This group includes celiac disease (CD), food protein-induced enterocolitis syndrome (FPIES), food protein-induced enteropathy (FPE), and food protein-induced allergic proctocolitis (FPIAP)
[1][33]. CD is an immune-mediated disorder triggered by dietary gluten, a protein found in cereals such as wheat, rye, and barley. CD is strongly dependent on the genetic background. The disorder is characterized by a small intestinal enteropathy leading to GI as well as extra-GI manifestations and the production of auto-antibodies besides anti-gliadin antibodies, such as anti-endomysium and anti-tissue transglutaminase (TTG) antibodies
[90]. The measurement of serum anti-TTG antibodies is very useful for the diagnosis and follow-up of these patients, since they disappear in patients following a gluten-free diet. The seroprevalence of CD was recently estimated at 1.4% worldwide, ranging from 1.1 to 1.8% across geographical areas
[91]. Besides CD (not further discussed in this review), this group of food protein-induced diseases have characteristically an early onset (within the first year of life) and GI clinical manifestations. Despite the significant, often dramatic clinical manifestations during the acute phases, the prognosis is favorable, with the majority of patients resolving by age 3–5 years. In acute FPIES, repetitive vomiting, lethargy, and paleness appear from 30 to 240 min after taking the triggering food, which is most commonly cow’s milk, soy, cereals, fish, eggs, poultry/meats, fruit, legumes, or other vegetables. Reactions to multiple foods are not uncommon. Diarrhea can also follow 5 to 10 h later, although this is a less common feature (25–50%). Symptoms may be quite severe, with up to 15% patients experiencing hemodynamic instability. Chronic FPIES typically occurs with persistent exposure to cow’s milk or soy-based formula, and it presents with chronic watery diarrhea (occasionally with blood or mucus), intermittent emesis, abdominal distension, and poor weight gain
[92][93][94][95]. Although FPIES generally occurs in early infancy, adult-onset FPIES is now also being increasingly recognized, most frequently triggered by seafood
[96][97][98][99]. Finally, a rare occurrence of symptomatic fetal and neonatal FPIES has recently been reported, due to intrauterine sensitization
[100][101]. In FPE, symptoms develop in infants shortly after the introduction of cow’s milk in the diet, with chronic diarrhea and features of malabsorption such as steatorrhea and failure to thrive. Vomiting is also frequently reported. FPE is usually transient and typically resolves by 1–2 years of age, as in the case of FPIAP, although in the latter case there is an increased risk of functional GI disorders. FPIAP most often occurs in exclusively breastfed infants within the first weeks of life, because of indirect exposure to maternal dietary protein via breastmilk, although direct feeding can also trigger symptoms. These infants present with bloody, loose stools, sometimes with mucus, but the infants generally appear in well-being
[102].
2.3.1. Diagnostic and Therapeutic Management of Non-IgE-Mediated Food Protein-Induced Allergy
Diagnosis of food protein-induced diseases relies, for the most part, on the clinical picture with the exception of FPE, in which histological confirmation is usually required when young infants (<9 months) present symptoms of vomiting and intestinal malabsorption
[103]. The diagnosis of FPIES and FPIAP relies on the appreciation of a constellation of concordant symptoms coupled with their resolution upon dietary restrictions of the offending food. An oral food challenge should be strongly considered when only a single episode has occurred, or when the causative food remains elusive, preferably with the documented recurrence of symptoms when foods are re-introduced. Two therapeutic strategies can be adopted, depending on the severity of the symptoms and on the quality of triggering foods: a “bottom-up approach”, in which only causal foods are eliminated, and a “top-down approach”, warranted in most severe cases where failure to thrive and dehydration are prominent. This latter approach consists of an initial avoidance of a wide variety of foods, sometimes starting with an elemental diet, followed by the sequential reintroduction of individual foods. Extensively hydrolyzed cow’s milk formula and amino acid-based formula (AAF) may be useful long-term management strategies for infants with IgE- or non-IgE-mediated cow’s milk allergy (but only 10–20% of the latter patients require AAF). Usually 50% of infants with FPIES caused by cow’s milk develop tolerance by 1 year and 90% by 3 years
[104]. FPIES caused by solid foods appears to persist for longer
[92]. Sometimes FPIES can arise during breastfeeding; in these cases, if a food trigger is identified, the mother has to follow an elimination diet.
2.3.2. Nutritional Concerns in Non-IgE-Mediated Food Protein-Induced Allergy
The three most common foods causing FPIES are milk, soy, and rice. Rare triggers include other cereals and legumes (peanut, green pea, string bean), sweet potato, squash, carrot, egg white, chicken, turkey, fish, and banana
[105]. When milk is the trigger food, it is important to supplement calcium and vitamin D. FPIAP is also typically induced by cow’s milk protein, which requires its elimination from the child’s diet or from the mother’s diet when breastfed infants are affected. Spontaneous resolution is acquired within 1–2 years of age. The most common triggers for FPE are cow’s milk, soy, and rarely chicken, rice and fish
[1]; this condition rarely persists beyond 3 years of age. Therefore, a strict surveillance for potential nutritional issues is only required for a limited period, and during follow-up visits it is key to address unnecessary restrictions of milk and dairy products that could further compromise health and quality of life, which is above and beyond the psychological price imposed by the prescribed dietary restriction
[106][107].
2.4. Pathophysiology of Immunologic Adverse Reactions to Food
Despite intensive research efforts over the past several years, the management of allergic diseases and chronic inflammatory conditions, which are on a steady rise in both prevalence and severity, still faces unmet challenges. Central to the pathogenesis of these diseases is the development of a T helper (Th) 2-biased, allergen-specific immune response, characterized by IgE synthesis, eosinophilia, and target organ hyperresponsiveness, and it results from a complex interplay of genetically controlled and environmental factors (). The hygiene hypothesis, invoked to explain the disproportionate rise in prevalence of allergic and other inflammatory disorders over the past 40 years, provides a conceptual framework to understanding how a modified environment may pave the way for abnormal, imbalanced immune reactivity in predisposed individuals
[108]. However, the factors and pathways mediating predisposition to allergic diseases, collectively referred to as atopy, remain elusive. It is thought that the sign and strength of an immune response to a given antigen (Ag) may reflect an intrinsic, predetermined bias in Ag-specific T cells, yet there is no definitive evidence in support of this theory. In fact, evidence suggests that exposure of non-allergic individuals to allergen may not result in allergen-specific protective, Th1-directed responses, and may instead result in specific tolerance or no response at all
[109]. This points to the involvement of an additional or alternative checkpoint(s) in Ag recognition, which may affect immune responses by controlling Ag availability and processing.
Figure 3. Regulation of immune tolerance in the gut mucosa. Upon processing of dietary fibers, bacterial metabolites, such as short chain fatty acids (SCFA) and retinoic acid (RA), direct the development and function of FoxP3+ Treg cells via the interaction with gut epithelial cells and tolerogenic dendritic cells (DCs) with naïve CD4+ T cells. The activation and expansion of Treg cells promote the production of the immune regulatory cytokine, IL-10, which foster IgG1 to IgG4 B-cell class switching. Allergen-specific IgG4 B cells produce high-affinity antibodies for food allergens, preventing allergen interactions with mast cell-bound IgE. Microbiota-delivered factors, such as tryptophan-indole catabolites, may directly activate ROR-γt+ type-3 innate lymphoid cells (ILC3), via the aryl-hydrocarbon receptor (AhR), and induce the production of IL-22, a cytokine promoting gut epithelial regeneration and barrier integrity. Conversely, upon exposure to pathobionts, DCs and epithelial cells receive danger signals and release cytokines, such as IL-25, IL-33, and thymic stromal lymphopoietin (TSLP); these promote the activation and expansion of ILC2s, which express Th2 cytokines, such as IL4, IL-5, and IL13. While IL-5 promotes eosinophil activation and differentiation and the production of profibrotic factors, such as transforming growth factor (TGF)-β1 and fibroblast growth factor (FGF)-9, IL-13 produced by Th2 cells and T follicular helper (Tfh) 13 cells, is critical for the expression of high-affinity antigen-specific IgE. IgE antibodies interact with FcεRI on mast cells and upon exposure to allergen triggers degranulation and release of histamine, which causes allergy and inflammation.
Under normal conditions, only minimal amounts of Ag can cross mucosal barriers through the paracellular pathway, a process typically associated with the development of immune tolerance. Ag exposure of inappropriate duration or magnitude may lead to immune-mediated diseases in genetically susceptible subjects. In a few instances, this has been suggested to reflect the intrinsic properties of the antigenic protein.
Dermatophagoides pteronyssinus (Der p) 1 from house dust mite (HDM), one of the most common indoor aeroallergens, has long been shown to be able to disrupt intercellular tight junctions (TJ) and increase Ag trafficking through bronchial epithelial monolayers
[110]. This property, and in general the ability to induce epithelial effector functions, is shared with other allergens—including certain food allergens—and less specific triggers such as detergents and microplastics
[111][112][113]. However, allergic sensitization may be facilitated in the presence of intrinsic barrier defects, as evidenced in genetic studies showing the significant association of filaggrin (FLG) loss-of-function mutants, impaired skin barrier function, and the development of allergic disease
[114]. Increasing evidence suggests that epicutaneous sensitization may in fact promote the subsequent development of allergic diseases, especially food allergy, in children with AD, thus contributing to the progression of the atopic march
[115][116].
As amply documented in a wealth of studies conducted over the past 20 years, the composition and diversity of the microbial communities lining all body surfaces, collectively referred to as microbiota, represent a major, critical variable in the regulation of barrier competence and adaptive and innate responses
[117][118]. The gut microbiota is successfully seeded early in life, by colonization from maternal vaginal and breast communities at birth and during lactation
[119]. Subsequently, throughout adult life, the microbiota is significantly influenced by dietary habits
[120]. Host-microbiota interactions are known to have a critical impact on multiple components of the immune system, contributing to immune homeostasis and susceptibility to infectious and inflammatory diseases. Allergic inflammation resulting from skewed activation of Th2 clones is typically enhanced in germ-free animals, suggesting a major role for gut colonization in the development of balanced Th1/Th2 responses
[121]. Likewise, reduced Th1 responses and an increased predisposition to develop allergic disease can be observed in infants delivered by cesarean section, associated with a delayed gut colonization of symbiont species and a less diverse microbial community
[122]. In fact, the risk of developing unbalanced immune responses in infancy and childhood, resulting in allergic and autoimmune disease, has also been linked to the maternal diet, particularly during pregnancy and lactation. In support of this theory, recent studies have documented the increased levels of food allergen-specific IgE and IgG antibodies in the offspring of mothers who were prescribed gestational-targeted or exclusion diets
[123], and, conversely, an overall reduced risk for immune dysfunction was found following maternal supplementation with probiotics (reviewed in
[124]). These overall studies provide factual evidence in support of the hygiene hypothesis, whereby exposure to declining environmental biodiversity, by adversely affecting the human microbiota and its central functions in immune regulation, would primarily account for the rising prevalence of allergic and other chronic inflammatory diseases
[125][126].
Highlighted in several studies are the interactions of FoxP3
+ T regulatory (Treg) cells, a CD4
+ T-cell subset critically involved in immune homeostasis and tolerance, with microbiota-delivered signals. A sufficiently diverse microbial community may promote the activation and expansion of Treg cells and the production of the immune regulatory cytokine, IL-10, via the interaction of certain bacterial components with Toll-like receptors (TLR) or other pattern recognition receptors (PRR)
[127]. Bacterial metabolites, e.g., butyrate and other short chain fatty acids (SCFA) generated upon the processing of dietary fibers, can also direct the development and function of Treg cells via the interaction with gut epithelial cells and dendritic cells and the induction of immunomodulatory mediators such as vitamin A metabolite retinoic acid (RA)
[128][129]. As documented in animal models of food allergy, concentrations of butyrate, such as those measured in mature human milk, are sufficient to promote gut barrier integrity and IL-10 production, reduce the allergic response, and enhance the desensitizing effect of allergen immunotherapy (AIT)
[130][131]. These findings are invoked to explain the beneficial anti-inflammatory, anti-allergic effects of
Lactobacillus and
Bifidobacterium probiotic mixtures and of a high-fiber diet
[120][128][130][132].
A connection between diet, gut microbiota composition, and allergic inflammation is postulated in several studies
[133][134]. Studies in germ-free mice demonstrated that microbiota-delivered factors can regulate Th2-driven immunity through the induction of Th17 cells and of a subset of Treg cells expressing the Th17 signature factor, retinoid-related orphan receptor (ROR)-γt
[135].
Lactobacillus strains and other symbiotic species, through the production of tryptophan-indole catabolites, may directly activate these cells, as well as ROR-γt
+ type 3 innate lymphoid cells (ILC3), via the aryl-hydrocarbon receptor (AHR), and induce the production of IL-22, a cytokine-promoting gut epithelial regeneration, barrier integrity, and the secretion of antimicrobial peptides
[132][136]. In addition, IL-1β produced by intestinal macrophages sensing microbial signals can induce the release of granulocyte-macrophage colony-stimulating factor (GM-CSF) from nearby ROR-γt
+ ILC3, which in turn upregulates RA and IL-10 production from dendritic cells and macrophages, further contributing in the maintenance of tolerance to dietary antigens
[137].
The central role of innate immunity in the integration of the environmental signals involved in the development and maintenance of natural tolerance to foods and other antigens has emerged convincingly in recent years. As shown in a cohort study of egg-allergic infants, a distinctive cytokine signature is detected early in life in circulating monocytes and dendritic cells, which is predictive of persistent food allergy in childhood
[138]. On the other hand, innate immune profiles in children eventually outgrowing their allergy were directly related to serum levels of vitamin D, further stressing the importance of this nutrient in the development of natural tolerance in childhood
[138]. Cytokines released by dendritic cells and epithelial cells upon exposure to pathobiont-delivered danger signals, including IL-25, IL-33, and thymic stromal lymphopoietin (TSLP)—collectively referred to as alarmins—directly contribute to allergic inflammation via the direct activation and expansion of ILC2, an innate lymphoid subset that expresses substantial amounts of Th2 cytokines
[139]. The pathophysiologic role of ILC2 in allergic disease has been demonstrated in several models
[140][141][142]. These cells, either on their own or in a complex amplifying loop with Th2 cells, may promote and enhance the expression of food-specific IgE in switched and unswitched B cells mainly via the production of IL-4 and IL-13
[141][142][143].
The critical role of IL-13 and related Th2 cytokines in allergic sensitization and its clinical manifestations is emphasized in a recent study documenting that a subset of IL-13-producing T follicular helper (Tfh) cells, termed Tfh13, is critically required for the expression of high-affinity specific IgE and the subsequent development of severe anaphylaxis
[144]. Tfh cells, a specialized CD4
+ T-cell subset defined by expression of the nuclear factor Bcl6 and of the cytokine, IL-21, are key players for the development of switched, memory B cells and plasma cells in germinal centers
[145]. While both IL-13 and IL-4 contribute to IgE class-switch recombination and allergic inflammation in part via the interaction with shared receptors, their expression is driven by diverging mechanisms, possibly reflecting their unique involvement in distinct aspects of the allergic response
[146][147][148]. Regardless, lineage tracing experiments conclusively demonstrated that in most instances, the switched, allergen-specific IgG
+ B cells are the precursors of IgE-expressing B cells and IgE antibody-secreting plasma cells
[149]. In situ IgE class switching and IgE production have been documented in the respiratory and gastroenteric mucosa in response to such environmental signals, as allergen exposure and microbial superantigens
[150][151]. Importantly, single-cell transcriptomic analyses reveal virtually absent IgE
+ memory B cells in most individuals, whereby most IgE-producing cells are represented by plasma cells
[152]. Taken together, these findings suggest that persisting levels of specific IgE in patients with chronic allergies may only be ensured by continuous plasma blast generation via sequential switching from an IgG
+ memory pool
[153][154].
A marked reduction in serum IgE titers was documented in AD patients treated with dupilumab, a monoclonal antibody that blocks IL-4 and IL-13 interaction with their shared receptor
[155]. This is consistent with the idea that a significant proportion of IgE are secreted from newly switched, short-lived plasma blasts, and that interfering with Th2 or ILC2 activation and downstream effector signals might hence greatly contribute to restoring tolerance to common allergens. Current AIT protocols, including OIT, are indeed aimed at counteracting these responses via the induction of allergen-specific Treg cells
[156]. However, a prospective decrease in specific IgE levels is not a sufficient predictor of clinical outcome in AIT protocols (reviewed in
[157]), whereas a rise in other antibody classes, namely IgA, IgG1, and especially IgG4, is more consistently observed
[158][159]. Switching to IgG4 is promoted by IL-10, a cytokine produced at higher levels in patients receiving AIT
[160]. Such allergen-specific IgG4 undergo increased somatic mutation relative to IgE in these patients, resulting in the expression of antibodies with higher affinity for allergens and is hence more effective at preventing allergen interactions with mast cell-bound IgE
[161].
IL-10-producing, immunosuppressive B regulatory cells (Breg) have been recently demonstrated, which were found to be expanded and contribute to peripheral allergen tolerance in patients receiving AIT
[162][163]. Interestingly, a subset of Breg cells, termed BR1, were found to be the exclusive source of specific IgG4 and a precursor of IgG4-secreting plasma blasts, in subjects displaying spontaneous or AIT-acquired tolerance to allergen
[164]. The recently discovered mutual interactions of effector and regulatory B cells with microbiota components further stress the relative importance of these cells in immune homeostasis in the gut and other mucosal surfaces and the pathophysiology of food allergy and other diseases associated with imbalanced, aberrant responses to environmental antigens.
While most immune-mediated food allergies are associated with predominant Th2-driven, IgE-mediated responses, the development of variably related conditions as eosinophilic esophagitis, FPIES, FPE, FPIAP, and CD, recognizes distinct and relatively complex immune mechanisms. EoE is also mediated by a prevalent, Th2-biased immune response
[165]. In particular, elevated levels of IL-5 promote eosinophil differentiation and trafficking to the esophagus
[166] and, together with IL-9, are responsible for the progressive eosinophilia and mastocytosis typically observed in the esophageal mucosa in affected patients. Activated eosinophils and mast cells can in turn produce profibrotic factors (such as the transforming growth factor (TGF)-β1 and the fibroblast growth factor (FGF)-9), which cause remodeling of the esophageal epithelium and subepithelium and are responsible for the characteristic symptoms and complications of this condition
[167]. However, the association of EoE’s clinical picture with IgE-dominated specific responses to food is not entirely clear. In some cases, EoE symptoms also appear to be triggered by airborne allergens, and in quite a few cases no clear trigger can be identified
[68]. Moreover, recent studies also highlighted the presence of IgG4 deposits in the esophageal mucosa, suggesting their possible contribution to the inflammatory response in EoE
[69].
In FPIES, a specific T-cell response to causative food antigens leads to TNF-α secretion, which initiates the systemic activation of monocytes, eosinophils, neutrophils, and natural killer cells, resulting in inflammation and increased permeability of the GI mucosa
[168][169]. In FPE, the jejunal mucosa is damaged by infiltrating T cells that mostly exhibit a cytotoxic, CD8
+ effector phenotype and a γδ TCR, causing malabsorption
[170]. FPIAP is characterized by a dense eosinophilic infiltration of the rectosigmoid mucosa and typically affects breastfed infants, suggesting the possible role of immunologic components found in breastmilk, such as secretory Igs specific for dietary proteins
[171]. Finally, in CD, deamidated α-gliadin-derived peptides are presented by HLA-DQ2/DQ8 complexes of APC in genetically predisposed individuals. Following activation, α-gliadin-specific T cells migrate from the lamina propria into the subepithelial area and begin to produce various pro-inflammatory cytokines, such as IFN-γ and TNF-α. Activated cytotoxic T cells also produce molecules, such as Fas ligand and granzymes, which promote apoptosis of nearby enterocytes. These events combined trigger an extensive immune reaction that causes pathological tissue alterations, resulting in damage of the small intestinal mucosa, villous atrophy, and malabsorption
[172]. The ensuing activation of B cells leads to the production of anti-gliadin, anti-endomysium, and anti-TTG antibodies
[173]. The presence of anti-TTG antibodies in serum is very useful for diagnosis and for times during the follow-up since they disappear from the serum of patients when they are on a gluten-free diet. However, it is unclear whether they are responsible for the damage to the mucosa or are rather its consequence
[172].