=§cur
nr‘ .bwwnsai
er; . . .a
“5?. Ant 2..

:2} ..

i, .35. J
‘JI ”- V
an 4.5.9.9.. m w

L
u‘

 

uni: . a ..
.f Wm»? mm... .

a? I rtnxflj V
\ a .m
\:Eu.:;..!,
I . . . 5‘
. . . ‘ . . v V fit!
. ‘ .. nIoJMJI...,

 

an.“ e at
5

{km

5.. 5.33%
. t... uu
$3

9”: «NM. cr 1*

_.,, .5. ffio... ..
3%....
«r,- .. . 3
é§w§.

 

. .(4ili
23";

 

L4l‘i'lll... r |

 

THEE}

3007

 

This is to certify that the
dissertation entitled

Development, Characterization and Validation of a mouse
model of tree-nut allergy using hazelnut as a model allergenic
food

presented by

Neil Patrick Birmingham

has been accepted towards fulfillment
of the requirements for the

Doctoral degree in The Department of Food

 

 

Science and Human Nutrition
\! l
\

Major Pilofeiéor’s Signature
£91) 2—] /7,w‘)’6
l I

Date

 

 

MSU is an Affirmative Action/Equal Opportunity Institution

 

LIBRARY
Michigan State
University

 

 

 

I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
l
I
.
I
.
l
.
.
.

 

 

PLACE IN RETURN BOX to remove this checkout from your record.
To AVOID FINES return on or before date due.
MAY BE RECALLED with earlier due date if requested.

 

DATE DUE DATE DUE DATE DUE

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

2/05 p:/C|RC/DaleDue.indd-p.1

 

 

DEVELOPMENT, CHARACTERIZATION AND VALIDATION OF
A MOUSE MODEL OF TREE-NUT ALLERGY USING HAZELNUT
AS A MODEL ALLERGENIC FOOD

By

Neil Patrick Birmingham

A DISSERTATION
Submitted to
Michigan State University
in partial fulfillment of the requirements
for'the degree of
DOCTOR OF PHILOSOPHY
Department of Food Science & Human Nutrition

2006

ABSTRACT
DEVELOPMENT, CHARACTERIZATION AND VALIDATION OF A MOUSE
MODEL OF TREE-NUT ALLERGY USING HAZELNUT AS A MODEL
ALLERGENIC FOOD
By
Neil Patrick Birmingham

Food allergy prevalence is increasing and the public has become increasingly aware of
the problem. The mechanisms of tree-nut allergy are not completely known and a mouse
model to study tree-nut allergy is unavailable. The overall hypotheses driving my
research were i) mice can develop tree-nut allergy that mimic certain phenotypes of
human tree-nut allergy and ii) a validated mouse model of tree-nut allergy is useful to
study impact of dietary modification on various markers of this disease. The aims of my
studies to test these hypotheses were 1) to develop an ELISA based method to measure
allergen specific IgE as an alternative to the passive cutaneous anaphylaxis assay (PCA);
2) to determine if hazelnut can directly elicit a specific IgE antibody response via
activating IL-4 in mice; 3) to characterize the systemic immune response following
transdermal hazelnut protein exposure in BALB/c mice; 4) to develop an adjuvant free
model of hazelnut protein induced systemic anaphylaxis; and 5) to study the effect of a
diet rich in eicosapentaenoic acid (EPA) and docosahexaenoic acid (DHA) on systemic
immune responses to hazelnut. During our studies we demonstrated that 1) Food specific
IgE levels can be measured by an ELISA based method that is comparable to PCA; 2)
Hazelnut itself can be an allergenic food, capable of directly eliciting hazelnut binding
specific IgE antibodies via activation of Type-2 cytokines in mice; 3) Mice can respond

in an allergic manner to transdermal hazelnut protein exposure without the use of

adj uvant; 4) Transdermal exposure of mice with hazelnut protein sensitizes them for
systemic anaphylaxis when challenged either i.p. or orally with hazelnut; and 5) EPA and
DHA supplementation together can enhance hazelnut specific IgE antibodies via altering
the INF -y / IL-4 ratio in favor of IL-4. Taken together, the mouse model described in this
thesis might be a useful tool for determination of mechanisms of tree-nut allergy. Future
studies that could be proposed using this model include using tools such as gene knock
out strains of mice and RNA interference to study mechanisms associated with tree-nut
allergy; to further therapeutic and prophylactic studies using pharmaceutical or dietary

interventions.

ACKNOWLEDGEMENTS

I gratefully acknowledge the support given to me both mentally as well as financially
from a great number of supporters, without whom, I would not have been able to achieve

these goals.

First and foremost, I thank my major advisor, Dr. Venu Gangur, for great guidance and
being a wonderful role model to follow. I would also like to thank the rest of my
guidance committee, Dr. James Pestka, Dr. Maurice Bennink and Dr. Jack Harkema for
wonderful advice both personally as well as scholarly. Besides my advisors, I would
like to thank all of my lab mates; Dr. Hanem Hassan (Post-Doc), Lalithia Navuluri,
Sitaram Parvataneni, Sridhar Sarnineni and Caleb Kelly (undergrad) for helping me with
everything and anything at a moments notice and other authors on papers including
Sandhya Payankaulam, Bill Stefura (University of Manitoba) and Professor Kent
Hayglass (Immunology Chair, University of Manitoba). I would also like to thank Dr.

Paul Satoh of Neogen.

I would like to express my extreme gratitude to all of the bodies that funded me through
this program; Department of Food Science and Human Nutrition, Kellogg Fellowship
Program, Dr. Venu Gangur (MSU Foundation, MAES), MSU Graduate School, College
of Agriculture and Natural Resources, College of Human Ecology, Dairy Plant
Fellowship Fund and the Rachel A. Schemmel Graduate Student Endowed Research

Scholarship.

iv

Finally, I would like to thank my family for all the love and support that guided me
through this journey, especially Janette for her love and encouragement, which drives me

to achieve the goals I set forth.

TABLE OF CONTENTS

Contents Page No.
List of tables ............................................................................. viii
List of figures ........................................................................... ix
Abbreviations used ................................................................... xiv
Chapter 1: Introduction ............................................................... 1
Chapter 1: References ................................................................
Chapter 2: Review of literature ...................................................... 9
Chapter 2.1: Animal models of food allergy ...................................... 9
Chapter 2.2: Rat model of Ovalbumin allergy .................................... 10
Chapter 2.3: The Guinea pig model of cow’s milk proteins ..................... 12
Chapter 2.4: Atopic dog model for multiple foods (Cow’s milk, beef,
ragweed, and wheat) .................................................................. 13
Chapter 2.5: Atopic dog model for multiple foods (Peanut, walnut, Brazil
nut) ..................................................................................... 15
Chapter 2.6: The Neonatal swine model of peanut allergy ..................... 17
Chapter 2.7: Mouse models of food allergy ....................................... 20
Chapter 2.8: Mouse model of Cow’s milk allergy ............................... 21
Chapter 2.9: Mouse model of peanut allergy ...................................... 23
Chapter 2.10: Mouse model of Ovalbumin induced food allergy .............. 25
Chapter 2.11: Anti-ulcer drugs promote oral sensitization to hazelnut
allergens ................................................................................ 27
Chapter 2.12: Gap in the knowledge ............................................... 28
Chapter 2.13: References ............................................................ 30

Chapter 3: An ELISA based method for food specific IgE antibody
measurement in mouse serum: An alternative to the passive cutaneous

anaphylaxis assay ...................................................................... 32
Chapter 3.1: Abstract ................................................................ 32
Chapter 3.2: Introduction ............................................................ 33
Chapter 3.3: Materials and Methods ............................................... 35
Chapter 3.4: Results .................................................................. 38
Chapter 3.5: Discussion ............................................................. 41
Chapter 3.6: Figures .................................................................. 44
Chapter 3.7: References ............................................................. 54

Chapter 4: Hazelnut allergy: Evidence that hazelnut can directly elicit

specific IgE antibody response via activating Type-2 cytokines in

mice ..................................................................................... 57
Chapter 4.1: Abstract ................................................................ 57

vi

Chapter 4.2: Introduction ............................................................ 59

Chapter 4.3: Materials and Methods ............................................... 61

Chapter 4.4: Results .................................................................. 63

Chapter 4.5: Discussion ............................................................. 65

Chapter 4.6: Figures .................................................................. 68

Chapter 4.7: References ............................................................. 75

Chapter 5: Characterization of the systemic immune response following

transdermal hazelnut protein exposure in BALB/c mice ......................... 78

Chapter 5.1: Abstract ................................................................ 78

Chapter 5.2: Introduction ............................................................ 79

Chapter 5.3: Materials and Methods ............................................... 80
Chapter 5.4: Results .................................................................. 82

Chapter 5.5: Discussion ............................................................. 84
Chapter 5.6: Figures .................................................................. 87

Chapter 5.7: References ............................................................. 98

Chapter 6: Development of an adjuvant-free model of tree-nut protein

induced systemic anaphylaxis ........................................................ 99

Chapter 6.1: Abstract ................................................................ 99

Chapter 6.2: Introduction ............................................................ 101
Chapter 6.3: Materials and Methods ............................................... 102
Chapter 6.4: Results .................................................................. 105
Chapter 6.5: Discussion ............................................................. 109
Chapter 6.6: Figures .................................................................. 112
Chapter 6.7: References ............................................................. 126
Chapter 7: Effect of a diet rich in EPA and DHA on systemic immune

responses to hazelnut: A pilot study ................................................. 128
Chapter 7.1: Abstract ................................................................ 128
Chapter 7.2: Introduction ............................................................ 130
Chapter 7.3: Materials and Methods ............................................... 132
Chapter 7.4: Results .................................................................. 135
Chapter 7.5: Discussion ............................................................. 138
Chapter 7.6: Figures .................................................................. 143
Chapter 7.7: References ............................................................. 180
Chapter 8: Future Studies ........................................................... 182

vii

Table No.

Table 1.1

Table 4.1

Table 6.1

Table 6.2

Table 6.3

Table 7.1

Table 7.2

Table 7.3

Table 7.4

LIST OF TABLES

Title

Prevalence of common food allergies in the United States. . . . . . . .

Hazelnut specific IgE antibody response in mice without the use of
alum adjuvant ...............................................................

I.P. challenge: study
design .........................................................................

Oral gavage challenge: study
design .........................................................................

Table 6.3 Hazelnut specific IgE antibody titers in BALB/c mice
pre-immune and at 6th response ............................................

Experiment used to assess DHA + EPA enriched oil on tree-nut
allergy ........................................................................

Fatty acid composition of experimental diets ............................

Spleen phospholipid fatty acid composition in mice fed various
PUFA: Preventative study .................................................

Spleen phospholipid fatty acid composition in mice fed various
PUFA: Therapeutic study ..................................................

viii

Page
N o.

74

102

102

105

153

154

155

168

Figure No.
Figures
3.6.1(A, B)

Figures 3.6.2
(A-B)

Figures 3.6.3
(A-B)
Figures 3.6.4
(A-B)

Figure 3.6.5
(A-C)

Figure 3.6.6
Figure 4.6.1
(A & B)

Figure 4.6.2A

Figure 4.6.23

Figure 4.6.3
(A-D)

Figure 4.6.4
(A-D)

Figure 5.1

LIST OF FIGURES

Title

Impact of coating antigen concentration on assay

sensitivity .........................................................

Immunoglobulin (Ig) epsilon isotype specificity of the

assay ...............................................................

Impact of Protein-G treatment on egg specific IgG1 and
IgG2a antibody levels in serum from mice sensitized

with egg ..........................................................

Comparison of assay sensitivity: ELISA vs. PCA .........

Inter and Intra assay variation of the ELISA method ......

Applications of the ELISA method to measure food

extract specific IgE Ab in mouse serum .....................

Immune response to hazelnut in C57BL/6 mice ............

Dose-response and time-course analyses of Type-2-
dependent allergenic response to hazelnuts in C57BL/6

mice ...............................................................

Dose-response and time-course analyses of Type-2-
dependent antigenic response to hazelnuts in C57BL/6

mice ...............................................................

Hazelnut specific antibody responses in mice strains
with differing MHC haplotype ...............................

Hazelnut specific Type-2 cytokine (IL-4 and IL-5)

responses in mice ................................................

Transdermal sensitization protocol ..........................

ix

Page
N o.

44

46

48

50

51

53

68

69

70

71

73

87

Figure 5.2 (A-
B)

Figure 5.3 (A-
B)

Figure 5.4 (A
& B)

Figure 5.5 (A-
B)

Figure 5.6 (A-
13)

Figure 5.7

Figure 6.1

Figure 6.2

Figure 6.3

Figure 6.4

Figure 6.5

Figure 6.6

Characterization of hazelnut specific IgE antibody
responses in BALB/c mice following repeated
transdermal exposure to hazelnut ............................

Characterization of hazelnut specific antibody IgGZa
antibody responses in BALB/c mice following repeated
transdermal exposure to hazelnut ............................

Dose-response and time-course analyses of IgE antibody
response to transdermal hazelnut exposure in BALB/c
mice ...............................................................

Hazelnut driven Type-2 cytokine (IL-4) responses in
mice transdermally sensitized with hazelnut or saline. .

Hazelnut driven Type-1 cytokine (INF-y) in mice
transdermally sensitized with hazelnut or saline. . . . . . . .

INF -y / IL-4 ratio in mice transdermally sensitized with
hazelnut ..........................................................

Transdermal sensitization protocol for systemic
anaphylaxis to hazelnut ........................................

Transdermal exposure to hazelnut sensitizes BALB/c
mice for systemic anaphylaxis when challenged by i.p.
injection: Clinical symptom scores ..........................

Transdermal exposure to hazelnut sensitizes BALB/c
mice for systemic anaphylaxis when challenged by i.p.
injection: Rectal temperature .................................

Transdermal exposure to hazelnut sensitizes BALB/c
mice for systemic anaphylaxis when challenged by i.p.
injection: Plasma histamine levels ...........................

Determination of threshold challenge dose of hazelnut
required to induce systemic anaphylaxis in hazelnut
sensitized BALB/c mice: Clinical symptom scores .......

Determination of threshold challenge dose of hazelnut
required to induce systemic anaphylaxis in hazelnut
sensitized BALB/c mice: Changes in rectal

temperature ......................................................

89

91

93

94

95

96

112

113

114

115

116

117

Figure 6.7

Figure 6.8

Figure 6.9 (A-
B)

Figure 6.10
(A-B)

Figure 6.11
(A-B)

Figure 6.12
(A-B)

Figure 6.13
(A-B)

Figure 6.14
(A-B)

Figure 7.1 (A-
B)

Figure 7.2

Figure 7.3 (A-
B)

Transdennal exposure to hazelnut sensitizes BALB/c
mice for systemic anaphylaxis when challenged by oral
gavage: Clinical symptom scores ............................

Transdermal exposure to hazelnut sensitizes BALB/c
mice for systemic anaphylaxis when challenged by oral
gavage: Changes in rectal temperature ......................

Transdermal exposure to hazelnut sensitizes BALB/c
mice for pathological changes in the gastrointestinal tract
when challenged by oral gavage ..............................

Transdermal exposure to hazelnut sensitizes BALB/c
mice for pathological changes in the gastrointestinal tract
when challenged by oral gavage ..............................

Transdermal exposure to hazelnut sensitizes BALB/c
mice for pathological changes in the gastrointestinal tract
when challenged by oral gavage ..............................

Transdermal exposure to hazelnut sensitizes BALB/c
mice for pathological changes in the gastrointestinal tract
when challenged by oral gavage ..............................

Transdermal exposure to hazelnut sensitizes BALB/c
mice for pathological changes in the gastrointestinal tract
when challenged by oral gavage ..............................

Transdermal exposure to hazelnut sensitizes BALB/c
mice for pathological changes in the gastrointestinal tract
when challenged by oral gavage ..............................

Comparison of hazelnut specific IgE antibody responses
in BALB/c mice following repeated transdermal
exposure to hazelnut protein in different control diets ......

Comparison of total IgE in BALB/c mice following
repeated transdermal exposure to hazelnut protein in
different control diets ...........................................

Comparison of hazelnut specific IgG2a antibody
responses in BALB/c mice following repeated
transdermal exposure to hazelnut protein in different
control diets ......................................................

xi

118

119

120

121

122

123

124

125

143

145

146

Figure 7.4

Figure 7.5

Figure 7.6

Figure 7.7

Figure 7.8 (A-
B)

Figure 7.8 (C-
I3)

Figure 7.9 (A-
D)

Figure 7.10

Figure 7.11
(A-B)

Figure 7.12
(A-B)

Figure 7.13

Comparison of different control diets: Hazelnut driven
Type-2 cytokine (IL-4) responses in mice transdermally
sensitized with hazelnut protein ..............................

Comparison of different control diets: Hazelnut driven
Type-1 cytokine (IFN-y) responses in mice
transdermally sensitized with hazelnut protein .............

IFN-y / IL-4 ratio in mice transdermally sensitized with
hazelnut protein: Comparison of control diets ..............

Transdermal sensitization protocol to hazelnut protein:
preventative study ..............................................

Characterization of hazelnut specific IgE antibody
responses in plasma from BALB/c mice fed the test diet
enriched with EPA and DHA vs. control diet
(preventative study) .............................................

Characterization of hazelnut specific IgE antibody
responses in plasma from BALB/c mice fed the test diet
enriched with EPA and DHA vs. control diet
(preventative study) .............................................

Characterization of hazelnut specific IgG2a antibody
responses in plasma from BALB/c mice fed the test diet
enriched with EPA and DHA vs. control diet
(preventative study) .............................................

Characterization of Total IgE responses in BALB/c mice
fed the test diet enriched with EPA and DHA vs. control
diet ................................................................

Hazelnut driven Type-2 cytokine (IL-4) responses in
mice transdermally sensitized with hazelnut protein:
effect of experimental diets ....................................

Hazelnut driven Type-l cytokine (IFN-y) responses in
mice transdermally sensitized with hazelnut protein:
effect of experimental diets ....................................

IFN-y / IL-4 ratio in mice transdermally sensitized with
hazelnut protein: Effect of experimental diets .............

xii

148

149

150

152

156

158

160

162

163

164

165

Figure 7.14

Figure 7.15
(A-B)

Figure 7.15
(GB)

Figure 7.16
(A-D)

Figure 7.17

Figure 7.18
(A-B)

Figure 7.19
(A-B)

Figure 7.20

Transdermal sensitization protocol to hazelnut protein:
therapeutic study ................................................

Characterization of hazelnut specific IgE antibody
responses in plasma from BALB/c mice fed the test diet
enriched with EPA and DHA vs. control diet
(therapeutic study) ..............................................

Characterization of hazelnut specific IgE antibody
responses in plasma from BALB/c mice fed the test diet
enriched with EPA and DHA vs. control diet

(therapeutic study) ..............................................

Characterization of hazelnut specific IgG2a antibody
responses in plasma from BALB/c mice fed the test diet
enriched with EPA and DHA vs. control diet

(therapeutic study) ..............................................

Characterization of Total IgE responses in BALB/c mice
fed the test diet enriched with EPA and DHA vs. control

diet (therapeutic study) .........................................

Hazelnut driven Type-2 cytokine (IL-4) responses in
mice transdermally sensitized with hazelnut protein:
Effect of experimental diets ....................................

Hazelnut driven Type-1 cytokine (DIP-y) responses in
mice transdermally sensitized with hazelnut protein:
Effect of experimental diets ....................................

lFN-y / IL-4 ratio in mice transdermally sensitized with
hazelnut protein .................................................

* Images in this thesis/dissertation are presented in color.

xiii

167

169

171

173

175

176

177

178

ABBREVATIATION S USED

ASIgE, Allergen specific IgE; Ig, immunoglobulin; OD, optical density; ELISA, enzyme
linked immunosorbent assay; PCA, passive cutaneous anaphylaxis assay; BSA, bovine
serum albumin; PBS, phosphate buffered saline; SD, standard deviation, SE, standard

error; RAST, radio allegro sorbent test

xiv

CHAPTER ONE

1.0 Introduction

Allergy or immediate hypersensitivity can be described as an abnormal response of the
immune system to substances that are normally harmless (pollens, foods, drugs). Food
allergy is an immune mediated adverse reaction to food that occurs when the immune
system misidentifies food as a foreign chemical in the body and responds robustly to one

or more specific proteins in that food.

An adverse reaction to food encompasses any abnormal reaction after the food is
ingested. It may be due to a food intolerance (adverse physiologic response), or to a food
allergy (immunological reaction) [1]. Food intolerances might be due to toxic
contaminants (e.g. histamine in scombroid fish poisoning) or pharmacologic properties of
the food (e. g. tyramine in aged cheese), or it may be due to metabolic disorders in the
host (e.g. lactase deficiency) or even possible idiosyncratic responses [2]. Food allergies,
however, are immune-mediated reactions to food, commonly called food
hypersensitivities. These can be due to IgE mediated or non-IgE mediated immune
mechanisms. Therefore, the involvement of the immune system is a key-deciding factor

to determine if an adverse reaction to food is truly a food allergy.

Hypersensitivity has been defined as an inappropriate response of the immune system to
antigens (a substance, usually a protein, that stimulates an antibody immune response).

According to the scheme of Gel] and Coombs [3, 4] , there are four types of

hypersensitivity reactions classified as Type-I to Type-IV depending on both types of
antigens involved and mechanisms of disease. Type I hypersensitivity, commonly called
“immediate type hypersensitivity”, is a rapid adverse reaction mediated by IgE antibody
against a soluble allergen (subclass of antigen, where antibody production is of the IgE
isotype). Here IgE antibodies are bound to high affinity receptors on mast cells and
basophils. Upon antigen cross-linking, cells degranulate, releasing various chemical
mediators (e. g. histamine), which results in onset of disease symptoms. Food allergy
belongs to this type of hypersensitivity with peanut allergy being an example widely
seen. Clinically, Type I hypersensitivities are expressed extremely fast with symptoms
such as hives, rashes, asthma, vomiting, diarrhea etc. Skin prick testing and serum

screening for specific IgE antibodies diagnose this type of hypersensitivity.

In Type II hypersensitivity, IgG, IgM and complement play key roles, here with an
insoluble antigen. Antibodies bind to the antigen and activate the complement system,
which in turn initiates the release of mediators of disease. Examples of type II
hypersensitivity include autoimmune hemolytic anemia, rheumatic fever, drug allergies

etc. [5]. Currently, it is unclear if this mechanism plays a role in food allergy.

Type III hypersensitivity is associated with IgG and IgM binding to circulating antigens,
then forming immune complexes. These immune complexes become large mesh works
with antigen bridges. Complement is activated and the immune complex is cleared. If

failure to clear these immune complexes persists, disease events occur. An example of

type III hypersensitivity is systemic lupus erythematosus. It is not known whether this

mechanism may play role in food allergies.

Type IV hypersensitivity, commonly called “delayed type hypersensitivity”, is mediated
by CD4+ T-cells and intact antigen presenting cells. Reactions such as poison ivy,
contact dermatitis and nickel allergy all are type IV hypersensitivities. Gluten
enteropathy (or Celiac disease) is also a Type IV hypersensitivity and is mediated by a

non-IgE mechanism involving T cells and monocytes [6, 7].

Food allergy has been estimated to affect 3.7% of American adults [8]. Its prevalence is
at its highest during the first few years of life, affecting about 6% of infants less than 3
years of age and then decreases during childhood [9]. Whereas some food allergies such
as cow’s milk are likely to be outgrown [1], several food allergies, such as tree-nut and
peanut allergy can be chronic, often lasting a lifetime [8]. Recently, Sampson assessed

the prevalence of food allergy in the United States and is summarized in Table 1.1 [8].

Table 1.1 Prevalence of common food allergies in the United States.

 

 

Food Young Children Adults
Milk 2.5% 0.3%
Egg 1.3% 0.2%
Peanut 0.8% 0.6%
Tree nuts 0.2% 0.5%
Fish 0.1% 0.4%
Shellfish 0.1% 2.0%
Overall 6.0% 3.7%

 

* From Sampson et a1. 2004 [8].

Food allergy remains the leading cause of systemic anaphylaxis treated in emergency
departments in a number of countries, including the United States and the public has
become increasingly aware of the problem [8]. Food—related allergic reactions account
for around 30,000 emergency room visits [2] and 150-200 deaths in America alone each
year [8]. Peanuts and tree-nuts (e.g., hazelnuts, almonds etc) are the major food types

that cause systemic anaphylaxis with fatal or near fatal consequences [2, 10].

The pathogenesis of food allergy may be broken down to two phases, the sensitization
phase and the effector phase. In the sensitization phase, allergen is encountered for the
first time, allergen is presented to the B-lymphocytes by antigen presenting cells
(macrophages, dendritic cells) that then produce allergen specific IgE with the aid of T-
lymphocytes (Th2) [11]. These IgE antibodies then bind to high affmity receptors

(F CeRI) on both mast cells and basophils. In the effector phase, recurrent exposure to the
allergen, causes cross-linking of two bound IgE antibodies causing degranulation of the
mast cells and basophils, releasing mediators of disease (histamine, prostaglandins,

leukotrienes etc. [12].

Most symptoms of food allergy are immediate, occurring within minutes to hours after
food ingestion [4]. They range from skin reactions (hives, itching, eczema, swelling), to
gastrointestinal distress (nausea, vomiting, diarrhea, cramps), to respiratory troubles

(wheezing, sneezing, asthma, rhinitis, laryngeal edema, labored breathing, tightness of

throat) and systemic symptoms including anaphylactic shock with decreased body

temperature and blood pressure, which could potentially lead to death [13, 14].

While more than 100 different food types have been documented to trigger allergic
reactions in sensitized humans, 90% of food allergies are caused by only eight food types,
often called the red-flag foods (Egg, milk, wheat, soy, peanuts, tree-nuts, fish and
shellfish) [15]. The scientific reasons as to why only eight food types account for a vast
majority of food allergies are unclear. Despite the potential for a fatal outcome, food
avoidance is the only sure way to prevent food allergy episodes. Epinephrine is usually

prescribed to patients with food and can be used to stop an anaphylactic reaction.

There is growing concern and indication that the prevalence of food allergy might be
increasing consistent with asthma and other allergic diseases for reasons that are not clear
[8, 16, 17]. Even though extensive research is ongoing about food allergies, mechanisms
driving this increasing trend are unclear at present. Therefore more research needs to be
done so that this increasing trend can be stopped. To do this, well-characterized animal
models need to be developed. Then potential therapies can be developed and studied.
Currently there are several animal models available to study food allergy, including
rodent models for several foods including peanut, cow’s milk and egg. However, a
mouse model to study tree-nut allergy was not available when we began our work in

2001.

With this gap in the knowledge, I sought out to develop and characterize a mouse model
of tree-nut allergy, using hazelnut as a model tree-nut, which might be useful to study
tree-nut allergy. In this study, my guiding hypotheses were i) mice can deve10p tree-nut
allergy that mimic certain phenotypes of human tree-nut allergy and ii) a validated mouse
model of tree-nut allergy is useful to study impact of dietary modification on various

markers of this disease.

The aims of my studies to test these hypotheses were 1) to develop an ELISA based
method to measure allergen specific IgE as an alternative to the passive cutaneous
anaphylaxis assay; 2) to determine if hazelnut can directly elicit a specific IgE antibody
response via activating IL-4 in mice; 3) to characterize the systemic immune response
following transdermal hazelnut protein exposure in BALB/c mice; 4) to develop an
adjuvant free model of hazelnut protein induced systemic anaphylaxis; 5) to study the

effect of a diet rich in EPA and DHA on systemic immune responses to hazelnut.

In the following chapters I review the published animal models of food allergy and then

show the detailed studies done to achieve my study aims.

1.1 References

l.

10.

ll.

12.

13.

14.

15.

Sampson, H.A., Food allergy. Part 1 : immunopathogenesis and clinical
disorders. J Allergy Clin Irnmunol, 1999. 103(5 Pt 1): p. 717-28.

Sampson, H.A., 9. Food allergy. J Allergy Clin Irnmunol, 2003. 111(2 Suppl): p.
S540-7.

Janeway, C., Immunobiology. 6th ed. 2005: Garland Science.

Stephen T. Holgate, M.K.C., and Lawrence M. Lichtenstein, Allergy. 2nd ed.
2001: Mosby.

Solensky, R., Drug hypersensitivity. Med Clin North Am, 2006. 90(1): p. 233-60.

Macdonald, T.T. and G. Monteleone, Immunity, inflammation, and allergy in the
gut. Science, 2005. 307(5717): p. 1920-5.

Sampson, H.A., Clinical manifestations of adverse food reactions. Pediatr Allergy
Irnmunol, 1995. 6 Suppl 8: p. 29-37.

Sampson, H.A., Update on food allergy. J Allergy Clin Irnmunol, 2004. 113(5): p.
805-19; quiz 820.

Sicherer, S.H., A. Munoz-Furlong, and H.A. Sampson, Prevalence of seafood
allergy in the United States determined by a random telephone survey. J Allergy
Clin Immunol, 2004. 114(1): p. 159-65.

Bernstein, J .A., et al., Clinical and laboratory investigation of allergy to
genetically modified foods. Environ Health Perspect, 2003. 111(8): p. 1114-21.

Helm, R.M., Food allergy animal models: an overview. Ann N Y Acad Sci, 2002.
964: p. 139-50.

Helm, R.M., R.W. Errnel, and CL. Frick, Nonmurine animal models of food
allergy. Environ Health Perspect, 2003. 111(2): p. 239-44.

Sampson, H.A., Anaphylaxis and emergency treatment. Pediatrics, 2003. 111(6 Pt
3): p. 1601-8.

Sampson, H.A., L. Mendelson, and J .P. Rosen, Fatal and near-fatal anaphylactic
reactions to food in children and adolescents. N Engl J Med, 1992. 327(6): p.
380-4.

Lehrer, S.B., R. Ayuso, and G. Reese, Current understanding of food allergens.
Ann N Y Acad Sci, 2002. 964: p. 69-85.

16.

17.

Sicherer, S.H., A. Munoz-Furlong, and H.A. Sampson, Prevalence of peanut and
tree nut allergy in the United States determined by means of a random digit dial
telephone survey: a 5 -year follow-up study. J Allergy Clin Irnmunol, 2003.

112(6): p. 1203-7.

Kagan, R.S., et al., Prevalence of peanut allergy in primary-school children in
Montreal, Canada. J Allergy Clin Irnmunol, 2003. 112(6): p. 1223-8.

CHAPTER TWO

2.0 Review of Literature

2.1 Animal models of food allergy

With this increasing trend in food allergy prevalence and the fact the human prospective
sensitization studies are not ethically possible, animal models that mimic human allergic
responses need to be studied and developed. Animal models hold the potential to be
extremely valuable tools for determining mechanisms, predicting triggers, and testing

possible treatments/therapies.

Development of an animal model for food allergy should take into account the following
parameters: 1) the concentration of the allergen (high doses are known to induce
tolerance); 2) the allergen should be taken in context with the food source; 3) the route
and duration of allergen exposure; 4) genetic predisposition (high and low IgE
responders); 5) the use of adjuvants (agents which not having any specific antigenic
effects itself, stimulates the immune system, increasing the immune response to what it is
in contex with) (natural or artificial-alum, cholera toxin, Bordetella pertussis, and
carrageenan are known IgE-selective adjuvants); 6) isotype specificity response (mice
respond with two anaphylactic antibodies, IgG] and IgE; rats with IgG2a and IgE; guinea
pigs with IgG1 and IgE; dogs with IgE; and pigs, likely with IgE; and 7) the Th1/Th2
polarization (mice have very delineated Th1/Th2 polarization, whereas in humans

polarization is not as discrete) [1].

Animal models currently available include mice, rats, guinea pig, atopic dog, and
neonatal swine. Reviewed here is a list of animal models that have been developed

for food allergy.

2.2. Rat model of Ovalbumin allergy

A number of studies have used the Brown Norway rat for development of a model of
OVA allergy [2-5]. A group headed by Leon Knippels at the TNO nutrition and Food
Research Institute have developed a rat model of OVA allergy [3, 4]. They used Brown
Norway rats and exposed to ovalbumin either ad libitum via the drinking water (0.002 to
20 mg/mL) continuously for 6 weeks or by gavage (1 mg/mL per rat) without the use of
adjuvant. Gavage was performed daily, twice a week, once a week or once every 2 weeks
during a period of 6 weeks. Following sensitization, animals were assessed for OVA

specific IgE and IgG].

Afier intra-gastric administration of ovalbumin once or twice a week or once every two
weeks, a very low frequency of ovalbumin-specific antibody responses were detected.
Daily intra-gastric dosing with ovalbumin resulted in antigen-specific IgG as well as IgE
responses in ahnost all animals tested (7/ 8 at day 42 of sensitization). With ad libitum
exposure, ovalbumin-specific IgG was noted, but ovalbumin-specific IgE was less than

detectable.

10

These studies show that the BN rat may provide a suitable animal model for inducing
specific IgG and IgE responses to OVA upon exposure via the enteral route without the

use of adjuvants.

Furthering the earlier work, efforts were made to characterize the models immune-
mediated effects after oral challenge with OVA [3]. Here Brown Norway rats were
exposed to ovalbumin (OVA) by daily gavage dosing (1 mg OVA/rat/day) for 6 weeks,
without the use of an adjuvant, or by i.p. injections (positive control) with OVA
(0.2mg/ml) together with alum (5mg). Subsequently, effects on breathing frequency,
blood pressure, and gastrointestinal permeability were investigated upon an oral

challenge with 10 to 100 mg OVA in vivo.

In both i.p. and orally sensitized rats, an increase in gut permeability (increased passage
of beta-lactoglobulin as bystander protein) was determined between 0.5 and l h after an
oral OVA challenge was given. An oral challenge with OVA did not induce a clear
effect on the respiratory system or blood pressure in the majority of the animals.
Whereas, i.p. sensitized animals had a significant drop in blood pressure following i.v.
challenge with OVA. Upon oral challenge with OVA of orally and parenterally
sensitized animals, local effects were observed in all animals whereas systemic effects
were observed at a low frequency, which reflects the situation in food allergic patients

after an oral challenge.

11

The authors conclude from these studies that the Brown Norway rat provides a suitable
animal model to study oral sensitization to food proteins as well as immune-mediated

effects after oral challenge with food proteins.

Limitations of these studies include the lack of cytokine data to support the Th2
dominated response. Also, systemic effects after oral challenge were minimal even afier
rather large challenge doses of 100 mg of a purified allergen (OVA). Further analysis
into the local effects could have been done using histological analysis on the

gastrointestinal tract.

2.3 The Guinea pig model of cow’s milk proteins

A group located at the University of Texas Medical Branch has developed a model to
study the allergenicity of cow’s milk proteins in a guinea pig model [6]. In this study, the
allergenicity of milk-based infant formula and cow’s milk proteins were evaluated by
examining altered intestinal permeability, intestinal anaphylaxis, and PCA after oral
sensitization with cow’s milk proteins and challenge with B-lactoglobulin. Colonic
segments from cow’s milk-sensitized or milk formula—sensitized animals were
challenged with B-lactoglobulin in Ussing chambers. Cow’s milk—sensitized animals
responded with an anti gen-induced, anaphylactically mediated elevation in the transmural
short-circuit current as measured by net chloride secretion, whereas only 60% of animals
fed infant formula responded to challenge. Bronchospasm developed in all animals fed
cow’s milk; however, only those animals fed infant formula that responded to intestinal

challenge developed bronchospasm.

12

The authors concluded that cow’s milk—based infant formula had less sensitizing power
than whole cow’s milk and that the model was effective in testing allergenicity at the

intestinal level.

The strengths of this study are the fact that they fed the allergen without the use of
adjuvant. Limitations of this study are that they did not assess antibodies driving these
responses or cytokine profile leading to the response, therefore a true allergy is not
evident unless further work is done. Furthermore difficulties associated with passive
cutaneous anaphylaxis (PCA) testing and the fact that the antibody response is of the
IgGla subtype and not IgE limit the use of guinea pigs as a suitable model with which to
study food allergy associated mechanisms [7 , 8]. Finally, reagents and gene knockout

strains are not readily available as they are for other species of animal.

2.4 Atopic dog model for multiply foods (Cow’s milk, beef, ragweed, and wheat)
An atopic dog colony at the University of California has been under development as a
model of food allergy [9-11]. This colony is a spaniel/basenji-type dog that was selected
with a genetic predisposition to allergy and that had histories of sensitivity to pollens and
foods. Ennel et a1. [11] used newborn pups that they subcutaneously injected into the
axilla a mixture of allergens containing 1 ug each (cow's milk, beef, ragweed, or wheat)
commercial extract, with alum (200 pl) as an adjuvant. At ages 3, 7, and 11 weeks, pups
were vaccinated with attenuated distemper-hepatitis vaccine. At 2 and 9 days after each
vaccination, pups received a boost of the same allergen: alum injection they received as

newborns. Booster injections were IOug of allergen extract in 200ul alum.

l3

Irnmunized pups responded with allergen-specific IgE by week 3, which peaked at week
26, and could be maintained with injections of antigen with alum every other month and
daily feeding of diet containing small amounts of the allergen. Skin tests were positive
when challenged with the immunized allergen as evidenced by a wheal-and-flare reaction
and negative with a control, un-immunized allergen. Analysis of gastric food sensitivity
was done through an endoscope by injecting allergenic food extracts into the gastric
mucosa after intravenous injection of Evans blue dye. Tissue examination showed
marked mucosal swelling and persistent erythema at the site of this allergen injection.
Furthermore, late biopsy specimens revealed eosinophil infiltration into the lamina
propria and migration through the endothelium. From these results the authors conclude

that the dog may serve as a useful model to study food allergy [11].

Strengths of this model include that the dog is one of the few species other than humans
in which allergies develop naturally on normal environmental exposure to a broad
spectrum of allergens, including pollens, house dust mites, human dander, fleas, and
foods [12-14]. Also, the dog can respond both vomiting and diarrhea in response to oral
challenge due to food allergy [15 , 16]. Furthermore, food allergy has identified by
single-ingredient elimination testing in 25 dogs with histories and cutaneous signs were

consistent with food-induced allergic dermatitis [17].

Some of the limitations associated with this model include the high cost of care and

housing of a large animal like a dog would limit the studies. Having a large number of

14

animals per study would be difficult. The ease of getting reagents and gene knock out
animals would be difficult. They did not perform cytokine analysis to see if a Th1/Th2
imbalance is driving this allergy. Therefore, using the atopic dog as a model of

mechanistic and therapeutic analysis of food allergy is challenging.

2.5 Atopic dog model for multiply foods (Peanut, walnut, Brazil nut)

Further work was done on the atopic dog colony at the University of California with the
goal being the development of an animal model for peanut and tree-nut allergy [16].
They used their colony of a spaniel/basenji-type dog that had been selected with a genetic
predisposition to allergy and that had histories of sensitivity to pollens and foods. Here
they used a group of eleven dogs and put them through the previous protocol, injecting
them subcutaneously with lug of peanut, English walnut, soy and Brazil nut (commercial
extract) with alum (200 pl) as an adjuvant. The dogs were also sensitized to either wheat
or barley (1 ug in 200p] alum). At ages 3, 7, and 11 weeks, pups were vaccinated with
attenuated distemper-hepatitis vaccine. At 2 and 9 days after each vaccination, pups
received a boost of the same allergen: alum injection they received as newborns. Booster
injections were lOug of allergen extract in 200ul alum. Skin testing, IgE

immunoblotting, and oral challenges to allergen were preformed.

Dogs responded at 6 months of age with positive intradermal skin reactions to the nut
allergens. IgE immunoblotting showed a strong recognition to peanut, walnut and Brazil
nut in the aqueous preparations. Proteins binding to the IgE are similar to the profile of

major allergens seen in human food allergy (peanut-Ara h 1, walnut-Jug r 2, Brazil nut-

15

Ber e 1). At 2 years of age, each of the 4 peanut and the 3 Brazil nut sensitized dogs and
3 out of the 4 walnut sensitized dogs reacted to oral challenge with the allergen they were

sensitized to with symptoms such as vomiting and lethargy.

Strengths of this study include that the dog is one of the few species other than humans in
which allergies develop naturally on normal environmental exposure to a broad spectrum
of allergens, including pollens, house dust mites, human dander, fleas, and foods [12-14].
Also, the dog can respond both vomiting and diarrhea in response to oral challenge due to

food allergy [15, 16].

Limitations of this study include the length between sensitization and oral challenge.
Dogs are challenged at 2 years of age, thus this model would be a long and costly one
that would delay scientific progress. Also, the high cost of care and housing of a large
animal like a dog would limit the studies making having a large number of animals per
study difficult. The ease of acquiring reagents and gene knock out animals would be
challenging. Mechanistic studies to assess cytokine profile were not done. Furthermore,
multiple allergens are injected at the same time, thus making it difficult to assess the true
allergenic potential of each individual food due to competition and possible cross-
reactivity. Therefore, using the atopic dog as a model of mechanistic and therapeutic

analysis of food allergy is challenging.

l6

2.6 The Neonatal swine model of peanut allergy

A group lead by Ricki Helm at the University of Arkansas has been using the neonatal
swine to develop a model for peanut allergy [18]. Initially, they used both intragastric

(i. g.) and intraperitoneal (i.p.) sensitizations followed by oral challenge with peanut to
optimize a sensitization/challenge protocol. From the early studies they found that
approximately 25% of i. g. sensitized animals and 75—90% of i.p. sensitized animals
responded to an oral challenge of peanut meal. Thus, they concluded that the optimal
experimental protocol was to use i.p. sensitization of peanut extract and oral challenge
with peanut meal. From that they came up with the following protocol. Out bred Large
White/Landrace pregnant sows at day 108 of gestation are allowed to nurse under normal
conditions on a soybean/peanut-free diet. Following birth, piglets at days 9—1 1, 17, and
25 of age were i.p. sensitized with 500 pg of peanut extract with 100 pg cholera toxin.
Random selection of animals in each litter to receive control treatments, either phosphate
buffered saline (PBS) or PBS with 100 pg of cholera toxin was done. I.g. challenge with
peanut meal and intradermal skin testing was performed every other week starting 2
weeks after the final sensitization. Blood was taken weekly to assess the immune

responses to the sensitization protocol.

Oral challenges were preformed on days 39 and 53 with 10 or 20 g of peanut meal and
resulted in symptoms in 75—100% of animals by the second oral challenge within 30—60
rrrinutes of the challenge. Symptoms following oral challenge included emesis, malaise,
tremors, and convulsions with major and minor rashes. There was evidence of respiratory

distress and anaphylactic shock in approximately 10—20% of sensitized animals whereas

17

no shock was noted in control animals. Animals entering shock were treated with
epinephrine to alleviate all symptoms. When oral challenge was repeated up to day 90
sensitized animals responded with increasing degrees of physical symptoms, whereas the
control animals challenged with peanut meal did not respond with any physical,
gastrointestinal or systemic sign of allergy. Peanut-sensitized animals challenged with a
irrelevant allergenic food (soybean/peanut-free diet) did not show any symptoms, thus

showing the specificity to the previous peanut challenge.

Skin testing was done on alternating weeks with peanut confirmed the persistence of the
allergic state throughout a 14-week period. Using either the native or recombinant forms
of the major peanut allergens, Ara h l and Ara h 2, induced a positive skin test when
compared to rice extracts, thus showing reactivity to allergens similar to what is seen in
human peanut allergy. Skin prick tests with peanut extracts intraderrnally were also
positive showing a wheal and flare > 5—15 mm. The PBS and PBS/cholera toxin control
groups skin prick tests were negative (2 mm) with peanut extract, whereas the histamine

positive control showed positive wheal and flare.

The animals were assessed for the production of antigen-specific IgG and IgE by passive
cutaneous anaphylaxis. Peanut-specific IgG values measured in peanut-sensitized
animals reached levels > 1,000 pg/mL (range, 26—7,700 pg/mL) by day 37 and
maintained values of > 500 pg/mL (range, 5 1-1,500 pg/mL) at day 60. Non-peanut-
sensitized animals had < 50 pg/mL antigen-specific IgG. To prove that peanut specific

IgE is the responsible isotype causing the allergic symptoms following oral challenge,

18

passive cutaneous anaphylaxis tests were performed in naive animals. One hundred micro
liters of unheated and serial heat-inactivated serum from peanut-sensitized pigs was
administered intraderrnally into the back of naive animals. 24 hours later, 5 mg of peanut
extract was administered by i.v. injection. After 30 minutes the responses were noted.
Intradermal skin sites with the unheat-inactivated serum responded with a wheal and flare
> 10 mm at the site of injection, whereas heat-inactivated serum showed no reaction,
confirming that native IgE was responsible for the reaction because the peanut specific

IgE is denatured in heat, while the IgG was left intact.

Following the last oral challenge, the gastrointestinal tract was taken and assessed
pathological alterations. The histologic findings were vascular congestion, hemorrhage,
and epithelial denudation that in the proximal small intestine. Other acute markers
included mucus extrusion and submucosal edema in the stomach. The colon seemed
normal in most piglets, with occasional vascular congestion and crypt abscesses.

From these results the authors conclude that the neonatal pig model of peanut allergy
mimics the physical and immunological characteristics of peanut allergy in humans.
Therefore, this model should be useful for determining IgE-mediated mechanisms and

immunotherapeutic intervention strategies with repeated allergen challenges.

Some of the strengths of this model include i) how the swine closely resemble humans in
gastrointestinal physiology and the development of mucosal immunity is also similar to
that seen in humans [1]; ii) The developing piglet has similar anatomy and nutritional

requirements, a distribution and maturation of intestinal enzymes, and an enteric

19

absorption of antibody that is similar to that of the developing infant [1]; iii) The
newborn piglets are born immunocompetent, thus allowing for assessment of immune
responses [19]. iv) Hypersensitivity responses similar to those of human allergic disease
have been demonstrated in swine [20]. Furthermore, studies in veterinary medicine have
shown swine to have an IgE-mediated-like response to parasites, legumes, and pollens

reminiscent of that in humans

Limitations of using the neonatal pig model include how the pig is not a routinely used
animal model; therefore reagents and gene knock out strains are sparse, making some
research impossible. Also the size of the pig makes studies with large numbers extremely
costly. Here, mechanisms (cytokine profile) driving this allergic response were not
studied. Finally, the use of adjuvant and route of sensitization are also drawbacks of this
model because human exposure is not likely by an injection, a better route of
sensitization would be desired. Therefore, the use of the neonatal pig as a model of food

allergy could pose to be difficult.

2.7 Mouse models of food allergy

The mouse is one of the widest used animals in laboratory sciences today. With the ever
increasing number of gene knock out strains available and reagents, more and more
molecular and mechanistic studies can be done that can not be done in other, larger
species. For reasons like these, mouse models of allergy are valuable tools in allergy

research and need to be profiled further.

20

There has been several strains of inbred mice that have been characterized as being either
high or low IgE-responder animals for food allergens [18]. As in humans, two separate
events are required: the first event is a sensitization phase and, in the case of mice, the
production of two anaphylactic antibodies, IgE and IgG1; the second phase (effector) is

characterized as the allergic response following allergen challenge [18].

There are numerous models of food allergy using different allergens, routes of exposure,
with or without adjuvant and strain of mice. Here I review the models most widely used

that encompass both phases of allergy, the sensitization phase and the effector phase.

2.8 Mouse model of Cow’s milk allergy

A group at The Mount Sinai School of Medicine headed by Hugh Sampson has
developed a mouse model of Cow’s milk allergy. To do this Li et al., [21] used several
different strategies to overcome oral tolerance in a mouse model and induce IgE-
mediated cow’s milk hypersensitivity. They used three-week-old C3H/HeJ female mice
and sensitized them intragastrically with Homogenized Cow’s milk (0.01 mg, 0.1mg, or
1.0 mg/g body weight) plus cholera toxin (0.3 pg/ g) as an adjuvant and were boosted five
times at weekly intervals with the same dose of allergen and cholera toxin. Six weeks
afier the initial sensitization dose, mice were fasted and then intragastrically challenged
with two doses of cow’s milk (30 mg/ml) given 30 minutes apart. Hypersensitivity
responses were assessed based on systemic anaphylaxis symptom scores, vascular
leakage, plasma histamine release, PCA, serum antibody titers, skin testing, and

histological examination.

21

Their findings were that these mice exhibit several characteristics of human IgE-mediated
cow’s milk—induced food allergy. Mice had elevated cow’s milk specific IgE antibody
levels at 3 weeks and peaked at 6 weeks afier initial sensitization. Elevated allergen-
specific IgE levels were shown to be associated with systemic anaphylaxis, whereas
levels of IgG] were not; heating of serum from sensitized mice eliminated PCA reactions
in naive mice; and mast cell degranulation was evident because of elevated plasma
histamine levels (CM-sensitized (1mg/ g plus cholera toxin) mice (4144 +/-1244) vs. (661
+/- 72 nmol/L) in sham sensitized mice), all of which are important features of IgE-
mediated food allergy. Levels of serum casein after oral challenge were consistent with
intestinal permeability studies and histological examination revealed changes in both the
GI (vascular congestion, edema, enterocyte sloughing) and respiratory (increased
perivascular and peribronchial lymphocytes, monocytes, and eosinophils systems.
Furthermore, the development of this IgE-mediated hypersensitivity is likely to be Th2
cell mediated because in vitro stimulation of spleen cells from sensitized mice to cow’s
milk induced significant increased in the levels of IL-4 (44 pg/ml), IL-5 (68 pg/ml) but

not INF-y (4 pg/ml).

The authors conclude fi'om this study that this model should provide a useful tool for

evaluating the immunopathogenic mechanisms involved in cow’s milk allergy and for

exploring new therapeutic approaches [21].

22

Strengths of this model include i) The use of a dose study to determine amount of
allergen to sensitize with; ii) Determination of cytokine profile, showing mechanism

driving IgE; iii) Did oral challenge after sensitization and studied systemic anaphylaxis.

Some limitations of this study include i) The use of cholera toxin as an adjuvant does not
mimic how humans are likely sensitized to peanut allergens; ii) A profile of the IL-4 to

INF-7 ratio could have been done to show the balance in Th1 and Th2 cytokines.

2.9 Mouse model of peanut allergy

As in the cow’s milk allergy model, in able to mimic the clinical and immunological
characteristics of peanut allergy, Sampson’s group used female C3H/HeJ mice and
sensitized them orally with freshly ground whole peanut and cholera toxin as adjuvant
[22]. Five-week-old mice were sensitized by intragastric gavage with either a low dose
of 5 mg/ml of ground whole peanut (1 mg of peanut protein) or high dose of 25 mg of
ground whole peanut (5 mg of peanut protein) together with 10 pg cholera toxin on day 0
and day 7. Three and five weeks after initial sensitization, mice were fasted overnight
and challenged intragastrically with 10 mg crude peanut extract divided into two doses at

30- to 40-minute intervals.

Following challenge mice had fatal or near fatal anaphylaxis that occurred in 12.5% of
sensitized mice at 3 weeks. At the second challenge (five weeks challenge) symptoms
following challenge were more severe, increasing the fatality rate to 21%., whereas the

colera toxin alone control group had no responses. Peanut-specific IgE and titers were

23

significantly increased at weeks 1—5 with the low dose eliciting more peanut specific IgE.
IgGl levels did not differ between low-dose and high-dose sensitizations, suggesting this
antibody did not play a significant role in inducing anaphylaxis in this model. PCA
reactions were done to confirm what isotypes were contributing to the systemic
anaphylaxis. Heat-inactivated serum did not cause a positive response, whereas serum
fi'om the peanut sensitized mice did, confirming that the anaphylactic response to be IgE-
induced and not IgGl. IgG2a levels were significantly higher in the high-dose versus low
dose sensitization, suggesting that IgG2a was inversely related to the severity of peanut
hypersensitivity. Mast cell degranulation and histamine levels in the plasma were also

assessed and both associated with peanut sensitization and peanut challenge.

From this study the authors concluded that this model of peanut allergy mimics the
clinical and immunological characteristics of peanut allergy in humans and should serve

as a useful tool for developing therapeutic approaches for the treatment of peanut allergy.

Strengths of this model include a number of hallmarks of human allergy are seen such as,
increased plasma histamine, allergen specific IgE levels, and anaphylaxis following
challenge. Also, a dose study was conducted to determine optimum dose of peanut

sensitization.

Again a major drawback of this model is the fact that they use cholera toxin as an

adjuvant to induce allergy. An adj uvant free-model could be better at getting to the true

sensitization seen in human allergy.

24

2.10 Mouse model of Ovalbumin induced food allergy

A group fi'om the National Taiwan University Hospital headed by Rong-Hwa Lin has
developed a mouse model of Ovalbumin food allergy. Wang et al. [23] for the first time
found that epicutaneous allergen can induce a Th2 response, without the use of adjuvant.
They exposed both C57BL/6J and BALB/c mice to various concentrations of OVA
(100mg/ml, 100pg/ml, and 10pg/ml) by a patch method of applying a patch with the

allergen and securing it with an elastic bandage.

They found that epicutaneous exposure to both strains of mice results in high OVA
specific IgE levels. They saw antibody titers of OVA specific IgE in BALB/c mice
between 0 and 1800 for lpg after 5 courses of immunization. Overall specific IgE levels
in BALB/c mice around 5000 in their 100 pg exposed group. A rather weak IgG2a
response was noted (peak IgG2a titer of around 2000 at 4th response in 100pg exposed
group) from the BALB/c strain with very few animals making even detectable levels.
They did not report on the C5 7BL/6 strain. They report a Th-2 predominant response,
but there was only a 4-5picogram increase in 1L-4 levels following OVA ex vivo
stimulation. Two courses of OVA epicutaneous stimulation lead to a barely detectable

level of INF-y from ex vivo lymph node cell stimulation with OVA.
They found that epicutaneous exposure to both strains of mice results in high OVA

specific IgE levels as well as ex vivo 1L-4 production by lymph node cells. Further very

little IgG2a and NF-y production was seen.

25

Hsieh et a1. 2003 in continuation of the earlier mentioned work confirmed that food
allergy may be through skin sensitization [24]. In this study they sensitized BALB/c
mice epicutaneously through the shaved skin of the back. A patch impregnated with
100pg of ovalbumin was applied for a l-week period and then removed. After three
courses of sensitization, OVA-specific antibodies were measured and then mice were
challenged with 50mg of OVA orally. Anaphylactic responses, plasma histamine levels,
and histology of the intestines and lungs were then preformed. They found that BALB/c
mice elicit OVA-specific IgE when a patch impregnated with OVA is applied to the
shaved dorsal skin. Following oral challenge with allergen, symptoms of systemic
anaphylaxis occurred, plasma histamine increased, and marked changes were seen in both
the intestine (vascular congestion, edema, enterocyte sloughing at villis tips) and lungs
(perivascular and peribronchial inflammatory infiltrates, which consisted of lymphycytes,

monocytes, and eosinophils).

The major strength of this model is the fact that the sensitization phase is adjuvant free.
Because no adjuvant is used therapeutic and prophylactic methods to fight food allergy
can be better studied without the adjuvant effect altering the response (is the therapeutic

effect actually weakening the adjuvant).

A possible weakness of this model is the fact that a purified protein was used and not a

food allergen and Th1/Th2 balance was not assessed. Further analysis of a time course of

26

ex vivo stimulation would give a better representation of the cytokine response in this

model.

2.1] Anti-ulcer drugs promote oral sensitization to hazelnut allergens

The only mouse based model using tree-nut and assessing hypersensitivity, use anti-ulcer
drugs to promote hypersensitivity [25]. In the recent years, parallel with our studies a
group from the Medical University of Vienna developed a protocol to induce immune
responses to oral hazelnut. In there studies, they feed BALB/c mice hazelnut (2mg) with
or with different antiulcer drugs (sucralfate 2mg). They report hazelnut specific IgGl,
but no detectable levels of IgE when mice were orally fed hazelnut extract with a
pretreatment of anti-ulcer drugs. Although oral sensitization is highly sought after,
without adjuvant there is no response and mice develop oral tolerance. They claim that
this treatment did sensitize mice for type I skin reactivity to hazelnut extract, as
evidenced by passive cutaneous anaphylaxis (PCA) reactions using na'r've mice and
hazelnut specific IgGl purified from plasma from the anti-ulcer group of mice. They do
show that the IgGl specific to hazelnut is anaphylactogenic in naive animals when

concentrated, but do not show if there is an in vivo consequence afier challenge.

Conclusions of this study are that the use of anti-ulcer drugs may promote the induction

of immediate type food hypersensitivity towards hazelnut by protecting against gastric

digestion.

27

The strength of this study is that they use the oral route of sensitization to induce
hypersensitivity. Although a novel and important finding, this model does not follow
classical allergy mechanisms, showing allergen specific IgE, therefore it is not a validated
tree-nut allergy model. Furthermore, they do not show an in vivo effector phase after
allergen challenge. Finally, they do not assess mechanisms associated with
hypersensitivity development; cytokine profile assessment would have been interesting if

profiled.

2.12 Gap in the knowledge

Even though extensive research is ongoing about food allergies, mechanisms driving this
increasing trend are unclear at present. Therefore more research needs to be done so that
this increasing trend can be stopped. To do this, well-characterized animal models need

to be developed. Then potential therapies can be developed and studied. Reviewed here
was a list of current animal models available to study food allergy. They include mouse
models for several foods including peanut, cow’s milk and egg. However, a mouse

model to study tree-nut allergy was not available when we began our work in 2001.

With this gap in the knowledge, I sought out to develop and characterize a mouse model
of tree-nut allergy, using hazelnut as a model tree-nut, which might be useful to study
tree-nut allergy. In the next five chapters, efforts were made to characterize a mouse
model of hazelnut allergy with objectives being 1) to develop an ELISA based method to
measure allergen specific IgE as an alternative to the passive cutaneous anaphylaxis
assay; 2) to determine if hazelnut can directly elicit a specific IgE antibody response via

activating IL-4 in mice; 3) to characterize the systemic immune response following

28

transdermal hazelnut protein exposure in BALB/c mice; 4) to develop an adjuvant free
model of hazelnut protein induced systemic anaphylaxis; 5) to study the effect of a diet

rich in EPA and DHA on systemic immune responses to hazelnut.

29

2.13 References

l.

10.

11.

12.

13.

Helm, R.M., R.W. Ermel, and CL. Frick, Nonmurine animal models of food
allergy. Environ Health Perspect, 2003. 111(2): p. 23944

Atkinson, H.A. and K. Miller, Assessment [correction of Asessment] of the brown

Norway rat as a suitable model for the investigation of food allergy. Toxicology,
1994. 91(3): p. 281-8.

Knippels, L.M., et al., Immune-mediated eflects upon oral challenge of
ovalbumin-sensitized Brown Norway rats: further characterization of a rat food
allergy model. Toxicol Appl Pharrnacol, 1999. 156(3): p. 161-9.

Knippels, L.M., et al., Oral sensitization to food proteins: a Brown Norway rat
model. Clin Exp Allergy, 1998. 28(3): p. 368-75.

Pilegaard, K. and C. Madsen, An oral Brown Norway rat model for food allergy:
comparison of age, sex, dosing volume, and allergen preparation. Toxicology,
2004. 196(3): p. 247-57.

Kitagawa, S., et al., Relative allergenicity of cow's milk and cow's milk-based
formulas in an animal model. Am J Med Sci, 1995. 310(5): p. 183-7.

Piacentini, G.L., et al., Allergenicity of a hydrolyzed rice infant formula in a
guinea pig model. Ann Allergy Asthma Irnmunol, 2003. 91(1): p. 61-4.

Poulsen, OM. and J. Hau, Murine passive cutaneous anaphylaxis test (PCA) for

the 'all or none’ determination of allergenicity of bovine whey proteins and
peptides. Clin Allergy, 1987. 17(1): p. 75-83.

Buchanan, RB, et al., Thioredoxin-linked mitigation of allergic responses to
wheat. Proc Natl Acad Sci U S A, 1997. 94(10): p. 5372-7.

del Val, G., et al., Thioredoxin treatment increases digestibility and lowers
allergenicity of milk. J Allergy Clin Irnmunol, 1999. 103(4): p. 690-7.

Ermel, R.W., eta1., The atopic dog: a model for food allergy. Lab Anim Sci,
1997. 47(1): p. 40-9.

Mueller, R.S., S.V. Bettenay, and L. Tideman, Aero-allergens in canine atopic
dermatitis in southeastern Australia based on 1000 intradermal skin tests. Aust
Vet J, 2000. 78(6): p. 392-9.

Paterson, 8., Food hypersensitivity in 20 dogs with skin and gastrointestinal signs.
J Small Anim Pract, 1995. 36(12): p. 529-34.

30

14.

15.

16.

17.

l8.

19.

20.

21.

22.

23.

24.

25.

Sture, G.H., et al., Canine atopic disease: the prevalence of positive intradermal
skin tests at two sites in the north and south of Great Britain. Vet Immunol
Irrnnunopathol, 1995. 44(3-4): p. 293-308.

Buchanan, BB. and CL. Frick, The dog as a model for food allergy. Ann N Y
Acad Sci, 2002. 964: p. 173-83.

Teuber, S.S., et al., The atopic dog as a model of peanut and tree nut food allergy.
J Allergy Clin Irnmunol, 2002. 110(6): p. 921-7.

J effers, J.G., E.K. Meyer, and E.J. Sosis, Responses of dogs with food allergies to
single-ingredient dietary provocation. J Am Vet Med Assoc, 1996. 209(3): p.
608-1 1.

Helm, R.M., Food allergy animal models: an overview. Ann N Y Acad Sci, 2002.
964: p. 139-50.

Phillpis RW, T.M., Swine in Biomedical Research. 1986, New York: Plenum
Press.

Barratt, M.E., P.J. Strachan, and P. Porter, Antibody mechanisms implicated in
digestive disturbances following ingestion of soya protein in calves and piglets.
Clin Exp Irnmunol, 1978. 31(2): p. 305-12.

Li, X., et al., Strain-dependent induction of allergic sensitization caused by
peanut allergen DNA immunization in mice. J Irnmunol, 1999. 162(5): p. 3045-52.

Li, X.M., et al., A murine model of peanut anaphylaxis: T - and B—cell responses to
a major peanut allergen mimic human responses. J Allergy Clin Irnmunol, 2000.
106(1 Pt 1): p. 150-8.

Wang, L.F., et al., Epicutaneous exposure of protein antigen induces a
predominant ThZ-like response with high IgE production in mice. J Irnmunol,
1996. 156(11): p. 4077-82.

Hsieh, K.Y., et al., Epicutaneous exposure to protein antigen and food allergy.
Clin Exp Allergy, 2003. 33(8): p. 1067-75.

Scholl, I., et al., Antiulcer drugs promote oral sensitization and hypersensitivity to
hazelnut allergens in BALB/c mice and humans. Am J Clin Nutr, 2005. 81(1): p.
1 54-60.

31

CHAPTER THREE

3.0 An ELISA based method for food specific IgE antibody measurement in mouse

serum: An alternative to the passive cutaneous anaphylaxis assay1

3.1 Abstract

Background: Passive cutaneous anaphylaxis (PCA) assay has been a gold standard
method to measure allergen specific IgE antibody levels in allergy mouse models. Many
factors including stringent guidelines for laboratory animal use make PCA a difficult
choice. Therefore, alternative methods are needed that can be readily applied for
measurement of specific IgE antibody levels in mouse serum.

AmrL The aim of this study was to develop and optimize an ELISA based system that is
comparable to PCA and can be used to measure specific IgE in mouse plasma or serum.
BM Herein we describe a novel ELISA based method that is more-sensitive in
comparison to PCA, IgE isotype specific (because it has little cross-reactivity with IgGl
or IgGZa isotype) and highly reproducible (<10% inter or intra assay variation).
Furthermore, we demonstrate the utility of this assay to measure specific IgE Ab against
a variety of food extracts including chicken egg, peanut, almond, filbert/hazelnut and
sweet potato.

Conclusions: These findings are of particular interest to those who are seeking (i) to
measure food extract specific IgE antibody in animal models and (ii) an alternative to the
animal based PCA method to measure mouse IgE antibodies.

1 This work was published in J Immunol Methods. 2003 Apr 1;275(1-2):89-98.

32

3.2 Introduction

Allergen specific IgE antibody (ASIgE Ab) production is a central event in the
pathogenesis of atopic disorders that include allergic asthma, -rhinitis, -dermatitis, -
conjunctivitis, food and drug allergies and anaphylaxis [1-3]. Consequently, presence of
elevated levels of specific IgE Ab in the serum is a diagnostic factor for immediate
hypersensitivity response to environmental antigens in humans and animal allergy models
[4]. Therefore, accurate, reliable and user-fiiendly methods are needed to measure serum

levels of allergen specific IgE antibodies.

Passive cutaneous assay (PCA) performed in mice or rats has been widely used to
measure allergen specific IgE antibody levels in animal models for half a century [5-8]
The PCA method has the advantage of measuring not only the biologically active IgE Ab
but also the consequence of allergen/IgE interaction leading to the inflammatory mediator
release from mast cells and the clinical expression of cutaneous anaphylaxis. However,
many factors including stringent guidelines for laboratory animal use, its labor-intensive
nature and the capacity of murine IgGl to trigger PCA responses, limit the utility of this
assay. Therefore, alternative in vitro methods that are inexpensive, easy to perform and

comparable in sensitivity to PCA are needed.

A number of ELISA based methods have been described for measuring IgE Ab specific
to individual purified proteins such as ovalbumin, milk proteins (casein, beta-lacto
globulin), peanut major allergens (Ara hl, Ara h2), or haptens [7, 9-17]. However, we are

not aware of methods available for measuring whole food extract specific IgE antibody in

33

mouse serum. In an effort to fill this need, here we describe an ELISA based method for
food extract specific IgE Ab detection that is comparable in sensitivity to PCA assay.
Furthermore, we demonstrate the utility of this method by performing specific IgE Ab
detection using a variety of food extracts including chicken egg, peanut, ahnond,

filbert/hazelnut and sweet potato.

34

3.3 Materials and methods

3.3.1 Materials

The following materials were purchased from sources as indicated in parenthesis. Food
extracts (Greer Labs, Lenoir, NC, USA); Biotin labeled anti-mouse IgG] (Southern
Biotech, Birmingham, AL), anti-mouse IgG2a (Southern Biotech, Birmingham, AL) and
anti-mouse IgE antibody (Serotec, Raleigh, NC; BD PharMingen, San Deigo, CA, USA);
Ig isotype standards (Southern Biotech, Birmingham, AL); p-nitro-phenyl phosphate
(Sigma, St Louis, MO, USA); Streptavidin alkaline phosphatase (Jackson
IrnmunoResearch, West Grove, PA); Protein-G Sepharose (Pharrnacia Biotech);

Ovalbumin (ICN Bio-medicals, Montreal).

3.3.2 Mice
All mice were purchased from The Jackson Lab (Bar Harbor, Maine, USA). Only adult
female mice (6-7 weeks age) were used in the study. All animal procedures used were in

accordance with the Michigan State University policies.

3.3.3 Animal immunization and serum collection

Various standardized food protein extracts purchased from Greer labs were sterilized by
filtration and total protein content was determined by Lowry’s method. A group of adult
mice (n=5) received intraperitoneal injection of 100 ug of protein extract from peanut,
almond, filbert/hazelnut, walnut and chicken egg plus alum (2.5 mg/mouse) as an
adjuvant. Another group (n=5) received 100 ug of protein extract fiom chicken egg,

wheat, soy, coffee and sweet potato plus alum (2.5 mg/mouse). Control mice (n=3)

35

received sterile saline injections. All rrrice were bled on day 11 after first injection.
Animals received booster injection and bled on days 7, 15, 32 and 62 after the booster

injection. The serum was used in antibody estimations.

3.3.4 Food specific IgE, IgG], IgG2a antibody measurement

Enzyme linked immunosorbent assay (ELISA) was optimized for each of the food
extracts taking into account the background activity (i.e., all reagents added except for the
sample). All reagents were used at a final volume of 50 uL/well except for blocking
buffer that was used at 75 uL/well. Washing was done with 200 uL/well using an
automatic ELISA washer (Dynex Technologies Inc, Ultra-wash Plus). Each food extract
was analyzed for protein content by Lowry’s method and then used in ELISA for coating
at concentrations ranging from 10 to 5000 ug/mL. Briefly, ELISA plates (96 well
EIA/RIA plate, 96 well easy wash TM, high binding, Corning Inc., NY) were coated with
food extracts diluted in carbonate buffer (0.05 M, pH 9.6) and incubated at 4 ° C, over
night. Unbound extract was discarded and the plates were blocked (0.17% BSA/PBS) at
37 ° C. For peanut IgE assay, blocking was performed with 5% gelatin. After washing
(0.05% Tween 20 in PBS), serum samples were added at various two-fold dilutions from
1/20 to 1/640 or in some experiments at 1/30 to 1/61,440 in dilution buffer (0.085% BSA,
0.05% Tween 20 in PBS). Following incubation, plates were washed and a biotin labeled
anti-mouse IgGl or IgGZa or IgE antibody added. After incubation, plates were washed
and streptavidin alkaline phosphatase conjugate was added at 1/4000 (in dilution buffer).
Subsequently, plates were washed and p-nitro phenyl phosphate (PNPP) substrate added

(1 tablet per 5 mL substrate buffer, according to manufacturers instruction; Sigma).

36

Reactions were allowed to develop at the room temperature in the dark and absorbance
measured in a microplate reader using dual mode of wavelength at 405 nm (peak) minus
690 nm (background) (Microplate ELISA Reader, SoftMax program, Molecular
Devices;). According to manufacturer’s instructions, dual mode provides relatively better
measurements since it adjusts the reading for background interference (Personal
communication, Technical Services, Molecular Devices). All plates included negative
controls (no mouse serum sample background and no antigen coating control) and a
positive internal control (A reference mouse serum sample containing known levels of
ovalbumin specific IgE antibody; a kind gift from Prof. Kent HayGlass, The University

of Manitoba). All samples were analyzed a minimum of 2 to 3 times.

3.3.5 Determination of assay sensitivity and serum antibody titer
The assay sensitivity was defined as the background optical density from wells to which
all reagents but no mouse serum sample had been added, + 3 SD. Antibody titer was

defined as the reciprocal of the serum dilution that closely matched this value.

3.3.6 Passive cutaneous anaphylaxis assay

Anti-ovalbumin IgE levels were determined by 48-hr PCA assay in female S-D rats as

previously described [18]. Means of duplicate or triplicate analyses are presented.

37

3.4 Results

3.4.1 Impact of coating antigen concentration on food specific IgE antibody
detection

In order to determine the optimal coating antigen concentration for IgE antibody
detection, initially an ELISA was set-up using various amounts of egg extract. As evident
from the results, a coating antigen concentration of 500 ug/ml yielded maximal
absorbance (OD). (Figure 1A). However, the assay took 24 hr incubation for complete
color development. Furthermore, the OD was off the scale implying off-accuracy and off
precision. We rationalized that increasing the concentration of coating antigen might
reduce the developing time it takes to get a reasonably good assay with a good detection
window. Therefore, we used 2 to 10 fold higher antigen amounts for coating. As evident
(Figure 1B), a combination of higher coating antigen concentration (at 5000 ug/mL
instead of 500 ug/mL) and a lower developing time of 2 hr (instead of 24 hr) provided an
assay with a reasonably good window of detection. Furthermore, the background activity
was not significantly enhanced with higher coating antigen concentrations. Thus, coating
food extract at a protein concentration of 5000 ug/mL and a developing time of 1-3 hours
was used in all subsequent assays. This data suggests that the antigen did not bind to the

microplate well in sufficient quantities except at high concentration.

3.4.2 Immunoglobulin epsilon isotype specificity of the assay
We tested the isotype specificity of the assay by two different ways: (i) first, we
compared the isotype specificity of the detection antibody. In this experiment, we coated

wells with mouse IgE (at 250 ng/mL), IgGl (at 500 ng/mL) and IgG2a (at 500 ng/mL) lg

38

isotype standards or egg and then added serum samples from eg sensitive mice to egg
coated wells and buffer to isotype standard coated wells. All wells were developed with a
biotin conjugated anti-IgE antibody. As evident from the results (Figure 2A), biotin
antibody detected egg specific IgE antibody. Furthermore, biotin antibody reacted with
IgE but not with IgGl or IgG2a isotype standards verifying the epsilon isotype specificity
of the assay; (ii) second, we depleted IgG1 and IgG2a from mouse serum by treating it
with protein-G Sepharose following manufacturer’s instructions (Pharmacia Biotech) and
tested its impact on IgE detection by the assay (Figure 2B). Removal of IgG1 and IgG2a
from mouse serum indeed enhanced the detection window of the assay (i.e., peak signal
minus the background OD) although it had no significant impact on the assay sensitivity.
A control experiment was performed to make sure that protein-G treatment indeed

removed most of the egg specific IgG1 and IgG2a from mouse serum (Figure 3).

3.4.3 Comparison between ELISA and PCA assays

In order to compare the relative sensitivity of ELISA vs. PCA to measure mouse IgE
antibodies, we measured ovalbumin specific IgE Ab titer of serum sample by both
methods. As evident (Figure 4), ELISA titer of the sample (15,360) was ~two titers
higher compared to that of PCA titer (3,900). These data suggest that the ELISA method
described here might be a suitable alternative to PCA method for measuring mouse IgE

antibodies.

3.4.4 Reproducibility of the ELISA method

39

We examined the inter-assay and intra-assay variation of this assay when performed by
two different individuals. As evident from the data, co-efficient of variation in all cases

was <10 % (Figure 5).

3.4.5 Application of the assay to measure food specific IgE antibody levels in mouse
serum

We tested the potential utility of the ELISA method to measure food extract specific IgE
antibody against a variety of food types using food sensitized mouse serum. As evident
in Figure 6 this method was useful to measure specific IgE Ab levels using extracts from
chicken egg, peanut, almond, hazelnut and sweet potato. The average background activity
(OD) for various food types at 2 hour reading was as follows: Egg ~</= 0.3, peanut

extract, </=0.26), filbert/hazelnut </= 0.38), almond </=0.3, and sweet potato </= 0.06).

40

3.5 Discussion

We have described an ELISA based method for measurement of food specific IgE
antibody in mouse serum using a variety of food types. The data presented demonstrates
that the ELISA based method is more sensitive than PCA, isotype specific (because it has
little cross reactivity with IgGl , IgG2a) and highly reproducible (<10% inter-assay and
intra-assay variation). We are not aware of any previously described ELISA methods for

food extract specific IgE antibody detection in mouse serum.

Three methods have been widely used to measure allergen specific IgE antibody in the
serum samples --PCA, RAST and ELISA, although several other methods (e.g., in vitro
histamine release assay, fluorometric reverse (IgE-capture) ELISA, resetting etc) have
been described in the literature [4, 7, 9-15, 19-25]. PCA has several advantages
including: (i) high sensitivity; (ii) high specificity (if challenged 48 hour post
sensitization); and (iii) measurement of biological consequence of allergen-IgE
interaction. However, it suffers from many practical disadvantages such as (i) high cost;
(ii) animal use that is labor intensive; and (ii) concerns about animal welfare. The RAST
is routinely used clinically for measurement of specific IgE Ab in the human serum [3, 4,
24, 26]. Although it is applicable to a mouse system, its utility is limited by the use of
radioisotopes and low level of precision (i.e., 1+, 2+ as readouts). In contrast, this ELISA

method addresses the many shortcomings of PCA and RAST methods.

Currently there are few ELISA based methods available for determination of specific IgE

Ab against purified proteins derived fi'om foods including ovalburrrin (chicken egg

41

derived), milk proteins (casein, beta-lacto globulin) and Ara hl , Ara h2 (peanut derived)
[9, 10, 16, 17]. While these methods are very useful for individual purified proteins, their
suitability for whole food extracts from different sources has not been demonstrated.
Furthermore, we are not aware of previous reports on measuring specific IgE antibody in
mouse serum for either whole extract or purified proteins fi'om almond, hazelnut and

sweet potato.

Some of the specific IgE antibody measuring ELISA methods employ use of protein-G to
remove competing IgGl from the serum [10]. We found that depleting IgG1 and IgG2a
from the mouse serum enhanced detection window of the assay (Figure 3 suggesting
competition between specific IgGl/IgGZa and IgE antibody for common epitopes.

However, the sensitivity of the assay apparently was not affected by this competition.

In the present method we used alkaline phosphatase-dependent substrate (p-nitro- phenyl
phosphate) for color development. It may be possible to further enhance the sensitivity of
this method using alkaline phosphatase dependent fluorescence substrates (e.g., 4-methyl

umbelliferylphosphate).

In the present study, the ELISA method described here was tested for general
applicability using five different food types. We noticed that peanuts consistently
exhibited higher background activity when BSA was used as a blocking agent. In
contrast, use of 5% gelatin as a blocking agent dramatically reduced background and
increased assay sensitivity. We have also found that the method described here is suitable

for a number of other food types (data not shown). We conclude that the method

42

described here may be applied to detect IgE antibodies in mouse serum against a variety

of food types—not necessarily all food types.

In summary, we have described an ELISA based assay that would be very useful to those
who are seeking: (i) to measure food extract specific IgE antibody in mouse models of food
allergy; and (ii) an alternative to animal based PCA method to measure mouse IgE

antibodies.

43

3.6 Figures

CD

Coating antigen concentration

 

 

 

4r
Warrant

<

§ :3

1 .00
50

(mu ass-sow 00) av 36: woods 683

 
  
 

        

 

     
  

  

 

Background

  

I
:6 //

     

uyxz
N
r

  

Z/g

(moose-901v 00) av 36: W 563

44

\ \4. ~.‘ \ \u
u t x \.\. \
...... v---------.--..~~----------v--.-
. O O C

o 9.. o’c‘o’o’.‘o’o‘o‘o‘o‘o‘o’o’o‘o‘o’o‘.’o‘o‘o’o 0‘0 0.0.0.0...0.o.o.¢.o..

. :1, \\.._
.t. ‘.~\.\

‘OUQOQQOVOOOQ'OOOVQ
I...‘...‘......“...‘.‘.‘.‘.‘...‘.-.‘

\ .\-\*\~\\* \\
I \‘\‘\\\ Q\\\
.................

o o o o
'D’Q‘O o 0.0.. o o'c‘o'o’o’o’o‘o’o

 
 
 

   
  
        

    
    

320 640 1280 2560

160

80

Reciprocal serum dilution

40

20

0

10 50 100 500
antigen (egg extract) concentration (ugImL)

33333 .«o Mm -\+§oE Bumbag «an .m 2:me E muouszu "c0322: «a cow: was» 295m 838 08.8 05. doze—6
ow: 3 tom: 33 58 .< oSwm com .96 2:: mane—gov Son N2 8 2V 2.9596 coca 38808 33 8:: 336:8 wwo Eon
85890 530m 05 E n< me 058% $0 93 SEE: ooméc Hombxo E305 wmo me $8325 3033 5:3 @880 053 £33 2V

3322.8 human .8 gush—.359 caucus ”533 be 3:95 Am .331“ 8.53%

45

8:26 528 .8880”. 8.5.... $28 .8298".

 

 

.. .. mm
ssmsmmmmasm mwmmmmmmm
._._i_4-+~_qpo z.__q._..

"" 10240

.2 8.0

  
   
   

9 m. imammfiamw ........
s l .................... mo
............... £03 0380 5E
0529.03 W
l a OOOOOOOOOOOOOOOOOOOOOO
.4 6m goafimmgw
. m
.. O
r O
r P m
m as, 858 8m. .
.. um. 0:03 @8000 m3

 

 

46

(mu ass-em 00) car 36: W8 603

<

.08: n< mmH 3:030 $0 0% 003065

32.2. $80335 8 0202330 .3 mm -1538 500.0080 «:5 {mam me 0500mm wmo 8 0200.33 :05 05 20085 3008.833
0:200:me 900500.358 wEBo=£ emohanom 0-5085 mo €30 $8 55 BEobba we? 00:: 032028-30 80¢ 880m

RC .3955 me omsoEfiSw .23 “0320.60 98 baa £03 3800 mwo 8 52200 A038 Basia—0m $0 805 8200 3 830:8
.35»: 03 a ma Jan: 0% a «83 Jam: 0% a 6% £828 aboe mm 8 subs was as sea a? 838 22» 2?,» 2C

.38: 05 .«0 530502: 25.03 5:30 A»: Ease—ueuafiam “33¢ N66 00.52.,—

47

   

 

 

 

 

d
e
m
m- 08 + - 08
m- 80. w - 82
e
m- 88 m m - 88
P m G
89. m a m - 80..
88 m m - 0000
0000. m - 008.
000% d 1 000mm
803 W 1 000%
0000wa x - 88w.
00008 - 800m.“
885 - 0085
E
A 89.8. B 0003.8
F r - b

E... 80.03 000 2 “on. 8.8% 8m

E: 08.03 000 2 «~00. 058m 8m

Reciprocal of serum dilution

48

Figures 3.6.3 (A-B): Impact of Protein-G treatment on egg specific IgG] and IgG2a
antibody levels in serum from mice sensitized with egg

Wells were coated with the egg extract followed by addition of serum with or without
Protein-G treatment (50% slurry), following manufacturer’s instructions (Pharmacia
Biotech) and developed with biotin labeled ant-mouse IgGl (Figure A) or IgGZa (Figure
B) antibody. Serum sample was identical to the one used in Figure 2. Data represent

mean+/- SE of duplicates or triplicates.

49

 

 

 

 

 

 

 

 

 

A B
5 20000 g ‘
310000 g ,
2 b
‘3 g
'6 8 1:
if, 1000 E E
g '3 Background +3SD
S _ “V
g 100 i ¢ Background
<
5 so" I l l | I l l l l I l l
20 §§§§9§§§§§88
a 8 “2 '~ 3 '-

Rechroealotmmm

PCA (OVA)
ELISA (OVA)

Figures 3.6.4 (A-B): Comparison of assay sensitivity: ELISA vs. PCA

(A) An identical mouse serum sample was subjected to ELISA and PCA and ovalbumin
(OVA) specific IgE antibody titer determined. (B). This figure demonstrates the method
used to deduce ovalbumin specific IgE Ab titer by the ELISA method that is shown in
Figure A. Data from triplicate experiments is presented (mean +/- SE). Arrow indicates

the OVA specific IgE Ab titer.

50

A02
5 a
01
9..
2
m
g
0
20.2
§ 3

 

 

 

L

l
1 I l l 1 LL L l J

 

 

MWMim 7030 3840

 

l.

B

 

 

 

 

mmmweeomauo1meeoaeowuooo so

 

 

 

 

«44030720150013» 3810

in $0 m 210 120 so an
Redprocalofserumdllutlon

51

 

Figure 3.6.5 (A-C): Inter and Intra assay variation of the ELISA method

(A) This Figure illustrates three sets of data obtained by screening an identical sample on
the same ELISA plate (Intra-assay variation). (B, C) Data shown illustrates three sets of
data obtained by screening an identical sample on different ELISA plates on different
days by two different persons (Inter-Assay variation). Data represent mean+/- SE of

triplicates. Arrow indicates the egg specific IgE Ab titer.

52

 

 

2A 20000

1110240
10000;
E ._ ;
.42 .
g 1.- 6 -
‘ n
8 '2’
v < 1000?
3 g I
2% g 1
g Bedtgrounhaso 10°
n.

    

I I'II'I'

 

 

 

 

 

0.2

.
;

Figure 3.6.6: Applications of the ELISA method to measure food extract specific IgE

10240
51 20 f
2560 r
1280 '

640
320 -
1 60 *-
w .—
40 .—
Eon
Peanut
Hazelnut
mend

Ab in mouse serum

(A) Data shown in this figure illustrates the utility of this method to determine peanut
specific IgE antibody titers in the serum of mice sensitized with peanut. Reciprocal of the
serum dilution with OD that closely matched the background OD + 3SD, was defined as
the IgE antibody titer of the serum sample (shown as an arrow); (B) Similar method as
shown in Figure A, was used to deduce IgE titers against other food types using serum

from mice sensitized with various food types. Data represent mean+/- SE.

53

3.7 References

1.

10.

11.

Babu, K.S., S.H. Arshad, and S.T. Holgate, Anti-IgE treatment: an update.
Allergy, 2001. 56(12): p. 1121-8.

Platts-Mills, T.A., The role of immunoglobulin E in allergy and asthma. Am J
Respir Crit Care Med, 2001. 164(8 Pt 2): p. Sl-S.

Wilson, D.R., et al., Increases in allergen-specific IgE in BAL after segmental
allergen challenge in atopic asthmatics. Am J Respir Crit Care Med, 2002.
165(1): p. 22-6.

Perelmutter, L., In vitro allergy testing. J Allergy Clin Immunol, 2001. 107(3): p.
560-1.

Chen, Y.L., F.E. Simons, and Z. Peng, A mouse model of mosquito allergz for
study of antigen-specific IgE and IgG subclass responses, lymphocyte
proliferation, and IL-4 and IFN-gamma production. Int Arch Allergy Immunol,
1998. 116(4): p. 269-77.

Gaboriau-Routhiau, V. and MC. Moreau, Oral tolerance to ovalbumin in mice:
induction and long-term persistence unaflected by Staphylococcus aureus
enterotoxin B and Clostridium perfringens type A enterotoxin. Pediatr Res, 1997.
42(4): p. 503-8.

Gaveriaux, C ., et al., A comparison of five diflerent methods for the detection of
TWP specific mouse IgE: ELISA, ELISA on cells, rosetting, granule enzyme
release assay and passive cutaneous anaphylaxis. J Immunol Methods, 1986.
93(1): p. 107-14.

Gieni, R.S., X. Yang, and KT. HayGlass, Allergen-specific modulation of
cytokine synthesis patterns and IgE responses in vivo with chemically modified
allergen. J Immunol, 1993. 150(1): p. 302-10.

Adel-Patient, K., et al., Evaluation of a high IgE-responder mouse model of
allergy to bovine beta-lactoglobulin (BLG): development of sandwich
immunoassays for total and allergen-specific IgE, IgG] and IgGZa in BLG-
sensitized mice. J Immunol Methods, 2000. 235(1-2): p. 21-32.

Bilenki, L., eta1., Chlamydia trachomatis infection inhibits airway eosinophilic
inflammation induced by ragweed. Clin Immunol, 2002. 102(1): p. 28-36.

Fox, D.A., N. Chiorazzi, and DH. Katz, Hapten specific IgE antibody responses
in mice. V. Differential resistance of IgE and IgG B lymphocytes to X-irradiation.
J Immunol, 1976. 117(5 Pt 1): p. 1622-8.

54

12.

13.

14.

15.

16.

17.

18.

19.

20.

21.

22.

23.

Hall, T.J. and MB. Rittenberg, A novel in vitro assay for mouse IgE. J Immunol
Methods, 1985. 80(2): p. 145-54.

Hanashiro, K., et al., Production of a monoclonal dinitrophenyl-specific rat IgE
and establishment of an IgE capture ELISA for estimating the concentration of rat
IgE antibodies to dinitrophenyl-Ascaris suum. Int Arch Allergy Immunol, 1996.
110(4): p. 371-9.

Harada, S., S. Hosoi, and H. Mikawa, A quantitative enzyme-linked
immunosorbent assay for Dermatophagoides pteronyssinus-specific
immunoglobulin G. J Biochem (Tokyo), 1987. 101(1): p. 95-102.

Hill, RN. and RT. Liu, A sensitive enzyme-linked immunosorbent assay for the
quantitation of antigen-specific murine immunoglobulin E. J Immunol Methods,
1981. 45(1): p. 51-63.

Li, X.M., et al., A murine model of IgE-mediated cow's milk hypersensitivity. J
Allergy Clin Immunol, 1999. 103(2 Pt 1): p. 206-14.

Li, X.M., et al., A murine model of peanut anaphylaxis: T - and B—cell responses to
a major peanut allergen mimic human responses. J Allergy Clin Immunol, 2000.
106(1 Pt 1): p. 150-8.

HayGlass, K.T. and G.H. Strej an, Suppression of the IgE antibody response by
glutaraldehyde-modified ovalbumin: dissociation between loss of antigenic
reactivity and ability to induce suppression. Int Arch Allergy Appl Immunol,
1984. 74(4): p. 332-40.

Bohn, A. and W. Konig, Monoclonal murine anti-DNP IgE: in vitro histamine

release of rat mast cells in the presence of reaginic rat and mouse sera. Agents
Actions, 1985. 16(6): p. 485-90.

Bozelka, B.E., et al., Soy- and shrimp-specific IgE responses in orally and
intraperitoneally immunized mice. Int Arch Allergy Appl Immunol, 1987. 83(3):
p. 271-7.

Cruz, M.J., et al., An amplified ELISA inhibition method for the measurement of
airborne soybean allergens. Int Arch Allergy Immunol, 2000. 122(1): p. 42-8.

Doekes, G., et al., Enzyme immunoassays for total and allergen specific IgE in
population studies. Occup Environ Med, 1996. 53(1): p. 63-70.

Peng, Z., F.E. Simons, and AB. Becker, Measurement of ragweed-specific IgE in

canine serum by use of enzyme-linked immunosorbent assays, containing
polyclonal and monoclonal antibodies. Am J Vet Res, 1993. 54(2): p. 23943.

55

24. Sakaguchi, M., et al., Measurement of antigen-specific mouse IgE by a
fluorometric reverse (IgE-capture) ELISA. J Immunol Methods, 1989. 116(2): p.

181-7.

25. Sedgwick, J .D. and PG. Holt, The ELISA-plaque assay for the detection and
enumeration of antibody-secreting cells. An overview. J Immunol Methods, 1986.

87(1): p. 37-44.

26. Sicherer, S.H., Diagnosis and management of childhood food allergy. Curr Probl
Pediatr, 2001. 31(2): p. 35-57.

56

CHAPTER FOUR

4.0 Hazelnut allergy: Evidence that hazelnut can directly elicit specific IgE antibody

response via activating Type-2 cytokines in mice2

4.1 Abstract

Backggound: Hazelnut is one of the major tree-nuts that potentially causes fatal food
allergy, with underlying mechanisms unclear at present. One suggestion is that hazelnut
allergy results from immune cross-reactivity of IgE antibodies produced against certain
aeroallergens.

Hypothesis, approach: Here we tested the hypothesis that hazelnut is intrinsically
capable of eliciting an allergic response using a mouse model. Groups of mice were
injected i.p. with hazelnut/filbert protein extract with or without alum as an adjuvant and
hazelnut specific antibody (IgE, IgGl) responses were examined using optimized ELISA.
Hazelnut specific Type-2 and Type-1 cytokine responses were evaluated by ex vivo
antigen mediated activation of spleen cells.

3% Hazelnut elicited robust IgE and IgG1 antibody responses. Time-course and
dose response analyses further provided evidence for memory Type-2 dependent
antibody responses to hazelnuts. Hazelnut specific IgE response in two strains of mice
with different MHC haplotypes and IgE response to hazelnut without the use of alum
adjuvant asserted that hazelnut is intrinsically an allergenic food. The Type-2 cytokine
analyses revealed that hazelnut sensitization results fi'om activation of IL-4 and IL-5, thus

providing mechanistic basis for hazelnut specific IgE response.

57

Conclusion: Our data argue that hazelnut—a widely consumed food, is intrinsically an
allergenic food capable of directly eliciting hazelnut binding specific IgE antibodies via

activation of Type-2 cytokines in mice.

2 This work has been published in Int Arch Allergy Immunol.2005 Aug;137(4):295-302.

58

4.2 Introduction

Immediate hypersensitivity to food, commonly called food allergy, afflicts 6-8 %
children and 2% adults in Westernized countries such as the United States[1, 2].
Incidence of these diseases is believed to be increasing consistent with asthma and other
allergic diseases for reasons that are not clear[2-4]. Food allergies typically start in
infancy and some food allergies such as tree-nut allergies tend to be chronic often lasting
for lifetime[2]. Peanuts and tree-nuts (e. g., hazelnuts, almonds etc) are the major food

types that often cause systemic anaphylaxis with fatal or near fatal consequences[1, 2].

Scientific evidence based on double blind placebo controlled oral hazelnut challenge test
clearly confirms that hazelnut can cause immediate hypersensitivity in humans[5-7].
Hazelnut allergy is more common in Europe than in the US. One reason often cited for
this geographic difference is cross-reactivity between birch pollen (and hazel pollen)
allergens and nut proteins[8, 9]. Given this cross-reactivity, it is assumed that many cases
of hazelnut allergy are actually secondary allergies, and that one or more of the
aeroallergens is/are the primary sensitizer. Nevertheless, it is not clear whether this is
always the case, particularly in the US where birch pollen allergy is relatively rare.
Therefore, it is important to know whether hazelnut proteins can act as primary
sensitizers. Thus, it is unclear at present whether hazelnut by itself is capable of directly

eliciting specific IgE antibodies.

In this study we tested the hypothesis that hazelnut by itself is capable of eliciting IgE
response via activating prototypic Type-2 cytokines (IL-4 and IL-5) in a mouse model.

Our results provide the evidence that hazelnut is capable of eliciting both hall-marks of

59

an allergic response—Le, hazelnut binding allergenic (IgE) antibodies as well as Type-2

cytokine response (IL-4 and IL-5) in mice.

60

4.3 Materials and Methods

The following materials were purchased from sources as indicated in the parenthesis:
hazelnut/filbert protein extract (Greer Labs, Lenoir, NC, USA); hazelnut extract was
tested for protein content by Lowry’s method; biotin conjugated Rat anti-mouse IgGl ,
IgG2a and IgE antibody; paired antibodies and recombinant standards for mouse IL-4,
IL-5 and IFN-y (BD PharMingen, San Deigo, CA, USA); p nitro phenyl phosphate
(Sigma, St Louis, MO, USA); Streptavidin alkaline phosphatase (Jackson

IrnmunoResearch, West Grove, PA); and Con-A (Sigma, St Louis, MO, USA).

4.3.] Mice immunization and bleeding

All mice were purchased from The Jackson Lab (Bar Harbor, Maine, USA). Only adult
female mice (6-7 weeks age) were used in the study. All animal procedures used were in
accordance with the Michigan State University policies. A group of adult mice (n=4)
received intraperitoneal (i.p.) injection of 100 pg of hazelnut protein extract plus alum
(2.5 mg) as an adjuvant. Another group (n=4) received 1000 pg of hazelnut protein
extract plus alum. A third control group of mice (n=6) received saline plus alum alone.
All mice were bled three days before injections (pre-immune bleeding), on day 11 afier
first injection (primary response). Animals received booster injections and were bled on
indicated days during secondary and tertiary immune responses. Experiments where

hazelnut alone was used involved 4 mice per group per dose per strain.

Epicutaneous exposure experiments were performed using a modified method described

for ovalbumin[10]. Groups of mice (n=8) were exposed to saline or hazelnut extract

61

(n=12; 500 ug/ 1 00 uL in saline per mouse per application); each mouse was applied with
the reagent on the back skin that had been clipped-off hair and covered with non-latex
bandage for 3 days. Mice were rested for 4 days. Then the cycle of exposure to saline or
hazelnut was continued for six weeks. Weekly blood samples were collected and plasma

was used in the antibody analysis.

4.3.2 Measurement of hazelnut specific IgGl, IgE, IgG2a antibodies in the plasma
We havepreviously described optimization of enzyme linked immunosorbent assay

(ELISA) for food specific IgGl, IgG2a and IgE antibody analyses (chapter 3) [11, 12].

4.3.3 Spleen cell culture and cytokine analyses

Spleen cells were harvested and standard cell cultures were setup previously as described
[13]. Briefly, spleen cells were cultured (7.5 million/mL) in the absence and presence of
hazelnut protein extract (100 and 500 pg/mL). Cell culture in presence of medium alone
served as a negative control and Con-A (2 pg/mL) served as a positive control. Cell culture
supematants were harvested at various time points for use in cytokine analyses using pre-
optimized ultrasensitive assays (Assay sensitivity: IL-4, 0.13 pg/mL; IL-5: 2.8 pg/mL; IFN-

7: 0.72 pg/mL).

4.3.4 Statistical Analysis
Student’s t test was used to evaluate significance using Analyse-ItTM software program

(Analyse-It software Ltd, Leeds, UK). The statistical significance level was set at 0.05.

62

4.4 Results

4.4.1 Characterization of hazelnut specific antibody responses in C57BL/6 mice
Mice were evaluated for the appearance of hazelnut binding Type-2-dependent specific
IgG] and IgE antibodies in the plasma during primary (day 11), secondary (days 7 and 15
after 2nd injection), and tertiary (day 40 after 3'd injection) immune responses. Pre-
immune serum had neither detectable hazelnut specific IgE antibodies (mean titer <40)
nor significant amounts of hazelnut binding IgG] antibodies (mean titer <800).
Following injection with hazelnut extract (100 pg), elevated levels of both IgE and IgG]
antibodies appeared (primary response: dayl 1, mean IgE titer: 160; mean IgGl titer:
1,024,000; secondary response: day 15, mean IgE titer: 452 +/- 160, mean IgGl titer:
32,768,000). We next examined the dose-response and time-course of Type-2-dependent
antibody responses over a three-month period. As evident from the results (Figure 2A)
low doses of hazelnut (100 pg) consistently elicited higher primary as well as higher
memory IgE antibody responses relative to the high dose of 1000 pg per mouse. In
contrast, although there was a memory IgGl response to hazelnut, the intensity of
response was comparable at both doses tested (Figure 2B). These data provide direct
evidence that hazelnut is capable of eliciting Type-2 dependent antibody responses in

mice.

4.4.2 IgE antibody response to hazelnut: Role of host strain, alum adjuvant and
route of exposure
In order to test whether hazelnut is intrinsically an allergenic food, immune response to

hazelnut was examined in two inbred strains of mice with different MHC haplotypes

63

(C57BL/6 H-Zb; and BALB/c H-2d). As evident, hazelnut elicited a robust primary as
well as memory IgE antibody response in both strains although BALB/c exhibited higher
IgE titers (Figure 3A). Hazelnut specific IgG1 and IgG2a antibody titers are shown in
Figure 33, C. Furthermore, we tested whether hazelnut can elicit IgE antibody response
in the absence of alum adjuvant. Results (Table 1) demonstrated that hazelnut in the
absence of adjuvant is capable of eliciting IgE antibody response in BALB/c but not
C57Bl/6 mouse when injected by i.p. route. We then found that exposure to hazelnut in
BALB/c mice by physiologically relevant epicutaneous route also elicits robust IgE

antibody response (Figure 3D) confirming that hazelnut is intrinsically an allergenic food.

4.4.3 Characterization of hazelnut specific Type-2 cytokine responses in mice

In order to study the mechanism of IgE and IgG] response to hazelnuts, we examined
Type-2 cytokine responses in hazelnut-sensitized mice. As evident, hazelnut significantly
activated the prototypic Type-2 cytokine (IL-4 and IL-5) responses in spleen cells from
hazelnut plus alum sensitized C57B1/6 mice (Figure 4A-B). Cytokine response fiom alum
alone control mice shown in Figure 4C-D. Similar responses were also evident in
BALB/c mice sensitized with hazelnut alone in the absence of alum (data not shown).
Hazelnut also elicited a modest IFN-y response in C57Bl/6 model of hazelnut plus alum
sensitization: Culture medium: 51.8 i4; hazelnut 100 pg/mL: 159 $2; hazelnut 500
pg/mL: 251 21:35 pg/mL (12hr); Culture medium: 87 21:6; hazelnut 100 pg/mL: 252 i2;

hazelnut 500 pg/mL: 404 3:23 (48hr).

64

4.5 Discussion

There are three important findings from this study: (i) Hazelnut can directly elicit
hazelnut binding IgE antibodies in mice; (ii) This ability of hazelnut is an intrinsic
property since IgE response was demonstrable in two different strains of mice with
different MHC haplotype; and in the absence of alum adjuvant in BALB/c mice by i.p.
and epicutaneous route; and (iii) hazelnut is capable of activating prototypic Type-2
cytokine (IL-4 and H.-5) responses thus explaining the mechanism underlying the

observed IgE and IgG1 responses.

In this study we chose C57BL/6 and BALB/c mice to examine allergenic immune
response to hazelnuts. This is because these strains have been widely used in the allergy
studies earlier although it was not clear whether they respond to hazelnut with specific

IgE antibodies and Type-2 cytokine responses [12, 14-16].

We chose to examine IgGl , IgE responses to hazelnut as readouts of Type-2 responses in
this study. This is because both IgG1 and IgE isotype of antibody production is
dependent on the prototypic Type-2 cytokine IL-4, although dependence of IgGl is not
complete[17 -19]. Interestingly, hazelnut also elicited Type-1 cytokine (IFN-y) dependent

IgG2a antibody isotype as well.

In the present study we found that hazelnut elicited very high levels of specific IgG1

antibodies with titers in several millions. In mice IgG1 antibodies participate in systemic

65

anaphylaxis [20]. However, at present it remains unclear whether hazelnut specific IgG

antibodies have any disease relevance in humans.

We chose to inject hazelnut to mice because, earlier studies in mice using other allergens
(such as ovalbumin) have shown that i.p. injection is a more reliable and sensitive method
for inducing IgE antibodies compared to oral exposure [21]. Since our objective was to
test whether hazelnut elicits IgE antibodies in mice, we decided to choose the i.p., route
over the oral exposure. Oral sensitization to hazelnut has not succeeded in our hands so
far using very young (day 8 of birth) C5 7B1/6 mice (Birmingham et al. unpublished data).
A very recent study (published subsequent to the submission of our manuscript) reported
that oral administration of hazelnut in BALB/c mice did not lead to sensitization unless
antiulcer drugs were administered [22]. Notably, this study demonstrated that antiulcer
drugs facilitate induction of hazelnut specific IgG1 antibodies with immediate
hypersensitivity skin reactions although IgE antibodies were not detectable. In contrast to
this study, besides i.p., route, using physiologically relevant epicutaneous route of
exposure to hazelnut, we found robust specific IgE antibody response confirming the

intrinsic allergenic nature of hazelnuts.

There is also a previous study examining antibody response to hazelnut in rabbits[23, 24].

They also reported that hazelnut elicited IgG antibodies. However they did not study IgE

or Type-2 cytokine responses to hazelnuts.

66

It is noteworthy that hazelnut is an allergenic food (not an allergen per se) as it represents
a source of allergens some of which have been characterized : Cor a 1 (a Bet v I
homologue)[8]; Cor a 2 profilin[8]; 11S albumin[25]; 2S albumin, lipid transfer protein
and proteins with 47 kDa, 35 kDa and 32 kDa identified as allergens from hazelnut[7].
For some of them it is known that they are present in the pollen as well as in the seed
explaining immune cross-reactivity. It should be noted that results from our study does
not provide evidence against the immune cross-reactivity hypothesis of hazelnut
sensitization. Rather, we have only demonstrated that hazelnut by itself is also capable of

eliciting specific IgE antibody responses in mice.

In summary, data from this study argue that hazelnut is capable of directly eliciting

hazelnut binding allergenic (IgE) antibodies via activating Type-2 cytokine responses in

mice.

67

4.6 Figures

Optical density (405-690 nm)

0.50

0.25

0.00

Figure 1A

 

+ Primary response
+ Secondary response
+ Tertiary response

 

IgE

 

 

 

 

 

  
     

 

 

5120- 1— -1
1260?

640} +- -1
320} 1- —11—— —1

10240” 1 l
2560 -

20480

160 "

Figure 1B

 

3.00

1.50

+ Primary response
+ Secondary response
—*— Tertiary response

  
   

 

lgG1

 

 

 

 

 

 

0.00

           

 

32768000
16384000
8192000 P
4096000 —
2048000
1024000
512000 +
256000 -
128000 r
64000 F
32000 +

Reciprocal plasma dilution

Figure 4.6.1 (A & B): Immune response to hazelnut in C57BL/6 mice.

Groups of mice (n=4) were injected with hazelnut protein extracts plus alum and hazelnut
binding specific IgE (Figure A) and IgG1 (Figure B) antibodies were measured. Primary
response: Antibody levels on day 11 after first injection; Secondary response: antibody
levels on day 7 after the second injection; Tertiary response: antibody levels on day 40

after third injection. Data shown is OD (mean +/- SE). At some points error bars are not

visible.

68

Figure2A

 

 

 

 

 

 

 

 

 

 

 

 

20000
Hazelnut Dose
10000 t T

,2 E + 100 m'crogram hazelnut +Alun
% _ + 1000 m'crogram hazlenut+Ale
.5 l -€-'- 0 nicmgram hazelnut (Alun alone)
.5 IgE A
’; 1000 g
E: E T
E A
< V l

t T T
2.5. _ 41
e l l J
'6
3 100 :
‘0 F
s :
5 -
I)
g .
I .—

Day-3 Day 11 Day 7 Day 15 Day40 Day70
Pre-imnne 1° WSW rm 3%
561111
(Dayzero)1stiriection (Day zero) 2rd irjection (Day m)3rdiriedion

Figure 4.6.2A: Dose-response and time-course analyses of Type-2-dependent
allergenic response to hazelnuts in C57BL/6 mice.

Groups of mice (n=4) were injected with hazelnut protein extracts plus alum and hazelnut
binding specific allergenic antibody, IgE, responses were measured over a three month
period with three injections of hazelnut extracts. Data shown is geometric mean +/- SE

specific antibody titers. Where titers were identical error bars are not shown.

69

Figure 28

 

 

 

 

 

  
    

 

 

 

 

 

 

 

 

 

: + 100 microgram hazelnut + Alum
F + 1000 mlcrogram hazelnut+A|um

’3‘ ”we 5 “A" 0 microgram hazelnut (Alum alone)
' 3 a A
.2 ~ Hazelnut Dose
‘5 191-07 5—
.2 3 I961
" l
3 1000000 g
'5
c .—
< I
3, 100000 :—
0 _
9: C
3 _
:3 10000 g
E l
g _
I 1000 L A= a a

1m P l a l J l l

Day-3 Day 11 Day 7 Day 15 Day 40
Pre-imnne 1° Response 2 Respome 3' Rspome
5311'"
(Day zero) 1st irjection (Day zero) 2rd irjection (Dayzem)3rd itjedion

Figure 4.6.2B: Dose-response and time-course analyses of Type-2-dependent
antigenic response to hazelnuts in C57BL/6 mice.

Groups of mice (n=4) were injected with hazelnut protein extracts plus alum and hazelnut
binding specific antigenic, IgGl , responses were measured over a three-month period
with three injections of hazelnut extracts. Data shown is geometric mean +/- SE specific

antibody titers. Where titers were identical error bars are not shown.

70

wnw + zeH olqleg
wnlv OIQIBB

u"NV + 23H LS'O
wnlv 1.9-0

wnw + zeH ONIBB
wnw omea

wnlv + 22H 19'!)
umlv [.9'0

wnlv + 2914 o/qleg
umlv OIClIBB

“MN 4' 28H 1.9'0
UNIV 190

mm; + 2214 olqlea
“1an Olqlefl

‘11an + 28H [9‘0
“’an [.90

§

 

 

IgE

 

n

 

 

 

 

W wnwmmee

v [,1 wmv OIQIBS
W wnwems-o
v l; wnlv Ls-o
m wnwmmee
v E wan Dimes
W‘ wmv + z... 1.9-0
v E ‘11an LS'O

m wnwmmee

Q wnlv OIQIEG

m wnwewso

l; umrv 1.9-0
E umlv + 22” ones
E wnlv olqleg
[,1 wmv + 29H zs-o
wnlv 1.9-3

 

 

10000”
1000—

8

 

Tertiary

Secondary

Primary

 

 

 

 

Pre-immune

 

 

 

 

 

 

 

Secondary Tertiary

Primary

 

 

Pre-immune

 

 

 

 

L

 

 

5000000

‘—

Jem Apoqnue ougoads inulezeH

71

NE a

- \
a llk\\\\\\\\\\\\\
o
E
(I)
3
o
a vE
a E
:3 .c: ""
.9 3 8
Q.
(U e-I
LU o. g
\
UJ a) —
9 E S v E
<0 <0
(I) I
‘3‘.- V
D co
.. h a
C u
L 1
O
§ § § 8
XXXXXXXXXXX an... mama
Vt: wnlv 010189
M W + an m
v { u”“lV 1.9'3
W wrw + an we
v umlv onuea
umlv + 23H 1.9-0
((3 v { wnlv L90
‘3, [222222 wrw + an om
— vfi wmv ONIBS
[2 wmv + 23H 1.9-0
vi wnlv 1.9-0
v { wnlv + 23H ones
7 v{ wnlv mm
0 i? v{ tuna/names
C v { umlv LS'O

pre-lmmme filthreeponse pre-lnmne fifthresponse

Tertiary

Primary

 

Pro-immune

.5505. go: 0.8 was Hobo

33:82 203 who“: 80:3 doueufimcom 30053350 m0 @603 PE Saw. 380:8 «Emma .w> «Ema—Q wanna—Foam E 28: ~38me
me 0&6on .3 mm -\+ 588 oEoEoow mm 859% Sam dombmcw 203 momcoamom €255 ”3 280%? was QUE 25% Ho ASHE
8532 8 8898 bmsoocfisoao 20>» 8F: Qmém mo 3:80 ADV .EoE 33:5 058% mm -\+ 538 oEonow mm 8505
8an .3855 55052 mo 28:8 “E 8:: Sam was Bowen 8:5on 825 828%»: «N03 98 ~03 .me oEoomm mEuEn 850an
98 one? 83?. 8 :53 min 3858 5805 353a: £3. 380.9: 895 $1.5 oomE Aaméd 05.35 use Pmiv ohmbmo me 3:80

Géo 6932:2— UIE ”5.8.5.. .53 33.5. 3:: 5 mom—898.. .3255“ 958% 2.53:: Anu<v méé Bang

72

Figure 4

 

 

25.00 2500

l
I

18.75 1875

    

  
      
 

f4

        
     
       
     
       
 

9
.0
A.‘

'
9.34
t
o
’0

.v.

...-
‘0
..

            
     
 

   

     

 

   
     

   
 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

12.50 " 1250 " .Q...Q..
,. :ozozozo:
T 0.9.0.98 0.9.0.90
. _ .9990 0990
e 6.25 A 625 9.0.0.0. 9.0,.“
E ‘ E ’o’o’o‘o‘ ’o’o’o’o‘
2 m N 8 m 3’3’3’3 3’3’3’3
‘; 0.00 ~ v 0 , ,
2 CM H2100 5 CM H2100 H2500
E -..
‘E g
g 25.00 3 2500
C
8 C 8 D
l 18.75 ~ 3 1875}
12-50 ” 1250 r
6.25 F —r- 625 ..
CM H21 00 H2500 CM H2100 H2500

Figure 4.6.4 (A-D): Hazelnut specific Type-2 cytokine (IL-4 and IL-5) responses in
mice.

Groups of C57B1/6 mice (n=4) were sensitized with hazelnut (100 ug per mouse) plus
alum or alum alone. Three days following a booster injection with hazelnut or saline,
spleen cells were harvested and cultured with indicated dose of hazelnut or culture
medium alone. Cell culture supematants were harvested at different time points over 3
days and analyzed for cytokines using optimized ELISA. Data shown is peak cytokine
response (average +/- SE of duplicate analyses of data) from hazelnut plus alum
sensitized mice (A, B) and alum alone injected mice (C, D). Student’s t-test: For IL-4:
Hazelnut 100 and 500 ug/mL vs. culture medium alone, P<0.05, Fig. A; For IL-S:
hazelnut 100 and 500 ug/mL vs. culture medium alone; p<0.05, Fig. B; Student’s t-test.
Con-A was used as a positive control (At 24 hr: [L-4: 10.11 +/- 0.42 pg/mL; IL-S: 658.88

+/- 8.13 pg/mL).

73

Table 4.1 Hazelnut specific IgE antibody response in mice without the use of alum

 

 

 

adjuvant"
Mouse Strain Time Point Dose of Hazelnut (pgjper mouse)
0 100 1000
BALB/c Pre- _<_20 5.20 _<_20
(n=4/group/dose) immune“ $20 2560 160
Post***

C—5 7BL/6 Pre-immune _<_20 $20 520
(n=4/group/dose) Post 520 _<_20 $20

 

* Groups of mice were injected i.p., with hazelnut at indicated doses six times over a four
week period and antibody response was analyzed at multiple time points

** Antibody levels in pre-immune plasma

*** Peak antibody levels in the plasma noted following six injections of hazelnut protein
extract is shown; error bars are not shown because antibody titers were identical

74

4.7 References

10.

11.

12.

Bernstein, J .A., et al., Clinical and laboratory investigation of allergy to
genetically modified foods. Environ Health Perspect, 2003. 111(8): p. 1114-21.

Sampson, H.A., Update on food allergy. J Allergy Clin Immunol, 2004. 113(5): p.
805-19; quiz 820.

Kagan, R.S., et al., Prevalence of peanut allergy in primary-school children in
Montreal, Canada. J Allergy Clin Immunol, 2003. 112(6): p. 1223-8.

Sicherer, S.H., A. Munoz-Furlong, and H.A. Sampson, Prevalence of peanut and
tree nut allergy in the United States determined by means of a random digit dial
telephone survey: a 5-year follow-up study. J Allergy Clin Immunol, 2003.
112(6): p. 1203-7.

Hansen, K.S., et al., Roasted hazelnuts--allergenic activity evaluated by double-
blind, placebo-controlled food challenge. Allergy, 2003. 58(2): p. 132-8.

Ortolani, C., et al., Hazelnut allergy: a double-blind, placebo-controlled food
challenge multicenter study. J Allergy Clin Immunol, 2000. 105(3): p. 577-81.

Pastorello, E.A., et al., Identification of hazelnut major allergens in sensitive
patients with positive double-blind, placebo-controlled food challenge results. J
Allergy Clin Immunol, 2002. 109(3): p. 563-70.

Hirschwehr, R., et al., Identification of common allergenic structures in hazel
pollen and hazelnuts: a possible explanation for sensitivity to hazelnuts in patients
allergic to tree pollen. J Allergy Clin Immunol, 1992. 90(6 Pt 1): p. 927-36.

Luttkopf, D., et al., Comparison of four variants of a major allergen in hazelnut
(Corylus avellana) Cor a 1.04 with the major hazel pollen allergen Cor a 1.01.
Mol Immunol, 2002. 38(7): p. 515-25.

Wang, L.F., et al., Epicutaneous exposure of protein antigen induces a
predominant Th2-like response with high IgE production in mice. J Immunol,
1996. 156(11): p. 4077-82.

Birmingham, N., et al., An ELISA-based method for measurement of food-
specific IgE antibody in mouse serum: an alternative to the passive cutaneous
anaphylaxis assay. J Immunol Methods, 2003. 275(1-2): p. 89-98.

Birmingham, N., S. Thanesvorakul, and V. Gangur, Relative immunogenicity of
commonly allergenic foods versus rarely allergenic and nonallergenic foods in
mice. J Food Prot, 2002. 65(12): p. 1988-91.

75

13.

14.

15.

16.

17.

18.

19.

20.

21.

22.

23.

24.

Rempel, J .D., I.P. Lewkowich, and K.T. HayGlass, Endogenous IL-12 synthesis
is not required to prevent hyperexpression of type 2 cytokine and antibody
responses. Eur J Immunol, 2000. 30(2): p. 347-55.

Kodama, T., et al., Role of interleukin-12 in the regulation of CD4+ T cell
apoptosis in a mouse model of asthma. Clin Exp Immunol, 2003. 131(2): p. 199-
205.

Tsuchiya, T., et al., Airway remodeling of murine chronic antigen exposure
model. J Asthma, 2003. 40(8): p. 935-44.

Yang, X. and KT. HayGlass, IFN-gamma, but not IL-4, synthesis by antigen-
primed murine T cells is IL-2 dependent. Differential requirements in de novo and
established responses. J Immunol, 1993. 150(10): p. 4354-63.

Finotto, S. and L. Glimcher, T cell directives for transcriptional regulation in
asthma. Springer Semin Irnmunopathol, 2004. 25(3-4): p. 281-94.

Romagnani, S., Immunologic influences on allergy and the TH1/TH2 balance. J
Allergy Clin Immunol, 2004. 113(3): p. 395-400.

Rothenberg, M.E., Eosinophilic gastrointestinal disorders (EGID). J Allergy Clin
Immunol, 2004. 113(1): p. 11-28; quiz 29.

Oettgen, H.C., et al., Active anaphylaxis in IgE-deficient mice. Nature, 1994.
370(6488): p. 367-70.

Atherton, K.T., R.J. Dearrnan, and I. Kimber, Protein allergenicity in mice: a
potential approach for hazard identification. Ann N Y Acad Sci, 2002. 964: p.
1 63-71 .

Scholl, I., et al., Antiulcer drugs promote oral sensitization and hypersensitivity to
hazelnut allergens in BALB/c mice and humans. Am J Clin Nutr, 2005. 81(1): p.
1 54-60.

Koppelman, S.J., eta1., Comparison of different immunochemical methods for the
detectionand quantification of hazelnut proteins in food products. J Immunol
Methods, 1999. 229(1-2): p. 107-20.

Scheibe, B., et al., Detection of trace amounts of hidden allergens: hazelnut and
almond proteins in chocolate. J Chromatogr B Biomed Sci Appl, 2001. 756(1-2):
p. 229-37.

76

25. Beyer, K., et al., Identification of an 11S globulin as a major hazelnut food
allergen in hazelnut-induced systemic reactions. J Allergy Clin Immunol, 2002.
110(3): p. 517-23.

77

CHAPTER FIVE

5.0 Characterization of the systemic immune response following transdermal

hazelnut protein exposure in BALB/c mice

5.1 Abstract

Background: Previously we found that in BALB/c mice, transdermal exposure to
hazelnut was capable of eliciting hazelnut specific (IgE) antibodies without the use of
adjuvant.

Hypothesis, obLectives, and approach: In this study we tested the hypothesis that
transdermal application of hazelnut protein activates Type-2 cytokines, thus promoting
hazelnut specific IgE formation. Groups of BALB/c mice were exposed to 4 different
doses of hazelnut protein transdermally over a six-week period. Hazelnut specific
antibody responses (IgE, IgGZa) were studied. Short-term antigen driven recall cytokine
responses were assessed using spleen cells ex vivo.

Regalia We confirmed our earlier finding, that exposure of BALB/c mice to LPS-free
hazelnut—a model tree nut, without the use of adjuvant, elicits a robust systemic IgE
response; the IgE response was dose-dependent and increased with boosting. Hazelnut
specific IgGZa was also detectable and increased with boosting. A marked hazelnut
protein driven recall IL-4 response by spleenocytes was found. Surprisingly, there was
also evidence for a modest induction of IgGZa and IFN-y responses.

Conclusion: Transdermal exposure to hazelnut protein without the use of adjuvant results

in activation of IL-4 response leading to IgE antibody response.

78

5.2 Introduction

Mechanisms underlying food allergy sensitization are not completely understood. A
major question still yet to be answered in food allergy is how are people sensitized to
food allergens? Is the gastrointestinal tract the site of primary sensitization, or are other
sites seeing allergen first? Since people as well as rodents usually develop oral tolerance

to ingested proteins [1], what is causing this breakdown in oral tolerance, or is there one?

A recent study found a significant relationship of peanut allergy with the use of skin
preparations for the treatment of diaper rashes, eczema, dry skin, and IFNlammatory skin
conditions in IFNancy containing peanut oil (odds ratio, 6.8; 95 percent confidence
interval, 1.4 to 32.9) [2]. Previous reports have shown that epicutaneous application of
OVA can lead to a Th2 type of response with subsequent OVA-specific IgE [3]. With
the skin being a possible being a new, emerging route of exposure and our previous data
showing BALB/c mice can make IgE without adjuvant by high number of i.p. injections,

we sought out to profile the immune response to transdermal hazelnut exposure.

In this study we tested the hypothesis that transdermal application of hazelnut protein

activates IL-4 cytokine, thus promoting hazelnut specific IgE formation.

79

5.3 Materials and Methods

5.3.1 Materials

The following materials were purchased from sources as indicated in the parenthesis.
Hazelnut/filbert protein extract (Greer Labs, Lenoir, NC, USA); hazelnut extract was
tested for protein content by Lowry’s method; LPS content of hazelnut protein extract
was measure by LAL assay (Cambrex Bio Science Walkersville, Inc., Walkersville, MD,
USA); Biotin conjugated Rat anti-mouse IgE antibodies; paired antibodies and
recombinant standards for mouse IL-4 and IFN-y (BD PharMingen, San Diego, CA,
USA); p nitro phenyl phosphate (Sigma, St Louis, MO, USA); Streptavidin alkaline
phosphatase (Jackson IrnmunoResearch, West Grove, PA); PMA, inomycin (Sigma, St

Louis, MO, USA).

5.3.2 Transdermal sensitization and bleeding

All mice were purchased from The Jackson Lab (Bar Harbor, Maine, USA). Only female
mice (6-8 weeks age) were used in the study. All animal procedures used were in
accordance with the Michigan State University policies. Transdermal exposure
experiments were performed using the method described by us before (Chapter 4) [5].

The experimental protocol for the sensitization process is shown in Figure 5.1.

5.3.3 Measurement of hazelnut specific IgE antibodies and total IgE in the plasma
We have previously described optimization of enzyme linked immunosorbent assay

(ELISA) for food specific IgE antibody analyses (Chapter 3) [6, 7].

80

5.3.4 Spleen cell culture and cytokine analyses

Spleen cells were harvested and standard cell cultures were setup essentially as described

(Chapter4) [5, 8].

5.3.5 Statistical Analysis
Student’s t-test and ANOVA tests were used to evaluate significance using Analyse-ItTM
software program (Analyse-It software Ltd, Leeds, UK). The statistical significance level

was set at 0.05.

81

5.4 Results

5.4.1 Characterization of systemic hazelnut specific antibody responses following
transdermal exposure to hazelnut protein in BALB/c mice

We used LPS-free preparations of hazelnut protein extract (Greer Labs) in this study LPS
level as measured by LAL assay was at <0.4 pg/mg of protein. Mice were evaluated for
the appearance of hazelnut binding IgE and IgG2a antibodies in the plasma before and
after each transdermal exposure. Pre-immune serum had less than detectable hazelnut
specific IgE antibodies (mean titer <1/20) as well as less than detectable levels of
hazelnut specific IgG2a (mean titer <1/250). Following transdermal exposure with
hazelnut (500 jig/mouse) but not saline, elevated levels IgE antibodies appeared after the
third exposure (Figure 5.2). Measurable levels of hazelnut specific IgG2a were seen after
the third exposure, but extremely modest until the sixth response (Figure 5.3). We next
examined the dose-response and time-course of IgE antibody responses during
transdermal exposure over a six-week period. As evident, a dose of 5 ug/mouse did not
elicit detectable hazelnut specific IgE response, but IgE response was clearly evident with
the 50 and 500 ug/mouse dose (Figure 5.4A). Further analysis of antibody titers revealed
progressively higher memory IgE antibody responses (response from activation with the
same allergen at a later time point). Furthermore, analysis of total serum IgE levels
further confirmed elevated, dose and time dependent induction of allergic antibody

response (Figure 5.4B).

5.4.2 Characterization of hazelnut driven recall cytokine (IL-4 and IFN-y) responses

following transdermal exposure to hazelnut in BALB/c mice

82

In order to study the mechanism of IgE response to hazelnuts, we examined both Type-2
cytokine (IL-4) responses and Type-1 cytokine (IFN-y) responses in hazelnut-sensitized
mice. As evident, hazelnut activated a Type—2 cytokine (E-4) response in spleen cells
isolated from hazelnut but not saline exposed mice (Figure 5.5). IL-4 response was dose

dependent with spleen cells isolated from hazelnut-sensitized mice responding to 500 pg

of hazelnut stronger than to 100 pg hazelnut (p<0.05)(Figure 5.5), whereas in spleen cells

isolated from control mice no dose response was noted.

There was a background level of IFN-y being produced by saline sensitized mice spleen
cells exposed to hazelnut (Figure 5.6). Further analysis of the cytokine profile was done
with determination of the IFN-y/IL-4 ratio for the hazelnut sensitized mice. There was a
push towards IL-4 in animals that were sensitized to hazelnut thus providing a

mechanism to favor IgE antibody development (Figure 5.7).

83

5.5 Discussion

In this study we profiled hallmarks of allergenic immune responses (specific and total
IgE, IL-4, and IFN-y / IL-4 ratio) to transdermal hazelnut protein in BALB/c mice. There
are several novel finding of these studies; i) Repeated transdermal exposure of BALB/c
mice to hazelnut (in the absence of adjuvant) can activate systemic antibody response
characterized by robust Type-2 dependent isotopes (IgE); ii) Transdermal exposure to
hazelnut activates memory Type-2 cytokine (IL-4) responses; iii) Transdermal exposure
to hazelnut alters the IFN-y / IL—4 ratio in favor of IL—4, thus promoting Type-2 antibody

development.

In a recent study Lack et a1. [2] used data from the Avon Longitudinal Study of Parents
and Children, a geographically defined cohort study of 13,971 preschool children, to first
identify children with history of peanut allergy and then find association within them.
From this, they found a significant relationship between peanut allergy and the use of
skin emollients (creams) containing peanut oil for the treatment of diaper rashes, eczema,
dry skin, and IFNlammatory skin conditions in IFNancy (odds ratio, 6.8; 95 percent
confidence interval, 1.4 to 32.9 [2]. With the skin possible being a new, emerging route
of exposure, our adjuvant free model of sensitization may be more physiologically

relevant than it would have been in the past.

Wang et a1. [3] for the first time found that epicutaneous allergen can induce a Th2

response, without the use of allergen. They exposed both C57BL/6J and BALB/c mice to

various concentrations of OVA (10mg, 100ug and lug) by a patch method of applying a

84

patch with the allergen and securing it with an elastic bandage. They found that
epicutaneous exposure to both strains of mice results in high OVA specific IgE levels.
We found our low dose of hazelnut (5 pg) in BALB/c failed to elicit a hazelnut specific
IgE titer (levels of <10) at our low dose of hazelnut), whereas they had antibody titers of
OVA specific IgE in BALB/c mice between 0 and 1800 for 1 pg after 5 courses of
immunization. Overall specific IgE levels in BALB/c mice were similar between our
study and theirs, with our peak hazelnut antibody titer being 5120 at the filth response in
the 500 pg exposed group, whereas they saw a peak IgE antibody titer of around 5000 in

their 100 pg exposed group.

In the OVA epicutaneous model a rather weak IgG2a response was noted (peak IgG2a
titer of around 2000 at 4th response in 100pg exposed group) from the BALB/c strain
with very few animals making even detectable levels. They did not report on the
C57BL/6 strain. Similarly, our model had a rather modest IgG2a response, with a peak

IgG2a titer of 8000 at 6th response in 500pg exposed BALB/c group.

The OVA epicutaneous model, they report a Th-2 predominant response, but there was
only a 4-5picogram increase in IL-4 levels following OVA ex vivo stimulation. Our
transdermal model shows a robust IL-4 response following ex vivo stimulation (around a
280picogram increase following hazelnut ex vivo stimulation). One reason for this
difference could be due to the cell type, spleen cells in our model vs. lymph node cells in
the OVA model. Also in the OVA model cells were collected only 24 hours after

stimulation, whereas our culture is done over 5 days and peak cytokine response is

85

reported. IFN-y responses in both models were assessed. Two courses of OVA
epicutaneous stimulation lead to a barely detectable level of IFN-y from ex vivo lymph
node cell stimulation with OVA, whereas spleen cell stimulation with hazelnut following
six courses of transdermal hazelnut exposure gave a more modest hazelnut specific IFN-y
level of around 500picograms. Differences here also could be due to cell type used or a
lack of time course analysis in the OVA model in both the collection of cell culture

supematants and when cells were harvested for culture (only after 2 immunizations).

Although groundbreaking work, showing that epicutaneous allergen can sensitize for a
Th2 cytokine (IL-4) and IgE response, further analysis of IL-4 to IFN-y ratio would have
been beneficial to assess the effect of the doses of OVA and cytokine profile leading to

allergy.

In summary, we found that transdermal sensitization of mice without the use of adjuvant
results in activation of hazelnut driven IL-4 response leading to IgE antibody response.
Surprisingly, there was also evidence for a modest induction of IgGZa and IFN-y

I'CSpOl'lSCS.

86

$6325
050on 5532 05.0.on

.m_o>o_ 05.25 9:89:
.9330 :00 cooam 05.3 .o
Scone; 5S, 2955 “moon

 

$5: m .8905
2x00? on.

mom 3
mxmoz, m © 02.5 was

:2 2.0 B8505

 

 

 

 

 

 

gun N 66m \ \ _
T \\ filo; P
>3 P gnu v gun N
82m :0 cofiaEEm 2 95mm

8 350ch >394
/ to 52mm oxmk
.8205 >_xoo>>

B 9:3

8.3m:— cna 83am. Wm

87

.565 2 833233 Sign: m0 83.3 no 2.60:8 2:. ”—8395 33.3338 _aEuocmufih "fim 0.53m

88

c0326 mEmmE

owN : —. 99 r own: 00 t v 0Q _. 0v;

 

 

 

 

 

 

 

 

on:
4 . ._, a. ._. u. . o
L. H
H .r L
._.
E .N
02039 Em lfil
omcoamopEm IrmT.
oczEEToi |®Ii m
m

 

 

$0.5 SCENE;

com 2 r 05:. ONE V cm I —. 0Q _.

03—. ON:

 

 

q _ a _ _

TTTTTI

3:032 50 Irwl
8530.. En lmTl
mczEETQQ Il®|

d fl

 

 

 

<

 

 

 

fins 95%

mm 9:9“.

m

(wu 069-9017 00) ewseld U! IeAeI av 35: amoeds inUIezeH

89

.0363
8: 000 was 8&0 $58 >52: 2 .AMm -\+ 9025 DO 00 £0>0~ 090me 80935 0E00Qm 55000: 86% Sen 8.5898 @8003qu 0
00% 0080:00 08003 ”088000 so $03890 3500880 m 00:0 880:8 08003 ”0800000 Em “0.5898 0080p 008230 0883
”0885705 .<m5m 008508 3 0050008 0003 86335 E .< .mev me 0500mm mEvEn 3508: was 0:29. .8 $0338 £088
“:50an 0009.9: .8 080831 cow 8 00898 30:003qu 0003 asepw\o-mnqv 00:: .8 8:80 4:508: 3 230:8 38.5.0895

000003.. ”5.50:8 00:: 05445 E 008030.. .325: ”mu— 05003 2.505:— ..0 8305—80225 ”EJJ N.m 0.5!...“

90

coco—.2. Doom: ooov: ooow:

3:26 058.0

00°F: com: 0mm:

 

 

 

 

 

020000. 50 lLfll
00:030. Em Imll
0388705 .mell

 

 

 

 

m

 

 

 

 

Aoncv 35050:

O

ooomw:

Doom: ooov: ooom: coo—.2.

com:

0mm:

 

 

 

00.00000. 50 Ilflll
00:030. Em Iml
0.388705 |¢||

 

 

 

 

I
(wu use-gov 00) ewseld U! IeAeI qv ezsfil omoeds InUIGZBH

N

 

Amncvmczmw

00 059“.

91

.0300? 00: 000 0000 00000
00800 808 2 .35 -\+ 508v no 00 £05. 005000 8000000 00.000000 005000: 03000 00D 0000000000 00800000000 0 0000 088:8
08003 ”00000000 50 0000000900 00800000000 m 00000 006230 08003 ”00000000 00m ”00000900 000.000 00000200 0800—0 ”0888_
-000 {mim 00008000 3 00000008 0003 000000080 Am .< 03% 0.0.03 00.000000 @0089 008000: 000 00:00 00 0000.800 800000
005000: 0000003 .00 0.000000%: com 00 0000900 b0800000000 0003 300030-?5 0080 .00 000000 000.000: 00 0000098 .080000005

0000000.. ”830:8 088 051:5 8 00000000.. 0:000:80 30»— 5000500 050090 28.000: ._0 00600200000000 "Am—-0; Wm 00:»?—

92

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

—e— Hazelnut (5 ug / mouse n=4) M Hazelnut (5 ug I mouse n=4)
—+— Hazelnut (50 ug / mouse n=4) / Hazelnut (50 ug I mouse n=4)
10000—V— Hazelnut (500 ug l mouse n=4) 50- Hazelnut (500 ug / mouse n=4)
C A -
.03
a E
o D)
g 1000 :- 3
c : 0
m T)
Lu >
2’ .9
UJ
.3 9
o
a) (U
3 100 g g
*5 : a
E _
g S
m C
I l—
10 : 3 % fl
5 : l l l l I J

 

 

 

Pre 1R 2R 3R 4R 5R
Timepoint of analysis Timepoint of analysis

Figure 5.4 (A & B): Dose-response and time-course analyses of IgE antibody
response to transdermal hazelnut exposure in BALB/c mice. Groups of mice
(n=4/group) were transdermally exposed to indicated dose of hazelnut protein extract (0,
5, 50 or 500 pg/mouse) and hazelnut binding specific IgE antibody titer (Fig. A; Data
shown is geometric mean +/- SE) and total plasma IgE (Fig. B) levels were measured
over a six week period. Where titers were identical error bars are not shown. AVOVA
test results: Panel B, a= p<0.00l vs. hazelnut 0 group at same time point; b= p<0.001 vs.

all other hazelnut doses at the same time point.

93

Figure 5.5

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Hazelnut (n=5) Saline (n=4)
400 A l 400 e 1
300 - 300 -
_l—

E
8’
V 200 - 200 L
Y
=1

100 - 100 r

0 L 0 m rm m1
0 100 500 0 100 500
Hazelnut concentration (ug/ml) Hazelnut concentration (ug/ml)

Figure 5.5 (A-B): Hazelnut driven T ype—2 cytokine (IL-4) responses in mice
transdermally sensitized with hazelnut or saline.

Groups of BALB/c mice (n=4-5/group) were sensitized with hazelnut (500 pg per mouse)
(Figs. A) or saline (Figs. B) by transdermal exposure as described in Figure 5.1. Three
days following a booster exposure with hazelnut or saline, spleen cells were harvested
and cultured with indicated dose of hazelnut protein or culture medium alone. Cell
culture supematants were harvested at different time points over 5 days and analyzed for
cytokines using optimized ELISA. Data shown is peak cytokine response (average +/- SE
of duplicate analyses of data) from hazelnut-sensitized mice (A) and saline exposed mice
(B). ANOVA test: graph A; for all comparisons between 100 and 0 and 500 and 0:

p<0.0l).

94

Figure 5.6
Hazelnut (n=5)

 

 

 

 

 

 

 

5000 A I
E
.3}
,r 2500”
2 a,b
E
9‘339‘9‘:
0
0 100

Hazelnut concentration (lg/ml)

Figure 5.6 (A-B): Hazelnut driven Type-1 cytokine (IFN-y) in mice transdermally

sensitized with hazelnut or saline.

Groups of BALB/c mice (n=4-5/group) were sensitized with hazelnut (500 pg per mouse)
(Figs. A) or saline (Figs. B) by transdermal exposure as described in Figure 5.1. Three

days following a booster exposure with hazelnut or saline, spleen cells were harvested

 

 

 

 

 

 

 

 

500‘ Saline (n=4)
714
2500"
a,b
’ a,b
0
O 100 500

Hazelnut concentration (lg/ml)

and cultured with indicated dose of hazelnut protein or culture medium alone. Cell

culture supematants were harvested at different time points over 5 days and analyzed for
cytokines using optimized ELISA. Data shown is Day 3 of cell culture (average +/- SE of
duplicate analyses of data) from hazelnut-sensitized mice (A) and saline exposed mice

(B). ANOVA test: Graph A and B, a=p<0.01 vs. 0 hazelnut group at same time point;

b=p<0.01 vs. same dose of hazelnut in other panel.

95

 

Figure 5.7

 

 

Hazelnut sensitized mice (n=5)

 

 

 

 

 

 

 

 

 

i.)
A-

V v v '

o’o’o

raga
q;

‘%

 

 

egooooooooo

 

60 O

I O'O’OOOOOOOOOOO
909099900000

0

O

O

O
O
.0

.9

.0

O
O

O
.9

O
9

99000
saves
save»

oqgg,
‘%%004

 

l

 

 

4O

30

O
N

(0:191) v-‘II / l—dNI

96

1O

Saline I 05 Hazelnut 100 ug I 05 Hazelnut 500 ug I 05

Hazelnut 100 ug I 03 Hazelnut 500 ug I 03

Saline I 03

Stimulation and Timepoint of anaylsis

.3980 25% >3 oEmm m: .m> c2638? 3533 w; cow was o2

Eon 8m modva 6:32 “we“ <>OZ< .933 mo mowing 83:96 me mm -\+ owfioév 832353 55053 mi cow was m1 2: .25%
com v-1: no we \ S23 mo mm 98 8:88.. wk: .8 ma 1.7520 mm mm 56% Sam .<m3m 338:3 mam: moaflofio .8.“ 38:28
98 mag m ~96 $53 08: «cooobfi um 863.8: 803 358533 05:3 :00 .AoE—mmv 2.83 8:62: 83:6 .8 £88m 3532
.«0 08¢ @835: 53, 3.523 23 353.8: Be? 230 52% 455053 5:5 cosmomxo 85003 a mama/0:8 93 ooh; Ad oSwE
E wontomow mm 2:898 _wgocmgb .3 8308 con w: 83 5532 £15 @3528 203 anewknav our: 8QO mo $580

.3205:— 5? 33:38 bfifiuoumunb 3:: 5 3:2 v.1: \ >.ZE “Em 9:33

97

5.7 References

l. Strobel, S., Immunity induced after a feed of antigen during early life: oral
tolerance v. sensitisation. Proc Nutr Soc, 2001. 60(4): p. 437-42.

2. Lack, G., et al., Factors associated with the development of peanut allergy in
childhood. N Engl J Med, 2003. 348(11): p. 977-85.

3. Wang, L.F., et al., Epicutaneous exposure of protein antigen induces a
predominant ThZ—like response with high IgE production in mice. J Immunol,
1996. 156(11): p. 4077-82.

4. Hsieh, K.Y., et al., Epicutaneous exposure to protein antigen and food allergy.
Clin Exp Allergy, 2003. 33(8): p. 1067-75.

5. Birmingham, N., et al., Hazelnut Allergy: evidence that hazelnut can directly
elicit specific IgE antibody response via activating type 2 cytokines in mice. Int
Arch Allergy Immunol, 2005. 137(4): p. 295-302.

6. Birmingham, N., S. Thanesvorakul, and V. Gangur, Relative immunogenicity of

commonly allergenic foods versus rarely allergenic and nonallergenic foods in
mice. J Food Prot, 2002. 65(12): p. 1988-91.

7. Birmingham, N., et al., An ELISA-based method for measurement of food-specific

IgE antibody in mouse serum: an alternative to the passive cutaneous anaphylaxis
assay. J Immunol Methods, 2003. 275(1-2): p. 89-98.

8. Rempel, J .D., I.P. Lewkowich, and KT. HayGlass, Endogenous IL-12 synthesis is

not required to prevent hyperexpression of type 2 cytokine and antibody
responses. Eur J Immunol, 2000. 30(2): p. 347-55.

98

CHAPTER SIX

6.0 Development of an adjuvant-free model of tree-nut protein induced systemic

anaphylaxis

6.1 Abstract

Background: Our previous studies indicated that transdermal application of LPS free

 

hazelnut in the absence of adj uvant can lead to the development of hazelnut specific
allergic (IgE) antibodies via activating H_.-4 cytokine. Food allergy is not only having
hallmarks of allergy, but also having an adverse response after challenge (disease). A
mouse model of tree nut induced systemic anaphylaxis, the focus of this study, was
unavailable.

Hypothesis, obiectives, and approach: Here we tested the hypothesis whether
transdermal application of hazelnut can prime BALB/c mice for systemic anaphylaxis
when challenged either systemically (i.p.) or orally. Further testing to assess histamine
release after challenge (i.p.) and pathological changes in the gastrointestinal tract were
also done.

Result; Repeated transdermal exposure of BALB/c mice to hazelnut, again, elicited a
robust systemic IgE Ab response. Transdermal sensitization also primed for systemic
anaphylaxis as evidenced by dose-dependent clinical scores, marked drop in rectal
temperature and elevated histamine levels with 0.1 mg/mouse being the threshold dose
when challenged by i.p. injection. Following oral challenge (13 mg/mouse) with

hazelnut, sensitized mice showed several signs of systemic anaphylaxis (diarrhea,

99

hypothermia, convulsions). Tissues from the gastrointestinal tract were collected and
fixed with 10% formalin in PBS and stained with hematoxylin-eosin. In hazelnut-
sensitized mice that underwent systemic anaphylaxis, the small intestine had marked
edema, vascular congestion and enterocyte sloughing, whereas in saline control mice, no
pathological lesions were noted.

Conclusion: We have developed an adjuvant-free mouse model of tree nut induced
systemic anaphylaxis that should serve as a useful tool not only to elucidate further
mechanisms of this disease but also to develop improved therapeutic and/or prophylactic

methods.

100

6.2 Introduction

In the past several years, a number of interesting mouse models «often using adjuvants
such as cholera toxin in the experimental protocols, to study food allergies have been
reported. These include peanut, egg and milk allergies [1-3]. Unlike human food allergy,
these models used cholera toxin to break down oral tolerance and achieve their food
allergy state. After sensitization, animals are challenged with intra-gastric allergen and

systemic anaphylaxis occurs.

Previously, we found that transdermal application of LPS free hazelnut in the absence of
adjuvant can lead to the development of hazelnut specific allergic (IgE) antibodies via
activating Type-2 cytokines. However, a mouse model of tree nut induced systemic
anaphylaxis with concurrent pathology was not available. Therefore, we here tested the
hypothesis that transdermal sensitization of BALB/c mice to hazelnut primes for systemic

anaphylaxis.

In this study we report on systemic anaphylaxis and pathological changes seen in an
adjuvant free model of tree nut allergy. Following transdermal sensitization mice
undergo systemic anaphylaxis reactions following i.p. or oral challenge with allergen.
With the current study, pathological changes were noted in the small intestine after oral
challenge. In summary, we have developed and characterized an adjuvant—free mouse
model of tree nut induced systemic anaphylaxis that should serve as a useful tool for

further mechanistic and applied studies.

101

6.3 Materials and Methods
6.3.] Materials

Materials were used as previously described in chapters 4 and 5.

6.3.2 Transdermal sensitization, bleeding, and challenge

Sensitization procedure was done as previously described in chapter 4. The experimental
protocol for the sensitization and challenge process is shown in Figure 6.1. Tables 6.1
and 6.2 show the number of animals in each sensitization group. Following challenge
animals were bleed for plasma histamine determination, observed for clinical scores of

systemic anaphylactic responses and had rectal temperature taken.

Table 6.1 I.P. challenge: study design

 

 

 

 

 

Group ID Number of mice Sensitized Challenged i.p.
transdermally with with
1 5 Saline Saline
2 5 Saline Hazelnut
3 5 Hazelnut Saline
4 5 Hazelnut Hazelnut

 

 

 

 

 

 

Table 6.2 Oral gavage challenge: study design

 

 

 

 

 

Group ID Number of mice Sensitized Challenged orally
transdermally with with
1 1 6 Saline Saline
2 1 6 Saline Hazelnut
3 1 6 Hazelnut Saline
4 16 Hazelnut Hazelnut

 

 

 

 

 

 

102

6.3.3 Induction of systemic anaphylaxis, clinical scoring, measurement of rectal
temperature

Groups of hazelnut sensitized vs. saline exposed mice challenged i.p. or orally with
indicated amount of hazelnut protein or saline. Mice were then observed for signs of
systemic anaphylaxis and video-recorded for symptoms during the next 4 hours. Clinical
scoring (on a scale of zero to 5) was performed by 3 to 4 individuals in a blinded manner
according to the method described previously [1]. Score of 0, no symptoms; 1, scratching
and rubbing around the nose and head; 2, puffiness around the eyes and mouth, diarrhea,
pilar erecti, reduced activity, and/or decreased activity with increased respiratory rate; 3,
wheezing, labored respiration, cyanosis around the mouth and the tail; 4, no activity after
prodding, or tremor and convulsion; and 5, death. Rectal temperature was measured using
a temperature probe (Yellow Springs Instrument Co., Yellow Springs, OH, USA) at

indicated time points before and afier challenge.

6.3.4 Histamine measurement

Plasma was collected 30 minutes after the i.p. challenge by cardiac puncture. Samples were
stored at ~80 ° C. Histamine content was measured using a Histamine ELISA Kit

developed by Neogen Inc. (A kind gift from Dr. Paul Satoh, Neogen Inc., Lansing, MI).

6.3.5 Pathological changes

103

Histological changes in the gastrointestinal tract were evaluated 1hr after one oral
challenge. Mice transdermally sensitized with hazelnut protein and saline control mice
challenged with hazelnut were sacrificed 1 hour after oral challenge by C02 euthanasia.
Tissues from the gastrointestinal tract were collected for histological analysis. The
gastrointestinal tract was divided into sections (esophagus, stomach, duodenum, jejunum,
ileum, and colon) and fixed in 10% neutral buffered formalin for 4 days. Of the sixteen
animals per group, six of them had their tissues taken for further analysis. Fixed tissues
were embedded in paraffin and cut into 3-5 pm thick sections and mounted on glass slides
(done by the Michigan State University Histology Lab). Slides were then stained with
hematoxylin-eosin (done by the Michigan State University Histology Lab) and analyzed for
histological changes by light microscopy. Each mouse tissue had one slide prepared with
both a longitudinal and cross-section. Therefore, there were six slides for each of the six

animals per group (esophagus, stomach, duodenum, jejunum, ileum, and colon).

6.3.6 Measurement of hazelnut antibodies in the plasma
We have previously described optimization of enzyme linked immunosorbent assay

(ELISA) for food specific IgE antibody analyses (chapter 3) [4, 5].

6.3.7 Statistical Analysis
Student’s t-test, ANOVA, Mann Whitney, and the Kruskal-Wallis tests were used to
evaluate significance using Analyse-ItTM software program (Analyse-It software Ltd,

Leeds, UK). The statistical significance level was set at 0.05.

104

6.4 Results

6.4.] Conformation of hazelnut specific antibody responses to transdermal exposure
in BALB/c mice

Mice were evaluated for the appearance of hazelnut binding allergic (IgE) antibodies
before and after each transdermal exposure. We used LPS-flee preparations of hazelnut
protein extract (Greer Labs) in this study LPS level as measured by LAL assay was at
<0.4 pg/mg of protein. As previously seen, pre—immune serum had no detectable hazelnut
specific IgE antibodies (mean titer <1/20). Similar to what we had previously seen,
following transdermal exposure with hazelnut (500 ug/mouse) but not saline, elevated
levels of IgE antibodies were noted (Table 6.3).

Table 6.3 Hazelnut specific IgE antibody titers in BALB/c mice pre-immune and at 6m

 

 

 

 

 

 

 

 

 

response.
Pre-
Challenge Group n= immune After 6‘h response
I.P. Saline transdermal/ Saline LP. 5 <20 <20
I.P. Saline transdermal/ Hazelnut LP. 5 <20 <20
I.P. Hazelnut transdermal/ Saline LP. 5 <20 7500 +/-650
I.P. Hazelnut transdermal/ Hazelnut LP. 5 <20 8300+I-280
Oral Saline transdermal/ Saline oral 16 <20 <20
Oral Saline transdermal/ Hazelnut oral 16 <20 <20
Oral Hazelnut transdermal/ Saline oral 16 <20 8600+I-5 70
Oral Hazelnut transdermal/ Hazelnut oral 16 <20 8200+I-610

 

 

 

 

 

 

 

*Titer is shown as geometric mean +/- SE.
** Pre-immune= Plasma from naive mice prior to sensitization.
*** 6‘h response is plasma from mice following six courses of transdermal hazelnut
exposure.

105

6.4.2 Transdermal exposure to hazelnut protein sensitizes BALB/c mice for systemic
anaphylaxis: Determination of threshold challenge dose

Hazelnut sensitized and saline exposed control mice were challenged systemically (i.p.)
with hazelnut or saline in a crisscross manner and observed for clinical signs of systemic
anaphylaxis during the next 4 hours. Hazelnut but not saline control mice exhibited signs
of systemic anaphylaxis (convulsions, hypothermia, labored breathing) that were scored
as described in the methods section (Figure 6.2). As evident, a significant difference in
the clinical score was seen between hazelnut sensitized vs. saline control mice both
challenged with hazelnut (Figure 6.2; mean score: HN group 3.67 +/-0.21; saline group:
0.0 +/- 0.0). Opposite challenge confirmed the need for prior-sensitization for systemic
anaphylaxis. We then examined rectal temperature as an additional indicator of systemic
anaphylaxis. As evident, hazelnut sensitized but not control mice exhibited a significant
drop in rectal temperature at 30 minutes (Figure 6.3). Furthermore, there was significant
elevation of systemic histamine levels at 30 minutes following systemic challenge with

hazelnut but not saline (Figure 6.4).

We then conducted a detailed study to determine the threshold of systemic dose of
hazelnut required to induce systemic anaphylaxis in previously sensitized mice. As
evident, 100 jig/mouse was sufficient to induce signs of systemic anaphylaxis (p<0.01;
Figure 6.5, 6.6), whereas a dose of 10 jig/mouse failed to induce any signs of systemic

anaphylaxis.

106

6.4.3 Transdermal exposure to hazelnut primes BALB/c mice for systemic
anaphylaxis when orally challenged

Hazelnut sensitized mice and saline control mice were challenged orally with hazelnut
(13 mg) and observed for clinical signs of systemic anaphylaxis. Following oral
challenge with hazelnut, hazelnut sensitized mice underwent systemic anaphylaxis, as
evident by convulsions, systemic anaphylactic scores (Figure 6.7) and diarrhea, whereas
saline control mice had no signs of anaphylaxis. Similar to systemic anaphylactic scores,
hazelnut sensitized mice challenged with hazelnut had a marked drop in rectal
temperature when comparing to saline control mice challenged with hazelnut, indicating

systemic anaphylaxis (Figure 6.8).

6.4.4 Transdermal exposure to hazelnut protein sensitizes BALB/c mice for
pathological changes in the small intestine when challenged orally with hazelnut
Hazelnut and saline exposed control mice were challenged orally one time with 13 mg of
hazelnut. Hazelnut sensitized mice underwent systemic anaphylaxis; whereas saline
exposed mice had no such effect. Tissues (esophagus, stomach, duodenum, jejunum,
ileum, and colon) were collected and fixed for pathological analysis. During tissue
collection severe edema of the small intestine was noted in hazelnut-sensitized mice
following oral challenge. There were marked pathological changes seen in mice that
were sensitized with hazelnut. Severe edema, vascular congestion and enterocyte
sloughing was seen in all three areas of the small intestine (duodenum, jejunum, ileum)
when compared to saline control mice (Figure 6.11, 6.12, 6.13 A-B). Mild edema was

noted in the esophagus of hazelnut-sensitized mice when compared to saline control mice

107

(Figure 6.9). No lesions were noted in the stomach or colon (Figure 6.10, 6.14 A-B) at

this time point.

108

6.5 Discussion

In this study we attempted to further our model of tree-nut allergy by profiling the
systemic anaphylactic responses (symptom scores, rectal temperature, plasma histamine)
and pathological changes of the gastrointestinal tract associated with tree-nut allergy.

The finding of these studies are i) Repeated transdermal exposure can sensitize mice for
systemic anaphylaxis (symptom scores, increased plasma histamine, decreased rectal
temperature) to hazelnut when challenged with an i.p. injection; ii) Repeated transdermal
exposure can sensitize mice for systemic anaphylaxis (symptom scores, decreased rectal
temperature, diarrhea) to hazelnut when challenged orally; and iii) Marked pathological
changes are seen in the small intestine at 1 hour after oral challenge with hazelnut in mice

transdermally sensitized mice.

To our knowledge, there is currently is not a validated mouse model of tree-nut allergy.
In the recent years, parallel with our studies, efforts to develop animal model of tree-nut
allergy have been reported. One that has been developed uses the atopic dog as a model
for peanut and tree-nut allergy [6]. This model has both strengths and weaknesses. One
strength is that the dog is one of the few species other than humans in which allergies
develop naturally on normal environmental exposure to a broad spectrum of allergens,
including pollens, house dust mites, human dander, fleas, and foods [7-9]. Also, the dog
can respond both vomiting and diarrhea in response to oral challenge due to food allergy
[6, 10]. A key weakness is the high expense of dog studies, having high enough 11
numbers in studies, and having knockout strains / molecular reagents available to study

mechanisms. Also, in this model animals are injected with an adjuvant (Alum), multiple

109

allergens are injected at a time (ex. Brazil nut, wheat, and soy all injected together) and

requires 2.5 years before animals are challenged to allergen, delaying scientific progress.

The only mouse based model of tree-nut allergy proposed uses anti-ulcer drugs to
promote hypersensitivity [11]. Here they report hazelnut specific IgGl, but no detectable
levels of IgE when mice were orally fed hazelnut extract with a pretreatment of anti-ulcer
drugs. Although oral sensitization is highly sought after, here without adjuvant (anti-
ulcer) there is no response and mice develop oral tolerance. They claim that this
treatment did sensitize mice for type I skin reactivity to hazelnut extract, as evidenced by
passive cutaneous anaphylaxis (PCA) reactions using naive mice and hazelnut specific
IgGl purified from plasma from the anti-ulcer group of mice. Here they do show that the
IgGl specific to hazelnut is anaphylactogenic in naive animals when concentrated, but do
not show if there is an in vivo consequence afier challenge. In mice there are two distinct
mechanisms that can induce anaphylaxis [12]. The first is the classical IgE-mast cell
mediated pathway that is associated with human allergy and the second is IgE
independent, using IgG and macrophages. Although a novel and important finding, this
model does not follow classical allergy mechanisms, showing allergen specific IgE,
therefore it is not a validated tree-nut allergy model. Therefore a model having the

classical immune markers seen in human allergy (IgE) is desirable.

Limitations of our study are that only a single dose of allergen was used in the oral

challenge study. A further profile of other doses could show a threshold dose with milder

110

symptoms. Plasma histamine could not be studied in the oral challenge model because of

difficulty in bleeding mice following oral challenge.

In summary, this method of sensitization can prime BALB/c mice for systemic
anaphylaxis when challenged systemically (i.p.) or orally with hazelnut protein. This
should serve as a useful tool not only to study further mechanisms of tree nut induced
systemic anaphylaxis but also to develop improved methods of treating/preventing this

immune mediated, potentially fatal, disorder.

111

Figures 6.6

Figure 6.1

Weekly protocol

 

 

 

 

Thursday
Patch off
Monday Friday
Patch on Bleed
Rest 1 week , i / / >
/ / Rest 2
Pre-Bleedl Do weekly days
shave back protocol 6
times . . .
I.P. Injection or oral gavage
challenge of hazelnut or saline.

 

Mice arrive @
6 weeks of age

Measure hazelnut

 

Study systemic anaphylaxis.

specific antibodies and

confirm positive
response.

Figure 6.1: Transdermal sensitization protocol for systemic anaphylaxis to hazelnut.

The schedule of cycles of hazelnut sensitization and testing for systemic anaphylaxis are

shown.

112

Figure 6.2

 

 

 

 

 

 

 

 

5
a)
L.
o
o
a:

4 - e o
E .
.o-o
o.
E 0
>.
m
.S 3 r '
E.
'6
.E’.’
x
5!
> 2 *‘
.c
o.
as
c
to
.9
E 1 L
a:
H
m
>.
(I)

O O
0 i
llllllllllllllllllllllllllllllllllllll
Saline Hazelnut Saline Hazelnut
Control mice challenged with Hazelnut sensitized mice challenged with
(n=5) (n=5)

Figure 6.2 Transdermal exposure to hazelnut sensitizes BALB/c mice for systemic
anaphylaxis when challenged by i.p. injection: Clinical symptom scores

Groups of BALB/c mice (n=5/group) were sensitized with hazelnut (500 pg per mouse)
or saline by transdermal exposure as described in Figure 6.1. After 6 cycles of exposure,
IgE induction was confirmed and then exposed to hazelnut (1,000 pg/mouse) or saline
systemically (i.p. injections) and examined for indicators of systemic anaphylaxis.
Clinical symptoms are shown as a scatter plot with each symbol representing one mouse.
Mann-Whitney test results: hazelnut sensitized mice with hazelnut challenge p<0.05 vs.

all other groups.

113

Figure 6.3

 

 

 

 

 

 

 

 

 

 

 

100
- Saline / Saline challenge
m Hazelnut / Hazelnut challenge
VIA Saline / Hazelnut challenge
1:: 96 — [:1 Hazelnut/Salinechallenge
°V
<9
$—
3
m
L.
a, +
o. +
E 92
2’.
E
o
a)
o:
88
84

   

30 min

 

Timepoint of analysis

Figure 6.3 Transdermal exposure to hazelnut sensitizes BALB/c mice for systemic
anaphylaxis when challenged by i.p. injection: Rectal temperature

Groups of BALB/c mice (n=5/group) were sensitized with hazelnut (500 pg per mouse)
or saline by transdermal exposure as described in Figure 6.1. After 6 cycles of exposure,
IgE induction was confirmed and then exposed to hazelnut (1,000 pg/mouse) or saline
systemically (i.p. injections) and examined for indicators of systemic anaphylaxis.
Changes in rectal temperature are shown. Data shown is mean +/- SE. Kruskal-Wallis

test results: P<0.001 vs. all other groups.

114

Figure 6.4

 

 

 

 

 

 

 

 

 

 

 

*
300 I“ Y

E

O)

C

3

(D

E 2.. L

d)

.E

E

g _—

(n

E

m

E loo —

(I)

E

o.
X XX)”

0 K 2< ’ xx \

Saline Hazelnut
(n=5) (n=4)

Figure 6.4 Transdermal exposure to hazelnut sensitizes BALB/c mice for systemic
anaphylaxis when challenged by i.p. injection: Plasma histamine levels

Groups of BALB/c mice (n=S/group) were sensitized with hazelnut (500 pg per mouse)
or saline by transdermal exposure as described in Figure 6.1. After 6 cycles of exposure,
IgE induction was confirmed and then exposed to hazelnut (1,000 pg/mouse) or saline
systemically (i.p. injections) and examined for indicators of systemic anaphylaxis.
Plasma histamine profile at 30 minutes following systemic exposure to hazelnut or saline

is shown as mean +/- SE. Student’s t-test results: * = p<0.001 for saline vs. hazelnut.

115

Figure 6.5

 

 

 

 

 

 

5
a)
L.
o
o
w
E 4 — 01—1—
.9 o
o.
E
>
w
.8 3 -
.E
T)
.2
x
2*
(U 2 t- O.
.c
a.
(U
c
to
.9
E 1 - o o
9.
w
>.
(I)

O
l—m—w
O lllllllllllllllLlLLLllllllllllllllJLlllllllll
Saline 10 ug/mouse 100 uglmouse 1000 ug/mouse
Control mice challenged with Hazelnut sensitized mice challenged with hazelnut
(n=6) (n=4/dose)

Figure 6.5 Determination of threshold challenge dose of hazelnut required to induce
systemic anaphylaxis in hazelnut sensitized BALB/c mice: Clinical symptom scores

Groups of BALB/c mice (n=4-6/group) were sensitized with hazelnut (500 pg per mouse)
or saline by transdermal exposure as described in Figure 6.1. After 6 cycles of exposure,
IgE induction was confirmed and then exposed to indicated challenge doses (10, 100, and
1,000 pg/mouse) or saline systemically (i.p. injections) and examined for indicators of
systemic anaphylaxis. Clinical symptoms are shown as a scatter plot with each symbol
representing one mouse. Kruskal-Wallis test results; hazelnut sensitized mice with

hazelnut 1,000-pg/ mouse challenge p<0.05 vs. all other groups.

116

Figure 6.6

 

    
 

100
- Saline / Hazelnut Dug/mouse challenge
””4 Hazelnut / Hazelnut 10 uglmouse challenge
Hazelnut I Hazelnut 100 uglmouse challenge
:1 Hazelnut/ Hazelnut 1000 uglmouse challenge
96 -
u’.‘
i,
9
B
9
cu
o. 'r
s //
6
‘6 Jr
0.)
tr
4

 

 

 

 

 

 

3
Timepoint of analvsis

0

min

 

Figure 6.6 Determination of threshold challenge dose of hazelnut required to induce
systemic anaphylaxis in hazelnut sensitized BALB/c mice: Changes in rectal
temperature

Groups of BALB/c mice (n=4-6/group) were sensitized with hazelnut (500 pg per mouse)
or saline by transdermal exposure as described in Figure 6.1. After ”*"Icles of exposure,
IgE induction was confirmed and then exposed to indicated challenge dose *‘ ", 100, and
1,000 pg/mouse) or saline systemically (i.p. injections) and examined for indicators of
systemic anaphylaxis. Changes in rectal temperature are shown as mean +/- SE.

ANOVA test results: * = p<0.01 vs. all other groups.

117

Figure 6.7

 

 

 

 

 

 

5
g 4 ~ W
S
a 2 ”
E. 1 ~
(I)
o W '
Saline Hazelnut saline Hazeant
(n=16) (n=16) (n=16) (n=16)
Saline exposed mice orally Hazelnut sensitized
challenged with mice orally challenged with

Figure 6.7 Transdermal exposure to hazelnut sensitizes BALB/c mice for systemic
anaphylaxis when challenged by oral gavage: Clinical symptom scores

Groups of BALB/c mice (n=16/group) were sensitized with hazelnut (500 pg per mouse)
or saline by transdermal exposure as described in Figure 6.1. After 6 cycles of exposure,
IgE induction was confirmed and then exposed to hazelnut (13 mg/mouse) or saline by
oral gavage and examined for indicators of systemic anaphylaxis. Clinical symptoms at
30 minutes post challenge are shown as a scatter plot with each symbol representing one
mouse. Mann-Whitney test results: hazelnut sensitized mice with hazelnut challenge

p<0.05 vs. all other groups.

118

Figure 6.8

 

  

  
    
       
     
     
      

  

 

    
 

 

 

 

 

 

 

 

 

C

O

V95

‘1’ ,__

a ' ' ' ’3

ego — ,.:.,.: /7

<13 ’e’o’o’q

0- ’o’o’o’c

E b’o’o’c 0.

085 ” ’o’o’e’c ’

... no.»

.—. :zzzzzzz //

$80 . . /

(I Sallne Hazelnut Sallne Hazelnut
(n=16) (n=16) (n=16) (n=16)

Saline exposed mice
orally challenged with

Hazelnut sensitized
mice orally challenged with

Figure 6.8 Transdermal exposure to hazelnut sensitizes BALB/c mice for systemic

anaphylaxis when challenged by oral gavage: Changes in rectal temperature

Groups of BALB/c mice (n=l6/group) were sensitized with hazelnut (500 pg per mouse)

or saline by transdermal exposure as described in Figure 6.1. After 6 cycles of exposure,

IgE induction was confirmed and then exposed to hazelnut (13 mg/mouse) or saline by

oral gavage and examined for indicators of systemic anaphylaxis. Change in rectal

temperature 30 minutes after oral challenge is shown as mean +/— SE. Before hazelnut

challenge (crossed lines) and 30 minutes after hazelnut challenge (single lines). ANOVA

test results: hazelnut sensitized mice challenged with hazelnut p<0.001 at 30 min vs. all

other groups.

119

 

Control Esophagus Treatment

Figure 6.9 (A-B) Transdermal exposure to hazelnut sensitizes BALB/c mice for
pathological changes in the esophagus when challenged by oral gavage

Groups of BALB/c mice (n=16/group) were sensitized with hazelnut (500 pg per mouse)
or saline by transdermal exposure as described in Figure 6.1. Afier 6 cycles of exposure,
IgE induction was confirmed and then mice were exposed to hazelnut (13 mg/mouse) by
oral gavage. The esophagus was taken 1 hr after hazelnut oral gavage and fixed for 4
days in 10% formalin in PBS. Fixed tissue was embedded in paraffin, cut into 3-5 pm
thick sections and mounted on slides. Slides were then stained with hematoxylin-eosin
and analyzed with a light microscope for histological changes. A) Example of an
esophagus from saline sensitized mice (n=16) orally gavaged with hazelnut; B) Example

of an esophagus from hazelnut sensitized mice (n=16) orally gavaged with hazelnut.

120

 

Control Stomach Treatment

Figure 6.10 (A-B) Transdermal exposure to hazelnut sensitizes BALB/c mice for
pathological changes in the stomach when challenged by oral gavage

Groups of BALB/c mice (n=16/group) were sensitized with hazelnut (500 pg per mouse)
or saline by transdermal exposure as described in Figure 6.1. Afier 6 cycles of exposure,
IgE induction was confirmed and then mice were exposed to hazelnut (13 mg/mouse) by
oral gavage. The stomach was taken 1 hr afier hazelnut oral gavage and fixed for 4 days
in 10% formalin in PBS. Fixed tissues were embedded in paraffin, cut into 3-5 pm thick
sections and mounted on slides. Slides were then stained with hematoxylin-eosin and
analyzed with a light microscope for histological changes. A) Example of a stomach
from saline sensitized mice (n=16) orally gavaged with hazelnut; B) Example of a

stomach from hazelnut sensitized mice (n=16) orally gavaged with hazelnut.

121

Figure 6.1 l

 

Control Duodenum Treatment

Figure 6.11 (A-B) Transdermal exposure to hazelnut sensitizes BALB/c mice for
pathological changes in the duodenum when challenged by oral gavage

Groups of BALB/c mice (n=16/group) were sensitized with hazelnut (500 pg per mouse)
or saline by transdermal exposure as described in Figure 6.1. After 6 cycles of exposure,
IgE induction was continued and then mice were exposed to hazelnut (13 mg/mouse) by
oral gavage. The duodenum was taken 1 hr after hazelnut oral gavage and fixed for 4
days in 10% formalin in PBS. Fixed tissues were embedded in paraffin, cut into 3-5 pm
thick sections and mounted on slides. Slides were then stained with hematoxylin-eosin
and analyzed with a light microscope for histological changes. A) Example of a
duodenum fi'om saline sensitized mice (n=16) orally gavaged with hazelnut; B) Example

of a duodenum fiom hazelnut sensitized mice (n=16) orally gavaged with hazelnut.

122

Figure 6.12

 

Control Jejunum Treatment

Figure 6.12 (A-B) Transdermal exposure to hazelnut sensitizes BALBIc mice for
pathological changes in the jejunum when challenged by oral gavage

Groups of BALB/c mice (n=1 6/ group) were sensitized with hazelnut (500 pg per mouse)
or saline by transdermal exposure as described in Figure 6.1. After 6 cycles of exposure,
IgE induction was confirmed and then mice were exposed to hazelnut (13 mg/mouse) by
oral gavage. The jejunum was taken 1 hr after hazelnut oral gavage and fixed for 4 days
in 10% formalin in PBS. Fixed tissues were embedded in paraffm, cut into 3-5 pm thick
sections and mounted on slides. Slides were then stained with hematoxylin-eosin and
analyzed with a light microscope for histological changes. A) Example of a jejunum
from saline sensitized mice (n=16) orally gavaged with hazelnut; B) Example of a

jejunum from hazelnut sensitized mice (n=16) orally gavaged with hazelnut.

123

 

Control lleum Treatment

Figure 6.13 (A-B) Transdermal exposure to hazelnut sensitizes BALB/c mice for
pathological changes in the ileum when challenged by oral gavage

Groups of BALB/c mice (n=16/group) were sensitized with hazelnut (500 pg per mouse)
or saline by transdermal exposure as described in Figure 6.1. After 6 cycles of exposure,
IgE induction was confirmed and then mice were exposed to hazelnut (13 mg/mouse) by
oral gavage. The ileum was taken 1 hr afier hazelnut oral gavage and fixed for 4 days in
10% formalin in PBS. Fixed tissues were embedded in paraffin, cut into 3-5 pm thick
sections and mounted on slides. Slides were then stained with hematoxylin-eosin and
analyzed with a light microscope for histological changes. A) Example of an ileum fiom
saline sensitized mice (n=16) orally gavaged with hazelnut; B) Example of an ilemn from

hazelnut sensitized mice (n=16) orally gavaged with hazelnut.

124

Figure 6.14

 

Control

Figure 6.14 (A-B) Transdermal exposure to hazelnut sensitizes BALB/c mice for
pathological changes in the colon when challenged by oral gavage

Groups of BALB/c mice (n=l6/group) were sensitized with hazelnut (500 pg per mouse)
or saline by transdermal exposure as described in Figure 6.1. Afier 6 cycles of exposure,
IgE induction was confu'med and then mice were exposed to hazelnut (13 mg/mouse) by
oral gavage. The colon was taken 1 hr after hazelnut oral gavage and fixed for 4 days in
10% formalin in PBS. Fixed tissues were embedded in paraffin, cut into 3-5 pm thick
sections and mounted on slides. Slides were then stained with hematoxylin-eosin and
analyzed with a light microscope for histological changes. A) Example of a colon from
saline sensitized mice (n=16) orally gavaged with hazelnut; B) Example of a colon from

hazelnut sensitized mice (n=16) orally gavaged with hazelnut.

125

6.7 References

10.

11.

Li, X.M., et al., A murine model of IgE-mediated cow's milk hypersensitivity. J
Allergy Clin Immunol, 1999. 103(2 Pt 1): p. 206-14.

Li, X.M., et al., A murine model of peanut anaphylaxis: T— and B-cell responses to
a major peanut allergen mimic human responses. J Allergy Clin Immunol, 2000.
106(1 Pt 1): p. 150-8.

Morafo, V., et al., Genetic susceptibility to food allergy is linked to differential
T H2- T H 1 responses in C3H/HeJ and BALB/c mice. J Allergy Clin Immunol,
2003. 111(5): p. 1122-8.

Birmingham, N., S. Thanesvorakul, and V. Gangur, Relative immunogenicity of
commonly allergenic foods versus rarely allergenic and nonallergenic foods in
mice. J Food Prot, 2002. 65(12): p. 1988-91.

Birmingham, N., et al., An ELISA-based method for measurement of food—specific
IgE antibody in mouse serum: an alternative to the passive cutaneous anaphylaxis
assay. J Immunol Methods, 2003. 275(1-2): p. 89-98.

Teuber, S.S., et al., The atopic dog as a model of peanut and tree nut food allergy.
J Allergy Clin Immunol, 2002. 110(6): p. 921-7.

Mueller, R.S., S.V. Bettenay, and L. Tideman, Aero-allergens in canine atopic
dermatitis in southeastern Australia based on 1000 intradermal skin tests. Aust

Vet J, 2000. 78(6): p. 392-9.

Paterson, S., Food hypersensitivity in 20 dogs with skin and gastrointestinal signs.
J Small Anim Pract, 1995. 36(12): p. 529-34.

Sture, G.H., et al., Canine atopic disease: the prevalence of positive intradermal
skin tests at two sites in the north and south of Great Britain. Vet Immunol
Immunopathol, 1995. 44(3-4): p. 293-308.

Buchanan, BB. and CL. Frick, The dog as a model for food allergy. Ann N Y
Acad Sci, 2002. 964: p. 173-83.

Scholl, I., et al., Antiulcer drugs promote oral sensitization and hypersensitivity to
hazelnut allergens in BALB/c mice and humans. Am J Clin Nutr, 2005. 81(1): p.
154-60.

126

l2. Finkelman, F.D., et al., Molecular mechanisms of anaphylaxis: lessons from
studies with murine models. J Allergy Clin Immunol, 2005. 115(3): p. 449-57;
quiz 458.

127

CHAPTER SEVEN

7.0 Effect of a diet rich in EPA and DHA on systemic immune responses to hazelnut:

A pilot study

7.1 Abstract

Background: Previously, we found that in BALB/c mice, transdermal hazelnut is capable
of directly eliciting hazelnut specific binding allergenic (IgE) antibodies via activating
Type-2 cytokine. EPA and DHA, the main n-3 polyunsaturated fatty acids of fish oil,
have a controversial role in several allergic disorders, with several groups showing a
beneficial role in asthma. Currently it is not known whether n-3 fatty acids from fish
(EPA, DHA) may have either therapeutic or preventative effects in food allergy.
Hypothesis, objective and approach: Here we tested if EPA + DHA together can have an
effect on systemic immune responses to transdermal hazelnut sensitization in either a
preventative or therapeutic manner in this model.

EMS; With the addition of EPA and DHA in the diet in a preventative manner,
hazelnut specific IgE responses were increased (p<0.05) in comparison to the control
diet. EPA and DHA supplementation also altered the [FN-y / IL—4 ratio in favor of IL-4
dominance, thus providing a mechanism of enhancement of IgE production seen here. In
the therapeutic study no significant alteration in specific IgE response was noted.
Conclusion: EPA and DHA supplementation seems to enhance hazelnut specific IgE

responses via reducing the IFN-y / IL-4 ratio in a preventative model. Further studies

128

(e. g. repeating with a larger number of mice per group (i.e. n=10), dose response, and

disease model studies) are suggested to test the impact of fish oil on tree-nut allergy.

129

7.2 Introduction

Currently, our modern diets differ in many aspects from our past traditional diets, with
more complex, processed foods coming to the forefront, replacing fruits and vegetables.
One change that has been identified is increased consumption of n-6 polyunsaturated
fatty acids and decreased consumption of n-3 polyunsaturated fatty acids [1]. It has been
estimated that the ratio of n-6 to n—3 fatty acids in the typical western diet now ranges
from approximately 20-30:1 instead of the traditional range of 1-2:1 [2]. This alteration
in fatty acid consumption is a major candidate factor, as both epidemiological and
experimental evidence suggest a link between the declining consumption of anti-

IF Nlammatory n-3 polyunsaturated fatty acids with the rise in allergenic diseases [3].
Two of these n-3 polyunsaturated fatty acids EPA and DHA have had controversial
outcomes in allergenic disorders. Hodge et al have shown positive observations,
observing that children who regularly ate oily fish were significantly less likely to
develop asthma, odds ratio 0.26 [4]. While others such as Woods et al in the Cochrane
review 2003 conclude that there is no evidence of consistent benefit of n-3 PUFAs in the
management of asthma. Currently it is not known weather EPA and DHA could benefit,

harm, or have no impact on food allergy sufferers.

A recent study published in the New England Journal of Medicine showed that monthly
injections of humanized recombinant anti-IgE antibodies could increase the amount of
peanut that is tolerable in peanut-sensitive subjects [5], showing IgE as a target
mechanism for food allergy. Although a great find, monthly injections are needed to

achieve this increased tolerance. Studies like this highlight the importance that therapies

130

targeted at IgE in food allergy therapies. Therefore new, non-invasive therapies and
strategies (e. g. dietary supplementation, pro-biotics) need to be developed / studied to

alter the immune response to food allergens.

Previously we characterized the systemic immune response to transdermal hazelnut
protein in BALB/c mice as having robust specific IgE, total IgE, hazelnut driven IL-4,

and an IFN-y/IL-4 ratio in favor of IL-4.

Here we test if EPA + DHA together can have an effect on systemic immune responses to
transdermal hazelnut sensitization in either a preventative or therapeutic manner in this
model. EPA and DHA supplementation seems to enhance hazelnut specific IgE
responses via reducing the [FN-y / IL-4 ratio in a preventative model. Further studies

(e. g. repeating with a larger number of mice per group (i.e. n=10), dose response, and

disease model studies) are suggested to test the impact of fish oil on tree-nut allergy.

131

7.3 Materials and Methods
7.3.1 Materials

Materials used here were as listed earlier in Chapter 5.

7.3.2 Transdermal sensitization and bleeding

All mice were purchased from The Jackson Lab (Bar Harbor, Maine, USA). Only adult
animals (6-8 weeks age) were used in the study. All animal procedures used were in
accordance with the Michigan State University policies. Transdermal exposure

experiments were performed using the method described by us before (Chapter 4) [6].

7.3.3 Diets and experimental design

Profiling the immune response to transdermal hazelnut was done using a commercially
available diet, Harlan Teklad 22/5 Rodent diet. Experimental diets were based on the
purified AIN-93G formulation [7], and had the following ingredients (per kg): 10g AIN-
93G mineral mix, 10g AIN 93G vitamin mix, 200g casein, 397.5 g cornstarch, 132 g
Dyetrose (dextrinized cornstarch), 50g cellulose, 3g L-cystine, 2.5g choline bitartrate, 14
mg tert-butylhydroquinine, and 100g sucrose (Dyets). Oils were then added at 70g/kg
described below. Corn oil (Dyets), oleic acid (Dyets) DHA-enriched fish oil (containing
604g/kg DHA, 71 g/kg EPA) (Ocean Nutrition) and EPA-enriched fish oil (containing
540g/kg EPA, 71 g/kg DHA) (Ocean Nutrition) were added to the basal diet to achieve
two experimental diets. Diet formulation was previously described by J ia O. et a1. [8].
Table 7.1 shows the experimental diets, AIN-93G diets containing 10 g/kg corn oil plus
60 g/kg oleic acid (control), 10 g/kg corn oil plus 37 g/kg oleic acid and 23 g/kg DHA +

EPA enriched fish oil (402 g/kg DHA, 341 g/kg EPA) (Ocean Nutrition). Dry ingredients

132

 

 

 

 

were added to a clean, Kitchen Aid mixer and mixed for five minutes. Oils were then
added and mixed for an additional thirty minutes to ensure even distribution of contents.
Diets were prepared fresh every 2 weeks, stored in aliquots at -20°C, and was provided
fresh daily to the mice. The final lipid profiles of the experimental diets are shown in
(Table 7.2). In the preventative study the respective diets were feed to their respective
groups for 4 weeks prior to hazelnut sensitization and continued throughout the study
(Figure 7.7). In the therapeutic study animals were on a chow diet during hazelnut
sensitization and then put on their respective diets after an established IgE response was
seen (Figure 7.14). Comparison of the immune response to the chow diet (Harlan Teklad

22/5) and the control diet for the fish oil experiment were done.

7.3.4 Measurement of hazelnut specific IgE antibodies and total IgE in the plasma
We have previously described optimization of enzyme linked immunosorbent assay

(ELISA) for food specific IgE antibody analyses (Chapter 3) [9, 10].

7.3.5 Spleen cell culture and cytokine analyses
Spleen cells were harvested and standard cell cultures were setup essentially as described

(Chapter4) [11].

7.3.6 Lipid extraction and fatty acid analysis

To confirm tissue incorporation of (n-3) PUFA after experiment is complete, spleen

phospholipid content was be measured by a modification of the method of Hasler et al

133

[12]. Briefly, pooled mouse spleens were homogenized with a chloroform: methanol

(2: 1) solution. Total phospholipids were extracted, separated, and collected using a silica
column. Phospholipid samples were dried and esterified with methanol: acetonitrile:
boron tri-fluoride (112425, by vol). The resulting FAME was then extracted with hexane.
After centrifugation at 1200 x g for 5 min, the hexane supernatant was decanted, dried,
and re-dissolved in chloroform and then analyzed by GC utilizing a Varian 3700 GLC.
Fatty acid profiles were then identified by comparing the retention times with those of
appropriate standard FAME (Nu-Check-Prep). Samples were run twice on the Varian
3700 GLC so that a standard error could be determined. Table 7.3 and 7.4 shows the
spleen phospholipid fatty acid profile for the preventative and therapeutic studies

respectively.

7.3.7 Statistical Analysis
Student’s t-test and ANOVA tests were used to evaluate significance using Analyse—ItTM
software program (Analyse-It software Ltd, Leeds, UK). The statistical significance level

was set at 0.05.

134

7.4 Results

7.4.1 Immune responses in mice following transdermal exposure to hazelnut protein
in BALB/c mice: comparison of different control diets

In order to study dietary changes in out model of hazelnut allergy, we first needed to
compare the immune responses in the chow diet (Harlen Teklad 22/5) and the AIN—93
based control diet to be used in the dietary modification studies. Hazelnut specific IgE
(Figure 7.1), total plasma IgE levels (Figure 7.2), hazelnut specific IgG2a (Figure 7.3),
hazelnut driven IL-4 recall response (Figure 7.4), hazelnut driven IFN-y (Figure 7.5), and

IFN-y/lL-4 ratio (Figure 7.6) were compared with no major alterations being noted.

7.4.2 EPA and DHA incorporation into the phospholipid fatty acid profile in mouse
spleens

Following completion of the experiment, spleens were collected from all of the animals
in the various groups. One half of the spleen was used for cell culture and the other half
for phospholipid fatty acid analysis to confirm EPA and DHA incorporation into the
membranes. Both EPA [C20:5(n-3)] and DHA [C22:6(n-3)] were increased in the fish
oil groups (p<0.05) after feeding, whereas in control animals there was very little EPA

and DHA found (Table 7.3 preventative study, Table 7.4 therapeutic study).

7.4.3 Characterization of hazelnut specific antibody responses to transdermal

exposure in BALB/c mice

135

Mice were evaluated for the presence of hazelnut binding specific antibodies (IgE) in the
plasma before and afier each transdermal exposure. We used LPS-free preparations of
hazelnut protein extract (Greer Labs) in this study (LPS level as measured by LAL assay:
<0.4 pg/mg of protein). Pre-immune serum did not have detectable hazelnut specific IgE
antibodies (mean titer <1/ 20) in both the preventative and therapeutic studies (Figure
7.8A, 7.15A). Following transdermal exposure with hazelnut (500 pg/mouse), both in
the test diet as well as the control diet, but not saline, elevated levels of hazelnut specific
IgE antibodies appeared (Figure 7.8B, 7.15B). In the test diet + HAZ group (preventative
study) animals had a significantly higher level (p<0.05) of hazelnut specific IgE (Figure
7.88, C, D) when compared to the control diet counterpart. In the therapeutic study there
was no significant effect seen on hazelnut specific IgE levels (Figure 7.15B, C, D). Mice
were also evaluated for the presence of hazelnut specific IgG2a before and after
transdermal exposure to hazelnut or saline. In the preventative study no effect of diet was
seen on hazelnut specific IgG2a (Figure 7.9), whereas in the therapeutic study there was
an increased presence of hazelnut specific IgG2a seen at several time points (Figure

7.16).

7.4.4 Measurement of total IgE levels in plasma of BALB/c

Mice were evaluated for total plasma IgE before and after transdermal exposure to
hazelnut (500 pg/mouse) or saline alone. As evident, pre-immune plasma had very little
total IgE (Figure 7.10, 7.17). Following transdermal application, in the hazelnut
sensitized groups but not saline alone, total IgE levels were increased. Total plasma IgE

was significantly higher (p<0.05) in the test diet groups when compared the control diet

136

hazelnut sensitized group in both the preventative and therapeutic studies (Figure 7.10,

7.17).

7.4.5 Characterization of hazelnut driven cytokine responses (IL-4, IFN-y) following
transdermal exposure to hazelnut protein in BALB/c mice

In order to study the mechanism of IgE response to hazelnuts, we examined the IL-4
cytokine responses in hazelnut-sensitized mice. As evident, hazelnut significantly
activated the Type—2 cytokine, IL-4, in spleen cells from hazelnut but not saline exposed
mice (Figure 7.11, 7.18). Spleen cells isolated from the test diet fed groups elicited
significantly less IL-4 in response to hazelnut (p<0.05) (both 100 and 500 pg stimulation)

in both the preventative as well as the therapeutic study (Figure 7.1 1, 7.18).

In spleen cells isolated from the test diet fed group, IFN-y levels were significantly lower
(p<0.05) than the control diet fed group (preventative study) (Figure 7.12). In the

therapeutic study a significant effect (p<0.05) was also evident (Figure 7.19).

7.4.6 Characterization of IFN-y / IL-4 ratio in BALB/c mice following transdermal
exposure to hazelnut protein: effect of dietary fatty acid

In order to further study the mechanism of IgE response to hazelnuts, we determined the
lFN-y / IL-4 ratio in mice fed both the control diet and also the DHA + EPA diet. In the
preventative study, the IFN-y / IL-4 ratio was significantly lower (p<0.05) in the fish oil
diet in both the 100 and 500 pg hazelnut stimulation, whereas in the therapeutic study no

such trend was seen (Figure 7.13, 7.20). ,

137

7.5 Discussion

In this study we profiled hallmarks of allergenic immune responses (specific and total

I gE, IL-4, and IFN-y / IL-4 ratio) to transdermal hazelnut protein in BALB/c mice that
were supplemented with a diet rich in EPA and DHA. Here, specific and total IgE are
used as markers of Type-I hypersensitivity. Cytokine profile and ratio is used as showing
the relative dominance of either T—helper 1 or T—helper 2 cytokines. A T-helper 2
dominated response would favor the development of Type I hypersensitivity. There are
several novel finding of these studies; i) EPA + DHA supplementation together can lead
to a increased specific IgE response when given in a preventative manner; ii) both IL-4
and IFN-y hazelnut specific recall responses are significantly lower with EPA + DHA
supplementation; iii) the IFN-y / IL-4 ratio is swayed towards IL-4 with EPA + DHA
supplementation, showing a mechanism driving the increase in IgE responses seen in the

preventative study.

To our knowledge, there is only one other study using fish oil as a therapy for food
allergy. Prickett et a1. [13] showed an enhancement in both IgE and IgG antibodies to
egg albumin in Sprague-Dawley rats in fish fat fed animals in comparison to beef fat.
This study used alum as an adjuvant to elicit a response by IR injection. Therefore, in
both species (rat or mouse) fish oil seems to drive an enhancement of IgE responses.
Whereas they did not address the mechanism of enhancement, we concluded that the

alteration of the IFN-yI / IL-4 ratio is driving this increase in IgE levels.

138

Further analysis of molecules involved in the IFNlammatory response need to be looked
at in this type of therapy for food allergy. Ecosinoids of the n-3 variety tend to be less

IF Nlammatory than their counterparts derived from arachadonic acid (n-6). Interestingly
is the fact is arachadonic acid in both the therapeutic study and the preventative study
was found to be higher in the fish oil groups than in the control diet groups. Normally in
diets rich in n-3 fatty acids, arachadonic acid is partially replaced by the n-3 fatty acids in

the diet.

Since activation of Type-2 cytokine response (IL-4) is critical to initiate an allergic
response, we tested the hypothesis that a diet rich in fish oil can alter the classical type-2
(IL-4) responses seen in food allergy. Transdermal exposure to hazelnut is capable of
activating these cytokines. Our data confirmed the hypothesis. Our data argue that
hazelnut activates Type-2 cytokine (IL-4) in a dose dependent manner and is significantly
lowered (p<0.05) by using a diet rich in fish oil. IL-4 is the key class switch factor
allowing IgE to be made; therefore we assessed both total and specific IgE. EPA and
DHA supplementation together had a significant effect on hazelnut specific IgE in the
preventative study. With IL-4 being lowered and hazelnut specific IgE increasing, we
did further analysis of the lFN-y / IL-4 ratio, EPA + DHA supplementation significantly
drove this ratio in favor of IL-4, thus explaining the rise in hazelnut specific IgE seen in

the preventative study.

Others have used this same type of dietary approach in battling food allergy. Their

approaches have been to use supplement like compounds in pharmaceutical ways in order

139

to help battle food allergy. These compounds include an edible-mushroom derived

protein and food allergy herbal formula-1 [14, 15].

Hsiesh et al. [14] investigated the use of an immunomodulatory protein (F IP-fve) isolated
from the edible mushroom F lammulina velutipes, in a mouse model (BALB/c) of OVA
allergy. They used this protein (FIP-fve 200 uglmouse) orally before and during OVA
i.p. sensitization using alum as an adjuvant. Following sensitization, OVA-specific
antibodies, cytokine profile, anaphylaxis symptoms following oral challenge, plasma
histamine and histology of the intestines were examined. They report that mice receiving
oral FIP-fve treatment during sensitization had an impaired OVA-specific IgE response
with a T-helper 1 dominated cytokine response. Further, mice were protected from signs
of anaphylaxis after oral challenge with OVA with significant changes seen in symptom
scores, plasma histamine levels and very mild mucosa] edema and epithelial damage
when compared the severe edema, vascular congestion and damage to the intestinal
lining. They then report that this protein then may have immunoprophylaxis properties
for food allergy. Although very interesting, how this protein can induce a T-helper 1
skewing response, the use of adjuvant here skews the results. It is unknown if FIP-fve is
altering the adjuvant reSponse or the hypersensitivity response. Therefore our model of
adjuvant free sensitization is more useful to elucidate potential therapies for use in

altering immune responses to food allergens.

Li et al. [15] has investigated using a traditional Chinese medicine (F ood Allergy Herbal

Formula-1, F AHF-l) as a therapy for peanut induced anaphylaxis in a mouse model of

140

peanut allergy. This model uses C3H/HeJ mice sensitized to freshly ground whole
peanut in the presence of cholera toxin. One week after the final sensitization dose, mice
received either 21 mg of FAHF-l or water (sham) treatment by intragastric gavage twice
daily for seven weeks. Mice were then challenged with 10mg of crude peanut extract.
Anaphylaxis symptom scores, body temperatures, plasma histamine, IgE levels, and
cytokine profiles were analyzed. They report that FAHF-l completely blocked peanut-
induced anaphylaxis symptom scores and significantly reduced histamine release.
Further, peanut specific IgE levels were significantly lower after 2 weeks of treatment
and stayed lower even after 4 weeks after discontinuation of treatment. T-helper 2-
cytokine synthesis (IL-4, IL-5, IL-1 3) was lower, but no alteration of IFN-y was seen.
From this they concluded that FAHF-l protected peanut-sensitized mice from
anaphylactic reactions and reversed established IgE-mediated peanut allergy. This
therefore may serve as a valuable treatment for peanut allergy. Again exciting results,
although the use of cholera toxin as an adjuvant, even though treatment was after
sensitization, may alter the immune response to this therapy. The continuing adjuvant
effect may be altered and not the actual Type-1 hypersensitivity, therefore an adjuvant

free model would be more representative to study dietary therapies.

Treatments such as these, while showing benefits require a constant need to take the
supplement, whereas in a dietary lifestyle change such as eating more fish does not
require anything else besides a commitment. Therefore more research needs to be done

in these non-invasive ways to fight disease.

141

In this pilot study we assessed the potential use of EPA and DHA as a prevention and/or
therapy for genetically prone populations for food allergy. Since this was a pilot study
and not actually looking at disease it cannot be concluded if a benefit or harm is evident.
After finishing this study our model of tree-nut allergy has progressed to include
anaphylactic responses (clinical scores, rectal temperature, plasma histamine and gut
pathology). With this we have also assessed dose of allergen needed to cause systemic
anaphylaxis in an LP. challenge. Using this enhanced model, encompassing both
hallmarks of allergy as well as clinical responses, a better understanding of how and why

EPA and DHA affect food allergy responses could be determined.

Limitations of this study include the relative low number of animals per group. Here we
only used one dose of DHA+EPA, a dose study would be desirable to see if an effect
could be achieved at a lower dose. We also only used one strain of mouse; further studies
on other strains could show different results. We only assessed immune response to
transdermal hazelnut protein; a further assessment of disease would transform responses

to actual disease state.

In summary, with the changes in the modem diet to a more processed and n-6 rich one
and the rise in allergenic disorders seeming to coincide, we looked at hallmarks of food
allergy (food specific IgE, total IgE and IL-4) after feeding a diet rich in EPA and DHA
in a mouse model of tree-nut allergy. We show here that EPA + DHA can have altering
effects with regards to immune responses to transdermal hazelnut protein in BALB/c

mice, but more research needs to be done to assess the implications of these effects.

142

comp:

ova:

own: 00 p: on: ow:

c0326 mEmmE

ON:

 

 

d

- u H q

-

 

 

m

 

 

<

 

3:039 25 lLflll
omcoamm. Em lml
ecaEE_-o._n_ l®l

mu

 

 

ans .6 cm: .2 86 .2200

o

owNCF coo: own: on F2. on:

03..

ON:

 

 

- u q - u

u

 

 

3:032 50 ll
mmcoawo. Em lml

 

 

<

 

—-l

 

 

Eng 86 26,6

‘- O
(wu 069-907 00) ewseld U! leAel qv afil OU!09dS anIezeH

N

(’0

as 2:9“.

mocswi 0.5

143

2:68 8: 8m 8:: 886

9:8: 688 3. .Amm -\+ 568V no we 26>“: 6938: 360:8: 6:6on 8868: 95:5 8an 38898 386388 c 8:: 386:8
«89:: “8:38: 5: 868898 388388 m 86:: 828:8 «82:: ”6.80:8: 8m 68898 68.3: 386:8 «89:: ”68.88_

-65 86:8 comm: 3-2:/x :m m: mo :2: :8 86 38:8 65 :8: 8:86 E:8 z: 8 com: 3:8 QNN 336:. :EEE 65 m: 86 30:0

.<m3m @6888? x: 38988 863 Am .< .mev 860:8: mm: 6:6on $68: 8:68: :8: $6838 8808 8868: autumn:

no 888%: com 8 @8098 3:88:88: 863 €=8w$u5 8:8 .8 3:80 .33: .288 893:? 8 808.3 8568: 8 08898

88.83:“... 68.8%.. ”832—8 038 85.75 8 8838.. .3258 m3 #:6on 853:: we 8:89:80 ”Am—é: fin 98”:—

144

Figure 7.2

 

22

Control diet for fish oil (n=6)
7////A Chow diet (n=4)

‘V

Total plasma IgE level ( pg/ml)
:1

 

 

 

 

k

V I7]

 

Pre—immune 3rd response 6th response

Timepoint

Figure 7.2: Comparison of total IgE in BALB/c mice following repeated transdermal
exposure to hazelnut protein in different control diets. Groups of mice (n=4-6/group)
were transdermaly exposed to 500 ug/mouse of LPS-free hazelnut protein extracts and
total IgE levels were measured by optimized ELISA. Chow diet is the Harlan Teklad diet
used in all other studies and the control diet for fish oil is an AIN-93 based diet. Pre-
immune: plasma collected before exposure; 3rd response: plasma collected after 3
transdermal exposures; 6th response: plasma collected after 6 transdermal exposures. Data
shows total plasma IgE levels as uglml (mean +/- SE). At some points error bars are not

visible.

145

8:26 «58.:

88v: 80w: gov: coca: coop; com: 0mm:

 

 

q u — fi _ — _

P D > b m

     

owcoamm. Em Iqu
3.532 Em Iml
m 958:.me |®I|

 

 

 

 

 

ans .6 cm: .2 $6 .928

o

0805 Doom: 0003.. Doom: coo—2.

89 —

omQ w

 

 

— q u _ 4

3:88.. :5 'all
@2039 En [ml
mcaEE_-wE Imlu

 

 

<

 

 

 

Gus 66 265

O

N

(wu 069-9017 00) ewseld U! IeAeI qv ezofil amoeds munezeH

mg 23m:

146

.:0_: :0 :0: 3.: 38:00 Me .m> :03 30:0 M3 3.: modv: 0:300: :00: <>OZ< .0363 :0: 0:0 80: :080
3:3: 0800 :< .3m -\+ :008: D0 00 0:050: 0:500: 32:80 0:30am 35000: 030:0 «:09 00.80098 3:80:88: 0 6:0 :0:00:00
0800:: ”00:0:00: 50 ”00.80098 3:80:88: m 8:: :0:00:00 08003 ”00:800.: Em “0.80098 033: :0:00:00 08003 ”08:88:
-05 .:0:: :000: 3-22 :0 m: :0 :0: 3.: :03 38:00 0:: :8: 00380 :0::0 :0 8 :00: flmm :03 8020:. 82:02 0:: m: :03 30:0
49.5 :0Nm8:::0 x: :08308 0:03 Am .< .03”: 830:8: «NOE 0500mm w8:8: 8808:: :8: 0:08:08 80:08 88000: 00:-mmq
.3 008.08%: cow 0: :00098 30:80:88: 0:03 3:0:w3H5 00:8 3 09:80 .33: 35:00 820:3 8 80:0...— :=:_0N0: 0: 080098

38:05.82: :0:00:0._ w:_30=0.: 00:8 054<m 8 80:030.. 30:58: «~03 0500:: 28.000: 3 8023800 "Am—AG mg. 0.53.:

147

FMme7A

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Chow diet (n=5) Control diet for fish oil (n=6)
400 400 ‘
A B
300 "' 300 _
E
a) —.—
9: 200 - 200 -
Y
=1
100 ~ 100 :
0 1 0 m
0 100 0 100 500
Hazelnut concentration (ug/ml) Hazelnut concentration (ug/ml)

Figure 7.4: Comparison of different control diets: Hazelnut driven Type-2 cytokine
(IL-4) responses in mice transdermally sensitized with hazelnut protein

Groups of BALB/c mice (n=5-6/group) were sensitized with hazelnut (500 pg per mouse)
by transdermal exposure. Mice were on either a commercially available rodent chow diet
(Harlan Teklad 22/5) or an AIN-93 based diet. Three days following a booster exposure
with hazelnut, spleen cells were harvested and cultured with indicated dose of hazelnut
protein or culture medium alone. Cell culture supematants were harvested at different
time points over 5 days and analyzed for cytokines using optimized ELISA. Data shown
is peak cytokine response (average +/- SE of duplicate analyses of data) from hazelnut-

sensitized mice. (ANOVA test results: p<0.05 100 pg hazelnut stimulation in chow diet

vs. 100 pg hazelnut stimulation in control diet for fish oil).

148

Figure 7.5

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Chow diet (n=5) Control diet for fish oil (n=6)
5000 ,_/L_J 5000 _B__]
E
U)
3
: 2500 ” 2500“
z
&
fi—
0 m , 0 m
0 100 500 0 100 500
Hazelnut concentration (uglml) Hazelnut concentration (uglml)

Figure 7.5: Comparison of different control diets: Hazelnut driven Type-l cytokine
(IFN-y) responses in mice transdermally sensitized with hazelnut protein

Groups of BALB/c mice (n=5-6/ group) were sensitized with hazelnut (500 pg per mouse)
by transdermal exposure. Mice were on either a commercially available rodent chow diet
(Harlan Teklad 22/5) or an AIN-93 based diet. Three days following a booster exposure
with hazelnut, spleen cells were harvested and cultured with indicated dose of hazelnut
protein or culture medium alone. Cell culture supematants were harvested at different
time points over 5 days and analyzed for cytokines using optimized ELISA. Data shown
is D3 of cell culture (average +/- SE of duplicate analyses of data) from hazelnut-
sensitized mice. ANOVA test results: p<0.05 for 100 and 500 ug stimulation vs. 0 ug

hazelnut stimulation for both diets.

149

Figure 7.6

 

 

Hazelnut sensitized mice

 

 

 

 

 

 

 

 

 

 

 

 

 

        
   
     

 
     
   

090000

   

      

   

    

             

 

0999909999909909
0"0’0’9’0’0’9’0’9’9’93:93:39:
'099099990990990
090009909099999
“0.9.9.5900; 090 0 0 9

 

 

 

100

(0:121) 7"" M—Ndl

150

Hazelnut 100 ug Hazelnut 500 ug Saline Hazelnut 100 ug Hazelnut 500 ug

Chow diet (n

Saline

 

 

:6)

Control diet for fishvoil (n

:5)

Stimulation

:3 :0:: :0: :03 3:800 0830 .0> :03 30:0 0830 :0: modv: ”0:300: :00: 4: >0 >< A03: :0 000308: 0:00:30:

:0 mm -\+ 0w0:0>0: 3303880 :0:—0N0: w: cow ::0 w: cc: .0830 :0: 00:0:00: :02: :0 mm \ 0.2% :0 mm 0: :30:0 0:0Q .<m5m
308830 83: 0080330 :0: :0:»33 ::0 0x0: m :0>0 0:80: 08:: ::0:0::3 :0 :0:00>:0: 0:03 08038080 0:830 :00 A0830:
0:03 8:308 08:30 :0 803:: 3:300: :0 000: :0:0038 3:3 38:30 ::0 :0:00>:0: 0:03 0:00 :0030 £8300: ::3 0:000:08
3000: 0 8323: 0:0: 008:. :03 :000: 3-75:: :0 :0 GEN :0E0H :0::0m: :03 30:0 80:0: 03033 33338800 0 :0::3
:0 0:03 00:): 0.80098 38:30:08 x: 30008 :0:: w: 83 38000: :33 3:30:00 0:03 3:0:w\:-mu:: 00:8 05.7%: :0 093:0

0:03 3:800 .3 80020::800 "803...: 28300: :53 30050800 $383382: 00:8 8 8:0: :11: 157:: 3.: 0::—wt:

151

Figure 7.7

 

 

 

 

 

 

 

 

 

 

W kl t ol
ee y pro oc Preventative study: Feed respective test diets for 4 weeks prior to
sensitization and continue the entire experiment.
Thursday
Patch off
Feed
res ctive .
diet? for Monday Fnday Weekly bleedings. check
four weeks Patch on Bleed how IgE levels fall.
2"" Pre-Bleed Do weekly
and clip hair protocol 6
times
Mice arrive @
5 weeks 0‘ Boost animals with hazelnut or
898. Pre-bleed saline. Sacrifice animals. spleen cell
culture (cytokine determination).
collect ns for fa acid anal is.
Measure hazelnut orga W ys
specific lgE and confirm
positive response.
Group n= $nsitized with Dot fed prior to and during sensitization Diet fed after sensitization
1 5 Hazelnut Test Get Test diet
2 6 Hazelnut Control diet Control diet
3 5 $Iine Control diet Control diet

Figure 7.7: Transdermal sensitization protocol to hazelnut protein: preventative
study. The schedule of cycles of hazelnut sensitization and testing are shown for the

preventative study.

152

Table 7.l

Experimental diets used to assess DHA + EPA enriched oil on tree-nut allergyl

 

 

Diet Corn oil Oleic acid (n-3) enriched oil2
Lg/kg diet)

Control diet 10 60 0

Test diet (EPA + DHA) IO 37 23

 

1 All diets were adjusted with oleic acid to have final oil concentrations of 70g/kg.
2 Concentrated oils from fish that are enriched for DHA and EPA. The DHAzEPA ratio was 1:1.

153

Table 7.2

Fatty acid composition of experimental diets

 

 

 

 

Control diet Test diet (DHA + EPA enriched)
(g/kg diet) (g/kg diet)

Fatty acid
14:0 0.29 0.31
16:0 6.45 5.75
16:1 0.16 0.21
18:0 4.41 3.29
18:1 72.5 50.1
18:1 <0.10 0.85
18:2(n-6) 13.6 12.1
20:0 0.34 0.44
18:3(n-6) <0.10 <0.10
20:1 0.2 0.82
18:3(n-3) 0.32 0.53
20:2(n-6) <0.10 0.15
22:0 0.89 0.64
22:1 <0.10 <0.10
20:3(n-3) <0.10 0.11
20:4(n-6) <0.10 0.44
24:0 0.22 <0.10
20:5(n-3) 0.28 9.31
24:1 <0.10 <0.10
22:5(n-3) <0.10 1.64
22:6(n-3) 0.26 10.9
Total (n-6) 13.6 12.7
Total (n-3) 0.86 22.5
(n-6):(n-3) 16:1 1:1.8

 

 

 

* From Jia, Q., et al., J Nutr, 2004. 134(6): p. 1353-61.

154

Table 7.3

Spleen phospholipid fatty acid composition in mice fed various PUFA"2

Control diet + hazelnut Control diet + hazelnut Test diet + Hazelnut

g/100g total phospholipids fatty acid

Preventative study

 

16:0 20.25 +/- 2.21 18.89 +/- 1.98 16.07 +/- 1.84
1820 7.52 +/- 1.38 9.09 +/- 1.11 8.89 +/- 1.01
18:1 3.80 +/- 0.31 3.64 +/- 0.21 3.43 +/- 0.15
18:2 (n:6) 4.07 +/- 0.11 3.83 +/- 0.18 4.29 +/- 0.21
20:4 (n:6) 6.65 +/- 0.13 6.54 +/- 0.23 8.26 +/- 0.31 "
20:5 (n23) ND ND 3.14 +/- 0.22 “
22:6 (n:3) 1.18 +/- 0.15 1.48 +/- 0.09 1.99 +/- 0.12 “
Total n26 10.72 10.37 12.55
Total n:3 1.18 1.48 5.13
n26r’nz3 ratio 9:1 7:1 2.5:1

 

1 Only the major fatty acids are shown.
2 ND, not detectable.
3 ANOVA test results. ‘=p<0.05 vs. control diet fed animals.

155

OD (405 -690 nm)

OD (405 -690 nm)

3.00

2.50

2.00

1.50

1.00

0.50

0.00

3.00

2.50

2.00

1.50

1.00

0.50

0.00

 

 

 

 

Preventative study A

Pre -immune samples Figure 7.8

 

 

 

 

 

 

+ Control diet+saline [E53
“‘“ Control diet+HAZ

1E5:

“"“ Test diet + HAZ

 

W 4 “"

l l 1 1 l l l

 

20 40 80 160 320 640 1280
Reciprocal plasma dllutlon

 

 

 

+ Control diet+saline
l Preventative study CE
. Third response samples __ __ .
i\ ‘ Control diet + HAZ
\
.\ 4-- T... d... + HA2 CE

 

 

 

 

 

 

 

 

 

\.
\
\.
\
\. B
\.
\T
1\\
I \‘\
_L\ ‘ T
$.\
\\ \~
\ \.
\\~_ \ ...
_... ‘\ !-\ \
\“““-~_ \IT"‘-—-.
“"‘A ~~~~~~~~ ‘i—-~ ~ _.
y F t 4W 4* ““““ +4“; 7 " “ I.
l l 1 l l l l
20 40 80 160 320 640 1280

Reciprocal plasma dilution

156

 

 

Figure 7.8 (A-B): Characterization of hazelnut specific IgE antibody responses in
plasma from BALB/c mice fed the test diet enriched with EPA and DHA vs. control
diet (preventative study).

Groups of mice (n=5-6/group) were started on their respective diets and feed for 4 weeks
prior to being transdermaly exposed to LPS-free hazelnut protein extracts or saline as
shown in figure 7.7. Optimized ELISA measured hazelnut binding specific IgE
antibodies. Pre-immune: plasma collected before exposure; 3rd response: plasma collected
after 3 transdermal exposures; 6th response: plasma collected after 6 transdermal
exposures; 7th week of rest: plasma samples after seven weeks of rest following the 6
weeks of transdermal exposure. Data shows hazelnut specific IgE antibody levels as OD
(mean +/- SE). At some points error bars are not visible. These animals were on their
respective diets over the entire experiment (-4weeks, until end of experiment). ANOVA
test results: p<0.05 for graphs B, C, and D for test diet vs. control diet, p<0.05 for graphs

B, C, and D for both the hazelnut sensitized groups vs. control (saline) sensitized group.

157

(405-690 nm)

OD

OD (405-690 nm)

3.00

200

1.50

100

050

0.00

3.00

2.50

2.00

1.00

 

 

 

 

 

 

 

l “4"“ Control diet + saline n=5
Preventative study

\ Sixth response samples _ _‘__ Control diet + HA2 “:6
l “ .1

—-I--— Test diet + HAZ CE
\
I
' C

_L\

Flgure 7.8

 

 

 

 

 

1 4 l l l l l

 

20 4O 80 160 320 640 1280

Reciprocal plasma dilution

 

 

 

Preventative study + Control diet + saline

After 7 weeks of

 

 

 

 

 

 

 

 

 

 

rest samp|es — “‘ “ Control diet '0' HA2
_.-_._ Test diet + HA2
0
l \
\.
\.
\.
\.
\l
NE
SN A \
\ ‘\ ~\.
\ \ \ \.\
x‘\ \ ‘ ."\
\‘1—~-2;j‘\l~-~ \_\
‘ ‘ fi‘ ‘ - — T” "‘ "' ;‘ *- —- ;:;—_‘;~:; ' “r;':“:_"::‘;‘:
e 4* 4 % L J; i
1 l l l 1 1 l
20 40 80 160 320 640 1280

Reciprocal plasma dilution

158

 

Figure 7.8 (C-D): Characterization of hazelnut specific IgE antibody responses in
plasma from BALB/c mice fed the test diet enriched with EPA and DHA vs. control
diet (preventative study).

Groups of mice (n=5-6/group) were started on their respective diets and feed for 4 weeks
prior to being transdermaly exposed to LPS-free hazelnut protein extracts or saline as
shown in figure 7.7. Optimized ELISA measured hazelnut binding specific IgE
antibodies. Pre-immune: plasma collected before exposure; 3rd response: plasma collected
after 3 transdermal exposures; 6th response: plasma collected after 6 transdermal
exposures; 7‘h week of rest: plasma samples after seven weeks of rest following the 6
weeks of transdermal exposure. Data shows hazelnut specific IgE antibody levels as OD
(mean +/- SE). At some points error bars are not visible. These animals were on their
respective diets over the entire experiment (-4weeks, until end of experiment). ANOVA
test results: p<0.05 for graphs B, C, and D for test diet vs. control diet, p<0.05 for graphs

B, C, and D for both the hazelnut sensitized groups vs. control (saline) sensitized group.

159

 

83% meme;

omwx F

 

ooomtr coon: ooov: coon: oootw com:

    

 

mu: N<I + 56 uwmh llfli
on: N<I 123928 lml
MHC @E—mm + umzu _OLHCOO i®.i

 

 

 

 

 

 

89 9.063 n

ooow :P coon: ooov: ooomxw coo CF com:

omQ w

 

 

 

mu: N<I +86 63 lhl
: N<I + 86 6:80 lml

 

o

c 95mm + 56 6:80 m

m

 

 

 

 

 

mmcoamm: Em

o

259.: coon: ooov: ooow: coat?

com:

omQ w

 

   

 

     

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

H

e

z

W

U

m

.w

- . m.

H:

0

mn: N<I + $6 7.8 lal 6m

mu: N<I + 66 65:00 imil Z

mu: 65.3 + 66 63:00 l®l o W

N 0..

3:039 :5 W

cone: coca: gov: ooomfi coat. com: 8N; O .nIIUI

. . U

.. d

a.

S

w

e

m

G

i _. V

0

n...

9

mu: N<I 123mm: lal %

on: N<I + 86 6:50 lm+l u

mu: 953 + 66 .0550 l®l < M
N

oczEE_-9n.

mg 9:9“.

160

.9on continue. Aoczmmv _obcoo .m> 3:8» coNEmcomasfion: 65 53 How D can U 215%

SM modva “$3on 58 < >0 Z< .95 5:: mxooav¢ €085on 2:5 2: 5.5 $06 338%“: .55 so 203 mEEEw 085. .0363
Ho: 2a was hobo 85oz 2:8 2 .Amm -\+ 525 GO mm $05. 3,555 me 36.25 853m: 250% «:5 23898 395352“

me £83 0 2.: wEBoz£ “we Mo 8695 55m Saw 8388 «Emma um»: a8 x83 ex. $830me 388353 0 Sam 386:8 «Emma
Homaoamou 50 $8385 38635.5 m Home 83230 «833 ”828%“: Em ”2:898 283 380:8 «Emma ”0:595:65 .362:ch
«~03 #:0on wives 3532 85.0.38 <m5m wonEcQO .55 8st 5 F502...” we 25mm 8 30988 E805 558m: ova

-mmA 2 8898 32:3me mass 8 coca 8603 v 8m woo.“ Ba 36% groommoc :2: so @836 803 Aa=o&\o-mu5 8:: mo 3:80
63:3. 2555.6..5 3:. 35:8 .9» <=: 25 SE .53 toner—5 8:.

:8 2: co.— 8_E fins—«5 ER... «Ema:— E 82:38.. .3255“ «N03 958% 2.505:— .«e net-“3.88.230 "aéc as Bani

161

 

 

 

 

 

 

 

 

 

Figure 7.10 25
+ Control diet + saline
“A“ Control diet + HAZ
20 ”
E -I~-- Test diet+ HAZ
\
g) Preventative Study
V
LIJ 15 r
2’ ‘4' \.
/ / \\\~
m /' / \ \‘
E ' ll \ \
CD / l/ \\ \.
_‘E 10 ~ / / \ \\
Q / // \\ \
_ / // \\ \
m /’ // \\ \.\
H /' / \ ~ T
o T - ,x \ y
+- 4 x . .
5 .— /' I, \\ *
’/' * l/ \\
,// ,..4/ \‘
./‘,//”””
,z/ —o
0 l 1 1
pre—immune 3rd response 6th response after 7 weeks rest
Apply hazelnut epicutaneously 6X

Start feeding

Figure 7.10: Characterization of Total IgE responses in BALB/c mice fed the test
diet enriched with EPA and DHA vs. control diet. Samples from the preventative
study were analyzed for total IgE. Total IgE is expressed in ug/ml of plasma. ANOVA
test results: *= p<0.05 for test diet vs. control diet at 3rd response and after 7 weeks of

rest.

162

Figure 7.11

 

   

 

  

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

+ Control diet + saline + Control diet + HAZ + Test diet + HAZ
I n=5 I l n=6 I n=5
300 _ .
Hazelnut 100 ug/ml stimulation
Preventative study 4.
200 r
A
E
100 -
‘5, -
3
c:
'g o l
g Day 1 Day 3 Day 5
c
8 300 . .
: Hazelnut 500 ug/ml stimulation
0
0 3 *
‘1' 200 _
a
100 - :
o . A“ 4.
Day 1 Day 3 Day 5

Time of culture

Figure 7.11 (A-B) Hazelnut driven Type-2 cytokine (IL-4) responses in mice
transdermally sensitized with hazelnut protein: effect of experimental diets

Groups of BALB/c mice (n=5-6/group) were sensitized with hazelnut (500 pg per mouse)
or saline by transdermal exposure as described in Figure 7.7. Three days following a
booster exposure with hazelnut or saline, spleen cells were harvested and cultured with
indicated dose of hazelnut protein or culture medium alone. Cell culture supematants
were harvested at different time points over 5 days and analyzed for cytokines using
optimized ELISA..Data shown is peak cytokine response (average +/- SE of duplicate
analyses of data) from the fish oil preventative study mice (A, B). ANOVA test results:
for all comparisons of hazelnut sensitized mice vs. saline control p<0.05. *= p<0.05 for

test diet vs. control diet at marked time points.

163

Figure 7.12
+ Control diet+saline + Control diet+HAZ + Test diet+HAZ

n=5 I n=6 I n=5

 

5000
Hazelnut 100 uglml stimulation

Preventative study

  
   

 

A

 

 

2500

 

 

 

 

Day 1 Day 3 Day 5

 

 

 

 

 

lFN-gamma Concentration (pg/mL)
O

 

 

 

 

 

 

5000 . .
Hazelnut 500 uglml stimulation
B >l<
2500 r *
0 #1/
Day 1 Day 3 Day 5

Time of culture

Figure 7.12 (A-B) Hazelnut driven Type-l cytokine (IFNay) responses in mice
transdermally sensitized with hazelnut protein: effect of experimental diets

Groups of BALB/c mice (n=5-6/ group) were sensitized with hazelnut (500 ug per mouse)
or saline by transdermal exposure as described in Figure 7.7. Three days following a
booster exposure with hazelnut or saline, spleen cells were harvested and cultured with
indicated dose of hazelnut protein or culture medium alone. Cell culture supematants
were harvested at different time points over 5 days and analyzed for cytokines using
optimized ELISA. Data shown is peak cytokine response (average +/- SE of duplicate
analyses of data) from the fish oil preventative study mice (A, B). ANOVA test results:

*= p<0.05 for D3 and D5 test diet + HAZ vs. control diet + HAZ.

164

ecumSEzw 55mch .6 a: com

55on1 \ .06 “mm... 55$le \ _Ohcoo

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

0..

ON

cozmsEzm 55¢ch D6 0: oo_.

SEmNmI :06 ”mm...

5588.... \ ~06 65:00

 

 

 

 

 

33w 939555

 

 

 

 

 

 

 

 

 

S
01191 17"" ”‘le

0N
MFR $59“.

165

.:o:::_:::_:m 5508: w: com :0:: 00:

::0: :o: :0:: 6:88 .m> :0:: :8: :o.: m0dV: H ... 82:8: :8:-.:. 00:: .830 0388083 :0 :mm: 0:: 88.: A90: (:0 833:: 08023:: .:o
mm -\+ 0m80>3 8:22:85 3508: w: com ::0 w: oo— fio: :o: 0838: :1: .:o mm \ TZE .:o mm: m: 550:: «::0 {mim 88:53:
36: 8:388 :o.: c0838: :0:: 3:: m :0>0 $83 08:: 888.2: :0 8:88: 0:03 8:08.838 0:330 :00 80:83 3508:

.:o 08: 8:865 5:3 85:30 :5 8:88: 0:03 0:00 80:30 .5508: 5:3 8:038 8:80: a 338:8 3:: 00:2. .5: 0::wE

:_ 8:88: 8 8:838 38038:: 3 080:: :3 w: 003 5508: 5:3 88:88 0:03 3:89:88: 08:: 05.35 .8 3:80

30:: 3:88:38 .3 80:0 "50:93 3508: 5?» 88:30.0. bis—8:38: 00:8 5 2:8 :2: \ rule: "2.: 0.53.:—

166

Figure 7.14

Weekly protocol

 

Therapeutic study: Apply patch six times. after an established lgE response
is seen start feeding the experimental diets and continue the rest of the

Thursday experiment.
Patch off

Monday Friday
Patch on Bleed

 

 

I Weekly bleedings. check
Rest 1 week / / . how lgE levels fall. ’
Pre-Bleed / clip Do weekly I
ha" P'°‘°C°' 6 Start feeding the
times

Mice arrive @
6 weeks of age

Group

'1:

respective diets (control
diet and test diet),

 

Boost animals with hazelnut or

saline. Sacrifice animals. spleen cell

culture (cytokine determination).
collect organs for fatty acid analysis.

Measure hazelnut

specific IgE and confirm

positive response.

 

Sensitized with Diet fed prior to and during sensitization Diet fed after sensitization
Hazelnut Chow Test diet
Hazelnut Chow Control diet
Saline Chow Control diet

Figure 7.14: Transdermal sensitization protocol to hazelnut protein: therapeutic

study. The schedule of cycles of hazelnut sensitization and testing are shown for the
therapeutic study.

167

Table 7.4

Spleen phospholipid fatty acid composition in mice fed various PUFA"2

Control diet + hazelnut Control diet + hazelnut Test diet + Hazelnut

g/100g total phospholipids fatty acid

Therapeutic study

 

16:0 1629 +/- 2.98 l 1.09 +/~ 2.38 22.22 +/— 1.82
18:0 6.56 +/- 1.14 7.4] +/- 0.79 8.83 +/- 0.92
18:1 2.94 +/- 0.23 4.68 +/- 0.49 4.57 +/- 0.28
18:2 (n:6) 2.92 +/- 0.15 4.07 +/- 0.21 5.0] +/- 0.31
20:4 (n:6) 4.48 +/- 0.21 4.07 +/- O. 12 9.23 +/- 0.42 "
20:5 (n23) ND ND 4.62 +/- 0.34 "‘
22:6 (n:3) 0.94 +/- 0.08 0.99 +/- 0.” 1.92 +/- 0. l4 ‘
Total n:6 7.4 8.14 14.24
Total n23 0.94 . 6.54
n:6/n:3 ratio 8:1 8:] 2.2:]

 

1 Only the major fatty acids are shown.

2 ND, not detectable.

3 ANOVA test results, *=p<0.05 vs. control diet fed animals.

168

OD (405-690 nm)

OD (405-690 nm)

3.00

2.50

2.00

1.00

0.50

0.00

3.00

2.50

2.00

1.50

1.00

0.50

0.00

 

 

 

Therapeutic study
Pre-immune samples

+ Control diet+saline
“““ Control diet+ HAZ

 

 

 

 

 

 

 

 

 

 

 

 

—‘I'“ Test diet+HAZ n=
A
Figure 7.15
40 80 160 320 640 1280 2560
Reciprocal plasma dilution
1.
\.
\.
\l
‘\ K + Control diet+saline
\\\ ‘\‘
\\i \\ ““‘ Control diet+HAZ
A .
\ \
B .l \\ *5“ Test diet+ HAZ

 

 

 

 

 

 

 

 

 

 

 

 

\ \ Therapeutic study
\ \\ sixth response samples
l \\ \,
\\\_{
\\I .
X.\\
.L \\ .\ T
\Ex.
E“~ \ \
\\‘;1\~: \_
* ‘4!
o F a——-————c 4 + ‘0
A 1 1 l l J 1
20 40 80 160 320 640 1280

Reciprocal plasma dilution

169

 

Figure 7.15 (A-B): Characterization of hazelnut specific IgE antibody responses in
plasma from BALB/c mice fed the test diet enriched with EPA and DHA vs. control
diet (therapeutic study).

Groups of mice (n=4/ group) were transdermaly exposed to LPS-free hazelnut protein
extracts or saline as shown in figure 7.14. Optimized ELISA measured hazelnut binding
specific IgE antibodies. Following transdermal exposure groups of animals were given
their respective diets. Pre—immune: plasma collected before exposure; 6‘h response:
plasma collected after 6 transdermal exposures; 4th week of rest: plasma samples after
four weeks of rest following the 6 weeks of transdermal exposure; 7th week of rest:
plasma samples after seven weeks of rest following the 6 weeks of transdermal exposure.
Data shows hazelnut specific IgE antibody levels as OD (mean +/— SE). At some points
error bars are not visible. These animals were on their respective diets only after the

sensitization process was over.

170

00 (405-690 nm)

OD (405-690 nm)

3.00

2.50

2.00

1.50

1.00

0.50

0.00

3.00

2.50

2.00

1.50

1.00

0.50

0.00

 

 

 

. . = ,
° COWG' d'el + saline Therapeutic study

after 4 weeks
"‘-- Control diet + HAZ of feeding

 

 

 

 

 

 

 

 

 

 

 

 

“l”— Test diet + HAZ
C
Figure 7.15
b-
\\ \
i ‘~~-;‘£
\E‘\‘
\\ \.
\\\ _‘ ~
‘5 ‘\\\:;::~'\\
..... T ;*LT;\ \
~ ‘ ‘ ;t":;Ti-T 3::
.———————o——————-—% o r .——————-—'
1 1 J l 1 l 1
20 4O 80 160 320 640 1280
Reciprocal plasma dilution
Therapeutic StUdY + Control diet + saline n=4

after 7 weeks

 

 

 

 

 

 

 

 

 

0f feeding ““‘ Control diet+HAZ
D 4— ”Salaam
‘LTL‘It‘C‘q‘Q
3»sz
“*‘fiu‘h “h‘
l l l l l L J
20 4O 80 160 320 640 1280

Reciprocal plasma dilution

171

 

 

Figure 7.15 (GB): Characterization of hazelnut specific IgE antibody responses in
plasma from BALB/c mice fed the test diet enriched with EPA and DHA vs. control
diet (therapeutic study).

Groups of mice (n=4/ group) were transdermaly exposed to LPS-free hazelnut protein
extracts or saline as shown in figure 7.14. Optimized ELISA measured hazelnut binding
specific IgE antibodies. Following transdermal exposure groups of animals were given
their respective diets. Pre-immune: plasma collected before exposure; 6th response:
plasma collected afier 6 transdermal exposures; 4th week of rest: plasma samples after
four weeks of rest following the 6 weeks of transdermal exposure; 7th week of rest:
plasma samples afier seven weeks of rest following the 6 weeks of transdermal exposure.
Data shows hazelnut specific IgE antibody levels as OD (mean +/- SE). At some points
error bars are not visible. These animals were on their respective diets only after the

sensitization process was over.

172

 

8:26 Emma

88?: coon: oociw 88: 08.5. 08:. 8N2.

 

 

 

$0 0 G G G G

      
 

 

D

V": N<I + “EU 53.
an: N<I 1.4% 6:80 ImT
V": ®C=mm + “$6 _ObCOO l¢ll

 

 

 

 

 

 

 

 

9.608 9.095 v Eta
88:9 88: 08v: Sow: 83: com: 8w:
4
Li

vuc N<I +36 “wok ilfll
an: N<I + as .928 lml
VHC mph—mm + H@_U _ObCOO Ili®l m

 

 

 

 

 

 

mmcoomm. 5v

0

 

 

 

 

 

 

 

 

 

 

 

   

 

 

 

 

88:9 88: 803.. ooow? 08?: com: 0mm; 0

OF G G 0 Or 291

H

e

Z

W

U

m

x

.. .

H:

0

an: N<I+z238e lei mould

vnc N<I+um_u_obcoo lml Z

v": mc__mm+flm_u.obcoo l 0 pucv

N a.

3:039 50 w.

e

ooolmtp 88:. 83> coon: 89: com: 8N: O WI.

_ .. lcl .il/ Wu:

8

S

w

e

O

G

1 r .7

0

n...

m

V": 211238;. l 0

en: N<I+Hm=2obcoo lml u

vn: mc__mm+am_n_o:coo l®l < (w\

N

 

 

 

cascaded

o: 9:9”.

173

.03 003 0008a 5:320:00 05 00am Eco £06 038090.00 :0:“ so 0003 £255 0005. .0363 Hoe 000 EB 0800 8:60 088
2 .35 -\+ 500.5 90 mm m_0>0_ .6893 «NOE 0Eo0Qm 550$: 026:0 80D 05898 $800855 .«o 3003 c 05 was/0:8 30:0
3:09.598 3:00.“ Mo 8003 .58 00.00 003800. 08003 “@5000.“ .8 0x003 v 00% 8030098 380032000 0 00cm 088:8 «Emma
”033000 so ”0030090 360000820 e 000% 00000.30 08003 “083000 5v M05890 8803 0080.30 08003 ”05:55-05 00:02:30

«NOE 2.060% @5005 350$: 0050008 <m3m 030550 .36 05mm E 550% 00 0:200 00 $0098 E085 Han—0N0: 00¢
-mmq 2 000098 3080000505 mass 2 Sta $003 v no.“ 000% 0:0 306 03003000 :05 :o 00:80 0003 Escawénev 00:: .«o 8:80
.35: 30:252.: 3:. .238 a» SS Ea S:— e? 85:5 .20

000. 05 .00.— 085 0544m— Eeb «Ema—a E 002530.. 3.25:0 «~03 0500mm 0::—0N0: he :euafiuouuuaanu "AA—-43 03. 0.5a:—

174

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

25
+ Control diet + salin-
Figure 7.17 Therapeutic Study
"A" Control diet + HAZ
A20 _ /.§{‘*" Test diet + HAZ ”:4
E /,///.|— \\ \ \
\ , // ‘\
D) /// \\\ \ \ *
a / .\ T
15 r / / \ ‘
“a, / x ix
_ l’ \\ ~\ 3|:
(0 / l/ \ ‘\
E /' / \\ \‘\
a) /. / \T .\ l
_g 10 ~ /'/ >1\ 1
o. ,/,/ \\ l
a o/// \\\\
a—o ,/,’ \\\ T
O // ‘A
'— L /'/ J-
5 //
/
,/
/
,/
ll
0 i We 9* 9
pre-immune 4th response 6th response after feeding 4 weeks after feeding 7 weeks
Apply hazelnut epicutaneously 6X

Start feeding

Figure 7.17: Characterization of Total IgE responses in BALB/c mice fed the test
diet enriched with EPA and DHA vs. control diet (therapeutic study). Samples from
the therapeutic study were analyzed for total IgE. Total IgE is expressed in ug/ml of
plasma. ANOVA test results: * = p<0.05 for test diet + HAZ vs. control diet + HAZ for

the after 4 weeks of feeding and after 7 weeks of feeding samples.

175

Figure 7.18 + Control diet+saline + Control diet+HAZ + Testdiet+HAZ
I n=4 l l n=4 | I n=4 |

 

 

 

 

 

 

 

 

 

 

  
   

 

 

 

 

300 _
Hazelnut 100 uglml stimulation
Therapeutic study * *
200 a
A
CT
is 100 -
c»
9;
a:
1g 0 #1 4
‘3 Day 1 Day 3 Day 5
E
m 300 , , _
g Hazelnut 500 uglml stimulation *
<3
0 B
*

‘f 200 -
2‘

100 ”

 

 

Day 1 Day 3 Day 5
Time of culture

 

 

Figure 7.18 (A-B) Hazelnut driven Type-2 cytokine (IL-4) responses in mice
transdermally sensitized with hazelnut protein: Effect of experimental diets.
Groups of BALB/c mice (n=4/group) were sensitized with hazelnut (500 pg per mouse)
or saline by transdermal exposure as described in Figure 7.14. Three days following a
booster exposure with hazelnut or saline, spleen cells were harvested and cultured with
indicated dose of hazelnut protein or culture medium alone. Cell culture supematants
were harvested at different time points over 5 days and analyzed for cytokines using
optimized ELISA. Data shown is peak cytokine response (average +/- SE of duplicate
analyses of data) from the fish oil therapeutic study mice (A, B). ANOVA test results:
for all comparisons of hazelnut sensitized mice vs. saline control p<0.05. Also, * =

p<0.05 for comparisons of test diet + HAZ vs. Control diet + HAZ for D3 and D5.

176

 

 

 

 

 

F'Qure 7-19 —O— Contrlll"il+saline+ Cont'mdjslg“ HAZ + Tes ' +HAZ
m. n=4 n=4

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

5000 . .
Hazelnut 100 uglml stimulation
A Therapeutic study
_I
E
g 2500 ”
C
.9
‘5
‘5
g 0 Day 1 Day 3 Day 5
O
5000
g Hazelnut 500 uglml stimulation
g a
C.”
2
EL. 2500 r * *
fl
0 Day 1 Day 3 Day 5

Time of culture
Figure 7.19 (A-B) Hazelnut driven Type-1 cytokine (IFN-y) responses in mice

transdermally sensitized with hazelnut protein: Effect of experimental diets.

Groups of BALB/c mice (n=4/group) were sensitized with hazelnut (500 pg per mouse)
or saline by transdermal exposure as described in Figure 7.14. Three days following a
booster exposure with hazelnut or saline, spleen cells were harvested and cultured with
indicated dose of hazelnut protein or culture medium alone. Cell culture supematants
were harvested at different time points over 5 days and analyzed for cytokines using
optimized ELISA. Data shown is peak cytokine response (average +/- SE of duplicate
analyses of data) from the fish oil therapeutic study mice (A, B). ANOVA test results:
for comparisons of test diet + HAZ vs. control diet + HAZ, * = p<0.05 for D3 and D5 for

500 ug stimulation and D3 for 100 ug stimulation.

177

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

00:03:50 550ch *0 as com 530.3630 550ch 00 a: cc?
5505...: 2% .8» 5508: \ as .380 m 550%: \ .06 a3 5508: \ as .380 m
ILII
r i
*
Ir .0
- r N
b
r or r or _H
V
w
m.
Ilrl
*
m >030 ongoomuoch <
ON ON

 

 

02 050E

178

8888283 2.8on: w: com 88 o2 Eon you modvm N<I + 86 .8800 .m> N<$ +

66 “mob M832 $89 .088 >83 883805 :0 8mm 08 808 A88 .80 mug—«8w 83:96 .80 mm -\+ owfiozwv 888885 8838 w:
com 88 w: o3 Eon .5.“ 88888 wk: mo mm 1-2%.? mm 8 8305 8mm {mim B88880 w8m: 88895 .5.“ 353mg new $38 m
b>o $83 088 880.88 “a @2388 803 358833 0883 :00 .283 8838 0885 Ho 8308 880ch we 30v @2868 :83
3883 van 82882 803 £3 52% .088 Ho 8838 883 08898 swoon m w8>>o=£ 3% 88H .36 0.5me 8 388ch

mm 088on 88.8388 3 08—8 Ho 3258 8m w: 003 88088 883 88883 803 Esopwéucv 088 o\mq<m mo 8:80

.8395 88mg: :83 3882.8 838.5352“ «88 8 8:: wk: 1‘-th 5mg. 0.58%

179

7.7 References

1.

10.

11.

Prescott, S.L. and RC. Calder, N-3 polyunsaturated fatty acids and allergic
disease. Curr Opin Clin Nutr Metab Care, 2004. 7(2): p. 123-9.

Simopoulos, A.P., Essential fatty acids in health and chronic disease. Am J Clin
Nutr, 1999. 70(3 Suppl): p. 5608-5698.

Dunstan, J .A. and S.L. Prescott, Does fish oil supplementation in pregnancy
reduce the risk of allergic disease in IFNants? Curr Opin Allergy Clin Immunol,
2005. 5(3): p. 215-21.

Hodge, L., et al., Consumption of oily fish and childhood asthma risk. Med J
Aust, 1996. 164(3): p. 137-40.

Leung, D.Y., et al., Eflect of anti-IgE therapy in patients with peanut allergy. N
Engl J Med, 2003. 348(11): p. 986-93.

Birmingham, N., et al., Hazelnut Allergy: evidence that hazelnut can directly

elicit specific IgE antibody response via activating type 2 cytokines in mice. Int
Arch Allergy Immunol, 2005. 137(4): p. 295-302.

Reeves, P.G., F.H. Nielsen, and G.C. Fahey, Jr., AIN.93 purified diets for
laboratory rodents: final report of the American Institute of Nutrition ad hoc
writing committee on the reformulation of the AIN- 76A rodent diet. J Nutr, 1993.
123(11): p. 1939-51.

J ia, Q., et al., Docosahexaenoic acid and eicosapentaenoic acid, but not alpha-
linolenic acid, suppress deoxynivalenol-induced experimental IgA nephropathy in

mice. J Nutr, 2004. 134(6): p. 1353-61.

Birmingham, N., S. Thanesvorakul, and V. Gangur, Relative immunogenicity of

commonly allergenic foods versus rarely allergenic and nonallergenic foods in
mice. J Food Prot, 2002. 65(12): p. 1988-91.

Birmingham, N., et al., An ELISA-based method for measurement of food-specific
IgE antibody in mouse serum: an alternative to the passive cutaneous anaphylaxis
assay. J Immunol Methods, 2003. 275(1-2): p. 89-98.

Rempel, J .D., I.P. Lewkowich, and KT. HayGlass, Endogenous IL—12 synthesis is

not required to prevent hyperexpression of type 2 cytokine and antibody
responses. Eur J Immunol, 2000. 30(2): p. 347-55.

180

 

 

 

 

 

 

 

12.

13.

14.

15.

Hasler, C.M., J .E. Trosko, and MR. Bennink, Incorporation of n-3 fatty acids
into WB-F 344 cell phospholipids inhibits gap junctional intercellular
communication. Lipids, 1991. 26(10): p. 788-92.

Prickett, J .D., D.R. Robinson, and K]. Bloch, Enhanced production of IgE and
IgG antibodies associated with a diet enriched in eicosapentaenoic acid.
Immunology, 1982. 46(4): p. 819-26.

Hsieh, K.Y., et al., Oral administration of an edible-mushroom—derived protein
inhibits the development of food-allergic reactions in mice. Clin Exp Allergy,
2003. 33(11): p. 1595-602.

Li, X.M., et al., Food Allergy Herbal Formula-1 (FAHF—I) blocks peanut-induced
anaphylaxis in a murine model. J Allergy Clin Immunol, 2001. 108(4): p. 639-46.

181

CHAPTER EIGHT

8.0 Working model and future studies

From this work, the working model so far is hazelnut applied to the skin of mice can be
taken up and presented via MHC II by Langerhans’ cells to CD4+ Th2-cells. This causes
the CD4+ Th2-cells to produce IL-4, the class switch factor for IgE synthesis, causing B-
cells to make IgE specific to hazelnut. This IgE specific to hazelnut can then bind to
mast cells and basophils. Upon oral allergen exposure, the IgE on these mast cells and
basophils is cross-linked causing them to degranulate, releasing mediators such as

histamine. These mediators cause the signs and symptoms of systemic anaphylaxis.

With a well-characterized model of tree-nut allergy, there are further studies that can now
be proposed. 1) Determine the molecular mechanisms underlying tree-nut allergy, using
gene knockout mice (e. g. stat KO) could assess the role of STAT6 in Th2 development as
well as systemic anaphylaxis. Also different techniques could be used, for example
RNAi (RNA interference) could be used to determine the exact signals needed to trigger
an allergic response. Using RNAi the role of GATA3 could be assessed to determine if
this model is GATA3 dependent. 2) With this model more studies could be done
assessing therapies of food allergy, either in a dietary role (n-3 fatty acids, pro-biotics) or
even a pharmaceutical avenue, possibly uncovering different approaches that could be
used to slow down or reverse the increasing trend of food allergy. After the n-3 (DHA
and EPA) study was done, we further characterized the systemic anaphylaxis and
threshold dose to cause an anaphylactic response, now we could determine if DHA and

EPA enrichment could affect clinical responses. 3) The effect of different food

182

processing techniques on tree-nuts could be assessed to see if tree-nuts could be
processed in a certain way to make them less allergenic, therefore trace amounts could
possibly be tolerable to allergenic patients. 4) Using a similar approach that we used for
hazelnut, the allergenicity of various genetically modified foods could be assessed.
Using a number of strains of mice and an array of doses, the relative allergenicity could
be determined. With this model both phases of allergy can be studied, the sensitization
phase, as well as the effector phase. If something can be done to decrease either phase,

there would be a win in the fight against food allergy.

183