Preventing Food Allergies by Tricking Dendritic Cells

Food allergies are adverse responses to components within the foods we eat that result in a self-damaging response from our immune system. These components are usually proteins or glycoproteins. The results of this misguided immune reaction can include nausea, vomiting, diarrhea, and, in severe cases, death. Most food allergies are associated with eight foods — milk, eggs, peanuts, tree nuts, fish, shellfish, wheat, and soybeans — although many other foods can also cause allergies. Approximately 8% of the US population under age eighteen suffers from food allergies, and recent reports provide evidence that peanut and tree nut allergies — two of the most severe types of food allergies — are becoming increasingly more common (Branum & Lukacs 2008; Sicherer et al. 2010; Gupta et al. 2011). Whereas milk and egg allergies are often outgrown, peanut and tree nut allergies typically persist throughout adulthood (Fleischer 2007). Indeed, the risk of unintended allergen exposure due to accidental ingestion is a constant fear for individuals with life-threatening food allergies. Our intestines (about 25 feet in length) are the largest immunologic organ in our bodies and would cover roughly 100 square feet if spread out; we are still learning what cells and signaling systems regulate their response to the multitude of different foods and potential pathogens that pass through them.

Several questions related to the causes of food allergies and their treatments are currently the focus of intense investigation. Why do some foods but not others cause an allergic response? What similarities exist between the response to food allergens and potential intestinal pathogens or other intestinal microbial flora? Here, we'll introduce you to an innovative approach researchers have taken to uncover important clues about food allergies and a possible new therapeutic mechanism to prevent them.

Food allergies are a type I hypersensitivity response, which means that they are mediated by IgE antibodies. In this situation, these types of antibodies cause an inflammatory response with the release of vasoactive compounds from basophils and mast cells. Are food allergies associated with genetic or environmental factors? The answer is yes, they are associated with both factors. Indeed, several variables have been associated with food allergies, including genetics; how, when, and in what amount the food is first encountered; geographic influences; methods of food processing; and the type of bacteria that normally reside in our gastrointestinal tract (Maleki 2004; Lack 2008; Hong et al. 2009; Round & Mazmanian 2009; Berin & Sicherer 2011; Vereda et al. 2011).

What makes certain foods allergenic, whereas others are not? The term "allergenicity" reflects the ability of a protein or glycoprotein to induce an allergic reaction exemplified by the production of IgE antibodies. Allergenicity is a reflection of several characteristics. Many food allergens belong to the same protein family and have similar structural features, such as the cupin and prolamin superfamilies. In addition, a protein's ability to survive digestion and pass through the intestines intact can affect how it will act as an allergen.

Although therapy for food allergies remains an active area of research and several methods are under evaluation, strict avoidance is the current standard of care (Boden & Wesley Burks 2011). Recent research has advanced our understanding of food allergies, facilitated the development of guidelines for accurate diagnosis, and allowed the characterization of many food allergens from both plant and animal sources (Jenkins et al. 2005; Sampson 2005; Jenkins et al. 2007).

How have researchers learned about food allergies? Some important findings have come from the development of animal models to study food allergens. This research has clarified our understanding of the mechanisms behind food allergies and has aided in the evaluation of potential new therapies (Dearman & Kimber 2009). For example, researchers have discovered that genetically similar pigs exhibited drastic differences in the bacteria that grow in their gut depending on whether the pigs were housed in hygienic (clean) or in natural environments, and these differences correlated with changes in the expression of immune system–related genes (Mulder et al. 2009). Further, mouse models of peanut or tree nut allergies have been used to characterize changes in cytokines and to evaluate potential test therapies (Kulis et al. 2011; Smarr et al. 2011).

Our bodies are conditioned to foods during development, and this process requires interactions between structural and cellular components. These interactions establish a delicate balance that allows individuals to become either tolerant of or sensitized to a given food allergen. The conditioning of the gut lining plays an essential role in developing food tolerance, and a complex array of factors within the gastrointestinal tract serves as the cellular and molecular basis of food allergies (Brandtzaeg 2010).

What are some of these factors? One major factor is the activity of dendritic cells, specialized cells in the gut that act as antigen presenting cells. These cells are important for intestinal immune regulation (Coombes & Powrie 2008). Dendritic cells bind potential allergens and act as reporters. The information they present can initiate a signaling cascade that leads to the production of IgE-specific antibodies by B-cell– and T-cell–priming events. Alternatively, the information presented by dendritic cells can induce tolerance to potential allergens. Dendritic cells migrate from the intestinal lining into lymph nodes of the gut, and this travel is essential for carrying information to develop tolerance (Worbs et al. 2006). Exactly how dendritic cells initiate the decision-making process either to generate an immune response or to induce tolerance is an area of intense interest.

Another major factor is a set of surface proteins on dendritic cells called C-type lectin receptors (CLRs). CLRs are highly conserved members of the pattern recognition family of proteins that contain carbohydrate recognition domains, and they contribute to dendritic cell–induced tolerance (Zelensky & Gready 2005). Dendritic cell CLRs interact with potential pathogens by binding to their surface carbohydrates, and this interaction is thought to function in antigen processing. It is an important step in the decision to attack or tolerate a foreign antigen, and new evidence suggests that interactions between CLRs and antigens may provide a route for the establishment of oral tolerance to foods.

Using a molecular trick, Yufeng Zhou and his colleagues found that food allergens disguised as bacteria can prevent food allergy development in a mouse model system (Zhou et al. 2010). How did they make this exciting discovery? They used a carbohydrate modification of a protein (mannosylation) as a method to mimic the molecular surface of a bacterium. Mannose is a carbohydrate residue often found in terminal positions of long sugar chains attached to microbial glycoproteins, such as those in the bacterium that causes tuberculosis (Mycobacterium tuberculosis), but such highly clustered residues of terminal mannose are not common in humans.

Zhou and his colleagues used two test groups of mice in their experiments. They fed the first test group an unmodified form of the protein allergen bovine serum albumin (BSA), which naturally lacks any carbohydrates, and they fed the second test group BSA to which mannose residues were covalently linked (mannosylated BSA). They fed the mice in the two experimental test groups small doses of one of the two types of BSA for several weeks to allow them to become sensitized (allergic) or not to the protein. Following this sensitization period, they fed the mice a larger amount of unmodified BSA as a challenge to see how their immune systems would react. Whereas the mice previously fed the unmodified BSA during the sensitization period had severe allergic responses to the BSA challenge with high IgE levels, the mice fed the mannosylated BSA during the sensitization period had relatively reduced IgE levels and reduced histamine release, and none had severe allergic responses to the subsequent BSA challenge (Figure 1).

Figure 1: Mannosylated BSA protects mice from hypersensitivity to a model food allergen.

In a mouse model of food allergy, mice were potentially exposed in three stages: treatment, sensitization, and challenge. (a) Here, mice did not receive a treatment. They were then sensitized and challenged. Mice were sensitized with PBS buffer, PBS + 200 mg BSA (BSA), or PBS + 200 mg mannosylated BSA (Man), together with 10 mg of cholera toxin (a commonly used adjuvant) once a week for seven weeks. One week following the final sensitization, the mice were challenged by intraperitoneal injection with 1 mg of BSA and were scored for allergic symptoms after 30 minutes. The reaction of the mice to the BSA challenge was scored according to the following criteria: 0, no sign of reaction; 1, scratching and rubbing around the nose and head; 2, decreased activity with an increasing respiratory rate, pilar erecti (stiffening of the hair), and/or puffing around the eyes; 3, labored respiration and cyanosis (blue or purple coloration of the skin) around the mouth and tail; 4, slight or no activity after prodding or tremors and convulsion; 5, death. (b) In a different experiment, mannosylated BSA protects against food allergy. The statistical significance is indicated by an asterisk (*). Mice were fed 200 mg of either BSA or mannosylated BSA for three consecutive days and then were sensitized and challenged, and their symptoms were evaluated as in (a).

© 2010 Nature Publishing Group Adapted from Zhou, Y. et al. Oral tolerance to food-induced systemic anaphylaxis mediated by the C-type lectin SIGNR1. Nature Medicine 16, 1128–1133 (2010). doi: 10.1038/nm.2201. All rights reserved.

Figure Detail

The researchers then asked, could dendritic cells in the gut preferentially endocytose (engulf) the mannosylated BSA? They hypothesized that this could be an important step in establishing the decreased allergic response. To test their hypothesis, they used fluorescent labeling of the different forms of BSA coupled with cell-sorting methods and discovered that their hypothesis was indeed the case: dendritic cells from the gut were able to endocytose the mannosylated BSA better than the unmodified BSA.

Figure 2: SIGNR1 binds specifically to mannosylated BSA.

Microplates coated with the indicated dilutions of BSA, galactosylated BSA, or mannosylated BSA in buffer were blocked with assay buffer containing goat serum, and the plates were incubated with 0.5 μg/ml recombinant SIGNR1 at room temperature for two hours. The recombinant SIGNR1 protein was expressed as a fusion protein to the constant region of human IgG (SIGNR1-Fc) to allow for easy detection by the secondary antibody. The plates were then incubated with peroxidase-conjugated goat anti-human IgG Fc antibodies for one hour at room temperature, and the signal was developed by the addition of tetramethylbenzidine as the substrate. The reactions were stopped by the addition of sulfuric acid and read in a microplate reader at an absorbance of 490 nm. The relative binding activity was expressed as the absorbance at 490 nm after subtracting the background signal. As shown, SIGNR1 shows specific binding to mannosylated BSA.

© 2010 Nature Publishing Group Adapted from Zhou, Y. et al. Oral tolerance to food-induced systemic anaphylaxis mediated by the C-type lectin SIGNR1. Nature Medicine 16, 1128–1133 (2010). doi: 10.1038/nm.2201. All rights reserved.

Figure Detail

How did the mannosylated BSA cause SIGNR1-expressing dendritic cells to initiate tolerance? To answer this question, they decided to characterize the cytokine expression profiles of the dendritic cells. They showed that treatment of SIGNR1-expressing dendritic cells with mannosylated BSA led to a striking increase in the expression of the cytokine interleukin-10 (IL-10). Next, the researchers suppressed the increase in IL-10 expression by incubating the cells with a blocking anti-SIGNR1 antibody. These results provided evidence that the SIGNR1-dependent pathway was specifically responsible for the release of IL-10.

What role does IL-10 play in the regulation of the immune response? It is actually quite important for the suppression of proinflammatory cytokines primarily associated with macrophages (Cyktor & Turner 2011). The immunosuppressive properties of IL-10 reduce the ability of the immune system to kill pathogens, and IL-10 is important for the control of intracellular bacterial, fungal, and parasitic infections. In some cases, pathogens have developed the ability to exploit IL-10 function to increase their own survival. In addition, researchers have observed an increase in IL-10 expression following pollen allergen immunotherapy that was associated with decreased IgE production (Platts-Mills & Woodfolk 2011).

Although the discovery of the SIGNR1-associated increase in IL-10 expression provided new insights into how these dendritic cells initiate tolerance, Zhou and his colleagues carried out additional experiments to further characterize the immune-regulatory signaling events and the role of T cells. They determined that culturing T cells with the SIGNR1-expressing dendritic cells exposed to mannosylated BSA caused the T cells to differentiate into regulatory type 1 T cells with elevated IL-10 and interferon-gamma (IFN-γ) expression levels. What role might IFN-γ play in this process? IFN-γ has immunoregulatory activity and is important for protection against intracellular bacterial and viral pathogens. In line with the results of their earlier experiments, Zhou and his colleagues found that the elevated levels of IL-10 and IFN-γ expression in T cells were greatly reduced when the dendritic cells were incubated with an anti-SIGNR1 antibody. Further, by testing mice deficient for the IL-10 or IFN-γ protein, they learned that the IL-10 and IFN-γ proteins were required for the tolerance-inducing response of the mannosylated BSA.

Next, the researchers asked, could regulatory type 1 T cells isolated from tolerant mice confer tolerance to sensitive mice? Remarkably, when the researchers transferred regulatory type 1 T cells isolated from tolerant mice (treated with mannosylated BSA) to sensitive mice, the cells induced tolerance to unmodified BSA in the previously sensitive mice. This illuminating result showed that these cells are able to retain and transmit the essential information required for tolerance to the allergenic protein.

When combined with findings from other research, the data presented by Zhou and his colleagues suggest a model whereby the binding and activation of the SIGNR1 receptor on dendritic cells by mannose-modified proteins initiate a signaling pathway that can induce tolerance to the bound proteins (Rabinovich 2010) (Figure 3). Although this research provides important clues about a potential route for the development of oral tolerance to food allergens, several questions remain. For example, does the human homolog of SIGNR1, called DC-SIGN, function in the same manner to promote tolerance? What crucial components of the SIGNR1 pathway allow oral tolerance, rather than sensitization, to develop? Can the SIGNR1-dependent pathway be tailored to induce tolerance to other food allergens but still permit the inflammatory pathways for pathogens to remain intact?

Figure 3: A model for SIGNR1-mediated oral tolerance

Mannosylated BSA (Man₅₁-BSA) is captured by SIGNR1 on a dendritic cell (DC) in the mucous membrane (lamina propria) of the intestines. This unleashes a cascade of events that promote tolerance to food allergens. This cascade involves the secretion of IL-10, which leads to the generation of regulatory type 1 T cells (Tr1) cells in the mesenteric lymph node. C-C chemokine receptor type 7 (CCR7) expression may direct lamina propria DCs to mesenteric lymph nodes to encounter T cells. Upon differentiation, Tr1 cells secrete IL-10 and interferon-γ (IFN-γ), which ultimately prevent systemic anaphylaxis and promote oral tolerance in response to food allergens.

© 2010 Nature Publishing Group Rabinovich, G. A. A sweet path toward tolerance in the gut. Nature Medicine 16, 1076–1077 (2010) doi: 10.1038/nm1010-1076. All rights reserved.

Figure Detail