GM Crop Database

The MS1 and RF2 canola lines (Brassica napus) were developed using genetic engineering techniques to provide a pollination control system for production of hybrid oilseed rape (MS1 x RF2). The novel hybridization system involves the use of two parental lines, a male sterile line MS1 and a fertility restorer line RF2. The transgenic MS1 plants do not produce viable pollen grains and cannot self-pollinate. In order to completely restore fertility in the hybrid progeny, line MS1 must be pollinated by a modified plant containing a fertility restorer gene, such as line RF2. The resultant F1 hybrid seed derived from cross between MS1 x RF2, produces hybrid plants that produce pollen and are completely fertile.

The transgenic line MS1 (B91-4) was produced by genetically engineering plants to be male sterile and tolerant to the herbicide glufosinate ammonium (as a selectable marker). The parental line MS1 contains the barnase gene for male sterility, isolated from Bacillus amyloliquefaciens, a common soil bacterium that occurs naturally in the soil and in various organisms including bacteria and plants, and frequently used as a source for industrial enzymes. The barnase gene encodes for a ribonuclease enzyme (RNAse) expressed only in the tapetum cells of the pollen sac during anther development. The RNAse affects RNA production, disrupting normal cell functioning and arresting early anther development, thus leading to male sterility.

The transgenic line RF2 (B94-2) was produced by genetically engineering plants to restore fertility in the hybrid line and to be tolerant to the herbicide glufosinate ammonium (as a selectable marker). Transgenic RF2 plants contain the barstar gene isolated from Bacillus amyloliquefaciens. The barstar gene codes for a ribonuclease inhibitor (barstar enzyme) expressed only in the tapetum cells of the pollen sac during anther development. The ribonuclease inhibitor (barstar enzyme) specifically inhibits barnase RNAse expressed by the MS1 line. Together, the RNAse and the ribonuclease inhibitor form a very stable one-to-one complex, in which the RNAse is inactivated. As a result, when pollen from the restorer line RF2 is crossed to the male sterile line MS1, the resultant progeny express the RNAse inhibitor in the tapetum cells of the anthers allowing hybrid plants to develop normal anthers and restore fertility.

Both transgenic canola lines MS1 and RF2 contain the bar gene that confers tolerance to the post-emergence, broad-spectrum phosphinothricin herbicides (Liberty®). The bar gene, isolated from the common soil microorganism Streptomyces hygroscopicus, encodes a phosphinothricin acetyl transferase (PAT) enzyme. The active ingredient in phosphinothricin herbicides is glufosinate ammonium which acts by inhibiting the plant enzyme glutamine synthetase, leading to the accumulation of phytotoxic levels of ammonia killing the plant within hours of application. PAT detoxifies glufosinate ammonium by acetylation into an inactive compound, eliminating its herbicidal activity. The herbicide tolerance trait was introduced into the canola lines as a selectable marker to identify transformed plants cells in the laboratory, and as a field selection method to obtain 100% hybrid seed.

The MS1 and RF2 canola lines were created by Agrobacterium-mediated transformation in which the transfer-DNA (T-DNA) contained either the barnase or barstar genes from the common soil bacterium, Bacillus amyloliquefaciens, under the control of the PTA29 anther-specific promoter from Nicotiana tabacum. In addition, each T-DNA contained a copy of the bar gene from S. hygroscopicus, which encodes the PAT enzyme, and sequences encoding the enzyme neomycin phosphotransferase II (NPTII) from the Tn5 transposon of Escherichia coli, strain K12. Expression of the bar gene was regulated by the PSsuAra promoter from Arabidopsis thaliana and post-translational targeting of the gene product to the chloroplast organelles was accomplished by fusion of the 5’-terminal coding sequence with the chloroplast transit peptide DNA sequence from A. thaliana. The expression of NPTII activity, under control of the nopaline synthase promoter from A. tumefaciens, was used as a selectable trait for screening transformed plants for the presence of the barnase and barstar genes, respectively.

The Introduced DNA
Southern Blot and segregation analyses demonstrated the stable integration of the inserted DNA at a single genetic locus in both MS1 and RF2. The barnase, bar and neo genes were integrated into MS1 and similarly, the barstar, bar and neo genes were integrated in RF2. Additional Souther blot analyses demonstrated that there was no incorporation of any plasmid sequences outside of the T-DNA region.

Genetic Stability of the Introduced Trait
MS1 and RF2 are several generations removed from the original transformants. Comparisons to the original transformants, in addition to data from several generations of backcrossing demonstrated the stable inheritance of the novel traits in lines MS1 and RF2. Segregation analysis confirmed the data from Southern blots and indicated that the novel traits in both transformation events segregated as a single locus according to the rules of Mendelian inheritance.

Expressed Material
Transcription of the bar gene was detected in both leaf and flower bud tissue by Northern blot analysis and was estimated to be 0.8-1.6 pg/mg total RNA and 0.1-0.2 pg/mg total RNA, respectively. Similar analyses performed on canola line RF2 demonstrated that the barstar gene was only transcribed in flower bud tissue at a level of 0.4-0.8 pg/mg total RNA. Both genes were expressed only during the early stages of anther development in the tapetum cell layer of the anther.

The NPTII encoding gene was linked to a weak constitutive promoter and expression of this enzyme was not detected in pollen samples or unprocessed honey.

Field Testing
The canola lines MS1 and RF2, and the hybrid MS1 x RF2 were field tested in Canada from 1991 to 1994. Agronomic and adaptation characteristics such as vegetative vigour, overwintering capacity, flowering period, seed production, seed dissemination, germination and establishment, and seed dormancy were within the normal range of expression of characteristics in unmodified counterparts. As expected, seed production of the hybrid line was greater than that of unmodified counterparts. Flowers of the MS1 line had undeveloped anthers, slightly smaller petals and did not produce fertile pollen, but nectar production remained unchanged and normal insect pollination was observed. Stress adaptation was evaluated, including its resistance or susceptibility to major B. napus pests and pathogens (e.g., blackleg, sclerotinia, flea beetles, diamondback moth larvae) and determined to fall within the ranges currently displayed by commercial varieties. The lines were tested under various environmental conditions, and showed no differences in agronomic performance when compared to unmodified counterparts under the same conditions. Overall the field data reports demonstrated that MS1 and RF2, and the hybrid MS1 x RF2 had no potential to pose a plant pest risk.

Outcrossing
The sterility of the B. napus line MS1 ensured that gene introgression from MS1 into wild or cultivated sexually-compatible plants was extremely unlikely. The MS1 plants can act as pollen recipients, but the progeny would be partially male sterile.

Transgenic line RF2 and hybrid plants displayed normal reproductive characteristics. Brassica napus plants are known to outcross up to 30% with other plants of the same species, and potentially with plants of related species B. rapa, B. juncea, B. carinata, B. nigra, Diplotaxis muralis, Raphanus raphanistrum, and Erucastrum gallicum. Previous studies have demonstrated that gene flow is most likely to occur with B. rapa.

The genes coding for male sterility or fertility restoration do not confer any ecological advantage to potential hybrid offspring of MS1 or RF2 plants. If glufosinate ammonium tolerant individuals arose through interspecific or intergeneric hybridization, the novel traits would confer no competitive advantage to these plants unless these populations were routinely subject to herbicide treatments. In the event that a glufosinate ammonium tolerant B. napus survived, these herbicide-tolerant individuals would be easily controlled using mechanical and other available chemical means. It was concluded that gene flow from the transgenic lines MS1 and RF2 or their hybrids to canola relatives was possible, but would not result in increased weediness or invasiveness of these relatives.

Weediness Potential
Field studies on invasiveness and survival characteristics comparing the MS1, RF2 and hybrid MS1xRF2 to unmodified plants determined that the transgenic lines were not different from their counterparts in this respect. Published data showed that seed survival of transgenic seeds expressing kanamycin resistance and glufosinate ammonium tolerance was significantly lower than seed survival of unmodified counterparts, when seeded under a variety of wild conditions. Competition studies, between transgenic lines and barley sown together at various ratios, demonstrated that the transgenic lines did not confer any novel competitive advantage.

Secondary and Non-Target Adverse Effects
It was determined that genetically modified lines MS1, RF2, and hybrid MS1 x RF2 did not have a significant adverse impact on nontarget organisms or organisms beneficial to plants or agriculture, and were not expected to impact on threatened or endangered species. The barnase and barstar proteins did not result in altered toxicity or allergenicity properties and are only produced in the tapetum cell layer of anthers at a specific developmental stage. In addition, detailed studies of pollination behavior in the field and in greenhouses showed no effects on honeybees. An avian dietary test was performed with the seed eating canary bird (Serinus canaria domestica) and a feeding study was performed with the domesticated rabbit (Oryctolagus cuniculus). Both studies determined no differences in food consumption, behavior and body weight between birds or rabbits fed with the transgenics or unmodified counterparts. Microorganisms living in the rhizosphere surrounding the roots were compared quantitatively and qualitatively and there were no differences observed between plots of transgenic plants versus plots containing unmodified B. napus.

Impact on Biodiversity
The transgenic lines MS1, RF2 and their hybrids have no novel phenotypic characteristics which would extend their use beyond the current geographic range of canola/rapeseed production. Since outcross species are only found in disturbed habitats, transfer of novel traits would not have an impact on unmanaged environments. Studies have shown that these lines are not invasive of natural habitats and that they are no more competitive than the unmodified counterparts, both in natural and managed ecosystems. It was determined that the relative impact on biodiversity of MS1, RF2 and MS1xRF2 was equivalent to that of currently commercialized canola lines.

Dietary Exposure
The human consumption of canola products is limited to the refined oil. Refined edible canola oil does not contain any detectable protein and consists of purified triglycerides (96-97%). As the introduced gene products are not detectable in the refined oil produced from transgenic canola, there will be no human exposure to these proteins based on normal consumption patterns.

Nutritional Data
The analysis of nutrients from transgenic canola lines MS1 and RF2 and non-transgenic canola did not reveal any significant differences in the levels crude protein, crude fat, crude fibre, ash and gross energy in either whole seed or processed meal. The fatty acid composition of oils extracted from both transgenic and non-transgenic canola was statistically the same and within the normal range for canola oil. The transgenic lines meet the standards for canola oil in Canada of containing less than 2% erucic acid and less than 30 micromoles/g glucosinolates in the oil-free meal. It was determined that the consumption of refined oil from MS1, RF2 or hybrids derived therefrom would have no significant impact on the nutritional quality of the food supply in Canada.

Toxicity and Allergenicity
Since only the processed oil from transgenic MS1, RF2, or hybrids derived therefrom (MS1xRF2), will be available for human consumption and the processing removes proteinaceous material, there are no additional toxicity or allergenicity concerns with this product. This was further demonstrated by examining the amino aid sequence homology and the characteristics of the novel proteins RNAse (barnase), RNAse inhibitor (barstar),PAT (bar) and neomycin phosphotransferase (NPTII).

Comparative analyses with amino acid sequences of known protein toxins and allergens in public domain databases did not reveal any homologies. The only sequence homologies were with other related proteins. For example, barnase encoding RNAse was similar to ribonucleases from other bacilli, PAT was similar to other phosphinothricin acetyltransferases originating from different organisms, and NPTII was found to be similar to other peptides that have a similar function (aminoglycoside 3'-phosphotransferases, kanamycin kinases, streptomycin 3'-kinases. No resemblance with potential toxins or allergens was observed.