Marker assisted selection of new high oleic and low linolenic winter oilseed rape (Brassica napus L.) inbred lines revealing good agricultural value

Development of oilseed rape (Brassica napus L.) breeding lines producing oil characterized by high oleic and low linolenic acid content is an important goal of rapeseed breeding programs worldwide. Such kind of oil is ideal for deep frying and can also be used as a raw material for biodiesel production. By performing chemical mutagenesis using ethyl methanesulfonate, we obtained mutant winter rapeseed breeding lines that can produce oil with a high content of oleic acid (C18:1, more than 75%) and a low content of linolenic acid (C18:3, less than 3%). However, the mutant lines revealed low agricultural value as they were characterized by low seed yield, low wintering, and high content of glucosinolates in seed meal. The aim of this work was to improve the mutant lines and develop high-oleic and low-linolenic recombinants exhibiting both good oil quality and high agronomic value. The plant materials used in this study included high-oleic and low-linolenic mutant breeding lines and high-yielding domestic canola-type breeding lines of good agricultural value with high oleic acid content and extremely low glucosinolates content. Field trials were conducted in four environments, in a randomized complete block design. Phenotyping was performed for wintering, yield of seed and oil, and seed quality traits. Genotype × environment interaction was investigated with respect to the content of C18:1 and C18:3 acids in seed oil. Genotyping was done for the selection of homozygous high oleic and low linolenic lines using allele-specific CAPS markers and SNaPshot assay, respectively. Finally, new high oleic and low linolenic winter rapeseed recombinant lines were obtained for use as a starting material for the development of new varieties that may be of high value on the oil crop market.


Introduction
Oilseed rape, OSR, (Brassica napus L. var. oleifera, ssp. napus) also called canola [1] is one of the most important oil crops contributing a huge volume to the global production of oil (68 million metric tons (MMT) in 2017) next to soybeans (330 MMT) and followed by oil palm (64 MMT) (http://www.fao.org/faostat/en/#home). Rapeseed oil has been used for many years mainly for human nutrition and major improvements have been made for over 40 years through intensive breeding for enhancing the quality of seed and oil [2][3][4]. Canola oil is beneficial to human health and nutrition; it includes a very low amount of saturated fatty acids, palmitic C16:0 and stearic C18:0 (in total about 7%), and mainly mono-and polyunsaturated fatty acids, oleic C18:1 (about 62%), linoleic C18:2 (20%), linolenic C18:3 (10%) and eicosenoic C20:1 (1%) [5][6][7], in addition to bioactive compounds and antioxidants, important components of functional food including natural biofortification [8,9]. Such oil is valuable as salad oil and salad dressing. However, due to the presence of a large amount of polyunsaturated fatty acids, the rapeseed oil is less suitable for deep frying and industrial purposes, as there is a possibility of the formation of trans-fatty acids and oil oxidation during high-temperature treatment or long-term storage [10]. As functional oils with an enhanced nutritional value are of great importance [11], multiple efforts are undertaken to improve the oil quality for specific purposes of the highly competitive global plant oil market.
The physical and chemical properties of canola oil depend on the composition of seed oil fatty acids [5,12,13], especially the major C18 mono-and polyunsaturated fatty acids [10]. High content of oleic acid in rapeseed oil is desirable for its long shelf life and stability at high temperature [14], which makes it an optimal raw material for biodiesel production [15]. Polyunsaturated linolenic acid (C18:3) is prone to oxidation and unstable during frying and its reduced level in rapeseed oil is required for long storage [16] and also optimal for deep-fat frying [17], thus making it a target for fast-food producers.
In addition to its food-use, rapeseed oil is also applied as a raw material for the production of lubricants, pharmaceuticals, tensides, biodegradable plastics, and varnishes [18]. Moreover, in recent years, there has been an increasing interest in the use of plant oils with high oleic acid and low linolenic acid as a renewable raw material in the production of biofuels [18,19]. Renewable energy sources are of significance [20,21] and are gaining huge support for use in energy sectors [11] due to the need for lowering the emission of carbon dioxide which has increased mainly as a result of the use of fossil fuels as an energy source, particularly by the transportation sector. The increasing demand for biodiesel reinforces an increase in the production and acreage of rapeseed which benefits local farmers and also oil factories. In the European Union, over 11 MMT of biodiesel was produced in 2016; it is noteworthy that the seed meal remaining after biodiesel processing can be used as feedstock and a natural resource of brassinosteroids and plant hormones, due to its beneficial properties for both humans and plants [22]. However, further efforts in improving Brassica oilseeds to produce oil with high oleic and low linolenic acids, so-called high stable oil [1], with a special focus on reducing the content of linolenic acid to prevent oil oxidation [23] are needed.
Improved long storage and thermo-stability of canola oil has been one of the special goals of breeding programs for about two decades [11,12,[24][25][26]. In 2007, the HOLL (high oleic low linolenic) canola oil was defined by Maher et al., in new spring canola varieties in Australia, as the oil characterized by more than 65% of oleic acid and less than 3% of linolenic acid [27]. Subsequently, Canadian and Australian open pollinated as well as hybrid spring HOLL varieties were characterized by 68% of oleic acid and 3% of linolenic acid in seed oil. They were released by Cargill and reported by Salisbury et al. [http://www.australianoilseeds. com/__data/assets/pdf_file/0005/7079/salisbury,_phil_00150_.pdf]. In Europe, the newly registered winter HOLL hybrid varieties were characterized by about 80% of oleic and less than 3% of linolenic acid in seed oil [https://www.dsv-seeds.com/oilseed-rape/winter-osr/holloilseed-rape.html; https://www.weloveholl.co.uk/]. A current standard industry-wide definition of so-called HOLL (or HOLLi) canola includes varieties producing oil with more than 75% of oleic acid and less than 3% of linolenic acid. It was defined by Barth [17], regarding nutritional value of vegetable oils in a case of their application for frying. Friedt and Snowdon [26] also stated that oil produced by so-called HOLL canola varieties is characterized by significantly higher oxidative stability while frying due to a strong reduction of trans-fatty acids formation, resulting from the low amount of linolenic acid (C18:3) in seed oil [28].
Initially, HOLL rapeseed genotypes, including breeding lines, were developed by chemical mutagenesis using ethyl methanesulfonate (EMS) [7,13,[29][30][31][32]. Subsequent selection of HOLL mutant genotypes in breeding programs by phenotyping using gas-liquid chromatography has been laborious and time-consuming due to the fact that the composition of fatty acids in seed oil is highly influenced by solar radiation and temperature [33] which makes it difficult to reliably compare genotypes from different experiments. A genetic-based approach to genotype selection seems more promising for breeding programs. Examples of successful approaches exploit, e.g., fatty acid desaturases FAD2 and FAD3, which are involved in the formation of oleic and linolenic acid, respectively [5,6,34]. QTL genetic mapping using several DNA markers, such as RAPD, AFLP, SSR, and SNP, revealed the loci linked to the content of oleic and linolenic acid on the A04 and C10 chromosomes of B. napus [13,35,36]. Allele-specific DNA markers were identified for an effective selection of homozygous mutated genotypes that can produce seed oil with changed content of oleic and linolenic acid in breeding programs. Restriction site generating-polymerase chain reaction (RG-PCR) assay was developed by Barret et al. [37] for the selection of low linolenic (LL) B. napus genotypes derived from the 'Oro' LL mutant developed by Rakow et al. [38], CAPS markers were developed by Falentin et al. [39] for determining the high oleic (HO) genotype, and Illumina qPCR-based assay was developed by Hu et al. [30] for the detection of both mutated FAD2 and FAD3 alleles as well as SNaPshot by Mikolajczyk et al. [40] for the detection of FAD3, and SNP markers by Yang et al. [13] for the detection of allelic variation of FAD2 and FAD3 genes. Such functional markers are highly useful for the direct selection of desirable fad2 and fad3 mutations in marker-assisted trait introgression and breeding of HOLL-type canola.
In this paper we report the development of HOLL winter oilseed rape (WOSR) breeding lines of good agricultural value. After combining two breeding programs, 24 genotypes of different origin and properties were selected, grown in four environments and characterized through plant phenotyping and DNA marker analysis. In this work, we analyse the results obtained from the field experiments and identify genotypes with the most favorable set of traits, in particular seed oil fatty acid composition, good yield and low glucosinolates content. We also investigate trait stability by analyzing genotype-environment interaction. The selected desired genotypes make a valuable starting material to be used for further breeding.

Plant material
In this study, we combined two independent breeding programs to obtain WOSR breeding lines characterized by high oleic and low linolenic acid content in seed oil in addition to low content of glucosinolates (LGLS) in seed meal, as well as good agricultural value including improved seed yield and wintering. One of the breeding programs aimed at the development of HOLL mutant recombinants by crossing HOmut (�75%) and LLmut (�3%) WOSR mutant lines obtained previously by Spasibionek [31] via chemical mutagenesis using ethyl methanesulfonate (EMS). The mutant HOmutLLmut F5 inbred recombinants obtained could not be used directly for pre-breeding as they were of low agricultural value characterized by low yield, low wintering, and increased GLS content in seed meal. Another breeding program aimed at recurrent selection of HO-and LGLS-type domestic WOSR accessions of good agricultural value [41]. Ten high oleic and low glucosinolates (HOLGLS) breeding lines were selected in several independent field trials. The selected lines were of HO type, which can produce oil characterized by about 78% of oleic acid (C18:1) in seed oil. In addition, their seed meal had extremely low GLS content (5-10 μmoles/g of seeds) and their seed yield was up to 48.7 dt ha -1 [41].
To combine the two programs, an F5 inbred mutant recombinant (HOmutLLmut_837) was crossed with the 10 selected HOLGLS domestic breeding lines to develop new HOLL-type recombinants of good agricultural value. The breeding scheme is presented in Fig 1. From 220 analyzed recombinants, ten breeding lines were selected for detailed assessment in field trials together with their ancestral genotypes. Thus, in this study, plant material from all stages of the breeding program is investigated: ten HOLGLS domestic breeding lines, seven LLmut&HOLGLS and three HOmut&HOLGLS inbreds, one HOLLmut mutant recombinant line in addition to its parental mutant lines, HOmut and LLmut, and also cv. Monolit, a highyielding Polish WOSR cultivar, as a reference ( Table 1). The HOLLmut_837 line was an F5 inbred derived from crossing between a low linolenic (�3%) mutant (LLmut_681) breeding line [31] and a high oleic (�75%) mutant (HOmut_10464) line [31] (Fig 1; Table 1). Phenotyping was performed for seed yield (dt ha -1 ), wintering (%), oil content of seed (% of dry seed weight), GLS content of seed (μmoles g -1 of seeds), and content of C18 unsaturated fatty acids in seed oil (C18:1 and C18:3; %). Wintering percent was measured by calculating the proportion of the number of plants in one row (3 m 2 ) before and after winter.

Field trials and phenotyping
Biochemical analysis of fatty acid composition in seed oil and GLS content in seed meal was done using a Hewlett Packard Agilent Technologies 6890N Network GC System. For GLS analysis, desulfoglucosinolates silyl derivatives were used [42,43], and the European Commission CRM366 standard of 12.1 μmoles g -1 of seeds with a tolerance limit of 0.8 μmoles g -1 of seeds was applied. The seed oil content was measured using a broadband mass NMR spectroscopy analyzer (Newport Instruments Ltd.) [44].

Statistical analysis
Statistical analysis of the phenotypic traits aimed at assessing the effects of genotypes and environments with respect to the observed traits, investigating the stability of oleic and linolenic acid content, and identifying genotypes with most favorable phenotypes. In the analyses, the combinations of year and location were regarded as independent environments. Significance of the effects of genotype, environment and their interaction was assessed by analysis of

PLOS ONE
variance (ANOVA). Genotypic means, variances, and least significant differences at two significance levels (0.05 and 0.01) for all traits were obtained in Tukey multiple comparison test. Computations were performed in R statistical software. Assessment of accessions' stability by means of analysis of the genotype-environment interaction (i.e., estimation of main and interaction effects for genotypes for C18:1 and C18:3 acid content as well as their visualization as biplots) was performed in SERGEN [45], according to the methodology developed by Kaczmarek [46] and Calinski et al. [47].

Total DNA extraction and genotyping of allelic variants of FAD2 and FAD3 desaturase genes
Genomic DNA was isolated from young leaves using the modified CTAB method [48], as described by Mikolajczyk et al. [49]. The quality and quantity of DNA samples were estimated on 0.8% agarose gels using 50 ng of undigested lambda DNA as a reference, and by A 260 and A 280 UV absorbance measurements [49]. For monitoring the mutated and wild-type alleles of the functional BnA.FAD2 desaturase gene, in the A genome of B. napus [50], involved in the synthesis of oleic acid, allele-specific CAPS markers [39] were used. PCR amplification products were analyzed using 1.4% agarose gel electrophoresis and scored as FAD2_A for wild-type and fad2_a for mutant alleles of BnA.FAD2 desaturase gene. Wild-type and mutant alleles of FAD3 desaturase genes in the A and C genomes of B. napus, BnA.FAD3 and BnC.FAD3, respectively, were scored as FAD3_A and FAD3_C for wild-type alleles and fad3_a and fad3_c for mutant alleles, using an allele-specific SNaPshot assay involving two steps: PCR amplification of short regions including possible mutation sites followed by microsequencing [40].

Seed yield and wintering
In total, 24 accessions were assessed in multi-environment field trials and the reference cv. Monolit was characterized by the highest seed yield (34.89 dt ha -1 ). Twelve accessions revealed slightly lower but comparable seed yield of 28 dt ha -1 and more (regarding the calculated Tukey's least significant difference (LSD) at the significance level of 0.01) (  Table 2). The best wintering accession, LLmut&HOLGLS_882, was characterized by 87.23% of wintering. Five other genotypes, including the reference cv. Monolit (81.30%), revealed more than 80%: two HOLGLS (480 and 535), the HOmutLLmut_837, and the HOmut&LGLS_852. The LLmut_681 genotype revealed the lowest 37.85% of wintering. Two other accessions, LLmut&HOLGLS_440, and HOLGLS_520 showed low wintering percentage (less than 60%). The remaining 15 genotypes overwintered comparably to the best wintering accession, regarding the LSD 0.01 (Table 2).
There was a significant differentiation among the analyzed accessions with respect to the content of both oleic acid and linolenic acid in seed oil. The HOLGLS_481 domestic accession showed the highest C18:1 acid content (79.59%) and two HOmut&HOLGLS lines (852, and 873) revealed more than 79% as well. Six other lines revealed more than 78% of oleic acid in seed oil: the HOmut_10464 (78.63%), the HOLGLS_480 domestic accession (78.59%), the HOmut&HOLGLS_850 (78.40%), three LLmut&HOLGLS lines (880-78.37%, 882-78.29%, and 899-78.30%), as well as the LLmut&HOLGLS_880 (78.37%). Interestingly, the reference cv. Monolit showed the lowest oleic acid content of 64.46% (Table 2). Regarding the content of linolenic acid in seed oil in the analyzed genotypes, the LLmut_681 line showed the lowest content of 2.34% and was used as a reference for the LLtype. Only three HOLL-type recombinant lines-LLmut&HOLGLS 878, 880, and 882 revealed less than 3.5% of linolenic acid. At the same time, they showed the homozygous low linolenic genotype ('FAD3_A/C haplotype' in Table 2).
The GLS content in seed meal was extremely low in the HOLGLS lines (including also the LLmut&HOLGLS_440 line), ranging from 5.34 to 9.21 μmoles g -1 , as compared to cv. Monolit (12.16 μmoles g -1 of seeds). Moreover, all the genotypes were characterized by less than 15.00 μmoles g -1 of seeds (except for the LLmut&HOLGLS_878 line, which showed 21.57 μmoles g -1 of seeds) ( Table 2).

Genotype × environment interaction
Genotypes were evaluated in four environments (two years in two locations) by the variance analysis (Table 3) and the estimation of mean genotype and genotype by environment effects (Table 4). ANOVA for seed yield and seed quality traits (oil content, C18:1 and C18:3 fatty acids-content in seed oil and total seed GLS content in seed meal) revealed a statistically Table 2 significant effects of genotype (G), environment (E), and the influence of the environment on genotypes (GxE) for all traits (Table 3). Genotype × environment interaction for C18:1 and C18:3 fatty acids in seed oil was assessed by estimating the main G effects and G × E interaction and their significance. As shown in Table 4, out of 24 tested accessions ten genotypes, including the reference cv. Monolit, revealed a significant interaction with the environment regarding the content of oleic acid: low linolenic mutant LLmut_681 line, high oleic and low linolenic mutant recombinant HOLLmut_837, two of ten analyzed HOLGLS lines (490 and 543) and three of seven LLmut&HOLGLS recombinants (882, 878, and 902), in addition to two of the three analyzed HOmut&HOLGLS recombinants (850 and 852) ( Table 4). With respect to C18:3 linolenic acid content, nine genotypes showed a significant interaction with the environment: low linolenic mutant LLmut_681 line, high oleic mutant HOmut_10464, high oleic and low linolenic mutant recombinant HOLLmut_837, three out of the 10 analyzed HOLGLS lines (519, 535, and 561) as well as-similarly to C18:1 oleic acid content-two of the three analyzed HOmut&HOLGLS recombinants: 850 and 852 (Table 4). At the same time, ten genotypes-five HOLGLS lines (480, 481, 520, 535, and 550), four LLmut&HOLGLS recombinants (440, 880, 888, and 899), and one HOmut&HOLGLS recombinant (873)-revealed no significant G × E interaction (Table 4), reflecting their stability.

ID
To visualize the 24 analyzed genotypes with respect to the composition of C18:1 and C18:3 fatty acids in seed oil regarding the G × E interaction for the resultant environments (B16, B17, L16, and L17), a principal component analysis of the interaction effects was performed and shown graphically as biplots (Figs 3 and 4). The approach allowed us to identify the genotypes that show similar response patterns across environments and to group the environments that induce similar responses. The distances between the positions of genotypes and the origin reflect their interaction with the environment (cf. significant G × E effects in Table 4); the longer the distance, the more specific reaction of a genotype to the environmental conditions. The proximity of genotype positions to the environment vector indicates an above-average performance of the genotype in this environment. For the genotypes that demonstrate significant G × E interactions (Table 4), the biplot analysis reveals which environments contribute to the non-standard genotype performance. Small differences in response patterns against the average can also be observed for the genotypes identified as 'stable'.

Discussion
Due to the application of canola oil as a raw material in various industries [1], multifaceted research and breeding has been performed at the PBAI-NRI Department of Genetics and Breeding of Oilseed Crops, Poznan, for further improvement of the oil quality to meet the demands of the plant oil crop market. One of the selection goals was to develop starting materials for the breeding of HOLL-type B. napus cultivars. Our first HO and LL mutant lines [31] were characterized by low agricultural value which was not only due to the toxic effects of the chemical mutagenesis (EMS treatment) but also a result of continuous inbreeding of canola over generations [18,51]. Therefore, domestic WOSR accessions of different genetic background and good agricultural value were included in our breeding programs using recurrent selection [44] and the obtained breeding lines were further crossed with HOLL mutant recombinants. As a result, several new recombinant lines characterized by the modified fatty acid composition in seed oil exhibited improved seed yield in field trials. As a spectacular example, the HO (over 79% of oleic acid in seed oil) cv. Polka was first assessed for variety value for cultivation and use (VCU) [52] and then added to the Research Centre for Cultivar Testing (COBORU) National List (NLT) of Agricultural Plant Varieties in 2018 (http://www.coboru.pl). Unfortunately, the selection procedure of the new recombinants with changed fatty acid composition in seed oil based on phenotypic analysis (gas liquid chromatography) was not efficient due to its high dependence on changing environmental conditions [52,53]. Therefore, for an effective selection of LL genotypes, application of allele-specific SNaPshot assay [40] was crucial.
The HOLL mutant recombinant line used in this work included a HO mutant line bearing one point mutation in the BnA.FAD2 desaturase gene [39]. The comparison of the sequences with the other published sequences of this gene [13,30] revealed that the described allele possesses a novel, previously uncharacterized point mutation G to A resulting in a STOP codon in the BnA.FAD2 gene localized on the chromosome A05 in the main functional locus described by other authors [13,30,54]. Our LL mutant genotype is characterized by two statistically important single nucleotide polymorphisms (SNPs): (1) a G to A transition in the 5'splice donor site of the sixth intron in the bnaC.fad3 gene disrupting the proper gene expression, and (2) a C to T substitution in the third position of the sixth codon of the seventh exon in the BnaA.fad3 gene, leading to amino acid substitution, which may influence protein folding decreasing its enzymatic activity [40].
Hu et al. [30] characterized the spring canola mutant HOLL DMS100 line. The authors mapped one major and one minor locus for BnaA.FAD2 genes on A05 and A01 B. napus chromosomes, respectively, as well as two major loci for BnaC.FAD3 genes on A04 and C05 chromosomes. In the bnaA.fad2 gene, a C to T single nucleotide substitution leading to the TAG stop codon in the bnaA.fad2 gene was identified. The mutation caused the premature termination of the open reading frame and was associated with high oleic acid content. In the bnaC. fad3 gene, a G to A transition in the 5' donor splice site of the sixth intron in the bnaC.fad3 gene was associated with LL genotype [30]. The point mutation was the same as in the LL mutant genotype in our previous study [40]. In the HOLL mutant variety released by SW Hickory, Yang et al. [13] identified one deleted mutation associated with high oleic acid content in the BnaA.FAD2.a locus which was not reported previously. At the same time, the authors identified two mutated FAD3 alleles for low linolenic acid content in SW Hickory in the BnaA.FAD3.b locus. The first one was a C to T substitution in exon 2 leading to a synonymous mutation, and the second was a C to T transition in exon 7, resulting in an amino acid substitution from arginine to cysteine. The latter was the same as identified previously in our mutant low-linolenic genotype [40]. Moreover, Hu et al. [30], Mikolajczyk et al. [40], and Yang et al. [13], independently identified one C to T point mutation in the 5' splice site of the sixth intron in the BnaC.FAD3.b locus.
Analysis of G × E interaction revealed different behavior of the analyzed genotypes in the four environments. In both growing seasons, during germination (August) and rosette formation (September-November) mean monthly temperatures were higher than the adequate multiyear means, whereas mean monthly rainfalls were significantly lower (except for Borowo in August 2015). As a consequence, notable differences in rosette development before winter were observed in each environment. During winter, temperature and precipitation were favorable for plant overwintering in each environment. During spring vegetation (March-April) both temperature and rainfall varied among environments (S1 Table), however, it did not influence the proper stem elongation nor inflorescence emergence. During flowering (May) in 2017 there were short periods of lack of rainfall (May 11 th -20 th ) in both locations (B17 and L17), but this did not significantly influence the length of flowering period, nor negatively influenced the analyzed phenotypic traits, as compared to B16 and L16 (S2 and S3 Tables). During development of siliques (June), there was enormously high rainfall in Borowo in 2016 (B16) leading to early stem lodging as well as the to the delayed development of siliques. It might have caused the lower mean seed yield in B16, as compared to the B17 (S4 Table). At the same time, it did not significantly influence the mean oil content as well as C18:1 and C18:3 content in seed oil (S4 Table). During seed maturing and seed fill period (July), there was high rainfall in both locations in 2016, unlike in 2017. However, it did not substantially influence mean values of oil content, nor C18:1 and C13:3 fatty acid content in seed oil between B16 and B17, and L16 and L17. At the same time, there were significant differences in C18:1 and C18:3 fatty acid content for certain genotypes regarding one location and different years (S3 Table).
Different G × E interaction patterns revealed by particular breeding lines may result from their origin and different genetic background of domestic accessions used for crossing in breeding trials. In this work, we were looking for stable HOLL-type canola genotypes revealing good agricultural value. This is why the LLmut&HOLGLS_880 line is of our special interest and will be included in further breeding programs involving development of high yielding F1 hybrid varieties with improved biotic stress resistance. Homozygotes selection with the use of functional markers specific for mutated alleles of the FAD3 desaturase genes was necessary to maintain the low linolenic acid content in seed oil. It is in accordance with our previous results on determination of fatty acid composition in canola seed oil by mutated alleles of the FAD3 genes in the A and C genomes of B. napus [55].
The development of HOLL-type breeding lines of B. napus remains one of the special selection goals worldwide [1], and our new HOLL recombinant genotypes make a valuable resource on the world oil crop market. It is noteworthy that two of the LLmut&HOLGLS recombinants, 880 and 882, were characterized by seed yield comparable to the reference cv. Monolit, in addition to a high percentage of wintering as well as high content of oleic acid and low content of linolenic acid, thus revealing their HOLL phenotype (Table 2). At the same time, 880 was stable in different environments for both fatty acid content traits, while 882 for C18:3 (Table 4). Moreover, according to the genotyping results these two recombinant lines revealed wild-type homozygous alleles of FAD2 desaturase gene involved in the synthesis of oleic acid and at the same time, mutant homozygous alleles of FAD3 desaturase genes involved in the synthesis of linolenic acid ( Table 2). The newly developed B. napus lines are genetic resources of seed oil with changed fatty acid composition and thus they can support the demands of the world oil crops market regarding the use of plant oils for deep frying and preparing healthy food as well as in biodiesel production.