FLAVONOIDS FOR ENVIRONMENTAL EQUIVALENCE PROFILING OF GE PLANTS

FLAVONOIDS FOR ENVIRONMENTAL EQUIVALENCE PROFILING OF GE PLANTS
Christof Sautter & Bartosz Urbaniak
December, 2007

Substantial equivalence (SE) is a parameter of biosafety concerning health and environmental interactions of genetically engineered plants (GEP), even though legally it is not a part of the deregulation procedure. SE in this context means that the expression profile, protein content, and metabolome of the GEP is exactly the same as the respective wild type (wt) plant, regardless of the desired effect of the transgene(s). Although SE might be an essential aspect of biosafety, it excludes a priori any undesired side effect on the health of a consumer, be it a human being or an animal, and moreover, any unintended interaction with the biotic or abiotic environment. Theoretically, SE should fulfil the strong criteria of even the over interpreted "precautionary principle."

In the European public debate about GEP, biosafety is a prominent topic. Highly sceptical parties and non-governmental organizations usually consider small differences between a GEP and its corresponding wt as a potential hazard, often regardless of the small scale of these differences and even lack of their significance. In none of these cases has the variation between conventionally bred varieties or the different environmental conditions been taken into account. However, scientifically sound risk assessments require such comparisons. Theoretically, the ability to judge a risk is limited when no direct effect is detectable. Instead, when assessing risks or indirect effects of a new technology, we can only compare them with risks associated with conventional systems, which are familiar to us. Only such a comparison makes rational decisions possible.

The main difficulty of SE is the uncountable number of compounds that constitute a plant and the impossibility of detecting them all. Therefore, recently published profiling studies focused on the parts of plant metabolism that the authors considered particularly important, such as the proteome or the metabolome^1,2,3. Our main goal was to study environmental interactions. Therefore, we choose the flavonoids, since these metabolites contribute to plant-environment interactions. These include defence against insects or microbes as well as abiotic stress reactions like photoprotection. Phenyl ammonium lyase and chalcone synthase are the enzymes catalyzing the first reactions of the flavonoid biosynthesis pathway. These are among the first enzymatic activities induced upon a microbe attack⁴.

Data
Varieties
For our studies we used three different Swiss spring wheat varieties: Golin, Greina, and Frisal. From all three varieties, genetically engineered lines were derived. Golin and Greina have been engineered with KP4, a viral protein that increases the resistance against seed transmitted fungal pathogens, known as smuts and bunts⁵. From the variety Frisal, we used two lines with a broad spectrum resistance against powdery mildew. The first one contained a chitinase and a glucanase from barley (line A5), the latter a barley ribosomal inhibiting protein (line B12)^6,7.

Grown under the same greenhouse conditions, the varieties exhibited significant differences in their flavonoid profile, whereas the differences between wt and genetically engineered lines were comparably negligible⁸. Principal component analysis revealed that the samples clustered according to the varieties. We therefore concluded that the differences between the conventional varieties are by far greater than the differences between wt and the respective genetically engineered lines⁸.

Environmental conditions
In addition to variety genotype, environment also influences the metabolic profile of plants. We compared the flavonoid content of these three varieties under different environmental conditions. We had three different climatic conditions: regular greenhouse, open field, and vegetation hall, i.e., a greenhouse that is open like a cabriolet as long as the weather is fine (Fig. 1a). At night and as soon as wind or rain arises, the roof closes automatically within two minutes (Fig 1b).

A comparison of the flavonoid content of the variety Greina under different environmental conditions is shown in Figure 2. Many of the 12 detectable flavonoids increase moderately from greenhouse to vegetation hall; the isoorientin, however, increases dramatically (Fig. 2, red). In Figure 3, we show the compiled results of three flavonoids (isoorientin, 8-C-pentosyl-6-C-hexosyl apigenin, and 6"-O-pentosyl isoorientin) in all three varieties in vegetation hall and field. The values are given in % against the greenhouse value, which is always 100%. This graph exhibits an 8-fold increase of isoorientin in the variety Golin between vegetation hall and field. In contrast, the variety Frisal increased the content of this compound only slightly (Fig. 3). This comparison of the effect of climatic conditions on the flavonoid content was done only with the wt varieties, since the GEP need a particular permission for field growing. We had this permission only for the year 2004 and only collected data from one of the KP4 GEP varieties⁹. We found no difference between GEP and wt in the field.

Implications
The molecular profiling data to date reveal dramatic differences in expression profiles, protein content, and metabolome, depending on the genetic background of the variety and environmental conditions. This is somehow expectable for a botanist, since plants have to deal with changing conditions at the place where they stay, due to their sessile lifestyle. Therefore plants have a rich variety of possibilities available to modify their proteome and metabolome, depending on environmental conditions. Furthermore, breeders have long selected for traits to meet particular desires and needs of farmers and consumers. This artificial selection exposes a huge variety of phenotypes behind the concept of a species that is morphologically stable due to natural selection. Most of the genetic basis for this variability was always present in the gene pool of the species. Consequently, the variation in the different genotypes of crop plants is not perceived as a risk by the public.

During artificial selection, breeders may remove toxins, bitterness, or otherwise undesired traits from crop varieties. Often the substances responsible for bad taste or toxicity also provide plant pathogen defence or pest resistance. For instance, alkaloid production is genetically suppressed in all crop varieties of potato tubers. As a consequence, farmers must increase care of potatoes in the field to compensate for the deficit of natural defences, often using externally added chemicals that act as fungicides or pesticides. Consequently, increasing resistance against pathogens and pests is an attractive pursuit for crop breeders. It would indeed be very advantageous if plants could be genetically engineered for improved fungal resistance, as they have been for pest resistance with the Bt system.

Several attempts to generate GEP expressing fungal resistance have been described in the literature (for a recent review see Collinge et al., 2007¹⁰). Fungal resistance has been attempted by engineering phytoalexin expression, as well as by production of direct-acting antifungal proteins. Examples are stilbene synthase, which increases the content of antifungal phytoalexin resveratrol¹¹; chitinase and glucanase, which degrade the fungal cell wall, with cell wall fragments acting as elicitors to trigger the defence reaction of the host plant^7,12,13; and the KP-system, providing a specific quantitative resistance against a seed-transmitted group of fungal diseases caused by members of the order Ustilaginales (smuts and bunts)^5,9.

In all publications in which SE studies in GEP were reported, including ours, the differences in the molecular profile between wt and GEP were by orders of magnitude smaller than the differences between varieties under the same environmental conditions or the differences between different environmental conditions in the same variety. In scientific risk assessments, one must compare the widely accepted risk of genetic variation between varieties or modifications of the metabolome profile under different environmental conditions with the differences between wt and GEP. The transgene itself and its phenotype is the only parameter that should be different between wt and GEP. The transgene-generated difference is the desired effect, which can be—and in most cases has been—studied exhaustively for biosafety issues. We therefore conclude that transgenes in GEP do not significantly affect SE or the molecular profile of the GEP, with the exception of the transgene itself and its phenotype, which is by far a smaller risk as compared to the effect of the genetic background of the variety or environmental effects.

We conclude that SE is a biosafety parameter that in certain cases might be of interest in a GEP, e.g., when a transgene interacts with a metabolic pathway such as the stilbene synthase pathway, but in general is not an issue to be more concerned about than in conventional breeding.
Outlook
Research in Europe with GEP is currently not very popular, resulting in strong legal regulation by over-interpretation of the precautionary principle. The bottleneck for research is field testing. The European regulations compel many biosafety measures in field testing that are far beyond scientifically justified considerations. SE is only one of a number of conceivable biosafety issues. Horizontal gene transfer of antibiotic resistance is another one, which led to a ban of antibiotic resistance genes by European legislation without sufficient scientific reason. Moreover, according to the Swiss regulations, a field test must also provide data about biosafety, even if this is not part of the project. This makes field testing for public research in Europe increasingly difficult. Progress in a research area like fungal resistance, however, depends on field experiments as the last step of proof of concept. We hope that our results on flavonoid profiling add to a facts-based public discussion in this field, and thus in the long run, convince the legislative body.
References

1. Catchpole GS, Beckmann M, Enot DP, Mondhe M, Zywicki B, Taylor J, Hardy N, Smith A, King RD, Kell DB, Fiehn O, Draper O (2005) Hierarchical metabolomics demonstrates substantial compositional similarity between genetically modified and conventional potato crops. Proc Natl Acad Sci USA 102, 14458-14462

2. Lehesranta SJ, Davies HV, Shepherd LV, Nunan NM, McNicol JW, Koistinen KM (2005) Comparison of tuber proteomes of potato varieties, landraces, and genetically modified lines. Plant Physiol 138, 1690-1699

3. Baudo MM, Lyons R, Powers S, Pastori G, Edwards K, Holdsworth MJ, Shewry P (2006) Transgenesis has less impact on the transcriptome of wheat than conventional breeding. Plant Biotechnology J. 4, 369-380

4. Cramer CL, Bell JN, Ryder TB, Bailey JA, Schuch W, Bolwell GP, Robbins MP, Dixon RA, Lamb CJ (1985) Co-ordinated synthesis of phytoalexin biosynthetic enzymes in biologically-stressed cells of bean (Phaseolus vulgaris L.). EMBO J 4(2), 285-289

5. Clausen M, Krauter R, Schachermayr G, Potrykus I, Sautter C (2000) Antifungal activity of a virally encoded gene in transgenic wheat. Nat Biotechnol 18, 446-449

6. Bieri S, Potrykus I, Fütterer J (2003) Effects of combined expression of antifungal barley seed proteins in transgenic wheat on powdery mildew infection. Mol Breed 11, 37-48

7. Bliffeld M, Mundy J, Potrykus I, Fütterer J (1999) Genetic engineering of wheat for increased resistance to powdery mildew disease. Theor Appl Genet 98, 1079-1086

8. Ioset JR, Urbaniak B, Ndjoko-Ioset K, Wirth J, Martin F, GruissemW, Hostettmann K, Sautter C (2007) Flavonoid profiling among wild type and related GM wheat varieties. Plant Molecular Biology 65 (5), 645-654

9. Schlaich T, Urbaniak BM, Malgras N, Ehler E, Birrer C, Meier L, Sautter C (2006) Increased field resistance to Tilletia caries provided by a specific anti-fungal virus gene in genetically engineered wheat. Plant Biotechnology Journal 4(1), 63-76

10. Collinge DB, Lund OS, Thordal-Christensen H (2007) What are the prospects for genetically engineered, disease resistant plants? Euop. J. Plant Pathol. 199, in press

11. Leckband G, Lörz H (1998) Transformation and expression of a stilbene synthase gene of Vitis vinifera L. in barley and wheat for increased fungal resistance. Theor. Appl. Genet. 96(8), 1004-1012

12. Anand A, Zhou T, Trick HN, Gill BS, Bockus WW, Muthukrishnan S (2003) Greenhouse and field testing of transgenic wheat plants stably expressing genes for thaumatin like protein, chitinase and glucanase against Fusarium graminearum. J. Experimental Botany 54, 1101-1111

13. Oldach KH, Becker D, Lörz H (2001) Heterologous expression of genes mediating enhanced fungal resistance in transgenic wheat. MPMI 14, 832-838.

Christof Sautter and Bartosz Urbaniak
Institute of Plant Science, ETHZ, Universitätsstr 2
8092 Zurich, Switzerland
csautter@ethz.ch