Abstract

A BC₃ population segregating for male fertility restoration in CMS tour 25-143 cytoplasm that had been transferred from B. tournefortii to rapeseed was analysed. The segregation of the BC₃ plants into two major classes indicated the fertility restoration to be the action of a single dominant or partially dominant gene. The pollen production of restored plants was somewhat less than in the reference variety ‘Duplo’ but still sufficient for self-pollination. Using bulked segregant analysis 11 AFLP markers linked to the restorer gene were identified. The recombination frequencies between the gene and these markers ranged from 0.02 to 0.13, together they formed a linkage group of 18.3 cM in length. The distribution of markers around the restorer gene indicated normal meiotic recombination in the region surrounding the gene. Accordingly, linkage drag should be no major problem in the application of the restorer gene for CMS tour 25-143 in hybrid breeding.

Introduction

A major prerequisite of hybrid breeding is a working system for pollination control. In rapeseed a number of cytoplasmic male sterility (CMS) systems have been described but many of these are incomplete, missing restorer genes or good maintainers, or cannot be used in hybrid breeding due to other problems. A new cytoplasmic male sterility has recently been developed by protoplast fusion between rapeseed and Brassica tournefortii. One of the fusion products, genotype 25-143, proved to be cytoplasmic male sterile. This genotype had retained the plastids of rapeseed but contained a mitochondrial genome that was the product of a recombination between the mitochondrial genomes of B. tournefortii and rapeseed (Stiewe and Röbbelen 1994). The new cytoplasm is here designated CMS tour 25-143 to distinguish it from CMS tour cytoplasm (syn. CMS juncea), a naturally occurring alloplasmic CMS cytoplasm derived from B. tournefortii (Pradhan et al. 1991) that was first found in B. juncea (Rawat and Anand 1979). Restoration of male fertility in CMS tour and CMS tour 25-143 cytoplasm was later transferred to rapeseed by an interspecific cross with B. tournefortii (Stiewe et al. 1995). Preliminary results from segregation analyses in early backcross generations indicated a monogenic inheritance of this trait, involving a single dominant or partially dominant gene (Sodhi, personal communication). Molecular markers linked to this restorer gene could help to speed up its transfer into new genotypes during the development of pollinator lines for hybrid breeding. Moreover, linked markers would allow an analysis of genetic recombination in the vicinity of the restorer gene to test early for possible problems with linkage drag.

Materials and Methods

Plant material

In the current study a BC₃ population of 89 plants segregating for male fertility restoration in CMS tour 25-143 cytoplasm was used. The male fertility restoration had been transferred from B. tournefortii to rapeseed via an allopolyploid hybrid (AATT) derived from a cross between B. tournefortii (TT) and B. rapa (AA) (Stiewe et al. 1995). The allopolyploid was first crossed with B. napus (AACC) cv. ‘Duplo’ in CMS tour cytoplasm. Repeated backcrosses were done by using first the resulting F₁ (AACT) and later male fertile plants from the previous backcross generation as pollen donor and ‘Duplo’ in CMS tour 25-143 cytoplasm as female parent.

MARKER ANALYSIS

For AFLP analysis the AFLP Kits from Life Technologies were used. Reagents for the digestion of DNA with restriction endonucleases EcoRI and MseI, adapter ligation and preamplification were from the AFLP Core Reagent Kit. Reagents for selective amplification were from the AFLP Starter Primer Kit with the exception of Taq DNA polymerase and γ-³³P-ATP, which were purchased from Pharmacia Biotech and ICN, respectively. All reactions were performed according to the manual included with the AFLP Kits. Amplification products were separated on 5% denaturing polyacrylamide geles and visualised by autoradiography. For linkage analyses and genetic mapping MapMaker Macintosh v. 2.0 (Proctor et al. 1993) was used.

Results and Discussion

Among the plants of the BC₃ population clear differences in male fertility were observed. In addition, all plants showed a deviation in flower morphology from the rapeseed variety ‘Duplo’. With short, narrow filaments and reduced anthers the flowers of sterile plants were similar to flowers of ‘Duplo’ in CMS tour 25-143 cytoplasm. Male fertile plants also had short filaments but fully developed anthers, indicating this distinct flower morphology with shortened filaments to be an effect of the cytoplasm. When the male fertility of the BC₃ plants was scored the majority of plants was found in two classes (Fig. 1): sterile plants with no pollen production (score 1) and fertile plants with a pollen production somewhat less than the variety ‘Duplo’ but quite sufficient for self-pollination (score 7). Only eight plants showed intermediate phenotypes, still male fertile but with less pollen production than plants with score seven. With the plants of the BC₃ population segregating in only two major classes, the most likely hypothesis on the inheritance of the fertility restoration is the involvement of a single gene, confirming preliminary results from earlier backcross generations. The reduced male fertility of the plants with intermediate phenotype may be due to environmental factors or effects from other genes. Generally, the occurrence of plants with reduced male or female fertility is not unusual in segregating populations derived from interspecific crosses. A χ² test with the male sterile plants as one class and all male fertile plants as the other class was in good agreement with the 1:1 ratio expected for the segregation of a single dominant gene in a backcross population (χ² = 0.55, P = 0,46).

To find markers linked with the restorer gene a bulked segregant analysis was performed with AFLP markers and two DNA pools of 15 plants each derived from sterile plants and fertile plants with score seven. A total of 64 primer combinations were tested and 26 markers derived from 17 primer combinations showed polymorphisms between the DNA pools. All positive markers, designated TR1 to TR26, were mapped in an initial set of 44 genotypes from the BC₃ population. In addition, some selected markers were mapped in all 89 plants of this population. A linkage analysis revealed 11 markers to be clearly linked (rf ≤ 0.3) to the restorer gene (Table 1). When all plants producing at least some pollen were classified as heterozygous and all sterile plants as homozygous for the recessive maintainer allele of the restorer gene, the recombination frequencies between the restorer gene and the linked markers ranged from 0.02 to 0.13. Fig. 2 shows the linkage group that could be constructed from these markers and the restorer gene. The linkage group has a total length of 18.3 cM. No markers completely cosegregating with the restorer gene were found. The nearest marker was at a distance of 3.4 cM from the gene. In addition, all markers were found to be on one side of the restorer gene. The latter observation might be due to the restorer gene being located near the end of a chromosome. On the other hand, a more likely explanation is a recombination on the distal site of the restorer gene in an earlier backcross generation. Such an event would produce a genotype homozygous for the recurrent parent in the affected region. Consequently, no marker segregation would occur in this region in further backcross generations.

Table 1: Recombination frequency (rf) between the restorer gene and markers positive in the bulked segregant analysis

The absence of markers completely cosegregating with the restorer gene and the distribution of the linked markers across more than 18 cM as well as the absence of markers on one side of the gene all indicate a normal meiotic recombination in the vicinity of the gene. This makes it likely that the transfer of the restorer gene from B. tournefortii to rapeseed occurred by homoeologous recombination, involving genomic regions with sufficient homology to ensure proper pairing during meiosis. Therefore, linkage drag should be no major problem in the application of this gene in hybrid breeding. In this the CMS tour 25-143 system may differ from the CMS Ogu-INRA system where meiotic recombination is suppressed around the corresponding restorer gene (Delourme et al. 1998) and linkage to a gene inducing elevated seed glucosinolate levels has long prevented its widespread application in hybrid breeding.

Conclusion

The results of this study confirm that the restoration of male fertility in the CMS tour 25-143 cytoplasm that was transferred from B. tournefortii to rapeseed is due to the action of a single dominant or partially dominant gene. The pollen production of plants heterozygous for this restorer gene proved to be sufficient to ensure pollination in hybrids. Markers closely linked to the restorer gene could be developed that will be useful to speed up its transfer into new genotypes during the development of pollinator lines for hybrid breeding. In addition, the distribution of these markers around the restorer gene is indicative of the absence of recombination suppression in the vicinity of the gene. Accordingly, linkage drag should be no major problem in the application of the restorer gene for CMS tour 25-143 in hybrid breeding.

Acknowledgements

The authors thank Y. S. Sodhi and G. Stiewe for providing the plant material analysed in this study. Special thanks are extended to Norddeutsche Pflanzenzucht Hans-Georg Lembke KG and Deutsche Saatveredelung GmbH for providing financial support to M. I. Uzunova.

References

1. Delourme R, Foisset N, Horvais R, Barret P, Champagne G, Cheung WY, Landry BS, Renard M (1998) Characterisation of the radish introgression carrying the Rfo restorer gene for the Ogu-INRA cytoplasmic male sterility in rapeseed (Brassica napus L.). Theor Appl Genet 97:129-134

2. Pradhan AK, Mukhopadhyay A, Pental D (1991) Identification of the putative cytoplasmic donor of a CMS system in Brassica juncea. Plant Breeding 106:204-208

3. Proctor JL, Rafalski A, Hubner R, Tingey S (1993) Engineering Development Laboratory, E.I. duPont de Nemours and Company, Wilminton, DE, USA

4. Rawat DS, Anand IJ (1979) Male sterility in Indian mustard. Indian J Genet Plant Breeding 39:412-414

5. Stiewe G, Röbbelen G (1994) Establishing cytoplasmic male sterility in Brassica napus by mitochondrial recombination with B. tournefortii. Plant Breeding 113:294-304

6. Stiewe G, Sodhi YS, Röbbelen G (1995) Establishment of a new CMS-system in Brassica napus. Proc 9th Int Rapeseed Cong, Cambridge, UK, pp 49-51