IL-4 Haploinsufficiency Specifically Impairs IgE Responses against Allergens in Mice

Interleukin 4 is an immunoregulatory cytokine produced by CD4⁺ Th2 and follicular Th cells (TFH), and influences numerous cellular and humoral aspects of type 2 immune responses mounted against parasites and allergens. Typically, IL-4 promotes tissue mastocytosis and eosinophilia, alters intestinal fluid flux, enhances smooth muscle cell hypercontractility, and increases vascular permeability in response to the vasoactive mediators of allergies (1–6). IL-4 also regulates B cell activation and class-switch recombination. Most notably, IL-4 is essential for IgE production (7–9), but it additionally increases naive B cell MHC class II and CD23 (FcεRII) expression (10), augments germinal center B cell expansion, and promotes IgG1 Ab production (8, 9, 11, 12). IL-4 may also contribute to Ab affinity maturation (12). The influences of IL-4 and an additional Th2 cell–produced cytokine, IL-13, on asthma pathogenesis have led to the development of clinical IL-4–receptor antagonists that are shown to diminish exacerbation frequencies in specific subgroups of asthma sufferers (13–16).

The importance of IL-4 in type 2 immune responses has generated much interest in defining its cellular sources in vivo. Tissue-localized CD4⁺ effector Th2 cells, lymphoid CD4⁺ CXCR5⁺PD-1^hi TFH, invariant NKT cells, basophils, and eosinophils have been identified as IL-4–producing cells (10, 12, 17–23). Many of these populations were first identified as IL-4 producers through evaluations of Il4 mRNA transcript frequencies, or IL-4 production after ex vivo restimulation; however, more sensitive studies utilizing IL-4–reporter mice were integral in validating these results. Reporter mice have further permitted sensitive probing of the comparative importance of distinct IL-4–producing populations during type 2 immune responses (21, 24). The two most widely used IL-4 reporter mice are the KN2 and G4 IL-4–substituting reporters (22, 25). Both reporters faithfully reflect commitment to IL-4 production in vivo, with the KN2 reporter expressing modified human CD2, and the G4 reporter expressing enhanced GFP, both in place of the native IL-4 protein. A third, more recently developed IL-4 reporter, the BAC transgenic 4C13R IL-4 reporter mouse (17, 26, 27), expresses AmCyan fluorescent protein under control of the Il4 gene regulatory elements without manipulation of the endogenous Il4 gene locus, which permits evaluation of commitment to IL-4 production without ex vivo restimulation, and leaves the endogenous Il4 gene locus unmodified.

IL-4–substituting reporter mice provide insight into complex, difficult-to-monitor features of IL-4 production. However, these mice carrying a single IL-4–producing allele present the challenge of determining whether this accurately reflects normal Th2 immune response biology, and thus the conditions under which it is appropriate to apply these tools. In a previous study in IL-4^G4/+ hemizygous mice, IgE responses seemed much lower than expected (3), which led us to consider the impact individual Il4 alleles have on the type 2 immune response to allergens. In this study, we use protein/alum immunizations and reporter mice to show that Il4-hemizygous mice exhibit substantially diminished IgE levels and curtailed IgE-dependent allergic disease sequelae without significant changes to the IgG responses. Our findings imply that the serum IgE isotype response is disproportionately affected by the loss of one IL-4–producing allele, whereas other features of the allergic immune response are more resistant to changes in the availability of IL-4.

IL-4^KN2/KN2 (22) backcrossed for 10 generations to BALB/c background were crossed to BALB/c and BALB/cByJ mice to generate IL-4^KN2/+ mice. IL-4^G4/G4 (25) mice crossed to BALB/c and C57BL/6 backgrounds were respectively crossed to BALB/c and BALB/cByJ or C57BL/6J mice to generate IL-4^G4/+ mice. 4C13R (27) reporter mice were bred and maintained on a C57BL/6 background in the Malaghan Institute of Medical Research Biomedical Research Unit. 4C13R mice were crossed with IL-4^G4/G4 mice to generate 4C13R × IL-4^G4/+ mice and 4C13R × IL-4^G4/+ mice were crossed with IL-4^G4/G4 mice to generate 4C13R × IL-4^G4/G4 mice. C57BL/6J background IL-4^+/+ and IL-4^KN2/+ Basoph8 (28) mice were maintained in facilities at the University of California, San Francisco, and were used for Ab-secreting cell (ASC), basophil, and mast cell analysis experiments. All procedures were performed in accordance with institutional guidelines.

Mice were i.p. injected on days 0 and 14 with 50 μg of grade V OVA (Sigma, St Louis, MO) plus 1 mg of Alu-Gel-S (Serva Electrophoresis, Heidelberg, Germany). To induce active systemic anaphylaxis, on day 28 following i.p. immunization (C57BL/6 background) mice were i.v. injected with 100 μg of OVA and monitored for temperature changes intrarectally. To induce intestinal anaphylaxis (29), mice on a BALB/c genetic background were i.p. immunized on days 0, 14, 28, and 42 prior to receiving 50 mg of OVA intragastrically (i.g.) after a 4 h fast, three times a week, every second day, beginning on day 56. Diarrhea was monitored 60–90 min after challenge.

For OVA-specific IgE ELISAs, wells of Immunosorp Maxi plates (Nunc) were coated with 100 μl of 5 μg/ml anti-mouse IgE (6HD5) overnight at 4°C in 0.1 M carbonate coating buffer (pH 9.6). For OVA-specific IgG2a and IgG1, wells were coated with 100 μl of 10 μg/ml OVA. For all OVA-specific ELISAs, the next day wells were washed and blocked with 200 μl of 10% FBS in PBS (blocking buffer) for 2 h at room temperature. Blocking buffer was replaced with a 100 μl sample, serially diluted in blocking buffer. After 2 h at room temperature, or overnight at 4°C, wells were washed and incubated with 100 μl of 20 μg/ml biotinylated OVA (for OVA-specific IgE), 1 μg/ml biotinylated anti-mouse IgG2α (for OVA-specific IgG2α; R19-15), or 1 μg/ml biotinylated anti-mouse IgG1 (for OVA-specific IgG1; A85.1) for 60–90 min at room temperature. Wells were washed and incubated with 100 μl of 1/1000 streptavidin-conjugated HRP (GE Healthcare, Auckland, New Zealand) for 1 h. Colorimetry was performed with TMB substrate (BD Pharmingen) and stopped with 1 M of H₂SO₄. Absorbances were read at 450 nm using a Versamax plate reader. IgG1 titer was the inverse serum dilution at the OD₅₀. Nitrophenyl (NP)-specific IgE and IgG1 ELISAs were performed as described (30).

Excised 1.5 cm jejunal sections were placed in 10% formalin solution (4% w/v; Sigma) and stored at 4°C until embedded in paraffin. Tissues were processed into 4 μm sections using standard histological techniques. Sections were stained with chloroacetate esterase and lightly counterstained with hematoxylin. Mast cells were easily identified as bright red cells as described (31).

Murine mast cell protease-1 (mMCP-1) ELISA was performed according to the manufacturer’s instructions (eBioscience).

Mice were anesthetized using xylazine and ketamine (Phoenix, New Zealand). For T cell analyses, 30 μl of a solution containing 50 μg OVA or NP₁₆OVA (Biosearch Technologies, Petaluma, CA) with 1 mg Imject Alum (Pierce) was injected into the ear pinnae as described (32). After euthanasia, auricular lymph nodes were isolated and stored in IMDM until cell harvest and isolation. For ASC analyses, 25 μl containing 50 μg NP₁₆OVA in a 1:1 PBS:Alhydrogel (Accurate Chemical & Scientific, Westbury, NY) solution was injected into the ear pinnae as described (32). After euthanasia, auricular lymph nodes were isolated and stored in handling medium [10% FBS, 1% penicillin-streptomycin-l-glutamine solution (Life Technologies), in DMEM (Corning Mediatech)] until cell isolation.

For T cell analyses, lymph nodes were disrupted using the rubber bung of a 1 ml tuberculin syringe (BD) and liberated cells were washed through 70 μm cell strainers into complete IMDM [5% FBS (Sigma), 100 U/ml penicillin/100 μg/ml streptomycin (Invitrogen), and 55 μM β-mercaptoethanol (Invitrogen) in IMDM]. CD4⁺ T cells were purified by positive selection using FlowComp Mouse CD4 Dynabeads (Invitrogen). Live cell counts were determined using a hemocytometer and trypan blue (Invitrogen) exclusion. For ASC and basophil analyses, processing was as for T cells, except tissues were harvested and cells resuspended in handling medium and counts determined with a Z-series Coulter Particle Counter after treatment with ZAP-OGLOBIN II (Beckman Coulter, Hialeah, FL). Mast cells were isolated by peritoneal lavage with PBS.

For T cell analyses, cells were resuspended in 0.01% NaN₃ (Sigma) and 2% FBS in PBS then incubated with anti-CD16/32 Ab (clone 2.4G2) before staining with cocktails of fluorophore-conjugated Abs. For ASC, basophil, and mast cell analyses, the staining buffer contained 0.1% NaN₃ (Andwin Scientific), 2% FBS, and 2 mM EDTA (Life Technologies), and cells were incubated with anti-CD16/32 (clone 93; BioLegend) before staining. For T cell analyses, lymphoid cells were stained with Abs against the following molecules (clone, conjugate; source): B220 (RA3-6B2, Horizon V450; BD), CD4 (RM4-5, Qdot605; BD), PD-1 (RMP1-30, PE/Cy7; BioLegend), CXCR5 (2G8, biotin; BD); streptavidin-PE-Texas Red (Life Technologies) or streptavidin-allophycocyanin (Invitrogen) were also used. Splenic basophils and peritoneal mast cells were stained with Abs against IgE (R35-118, biotinylated; BD), cKit (2B8, PE/Cy7; BD), and IgD (1126c.2a, BV510; BD) or CD49b (DX5, PE; BD), FceR1α (MAR-1, APC; BioLegend), and cKit (2B8, APC/Cy7; BioLegend). Also used were streptavidin Qdot605 (Life Technologies) and NP-allophycocyanin [prepared in house as described (30, 33)]. ASC staining was performed as described (30) using purified Abs against IgE (RME-1; BioLegend), and fluorophore labeled Abs against IgE (RME-1, Fluorescein; BD), IgG1 (A85.1, Horizon V450; BD), IgD (1126c.2a, BV510; BD), B220 (RA3-6B2, Qdot 655; Life Technologies), CD138 (281-2, BV711; BD), IgM (II/41, PE/Cy7; eBioscience), and CD38 (90, AF700; eBioscience); additional staining reagents were biotinylated peanut agglutinin (Vector Laboratories, Burlingame, CA), Qdot605 streptavidin conjugate (Invitrogen), Fixable Viability Dye eFluor780 (eBioscience), and NP-allophycocyanin. T cell data were acquired on a BD LSRII SORP flow cytometer (Becton Dickinson, San Jose, CA); ASC, basophil, and mast cell data were acquired on an LSR Fortessa (Becton Dickinson). Gates excluded doublets and DAPI⁺, propidium iodide⁺ or eFluor780⁺ events. CD4⁺ T cells were gated as CD4⁺B220⁻ events; TFH were CXCR5⁺PD-1^hi CD4⁺ T cells. NP-binding ASC were gated as CD138⁺B220^mid/loIgD⁻IgM⁻CD38⁻NP_{(surface+intracellular)}⁺ cells. Basophils were identified in Basoph8 mice as yellow fluorescent protein (YFP)⁺cKit⁻IgD⁻ and peritoneal mast cells as side-scatter^hi YFP⁻cKit⁺. Splenic basophils from C57BL/6 background IL-4^+/+ and IL-4^G4/+ mice were identified as DX5⁺FceR1α⁺cKit⁻ and peritoneal mast cells as DX5⁺FceR1α⁺cKit⁺. Flow data were analyzed using FlowJo software (Treestar).

Wells of Immunosorp Maxi plates (Nunc) were coated with 100 μl of 2 μg/ml anti-mouse IL-4 (11B11) overnight at 4°C in 0.1 M carbonate coating buffer (pH 9.6). Wells were washed and blocked with 200 μl of 1% BSA in PBS (blocking buffer) for 1 h at room temperature. Blocking buffer was replaced with a 100 μl sample or IL-4 standard, serially diluted in blocking buffer. After 2 h at room temperature wells were washed and incubated with 100 μl of 2 μg/ml biotinylated anti-mouse IL-4 (Pharmingen) for 2 h at room temperature. Wells were washed and incubated with 100 μl of 1/1000 streptavidin-conjugated HRP (BioLegend) for 1 h. Colorimetry was performed with BD OptEIA substrate reagent set and stopped with 1 M of H₂SO₄. Absorbances were read at 450 nm using a Versamax plate reader.

Statistical analyses were performed using Graphpad Prism software (Graphpad Software, San Diego, CA). Statistical comparisons used are specified in the figure legends. Data shown on log scales were log-transformed prior to analysis. A p value ≤ 0.05 was considered significant.

We sought to determine whether the IL-4 haploinsufficiency resulting from the Il4-hemizygous state affected Ag-specific Ab production. Mice that were IL-4^+/+, IL-4^G4/+, or IL-4^G4/G4 were immunized with OVA/Alum and the OVA-specific IgG1, IgE, and IgG2a serum Ab titers evaluated by ELISA. Equivalent OVA-specific IgG1 titers of up to 10⁶ were detected in IL-4^+/+ and IL-4^G4/+ mice following allergen challenge, whereas the IL-4 deficient, IL-4^G4/G4 mice had a 4-fold reduced serum titer (Fig. 1A). OVA-specific IgE responses were also significantly induced in IL-4^+/+ mice, reaching serum titers of 10³ (Fig. 1B). However, OVA/Alum immunized IL-4^G4/+ hemizygous mice had 100-fold lower OVA-specific IgE responses. No OVA-specific IgE could be detected in immunized IL-4^G4/G4 mice, although these mice had elevated serum titers of OVA-specific IgG2a Ab compared with both IL-4^+/+ and IL-4^G4/+ mice (Fig. 1C). Total IgE levels in mice before and after immunization of IL-4^G4/+ mice were also significantly lower than in IL-4^+/+ mice (Fig. 1D).

The reduced IgE levels in Il4-hemizygous mice were not caused by the abundance of serum OVA-specific IgG Abs competing for allergen binding, because the ELISA we employed first captures total IgE and then uses an OVA-biotin conjugate to reveal the OVA-specific IgE serum Ig.

We next examined hapten-specific Ab responses at the level of responding B cells using NP conjugated OVA (NP-OVA) as the immunizing Ag. Seven days after immunization in the ear with NP-OVA/Alum, draining lymph nodes were excised and numbers of ASC binding the dominant hapten, NP, were evaluated by FACS (Fig. 2A). Similar total cell, NP-binding ASC and NP-binding IgG1 ASC numbers were detected in draining auricular lymph nodes of IL-4^+/+ and Il4-hemizygous IL-4^KN2/+ mice (Fig. 2B–D). In contrast, NP-binding IgE ASC were around 6-fold less abundant in IL-4^KN2/+ mice (Fig. 2E). These data demonstrate that the genesis of Ag-specific IgE ASC was impaired in the Il4-hemizygous mice.

To determine whether the IgE production deficiency affected the amount of IgE bound by high affinity IgE receptor (FcεRI)-expressing cells, we examined surface IgE expression and NP-binding levels on basophils from NP-OVA immunized IL-4^+/+ and IL-4^KN2/+ hemizygous Basoph8 reporter mice, in which basophils can be identified through the expression of YFP (28). As expected, YFP⁺ splenic basophils from IL-4^KN2/+ Il4-hemizygous mice had lower surface IgE levels and bound less NP-allophycocyanin than did basophils from the IL-4^+/+ strain (Fig. 3A–D). The geometric mean fluorescence intensity (gMFI) of IgE- and NP-binding on basophils from the Il4-hemizygous mice were 3–10-fold lower than in the IL-4^+/+ Basoph8 reporter mice (Fig. 3B–D), similar to the fold reduction in the number of NP-specific IgE ASC that were generated in these mice (Fig. 2E). Surface-bound IgE and NP staining were similarly affected on peritoneal mast cells (Fig. 3B–D). The reduced levels of surface IgE and NP-binding in Il4-hemizygous mice was not due to reduced levels of FcεR1 receptor on the surface of these cells, as the gMFI of FcεRIα on basophils (Fig. 3E) and mast cells (Fig. 3F) was comparable in IL-4^+/+ and IL-4^G4/+ mice. These data demonstrate that basophils and mast cells from Il4-hemizygous mice captured less NP-specific IgE than those from IL-4^+/+ mice.

These data indicate that immunized Il4-hemizygous mice generate fewer IgE ASC than IL-4^+/+ mice, resulting in lower levels of IgE both in circulation and bound to the surface of basophils and mast cells.

We next examined the impact of Il4-hemizygosity on Ab-mediated allergic immune responses. To characterize Ab-mediated active systemic anaphylaxis, mice were immunized i.p. with OVA/Alum and then challenged i.v. with OVA. Upon i.v. challenge, IL-4^+/+, IL-4^G4/+, and IL-4^G4/G4 mice became hypothermic, a characteristic feature of systemic anaphylaxis (Fig. 4A). Allergen challenge–induced hypothermia was more severe in IL-4^+/+ than IL-4^G4/G4 mice, whereas IL-4^G4/+ mice presented intermediate decreases in temperature (Fig. 4B). These data indicate that active systemic anaphylaxis was affected by Il4-hemizygosity in relative proportion with Il4 allele dose.

We also investigated whether Il4-hemizygosity affected disease development in a mouse model of IgE-dependent oral allergen–induced intestinal anaphylaxis (29). In this model, OVA-sensitized mice exhibit diarrhea after successive i.g. OVA challenges. Protection from oral allergen–induced diarrhea occurred in two independently developed strains of Il4-hemizygous mice, as 0/4 IL-4^G4/+ and 0/3 IL-4^KN2/+ hemizygous mice exhibited diarrhea when primed i.p. four times prior to i.g. challenge compared with 4/4 IL-4^+/+ mice (Fig. 4C). Similar observations were made in two additional experiments with mice primed i.p. twice prior to i.g. challenge as is standard in this model (M.J. Robinson, unpublished observations). Protection from diarrhea was associated with reduced frequencies of mast cells in intestinal jejuna and ≈30-fold lower serum mMCP-1 levels in the IL-4^G4/+ hemizygous mice compared with IL-4^+/+ mice after 10 i.g. challenges (Fig. 4D, 4E). These data confirm that Il4-hemizygosity impaired disease development in an IgE dependent model of allergic disease.

Allergen-specific IgE production is dependent on the allergen-induced expansion of IL-4–producing CD4⁺ T cells in the draining lymph node (11, 34). We therefore sought to establish the extent to which Il4-hemizygosity affected the frequency and quality of IL-4–producing CD4⁺ T cells in the draining lymph nodes of allergen-immunized mice. To determine the effect of Il4-hemizygosity on the in vivo genesis of IL-4–producing CD4⁺ T cells in the draining lymph nodes of immunized mice, we used BAC transgenic 4C13R reporter mice (17, 27). In these mice, the developing CD4⁺ Th cell that commits to IL-4 production can simultaneously produce IL-4 from its endogenous Il4 gene locus and the AmCyan fluorophore located in the Il4 locus of the BAC transgene. Using this system we were able to evaluate the impact of Il4-hemizygosity on CD4⁺ T cell IL-4 production by manipulating the number of intact wild type Il4 alleles and using the expression of AmCyan from the BAC transgene to identify those T cells that had become committed to IL-4 production. This was determined by comparing the frequencies of AmCyan⁺ CD4⁺ T cells in immunized 4C13R-IL-4^+/+, 4C13R-IL-4^G4/+, and 4C13R-IL-4^G4/G4 strains of mice.

Firstly, the previously observed effect of Il4-hemizygosity on OVA-induced IgE responses was confirmed in the 4C13R mouse strain using NP-OVA as the immunizing allergen. Serum levels of NP-specific IgE Ab were reduced 10-fold in 4C13R-IL-4^G4/+ mice compared with IL-4 sufficient 4C13R-IL-4^+/+ mice (Fig. 5A). Serum NP-specific IgG1 titers were similar in 4C13R-IL-4^+/+ and 4C13R-IL-4^G4/+ mice, but reduced 3–4 fold in IL-4 deficient 4C13R-IL-4^G4/G4 mice (Fig. 5B). When draining lymph node CD4⁺ T cell responses were evaluated in these mice, all three strains (4C13R-IL-4^+/+, 4C13R-IL-4^G4/+, and 4C13R-IL-4^G4/G4 mice) generated similar numbers of CD4⁺ T cells, AmCyan⁺ CD4⁺ T cells, and CXCR5⁺PD-1^hi TFH (Fig. 6A–E), indicating that even though Il4-hemizygosity impaired IgE production, similar numbers of CD4⁺ T cells were genetically committed to the production of IL-4. These AmCyan⁺ CD4⁺ T cells in the lymph node likely represent TFH (10, 12). This finding suggests that rather than Il4-hemizygosity impairing the development of IL-4 producing CD4⁺ TFH, it may be the reduced production of IL-4 by these TFH that affects IgE levels in the Il4-hemizygous mice.

To evaluate the ability of lymph node CD4⁺ Th cells to produce IL-4, we examined IL-4 production in NP-OVA/Alum immunized C57BL/6J IL-4^+/+, IL-4^G4/+ and IL-4^G4/G4 mice (Fig. 7A) and OVA/Alum immunized BALB/c-ByJ IL-4^+/+, IL-4^G4/+ and IL-4^G4/G4 mice (Fig. 7B). CD4⁺ T cells were purified by positive selection and cultured on anti-CD3 coated plates, supernatants were harvested and examined for IL-4 production by ELISA. Cultures of CD4⁺ T cells from either NP-OVA/Alum or OVA/Alum immunized IL-4^G4/+ mice produced between 2- and 6-fold less IL-4 than CD4⁺ T cell cultures from IL-4^+/+ mice. Because similar numbers of CD4⁺ T cells were shown to become committed to Il4 gene expression in IL-4^+/+ and IL-4^G4/+ mice, it would appear that the 2-fold reduction in the amount of IL-4 produced by Il4-hemizygous CD4⁺ T cells is sufficient to profoundly affect the subsequent serum IgE response.

We report that Il4-hemizygous mice exhibited severely compromised IgE responses but not IgG responses against OVA or hapten in OVA/Alum or hapten-OVA/Alum immunization models. Serum OVA or hapten-specific IgE was diminished by 10–100 fold in Il4-hemizygous mice, 6-fold fewer hapten-specific IgE ASC were generated, and hapten-specific IgE levels on mast cells and basophils were reduced 3–10 fold following immunization of Il4-hemizygous mice. The reduced levels of IL-4 produced by in vitro restimulated IL-4^G4/+ CD4⁺ T cells from allergen-immunized Il4-hemizygous mice supports the notion that it is the reduced amount of IL-4 produced by the lymph node CD4⁺ T cell compartment that leads to the IgE deficiency.

In addition to reduced IgE production, Il4-hemizygous mice were also protected from IgE-dependent oral allergen–induced diarrhea, but remained susceptible to i.v.-induced Ab-mediated active systemic anaphylaxis. Given that both the production of IgE and mastocytosis is a necessary step for diarrhea development in this model (1, 2, 35), it was not surprising that the attenuated mastocytosis combined with the deficiency in OVA-specific IgE production resulted in the Il4-hemizygous mice being protected from disease (31, 36–38). In the i.v. allergen model, Il4-hemizygous mice exhibited some signs of active systemic anaphylaxis when challenged i.v. with OVA. However, the allergen-induced hypothermia was generally less severe in Il4 deficient and hemizygous mice. In addition to influences of IgE on these responses, the reduced hypothermia may reflect an influence of Il4-hemizygosity on the IL-4–dependent sensitivity of vascular smooth muscle to mediators of anaphylaxis, including platelet-activating factor (6), which plays an important role in systemic anaphylaxis mediated by IgG (39).

The observation that normal levels of activation and development of IgG1 producing ASC occurred in Il4-hemizygous mice, whereas the development of IgE ASC was profoundly reduced, indicated that each response had different requirements for IL-4 and that a lower amount of IL-4 is sufficient for normal IgG1 production to protein/alum immunization, whereas IgE responses are very sensitive to reduced levels of IL-4. Consistent with this, anti-IL-4 Ab treatment can ablate the in vivo IgE response in mice without significantly impacting the IgG Ab response (7). Further, IgG1 Ab production is also unaffected by hemizygosity for the genes encoding the IL-4 signaling components IL4Rα and STAT6, whereas—similar to our findings for IL-4^+/− mice—haploinsufficiencies have been described for IgE responses in both Il4ra^+/− and Stat6^+/− mice (40, 41). Taken together, these data indicate there are critical threshold IL-4 and IL-4 signaling requirements for IgE synthesis that cannot be achieved when IL-4 and its signaling pathways are reduced by half.

The IgE-restricted haploinsufficiency we observed implies that some IL-4 dependent features of the immune response against allergens are compromised more significantly than would be predicted for the Il4-hemizygous state. Therefore, we sought to determine whether either the commitment of CD4⁺ T cells to IL-4 production or the expansion of TFH was affected by Il4-hemizygosity. In our analysis of IL-4–producing CD4⁺ T cell development in immunized Il4-hemizygous mice, we observed that there was a normal commitment of CD4⁺ T cells to Il4 gene expression, suggesting that the IgE deficiency was not caused by the impaired expansion of lymphoid CD4⁺ TFH. Instead, it seems likely that the amount of IL-4 the lymphoid CD4⁺ T cells produce does not reach the threshold required for stimulating the IL-4–dependent signals required for IgE class-switching.

How a threshold IL-4 signal for IgE switching could be so sensitive to the 2–6-fold reduction in IL-4 is not entirely clear. However, as IL-4 is reportedly secreted synaptically (42), the IL-4 concentration to which cognate B cells are exposed may be many times higher than the concentration to which non-cognate B cells in distal parts of the follicle are exposed; high level IL-4 exposure resulting from focused synaptic secretion may be critical for the induction of IgE class-switch recombination. Such super Th2 cells that could contribute high levels of IL-4 to intercellular synaptic events have been demonstrated to occur as part of the normal distribution of T cells expressing the Il4 gene either biallelically or monoallelically when activated. Although the majority of activated Th2 cells have been reported to follow a monoallelic expression pattern, a variable proportion (10–50%) of cells express IL-4 biallelically (25, 43–45). These biallelic Th2 cells would have the potential for very high IL-4 production, however, they would require expression from both Il4 alleles, and as such, would not be able to occur in Il4-hemizygous mice, perhaps explaining the profound loss of IgE production.

Our finding that IgE production is disproportionately affected by the genetic loss of one IL-4 producing Il4 gene copy may be relevant for why some individuals become sensitized to allergens. In humans, numerous polymorphisms and chromatin modifications surrounding IL-4 pathway genes correlate with serum IgE levels and allergic disease risk (for examples, see Ref. 46–51). It may be that some of the polymorphisms enhance or prolong IL-4/STAT-6 signaling to increase allergic disease risk, as has been suggested previously (52, 53). Minor changes in IL-4 signaling events may also be relevant for the associations observed between some IL-4Rα polymorphisms and IL-4Rα antagonist efficacy (13, 54). It also follows that more potent Th2 immunogens, such as parasitic hookworms, would generate significantly more IL-4 and make type 2 immune responses against these parasites more stable in the face of minor changes to IL-4 signaling thresholds.

In conclusion our findings suggest that Il4-hemizygous mice exhibit significantly attenuated IgE responses. Such observations indicate that agents able to reduce IL-4 to such levels have the potential to significantly attenuate allergen-induced allergic disease.

The authors have no financial conflicts of interest.