5.1. Allergen-specific to T cell activation
Antibody responses to proteins, such as food allergens, are strictly T-dependent [
91]. While B cells recognize three-dimensional epitopes Th cells are specific for small peptides presented in MHCII molecules. The initial entrance of the allergen leading to specific sensitization may occur through the skin, gastrointestinal tract, airway or damaged epidermal barrier [
92],[
93],[
94],[
95]
. In allergic individuals, sensitization results in the formation of Th2 and T follicular helper (Tfh) 13 cells and the formation of allergenic IgE [
50]. Subsequently, allergenic re-encounter through food ingestion via the gastrointestinal tract leads to IgE-mediated activation of mast cells and basophils, eventually driving acute allergic symptoms.
Food allergens can access a dysfunctional epithelial barrier and trigger the production of alarmins such as Interleukin 33 (IL33), Interleukin 25 (IL25) and Thymic stromal lymphopoietin (TSLP). These alarmins mediate type 1 hypersensitivity [
96] by skewing the T cell response towards the Th2 axis and activating mast cells, dendritic cells, innate lymphoid cells and eosinophils [
97],[
98]. B cells may also support the generation of Th2 cells and allergy development through the production of IL4, as recently shown in a murine model for allergic asthma [
28].
5.3. Production of mutated, high affinity IgE
Expressed in a membrane-bound form on the cell surface, antibodies serve as antigen-specific B cell receptor (BCR) which determine the cellular fate at all stages of development [
101]. Hypermutated, high-affinity antibodies are the product of the follicular pathway. B cells following the follicular pathway transform primary B cell follicles within secondary lymphoid tissues into germinal centers (GC), where B cells undergo hypermutation, affinity maturation and differentiation into long-lived plasma cells and memory B cells [
102],[
103] [
Figure 2].
GC development requires help from Tfh cells, which provide stimuli such as IL-21 and CD40 essential for induction of hypermutation and positive selection of B cells that have acquired BCR of higher affinity [
104],[
105]. While Th-derived IL4 is sufficient to induce class switch to IgE and the formation of low-affinity IgE, the generation of high-affinity IgE with anaphylactic properties depends on help from Tfh13 cells which additionally produce high levels of IL-13 and IL-5 together with some IL-21 [
50]. Tfh13 cells regulate germinal center responses in type 2 immune reactions and appears to be important for the generation of hypermutated high-affinity IgE and the development of asthma [
50]. Genetically modified mice lacking Tfh13 cells show only very low levels of anaphylactic, high-affinity IgE. Tfh13 cells are found in allergic mice and humans with high-affinity IgE to allergens and are further characterized by the expression of the transcription factors BCL6 and GATA3. These cells may represent an interesting target for future therapies of food allergy [
106].
A key property of the GC reaction is the generation of memory, mediated by memory B cells and long-lived plasma cells [
107]. The pool of long-lived plasma cells secretes antibodies of very high affinity but consist of only a low number of distinct clones. Therefore, they have a limited antigen-binding repertoire [
108]. In contrast, memory B cells do not secrete antibodies but provide a backup. Upon antigenic re-stimulation, they can undergo rapid differentiation into antibody-secreting plasma cells. Though the affinities of their antibodies are lower compared to that of long-lived plasma cells, they consist of a higher number of clones that cover a large antigen-binding repertoire [
108].
Despite the high-affinity IgE is derived from GCs, the existence of IgE+ memory B cells and long-lived plasma cells is a matter of debate. At least the majority of IgE cells seem to be excluded from these memory compartments. Because of their very low frequency and the possibility of confusion with B cells that bind IgE via their low-affinity receptor CD23, IgE-expressing B cells are difficult to detect without doubt. In addition, GCs typically stain brightly for IgG, but only some IgE is detectable in these tissue structures. Thus, the production of IgE appears to be tightly regulated [
109]. Different antibody classes exhibit qualitatively distinct signaling properties. In a model using forced IgE BCR signaling has been shown to induce apoptosis, independent from antigenic stimulation [
41]. Accordingly, in the same study, primary IgE+ cells showed a higher rate of apoptosis than IgG1+ cells. Noteworthy, another study confirmed the finding that IgE expression mediates a tonic, antigen-independent signal. This study could not confirm that IgE directly promotes B cell apoptosis, instead independent of antigen binding, it was found to support terminal differentiation into plasma cells, which involves multiple parts of the IgE BCRs as well as Syk, CD19, BLNK, Btk, and IRF4 [
39]
.
Another study showed that though IgE is formed by reactivation of IgG memory cells, signaling of the membrane IgE BCR, but not of the murine IgG1 BCR is required to yield high IgE levels [
110]. IgE BCR expression was still found on plasma cells, which is different to IgG which are not expressed any more on the surface of long-lived plasma cells and IgE signaling on plasma cells was found to be relevant for the production of serum IgE.
Together, IgE BCR expression mediates a tonic signal even in the absence of external stimulation by antigen, if that promotes apoptosis or terminal differentiation might be dependent on the model, and under physiological conditions, on additional factors such as the availability of an anti-apoptotic environment. Some IgE+ B cells seem to survive and signal via their membrane-bound IgE receptor which is a crucial regulator of IgE production.
Comparison of IgE+ and IgG1+ murine B cells by whole-genome CRISPR screening showed that IgE+ B cells have distinct properties [
111]. Different form IgG+ cells, IgE+ B cells and IgE+ plasma cells showed chronic calcium signaling eventually resulting in BCL2L11-dependent apoptosis. Moreover, there is evidence that after repeated antigenic stimulation in mice, high-affinity IgE-secreting plasma cells are generated through reactivation and further class-switch recombination of IgG1 memory B cells [
34]. Together, these findings indicate that the majority of IgE is formed by short-lived plasma cells generated from IgG+ memory B cells.
Nevertheless, next-generation sequencing of the IgG1 and IgE repertoires in a murine model of food allergy indicates that IgE+ cells may follow individual fates. While most IgE clones showed little clonal expansion, a small proportion of highly hypermutated IgE clones exhibited massive clonal expansion, comparable to that of the most expanded IgG1 clones [
33]. These data are in accordance with the view that most IgE+ cells show little proliferation and exhibit a short lifetime. But a few hypermutated, GC-derived IgE+ B cells seem to undergo positive selection and proliferation. If these expanded IgE clones that escape from early apoptosis can survive on the long run to enter the memory compartment remains to be elucidated.
In favour of such an idea there is some evidence for the existence of IgE+ memory B cells and long-lived plasma cells in mice and humans [
112],[
113],[
114].
5.4. Regulation of IgG to IgE ratios
Most of the class-switched IgE+ cells are derived from IgG+ B cells that underwent further (sequential) class switch to IgE in a follicular B cell response [
31],[
33],[
115]. The mechanisms controlling the relative IgE to IgG production during the allergen-specific immune response are only partly understood so far. Studies from our laboratory investigating the antibody response to hen´s egg in a murine food allergy model indicate that IgE to IgG ratios are controlled on the level of single B cell clones [
33]. Most individual clones containing both IgE and IgG1, showed a several-fold excess of IgG1 compared to IgE, i.e., a high IgG1 to IgE ratio. However, fewer, but still a considerable proportion of clones showed a massive excess of IgE, with ratios above 5-fold more IgE than IgG1. Evidence was provided that the differential IgG1 to IgE ratios are due to individual BCR signaling strength which had two consequences. First, strong BCR signaling inhibited sequential class switch from IgG1 to IgE. Second, BCR crosslinking could optimize help from T follicular helper cells producing IL-21, a cytokine that was found to favor IgG1 over IgE production. Hence, on a clonal level, IgG1 to IgE ratios seem to be strongly affected by the individual antibody affinities [
33] [
Figure 3]. There is increasing evidence that nutrition and metabolic factors can have a strong impact on B cell activation [
116], however, their role on class switch, and the relative ratios of allergenic IgE and anti-allergenic IgG remains to be elucidated.