Managing Allergenic Proteins
As a preliminary evaluation of the safety of transgenic plants, the verification of
substantial equivalence with the genetically unmodified counterpart is now
widely employed (Kuiper et al.
, 2001). Modern, transcriptomic, proteomic, and
metabolomic profiling techniques can be a vital part of such testing. Although
substantial equivalence measurements are not safety assessments in themselves,
they can reveal biochemical differences that can then be subjected to more
rigorous toxicological and immunologic testing.
A potential consequence of the genetic modification of crop plants is introduction
or creation of allergens. This could occur in several possible ways, including
introduction of unknown allergens with the transgenic protein itself, modification
by the host transgenic plant of the immunogenic properties of the transgenic
protein, modification of the immunogenicity of endogenous proteins in the transgenic
plant, and dissemination of an allergen through pollen that induces respiratory
sensitization (Moneret-Vautrin, 2002). Such risks need to be evaluated prior
to widespread use of a transgenic crop plant. Unfortunately, the possibility of
allergen induction can be exaggerated to the general public and used to fuel the idea that genetic modification is an unpredictable and irresponsible science. It is
true that the allergenicity of proteins, such as BNA, may not be widely known
before their introduction into a crop plant. However, the scientific community
quickly becomes aware of such potential problems (Nordlee et al.
, 1996) and acts
appropriately. For example, the transgenic soybean plants expressing BNA were
never commercially developed. As we gain a better understanding of the identity
and epitopic composition of common allergenic proteins, their selective modification
or elimination becomes feasible, and this could lead to the development of
Soybean consumption is a problem for some people and animals as it contains
several dominant allergenic proteins: Gly m Bd 68K, Gly m Bd 28K, and Gly m Bd
30K (P34) (Ogawa et al.
, 2000). The widespread use of soybean in the human foods
and animal feeds makes it an obvious target for genetic engineering to remove or
reduce these allergens. Gly m Bd 68K and Gly m Bd 28K are seed storage proteins,
and some reduction of their levels has been achieved through the development of
mutant lines (Ogawa et al.
, 2000). However, such a strategy has not been successful
with P34, which is an albumin and a member of the papain family of cysteine
proteases (Ogawa et al.
, 2000). Although this protein is a minor seed constituent, it
most dominant soybean allergen (Yaklich et al.
, 1999). While considered an
albumin, P34 partitions into oil body membranes during processing, as well as with
the globulin fraction (Kalinski et al.
, 1992). Consequently, it is almost impossible to
completely remove this protein from soybean isolates. Furthermore, its ubiquitous
presence in cultivated and wild soybean varieties suggests that it will not be possible
to reduce its level through conventional breeding (Yaklich et al.
, 1999). However,
through the sense expression of a Gly m Bd 30K cDNA, transgenic lines have been
developed in which the endogenous Gly m Bd 30K
gene is completely silenced
(Herman et al.
, 2003). A function for this protein has not been demonstrated but no
overt phenotypic change was observed in the gene-silenced plants. These transgenic
soybeans are currently being further evaluated in field trials (Herman et al.
Rice induces allergic reactions in some people and this is a growing problem in
some countries, like Japan (Watanabe, 1993). One of the major rice allergens was
identified as a 16-kDa albumin (Matsuda et al.
, 1988; Urisu et al.
, 1991). This
protein is encoded by a multigene family composed of at least ten members
(Tada et al.
, 2003), each of which has allergenic properties (Matsuda et al.
An antisense strategy was used to reduce the abundance of the 16-kDa albumin as
well as other gene family members (Tada et al.
, 1996). An 80% reduction in
abundance of the 16-kDa rice allergen was achieved (Tada et al.
, 1996), and the
reduction in protein levels of other family members was proportional to their
degree of nucleotide sequence identity with the transgene. Highly homologous
proteins were markedly lowered, and proteins with less identity were hardly
reduced at all (Tada et al.
, 2003). Highly immunogenic proteins need only be
present in minute quantities in order to elicit an immune response. Thus, these
results suggest that the antisense strategy may not be suitable for complete
removal of allergenic proteins, especially if they are encoded by divergent multigene
families. In this regard, gene silencing strategies (Waterhouse et al.
, 1998) that
require smaller regions of DNA sequence identity may prove to be more suitable.