Induction of IgG₂ and IgG₄ B‐cell memory following sublingual immunotherapy for ryegrass pollen allergy

Abstract

Background

While treatment for atopic rhinitis is aimed mostly to relieve symptoms, only allergen‐specific immunotherapy (AIT) is targeted to modify the natural history of allergic diseases. This results in sustained clinical tolerance, even when treatment has stopped. The immunomodulatory effects of AIT are attributed mainly to increased regulatory T‐cell function and increased allergen‐specific IgG₄, yet little is known about the effect on the memory B‐cell compartment.

Objective

We aimed to examine the effects of AIT on the IgE‐ and IgG subclass‐expressing memory B cells.

Methods

We recruited 29 patients with atopic seasonal rhinoconjunctivitis and performed a longitudinal analysis of the peripheral immune compartment before, during, and after sublingual immunotherapy (SLIT) for allergy to temperate grass pollen, predominantly to ryegrass pollen (RGP; Lolium perenne). Using flow cytometry on peripheral blood mononuclear cells and serum immunoassays, we analyzed the effects of a 4 months preseasonal treatment regimen comprising two or three courses in consecutive years on circulating IgE⁺ and IgG⁺ memory B cells and allergen‐specific Ig levels.

Results

SLIT increased RGP‐specific serum IgG₂ and IgG₄, as well as the frequencies of IgG₂ ⁺ and IgG₄ ⁺ memory B cells, whereas no effect was observed on the IgE⁺ memory B‐cell compartment. Furthermore, SLIT enhanced proportions of regulatory T cells specific to RGP. These changes were associated with clinical improvement.

Conclusion

Our data provide evidence for immunological effects of SLIT on B‐cell memory. Skewing responses toward IgG₂ and IgG₄ subclasses might be a mechanism to suppress IgE‐mediated allergic responses.

This study examines the effect of ryegrass pollen AIT on B‐cell responses in a population of 29 patients with allergic rhinitis. Successful immunotherapy for ryegrass pollen allergy increases allergen‐specific IgG₂ and IgG₄ serum levels, and proportions of IgG_2‐ and IgG₄‐expressing memory B cells. Skewing toward the anti‐inflammatory IgG₂ and IgG₄ subclasses might be a mechanism to suppress IgE‐mediated allergic responses.

Abbreviations

AIT

allergen‐specific immunotherapy

FeNO

fractional exhaled nitric oxide

IL

interleukin

RGP

ryegrass pollen

SCIT

subcutaneous immunotherapy

SHM

somatic hypermutation

SLIT

sublingual immunotherapy

SPT

skin prick test

Th1/2

T helper 1/2

Treg

regulatory T cell

VAS

visual analog score

1. INTRODUCTION

Rhinoconjunctivitis and other IgE‐mediated allergies are an increasing disease burden globally.1 Most therapies for allergies are directed at relieving symptoms, but allergen‐specific immunotherapy (AIT) is the only current therapy that modifies the natural course of allergic diseases. Subcutaneous immunotherapy (SCIT) and sublingual immunotherapy (SLIT) are both proven effective treatments for grass pollen‐induced rhinoconjunctivitis.2, 3, 4 The therapeutic effect is maintained beyond the conclusion of treatment.5, 6, 7 In patients with allergic rhinitis, AIT can prevent the onset of new sensitizations8 and decrease the likelihood of developing asthma.9 The immunomodulatory properties of AIT affect local and systemic immune responses, with an impact on the number and function of mast cells, basophils, antigen‐presenting cells, T cells, and B cells.10, 11

Allergic patients manifest sensitization by means of allergen‐specific IgE bound to effector cells, particularly mast cells and basophils.12 The underlying mechanism is thought to be a shifted T‐cell balance toward a T helper 2 (Th2) phenotype, and these cells produce interleukin (IL)‐4 and IL‐13 that direct allergen‐specific B cells to produce IgE.13 Furthermore, Th2 cells produce IL‐5 which promotes the involvement of eosinophils in the pathogenesis of allergic diseases.14 In contrast, Th1 responses are promoted by IFN‐γ and skew away from a Th2 phenotype.15 Effective immunotherapy has been shown to reverse the Th2 dominance and to result in anergy of allergen‐specific T cells,16, 17 induction of regulatory T cells (Treg),18, 19, 20, 21 and production of blocking antibodies of the IgG and IgA isotypes.22, 23 Specifically, TGF‐β and IL‐10 produced by Treg are pivotal for the successful immune deviation in AIT.24, 25

The tolerogenic functions of IL‐10 are extensive, but mainly encompass the inhibition of mast cell activity,26 suppression of IL‐5 production by Th2 cells,27 and cell death induction in eosinophils.28 Furthermore, IL‐10 in combination with IL‐4 and IL‐13 directs B‐cell immunoglobulin class switching to IgG₄ instead of IgE.29 Indeed, one of the known effects of AIT is an increase in allergen‐specific serum IgG₄ and an increased serum IgG₄/IgE antibody ratio that is associated with clinical efficacy.30

SCIT and SLIT have distinct immunomodulatory capabilities that appear related to the different routes of administration. Sublingual administration results in fewer systemic adverse effects, but some studies indicate diminished clinical and immunological efficacy compared with subcutaneous administration.2, 31 SLIT results in increased numbers of FoxP3⁺ Treg both in the oral epithelium and in the peripheral blood.23, 32 Further systemic alterations are more diverse. Some studies report an initial increase in allergen‐specific IgE serum levels, followed by a decrease after 1 month.33 Furthermore, allergen‐specific IgG₂, IgG_4, and IgA serum levels are reported to increase in as little as 1 day after the start of therapy.33, 34, 35 However, other studies detected no systemic alterations with regard to allergen‐specific lymphoproliferation, cytokine secretion, or Ig serum levels.36, 37

IgG₂ and IgG₄ heavy chain constant regions are encoded by genes in the IGH locus. Ig class switching to IgG₂ and IgG₄ frequently occurs indirectly following a switch from IgM to the more proximal IgG₃ and IgG₁ genes rather than directly from IgM to IgG₂ or IgG₄.38 Given the higher loads of somatic hypermutation (SHM) in variable regions of IgG₂ and IgG₄ transcripts, it has been suggested that B cells expressing these transcripts have spent more time in the germinal center response.39 In addition, the majority of IgG₂‐ and IgG₄‐expressing B cells co‐express CD27, and their frequencies increase with age.40, 41 Hence, it appears that these Ig class switches occur following repeated exposure to the same antigen.

Since AIT has been shown to have long‐lasting beneficial effects, it is important to determine whether this is the result of changes in immunological memory. We here address this question in our cohort of patients with moderate‐to‐severe seasonal allergic rhinitis, studied longitudinally before, during, and after SLIT for grass pollen allergy.42 As published previously,42, 43 SLIT in our cohort resulted in allergic rhinitis symptom relief and conferred significant protection from epidemic thunderstorm asthma, making this an ideal cohort to examine the effects of a 4‐month treatment regimen and the subsequent effects of two further courses of treatment over 3 years on circulating IgE⁺‐ and IgG subclass‐expressing memory B cells and allergen‐specific Ig levels.

2. METHODS

2.1. Study design

Using an open‐label longitudinal design (ClinicalTrials.gov identifier: NCT02014623), 29 participants were recruited for treatment with a commercial 5‐grass pollen SLIT tablet (Oralair^®; Stallergenes) using a 4‐month (May‐September) regimen completed prior to the Australian pollen season, for 3 consecutive years (2014‐2016; subject numbers at each time point shown in Figure 1A). Treatment with Oralair^® involved dissolution under the tongue (at least 2 minutes) followed by swallowing the residue. The treatment regimen comprised the following: day 1—1 tablet 100 IR (index of reactivity); day 2—2 tablets 100 IR; and day 3 to day 120—1 daily tablet 300 IR. Blood samples were collected immediately before initial treatment (May 2014) and after the first 4 months of treatment (September 2014), followed by annual collections in May 2015 and May 2016 (prior to commencement of 2nd and 3rd courses of SLIT), and May 2017 (Figure 1A).

Figure 1.

Study design and clinical parameters of allergic rhinitis decreased after SLIT. A, Timeline of SLIT for grass pollen allergy between May 2014 and 2017. Time points indicate blood sampling. B, Allergic rhinitis symptoms by visual analog scale measured during peak pollen season. C, Wheal diameter (in mm) from skin prick test (SPT) with RGP. D, Fractional exhaled nitric oxide (FeNO) measured immediately prior to starting SLIT. E, Total IgE in serum. Each dot represents one individual; red lines indicate median values. Statistical analysis was performed between baseline and each follow‐up time point to assess changes induced by SLIT using the Wilcoxon signed‐rank test; *P < .05, **P < .01, and ***P < .001

2.2. Participant characteristics

Participants were recruited from The Alfred Hospital Allergy Clinic, Melbourne, Victoria, Australia. All had well‐characterized moderate‐to‐severe seasonal allergic rhinitis (plus or minus asthma) due to RGP allergy with positive serum RGP‐specific IgE (≥0.35 kU/L; ImmunoCAP, Phadia). Exclusion criteria were a co‐existing immunodeficiency, previous immunotherapy within the last 5 years, ongoing immunotherapy with other allergens, and treatment with continuous oral corticosteroids and/or β‐blockers. The use of usual medications for allergic rhinitis was permitted, including antihistamines and topical corticosteroids. Alfred Hospital Research and Ethics Committee approval and written informed consent from each participant were obtained prior to inclusion (project number 514/13). Twenty‐nine participants (12 males) were recruited for treatment, with a mean age of 35 years (range 18‐59 year) and mean serum RGP‐specific IgE of 52 kU/L. Five withdrew after the first (baseline) time point (n = 4 tongue swelling, upset stomach; n = 1 failed to attend) leaving 24 participants who commenced SLIT (Table S1). At subsequent time points, three participants failed to attend after 4 months of treatment and a further seven participants were excluded after 1 year (n = 2 opted to receive SCIT, n = 2 opted to receive sublingual drops, and n = 3 withdrew). Blood samples for serum and flow cytometric analysis were obtained from n = 24 patients in May 2014 prior to starting SLIT, n = 24 in September 2014, n = 21 in May 2015, n = 14 in May 2016, and n = 14 in May 2017. Details on sample numbers for each analysis are included in Table S2. A further 5 RGP‐allergic subjects who did not receive SLIT (ie, received usual medication alone) were included as untreated patients at baseline and 4 months.

2.3. Clinical parameters of allergic rhinitis

Allergic rhinitis symptoms during the peak RGP season were recorded by the participants using a visual analog score (VAS; scale, 0‐100). Fractional exhaled nitric oxide (FeNO, in parts per billion [ppb]; HypoAirFeNO) was measured according to the manufacturer's instructions (NIOX, Uppsala, Sweden). FeNO was measured immediately before the start of SLIT therapy outside of the grass pollen season to minimize effects of daily fluctuations in pollen levels.

2.4. Quantification of serum total IgE and allergen‐specific IgE, IgG₂, and IgG₄

Serum total IgE, RGP‐specific‐IgE, and ‐IgG₄ levels were measured by ImmunoCAP. Serum RGP‐specific IgG₂ antibodies were measured by in‐house ELISA, as described previously.44 Briefly, ELISA plate wells were coated with an aqueous RGP extract (Stallergenes Greer), blocked with 2% bovine serum albumin in PBS (Sigma‐Aldrich), and incubated with serial dilutions of serum samples. Separate wells were coated with serial dilutions of purified human IgG₂ (Sigma‐Aldrich, #I5404) to generate a standard curve for quantification of IgG₂ in serum samples. Bound IgG₂ was detected using biotinylated anti‐hIgG₂ (clone HP6002; Thermo Scientific) followed by Pierce High Sensitivity Streptavidin‐HRP (Thermo Scientific). ELISA was developed using TMB (Thermo Scientific), and the reaction stopped with 1 mol/L HCl. Absorbance (OD 450 nm) was measured using a FLUOstar Optima plate reader (BMG Labtech).

2.5. In vitro RGP stimulation of PBMC, Treg staining, and measurement of cytokines

PBMC were isolated by Ficoll‐paque density centrifugation. Fresh PBMC were used for in vitro culture, and the remaining cells stored in liquid nitrogen. PBMC were labeled with CFSE (0.5 µmol/L CFSE/10⁷ PBMC; Molecular Probes) and cultured with an aqueous RGP extract (50 μg/mL; Stallergenes Greer) or tetanus toxoid (20 Lfu/mL; Statens Serum Institut, Copenhagen, Denmark). On day 7, cells were stained with CD4‐PE Cy7, CD25‐PE (both from BD Biosciences), FoxP3‐APC (eBioscience), and aqua live/dead dye (Life Technologies). The Treg gating strategy is shown in Figure S1. Data were acquired using an LSR‐II flow cytometer (BD Biosciences).

The levels of IFN‐γ, IL‐5, IL‐10, and IL‐13 in 7‐day culture supernatants were determined using a Luminex human premixed multi‐analyte kit (R&D Systems Inc) according to the manufacturer's instructions. Due to changes in IL‐5 production observed after 3 years of SLIT, IL‐13 was also assessed at the same time point to further investigate Th2 cytokine production. Tetanus toxoid was included as a control antigen to determine RGP specificity. “No antigen” values were subtracted from test values.

2.6. B‐cell subset analysis by flow cytometry

One million thawed PBMC were incubated with 11‐color antibody cocktails against B‐cell markers for 15 minutes at room temperature in 100 µL total volume (Table S3). Flow cytometric analyses were performed on a 4‐laser LSRFortessa (BD Biosciences), and data were analyzed using FACSDiva V8.0 (BD Biosciences). B‐cell subsets were defined as described previously.41, 45, 46 Briefly, within the CD19⁺ B‐cell population, the proportions were determined of plasmablasts (CD27⁺CD38^high), transitional (CD27⁻CD38^high), naive mature (CD27⁻IgM⁺IgD⁺), natural effector memory B cells (CD27⁺IgM⁺IgD⁺), and IgM‐only memory B cells (CD27⁺IgM⁺IgD^–). Furthermore, we analyzed Ig switched CD27⁻CD38^dim and CD27⁺CD38^dim memory B cells expressing IgA, IgE, IgG, or each of the 4 IgG subclasses.

2.7. Molecular analysis of Ig gene rearrangements

RNA was isolated from PBMC from a limited cohort of 5 subjects treated with SLIT (Table EI; patient no. 1, 11, 13, 15, and 19) with a GenElute mammalian RNA kit (Sigma‐Aldrich) and reverse transcribed to cDNA with random primers (Invitrogen Life Technologies, Waltham, MA). Rearranged IgG transcripts were amplified in a multiplex PCR approach using 4 different IGHV family leader forward primers in combination with an IGHG‐consensus reverse primer.47 PCR products were cloned into a pGEMT easy vector (Promega), amplified by colony PCR, and sequenced by the Micromon facility of Monash University on an Applied Biosystems 3730s DNA Analyzer (Thermo Scientific). Obtained sequences were analyzed using the IMGT database (http://www.imgt.org) to assign the IGHV, IGHD, and IGHJ gene alleles and to identify SHM. For each unique clone, the position and frequency of mutations were determined within the entire IGHV gene (FR1‐CDR1‐FR2‐CDR2‐FR3). SHM was determined as variations on the best‐matched V‐gene and represented as the percentage of mutations of the total sequenced V‐gene nucleotides. The IgG subclasses were determined using the IGH reference sequence (NG_001019).

2.8. Statistical analysis

Differences in symptom scores, serum Ig values, cytokines, and B‐ and T‐cell subsets before, during, and after treatment were analyzed with the Wilcoxon signed‐rank test. All analyses were two‐tailed, and differences were considered statistically significant if P‐values were <.05. Due to missing values at year 1, 2, and 3 measures, it was not possible to use repeated measures ANOVA. Therefore, we performed pairwise analysis between time point 0 and each follow‐up sample. Differences in IgG subclass usage of unique IGH transcripts were statistically analyzed with the chi‐squared test. Statistical analysis was performed using GraphPad Prism software, version 7.01 (GraphPad Software).

3. RESULTS

3.1. SLIT reduces symptoms of allergic rhinitis

To study the clinical effects of SLIT, we assessed the severity of symptoms for allergic rhinitis using a VAS. Before the start of treatment, participants reported a median VAS of 80 mm for the 2013 pollen season (Figure 1B). In the first pollen season after commencing SLIT, participants experienced fewer symptoms (median VAS 40 mm, P < .001), and these remained low for the second and third seasons following repeat SLIT courses (median VAS 20 mm at 2 years, P < .001; median VAS 35 mm at 3 years, P < .01), confirming sustained clinical efficacy.

The level of RGP sensitization was monitored by skin prick tests (SPT) with wheal diameters (mm) positively correlating with symptoms of allergic disease.48 SLIT significantly decreased SPT wheal diameter in response to SPT with RGP extract within 1 year of commencing therapy (Figure 1C). Wheal size remained low on retesting after the third year of SLIT. In addition, the severity of airway inflammation and bronchial hyperreactivity was assessed by measurement of FeNO.49, 50 SLIT significantly decreased FeNO from baseline at 1, 2, and 3 years after commencing SLIT (Figure 1D). The significant decreases in SPT wheal diameter and FeNO were consistent with decreased symptoms of allergic rhinitis 1 year after commencing treatment as well as after the second and third successive years of SLIT. The reduction in VAS and FeNO after SLIT did not correspond with any changes in total serum IgE levels (Figure 1E).

3.2. Induction of RGP‐specific Treg and IL‐10 production following SLIT

Successful AIT has been associated with proliferation of allergen‐specific Treg.50 To assess Treg proliferation in response to allergen, we stimulated PBMC with RGP and measured the proliferation of activated Treg (CD4⁺CD25⁺FoxP3⁺) using CFSE. RGP‐induced proliferation of Treg was enhanced by SLIT after 4 months and remained raised throughout subsequent years (Figure 2A).

Figure 2.

Sublingual immunotherapy (SLIT) alters in vitro Treg proliferation and cytokine production in response to RGP. PBMC stimulated with RGP were assessed for A, Treg proliferation and production of B, IL‐10, C, IFN‐γ, and D, IL‐5 were determined for all patients included at t = 0, 1, 2, 3 y. E. Paired analysis of IL‐5 and F. IL‐13 at t = 0 and t = 3 y. The lower limit of detection of IL‐10 levels (0.5 pg/mL) in panel B is depicted by a dashed line and datapoints representing undetectable levels are placed below it. Statistical analysis was performed between baseline and each follow‐up time point to assess changes induced by SLIT using the Wilcoxon signed‐rank test; *P < .05, **P < .01, ***P < .001 and ****P < .0001

Given the role of cytokines in skewing T‐cell responses, we quantified IL‐5, IL‐10, IL‐13, and IFN‐γ production from RGP‐stimulated PBMC. After 3 years of SLIT, T cells produced significantly more IL‐10 and IFN‐γ and significantly less IL‐5 and IL‐13 (Figure 2B‐E). SLIT did not alter the PBMC cytokine response to tetanus toxoid (Figure S2). Taken together, the data from our study suggest SLIT enhances Treg response to allergen within 4 months without impacting the Treg response to other antigens. Furthermore, SLIT may alter the Th1/Th2 cytokine profile away from pro‐allergic Th2 cytokines toward a regulatory and Th1‐biased response.

3.3. Increased IgG₄ serum levels and IgG₄ ⁺ memory B‐cell frequencies after 4 months SLIT

To study the short‐term effects of SLIT on the immune system, we analyzed serum Ig levels and B‐cell subsets before and directly after the first 4 months of therapy. The gating strategies for flow cytometric detection of memory B cells expressing the four IgG subclasses are shown in Figure 3A. We observed that after 4 months of immunotherapy, RGP‐specific serum IgG₂ increased from a median of 2.46 to 5.08 μg/mL (Figure 3B). Furthermore, all participants showed an increase in RGP‐specific serum IgG₄ from a median of 0.37 μg/mL pretreatment to 1.16 μg/mL post‐treatment after 4 months of SLIT. This was accompanied by a significant increase in the frequency of IgG⁺ memory B cells (CD19⁺CD38^dim) expressing IgG₄ (Figure 3C). The increase in the IgG₄ ⁺ memory B‐cell frequencies was not directly correlated with the increase in RGP‐specific serum IgG₄ (p > .05). SLIT did not change IgE⁺ memory B‐cell frequencies (Figure 3D). However, the increase in IgG₄ ⁺ memory B cells resulted in a significantly higher IgG₄ ⁺/IgE⁺ memory B‐cell ratio following 4 months of treatment (Figure 3E). The frequencies of all other B‐cell subsets, including transitional, naive mature, memory, and plasmablasts, remained unchanged after 4 months of SLIT (Figure S3). Thus, 4 months SLIT quite specifically affected allergen‐specific IgG₄ serum levels and the frequencies of IgG₄‐expressing memory B cells.

Figure 3.

RGP‐specific IgG₄ and IgG₄ ⁺:IgE⁺ memory B‐cell ratio increased after a 4‐mo course of SLIT. A, Gating strategy for Ig isotype and IgG subclass‐expressing memory B cells by flow cytometry. B, RGP‐specific IgG₂ and IgG₄ in sera and C, proportion of IgG₂ ⁺ and IgG₄ ⁺ memory B cells as a percentage of IgG⁺ population after 4 mo of SLIT. D, Proportion of IgE⁺ memory B cells as a percentage of total CD19⁺ B cells after 4 mo of SLIT. E, Ratio of IgG₄ ⁺ to IgE⁺ B‐cell percentages after 4 mo of SLIT (from C and D). Each dot represents one individual; red lines indicate median values. Statistical analysis was performed between baseline and each follow‐up time point to assess changes induced by SLIT using the Wilcoxon signed‐rank test. *P < .05 and ****P < .0001

3.4. SLIT has persistent long‐term effects on IgG₂ and IgG₄ memory B cells

In addition to short‐term effects of SLIT, we studied the longer‐term effects of SLIT, that is, 1, 2, and 3 years after the start of the first treatment course. SLIT did not significantly alter serum RGP‐specific IgE levels (Figure 4A). RGP‐specific IgG₂ levels increased after a total of 3 courses of SLIT (Figure 4A). RGP‐specific IgG₄ increased after each consecutive course of treatment at 1, 2, and 3 years (Figure 4A). Similar to RGP‐specific IgE and IgG₂ antibodies, frequencies of IgE⁺ memory B cells were unchanged by SLIT, while IgG₂ ⁺ memory B cells were significantly increased 3 years after commencing SLIT (Figure 4B). Frequencies of IgG₄ ⁺ memory B cells were increased 2 years after commencing SLIT.

Figure 4.

Persistent increase in IgG₄ and late rise in IgG₂ after three 4‐mo SLIT courses. A, RGP‐specific IgE, IgG₂ and IgG₄ in serum. B, Proportions of IgE⁺, IgG₂ ⁺, and IgG₄ ⁺ memory B cells in peripheral blood. IgE⁺ memory B cells presented as a percentage of CD19⁺ B cells. IgG₂ ⁺ and IgG₄ ⁺ memory B cells presented as a percentage of IgG⁺ memory B cells. Baseline data for IgE⁺ and IgG4⁺ B cells are the same as those in Figure 3C and D. Statistical analysis was performed between baseline and each follow‐up time point to assess changes induced by SLIT using the Wilcoxon signed‐rank test; *P < .05 and **P < .01

3.5. Molecular analysis of Ig gene rearrangements

Given that SLIT increased IgG₂ and IgG₄ antibodies and memory B‐cell proportions, we investigated whether these changes were reflected in the proportions of unique IgG transcripts from blood B cells for a subgroup of 5 participants with an increased percentage of IgG₂ ⁺ memory B cells after SLIT. Median frequencies of somatic hypermutations did not differ between the proximal IgG₁ and the distal IgG₂ subclasses (Figure 5A and B) nor were significantly different after 3 years SLIT. However, after 3 years SLIT the relative usage of the IgG₂ and IgG₄ subclasses were significantly increased at the expense of IgG₁ (Figure 5C). Taken together, these data demonstrate that repeat courses of SLIT for grass pollen allergy induce allergen‐specific IgG₂ and IgG₄ responses, evidenced by an increase in IgG₂ ⁺ and IgG₄ ⁺ B‐cell proportions and skewing toward unique IgG₂ and IgG₄ transcripts.

Figure 5.

Sublingual immunotherapy (SLIT) increases frequency of unique IgG2 transcripts. A, Schematics of the human IGH locus depicting the positioning of the constant gene regions relative to the rearranged VDJ exon. B, Somatic hypermutation frequencies of unique IgG transcripts obtained from 5 patients before (2014) and after SLIT (2017) and grouped per IgG subclass. C, Relative isotype distribution of unique IgG transcripts. Central number indicates total unique IgG sequences identified. Significance was determined by chi‐squared test; **P = .0011

4. DISCUSSION

We here report that SLIT for grass pollen allergy not only has long‐term beneficial clinical effects, but also results in sustained systemic effects on the immune system. SLIT induced a rapid and prolonged increase in RGP‐specific serum IgG₄ accompanied by an increase in the frequency of peripheral blood IgG₄ ⁺ memory B cells. Furthermore, repeat courses of SLIT resulted in a similar increase in RGP‐specific IgG₂ in serum corresponding with increased frequency of IgG₂ ⁺ memory B cells in the blood.

Currently, grass pollen SLIT is recommended as a preseasonal and co‐seasonal course starting 4 months prior to the hay fever season, confirmed by meta‐analyses as clinically effective.2 Yet, long‐term treatment regimens are costly and discourage treatment adherence.51 As patients are exposed to grass pollens during the spring season, we reasoned that a 4 months preseasonal treatment regimen would avoid the risk of adding to excessive and unpredictable allergen loads during the Melbourne Spring. Based on our analysis of symptom scores, this approach is highly effective.42, 43 Prolonged treatment (duration > 12 months) is known to have beneficial effects on symptom and medication scores.52 The fact that some immunological effects are delayed, only occurring after the second or third treatment year as observed for serum RGP‐specific IgG₂ levels, or continuing to rise after consecutive treatment as for serum RGP‐specific IgG₄ levels, supports these premises.

In particular, we observed a marked increase in RGP‐specific IgG. Previously, allergen‐specific immunotherapy, either SCIT or SLIT, has already been demonstrated to result in increased allergen‐specific IgG₄ serum levels.53, 54 Increased allergen‐specific IgG₄ has been postulated as one of the explanations for the beneficial effects of immunotherapy and has been observed as a natural effect in beekeepers exposed to bee venom for prolonged periods,55 yet the exact desensitizing effect of specific IgG₄ in immunotherapy remains unclear. Allergen‐specific IgG₄ can competitively inhibit IgE from binding to allergens and may subsequently reduce allergic responses by preventing FcεR‐mediated activation of granulocytes.56 Furthermore, IgG₄ antibody has been proposed to inhibit inflammatory responses by preventing C1q complement activation and binding to the inhibitory receptor FcγRIIb (CD32b).57, 58

SLIT also increased RGP‐specific IgG₂ after three consecutive courses of SLIT. This suggests repeated or high‐dose exposure to RGP from SLIT is required to enhance RGP‐specific IgG₂ beyond that which is generated from annual RGP exposure during the pollen season. Furthermore, sublingual administration of RGP may have preferentially induced an IgG₂ response not seen from environmental exposure through the airway.

The immune mechanisms by which allergen‐specific IgG₂ may contribute to the benefits of immunotherapy remain unclear. IgG₂ has been shown to inhibit histamine release from basophils by activating FcγRIIb and may reduce allergic symptoms by this mechanism.59 In a similar manner to IgG₄, IgG₂ may also bind allergen and prevent effector cell degranulation by masking IgE epitopes.

The source of increased serum IgG₂ and IgG₄ levels is IgG₂‐ and IgG₄‐producing plasma cells, respectively. As the majority of serum IgG is produced by bone marrow residing plasma cells, we were unable to assess these. However, we did assess the more immature plasmablasts in blood, finding that their frequencies were not affected by SLIT. Further characterization of Ig isotypes and IgG subclasses was not possible due to our approach for membrane staining, since the majority of plasmablasts (especially those producing IgG) lack surface Ig expression. Hence, we focused on the analysis of memory B cells, which are abundant in the blood due to their circulatory nature, and their capacity to quickly differentiate into plasma cells in subsequent antigen responses.60

We observed that SLIT drives increased frequencies of IgG₂ ⁺ and IgG₄ ⁺ memory B cells, whereas there was no effect on frequencies of IgE⁺ memory B‐cell subsets. The latter observation can explain the absence of a decline in IgE serum levels as also demonstrated by others.52, 61, 62 Since we observed the increase in IgG₄ ⁺ memory B cells after 4 months of treatment with SLIT, and before the pollen season, this effect can be directly attributed to the treatment with Oralair^®. Our observation that frequencies of IgG₄ ⁺ memory B cells remain increased for at least 3 years can be an explanation for the long‐lasting effects attributed to immunotherapy.5, 6, 7

Allergen‐specific IgG₂ and IgG₄ appear to be robust markers of repeated allergen exposure. Increases in allergen‐specific IgG₂ and IgG₄ alongside increased frequency of IgG₂ ⁺ and IgG₄ ⁺ memory B cells may arise from class switching of allergen‐specific IgG₁ ⁺ memory B cells upon repeated exposure to allergen. Sequential Ig class switching can only occur 5’–3’ along the IGH locus, which is arranged in the following order: IgG₃ > IgG₁ > IgG₂ > IgG₄ > IgE > IgA (Figure 5A). As such, IgG₁ ⁺ B cells may switch to IgG₂ or IgG₄ but not vice versa. In a study of allergen‐specific antibodies in children from birth to 10 years old, IgE responses to aeroallergens were typically preceded by IgG.63 Allergen‐specific IgG⁺ memory B cells may therefore provide a reservoir for switching to IgE that can be induced by repeated exposure to the allergen and subsequently cause allergic sensitization. A similar pathway may give rise to allergen‐specific IgG₂ and IgG₄, whereby allergen exposure promotes class switching of allergen‐specific IgG₁ ⁺ memory B cells to IgG₂ and IgG₄. Switching to IgG₂ and IgG₄ may help explain why many allergies subside with age, perhaps due to repeated allergen exposure throughout childhood. However, it remains to be determined whether allergen‐specific IgG₂ and IgG₄ are indispensable for generating clinical benefit by SLIT in lieu of T cell–mediated tolerance.

In line with previous reports, we observed that proliferation of Treg from patients after SLIT was increased in response to in vitro stimulation with RGP. Previous studies have observed new generation of allergen‐specific Treg, as well as clonal expansion of allergen‐specific Treg in response to AIT.56, 64 Furthermore, we observed that SLIT increased IL‐10 production from RGP‐stimulated PBMC, whereas IL‐13 was diminished. The cytokines IL‐4 and IL‐13 promote Ig class switch recombination (CSR) to IgE, and these are predominantly produced by Th2 cells. In addition to Th2 cytokines, CSR to IgG₄ is also regulated by IL‐10 which is predominantly secreted by Treg. Our data are consistent with SLIT‐induced proliferation of allergen‐specific Treg, whereby increased IL‐10 production induces Ig class switching of allergen‐specific B cells to IgG₄. B regulatory cells (Breg) and monocyte‐derived macrophages may also be an alternative source of IL‐10 in our in vitro assay, further enhancing IgG₄ class switching in response to RGP.65, 66

In conclusion, our data provide evidence for long‐lasting effects of allergen SLIT on the memory compartment of the immune system. Increased Treg frequencies and increased IL‐10 production were associated with increased frequency of IgG₄ ⁺ memory B cells and a beneficial shift in the IgG₄ ⁺/IgE⁺ memory B‐cell ratio, reflecting the increased IgG₄/IgE antibody fraction in serum and resultant clinically favorable outcome. Moreover, to our knowledge, our study is the first to demonstrate increases in memory B cells expressing IgG₂ or IgG₄ following allergen immunotherapy. As IgG₂ and IgG₄ have anti‐inflammatory properties and are induced following repeated antigen‐exposure,41 this B‐cell memory compartment is a potential mechanism by which allergen immunotherapy modifies the natural course of disease. In future studies, it would therefore be of interest to examine the functional properties of allergen‐specific B cells, as well as the effector functions of allergen‐specific IgG subclasses.67

CONFLICT OF INTEREST

All authors declare that no conflict of interest exists.

Supporting information

 

 

 

 

Heeringa JJ, McKenzie CI, Varese N, et al. Induction of IgG₂ and IgG₄ B‐cell memory following sublingual immunotherapy for ryegrass pollen allergy. Allergy. 2020;75:1121–1132. 10.1111/all.14073