ABSTRACT

Sinapis alba L. (white mustard) shows considerable promise as an alternative oilseed crop for dry temperate climates. It possesses many beneficial characteristics such as drought and heat tolerance, pest resistance and a short growing season. We present the first genetic linkage maps of S. alba. These high density RFLP maps will assist in the rapid development of S. alba as a new oilseed crop. A four-parent cross using mustard lines of diverse origins produced two reciprocal, highly polymorphic mapping populations and these were probed with Brassica RFLP clones. In all, 380 loci were mapped on 12 linkage groups. Detailed analysis of the RFLP maps revealed large segments of chromosome that were duplicated and often triplicated. The use of a common set of RFLP probes allowed the comparative mapping of S. alba with related Brassica species. The ability to assay separately the inheritances of male and female gametes from two different hybrid genotypes of S. alba allowed the rate and distribution of recombination in male and female meioses to be compared and the observed differences have implications for mustard breeding. Single-seed descent is being used to develop 360 inbred lines for the field testing of quality and yield characters. Genotyping of these lines with RFLP probes will allow the mapping of important QTLs, particularly those controlling oil to protein ratio.

INTRODUCTION

Sinapis alba has been cultivated for centuries as a condiment crop. It has a number of favourable agronomic traits that makes it attractive as an alternative oilseed crop to rapeseed. For example, S. alba produces large, yellow seeds, performs well in dry, hot conditions, and shows excellent resistance to pests and diseases.

However, the grain yield and seed oil content is relatively low compared to rapeseed, and seeds generally contain high levels of erucic acid and glucosinolates which are considered detrimental to human and animal nutrition.

Restriction Fragment Length Polymorphism (RFLP) technology is well established in Brassica species and has been used to produced high-density linkage maps of the agriculturally important species such as Brassica napus (Parkin et al, 1995, Sharpe et al, 1995) and B. oleracea (Bohuon et al, 1996). Such a map would be very useful for understanding the structure of the S. alba genome and for assisting in its development as an oilseed crop.

MUSTARD BREEDING

Research to establish rigorous genetic analysis of S. alba is being carried out collaboratively by AAFC Saskatoon Research Centre and the John Innes Centre in order to assist in the development of white mustard as a new canola crop. Figure 1 presents the four-way crossing strategy employed to maximise genetic variation in the two reciprocal F1-mapping populations (F1-1 and F1-2). The four “grandparental” lines were of diverse origin and contrasting seed oil characteristics – including one low yielding canola-quality line. Single-seed descent is being used to produce 360 F5 inbred lines for field testing in seasons 1999/2000. RFLP-genotyping of these lines will allow the mapping of QTLs (Quantitative Trait Loci) governing important agronomic characters such as seed oil content and quality.

Figure 1 Four-parent crossing strategy employed to produce two reciprocal F1 mapping populations and 360 F5 inbred lines for field testing and QTL analysis.

RFLP MAPPING

The RFLP methodology and probes used were essentially as described by Sharpe et al (1995). 160 Brassica clones were used to probe DNA from 128 F1 individuals cut with EcoRI, EcoRV or HindIII. In total, 545 loci were scored and assigned to either the P1 or P2 map. 165 loci were common to both maps, 98 were mapped only in P1 and 117 were mapped only in P2. This gave a total of 380 different loci mapped over both maps, an average of 2.4 polymorphic loci per probe.

The loci formed 12 linkage groups in both maps and those common to both maps were used to identify equivalent linkage groups. The total map lengths differed slightly – 955cM for P1 and 840cM for P2 (Kosambi mapping function). The average interval between adjacent loci was 3.6 cM in the P1 map and 3.0cM in P2.

Comparative mapping

The Brassica clones used in mapping S. alba have previously been used to produce maps for several Brassica species, for example, B. napus (Parkin et al, 1995), B. oleracea (Bohuon et al, 1996) and B. nigra (Lagercrantz and Lydiate, 1995). Lagercrantz and Lydiate (1996) made an in-depth comparison of the B. rapa, B. nigra and B. oleracea linkage maps. Such a comparison is being carried out between S. alba and related crucifer species, with the closely-related B. nigra being of particular interest. Not only are these studies useful for increasing our understanding of the evolutionary relationships between crucifer species, but should also assist the effective transfer of favourable alleles between agriculturally important crucifer crops.

Male and female meioses

Recombination in male and female gametes can differ considerably in some species. In B. nigra, the rate of recombination in male gametes is higher close to the ends of chromosomes, and in female gametes the rate is higher in the centromeric regions (Lagercrantz and Lydiate, 1995). However no such differences were found in B. napus (Kelly et al, 1996).

Differing distributions of recombination rates in male and female meioses can influence the choice of which parent to use as the pollen donor in a breeding programme, for example in the introgression of a disease resistance gene from a wild relative into an agronomically superior cultivar by recurrent backcrossing to the cultivar. A high recombination rate is desirable near the gene of interest as this will reduce the genetic drag of undesirable genes linked to the gene of interest. However, for the rest of the genome, a lower recombination rate is desirable in order to maximise the rate of enrichment of the superior recurrent parent genotype.

In order to test for differences in the distribution of recombination in male and female gametes of S. alba, the F1 mapping populations were increased in size to 188 F1-1 individuals (where P1 was the female parent) and 194 F1-2 individuals (where P2 was the female parent). These additional individuals were genotyped with a subset of probes selected to give even marker coverage across the P1 and P2 maps. The significance of differences in the frequency of recombination between adjacent markers in male and female meioses is being assessed.

CONCLUSIONS

The aim of this project was to increase our understanding of the S. alba genome. This was done by studying the control of recombination during male/female meioses and comparing the S. alba maps with other crucifer species. These studies will assist the rapid development of S. alba as a new canola crop.

ACKNOWLEDGEMENTS

This research was supported by the John Innes Foundation and Agriculture and AgriFood Canada.

REFERENCES

1. Bohuon EJR, Keith DJ, Parkin IAP, Sharpe AG, and Lydiate DJ, 1996. Alignment of the conserved C genomes of Brassica oleracea and Brassica napus. Theor. Appl. Genet. 93: 883-839

2. Kelly AL, Sharpe AG, Nixon JH, Evans EJ, and Lydiate DJ, 1997. Indistinguishable patterns of recombination resulting from male and female meioses in Brassica napus (oilseed rape). Genome 40: 49-56

3. Lagercrantz U and Lydiate DJ, 1995. RFLP mapping in Brassica nigra indicates differing recombination rates in male and female meioses. Genome 38: 255-264

4. Lagercrantz U and Lydiate DJ, 1996. Comparative mapping in Brassica. Genetics 144: 1903-1910

5. Parkin IAP, Sharpe AG, Keith DJ, and Lydiate DJ, 1995. Identification of the A and C genomes of amphidiploid Brassica napus (oilseed rape). Genome 38: 1122-1131

6. Sharpe AG, Parkin IAP, Keith DJ, and Lydiate DJ, 1995. Frequent nonreciprocal translocations in the amphidiploid genome of oilseed rape (Brassica napus). Genome 38: 1112-1121