in.”
. $3.

. EH
Lug-$1.... 31:.” .

4.1“er s2! 9&1.
n

,2!

?

1!! I w 6-. P"
r. a... G!
...um.mm.fi..a..fi .
. . lo. 7 .
mWMWM: ‘ . Lie...
Ewan”... 4%...” A
‘ A I .\
3.50451?! A

«an wwfiuugr.

1.: d!

.15
1.... .

Ixxlkcg.

.nflnermq: ..
JR.

.. 3. a i
A...
i {fluid}

:54. ‘
lit}...
4 .

lit .. vL
0x4.

:25?!

 

male

200?

This is to certify that the
dissertation entitled

ALLERGIC DISORDERS
AND
SEX HORMONE DISRUPTERS

presented by

KEVIN ROYD BROOKS

has been accepted towards fulfillment
of the requirements for the

Doctoral degree in Epidemiology

 

 

O m L__,,

 

Major Professor’s Signature

January , 2007

 

Date

MSU is an Affirmative Action/Equal Opportunity Institution

._.—.-.-._._.__._._.__-.-.__-.-._.___,_._._,-_-._._,_._.-._,_._.-.-.-._.-.-.-._.

-- -—._----.—._._-—._._.-—-‘—.-.--nuu

 

 

LiBRARY
Michigan State
University

 

 

 

PLACE lN 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

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

6/07 plelRCIDateDue.indd-p.1

 

 

 

EJ/f

 

ALLERGIC DISORDERS AND SEX HORMONE DISRUPTERS

By

Kevin Royd Brooks

A DISSERTATION

Submitted to
Michigan State University
In partial fulfillment of the requirements
for the degree of
DOCTORAL DEGREE
DEPARTMENT OF EPIDEMIOLOGY

2007

ABSTRACT

ALLERGIC DISORDERS AND SEX HORMONE DISRUPTERS

By

Kevin Royd Brooks

Allergic disorders are among the most disruptive of childhood. Further, the

prevalence and severity of these disorders are increasing.

Great strides have been made toward disentangling the etiology of allergies;
however a clear understanding is still elusive. Recently, the role of sex hormone
disrupters (SHD) in triggering childhood allergies have gained increased
attention. Specifically, halogenated organochlorine compounds (HOCs) such as
dichlorodiphenyl dichloroethene (DDE) and polychlorinated biphenyls (PCBs) in
addition to synthetic hormones such as estrogen and progesterone in the form of

contraception have been implicated.

We investigated the role of SHD in the etiology of allergic diseases. First we
investigated the association between placental p,p’-DDE and cord blood immune
markers. p,p'-DDE concentration and cord plasma interleukin (IL)-13, lL-4 and
INF-y were determined in 19 neonates from the ongoing Pregnancy Environment
and Child Health (PEACH) study. Results showed increased placental p,p’-DDE
to be associated with a statistically significant increase in cord plasma interleukin
(IL)-13. Furthermore, both cord plasma IL-4/lNF-y and lL-13/lNF-y ratios were

significantly positively associated with placental p,p’-DDE concentration.

 

 

(‘2

F .

Second, we estimated associations between maternal oral contraceptive (OCp)
use and humoral immune markers in offspring. We hypothesized that maternal
OC use increases humoral immune markers of allergy in offspring. Toward this
end, data from a cross-sectional investigation including 334 mother child (aged
7—10 years) pairs from Hesse, Germany were used. Results showed that female
offspring of mothers who used 00 had significantly lower (p<0.05): lgA (123.43
mg/dL vs. 150.52 mg/dL), and lgE (22.96 kU/L vs. 50.83 kU/L) levels as well as
basophilic surface lgE counts (783 vs. 842), compared to those of mothers who
did not use 00. For male offspring, statistical significance was only seen in an

increased number of basophilic surface lgE (911 vs. 876).

Third, using data from a 1986—1987 survey of 11-12 years old Jamaican children,
an investigation of the association between maternal OCp use before pregnancy
and childhood allergic phenotypes suggest that offspring’s asthma or wheezing
may be related to 00p use (aOR: 1.81; 95% CI (1 .25- 2.61)).

Fourth, the relationship between childhood DDE and immune markers was
quantified using data from the above-mentioned Germany cohort.

Higher DDE concentration was significantly related to increased serum lgE

levels, lgE count on basophils, and the reduced eosinophilic granula.

These findings suggest that SHD may be important in the in etiology of childhood
allergy. Of note, SHD may play a role in the prenatal priming of childhood allergy.
Exploring the specific immune mechanism behind the possible effect of SHD on

allergy may be worthwhile.

DEDICATIONS

This work is especially dedicated first to my wife Valrie, daughter Nessia, son
Kevin and mother Patricia. I also dedicate this monograph to my very promising

nephew Jevon 0. Brooks.

iv

ACKNOWLEDGEMENTS
The completion of this work is the result of a team of enthusiastic and devoted
individuals. Sincere gratitude to my advisor, Dr. Wilfried Karmaus, for his

prudence, and moral support in the process of completing this work.

I also thank the other members of my committee, Drs. Joseph Gardiner,
David Todem and Venugopal Gangur for taking time from their busy schedule

to provide unswerving intellectual support.

For their many years of encouragement, example and technical support, I
thank Drs. Maureen Samms-Vaughn and Rainford Wilks. It would be remiss
of me not to acknowledge my collegues Kathy, Ali and Azfar for their empathy

and friendship.

This work would not be possible without the contribution of all study

participants. I thank you all for your time and patience.

TABLE OF CONTENTS

LIST OF TABLES ............................................................................................. Vii
LIST OF FIGURES ........................................................................................... ix
OVERVIEW ......................................................................................................... 1

CHAPTER1
ALLERGIC DISORDERS AND SEX HORMONES DISRUPTERS ................... .5
11Allergy ..........6
1.2Asthma .......................................................................................... 7
1.3Sex hormones disrupters 21
1.4Time-windowofexposure.. 28
1. 5Components of the conceptual model ......29
1. 6Literature review of components comprising the conceptual model ...... 31
References” .36

CHAPTER 2
BACKGROUND AND METHODS FOR THE RESPECTIVE STUDIES ........... .50
2.1 The PEACH study. .. .51
2. 2 Child Health and Environment Cohort Study Germany ..................... 59
2. 3 The Jamaican Perinatal Morbidity, Mortality Survey. .. .61
References. . .. . . 65

CHAPTER3

RESULTS ............................................................................................. 66
Component one ................................................................................ 68
References ....................................................................................... .78
Component two 79
References ...................................................................................... 92
Component three .102
References ................................................................................... .124
Component four .........136
References .................................................................................... 158

CHAPTER4
Future work 174
References 176

vi

 

CO
In L

T’

-r

LIST OF TABLES

COMPONENT ONE
In utero exposure to SHD affects perinatal biomarkers of allergy

Table 1. Descriptive characteristics of the cohort .................................... 76

Table 2. Association (betas ((3) and standard errors (SE)) between
p,p'-DDE concentrations and lL—4, lL-13, INF-y, and the ratio
of lL-4/ INF-y, and lL-13/ INF-y ................................................... 77

COMPONENT TWO
In utero SHD affects postnatal biomarkers of allergy

Table 1. Descriptive characteristics of the children’s cohort ............... 95.96

Table 2: Gender specific geometric mean and 5-, 95 values for
immunoglobulins by covriates. ................................................. 97

Table 3: Gender specific geometric mean and 5-, 95 values for

white blood cell, eosinophilic characteristics, and basophilic
surface lgE by covriates. ...................................................... 98,99

Table 4: Maternal oral contraceptive use by immunoglobulins .............. 100

Table 5: Maternal oral contraceptive use by white blood cell,
eosinophilic characteristics, and basophilic surface lgE .......... 101

COMPONENT THREE
In utero exposure to SHD affects allergic outcomes in infancy

Table 1: Comparison characteristics of exposure variables in baseline
and linked datasets ................................................................. 129

Table 2: Proportions of simultaneous occurrence of the different
atopic manifestations ............................................................... 130

Table 3: Proportion of atopic disorders in relation to their in utero
risk factors .............................................................................. 131

Table 4: Proportion of atopic disorders in relation to their perinatal
potential risk factors ........................................................ 132, 133

Table 5: Odds ratios and 95% confidence intervals (CI) of antenatal

and perinatal risk factors for asthma or wheezing, and
frequent nighttime or early morning cough .............................. 134

vii

Table 6.

LIST or TABLES (CONTINUED)

Odds ratios and 95% confidence intervals (CI) of antenatal
and perinatal risk factors for eczema, and hay fever or sinus

problem or some other allergy .............................................. 135
COMPONENT FOUR
Postnatal exposure to SHD affects biomarkers of allergy in infancy
Table 1: Descriptive characteristics of the study cohort. ....................... 162
Table 2: Geometric mean and 5-, 95% values for whole blood OC
and Pb by covariates ............................................................... 163
Table 3: Spearman correlation coefficients between organochlorine
compounds and their geometric means. ................................. 164
Table 4: White blood cell, eosinophilic characteristics, and basophilic
surface lgE by DC and Pb (geometric mean) .......................... 165
Table 5: Lymphocyte phenotypes by whole blood DDE, PCBs, HCB,
y-HCH and Pb concentration (geometric mean) ...................... 166
Table 6: lmmunoglobulins by whole blood DDE, PCBs, HCB, y—HCH
and Pb concentration in children (geometric mean) ................ 167

viii

LIST OF FIGURES

Figure 1: Sex Hormone Disrupters and Childhood Allergy —

Conceptual model .................................................................... 2
Figure 2: Various definitions of asthma ..................................................... 7
Figure 3: Theoretical immune responses .................................................. 8
Figure 4: Map of Michigan showing PEACH Study Site .......................... 52
Figure 5: The Rhine Valley ...................................................................... 59

Figure 6: Jamaica: parishes, regional administrative areas and
Distribution of hospitals ............................................................ 62

Figure 7: Diagrammatic representation of the breastfeeding, childhood
exposures and immune markers associations ....................... 146

Figure 8: The combined effect of increasing DDE and lead (Pb) on
lgE serum levels .................................................................... 150

ix

 

(I)

I)

(D

OVERVIEW

Childhood allergies are among the most vexing diseases: purported to be
responsible for enormous personal and public health burden. In spite of
numerous efforts, the etiology of these diseases continues to evade researchers.
The purpose of this monograph is to investigate the role of sex hormone
disrupters (SHD) in the etiology of allergic diseases. Specifically, this work
explores the prenatal origins of childhood allergies (in utero programming) by
probing the possible role of SHD in this process. Next, the association between

exposure to SHD and biomarkers of allergic susceptibility in infancy is assessed.

The quest for appropriate pieces of the SHD-allergic disorders puzzle started with
the development of a conceptual model (Figure 1; more details provided in the
“Components of the conceptual model” section). This model depicts areas that
speak to the purpose of epidemiology, which is to provide clues to the cause of a
health related problem. An inventory of suitable data was then undertaken and
the following four components developed.

1. In utero exposure to SHD affects perinatal biomarkers of allergy.

2. In utero SHD affects postnatal biomarkers of allergy.

3. In utero exposure to SHD affects allergic outcomes in infancy.

4. Postnatal exposure to SHD affects biomarkers of allergy in infancy.

"Y!

Chil:

b...
.VfV-

5‘
(D

 

.D-PC Vi;

 

 

      

 

 

 

 

. Serum
rogesterone
markers
In utero and childhood immune development
Pregnancy (in utero environment) Childhood

 

 

 

 

  
 

   
   

Cord
plasma
IL-13, lL-4
& INF-y

  
  

Placental
OC

    
 
 

Immune
markers

  

 

 

 

* Maternal oral contraceptive use before pregnancy

 

Figure 1 Sex Hormone Disrupters and Childhood Allergy - Conceptual model

Data from three cohorts were used to address the above components. These
cohorts are: the ongoing Pregnancy Environment and Child Health (PEACH)
study from which information on placental dichlorodiphenyl dichloroethene (DDE)
concentration and cord plasma interleukin (lL)-13 were derived (component 1).
Data on maternal oral contraceptive use and on childhood biomarkers of allergic
susceptibility (lmmunoglobulins (lgs), white blood cell, eosinophils, and
basophilic surface lgE) were obtained from the Child Health and Environment
Cohort Study conducted in 1995 (component 2). This study also provided data on
childhood organochlorine exposure and the above mentioned childhood
biomarkers of allergic susceptibility (component 4). A geographic sub-cohort from
the Jamaican Perinatal Morbidity, Mortality Survey conducted in 1986-1987

provided data on maternal oral contraceptive use before pregnancy and

childhood biomarkers of effect such as asthma, wheezing and eczema
(component 3). Detailed description for each respective study is presented in

Chapter 2.

Ethical approval for this work was obtained from the Institutional Review Board of

the Michigan State University.

This first chapter presents the definition and immunology of asthma, its
prevalence and burden, risk factors and the possible effects of various SHD (sex
hormone disrupters) on childhood allergy. Next, the importance of time windows
of exposures and the main concepts that may explain the increase in the
prevalence of asthma are introduced. Then the conceptual model with its four
components (see above) followed by their respective backgrounds are
presented. Chapter 2 - ‘Methods’ - describes different aspects of the materials
and methods not discussed comprehensively in the manuscripts to be found in
Chapter 3 - ‘Results’. Additional details on materials and methods can be found
in the respective manuscripts in Chapter 3. Furthermore, Chapter 3 showcases
four manuscripts; two published, one submitted and one in preparation for
publication. These manuscripts provide findings that address the following
research questions:

1. Is there evidence for the immunemodulation of cord immune markers of

allergy by in utero exposure to sex hormone disrupters?

2. Is maternal oral contraceptive use (a proxy for in utero
estrogen/progesterone exposure) associated with immune markers of

allergy in the offspring?

3. Is maternal oral contraceptive use a risk factor for allergic diseases in

offspring?
4. Does serum organochlorine alters immune markers in children?

The fourth and last chapter of this monograph, ‘Discussion and Conclusion’,
discusses and present concluding remarks of the results found in chapter 3.
Practical public health issues are here presented along with future scientific steps

toward a better understanding and early life prevention of childhood asthma.

CHAPTER I

ALLERGIC DISORDERS

AND

SEX HORMONE DISRUPTERS

 

else

a

 

1.1 ALLERGY

Allergy is a hypersensitivity reaction initiated by immunologic mechanismsz.

Allergy can be antibody- or cell-mediated. In most patients, the antibody typically
responsible for an allergic reaction belongs to the lgE isotype and these patients
may be said to suffer from lgE-mediated allergy. In cell-mediated allergy, as in

allergic contact dermatitis, immunologically sensitized lymphocytes play a major

role 2. Allergic disease represents a dysregulation of the intact immune system.

 

Box 1 Definitions

ALLERGY is a hypersensitivity reaction initiated by immunologic mechanisms 2.
The term allergy is also used to denote allergic disorders such as asthma and

allergic rhinitis .
ASTHMA is a chronic inflammatory disorder of the lungs that is characterized by a
reversible obstruction of the airways

ATOPY is a personal and/or familial tendency to become sensitized and produce
lgE antibodies in response to ordinary exposures t; allergens. As a consequence,

asthma, rhinoconjunctivitis, or eczema may occur .
WHEEZE is a high pitched sound made when breathing, which probably results
from turbulence through narrowed tubes

SENSITIZATION is the state where, following exposure to allergens, the immune
system is primed to produce IgE antibodies specific to that allergen.

 

 

 

Patients usually develop a hypersensitive immune response to environmental

antigens, resulting in local tissue damage and Inflammation . Diseases such as

allergic rhinitis, dermatitis, and asthma represent a significant personal and

public health burden. Approximately 25% of all children experience at least one

allergic manifestation 8. Asthma however, has received the most attention seeing

that it is the most common chronic disease of childhood and is considered the

most debilitating 9.

1.2 Asthma

Definition and immunology

Definition

A universally accepted definition of asthma has been frustrated by a lack of
understanding of the mechanisms involved in the disease. The American
Thoracic Society defines asthma as "a disease characterized by an increased
responsiveness of the airways (bronchial hyperresponsiveness) to various stimuli
and manifested by slowing of forced expiration (due to narrowing of the airways)
which changes in severity either spontaneously or as a result of therapy". This
narrowing of the airways ultimately produces “wheeze”, which is considered to be

the necessary clinical criterion for asthma. Of note; not all that wheezes is

 

10
asthma
Asthma
Recently, the Nomenclature
, , Allergic asthma Nonallergic asthma
ReVIew Committee of the
. . 2 _ - _ _ . ,
WOFId Allergy Organization lgE mediated asthma Non lgE mediated allergic asthma
Figure 2 various definitions of asthma

 

 

 

proposed the use of the term
allergic asthma as the basic term for asthma mediated by immunologic
mechanisms (Figure 2). The organization further suggests that when there are

indications of lgE-mediated mechanisms, the term should be lgE-mediated

 

 

 

Ii

-I\
.Ih—P
‘0.
“'5
”P-

S/Pt:

asthma 2. In addition to lgE, several other immunologic factors such as lL-4 and

lL-13 are involved in the etiology of asthma.

Immunology
For the purpose of this work, the immunology of asthma is here discussed within
the framework of the type 1 and 2 helper T cells (Th1/Th2) model. This

model/paradigm was born out of a series of experiments developed by Tom

Wegmann 11 as he perused the placenta in search of clues that migt explain

spontaneous abortion. To date, mechanistic explanations for allergy have often

invoked the Th1/Th2 paradigm ”'13 ”'15.
Theoretically, the immune system ‘idles' in a Th0 or na'i've state until it is required
to change response by some external stimuli (Figure 3). These stimuli may be in

the form of environmental antigens in which case non-allergic Th1 response is

 

initiated ieading ICO Adopted from Whitacre et al. Science 1999;
Non-allergic state

. Non- Allergic
secretion of lgG W a immune markers
/ lFN-y, lL-2, etc.

antibodies, removal of the

allergen, and the

Antigen -—> Naive state

Allergic immune

suppression of lgE ”by," markers:
, lL-4, lL-13 etc.
Allergic state

. 16,17 . .
SyntheSIS - In add'tlon Figure 3 Theoretical immune responses

 

 

 

 

 

II

AL

to antigens, hormones may cause a shift in immune response. However, it is

believed that sex hormones such as progesterone and high concentration of

estrogen may result in an allergic — Th2 type response 1 (Figure 3).

There is a general consencious in the literature that a shift to a Th2 polarize state
occurs during dramatic changes in immune responses associated with

pregnancy. To prevent termination of the fetus (removal of the allergen), the feto-

maternal immune system shifts to a Th2 direction 18’19. This Th2 shift begins

with the secretion of human chorionic gonadotropin, which in turn stimulates

progesterone production. Progesterone then promotes the secretion of Th2
cytokines over the secretion of phagocytic 20 Th1 cytokines. In other words,
progesterone is believed to prime the immature fetal immune system towards an

allergic direction which may be manifed as allergic diseases (such as asthma)

during infancy.

 

Fundamentally, the underlying problem in asthma Box 2 Select markers of

immune response

 

appears to be immunologic with children in the early Th1 response:

lgG - immunoglobulin G

sta es of asthma showin si ns of excessive
g g g lFN-y - interferon gamma

imflamation in their ainivays. While chemokines (a "=2 - interleukin 2
IgA - immunoglobulin A

subset of cytokines that chemically attracts immune ,gM _ immunoglobunn M

cells) are important in the recruitment of imflamatory Th2 res onse,

lgE — immunoglobulin E
lL-4 - interleukin 4
those producing Th2 cytokines — are responsible for lL-13 - interleukin 13

cells to the airway, the lymphocytes — particularly

 

 

 

directing and maintaining this inflammatory process 21. Therefore, the cardinal

phenotypic features of allergy are associated with the production of the Th2
cytokines lL-4 and IL-13 which are central to the synthesis of IgE; the forerunner
of allergic manifestations (Box 2). Of note, recent reports indicate that cord blood

mononuclear cells (CBMC) produce lL-13 and that cord blood lL-13 is associated

22,2

with allergy in childhood 3. Further evidence suggests that lL-13 can

participate in allergies independent of lL-4 24.

As mentioned earlier, another important marker of a Th2 response is lgE, which
is central to the etiology of allergic diseases such as allergic asthma. In allergic
individuals, specific lgE is bound to high affinity receptors on mast cells and is
cross-linked by antigen; mast cell activation/degranulation occurs and sets in

motion a cascade of events resulting in the clinical manifestations of allergic

disease 25. The interaction of IgE and antigen results in an immediate

hypersensitivity reaction may be responsible for the classic asthma symptoms
exhibited during acute exacerbation, for example, mucosal edema, mucus
production, and smooth muscle constriction. Eventually, these and other reaction

cascades can induce the production of cells responsible for airway inflammation

that underlies allergic asthma 26. Nonallergic-asthma, conversely, manifests all

the clinical feature of allergic asthma without the lgE mediated immunologic

aspect. At a population level, it is estimated that 50% of all asthma cases have

10

 

 

 

 

 

.5

ad

 

 

an allergic etiology 8'27. Regarding Non-IgE-mediated asthma (Figure 2) this

subtype may be characterized by the following four criteria:
1) clinical phenotype of lgE mediated asthma;
2) absence of other allergic diseases such as allergic
dermatitis and rhinoconjunctivitis;
3) negative sensitization (Box 1) to common inhalant and food allergens ; and
4) normal total serum IgE levels.
In contrast to a Th2 response, a Th1 response leads to secretion of IgG

antibodies, removal of the allergen, and the suppression of lgE synthesis 28-30.

Other immune markers are associated with a Th1 type response (Box 2). In
general, Th1-responses are highly protective against infections mounted by the

majority of microbes.

It is widely accepted that changes leading to allergic disorders such as allergic
asthma begin in childhood, thus it is important to study the natural history of
childhood asthma in an effort to identify risk factors associated with its

development.

Asthma in children

Asthma may develop during the first few months of life, but it is often difficult to

make a definite diagnosis until the child is older 4. Approximately 30 % of

asthmatic patients are symptomatic by one year of age, 80-90 % have their first

symptoms before four to five year of age. The majority of affected children have

Il

only occasional attacks of slight to moderate severity, which are managed with
relative ease. A minority experience severe, intractable asthma, usually
perennial rather than seasonal. The relationship of age of onset to prognosis is
uncertain. Most severely affected children have an onset of wheezing during the

first year of life and a family history of asthma and other allergic diseases

(particularly atopic dermatitis) 31

Prevalence and burden of asthma

. . 32
Asthma remains an enormous personal and public health concern . Asthma

symptoms in children vary in prevalence from 0% to 35% in different populations

33. The prevalence of asthma symptoms is on average 14% among children

aged 6-7 years and 18% among those aged 13-14 years 33. Overall, the

. . . . . . 4,34
prevalence and severity of asthma and other allergic diseases is increasmg

Globally, approximately 300 million people currently have asthma and it is

estimated that there may be an additional 100 million more asthmatics by 2025 9.

Furthermore, asthma accounts for about one in 250 deaths worldwide.

In the United States (US), asthma affects more than 17 million individuals 35 and

results in approximately 10 million visits to physicians, 2 million emergency room

visits and approximately 7,000 deaths annually 36. The annual cost of caring for

12

asthmatics exceeds six billion dollars per year in the US, and the worldwide

market for asthma medication is currently valued at 5.5 billion dollars each year

37. It is noteworthy that the cost of asthma should be divided into direct and

indirect cost. Direct costs include the costs of medication, medical bills, and
documented episodes of health service utilization such as clinic visits and
hospital admissions. Indirect costs include the adverse economic impact of the
disease on an individual, family, and society. This includes the “cost" of

premature mortality and productivity loss. In the US, the cost of medical

treatments for asthma is on average 10% of total family income 38 39.

The economic cost of asthma is alarming both in terms of direct medical costs
(drugs, hospitalization) and indirect medical costs (time lost from work and
premature death). The number of disability-adjusted life years (DALYs) lost due
to asthma worldwide is currently estimated about 15 million/year, and asthma

accounts for around 1% of all DALYs lost, reflecting the high prevalence and

severity of asthma 9. Asthma in general is incapacitating and interferes with

school attendance, play activity, and day-to-day functioning. In a 1988 National
Health Interview Survey, children with asthma missed 10.1 million days from

school (2 times the number of days missed by children without asthma), and had

12.9 million contacts with medical doctors and 200,000 hospitalizations 40. In the

same survey, it was determined that almost 30% of children with asthma

experienced some limitation of physical activity, compared to only 5% of children
without asthma 41.
Risk factors for asthma

Risk factors for asthma may be classified as host factors that predispose
individuals to developing asthma, and environmental factors that influence the

susceptibility to the development of asthma in predisposed individuals,

precipitate asthma exacerbations, and/or cause symptoms to persist 4.

Host factors
These include sex and race along with the genetic predisposition to the

development of either asthma or allergic sensitization.

Sex

Sex biases in the susceptibility to, and severity of, autoimmune and allergic

. . 7 .
diseases are well-recognized . Preadolescent boys have about 1.5 times the
. 42-44 .
asthma prevalence of girls . However, after adolescence, the sex difference

is reversed 45. Some authors suggest that the increased risk for males in

childhood is probably related to narrower airways, increased airway tone 46—48,

and possibly higher lgE 44 in boys. These factors, it is believed, predispose boys

to enhanced airflow limitation in response to a variety of insults. More recently,

other studies have suggested that the observed difference may be due to

. . 1,7 . . . 49-
endocrine influences . In an effort to explain this difference, some authors

51 postulate that the acquired immune system of females differs from that of

males, because estrogens stimulate immunologic processes driven by CD4 +
TH2 cells and B cells, whereas androgens enhance CD4 + TH1 and CD8 + cell

14

activity. Androgens are also known to promote production of lL-2 by TH1 cells.
Interleukin 2 reduces TH2 activity and stimulates CD8 + T cells 52. Further

discussion about the possible role of estrogen as an immune modulator is

presented in section ‘Estrogen and progesterone.’

Race

Most experts agree that race as a social construct represents a continuum rather

than a term with clear-cut boundaries, a point that has been illustrated by large-

scale genotyping among diverse biogeographic populations 53

Asthma morbidity and mortality is disproportionately high and continues to
increase among African Americans 54. As documented by national surveys, there
are higher rates of asthma in black than White children, a disparity that has
widened since 1980 55. Simon and co-workers also found that among children
with asthma, Blacks were more likely than Whites to report asthma-related
limitations in physical activity and need for urgent medical services 56

Furthermore, at the molecular level, authors have detected increased expression

of the CD80/CDB6 lymphocytes in African American adults following mitogen and

antigen stimulation 57. CD80/86 are crucial molecules in T-cell activation 58. This

finding was supported by that of Willwerth et al. 58 which showed a more

generalized increased T-cell responsiveness that included responsiveness to the

mitogen phytohemagglutinin and to the allergens in children of black ethnicity.

15

These reports give credence to the idea that differential neonatal immune

responses may underlie the higher incidence of asthma in blacks.

Genetic predisposition

There is evidence that asthma has genetic components 4. A number of studies

have shown an increased prevalence of asthma phenotypes among the offspring

of subjects with asthma compared to the offspring of subjects without 59

60. Family studies have convincingly indicated that atopy (as measured by

allergen skin tests, total lgE, and/or specific lgE), airway hyper-responsiveness,

and asthma as diagnosed by questionnaire are at least partly under genetic

59 61
control .

Numerous studies of twins have demonstrated that concordance rates for
asthma phenotypes are all substantially higher for monozygotic than for dizygotic
twins, suggesting a strong genetic contribution. In population—based studies of

twins, the estimated effect of genetic factors is about 35 to 70 %, depending on

the population and the design of the study 59’60.
Environmental factors

Early attention to environmental factors that may influence the susceptibility to
the development of asthma in predisposed individuals focused on the effect of
allergens, air pollution and tobacco smoke. Recently, the role of environmental
SHD (synthetic hormones: estrogen and progesterone as well as toxicants:

halogenated organochlorine compounds (HOCs) such as dichlorodiphenyl

l6

dichloroethene (DDE) and polychlorinated biphenyls (PCBs)) in the etiology of

. . . 62-69
allergy has gained increased attention

Exposure to tobacco smoke

Tobacco burning, a ubiquitous source of irritants, produces a large and complex
mixture of gases, vapors, and particulate matter. More than 4,500 compounds
and contaminants have been identified in tobacco smoke, among them respirable

particles, polycyclic hydrocarbons, carbon monoxide, carbon dioxide, nitric oxide,

. . . . . 4
nitrogen OdeeS, nicotine, and acrolein .

There is evidence that both maternal smoking during pregnancy and passive

smoke (environmental tobacco smoke (ETS)) exposure after pregnancy affect
. . 70—74 .

asthma later In offspring . Actually, ETS was reported to increase both

prevalence and severity of asthma, as judged by increases in the frequency of

attacks, the number of emergency room visits and the risk of intubations 75

Using data from a large cohort study (n~60,000), Jaakkola et al. demonstrated

that maternal smoking in pregnancy increases the childhood risk of asthma 76

A large study of 20,000 children (aged 6 to 12 years) from nine countries in

Europe and North America, reported that smoking during pregnancy decreases

lung function parameters 77. The effects of past and current passive exposure to

tobacco smoke were smaller than the effect of smoking during pregnancy. The

aforementioned study strongly suggests that maternal smoking during pregnancy

17

subsequently affects the lung function and intuitively asthma in children.

However, in a cohort of New Zealand children, Sherill et al., 78 found no

significant detrimental effects for absolute forced expiratory volume (FEV), or

vital capacity related to ETS (including in utero) exposure. Likewise,
Cunningham et al., 79 compared lung function, measured by FEV1, and were

unable to find significant reduction.

Breast-feeding

Ever since Grulee and Sanford in 1936 80 first reported that breastfeeding

protects against an infantile allergic response, researchers have been exploring

this topic. Some authors reported that breastfeeding appears to protect against

87-92

atopic manifestations 81-86 while others found no such association To-

date, despite numerous studies, the search for consensus continues 81’82'93.

Breast milk, as a functional food, is thought to contain its own complex immune

system 94. The selective colonization of the mammary gland during lactation by a

population of T lymphocytes which displays the phenotype and functional

characteristics of memory T cells may be one of the mechanisms whereby the

suckling infant benefits from its mother's immunological history 95. Cytokines and

chemokines derived from milk cells and mammary gland epithelium may

18

contribute to the activation of intestinal T lymphocytes to enhance immunity

during the early neonatal period 96.

Findings from a population-based case-contral study (cases=2,315 and control
=21 ,513) of 6—15 year old Japanese children showed a significantly higher
prevalence of asthma among children who had been breastfed (adjusted OR =
1.2; 95% Cl: 1.05, 1.36). Conversely, a systematic review of prospective studies
evaluating the association between exclusive breast-feeding during the first 3-
months after birth and asthma, reported opposite findings. The aforementioned
study found summary odds ratio (OR) for the protective effect of breast-feeding
to be 0.70 (95% Cl 0.60, 0.81) thus concluding that exclusive breast feeding

during the first months after birth is associated with lower asthma rates during

childhood 81.

Maternal allergy

Children born to parents with allergic diseases present an increased risk of

developing similar diseases 75’97. The proportion of children of two allergic

parents can be as high as 58 %, when one parent is allergic it is about 29% and

a child with neither parents having history of allergy still has about 13 % risk of

developing allergic disease sometime during his life 98. However, maternal atopy

is a stronger predictor of childhood allergy that paternal. Mother’s atopic history

was found to be significantly associated with cord-blood lgE 99. Likewise,

maternal history of atopy was associated with an elevated lgE among newborns.

l9

For maternal asthma, this association was only evident in infant girls 100.

Recently, Kuiper et al. found that maternal asthma was associated with elevated

total lgE at all total IgE cut-off levels studied except 0.5 lU/mL 101. This finding

was supported by Shah et al. who report a significant association between

maternal history of allergic disease and elevated cord serum lgE among

98
newborns

Environmental sex hormone disrupters
A balanced endocrine system is essential for optimum health. Sex hormones are
considered crucial in maintaining this balance. Therefore any disruption/alteration

of these hormones could intuitively bring about adverse health effects such as

. . 7 . . .
allergic diseases or even death. A more extensrve CIISCUSSIOTT on sex hormone

disrupters and their possible role in the etiology of diseases is presented on the

next page.

Summary
The prevalence and severity of allergic diseases continue to increase. Sex is an

important risk factor with evidence showing that asthma is more prevalent in boys
throughout childhood, but in adolescence a gender reversal occurs. African-
Americans are disproportionately affected while the idea of a genetic
predisposition to allergy continues to intrigue researchers. There is virtually
universal agreement that tobacco smoke represents a significant risk factor for

allergies. Conversely, the ‘verdict’ on breast-feeding is inconclusive. Mothers

2O

with allergies tend to have children with allergies: clinical and/or immunologic
(increased lgE). Sex hormone disrupters have been implicated in the etiology of
allergic disorders. Recently, a plethora of studies have focused on this area in

an effort to explain the rising prevalence of allergies.

1.3 Sex hormone disrupters

One difficulty experienced when considering endocrine disruption is the lack of
consensus as to an appropriate definition for the term. Various definitions have
been suggested including those of the US EPA (Kavlock et al., 1996) and a
European workshop (EC, 1997). For this dissertation, the approach taken by the
European workshop was adopted and modified to apply specifically to chemical
disruption of the endocrine system: “A sex hormone disrupter is an exogenous
substance that causes adverse health effects of an intact organism or its

progeny, consequent to changes in endocrine function.”

Sex hormones are synthesized mainly by endocrine glands such as the gonads
(testis and ovary), the adrenals and (during gestation) by the fetoplacental unit,

and are then released into the blood circulation.

The biological hormonal properties of these substances have earned them a
great deal of attention in the past decade as possible etiologic agents of certain
diseases. Some of these chemicals display estrogen like activity, hence their
importance in endocrine disruption and subsequent immune modification is

closely scrutinize. Sources of these xenobiotics includes life style practices such

21

as diet (e.g., phytoestrogens coumestrol and genestein), oral contraceptive use
or from the environment (e.g., dichlorodiphenyl dichloroethene (DDE), and
polychlorinated biphenyls (PCBs). Sex hormones or similar substances from the
above-mentioned sources may be classified as endocrine disrupters (EDs).
These endocrine disrupters alter normal hormonal regulation and may be

naturally occurring or environmental contaminants‘”.

There are two classes of substances that may cause disruption: 1) Synthetic
hormones; mainly estrogen and progesterone as in the case of oral
contraceptives (Ocp), and 2) toxicants including the halogenated organochlorine
compounds (HOCs): DDE & PCBs. It has been broadly suggested that these
substances bring about adverse effects in at least two ways: by mimicking the
action of a naturally-produced hormone and thereby setting off similar bio-
chemical reactions and/or by affecting the synthesis, transport, metabolism and
excretion of hormones, thus altering the concentrations of natural hormones‘o‘.
However, there is yet to be a clear mechanistic understanding of how these
proposed sex hormones disrupt the immune system. Recent attention has
focused on the role of the ‘promiscuous’ estrogen receptors (ERs) through which
sex hormone disrupters may communicate with cells prior to altering immune

response.

22

Estrogen and progesterone

There is mounting evidence that the immune system is regulated in part by sex
steroid hormones such as estrogen and progesterone. It is known that proper
immune function is critical for the delivery of a healthy baby a process that is also
believed to be of importance in the etiology of immune diseases. The
involvement of estrogen is hinged on the fact that estrogen receptors are found
on T lymphocytes: Families of specialized white blood cells that help orchestrate

the body's immune responses. Generally, estrogens are seen as enhancers at

least of the humoral immunity 49. Furthermore, many of the allergic family of

cytokines are under the control of estrogen 105. Estrogen and progesterone

appears to work in tandem. In order for progesterone to work on the immune
system, progestin receptors must be synthesized on the reticuloendothelial cells

(cells which trap and consume foreign agents, except leukocytes circulating in

the bloodstream) of the thymus 106. Such receptors are usually the result of

estrogen stimulation.

Under normal circumstances, increased estrogen leads to an increase in INF-y

concentration, presumably as a mechanism of feedback inhibition to prevent

unopposed stimulation of TH2 cells 107. Estrogen also stimulates secretion of IL-

4, -5, -6, and -10 by TH2 lymphocytes. These cytokines are potent stimulators of

B-cell proliferation, maturation into plasma cells, and synthesis of antibody.

23

 

 

 

III

ac

Hal

.4
O

 

Interleukins 4, 5, 6, and 10 are expressed in greater quantity in an estrogen-

. .. 50,52,108,109
dominant hormonal milieu .

Progesterone seems to promote the preferential development of Th2 cells and to

induce lL-4 production 110-113. Th2 type cytokines are considered vital for the

maintenance of pregnancy by controlling the immune and endocrine systems

19’114. Simultaneously, it is this ability of progestin to invoke this Th1/Th2 shift

that makes it important in the etiology of allergic diseases. However, the tight
interdependence between estrogen and progestin secretions during the ovarian
cycle and concomitant sex hormone hypersecretion during pregnancy pose
substantial difficulties for the assessment of the individual contribution of

physiological changes in progestin levels to immune regulation.

Epidemiologically, Michel and colleagues reported that progestin administered
during pregnancy increased the levels of cord blood lgE in 136 neonates: when
mothers received progesterone, 53% of the newborns had detectable lgE (<0.5

IU/ml) versus only 24% of the newborns in mothers without progesterone

. . . 97
administration

Halogenated organochlorine compounds (HOCs)
To date studies assessing the role of HOCs in altered immune response have
focused mainly on their estrogen-like effects. However, some authors have

proposed that the estrogenic activity of environmental toxicants is very low,

24

relative to estradiol, and thus unlikely to make significant contribution to the

115J16

etiology of diseases . This view has been challenged by reports of low

dose effects of estrogenic toxicants 117'118 while others have been unable to

119fl20

confirm these effects . Nonetheless, environmental toxicants continue to

be regarded as important contributing factors in the pathogenesis of allergic

diseases.

Dichlorodiphenyldichloroethene (DDE)
Dichlorodiphenyltrichloroethane (DDT) was originally prepared in 1873, but it was

not until 1939 that Paul Muller of Geigy Pharmaceutical in Switzerland

discovered the effectiveness of DDT as an insecticide 121. The use of DDT

increased enormously worldwide after World War II, primarily because of its
effectiveness against the mosquito and lice. The World Health Organization
estimates that during the period of its use approximately 25 million lives were
saved predominantly from malaria and typhus. However, many species of insects
developed resistance to DDT; it proved to have a high toxicity toward fish; and it
was responsible for the near extinction of several bird species because of its
interference with the formation of egg shells. For these reasons and because of
its environmental persistence, the use of DDT was banned in the United States in

1972. However, DDT is still in use in many countries (including some South

. . . . .. 122
American, African countries) as an InsectICIde

25

Humans are exposed to DDT mainly through foods, and infants through the

placenta and breast-feeding 123. DDE, a metabolite of DDT, is persistent and is

stored in fat tissue. The long half-life of DDE accounts for its ubiquity in the

general population 124’125. Though exposure is of relatively low toxicity, DDE

does have troubling effects as it is known to have weak estrogenic 126 and

considerable anti-androgenic activity resulting in its endocrine disrupting

capabilities 127-131. A more detailed description of DDE possible effect on the

etiology of allergic diseases is presented in the ‘Literature Review of

Components Comprising the Conceptual model’ section.

Polychlorinated biphenyls (PCBs)

PCBs are mixtures of up to 209 individual chlorinated compounds (known as
congeners). There are no known natural sources of PCBs. PCBs are either oily
liquids or solids that are colorless to light yellow. Some PCBs can exist as a
vapor in air. PCBs have no known smell or taste. Many commercial PCB

mixtures are known in the US. by the trade name Aroclor.

PCBs have been used as coolants and lubricants in transformers, capacitors,
and other electrical equipment because they do not burn easily and are good
insulators. The manufacture of PCBs was stopped in the US. in 1977 because of
evidence they build up in the environment and can cause harmful health effects.

Products made before 1977 that may contain PCBs include old fluorescent

26

lighting fixtures and electrical devices containing PCB capacitors, and old

. . . 132
microscope and hydraulic OIIS

Residues have been detected in foods and in human adipose tissue, milk, and

serum fat 133. Reports of PCBs having estrogenic effects 129 and altering

immune responses have been recorded.

Reichrtova et al. 65 reported that higher levels of the congener PC8118 and DDE

in human placentas are associated with increased cord blood lgE levels.
Specific to PCBs and various congeners: A study using pregnant minks

documented that serum progesterone concentrations were significantly increased

by experimental exposure to PCB-153 134. Similarly, an in vitro study

documented increased secretion of progesterone from ovaries after exposure to

135136

PCB-153 . In seals, progesterone metabolism was significantly decreased

with increasing liver PCB concentrations”? In polar bear dams, the plasma

concentration of total PCBs is associated with an increase in placenta

progesterone and accounts for 27% of the variance in the progesterone

. 138
concentration

Sex hormones and immunophysiology - Conclusion
There is an abundance of evidence suggesting that the sex hormone disrupters
DDE, PCBs, and synthetic progesterone, and estrogen may bring about

alteration of normal endocrine functions resulting in allergic manifestations.

27

However, no clear unifying hypothesis exists to explain how sex hormone

disrupters affect the development and function of the immune system.

1 .4 Time-window of exposure
In an attempt to sort out the complex interrelationships between sex steroid
hormones and allergic diseases, this work will focus on two important time-

windows: the in utero and perinatal.

The in utero environment sets the stage for prenatal priming of allergic diseases.
This concept (prenatal priming) considers fetal life as the critical period of
exposure and suggests that in utero exposure to environmental factors such as

sex steroid hormones are important for the rise in the prevalence of asthma and

139-141
allergy .

The prenatal priming concept

‘Priming or programming’ describes the process whereby a factor at a critical

period of development has lasting effects 142. As an example, dramatic changes

in maternal immune responses occur during pregnancy. To avoid aborting the

fetus, the feto-maternal immune system is shifted in an allergic/Th2 direction
18’19. This Th2 reaction is characterized by the secretion of cytokines which then

promote lgE production, resulting finally in allergies later in life. Hormones are

believed to be responsible for controlling this shift 105.

28

 

 

 

T)

In assessing the in utero impact of sex steroid hormones on allergies, this
dissertation investigates associations between: placental organochlorine and
cord blood markers of allergy; maternal oral contraceptive use and offspring’s
serum allergic markers followed by an investigation of maternal oral

contraceptive use and allergic phenotypes.

The postnatal priming concept

Seeing that hormones may play an important in the prenatal priming of allergies,
it is plausible to postulate that postnatal exposure may also have a role in the
etiology of allergies. Using serum 00, this work further determined the
importance of postnatal exposure to sex steroid hormones in altering allergic

markers and henceforth the development of asthma.

A theoretical construct outlining the components of both processes (pre- and

postnatal), are presented in the forthcoming section.

1.5 Components of the conceptual model

The etiology of asthma, for example, is the result of a complex conflation of
events. This section presents a model for assessing one of these complex
processes: the involvement of sex hormone disrupters. The components of this
model span the pre - and postnatal periods while using a combination of
qualitative and quantitative measures of exposure and outcome. Specific pieces
of evidence provided are: First, evidence that prenatal SHD exposure can
disrupt allergic markers measured in utero. For this piece of evidence, placental

p, p’ DDE was used as measure of exposure while cord plasma lL-4, lL-13 and

29

INF-y (cytokines typical of the atopic response) as measures outcome. This will
show the effect of SHD on allergic markers that cannot be attributed to postnatal
influences. Note, both exposure (placental p,p’ DDE) and outcomes (IL-4, lL-13
and INF-ry) are quantitative measures. The hypotheses behind this piece of
evidence are: Increased placental p,p’-DDE concentration is associated with an
increase in cord plasma IL—4 and lL-13, and that increased placental p,p’-DDE
concentration will lead to a reduction in cord plasma lNF-wy. Second, existence of
an association between proxy measures of in utero exposure to sex hormone
disrupters and postnatal markers of allergic susceptibility. This bit of evidence
suggests that the possible in utero immunomodulating effect of sex hormone
disrupters may persist into child hood. Third, as an extension of the second
condition above, evidence is required linking proxy measures of in utero
exposure to sex hormone disrupters (maternal oral contraceptive use) with
markers of effect: clinical manifestations of allergy such as asthma-like
symptoms and atopic eczema in childhood. This along with evidence from the
second condition will provide further support for the idea that sex steroid
hormones may be associated with allergic disorders by modulating the immune
status, through endocrine disruption. Fourth, serum organochlorines stimulate
an allergic response. Results from the serum DDE / markers of allergic
susceptibility association add a new level of support to the sex hormone
disrupters / immune modulating idea in that both exposure and outcome are
measured postnatally. This suggests that the effects of sex hormone disrupters

do not only occur during the early stages of immune development (in utero) when

30

 

      

 

 

 

*Estrogen I 37:39:“:
Progesterone markers
@childhood immune WE

 

 

 

Pregnancy (in utero environment) I a ChildhoOd

log/e

* Maternal oral contraceptive use before pregnancy

 

 

    
     
    
   

  
  

Cord
plasma
IL-13, lL-4
& INF-q

Placental
OC

 
 

 

 

 

Figure 1 Sex Hormone Disrupters and Childhood Allergy - Conceptual model
the cells are more susceptible to change but may also produce change in
immune development postnatally. A schematic representation of a proposed
model that encapsulates the overarching theme of SHD and childhood allergy is

presented in Figure 1 and discussed in the following background section.

1.6 Literature review of components comprising the conceptual model.
The effect of placental p,p’ DDE on cord plasma lL-4, lL-13 & INF-y.

There is evidence that the cytokine lL-13 is both necessary and sufficient to

induce all features of allergic asthma 143. In humans, this was supported by

consistent associations between tissue lL-13 levels and genetic variants in the

lL-13 gene with asthma and related traits 143. Cellular response (including lL-4,

lL-13, and lFNq,) to allergens have been reported at week 23 gestation 144

31

 

 

 

('1

Recent findings suggest that in the etiology of allergic diseases, the Th2/Th1

ratio may be of more importance that individual cytokines 145

DDE is known to cross the placenta 146 potentially interfering with fetal immune

development. This interference is thought to be via its ability to reduce the

binding of progesterone to its receptor 147

Prenatally, Noakes and co-workers suggest a possible immunomodulating effect

of cord blood DDE on phytohaemagglutinin mitogen stimulated cord blood lL-13

and lFN-y 148. However, findings on the effect of placental p,p’ DDE on cord

plasma cytokines and their ratios (Th1/Th2) are yet to be reported.

In utero exposure to sex hormone disrupters is related to immune markers
of allergic susceptibility in childhood.

To date, Frye et al. is the only team to report on the association between
maternal OCp use and immune markers (specific immunoglobulin E against
common inhalant allergens) in offspring. For atopic sensitization, they reported

similar effects for maternal OCp use before and after birth. This finding raises

93,105,149

doubts in the previously reported association between maternal

hormone level and allergic diseases.

In an effort to allay these doubts, this dissertation investigates the association
between maternal oral contraceptive use and humoral immune markers (such as

serum immunoglobulin (lg)-A, lgE and basophilic surface lgE counts) in offspring.

32

In utero exposure to sex steroid hormones and their relation to childhood
markers of allergic effect.

Studies assessing the association between in utero exposure to sex hormone
disrupters (using maternal oral contraceptive use as proxy) and allergic

phenotypes have reported inconsistent findings.

In 1987, Peters and Golding, presented findings suggestive of an increased risk

of childhood eczema for children of mothers who used OC in the 18 months prior

to pregnancy in a cohort of 10,900 5-year-old children 93

Approximately ten years later, using data from an aggregative (ecological) study,
stt et al. found a geographic trend between asthma prevalence in children and
maternal OCp use. This similarity was most notable during 1970 when a sharp
decrease in OCp use was followed by a similar decrease in rate of hospital
discharge with a diagnosis of asthma. They concluded that mother’s OCp use fits

well into the geographic and temporal background of this increase in asthma

105
prevalence

In addition, findings from a geographically defined cohort of 5,188 subjects

showed that the prevalence of atopy at age 31 years was lower in children whose

mothers reached menarche at a later age, especially after age 15 149. Early age

of menarche is associated with higher levels of estrogen in adulthood 150452.

33

Contrary to the above findings, Xu et al. 153 found no differences in the mean

concentrations of maternal sex steroid hormones according to the presence of
allergic rhinitis or atopic eczema among offspring. In addition, while Frye et al.

found maternal OC use before birth to be risk factor for asthma, allergic rhinitis,

and atopic eczema; they also found similar effects for 00 use after birth 69.

To date, these associations have only been tested using data from predominantly
Caucasian cohorts. Assessing this association in a cohort with a different ethnic
profile may provide cogency to the controversy. It is in this spirit that we
investigated the in utero sex hormone exposure / allergic phenotypes association

using data form an African-Caribbean cohort.

Immunologic consequence of childhood exposure to organochlorine.
Previously, in this monograph, we proposed the assessment of prenatal
exposure to DDE and its association with allergic markers in childhood. This, as
mentioned, provided evidence in favor of the prenatal priming concept; purporting
that DDE is able to alter immune status during its early stages of development. In
this section we extend this idea by determining whether DDE is also able to

impose a similar effect postnatally.

There are reports of DDE being associated with changes in cellular and humoral

154,155

immunity , in particular cytokines related to allergy, such as interleukin-4

156'157. In addition, using data from the Child Health and Environment Cohort

34

Study - Germany, Karmaus et al. showed serum DDE to be strongly related with

increases in total lgE 66. All the above-mentioned studies focused on adult

populations, with the exception of that of Karmaus et al. However, this study in
children only reported on one allergic marker -— total lgE. Here, using the same
study of Germany school children, we investigated whether serum DDE
concentrations is associated with several immune markers including lgE, IgG,
and IgA levels, lgE count on basophils along with eosinophilic granula, and white

blood cell count.

35

References

10.

11.

12.

Whitacre CC, Reingold SC, O_Looney PA. A gender gap in autoimmunity.
Science 1999;283(5406):1277-8.

Johansson SG, Bieber T, Dahl R, Friedmann PS, Lanier BQ, Lockey RF,
Motala C, Ortega Martell JA, Platts-Mills TA, Ring J, Thien F, Van
Cauwenberge P, Williams HC. Revised nomenclature for allergy for global
use: Report of the Nomenclature Review Committee of the World Allergy
Organization, October 2003. J Allergy Clin lmmunol 2004;113(5):832-836.

Galli SJ. Allergy. Curr Biol 2000;10(3):R93-5.
National institutes of Health, LaBl. NH. GLOBAL STRATEGY FOR
ASTHMA MANAGEMENT AND PREVENTION, 2005.

MERCKMedicus. Updated 2006. Available at:
http://www.merckmedicus.com

Elphick HE, Sherlock P, Foxall G, Simpson EJ, Shiell NA, Primhak RA,
Everard ML. Survey of respiratory sounds in infants. Arch Dis Child
2001;84(1):35—39.

Shames RS. Gender differences in the development and function of the
immune system. Journal of Adolescent Health 2002;30(4):59—70.

Kuehr J, Frischer T, Karmaus W, Meinert R, Barth R, Urbanek R. Clinical
atopy and associated factors in primary-school pupils. Allergy
1992;47(6):650-5.

Humbert M. Asthma, a priority for the allergist. Allergy 2006;61(5):515-7.

Murray DM, Lawler PG. All that wheezes is not asthma. Paradoxical vocal
cord movement presenting as severe acute asthma requiring ventilatory
support. Anaesthesia 1998;53(10):1006-11.

Chaouat G, Ledee-Bataille N, Dubanchet S, Zourbas S, Sandra 0, Martal
J. TH1/T H2 paradigm in pregnancy: paradigm lost? Cytokines in
pregnancy/early abortion: reexamining the TH1/'1' H2 paradigm. Int Arch
Allergy lmmunol 2004;134(2):93—119.

Maggi E. The TH1/T H2 paradigm in allergy. lmmunotechnology
1998;3(4):233-44.

36

 

13.

14.

15.

16.

17.

18.

19.

20.

21.

22.

23.

24.

25.

Wills-Karp M, Santeliz J, Karp CL. The germless theory of allergic
disease: revisiting the hygiene hypothesis. Nat Rev lmmunol
2001;1(1):69-75.

Robinson 08, Larche M, Durham SR. Tregs and allergic disease. J Clin
Invest 2004;114(10):1389-97.

Romagnani S. Immunologic influences on allergy and the TH1/T H2
balance. J Allergy Clin Immunol 2004;113(3):395-400.

Beeh KM, Ksoll M, Buhl R. Elevation of total serum immunoglobulin E is
associated with asthma in nonallergic individuals. Eur Respir J
2000;16(4):609-14.

Baena-Cagnani CE, Teijeiro A. Role of food allergy in asthma in
childhood. Curr Opin Allergy Clin lmmunol 2001 ;1(2):145-9.

Gaunt G, Ramin K. Immunological tolerance of the human fetus. Am J
Perinatol 2001;18(6):299-312.

Thellin O, Coumans B, Zorzi W, lgout A, Heinen E. Tolerance to the foeto-
placental ’graft': ten ways to support a child for nine months. Curr Opin
lmmunol 2000;12(6):731-7.

Romagnani S. The role of lymphocytes in allergic disease. journal of
allergy and clinical immunology 2000;105(3):399-408.

Monaco C, Andreakos E, Kiriakidis S, Feldmann M, Paleolog E. T-cell-
mediated signalling in immune, inflammatory and angiogenic processes:
the cascade of events leading to inflammatory diseases. Curr Drug
Targets lnflamm Allergy 2004;3(1):35-42.

Ohshima Y, Yasutomi M, Omata N, Yamada A, Fujisawa K, Kasuga K,
Hiraoka M, Mayumi M. Dysregulation of lL-13 production by cord blood
CD4+ T cells is associated with the subsequent development of atopic
disease in infants. Pediatr Res 2002;51(2):195-200.

Williams TJ, Jones CA, Miles EA, Warner JO, Warner JA. Fetal and
neonatal IL-13 production during pregnancy and at birth and subsequent
development of atopic symptoms. J Allergy Clin lmmunol
2000;105(5):951-9.

Wynn TA. lL-13 effector functions. Annu Rev lmmunol 2003;21:425-56.

Siraganian RP. Allergy: principles and practice. . 5th ed. ed. St.Louis: C V
Mosby, Inc, 1998.

37

26.

27.

28.

29.

30.

31.

32.

33.

34.

35.

36.

37.

38.

Oettgen HC, Geha RS. lgE in asthma and atopy: cellular and molecular
connections. J Clin Invest 1999;104(7):829-35.

Ponsonby AL, Gatenby P, Glasgow N, Mullins R, McDonald T, Hurwitz M.
Which clinical subgroups within the spectrum of child asthma are
attributable to atopy? Chest 2002;121(1):135-42.

Renz H. The central role of T-cells in allergic sensitization and lgE
regulation. Exp Dermatol 1995;4(4 Pt 1):173-82.

Daser A, Meissner N, Herz U, Renz H. Role and modulation of T-cell
cytokines in allergy. Curr Opin lmmunol 1995;7(6):762-70.

Erb KJ. Atopic disorders: a default pathway in the absence of infection?
lmmunol Today 1999;20(7):317-22.

Behrman RE, Kiliegman, R. M., Jenson, H. B. Nelson Textbook of
Pediatrics. 17 ed W.B. Saunders Company, 2004.

Guilbert T, Krawiec M. Natural history of asthma. Pediatric Clinics of North
America 2003;50(3):523-38.

Johnson CC, Ownby DR, Zoratti EM, Alford SH, Williams LK, Joseph
CLM. Environmental epidemiology of pediatric asthma and allergy.
Epidemiologic Reviews 2002;24(2):154-175.

Beasley R, Crane J, Lai CK, Pearce N. Prevalence and etiology of
asthma. The Journal of Allergy and Clinical Immunology 2000;105(2 Pt
2):S466-72.

MMWR Morb Mortal Wkly Rep. Forecasted state-specific estimates of
self-reported asthma prevalence-United States. MMWR Morb Mortal Wkly
Rep, 1998;47:1022—5.

Mannino DM, Homa DM, Akinbami LJ, Moorman JE, Gwynn C, Redd SC.
Surveillance for asthma—United States, 1980-1999. MMWR Surveill
Summ 2002;51(1):1-13.

Palmer LJ, Cookson W. Genomic approaches to understanding asthma.
Genome Research 2000; 10(9):1280-1287.

Marion RJ, Creer TL, Reynolds RVC. Direct and Indirect Costs Associated

with the Management of Childhood Asthma. Annals of Allergy
1985;54(1):31-34.

38

 

 

39.

40.

41.

42.

43.

44.

45.

46.

47.

48.

49.

50.

Vance VJ, Taylor WF. The financial cost of chronic childhood asthma. Ann
Allergy 1971 ;29(9):455-60.

Department of Health and Human Services. Action against asthma: A
strategic plan for the Department of Health and Human Services. 2000.
Taylor WR, Newacheck PW. Impact of childhood asthma on health.
Pediatrics 1992;90(5):657-62.

Taussig LM, Wright AL, Holberg CJ, Halonen M, Morgan WJ, Martinez
FD. Tucson Children's Respiratory Study: 1980 to present. J Allergy Clin
Immunol 2003;111(4):661-75; quiz 676.

Kuehr J, Frischer T, Karmaus W, Meinert R, Barth R, Herrmann-Kunz E,
Forster J, Urbanek R. Early childhood risk factors for sensitization at
school age. Journal Of Allergy And Clinical Immunology 1992;90(3 Pt
1):358-63.

Sears MR, Burrows B, Flannery EM, Herbison GP, Holdaway MD. Atopy
in childhood. 1. Gender and allergen related risks for development of hay

fever and asthma [see comments]. Clinical and Experimental Allergy
1993;23(11):941-8.

Camargo CA, Jr., Schatz M. The relationship of gender to asthma
prevalence, healthcare utilization, and medications in a large managed
care organization. Acad Emerg Med 2003;10(5):508.

Gissler M, Jarvelin MR, Louhiala P, E. H. Boys have more health
problems in childhood than girls: follow-up of the 1987 Finnish birth cohort.
Acta Paediatr 1999;88:310-4.

LeSouef PN. Expression of predisposing factors in early life. Asthma:
physiology, immunopharmacology and treatment. . London: Academic
Press 1993.

Smith JM, Harding LK, G. C. The changing prevalence of asthma in
school children. . Clin Allergy 1971;1z57-61.

Cutolo M, Sulli A, Capellino S, Villaggio B, Montagna P, Seriolo B, Straub
RH. Sex steroid hormones influence on the immune system: basic and
clinical aspects in autoimmunity. Lupus 2004;13(9):635-638.

Beagley KW, Gockel CM. Regulation of innate and adaptive immunity by

the female sex steroid hormones oestradiol and progesterone. FEMS
Immunol Med Microbiol 2003;38(1):13—22.

39

 

51.

52.

53.

54.

55.

56.

57.

58.

59.

60.

61.

62.

Sell E, Arici A. Sex steroids and the immune system. Immunology and
Allergy Clinics of North America 2002;22(3):407-433.

Parham P. The Immune System. NY: Elsevier Science Inc, 2000.

Barnes KC. Genetic epidemiology of health disparities in allergy and
clinical immunology. Journal of Allergy and Clinical Immunology
2006; 1 1 7(2):243-254.

Joseph CLM, Williams LK, Ownby DR, Saltzgaber J, Johnson CC.
Applying epidemiologic concepts of primary, secondary, and tertiary
prevention to the elimination of racial disparities in asthma. Journal of
Allergy and Clinical Immunology 2006;117(2):233-240.

Centers for Disease Control and Prevention. Measuring childhood asthma
prevalence before and after the 1997 redesign of the National Health
Interview Survey - United States (Reprinted from MMWR, vol 49, pg 908-
911, 2000). Jama-Journal of the American Medical Association
2000;284(18):2312-2313.

Simon PA, Zeng ZW, Wold CM, Haddock W, Fielding JE. Prevalence of
childhood asthma and associated morbidity in Los Angeles County:
Impacts of race/ethnicity and income. Journal of Asthma 2003;40(5):535-
543.

Hutchings A, Purcell WM, Benfield MR. Peripheral blood antigen-
presenting cells from African-Americans exhibit increased CD80 and

CD86 expression. Clinical and Experimental Immunology
1999;118(2):247-252.

Willwerth BM, Schaub B, Tantisira KG, Gold DR, Palmer LJ, Litonjua AA,
Perkins DL, Schroeter C, Gibbons F K, Gillman MW, Weiss ST, Finn PW.
Prenatal, perinatal, and heritable influences on cord blood immune

responses. Annals of Allergy Asthma & Immunology 2006;96(3):445-453.

Holloway JW, Beghe B, Holgate ST. The genetic basis of atopic asthma.
Clin Exp Allergy 1999;29(8):1023-32.

Wiesch DG, Meyers DA, Bleecker ER. Genetics of asthma. J Allergy Clin
lmmunol 1999; 1 04(5):895-901.

Holgate ST. Genetic and environmental interaction in allergy and asthma.
J Allergy Clin Immunol 1999;104(6):1139-46.

Weisglas_Kuperus N, Sas TC, Koopman_Esseboom C, van_der_Zwan
CW, De_Ridder MA, Beishuizen A, Hooijkaas H, Sauer PJ. Immunologic

40

63.

64.

65.

66.

67.

68.

69.

70.

71.

72.

 

effects of background prenatal and postnatal exposure to dioxins and
polychlorinated biphenyls in Dutch infants. Pediatric Research
1995;38(3):404-10.

Dewailly E, Ayotte P, Bruneau S, Gingras S, Belles_lsles M, Roy R.
Susceptibility to infections and immune status in Inuit infants exposed to
organochlorines. Environmental Health Perspectives 2000;108(3):205—1 1.

Belles_lsles M, Ayotte P, Dewailly E, Weber JP, Roy R. Cord blood
lymphocyte functions in newborns from a remote maritime population
exposed to organochlorines and methylmercury. 2002;65(2):165-82.
Reichrtova E, Ciznar P, Prachar V, Palkovicova L, Veningerova M. Cord
serum immunoglobulin E related to the environmental contamination of
human placentas with organochlorine compounds. Environ Health
Perspect 1999;107(11):895-9.

Karmaus W, Kuehr J, Kruse H. Infections and atopic disorders in
childhood and organochlorine exposure. Arch Environ Health
2001;56(6):485-92.

Karmaus W, Brooks KR, Nebe T, Witten J, Obi-Osius N, H. K. Immune
function biomarkers in children exposed to lead and organochlorine
compounds: a cross-sectional study. Environ Health. 2005 Apr 14;4(1):5. .
Environ Health. 2005;I4(1):5.

Brooks K, Samms-Vaughan M, Karmaus W. Are oral contraceptive use
and pregnancy complications risk factors for atopic disorders among
offspring? Pediatric Allergy and Immunology 2004;15(6):487-496.

Frye C, Mueller JE, Niederrneier K, stt M, Heinrich J. Maternal oral
contraceptive use and atopic diseases in the offspring. Allergy
2003;58(3):229-32.

Gold DR, Burge HA, Carey V, Milton DK, PIatts-Mills T, ST. W. Predictors
of repeated wheeze in the first year of life: the relative roles of cockroach,
birth weight, acute lower respiratory illness, and maternal smoking. . Am J
Respir Crit Care Med 1999;160:227-36.

Nafstad P, Kongerud J, Botten G, Hagen JA J, JJ. a. The role of passive
smoking in the development of bronchial obstruction during the first 2
years of life. . Epidemiology 1997;82293—7.

Jaakkola JJ, MS. J. Effects of environmental tobacco smoke on the

respiratory health of children. . Scand J Work Environ Health
2002;28(Supp| 2):71-83.

41

73.

74.

75.

76.

77.

78.

79.

80.

81.

82.

83.

London SJ, James Gauderrnan W, Avol E, Rappaport EB, JM. P. Family
history and the risk of early-onset persistent, early-onset transient, and
late-onset asthma. . Epidemiology 2001;12:577-83.

Gilliland FD, Li YF, JM. P. Effects of maternal smoking during pregnancy
and environmental tobacco smoke on asthma and wheezing in children.
Am J Respir Crit Care Med 2001;163:429-36.

Arruda LK, Sole D, Baena-Cagnani CE, Naspitz CK. Risk factors for
asthma and atopy. Curr Opin Allergy Clin Immunol 2005;5(2):153-9.

Jaakkola JJ, Gissler M. Maternal smoking in pregnancy, fetal
development, and childhood asthma. . Am J Public Health 2004;94:136-
40.

Moshammer H, Hoek G, Luttmann—Gibson H, al. e. Parental Smoking and
Lung Function in Children: an International Study. . Am J Respir Crit Care
Med 2006.

Sherrill DL, Martinez FD, Lebowitz MD, Holdaway MD, Flannery EM,
Herbison GP, Stanton WR, Silva PA, Sears MR. Longitudinal effects of
passive smoking on pulmonary function in New Zealand children. Am Rev
Respir Dis 1992; 1 45(5):1 1 36-41.

Cunningham J, Dockery D, Speizer FE. Maternal Smoking during
Pregnancy as a Predictor of Lung Function in Children American Journal
of Epidemiology 1994;139(12):1139-1152

Takemura Y, Sakurai Y, Honjo S, Kusakari A, Hara T, Gibo M, Tokimatsu
A, Kugai N. Relation between breastfeeding and the prevalence of asthma
: the Tokorozawa Childhood Asthma and Pollinosis Study. American
Journal of Epidemiology 2001;154(2):115-9.

Gdalevich M, Mimouni D, David M, Mimouni M. Breast-feeding and the
onset of atopic dermatitis in childhood: A systematic review and meta-
analysis of prospective studies. Journal of the American Academy of
Dermatology 2001;45(4):520-7.

Gdalevich M, Mimouni D, Mimouni M. Breast-feeding and the risk of
bronchial asthma in childhood: a systematic review with meta-analysis of
prospective studies. Journal of Pediatrics 2001;139(2):261-6.

Saarinen UM, Kajosaari M. Breastfeeding as prophylaxis against atopic

disease: prospective follow-up study until 17 years old. Lancet
1995;346(8982):1065-9.

42

84.

85.

86.

87.

88.

89.

90.

91.

92.

93.

94.

95.

Peat JK. Prevention of asthma. Eur Respir J 1996;9(7):1545—55.

Oddy WH, Holt PG, Sly PD, Read AW, Landau Ll, Stanley FJ, Kendall
GE, Burton PR. Association between breast feeding and asthma in 6 year
old children: findings of a prospective birth cohort study. ij
1999;319(7213):815-9.

Wafula EM, Limbe MS, Onyango FE, Nduati R. Effects of passive smoking
and breastfeeding on childhood bronchial asthma. East Afr Med J
1999;76(11):606-9.

Taylor B, Wadsworth J, Golding J, Butler N. Breast feeding, eczema,
asthma, and hayfever. Journal Of Epidemiology And Community Health
1983;37(2):95-9.

Midwinter RE, Morris AF, Colley JR. Infant feeding and atopy. Arch Dis
Child 1987;62(9):965-7.

Pratt WR. Allergic diseases in pregnancy and breast feeding. Ann Allergy
1981 ;47(5 Pt 1):355-60.

Pratt HF. Breastfeeding and eczema. Early Hum Dev 1984;9(3):283-90.

stt M, Dold S, Reitmeier P, Wulff A, Nicolai T, von Mutius E. [Does
breast feeding prevent asthma and allergies? Results of the Munich
asthma and allergy study]. Monatsschr Kinderheilkd 1992;140(10):769—74.

Rusconi F, Galassi C, Corbo GM, Forastiere F, Biggeri A, Ciccone G,
Renzoni E. Risk factors for early, persistent, and late-onset wheezing in
young children. SIDRIA Collaborative Group. Am J Respir Crit Care Med
1999;160(5 Pt 1):1617-22.

Peters TJ, Golding J. The epidemiology of childhood eczema: lI. Statistical
analyses to identify independent earIy predictors. Paediatric and Perinatal
Epidemiology 1987; 1 (1 )280-94.

Hanson LA, Korotkova M, Haversen L, Mattsby-Baltzer l, Hahn-Zoric M,
Silfverdal SA, Strandvik B, Telemo E. Breast-feeding, a complex support
system for the offspring. Pediatr Int 2002;44(4):347-52.

Bertotto A, Gerli R, Fabietti G, Crupi S, Arcangeli C, Scalise F, Vaccaro R.

Human breast milk T lymphocytes display the phenotype and functional
characteristics of memory T cells. Eur J lmmunol 1990;20(8):1877-80.

43

96.

97.

98.

99.

100.

101.

102.

103.

104.

105.

106.

107.

Takahata Y, Takada H, Nomura A, Nakayama H, Ohshima K, Hara T.
Detection of interferon-gamma-inducible chemokines in human milk. Acta
Paediatr 2003;92(6):659-65.

Michel FB, Bousquet J, Greillier P, Robinet_Levy M, Coulomb Y.
Comparison of cord blood immunoglobulin E concentrations and maternal
allergy for the prediction of atopic diseases in infancy. The Journal of
Allergy and Clinical Immunology 1980;65(6):422-30.

Shah 8, MM. B. Parental history of allergy, maternal serum lgE & cord
serum lgE. Indian J Med Sci. 2006 60(1):13-8.

Bergmann RL, Schulz J, GUnther S, Dudenhausen JW, Bergmann KE,
Bauer CP, Dorsch W, Schmidt E, Luck W, Lau S, al e. Determinants of
cord-blood lgE concentrations in 6401 German neonates [see comments].
Allergy 1995;50(1):65-71.

Johnson CC, Ownby DR, Peterson EL. Parental history of atopic disease
and concentration of cord blood lgE. Clin Exp Allergy 1996;26(6):624-9.

Kuiper S, Muris JWM, Dompeling E, van Schayck CP, Schonberger
HJAM, Wesseling G, Knottnerus JA. Association between first-degree
familial predisposition of asthma and atopy (total lgE) in newborns. Clinical
& Experimental Allergy 2006;36(5):594-601.

Steimer T. Reproductive health. Ares-Serono Symposia Series - Frontiers
in Endocrinology. Vol. 2. Rome: Ares Serono Symposia Publications,
1993.

Waring RH, Harris RM. Endocrine disrupters: a human risk? Mol Cell
Endocrinol 2005;244(1-2):2-9.

Choi SM, Yoo SD, Lee BM. Toxicological characteristics of endocrine-
disrupting chemicals: Developmental toxicity, carcinogenicity, and
mutagenicity. Journal of Toxicology and Environmental Health-Part B-
Critical Reviews 2004;7(1):1-32.

stt M, Dold S. Is asthma an endocrine disease? Pediatric Allergy and
Immunology 1997;8(4):200-4.

Druckmann R. Review: female sex steroid hormones, autoimmune
diseases and immune response. Gynecol Endocrinol 2001;15 Suppl 6:69-
76.

Ackerman LS. Sex steroid hormones and the genesis of autoimmunity.
Arch Dermatol 2006;142(3):371-6.

44

108.

109.

110.

111.

112.

113.

114.

115.

116.

117.

Verthelyi D. Sex steroid hormones as immunomodulators in health and
disease. INTERNATIONAL IMMUNOPHARMACOLOGY 2001;1(6):983-
993.

Wilder RL, Elenkov IJ. Hormonal regulation of tumor necrosis factor-alpha,
interleukin-12 and interleukin-10 production by activated macrophages. A
disease-modifying mechanism in rheumatoid arthritis and systemic lupus
erythematosus? Ann N Y Acad Sci 1999;876:14-31.

Szekeres-Bartho J, Wegmann TG. A progesterone-dependent
immunomodulatory protein alters the Th1/Th2 balance. J Reprod lmmunol
1996;31(1-2):81-95.

Piccinni MP, Giudizi MG, Biagiotti R, Beloni L, Giannarini L, Sampognaro
S, Parronchi P, Manetti R, Annunziato F, Livi C. Progesterone favors the
development of human T helper cells producing Th2-type cytokines and
promotes both lL-4 production and membrane CD30 expression in
established Th1 cell clones. Journal of Immunology (Baltimore, Md. :
1950) 1995;155(1):128-33.

Szekeres-Bartho J, Par G, Szereday L, Smart CY, Achatz I. Progesterone
and non-specific immunologic mechanisms in pregnancy. American
Journal Of Reproductive Immunology 1997;38(3):176-82.

Hamano N, Terada N, Maesako K, Hohki G, Ito T, Yamashita T, Konno A.
Effect of female hormones on the production of lL-4 and lL-13 from
peripheral blood mononuclear cells. Acta Otolaryngol Suppl 1998;537:27-
31.

Piccinni MP, Maggi E, Romagnani S. Role of hormone-controlled T-cell
cytokines in the maintenance of pregnancy. Biochem Soc Trans
2000;28(2):212-5.

Safe SH, Zacharewski T. Organochlorine exposure and risk for breast
cancer. Progress In Clinical And Biological Research 1997;396:133-45.

Safe SH. Xenoestrogens and breast cancer. N Engl J Med
1997;337(18):1303-4.

vom Saal FS, Timms BG, Montano MM, Palanza P, Thayer KA, Nagel SC,
Dhar MD, Ganjam VK, Parmigiani S, Welshons VW. Prostate enlargement
in mice due to fetal exposure to low doses of estradiol or diethylstilbestrol
and opposite effects at high doses. Proc Natl Acad Sci U S A
1997;94(5):2056-61.

45

118.

119.

120.

121.

122.

123.

124.

125.

126.

127.

128.

129.

130.

Welshons WV, Nagel SC, Thayer KA, Judy BM, Vom Saal FS. Low-dose
bioactivity of xenoestrogens in animals: fetal exposure to low doses of
methoxychlor and other xenoestrogens increases adult prostate size in
mice. Toxicol Ind Health 1999;15(1-2):12-25.

Ashby J, Tinwell H, Haseman J. Lack of effects for low dose levels of
bisphenol A and diethylstilbestrol on the prostate gland of CF1 mice
exposed in utero. Regul Toxicol Pharmacol 1999;30(2 Pt 1):156-66.

Odum J, Ashby J. Neonatal exposure of male rats to nonylphenol has no
effect on the reproductive tract. Toxicol Sci 2000;56(2):400-4.

Africa). ACfEAaGC. Approaches to Effective Malaria Control that Avoid
DDT in Kenya: Use of Bacillus thuringiensis israelensis (BTi). 2006.

ATSDR. Toxicological Profile for DDT/DDD/DDE. Update. (Draft for Public
Comment). In: Registry AfTSaD, ed. Atlanta, GA, 2000;433 pp.

Sunyer J, Torrent M, Munoz-Ortiz L, Ribas-Fito N, Carrizo D, Grimalt J,
Anto JM, Cullinan P. Prenatal dichlorodiphenyldichloroethylene (DDE) and
asthma in children. Environ Health Perspect 2005;113(12):1787-90.

Woodruff T, Wolff MS, Davis DL, Hayward D. Organochlorine exposure
estimation in the study of cancer etiology. Environ Res 1994;65(1):132-44.

Longnecker MP, Rogan WJ, Lucier G. The human health effects of DDT
(dichlorodiphenyltrichloroethane) and PCBS (polychlorinated biphenyls)
and an overview of organochlorines in public health. Ann Rev Public
Health 1997;18:211-44.

Chedrese PJ, Feyles F. The diverse mechanism of action of
dichlorodiphenyldichloroethylene (DDE) and methoxychlor in ovarian cells
in vitro. Reprod Toxicol 2001;15(6):693-8.

Kelce WR, Wilson EM, Lambright CR, Gray LE, Jr., Roberts KP, Stone
CR, Laws SC, Gray LE, Kemppainen JA.. J Mol Med 1997;75(3):198-207.

Kelce WR, Lambright CR, Gray LE, Jr., Roberts KP, Stone CR, Laws SC,
Gray LE, Kemppainen JA, Wilson EM. Toxicol Appl Pharmacol
1997;142(1):192-200.

Hansen LG. Stepping backward to improve assessment of PCB congener
toxicities. Environmental Health Perspectives 1998;106 Suppl 1:171-89.

Ulrich EM, Caperell-Grant A, Jung SH, Hites RA, Bigsby RM.
Environmentally relevant xenoestrogen tissue concentrations correlated to

46

131.

132.

133.

134.

135.

136.

137.

138.

139.

140.

141.

biological responses in mice. Environ Health Perspect 2000;108(10):973-
7.

Burow ME, Tang Y, Collins-Burow BM, Krajewski S, Reed JC, McLachlan
JA, Beckman BS. Effects of environmental estrogens on tumor necrosis
factor alpha- mediated apoptosis in MCF-7 cells. Carcinogenesis
1999;20(11):2057-61.

ATSDR. ToxFAQsTM for Polychlorinated Biphenyls (PCBs)

Safe 8. Polychlorinated biphenyls (PCBs), dibenzo-p-dioxins (PCDDs),
dibenzofurans (PCDFs), and related compounds: environmental and
mechanistic considerations which support the development of toxic
equivalency factors (TEFs). Crit Rev Toxicol 1990;21 (1 ):51-88.

Patnode KA, Curtis LR. 2,2',4,4’,5,5'- and 3,3',4,4',5,5'-hexachlorobiphenyl
alteration of uterine progesterone and estrogen receptors coincides with
embryotoxicity in mink (Mustela vision). Toxicol Appl Pharmacol
1994;127(1):9-18.

Wojtowicz AK, Gregoraszczuk EL, Lyche JL, Ropstad E. Time dependent
and cell-specific action of polychlorinated biphenyls (PCB 153 and PCB
126) on steroid secretion by porcine theca and granulosa cells in mono-
and co-culture. J Physiol Pharmacol 2000;51 (3)1555-68.

Wojtowicz A, Ropstad E, Gregoraszczuk E. Estrous cycle dependent
changes in steroid secretion by pig ovarian cells in vitro to polychlorinated
biphenyl (PCB 153). Endocr Regul 2001;35(4):223-8.

Troisi GM, Mason CF. PCB-associated alteration of hepatic steroid
metabolism in harbor seals (Phoca vitulina). J Toxicol Environ Health A
2000;61(8):649-55.

Haave M, Ropstad E, Derocher AE, Lie E, Dahl E, Wiig O, Skaare JU,
Jenssen BM. Polychlorinated biphenyls and reproductive hormones in
female polar bears at Svalbard. Environ Health Perspect 2003;111(4):431-
6.

Holt PG, Macaubas C. Development of long-term tolerance versus
sensitisation to environmental allergens during the perinatal period. Curr
Opin lmmunol 1997;9(6):782-7.

Strachan DP. ls allergic disease programmed in early life? Clinical And
Experimental Allergy 1994;24(7):603—5.

Bjorksten B, Kjellman NI. Perinatal environmental factors influencing the
development of allergy. Clin Exp Allergy 1990;20 Suppl 3:3-8.

47

—_.._

--.“--7

142.

143.

144.

145.

146.

147.

148.

149.

150.

151.

152.

Barker DJ. In utero programming of chronic disease. Clin Sci (Lond)
1998;95(2):115—28.

Wills-Karp M, Chiaramonte M. Interleukin-13 in asthma. Curr Opin Pulm
Med 2003;9(1):21-7.

Evans PC, Lambert N, Maloney S, Furst DE, Moore JM, Nelson JL. Long-
term fetal microchimerism in peripheral blood mononuclear cell subsets in ‘
healthy women and women with scleroderma. Blood 1999;93(6):2033-7.

Gabrielsson S, Soderlund A, Nilsson C, Lilja G, Nordlund M, Troye-
Blomberg M. Influence of atopic heredity on |L-4-, IL-12- and IFN-gamma-
producing cells in in vitro activated cord blood mononuclear cells. Clinical
and Experimental Immunology 2001;126(3):390-396.

t;.‘___ 1'

Cohn B. DDT and DDE: Second Generation Time to Pregnancy Effects.

Lundholm CE. The effects of DDE, PCB and chlordane on the binding of
progesterone to its cytoplasmic receptor in the eggshell gland mucosa of
birds and the endometrium of mammalian uterus. Comp Biochem Physiol
C 1988;89(2):361-8.

Noakes PS, Taylor P, Wilkinson S, Prescott SL. The relationship between
persistent organic pollutants in maternal and neonatal tissues and immune
responses to allergens: A novel exploratory study. Chemosphere
2006;63(8):1304-1311.

Xu B, Jarvelin MR, Hartikainen AL, Pekkanen J. Maternal age at
menarche and atopy among offspring at the age of 31 years. Thorax
2000;55(8):691-3.

Vihko R, Apter D. Endogenous steroids in the pathophysiology of breast
cancer. Critical Reviews in Oncology/Hematology 1989;9(1):1-16.

Apter D, Reinila M, Vihko R. Some endocrine characteristics of early
menarche, a risk factor for breast cancer, are preserved into adulthood.
International Journal of Cancer. Journal International Du Cancer
1989;44(5):783-7.

Moore JW, Key TJ, Wang DY, Bulbrook RD, Hayward JL, Takatani 0.
Blood concentrations of estradiol and sex hormone - binding globulin in
relation to age at menarche in premenopausal British and Japanese
women. Breast Cancer Research and Treatment 1991 ;18 Suppl 12847-50.

48

153.

154.

155.

156.

157.

158.

159.

 

Xu B, Pekkanen J, Husman T, Keski-Nisula L, Koskela P. Maternal sex
steroid hormones in early pregnancy and asthma among offspring: a case-
control study. The Journal of Allergy and Clinical Immunology
2003;112(6):1101-4.

Cooper GS, Martin SA, Longnecker MP, Sandler DP, Gerrnolec DR.
Associations between plasma DDE levels and immunologic measures in
African-American farmers in North Carolina. Environ Health Perspect
2004;112(10):1080-4.

Vine MF, Stein L, Weigle K, Schroeder J, Degnan D, Tse CK, Backer L.
Plasma 1,1-dichloro-2,2-bis(p-ch|orophenyl)ethylene (DDE) levels and
immune response. Am J Epidemiol 2001;153(1):53-63.

Bilrha H, Roy R, Moreau B, Belles-Isles M, Dewailly E, Ayotte P. In vitro
activation of cord blood mononuclear cells and cytokine production in a
remote coastal population exposed to organochlorines and methyl
mercury. Environ Health Perspect 2003;111(16):1952-7.

Daniel V, Huber W, Bauer K, Suesal C, Conradt C, Opelz G. Associations
of dichlorodiphenyltrichloroethane (DDT) 4.4 and
dichlorodiphenyldichloroethylene (DDE) 4.4 blood levels with plasma IL-4.
Arch Environ Health 2002;57(6):541-7.

Kavlock RJ, Daston GP, Derosa C, et al., Research needs for the risk
assessment of health and environmental effects of endocrine disruptors
Environ Health Perspect 1996;104 (suppl 4), 715-740.

EC (1997) European Workshop on the Impact of Endocrine Disrupters on

Human Health and Wildlife: Report of the Proceedings (EUR 17549),
Weybridge, UK, European Commission

49

kw...”

CHAPTER 2

BACKGROUND AND METHODS
FOR THE

RESPECTIVE STUDIES

50

The four research questions posed in this monograph utilizes data from three
cohorts: namely the Pregnancy Environment and Child Health (PEACH) study,
the Jamaican Perinatal Morbidity, Mortality Survey and the Child Health and

Environment Cohort Study-Germany. This chapter describes the background and

methods used the respective studies. Methodologic details not covered in this

hf-‘IW'I .

chapter are presented in the four respective manuscripts found in the ‘Results’

secfion.

2.1 The PEACH study - Background
PEACH is a multi-center longitudinal study of mother-infant pairs, which had its
inception in Berrien County Michigan (Figure 4) at the Benton Harbor and St.

Joseph study sites.

Situated along the South East shore of Lake Michigan in Berrien County, Benton

Harbor and St. Joseph are two adjoining but very different cities. According to

City-Data.com 1, in 2004, Benton Harbor had over 10, 000 residence (92.4% are

African-American) with a median household income 0f approximately $17,500.

51

 

   
 

E Michigan State University
_ Berrien County - PEACH Study Site
Other Counties (Not Involved)

 

 

 

Figure 4. Map of Michigan showing PEACH Study Site

52

In 1894, the Baushke family, prominent citizens in Benton Harbor, built America’s
first car, with 4-wheels and seating for 5 people. This was the inception of Benton
Harbor’s industries which now includes the headquarters for the world renowned
Whirlpool Corporation formally known as the Upton Machine Co. As in any other
towns or cities, the introduction of industries is good news for the economy; l
however their possible impact on public health is of grave concern.

St. Joseph on the other hand in 2004 had an estimated population of 9,000 of

aim” .1.

which 89.6% are Caucasian. Median household income for the year 2000 was

$37,032 1. Its ideal location at the mouth of the St. Joseph River (linking it to

Chicago) earned the city the title ‘The Riviera of the Midwest’. This location
encourages much commerce and tourism. Fertile lands boast orchards full of
apples, peaches and cherries along with vineyards of grapes, fields of corn, rows
of strawberries and blueberries. Little wonder St. Joe (as it is affectionately
called) is the largest non-citrus fruit-growing region in the United States. There
are also several industries (In 1911 the Uptons started their company, now
Whirlpool - later moved to Benton Harbor) dotting the landscape, hence as is the

case of its ‘sister’ city Benton Harbor, a possible risks to public health looms.

The cities of Benton Harbor and St. Joseph are ideal sites for studying the
human health effects of exposure to environmental sex steroid hormones for the
following reasons. First, both cities are adjacent to each other; second, they are

located along the South East shore of Lake Michigan, a known ‘reservoir’ for

environmental toxicants and; third, the discrete ethnic and socio-economic
differences between cities.

It is in this setting that the PEACH study had its ‘genesis’ in May of 2003. The
target population was women in their first trimester of pregnancy living in Benton
Harbor or St. Joseph. The study population was women from target population
who met the inclusion criteria of being primiparous (expecting first live birth),
having no previous stillbirth, no multiple births, no diabetes, thyroid or adrenal

disorders, and age 18-34 years.

The PEACH study - Methods

Questionnaires

At the time of enrollment, a brief face-to-face interview was conducted to collect
data on the mother’s lifestyle and mediation usage. A more detailed telephone
interview was then conducted two weeks after to collect data on participants’
general and reproductive health. Approximately two weeks after delivery, another
telephone interview was conducted that focuses on the mother’s pregnancy
experience such as duration of pregnancy and delivery, and complications during
pregnancy. Information on the mother’s general health along with that of her

newborn was also ascertained.

Cord blood and placenta tissue sampling
Following delivery of the placenta, the external surface of the cord was washed
to avoid contamination with maternal cells, and then approximately 1 OmL cord

blood was collected by cordocentesis in heparinized tubes. The tube with blood

54

is then gently shaken to allow proper mixing of heparin and cord blood. The
mixture was then refrigerated and prepared for mononuclear cell separation

within 24 hours of collection.

Within an hour of delivery, approximately one quarter of placental tissue (1009)

was cut and stored in a glass container for the determination of DDE and other

organochlorine compounds. The sample was then stored at —20°C until analyzed.

Laboratory analyses of placental samples
All organochlorine analyses were done at the Analytical Chemistry Section
(ACS), Michigan Department of Community Health. Compounds were

extracted, fractionated and quantified according to procedures described by

Najam et al. 2, with modifications.

Laboratory analyses of cord plasma samples

Fresh cord blood was diluted with equal volume of sterile saline, carefully layered
onto Histopaque-1077 (Sigma Chemical Company, St. Louis, MO) and
centrifuged 30 minutes (at 600 G) at room temperature. The cells harvested from
the interface were washed in serum free saline, then re-suspended into 2-3 ml of
RPMI 10 (Gibco BRL, Life Technologies Inc., New York, USA) and the number of
viable cells determined using trypan blue exclusion. CBMC staining will also be
performed with Turk’s stain to exclude nucleated erythrocytes from cell counts

and determine the degree of contamination. PHA-P (Sigma Chemical Company),

55

as a polyclonal activator at 5 ug/mL in culture, and dust mite allergen (Greer
Labs), at 100 and 500 ug/mL, were used in cell cultures. Cells were cultured in
triplicates using 96-well flat bottom plates in a final volume of 200 ill/well at a final

concentration of 1 x 106 leukocytes/mL.

Cytokine assays: Cytokine protein levels were measured using cytokine-
specific, sandwich enzyme-linked immunoassays (ELISAs).

IL-4 and lL-13: Recombinant human lL-4 and lL-13 standards of specific activity
equivalent to that of WHO standard 88/656 (21) were included in each ELISA.
The sensitivity of this assay was 0.9 to 1.9 pg/ml and the linear range of

detection between 0.9 to 250 pg/mL.

IFN-y. Similar procedure to that used for lL-13 was used to quantify IF N-y.
However, the lower limit of detection for the assay was typically 0.3 mm and the
linear range for quantitation was between 0.3 to 25 UlmL. This assay was
calibrated using human lFN-y reference reagent G923—901—530 (specific activity
7 x 105 U/mg, 1 NIH unit = 115 pg, provided by Dr. C. Laughlin, NIAID,

Bethesda, Md.).

The Institutional Review Board of the Michigan State University approved the

study in addition; all participants gave written, informed consent.

56

firm—- *1!

Statistical analysis

Adjusted betas (l3) and standard errors (SE) for the associations between
exposure and outcomes were determined using linear regression models.
Statistical tests are 2-sided and declared significant if p value < 0.05. The

statistical software package SAS (version 9.1; SAS Institute Inc, Cary, NC)

3
was used for all analyses .

Linear regression

In its simplest form a linear regression analysis involves finding the best straight
line relationship that explains how the variation in an outcome (or dependent)
variable, Y, depends on the variation in a predictor (or independent) variable, X,
and is expressed by the following equation:

Y = b0 + b1X

where b1 is called the slope of the equation and b0 the intercept.

From a sample {(Yi, Xi) : | = 1, ...,n}, b1, be are estimated by
3 XXX. — (2X.- KEY.)
1 2
2X12 — (2X: )2

 

be is called the intercept and estimated by the following formula;

A ZY—QZX
b0:

 

n
This method is called the process of Least Squares that calls for minimization

n n
with respect to (b0, b,) of the error sum of squares; Zeiz : 20”. — b0 — b139,)2

i=1 i=1

57

'

Linn

-__.

which leads to the expression forbAO. Ii, . Since most outcomes are dependent on

more than one factor (variable) it is necessary to explain the outcome in terms of
the combined effect of these multiple independent factors or variables. This is
done by multiple regression which is an extension of simple linear regression in
which the dependent variable is predicted by a linear combination of the
explanatory variables. For example, if an outcome Y depends on two
independent variables X1 and X2, the regression equation will be Y = be + b1X1 +
bzxz where the Do is the intercept, and b1 and b2 are slopes of X1 and X2
respectively.

The amount of variation in the dependent variable that is accounted for by
variation in the predictor variables is measured by the value of the coefficient of
determination, or R2 (R squared). The R2 from multiple regression models is
called Adjusted R2 when penalized for the number of variables in the model. This
Adjusted R2 is the total percent variation in the dependent variable that is

explained by the predictor variables together.

58

2.2 Child Health and Environment Cohort Study - Germany.

The Child Health and Environment Cohort Study was conducted in 1994,

targeting children residing in 18 townships in Hesse, Germany.

Hesse is one of Germany's sixteen federal states and has an area of 21,1 10 km2

and just over six million inhabitants. Most inhabitants live in the southernmost

part of Hesse between the rivers Main and Rhine. The latter is one of the longest

and most important rivers in
Europe and borders Hesse
on the southwest without
running through the state.
The valley formed by the
Rhine (Rhine Valley) in
Hesse is where two of the

regions targeted in this

study are situated 4. Figure

5 shows the location of

 

Worms

 

 

Rhine Valley-control
Toxlc Waste lnclnerator-reglon
Odenwald-control

 

 

O Darmstadt

 

Figure 5: The Rhine Valley

study sites and the toxic waste incinerator in the Rhine Valley. Two regions are

situated in the Rhine Valley with low mountains on both sides.

59

 

For details regarding the Questionnaires and Laboratory procedures used in
this study, please see the manuscript entitled ‘Immune function biomarkers in
children exposed to lead and organochlorine compounds: a cross-sectional

study in the ‘Results’ section.

Statistical analysis

The statistical software package SAS (version 9.1; SAS Institute Inc, Cary, NC)

was used for all analyses 3. The linear regression method is described in pages

39 and 40. Statistical tests are 2-sided and decleared important if p value < 0.05.

60

2.3 The Jamaican Perinatal Morbidity, Mortality Survey - Background

The Jamaican Perinatal Morbidity, Mortality Survey targeted all women who had
a live birth, or stillbirth of 500 g or more in Jamaica during the period September

1, 1986 to October 31, 1986.

Jamaica is an island nation 240 kilometres (150 miles) in length and 85
kilometres (50 miles) in width, which is slightly smaller than Connecticut-USA.
The island is situated in the Caribbean Sea, approximately 804 kilometres (500
miles) south south east of the southern tip of Florida (Figure 5). Jamaica, with its
mostly mountains interior and narrow discontinuous coastal plains, enjoys a
mainly tropical climate with a temperate interior (average minimum temperature
220°C (71.6°F) and average maximum of 303°C (86.5°F)). The population
(N=2,758,124) is mainly African-Caribbean (90.9%) of West African descent with
a labor force (approximately 1.2 million) made up of workers in services (64.1%;
mainly tourism), agriculture (19.3%) and industry (16.6%; mainly

bauxite/alumina).

The Jamaican Perinatal Morbidity, Mortality Survey was motivated by results of
the Child Mortality Study of 1972 that clearly demonstrated the extent of under-
reporting of early infant deaths. Shortly after the island wide survey (Figure 6)

was designed using the First British Perinatal Mortality Survey 5 as a blue print.

61

 

.quoE. “mun ”322.4 5 ecu :33:qu _. ”octofino «mu—z. "Summ 530m .ucqun—ue ”has. “mum
Ec< umuo swam 5.52 doused—On: “Layne—.052"? ”finned—m ”mu—... ”530w .Eiflofinh HoEu... «mum ro>oca1um
65.955335qu Soc; .3533: .0 5:45:35 use use; 3:23.58? .2369. 623:3 ”saunas .0 2:9“.

 

I

 

 

00:8 O_oow

 

 

 

       

(mm Zdwmmamdo 08

Any

40323..

  

88 2.538

    

952 23238:
100m 0 . u
~.¢§ e 0
In,
258 8238 .4...
3535 a o
33:43.
p .o

  

oOh
—

 

 

 

 

 

 

62

The overall aim of the study was to provide information to help improve maternal

and childcare services and reduce perinatal morbidity and mortality in Jamaica.

The Jamaican Perinatal Morbidity, Mortality Survey - Methods

At enrollment, a face-to-face interview was conducted with each prticipant and
their babies examined, usually within the first 48 hours after delivery. A
standardized questionnaire was used to get at data during the antenatal, labor
and delivery, and perinatal periods. Data was also abstracted from the newborn
nursery charts and mother’s medical record to supplement interview data where

necessary. Overall, 10,310 (94%) of the births in the two-month period were

identified and included in the study (main cohort study) 5.

The first follow-up occurred six weeks after delivery when another face-to-face
interview was conducted aimed at information about breastfeeding practices and

the infant’s health status.

The study included in this monograph utilizes data from a geographic sub-sample
(n= 1,720) of eleven to twelve year old children, representing those born in
Kingston and St. Andrew. In a cross-sectional survey of these children, data on
health outcomes were collected from the guardian or parent; who was often time

the mother.

The Research Ethics Committees of the University of the West Indies Jamaica
and Michigan State University approved the assembly of the cohort and analysis

of data for the included manuscript, respectively.

63

 

 

Statistical analysis

The statistical software package SAS (version 9.1; SAS Institute Inc, Cary,

NC) was used for all analyses 3. The method of statistical analysis used

was logistic regression with 2-sided statistical tests declaring associations

significant if p value < 0.05.

Logistic regression

Logistic regression is an adaptation to dichotomous outcomes of the classical
linear regression model for continuous outcomes. It is primarily used to describe
the relationship to the outcome of a primary exposure after adjusting for the
effects of other (independent) variables that may influence both the outcome and

exposure.

Formally, in formulating a logistic model let Y (0 or 1) denoting the dichotomous
outcome of interest with Y =1 labeling the presence of the condition under study.
Then the logistic model for P[Y = l | X 1 , X 2 X k] is written simply as P(X)
where X is a shorthand for X1 through Xk, The model formula is given by

__ 1
PL)_ -(0t+Zl3iXi)

I+e

 

in which or and B, are unknown parameters to be

estimated based on the observed exposure variables X1 through Xk and the
outcome variable Y. The coefficients [31 through [3,. can be interpreted as adjusted
log-odds ratios. For example, if X1 is the primary exposure, coded X1 = 1 for
“exposed” and X1 =0 for “not exposed”, then assuming the variables X2 through

Xk do not include X1, the adjusted odds ratio is given by exp(i31).

64

References

1.

2.

City-Data.com. www.City-Data.com.

Najam AR, Kover MP, Williams CC, Burse VW, Needham LL. Analysis of
a Mixture of Polychlorinated Biphenyls and Chlorinated Pesticides in
Human Serum by Column Fractionation and Dual-Column Gas
Chromatography with Electron Capture Detection. Journal of the AOAC
International 1999;82:177-185.

SAS Institute. Statistical Analysis System, Version 9.1. In: Institute S, ed.
Gary, NC, 2002-2003.

Wikipedia. Hesse. 2006.
Ashley D, McCaw-Binns A, Golding J, Keeling J, Escoffery C, Coard K,

Foster-Williams K. Perinatal mortality survey in Jamaica: aims and
methodology. Paediatr Perinat Epidemiol 1994;8 Suppl 1:6—16.

65

'*

h;m w .'
,1 A

CHAPTER 3
RESULTS

66

 

In this chapter, four manuscripts are presented. Their titles and order are:

. Placental p, p'-dichlorodiphenyldichloroethylene and cord blood immune

markers (forthcoming).

Maternal oral contraceptive use and immune markers in the offspring (In
progress).

Are oral contraceptive use and pregnancy complications risk factors for
atopic disorders among offspring? (Published)

Immune function biomarkers in children exposed to lead and

organochlorine compounds: a cross-sectional study (Published).

These manuscripts correspond to the four research questions:

1.

Is there evidence for the immunemodulation of cord immune markers of

allergy by in utero sex hormone exposure?

. Is maternal oral contraceptive use (a proxy for in utero

estrogen/progesterone exposure) associated with immune markers of

allergy in the offspring?

. Is maternal oral contraceptive use (a proxy for in utero

estrogen/progesterone exposure) a risk factor for markers of allergic

susceptibility in offspring?

4. Does serum organochlorine alters immune markers in children?

67

COMPONENT ONE

In utero exposure to SHD affects perinatal biomarkers of allergy

68

Placental p, p'-dichlorodiphenyldichloroethylene and cord blood immune
markers

Kevin Brooks1, Hanem Hasanz, Sridhar Samineniz, Venu Gangur2 and Wilfried
Karmaus1

1 Department of Epidemiology and Biostatistics,
Norman J. Arnold School of Public Health,
University of South Carolina

800 Sumter Street, Columbia, SC., 29208-0001

2 Food Allergy & Immunology Laboratory, Food Science 8. Human Nutrition;
National food Safety & Toxicology Center, Michigan State University, East
Lansing, MI 48824, USA

Abstract

Placental p, p'-dichlorodiphenyldichloroethylene (p,p’-DDE) concentration and
cord blood atopic markers were determined in 19 neonates. Increased placental
p,p’-DDE was associated with a statistically Significant increase in cord plasma
interleukin (lL)-13.

Furthermore, both cord plasma IL-4/lNF-y and lL-13/lNF-y ratios were

Significantly positively associated with placental p,p’-DDE concentration.

Key words:

DDE, placenta, cytokines, cord, allergy

69

The possible immune modulating role of ‘hormone-like substances’ is under

investigation 1'2. Previously, in a cohort of German school children, we
demonstrated significant associations between serum DDE and biomarkers of
allergic susceptibility; including serum immunoglobulin E 2. In addition, for a
group of 26 mother-infant pairs, Noakes et al. reported on the ability of maternal
DDE and other ‘hormone-like’ substances to alter the fetal immune response 1.
More recently, Sunyer and co-workers showed that in utero DDE exposure is

related to increased incidence of asthma in children 3. It is therefore very timely

to present the yet-to-be-reported relation between placental p,p’-DDE and cord

plasma levels of prototypic type-1 and type-2 cytokines.

Our objective was to test the hypothesis whether increased placental p,p’-DDE
concentration is associated with an increase in cord plasma IL-4 and lL-13. In
addition, since type-1 and type-2 cytokines often antagonize each others activity,
we hypothesize that increased placental p,p’-DDE concentration is associated

with a reduction in cord plasma INF-y. The Th2/Th1 cytokine ratio has been used
as a measure of relative atopic reactivity 4. We therefore further hypothesize that

increased placental p,p’-DDE concentration is associated with an increase in the

ratio of cord plasma lL-4/lNF-y as well as the ratio of lL-13/lNF-y.

70

 

 

Methods and results

Primiparous women in their first trimester were enrolled in the ongoing
Environment and Child Health (PEACH) study. Ethical approval was granted by
the Michigan State University Committee on Research Involving Human

Subjects.

Telephone interview data on passive smoking (ETS), mother’s age, and allergic
status (defined as mother ever had one or any combination of: asthma, hay
fever, itchy rash, or wheezing) were collected. Cord blood samples were drawn
from the umbilical vein and transported to the lab within 24 hours. Cytokine
protein levels were measured using ultrasensitive Elisa based assays optimized
using paired monoclonal antibodies (PharMingen). Typical assay sensitivities; IL-

4: 0.05pg/ml; |L-13:1.5 pg/ml; and INF-y. 0.5pg/ml).

Placental p,p’-DDE was extracted into diethyl ether-hexane (1 :1 v/v), passed
over a Florisil column and quantified by high-resolution gas chromatography with
electron capture detection. p,p’-DDE values below the detection limit (DL = 0.50

pg/g) were replaced with half the DL and all concentrations were lipid adjusted.

We used linear regression models (PROC GLM) with cytokines as dependent
and p,p’-DDE as independent variables; while controlling for mother's age, atopic
status, and ETS. In order to obtain a multivariate normal distribution we log-

transformed cytokine levels.

71

Complete data (interview, cytokine and DDE concentration) are available for 19
mother-infant pairs. Mothers were on average 29 years of age while 2 of 19 were
exposed to passive cigarette smoke in their homes during pregnancy (Table 1).
Median value for placental p,p’-DDE was 58.3 pg/g. For lL-4 and lL-13 their
median values were 4.2 pg/mL and 61.4 pg/mL respectively (Table 1).

Adjusting for maternal age, allergy and exposure to tobacco smoke, we found

that placental p,p’-DDE statistically significantly increase the level of cord plasma

Phil-HM fl

lL-13 (B = 0.007 pg/g; p=0.03, Table 2). Placental p,p’-DDE was also found to
increase lL-4 levels while reducing INF-y; however these associations were not
statistically significant. Nevertheless, both cord plasma lL—4/INF-y and lL-13/INF-
y ratios were Significantly increased with increasing placental p,p’-DDE

concentration (8 = 0.03 pg/g; p=0.007 and i3 = 0.17 pg/g; p=0.02 respectively;

Table 2)

Discussion

Using placental and cord blood samples of 19 mother-infant pairs we
demonstrated, a statistically significant association between placental p,p’-DDE
and cord plasma lL-13. The results support our hypotheses that placental DDE is
related to increased cord plasma lL-13 and may be associated with increased IL-
4 and reduced INF-y. Considering the aforementioned results for lL-4, |L-13 and
INF-y, we expected placental DDE to be associated with the ratio of lL-4/lNF-y
along with that of lL-13/lNF-y. The associations (increased DDE and increased

Th2fTh1 ratios) were found to be statistically Significant.

72

DDE is known to cross the placenta. However, in their Western Australian
samples, Noakes et al. were unable to detect placental p,p’-DDE and used
maternal levels instead. This was not the case in our North American samples.
The fact that we determined p,p’-DDE in placental samples may be explained by
different exposure in the southern and northern hemisphere, resulting from a

long-range atmospheric transport of organochlorines enhanced by the meridional

flow 5.

Whereas both |L-4 and IL-13 play critical roles in the initiation as well as

maintenance of allergic inflammation and asthma 6, INF-y dominated responses

are associated with a ‘non-allergic’ response 7. Notably, lL-4 and lL-13 Share

many but not all proallergenlc activities. Recent studies suggest that lL-13 is a

key player in asthma independent of lL-4 6. It is suggested that decreased lFN-y
in cord blood mononuclear cells is the hallmark of newborns from high-risk

population 8. We are uncertain whether the relationships observed in this study
between cytokines and placental p,p’-DDE are direct or indirect. Maternal and

. . . . 9
fetal immune responses during pregnancy are under endocrine Influence . Thus,

it is possible that an explanation for our finding may be that DDE introduce an

endocrine disrupting effect 10, which in turn might alter the fetal T-cell cytokine

profile.

73

We are not aware of studies implicating direct effect of DDE on Th2fih1 cell
development or cytokine responses. Our findings suggest that the assessment of
DDE exposure on Th2/Th1 ratios may be of more importance than the

investigation of individual cytokines.

Noakes et al. reported a significant inverse correlation (r =0.406, p = 0.049)

between maternal adipose tissue levels of OC p,p’-DDE and maternal INF-y

reaction to mitogens 1. The authors also found increased phytohaemagglutinin

mitogen stimulated |L-13 and decreased IF N-y with increasing cord blood p,p’-
DDE levels. Though these findings, from their small sample (n=26), were not
statistically significant; they were in the same direction as those we presented
using a similar size cohort (n=19). Since these small studies reported
comparable findings, the effects are likely to gain statistical significance in a

study with a larger sample size and improved statistical power.

Our findings suggest that DDE may play a role in the prenatal priming of allergic
diseases via promoting lL-13 dominance. This priming — to a greater extent -
may also be through the overall dominance of the ratio of Type-2 cytokines (both
lL-4 and lL-13) over Type-1 (lFN-y) cytokine. Larger epidemiologic studies are

needed to explicate possible mechanistic explanations for our findings.

74

Acknowledgement
This work was supported by United States Environment Protection Agency STAR

grant number R830825—01-0.

75

Table 1. Descriptive characteristics of the cohort

 

 

 

Stuchohort
(n=19)
Maternal
Everhad:
Asthma Yes 1/19
Hay fever Yes 3/19
Itchy rash Yes 2/19
Itchy eyes or stuffy nose Yes 7/19
Wheezing Yes 3/19
Allergy Yes 10/19
Exposure to passive smoke in Yes 2/19
home
Ethnicity Caucasian 15/19
Age (mean, range, years) 29.3 (21-42)

Placental p,p’-DDE (median, 5-95% value, pg/g) 58.3 (40.2 — 388.4)

Cord plasma (median, 5-95% value, pg/mL):

lL-4 4.2 (0.7 — 29.8)
lL-13 61.4 (9.2 - 963)
INF-y 119.8 (0.35 — 1398.1)
lL-4/ INF-y ratio 0.05 (0.003 - 15.41)
lL-13/ INF-y ratio 0.72 (0.04 — 614.14)

 

76

 

683 a co ummmn mod v Q ..
8:5: E 96:5 8283 2 mSmoaxm new >925 .mmm 9559: Lo.— noumaua. m

 

 

 

68.8 .too 88.3 .83 68.923. 88.9 Loco 88.983 a 8513.
88.9 Sod 83.9 Rod 68980.0- $8.8 Sod $8.8 Boo 820
33 moaned
Emu a $9 a me a mg a mg n
oee $2. a 3. 25. i2. 21.: 225.9 I2. SEES 2-... SEE: E.

 

875 $2. 83. as... .52. :1. So one 9: oco
.rmz. .35: T... ucm mcosmbcmocoo moaned cmmzcmn 2mg Bate Emucflm vcm av mmumnv cozfloommd. .N 833.

77

References

10.

Noakes PS, Taylor P, Wilkinson 8, Prescott SL. The relationship between
persistent organic pollutants in maternal and neonatal tissues and immune
responses to allergens: A novel exploratory study. Chemosphere 2006:
63: 1304-1311.

Karmaus W, Brooks KR, Nebe T, Witten J, Obi-Osius N, Kruse H. Immune
function biomarkers in children exposed to lead and organochlorine
compounds: a cross-sectional study. Environ Health 2005: 4: 5.

Sunyer J, Torrent M, Garcia-Esteban R, et al. Early exposure to DDE,
breastfeeding and asthma at age six (in print). Clin Exp Allergy 2006.

Tanaka T, Kouda K, Kotani M, et al. Vegetarian diet ameliorates
symptoms of atopic dermatitis through reduction of the number of
peripheral eosinophils and of PGE2 synthesis by monocytes. J Physiol
Anthropol Appl Human Sci 2001: 20: 353-361.

Tanabe S, Tatsukawa R, Kawano M, Hidaka H. Global distribution and
atmospheric transport of chlorinated hydrocarbons: HCH (BHC) isomers
and DDT compounds in the Western Pacific, Eastern Indian and Antarctic
Oceans. Journal of Oceanography 1982: 38: 137-148.

Wills-Karp M. Interleukin-13 in asthma pathogenesis. lmmunol Rev 2004:
202: 175—190.

Romagnani S. Immunologic influences on allergy and the TH1/1' H2
balance. J Allergy Clin lmmunol 2004:1132 395-400.

Kondo N, Kobayashi Y, Shinoda S, et al. Reduced interferon gamma
production by antigen-stimulated cord blood mononuclear cells is a risk
factor of allergic disorders—6-year follow- up study. Clin Exp Allergy 1998:
28: 1340-1344.

Piccinni MP, Maggi E, Romagnani S. Role of hormone-controlled T-cell
cytokines in the maintenance of pregnancy. Biochem Soc Trans 2000: 28:
212-215.

Sonnenschein C, Soto AM. An updated review of environmental estrogen

and androgen mimics and antagonists. J Steroid Biochem Mol Biol 1998:
65: 143-150.

78

COMPONENT TWO

In utero SHD affects postnatal biomarkers of allergy

79

rfiha‘.‘--- -' 7‘

 

Maternal oral contraceptive use and immune markers in the offspring.

BACKGROUND: Maternal oral contraceptive (OC) use is Shown to be
associated with clinical manifestations of allergic diseases in offspring. The aim
of this study was to assess the, yet to be reported, association between OC use
and humoral immune markers in offspring. We hypothesized that maternal 00
use increases humoral immune markers of allergy in offspring.

METHODS: A cross-sectional investigation included 334 mother child (aged 7—
10 years) pairs from Hesse, Germany. In 1995, detailed self-administered
questionnaires were used to collect information on maternal OC use and atopy,
passive smoking, breastfeeding, offspring’s gender and number of infections in
the previous 12 months. lmmunoglobulins (lgs) along with white blood cell,
eosinophils, and basophilic surface lgE were also quantified. The data was
analyzed using linear regression, while controlling for confounders. Since allergic
reactions are different for boys and girts, we stratified by sex.

RESULTS: In children ages 7 to 10, female offspring of mothers who used OC
had Significantly lower (p<0.05): IgA (123.43 mg/dL vs. 150.52 mgldL), and lgE
(22.96 kU/L vs. 50.83 kU/L) levels as well as basophilic surface lgE counts (783
vs. 842), compared to those of mothers who did not use 00. For male offspring,
statistical Significance was only seen in an increased number of basophilic

surface lgE (911 vs. 876).

80

 

 

CONCLUSIONS: This study suggests that maternal 00 use may result in sex-
related differences of the immunomodulating effects in offspring: a finding that

could play a role in the etiology of allergic diseases.

81

Allergic/atopic diseases are among the most vexing disorders of childhood. The

. . . . . 1,2
IncreaSIng prevalence and severity of asthma and other allergic diseases

render them of enormous personal and public health concern 3. Early efforts to

elucidate the etiology of these conditions focused on factors such as genetic

predisposition, air pollution and tobacco smoke.

The role of synthetic hormones (estrogen and progesterone) as risk factors for

allergic diseases has gained increased traction 46. Specifically, the importance

of prenatal (in utero) exposure to these hormones has been presented 1’7.

Xu and co-workers assessed the effect of maternal age at menarche on the
occurrence of atopy among offspring. In a follow-up study of 5188 subjects, they
collected data during pregnancy and a follow up examination was completed at
31 years of age. Atopy was determined by skin prick tests while maternal age at
menarche was obtained from perinatal data. The authors reported that the

prevalence of atopy was lower in children whose mothers reached menarche at a

later age 7: a finding that may be explained by the fact that early age of

menarche is associated with higher levels of estrogen in adult women 8'10.

Further support for the possible prenatal endocrine disruption-allergic diseases
idea comes from the work of Frye et al. They reported that maternal oral

contraceptive (00) use before birth was associated with a higher risk of atopic

82

diseases in the offspring compared with children of mothers who had never taken

00 4. Moreover, in a recent study, we demonstrated significant associations for

asthma or wheezing, and cough between 11-12 year old offspring of 00 users

5 . . . . .
and non-users . These studies Show aSSOCIatIons between OC use and clinical

manifestations of allergic diseases in offspring; however, a relationship between

OC use and humoral immune markers in offspring is yet to be presented.

Using data from a cross-sectional investigation conducted in Hesse, Germany,
we assess the relationship between maternal OC use and immune markers in
offspring. We hypothesized that maternal 00 use Significantly increases

humoral allergic markers in offspring.

Methods

Subjects and materials

In 1995, parents of 1091 second-grade school children in 18 townships of Hesse,
Germany were invited to participate in the study. Children were allowed to
participate only if passive smoking in the private household did not exceeded 10
cigarettes per day during the previous 12 months. The study was approved by
the Data Protection Agency of the State of Hamburg, Germany, the Ministry of
Cultural Affairs of Hesse, Germany, and the participating local school

committees.

83

Questionnaires

Four self-administered parental questionnaires were used in the survey: one
regarding the living condition and nutrition of the family, one for each parent, and
one regarding information on the child. Maternal oral contraceptive use was
determined in an interview on reproductive history. The question posed was “In
the 12 months before conception of the index child, did you use oral
contraceptives at any time (yes\no)?” Duration of breastfeeding was recorded in
weeks of total and in weeks of exclusive nursing. Environmental tobacco smoke
(ETS) was graded as smoking in the child’s home in the previous 12 months (no
cigarettes, 1-20 cigarettes, 11-20 cigarettes, 21-30 cigarettes, more than 30
cigarettes per day). We recorded age, gender, and the number of infections,
defined as cold, coughing, and sore throat with or without fever in the last 12

months (none, less than 5 infections, 5-10 infections, more than 10 infections).

Immunologic assays

For phlebotomy, we used the 'Vacutainer System' (Becton, Dickinson 8.
Company, San Jose, California). Approximately 25 mL were drawn and
separated into different aliquots. lmmunoglobulin (lg) E in serum was quantified
at the Medical, Alimentary and Veterinary Institute for Research Middle Hesse,
Division of Human Medicine, Dillenburg, Germany, using a florescence-
immunoassay (CAP, Pharmacia, Uppsala, Sweden). We also measured IgA, G,
and M by laser immunonephelometry (Dade Behring, Liederbach, Germany). The

results for IgA, G and M were provided in mg/dL and for lgE in kU/L serum.

84

For leukocyte subsets, we collected 8 mL of blood in tubes containing EDTA and
mixed them to prevent clotting. This aliquot was transported to the Central
Laboratory of the University Clinic of Mannheim and analyzed on the same day.
We used 200 pL of blood for the automated differential (laser-based hematology
analyzer CD3500, Abbott Diagnostics, Santa Clara, California), and 100 p.L for
each of the nine three-color test tubes analyzed by flow cytometry (FACScan,
Becton, Dickinson, 8. Company, San José, California, equipped with a 488 nm

air-cooled argon ion laser).

Eosinophils were determined according to their specific depolarisation
characteristics and their eosinophilic granula content by the intensity of light
scatter by flow cytometry. lgE on basophils were identified by their high density
on the basophil surface using immunofluorescence with a Phycoerythrin labeled

anti-lgE antibody.

Statistical methods

Since data for leukocytes (VVBC) and their subsets (eosinophils),
immunoglobulins were not normally distributed, the geometric mean, 5-, 95-
percentiles are provided. In order to obtain a multivariate normal distribution, we
log-transformed the number of cells and the immunoglobulins before testing

associations with possible predictors.

All statistical analyses were performed using SAS software versions 9.1 11. We

used multiple linear regression models (PROC GLM) stratified by sex with

85

immune markers of allergic diseases as dependent variables. The potential
confounders; maternal atopy (positive if mother reported a history of asthma,
Eczema, or hay fever), age of offspring, ETS, number of infections during the last
12 months, and breastfeeding were controlled for in our models. Age of the child
was divided into three groups: 7, 8 and 9-10 years of age. Our main exposure of
interest, OC use in the 12 months before conception of index child, was coded
‘yeS/no’. Information on passive smoking (ETS) in the child’s home in the
previous 12 months was divided into three categories (no cigarettes, 1-20
cigarettes (a pack), 20 cigarettes per day and more). For the number of infections
we considered three categories (none, 1-5, more than 5). Duration of
breastfeeding was considered as either: 0 weeks, 1-5 weeks or more than 5

weeks.

From the results of the regression analyses, we calculated adjusted geometric
means of immune markers for the categories within each covariate. T-tests were
used to compare the statistical effect of maternal OC use with no maternal 00

use (reference).

Results

Of 1091 subjects invited, 671 (61.5%) agreed to participate. Of these, only
‘natural' mother-child pairs (n=663) were selected for this paper. Phlebotomy was
performed on 350 of the 663 children; 340 of whom had immunoglobulins
quantified. Complete information (questionnaires and immune markers) was

available for 334 children.

86

A comparison of potential confounders’ characteristics between the sub-cohort of
‘natural’ mother-child pairs and that with complete information is presented in
Table 1. Approximately 40 % of mothers who gave birth to boys used OC before
pregnancy compared to about a third of those who gave birth to girls. Of the
maternal atopy ‘family’ of variables, hay fever was the most prevalent at 15%.
Ninety Six percent of the children were between 7 and 8 years of age and
approximately 74 % experienced 1 to 5 infections during the last 12 months
(Table 1). Due to the eligibility criterion of passive smoking of less than 10
cigarettes in the child’s home for being included in blood sampling, the
prevalence of passive smoking was also lower in the group with phlebotomy than
in the total group (Table 1). Nevertheless, the fact that parents were separated or
divorced and Shared cohabitation for their child, resulted in a re-assessment of
the passive smoking status after phlebotomy. Eligibility was determined on the
information provided by one parent (mother or father) for their household. In the
case of separate dwellings, we re-assessed the exposure by taking the average
number of cigarettes smoked in both homes. As a consequence, 24 (7.5%)
children who were exposed to more than 10 cigarettes per day at home had a

phlebotomy and were included in the analyses.

Tables 2 reports that for girls whose mothers used OC before pregnancy, their
level of lgE was half that of girls whose mothers did not used OC before
pregnancy. The level of IgA was also decreased in girls of 00 users compared
to girls of none-OC users. This decreased level however was less dramatic than

that reported for lgE. lgG, 19M and lgE showed an increase with increasing age

87

in boys while only lgM showed this pattern in girls. lgE showed an increase with

age in girls but not in a dose dependent fashion.

Regarding lgE count on basophils; a higher number was evident for boys of
mother who use OC compared to boys of mother who were none-OC users. The
opposite was true for girls: higher in girls of none-OC-users compared to OC '
users. In addition, lgE count on basophils was found to increase with increasing

age and, conversely, decrease with increasing number of infections (Table 3).

Results from adjusted regression models showed significantly decreased IgA
(p=0.04; 123.43 mg/dL vs. 150.52 mg/dL) and lgE (p=0.04; 22.96 kU/L vs. 50.83
kU/L ) in girls ofimothers who used 00 before pregnancy (Table 5). In boys of
OC users, lgE count on basophils was significantly increased (p=0.04; 911 vs.
876); concomitantly, they were significantly decreased in girls (p<0.01; 783 vs.

842; Table 5).

Discussion

The data for children 7-10 years of age suggest, for the first time, sex-related
differences in the association between maternal OC use before pregnancy and
serum immune markers in offspring. Decreased serum IgA, and lgE levels as
well as basophilic surface lgE counts were observed in female offspring of OC
users compared to their female counterpart. In addition, increased basophilic
surface lgE counts was present for boys if their mother used 00 before

pregnancy.

88

The cross-sectional design of the study is a limitation as it brings into focus the
possibility of recall bias. However, mothers did not know the result from their
child’s serum immunologic analyses therefore the possibility of recall bias in this

instance was minimized. Furthermore there is no major public interest about DC

. . 4
use and atopic diseases .

Contrary to our a priori hypothesis, the data did not reveal an overall significant
association between maternal OC use and humoral immune markers in offspring.
Similarly, Xu and co-workers failed to find significant associations between

maternal sex hormone concentrations (in serum) during early pregnancy and

onset of allergic disease in early childhood 12. Furthermore, a study by Frye et al.

found no significant difference for atopic manifestations between offspring of

mothers who used 00 before and those of mothers who used 00 after

pregnancy 4. Nonetheless, our significant sex-related findings are intriguing and

consistent with the established hypothesis that there are sex-related differences

. . 13-16 . . .
in immune response ,Suggesting the Involvement of sex sterOId hormones

1647

OCp have two major components: progesterone, which is known to stimulate the

development of TH2 (allergic type) cells 18, and estrogen that is suppose to

increase the production of TH1 (non-allergic type) cells 19’20. To date, there is no

clear explanation of how sex steroid hormones affect the development and

89

function of the immune system. Nonetheless, it is know that some cells have

estrogen and progesterone receptors; creating binding sites for these hormones

. . . 14,15
and ultimately altering Immune responses .
Regarding IgA, Sell and Arici suggest that estrogen may increase or decrease

IgA levels in an organ dependent manner 15. We found that maternal OC use

may decrease serum IgA levels in girls. IgA was also decreased in boys of 00

users though not statistically significant.

For lgE, our finding of decreased levels in girls of mothers who used 00

suggests that hormones may be influencing lgE levels. Since lgE iS a strong

determinant of asthma 34, this finding supports the existing hypothesis that

hormonal factors may be involved in the pathogenesis of asthma.

Concerning basophil-bound lgE, studies have Shown increased counts in atopic

children 29'30 and deduced that they can serve as an indicator of allergic

sensitization 31. Our finding of decrease counts in girls of mothers who used 0C

is in concordance with our findings for lgE. Similar explanation may apply. On
the other hand, boys of OC users had an increased number of basophil-bound
lgE. An increase was also seen for lgE in boys of OC users, however this was
not statistically significant. These findings may be due to the estrogen/androgen

ratio in boys, which is known to modulate the activities of all cells involve in an

90

. 32 . . . .
Immune response . The pattern of androgenic and estrogenic stimulation

during early development (in utero) 33 may also be of importance. Further studies

are needed to examine these hypotheses.

In conclusion, our study supports the notion that there are sex-related differences
in the susceptibility to allergic diseases. Furthermore, the study adds evidence to
the idea that maternal OC (oral contraceptive) use may have immunomodulatory

effects in offspring.

There are yet unanswered questions. Studies are needed to elucidate the
mechanisms by which sex steroid hormones alter immune function. In addition,
clarity is needed on how the levels of sex steroids correlate with disease severity.
An idyllic setting to begin to answer these questions is that of a follow-up study
where women are enrolled early in pregnancy and hormone levels determined
from samples taken at different time-points (at least 3) during pregnancy. Cord
immune marker of allergy should then be quantified and the association between
hormone levels and these markers assessed. In addition to testing for an overall
effect, analyses should be stratified by sex to get at possible sex-related

differences.

9]

References

10.

11.

Beasley R, Crane J, Lai CK, Pearce N: Prevalence and etiology of
asthma. The Journal of Allergy and Clinical Immunology 2000;105:8466-
472.

National institutes of Health, LaBl. NH: GLOBAL STRATEGY FOR
ASTHMA MANAGEMENT AND PREVENTION. In, 2005.

Guilbert T, Krawiec M: Natural history of asthma. Pediatric Clinics of North
America 2003;50:523-538.

Frye C, Mueller JE, Niedermeier K, stt M, Heinrich J: Maternal oral
contraceptive use and atopic diseases in the offspring. Allergy
2003;58:229-232.

Brooks K, Samms-Vaughan M, Karmaus W: Are oral contraceptive use
and pregnancy complications risk factors for atopic disorders among
offspring? Pediatric Allergy and Immunology 2004;15:487-496.

stt M, Dold S: Is asthma an endocrine disease? Pediatric Allergy and
Immunology 1997;8z200—204.

Xu B, Jarvelin MR, Hartikainen AL, Pekkanen J: Maternal age at
menarche and atopy among offspring at the age of 31 years. Thorax
2000;55:691-693.

Vihko R, Apter D: Endogenous steroids in the pathophysiology of breast
cancer. Critical Reviews in Oncology/Hematology 1989;921-16.

Apter D, Reinila M, Vihko R: Some endocrine characteristics of early
menarche, a risk factor for breast cancer, are preserved into adulthood.
International Journal of Cancer. Journal International Du Cancer
1989;44:783-787.

Moore JW, Key TJ, Wang DY, Bulbrook RD, Hayward JL, Takatani 0:
Blood concentrations of estradiol and sex hormone -binding globulin in
relation to age at menarche in premenopausal British and Japanese
women. Breast Cancer Research and Treatment 1991;18 Suppl 1:S47-50.

SAS Institute: Statistical Analysis System, Version 8. In Institute S
(ed):Gary, NC, 2002-2003.

92

12.

13.

14.

15.

16.

17.

18.

19.

20.

21.

22.

23.

Xu B, Pekkanen J, Husman T, Keski-Nisula L, Koskela P: Maternal sex
steroid hormones in early pregnancy and asthma among offspring: a case-
control study. The Journal of Allergy and Clinical Immunology
2003;112:1101-1104.

Inman RD: Immunologic sex differences and the female predominance in
systemic lupus erythematosus. Arthritis Rheum 1978;21:849-852.

Bouman A, Heineman MJ, Faas MM: Sex steroid hormones and the
immune response in humans. Hum Reprod Update 2005;11:411-423.

Seli E, Arici A: Sex steroids and the immune system. Immunology and
Allergy Clinics of North America 2002;22:407-433.

Shames RS: Gender differences in the development and function of the
immune system. Journal of Adolescent Health 2002;30:59-70.

Whitacre CC, Reingold SC, O_Looney PA: A gender gap in autoimmunity.
Science 1999;283:1277-1278.

Piccinni MP, Giudizi MG, Biagiotti R, Beloni L, Giannarini L, Sampognaro
S, Parronchi P, Manetti R, Annunziato F, Livi C: Progesterone favors the
development of human T helper cells producing Th2-type cytokines and
promotes both lL-4 production and membrane CD30 expression in
established Th1 cell clones. J Immunol 1995;155:128-133.

Fox HS, Bond BL, Parslow TG: Estrogen regulates the IFN-gamma
promoter. J lmmunol 1991;146:4362-4367.

Whitacre CC: Sex differences in autoimmune disease. Nature Immunology
2001;2:777-780.

Klein KO, Baron J, Colli MJ, McDonnell DP, Cutler GB, Jr.: Estrogen
levels in childhood determined by an ultrasensitive recombinant cell
bioassay. J Clin Invest 1994;94:2475-2480. '

Garnett SP, Hogler W, Blades B, Baur LA, Peat J, Lee J, Cowell CT:
Relation between hormones and body composition, including bone, in
prepubertal children. American Journal of Clinical Nutrition 2004;80:966-
972.

Sears MR, Burrows B, Flannery EM, Herbison GP, Holdaway MD: Atopy
in childhood. |. Gender and allergen related risks for development of hay
fever and asthma [see comments]. Clinical and Experimental Allergy
1993;23:941-948.

93

24.

25.

26.

27.

28.

29.

30.

31.

32.

33.

34.

Maziak W, Behrens T, Brasky TM, Duhme H, Rzehak P, Weiland SK, Keil
U: Are asthma and allergies in children and adolescents increasing?
Results from ISAAC phase I and phase III surveys in Munster, Germany.
Allergy 2003;58:572-579.

Obendorf M, Patchev VK: Interactions of sex steroids with mechanisms of
inflammation. Curr Drug Targets Inflamm Allergy 2004;3z425—433.

Akdis M, Blaser K, Akdis CA: T regulatory cells in allergy: novel concepts
in the pathogenesis, prevention, and treatment of allergic diseases. J
Allergy Clin lmmunol 2005;116:961-968; quiz 969.

Stock P, Akbari O, DeKruyff RH, Umetsu DT: Respiratory tolerance is
inhibited by the administration of corticosteroids. Journal of Immunology
2005;175:7380-7387.

Jansson L, Holmdahl R: Estrogen-mediated immunosuppression in
autoimmune diseases. Inflamm Res 1998;47:290-301.

Stallman PJ, Aalberse RC: Quantitation of basophil-bound lgE in atopic
and nonatopic subjects. Int Arch Allergy Appl lmmunol 1977;54:114-120.

Wada T, Toma T, Shimura 8, Kudo M, Kasahara Y, Koizumi S, Ra C, Seki
H, Yachie A: Age-dependent increase of lgE-binding and Fcepsiloan
expression on circulating basophils in children. Pediatr Res 1999;46:603-
607.

Lander F, Meyer HW, Norn S: Serum lgE specific to indoor moulds,
measured by basophil histamine release, is associated with building-
related symptoms in damp buildings. Inflamm Res 2001;50:227-231.

Cutolo M, Seriolo B, Villaggio B, Pizzorni C, Craviotto C, Sulli A:
Androgens and estrogens modulate the immune and inflammatory
responses in rheumatoid arthritis. Ann N Y Acad Sci 2002;966:131-142.

Martin JT: Sexual dimorphism in immune function: the role of prenatal
exposure to androgens and estrogens. Eur J Pharmacol 2000;405:251-
261.

Jaakkola MS, Ieromnimon A, Jaakkola JJ.K: Are atopy and specific lgE to

mites and molds important for adult asthma? J Allergy Clin lmmunol 2006;
117:642-648

94

 

 

 

 

0.0 E 28> 2-0
00¢ o.m~ 900> w
0.0.. 0.8 98> >
00:06-02
08 mg 00> >82
0. E 0.0 00> 05080
.0>0..
02 0.2 9% >0:
3.. 0.0 we, 0552
.0505).
0N we 00.00:).
0.8 in 00> 20
ed m.v 9.00.5.
0.? 0.00 00> >00
5000090 0.0.00 00:
02000090000 .90 80.0.05.
:0 «.8 >00
x0m
as as
€89. somuzv
900

.90me 002:8. 0:0 02000090000
.90 .0599: 5.2, 030505

9.00 2.0000508
.930: .0 030.0

 

t0...00 0.00.0..00 05 .0 8.6.0.0905 0200:0000 ... 0.00...

95

 

0.0 0.0 9.00.5.

0.00 0.00 0 :05 0.0.2
0.0V 0.0.. 0-?
...: 0.0. 0
00.003.
0500000090 .900 00 :00930
0.0V 00? 0A
0.: 0.: 0 -F
0.0 0.0 0:02

05:2: N. .00.
05 0:000 0:00002. .0 09:32

me 0.. 0500.2
0.0 for 00 :05 0.0.2
500 ......0 om-..
N00 0. .0 0:02
$00 .00 0000006.

05:08 N? 000. 05 0:030 0600
0.2.00 05 :_ 0:0.0E0 0>.000n.

 

 

o\o o\o
.vmmuc. .000nz.
900
0:00.008 0::EE. 0:0 0>00009E00 00.00 0.....000508
.90 .0599: 5.? 020.0000 .930: n.0 0:000

 

:0000 0:90:00 000 n.0 000000009000 0200:0000 0000:3000. ., 0.00 ..

96

 

...000 I 0.0. ..00
...000 I ..0. 0.00
0.007000. 0.00

.0.0.0 I ..0. 0.00
.0000 I 0.0. 0.0
.0000 I 0... ..00

.000. I 0.0. 0.00
.0000 I 0... 0.00
.0.0.0 I 0.0. 0.00

0000.00. ...0
0000-00. ..00
...0000... 0.0

.000 I 00. 00.
.000 I 00. 0....
.000 I 00. ...0.

.000 I .0. ..00.
.000 I 00. 0.00.
.0000I00. 0.00.

.000 I 00. 0.00.
.0.0 I 00. 0.0.
..00 I 00. ..00.

.00000. 0.00.
.00000. 0.00.
.00000. 0. .0.

. .00 I 00. ..00.
.000 I 00. 0.00.
.000 I 00. 0.00.

.000 I 0.. 0.00.
.000 I 00. 0.00.
.000 I 00. 0.00.

.00. I .0. 00.
.000 I 00. 0.00.
.00 I 00. 000.

000-00. 0.00.

.000. I 000. 0.0.. .0.0. 0 :9. 0.0.2
.000. I 000. 0.00.. .00. 0-.

.000. I 000. 0.00.. .00. 0
00.003. 0500000090 .90. .0 :0..90n.

.000. I 000. 0.0... .00. 0A
.000. I 000. 0.00.. .000. 0 -.
.0000. I .00. ..000. .00. 0.52

05:0...
0. .00. 0... 0:..00 0:0..00.:. .0 00:52

.000. I 000. 0.000. .0. 00 :0... 0.0.2
.000. I 00.. 0.00.. .00. 00-.

.000. I 00.. 0. . . .. .000. 0002
900 .00 0000.006. 00.:0E N. .00.

0... 0:..00 0E0: 0.0..00 0... :. 0:...0E0 020000

800.000. 0.00.. .0.. 98> 0.0

 

 

.00... M9

30.00.. .29

.0000. 0.00. 800.000. .. .0.. .00.. 38> 0
E000. 0.00. 600.00.. 0.00.. .00.. 28> 0
0000.0-00<
30.0.... <0_ 30.0.... 00_ 0.8800

 

.00.0..>00 >0 0:..000.00:0EE. .0. 000.05 00 .-0 0:0 :00... 0.000.000 00.0000 00:00 ”N 0.00...

97

 

.000 I .00. 0.00
.000 I 00. 00.
.0000 I 00. 00.

.000 I 000. 000 £00 I 000. 0.000
.000 I 000. 0. .00 .0. .00 I .00. 0.000
..00 I 000. 0.000 ..00 I 000. 000

.0000. I 00.0. 0.0.00
.0000. I 0000. 0.0.00
.0000. I 00.0. 0.0000

.0. 00 :0... 0.0.2
.00.. 00-.
.000.0002

900 .00 0000.000. 00.00... N. .00_

.000 I 000. .00
. .00 I 000. 0.000
.000 I 000. 0.000

.000 I 0.000. 0.000
.000 I 0.000. 0.000

.000 I 000. 0.000
.000 I 00 0.000

.
..00 I 00 .0000
.000 I 000. 0.000

m
m

.000 I 000. 0.000
.000 I 000. 0.000

.000 I 000. 0.0 .0
.000 I 000. 000
.000 I 000. 0.000

.000 I 000. 0.000
..00 I 000. 0.000

.000 I 000. 0.000
.000 I 000. .00

.000 I 000. 0.000
.000 I 000. ..000

.000 I 000. 0.000
.000 I 000. 0.000

.000. I 00. 0.0.0
.000 I .0. 0.0.
.00 I 00. 000.

.000 I 00. 0.00.
.000 I 00. 0.00.

.000 I .0. ...0.
.000 I 00. ..00.

.000 I 00. 0. .00
.000 I 00. 0.000

.000 I .0. 0.00.
.000 I 00. 0.000

00. 00000 00.00 00:00 00. 0. 00200.0 020000

.0000. I 0000. 0.. .00
.0000. I 0000. 0.00.0
.0000. I 0000. ..0000

.0000. I 0000. 0.0000
.000.. I 00.0. 0.0000

.000.. I 0000. 0.0000
.0000. I 0000 0.0000

.
.00.0. I 0000. 0.0.00
.0000. I 0000. 0.0000

.0.. 98> 0.0
.00.. 98> 0
.00.. 98> 0

000 02.0000..000 _0.0 050.02

.000.. I 0000. 0.0000
.0000. I 0000. 0.0000

 

0:000000
00 .0000 m0_

0300.0 0___0000_00m

3.0.x .008
:00 __..000_000

3.0.x 0:00 0020 0.22.

0000.0-00<
..00. oz
..0. 00>
>00.0 .00.0.02
.00. oz
.00. 00> _._0
.00.. 02 >00
..0. 00>
.00.. _._0
.00..>00
x0w
0.000>00

 

.00.00>00 >0 m9 000000 05000000
000 02.000.00.000 050005000 .__00 0020 0.0.3 .0. 000_0> 00 .-m 000 0000. 00.0.0000 000000 .00000 .0 0.00-.-

98

 

.000 I 000. 0.000 .000 I 000. 0.000 .000 I 00. 0. .0. 400.0. I 0000. 0.0000 .0.0. 0 :0... 0.05.
.000 I 0.0. 0.000 .000 I 000. ..000 .000 I 00. 0.00. .0000. I 0000. ..0000 .00. 0-.
.000 I ..000. 000 ..00 I 000. 0.000 .000. I 00. 0.00. .0000. I 0000. 0.0000 .00. 0

.3003. 0500000005 _0.0. .0 0000.00

.000 I 000. 0.000 .000 I 000. 0.000 .000 I 00. 0. .00 .0000. I 0000. 0000 .00. 0A
.000 I0.000. 0.000 .000 I 000. 0.000 .000 I 00. ..00. .000.. I 0000. 0.0000 .000. 0 -.

.000 I 000. 0.000 .000 I 000. 000 .0.000 I 0.00. 0.00. .000.. I 00.0. 0.0000 .00. 0002
00000 00.000. N. .00_ 00. 05000.5 .0 000.02

 

0:000000

01.0.0... .008 30,0...
00 .0000 m0_

0_000.0 0___0000_00m =00 __0000_00m 0:00 0020 0:03 0.00050

 

.00.00>00 >0 m9 000000 05000000 000 02.000.00.000
050005000 .__00 0020 0.0.3 .0. 000_0> 00 .-m 000 000E 00.00.000 050000 .00000 ".0000_.000. 0 0.00...

00020.0. 00. 00 000 0>_.0000..000 _0.0 _00.0.00. 00 0. 00.000.00.00.-. 0 00 00000 00.90 ..

0500000005 _0.0. .0 0000.00 000 00.00.: .0.. .00_ 00. 00000

00080.5 .0 .00.000

00.00:. N. .00_ 00. 00000 00.00 00:00 00. 0_ 00200.0 020000 .000 00:00 $00.0 _00.0.0E .0. 00.00.00. 0

 

 

 

 

.m 00 mm. mm. .0. S. .00. 3.. 0z 0 00.00.04.
.00 m... .0. mm. .mm. 00. .mm. mm .. 00>
0m 0m 0m. 0.. 0m. 0m. 00.. 000. 02 000.0
v. 00 mm. 0.. 0.. 00. 00.. 000. 00>
000
0>_.0000..000 _0.0 00.0.02
2
005 0>0m 005 0>0m 005 0>0m 005 0>0m
00.50.00
.00... 09 .00.0.... 29 .00.0.... <9 30.00.. 00.

 

00__000_00000.0._ >0 000 0>_.0000..000 _0.0 00.0.0.2 "v 0.00...

100

0800.0. 00. 00 000 0>_.0000..000 _0.0 _00.0.0E 00 0. 00.000.00.00.-. 0 00 00000 00.0V0 .
020000000... _0.0. .0 0000.00 000 00.00.: N. .00_ 00. 00000

00260.0. .0 .0060:

00.00:. N. .00_ 00. 00000 00.00 00:00 00. 0_ 02200.0 020000 .000 00:00 .>00.0 _00.0.00. .0. 00.00.00. 0

 

 

 

 

 

N00 000 000 000 0N. 00N 0000 0000 02 a 00.00_0<
.00. .. .0 ..00 000 .N. .mN .0000 N. .0 00>
000 000 N00 .00 .0. NON 0N8 0.00 02 000.0
..00 000 .00 000 N0. 00N 0000 00N0 00>
000
0>00000..000 _0.0 00.0.00.
2
005 0>0m 005 0>0m 005 0.80 005 0>0m
2000000 0000.0 .100... .008 3.0... 00.00.00
00 .0000 09 2__0000_00m_ =00 __0000_0om 0:00 0020 0._0>>
00_I

000000 2:000000 000 02.000.00.000 050000000 .__00 0020 0002. >0 000 0>00000..000 _0.0 _00.0.0_>_ ”0 200...

101

COMPONENT THREE

In utero exposure to SHD affects allergic outcomes in infancy

102

Are oral contraceptive use and pregnancy complications risk factors for

atopic disorders among offspring?

Kevin Brooks 1, Maureen Samms-Vaughan 2, Wilfried Karmaus 1

1 Department of Epidemiology, School of Human Medicine, Michigan State

University, East Lansing, United States;

Department of Obstetrics, Gynecology and Child Health, University of the

West Indies, Mona campus, Jamaica.

103

Abstract

In utero programming of atopic manifestations has been suggested. We
investigated the association between oral contraceptive (OC) use before, and
complications during pregnancy (GDP) and asthma, along with other atopic
manifestations. The study is based on neonates from Kingston and St. Andrew, a
geographic sub-cohort from the Jamaican Perinatal Morbidity, Mortality Survey
conducted in 1986-1987. Information on 00 use and CDP was extracted from
maternal interviews and medical records. In a follow-up in 1997-1998, via
interviews with mothers, trained nurses collected information on
asthma/wheezing, coughing, eczema, and hay fever. Data, specific to this paper,
from birth and 11 - 12 years of age was available for a total of 1040 of the 1720
members of the geographic sub-cohort. Using logistic regression, controlling for
confounders, we estimated adjusted odds ratio (aOR) and corresponding 95 %
confidence intervals (CI). For asthma or wheezing, and coughing, aOR for 00
use were 1.81 (95% Cl 1.25- 2.61), and 2.72 (95% CI 1.41-5.24), respectively.
CDP was only shown to be a significant risk factor for hay fever. Additionally, a
higher number of older siblings were protective for hay fever. The results suggest
that asthma in childhood may be programmed in utero.

Key words: asthma; eczema; hay fever; atopy; oral contraceptives; pregnancy

complications

104

The prevalence of allergies is increasing in many parts of the world, and asthma
has become the most common chronic disease of childhood. The etiology of

asthma and allergic disease remains poorly understood, despite considerable

research 1. Recent work has expanded to include the study of novel factors such

as in utero and perinatal exposures that may program initial susceptibility 2.
Some authors have reported that maternal estrogen levels and maternal health
complications during pregnancy are associated with asthma and other atopic

manifestations later in life 3'6.

The most appropriate design to test an association between perinatal conditions
and asthma in offspring is a longitudinal study commencing with mothers in the

antenatal period and following their offspring through childhood. The Jamaican

Perinatal Mortality and Morbidity Survey (JPMMS) 7, and the subsequent follow—

up studies offered a unique opportunity for testing these hypotheses.

The JPMMS, originally designed to ascertain causes of maternal and neonatal
mortality and morbidity, obtained detailed information about maternal lifestyle
before and during pregnancy, the labor and delivery process. The subsequent
follow-up studies of the children were designed to identify factors promoting and

preventing optimum childhood outcomes, including child health.

105

Women who participated in the JPMMS in 1986 were contacted at six weeks
postpartum and again eleven to twelve years later. We investigated whether in
utero and perinatal factors contributed to the development of asthma and other
atopic manifestations in children 11 to 12 years of age. Specifically, we tested the
association between two risk factors:

(1) Maternal use of oral contraceptives (OC) before pregnancy

(2) Maternal complications during pregnancy (GDP)

and the prevalence of four atopic manifestations:

(a) Asthma/wheezing,

(b) frequent nighttime/early morning cough,

(0) eczema, and

(d) hay fever/sinus problems/other allergy

Methods

Study population
The Jamaican Perinatal Morbidity and Mortality Survey included all pregnant

women who had a live or stillbirth during the two-month period from September
1St to October 315‘, 1986. Women were interviewed and their babies examined,
usually within the first 48 hours after delivery. Data were gathered during this
phase using the main questionnaire (MQ). A total of 10392 babies or 94 % of the

births in the two-month period were identified and included in the study (main

cohort study) 8. The first follow-up study of the children occurred at six weeks of

age when 8,800 (84.7 %) mother-child pair from the main cohort were

106

interviewed. Information on breastfeeding and the infant’s health was obtained
using the first follow-up questionnaire (FF UQ). These 8800 mother-child pairs
were used as the baseline cohort (entire island) for the analyses.

Follow-up

A geographic sub-sample of 1,720 eleven to twelve-year-old children,
representing those residing in Kingston or St. Andrew, the two most urban of
Jamaica’s fourteen parishes, was selected for a second follow-up study. In a
cross-sectional survey of these children, data on health outcomes were collected
from participating parents or guardians (n=1,163), mostly the mothers, and
recorded using a second follow-up questionnaire (SFUQ). We linked the three
datasets using a unique identification number assigned at birth and dates of birth
for both mother and child.

Questionnaires and examinations

During the JPMMS, trained nurse midwives administered face-to-face interviews
with mothers and conducted a brief examination of the babies shortly after birth.
Babies were also weighed and measured. Hospital delivery notes and maternal
health records were consulted to verify and supplement the data obtained by
interview. The MO provided data on past obstetric history and complications

during pregnancy (CDP), labor and delivery.

Data on the current status of the baby (alive/healthy, alive/ill, dead or don’t know)

and what the baby was currently being fed (breast, breast and formula or formula

only) was extracted from the FFUQ. Information on the presence of asthma or

107

wheezing, coughs, eczema and hay fever/ sinus problem/ other allergy in the
children at 11 to 12 years of age was collected from the general health section of
the SFUQ. Trained nurse interviewers administered the questionnaire to either
the parents or guardians.
Definition of variables
Maternal age at birth was calculated from the mother’s date of birth and child’s
date of delivery. Based on questions from the MD, a yes, no, or don’t know
response was used to define the following:
- Did you smoke tobacco regularly at any time during this pregnancy?
- Have you ever used contraception? If yes, which methods have you used?
(Contraceptive pill was among a list of contraception.)
- Were you admitted to hospital or rural maternity center during this pregnancy
but before going into labor? (Used to determine maternal hospital admissions.)
- Are any of the following services available within 1-2 miles walking distance of
where you live?’ (Health center was among a list of services. - Used to
determine access to health care.)
Mother’s education level was grouped into three categories for the purpose of
this study (less than secondary, secondary, and more than secondary
educafion)

Maternal CDP was defined as a positive response to any of the following
quesfions:
1) Did you have vaginal bleeding in the first 28 weeks of pregnancy?

2) Did you have vaginal bleeding after 28 weeks (7 months)?

108

3)

4)

5)

5)

7)

After starting antenatal care, were you referred for treatment during
pregnancy for any reason?

Were you referred during pregnancy for any reason?

Was the mother diagnosed as having hypertension, pre-eclampsia or
eclampsia during this pregnancy?

Did the mother have any of the following during this pregnancy; epilepsy,
eclamptic fits or heart disease and lastly,

Did the mother have any other complications, disorder or serious illness

during this pregnancy?

Responses to the first five questions above were obtained by interview, while the

remainder was obtained from health records.

Maternal infection during pregnancy was defined as a positive answer to any of

the following question based on information from medical records:

1)

2)

Did the mother have a vaginal discharge or infection during this
pregnancy?

Did the mother have urinary tract infection, tuberculosis, rubella (German
measles), gonorrhea, syphilis, a positive Venereal Disease Research

Laboratory test or genital sores or blisters during pregnancy?

The following variables were used to define maternal health complications during

labor or delivery (positive if indicated in the medical records):

1)
2)

3)

Did the mother have eclamptic fits?
Did she hemorrhage?

Was labor obstructed?

109

4) Did the uterus rupture?

5) Was mother transferred during labor?

6) Did the cord prolapse?

7) Was there meconium in the liquor?

8) Were there any other complications?

The number of preceding live births determined birth order of the index child. The
child’s birth weight was also recorded during this period (MO) and grouped into
three categories for the purpose of our analyses (<2.5kg, 2.5-4.0 kg and >40
kg).

The variable child's illness or health problems in first week of life was defined as
a positive response to any of the following questions from the FF U0:

1) Was the baby referred to a medical officer?

2) Was the baby admitted to hospital or special care baby unit?

3) Were any abnormal symptoms noted in the baby?

Outcomes of interest at 11 to 12 years of age were ascertained from the SF U0.
The original question to the parents was, ‘Has your child ever had...’ This
question was followed by a list of conditions; including asthma or wheezing,
frequent nighttime or early morning cough, eczema, and hay fever sinus problem

or other allergy.

Statistical analysis

We tabulated the prevalence of the four atopic manifestations for the different
ante- and perinatal exposures. Differences in proportions between the different

sub-samples were assessed. Adjusted odds ratios (aOR) and their 95 %

110

confidence intervals (95% Cl) were estimated by logistic regression analysis.
Statistically, we controlled for maternal infection, hospital admissions and
smoking during pregnancy, as well as maternal complications during
labor/delivery, induced labor, anesthetic/analgesic during labor, gender, birth
weight, breastfeeding, child’s illness in the first week of life, access to health care
and mother’s level of education. These potential confounders were included

based on previous publications and after satisfactorily, checking for collinearity

between all predictors (risk factors and all confounders) 9. We estimated aOR for

all four outcome variables independently.

Further steps were taken to determine whether having both, asthma or wheezing
and frequent nighttime or early morning cough, was associated with 00 use and

complications during pregnancy. All calculations were carried out using SAS for

Microsoft Windows (release 8.2) 10.

Results

From the three data sets, we successfully linked 1040 (60.4 %) of the 1720
available mother-child pairs from the Kingston and St. Andrew sub-cohort.
Compared characteristics between exposure variables from the baseline cohort
(entire island) and the linked data set (urban population) are presented in Table
1. Approximately 50 % of mothers used OC, and approximately, 50 % had CDP.
In the baseline sample, maternal education was lower than in our sub-sample

(Table 1: less than secondary level of education - baseline: 35.6%, subsample:

111

25.4%). The proportion of children born with low birth weight (<2.5 kg) was higher
in the more urban setting of Kingston and St. Andrew (12.2 %), where the second
follow-up cohort originated. A significant rural-urban difference was also indicated

by the difference in breastfeeding categories.

When comparing the follow-up group for whom health information was available,
(n=1163 of 1720) with the linked sample (n=1040) there was no substantial
difference (asthma or wheezing: 17.0 % vs. 16.8 %, frequent nighttime or early
morning cough: 4.8 % vs. 5.3 %, eczema: 6.5 % vs. 6.3 %, hay fever or sinus

problem: 19.6 % vs. 20.1 %),

Table 2 presents the extent to which atopic manifestations occur with each other.
The proportion of atopic disorders in relation to in utero risk factors is presented
in Table 3. Children of mothers, who used, compared to those who did not use
OC before pregnancy, had a higher prevalence of asthma (20 °/o vs. 13.4 %),
frequent nighttime or early morning cough (7.4 % vs. 3.6 %) and eczema (7.4 %
vs. 5.5 %), Asthma and wheezing, as well as hay fever or sinus problems were
more common in children whose mothers had complications during pregnancy

(CDP) than in those who did not.

Of the perinatal risk factors, induced labor was related to a higher prevalence of

eczema (11.4 % vs. 6 %, Table 4). Children of mothers who had complications

during labor or delivery had a greater proportion of frequent nighttime or early

112

morning cough (8.1 °/o vs. 4.8 %), In children who had illness or health problems
in their first week of life, eczema (10.5 % vs. 5.3 %) and hay fever sinus problem
or some other allergy (27.4 °/o vs. 18.3 %) were more frequently reported at 11-

12 years of age.

Information on breastfeeding was available for only 763 children (Table 4). For
this reason, breastfeeding was not considered when testing the hypotheses with
a full model. A possible confounding effect of breastfeeding was investigated in

additional analyses.

Adjusted odds ratios for asthma or wheeze, and frequent nighttime or early
morning cough are reported in Table 5 while those for eczema and hay fever,
sinus problems or other allergy are reported in Table 6. Maternal 00 use before
pregnancy was significantly associated with both asthma (aOR = 1.81) and
frequent nighttime or early morning cough (aOR = 2.72) in the offSpring. The
association between maternal OC use before pregnancy and asthma was
stronger for girls (aOR = 2.02, 95% Cl 1.21 -3.37) than for boys (aOR = 1.60,
95% CI 0.92- 2.76). Regarding eczema, hay fever, and sinus problem or other
allergy at age 11-12 years of age, the confidence limits of the aOR for maternal

OC use included the null-value (Table 6).

Additionally, it is well established that ‘All that wheezes is not asthma’ 11

therefore we conducted further analysis with a combined variable, defined as the

113

conditions asthma or wheezing and frequent nighttime or early morning cough
occurring concurrently. Of children whose mothers used CC, 5.1 % had asthma
or wheezing, and coughing compared to 1.9 % of children of mothers who did
not. This association was then assessed controlling for confounders in the full
model. When compared to the two separate effects, the association of the

combined outcome was stronger (aOR=3.32, 95% CI 1.33-8.32).

Complications during pregnancy gained statistical importance only for hay fever
or sinus problems (Table 6). When type of feeding was introduced into the
models, analyses were then based on 763 children. For CDP, this resulted in an
increased odds ratio for hay fever (aOR=1.72, 95% CI 1.13-2.63) (data not
shown). However, the significant odds ratio for oral contraceptive use did not

change.

A higher number of older siblings (birth order) showed a significant protective
effect for hay fever (Table 6). For asthma or wheezing and for coughing, similar

association was found though not statistically significant (Table 5).

Discussion

The findings support the hypothesis that children whose mothers used OC (oral
contraceptive) before pregnancy, are at a higher risk for developing asthma or

wheezing, and frequent nighttime or early morning cough at 11 to 12 year of age.
The data also supported the hypothesis that a maternal complication during

pregnancy is a risk factor for atopic manifestations in offspring.

114

The association between OC use and asthma or wheezing and coughing is in
concordance with a suggestion of stt et al. who, based on aggregative data on

OC sales and hospital discharge data, were the first to hypothesize that maternal

use of OC is a risk factor for asthma in the offspring 3. To the best of our

knowledge, the present study is the first to prospectively assess the association
between, maternal OC use and asthma or wheeze in offspring, using individual

data.

The most frequently used 00 in Jamaica during the study period was the Pearl
®, in which the active components were ethinyloestradiol (30 pg) and
Ievonorgestrel (150 pg), an estrogen and progesterone respectively. It has been

suggested that women who use OC have higher estrogens levels after

discontinuing their use 12, however, the evidence is not consistent 13.

The OC-atopy association may be explained by endocrine effects on T-helper 2
(Th2) cells. Th2 cells produce interleukin (IL)-4, lL-5, and lL-13, which promote

the production of IgE and eosinophil infiltration in the ainivays and ultimately

orchestrate the development of allergy and asthma 14-16. stt and co-workers

suggest that estrogen might enhance the activity of Th2 type cytokines 3,

predisposing the fetus to atopic disorders later in life. Furthermore, in vitro
studies suggest that progesterone promotes the preferential development of Th2-

type cells and to function as a potent inducer of the production of the Th2 type

115

cytokines 1749. Michel et al. reported that lgE levels in cord blood were

significantly increased among neonates whose mothers had taken progesterone

. 20
during pregnancy

We suggest two possible scenarios for the endocrine mechanism, which may
explain our OC-atopy association. Either maternal exposure to estrogen via OC
use may trigger an increase Th2 activity continuing into pregnancy (prolonged

Th2 effect), consequently predisposing the fetus to a higher risk of developing

atopy later in life, a Th2-mediated disorder 15’21’22. Alternately, women who used

OC continuously, discontinuing before becoming pregnant, may have created an
environment of higher estrogen or progesterone levels (prolonged endocrine

effect), resulting in an increased Th2 activity impacting the fetus transplacentally.

Evidence for the prolonged Th2 effect is supported by findings that women

exposed in utero to the estrogen diethylstilbestrol, were found to have alterations
in their T-cell-mediated immunity 23. In addition, children of mothers who used
OC in the 18 months prior to being pregnant were found to be more likely to have

eczema at age five 24. This finding was later corroborated by Xu et al. 25.

Evidence in support of an prolonged endocrine effect: Xu et al. reported that

atopy in adults was more common among those whose mothers experienced an

early age of menarche 4. A likely explanation for the atopy — age at menarche

116

association is that early age of menarche is associated with higher levels of

. 1 - -
estrogen In adult women 3’ 26 28. Hence, the fetus of women wrth an early age

at menarche may experience higher exposure to estrogens. This could then lead

to the programming of the immune system of the fetus affecting atopic status

later in life 29.

In a cross-sectional study, Frye et al. 30 provided data to support the prevailing

OC - atopy association. They however downplayed the significance of this
finding on the basis that similar results were found in the offspring of women who
used OC exclusively after they delivered. It is noteworthy to accentuate the fact
that Frye and co-workers used a cross-sectional design wherein OC use was
assessed retrospectively. This design will no doubt be plagued with biases.
These results may be an artifact of ‘reverse causation’ introduced by behavior
modification of the parents. An example of this is women in the ‘exclusive OC
use after birth’ group may have chosen to resume the use of OC after the index
child developed atopic eczema. The reason is that a child with eczema will
require more attention at nights. Therefore, a mother with a demanding child

may decide to delay having another by using OC.

There is a paucity of studies on the relationship between maternal complications
during pregnancy and the risk of having asthma or any atopic disorders. Nafstad

et al. in a population-based, four-year, cohort study involving 2531 children.

Asthma was assessed at age four 5. The authors reported that uterus-related

117

complication (antepartum hemorrhage, preterm contractions, insufficient
placenta, and restricted growth of the uterus) increased the risk of having asthma
(OR = 3.0, 95 % CI 1.8-5.4) and allergic rhinitis (OR = 2.9, 95 % CI 1.6-5.2).
Annesi-Maesano et al. also found maternal complications during pregnancy to be

a risk factor of asthma (OR = 2.01, 95 % CI 1.52-2.67) in the offspring using a

large British birth cohort (4065 natural children of 2583 mothers) 6. More

31,32

recently, two studies have supported these findings , one by using data on

Norwegian live births (1967-1993, n=1,548,429) 31. Our study found maternal

complications during pregnancy (CDP) to be a significant risk factor for hay fever

and sinus problems (aOR = 1.5, 95 % CI 1.05-2.16, p=0.03).

Further analyses for items included in the predictor complication during
pregnancy (CDP) indicated that only the variable ‘have there been any other
complications, disorder or serious illness during this pregnancy?’ conferred a
significantly increased risk for asthma in the child (p< 0.001), but none of the
specific items. This finding is in agreement with a recent report by Nafstad et al.

who also found that pregnancy complications in general might present risk
factors of stress for the developing fetus 31. It might be that CDP (complications
during pregnancy) is reflective of factors such as a threatened pregnancy, found

to be associated with frequent wheezing 33, which will alter the Th1 /Th2 balance

34'35. This hypothesis warrants further investigations with adequate perinatal and

breastfeeding information.

118

Our study adds to the body of evidence, which suggest that the number of

siblings, in particular older siblings, is important in the etiology of atopic disorders

28. We found number of older siblings to be protective for hay fever or sinus

problems (Table 6). Number of older siblings and oral contraceptive use were

statistically significantly correlated (rspe,,.rm,,.n = 0.3, p=0.0001, n=1020).

The lack of association between low birth weight and asthma is in contrast to

some previous studies 636-38 and in agreement with others 3942. If is worth

mentioning that the finding from our Afro-Caribbean cohort is similar to that of an

African-American cohort 40.

The strengths of the current investigation are in its prospective design, being
derived from a large population based cohort, with all the exposure information
collected during pregnancy, at birth and shortly after. This design presents a
clear time-order, as outcomes were not assessed until approximately 11 years
after. The impact of recall bias is also reduced. Additionally, controlling for a
broad range of confounders minimize the possibility of these findings occurring

purely by chance.

A limitation of this study is the follow-up proportion of 60.4 %. This is largely

explained by parental participation in the second follow-up study. Though

119

parental consent was obtained for the 1720 children evaluated using the child’s
completed reports, only 1163 parents provided interview and questionnaire data.
The significant differences found between the characteristics of some of the
variables from the baseline cohort when compared with those of the linked
dataset were expected. This, we infer, is due to the rural-urban difference

between these groups.

The difference between reported occurrence of CDP in the baseline cohort
(38.4%) and our sub-sample from Kingston/St. Andrew (49.2%) is a possible
source of selection bias. The baseline cohort has both rural and urban
populations, whereas the sub-sample if predominantly urban. Different
proportions of maternal complications in these samples can be explained by two
factors, namely maternal education level and access to health care. For instance,
of the women in our sub-sample who reported having complications during
pregnancy, 3.6% were characterize as having low level education (less than
secondary) and less access to health care compared to 8.9% of those with
similar characteristics from the baseline cohort. Recently, McCaw—Binns et al

reported that access to health care in Jamaica differs from rural to urban areas

43. Since these two factors (maternal education level and access to health care)
seem to affect the reported proportion of maternal complications, and as the
prevalence of asthma, eczema, and sinus problems/allergies increase with level

of education, they were included as confounders in our models to minimize the

impact of a selection effect (Tables 5 and 6).

120

Another limitation is the absence of information on how long before pregnancy
OC was used and if it was the last method of birth control used. This limitation
introduces a non-differential misclassification, which tends to underestimate the
association between OC and asthma or wheezing, and coughing. To evaluate,
whether the OC hypothesis holds true, we investigated data of children from
pregnancies that occurred despite birth control (n=90), 42 with and 35 without
using oral contraceptives. Asthma was more frequently reported at 11-12 years
of age in children whose mothers used 00, 23.6 %, compare to 17.1 % in

children whose mothers did not use OC when they conceived.

We did not collect information on parental atopy and thus could not adjust for it in
our model. While there is data showing parental history of atopy to be a risk
factor for atopic disorders in the offspring, data supporting an association
between parental history of atopy and oral contraceptive use is lacking and the
plausibility of such an association is questionable. Hence, it is highly unlikely that
parental history of atopy would have confounded the association between oral

contraceptive and atopy, since confounding requires both, association with

44
exposure and outcome

The present study was not originally designed to address our hypotheses.
However, an advantage of the cohort study design is that data collected can be

used to provide answers to newly emerging research questions. We therefore

121

lJfi-_

 

took advantage of the utility and efficiency of this study to test our respective
hypotheses. Additionally, the unique nature of this cohort makes this the first

study to test perinatal risk factors for asthma in an Afro-Caribbean population.

Regarding coughing, we asked for ‘frequent night time or early morning
coughing’, but did not include the adjective ‘dry'. Nevertheless, the validity of

“early morning coughing’ as a marker for asthma has been found to be sufficient

45'46. On the other hand, it has been shown that bronchial hyperreactivity was

not related to dry cough or to nocturnal cough 47. According to Chang, coughing

and bronchoconstriction may be caused by different mechanisms. Hence,

different asthma-like symptoms may indicate different features 48

The definition or measurement of asthma or wheezing may also be of concern as

wheezing is a heterogeneous disorder 49 and not all that wheezes is asthma 11.

The impact this misclassification may have, was assessed by combining the
variables asthma or wheezing and frequent nighttime or early morning cough to
create a compound outcome variable. The combined condition was less
prevalent and represented a more specific state of bronchial reactivity. The
association between the combined variable and maternal OC use before
pregnancy was assessed and results showed a stronger association (aOR =
3.15, 95 % Cl 1.29- 7.71). Additionally, we investigated whether the association

with asthma was more likely due to an atopic response. For this purpose, we

122

combined two variables: having both, asthma and allergies. The association
between maternal OC and the new combined variable was stronger (aOR 2.26,
95% CI 1.18-4.35) compared to the association with only asthma (aOR 1.81,
95% CI 1.25-2.61). This strengthening suggests that the association is likely to

have been the result of an atopic mechanism.

In summary, this data contributes to existing literature suggesting that maternal
oral contraceptive use may be important in the development of asthma in

offspring. In particular, the data helps to address the ongoing question as to

whether it is ‘hormones or hygiene’ 50 that is responsible for atopy. There are

only a few existing studies that allow for the investigation of maternal OC use and

the child’s risk of atopic manifestation. The hypothesis that oral contraceptive use

may contribute to an increase of asthma in children as put forward by stt 3 is

very thought provoking. We suggest this hypothesis be tested in future studies

with more precise exposure assessment.

Acknowledgements

Funding for the Jamaican Perinatal Morbidity, Mortality Survey was provided by
the International Development Research Center of Canada while the Jamaica
Cohort study was supported by the Policy Development Unit of the Planning

Institute of Jamaica.

123

References

10.

11.

12.

ASHER Ml, KEIL U, ANDERSON HR, et al. International Study of Asthma and
Allergies in Childhood (ISAAC): rationale and methods. The European
Respiratory Journal : Official Journal of the European Society For Clinical
Respiratory Physiology 1995;82483-91.

BEASLEY R, CRANE J, LAI CK, PEARCE N. Prevalence and etiology of
asthma. The Journal of Allergy and Clinical Immunology 2000;105:8466-
72.

WJST M, DOLD S. ls asthma an endocrine disease? Pediatric Allergy and
Immunology 1997;82200-4.

XU B, JARVELIN MR, HARTIKAINEN AL, PEKKANEN J. Maternal age at
menarche and atopy among offspring at the age of 31 years. Thorax
2000;55:691-3.

NAFSTAD P, MAGNUS P, JAAKKOLA JJ. Risk of childhood asthma and allergic
rhinitis in relation to pregnancy complications. J Allergy Clin lmmunol.
2000;106:867-73.

ANNESl-MAESANO l, MOREAU D, STRACHAN D. In utero and perinatal
complications preceding asthma. Allergy 2001;56:491-7.

ASHLEY D, MCCAW-BINNS A, FOSTER-WILLIAMS K. The perinatal morbidity
and mortality survey of Jamaica 1986-1987. Paediatr Perinat Epidemiol
1988;2z138-47.

ASHLEY D, MCCAW-BINNS A, GOLDING J, et al. Perinatal mortality survey in
Jamaica: aims and methodology. Paediatr Perinat Epidemiol 1994;8 Suppl
1:6-16.

KLEINBAUM K, MULLER. Applied Regresion Analysis and Other Multivariable
Methods, 1988.

SAS INSTITUTE. Statistical Analysis System, Version 8. Gary, NC, 2000.
MURRAY DM, LAWLER PG. All that wheezes is not asthma. Paradoxical
vocal cord movement presenting as severe acute asthma requiring
ventilatory support. Anaesthesia 1998;53:1006-1 1.

BARBIERI RL, GAO X, XU H, CRAMER DW. Effects of previous use of oral

contraceptives on early follicular phase follicle-stimulating hormone.
Fertility and Sterility 1995;64:689-92.

124

13.

14.

15.

16.

17.

18.

19.

20.

21.

22.

MOORE JW, KEY TJ, WANG DY, BULBROOK RD, HAYWARD JL, TAKATANI 0.
Blood concentrations of estradiol and sex hormone -binding globulin in
relation to age at menarche in premenopausal British and Japanese
women. Breast Cancer Research and Treatment 1991;18 Suppl 1:S47-50.

MOVERARE R, ELFMAN L, STALENHEIM G, BJORNSSON E. Study of the
Th1fTh2 balance, including IL-10 production, in cultures of peripheral
blood mononuclear cells from birch-pollen-allergic patients. Allergy
2000;55:171-5.

MAZZARELLA G, BIANco A, CATENA E, DE PALMA R, ABBATE GF. Th1/Th2
lymphocyte polarization in asthma. Allergy 2000;55:6-9.

KUO ML, HUANG JL, YEH KW, Ll PS, HSIEH KH. Evaluation of Th1/Th2 ratio
and cytokine production profile during acute exacerbation and
convalescence in asthmatic children. Ann Allergy Asthma lmmunol
2001;86:272-6.

PICCINNI MP, GIUOIZI MG, BIAGIOTTI R, et al. Progesterone favors the
development of human T helper cells producing Th2-type cytokines and
promotes both lL-4 production and membrane CD30 expression in
established Th1 cell clones. Journal of Immunology (Baltimore, Md. :
1950) 1995;155:128-33.

SZEKERES-BARTHO J, WEGMANN TG. A progesterone-dependent
immunomodulatory protein alters the Th1fl’h2 balance. J Reprod lmmunol
1996;31:81-95.

HAMANO N, TERADA N, MAESAKO K, et al. Effect of female hormones on the
production of lL-4 and lL-13 from peripheral blood mononuclear cells. Acta
Otolaryngol Suppl 1998;537:27-31.

MICHEL FB, BOUSOUET J, GREILLIER P, ROBINET_LEVY M, COULOMB Y.
Comparison of cord blood immunoglobulin E concentrations and maternal
allergy for the prediction of atopic diseases in infancy. The Journal of
Allergy and Clinical Immunology 1980;65:422-30.

WARNER JA, JONES CA, WILLIAMS TJ, WARNER JO. Maternal programming
in asthma and allergy. Clin Exp Allergy 1998;28 Suppl 5235-8; discussion
50-1.

JONES CA, HOLLOWAY JA, WARNER JO. Does atopic disease start in foetal
life? Allergy 2000;55:2-10.

125

23.

24.

25.

26.

27.

28.

29.

30.

31.

32.

33.

34.

BURKE L, SEGALL-BLANK M, LORENZO C, DYNESIUS-TRENTHAM R, TRENTHAM
D, MORTOLA JF. Altered immune response in adult women exposed to
diethylstilbestrol in utero. Am J Obstet Gynecol 2001;185:78-81.

PETERS TJ, GOLOING J. The epidemiology of childhood eczema: ll.
Statistical analyses to identify independent early predictors. Paediatric and
Perinatal Epidemiology 1987;1z80-94.

Xu B, JARVELIN MR, PEKKANEN J. Prenatal factors and occurrence of
rhinitis and eczema among offspring. Allergy 1999;54:829-36.

VIHKO R, APTER D. Endogenous steroids in the pathophysiology of breast
cancer. Critical Reviews in Oncology/Hematology 1989;9:1-16.

APTER D, REINILA M, VIHKO R. Some endocrine characteristics of early
menarche, a risk factor for breast cancer, are preserved into adulthood.
International Journal of Cancer. Journal International Du Cancer
1989;44:783-7.

KIRCHENGAST S, HARTMANN B. Association between maternal age at
menarche and newborn size. Soc Biol 2000;47:114-26.

WESTERGAARD T, BEGTRUP K, ROSTGAARO K, KRAUSE TG, BENN CS,
MELBYE M. Reproductive history and allergic rhinitis among 31145 Danish
women. Clin Exp Allergy 2003;33:301-5.

F RYE C, MUELLER JE, NIEDERMEIER K, WJST M, HEINRICH J. Maternal oral
contraceptive use and atopic diseases in the offspring. Allergy
2003;58:229-32.

NAFSTAD P, SAMUELSEN SO, IRGENS LM, BJERKEDAL T. Pregnancy
complications and the risk of asthma among Norwegians born between
1967 and 1993. Eur J Epidemiol 2003;18:755-61.

CALVANI M, ALESSANDRI C, SOPO SM, et al. Infectious and uterus related
complications during pregnancy and development of atopic and nonatopic
asthma in Children. Allergy 2004;59:99-106.

STAZI MA, SAMPOGNA F, MONTAGANO G, GRANDOLFO ME, COUILLIOT MF,
ANNESI-MAESANO l. Early life factors related to clinical manifestations of
atopic disease but not to skin-prick test positivity in young children. Pediatr
Allergy lmmun012002;13:105-12.

NILSSON L, KJELLMAN NI, LOFMAN O, BJORKSTEN 8. Parity among atopic
and non-atopic mothers. Pediatr Allergy Immunol 1997;8z134-6.

126

35.

36.

37.

38.

39.

40.

41.

42.

43.

44.

45.

PICCINNI MP, MAGGIE, ROMAGNANI 8. Role of hormone-controlled T-Cell
cytokines in the maintenance of pregnancy. Biochem Soc Trans
2000;28:212-5.

SCHWARTZ J, GOLD D, DOCKERY DW, WEISS ST, SPEIZER FE. Predictors of
asthma and persistent wheeze in a national sample of Children in the
United States. Association with social Class, perinatal events, and race.
Am Rev Respir Dis 1990;142:555-62.

SVANES C, OMENAAS E, HEUCH JM, IRGENS LM, GULSVIK A. Birth
Characteristics and asthma symptoms in young adults: results from a
population-based cohort study in Nonivay. Eur Respir J 1998;12:1366-70.

STEFFENSEN FH, SORENSEN HT, GILLMAN MW, et al. Low birth weight and
preterm delivery as risk factors for asthma and atopic dermatitis in young
adult males. Epidemiology 2000;11:185-8.

GREGORY A, DOULL I, PEARCE N, et al. The relationship between
anthropometric measurements at birth: asthma and atopy in childhood.
Clin Exp Allergy 1999;29:330-3.

OLIVETI JF, KERCSMAR CM, REDLINE 8. Pre- and perinatal risk factors for
asthma in inner City African-American children. American Journal of
Epidemiology 1996;143:570-7.

SEARS MR, HOLDAWAY MD, FLANNERY EM, HERBISON GP, SILVA PA.
Parental and neonatal risk factors for atopy, airway hyper-responsiveness,
and asthma. Arch Dis Child 1996;75:392-8.

KATz KA, POCOCK SJ, STRACHAN DP. Neonatal head Circumference,
neonatal weight, and risk of hayfever, asthma and eczema in a large
cohort of adolescents from Sheffield, England. Clin Exp Allergy
2003;33:737-45.

MCCAW_BINNS A, STANDARD_GOLDSON A, ASHLEY D, WALKER G,
MACGILLIVRAY I. Access to care and maternal mortality in Jamaican
hospitals: 1993-1995. International Journal of Epidemiology 2001;30:796-
801.

ROTHMAN KJ. Modern epidemiology, 1986.
KUEHR J, FRISCHER T, KARMAUS W, MEINERT R, BARTH R, URBANEK R.

Clinical atopy and associated factors in primary-school pupils. Allergy
1992;47:650-5.

127

46.

47.

48.

49.

50.

FERRIS BG. Epidemiology Standardization Project (American Thoracic
Society). Am Rev Respir Dis 1978;118:1-120.

STRAUCH E, NEUPERT T, IHORST G, et al. Bronchial hyperresponsiveness to
4.5% hypertonic saline indicates a past history of asthma-like symptoms in
Children. Pediatr Pulmonol 2001;31:44-50.

CHANG AB. Cough, cough receptors, and asthma in Children. Pediatr
Pulmonol 1999;28:59-70.

DELL S, TERESA, T. Breastfeeding and asthma in young children. Arch of
Pediatric Adolescence medicine 2001;155:1261-1265.

RANGARAJ S, DOULL l. Hormones not hygeine? Birth order and atopy. Clin
Exp Allergy 2003;33:277-8.

 

 

mdm 3:. GE B x33 “me E mEman 5.8: Lo 305: 9230

4.4 .3 9. 3x

SS :3 9. 3. - 9. ma

N.N_. Nov 9. m.~v - Ego; :Em

v.0 me >Eo mmSECon.

odm mdv 0.2259 ucm Emmi

mém NS. 2:0 685 I 9.68;

mom Own 9E8 5316 8:2. N4 55:5 m>_._

m9. mg. E3503 :9: 8:9:

vhm Qmm Emucoomm :9: $3 - cozmosum .3562

03 m. I. bo>=mu Co Loam. mctzu wcozmgano 5.3: _mEQNE

QNN o.N_. Long 0.5.3 :96 28535 5 2392.4

No me Loom. Bonus

mar Qm >ocmcmoa act—6 co_mw_Eum _ano; _mEQms.

5mm mNm 5:385 mESU 25:0th _mEofis.

va v.8 55:85 9:28 wcozmo__ano _mEQmE

adv Emv mm: 3:302:00 _Eo _mEoumE
83ers 68qu 292$

3992 um new
commas. E O_QEmmbsm

9.: 80: %sz Ema 0925

33.329 xmwzifi ucm _EmELmdv
@ev tocoo 05.3mm

 

3353 09:5 ucm 958mm 5 mm_nm_._m> osmoaxo Co 83:209ch cowtquoO ... 29¢...

129

«Me M 952835 :32;

 

 

 

xmao=m
2: 3: qt. t: is 3: QB .26 ES 5 E285
35m .0 .96“. $1
:98 mEEoE 2:8 .0
mm _..m ...? me Q? 9:329: Emscmi
0.9 «.8 men @5322, Lo nEEma.
avers Scans anus Ears 321: «means sons
02 mo> oz mo> oz wm>
go
965 :
mEEoE 25m .0 m5~mmn>> Lo mE£w< emEmNom

_Eo._. oEficmE Emzcmi
mcozfimmL—EmE 062m EmEEn m5 h_o 85:38 9853366 ho mcoanoi .N 2an

130

_ I ‘ .
. ‘
Ell-I... $1.11..“

 

xmflém 550
wdm N: wdm on. m? o. E 55 NNN odw mdm :N v.3 mEom 5 E039:
0:50 :05.— >01

 

 

3 4.0 we «.8 00 S we 2 on E :0 3 Emma
:maou
me on ..m E ..m we we mam on E 0.0 3. SEE E8 5
0.55:9: E0309...
. . . . . . . . . . . . 3.322,
N: m 3 :0. I: .0. am. me. am. .2 com me. N0. 59:59..
381: «ENS Roars «RYE SKIS nouns gens aware 38".: fans Remus swans 0252.6
02 00> 02 00> 02 00> oz mm; oz 00> O_mEmu 22>.
6:039: 5:209: @550 mm: 02300928
gmwcmfimmfi: @550 m:o_mm_Eum mmwdwwmmmqmmficmc m:o_.mo__an0 _m:o _mEBmE 50:00
. 0 .x m _mgamo: _mEBms. .. a. ._ u _>_ _mefis.

 

9900... ._m: 803 E :05 9 830.9 :_ 909006 0590 .o :oquoi .m 052.

131

 

 

35:0 550 0E00 5

 

 

0.2 mew N t. 0 mm wow 08 E039: 0:50 55. >01
«.0 ms 5. ~.~_. 0.0 v. E 080~0m

:mzoo mEEoE
0.4 ..0 0.0 0.0 0.0 ...v 2.00 5 0E:

:5: E0309:
0.9 0.3 0.9 0.0? 53 0.9 m:_~00:>> 5 0E£0<

a Bus 39".: €055 ammus 63".: 855 as 0:80.30
02 00> 02 00> 02 00>
502.00 5 5:0. 5:0_ @550 520 5:0_ 00030:.

@550 0:0:00__an0 55902

259005 5 0009094

05.00.. .0: .0355: .9055: :55 9 520.9 :_ 050506 0590 5 :25qu .v 030...

132

 

35:0 550 0:50 5

 

 

0.0? 0.0.0. 0.0V :0 0.0.. 0.00 0.00 0.0.. E0305 00:.0 5>0._ >0...
0.0 0.3 0.0 0.0 0.0 0.0 v.0 0.0 0E0N0m.
. . . . . . . . 00:00 0.:_:5E >000
N0 00 00 F0 F0 00 00 00 5085;050:0002“.
0.0? ND. 0.: 0.: 0.3 0.5 50.. 0.3 05000:; 5 0E£0<
ammus Emu... fins @0010 38".: 810 sins Gowns .x. 05850
00. 9. 0.0 00. 2:0 00.0.05”. 2:0
02 00> 0.0 A 9. 00 0.0 v 00.:E5u. .0 .0005 00005
0.... 0o 9.82 £00; 55 0:000“.

0 00:: :_ 005.005
5.00: 5 000:... 0.2.00

0550.. 0.0: .0055: 55:50 :55 2 500.0: :_ 05050.0 0.0000 5 5.0002,... 8005508 .0 0.00...

133

 

 

 

00.0 0 029.0 .
.0:00 5.00: 2 000000 0:0 5000000 .000 5:55.: 0:0 50:00 5.. >._0:00.000 0:0 0.000 05 :. 00.0009 5.. 00000.5( ..
00.0-8.0 00.0 2.0.00.0 00.0 3.0.00.0 00., 000-000 00.0 00. 0.?

0 0 P F P F 0 F .9. 0.0 I 9. 0.0 8020000
000.000 00.0 3.0-00.0 00.0 2.0-00.0 «00 00.7000 00.0 9.0.0v 000.0; 500
00. 0.000 00.0 00. 7R0 00.0 B. 30.0 00.0 00. 700.0 00.0 00:00.0 .020 00 00:52

.00_ 0o 0.8; 00:0 0.0

00. 0.000 00.0 2.0-00.0 0 2 00.300 00.0 00.0-2.0 00.0 0. 052020 5.8,. :o 0000:. 0.2.00
>5>..00

vmduomd oné 00600.0 3.; No. .3506 or... 00. Tend o _.._. 9:030 000.002.9000 50:205.

. . . . :80.
00 0 S 0 00 0 0c 0 00 0 0.0 F 00 F 00 0 00 0 v0 _ 00 0 : F 05:00 0020 00050000 :0 2390.2
00. 0.000 00.0 5.0-00.0 50 00. 700.0 00.0 2.0-00.0 «I :80 08:00.

0050050 .0555; M...
000-000 00.0 000-000 00.0 00. 0-00.0 00.0 00. 30.0 00.0 5000090 00.50 000.050 1
3.0-00.0 03 8.0-00.0 03 00. 0.000 :0 05-000 0: 0.002020
0:000 5.00.800 .00.0000 5:555.
000.000 00.0 3.0-00.0 0: 00. 7:0 00._ {.300 00.0 502020 00.50 00000000. 00:90.2
>0:0:m0:0
05.000 00.0 00.0-03 00.0 5.0-00.0 00.0 00.0-8.0 .03 00.50 0808.058 00:20.2
«0.0.3.0 .000 00.0-00; .00.0 5.0-00.0 .0: 000.0: .00.? 00: 20080008 .90 .0520:
0080050 55:05.0.
_0 .000 000 .0 0.00 000 _0 $00 000 .0 $00 000 920$
"2
.05 0 000%

00:00 0552: >000 5
0.0.0.5.: 50:00:”.

0500002, 5 00:50.0.

 

00:00 05:50: >000 5 08.0094

E00000 0:0 0500003 5 00:500. 50 050000 0.0.: 55:50 0:0 55:25 5 :00 0.0205. 00:00::00 $00 0:0 00000: 0000 .0 0.00...

Ed!» 405‘. 1 Inn.“—

mod w 00.0.4.0 ...
.0000 5.00: 0. 000000 0:0 5000000 .000 5505:. 0:0 50:00 5. >__0:o_._000 0:0 0.00. 0:. :_ 00.:0:0> 50 00.00.00. ._-

 

 

 

00..-00.0 00.0 00..-00.0 00.0 00.0-00.0 .0.. 000.000 00.. 0.. 0.0A
0 _- . 0 F . . . .00. o..0l0x 0.0 00:05.50.
00.7000 .00 0.7.0.0 00.0 0.70.0 00.0 .0.-0.0 00.0 000.0v .00.00>00.0
00.0-00.0 .000 000.000 .000 00.7000 00.0 00.0-00.0 .000 00002000280000.0002
.- . . .- . . .- . . .- . . 00:50.00;
0.0000 .0. 0000.. .00. 000000 00. 00000. .0.0 0000.00.03.0000.000000005000000
00..-00.0 00.0 00.7000 0... 00.7000 00.0 00.0-.00 0... 0002.00 00.0000000000000000000.02
00.7000 00.. 00.0.00. .00.. 000.000 00.. .0700. ..00 0000.00.30 002000050000000.000.000.
00..-00.0 00.0 00.7000 .0.. 000.00 0... 00.7000 00.0 0000.88.00.
0080200 .000550.
007000 00.0 00..-00.0 00.0 00.0-00.0 00.. 000-000 .0.. 0.0000090 000000020000
. . . . . . . . . . . . .>0:0:000n.
0.7000 000 0.7000 000 000-000 00. 000-000 00. 00000 00.00.500.00000000.0.0.2
00.7000 00.. 00.7000 00.. 00.7000 00.0 007000 00.0 00000090 00000000000.0__0000002
0.0-00. ..00. 00.7000 00.. .00-000 00.. 00.0000 00.. 000000000 00.00000000000000000000.05.
00.7.0.0 .0.. 00.7000 00.0 0.0-00.0 .0.. 00.0-.00 00.. 00: 0200800080000000.00,.
00E00.0o _0.0:0.:<
.0 00.00 0.00 .0 00.00 000 500.00 0.00 .0 00.00 0.00 00200“.

.000n0. .000u0.

>05__0 5:5 0:50 5

0:500
E0305 0.55 5 55.. >0... m.

 

>05__0 5:5 00:00 5 E03900
000.0 5 55.. >0: 0:0 0.00000 5.. 05.000 0.00 5.00.50 0:0 50000.00 5 ..0. 05:00:. 00:00:08 00000 0:0 .0000. 0000 .0 0.000

135

COMPONENT FOUR

Postnatal exposure to SHD affects biomarkers of allergy in

infancy

136

1-

m _. I. n'AI

Immune function biomarkers in children exposed to lead and

organochlorine compounds: a cross-sectional study

Wilfried Karmaus”, Kevin R. Brooks“, Thomas Nebez', Jutta Witten3', Nadia Obi-

Osius“, Hermann Kruses'

1 Department of Epidemiology, Michigan State University, 8601 West Fee Hall,
East Lansing, MI 48824, USA

2 Central Laboratory, University Hospital Mannheim, Germany

3 Ministry of Social Welfare Hesse, Department of Health, Wiesbaden, Germany
4 Epidemiological Working Group of the Ministry of Environment and Health and
the Institute for Mathematics and data management in Medicine, University
Hospital Hamburg-Eppendorf and

5 Institute of Toxicology, Christian-Albrecht University, Kiel, Germany

*These authors contributed equally to this work

§Corresponding author

Abstract

Background: Different organochlorines and lead (Pb) have been shown to
have immunomodulating properties. Children are at greater risk for exposure to
these environmental toxicants, but very little data exist on simultaneous

exposures to these substances.

Methods: We investigated whether the organochlorine compounds (00)
dichlorodiphenylethylene (DDE), hexachlorobenzene (HCB),

hexachlorocyclohexane (y-HCH), the sum of polychlorinated biphenyls (ZPCBs)

137

and Pb were associated with immune markers such as immunoglobulin (lg)
levels, white blood cell (WBC), counts of lymphocytes; eosinophils and their
eosinophilic granula as well as lgE count on basophils. The investigation was
part of a cross-sectional environmental study in Hesse, Germany. In 1995,
exposure to OC and Pb were determined, questionnaire data collected and
immune markers quantified in 331 children. For the analyses, exposure (00 and
Pb) concentrations were grouped in quartiles (y-HCH into tertiles). Using linear
regression, controlling for age, gender, passive smoking, serum lipids, and
infections in the previous 12 months, we assessed the association between
exposures and immune markers. Adjusted geometric means are provided for the

different exposure levels.

Results: Geometric means were: DDE 0.32pg/L, XPCBs 0.50 pg/L, HCB
0.22 pg/L, y-HCH 0.02 pg/L and Pb 26.8pg/L. The ZPCBs was significantly
associated with increased lgM levels, whereas HCB was inversely related to lgM.
There was a higher number of NK cells (CD56+) with increased y-HCH
concentrations. At higher lead concentrations we saw increased lgE levels. DDE
showed the most associations with significant increases in WBC count, in lgE
count on basophils, lgE, lgG, and IgA levels. DDE was also found to significantly

decrease eosinophilic granula content.

Conclusion: Low-level exposures to OC and lead (Pb) in children may have
immunomodulating effects. The increased lgE levels, lgE count on basophils,

and the reduction of eosinophilic granula at higher DDE concentrations showed a

138

1: ..i‘yfl 1 in..."
.A

most consistent pattern, which could be of clinical importance in the etiology of

allergic diseases.

139

fz-QHI‘ .,
H '

Background

Environmental toxicants such as organochlorine compounds (00) and lead (Pb)
may alter immune responses. There is a paucity of studies reporting associations
between organochlorine [1-4] and lead [5-8] exposures and immune function

biomarkers in children.

We conducted a large-scale environmental study of second-grade school

children in three regions south of the Federal State of Hesse, Germany in 1995. ;Li
Two of the regions are situated in the Rhine Valley with low mountains on both “1“
sides. One of these areas with several municipalities is located within a 10 km
radius around an industrial waste incinerator and other industries, such as
chemical plants. One plant was associated with dichlorodiphenylethylene (DDE),
hexachlorobenzene (HCB), and hexachlorocyclohexane (y-HCH) pollution [9].
The other region, also industrial, is 15 km north (downwind) of the incinerator.
Both Rhine valley regions are also intensively used for the production of
vegetables. The third study region is located in low mountains (about 0.4 km
above sea level) that separate it from the industrial area. Blood concentrations of
PCBs were shown to be higher in children living close to the toxic waste
incinerator [10]. Results on P085 and thyroid hormones, chromium and
lymphocytes, DDE and breastfeeding and asthma have been published

elsewhere [4, 11-15].

Considering infection and atopic disorder in children, we have previously shown
an association between DDE blood levels; asthma and one immunoglobulin (lg),

namely lgE [4]. However, the potential effects of organochlorines on other lgs

140

and cellular defense were not reported. Hence, the focus of this paper is to
investigate the impact of organochlorine compounds and Pb on humoral immune
markers and cell-mediated immune responses. Specifically, for immune
responses we focus on leukocytes, lymphocytes, B-cell, T-cells and their
subsets. Assuming a concurrent effect of QC on immune markers, we conducted
cross-sectional analyses of the data from the first of three surveys conducted in
1994/1995, 1996, and 1997. Only the first investigation included an extensive

clinical assessment of immune markers. ..

Methods

Study population

After obtaining approval from the Data Protection Agency of Hamburg, Germany,
the Ministry of Cultural Affairs of Hesse, Germany, and the local school
committees, we invited the parents of 1,091 second-grade school children in 18
townships to participate in our study. We obtained informed consent from all
participating parents, according to the requirements of the Ethical Committee of
the Board of Physicians, the Helsinki Declaration, and the Data Protection
Agency of the State of Hamburg. We asked each parent to allow their child to
participate in phlebotomy only when passive smoking in the private household

did not exceeded 10 cigarettes per day during the previous 12 months.

Questionnaires
We used four self-administered parental questionnaires in the survey: one

regarding the living condition and nutrition of the family, one for each parent, and

141

one regarding information on the child. Duration of breastfeeding was recorded in
weeks of total and in weeks of exclusive nursing. Environmental tobacco smoke
(ETS) was graded as smoking in the child‘s home in the previous 12 months (no
cigarettes, 1-10 cigarettes, 11-20 cigarettes, 21-30 cigarettes, more than 30
cigarettes per day). We recorded age, gender, and the number of infections,
defined as cold, coughing, and sore throat, with or without fever, in the last 12

months (none, less than 5 infections, 5-10 infections, more than 10 infections).

Laboratory analyses of blood samples

One parent accompanied each child in the medical examination. For blood
sampling, we used the 'Vacutainer System' (Becton, Dickinson & Company, San
José, California) Approximately 25 mL were drawn and separated into different
aliquots. lmmunoglobulin (lg) E in serum was quantified at the Medical,
Alimentary and Veterinary Institute for Research Middle Hesse, Division of
Human Medicine, Dillenburg, Germany, using a florescence-immunoassay (CAP,
Pharmacia, Uppsala, Sweden). To determine levels of specific lgE against
inhalant allergens (aeroallergens), we incubated serum with immunocaps
containing a mixture of aeroallergens and determined the reactivity using a
fluorescence measurement (UNICAP Pharmacia, Uppsala, Sweden). Results
from this method were provided in semi-quantitative format. We also measured
lgA, G, and M by laser immunonephelometry (Dade Behring, Liederbach,
Germany). The results for IgA, G and M were provided in mg/dL and for lgE in
kU/L serum. Triglycerides and cholesterol were measured on a clinical chemistry

analyzer according to IFCC methods (Hitachi 717, Boehringer Mannheim).

142

 

Leukocyte subsets

We collected 8 mL of blood in tubes containing EDTA and mixed them to prevent
clotting. This aliquot was transported to the Central Laboratory of the University
Clinic of Mannheim and analyzed on the same day. We used 200 pL of blood for
the automated differential (laser-based hematology analyzer CDSSOO, Abbott
Diagnostics, Santa Clara, California), and 100 uL for each of the nine three-color
test tubes analyzed by flow cytometry (FACScan, Becton, Dickinson, &
Company, San José, California, equipped with a 488 nm air-cooled argon ion
laser). Eosinophils were determined according to their specific depolarisation
characteristics and their eosinophilic granula content by the intensity of light
scatter by flow cytometry. Basophils were identified by their high lgE density on
the cell surface using immunofluorescence with a Phycoerythrin (PE) labeled

anti-lgE antibody.

We used monoclonal antibodies directed against specific cell surface antigens to
differentiate cell populations by multicolour immunofluorescence. Three
antibodies were simultaneously applied with the fluorochrome combination
FlTC/PE/PE-Cy5. CD4/CD8/CD3 was used to detect absolute number of
lymphocytes, T-helper cells and cytotoxic T-cells; CD19/CD5/IGE was used to
differentiate B-cell subsets and basophils; CD3/CD16 and CD56/CD57 were
used for natural killer cells. CD45RO defines memory T-helper cells. The CD
nomenclature assigns the antibodies to clusters of differentiation, according to

the International Workshop on Human Leukocyte Differentiation Antigens [16].

143

 

Organochlorine compounds (GO) in blood

00 including eight PCB congeners (101, 118, 138, 153, 170, 180, 183, 187),
DDE, HCB, and three HCH congeners (a-, B- and y) were determined (in pg/L) at
the Institute of Toxicology, University of Kiel, Germany. OC were analyzed in 5
mL samples of whole blood by high resolution gas chromatography (HRGC,
Model 3400 by Varian Inc., Palo Alto, California) with a 63Ni—electron-capture-
detector. The detection limit (DL) (two times the signal/low-noise ratio) was 0.02
pg/L for B- and y-HCH, DDE and each PCB congener, and 0.01 ug/L for HCB
and a-HCH. For extraction and clean-up procedures, we used florisil and n-
hexane for elution (9 g florisil were deactivated with 3% H20 and placed in a
chromatography column 22 mm in diameter and 48 mm in length). The capillary
column amounted to 30 mm in length and 0.25 mm in diameter; nitrogen was
used as a carrier gas. We determined the PCB congeners by retention times on
the chromatograms and identified them by comparison with known standards.
Additionally, we tested reliability with gas chromatography-mass spectroscopy
(GC/MS). The laboratory successfully participated in nationwide quality

assessments for the determination of these OC.

Lead in blood

Lead (Pb) analysis was done at the Institute of Toxicology, University of Kiel,
Germany. The determination in whole blood samples was by flow injection
atomic absorption spectroscopy (Perkin Elmer) after adding 0.1% Triton —X-1-
solution and 15 mol nitric acid to from a solution. This solution was then

centrifuged at 3000 rpm. The DL for Pb was 9 pg/L (48 nmol/l; atomic weight:

144

 

207.19).

Data analyses

Since the data for leukocytes (WBC) and their subsets (lymphocytes and
eosinophils), immunoglobulins, DDE, PCB congeners, HCB, y-HCH and Pb were
not normally distributed, the geometric mean, 5-, 95-percentiles are provided. In
order to obtain a multivariate normal distribution, we log-transformed the number

of cells and immunoglobulins before testing associations with possible predictors

- wuLTr WW'

by multiple linear regression models.

1*

All statistical analyses were performed using SAS software [17]. We calculated
the sum of the PCB congeners (ZPCBs = sum of seven congeners, the congener
PCB101 was not detected). For descriptive purposes, we substituted values of
00 below detection limit with one half of the detection limit. The statistical
procedure (PROC RANK) was used to group exposure variables into quartiles
(DDE, PCBs, HCB and Pb) or tertiles (y-HCH). All observations below the
detection limit were part of the lowest level group (reference). To account for the
influence of lipids on the concentration of OC, we controlled for the sum of
triglycerides and cholesterol in the regression analyses. Further steps were taken
to determine whether our results were different when lipids were represented as
sum of triglycerides and cholesterol as opposed to triglycerides and cholesterol

as individual variables.

We used linear regression models (PROC GLM) with immune markers as

dependent variables and all organochlorine compounds and lead as independent

145

variables in each model. We also controlled for potential confounders (age,
gender, environmental tobacco smoke (ETS), number of infections during the last
12 months, and lipid concentration). Information on passive smoking (ETS) in the
child’s home in the previous 12 months was divided into four categories (no
cigarettes, 1-10 cigarettes, 11-20 cigarettes, 21 cigarettes per day and more).

For the number of infections we considered four categories (none, less than 5, 5-
10, more than 10). Age of the child was divided into three groups; 7, 8 and 9-10

years.

From the results of the regression analyses, we calculated adjusted geometric
means for leukocyte subsets and immunoglobulins for increasing categories of
exposure. T-tests were used to compare the statistical effect of higher exposure

group to the lowest (reference).

 

Since one major route of exposure to

 

Child's exposure
the pollutants analyzed is breast concentration

feeding [18-21] and breastfeeding \

provides passive immunity [22-24], [ Breastfeeding I

immune markers and pollutants could

 

 

 

 

 

 

 

Figure 7: Diagramatic representation of
be spuriously correlated if breast the breastfeeding, childhood exposures

and immune markers associations
feeding is not controlled for.
However, this triangle (Figure 7) cannot be tested with linear regression models,

as intervening variables do not qualify as confounders [25]. Controlling will

146

VL'T "I,

 

reduce the initial association between the risk factor and the marker, as one
causal chain is split into two associations. Thus, we explored the relationship
between childhood breastfeeding (total duration of breastfeeding in weeks), the
concentration of OC, and immune response by path analysis [26], using the

CALIS procedure SAS Institute [17].

Results

The proportion of participation was 61.5 % (671 of 1091). We obtained blood
samples from 350 children, conducted 00 and Pb analyses on 343 samples, and
quantified immunoglobulins in 340. Overall, information (i.e., questionnaires,
exposure biomarkers, and immune markers) was available for 331 children.
Fewer girls than boys participated in phlebotomy; and 96 % of the children were
7 to 8 years of age (Table 1). Due to the inclusion criterion for blood sampling
(passive smoking of less than 10 cigarettes in the child’s home), the prevalence
of passive smoking was also lower in the group with phlebotomy than in the total
group (Table 1). Nevertheless, the fact that parents were separated or divorced
and shared cohabitation for their child, resulted in a re-assessment of the passive
smoking status after phlebotomy. Eligibility was determined on the information
provided by one parent (mother or father) for their household. In the case of
separate dwellings, we re-assessed the exposure by taking the average number
of cigarettes smoked in both homes. As a consequence, 26 (7.9%) children who
were exposed to more than 10 cigarettes per day at home had a phlebotomy and

were included in the analyses.

147

 

For y-HCH, 27.7 "/0 of the observations were below the detection limit, 2.9% for
Pb, whereas none for DDE and HCB. At least one of seven PCB congeners was
detected in each sample. Whole blood concentration for the sum of PCB
congeners (118, 138, 153, 170, 180, 183, 187), HCB and of Pb showed a decline
with increasing age (Table 2). DDE, PCB, and HCB concentrations were lower in

children with higher passive smoking exposure. Regarding infections, lead

- TIIHTV

concentration was higher in children with more than 10 infections during the last

 

12 months, whereas DDE concentration was lower in this group (Table 2).

The concentrations of DDE, EPCBs (sum of PCBs), and HCB were all correlated
(Table 3). However, we used categorized levels of OC, which were then only
marginally correlated; the highest rank correlation was for the PCB and HCB
groups (rspearman=0.46). These correlations did not result in multicollinearity since
the tolerance (variance of 00 not explained by other predictors) was at least
53%. The volume-based organochlorine concentrations were only marginally
correlated with the lipid serum levels. To adjust for lipid concentrations, we
included lipid concentrations as a confounder in the explanatory models for
leukocyte subsets and immunoglobulins. Results derived from models using the
sum of triglycerides and cholesterol compared to triglycerides and cholesterol as
individual variables did not reveal any substantial difference (data not shown).
We therefore reported results from models using the sum of triglycerides and

cholesterol.

148

Regarding lead in whole blood, we found weak correlations with whole blood
levels of OC (DDE: r= 0.15, n = 331, p < 0.01; HCB: r= 0.14, n = 331, p < 0.01;

y-HCH: r= -0.02, n = 331, p < 0.70; EPCBs: r= 0.14, n = 331, p < 0.01)

Increased white blood cell count (WBC; total leukocytes) was evident in the
group with highest DDE level, whereas Pb, at the second, along with PCB at the
highest level was associated with a reduction in WBC count. An increase in the
number of eosinophils — a leukocyte subset - was identified in the highest DDE
category, but not statistically significant, Table 4. However, eosinophilic granula
content was significantly reduced at the upper DDE levels. In addition, lgE count
on basophils was increased at higher DDE exposure, being statistically

significant for the 0.3-0.43pg/L category.

Regarding lymphocytes and specific lymphocyte subsets (B-cells, T-cells), the
number of T-cells (CD3+), cytotoxic T-cells (CD8+) and B-cells (CD19+) were all
significantly reduced in the median Pb category (Table 5). Both natural killer
(NK) cells (CD56+) and a NK cells subset (CD57+) were significantly associated

with y-HCH. However, these associations did not reveal dose-dependency.

All four immunoglobulins were associated in a virtually dose dependent fashion
to either DDE, HCB or P085 (Table 6). lgM serum levels increased with the
concentration of PCBs (F-test, p<0.01) but decreased with increasing
concentration of HCB (F-test, p<0.01). In the two upper quartiles of DDE
exposures, IgA levels were significantly higher, but lower in the upper quartile of

HCB. DDE was not associated in a dose-response mode with lgG (F-test,

149

 

p=0.14), however, compared to the reference, the highest DDE exposure group
showed a significantly elevated lgG level (t—test, p=0.04). IgE levels more than
doubled as DDE concentration increased (F-test, p=0.02). The Pb serum levels
were related to a significant differences in lgE (F-test: p=0.028), but not in a dose

dependent fashion (Table 6).

Figure 8 shows that both 70 mew—m—

 

 

 

 

 

 

__ l
DDE and lead were 50 * “ “W . “‘"wj
...... ;
associated with higher 50 ’ ‘ l
. Q 40 - ' -_‘
serum lgE levels in 3 '
. , E 30 ii _
children. In groups wrth 9
4i ------------- a
20 ,
lower DDE blood __,_ _Pb <= 28 ug/L
1o .......
concentrations, Pb ' Pb >28 “91"
0 ’ .
concentrations above the <=o.2 0.21 -o.29 o.3-<0.43 >043
DDElug/L)
median (28IJ9/L) were Figure 8: The combined effect of increasing DDE

and lead (Pb) on lgE serum levels
related to increase lgE

levels. In groups with higher DDE, there was no additional effect of Pb.

Statistically, the combined effect of DDE and Pb on lgE was not significant.

In order to determine whether breastfeeding confounded the associations
identified in linear regression models (Figure 7), we repeated our analyses using
structural model (path analysis) for exposures determined as significant in linear

regressions. Inclusion of breast feeding did not substantially change our findings.

150

 

Discussion

In 331 school children, age 7-10 years, we demonstrated significant relationships
between 00 and Pb whole blood concentration and cellular and humoral immune
markers. First, modest associations were found between NK cells (CD3-
CD16+CD56+) and a subset of natural killer cells (CD3-CD16+CD56+CD57+)
and y-HCH (Table 5). Second, HCB was inversely related to lgM. Third, ZPCBs

were directly related to lgM. Fourth, our data showed that Pb decreased the

 

count of T-cells (003+), cytotoxic T-cells (CD3+CD8+), and B-cells (c03+005+ .,
CD19+). This reduction was most evident at the 22.1 — 28.3 pg/L Pb
concentration, though not in a dose response fashion. Lastly, DDE was inversely
related to all immunoglobulins, except lgM (Table 6). However, DDE was not
associated with total serum protein (data not shown). The DDE effect was
strongest for lgE - more than twofold increase - which also corresponded to an
increased count of lgE on basophils. We did not detect a significant relationship
between DDE and eosinophils, nevertheless, the number of eosinophils was
positively correlated with lgE (rspeaman = 0.4, p < 0.01). However, high DDE
levels were found to be significantly associated with lower eosinophilic granula
content. The granula contains basic proteins which are cytotoxic and part of the

inflammatory response [27].

The cross-sectional nature of the study limits conclusions on whether exposure
occurred before immune responses. We can assume that organochlorine
concentrations do not vary substantially in childhood, post breastfeeding. There

is a decline of PCBs and HCB with age (Table 2), however the assumption of the

151

stability is supported by a follow-up of the same children and OC determined in
1997. The Spearman rank correlation between the two successive
measurements were high, with the exception of y-HCH: DDE: r = 0.86, n = 274, p
< 0.01; HCB: r= 0.74, n = 274, p < 0.01; y-HCH: r= 0.1, n = 270, p = 0.11;

ZPCBs: r = 0.82, n = 274, p < 0.01.

The reported concentrations for DC were not lipid-based. In this cohort, there is a
high correlation between lipid- and non lipid-based concentrations for CC (Table
3). Thus, our findings are independent of lipid- or wet weight-based
determinations. In our models we controlled for lipids instead of dividing the
concentration of 00 by the lipid concentration for three reasons. First, a simple
division assumes a monotonous linear relation between lipids and
organochlorines. Although Phillips and co-workers reported for 20 adults that
division by lipids reduces the difference between fasting and non-fasting
concentration of CC [28], there is no data to justify a linear relation. Our data in
children showed only weak correlations between OC and the sum of cholesterol
and triglycerides (Table 3). This correlation did not increase when the sum of
lipids were derived by using the 2nd formula proposed by Phillips et al. [28].
Second, there is no standard approach to adjust concentrations below the limit of
detection for lipids. In particular, the probability of detection may be influenced by
the individual lipid concentration of a child. Third, division by lipids does not take
into account that they may confound the organochlorine - immune response

relationships. Confounding is likely since lipids and DC are correlated, plus lipids

152

are, for example, associated with the count of lymphocytes [29, 30].

There is evidence that breast milk is a significant source of OC, Pb [18-21], and
passive immunity [22-24]. Path analytical techniques (Figure 7) were used to
verify whether breastfeeding as an intervening variable confounded our
associations. The inclusion of breastfeeding in the path analysis did not reveal
results different from the linear regression models. Hence, the associations

between pollutants and immune markers were independent of breastfeeding.

We found whole blood concentrations of OC in our cohort comparable to similar
children in Germany [31]. Compared to children in the United States, age 12-19
years (NHANES - 1999-2000) [32], our DDE values were lower though still within
the 95% confidence interval. However, when comparing our results (in whole
blood) with those of NHANES (in serum), we have to consider differences
between serum and whole blood concentrations. Mes et al. reported that DDE
was higher in sera and plasma than in whole blood samples [33]. Conversely,
PCBs were higher in whole blood samples. No other comparison with NHANES
data was possible as the values for PCB congeners and other OC were below

the limit of detection.

Regarding lead (Pb), the geometric mean of 27ug/L in our investigation was
similar to the 33pg/L found in a study of 797 East-German children 5-14 years of
age [34]. Against that, the 1999-2000 NHANES study showed a lower geometric
mean (15.1 pg/L) in 905 children 6-11 years of age [32]. However other studies in

areas of higher exposure, reported average concentrations above the NHANES

153

 

value: 40 pg/L for children, 6 to 15 years of age in four communities with mining
and smelting operations and two control groups in the United States [6], and

95pg/L in Chinese children 3-6 years old [8].

We selected a subgroup for blood analyses due to budget constraints. The group
having a lower ETS exposure in their homes was selected to reduce the

potentially confounding effect of ETS. This group did not significantly differ from

other participating children (Table 1). Parents did not know the individual results

m—

of the blood analyses, when they provided information on their children, thus

reducing recall bias.

The inverse association between DDE and the number of infections 12 months
prior to the interview is surprising (Table 2). However, in a logistic regression
model the number of infections reported did not show a significant protective
effect of DDE. Additionally, when infection was eliminated from the models, there

were no major changes in the OC - immune markers association.

The few existing studies estimating the immunotoxicity of lead (Pb) in children,
measured by immune markers, are inconsistent in their findings. Regarding
immunoglobulins, our positive relation between Pb and lgE was consistent with
that of Lutz et al. [7]. However, Sun and co-workers had different results [8].
Concerning lymphocytes, we found that the number of B-cells was significantly
reduced with increased Pb concentration. Conversely, Sarasua et al. reported an

increase in the number of B-cells for children less than 3 years old [6].

154

Studies assessing the relation between organochlorine and immune markers,
determined in our study, also showed conflicting results and focused mostly on
adults [35-37]. In comparison with these adult studies, Vine et al. reported similar
modest findings for immunoglobulins and DDE. However, only results for IgA
showed statistical significance [35]. Our findings regarding lgE and eosinophilic
granula suggest that DDE shifts the immune response into a T helper (Th) 2 F

direction [38]. Mechanistically, immune responses have been polarized into Th1

9 A

and Th2 reactions. Th1 responses lead to the secretion of immunoglobulin G
(IgG) and removal of the allergen. The allergic Th2 phenotype is characterized by
secretion of cytokines that promote immunoglobulin E (lgE) production resulting
in allergies. This suggestion is in agreement with findings of Daniel and co-
workers, who reported an association between DDE and interleukin-4, a Th2
cytokine [39]. In addition, our interpretation that DDE may be associated with an
allergy-like response is supported by the distribution of aeroallergen-specific lgE
results over the four DDE exposure levels. In the lowest DDE exposure group
11.3% of the children showed a positive specific lgE, 10.9% and 12.2% in the

two intermediate groups, but 23.0% in the highest DDE exposure group (p=0.03).

Interestingly, the effects of lead (Pb) and DDE on lgE seems to be competitive.
At lower DDE exposure, Pb seems to increase lgE concentrations (Figure 8).
There was no additional effect of the other pollutant if one is high; therefore it is
possible that both pollutants are involved in the same mechanism. Indeed,

studies have surmised that Pb may also shift the immune responses in a Th2

155

direction [40-42].

There are only few studies on OC blood/serum concentration and immune
responses in children. Weisglas-Kuperus et al. reported that prenatal PCB
exposure was associated with an increase in the T-cell markers CDBCD8+ and
CD4+CD45RO+ [2]. Our data did not support these findings. In another study
with prenatal exposures to PCBs, HCB, and DDE, Dewailly et al. did not identify
significant associations with immune markers including CD3+, CD4+, CD8+
lymphocytes nor with IgA, IgG, and lgM [3]. However, we found significant
relationships between PCBs and HCB with lgM (Table 6). Reichrtova et al. have
shown that in utero exposure to DDE is positively correlated with cord serum lgE
[43]. No other study of children has investigated the relationship between DDE
determined postnatally and Th2 markers such as lgE and eosinophilic granula.
This is the second publication showing an association between DDE and serum
lgE [4] and the first to report associations between Pb, and DDE and lgE count

on basophils and eosinophilic granula.

Conclusion

In conclusion, our study suggests a non-linear association between lgE and Pb
concentration. Regarding OC, our data indicated an increase of lgE related to
DDE serum concentrations. A parallel association between DDE, lgE count on
basophils, and reduction of eosinophilic granula contents further supports a

potential stimulation of a Th2 response related to DDE exposure.

156

Prospective studies should determine more than one DC in a scenario with
multiple exposures in order to prevent spurious correlations and include repeated
determinations of immune responses to determine changes in immune
development during childhood. Furthermore, studies are warranted that

determine allergic susceptibilities following DDE and Pb exposure in children.

Acknowledgements i
This study was supported by the Ministry of Environment, Energy, Youth, Family

 

and Health Hesse, Germany. We acknowledge Dr. Rauterberg for the analyses to
of immunoglobulins. The analyses were conducted in preparation of an EPA
STAR grant (R830825). KRB was a graduate assistant in the EPA grant while

preparing the final manuscript.

157

References

1.

10.

11.

Marth E, Sixl W, Bencko V, Medwed M, Lapajne S, Voncina E, Brumen S:
People on the garbage dumps of Cairo: a toxicological in vivo model?
Cent EurJ Public Health 1995, 3:154-157.

Weisglas-Kuperus N, Patandin S, Berbers GA, Sas TC, Mulder PG, Sauer
PJ, Hooijkaas H: Immunologic effects of background exposure to
polychlorinated biphenyls and dioxins in Dutch preschool children. Environ
Health Perspect 2000, 108:1203-1207.

Dewailly E, Ayotte P, Bruneau S, Gingras S, Belles-Isles M, Roy R:
Susceptibility to infections and immune status in Inuit infants exposed to
organochlorines. Environ Health Perspect 2000, 108:205-211.

Karmaus W, Kuehr J, Kruse H: Infections and atopic disorders in
childhood and organochlorine exposure. Arch Environ Health 2001,
56:485-492.

Rabinowitz MB, Allred EN, Bellinger DC, Leviton A, Needleman HL: Lead
and childhood propensity to infectious and allergic disorders: is there an
association? Bull Environ Contam Toxicol 1990, 44:657-660.

Sarasua SM, Vogt RF, Henderson LO, Jones PA, Lybarger JA: Serum
immunoglobulins and lymphocyte subset distributions in children and
adults living in communities assessed for lead and cadmium exposure. J
Toxicol Environ Health A 2000, 60:1-15.

Lutz PM, Wilson TJ, Ireland J, Jones AL, Gorman JS, Gale NL, Johnson
JC, Hewett JE: Elevated immunoglobulin E (lgE) levels in children with
exposure to environmental lead. Toxicology 1999, 134:63-78.

Sun L, Hu J, Zhao Z, Li L, Cheng H: Influence of exposure to
environmental lead on serum immunoglobulin in preschool children.
Environ Res 2003, 92:124-128.

Forth W: Hexachlorcyclohexan - Gift in den Lebensmitteln? Deutsches
Aerzteblatt 1980, 83:2169-2176.

Holdke B, Karmaus W, Kruse H: [Body burden of PCB in whole human
blood of 7-10-year—old children in the vicinity of a hazardous waste
incinerator]. Gesundheitwesen 1998, 60:505-512.

Osius N, Karmaus W, Kruse H, Witten J: Exposure to polychlorinated

biphenyls and levels of thyroid hormones in children. Environ Health
Perspect 1999, 107:843-849.

158

.I’F .—
d

12.

13.

14.

15.

16.

17.

18.

19.

20.

21.

22.

23.

Karmaus W, Huang S, Osius N, Nebe T: Chromium urine concentration
and effects on lymphocyte subpopulations in children. J Environ Med
1999,11153-161.

Karmaus W, DeKoning EP, Kruse H, Witten J, Osius N: Early childhood
determinants of organochlorine concentrations in school-aged children.
Pediatr Res 2001, 50:331-336.

Karmaus W, Davis S, Chen Q, Kuehr J, Kruse H: Atopic manifestations,
breast-feeding protection and the adverse effect of DDE. Paediatr Perinat
Epidemiol 2003, 17:212-220.

Obi-Osius N, Misselwitz B, Karmaus W, Witten J: Twin frequency and
industrial pollution in different regions of Hesse, Germany. Occup Environ
Med 2004, 61:482-487.

International workshop on human leukocyte differentiation antigens
[http:l/www.ncbi.nlm.nih.gov/prow/guide/45277084.html

 

SAS Institute. Statistical Analysis System, Version 9.1. Gary, NC, 2002-
2003.

Abraham K, Papke 0, Gross A, Kordonouri O, Wiegand S, Wahn U, Helge
H: Time course of PCDD/PCDF/PCB concentrations in breast-feeding
mothers and their infants. Chemosphere 1998, 37:1731-1741.

Lackmann GM, Schaller KH, Angerer J: Organochlorine compounds in
breast-fed vs. bottle-fed infants: preliminary results at six weeks of age.
Sci Total Environ 2004, 329:289-293.

Rabinowitz M, Leviton A, Needleman H: Lead in milk and infant blood: a
dose-response model. Arch Environ Health 1985, 40:283-286.

Ettinger AS, Tellez-Rojo MM, Amarasiriwardena C, Bellinger D, Peterson
K, Schwartz J, Hu H, Hernandez-Avila M: Effect of breast milk lead on
infant blood lead levels at 1 month of age. Environ Health Perspect 2004,
112:1381-1385.

Hanson LA: Breastfeeding provides passive and likely long-lasting active
immunity. Ann Allergy Asthma Immunol 1998, 81 :523-533; quiz 533-524,
537.

Hanson LA, Korotkova M, Lundin S, Haversen L, Silfverdal SA, Mattsby-

Baltzer l, Strandvik B, Telemo E: The transfer of immunity from mother to
child. Ann N YAcad Sci 2003, 987:199-206.

159

 

24.

25.

26.

27.

28.

29.

30.

31.

32.

33.

34.

35.

Van de Perre P: Transfer of antibody via mother's milk. Vaccine 2003,
21 :3374-3376.

Rothman KJ: Modern Epidemiology. Boston/Toronto: Little Brown and
company; 1986.

Hatcher L: A step-by-step Approach to Using SAS for Factor Analysis and
Structural Equation Modeling. 6 edn. Cary, NC: SAS Institute Inc.; 1994.

Leung DY: Molecular basis of allergic diseases. Mol Genet Metab 1998,
63:157-167.

 

Phillips DL, Pirkle JL, Burse VW, Bernert JT, Jr., Henderson LO, IE...
Needham LL: Chlorinated hydrocarbon levels in human serum: effects of
fasting and feeding. Arch Environ Contam Toxicol 1989, 18:495-500.

Moreno LA, Sarria A, Lazaro A, Lasierra MP, Larrad L, Bueno M:
Lymphocyte T subset counts in children with hypercholesterolemia
receiving dietary therapy. Ann Nutr Metab 1998, 42:261-265.

Ogawa Y, lmaki M, Yoshida Y, Shibakawa M, Tanada S: An
epidemiological study on the association between the total leukocyte and
neutrophil counts, and risk factors of ischemic heart disease by smoking
status in Japanese factory workers. Appl Human Sci 1998, 17:239-247.

Landesgesundheitsamt Baden-Wuerttemberg: Projekt
Beobachtungsgesundheitsaemter. Belastungs- und Wirkungsmonitoring.
Stuttgart; 2000.

Second national report on human exposure to environmental chemicals
lhttp:l/www.cdc.qov/exposurereport/Znd/pdf/secondner.pdfl

Mes J, Marchand L, Turton D, Lau P, Ganz P: The determination of
polychlorinated biphenyls congeners and other chlorinated hydrocarbon
residues in human blood, serum and plasma. A comparative study. Intem
J Environ Anal Chem 1992, 48: 1 75-1 86.

Jacob B, Ritz B, Heinrich J, Hoelscher B, Wichmann HE: The effect of
low-level blood lead on hematologic parameters in children. Environ Res
2000, 82:150-159.

Vine MF, Stein L, Weigle K, Schroeder J, Degnan D, Tse CK, Backer L:

Plasma 1,1-dichloro-2,2-bis(p-chlorophenyl)ethylene (DDE) levels and
immune response. Am J Epidemiol 2001, 153:53-63.

160

36.

37.

38.

39.

40.

41.

42.

43.

Daniel V, Huber W, Bauer K, Suesal C, Conradt C, Opelz G: Associations
of dichlorodiphenyltrichloroethane (DDT) 4.4 and
dichlorodiphenyldichloroethylene (DDE) 4.4 blood levels with plasma |L-4.
Arch Environ Health 2002, 57:541-547.

Cooper GS, Martin SA, Longnecker MP, Sandler DP, Germolec DR:
Associations between plasma DDE levels and immunologic measures in
African-American farmers in North Carolina. Environ Health Perspect
2004, 112:1080-1084.

Mosmann TR, Sad S: The expanding universe of T-cell subsets: Th1, Th2
and more. lmmunol Today 1996, 17:138-146.

Daniel V, Huber W, Bauer K, Suesal C, Conradt C, Opelz G: Associations
of blood levels of PCB, HCHs, and HCB with numbers of lymphocyte
subpopulations, in vitro lymphocyte response, plasma cytokine levels, and
immunoglobulin autoantibodies. Environ Health Perspect 2001, 1092173-
178.

McCabe MJ, Jr., Lawrence DA: Lead, a major environmental pollutant, is
immunomodulatory by its differential effects on CD4+ T cells subsets.
Toxicol Appl Pharmacol 1991 , 1 1 1:13-23.

Heo Y, Lee BK, Ahn KD, Lawrence DA: Serum lgE elevation correlates
with blood lead levels in battery manufacturing workers. Hum Exp Toxicol
2004, 23:209-213.

Heo Y, Lee WT, Lawrence DA: In vivo the environmental pollutants lead
and mercury induce oligoclonal T cell responses skewed toward type-2
reactivities. Cell lmmunol 1997, 179:185-195.

Reichrtova E, Ciznar P, Prachar V, Palkovicova L, Veningerova M: Cord
serum immunoglobulin E related to the environmental contamination of
human placentas with organochlorine compounds. Environ Health
Perspect 1999, 107:895-899.

161

 

Table 1: Descriptive characteristics of the study cohort.
Total group Subgroup with CO

 

 

and immune
markers
(N=671) (n=331)
% %
Boys 53.1 56.8
Age 7 years 45.8 46.2
8 years 50.2 50.2
9-10 years 4.1 3.6
Passive smoking in the child’s home
during the last 12 months(cigarettes per day)
None 52.2 66.5
1-10 23.4 24.8
11-20 14.3 5.3
more than 30 10.1 2.4
Number of infections during the last 12
months
None 6.0 5.7
Less than 5 74.7 74.8
5 to 10 17.2 17.4
more than 10 2.1 2.1
Duration of total breastfeeding (weeks)
0 19.1 15.1
less than 5 17.9 15.4
5 — 8 12.5 12.1
9- 12 10.6 11.8
more than 12 34.7 41.1
Missing 5.2 4.5

Serum cholesterol concentration

(mean, 5-95%-value, mg/dL)
Triglyceride concentration
imean, 5-95%-value, mg/dL)

162

186 (143-235)

130 (53-262)

 

 

 

 

 

$.27 38 v.2“
8.213: new
:2: I 3: New
22% Isa: 33

8:14.28 .3
2:: $2802
8.? I cs: 0.3
2.: I 3: new
3.3 $284.2
8.: I no: new
$.27 e28 new
3.3 I 3: EN
:2 «If :28

$329888
23829888
3328888
83.329888

$320888
23829888
$349888
6329888
3329888
288228.888
832983.:
$329888
33.323888

Charm—.8 ad
afiolfi _dv mmd
23.0.; :8 mmd
Smdlo— .8 $6

:2: 2.828
aseI882 .o
GE: 288.0
32; :33
84888828
:25; _8 .3
627 2 _8 3o
comer _ 18mg
223; _ .8 :8

EST :8 one
:N.NI 8.8 28
a? I 2.8 83
2&7 8.8 as

$.de 2.8 Re
8.2 I 2.8 83
E: I 2.8 one
8: I :8 SB
327 88 Rd
33 I 2.8 :8
so; I 2.8 who
33 I 2.8 and
33 I28 28

5.8 I 2.8 «we
a; I 2.8 one
:23 I 2.8 :3
a: I 2.8 88

2 I I 2.8 23
33 I 8.8 3o
:23 I 2.8 e3
22: I 3.8 28
is I 8.8 :8
23.9 I 2.8 is
28.: I 2.8 NS
See I 2.8 23
E: I 2.8 22.

5 8n
:8 EA

:1me v
3: 9:02

A287:
A237:

:8 8 I _
528 252

a: as; 8 d

Go: 28% w
an: as; e

828 mom
So: 25

23:92

N. 22s. 2:
wztsn 2.2585
mo $3552

:8

com 329833
mfizoE 2 a2
2: mats—o 0E0:
PEEU 2: E
wExcEm 332$

masocwIQw/x

Sacco

 

338 ca

388 :81?

3&3 mu:

3&8 emote N

388 man

A5 @0350

828.50

 

.mouatgoo .3 am can DO coo—a 9.2—3 .5.— mo=_a> Roma ..m ...—a 58:. 2.52.806 "N 235.

163

 

n3 .30 .03 6: 69. d? _w: .2: mmeomcoo mom ho Ezm mmOQN

.Gmdnseaam: mm; 00.3808 500 So: 020: .29 0858 00:90:00 x02

 

 

.82....“ so mezzo 20 N 0.05.2 0500 02.0.8.8 022. .002 0
002.8sz ucm 602.0205 6 E30 mm 62200.8 028. .99. 3
$08 3 082.20:
00.0 .10:
0.00 20208 a 00000-20:
2.2 02.0 .00...
00.00 8.00 20208 3 0008-20:
00.3: 3.0 5.0 08%
59¢ 800 20.00 608 a 808.20:
00.02 00.0 00.0 00.0 .moo
20.00 00.00 00.00 N000 00.0V0 0.08
«0.0 00.0 00.0 00.0 2.0 :0 2.0 :08 10:-»
:00 20.00 20.00 000 $00 0000 200
~00 00.0 00.0 E0 00.0 00.0 00.0 8.0 :03 mo:
:00 S00 5.00 5.03 «0.0...0 00.00 tone 5.00
00.0 00.0 05.0 00.0 00.0 00.0 00.0 00.0 050 :08 000de
E00 S00 000 S00 00.00 800 20.00 20.00 5.0%.
00.0 3.0 020 5.0 00.0 00.0 00.0 9.0 00.0 20.0 :08 moo
cmmE

050580 05:80:-» 29:00: 29.3092 055000 0020: 9020: 10:-»

m0: mmOQN
.mcuoE o_.=oEoom :2: new

:3": €803-23. can com—3.35 mucsonEoo 8500200590 2353 3:22:30 cozflotoo cant—03m “a min...

164

 

 

Table 4: White blood cell, eosinophilic characteristics, and basophilic surface lgE by CC and Pb (geometric mean}
HCB (pg/L)

 

 

 

DDE (pg/L) Sum of PCBs (pg/L) y-HCH (pg/L) Pb (pg/L)
Outcome 3 0.2 0.21—0.29 0.30- 0.43 >043 50.30 0.31—0.48 0.49— 0.75 >075 s 0.15 0.16 —02 0.21—0.27 >027 0.01 0.02 >002 <22.0 22.1—28.3 28.4— 34.1 >34.1
N 78 89 79 85 80 86 82 83 84 77 86 84 91 130 110 82 81 86 82
White blood cells, x1 03/10
Total: crude 8136 8579 8131 8555 8373 8456 8528 8074 8456 8354 8383 8233 8062 8512 8421 8612 7907 8376 8552
Adjusted § 7782 8275 7970 8584’ 8354 8318 8447 7488‘ 8280 8037 8141 8131 7920 8284 8247 8397 7730' 8167 8290
F-test: p=0.09 F-test: p=0.02 F-test: p=0.92 F—test: p=0.09 F-test: p=0.33
Eosinophil cell count, x103/pL
Total: crude 182 206 188 207 178 182 197 210 192 187 193 190 186 206 177 176 166 210 214
Adjusted § 176 220 208 223 181 194 217 235 245 217 196 172 199 225 194 190 185 233 218
F-test: p=0.37 F-test: p=0.61 F-test: p=0.31 F-test: p=0.32 F-test: p=0.22
Eosinophilic granula
Total; crude 902 903 901 893 897 901 896 904 899 897 905 897 906 896 898 890 904 900 904
AdeSted § 919 913 910 895' 902 907 905 923 908 907 916 906 915 906 906 898 913 911 914
F-test: p=0.22 F-test: p=0.37 F-test: p=0.65 F-test: p=0.47 F-test: p=0.27
lgE count on basophils
Total: CTUde 810 822 863 859 807 842 838 866 808 841 839 867 845 838 833 836 811 845 860
Adjusted § 851 851 896' 884 860 884 861 877 852 879 866 886 880 868 863 871 845 878 888
F-test: p=0.67 F-test: p=0.50 F-test: p=0.41 F-test: p=0.60 F-test: p=0.10

 

'1 absolute number of cells/0L.
p s 0.05 based on a t-test compared with the lowest exposure category as the reference.

§ Adjusted for all exposures in the table (OC & Pb) in addition to gender, age, number of infections in the last 12 months, passive smoke exposure in the child's home in the last 12 months, and lipids

(sum of cholesterol and triglycerides)

165

 

 

 

 

 

 

 

Table 5: Lymphocyte phenotypes by whole blood DDE, PCBs, HCB, y-HCH and Pb concentration (geometric mean)
DDE (pg/L) Sum of PCBs (pg/L) HCB (119/L) y—HCH (pg/L) Pb (pg/L)
<07 071-079 030 04‘ >043 <030 031048 049- > <
Cells _ .- .- .2 . - . 3 . _ . . - . 0.75 0.75 _0.15 0.16 - 0.2 0.21- 0.27 >027 0_01 002 >002 <22.0 22.1-28.3 28.4-34.1 >34.1
N 78 89 79 85 80 86 82 83 84 77 86 84 91 130 110 82 81 86 82
T—cells (CD3+)
Total: 7
crude -193 2183 2139 2242 2286 2179 2144 2156 2289 2170 2155 2146 2131 2212 2213 2318 2101 2160 2184
Adjusted § 1950 2005 1980 2076 2092 1998 1990 1932 2082 1975 1986 1968 1967 2020 2021 2118 1919' 1979 1999
F-test: p=0.71 F-test: p=0.74 F-test: p=0.76 F-test: p=0.77 F-test: p=0.17
T-helper cells CD3+CD4+)
Total
crude: 1204 1209 1205 1251 1276 1212 1198 1189 1293 1190 1209 1180 1155 1232 1255 1297 1182 1203 1192
AdeSted§ 1087 1138 1147 1200 1183 1145 1141 1100 1220 1113 1139 1100 1101 1150 1176 1214 1106 1128 1123
F-test: p=0.51 F-test: p:0.83 F-test: p=0.37 F-test: p20.37 F-test: p=0.24
Cytotoxic T-cells
(CD3+CD8+)
Totali
crude 753 743 727 778 773 740 739 752 769 744 740 751 750 753 748 799 711 746 750
AdeSted § 660 665 649 692 697 660 668 642 678 658 662 669 671 669 659 712 634‘ 661 662
F-test: p=0.78 F-test: p:0.80 F--test: p=0.97 F-test: p=0.93 F—test: p=0.23
Memory T-helper cells (CD4+CD45RO+)
Total
crud: 315 249 352 325 316 277 308 325 265 349 311 307 331 274 325 317 304 327 341
Adjusted § 332 323 363 362 343 342 347 347 354 363 345 317 348 341 345 358 321 348 351
F-test: p=0.18 F -test: p=1.00 F-test: p=0.34 F—test: p=0.91 F-test: p=0.22
Natural killer cells (CD16+CD56+)
Total .
crud: 384 367 350 372 382 378 350 364 369 362 394 348 328 395 373 371 348 365 389
Adjusted § 371 338 334 362 378 366 317 345 326 348 390 342 322 368' 305' 350 322 354 378
F-test: p:0.48 F-test: p=0.24 F-test: p=0.16 F-test: p<0.01 F-test: p=0.20 >
Natural killer cells subset (CD16+CD56+CD57+) 3
Tot
cruzl: 156 153 148 148 159 156 142 147 155 144 164 142 137 162 151 154 148 148 156
Adjusted§ 168 157 158 162 177 171 142 156 148 153 183 162 143 175‘ 167 162 152 161 169
F-test: p=0.89 F-test: p=0.21 F-test: p=0.18 F-test: p=0.03 F-test: p=0.69
B-cells (CD3+CD5+CD19+)
T
c352]: 457 474 456 468 463 461 464 469 473 482 45 8 446 445 474 469 505 423 462 469
Adjusted§ 378 398 382 393 369 372 395 416 420 411 375 349 371 395 397 418 353' 389 393
F-test: p=1.00 F-test: p=0.67 F ~test: p=0.26 F-test: p=0.53 F-test: p=0.10

 

I absolute number of subtype cells/uL based on percent of lymphocytes x total lymphocyte count.
p- < 0. 05 based on a t-test compared with the lowest exposure category as the refe
§ Adjusted for all exposures in the table (0C & Pb) in addition to gender, age, num

and triglycerides)

166

rence.
ber of infections in the last 12 months, passive smoke exposure in the child's home in the last 12 months and lipids (sum of cholesterol

 

 

 

 

0002.33wc. .050 300.00.00.93 8:0. 020: “.50 0500:. N. .00_ on. E 0:50. 0.2.0.0 2.. ... 0.000000

3.60:0 0.0.0000. .0502... N. .00. 2.. :. 0:0..00..E ..o .5953: 60.0 doucow o. 02.6.00 ... 30 0% 00. 0.0.0. 2.. E 00.200008 :0 81. 8.00.030 a
095.0%. 0... 0.0 0.00.0.8 0.00096 .0030. 2.. 0...? 3.00.0.3 .00.-. .0 so .0003 3.3 w 00 .

 

00.010 009.0 00.010 003.0 00.010 002.0 00010380-“. 00.010 080-0
00 00 00 00 .0 00 .0 00 00 .0 00 00 .0 00 00 .00 00 00 00 08.00.00.
00 00 00 .0 00 00 00 00 .0 00 00 00 00 00 00 .0 00 00 00 0.02.3
.0:..00.

00010 08.-.. 00.010 00.2-.. .000 08.-.. .000 002-0 00.010 009-0
00. 00. 00. 00. 00. 00. 00. .00. .0. 00. 00. .00. 00. 00. 00. 00. 00. 00. 00.000.00.03
00. 00. 0.. 00. 00. 0.. 00. 0.. 00. 00. 00. .0. 0.. 00. 00. 00. 00. .0. 00. 00.5
30.08.20.

00.010 08.-.. .0010 000... 00.010 080-0 00.010 080-... 00.010 080-0
00. 00. .0. 00. 00. .0. 00. .... 00. 00. 00. 00. 0.. 00. .0. ..0. .00. 00. 0:03.903.
005.0

00. 00. 00. 0.0. .0. 00. 00. 00. 00. 00. 00.1mm.-. 00. 00. 00. 00. 00. 00. 00.

30.08. <0.

00.010 003-... 00.010 0000.0 00010 009.0 00.010 08.-.. 0.010 0000.0
.00. .00. 0.0.. 0.0. 000. 000. .00. 00: 00: 80. 000. 000. 0.0. 00: 000. .000. 00.. 00.. E. 0.8.9.3.
203.0

0.: .0.. 0:. 00.. 00.. .... .0.. 00: 000. 00.. 00: 00.. 00.. 00.. 00.. 00.. 00.. 0... .0:

30.08.00.
00 00 .0 00 0.. 00. .0 00 00 00 00 00 00 00 00 00 00 00 00 z

..00 0.00 00.0 0.0 00.0 00.0 00.0 00.0
..00A -000 -..00 0.00v 00.0A 00.0 .00 00.0A -.00 0.0 0.00 00.0A -000 -.00 00.00 00.0A -000 -.00 0.00 088.00

 

3.00. 00

3.00. 0.0:-..

30.... mo:

3.00. 0.0.00.0 800

3.00. .00.0.
Enos. £50582

00.2.8 ... 00.02.0808 00 0:... :01... do... 8.000 Mon 08... 0.2.; .3 05.000.020.08. ”0 0.000

167

CHAPTER 4

DISCUSSION

168

 

 

The pieces of the sex hormone disrupters — childhood allergy puzzle presented in
this monograph suggest an immunomodulating role of sex hormone disrupters in
the etiology of childhood allergic disorders. First, evidence was presented in
support of DDE playing a role in the prenatal priming of allergic diseases through
the alteration of cord plasma cytokines. Second, an association between prenatal
exposure to exogenous sex steroid hormones and biomarkers of allergic
susceptibility was demonstrated. Third, prenatal exposure to exogenous sex
steroid hormones appears to be associated with allergic phenotypes such as
asthma; and fourth, postnatal organochlorine exposure seems to alter serum

biomarkers of allergy.

The motivation for this work stems from my curiosity to investigate the
speculation that sex steroid hormones disrupters may be important in the etiology
of childhood allergy. This effort was represented by the use of both questionnaire
and lab data for various exposures and outcomes. In addition, the data used
were from cohorts of varying backgrounds (Caucasian and African-Caribbean)

and study designs (cross-sectional and follow-up).

The pathways explored in the first three manuscripts highlights the importance of
in utero exposure by ‘connecting the dots' showing a link between: 1) placental
p,p’ DDE and cord plasma lL-4, |L-13, and INF-y (cytokines suggesting allergy
susceptibility); 2) Maternal oral contraceptive use before pregnancy (a proxy for
in utero progesterone/estrogen exposure) and serum lgE, IgA, and basophilic

surface lgE counts; 3) Maternal contraceptive use and asthma and hay fever in

169

 

 

offspring. These three pieces of work suggest that in utero exposure to sex
steroid hormones may alter the immune status at birth, which may result in
childhood allergic disorders. This suggests that the in utero environment may
play a major role in the pathogenesis of childhood allergy.

To the best of our knowledge, there is no report of an association between

placental p,p' DDE and cord plasma cytokines. Noakes and co-workers 1 were

unable to detect placental p,p’-DDE and used maternal levels instead. Though
their results for cord lL-13 and INF-y were not significant, they tended in the
same direction (increased |L-13, decreased INF-y) as our findings for these

cytokines.

This overall sex hormone — childhood allergy puzzle was made clearer with
findings from the fourth manuscript. In children ages 7 to 10 years, serum DDE
was shown to significantly increase humoral WBC count, lgE levels, lgE count on

basophils, 196, and IgA: all considered important immune markers of allergy.

This work would have benefited from the use of actual blood concentrations of

estrogen and progesterone since previous findings on DC use and hormone

levels are not consistent 2’3. However, since DDE is known to have weak

estrogenic properties, the association between placental DDE and cord cytokines
provides useful clues of possible results. Another limitation is the fact that, the

duration between 0C use and pregnancy was not taken into consideration. The

170

 

 

 

effect of 00 use on childhood allergy may be minuscule if the period between

00 use and pregnancy is long.

The fact that data from studies of different ethnic groups and study designs was
used is noteworthy. Data from an African-Caribbean cohort (component 3)
provided consistent findings to those from Caucassian cohorts. Another strength
of this work is the different time-windows in which exposures and outcomes were

assessed.

On a whole, the effects of sex hormone disrupters on childhood allergic disorders
are pervasive but unexplained. This compilation of four studies from three
diverse cohorts suggests possible explanations. The first component (In utero
exposure to SHD affects perinatal biomarkers of allergy) implies that placental
DDE (a SHD) concentration may influence the priming of cord blood immune
marker towards an allergic state. Whilst important, on its own this component
may not be sufficient to make the leap from placental DDE to childhood allergy.
This gap was narrowed by findings from component two (In utero SHD affects
postnatal biomarkers of allergy), which indicate a possible relation between
maternal oral contraceptive use before pregnancy and serum immune markers of
allery in offspring. Furthermore, the data suggest that maternal 00 use may
have sex-related differences of the immunomodulating effects in offspring. The
first two components breached the gap between the in utero and postnatal

environment. However, though immune markers (especially lgE levels) are good

predictors of disease outcome they do not always provide consistent results 4'8.

171

 

 

 

 

The investigation was therefore taken to another level with component three (In
utero exposure to SHD affects allergic outcomes in infancy) providing evidence
of a possible involvement of maternal oral comtraceptive use in the etiology of
childhood allergic diseases including asthma. By connecting the clues embedded

in these three studies, a picture is painted that implies that in utero exposure to

 

SHD may be a risk factor for childhood asthma (for example) by first primining

the fetal immune response. This priming may be sustained into childhood,

 

evident by altered serum allergic markers, and ultimately clinical allergic
manesfestations. The findings of component four (Postnatal exposure to SHD
affects biomarkers of allergy in infancy) was consistent with previous studies that
inferred a link between postnatal HOC exposure and serum immune markers.
These results should be interpreted with caution since HOC concentrations were
measured once at 7-10 years of age. Hence we do not know whether the
concentrations were higher before this age and are therefore responsible for the
observed immune response. For example, it is possible that the increased serum
immune levels may be the result of in utero and not postnatal exposures (see

component 2).

it would be advantageous to test all four components in one study; collecting all
relevant biologic samples and questionnaire data at the appropriate time.
However, since no such study exist, and may not be realistic, the present work
provides important stepping stones along the path of better understanding the

role of SHD (sex hormone disrupters) in the etiology of childhood allergy.

172

 

Future work

The mechanism through which these compounds (organochlorines, exogenous
estrogen and progesterone) disrupt normal sex hormone function is unclear. The
epigenetic concept has been proposed as a credible explanation. This concept

posits that adverse effects can result from enduring changes in gene expression

9. This means that the active portion of the genetic information is altered without

a change in the genetic blueprint (epigenesis). The expression or silencing of
genes then results in cells/tissue with permanent (irreversible) committed

activities.

Epigenetic mechanisms have been suggested as the underlying reason for the

observed increased risk of several chronic diseases in adulthood associated with

. . . . 1 -12
xenobrotrc exposures early in life 0 . For example, Fukuzawa et al. reported

that (coplanar) PCB affects the expression of steroidogenic enzyme genes 13. In

addition, Goldman et al. found that PCBs (Aroclor 1254) inhibit the cytochrome

P—450 21-hydroxylase activity, thus blocking the branch from progesterone to
glucocorticoid production 14. Furthermore, Chen et al. reported that DDT and its
metabolites (including DDE) down-regulate the gene expression of the estrogen

1 5
receptor

Future studies aimed at explicating the mechanism through which sex hormones
are disrupted should embrace the epigenetic concept. Since gene expression

can be detected as early as in fetal tissues of the placenta, it may be prudent to

173

 

 

 

assess the variations in gene expression in neonates with varying concentrations
of exposure to DDE, PCB, exogenous estrogen, and progesterone. The effect of
sex hormone disrupters on allergy may be from their direct impact on the gene

expression of allergy related cytokines such as lL—4, lL-13, and INF-y.

In summary, it is of personal and public health importance to consider the
potential impact of sex hormone disrupters on the pathogenesis of allergy. The
prenatal period is of particular importance since the fragile fetal immune system
is more vulnerable to exogenous exposures. Therefore, intervention through
education regarding exposure prevention should be implemented before
pregnancy. Further exploration of the epigenetic concept is a fertile area for

future research.

174

 

 

 

References

1.

10.

Noakes PS, Taylor P, Wilkinson S, Prescott SL. The relationship between
persistent organic pollutants in maternal and neonatal tissues and immune

responses to allergens: A novel exploratory study. Chemosphere
2006;63(8):1304-1311.

Barbieri RL, Gao X, Xu H, Cramer DW. Effects of previous use of oral
contraceptives on early follicular phase follicle-stimulating hormone.
Fertility and Sterility 1995;64(4):689-92.

Moore JW, Key TJ, Wang DY, Bulbrook RD, Hayward JL, Takatani 0.
Blood concentrations of estradiol and sex hormone -binding globulin in
relation to age at menarche in premenopausal British and Japanese
women. Breast Cancer Research and Treatment 1991;18 Suppl 12847-50.

Eiriksson TH, Sigurgeirsson B, Ardal B, Sigfusson A, Valdimarsson H.
Cord blood lgE levels are influenced by gestational age but do not predict
allergic manifestations in infants. pediatric allergy and immunology
1994;5(1):5-10.

Klink M, Cline MG, Halonen M, Burrows B. Problems in defining normal
limits for serum lgE. J Allergy Clin lmmunol 1990;85(2):440-4.

Bergmann RL, Edenharter G, Bergmann KE, Guggenmoos-Holzmann I,
Forster J, Bauer CP, Wahn V, Zepp F, Wahn U. Predictability of early
atopy by cord blood-lgE and parental history. Clin Exp Allergy
1997;27(7):752-60.

Perkin MR, Strachan DP, Hc W, Lack G, Golding J. The predictive value
of early life total immunoglobulin E measurement in identifying atopic
children in a population-based birth cohort study. Pediatr Allergy lmmunol
2006;17(2):118-24.

Croner S, Kjellman Nl, Eriksson B, Roth A. lgE screening in 1701 newborn
infants and the development of atopic disease during infancy. Arch Dis
Child 1982;57(5):364-8.

Tchernitchin AN, Tchernitchin NN, Mena MA, Unda C, Soto J. Imprinting:
perinatal exposures cause the development of diseases during the adult
age. Acta Biol Hung 1999;50(4):425-40.

Jirtle RL, Sander M, Barrett JC. Genomic imprinting and environmental
disease susceptibility. Environ Health Perspect 2000;108(3):271-8.

175

 

 

11.

12.

13.

14.

15.

Li S, Hursting SD, Davis BJ, McLachlan JA, Barrett JC. Environmental
exposure, DNA methylation, and gene regulation: lessons from
diethylstilbesterol-induced cancers. Ann N Y Acad Sci 2003;983:161-9.

Li S, Washburn KA, Moore R, Uno T, Teng C, Newbold RR, McLachlan
JA, Negishi M. Developmental exposure to diethylstilbestrol elicits

demethylation of estrogen-responsive lactoferrin gene in mouse uterus.
cancer research 1997;57(19):4356-9.

Fukuzawa NH, Ohsako S, Nagano R, Sakaue M, Baba T, Aoki Y,
Tohyama C. Effects of 3,3',4,4',5-pentachlorobiphenyl, a coplanar
polychlorinated biphenyl congener, on cultured neonatal mouse testis.
Toxicol In Vitro 2003;17(3):259-69.

Goldman D, Yawetz A. The interference of aroclor 1254 with progesterone
metabolism in guinea pig adrenal and testes microsomes. Journal of
Biochemical Toxicology 1990;5(2):99-107.

Chen CW, Hurd C, Vorojeikina DP, Arnold SF, Notides AC. Transcriptional
activation of the human estrogen receptor by DDT isomers and
metabolites in yeast and MCF-7 cells. Biochem Pharmacol
1997;53(8):1161-72.

176