of transgenic food crops for the consumer and the food industry in the Community
The main objective of the project was the analysis of the food quality
of transgenic crops and their derivatives. The two different aspects to
be analysed in detail were agronomic and processing quality and food safety.
Transgenic tomato was chosen as a model crop.
We were primarily interested in analysing the insect resistance trait
encoded by the Cry1Ab gene. However, as most, if not all, transgenic plants
will, in addition to the trait of interest, also contain and express selectable
marker genes which are necessary for the identification of transgenic
cells in vitro, we have also included in the study one of the most
frequently used examples of such genes, namely, neo (NPTII).
the right: tomatoes containing the
On the left: control tomatoes.
Approach and methodology
Production and analyses of transgenic tomatoes
Based on greenhouse evaluations, two transgenic lines were chosen for
the agronomic evaluation of the crop in the field. A field trial was done
during the 1993 season in Piana di Monte Verna (Caserta, Italy). The harvested
fruit was used for two purposes. Part was used for the preparation of
different types of processed products: peeled tomatoes, chopped tomatoes
and tomato concentrate. The remaining part of the harvest was concentrated
by freeze-drying and used in a feeding trial. The composition of the freshly
harvested tomatoes, including parameters important for processing such
as percentage dry matter, pH and sugar content, was determined. Expression
levels of Cry1Ab and neo were measured. Relevant chemical and physicochemical
data, as well as Cry1Ab and NPTII content, were also determined on the
The safety assessment focused on the analysis of proteins purified from
recombinant over-expressing micro-organisms. As Cry1Ab has known toxic
effects in insects, emphasis was placed on studying this protein. Less
exhaustive treatment of the marker protein was performed. A series of
in vitro characteristics of the proteins were measured and the
potential in vivo effects were estimated. This strategy was designed
to answer the following questions:
Do the newly introduced proteins cause general toxic effects in mammals?
To answer this question both in vitro and in vivo approaches
were designed. Degradation of the proteins was followed in the presence
of digestive enzymes (pepsin, trypsin and chymotrypsin) in simulated conditions
of the gastrointestinal tract. Digestibility of Cry1Ab has also been studied
in vivo through feeding of unique high doses of Cry1Ab protein.
General toxic effects were, in addition, analysed by short-term oral feeding
in rats and rabbits. Effects on blood cells and human cell lines were
Does the introduced Cry1Ab protein show specific toxic effects in mammals
similar to the effect observed in insects?
This analysis was based on the observation that the first step in the
interaction between Bacillus thuringiensis insecticidal crystal
proteins and the target insects consists in the interaction of the protein
with specific receptors. Different tissues of the gastrointestinal tract
of rodents and primates, including humans, were screened for the presence
of receptors to these toxins.
Does the genetic modification of the tomato result in changes in nutrient
content and the presence of naturally occurring toxins?
Lyophilised powders of tomatoes harvested were submitted to a detailed
analysis. Beside the content of the newly introduced protein, total protein
content, fat, carbohydrate, fibre, vitamin C and mineral content were
compared. The lyophilised tomato powders were also mixed with rat feed
at a concentration of 10 % and were used in a 90-day feeding trial.
Main findings and outcome
No differences have been found between the characteristics of both transformants
and their non-transgenic counterparts in fresh or processed products.
Both Cry1Ab and NPTII proteins are unstable in simulated gastrointestinal
conditions. They degrade into fragments with molecular weights below 10
kDaltons. The toxic effect of Cry1Ab on insects is highly specific and
no specific receptors were found along the gastrointestinal tract of mammals.
In addition, no histopathological effects were found upon ingestion of
Cry1Ab by mammals. Orally administrated Cry1Ab does not exert adverse
systemic effects in rats or mice. No indications of immunotoxic effects
were found upon histological examination of lymph nodes, spleen and Peyer's
patches of treated animals. In addition, no specific antibodies against
Cry1Ab could be detected, nor was there a general increase in total IgG.
No haemolytic potency of Cry1Ab was observed in in vitro experiments.
No major changes were observed in the chemical composition of transgenic
tomatoes as a result of the insertion of the novel genes. Transgenic tomato
powders mixed into the rat diet had no adverse effect on the development
of the animals. The average daily intake of tomato powders corresponded
to a daily human consumption of 13 kg fresh tomatoes. Food intake, body
and organ weights were normal and examination of the tissues showed no
indications of toxic effects.
It is strongly suggested that neither NPTII nor Cry1Ab presents major
risks to the consumer in food, even at elevated concentrations.
April 1991 - March 1994
Aventis CropScience N.V.
Follow-up of the project
This project was continued in EC project: AIR3-CT94-2311.
SME Ricerche S.c.p.A.
Piana di Monte Verna (IT)
State Institute for Quality Control of Agricultural Products (RIKILT)
Università degli Studi di Genova (IT)