University of Southern California Dissertations and Theses

Monoamine oxidase A: (1) As one of many candidate genes for preeclampsia and (2) as a candidate gene for tobacco addiction

(USC Thesis Other)

Monoamine oxidase A: (1) As one of many candidate genes for preeclampsia and (2) as a candidate gene for tobacco addiction

PDF

Download a page range

Download transcript

Copy asset link

Request this asset

Transcript (if available)

Content NOTE TO USERS
This reproduction is the best copy available.
®
UMI
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
MONOAMINE OXIDASE A:
1). AS ONE OF MANY CANDIDATE GENES FOR PREECLAMPSIA
AND
2) AS A CANDIDATE GENE FOR TOBACCO ADDICTION
by
Melissa Lee Wilson
A Dissertation Presented to the
FACULTY OF THE GRADUATE SCHOOL
UNIVERSITY OF SOUTHERN CALIFORNIA
In Partial Fulfillment of the
Requirements for the Degree
DOCTOR OF PHILOSOPHY
(EPIDEMIOLOGY)
December 2004
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
UMI Number: 3155498
INFORMATION TO USERS
The quality of this reproduction is dependent upon the quality of the copy
submitted. Broken or indistinct print, colored or poor quality illustrations and
photographs, print bleed-through, substandard margins, and improper
alignment can adversely affect reproduction.
In the unlikely event that the author did not send a complete manuscript
and there are missing pages, these will be noted. Also, if unauthorized
copyright material had to be removed, a note will indicate the deletion.
®
UMI
UMI Microform 3155498
Copyright 2005 by ProQuest Information and Learning Company.
All rights reserved. This microform edition is protected against
unauthorized copying under Title 17, United States Code.
ProQuest Information and Learning Company
300 North Zeeb Road
P.O. Box 1346
Ann Arbor, Ml 48106-1346
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
DEDICATION
To my grandfather,
Samuel Pavlovic
— A man whose continued quest for knowledge inspired me to do the
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
ACKNOWLEDGEMENTS
This work was done under the direction and supervision of Dr. Sue Ann Ingles. I am
grateful to Dr. Ingles for her insights, experience, knowledge, and her belief in my
abilities. Dr. Ingles has always been generous with her time and the many hours she
has spent working with me has profoundly impacted the quality of my work. I
consider Dr. Ingles to be not only a professional mentor, but also a good friend.
I would like to thank Dr. Mimi Yu for providing access to her incredible Singapore
cohort dataset as well as providing support and ideas that helped to further my work
on tobacco addiction. Her emphasis on understanding the context and social
practices of the population being studied has deeply influenced my work.
My undiluted appreciation also extends to the other members of my committee, Dr.
Giske Ursin, Dr. Jim Gauderman, and Dr. Jean Shih for their valuable input and
support. I have been blessed with many excellent role models!
Last, but certainly not least, I want to thank my labmates (and friends), Hui-lee
Wong, Wei Wang, and Jun Wang for teaching me virtually everything I know in the
lab and for being so ridiculously selfless with both their knowledge and their time. I
also want to thank Marina Vigen for her help in genotyping the Singapore cohort
samples.
iii
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
TABLE OF CONTENTS
Page
DEDICATION ii
ACKNOWLEDGEMENTS iii
LIST OF TABLES ix
LIST OF FIGURES xi
ABSTRACT xii
CHAPTER 1: INTRODUCTION 1
CHAPTER 2: MONOAMINE OXIDASE MINI-REVIEW 3
2.1 Background 3
2.2 Polymorphisms 6
2.3 MAO-A and Disease 9
2.4 MAO-A Resequencing 1 1
2.4.1 Methods 12
2.4.2 Results and Discussion 12
CHAPTER 3: LITERATURE REVIEW (PREECLAMPSIA) 16
3.1 Introduction 16
3.2 Methods 17
3.3 Background 18
3.4 Etiology of Preeclampsia 19
3.4.1 Immune Maladaptation 21
3.4.2 Placental Ischemia 22
3.4.3 Oxidative Stress 23
3.4.4 Preeclampsia as a Genetic Disease 24
3.5 Candidate Genes 26
3.5.1 Blood Pressure Regulation 29
3.5.1.1 Angiotensinogen 29
3.5.1.2 Renin 32
3.5.1.3 Angiotensin-Converting Enzyme 3 6
iv
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
3.5.1.4 Angiotensin II Type 1 Receptor 37
3.5.1.5 Endothelin-1 38
3.5.1.6 Estrogen Receptor « 39
3.5.2 Vascular Remodeling/Vascular Injury 40
3.5.2.1 Endothelial NO Synthase 40
3.5.2.2 Prothrombin 42
3.5.2.3 Factor V Leiden Mutation 46
3.5.2.4 5, 10-Methylenetetrahydrofolate Reductase 47
3.5.2.5 Cystathionine P-Synthase 51
3.5.2.6 P3 Integrin Glycoprotein Ilia 55
3.5.2.7 Matrix Metalloproteinase-1 56
3.5.3 Endothelial Cell Health 57
3.5.3.1 Epoxide Hydrolase 57
3.5.3.2 Lipoprotein Lipase 58
3.5.3.3 Superoxide Dismutase 60
3.5.3.4 Long-Chain 3-Hydroxyacyl-CoA
Dehydrogenase 61
3.5.3.5 Apolipoprotein E 62
3.5.4 Immune Maladaptation 63
3.5.4.1 Human Leukocyte Antigens (HLA) 63
3.5.4.2 HLA-G 64
3.5.4.3 HLA-DR 65
3.5.4.4 Other HLA 66
3.5.4.5 Tumor Necrosis Factor Alpha 66
3.5.4.6 Interleukin-ip 68
3.5.5 Placentation 70
3.5.6 Other 71
3.5.6.1 Insulin-Like Growth Factor 71
3.5.6.2 Mitochondrial DNA 72
3.6 Discussion 73
CHAPTER 4: A PILOT STUDY OF NOVEL CANDIDATE GENES
FOR PREECLAMPSIA 77
4.1 Specific Aims 78
4.2 Background and Significance 81
4.2.1 Introduction 81
4.2.2 Preeclampsia as a Genetic Disease 83
4.2.3 Population Stratification in Candidate Gene Studies 86
4.2.4 Susceptibility Genes 87
4.2.4.1 Placentation Genes 87
v
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
4.2.4.1.1 Proliferation vs. Differentiation:
HAND 1 and HASH2 89
4.2.4.1.2 Syncytiotrophoblast Formation:
GCM1 91
4.2.4.1.3 Hypoxia and Trophoblast Invasion:
HEF-loc andTGFp3 92
4.2.4.2 Angiogenesis and Endothelial Cell
Maintenance: VEGF, PGF, and sFLTl 93
4.2.4.3 Utero-Placental Blood Pressure Regulation:
MAO-A 96
4.2.5 Known Risk Factors 99
4.3 Preliminary Studies 100
4.3.1 The Investigative Team 100
4.3.1.1 Dr. Sue Ingles 100
4.3.1.2 Dr. T. Murphy Goodwin 100
4.3.1.3 Dr. Duncan Thomas 100
4.3.1.4 Dr. David Conti 101
4.3.1.5 Dr. Noah Rosenberg 101
4.3.1.6 Dr. Jean Shih 101
4.3.1.7 Melissa Wilson 102
4.3.2 Number of Potential Participants 102
4.3.3 Tracing of Eligible Cases 102
4.3.4 Buccal Cell DNA Extraction and Whole Genome
Amplification 103
4.4 Research Design and Methods 104
4.4.1 Epidemiologic Methods 104
4.4.1.1 Rationale and Epidemiologic Study Design 104
4.4.1.2 Study Population 105
4.4.1.3 Eligibility Criteria 105
4.4.1.4 Selection of Cases 106
4.4.1.5 Selection of Controls 107
4.4.1.6 HIPAA Considerations 107
4.4.1.7 Interview Protocol and Informed Consent 107
4.4.1.8 Risk Factor Questionnaire 108
4.4.1.9 Clinical Outcomes Data 108
4.4.2 Laboratory Methods 109
4.4.2.1 Biological Specimen Processing 109
4.4.2.2 DNA Extraction and Whole Genome
Amplification 109
4.4.2.3 Polymorphism Discovery (Re-Sequencing) 110
v i
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
4.4.2.4 Genotyping of Microsatellites and Other
Length Polymorphisms (GeneScan Method) 111
4.4.2.5 SNP Genotyping (TaqMan Assay) 112
4.4.2.6 Genotyping Quality Control 113
4.4.3 Data Management and Statistics 113
4.4.3.1 Database Management and Data Processing 113
4.4.3.2 Statistical Analysis and Power Calculations 114
4.4.3.2.1 Aim la 114
4.4.3.2.2 Aim lb&c 114
4.4.3.2.3 Aim 2a&b 115
4.4.3.2.4 Aim 3a 117
4.4.3.2.5 Aim 3b 117
4.4.4 Proposed Time Table 121
4.4.4.1 Year One 121
4.4.4.2 Year Two 121
4.4.5 Study Strengths and Limitations 121
4.5 Human Subjects Research 123
4.5.1 Protection of Human Subjects 123
4.5.1.1 Risks to Subjects 123
4.5.1.2 Adequacy of Protection Against Risks 124
4.5.1.3 Potential Benefits of the Proposed Research
to the Subjects and Others 125
4.5.1.4 Importance of the Knowledge to be Gained 125
4.5.2 Inclusion of Women 126
4.5.3 Inclusion of Minorities 126
4.5.4 Inclusion of Children 126
4.6 Vertebrate Animals 127
4.7 Literature Cited 127
4.8 Consortium/Contractual Arrangements 127
4.9 Consultants 127
CHAPTER 5: MONOAMINE OXIDASE A AND TOBACCO
ADDICTION 128
5.1 Background 128
5.1.1 Genetics of Tobacco Addiction 128
5.1.2 Mechanism of Tobacco Addiction 129
5.1.3 The MAO-A Connection 131
5.2 Methods 136
5.2.1 Study Population 136
5.2.2 Laboratory Methods 137
vii
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
5.2.3 Statistical Methods 138
5.3 Results 140
5.3.1 Demographics 140
5.3.2 Personal History of Illness 145
5.3.3 Age at Diagnosis of Illness 149
5.3.4 Alcoholic Beverages 153
5.3.5 Passive Smoking 156
5.3.6 Occupation 160
5.3.7 Physical Activity 162
5.3.8 MAO-A and Tobacco Addiction 167
5.4 Discussion 174
5.5 Conclusions and Future Directions 182
CHAPTER 6: SUMMARY 188
REFERENCES 191
APPENDIX A: PERMISSION TO REPRODUCE ARTICLE 224
APPENDIX B: RISK FACTOR QUESTIONNAIRE 225
APPENDIX C: REVIEWER COMMENTS 246
viii
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
LIST OF TABLES
Page
Table 2.1 Variations in the gene for Monoamine Oxidase A 7
Table 2.2 Observed MAO-A uVNTR repeats and allele frequency in the
Singapore Chinese Health Study (n=955) compared to
terminology and frequency in selected other publications 15
Table 3.1 Genes hypothesized to play a role in development of PE 27
Table 3.2 Molecular epidemiology studies of the T235 AGT polymorphism
and risk of PE 33
Table 3.3 Molecular epidemiology studies of Prothrombin G20210A
polymorphism and risk of PE 44
Table 3.4 Molecular epidemiology studies of the factor V Leiden mutation
and risk of PE 48
Table 3.5 Molecular epidemiology studies of the MTHFR C677T
polymorphism and risk of PE 52
Table 4.1 Power for detecting a main effect of genotype (or haplotype)
on PE risk 117
Table 5.1 DSM-IV criteria for substance abuse and dependence 133
Table 5.2A Baseline demographic variables by smoking status 142
Table 5.2B Baseline demographic variables by genotype 143
Table 5.2C Baseline demographic variables by smoking status and genotype 144
Table 5.3A Proportion of subjects with a personal history of illness at
follow-up by smoking status 146
Table 5.3B Proportion of subjects with a personal history of illness at
follow-up by genotype 147
ix
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Table 5.3C Proportion of subjects with a personal history of illness at
follow-up by smoking status and genotype 148
Table 5.4A Mean age at diagnosis by smoking status 150
Table 5,4B Mean age at diagnosis by genotype 151
Table 5.4C Mean age at diagnosis by smoking status and genotype 152
Table 5.5 A Use of alcoholic beverages at baseline by smoking status 153
Table 5.5B Use of alcoholic beverages at baseline by genotype 154
Table 5.5C Use of alcoholic beverages at baseline by smoking status and
genotype 155
Table 5.6A Passive smoking exposure by smoking status 157
Table 5.6B Passive smoking exposure by genotype 158
Table 5.6C Passive smoking exposure by smoking status and genotype 159
Table 5.7A Proportion of subjects in various occupational categories by
smoking status 161
Table 5.7B Proportion of subjects in various occupational categories by
genotype 161
Table 5.7C Proportion of subjects in various occupational categories by
smoking status and genotype 162
Table 5.8 A Physical activity by smoking status 164
Table 5.8B Physical activity by genotype 165
Table 5.8C Physical activity by smoking status and genotype 166
Table 5.9 Estimated odds rations and 95% confidence intervals for being a
current/current smoker by MAO-A genotype 169
Table 5.10 Active smoking among current/current smokers by genotype 169
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Table 5.11 Estimated coefficients and 95% confidence intervals for short
MAO-A genotype using linear regression models for age
at smoking initiation/ regular smoking among current/current
smokers
Table 5.12 Odds rations and 95% confidence intervals from a univariate
analysis of active smoking among current/current smokers
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
LIST OF FIGURES
Page
Figure 2.1 Gene structure of Monoamine Oxidase A 6
Figure 2.2 Sequencing results for 3.5+ repeat allele from Singapore Chinese
Health Study 14
Figure 3.1 Postulated interplay between the various etiologic factors 20
Figure 3.2 The renin-angiotensin system 29
Figure 4.1 Inferred population structure 120
Figure 5.1 Proposed mechanism for role of MAO-A in smoking initiation and
addiction 135
Figure 5.2 Age at smoking initiation by genotype (reported at baseline) 170
Figure 5.3 Number of cigarettes per day by genotype (reported at baseline) 170
Figure 5.4 Number of years smoking regularly by genotype (reported
at baseline) 171
xii
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
ABSTRACT
Monoamine oxidase A (MAO-A) is an enzyme in the mitochondrial membrane that
degrades norepinephrine, serotonin, and dopamine. The gene for MAO-A is located
on the X chromosome at Xp21-ll. It spans 90 kb and has 15 exons and 14 introns.
It is highly polymorphic with at least one functional polymorphism in the proximal
promoter region, termed the upstream variable number tandem repeat (uVNTR).
This repeat polymorphism consists of a 30 bp repeat reportedly present in 2, 3, 3.5, 4
or 5 copies.
The gene for MAO-A is a candidate gene for predisposing to preeclampsia (PE), a
condition of high blood pressure and proteinuria during pregnancy which, if left
untreated, can lead to maternal organ failure and preterm delivery. This dissertation
contains a comprehensive review of the molecular epidemiology of PE and a grant
proposal to examine the possible role of the maternal and/or fetal gene for MAO-A,
genes involved in angiogenesis (VEGF, PGF, sFLTl), and a series of placentation
genes (HAND1, HASH2, GCM1, HIP-la, and TGF(33) in the development of PE.
Additionally, MAO-A has also been examined as a candidate gene for numerous
mental disorders including substance abuse and addiction. To examine the possible
effect of MAO-A in tobacco addiction, we conducted a nested case-control study
utilizing subjects from the Singapore Chinese Health Study. We selected 490 current
xiii
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
daily smokers smoking at least 7-12 cigarettes per day and 490 nonsmokers who
reported being current daily smokers or nonsmokers at both baseline and follow-up.
After excluding genotyping failures, 486 current/current smokers and 469
never/never smokers remained for analysis.
Although MAO-A genotype did not predict smoking status it did appear to influence
the age at smoking initiation among the current/current smokers. Specifically,
subjects with the short allele were more likely to begin smoking while still a teenager
compared to subjects with the long allele (p = 0.007). There also appeared to be an
increased risk of carrying the short allele for subjects who smoked more than two
packs of cigarettes per day (ORu n a d j = 1.97, 95% Cl: 0.70, 5.54), though there were
not enough subjects in this category to reach statistical significance.
permission of the copyright owner. Further reproduction prohibited without permission.
CHAPTER 1: INTRODUCTION
In a quiet, residential neighborhood in Santa Monica, a molecular epidemiologist and
a molecular pharmacologist began a conversation that would dictate how I was to
spend the next four years of my life. What began as polite small talk at a social
event led to the disclosure that the offspring of preeclamptic pregnancies tended to
resemble MAO-A knockout mice, at least in terms of their being small for their
gestational age. In the following six chapters, I will share with you the fruits of my
labor (pun intended) and establish the potential importance of the gene for MAO-A
with respect to preeclampsia as well as its possible role in tobacco addiction and/or
smoking related behaviors.
In order to provide background on the relationship between MAO-A and disease, I
will review the relevant human biology and genetics of MAO-A in Chapter 2. I will
also detail results of resequencing an approximate 500 base pair region of the MAO-
A gene promoter.
A comprehensive review of the molecular epidemiology of preeclampsia (PE) is
presented in Chapter 3. This includes a critique of all candidate genes studies
published through the end of 2002 as well as a brief review of current etiologic
hypotheses for PE.
Chapter 4 contains a National Institute of Child Health and Development (NICHD)-
funded proposal to examine selected placentation genes, as well as MAO-A, as
candidates for predisposing to PE. Also included are questionnaires being used in
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
the ongoing pilot study and the reviewers’ comments regarding the proposal. Lastly,
evidence supporting a possible role for MAO-A in the etiology of PE is outlined in
this chapter.
In Chapter 5 ,1 will switch outcomes of interest from PE to tobacco addiction. In this
chapter, I will present my findings with respect to MAO-A genotype and variables
related to tobacco use and addiction in the Singapore Cohort. The format of Chapter
5 will be similar to that of a peer-reviewed publication and, in fact, will form the
basis of a paper yet to be submitted for publication. It should be noted that Chapter 5
will be much more comprehensive than the paper that will be derived from it,
including expanded introduction, results and discussion sections.
In the last chapter, Chapter 6 ,1 will summarize my work on MAO-A with respect to
both PE and tobacco addiction, emphasizing future directions and what was learned
from each step in the process. In this chapter, I will outline the work I will be doing
post-graduation as well as suggest possible areas of interest and future grant
applications.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
CHAPTER 2: MONOAMINE OXIDASE MINI-REVIEW
2.1 BACKGROUND
Monoamine oxidase (MAO) is an enzyme of the mitochondrial outer layer that
degrades biogenic amines (Greenawalt and Schnaitman 1970). There are two
iso forms of MAO, termed A and B, which have different substrate and inhibitor
specificities (Johnston 1968; Knoll and Magyar 1972; Fowler and Callingham 1978),
immunological properties (Denney, Fritz et al. 1982), tissue and cell distribution
(Westlund, Denney et al. 1985), promoter organization (Shih, Grimsby et al. 1993),
and are encoded by separate genes (Bach, Lan et al. 1988). MAO-A preferentially
oxidizes neurotransmitters such as norepinephrine (NE) and serotonin (5-FIT) and is
inhibited by clorgyline while MAO-B prefers phenylethylamine and benzylamine
and is inactivated by pargyline and deprenyl (Grimsby, Chen et al. 1991). Both
MAO-A and MAO-B can degrade dopamine (DA), tyramine, and tryptamine
(Grimsby, Chen et al. 1991). MAO-A activity precedes MAO-B activity during
human development, with MAO-A activity being higher in the fetal brain
(Lewinsohn, Glover et al. 1980) and MAO-B activity being higher in the adult brain
(Garrick and Murphy 1982). This review will focus on MAO-A, which is found
primarily in cholaminergic neurons (Fowler, MacGregor et al. 1987; Thorpe,
Westlund et al. 1987), but can also be found in numerous other tissues including the
placenta (Weyler and Salach 1985) and cultured skin fibroblasts (Breakefield,
Castiglione et al. 1976).
The gene for MAO-A is located on the X chromosome and has been mapped to
Xp21-pll (Ozelius, Hsu et al. 1988). The gene spans 90 kilobases (kb) and is
3
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
comprised of 15 exons and 14 introns (Grimsby, Chen et al. 1991). Most tissues
express a 5 kb mRNA transcript but placenta and the small intestine express a 2 kb
mRNA transcript, possibly resulting from an alternative polyadenylation site
(Grimsby, Lan et al. 1990). The longer mRNA transcript contains an extension of
2.2 kb in the 3' noncoding region that is contained entirely within exon 15 (Chen,
Hotamisligil et al. 1991). It appears that population variation in MAO-A activity
levels, at least in human cultured fibroblasts, are determined through variations in
non-coding, regulatory elements (Hotamisligil and Breakefield 1991). Moreover,
MAO-A levels in monozygotic twins are highly correlated and levels in individual
cell lines are stable even after multiple subcultures (Breakefield, Giller et al. 1980),
providing further evidence that activity is regulated through polymorphisms within
the regulatory elements of this gene.
There is a strong degree of conservation between human MAO-A and other
mammalian species. Specifically, the amino acid sequence identity of human and
bovine MAO-A and human and rat MAO-A are 88% and 87%, respectively (Shih,
Grimsby et al. 1993). The identity between mammalian species is even greater than
the identity between human MAO-A and MAO-B (73%). Such strong conservation
between species suggests that evolutionary pressure is at work to maintain the
physiological function of MAO.
The core promoter region of MAO-A contains two 90 bp repeats (-268 to -179 and
-178 to -89), each consisting of one 42 bp and one 48 bp repeat in tandem (Zhu,
Chen et al. 1994). Each smaller repeat contains an Spl site, for a total of four Spl
sites. These Spl sites are necessary for MAO-A promoter activity (Zhu, Grimsby et
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
al. 1992) and Spl concentration is an important factor in controlling MAO-A
expression (Zhu, Chen et al. 1994). Furthermore, it appears that sequences upstream
from the core promoter can down-regulate MAO-A gene expression (Zhu, Grimsby
et al. 1992), but no inhibitory czs-elements have been found between -200 and at
least -935 (Chen 2004). Nevertheless, polymorphisms in the promoter region of
MAO-A are likely to result in different levels of MAO-A expression (Zhu, Chen et
al. 1994).
Regulation of MAO-A activity appears to be at least partially regulated by ovarian
steroids (Girgin Sagin, Sozmen et al. 2004). For example, estrogen appears to have
a tissue-specific effect on MAO-A activity in the rat (Holschneider, Kumazawa et al.
1998). Upon administration of high dose estrogen to ovariectomized adult female
rats, Holschneider et. al. found that MAO-A activity was decreased in the
hypothalamus, amygdala, uterus, and kidney while remaining essentially unchanged
in the liver, lung, small intestines, heart, cerebral cortex, and adrenal tissue. This
finding may at least partially explain the observed differences in MAO-A activity by
gender or the observed effect modification by gender in numerous MAO-A
candidate gene association studies (Deckert, Catalano et al. 1999; Ibanez, de Castro
et al. 2000; Desautels, Turecki et al. 2002; Eley, Tahir et al. 2003). It has even been
suggested that the differential effect of MAO-A genotype on disease between males
and females might be a result of estrogen interacting with cA-acting regulatory
elements on specific MAO-A alleles (Desautels, Turecki et al. 2002).
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Figure 2.1: Gene Structure of Monoamine Oxidase A
H C_000Qg3
[ 4-2636593 ►
5 ' I -----
NH_00024(l H h + Hh
[42652050 ►
H 3'
NP_000231
- co d in g r-e-gion I - uniroinslflte-d re g io n
Source: NCBI Entrez Gene Website:
http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?db=gene&cmd=Retrieve&dopt=Graphics&list_uids=4128
2.2 POLYMORPHISMS
MAO-A activity levels across subjects vary as much as 100-fold (Tivol, Shalish et
al. 1996) Only six polymorphisms have been found throughout the coding region of
the gene (Table 2.1). Although they do not alter the amino acid sequence, two
polymorphisms, T-G at 941 and O T at 1460, result in alterations to restriction
enzyme sites (Fnu4HI and EcoRV, respectively) and are in complete linkage
disequilibrium (Hotamisligil and Breakefield 1991). Tivol and colleagues found two
other variants that did not alter the amino acid sequence (A435C, arg-arg and
A1076T, pro-*pro) and a third that resulted in a neutral amino acid substitution
(A1609Q lys-*arg) that is unlikely to alter the protein function (Tivol, Shalish et al.
1996). One paper examined the functional impact of alleles carrying the iscoRV,
FnuRl, and a third intronic restriction site polymorphism, Mspl and found
statistically significant associations with activity levels as measured in human skin
fibroblasts (Hotamisligil and Breakefield 1991). This finding suggests that these
polymorphisms are in linkage disequilibrium with some other functional variant.
6
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Table 2.1: Variations in the gene for Monoamine Oxidase A
Variant Description Location Functional? Reference
30 bp upstream VNTR Promoter Yes (Sabol, Hu et al.
1998)
(GT)n + 23 bp VNTR Intron 1 Not likely (Hinds,
Hendriks et al.
1992)
Variable Dinucleotide
(GT)n
Intron 2 Not likely (Black, Chen et
al. 1991)
435, A -C , arg-arg Exon 4 Not likely (Tivol, Shalish
et al. 1996)
936, C-T, gin-STOP Exon 8 Yes (Brunner, Nelen
et al. 1993)
941, T-G, arg-arg
(Fnu4m)
Exon 8 Not likely (Hotamisligil
and Breakefield
1991)
1076, A -T, pro-pro Exon 9 Not likely (Tivol, Shalish
et al. 1996)
1460, C-T, asp-asp
(EcoRV)
Exon 14 Not likely (Hotamisligil
and Breakefield
1991)
1609, A -Q lys-arg Exon 15 Not likely (Tivol, Shalish
et al. 1996)
Mspl RFLP
Intron Not likely (Ozelius,
Gusella et al.
1989)
Brunner and colleagues described a rare variant discovered in a Dutch family in
which several males display mild mental retardation and impulsive aggression
including arson, attempted rape and exhibitionism (Brunner, Nelen et al. 1993).
Urinary analysis of these subjects uncovered substantially altered monoamine
metabolism. Upon further investigation, it was discovered that they carried a point
mutation in exon 8 that resulted in a termination codon and subsequent complete
MAO-A deficiency.
A variable dinucleotide repeat in intron 2 has also been reported (MAOA-CA)
(Black, Chen et al. 1991). Although this repeat is intronic, the observation that it is
conserved between MAO-A and MAO-B suggests that it may possibly be
functionally important (Grimsby, Chen et al. 1992).
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
An upstream variable number tandem repeat (uVNTR) polymorphism has been
described 1.2 kb upstream of the MAO-A coding sequence consisting of a 30-bp
repeat that can reportedly be present in 2, 3, 3.5, 4 or 5 copies (Sabol, Hu et al.
1998). However, the 3-repeat and 4-repeat alleles are by far the most common.
Alleles with 3.5 or 4 repeats are more efficiently transcribed than alleles with 3 and
possibly those with 5 repeats and result in higher MAO-A levels (Sabol, Hu et al.
1998; Deckert, Catalano et al. 1999; Denney, Koch et al. 1999). It is not yet known
how the uVNTR influences functionality. Based on the functional characterization,
uVNTR alleles are often dichotomized into a short allele (<3 repeats) and a long
allele (>3 repeats). Strong linkage disequilibrium exists between the uVNTR and the
dinucleotide repeat in intron 2 (MAOA-CA), with the 3-repeat allele tending to
segregate with the shorter dinucleotide repeats of the MAOA-CA polymorphism
(Sabol, Hu et al. 1998).
A novel polymorphism was discovered by Hinds et. al. that consists of a GT
microsatellite repeat directly adjacent to an imperfectly duplicated 23 bp VNTR in
intron 1 (Hinds, Hendriks et al. 1992). Both the microsatellite and the VNTR differ
in the number of repeats, creating a highly informative polymorphic region. This
polymorphism exhibits a high degree of linkage with the uVNTR (Balciuniene,
Emilsson et al. 2002).
Comings and others have suggested that repeat polymorphisms, such as the uVNTR,
may be good candidates to explain a large portion of the variation observed in
complex, polygenic disorders since such variants often play an important role in
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
gene expression and are more commonly found in control regions than coding
regions (Comings 1998; Sabol, Hu et al. 1998).
More than forty single nucleotide polymorphisms (SNPs) have been reported on the
National Center for Biotechnology Information (NCBI) website. Most of these
polymorphisms are either intronic, and are therefore unlikely to be functional, or
located in the untranslated region. Only those that have been examined in the
context of an association study appear in Table 2.1
2.3 MAO-A AND DISEASE
MAO-A inhibition in rodents leads to accumulation of 5-HT, NE and, DA in brain
tissue (Campbell, Robinson et al. 1979) (Kim, Shih et al. 1997). Alterations in
behavior, including increased aggressiveness and increased classical fear
conditioning, have been observed in MAO-A deficient mice (Cases, Seif et al. 1995)
(Kim, Shih et al. 1997). In humans, low MAO-A activity or MAO-A inhibition has
been associated with mood elevation (Quitkin, Rifkin et al. 1979), loss of REM sleep
(Kupfer and Bowers 1972), increased impulsivity (Eensoo, Paaver et al. 2004),
sexual dysfunction (Friedman 1983; Mitchell and Popkin 1983), and hypertension
(Lavin, Mendelowitz et al. 1993). Complete loss of MAO-A function, resulting
from a point mutation in the human MAO-A gene (described in section 2.2) has been
reported in the male members of a Dutch family and results in Brunner Syndrome, a
condition marked by mild mental retardation and impulsive aggressiveness (Brunner,
Nelen et al. 1993). Complete loss of both forms of MAO have been implicated in
Norrie disease, a condition usually characterized by blindness and severe mental
retardation (Collins, Murphy et al. 1992).
9
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
A wide variety of disorders have been found to be associated with alterations in
MAO-A activity. For example, low MAO-A activity has been found to be
associated with schizophrenia (Jonsson, Norton et al. 2003), obsessive compulsive
disorder (Camarena, Cruz et al. 1998), antisocial behavior among male alcoholics
(Samochowiec, Lesch et al. 1999), bipolar affective disorder (Lim, Powell et al.
1995; Rubinsztein, Leggo et al. 1996), autism severity (Cohen, Liu et al. 2003),
alcoholism (Saito, Lachman et al. 2002), and antisocial personality disorder among
male alcoholics (Schmidt, Sander et al. 2000) and male adults who were maltreated
as children (Caspi, McClay et al. 2002). In contrast, high MAO-A activity has been
implicated in depression (Sandler, Bonham Carter et al. 1975), panic disorder
(Deckert, Catalano et al. 1999), restless leg syndrome among females (Desautels,
Turecki et al. 2002), and attention deficit hyperactivity disorder (Manor, Tyano et al.
2002). MAO-A may also be important in the development and/or progression of
Alzheimer’s Disease (Takehashi, Tanaka et al. 2002) and Parkinson’s Disease
(Nakatome, Tun et al. 1998).
MAO-A is expressed in the placenta, strongly suggesting that it may play some role
in the normal development or function of the placenta. Specifically, MAO-A
activity has been detected in cryostat sections of syncytiotrophoblasts , indicating a
role in maternal-fetal circulation (Yoshimoto, Sakumoto et al. 1986). Moreover, the
metabolism of amines by MAO-A is known to result in the subsequent generation of
oxygen free radicals, which then have the potential to create substantial oxidative
damage (Girgin Sagin, Sozmen et al. 2004). Since preeclampsia is known to be a
disease of placental dysfunction and since it has been hypothesized that preeclampsia
might be at least partially caused by oxidative stress (Roberts and Cooper 2001), the
10
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
gene for MAO-A is a good candidate for consideration in this complex, multigenic
disorder (see Chapters 3 and 4 for more on Preeclampsia).
As early as the 1960's, low MAO-A expression became a proposed risk factor in the
development of PE due to an observed deficit in placental MAO-A among women
with PE (Sandler and Coveney 1962). Subsequent studies also found decreased
MAO-A in the placentas of women with PE (Gujrati, Shanker et al. 1996;
Sivasubramaniam, Finch et al. 2002). Furthermore, circulating serotonin levels are
increased in women with PE, consistent with a decrease in MAO-A activity (Weiner
1987; Hutter and Filshie 1993; Carrasco, Cruz et al. 1998). However, it has recently
been discovered that, rather than reduced levels of the MAO-A protein or MAO-A-
specific mRNA, there was a 60% reduction in catalytic efficiency of the MAO-A
enzyme in the placentae from preeclamptic women compared to the normotensive
controls, possibly resulting from increased hypoxia and oxidative damage
(Sivasubramaniam, Finch et al. 2002). Thus, it is possible that the reduced MAO-A
activity observed in preeclamptics is a result of the disease, rather than a factor
contributing to it. Gene-association studies might be able to better clarify this issue.
2.4 MAO-A RESEQUENCING
To determine if there were any unknown common polymorphisms in an
approximately 500 bp region of the promoter (-1567 to -1070), we sequenced forty
six cases of advanced prostate cancer, including 26 Caucasians and 20 African
Americans. Additionally, two samples from the Singapore Chinese Health Study
were sequenced to confirm what appeared to be a novel allele. These two sets of
subjects will be addressed separately.
11
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
2.4.1 Methods
Phenylchloroform extraction was used to obtain DNA from buffy coat stored at
-70°C. DNA amplification was achieved by polymerase chain reaction (PCR) using
the following primers: 5'-GCGGCTACACC CACATCTAC-3' (forward) and 5'-
GTAGGAGGTGTCGTCCAAGC-3' (reverse). PCR was performed in a total
volume of 25 pi with 2 pi of DNA,, 0.2 pi of DMSO, and a final concentration of
3.0 pmol of each primer, 0.25 mM of dNTPs, 2.0 mM of MgCl2 , 1 U Taq DNA
polymerase (Promega Biosciences, Inc., San Luis Obispo, CA), lx buffer (Promega
Biosciences, Inc., San Luis Obispo, CA). PCR thermal cycling conditions included
an initial denaturation at 94°C for 5 minutes, followed by 40 cycles of denaturation
at 94°C for 30 seconds, annealing at 53°C for 30 seconds, and extension at 72°C for
30 seconds. A final extension at 72°C for 5 minutes was conducted. Sequencing was
carried out using the ABI Prism 3700 DNA Sequencer (Applied Biosystems, Foster
City, CA) and read using Sequence Analysis 3.7 software (Applied Biosystems,
Foster City, CA).
2.4.2 Results and Discussion
Previously reported alleles included 2, 3, 3.5, 4, and 5 repeats, with the 3-repeat and
the 4-repeat alleles being the most commonly observed alleles (Deckert, Catalano et
al. 1999). Interestingly, I did not observe any alleles which did not include a half
repeat. Therefore, among the 46 advanced prostate cancer subjects, the alleles
observed included one with 2.5 repeats, seventeen with 3.5 repeats, 26 with 4.5
repeats, and two for whom sequence data was not obtained. The discrepancy
between the previously reported allele sizes and the results presented above has been
alluded to in previous publications. Specifically, Sabol et. al. describe their “allele
12
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
3" as being both a 4-repeat and being “identical to the published sequence” (Sabol,
Hu et al. 1998). However, examination of the published sequence shows that it
contains 4 repeats followed by the first 15 base pairs of the repeat, and is, in fact, a
4.5 repeat. Furthermore, Eley et. al. note that “each allele ends with an additional 15
bp half-repeat section,” consistent with my observation that all sequenced alleles
contain a half-repeat (Eley, Tahir et al. 2003). However, in order to maintain
consistency across the numerous studies of the MAO-A uVNTR, I will refer to each
allele excluding the additional half-repeat. Thus, the observed alleles in the
advanced prostate cancer cases included 2, 3, and 4-repeat alleles.
Analysis by race indicates that allele frequency differs between blacks and whites.
Specifically, there was some suggestion that shorter repeats (2 and 3 repeats) were
more common among blacks than whites, but this difference was not statistically
significant, probably due to the small sample size (blacks: 56% short alleles vs.
whites: 31% short alleles, p = 0.15). These allele frequencies are very consistent
with those reported by Sabol and colleges (Sabol, Hu et al. 1998) (Table 2.2).
Using GeneScan technology, two subjects from the Singapore Chinese Health Study
(see Chapter 5) were found to have PCR product of an unusual length. These two
samples were sequenced in order to confirm their genotype. In doing so, I
discovered a rather interesting and novel allele containing two full repeats, followed
by the last 12 base pairs of the full repeat, followed by another full repeat, and
culminating in a half-repeat (Figure 2.2). This allele is very rare, having been
observed only twice out of 955 samples. Since the frequency of this allele, termed
the 3.5+ allele, is similar to that reported for the 3.5 allele (Sabol, Hu et al. 1998), it
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
seems reasonable to suspect that the previously reported 3.5 allele is actually a 3.5+
allele. Table 2.2 provides the allele frequencies obtained from the Singapore
Chinese Health Study and indicates how these alleles are referred to in selected other
publications.
Figure 2.2 Sequencing Results for 3.5+ repeat allele from Singapore Chinese Health Study
teuti'iua&fcxa ««m *
S arckM A O IF _______L ire : 1 3 B * « * p a c m g l4 i8 4 913ba»m lM 3Q «aai tolcf4
t i t i & i
r
Note: The repeat sequence begins about 30 bp into the above sequence and has the following series of
30 base pairs: ACCGGC ACCGGC ACCAGT ACCCGC ACCAGT.
14
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Table 2.2 Observed MAO-A uVNTR Repeats and Allele Frequency in the Singapore Chinese Health
Study (n = 955) Compared to Terminology and Frequency Reported in Selected other Publications
Number of
Repeats
Observed
Frequency (%) Allele Frequency (%) &
Terminology used in Sabol
et. al. (Asian
Population)(Sabol, Hu et
al. 1998)
Allele Frequency (%) &
Terminology used in
Deckert et. al. (German &
Italian Controls)(Deckert,
Catalano et al. 1999)
1.5 1 (0.1) Not Observed Not Observed
2.5 5 (0.5) Not Observed 2-repeat; 3 (1.0)
3.5 558 (58.4) allele 1; 50 (61.0) 3-repeat; 119 (38.3)
3.5+ 3 (0.3) allele 2; 1 (1.2) 3a-repeat*; 1 (0.3)
4.5 387 (40.5) allele 3; 31 (37.8) 4-repeat; 182 (58.5)
5.5 1 (0.1) allele 4; Not Observed 5-repeat; 6 (1.6)
* Allele 3a is described as 3 full repeats plus 18 bp of the repeated motif
15
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
CHAPTER 3: LITERATURE REVIEW
The following chapter contains a literature review entitled “Molecular Epidemiology
of Preeclampsia.” This literature review was undertaken to fill a tremendous gap in
the literature, namely, a comprehensive review of the work that had been done to date
examining a role for genetics in the development of preeclampsia (PE).
Although many candidate gene studies have been conducted, a consensus as to what
genes might be involved in conferring increased risk of PE remains elusive. In
writing this review, I hoped to clarify the work that had already been completed,
provide a critical review of that work, and outline areas of future research that should
be undertaken.
The final document appeared in Obstetrical & Gynecological Survey in the January,
2003 issue.
3.1 INTRODUCTION
Preeclampsia (PE) is a hypertensive disorder specific to pregnancy and is a major
cause of maternal and neonatal death and morbidity worldwide, affecting nearly 6%
of all pregnancies (Kaunitz, Hughes et al. 1985; WHO 1987). As much as 15-20% of
maternal mortality in developed countries can be attributed to PE (Pregnancy 1990).
Severity ranges from a mild disorder associated with transient hypertension in the
later part of pregnancy to a life-threatening disorder with seizures, HELLP
(hemolysis, elevated liver enzymes and low platelets) syndrome, fetal hypoxia and
growth retardation (Amgrimsson, Hayward et al. 1997). These more severe sequelae
16
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
are much less common than mild disease, affecting approximately 0.56 pregnancies
per 1000 deliveries (Saftlas, Olson et al. 1990). PE also predisposes women to other
serious pregnancy complications such as abruptio placentae, acute renal failure,
cerebral hemorrhage, disseminated intravascular coagulation and circulatory collapse
(Chesley 1984). Despite the large number of studies focusing on PE, the etiology of
this disease remains to be elucidated. Identification of genetic risk factors, or of a
population with an exceptionally high risk of disease, could aid substantially in the
understanding of this important public health problem and provide clues for the
prevention or treatment of PE.
We undertook a literature review of preeclampsia and genetics, focusing on studies
which used epidemiologic methods to conduct their research. Specifically, our aim is
to provide a review of the candidate genes examined thus far, discuss how they fit
into the prevailing etiologic hypotheses and to evaluate the quality of the evidence
that these genes affect the risk of PE. By doing so, we hope to provide a clearer
direction for future research.
3.2 METHODS
Three methods were used to select papers for this review. First, we undertook a
MEDLINE search using the keywords preeclampsia, genetics and epidemiology. We
focused on peer-reviewed papers published between January 1, 1990 and December
31, 2001, written in the English language and limited to humans. Next, we
performed a search using Web of Science (v4.3.1) using the keywords preeclampsia
and genetics and focusing the search on papers published after 1990 in the English
language. From the list generated, we selected papers that conducted their research
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
in humans and performed their studies using epidemiological methods. Lastly, we
selected papers from the bibliography sections of the papers identified above which
dealt directly with the topics of interest.
In evaluating the quality of the evidence, we focused primarily on issues of sample
size, case definitions and possible confounding and bias. We have categorized the
candidate genes into six categories, based on their hypothesized mechanism of action.
Results for each candidate gene will be summarized and, where applicable,
assessments of study quality will be included. We have only attempted to assess the
quality of the epidemiological evidence, although other evidence may be cited.
3.3 BACKGROUND
The epidemiology of PE has been complicated by the varying definitions and criteria
used to diagnose it. In order to allow for comparisons across studies, a strict,
consistent definition of the disease requiring both hypertension and proteinuria is
essential. To that end, PE is best defined as an increase in blood pressure to greater
than or equal to 140/90 mmHg or a 30/15 mmHg increase over baseline, on at least
two occasions at least 6 hours apart after 20 weeks of gestation and resolving shortly
after delivery, accompanied by significant proteinuria (at least 30 mg/dl on a random
urine sample or 300 mg in 24 hours) (Chesley 1980; Liston and Kilpatrick 1991;
Dekker, de Vries et al. 1995). Although having a history of severe or early PE
increases one's risk for developing PE in a subsequent pregnancy (Dekker and Sibai
1999), PE is considered to be primarily a disease of first pregnancies (MacGillivray
1983).
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Further complicating matters, evidence suggests that PE is likely to be a two-stage
process (Ness and Roberts 1996; Roberts 2000), requiring input from both the fetus
(via the placenta) and the mother (via underlying susceptibility). According to Ness
and Roberts, PE can result from poor placental perfusion (a component that is likely
to be hereditary) combined with an underlying maternal condition which may or may
not be diagnosed prior to pregnancy and may or may not have a heredity component.
Ideally, a study of genetic susceptibility to PE would focus on hypertension and
proteinuria that is primarily of placental origin since it is this form that holds the key
to the unique contribution of pregnancy to this disease.
3.4 ETIOLOGY OF PREECLAMPSIA
The etiology of PE is unknown. However, PE is known to be the result of
pathological changes in placental development with subsequent endothelial cell
dysfunction, which accounts for its clinical signs. To better understand what is
abnormal about preeclamptic placentation, a brief review of normal placetation is
necessary.
The placenta develops primarily from fetally derived cells known as trophoblasts.
The trophoblasts initially differentiate into two types, the cytotrophoblasts, which are
the precursors to all subsequent trophoblast cells and the synctiotrophoblasts, which
are responsible for the invasion into the decidua, and in particular, into the maternal
spiral arteries. There are two waves of trophoblastic invasion, one at the beginning
of pregnancy and another later in pregnancy, around 14 to 16 weeks gestation
(Robillard 2002). The invasion of the synctiotrophoblasts into the spiral arteries
results in a widening of these arteries to approximately 4-6 fold their width in
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
nonpregnant women, thereby increasing the blood flow available to the developing
fetus and placenta. In PE, trophoblastic invasion and subsequent remodeling of the
spiral arteries, especially during the second wave of invasion, is deficient, resulting in
spiral artery diameters that are only about 40% as wide as those in a normal
pregnancy. The result is placental ischemia and poor placental perfusion in women
who will eventually develop the clinical signs of PE.
There are four main etiologic factors believed to be involved in the development of
PE (Dekker and Sibai 1998; Roberts 2000): (1) immune maladaptation, (2) placental
ischemia, (3) oxidative stress and (4) genetic susceptibility (Dekker and Sibai 1998;
Dekker 1999; Roberts 2000). These categories are not mutually exclusive and in
reality, the etiology is likely to be a combination of the four (Figure 3.1).
Figure 3.1 Postulated interplay between the various etiologic factors
G enetic Predisposition
Immune
|jj||ladaptat
Abnormal
Placentation
1 0 1 1
Placental
g Jgc|emkL
Cytokine
Changes
>
Oxidative
Endothelial
Dysfunction
>
< PE
20
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
3.4.1. Immune Maladaptation
PE may result from an abnormal maternal immune response to paternally derived
antigens on the trophoblast. The evidence for the involvement of an abnormal
immune response in PE development comes from epidemiologic research, including
the following observations: (1) the risk of preeclampsia is decreased after the first
pregnancy, (2) the protective effect of multiparity is largely lost with a change in
partner, (3) prior abortion or blood transfusion protects against preeclampsia (4)
artificial donor insemination and oocyte donation lead to an increase in risk of PE
and (5) increased exposure to semen (e.g., length of cohabitation, use of oral
contraceptives) may be protective.
Evidence for the role of immune maladaptation in the etiology of PE also comes from
reports of immunologic phenomena occurring in women with PE. These include
antibodies against endothelial cells; increased circulating immune complexes;
complement activation; complement and immune complex deposition in spiral
arteries, placenta, liver, kidney and skin; altered TH1:TH2 profile; decreased
suppression of T-cell receptor chain CD3£; and elevated concentrations of
proinflamatory cytokines (Dekker and Sibai 1999). Thus, it is certain that women
with PE manifest immunopathology, however it is not certain whether this
immunopathology is the cause or the result of PE.
The mechanism by which immune maladaptation is related to the endothelial cell
dysfunction seen in PE is uncertain, but it is postulated that activated immune cells
from the decidua may release mediators that act on the endothelial cells. Several
possible mediators have been suggested. Specifically, plasma elastase levels have
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
been found to be elevated in PE (Dekker and Sibai 1998). Elastase and other toxic
proteases are released by activated neutrophils. Alternatively, cytokines (particularly
TNF-a and IL-1) have been suggested as the cause of the endothelial dysfunction
seen in PE since both may be elevated in plasma and/or amniotic fluid of
preeclamptics and can have endothelial effects that resemble the changes observed in
PE (Dekker and Sibai 1998).
3.4.2. Placental Ischemia
The placental ischemia hypothesis suggests that the disease process begins with the
failure of the spiral arteries to widen in response to the increased vascular demands of
pregnancy, leading to a deficient blood supply to the placenta (reviewed in (Dekker
and Sibai 1998; van Beck and Peeters 1998; Dekker 1999).
At least two theories have been put forth to link placental ischemia to endothelial cell
dysfunction. First, increased deportation of syncytiotrophoblast microvillous
membrane (STBM) particles from preeclamptic compared to normal placentas has
been demonstrated to disrupt endothelial cells and inhibit their proliferation (Chua,
Wilkins et al. 1991; Smarason, Sargent et al. 1993). Alternatively, oxidative stress
secondary to placental ischemia might lead to endothelial cell dysfunction (Roberts
2000). Reduced organ perfusion followed by the return of normal oxygenation is
known to cause the formation of reactive oxygen species such as superoxide radicals
(Roberts 2000). In concert with maternal changes in lipid metabolism and/or a
maternal predisposition such as hyperhomocysteinemia or antioxidant deficiency,
reduced placental perfusion might result in oxidative stress that the mother is not
capable of counteracting, thereby leading to endothelial cell dysfunction and PE.
22
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
3.4.3. Oxidative Stress
Pregnancy increases energy demands, a fact that is reflected by the accumulation of
very low-density lipoproteins (VLDL) throughout pregnancy (Arbogast BW 1996).
Free fatty acids (FFA) are increased in women with PE as much as 15-20 weeks
before the onset of clinical disease (Lorentzen and Henriksen 1994). Plasma
albumin, which normally exerts a toxicity-preventing activity, is less effective in this
respect when there are increased amounts of circulating FFA. Thus, triglycerides
tend to accumulate in the endothelial cells of preeclamptic women.
According to the oxidative stress hypothesis, oxidative stress generated in the
hypoxic placenta is transferred to the systemic circulation, resulting in oxidative
damage to the vascular endothelial cells throughout the body (Roberts 2000; Roberts
and Cooper 2001). Short-lived reactive oxygen species may interact with lipids to
form stable lipid peroxidation products that are potentially very damaging to cell
structures. Dislipidemia early in preeclamptic pregnancies can lead to the
accumulation of small, dense LDL in the subendothelial space which are easily
oxidized to form highly reactive oxidized LDL (Roberts and Cooper 2001). While
lipid peroxides are known to increase in normal pregnancy, this increase is usually
offset by an increase in antioxidant activity. In preeclamptic pregnancies, there is a
decrease in net antioxidant activity and thus, the potential for damage by oxygen free
radicals (Dekker and Sibai 1998).
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
3.4.4. Preeclampsia as a Genetic Disease
Evidence for a genetic component comes from the observation that there is a marked
increase in preeclampsia among mothers, daughters, sisters and granddaughters of
women who have had preeclampsia, but not in women related through marriage (i.e.,
in-laws) (Sutherland, Cooper et al. 1981; Amgrimsson, Bjomsson et al. 1990; Cooper
1993). Some studies have found that the increased risk is greatest for the daughters
of a preeclamptic pregnancy, even greater than the risk for a sister bom of a
normotensive pregnancy (Cooper, Hill et al. 1988) while others have found a similar
increase in risk for all daughters bom to a mother with a history of PE (Amgrimsson,
Bjomsson et al. 1990; Mogren, Hogberg et al. 1999).
Additionally, higher concordance rates among monozygotic twins compared to
dizygotic twins suggests a role for genetics in the development of disease (Ros 2000).
However, the observation that a high number of monozygotic twin sets are discordant
for the development of PE during their own pregnancies suggests that the fetal
genotype, as well as environmental factors, may also be important in determining
susceptibility.
There are several other lines of evidence that suggest a fetal (paternal) component to
PE susceptibility. For instance, the association between PE and fetal chromosomal
abnormalities supports a fetal contribution to etiology (Boyd, Lindenbaum et al.
1987) as does the observation that the risk of developing PE is increased in women
with complete hydatidiform moles, which are entirely of paternal origin (Goldstein
and Berkowitz 1994). In addition, the small but statistically significant increase in
incidence among daughters-in-law of index cases (Amgrimsson, Bjomsson et al.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
1990) and the observation that men bom of preeclamptic pregnancies were more
likely to father a preeclamptic pregnancy than their matched controls (Esplin, Fausett
et al. 2001) also support the idea of a fetal/patemal contribution to risk. Lastly,
multiparous women who change partners are at increased risk of PE, especially if
their new partner is known to have fathered a preeclamptic pregnancy with another
woman (Need 1975).
While the role of genetic factors in the etiology of PE is widely accepted, the mode of
inheritance is still the subject of vigorous debate. Some researchers have suggested
that susceptibility to PE could be inherited via a single (usually maternal) autosomal
recessive gene (Cooper and Liston 1979; Sutherland, Cooper et al. 1981; Chesley
1986; Amgrimsson, Bjomsson et al. 1990; Chesley 1993) or a dominant gene with
incomplete penetrance (Amgrimsson, Bjomsson et al. 1990; Chesley 1993;
Amgrimsson, Connor et al. 1994; Dekker and Sibai 1999). A role for the fetal
genotype has also been hypothesized (Cooper, Hill et al. 1988; Amgrimsson,
Bjomsson et al. 1990). It has been suggested that PE susceptibility is due to complex
interactions between two or more maternal genes, environmental factors and fetal
genotypes (Amgrimsson, Hayward et al. 1997; Mogren, Hogberg et al. 1999; Pipkin
1999; Walker 2000), a combination of maternal, fetal and paternal (via fetus) genetic
contributions (Lie 1998; Kilpatrick 1999) or maternal-fetal interactions (Liston and
Kilpatrick 1991; Kilpatrick 1999).
Intuitively, the genetic model for a condition that is restricted to pregnancy would
include components from both the mother and the fetus. Therefore, it has been
suggested that the PE phenotype is due to a maternal-fetal genotype-by-genotype
25
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
interaction either at the same locus or at separate ones. It is probable that
susceptibility to PE is due to one or more genes, acting in both the mother and her
fetus, modified by various environmental factors (Morgan, Crawshaw et al. 1999).
Five genome-wide scans have provided indirect support for the hypothesis that PE is
genetically heterogeneous and is not inherited in a simple Mendelian manner
(Hayward, Livingstone et al. 1992; Harrison, Humphrey et al. 1997; Amgrimsson,
Sigurard ttir et al. 1999; Moses, Lade et al. 2000; Lachmeijer, Amgrimsson et al.
2001). Only one scan has identified a locus (on chromosome 2p) meeting the criteria
for genome-wide statistical significance, however, this result was generated largely
by two Icelandic families (Amgrimsson, Bjomsson et al. 1990) and has not been
confirmed in other populations. Although several other loci have been identified that
are suggestive of linkage, little overlap has been observed from study to study. Such
inconsistency is not surprising and is even to be expected when conducting genome-
wide scans for complex traits.
Since it is likely that no one major gene determines PE risk and since most genetic
studies of preeclampsia to date have concentrated on the maternal gentotype alone
(Walker 2000; Esplin, Fausett et al. 2001), more research needs to be conducted
examining both the maternal and fetal genotypes and possible matemal-fetal
genotypic interactions at multiple loci.
3.5 CANDIDATE GENES
It appears likely that no one gene can account for all of the genetic risk in all women.
More likely, polymorphisms in a number of genes can affect PE risk. The specific
gene(s) involved may depend, at least in part, on the characteristics of the population
26
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
being studied (e.g., ethnicity, severity of the disease, maternal age or gestational age
at onset). As is illustrated in Figure 3.1, genetic predisposition could be involved in
any aspect of PE etiology: immune maladaptation, placental ischemia or oxidative
stress. Genes involved in blood pressure regulation, placentation and vascular
remodeling/injury may be involved in causing placental ischemia while genes in the
endothelial cell health category may lead to oxidative stress. A summary of
candidate genes, their proposed mechanism of action and previously published allele
frequencies appears in Table 3.1.
Table 3.1: Genes hypothesized to play a role in development of PE (continued on
next page)
Gene Maternal/
Fetal
High Risk
Genotype
Allele
Frequency (%)
Proposed Mechanism
for Increased Risk of PE
Selected
References
AGT Maternal T235 35-73% Elevated AGT levels lead to
atherotic changes and abnormal
spiral artery remodeling
Morgan, 1995
Morgan, 1997
Renin Both Increased renin activity leads to
increased angiotensin II and
placental vascular contraction
Amgrimsson,
1994
Shah, 2000
ACE Both Deletion 19-29% Increased ACE activity leads to
increased angiotensin II and
placental vascular contraction
Goldkrand, 1986
Li, 1992
Kalenga, 1996
Morgan, 1999
ATI Both T573
G1062
Cl 166
A4 Repeat
47%
11%
27%
46%
Reduced placental expression may
lead to impaired prostaglandin
secretion, inadequate dilation and
ischemia or, low receptor expression
may lead to low trophoblast
responsiveness to angiotensin II and
impaired placentation
Morgan, 1998
ENOS Both T 9-13% May be linked to a polymorphism
which causes reduced eNOS activity
and a lack of vasodilation
Lyall, 1999
Yoshimura, 2000
Prothrombin Both A20210 2.5-4.1% Increased concentrations of
prothrombin leads to increased risk
of thrombosis
Kupferminc,
1999
Higgins, 2000
FVL Both FVL+ 0 .6 -7 % Predisposes to thrombophilic events
and could lead to placental
infarctions
Rees, 1995
Mello, 1999
MTHFR Maternal T 15.2% Reduced MTHFR activity increases
levels of homocysteine, which then
causes vascular injury
Dekker, 1995
Levine, 2000
CBS Maternal? 844ins68 13.5-37.7% CBS deficiency may increase
plasma homocysteine and lead to
vascular injury
Tsai, 1996
Franco, 1998
Kim, 2001
27
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Table 3.1 Genes hypothesized to play a role in development of PE (continued)
Gene M aternal/
Fetal
High Risk
Genotype
Allele
Frequency (%)
Proposed Mechanism
for Increased Risk of PE
Selected
References
LPL Both? S291
N9/-
93G
S447
1.5%
0.7%
9.4%
Reductions in LPL activity lead to
dyslipidemia and endothelial cell
dysfunction
Fisher, 1997
Hubei,
1999
SOD Decreased activity leads to increased
superoxides and oxidative stress
Chen, 1994
Wang,
2001
LCHAD Both Impaired oxidation of long-chain
fatty acids leads to triglyceride
accumulation and endothelial cell
injury
Ijlst, 1996
Den Boer, 2000
HLA-G Both T107
A110
Deletion 62%
Impaired immune tolerance leads to
rejection of baby by mother or vice
versa. May also be important in
adequate trophoblast invasion
Yelavarthi, 1991
Kilpatrick, 1999
Bermingha
m, 2000
HLA-DRb Both Impaired immune tolerance leads to
rejection of baby by mother or vice
versa
Wilton, 1990
Kilpatrick,
1999
HLA-DR4 Both DR4
Present?
15-28% Impaired immune tolerance leads to
rejection of baby by mother or vice
versa
Kilpatrick, 1990
Kilpatrick, 1999
TNF-a Maternal TNFA-2 16-23% Higher TNF-a secretion may lead to
higher levels of thromboxane,
causing vasoconstriction. TNF-a is
important in immune function and
may also alter endothelial cell
function via its ability to generate
reactive oxygen species.
Bouma, 1996
Couchane,
1997
Dizon-
Townson,
1998
IGF-II Both Overexpression may have negative
effects on intrauterine fetal growth
Guidice, 1997
Irwin, 1999
Bermingha
m ,2000
Mitochon
drial DNA
Maternal Defective mitochondria result in
decreased energy supply and
inadequate trophoblast invasion
Torbergsen, 1989
Furui, 1994
Folgero,
1996
MAO-A Both Short (£3
repeats)
36% Reduced levels of MAO-A and
increased levels of serotonin lead to
vasoconstriction
Gujrati, 1985
Weiner,
1987
Carrasco,
2000
28
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
3.5.1 Blood Pressure Regulation:
Genes that encode aspects of the renin-angiotensin system (RAS) (Figure 3.2) appear
to be good candidates for involvement in PE etiology due to their role in regulating
blood pressure, body-fluid volume and vascular remodeling during pregnancy. Since
the placenta has no autonomic nerve supply, it depends on humoral factors, such as
angiotensin II to regulate vascular resistance. Briefly, renin, an enzyme present in
the uteroplacental unit, converts angiotensinogen (produced in the liver) into
angiotensin I. Next, angiotensin converting enzyme (ACE) converts angiotensin I to
angiotensin II, a powerful vasoconstrictor. Increased sensitivity to angiotensin II as
well as increased angiotensin II type I receptor (ATI) expression in placental tissue
could lead to placental ischemia and play a part in the development of PE. It is also
possible, however, that the effects of the RAS could be a secondary response to some
primary maternal or fetal stimulus.
Figure 3.2 The renin-angiotensin system
I . AngioLensin
\ ■ C o n v e r t i n g F n z ' Converting In/\mc
giolcnsin I iotcnsinogen Angiotensin II
\n gioien sin II
Receptors
Y
Vasoconstrictio-
29
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
3.5.1.1 Angiotensinogen (AGT): One coding region polymorphism, resulting in a
substitution of threonine for methionine at amino acid 235, has been extensively
studied (Table 3.2). The frequency of the variant T235 allele varies greatly by
ethnicity, occurring in approximately 40% of whites, 71% of Hispanics, 75% of
Asians, 75% of African Americans and >90% of Africans (Nakajima, Jorde et al.
2002). In addition, polymorphisms have been identified in the 5’and 3’ flanking
regions. A common variant, G(-6)A, in the proximal promoter region has been
reported to affect the basal transcription rate and the A(-6) allele is in complete
linkage disequilibrium with the T235 variant (Morgan, Craven et al. 1997). Two
other polymorphisms at 20 bp and 18 bp upstream of the initiation site are also in
linkage disequilibrium with M235T. All of these polymorphisms have been
associated with essential hypertension but since they are linked, it is difficult to
determine which might be functionally important. Although a coding region variant
such as the T235 variant would not be expected to increase AGT levels on its own,
linkage with a promoter mutation such as G(-6)A could explain observed associations
between increased AGT levels and the T235 polymorphism. Finally, there is also a
highly polymorphic CA-repeat site in the 3’ flanking region of the AGT gene. Thus
far, eleven alleles have been identified and are designated A l-A ll, with the most
common allele being A7 (Morgan, Crawshaw et al. 1999).
Evidence for a role for AGT variants in the development of PE comes from several
sources. In a study of decidual spiral artery cross-sections from normal pregnancies
terminated at 8 weeks’ gestation, women homozygous for the T235 allele had a
statistically smaller external diameter and greater area-to-diameter ratio than woman
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
homozygous for the M235 allele (Morgan, Craven et al. 1999). This strongly
suggests that T235 homozygotes have abnormal spiral artery remodeling, a hallmark
sign of PE. Additionally, T235 expression in heterozygous women was highly
statistically significantly elevated in decidual spiral arteries relative to the M235
allele. This observation lends support to the hypothesis that a promoter
polymorphism in linkage disequilibrium with the T235 variant is in fact responsible
for the observed elevation of AGT levels seen in women with the T235 variant.
Linkage analyses using the CA-repeat in the 3’ flanking region have shown mixed
results. One study reported no evidence of cosegregation between PE and the repeat
(Wilton 1995), while another found evidence of linkage (Amgrimsson, Purandare et
al. 1993).
Epidemiological evidence is also mixed, varying substantially by population studied.
Several studies in Caucasian (Ward, Hata et al. 1993) and Japanese women (Ward,
Hata et al. 1993; Kobashi 1995; Kobashi, Hata et al. 1999; Kobashi, Shido et al.
2001) have found a statistically significant increase in the frequency of the T235
variant among women with PE compared to normotensive controls. However, the
two studies conducted by Kobashi and colleagues appear to include some of the same
subjects. In particular, the more recent study (Kobashi, Shido et al. 2001) included a
more restrictive study population and as such, is the more rigorous of the two.
Additionally, Morgan and colleagues found that the A9 allele of the C A repeat is
transmitted from the mother to the fetus significantly more frequently than would be
expected by chance, suggesting that matemal-fetal allele sharing might be involved
in PE (Morgan, Crawshaw et al. 1999). Of the studies reporting a positive
31
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
association between PE and AGT genotype, two have either failed to provide details
of the PE definition used or they included women with nonproteinuric hypertension,
suggesting that the observed association could be due to the known effect of AGT
genotype on chronic hypertension. The association between AGT genotype and PE
was not supported by several other studies in white (Morgan, Baker et al. 1995; Guo,
Wilton et al. 1997; Morgan, Crawshaw et al. 1999), Japanese (Suzuki 1999), Chinese
(Guo, Wilton et al. 1997; Yamada 1997), and Hispanic (Bashford, Hefler et al. 2001)
populations.
The lack of conclusive findings can be attributed to a number of common problems
encountered in the study of PE. For example, various PE definitions are used,
making it difficult to compare results across studies. Lack of adjustment for parity is
another possible explanation for inconsistent results. Lastly, differences in exclusion
criteria can result in studies that are not comparable since the study populations can
be quite different. While some researchers do not appear to have excluded any
subjects a priori (Morgan, Baker et al. 1995; Guo, Wilton et al. 1997; Morgan,
Crawshaw et al. 1999), others have quite an extensive list of exclusion criteria (Ward,
Hata et al. 1993) (Table 3.2).
3.5.1.2 Renin Gene. Renin is expressed in the first trimester decidua (Morgan,
Craven et al. 1998) as well as in the chorionic villi (Hagemann, Nielsen et al. 1994),
suggesting that it may be important in pregnancy development. Renin gene
expression in the decidua vera (DV) but not in the decidua basalis (DB) or chorionic
villi (CV) portions of the placenta has been observed to be higher among
32
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Reproduced w ith permission o f th e copyright owner. Further reproduction prohibited without permission.
Table 3.2 Molecular epidemiology studies of the T235 AGT polymorphism and risk of PE (next 3 pages)
Reference &
Country of Study
Type of
Cases
Type of
Controls
Parity PE Definition Covariates
(Adjustments)
Exclusions Cases/
Controls
T235 Allele
Frequency
% (Cases/
Controls)
P
Ward, Hata et al.
1993
(Utah & Japan)
Incident cases
of any type
PIH1
Normo-
tensive
pregnant
controls
Multiparous
cases &
controls
Not specified None, but
stratified on
parity
Cadaveric renal transplant
recipient, multiple pregnancies,
delivery at < 20 weeks gestation
(Utah cases); premature delivery,
term infant < 2500 g, diastolic BP
> 85, systolic BP > 140, chronic
HT, active renal disease (Utah
controls)
Multiple pregnancies, history of
HT, diabetic nephropathy, renal
disease (Japanese cases & controls)
149/571
(Utah)
41/80
(Japan)
NOTE:
Analysis
restricted to
Utah Whites
& Japanese
cases with
similar PE
definition
50.0/41.0
(Utah)
90.0/71.0
(Japan)
<0.05
0.0006
Morgan, Baker et
al. 1995
(UK)
Incident PE Normo-
tensive
pregnant
controls
Not
specified
“Strict criteria”
not specified,
but must have
been
normotensive
before week 20
and by 6 weeks
postpartum
None None specified 15/15 43.0/47.0 >0.05
Guo, Wilton et al.
1997
(Australia &
China)
Severe PE or
E cases
“Normal
maternal
controls”
Not
specified
BP > 140/90
mmHg,
proteinuria (>
0.3 g/L/24 h) &
generalized
edema after 20
weeks
gestation
None None specified 106/81
(Australian
Whites)
72/48
(Chinese)
47.0/38.0
(Australia)
78.0/75.0
(Chinese)
0.08
0.53
u >
Reproduced w ith permission o f th e copyright owner. Further reproduction prohibited without permission.
Table 3.2 Molecular epidemiology studies of the T235 AGT polymorphism and risk of PE (continued)
TJpeoT
Cases
^Tases^
Controls
Reference &
Country of Study
Type of
Controls
Parity PE Definition Covariates
(Adjustments)
Exclusions T235 Allele
Frequency
(Cases/
Controls)
Suzuki 1999
(Japan)
Incident cases Normo-
ofPIH tensive
remainder
o f cohort
Multiparous BP > 140/90 None History of HT, HT prior to
cases & mmHg week 20, diabetic nephropathy,
controls elevated over renal disease & proteinuria in
two visits & early pregnancy, multiple
proteinuria (> pregnancies, DNA sample
0.3 g/L/24 h) unavailable, subject delivered
after week 20, at other hospital, incomplete
& normo medical records, hypertensive
tensive with no
proteinuria at 6
weeks
postpartum
Note: If no
significant
proteinuria,
subjects were
classified as
GH
6 weeks postpartum
313
subjects; 33
developed
PIH
72.2 (PE
only) /80.4
"43784
(Maternal)
41/81
(Fetal)
>0.05
Morgan,
Crawshaw et al.
1999
(UK)
PE cases
(White)
Normo-
tensive
pregnant
controls
(White)
Multiparous
cases &
controls
BP > 140/90
mmHg in
previously
normotensive
woman,
proteinuria (>
0.3 g/L/24 h or
++ dipstick) &
normotensive 3
mo. pospartum
None None specified Maternal:
48.0/48.0
Fetal:
48.0/40.0
>0.05
>0.05
Kobashi, Hata et Cases of PE, Normo Multiparous > 30/15 mmHg None, but stratified Multiple pregnancies, chronic 115/381 Primigravid < 0.001
al. 1999 eclamp-sia or tensive cases & increase in BP by age, gravidity & HT, HELLP syndrome, renal PE/E vs.
(Japan) transient HT pregnant
controls
controls over average
before 20
weeks or BP a
140/90 after
week 20
(& proteinuria
30 mg/dL or +
dipstick for
PE/E)
diagnosis (PE/E vs.
transient HT)
diseases, diabetes, amniotic
volume abnormalities, fetal
abnormalities, HT or
proteinuria prior to week 20 or
4 weeks postpartum
primigravid
controls:
93.0/77.0
u >
■ f c .
Reproduced w ith permission o f th e copyright owner. Further reproduction prohibited without permission.
Table 3.2 Molecular epidemiology studies of the T235 AGT polymorphism and risk of PE (continued)
Reference Type of
Cases
!Vpe of
Controls
Parity PE Definition Covariates
(Adjustments)
Exclusions Cases/
Controls
T235 Allele
Frequency
(Cases/
Controls)
P
Kobashi, Shido et
al. 2000
(Japan)
NOTE: May
include some of
the same subjects
as Kobashi, Hata
etal. 1999
Primi-parous
PE cases
Normo
tensive
pregnant
controls
Primiparous
cases &
controls
> 30/15 mmHg
increase in BP
over average
before 20
weeks or BP >
140/90 after
week 20 &
proteinuria (>
30 mg/dL or +
dipstick)
Prepregnancy BM1
> 24, mentally
stressful condition
during pregnancy,
salty dishes
preferred during
pregnancy
Multiple pregnancies, renal
disease, diabetes, amniotic
volume abnormalities,
preexisting HT, fetal
abnormalities, HT or
proteinuria prior to week 20 or
4 weeks postpartum
58/164 Unable to
calculate
from data
given
OR = 2.5 for
homozygous
T235/T235
genotype
<0.05
Bashford, Hefler
et al. 2000
(US Mexican/
Central American
Hispanic)
PE cases Women
with at
least 2
prior
normo
tensive
term
pregnan
cies
Multiparous
cases &
controls
> 30/15 mmHg
increase in BP
over average
before 20
weeks or BP >
140/90 &
proteinuria (>
300 mg)
None specified (for
association between
disease status &
genotype)
For cases: HT & proteinuria
before 20 weeks gestation or
other significant medical
conditions that can cause
hypertension & proteinuria
For Controls: essential HT,
chronic renal disease, diabetes,
platelet disorders or
autoimmune conditions
68/50 72.0/70.0 0.84
preeclamptics than controls (Shah, Banu et al. 2000). Additionally, the presence of
renin mRNA in the fetal side of the placenta indicates that the fetal genotype may be
important in determining risk of PE.
Though several restriction length polymorphisms are now known (Bgll, TaqI and
Hinfl) (Ballantine, Klemm et al. 1994), no one has yet examined them in an
association study. An earlier linkage study found no evidence of linkage or distorted
allele distribution using a diallelic genomic probe containing exon 1 (Amgrimsson,
Geirsson et al. 1994). However, several limitations of this study may have
contributed to the lack positive findings. First, the study did not require proteinuria
as part of the PE definition and may have included some women with conditions
other than PE (e.g., essential or gestational hypertension), creating a heterogeneous
population. Additionally, this study was likely to have low power due to the limited
sample size and the use of a diallelic probe.
3.5.1.3 Angiotensin-converting enzyme (ACE). An insertion-deletion polymorphism
in intron 16 is associated with changes in ACE activity. Subjects homozygous for the
deletion (DD) have the highest ACE levels and subjects homozygous for the insertion
(II) have the lowest levels (Rigat, Hubert et al. 1990). The frequency of this
polymorphism varies significantly by ethnic group with 29% of African Americans,
19% of Indians (Indian or Pakistani) and 29% of Whites being homozygous for the
deletion (Mathew, Basheeruddin et al. 2001).
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
There is some evidence that serum ACE activity may be higher in women with
pregnancy-induced hypertension (PIH) (a hypertensive disorder of pregnancy related
to PE), with and without proteinuria, compared to normotensive pregnancies
(Goldkrand and Fuentes 1986; Li, Hu et al. 1992). However, there was no evidence
for an association between the insertion-deletion polymorphism and PE in either
maternal or fetal samples from a British population (Morgan, Foster et al. 1999) or in
maternal samples from a Finnish population (Heiskanen, Pirskanen et al. 2001). Both
studies included nulliparous and multiparous women. Additionally, both studies
lacked the statistical power to detect the relatively small difference between PE cases
and controls that would be expected given our understanding of genetic risk factors.
3.5.1.4 Angiotensin II type 1 receptor gene (ATI). ATI receptors, which are found in
the resistance vessels of full-term placentas and are most probably encoded by the
fetal gene (Morgan, Crawshaw et al. 1998), may contribute to fetoplacental blood
flow regulation and may impact placental perfusion (Knock, Sullivan et al. 1994;
Bird, Zheng et al. 1997; Morgan, Crawshaw et al. 1998). Upregulation of placental
ATI receptor activity has been found among women with PE compared to their
normotensive counterparts (Leung, Tsai et al. 2001). Since ATI receptors coupled
with angiotensin II are strong vasoconstrictors, it seems reasonable that upregulation
of these receptors could be involved in the pathophysiology of PE. In contrast,
Morgan and colleagues have suggested that a reduction in ATI receptors or receptor
activity could lead to decreased trophoblast responsiveness to angiotensin II as well
as reduced prostaglandin secretion, inadequate vasodilation, placental ischemia and
PE (Morgan, Crawshaw et al. 1998).
37
permission of the copyright owner. Further reproduction prohibited without permission.
Morgan et al. examined two diallelic nonfunctional polymorphisms in the coding
region (C573T & A1062G), one in the 3' untranslated region (A1166C) and one
dinucleotide repeat in the 3' flanking region (Morgan, Crawshaw et al. 1998). Allele
frequencies for the “at-risk” alleles were 47% for T573, 11% for 1062G, 27% for
1166C and 27% for the A4 repeat polymorphism in a sample of healthy British
women (Morgan, Crawshaw et al. 1998). Although Morgan and colleagues failed to
find any differences in fetal or maternal ATI receptor allele or genotype frequencies
between PE cases and controls, they did find statistically significant distortions in
maternal-fetal transmission for both the dinucleotide repeat (A4 allele) and the
C573T (T573) polymorphisms, suggesting that these nonfunctional polymorphisms
may be in linkage disequilibrium with a functional site (Morgan, Crawshaw et al.
1998). However, 11% of the controls and nearly 20% of the cases were multigravid.
The inclusion of women with previous pregnancies is significant since multigravidas
with previous PE may have a different mechanism of disease than primiparous
women. Thus, any association found between the ATI receptor polymorphism and
PE, defined to include multigravid women, could in fact be an association with an
underlying maternal condition rather than PE itself.
3.5.1.5 Endothelin-1 Gene (ET-1): ET-1 is a vasoconstrictor expressed in the
endothelium and smooth muscle. ET-1 levels have been found to be increased among
women with PE (Barden, Herbison et al. 2001), but it is uncertain whether the
increase occurs before or after development of disease.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
A substitution of asparagine for lysine at codon 198 (Lysl98Asn) in the ET-1 gene
has been found to be associated with higher resting blood pressures among
overweight people (Tiret, Poirier et al. 1999). Barden et al. found no association
between the polymorphism and PE (Barden, Herbison et al. 2001). However, in both
the normotensive and preeclamptic groups, women with at least one variant (Asnl98)
allele had increased systolic blood pressure compared to the wild type homozygotes
(Lysl98/Lysl98). Additionally, pregnant women homozygous for the variant allele
(Asnl98/Asnl98) had increased ET-1 levels. Thus, although the Lysl98Asn
polymorphism appears to be associated with both ET-1 levels and systolic blood
pressure, it does not appear to directly affect PE risk. However, more research is
needed to clarify if fetal genotype, in combination with maternal genotype, can
impact the risk of PE.
3.5.1.6 Estrogen Receptor a (ERa): ERa can regulate vascular tone and structure
through its ability to induce gene expression of vasoactive substances in vascular
tissues (Malamitsi-Puchner, Tziotis et al. 2001). Polymorphisms in exons 1 and 2
encoding the NH2 -terminal portion of ERa have been associated with various
pathologies, including essential hypertension (Lehrer, Rabin et al. 1993). Thus, it has
been hypothesized that variation in the NH2 -terminal portion could be related to PE
risk.
Malamitsi-Puchner and colleagues are the only group thus far to address the possible
relationship between PE and the ERa gene (Malamitsi-Puchner, Tziotis et al. 2001).
In a very small sample of preeclamptic women (n=16) and pregnant controls (n=20),
39
permission of the copyright owner. Further reproduction prohibited without permission.
they found two polymorphisms, a T— »C substitution in codon 10 and a G->C
substitution in codon 87. Neither of these polymorphisms alters the amino acid
sequence and neither was associated with PE. However, the extremely small sample
size may have precluded a statistically significant finding.
3.5.2 Vascular Remodeling/Vascular Injury
Genes in this category affect aspects of normal vascular remodeling during pregnancy
or cause vascular injury by increasing levels of homocysteine or by increasing the
risk of a thrombotic event. Inadequate vascular remodeling or vascular injury would
likely lead to placental ischemia, and possibly to an increase in PE risk. Both fetal
and maternal thrombosis may predispose to placental thrombosis, and thus, the
putative genes in this category may be of either fetal or maternal origin.
3.5.2.1 Endothelial NO synthase (eNOS): Nitric oxide (NO) increases in the placental
tissue during normal pregnancy, contributing to vasodilation, vascular remodeling
and inhibition of platelet aggregation (Forstermann 1994; Sladek, Magness et al.
1997). However, NOS activity has been shown to be decreased in the placentas of
women with PE (Brennecke, Gude et al. 1997). Since the syncytiotrophoblasts (fetal)
and the endothelium of blood vessels within the myometrium (maternal) both express
eNOS, fetal and maternal eNOS genotypes might both be important in determining
nitric oxide synthase activity (Lyall, Bulmer et al. 1999).
A G -> T transversion in exon 7 results in the replacement of glutamic acid with
aspartic acid at codon 298; however, no functional change in eNOS activity is yet
40
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
known to result from this polymorphism (Yoshimura, Yoshimura et al. 2000). The
variant allele (T) has been found in 9-13% of the general population and has been
shown to be associated with spastic angina (Yoshimura, Yasue et al. 1998),
myocardial infarction (Yoshimura, Yasue et al. 1998) and essential hypertension
(Miyamoto 1998). In a sample of normal pregnant women, Savvidou et al. found that
maternal flow-mediated vasodilation (FMD), was 21% lower among homozygotes for
the variant allele (TT) than in women homozygous for the wild type (GG) allele
(Savvidou, Vallance et al. 2001). Heterozygotes showed intermediate FMD. These
results suggest that carriers of the exon 7 variant have diminished NO bioactivity,
possibly resulting in a lowered threshold to PE development.
The results of linkage analyses have been contradictory and inconclusive. While
linkage analyses in Scotland, Iceland and Australia have suggested that the region of
chromosome 7q36 (that contains the eNOS gene) may contain a PE susceptibility
locus (Amgrimsson, Hayward et al. 1997; Guo, Lade et al. 1999). However, strong
evidence against linkage to markers in and near the eNOS gene itself was found in
this same Australian population (Lade, Moses et al. 1999). Nevertheless, it should be
noted that negative linkage results do not necessarily exclude the possibility that
eNOS is a susceptibility gene for PE. First, the microsatellite marker (CA-repeat),
for which strong evidence against linkage was found, is intronic and may not be in
linkage disequilibrium with potentially causal variants. Second, linkage analyses
may not be able to identify the involvement of the eNOS gene due to the complex
(multigenic/multifactorial) etiology of PE.
41
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
In association studies, the exon 7 variant has been found to be associated with PE. In
a Japanese population the variant allele (genotypes TG + TT) was found more
frequently among cases than controls (Yoshimura, Yoshimura et al. 2000). A second
study of PIH reported similar results (Kobashi, Yamada et al. 2001), however, this
finding failed to reach statistical significance when restricted to primiparous women
meeting the more rigorous definition of PE, possibly due to the reduction in sample
size. Results using intronic markers are more mixed. An eNOS intron 13 CA-repeat
was not associated with PE in either Chinese or Australian subjects (Guo, Lade et al.
1999). A 27 base pair repeat polymorphism in intron 4, previously associated serum
NO levels and blood pressure among nonpregnant women, was found to be
associated with PE among Hispanic women (Bashford, Hefler et al. 2001). Further
research is needed to clarify the role, if any, of the eNOS gene in the development of
PE.
3.5.2.2 Prothrombin Gene. Placental infarctions and intervillous thrombosis are
common events among women with PE, suggesting that genes predisposing to
thrombophilias are also candidates for susceptibility to PE. Specifically, in the
prothrombin gene, a G— >A mutation at nucleotide 20210 in the 30 untranslated region
is associated with higher plasma concentrations of prothrombin and leads to increased
risk of venous thromboembolism, myocardial infarction and cerebral-vein thrombosis
(Kupferminc, Eldor et al. 1999). Heterozygous carrier rates among Caucasian
controls range from 2.5% to 6%.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
The epidemiologic data on whether or not this mutation is associated with PE is
mixed (Table 3.3). One study in Italy found a statistically significant association
between PE and the prothrombin gene mutation (Grandone, Margaglione et al. 1999):
11.4% of PE cases but only 4.1% of controls carried the mutation. This difference
remained statistically significant even after adjustment for maternal age, parity, FVL
and MTHFR genotypes. Similarly, investigators in Israel found that the prothrombin
mutation was statistically significantly associated with obstetric complications,
including severe PE, but they did not have the power to detect an association with PE
alone (Kupferminc, Eldor et al. 1999; Kupferminc, Peri et al. 2000). In a later study
these same investigators reported a statistically significant association with severe
PE, but only after adjusting for other covariates (Kupferminc, Fait et al. 2000).
In contrast, PE has not been found to be associated with the prothrombin mutation
among Australian and New Zealand women (Higgins, Kaiser et al. 2000), women in
the Netherlands (De Groot 1999), white and African-American women and their
fetuses (Livingston, Barton et al. 2001) or among a separate group of white northern
European women (O'Shaughnessy, Fu et al. 2001). However, the study by Higgins et
al was powered to find a 3-fold increase in risk and any effect less than that might
easily go undetected. Additionally, more than half of the participants in the study by
Livingston et al. were of African-American descent. Since the prothrombin mutation
is virtually non-existent in non-white populations (O'Shaughnessy, Fu et al. 2001),
any effect of this mutation would necessarily be restricted to whites. While the
authors might have simply stratified by ethnicity, this would result in very small
numbers of subjects in each ethnic group, thereby limiting the power to detect an
association.
43
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Reproduced w ith permission o f th e copyright owner. Further reproduction prohibited without permission.
Table 3.3 Molecular epidemiology studies of the Prothrombin G20210A polymorphism and risk of PE (next 2 pages)
Reference &
Country of Study
Type of
Cases
Type of
Controls
Parity PE Definiton Covariates
(Adjustments)
Exclusions Cases/
Controls
20210A Allele
Frequency %
(Cases/ Controls)
P
De Groot 1999
(Netherlands)
PE history Age &
delivery date
matched;
normo
tensive
pregnancy
history
Primiparous
cases &
controls
a 30/1 i mmHg
rise in BP or
BP a 140/90
mmHg &
proteinuria
(>30 mg/dL)
Family itx. of
thrombosis or
HT, body mass,
smoking,
coagulation
mutations
(FVL+)
Multiple pregnancies,
chronic HT, renal disease,
diabetes, collagen vascular
diseases, cancer, thrombosis
history
lt>3/l63 3.1/3.7 >0.05
Grandone,
Margaglione et al.
1999
(Italy)
NOTE: May
include some of
the same subjects
as Grandone,
Margaglione et al.
1997
Incident GH
(with &
without
proteinuria)
Normo
tensive
gravid
women
matched on
ethnicity
Multiparous
cases &
controls
BP a 140/90
mmHg on 2
occasions a 4 h
apart & with
(n = 70) or
without
(n = 70) pro
teinuria (> 300
mg/24 h)
Maternal age,
parity, FVL
carrier status,
677T MTHFR
homozygosity
Transient HT, diabetes,
autoimmune disease,
chronic renal or pulmonary
disease & chronic essential
HT prior to pregnancy
140/216 11.4/4.2 <0.05
Kupferminc,
Eldoretal. 1999
(Israel)
Incident
severe PE,
AP, FGR or
stillbirth
Age &
geograph
ically
matched
normo
tensive
pregnancy
Multiparous
cases &
controls
Severe PE: BP
> 160/110
mmHg,
proteinuria (>
500 mg/24 h),
HELUPor
eclampsia
None, but
stratified by
diagnosis
Varied by diagnosis - none
specified for severe PE
110/110 For severe PE
(n=34):
5.9/3.0 >0.05
Higgins, Kaiser et
al. 2000
(Australia & New
Zealand)
Cases of
eclampsia or
PE
(84/189 of
cases
studied were
from
multicase
families)
Parous
women with
normo
tensive
pregnancy
history
Not
specified
a 25/15 mmHg
rise in BP from
baseline or BP
a 140/90
mmHg on at
least 2
occasions 6 h
apart &
proteinuria (>
0.3 g/1/24 h)
None History of chronic HT,
diabetes or renal disease
189/119 3.6 (combined
cases)
/2.5
0.73
Kupferminc, Fait
et al. 2000
(Israel)
Incident
severe PE
Age &
ethnicity
matched
women with
i 1 normal
pregnancy
Multiparous
cases &
controls
BP > 160/110
mmHg&
proteinuria (>
500 mg/24 h),
HELLP or
eclampsia
Not specified History of thrombophilic
event, HT in early
pregnancy, multiple
pregnancies
63/126 8.0/3.0 0.14
(unadj)
0.03 (adj)
4^
Reproduced w ith permission o f th e copyright owner. Further reproduction prohibited without permission.
Table 3.3 Molecular epidemiology studies of the Prothrombin G20210A polymorphism and risk of PE (continued)
Reference &
Country of Study
Type of
Cases
Type of
Controls
Parity PE Definiton Covariates
(Adjustments)
Exclusions Cases/
Controls
20210A Allele
Frequency %
(Cases/ Controls)
P
Kupferminc, Peri
et al. 2000
(Israel)
Note: Included 90
subjects (controls)
also described in
Kupferminc,
Eldoretal. 1999
Women with
incident
obstetric
complica
tions
Healthy
women with
k 1 normal
pregnancy
Multiparous
cases &
controls
Per American
College of
Obstetricians
and
Gynecologists
None No history of a
thrombophilic event
222/156 For PE only
(n = 80):
8.8/3.2 >0.05
Livingston,
Barton et al. 2001
(U.S.)
Incident
severe PE
Normo
tensive
throughout
pregnancy
with no
history of
PE
Multiparous
cases &
controls
(implied)
BP 2 160/110
mmHg &
proteinuria (>
300 mg/24 h),
eclampsia or
HELLP
None Chronic HT, diabetes,
preexisting renal disease,
history of
thromboembolism,
multifetal gestation, major
fetal congenital anomaly
Maternal
samples:
110/97
Fetal
samples:
75/92
Maternal samples:
0/1.0
Fetal samples:
1.3/2.2
>0.05
> 0.05
O ’Shaughnessy,
Fu et al. 2001
(UK)
NOTE: Includes
200 subjects
(controls) & may
also include some
of the same cases
as described in
O’Shaughnessy,
Fuetal. 1999
Incident PE Age-
matched
normo
tensive
pregnant
(n=100) &
normo
tensive
nonpregnant
with no
PE/PIH
history &
age < 50 at
screen
(n=100)
Multiparous
cases &
controls
BP > 140(90
mmHg after
week 20 with
at least 25
mmHg increase
in diastolic BP
& proteinuria
(> 300 mg/24h
or ++ dipstick);
resolved by 3
months
pospartum
None Multiple births, diabetes,
renal disease, essential HT
356/200 2.0/1.1 > 0 .0 5
4^
3.5.2.3 Factor VLeiden Mutation (FVL): The Leiden mutation is a G— »A substitution
at nucleotide position 1691 in exon 10 of the factor V gene. It results in the
replacement of glutamine for arginine at position 506 at the cleavage site for
activated protein C (APC) (Bertina, Koeleman et al. 1994; Dahlback 1994; Svensson
and Dahlback 1994), causing activated protein C resistance (APCR), retention of
procoagulant activity and thus, increased susceptibility to thrombophilic events
(Mello 1999). Pregnant women with APCR are at increased risk of PE (Dekker, de
Vries et al. 1995). Additionally, placentas with greater than 10% placental infarction
have been associated with a more than 10-fold increase in fetal FVL carrier rate,
suggesting that fetal, as well as maternal, FVL genotype may be of importance
(Dizon-Townson, Meline et al. 1997). The frequency of the Leiden mutation ranges
from 0.6% in Asia Minor to 7% among Greeks, with a frequency among those in the
United Kingdom of 4.4% (Rees, Cox et al. 1995).
The majority of epidemiological research also suggests an association between FVL
and PE (Table 3.4). Studies of women in Hungary (Nagy, Toth et al. 1998), Israel
(Kupferminc, Eldor et al. 1999; Kupferminc, Fait et al. 2000), Italy (Grandone,
Margaglione et al. 1997; Grandone, Margaglione et al. 1999; Mello 1999) and Utah
(Dizon-Townson, Nelson et al. 1996) have all found a statistically significant striking
increase in risk of PE associated with either being a homozygote or a heterozygote
carrier of FVL. In contrast, several studies have failed to find an association (De
Groot 1999; Kobashi 1999; O'Shaughnessy, Fu et al. 1999; Kim 2001; Livingston,
Barton et al. 2001). However, in a Japanese population, none of the cases or controls
carried the FVL mutation (Kobashi 1999). One European study that failed to find an
46
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
association found a high FVL carrier rate among both cases (9.8%) and control
women (9.2%) (De Groot 1999), suggesting that selection bias among controls may
be responsible for the negative results. One U.S. study found a statistically
nonsignificant elevation in carrier rate among cases (Kim 2001), while one European
(O'Shaughnessy, Fu et al. 2001) and one American (Livingston, Barton et al. 2001)
study found no association at all. Additionally, Livingston and colleagues examined
the possible role of fetal FVL mutation and found no association. Although it seems
clear that FVL is associated with increased risk of thrombophilia, the strength of its
association with PE remains to be clarified.
3.5.2.4 5,10-Methylenetetrahydrofolate Reductase (MTHFR): Increased
homocysteine levels are known to cause vascular injury (McCully 1996) and
hyperhomocysteinemia has been reported in preeclamptic patients (Dekker, de Vries
et al. 1995). The mutation most commonly reported to be associated with PE is a
cytosine-to-thymine substitution at nucleotide 677, which results in an amino acid
substitution (alanine-*valine). The two alleles are commonly referred to as C (wild
type) and T (variant). The variant causes reduced MTHFR activity and modestly
increases circulating homocysteine levels (Frosst, Blom et al. 1995; O'Shaughnessy,
Fu et al. 1999). In a Los Angeles population, the frequency of the TT genotype was
highest among Hispanics with 15.2% having the high-risk genotype, compared to
10.2% among Caucasians, 8.8% among Asians and 2.4% among African Americans
(Levine, Siegmund et al. 2000).
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Reproduced w ith permission o f th e copyright owner. Further reproduction prohibited without permission.
Table 3.4 Molecular epidemiology studies o f the factor V Leiden mutation and risk o f PE (next 2 pages)
Reference &
Country of Study
Type of
Cases
Type of
Controls
Parity PE Definiton Covariates
(Adjustments)
Exclusions Cases/
Controls
FVL
Carrier
Rate %
(Cases/
Controls)
P
Dizon-Townson,
Nelson etal. 1996
(Utah)
Incident
severe PE
Normotensive
term gravid
women
Not
specified
Sustained HT (BP >
160/110 mmHg) &
proteinuria (> 5 g
gm/24 h or 3+ or 4+)
None Preexisting HT, renal disease,
HELLP syndrome
158/403 8.9/4.2 0.03
Grandone,
Margaglione et al.
1997
(Italy)
Incident PIH Healthy parous
pregnant
controls;
matched on
ethnicity
Multiparous
cases &
controls
BP > 140/90 mmHg
on at least 2 occasions
> 4 h a p a rt (n=51)
Plus proteinuria (>
300 mg/24 h) (n=45)
None Transient HT, diabetes,
chronic renal or pulmonary
disease, chronic essential HT
prior to pregnancy (possibly
excluded in cases only)
95/128 10.5/2.3 0.02
Nagy, Toth et al.
1998
(Hungary)
Incident
severe PE
Healthy NP
(n=58) &
healthy P
(n=71)
Not
specified,
but
multiparous
implied
BP > 160/90 mmHg &
proteinuria (1000
mg/24 h)
None None specified 69/129 18.8/
7.0 (P) &
5.2 (NP)
<0.05
Kobashi 1999
(Japan)
Incident PIH Normotensive
pregnant
controls
Primiparous
cases &
controls
One or more of the
following: (1) 30/15
mmHg increase in BP
from baseline (2) BP
> 140/90 mmHg or (3)
proteinuria (1+)
None Multiple births, renal disease,
diabetes, amniotic volume
abnormalities, preexisting HT,
fetal anomalies
71/109 0/0 NA
Mello 1999
(Italy)
PE History Healthy
women with
normal preg
nancy history;
matched for
age, gravidity
& parity
Multiparous
cases &
controls
Sustained HT (BP >
140/90 mmHg) &
proteinuria (> 0.3
g/1/24 h)
None No history of HT before week
20 in prior pregnancies,
diabetes, not taldng
estroprogestinic preparations
All subjects were Caucasian
46/80 26.1/3.8 <0.04
O ’Shaughnessy,
Fu et al. 1999
(UK)
Incident PE Age-matched
normotensive P
controls
(n=100) &
normotensive
NP controls
with no PE
history(n=100)
Multiparous
cases &
controls
BP > 140/90 mmHg
after 20 weeks with at
least 25 mmHg rise in
diastolic BP &
proteinuria (> 300
mg/24 h or ++
dipstick); resolved by
3 months postpartum
None Multiple births, concurrent
diabetes, renal disease,
essential HT
283/200 53/5.5
(pooled
controls)
>0.05
4^
00
Reproduced w ith permission o f th e copyright owner. Further reproduction prohibited without permission.
Table 3.4 Molecular epidemiology studies of the factor V Leiden mutation and risk of PE (continued)
Reference &
Country of Study
Type of
Cases
Type of
Controls
Parity PE Definiton Covariates
(Adjustments)
Exclusions Cases/
Controls
FVL
Carrier
Rate %
(Cases/
Controls)
P
DeGroot 1999
(Netherlands)
PE history Age & delivery
date matched;
normotensive
pregnancy
history
Primiparous
cases &
controls
>30/15 mmHg rise in
BP or > BP 140/90
mmHg & proteinuria
(> 30 mg/dL)
Family hx. of
thrombosis or
HT, body mass,
smoking,
coagulation
mutations
Multiple pregnancies, chronic
HT, renal disease, diabetes,
collagen vascular diseases,
cancer, thrombosis history
163/163 9.8/9.2 >0.05
Grandone,
Margaglione et al.
1999
(Italy)
NOTE: May
include some of
the same subjects
as Grandone,
Margaglione et al.
1997
Incident GH
(with &
without
proteinuria)
Normotensive
gravid women
matched on
ethnicity
Multiparous
cases &
controls
BP > 140/90 mmHg
on 2 occasions > 4 h
apart & with (n = 70)
or without (n = 70)
proteinuria (> 300
mg/24 h)
Maternal age,
parity,
prothrombin
A20210 carrier
status, 677T
MTHFR
homozygosity
Transient HT, diabetes,
autoimmune disease, chronic
renal or pulmonary disease &
chronic essential HT prior to
pregnancy
140/216 7.9/1.9 <0.05
Kupferminc,
Eldoretal. 1999
(Israel)
Incident
severe PE,
AP, FGR or
stillbirth
Age &
geographically
matched
normotensive
pregnancy
Multiparous
cases &
controls
Severe PE: BP >
160/110 mmHg,
proteinuria (> 500
mg/24 h), HELLPor
eclampsia
None, but
stratified by
diagnosis
Varied by diagnosis - none
specified for severe PE
110/110 For severe
PE (n=34):
26.5/6.0
<0.05
Kupferminc, Fait
et al. 2000
(Israel)
Incident
severe PE
Age &
ethnicity
matched
women with >
1 normal preg.
Multiparous
cases &
controls
BP > 160/110 mmHg
& proteinuria (> 500
mg/24h), HELLPor
eclampsia
Not specified History of thrombophilic
event, HT in early pregnancy,
multiple pregnancies
63/126 24.0/6.3 0.001
(unadj)
<0.05
(adj)
Kim 2001
(Iowa & North
Carolina)
Incident PE History of i 2
normotensive
term pregnan
cies without
HT or family
histoiy of PE
Multiparous
cases &
controls
Mild PE: BP > 140/90
mmHg & proteinuria
(> 300 mm/24-h or +
dipstick)
None, but
stratified on
parity, PE
severity &
HELLP status
None (except as listed for
controls)
281/360 3.0/2.4
(for all PE
cases
combined)
0.53
Livingston,
Barton et al. 2001
(U.S.)
Incident
Severe PE
Normotensive
throughout
pregnancy with
no history of
PE
Multiparous
cases &
controls
(implied)
BP >160/110 mmHg
& proteinuria (> 300
mg/24 h), eclampsia
or HELLP
None Chronic HT, diabetes,
preexisting renal disease,
history of thromboembolism,
multifetal gestation, major
fetal congenital anomaly
Maternal
Samples:
110/97
Fetal
Samples:
75/92
Maternal
Samples
2.7/1.5
Fetal
Samples:
2.0/0
>0.05
>0.05
4^
v©
The epidemiologic data are mixed (Table 3.5). Studies in Israel (Kupferminc, Eldor
et al. 1999; Kupferminc, Fait et al. 2000), Japan (Sohda, Arinami et al. 1997), and
Italy (Grandone, Margaglione et al. 1997) have all found that the TT genotype is
associated with a statistically significant increase in risk of PE. Interestingly, the PE
risk conferred by the TT genotype was highest among proteinuric nulliparae cases
(Grandone, Margaglione et al. 1999). In contrast, studies in the United States
(Powers 1999; Kim 2001; Livingston, Barton et al. 2001), Australia (Kaiser,
Brennecke et al. 2000), the United Kingdom (O'Shaughnessy, Fu et al. 1999), Finland
(O'Shaughnessy, Fu et al. 1999; Laivuori, Kaaja et al. 2000) and The Netherlands
(Lachmeijer, Amgrimsson et al. 2001) have all failed to find a direct association of
the C677T mutation with PE in either maternal (Powers 1999; Kaiser, Brennecke et
al. 2000; Kim 2001) or fetal (Livingston, Barton et al. 2001) samples. It would be
useful to know what percentages of each of the case groups used in this study were
primiparous and to have information on possible underlying maternal conditions.
A second variant, an adenine-to-cytosine substitution at base 1298 (A1298C), has
recently been described (van der Put 1998). The variant (C allele) also results in
decreased MTHFR activity but is not associated with increased plasma homocysteine
or decreased plasma folate (Lachmeijer, Amgrimsson et al. 2001). The frequency of
the C allele has been reported to be 33% in a European population (van der Put
1998). To date, only one study has examined the effect of the A1298C mutation on
PE risk, finding no difference in mutation frequency between PE cases and controls
(Lachmeijer, Amgrimsson et al. 2001). However, the controls used in this study were
not screened for PE history, and thus, could have included some women with a
50
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
history of PE. The inclusion of such women in the control group might have
obliterated a true difference in mutation rate between women with normotensive
pregnancy histories and those with a history of PE.
Since folate can decrease homocysteine levels irrespective of MTHFR genotype, it is
important to address the possibility of confounding due to folate status. Folate
supplementation is more common in the U.S. than in many other countries, and is
actively encouraged in women of childbearing age. Therefore, differences in the
folate status of the study population could at least partially explain the differing
results obtained.
3.5.2.5 Cystathionine /3-Synthase (CBS). CBS is an enzyme that catalyzes the
condensation of serine with homocysteine to produce cystathionine (Kraus,
Oliveriusova et al. 1998). It has been suggested that CBS deficiency leads to high
plasma levels of homocysteine and might predispose to thrombophilic events (Kim
2001). A T-»C substitution at nucleotide 833 results in the substitution of threonine
for isoleucine and has been shown to be associated with mild hyperhomocysteinemia
(Kim 2001), but is relatively rare in the general population (<1%) (Silaste, Rantala et
al. 2001). This mutation is known to cosegregate in cis with a common 68-bp
insertion in the coding region of exon 8 (844ins68) (Franco, Elion et al. 1998). Allele
frequencies for the insertion were 6.7% among whites, 0.45% among Amerindians
and 11% among Blacks (Franco, Elion et al. 1998).
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Reproduced w ith permission o f th e copyright owner. Further reproduction prohibited without permission.
Table 3.5 Molecular epidemiology studies of the MTHFR C677T polymorphism and risk of PE (next 3 pages)
Reference &
Country of Study
Type of
Cases
Type of
Controls
Parity PE Definiton Covariates
(Adjustments)
Exclusions Cases/
Controls
IT Homo
zygosity
% (Cases/
Controls)
P
Grandone,
Margaglione et al.
1997
(Italy)
Incident PIH Healthy
parous P
controls;
matched on
ethnicity
Multiparous
cases &
controls
BP > 140/90
mmHg on at least 2
occasions > 4 h
apart (n=51)
Plus proteinuria (>
300 mg/24 h)
(n=45)
None Transient HT, diabetes,
chronic renal or pulmonary
disease, chronic essential
HT prior to pregnancy
(possibly excluded in cases
only)
94/129 29.8/18.6 A . f o
Sohda, Annami et al.
1997
(Japan)
PE cases “Matched”
healthy P
controls
(n=98) &
normotensive
healthy NP
controls
(n=260)
Not
specified
According to
American College
of Obstetricians
and Gynecologists
(1990) criteria
None None 67/358 24.0/11.0
(all
controls)
< 0.004
O'Shaughnessy, Fu
etal. 1999
(UK)
Incident PE Age-matched
normotensive
P controls
(n=100) &
normotensive
NP controls
with no PE
history & age
< 50 at screen
(n=100)
Multiparous
cases &
controls
BP > 140/90
mmHg after 20
weeks with at least
25 mmHg rise in
diastolic BP &
proteinuria (> 300
mg/24 h or ++
dipstick); resolved
by 3 months
postpartum
None Multiple births, concurrent
diabetes, renal disease,
essential HT
283/200 11.0/11.5
(all
controls)
>0.05
Powers 1999
(Pennsylvania)
Incident PE
(n=99) &
Incident GH
(n=24)
Normotensive
P controls
Primiparous
cases &
controls
GH: BP > 140/90
mmHg or increase
of > 30/15 mmHg
over BP at 20
weeks
PE: GH &
proteinuria (> 500
mg/24 h or > +
dipstick) & hyper
uricemia (> 5.5
mM) all resolving
after pregnancy
None, but
stratified by
diagnosis
Chronic HT, renal disease,
metabolic disease
123/114 For PE vs.
Controls:
15.0/12.2
*Note: also
found no
difference
between
infants of
cases &
controls
0.21
Reproduced w ith permission o f th e copyright owner. Further reproduction prohibited without permission.
Table 3.5 Molecular epidemiology studies of the MTHFR C677T polymorphism and risk o f PE (continued)
Reference &
Country of Study
Type of
Cases
Type of
Controls
Parity PE Definition Covariates
(Adjustments)
Exclusions Cases/
Controls
TT Homo
zygosity
% (Cases/
Controls)
P
Grandone,
Margaglione et al.
1999
(Italy)
Note: May include
some of the same
subjects as
Grandone,
Margaglione et al.
1997
Incident GH
(with &
without
proteinuria)
Normo
tensive gravid
women
matched on
ethnicity
Multiparous
cases &
controls
BP a 140/90
mmHg on 2
occasions a 4 h
apart & with (n =
70) or without (n =
70) proteinuria (a
300 mg/24 h)
Maternal age,
parity, FVL
carrier status,
prothrombin
A20210 carrier
status
Transient HT, diabetes,
autoimmune disease,
chronic renal or pulmonary
disease & chronic essential
HT prior to pregnancy
139/216 24.5 (all
cases)
/16.7
>0.05
Kupferminc, Eldor et
al. 1999
(Israel)
Incident
severe PE,
AP, FGRor
stillbirth
Age &
geographical
ly matched
normotensive
pregnancy
Multiparous
cases &
controls
Severe PE: BP >
160/110 mmHg,
proteinuria (> 500
mg/24 h), HELLP
or eclampsia
None, but
stratified by
diagnosis
Varied by diagnosis - none
specified for severe PE
110/110 For Severe
PE (n=34):
20.6/8.0 0.05
Kupferminc, Fait et
al. 2000
(Israel)
Incident
severe PE
Age &
ethnicity
matched
women with
> 1 normal
pregnancy
Multiparous
cases &
controls
BP > 160/110
mmHg &
proteinuria (> 500
mg/24 h), HELLP
or eclampsia
Not specified History of thrombophilic
event, HT in early
pregnancy, multiple
pregnancies
63/126 24.0/10.0 0.008
(unadj)
<0.05
(adj)
Laivuori, Kaaja et al.
2000
(Finland)
PE (n=13) &
Severe PE
(n=100)
Healthy
women with
history of > 1
normotensive
pregnancies
Multiparous
cases &
controls
PE: BP > 140/90
mmHg on 2
occasions > 6 h
apart & proteinuria
(> 300 mg/24 h or
+ dipstick),
resolved by 12
weeks postpartum
Severe PE: >
160/110 &
proteinuria (> 200
mg/24 h), resolved
by 12 weeks
postpartum
None Renal or autoimmune
disease
113/103 3.0/7.0 0.55
Reproduced w ith permission o f th e copyright owner. Further reproduction prohibited without permission.
Table 3.5 Molecular epidemiology studies of the MTHFR C677T polymorphism and risk of PE (continued)
Reference &
Country of Study
Type of
Cases
Type of
Controls
Parity PE Definiton Covariates
(Adjustments)
Exclusions Cases/
Controls
'IT Homo
zygosity
% (Cases/
Controls)
P
Lachmeijer,
Amgrimsson et al.
2001
(Netherlands)
PE history;
consecutive
patients
(n=47),
affected sib-
pair families
(n=127) &
patients with
known
homocys
teine status (n
= 85)
Healthy
Dutch blood
donors
Not
specified for
cases,
unknown for
controls
Diastolic BP > 90
mmHg on at least 2
occasions > 6 h
apart or increase of
20 mmHg over
first-trimester
levels & proteinuria
(> 300 mg/24 h)
Stratified on
severity (PE,
HELLP or
eclampsia);
adjusted model
with
homocysteine
& vitamin
levels as
dependent
variables &
genotypes as
independent
variables
None 259/120 6.5
(consecu
tive cases);
7.9 (sib-
pair
cases)/9.2
(controls)
32.9
(hyper-
homocys-
teinemia)
/5.6
(normal
homocys
teine)
0.97
(con
secutive
cases)&
0.06
(sib-pair
cases)
.001
Kaiser, Brennecke et
al. 2000
(Australia)
Severe PE
(n=116) &
eclampsia
(n=40) cases
Normal P
controls
Primiparous
cases &
controls
BP > 140/90
mmHg or rise of
25/15 mmHg from
baseline on > 2
occasions > 6 h
apart & proteinuira
(> 300 mg/24 h or
++ dipstick)
None Chronic HT, renal disease,
metabolic disease
156/79 12.2/13.9 >0.05
Kim 2001
(Iowa & North
Carolina)
Incident PE History of > 2
normotensive
term
pregnancies
without HT or
family history
of PE
Multiparous
cases &
controls
Mild PE: BP >
140/90 mmHg &
proteinuria (> 300
mg/24 h or +
dipstick)
None, but
stratified on
parity, PE
severity &
HELLP status
None (except as listed for
controls)
281/360 11.7 (all
PE cases)
/11.4
0.98
Livingston, Barton et
al. 2001
(U.S.)
Incident
Severe PE
Normotensive
throughout
pregnancy
with no
history of PE
Multiparous
cases &
controls
(implied)
BP >160/110
mmHg&
proteinuria (> 300
mg/24 h),
eclampsia or
HELLP
None Chronic HT, diabetes,
preexisting renal disease,
history of
thromboembolism,
multifetal gestation, major
fetal congenital anomaly
Maternal
Samples:
110/97
Fetal
Samples:
75/92
Maternal
Samples
9.0/7.0
Fetal
Samples:
9.0/4.0
>0.05
>0.05
The increased incidence of PE among women with hyperhomocysteinemia has led to
speculation about a potential role for CBS in PE development. Thus far, only one
study has examined the role of the CBS insertion mutation in relation to PE and has
found no association with disease among nulliparous white carriers in the United
States (Kim 2001). Tsai and colleagues have provided a possible explanation for the
lack of association between the insertion polymorphism and PE. They suggest that
the mutation is not related to homocysteine levels since it creates an alternate splicing
site that eliminates the T833C polymorphic site (which is associated with
homocysteine levels) and produces a normal mRNA and CBS enzyme (Tsai, Bignell
et al. 1996). This assertion is supported by the finding that both heterozygous
carriers of the insertion as well as homozygous subjects do not have lowered CBS
enzyme activity (Tsai, Bignell et al. 1996) nor were plasma total homocysteine levels
different between heterozygous carriers and noncarriers (Silaste, Rantala et al. 2001).
3.5.2.6 f33 Integrin Glycoprotein Ilia (GPIIIa): The GPIIIa gene encodes the beta
subunit of the GP Ub/IIIa and otv(53 complexes that belong to a class of receptors that
bind cell adhesion molecules. The GP Ub/IIIa receptor is found only in platelets and
megakaryocytes and is important in platelet aggregation. The avp3 integrin is
expressed by invading trophoblasts, suggesting it may be important in placentation.
Additionally, the (53 integrin has been implicated in the failure of the cytotrophoblasts
to adopt a vascular phenotype in women with PE (Zhou, Damsky et al. 1997).
A coding variant (C98T) in exon 2 of the GPIIIa gene creates two antigenically
distinct forms of the mature GPIIb/IIIa antigen on platelets (P1(A) antigens 1 and 2)
55
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
(O'Shaughnessy, Fu et al. 2001). The variant (T98) has been associated with risk of
premature acute coronary syndromes and strokes in young white women (Carter,
Catto et al. 1998). One association study has examine the role of the variant in PE
and found a statistically significant excess of homozygotes (T98/T98) among white
northern European women with PE . These results need to be confirmed in other
studies and in other ethnic groups.
3.5.2.7 Matrix Metalloproteinase-1 (MMP1): MMP1 is produced by decidual,
endothelial and trophoblastic cells and is involved in the process of interstitium
remodeling and degrading collagens I, II, III, VII and X (Jurajda, Kankova et al.
2001). Interstitium remodeling occurs very early in pregnancy, prior to trophoblast
invasion, and is a necessary part of the normal arterial changes that occur during
pregnancy. Among women with PE, the amount of MMP1 in decidual artery
endothelial cells is reduced in comparison to normotensive pregnant women,
potentially explaining the inhibited vascular invasion by cytotrophoblasts in
preeclamptics (Gallery, Campbell et al. 1999).
An insertion polymorphism (1G/2G) at position -1607 in the MMP1 gene creates an
Ets binding site and is associated with higher transcriptional activity (2G) (Rutter,
Mitchell et al. 1998). A single study has examined the possible role of this
polymorphism in PE etiology, but found no increase in PE, eclampsia or PIH risk
associated with the variant allele (Jurajda, Kankova et al. 2001). However, since the
sample size was small (58 cases), the failure to find an association may have been
due to lack of power. Moreover, the study included primarily multiparous women
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
who had a wide range of hypertensive pregnancy disorders, creating a population
potentially too heterogeneous to find an effect. Lastly, fetal MMP1 genotype could
be important in the remodeling process, but was not considered in the single study
examining MMP1 genotype and risk of PE.
3.5.3 Endothelial Cell Health
Genes categorized as affecting endothelial cell health generally do so through their
effect on oxdative stress. Specifically, small dense LDLs, which are increased as part
of the dyslipidemia present in women prior to the development of PE, are highly
oxidizable and can lead to membrane damage. Thus, genes involved in the
production or metabolism of reactive oxygen species (ROS) may lead to increased
oxidative stress and endothelial cell dysfunction.
3.5.3.1 Epoxide Hydrolase Gene (EPHX). Microsomal epoxide hydrolase (EPHX) is
involved in the hydrolysis of certain oxides to form less toxic products, but the
process may also produce toxic intermediates that could contribute to PE
development (Zusterzeel, Peters et al. 2001). Two relevant polymorphisms have been
described: Tyrll3His in exon 3 has been associated with decreased enzyme activity
and Hisl39Arg in exon 4 has been associated with increased enzyme activity
(Zusterzeel, Peters et al. 2001). The fast alleles that produce more active enzymes,
Tyrl 13 and Argl39, are hypothesized to increase the risk of PE by producing more
toxic intermediates and thus inducing endothelial cell damage. Allele frequencies for
the Hisl39Arg polymorphism are relatively stable across populations but vary greatly
by population for the Tyrl 13His polymorphism. For example, the frequency of
57
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Tyrl 13/Tyrl 13 homozygotes ranges from 26-69%, depending on the ethnicity and
geography of the sample.
Thus far, only one study has addressed the possibility that one or both of these
variants could play a role in PE development. Zusterzeel and colleagues found no
significant differences between cases and controls for genotypes at the 139 locus,
however, at the 113 locus the fast genotype was two times more common in women
with a history of PE than in women with a normotensive pregnancy history
(Zusterzeel, Peters et al. 2001). Additionally, when genotypes at the two
polymorphic sites were combined and categorized into high, intermediate and low
overall EPHX activity, Zusterzeel et al. found that women with high EPHX activity
were overrepresented in the PE cases compared to the controls.
A source of concern is the exclusion criteria used in this study. Controls with known
predisposing conditions (e.g., hypertension, renal disease, etc.) were excluded from
the study population while cases with these conditions were not. In theory, bias away
from the null can occur when exclusions are made among the controls that are not
made among the cases. Such bias could explain the associations seen in this study,
therefore, more research on this gene is needed before it can be dismissed or accepted
as a potential candidate gene for PE.
3.5.3.2 Lipoprotein lipase gene (LPL). Triglyceride and free fatty acid accumulation
in women with PE has led to speculation about the potential role of the lipoprotein
lipase (LPL) gene. LPL is responsible for mediating the clearance of circulating
58
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
lipoprotein triglyceride and a decrease in its activity could lead to endothelial cell
dysfunction. Four relatively common polymorphisms in the coding region have been
identified thus far which are associated with alterations in plasma lipids: Asn291Ser
(exon 6), Asp9Asn (exon 2), -93T -> G (promoter region), and Ser447X (Fisher,
Humphries et al. 1997). The Asn9 and -93G variants are in strong linkage
disequilibrium and are inherited together in Caucasians. Since the Ser291 and Asn9/-
93G variants result in lower LDL activity and increased dyslipidemia, they are
expected to increase the risk of PE. Conversely, the X447 variant results in lower
triglycerides and higher HDL concentrations and should therefore result in less risk
of PE (Hubei, Roberts et al. 1999). Allele frequencies for the high-risk alleles were
1.5% for Ser291, 0.7% for Asn9/-93G and 9.4% for Ser447 in a random sample of
population controls from Pennsylvania (Hubei, Roberts et al. 1999).
To date, only two studies have examined the possible role of LPL gene
polymorphisms on the risk of PE. In one well-designed study, the Ser291 and Asn9/-
93G alleles were found statistically significantly more frequently among cases of PE
than among either pregnancy or population controls while the frequency of the X447
allele did not vary (Hubei, Roberts et al. 1999). Conversely, Kim et al. failed to find
a statistically significant difference in allele frequencies between cases of PE or for
the offspring of PE cases compared to controls for the Asp9Asn, -93T -> G or
Asn291Ser polymorphisms (Kim, Williamson et al. 2001). However, among
nulliparous women with HELLP syndrome (n=12), there was a statistically
significant increase in the frequency of the Ser291 allele compared to controls (8.3%
vs. 1.5% respectively, p < 0.05) (Kim, Williamson et al. 2001). Since both study
59
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
populations consisted entirely of Caucasians and patterns of linkage disequilibrium
may differ by ethnicity, more research should be conducted in other ethnic groups.
However, the relative rarity of these polymorphisms suggests that they cannot
possibly account for a large proportion of PE cases.
3.5.3.3 Superoxide dismutase (SOD). SOD catalyzes the dismutation of superoxide
into oxygen and hydrogen peroxide, thereby protecting against oxidative damage
from superoxide radicals (Fridovich 1978; Getzoff, Tainer et al. 1983). Decreased
SOD activity results in increased levels of superoxides, which could reasonably be
expected to increase oxidative stress and possibly result in PE.
There are three types of SOD in eukaryotic cells. The predominant type, encoded by
the SOD1 gene, is found in the cytosol and contains copper and zinc at the catalytic
site (CuZn-SOD) while the manganese-containing enzyme (Mn-SOD), encoded by
SOD2, is found in the mitochondrial matrix (Michelson, McCord et al. 1977). A
third unrelated gene, SOD3, encodes an extracellular enzyme also containing copper
and zinc (Hjalmarsson, Marklund et al. 1987). Evidence suggests that trophoblasts
from preeclamptic placentae generate statistically significantly more superoxide than
trophoblasts from normal placentae (Wang and Walsh 2001). Likewise, SOD activity
and relative mRNA expression for CuZn-SOD were statistically significantly
decreased in trophoblast cells from preeclamptic compared to normal placentae
(Wang and Walsh 2001). Thus, increased superoxide generation appears to be
associated with decreased CuZn-SOD mRNA expression and enzyme activity in
trophoblast cells from preeclamptic placentae.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
To date, only one study has examined the role of the SOD1 gene in relation to PE risk
and no studies have examined the role of SOD2 or SOD3. In a very small case-
control study, Chen and colleagues found that SOD activity was statistically
significantly decreased in women with PIH compared to pregnant controls (Chen,
Wilson et al. 1994). They found no evidence of polymorphism in EcoRl cutting sites
in or near the CuZn-SOD gene, however the possibility that other types of
polymorphisms in the gene might be responsible for the decreased SOD activity
remains to be investigated. In addition, nine of the 14 women included in the study
did not have proteinuria, thus, they did not meet the criteria for PE and the authors
failed to exclude conditions known to be associated with an increased risk of PE,
creating a very heterogeneous population. Thus, more epidemiological data is needed
before a role for the CuZn-SOD gene can be entirely dismissed and data needs to be
accumulated to assess the role of the other SOD genes.
3.53.4 Long-chain 3-hydroxyacyl-CoA dehydrogenase (LCHAD). Patients with
LCHAD deficiency have impaired oxidation of long-chain fatty acids and severe
pregnancy complications, including HELLP syndrome, have been reported in
heterozygous carriers of a mutation in the LCHAD gene (Wilcken, Leung et al. 1993;
Treem, Shoup et al. 1996). The mutation, a G-»C substitution at position 1528 was
shown to be directly responsible for the loss of LCHAD activity (Ijlst, Ruiter et al.
1996). While mutation rates in subjects with LCHAD deficiency have been reported
to be as high as 87% (Ijlst, Ruiter et al. 1996), the mutation appears to be very rare in
the general population (den Boer, Ijlst et al. 2000).
61
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
The G1528C variant was studied in a Dutch population, but not found to be
associated with HELLP syndrome (den Boer, Ijlst et al. 2000). However, it has been
suggested that only women carrying fetuses homozygous for the LCHAD mutation
are at increased risk of HELLP (den Boer, Ijlst et al. 2000). Unfortunately, Den Boer
et al. examined maternal genotypes without consideration of fetal genotypes.
Additionally, the lack of information on how the cases and controls were selected for
the study makes these results difficult to interpret.
3.5.3.5 Apolipoprotein E (apoE). ApoE is involved in the clearance of atherogenic
remnants of triglyceride rich lipoproteins, and the ApoE gene has three common
alleles e2, s3 and s4 (Nagy, Rigo et al. 1998). Cholesterol absorption efficiency in
the intestines increases in allelic order (s2 < s3 < e4) (Miettinen 1991). Since the
apoE e4 allele is an established risk factor for dyslipidemia (Sattar, Bendomir et al.
1997) and women with PE are known to have alterations in their lipid profiles, the
apoE s4 allele would be expected to increase the risk of PE.
Nagy et al. unexpectedly found a statistically significant increase in PE risk
associated with the s2 allele and a non-significant reduction in s4 allele frequency
among preeclamptics compared to healthy pregnant controls (Nagy, Rigo et al. 1998).
In contrast, Makkonen and colleagues failed to find support for a role of the apoE
gene in PE development (Makkonen, Heinonen et al. 2001). However, the study by
Nagy and colleagues was limited by a fairly small sample size (52 healthy pregnant
women, 54 severe preeclamptics and 101 healthy nonpregnant women). More
importantly, since the range of gestational ages in the normal, healthy pregnancy
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
group includes gestations which are prior to the gestational age at which PE is most
commonly diagnosed, we do not know if some of these control women went on to
develop PE. Additionally, both studies included a control group of healthy blood
donors, some of whom may have a history of PE. If some women with PE were
included in the control groups, it could obliterate a possible true association between
the ApoE e4 allele and PE as well as create spurious associations with other alleles.
Thus, more research is needed to clarity the possible role of ApoE genotype in PE
development.
3.5.4 Immune Maladaptation
Immune function is normally suppressed during pregnancy, ostensibly as an adaptive
measure meant to protect the fetal allograft from maternal attack. As mentioned
previously, immune function appears to be abnormal in women who develop PE.
Thus, genes that encode the various aspects of the immune system may be good
candidates for involvement in PE development.
3.5.4.1 Human Leukocyte Antigens (HLA). Human leukocyte antigens (HLA) are
involved in immune tolerance and thus, deviation from normal expression may play a
role in the immune maladaptation believed to be important in PE development.
Specifically, HLA is thought to be involved in the rejection of the fetus by the
mother, or vice versa. To date, the literature in this area is rather muddled with most
studies failing to prove replicable (Cooper 1993). The usual difficulties in studying
genetic susceptibility to PE are compounded by the fact that there are a large number
of antigens at each HLA locus, making it very difficult to study them effectively in
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
small studies. In addition, studies attempting to investigate numerous antigens suffer
from the problem of multiple comparisons and uninterpretable older studies in which
laboratory methodologies were less reliable in distinguishing between homozygotes
and heterozygotes (Cooper 1993). Since comprehensive reviews have already been
written to address the findings in the HLA system (Cooper 1993; Kilpatrick 1999;
Talosi, Endreffy et al. 2000), we focus here on only a few of the more recent findings.
3.5.4.2 HLA-G. HLA-G is highly expressed in first trimester but greatly reduced in
third trimester cytotrophoblast cells suggesting that it may be important in placental
development (Yelavarthi, Fishback et al. 1991) and in adequate trophoblast invasion
(O'Brien, Dausset et al. 2000). HLA-G has also been shown to help regulate the
release of cytokines and thus might be important in blocking a maternal immune
response to the placenta and ensuring a viable pregnancy (Humphrey, Harrison et al.
1995; Bermingham, Jenkins et al. 2000).
Despite evidence suggesting that HLA-G expression in trophoblasts is reduced in
preeclamptic pregnancies (Goldman-Wohl, Ariel et al. 2000), the results of both
linkage and candidate gene analyses fail to support the hypothesized role of HLA-G
in genetic susceptibility to PE. Specifically, Bermingham et al. found no difference
between PE offspring and control offspring in a Caucasian population for three
polymorphisms studied: A-»T substitution at codon 107, C-»A substitution at codon
110 and a 14 bp insertion/deletion polymorphism in exon 8 (Bermingham, Jenkins et
al. 2000). An excess of heterozygotes for the deletion polymorphism was found in
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
PE offspring, but the relevance of this deviation has yet to be determined. Moreover,
a well-designed study examined the role of both maternally and fetally expressed
HLA-G on the risk of PE, but found that neither showed linkage to PE or eclampsia
(E) (Humphrey, Harrison et al. 1995).
Unfortunately, the Bermingham study failed to describe the source population for its
cases and controls and therefore, it is impossible to assess whether selection bias may
be affecting the results. Additionally, proteinuria was not required for a PE
diagnosis, suggesting that the study population may be heterogeneous. Lastly, a
relatively small sample size (n = 68) could have limited their power to detect an
effect.
3.5.43 HLA-DR. The HLA-DR antigens play a role in the recognition of self versus
non-self and have been the focus of several studies (de Luca Brunori, Battini et al.
2000). To date, only the role of maternal HLA-DRp has been examined; the possible
role of fetal HLA-DRp genotype PE risk has not be studied. The only study to
examine this locus failed to find linkage between maternal HLA-DRp and PE (Wilton
and Cooper 1990).
Among the more promising of the HLA-DR alleles is HLA-DR4, though the results
are hardly conclusive. A study conducted by Kilpatrick et al. found that the
frequency of HLA-DR4 was statistically significantly increased in preeclamptic
women as well as their babies compared to controls (Kilpatrick, Gibson et al. 1990).
The strongest finding was that HLA-DR4 sharing between a preeclamptic mother and
65
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
her fetus was statistically significantly higher than between controls and their babies
(26% vs. 7.7%, respectively). Nonetheless, other studies found no association
(Hayward, Livingstone et al. 1992) or linkage (Wilton and Cooper 1990) between PE
and HLA-DR4.
3.5.4.4 Other HLA. HLA-A, -B and -C genes produce products that stimulate graft
rejection (of fetus by mother or vice versa) and are blocked in placental trophoblast
cells (Hunt and Orr 1992). In a relatively large prospective study (n=712
primigravids), women with the A23/29, B44 and DR7 haplotypes had a statistically
significant increase in incidence of PE (Peterson 1994). Despite these highly
suggestive findings, the frequency of these haplotypes in the general population are
rather low (<5% of whites and <1% of blacks), ruling out the possibility of a major
contribution to disease.
3.5.4.5 Tumor Necrosis Factor Alpha (TNF-a)\ Several lines of evidence support a
role for TNF-a in the development of PE. First, PE has been associated with
increased plasma and amniotic levels of TNF-a and TNF receptors (Kupferminc,
Peaceman et al. 1994; Visser 1994). Second, TNF-a has been detected during
fertilization and is thought to interact with other cytokines in early pregnancy to
promote growth and differentiation and normal placentation (Visser 1994; Dizon-
Townson, Major et al. 1998). Moreover, The blastocyst bears receptors for TNF-a,
suggesting that TNF-a is important in early embryo development (Hunt, Chen et al.
1996). Third, an increased thromboxane-to-prostacyclin ratio has been suggested to
play a role in the pathophysiology of PE. Increased levels of TNF-a may contribute
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
to the increased levels of thromboxane, which then leads to vasoconstriction and
platelet aggregation (Chen, Wilson et al. 1996; Dizon-Townson, Major et al. 1998).
Lastly, the endothelial cell damage observed in PE is thought to be a result of attack
by ROS. TNF-a can not only generate ROS but may also interfere with the buffering
capacity of intracellular components so that the endothelium becomes more
susceptible to oxidant-mediated injury (Chen, Wilson et al. 1996).
A substitution polymorphism at position -308 in the promoter of the TNF-a gene
produces two alleles known as TNFA-1 (G) and TNFA-2 (A). The TNFA-2 allele
disrupts an AP-2 binding site, resulting in 6-7 fold higher levels of reporter gene
transcription in both mitogen-stimulated and unstimulated cells (Wilson 1994).
Moreover, individuals homozygous for the TNFA-2 allele have significantly higher
constitutive and inducible levels of TNF-a secretion compared to TNFA-1
homozygotes, with heterozygous subjects having intermediate levels (Bouma,
Crusius et al. 1996; Stuber, Petersen et al. 1996). Thus, we expect women who are
homozygotes or heterozygotes for TNFA-2 to have increased risk of PE. Reported
allele frequencies for the TNFA-2 allele range from 16% in a Tunisian study
(Chouchane, Ahmed et al. 1997) to 23% in a Dutch study (Bouma, Crusius et al.
1996).
Epidemiological studies have had mixed results, with one study suggesting that the
TNFA-1 allele is increased in PE (Chen, Wilson et al. 1996) while another found no
significant differences in allele frequencies between preeclamptic women and
normotensive gravid controls or published allele frequencies (Dizon-Townson, Major
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
et al. 1998). The study by Chen et al. is limited by its small number of PE cases (n =
14), the lack of exclusions for known underlying conditions (e.g., renal disease,
diabetes, etc.) and the inclusion of multiparas. Moreover, the finding of increased
TNFA-1 frequency is counterintuitive to what would be expected given our
understanding of the mechanisms involved and outlines the need for further research
in this area.
Additionally, a second polymorphism exists at position -238 in the promoter region,
but does not appear to be associated with PE (Lachmeijer, Crusius et al. 2001). The
TNF-a gene is closely linked to the gene for lymphotoxin-a (also called TNF-P). The
two polymorphisms described for the TNF-a gene in combination with two
polymorphisms in the TNF-P gene define five haplotypes, TNF-C, -E, -H, -I and -P
(Bouma, Xia et al. 1996). Lachmeijer et al. found an increased risk of PE or HELLP
associated with having at least one copy of the TNF-I haplotype (Lachmeijer, Crusius
et al. 2001). However, this association was only found among index cases,
particularly index cases with PE only, but not among their sisters, who met similar
disease criteria. Other than suggesting that PE and HELLP are distinct diseases,
these results remain puzzling since any association with a haplotype observed in the
index cases should have also been seen among the similarly afflicted sisters of the
index cases.
3.5.4.6 Interleukin-1 f3 (IL-1P): IL-ip, a cytokine produced by endothelial cells,
monocytes and macrophages is involved in the initiation of a proinflamatory
response. Both IL-ip and interleukin-1 receptor antagonist (IL-1RA) bind to the
68
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
interleukin-1 (IL-1) receptor, the first initiating a proinflamatory response and the
second terminating the inflammatory event. Although circulating levels of IL-1 ( 3 and
IL-1RA have not been consistently elevated in preeclamptics, IL-1 ( 3 expression in
placentas of preeclamptic patients has been found to be upregulated (Rinehart,
Terrone et al. 1999).
Two genes within the interleukin beta cluster have been examined with respect to PE
risk: the genes encoding IL-ip and IL-1RA. Two polymorphisms in the IL-1 [ 3 gene,
one in the promoter at position -511 (C-»T) and another in exon 5 correlate with
altered IL-ip protein expression (Hefler, Tempfer et al. 2001). Additionally, the
second intron of the IL-1RA gene contains a 86 bp repeat polymorphism that varies
between 2-6 repeats, with 4 repeats being the most common (IL-1RA*1) and 2
repeats (IL-1RA*2) being associated with prolonged inflammatory response (Witkin,
Gerber et al. 2002). Women who are homozygous for IL-1RA*2 would be expected
to be at increased risk of PE as would women with the IL-ip promoter variant (-
51 IT) and the IL-1 P exon 5 variant (E2).
One study examined the role of all three polymorphisms in relation to PE risk and
found that none of the polymorphisms appeared to increase the risk of PE (Hefler,
Tempfer et al. 2001). However, the power to detect an effect for these variants was
very low, thus a small to moderate effect would likely go undetected in this study.
Nevertheless, the results did suggest that disease severity might be influenced by
variations in these two genes. Specifically, preeclamptics with the IL-1RA*2 variant
had statistically significantly higher mean systolic blood pressure upon admission
69
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
compared to normotensive pregnant controls (Hefler, Tempfer et al. 2001).
Moreover, when all three polymorphisms were combined, preeclamptic women with
at least three mutant alleles had statistically significantly higher mean systolic blood
pressure than other preeclamptics (Hefler, Tempfer et al. 2001).
3.5.5 Placentation
There are a large number of genes that regulate all aspects of placental growth,
development and function. It has been suggested that these genes are good
candidates for predisposing to PE due to their obvious connection to placental
function (Roberts and Cooper 2001). Studies conducted in mice have shown that
mutations in placentation genes can affect signaling interactions between embryonic
and trophoblast tissues and lead to abnormal vascularization of the placenta (Rossant
and Cross 2001).
Although functional data on the role of specific genes in human placental
development is very limited, what is known appears to correspond well with the
findings in mice. Genes involved in the differentiation and maintenance of cells in
the trophoblast lineage appear to be the best candidates for a role in PE development.
Specifically, human homologues for Mash2, which is involved in the maintenance of
trophoblast stem cells and Handl, which promotes the differentiation of trophoblast
giant cells may prove to be important in human placental development (Rossant and
Cross 2001). Additionally, the chorionic trophoblast cells in Gem 1-deficient mice
fail to undergo differentiation into syncytiotrophoblasts (Anson-Cartwright 2000),
which could theoretically lead to inadequate spiral artery invasion and eventually, PE.
70
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Of these genes, only Hash2, the human homologue to Mash2, is known to be
polymorphic, containing a polymorphism in the 3’ UTR of the gene. To date, no
molecular epidemiology studies have been conducted to examine the role of
polymorphisms in placentation genes and the risk of PE but it is an area of research
that deserves attention.
3.5.6 Other
3.5.6.1 Insulin-like Growth Factor (IGF). Insulin-like growth factor-II (IGF-II) is
involved in mammalian growth and has an important influence on fetal cell division
and differentiation (McCowan and Becroft 1994). Moreover, trophoblast-derived
IGF-II may be important for invasion and for both trophoblast and decidual function
(Irwin, Suen et al. 1999). Beckwith-Wiedemann Syndrome, a condition of prenatal
overgrowth and predisposition to embryonic malignancies, has been associated with
both IGF-n overexpression and severe, early onset PE (McCowan and Becroft 1994).
This has prompted the hypothesis that genetic variation in IGF-II may be involved in
the development of PE by restricting intrauterine fetal growth.
An Apal polymorphism in the 3’ flanking region of the IGF-II gene produces two
alleles, A and B. Allele frequencies for the B allele was approximately 40% among
an Irish population (Bermingham, Jenkins et al. 2000). Bermingham and colleagues
found no association between the IGF-II polymorphism and PE using a mother-
father-child trio design (Bermingham, Jenkins et al. 2000). However, this study
suffers from some methodologic issues, as described earlier in this paper (HLA-G),
which make the significance of these results difficult to interpret.
71
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
In addition to IGF-II, insulin-like growth factor binding protein-1 (IGFBP1) and
insulin-like growth factor-I (IGF-I) have hypothesized etiological roles in PE.
IGFBP1 is thought to inhibit cytotrophoblast invasion into the endometrial stromal
multilayers and IGF-I is thought to be important in fetal growth (Giudice, Martina et
al. 1997). Guidice and colleagues found that in women with severe PE, circulating
levels of IGFBP-1 is significantly elevated, IGF-I is significantly lower and IGF-II is
no different in cases versus premature labor controls. However, several concerns,
including lack of control for potential confounders (e.g., intra-uterine growth
restriction) and the use of a potentially inappropriate comparison group (pre-term
labor controls might not be representative of the base population from which the
cases arose) may compromise the validity of these results. Thus, no definitive
conclusions can be drawn from these data.
3.5.6.2 Mitochondrial DNA. The importance of energy for the transport of nutrients
across the placenta to the fetus has led some to hypothesize a role for reduced
mitochondrial gene expression in the development of PE (Furui, Kurauchi et al.
1994). Many of the features of PE can be explained by mitochondrial dysfunction,
including vasoconstriction, platelet aggregation, disturbed ion transport, reduced
prostaglandin synthesis and hyperuricemia (Torbergsen, Oian et al. 1989). Moreover,
uterine and placental tissues obtained from women with PE showed engorgement in
the endothelial mitochondria with loss of cristae and other mitochondrial changes
suggestive of a systemic metabolic disorder (Shanklin and Sibai 1990).
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
However, the evidence for a mitochondrial contribution to PE development is mixed.
Although one European study found no evidence of reduced respiratory chain
enzyme activities in preeclamptic placentae (Vuorinen, Remes et al. 1998), a very
small study in Japan found a reduced expression of cytochrome c oxidase and
cytochrome oxidase subunit I (both of which are encoded by the mitochondrial
genome) in placental tissue from preeclamptic patients compared to women whose
gestations were appropriate for their gestational age. (Furui, Kurauchi et al. 1994).
Another study examined two Norwegian families with a high incidence of
preeclampsia/eclampsia to determine if there were any mutations in their
mitochondrial DNA (Folgero, Storbakk et al. 1996). They found two different single
nucleotide polymorphisms (SNPs) in these families, but no causal relationship could
be established since a high prevalence of a given polymorphism within affected
families does not necessarily imply causation. Additionally, the authors included
women who had PE/E in a second or later pregnancy, indicating that the etiology of
their PE may be different than primiparous PE. While it is conceivable that
mitochondrial dysfunction plays a role in a small subset of women with PE, it is
unlikely to be involved in the majority of cases.
3.6 DISCUSSION
A common theme in almost all of the epidemiological evidence produced thus far is
methodological problems. A wide range of PE definitions is used, making it very
difficult to compare studies. First, the inclusion of multiparous women in the study
of a disease that is considered to be primarily one of first-time mothers has created
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
heterogeneous study populations, making it difficult to understand the genetic factors
that may be at play. Moreover, multiparous women are more likely to have
developed PE as a result of an underlying medical condition which may or may not
have a genetic basis but which is brought out by the physical demands of pregnancy.
These etiologic differences between parous and nonparous women can and probably
do lead to spurious associations between PE and genes that are actually related to
another disease.
Second, the failure to exclude women with essential hypertension or hypertension
prior to the 20th week of gestation could lead to an inflated odds ratio for genes, such
as AGT, that are known to be associated with hypertension. Therefore, studies of PE
should exclude women with underlying conditions that resemble PE and either
restrict enrollment to nulliparous women or stratify by parity at the analysis stage.
However, it should be noted that stratifying would substantially decrease the sample
size and thus the power to detect an effect within a given stratum.
The importance of fetal genotype, in addition to maternal genotype, is fairly well
recognized, yet the actualization of including fetal genotype assessment has, for the
most part, not yet occurred. One of the barriers to including fetal genotype is that, in
order to look at genotype-by-genotype interactions, a very large sample size is
needed. A large number of small studies simply cannot address the question of
maternal-fetal genotypic interactions and, as such, large studies should be conducted
on multiple genes. The inclusion of non-white participants in these studies is
74
permission of the copyright owner. Further reproduction prohibited without permission.
essential, provided that large enough numbers of minority subjects can be recruited to
provide a meaningful analysis.
With such an extensive list of candidate genes and many others not yet studied, the
challenge now becomes determining which of them hold the most promise for future
research. If our current understanding of the disease process is correct, then the most
promising candidates may be those that are involved in placentation, vascular
remodeling and endothelial cell health. Placentation genes have not been included in
any epidemiologic studies to date and thus deserve special attention. For all of these
genes, studies are needed to better characterize the subpopulations that may be
affected. Some genes that have been associated with PE, such as those in the
thrombophilic pathway (prothrombin, FVL) may characterize a subpopulation at
elevated risk, but do not appear to be independent risk factors for PE. MTHFR may
be an important gene among folate deficient women and therefore should be
considered when examining poor or undernourished populations. Polymorphisms in
LPL and LCHAD are too rare to account for a large number of PE cases, but may be
important in certain subsets of women. Evidence for an association between PE and
genes in the blood pressure regulation pathway is fairly weak, but these genes cannot
be ruled out until an examination of maternal-fetal genotypic interactions has been
completed. Additionally, limited research on the genes for GPIIIa, MMPI, EPHX,
SOD and ApoE have had promising results and suggest that they might be important
in at least some populations. Due to the failure to include fetal genotype in most
molecular epidemiology studies, it seems premature to rule out any of the potential
candidates based on results obtained in maternal samples alone.
75
permission of the copyright owner. Further reproduction prohibited without permission.
In conclusion, more molecular epidemiology studies need to be conducted utilizing
larger sample sizes where genotype-by-genotype interactions can be examined,
especially maternal-fetal genotypic interactions. Additionally, placentation genes
must be added to the list of candidate genes.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
CHAPTER 4: A PILOT STUDY OF NOVEL CANDIDATE GENES FOR
PREECLAMPSIA
The following chapter contains an R21 grant proposal entitled, “A Pilot Study of
Novel Candidate Genes for Preeclampsia.” The grant was funded by the National
Institute of Child Health and Development (NICHD) beginning April 15, 2004.
Several versions of this grant existed prior to the version which appears here.
Initially, we proposed to conduct a candidate gene association study of previously
studied polymorphisms for which the data was the most supportive but not yet
conclusive. Previous candidate genes proposed included platelet glycoprotein Ilia
(GPIHa), matrix metalloproteinase-1 (MMP1), apolipoprotein E (ApoE), epoxide
hydrolase (EPHX), tumor necrosis factor alpha (TNF-«), endothelial nitric oxde
synthase (eNOS), angiotensinogen (AGT), Factor V Leiden mutation (FVL+),
methylenetetrahydrofolate reductase (MTHFR), and monoamine oxidase A (MAO-
A). Of these, only MAO-A was retained in the funded version of the proposal.
It should be noted that several sections of the funded proposal were written by other
co-investigators listed on the proposal. Specifically, the portion of the grant
addressing population stratification was written by Drs. Noah Rosenberg, Sue Ann
Ingles, and David Conti while the sections describing the TaqMan assay and whole
genome amplification were written by Dr. Sue Ann Ingles.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
4.1 SPECIFIC AIMS
Preeclampsia (PE), a form of pregnancy-induced hypertension, is believed to have a
genetic component with both the mother and the fetus contributing to the risk of PE
(Liston and Kilpatrick 1991; Cooper 1993; Dekker and Sibai 1998; Pipkin 1999;
Esplin, Fausett et al. 2001; Roberts and Cooper 2001). Although numerous studies
have been conducted to investigate maternal genetic contributions to PE
susceptibility, most previous studies have failed to consider the role of fetal genotype
(Wilson, Goodwin et al. 2003). Our long-term goal is to conduct a large-scale
candidate gene study to examine the roles of maternal and fetal genotypes and
matemal-fetal gene-gene (GxG) interaction in PE susceptibility. Among the
candidate genes to be examined, we propose to include a promising class of genes
that has been previously overlooked: genes involved in vascularization and
development of the placenta. To date, no placentation genes have been included in
candidate gene studies, and very little is known about polymorphism in these genes.
In order to establish the feasibility of conducting such a study, we must first assess
the extent of polymorphism in the candidate genes of interest. These genes include
several involved in placentation (HAND1, HASH2, GCM1, HIF-la and TGFp3) and
angiogenesis (VEGF, sFLTl, PGF), and one gene involved in utero-placental blood
pressure regulation ( MAO-A). First, we propose to resequence the coding and
regulatory regions of these genes to identify all common variants. Once all common
polymorphic variants have been identified, we will determine allele and haplotype
frequencies and will attempt to generate preliminary data linking polymorphism in
these genes to PE risk. We will accomplish this aim by conducting a pilot case-
permission of the copyright owner. Further reproduction prohibited without permission.
control study of women with a history of PE (during their first pregnancy with a
specific partner) and controls with a normotensive pregnancy history, matched on
maternal age, race, gestational age and year/month of delivery. We intend to identify
and recruit 120 case/child pairs and 240 control/child pairs using the delivery logs
from the Los Angeles County/University of Southern California (LAC+USC)
Women’s and Children’s Hospital.
This case-control study conducted at LAC+USC will also serve as a feasibility study
for conducting a large-scale study, powered to detect GxG interaction, that draws on
the patient populations of the five hospitals (Hollywood Presbyterian, Good
Samaritan, Queen of Angels, White Memorial, and.LAC+USC) covered by the USC
faculty practice. If we are successful at tracing and recruiting women from the highly
mobile, largely immigrant population of LAC+USC, we should also be successful in
the more easily traceable populations covered by the remaining hospitals.
Finally, because the patient population of LAC+USC is predominantly Latino and
because Latinos are an ethnically admixed population, we will also determine the
extent of population statification in this population to assess whether stratification is
likely to lead to biased gene-disease associations. While the likely severity of such
bias has been debated in the epidemiologic literature (Thomas and Witte 2002;
Wacholder, Rothman et al. 2002), there have been no studies to assess the extent of
this problem in Latinos, one of the populations that is most likely to be affected.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Specifically, our aims are as follows:
1. To determine the extent of polymorphism in proposed PE candidate genes
(HAND1, HASH2, GCM1, HIF-lcc, TGFp3, VEGF, sFLTl, PGF, and MAO-A).
la. To identify all common polymorphic variants in the genes,
lb. To determine allele frequencies for all polymorphisms in these genes,
lc. To determine haplotype frequencies for those genes that contain more
than one polymorphic site.
2. To generate preliminary data linking polymorphism in these genes to PE risk, by
testing the following hypotheses:
2a. Maternal genotypes &/or haplotypes influence the risk of PE.
Specifically, mothers who carry a “variant” form of these genes are more
likely to have developed PE and/or have had earlier onset and more severe
disease than mothers who carry the “wild-type” allele.
2b. Fetal genotypes &/or haplotypes influence the risk of PE. Specifically,
mothers who carried a fetus with a “variant” form of these genes are more
likely to have developed PE and/or have had earlier onset and more severe
disease than mothers who carried a fetus with the “wild-type” allele.
3. To establish the feasibility of conducting a large-scale case-control study of PE in
USC-associated hospitals.
3a. To determine recruitment rates among eligible study subjects delivering
at LAC+USC Hospital, the USC-associated hospital likely to have the most
difficult to recruit patient population.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
3b. To estimate the degree of population stratification among Latino patients
at LAC+USC, and thus the extent to which gene-disease associations may be
confounded by ethnic admixture in this population.
4.2 BACKGROUND AND SIGNIFICANCE
4.2.1 Introduction
Preeclampsia (PE), a hypertensive disorder specific to pregnancy, is the leading cause
of maternal and neonatal death and morbidity in developed countries, affecting nearly
6% of all pregnancies (Chesley 1984). This condition predisposes women to other
serious pregnancy complications such as abruptio placentae, acute renal failure,
cerebral hemorrhage, disseminated intravascular coagulation, circulatory collapse
and maternal death (MacKay, Berg et al. 2001). PE accounts for 15-20% of maternal
mortality and 10% of perinatal deaths in developed countries (Cooper 1993; Dekker
and Sibai 1999). In a recent study conducted using data from the Center for Disease
Control’s Division of Reproductive Health, they found that 19.6% of pregnancy-
related deaths to women at 20 weeks or greater gestation that occurred from 1979 to
1992 were from complications of PE or its more severe form, eclampsia (E ). With
rates in the United States on the rise more research is urgently needed to identify
possible genetic and environmental exposures that could predispose a woman to PE.
Disease severity can range widely from a mild disorder associated with transient
hypertension in the latter part of pregnancy to a life-threatening disorder with
convulsions (eclampsia), thrombocytopenia, HELLP (hemolysis, elevated liver
enzymes and low platelets) syndrome, fetal hypoxia and growth retardation
81
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
(Amgrimsson, Hayward et al. 1997). Rates for these more severe sequelae are much
less common than mild disease, affecting approximately 0.56 pregnancies per 1000
deliveries (Saftlas, Olson et al. 1990). Due to the complications associated with
preeclampsia and eclampsia, many more labor inductions, Cesarean sections and
preterm deliveries are performed, resulting in more than $30 million U.S. dollars
being spent annually to treat these conditions (Cooper 1993).
PE is most often defined as in increase in blood pressure to greater than 140/90
mmHg or a 30/15 mmHg increase over baseline on at least two occasions at least 6
hours apart accompanied by significant proteinuria (at least 30 mg/dl on a random
sample or 300 mg in 24 hours) after 20 weeks of gestation and resolving by three
months postpartum (Chesley 1980; Liston and Kilpatrick 1991; Hayward,
Livingstone et al. 1992; Amgrimsson, Hayward et al. 1997). PE is considered to be
primarily a disease of first pregnancies, although having a history of severe or early
PE increases one's risk for a subsequent PE diagnosis (Dekker and Sibai 1999).
The epidemiology of PE has been complicated by the varying definitions and criteria
used to diagnose it. This has lead to flawed studies and the potential for substantially
misclassified disease status (Chesley 1984; Landsbergis and Hatch 1996; Wilson,
Goodwin et al. 2003). Further complicating matters, evidence suggests that PE is
likely to be a two stage process (Ness and Roberts 1996; Roberts 2000), requiring
input from both the fetus (via the placenta) and the mother (via underlying
susceptibility). According to Ness & Roberts (1996), PE can result from poor
placental perfusion (a component that is likely to be hereditary) combined with an
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
underlying maternal condition that may or may not be diagnosed prior to pregnancy
and may or may not have a heredity component. Ideally, a study of genetic
susceptibility to PE would focus on hypertension and proteinuria that is primarily of
placental origin, since it is this form that holds the key to the unique contribution of
pregnancy to this disease.
Despite the large number of experimental and epidemiological studies focusing on
PE, the etiology of this disease remains to be elucidated. Identification of risk
factors, including predisposing genetic variants, could aid substantially in our ability
to identify those women who are at exceptionally high risk of disease. In addition,
knowledge of predisposing genotypes would most certainly advance our
understanding of the etiology and pathogenesis of the disease process. Such
knowledge would likely facilitate the development of new prophylactic and/or
therapeutic interventions and lead to an improvement in maternal and perinatal
outcomes (Cooper 1993). Since PE affects a substantial proportion of pregnant
women, overall public health would surely benefit from any treatments derived and
from a greater understanding of this disease.
4.2.2 Preeclampsia as a Genetic Disease
It is the potential for a genetic component to PE that is the focus of this proposal.
Evidence for a genetic component comes from the observation that there is a marked
increase in preeclampsia among mothers, daughters, sisters and granddaughters of
women who have had preeclampsia, but not in women related through marriage (i.e.,
in-laws) (Amgrimsson, Bjomsson et al. 1990; Widschwendter, Schrocksnadel et al.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
1998; Morgan and Ward 1999). Some studies have found that the increased risk is
greatest for the daughters of a preeclamptic pregnancy, greater even than that of her
sister bom of a normal pregnancy (Widschwendter, Schrocksnadel et al. 1998) while
others have found a similar increase in risk for all daughters bom to a mother with a
history of PE, (Amgrimsson, Bjomsson et al. 1990; Mogren, Hogberg et al. 1999).
Additionally, higher concordance rates among maternal monozygotic twins compared
to dizygotic twins suggest a role for genetics in the development of disease (Ros
2000). However, the observation that a high number of maternal monozygotic twin
sets are discordant suggests that the fetal genotype, as well as environmental factors,
may also be important in determining susceptibility (Cincotta and Brennecke 1998).
There are several other lines of evidence that suggest a fetal (paternal) component to
PE susceptibility. For instance, the association between PE and fetal chromosomal
abnormalities supports a fetal contribution to etiology (Dekker and Sibai 1999) as
does the observation that PE risk is increased in women with hydatidiform moles,
which arise from entirely paternal origin (Goldstein and Berkowitz 1994). In
addition, the small but statistically significant increase in incidence among daughters-
in-law of index cases (Amgrimsson, Bjomsson et al. 1990) and the observation that
men bom of preeclamptic pregnancies were more likely to father a preeclamptic
pregnancy than their matched controls (OR = 2.1, Cl: 1.0-4.3) (Esplin, Fausett et al.
2001) also support the idea of a fetal contribution to risk. Lastly, multiparous women
who change partners are at increased risk of PE, especially if their new partner is
known to have fathered a preeclamptic pregnancy in another woman (Cooper 1993).
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
While the role of genetic factors in the etiology of PE is widely accepted, the mode of
inheritance is still the subject of vigorous debate. Some researchers have suggested
that susceptibility to PE could be inherited via a single, autosomal recessive gene
(Cooper and Liston 1979; Chesley 1986; Ros 2000) or a dominant gene with
incomplete penetrance (Dekker and Sibai 1999). Others believe that it is more likely
due to complex interactions between two or more genes along with environmental
factors (Amgrimsson, Hayward et al. 1997; Mogren, Hogberg et al. 1999; Pipkin
1999; Walker 2000). Still others believe that susceptibility to PE is due to an
interaction between the maternal and fetal genotypes (Lie 1998; Kilpatrick 1999) or
matemal-fetal allele sharing (Liston and Kilpatrick 1991). The most natural genetic
model for a condition that only arises during pregnancy is one in which both the
maternal and fetal genes play a role (Cooper 1993). Specifically, it has been
suggested that the PE phenotype is due to a matemal-fetal genotype-by-genotype
interaction either at the same locus or at separate ones. It is probable that
susceptibility to PE is due to one or more genes, acting in both the mother and her
fetus, modified by various environmental factors (Morgan, Crawshaw et al. 1999).
Since most genetic studies of preeclampsia to date have concentrated on the maternal
genotype (Walker 2000; Esplin, Fausett et al. 2001) more research needs to be
conducted where both the maternal and fetal genotypes are considered.
Although linkage studies are generally useful for the identification of major genes
(i.e., genes which have a large effect on disease), association studies which utilize the
candidate gene approach are much more powerful for studying genes that may have a
modest effect on disease risk (Risch and Merikangas 1996), as is thought to be the
85
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
case in PE. Additionally, linkage and allele-sharing methodologies are not especially
useful in determining if there is a fetal genetic component to PE development since
incidence is greatly decreased in second and later pregnancies (Wilton 1995; Roberts
and Cooper 2001). Moreover, the fact that only females who have reproduced will
have a known phenotype makes traditional genetic analyses difficult. Lastly, the
difficulty in locating a large family in which diagnosis of PE is consistent across
family members would be a slow and expensive task, suggesting that the candidate
gene approach is preferable to linkage analysis for the study of PE (Roberts and
Cooper 2001). Thus, we propose a pilot study to determine the feasibility of
conducting a large-scale association (case-control) study of the effects of selected
maternal and fetal candidate genes on the risk of PE.
4.2.3 Population stratification in candidate gene studies
The case-control design, while possessing many advantages, is potentially susceptible
to a form of confounding known in the genetics literature as "population
stratification" (Khoury and Beaty 1994; Lander and Schork 1994; Ewens and
Spielman 1995; Altshuler, Kruglyak et al. 1998; Khoury 1998; Witte, Gauderman et
al. 1999). Confounding occurs when an observed association between disease status
and a specific allele of a candidate gene is not causal. Rather, an unidentified causal
factor is associated with ethnicity, which in turn is associated with candidate gene
allele frequencies. If ethncity is also associated with disease incidence, then false
associations can be generated. Some critics of candidate gene association studies
contend that since ethnicity can never be adequately measured and adjusted for, the
case-control design should not be used for genetic studies. Others have shown that,
86
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
at least in Caucasian populations of mixed European heritage, population
stratification is not likely to be a serious problem (Wacholder, Rothman et al. 2000).
Although the seriousness of the population stratification problem has been the subject
of vigorous debate in the epidemiologic literature (Thomas and Witte 2002;
Wacholder, Rotherman et al. 2002), it is generally accepted that the potential for bias
is greatest when both disease prevalence and allele frequencies differ greatly among
population strata (Wacholder, Rothman et al. 2000). The Latino population, being an
admixture of European, Amerindian, and African populations is one population that
is potentially most susceptible to confounding by stratification. To our knowledge,
no one has generated empirical data to assess the severity of this problem in a Latino
population.
4.2.4 Susceptibility Genes
4.2.4.1 Placentation genes
The failure to form a fully functional placenta may result in a range of pathologies,
including recurrent miscarriage and PE (Amgrimsson, Connor et al. 1994; Knofler,
Kalionis et al. 2000). Nearly all aspects of placental development are believed to be
genetically determined (Knofler, Kalionis et al. 2000). Therefore, genes responsible
for placental development are obvious candidates for susceptibility to a disease such
as PE, which is widely believed to result from abnormal placentation (Amgrimsson,
Connor et al. 1994; Lyall and Greer 1994). However, very little is currently known
about which genes are critical for normal human placentation and, for the few that
have been identified, almost nothing is known about variation in these genes.
87
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Most of what is known about placental function and the genes that regulate it comes
from observations made in mice. Although the superficial placental structure (Flint
1994) and terminology used to describe placental anatomy vary greatly among
different types of mammals, many aspects are conserved between different
mammalian species (Cross, Baczyk et al. 2003). For example, Cross et al. describe
that in all species the trophoblast performs the same basic functions: it generates a
large surface area for the exchange of nutrients (the ‘transport and barrier trophoblast
function’), and it invades the uterus, subsequently producing growth factors,
cytokines and hormones that increase the blood flow and nourishment to the
developing fetus and placenta ( the ‘invasive and endocrine trophoblast function’).
Since trophoblasts are functionally similar across many different species of
mammals, data obtained regarding the genetic determinants of trophoblast
differentiation and invasion in mice should be largely extractable to humans as well
(Cross, Baczyk et al. 2003). Thus, the human homologues of genes that regulate
trophoblast function in mice are good candidates for study in humans since it is the
trophoblasts that are responsible for adequate uterine invasion and subsequent normal
placental development (Rinkenberger and Werb 2000).
Placentas of mice differ from human placentas in several structural aspects (Cross,
Baczyk et al. 2003). We briefly review these differences here to facilitate
comparisons between analogous structures in our later discussions. First, the
structure called the chorionic villi in humans, containing the synctiotrophoblasts,
villous cytotrophoblasts, stroma and blood vessels has a somewhat different
appearance in mice where it is called the labyrinth. The chorionic villi (in humans)
88
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
and labyrinth (in mice) provide organization for the ‘transport and barrier
trophoblast’ subpopulation. The ‘invasive and endocrine’ trophoblast subpopulation
also exhibits some differences in humans when compared to mice. For instance,
while in mice, the trophoblast giant cells and glycogen trophoblast cells migrate as
much as several hundred microns into the uterus, the extravillous cytotrophoblast
cells, the analogous cell type in humans, migrate a much greater distance into the
endometrium. The spongiotrophoblast is a layer of proliferated trophoblasts and is
present only in mice.
Since placentation genes are an intuitive choice and since none of these genes have
yet been studied in relation to PE, we propose to study a handful of novel candidate
genes, selected on the strength of the existing evidence in humans, to determine
whether they deserve further attention in a future large case-control study.
Please note that throughout this proposal, genes that are found in mice will be
represented by lower case letters except for the initial letter (e.g., Handl) while genes
found in humans will appear in all upper case letters (e.g., HAND1).
4.2.4.1.1 Proliferation vs. differentiation: HAND1 and HASH2
Murine placental development is known to be regulated through the basic helix-loop-
helix (bHLH) family of transcription factors that includes Mash2 and Handl (Cross,
Baczyk et al. 2003), both which are essential in the regulation and development of
the placenta in mice (Scott, Anson-Cartwright et al. 2000). Mash2 and Handl have
opposing functions in the mouse placenta, with Mash2 favoring trophoblast
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
proliferation and Handl promoting differentiation toward trophoblast giant cells (the
invasive cell type) (Kraut, Snider et al. 1998; Cross, Baczyk et al. 2003). Thus, both
Mash2 and Handl are involved in maintaining the delicate balance between
proliferation versus differentiation to trophoblast giant cells (Scott, Anson-Cartwright
et al. 2000).
Placental expression of Handl is limited to the trophoblasts (Riley, Anson-Cartwright
et al. 1998). Handl knockout mice exhibit an upregulation of Mash2 and a resulting
decrease in the numbers of differentiated trophoblast giant cells (the invading type of
trophoblast) (Riley, Anson-Cartwright et al. 1998), thereby impeding uterine
invasion. Thus, Handl is critical to the development of the murine placenta (Knofler,
Meinhardt et al. 1998; Riley, Anson-Cartwright et al. 1998). HAND1, the human
homologue to Handl, shares 93% overall sequence identity with its counterpart in the
mouse. Knofler and colleagues found production of HAND 1 mRNA and polypeptide
in the trophoectodermal cell layer and concluded that it may play an important role in
differentiation of the amniotic membrane and pre-implanting trophoblast (Knofler,
Meinhardt et al. 2002). Furthermore, HAND1 is expressed in large quantities in
cytotrophoblastic tumor cells, suggesting a role in early trophoblast differentiation
(Knofler, Meinhardt et al. 1998; Vasicek, Meinhardt et al. 2003). Thus, HAND1
appears to play an important role in human placental development.
Mash2, the gene favoring proliferation, is expressed in the chorion, ectoplacental
cone and later, in the spongiotrophoblast (Cross, Baczyk et al. 2003). Mash2
knockout mice fail to form a spongiotrophoblast (proliferated layer) which then
90
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
results in the impairment of a critical step in placental development, labyrinth
formation (Guillemot, Nagy et al. 1994; Tanaka, Gertsenstein et al. 1997). HASH2,
the human homologue, is expressed in only the extravillous (invasive type)
trophoblasts of the developing placenta (Alders, Hodges et al. 1997), suggesting an
important role for this gene in human placentation.
4.2.4.1.2 Syncytiotrophoblast Formation: GCMI
The transcription factor glial cells missing 1 (Gcml) is the only gene thus far that has
been shown to be necessary for syncytiotrophoblast formation (Anson-Cartwright
2000; Cross, Baczyk et al. 2003). In fact, mice lacking Gcml were unable to form
the labyrinth, where Gcml is expressed, and died in utero of placental insufficiency
(Schreiber, Riethmacher-Sonnenberg et al. 2000). In all mammals, the placenta is the
only known site of Gcml expression (Basyuk, Cross et al. 1999), strongly suggesting
that it plays an important part in mammalian placentation.
In humans, GCMI is expressed in the chorionic villi, the analogous structure to the
labyrinth, and abnormal development of the chorionic villi has been found to be
associated with fetal death and IUGR in humans (Sagol, Sagol et al. 2002; Bane and
Gillan 2003). Moreover, as human cytotrophoblasts differentiate into invasive
extravillous cytotrophoblasts, they express high levels of GCMI in vitro (Janatpour,
Utset et al. 1999), suggesting that GCMI is an important determinant of the invasive
phenotype. Lastly, evidence that GCMI regulatory networks engage in cross talk
with bHLH transcription factors further supports a role for GCMI in human placental
development.
91
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
4.2.4.1.3 Hypoxia and Trophoblast Invasion: HIF-la and TGF03
Early placentation occurs in an environment that is low in oxygen (0 2 ). This hypoxic
environment favors trophoblast proliferation whereas high levels of 0 2 favor
differentiation toward an invasive phenotype (Genbacev, Joslin et al. 1996;
Genbacev, Zhou et al. 1997). This change from low to high 0 2 tension is essential for
proper placental development. An overabundance of immature trophoblast cells
(characteristic of a low 0 2 environment) have been found in preeclamptic placentae
(Redline and Patterson 1995). The failure of trophoblast cells to take on an invasive
phenotype and adequately invade the myometrium suggests that genes involved in
hypoxia-induced regulation of trophoblast differentiation may be important
susceptibility factors in PE development.
Hypoxia-inducible factor-1 (HIF-1) is a bHLH transcription factor that is found in
mammalian cells cultured under hypoxic conditions. There are two subunits of this
gene, alpha and beta, with the oxygen-regulated alpha subunit controlling HIF-1
activity. HIF-1 is expressed by trophoblast cells and regulates the cellular response
to hypoxia (Seligman, Nishiwaki et al. 1997), including the regulation of genes
involved in angiogenesis (e.g., VEGF). Transforming growth factor type P-3
(TGFP3), an inhibitor of trophoblast differentiation, parallels HIF-1 expression. Both
of these proteins remain at high levels in the early hypoxic plactental environment,
decreasing once 0 2 tension increases after about 10 weeks of gestation.
One recent experiment found that villous explants from women with PE do not
adequately downregulate HIF-1 protein expression when 0 2 levels increase
92
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
(Rajakumar, Doty et al. 2003). Likewise, TGF03 expression is abnormally high in
women with PE . Both of these have the effect of inhibiting differentiation of
trophoblasts to the invasive phenotype. Caniggia et al. found that downregulation of
HIF-1 a expression in hypoxic placental explants resulted in inhibition of TGF03
expression and restored the invasive capacity of preeclamptic trophoblasts (Caniggia,
Mostachfi et al. 2000). These findings led Caniggia and colleagues to speculate that
if 0 2 tension does not increase or the trophoblasts fail to respond to this increase,
HIF-1 a and TGFP3 expression remains high, trophoblasts do not differentiate to
invasive phenotype and invasion is shallow, leading to PE. Polymorphisms in one or
both of these genes could explain why the expression fails to downregulate in
response to increased 0 2 tension and thus, could predispose to PE.
4.2.4.2 Angiogenesis and endothelial cell maintenance: VEGF, PGF & sFLTl
Genes involved in the development of maternal and placental vasculature are
excellent candidates for study since deficient vascularization and subsequent
placental ischemia are believed to be an important part of the PE etiologic pathway.
A recent study has implicated vascular endothelial growth factor (VEGF), placental
growth factor (PGF), and their receptor, soluble fms-like tyrosine kinase 1 (sFLTl),
in the etiology of PE (Maynard, Min et al. 2003). Maynard et al. found that women
with PE exhibit substantially higher levels of sFltl and lower levels of free VEGF
and PGF in maternal serum than women with a normotensive pregnancy (Maynard,
Min et al. 2003).
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
VEGF and PGF belong to a family of regulatory peptides which controls blood vessel
formation in, among other places, the placenta (Reynolds, Grazul-Bilska et al. 2002).
VEGF is a well-known promoter of angiogenesis and its expression is upregulated
under hypoxic conditions (Millauer, Shawver et al. 1994), such as those favored by
early placentation. Whereas VEGF expression can be found in many tissues, PGF
expression is virtually absent in the nonpregnant state. Pgf deficient mice were found
to have impaired angiogenesis (Carmeliet 2001), however, very little is known about
PGF outside of its importance in blood vessel formation during pregnancy.
sFLTl is expressed in the normal human placenta (Clark, Smith et al. 1998) but
expression is increased in preeclamptic placentas, with expression returning to more
normal levels within 48 hours of delivery (Maynard, Min et al. 2003). Increased
levels of sFLT have also been detected at higher levels in amniotic fluid of women
with PE compared to healthy pregnant controls (Vuorela, Helske et al. 2000). When
maternal blood from preeclamptics was assayed, in vitro blood vessel formation was
severely hampered due to high levels of sFLTl and restored when VEGF and PGF
were added to the serum (Maynard, Min et al. 2003). These findings suggest that the
antiangiogenic properties of preeclamptic serum is likely due to the effect of sFLTl
in blocking normal VEGF and PGF activity. VEGF and PGF are also responsible for
inducing vasodilation in vitro, an action that was effectively blocked by sFLTl
(Maynard, Min et al. 2003).
sFLTl is the soluble form of the receptor, FLT1, that mops up VEGF and PGF in the
circulation. Low maintenance levels of VEGF and PGF are needed to maintain
94
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
normal endothelial cell function in the adult. If circulating levels fall too low,
endothelial cell dysfunction can result. Elevated circulating levels of sFLTl have
been shown to lead to widespread endothelial cell dysfunction (Maynard, Min et al.
2003), a hallmark of PE. Some of the typical symptoms of PE - elevated blood
pressure and proteinuria - may also be explained by decreased VEGF (Maynard, Min
et al. 2003) since VEGF has been implicated in glomerular endothelial repair
(Ostendorf, Kunter et al. 1999) and VEGF inhibitors have been shown to cause
hypertension and proteinuria in humans (Yang 2002). Lastly, one study has been
able to reproduce (in mice) the typical renal lesions seen in preeclamptic women by
producing mice lacking one VEGF-A allele in renal podocytes (Eremina, Sood et al.
2003). Maynard and colleagues were also able to induce hypertension, proteinuria
and renal lesions in rats, by administering sFLTl, even at doses similar to those
experienced by women with PE (Maynard, Min et al. 2003). These findings are
specific to sFLT since the alternate VEGF and PGF receptor, soluble fetal liver
kinase 1 (sFLKl) failed to produce similar results (Maynard, Min et al. 2003).
VEGF and sFLT can also explain one of the most puzzling aspects of PE to date, the
fact that smoking decreases the risk of developing disease. Smoking has been found
to increase endothelial VEGF expression (Conklin, Zhao et al. 2002) and sFLTl has
been shown to be lower in smokers compared to nonsmokers (Belgore, Lip et al.
2000). Taken together, these two findings suggest that unbound VEGF levels in
smokers would be higher than in nonsmokers.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
4.2.4.3 Utero-placental blood pressure regulation: MAO-A
MAO-A is responsible for catalyzing the oxidative deamination of a number of
biogenic amines (Shih, Chen et al. 1999). In particular, MAO-A degrades serotonin,
epinephrine and norepinephrine and is predominantly expressed in the human
placenta and fibroblasts (Shih, Grimsby et al. 1993). Serotonin plays an important
role in pregnancy, regulating vasodilation in umbilical arteries (Gujrati, Shanker et al.
1996; Carrasco, Cruz et al. 1998; Haugen and Moe 1999). In fact, serotonin is a
more powerful vasoconstrictor of umbilical arteries in vitro than either angiotensin II
or epinephrine and has been shown to augment the vasoconstriction produced by the
latter (Weiner 1987). Placental monoamine oxidase A (MAO-A) plays a major role
in the regulation and metabolism of fetal serotonin (Gujrati, Shanker et al. 1985). It
has been suggested that the fetus may contribute to the observed increases in
serotonin levels in maternal blood and thus, both maternal and fetal MAO-A
genotype may be an important determinant of maternal plasma serotonin levels.
Decreased MAO-A activity in both the placenta (fetus) and the mother, and the
subsequent increase in serotonin levels may lead to vasoconstriction of placental
vessels and platelet aggregation, thereby impeding normal blood flow to the placenta
and fetus and resulting in PE (Carrasco, Cruz et al. 2000).
Additionally, platelets, the cells responsible for the uptake and transport of serotonin,
are excessively activated in preeclamptic pregnancies compared to normal
pregnancies (Carrasco, Cruz et al. 1998). As a result, the platelets tend to aggregate
and adhere to the endothelium, causing the release of serotonin. If the serotonin is
not degraded at an optimal rate (via MAO-A), it could induce fibrin deposits, which
96
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
then occlude maternal blood flow and lead to placental infarction (Whigham, Howie
et al. 1978; Carrasco, Cruz et al. 1998). The resultant vasoconstriction and ischemia
associated with increased serotonin are hallmarks of preeclampsia/eclampsia. In fact,
many of the clinical features of PE can be explained by the serotonin hypothesis
(Gujrati, Shanker et al. 1985).
The MAO-A gene is located on the X chromosome (Shih, Grimsby et al. 1993). The
promoter activity of MAO-A is contained in two 90 bp repeats (Zhu, Grimsby et al.
1992) and a functional polymorphism has been identified, consisting of a 30-bp
repeated sequence which can be present in either 2 (rare), 3, 3.5, 4 or 5 copies (Sabol,
Hu et al. 1998). In general, longer repeats (> 3 repeats) have higher promoter activity
and therefore greater MAO-A activity (Deckert, Catalano et al. 1999) while short
repeats (< 3 repeats) have decreased MAO-A activity (Denney, Koch et al. 1999),
leading to increased serotonin levels and possibly an increased risk of PE.
Abundant evidence exists that serotonin and MAO-A might play a role in the
development of PE. First, a state of hyperserotonemia has been associated with
preeclampsia and eclampsia (Gujrati, Shanker et al. 1985; Gujrati, Shanker et al.
1996), suggesting that either too much serotonin is being produced or not enough
MAO-A is produced to degrade the serotonin. Gujrati et. al. found that, compared to
normal placentas, preeclamptic/eclamptic placentas had statistically significantly
higher placental serotonin levels and lower monoamine oxidase activity (Gujrati,
Shanker et al. 1985; Gujrati, Shanker et al. 1996). Both the increase in serotonin and
the decrease in monoamine oxidase activity are correlated with systolic blood
97
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
pressure (Gujrati, Shanker et al. 1996; Carrasco, Cruz et al. 1998). Gujrati and
colleagues also found that severity of PE increases with increasing placental
serotonin and decreasing placental MAO-A activity (Gujrati, Shanker et al. 1996).
Similar increases in maternal and/or cord blood serotonin levels and decreases in
monoamine oxidase activity were found among women with PE (defined as
hypertension plus proteinuria and/or edema) (Gonzalez 1990; Gujrati, Shanker et al.
1996; Middelkoop, Dekker et al. 1993; Carrasco, Cruz et al. 1998). Furthermore,
placental serotonin metabolism has been found to be significantly higher in normal
pregnancies than in PE. Specifically, after 60 minutes of incubation with serotonin,
placental homogenates from severely preeclamptic pregnancies failed to metabolize
more than 94% of the serotonin, compared to 15% remaining unmetabolized in
homogenates from normotensive pregnancies (Carrasco, Cruz et al. 2000). Carrasco,
concluded that the higher serotonin levels observed in severe PE are primarily a
result of a reduction in MAO-A activity rather than a limited rate of serotonin uptake
into the cells.
Second, ketanserin, a selective blocker of serotonin, has been shown to dramatically
eliminate the hypertension associated with PE, while having little effect on the
arterial pressure in normotensive patients (Weiner 1987; Denney, Koch et al. 1999).
This suggests that ketanserin is acting on the preeclamptic component of the
hypertension (serotonin). Likewise, cyproheptadine, a serotonin antagonist, has been
shown to improve the clinical symptoms of PE (Chandravati 1979).
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Third, serotonin has been shown to cause lesions in the kidneys and placentae of
animal models which are similar to those commonly seen in women suffering from
severe PE or eclampsia (Weiner 1987; Poulson 1963). Prior administration of
serotonin antagonists can prevent the development of these lesions (Poulson 1963).
It is possible that decreased MAO-A activity and increased levels of serotonin are
merely a result of the disease, rather than a contributing factor to its development.
Prospective data or an association study of MAO-A genotype, which is not
modifiable by disease diagnosis, could help to distinguish between these possibilities.
4.2.5 Known Risk Factors
Numerous risk factors have been identified that increase the risk of developing PE.
Well established risk factors include: prepregnancy obesity, primiparity
(primipatemity), family history of PE, black race, Hispanic ethnicity, multiple
gestation pregnancies and a personal history of PE, diabetes, chronic hypertension,
vascular disorders or renal disease. Other possible risk factors include: employment
during pregnancy/occupational stress, urinary tract infection, <16 years of education,
alcohol, caffeine and drug use during pregnancy, marital status, ABO blood group
incompatibility, maternal blood group 0 , fewer than three prenatal care visits, first
prenatal visit after 28 weeks, maternal age < 20, sickle cell trait/disease, lupus, lipid
abnormalities, elevated homocysteine concentrations, donor sperm or eggs, and
depression and/or anxiety in early pregnancy. Protective factors may include
smoking, oral sex, longer cohabitation before index pregnancy, non-barrier methods
99
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
of birth control and a history of a previous abortion or miscarriage with the same
father as the index pregnancy.
4.3. PRELIMINARY STUDIES
4.3.1 The investigative team
4.3.1.1 Dr. Sue Ingles is Assistant Professor of Preventive Medicine, Co-Director of
the USC Program in Molecular Epidemiology and Director of the Molecular
Epidemiology Laboratory in the Norris Comprehensive Cancer Center. She has
formal training and experience in Epidemiology, Biostatistics, and Laboratory
Sciences. Her research has been focused primarily on susceptibility genes that
mediate the effects of nutritional and hormonal exposures in cancer etiology.
Additionally, her laboratory collaborates with a number of investigators in analyzing
a wide array of other genetic polymorphisms.
4.3.1.2 Dr. T. Murphy Goodwin is Associate Professor of Obstetrics and
Gynecology, Director of Maternal & Fetal Medicine at USC and Co-Director of the
Institute for Maternal Fetal Health at USC and Childrens Hospital Los Angeles. He
has published widely in the area of medical complications of pregnancy with an
emphasis on both maternal and fetal outcomes.
4.3.1.3 Dr. Duncan Thomas is Professor in the Department of Preventive Medicine,
Director of the Biostatistics Division, and holder of the Verna Richter Chair in
Cancer Research. His research has been primarily devoted to the development of
statistical methods for environmental and genetic epidemiology, with a wide range of
100
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
applications, primarily in cancer etiology. He is a former President of the
International Genetic Epidemiology Society and author of Statistical Methods in
Genetic Epidemiology soon to be published by Oxford University Press.
4.3.1.4 Dr. David Conti is Assistant Professor in the Division of Biostatistics and
was awarded the 2001 Roger Williams award from the International Genetic
Epidemiology Society. His particular interests relevant to this application include
analysis of candidate gene studies and statistical methods for detecting and correcting
for population stratification.
4.3.1.5 Dr. Noah Rosenberg is a Research Associate in the Program in Molecular
and Computational Biology and an NSF Postdoctoral Fellow in Biological
Informatics. His research focuses on theoretical population genetics, with emphasis
on the effects of human population history and structure on the design of association
studies. He is the lead author on a paper entitled "Genetic structure of human
populations" (see Appendix B), which illustrates how panels of microsatellite
markers can be used to identify genetic clustering within populations
4.3.1.6 Dr. Jean C. Shih is a University Professor at the USC Keck School of
Medicine with appointments in the departments of Pharmacy and Cell and
Neurobiology and is holder of the Boyd and Welin Chair in Molecular Pharmacology
and Toxicology. Her research is focused on monoamine oxidases (including MAO-A
which is of interest in this application), their transcriptional regulation, structure and
functions and their clinical implications.
101
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
4.3.1.7 Melissa Wilson is a Ph.D. candidate in Epidemiology whose dissertation
focus has been on the molecular epidemiology of preeclampsia. She is the lead
author on a comprehensive review paper (see Chapter 3) entitled "Molecular
Epidemiology of Preeclampsia" and is an experienced study manager, having worked
in that area for five years prior to returning to graduate school.
4.3.2 Number of Potential Participants
In order to estimate the number of eligible patients that can be recruited for our pilot
study, we performed a search of the LAC + USC Women’s and Children’s Hospital
delivery log database. Among other things, the delivery log includes information on
age, parity, labor and delivery characteristics, antepartum and intrapartum procedures
and medications, maternal medical problems and obstetric and fetal complications.
We identified all cases of mild PE, severe PE, eclampsia or HELLP, excluding cases
also diagnosed with diabetes, chronic hypertension, renal disease and lupus.
During the four year period January 1,1999 to December 31,2002, there were
593 women diagnosed with PE (and meeting the above exclusions) out of 8212
deliveries (7.2%). Assuming a 50% participation rate, and an estimated 1667
deliveries per year (based on 2002 deliveries), we should be able to recruit 60
participants per year, or 120 over two years.
4.3.3 Tracing of eligible cases
To determine whether women from the LAC + USC population can be located
retrospectively after delivery, we attempted to trace 196 women, using cost-free
internet resources, who were diagnosed between January 1,1999 and December 31,
102
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
2001. Women whom we were not able to contact via telephone or letter were
considered unlocatable. The numbers of unbeatable women were: 39/108 (36%),
20/49 (41%), and 12/39 (31%) for those diagnosed in 1999, 2000, and 2001,
respectively. Since we could not locate more than 30% of women diagnosed only 1.5
years ago, we now propose to contact elligible subjects within 1 month of giving
birth.
4.3.4 Buccal cell DNA extraction and whole genome amplification
To insure that we have sufficient infant DNA to genotype multiple markers, we will
use whole genome amplification of buccal DNA samples. For more than 500 buccal
samples extracted in our laboratory, median DNA yield from two swabs was 1.4
micrograms. Using the Repli-G kit from Molecular Staging (New Haven, CT) to
perform whole genome amplification, and starting with 50 ng of DNA in a 100
micro liter reaction, typical yield of amplified DNA was 10-20 micrograms. When
scaling down to a 10 micro liter reaction, we obtain a proportional amount of
amplified DNA. Therefore, even on samples with low yield, we should be able to
obtain at least 1 microgram of amplified DNA, which is sufficient to carry out 50-100
PCR reactions. Amplified DNA has been tested on a SNP panel and a microsatellite
panel, and the amplified DNA performed one-to-one with the unamplified DNA.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
4.4. RESEARCH DESIGN AND METHODS
4.4.1 Epidemiologic Methods
4.4.1.1 Rationale and Epidemiologic Study Design
Our long-term goal is to conduct a large-scale candidate gene study to examine the
roles of maternal and fetal genotypes and matemal-fetal gene-gene (GxG) interaction
in PE susceptibility. To study the complex interactions among risk factors, including
GxG interactions thought to be important in PE, a large sample will be needed. Thus,
in this proposal, we will collect pilot data to support a future large-scale study.
The target population of the future large-scale study will be patients delivering at any
one of five USC associated hospitals (Hollywood Presbyterian, Good Samaritan,
Queen of Angels, White Memorial, and LAC+USC). We have chosen to base our
pilot study in the LAC+USC Hospital which serves a highly mobile immigrant
popoulation. If we are successful in tracing and recruiting women from this
population, we should also be successful in the more easily traceable populations
covered by the remaining hospitals. In addition, because the patient population of
LAC+USC is predominantly Latino this pilot study will allow us to determine the
extent of population statification, and the extent to which stratification is likely to
lead to biased gene-disease associations in the Latino population. While it has been
argued that stratification is not a serious problem in white populations of mixed
European heritage (Wacholder, Rothman et al. 2000), there have been no studies to
assess the extent of this problem in Latino populations which, due to ethnic
admixture, are more likely to be affected.
104
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
In this pilot case-control study, we propose to enroll women with a history of PE and
their matched controls. Maternal susceptibility genotypes will be determined via
blood samples obtained during a post-natal interview while fetal susceptibility
genotypes will be determined via buccal swabs from the infant obtained during the
interview. Risk factors will also be assessed during the interview via questionnaire.
4.4.1.2 Study Population
The source population for the proposed study will include all women delivering at the
LAC + USC Women’s and Children’s Hospital during the two year study period. To
identify women diagnosed with PE, delivery logs will be searched on a weekly basis
so that cases can be contacted within one month of delivery. We will include all
cases who meet eligibility criteria described below and who agree to participate. The
population is expected to be at least 85% Latino. The remaining 15% will be
comprised of Caucasians, Asians and African-Americans.
4.4.1.3 Eligibility Criteria
All pregnant women who delivered single babies at the LAC + USC Women’s and
Children’s Hospital and are between the ages of 18 and 40 at the time of delivery,
had at least one prenatal visit prior to 20 weeks gestation and who are either
nulliparous or are having their first child with a new partner are potentially eligible to
participate. Cases must also have had a PE diagnosis made after 20 weeks gestation.
Cases and controls diagnosed with any of the following predisposing conditions will
be excluded: history of chronic hypertension or hypertension (greater than 140/90
mmHg) diagnosed prior to the 20th week of gestation, diabetes (type I, type II and
105
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
gestational), chronic renal disease, lupus, sickle cell disease/trait, fetal malformations
or chromosome abnormalities for the index pregnancy and positive drug screen
during the index pregnancy.
4.4.1.4 Selection of Cases
Cases will be those women meeting the following criteria: (1) have at least two blood
pressure readings (at least 6 hours apart) of 140/90 mmHg or an increase in blood
pressure of 30/15 mmHg over baseline, (2) have proteinuria greater than 300 mg/24
hours or greater than 100 mg/dl on two random specimens, and (3) symptoms must
resolve by 12 weeks after delivery.
Cases will be ascertained and recruited as follows. (1) The database containing
information completed by the delivering obstetrician (delivery logs) will be queried
on a weekly basis to determine the names of women who received a diagnosis of
mild PE, severe PE, eclampsia or HELLP during delivery but excluding women also
diagnosed (on delivery logs) with diabetes, chronic hypertension, renal disease or
lupus. (2) Within one month of delivery, eligible cases will receive a brochure
describing the study and a letter inviting them to participate from Dr. T. Murphy
Goodwin. (3) Letters will be followed by a telephone call from the interviewer
approximately one week later. (4) For those who agree to participate, a chart
abstraction will be made to verify eligibility (see eligibility criteria above).
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
4.4.1.5 Selection of Controls
Two controls will be selected for each case. Controls will be women who were not
diagnosed with PE during the index pregnancy and will be selected from the same
population that gave rise to the cases. Controls will be identified using the delivery
logs and will be matched to cases on the basis of maternal age (± 5 years), gestational
age at case presentation (± 2 weeks), race and year/month of delivery. Controls will
be the first two normotensive deliveries at the LAC + USC Women’s and Children’s
Hospital after the case delivery that meet the matching criteria. Controls will be
recruited in the same manner as cases (see 2-4 in section 4.4.1.4 above).
4.4.1.6 HIPAA Considerations
To address the recent enactment of legislation restricting access to protected health
information (PHI), we have applied for a waiver from the Institutional Review Board,
requesting access to delivery logs to identify elligible study subjects. All women
who decline to participate in the study will be deleted from the study data base.
4.4.1.7 Interview Protocol and Informed Consent
Women who agree to participate will be scheduled for an in-home interview that will
take approximately 1 hour and will be conducted at a time convenient to them.
Potential subjects will be given an informed consent for both themselves and their
infant during the postnatal interview. The interviewer will read the consent with the
participant and answer any questions they may have before obtaining their signature
and witnessing the consent. A blood sample of approximately 30-ml (about 2
tablespoons) will be collected from the mother and two buccal swabs will be obtained
107
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
from the infant. Upon completion of the interview, mothers will receive $20 and a
small gift (baby wash, shampoo, and powder donated by Johnson & Johnson) as
compensation for their time and effort.
4.4.1.8 Risk Factor Questionnaire
Known and potential risk factors for PE will be assessed via questionnaire. The
questionnaire will be administered by a bilingual (English/Spanish) interviewer/
phlebotomist during a postnatal interview. The risk factor questionnaire used in this
study is a modified compilation of questionnaires from two different epidemiological
studies of PE (R. Ness, Personal Communication). The questionnaire will include
questions on basic demographics, obstetric and reproductive history, medical history,
family medical history, employment status, type(s) of birth control used, tobacco and
alcohol exposure and physical activity (see Appendix B). In order to differentiate
between different subpopulations of Latinos, we have included questions on
birthplace of parents and grandparents.
4.4.1.9 Clinical Outcomes Data
In addition to the risk factor questionnaire, data on clinical outcomes (disease
severity, gestational age at onset of PE, length of gestation, maternal weight,
maternal blood pressure, birth weight, etc.) will be collected via review of medical
records.
108
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
4.4.2 Laboratory Methods
4.2.2.1 Biological Specimen Processing
Biological samples will be transported to the lab and processed immediately
following the interview. In order to maintain flexibility for later studies, both plasma
and serum samples (1 EDTA, 1 ACD, 1 clot) will be obtained from the mother.
Plasma and serum will be removed and aliquoted into 1 ml. freezer tubes and stored
at -80°C. Sample tubes will be labeled by study ID number, using pre-printed
freezer-resistant labels, and sample characteristics (e.g., volume, number of tubes,
presence of hemolysis) will be entered into the sample inventory database. Bufify
coat will be removed from the unclotted tubes (EDTA and ACD) to be used for DNA
extraction. Cellular material from infant buccal swabs will be lysed and the elluate
will be stored at -80 degrees Celcius for DNA extraction.
4.4.2.2 DNA extraction and whole genome amplification
DNA will be extracted from maternal bufify coat and from infant buccal samples
using QLAamp DNA blood mini kits (QIAGEN). Extracted DNA will be quantitated
using the Hoescht dye method. Working dilutions of maternal DNA will be prepared
to a concentration of 10 ng/ml and stored at -20°C. The remaining stock DNA will
be inventoried and stored at -80°C.
Previous studies utilizing buccal cell sampling in children have found that DNA
yields were lower than expected (Zheng 2001). In order to insure sufficient quantity
of infant DNA, we will perform whole genome amplification using the MDA
(multiple displacement amplification) method (Dean, Hosono et al. 2002).
109
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Approximately 50 nanograms of infant DNA will be amplified using the Repli-G kit
from Molecular Staging (New Haven, CT) according to manufacturer's instructions.
Working dilutions of amplified DNA will be prepared as described for maternal
DNA above.
4.4.2.3 Polymorphism Discovery (Resequencing)
With the exception of VEGF and MAO-A, little is known about polymorphism in the
candidate genes of interest (HAND1, HASH2, GCM1, HIF-la, TGFp3, VEGF,
sFLTl, PGF, and MAO-A). Furthermore, data reported in public polymorphism
databases contain a high rate of false positives, and many common polymorphisms
are not reported. Therefore, we will systematically resequence each candidate gene to
identify all common coding region and regulatory region polymorphisms.
Because our long-term goal is to conduct a larger study that includes women from a
range of racial/ethnic groups, we will conduct polymorphism discovery studies using
available DNA samples from four groups (Asian, White, Black, Latino). These
samples were collected for use in previous epidemiologic studies from study subjects
who agreed to allow their DNA to be saved and used for other research purposes.
For each of the candidate genes of interest, we will sequence DNA from 192
chromosomes from 96 individuals (24 from each of the four racial/ethnic groups).
Regions to be sequenced include all exons and the 5' flanking region, extending at
least 1 kilobase upstream of the transcription start site, additionally including any
known regulatory regions and highly conserved regions. Each exon or regulatory
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
region will be PCR amplified, and PCR products will be sequenced on the ABI prism
3700 capillary sequencer. Both forward and reverse strands will be sequenced. We
will genotype all polymorphisms that are potentially functionally significant. This
includes all non-synonymous SNPs (single nucleotide polymorphisms), exonic indels
(insertion/deletions), splice site SNPs, and all regulatory region (5'UTR, 3TJTR,
5'flanking region) polymorphisms.
4. 4. 2.4 Genotyping of microsatellites and other length polymorphisms (Genescan
method)
As an example of the Genescan method for genotypng length polymorphisms, we
describe the assay for genotyping the 30-bp repeat polymorphism in the MAO-A
gene. This polymorphic variant consists of a 30-bp repeated sequence which can be
present in either 2 (rare), 3, 3.5,4 or 5 copies (Sabol, Hu et al. 1998). First, the
genomic region containing the polymorphism will be PCR amplified using one
unlabeled and one fluorescently-labeled primer. The forward primer is 5'-
TGCTCCAGAAACATGAGCAC-3' and the FAM labeled reverse primer is 5'-
GTAGGAGGTGTCGTCCAAGC-3'. The 10 ul PCR reaction mixture contains 2.0
mM MgCl2 and 2% DMSO. The PCR cycling conditions are: 5 min initial
denaturation at 94°C followed by 35 cycles of: denaturation at 94°C for 1 min,
annealing at 53°C for 1 min, elongation at 72°C for 1 min. Final elongation is 5 min
at 72°C. PCR products are run on the ABI 3700 Sequencer (Applied Biolsystems,
Carson City, CA) with a fluorescently-labeled size standard. Results are analyzed
using GeneScan software (Applied Biosystems, Carson City, CA).
Ill
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Other length polymorphisms (eg. indels, microsatellites, repeated sequences) that are
discovered will be genotyped using a similar procedure. In addition, 50
microsatellite polymorphisms will be genotyped to assess population stratification
(see Section 4.4.3.2.5 below).
4.4.2.5 SNP Genotyping (TaqMan Assay)
Genotyping of single nucleotide polymorphisms (SNPs) will be performed by the
TaqMan assay using the TaqMan Core Reagent Kit (Applied Biosystems, Foster
City, CA). Two oligonucleotide probes, one specific for each allele, are labeled with
different fluorophores and are included during the PCR amplification. Hybridization
of the reporter probe to its target sequence leads to nucleolytic cleavage of the probe
during PCR amplification/DNA synthesis as a result of the 5' > 3' nuclease activity of
Taq polymerase. The fluorescent signal is measured in an Applied Biosystems
Sequence Detection System model 7900HT. Multicolor analysis is used to detect
both alleles of the biallelic system. Experimental samples are compared to 9
previously sequenced controls (3 of each genotype) to identify the 3 genotypes at
each locus. Any samples that are outside of the parameters defined by the controls
are identified as non-informative.
For SNPs that are difficult to optimize using the TaqMan methodology, we will use
alternate methods that are in use in our laboratory, such as primer extension analysis,
sequencing, or manual methods (eg. RFLP).
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
4.4.2.6 Genotyping Quality Control
All genotyping assays will include control samples that have had genotype previously
confirmed by sequencing. All PCR assays include a "water blank" to guard against
contamination of the PCR reaction. To assure assay reproducibility, 5% of samples
will be repeated with laboratory personnel blinded as to repeat status.
4.4.3 Data Management and Statistics
4.4.3.1 Database Management and Data Processing
Three separate databases will be maintained during the course of the study. First, all
potential cases and controls will be entered into a tracking database which will be the
only database containing personal identifiers and contact information as well as the
study identification number. Women who decline to participate will be deleted from
the tracking database. The second database will include only study participants and
will contain questionnaire data as well as data obtained via medical records. In this
database, subjects will be identified only by study identification number. Prior to data
entry, questionnaire data will be edited and checked for completeness by the data
manager. If necessary, study participants will be recontacted to clarify inconsistent
or missing data. The third database will contain laboratory data, including sample
inventory data and genotype results and will also be identified only through study
identification number. Genotype data from the TaqMan or ABI sequencer will be
captured into Excel spreadsheets and merged into the laboratory database. All data
will be backed up frequently. For the questionnaire and laboratory databases, the
programmer will run cleaning and consistency checks and examine the distributions
113
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
of all variables on a regular basis. Only study personnel will have access to the data
files.
4.4.3.2 Statistical Analysis and Power Calculations.
4.4.3.2.1 Aim la: To identify all common polymorphic variants in the candidate
genes of interest.
We will sequence 192 chromosomes (96 individuals, 24 from each of 4 racial/ethnic
groups). Power to detect polymorphisms depends on polymorphism prevalence and
on the sensitivity of the screening test: N = ln(l-power) / ln(l-(prevalence x
sensitivity)). Since sequencing is expected to be nearly 100% sensitive, screening 24
subjects (48 chromosomes) in each ethnic group should give us greater than 90%
power to detect polymorphisms having a prevalence of 5% within an ethnic group.
With all ethnic groups combined, we will have greater than 85% power to detect
polymorphisms having a prevalence of 1%.
4.4.3.2.2 Aim lb&c: To determine allele frequencies for all polymorphisms in these
genes. To determine haplotype frequencies for those genes that contain more than
one polymorphic site.
For genes that contain more than one polymorphism, we will estimate haplotype
frequencies using the EM algorithm as implemented in the hapif command in Stata
(Stata Corporation, College Station, TX) (Mander 2001).
114
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
We will be able to estimate the allele and haplotype frequencies ± 5% or better,
depending upon the allele frequency. The lowest precision occurs when the two
alleles are equally frequent. With an allele frequency of 0.5 and 480 alleles observed
(240 controls), we obtain the following confidence interval: Cl = 0.5 ± 1.96
[(0.5)(0.5)/480]°5 = 0.5 ± 0.05 = (0.45, 0.55).
4.4.3.2.3 Aim 2a&b: To generate preliminary data linking polymorphism in
candidate genes to PE risk
For preliminary descriptive analysis, genotype frequencies and basic demographic
variables will be examined separately for cases and controls. Univariate analyses
will be performed using the McNemar’s or paired t-tests.
For multivariate analyses, standard analyses for matched case-control studies will be
conducted (Breslow and Day 1980). Maternal age, race/ethnicity, gestational age and
year/month of delivery will be controlled for by matching on these variables and
employing a matched analysis (conditional logistic regression on the matched
case/control sets). Other potential confounders will be included in the conditional
logistic regression model. A variable will be considered a confounder only if it is
either a well-established confounder (based on the literature) or if the adjusted and
unadjusted odds ratios differ in value by at least 10%.
Since the power to assess statistical interaction between fetal and maternal genotypes
will be very low in this pilot study, we plan to examine these potential effect
115
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
modifiers in a future study. The general approach to interaction that we will take in
future studies will be similar in spirit to the log-linear models discussed by Weinberg
et al. (Weinberg, Wilcox et al. 1998), Kraft and Wilson (Kraft and Wilson 2002), and
Sinsheimer et al. (Sinsheimer, Palmer et al. 2003) for estimating maternal-child
genotype effects for case-parent triads. However, by sampling matched case/child
and control/child pairs we can estimate effects and interactions directly within the
conditional logistic framework.
Power for detecting a main effect of genotype (or haplotype) on PE risk is given in
the table below for a range of genotype frequencies and a range of odds ratios (Table
4.1). Power calculations were calculated using a matched analysis assuming a type I
error rate of 0.05,120 cases and 240 controls, and no correlation of exposures
between matched mother/child sets. We will have adequate power to detect a
moderate sized effect (OR=2.0) when the at-risk genotype is common (20% or higher
prevalence) or a larger effect (OR=2.5) when the at-risk genotype frequency is as low
at 10%. For uncommon genotypes (5% frequency), we will only be able to detect a
relatively large effect (OR=3.0), however; an odds ratio of this magnitude is not
unexpected with uncommon alleles, since uncommon alleles are more likely to be
functionally significant (Cargill, Altshuler et al. 1999) (and thus produce larger odds
ratios).
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Table 4.1 Power for detecting a main effect of genotype (or haplotype) on PE risk
Probabiltiy of exposure in controls OR______________________________
(i.e. variant genotype/haplotype) 1.5 2.0 2.5 3.0
0.05 0.16 0.38 .62 0.81
0.10 0.23 0.59 0.85 0.96
0.15 0.30 0.71 0.93 0.99
0.20 0.35 0.78 0.96 1.00
0.30 0.41 0.85 0.98 1.00
0.40 0.44 0.87 0.98 1.00
0.50 0.43 0.86 0.98 1.00
4.4.3.2.4 Aim 3a: To determine recruitment rates among eligible study subjects
delivering at LAC+USC.
We estimate that there will be 240 eligible cases (see Section 4.3.2 above). We will
be able to estimate the recruitment rate ± 6% or better. The lowest precision occurs
for a participation rate of 50%. A participation rate of 50% from 240 eligible cases
gives the following confidence interval: Cl = 0.5 ± 1.96 [(0.5)(0.5)/240]°'5 = 0.5 ±
0.06 = (0.44, 0.56).
4.4.3.2.5 Aim 3b: To estimate the degree of population stratification among Latino
patients at LAC+USC, and thus the extent to which gene-disease associations may be
confounded by ethnic admixture in this population.
Using a worldwide sample that included individuals from 52 populations, Rosenberg
et al. (Rosenberg, Pritchard et al. 2002) found that continental ancestry of
individuals, and in many cases, populational ancestry, could be inferred accurately
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
using 377 microsatellite markers. Additional work using these data suggests that
similar results are obtained if analysis is restricted to the 40 to 50 markers whose
allele frequencies differ most substantially across populations (N. A. Rosenberg, L.
Li, R. Ward and J. K. Pritchard, submitted) (See Figure 4.1). Among the 377
markers in the Rosenberg et al. (2002) data set, the 50 whose frequencies are most
strongly diverged across the relevant regions (eg. the Americas, Europe, and Africa)
will be identified. The choice of markers will be based on allele frequencies from
those populations in the Rosenberg et al. data that are closely related to potential
source groups for the individuals in the study sample.
This set of 50 markers will provide a flexible framework in which to 1) test for the
potential impact of population stratification in our case-control sample, 2) estimate
genetic subpopulations, and 3) obtain unbiased risk estimates with valid tests of
association. In the presence of population structure, the usual test of association is
overdispersed and does not follow the assumed null distribution. However, by
examining the distribution of tests of association for all the unlinked markers in the
set, we can check if the population structure is substantial enough to alter the null test
statistic distribution (Devlin and Roeder 1999; Pritchard and Rosenberg 1999; Reich
and Goldstein 2001). Additionally, it has been demonstrated that the impact of
population structure on the degree of departure from the null distribution increases as
the sample size increases (Pritchard and Donnelly 2001). Therefore, selecting
markers that are highly informative for individual ancestry ensures the maximum
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
potential for using our initial sample to establish the feasibility of conducting a
larger-scale case-control study in the patient population.
These highly informative markers are ideal in testing for stratification (Pritchard and
Rosenberg 1999), because (1) if no stratification is detected using a set of extremely
informative markers, it can be concluded with confidence that stratification has little
or no effect on an association study; (2) these markers have the greatest potential for
identifying stratification if it is present; and (3) they are ideal for use in methods that
estimate individual ancestry coefficients to control for the effects of stratification, if
it is found (Pritchard, Stephens et al. 2000).
Identifying the potential for population stratification is only the initial step, as we
would also like to gain knowledge regarding the underlying population structure.
The computer program structure (Pritchard, Stephens et al. 2000) uses a set of
unlinked markers to estimate the probability that an individual’s ancestry is from two
or more genetic subpopulations. This “ancestry coefficient” may then be used to
calculate a valid stratified test statistic that models allele frequencies for a candidate
gene variant conditioned upon both disease status and subpopulation (Pritchard,
Stephens et al. 2000). Extending this idea, the “ancestry coefficient” can also be used
as a latent stratification variable in the generalized linear framework. Therefore, we
can use conditional logistic regression to perform a matched multivariate analysis
adjusting for potential confounders and individual ancestry. This provides a
mechanism to obtain unbiased estimates of risk and valid tests of association in
structured populations.
119
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Pritchard et al. (2000) and Pritchard and Donnelly (2001) argue that the degree to
which confounding due to stratification can be corrected is based largely on the
degree to which ancestry coefficients can be accurately estimated. The attached
figure demonstrates that ancestry coefficients can be estimated accurately using 40-
50 highly informative markers. Because we are using at least this number of
markers, we will have power to identify and correct for stratification if it is present
(Figure 4.1).
Figure 4.1; Inferred population structure using five clu sters for m arers of highest an d low est
inform ativeness. E ach individual is rep resented by a thin vertical line, which is partitioned into
5 colored se g m e n ts that rep re sen t th e individual's estim ated m em bership fractions in 5
clusters. Black lines s e p a ra te individualsof different populations.
Loci of highest informativeness
[T
Loci of lowest Informativeness
10 loci
20 loci
M 1 '/ , ,| M <’ ' , - , f” |l||l< I ,1 - f
40 loo
80 loci
120 loo
180 oci
260 loci
377 loci
|j B J iu 11‘hi I
S j i ! , i i I i n * } i j f ! S|
120
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
4.4.4 Proposed Time Table
4.4.4.1 Year One. Identification and recruitment of eligible study subjects will
begin. Delivery logs will be queried on a weekly basis and interviews will take place
as participants are identified. As cases are recruited, two appropriately matched
controls will be selected from the delivery logs. Interviews will take place throughout
the two year duration of the study. As batches of biological samples accrue, DNA
extraction and quantitation will be performed.
In the first six months, DNA from four ethnic groups (see Section 4.4.2.3 above) will
be sequenced to identify all common polymorphisms in the candidate genes of
interest. In the second six months, we will begin genotyping these newly-identified
polymorphisms using DNA from case/child and control/child pairs.
4.4.4.2. Year Two. Interviews and genotyping will continue until the target sample
size is achieved. Towards the end of year two, data will be analyzed. Analysis of
population stratification will be conducted by Drs. Conte and Rosenberg under the
supervision of Dr. Thomas. Case-control analysis will be personally supervised by
Dr. Ingles. All investigators will participate in the interpretation of the results and
manuscript preparation.
4.4.5 Study Strengths and Limitations
Several inherent limitations exist as a result of the case-control design proposed here.
For example, selection bias can result from a failure to identify the base population
from which cases arise. We plan to minimize this occurence by using delivery logs
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
to define a base population and select both cases and controls from this population.
Thus, we will be able to make causal inferences with more confidence. In order to
properly control for measurable confounders, we will collect data on all known or
suspected confounders and adjust for them in our statistical analysis. Confounding
by ethnic admixture is a potentially serious concern and cannot be completely
adjusted for by using questionnaire data. Therefore, we will utilize newly developed
methodologies for assessing the extent of population stratification. If substantial
stratification is found, we will use novel statistical methods to correct for bias.
Additionally, we will collect data on birthplace of parents and grandparents to help
identify the underlying ethnic origins of any detected population clusters.
One of the most significant potential limitations of studying PE is the difficulty in
properly classifying disease status, leading to misclassification. Preeclampsia was
historically defined as a very heterogeneous disease, encompassing what is now
considered gestational hypertension (most likely underlying essential hypertension)
and “pure type” PE. To address this problem, more stringent diagnostic criteria have
been developed to differentiate between essential hypertension and “pure type”
preeclampsia. Since it is our intention to study only “pure type” PE, rather than the
entire spectrum of hypertensive disorders of pregnancy, in our proposal we utilize a
very strict definition of disease that excludes women with underlying maternal
conditions (e.g., essential hypertension and diabetes). By doing so, we reduce the
possibility that disease status will be misclassified on the basis of polymorphisms in
genes that might be involved in essential hypertension but not necessarily with PE.
Such exclusions will affect the generalizability of our results, however, it is our
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
intention to generalize only to women with “pure type” PE. Nondifferential
misclassification (misclassification not dependent on genotype) could still exist,
however, it would result in an attenuated odds ratio. Therefore, if an effect of
genotype is found on the risk of PE, it is likely that the true effect is even stronger.
Strengths of the proposed study include the efficiency of the case-control design and
the lack of temporal ambiguity since the exposures of interest (genotypes) are not
modifiable due to disease diagnosis. More importantly, the proposal of a series of
novel PE candidate genes, the interrogation of these genes to detect previously
unidentified polymorphism, and the assessment of these polymorphisms for
association with PE risk will be a substantial contribution to the literature. Also,
since allele frequencies are likely to vary by ethnicity, focusing on an ethnic group
not commonly studied (e.g., Latinos) can provide novel information. Finally, the
investigation of potential population stratification in the Latino population has long
been needed. Although people have speculated that this population is especially
susceptible to confounding by ethnic admixture, no previous empirical data have yet
been published. The Southern California population is ideal for conducting such a
study and the data generated will have wide-reaching applications.
4.5 HUMAN SUBJECTS RESEARCH
4.5.1 Protection of Human Subjects
4.5.1.1 Risks to the Subjects
We plan to recruit 120 cases of PE and 240 matched normotensive controls between
the ages of 18 and 40. All participants will be interviewed to obtain information on
123
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
possible risk factors for PE and approximately 30 cc of blood will be collected from
the mother and buccal cell samples will be collected from infants to assess genotypes
for a series of candidate genes. PE diagnosis and study eligibility will be evaluated
using medical records. All data and blood samples collected will be used solely for
research purposes and plasma and DNA will be stored for possible future studies of
PE.
The physical risks of venipuncture are minimal and include possible discomfort,
hematoma and/or fainting. Psychological risks from the interview are extremely
minimal. We have asked similar questions in previous studies and in most instances,
subjects readily responded to the questions and found the experience to be positive
overall. Standard treatment for PE will not be affected by participation in this study
and an alternative to participation in the proposed study is nonparticipation.
4.5.1.2 Adequacy of Protection Against Risks
Recruitment will be facilitated through delivery logs from the LAC + USC Women’s
and Children’s Hospital. We have applied for a HEPAA waiver to access this
protected health information. Women who decline to participate in the study will be
deleted from the database. The informed consents used in this study have been
approved by the USC IRB committee and informed consent will be obtained from
study participants by the interviewer/phlebotomist during the in-home interview. To
help protect the confidentiality of any genetic information that is generated as a result
of this study, a Certificate of Confidentiality has been issued by the NIH. This
certificate allows researchers to avoid involuntary disclosures of information.
124
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
To minimize the risks associated with venipuncture, we will ensure that all study
personnel are adequately trained. In order to maintain participant confidentiality,
especially with respect to genetic information, we will be applying for an NIH
Certificate of Confidentiality. This is in addition to the standard procedures of
maintaining confidentiality, such as utilizing locked cabinets for data storage and
identifying data via a coded identification number only.
4.5.1.3 Potential Benefits of the Proposed Research to the Subjects and Others
Because the risks associated with participation in this study are minimal, we believe
that the advances in knowledge that stand to be gained from conducting this study are
well justified.
4.5.1.4 Importance of the Knowledge to be Gained
Knowledge of predisposing PE genotypes would most certainly advance our
understanding of the etiology and pathogenesis of the disease process and would
allow the identification of a subgroup of women at unusually high risk of disease.
Such knowledge would likely facilitate the development of new prophylactic and/or
therapeutic interventions and lead to an improvement in maternal and perinatal
outcomes by advancing our understanding of PE etiology. Since PE affects a
substantial proportion of pregnant women, overall public health would surely benefit
from any treatments derived and from a greater understanding of this disease. In this
light, the very minimal risks associated with the proposed study are well worth the
potential advancement in knowledge and treatment.
125
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Additionally the knowledge gained from polymorphism discovery, determination of
allele and haplotype frequencies in a predominately Latino population, and the
assessment of population stratification among Latinos will have wide ranging
applications in a number of disciplines.
4.5.2 Inclusion of Women
All women between the ages of 18 and 40 at the time of the index pregnancy who
delivered at the LAC + USC Women’s and Children's Hospital will be eligible to
participate.
4.5.3 Inclusion of Minorities
Since the large majority of our population is of Latino origin, we will be including a
considerable number of minorities in our study population. We will not exclude
anyone on the basis of ethnicity. In order to facilitate the inclusion of these
minorities, we will be hiring an English/Spanish bilingual interviewer.
4.5.4 Inclusion of Children
Children under the age of 18 will not be eligible to participate as maternal subjects
since PE that occurs at the extremes of age may have a different etiology. Thus, by
excluding children under the age of 18 and women over the age of 40, we hope to
obtain a more homogeneous population. However, we will be including children
bom of the index pregnancy in the assessment of fetal susceptibility genotypes for
PE.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
4.6. VERTEBRATE ANIMALS
None.
4.7. LITERATURE CITED
See Reference Section located at the end of the dissertation.
4.8. CONSORTIUM/CONTRACTUAL ARRANGEMENTS
None.
4.9. CONSULTANTS
Noah Rosenberg, Duncan C. Thomas
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
CHAPTER 5: MONOAMINE OXIDASE A AND TOBACCO ADDICTION
5.1 BACKGROUND
As the most prevalent addiction in modem society (Kandel, Chen et al. 1997),
tobacco addiction is a serious public health problem, with smoking being an
important risk factor for lung and other cancers (reviewed in Stein and Colditz
2004), cardiovascular disease (reviewed in Critchley and Capewell 2003), oral
diseases (Reibel 2003), ocular diseases (reviewed in Cheng, Pang et al. 2000) and
pulmonary disease (reviewed in Mannino 2003), to name but a few.
Nicotine is a highly addictive dmg, meeting the same criteria for addiction as other
addictive substances such as cocaine, alcohol and heroin (Services 1988; US Food
and Dmg Administration (21 CFR Part 801 1996) (Table 5.1). Since nicotine is
generally considered to be the addictive component of tobacco, I refer to tobacco
addiction and nicotine addiction interchangeably throughout this chapter.
5.1.1 Genetics and Tobacco Addiction
There is good evidence that smoking behaviors are at least partially under genetic
control (Heath and Madden 1995; Tyndale 2003). In one study examining male
twins, higher concordance rates were found for smoking and for quitting smoking
among monozygotic compared to dizygotic twins, suggesting that a genetic
predisposition may be involved (Carmelli, Swan et al. 1992). Moreover, the risk of
nicotine dependence (as measured by standardized questionnaires designed
specifically to assess it) is statistically significantly greater among second siblings if
128
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
the first sibling is nicotine dependent (Niu, Chen et al. 2000). While it remains
difficult to separate the effects of genetic predisposition from environmental
similarities within families, studies of twins reared apart suggest a genetic
component to drug-related problems and antisocial personality (Grove, Eckert et al.
1990) as well as smoking habits (Hayakawa 1987; Kendler, Thornton et al. 2000). A
recent study of adoptees further supports a role for a genetic component in smoking
behaviors (Osier, Holst et al. 2001).
It is generally accepted that there are three stages of tobacco addiction - initiation,
continuation and cessation. Roughly, initiation refers to whether or not an individual
ever begins to smoke while continuation (i.e., persistence) refers to whether or not
the smoking activity is maintained after the first trial. Cessation refers to whether or
not the individual is able to quit smoking once they have become addicted. All
three stages of tobacco addiction appear to be under genetic control (True, Heath et
al. 1997), with the different stages being controlled by different genetic mechanisms,
although there maybe some overlap (Heath and Martin 1993; Kendler, Neale et al.
1999). Heritability estimates for smoking initiation range from 46-84%, while those
for continuation range from 53-70% (Eaves and Eysenck 1980; Hannah, Hopper et
al. 1985; Edwards, Austin et al. 1995; True, Heath et al. 1997; Kendler, Neale et al.
1999).
5.1.2 Mechanism of Tobacco Addiction
In order to become addicted to nicotine, one must first be exposed to the substance.
The question then becomes, what factors predispose to smoking initiation? Research
129
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
on personality traits has defined a characteristic known as sensation-seeking. People
with the sensation-seeking temperament tend to require heightened levels of novelty
and variety and will take physical and social risks in order to achieve them (Gerra,
Avanzini et al. 1999). Thus, people who score high on sensation-seeking indices are
more likely to try smoking than those with low scores (Etter, Pelissolo et al. 2003).
Moreover, it is believed that personality traits, including the novelty-seeking
phenotype (novelty-seeking is part of sensation-seeking), have a genetic basis
(Cloninger 1986). Importantly, decreased MAO-A activity has been linked to
enhanced reward and the sensation-seeking temperament (Johansson, Von Knorring
et al. 1983; Hallman, Sakurai et al. 1990).
In order to proceed along the pathway toward nicotine addiction, an individual must
continue to smoke after the initial trial. Smoking, as in other drugs of abuse, invokes
reward pathways which then promote further consumption of nicotine (Koob and
Nestler 1997). Specifically, nicotine triggers the reward pathway through activation
of the acetylcholine receptors on dopaminergic neurons in the ventral tegmental
region of the brain (Comings and Blum 2000). The resultant release of dopamine
(DA) reinforces the behavior and contributes to nicotine addiction in rats (Corrigall,
Franklin et al. 1992) as well as humans (Laviolette and van der Kooy 2003). In fact,
it has been suggested that addicted individuals may suffer from “reward deficiency
syndrome” whereby functional genetic variations lead the individual to search out
“unnatural rewards” such as alcohol, drugs of abuse (including nicotine), or risky
sports (Comings and Blum 2000).
130
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
An individual who is unable to readily quit smoking may be considered addicted if
they also meet DSMIV criteria for addiction (Table 5.1). The ability to stop
smoking may also be under genetic influence and genes in the dopaminergic
pathway have been implicated (Lerman, Shields et al. 2003).
5.1.3 The MAO-A connection
Levels of DA, 5-HT, and NE are all affected by nicotine administration, suggesting a
possible role for MAO-A in nicotine addiction. Early studies of candidate genes for
nicotine addiction focused primarily on the DA stimulation pathway since
dopaminergic neurons express nicotinic acetlycholine receptors (Pidoplichko,
DeBiasi et al. 1997). Moreover, DA in the mesolimbic region of the brain (a portion
of the midbrain important for memory and motivating behaviors) is released in
response to nicotine administration (Imperato, Mulas et al. 1986; Pontieri, Tanda et
al. 1996), much as it is with other drugs of abuse. Indeed, statistically significant
associations have been found between cigarette smoking and polymorphisms in
several genes in this pathway, including DRD2 (Noble, St Jeor et al. 1994; Comings,
Ferry et al. 1996), SLC6A3 (Lerman, Caporaso et al. 1999; Sabol, Nelson et al.
1999) and DRD4 (Shields, Lerman et al. 1998).
There is substantial evidence from animal models that serotonin (5-HT) and
norepinephrine (NE), both of which are released as a result of nicotine
administration (Ribeiro, Bettiker et al. 1993; Summers and Giacobini 1995; Reuben
and Clarke 2000), may be involved in nicotine addiction as well. Specifically,
administration of the NE reuptake inhibitor reboxetine has been found to reduce
131
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
nicotine self-administration in rats (Rauhut, Mullins et al. 2002). Moreover, rats
chronically self-administering nicotine experienced sustained NE secretion in the
hypothalamic paraventricular nucleus (PVN), a region of the brain that mediates
neuroendocrine and behavioral responses (Fu, Matta et al. 2001), suggesting that NE
is important in smoking related behaviors.
A role for 5-HT in nicotine addiction is also supported by several sources In fact,
MAO-A activity in the brain (Fowler, Volkow et al. 1996; Fowler, Volkow et al.
1998) is reduced in response to tobacco use in a dose-dependent manner. Likewise,
lower levels of a dopamine metabolite, homvanillic acid (HVA) (Geracioti, West et
al. 1999), have been observed in smokers compared to nonsmokers, indicating a
reduction in dopamine metabolism among smokers. The net effect of decreased
MAO-A is to further increase 5-HT, NE and DA levels in response to smoking
(Fowler, Volkow et al. 1996).
Alterations in normal MAO-A levels or activity have been implicated in a variety of
psychiatric disorders, including alcoholism (Vanyukov, Moss et al. 1995; Hsu, Loh
et al. 1996), borderline mental retardation and impulsive aggression (Brunner, Nelen
et al. 1993; Brunner, Nelen et al. 1993), depression (Glassman, Helzer et al. 1990;
Quattrocki, Baird et al. 2000), schizophrenia (Jonsson, Norton et al. 2003), drug
abuse (Gade, Muhleman et al. 1998), panic disorder (Deckert, Catalano et al. 1999)
and risk taking behaviors such as gambling (Gade, Muhleman et al. 1998). In
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Table 5.1: DSM-IV Criteria for Substance Abuse and Dependence (First 2000)
^ _ _ _ _ _ _ _ _ N icotine D ependence _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ N icotine W ithdraw al
D ependence is defined as a cluster of three or m ore of the
sym ptom s listed below occurring at any tim e in the sam e
12-m onth period
1. Tolerance, as d efin ed b y eith er o f the follow ing:
a), a n eed fo r m ark ed ly in creased am o u n ts o f th e
substance to ach iev e d esired effect
b). m ark ed ly d im in ish ed e ffect w ith continued use o f
the sam e a m o u n t o f th e su b stan ce
2. W ithdraw al, as m an ifested b y eith er o f the follow ing:
a), the characteristic w ithdraw al syndrom e for the
substance (see right)
b). the sam e (o r closely related ) su b stan ce is taken to
relieve o r avoid w ithdraw al sym ptom s
3. T h e su b stan ce is often taken in larger am o u n ts or over a
longer p erio d than w as in tended
4. T here is a p e rsisten t desire o r u nsuccessful efforts to cu t
dow n or control su b stan ce use
5. A g reat deal o f tim e is sp en t in activities n ecessary to
obtain the su b stan ce, use th e su b stan ce (e.g., chain
sm oking), o r reco v er fro m its effects
6. Im portant social, o ccupational, o r recreational activities
are given up o r red u ced b e cau se o f substance u se (e.g.,
sm o king-restricted area cau ses individual to forgo activity)
7. T h e su b stan ce u se is contin u ed despite know ledge o f
having a p ersisten t o r rec u rre n t physical o r psychological
p roblem th at is likely to h av e b een caused o r exacerbated
by th e substance (e.g ., continued use desp ite know ledge o f
sm oking-related h ealth p ro b lem )_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _
N /A
1. D aily u se o f n icotine for a t least several w eeks
2. A b ru p t c essatio n o f nico tin e use, o r red u ctio n in the
am o u n t o f nico tin e used, follow ed w ithin 2 4 ho u rs b y four
(o r m o re) o f th e follow ing signs:
a), d y sphoric o r dep ressed m ood
b). inso m n ia
c). irritability, fru stratio n , o r anger
d). an xiety
e). d ifficulty concen tratin g
f). restlessness
g). d ecreased h eart rate
h ). increased ap petite o r w eig h t gain
3. T he sym ptom s listed above cau se clinically significant
distress o r im p airm en t in social, occupational, o r o ther
im p o rtan t areas o f fu nctioning
4. T he sym ptom s are n o t d u e to a general m edical condition
and are n o t b e tte r accounted fo r b y an o th er m ental d isorder
133
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
particular, it has been suggested that depressives, who tend to be smokers more often
than their mentally healthy counterparts (Lenz 2004), may smoke as a means of self-
medication (Berlin, Said et al. 1995; Quattrocki, Baird et al. 2000) since smoking
reduces MAO-A activity and increases levels of neurotransmitters (NTs) as do many
antidepressants. This observation is supported by the fact that the antidepressant
moclobemide, a drug that inhibits MAO-A, has been successfully used for smoking
cessation, producing better cessation rates for up to six months post-treatment
(Berlin, Said et al. 1995).
As outlined in Chapter 2, a functional variable number tandem repeat polymorphism
exists in the MAO-A gene promoter region (uVNTR). Although at least six possible
alleles have been identified, most individuals have either a 3-repeat allele or a 4-
repeat allele, allowing the genotype to be dichotomized into a “short” (^3 repeats)
allele and a “long” (>3 repeats) allele. Functional studies have shown that the long
allele is more active than the short allele (Deckert, Catalano et al. 1999; Denney,
Koch et al. 1999). Lower MAO-A activity results in increased levels of DA and
thus, more reward-producing effects (Zuckerman 1993; Netter, Hennig et al. 1996).
Therefore, it can be expected that subjects with short repeats (and thus lower MAO-
A activity) would have an increased risk of becoming addicted to tobacco (Whitfield,
Pang et al. 2000).
Figure 5.1 summarizes the hypothesized role of MAO-A in tobacco initiation and
addiction. Briefly, low levels of MAO-A, possibly due to having the short repeat
allele in the MAO-A promoter, results in an increased tendency to seek out novel
134
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
experiences, including smoking. Continued smoking leads to further decreased
MAO-A and increased levels of DA, 5-HT and NE, which then activate the reward
pathway and reinforce the smoking behavior. The end result is outright nicotine
addiction.
Figure 5.1: Proposed Mechanism for Role o f MAO-A in Smoking Initiation
and Addiction
Sensation-
; ^Seeking
Addiction
Maintain
Smoking
Initiate
Smoking
. ■ ■ i
t NTs
■ y Reward
I MAO-A
1:
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
5.2 METHODS
5.2.1 Study Population
The study population was drawn from the Singapore Chinese Health Study described
in detail elsewhere (Seow, Shi et al. 1998). Briefly, between April 1993 and
December 1998, 27,959 Chinese men and 35,298 Chinese women (63,257 total)
aged 45 to 74 were recruited from government housing estates (86% of the
Singapore population live in such facilities). In-person interviews were conducted
by a trained interviewer using a structured questionnaire that collected data on
demographics, smoking history, diet, medical history, reproductive history (women
only), occupational exposure, physical activity and family history of cancer. Blood
and urine specimens were obtained from a random 3% sample of study subjects
beginning one year after study inception. Beginning in January, 2000, biospecimen
collection was extended to all consenting cohort subjects. This component of the
Singapore Chinese Health Study is ongoing. A follow-up questionnaire was
completed by 52,326 of the 56,965 eligible cohort participants (91.9%) after
accounting for deletions due to death (n=4,715), migration out of Singapore (n=5)
and contact failure (n=l,572). Follow-up data collected included updated smoking
information (active and passive), alcohol use, medical history, medication use,
menopausal status (women only), job history and current weight/height.
All subjects selected for this study were part of the 3% random sample of healthy
cohort participants who provided blood and urine specimens. Additionally, subjects
were selected only if both baseline and follow-up data were available. I selected
only male subjects since smoking is extremely rare in Singapore females. Subjects
136
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
with a positive cancer history were excluded. Study subjects included 490 men who
reported being daily smokers at both baseline and follow-up (current/current
smokers) and 490 men who reported being nonsmokers at both time points
(never/never smokers) and were matched on age. There were 4 genotyping failures
among the current/current smokers and 21 among the never/never smokers, leaving
486 current/current smokers and 469 never/never smokers available for analysis
(failure rate of 2.5% overall).
5.2.2 Laboratory Methods
DNA was extracted from diluted buffy coat stored using QIAamp DNA Blood
BioRobot Kits (Qiagen, Valencia, CA, USA). MAO-A uVNTR genotype was
determined by polymerase chain reaction (PCR) using the following primers: 5'-
TGCTCCAGA AACATGAGCAC-3' (forward) and 5'-GTAGGAGGTGTCGTCCA
AGC-3' (reverse). The reverse primer was labeled with fluorescent dye (6-FAM) to
enable usage of GeneScan technology. PCR was performed in a total volume of 25
pi with 2 pi of DNA, 0.2 pi of DMSO, and a final concentration of 3.0 pmol of each
primer, 0.25 mM of dNTPs, 2.0 mM of MgCl2 , 1 U Taq DNA polymerase (Promega
Biosciences, Inc., San Luis Obispo, CA), and lx buffer (Promega Biosciences, Inc.,
San Luis Obispo, CA). PCR thermal cycling conditions included an initial
denaturation at 94°C for 5 minutes, followed by 40 cycles of denaturation at 94°C
for 30 seconds, annealing at 53°C for 30 seconds, and extension at 72°C for 30
seconds. A final extension at 72°C for 5 minutes was conducted. One pi of 10-fold
diluted PCR product was combined with 10 pi of formamide and HD-400 size
137
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
standard and run through the ABI Sequencer 3700 (Applied Biosystems, Foster City,
CA). GeneScan software (Applied Biosystems, Foster City, CA) (version 3.5) was
used to read the size of the PCR product. The 1-repeat allele (1.5 repeats, see
Chapter 2) produced a PCR product 180 base pairs (bp) in length, 2 repeats produced
product of 210 bp, 3 repeats produced product of 240 bp, 3.5+ repeats (novel allele)
produced product of 252 bp, 4 repeats produced product of 270 bp and 5 repeats
produced product of 300 bp. Genotyping was conducted without knowledge of
smoking status. PCR was duplicated for 23% of the samples to corroborate initial
genotype. Samples with novel genotypes were confirmed via direct sequencing.
5.2.3 Statistical Methods
Continuous variables were compared using t-tests and categorical variables were
compared using Chi-squared tests. Kruskal-Wallis equality of populations rank tests
were used to compare categorical baseline smoking variables. All p-values were
two-sided and p < 0.05 was considered statistically significant. Where sparse data
were observed, variables were collapsed as necessary to create more robust
categories while attempting to maintain any biological differences that may be
present between categories.
Although at least six possible alleles exist for the uVNTR polymorphism in MAO-A,
the genotype was dichotomized into short (<3) and long alleles (>3.5), as suggested
by functional studies indicating that the longer alleles are more active than the
shorter alleles (Deckert, Catalano et al. 1999; Denney, Koch et al. 1999). However,
there remains some uncertainty regarding the activity level of the 5-repeat allele.
138
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Whereas Deckert et. al. found that the longer alleles (3.5,4, and 5 repeats) had
higher activity levels than the 3-repeat allele, Sabol et. al. found that the 5-repeat
allele had an activity level closer to the 3-repeat allele (Sabol, Hu et al. 1998).
Nevertheless, since the 5-repeat allele is very rare, I chose to dichotomize on the
basis of “long” versus “short” alleles for the sake of simplicity. Moreover, analyses
conducted placing the 5-repeat allele subjects in the “short” (i.e., low activity)
category did not differ from those presented (data not shown).
Logistic regression was used to model the association between smoking status (the
dependent variable) and the dichotomized genotype variable, adjusting for
confounding variables. All variables showing an association with genotype or
smoking status during the preliminary analysis as well as those variables that were
considered a priori to be potential confounders were examined. Variables were
included in the final model if they changed the effect estimate by > 10% when
included or if they are generally considered to be confounders of the tobacco-
genotype association in most circumstances.
Linear regression was used to model the associations between genotype and follow-
up smoking variables including age at first cigarette, age at regular smoking and
number of cigarettes smoked. Confounders were assessed in a manner similar to that
used for logistic regression, although some effort was made to select a group of
covariates that could be used for all three outcomes.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
5.3 RESULTS
The results appearing in Tables 5.2ABC through tables 5.8ABC present preliminary
analyses conducted to (1) identify potential confounding variables and (2)
familiarize myself with the data set. Therefore, these results will be addressed only
briefly to highlight statistically significant findings. Recall that MAO-A is located
on the X chromosome and that, for the male subjects in this study, only one MAO-A
allele is present since there is only one X chromosome in males. Thus, for the
purposes of these results, allele = genotype. Additionally, no adjustment for
multiple comparisons has been made and thus, some statistically significant
associations would be expected by chance alone. Tables 5.9-5.12 contain the results
which are the main focus of my work and will be discussed in greater detail.
5.3.1. Demographics
Table 5.2A shows the differences in demographic variables by smoking status. As
expected, current/current smokers (hereafter referred to as smokers) weigh
statistically significantly less than and have a smaller body mass index (BMI)
compared to never/never smokers (hereafter referred to as nonsmokers) (p<0.001 for
both comparisons). This difference in weight can be explained by the appetite
suppressing effects of nicotine (Perkins, Sexton et al. 1996) and is commonly
observed in studies comparing smokers to nonsmokers (Erikssen and Enger 1978;
Kok, Matroos et al. 1982; Staley, Krishnan-Sarin et al. 2001). A statistically
significant difference between smokers and nonsmokers was also observed with
respect to educational status, with a much higher percentage of nonsmokers having a
university education compared to smokers (7.8% vs. 0.6% respectively, p < 0.001).
140
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Additionally, a statistically significant difference in smoking status was observed for
dialect group, with more smokers being from the Hokkien rather than Cantonese
group (60.8% vs. 39.2% respectively, p < 0.001). This difference by dialect group
can be explained by the fact that subjects in the Hokkien group are substantially less
educated than those from the Cantonese group (data not shown). No other
differences in demographics were observed by smoking status.
Table 5.2B presents the demographic data by genotype. The only finding of interest
may be the observed difference in allele frequency by family history of cancer.
Specifically, subjects with the short allele appear to have a marginally statistically
significant increase in proportion of subjects with a family history of any cancer
(short: 17.7% vs. long: 13.0%, p = 0.051). Given that “family history of cancer” is a
rather nonspecific exposure (not all cancers are smoking-related) and the problem of
multiple comparisons, this finding should be regarded with some suspicion.
Demographic results for genotype, stratified by smoking status can be found in Table
5.2C. These data suggest that both mean weight and height may differ by genotype,
after adjusting for smoking status. In fact, there appears to be some effect
modification by smoking status, since weight and height are statistically significantly
greater in subjects with the long allele, but only among the nonsmokers (p = 0.041
and p = 0.002, respectively). No such effect was observed among smokers. A
marginally statistically significant increase in family history of any cancer was
observed for subjects with the short allele, but again, only among the nonsmokers
(short: 18.8% vs. long: 12.4%, p - 0.064).
141
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Table 5.2A: Baseline Demographic Variables By Smoking Status (Means presented as ±
(SD); categorical variables presented as N, (column percentage))
Demographic
Variable
Never/Never Smokers
(n = 490)
Current/Current
Smokers
(n = 490)
p-value*
Mean Weight (kg) 63.1 (9.8) 60.8 (8.7) <0.001
Mean Height (cm) 165.1 (6.3) 165.2 (6.4) 0.932
Mean Age
(at interview)
55.4 (7.4) 55.8 (7.1) 0.373
Mean BMI 23.2 (3.4) 22.3 (2.9) <0.001
BMI
<20 72 (14.7) 102 (20.8)
20 - <24 243 (49.6) 284 (58.0)
24 - < 28 144 (29.4) 84(17.1)
28+ 31 (6.3) 20(4.1) <0.001
Education
No Formal Education 34 (6.9) 65 (13.3)
Primary School 196 (40.0) 288 (59.5)
Secondary School 179 (36.5) 115(23.5)
A Level/Vocational 43 (8.8) 19(3.9)
University 38 (7.8) 3 (0.6) <0.001
Marital Status
Married 450 (91.8) 446(91.0)
Separated/Divorced 5(1.0) 10 (2.0)
Widowed 13 (2.7) 13 (2.6)
Never Married 22 (4.5) 21 (4.3) 0.635
Birthplace
Singapore 383 (78.2) 372 (75.9)
Malaysia 63 (12.9) 63 (12.9)
China/Hong Kong/
Macaw
39 (8.0) 50 (10.2)
Other 5(1.0) 5(1.0) 0.678
Dialect Group
Cantonese 265154.1) 192 (39.2)
Hokkien 225 (45.9) 298 (60.8) <0.001
Years living in
Singapore
<25 7(1.4) 9(1.8)
25+ 483 (98.6) 481 (98.2) 0.614
Family History, Lung
Cancer
12 (2.5) 13 (2.7) 0.839
Family History,
Any Cancer
80(16.3) 76(15.5) 0.727
'Chi-square test for categorical variables and t-test for continuous variables
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Table 5.2B: Baseline Demographic Variables By Genotype (Means presented as ±
(SD); categorical variables presented as N, (column percentage))
Demographic Variable Long Allele
(n = 391)
Short Allele
(n = 564)
p-value1
Mean Weight (kg) 62.2 (9.3) 61.9X9.3) 0.632
Mean Height (cm) 165.6 (6.3) 164.9 (6.4) 0.089
Mean Age
(at interview)
55.3 (7.2) 55.8 (7.3) 0.327
Mean BMI 22.7 (3.2) 22.8 (3.2) 0.683
BMI
<20 71 (18.2) 97 (17.2)
20 - <24 206 (52.7) 307 (54.4)
24 - < 28 93 (23.8) 131 (23.2)
28+ 21 (5.4) 29 (5.1) 0.959
Education
No Formal Education 42 (10.7) 54 (9.6)
Primary School 182(46.6) 290 (51.4)
Secondary School 120 (30.7) 166 (29.4)
A Level/Vocational 29 (7.4) 31 (5.5)
University 18(4.6) 23 (4.1) 0.550
Marital Status
Married 358 (91.6) 517(91.7)
S eparated/Di vorced 5(1.3) 9(1.6)
Widowed 9 (2.3) 16(2.8)
Never Married 19 (4.9) 22 (3.9) 0.824
Birthplace
Singapore 308 (78.8) 427 (75.7)
Malaysia 47 (12.0) 76 (13.5)
China/Hong Kong/ Macaw 35 (9.0) 52 (9.2)
Other 1 (0.3) 9(1.6) 0.200
Dialect Group
Cantonese 185 (47.3) 255 (45.2)
Hokkien 206 (52.7) 309 (54.8) 0.522
Years living in Singapore
<25 5(1.3) 11 (2.0)
25+ 386 (98.7) 553 (98.1) 0.427
Family History, Lung
Cancer
9 (2.3) 14 (2.5) 0.858
Family History, Any Cancer 51 (13.0) 100(17.7) 0.051
'Chi-square test for categorical variables and t-test for continuous variables
143
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Table 5.2C: Baseline Demographic Variables By Smoking Status and Genotype (Means presented as ± (SD);
categorical variables presented as N, (column percentage))
Never/Never Smokers Current/Current Smokers
Demo-graphic
Variable
Long Allele
(n = 193)
Short Allele
(n = 276)
P-
value1 ,2
Long Allele
(n = 198)
Short Allele
(n = 288)
P-
value1 ,3
M e a n W e ig h t
(k g )
6 4 .4 ( 9 .4 ) 6 2 .5 ( 9 .9 ) 0 .0 4 1 6 0 . 0 ( 8 . 8 ) 6 1 .3 (8 .7 ) 0 .1 2 2
M e a n H e ig h t
(c m )
1 6 6 .2 ( 6 .4 ) 1 6 4 .4 (6 .2 ) 0 .0 0 2 1 6 5 .0 ( 6 .1 ) 1 6 5 .3 (6 .6 ) 0 .5 4 9
M e a n A g e
( a t in te r v ie w )
5 5 .3 ( 7 .5 ) 5 5 .4 (7 .4 ) 0 .9 0 3 5 5 .3 (6 .9 ) 5 6 .1 (7 .2 ) 0 .2 0 2
M e a n B M I 2 3 .3 ( 3 .2 ) 2 3 .1 (3 .6 ) 0 .5 8 3 2 2 .1 (3 .0 ) 2 2 .4 (2 .8 ) 0 .1 9 3
B M I
< 2 0 2 7 ( 1 4 .0 ) 3 9 ( 1 4 .1 ) 5 8 ( 2 0 .1 ) 4 4 ( 2 1 .8 )
2 0 - < 2 4 9 0 ( 4 6 .6 ) 143 ( 5 1 .8 ) 1 6 4 (5 6 .9 ) 1 2 0 ( 5 9 .4 )
2 4 - < 2 8 6 3 ( 3 2 .6 ) 7 7 ( 2 7 .9 ) 5 4 ( 1 8 .8 ) 3 0 ( 1 4 .9 )
2 8 + 13 ( 6 .7 ) 1 7 ( 6 .2 ) 0 .6 7 5 1 2 ( 4 .2 ) 8 ( 4 .0 ) 0 .7 1 9
E d u c a tio n
N o F o r m a l
E d u c a tio n
15 ( 7 .8 ) 1 7 ( 6 .2 ) 2 7 ( 1 3 .6 ) 3 7 ( 1 2 .9 )
P r im a r y S c h o o l 6 3 ( 3 2 .6 ) 1 23 ( 4 4 .6 ) 1 1 9 ( 6 0 .1 ) 1 6 7 ( 5 8 .0 )
S e c o n d a r y
S c h o o l
7 9 ( 4 0 .9 ) 9 3 ( 3 3 .7 ) 4 1 ( 2 0 .7 ) 7 3 ( 2 5 .4 )
A L e v e l/
V o c a tio n a l
18 ( 9 .3 ) 2 3 (8 .3 ) 11 ( 5 .6 ) 8 ( 2 . 8 )
U n iv e r s ity 1 8 ( 9 .3 ) 2 0 (7 .3 ) 0 .1 4 3 0 ( 0 ) 3 ( 1 . 0 ) 0 .2 3 1
M a rita l S ta tu s
M a r r ie d 1 7 6 ( 9 1 .2 ) 2 5 7 ( 9 3 .1 ) 1 8 2 ( 9 1 .9 ) 2 6 0 ( 9 0 .3 )
S e p a r a te d /
D iv o r c e d
1 (0 .5 ) 3 ( 1 . 1 ) 4 ( 2 .0 ) 6 ( 2 . 1 )
W id o w e d 7 ( 3 . 6 ) 5 ( 1 . 8 ) 2 ( 1 . 0 ) 11 ( 3 .8 )
N e v e r M a r r ie d 9 ( 4 .7 ) 11 (4 .0 ) 0 .5 6 0 1 0 ( 5 .1 ) 11 (3 .8 ) 0 .2 7 1
B ir th p la c e
S in g a p o r e 151 ( 7 8 .2 ) 2 1 5 (7 7 .9 ) 1 5 7 ( 7 9 .3 ) 2 1 2 ( 7 3 .6 )
M a la y s ia 2 6 ( 1 3 .5 ) 3 5 ( 1 2 .7 ) 21 ( 1 0 .6 ) 41 ( 1 4 .2 )
C h in a /
H o n g K o n g /
M a c a w
15 (7 .8 ) 2 2 (8 .0 ) 2 0 ( 1 0 .1 ) 3 0 ( 1 0 .4 )
O th e r 1 ( 0 .5 ) 4 ( 1 . 5 ) 0 .8 0 5 0 ( 0 ) 5 ( 1 . 7 ) 0 .1 6 0
D ia le c t G r o u p
C a n to n e s e 1 1 0 ( 5 7 .0 ) 1 4 0 ( 5 0 .7 ) 7 5 ( 3 7 .9 ) 115 ( 3 9 .9 )
H o k k ie n 8 3 ( 4 3 .0 ) 1 3 6 ( 4 9 .3 ) 0 .1 8 0 1 2 3 ( 6 2 .1 ) 17 3 ( 6 0 .1 ) 0 .6 4 9
Y e a rs liv in g in
S in g a p o r e
< 2 5 2 ( 1 . 0 ) 5 ( 1 . 8 ) 3 ( 1 . 5 ) 6 ( 2 . 1 )
2 5 + 191 ( 9 9 .0 ) 2 7 1 ( 9 8 .2 ) 0 .4 9 6 1 9 5 ( 9 8 .5 ) 2 8 2 ( 9 7 .9 ) 0 .6 4 8
F a m ily
H is to ry , L u n g
C a n c e r
3 ( 1 . 6 ) 7 (2 .5 ) 0 .4 6 9 6 ( 3 .0 ) 7 ( 2 .4 ) 0 .6 8 7
F a m ily
H is to ry , A n y
C a n c e r
2 4 ( 1 2 .4 ) 5 2 ( 1 8 .8 ) 0 .0 6 4 2 7 ( 1 3 .6 ) 4 8 ( 1 6 .7 ) 0 .3 6 4
'C h i- s q u a r e te s t f o r c a te g o r ic a l v a r ia b le s a n d t- te s t f o r c o n tin u o u s v a r ia b le s
" C o m p a r e s lo n g v s . s h o r t a lle le s a m o n g N e v e r /N e v e r S m o k e r s
" C o m p a r e s lo n g v s. s h o r t a lle le s a m o n g C u r r e n t/C u r r e n t S m o k e r s
144
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
5.3.2 Personal History o f Illness
I also examined the role o f smoking status (Table 5.3A), genotype (Table 5.3B) and
genotype stratified by smoking status (Table 5.3C) on the personal history o f various
illnesses. Specifically, history o f hypertension, diabetes, arthritis, hip fracture,
allergic rhinitis, eczem a and enlarged prostate are all statistically significantly more
common among nonsmokers than smokers (p < 0.05 for all comparisons). It remains
unclear as to why any o f these disorders might be more common among
nonsmokers. A s might be predicted, history o f chronic cough and/or phlegm was
statistically significantly more common among smokers than nonsmokers (p < 0.002
for all comparisons) as was history o f shortness o f breath (p = 0.008).
Statistically significant differences in personal history o f illness were also observed
by genotype (Table 5.3B). Specifically, history o f heart attack/angina, diabetes and
gout were all more common among subjects carrying the long allele than subjects
with the short allele (p = 0.041, p = 0.053 and p = 0.012, respectively). In contrast,
allergic rhinitis, day/night cough and cough for more than three months per year
were more common among subjects with the short allele (p = 0.028, p = 0.007 and p
= 0.058, respectively).
After controlling for smoking status, only heart attack/angina and gout remained
statistically significantly associated with carriers o f the long allele (p = 0.042 and p =
0.049, respectively) and only among nonsmokers. In contrast, only day/night cough
and cough lasting at least three months per year remained associated with the short
allele (p = 0.002 and p = 0.055, respectively) and only among the smokers.
145
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Table 5.3A: Proportion of Subjects with a Personal History of Illness at Follow-up
by Smoking Status (Presented as N, (column percentage))
Condition Never/Never
Smokers
(n = 490)
Current/Current
Smokers
(n = 490)
p-value1
Hypertension 209 (42.7) 123 (25.1) <0.001
Heart Attack/Angina 41 (8.4) 36 (7.4) 0.553
Stroke 16(3.3) 17 (3.5) 0.859
Diabetes 88(18.0) 44 (9.0) <0.001
Arthritis 44 (9.0) 24 (4.9) 0.012
Gout 23 (4.7) 15(3.1) 0.186
Ulcer 35 (7.1) 46 (9.4) 0.202
Polyps 7 (1 .4 ) 5 (1 .0 ) 0.561
Hip Fracture 4 (0.8) 0 (0 ) 0.045
Other Bone Fracture 51 (10.4) 63 (12.9) 0.232
Glaucoma 7 (1 .4 ) 5 (1 .0 ) 0.561
Cataract 86 (17.6) 85 (17.4) 0.933
Parkinson’s Disease 2 (0.4) 0 (0 ) 0.157
Allergic Rhinitis 12 (2.5) 3 (0.6) 0.019
Sinusitis 8 (1 .6 ) 12(2.5) 0.366
Hay Fever 0 (0) 1 (0.2) 0.317
Eczema 41 (8.4) 2214.5) 0.013
Cholecystectomy 10(2.0) 14 (2.9) 0.408
Enlarged Prostate 24 (4.9) 11 (2.2) 0.025
Morning Cough 17(3.5) 49J10.0) <0.001
Day/Night Cough 20 (4.1) 44 (9.0) 0.002
Cough > 3 mo/year 19 (3.9) 48 (9.8) <0.001
> 5 Years Coughing 8 (1 .6 ) 27 (5.5) <0.001
Morning Phlegm 33 (6.7) 91 (18.6) <0.001
Day/Night Phlegm 15 (3.1) 62 (12.7) <0.001
Phlegm > 3 mo/year 27 (5.5) 92(18.8) <0.001
> 5 Years w / Phlegm 18 (3.7) 46 (9.4) <0.001
Asthma 22 (4.5) 1H 2 .7 ) 0.121
Asthma Attack in Last ^
12 mo.
(n = 35)
8 (36.4) 6 (46.2) 0.568
Shortness o f Breath 24 (4.9) 48 (9.8) 0.008
Blood Transfusion 46 (9.4) 36 (7.4) 0.249
'Chi-Square Test
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Table 5.3B: Proportion of Subjects with a Personal History of Illness at Follow-up
by Genotype (Presented as N, (column percentage))
Condition Long Allele
(n = 391)
Short Allele
(n = 564)
p-value1
Hypertension 125 (32.0) 199 (35.3) 0.287
Heart Attack/Angina 40 (10.2) 37 (6.6) 0.041
Stroke 12(3.1) 20 (3.6) 0.687
Diabetes 62(15.9) 65 (11.5) 0.053
Arthritis 27 (6.9) 39 (6.9) 0.995
Gout 23 (5.9) 15 (2.7) 0.012
Ulcer 30 (7.7) 49 (8.7) 0.575
Polyps 6 (1 .5 ) 6 (1 .1 ) 0.521
Hip Fracture 1 (0.3) 3 (0.5) 0.516
Other Bone Fracture 46(11.8) 66(11.7) 0.976
Glaucoma 5 (1 .3 ) 7 (1 .2 ) 0.959
Cataract 71 (18.2) 96 (17.0) 0.649
Parkinson’s Disease 0 (0 ) 1 (0.2) 0.405
Allergic Rhinitis 2 (0.5) 13 (2.3) 0.028
Sinusitis 6 (1 .5 ) 13 (2.3) 0.402
Hay Fever 0 (0 ) 1 (0.2) 0.405
Eczema 26 (6.7) 34 (6.0) 0.697
Cholecystectomy 9 (2.3) 15 (2.7) 0.728
Enlarged Prostate 17(4.4) 17(3.0) 0.274
Morning Cough 21 (5.4) 42 (7.5) 0.204
Day/Night Cough 15 (3.8) 46 (8.2) 0.007
Cough > 3 mo/year 19 (4.9) 45 (8.0) 0.058
>5 Years Coughing 9 (2.3) 25 (4.4) 0.133
Morning Phlegm 49(12.5) 74(13.1) 0.789
Day/Night Phlegm 29 (7.4) 48 (8.5) 0.542
Phlegm > 3 mo/year 49(1 2 .5 ) 69 (12.2) 0.891
>5 Years w / Phlegm 25 (6.4) 38 (6.7) 0.923
Asthma 14(3.6) 21 (3.7) 0.908
Asthma Attack in Last
12 mo.
(n = 35)
7 (50.0) 7 (3 3 .3 ) 0.324
Shortness o f Breath 31 (7.9) 40 (7.1) 0.828
Blood Transfusion 34 (8.7) 44 (7.8) 0.620
'Chi-Square Test
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Table 5.3C: Proportion of Subjects with a Personal History of Illness at Follow-up by Smoking Status and
Genotype (Presented as N, (column percentage))
N e v e r /N e v e r S m o k e r s C u r r e n t/C u r r e n t S m o k e r s
C o n d itio n Long Allele
(n = 1931
Short Allele
(n = 276)
P-
value1 ,2
Long Allele
(n = 198)
Short Allele
(n = 288)
P-
value1 ,3
H y p e r te n s io n 8 3 ( 4 3 .0 ) 1 1 8 ( 4 2 .8 ) 0 .9 5 7 4 2 ( 2 1 .2 ) 81 ( 2 8 .1 ) 0 .0 8 5
H e a r t A tta c k /
A n g in a
2 3 ( 1 1 .9 ) 1 8 ( 6 .5 ) 0 .0 4 2 1 7 ( 8 .6 ) 1 9 ( 6 .6 ) 0 .4 1 1
S tro k e 5 ( 2 .6 ) 1 0 ( 3 .6 ) 0 .5 3 2 7 (3 .5 ) 1 0 ( 3 .5 ) 0 .9 7 0
D ia b e te s 4 1 ( 2 1 .2 ) 4 3 ( 1 5 .6 ) 0 .1 1 5 2 1 ( 1 0 .6 ) 2 2 ( 7 .6 ) 0 .2 5 8
A r th r itis 1 6 ( 8 .3 ) 2 8 ( 1 0 .1 ) 0 .4 9 8 11 (5 .6 ) 11 ( 3 .8 ) 0 .3 6 6
G o u t 1 4 ( 7 .3 ) 9 ( 3 . 3 ) 0 .0 4 9 9 (4 .6 ) 6 ( 2 . 1 ) 0 .1 2 3
U lc e r 1 2 ( 6 .2 ) 2 2 (8 .0 ) 0 .4 7 1 18 ( 9 .1 ) 2 7 (9 .4 ) 0 .9 1 5
P o ly p s 4 ( 2 . 1 ) 3 ( 1 . 1 ) 0 .3 8 6 2 ( 1 . 0 ) 3 ( 1 . 0 ) 0 .9 7 3
H ip F r a c tu r e 1 ( 0 .5 ) 3 ( 1 . 1 ) 0 .5 1 0 0 ( 0 ) 0 ( 0 ) 1 N A
O th e r B o n e
F r a c tu r e
21 ( 1 0 .9 ) 2 8 ( 1 0 .1 ) 0 .7 9 8 2 5 (1 2 .6 ) 3 8 ( 1 3 .2 ) 0 .8 5 5
G la u c o m a 3 ( 1 . 6 ) 4 ( 1 . 5 ) 0 .9 2 6 2 ( 1 . 0 ) 3 ( 1 . 0 ) 0 .9 7 3
C a ta r a c t 3 3 ( 1 7 .1 ) 4 9 ( 1 7 .8 ) 0 .8 5 4 3 8 ( 1 9 .2 ) 4 7 ( 1 6 .3 ) 0 .4 1 3
P a r k in s o n ’s
D is e a s e
0 ( 0 ) 1 (0 .4 ) 0 .4 0 3 0 ( 0 ) 0 ( 0 ) N A
A lle r g ic R h in itis 2 ( 1 . 0 ) 1 0 ( 3 .6 ) 0 .0 8 1 0 ( 0 ) 3 ( 1 . 0 ) 0 .1 5 0
S in u s itis 3 ( 1 . 6 ) 5 ( 1 . 8 ) 0 .8 3 2 3 ( 1 . 5 ) 8 ( 2 .8 ) 0 .3 5 8
H a y F e v e r 0 ( 0 ) 0 ( 0 ) N A 0 ( 0 ) 1 ( 0 .4 ) 0 .4 0 7
E c z e m a 15 ( 7 .8 ) 2 3 (8 .3 ) 0 .8 2 6 1 1 ( 5 .6 ) 11 ( 3 .8 ) 0 .3 6 6
C h o le c y s t
e c to m y
3 ( 1 . 6 ) 7 (2 .5 ) 0 .4 6 9 6 ( 3 .0 ) 8 ( 2 .8 ) 0 .8 7 0
E n la r g e d
P r o s ta te
1 0 ( 5 .2 ) 13 (4 .7 ) 0 .8 1 6 7 ( 3 .5 ) 4 ( 1 . 4 ) 0 .1 1 8
M o r n in g C o u g h 7 ( 3 .6 ) 8 (2 .9 ) 0 .6 5 9 1 4 ( 7 .1 ) 3 4 ( 1 1 .8 ) 0 .0 8 6
D a y /N ig h t
C o u g h
7 ( 3 .6 ) 11 (4 .0 ) 0 .8 4 2 8 (4 .0 ) 3 5 ( 1 2 .2 ) 0 .0 0 2
C o u g h > 3
m o /y e a r
6 ( 3 . 1 ) 11 (4 .0 ) 0 .6 1 7 13 (6 .6 ) 3 4 ( 1 1 .8 ) 0 .0 5 5
> 5 Y e a rs
C o u g h in g
2 ( 1 . 0 ) 6 (2 .2 ) 0 .5 9 1 7 ( 3 . 5 ) 1 9 ( 6 .6 ) 0.111
M o r n in g P h le g m 1 2 ( 6 .2 ) 21 (7 .6 ) 0 .5 6 2 3 7 ( 1 8 .7 ) 5 3 ( 1 8 .4 ) 0 .9 3 7
D a y /N ig h t
P h le g m
6 ( 3 . 1 ) 9 ( 3 .3 ) 0 .9 2 7 2 3 ( 1 1 .6 ) 3 9 ( 1 3 .5 ) 0 .5 3 2
P h le g m > 3
m o /y e a r
1 0 ( 5 .2 ) 1 7 ( 6 .2 ) 0 .6 5 5 3 9 ( 1 9 .7 ) 5 2 ( 1 8 .1 ) 0 .6 4 9
> 5 Y e a rs w /
P h le g m
6 ( 3 . 1 ) 1 2 ( 4 .4 ) 0 .7 8 8 1 9 ( 9 .6 ) 2 6 (9 .0 ) 0 .9 0 9
A s th m a 8 ( 4 .2 ) 1 4 ( 5 .1 ) 0 .6 4 0 6 (3 .0 ) 7 ( 2 .4 ) 0 .6 8 7
A s th m a A tta c k
in L a s t 12 m o .
(n = 3 5 )
4 ( 5 0 .0 ) 4 ( 2 8 .6 ) 0 .3 1 5 3 ( 5 0 .0 ) 3 ( 4 2 .9 ) 0 .7 9 7
S h o r tn e s s o f
B re a th
1 4 ( 7 .3 ) 9 ( 3 . 3 ) 0 .0 6 8 1 7 ( 8 .6 ) 31 ( 1 0 .8 ) 0 .7 0 3
B lo o d T r a n s
fu s io n
19 (9 .8 ) 2 4 (8 .7 ) 0 .6 7 1 15 ( 7 .6 ) 2 0 (6 .9 ) 0 .7 9 1
‘C h i- S q u a r e T e s t
2C o m p a r e s lo n g v s. s h o r t a lle le s a m o n g N e v e r /N e v e r S m o k e r s
3C o m p a r e s lo n g v s . s h o r t a lle le s a m o n g C u r r e n t/C u r r e n t S m o k e r s
148
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
5.3.3 A ge at Diagnosis o f Illness
Next, I tested whether the mean age at diagnosis o f the above illnesses was
associated with smoking status (Table 5.4A), genotype (Table 5.4B) or genotype
stratified by smoking status (Table 5.4C). Briefly, both heart attack/angina and
sinusitis were diagnosed at a more advanced age among nonsmokers (p = 0.054 and
p = 0.004, respectively) while hypertension was diagnosed at a younger age
compared to smokers (p = 0.005).
When examined with respect to genotype (Table 5.4B), only age at diagnosis o f
intestinal polyps shows a statistically significant association. Specifically, subjects
with the short allele are diagnosed nearly 12 years earlier than subjects with the long
allele (p = 0.003). After adjusting for smoking status (Table 5.4C), the association
between polyps and genotype remains statistically significant only among the
nonsmokers (p = 0.017). Additionally, a statistically significant association is
observed for age at cholecystectomy among nonsmokers but not smokers, with
carriers o f the long allele being diagnosed nearly 23 years earlier than those with the
short allele. Caution is needed in interpreting these findings since they are based on
very small sample sizes and are thus strongly affected by outliers.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Table 5.4A: Mean Age (± SD) at Diagnosis by Smoking Status
Condition Never/Never Smokers
(n=490)
Current/Current
Smokers
(n=490)
p-value'
Hypertension 51.5(10.1) 54.8(10.2) 0.005
(n = 209) (n = 123)
Heart Attack/ Angina 59.8 (8.9) 56.0 (8.1) 0.054
(n = 41) (n = 36)
Stroke 57.7(10.6) 59.6(8.6) 0.565
(n = 16) (n = 17)
Diabetes 51.7(10.9) 54.7 (9.0) 0.115
(n = 88) (n = 44)
Arthritis 57.7 (10.4) 54.5 (8.9) 0.202
(n = 44) (n = 24)
Gout 55.0(10.8) 51.4(6.2) 0.247
(n = 23) (n = 1 5 )
Ulcer 47.3 (14.6) 47.6(11.0) 0.920
(n = 35) (n = 46)
Polyps 58.6(8.2) 61.0 (8.3) 0.626
(n = 7) (n = 5)
Hip Fracture 44.8 (14.2) NA NA
(n = 4) (n = 0)
Other Bone Fracture 42.5 (16.0) 39.7(18.2) 0.405
(n = 51) (n = 63)
Glaucoma 64.6(11.2) 63.0 (9.2) 0.803
(n = 7) (n = 5)
Cataract 61.7(8.4) 60.5 (10.2) 0.387
(n = 86) (n = 85)
Parkinson’s Disease 61.5 (7.8) NA NA
(n = 2) (n = 0)
Allergic Rhinitis 35.6(13.7) 25.3 (10.1) 0.252
(n = 12) (n = 3)
Sinusitis 51.6(14.2) 31.1 (13.5) 0.004
(n = 8)
(n=12)
Hay Fever NA 61.0 (NA) NA
(n = 0) (n = 1)
Eczema 49.7(16.3) 48.0(11.8) 0.675
(n = 41) (n = 22)
Cholecystectomy 53.8(13.9) 52.7(11.3) 0.835
(n = 10) (n = 14)
Enlarged Prostate 60.3 (7.3) 63.0(6.7) 0.304
(n = 24) (n=ll)
Asthma 29.4(19.4) 37.9 (26.5) 0.282
(n = 22) (n = 13)
Asthma Attack in 33.2(15.4) 27.7 (24.2) 0.531
Last 12 mo. (n = 14) (n = 7)
First Blood 41.5 (17.8) 41.7(16.7) 0.966
Transfusion (n = 46) (n = 36)
Number of Blood 1.5 (2.8) 1.3 (0.6) 0.542
Transfusions (n = 46) (n = 36)
'T-test
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Table 5.4B: Mean Age (± SD) at Diagnosis by Genotype
Condition Long Allele
(n = 391)
Short Allele
(n = 564)
p-value
Hypertension 52.3 (10.3) 53.1 (10.2) 0.499
(n = 125) (n = 199)
Heart Attack/Angina 58.7(9.1) 57.3 (8.3) 0.493
(n = 40) (n = 37)
Stroke 57.5 (8.9) 59.5 (10.3) 0.581
(n = 12) (n = 20)
Diabetes 53.7 (9.7) 51.6(10.7) 0.256
(n = 62) (n = 65)
Arthritis 57.3 (10.7) 56.3 (9.6) 0.685
(n = 27) (n = 39)
Gout 55.2(8.4) 51.2(10.6) 0.206
(n = 23) (n = 15)
Ulcer 48.5 (12.5) 46.9 (12.7) 0.589
(n = 30) (n = 49)
Polyps 53.7 (4.3) 65.5 (6.1) 0.003
(n = 6) (n = 6)
Hip Fracture 33.0 (NA) 48.7(14.5) NA
(n = 1) (n = 3)
Other Bone Fracture 39.9(16.4) 41.4(18.1) 0.667
(n = 46) (n = 66)
Glaucoma 66.4 (9.6) 62.1 (10.6) 0.494
(n = 5) (n = 7)
Cataract 60.5 (9.4) 61.8 (9.4) 0.376
(n = 71) (n = 96)
Parkinson’s Disease NA 56.0 (NA) NA
(n = 0) (n = 1)
Allergic Rhinitis 27.5 (0.7) 34.5 (14.3) 0.516
(n = 2) (n=13)
Sinusitis 42.0(19.4) 38.0(17.1) 0.654
(n = 6) (n=13)
Hay Fever NA 61.0 (NA) NA
(n = 0) (n = 1)
Eczema 50.0(15.2) 48.8 (14.3) 0.746
(n = 26) (n = 34)
Cholecystectomy 50.6(11.9) 54.7(12.5) 0.428
(n = 9) (n = 15)
Enlarged Prostate 59.8 (6.1) 62.8 (8.0) 0.220
(n = 17) (n = 17)
Asthma 36.2 (22.6) 30.1 (22.3) 0.439
(n = 14) (n = 21)
Asthma Attack in Last 35.7 (20.2) 29.2(17.7) 0.458
12 mo. (n = 7) (n = 14)
First Blood Transfusion 42.5(18.7) 42.0(16.6) 0.898
(n = 34) (n = 44)
Number of Blood 1.3 (0.7) 1.5 (2.9) 0.610
Transfusions (n = 34) (n = 44)
'T-test
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Table 5.4C: Mean Age (± SD) at Diagnosis by Smoking Status and Genotype
Never/Never Smokers C urrent/Current Smokers
Condition Long Allele
(n = 193)
Short Allele
(n = 276)
P -
value1 ,2
Long Allele
(n = 198)
Short Allele
(n = 288)
p " ,
yalue ’
H y p e r
te n s io n
5 1 . 4 ( 1 0 . 6 )
(n = 8 3 )
5 1 .7 (9 .7 )
( n = 1 1 8 )
0 .8 4 6 5 4 .1 (9 .5 )
(n = 4 2 )
5 5 .1 ( 1 0 .6 )
( n = 8 1 )
0 .5 8 6
H e a r t
A tta c k /
A n g in a
6 1 . 0 ( 9 . 6 )
(n = 2 3 )
5 8 .4 ( 8 .0 )
(n = 1 8 )
0 .3 6 8 5 5 . 6 ( 7 . 6 )
( n = 1 7 )
5 6 .3 ( 8 .7 )
( n = 1 9 )
0 .8 0 9
S tr o k e 5 4 . 8 ( 1 2 . 5 )
( n = 5 )
5 9 . 2 ( 1 0 . 6 )
(n = 1 0 )
0 .4 8 6 5 9 .4 (5 .7 )
(n = 7 )
5 9 . 8 ( 1 0 . 6 )
( n = 1 0 )
0 .9 3 4
D ia b e te s 5 2 . 9 ( 1 0 . 5 )
(n = 4 1 )
5 0 .5 (1 1 .1 )
(n = 4 3 )
0 .2 9 5 5 5 .1 (8 .2 )
( n = 2 1 )
5 3 . 8 ( 9 . 8 )
(n = 2 2 )
0 .6 4 6
A r th r itis 5 9 . 9 ( 1 1 . 7 )
( n = 1 6 )
5 6 .4 (9 .6 )
(n = 2 8 )
0 .2 8 8 5 3 .5 (8 .2 )
( n = l l )
5 6 . 0 ( 9 . 9 )
( n = l l )
0 .5 3 3
G o u t 5 6 .4 ( 9 .4 )
( n = 1 4 )
5 2 . 9 ( 1 3 . 1 )
(n = 9 )
0 .4 5 8 5 3 .2 (6 .5 )
( n = 9 )
4 8 . 7 ( 5 . 1 )
(n = 6 )
0 .1 7 5
U lc e r 4 9 .3 ( 1 5 .6 )
( n = 1 2 )
4 6 . 8 ( 1 4 . 4 )
( n = 2 2 )
0 .6 4 0 4 7 . 9 ( 1 0 . 3 )
(n = 1 8 )
4 6 . 9 ( 1 1 . 5 )
( n = 2 7 )
0 .7 7 6
P o ly p s 5 3 . 0 ( 3 . 6 )
( n = 4 )
6 6 .0 (6 .2 )
( n = 3 )
0 .0 1 7 5 5 .0 (7 .1 )
( n = 2 )
6 5 .0 ( 7 .2 )
( n = 3 )
0 .2 2 4
H ip F r a c tu r e 3 3 .0 (N A )
(n = 1)
4 8 . 7 ( 1 4 . 5 )
(n = 3 )
N A N A
(n = 0 )
N A
(n = 0 )
N A
O th e r B o n e
F r a c tu r e
4 0 .3 ( 1 2 .1 )
(n = 2 1 )
4 3 .5 ( 1 8 .8 )
(n = 2 8 1
0 .4 9 5 3 9 . 6 ( 1 9 . 5 )
(n = 2 5 )
3 9 .8 ( 1 7 .6 )
(n = 3 8 )
0 .9 7 5
G la u c o m a 6 5 .3 ( 1 0 .0 )
(n = 3 )
6 4 . 0 ( 1 3 . 6 )
(n = 4 )
0 .8 9 3 6 8 . 0 ( 1 2 . 7 )
( n = 2 )
5 9 . 7 ( 6 . 8 )
(n = 3 )
0 .3 9 5
C a ta r a c t 6 2 .2 (8 .0 )
(n = 3 3 )
6 1 . 9 ( 8 . 9 )
(n = 4 9 )
0 .8 7 9 5 9 . 0 ( 1 0 . 3 )
( n = 3 8 )
6 1 .7 ( 1 0 . 0 )
( n = 4 7 )
0 .2 3 0
P a r k i n s o n ’s
D is e a s e
N A
( n = 0 )
5 6 .0 (N A )
( n = l )
N A N A
( n = 0 )
N A
(n = 0 )
N A
A lle r g ic
R h in itis
2 7 .5 ( 0 .7 )
( n = 2 )
3 7 . 2 ( 1 4 . 6 )
f n = 1 0 )
0 .3 8 7 N A
( n = 0 )
2 5 .3 ( 1 0 .1 )
( n = 3 )
N A
S in u s itis 5 4 .3 ( 1 5 .0 )
( n = 3 )
5 0 . 0 ( 1 5 . 3 )
( n = 5 )
0 .7 0 9 2 9 .7 ( 1 6 . 2 )
( n = 3 )
3 0 .5 ( 1 4 .1 )
(n = 8)
0 .9 3 5
H a y F e v e r N A
(n = 0 )
N A
(n = 0 )
N A N A 6 1 .0 (N A )
(n = 1)
N A
E c z e m a 5 0 . 2 ( 1 8 . 7 )
( n = 1 5 )
5 0 . 0 ( 1 4 . 4 )
(n = 2 3 )
0 .9 7 7 4 9 . 8 ( 9 . 1 )
( n = 11)
4 6 . 2 ( 1 4 . 3 )
(n = 11)
0 .4 8 5
C h o le c y s t
e c to m y
3 8 .0 ( 7 .0 )
(n = 3 )
6 0 .6 ( 9 .7 )
(n = 7 )
0 .0 0 7 5 6 .8 (8 .0 )
(n = 6 )
4 9 . 6 ( 1 2 . 9 )
( n = 8 )
0 .2 5 3
E n la r g e d
P r o s ta te
5 9 .2 (6 .4 )
(n = 1 0 )
6 1 .5 (8 .2 )
( n = 1 3 )
0 .4 8 2 6 0 .6 (6 .0 )
( n = 7 )
6 7 .3 ( 6 .3 )
( n = 4 )
0 .1 1 4
A s th m a 3 0 . 5 ( 1 9 . 2 )
(n = 8 )
2 8 .8 ( 2 0 .2 )
( n = 1 4 )
0 .8 4 8 4 3 .8 ( 2 6 .3 )
(n = 6 )
3 2 .9 ( 2 7 .6 )
( n = 7 )
0 .4 8 1
L a s t A s th m a
A tta c k
3 6 .5 ( 1 8 .7 )
(n = 4 )
3 1 . 9 ( 1 4 . 8 )
(n = 10)
0 .6 3 2 3 4 .7 ( 2 6 .5 )
( n = 3 )
2 2 .5 ( 2 4 .8 )
( n = 4 )
0 .5 5 9
F ir s t B lo o d
T r a n s f u s io n
4 4 . 2 ( 1 9 . 7 )
( n = 1 9 )
4 1 . 2 ( 1 6 . 5 )
( n = 2 4 )
0 .5 8 7 4 0 .3 ( 1 7 .3 )
(n = 1 5 )
4 2 . 9 ( 1 7 . 0 )
J n = 2 0 )
0 .6 5 6
N o . o f B l o o d
T r a n s
f u s io n s
1.2 ( 0 .5 )
( n = 1 9 )
1.8 (3 .9 )
(n = 2 4 )
0 .4 5 6 1 .4 (0 .8 )
( n = 1 5 )
1 .2 (0 .5 )
(n = 2 0 )
0 .2 7 2
'T -te s t
2C o m p a r e s lo n g v s. s h o r t a lle le s a m o n g N e v e r /N e v e r S m o k e r s
3C o m p a r e s lo n g v s . s h o r t a lle le s a m o n g C u r r e n t/C u r r e n t S m o k e r s
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
5.3.4 Alcoholic Beverages
As expected, smoking status was highly associated with use o f alcoholic beverages,
especially beer and hard liquor (Table 5.5A). However, there were no apparent
associations between genotype and alcohol consumption (Table 5.5B) and no
associations between genotype and alcohol after adjusting for smoking status (Table
5.5C).
Table 5.5 A: U se o f Alcoholic Beverages at Baseline by Smoking Status (Means
presented as ± (SD); alcohol status presented as N , (column percentage))
Alcohol Never/Never
Smokers
(n = 490)
Current/Current
Smokers
(n = 490)
p-value
Alcohol Status
Nondrinker 351 (71.6) 300 (61.2)
Monthly Drinker 62 (12.7) 63 (12.9)_
Weekly Drinker 61 (12.5) 78 (15.9)
Daily Drinker 16(3.3) 49 (10.0) <0.001
Mean Alcohol
Frequency
(# times/ week)
0.6 (1.8) 1.3 (2.6) <0.0001
Mean Total
Alcohol
(# drinks/week)
1.1 (3.8) 3.0 (6.9) <0.0001
Mean Beer
(# drinks/week)
0.9 (3.3) 2.6 (6.5) <0.0001
Mean Rice Wine
(# drinks/week)
0.01 (0.3) 0.04 (0.5) 0.271
Mean Grape Wine
(# drinks/week)
0.03 (0.3) 0.04 (0.5) 0.856
Mean Hard Liquor
(# drinks/week)
0.1 (1.0) 0.3 (1.7) 0.049
'Chi-square test for categorical variables and t-test for continuous variables
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Table 5.5B: U se o f Alcoholic Beverages at Baseline by Genotype (Means presented
as ± (SD); alcohol status presented as N, (column percentage))
Alcohol Long Allele
(n = 391)
Short Allele
(n = 564)
p-value
Alcohol Status
Nondrinker 261 (66.8) 371 (65.8)
Monthly Drinker 4 9 (1 2 .5 ) 73 (12.9)
Weekly Drinker 57(14 .6) 81 (14.4)
Daily Drinker 24 (6.1) 39 (6.9) 0.963
Mean Alcohol
Frequency (#
times/week)
1.0 (2.4) 1.0 (2.2) 0.876
Mean Total
Alcohol
(# drinks/week)
2.1 (5.9) 2.0 (5.5) 0.729
Mean Beer (#
drinks/week)
1.8 (5.5) 1.7 (5.2) 0.783
Mean Rice Wine
(# drinks/week)
0.02 (0.4) 0.03 (0.4) 0.755
Mean Grape Wine
(# drinks/week)
0.03 (0.4) 0.04 (0.4) 0.771
Mean Hard Liquor
(# drinks/week)
0.3 (1.8) 0.2 (1.1) 0.591
'Chi-square test for categorical variables and t-test for continuous variables
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Table 5.5C: Use of Alcoholic Beverages at Baseline by Smoking Status and Genotype
(Means presented as ± (SD); alcohol status presented as N, (column percentage))
Never/Never Smokers Current/Current
Smokers
Alcohol Long
Allele
(n = 193)
Short
Allele
(n = 276)
P-
value1 ’ 2
Long
Allele
(n = 198)
Short
Allele
(n = 288)
p' ,
yalue
Alcohol
Status
Nondrinker 136 (70.5) 200 (72.5) 125 (63.1) 171 (59.4)
Monthly
Drinker
26(13.5) 33 (12.0) 23 (11.6) 40 (13.9) .
Weekly
Drinker
26(13.5) 34(12.3) 31 (15.7) 47(16.3)
Daily Drinker 5 (2.6) 9 (3.3) 0.906 19(9.6) 30 (10.4) 0.838
Mean Alcohol
Frequency
(# times/week)
0.6 (2.1) 0.6 (1.6) 0.607 1.3 (2.6) 1.3 (2.6) 0.895
Mean Total
Alcohol
(# drinks/week)
1.0 (3.5) 1.1 (4.1) 0.781 3.2 (7.5) 2.9 (6.5) 0.564
Mean Beer
(# drinks/week)
0.8 (2.8) 1.0 (3.7) 0.571 2.8 (7.1) 2.4 (6.1) 0.534
Mean Rice
Wine
(# drinks/week)
0.04 (0.5) 0(0) 0.232 0.01 (0.2) 0.06 (0.6) 0.245
Mean Grape
Wine
(# drinks/week)
(0.2) 0.04 (0.4) 0.757 0.04 (0.5) 0.04 (0.5) 0.884
Mean Hard
Liquor
(# drinks/week)
0.2 (1.1) 0.1 (1.0) 0.599 0.4 (2.2) 0.3 (1.2) 0.748
'Chi-square test for categorical variables and t-test for continuous variables
2 Compares long vs. short alleles among Never/Never Smokers
3 Compares long vs. short alleles among Current/Current Smokers
155
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
5.3.5 Passive Smoking
The effect o f passive smoking was also examined with respect to smoking status
(Table 5.6A), genotype (Table 5.6B) and genotype stratified by smoking status
(Table 5.6C). Not surprisingly, statistically significant associations were observed
between smoking status and passive smoking exposure under the age o f 18, over the
age o f 18, current exposure and job exposure (p < 0.001 for all comparisons). In
general, a greater proportion o f the smokers reported being exposed to passive
smoke than did nonsmokers. In particular, passive smoking under the age o f 18 was
statistically significantly more prevalent among smokers if the source was maternal
(p = 0.002) or from siblings (p < 0.001) whereas passive smoking over the age o f 18
was more prevalent among smokers for all sources except parents/in-laws (p < 0.001
for spouse, children or other sources). Moreover, the proportion o f smokers with
greater than 25 years o f passive smoking exposure over the age o f 18 was double
that o f nonsmokers (p < 0.001). Current passive smoking exposure was statistically
significantly more common among smokers if the source was from the spouse (p =
0.007). Last, job-related passive smoking exposure was much greater among
smokers, with a higher proportion o f them being exposed for greater than 20 years
and for more than four hours per day (p < 0.001).
When examined by genotype (Table 5.6B), passive smoking exposure under the age
o f 18 was statistically significantly more common among carriers o f the short allele
(p = 0.016) w hile the number o f years o f daily occupational passive smoking
exposure was marginally higher among those with the short allele (p - 0.053). After
stratifying by smoking status (Table 5.6C), the association between passive smoking
156
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
exposure under the age o f 18 and the short allele remained statistically significant
among smokers only (p = 0.026). In contrast, the long allele was associated with
four or more hours o f job exposure to passive smoking, but only among the
nonsmokers (p = 0.034).
T a b le 5 .6 A : P a s s iv e S m o k in g E x p o s u r e b y S m o k in g S ta tu s ( P r e s e n te d a s N , ( c o lu m n p e r c e n ta g e ) )
Passive Smoking Never/Never Smokers
(n = 490)
Current/Current Smokers
(n = 490)
p-value‘
Exposure Before Age 18
A n y H o m e E x p o s u r e 3 6 2 /4 9 0 ( 7 3 .9 ) 4 3 2 /4 9 0 ( 8 8 .2 ) < 0 .0 0 1
F a th e r D a ily S m o k e r 2 8 0 /3 6 2 ( 7 7 .4 ) 3 4 4 /4 3 2 ( 7 9 .6 ) 0 .4 3 5
M o th e r D a ily S m o k e r 8 6 /3 6 2 ( 2 3 .8 ) 1 4 5 /4 3 2 ( 3 3 .6 ) 0 .0 0 2
G r a n d p a r e n t( s ) D a ily S m o k e r 3 7 /3 6 2 ( 1 0 .2 ) 6 2 / 4 3 2 ( 1 4 .4 ) 0 .0 7 9
S ib lin g ( s ) D a ily S m o k e r 6 9 / 3 6 2 ( 1 9 .1 ) 1 6 5 /4 3 2 ( 3 8 .2 ) < 0 .0 0 1
O th e r D a ily S m o k e r 1 2 9 /3 6 2 ( 3 5 .6 ) 1 8 2 / 4 3 2 ( 4 2 .1 ) 0 .0 6 2
Y e a rs D a ily E x p o s u r e
< 1 3 /3 6 2 ( 0 .8 ) 1 /4 3 2 (0 .2 )
2 -5 1 2 /3 6 2 ( 3 .3 ) 8 / 4 3 2 ( 1 .9 )
6 -1 1 2 1 /3 6 2 (5 .8 ) 2 0 /4 3 2 ( 4 .6 )
1 2 + 3 2 6 /3 6 2 ( 9 0 .1 ) 4 0 3 /4 3 2 ( 9 3 .3 ) 0 .2 8 2
Exposure After Age 18
A n y H o m e E x p o s u r e 3 1 7 /4 9 0 ( 6 4 .7 ) 4 2 7 /4 9 0 ( 8 7 .1 ) < 0 .0 0 1
S p o u s e D a ily S m o k e r 2 /3 1 7 (0 .6 ) _ 3 6 /4 2 7 ( 8 .4 ) < 0 .0 0 1
P a r e n t/I n - la w s D a i l y S m o k e r 2 3 9 /3 1 7 ( 7 5 .4 ) 2 9 6 /4 2 7 ( 6 9 .3 ) 0 .0 6 8
C h ild ( r e n ) D a ily S m o k e r 3 4 / 3 1 7 ( 1 0 .7 ) 1 1 5 /4 2 7 ( 2 6 .9 ) < 0 .0 0 1
O th e r D a ily S m o k e r 1 3 3 / 3 1 7 ( 4 2 .0 ) 2 7 0 /4 2 7 ( 6 3 .2 ) < 0 .0 0 1
Y e a rs D a ily E x p o s u r e
< 1 2 / 3 1 7 ( 0 .6 ) 1 /4 2 7 (0 .2 )
2 -4 3 6 / 3 1 7 ( 1 1 .4 ) 2 3 /4 2 7 (5 .4 1
5 - 1 4 1 6 9 /3 1 7 ( 5 3 .3 ) 1 5 6 /4 2 7 ( 3 6 .5 )
1 5 -2 4 6 1 / 3 1 7 ( 1 9 .2 ) 1 0 7 /4 2 7 ( 2 5 .1 )
2 5 + 4 9 / 3 1 7 ( 1 5 .5 ) 1 4 0 /4 2 7 ( 3 2 .8 ) < 0 .0 0 1
Current Daily Exposure
A n y H o m e E x p o s u r e 2 8 /4 9 0 ( 5 /7 ) 1 2 7 /4 9 0 ( 2 5 .9 ) < 0 .0 0 1
S p o u s e 0 /2 8 (0 ) 2 7 /1 2 7 ( 2 1 .3 ) 0 .0 0 7
P a r e n ts /I n - la w s 0 /2 8 (0 ) 6 / 1 2 7 ( 4 .7 ) 0 .2 4 1
C h ild ( r e n ) 2 4 /2 8 ( 8 5 .7 ) 8 6 /1 2 7 ( 6 7 .7 ) 0 .0 5 8
O th e r 3 /2 8 ( 1 0 .7 ) 6 /1 2 7 (4 .7 ) 0 .2 2 0
J o b E x p o s u r e
A n y J o b E x p o s u r e 2 2 7 /4 9 0 (4 6 .3 ) 3 2 1 /4 9 0 ( 6 5 .5 ) < 0 .0 0 1
Y e a rs D a ily J o b E x p o s u r e
< 1 6 /2 2 7 (2 .6 ) 3 /3 2 1 (0 .9 )
2 -4 1 4 /2 2 7 (6 .2 ) 6 /3 2 1 (1 .9 )
5 -9 2 3 /2 2 7 ( 1 0 .1 ) 1 3 /3 2 1 ( 4 .1 )
1 0 -1 9 6 3 /2 2 7 ( 2 7 .8 ) 5 0 /3 2 1 ( 1 5 .6 )
2 0 + 1 2 1 /2 2 7 (5 3 .3 ) 2 4 9 /3 2 1 ( 7 7 .6 ) < 0 .0 0 1
H o u r s /D a y J o b E x p o s u r e
< 1 1 0 5 /2 2 7 (4 6 .3 ) 9 7 /3 2 1 ( 3 0 .2 )
1-3 6 6 /2 2 7 ( 2 9 .1 ) 7 1 /3 2 1 ( 2 2 .1 )
4 + 5 6 / 2 2 7 ( 2 4 .7 ) 1 5 3 /3 2 1 ( 4 7 .7 ) < 0 .0 0 1
C u r r e n t D a ily J o b E x p o s u r e 4 4 /2 2 7 ( 1 9 .4 ) 7 2 /3 2 1 ( 2 2 .4 ) 0 .3 9 0
'C h i- s q u a r e te s t
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Table 5.6B: Passive Smoking Exposure by Genotype (Presented as N, (column percentage))
Passive Sm oking Long Allele
(n = 391)
Short Allele
(n = 564)
p-value1
Exposure at < 18 Years
of Age
A n y H o m e E x p o s u r e 3 0 3 / 3 9 1 ( 7 7 .5 ) 4 7 2 / 5 6 4 ( 8 3 .7 ) 0 .0 1 6
F a t h e r D a i l y S m o k e r 2 3 7 / 3 0 3 ( 7 8 .2 ) 3 7 1 / 4 7 2 ( 7 8 .6 ) 0 .8 9 9
M o t h e r D a i l y S m o k e r 8 9 / 3 0 3 ( 2 9 .4 ) 1 3 9 / 4 7 2 ( 2 9 .5 ) 0 .9 8 2
G r a n d p a r e n t ( s ) D a i l y
S m o k e r
3 2 / 3 0 3 ( 1 0 .6 ) 6 2 / 4 7 2 ( 1 3 . 1 ) 0 .2 8 4
S i b l i n g ( s ) D a i l y S m o k e r 8 8 / 3 0 3 ( 2 9 .0 ) 1 4 2 / 4 7 2 ( 3 0 .1 ) 0 .7 5 7
O t h e r D a i l y S m o k e r 1 2 5 /3 0 3 ( 4 1 .3 ) 1 7 7 / 4 7 2 ( 3 7 .5 ) 0 .2 9 6
Y e a r s D a i l y E x p o s u r e
< 1 0 / 3 0 3 ( 0 ) 4 / 4 7 2 ( 0 .9 )
2 - 5 7 / 3 0 3 ( 2 .3 ) 1 2 / 4 7 2 ( 2 . 5 )
6 -1 1 1 3 /3 0 3 ( 4 .3 ) 2 5 / 4 7 2 ( 5 .3 )
1 2 + 2 8 3 / 3 0 3 ( 9 3 .4 ) 4 3 1 / 4 7 2 ( 9 1 .3 ) 0 .3 8 0
Exposure at > 18 Years
of Age
A n y H o m e E x p o s u r e 2 8 6 / 3 9 1 ( 7 3 .2 ) 4 3 8 / 5 6 4 ( 7 7 .7 ) 0 .1 0 9
S p o u s e D a i l y S m o k e r 1 3 /2 8 6 ( 4 .6 ) 2 4 / 4 3 8 ( 5 .5 ) 0 .5 7 7
P a r e n t / I n - l a w s D a i l y
S m o k e r
1 9 8 / 2 8 6 ( 6 9 .2 ) 3 2 2 / 4 3 8 ( 7 3 .5 ) 0 .2 1 0
C h i l d ( r e n ) D a i l y S m o k e r 5 1 / 2 8 6 ( 1 7 .8 ) 9 5 / 4 3 8 ( 2 1 .7 ) 0 .2 0 6
O t h e r D a i l y S m o k e r 1 5 5 / 2 8 6 ( 5 4 .2 ) 2 3 9 / 4 3 8 ( 5 4 .6 ) 0 .9 2 2
Y e a r s D a i l y E x p o s u r e
< 1 2 / 2 8 6 ( 0 .7 ) 1 /4 3 8 ( 0 .2 )
2 - 4 2 5 / 2 8 6 ( 8 .7 ) 3 1 / 4 3 8 ( 7 .1 )
5 - 1 4 1 3 4 / 2 8 6 ( 4 6 . 9 ) 1 8 5 / 4 3 8 ( 4 2 .2 )
1 5 - 2 4 5 9 / 2 8 6 ( 2 0 .6 ) 1 0 2 / 4 3 8 ( 2 3 .3 )
2 5 + 6 6 / 2 8 6 ( 2 3 .1 ) 1 1 9 /4 3 8 ( 2 7 .2 ) 0 .3 9 7
C urrent Daily Exposure
A n y H o m e E x p o s u r e 5 6 /3 9 1 ( 1 4 .3 ) 9 7 / 5 6 4 ( 1 7 .2 ) 0 .2 3 3
S p o u s e 1 1 / 5 6 ( 1 9 .6 ) 1 6 /9 7 ( 1 6 .5 ) 0 .6 2 3
P a r e n t s / I n - l a w s 2 / 5 6 ( 3 . 6 ) 4 / 9 7 ( 4 .1 ) 0 .8 6 5
C h i l d ( r e n ) 3 7 / 5 6 ( 6 6 .1 ) 7 2 / 9 7 ( 7 4 .2 ) 0 .2 8 3
O t h e r 5 /5 6 ( 8 .9 ) 4 / 9 7 ( 4 . 1 ) 0 .2 2 4
J o b E x p o s u r e
A n y J o b E x p o s u r e 2 1 2 / 3 9 1 ( 5 4 .2 ) 3 2 4 / 5 6 4 ( 5 7 .5 ) 0 .3 2 3
Y e a r s D a i l y J o b E x p o s u r e
< 1 3 / 2 1 2 ( 1 . 4 ) 4 / 3 2 4 ( 1 . 2 )
2 - 4 4 / 2 1 2 ( 1 . 9 ) 1 4 /3 2 4 ( 4 .3 )
5 - 9 9 / 2 1 2 ( 4 . 3 1 2 7 / 3 2 4 ( 8 .3 )
1 0 - 1 9 5 4 / 2 1 2 ( 2 5 .5 ) 5 7 / 3 2 4 ( 1 7 .6 )
2 0 + 1 4 2 / 2 1 2 ( 6 7 . 0 ) 2 2 2 / 3 2 4 ( 6 8 .5 ) 0 .0 5 3
H o u r s / D a y J o b E x p o s u r e
< 1 7 3 / 2 1 2 ( 3 4 . 4 ) 1 2 3 / 3 2 4 ( 3 8 .0 )
1 -3 5 1 / 2 1 2 ( 2 4 . 1 ) 8 5 / 3 2 4 ( 2 6 .2 )
4 + 8 8 / 2 1 2 ( 4 1 . 5 ) 1 1 6 /3 2 4 C 3 5 .8 ) 0 .4 1 2
C u r r e n t D a i l y J o b
E x p o s u r e
4 1 / 2 1 2 ( 1 9 . 3 ) 7 5 / 3 2 4 ( 2 3 .2 ) 0 .2 9 5
'C h i - s q u a r e t e s t
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
T a b le 5 .6 C : P a s s iv e S m o k in g E x p o s u r e b y S m o k in g S ta tu s a n d G e n o ty p e ( P r e s e n te d a s N , ( c o lu m n p e r c e n ta g e ) )
( c o n tin u e d o n n e x t p a g e )
Never/Never Smokers Current/Current Smokers
Passive
Smoking
Long Allele
(n = 193)
Short Allele
(n = 276)
P-
value1 ’ 2
Long Allele
(n = 198)
Short Allele
(n = 288)
P-
value1 ’ 3
<18 Years
of Age
A n y H o m e
E x p o s u r e
1 3 6 /1 9 3 ( 7 0 .5 ) 2 1 0 /2 7 6 ( 7 6 .1 ) 0 .1 7 3 1 6 7 /1 9 8 ( 8 4 .3 ) 2 6 2 /2 8 8 ( 9 1 .0 ) 0 .0 2 6
F a th e r D a ily
S m o k e r
1 0 6 /1 3 6 ( 7 7 .9 ) 1 6 1 /2 1 0 ( 7 6 .7 ) 0 .7 8 3 1 3 1 /1 6 7 ( 7 8 .4 ) 2 1 0 /2 6 2 ( 8 0 .2 ) 0 .6 6 9
M o th e r
D a ily
S m o k e r
3 7 /1 3 6 ( 2 7 .2 ) 4 8 / 2 1 0 ( 2 2 .9 ) 0 .3 5 9 5 2 /1 6 7 (3 1 .1 ) 9 1 /2 6 2 ( 3 4 .7 ) 0 .4 4 1
G r a n d
p a r e n t s )
D a ily
S m o k e r
1 1 /1 3 6 ( 8 .1 ) 2 1 / 2 1 0 ( 1 0 .0 ) 0 .5 4 9 2 1 / 1 6 7 ( 1 2 .6 ) 4 1 / 2 6 2 ( 1 5 .7 ) 0 .3 7 7
S ib lin g ( s )
D a ily
S m o k e r
2 6 /1 3 6 ( 1 9 .1 ) 4 1 / 2 1 0 ( 1 9 .5 ) 0 .9 2 6 6 2 /1 6 7 ( 3 7 .1 ) 1 0 1 /2 6 2 ( 3 8 .6 ) 0 .7 6 7
O th e r D a ily
S m o k e r
5 0 / 1 3 6 ( 3 6 .8 ) 7 1 / 2 1 0 ( 3 3 .8 ) 0 .5 7 3 7 5 / 1 6 7 ( 4 4 .9 ) 1 0 6 /2 6 2 ( 4 0 .5 ) 0 .3 6 3
Y e a rs D a ily
E x p o s u r e
< 1 0 / 1 3 6 ( 0 ) 3 / 2 1 0 ( 1 .4 ) 0 /1 6 7 (0 ) 1 /2 6 2 (0 .4 )
2 -5 4 /1 3 6 ( 2 .9 ) 7 / 2 1 0 ( 3 .3 ) 3 / 1 6 7 ( 1 .8 ) 5 / 2 6 2 ( 1 .9 )
6 -11 5 / 1 3 6 ( 3 .7 ) 1 3 / 2 1 0 ( 6 .2 ) 8 /1 6 7 (4 .8 ) 1 2 /2 6 2 ( 4 .6 )
1 2 + 1 2 7 /1 3 6 ( 9 3 .4 ) 1 8 7 /2 1 0 ( 8 9 .1 ) 0 .3 6 8 1 5 6 /1 6 7 ( 9 3 .4 ) 2 4 4 /2 6 2 ( 9 3 .1 ) 0 .8 8 4
>18 Years
of Age
A n y H o m e
E x p o s u r e
1 2 0 /1 9 3 ( 6 2 .2 ) 1 8 1 /2 7 6 ( 6 5 .6 ) 0 .4 4 9 1 6 6 /1 9 8 ( 8 3 .8 ) 2 5 7 /2 8 8 ( 8 9 .2 ) 0 .0 8 2
S p o u s e
D a ily
S m o k e r
0 /1 2 0 (0 ) 1 /1 8 1 (0 .6 ) 0 .4 1 5 1 3 /1 6 6 (7 .8 ) 2 3 /2 5 7 ( 9 .0 ) 0 .6 8 7
P a r e n t/I n
la w s D a ily
S m o k e r
9 1 /1 2 0 ( 7 5 .8 ) 1 3 6 /1 8 1 ( 7 5 .1 ) 0 .8 9 1 1 0 7 /1 6 6 ( 6 4 .5 ) 1 8 6 /2 5 7 ( 7 2 .4 ) 0 .0 8 5
C h ild ( r e n )
D a ily
S m o k e r
1 0 /1 2 0 ( 8 .3 ) 2 2 /1 8 1 ( 1 2 .2 ) 0 .2 9 2 4 1 /1 6 6 ( 2 4 .7 ) 7 3 /2 5 7 ( 2 8 .4 ) 0 .4 0 2
O th e r D a ily
S m o k e r
4 8 /1 2 0 ( 4 0 .0 ) 7 8 /1 8 1 ( 4 3 .1 ) 0 .5 9 4 1 0 7 /1 6 6 ( 6 4 .5 ) 1 6 1 / 2 5 7 ( 6 2 .7 ) 0 .7 0 6
Y e a rs D a ily
E x p o s u r e
< 1 1 /1 2 0 ( 0 .8 ) 1 /1 8 1 ( 0 .6 ) 1 /1 6 6 (0 .6 ) 0 /2 5 7 (0 )
2 - 4 1 5 / 1 2 0 ( 1 2 .5 ) 1 8 /1 8 1 ( 9 .9 ) 1 0 /1 6 6 (6 .0 ) 1 3 /2 5 7 (5 .1 )
5 - 1 4 6 5 /1 2 0 ( 5 4 .2 ) 9 9 /1 8 1 ( 5 4 .7 ) 6 9 /1 6 6 ( 4 1 .6 ) 8 6 /2 5 7 ( 3 3 .5 )
1 5 -2 4 2 3 / 1 2 0 ( 1 9 .2 ) 3 3 /1 8 1 ( 1 8 .2 ) 3 6 /1 6 6 ( 2 1 .7 ) 6 9 /2 5 7 ( 2 6 .9 )
2 5 + 1 6 / 1 2 0 ( 1 3 .3 ) 3 0 /1 8 1 ( 1 6 .6 ) 0 .9 0 2 5 0 /1 6 6 (3 0 .1 ) 8 9 /2 5 7 ( 3 4 .6 ) 0 .2 6 4
159
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Table 5.6C: Passive Smoking Exposure by Smoking Status and Genotype (Presented as N, (column
percentage)) (continued)
Never/Never Smokers Current/Current Smokers
Passive
Smoking
Long Allele
(n = 193)
Short Allele
(n = 276)
P-
value1 ’ 2
Long Allele
(n = 198)
Short Allele
(n = 288)
P-
value1 ,3
Current
Daily
Exposure
A n y H o m e
E x p o s u r e
9 /1 9 3 ( 4 .7 ) 1 9 /2 7 6 (6 .9 ) 0 .3 1 8 4 7 /1 9 8 ( 2 3 .7 ) 7 8 /2 8 8 ( 2 7 .1 ) 0 .4 0 7
S p o u s e 0 /9 (0 ) 0 / 1 9 ( 0 ) N A 1 1 /4 7 ( 2 3 .4 ) 1 6 /7 8 ( 2 0 .5 ) 0 .7 0 4
P a r e n ts /I n
la w s
0 /9 (0 ) 0 / 1 9 ( 0 ) N A 2 /4 7 (4 .3 ) 4 /7 8 ( 5 .1 ) 0 .8 2 5
C h ild ( r e n ) 7 /9 ( 7 7 .8 ) 1 7 / 1 9 ( 8 9 .5 ) 0 .4 0 9 3 0 /4 7 (6 3 .8 ) 5 5 /7 8 ( 7 0 .5 ) 0 .4 3 8
O th e r 2 /9 ( 2 2 .2 ) 1 / 1 9 ( 5 .3 ) 0 .1 7 5 3 /4 7 (6 .4 ) 3 /7 8 (3 .9 ) 0 .5 2 0
J o b
E x p o s u r e
A n y J o b
E x p o s u r e
8 7 /1 9 3 ( 4 5 .1 ) 1 3 0 /2 7 6 ( 4 7 .1 ) 0 .6 6 5 1 2 5 /1 9 8 ( 6 3 .1 ) 1 9 4 /2 8 8 ( 6 7 .4 ) 0 .3 3 5
Y e a rs
D a ily J o b
E x p o s u r e
< 1 1 / 8 7 ( 1 .2 ) 3 /1 3 0 (2 .3 ) 2 /1 2 5 ( 1 .6 ) 1 /1 9 4 ( 0 .5 )
2 - 4 3 /8 7 ( 3 .5 ) 9 / 1 3 0 ( 6 .9 ) 1 /1 2 5 ( 0 .8 ) 5 /1 9 4 ( 2 .6 )
5 -9 6 /8 7 ( 6 .9 ) 1 7 / 1 3 0 ( 1 3 .1 ) 3 /1 2 5 ( 2 .4 ) 1 0 /1 9 4 (5 .2 )
1 0 -1 9 3 1 /8 7 ( 3 5 .6 ) 3 0 / 1 3 0 ( 2 3 .1 ) 2 3 /1 2 5 ( 1 8 .4 ) 2 7 /1 9 4 ( 1 3 .9 )
2 0 + 4 6 /8 7 ( 5 2 .9 ) 7 1 / 1 3 0 ( 5 4 .6 ) 0 .1 7 5 9 6 /1 2 5 ( 7 6 .8 ) 1 5 1 /1 9 4 ( 7 7 .8 ) 0 .3 2 8
H o u r s /D a y
J o b
E x p o s u r e
< 1 3 9 /8 7 ( 4 4 .8 ) 6 0 / 1 3 0 ( 4 6 .2 ) 3 4 /1 2 5 ( 2 7 .2 ) 6 3 /1 9 4 ( 3 2 .5 )
1-3 2 0 /8 7 ( 2 3 .0 ) 4 6 /1 3 0 ( 3 5 .4 ) 3 1 /1 2 5 ( 2 4 .8 ) 3 9 /1 9 4 ( 2 0 .1 )
4 + 2 8 /8 7 ( 3 2 .2 ) 2 4 /1 3 0 ( 1 8 .5 ) 0 .0 3 4 6 0 /1 2 5 ( 4 8 .0 ) 9 2 /1 9 4 ( 4 7 .4 ) 0 .4 8 1
C u r r e n t
D a ily J o b
E x p o s u r e
1 4 / 8 7 ( 1 6 .1 ) 3 0 /1 3 0 ( 2 3 .1 ) 0 .2 1 0 2 7 /1 2 5 ( 2 1 .6 ) 4 5 /1 9 4 ( 2 3 .2 ) 0 .7 3 9
'C h i- s q u a r e te s t
2C o m p a r e s lo n g v s. s h o r t a lle le s a m o n g N e v e r /N e v e r S m o k e r s
3C o m p a r e s lo n g v s. s h o r t a lle le s a m o n g C u r r e n t/C u r r e n t S m o k e r s
5.3.6 Occupation
The relationship between occupation and smoking status (Table 5.7A), genotype
(Table 5.7B) and genotype stratified by smoking status (Table 5.7C) was also
assessed. Occupation was statistically significantly associated with smoking status
(p < 0.001), but not with genotype either before or after adjusting for smoking status.
In general, smokers were more likely to be craftsmen, laborers or plant workers
while nonsmokers were more likely to be employed as professionals or office
160
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
workers. This finding most probably reflects the difference in educational status
observed earlier (Table 5.2A).
T a b l e 5 . 7 A : P r o p o r t i o n o f S u b j e c t s i n V a r i o u s O c c u p a t i o n a l C a t e g o r i e s b y S m o k i n g S t a t u s ( P r e s e n t e d
a s N , ( c o l u m n p e r c e n t a g e ) )
Occupation Never/Never
Smokers
(n = 490)
Current/Current
Smokers
(n = 490)
p-value1
M a n a g e r o r a d m i n i s t r a t o r 1 2 ( 2 .5 ) 4 ( 0 . 8 )
P r o f e s s i o n a l 1 5 ( 3 . 1 ) 2 ( 0 . 4 )
T e c h n i c i a n o r
a s s o c i a t e p r o f e s s i o n a l
5 4 ( 1 1 . 0 ) 1 6 ( 3 . 3 )
C l e r i c a l w o r k e r 5 6 ( 1 1 . 4 ) 2 3 ( 4 . 7 )
S e r v i c e o r s a l e s w o r k e r 9 1 ( 1 8 . 6 ) 8 5 ( 1 7 . 4 )
A g r i c u l t u r a l o r
f i s h e r y w o r k e r
1 2 ( 2 .5 ) 9 ( 1 . 8 )
P r o d u c t i o n c r a f t s m a n o r
r e l a t e d t r a d e w o r k e r
1 2 8 ( 2 6 . 1 ) 1 5 9 ( 3 2 .5 )
P l a n t / m a c h i n e o p e r a t o r o r
a s s e m b l e r
9 2 ( 1 8 . 8 ) 1 0 0 ( 2 0 .4 )
C l e a n e r , l a b o r e r o r
r e l a t e d w o r k
3 0 ( 6 . 1 ) 9 1 ( 1 8 . 6 )
N e v e r w o r k e d 0 ( 0 ) ] 1 ( 0 . 2 ) < 0 . 0 0 1
'C h i - s q u a r e t e s t
T a b l e 5 . 7 B : P r o p o r t i o n o f S u b j e c t s i n V a r i o u s O c c u p a t i o n a l C a t e g o r i e s b y G e n o t y p e ( P r e s e n t e d a s N ,
( c o l u m n p e r c e n t a g e ) )
Occupation Long Allele
(n = 391)
Short Allele
(n = 564)
p-value
M a n a g e r o r a d m i n i s t r a t o r 1 0 ( 2 . 6 ) 6 ( 1 . 1 )
P r o f e s s i o n a l 9 ( 2 .3 ) 8 ( 1 . 4 )
T e c h n i c i a n o r
a s s o c i a t e p r o f e s s i o n a l
2 8 ( 7 . 2 ) 4 1 ( 7 .3 )
C l e r i c a l w o r k e r 3 2 ( 8 . 2 ) 4 3 ( 7 .6 )
S e r v i c e o r s a l e s w o r k e r 6 9 ( 1 7 . 7 ) 1 0 3 ( 1 8 . 3 )
A g r i c u l t u r a l o r
f i s h e r y w o r k e r
9 ( 2 .3 ) 1 2 ( 2 . 1 )
P r o d u c t i o n c r a f t s m a n o r r e l a t e d t r a d e
w o r k e r
1 1 0 ( 2 8 . 1 ) 1 7 0 ( 3 0 . 1 )
P l a n t / m a c h i n e o p e r a t o r o r a s s e m b l e r 7 6 ( 1 9 . 4 ) 1 1 0 ( 1 9 . 5 )
C l e a n e r , l a b o r e r o r
r e l a t e d w o r k
4 8 ( 1 2 . 3 ) 7 0 ( 1 2 . 4 )
N e v e r w o r k e d 0 ( 0 ) 1 ( 0 .2 ) 0 . 8 0 9
'C h i - s q u a r e t e s t
161
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Table 5.7C: Proportion of Subjects in Various Occupational Categories by Smoking
Status and Genotype (Presented as N, (column percentage))
Never/Never Smokers Current/Current Smokers
Occupation Long
Allele
(n = 193)
Short Allele
(n = 276)
p " ,
yalue ’
Long
Allele
(n = 198)
Short
Allele
(n = 288)
P-
j^lue
Manager or
administrator
7 (3.6) 5 (1 .8 ) 3 (1 .5 ) 1 (0.4)
Professional 8 (4.2) 7 (2.5) 1 (0.5) 1 (0.4)
Technician or
associate
professional
22(11.4) 31 (11.2) 6 (3.0) 10(3.5)
Clerical
worker
24 (12.4) 29(10.5) 8 (4.0) 14 (4.9)
Service or
sales worker
33 (17.1) 54(19.6) 36(1 8 .2 ) 49 (17.0)
Agricultural
or
fishery worker
5 (2.6) 7 (2.5) 4 (2.0) 5 (1 .7 )
Production
craftsman or
related trade
worker
47 (24.4) 75 (27.2) 63 (31.8) 95 (33.0)
Plant/machine
operator or
assembler
37(1 9 .2 ) 49(17.8) 39(19.7) 61 (21.2)
Cleaner,
laborer or
related work
10(5.2) 19 (6.9) 38 (19.2) 51 (17.7)
Never worked 0(0) 0 (0 ) 0.845 0 (0 ) 1 (0.4) 0.946
'Chi-square test
2 Compares long vs. short alleles among Never/Never Smokers
3Compares long vs. short alleles among Current/Current Smokers
5.3.7 Physical Activity
I also examined the relationship between physical activity and smoking status (Table
5.8A), genotype (Table 5.8B) and genotype stratified by smoking status (Table 5.8C).
As expected, smokers were statistically significantly less likely than nonsmokers to
engage in strenuous sports (p < 0.001) or moderate activity (p < 0.001) and also spent
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
fewer hours per week so engaged (p < 0.001 for both comparisons). In contrast,
smokers spent less time each day sitting (p < 0.001) and were more likely to be
involved in vigorous work (p = 0.031), probably related to having occupations that
require more physical work than nonsmokers. There were no statistically significant
differences by genotype (Table 5.8B) or by genotype adjusted for smoking status
(Table 5.8C).
Based on the above analyses, it appears that a number o f variables require further
evaluation as possible confounders. Any variable that was statistically significantly
associated with either the genotype or smoking status during the preliminary analysis
was considered a potential confounder, except where the relationship appeared to be
causal (e.g., smoking status and morning cough). For example, body mass index,
education, dialect group, birthplace, and family history o f cancer (lung or any type)
were all considered potential confounders based on preliminary analyses as w ell as
prior knowledge. The preliminary analysis served to highlight several other variables
as possible confounders for which there was little to no reason to suspect their
involvement a priori. For instance, history o f diabetes and exposure to passive
smoking at hom e before age 18 were statistically significantly associated with both
the exposure (genotype) and the disease (smoking status), indicating a need for
further evaluation although I might not have selected them as possible confounders
based on prior knowledge alone. From the preliminary analyses, a list o f potential
confounders was generated for further evaluation as described in Section 5.2.3.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Table 5.8A: Physical Activity by Smoking Status (Presented as N, (column percentage))
Physical Activity Never/Never Smokers
(n = 490)
Current/Current Smokers
(n = 490)
p-value'
S tr e n u o u s s p o r ts ( Y /N ) 8 7 ( 1 7 .8 ) 2 5 ( 5 .1 ) < 0 .0 0 1
V ig o r o u s w o rk _ (Y /N ) 5 4 ( 1 1 .0 ) 7 7 ( 1 5 .7 ) 0 .0 3 1
M o d e r a te a c tiv ity
N o n e 3 3 9 ( 6 9 .2 ) 4 1 3 ( 8 4 .3 )
0 .5 - 3 h o u r s /w e e k 1 0 3 ( 2 1 .0 ) 4 9 ( 1 0 .0 )
4 + h o u r s /w e e k 4 8 (9 .8 ) 2 8 (5 .7 ) < 0 .0 0 1
V ig o r o u s w o r k o r s tr e n u o u s
s p o r ts (Y /N ) 1 3 5 ( 2 7 .6 ) 9 8 ( 2 0 .0 ) 0 .0 0 5
A v e ra g e H o u r s / d a y s le e p in g
5 3 5 (7 .1 ) 3 6 ( 7 .4 )
6 1 2 9 ( 2 6 .3 ) 113 ( 2 3 .1 )
7 171 ( 3 4 .9 ) 1 7 9 ( 3 6 .5 )
8 1 2 6 ( 2 5 .7 ) 131 ( 2 6 .7 )
9 1 9 ( 3 .9 ) 2 0 ( 4 .1 )
10 1 0 (2 .0 ) 11 ( 2 .2 ) 0 .9 2 2
A v e ra g e H o u r s /d a y s ittin g
< 2 1 0 ( 2 .0 ) 1 6 ( 3 .3 )
< 4 6 5 ( 1 3 .3 ) 113 ( 2 3 .1 )
< 6 9 5 ( 1 9 .4 ) 1 1 2 ( 2 2 .9 )
< 8 121 ( 2 4 .7 ) 9 0 ( 1 8 .4 )
< 10 6 6 ( 1 3 .5 ) 6 3 ( 1 2 .9 )
< 12 6 0 ( 1 2 .2 ) 3 2 ( 6 .5 )
< 14 3 9 (8 .0 ) 2 8 (5 .7 )
< 16 1 6 ( 3 .3 ) 2 0 ( 4 . 1 )
< 18 5 ( 1 . 0 ) 5 ( 1 . 0 )
> 18 13 (2 .7 ) 11 ( 2 .2 ) < 0 .0 0 1
A v e ra g e h o u r s /w e e k o f
m o d e r a te a c tiv ity
0 3 3 9 ( 6 9 .2 ) 4 1 3 ( 8 4 .3 )
0 .7 5 4 0 (8 .2 ) 18 (3 .7 )
2 .5 6 3 ( 1 2 .9 ) 31 (6 .3 )
5 2 8 ( 5 .7 ) 15 ( 3 .1 )
12 1 9 ( 3 .9 ) 1 2 ( 2 .5 )
3 0 .5 1 (0 .2 ) 1 ( 0 .2 ) < 0 .0 0 1
A v e ra g e h o u r s /w e e k o f
v ig o r o u s w o r k
0 4 3 6 ( 8 9 .0 ) 4 1 3 ( 8 4 .3 )
0 .7 5 1 2 ( 2 .5 ) 18 ( 3 .7 )
2 .5 1 5 ( 3 .1 ) 1 4 ( 2 .9 )
5 11 (2 .2 ) 1 7 ( 3 .5 )
12 13 (2 .7 ) 2 0 ( 4 .1 )
3 0 .5 3 (0 .6 ) 8 ( 1 . 6 ) 0 .2 2 8
A v e ra g e h o u r s /w e e k o f
s tr e n u o u s s p o r ts
0 4 0 3 ( 8 2 .2 ) 4 6 5 ( 9 4 .9 )
0 .7 5 31 ( 6 .3 ) 6 ( 1 . 2 )
2 .5 2 9 (5 .9 ) 14 (2 .9 )
5 2 0 (4 .1 ) 3 ( 0 .6 )
12 7 ( 1 . 4 ) 2 ( 0 .4 )
3 0 .5 0 ( 0 ) 0 ( 0 ) < 0 .0 0 1
'C h i- s q u a r e te s t
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Table 5.8B: Physical Activity by Genotype (Presented as N, (column percentage))
Physical Activity Long Allele
(n = 391)
Short Allele
(n = 564)
p-value‘
S tr e n u o u s s p o r ts ( Y /N ) 3 7 (9 .5 ) 7 0 ( 1 2 .4 ) 0 .1 5 5
V ig o r o u s w o r k ( Y /N ) 5 6 ( 1 4 .3 ) 7 3 ( 1 2 .9 ) 0 .5 4 0
M o d e r a te a c tiv ity
N o n e 3 0 2 ( 7 7 .2 ) 4 3 7 ( 7 7 .5 )
0 .5 - 3 h o u r s /w e e k 6 3 ( 1 6 J ) 7 9 ( 1 4 .0 )
4 + h o u r s /w e e k 2 6 (6 .7 ) 4 8 (8 .5 ) 0 .4 2 3
V ig o r o u s w o r k o r s tr e n u o u s s p o r ts
( Y /N ) 9 0 ( 2 3 .0 ) 1 3 6 ( 2 4 .1 ) 0 .6 9 5
H o u r s /d a y s le e p in g
5 3 0 ( 7 .7 ) 4 0 ( 7 .1 )
6 9 5 ( 2 4 .3 ) 1 3 7 ( 2 4 .3 )
7 1 4 4 ( 3 6 .8 ) 1 9 7 ( 3 4 .9 )
8 9 7 ( 2 4 .8 ) 1 5 5 ( 2 7 .5 )
9 17 (4 .4 ) 2 2 (3 .9 )
10 8 ( 2 .1 ) 13 (2 .3 ) 0 .9 5 0
H o u r s /d a y s ittin g
< 2 11 (2 .8 ) 15 (2 .7 )
< 4 7 2 ( 1 8 .4 ) 101 ( 1 7 .9 )
< 6 7 6 ( 1 9 .4 ) 1 2 5 ( 2 2 .2 )
< 8 9 9 ( 2 5 .3 ) 1 0 7 ( 1 9 .0 )
< 10 5 2 ( 1 3 .3 ) 7 3 (1 2 .9 )
< 12 3 0 ( 7 .7 ) 6 0 ( 1 0 .6 )
< 14 2 7 ( 6 .9 ) 3 9 (6 .9 )
< 16 1 2 ( 3 .1 ) 2 2 (3 .9 )
< 18 4 ( 1 . 0 ) 6 ( 1 . 1 )
> 18 8 ( 2 .1 ) 1 6 ( 2 .8 ) 0 .4 9 6
A v e ra g e h o u r s /w e e k o f m o d e r a te
a c tiv ity
0 3 0 2 ( 7 7 .2 ) 4 3 7 ( 7 7 .5 )
0 .7 5 21 (5 .4 ) 3 2 (5 .7 )
2 .5 4 2 ( 1 0 .7 ) 4 7 (8 .3 )
5 13 (3 .3 ) 3 0 ( 5 .3 )
12 13 (3 .3 ) 1 6 ( 2 .8 )
3 0 .5 0 ( 0 ) 1 2 (0 .4 ) 0 .4 0 6
A v e ra g e h o u r s /w e e k o f v ig o r o u s w o r k
0 3 3 5 ( 8 5 .7 ) 4 9 1 ( 8 7 .1 )
0 .7 5 15 (3 .8 ) 15 (2 .7 )
2 .5 1 3 ( 3 .3 ) 15 (2 .7 )
5 13 ( 3 .3 ) 15 (2 .7 )
12 1 2 ( 3 .1 ) 2 0 ( 3 .6 )
3 0 .5 3 (0 .8 ) 8 ( 1 .4 ) 0 .7 3 5
A v e ra g e h o u r s /w e e k o f s tr e n u o u s
s p o r ts
0 3 5 4 (9 0 .5 ) 4 9 4 ( 8 7 .6 )
0 .7 5 9 (2 .3 ) 2 4 (4 .3 )
2 .5 1 6 ( 4 .1 ) 2 7 (4 .8 )
5 8 ( 2 .1 ) 15 (2 .7 )
12 4 ( 1 . 0 ) 4 (0 .7 )
3 0 .5 0 ( 0 ) 0 ( 0 ) 0 .4 5 4
'C h i- s q u a r e te s t
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
T a b le 5 .8 C : P h y s ic a l A c tiv ity b y S m o k in g S ta tu s a n d G e n o ty p e ( P r e s e n te d as N , ( c o lu m n p e r c e n ta g e ) ) ( c o n tin u e d
o n n e x t p a g e )
Never/Never Smokers C urrent/Current Smokers
Physical
Activity
Long Allele
(n = 193)
Short Allele
(n = 276)
P-
value1 ’ 2
Long Allele
(n = 198)
Short Allele
(n = 288)
P-
value1 ,3
S tr e n u o u s
s p o r ts ( Y /N )
2 9 ( 1 5 .0 ) 5 4 ( 1 9 .6 ) 0 .2 0 5 8 ( 4 .0 ) 1 6 ( 5 .6 ) 0 .4 4 9
V ig o r o u s w o r k
(Y /N )
2 4 ( 1 2 .4 ) 2 8 ( 1 0 .1 ) 0 .4 3 7 3 2 ( 1 6 .2 ) 4 5 ( 1 5 .6 ) 0 .8 7 4
M o d e r a te
a c tiv ity
N o n e 1 3 2 ( 6 8 .4 ) 1 9 7 ( 7 1 .4 ) 1 7 0 ( 8 5 .9 ) 2 4 0 ( 8 3 .3 )
0 .5 - 3 h /w e e k 41 ( 2 1 .2 ) 5 3 ( 1 9 .2 ) 2 2 ( 1 1 .1 ) 2 6 ( 9 .0 )
4 + h /w e e k 2 0 ( 1 0 .4 ) 2 6 (9 .4 ) 0 .7 8 5 6 ( 3 .0 ) 2 2 ( 7 .6 ) 0 .0 8 5
V ig o r o u s w o r k
o r s tr e n u o u s
s p o r ts ( Y /N )
5 0 ( 2 5 .9 ) 7 9 ( 2 8 .6 ) 0 .5 1 7 4 0 ( 2 0 .2 ) 5 7 ( 1 9 .8 ) 0 .9 1 1
H o u r s /d a y
s le e p in g
5 13 ( 6 .7 ) 21 (7 .6 ) 1 7 ( 8 .6 ) 1 9 ( 6 .6 )
6 4 8 ( 2 4 .9 ) 7 2 ( 2 6 .1 ) 4 7 ( 2 3 .7 ) 6 5 ( 2 2 .6 )
7 7 7 ( 3 9 .9 ) 8 7 ( 3 1 .5 ) 6 7 ( 3 3 .8 ) 1 1 0 ( 3 8 .2 )
8 4 5 ( 2 3 .3 ) 7 7 ( 2 7 .9 ) 5 2 ( 2 6 .3 ) 7 8 ( 2 7 .1 )
9 7 ( 3 .6 ) 1 2 ( 4 .4 ) 1 0 ( 5 .1 ) 1 0 ( 3 .5 )
10 3 ( 1 . 6 ) 7 (2 .5 ) 0 .5 4 4 5 (2 .5 ) 6 ( 2 . 1 ) 0 .8 2 8
H o u r s /d a y
s ittin g
< ,2 4 ( 2 . 1 ) 6 ( 2 .2 ) 7 ( 3 .5 ) 9 ( 3 . 1 )
< 4 3 2 ( 1 6 .6 ) 2 9 ( 1 0 .5 ) 4 0 ( 2 0 .2 ) 7 2 ( 2 5 .0 )
< 6 31 ( 1 6 .1 ) 5 9 ( 2 1 .4 ) 4 5 ( 2 2 .7 ) 6 6 ( 2 2 .9 )
< 8 51 ( 2 6 .4 ) 6 6 ( 2 3 .9 ) 4 8 ( 2 4 .2 ) 4 1 ( 1 4 .2 )
< 10 2 3 ( 1 1 .9 ) 4 0 ( 1 4 .5 ) 2 9 ( 1 4 .7 ) 3 3 (1 1 .5 )
< 12 2 4 ( 1 2 .4 ) 3 4 ( 1 2 .3 ) 6 ( 3 .0 ) 2 6 (9 .0 )
< 14 15 ( 7 .8 ) 2 3 (8 .3 ) 1 2 ( 6 .1 ) 16 (5 .6 )
< 16 6 ( 3 . 1 ) 8 ( 2 .9 ) 6 ( 3 .0 ) 1 4 ( 4 .9 )
< 18 2 ( 1 . 0 ) 3 ( 1 . 1 ) 2 ( 1 . 0 ) 3 ( 1 . 0 )
> 18 5 ( 2 .6 ) 8 (2 .9 ) 0 .7 5 5 3 ( 1 . 5 ) 8 (2 .8 ) 0 .0 5 1
A v e ra g e
h o u r s /w e e k o f
m o d e r a te
a c tiv ity
0 1 3 2 ( 6 8 .4 ) 1 9 7 ( 7 1 .4 ) 1 7 0 ( 8 5 .9 ) 2 4 0 ( 8 3 .3 )
0 .7 5 13 ( 6 .7 ) 2 2 (8 .0 ) 8 ( 4 .0 ) 1 0 ( 3 .5 )
2 .5 2 8 ( 1 4 .5 ) 31 ( 1 1 .2 ) 1 4 ( 7 .1 ) 1 6 ( 5 .6 )
5 9 (4 .7 ) 1 9 ( 6 .9 ) 4 (2 .0 ) 11 ( 3 ,8 )
12 1 1 ( 5 .7 ) 6 (2 .2 ) 2 ( 1 . 0 ) 1 0 ( 3 .5 )
3 0 .5 0 ( 0 ) 1 1 (0 .4 ) 0 .2 3 0 0 ( 0 ) 1 ( 0 .4 ) 0 .3 6 6
A v e ra g e
h o u r s /w e e k o f
v ig o r o u s w o r k
0 1 6 9 ( 8 7 .6 ) 2 4 8 ( 8 9 .9 ) 1 6 6 ( 8 3 .8 ) 2 4 3 ( 8 4 .4 )
0 .7 5 5 ( 2 .6 ) 7 (2 .5 ) 1 0 ( 5 .0 ) 8 (2 .8 )
2 .5 6 ( 3 . 1 ) 8 (2 .9 ) 7 ( 3 . 5 ) 7 (2 .4 )
5 5 ( 2 .6 ) 6 (2 .2 ) 8 (4 .0 ) 9 ( 3 . 1 )
12 7 ( 3 . 6 ) 5 ( 1 .8 ) 5 (2 .5 ) 1 5 ( 5 .2 )
3 0 .5 1 ( 0 .5 ) 2 (0 .7 ) 0 .8 8 8 2 ( 1 . 0 ) 6 ( 2 . 1 ) 0 .3 8 1
166
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
T a b le 5 .8 C : P h y s ic a l A c tiv ity b y S m o k in g S ta tu s a n d G e n o ty p e ( P r e s e n te d a s N , ( c o lu m n p e r c e n ta g e ) )
( c o n tin u e d )
Never/Never Smokers C urrent/Current Smokers
Physical
Activity
Long Allele
(n = 193)
Short Allele
(n = 276)
P-
value1 ’ 2
Long Allele
(n = 198)
Short Allele
(n = 288)
P-
value1 ’ 3
A v e ra g e
h o u r s /w e e k o f
s tr e n u o u s
s p o r ts
0 1 6 4 ( 8 5 .0 ) 2 2 2 ( 8 0 .4 ) 1 9 0 ( 9 6 .0 ) 2 7 2 ( 9 4 .4 )
0 .7 5 8 ( 4 .2 ) 1 9 ( 6 .9 ) 1 ( 0 .5 ) 5 ( 1 . 7 )
2 .5 1 2 ( 6 .2 ) 1 7 ( 6 .2 ) 4 ( 2 .0 ) 1 0 ( 3 .5 )
5 6 ( 3 . 1 ) 1 4 ( 5 .1 ) 2 ( 1 . 0 ) 1 (0 .4 )
12 3 ( 1 . 6 ) 4 ( 1 . 5 ) 1 ( 0 .5 ) 0 ( 0 )
3 0 .5 0 ( 0 ) 0 ( 0 ) 0 .5 9 2 0 ( 0 ) 0 ( 0 ) 0 .3 2 9
'C h i- s q u a r e te s t
2C o m p a r e s lo n g v s . s h o r t a lle le s a m o n g N e v e r /N e v e r S m o k e r s
3C o m p a r e s lo n g v s . s h o r t a lle le s a m o n g C u r r e n t/C u r r e n t S m o k e r s
This list included, but was not limited to passive smoke exposure, activity levels,
alcohol usage, occupation, weight, BMI, education, dialect group, family history o f
cancer, birthplace, history o f and age at diagnosis o f eczema, enlarged prostate,
allergic rhinitis, hypertension, arthritis, hip fracture, Parkinson’s Disease, asthma,
heart attack/angina, and stroke.
5.3.8 M AO -A and Tobacco Addiction
This study was undertaken to assess whether or not M AO -A genotype might affect an
individual’s predisposition to take up smoking. Therefore, the results presented in
Tables 5.9 - 5.12 are o f primary interest.
Overall, M AO -A genotype did not predict smoking status (Table 5.9). Even after
adjusting for birthplace, dialect group, education, alcohol use, history o f hyper-
1 6 7
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
tension, history o f diabetes, passive smoke exposure at home before age 18, dialect
by history o f hypertension and history o f diabetes by history o f hypertension, subjects
with the short allele were 0.85 times as likely to be smokers as were subjects with the
long allele (95% Cl: 0.63, 1.14). Additional adjustment for age, year o f birth and
year o f interview made no appreciable difference in the estimate (data not shown).
Table 5.10 presents the unadjusted results o f various smoking-related variables at two
time points: at initial recruitment (baseline) and again at follow-up approximately 5-6
years later. At baseline, the age at which subjects began to smoke showed a
statistically significant trend by genotype. Specifically, subjects with the short allele
were more likely to begin smoking at an earlier age than subjects with the long allele
(Figure 5.2). Overall, 65.3% o f subjects with the short allele began smoking when
they were 19 years o f age or younger while only 53.6% o f subjects with the long
allele began smoking while still a teenager, a difference which was highly
statistically significant (p = 0.007). The number o f years that an individual smoked
regularly was not statistically significantly associated with M AO -A genotype,
however, there was some suggestion that very long-term smokers (40+ years) were
over-represented among those with the short allele (short: 43.1% vs. long: 36.9%)
(Figure 5.3). Likewise, subjects with the short allele were somewhat more likely to
be very heavy smokers (43+ cigs/day) compared to subjects with the long allele
(short: 4.9% vs. long: 2.5%), though the numbers in this category were quite small
and this difference was not statistically significant (Figure 5.4).
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
T a b l e 5 .9 : E s t i m a t e d O d d s R a t i o s a n d 9 5 % C o n f i d e n c e I n t e r v a l s f o r b e i n g a C u r r e n t / C u r r e n t S m o k e r
b y M A O - A G e n o t y p e ( n = 9 5 5 )
Variable Odds Ratio 95% Confidence Interval p-value
M A O - A G e n o t y p e
( u n a d j u s t e d )
L o n g 1 .0 0
S h o r t 1 .0 2 0 . 7 9 , 1 .3 2 0 . 8 9 7
M A O - A G e n o t y p e 1
L o n g 1 .0 0
S h o r t 0 .8 5 0 . 6 3 , 1 . 1 4 0 . 2 7 2
'A d j u s t e d f o r b i r t h p l a c e ( S i n g a p o r e , M a l a y s i a , O t h e r ) , d i a l e c t g r o u p ( C a n t o n e s e , H o k k i e n ) , e d u c a t i o n
( n o f o r m a l , p r i m a r y , s e c o n d a r y , A l e v e l / v o c a t i o n a l o r u n i v e r s i t y ) , a l c o h o l u s e ( n o n d r i n k e r , m o n t h l y
d r i n k e r , w e e k l y d r i n k e r , d a i l y d r i n k e r ) , h i s t o r y o f h y p e r t e n s i o n ( y e s / n o ) , h i s t o r y o f d i a b e t e s ( y e s / n o ) ,
p a s s i v e h o m e s m o k e e x p o s u r e b e f o r e a g e 1 8 ( y e s / n o ) , d i a l e c t X h i s t o r y o f h y p e r t e n s i o n , a n d h i s t o r y o f
d i a b e t e s X h i s t o r y o f h y p e r t e n s i o n
T a b l e 5 .1 0 : A c t i v e S m o k i n g a m o n g C u r r e n t / C u r r e n t S m o k e r s ( n = 4 9 0 ) b y G e n o t y p e ( M e a n s p r e s e n t e d
a s ± S D ; c a t e g o r i c a l v a r i a b l e s p r e s e n t e d a s N , ( c o l u m n p e r c e n t a g e ) )
Smoking Variable Long Allele
(n = 198)
Short Allele
(n = 288)
p-value
B a s e l i n e
A g e s t a r t e d s m o k i n g r e g u l a r l y 1
< 1 4 3 2 ( 1 6 . 2 ) 6 5 ( 2 2 .6 )
1 5 - 1 9 7 4 ( 3 7 . 4 ) 1 2 3 ( 4 2 . 7 )
2 0 - 2 9 8 0 ( 4 0 .4 ) 8 8 ( 3 0 . 6 )
3 0 4 - 1 2 ( 6 . 1 ) 1 2 ( 4 .2 ) 0 . 0 0 7
N u m b e r o f y e a r s s m o k e d r e g u l a r l y 1
< 2 0 8 ( 4 . 0 ) 8 ( 2 .8 )
2 0 - 2 9 3 8 ( 1 9 . 2 ) 4 7 ( 1 6 . 3 )
3 0 - 3 9 7 9 ( 3 9 . 9 ) 1 0 9 ( 3 7 . 9 )
4 0 4 - 7 3 ( 3 6 .9 ) 1 2 4 ( 4 3 . 1 ) 0 . 1 3 3
N u m b e r c i g s / d a y 1
7 - 1 2 4 9 ( 2 4 .8 ) 6 8 ( 2 3 . 6 )
1 3 - 2 2 1 0 6 ( 5 3 .5 ) 1 4 4 ( 5 0 . 0 )
2 3 - 3 2 1 9 ( 9 .6 ) 3 1 ( 1 0 . 8 )
3 3 - 4 2 1 9 ( 9 .6 ) 3 1 ( 1 0 . 8 )
4 3 + 5 ( 2 .5 ) 1 4 ( 4 . 9 ) 0 . 3 2 9
F o l l o w - u p
M e a n a g e a t f i r s t c i g a r e t t e 2 1 8 .1 ( 5 .0 ) 1 7 .4 ( 5 . 3 ) 0 . 1 2 4
M e a n a g e s m o k e d r e g u l a r l y 2 1 9 .8 ( 5 .8 ) 1 9 . 0 ( 6 . 0 ) 0 .1 9 1
M e a n n u m b e r c i g s / d a y 2 2 1 . 0 ( 1 0 . 0 ) 2 1 .1 ( 1 0 . 5 ) 0 . 8 9 5
'K m s k a l - W a l l i s e q u a l i t y o f p o p u l a t i o n s r a n k t e s t , a d j u s t e d f o r t i e s
2T - t e s t f o r c o n t i n u o u s v a r i a b l e s
169
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Figure 5.2: Age at Smoking Initiation by Genotype (Reported at Baseline)
Age at Starting to Smoke (at Baseline) by Genotype
14 o r y o u n g e r
% Long A llele
Figure 5.3: Number o f Cigarettes per Day by Genotype (Reported at Baseline)
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Figure 5.4: Number of Years Smoking Regularly by Genotype (Reported at Baseline)
No. of Years Smoking Regularly (at Baseline) by Genotype
<20 20-29 30-39 40+
% Long A llele m g % S h o rt A llele
The follow-up data showed a similar pattern as the baseline data, however, no
statistically significant results were observed (Table 5.10). In general, subjects with
the short allele had their first cigarette an average o f 8.4 months earlier and began
smoking regularly 9.6 months earlier than subjects with the long allele (p = 0.124 and
p = 0.191, respectively). After adjusting for education, passive smoke exposure at
home before age 18, passive smoke exposure at work, year o f birth, birthplace,
alcohol use and history o f blood transfusion, subjects with the short allele are
estimated to initiate smoking an average o f 10.4 (-0.87*12) months earlier (p =
0.068) and to begin smoking regularly 9.1 (-0.76*12) months earlier (p = 0.165) than
subjects with the long allele (Table 5.11). The average number o f cigarettes smoked
171
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
per day did not differ between the two genotypes, with subjects carrying both short
and long allele smoking approximately one pack per day (p = 0.895) (Table 5.10).
Adjusting for education, passive smoke exposure at home before age 18, passive
smoke exposure at work, year o f birth, birthplace, alcohol use, history o f heart
attack/angina, and history o f arthritis suggested that subjects with the short allele
smoke 0.32 more cigarettes than subjects with the long allele, a difference that was
not statistically significant (p = 0.733) (Table 5.11). Further adjustments for age and
year o f interview did not alter the estimate notably (data not shown).
At baseline, the unadjusted odds o f starting to smoke during adulthood rather than
during adolescence was reduced among subjects with the short allele (Table 5.12). In
particular, subjects with the short allele were 0.66 times as likely to begin smoking at
the age o f 30 years or greater compared to those with the long allele (p = 0.319).
Although none o f the odds ratios were statistically significant, there was a statistically
significant trend in reduced risk o f being in the older age groups at smoking initiation
for carriers o f the short allele (ptr e n d = 0.008). The follow-up data, while not
statistically significant, show a similar pattern in that subjects with the short allele
appear to be at a slightly reduced risk o f smoking initiation as they age (ORF ir s tC ig a r e tte
= 0.97, 95% C.I. = 0.94, 1.01; ORS m o k e R e g u la r I y = 0.98, Cl: 0.95, 1.01). Taken together,
these findings suggest that subjects with the short allele begin to smoke at earlier ages
than subjects with the long allele, therefore, their risk o f starting to smoke during
adulthood is reduced compared to those carrying the long allele.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
As noted previously, subjects who reported at baseline having smoked for forty or
more years are somewhat more likely to carry the short allele, but this difference was
not statistically significant (OR = 1.20, Cl: 0.85, 1.69) (Table 5.12). Moreover,
subjects who report smoking more than two packs o f cigarettes per day at baseline are
also more likely to have the short allele than the long allele, but again this difference
did not reach statistical significance (OR = 1.97, Cl: 0.70, 5.54), possibly due to the
small number o f subjects in this category. This effect was not observed in the follow-
up data.
Table 5.11: Estimated Coefficients and 95% Confidence Intervals for Short M AO-A
Genotype using Linear Regression M odels for A ge at Smoking Initiation/Regular
Smoking among Current/Current Smokers (n=486)
Variable Estimated
Coefficient
95%
Confidence
Interval
p-value
Age at First Cigarette1 -0.87 -1.81,0.07 0.068
Age Smoked Regularly1 -0.76 -1.84, 0.32 0.165
Number o f Cigarettes
Smoked2
0.32 -1.54, 2.19 0.733
'Adjusted for education (no formal, primary, secondary, A level/vocational, or university),
passive home smoking exposure before age 18 (yes/no), passive job smoking exposure
(yes/no), year of birth (in quartiles), birthplace (Singapore, Malaysia, Other), alcohol use
(nondrinker, monthly drinker, weekly drinker, or daily drinker) and history of blood
transfusion (yes/no)
2 Adjusted for education (no formal, primary, secondary, A level/vocational/university),
passive home smoking exposure before age 18 (yes/no), passive job smoking exposure
(yes/no), year of birth (in quartiles), birthplace (Singapore, Malaysia, Other), alcohol use
(nondrinker, monthly drinker, weekly drinker, or daily drinker), history of heart attack/angina
(yes/no), and history of arthritis (yes/no)
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
T a b le 5 .1 2 : O d d s R a t i o s ' a n d ( 9 5 % C o n f id e n c e I n te r v a ls ) f r o m a U n iv a r ia te A n a ly s is o f A c tiv e S m o k in g a m o n g
C u r r e n t/C u r r e n t S m o k e r s (n = 4 9 0 )
S m o k i n g V a r i a b l e N O d d s R a t i o 9 5 % C o n f i d e n c e I n t e r v a l p - v a l u e 4
B a s e lin e
A g e s ta r te d s m o k in g
r e g u la r ly
< 14 9 9 1 .0 0 (re f)
1 5 -1 9 1 98 1 .1 0 ( 0 .7 9 , 1 .5 3 ) 0 .5 8 8
2 0 - 2 9 1 6 9 0 .7 3 ( 0 .5 1 , 1 .0 3 ) 0 .0 7 0
304- 2 4 0 .6 6 ( 0 .2 9 , 1 .4 9 ) 0 .3 1 9
P tr e n f = 0 .0 0 8
N u m b e r o f y e a r s s m o k e d
r e g u la r ly
< 2 0 16 1 .0 0 (re f )
2 0 - 2 9 85 0 .8 8 ( 0 .5 5 , 1 .3 9 ) 0 .5 7 4
3 0 - 3 9 1 8 9 0 .9 8 ( 0 .6 9 , 1 .3 7 ) 0 .8 9 1
4 0 + 2 0 0 1 .2 0 ( 0 .8 5 , 1 .6 9 ) 0 .2 9 0
P . ™ / = 0 . 5 3 0
N u m b e r c ig s /d a y
7 - 1 2 118 1 .0 0 (re f)
1 3 -2 2 2 5 2 0 .9 6 ( 0 .7 1 , 1 .2 9 ) 0 .7 6 7
2 3 - 3 2 51 1 .1 5 ( 0 .6 3 , 2 .0 8 ) 0 .6 4 9
3 3 - 4 2 5 0 1 .1 5 ( 0 .6 3 , 2 .0 8 ) 0 .6 4 9
4 3 + 19 1 .9 7 ( 0 .7 0 , 5 .5 4 ) 0 .1 9 9
P m J = 0 .4 0 0
F o llo w - u p
A g e a t f ir s t c ig a r e tte 4 4 8 6 0 .9 7 0 .9 4 , 1.01 0 .1 2 5
A g e s m o k e d r e g u la r ly 4 4 8 6 0 .9 8 0 .9 5 , 1.01 0 .1 9 2
N u m b e r c ig s /d a y 4 4 8 6 1 .0 0 0 .9 8 , 1 .0 2 0 .8 9 4
'I n d ic a te th e o d d s o f b e in g g e n o ty p e = s h o r t
" U n iv a r ia te lo g is tic r e g r e s s io n
" U n iv a ria te lo g is tic r e g r e s s io n u s in g c a te g o r y m id p o in ts (a s c o n tin u o u s v a r ia b le )
4C o n tin u o u s v a r ia b le s
5.4 DISCUSSION
Although our understanding o f addiction is still rather theoretical and incomplete,
most researchers agree that some elements o f personality play an important role in the
addictive process. In particular, the sensation-seeking/novelty-seeking personality
trait has long been implicated in predisposing to substance abuse (Franques,
Auriacombe et al. 2000; Franques, Auriacombe et al. 2003) and has also been found
to be associated with lower platelet MAO (Zuckerman 1993) and higher plasma NE
(Gerra, Avanzini et al. 1999). Furthermore, both D A and 5-HT act as mediators in
the reward pathway that is activated when drugs o f abuse are consumed (Netter,
174
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Hennig et al. 1996), producing long-term alterations in the mesocorticolimbic DA
system (Merlo Pich, Chiamulera et al. 1999). Thus, the rationale for examining the
role o f the M AO -A gene in tobacco addiction appears to rest on a solid scientific
foundation.
Our finding o f a statistically significant trend in younger age at smoking initiation for
subjects with the short allele is in line with what is understood about the role o f
M AO-A in the development o f addictive disorders. Few studies have examined the
role o f M AO -A polymorphisms in relation to tobacco addiction directly, but those
that have produced relatively consistent results.
One study examined the role o f a dinucleotide CA repeat in intron 2 (Vanyukov,
M oss et al. 1995). The functional significance o f this repeat is unknown, however, it
is conserved in both M AO -A and M AO-B, suggesting a potential functional
importance (Shih, Grimsby et al. 1993). Vanyukov and colleagues found that the
“long” allele (> 1 1 5 bp) was overrepresented among males who were alcohol or
substance abusers, but this difference did not reach statistical significance (OR =
2.15, p > 0.17), possibly due to the very small sample size used in this study.
Additionally, the age o f drug abuse onset was statistically significantly lower among
subjects with the long allele (13.9 ± 0.78 vs. 15.6 ± 0.27, p = 0.03). Although the
functional relevance o f this polymorphism is unknown, it is in strong linkage
disequilbrium with the uVNTR polymorphism, with “short” dinucleotide repeats
tending to segregate with the 3-repeat allele o f the uVNTR. Although these results
are not consistent with our observation o f an earlier age at smoking initiation among
1 7 5
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
subjects with the short uVNTR allele, they do suggest that variability in the gene for
M AO-A m ay be important in age at onset o f drug abuse.
McKinney et. al. examined the possible relationship between the M AO -A T1460C
(FcoRV) & G941T (FnuHY) polymorphisms and tobacco addiction (McKinney,
Walton et al. 2000). Although these polymorphisms are conservative substitutions
and unlikely to be functional, they are thought to be in linkage disequilibrium with
the uVNTR in the promoter and are known to be in linkage disequilibrium with one
another. The T allele o f both polymorphisms has been found to be associated with
reduced enzyme activity in cultured skin fibroblasts (Hotamisligil and Breakefield
1991), and is therefore expected to be comparable to our short allele. M cKinney and
colleagues found a statistically significant increase in number o f cigarettes smoked
for subjects with the 1460T allele compared to those hom ozygous for the 1460C
allele, even after adjustment for gender and alcohol consumption (p = 0.046).
Furthermore, heavy smokers (20+ cigarettes/day) were less likely to have the 1460C
allele (RR = 0.31, Cl: 0.13, 0.74). These findings are consistent with our results and
suggest that M AO -A polymorphisms may be important influences on smoking
behavior and in particular, that low M AO-A activity may be related to heavy
smoking.
Hsu et. al. found a statistically significant association between both the AcoRV
polymorphism and an unspecified dinucleotide repeat polymorphism with alcohol
dependence among Chinese males in Taiwan (Hsu, Loh et al. 1996). Although the
functional significance o f these two polymorphisms remains unclear, these results
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
also suggest the importance o f M AO-A in predisposing to addictive disorders. This
study is also important in that it shows that polymorphisms in the gene for M AO-A
are important in Chinese subjects o f Han ancestry, a population that is genetically
very similar to the Hokkien dialect group in Singapore.
Gade and colleagues investigated the possible role o f a complex polymorphism in the
first intron o f the M AO -A gene (Gade, Muhleman et al. 1998). This polymorphisms
consists o f a dinucleotide (GT) repeat plus an imperfectly duplicated 23-bp VNTR
motif. They found that their longest allele (>335 bp) was statistically significantly
more common among subjects undergoing treatment for drug or alcohol abuse (6.3%
vs. 21.7%, p = 0.00006). Specifically, subjects with a revised Michigan Alcoholism
Severity Test score indicating drug abuse were more likely to carry the >335 bp allele
than control subjects (p = 0.001). The effect o f this allele was stronger among drug
abusers than alcohol abusers. In order to indirectly assess the functional relevance o f
this polymorphism, Gade and colleagues compared alleles with < 320 bp to the
FnuAYil 2 allele, which was known to possess higher activity than the alternate 1
allele. They found that alleles with 320 bp or less were statistically significantly
associated with the Fnu4Hl 2 allele and concluded that the >335 bp allele group was
associated with the lowest M AO-A activity. Thus, the results o f Gade et. al. are
entirely consistent with our finding that the uVNTR short allele (low activity) is
somewhat overrepresented among subjects with earlier age at smoking initiation, 40+
years o f smoking regularly, and 43+ cigarettes smoked/day, although only age at
smoking initiation reached statistical significance in our sample.
1 7 7
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
More recently, the long allele, defined as 4 repeats, was found to be protective against
being a current smoker among females but not males in a Japanese population (OR =
0.49, Cl: 0.26, 0.93) (Ito, Hamajima et al. 2003). Additionally, the haplotype
consisting o f three repeats o f M AO-A and the G allele o f M AO-B (A644G) was
found to be associated with a later age o f smoking initiation among males compared
to male subjects with one o f the other haplotypes (21.7 vs. 19.5, respectively, p <
0.05). The 3G haplotype was shown to be associated with lower M AO -A activity
(Balciuniene, Em ilsson et al. 2002) and is thus comparable to our short genotype, for
which an earlier age o f smoking initiation was found. The lack o f consistency
between our results and those o f Ito et. al. may be less bothersome when one
considers that there were only six subjects in the 3G haplotype group. Thus, chance
alone could account for the latter’s findings.
There are a number o f constitutional strengths in the current investigation. First, the
availability o f data at two points in time (baseline and follow-up) allow us to refine
the smoking phenotype. B y using subjects who self-identified as smokers on two
separate questionnaires w e are able to say with more confidence that our current
smokers truly represent the population o f long-term smokers in the base population.
Additionally, all o f the current smokers in our sample were daily smokers who
smoked at least 7-12 cigarettes per day, thereby producing a somewhat less
heterogeneous group o f smokers than is commonly employed in similar studies.
Furthermore, the nonsmokers in our study were defined so that all subjects had at the
potential to be exposed to cigarettes. That is, any subject exposed to less than 100
cigarettes in their lifetime would still be considered a never smoker. Prior
178
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
opportunity to take up smoking is an important consideration since any genetic
predisposition toward addiction that is present would then have the chance to be
expressed.
Another strength o f our study is the large number o f covariates available for
consideration as possible confounders and/or effect modifiers. The Singapore
Chinese Health Study has an abundance o f data on topics ranging from diet and
physical activity to family history o f illness and passive smoking exposure. As a
result, we were able to m odel the association between M AO -A genotype and various
smoking variables, including aspects o f smoking behavior, more completely than is
possible in many gene association studies.
Additionally, by using a population that is ethnically very homogeneous, w e were
able to essentially bypass the problem o f population stratification. Cohort subjects
were drawn from the two major dialect groups o f Chinese in Singapore, the Hokkiens
and the Cantonese, who originated from two contiguous prefectures in southern
China. Until very recently (and thus, irrelevant to our cohort), marriage across
dialect groups in Singapore was virtually nonexistant. Although the seriousness o f
this problem has been much debated (Thomas and Witte 2002; Wacholder, Rothman
et al. 2002), bias due to population stratification has been observed in gene
association studies in the past (Knowler, Williams et al. 1988) and avoiding ethnic
admixture is one way o f assuring that such bias cannot explain an observed
association with a candidate gene.
1 7 9
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Last, the legislative climate in Singapore provides a unique opportunity to study
tobacco addiction. Beginning in 1970, the Singapore government enacted a series o f
progressively more restrictive laws limiting where a person could smoke. These
laws, in combination with a conformist culture, created a climate where smoking has
become very difficult and smokers are viewed with suspicion. Therefore, subjects
who continue to smoke in Singapore are most likely very heavily addicted. Clearly,
tobacco smoking, like other drugs o f abuse, cannot be eradicated simply by
legislative efforts.
A number o f limitations also apply to this study. First, the use o f only current/current
smokers and never/never smokers essentially restricts our investigation to smoking
initiation. Since w e have not included ex-smokers in the study population, w e cannot
examine the role o f M AO -A genotypes in smoking cessation or continuation. Our
rationale for restricting the study population in this way was to contrast the most
extreme smoking groups as part o f a preliminary analysis using the Singapore cohort.
By doing so, w e expected to assess the usefulness o f the Singapore data for
examining smoking-related outcomes (it was originally designed to investigate cancer
outcomes) and to provide the greatest opportunity to find a difference between these
two extreme smoking groups. Based on our findings, w e feel that expansion o f the
study to include additional smoking categories is warranted and w ill be conducted in
the near future.
A second limitation arises due to the lack o f data on concurrent mental illnesses. As a
result, no adjustments have been made for history o f schizophrenia, depression,
180
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
anxiety, or any o f the other conditions that may be related to both'smoking status and
M AO-A genotype. However, the rate o f reported mental illness in elderly
Singaporeans is relatively low (Kua 1998), suggesting that even if the data were
available for this population, the impact o f adjusting for mental illness would be
small.
A potential limitation arises as a result o f revisions made to the questionnaire after the
baseline interview affecting how smoking exposure was assessed. W hile the baseline
questionnaire defined current smokers as people who has smoked at least one
cigarette a day for one year or longer (thus, excluding non-daily smokers), the follow-
up questionnaire defined smokers so that non-daily smokers could also qualify as
current smokers. However, we avoided any potential problems with these differing
definitions by restricting our current smoking population to subjects who reported
smoking at least 7-12 cigarettes daily at both timepoints. Additionally,
the baseline questionnaire used pre-defined categories to classify age at smoking
regularly, number o f years o f smoking regularly and the number o f cigarettes smoked
while the follow-up questionnaire allowed subjects to complete the questionnaire
with a number o f their choosing. Since w e were unsure o f how this would effect the
results, we chose to analyze the baseline and follow-up data separately in the
preliminary analysis. We used the follow-up data for our adjusted m odels o f
smoking behaviors (age at first cigarette, age at regular smoking, and number o f
cigarettes smoked) since this data was continuous and provided the most information.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Despite these limitations, w e believe that the current study offers some possible
insights into tobacco addiction that are useful and advance the current knowledge
regarding the role o f M AO -A in smoking initiation. The finding o f an earlier age at
smoking initiation in subjects carrying a short uVNTR allele is consistent with our
understanding o f the biological mechanism o f tobacco addiction (Figure 5.1).
Tobacco addiction is a serious public health challenge. According to the Morbidity
and Mortality W eekly Report, nearly 46.5 m illion adults are current smokers (2001).
A better understanding o f the underlying susceptibility to tobacco addiction could aid
in the development o f more effective prevention and cessation programs. The ability
to identify populations at increased risk for starting smoking would allow public
health officials to target specific groups and could increase effectiveness o f
prevention programs as w ell as reduce the costs associated with them. More personal
and tailored prevention programs could be designed to specifically address some o f
the biological mechanisms underlying smoking initiation and potentially prevent
many o f the subsequent diseases for which smoking is an important risk factor.
Lastly, improving our understanding o f the biological basis o f tobacco addiction has
implications for our understanding o f addiction in general since many o f the
processes are believed to be the same.
5.5 CONCLUSIONS AND FUTURE DIRECTIONS
Based on the results presented here, it appears that M AO-A genotype may primarily
affect smoking initiation and/or smoking intensity rather than determine overall
182
permission of the copyright owner. Further reproduction prohibited without permission.
smoking status. The finding o f an earlier age at first cigarette among current/current
smokers who carry the short uVNTR allele and lack o f an overall association between
M AO-A genotype and smoking status supports this assertion. Further support is
provided by the suggestion that current/current smokers with the short allele might
also be more likely to be heavy smokers (43+ cigs/day) and might be more likely to
smoke for 40+ years. In this respect, our results are consistent with observations in
the literature indicating that M AO-A genotype may be involved in determining the
number o f cigarettes smoked (McKinney, Walton et al. 2000) but not in predicting
smoking status, at least among Japanese males (Ito, Hamajima et al. 2003). Because
MAO appears to be involved in influencing some aspects o f an individual’s behavior
(e.g., whether or not they tend to seek out novel experiences), it seems reasonable to
infer that the role o f the monoamine degradation pathway in the development o f
tobacco addiction might be restricted to those aspects o f behavior that result in
smoking initiation or determine smoking intensity. However, since w e did not
include ex-smokers in the current study population, w e were unable to assess the
impact o f M AO -A genotype on smoking continuation or ability to quit. Future
studies w ill expand to include ex-smokers so that w e can examine the potential role
o f M AO -A in other aspects o f tobacco addiction.
In order to more fully examine the possible role o f M AO in tobacco addiction, it will
be necessary to also examine polymorphisms in M AO-B since the same biological
rationale applies for both forms o f MAO. Several polymorphisms have been
identified in M AO-B (Grimsby, Chen et al. 1992; Konradi, Ozelius et al. 1992) and
research has suggested that M AO-B might also be important in tobacco addiction
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
(Checkoway, Franklin et al. 1998). Future studies w ill also include polymorphisms
in the M AO-B gene.
In addition to genes in the monoamine degradation pathway, genes involved in the
synthesis and transport o f neurotranmitters, nicotine receptors, and nicotine
metabolism may also be important in specific stages o f tobacco addiction (Tyndale
2003). From the available data, it appears that neurotransmitter systems affect the
propensity toward tobacco addiction through their effects on personality traits (e.g.,
neuroticism, novelty-seeking) rather than through a direct effect o f nicotine itself.
Thus, in order to obtain a more complete picture o f the genetic predisposition toward
tobacco addiction, including those aspects that cause an individual to try smoking,
continue to smoke, or be unable to quit, w e w ill need to examine polymorphisms in
genes in the dopaminergic, serotonergic, and GABAergic systems as w ell as genes
involved in the metabolism o f nicotine.
Genes in the D A system have received the vast majority o f attention due to the
importance o f D A in rewarding and reinforcing nicotine use (Merlo Pich, Chiamulera
et al. 1999). Although the results have not been entirely consistent, it appears that the
D l, D2, D4, and D5 dopamine receptors play some role in tobacco addiction, with
some variants o f D 2 being implicated in smoking initiation (Spitz, Shi et al. 1998;
Anokhin, Todorov et al. 1999) and smoking intensity (Wu, Hudmon et al. 2000).
Genetic variants o f the dopamine transporter gene (SLC6A3) may also play a role in
tobacco addiction, having been implicated in both smoking initiation and cessation
(Lerman, Caporaso et al. 1999; Sabol, Nelson et al. 1999).
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Genes in the serotonergic system are also good candidates for involvement in tobacco
addiction due to the role o f serotonin in mood regulation and impulse control
(Veenstra-VanderWeele, Anderson et al. 2000). To date, none o f the known variants
in serotonin receptors have been implicated in tobacco addiction, but polymorphisms
in tryptophan hydroxylase, a rate-limiting enzyme in serotonin synthesis, have been
found to be associated with age at smoking initiation (Lerman, Caporaso et al. 2001;
Sullivan, Jiang et al. 2001).
Inhibition o f the GABAergic system has been shown to alter the amount o f dopamine
released at the nucleus acumbens, also known as the “reward site” o f the brain.
Although no studies to date have examined the role o f genes from the GABAergic
system in tobacco addiction, some studies have found an association with various
GAB A receptor gene variants and alcohol dependence (Parsian and Cloninger 1997;
Noble, Zhang et al. 1998; Parsian and Zhang 1999; Sander, Sam ochowiec et al. 1999;
Schuckit, Mazzanti et al. 1999).
Lastly, genes encoding nicotinic receptors and those involved in nicotine metabolism
are most probably important determinants o f brain nicotine levels. Thus,
polymorphisms in these genes are likely to explain at least som e o f the variation in
smoking behavior. In fact, polymorphisms in CYP2A6, the enzyme responsible for
approximately 90% o f the inactivation o f nicotine to cotinine, have been found to
affect the risk o f tobacco addiction, the age at smoking initiation, smoking duration
and intensity, and the ability to quit (Gu, Hinks et al. 2000; Rao, Hoffmann et al.
2000; Tyndale and Sellers 2002).
1 8 5
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
In summary, tobacco addiction is a complex state requiring input from numerous
physiological systems in order to manifest. Furthermore, it is likely that there are
multiple pathways to becoming nicotine-addicted. Developing even a quasi-
comprehensive m odel for tobacco addiction w ill require a much greater
understanding o f the neurological basis o f addiction as w ell as a more complete
picture o f the underlying genetic susceptibilities. To that end, future studies should
examine multiple loci simultaneously, including genes from not only the various
neurotransmitter systems but also those involved in nicotine metabolism.
In conclusion, I plan to expand the present study in several directions. First,
additional categories o f smokers, nonsmokers, and ex-smokers need to be examined
in order to assess the possible role o f MAO in all three aspects o f tobacco addiction -
initiation, continuation, and cessation. Specifically, in addition to the never/never
and current/current smokers, I plan to include groups that can give further insight into
smoking initiation (e.g., never/current smokers) as w ell as those that can provide
information regarding cessation (e.g., current/ex smokers, ex/ex smokers, and
ex/current smokers). Second, I plan to expand the list o f candidate genes to initially
include M AO-B and, eventually, to include other genes in the dopaminergic,
serotinergic, and GABAergic systems.
It is important to note that because the Singapore Chinese Health Study was not
originally designed to study tobacco addiction, it is not the ideal population to
investigate smoking continuation since this would require a more rapid reassessment
o f smoking status after a subject has had their first cigarette. In the Singapore cohort,
186
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
although smoking status was assessed twice it is not possible to determine which
subjects went on to becom e regular smokers after having their first cigarette.
Nevertheless, the Singapore Chinese Health Study, because o f the reasons outlined in
Section 5.4, does have many strengths and appears to be a good population in which
to study nicotine addiction.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
CHAPTER 6: SUMMARY
As I near the end o f this long yet exciting journey, I find m yself reflecting back to the
goals I had when I first arrived at the University o f Southern California in the fall o f
1998. I came to USC with the hope that I might grow as a scientist, gaining not only
in classroom knowledge, but also in practical experience. It is the latter which I feel
is the most valuable, but also the most difficult to obtain in an academic program.
Nevertheless, I feel that I have received precisely the practical experience that I had
hoped to accrue during m y years at USC.
While choosing a dissertation project outside the area o f interest for most faculty
members in the Department o f Preventive M edicine almost certainly increased the
time I have spent as a graduate student, it also provided me the opportunity to gain
real-world experience as an epidemiologist. I began m y project not knowing
anything about preeclampsia, tobacco addiction, or monoamine oxidase A but have
finished feeling like I have gained a great deal o f insight into these topics. But more
importantly, I now know much more about how to approach a research question,
including how to take an abstract idea and turn it into a funded research project.
In writing the literature review “Molecular Epidemiology o f Preeclampsia,” (Chapter
3) I learned several important lessons. First, I learned that no one single literature
database can be relied upon to find a majority o f the important publications in a
given field. Rather, a combination o f library databases and reference sections from
important review papers is a more thorough method o f locating papers on a specific
188
permission of the copyright owner. Further reproduction prohibited without permission.
topic. Second, I discovered that a comprehensive literature review is an essential
starting point for educating onself, determining where the gaps in knowledge are, and
generating ideas on how these gaps might be addressed in future studies. Third, and
most importantly, I learned how to read and critically evaluate a large body o f
literature, summarize the current state o f knowledge in a succinct manner, and
formulate an overall opinion based on the available data.
As an academic researcher, there can be no more important skill than the ability to
write a fundable grant proposal. The process o f writing, submitting, resubmitting,
and re-resubmitting can be a daunting task to even an experienced principal
investigator, let alone an inexperienced graduate student. Having gone through this
process once has been a great boon to my skill set. Again, I have learned many
invaluable lessons from this experience. For example, I understand that the process
is not always as fair or straight-forward as one might like. Reviewers are not blank
slates who read a proposal with a completely open mind nor are they unmoved by
current research trends or “hot topics.” Writing a successful grant application
requires that the investigator be familiar with not only the relevant scientific
material, but also the current political climate as w ell as the funding agency’s
specific biases and concerns.
The data analysis presented in Chapter 5, while limited in scope, served two essential
purposes. First, the consistency o f the findings with those from other populations
suggest that the Singapore cohort is a reasonable population to study when
examining the genetic predisposition to smoking and smoking-related behaviors.
189
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Although the Singapore cohort was not originally designed to investigate tobacco
addiction, the questionnaire data on smoking and smoking-related behaviors is
relatively complete and reliable. Additionally, the Singapore cohort has the added
benefit o f having been subjected to social and legislative pressures intended to
discourage smoking and thus, is an especially good population in which to address
the addictive power o f nicotine. Second, the finding o f a younger age at smoking
initiation among men who carried the short uVNTR allele provides justification for
further studies to examine the monoamine degradation pathway and as such, can
form the basis o f a future grant application. Future avenues o f research include a
planned expansion o f the tobacco addiction study to include additional groups o f
smokers and ex-smokers as w ell as a more complete genetic picture to include
additional M AO -A polymorphisms, M AO-B polymorphisms, and possibly other
candidate genes including those in the dopamine receptor family.
Rather than an ending, I view the writing o f this dissertation as a starting point from
which many roads emanate. Along one road lies the possible answer to the
preeclampsia puzzle. Our ongoing pilot study to examine the effects o f specific
genes involved in placentation may help to explain part o f that puzzle. Moreover,
the data generated as part o f this pilot study to assess population stratification among
Latinas has potentially far-reaching applications. Furthermore, our ability to recruit
subjects from the LAC + USC Women and Children’s Hospital (WCH) opens the
doors to a w hole new area o f research at USC. I plan to remain at USC post
graduation to complete the preeclampsia pilot study and, hopefully, to continue
working with investigators at the WCH on future projects o f mutual interest.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
REFERENCES
(2000). "Working Group Report on High Blood Pressure in Pregnancy." American
Journal Obstetrics Gynecology 183(1).
(2001). Cigarette smoking among adults — United States, 1999. Morbidity and
Mortality W eekly Report: 869-873.
Alders, M ., M. Hodges, et al. (1997). "The human Achaete-Scute hom ologue 2
(ASCL2,HASH2) maps to chromosome 1 l p l 5.5, close to IGF2 and is
expressed in extravillus trophoblasts." Hum M ol Genet 6(6): 859-67.
Altshuler, D., L. Kruglyak, et al. (1998). "Genetic polymorphisms and disease." N
E n g U M ed 338(22): 1626.
Anokhin, A. P., A. A. Todorov, et al. (1999). "Brain event-related potentials,
dopamine D2 receptor gene polymorphism, and smoking." Genet Epidemiol
17 Suppl 1: S37-42.
Anson-Cartwright, L., Dawson, Kerri, Holmyard, Doug, Fisher, Susan J, Lazzarini,
Robert A., Cross, James C (2000). "The glial m issing-1 protein is essential
for branching morphogenesis in the chorioallantoic placenta." Nature
Genetics 25: 314-315.
Arbogast BW, L., S.C., Meerrick, R.D., Olive, K.E., Taylor, R.N. (1996). "Plasma
factaors that determine endothelial cell lipid toxicity in vitro correctly
identify wom en with preeclampsia in early and late pregnancy."
Hypertension in Pregnancy 15(3): 263-279.
Amgrimsson, R., S. Bjom sson, et al. (1990). "Genetic and familial predisposition to
eclampsia and pre-eclampsia in a defined population." British Journal o f
Obstetrics & Gynaecology 97(9): 762-9.
Amgrimsson, R., J. M. Connor, et al. (1994). "Is genetic susceptibility for pre
eclampsia and eclampsia associated with implantation failure and fetal
demise? [letter]." Lancet 343(8913): 1643-4.
Amgrimsson, R., R. T. Geirsson, et al. (1994). "Renin gene restriction fragment
length polymorphisms do not show linkage with preeclampsia and
eclampsia." Acta Obstetricia et Gvnecologica Scandinavica 73(1): 10-3.
Amgrimsson, R., C. Hayward, et al. (1997). "Evidence for a familial pregnancy-
induced hypertension locus in the eNOS-gene region." American Journal o f
Human Genetics 61(2): 354-62.
191
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Amgrimsson, R., S. Purandare, et al. (1993). "Angiotensinogen: a candidate gene
involved in preeclampsia? [letter; comment]." Nature Genetics 4(2): 114-5.
Amgrimsson, R., S. Sigurard ttir, et al. (1999). "A genom e-wide scan reveals a
maternal susceptibility locus for pre-eclampsia on chromosome 2pl3."
Human Molecular Genetics 8(9): 1799-805.
Bach, A. W., N. C. Lan, et al. (1988). "cDNA cloning o f human liver monoamine
oxidase A and B: molecular basis o f differences in enzymatic properties."
Proc Natl Acad Sci U S A 85(13): 4934-8.
Balciuniene, J., L. Emilsson, et al. (2002). "Investigation o f the functional effect o f
monoamine oxidase polymorphisms in human brain." Hum Genet 110(1): 1-
7.
Ballantine, D. M ., S. A. Klemm, et al. (1994). "Renin gene polymorphism associated
with aldosterone responsiveness to the renin-angiotensin system in patients
with aldosterone-producing adenomas." Clin Exp Pharmacol Physiol 21(3):
215-8.
Bane, A. L. and J. E. Gillan (2003). "Massive perivillous fibrinoid causing recurrent
placental failure." Biog 110(3): 292-5.
Barden, A. E., C. E. Herbison, et al. (2001). "Association between the endothelin-1
gene L ysl98A sn polymorphism blood pressure and plasma endothelin-1
levels in normal and pre-eclamptic pregnancy." J Hvpertens 19(10): 1775-82.
Bashford, M. T., L. A. Hefler, et al. (2001). "Angiotensinogen and endothelial nitric
oxide synthase gene polymorphisms among Hispanic patients with
preeclampsia." Am J Obstet Gvnecol 184(7): 1345-50; discussion 1350-1.
Basyuk, E., J. C. Cross, et al. (1999). "Murine G cm l gene is expressed in a subset o f
placental trophoblast cells." Dev Dvn 214(4): 303-11.
Belgore, F. M ., G. Y. Lip, et al. (2000). "Vascular endothelial growth factor and its
receptor, Fit-1, in smokers and non-smokers." Br J Biom ed Sci 57(3): 207-13.
Berlin, I., S. Said, et al. (1995). "A reversible monoamine oxidase A inhibitor
(m oclobemide) facilitates smoking cessation and abstinence in heavy,
dependent smokers." Clin Pharmacol Ther 58(4): 444-52.
Bermingham, J., D. Jenkins, et al. (2000). "Genetic analysis o f insulin-like growth
factor II and HLA-G in pre-eclampsia." Biochemical Society Transactions
28(2): 215-9.
Bertina, R. M., B. P. Koeleman, et al. (1994). "Mutation in blood coagulation factor
V associated with resistance to activated protein C." Nature 369(6475): 64-7.
1 9 2
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Bird, I. M., J. Zheng, et al. (1997). "Pregnancy induces an increase in angiotensin II
type-1 receptor expression in uterine but not systemic artery endothelium."
Endocrinology 138(1): 490-8.
Black, G. C., Z. Y. Chen, et al. (1991). "Dinucleotide repeat polymorphism at the
M AO A locus." Nucleic Acids Res 19(3): 689.
Bouma, G., J. B. Crusius, et al. (1996). "Secretion o f tumour necrosis factor alpha
and lymphotoxin alpha in relation to polymorphisms in the TNF genes and
HLA-DR alleles. Relevance for inflammatory bowel disease." Scand J
Immunol 43(4): 456-63.
Bouma, G., B. Xia, et al. (1996). "Distribution o f four polymorphisms in the tumour
necrosis factor (TNF) genes in patients with inflammatory bow el disease
(IBD)." Clin Exp Immunol 103(3): 391-6.
Boyd, P. A., R. H. Lindenbaum, et al. (1987). "Pre-eclampsia and trisomy 13: a
possible association." Lancet 2(8556): 425-7.
Breakefield, X. O., C. M. Castiglione, et al. (1976). "Monoamine oxidase activity
decreased in cells lacking hypoxanthine phosphoribosyltransferase activity."
Science 192142431: 1018-20.
Breakefield, X. O., E. L. Giller, Jr., et al. (1980). "Monoamine oxidase type A in
fibroblasts from patients with bipolar depressive illness." Psychiatry Res
2(3): 307-14.
Brennecke, S. P., N. M. Gude, et al. (1997). "Reduction o f placental nitric oxide
synthase activity in pre-eclampsia." Clin Sci (Colch) 93(1): 51-5.
Breslow, N. E. and N. E. Day (1980). "Statistical methods in cancer research.
Volume I - The analysis o f case- control studies." IARC Sci Publ 32: 5-338.
Brunner, H. G., M. Nelen, et al. (1993). "Abnormal behavior associated with a point
mutation in the structural gene for monoamine oxidase A." Science 262: 578-
80.
Brunner, H. G., M. R. Nelen, et al. (1993). "X-linked borderline mental retardation
with prominent behavioral disturbance: phenotype, genetic localization, and
evidence for disturbed monoamine metabolism." Am J Hum Genet 52(6):
1032-9.
Camarena, B., C. Cruz, et al. (1998). "A higher frequency o f a low activity-related
allele o f the M AO -A gene in females with obsessive-com pulsive disorder."
Psvchiatr Genet 8(4): 255-7.
193
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Campbell, I. C., D. S. Robinson, et al. (1979). "The effects o f chronic regimens o f
clorgyline and pargyline on monoamine metabolism in the rat brain." J
Neurochem 32(1): 49-55.
Caniggia, I., H. M ostachfi, et al. (2000). "Hypoxia-inducible factor-1 mediates the
biological effects o f oxygen on human trophoblast differentiation through
TGFbeta(3)." J Clin Invest 105(5): 577-87.
Cargill, M ., D. Altshuler, et al. (1999). "Characterization o f single-nucleotide
polymorphisms in coding regions o f human genes." Nat Genet 22(3): 231-8.
Carmeliet, R, M oons, L., Luttun, A., Vincenti, V., Com pem olle, V., D e M ol, M.,
Wu, Y., Bono, R, Devey, L., Beck, H., Scholz, D ., Acker, T., et al. (2001).
"Synergism between vascular enothelial growth factor and placental growth
factor contributes to angiogenesis and plasma extravsation in pathological
conditions." Nature M edicine 7: 575-583.
Carmelli, D., G. E. Swan, et al. (1992). "Genetic influence on sm ok ing-a study o f
male twins." N Engl J Med 327(12): 829-33.
Carrasco, G., M. A. Cruz, et al. (2000). "Transport and metabolism o f serotonin in
the human placenta from normal and severely pre-eclamptic pregnancies."
Gynecologic & Obstetric Investigation 4913): 150-5.
Carrasco, G., M. A. Cruz, et al. (1998). "Plasma and platelet concentration and
platelet uptake o f serotonin in normal and pre-eclamptic pregnancies." Life
S d 62(15): 1323-32.
Carter, A. M., A. J. Catto, et al. (1998). "Platelet GP Ilia P1A and GP lb variable
number tandem repeat polymorphisms and markers o f platelet activation in
acute stroke." Arterioscler Thromb Vase Biol 18(7): 1124-31.
Cases, O., I. Seif, et al. (1995). "Aggressive behavior and altered amounts o f brain
serotonin and norepinephrine in m ice lacking MAO A." Science 268(5218):
1763-6.
Caspi, A., J. McClay, et al. (2002). "Role o f genotype in the cycle o f violence in
maltreated children." Science 297(5582): 851-4.
Checkoway, H., G. M. Franklin, et al. (1998). "A genetic polymorphism o f MAO-B
m odifies the association o f cigarette smoking and Parkinson's disease."
Neurology 5015): 1458-61.
Chen, G., R. W ilson, et al. (1994). "Normal superoxide dismutase (SOD) gene in
pregnancy-induced hypertension: is the decreased SOD activity a secondary
phenomenon?" Free Radic Res 21(2): 59-66.
194
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Chen, G., R. W ilson, et al. (1996). "Tumour necrosis factor-alpha (TNF-alpha) gene
polymorphism and expression in pre-eclampsia." Clin Exp Immunol 104(1):
154-9.
Chen, K. (2004). "Organization o f MAO A and M AOB Promoters and Regulation o f
Gene Expression." NeuroToxicology 25: 31-36.
Chen, Z. Y., G. S. Hotamisligil, et al. (1991). "Structure o f the human gene for
monoamine oxidase type A." Nucleic Acids Res 19(16): 4537-41.
Cheng, A. C., C. P. Pang, et al. (2000). "The association between cigarette smoking
and ocular diseases." Hong Kong Med J 6(2): 195-202.
Chesley, L. C. (1980). "Hypertension in pregnancy: Definitions, familial factor, and
remote prognosis." Kidney International 18: 234-240.
Chesley, L. C. (1984). "History and epidemiology o f preeclampsia-eclampsia." Clin
Obstet Gvnecol 27(4): 801-20.
Chesley, L. C. (1993). "The genetics o f preeclampsia." Hypertension in Pregnancy
12: vii-x.
Chesley, L. C., Cooper, D.W , (1986). "Genetics o f hypertension in pregnancy:
possible single gene control o f pre-eclampsia and eclampsia in the
descendants o f eclamptic women." British Journal o f Obstetrics &
G ynaecology 93: 898-908.
Chouchane, L., S. B. Ahmed, et al. (1997). "Polymorphism in the tumor necrosis
factor-alpha promotor region and in the heat shock protein 70 genes
associated with malignant tumors." Cancer 80(8): 1489-96.
Chua, S., T. Wilkins, et al. (1991). "Trophoblast deportation in pre-eclamptic
pregnancy." Br J Obstet Gynaecol 98(10): 973-9.
Cincotta, R. B. and S. P. Brennecke (1998). "Family history o f pre-eclampsia as a
predictor for pre-eclampsia in primigravidas." International Journal o f
Gynaecology & Obstetrics 60(1): 23-7.
Clark, D. E., S. K. Smith, et al. (1998). "A vascular endothelial growth factor
antagonist is produced by the human placenta and released into the maternal
circulation." B iol Reprod 59(6): 1540-8.
Cloninger, C. R. (1986). "A unified biosocial theory o f personality and its role in the
development o f anxiety states." Psvchiatr D ev 4(3): 167-226.
Cohen, I. L., X. Liu, et al. (2003). "Association o f autism severity with a monoamine
oxidase A functional polymorphism." Clin Genet 64(3): 190-7.
1 9 5
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Collins, F. A ., D. L. Murphy, et al. (1992). "Clinical, biochemical, and
neuropsychiatric evaluation o f a patient with a contiguous gene syndrome due
to a microdeletion X p ll.3 including the Norrie disease locus and monoamine
oxidase (M AOA and M AOB) genes." Am J M ed Genet 42(1): 127-34.
Comings, D. E. (1998). "Polygenic inheritance and micro/minisatellites." M ol
Psychiatry 3(1): 21-31.
Comings, D. E. and K. Blum (2000). "Reward deficiency syndrome: genetic aspects
o f behavioral disorders." Prog Brain Res 126: 325-41.
Comings, D. E., L. Ferry, et al. (1996). "The dopamine D2 receptor (DRD2) gene: a
genetic risk factor in smoking." Pharmacogenetics 6(1): 73-9.
Conklin, B. S., W. Zhao, et al. (2002). "Nicotine and cotinine up-regulate vascular
endothelial growth factor expression in endothelial cells." Am J Pathol
160(2): 413-8.
Cooper, D. W., Brennecke S.P., Wilton A.N. (1993). "Genetics o f Pre-eclampsia."
Hypertension in Pregnancy 12(1): 1-23.
Cooper, D. W., J. A. Hill, et al. (1988). "Genetic control o f susceptibility to
eclampsia and miscarriage." Br J Obstet Gvnaecol 95(7): 644-53.
Cooper, D. W. and W. A. Liston (1979). "Genetic control o f severe pre-eclampsia."
Journal o f M edical Genetics 16(6): 409-16.
Corrigall, W. A., K. B. Franklin, et al. (1992). "The m esolim bic dopaminergic
system is implicated in the reinforcing effects o f nicotine."
Psvchopharmacology 107(2-3): 285-9.
Critchley, J. and S. Capewell (2003). "Smoking cessation for the secondary
prevention o f coronary heart disease." Cochrane Database Svst R ev(4):
CD003041.
Cross, J. C., D. Baczyk, et al. (2003). "Genes, development and evolution o f the
placenta." Placenta 24(2-3): 123-30.
Dahlback, B. (1994). "Inherited resistance to activated protein C, a major cause o f
venous thrombosis, is due to a mutation in the factor V gene." Haemostasis
24(2): 139-51.
D e Groot, C. J. M ., Bloemenkamp, K.W.M., Duvekot, E.J., Helmerhorst, F.M.,
Bertina, R.M ., Van Der Meer, F., De Ronde, H., Oei, S.G., Kanhai, H.H.,
Rosendaal, F.R. (1999). "Preeclampsia and genetic risk factors for
thrombosis: A case-control study." American Journal o f Obstetrics &
G ynecology 181: 975-980.
196
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
de Luca Brunori, I., L. Battini, et al. (2000). "Increased HLA-DR hom ozygosity
associated with pre-eclampsia." Human Reproduction 15(8): 1807-12.
Deckert, J., M. Catalano, et al. (1999). "Excess o f high activity monoamine oxidase
A gene promoter alleles in female patients with panic disorder." Hum Mol
Genet 8(4): 621-4.
Dekker, G. A. (1999). "Risk factors for preeclampsia." Clinical Obstetrics &
G ynecology 42(3): 422-35.
Dekker, G. A., J. I. de Vries, et al. (1995). "Underlying disorders associated with
severe early-onset preeclampsia." Am J Obstet Gynecol 173(4): 1042-8.
Dekker, G. A. and B. M. Sibai (1998). "Etiology and pathogenesis o f preeclampsia:
current concepts [see comments]." American Journal o f Obstetrics &
G ynecology 179(5): 1359-75.
Dekker, G. A. and B. M. Sibai (1999). "The immunology o f preeclampsia." Semin
Perinatol 23(1): 24-33.
den Boer, M. E., L. Ijlst, et al. (2000). "Heterozygosity for the common LCHAD
mutation (1528g>C) is not a major cause o f HELLP syndrome and the
prevalence o f the mutation in the Dutch population is low." Pediatr Res
48(2): 151-4.
Denney, R. M ., R. R. Fritz, et al. (1982). "Human liver M AO -A and M AO-B
separated by immunoaffinity chromatography with M AO -B-specific
m onoclonal antibody." Science 215(4538): 1400-3.
Denney, R. M ., H. Koch, et al. (1999). "Association between monoamine oxidase A
activity in human male skin fibroblasts and genotype o f the M AOA
promoter-associated variable number tandem repeat." Hum Genet 105(6):
542-51.
Desautels, A., G. Turecki, et al. (2002). "Evidence for a genetic association between
monoamine oxidase A and restless legs syndrome." Neurology 59(2): 215-9.
Devlin, B. and K. Roeder (1999). "Genomic control for association studies."
Biometrics 55(4): 997-1004.
Dizon-Townson, D. S., H. Major, et al. (1998). "A promoter mutation in the tumor
necrosis factor alpha gene is not associated with preeclampsia." Journal o f
Reproductive Immunology 38(11: 55-61.
Dizon-Townson, D. S., L. M eline, et al. (1997). "Fetal carriers o f the factor V Leiden
mutation are prone to miscarriage and placental infarction." Am J Obstet
Gvnecol 177(2): 402-5.
1 9 7
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Dizon-Townson, D. S., L. M. Nelson, et al. (1996). "The factor V Leiden mutation
may predispose wom en to severe preeclampsia." American Journal o f
Obstetrics & Gynecology 175(4 Pt 1): 902-5.
Eaves, L. J. and H. J. Eysenck (1980). N ew approaches to the analysis o f twin data
and their application to smoking behavior. The Causes and Effects o f
Sm oking. H. J. Eysenck. London, Maurice Temple Smith: 140-314.
Edwards, K. L., M. A. Austin, et al. (1995). "Evidence for genetic influences on
smoking in adult women twins." Clin Genet 47(5): 236-44.
Eensoo, D., M. Paaver, et al. (2004). "Low platelet MAO activity associated with
high dysfunctional impulsivity and antisocial behavior: evidence from drunk
drivers." Psvchopharmacology (Berl) 172(3): 356-8.
Eley, T. C., E. Tahir, et al. (2003). "Association analysis o f M AO A and COMT with
neuroticism assessed by peers." Am J Med Genet 120B(1): 90-6.
Eremina, V., M. Sood, et al. (2003). "Glomerular-specific alterations o f VEGF-A
expression lead to distinct congenital and acquired renal diseases." J Clin
Invest 111(5): 707-16.
Erikssen, J. and S. C. Enger (1978). "The effect o f smoking on selected coronary
heart disease risk factors in middle-aged men." Acta M ed Scand 203(1-2):
27-30.
Esplin, M. S., M. B. Fausett, et al. (2001). "Paternal and maternal components o f the
predisposition to preeclampsia." N ew England Journal o f M edicine 344(12):
867-72.
Etter, J. F., A. Pelissolo, et al. (2003). "Associations between smoking and heritable
temperament traits." Nicotine Tob Res 5(3): 401-9.
Ewens, W. J. and R. S. Spielman (1995). "The transmission/disequilibrium test:
history, subdivision, and admixture." Am J Hum Genet 57(2): 455-64.
First, M. B., Ed. (2000). Diagnostic and Statistical Manual - Text Revision (DSM -
IV-TR. 20001. Washington, D.C., American Psychiatric Association.
Fisher, R. M., S. E. Humphries, et al. (1997). "Common variation in the lipoprotein
lipase gene: effects on plasma lipids and risk o f atherosclerosis."
Atherosclerosis 135(2): 145-59.
Flint, W. (1994). Placentation. Marshall's Physiology o f Reproduction. 4th Edition.
L. GE. N ew York, Chapman and Hall: 233-460.
198
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Folgero, T., N. Storbakk, et al. (1996). "Mutations in mitochondrial transfer
ribonucleic acid genes in preeclampsia." American Journal o f Obstetrics &
G ynecology 174(5): 1626-30.
Forstermann, U. (1994). "Biochemistry and molecular biology o f nitric oxide
synthases." Arzneimittelforschung 44(3A): 402-7.
Fowler, C. J. and B. A. Callingham (1978). "Substrate-selective activation o f rat liver
mitochondrial monoamine oxidase by oxygen." Biochem Pharmacol 27(16):
1995-2000.
Fowler, J. S., R. R. MacGregor, et al. (1987). "Mapping human brain monoamine
oxidase A and B with llC -labeled suicide inactivators and PET." Science
235(4787): 481-5.
Fowler, J. S., N. D. Volkow, et al. (1996). "Brain monoamine oxidase A inhibition in
cigarette smokers." Proc Natl Acad Sci U S A 93(24): 14065-9.
Fowler, J. S., N. D. Volkow, et al. (1998). "Visualization o f monoamine oxidase in
human brain." Adv Pharmacol 42: 304-7.
Franco, R. F., J. Elion, et al. (1998). "Heterogeneous ethnic distribution o f the
844ins68 in the cystathionine beta-synthase gene." Hum Hered 48(6): 338-42.
Franques, P., M. Auriacombe, et al. (2003). "Sensation seeking as a common factor
in opioid dependent subjects and high risk sport practicing subjects. A cross
sectional study." Drug Alcohol Depend 69(2): 121-6.
Franques, P., M. Auriacombe, et al. (2000). "[Addiction and personality]." Encephale
26(1): 68-78.
Fridovich, I. (1978). "The biology o f oxygen radicals." Science 201(4359): 875-80.
Friedman, S. (1983). "Female anorgasmia and MAOIs." Am J Psychiatry 140(5):
665.
Frosst, P., H. J. Blom , et al. (1995). "A candidate genetic risk factor for vascular
disease: a comm on mutation in methylenetetrahydrofolate reductase." Nat
Genet 1011): 111-3.
Fu, Y., S. G. Matta, et al. (2001). "Norepinephrine secretion in the hypothalamic
paraventricular nucleus o f rats during unlimited access to self-administered
nicotine: An in vivo microdialysis study." J Neurosci 21(22): 8979-89.
Furui, T., O. Kurauchi, et al. (1994). "Decrease in cytochrome c oxidase and
cytochrome oxidase subunit I messenger R N A levels in preeclamptic
pregnancies." Obstetrics & Gynecology 84(2): 283-8.
199
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Gade, R., D. Muhleman, et al. (1998). "Correlation o f length o f VNTR alleles at the
X-linked M AO A gene and phenotypic effect in Tourette syndrome and drug
abuse." M ol Psychiatry 3(11: 50-60.
Gallery, E. D ., S. Campbell, et al. (1999). "Preeclamptic decidual microvascular
endothelial cells express lower levels o f matrix metalloproteinase-1 than
normals." M icrovasc Res 57(3): 340-6.
Garrick, N. A. and D. L. Murphy (1982). "Monoamine oxidase type A: differences in
selectivity towards 1-norepinephrine compared to serotonin." Biochem
Pharmacol 31(241: 4061-6.
Genbacev, O., R. Joslin, et al. (1996). "Hypoxia alters early gestation human
cytotrophoblast differentiation/invasion in vitro and m odels the placental
defects that occur in preeclampsia." J Clin Invest 97(2): 540-50.
Genbacev, O., Y. Zhou, et al. (1997). "Regulation o f human placental development
by oxygen tension." Science 277(5332): 1669-72.
Geracioti, T. D., Jr., S. A. West, et al. (1999). "Low CSF concentration o f a
dopamine metabolite in tobacco smokers." Am J Psychiatry 156(1): 130-2.
Gerra, G., P. Avanzini, et al. (1999). "Neurotransmitters, neuroendocrine correlates
o f sensation-seeking temperament in normal humans." Neuropsvchobiologv
39(4): 207-13.
Getzoff, E. D., J. A. Tainer, et al. (1983). "Electrostatic recognition between
superoxide and copper, zinc superoxide dismutase." Nature 306(5940): 287-
90.
Girgin Sagin, F., E. Y. Sozmen, et al. (2004). "Link between monoamine oxidase and
nitric oxide." Neurotoxicology 25(1-2): 91-9.
Giudice, L. C., N. A. Martina, et al. (1997). "Insulin-like growth factor binding
protein-1 at the maternal-fetal interface and insulin-like growth factor-I,
insulin-like growth factor-II, and insulin-like growth factor binding protein-1
in the circulation o f wom en with severe preeclampsia." Am J Obstet Gvnecol
176(4): 751-7; discussion 757-8.
Glassman, A. H., J. E. Helzer, et al. (1990). "Smoking, smoking cessation, and major
depression." Jama 264(12): 1546-9.
Goldkrand, J. W. and A. M. Fuentes (1986). "The relation o f angiotensin-converting
enzyme to the pregnancy-induced hypertension-preeclampsia syndrome." Am
J Obstet Gvnecol 154(4): 792-800.
200
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Goldman-Wohl, D. S., I. Ariel, et al. (2000). "Lack o f human leukocyte antigen-G
expression in extravillous trophoblasts is associated with pre-eclampsia." Mol
Hum Reprod 6(1): 88-95.
Goldstein, D. P. and R. S. Berkowitz (1994). "Current management o f complete and
partial molar pregnancy." Journal o f Reproductive M edicine 39(3): 139-46.
Grandone, E., M. Margaglione, et al. (1997). "Factor V Leiden, C > T MTHFR
polymorphism and genetic susceptibility to preeclampsia." Thrombosis &
Haemostasis 77f6): 1052-4.
Grandone, E., M. Margaglione, et al. (1999). "Prothrombotic genetic risk factors and
the occurrence o f gestational hypertension with or without proteinuria."
Thromb Haemost 81(3): 349-52.
Greenawalt, J. W. and C. Schnaitman (1970). "An appraisal o f the use o f monoamine
oxidase as an enzyme marker for the outer membrane o f rat liver
mitochondria." J Cell Biol 46(1): 173-9.
Grimsby, J., K. Chen, et al. (1992). "Dinucleotide repeat (TG)23 polymorphism in
the M AOB gene." Nucleic Acids Res 20(4): 924.
Grimsby, J., K. Chen, et al. (1991). "Human monoamine oxidase A and B genes
exhibit identical exon-intron organization." Proc Natl Acad Sci U S A 88(9):
3637-41.
Grimsby, J., N. C. Lan, et al. (1990). "Tissue distribution o f human monoamine
oxidase A and B mRNA." J Neurochem 55(4): 1166-9.
Grove, W. M., E. D. Eckert, et al. (1990). "Heritability o f substance abuse and
antisocial behavior: a study o f m onozygotic twins reared apart." Biol
Psychiatry 27(12): 1293-304.
Gu, D. F., L. J. Hinks, et al. (2000). "The use o f long PCR to confirm three common
alleles at the CYP2A6 locus and the relationship between genotype and
smoking habit." Ann Hum Genet 64(Pt 5): 383-90.
Guillemot, F., A. Nagy, et al. (1994). "Essential role o f Mash-2 in extraembryonic
development." Nature 371(6495): 333-6.
Gujrati, V. R., K. Shanker, et al. (1985). "Serotonin in toxaemia o f pregnancy."
Clinical & Experimental Pharmacology & Physiology 12(1): 9-18.
Gujrati, V. R., K. Shanker, et al. (1996). "Novel appearance o f placental nuclear
monoamine oxidase: biochemical and histochemical evidence for
hyperserotonomic state in preeclampsia-eclampsia." Am J Obstet Gvnecol
175(6): 1543-50.
201
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Guo, G., J. A. Lade, et al. (1999). "Genetic susceptibility to pre-eclampsia and
chromosome 7q36." Human Genetics 105(6): 641-7.
Guo, G., A. N. Wilton, et al. (1997). "Angiotensinogen gene variation in a population
case-control study o f preeclampsia/eclampsia in Australians and Chinese."
Electrophoresis 18(9): 1646-9.
Hagemann, A., A. H. Nielsen, et al. (1994). "The uteroplacental renin-angiotensin
system: a review." Exp Clin Endocrinol 102(3): 252-61.
Hallman, J., E. Sakurai, et al. (1990). "Blood platelet monoamine oxidase activity,
serotonin uptake and release rates in anorexia and bulimia patients and in
healthy controls." Acta Psvchiatr Scand 81(1): 73-7.
Hannah, M. C., J. L. Hopper, et al. (1985). "Twin concordance for a binary trait. II.
Nested analysis o f ever-smoking and ex-sm oking traits and unnested analysis
o f a "committed-smoking" trait." Am J Hum Genet 37(1): 153-65.
Harrison, G. A., K. E. Humphrey, et al. (1997). "A genom ewide linkage study o f
preeclampsia/eclampsia reveals evidence for a candidate region on 4q."
American Journal o f Human Genetics 60151: 1158-67.
Haugen, G. and N. M oe (1999). "In vitro perfusion studies o f single umbilical artery
cords: the vasoactive effects o f serotonin." Acta Obstet Gvnecol Scand 78(4):
285-9.
Hayakawa, K. (1987). "Smoking and drinking discordance and health condition:
Japanese identical twins reared apart and together." Acta Genet Med
G em ellol (Roma) 36(4): 493-501.
Hayward, C., J. Livingstone, et al. (1992). "An exclusion map for pre-eclampsia:
assuming autosomal recessive inheritance [see comments]." American
Journal o f Human Genetics 50(4): 749-57.
Heath, A. C. and P. G. Madden (1995). Behavior genetic approaches in behavioural
medicine. Behavior Genetic Influences on Smoking Behavior. J. R. Turner,
L. R. Cardon and J. K. Hewitt. N ew York, Plenum: 37-48.
Heath, A. C. and N. G. Martin (1993). "Genetic models for the natural history o f
smoking: evidence for a genetic influence on smoking persistence." Addict
Behav 18(1): 19-34.
Hefler, L. A., C. B. Tempfer, et al. (2001). "Polymorphisms within the interleukin-1
beta gene cluster and preeclampsia." Obstet Gvnecol 97(5 Pt 1): 664-8.
202
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Heiskanen, J. T., M. M. Pirskanen, et al. (2001). "Insertion-deletion polymorphism in
the gene for angiotensin-converting enzyme is associated with obstetric
cholestasis but not with preeclampsia." Am J Obstet Gvnecol 185(3): 600-3.
Higgins, J. R., T. Kaiser, et al. (2000). "Prothrombin G 20210A mutation: is it
associated with pre-eclampsia?" Gynecologic & Obstetric Investigation
50(4): 254-7.
Hinds, H. L., R. W. Hendriks, et al. (1992). "Characterization o f a highly
polymorphic region near the first exon o f the human M AO A gene containing
a GT dinucleotide and a novel VNTR motif." Genomics 13(3): 896-7.
Hjalmarsson, K., S. L. Marklund, et al. (1987). "Isolation and sequence o f
complementary D N A encoding human extracellular superoxide dismutase."
Proc Natl Acad Sci U S A 84(18): 6340-4.
Holschneider, D. P., T. Kumazawa, et al. (1998). "Tissue-specific effects o f estrogen
on monoamine oxidase A and B in the rat." Life Sci 63(3): 155-60.
Hotamisligil, G. S. and X. O. Breakefield (1991). "Human monoamine oxidase A
gene determines levels o f enzyme activity." Am J Hum Genet 49(2): 383-92.
Hsu, Y. P., E. W. Loh, et al. (1996). "Association o f monoamine oxidase A alleles
with alcoholism among male Chinese in Taiwan." Am J Psychiatry 153(9):
1209-11.
Hubei, C. A., J. M. Roberts, et al. (1999). "Association o f pre-eclampsia with
common coding sequence variations in the lipoprotein lipase gene." Clinical
Genetics 56(4): 289-96.
Humphrey, K. E., G. A. Harrison, et al. (1995). "HLA-G deletion polymorphism and
pre-eclampsia/eclampsia." British Journal o f Obstetrics & Gynaecology
102(9): 707-10.
Hunt, J. S., H. L. Chen, et al. (1996). "Tumor necrosis factors: pivotal components o f
pregnancy?" Biol Reprod 54(3): 554-62.
Hunt, J. S. and H. T. Orr (1992). "HLA and maternal-fetal recognition." Faseb J 6(6):
2344-8.
Hutter, J. and M. Filshie (1993). "Role o f serotonin in preeclampsia and eclampsia."
Contemp Rev Obstet Gvnecol 5: 199-201.
Ibanez, A., I. P. de Castro, et al. (2000). "Pathological gambling and D N A
polymorphic markers at M AO-A and M AO-B genes." M ol Psychiatry 5(1):
105-9.
2 0 3
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Ijlst, L., J. P. Ruiter, et al. (1996). "Long-chain 3-hydroxyacyl-CoA dehydrogenase
deficiency: a new method to identify the G1528C mutation in genomic D N A
showing its high frequency (approximately 90%) and identification o f a new
mutation (T2198C)." J Inherit Metab P is 19(2): 165-8.
Imperato, A., A. Mulas, et al. (1986). "Nicotine preferentially stimulates dopamine
release in the limbic system o f freely m oving rats." Eur J Pharmacol 132(2-
3): 337-8.
Irwin, J. C., L. F. Suen, et al. (1999). "Role o f the IGF system in trophoblast invasion
and pre-eclampsia." Hum Reprod 14 Suppl 2: 90-6.
Ito, H., N. Hamajima, et al. (2003). "Monoamine oxidase polymorphisms and
smoking behaviour in Japanese." Pharmacogenetics 13(2): 73-9.
Janatpour, M. J., M. F. Utset, et al. (1999). "A repertoire o f differentially expressed
transcription factors that offers insight into mechanisms o f human
cytotrophoblast differentiation." D ev Genet 25(2): 146-57.
Johansson, F., L. Von Knorring, et al. (1983). "Platelet MAO activity in patients with
chronic pain syndrome. Relationship to personality traits, endorphins in CSF
and plasma cortisol." Med Biol 61(2): 101-4.
Johnston, J. P. (1968). "Some observations upon a new inhibitor o f monoamine
oxidase in brain tissue." Biochem Pharmacol 17(7): 1285-97.
Jonsson, E. G., N. Norton, et al. (2003). "Association between a promoter variant in
the monoamine oxidase A gene and schizophrenia." Schizophr Res 61(1): 31-
7.
Jurajda, M., K. Kankova, et al. (2001). "Lack o f an association o f a single nucleotide
polymorphism in the promoter o f the matrix metalloproteinase-1 gene in
Czech wom en with pregnancy-induced hypertension." Gvnecol Obstet Invest
52(2): 124-7.
Kaiser, T., S. P. Brennecke, et al. (2000). "Methylenetetrahydrofolate reductase
polymorphisms are not a risk factor for pre-eclampsia/eclampsia in
Australian women." Gynecologic & Obstetric Investigation 50(2): 100-2.
Kandel, D., K. Chen, et al. (1997). "Prevalence and demographic correlates o f
symptoms o f last year dependence on alcohol, nicotine, marijuana and
cocaine in the U.S. population." Drug Alcohol Depend 44(1): 11-29.
Kaunitz, A. M ., J. M. Hughes, et al. (1985). "Causes o f maternal mortality in the
United States." Obstet Gvnecol 65(5): 605-12.
204
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Kendler, K. S., M. C. Neale, et al. (1999). "A population-based twin study in women
o f smoking initiation and nicotine dependence." Psychol M ed 29(2): 299-308.
Kendler, K. S., L. M. Thornton, et al. (2000). "Tobacco consumption in Swedish
twins reared apart and reared together." Arch Gen Psychiatry 57(9): 886-92.
Khoury, M. J. and T. H. Beaty (1994). "Applications o f the case-control method in
genetic epidemiology." Epidemiol Rev 16(1): 134-50.
Khoury, M. J., Yang, Q. H. (1998). "The future o f genetic studies o f complex human
diseases: An epidemiologic perspective." Epidem iology 9(31: 350-354.
Kilpatrick, D. C. (1999). "Influence o f human leukocyte antigen and tumour necrosis
factor genes on the development o f pre-eclampsia." Human Reproduction
Update 5(21: 94-102.
Kilpatrick, D. C., F. Gibson, et al. (1990). "Pre-eclampsia is associated with HLA-
DR4 sharing between mother and fetus." Tissue Antigens 35(4): 178-81.
Kim, J. J., J. C. Shih, et al. (1997). "Selective enhancement o f emotional, but not
' motor, learning in monoamine oxidase A-deficient mice." Proc Natl Acad Sci
U S A 941111: 5929-33.
Kim, Y. J., R. A. W illiamson, et al. (2001). "Lipoprotein lipase gene mutations and
the genetic susceptibility o f preeclampsia." Hypertension 38(5): 992-6.
Kim, Y. J., W illiamson, R.A., Murray, J.C., Andrews, J., Pietscher, J.J., Peraud, P.J.,
Merrill, D.C. (2001). "Genetic susceptibility to preeclampsia: Roles o f
cytosine-to-thymine substitution at nucleotide 677 o f the gene for
methylenetetrahydrofolate reductase, 68-base pair insertion at nucleotide 844
o f the gene for cystathionine beta-synthase, and factor V Leiden mutation."
Americal Journal o f Obstetrics and Gynecology 184(1211-1217).
Knock, G. A., M. H. Sullivan, et al. (1994). "Angiotensin II (ATI) vascular binding
sites in human placentae from normal-term, preeclamptic and growth
retarded pregnancies." J Pharmacol Exp Ther 271(2): 1007-15.
Knofler, M ., B. Kalionis, et al. (2000). "Novel genes and transcription factors in
placental development— a workshop report." Placenta 21 Suppl A: S71-3.
Knofler, M ., G. Meinhardt, et al. (2002). "Human Handl basic helix-loop-helix
(bHLH) protein: extra-embryonic expression pattern, interaction partners and
identification o f its transcriptional repressor domains." Biochem J 361 (Pt 3):
641-51.
205
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Knofler, M ., G. Meinhardt, et al. (1998). "Molecular cloning o f the human Handl
gene/cD N A and its tissue- restricted expression in cytotrophoblastic cells and
heart." Gene 224(1-2): 77-86.
Knoll, J. and K. Magyar (1972). "Some puzzling pharmacological effects o f
monoamine oxidase inhibitors." Adv Biochem Psvchopharmacol 5: 393-408.
Knowler, W. C., R. C. W illiams, et al. (1988). "Gm3;5,13,14 and type 2 diabetes
mellitus: an association in American Indians with genetic admixture.” Am J
Hum Genet 43(4): 520-6.
Kobashi, G. (1995). "[A case-control study o f pregnancy-induced hypertension with
a genetic predisposition: association o f a molecular variant o f
angiotensinogen in the Japanese women]." Hokkaido Igaku Zasshi -
Hokkaido Journal o f M edical Science 70(4): 649-57.
Kobashi, G., A. Hata, et al. (1999). "Association o f a variant o f the angiotensinogen
gene with pure type o f hypertension in pregnancy in the Japanese:
implication o f a racial difference and significance o f an age factor."
American Journal o f M edical Genetics 86(3): 232-6.
Kobashi, G., K. Shido, et al. (2001). "Multivariate analysis o f genetic and acquired
factors; T235 variant o f the angiotensinogen gene is a potent independent risk
factor for preeclampsia." Semin Thromb Hemost 27(2): 143-7.
Kobashi, G., H. Yamada, et al. (2001). "Endothelial nitric oxide synthase gene
(NO S3) variant and hypertension in pregnancy." Am J M ed Genet 103(3):
241-4.
Kobashi, G., Yamada, H., Asano, T., Nagano, S., Hata, A., Kishi, R., Kondo, K.,
Fujimoto, S. (1999). "The factor V leiden mutation is not a common cause o f
pregnancy-induced hypertension in Japan." Seminars in Thrombosis &
Hem ostasis 25(5): 487-489.
Kok, F. J., A. W. Matroos, et al. (1982). "Correlates o f body mass index in 1926
Dutch men and w om en.” Hum Nutr Clin Nutr 36(2): 155-65.
Konradi, C., L. Ozelius, et al. (1992). "Highly polymorphic (GT)n repeat sequence in
intron II o f the human M AOB gene." Genomics 12(1): 176-7.
Koob, G. F. and E. J. Nestler (1997). "The neurobiology o f drug addiction." J
Neuropsychiatry Clin Neurosci 9(3): 482-97.
Kraft, R and M. W ilson (2002). "Family-based association tests incorporating
parental genotypes." Am J Hum Genet 71(5): 1238-9; author reply 1240-2.
206
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Kraus, J. P., J. Oliveriusova, et al. (1998). "The human cystathionine beta-synthase
(CBS) gene: complete sequence, alternative splicing, and polymorphisms."
Genomics 52(3): 312-24.
Kraut, N ., L. Snider, et al. (1998). "Requirement o f the mouse I-mfa gene for
placental development and skeletal patterning." Embo J 17(21): 6276-88.
Kua, E. H. (1998). "Mental health o f the elderly Chinese population in Singapore."
Psychiatry Clin Neurosci 52 Suppl: S297-9.
Kupfer, D. J. and M. B. Bowers, Jr. (1972). "REM sleep and central monoamine
oxidase inhibition." Psvchopharmacologia 27(3): 183-90.
Kupferminc, M. J., A. Eldor, et al. (1999). "Increased frequency o f genetic
thrombophilia in wom en with complications o f pregnancy." N ew England
Journal o f M edicine 340(1): 9-13.
Kupferminc, M. J., G. Fait, et al. (2000). "Severe preeclampsia and high frequency o f
genetic thrombophilic mutations." Obstetrics & Gynecology 96(1): 45-9.
Kupferminc, M. J., A. M. Peaceman, et al. (1994). "Tumor necrosis factor-alpha is
elevated in plasma and amniotic fluid o f patients with severe preeclampsia."
Am J Obstet Gvnecol 170(6): 1752-7; discussion 1757-9.
Kupferminc, M. J., H. Peri, et al. (2000). "High prevalence o f the prothrombin gene
mutation in wom en with intrauterine growth retardation, abruptio placentae
and second trimester loss." Acta Obstet Gvnecol Scand 79(11): 963-7.
Lachmeijer, A. M., R. Amgrimsson, et al. (2001). "A genom e-wide scan for
preeclampsia in the Netherlands." Eur J Hum Genet 9(10): 758-64.
Lachmeijer, A. M ., R. Amgrimsson, et al. (2001). "Mutations in the gene for
methylenetetrahydrofolate reductase, homocysteine levels, and vitamin status
in wom en with a history o f preeclampsia." Am J Obstet Gvnecol 184(3): 394-
402.
Lachmeijer, A. M ., J. B. Cmsius, et al. (2001). "Polymorphisms in the tumor
necrosis factor and lymphotoxin-alpha gene region and preeclampsia." Obstet
Gvnecol 98(4): 612-9.
Lade, J. A., E. K. M oses, et al. (1999). "The eNOS gene: a candidate for the
preeclampsia susceptibility locus?" Hypertension in Pregnancy 18(1): 81-93.
Laivuori, H., R. Kaaja, et al. (2000). "677 C -> T polymorphism o f the
methylenetetrahydrofolate reductase gene and preeclampsia." Obstetrics &
G ynecology 96(2): 277-80.
207
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Lander, E. S. and N. J. Schork (1994). "Genetic dissection o f complex traits."
Science 265(5181): 2037-48.
Landsbergis, P. A. and M. C. Hatch (1996). "Psychosocial work stress and
pregnancy-induced hypertension [see comments]." Epidem iology 7(4): 346-
51.
Lavin, M. R., A. M endelowitz, et al. (1993). "Spontaneous hypertensive reactions
with monoamine oxidase inhibitors." Biol Psychiatry 34(3): 146-51.
Laviolette, S. R. and D. van der K ooy (2003). "Blockade o f m esolim bic dopamine
transmission dramatically increases sensitivity to the rewarding effects o f
nicotine in the ventral tegmental area." M ol Psychiatry 8(1): 50-9, 9.
Lehrer, S., J. Rabin, et al. (1993). "Estrogen receptor variant and hypertension in
women." Hypertension 21(4): 439-41.
Lenz, B. K. (2004). "Tobacco, depression, and lifestyle choices in the pivotal early
college years." J Am Coll Health 52(5): 213-9.
Lerman, C., N. E. Caporaso, et al. (1999). "Evidence suggesting the role o f specific
genetic factors in cigarette smoking." Health Psychol 18(1): 14-20.
Lerman, C., N. E. Caporaso, et al. (2001). "Tryptophan hydroxylase gene variant and
smoking behavior." Am J Med Genet 105(6): 518-20.
Lerman, C., P. G. Shields, et al. (2003). "Effects o f dopamine transporter and
receptor polymorphisms on smoking cessation in a bupropion clinical trial."
Health Psvchol 22151: 541-8.
Leung, P. S., S. J. Tsai, et al. (2001). "The upregulation o f angiotensin II receptor
AT(1) in human preeclamptic placenta." M ol Cell Endocrinol 184(1-2): 95-
102.
Levine, A. J., K. D. Siegmund, et al. (2000). "The methylenetetrahydrofolate
reductase 677C-->T polymorphism and distal colorectal adenoma risk."
Cancer Epidemiol Biomarkers Prev 9(7): 657-63.
Lewinsohn, R., V. Glover, et al. (1980). "Development o f benzylamine oxidase and
monoamine oxidase A and B in man." Biochem Pharmacol 29(9): 1221-30.
Li, J., H. Y. Hu, et al. (1992). "Serum angiotensin-converting enzyme activity in
pregnancy-induced hypertension." Gvnecol Obstet Invest 33(3): 138-41.
Lie, R., Rasmussen HB, Gjessing HK, N ielsen EL, Irgens LM (1998). "Fetal and
maternal contributions torisk o f pre-eclampsia: population based study." BMJ
316: 1343-7.
208
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Lim, L. C., J. Powell, et al. (1995). "Evidence for a genetic association between
alleles o f monoamine oxidase A gene and bipolar affective disorder." Am J
M ed Genet 60141: 325-31.
Liston, W. A. and D. C. Kilpatrick (1991). "Is genetic susceptibility to pre-eclampsia
conferred by hom ozygosity for the same single recessive gene in mother and
fetus?" British Journal o f Obstetrics & Gynaecology 98(11): 1079-86.
Livingston, J. C., J. R. Barton, et al. (2001). "Maternal and fetal inherited
thrombophilias are not related to the development o f severe preeclampsia."
Am J Obstet Gynecol 185111: 153-7.
Lorentzen, B. and T. Henriksen (1994). "[Acetylsalicylic acid and pre
eclampsia/growth retardation (II)]." Tidsskr Nor Laegeforen 114(17): 2033.
Lyall, F., J. N. Bulmer, et al. (1999). "Human trophoblast invasion and spiral artery
transformation: the role o f nitric oxide." Am J Pathol 154(4): 1105-14.
Lyall, F. and I. A. Greer (1994). "Pre-eclampsia: a multifaceted vascular disorder,of
pregnancy." JHypertens 12(12): 1339-45.
MacGillivray, I. (1983). Epidem iology o f Preeclampsia and Eclampsia.
Preeclampsia. I. MacGillivray. London, WB Saunders: 8-22.
MacKay, A. P., C. J. Berg, et al. (2001). "Pregnancy-related mortality from
preeclampsia and eclampsia." Obstetrics & Gynecology 97(4): 533-8.
Makkonen, N ., S. Heinonen, et al. (2001). "Apolipoprotein E alleles in wom en with
pre-eclampsia." J Clin Pathol 54(8): 652-4.
Malamitsi-Puchner, A., J. Tziotis, et al. (2001). "Gene analysis o f the N-terminal
region o f the estrogen receptor alpha in preeclampsia." Steroids 66(9): 695-
700.
Mander, A. P. (2001). "Haplotype Analysis in Population-Based Association
Studies." The Stata Journal 1: 58-75.
Mannino, D. M. (2003). "Chronic obstructive pulmonary disease: definition and
epidemiology." Respir Care 481121: 1185-91; discussion 1191-3.
Manor, I., S. Tyano, et al. (2002). "Family-based and association studies o f
monoamine oxidase A and attention deficit hyperactivity disorder (ADHD):
preferential transmission o f the long promoter-region repeat and its
association with impaired performance on a continuous performance test
(TOVA)." M ol Psychiatry 7(6): 626-32.
209
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Mathew, J., K. Basheeruddin, et al. (2001). "Differences in frequency o f the deletion
polymorphism o f the angiotensin-converting enzyme gene in different ethnic
groups." Angiplogy 52(6): 375-9.
Maynard, S. E., J. Y. Min, et al. (2003). "Excess placental soluble fms-like tyrosine
kinase 1 (sF ltl) may contribute to endothelial dysfunction, hypertension, and
proteinuria in preeclampsia." J Clin Invest 111(5): 649-58.
McCowan, L. M. and D. M. Becroft (1994). "Beckwith-Wiedemann syndrome,
placental abnormalities, and gestational proteinuric hypertension." Obstet
G ynecol 83(5 Pt 2): 813-7.
M cCully, K. S. (1996). "Homocysteine and vascular disease." Nat M ed 2(4): 386-9.
McKinney, E. F., R. T. Walton, et al. (2000). "Association between polymorphisms
in dopamine metabolic enzymes and tobacco consumption in smokers."
Pharmacogenetics 10(6): 483-91.
M ello, G., Parretti, E., Martini, E., M ecacci, F., La Torre, P., Cioni, R., Lucchetti, R.,
Fedi, S., Gori, A.M ., Pepe, G., Prisco, D., Abbate, R. (1999). "Usefulness o f
screening for congenital or acquired hemostatic abnormalities in wom en with
previous complicatd pregnancies." Haemostasis 29: 197-203.
Merlo Pich, E., C. Chiamulera, et al. (1999). "Molecular mechanisms o f the positive
reinforcing effect o f nicotine." Behav Pharmacol 10(6-7): 587-96.
M ichelson, A. M., J. M. McCord, et al. (1977). Superoxide and superoxide
dismutases. N ew York, Academic Press.
Miettinen, T. A. (1991). "Impact o f apo E phenotype on the regulation o f cholesterol
metabolism." Ann Med 23121: 181-6.
Millauer, B., L. K. Shawver, et al. (1994). "Glioblastoma growth inhibited in vivo by
a dominant-negative Flk-1 mutant." Nature 367(6463): 576-9.
Mitchell, J. E. and M. K. Popkin (1983). "Antidepressant drug therapy and sexual
dysfunction in men: a review." J Clin Psvchopharmacol 3(2): 76-9.
Miyamoto, Y., Saito, Y., Kajiyama, N ., Yoshimura, M ., Shimasaki, Y., Nakayama,
M., Kamitani, S., Harada, M., Ishikawa, M., Kuwahara, K., Ogawa, E.,
Hamanaka, I., Takahashi, N., Kaneshige, T., Teraoka, H., Akamizu, T.,
Azuma, N ., Yoshimasa, Y., Yoshimasa, T., Itoh, H., Masuda, I., Yasue, H.,
Nakao, K. (1998). "Endothelial nitric oxide synthase gene is positively
associated with essential hypertension." Hypertension 32(1): 3-8.
Mogren, I., U. Hogberg, et al. (1999). "Familial occurrence o f preeclampsia."
Epidem iology 10(5): 518-22.
210
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Morgan, L., P. Baker, et al. (1995). "Pre-eclampsia and the angiotensinogen gene."
British Journal o f Obstetrics & Gynaecology 102(6): 489-90.
Morgan, L., S. Crawshaw, et al. (1998). "Distortion o f maternal-fetal angiotensin II
type 1 receptor allele transmission in pre-eclampsia." Journal o f Medical
Genetics 35(8): 632-6.
Morgan, L., S. Crawshaw, et al. (1999). "Maternal and fetal angiotensinogen gene
allele sharing in pre-eclampsia." British Journal o f Obstetrics & Gynaecology
106(3): 244-51.
Morgan, L., F. Foster, et al. (1999). "Angiotensin-converting enzyme insertion-
deletion polymorphism in normotensive and pre-eclamptic pregnancies." J
Hvpertens 17(6): 765-8.
Morgan, T., C. Craven, et al. (1999). "Angiotensinogen Thr235 variant is associated
with abnormal physiologic change o f the uterine spiral arteries in first-
trimester decidua." Am J Obstet Gvnecol 180(1 Pt 1): 95-102.
Morgan, T., C. Craven, et al. (1997). "Angiotensinogen T235 expression is elevated
in decidual spiral arteries." Journal o f Clinical Investigation 100(6): 1406-15.
Morgan, T., C. Craven, et al. (1998). "Human spiral artery renin-angiotensin
system." Hypertension 32(4): 683-7.
Morgan, T. and K. Ward (1999). "New insights into the genetics o f preeclampsia."
Seminars in Perinatology 23(11: 14-23.
M oses, E. K., J. A. Lade, et al. (2000). "A genome scan in families from Australia
and N ew Zealand confirms the presence o f a maternal susceptibility locus for
pre-eclampsia, on chromosome 2." Am J Hum Genet 67(6): 1581-5.
Nagy, B., J. Rigo, Jr., et al. (1998). "Apolipoprotein E alleles in wom en with severe
pre-eclampsia." Journal o f Clinical Pathology 51(4): 324-5.
Nagy, B., T. Toth, et al. (1998). "Detection o f factor V Leiden mutation in severe
pre-eclamptic Hungarian women." Clinical Genetics 53(6): 478-81.
Nakajima, T., L. B. Jorde, et al. (2002). "Nucleotide diversity and haplotype
structure o f the human angiotensinogen gene in two populations." Am J Hum
Genet 7011): 108-23.
Nakatome, M ., Z. Tun, et al. (1998). "Detection and analysis o f four polymorphic
markers at the human monoamine oxidase (MAO) gene in Japanese controls
and patients with Parkinson's disease." Biochem Biophvs Res Commun
247(2): 452-6.
211
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Need, J. A. (1975). "Pre-eclampsia in pregnancies by different fathers:
imm unological studies." Br Med J 1(5957): 548-9.
Ness, R. B. and J. M. Roberts (1996). "Heterogeneous causes constituting the single
syndrome o f preeclampsia: a hypothesis and its implications [see
comments]." American Journal o f Obstetrics & Gynecology 17515): 1365-70.
Netter, P., J. Hennig, et al. (1996). "Serotonin and dopamine as mediators o f
sensation seeking behavior." Neuropsvchobiology 34(3): 155-65.
Niu, T., C. Chen, et al. (2000). "Nicotine dependence and its familial aggregation in
Chinese." Int J Epidemiol 29(2): 248-52.
Noble, E. P., S. T. St Jeor, et al. (1994). "D2 dopamine receptor gene and cigarette
smoking: a reward gene?" Med Hypotheses 42(4): 257-60.
Noble, E. P., X. Zhang, et al. (1998). "D2 dopamine receptor and G ABA(A) receptor
beta3 subunit genes and alcoholism." Psychiatry Res 81(2): 133-47.
O'Brien, M ., J. Dausset, et al. (2000). "Analysis o f the role o f HLA-G in
preeclampsia." Hum Immunol 61(11): 1126-31.
O'Shaughnessy, K. M., B. Fu, et al. (2001). "Thrombophilic polymorphisms in pre
eclampsia: altered frequency o f the functional 98C>T polymorphism o f
glycoprotein Ilia." J Med Genet 38(11): 775-7.
O'Shaughnessy, K. M., B. Fu, et al. (1999). "Factor V Leiden and thermolabile
methylenetetrahydrofolate reductase gene variants in an East Anglian
preeclampsia cohort." Hypertension 33(6): 1338-41.
Osier, M., C. Holst, et al. (2001). "Influence o f genes and family environment on
adult smoking behavior assessed in an adoption study." Genet Epidemiol
21(3): 193-200.
Ostendorf, T., U. Kunter, et al. (1999). ”VEGF(165) mediates glomerular endothelial
repair." J Clin Invest 104(7): 913-23.
Ozelius, L., J. F. Gusella, et al. (1989). "MspI RFLP for human M AO A gene."
Nucleic Acids Res 17(24): 10516.
Ozelius, L., Y. P. Hsu, et al. (1988). "Human monoamine oxidase gene (MAOA):
chromosome position (X p 21-p ll) and D N A polymorphism." Genomics 3(1):
53-8.
Parsian, A. and C. R. Cloninger (1997). "Human GABAA receptor alpha 1 and alpha
3 subunits genes and alcoholism." Alcohol Clin Exp Res 21(3): 430-3.
212
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Parsian, A. and Z. H. Zhang (1999). "Human chromosomes l lp l 5 and 4 p l2 and
alcohol dependence: possible association with the GABRB1 gene." Am J
M ed Genet 88(51: 533-8.
Perkins, K. A., J. E. Sexton, et al. (1996). "Acute thermogenic effects o f nicotine and
alcohol in healthy male and female smokers." Physiol Behav 60(1): 305-9.
Peterson, R., Tuck-Muller, C.M., Spinnato, J.A., Peevy, K., Giattina, K., et al.
(1994). "Short Communication an HLA-Haplotype Associated With
Preeclampsia and Intrauterine Growth Retardation." American Journal o f
Reproductive Immunology 31: 177-179.
Pidoplichko, V. I., M. DeBiasi, et al. (1997). "Nicotine activates and desensitizes
midbrain dopamine neurons." Nature 390(6658): 401-4.
Pipkin, F. B. (1999). "What is the Place o f Genetics in the Pathogenesis o f Pre-
Eclampsia?" Biology o f the Neonate 76: 325-330.
Pontieri, F. E., G. Tanda, et al. (1996). "Effects o f nicotine on the nucleus accumbens
and similarity to those o f addictive drugs." Nature 382(6588): 255-7.
Powers, R. W., M inich, L.A., Lykins, D.L., N ess, R.B., Crombleholme, W.R.,
Roberts, J.M. (1999). "Methylenetetrahydrofolate reductase polymorphism,
folate, and susceptibility to preeclampsia." J Soc Gynecol Invest 6: 74-79.
(1990). "National High Blood Pressure Education Program Working Group report on
high blood pressure in pregnancy: Consensus report." Americal Journal o f
Obstetrics and Gynecology 163: 1689-1712.
Pritchard, J. K. and P. Donnelly (2001). "Case-control studies o f association in
structured or admixed populations." Theor Popul Biol 60(3): 227-37.
Pritchard, J. K. and N. A. Rosenberg (1999). "Use o f unlinked genetic markers to
detect population stratification in association studies." Am J Hum Genet
65(1): 220-8.
Pritchard, J. K., M. Stephens, et al. (2000). "Inference o f population structure using
multilocus genotype data." Genetics 155(2): 945-59.
Pritchard, J. K., M. Stephens, et al. (2000). "Association mapping in structured
populations." Am J Hum Genet 6711): 170-81.
Quattrocki, E., A. Baird, et al. (2000). "Biological aspects o f the link between
smoking and depression." Harv Rev Psychiatry 8(3): 99-110.
Quitkin, F., A. Rifkin, et al. (1979). "Monoamine oxidase inhibitors. A review o f
antidepressant effectiveness." Arch Gen Psychiatry 36(7): 749-60.
213
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Rajakumar, A., K. Doty, et al. (2003). "Impaired Oxygen-dependent Reduction o f
HIF-1 alpha and -2alpha Proteins in Pre-eclamptic Placentae." Placenta 24(2-
3): 199-208.
Rao, Y., E. Hoffmann, et al. (2000). "Duplications and defects in the CYP2A6 gene:
identification, genotyping, and in vivo effects on smoking." M ol Pharmacol
58(4): 747-55.
Rauhut, A. S., S. N. M ullins, et al. (2002). "Reboxetine: attenuation o f intravenous
nicotine self-administration in rats." J Pharmacol Exp Ther 303(2): 664-72.
Redline, R. W. and P. Patterson (1995). "Pre-eclampsia is associated with an excess
o f proliferative immature intermediate trophoblast." Hum Pathol 26(6): 594-
600.
Rees, D. C., M. Cox, et al. (1995). "World distribution o f factor V Leiden." Lancet
346(8983): 1133-4.
Reibel, J. (2003). "Tobacco and oral diseases. Update on the evidence, with
recommendations." Med Princ Pract 12 Suppl 1: 22-32.
Reich, D. E. and D. B. Goldstein (2001). "Detecting association in a case-control
study while correcting for population stratification." Genet Epidemiol 20(1):
4-16.
Reuben, M. and P. B. Clarke (2000). "Nicotine-evoked [3H]5-hydroxytryptamine
release from rat striatal synaptosomes." Neuropharmacologv 39(2): 290-9.
Reynolds, L. P., A. T. Grazul-Bilska, et al. (2002). "Angiogenesis in the female
reproductive organs: pathological implications." Int J Exp Pathol 83(4): 151-
63.
Ribeiro, E. B., R. L. Bettiker, et al. (1993). "Effects o f systemic nicotine on serotonin
release in rat brain." Brain Res 621(2): 311-8.
Rigat, B., C. Hubert, et al. (1990). "An insertion/deletion polymorphism in the
angiotensin I-converting enzyme gene accounting for half the variance o f
serum enzyme levels." J Clin Invest 86(4): 1343-6.
Riley, P., L. Anson-Cartwright, et al. (1998). "The Handl bHLH transcription factor
is essential for placentation and cardiac morphogenesis." Nat Genet 18(3):
271-5.
Rinehart, B. K., D. A. Terrone, et al. (1999). "Expression o f the placental cytokines
tumor necrosis factor alpha, interleukin lbeta, and interleukin 10 is increased
in preeclampsia." Am J Obstet Gvnecol 181(4): 915-20.
214
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Rinkenberger, J. and Z. Werb (2000). "The labyrinthine placenta [news; comment]."
Nature Genetics 25(3): 248-50.
Risch, N. and K. Merikangas (1996). "The future o f genetic studies o f complex
human diseases." Science 273(5281): 1516-7.
Roberts, J. M. (2000). "Preeclampsia: What we know and what we do not know."
Semin Perinatal 24: 24-28.
Roberts, J. M. and D. W. Cooper (2001). "Pathogenesis and genetics o f pre
eclampsia." Lancet 357(9249): 53-6.
Robillard, P. Y. (2002). "Interest in preeclampsia for researchers in reproduction." J
Reprod Immunol 53(1-2): 279-87.
Ros, H. S. S., Lichtenstein, P., Lipworth, L., Cnattingius, S. (2000). "Genetic Effects
on the Liability o f Developing Preeclampsia and Gestational Hypertension."
American Journal o f M edical Genetics 91: 256-260.
Rosenberg, N. A., J. K. Pritchard, et al. (2002). "Genetic structure o f human
populations." Science 298(5602): 2381-5.
Rossant, J. and J. C. Cross (2001). "Placental development: lessons from mouse
mutants." Nat Rev Genet 2(7): 538-48.
Rubinsztein, D. C., J. Leggo, et al. (1996). "Genetic association between monoamine
oxidase A microsatellite and RFLP alleles and bipolar affective disorder:
analysis and meta-analysis." Hum M ol Genet 5(6): 779-82.
Rutter, J. L., T. I. M itchell, et al. (1998). "A single nucleotide polymorphism in the
matrix metalloproteinase-1 promoter creates an Ets binding site and augments
transcription." Cancer Res 58(23): 5321-5.
Sabol, S. Z., S. Hu, et al. (1998). "A functional polymorphism in the monoamine
oxidase A gene promoter." Hum Genet 103(3): 273-9.
Sabol, S. Z., M. L. Nelson, et al. (1999). "A genetic association for cigarette smoking
behavior." Health Psychol 18(1): 7-13.
Saftlas, A. F., D. R. Olson, et al. (1990). "Epidemiology o f preeclampsia and
eclampsia in the United States, 1979-1986 [see comments]." American
Journal o f Obstetrics & Gynecology 163(2): 460-5.
Sagol, S., O. Sagol, et al. (2002). "Stereological quantification o f placental villus
vascularization and its relation to umbilical artery Doppler flow in
intrauterine growth restriction." Prenat Diagn 22(5): 398-403.
215
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Saito, T., H. M. Lachman, et al. (2002). "Analysis o f monoamine oxidase A
(M AOA) promoter polymorphism in Finnish male alcoholics." Psychiatry
Res 109(2): 113-9.
Sam ochowiec, J., K. P. Lesch, et al. (1999). "Association o f a regulatory
polymorphism in the promoter region o f the monoamine oxidase A gene with
antisocial alcoholism." Psychiatry Res 86(1): 67-72.
Sander, T., J. Sam ochowiec, et al. (1999). "Association analysis o f exonic variants o f
the gene encoding the GABAB receptor and alcohol dependence." Psvchiatr
Genet 9(2): 69-73.
Sandler, M., S. Bonham Carter, et al. (1975). "Is there an increase in monoamine
oxidase activity in depressive illness?" Lancet I: 1045-9.
Sandler, M. and J. Coveney (1962). "Placental monoamine-oxidase activity in
toxaemia o f pregnancy." Lancet 1: 1096-8.
Sattar, N ., A. Bendomir, et al. (1997). "Lipoprotein subtraction concentrations in
preeclampsia: pathogenic parallels to atherosclerosis." Obstet Gvnecol 89(3):
403-8.
Savvidou, M. D., P. J. Vallance, et al. (2001). "Endothelial nitric oxide synthase gene
polymorphism and maternal vascular adaptation to pregnancy." Hypertension
38(6): 1289-93.
Schmidt, L. G., T. Sander, et al. (2000). "Different allele distribution o f a regulatory
M AO A gene promoter polymorphism in antisocial and anxious-depressive
alcoholics." J Neural Transm 107(6): 681-9.
Schreiber, J., E. Riethmacher-Sonnenberg, et al. (2000). "Placental failure in mice
lacking the mammalian hom olog o f glial cells missing, GCMa." M ol Cell
Biol 20(7): 2466-74.
Schuckit, M. A., C. Mazzanti, et al. (1999). "Selective genotyping for the role o f 5-
HT2A, 5-HT2C, and G ABA alpha 6 receptors and the serotonin transporter in
the level o f response to alcohol: a pilot study." Biol Psychiatry 45(5): 647-51.
Scott, I. C., L. Anson-Cartwright, et al. (2000). "The HAND1 basic helix-loop-helix
transcription factor regulates trophoblast differentiation via multiple
mechanisms." M ol Cell Biol 20(2): 530-41.
Seligman, S. P., T. Nishiwaki, et al. (1997). "Hypoxia stimulates ecNOS mRNA
expression by differentiated human trophoblasts." Ann N Y Acad Sci 828:
180-7.
216
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Seow, A., C. Y. Shi, et al. (1998). "Isoflavonoid levels in spot urine are associated
with frequency o f dietary soy intake in a population-based sample o f middle-
aged and older Chinese in Singapore." Cancer Epidemiol Biomarkers Prev
7(2): 135-40.
Services, U. D. o. H. a. H. (1988). The health consequences o f smoking: nicotine
addiction. A report o f the Surgeon General, 1988. Rockville, M D, Public
Health Service, Centers for Disease Control, Office on Smoking and Health.
Shah, D. M., J. M. Banu, et al. (2000). "Reproductive tissue renin gene expression in
preeclampsia." Hvpertens Pregnancy 19(3): 341-51.
Shanklin, D. R. and B. M. Sibai (1990). "Ultrastructural aspects o f preeclampsia. II.
Mitochondrial changes." Am J Obstet Gynecol 163(3): 943-53.
Shields, P. G., C. Lerman, et al. (1998). "Dopamine D 4 receptors and the risk o f
cigarette smoking in African-Americans and Caucasians." Cancer Epidemiol
Biomarkers Prev 7(6): 453-8.
Shih, J. C., K. Chen, et al. (1999). "Monoamine oxidase: from genes to behavior."
Annu Rev Neurosci 22: 197-217.
Shih, J. C., J. Grimsby, et al. (1993). "Structure and promoter organization o f the
human monoamine oxidase A and B genes." J Psychiatry Neurosci 18(1): 25-
32.
Silaste, M. L., M. Rantala, et al. (2001). "Polymorphisms o f key enzymes in
hom ocysteine metabolism affect diet responsiveness o f plasma homocysteine
in healthy women." J Nutr 131(10): 2643-7.
Sinsheimer, J. S., C. G. Palmer, et al. (2003). "Detecting genotype combinations that
increase risk for disease: Maternal-Fetal genotype incompatibility test."
Genet Epidemiol 24(T): 1-13.
Sivasubramaniam, S. D., C. C. Finch, et al. (2002). "Monoamine oxidase expression
and activity in human placentae from pre-eclamptic and normotensive
pregnancies." Placenta 23(2-3): 163-71.
Sladek, S. M ., R. R. M agness, et al. (1997). "Nitric oxide and pregnancy." Am J
Phvsiol 272(2 Pt 2): R441-63.
Smarason, A. K., I. L. Sargent, et al. (1993). "The effect o f placental
syncytiotrophoblast microvillous membranes from normal and pre-eclamptic
wom en on the growth o f endothelial cells in vitro." Br J Obstet Gynaecol
100(10): 943-9.
217
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Sohda, S., T. Arinami, et al. (1997). "Methylenetetrahydrofolate reductase
polymorphism and pre-eclampsia." Journal o f M edical Genetics 34(6): 525-6.
Spitz, M. R., H. Shi, et al. (1998). "Case-control study o f the D 2 dopamine receptor
gene and smoking status in lung cancer patients." J Natl Cancer Inst 90(5):
358-63.
Staley, J. K., S. Krishnan-Sarin, et al. (2001). "Sex differences in [123I]beta-CIT
SPECT measures o f dopamine and serotonin transporter availability in
healthy smokers and nonsmokers." Synapse 41(4): 275-84.
Stein, C. J. and G. A. Colditz (2004). "Modifiable risk factors for cancer." Br J
Cancer 90(2): 299-303.
Stuber, E , M. Petersen, et al. (1996). "A genomic polymorphism within the tumor
necrosis factor locus influences plasma tumor necrosis factor-alpha
concentrations and outcome o f patients with severe sepsis." Crit Care Med
24(3): 381-4.
Sullivan, P. R, Y. Jiang, et al. (2001). "Association o f the tryptophan hydroxylase
gene with smoking initiation but not progression to nicotine dependence."
Am J M ed Genet 105(5): 479-84.
Summers, K. L. and E. Giacobini (1995). "Effects o f local and repeated systemic
administration o f (-)nicotine on extracellular levels o f acetylcholine,
norepinephrine, dopamine, and serotonin in rat cortex." Neurochem Res
20(6): 753-9.
Sutherland, A ., D. W. Cooper, et al. (1981). "The indicence o f severe pre-eclampsia
amongst mothers and mothers-in-law o f pre-eclamptics and controls." British
Journal o f Obstetrics & Gynaecology 88(8) : 785-91.
Suzuki, Y., Tanemura, M., Suzuki, Y., Murakami, I. (1999). "Is angiotensinogen
gene polymorphism associated with hypertension in pregnancy?"
Hypertension in Pregnancy 18(3): 261-271.
Svensson, P. J. and B. Dahlback (1994). "Resistance to activated protein C as a basis
for venous thrombosis." N Engl J Med 330(8): 517-22.
Takehashi, M., S. Tanaka, et al. (2002). "Association o f monoamine oxidase A gene
polymorphism with Alzheimer's disease and Lewy body variant." Neurosci
Lett 327(2): 79-82.
Talosi, G., E. Endreffy, et al. (2000). "Molecular and genetic aspects o f
preeclampsia: state o f the art." Molecular Genetics & M etabolism 71(4): 565-
72.
218
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Tanaka, M., M. Gertsenstein, et al. (1997). "Mash2 acts cell autonomously in mouse
spongiotrophoblast development." D ev Biol 190(1): 55-65.
Thomas, D. C. and J. S. Witte (2002). "Point: population stratification: a problem for
case-control studies o f candidate-gene associations?" Cancer Epidemiol
Biomarkers Prev 11(6): 505-12.
Thorpe, L. W., K. N. Westlund, et al. (1987). "Immunocytochemical localization o f
monoamine oxidases A and B in human peripheral tissues and brain." J
Histochem Cvtochem 35(1): 23-32.
Tiret, L., O. Poirier, et al. (1999). "The Lysl98A sn polymorphism in the endothelin-
1 gene is associated with blood pressure in overweight people." Hypertension
33(5): 1169-74.
Tivol, E. A., C. Shalish, et al. (1996). "Mutational analysis o f the human M AOA
gene." Am J M ed Genet 67(1): 92-7.
Torbergsen, T., P. Oian, et al. (1989). "Pre-eclampsia— a mitochondrial disease?"
Acta Obstetricia et Gvnecologica Scandinavica 68(21: 145-8.
Treem, W. R., M. E. Shoup, et al. (1996). "Acute fatty liver o f pregnancy, hemolysis,
elevated liver enzymes, and low platelets syndrome, and long chain 3-
hydroxyacyl-coenzyme A dehydrogenase deficiency." Am J Gastroenterol
91(11): 2293-300.
True, W. R., A. C. Heath, et al. (1997). "Genetic and environmental contributions to
smoking." Addiction 92(10V . 1277-87.
Tsai, M. Y., M. Bignell, et al. (1996). "High prevalence o f a mutation in the
cystathionine beta-synthase gene." Am J Hum Genet 59(6): 1262-7.
Tyndale, R. F. (2003). "Genetics o f alcohol and tobacco use in humans." Ann Med
35(2): 94-121.
Tyndale, R. F. and E. M. Sellers (2002). "Genetic variation in CYP2A6-mediated
nicotine metabolism alters smoking behavior." Ther Drug Monit 24(1): 163-
71.
US Food and Drug Administration (21 CFR Part 801, e. a. (1996). "Regulations
restricting sale and distribution o f cigarettes and smokeless tobacco products
to protect children and adolescents; final rule." Federal Register 61: 44395-
618.
van Beck, E. and L. L. Peeters (1998). "Pathogenesis o f preeclampsia: a
comprehensive model." Obstetrical & Gynecological Survey 53(4): 233-9.
219
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
van der Put, N. M. J., Gabreels, F., Stevens, Smeitink J.A.M., Trijbels,
F.J.M., Eskes, T.K.A.B., van den Heuvel L.P., Blom H.J. (1998). "A second
comm on mutation in the methylenetetrahydrofolate reductase gene: an
additional risk factor for neural-tube defects?" American Journal o f Human
Genetics 62: 1044.
Vanyukov, M. M., H. B. M oss, et al. (1995). "Preliminary evidence for an
association o f a dinucleotide repeat polymorphism at the M AO A gene with
early onset alcoholism/substance abuse." Am J Med Genet 60(2): 122-6.
Vasicek, R., G. Meinhardt, et al. (2003). "Expression o f the human Handl gene in
trophoblastic cells is transcriptionally regulated by activating and repressing
specificity protein (Sp)-elements." Gene 302(1-2): 115-27.
Veenstra-VanderWeele, J., G. M. Anderson, et al. (2000). "Pharmacogenetics and
the serotonin system: initial studies and future directions." Eur J Pharmacol
410(2-3): 165-181.
Visser, W., Beckmann, I., Bremer, H. (1994). "Bioactive tumour necrosis factor @ in
pre-eclamptic patients with and without the HELLP syndrome." British
Journal o f Obstetrics & Gynaecology 101: 1081-1082.
Vuorela, P., S. Helske, et al. (2000). "Amniotic fluid-soluble vascular endothelial
growth factor receptor-1 in preeclampsia." Obstet Gynecol 95(3): 353-7.
Vuorinen, K., A. Remes, et al. (1998). "Placental mitochondrial D N A and respiratory
chain enzymes in the etiology o f preeclampsia." Obstet Gvnecol 91(6): 950-
5.
Wacholder, S., N. Rothman, et al. (2000). "Population stratification in epidemiologic
studies o f comm on genetic variants and cancer: quantification o f bias." J Natl
Cancer Inst 921141: 1151-8.
Wacholder, S., N. Rothman, et al. (2002). "Counterpoint: bias from population
stratification is not a major threat to the validity o f conclusions from
epidem iological studies o f common polymorphisms and cancer." Cancer
Epidemiol Biomarkers Prev 11(6): 513-20.
Walker, J. J. (2000). "Pre-eclampsia." The Lancet 356: 1260-65.
Wang, Y. and S. W. Walsh (2001). "Increased superoxide generation is associated
with decreased superoxide dismutase activity and m RNA expression in
placental trophoblast cells in pre-eclampsia." Placenta 22(2-3): 206-12.
Ward, K., A. Hata, et al. (1993). "A molecular variant o f angiotensinogen associated
with preeclampsia." Nature Genetics 4(1): 59-61.
220
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Weinberg, C. R., A. J. W ilcox, et al. (1998). "A log-linear approach to case-parent-
triad data: assessing effects o f disease genes that act either directly or through
maternal effects and that may be subject to parental imprinting." Am J Hum
Genet 62(4): 969-78.
Weiner, C. R (1987). "The role o f serotonin in the genesis o f hypertension in
preeclampsia." Am J Obstet Gynecol 156(4): 885-8.
Westlund, K. N ., R. M. Denney, et al. (1985). "Distinct monoamine oxidase A and B
populations in primate brain." Science 230(4722): 181-3.
Weyler, W. and J. I. Salach (1985). "Purification and properties o f mitochondrial
monoamine oxidase type A from human placenta." J Biol Chem 260(24):
13199-207.
Whigham, K. A., P. W. Howie, et al. (1978). "Abnormal platelet function in pre
eclampsia." British Journal o f Obstetrics & Gynaecology 85(1): 28-32.
Whitfield, J. B., D. Pang, et al. (2000). "Monoamine oxidase: associations with
alcohol dependence, smoking and other measures o f psychopathology."
Psychol M ed 30(2): 443-54.
WHO (1987). The hypertensive disorders o f pregnancy. World Health Organization
Technical Report Series 758. Geneva, World Health Organization.
Widschwendter, M ., H. Schrocksnadel, et al. (1998). "Pre-eclampsia: a disorder o f
placental mitochondria?" Molecular M edicine Today 4(7): 286-91.
Wilcken, B., K. C. Leung, et al. (1993). "Pregnancy and fetal long-chain 3-
hydroxyacyl coenzyme A dehydrogenase deficiency." Lancet 341(8842):
407-8.
Wilson, A. G., Symons, J.A., M cDowell, T.L., di Giovine, F.S., Duff, G.W. (1994).
"Effects o f a tumor necrosis factor alpha (TNF alpha) promoter base
transition on transcriptional activity." British Journal o f Rheumatology
33(Suppl 1): 89.
Wilson, M. L., T. M. Goodwin, et al. (2003). "Molecular epidem iology o f
preeclampsia." Obstet Gynecol Surv 58(1): 39-66.
Wilton, A. N. and D. W. Cooper (1990). "Pre-eclampsia and HLA-DR4 [letter;
comment]." Lancet 336(8710): 323.
Wilton, A. N., Kaye, J.A., Gue, G.L., Brennecke, S.P., Cooper, D.W. (1995). "Is
angiotensinogen a good candidate gene for preeclampsia?" Hypertension in
Pregnancy 14(2): 251-260.
221
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Witkin, S. S., S. Gerber, et al. (2002). "Influence o f interleukin-1 receptor antagonist
gene polymorphism on disease." Clin Infect P is 34(2): 204-9.
Witte, J. S., W. J. Gauderman, et al. (1999). "Asymptotic bias and efficiency in case-
control studies o f candidate genes and gene-environment interactions: basic
family designs." Am J Epidemiol 149(8): 693-705.
Wu, X., K. S. Hudmon, et al. (2000). "D2 dopamine receptor gene polymorphisms
among African-Americans and Mexican-Americans: a lung cancer case-
control study." Cancer Epidemiol Biomarkers Prev 91101: 1021-6.
Yamada, N ., Sohda, S., Hamada, H., Saitoh, M., Hamaguchi, H., Kubo, T. (1997).
"Polymorphism o f the angiotensinogen genes in Japanese patients wiht
pregnancy-induced hypertension P52.41." Acta Obstetrics et Gvnecologica
Scandinavica 76fSupplement 167): 54.
Yang, J. C., et al. (2002). "A randomized double-blind controlled trial o f
bevacizumab (anti-VEGF antibody) demonstrating a prolongation in time to
progression in patients with metastatic renal cancer." ASCQ Meeting
Abstract 21: A15 (Abstr.).
Yelavarthi, K. K., J. L. Fishback, et al. (1991). "Analysis o f HLA-G m RNA in
human placental and extraplacental membrane cells by in situ hybridization."
J Immunol 146(8): 2847-54.
Yoshimoto, Y., T. Sakumoto, et al. (1986). "Histochemical studies on human
placental and chorionic monoamine oxidase." Obstet Gvnecol 67(3): 344-8.
Yoshimura, M., H. Yasue, et al. (1998). "A m issense Glu298Asp variant in the
endothelial nitric oxide synthase gene is associated with coronary spasm in
the Japanese." Hum Genet 103(1): 65-9.
Yoshimura, T., M. Yoshimura, et al. (2000). "Association o f the m issense
Glu298Asp variant o f the endothelial nitric oxide synthase gene with severe
preeclampsia." Journal o f the Society for Gynecologic Investigation 7(4):
238-41.
Zhou, Y., C. H. Damsky, et al. (1997). "Preeclampsia is associated with failure o f
human cytotrophoblasts to mimic a vascular adhesion phenotype. One cause
o f defective endovascular invasion in this syndrome?" J Clin Invest 99(9):
2152-64.
Zhu, Q. S., K. Chen, et al. (1994). "Bidirectional promoter o f human monoamine
oxidase A (MAO A) controlled by transcription factor Spl." J Neurosci
14(12): 7393-403.
222
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Zhu, Q. S., J. Grimsby, et al. (1992). "Promoter organization and activity o f human
monoamine oxidase (MAO) A and B genes." J Neurosci 12(11): 4437-46.
Zuckerman, M. (1993). "P-impulsive sensation seeking and its behavioral,
psychophysiological and biochemical correlates." Neuronsvchobiology 28(1-
2): 30-6.
Zusterzeel, P. L., W. H. Peters, et al. (2001). "A polymorphism in the gene for
microsomal epoxide hydrolase is associated with pre-eclampsia." J Med
Genet 38(4): 234-7.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
LIPPINCOTT- . „ r,r, A , .
WlL.LlyS.M5 & W i l k i n s A P P E N D IX A
A. V ^ itW r i KfcrwNw Cnm|M n>>
June 18, 2004
University of Southern California
USC/Norris Cancer Center
1441 Eastlake Ave., MS44
Los Angles, CA 90089
Melissa Wilson
VIA EMAIL TO: melisslw@usc.edu June 18, 2004
FEE: NONE
RE: Melissa L. Wilson, T. Murphy Goodwin, Vivien Pan, Sue Ann Ingles “Moleculai Bttdemiology of
Preeclampsia”
Obstetrical and Gynecological Survey 2002;58(l):39-66
USE: Thesis
CONDITION OF AGREEMENT
Permission is granted upon the return of this signed agreement to Lippincott Williams & Wiftjns (LWW). Please
sign and date this form and return to:
Lippincott Williams & Wilkins
David O’Brien, Worldwide Copyright Management
351 W Camden Street, 4 North
Baltimore, MD 21201
USA
Permission is granted and is subject to the following conditions:
1) A credit line will be prominently placed and include the journal article author and artfcll litle, journal title, •
volume number, issue number, and the inclusive pages.
2) The requestor warrants that the material shall not be used in any manner, which may befjerogatory to the
title, content, or authors of the material or to LWW.
3) Permission is granted for one time use only as specified in your correspondence. Rigfct^flerein do not apply
to future reproductions, editions, revisions, or other derivative works. :
4) Permission granted is non-exclusive, and is valid throughout the world in the English)!lajfeuage.
5) LWW cannot supply the requestor with the original artwork or a “clean copy”.
6) Permission is valid if the borrowed material is original to a LWW imprint (Lippincott-l*Aen Publishers,
Williams & Wilkins, Lea & Febiger, Harwal, Igaku-Shoin, Rapid Science, Little Bro^vilt Company,
Harper & Row Medical, American Journal of Nursing Co., and Urban & Schwarzentlbrf {English language.)
Requestor accepts: Date:
351 West Camden Street Baltimore, MD 21201
2 2 4
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
PILOT STUDY FOR PREECLAM PSIA/ECLAM PSIA APPENDIX B
BASELINE INTERVIEW - DEMOGRAPHICS
THANKS FOR AGREEING TO PARTICIPATE IN THIS IMPORTANT STUDY OF WOMEN’S HEALTH. IN THIS INTERVIEW,
WE’LL BE DISCUSSING A NUMBER OF TOPICS INCLUDING YOUR LIFESTYLE HABITS AND MEDICAL HISTORY.
PLEASE FEEL FREE TO INTERRUPT AND ASK ABOUT ANYTHING THAT IS NOT CLEAR.
1. Interviewer ID
3. Patient Study ID
2. Time Start
4. Patient Medical Record Number
't- (USC WCH have 7 digits)
5. Patient Date of Birth:
Month
6A. Are you of Hispanic/Latino ethnicity?
Hrs. Min.
□ am □ pm
Day
□Yes
6B. Is the baby’s father of Hispanic/Latino ethnicity? DYes
Year
□No
□No
7A. Race (check all that apply): □American Indian/Alaska Native
□Asian
□ B lack/African- American
□Native Hawaiian/Pacific Islander
□White
□ Other:
7B. Race of the baby’s father (check all that apply): □American Indian/Alaska Native
□Asian
□Black/African-American
□Native Hawaiian/Pacific Islander
□White
□ Other:
8A. Where were you bom?
□ United States
□ Other:_________ 8B. Age moved to U.S.: yrs.
9. Where was your mother bom?
□ United States
□ Other:
10. Where was your father bom?
□ United States
□ Other:
225
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
APPENDIX B
11. Where was your mother’s mother bom? 12. Where was your mother’s father born?
□ United States □ United States
□ Other: □ Other:__________________________
13. Where was your father’s mother bom? 14. Where was your father’s father bom?
□ United States □ United States
□ Other: □ Other:______________________
15. What is your current marital status? □ Never married
(SELECT BEST DESCRIPTION) □ Presently married
□ Marriage-like relationship
16. With whom do you currently live? (SELECT ALL THAT APPLY)
□ Alone
□ With husband or partner
□ With 1 or more children How many?
□ With 1 or more parents How many?
□ With other family How many?
□ With 1 or more friends How many?
□ Group home
□ Widowed
□ Other
□ Divorced/Separated
17. How many people currently live in your household?
□ 1 0 2 □ 3 0 4 □ 5 □ 6 □ 7 □ 8+
18. About how many times have you moved during the past 12 months? Times
19. What is the highest grade or year of school you have completed? Please include trade or vocational
school that was not part of high school.
□ Completed 8th grade
□ 8-11 years (some HS)
□ HS Diploma/Certificate’ 1 '
□ GED*
□ V/T without HS/GED
□ V/TwithHS/GED
□ Some college
□ Associates degree
□ Bachelors (or equivalent)
□ Masters (or equivalent)
□ PhD or EdD
□ MD, DMD, JD, LLB, LLD
□ Other _______________
INTERVIEWER CAN CODE
FOLLOWING PARTICIPANT
RESPONSE, NOT NECESSARY
TO READ ALL POSSIBLE
RESPONSES.
*PROMPT: ANY
VOCATIONAL SCHOOL?
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
20. What is the highest grade your husband/partner/head of household has
trade or vocational school that was not a part of high school.
APPENDIX B
completed? Include
□ 8U l grade □ Associates degree
□ 8-11 years (some HS) □ Bachelors
□ Diploma/Certificate □ Masters
□ GED □ PhD or EdD
□ V/T no HS/GED □ MD, DMD, JD, LLB, LLD
□ V/T with HS/GED □ Other
□ Some college
21. What is your current employment status? Select all applicable categories.
O Part time □ Homemaker -* □ Full time 0
□ Student -» □ Full time □
□ Employed -» □ Full time □
□ Temporary medical leave
□ Permanent disability
□ Unemployed
□ L o o k i n g f o r w o r k
□ Other ______
22. Current/previous occupation. SEE/SHOW JOB CARD.
□ Homemaker □ Student
□ Manage/professional □ Technical
□ Never employed □ Other_____
□ Service
23. Other than yourself, what is the current employment status of the
person in your household who is the top breadwinner? Select all
applicable categories.
□ Homemaker -> □ Full time □ Part time
□ Student - > □ Full time □ Part time
0 Employed — > □ Full time □ Part time
□ Temporary medical leave
□ Permanent disability
□ Unemployed
□ Looking for work
□ Other__________________________
□ Lives alone
IF REPORTED MARRIED
OR IN MARRIAGE-LIKE
RELATIONSHIP, ASK
ABOUT HUSBAND/
PARTNER.
IF REPORTED SINGLE &
LIVING WITH PARENTS,
ASK ABOUT HEAD OF
HOUSEHOLD (MOTHER/
FATHER).
SHOW JOB CARD &
PROMPT: WHICH JOB
BEST DESCRIBES YOU?
IF SUBJECT NOT
CURRENTLY WORKING,
ASK WHAT TYPE OF JOB
DID THEY DO OR WHAT
JOB HELD FOR LONGEST.
IF SUBJECT HOMEMAKER
OR STUDENT & WORKS
PART TIME, SELECT BOTH.
227
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
A P P E N D IX B
24. Other than yourself, which of the descriptions best describes the job
of the person in your household who is the top breadwinner?
□ Homemaker □ Operators
□ Manage/professional □ Student
□ Technical □ Never employed
□ Service □ Other
□ Lives alone
SHOW JOB CARD &
PROMPT: WHAT
TYPE OF JOB DID
THIS PERSON DO, OR
WHAT JOB DID THIS
PERSON HOLD THE
LONGEST?
INTERVIEWER MAY
CLASSIFY THE JOB IF
GIVEN JOB TITLE.
25. What was your approximate total combined household income last year (in thousands o f US
dollars)? Point to one that is the best guess. Include money from jobs, unemployment payments,
public assistance, or any other money income received before taxes.
□ Less than 10 □ 100 to less than 150
□ 10 to less than 20 □ 150 or more
□ 20 to less than 35 □ Don’t know
□ 35 to less than 50 □ Refused
□ 50 to less than 75
□ 75 to less than 100
PROMPT: THIS
INFORMATION IS
IMPORTANT FOR
DESCRIBING THE WOMEN IN
THIS STUDY AND IS KEPT
STRICTLY CONFIDENTIAL.
PROMPT: THIS MAY
INCLUDE SOCIAL SECURITY,
RETIREMENT INCOME,
INTEREST, DIVIDENDS, NET
INCOME FROM BUSINESS,
FA R M OR R F.N T
26. Where were you living when the (index baby) was bom? Please provide the street address, city,
state, zip code and country.
A. Street:__________________________________________
City:_______________________________
State:__________________ Zip Code:_________________ Country:_____________ _
B. Did you move at any time during your pregnancy with (index baby)?
0 Yes (COMPLETE CHART BELOW)
□ No (SKIP TO #27)
Address (Street, City, State, Zip Code & Country) M onth & Year
of Move
M onth of
Pregnancy a t
M ove
C.
D.
E.
F.
G.
H.
I.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
INDEX PREGNANCY
A P P E N D IX B
NOW I'D LIKE TO ASK YOU SOME QUESTIONS ABOUT YOUR PREGNANCY WITH (INSERT
CHILD'S NAME). THIS IS CALLED THE INDEX PREGNANCY AND IS THE PREGNANCY THAT
LED TO YOUR IDENTIFICATION AS A STUDY PARTICIPANT.
NOTE: INTERVIEWER SHOULD BE AWARE OF WHICH PREGNANCY IS THE INDEX
PREGNANCY & REFER TO IT BY THE DATE OF BIRTH OR CHILD’S NAME.
27. Delivery date (index pregnancy)
Month Day
28. Did you work during the index pregnancy? □ Yes
Year
□ No (SKIP TO #31)
29. How much did you work?
, □ Less than 10 hours/week
□ 10-20 hours/week
□ 21-30 hours/week
□ 31-40 hours/week
□ More than 40 hours/week
30. For how much of the index pregnancy did you work?
□ Less than 1 month □ 4-6 months
□ 1-3 months □ 7-9 months
31. During the index pregnancy, how many weeks pregnant were you when you first saw the OB/
GYN?
weeks
32. Before becoming pregnant with the index pregnancy, how long were you with the father o f the
baby?
months years
3 3 . Did you drink caffmated coffee, tea or sodas during the index pregnancy?
□ Yes
□ No ( S K I P T O # 3 5 )
3 4 . How often did you drink caffmated beverages (coffee, tea or soda) during the index pregnancy?
□ Never
□ Less than once per month
□ 1 -3 drinks per month
□ 1 -3 drinks per week
□ 4-6 drinks per week
□ 1-3 drinks per day
□ 4 or more drinks per day
229
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
R E P R O D U C T I V E H I S T O R Y
A P P E N D IX B
NOW I’D LIKE TO ASK YOU ABOUT YOUR ENTIRE PREGNANCY HISTORY, INCLUDING THOSE THAT DID
NOT GO TO TERM.
3 5 . A g e a t f i r s t m e n s t r u a l p e r i o d ? Y r s . ( R E C O R D Y O U N G E S T A G E N O T E D )
3 6 . N u m b e r o f p r i o r p r e g n a n c i e s , e x c l u d i n g i n d e x p r e g n a n c y ?
3 7A . D a te f i r s t p r e g n a n c y e n d e d
M o n t h
3 7 B . W h a t w a s t h e o u t c o m e o f t h e f i r s t p r e g n a n c y ?
□ M i s c a r r i a g e □ S till b i r t h
□ I n d u c e d a b o r t i o n □ L iv e b i r t h
□ T u b a l o r e c t o p i c □ M o l a r P r e g n a n c y
□ O t h e r
3 7 C . A t w h a t w e e k d i d t h e f i r s t p r e g n a n c y e n d ?
3 7 D . W a s th is th e i n d e x p r e g n a n c y ?
Y e a r
NOTE: QUESTION 37-
40D ARE NOT TO BE
ASKED OF THE
PARTICIPANT BUT
RATHER NOTED BY
THE INTERVIEWER
□ Y e s
( R E C O R D G R E A T E S T # )
□ N o
3 8A . D a t e s e c o n d p r e g n a n c y e n d e d
M o n th Y e a r
3 8 B . W h a t w a s t h e o u tc o m e o f t h e s e c o n d p r e g n a n c y ?
□ M i s c a r r i a g e □ S till b i r t h
□ I n d u c e d a b o r t i o n □ L iv e b i r t h
□ T u b a l o r e c t o p i c □ M o l a r P r e g n a n c y
□ O th e r
3 8 C . A t w h a t w e e k d i d t h e s e c o n d p r e g n a n c y e n d ?
3 8 D . W a s t h i s th e i n d e x p r e g n a n c y ? □ Y e s
( R E C O R D G R E A T E S T # )
□ N o
230
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
A P P E N D IX B
3 9 A . D a te t h i r d p r e g n a n c y e n d e d
M o n t h Y e a r
3 9 B . W h a t w a s th e o u tc o m e o f th e t h i r d p r e g n a n c y ?
□ M i s c a r r i a g e □ S till b i r t h
□ I n d u c e d a b o r t i o n □ L iv e b i r t h
□ T u b a l o r e c to p ic □ M o l a r p r e g n a n c y
□ O t h e r _ _ _ _ _ _ _ _ _ _ _ _ _ _
3 9 C . A t w h a t w e e k d i d t h e t h i r d p r e g n a n c y e n d ? _ _ _ _ _ _ _ _
3 9 D . W a s th is t h e in d e x p r e g n a n c y ? □ Y e s □ N o
4 0 A . D a te f o u r t h p r e g n a n c y e n d e d _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _
M o n t h Y e a r
4 0 B . W h a t w a s t h e o u t c o m e o f th e f o u r t h p r e g n a n c y ?
□ M i s c a r r i a g e □ S t i l l b i r t h
□ I n d u c e d a b o r t i o n □ L iv e b i r t h
□ T u b a l o r e c t o p i c □ M o l a r p r e g n a n c y
□ O t h e r
4 0 C . A t w h a t w e e k d i d th e f o u r t h p r e g n a n c y e n d ? _ _ _ _ _ _ _
4 0 D . W a s th is t h e i n d e x p r e g n a n c y ? □ Y e s □ N o
U S E C O N T I N U A T I O N S H E E T S F O R A D D I T I O N A L P R E G N A N C I E S
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
SUPPLEMENTS & MEDICATIONS A P P E N D IX B
NEXT I WILL ASK ABOUT SUPPLEMENTS AND MEDICATIONS THAT YOU TOOK DURING THE INDEX
PREGNANCY.
4 1 . D i d y o u ta k e a n y m u l t i v i t a m i n s o r p r e n a t a l v ita m in s at least once/week d u r i n g t h e i n d e x
p r e g n a n c y ?
□ Y e s ( G O T O 4 1 A )
□ N o ( S K I P T O # 4 4 )
4 1 A . D u r i n g w h i c h tr i m e s t e r s o f th e i n d e x p r e g n a n c y d i d y o u ta k e p r e n a t a l v i t a m i n s at least
once/week? ( C H O O S E A L L T H A T A P P L Y )
□ F i r s t t r i m e s t e r ( m o n th s 1 -3 )
□ S e c o n d t r i m e s t e r ( m o n th s 4 - 6 )
□ T h i r d t r i m e s t e r ( m o n th s 7 - 9 )
4 2 . H o w m a n y v i t a m i n t a b l e t s d i d y o u ta k e ?
□ 1 to 3 p e r w e e k □ 2 p e r d a y
□ 4 t o 6 p e r w e e k □ 3 + p e r d a y
□ 1 p e r d a y □ D o n ’t k n o w
4 3 . D i d th e v i t a m i n y o u t o o k during the index pregnancy r e q u i r e a d o c t o r ’s p r e s c r i p t i o n ?
□ Y e s
□ N o
4 4 . During the index pregnancy, d i d y o u ta k e a n y o f t h e f o l l o w i n g in d i v i d u a l s u p p l e m e n t s a t l e a s t
o n c e p e r w e e k ?
4 4 A . C a l c i u m , i n c l u d i n g T U M S □ Y e s
□ N o ( S K I P T O # 4 4 C )
4 4 B . H o w m a n y c a l c i u m t a b l e t s o r T U M S d i d y o u ta k e ?
□ 1 to 3 p e r w e e k □ 2 p e r d a y
□ 4 t o 6 p e r w e e k □ 3 + p e r d a y
□ 1 p e r d a y □ D o n ’t k n o w
4 4 C . F o l i c A c i d □ Y e s
□ N o ( S K I P T O # 4 4 E )
PROMPT: WE ARE
INTERESTED IN WHETHER YOU
TOOK EACH OF THESE BY
THEMSELVES, NOT AS PART OF
A MULTIVITAMIN OR
COMBINATION PREPARATION.
4 4 D . H o w m a n y f o lic a c i d t a b le ts d id y o u ta k e ?
□ 1 t o 3 p e r w e e k □ 2 p e r d a y
□ 4 to 6 p e r w e e k □ 3 + p e r d a y
□ 1 p e r d a y □ D o n ’t k n o w
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
4 4 E . H e p a r i n o r o t h e r b l o o d th i n n e r s □ Y e s A P P E N D IX B
□ N o ( S K I P T O # 4 4 H )
4 4 F . H o w m a n y h e p a r i n o r o t h e r b lo o d th in n e r i n je c tio n s d id y o u ta k e ?
□ 1 to 3 p e r w e e k □ 2 p e r d a y
□ 4 to 6 p e r w e e k □ 3 + p e r d a y
□ 1 p e r d a y □ D o n ’t k n o w
4 4 G . G iv e r e a s o n f o r t a k i n g h e p a r i n o r b l o o d th in n e r :_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _
4 4 H . B a b y a s p i r i n
4 4 1 . H o w o f t e n d i d y o u t a k e b a b y a s p ir in ?
□ N e v e r □ 4 - 6 w e e k
□ < 1 / m o n t h □ 1 p e r d a y
□ 1 -3 m o n t h □ 2 p e r d a y
□ 1 -3 w e e k □ D o n ’t k n o w
4 4 J . G i v e r e a s o n f o r t a k in g b a b y a s p ir in :_ _ _ _ _ _ _
□ Y e s
□ N o ( S K I P T O # 4 4 K )
DO NOT INCLUDE
NON-ASPIRIN
ANALGESICS SUCH
AS ACETAMINOPHEN
OR IBUPROFEN.
4 4 K . N o n - S t e r o i d a l A n t i - I n f l a m a t o r y D r a g s ( N S A I D s ) :
( f o r e x a m p l e , a d u l t a s p i r i n o r ib u p r o f e n )
4 4 L . H o w o f t e n d i d y o u ta k e N S A I D s ?
□ N e v e r □ 4 - 6 w e e k
□ < 1 / m o n t h □ 1 p e r d a y
□ 1 -3 m o n t h □ 2 p e r d a y
□ 1 -3 w e e k □ D o n ’t k n o w
□ Y e s
□ N o ( S K I P T O # 4 5 )
4 5 . H a v e y o u e v e r t a k e n m e d i c a t i o n ( s ) , h e r b s o r te a s f o r d e p r e s s io n ?
□ Y e s
□ N o ( S K I P T O # 4 8 )
4 6 . W h e n d id y o u t a k e m e d i c a t i o n , h e r b s o r te a s f o r d e p r e s s i o n ( C H E C K A L L T H A T A P P L Y ) ?
□ B e f o r e i n d e x p r e g n a n c y
□ D u r i n g i n d e x p r e g n a n c y
□ A f t e r i n d e x p r e g n a n c y
4 7 . P l e a s e l is t a l l m e d i c a t i o n s , h e r b s o r t e a s ta k e n f o r d e p r e s s i o n : _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _
4 8 . H a v e y o u e v e r t a k e n m e d i c a t i o n ( s ) , h e r b s o r te a s f o r a n x ie ty ?
□ Y e s
□ N o ( S K I P T O # 5 1 )
4 9 . W h e n d id y o u t a k e m e d i c a t i o n ( s ) , h e r b s o r te a s f o r a n x i e t y ( C H E C K A L L T H A T A P P L Y ) ?
□ B e f o r e i n d e x p r e g n a n c y □ D u r in g i n d e x p r e g n a n c y □ A f t e r i n d e x p r e g n a n c y
5 0 . P l e a s e lis t a ll m e d i c a t i o n s , h e r b s o r te a s ta k e n f o r a n x ie ty :
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
A P P E N D IX B
5 1 . P le a s e l i s t a ll o th e r m e d i c a t i o n s ta k e n d u r i n g t h e i n d e x p r e g n a n c y , a s w e ll a s t h e f r e q u e n c y a n d
d u r a tio n o f u s e :_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _
W EIGHT
5 2 . W h a t w a s y o u r w e i g h t w h e n y o u w e r e b o m ?
□ < 5 lb s □ A b o v e a v e r a g e
□ 5 to 5 .5 lb s □ B e l o w a v e r a g e
□ > 5 .5 to 7 lb s □ A v e r a g e
□ > 7 t o 8 .5 l b s □ D o n ’t k n o w
□ > 8 .5 to 1 0 lb s
□ > 1 0 lb s
5 3 . W h a t w a s y o u w e i g h t b e f o r e t h e i n d e x p r e g n a n c y ? P o u n d s
5 4 . H o w m a n y p o u n d s d i d y o u g a i n d u r in g t h e e n tir e i n d e x p r e g n a n c y ? P o u n d s
234
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
BIRTH CONTROL A P P E N D IX B
NOW I’LL BE ASKING SOME QUESTIONS ABOUT THE TYPES OF BIRTH CONTROL YOU USED DURING THE SIX
MONTHS PRIOR TO BECOMING PREGNANT WITH THE INDEX PREGNANCY. REMEMBER, EVERYTHING YOU
TELL ME IS CONFIDENTIAL.
5 5 . W h e n w a s the first time you had intercourse w ith t h e m a n w h o f a th e r e d t h e i n d e x p r e g n a n c y ?
M o n t h Y e a r
5 6 . W h e n d id y o u f i r s t h a v e unprotected sex w i t h th e m a n w h o f a th e r e d th e i n d e x p r e g n a n c y ?
PROMPT: BY UNPROTECTED SEX, WE MEAN
VAGINAL INTERCOURSE WITHOUT RUBBERS,
DIAPHRAGM, CERVICAL CAP/SPONGE OR ANY
OTHER TYPE OF BARRIER METHOD OF BIRTH
CONTROL. USING THIS DEFINITION, TAKING
THE PILL WOULD BE CONSIDERED
UNPROTECTED SEX SINCE IT DOES NOT KEEP
THE EGG AND THE SPERM FROM COMING INTO
CONTACT.
M onth Y e a r
5 7 . D u r i n g t h e 6 m o n t h s b e f o r e y o u b e c a m e p r e g n a n t w i t h t h e i n d e x p r e g n a n c y , d i d y o u o r y o u r
s e x u a l p a r t n e r s u s e a n y o f t h e f o l l o w i n g to p r e v e n t o r d e l a y p r e g n a n c y o r t o p r e v e n t p a s s i n g
in f e c tio n ( R E A D A L O U D & S E L E C T A L L T H A T A P P L Y ) ?
□ W it h d r a w a l b e f o r e e ja c u l a t i o n
□ D i a p h r a m
□ C o n d o m u s e d b y t h e m a n
□ C e r v ic a l c a p o r r i n g , f e m a l e c o n d o m
□ S p o n g e
□ B i r t h C o n t r o l P i l l s
□ D e p o - P r o v e r a o r N o r p l a n t
□ S p e r m i c i d e - f o a m
□ O t h e r _ _ _ _ _ _ _ _ _ _ _ _ _ _ _
□ N o n e
5 8 . Is th is th e f i r s t p r e g n a n c y , i n c l u d i n g a n y p r e g n a n c ie s t h a t m a y h a v e m i s c a r r i e d o r b e e n t e r m i n a t e d ,
t h a t y o u h a v e h a d together?
□ Y e s
□ N o
□ D o n ’t k n o w
235
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
PHYSICAL ACTIVITY A P P E N D IX B
NOW I’D LIKE TO ASK YOU ABOUT YOUR PHYSICAL ACTIVITY LEVEL DURING THE YEAR BEFORE YOU
BECAME PREGNANT WITH [INDEX PREGNANCY. PLEASE TRY TO BE AS ACURATE AS POSSIBLE AND
GIVE YOUR BEST GUESS WHERE NECESSARY.
5 9 . D u r i n g t h e y e a r b e f o r e t h e i n d e x p r e g n a n c y , h o w m a n y hoars per day d id y o u s p e n d s i t t i n g a t a
d e s k , w a tc h in g t e l e v i s i o n , r e a d i n g , e a t i n g o r a n y o t h e r a c tiv ity t h a t is d o n e w h i l e s i t t i n g ?
□ L e s s th a n 1 h o u r / d a y □ 8 -1 1 h o u r s / d a y
□ 1 -3 h o u r s / d a y □ 1 2 - 1 6 h o u r s / d a y
□ 4 - 7 h o u r s / d a y □ 1 7 o r m o r e h o u r s / d a y
6 0 . D u r i n g t h e y e a r b e f o r e t h e i n d e x p r e g n a n c y , d i d y o u p l a y s p o r ts o r h a v e a r e g u l a r l e i s u r e t im e
p h y s i c a l a c tiv ity , s u c h a s w a l k i n g , d a n c i n g , j o g g i n g , a e r o b ic s , b a s k e tb a ll, s o f t b a l l , o r o t h e r s p o r ts ?
□ Y e s
□ N o ( S K I P T O # 6 4 )
6 1 . W h i c h a c tiv ity d i d y o u d o m o s t f r e q u e n t l y ?
I n te n s ity : □ L o w ( 0 .7 6 ) □ M o d e r a t e ( 1 .2 6 ) "
BASED ON PARTICIPANT
RESPONSE, INTERVIEWER
CODES INTENSITY LEVEL
OF THAT ACTIVITY.
□ H i g h ( 1 .7 6 )
6 2 . H o w m a n y hours per week d i d y o u e x e r c i s e a t t h a t a c t i v i t y le v e l? P l e a s e g iv e y o u r b e s t e s t i m a t e .
□ < 1 ( 0 . 5 ) □ 1 -2 ( 1 .5 ) □ 2 - 3 ( 2 .5 ) □ 3 -4 ( 3 .5 ) 0 > 4 ( 4 . 5 )
6 3 . H o w m a n y months d u r i n g t h e y e a r b e f o r e t h e i n d e x p r e g n a n c y d id y o u e x e r c is e a t t h a t a c t i v i t y
l e v e l ? . P le a s e g iv e y o u r b e s t e s tim a te .
□ < 1 ( 0 . 0 4 ) □ 1 -3 ( 0 .1 7 ) □ 4 - 6 ( 0 .4 2 ) □ 7 - 9 ( 0 .6 7 ) □ > 9 ( 0 .9 2 )
6 4 . D u r i n g t h e y e a r b e f o r e t h e i n d e x p r e g n a n c y , h o w m a n y minutes d i d y o u w a l k per day to a n d
f r o m w o r k , t h e b u s , s c h o o l a n d s h o p p i n g . P l e a s e g i v e y o u r b e s t e s tim a te . ( R E C O R D A S
M I N U T E S P E R D A Y )
□ < 5 0 5 - 1 5 0 1 6 - 3 0 0 3 1 - 4 5 □ > 4 5
236
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
MEDICAL HISTORY
A P P E N D IX B
NOW I’D LIKE TO ASK ABOUT YOUR MEDICAL HISTORY. HAS A DOCTOR OR NURSE EVER TOLD YOU THAT
YOU HAD:
6 5 . H e a r t d is e a s e
□ N o ( S K I P T O # 6 6 ) - - - - - - - - - - - - - - - - - - - - - - - - - - - -
□ Y e s - » 6 5 A . W h e n w a s th is f ir s t d ia g n o s e d ?
M onth Year
6 5 B . D o y o u t a k e m e d ic a tio n s ( o r a l o r i n j e c t i o n ) f o r y o u r h e a r t d z ?
□ N o
□ Y e s
6 6 . T h y r o i d p r o b l e m
□ N o ( S K I P T O # 6 7 )
□ Y e s - » 6 6 A . W h e n w a s th is f u s t d ia g n o s e d ?
onth Y e a r
6 6 B . D o y o u t a k e m e d i c a t i o n s ( o r a l o r i n j e c t i o n ) f o r y o u r ty r o id ?
□ N o
□ Y e s
6 6 C . S p e c if y : □ O v e r a c tiv e ( h y p e r t h y r o i d i s m )
□ U n d e r a c t i v e ( h y p o t h y r o i d i s m )
□ D o n ’t k n o w
67. B l o o d c lo t o r s tr o k e
□ N o ( S K I P T O # 6 8 ) - j -
□ Y e s - > 67A. W h e n w a s t h i s f i r s t d ia g n o s e d ?
M onth Y e a r
67B. D o y o u t a k e m e d ic a tio n s ( o r a l o r i n j e c t i o n ) f o r t h i s c o n d itio n ?
□ N o
□ Y e s
6 8 . C a n c e r ( s p e c i f y p r i m a r y s ite :_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _
□ N o ( S K I P T O # 6 9 )
□ Y e s - » 6 8 A . W h e n w a s t h i s f u s t d ia g n o s e d ?
Month Year
6 8 B . D o y o u t a k e m e d ic a tio n s ( o r a l o r i n j e c t i o n ) f o r y o u r c a n c e r ?
□ N o
□ Y e s
237
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
6 9 . P r e e c la m p s ia , to x e m ia o r h i g h b l o o d p r e s s u r e d u r in g p r e g n a n c y
□ N o ( S K I P T O # 7 0 )
□ Y e s - > 6 9 A . W h e n w a s th is f ir s t d ia g n o s e d ?
A P P E N D IX B
M ontli W e a r
6 9 B . D i d y o u ta k e m e d i c a t i o n s ( o r a l o r in je c tio n ) f o r y o u r P E ?
□ N o
□ Y e s
6 9 C . A t w h a t w e e k o f y o u r p r e g n a n c y w a s i t d ia g n o s e d ? W k s .
F A M I L Y H I S T O R Y
NOW I’D LIKE TO ASK YOU A FEW QUESTIONS ABOUT YOUR FAMILY MEDICAL HISTORY. HAS
ANYONE IN YOUR FAMILY, THAT IS YOUR GRANDPARENTS, MOTHER, FATHER, SISTER(S),
BROTHER(S) EVER BEEN TOLD BY A DOCTOR OR NURSE THAT THEY HAD:
7 0 . D ia b e te s ( s u g a r in u r i n e o t h e r t h a n d u r i n g p r e g n a n c y )
7 0 A . W h ic h f a m i l y m e m b e r ?
□ G r a n d m o t h e r - > H o w m a n y ?
□ G r a n d f a t h e r —> H o w m a n y ?
□ M o t h e r
□ F a t h e r
□ S is te r s —» H o w m a n y ?
□ B r o t h e r s - > H o w m a n y ?
7 1 . H y p e r te n s io n ( h i g h b l o o d p r e s s u r e )
7 1 A . W h ic h f a m i l y m e m b e r ?
□ G r a n d m o t h e r - » H o w m a n y ?
□ G r a n d f a t h e r — > H o w m a n y ?
□ M o t h e r
□ F a t h e r
□ S is te r s — » H o w m a n y ?
□ B r o t h e r s —> ■ H o w m a n y ?
7 2 . H e a r t D is e a s e
7 2 A . W h ic h f a m i l y m e m b e r ?
□ G r a n d m o t h e r - > H o w m a n y ?
□ G r a n d f a t h e r — » H o w m a n y ?
□ M o t h e r
□ F a t h e r
□ S is te r s - » H o w m a n y ?
□ B r o t h e r s - » H o w m a n y ?
□ N o ( G O T O # 7 1 )
□ Y e s ( G O T O # 7 0 A )
□ N o ( G O T O # 7 2 )
□ Y e s ( G O T O # 7 1 A )
□ N o ( G O T O # 7 3 )
□ Y e s ( G O T O # 7 2 A )
238
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
A P P E N D IX B
7 3 . K id n e y D is e a s e 0 N o ( G O T O # 7 4 )
□ Y e s ( G O T O # 7 3 A )
7 3 A . W h ic h f a m i l y m e m b e r ?
□ G r a n d m o t h e r — > H o w m a n y ?
□ G r a n d f a t h e r H o w m a n y ?
□ M o t h e r
□ F a t h e r
□ S is te r s — > H o w m a n y ?
□ B r o t h e r s — > H o w m a n y ?
7 4 . C a n c e r □ N o ( G O T O # 7 5 )
□ Y e s ( G O T O # 7 4 A )
7 4 A . W h ic h f a m i l y m e m b e r ( S P E C I F I Y P R I M A R Y S I T E O R T Y P E F O R E A C H ) ?
□ G r a n d m o t h e r — >
□ G r a n d f a t h e r ->■
□ M o t h e r
□ F a t h e r
□ S is te r s — >
□ B r o t h e r s — >
H o w m a n y ?
H o w m a n y ?
H o w m a n y ?
H o w m a n y ?
S ite _
S ite _
S ite _
S ite _
S ite _
S ite
7 5 . W a s y o u r g r a n d m o th e r , m o t h e r o r w e r e y o u r s is te r ( s ) e v e r t o l d t h a t th e y h a d t o x e m i a ,
p r e e c l a m p s i a o r h i g h b l o o d p r e s s u r e d u r i n g p r e g n a n c y ?
7 5 A . W h ic h f a m i l y m e m b e r ?
□ G r a n d m o t h e r — >
□ M o t h e r
□ S is te r s - >
7 6 . B l o o d c lo t o r s tr o k e
7 6 A . W h ic h f a m i l y m e m b e r ?
□ G r a n d m o t h e r
□ G r a n d f a t h e r - >
□ M o t h e r
□ F a t h e r
□ S is te r s — >
□ B r o t h e r s - >
H o w m a n y ?
H o w m a n y ?
H o w m a n y ?
H o w m a n y ?
H o w m a n y ?
H o w m a n y ?
□ N o ( G O T O # 7 6 )
□ Y e s ( G O T O # 7 5 A )
□
□
□ N o ( G O T O # 7 7 )
□ Y e s ( G O T O # 7 6 A )
239
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
A P P E N D IX B
7 7 . T h y r o id P r o b le m ( c o n ’t o n n e x t p a g e ) □ N o ( G O T O # 7 8 )
□ Y e s ( G O T O # 7 7 A )
7 7 A . W h ic h f a m i l y m e m b e r ?
□ G r a n d m o t h e r —> M a te r n a l — >
P a t e r n a l - »
□ H y p e r
□ H y p e r
□ H y p o
□ H y p o
□ D o n ’t k n o w
□ D o n ’t k n o w
□ G r a n d f a t h e r - » M a t e r n a l — >
P a t e r n a l - »
□ H y p e r
□ H y p e r
□ H y p o
□ H y p o
□ D o n ’t k n o w
□ D o n ’t k n o w
□ M o t h e r - > □ H y p e r □ H y p o □ D o n ’t k n o w
□ F a t h e r - » □ H y p e r □ H y p o □ D o n ’t k n o w
□ S is te r ( s ) — >
(1 )
- > □ H y p e r □ H y p o □ D o n ’t k n o w
(2 ) □ H y p e r □ H y p o □ D o n ’t k n o w
(3)
□ H y p e r □ H y p o □ D o n ’t k n o w
□ B r o t h e r ( s ) — >
(1 )
□ H y p e r □ H y p o □ D o n ’t k n o w
(2 )
□ H y p e r □ H y p o □ D o n ’t k n o w
(3)
- » □ H y p e r □ H y p o □ D o n ’t k n o w
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
T O B A C C O - A C T I V E
A P P E N D IX B
NOW I WANT TO ASK YOU SOME QUESTIONS ABOUT YOUR LIFESTYLE HABITS. ALL RESPONSES WILL
BE KEPT ENTIRELY CONFIDENTIAL.
7 8 . D u r i n g y o u r e n t i r e lif e , h a v e y o u s m o k e d a t l e a s t 1 0 0 c ig a r e tte s ?
□ Y e s
□ N o ( G O T O # 8 1 )
PROMPT: THAT IS ABOUT 5
PACKS OF CIGARETTES (1 PACK =
20 CIGS).
7 9 . D u r i n g t h e i n d e x p r e g n a n c y , d id y o u s m o k e a t a ll? □ Y e s □ N o
8 0 . D u r i n g t h e i n d e x p r e g n a n c y , h o w m a n y c ig a r e tte s d id y o u u s u a l l y s m o k e per dayl
C i g s / d a y
IF SUBJECT WAS NOT A REGULAR SMOKER,
ASK HOW MANY CIGS/WEEK AND DIVIDE BY 7.
THE NUMBER OF CIGS/DAY CAN BE A
FRACTION.
T O B A C C O - P A S S I V E
8 1 . D u r i n g t h e i n d e x p r e g n a n c y , o n a v e r a g e , h o w m a n y h o u r s / w e e k w e r e y o u exposed t o c ig a r e tte ,
c i g a r o r p ip e s m o k e b e c a u s e o f s m o k i n g b y o th e r s ?
h r s / w e e k
8 2 . P r i o r t o i n d e x p r e g n a n c y , d i d t h e f a t h e r o f b a b y s m o k e c ig a r e tte s , c ig a r s o r t o b a c c o i n a p ip e ?
□ N o
□ Y e s A □ C i g a r e t t e s S E L E C T A L L T H A T A P P L Y
□ C ig a r s
□ P i p e ( to b a c c o )
241
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
ALCOHOL A P P E N D IX B
NEXT, I’LL BE ASKING SOME QUESTIONS ABOUT YOUR DRINKING HABITS WHILE YOU WERE PREGNANT
WITH THE INDEX PREGNANCY.
8 3 . I n y o u r e n t i r e l i f e h a v e y o u h a d a t l e a s t 1 0 d r in k s o f a n y k i n d o f a lc o h o l, i n c l u d i n g b e e r , m a l t
liq u o r , w in e , w in e c o o l e r s o r l i q u o r ( N O T I N C L U D I N G W I N E D U R I N G R E L I G I O U S
S E R V I C E S ) ?
□ Y e s
□ N o
□ D o n ’t k n o w
8 4 . D u r i n g t h e i n d e x p r e g n a n c y , d id y o u h a v e a n y a lc o h o l i c b e v e r a g e s ?
□ Y e s
□ N o
□ D o n ’t k n o w
PROMPT: AGAIN, THIS MEANS A DRINK OF BEER, WINE, WINE COOLER OR
LIQUOR. IT DOES NOT INCLUDE WINE DURING RELIGIOUS SERVICE.
8 5 . D u r i n g t h e i n d e x p r e g n a n c y , h o w o f t e n d i d y o u d r i n k a n y k i n d o f a lc o h o l, i n c l u d i n g b e e r , m a l t
li q u o r , w in e , w in e c o o l e r s o r l i q u o r ( N O T I N C L U D I N G W I N E D U R I N G R E L I G I O U S
S E R V I C E S ) ?
□ N e v e r
□ S o m e d a y s
□ D a ily
- > ti m e s p e r □ W e e k
□ M o n t h
□ Y e a r
8 6 . O n d a y s y o u d r a n k a l c o h o l d u r i n g t h e i n d e x p r e g n a n c y , h o w m a n y 1 2 o z c a n s , b o t t l e s o r g la s s e s
o f b e e r , 4 o z . g la s s e s o f w i n e o r 1 o z . s h o ts d id y o u d r i n k ( M I X E D D R I N K S U S U A L L Y H A V E
O N E S H O T P E R D R I N K ) ?
• _ _ _ _ D r in k s
INTERVIEWER SHOULD ADJUST NUMBER IF
PARTICIPANT REPORTS LARGER OR SMALLER
DRINK SIZE. FOR EXAMPLE, 40 OZ. BOTTLE OF
BEER = 3.33 STANDARD DRINKS
242
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
INTERVIEWER ASSESSMENT A P P E N D IX B
8 7 . T im e i n t e r v i e w e n d e d : □ A M □ P M
8 8 . W h o e l s e w a s p r e s e n t i n t h e r o o m d u r in g th e in te r v ie w ?
□ N o o n e , a l o n e w i t h p a r t i c i p a n t ( S K I P T O # 9 0 )
□ M a l e p a r t n e r
□ M o t h e r o r li k e p e r s o n
□ F e m a l e f r i e n d o r r e la tiv e
□ O th e r _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _
"i
8 9 . H o w m u c h d id t h e o t h e r p e r s o n c o n t r ib u te to th e in f o r m a t i o n ?
□ N o n e □ S o m e
□ A little □ A lo t
9 0 . A n y in te r r u p tio n s o r d i s t r a c t i o n s ? □ N o □ Y e s
( D e s c r i b e :_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _
)
9 1 . P a r t i c i p a n t ’s c o o p e r a t i o n d u r in g t h e i n t e r v i e w w a s :
□ V e r y g o o d □ F a i r
□ G o o d □ P o o r
9 2 . T h e o v e r a l l q u a l i t y o f t h i s i n te r v ie w is:
□ H i g h q u a l i t y □ Q u e s tio n a b le
□ G e n e r a l l y r e l i a b l e 0 U n s a t i s f a c t o r y
9 3 . O t h e r c o m m e n ts :
243
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
PHYSICAL MEASUREMENTS A P P E N D IX B
NOW I’D LIKE TO TAKE SOME PHYSICAL MEASUREMENTS. PLEASE TRY TO RELAX AND REMAIN
QUIET DURING THE MEASUREMENTS. I WILL TAKE EACH MEASUREMENT TWO OR THREE TIMES.
1. P a t i e n t S tu d y ID : 2 . E x a m i n e r I D :
3. D a te o f e x a m :
M o n t h D a y
4 . H e ig h t:
( c m )
6 . W a i s t c i r c u m f e r e n c e ( t o n e a r e s t 0 .5 c m ):
7. H i p c i r c u m f e r e n c e ( t o n e a r e s t 0 .5 c m ):
8. B l o o d p r e s s u r e :
9. P u ls e :
C D
5 . W e ig h t:
( lb s )
Y e a r
c m
c m
c m
c m
c m
c m
r n m H g
m m H g
b e a t s / m i n u t e ( 2 ) b e a t s / m i n u t e
WE’RE ALL DONE NOW. IS THERE ANYTHING ELSE YOU WOULD LIKE TO ADD OR ANY QUESTIONS YOU
HAVE FOR ME?
YOUR COOPERATION, TIME AND EFFORT IN THIS STUDY WILL HELP INCREASE THE MEDICAL COMMUNITY’S
UNDERSTANDING AND KNOWLEDGE ABOUT PREECLAMPSIA. SHOULD YOU EVER BE CONTACTED AGAIN TO
PARTICIPATE IN A RESEARCH STUDY, WE HOPE THAT YOU WILL BE AS GENEROUS WITH YOUR TIME AND
INTEREST ON THOSE INVESTIGATIONS AS YOU HAVE BEEN WITH US. YOUR CONTRIBUTION TO THIS STUDY
CANNOT BE REPLACED BY ANYONE ELSE AND WE SINCERELY APPRECIATE YOUR PARTICIPATION. WE (I)
HOPE THIS HAS BEEN A PLEASANT EXPERIENCE FOR YOU.
MY SUPERVISOR AND/OR I MAY CONTACT YOU IN THE NEAR FUTURE TO CLARIFY A QUESTION. THANK
YOU AGAIN COFFER BROCHURE).
244
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
CONTACT INFORMATION A P P E N D IX B
P a t i e n t S tu d y I D :
P a tie n t M e d i c a l R e c o r d N u m b e r :
Y o u r c u r r e n t f u ll le g a l n a m e :_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _
U n d e r w h a t n a m e is y o u r p h o n e n u m b e r lis te d in t h e p h o n e b o o k ? _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _
C u r r e n t A d d r e s s :
C u r r e n t T e l e p h o n e N u m b e r :
A r e y o u p l a n n i n g to m o v e i n th e n e x t 6 m o n th s ?
N o
Y e s — > W h e n a n d w h e r e a r e y o u p l a n n i n g to m o v e ? D a te :_ _ _ _ _ _ _ _ _ _ ( m m /d d /y y )
A d d r e s s :
P le a s e t e l l m e t h e n a m e o f a r e l a t i v e a n d a f r ie n d , n o t l i v in g i n y o u r h o u s e h o ld , w h o w o u l d b e l i k e l y to
k n o w h o w t o c o n t a c t y o u i f w e c a n n o t c o n ta c t y o u d ir e c tly . T h is p e r s o n d o e s n o t n e e d t o l i v e i n t h e L o s
A n g e le s a r e a .
N a m e :
A d d r e s s :
T e le p h o n e : R e la t i o n s h i p :
N a m e :
A d d r e s s :
T e le p h o n e : R e la tio n s h ip :
245
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
APPENDIX C
JOHN ILEKIS Release Date:
(301)435-6895 11/23/2003
ilekisj@ m ail.nih.gov
SUMMARY STATEMENT
( Privileged C o m m u n icatio n )
INGLES, SUE A DPH, DRPH
USC / NORRIS COMPREHEN CANCER CTR
DEPT OF PREVENTIVE MEDICINE
1441 EASTLAKE AVE/ MS44, RM6419
LOS-ANGELES, CA 90089-9176
Application Number: 1 R21 HD046624-01
Review Group: MGN
M am m alian G en etics S tudy Section
Meeting Date: 10/23/2003 RFA/PA: PA03-107
Council: JAN 2004 PCC: PP -Jl
Requested Start: 04/01/2004
Project Title: A Pilot S tu d y of Novel C andidate G en es for P reeclam p sia
SRG Action: Priority S core: 152 Percentile: 7.5 +
Human Subjects: 30-H um an s u b je c ts involved - Certified, no SRG co n c e rn s
Animal Subjects: 10-No live v erteb rate anim als involved for com peting appl.
Gender: 1 A-Both g e n d e rs, scientifically accep tab le
Minority: 1 A-M inorities an d non-m inorities, scientifically accep tab le
Children: 1 A-Both C hildren and A dults, scientifically accep tab le
Clinical R esearch - not NIH-defined P h ase III Trial
Project Direct C o sts E stim ated
Year R eq u ested Total C ost
1 125,000 203,054
2 150,000 243,665
TOTAL 275,000 446,719
ADMINISTRATIVE BUDGET NOTE: The b u d g et sh o w n is th e re q u ested b u d g et and
h a s n o t b een ad ju sted to reflect any reco m m en d atio n s m ade by review ers. If an aw ard
is planned, th e c o s ts will be calculated by Institute g ra n ts m an ag em en t staff b ased on
th e reco m m en d atio n s outlined below in th e COMMITTEE BUDGET
RECOMMENDATIONS section.
BUDGET MODIFICATIONS
246
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
APPENDIX C
MGN 2 1 R21 HD046624-01
INGLES, S
R21HD046624-01 INGLES, S
COMMITTEE BUDGET RECOMMENDATIONS
RESUME AND SUMMARY OF DISCUSSION: This is a well-written R21 pilot application by
Dr. Ingles aiming to identify preeclampsia (PE) susceptibility genes contributed by the
mother as well as by the fetus. The significance of the problem is high, given that PE is the
leading cause of maternal and neonatal death and morbidity in developed and
underdeveloped countries, with a rising rate in the United States. The investigators'
emphasis on the role played by the fetus in maternal-fetal gene-gene interaction in the
genetic predisposition to PE could make an important contribution in the area of PE
research, and is highly appreciated by the Study Section. Another strength of this application
is that the investigators are aggressively studying the genetics of PE in an underserved
Hispanic population and developing a better understanding of the effect of an ethnically
admixed population. Additional strengths of the application include the innovative approach
of the study, the dedication of the research team, and the strong epidemiological, biological
and clinical support available. Although this is ambitious for an R21 proposal and there are
some minor weaknesses noted in the study design, the Study Section has very high
enthusiasm for this project. It is likely that important results will ensue; the research team is
extremely well qualified; and they address a timely and cutting edge issue.
DESCRIPTION (provided by applicant): Preeclampsia (PE), a form of pregnancy-induced
hypertension, is believed to have a genetic component with both the mother and the fetus
contributing to the risk of PE. Although numerous studies have been conducted to
investigate maternal genetic contributions to PE susceptibility, most previous studies have
failed to consider the role of fetal genotype. Our long-term goal, to be pursued in future
studies, is to conduct a large-scale candidate gene study to examine the roles of maternal
and fetal genotypes and maternal-fetal gene-gene (GxG) interaction in PE susceptibility.
Among the candidate genes to be examined, we propose to examine a promising class of
genes that has been previously overlooked: genes involved in vascularization and
development of the placenta. To date, no placentation genes have been included in
candidate gene studies, and very little is known about polymorphism in these genes.
Specifically, we propose a feasibility study to examine several genes involved in placentation
(HAND1, HASH2, GCM1, HIF-1 alpha and TGFbeta3) and angiogenesis (VEGF, sFLT1,
PGF), and one gene involved in utero-placental blood pressure regulation (MAO-A). We
propose to resequence the coding and regulatory regions of these genes to identify all
common variants, determine allele and haplotype frequencies and attempt to generate
preliminary data linking polymorphism in these genes to PE risk. We will accomplish this
aim by conducting a pilot case-control study of women with a history of PE (during their first
pregnancy with a specific partner) and controls with a normotensive pregnancy history,
matched on maternal age, race, gestational age and year/month of delivery. We intend to
identify and recruit 120 case/child pairs and 240 control/child pairs using the delivery logs
from the Los Angeles County/University of Southern California (LAC+USC) Women's and
Children's Hospital. Finally, because the patient population of LAC+USC is predominantly
Hispanic, and because Hispanics are an ethnically admixed population, we will also
determine the extent of population stratification in this population to assess whether
stratification is likely to lead to biased gene-disease associations. While the likely severity of
2 4 7
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
APPENDIX C
such bias has been debated in the epidemiologic literature, there have been no studies to
assess the extent of this problem in Hispanics, one of the populations that is most likely to
be affected.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
APPENDIX C
MGN 3 1 R21 HD046624-01
INGLES, S
CRITIQUE 1:
SIGNIFICANCE: This proposal will look at the possible role that the developing fetus plays in
generating genetic predisposition to preeclampsia. Since preeclampsia is a major cause of
morbidity and mortality of mothers and infants in both developed and underdeveloped
countries, additional insights into its causal nature would be of substantial importance. In
addition, this study also holds the prospect for focusing on a Hispanic population and a
developing a better understanding of the effect of admixture on case control studies which
would also be of significant utility.
APPROACH: This is a well-written grant from a group that has assembled substantial
expertise in issues related to preeclampsia. The fundamental hypothesis that they are
addressing as to whether the fetus plays a role in genetic predisposition to preeclampsia is
an important one and as they state has been if not ignored, at least less studied than work
that has focused on the mother's genetic background. This is a good strength of this
proposal as is the strong epidemiologic and biostatistical support, the ability to select
relevant candidate genes for placental development and the involvement of clinicians to help
confirm the diagnosis. Strengths also include the desire to generate SNPs by resequencing
risk populations and using these in haplotype constructions and the overall analytical
approaches.
There are several areas of concern. First, buccal swabs are justified for collection from
infants and they indeed work but given that the cases will be identified from hospital records,
it may well be possible to obtain cord blood which is stored in many delivery systems for up
to a week to ten days after the baby's delivery before it is discarded. Cord blood could
provide both plasma and large quantities of DNA and obtaining it should at least be
considered. A parallel concern is that there has been no attempt to collect blood or other
biological material that could be used for DNA analysis from the fathers. This group has
extensive expertise in carrying out the case-control study as they outline and they also
discuss many of the competing and complementary methodologies such as Weinberg's log
linear approaches that can make use of paternal as well as maternal samples. Since they
will have close point of contact to patients surrounding delivery, they may well regret not
collecting father samples which could be of particular use in a strategy looking at the role of
the fetus and for example, using transmission distortion of alleles coming from the father's
side. Although there will be sequencing of risk-populations, there might be some additional
consideration to sequencing actual DNA samples from individuals who have experienced
preeclampsia, which would future enrich the possibility if they identify relevant and maybe
functional variants. The overall plan to sequence these genes for both coding and regulatory
regions is extremely ambitious and will likely not be possible with the scope of the budget
that is outlined in this particular proposal. A number of these genes are quite complicated
and have many exons and since transgenic animal models with knockouts and
heterozygotes have for almost all of these genes failed to identify anything resembling
preeclampsia in the mouse, it would seem quite likely that it will be regulatory mutations that
are of most interest. Thus, a better defined plan for identifying these which may be found in
some cases hundreds of kilo bases away from the coding sequence proper and how it might
actually be a more effective strategy to use haplotype blocks to try and serve as surrogates
249
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
APPENDIX C
for likely functional polymorphisms would seem to be a much more efficient and cost
effective plan. The Hapmap project will be generating data within the time frame of this
project and a better focus on studying haplotypes rather than all individual polymorphisms
should be considered. This, in addition, will also generate a substantial expense for a
program of this type. The interview instrument also appears to overlook a few areas that
MGN 4 1 R21 HD046624-01
INGLES, S
might benefit the study as well. For example, there do not appear to be any questions asking
the mother about whether her previous pregnancies had preeclampsia or not and also
whether there is any family history of preeclampsia. There are also almost no questions on
infant outcome and even minimal data such as birth weight, APGAR scores, medical
complications following delivery and so on might be of future use in identifying these
components of risk. Since much of the focus of the study is on the placenta and the fetus,
identification of risk factors that may only be observable to outcome of the infant would be a
useful consideration. There are many limitations and modifications in the way in which the
study might be optimized. The group is nonetheless extremely well qualified to carry these
out and using a pilot study of this type to demonstrate the ability to ascertain cases
identifying characterized polymorphisms and look at population admixture and stratification
is a worthy set of preliminary goals for a very important problem.
INNOVATION: There are several aspects to this study that would score high points for
innovation, including the goal to focus on fetal and placental effects as well as the efforts to
better study population admixture and stratification with a special focus on Hispanic
populations.
INVESTIGATOR: Dr. Ingles and her team are very well suited to carry out this study with a
wide range of competencies including epidemiologic, biostatistical, molecular and clinical
being available to the team as a whole.
ENVIRONMENT: The environment including access to clinical case material, laboratory, and
statistical support is first rate throughout.
OVERALL EVALUATION: As an R21 proposal this is very ambitious but the group is very
well positioned, has considered the approach carefully and will almost certainly make
substantial progress on an important question.
PROTECTION OF HUMAN SUBJECTS FROM RESEARCH RISK: Adequate plans are in
place for protection of research risks including attention to the role that HIPAA regulations
will play. Some concern is raised about the use of Johnson & Johnson materials to provide a
form of reimbursement to the families as this might be interpreted as an endorsement of
Johnson & Johnson products and this should be revisited by the IRB.
INCLUSION OF WOMEN PLAN: Since preeclampsia occurs in women, the basic focus of
this is on women, and thus it would be rated G2.
INCLUSION OF MINORITIES PLAN: Minorities with a special emphasis on Latinos will be
included and so it would be rated M1.
INCLUSION OF CHILDREN PLAN: As children are being under age 21 by the NIH, and this
population will include 18 year olds and above. Some children will be included but since it
250
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
APPENDIX C
involves a disease of pregnancy, it is appropriate that numbers of children are relatively
minimal.
BUDGET: The enormous plan undertaken in this proposal is unlikely to be achieved with the
budget being requested and if possible some increase in budget in both years available
under the R21 mechanism should be considered.
MGN 5 1 R21 HD046624-01
INGLES, S
CRITIQUE 2:
SIGNIFICANCE: Preeclampsia (PE) is a pregnancy induced form of hypertension where
both genes and environmental risk factors are thought to be important in the etiology and
severity of the disease. PE is estimated to affect 6% of all pregnancies and accounts for 15-
20% of all maternal mortality. There is strong evidence of familial aggregation of risk to PE,
although other risk factors such as parity, obesity, smoking, etc. also influence risk. A key
complication in identifying causal genes for PE is the necessity of separating the effects of
the mother's genotype from that of the developing fetus. Constraints on the occurrence of
PE (i.e. the phenotype is limited to pregnant females) make standard family studies difficult
and the choice of a case-control design is quite reasonable. Also, the proposal notes that
this group of investigators plans to conduct a larger multi-center study of PE in the future,
where more candidate genes will be examined, but the current proposal is limited to
recruiting from two hospitals and will focus on 9 specific genes.
APPROACH: This proposal for an R21 pilot grant describes how single nucleotide
polymorphic (SNP) markers in 9 candidate genes will be tested for association with risk of
PE using a case-control design. The proposal to conduct a pilot study by collecting 120 PE
cases and 240 controls from two large hospitals in Los Angeles (Los Angeles County and
USC Women's & Children's Hospital) seems rather ambitious for an R21, but the high
prevalence of PE (7.2% of 8,212 deliveries over a 4 year period) in this clinic based
population makes it seem feasible. The cases will be matched for age (within 5 years),
gestational age at case diagnosis (within 2 weeks), date of delivery (within a month), plus
race/ethnic group. Both cases and controls will be interviewed to collect information on risk
factors and DNA will be collected from both mothers and infants (the former via venipuncture
and the latter by buccal brush). Both the ascertainment scheme and matched case-control
analysis proposed here are solid, and should provide adequate statistical power to a
doubling of risk for PE (OR=2.0) when the 'high risk' genotype is very common (20%) of the
population. For less common alleles, however, this sample size will only provide adequate
power to detect much larger effects. Since this is an R21 grant, however, the limited power
should not be viewed as a major weakness. Specific aim 3b is designed to estimate the
genetic admixture among both cases and controls, who are expected to be 85% Hispanic. A
set of 50 microsatellite markers selected for differing widely among the 'continental races'
(Europe, Africa and Asia/ Americas) will be genotyped and used to identify genetically
distinct sub-groups among Hispanic cases and controls using the program Structure. Since
the Hispanic population is expected to show admixture of European and Amerindian genes,
this is a reasonable analysis.
251
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
APPENDIX C
INNOVATION: The 9 candidate genes to be tested here have not been previously
considered to play a causal role in the etiology of PE, and this may be the most innovative
component of this proposal. The reliance on whole genome amplification to generate
sufficient DNA from buccal brush samples obtained from infants is another innovative aspect
to this proposal. If this were a full scale R01, preliminary data on the validity of SNP
genotyping from these amplification products might be deemed necessary, but not for an
R21 proposal. The study design is conventional and very appropriate for this complex
disease.
INVESTIGATOR: Dr. Sue Ann Ingles, P.I. (7% effort; 7% FTE), is an Assistant Professor of
Preventive Medicine, Keck School of Medicine at USC. Dr. Ingles has worked largely in
cancer epidemiology in the past, but is well qualified to direct this study. Dr. T. Murphy
Goodwin, M.D. (3% effort; 3% FTE) is a Professor in the Department of OB-GYN at USC,
and is chief of the Maternal-Fetal Medicine Division.
MGN 6 1 R21 HD046624-01
INGLES, S
Dr. Jean C. Shih, Ph.D. (3% effort; 3% FTE) is a Professor of Molecular Pharmacology and
Toxicology at USC. Dr. Shih will oversee some of the genotyping (the MAO genes), but it is
not clear who will be doing the bulk of the genotyping. Dr. David Conti, Ph.D. (5% effort; 5%
FTE) is an Assistant Professor of Biostatistics at USC, and will be supervising the analysis to
detect population structure.
ENVIRONMENT: The research environment is excellent, but additional details about where
the genotyping will be done would have helped. Since this is an R21, this is not a
requirement..
OVERALL EVALUATION: This proposal is somewhat ambitious in terms of the number of
cases and controls to be recruited in only 2 years, but the design of this study is solid and an
innovative list of candidate genes is presented. The use of both maternal and child
genotypes is an important plus. This proposal, if successfully implemented, would generate
valuable data and resources for a larger study of this complex disease.
PROTECTION OF HUMAN SUBJECTS FROM RESEARCH RISK: Adequate
INCLUSION OF WOMEN PLAN: G1 A: Both genders of infants are included, but all the
mothers are females. Since this is a disease of pregnancy, this is acceptable.
INCLUSION OF MINORITIES PLAN: M1 A: All available racial and ethnic minorities will be
included, although the bulk of the patient population in these 2 hospitals is Hispanic (largely
Mexican American). The study design carefully accommodates this fact.
INCLUSION OF CHILDREN PLAN: C1 A: Infants are included in this study, and the effect of
their genotypes is deliberately included in the study design.
BUDGET: Fixed.
THE FOLLOWING RESUME SECTIONS WERE PREPARED BY THE SCIENTIFIC
REVIEW ADMINISTRATOR TO SUMMARIZE THE OUTCOME OF DISCUSSIONS OF
THE REVIEW COMMITTEE ON THE FOLLOWING ISSUES:
252
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
APPENDIX C
PROTECTION OF HUMAN SUBJECTS (R esum e): ACCEPTABLE. There are acceptable
risks and adequate protections for this project.
INCLUSION OF WOMEN PLAN (R esum e): ACCEPTABLE. Since preeclampsia occurs in
women, the basic focus of this is on women. However, both genders of infants will be
included in this study.
INCLUSION OF MINORITIES PLAN (R esum e): ACCEPTABLE. Minorities, with a special
emphasis on Latinos, will be included.
INCLUSION OF CHILDREN PLAN (R esum e): ACCEPTABLE. The mothers will be ages
18-40 years old. In addition, infants born to the index pregnancies will be included.
COMMITTEE BUDGET RECOMMENDATIONS: T he following c h a n g e s w ere
recom m ended: In order to ensure that the investigator has sufficient resources to carry out
the large amount of work proposed here, the Study Section recommends that the budget be
increased to the maximum permitted under an R21. _______________________________ __
+ Derived from the range of percentile values calculated for the study section that reviewed
this application.
NOTICE: The NIH has modified its policy regarding the receipt of amended applications.
Detailed information can be found by accessing the following URL address:
http://grants.nih.gov/grants/policy/amendedapps.htm
NIH announced implementation of Modular Research Grants in the December 18, 1998
issue of the NIH Guide to Grants and Contracts. The main feature of this concept is that
grant applications (R01, R03, R21, R15) will request direct costs in $25,000 modules,
without budget detail for individual categories. Further information can be obtained from the
Modular Grants Web site at http://grants.nih.gov/grants/funding/modular/modular.htm
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
APPENDIX C
MEETING ROSTER
M ammalian G enetics S tudy S ection
G enetic S cien ce s Integrated Review G roup
CENTER FOR SCIENTIFIC REVIEW
MGN
O ctober 23, 2003 - O cto b er 24, 2003
CHAIRPERSON
MURRAY, JEFFREY C „ MD
PROFESSOR
DEPARTMENT OF PEDIATRICS
COLLEGE OF MEDICINE
UNIVERSITY OF IOWA
IOWA CITY, IA 52242
MEMBERS
BACHMANOV, ALEXANDER A., PHD *
ASSOCIATE MEMBER
MONELL CHEMICAL SENSES CENTER
PHILADELPHIA, PA 19104
BARTOLOMEI, MARISA S., PHD *
ASSOCIATE PROFESSOR
DEPARTMENT OF CELL
AND DEVELOPMENTAL BIOLOGY
SCHOOL OF MEDICINE
UNIVERSITY OF PENNSYLVANIA
PHILADELPHIA, PA 161046148
BEATY, TERRI H„ PHD *
PROFESSOR
DEPARTMENT OF EPIDEMIOLOGY
BLOOMBERG SCHOOL OF PUBLIC HEALTH
JOHNS HOPKINS UNIVERSITY
BALTIMORE, MD 212052103
BRZUSTOWICZ, LINDA M„ MD
ASSOCIATE PROFESSOR
DEPARTMENT OF GENETICS
RUTGERS UNIVERSITY
PISCATAWAY, NJ 08854
CHAILLET, J. RICHARD, MD, PHD *
ASSOCIATE PROFESSOR
DEPARTMENT OF MOLECULAR GENETICS AND
BIOCHEMISTRY
UNIVERSITY OF PITTSBURGH
SCHOOL OF MEDICINE
PITTSBURGH, PA 15213
254
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
APPENDIX C
CRUSIO, WIM E„ PHD*
PROFESSOR
BRUDNICK NEUROPSYCHIATRIC RESEARCH INSTITUTE
UNIVERSITY OF MASSACHUSETTS MEDICAL SCHOOL
WORCESTER, MA 01604
Dl RIENZO, ANNA , PHD *
ASSOCIATE PROFESSOR
DEPARTMENT OF HUMAN GENETICS
UNIVERSITY OF CHICAGO
CHICAGO, IL 60637
DIETZ, HARRY C„ MD
PROFESSOR
DEPARTMENTS OF PEDIATRICS, MEDICINE, MOLECULAR
BIOLOGY AND GENETICS, AND NEUROSURGERY
ASSOCIATE INVESTIGATOR, HHMI
JOHNS HOPKINS UNIVERSITY SCHOOL OF MEDICINE
BALTIMORE, MD 21205
FEINGOLD, ELEANOR , PHD *
ASSOCIATE PROFESSOR
DEPARTMENT OF HUMAN GENETICS
UNIVERSITY OF PITTSBURGH
PITTSBURGH, PA 15261
FOROUD, TATIANA M„ PHD
ASSOCIATE PROFESSOR
DEPARTMENT OF MEDICAL AND MOLECULAR GENETICS
INDIANA UNIVERSITY SCHOOL OF MEDICINE
INDIANAPOLIS, IN 46202
GEJMAN, PABLO V., MD *
PROFESSOR OF PSYCHIATRY
DIRECTOR, CENTER FOR PSYCHIATRIC GENETICS
NORTHWESTERN UNIVERSITY
EVANSTON, IL 60201
GILLIAM, T. CONRAD, PHD
BORNE PROFESSOR OF GENETICS AND DEVELOPMENT
DEPARTMENT OF GENETICS AND DEVELOPMENT
COLLEGE OF PHYSICIANS AND SURGEONS
COLUMBIA UNIVERSITY
NEW YORK, NY 10032
HAMMER, MICHAEL F., PHD *
ASSOCIATE PROFESSOR
DEPARTMENT OF ECOLOGY AND EVOLUTIONARY
BIOLOGY
UNIVERSITY OF ARIZONA
TUCSON, AZ 85721
255
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
APPENDIX C
HERMAN, GAIL E„ MD, PHD
PROFESSOR AND DIRECTOR
DIVISION OF MOLECULAR AND HUMAN GENETICS
DEPARTMENT OF PEDIATRICS
CHILDREN'S RESEARCH INSTITUTE
THE OHIO STATE UNIVERSITY
COLUMBUS, OH 43205
HORWITZ, MARSHALL S., MD, PHD *
ASSOCIATE PROFESSOR
DIVISION OF MEDICAL GENETICS
DEPARTMENT OF MEDICINE
UNIVERSITY OF WASHINGTON SCHOOL OF MEDICINE
SEATTLE, WA 981957720
JACOB, HOWARD J„ PHD *
DIRECTOR, HUMAN AND MOLECULAR GENETICS
CENTER
WARREN P. KNOWLES CHAIR OF GENETICS
PROFESSOR OF PHYSIOLOGY
MEDICAL COLLEGE OF WISCONSIN
MILWAUKEE, Wl 53226
JASIN, MARIA , PHD *
MEMBER
CELL BIOLOGY PROGRAM
SLOAN-KETTERING INSTITUTE
NEW YORK, NY 10021
JOYNER, ALEXANDRA L„ PHD
PROFESSOR
SKIRBALL INSTITUTE DEVELOPMENTAL GENETICS
PROGRAM
DEPT. OF CELL BIOLOGY & PHYSIOLOGY &
NEUROSCIENCES
HOWARD HUGHES MEDICAL INSTITUTE
NEW YORK UNIVERSITY SCHOOL OF MEDICINE
NEW YORK, NY 10016
LEDBETTER, DAVID H„ PHD *
ROBERT W. WOODRUFF PROFESSOR OF HUMAN
GENETICS
DIRECTOR, DIVISION OF MEDICAL GENETICS
DEPARTMENT OF HUMAN GENETICS
EMORY UNIVERSITY SCHOOL OF MEDICINE
ATLANTA, GA 30322
LONG, JEFFREY C „ PHD
PROFESSOR
DEPARTMENT OF HUMAN GENETICS
UNIVERSITY OF MICHIGAN
2 5 6
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
APPENDIX C
ANN ARBOR, Ml 48109
MORAES, CARLOS T „ PHD
ASSOCIATE PROFESSOR
DEPARTMENT OF NEUROLOGY
UNIVERSITY OF MIAMI
SCHOOL OF MEDICINE
MIAMI, FL 33136
NEUMAN, ROSALIND J„ PHD *
RESEARCH ASSOCIATE PROFESSOR
DEPARTMENT OF PSYCHIATRY
WASHINGTON UNIVERSITY
SCHOOL OF MEDICINE
ST. LOUIS, MO 63110
OLSON, JANE M„ PHD *
ASSOCIATE PROFESSOR
DEPARTMENT OF EPIDEMIOLOGY AND BIOSTATISTICS
METRO HEALTH MEDICAL CENTER
CASE WESTERN RESERVE UNIVERSITY
CLEVELAND, OH 44109
PFEIFER, KARL , PHD *
INVESTIGATOR
UNIT ON GENOMIC IMPRINTING
LABORATORY OF MAMMALIAN GENES & DEVELOPMENT
NATIONAL INSTITUTE OF CHILD HEALTH AND HUMAN
DEVELOPMENT
NATIONAL INSTITUTES OF HEALTH
BETHESDA, MD 20892
PICKAR, DAVID , MD *
PRESIDENT
GABRIEL PHARMA
CABIN JOHN, MD 20818
PULST, STEFAN M„ MD *
WARSCHAW CHAIR AND DIRECTOR
DIVISION OF NEUROLOGY
PROFESSOR OF MEDICINE
UCLA SCHOOL OF MEDICINE
LOS ANGELES, CA 90048
RIED, THOMAS , MD *
CHIEF, SECTION OF CANCER GENOMICS
CENTER FOR CANCER RESEARCH
NATIONAL CANCER INSTITUTE
NATIONAL INSTITUTES OF HEALTH
BETHESDA, MD 20892
257
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
APPENDIX C
SAVITZ, DAVID A., PHD *
PROFESSOR AND CHAIR
DEPARTMENT OF EPIDEMIOLOGY
SCHOOL OF PUBLIC HEALTH
UNIVERSITY OF NORTH CAROLINA, CHAPEL HILL
CHAPEL HILL, NC 275997435
SIRACUSA, LINDA D„ PHD *
ASSOCIATE PROFESSOR
KIMMEL CANCER CENTER
DEPARTMENT OF MICROBIOLOGY AND IMMUNOLOGY
JEFFERSON MEDICAL COLLEGE
THOMAS JEFFERSON UNIVERSITY
PHILADELPHIA, PA 19107
SPEER, MARCY C., PHD *
ASSOCIATE RESEARCH PROFESSOR
DEPARTMENT OF MEDICINE
DUKE UNIVERSITY MEDICAL CENTER
DURHAM, NC 27710
SCIENTIFIC REVIEW ADMINISTRATOR
CORSARO, CHERYL M„ PHD
GENETIC SCIENCES IRG
CENTER FOR SCIENTIFIC REVIEW
NATIONAL INSTITUTES OF HEALTH
BETHESDA, MD 20892
GRANTS TECHNICAL ASSISTANT
MORRIS, MARCIA B.
GENETIC SCIENCES IRG
CENTER FOR SCIENTIFIC REVIEW
NATIONAL INSTITUTES OF HEALTH
BETHESDA, MD 20892
* Temporary Member. For grant applications, temporary members may participate in the
entire meeting or may review only selected applications as needed.
Consultants are required to absent themselves from the room during the review of any
application if their presence would constitute or appear to constitute a conflict of interest.
258
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.

Linked assets

University of Southern California Dissertations and Theses

Conceptually similar

PDF

Breast cancer in the multiethnic cohort study: Genetic (prolactin pathway genes) and environmental (hormone therapy) factors

PDF

Progression of subclinical atherosclerosis in type 2 diabetes: Therapeutic and metabolic influences

PDF

Risk factors, sequellae and intervention studies for the treatment of diabetes mellitus in Native American populations

PDF

Molecular and genetic epidemiology of Chlamydia trachomatis in the United States

PDF

Insulin-like growth factor 1 genotype, phenotype and breast cancer risk, by racial/ethnic group

PDF

Effect of hormone therapy on the progression of carotid-artery atherosclerosis in postmenopausal women with and without established coronary artery disease

PDF

Lymphedema: Impact on breast cancer survivors

PDF

Single-nucleotide polymorphisms in 17beta-hydroxysteroid dehydrogenase type III and prostate cancer risk

PDF

Dietary fats, fat metabolizing genes, and the risk of breast cancer

PDF

Colorectal cancer risks in Singapore Chinese: Polymorphisms in the insulin-like growth factor-1 and the vitamin D receptor

PDF

The role of genetic variation in insulin -like growth factors in prostate cancer risk: The multiethnic cohort

PDF

Identifying predictors of cytogenetic relapse and death for patients undergoing allogeneic blood or marrow transplantation for treatment of chronic myeloid leukemia in chronic phase: The City of...

PDF

Physical activity and adenomatous polyps: Measures of association and impact

PDF

BRCA1 mutations and polymorphisms in African American women with a family history of breast cancer identified through high throughput sequencing

PDF

The effects of dietary and other lifestyle behaviors on the risk of colorectal cancer among the Mexican -origin Latino population

PDF

Colorectal cancer: genomic variations in insulin-like growth factor-1 and -2

PDF

A case/parental/sibling control study of Ewing's sarcoma/peripheral primitive neuroectodermal tumor (pPNET)

PDF

Validation of serum cotinine as a biomarker of environmental tobacco smoke exposure: Validation with self-report and association with subclinical atherosclerosis in non-smokers

PDF

Agreement between administrative and clinical data for term newborns with congenital malformations

PDF

Infrequent androgen receptor mutations in primary prostate tumors from men residing in Singapore and Los Angeles

Asset Metadata

Creator Wilson, Melissa Lee (author)

Core Title Monoamine oxidase A: (1) As one of many candidate genes for preeclampsia and (2) as a candidate gene for tobacco addiction

Degree Doctor of Philosophy

Degree Program Epidemiology

Publisher University of Southern California (original), University of Southern California. Libraries (digital)

Tag biology, genetics,biology, molecular,health sciences, medicine and surgery,health sciences, pathology,OAI-PMH Harvest

Language English

Contributor Digitized by ProQuest (provenance)

Permanent Link (DOI) https://doi.org/10.25549/usctheses-c16-329800

Unique identifier UC11340019

Identifier 3155498.pdf (filename),usctheses-c16-329800 (legacy record id)

Legacy Identifier 3155498.pdf

Dmrecord 329800

Document Type Dissertation

Rights Wilson, Melissa Lee

Type texts

Source University of Southern California (contributing entity), University of Southern California Dissertations and Theses (collection)

Access Conditions The author retains rights to his/her dissertation, thesis or other graduate work according to U.S. copyright law. Electronic access is being provided by the USC Libraries in agreement with the au...

Repository Name University of Southern California Digital Library

Repository Location USC Digital Library, University of Southern California, University Park Campus, Los Angeles, California 90089, USA