Community Research and Development Information Service - CORDIS

Final Report Summary - GMSAFOOD (Biomarkers for post market monitoring of short and long-term effects of genetically modified organisms (GMOs) on animal and human health)

Executive Summary:
The main objective of GMSAFOOD was to establish an approach to identify potential biomarkers for Genetically Modified Organisms (GMOs) after authorisation. The work was organised in 6 experimental and 1 analytic workpackages including:

Molecular analysis of the sugars attached to the alpha-amylase inhibitor (AAI) in transgenic peas, cowpeas, chickpeas as well as in several bean varieties (Australia)
AAI Pea and Bt-maize feeding studies in pigs (Ireland)
AAI Pea and Bt-maize feeding studies in salmon and zebrafish (Norway)
Pea and Bt-maize feeding studies in mice and human-severe combined immunodeficiency (SCID) mice (Austria)
Food chain studies where rats were fed pig meat and salmon raised on Bt-maize (Norway)
Epitope mapping and antibody determinations (Hungary)
Generation of a novel bioinformatics strategy for identifying biomarkers for post market monitoring of authorised GMOs

The objectives of the GMSAFOOD project achieved:

Molecular analysis of the sugars attached to the alpha-amylase inhibitor revealed significant variation in the glycosylation patterns in transgenic pea, cowpea, chickpea, Tendergreen and Pinto bean varieties demonstrating that no specific post translational modification can be attributed to transgenesis
AAI Pea and Bt-maize feeding studies in pigs revealed no adverse health effects in adults and offspring
AAI Pea and Bt-maize feeding studies salmon and zebrafish demonstrated, with few exceptions, that there were no adverse GMO health effects
AAI Pea and Bt-maize feeding studies in mouse models revealed no GMO-specific allergenic or adjuvant effects of amylase inhibitor peas
Food chain studies where rats were fed pig meat and fish from pigs and salmon which were raised on Bt-maize demonstrated minor but no adverse GMO health effects
Defined epitopes for AAI and Bt and AAI antibody determinations
Algorithmic strategy for the identification of potential predictive biomarkers for GMOs for post market monitoring
Generation of a GMO experimental database core for an open source Public Repository for GMOs for post market monitoring

The main project outcomes were presented at the GMSAFOOD GMO safety and post market monitoring conference in Vienna, March 2012. Detailed information on the project results and the teams are available on the GMSAFOOD website http://www.GMSAFOODproject.eu.

Project Context and Objectives:
Unintended consequences associated with the consumption of food containing GMO by a genetically varied population of humans and animals cannot adequately be evaluated during pre-market risk assessment. Moreover, monitoring of unexpected adverse health effects after a GMO is authorized is currently not done. Because of a lack of reporting systems and limited GMO content labeling, it is neither possible to quantify GMO exposure nor correlate exposure with disease. GMSAFOOD is dedicated to designing a biomarker strategy for post market monitoring Biomarkers are anatomical, physiological, biochemical, or molecular characteristics associated with an underlying physiological state, detectable and measurable by a variety of methods, and are associated with disease, exposure, response, and susceptibility. Our consortium established a bioinformatics strategy for the identification of biomarkers for authorised GMOs. We used a prototype alpha-amylase inhibitor pea created by CSIRO (Australia) and the authorised Bt-maize (Mon810) as GMOs. Data from feed studies with rats, mice, fish and pigs were entered into a newly established database containing raw data. These data from multiple species were then analysed using machine learning methodology for the identification of biomarkers associated with these GMOs. The algorithmic analysis derived predictive mathematical models for GMOs. The database with all feed study data may be converted to an open source database and an eventual public repository in which data can be accessed for further analysis and new data from public and private institutions can be added. This analytical approach and the availability of a broad public repository with datasets for authorised GMOs will increase the power to identify untoward GMO effects and related biomarkers.

The specific objectives of the GMSAFOOD project were to

1. To identify and qualify Biomarkers for an allergenic prototype GMO for use as a tool for post market monitoring of GMOs

2. To relate these Biomarker profiles to developmental stages including gestation, growth, maturation, and adult-life

3. To use Biomarkers to identify the movement and effects of GMOs in the food chain,

4. Establish Biomarkers for GMO-induced immunogenicity and allergenicity in animals and humans

5. To test Bt-maize as an authorised GMO to determine whether there was unexpected health affects in feed trials

Alpha-amylase inhibitor peas

Peas (Pisum sativum L.) expressing a gene, alpha-amylase inhibitor (AAI) from the common bean (Phaseolus vulgaris L. cv. Tendergreen) were generated at CSIRO in Australia and are able to protect peas from damage by inhibiting the alpha-amylase enzyme in the pea weevil (Bruchus pisorum). The pea weevil is one of the major pests of pea crops in Australia and the presence of the seed-specific gene in peas confers resistance to the pea weevil thus preventing damage to the pea seed. The AAI protein inactivates the pea weevilâ??s starch-degrading enzyme, alpha-amylase. The gene was reconstructed for high-level expression (4% of seed protein) specifically in seed and transferred to the pea variety called excell using antibiotic (kanamycin) selection in tissue culture. The transgenic pea line has been tested in the laboratory, greenhouse and field over several generations, and at many sites for efficacy against the insect pests. AAI pea is a good model system because 1) It is a seed-specific transgenic plant, 2) It is apparently allergenic in animal models (Prescott et al. 2005), and 3) The gene codes for a protein that, unlike Bt and other commercialised GM plants, is processed by the endomembrane system in the seed. The latter reason is important because the protein travels through the endoplasmic reticulum, the Golgi apparatus, and accumulates in the seed storage protein vacuoles. Thus, the combination of the protein being only found in the seed and a differential pattern of storage in seed vacuoles may result in the production of proteins that are handled differentially in the digestive tract and by the immune system of animals and humans consuming the GMO.
GMOs in Europe

The EU has a deficit in protein rich feedstuffs and this makes it the world's largest importer of soybean meal and second-largest importer of soybeans. Due to reforms in the Common Agricultural Policy (CAP) and in particular the reduction in subsidies paid to oilseed producers within the EU, its reliance on imported soybean meal has been accentuated further. The pig and poultry sectors in the EU are particularly reliant on imported soybean meal as quality protein source for diet formulation. New GM soybean and maize events are now planted in the Americas for which authorisation in the EU has not yet been sought. Based on previous experience with unauthorised GM events, such as Bt10 maize and LL rice that were discovered in EU member states it is likely that the same could happen with the as yet unauthorised soybean and maize. These new, unauthorised events limit the supply of maize and maize by-products to importers in the Europe. The effect is that the price of compound animal feed has and will increase. Peripheral countries such as Ireland and Norway will be most affected as most of maize and protein rich feed legumes are imported. Twenty six percent of all maize by-products entering the EU go to the island of Ireland. Peas used in this project are relevant for Europe. They are a valuable, alternative protein/nutrient source suitable for feeding monogastric animals and for human consumption. They have the added advantage that they are suitable for cultivation in many regions of Europe unlike soybean, the more traditional source of protein in non-ruminant diets, which must be imported into Europe. This has the potential to greatly insure Europe's supply of protein.

GMO concerns

The promise of foods based on GMOs has led to an increase in their cultivation. In the second generation of GM plants, there are approaches aimed at eliminating known allergens (e.g. in peanuts, soybeans and rice), provide nutrients, vitamins, etc. Moreover, GMO-based foods have also been proposed for the treatment of allergic disease and for vaccine production and application for protection against bacterial and viral infections as well as autoimmune disease. Although these crops may have considerable positive impact on the global economy, due to unidentified impacts on the economy and welfare of individuals in under-developed countries as well as on general health and well being, the risk to benefit ratio of GMO-based foods is not yet known. The increased usage of GMOs for foods for direct human consumption and for feeding to meat and milk producing animals has led to some public concerns. Together with the environmental risks associated with such crops, there is also increasing concern associated with the GM crops such as toxicity and the potential of the introduced proteins to elicit a potentially harmful immunological response including allergenic hypersensitivity.

Allergenicity of GMOs

Allergic responses to foreign ingested proteins are a concern that has led to increased awareness and testing for allergenic potential of proteins established in new hosts. The determination of allergenicity of GM proteins and of the whole GM plant and derived foods is done with an integrated, stepwise, case-by-case approach, which is continually assessed by a working group of the EFSA GMO panel. Allergenicity testing is generally evaluated in vitro using serum from individuals known to be allergic to the gene donor. An example of such testing led to the finding that a Brazil nut gene inserted into soybean caused allergic reactions in humans. Additionally, the assessment involves full characterisation of the introduced gene and expressed protein in the GM crop. The sequence of the protein is compared with known allergens and if homology with known allergens is detected, serum IgE assays (RAST, ELISA or Western blots), in vitro cell-based histamine release assays are done with donor sera and cells, in addition to skin-prick testing. Protein stability in pepsin is also tested as a predictor. These are considered useful tests but do not perfectly predict allergenicity with 100% certainty. For example, AAI is resistant to pepsin hydrolysis. Importantly, when no adverse health effects are revealed during testing for substantial equivalence to the non transgenic variety, the GM is authorised for the market and there is no subsequent government regulation.

Biomarkers

Biomarkers can indicate exposure, can indicate risk of adverse outcomes resulting from exposure, and can demonstrate risk of development of a disease in the context of exposure, if the disease is altered for better or worse, and can assess risk of one or another pathway of metabolism, etc. To the extent that they are correlated with outcomes, biomarkers all describe risk. Although no untoward effects were discovered in experiments from this project, a machine learning methodology would enable the identification of potential biomarkers associated with the potential adverse health effects related to GMO exposure. This broad approach will allow for the collection and analysis of vast quantities of data for biomarker identification.

Pre-market risk assessment and post market monitoring of GMOs

During pre-market risk assessment, GMO safety testing is focused on the risk that untoward responses will result from GMO exposure. It is partly based on a concept of substantial equivalence to unmodified organisms that was established in 1991 by the FAO-WHO. It includes acute toxicity testing with consumed doses 100 times the expected exposure of humans and/or repeated dose toxicity. FAO/WHO concluded that pre-market allergenicity assessment of the genetically modified food gives a satisfactory safety assurance, but it recognised that further evaluation for adverse effects of GMOs should be considered once the product has reached the market. The main concern of the pre-market risk assessment is that long-term or chronic untoward effects of GMOs may not be recognised until after the product has been on the market for a period of time and been used by many people. However, during pre-market risk assessment, human consumption is not allowed, which is where post market monitoring begins. The main objective of post market monitoring is the collection and analysis of data on a product after market release and the reporting of this information to identified stakeholders. Post market monitoring is not a substitute for pre-market safety assessment but complements it in order to confirm the pre-market risk assessment. Post market monitoring may reveal long-term effects GM food and feed on more diverse populations than those consuming the GMO during pre-market testing, such as individuals with underlying pathological abnormalities that increase the likelihood of immunogenic disease and/or cancer. This also includes the risk of health problems related to repeated, high-dose exposure, which cannot be fully controlled. An interesting example that illustrates this problem is Aspartame. It was subjected to controlled clinical studies and was considered safe before market release; the adverse health effects were recognised during post marketing surveillance on a small proportion of population unable to metabolise phenylalanine, a component of Aspartame.

The main objectives tackled and achieved by the GMSAFOOD consortium include:

Molecular analysis of the sugars attached to the alpha-amylase inhibitor (AAI) in transgenic peas, cowpeas, chickpeas as well as in several bean varieties (Australia)
AAI Pea and Bt-maize feeding studies in pigs (Ireland)
AAI Pea and Bt-maize feeding studies in salmon and zebrafish (Norway)
Pea and Bt-maize feeding studies in mice and human-severe combined immunodeficiency (SCID) mice (Austria)
Food chain studies where rats were fed pig meat and salmon raised on Bt-maize (Norway)
Epitope mapping and antibody determinations (Hungary)
Generation of a novel bioinformatics strategy for identifying biomarkers for post market monitoring of authorised GMOs

The main objectives of the project were achieved. The consortium has improved knowledge and provided an approach to address GMO effects on human and animal health after authorisation. Our improved techniques for post market monitoring of GMOs can be a significant support to policy needs of the European Food Safety Authority, the European Commission Health and Consumer Protection Directorate General, and the European Commission Environment Directorate General.

Project Results:
Increase knowledge on AAI and/or Bt-maize GM effect on adult mice, development of mice and on mice with pre-existing allergic disease

The gene for alpha-amylase inhibitor (AAI) from beans was transferred to peas for the protection of the peas against pea weevils. Importantly, this transgenic protein is found in the seed in high concentrations, unlike most authorised GMOs, and is synthesised on the endomembrane system, which leads to processing following translation In our studies, we evaluated the allergenicity and immunogenicity of AAI peas, cowpeas and chickpeas vs. their non-transgenic controls and bean AAI controls.

Hypothesis I. To address hypothesis I, AAI proteins expressed in AAI GMOs are differentially glycosylated and are therefore more immunogenic and allergenic than the bean AAI, we studied the in vivo immunogenicity of AAIs extracted from the three transgenic legumes and two different bean varieties. We immunised BALB/c mice with AAI isolated from Tendergreen and Pinto bean and cowpea, pea and chickpea and evaluated sera for anti-AAI antibodies to each purified protein. We found that immunising with AAI from beans and peas led to IgG1and IgE responses against AAI. The antibody titres varied but this variation was unrelated to the source of AAI. The bean, chickpea and cowpea versions generated high antibody responses, while the response to pea AAI was low. In contrast, the pea and Pinto bean AAIs used in the Prescott et al. (2005) J Agri Food Chem 53: 9023-9030 showed that the pea AAI generated an immune response but the bean AAI protein did not. We additionally found that chickpea AAI immunisation caused the highest IgE titres compared to the other AAIs. The differential antibody responses to different AAIs demonstrated altered immunogenicity but there was no correlation between antibody response and GM vs. non GM AAIs.

Hypothesis II. After having established that differences in the AAIs were not responsible for the initial findings in the Prescott et al. (2005) J Agri Food Chem 53: 9023-9030 study, we sought to address hypothesis II: AAI pea feeding induces allergy to the novel transgenic protein. We used the same lot of AAI and non transgenic pea and Pinto bean seed meals, in addition to newly generated transgenic cowpeas and chickpeas and also added Tendergreen bean as a control that was not tested in the original Prescott study. The experimental approach that we took was otherwise similar to the Prescott study. The seed meals were prepared as a suspension and fed to mice twice a week for 4 weeks. In some experiments, we fed the mice three times a week for 4 weeks. Additionally, we used cooked (100oC) and raw seed meal to determine the effect on heat treatment on the AAIs. Two days after the last feeding, the animals received purified intranasal AAI as a read out of an in vivo immune response.

Hypothesis III. To address hypothesis III, AAI peas and Bt-maize worsen allergies to other allergens, we immunised mice with the chicken egg protein, ovalbumin (OVA) to induce an acute onset of allergic asthma and followed that some months later with another challenge with OVA to induce a disease exacerbation. To test whether egg allergic mice were influenced by the feeding with AAI peas, we force fed mice twice a week for 4 weeks with the non transgenic peas, the transgenic peas, PBS and Tendergreen beans before the induction of disease and before the induction of exacerbation. Immunising and challenging mice with OVA generates an intense allergic response. The airways and lungs are inflamed and there is an increase in mucus production. The antibody responses to OVA were elevated at the onset of acute allergic disease. After the mice recuperated from the acute immune response, they were left with inflammatory infiltrates in their lungs that responded to exposure to OVA for the rest of their lifetime. A re-exposure with OVA induced an exacerbation of allergic disease. We tested the effect of eating GM and non GM peas during the initiation of allergic disease and during the recuperative phase, to determine whether there was an influence of the GMO on disease exacerbation.

Establish Hu-SCID chimera model for testing AAI pea effects

The gene for amylase inhibitor (AAI) from beans was transferred to peas for the protection of the peas against pea weevils. Importantly, this transgenic protein is found in the seed in high concentrations, unlike most authorised GMOs, and is synthesised on the endomembrane system, which leads to processing following translation. In our studies, we evaluated the allergenicity and immunogenicity of AAI peas, cowpeas and chickpeas vs. their non-transgenic controls and bean AAI controls in human severe combined immunodeficiency (SCID) chimera. The SCID mice have no B and T cells and are unable to generate immune responses. However, it is possible to inject immune cells from other animals and humans, which develop an immune system within the mouse. There is no consideration of MHC and anti-donor cell responses because there is no immune system. The cells can be maintained in the mouse. To generate an allergic patient-mouse chimera, we selected legume allergic patients as donors for the chimera. Our hypothesis is that the resulting mouse will contain the immune system of the allergic donor and will respond to allergens similarly. This provides a novel model for testing the ingestion of legume GMOs. In testing GMOs for allergenicity in humans, there is the matter of ethics. While there are numerous in vitro tests for allergenicity of proteins and animal models and animals (livestock) to test for untoward effects, it would be ideal to have a model that could reliably test the potential for allergenicity in humans. To address this issue, we established Hu-SCID chimera as a model for human allergy.

Bt-maize and AAI-pea fed fish

Maize and maize gluten are common sources of starch and protein, respectively, in commercial diets for fish, including Atlantic salmon (Salmo salar L.). The market availability of these products absent of GM maize is rapidly diminishing. Thus, the main aim of these studies was to evaluate the effects of whole meal Bt-maize expressing Cry1Ab protein as well as whole meal GM AAI peas in diets for Atlantic salmon at various stages of development. Both healthy and sensitised salmon were used in the trials when possible, the latter group to also assess effects in fish afflicted with an intestinal soybean meal-induced type IV hypersensitivity reaction. Responses in salmon were measured in the short (33 days) and medium (up to 99 days). A longer term cross-generational study in zebrafish (Danio rerio) as a model was also conducted with Bt-maize, assessing effects in both parents and offspring.

Bt-maize expressing Cry1Ab

In the first experiment, rapidly growing post-smolt salmon in seawater, approximately 100-125 g initial body weight, were fed fishmeal-based diets containing 1) 20% non-GM maize, 2) 20% Bt-maize, 3) 20% non-GM maize with 15% extracted soybean meal, or 4) 20% GM maize with 15% extracted soybean meal for 33 days and 97 days, triplicate tanks per diet group. The results indicated that salmon fed Bt-maize utilised feed less efficiently as revealed by lower protein (p=0.0456) and mineral (p=0.0261) digestibilities, lower whole body lipid deposition (p=0.0012), and lower lipid (p=0.0203) and energy (p=0.0550) retention efficiencies. These may be associated with decreased digestive enzyme leucine aminopeptidase specific activity (p=0.0368) and total enzymatic capacity (p=0.0724) in the proximal intestine and enhanced intestinal growth and immune responses, as indicated by a tendency toward transiently elevated proximal (p=0.0752) and total intestinal weights (p=0.0924) observed on day 33, increased distal intestinal interferon-gamma (IFNγ) mRNA expression in healthy fish (p>0.05) and transiently higher T helper cell marker CD4 protein level (p=0.0436) caused by feeding Bt-maize. This suggests that there may be metabolic costs associated with feeding salmon Bt-maize. The increases in CD4 protein and IFNγ mRNA expression in the distal intestine of Bt-maize-fed compared to normal fish, suggest that Cry1Ab protein or other antigen(s) produced due to the genetic modification have potential immunogenic effects in the gastrointestinal tracts in Atlantic salmon and may function as biomarkers for MON810 event maize exposure for this species. The immune responses may only be local, however, since no specific antibodies against Cry1Ab were detected in plasma. Bt-maize potentiated oxidative cellular stress in the distal intestine of fish afflicted with an intestinal hypersensitivity reaction as indicated by increases in superoxide dismutase (33 day; p=0.0271) and heat shock 70 (97 day; p=0.0366) mRNA expression.

AAI peas expressing alpha-amylase inhibitor

AAI peas expressing alpha-amylase inhibitor (AAI) and their isogenic non-GM counterparts were grown and harvested, and a feeding trial with salmon post-smolts was carried out during the fall of 2011. Duplicate tanks of fish were fed nonGM and AAI peas for 10 weeks, 20% inclusion levels in the diets. Analyses conducted indicate that the peas had no detrimental effects on growth or protein, amino acid or lipid digestibilities of the diets. The relative weight of the distal intestine of GM pea-fed salmon tended to be slightly increased (p=0.0753), but no other organ weights were significantly affected. Histomorphological evaluation of the various regions of the gastrointestinal tract, liver, spleen and head kidney with light microscopy did not reveal any differences between nonGM and AAI pea fed salmon. Digestive enzyme capacity in the distal intestine was significantly higher in the AAI pea fed groups (p=0.0127), possibly due to the higher distal intestinal tissue weight. No effects on plasma clinical chemistry were detected.

Food chain trials in rats

In order to assess introduction of GM crops in the food chain, a feeding trial with brown Norway rats (Rattus norvegicus) fed six diets containing: 1) non-GM maize, 2) Bt-maize, 3) filet from non-GM maize fed salmon, 4) filet from Bt-maize-fed salmon, 5) pork from non-GM maize fed pigs, and 6) pork from Bt-maize-fed pigs. Ten growing male rats per diet group were fed these diets for 90+ days immediately following weaning from their mothers. Body weight gain (p=0.0614) and whole body crude protein levels (p=0.0574) tended to be higher in GM-fed groups (direct and indirect GM consumption) than non-GM groups. Slightly higher feed intake (p=0.0525) in the rats fed Bt-maize and flesh from Bt-maize-fed salmon and pigs may help explain these tendencies. However, no significant differences in whole body crude lipid or dry matter contents were observed. No significant effects on nutrient digestibilities, fecal dry matter, intestinal digestive enzyme activities (leucine aminopeptidase), haematological parameters, organ (thymus, spleen, liver, kidney, jejunum, caecum and colon) PCNA, HSP70 or EMBP (eosinophil major basic protein) protein expressions, or most plasma clinical chemistry and organosomatic indices were detected. Tendencies of lower relative stomach (p=0.0690), colon (p=0.0765) and brain (p=0.0740) weights, and plasma amylase (p=0.0513) and triglycerides (p=0.0823) were observed in GM fed groups. On the other hand, relative heart (p=0.0336) and absolute testis (p=0.0183) weights and plasma total bilirubin concentrations (p=0.0774) were higher in GM fed groups, although differences were small. Urine analysis revealed a higher number of rats in the GM-fed groups that demonstrated the presence of low to moderate levels of erythrocytes (p=0.0297). Histomorphological evaluations of various organs did not reveal any gross pathological changes.

The effects of short-term feeding of Bt-maize to weanling pigs

Male weanling pigs (n=32) with a mean initial body weight of 7.5 kg and a mean weaning age of 28 d were used in a 31 d study to investigate the effects of feeding genetically modified (GM) Bt MON810) maize on growth performance, intestinal histology, immune response, intestinal microbiota and organ weight and function. At weaning, pigs were fed a non-GM starter diet during a 6 day acclimatization period. Pigs were then blocked by weight and litter ancestry and assigned to diets containing 38.9% GM (Bt MON810) or non-GM isogenic parent line maize for 31 days. Body weight and feed disappearance were recorded on a weekly basis (n=16/treatment) and pigs (n=10/treatment) were sacrificed on d 31 for collection of organ, tissue and blood samples. Effects on the porcine intestinal microbiota were assessed through culture-dependent and -independent approaches.

The effects of medium-term feeding of Bt-maize to pigs

Seventy-two male weaned pigs were used in a 110 day study to investigate the effect of feeding genetically modified (GM) Bt MON810 maize on selected growth, intestinal microbiota, immune response and health indicators. Following a 12 day basal period, pigs (10.7 -1.9 kg body weight; approximately 40 days old) were blocked by weight and ancestry and randomly assigned to the following treatments: 1) non-GM maize diet for 110 days (non-GM), 2) GM maize diet for 110 days (GM), 3) non-GM maize diet for 30 days followed by GM maize diet to day 110 (non-GM/GM) and 4) GM maize diet for 30 days followed by non-GM maize diet to day 110 (GM/non-GM). Body weight and daily feed intake were recorded on days 0, 30, 60 and 110 (n=15). Following slaughter on day 110, organs and intestines were weighed and sampled for histological analysis and urine was collected for biochemical analysis (n=10). Serum biochemistry analysis was performed on days 0, 30, 60, 100 and 110. Using culture-based methods, Enterobacteriaceae, Lactobacillus and total anaerobes were enumerated in the faeces on days 0, 30, 60 and 100 of the study and in ileal and cecal digesta on day 110. Cecal microbial communities were also analysed by high throughput 16S rRNA gene sequencing on day 110.

In conclusion, medium-term feeding of GM maize to pigs did not adversely affect growth rate or the selected health indicators investigated. Feeding pigs a Bt-maize-based diet for 110 days did not affect counts of any of the culturable bacteria enumerated in the faeces, ileum or cecum. Neither did it influence the composition of the cecal microbiota with the exception of a minor increase in the genus Holdemania. As the role of Holdemania in the intestine is still under investigation and no health abnormalities were observed, this difference is not likely to be of clinical significance. These results indicate that feeding of Bt-maize to pigs in the context of its influence on porcine intestinal microbiota is safe. Perturbations in peripheral immune response were thought not to be age-specific and were not indicative of Th2 type allergenic or Th1 type inflammatory responses.

The effects of long-term feeding of Bt-maize to pigs

Twenty-four sows and their offspring were used to investigate the effect of feeding genetically modified (GM) Bt MON810 maize to sows during gestation and lactation and to their offspring from weaning for 115 days on offspring growth and health. Sows were fed diets containing GM or non-GM maize from service to the end of lactation. Following weaning at approximately 28 d of age, mixed sex offspring from sows fed non-GM or GM maize diets were blocked by body weight, sex, sow treatment and litter ancestry and randomly assigned to diets containing GM or non-GM maize to give four treatment groups; 1) non-GM maize-fed sow/non-GM maize-fed offspring (non-GM/non-GM); 2) non-GM maize-fed sow/GM maize-fed offspring (non-GM/GM); 3) GM maize-fed sow/non-GM maize-fed offspring (GM/non-GM); 4) GM maize-fed sow/GM maize fed offspring (GM/GM). Growth performance was measured (n=15/treatment) and blood samples were taken for biochemical analysis on d 0, 30, 70, 100 and 115 of the study (n=10/treatment). Pigs were slaughtered on d 115 (n=10/treatment) and carcass weight, dressing percentage and back-fat depth were measured. Organs (heart, kidney, spleen, liver) were also weighed at this time. Kidney, liver, mesenteric lymph nodes and small intestine were collected for histological analysis.

Increase knowledge on the food chain by establishing effects of feeding Bt-maize-fed fish and pork to rats

A feeding trial with brown Norway rats (Rattus norvegicus) fed six diets containing: 1) filet from non-GM maize fed salmon, 2) filet from Bt-maize-fed salmon, 3) pork from non-GM maize fed pigs, 4) pork from Bt-maize-fed pigs, 5) non-GM maize, and 6) Bt-maize. Ten growing male rats per diet group were fed these diets for 90+ days immediately following weaning from their mothers. Final body weights, growth and feed intake tended to be higher in GM-fed groups (direct and indirect GM consumption) than non-GM. Whole body crude protein levels were slightly higher in the rats fed GM maize and flesh from GM maize-fed salmon and pigs, which may help explain the higher body weight gain observed in these groups. However, no significant differences in whole body crude lipid or dry matter contents were observed. No significant affects on nutrient digestibility, fecal dry matter, intestinal digestive enzyme activities (leucine aminopeptidase) haematological parameters, organ (thymus, spleen, liver, kidney, jejunum, caecum and colon) PCNA, HSP70 or EMBP (eosinophil major basic protein) protein expressions, or most plasma clinical chemistry or organosomatic indices were detected. Tendencies of lower relative stomach, colon and brain weights, and plasma amylase and triglycerides were observed in GM fed groups. On the other hand, heart and testis weights and plasma total bilirubin concentrations were significantly higher in GM fed groups, although differences were small. Urine analysis revealed a higher number of rats in the GM-fed groups that demonstrated the presence of low to moderate levels of erythrocytes. Histomorphological evaluations of various organs revealed longer gastric pits and a tendency of taller absorptive colonocytes in GM fed groups. No GM effects were detected in the number of rats with histomorphological signs of focal glomerulonephrosis or myocarditis.

A main research objective was to determine the fate of ingested recombinant DNA (cry1Ab gene) and protein (Cry 1Ab) in gastrointestinal digesta, and assess their potential for translocation to blood and organ tissues and evaluate the antibody responsiveness to transgenic protein in sera of laboratory and farm animals exposed short, medium, and long-term to Bt-maize (MON 810) diets in comparison with animals fed non-GM diets. Another objective was focused on determination of the fate of ingested recombinant protein (AAI) in gastrointestinal digesta, and assess their potential for translocation to blood and organ tissues and evaluate the antibody responsiveness to transgenic protein in sera of laboratory and in pigs exposed for short-term thus allowing a clearer assessment of the safety of Bt-maize. As AAI peas were considered allergenic (Prescott et al. (2005) J Agri Food Chem 53: 9023â??9030), we focused on the evaluation of IgE reactive epitopes in legume allergic humans with AAI sensitivity, and compared these data with IgE reactivity in Hu-SCID mice generated from these patients following AAI pea consumption.

In short, medium and long-term experiments, pigs fed Bt-maize had transgenic DNA and protein only in the gastrointestinal digesta. Neither DNA or protein were found in the kidneys, liver, spleen, muscle, heart or blood. A trans-generational study also found that the offspring piglets had no transgenic DNA and protein in their organs and blood. The transgenic DNA was only observed in the gastrointestinal digesta when feeding Bt-maize to sows during gestation and lactation. Bt-maize consumption did not induce Bt-specific IgG and IgA serum antibodies. In a short-term AAI pea feeding trial, AAI DNA and protein was limited to the gastrointestinal digesta and was not found in the kidneys, liver, spleen, muscle, heart or blood. AAI pea consumption did not induce AAI-specific IgG and IgA antibodies in pigs.

Patients with a history of gastro-intestinal, skin and respiratory symptoms in response to legumes were compared with healthy controls without allergy. Patients included in the study had clinical symptoms after eating legume diet and positive allergen specific IgE-RAST or Prick to Prick test, and severe clinical symptoms after elimination diet and repeated legume food challenges. Patient sera were shown to contain antibodies against AAI from pea and beans. AAI was identified on SDS PAGE gels and could separate allergic patients from healthy controls on western blots. To exclude carbohydrate epitopes were destroyed in the highly glycosylated AAI proteins. The selected sera recognised 3-10 polypeptides in 2-DE separated, native bean and transgenic pea meals but were unable to identify IgE reactive epitopes in the cooked seed meals. These data were comparable with literature data.

Machine Learning Analysis and database generation

The next step is validation: Only parameters with high matching factors are used for validation. To validate the results of clustering, we use the MLP neural network method. This method knows which food type each mouse received (supervised learning). The goal of this method is to validate the results of clustering methods. In other words, the neural network should reproduce the results from the clustering method. If the network, which knows which mouse ate which food, produces the same result as with the clustering method, then the plausibility value is high (near 1). If the network derives different results than the clustering method, then the plausibility is low. Therefore, the plausibility value is a measure how well a classifier based on a specific parameter would work. Importantly, the more samples tested / parameter, the better the results, much the same as it is with statistical analysis.

GMO Database Public experimental GMO data repository

The GMSAFOOD has designed and created a database with raw and machine learning datasets. The intention of this core database is to eventually establish an international centralised repository of experimental GMO data. It would be available as an open source web application or potentially integrated into cloud-based applications. The idea would be to provide GMO data available via portals, metadata, and search tools for use by researchers, citizens, government, regulators, risk assessors, voluntary organisations, and the private sector.

A public repository could be established to provide access to scientific experimental data to forge new collaborations, technology development, new probability of early identification of adverse health effects of specific GMOs, and improved biomarker discovery. GMO safety, risk assessment and post market monitoring could be facilitated by a centralised collection of datasets further data mining and analytic methods like other databases, e.g. Human Genome project that enable anyone to easily locate, download, customise, and analyse data. We could provide researchers across disciplines and industries with tools to enable early detection of GMO related untoward effects.

The aims of such a public repository of experimental GMO data are:

To support innovation by improving access to knowledge resources for the private and public sector by reducing duplication and promoting the availability of existing resources
To advance research, education, and training, with the ultimate goal of improving human and animal health
To provide open and unrestricted access to scientific data for statistical, algorithmic and scientific use to capitalise on its use and value
To support research by providing evidence-based primary research internationally to academia, the public sector, and industry-based research communities
To provide a centralised place for the addition of new data enabling others to reuse existing data to reduce the time necessary to fully evaluate GMOs after authorisation
To provide data necessary for informed evidence-based analysis and decisions for regulators and policymakers and to help form the basis of new legislation
To provide a comprehensive resource to which data can be contributed by government agencies, academic institutions, non-government organisations, and private industry
Potential Impact:
GMSAFOOD has made an impact in the area of GMO safety and post market monitoring. We improved the knowledge of GMO effects on health for a prototype GMO, the alpha-amylase inhibitor (AAI) pea and Bt-maize (Mon810). These data have and will soon be published in peer-reviewed journals, which adds to scientific knowledge of GMO safety. Additionally, these results have made their way to the public and to the European Food Safety Authority (EFSA), the European Commission Health and Consumer Protection Directorate General, and the European Commission Environment Directorate General. Our findings that there are no adverse health effects of Bt-maize in farmed salmon and pigs will impact the legislation for Bt-maize (Mon810) in Europe. Additionally, the EFSA GMO risk assessment network is aware that GMSAFOOD project data refute the earlier Prescott et al. (2005) J Agri Food Chem 53: 9023 study by showing that not only are AAI peas not allergenic, but neither are AAI cowpeas and AAI chickpeas. Additionally, we have widely disseminated information about our novel strategy for post market monitoring of GMOs using bioinformatics methodology and our aim to use the database generated in the project as a core for an open access Experimental GMO Data Public Repository for post market monitoring, which would directly support the policy needs of the main regulatory and legislative bodies in Europe and elsewhere.

GMSAFOOD impact on improving knowledge

Impact on the alpha-amylase inhibitor protein studies

A detailed picture of the glycans attached to AAI from three different transgenic legumes as well as several non-transgenic bean varieties has emerged from this study. The composition of the glycans was determined together with the quantification of the different types of glycan for each of the AAI subunits. Although there were seven different glycan families, they were all present in the AAIs of each of the transgenic and non-transgenic legumes. There was variation in the levels of the different glycans on AAI in each of the transgenic legumes. Furthermore, there was variation in the levels of the different glycans on AAI in each of the non-transgenic bean varieties. The variation in the glycan levels in the transgenic AAIs overlapped with the variation in the glycan levels in the non-transgenic AAIs. This result led to the conclusion that the two types of AAIs (transgenic and non-transgenic) were essentially equivalent to each other.

The scientific implication of this research is that any N-glycosylated protein that is expressed in a transgenic host is likely to have small but discernible changes in the levels of each of the glycans attached to that protein. This information is useful for regulators who are analyzing the differences between transgenic and non-transgenic versions of a protein with a view to ascertaining the consequences of such small variations. The information is also useful to any entity proposing to produce glycoprotein therapeutics such as antibodies in any part of any plant.

Impact for animal production and animal feed industry in the EU

2012 marks the 16th anniversary of the commercialisation of GM crops. During these 16 years, the cultivation area of GM crops worldwide has increased 87-fold and reached 148 million hectares in 2010. GM crops/feed are subject to rigorous risk assessment before they can be authorised for use within the EU. Even though many EU countries do not cultivate GM crops, authorised GM feed ingredients are imported into these countries and used for animal feeding. Public opinion regarding GM technology in the EU is somewhat divided. On one hand, GM technology has enormous potential to stabilise world food supply and price into the future by offering resistance to environmental stresses, such as disease and weather conditions. However, some consumers are concerned about potential risks associated with this technology. Genetic engineering is a tool employed by plant breeders, which allows faster genetic improvement than achievable with traditional plant breeding technologies. It is mainly used to confer herbicide resistance or insect resistance or both to a crop and the technology has been widely adopted by the main grain exporting countries. The European animal feed industry is highly reliant on imported feed ingredients, particularly soya and maize by-products as a source of protein. The unprecedented growth in the use of GM technology in the grain producing areas of North and South America has been such that it is now difficult to source nonGM alternatives.

Impact of Bt-maize (MON810) and AAI peas in pig feed trials

We studied effects of feeding GM maize to pigs during short-, medium- and long-term trials. Experiments were conducted using pigs to determine the growth, health and immunological consequences of feeding Bt (MON810) maize. The effects of feeding AAI peas to weaned pigs were also evaluated in a 30-day study. These trials were used to investigate the effects of Bt-maize on growth performance, intestinal histology, immune response, intestinal microbiology and organ weight and function. Analysis was also performed to determine if the gene encoding the toxin responsible for the genetic modification of the maize or the toxin itself moved from the animal's digestive tract into blood or organs. The main results of this work can be summarised in the following bullet points:

Growth rate was not adversely affected when pigs were fed Bt-maize
Organ (liver, heart, kidneys and spleen) structure and function were not adversely affected by feeding Bt-maize
No major changes to Intestinal structure and function were observed and small changes observed did not occur consistently and are not believed to be of biological importance
Some minor alterations in immune response occurred following Bt-maize exposure, however, it is believed that similar differences might be observed if two varieties of conventional maize were compared. The results do tell us however, that there was no allergic-type immune response to the Bt toxin
We found that feeding Bt-maize to pigs resulted in only minimal impact on microbial community structure in the caeca of pigs. Bacteria within the digestive systems of pigs fed Bt-maize appear to tolerate well the Bt toxin
The Bt toxin or gene did not migrate from the digestive system of the animal into other body organs
The gene encoding the Bt toxin was broken down as it moved through the digestive system with DNA from the Bt toxin found in the stomach contents of pigs fed Bt-maize but not in the colon

Impact on tracking GMO proteins and transgenes in animals

The tracking of proteins and transgenes in vivo have yielded completely different results from published data derived from in vitro digestibility tests. The impact of these findings is Europe-wide because risk assessors will now need to consider all related information along with published data in relation to in vitro digestibility in their assessment of allergenic risk of GMO containing food and feed.

Tracking antibody responses to GMO proteins highlights the importance of sound methodology for the assessment of peptide-specific IgE recognition in glycosylated protein for allergenicity and the evaluation of the impact of processing in GMO risk assessment.

Impact of the mouse models for testing GMO safety and allergenicity

The impact of the GMSAFOOD results is clarity for scientists, regulators and the public regarding the use of mouse models for the assessment of GMO safety and allergenicity. In 2005, Prescott et al. published studies showing that the AAI peas were allergenic while the bean was not (J Agri Food Chem 53: 9023). This caused reactions from the public about GMOs being unsafe and allergenic. Regulators and policy makers made decisions affecting GMO legislation and changes in the regulatory process. The GMSAFOOD project results refute this study. We found that AAI peas as well as AAI cowpeas and AAI chickpeas are not specifically allergenic in mouse models. Taken together, the disparate results between these two studies underscore the importance of doing repeat experiments in independent laboratories to confirm findings. This is especially important before definitive risk assessment and the consequent regulatory and legislative decisions are made. These findings have significant impact because scientists, regulators, policy makers and the public have to consider that the original data were misleading and re-consider the approach to GMO risk assessment.

Impact of the Bt-maize and AAI pea feed trials in farmed Atlantic salmon

The feeding trials conducted and the data collected from the feeding trials with Bt-maize for Atlantic salmon and zebrafish and AAI peas for salmon contribute to the body of knowledge available on effects of GM feed ingredients in formulated diets for cultured fish. Bt-maize and AAI peas may have local (intestinal) effects in the form of mild immune responses and/or enhanced tissue growth. This may somewhat affect intestinal function and ultimately have a small metabolic cost. However, the changes were small and systemic health was apparently not detrimentally affected. A cross-generational study conducted in zebrafish as a model organism indicated that feeding Bt-maize to the parents had no adverse health effects for the offspring. Thus, long-term health implications of Bt-maize or AAI peas in fish feeds, also across generations, are not predicted.

Impact of the Bt-maize and food chain studies in experimental rats

The food chain study conducted by GMSAFOOD using rats as a model consumer, appear to be among the first controlled feeding trials of this kind. The data indicate that some factors in GM feed ingredients fed to production animals may be transferred to consumers via their animal products. However, although some differences in responses were observed between nonGM and GM-related foods, these were small and were well within the normal physiological ranges of the parameters. Furthermore, differences observed between the main ingredients maize, salmon filet and pork were greater and more numerous than those registered between nonGM and GM-related groups. Thus, long-term health implications of Bt-maize in the food chain are not predicted.

Impact of bioinformatics on GMO research and GMO public data repository

Currently, no adequate system is in place for monitoring the effects of GMO consumption on animal and human health once a crop has been approved for the market. The main emphasis is placed almost entirely on pre-market testing. Given the difficulties and expense of epidemiological approaches for post market surveillance, we propose a novel strategy in which experimental data undergo machine learning clustering and neural network analyses to identify potential biomarkers capable of detecting unforeseen health risks related to GMOs. A bioinformatics strategy is advantageous because analyses for the identification of biomarkers can be applied to individual and disparate experiments and can even analyse data across species. GMSAFOOD has submitted all experimental data in the form of a core database for a Public Data Repository containing data on GMOs. We envisage a Public Experimental GMO Data Repository for researchers from private and public sectors worldwide. An open access web-based GMO Database for post market monitoring would complement pre-market risk assessment and offer invaluable opportunities for early detection of adverse health effects related to human and animal GMO consumption and their associated biomarkers for post market monitoring. The impact of delivering a database and strategy for a public repository that could include data from public and private labs for combined analysis to identify potential biomarkers is a great advance in the current approaches and could potentially improve the regulatory process and public safety. Evidence-based science is the only way to truly address whether GMOs are safe for animal and human consumption. For those regulators and members of the EC and EFSA who publish guidelines for management of GMOs and assess GMO risk, it is crucial that they have the necessary tools for informed decisions. The impact of the results from GMSAFOOD is that they will provide evidence that will be utilised in risk assessment analysis, subsequent governmental guidelines and eventual legislation.

Impact on policy and post market monitoring

Ensuring safe food for EU citizens is a main concern of the European Union. Such public concerns have translated into key objectives for the European research community in the agriculture sciences, as well as in EU policy and consumer associations. Biomedical and agricultural technology has made major contributions to novel GM crops that have the potential to reduce allergy, to prevent crop devastation, to vaccinate against a myriad of diseases within the last 20 years. The impact of novel GM crops will be diminished if there are no strategies in place for prevention and early detection of untoward health effects of GMOs on animals and humans.

Today's regulations regarding post market monitoring of effects of GMOs are rather general. Further knowledge on suitable biomarkers may allow a higher degree of specifications and aid in development of robust post market monitoring. A strengthened basis for developing biomarker protocols will support the policy needs of the European Food Safety Authority, the European Commission Health and Consumer Protection Directorate General, and the European Commission Environment Directorate General.

According to European regulations following the market release of a GMO, the notifier, under Article 20(1) of the Directive on the deliberate release into the environment of genetically modified organisms http://www.biosafety.be/GB/Dir.Eur.GB/Del.Rel./2001_18/2001_18_TC.html, has a legal obligation to ensure that monitoring and reporting are carried out. The Council Decision 2002/811/EC (EC, 2002b) supplements Annex VII (Monitoring plan) to Directive 2001/18/EC.

The objectives of post market monitoring, as detailed under Annex VII, are to:
confirm that any assumptions regarding the occurrence and impact of potential adverse effects of the GMO or its use in the environmental risk assessment are correct, and identify the occurrence of adverse effects of the GMO or its use on human health or the environment, which were not anticipated in the environmental risk assessment.

Monitoring should not be regarded as research per se but as a means to evaluate or verify results and assumptions arising from previous research and evaluation of potential risk and research. Against this background, it is essential to identify the type of effects or variables to be monitoring and importantly, the tools and systems to measure them and an appropriate time-period for measurements. Monitoring results may be important in the development of further research. Effective monitoring and general surveillance requires that appropriate methodology has been developed and is available prior to the commencement of monitoring programmes.

As part of the authorization process for a new GM containing food, EFSA's recommendation may include restrictions or conditions such as post market monitoring in response to results of the safety assessment. According to EFSA, post market monitoring is not a substitute for pre-market risk assessment and should be considered on a case-by-case basis. EFSA considers it as a later step used to confirm pre-market risk assessment with the aim to identify rare untoward health effects.

GMSAFOOD project has developed a strategy for early identification of suitable biomarkers for use in post market monitoring for GMOs. There are few reliable biomarkers available for the detection of harmful effects of GM foods during post market monitoring. Currently, the knowledge gained through post market monitoring might at best describe broad patterns following human nutritional exposure EFSA. Thus, in general, it cannot be relied upon as a technique for monitoring adverse health events or other health outcomes related to the consumption of GM plant derived foods. EFSA considers that specific hypothesis-driven studies may be required to relate adverse health effects to the consumption of these foods.

A public repository for GMO data has the follow impacts on:

Innovation by improving access to knowledge resources for the private and public sector by reducing duplication and promoting the availability of existing resources
Advancing research, education, and training
Data access to scientific data for statistical, algorithmic and scientific use to capitalize on its use and value.
Research by providing evidence-based primary research internationally to academia, the public sector, and industry-based research communities
A the initiation of a centralized repository for the addition of new data enabling others to reuse existing data to reduce the time necessary to fully evaluate GMOs after authorization
On GMO risk assessment, decisions for regulators and policymakers, new legislation, etc.
Government agencies, academic institutions, non-government organizations, and private industry

The following organizations are likely to benefit from the findings from the project results:

Community Reference Laboratory (CRL), which evaluates and validates quantitative GMO detection methods.
The European Commission Joint Research Centre's GMO Methods Database
The Biotechnology and GMOs Unit of the European Commission Joint Research Centre's Institute for Health and Consumer Protection (IHCP).
The European Network of GMO Laboratories (ENGL)
The Institute for Reference Materials and Measurements (IRMM) of the European Commission's Joint Research Centre.
CEN, the European standards body, which is preparing standards for the analysis of GMOs and derived products in foods through its Technical Committee 275, working group 11.
ISO: Technical Committee 34 covers the standardisation of human and animal foods

Dissemination Activities

Due to the politically and scientifically charged subject of GMO safety, it was important to disseminate project results when they were completed. This was planned to avoid potentially erroneous conclusions before all data were analysed. For that reason, intense project dissemination was planned for the end of the project, especially for presentation at the GMSAFAOOD Final Conference and after the termination of the project. Specific, broad dissemination of the project results and up-to-date information about the current state-of-the-art as well as dissemination of the project conference proceedings are on the project website: www.GMSAFOODproject.eu. E-newsletters were and will continue to be sent to a recipient mailing list of more than 1800 recipients in the GMO and post market monitoring fields.

The GMSAFOOD GMO Safety and Post Market Monitoring Conference was held in Vienna, Austria on March 6-8th, 2012 at the Medical University of Vienna. The theme of this conference was GMO safety and post market monitoring and the main objective was to have a science-based discussion regarding applicability, practicality and utility of post market monitoring for foods derived from GMOs, to help regulators consider post market monitoring as a tool for regulatory decision making, to provide a venue for scientists in the public and private sectors to communicate. Additionally, parts of the final conference were open to the press and public to ensure broad stakeholder dissemination.

To view the conference Report:
http://www.gmsafoodproject.eu/Conference/MagazineGMSAfood_webQ.pdf

To download lecture PDFs:
http://www.gmsafoodproject.eu/Sections.aspx?section=463.527

To view the videos of selected talks, go to:
GMSAFOOD website:
http://www.gmsafoodproject.eu/
or
GMSAFOOD Channel on YouTube.com:
http://www.youtube.com/user/gmsafood?feature=results_main

The main audience for the projectâ??s results includes:
SCIENCE
Scientific Magazines and Journals

INDUSTRY
Biotech
GM producing companies
Agricultural sector

GOVERNMENT
Decision makers
Policy makers
Funding bodies

PUBLIC
Press Agencies
Newspapers
Internet University website newsletters
Broadcast (television science and information technology programmes and radio news, science and information technology programmes)

RESEARCH AND EDUCATION COMMUNITIES
Universities
Schools
Research Institutes

To date, 9 peer-reviewed publications have been published from the GMSAFOOD project and another 12 will follow in the coming months.

Dissemination during the conference and following the conference has resulted in increasing awareness about the project's results:

Dr. TJ Higgins was invited to report on the conference and the project's findings by the Australian Foods Standard Agency.

Dr. Epstein has been invited to the EFSA network on GMO risk assessment in Parma, Italy on May 3, 4th, 2012 as a direct result of the conference dissemination.

Exploitation of results

There are two main results for exploitation from the project. One is the GMSAFOOD GMO Database and the other are the transgenic cowpea and chickpeas.

The plan is to exploit the GMSAFOOD database as an open source public data repository for the identification of untoward effects of GMOs on human and animal health after authorization. A public data repository will provide an opportunity for improved knowledge in the GMO field. It will allow regulators and risk assessors access to raw data to make accurate predictive analysis on risk and assist in the identification of untoward effects and potential biomarkers predictive of adverse reactions to GMOs.

Website: http://www.GMSAFOODproject.eu

Contact details; Secretariat@GMSAFOODproject.eu

List of Websites:
http://www.GMSAFOODproject.eu

Secretartiat@GMSAFOODproject.eu

Project videos: on the GMSAFOOD website and GMSAFOOD channel at YouTube.com

TJ Higgins lecture:
http://www.youtube.com/watch?v=yKda722clq8

Teagasc, Maria Walsh and Stefan Buzoianu lecture:
http://youtu.be/uaN_ajxzZGU

Press conference GMSAFOOD conference Ashild Krogdahl:
http://www.youtube.com/watch?v=7nYiRJS-CZM&feature=relmfu

Press conference GMSAFOOD conference TJ Higgins:
http://www.youtube.com/watch?v=gHUKE_luMR8&feature=relmfu

Press conference GMSAFOOD conference Peadar Lawlor:
http://www.youtube.com/watch?v=JRKJmSU0Eq0&feature=relmfu

Press conference GMSAFOOD conference Michelle Epstein:
http://www.youtube.com/watch?v=JkivByhMNCo&feature=relmfu

GMSAFOOD conference Panel discussion on GMO safety and post market monitoring:
http://www.youtube.com/watch?v=_1iK9NPb64A&feature=relmfu

List of Beneficiaries

Peadar Lawlor
Teagasc, Irish Agriculture and Food Development Authority, Pig Development Department, Animal and Grassland Research & Innovation Centre,
Moorepark Fermoy, Co.Cork, Ireland
Phone: +353 25 42217
Fax: +353 25 42340
Email: peadar.lawlor@teagasc.ie

A?shild Krogdahl
Norwegian School of Veterinary Medicine
Gut and Health Group of the Aquaculture Protein Centre
P.O. Box 8146 Dep,
NO-0033 Oslo, Norway
Phone: +47 22964534
Fax.: +47 22597310
Emil: ashild.krogdahl@nvh.no

TJ Higgins
CSIRO Plant Industry
Clunies Ross St, Black Mountain
Canberra ACT 2601 Australia
Phone: +61 2 6246 5001
Fax: +61 2 6246 5000
Email: Tj.Higgins@csiro.au

Central Food Research Institute
Department of Food Safety
H-1022 Budapest, Herman Ottat 15.
Phone: +36 1 225 3343
Fax: +36 1 225 3342
Email: e.gelencser@cfri.hu

Michelle Epstein
Medical University of Vienna
Department of Dermatology, DIAID
Experimental Allergy
Wahringer Gartel 18-20, Room 4P9.02,
A1090, Vienna, Austria
Phone: +431 40160 63012
Fax: +431 40160 963012
Email: michelle.epstein@meduniwien.ac.at

Ahmet Sümer
Troyka Ltd
MetuTechnopolis (Odtu Teknokent)
Ikizler Binasi, Zemin Kat Ara-2
Middle East Technical University
ODTU, 06531 Ankara, Turkey
Phone: +90 312 210 1147
Fax: +90 312 210 1151
Email: ahmet.sumer@troyka-ltd.com

Related information

Reported by

THE MEDICAL UNIVERSITY OF VIENNA
1090 VIENNA
Austria
Follow us on: RSS Facebook Twitter YouTube Managed by the EU Publications Office Top