Etiology of Epithelial Barrier Dysfunction in Type 2 Inflammatory Diseases

Abstract

Epithelial barriers of the skin, gastrointestinal tract and airway serve common critical functions such as maintaining a physical barrier against environmental insults and allergens, as well as providing a tissue interface balancing the communication between the internal and external environments. We now understand that in allergic disease, regardless of tissue location, the homeostatic balance of the epithelial barrier is skewed towards loss of differentiation, reduced junctional integrity and impaired innate defense. Importantly, epithelial dysfunction characterized by these traits appears to pre-date atopy and development of allergic disease. Despite our growing appreciation of the centrality of barrier dysfunction in the initiation of allergic disease, many important questions remain to be answered regarding the mechanisms disrupting the normal barrier function. Although our external environment (proteases, allergens, injury) is classically thought of as a principal contributor to barrier disruption associated with allergic sensitization, there is a need to better understand contributions of the internal environment (hormones, diet, circadian clock). Systemic drivers of disease, such as alterations of the endocrine system, metabolism and aberrant control of developmental signaling, are emerging as new players in driving epithelial dysfunction and allergic predisposition at various barrier sites. Identifying such central mediators of epithelial dysfunction, using both systems biology tools and causality-driven laboratory experimentation, will be essential in building new strategic interventions to prevent or reverse the process of barrier loss in allergy.

I. Anatomy and physiology of normal barrier - the Immune Barrier

The epithelial barrier is diverse, occurring, inter alia, in the gastrointestinal tract, the urogenital tract, the respiratory system, the eyes and the skin. Epithelial cells are specialized in each of these organ systems and in the various geographic regions of each system. The main functions of epithelia are to present a physical and immune barrier, maintain a surface that accommodates commensal organisms but not pathogens, remove, degrade or neutralize environmental toxins and particulates, and maintain the balance of water and electrolytes. Specialized functions of the physical barrier include maintenance of the mucociliary escalator in airways, production of surfactants in the alveoli and small airways and production of a protective mucus blanket in the gastrointestinal and urogenital tracts. Immune Barrier refers to epithelial elements that are essential for innate immune resistance to potential pathogens, including constitutive and inducible expression of enzymes, peptides, proteins, lipids, ions and pathogen recognition receptor (PRR) systems designed to protect the host. Combined with epithelial tight junctions, these systems present a vigilant and effective barrier to microbial invasion. Many allergic diseases are now known to manifest clear disorders of the immune barrier. The purpose of this review is to discuss the presence of dysfunction of barrier, its role in allergic disease, and the molecular and endocrinological mechanisms that are underlying causes.

II. Barrier defects in type 2 diseases

Introduction

When the immune barrier is predisposed to disruption, microorganisms and antigens can gain access between epithelial cells through the basement membrane to the underlying lamina propria and connective tissue. Penetration of microbes triggers strong innate immune responses by PRR on epithelial cells and immune cells such as macrophages, ILCs and mast cells, residing among epithelial cells or on the basolateral side of the lamina reticularis. Subsequent sensitization and activation of adaptive immune responses can result, including type 2 immune responses of relevance to the allergic diathesis. Barrier loss can result from defects in several essential components, including tight junction proteins, protective antiproteases, structural elements such as filaggrin in the skin, expression of antimicrobial products, transport of ions, protons, water or antimicrobial materials and other mechanisms. Associated with loss of barrier and sensitization is also the profound activation of sensory nerves important in manifestation of disease experienced by the patient. In this section we briefly discuss some of the salient mechanisms and features of barrier loss commonly shared by all allergic diseases (summarized in Figure 1).

Figure 1. Key features of epithelial dysfunction common to all allergic disease.

Epithelial barriers predisposed to Type 2 allergic disease are characterized by increased permeability and aberrant behavior of morphogenetic programs that maintain epithelial homeostasis. Both exogenous (inhaled allergens, respiratory viruses, chemical sensitizers, air pollutants) and endogenous (hormones, dietary factors, altered circadian clock) disruptors of epithelial homeostasis may drive predisposition to allergic sensitization by altering homeostatic activity of developmental pathways (WNT, Notch, Hedgehog) that maintain proper epithelial-to-mesenchymal communication, epithelial differentiation, and barrier integrity. Transition to a remodeling state in disease typically follows allergic sensitization and inflammatory responses. This transition features further loss of differentiation signals, downregulation of innate defense molecules, pro-fibrotic processes, deposition of extracellular matrix, and potentiated activity of the mesenchymal unit.

Atopic Dermatitis

Atopic dermatitis (AD) is characterized by loss of barrier function of the skin culminating in a clinical phenotype characterized by formation of skin lesions. Skin barrier dysfunction is induced by disruption of the stratum corneum, a dense protein–lipid matrix, which functions as a barrier to water loss, environmental insults and allergens. Filaggrin (FLG), a filament-associated “epidermal differentiation complex” (EDC) protein essential for the regulation of epidermal homeostasis, is highly deficient in skin of many subjects with AD¹. Similarly, filaggrin-like proteins hornerin and filaggrin-2 are detected at significantly lower levels in the skin of many patients with AD, irrespective of FLG genotype². Dysregulated filaggrin is frequently discussed as central to the origins of disease³; however, barrier defects in atopic dermatitis go far beyond filaggrin deficiency. Defects in claudins 1, 4 and 8 have also been associated with the development of atopic dermatitis^4–7. Disruption of the skin barrier involves defects in the entire keratinocyte terminal differentiation program via dysregulation of multiple EDC genes on human chromosome region 1q21⁸, reduced expression of epithelial tight junction proteins⁹, increased transepidermal water loss (TEWL)^{10, 11}, and reduction in epidermal natural moisturizing factors¹². The impaired skin barrier function in AD also involves altered lipid profiles, including shortening of carbon chain length of stratum corneum lipids (ceramides and free fatty acids), an event that exhibits strong correlations with TEWL and occurs independently of filaggrin mutations^13–16. Intriguingly, clinically unaffected, non-lesional skin in subjects with AD similarly exhibits defects in terminal keratinocyte differentiation¹⁷, as well as reduced filaggrin and lipids^{14, 15}, suggesting a global cutaneous predisposition to barrier dysfunction in AD patients.

Asthma

An intact functional mucosal barrier is crucial for the maintenance of airway homeostasis. Despite the existence of multiple clinical endotypes, most forms of asthma exhibit a dysregulated epithelial barrier. Asthmatic epithelium is characterized by an increase in basal and goblet cells¹⁸ and a decrease in terminally differentiated ciliated cells, frequently accompanied by basement membrane thickening^{19, 20} and epithelial shedding with the formation of Creola bodies consisting of clusters of shed epithelium (even in mild forms of disease)^21–23. Disruption of epithelial tight and adherens junctions is typical for asthma, with marked loss of E-cadherin²⁴ and claudin-18²⁵. The underlying extracellular matrix, which supports homeostasis and repair of the epithelium, also undergoes significant remodeling, characterized by increased deposition of provisional matrix components, such as the glycoproteins fibronectin²⁶, periostin²⁷, tenascin-C^{28, 29}, hyaluronan and versican³⁰. Epithelial-to-mesenchymal transition (EMT) was proposed as a mechanism underlying epithelial dedifferentiation and perpetuated remodeling³¹; however, this remains a topic of debate, due to the lack of markers accurately defining this process in asthma³². Mucociliary clearance, which depends on the cooperation between submucosal glands, goblet cells and ciliated cells, and normal chemical barrier function of the asthmatic epithelium are also compromised. There are profound changes in relative proportions and viscosity of mucins MUC5AC and MUC5B that in turn contribute to airway obstruction^{33, 34}, and downregulation of lipoxins³⁵. Perturbations in sphingolipid balance with increased levels of ceramides have also been reported in asthmatic airway epithelium ³⁶.

Allergic Rhinitis

Mucosal epithelial barrier disruption is observed in models of allergic rhinitis and in patients³⁷. Allergens can contain proteases and have been shown to disrupt epithelial tight junctions^{38, 39}. Nasal challenge with histamine, or with antigen in a sensitized subject, can cause significant plasma exudation across the epithelial barrier. This is a non-destructive process and may be a first line defense of respiratory mucosa⁴⁰. A proteomic study detected changes reflecting altered epithelial permeability in patients with allergic rhinitis, including increased α2 macroglobulin, a measure of vascular and epithelial leak, and decreased levels of the protective antileukoproteinase SLPI⁴¹. Reduced levels of barrier proteins E-cadherin and zonula occludens-1 (ZO-1) were observed in nasal mucosal tissue from allergic rhinitis patients compared to healthy controls using real time PCR, Western blot and immunohistochemistry⁴². In vitro, IL-4 and TNF induced loss of these markers in epithelium. Steelant et al. observed decreased barrier function measured in vitro using primary epithelial cells collected in patients with allergic rhinitis compared to healthy controls and found decreased expression of the tight junction proteins occludin and ZO-1⁴³.

Chronic rhinosinusitis

Early studies provided evidence for defective barrier in patients with chronic rhinosinusitis (CRS) using immunohistochemistry, bioelectric evaluation and ion transport function as indicators^{44, 45}. Using air-liquid interface epithelial cultures, Soyka et al. showed reduced barrier function using tissues collected from patients with or without CRS⁴⁶. A number of studies have focused on the loss of epithelial tight junctions in CRS patients by assessing levels of individual tight junction proteins^{37, 47}. Alterations of expression of ZO-1 and E-cadherin were reported by Jang et al.⁴⁸. Epithelial damage, including shortening of desmosomes, was reported by Shahana et al. in a study using electron microscopy in polyp tissue derived from asthmatic patients⁴⁹. Reductions of the tight junction proteins claudin-1 and occludin in CRS tissues were reported in another study⁵⁰. The group of Bachert reported that ZO-1, occludin and E-cadherin were all reduced in mature polyps removed from CRSwNP patients⁵¹. Moreover, aquaporin 5, a marker of epithelial differentiation, was significantly reduced in sinonasal samples of CRSwNP subjects compared to CRSsNP or controls⁵². Physical evidence of disruption of epithelial barrier in CRS includes prominent acanthosis and acantholysis⁴⁷. The loss of differentiation and acanthosis may result from a cycle of ongoing injury and repair associated with epithelial to mesenchymal transition (EMT). Factors that have been identified in tissues from CRS patients that might stimulate barrier loss and acanthosis include TGFα, oncostatin M, epiregulin, hypoxic activation of HIF1α and endocrine deficiency (see below)^{47, 53–55}.

Eosinophilic Esophagitis (EoE)

The esophagus is composed of stratified squamous epithelium remarkably similar to the epithelium of the skin, only differing in its lack of a cornified layer and its possession of a mucous layer⁵⁶. As occurs in other Type 2 diseases, the normal structure of the epithelium is disrupted in EoE, including basal cell hyperplasia, dilated intracellular spaces and impaired barrier function and cell junctions^{57, 58}. These changes are thought to be associated with dysregulated epithelial differentiation and impaired epithelial barrier formation. As in the skin, epithelium in EoE is characterized by dysregulation of genes of the epidermal differentiation complex on human chromosome 1q21, including filaggrin, involucrin and several small proline-rich repeat family members⁵⁹. In esophageal epithelial cells, filaggrin was shown to be negatively regulated in response to IL-13 ex vivo⁵⁹. Desmosomal and tight junctional proteins desmoglein-1, E-cadherin and claudin-1 are reduced in active EoE^{60, 61}. Epithelial barriers in eosinophilic esophagitis also exhibit characteristic features of EMT, which correlate with subepithelial fibrosis and eosinophil counts in human biopsies⁶².

III. Mechanisms of barrier defects

Introduction

Barrier defects can be induced by loss or defects in key proteins that comprise tight or adherens junctions, disruption of barrier by environmental exposures (including proteases, chemical injury or trauma), inflammatory responses that induce barrier disrupting Th2 cytokines and endogenous mechanisms such as altered central metabolism and imbalance of hormones that regulate epithelial homeostasis. The purpose of this section is to briefly consider some of the external and internal factors that are thought to be of particular importance in the disruption of barriers in allergic diseases.

Genetic and epigenetic barrier defects

Genome-wide association studies (GWAS) and positional cloning have successfully identified several risk alleles and loci reproducibly associated with atopic dermatitis, asthma and eosinophilic esophagitis. Interestingly, the majority of allergic susceptibility candidate genes control epithelial barrier homeostasis. Null gene mutations of filaggrin (FLG) (as mentioned previously, an important component of terminal keratinocyte differentiation) are the most significant known risk factor for atopic dermatitis and eosinophilic esophagitis, along with mutations of other epidermal differentiation cluster (EDC) genes⁵⁹. Polymorphisms in asthma susceptibility genes ORMDL3, GSDMB, PCDH1, CDHR3, ADAM33, SMAD3, IL1RL1 and IL18R1 are now thought to be linked to aberrant epithelial remodeling, the unfolded protein response and lipid biosynthesis^63–65. A recent pathway-based association study by Barreto-Luis et al.⁶⁶ revealed novel asthma susceptibility loci near WNT pathway genes that regulate barrier morphogenesis. Even though genetic associations are stronger within clinically well-defined subgroups of disease⁶⁷, overall allergic disease-associated alleles typically have small effect sizes and cannot account for the rapidly rising prevalence of allergy⁶⁸, reinforcing the significance of environmental factors in development of allergic diseases. Significantly, environmental inputs can affect gene transcription via heritable epigenetic regulation that does not require alterations in gene sequence, including DNA and histone modifications, as well as changes in noncoding RNAs⁶⁹. One of the notable epigenetic changes in asthmatic epithelium involves hypo-methylation of KRT5⁶⁹, which increases keratin 5 expression in basal epithelium and may therefore lead to dysregulated epithelial differentiation^70–72. The potential for epigenetic “reprogramming” is evident when asthmatic epithelium cultured in normal media still retains persistent indicators of defects in junctional maintenance, “immaturity” and repair ex vivo⁷³. The topic of epigenetic mechanisms of epithelial remodeling is a still-emerging area of investigation of epithelial barrier dysfunction, and ongoing work is likely to produce novel insights to aberrant regulation of epithelial differentiation programs in allergic disease.

Environmental contributions to barrier loss

Barriers of the skin, airways and gastrointestinal tract constantly experience biological and chemical insults from the surrounding environment. Allergens, such as house dust mite or pollens, are capable of disrupting physical integrity of the barrier via their protease activity, which degrades adhesion proteins and triggers epithelial alarmin cytokine response^74–78. Respiratory viruses promote airway epithelial barrier dysfunction by disrupting epithelial junctions, which may represent an inciting or sustaining event linking viral infections and allergic inflammation^{79, 80}. Interestingly, asthma susceptibility barrier gene CDHR3 is a receptor for rhinovirus C⁶⁵. Many chemicals and irritants prevalent in industrial environments act as adjuvants and sensitizers, disrupting normal functions of the epithelial barrier, setting off alarmin responses and promoting allergic sensitization^{81, 82}. Air pollution associates with the development of atopic dermatitis in both children and adults^{83, 84}. Less obvious, but equally important, environmental contributors to barrier loss do not disrupt barrier externally, but act as endogenous factors in dysregulation of epithelial homeostatic processes. Stimulation of the aryl hydrocarbon receptor (AHR) by ingested or inhaled xenobiotics, such as polycyclic aromatic hydrocarbons, is now known to affect epidermal differentiation and skin barrier formation⁸⁵. Xenoestrogens linked to development of allergy, such as bisphenols and phthalates, mimic the natural action of nuclear hormonal receptor ligands (estrogens, androgens) and thus disrupt normal epithelial homeostatic processes^86–88. A multitude of dietary factors have the capacity to alter epithelial behavior via direct receptor action or indirect regulation of tissue metabolism. For example, Fischer et al. demonstrated that vitamin D supplementation, acting via the vitamin D receptor pathway, reduces EMT processes and improves barrier function in several clinically relevant murine models of asthma⁸⁹. Fatty acid-deficient diets spontaneously induce skin barrier disruption via alteration of skin metabolism^{90, 91}. Moreover, the intestinal microbiome is now emerging as an important regulator of metabolism with consequences for epithelial and immune homeostasis⁹².

Defects in hormonal signaling as early events

Contributions of biological systems other than the immune system to disruption of epithelial homeostasis and priming of allergic response are currently poorly understood. Several studies report insulin resistance in children and adults with asthma^{93, 94}, as well as association of asthma and atopic eczema with pre-diabetes and metabolic syndrome⁹⁵. We found profound changes in serum hormonal profiles of pre-pubertal non-obese allergic children compared to non-allergic controls, as well as chronic rhinosinusitis patients compared to healthy controls, including decreased levels in serum insulin and increased output of thyroid and growth hormones⁹⁶. Hormones play a critical role in maintenance of epithelial barrier homeostasis and integrity via their integration with epithelial morphogenetic programs^{88, 97}, evidenced by deficiencies in wound healing and epithelial dysfunction in patients with thyroid disease and diabetes^{98, 99}. Insulin and IGF-1 are essential drivers of epithelial differentiation and regulators of energy metabolism. Double transgenic mice lacking both insulin and IGF-1 receptors are characterized by severe defects in epithelial differentiation, severely impaired stratification of the epidermis, spontaneous lesions and overall loss of skin barrier function¹⁰⁰. Sex steroids, and estrogen in particular, are recently receiving more attention for their potential to explain sex bias in prevalence of allergic disease¹⁰¹. Glucocorticoids have profound effects on epithelial differentiation^{102, 103} and exemplify the therapeutic potential of manipulating the endocrine system in treatment of allergic disease.

Disruption of normal epithelial development

The significance of early life events in initiation and propagation of allergic disease is now widely recognized. The first two years of life represent a window of susceptibility for development of asthma. Importantly, epithelial barriers in childhood are shaped and regulated by active on-going developmental programs. In particular, morphogenesis of lung epithelial barriers continues throughout normal postnatal development, characterized by ongoing septation and epithelialization of the lung¹⁰⁴. Consequently, perturbation of barrier morphogenesis in childhood may have a lasting effect on adult epithelium via alteration of developmental programs and epigenetic reprogramming at developmental checkpoints^105–107. Multiple lines of evidence now implicate active re-engagement of morphogenetic programs in adult disease, typically not seen in adult homeostatic tissue^108–110. Alterations of the Wnt⁶⁶, Hippo¹¹¹, Notch/Jagged^{112, 113} and Hedgehog¹¹⁴ developmental pathways all exhibit strong association with epithelial remodeling and allergy. Early disruption of normal barrier development frequently pre-dates atopy^{11, 44} and is emerging as central to initiation of the “atopic march”. Mouse models of the atopic march demonstrate conclusively that sensitization to allergens via disrupted skin barriers is sufficient to elicit an immune response at other barrier sites^{115, 116}. Causes of disruption of epithelial morphogenesis early in life are not well understood. Early life viral exposures^{117, 118} and changes in microbiota¹¹⁹ are some of the better understood factors driving aberrant epithelial responses and a predisposition to allergic sensitization.

The Injury-Repair cycle and EMT

Exposure to environmental factors causes injury or disruption of normal homeostasis of the epithelium followed by engagement of robust repair processes to minimize further damage from microbes and environment. Mature differentiated epithelial cells undergo an epithelial to mesenchymal transition (EMT) in which the cells lose their attachments to basement membrane and each other, lose their polarity, begin to divide and become migratory in order to rapidly cover the injured area. During this process, epithelial cells lose expression of tight junction proteins ZO-1, occludin, E-cadherin and other markers of mature differentiated epithelium. The mesenchymal cells derived from the basal epithelial cells begin to produce desmin, fibronectin, tenascin, laminin, collagens and others proteins that form a provisional matrix to protect the exposed basement membrane or lamina propria and express differentiation markers such as α-smooth muscle actin. Evidence for EMT usually consists of loss of tight junction proteins, gain of provisional matrix proteins or biomarkers, or both. Early studies by Davies, Holgate, Hackett, Knight and others found evidence for persistence of EMT in asthma^120–123, although it is more often described as a mild, “partial EMT” phenotype, which could contribute to poor expression of conventional EMT markers in asthmatic airways³². More recent studies found similar evidence for increased EMT in allergic rhinitis^{42, 43}. Several studies have demonstrated ongoing EMT in CRS^{37, 47, 49–51}. Two studies have demonstrated ongoing EMT in eosinophilic esophagitis^{62, 124}. However, to date, the authors have not found any reports of EMT (as classically defined by developmental biologists) in atopic dermatitis, despite abundant evidence for loss of epithelial differentiation programs in the AD skin. Most chronic type 2 inflammatory diseases thus often appear to have a chronic EMT-based ongoing injury-repair cycle.

Inflammation and barrier - a vicious cycle

Whether inflammation is the primary cause of barrier disruption or whether it is barrier dysfunction that leads to sensitization and aberrant inflammatory response poses a long-standing “chicken-egg” dilemma (Figure 2). Aberrant innate immune activity at barrier sites favors sensitization to innocuous antigens and if there is an ensuing adaptive immune response, it can lead to a full-scale inflammatory attack on epithelial cells. During chronic inflammation associated with repeated or persistent antigen exposure, inflammation can lead to persistent disruption of epithelial junctions, as well as epithelial remodeling and sub-epithelial fibrosis driven by chronic perpetuated repair. Conversely, an accumulating body of evidence points to a causal role for barrier dysfunction as a primary driver of the allergic response and a vicious cycle of barrier leak^125–127. Strikingly, only 30% to 40% of cases of allergic disease (asthma, eczema, allergic rhinitis) in early childhood are attributable to atopy and 60% to 70% of cases result from yet unknown factors¹²⁸. In most “atopic march” cases, increased skin permeability at birth and non-atopic eczema frequently predate development of atopy^{11, 127}, regardless of filaggrin mutations¹⁰. Moreover, although the atopic march assumes that one allergic disease brings risk of acquisition of another allergic disease later in life, it may involve sensitization to antigens different from those that triggered the initial disease^{129, 130}. From this perspective, an alternative view might be that atopy and allergic inflammation are both secondary to a yet-unknown systemic process progressively afflicting epithelial barriers of the skin, gut and airway. Based on our observations of hormonal imbalance in allergic children (described above), we hypothesize that dysregulation of the endocrine system could represent a missing systemic link underlying susceptibility to initial barrier dysfunction. Remarkable similarities in the biology of barrier dysfunction across different allergic diseases (Figure 1), including upper and lower “unified” airway responses and non-lesional defects in clinically unaffected skin of atopic dermatitis patients, all support the existence of such a systemic trigger of disease^131–133. Transgenic mouse models with defects in proteins maintaining epithelial homeostasis, such as filaggrin, are also characterized by spontaneous allergic sensitization and enhanced allergic inflammation^134–137. This notion of causality of a pre-existing barrier problem fits well with an emerging view of Type 2 immunity as a “restorative” response evolved to maintain tissue homeostasis and assist with repair and remodeling^{138, 139}. For example, Huang et al.¹⁴⁰ provide supporting evidence suggesting that Th2 immune responses and homeostatic eosinophil activity¹⁴¹ may have evolved not to expel parasites, but rather to remodel tissue that contains parasitic infestation and minimize potential damage to the host.

Figure 2. Changing states of the epithelial barrier on a homeostasis-allergic disease continuum.

Dysregulation of normal epithelial barrier function is a gradual process involving multiple causal factors. This includes genetics, epigenetics and influences of the external and internal epithelial environments.

IV. Identifying key drivers of defects and the basis for predisposition to barrier disease

The existence of a dilemma about whether it is barrier loss or sensitization that is the primary cause of this vicious cycle illustrates well the complexity of allergic pathogenesis (Figure 2). In health, multiple biological systems and processes must work in unison to maintain tissue homeostasis at barrier sites. This includes morphogenetic/developmental programs, metabolism, the endocrine and the immune systems. There is a growing number of molecules that we now believe are of critical importance in regulating barrier biology; however, despite exhibiting strong phenotypes in models of disease, they may not be central drivers of disease pathogenesis. Filaggrin, for example, is only one of many molecules in epidermal differentiation complex (EDC)¹⁴². Knock-down of filaggrin alone was not sufficient to affect lipid composition and permeability in an ex vivo skin model¹⁴³. Several other EDC molecules play an equally critical role in epithelial homeostasis and are frequently downregulated with filaggrin in AD, independently of filaggrin genotype or atopic status⁸. Such changes in differentiation status of cells are intricately intertwined with changes in tissue metabolic demands¹⁴⁴, aberrant immune responses¹⁴⁵ and epigenetic rewiring¹⁴⁶. The question then becomes which non-genetic factors have capacity to dysregulate the entire EDC cluster, affecting the balance of the whole system? Identifying such central mediators is the key to finding strategies to restore the homeostatic state of the epithelial barrier. The endocrine system and changes in central metabolism have the potential to be central drivers of the predisposition to barrier defects, given the systemic regulatory effects of hormones (insulin, EGF-1) on epithelial morphogenetic programs of all barrier sites^147–149, and their regulatory roles for metabolism and immunity. Our understanding of central mediators of allergic disease is currently impeded by use of reductionist approaches and limited focus on processes of the immune system, insufficient understanding of complexity of biological interactions that control normal epithelial homeostasis, the dynamic nature of biological systems and the extreme clinical heterogeneity of asthma and other allergic diseases. Studies of systems and integrative biology address exactly these challenges, and are now emerging as critical in advancing all branches of biomedical science. Next generation sequencing technology, shared availability of “big data” and advances in computational biology are new promising tools for discovery of novel drivers of barrier dysfunction^{150, 151}.

V. Targets for therapeutic intervention

There are severe forms of the diseases discussed in this review, asthma, allergic rhinitis, chronic rhinosinusitis, eosinophilic esophagitis and atopic dermatitis, that are difficult to manage and for which there is a serious unmet medical need for new and more effective medications. At present, one of the most highly effective classes of drugs for these diseases is topical glucocorticosteroids. These drugs improve elements of innate immunity as well as barrier function, as measured by a number of criteria, including markers of tight junction presence and function. Although the benefit of corticosteroids derives from suppressing the inflammatory response, e.g. expression of IL-13 and other cytokines, as well as direct effects on epithelial cells to promote barrier integrity, the relative contribution of these two effects is not clear. New classes of drugs are rapidly emerging to help manage disease in patients for whom glucocorticosteroids are inadequate to control disease. This includes small molecule drugs that block signaling pathways (CRTH2, GATA3, Jak and other kinases, etc.) and monoclonal antibodies that block cytokine signaling or eliminate key allergic effector cells such as eosinophils and mast cells (IL-5, IL-13, IL-5Rα, IL-4/IL-13Rα, Siglec-8 etc.). The pursuit of barrier restorative therapies is another emerging field¹⁵². Based on the hypothesis advanced here that endocrine defects are important in disease pathogenesis, it will be worthwhile to test manipulation of signaling of other hormones (insulin, growth hormone, thyroid hormone etc.), seeking benefit for patients.

VI. Summary and conclusions

Many important questions remain to be answered regarding the mechanisms by which the epithelial barrier becomes disrupted in allergic diseases. We also need to better understand the role that barrier loss plays as both an initiator and as a consequence of the inflammatory process. We need to know more about genetic and epigenetic factors that cause barrier loss, as well as the environmental triggers and the cellular and molecular mediators and targets of the process. Identifying the central mediators of the response and the consequences of barrier dysfunction in disease will be essential in building a rationale and in strategic design of new interventions to prevent or reverse the process.