Abstract:

Winter oilseed rape, Brassica napus L., with high oleic (HO) acid contents is of interest for nutritional as well as for industrial purposes. Only very limited information is available on the environmental influence on the expression of oleic acid content. A population of 60 doubled haploid (DH) lines segregating for oleic acid content was grown 1998 at three locations in Northern Germany. The DH population was derived from a cross between a HO-mutant (76% C18:1) and the DH line 11.4 of cv. Samourai (55% C18:1). Fatty acid composition, oil, protein and glucosinolate contents were determined of selfed seeds of three plants per line, in two replications and three locations. Analysis of variance revealed a high genetic variability (V_G=93,1%** for 100%= total variation) of the population for oleic acid content and a small but significant environmental variation (V_E=0,8%*) and GxE-Interaction (V_GxE=0,8%**).Seperation of the population into a high (>64% C18:1) and a low (<64% C18:1) class showed high heritabilities (h²=0,94) for C18:1 contents of the high and low oleic acid types. This shows clearly, that oleic acid contents in the seed oil are environmentally stable expressed.

Introduction:

Brassica napus L. is the most important oilseed crop of the temperate climate. The standard rapeseed oil contains about 60% oleic acid (C18:1), 20% linoleic acid (C18:2) and 10% linolenic acid (C18:3). High oleic (HO) rapeseed oil (<75% C18:1) is like HO sunflower and HO soybean oil of interest for nutritional as well as for industrial purposes. This study investigates the environmental influence on oleic acid content in winter rapeseed. Therefore, a population of 60 homozygous DH lines and its parental lines are tested in a two year, three location trial. Here we show the results of the first year.

Materials and methods:

Two DH populations were developed from crosses between the DH line 11.4 derived from cv. Samourai (Uzunova et al. 1995), which has a low oleic acid (C18:1) content in the seeds, and two mutant lines, M19508 and M19566 (M3), selected in an ethylmethansulphonate mutation program of cv. Wotan for high contents of C18:1 in the seed oil. The increase in C18:1 contents in the seed oil of the mutants is monogenically inherited and the mutated genes proved to be allelic (data not shown). Therefore the two DH populations are treated as one.

In total 60 DH lines were produced, 21 lines from the cross M19508 x DH 11.4 and 39 lines from the cross M19566 x DH 11.4. Plants were grown and selfed in the greenhouse (March-June 1996). 60 DH lines and their parental lines were propagated in the field in Göttingen in 1996/97. In 1997/98 60 DH lines and their parental lines were tested at three locations in Northern Germany (Göttingen, Hohenlieth, Thüle). The trial was sown in Göttingen on August, 22nd, in Hohenlieth on August , 20th, and in Thüle on september, 9th, 1997. In each location a randomized block design with two replications was used. Three plants were selfed in each DH line per replication and location. Selfed seeds were analysed for their fatty acid composition by gas chromatography following Thies (1971) and by NIRS for oil protein, and glucosinolate (GSL) content.

Analysis of variance was carried out by PLABSTAT (Utz 1994). Heritability in the broad sense (h²) was calculated following Hill et al. (1998). Components of variance were obtained from the ANOVA: h²= σ²G /( σ²G + σ²GE/E + σ²GER/ER) (G=genotype; E=environment; R=replication). Probability levels are indicated as follows: ** for P≤0,01; * for P≤0,05 and + for P≤0,10.

Results:

The parental lines of the DH population differ significantly for C18:1**, C18:2**, C18:3** and protein* contents (Tab. 3). Mean values of C18:1 contents of the parental lines are slightly higher than the mean of the DH population (Tab.2; 3). No transgression for C18:1 has been monitored in the DH population (Tab. 1).

Due to the monogenic segregation of the high oleic acid content (Fig. 1), the population was seperated in a high oleic acid class (29 DH lines) and a low oleic acid class (31 DH lines) (Tab. 4). Means of the high oleic class differ significantly from the low oleic acid class for C18:1** and oil content*, while protein and glucosinolate contents do not reveal any significant differences.

The analysis of variance of the segregating DH population shows significant variation for all fatty acids, oil, protein and glucosinulate contents. Environment (E) influences significantly C18:1*, C18:2**, C18:3**, oil**, protein** and glucosinulate* contents and for all factors GxE-interactions are significant (*, oil+). Calculation of ANOVA for the high and low oleic acid class of the DH population for C18:1 contents reveals a minor environmental influence (+) and only for the low C18:1 class significant GxE effects**. The main relative component of variation for C18:1 content in the seeds is the genotypic effect (93%) while there is only a minor environmental effect (0,84%) (Tab. 5b). In contrast to that, oil contents show a high environmental effect (47,4%). Heritability of C18:1 in the whole DH population is very high (0,99) and only slightly lower (0,94) when calculated for the high and low C18:1 classes seperately (Tab. 4)

The C18:1 content is negatively correlated with C18:2 (r²=-0,98**) and C18:3 (r²=-0,59*) contents and positively correlated (r²=0,39**) with oil content.

Tab. 1: Mean values, standard deviation, range of seed oil oleic acid contents (%C18:1) of the DH population in 1997/98 at three locations. Minimum and maximum values are related to single selfed plants. Parental means are calculated for Göttingen.

Tab. 2: Mean values of fatty acid composition, oil, protein and glucosinolates (GSL) of 60 DH lines tested in three locations in 1997/98.

Tab. 3: Seed composition of the parental lines of the DH population. Mean values of lines, replications and environments.

Tab. 4: Means and heritabiliy of oleic acid, oil, protein and GSL contents in the high and low oleic acid class of the DH population, tested in three locations.

Tab. 5: Variance components from analysis of variance of the DH population, tested in three environments

F-test of respective mean squares are indicated as follows: ** for P≤0,01; * for P≤0,05 and + for P≤0,10.

Discussion:

The aim of this experiment is a quantification of the environmental effects on the trait oleic acid content in winter rapeseed oil. A low environmental stability of this trait would (1) require a high extent of testing in selection and (2) would make an environmentally independent and stable production of high oleic acid contents in rapeseed difficult.

Environmental effects on the fatty acid composition in Brassica napus L. have been reported earlier. For low C18:3 contents in the seed oil there have been estimated heritabilities of 0,82 (Rajcan et al. 1997), 0,92 (Pleines and Friedt 1988) and 26 - 59% (Kondra and Thomas 1975), all in spring rapeseed. For C18:3 contents at least three major genes (Somers et al. 1998) and probably severall minor genes (Rücker and Röbbelen 1996) are involved in the desaturation step of C18:2 to C18:3. In contrast to this the mutants M19508 and M19566 show a clear monogenic segregation of the oleic acid content. Since the heritability of monogenic traits is higher than that of polygenic traits a high heritability is expected for C18:1 contents. This has been confirmed by a heritability of h²=0.99 of the whole DH population. Separation of the population in a high (>64% C18:1) and a low oleic acid class (<64% C18:1) shows also a high heritability of h²=0,94 in the high and in the low class. This might be due to the non discrete segregation of the DH population. Assuming a monogenic segregation, all high types and all low types, resp., should have the same ‘oleic acid genotype’, resulting in two distinct groups. But since there is variation in the class, the heritability in the class is high. This indicates, that there are modifying effects influencing the oleic acid content in the seed oil.

Low heritability and significant environmental effects on the oil content might be the consequence of the late sowing time in Thüle. Even though the oil content is significantly reduced in this location, the oleic acid content in the seeds remains stable and is not influenced by the sowing time.

The high heritability of C18:1 contents in the seedoil shows clearly, that oleic acid contents in the seed oil are environmentally stable expressed. The stability of this trait not only over locations, but over years will have to be demonstrated in a second year of testing.

Acknoledgements: We thank Christian Möllers and Beate Rücker for the development of the doubled haploids, Fachagentur Nachwachsende Rohstoffe (FNR) and Gemeinschaft zur Förderung der privaten deutschen Pflanzenzüchtung (GFP) for financial support and Norddeutsche Pflanzenzucht (NPZ) and Deutsche Saatveredlung (DSV) for carrying out field trials.

References:

1. Hill, J.; Becker, H.C.; Tigerstedt, P.M.A. (1998): Quantitative and Ecological Aspects of Plant Breeding. Chapman and Hall

2. Kondra, Z.P.; Thomas, P.M. (1975): Inheritance of linoleic and linolenic acids in seed oil of rapeseed Brassica napus. Can. J. Plant Science 55 p. 205-210.

3. Pleines S.; Friedt W. (1988): Breeding for improved C18-fatty acid composition in rapeseed (Brassica napus L.). Fat Science Technology 90 (5) p. 167-171.

4. Rajcan, I.; Kott, L.A.; Beversdorf, W.D.;Kasha, K.J. (1997): Performance of doubled haploid populations segregating for linolenic acid levels in spring rapeseed. Crop Science 37 p.1438-1442.

5. Rücker, B.; Röbbelen, G. (1996): Impact of low linolenic acid content on seed yield of winter oilseed rape (Brassica napus L.). Plant Breeding 115 p. 226-230

6. Somers, D.J.; Friesen, K.R.D.; Rakow, G. (1998): Identification of molecular markers associated with linoleic acid desaturation in Brassica napus. TAG 96 p. 897-903

7. Thies, W. (1971): Schnelle und einfache Analysen der Fettsäurezusammensetzung in einzelnen Rapskotyledonen. I. Gaschromatographische und papierchromatographische Methode. Z. Pflanzenzüchtung 65 p. 181.202

8. Utz, H.F. (1994): Plabstat, ein Computerprogramm für statistische Analysen von Pflanzenzüchtungsexperimenten. Universität Hohenheim.

9. Uzunova, M.; Ecke, W.; Weissleder, K.; Röbbelen, G. (1995): Maping the genome of rapeseed (Brassica napus L.). I. Construction of an RFLP linkage map and localization of QTLs for seed glucosinulate content. TAG 90 p.194-204.