ABSTRACT

The cross between B. napus and B. oleracea is normally infertile, but the use of embryo culture techniques can produce viable hybrids. Hybrids have been produced between forage rape (B. napus ssp. biennis) and kale (B. oleracea ssp. acephala) and also between rape and cauliflower (B. oleracea ssp. botrytis). The main programme has involved doubling the chromosomes of the hybrids with colchicine to produce the possible new species hybrid B. napoleracea. The hybrids are usually highly fertile when back-crossed to rape but produce only occasional seed when back-crossed to kale. Aphid resistance and self-incompatibility have been transferred to rape, and new combinations of glucosinolates in the plant tissues have been obtained.

INTRODUCTION

Brassica napus is an allotetraploid species derived from the parental species B. oleracea (kale/cabbage group) and B. campestris (syn. B. rapa, turnip group). The cross is naturally highly infertile. Fertilisation may take place, but abortion occurs early in the development of the embryo. Embryo rescue techniques can produce viable hybrids. Embryo culture was first used by Nishi et al. (1959) and ovary and ovule culture has since been used to increase the efficiency of the cross (Diederichsen and Sacristan, 1994). Desired characters can, therefore, be transferred to B. napus from either parent by resynthesis.

Characters could be more directly transferred from B.oleracea to B. napus by crossing the two species together. This cross is also normally infertile, but again embryo rescue techniques can produce hybrid plants (Ayotte et al., 1987). The hybrids can be backcrossed to either parent and act as a bridging cross to transfer characters between the species (Quazzi, 1988). Chromosome doubling of the hybrids produces the allohexaploid B. napoleracea (Chiang, 1985) and the possibility exists of producing a fertile new species hybrid. Backcrosses can be made between the triploid or hexaploid hybrid with either parent.

MATERIALS AND METHODS

Initial crosses were made between rape and kale using rape breeding lines H93 and H95. These lines were produced from (rape x kale) x rape crosses by Quazzi (1988) and, therefore, may have been ‘pre-conditioned’ for making this cross. The kale cultivars used in the crosses were Merlin and Condor. A second set of crosses used rape cultivars Arran, Emerald and Sparta, which were crossed with kale cultivars Curlew and Rawara. One other cross was made initially - rape H95 was crossed with All Year Round Cauliflower.

Pollinations were carried out on the largest unopened flower buds, which had been destaminated, and were then covered with glassine paper bags to prevent uncontrolled pollinations. The pollinations were carried out at Gore, Southland, and the ovule culture at Crop & Food Research headquarters, Lincoln, near Christchurch.

Developing capsules were excised from the plants 12 - 28 days after pollination, sealed in polythene bags and sent to Lincoln. On arrival, the capsules were surface sterilised and ovules were dissected out and cultured on Linsmaier-Skoog (LS) medium (Linsmaier and Skoog, 1965) containing 1 mg/litre benzyladenine. After seven weeks a mass of fused leaf tissue was present from which normal shoots developed. The shoots were transferred to hormone-free LS medium and multiplied by nodal and apical cuttings. Shoots were transferred to soil once there was sufficient root and shoot development. With the second set of crosses, the ovaries were first cultured on Nitsch and Nitsch (1956) medium for 28 days, before ovules were dissected and cultured as before.

Chromosome doubling was initially attempted by placing drops of 0.1% colchicine in the leaf axils of mature plants, after first scraping away the epithelium in the axil. The apical meristem was also removed. From 14 plants, with several axils treated per plant, only one tetraploid shoot was produced. The embryo culture technique allowed another batch of seedlings to be produced from two of the embryos. The seedlings were treated when young plants by immersing the roots in 0.05% colchicine for 12 hours, rinsing overnight in running water, and transplanting to compost in pots. This treatment produced several tetraploid branches on four plants from each embryo. The hybrids from the second set of crosses were doubled using this method.

RESULTS

For the first set of crosses, 480 pollinations were made, 213 developing capsules were dissected and 570 ovules were cultured; three hybrids were obtained. As colchicine treatment of the hybrids was mostly unsuccessful, 14 triploid plants produced from them were allowed to open pollinate in an insect-proof cage with bumble bees as pollinators. Six seed were obtained, but none of them were spontaneously doubled hybrids. Plants from these seeds were backcrossed to rape. Selfed and second backcross lines have been grown in the field and tested agronomically as forage rapes.

In the second set of crosses, approximately a third of the developing capsules were left on the plant to mature as a control; the rest were excised and sent for culture. From 71 pollinations, 62 capsules were allowed to develop on the plant and four seeds were produced - three large and a smaller one. It was expected that these would be maternally derived; this appeared to be the case for the large seeds, but the smaller seed was a triploid hybrid. Five hybrids were produced from culture, and these were obtained from 140 capsules produced by 150 pollinations.

The second set of crosses produced several lines of putative hexaploids, which have been selected on pollen size. Several of these lines were highly fertile. The hybrids were generally intermediate between rape and kale for most morphological characters. The hexaploids from cabbage described by Chiang (1995) were self-compatible, but the hybrids developed from kale were mostly self-incompatible. With the present material, crosses to rape were highly fertile, while those with kale were not.

After backcrossing to rape, stable, fertile rape lines have been obtained after the first generation of selection and selfing. A wide range of variation is present in these lines, including aphid resistance and self-incompatibility. Glucosinolates have been tested in some of these lines (Table 1, from Muhammad et al., 1997). A wider glucosinolate spectrum has been produced, and lines with both higher and lower levels of total glucosinolate than normal have been obtained.

Table 1: Glucosinolate content of rape lines derived from rape x kale interspecific crosses and their parents.

DISCUSSION AND CONCLUSIONS

The variation between subspecies of B. napus is widely divergent, ranging from oil-seed forms, through forage rapes, to swedes. A corresponding range occurs within B. campestris. To transfer a character from B. oleracea to B. napus through resynthesis, it would be logical to use the type of B. campestris that was desired in B. napus. However, it is even more logical to use an adapted cultivar of B. napus and cross this with the desired line of B. oleracea.

Initial crosses at Crop & Food Research were made using oil-seed rape (Quazzi, 1988), because at the time there was a oil-seed breeding programme. Since that programme terminated, the material produced has been used in the forage rape breeding programme. However, because of the origin of the material, there have been problems with early flowering when used for a forage crop, even though the hybrids were backcrossed immediately to forage types. To avoid this problem, the more recent work has used standard forage rapes as the initial parents in the crosses.

The production of triploid hybrids between rape and kale allows backcrossing to take place with the parental species. It is limited, however, to crosses that can be made immediately, as the triploids are highly infertile and cannot be maintained. Doubling the triploids to produce fertile hexaploids, even though these may not be completely stable, enables the hybrids to be multiplied and maintained. Numerous backcrosses can be made to these hybrids over a period of several years, allowing a wider and more varied range of rape lines to be used for crosses during this time.

REFERENCES

1. Ayotte R., Harney P.M. and Souza Machado V. 1987. The transfer of triazine resistance from Brassica napus L. to B. oleracea L. I. Production of hybrids through embryo culture. Euphytica 36: 615-624.

2. Chiang M.S. 1995. Brassica napoleracea Cruciferae Newsletter 10: 25.

3. Diederichsen E. and Sacristan M.D. 1994. The use of ovule culture in reciprocal hybridization between B. campestris L. and B. oleracea L. Plant Breeding 113: 79-82.

4. Muhammad S., Kirkegaard J., Gowers S. and Garrett B.C. 1997. Glucosinolate content and animal acceptability of forage brassicas. Cruciferae Newsletter 19: 13-14.

5. Nishi S., Kawata J. and Toda,M. 1959. On the breeding of interspecific hybrids between two genomes, c and a of Brassica through the application of embryo culture techniques. Japanese Journal of Breeding 8: 215-222.

6. Nitsch J.P. and Nitsch C. 1956. Auxin dependant growth of excised Helianthus tuberosus tissues. American Journal of Botany 43: 839-851.

7. Quazzi M.H. 1988. Interspecific hybrids between Brassica napus L. and B.oleracea L. developed by embryo culture. Theoretical and Applied Genetics 75: 309-318.