DEVELOPING AN ANIMAL MODEL TO PREDICT ALLERGENICITY OF GENETICALLY MODIFIED FOODS

• Sponsoring Institution: National Institute of Food and Agriculture
• Project Status: COMPLETE
• Funding Source: OTHER GRANTS
• Reporting Frequency: Annual
• Accession No.: 0189114
• Grant No.: 2001-52100-11249
• Proposal No.: 2001-04239
• Project Start Date: Sep 15, 2001
• Project End Date: Sep 14, 2005
• Grant Year: 2001
• Program Code: [(N/A)]- (N/A)

Recipient Organization
UNIV. OF ARKANSAS FOR MEDICAL SCIENCES
4301 W. MARKHAM STREET
LITTLE ROCK,AR 72205

Performing Department
PEDIATRICS

Non Technical Summary
The modern techniques of biotechnology offer great potential to further improve the abundance, availability, and nutritional quality of our food supply. However, biotechnological manipulation of food sources raises the concern that food allergens may be moved between foods or that proteins with no significant history of consumption could prove to be allergenic when introduced into a food crop. Methods need to be developed to accurately identify and characterize potential protein allergens and for estimating their relative potency as the first step towards risk assessment. The newborn pig mimics many of the features of the human newborn, particularly with regard to digestive and immunologic function. Using the neonatal pig as a food allergy model, investigations of known food allergens and non-allergen food proteins are being used to validate the allergenic response to strong (peanut), moderate (soybean), mild (fruit) and non-allergenic food sources (Rubisco). In this manner we will establish criteria for protein sensitization and challenge responses that can be used to determine the relative allergenicity of novel proteins. The goal of this project is to develop an animal model that can be used to assess the potential allergenicity of novel proteins introduced into foods by genetically modified organisms (GMOs).

Animal Health Component

50%

Research Effort Categories

Basic

(N/A)

Applied

50%

Developmental

50%

Classification

Knowledge Area (KA)	Subject of Investigation (SOI)	Field of Science (FOS)	Percent
712	5010	1090	70%
723	3599	1150	30%

Knowledge Area
712 - Protect Food from Contamination by Pathogenic Microorganisms, Parasites, and Naturally Occurring Toxins; 723 - Hazards to Human Health and Safety;

Subject Of Investigation
3599 - Swine, general/other; 5010 - Food;

Field Of Science
1090 - Immunology; 1150 - Toxicology;

Keywords

animal models

allergens

genetic engineering

food quality

food safety

predictive models

food contamination

swine

immunology

toxicology

allergies

risk assessment

information dissemination

biotechnology

elisa (assay)

bioassays

inflammation

comparative analysis

performance evaluation

Goals / Objectives
The outstanding challenge in food allergy is to understand what makes a common protein or peptide behave as an allergen for some individuals or why it may become a much more potent allergen than usual. Ethical and practical reasons prevent experimental testing of newly derived protein sources as potential allergens in humans. The overall objective of this project is to develop an animal model that can be used to assess the potential allergenicity of novel proteins introduced into foods by genetically modified organisms (GMOs) and educate the public regarding the safety of GM-food products. This will be achieved via 3 supporting objectives: (1) establish a useful and validated food allergy animal model, (2) characterize the molecular features of allergens using the swine food allergy model, and (3) disseminate to the public the knowledge we obtain about the relative risk of GM-foods. With the expanded use of biotechnology, crop plants that express novel proteins introduced from other sources and the issues of protein allergenicity assume toxicological significance. Although known allergens in genetically engineered or altered foods can be identified using traditional immunological assays with sera from allergen-sensitive individuals, such as radioallergosorbent tests, enzyme-linked immunosorbent assays or immunoblotting procedures, these methods are inadequate because they have only a 50% positive predictive value for clinical allergic food allergy. A valid food allergy animal model that more accurately simulates the human allergenic response is clearly needed. Using the swine neonatal food allergy model we will investigate known food allergens and non-allergen food proteins to determine the allergenic response from strong (peanut), moderate (soybean), mild (fruit) and non-allergenic food sources (Rubisco). In this manner we will establish criteria for protein sensitization and challenge responses that can be used to determine the relative allergenicity of novel proteins that are introduced into the food chain. Thus, the intestinal inflammatory response and the immunogenicity/allergenicity of potential allergen sources derived from GMO' and GM-foods will be compared to a profile of known food allergens and non-food allergens in the swine animal model.

Project Methods
Controversy continues to build over the safety of genetically modified crops as biotech and plant geneticists invest effort in developing GM-food crops (e.g., weed- and insect-resistant lines, increased nutrient sources), consumers demand zero tolerance for GM-foods, and regulatory agencies (FDA, USDA, EPA) grapple with the safety issues of GM-crops that could be introduced into the food chain. We will develop the newborn pig as a food allergy animal model using intraperitoneal and intragastric administrations of food allergens according to a sensitization/challenge protocol we have developed. The establishment of a food allergen profile using potent allergens that can induce symptoms of anaphylaxis (e.g., peanuts), moderate or transient allergy symptoms (e.g., soybeans) and mild oral allergy symptoms (e.g., fruits), and non-food allergens (e.g., Rubisco) will enable us to categorize the level of potential allergenicity of novel proteins that could be introduced into the diet. The assessment of GMO proteins and GM-foods for inflammatory responses under natural dietary conditions can be assessed by comparing GM-food sources to known food allergens and non-food allergens to determine their relative allergenic potential. Intestinal hypersensitivity is characterized by symptoms ranging from episodes of diarrhea to anaphylactic shock that are physiologically similar in pigs and man. In the pig model, observations of diarrhea and changes in intestinal morphology (villus height and maturation) will be used as measures of intestinal damage in response to a GM-food diet. By placing small tubing in the neonatal pig gastrointestinal tract, samples of digested food can be obtained to determine the level of enzymatic digestion of food allergens and proteins. This will enable us to provide a more accurate characterization of known food allergen enzymatic stability, which has routinely been performed in the test tube and does not mimic natural digestive processes. This will contribute knowledge to both sensitization and threshold levels that provide characteristic IgE-mediated responses that may be correlated to the human response. We will compare the allergenic versus immunogenic components of native and an allergen null soybean seed products using this animal model to determine the potential allergenicity of unknown food sources. Finally, we will disseminate to the public the knowledge we obtain about the relative risk of GM-foods. The education and outreach objectives will deliver science-based communication on various issues concerning the benefits and safety of GM-crops and foods. The information disseminated will include what the issues are, how they affect the various groups targeted to receive the information and how scientists and regulatory agencies assess safety risks.

Progress 09/15/01 to 09/14/05

Outputs
Efforts were continued to determine the relative sensitivity of newborn pigs to soybean and egg proteins in the Yorkshire/Duroc/Landrace breed. Intermittent success led to considerable discussion with Purdue geneticists that resulted in collaborative efforts to identify high and low IgE responders in different swine breeds. Selection criteria for various gilt strains and two boar donors have been designed to determine primary candidates that score negative for both soy and peanut allergen and exhibit minimal diarrhea scores (low IgE responders) and pigs that score positive for both soy and peanut allergen and exhibit high diarrhea scores (high IgE responders). The PI is collaborating with Alan Schinkel and Terry Stewart at Purdue by providing a boar with peanut sensitivity for breeding purposes and directing the skin prick and oral challenges. The PI is also collaborating with Niels Nielsen, now at NCS, and Eliot Herman at Danforth, in an effort to determine the relative allergenicity of various soybean varieties that are being modified to reduce the transient hypersensitivity in weaning pigs. With regard to potential allergenicity, it is generally thought that proteins susceptible to gastrointestinal (GI) digestion are inherently safer than those that are stable. Studies have been conducted comparing the digestibility of known food and non-food allergens with common non-food proteins using pepsin or a simulated gastric fluid and simulated intestinal fluid as described in the United States Pharmacopoeia. However, there is currently no reliable single or combined method to evaluate or predict the allergenicity of a protein using these methods. . Experiments were conducted in neonatal pigs to determine the relative stability of peanut allergens in the context of a peanut meal gastrointestinal digestion. Soluble proteins were rapidly digested and insoluble material continued to be released as IgE-reactive proteins throughout the gastrointestinal tract. Thus, gastrointestinal digestion of food allergens can play a prominent role when assessing major and minor allergens within the context of a food matrix or meal and during the sensitization phase of IgE-mediated allergy.

Impacts
Validation for natural high and low IgE responders in the swine neonatal model will extrapolate to human immediate hypersensitivity in both anatomical and physiological areas. Strain variations have been identified that suggest that swine are genetically programmed to soybean and peanut allergy during sensitization. Allergen stability to gastrointestinal digestion verified the continued release of peanut proteins in the small intestine that could lead to IgE sensitization by gastrointestinal associated lymphoid tissues. Assessment of modified soybean meal has been proposed to determine the transient hypersensitivity in weaning piglets. Successful results will benefit both the swine industry and the basic scientist interested in a more validated natural hypersensitivity animal model.

Publications

• Kopper RA, West CM, Helm RM. 2005. Physiological digestion of peanut proteins in the swine model of food allergy. J Allergy Clin Immunol 33:abst 133.
• Kopper RA, Odum NJ, Sen M, Helm RM, Stanley JS Burks AW. 2005. Peanut protein allergens: the effect of roasting on solubility and allergenicity. In Arch Allergy Immunol. 136(1):16-22.
• Kopper RA, West CM, Helm RM. 2006. Comparison of Physiological and In Vitro Porcine Gastric Fluid Digestion. Int Arch Allergy Immunol Submitted
• Golden C, McMahon M, Thampi P, Badger TM, Nagarajan S, Helm RM. 2006. Histopathological and Immunohistochemical Changes in Gut-associated Lymphoid Tissue (GALT) in Neonatal Piglets Fed Casein. Biol Neonate. Manuscript in preparation
• Helm RM. 2005. Animal models for the study of allergy. IN: Handbook of Laboratory Animal Science. Hau J, Van Hoosier GL Eds. CRC Press.VIII;pp129-149.
• Helm RM. 2005. Hypersensitivity reactions: Non-rodent animal models. IN: Investigative Immunotoxicology. Tryphonas H, Fournier M, Blakley BR, Smits JEG, Bousseau P Eds. Taylor & Francis, CRC Press. ; Chapter 18,pp 281-289.
• Helm RM. 2005. Porcine Immune System. IN: Encyclopedic Reference of Immunotoxicology. Springer-Verlag.

Progress 10/01/03 to 09/30/04

Outputs
Continuation of the Yorkshire/Landrace, 0.5 Yorkshire and 0.5 Landrace, neonatal peanut allergy model was interrupted by an alteration in the source of genetic stock and supplier to a more modified Yorkshire/Duroc/Landrace, 0.25 Yorkshire, 0.25 Duroc, and 0.25 Landrace. This change in genetic stock resulted in re-establishment that genetics plays an important role in the sensitization of the breed. Our course has subsequently changed to identify the high and low IgE responders by skin test/oral response to peanut, soybean and egg allergens. An additional aim was incorporated that would establish the transient hypersensitivity following soybean feeding at weaning. Our success in these two aspects is continuing with variable results: peanut sensitization skin testing is positive; however, oral challenge did not result in emesis. There is evidence of diarrhea at 24-48 hours in a number of the piglets. In the transient hypersensitization to soybean meal at weaning, we also have met with moderate success. The problems are believed to be a result of breed selection for production that minimizes the normal transient hypersensitivity to soybean thereby increasing growth rate and weight. Both aims are still under investigation with different breeding stock to determine the optimal sensitization/response to food allergen sensitizations in different swine breeds beginning with the Landrace and crossing with the various other breeds used in swine production facilities. We also undertook investigations to determine the allergenic potential of peanut allergens. Correlations between simulated gastric conditions, pepsin digestion, and actual gastric fluid from piglets were used to determine the physiological digestion and stability of peanut allergens. In vitro digestion of Ara h1 with pepsin and porcine gastric fluid resulted in virtually identical hydrolysis patterns as observed on SDS-PAGE. The protease activity of both pepsin and gastric fluid were inhibited at high pH and in the presence of pepstatin. The results of this study demonstrated that protein digestion in the porcine stomach is carried out exclusively by pepsin. Thus, in vivo gastric digestion is modeled accurately by peptic hydrolysis. In subsequent studies, the in vitro digestion was compared with actual in vivo digestion to compare both simulation gastrointestinal and intestinal fluid digestion models. To perform these studies, stomach and small intestinal cannulation was performed and samples were taken at 0, 15, 30, 60, 120, 180 and 240 minutes following a 20 gram bolus of peanut meal. Peanut meal neutralized the stomach contents resulting in inactivation of pepsin for approximately 30 minutes after which acidification resulted in rapid digestion of soluble intact proteins/allergens. Particulate material and continued release of soluble material was evident in the stomach and small intestine throughout the sampling period suggesting that gastric and intestinal enzymatic digestion in the gut may play a more prominent role when assessing allergens in the context of a food matrix or meal both during sensitization and response phases of IgE-mediated food allergy.

Impacts
The neonatal swine model continues to be useful in the identification of allergen sensitization that correlates with human food-sensitive individuals from a clinicopathologic and genetic background. Additional studies indicate that physiological digestion of allergens in the food matrix and diet will contribute to the understanding of allergen sensitization.

Publications

• Kopper RA, Odum NJ, Sen M, Helm RM, Stanley JS, Burks AW. 2004 Peanut Protein Allergens: Gastric Digestion Is Carried Out Exclusively By Pepsin. J Allergy Clin Immunol. 114(3):614-8.
• Kopper RA, Odum NJ, Sen M, Helm RM, Stanley JS, Burks AW. 2005. Peanut Protein Allergens: The Effect of Roasting on Solubility and Allergenicity. Int Arch Allergy Immunol. In press.
• Ladics GS, Holsapple MP, Astwood JD, Kimber I, Knippels LM, Helm RM, Dong W. 2000. Workshop overview: approaches to the assessment of the allergenic potential of food from genetically modified crops. Toxicol Sci 73:8-16.
• Helm RM, Ermel RW, Frick OL. 2003. Nonmurine models of food allergy. Environ Health Perspect 111:239-244.
• Kimber I, Dearman RJ, Penninks AH, Knippels LMJ, Buchanan RB, Hammerberg B, Jackson HA, Helm RM. 2003. sment of protein allergenicity on the basis of immune reactivity: animal models. Environ Health Perspect 111:1125-30.
• Herman EM, Helm RM, Jung R, Kinney AJ. 2003. ic modification removes an immunodominant allergen from soybean. Plant Physiol 132:36-43.
• Helm RM. 2003. biotechnology: is this good or bad? Implications to allergic disease. Ann Allergy Asthma Immunol S1.
• Helm RM. 2003. Porcine Immune System. IN: Encyclopedic Reference of Immunotoxicology. Springer-Verlag.
• Helm RM. 2003. Detecting allergens in food. IN: Detecting Allergenic Residues in Food. Woodhead Publishing Limited. 2003. Submitted.
• Helm RM. 2003. Animal Models for the Study of Allergy. IN: Handbook of Laboratory Animal Science, Vol III. Animal Models. CRC Press.
• Helm RM. 2003. Hypersensitivity reactions: Non-rodent animal models. IN:Investigative Immunotoxicology. Eds. Tryphonas H, Fournier M, Blakley BR, Smits JEG, Brousseau P. Taylor & Francis.

Progress 10/01/01 to 09/30/02

Outputs
The Yorkshire/Landrace strain of swine is being used to determine the immunoregulatory mechanisms of clinically relevant GI immediate hypersensitivity; developing a model for threshold food immune sensitization/allergic response levels for prediction of novel proteins potential allergenicity; and to offer therapeutic options for treatment of food allergy. Neonatal piglets reared on a peanut/soybean-free diet were intraperitoneally sensitized to peanut, soybean or egg in the presence of cholera toxin as adjuvant. PBS/sham injected orally challenged animals or non-challenged animals were used as negative food allergy controls. Skin tests and oral challenge by gastric intubation performed with the respective foods established an allergen specific symptomatic response. Oral challenged piglets responded with no symptoms (controls), mild symptoms (egg and soybean - body rashes, lethargy), and more severe symptoms (peanut- vomiting, diarrhea, respiratory distress) compatible with human food allergic reactions. Skin tests confirmed sensitization to the respective food with little or no response to non-sensitizing foods. Endoscopic analysis was performed pre-, 2, and 4 hours post challenge; gross mucosal inspection and mucosal tissue samples from esophagus, stomach, and duodenum were obtained with a GIF-XQ40 gastroscope. At endoscopy, both the stomach and small intestine were nearly devoid of peanut meal by two hours. At these early time points, the mucosal surface of esophagus, stomach and small intestine appeared normal. Analysis of tissues at 2 hours showed normal esophageal, gastric and duodenal histology; at 4 hours, pigs showed evidence of vascular congestion in the gastric and duodenal lamina propria. At 24 hours, gastric tissues showed lamina propria congestion and massive mucus extrusion into the luminal space. In the proximal small intestine, vascular congestion and hemorrhage was identified in the lamina propria and focal villus denudation was observed. Small intestinal lamina propria showed an expansion of lymphocytes with focal collections of eosinophils. Challenged animals showed and increased expression of IL-10 and INF-gamma but not TNF-alpha or IL-4. Collection of orally administered radiopaque rings from intestinal sites showed increased clearance of rings to distal sites in challenged pigs compared to controls. Gastric motility is increased as evidenced by the lack of peanut chow in the stomach and small intestine following oral challenge. Confirmed motility alteration was in evidence with increased numbers of radio-opaque rings evident in the colon of peanut sensitive animals compared to more even distribution throughout the GI tract in controls. Mucus extrusion, vascular congestion and edema of stomach and small intestine represent acute features of gastrointestinal anaphylactic mucosal responses. Data suggests an altered cytokine profile consistent with Th2 type responses. This model serves as a prototype for determining the short and long term clinicopathologic features of food hypersensitivity reactions, understanding pathogenic mechanisms and developing future therapeutic tools.

Impacts
The swine food allergy model can be used for additional food allergen sensitization. The clinical symptoms and allergic response to peanut, soybean and egg correlates with that seen in food allergic responses in food-sensitive individuals. Extended food and non-food investigations will establish a food allergen profile - serious to mild to no allergic response - from which novel proteins identified in GMOs can be compared.

Publications

• References Agriculture Research. September 2002. Researchers develop first hypoallergenic soybeans. pp 16-17
• Helm RM, Burks AW, Herman E. Hypoallergenic Foods - Soybean and peanuts. Food Technology.ISB News Report. Ag Environ Biotech Develop. Oct 2002, Page 3.
• Presentations: September 24-27, 2001: NIEHS Meeting, Durham, NC. Presentation on the Swine Model for Food Allergy
• November 13-14, 2001: Health Canada Workshop, Ottawa, Canada. Assessment of Allergenicity using swine model.
• December 10-12, 2001..Chapel Hill, NC, USA. NIH - EPA - FDA: Assessment o the Peanut allergy in a swine model.
• January 14. 2002. Washington, DC, USA.FDA - CREES: Assessment of peanut allergens in a neonatal swine model.
• March 19-22, 2002: Society of Toxicology Annual Meeting. Nashville, TN, USA Approaches to the Assessment of Food Allergenicity. "Non-Rodent Animal Models for Assessing Protein Allergenicity".
• Journal Articles Helm RM. Furuta GT. Stanley JS. Ye J. Cockrell G. Connaughton C. Simpson P. Bannon GA. Burks AW. A neonatal swine model for peanut allergy. Journal of Allergy & Clinical Immunology. 109(1):136-42, 2002.
• Helm RM. Food allergy animal models: an overview. Ann NY Acad Sc 2002;964:139-50.
• Helm RM. Biotechnology and food allergy. Curr Allergy Asthma Rep 2002;2:55-62.