Development of Cucumber Cultivars Resistant to
Root-knot Nematodes

S. Alan Walters and Todd C. Wehner

Department of Horticultural Science, North Carolina State University, Raleigh, NC 27695

Kenneth R. Barker

Department of Plant Pathology, North Carolina State University, Raleigh, NC 27695

Additional index words. Meloidogyne spp, Meloidogyne arenaria , Meloidogyne javanica , Meloidogyne hapla, Cucumis sativus, Cucumis sativus var. hardwickii, Cucumis metuliferus, heritability

Abstract. Root knot caused by Meloidogyne sp. is the most important disease of cucumber (Cucumis sativus L.) in North Carolina, a leading producer of cucumbers in the United States. In order to develop root-knot nematode resistant cultivars, a research program was initiated to 1) screen the C. sativus and C. metuliferus germplasm collection for resistance, 2) develop effective nematode testing methods, 3) study interspecific transfer of M. incognita resistance from C. metuliferus to C. sativus, 4) measure the inheritance of resistance to M. javanica, 5) measure the inheritance of resistance to M. arenaria, 6) study the mechanism of resistance, and 7) develop resistant germplasm

Screening the germplasm collection.
In a screening of 800 cultivars, breeding lines, and plant introduction accessions (hereafter referred to as cultigens) in the greenhouse, resistance to Meloidogyne arenaria races 1 and 2, and M. javanica was found in Cucumis sativus var. hardwickii line LJ 90430. Moderate resistance to M. arenaria race 2 in was found in C. sativus var. sativus 'Southern Pickler' and 'Mincu' as well. Resistance was verified in a retest in the greenhouse. Another cultigen of C. sativus var. hardwickii, PI 215589, was found to be resistant to M. arenaria race 2, but not to any other species or races.

Field vs. greenhouse testing. For resistance to be effective, it must control the pathogen/pest under field conditions. Therefore, a greenhouse­field correlation test was conducted over 2 years. Ten cultigens were evaluated for resistance to M. arenaria races 1 and 2, and M. javanica under greenhouse and field conditions. Those tests indicated that the cultigens found to be resistant under greenhouse conditions were also resistant in the field. Over both years, greenhouse and field data were highly correlated (r = 0.74), with repeatabilities ranging from 0.83 to 0.99. This study indicated that the response of a cucumber plant to root-knot nematode infection under greenhouse conditions is a good indicator of its reaction in the field.

Interspecific transfer of resistance to M. incognita.
Cucumis metuliferus has a high level of resistance to M. incognita compared to C. sativus, which has no resistance to this nematode in its germplasm pool. Autotetraploids of C. sativus and C. metuliferus were, therefore, created in an attempt at interspecific hybridization between the two species. Best results for the production of tetraploids for both C. sativus and C. metuliferus were obtained when seeds were immersed in 0.5 % colchicine for 6 to 8 hr. Crosses were made in all combinations of diploid and tetraploid lines of C. sativus with diploid and tetraploid lines of C. metuliferus. In crosses involving C. sativus as the female parent, fruit development often occurred. All seeds within the fruit were, however, flat and nonviable. No fruit development occurred in all crosses made using C. metuliferus as the female parent.

Inheritance of resistance to M. javanica. Resistance to M. javanica was identified in the C. sativus var. hardwickii line LJ90430. Parents, F1, F2, and BC1 to both parents of a cross between 'Sumter' (C. sativus var. sativus) and LJ 90430 were evaluated in two greenhouse experiments. All F1 progeny were susceptible, and segregation ratios in the F2 resulted in 1resistant : 3 susceptible. Backcross progeny to the susceptible parent were susceptible, and the BC1 to the resistant parent segregated 1 resistant : 1 susceptible. Reciprocal crosses

Cucurbitaceae '98


did not differ and showed, therefore, no evidence of maternal or cytoplasmic effects. Results from the crosses of several inbreds ('Addis', Gy14, Gy 57u, 'Poinsett 87', and 'Sumter') with LJ90430 indicated that use of those five different genetic backgrounds had no influence on gene expression. Thus, resistance to M. javanica is conferred by a single recessive gene for which we propose the symbol mj.

Inheritance of resistance to M. arenaria. Resistance in cucumber to M. arenaria races 1 and 2 was studied in four families ('Sumter' x LJ 90430, LJ 90430 x 'Sumter', 'Addis' x LJ 90430, and 'Poinsett 87' x LJ 90430) under greenhouse and/or field microplot conditions. Based on genetic and phenotypic correlations between M. arenaria races 1 and 2, a split-root experiment with different races on different parts of the same root system for the cross 'Sumter' x LJ 90430 indicated that the same genes were conditioning resistance to both races of M. arenaria. Greenhouse and field microplot experiments gave similar results with the additive genetic variance being the largest component of genetic variance. High estimates of broad-sense (HB) and narrow-sense (hN2) heritabilities were obtained in all families tested [ranging from 0.49 to 0.83 for (HB) and (hN2) for M. arenaria race 1, and from 0.61 to 0.90 for (HB) and (hN2) for M. arenaria race 2]. The reciprocal test indicated that maternal and cytoplasmic effects may be present since both F1 and F2 generation plants were more resistant to M. arenaria race 2 when LJ 90430 was used as the maternal parent. The number of effective factors involved in the resistance was estimated and ranged from 2 to 5. The resistance to M. arenaria races 1 and 2 appears to be highly heritable, and breeding methods that best utilize additive genetic variance should be used.

Mechanism of resistance. Relative rates of root-knot nematode penetration and the histopathology of the infections were studied in susceptible and resistant Cucumis sp. Five species and/or races of Meloidogyne tested were able to penetrate resistant as well as susceptible Cucumis cultigens. Significant differences (P 0.05) were detected between nematodes with respect to cultigen penetration. In LJ 90430, numbers of stage 2 juveniles (J2s) of M. arenaria races 1 and 2, M.

javanica, and M. hapla (the nematodes to which LJ 90430 has resistance), remained relatively high at 15 days after inoculating, which indicated that they were not developing normally. Also, no J3s or J4s of M. hapla were present in roots at 15 days after inoculation, which also indicated that the development of M. hapla in LJ 90430 was being delayed or terminated, or that J2s were moving out of the roots. Results were similar for C. metuliferus PI 482454. However, in 'Sumter', all root-knot nematodes developed normally from J2 to J3 or J4 stages except for M. hapla, which had arrested development. Histology of roots indicated that the resistance observed could not be explained by a hypersensitive reaction. Thus, resistance is due to improper development of the parasite due to the host not providing proper conditions at the feeding sites.

Development of resistant germplasm. An original bulk of LJ 90430 (selection from PI 183967) was tested using a split-root technique for resistance to M. javanica and M. arenaria races 1 and 2; resistant plants were self-pollinated in the greenhouse. The homozygous resistant inbred that resulted after five generations of selection and self-pollination was released as NC-42. An original bulk of 'Southern Pickler' (developed from a cross of 'Producer' x 'Ohio MR 17') was tested for resistance to M. arenaria race 2; resistant plants were self-pollinated in the greenhouse. 'Mincu' is its original source of nematode resistance. The homozygous resistant inbred that resulted after six generations of selection and self-pollination was released as NC-43. A population was developed from LJ 90430 crossed with 12 cultivars, breeding lines, and plant introductions. The population, designated NCH1, was improved for yield, earliness, and quality of American pickling type fruit for nine cycles of recurrent selection. Plants from cycle nine of NCH1 were tested for resistance to M. javanica and M. arenaria races 1 and 2 and self-pollinated. After seven generations of selection for nematode resistance and self-pollination, inbreds were field tested for yield, earliness, and quality. Three inbreds with high performance were released for use by the industry: 'Manteo' (NC-44), 'Shelby' (NC-45), and 'Lucia' (NC-46).

Cucurbitaceae '98