IgG1 B‐cell immunity predates IgE responses in epicutaneous sensitization to foods

Abstract

Background
The generation of IgE‐mediated food allergy in humans is silent and only diagnosed upon manifestation of clinical symptoms. While experimental models have been used to investigate some mechanisms of allergic sensitization, the generation of humoral immunity and memory remains to be elucidated. Here, we defined the evolution of allergen‐specific B‐cell responses during epicutaneous sensitization to foods.

Methods
Wild‐type and genetic knockout animals, and drug or antibody strategies for cell depletion and immunoglobulin signaling blockade were used to investigate epicutaneous sensitization and disease progression; we analyzed allergen‐specific germinal centers and IgG1⁺ memory B cells by flow cytometry, evaluated humoral responses, and determined clinical reactivity (anaphylaxis).

Results
Epicutaneous sensitization caused microscopic skin damage, inflammation, and recruitment of activated dendritic cells to the draining lymph nodes. This process generated allergen‐specific IgG1⁺ germinal center B cells, serum IgG1, and anaphylaxis that was mediated by the alternative pathway. Whether we used peanut and/or ovalbumin from the egg white for sensitization, the allergen‐specific IgG1⁺ memory compartment predominantly exhibited an immature, pro‐germinal center phenotype (PDL‐2⁻CD80⁻CD35⁺CD73⁺). Subsequent subclinical exposures to the allergen induced IgE⁺ germinal center B cells, serum IgE, and likely activated the classical pathway of anaphylaxis.

Conclusions
Our data demonstrate that IgG1⁺ B‐cell immunity against food allergens in epicutaneous sensitization precedes the generation of IgE responses. Therefore, the assessment of allergen‐specific cellular and humoral IgG1⁺ immunity may help to identify individuals at risk of developing IgE‐mediated food allergy and hence provide a window for therapeutic interventions.

Full text

ORIGINAL ARTICLE
Food Allergy and Gastrointestinal Disease
IgG1
+
B-cell immunity predates IgE responses in epicutaneous
sensitization to foods
R. Jim
enez-Saiz
1
|
Y. Ellenbogen
1
|
J. F. E. Koenig
1
|
M. E. Gordon
1
|
T. D. Walker
1
|
D. Rosace
1
|
P. Spill
1
|
K. Bruton
1
|
J. Kong
1
|
K. Monteiro
1
|
J. Wen
1
|
E. I. Tuomanen
2
|
R. Kolbeck
3
|
D. K. Chu
4
|
S. Waserman
4
|
M. Jordana
1
1
Department of Pathology & Molecular
Medicine, McMaster Immunology Research
Centre (MIRC), McMaster University,
Hamilton, ON, Canada
2
Department of Infectious Diseases, St.
Jude Children’s Research Hospital,
Memphis, TN, USA
3
Department of Respiratory, Inflammation
& Autoimmunity, MedImmune LLC,
Gaithersburg, MA, USA
4
Department of Medicine, McMaster
University, Hamilton, ON, Canada
Correspondence
Manel Jordana, Department of Pathology &
Molecular Medicine, McMaster Immunology
Research Centre, McMaster University,
Hamilton, ON, Canada.
Email: jordanam@mcmaster.ca
Funding information
This work was supported by grants from
Food Allergy Canada, MedImmune LLC,
AllerGen NCE, the Delaney family, the Zych
family, and the Walter and Maria Schroeder
Foundation.
Abstract
Background: The generation of IgE-mediated food allergy in humans is silent and only
diagnosed upon manifestation of clinical symptoms. While experimental models have
been used to investigate some mechanisms of allergic sensitization, the generation of
humoral immunity and memory remains to be elucidated. Here, we defined the evolu-
tion of allergen-specific B-cell responses during epicutaneous sensitization to foods.
Methods: Wild-type and genetic knockout animals, and drug or antibody strategies
for cell depletion and immunoglobulin signaling blockade were used to investigate
epicutaneous sensitization and disease progression; we analyzed allergen-specific
germinal centers and IgG1
+
memory B cells by flow cytometry, evaluated humoral
responses, and determined clinical reactivity (anaphylaxis).
Results: Epicutaneous sensitization caused microscopic skin damage, inflammation,
and recruitment of activated dendritic cells to the draining lymph nodes. This pro-
cess generated allergen-specific IgG1
+
germinal center B cells, serum IgG1, and ana-
phylaxis that was mediated by the alternative pathway. Whether we used peanut
and/or ovalbumin from the egg white for sensitization, the allergen-specific IgG1
+
memory compartment predominantly exhibited an immature, pro-germinal center
phenotype (PDL-2

CD80

CD35
+
CD73
+
). Subsequent subclinical exposures to the
allergen induced IgE
+
germinal center B cells, serum IgE, and likely activated the
classical pathway of anaphylaxis.
Conclusions: Our data demonstrate that IgG1
+
B-cell immunity against food aller-
gens in epicutaneous sensitization precedes the generation of IgE responses. There-
fore, the assessment of allergen-specific cellular and humoral IgG1
+
immunity may
help to identify individuals at risk of developing IgE-mediated food allergy and
hence provide a window for therapeutic interventions.
KEYWORDS
anaphylaxis, epicutaneous sensitization, IgE-mediated food allergy, IgG1
+
memory B cells, Th2
immunity
Abbreviations: Ag, antigen; DCs, dendritic cells; FMO, fluorescence minus one; GC, germinal center; Ig, immunoglobulin; iLN, inguinal lymph node; KO, knockout; MC, mast cell; OVA,
ovalbumin; PAF, platelet-activating factor; PCs, plasma cells; Th2, T helper type 2; WT, wild type; CPE, crude peanut extract; Ig, immunoglobulin.
Accepted: 14 May 2018
DOI: 10.1111/all.13481
Allergy. 2019;74:165–175. wileyonlinelibrary.com/journal/all ©2018 EAACI and John Wiley and Sons A/S.
Published by John Wiley and Sons Ltd.
|
165

1
|
INTRODUCTION
Food allergy is a worldwide health and economic concern. It currently
affects ~15 million Americans, and its prevalence is increasing.
1-3
The
clinical manifestations of food allergy are primarily mediated by
immunoglobulin (Ig) E with symptoms ranging from hives, pruritus,
wheezing, vomiting, and diarrhea to anaphylaxis: a rapid, systemic
reaction that can lead to hypotension, respiratory failure, and death.
4-6
Allergic sensitization—the series of events leading to the formation of
allergen-specific IgE—is a clinically silent process; indeed, the majority
of patients allergic to peanut, tree nuts, and egg react upon their first
known exposure.
7,8
Therefore, interventions to prevent or dismantle
food allergy require understanding of the events (mechanisms) leading
to the generation of IgE against foods.
Over the past 15 years, several clinical studies have proposed
that the skin is a site for Th2 sensitization to foods.
9-12
Collectively,
these studies postulate that defects in the integrity of the epidermis,
whether due to gene mutations (eg, filaggrin) and/or damage (eg,
eczema), are associated with a higher risk of developing food allergy.
A number of animal models have been established to investigate the
immunological basis of food allergy induced via the epicutaneous
route.
13-19
These models have been shown to produce allergen-speci-
fic IgE against common food allergens such as peanut or ovalbumin
(OVA)—the major allergen of the chicken egg white.
20,21
However,
the series of events that precede the generation of IgE-secreting
cells plasma cells (PCs) remain unexplored. The primary objective of
this study was to decipher the evolution of humoral immunity and
immunological memory in a model of epicutaneous sensitization to
foods.
We found that IgG1
+
B-cell immunity prevailed in the initial
stages of epicutaneous sensitization to food allergens in the germinal
center (GC) as well as in the effector (ie, PC and clinical reactivity)
and memory compartments. The memory compartment had predomi-
nantly an immature and pro-GC phenotype regardless of antigen
specificity (OVA vs peanut). Accordingly, clinical reactivity at this
stage was driven by the alternative pathway, that is, IgG, macro-
phages, and platelet-activating factor (PAF).
22,23
Following subclini-
cal, nonsensitizing allergen exposures, allergen-specific GCs formed,
IgE
+
GC B cells and PCs were generated, and allergen-specific IgE
was detected in circulation.
GRAPHICAL ABSTRACT
IgG1
+
B-cell immunity against food allergens in epicutaneous sensitization precedes the generation of IgE responses. Immature, pro-germinal
centre, allergen-specific IgG1
+
memory B cells are formed at the incipient stages of skin sensitization to foods. The assessment of allergen-spe-
cific cellular and humoral IgG1
+
immunity may help to identify individuals at risk of developing IgE-mediated food allergy.
166
|
JIM

ENEZ-SAIZ ET AL.

These data demonstrate that IgG1
+
B-cell immunity against food
allergens in epicutaneous sensitization precedes the generation of
IgE responses. Therefore, the assessment of cellular and humoral
allergen-specific IgG immunity may help to identify individuals at risk
of developing IgE-mediated food allergy and hence provide a win-
dow for immunological interventions to prevent disease progression.
2
|
METHODS
2.1
|
MICE
Six- to eight-week-old females and vendor- and strain-matched
controls were used in all the experiments. IgE-deficient mice
(Igh-7
tm1Led
)
24
and PAF receptor KO (PAF-R
/
)
25
mice were kindly
provided by Dr. H. Oettgen (Harvard Medical School, Boston, MA)
and Dr. E. I. Tuomanen (St. Jude Children’s Research Hospital, Mem-
phis, TN), respectively, and bred in house. C57BL/6 and Balb/c mice
were purchased from Charles River, and mast cell (MC)-deficient
mice (B6.Cg-Kit
W-sh
/HNihrJaeBsmGlli) and CCR2
/
mice (B6.129S4-
Ccr2
tm1lfc
/J) were purchased from The Jackson Laboratory. IL-33 KO
mice were generously provided by MedImmune (Gaithersburg, MD)
from Taconic (Derwood, MD). All procedures were approved by the
McMaster University Research Ethics Board.
2.2
|
Food allergy model
Epicutaneous sensitization was conducted as previously published.
26
Serum was collected by retro-orbital bleeding and analyzed for pea-
nut and/or OVA-specific Igs via sandwich ELISA.
27,28
Mice were sys-
temically challenged via intraperitoneal administration crude peanut
extract (CPE) (5 mg). Core temperature was assessed telemetrically
(see below), or rectally with a digital thermometer (23226-656,
VWR), and hematocrit by centrifuging (HemataSTAT II, Fisher Scien-
tific) anticoagulated blood 40 minutes after challenge.
2.3
|
Telemetric blood pressure and body
temperature measurements
Telemetric transmitters were used to measure mean arterial blood
pressure (PA-C10, Data Science International) and core body tem-
perature (TA-F10, Data Science International) in conscious unre-
strained mice. The devices were implanted following the
manufacturer’s protocol as described.
29
Telemetric data were col-
lected and analyzed with Dataquest A.R.T software (version 4.0;
Data Science International).
2.4
|
Peritoneal lavage
Peritoneal lavage was collected and processed as reported.
30
Cytos-
pins of the cells were stained 24 hours later with the Hema 3 stain
set (Fischer scientific). A minimum of 500 cells were counted to
quantify percentages of MCs, mononuclear cells, neutrophils, and
eosinophils.
2.5
|
Histology
Skin segments were collected and fixed in 10% formalin for
24 hours and then washed with 70% ethanol and paraffin embed-
ded. Sections were stained with hematoxylin and eosin (HE) for
microscopic analysis (Nikon X-Cite 120 LED) using NIS-ELEMENTS
software.
2.6
|
Inguinal lymph nodes processing
Inguinal lymph nodes (iLNs) were triturated between frosted slides in
Hank’s balanced salt solution, washed, and filtered (40 lm).
2.7
|
Flow cytometry
Antibodies were obtained from eBioscience, BD, or BioLegend.
OVA-FITC was obtained from Molecular Probes (023020, Thermo
Fisher), and biotinylated Ara h 1 and Ara h 2 from Indoor Biotech-
nologies Inc. In all assays, cells were incubated with anti-FccRII/III
before incubation with fluorochrome-conjugated antibodies. Dead
cells were excluded by propidium iodide uptake (Sigma) or fixable
viability dye eFluor780 (eBioscience) and gated on singlets. On aver-
age, a minimum of 10
6
and 300 000 alive and singlet cells were ana-
lyzed for experiments evaluating GCs, PCs, and memory B cells, and
dendritic cells (DCs), respectively. Experiments that involved detec-
tion of IgE
+
B cells or PCs included unconjugated anti-IgE (clone
RME-1) together with anti-FccRII/III and exclusive detection of intra-
cellular IgE via IgE-PE (clone RME-1) upon fixation/permeabilization
of cells (BD Biosciences).
31
Fluorescence minus one (FMO) and iso-
type controls were used for gating. Data were acquired on a LSR II
or Fortessa (BD) and analyzed using FlowJo (Treestar).
2.8
|
Antibody and drug administration
Blockade of histamine receptors and IgG-mediated anaphylaxis was
conducted as reported.
32,33
Depletion of basophils, phagocytes, and
neutrophils was conducted as previously shown.
33,34
2.9
|
Spleen cell cultures and assessment of
cytokine production
Spleens were processed and cultured as reported.
31
After 4-5 days
of culture, supernatants were stored at 80°C until further analysis.
Cytokines in cell-free supernatants were quantified with the
MAGPIX
â
system using MILLIPLEX
â
MAP magnetic bead-based
multi-analyte panels (Millipore).
2.10
|
Statistics
Data were analyzed and graphed with GraphPad Prism 6 software
(GraphPad Software). Continuous data are expressed as
means SEMs and were analyzed using 1-way ANOVA with Bon-
ferroni post hoc tests and unpaired Student’sttest. Ordinal data are
JIM

ENEZ-SAIZ ET AL.
|
167

shown as individual mice with medians and were analyzed with
Mann-Whitney Utests. Differences were considered statistically sig-
nificant at a Pvalue of <.05 or as indicated. Half-life was calculated
using the following equation: (Elapsed time 9log
2
)/(log[beginning
amount/ending amount]).
31
3
|
RESULTS
3.1
|
Epithelial damage causes DC activation and
migration to lymph nodes
We employed a murine model to investigate the humoral arm of
food sensitization via the epicutaneous route.
26
This 10-day model
involves application of allergen onto previously shaved and tape-
stripped skin on the lower backs of the mice (Figure 1A). This is
shorter than other models in which the length of allergen exposure
ranges from 14 to 49 days.
13-19
While tape-stripping did not cause
macroscopic damage (Figure S1A), histologic analysis revealed the
signs of microscopic damage and inflammation of the skin epithelium
(Figure 1B-E). Shortly after tape-stripping (30 minutes), the stratum
corneum (the outer layer of the epidermis) was absent, the extracel-
lular space stained intensely with eosin—which is suggestive of the
release of cellular contents and damage—and the normal morphol-
ogy of the cells in the epidermis (ie, shape and nuclear staining) was
altered (Figure 1C); 6 hours following tape-stripping, the epidermis
remained altered, and immune cell recruitment into the dermis was
observed (Figure 1D); 24 hours after tape-stripping, the inflamma-
tion of the dermis was no longer apparent, while the epidermis
showed hyperplasia and partial regeneration of the stratum corneum
(Figure 1E).
We next assessed the impact of this local inflammatory response
on the initiation of adaptive immunity by evaluating DC migration
and activation in the draining inguinal lymph nodes (iLNs). After
3 days of tape-stripping and allergen exposure, the absolute number
of DCs was significantly higher in mice undergoing epicutaneous
sensitization compared to na
€
ıve mice (Figure 1F-G); DCs in mice
undergoing sensitization were activated, borne out by upregulation
of the costimulatory markers CD80, CD86, and CD40
35,36
(Fig-
ure 1H-I). These data suggest that microscopic skin damage and local
inflammation lead to the migration of activated DCs to the iLNs,
thus indicating an ongoing immune response.
3.2
|
Allergen-specific GC activity and humoral
IgG1 immunity in epicutaneous sensitization
We then assessed whether the immune response initiated during
epicutaneous sensitization entailed GC activity, as this process is
central to the evolution of humoral responses and memory.
37
To
accomplish this, we delineated the kinetics of allergen-specific GC B
cells in the spleen and iLNs using a validated flow cytometric
method
31
(Figure 2A,B). Allergen-specific GC B cells were not
detected in the spleen at any time point. Conversely, the number of
(A)
(B)
(F)
(G) (I)
(H)
(C)
(D)
(E)
FIGURE 1 Epithelial damage causes an increase in activated DCs to iLNs. Separate groups of C57BL/6 and BALB/c mice were
epicutaneously sensitized to peanut (PN) (A). Biopsies of the skin were taken (B-E) from naive mice (B), or mice after 30 min (C), 6 h (D), and
24 h (E) of being tape-stripped (TS), and HE-stained. Following 3 d of sensitization, the draining lymph nodes (iLN) were collected to assess
DC migration (F,G) and activation (H,I) by flow cytometry. Representative pictures from 3 mice were taken; scale bar is 100 lm; arrows
indicate significant features (B-E). Representative data from 2-3 independent experiments (F-I) represented as mean SEM, n =3-6. *P<.05
vs na
€
ıve [Colour figure can be viewed at wileyonlinelibrary.com]
168
|
JIM

ENEZ-SAIZ ET AL.

allergen-specific GC B cells increased in the iLNs after 7 days. They
peaked after 10 days and returned to baseline 1 week later (day 17).
The calculated half-life of allergen-specific GC reactions in this sys-
tem was 2.48 days (95% CI, 0.98-4.63 days). The majority (≥95%) of
class-switched allergen-specific GC B cells generated at 11 days
expressed cell-surface IgG1 (Figure 2B), while IgE
+
GC B cells were
undetectable (data not shown). We also assessed serum Igs at 2 and
6 weeks after sensitization by ELISA (Figure 2C). Mirroring the GC
responses, allergen-specific IgE levels were not significantly
increased compared to untreated mice and allergen-specific IgG1
levels were elevated for at least 6 weeks (Figure 2C). Altogether,
these data show that epicutaneous exposure to food allergens
induces allergen-specific GC activity, which is dominated by IgG1-
expressing B cells, and a serum allergen-specific IgG1 humoral
response.
3.3
|
Epicutaneously sensitized mice undergo
anaphylaxis by the alternative pathway
To evaluate the functionality of the effector B-cell response gener-
ated on epicutaneous sensitization, we assessed clinical reactivity on
allergen challenge 2 weeks after the end sensitization. Compared to
na
€
ıve mice, epicutaneously sensitized mice underwent anaphylaxis,
as evidenced by the development of hypothermia (Figure 3A),
hypotension (Figure 3B), hemoconcentration (Figure 3C), and clinical
signs (Figure 3D). In addition, inflammation (Figure S1C) and eosino-
philia (Figure S1D) were detected in the peritoneal cavity (the site of
allergen challenge) of epicutaneously sensitized mice 72 hours
postchallenge indicating a late phase, T cell–mediated allergic
response.
As the Ig response was mainly comprised of IgG1, we tested the
importance of the classical and alternative pathways of anaphy-
laxis
32,33
in epicutaneously sensitized mice. Therefore, we assessed
clinical reactivity in IgE-deficient mice and mice treated with anti-
FccRII/III prior to challenge (Figure 3E). IgE-deficient mice under-
went anaphylaxis comparable to wild-type animals. However, block-
ade of IgG signaling significantly reduced the allergic reaction, which
was fully abrogated in IgE-deficient mice treated with anti-FccRII/III.
Next, we evaluated the contribution of immune cells potentially rele-
vant to clinical reactivity. The depletion of MCs, basophils, or neu-
trophils did not significantly ameliorate anaphylaxis (Figure 3F). In
contrast, depletion of phagocytes using clodronate-containing lipo-
somes completely abrogated clinical reactivity (Figure 3F). However,
CCR2
/
mice, which are monocytopenic due to defective bone
marrow egress of these cells,
38
developed anaphylaxis similar to
wild-type animals (Figure S2). Because CCR2
/
mice have an intact
FIGURE 2 Epicutaneous sensitization entails IgG1
+
GC B-cell activity and humoral IgG1. Gating strategy to identify OVA-specific, class-
switched, GC B cells by flow cytometry (A). Separate groups of C57BL/6 wild-type (WT) mice were epicutaneously sensitized to peanut (PN)
or OVA, and the kinetics of allergen-specific GC activity were assessed in the draining iLNs and the spleen (B). Serum levels of PN-specific IgE
and IgG1 were measured at different times postsensitization by ELISA (C). Representative (A,C) or pooled (B) data from 3 independent
experiments represented as mean SEM, n =3-6. *P<.01 vs na
€
ıve (ie, day 0) [Colour figure can be viewed at wileyonlinelibrary.com]
JIM

ENEZ-SAIZ ET AL.
|
169

macrophage compartment,
39
this indicates that macrophages were
necessary to cause anaphylaxis.
We then proceeded to investigate the contribution of a number
of anaphylactic mediators potentially relevant to epicutaneous sensiti-
zation.
16,32,40
Clinical reactivity was significantly reduced in PAF-R
/
mice (Figure 3G) despite having the same serum levels of peanut-
specific IgG1 than wild-type mice (data not shown). The co-
administration of H1 and H2 receptor antagonists had a modest,
nonsignificant effect on clinical reactivity. However, the combined
blockade of H1 and H2 receptors in PAF-R
/
mice fully prevented
clinical reactivity (Figure 3G), which is consistent with our previous
work in a model of oral sensitization.
32
Galand et al
16
recently showed
that IL-33 promotes IgE- and MC-mediated anaphylaxis. Expectedly,
we found that IL-33 KO mice underwent anaphylaxis similar to wild-
type mice on allergen challenge (Figure 3H). Collectively, these data
show that the IgG-macrophage-PAF axis is central in mediating clinical
reactivity at the early stages of epicutaneous sensitization to foods.
3.4
|
Phenotypic characterization of allergen-
specific IgG1
+
memory B cells
To better understand the B-cell compartment driving IgG1-dominant
epicutaneous sensitization to foods, we analyzed allergen-specific
memory B cells generated against major peanut (Ara h 1 and 2)
41
and egg (OVA)
42
allergens. To identify features in the memory
compartment that are inherent to the allergen, mice were sensitized
to peanut extract or OVA, either alone or concomitantly. Mice were
rested after sensitization for 3-4 weeks—well beyond the duration
of allergen-specific GC activity (2.48 days (95% CI, 0.98-4.63 days);
Figure 2B)—before the memory compartment in the iLNs was ana-
lyzed (Figure 4). Following exclusion of lymphocytes expressing IgD,
IgM, CD138, and CD4, a clear population of IgG1
+
memory B cells
was found while class-switched B cells expressing IgE were unde-
tectable (Figure 4A). Then, we assessed allergen specificity via bind-
ing to either Ara h 1 and 2, or OVA. Allergen-specific IgG1
+
memory
B cells were undetected in na
€
ıve mice, and peanut-specific popula-
tions were not observed in mice sensitized to OVA and vice versa
(Figure 4A). Allergen-specific IgG1
+
memory B cells were rare
(~0.002% of total cells). The frequency of OVA-specific from IgG1
+
memory B cells was significantly higher in OVA-sensitized mice
(16.2%) than in cosensitized mice (5.4%), while the frequency of pea-
nut-specific from IgG1
+
memory B cells was comparable between
peanut (6.4%) and cosensitized mice (5.8%). Thus, the addition of
OVA in the cosensitized mice did not significantly alter the relative
availability of Ara h 1 and 2, which were used to measure peanut-
specific IgG1
+
B cells, whereas the presence of peanut did signifi-
cantly reduce the relative availability of OVA compared to just the
OVA-sensitized mice.
We examined the identity and functionality of allergen-specific
IgG1
+
memory B cells from epicutaneously sensitized mice by
FIGURE 3 Epicutaneously sensitized mice undergo anaphylaxis mediated by IgG-macrophage-PAF and independent of IL-33. Separate
groups of C57BL/6 and BALB/c wild-type (WT) mice, and mice deficient in IgE (E, BALB/c), MCs (F, C57BL/6), PAF-R (G, C57BL/6), and IL-33
(BALB/c) were epicutaneously sensitized to peanut. Clinical reactivity was assessed by measuring hypothermia (A), blood pressure (B),
hemoconcentration (C,E-H), and clinical signs (D). Some groups of sensitized mice were treated prior to allergen challenge with drugs or
antibodies to inhibit IgG signaling (E, anti-FccRII/III), to deplete basophils (F, anti-Ba103), neutrophils (F, anti-1A8), and phagocytes (F,
clodronate-containing liposomes) or to block histamine receptors 1 (mepyramine) and 2 (cimetidine) (G). Pooled data from 2 to 3 experiments
represented as mean SEM (A-C,E-H), or individual mice with medians (D), n =6-15. *P<.05 vs na
€
ıve or as indicated [Colour figure can be
viewed at wileyonlinelibrary.com]
170
|
JIM

ENEZ-SAIZ ET AL.

assessing PDL-2, CD80, CD73, and CD35 expression
43-46
(Figure 4B).
Regardless of allergen specificity (peanut vs OVA) and diversity (sin-
gle allergen vs complex mixture), the phenotype of the allergen-
specific IgG1 memory compartment was consistently the same, domi-
nated by cells negative for PDL-2 and CD80, and positive for CD73
and CD35. However, additional exposures to the allergen, under
FIGURE 4 Phenotypic characterization of allergen-specific IgG1
+
memory B cells. Separate groups of C57BL/6 mice were epicutaneously
sensitized to peanut (PN), OVA, or both, and rested for 3-4 wk. ILNs were collected and stained for flow cytometry analysis of the frequency
(A) and phenotype (B) of allergen-specific IgG1
+
memory B cells (every histogram represents a single mouse). Representative data from 3 to 4
independent experiments; n =4-6 per experiment [Colour figure can be viewed at wileyonlinelibrary.com]
FIGURE 5 Phenotypic changes of allergen-specific IgG1
+
memory B cells following allergen exposures. Separate groups of C57BL/6 mice
were epicutaneously sensitized to peanut (PN) and OVA, rested for 3-4 wk, and subcutaneously (s.c.) exposed to subclinical doses of PN. Then,
4-6 wk later, the iLNs were collected (A) and stained for phenotypic analysis by flow cytometry (B, every histogram represents a single mouse).
Representative data from 2 independent experiments; n =4-5 per experiment [Colour figure can be viewed at wileyonlinelibrary.com]
JIM

ENEZ-SAIZ ET AL.
|
171

subclinical conditions (Figure S3), modified the phenotype of this
population, which partially increased PDL-2 and CD80 expression
and decreased CD35, while remained positive for CD73 (Figure 5).
Collectively, these data show that the incipient stages of epicuta-
neous sensitization to foods generate a memory population of IgG1
+
B cells with an immature and pro-GC phenotype (PDL-
2

CD80

CD35
+
CD73
+
).
3.5
|
Allergen re-exposures in epicutaneously
sensitized mice generate IgE
+
GC B cells and humoral
IgE
Lastly, we sought to evaluate the functional impact of allergen re-
exposures after skin sensitization. To do this, mice were epicuta-
neously cosensitized to peanut and OVA, and then rested for 3-
4 weeks prior to being re-exposed to the allergen 3 times at subclin-
ical doses (Figure S3). The B-cell compartment was analyzed 2 days
after the last exposure (Figure 6A). We only detected significant
peanut-specific GC activity in the iLNs of epicutaneously sensitized
mice that had been subclinically re-exposed to the allergen (Fig-
ure S4A). Interestingly, the subclinical exposures to peanut—in mice
cosensitized to peanut and OVA—induced bystander OVA-specific
GC activity in a nonspecific manner (Figure S4B). We employed a
well-established flow cytometry method to detect bona fide, IgE
+
GC
B cells and PCs.
31,47
A clear population of IgE
+
GC B cells (Fig-
ure 6B,C) and PCs (Figure 6D,E) was exclusively detected in the iLNs
in sensitized mice subjected to allergen re-exposure. Accordingly,
peanut-specific IgE levels in circulation were detected in these mice
compared to sensitized but not re-exposed mice. Furthermore,
splenocyte cultures with peanut extract from allergic mice exposed
to the allergen, but not those from na
€
ıve mice, exhibited a Th1:Th2
ratio marked by IL-4 production (Figure 6G). In a similar fashion, we
found that oral exposures to allergen after epicutaneous sensitization
induced allergen-specific IgE responses in ~40% of the mice (Fig-
ure S5A). Furthermore, these mice underwent anaphylaxis despite
macrophage depletion (Figure S5B), which suggests activation of the
classical pathway. Collectively, these data show that subclinical
exposures to allergen in epicutaneously sensitized mice induce aller-
gen-specific GC activity, generation of IgE
+
GC B cells and PCs, and
allergen-specific IgE, thus likely enabling classical IgE-mediated food
allergy.
4
|
DISCUSSION
Defining the evolution of IgE-mediated food allergy in humans is
exceedingly difficult because the immunological events underlying
this process are clinically silent. The proposition that the skin
epithelium can be a relevant site for Th2 sensitization to foods in
humans
9-12
has prompted the development of experimental models
of epicutaneous sensitization to investigate the underlying mecha-
nisms leading to the generation of IgE responses.
13-19
While the
role of a number of cytokines
13,19,48
and innate cells
18,49
in epicu-
taneous sensitization has been reported, the events pertaining to
the generation of humoral immunity and memory remain unex-
plored. Here, we investigated the nature of the allergen-specific B-
FIGURE 6 Allergen re-exposures generate IgE
+
GC B cells and humoral IgE in epicutaneously sensitized mice. Separate groups of C57BL/6
mice were epicutaneously sensitized (WT) to peanut (PN) and OVA, and rested for 3-4 wk. Some groups were subcutaneously (s.c.) exposed to
100 lg of PN extract, 3 times (A). Two days after the last exposure, we assessed IgE
+
GC B cells (B,C) and PCs (D,E) by flow cytometry, serum
levels of allergen-specific IgE by ELISA (F), and cytokine secretion in supernatants from splenocyte cultures by MAGPIX (G). Representative (B,
D,G) or pooled (C,E,F) data from 2-3 independent experiments represented as mean SEM, n =6-15. *P<.05 vs na
€
ıve [Colour figure can be
viewed at wileyonlinelibrary.com]
172
|
JIM

ENEZ-SAIZ ET AL.

cell response and its evolution from the initial stages of food sensi-
tization.
We found that, as previously reported, tape-stripping of the skin
induced damage and inflammation.
50,51
The generation of damage
(via mechanical, chemical, or biological stress) and subsequent
release of alarmins, such as uric acid crystals, IL-25 or TSLP,
17,26,48,52
that precondition DCs to initiate Th2 immunity is a common feature
of allergic sensitization regardless of the route of allergen expo-
sure.
53
Our data showed that the application of food allergens onto
the tape-stripped skin induced allergen-specific GC activity. Its dura-
tion was shorter than that reported in models of Th2 immunity
against helminths (eg, Nippostrongylus brasiliensis),
44,54,55
likely
because these models involve extended antigen availability.
56
We
found that practically the entire GC B-cell compartment was
expressing IgG1 throughout all stages of sensitization. This resulted
in an allergen-specific IgG1 humoral response, likely maintained by
long-lived PCs as the titers were sustained for, at least, 6 weeks fol-
lowing sensitization.
Mice underwent anaphylaxis on allergen challenge, and clinical reac-
tivity was primarily effected by the alternative pathway—IgG, macro-
phages, and PAF. However, subsequent subclinical, nonsensitizing
allergen exposures facilitated the evolution of the humoral response
from an IgG1 (at the incipient stages of sensitization) to an IgE-domi-
nated response (established allergy) thus likely enabling the classical
pathway to mediate clinical reactivity. While the ability of the alternative
pathway to mediate anaphylaxis in mice is well established,
6,57
its contri-
bution to human anaphylaxis is debatable and appears to be restricted to
patients treated parentally with large quantities of biologics, drugs, or
transfusions (eg, dextran, aprotinin, Willebrand factor, and immunoglobu-
lin treatments).
58
Here, we show that IgG1 responses developed at the
incipient stages of food sensitization, as it has been reported in birch-
sensitized children,
59,60
but that they may remain clinically undetectable.
Allergen-specific IgG1
+
memory B cells established at the incipi-
ent stages of food sensitization were predominantly PDL-
2

CD80

CD73
+
CD35
+
. This intimates that its generation (ie, GC
history, and maturity), regardless of the allergen used for sensitiza-
tion (peanut and/or OVA), was comparable. Recently, Weisel et al
46
described a population of early memory B cells that were PDL-
2

CD80

CD73
+
in a model of systemic immunization to a hapten
(ie, NP) suggesting that the memory compartment originated during
epicutaneous sensitization is rather immature. Anderson et al
45
reported a population of CD80

CD35
+
memory B cells that were
unmutated in a model of NP-alum, which is consistent with the brief
GC reaction following epicutaneous sensitization. In experiments in
which mice were cosensitized to peanut and OVA, we observed that
the number of allergens at the time of sensitization—Ara h 1-17 for
peanut vs OVA—limited the magnitude of the immune response (ie,
clonal expansion, and GC activity) to each individual allergen, inti-
mating the existence of mechanisms that regulate clonal expansion
and diversity in the GC.
37,61
It is known that allergens differ in their
innate immunostimulatory properties
15,62
; however, the features of
the memory compartment that we identified were consistent regard-
less of the allergen specificity (ie, OVA, peanut, or cosensitization),
which may indicate that immunological features intrinsic to the skin
epithelium govern the nature of the B-cell response.
Following nonsensitizing, subclinical allergen exposures, memory
B cells generated allergen-specific GCs, in agreement with the func-
tionality attributed to PDL-2

CD80

memory B cells.
43
Interestingly,
secondary responses to peanut caused bystander OVA-specific GC
activity in mice that were sensitized to both allergens. This might
constitute a mechanism that perpetuates Th2 immunity in a non-
specific manner, as reported by Bernasconi et al,
63
in the context of
human vaccination. We employed a well-established flow cytometry
method to detect bona fide, IgE
+
GC B cells and PCs.
31,47
The forma-
tion of IgE
+
GC B and PCs, and production of IgE following allergen
exposure indicate that IgG1
+
B-cell immunity predates IgE responses
in food allergy; a process that we show that is dominated by IL-4,
and that is likely dependent on CD4 T-cell help and sequential class-
switching.
31,44,54,64
Importantly, a recent analysis of Ig heavy regions
in peripheral blood of allergic patients demonstrated that a majority
of IgE
+
cells arise from antigen-experienced, somatically hypermu-
tated, IgG1-expressing cells
65
suggesting that the IgG1 memory com-
partment matures likely due to allergen stimulation. This is
supported by our observation that the predominantly immature,
pro-GC, IgG1
+
memory B-cell compartment (PDL-2

CD80

CD73
+
CD35
+
) that exists in the preclinical stages of epicutaneous sensitiza-
tion partially shifted to a more mature PDL2
+
CD73
+
CD80
+
pheno-
type on further allergen exposures.
Our findings indicate that IgG1
+
B-cell immunity against food
allergens dominates the initial stages of epicutaneous sensitization.
This process results in the generation of an immature IgG1
+
memory
B-cell compartment that on subclinical, repeated allergen exposure
predates IgE responses. The assessment of allergen-specific IgG
+
memory B cells and humoral IgG may help to predict the risk for
developing clinical allergy, thus providing a window for immunologi-
cal interventions to prevent disease progression.
ACKNOWLEDGMENTS
We thank Dr. H. Oettgen for providing IgE-deficient mice. We rec-
ognize He Tian Chen for graphical design. We thank all the members
of the Jordana-Waserman Lab for technical help and scientific input.
RJS and JFEK are holders of a Mitacs Postdoctoral Fellowship and a
CIHR Graduate Student Scholarship, respectively. MJ holds a senior
Canada Research Chair. RK is an employee of MedImmune LLC.
CONFLICTS OF INTEREST
The authors declare that they have no conflicts of interest.
AUTHOR CONTRIBUTIONS
RJS designed and conducted experiments, analyzed data, and wrote the
manuscript. YE, JFEK, MEG, TDW, DR, PS, KB, and JW helped with
experiments. JK and KF performed and analyzed data from experiments
that involved telemetry measurements and IL-33 KO mice, respectively.
JIM

ENEZ-SAIZ ET AL.
|
173

EIT, RK, and DKC provided reagents and/or input. SW and MJ obtained
funding. MJ oversaw the project and edited the manuscript.
ORCID
R. Jim
enez-Saiz http://orcid.org/0000-0002-0606-3251
REFERENCES
1. Jones SM, Burks AW. Food allergy. N Engl J Med. 2017;377:1168-
1176.
2. Sicherer SH, Sampson HA. Food allergy: epidemiology, pathogenesis,
diagnosis, and treatment. J Allergy Clin Immunol. 2014;133:291-307.
3. Tordesillas L, Berin MC, Sampson HA. Immunology of food allergy.
Immunity. 2017;47:32-50.
4. Brown SG, Stone SF, Fatovich DM, et al. Anaphylaxis: clinical patterns,
mediator release, and severity. J Allergy Clin Immunol. 2013;132:1141-
1149.
5. Galli SJ, Tsai M. IgE and mast cells in allergic disease. Nat Med.
2012;18:693-704.
6. Reber LL, Hernandez JD, Galli SJ. The pathophysiology of anaphy-
laxis. J Allergy Clin Immunol. 2017;140:335-348.
7. Sicherer SH, Burks AW, Sampson HA. Clinical features of acute aller-
gic reactions to peanut and tree nuts in children. Pediatrics.
1998;102:e6.
8. Monti G, Muratore MC, Peltran A, et al. High incidence of adverse
reactions to egg challenge on first known exposure in young ato-
pic dermatitis children: predictive value of skin prick test and
radioallergosorbent test to egg proteins. Clin Exp Allergy.
2002;32:1515-1519.
9. Lack G, Fox D, Northstone K, Golding JAvon Longitudinal Study of
P, Children Study T. Factors associated with the development of
peanut allergy in childhood. N Engl J Med. 2003;348:977-985.
10. Ashley SE, Tan HT, Vuillermin P, et al. The skin barrier function gene
SPINK5 is associated with challenge-proven IgE-mediated food
allergy in infants. Allergy. 2017;72:1356-1364.
11. Brown SJ, Asai Y, Cordell HJ, et al. Loss-of-function variants in the
filaggrin gene are a significant risk factor for peanut allergy. J Allergy
Clin Immunol. 2011;127:661-667.
12. Brough HA, Liu AH, Sicherer S, et al. Atopic dermatitis increases the
effect of exposure to peanut antigen in dust on peanut sensitiza-
tion and likely peanut allergy. J Allergy Clin Immunol. 2015;135:164-
170.
13. Hsieh KY, Tsai CC, Wu CH, Lin RH. Epicutaneous exposure to
protein antigen and food allergy. Clin Exp Allergy. 2003;33:1067-
1075.
14. Bartnikas LM, Gurish MF, Burton OT, et al. Epicutaneous sensitiza-
tion results in IgE-dependent intestinal mast cell expansion and
food-induced anaphylaxis. J Allergy Clin Immunol. 2013;131:451-460.
15. Tordesillas L, Goswami R, Benede S, et al. Skin exposure promotes a
Th2-dependent sensitization to peanut allergens. J Clin Invest.
2014;124:4965-4975.
16. Galand C, Leyva-Castillo JM, Yoon J, et al. IL-33 promotes food ana-
phylaxis in epicutaneously sensitized mice by targeting mast cells. J
Allergy Clin Immunol. 2016;138:1356-1366.
17. Noti M, Kim BS, Siracusa MC, et al. Exposure to food allergens
through inflamed skin promotes intestinal food allergy through the
thymic stromal lymphopoietin-basophil axis. J Allergy Clin Immunol.
2014;133:1390-1399.
18. Hussain M, Borcard L, Walsh KP, et al. Basophil-derived IL-4 pro-
motes epicutaneous antigen sensitization concomitant with the devel-
opment of food allergy. J Allergy Clin Immunol. 2018;141:223-234.
19. Jin H, Oyoshi MK, Le Y, et al. IL-21R is essential for epicutaneous
sensitization and allergic skin inflammation in humans and mice. J
Clin Invest. 2009;119:47-60.
20. Jimenez-Saiz R, Pineda-Vadillo C, Lopez-Fandino R, Molina E.
Human IgE binding and in vitro digestion of S-OVA. Food Chem.
2012;135:1842-1847.
21. Jim
enez-Saiz R, Ruiz-Henestrosa VMP, L
opez-Fandi~
no R, Molina E.
In vitro digestibility and allergenicity of emulsified hen egg. Food Res
Int. 2012;48:404-409.
22. Khodoun MV, Strait R, Armstrong L, Yanase N, Finkelman FD. Iden-
tification of markers that distinguish IgE- from IgG-mediated anaphy-
laxis. Proc Natl Acad Sci USA. 2011;108:12413-12418.
23. Escribese MM, Rosace D, Chivato T, Fernandez TD, Corbi AL, Barber
D. Alternative anaphylactic routes: the potential role of macro-
phages. Front Immunol. 2017;8:515.
24. Oettgen HC, Martin TR, Wynshaw-Boris A, Deng C, Drazen JM, Leder
P. Active anaphylaxis in IgE-deficient mice. Nature. 1994;370:367-370.
25. Radin JN, Orihuela CJ, Murti G, Guglielmo C, Murray PJ, Tuomanen
EI. beta-Arrestin 1 participates in platelet-activating factor receptor-
mediated endocytosis of Streptococcus pneumoniae. Infect Immun.
2005;73:7827-7835.
26. Kong J, Chalcraft K, Mandur TS, et al. Comprehensive metabolomics
identifies the alarmin uric acid as a critical signal for the induction of
peanut allergy. Allergy. 2015;70:495-505.
27. Chu DK, Mohammed-Ali Z, Jimenez-Saiz R, et al. T helper cell IL-4
drives intestinal Th2 priming to oral peanut antigen, under the con-
trol of OX40L and independent of innate-like lymphocytes. Mucosal
Immunol. 2014;7:1395-1404.
28. Swirski FK, Sajic D, Robbins CS, Gajewska BU, Jordana M, Stampfli
MR. Chronic exposure to innocuous antigen in sensitized mice leads
to suppressed airway eosinophilia that is reversed by granulocyte
macrophage colony-stimulating factor. J Immunol. 2002;169:3499-
3506.
29. Korhonen H, Fisslthaler B, Moers A, et al. Anaphylactic shock
depends on endothelial Gq/G11. J Exp Med. 2009;206:411-420.
30. Sun J, Arias K, Alvarez D, et al. Impact of CD40 ligand, B cells, and
mast cells in peanut-induced anaphylactic responses. J Immunol.
2007;179:6696-6703.
31. Jimenez-Saiz R, Chu DK, Mandur TS, et al. Lifelong memory
responses perpetuate humoral TH2 immunity and anaphylaxis in
food allergy. J Allergy Clin Immunol. 2017;140:1604-1615.
32. Arias K, Baig M, Colangelo M, et al. Concurrent blockade of platelet-
activating factor and histamine prevents life-threatening peanut-induced
anaphylactic reactions. J Allergy Clin Immunol. 2009;124:307-314.
33. Arias K, Chu DK, Flader K, et al. Distinct immune effector pathways
contribute to the full expression of peanut-induced anaphylactic
reactions in mice. J Allergy Clin Immunol. 2011;127:1552-1561.
34. Daley JM, Thomay AA, Connolly MD, Reichner JS, Albina JE. Use of
Ly6G-specific monoclonal antibody to deplete neutrophils in mice. J
Leukoc Biol. 2008;83:64-70.
35. MacDonald AS, Straw AD, Dalton NM, Pearce EJ. Cutting edge: Th2
response induction by dendritic cells: a role for CD40. J Immunol.
2002;168:537-540.
36. Hussaarts L, Yazdanbakhsh M, Guigas B. Priming dendritic cells for
th2 polarization: lessons learned from helminths and implications for
metabolic disorders. Front Immunol. 2014;5:499.
37. Dufaud CR, McHeyzer-Williams LJ, McHeyzer-Williams MG. Decon-
structing the germinal center, one cell at a time. Curr Opin Immunol.
2017;45:112-118.
38. Tsou CL, Peters W, Si Y, et al. Critical roles for CCR2 and MCP-3 in
monocyte mobilization from bone marrow and recruitment to inflam-
matory sites. J Clin Invest. 2007;117:902-909.
39. Hashimoto D, Chow A, Noizat C, et al. Tissue-resident macrophages
self-maintain locally throughout adult life with minimal contribution
from circulating monocytes. Immunity. 2013;38:792-804.
174
|
JIM

ENEZ-SAIZ ET AL.

40. Vadas P, Perelman B, Liss G. Platelet-activating factor, histamine,
and tryptase levels in human anaphylaxis. J Allergy Clin Immunol.
2013;131:144-149.
41. Jimenez-Saiz R, Benede S, Molina E, Lopez-Exposito I. Effect of pro-
cessing technologies on the allergenicity of food products. Crit Rev
Food Sci Nutr. 2015;55:1902-1917.
42. Mine Y, Yang M. Recent advances in the understanding of egg aller-
gens: basic, industrial, and clinical perspectives. J Agric Food Chem.
2008;56:4874-4900.
43. Zuccarino-Catania GV, Sadanand S, Weisel FJ, et al. CD80 and PD-
L2 define functionally distinct memory B cell subsets that are inde-
pendent of antibody isotype. Nat Immunol. 2014;15:631-637.
44. He JS, Subramaniam S, Narang V, et al. IgG1 memory B cells keep
the memory of IgE responses. Nat Commun. 2017;8:641.
45. Anderson SM, Tomayko MM, Ahuja A, Haberman AM, Shlomchik
MJ. New markers for murine memory B cells that define mutated
and unmutated subsets. J Exp Med. 2007;204:2103-2114.
46. Weisel Florian J, Zuccarino-Catania Griselda V, Chikina M, Shlomchik
Mark J. A temporal switch in the germinal center determines
differential output of memory B and plasma cells. Immunity.
2016;44:116-130.
47. Yang Z, Sullivan BM, Allen CD. Fluorescent in vivo detection reveals
that IgE(+) B cells are restrained by an intrinsic cell fate predisposi-
tion. Immunity. 2012;36:857-872.
48. Muto T, Fukuoka A, Kabashima K, et al. The role of basophils and
proallergic cytokines, TSLP and IL-33, in cutaneously sensitized food
allergy. Int Immunol. 2014;26:539-549.
49. Deckers J, Sichien D, Plantinga M, et al. Epicutaneous sensitization to
house dust mite allergen requires interferon regulatory factor 4-depen-
dent dermal dendritic cells. J Allergy Clin Immunol. 2017;140:1364-1377.
50. Surber C, Schwarb FP, Smith EW. Tape-stripping technique. J Toxicol
Cutaneous Ocul Toxicol. 2001;20:461-474.
51. Weigand DA, Gaylor JR. Removal of stratum corneum in vivo: an
improvement on the cellophane tape stripping technique. J Invest
Dermatol. 1973;60:84-87.
52. Lee JB, Chen CY, Liu B, et al. IL-25 and CD4(+) TH2 cells enhance
type 2 innate lymphoid cell-derived IL-13 production, which pro-
motes IgE-mediated experimental food allergy. J Allergy Clin Immunol.
2016;137:1216-1225.
53. Jim
enez-Saiz R, Chu DK, Waserman S, Jordana M. Initiation, persis-
tence and exacerbation of food allergy. In: Schmidt-Weber CB, eds.
Allergy Prevention and Exacerbation: The Paradox of Microbial Impact
on the Immune System. Cham, Switzerland: Springer International
Publishing; 2017:121-144.
54. Turqueti-Neves A, Otte M, Schwartz C, et al. The extracellular domains
of IgG1 and T cell-derived IL-4/IL-13 are critical for the polyclonal mem-
ory IgE response in vivo. PLoS Biol. 2015;13:e1002290.
55. Talay O, Yan DH, Brightbill HD, et al. IgE(+) memory B cells and
plasma cells generated through a germinal-center pathway. Nat
Immunol. 2012;13:396-404.
56. Cirelli KM, Crotty S. Germinal center enhancement by extended
antigen availability. Curr Opin Immunol. 2017;47:64-69.
57. Finkelman FD. Anaphylaxis: lessons from mouse models. J Allergy
Clin Immunol. 2007;120:506-515.
58. Finkelman FD, Khodoun MV, Strait R. Human IgE-independent sys-
temic anaphylaxis. J Allergy Clin Immunol. 2016;137:1674-1680.
59. Hofmaier S, Hatzler L, Rohrbach A, et al. “Default”versus “pre-ato-
pic”IgG responses to foodborne and airborne pathogenesis-related
group 10 protein molecules in birch-sensitized and nonatopic chil-
dren. J Allergy Clin Immunol. 2015;135:1367-1374.
60. Huang X, Tsilochristou O, Perna S, et al. Evolution of the IgE and
IgG repertoire to a comprehensive array of allergen molecules in the
first decade of life. Allergy. 2018;73:421-430.
61. Mesin L, Ersching J, Victora GD. Germinal center B cell dynamics.
Immunity. 2016;45:471-482.
62. Ruiter B, Shreffler WG. Innate immunostimulatory properties of
allergens and their relevance to food allergy. Semin Immunopathol.
2012;34:617-632.
63. Bernasconi NL, Traggiai E, Lanzavecchia A. Maintenance of serologi-
cal memory by polyclonal activation of human memory B cells.
Science. 2002;298:2199-2202.
64. Xiong H, Dolpady J, Wabl M, Curotto de Lafaille MA, Lafaille JJ.
Sequential class switching is required for the generation of high
affinity IgE antibodies. J Exp Med. 2012;209:353-364.
65. Looney TJ, Lee JY, Roskin KM, et al. Human B-cell isotype switching
origins of IgE. J Allergy Clin Immunol. 2016;137:579-586.
SUPPORTING INFORMATION
Additional supporting information may be found online in the
Supporting Information section at the end of the article.
How to cite this article: Jim
enez-Saiz R, Ellenbogen Y, Koenig
JFE, et al. IgG1
+
B-cell immunity predates IgE responses in
epicutaneous sensitization to foods. Allergy. 2019;74:165–
175. https://doi.org/10.1111/all.13481
JIM

ENEZ-SAIZ ET AL.
|
175