Food Safety - The Oxford Topic of Food, Politics, and Society - page 595

will not be described here in detail (König et al., 2004; Delaney et al., 2008). Suffice it to
say that methods exist with which it is possible to determine if a protein will be digested
in the stomach or denatured by food processing—both processes that could lead to loss
of functional activity—or, by comparison, to databases if the protein bears any resem-
blance to known toxins or allegens. Composition studies, animal studies, and the evalu-
ation of protein toxicity or allergenicity are discussed in subsequent sections.
Unintended changes occur in all types of plant breeding (Cellini et al., 2004; Parrott,
2005; Weber et al., 2012). Insertion of DNA into chromosomes can theoretically interrupt
the function of genes, create new genes and/or pathways that encode novel products, acti-
vate cryptic pathways leading to the formation of potentially toxic or allergenic products,
or through direct or indirect effects alter gene expression and cell composition. The chal-
lenge for the risk assessor is that none of these unforeseen changes can be anticipated. The
strategy for identifying unintended effects is to screen carefully for new hazards, as, for
example, a meaningful change in composition or phenotype of the crop plant. The ultimate
challenge is to determine if any of the unknown changes that occurred in plant breeding
has created a potentially harmful risk exposure. Changes per se are not dangerous.
Skeptics and critics have advocated a highly precautionary approach to GM crops
approval. They argue, for example, that the inserted DNA may cause adverse effects such
as the spread of antibiotic resistance genes or that the insertion of DNA elements into the
human genome that could cause cancer, that the inserted proteins could be allergens, that
alternative RNA splicing or unintended DNA fragment insertions could produce novel
and potentially harmful proteins, and that new or cryptic pathways could be activated to
produce toxic molecules (Latham et al., 2006; Smith, 2007; Dona and Arvanitoyannis,
2009). These are exactly the same hazards that can be found on the risk assessors' hazard
scan. However, whereas skeptics worry about hypothetical hazards, risk assessors evaluate
laboratory-derived data to determine which, if any, of the potential hazards have material-
ized in a newly formed transgenic variety creating a real risk of harm upon consumption.
Critics also argue that not enough safety research has been done and that we do not
understand enough about the complex interactions and regulatory circuits that control
living cells (Latham et al., 2006; Smith, 2007; Dona and Arvanitoyannis, 2009). What
is typically not explained is why the same argument is not made against other modali-
ties of breeding that have as a goal the production of changes in the DNA that will give
rise to altered and/or novel phenotypes. Critics further assert, without providing direct
evidence, that transgene insertion is unnatural in that it does not occur in nature and
that it is fundamentally different than other modalities of breeding. However, hori-
zontal gene-transfer between sexually incompatible species has been demonstrated to
commonly occur in nature (Parrott, 2005).8 The incorporation of heterologous genes
in nature is mediated by the same systems that integrate transgenic DNA sequences in
the laboratory. When breeders make crosses between species, heterologous genes are
transferred and thousands of DNA insertions occur as whole chromosomes recom-
bine (Chrispeels and Sadava, 2003). Genomic DNA sequencing has revealed evidence
of numerous DNA insertions in many plant species. Insertions can be caused by DNA
transposition mediated by transposition and insertion sequences, as well as by muta-
tions that result in DNA duplications, inversions, and deletions (Weber et al., 2012).