-31, B-66). In addition, the B-31 marker was polymorphic between 'NK-1' and 'KM-25'. The four polymorphic RFLP markers were codominant, and the three polymorphic RAPD markers were dominant. These seven markers segregated independently (data not shown).

Two markers, C-396 and B-31, were associated with at least one of the traits that are shown in Table 2. C-396 was associated with true leaf width and length, internode lengths of the main and lateral stems, and numbers of pistillate flowers. B-31 was associated with true leaf width and length, internode lengths of main and lateral stems, and petiole length.


Three-point analyses detected linkage between ll and C-396, and between ll and B-31 (Tables 3, 4). The recombination value between ll and C-396 is 34.4% ± 0.06%, and that between ll and B-31 is 18.1% ± 0.03%. Their order is shown in Figure. 1.

We postulate the following. 1) That a QTL controlling the number of pistillate flowers is located between ll and C-396 based on the relationship between true leaf size and number of pistillate flowers (Mizusawa et al., 1996). 2) That a QTL controlling petiole length is located near B-31. C-396 and B-31, and that it may be used for MAS selection of leaf size, number of pistillate flowers, and petiole length. Further markers will be evaluated in order to find QTLs for fruit-setting ability.


Cucurbitaceae '98

Figure 1. Linkage relationships and rcv (%) among ll locus, and C-396 RFLP and B-31 RAPD markers resulting from the linkage analyses in Tables 3 and 4.

In: G.E. Lester and J.R. Dunlap (eds.) Proc. Cucurbitaceae 94: Evaluation and enhancement of cucurbit germplasm.

Matsuura, S. and Y. Fujita. 1995b. RFLPs in Japanese cucumber varieties. Breeding Sci. 45:91­95.

Meglic, V. and J.E. Staub. 1996. Inheritance and linkage relationships of isozyme and morphological loci in cucumber (Cucumis sativus L.). Theor. Appl. Genet. 92:865­872.

Mizusawa, H., N. Hirama, and S. Matsuura. 1996. Correlation between leaf size and horticultural characteristics in cucumber. Cucurbit Genet. Coop. Rpt. 19:21­22.

Mizusawa, H., N. Hirama, and S. Matsuura. 1997. Heritability of parthenocarpic ability in the cucumber. Proc. 8th SABRAO Gen. Congr. 252­253.

Murrey, M.G. and W.F. Thompson. 1980. Rapid isolation of high molecular weight plant DNA. Nucleic Acid Res. 8:4321­4325.

Sambrook, J., E.F. Fritsch, and T. Maniatis. 1989. Molecular cloning: A laboratory manual. Cold Spring Harbor Laboratory Press, Cold Spring Harbor, N.Y.

Wehner, T.C., J.E. Staub and C.E. Peterson. 1987. Inheritance of little leaf and multibranched plant type in cucumber. Cucurbit Genet. Coop. Rpt. 10:33­34.

Literature cited

Allard, R.W. 1956. Formulas and tables to facilitate the calculation of recombination values in heredity. Hilgardia 24(10):235 278.

Edwards, M.D. and N.J. Johnson. 1994. Evaluation of marker-assisted selection through computer simulation. Theor. Appl. Genet. 88:376­382.

Ito, K., S. Kamimura, D. Ishiuchi, and T. Mochizuki. 1989. Application of small leaf Characteristic to cucumber breeding. J. Jpn. Soc. Hort. Sci. 58:174­175.

Kennard, W.C., K. Poetter, A. Dijkhuizen, V. Meglic, J.E. Staub, and M.J. Havey. 1994. Linkage among RFLP, RAPD, isozyme, disease-resistance and morphological markers in narrow and wide crosses of cucumber. Theor. Appl. Genet. 89:42­48.

Knerr, L.D. and J.E. Staub. 1992. Inheritance and linkage relationships of isozyme loci in cucumber (Cucumis sativus L.). Theor. Appl. Genet. 84:217­224.

Matsuura, S. and Y. Fujita. 1995. Correlation between agronomic characters of cucumber cultivars grown in Japan. J. Jpn. Soc. Hort. Sci. 64:305­313.

Matsuura, S. and Y. Fujita. 1995a. RFLP mapping of locus F controlled sex expression in cucumber, p. 172­174.

Cucurbitaceae '98