Food Safety - The Oxford Topic of Food, Politics, and Society - page 598

Compositional analysis also provides a window into the presence of potentially
adverse unintended effects and the presence of new and unintended novel metabo-
lites. A major challenge in making such compositional comparisons is definition of the
normal range of metabolites for each crop. It has been observed that macronutrients
can vary by twofold or greater in different varieties of the same crop and micronutri-
ents and other minor components may vary by as much as 100-fold or more (Harrigan
et al., 2010; Harrigan and Chassy, 2012). Plant composition is highly variable (for fur-
ther references see in Herman, Chassy, and Parrott 2009; Herman and Price, 2013; and
Ricroch et al., 2011). To the question of how much difference in composition should trig-
ger further investigation, it has been suggested that differences ranging from 3 percent
to 20 percent in a component should be a trigger for further research. It is pointless to
pick an arbitrary percentage difference since the biological significance of a difference
is dependent on the role of that crop in the human diet, how much and how often it is
consumed, and what percentage of the required intake of that nutrient is derived from
that crop. A similar argument can be made with respect to antinutrients and toxicants;
the potential harmfulness depends directly on dose and exposure and not on percent
change. Risk assessors have established methods for making an analysis of the biological
significance of differences in composition.
The concern has been expressed that transgene insertion could activate a cryptic bio-
synthetic pathway or in some other way lead to the production of a novel toxic compound.
The unknown compound would not be analyzed since its presence could not be predicted.
This is a possible but highly unlikely scenario for several reasons. The targeted composi-
tional analysis that is applied to new transgenic varieties typically accounts for 85-99%
of the biomass, thus only a small percentage of the crop mass could be present as a novel
component. Based on the projected dietary consumption of the new variety the maximum
dose and exposure to a novel metabolite can be calculated (Chassy et al., 2004). Such cal-
culations reveal that a novel compound would need to be very highly toxic to have any
adverse effect (Chassy et al. 2004). Moreover, it is difficult to imagine such a scenario since
no novel toxic compound of this type has ever been identified in a crop plant either as a
spontaneous occurrence or as a result of breeding, including transgene insertion. There is
also no reason to believe that the chances of evolution of a toxic compound would be any
greater in transgenic breeding than it would be in conventional breeding.
Some have expressed the concern that novel toxic molecules could spontaneously
appear in transgenic crops. The production of novel toxic molecules requires the cre-
ation of new metabolic pathways for its biosynthesis. It is, however, very difficult for
organisms to evolve novel biosynthetic pathways that are encoded by multiple genes.
Evolving pathways using older mutagenic methods confronts the reality that great
majority of mutations are cryptic or cause loss of function. Although novel single genes
have appeared in crops, the spontaneous evolution of a novel multigene biosynthetic
pathway has not been reported in any crop plant. Plant breeders have had difficulty
evolving new pathways in order to produce desired metabolites, despite great effort.
One motivation for using genetic engineering is that the technique allows the introduc-
tion of multiple genes, opening up the possibility of pathway engineering.