Eosinophilic esophagitis is rapidly increasing in incidence. It is associated with food antigen–triggered, eosinophil-predominant inflammation, and the pathogenic mechanisms have many similarities to other chronic atopic diseases. Studies in animal models and from patients have suggested that allergic sensitization leads to food-specific IgE and T-helper lymphocyte type 2 cells, both of which seem to contribute to the pathogenesis along with basophils, mast cells, and antigen-presenting cells. In this review our current understandings of the allergic mechanisms that drive eosinophilic esophagitis are outlined, drawing from clinical and translational studies in humans as well as experimental animal models.
Key points
- •
Eosinophilic esophagitis shares a clinical link with other atopic diseases and is caused by immune dysregulation secondary to allergic sensitization to dietary or aeroallergens.
- •
Allergic sensitization, which may occur via multiple routes, drives the formation of allergen-specific IgE and T cells, which seem to participate in the esophageal hypersensitivity response.
- •
Loss of tolerance is likely critical to pathogenesis, and may be secondary to regulatory T-cell imbalance.
- •
Eosinophilic esophagitis is dominated by T-helper lymphocyte type 2–mediated eosinophil-predominant inflammation, with key contributions from mast cells, basophils, epithelial cells, and dendritic cells.
Introduction
Chronic tissue infiltration of eosinophils is the hallmark of allergic inflammatory diseases, which include asthma, atopic dermatitis, and eosinophilic gastrointestinal diseases (EGIDs). Eosinophilic esophagitis (EoE), a type of EGID, is characterized by eosinophil-predominant inflammation isolated to the esophagus, the only organ in the gastrointestinal (GI) tract that is homeostatically devoid of eosinophils. There is significant evidence of a role for allergic mechanisms as the driving force in EoE. The foundation for this understanding began almost 20 years ago, when Kelly and colleagues reported that children with EoE resolved inflammation and clinical symptoms on amino acid–based diets. EoE is strongly linked to atopic disease and most often occurs comorbidly with asthma, eczema, allergic rhinitis, and anaphylactic food allergy, and a strong family history of such disorders is equally as common. The tissue inflammatory response in EoE is similar to other allergic inflammatory disorders and is characterized by a dysregulated immune response, with both IgE-mediated and non–IgE-mediated T-helper cell type 2 (Th2) responses. Cross-sensitization to aeroallergens seems to be common, and recent literature has begun to define antigen presentation, as well as the role of mast cells and basophils. In this review, our understanding of the allergic mechanisms involved in the pathogenesis of EoE are examined, based on clinical and translational studies in humans as well as experimental models in genetically modified animals.
Introduction
Chronic tissue infiltration of eosinophils is the hallmark of allergic inflammatory diseases, which include asthma, atopic dermatitis, and eosinophilic gastrointestinal diseases (EGIDs). Eosinophilic esophagitis (EoE), a type of EGID, is characterized by eosinophil-predominant inflammation isolated to the esophagus, the only organ in the gastrointestinal (GI) tract that is homeostatically devoid of eosinophils. There is significant evidence of a role for allergic mechanisms as the driving force in EoE. The foundation for this understanding began almost 20 years ago, when Kelly and colleagues reported that children with EoE resolved inflammation and clinical symptoms on amino acid–based diets. EoE is strongly linked to atopic disease and most often occurs comorbidly with asthma, eczema, allergic rhinitis, and anaphylactic food allergy, and a strong family history of such disorders is equally as common. The tissue inflammatory response in EoE is similar to other allergic inflammatory disorders and is characterized by a dysregulated immune response, with both IgE-mediated and non–IgE-mediated T-helper cell type 2 (Th2) responses. Cross-sensitization to aeroallergens seems to be common, and recent literature has begun to define antigen presentation, as well as the role of mast cells and basophils. In this review, our understanding of the allergic mechanisms involved in the pathogenesis of EoE are examined, based on clinical and translational studies in humans as well as experimental models in genetically modified animals.
Clinical associations of EoE with allergic diseases
Despite being frequently managed by gastroenterology specialists, there are several unique features of EoE that reflect the allergic nature of this disorder. During the last few decades, the prevalence of allergic diseases has been increasing fast: in Westernized countries, more than a quarter of the population has allergic eczema, allergic rhinitis, or food allergy. Paralleling this trend, epidemiologic studies show an increase in the number of children and adults with EoE. In 2006, the prevalence of pediatric EoE in Australia was estimated to have increased 18-fold over the previous 10 years, and a US study reported a 35-fold increase over a similar time period. This increasing prevalence is supported by an increasing incidence; population-based data from Olmsted County, MN show that the incidence of EoE increased from 0.35 to 9.45 per 100,000 persons from 1991 to 1995 to 2001 to 2005. Similarly, a population-based study of pediatric and adult patients with EoE in Switzerland reported increased incidence from 2 to 6 per 100,000 persons from 1989 to 2004, and findings from the Netherlands reported an increased incidence from 0.01 per 100,000 in 1996 to 1.31 per 100,000 in 2010. This worldwide increase in EoE mirrors the trends in atopic diseases, although whether this is caused by a true increase in disease incidence, increased disease recognition, or improved access to endoscopy remains unclear.
Beyond these parallels in disease incidence, several clinical studies have reported significant comorbidity of EoE with other atopic diseases. Although the prevalence of comorbid atopic disease varies between these studies, perhaps because of regional population differences, they show that most (50%–80%) patients with EoE also have atopy and other allergic diseases, including rhinitis, asthma, and eczema. Although these allergic diseases can commonly be seen to follow the atopic march, whereby early in life, appearance of eczema and food allergy are observed and predispose for rhinitis and asthma several years later. EoE has a profoundly broad age range of onset and so seems to not adhere to this established concept. However, the strong clinical links between EoE and other well-described allergic diseases seem to imply shared pathogenic mechanisms.
Hypersensitivity in EoE: immediate or delayed?
According to the National Institute of Allergy and Infectious Diseases–convened Guidelines for the Diagnosis and Management of Food Allergy in the United States , food allergy was broadly defined as an abnormal immunologic hypersensitivity response to specific food proteins that leads to adverse clinical reactions. Within this definition, EoE is clearly an example of food antigen–driven hypersensitivity. However, hypersensitivity reactions can be further classified based on the mechanisms of antigen recognition: IgE (immediate type) or the T-cell receptor (delayed type).
In immediate-type IgE-mediated hypersensitivity, typically associated with anaphylaxis or urticaria, cross-linking of IgE and its receptor by antigen leads to the rapid release of preformed mediators from mast cells and basophils. Histamine is a key molecule in this process, and histamine receptors are important in regulating the physiologic responses. This early antigen-specific rapid response is followed by the subsequent de novo synthesis and release of lipid mediators, cytokines, and chemokines, which can drive a late phase response in some individuals, with infiltration of inflammatory cells, including eosinophils.
Most patients with EoE have compelling evidence of IgE-mediated hypersensitivity to foods, as determined by increased food-specific IgE or abnormal skin prick test (SPT), despite food-induced anaphylaxis occurring in only around 15% of these patients. Mechanistically, it has been shown that IgE-bearing mast cells are increased in the esophageal mucosa of patients with EoE, particularly those who are atopic. Thus, it may be that the immediate hypersensitivity response in EoE occurs in a localized fashion exclusively in the esophagus, similar to what is seen in oral allergy syndrome. Although the involvement of IgE-mediated activation of mast cells in responses in the esophagus of patients with EoE remains to be defined, the early phase reaction could enhance blood flow and muscle contractility via release of histamine, whereas the late phase reaction could contribute to the recruitment of eosinophils, similar to processes that have been noted in allergen-induced eosinophil recruitment in atopic dermatitis.
Although the role of IgE-mediated hypersensitivity remains unclear, non–IgE-mediated reactions are increasingly understood to participate in EoE. These delayed-type reactions, often referred to as T-cell–mediated hypersensitivity, are characterized by the activation of antigen-specific T cells and subsequent recruitment of inflammatory cells. Delayed-type hypersensitivity (DTH) associated with allergic inflammatory disease is classically characterized by a Th2-predominant immune response, with increased interleukin 4 (IL-4), IL-5, and IL-13 levels, along with eosinophilic inflammation. In clinical diagnosis, patch testing, whereby antigen is applied to the skin so as to elicit a DTH-associated response, has been shown to significantly improve predictive values over SPT alone, highlighting the likely contribution of this arm of the immune response in responses of patients with EoE. The IgE-mediated and T-cell–mediated arms may intersect, because IgE has been shown to enhance DTH responses in mice.
Allergic sensitization
Dependent on IgE or T Cells?
The loss of tolerance and subsequent sensitization to antigen are critical events in the initiation of allergic conditions, involving coordinated involvement of antigen-presenting cells (APCs), T cells, and B cells, to prime the adaptive immune system for subsequent responses to antigen exposures. In particular, allergic sensitization associates with the generation of allergen-specific Th2 cells, which proliferate and differentiate into antigen-specific effector and memory T cells. In addition, these Th2 cells play a critical role in B-cell production of allergen-specific IgE, through their ability to generate IL-4.
In EoE, allergic sensitization is clearly evident: regardless of atopic status, patients with EoE have increased density of B cells and expression of IgE in the esophagus along with evidence of local class switching. Specific IgEs for foods that trigger active disease are commonly detected in patients with EoE in the absence of anaphylaxis, although they may be present at low levels, perhaps reflecting local production. Peripheral blood mononuclear cells from patients with EoE show allergen-specific cytokine responses that correlate with this increase in specific IgE (although some patients have allergen-specific cytokine responses without increased specific IgE levels, consistent with non-IgE–mediated allergic sensitization). In addition, mouse models of EoE-like disease, whereby sensitization is elicited via cutaneous or respiratory allergen exposure, show increased antigen-specific IgE levels, and a clear dependency on T cells but are still able to traffic eosinophils to the esophagus in the absence of either B cells or IgE. Thus, allergic sensitization in EoE drives the formation of allergen-specific IgE and T cells; however, they potentially have independent roles in the underlying disease pathogenesis.
Tolerance
In animal models, allergic sensitization commonly occurs to an antigen for which the animal is naive; however, in EoE, in which the mean age at diagnosis is 33.5 years, sensitization often occurs to antigens already in the diet or environment that have been tolerated. This finding suggests that loss of tolerance may be critical to facilitate the subsequent immune sensitization. Several early-life risk factors have been found for EoE, including antibiotic use in infancy, cesarean delivery, preterm birth, and lack of breastfeeding, all of which have been suggested by other studies to alter the development of tolerance. One critical cell type that seems to play a role in the maintenance of tolerance is the regulatory T (Treg) cell, characterized by expression of Foxp3. In EoE, an imbalance between effector and Treg cells was shown by Stuck and colleagues, who found the proportion of Foxp3 + CD3 + T cells was 50% reduced in EoE compared with healthy controls. Using an intranasal aeroallergen-induced murine model of EoE, Zhu and colleagues found a similar alteration in the frequency of CD4 + T-cell subsets in the esophagus of allergen-challenged mice compared with saline-challenged mice. Not surprisingly, it was FOXP3 + cells that were critically reduced among these cell populations. Although these studies suggest that there may be a relative lack of Treg-type T cells at the site of inflammation, it remains to be determined whether this alteration in Treg cells has simply a sustaining effect in EoE or whether Treg imbalances are critical to promoting allergic sensitization.
Route of Sensitization
An interesting question that remains in EoE is the site at which allergic sensitization occurs. Clinical studies of early-life risk factors have not supported a specific route of sensitization in EoE. However, a variety of animal models of allergic disease have shown that allergic sensitization can occur via the skin, airway, or gut. Experimental animal models of EoE suggest that sensitization may not require esophageal allergen exposure. Akei and colleagues found that epicutaneous antigen sensitization in the form of repeated allergen administration to the shaved back of mice followed by an intranasal allergen challenge facilitated an EoE-like disease that was IL-5 dependent and, to a lesser extent, IL-4 and IL-13 dependent. More recently, Noti and colleagues found that sensitization to egg or peanut protein could occur during skin inflammation or injury (tape stripping) in a thymic stromal lymphopoietin (TSLP)-dependent, basophil-dependent, and IgE-independent manner. Sensitization also seems to be effective via the lungs, because repeated exposure to aeroallergens via intranasal administration increased antigen-specific IgG1 levels and was sufficient to induce eosinophilic inflammation in the esophagus. Neither intragastric nor oral administration of aeroallergens was sufficient to induce EoE-like disease. It is likely, therefore, that antigen sensitization can occur at several sites, including lung and skin, and likely depends on a variety of host and environmental factors, but further studies are required to determine what is clinically relevant in EoE.
Aeroallergens
Many patients with EoE show evidence of polysensitization, not just to multiple foods but also to environmental aeroallergens. Although the functional role of allergic sensitization to aeroallergens is not entirely clear, there is evidence both in animal models and clinically that it may participate in driving elements of the EoE disease. Several studies have described seasonal variation in a subset of patients in whom disease worsening occurred during pollen season, regardless of therapy. It remains unclear whether swallowed aeroallergen directly promotes active disease or inhaled aeroallergen exacerbates concomitant airway disease, as could be interpreted from the murine studies with intranasal aeroallergen. Primary allergic sensitization to aeroallergens may also contribute to food sensitization as a result of cross-reactivity or cross-sensitization. Detection of this phenomenon has become possible with the use of protein microarrays, which allow for simultaneous assessment of specific IgE antibodies against multiple recombinant or purified natural allergen components, so-called component-resolved diagnostics. Using this technique, Simon and colleagues found IgE antibodies against food-specific allergen components were rare in Swiss patients with EoE and cross-reactive responses were common. These investigators noted that the dominant pattern of cross-reactivity was to profilins, pathogenesis-related (PR)-10 and lipid transfer proteins. These findings were validated by a group in the Netherlands, who also found that most food sensitizations in patients with EoE are a result of cross-sensitization to PR-10 proteins present in birch pollen. These proteins can pass through the esophagus intact, but are degraded in the stomach, which could limit inflammation to the esophagus. Thus, sensitization to aeroallergens, which can occur via the skin or airway, may be a significant factor in the development of EoE and explain some of the comorbidity with atopic disease.
Antigen presentation in the esophagus
Antigen presentation plays a crucial role in initiating a highly specific immune response to a foreign protein. APCs engulf, process, and show peptides coupled to major histocompatibility complex (MHC) class II peptides on their cell membrane. Cell surface costimulatory molecules also help determine whether presented antigen provokes an immunogenic or tolerogenic T-lymphocyte response. In EoE, the mechanistic understanding of this process is limited, but it seems that both professional APCs such as dendritic cells (DCs) and nonprofessional APCs such as epithelial cells play a role.
The primary professional APC in the esophagus seems to be the Langerhans cell, a type of DC found in all squamous epithelia, particularly the epidermis. Langerhans cells of the esophagus are structurally similar to those in the skin and are located along the papillae of the lamina propria and in the suprabasal region. These myeloid DCs, identified by the surface marker CD1a, are increased in density in children with EoE compared with controls and are reduced after treatment. However, there are contradictory data from a study in adults with EoE, which found no differences in CD1a density before/after treatment or with controls. This finding may be because of histopathologic variability or may represent a key etiologic difference between children and adults with EoE. In addition, Langerhans cells of the upper GI tract express the high-affinity IgE receptor, FcεRI, for which expression increases in active EoE. IgE signaling on APCs has been proposed to enhance antigen uptake and enhance the development and activation of allergen-specific T cells. Although the role of Langerhans cells in EoE is unclear, they are critical to the pathogenesis of atopic diseases such as eczema, and their proximity to T cells in the esophagus suggests a possible interaction and pathologic function.
Under pathologic conditions, epithelial cells at mucosal surfaces act as nonprofessional APCs and can regulate immune responses at the site of exposure. Antigen presentation by small bowel epithelium is well established and likely plays a role in food hypersensitivity. Mulder and colleagues found that basal epithelial cells in EoE biopsies express the MHC class II protein HLA-DR. Using the human esophageal epithelial HET-1A cell line, which maintains characteristics of basal esophageal epithelium, these investigators showed the ability of these cells to engulf, process, and present antigen in an interferon γ (IFN-γ)-dependent manner, as well as stimulate T-helper cell activation. IFN-γ, which is increased in biopsy tissue of patients with EoE, enhanced expression of MHC class II, whereas IL-4 enhanced costimulatory molecule expression. Thus, although the esophageal epithelium is unlikely to play a role in the early initiation of an immune response, given the dependence on cytokine priming, it may play a role in perpetuating EoE-associated inflammation.
Controversially, recent literature has suggested that eosinophils may also function as APCs at the site of inflammation. In asthmatics, eosinophils seem to express the MHC class II protein, HLA-DR, dependent on stimulation by granulocyte-macrophage colony-stimulating factor, as well as costimulatory molecules CD40, CD80, and CD86, and can traffic to regional lymph nodes after exposure to antigen, bringing them close to T cells for presentation. Likewise, in EoE, tissue eosinophils have been shown to express increased HLA-DR, as well as CD40 and CD80, supporting the potential capacity for antigen presentation. However, the evidence to support the ability of eosinophils to engulf and process protein antigens seems lacking and limits their ability to function in similar ways to professional APCs, such as DCs. One postulated mechanism, whereby MHC II loading occurs from exogenous peptides generated from the protease-rich milieu in EoE, remains to be fully established. In addition, the nature of the interaction between T cells and antigen-presenting eosinophils is not well understood; thus, the extent to which eosinophils act as APCs to initiate or even sustain the inflammatory process is unclear.
T-lymphocyte immune responses
Murine studies on mice lacking various components of the adaptive immune system have established a critical role for T cells in EoE. Similar to other atopic diseases, such as allergic asthma and eczema, tissue inflammation in EoE is characterized by a Th2-type inflammatory response. This finding was initially described by Straumann and colleagues in 2001, who observed increased T cells and IgE + mast cells in esophageal biopsies of patients with EoE associated with IL-5 expression in the infiltrating inflammatory cells. Since that initial study, several reports have confirmed these findings and described increased levels of IL-4 and IL-13 in biopsy samples. Peripheral blood mononuclear cells from patients with EoE also produce IL-5 and IL-13 in response to specific allergen stimulation, and murine models of EoE have provided additional support that a Th2-mediated response is required for pathogenicity. Blanchard and colleagues examined a large cohort of patients with EoE and found enhanced expression of both IL-4 and IL-5 in atopic individuals with EoE compared with nonatopic individuals. In addition, the study noted concerted expression of IL-5 and IL-13, suggesting that a common cell type is responsible for their production.
Although much of the focus has been on the Th2 cell in EoE, Th1-associated cytokines are also an important part of the inflammatory response, and likely play a critical role in pathogenesis. This category includes tumor necrosis factor, which is expressed by esophageal epithelial cells, and is involved in remodeling, and IFN-γ, which is expressed by T cells after stimulation with IL-15 and is involved in priming the epithelium for antigen presentation.
IL-4
Critical to the initiation of a Th2 response, IL-4 promotes differentiation of naive T-helper cells into Th2 cells as well as B-cell class switching to produce IgE. Although the initial source of IL-4 in atopic disease is not entirely clear, recent work has proposed that TSLP-elicited basophils may be important. In the esophagus, IL-4 not only sustains the Th2 response but also contributes directly to recruitment of eosinophils by stimulating eotaxin production in esophageal epithelial cells. Th2 cells are considered to be an important source of IL-4 in EoE, which is enhanced by IL-15 stimulation. Countering this concept, IL-4 expression levels correlate poorly with the other Th2-associated cytokines, IL-5 and IL-13, and suggest that other cells may be relevant sources.
IL-5
Of the Th2 cytokines, IL-5 is the most well studied in EoE and seems to be central to the disease. In the GI tract, IL-5 + allergen-specific T-cell responses differentiate EGIDs from IgE-mediated immediate hypersensitivity, characterized by IL-5 − Th2 responses. IL-5 is produced primarily by Th2 cells, although additional sources include mast cells and eosinophils. It acts on the bone marrow to stimulate eosinophil proliferation and differentiation and regulates survival and activation of eosinophils. Mice lacking IL-5 fail to recruit eosinophils to the esophagus in intranasal aeroallergen-induced EoE, whereas mice transgenic for IL-5 under the control of the T-cell–specific CD2 promoter (CD2-IL5) develop chronic esophageal eosinophilia and mastocytosis. Trafficking of eosinophils by IL-5 likely occurs by priming of eosinophil responses to chemokines such as eotaxins (CCL11, CCL24, and CCL26), or by upregulating homing receptors. IL-5 also has a role in tissue remodeling, because mice with CD2-IL-5–mediated esophageal eosinophilia have increased collagen accumulation in the lamina propria and extended stromal papillae, whereas IL-5–deficient mice do not in similar model studies. Although clinical and murine studies has shown a central role for IL-5 in the allergic mechanisms of EoE, and there is variable downregulation of IL-5 expression after treatment with fluticasone, anti-IL-5 biological therapy has shown limited clinical efficacy, suggesting that, although IL-5 plays an important role in EoE disease, blocking IL-5 may not be sufficient to prevent the immunopathology of EoE in humans.
IL-13
IL-13 is a pleotropic cytokine that exerts pathologic effects when excessively produced by activating local tissue inflammatory responses; its cellular sources include Th2 cells and activated eosinophils. In EoE, active inflammation is self-perpetuating, because infiltrating eosinophils secrete IL-13, which acts to enhance further recruitment of eosinophils to the esophagus by inducing STAT6-dependent eotaxin expression in the epithelium. The critical interaction between IL-13 and the esophageal epithelium in driving EoE was highlighted by Blanchard and colleagues, who found a significant overlap between the transcriptome of primary esophageal epithelial cells treated with IL-13 and total RNA from biopsies of patients with EoE, which was reversible with steroid therapy. IL-13 also recruits eosinophils by promoting fibroblasts to produce periostin, which increases eosinophil adhesion to fibronectin. Intratracheal IL-13 was sufficient to induce experimental EoE in mice in an eotaxin/IL-5/STAT-6–dependent manner but was not required, because intranasal aeroallergen-induced EoE had only mildly reduced esophageal eosinophilia in IL-13–deficient mice, which may be caused by exaggerated Th17 responses. IL-13 also plays key roles in barrier function, by downregulating genes involved in the epithelial cell differentiation, such as desmoglein-1, filaggrin, and involucrin, and eosinophil-independent tissue remodeling by promoting collagen deposition, angiogenesis, and epithelial hyperplasia. IL-13 is a prominent T-cell mediator in EoE pathogenesis with broad function and may be predicted as a likely factor that prevented efficacy from targeted IL-5 therapy. Ongoing clinical trials with anti-IL-13 will serve to better understand its role in pathogenesis and relevance as a therapeutic target.
Related posts:
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
