Selection for high oleic acid in ‘zero’ erucic acid

Sinapis alba


J Philip Raney, Gerhard FW Rakow and Todd V Olson


Agriculture and Agri-Food Canada, Saskatoon Research Centre, 107 Science Place, Saskatoon, SK., Canada, S7N 0X2




As part of the Saskatoon Research Centre’s (AAFC-SRC) ongoing program to develop Sinapis alba germplasm which will meet the Canadian canola definition for seed quality, half-seed selections were made within a low (less than 5%) erucic acid line of S. alba for ‘zero’ erucic acid lines. Lines were found which contained less than 0.1% erucic acid. Lines containing approximately 1% erucic acid were also identified which could be distinguished from the ‘zero’ erucic acid lines at the half-seed level and confirmed at the single plant level. However, within the ‘zero’ erucic acid lines there was observed at the half-seed level considerable variation for oleic acid, from less than 60% to greater than 80%. Linolenic acid content also varied from 7% to 15%. In some half-seeds linolenic acid was much higher than linoleic acid, while in others the reverse was true. Because of the recent interest in high oleic/low linolenic acid canola oil, selection for a high oleic/low linolenic acid phenotype was carried out for six generations in this germplasm. Selection was accomplished by gas chromatographic analysis of the fatty acid methyl esters derived from the outer cotyledon of germinated seeds. These profiles were later confirmed by analysis of seed of individual plants grown from selected half-seeds. Plants with the highest levels of oleic acid were then selected for another cycle of half-seed selection. Composites of the selected plants of each of the six generatons were grown in isolation tents in 1996 and 1998. The results indicate that selection of a high oleic/low linolenic acid phenotype is possible in S. alba by this method.


KEYWORDS: fatty acid composition, half-seed selection, yellow mustard




Yellow mustard (Sinapis alba L) is successfully grown as a condiment crop on the Canadian prairies, especially in Saskatchewan and Alberta. It has superior heat and drought tolerance in comparison to Brassica napus and B. rapa canola and is therefore well adapted for production in this region. Yellow mustard is also highly shatter resistant and has a large bright yellow seed. Further advantages include tolerance to blackleg and flea beetle attack. Despite these many advantages yellow mustard is not grown as an oilseed crop in Canada because the seed normally contains relatively high concentrations of erucic acid and glucosinolate and is low in oil content (25-28%). However, it has been grown as an oil crop in Sweden (Olsson 1974). It is now possible to develop a canola quality S. alba, because of the existence of a low erucic acid line developed at Saskatoon, a low glucosinolate line developed in Poland (Krzymanski et al. 1991) and a high oil content (36-38% oil) line developed in Sweden (Olsson 1974). One of our objectives for S. alba is to develop high oil content, low erucic and low glucosinolate cultivars suitable for production on the Canadian prairies as an oilseed crop (Raney et al. 1995). In order to achieve this it was necessary to reselect for ‘zero’ erucic acid lines from the Saskatoon low erucic acid line utilizing the half-seed technique of Downey and Harvey (1963). In doing so, lines containing less than 0.1% erucic acid were identified which could be distinguished from lines containing approximately 1% erucic acid. In addition, considerable variation in the oleic acid content of these half-seeds was observed. Because of the current interest in high oleic acid types of edible oils, selection for a very high oleic acid content line was attempted. This paper describes the results of that effort.




At each generation selection pressure was applied for a high oleic and low linolenic fatty acid profile. All selections were based on half-seed analysis. The selections were then confirmed by analysis of a five, ten or fifty seed ‘bulk’ analysis of seed produced on the plants grown from selected half-seeds. Growth of the plants in the greenhouse or growth chambers followed standard greenhouse practices. Open pollinated (OP) seed was produced by brush pollination of selected plants grown in isolation. Selfed seed was produced by bud pollination. In 1996 and 1998 bulk composites of selected lines of all six generations of ‘zero’ erucic acid S. alba were grown at the AAFC-SRC farm site as 1 meter by 6 meter plots under isolation tents to prevent cross pollination with other lines of S. alba.


The fatty acid composition of whole seeds and germinated half-seeds (18 hours at RT) was determined by a method similar to that of Thies (1971) as follows. The oil of ‘bulk’ samples (up to 2 grams) was first extracted by shaking the seed, a metal rod and hexane together in a 20 ml PET scintillation vial in reciprocating shaker. An appropriate amount of oil and solvent was then transferred to an autosampler vial. Transesterification was accomplished in 0.1 ml 0.8% sodium metal in methanol and 0.05 ml heptane for 20 minutes at RT. The outer cotyledon of germinated seeds was transesterified by grinding it with a glass rod in a 1.5 ml autosampler vial in the presence of transesterification reagent. After transesterication, 0.1 ml of 0.2 M sodium phosphate, pH 7.0 was added to vial and the sample briefly dried bunder a stream of air (1 minute) and 0.5 ml of heptane added. Chromatography of the fatty acid methyl esters was performed with a Hewlett Packard 6890 gas chromatograph as described in Figure 1.



















Figure 1. Gas chromatogram of seed oil from the high oleic acid S. alba line grown in 1998.





The low erucic acid (<5%) parent line (BHL-926) was half-seed selected for lowest possible erucic acid content (16 out of 272 with <1%). From the bulk analysis of the OP seed, the 6 lines of the 16 were separated into two groups. Group 1 averaged 0.4% erucic acid and group 2 averaged 1.4% erucic acid. Half-seed selection was carried out on each of the two groups (20 seeds from each plant. Plants of group 1 segregated seeds that contained no erucic acid by our analysis. Plants of group 2 did not segregate ‘zero’ erucic acid half-seeds. The selected ‘zero’ or low erucic acid half-seed progeny from each of the two groups were grown in isolation in the greenhouse and OP F2 seed harvested. Upon half-seed analysis harvested seed of group 1 plants were found to be all 0% erucic acid, whereas seed from group 2 plants was found to contain between 1% and 5% erucic acid. Therefore group 2 plants were discontinued. Selected half-seeds from group 1 contained oleic acid contents between 67% and 71%. Selected plants (n = 54) were bud selfed and F3 seed harvested. Based on the ‘bulk’ analysis 17 lines were selected for continuation by half-seed selection (12 seeds analyzed per line). Selected half-seeds (52) contained between 70% and 77% oleic acid. Selected plants were bud selfed and F4 seed harvested. The seed from 16 lines was bulked together and grown at the farm site in 1994. Analysis of the harvested seed from the field demonstrated that it contained significant amounts of erucic acid, probably as the result of cross-pollination with the large condiment yellow mustard nursery grown at the same site. It was then decided that thereafter this material would be grown under isolation tents to ensure isolation. The bulked F4 seed was also half-seed selected and ten seeds out of 30 were selected with oleic acid contents between 70% and 80%. Bulk analysis of F5 seed from nine lines harvested revealed that actual oleic acid content ranged between 70% and 76%. Three lines were selected for continuation. 48 seeds of each were analyzed and a total of 45 seeds selected. In this generation several plants were lost due to a severe aphid infestation in the growth chamber used. Bulk analysis of F6 seed from 26 lines revealed nine lines with >78% oleic acid. 24 half-seeds from each of these nine lines were analyzed and 30 selected for selfing. Bulk analysis of the harvested F7 seed from these plants revealed that the oleic acid was not higher than the F6 seed analysis. Composites from all plants grown at each generation were then made and these were grown at the AFFC-SRC farm site in 1996 in isolation tents. In 1998 seed from the 1996 isolation tents was used to repeat the 1996 field trial. The fatty acid composition of the seed grown from these composites is shown in Table 1. The results indicate that selection for higher oleic acid (C18:1) contents was successful. At the same time selection for a lower content of linolenic acid (C18:3) and ‘zero’ erucic acid (C22:1) was also successful. The oleic acid content was elevated at least 5%, the linolenic acid content lowered by 3%, and erucic acid content reduced to 0.1% or less.


Table 1. Fatty acid profiles of six generations of selection for high oleic acid and parental lines


Line                Gen.   Year  C16:0  C16:1  C18:0   C18:1  C18:2  C18:3  C20:0   C20:1  C22:0  C22:1

BHL-9261      Parent   GH       3.6       0.3       1.8     57.3     11.8     13.6       0.5       6.2       0.3       4.5

Group 11         F1         GH       5.1       0.5       1.7     58.9     17.1     13.6       0.5       1.7       0.3       0.4

Group 21         F1         GH       4.9       0.5       1.9     58.1     15.9     13.0       0.6       3.4       0.3       1.4


TO93-08762   F2      1996       3.7       0.2       1.9     69.2     10.5     12.0       0.5       1.1       0.2       0.0

                                1998       4.1       0.3       1.9     68.7     12.2       9.1       0.6       1.2       0.3       0.1

TO96-17002   F3      1996       3.7       0.2       1.8     71.3       9.8     10.7       0.5       1.1       0.2       0.1

                                1998       4.2       0.3       1.9     70.1     11.6       8.3       0.6       1.3       0.3       0.1

TO96-17012   F4      1996       3.9       0.3       1.9     71.6     10.1       9.7       0.5       1.1       0.2       0.0

                                1998       4.2       0.3       1.9     72.6     10.2       7.4       0.5       1.2       0.2       0.0

TO96-17022   F5      1996       3.8       0.3       1.7     74.3       9.1       8.6       0.4       1.0       0.2       0.0

                                1998       4.1       0.3       1.6     71.8     10.2       7.7       0.5       1.6       0.2       1.0

TO96-17032   F6      1996       3.7       0.3       2.0     74.7       8.9       8.0       0.5       1.0       0.2       0.0

                                1998       4.2       0.3       1.8     73.9     10.0       6.8       0.5       1.2       0.2       0.1

TO96-17042   F7      1996       3.7       0.3       1.9     74.9       8.9       8.1       0.5       1.0       0.2       0.1

                                1998       4.1       0.3       1.7     74.3       9.6       6.8       0.5       1.2       0.2       0.1

1.        Bulk analysis of greenhouse produced seed.

2.        Bulk analysis (400 seeds) of seed produced in 1996 and 1998 in isolation tents.







Selection of a ‘zero’ erucic acid S. alba line with nearly 75% oleic acid and less than 8% linolenic acid in its seed oil was achieved. This indicates that the oil quality of future canola quality S. alba cultivars can be competitive with the canola that is presently grown on the Canadian prairies. Other hurdles, such as the low oil content of S. alba relative to canola, will have to be overcome before it can be grown successfully as an oilseed crop.




This project was supported in part by grants from the Western Development Grain Research Fund and the Agri-Food Innovation Fund.




Downey, R.K. and Harvey, B.L. 1963. Methods of breeding for oil quality in rape. Canadian Journal of Plant Science 43, 271-275.


Krzymanski, J., Pietka, T., Ratajska, I., Byczanska, B. and Krotka, K. 1991. Development of low glucosinolate white mustard (Sinapis alba syn Brassica hirta). Proceedings of the Eighth International Rapeseed Congress, Saskatoon, 5, 1545-1548.


Olsson, G. 1974. Continuous selection for seed number per pod and oil content in white mustard. Hereditas, 77, 197-204.


Raney, P., Rakow, G. and Olson, T. 1995. Development of low erucic, low glucosinolate Sinapis alba. Proceedings of the Ninth International Rapeseed Congress, Cambridge, UK, 2, 416-418.


Thies, W. 1971. Schnelle und einfache Analysen der Fettsäurezusammensetzung in einzelnen Raps-kotyledonen I. Gaschromatographische und papierchromatographische Methoden. Zeitschrift für Pflanzenzüchtung, 65, 181-202.