University of California, San Diego

UNIVERSITY OF CALIFORNIA, SAN DIEGO

The Cholinesterases: A Study in Pharmacogenomics

A dissertation submitted in partial satisfaction of the requirements for the degree

Doctor of Philosophy

Biomedical Sciences

Anne Marie Valle

Committee in charge:

Professor Palmer Taylor, Chair Professor Philip Bourne Professor Mark A. Lawson Professor Daniel T. O’Connor Professor Nicholas J. Schork

2008

This Dissertation of Anne Marie Valle is approved, and it is acceptable in quality and form for publication in microfilm:

Chair

University of California, San Diego 2008

iii

To my family: my mother Connie, my husband George, my beautiful children Alethea, Krista, Tammy, Georgie and Carly my wonderful grandchildren Ashley, Kevin, Kristopher, Ezra, Fernando, Diego, Julia, and last but not least Gavin

TABLE OF CONTENTS

Signature Page………………………………..………………………………….……….iii

Dedication………………………………………………………………..………….……iv

Table of Contents………………………………………………………………...... v

List of Abreviations……………………………………………………………………...vii

List of Figures………………………………………………………………………...... xi

List of Tables and Schemes……………………………………………..………………xiii

Acknowledgements…………………………………………………………………...... xv

Vita and Publications…………………………………………………………………..xviii

Abstract of Dissertation……………………………………………………………..…...xx

Chapter I Introduction to the Fields of Pharmacogenetics/Pharmacogenomics, Twin Studies, Cholinergic Control of Cardiovascular Function, and the Cholinesterases……………………………………………………………1 A. Pharmacogenetics/Pharmacogenomics and the Promise of Personalized Medicine..….……………………………………………………………...1 B. Twin Studies….………………………………………………………..9 C. Cholinergic Control of Cardiovascular Function.……………..……..13 D. The Cholinesterases...... 22 E. Dissertation Overview………………………………………………..38

Chapter II Butyrylcholinesterase: Association with the Metabolic Syndrome and Identification of Two Gene Loci Affecting Activity…….………………40 A. Abstract………………………………………………………………40 B. Introduction…………………………………………………………..41 C. Participants and Methods…………………………………………….42 D. Results………………………………………………………………..47 E. Discussion…………………………………………………………….54 F. Acknowledgements…………………………………………………...58

Chapter III A Pharmacogenomic Study: Investigating Naturally Occurring Variations in the Acetylcholinesterase Gene …...... 60 A. Abstract………………………………………………………………60 B. Introduction…………………………………………………………..61 C. Material and Methods………………………………………………...63 D. Results………………………………………………………………..75 E. Discussion…………………………………………………………...115 F. Acknowledgements………………………………………………….118

Chapter IV Summary and Closing Remarks……..……………………………….....119 A. The Butyrylcholinesterase Story……………………………………120 B. The Acetylcholinesterase Story……………………………………..125 C. Closing Remarks……………………………………………………126

References …………………………………………………………………………..127

LIST OF ABBREVIATIONS

2-PAM pyridinium aldoxime

ACh acetylcholine

AChE acetylcholinesterase

ADR adverse drug response

ANOVA analysis of variance

ANS autonomic nervous system

AV atrioventricular

BChE butyrylcholinesterase

BMI body mass index

BP blood pressure

BSA bovine serum albumin cDNA complementary deoxyribonucleic acid

CI confidence interval

CV cardiovascular

ChE cholinesterase

CMV cytomegalovirus

CNS central nervous system

CYP cytochrome P450

CVLM caudal ventrolateral medulla

DMEM Dulbecco’s modified Eagles medium

DNA deoxyribonucleic acid

DTNB 5,5’-dithio-bis(2-nitrobenzoic acid

vii

DZ dizygotic

EPI epinephrine

FBS fetal bovine serum

G6PD glucose-6-phosphate-dehydrogenase gDNA genomic deoxyribonucleic acid

HEK human embryonic kidney

HR heart rate

HTN hypertension

LOD logarithm of odds mAChR muscarinic acetylcholine receptor mRNA messenger ribonucleic acid

MZ monozygotic

NAT2 N-acetyltransferase-2 nAChR nicotinic acetylcholine receptor

NCBI National Center for Biotechnology Information

NE norepinephrine

NTS nucleus of the solitary tract

OMIM Online Mendelian Inheritance in Man

OP organophosphates

PAS peripheral anionic site

PCR polymerase chain reaction

PDB Protein Data Bank

viii

PTC phenylthiocarbamide

QTL quantitative trait loci

RVLM rostral ventrolateral medulla

SA sinoarterial

SD standard deviation

SNP single nucleotide polymorphism

SpDMB (Sp)-3,3dimethylbutyl methylphosphonothiocholine

SOLAR Sequential Oligogenic Linkage Analysis Routines

TDT transmission disequilibrium test

TPMT thiopurine s-metyl transferase

UTR untranslated region

VMC vasomotor center wt wildtype

Amino Acid Residues

Ala or A alanine

Arg or R arginine

Asn or N asparagines

Asp or D aspartate

Cys or C cysteine

Glu or E glutamate

Gln or Q glutamine

Gly or G glycine

His or H histidine

Ile or I isoleucine

Lys or K lysine

Leu or L leucine

Met or M methionine

Phe or F phenylalanine

Pro or P proline

Ser or S serine

Thr or T threonine

Trp or W tryptophan

Tyr or Y tyrosine

Val or V valine

LIST OF FIGURES

Chapter I Figure I.1 Contribution of Twins to the Study of Complex Traits and Diseases…...11

Figure I.2 Sympathetic and Parasympathetic Neurotransmitters…………………...14

Figure I.3 Sympathetic and Parasympathetic Regulation of Cardiac Function……..16

Figure I.4 AChE and BChE Gene Structures………………………………………..23

Figure I.5 hAChE and hBChE Crystal Structures…………………………………..27

Figure I.6 Conformation of ChE Proteins with Bound BW284c51…………………28

Figure I.7 ChEs Activity Curves……………………………………………………29

Figure I.8 AChE Inhibition and Reactivation……………………………………….37

Chapter II Figure II.1 Cumulative Sum of Deviations from Overall Mean……………………..48

Figure II.2 Results of Linkage Analysis for the Detection of Loci Affecting Plasma Cholinesterase…………..………………………………………………..53

Chapter III Figure III.1 Frequency Distribution of Cholinesterase Activity……………………...77

Figure III.2 Linear Regression Graphs of MZ and DZ Twin Pairs…………………...78

Figure III.3 SNP Discovery: SNPs Found in the Unrelated Panel Mapped to the AChE gene………………………………………………………………………80

Figure III.4 cSNPs Modeled on mAChE Crystal Structure…………………………..91

Figure III.5 Comparison of wtT547 and D134H Gene Expression…………………..95

Figure III.6 D134H Temperature Sensitivity………………………………………....95

Figure III.7 pS Curves: A = wtT547 B = R3Q C = D134H D = H322N…………..98

Figure III.8 Comparison of pS Curves for wtT547 AChE and Mutants……………...99

Figure III.9 Exponential Decay Curves from Stability Assays……………………...102

Figure III.10 Human AChE Structure………………………………………………...104

Figure III.11 Compensating Double Mutant (D134H/R136Q)……………………….105

Figure III.12 Protein Expression Normalized to the wtT547…………………………107

Figure III.13 D134H/R136Q pS Curve (Top Panel) and Summary of pS Curves (Bottom Panel)…………………………………………………………………...108

Figure III.14 Half-Life (t50) Parameter Normalized to the wtT547…………………..109

Figure III.15 Paraoxon Inhibition Curves for wtT547 and D134H Proteins………….112

Figure III.16 Oxime-Assisted Reactivation Curves for the wtT547 and D134H Proteins …...……………………………………………………………………...114

Chapter IV Figure IV.1 Ideogram of Genes Located Around GATA12G02…………………….122

xii

LIST OF TABLES AND SCHEMES

Chapter I Table I.1 Key Discoveries of the 1950s……………………………………………..4

Table I.2 Genetic Basis of Variability in Drug Response…………………………...8

Table I.3 Twin Studies and Their Applications……………………………………12

Table I.4 In vivo Studies Implicating a Role for the Cholinergic Control of Cardiovascular Function…………………………………………………20

Chapter II Table II.1 Effects of Method Variation, Sex and Sex-Specific Age Variation on Plasma Cholinesterase Activity…………………………………………..47

Table II.2 Correlation between Plasma Cholinesterase Activity (adjustment for method variation) and Variables Related to Cardiovascular Risk……….50

Table II.3 Pairwise Similarity of Twin Plasma Cholinesterase and Models for Fitting Data………………………………………………………………………51

Chapter III Table III.1 Mutagenesis Primers with Corresponding Template Sequence……….....70

Table III.2 Correlation Analysis of Cholinesterase Enzymatic Activity…………….81

Table III.3 SNP Results: Resequencing of AChE in the Unrelated Panel…………..82

Table III.4 Minor Allelic Frequencies in the Unrelated Panel (Contig NT_007933.13 Reverse Complement)…………………………………………………....84

Table III.5 Haplotype Reconstruction using the Nonsynonymous cSNPs…………..85

Table III.6 Haplotype (HAP) Reconstruction using SNPs 5, 8, 9, 12, 16…………...86

Table III.7A Heritability and Association Study (Complete Twin Panel)…………….88

Table III.7B Heritability and Association Study (Eur-Am Panel)………………….....89

Table III.8 Vector pcDNA3 + hAChE cDNA (7224 bp)…………………………....93

Table III.9 Kinetic Parameters for wtT547 and Mutant Protein Enzymes...... 100

Table III.10 Relative Stability of the Mutant Protein Enzymes……………………...110

xiii

Chapter IV Table IV.1 Links to Identified Genes around GATA126G02 on the UCSC Genome Browser…………………………………………………………………123

LIST OF SCHEMES

Chapter I Scheme I.1 Scheme and Equation I.1………………………………………………...31

Scheme I.2 Scheme and Equation I.2………………………………………………...32

Chapter III Scheme III.1 Scheme and Equation III.1……………………………………………....96

Scheme III.2 Scheme and Equation III.2……………………………………………...111

Scheme III.3 Scheme and Equation III.3……………………………………………...113

xiv

ACKNOWLEDGEMENTS

First and foremost I would like to give my thanks and gratitude to my mentor and

dissertation chair Dr. Palmer Taylor not only for accepting me into his lab but most

importantly, for having the patience and understanding to help me surmount my learning curve and lack of confidence. I feel very fortunate to have worked with one of the most brilliant and hard-working scientist in this era. I am grateful to all my committee members for their advice and support. My project was a collaborative effort and I could not have done it without the assistance and advice of Drs. Phil Bourne, Dan O’Connor,

Nik Schork, and Brinda Rana. I am grateful to Dr. Mark Lawson for agreeing to step in at the last moment when a member of my committee retired. I am also very grateful for all the help and advice I have received from Dr. Kim Barrett, my SPAC advisor, and

Gina Butcher and Leanne Nordeman from the BMS program.

I am profoundly grateful to my second family all the members of the Taylor lab

past and present: Shelley Camp for her help in molecular biology and general lab

management, Joannie K.-Y. Ho for running a million assays and helping me in general,

Jennifer Wilson for always saving me with my last minute ordering, Akos Nemecz and

Dr. Todd Talley for helping me with computer nonsense, Dr. Davide Comoletti especially for not letting me mess up my very first transfection experiment, Dr. Antonella

DeJaco, Pam Tetu, Wenru Yu, Meghan Miller, John Yamuchi, Neveen Barakat, Robin

Flynn, Diana Nguyen, Brian de la Torre, Michael Marquez, Cindy Garcia and Helen

Newlin who all contributed to my successful graduate experience. I have to give special thanks to my friend and swimming partner Limin Zhang who helped me to keep my sanity and for my old graduate partners Dr. Scott Hansen and Dr. Ryan Hibbs. I have left

Dr. Zoran Radić for last because to him I have to give my utmost gratitude. Without his

assistance, patience, and continued support in all aspects of my research, I could never

have accomplished this project, thank you Zoran for always helping me out no matter

how busy you were!

I have to give a very special thanks to Nichol Goodman, Nancy Hurtado, Kathryn

Nguyen, and Melissa Passino my fellow BMSers who picked me up whenever I was at

my lowest, worked with me to climb the highest learning curve ever, kept me focused,

and helped me to enjoy my life here at UCSD. Also, special thanks goes to Swami

Suryadevananda who taught me understanding and Julie Onton, my meditation partner

and friend who helped me to stay sane.

I have left my family for last because there are no words that can adequately

express my love and gratitude for their help, encouragement, continued support, and

unfailing love. My mom, Connie who gave me her unswerving faith and support, my

husband George not only for his support but also for always coming to lift me up and

make me laugh, for my beautiful children Alethea, Krista, Tammy, Georgie and Carly

who gave me the love and will to succeed, and last but not least all eight of my wonderful

grandchildren Ashley, Kevin, Kristopher, Ezra, Fernando, Diego, Julia, and Gavin who helped grandma with their love and beautiful smiles.

The material contained in Chapter II is, in full, from the publication

“Butyrylcholinesterase: Association with the Metabolic Syndrome and Identification of 2

Gene Loci Affecting Activity,” Anne Valle, Daniel T. O’Connor, Palmer Taylor, Gu Zhu,

Grant W. Montgomery, P. Eline Slagboom, Nicholas G. Martin, and John B. Whitfield.

xvi

Clinical Chemistry, 2006, 52(6):1014-1020. The dissertation author was a primary

researcher and the author of the work; the co-authors listed in this publication assisted

with the research which forms the basis for this chapter. Samples were obtained from the

Australian Twin Registry by John B. Whitfield and Nicholas G. Martin. All cholinesterase assays were run in Palmer Taylor’s lab and analysis of data was performed

in the labs of John Whitfield, Palmer Taylor and Daniel T. O’Connor. All results were

analyzed and presented jointly.

xvii

VITA

2008 Ph.D., Biomedical Sciences, University of California, San Diego

2000 B.S., Biology, California State University, Dominguez Hills

AWARDS AND HONORS

2006-present NIH Minority Supplemental Grant Fellow, UC San Diego

2005-2006 Pharmacology Training Grant Fellow, UC San Diego

2004-2005 Genetics Training Grant Fellow, UC San Diego

2003-2004 Pharmacology Training Grant Fellow, UC San Diego

2003-2004 SREB-AGEP Doctoral Scholars Program, UC San Diego

2001-2003 NIH Scholars Program, Cota Robles Fellow, UC San Diego

2000 California State University Forgivable Loan / Doctoral Program

2000 Alumni Association's Outstanding Achievement Award, CSU Dominguez Hills

2000 1st place Biological & Agricultural Sciences Undergraduate Category, CSU Research Competition

2000 Outstanding Undergraduate Research Award, Sigma Xi Research Society

1998-2000 Undergraduate Student Training in Academic Research (USTAR) Scholarship Program, CSU Dominguez Hills

PUBLICATIONS

A.M. Valle, Z. Radić, B.K. Rana, J.B. Whitfield, D.T. O'Connor, N.G. Martin, and P. Taylor. The Cholinesterases: Analysis by Pharmacogenomics in Man. Chemico- Biological Interactions, submitted.

xviii

Valle A, O’Connor DT, Taylor P, Zhu G, Montgomery GW, Slagboom PE, Martin NG, Whitfield JB. Butyrylcholinesterase: Association with the Metabolic Syndrome and Identification of 2 Gene Loci Affecting Activity. Clinical Chemistry 52(6):1014–1020 (2006)

ABSTRACTS

A.M. Valle, Z. Radić, B.K. Rana, J.B. Whitfield, D.T. O'Connor, N.G. Martin, and P. Taylor. The Cholinesterases: Analysis by Pharmacogenomics in Man. IXth International Meeting on Cholinesterases, 2007

Valle AM, Ho KY, Mahboubi V, Barragan MT, Radic Z, Rana BK, O’Connor DT, Taylor P. A Pharmacogenomic Study of Cholinergic Target Proteins in the Control of Cardiovascular Function. The American Society of Human Genetics 56th Annual Meeting, 2006

Valle A, Mahboubi V, Barragan M, Rana BK, O'Connor DT, Taylor P. Polymorphism in cholinergic target proteins in autonomic control of cardiovascular function. PGRN and PharmGKB 4th Scientific Meeting, 2004

*Resta-Lenert SC, Valle AM, Barrett KE. Probiotics Prevent Apoptosis of Intestinal Epithelial Cells Induced by Enteroinvasive Pathogens. Digestive Disease Week Annual Meeting, 2002 *Poster of Distinction

xix

ABSTRACT OF THE DISSERTATION

The Cholinesterases: A Study in Pharmacogenomics

Anne Marie Duran Valle

Doctor of Philosophy in Biomedical Sciences

University of California, San Diego, 2008

Professor Palmer Taylor, Chair

The cholinesterases are serine hydrolases that catalyze the hydrolysis of acetylcholine (ACh), the cholinergic neurotransmitter in the central and autonomic nervous systems and at neuromuscular synapses. Peripheral and central nervous system control of cardiovascular (CV) function mediated through cholinergic pathways is critical in the homeostatic maintenance of blood pressure and responsiveness to stress. The specific role for acetylcholinesterase (AChE; EC 3.1.1.7) lies in its ability to rapidly catalyze the hydrolysis of ACh. Butyrylcholinesterase (BChE; EC 3.1.1.8) catalyzes the hydrolysis of esters of choline including ACh, although it differs in its substrate and inhibitor specificities. The physiological role for BChE remains uncertain, but studies have shown a strong correlation between BChE activity and the metabolic syndrome.

This dissertation investigates the role variations in cholinesterase genes play in relation to cardiovascular function and the metabolic syndrome. AChE and BChE can be found in whole blood enabling a biochemical phenotypic characterization in addition to correlation of genotype with phenotypic physiologic responses. Analysis of enzymatic

activity was determined spectrophotometrically in plasma and blood cells of twin subject

registries from Australia and San Diego along with a general population subject registry

from San Diego. Association studies revealed significant relationships between

cholinesterase activity and certain cardiovascular endpoints

For AChE, I looked at naturally occurring single nucleotide polymorphisms

(SNPs) in the AChE gene in a human population in relation to catalytic properties and cardiovascular function. SNP discovery by re-sequencing of the AChE gene using genomic DNA of the general population registry and SNP genotyping was performed using genomic DNA of the San Diego twin subject registry. Nineteen SNPs have been identified: 7 SNPs in the coding region (cSNPs), 4 non-synonymous encoding for a different amino acid and 3 synonymous encoding the same amino acid; 12 are in untranslated regions (UTR) of the gene with 3 of these in a conserved region of intron 1.

The non-synonymous cSNPs were inserted into a human AChE cDNA vector and transfected into human embryonic kidney (HEK) cells for protein expression.

Characterization of the purified mutant enzymes encoded by the SNP polymorphisms revealed significant thermal and chemical stability differences when compared with the predominant AChE species.

For BChE, I examined whether BChE activity correlated with parameters of the

metabolic syndrome and cardiovascular function. Linkage analysis with data from a

dizygotic twin set showed suggestive linkage at the BChE locus, and statistical analysis

revealed a high correlation between BChE activity and variables associated with

cardiovascular risk and the metabolic syndrome. Statistical analysis revealed a

significant relationship between two BChE SNPs and BChE activity. The pattern of

xxi

within-pair twin correlations by zygosity and the ACE model-fitting findings suggested the major source of variation (65%) in BChE activity was attributable to additive genetic influences.

In summary the research presented here is important in defining the role of cholinesterase SNPs in disease susceptibility particularly in the realm of cardiovascular diseases and the metabolic syndrome, and individual risk associated with chemical terrorism with agents affecting cholinergic function. These enzymes, AChE and BChE, are readily accessible in blood samples however the tissues where activity most likely affects physiologic function are typically not accessible. However, my studies should reveal intrinsic differences in generalized expression parameters. Furthermore, this work underscores the importance of structure-based modeling for analytical and theoretical insights that can be ascertained based on the modeling of SNPs to a protein’s crystal structure.

xxii

CHAPTER I

Introduction to the Field of Pharmacogenetics/Pharmacogenomics: Twin Studies,

Cholinergic Control of Cardiovascular Function, and The Cholinesterases

In this dissertation, I will present a study in pharmacogenomics on individual genetic variations and the effect these polymorphisms have on the cholinesterases.

Specifically, I investigated the role of naturally occurring variations in a human population on the cholinesterase genes and their enzyme products, particularly in response to stress. In this introductory chapter, I hope to familiarize the reader with background information and the relevance of my research, which is especially noteworthy in today’s arena of chemical terrorism aimed at the cholinesterase function.

A. From Pharmacogenetics to Pharmacogenomics and the Promise of

Personalized Medicine

The scienctific fields of pharmacology and genetics arose as two distinct disciplines. While the knowledge of plants with medicinal properties has been around for thousands of years (the ancient Greeks used the word Pharmacon meaning poison or a drug [1], knowledge of the intrinsic active chemicals responsible for their effect(s) were not reported until the 19th century. The science of modern genetics had its beginnings in the 19th century with Gregor Mendel’s report in 1865 on his historical experiments in which he demonstrated the inheritance of traits in pea plants (although at that time it was largely ignored by the scientific community). Francis Galton in

1875 described the concept of using twins that were similar (monozygotic) but raised

1 2 separately and twins that were dissimilar (dizygotic) but raised together to study the inheritance of ability. The marriage of these two disciplines pharmacology and genetics came about through insightful clinical observations of patients with variable plasma or urinary drug concentrations and that variation in concentration had a high incidence of familial inheritance. Around 1898 Sir Archibald Garrod, a perceptive physician interested in urinary pigments, studied patients with alcaptonuria and patients with drug induced porphyria from St. Bartholomew’s Hospital in London. He especially noticed a greater incidence of consanguinity among parents of children with alcaptonuria and in 1902 reported his observations to the medical community [2]. He then went on to develop and report on the concept of “chemical individuality” where he noted “in some subjects a dose that is innocuous to the majority has toxic effects, whereas others show exceptional tolerance of the same drug” [3]. But it was the biologist William Bateson who had popularized Mendel’s work (after its rediscovery by the botanists Hugo de Vries and Carl Correns) that interpreted Garrods’ reports as recessive inheritance. Bateson introduced the term genetics (from the Greek word genno meaning to give birth) in the early 1900s, and along with Reginald Crundall

Punnett (creator of the Punnett Square), discovered genetic linkages. Together they co-founded the Journal of Genetics in 1910.

Insights from these pioneers were the basis for pharmacogenetics but the first actual project that established the prototype for a pharmacogenetic study was done not in comparison of individual differences in drug response, but in the genetic variation of taste. In 1931 A.L. Fox, a Dupont chemist, while searching for a sugar substitute found that some people detected a bitter taste for phenylthiocarbamide (PTC), while

3 others noted only a slight taste or no taste. In a large study of 800 families L.H.

Synder found this taste variation was inherited in an autosomal-recessive manner and the frequency of non-tasters varied according to ethnicity [4].

The 1950s saw the advent of key discoveries (Table I.1 ) by researchers such as Alving and co-workers who found primiquine-induced hemolysis was associated with glucose-6-phosphate-dehydrogenase (G6PD), Bonicke and Reif, and Hughes et al. described differences in isoniazid metabolism, while Kalow and Staron characterized the atypical forms of human serum-cholinesterase in patients with succinylcholine apnea [5-8, 9]. These historic studies helped to acknowledge pharmacogenetics as a new distinct field of science. New advancements in diagnostic techniques enabled more precise measurements of enzyme activity, drug metabolism and drug responses and were vital in the formation of these insightful observations and discoveries. The field of pharmacogenetics as a field of science became firmly established with the seminal paper by Arno Motulsky in 1957 proposing that the underlying cause of individual differences in response to drug administration could be due to genetic traits that were not apparent until exposure to a given drug [10]. In 1959

Friedrich Vogel coined the term ‘pharmacogenetics’ [11].

The first half of the 20th century was important in laying the foundation and basis for pharmacogenetic research, but it wasn’t until the ensuing decades that genetic variations were found to be the underlying cause of differential drug responses.

In the 1960s numerous publications arose on the genetic control of drug responses and in 1967 the first international conference on pharmacogenetics was held at the New

York Academy of Science. This conference brought together investigators from many newly developing areas of research into gene-drug interactions. In addition to G6PD deficiency, some of the topics presented included studies on malignant hyperthermia in response to a general anesthetic, familial dysautonomia, porphyria and enzymatic defects associated with drug-induced hemolysis, hereditary resistance to coumarin anticoagulants, reduced activity of alcohol dehydrogenase in the liver, adverse drug response to acetophenetidin and genetic aspects of allergic reactions to drugs [12].

Many of these studies used monozygotic and dizygotic twins in an attempt to determine the genetic contribution to drug metabolism and drug therapy [13, 14]. The momentous discovery in the 70s by two independent pharmacokinetic studies in which several volunteer participants experienced unexpected adverse drug reactions to both debrisoquine and sparteine sparked the field of pharmacogenetics, attracted international attention and eventually led to the discovery of the cytochrome P450 superfamily. Variations in the cytochrome P450 enzymes have become one of the most studied pharmacogenetic traits with over 2500 publications. The home page of the CYP allele nomenclature (http://www.cypalleles.ki.se/) lists 44 alleles with numerous identified variants including gene deletions and duplications with a high degree of racial heterogeneity within the populations studied [12].

Researchers were then beginning to elucidate the genetic basis for differential drug disposition which began with a focus on drug metabolism and adverse drug response. These studies are far too numerous to cite, but a few important examples are

6 listed in Table I.2. Initially these studies concentrated on clinical and biochemical assessments of individuals and their families on an observed gene-drug phenotype for a monogenetic trait, oftentimes involving drug metabolism enzymes with measurable drug metabolites. With the rapidly emerging science of genomics and advancements in molecular genetics, bioinformatics, and biochemical and molecular techniques, the field has expanded to include complex polygenic (group of genes with additive effect to influence a complex trait) disorders with involvement of polygenes, polygenetic pathways, and gene(s)-environmental interactions leading to the evolution of pharmacogenomics. In 1997 the term “Pharmacogenomics” first arose in the literature

[12] but to date little distinction is given between the two. The terms pharmacogenetics and pharmacogenomics are used interchangeably. With the knowledge derived from clinical and biochemical assessment of individuals, the science of pharmacogenetics evolved to establish that a single therapeutic agent can not be expected to be an effective treatment across diverse populations, especially in polygenic disorders. This brings us to the 21st century and the prospect of personalized medicine in the pharmacogenomics era.

In 2003 completion of the Human Genome Project was announced, and with this announcement came an expectation of clinical applications for the near future with potential to base a patient’s therapy on his/her genetic background.

Pharmacogenomics in the postgenomic era was perceived as the arrival of the promise of personalized medicine, not only to improve outcome of drug therapy based on genetic risk, but also to identify new drug targets with maximal efficacy while minimizing adverse drug reactions. Adverse drug reactions are a major cause of

7 morbidity and mortality. Identifying genetic risk factors could significantly decrease healthcare costs and improve drug development [15].

The International HapMap Project (http://www.hapmap.org/) a multi-country effort to find and catalog genetic variations in the human population provides an extensive resource researchers can use to discover genetic variants that may be involved in disease and variable responses to drug therapy [16, 17]. More importantly the HapMap’s goal to provide a haplotype map that would identify ‘tag’ SNPs to capture SNP variation in haplotype blocks would greatly reduce sequencing and genotyping costs. This would then enable large-scale genotyping projects that would otherwise be too costly and labor intensive. Another important feature in todays postgenomic era is the accessibility to bioinformatics databases with resources for computational biology and tools for analyzing genome data such as the National

Center for Biotechnology Information (NCBI: http://www.ncbi.nlm.nih.gov/).

One of the goals of my project is to identify variations in the cholinesterase genes that could potentially affect either expression of the protein product or the catalytic function and/or stability of the enzyme. This was done in large cohorts of twin subjects and also in an unrelated individual panel. Access to registries and to genotyping data from these registries has allowed linkage analysis from previously genotyped data with interesting results that will be outlined and discussed in subsequent chapters.

B. The Relevance of Twin Studies in Pharmacogenomics

Francis Galton first coined the term “nature versus nurture” when he attempted to determine if human ability was innate (inherited) in men of scientific eminence by looking at their families and relatives. Realizing the limitation of this method he proposed twin studies to tease apart the inherited genetic component of intelligence from the environmental component by comparing twins similar at birth (monozygotic) reared in different environments and twins dissimilar at birth (dizygotic) reared in similar environments. Galton used questionnaires to gather data that he tabulated and outlined in his paper "The History of Twins" to conclude that “nature was far stronger then nurture” [18]. This is often cited as the first paper describing the classical twin method although it was actually Hermann Siemens, a dermatologist who in 1924 used correlation of the concordance rate of mole counts in monozygotic (MZ) versus dizygotic (DZ) twin pairs to establish the twin rule of pathology: a heritable disease would be more concordant (occurrence of the same trait in both twin members) in identical twins versus non-identical twins, and concordance would be even lower

(discordant) in non-siblings [19]. For monogenic diseases with Mendelian patterns of inheritance twins have not traditionally been used; instead family pedigrees with affected sibling pairs have generally been studied with much success. But for complex traits such as cardiovascular disease involving a combination of genetic and environmental factors with perhaps small but additive effects, this approach has run into problems. Twin studies through their common birthdates and shared environments (intrauterine and postnatal) have become an invaluable source for

10 investigating the relative importance of genetic and environmental factors in complex traits and disease susceptibility (Fig. I.1) [20].

An important resource in the dissection of complex traits is the availability of large worldwide registers of data on twins and their relatives that can be used to understand the genetic epidemiology of diseases and the interaction of genotype and environmental risk factors by comparing in MZ and DZ twin pairs the co-morbidity of traits and diseases [21]. Expansive twin studies have been completed, perhaps due to recent advances not only in molecular genetic tools but also in accessibility to bioinformatics databases such as the NCBI mentioned above that allows public accessibility of mining genome data. The development of powerful statistical programs that can accommodate the complexity of analyzing data between intra-twin pairs such as Sequential Oligogenic Linkage Analysis Routines (SOLAR) [22]and Mx

Statistical Modeling [23]have opened up new avenues of research in analyzing twin genetics. Comparing concordance and discordant rates between MZ and DZ twins, twin registers provide an excellent resource for evaluating the importance of genetic variation in disease susceptibility and environmental stress. Examples and types of twin studies and their application adapted from Boomsma et al. are listed in Table I.3.

In my research I have used the San Diego Twin Registry

(http://elcapitan.ucsd.edu/hyper) to analyze and characterize AChE polymorphisms found in samples of genomic DNA from unrelated and twin pair subjects and the

Australian twin registry (http://www.twins.org.au) for linkage analysis and association studies with BChE activity and the metabolic syndrome. Studies and results will be presented and discussed in Chapters II and III respectively.

TDT- transmission disequilibrium test adapted from MacGregor et al. 2000

Figure I.1 Contribution of Twins to the Study of Complex Traits and Diseases: For complex traits twins provide an estimate of the genetic component to disease through estimation of heritability, enhance strategies to detect genes through linkage and association studies, provide a matched setting (same birthdate and environment) in which to assess the risk that is associated with environmental exposure, and provide insight into the way in which genes and the environment act together to cause disease.

C. Cholinergic Control of Cardiovascular Function

The autonomic nervous system (ANS) part of the peripheral nervous system plays an important role in maintaining homeostasis in the body. The sympathetic and parasympathetic divisions of the ANS typically function in opposition to each other in a feedback loop to achieve homeostasis. The neurotransmitter in the ANS for all preganglionic fibers, all postganglionic parasympathetic fibers, and a few postganglionic sympathetic fibers is ACh and the neurotransmitter for postganglionic sympathetic fibers is norepinephrine (Fig. I.2). Chromaffin cells in the adrenal medulla are innervated by preganglionic fibers that release ACh. This in turn activates the cells to release epinephrine and so adrenomedullary cells act like modified postganglionic fibers only that they release mainly epinephrine instead of norepinephrine. The sympathetic neurons under conditions of stress mediate the fight or flight response (pupil dilation, bronchi dilation, increased heart rate), while the parasympathetic neurons mediate the opposite response (pupil constriction, bronchi constriction, decreased heart rate) and is concerned mainly with conservation of energy and maintenance of normal function during periods of minimal activity [37].

The sympathetic division can therefore be thought of as the typical accelerator and the parasympathetic division as the brake.

The cardiovascular (CV) system is under reflex regulation mediated through the parasympathetic and sympathetic systems to ensure an appropriate supply of oxygenated blood to different body tissues under a wide range of conditions. This

Preganglionic Neuron

Postganglionic Neuron

Adrenergic Tissue Receptor Muscarinic Muscarinic*

Adapted from Principles of Pharmacology The Pathophysiologic Basis of Drug Therapy

Figure I.2 Sympathetic and Parasympathetic Neurotransmitters: Acetylcholine is released by both sympathetic and parasympathetic preganglionic neurons, parasympathetic postganglionic neurons, and sympathetic postganglionic neurons in sweat glands (*muscarinic receptors). All other sympathetic postganglionic neurons release norepinephrine (except for the adrenal medulla that releases primarily epinephrine) stimulating α and β adrenergic receptors. ACh stimulates nAChRs on sympathetic and parasympathetic postganglionic neurons and at neuromuscular junctions (NMJ). ACh stimulates mAChRs on sweat glands and on tissues innervated by parasympathetic postganglionic neurons. Presynaptic nicotinic and muscarinic receptors are also found primarily in the central nervous system and secondarily in the peripheral nervous system.

15 essential homeostasis process entails sensory monitoring primarily through baroreceptors that control arterial pressure and secondarily via chemoreceptors that control the level of oxygen and carbon dioxide in the blood. Collectively they are simply referred to as baroreceptors (Fig. I.3). The baroreceptors provide a negative feedback loop that activates the parasympathethetic response in elevated blood pressure and conversely decreased blood pressure depresses the reflexes activating the sympathetic response and causing blood pressure to rise. The baroreceptors are located in the heart and major blood vessels and the chemoreceptors are primarily in the carotid sinuses in the carotid arteries. Nerve endings in the baroreceptors are activated by stretch as the vessel walls expand and contract. Chemoreceptors in the carotid sinuses and aorta respond to the partial pressure of oxygen and carbon dioxide in the blood. The carotid sinus chemoreceptors are innervated by the glossopharyngeal nerve (IX) and the baroreceptors are innervated by the vagus nerve

(X) [38].

Afferent action potentials from the receptors are sent along the appropriate nerves to the nucleus of the solitary tract (NTS) in the brainstem. The NTS sends excitatory fibers to the caudal ventrolateral medulla (CVLM), the activated CVLM sends inhibitory fibers to the rostral ventrolateral medulla (RVLM). The RVLM is the primary regulator of the sympathetic system sending catcholaminergic

(norepinephrine and epinephrine) excitatory fibers to the preganglionic neurons in the spinal cord. Elevated blood pressure stretches the carotid and aortic baroreceptors, they in turn send action potentials relayed to the NTS inhibiting the vasomotor center

ACh acetylcholine; SA sino-atrial node; AV atrioventricular node; HR heart rate; NE norepinephrine EPI epinephrine Adapted from Netter’s Illustrated Pharmacology First Edition

Figure I.3 Sympathetic and Parasympathetic Regulation of Heart Function: The Sympathetic and Parasympathetic systems innervate the heart regulating cardiac function. They act in opposition to each other in a feedback loop the sympathetic system as the accelerator and the parasympathetic system as the brake. Increased sympathetic drive activates β1 adrenergic receptors in the SA node and increases pacemaker cell depolarization rate, heart rate and contractile strength. Parasympathetic activation through vagus nerves reduces heart rate, AV node conduction, and contraction force by ACh release activating muscarinic M2 receptors.

17 and activating the parasympathetic vagal nuclei release of ACh. This results in inhibition of the sympathetic branch and a decrease in haeart rate and blood pressure.

In turn low blood pressure leads to sympathetic activation with parasympathetic inhibition and an increase in blood pressure [38].

As mentioned above, activation of the sympathetic system increases heart rate and contractile force by releasing norepinephrine and epinephrine while activation of the parasympathetic system stimulates ACh release to reduce heart rate and contractile strength. Sympathetic efferent response activates β1 receptors in the sinoarterial (SA) node in the heart to increase pacemaker cell depolarization rate, heart rate, and contraction strength. The pacemaker cells of the SA node depolarize and promote atrial contraction while ventricular contraction is due to conduction from the artioventricular (AV) node to the AV bundle to Purkinje fibers. Parasympathetic efferent impulses sent via the vagus nerves release ACh that activates muscarinic M2 receptors. Receptor activation reduces cellular cAMP levels and increases K+ conductance and leading to pacemaker cell hyperpolarization to reduce heart rate, AV node conduction, and contraction force [39].

The two branches of the ANS have opposing effects on blood pressure in the homeostatic maintenance of CV function, sympathetic activation leads to an elevation of total peripheral resistance and cardiac output, while parasympathetic activation leads to a decrease in cardiac output. Because the sympathetic system is involved in activation of the fight or flight responses to stress, this is the area that has been targeted in the control of hypertension (HTN). HTN is a medical condition in which

18 the arterial blood pressure is constantly elevated. HTN can be classified as essential

(primary) when there is no specific medical cause to explain a patient's condition or secondary when elevated blood pressure is a result of another condition, such as kidney disease or certain tumors (especially of the adrenal gland). HTN treatment by pharmacological means involves 3 major drug classes: diuretics, ACE inhibitors, and

β and α adrenergic receptor blockers that target the sympathetic influence on Ca+ channel blockers norepinephrine and epinephrine neurotransmitter release [39].

Persistent HTN is one of the major risk factors associated with CV diseases but despite many classes of antihypertensives for treating HTN, substantial morbidity and mortality associated with CV diseases continues [40-44]. According to a detailed review written by Buccafusco in 1996 part of the reason may be due to the fact that effects of enhanced central sympathetic drive to peripheral organs and the vasculature are not always helped by antihypertensive drugs and in some cases may actually exaggerate sympathetic activity leading to chronic heart failure [45]. In his review

Buccafusco presents historical work performed by many researchers in both normotensive and hypertensive animal models beginning with the first in-depth study by Suh et al. in 1936, implicating the role of central cholinergic mechanisms in the neural control of CV function. Extensive ongoing research has added to a growing body of evidence that is consistent with the role of the central cholinergic pathway in the neural control of CV function. There are too many references to cite, but a few examples of relevant in vivo research [46-55] performed in this century is outlined in

Table I.4. Evidence gathered from these and other studies using experimentally hypertensive animals has implicated an important role of central cholinergic neurons

19 in the control of cardiovascular function and in the hypertensive condition opening up a novel approach to designing therapeutic treatments in ameliorating CV risk factors

[56].

In summary, central cholinergic pathways play a critical role in the regulation of CV function and in the maintenance of CV homeostasis. As discussed above ACh is the neurotransmitter of all preganglionic autonomic fibers, all postganglionic parasympathetic fibers, and a few postganglionic sympathetic fibers. In preganglionic sites and in the adrenal medulla ACh binds to its nicotinic receptor (nAChR) initiating cholinergic transmission and in postganglionic sites binds to muscarinic receptors

(mAChR) potentiating parasympathetic signaling. AChE is found in the presynaptic

(cholinergic) and postsynaptic (cholinoceptive) component of the cholinergic pathway where it plays an essential role in termination of neurotransmission by catalyzing the rapid hydrolysis of ACh. Toxicity of irreversible AChE inhibition and the compromised condition of AChE knockout mice have demonstrated the critical role of

AChE in cholingergic pathways. Cholinergic proteins therefore could and should be considered as potential therapeutic targets in ameliorating CV risk factors such as

HTN.

D. The Cholinesterases

Acetylcholinesterase (AChE; EC 3.1.1.7) and butyrylcholinesterase (BChE;

EC 3.1.1.8) are the two enzymes that form the cholinesterase (ChE) family. A tremendous amount of effort has been put into the study of these two enzymes, and as a result many important insights have been gained starting from the molecular level of the gene to the enzymatic function and structure of the proteins. I will attempt to summarize important findings in the field and will elaborate only on aspects that are relevant to my project.

The mammalian AChE gene is encoded within 7.5 kilobases and in the human genome is located on chromosome 7q22 encompassing six exons and four introns [57,

58]. The protein product, AChE, is differentially expressed in various tissues as a result of alternative splicing at the 3’ end of the open reading frame (Fig. I.4)[59, 60].

The mammalian BChE gene spans over 70 kb of the human genome and is located on chromosome 3q26 with four exons and three very large introns [61, 62]. No evidence suggesting alternative splicing in the BChE gene has been reported. Mammalian

AChE and BChE share ~55% amino acid sequence identity and identical disulfide bond arrangements [63, 64]. They are present in most tissues, although BChE is present more abundantly than AChE except in brain and muscle [63].

The specific function of AChE is in catalyzing the hydrolysis of acetylcholine

(ACh) at cholinergic synapses, the true biological function of BChE remains unclear,

A. AChE Gene (Pictorial representation by Shelley Camp) yg

ATG TAA TAA TGA AATAA AATAA E1 E2 E3 E4 E5 E6

HindIII HindIII RNA: PROTEIN: alternative c-terminal alternative 3’ end splicing anchors Read-through Exon 4 ATG TAA AAAAAAAAAAA Soluble monomer Splice to Exon 5

ATG TAA Glycophospholipid anchor to lipids AAAAAAAAAAA -S-S-

Splice to Exon 6

S - Attaches to ATG TGA S collagen and lipid linked AAAAAAAAAAA S-S subunits S-S S-S

B. BChE Gene

Figure I.4 AChE and BChE Gene Structures: A. AChE is encoded within 7.5 kilobases and in the human genome is located on chromosome 7q22 (Getman et al, 1992). The protein product, AChE, is a serine hydrolase that is differentially expressed in various tissues as a result of alternative splicing at the 3' end of the open reading frame (Li et al 1991). The Read-through form results in a soluble monomer; the Exon 4 to Exon 5 splice results in GPI anchored dimers found on RBCs; the Exon 4 to Exon 6 splice results in the synaptic form found in brain and muscle. B. BChE (bottom panel) spans over 70 kb and is located on chromosome 3q22. No evidence of alternative splicing for the BChE gene has been noted, although the basic rotein forms parallel the AChE forms. The protein products, AChE and BChE, contain an invariant catalytic core comprised of exon 2 through exon 4.

24 although it is capable of hydrolyzing ACh; but to date, a natural endogenous substrate has not been identified. It has been suggested that BChE acts as a scavenger enzyme capable of detoxifying natural compounds such as ingested plant esters and in the degradation of toxic and pharmacological agents [65].

ChE structural variants in the human population were initially found for the

BChE gene product but not for AChE. Individuals with inherited BChE mutations are phenotypically normal and were only discovered due to prolonged apnea associated with administration of the muscle relaxant, succinylcholine, administered during surgical procedures[66]. BChE is highly polymorphic and numerous BChE variants have been reported beginning in 1957 with Kalow and Staron’s report on atypical forms of serum cholinesterase seen in subjects that suffer postsuccinylcholine apnea

[9]. ESTHER, a database of the alpha/beta-hydrolase fold superfamily of proteins

(www.ensam.inra.fr/cgi-bin/ace/search/Achedb?class=Mutation&query=%2Ahuman- buche%2A) [67] and the Online Mendelian Inheritance in Man (OMIM)

(http://www.ncbi.nlm.nih.gov/entrez/dispomim.cgi?id=177400) [McKusick-Nathans

Institute of Genetic Medicine, Johns Hopkins University (Baltimore, MD) and

National Center for Biotechnology Information, National Library of Medicine

(Bethesda, MD)] websites contain annotated and periodically updated sources of information on BChE mutations.

In general BChE mutations produce enzymes with differing levels of catalytic activity. Two of the most common nonsynonymous coding single nucleotide polymorpisms (cSNPs) in this gene are the ‘atypical’ D70G variant that has ~30% lower enzyme activity than the most common wildtype BChE form [68] and the

A539T substitution known as the K variant with ~33% less activity [69]. Individuals homozygous for these mutations have prolonged apnea after succinylcholine administration, and carriers of atypical or silent phenotypes have been found to suffer from chemical hypersensitivity when exposed to anticholinesterase pesticides [70, 71].

Until recently it was a different story for the AChE enzyme, except for the well known and biochemically neutral H322N cSNP known as the YT blood group antigen

[72] structural variants for this enzyme had not been reported. Given the essential role of AChE in cholinergic synaptic regulation and the lack of detected variants, the absence of distinct phenotype variations led to the view that nonsynonymous polymorphisms would be too deleterious and potentially result in either a severely compromised state or not be conducive to life [73]. Recently that view has changed with the report of structural variants found on residues positioned on the AChE surface having no apparent affect to the catalytic properties of the mutant enzyme proteins, but potentially since they are on the surface of the protein, may have antigenic properties as seen for the H322N YT blood group antigen [74].

Structurally AChE and BChE belong to the superfamily of α/β hydrolase fold proteins [75]. As referenced above ESTHER is a database dedicated to the analysis of genes and protein sequences belonging to this superfamily of alpha/beta hydrolases

(http://bioweb.ensam.inra.fr/ESTHER/general?what=index). Crystal structures of human AChE and human BChE proteins complexed with either inhibitors or substrate and products have been solved (PDB# 1B41 and 2PM8, respectively)[76-79] and are depicted in Figure I.5 individually and overlapped to illustrate structural similarity between the proteins.

Although AChE and BChE share structural and functional similarity they differ in their substrate selectivity and inhibitor specificity [80-84]. AChE and BChE both catalyze the hydrolysis of the neurotransmitter acetylcholine (ACh) with similar efficiency, but differ in their ability to catalyze the hydrolysis of larger acyl group size such as butyrylcholine or benzoylthiocholine. BChE is able to accommodate larger substrates due to small but significant amino acid residue differences that confer a larger acyl pocket. Inhibitor specificity for BChE is based on its larger acyl pocket that allows binding of bulky ligands such as the covalent inhibitor tetraisopropyl pyrophosphoramide (iso-OMPA), and the absence of an aromatic group (tyrosine or phenylalanine) in the choline site of BChE for ethopropazine inhibition. AChE specificity is due to a site of aromatic residues near the mouth of the active-center gorge that confers favored inhibition by fasciculin, BW284c51, and propidium [82,

85-87]. Figure I.6 illustrates how binding of BW284c51 in the hAChE enzyme confers a tight and favorable fit resulting in a tight inhibitory complex compared to the binding of BW284c51 in hBChE that depicts a loose and unfavorable fit due to the lack of aromatic residues at the rim of the gorge.

Overlay of hAChE (red) and hBChE (blue) proteins illustrates the structural similarity between the two enzymes. View is looking directing into the gorge.

Figure I.5 hAChE and hBChE Crystal Structures: hAChE structure in red with catalytic triad depicted in white; hBChE structure in blue & catalytic triad (Glu, His, Ser) in yellow. Visualized by WebLab Viewer software (Accelyrs, San Diego)

Figure I.6 Conformation of ChE Proteins with Bound BW284c51: In the top panel hAChE protein illustrates a stable conformation with BW284c51 tightly bound. In the bottom panel BW284c51 is loosely bound to hBChE with an unstable conformation due to the lack of aromatic residues in the peripheral site at the rim of the gorge. Visualized by WebLab Viewer software (Accelrys, San Diego)

The catalytic site for both enzymes is found at the bottom of a deep (~20 Å) and narrow gorge lined predominantly with aromatic residues [88]. The active site for both AChE and BChE consists of a catalytic triad with an active serine, a histidine and a glutamate. The catalytic triad residues for AChE are at Ser203, His447, and Glu334 and for BChE the corresponding residues are at Ser198, His438 and Glu325 (Fig. I.5) [64, 78,

88, 89]. In order to maintain catalytic activity a number of subsites are important in the catalytic process. Steps involved in ester hydrolysis begin with transacylation in which the acyl group is transferred from the substrate to the active serine of the enzyme followed with addition of water and a subsequent deacylation of the enzyme.

The anionic subsite of the AChE enzyme consisting of aromatic residues (Trp86,

Tyr133, Tyr337, and Phe338) binds the quaternary trimethylammonium moiety of the choline portion of the substrate optimally positioning the carbonyl carbon of the ester at the acylation site [90, 91]. The catalytic process is facilitated by the orientation of the substrate’s carbonyl oxygen toward the hydrogens of two glycines and an alanine on the amide backbone of the oxyanion hole that stabilizes the tetrahedral transition state of the substrate [92]. The AChE residues are Gly120 and Gly121 and at Ala204 and the corresponding residues for BChE are Gly116 and Gly117 and Ala199. The binding pocket in the AChE enzyme as referenced above is much smaller then the binding pocket in the BChE enzyme. This is due to aromatic phenylalanine residues (Phe295 and Phe297) that line the acyl pocket of AChE. Corresponding residues in the BChE enzyme are Leu286 and Val288 that provide a hydrophobic, but less constrained acyl pocket. This is why BChE can accommodate larger substrates.

Substrate catalysis in the absence of inhibition follows the Michaelis-Menten kinetics (see Scheme I.1) and can be summarized as follows: 1) Substrate enters the catalytic site binds to form a covalent complex with the active serine (tetrahedral intermediate); 2) Collapse of tetrahedral intermediate with concomitant release of choline. The active center serine is acetylated and choline is bound to the anionic subsite; 3) Choline leaves the subsite, and exits through the gorge; 4) a water molecule attacks the carbon of the acyl group forming a new tetrahedral intermediate;

5) Collapse of tetrahedral intermediate with acetate bound in the active site; 6) Acetate leaves the active site and exits the enzyme through the gorge [93]. The kinetic equation derived for the above mechanism is detailed in scheme I.I [82] for substrate catalysis in the absence of inhibition, where E stands for enzyme, S for substrate, ES is the enzyme- substrate complex, v is the initial enzyme activity at concentration of substrate, V is maximal activity of the enzyme, and Km is the Michaelis-Menten constant.

uation I.1 q Scheme and E

However, substrate hydrolysis by the ChEs differ from Michaelis-Menten kinetics in that AChE activity is inhibited by excess substrate and BChE activity is activated by excess substrate. These phenomena known by the terms substrate inhibition and substrate activation can be described as a consequence of formation of a ternary complex. The proposed mechanism of action between the enzyme and two substrate molecules, one binding allosterically to the enzyme forming either an active or inhibited ternary complex is outlined below in Scheme I.2 [82]. In AChE the ternary complex has very little or no activity whereas in BChE it appears to be active.

Scheme I.2

In this Scheme I.II, as described by Radic and Taylor: E stands for free enzyme and S stands for substrate molecules; ES is the Michaelis-Menten complex; SE is substrate bound to an allosteric, peripheral site on the enzyme; and SES is the ternary complex with two substrate molecules, one bound to the active site and one to peripheral site.

Kss is the dissociation constant for substrate bound to the peripheral site; kcat is the turnover number for the Michaelis-Menten complex and bkcat is the turnover number for the ternary complex [94]. From this equation when the enzyme turnover by the factor b = 1, the reaction is the same as from Michaelis-Menten kinetics and binding of a second substrate molecule to the peripheral site does not influence catalytic

33 turnover. When b < 1, the enzyme is inhibited by excess substrate and when b > 1, the enzyme activity is increased. For inhibition a curve generated by the relationship between enzyme and substrate concentration (pS Curve) is bell-shaped as is seen in

Chapter III for my recombinant AChE enzyme proteins. In the BChE enzyme substrate activation gives a sigmoidal curve. Example curves taken from Radić and

Taylor are shown below:

Figure I.7 ChEs Activity Curves: Curves are generated as a function of substrate acetylthiocholine concentration. Curves were generated using the equation from Scheme II. Catalytic constants for human AChE (hAChE) and hBChE obtained from Kaplan et al (2001), Torpedo AChE (TAChE) and mouse AChE (mAChE) from Radić et al. (1992 & 1993), and Drosophila AChE (DAChE) from Stojan et al. (1998). Constants were determined at room temperature except for hAChE and hBChE determined at 27 °C which contributes to their relatively higher kcat values.

Allosteric binding to the peripheral anionic site with concomitant modulation of AChE catalytic activity was first proposed by Changeux in 1966 [95] and was corroborated by subsequent studies [96-98]. The peripheral anionic site (PAS) in the

AChE enzyme was identified by titration with the inhibitor propidium and confirmed with the resolution of the crystal structure placing the site at the rim of the gorge entrance [88, 98]. The PAS consists of 5 residues and in the human AChE enzyme are Tyr72, Asp74, Tyr124, Trp286, and Tyr341 [82, 91, 99, 100]. Corresponding residues in the human BChE enzyme are Asn68, Asp70, Gln119, Ala277 and Tyr332 [101, 102].

Differences in the PAS residues between AChE and BChE are thought to account structurally for ligand binding specificity (Fig. I.6) and to play a significant role in excess substrate concentration kinetic profiles that lead to substrate inhibition as reflected in the bell-shaped curve for the AChE enzyme and in substrate activation as seen by the sigmoidal curve for the BChE enzyme (Fig. I.7). The PAS aromatic cluster in the AChE enzyme specifically binds cationic and aromatic inhibitors that are too large to enter the gorge such as in the propidium and fasciculin ligands and also in binding bisquaternary ligands such as BW286c51 and decamethonium that are long and slender and can extend from the gorge to the PAS [103-105]. In BChE a lack of aromatic residues confer differential binding so that BChE binds most bisquaternary and bifunctional inhibitors, and aromatic inhibitors with several orders of magnitude lower affinity than seen for binding with AChE [82, 87].

Organophosphates (OPs) are nonselective and potent inhibitors of both AChE and BChE and are of special interest due to their use as pesticides and in the current political arena as potential agents of chemical terrorism. Pharmacological effects of

35 inhibition of AChE are due primarily to the resulting accumulation of ACh at sites of cholinergic transmission. Accumulation of transmitter leads to enhanced stimulation at cholinergic nerve endings associated with the preservation of excess released ACh.

OP toxicity can produce overstimulation to effector organs at any and all cholinergic synapses. As listed in Goodman and Gilman (10th edition) the anti-ChE agents can potentially produce the following effects: 1) stimulation of muscarinic receptor responses at autonomic effector organs; 2) stimulation, followed by depression or paralysis, of all autonomic ganglia and skeletal muscle (nicotinic actions); and 3) stimulation, with occasional subsequent depression, of cholinergic receptor sites in the

CNS. Accumulation of ACh at postganglionic muscarinic synapses leads to parasympathetic activity of smooth muscle in the lungs, GI tract, heart, eyes, bladder, and secretory glands and increased activity in postganglionic sympathetic receptors for sweat glands. This can lead to cardiovascular effects such as bradycardia and hypotension, severe respiratory distress, gastrointestinal effects, such as nausea and vomiting, abdominal pain, diarrhea and fecal incontinence, blurred vision and miosis, urinary incontinence and excess sweating and tearing. Nicotinic effects can include muscle fasciculations, cramping, weakness, and diaphragmatic failure. Autonomic nicotinic effects can include hypertension, tachycardia, and mydriasis. CNS effects include anxiety, emotional lability, restlessness, confusion, ataxia, tremors, seizures, and coma [106].

OP compounds act directly with the catalytic site of the ChEs and progressive inhibition is due to phosphonylation and phosphorylation of the active center serine by the OP forming stable serine-conjugated OPs which react very slowly with water

[107-109]. Reactivation of phosphorylated AChE can occur slowly by hydrolytic cleavage if aging has not occurred. Aging occurs when an OP-cholinesterase conjugate undergoes an internal dealkylation reaction rendering the conjugated complex largely resistant to oxime-assisted reactivation.

Early studies by Wilson and Ginsburg showed that by directing nucleophiles to the active site to break the phosphoester bond, conjugated OPs could be released regenerating the enzyme [110]. This approach led to development of the first pyridinium aldoxime (2-PAM) used as an antidote in OP poisoning. Oximes are strong nucleophiles that are effective in reactivating OP-cholinesterase conjugates.

Nucleophilic strength of the oxime, orientation of the nucleophile with respect to the inhibited enzyme, and rate of aging, all affect reactivation [111]. In Figure I.8 a reaction cycle for alkyl phosphate hydrolysis catalyzed by AChE and an oxime is depicted with three subsequent fates: oxime-assisted reactivation, spontaneous reactivation by hydrolysis, and formation of an aged enzyme conjugate resistant to reactivation [112]. Studies are now underway using AChE as a template to design and develop modified AChE enzymes that are 1) more susceptible to oxime reactivation to provide a scavenging agent that can catalyze the hydrolysis of OPs in plasma before they reach their cellular target site; 2) in the synthesis and selection of more effective reactivating agents; and 3) as a diagnostic in the detection of OP exposure [112]. This work is essential especially in today’s environment of widespread use of pesticides that has not only led to the development of pesticide resistance in some insects [113] but also in cases of inadvertent or intentional misuse of OPs which warrants

37 immediate research into the development of designing new methods of detection and more effective treatments against OP toxicity.

Figure I.8 AChE Inhibition and Reactivation: Depicted here is the phosphorylation step by an organophosphate rendering the acetylcholinesterase inactive followed by spontaneous reactivation or assisted-reactivation promoted by an administered oxime [112]

D. Dissertation Overview

At the start of my thesis project genetic variations in the cholinesterases were thought to occur mainly in the BChE gene with the view that variations in the AChE gene would be very rare given its essential role in cholinergic neurotransmission.

Given the difficulty in mining for variants in AChE due to the uneven distribution of

GC content making GC rich regions hard to sequence and given that this gene is highly conserved throughout nature in all species, this view was not surprising.

Previously there had been only two reported studies, one by Bartels et al. in 1999 and another by Ehrlich et al. in 1994 with identification of only four polymorphisms [72,

114]. Out of the four, only two were significant and of clinical relevance. However since the start of my project variants including structural substitutions have been discovered [74] and in the work presented in this dissertation I will demonstrate polymorphisms in the AChE gene that confer an unstable phenotype when compared to the wildtype species.

Objectives of this dissertation were: 1) To discover naturally occurring polymorphisms in the AChE gene of a human population; 2) To identify and characterize the effect variations in the AChE gene may have on gene expression, the mature proteins structure and enzymatic parameters, and their effect in relation to cardiovascular function; 3) To understand and characterize the relationship between

BChE activity and the metabolic syndrome; and 4) Linkage analysis to assess quantitative trait loci affecting BChE activiy. Results and conclusions obtained for the

1st and 2nd objectives will be outlined and presented in chapter III. Results and

39 conclusions obtained for the 3rd and 4th objectives will be outlined and presented in chapter II.

My project is part of a large collaborative effort to understand the affect polymorphisms may have on CV function and the metabolic syndrome. As outlined in this chapter the cholinergic nervous system is critical in the homeostatic maintenance of CV function and as such the cholinergic proteins were selected for initial analysis.

AChE was selected as the initial candidate gene of interest for several reasons. In contrast to the great diversity in cholinergic receptors, a single gene, AChE, subserves the entire cholinergic nervous system as a key modulator of cholinergic neurotransmission. Although a natural substrate for BChE has not been identified it can also catalyze the hydrolysis of the cholinergic neurotransmitter, acetylcholine, almost as efficiently as AChE. To this end I assessed ChE activity to study BChE activity covariation with CV risk factors and components of the metabolic syndrome and I investigated the role of naturally occurring AChE polymorphisms on the gene and gene product.

CHAPTER II

Butyrylcholinesterase: Association with the Metabolic Syndrome and

Identification of 2 Gene Loci Affecting Activity

A. Abstract

Plasma cholinesterase activity is known to be correlated with plasma triglycerides, HDL- and LDL-cholesterol, and other features of the metabolic syndrome. A role in triglyceride metabolism has been proposed. Genetic variants that decrease activity have been studied extensively, but the factors contributing to overall variation in the population are poorly understood. We studied plasma cholinesterase activity in a sample of 2200 adult twins to assess covariation with cardiovascular risk factors and components of the metabolic syndrome, to determine the degree of genetic effects on enzyme activity, and to search for quantitative trait loci affecting activity.

Cholinesterase activity was lower in women than in men before the age of 50, but increased to activity values similar to those in males after that age. There were highly significant correlations with variables associated with the metabolic syndrome: plasma triglyceride, HDL- and LDL-cholesterol, apolipoprotein B and E, urate, and insulin concentrations; -glutamyltransferase and aspartate and alanine aminotransferase activities; body mass index; and blood pressure. The heritability of plasma cholinesterase activity was 65%. Linkage analysis with data from the dizygotic twin pairs showed suggestive linkage on chromosome 3 at the location of the cholinesterase (BCHE) gene and also on chromosome 5.

40 41

Our results confirm and extend the connection between cholinesterase, cardiovascular risk factors, and metabolic syndrome. They establish a substantial heritability for plasma cholinesterase activity that might be attributable to variation near the structural gene and at an independent locus.

B. Introduction

Butyrylcholinesterase (plasma or BChE) (EC 3.1.1.8; butyrylcholinesterase, pseudocholinesterase) has been studied extensively because of the associations between low activity and delayed metabolism of the muscle relaxant succinylcholine

[115-117] and because it is a marker of exposure to organophosphate chemicals [118].

The physiologic function of plasma cholinesterase remains unresolved, however, and it is not clear whether the known sequence variations in the butyrylcholinesterase gene

(BCHE) account for all of the genetic variations in the population. The causes of variation take on an increased importance in the light of reports that cholinesterase is involved in triglyceride metabolism; that its activity is correlated with plasma LDL- cholesterol (LDL-C), HDL-cholesterol (HDL-C), and triglyceride concentrations [119-

123]; and that it shows significant associations with components of the metabolic syndrome [124].

Although most of the evidence points to plasma cholinesterase acting as a marker (rather than a cause) of cardiovascular risk, metabolic syndrome, or diabetes, its proposed role in triglyceride metabolism might mean that natural variations in cholinesterase activity contribute to variations in risk. In addition to associations with cardiovascular risk factors, the metabolic syndrome, and possibly, type 2 diabetes

[125, 126], there are conflicting reports of a causative role for the low-activity K

42 variant in Alzheimer disease [127-134]. The biomedical importance of plasma cholinesterase is therefore wider than the pharmacogenetic phenomenon of delayed metabolism of succinylcholine or other drugs. The gene for plasma cholinesterase

(BCHE) is on chromosome 3, at bp 166973394–167037952, and many comparatively rare genetic variants leading to low activity are now known [135]. In addition to the known gene variants leading to low activity, a proportion of individuals show an additional cholinesterase band on electrophoresis and increased enzyme activity. This occurs with a frequency of 8% to 10% among Europeans [136], and is ascribed to the effects of another gene, cholinesterase (serum), 2 (CHE2), whose location is uncertain

[137]. A recent report [138] has shown that the increased activity is not attributable to increased cholinesterase protein concentration, but rather to increased specific activity.

Variations in plasma cholinesterase activity are therefore associated with variations in risk factors for cardiovascular and metabolic disease and are at least partly under genetic control. We studied variations in plasma cholinesterase activity in a sample of adult twins to assess the covariation with cardiovascular and metabolic disease risk factors, the magnitude of genetic effects on variation within the population, and the location of genes that determine or modify cholinesterase activity.

C. Participants and Methods

Participants in this study were described in a previous report [139]. They completed a questionnaire in 1989 and a telephone interview in 1993–1994, and provided a blood sample in 1993–1996. All participants were twins, born between

1903 and 1964, but in some cases only one member of a twin pair provided blood.

Zygosity was determined from responses to questions about physical similarities and the inability of others to tell them apart, supplemented by blood group information and

(for pairs included in linkage studies) extensive microsatellite genotyping. Participants gave informed consent to the questionnaire, interview, and blood collection, and the studies were approved by appropriate Ethics Review Committees. We collected blood samples from 1134 men and 2241 women. Plasma and serum were separated and stored at –70 °C until analyzed. Immediately before blood collection, participants completed a brief questionnaire reporting their alcohol consumption over the previous week. They also reported the time of their last meal, and the time of blood collection was noted. At the same visit, their height and weight were measured. Body mass index was calculated from weight and height as weight in kilograms (kg)/[height in meters

(m)]2. Systolic and diastolic blood pressures were measured, with the participants sitting, by use of an automated blood pressure recorder (Dynamap 845 Vital Signs

Monitor; Critikon Inc.). The mean of 2 results taken at 1-min intervals was calculated.

Blood pressure results were available for 1666 of the participants.

Plasma samples were analyzed for cholinesterase activity by measurement of the absorbance increase at 412 nm on addition of the substrate acetylthiocholine at a final concentration of 0.5 mmol/L, according to the colorimetric method of Ellman et al. [140]. 5,5'-Dithio-bis(2-nitrobenzoic acid) at a final concentration of 0.3 mmol/L was used as the chromogenic indicator of thiocholine formation. Samples were either measured in 1-cm optical-path cuvettes in a spectrophotometer (Response; Gilford

Instrument Laboratories) for 5 min or in a 0.73-cm optical-path microtiter plate reader

(Safire; Tecan Systems Inc.) for 3 min and 35 s. In either case, activity was recorded as a change in the absorbance at 412 nm in 1 min per microliter of serum.

Butyrylcholinesterase activity was determined by including the specific acetylcholinesterase inhibitor BW284c51 (1 µmol/L final concentration) in the reaction mixture. Activity was expressed in international units as µmoles of acetylthiocholine hydrolyzed per milliliter of sample per minute.

Serum -glutamyltransferase (GGT), aspartate aminotransferase (AST), alanine aminotransferase (ALT), cholesterol, glucose, triglycerides, and urate were measured by Boehringer Mannheim reagents and methods on a Hitachi 747 analyzer. HDL-C

was measured by precipitation of non-HDL lipoproteins with dextran/MgSO4 followed by enzymatic cholesterol assay. Apolipoproteins A-I, A-II, B, and E were measured by immunonephelometry on a Behring nephelometer using Behring reagents. Plasma insulin was measured by RIA (Diagnostic Products Corporation).

Several of the measured variables were log-transformed because their frequency distributions were skewed. All references to serum GGT, AST, ALT, triglycerides, and insulin and to the quantity of alcohol consumed per week are to the log-transformed values unless specified otherwise. LDL-C was calculated from the total cholesterol, HDL-C, and triglyceride values by the Friedewald equation if triglycerides were <4.5 mmol/L. If the serum triglyceride concentration was above this limit, LDL-C was treated as missing. The samples were not taken in the fasting state, but participants reported the time of their last meal, and the triglyceride, glucose, and

45 insulin results were adjusted for the elapsed time between the last meal and blood collection.

Exploratory analysis was carried out with SPSS, Ver. 13 (SPSS Inc.). Because the participants were twins and therefore not genetically independent, the effective number of individuals for any characteristic with substantial heritability would be less than the actual number of participants, and the significance (but not the magnitude) of correlations may be overestimated. More detailed examination of the effects of covariates and the sources of variation in cholinesterase was performed with the Mx program, Ver. 1.50 [23], which is designed for analysis of twin and family data and overcomes this problem.

We performed a genome-wide linkage analysis for loci affecting plasma cholinesterase activity on the dizygotic twin pairs. DNA was extracted from blood or buccal swabs according to standard procedures [141]. Genotype data were assembled from 4 genome scans that had been done previously for other projects by the

Mammalian Genotyping Service (Marshfield, WI), Leiden University Medical Centre

(Leiden, The Netherlands) [142], Sequana Inc., and Gemini plc (United Kingdom).

Pedigree structures for each scan were examined to identify inconsistencies between the genotypic data and pedigree relationships. Once any discrepancies were resolved, data for the 4 scans were merged and then checked again for pedigree errors. The combined genome scan data included 458 markers that were typed in 2 or more scans, which were included separately on the genetic map for the scan, separated by a very small distance [0.001 centimorgans (cM)]. The consistency of genotype information

46 among these 458 markers was checked via cross-tabulations of allele calls between different scans. Markers with genotypic data inconsistent between different genome scans were removed from further analysis. Map positions were in Kosambi cM, estimated via locally weighted linear regression from the National Center for

Biotechnology Information build 34.3 physical map positions and from published deCODE and Marshfield genetic maps. The procedures for combining and checking the genotype data and for the linkage analysis are described in Ref. [143].

Trait-specific empirical genome-wide suggestive and significant thresholds were calculated through use of 1000 gene-dropping simulations as described by

Abecasis et al. [144]. Details are given in Ref. [143]. The empirical genome-wide thresholds for suggestive or significant linkage [145] were defined as the thresholds for which we observed, on average, 1 or 0.05 peaks per simulation with a logarithm of odds (LOD) score at or above the threshold, respectively. After the initial simulation results, which produced unusually high significance thresholds, we applied winsorization to reduce the impact of outliers on the linkage analysis [146]. This was done by setting values for all cholinesterase residuals greater than 3 SD above or below the mean to values equivalent to 3 SD above or below the mean, respectively.

Linkage analysis and the results of simulations to determine the genome-wide empirical P value are reported for the winsorized dataset.

D. Results

Cholinesterase results were obtained for 2237 samples, including 564 complete monozygotic and 518 dizygotic twin pairs. The mean (SD) activity was 954 (272) U/L, with a nonparametric 95% range of 505-1539 U/L. Because the method was changed from measurement of absorbance with a spectrophotometer to use of a microtiter plate reader during the course of the study, we examined the running mean of the results across time and also calculated and graphed the cumulative sum of the differences from the mean (cusum) [147]. This plot (Fig. II.1) showed evidence of 5 periods with different mean values during the analysis of the samples, which was conducted over 15 months from January 2003 to April 2004. To reduce the effects of this analytical variation, dummy variables were created for the second to fifth periods and included as definition variables in an analysis using Mx. The results are summarized in Table II.1.

Significant effects of variation in the method over time were confirmed, together with significant effects of sex, and of age in women but not men.

Table II.1 Effects of Method Variation, Sex and Sex-Specific Age Effects on Plasma Cholinesterase Activity

Variables added -2LL Df ∆χ² ∆df p

None 8410.15 2160 - - -

Method 8080.05 2156 330.1 4 < 10-12

Sex 7992.08 2155 88.0 1 < 10-12

Age (female) 7933.25 2154 58.8 1 < 10-12

Age (male) 7933.20 2153 0.04 1 0.84

Cusum

1000 900 800 700 600 500 400 300 200 100 0

2201 2101

2001 1901

1801

rom first to last) on the x- to last) rom first 1701

1601 ily seen as changes in the slope 1623

to divide the results into five five into the results to divide Temporal variation (by date of differences from the overall mean 1501

1401 1301

CUSUM deviations sum of (Cumulative from overall mean) 1201

red by date of analysis (f 1101

990 List order

1001 ) 901 to the model of sources variation.

804 801 anges in the process are more read

2 = Safire 2 = results and the cumulative sum of

, Deviations from Overall Mean: Overall Deviations from ing mean. This information led us 701

Running Mean Running 601

1 = Gilford1 =

( 501

Method 401

458 301

201 101

8 6 4 2 0

(cusum) are plotted on the y-axis. Ch analysis) in cholinesterase results. Results were orde were Results results. in cholinesterase analysis) axis; both the running mean of 20 of the cusum plot than in runn Figure II.1 Cumulative Sum of groups and to incorporate group effects in 14 12 10 thod and Running Mean (20) Mean Running and thod Me

Exploratory analysis of the correlations between cholinesterase activity, adjusted for variation over time, and other variables gave the results shown in Table

II.2. There were multiple significant correlations with known cardiovascular risk factors and variables associated with the metabolic syndrome. There was no significant correlation, in either men or women, with alcohol intake or smoking.

The pairwise correlations by zygosity after adjustment for method variation, sex, and age are shown in Table II.3, together with the results of testing models of genetic and environmental sources of variation. We found that 65% (95% confidence interval, 50%–75%) of the variation in plasma cholinesterase activity was attributable to additive genetic effects. Although the model including only additive genetic and nonshared environmental sources of variation fitted the data satisfactorily, a small shared environmental effect cannot be excluded.

Table II.2 Correlations between Plasma Cholinesterase Activity (adjusted for method variation) and Variables Related to Cardiovascular Risk.a Males Females Age 0 0.24b Apo5 A1 0 –0.13c Apo A2 0.27b 0.06 Apo B 0.30b 0.34b Apo E 0.28b 0.27b Cholesterol (total) 0.28b 0.32b HDL-C –0.13c –0.23b LDL-C (calculated) 0.19c 0.29b Triglycerides (log) 0.33b 0.34b Urate 0.19c 0.25b BMI 0.26b 0.28 BP, systolic 0.28b 0.24b BP, diastolic 0.32b 0.23b GGT (log) 0.27b 0.24b AST (log) 0.16c 0.11c ALT (log) 0.22b 0.17b Glucose 0.07 0.15c Insulin (log) 0.12d 0.18b Alcohol intake (previous week) 0.06 –0.11c Smoker (Yes/No) –0.01 –0.04 a Triglyceride, glucose, and insulin values are adjusted for the reported time since the last meal. b-d P values are calculated on the highly conservative assumption that the effective number of cases is one half the actual number, to allow for any effects of the twin status of participants. Unless indicated, the correlation is not significant: b P <0.0001; c P <0.01; d P <0.05 e Apo, apolipoprotein; BMI, body mass index; BP, blood pressure.

Table II.3 Pairwise Similarity of Twin Plasma Cholinesterase Activity and Models for Fitting Data

A. MZF MZM DZF DZM DZOS N of pairs 408 158 224 77 217 Correlation of plasma cholinesterase activity 0.73 0.65 0.4 0.44 0.43 B. Percentage of variance due to: Models -2LL df ∆χ² ∆df P A C E 65 7 28 ACE 7922.5 2157 - - - (50-75) (0-21) (25-32) AE 7923.5 2158 1.01 1 0.315 72 - 28 CE 8007.6 2158 85.2 1 < 10-18 - 58 42 LL = log likelihood, df = degrees of freedom

A. Pairwise similarity of twin plasma cholinesterase results, by zygosity, after adjustment for method variation, sex and sex-specific age effects; B. results of fitting the data to models containing sources of variation due to A, additive genetic effects, C, shared-environmental effects, and E, non-shared environmental effects. The 95% confidence intervals for estimates of A, C and E under the ACE model are also shown. Comparison of the goodness-of-fit between the data and the models shows that the AE model is not significantly worse than the full ACE model, but that the CE models is strongly rejected.

The results of linkage analysis on 368 dizygotic twin pairs with genome-scan data are shown in Fig. II.1. Two peaks with LOD scores of 3.0 or greater were found, on chromosomes 3 and 5. The empirical significance thresholds determined by simulation on these data were 3.7 (for 1 occurrence in every 20 simulations, genome- wide; P = 0.05) and 1.8 (for an average of 1 occurrence per simulation), however; therefore, both peaks must be considered suggestive by the criteria of Lander and

Kruglyak [145] (expected to occur less than once per genome scan) rather than significant. The peak on chromosome 3 (peak LOD score, 3.00; empirical genome- wide significance, P = 0.241) at GATA3H01 (172.3 cM, or 168.7 Mb from the p- terminal end of the chromosome) coincided with the location of the BCHE gene (167.0

Mb). The peak on chromosome 5 (peak LOD score, 3.34; empirical genome-wide significance, P = 0.135) was at GATA12G02 (106.1 cM, or 91.0 Mb), and the 1-LOD interval ran from 98.4 to 123.1 cM (82.0–114.1 Mb). No other linkage peaks besides these two peaks exceeded the suggestive threshold of LOD score (>1.8).

Results ance from the p-terminal ance from the p-terminal mosome panel (chromosome ffecting plasma cholinesterase: plasma cholinesterase: ffecting axis and the genetic dist y ysis for the detection of loci a axis. x LOD score is plotted on the adjusted for method variation, sex, and sex-specific ag e effects. For each chro number listed at the top), the top), at the number listed Figure II.2 Results of linkage anal end of the chromosome on

E. Discussion

The observed plasma cholinesterase values covered a 3-fold range, from 505 to

1539 U/L, and the mean values varied by sex and, for women, by age (see Table II.1).

We confirmed that there are many significant correlations between cholinesterase activity and variables associated with cardiovascular disease, metabolic syndrome, and diabetes. We also established a significant degree of heritability for plasma cholinesterase activity in the general population and identified 2 chromosomal locations that contribute to this heritability. We detected variation in the method across time by cusum analysis to define the points at which changes had occurred. This permitted adjustment of the results for method variation as well as for age and sex effects. This was done before assessment of correlations with other variables, sources of variation, and linkage. Residual analytical variation would attenuate the correlations

(which were nevertheless highly significant) and inflate either nonshared or shared environmental sources of variation. Because both samples from twin pairs were generally analyzed at the same time, estimates of shared environment are more likely to be affected, and we did in fact find that if method variation was not taken into account the shared environmental effects were significant (data not shown). Because linkage analysis is based on within-pair differences, it will not be substantially affected by between-batch analytical variation.

As noted above, several previous investigators have found significant relationships between cholinesterase activity and triglycerides, HDL-C, and LDL-C

[119-123]. A recent report [124] has extended this to a wider range of variables

55 associated with the metabolic syndrome. Our results agree with this, and it is now clear that plasma cholinesterase clusters with a wider range of characteristics, including body mass index, apolipoprotein concentrations, insulin, liver enzymes, and blood pressure. Although many of these associations can be related to lipid or lipoprotein metabolism, the association with blood pressure does not easily fit into a concept of cholinesterase as an esterase involved in the metabolism of triglycerides and VLDL.

There seems to be a broader involvement of cholinesterase with the metabolic syndrome, extending to normal variation between people in their blood pressure and

(in the extreme case) to hypertension. The associations with AST, ALT, and GGT activities, which are known to be associated with insulin resistance [148] and increased in the metabolic syndrome [149], probably reflect an association between cholinesterase activity and the metabolic syndrome, of which fatty liver is a feature.

One important issue arising from these findings, and previous similar ones, is whether higher cholinesterase activity is a cause or a consequence of dyslipidemia and metabolic syndrome. Several published reports of animal [121, 150, 151] or human

[123] studies are relevant, but unfortunately, the results do not give a clear answer.

Interventions that primarily increase or decrease lipids tend to have the same effect on cholinesterase, but inhibition of cholinesterase activity in vivo has been shown to decrease lipid concentrations. For example, mice with streptozotocin-induced diabetes showed increased serum LDL, triglycerides, and cholinesterase [121], which decreased with insulin treatment, suggesting that the insulin-deficient state led to changes in cholinesterase activity. However, in the same study, inhibition of cholinesterase

56 activity with tetraisopropylpyrophosphoramide led to a decrease in serum LDL and triglyceride concentrations. These results appear to place cholinesterase in the chain of events leading to changes in lipid values, rather than being a consequence. If this is the case, and particularly if this reasoning also applies to the other features of the metabolic syndrome that showed significant correlations with cholinesterase activity, then the sources of variation in cholinesterase activity between people take on an added significance. Blood pressure, as well, might be influenced causally by cholinesterase because the cholinesterase substrate acetylcholine causes vasodilation when infused into the human vasculature, triggering nitric oxide release via endothelial muscarinic cholinergic receptors [152]. Thus, an excess of cholinesterase activity in the metabolic syndrome could also adversely affect endothelial function and ultimately increase blood pressure.

The pattern of within-pair twin correlations by zygosity and the model-fitting results, shown in Table II.3 suggest contributions to variation from additive genetic effects and nonshared environmental effects. Some shared environmental effects, possibly related to batch effects on the measurement of cholinesterase activity, cannot be excluded but are estimated at only 7% of the total variance. The genetic effects are substantially greater, at 65% (95% confidence interval, 50%–75%). Therefore, the major source of variation is genetic, and the location of the relevant genes can be assessed from the linkage analysis carried out on a subset of the dizygotic pairs.

This linkage analysis revealed 2 suggestive peaks, on chromosomes 3 and 5.

The localization of a gene or genes whose variation affects plasma cholinesterase

57 activity to chromosome 3 is to be expected, as this is the location of the BCHE gene itself. Nevertheless, it is gratifying to be able to identify linkage in the appropriate region, and this shows that linkage analysis for genes affecting quantitative traits can be done with comparatively small numbers of sibling pairs. Several variants of the

BCHE gene that affect cholinesterase activity are already known; these include the common K variant, which produces an 20% decrease in activity, and the much rarer variants with major effects. Given the high heritability of cholinesterase activity and the linkage peak on chromosome 3 at the BCHE locus, it is likely that there are other sequence variations in or near the BCHE gene that affect activity, and a search using single-nucleotide variations and haplotype analysis will help to define them.

The chromosome 5 linkage peak is less readily explained. There are 70 genes or possible genes (see Table IV.1) in the region indicated by this peak, but none has obvious relevance to plasma cholinesterase activity. A search through this region with current techniques would be time-consuming and expensive and not justified until the linkage result in this region has been replicated. One possibility that should be considered is that this region of chromosome 5 contains a gene for a protein that binds to cholinesterase and increases its activity [138]; efforts to identify the nature of this protein by conventional biochemical means have been unsuccessful to date. Another possibility is that genes in this region affect the risk of insulin resistance and metabolic syndrome and that the linkage for cholinesterase activity is a consequence of this.

However, other linkage studies on the metabolic syndrome [153, 154] do not support

58 linkage to this region of chromosome 5, nor do our own results on metabolic syndrome components using this cohort of persons [Ref. [155]and our unpublished data].

In summary, our results emphasize the relevance of cholinesterase activity to cardiovascular risk and extend knowledge of its sources of variation. Further characterization of the chromosome 3 BCHE locus might elucidate the effects of known variants against the overall heritability and the linkage peak. Such investigations might also determine whether genetic causes of low activity lead to lower values for the cardiovascular risk factors, a finding that would clarify the practical significance of plasma cholinesterase for cardiovascular risk and cardiovascular disease. The larger task of detailed examination of genes under the chromosome 5 peak must await identification of candidate genes in this region, or replication of our linkage result.

F. Acknowledgments

The material contained in Chapter II is, in full, from the publication

“Butyrylcholinesterae: Association with the Metabolic Syndrome and Identification of 2 Gene Loci Affecting Activity,” Anne Valle, Daniel T. O’Connor, Palmer Taylor,

Gu Zhu, Grant W. Montgomery, P. Eline Slagboom, Nicholas G. Martin, and John B.

Whitfield. Clinical Chemistry, 2006, 52(6):1014-1020. The dissertation author was a primary researcher and the author of the work; the co-authors listed in this publication assisted with the research which forms the basis for this chapter. Samples were obtained from the Australian Twin Registry by John B. Whitfield and Nicholas G.

Martin. All cholinesterase assays were run in Palmer Taylor’s lab and analysis of data was performed in the labs of John Whitfield, Palmer Taylor and Daniel T. O’Connor.

All results were analyzed and presented jointly.

Parts of this work were supported by the National Institutes of Health, the

Department of Veterans’ Affairs, and the National Heart, Lung, and Blood

Institute/Marshfield Mammalian Genotyping Service; by an GenomEUtwin grant

(EU/QLRT-2001-01254); and the Australian National Health and Medical Research

Parts of this work were supported by the National Institute of Health grant

R37-GM18360.

Chapter III

A Pharmacogenomic Study: Investigating Naturally Occurring Variations in the

Acetylcholinesterase Gene

A. Abstract

In this study I have taken a candidate gene approach to identify polymorphisms in the gene encoding for the enzyme acetylcholinesterase (AChE; EC: 3.1.1.7) to ascertain whether they adversely affect cholinesterase function, particularly in reference to the cholinergic control of cardiovascular (CV) function. Cholinergic neurotransmission plays an essential regulatory role in the homeostatic maintenance of

CV function mediated through the autonomic nervous system (ANS). AChE catalyzes hydrolysis of the cholinergic neurotransmitter acetylcholine (ACh), and as such, is a key modulator in cholinergic pathways. I examined naturally occurring single nucleotide polymorphisms (SNPs) in the acetylcholinesterase (AChE) gene in a human population in relation to catalytic properties and cardiovascular function. AChE can be found in whole blood enabling a biochemical phenotypic characterization in addition to the correlation of genotype with phenotypic physiologic responses.

Analysis of enzymatic activity was determined spectrophotometrically in plasma and blood cells of a twin subject registry and a general population subject registry.

Correlation analysis revealed significant relationships between AChE activity and certain cardiovascular endpoints. SNP discovery was performed by the re-sequencing of the AChE gene using genomic DNA (gDNA) of the general population registry, while SNP genotyping was performed using gDNA of the twin subject registry. To

60 61 date 19 SNPs have been identified: 7 SNPs in the coding region (cSNPs), 4 nonsynonymous encoding for a different amino acid and 3 synonymous encoding the same amino acid; 12 are in untranslated regions (UTR) of the gene with 3 of these in a conserved region of intron 1. Mutagenesis was conducted to introduce the nonsynonymous cSNPs into a human AChE cDNA vector. Upon successful mutagenesis, each cDNA vector with the incorporated nonsynonymous cSNP was transfected into

HEK cells for protein expression. Characterization of the purified mutant enzymes encoded by the SNP polymorphisms revealed significant thermal stability differences when compared with the predominant AChE species. Of particular importance is a cSNP at amino acid residue 134 where substitution of a single guanine (G) nucleotide to a cytosine (C) changed the aspartic acid codon to read as a histidine adversely affecting the stability of the expressed protein enzyme. With the structure of cholinesterase (ChE) enzymes and genes delineated, my research approaches a contemporary topic of analyzing disease susceptibility particularly in the realm of cardiovascular diseases, hypersensitivity to pesticides, and also the potential risk to individuals associated with chemical terrorism with agents affecting cholinergic function.

B. Introduction

As demonstrated in animal models, peripheral and central nervous system control of CV function mediated through the autonomic nervous system is critical in the homeostatic maintenance of blood pressure and responsiveness to exercise, postural alterations, and stress [45, 156-160]. Central cholinergic pathways in the spinal cord and higher brain centers modulate CV responses to influence basal blood

62 pressure and baroreflex pressor responses via the cholinergic neurotransmitter acetylcholine (ACh). Increased arterial pressure activates stretch receptors in the aortic arch and carotid sinus. Baroreceptor activation initiates afferent impulses to the vasomotor center (VMC) in the medulla of the brain stem. Stimulation of the vagus nerve in the VMC increases vagal parasympathetic release of ACh causing change in potential in the sino-atrial node (pacemaker cell of the heart) and atrio-ventricular conduction, thereby resulting in a decrease in heart rate that diminishes cardiac output

[161]. In turn, peripheral responses are controlled by nicotinic receptors in ganglia and the adrenal medulla as well as muscarinic receptor sites in postganglionic parasympathetic systems. Both the binding affinity of ACh to its receptors, and in turn, rapid ACh turnover are essential in the regulation of cholinergic neurotransmission and the cholinergic control of CV function. ACh binds to nicotinic and muscarinic receptors initiating cholinergic signaling, while AChE plays a key role in modulating cholinergic transmission by catalyzing the rapid hydrolysis of released

ACh. In the cholinergic control of CV function, it has been demonstrated using neostigmine (an inhibitor of AChE) that the rate of ACh degradation plays a greater role in determining the properties of transduction from vagus nerve activity to heart rate rather then the concentration of available ACh at the neuroeffector junction [162].

Outlined in detail in my introductory chapter and summarized here, the ChEs are serine hydrolases and are members of the α/β-hydrolase fold superfamily of proteins [75]. The mammalian AChE gene is encoded within 7.5 kilobases and in the human genome is located on chromosome 7q22 [57, 58]. The protein product, AChE, is differentially expressed in various tissues as a result of alternative splicing at the 3’

63 end of the open reading frame [59, 60]. The mammalian butyrylcholinesterase

(BChE) gene spans over 70kb of the human genome and is located on chromosome

3q26 [61, 62]. AChE and BChE share about 54% amino acid sequence identity and are present in most tissues although BChE is present in greated abundance than AChE except in the brain [63]. The catalytic sites for both AChE and BChE contain a catalytic triad of serine, histidine and glutamate residues [64, 88, 89]. While the specific function of AChE is in catalyzing the hydrolysis of ACh at cholinergic synapses, the true biological function of BChE primarily in liver, plasma and the intestine remains unclear although it catalyzes the hydrolysis of choline esters including ACh [83].

C. Material and Methods

Human Subjects: Participants included 80 unrelated subjects recruited by advertisement and referral from a general population in the San Diego area. Twin subjects were recruited by access to a population birth record-based twin registry

[163], as well as by newspaper advertisement. Dr. Dan O’Connor and his collegues recruited n = 478 twin individuals: n =167 monozygotic (MZ) pairs (33 male pairs,

134 female pairs), and n = 72 dizygotic (DZ) pairs (14 male pairs, 39 female pairs, and

19 male/female DZ pairs). Twin zygosity assignment was based on self-identification, with further confirmation by the presence or absence of heterozygosity at the tyrosine hydrolase gene microsatelite [164].

Ethnicities of twin and unrelated subjects were established by self- identification, as well as the ethnicities reported for both parents and all four grandparents. Twin subjects consist of 181 pairs of European ancestry, 12 pairs

African Americans, 15 pairs Hispanics, 2 pairs Filipinos, 5 pairs East Asian, 1 pair

Native American, and 23 pairs of mixed ancestry; unrelated subjects consist of 30

African-Americans, 8 Hispanics (Mexican-American), 6 Filipinos, 2 east Asians, and

34 of European ancestry. To address differences arising from ethnic backgrounds that may either mask an association or give a false-positive result, genetic association analysis was conducted on the twin panel, again using only twin subjects of European ancestry to address differences. Twin ages were 14 to 84 years. Family histories for hypertension [in a first-degree relative before the age of 60 yr] were as follows: 111 pairs were positive (one or both parents); 102 pairs were negative; and 26 pairs were indeterminate/unknown. There were 429 individuals that were normotensive, and 49 were hypertensive. None of the subjects had a history of renal failure. Subjects were volunteers from urban southern California, and each subject gave informed, written consent. Protocols were approved by the Institutional Review Board.

Sequencing Reaction: gDNA was extracted from leukocytes in EDTA- anticoagulated blood using a PureGene® DNA Purification Kit (Gentra Systems Inc,

Minneapolis, MN) and purified according to the manufacturer’s protocol. PCR reactions were performed in MJ Research Dyad® thermal cyclers (Waltham, MA).

Twenty-five microliter PCR reactions were performed using 25 ng of DNA (or H2O for negative controls), 25 mM of MgCl2, 10 mM of dNTPs, 20 µM of primer, and 0.5

U/reaction of Amplitaq Gold® Taq DNA Polymerase (Applied Biosystems, Foster

City, CA, www.appliedbiosystems.com). Taq polymerase was heat activated by incubation at 95°C for 15 min. The reaction was cycled 35 times with a denaturation step of 95°C for 30 seconds, an annealing step of 65.8°C for 1 min, an elongation step

65 of 72°C for 1 min, and a final elongation step of 72°C for 8 min. Ten microliters of

PCR product was used to verify amplification (or the absence of amplified product in negative controls) on a 1% agarose gel using electrophoresis. Fifteen microliters of

PCR products were purified with exonuclease I (3 U/reaction) and shrimp alkaline phosphatase (0.8 U/reaction) by incubation at 37°C for 30 min, and then at 85°C for

15 min.

Sequencing reactions were performed according to Applied Biosystems

BigDye® Terminator v3.1 Cycle Sequencing Kit Protocol (2002). Sequencing reaction mix was purified using Sephadex® G-50 DNA Grade beads (Sigma Scientific, St.

Louis, MO, www.sigmaaldrich.com). Ten microliters of Hi-Di Formamide was added to the purified sequencing reaction. Sequencing was performed on an ABI Prism®

3100 Genetic Sequencer (www.appliedbiosystems.com; Applied Biosystems).

Genotyping: Twin samples were genotyped using a PyrosequencingTM

HSA96A (Pyrosequencing AB, Uppsala, Sweden, www.pyrosequencing.com) instrument according to standardized protocol [165, 166]. A bound biotinylated single- stranded DNA was generated for using a sense biotinylated PCR primer (HPLC purified) and an anti-sense primer pair for both SNPs. Biotinylated PCR product (10

µl) was bound to streptavidin-coated Sepharose® HP beads by incubation with 30 µl binding buffer pH 7.6 (10 mM Tris-HCl, 2M NaCl, 1 mM EDTA, 0.2% Tween 20) for

10 min at 28oC with mixing and excess fluid removed by vacuum filtration. Beads were incubated for 1 min in 50 µl of denaturation solution (0.2 M NaOH) to remove non-biotinylated strands. Pyrosequencing reactions conducted by stepwise elongation of the primer strand by nucleotides (deoxynucleoside triphosphates: dCTP, dGTP,

66 dTTP, deoxyadenosine α-thiotriphosphate) followed by degradation of excess nucleoside triphosphates after each elongation step by apyrase. Sequential addition of the dNTPs followed the dispensation order determined by flanking sequence of the

SNP. The sequence at each elongation step was inferred by measuring light emission as an indicator of nucleotide incorporation, and the resulting sequences were analyzed by visual inspection and automatically by the SNP software.

Sequence and Haplotype Analysis: The Phred, Phrap, Consed suite of sequence analysis software was used to automate base calling, assemble sequence fragments, and visualize sequence data. PolyPhred was used to detect heterozygous sites, SNPs were re-verified using Consed graphical interface and also manually on the

ABI chromatogram. Haplotype inference on unrelated and twin subject panels was performed on Phase v.2 analysis software [167, 168].

Physiological/autonomic phenotyping in vivo: Subjects were studied prospectively, before genotyping. Wild-type and variant subjects were studied during the same time interval. Blood pressure (in mmHg) and pulse interval (R-R interval or heart period, in ms/beat) were recorded continuously and noninvasively for 5 min in seated subjects with a radial artery applanation device and dedicated sensor hardware

(Colin Pilot; Colin Instruments, San Antonio, TX) and software [ATLAS from WR

Medical, Stillwater, MN; and Autonomic Nervous System/Tonometric Data Analysis

(ANS/TDA) from Colin Instruments], calibrated every 5 min against ipsilateral brachial arterial pressure with a cuff sphygmomanometer.

Measurements of In-Vivo and Ex-Vivo Cholinesterase Activity: Whole blood samples obtained from the 80 unrelated individual subjects and the 478 twin subjects

67 were analyzed for ChE activity using a modification of the colorimetric method of

Ellman et al. [140]. To alleviate interference of hemoglobin absorbance from erythrocyte AChE in whole blood, the wavelength was adjusted to 436 nm [169]. At this wavelength hemoglobin absorption was reduced by 75% and a reduction of pH further amplified the signal-to-noise ratio [169]. Each sample was diluted 100X in blood diluting solution (300 ul of Triton X-100 in 1 liter of 0.1M NaP buffer pH 7) and incubated at 37 °C for ten minutes. Increases of the absorbance at 436 nm were measured upon addition of the substrate acetylthiocholine (0.5 mM final concentration). 5,5'-Dithio-bis(2-nitrobenzoic acid) (DTNB) at a final concentration of 0.33 mM was used as the chromogenic indicator of thiocholine formation. Samples were measured in 1 cm optical path curettes in a spectrophotometer (Response,

Gilford Instrument Laboratories, Oberlin, Ohio) for 5 minutes. ChE activity was recorded as the mean change in absorbance at 436 nm in 1 minute per μliter of whole blood. Each sample was measured for total ChE activity and for each specific enzyme acetylcholinesterase (AChE) and butyrylcholinesterase (BChE) activity by including the specific inhibitors BW284c51 and ethopropazine (respectively) at 1 μM final concentration in the reaction mixture.

Purified AChE enzyme activity was determined by the Ellman method at room temperature. Increase of the absorbance at 412 nm was measured upon addition of the substrate acetylthiocholine (1 mM final concentration). DTNB at a final concentration of 0.3 mM was used as the chromogenic indicator of thiocholine formation. Samples were measured in 1 cm optical path curettes in a spectrophotometer (Beckman

Coulter, Fullerton, CA) for 1 minute. Enzyme activity was recorded as the mean change in absorbance per minute of reaction.

Statistical Analysis: Descriptive statistics (means ± SD) and linear regression and correlation statistics were completed on the unrelated and twin panels using

GraphPad Prism v.4 software (San Diego, CA). ANOVA and t tests analyses for all enzyme denaturation results were also conducted using GraphPad Prism. Estimates of heritability (h2 ± VG/VP, where VG is additive genetic variance and VP is total phenotypic variance) were obtained using the variance-component methodology implemented in the SOLAR (“Sequential Oligogenic Linkage Analysis Routines”) package [22], available at the SOLAR website (http://www.sfbr.org/solar/). This method maximizes the likelihood assuming a multivariate normal distribution of phenotypes in twin pairs (monozygotic vs. dizygotic) with a mean dependent on a particular set of explanatory covariates. The null hypothesis (Ho) of no heritability is tested by comparing the full model, which assumes genetic variation (VG), and a reduced model, which assumes no genetic variation, using a likelihood ratio test.

SOLAR was also used to evaluate whether allelic variation at the locus contributed to a significant fraction of the trait heritability (i.e., locus-specific VG), by comparing models including or excluding the genotype as a covariate.

Plasmids and Mutagenesis: A cDNA encoding human AChE (hAChE) splice form exon 4 to exon 6 (kindly provided by Dr. Oksana Lockridge, Eppley Institute,

UNMC) was subloned into a pcDNA3 vector (pcDNA_hAChE) for expression and purification. Soluble monomeric hAChE was generated by introducing an early stop codon at residue Glu-548 by site-directed mutagenesis using the Stratagene

QuikChange Mutagenesis Kit (Stratagene, San Diego) as described previously for the soluble monomeric mouse AChE [170] and verified by automated sequencing. This recombinant protein referred to as the wildtype form (wtT547) was used as the template for all single cSNP mutations. Single base mutations were introduced into the wtT547 cDNA template using the Stratagene QuikChange mutagenesis kit for the

Asp-165 Æ His (D165H) cSNP, and the His-353 Æ Asn (H353N) cSNP. The Arg-34

Æ Gln (R34Q) cSNP required a 2-Stage PCR Protocol [171] using the Stratagene mutagenesis kit. A double mutant was constructed using the D165H cDNA template to generate the Arg-167 Æ Gln (D165H/R167Q) construct. DNA fragments containing each specific mutation were subcloned back into the wtT547 naïve template for all single mutants and the D165H template for the double mutant.

Plasmids were purified using the Plasmid Maxiprep System Kit (GilbcoBRL, Life

Technologies, San Diego). Sense and antisense strand mutagenesis primers for all above mutagenesis reactions are listed in Table III.1. All cSNP mutations were verified manually on the ABI chromatogram before subcloning, and the resulting subcloned sequences encoding each cSNP were again verified manually (ABI Prism®

310 Sequencer).

Table III.1 Mutagenesis Primers with Corresponding Template Sequence

Name Primer Sequence

Template pcDNA3_hAChE 5'GAC ACG CTC GAC GAG GCG GAG CGC C-3'

Mutation wtT547 Stop 5'GAC ACG CTC GAC TAG GCG GAG CGC C-3' 3'CTG TGC GAG CTG ATC CGC CTC GCG G-5'

Template wtT547 5'GCT GAG GGC CGG GAG GAT GCA GAG C-3' Mutation R3Q 5'GCT GAG GGC CAG GAG GAT GCA GAG C-3' 3'CGA CTC CCG GTC CTC CTA CGT CTC G-5'

Template wtT547 5'-CC TTG GAC GTG TAC GAT GGC CGC TTC TTG3'

Mutation D134H 5'-CC TTG GAC GTG TAC CAT GGC CGC TTC TTG3' 3'-GG AAC CTG CAC ATG GTA CCG GCG AAG AAC5'

Template wtT547 5'-C AAC GCG GGA GAC TTC CAC GGC CTG CAG3'

Mutation H322N 5'-C AAC GCG GGA GAC TTC AAC GGC CTG CAG3' 3'-G TTG CGC CCT CTG AAG TTG CCG GAG GTC5'

Template D134H 5'-C CAT GGC CGC TTC TTG GTA CAG G-3'

Mutation D134H/R136Q 5'-C CAT GGC CAG TTC TTG GTA CAG G-3' 3'-G GTA CCG GTC AAG AAC CAT GTC C-5'

Nucleotide substitutions are underlined and in italics. Listed above each mutation are the referenced wildtype cDNA sequences used as a template.

Cell Culture and Transfection: Human embryonic kidney (HEK293) cells were plated in 10cm dishes and maintained at 37 °C and 10% C02 in Dulbecco’s modified Eagle medium containing 10% fetal bovine serum (FBS) for 24 hours prior to transfection or when the cell density was 50 - 80% confluent. Cells were transfected either by Ca3(PO4)2 precipitation or with FuGENE 6 Transfection Reagent

(Roche Applied Science) using respectively 10ug or 6ug of plasmid vector with the incorporated SNP.

Clonal selection was dependent on incorporation of the neomycin-resistance gene contained in the plasmid and selected by growth in G418 (Geneticin, Sigma).

Clones with the highest relative AChE activity were selected for large scale protein production and purification. Stable cell lines were frozen in media with 20% serum and 5% DMSO for future use or grown in 10% serum to confluency in 10cm plates for expansion into flasks. Cell lines were periodically tested and shown to be free from mycoplasma contamination.

Real-time Reverse Transcription-PCR (RT-PCR): Total RNA was isolated using the RNeasy Mini Kit (Qiagen) according to manufacturer’s instructions from cultured HEK 293 cells stably transfected with either wtT547 or mutant D165H constructs. To avoid contaminating DNA, Turbo DNase treatment was performed according to manufacturer’s protocol (Ambion, ABI Systems). RNA was reverse transcribed into single stranded cDNA using Superscript First-strand System

(Invitrogen) according to manufacturer’s instructions. Real-time PCR analysis was performed using the Opticon DNA Engine 2 (MJ Research) and the Quantitect SYBR

Green PCR kit (Qiagen) using 1.5 μL of cDNA template in a 25 μL reaction. PCR

72 efficiencies of the primers were calculated by serial dilution of template and no significant differences in efficiency were found between the target genes and the housekeeping genes. All PCR assays were run in duplicate. Results were analyzed with the Opticon 2 Software using the comparative CT method as described [172].

Data were expressed as 2─ΔΔCT for the mutant D165H gene normalized against the housekeeping gene and the relevant positive control (wtT547) was also normalized against the housekeeping gene. The following primers [173] were used for the housekeeping gene HPRT: Forward primer 5' GTT AAG CAG TAC AGC CCC AAA

3' and Reverse primer 5' AGG GCA TAT CCA ACA ACA AAC TT 3'. Primers used for the hAChE gene: Forward primer 5’GAG GGC TCG TAT TTT CTG GTT T 3’ and Reverse primer 5’ CAG TCT GTG TAA TGC AGG ACC

Purification of Recombinant Enzymes: Stable cell lines grown to confluency in 10cm plates were transferred into three-layer flasks (2 plates per flask). The cells were grown to confluency in serum-containing media (120 ml/flask). Upon confluency the serum-containing media was discarded, and the flasks were rinsed with

50 mls of PBS and replaced with serum-free media (Ultra Culture, Biowhittaker or

SM4HiCell HyClone, Fisher) and maintained at 37 °C and 5% C02. Serum-free media containing the expressed recombinant enzymes were harvested and centrifuged (5000 g, 15 min, 4 °C) to remove cell debris and assayed for AChE activity using the Ellman method. Media was harvested until cell death (when cells began to visibly peel off the flask surface) generally from 4-6 weeks except for flasks with cells expressing the

D165H mutation (explained in results section).

Purification of the recombinant enzymes was performed by affinity chromatography utilizing a meta-trimethyl(aminophenyl)ammonium chloride inhibitor resin [174] typically in milligram amounts. A resin suspension (1ml for each mg of

AChE protein) was added to the harvested media and the resin-enzyme mixture was allowed to stir in a spinner flask (Bellco Glass, Inc., USA) overnight at 4 °C in the presence of 0.02% NaN3. The mixture was then assayed for residual enzyme activity and, if required, supplemented with additional resin until 80% of the enzyme was inhibited. The slurry media and resin was poured onto a Bio-Rad Econo-column, allowed to pack by sedimentation while stirring at 4 °C. Three successive washes were applied (first with 10mM bicarbonate loading buffer pH 8.0, next loading buffer with added NaCl increased to 100 mM, and the last wash again with loading buffer).

Elution by competition was accomplished using 100 mM decamethonium bromide dissolved in loading buffer at a slow flow rate (1-2 ml h-1). The purified enzyme was dialyzed using the 14-16 kDa cutoff dialysis tubing (Spectrapore, USA) in 4 liters of

Tris dialysis buffer (10 mM Tris HCl, 100 mM NaCl, 40 mM MgCl2, 0.02% NaN3, pH

8) for 8 hours 4 times. Activity was measured in 0.1 M NaPO4 buffer, pH 7.0 at room temperature by the Ellman assay using acetylthiocholine (ATCh) as the substrate and related to protein concentration. Proteins were concentrated to ~100 uM using

Centriprep 30 or Amicon Ultra-15 (Millipore, Bedford, MA) regenerated cellulose low binding membrane with a 30,000 NMWL (nominal molecular weight limit).

Determination of Kinetic Parameters for Enzyme Catalytic Activity: AChE enzyme activity was measured as a function of substrate concentration ranging from

0.01 to 100mM ATCh. Individual kinetic constants were determined by nonlinear

fitting of activities versus substrate concentration [82]. The Vm obtained from this experiment was used to determine the AChE catalytic turnover constant kcat by normalizing Vm to the concentration of AChE active sites. Active sites were quantitated by titration using the covalent, “irreversible” organophosphate inhibitor,

(Sp)-3,3dimethylbutyl methylphosphonothiocholine (SpDMB), to ascertain the minimal concentration required for complete inhibition, kcat was determined by comparing Vmax to the number of titratable active sites [82].

Thermal and Chemical Stability Assays: Stability assays were performed on the wildtype and mutant recombinant protein enzymes using 1 μM enzyme concentration in 0.1 M phosphate buffer, pH 7, for the thermodenaturation assay and

1 μM enzyme in 3 M urea (UltraPureTM Urea, Invitrogen) for the chemical denaturation assay. Thermal stability experiments were conducted at 50 °C for 60 minutes and the urea stability experiments were conducted at room temperature.

Aliquots were taken at regular intervals diluted 30-fold in phosphate buffer with

0.01% BSA to stop the denaturation process and remaining enzyme activity was measured using the Ellman method. Using GraphPad Prism nonlinear regression analysis of relative measured activity for the different time points was used to generate one phase exponential decay curves for all the recombinant protein enzymes.

Cholinesterase Inhibition by Paraoxon: Inhibition experiments were run using 40 nM wtT547 or D134H recombinant protein enzymes (400 pM final concentration in assay). Enzyme samples were incubated with paraoxon (300 nM) in

100 mM phosphate buffer, pH 7.4, containing 0.01% BSA for up to 40 minutes at 25

°C. Uninhibited enzyme samples were run in parallel as a control. The inhibition

75 reaction was stopped by addition of 1.0 mM ATCh, and the extent of inhibition was determined by measuring the residual activity using the Ellman method. Aliquots were taken at specified time points for the inhibited samples and uninhibited controls.

Oxime Reactivation of Phosphorylated Cholinesterase: Reactivation experiments for the paraoxon inhibited enzymes were performed using pyridinium aldoxime (2-PAM) as reactivator. Wildtype and mutant D134H enzymes at concentration of 33 μM were inhibited with paraoxon (45 μM) for 30 minutes at room temperature. Inhibited enzymes were diluted 10 times and passed through a Sephadex

G-50 spin column (Pharmacia) to remove excess unconjugated organophosphate. The enzymes were reactivated using 2-PAM (1 mM) in 100 mM phosphate buffer, pH 7.4, containing 0.01% BSA at 25 °C. At designated time intervals, aliquots of reactivation mixture were taken, and activity was measured using the Ellman method. As a control an equivalent sample of uninhibited enzyme was passed through a parallel column, diluted to the same extent as the inhibition mixture, and control activity was measured in the presence of oxime at the same concentration used for the reactivation of the inhibited enzymes.

D. Results

Cholinesterase phenotyping: In vivo ChE activity was ascertained for the 80 unrelated samples and the 478 twin samples (167 monozygotic and 72 dizygotic twin pairs). The AChE mean and standard deviation (SD) activity for the unrelated samples was 42.92 (7.2) ΔA/min/ul with a 95% confidence interval (CI) of 41.32 –

44.52, and the BChE activity mean was 9.82 (2.4) ΔA/min/ul with a 95% CI of 9.28 –

10.36. AChE mean (SD) activity for the twin samples was 42.04 (7.2) ΔA/min/ul with

76 a 95% CI of 41.39 – 42.69 and the BChE mean (SD) activity was 9.77 (2.3)

ΔA/min/ul with a 95% CI of 9.56 – 9.98. Frequency distribution of AChE and BChE activity for both the unrelated panel and the twin panel displayed an approximate

Gaussian distribution (Fig. III.1)

Linear regression and correlation analyses of ChE activity were performed on the monozygotic and dizygotic twin pairs to determine if there was a shared relationship within the twin pairs. Linear regression and correlation are not the same but they are related. Linear regression finds the line that best predicts Y from X, while correlation quantifies how well X and Y vary together. In linear regression the slope quantifies the steepness of the line if the slope is equal to zero the line is horizontal and you cannot predict Y by knowing X if it is 1 the line is at a 45° angle and by knowing X you can predict Y. In the linear regression analysis the ChE enzymatic activity of the older sibling within each twin pair was plotted on the X axis and the younger twin on the Y axis. Results of this analysis revealed significant slopes for all twin pairs ranging from 0.48 to 0.79 with slopes for the monozygotic twin pairs for each enzyme consistently closer to one than for the dizygotic twin pairs (Fig. III.2).

Unrelated Individuals

San Diego Twin Registry

Figure III.1 Frequency Distribution of Cholinesterase Activity: Distribution for the unrelated individuals (top panels) and twins from the San Diego registry (bottom panels) display a normal or approximate Gaussian distribution.

Figure III.2 Linear Regression Graphs of MZ and DZ Twin Pairs: These graphs display the linear regression analysis performed for the intra-pair twin subjects (167 monozgotic twin pairs and 72 dizygotic twin pairs). For each pair the older twin is designated as Twin 1 and enzyme activity for Twin 1 is plotted on the X-axis. Twin 2 enzyme activity is plotted on the Y-axis. The top 2 panels are graphs of AChE activity for the monozygotic and dizygotic intra-pair twins (respectively). The bottom 2 panels reflect graphs for BChE activity from the same intra-pair twins.

As described above when 2 variables (X and Y) vary together there is a lot of covariation or correlation between the variables. The correlation coefficient r quantifies the direction and magnitude of the correlation or relationship between X and Y. When r is equal to zero, the 2 variables do not vary together and when r is equal to 1, there is perfect correlation between X and Y. In this analysis I used the

Pearson r correlation coefficient which is based on the assumption that both X and Y values are sampled from a Gaussian or approximately Gaussian distribution (If the sample size is large enough this assumption is not critical). The best way to understand the value of r is to square it to calculate the coefficient of determination or r2. r2 is the fraction of the variance in the 2 variables that is shared. For example, if you calculate an r2 of 0.45 then 45% of the variance in X can be explained by the variance in Y. R squared values derived from the r values of the intra-twin pairs listed below were significant and ranged from 0.22 (22%) to 0.62 (62%).

SNP Discovery and Genotyping: Results of DNA resequencing of the AChE gene on the panel of 80 unrelated individuals have identified nineteen SNPs and potentially two microsatelite variations in the 3’ untranslated region (Fig. III.3).

Seven of the identified SNPs are in the coding region (cSNPs); four are nonsynonymous cSNPs encoding for a different amino acid including the well known YT blood group antigen and 3 are synonymous cSNPs encoding the same amino acid.

Twelve SNPs are in untranslated regions of the gene with three of those in a highly conserved region of intron 1 [175]. SNPs are listed in Table III.3 with identifying numbers for SNPs referenced in Build 128 (as of 10/23/07) of the public NCBI SNP database.

AATAA

3’UTR

AATAA MS– 1 GGGGA MS– 2 GGGGC Microsatellite (MS): SNP 15 16 17 18 19 17 18 SNP 15 16 3’ Untranslated Region TGA E 6

ssible microsatelite SNP14 Intron5 TAA P592R SNP13 TAA E 4 E 5 E 4

ck brackets); 2 po

P477P SNP12 E 3 E SNP11 Intron2

SNP10 Intron2 SNP9 H353N SNP8 E344E antigen; 3 synonymous cSNPs (in blue boxes); 12 SNPs in untranslated antigen; 3 synonymous cSNPs (in blue boxes); 12 SNPs SNP7 T269T E 2 the conserved region of Intron 1 (bla D165H SNP5 R34G SNP6 ATG gene SNP Discovery Findings: 19 SNPs – 4 nonsynonymous cSNPs (yellow boxes) ted region (blue brackets). AChE SNP4 SNP4 SNP3 SNP2 SNP1 SNP1 Conserved region: region: Conserved Intron 1 5’UTR E 1 Figure III.3 SNP Discovery: SNPs Found in the Unrelated Panel Mapped to AChE Gene Resequencing of the including the well-known YT blood group including the well-known or intronic regions including 3 SNPs in repeats in the 3’ untransla

Table III.2 Correlation Analysis of ChE Enzymatic Activity

MZ Twin Pairs DZ Twin Pairs AChE N = 167 pairs N = 72 pairs Pearson r = 0.49 Pearson r = 0.47

P<0.0001 P<0.0001

Significant Significant r2 = 0.24 r2 = 0.22

BChE N = 167 pairs N = 72 pairs

Pearson r = 0.79 Pearson r = 0.50

P<0.0001 P<0.0001 Significant Significant r2 = 0.62 r2 = 0.25

Pearson correlation coefficient r was used to determine the magnitude of correlation of ChE activity within the intra-twin pairs. Results shown here are significant with p values of < 0.0001. R squared values calculated from r indicate a significant degree of shared variance from 22% to 62% for the intra-twin pairs.

(http://www.ncbi.nlm.nih.gov/SNP/snp_ref.cgi?locusId=43). All of the minor allelic frequencies were computed for each SNP relative to the unrelated panel (Table III.4).

Several SNPs showed significant differences in allelic frequency (up to a 5-fold difference) between individuals of African American (AA) ancestry compared to the allelic frequency for individuals of European (Eur) ancestry.

Haplotype Reconstruction: AChE haplotypes were inferred in the unrelated and twin subject populations using PHASE v.2 software. Phase uses a Bayesian statistical approach for reconstructing haplotypes from population genotype data

[167]. Haplotype reconstruction using genotyped data from the unrelated panel incorporating all 19 SNPs revealed 30 inferred haplotypes with numerous 2N haplotype compilations. However, reconstruction using only the nonsynonymous cSNPs (R34Q, D165H, H353N, and P592R) on the unrelated genotyped data identified five haplotypes. Haplotype frequency between African-American individuals and European-American individuals was significantly different especially as noted in the two most common haplotypes #1 and #2 showing a 25 to 30%

(respectively) frequency difference (Table III.5).

Haplotype reconstruction using genotyped data acquired from the twin subject panel revealed 9 inferred haplotypes incorporating SNPs 5 (R34Q), 8 (E269E), 9

(H353N), 12 (P477P), and 16 (3’ UTR) (Table III.6). The wildtype haplotye contained no SNPs and indicated by a 1 was the most common with 426 individuals having a 2N haplotype of 1,1. This was followed by the 2N haplotype of 2,7 with 22

ber of individuals ber of individuals allele Freq = frequency of minor HTZ = heterozygotes HMZ = homozygotes AA = African Americans ancestry Eur = European *aa = numbering reflects amino acid in mature protein *aa = numbering reflects amino acid in N = num

Table III.5 Haplotype Reconstruction using the Nonsynonymous cSNPs

Haplotypes African-Americans European-Americans SNP Haplotype: 2N = 60 2N = 68 R34Q-D165H-H353N-P592R

#1 -arg-asp-his-pro- 76.7% 51.5% #2 -arg-asp-his-arg- 11.7% 41.2% #3 -gln-asp-his-pro- 8.3% 0.0% #4 -arg-his-his-pro- 0.0% 4.4% #5 -arg-his-asn-pro- 1.7% 2.9%

Using Phase software five haplotypes incorporating the four nonsynonymous cSNPs were constructed from genotyped data of African-American and European-American populations in the unrelated panel. Two common haplotypes (#1 and #2) were found in both populations but at significantly different frequencies.

86 individuals heterozygous for SNPs 9 and 12 and homozygous for SNP 16. Ten individuals with a 2N haplotype of 2,4 were heterozygous for SNP 12 and homozygous for SNP 16.

Table III.6 Haplotype (HAP) Reconstruction using SNPs 5, 8, 9, 12, 16

Nonsynonymous cSNPS: 5 – R34Q (C/T) 9 – H353N (G/T) Synonymous cSNPs: 8 – E259E(C/T) 12 – P477P (C/T) Noncoding SNP: 16 – 3’ UTR (C/A)

SNP Twin HAP # Haplotype N 2N HAP Frequency 5,8,9,12,16

1 CCGCC 865 1,1 426 2 CCGCA 36 1,2 1

3 CCGTC 6 1,3 5

4 CCGTA 15 1,6 3 5 CCTCA 1 1,7 1 6 CCTTC 3 1,8 3 7 CCTTA 26 2,2 1 8 CTGCC 3 2,4 10 9 TCGCC 1 2,5 1

2,7 22

3,9 1 4,4 2 4,7 1 7,7 1

Phase analysis software was used to infer haplotypes using genotyped data from the twin panel. Nine haplotypes were inferred with the most common or wildtype haplotype (HAP #1) with no SNPs referenced. The most common 2N haplotype (1,1) contains the wildtype haplotype at each allele. The 2N haplotypes with a frequency of 2% and 5% are 2,4 and 2,7. They each contain one SNP at one allele (HAP #2) and either 2 or 3 SNPs (HAP #4 and #7 respectively) at the other allele position.

Cholinesterase Heritability - Physiological and Allelic Association Studies:

Implementing SOLAR software both heritability and association analyses were conducted on the complete twin sample set (n = 478) and again on a subset of 362 twins of European ancestry (Table III.7A & 7B). Analysis was rerun on the twins of

European ancestry to address any analytic bias arising from allelic and physiological differences due to different ethnic backgrounds. Due to the limited number of twins of other ethnicities analysis based on ethnicity could be performed solely on the twins of

European ancestry. Studies on both twin panels revealed a significant degree of heritability (h2) of the biochemical phenotype (ChE activity) along with significant association with several physiological traits including CV phenotypic traits.

Heritability studies on the complete twin panel set revealed an h2 of 52% for AChE activity and an h2 of 80% for BChE activity. Rerunning the same analysis on the panel of twins with European ancestry gave similar values for ChE heritability with an h2 of

49% for AChE activity and an h2 of 81% for BChE activity.

Significant association was found in the complete twin panel set between

AChE enzyme activity and ethnicity, familial history of hypertension, and both systolic and diastolic blood pressure with covariate p values ranging from 0.02 to 0.09.

For BChE significant association was found between enzyme activity and blood pressure status, weight, body mass index (BMI) and systolic blood pressure (SBP) with covariate p values ranging from 3.6 E-10 to 0.08. Analysis on the subset of twins of European ancestry revealed significant association between AChE enzyme activity and blood pressure status and systolic blood pressure with a p value of 0.08 and 0.02

Table III.7A Heritability & Association Study (Complete Twin Panel)

Trait: ChE activity Trait/Covariant H2r H2r SE p value covar p value ACHE 0.52 0.05 8.1E-15 - Gender 0.50 0.05 1.9E-14 5.2E-06* BChE 0.52 0.05 3.5E-15 0.0056* Ethnicity 0.51 0.05 2.5E-14 0.0169* BP Status 0.52 0.05 5.7E-15 0.1673 Familial HTN 0.51 0.05 6.7E-15 0.0879* Weight (kg) 0.51 0.05 9.0E-15 0.1701 Height (m) 0.52 0.05 8.2E-15 0.8472 TD SBP Mean 0.50 0.06 8.6E-13 0.0202* TD DBP Mean 0.50 0.06 7.7E-13 0.0596* BMI 0.51 0.05 8.8E-15 0.3745 SNP 5 R34Q 0.67 0.06 1.1E-11 0.3356 SNP 8 E344E 0.49 0.07 4.3E-08 0.3655 SNP 9 H353N 0.64 0.05 1.1E-15 0.0504* SNP12 P477P 0.58 0.06 9.1E-13 0.0891* SNP13 P561R 0.56 0.06 4.4E-11 0.1551 SNP16 3'UTR 0.62 0.06 5.5E-13 0.2177 H1_1 0.51 0.05 1.3E-14 0.0672* H2_4 0.52 0.05 8.2E-15 0.0295* H2_7 0.51 0.05 2.9E-14 0.0293*

Trait/Covariant H2r H2r SE p value covar p value BCHE 0.80 0.03 1.9E-39 - Age 0.79 0.03 4.4E-37 0.0001* AChE 0.81 0.03 2.1E-40 0.0012* Ethnicity 0.80 0.03 2.1E-39 0.5767 BP Status 0.80 0.03 6.9E-38 0.0847* Familial HTN 0.80 0.03 2.6E-39 0.2528 Weight (kg) 0.79 0.03 7.0E-37 1.3E-09* Height (m) 0.80 0.03 3.0E-39 0.3473 BMI 0.79 0.03 1.9E-38 3.6E-10* TD SBP Mean 0.79 0.03 8.0E-31 0.0098* TD DBP Mean 0.79 0.03 5.6E-33 0.2797

Table III.7B (continued) Heritability & Association Study (Euro-Am Panel) Trait: ChE activity Trait/Covariant H2r H2r SE p value covar p value ACHE 0.49 0.06 9.4E-11 - Age 0.49 0.06 1.1E-10 0.2729 BChE 0.49 0.06 6.2E-11 0.0454* BP Status 0.49 0.06 8.9E-11 0.0791* Family History 0.49 0.06 6.0E-11 0.1284 Weight (kg) 0.49 0.06 9.7E-11 0.2314 Height (m) 0.48 0.06 1.4E-10 0.2277 TD SBP Mean 0.48 0.06 1.1E-09 0.0193* TD DBP Mean 0.48 0.06 1.0E-09 0.1179 BMI 0.49 0.06 1.2E-10 0.2041 SNP5 R34Q 0.71 0.06 3.8E-11 0.2932 SNP 8 E344E 0.49 0.08 1.2E-06 0.4861 SNP 9 H353N 0.66 0.06 1.5E-12 0.1585 SNP12 P477P 0.60 0.07 2.1E-10 0.1035 SNP13 P561R 0.55 0.07 2.4E-08 0.0126* SNP16 3'UTR 0.61 0.06 1.6E-10 0.0811* H1_1 0.48 0.06 3.3E-10 0.0394* H2_4 0.49 0.06 1.7E-10 0.0129* H2_7 0.48 0.06 4.1E-10 0.0514*

Trait/Covariant H2r H2r SE p value covar p value BCHE 0.81 0.03 1.1E-32 - Age 0.81 0.03 4.4E-31 0.0031* AChE 0.82 0.03 2.2E-33 0.0113* BP Status 0.81 0.03 5.6E-32 0.1694 Family History 0.81 0.03 1.3E-32 0.6713 Weight (kg) 0.80 0.03 1.2E-30 6.2E-09* Height (m) 0.81 0.03 7.9E-32 0.0278* BMI 0.79 0.03 1.4E-29 5.0E-12* TD SBP Mean 0.81 0.03 1.6E-26 0.0331* TD DBP Mean 0.81 0.03 1.9E-28 0.3184

H2r: Heritability SE: standard error *Significant covar p values

A. Entire twin set. B. European-American Twins (Euro-Am) SOLAR Studies: Results for heritability of ChE activity on the San Diego twins show a significant h2 of 49 - 82% for AChE and BChE. ChE activity and several physiologic traits as covariates show strong association as indicated here by the significant covar p values.

(respectively). Association of BChE enzyme activity was highly significant with weight, height, BMI, and SBP with covariate p values ranging from 5.0 E-12 to 0.03.

Genetic allelic associations between ChE enzyme activity and SNPs 5 (R34Q),

8 (E344E), 9 (H322N), 12 (P477P), 13 (P561R), and 16 (3’ UTR) were determined for both twin panel sets. In the complete twin panel set SNPs 9 and 12 showed significant associations with covariate p values of 0.05 and 0.09 respectively. In the European-

American twins SNPs 13 and 16 showed significant association with covariate p values of 0.01 and 0.08 respectively.

Haplotype association studies were also run on both twin panel sets between

ChE activity and the 3 most common haplotypes referenced above (1,1; 2,4; and 2,7).

Interestingly both twin panel sets showed a significant relationship between all 3 haplotypes and ChE activity with covariate p values of 0.07, 0.03, and 0.03 respectively in the complete twin panel and 0.04, 0.01, and 0.05 respectively in the

European-American twin panel.

Production of AChE Mutant Proteins: All but one of the nonsynonymous cSNPs were selected for protein production, the nonsynonymous cSNP Pro592 Æ Arg positioned in exon 5 was not produced because the hAChE cDNA used is the exon 4 to exon 6 splice form and therefore does not include exon five. The remaining cSNPs are shown in Figure III.4 modeled on the mouse AChE (mAChE) crystal structure

[176]. This structure (http://www.pdb.org/pdb/explore/explore.do?structureId=1J06) was obtained from the Research Collaboratory for Structural Bioinformatics (RCSB)

Protein Data Bank (PDB) a non-profit international resource set up as a database of

Figure III.4 cSNPs Modeled on mAChE Crystal Structure: The cSNPs are illustrated here in relation to the active serine at residue 203. The cSNPs are not in close proximity to the active serine. Although the R3Q cSNP shown at the end of the protein’s N-terminus could potentially elicit antigenic properties in vivo. Visualized using the WebLab Viewer software (Accelrys, SanDiego)

3-D structures to further the understanding of biological systems through the study of related macromocules [177]. The mouse structure was initially used in order to model the R3Q cSNP as the N-terminus of the human structure has not been fully defined and a proposed cleavage of the leader would place residue 5 as the N-terminus.

Mutations for all recombinant proteins were generated using the wtT547 cDNA template (exon 4 to exon 6 hAChE splice form). The nonsynonymous cSNPs

Arg34 Æ Gln, Asp165 Æ His, and H353 Æ Asp all located within exon 2 were successfully incorporated by site-directed mutagenesis. All introduced mutations were first verified by sequencing then cut out and subcloned back into the parental wtT547 hAChE pcDNA3 plasmid. Restriction enzymes used for subcloning are outlined in

Table III.8. Prior to large scale plasmid preparation, each ligation site and the entire cassette specific for each mutation was sequenced and manually checked on the ABI chromatogram to ensure the presence of the desired mutation and to verify that spurious mutations were not inadvertently introduced into the coding sequence.

HEK cells stably transfected with plasmids specific for each SNP were grown in the presence of serum until confluent and switched to serum-free medium for AChE expression. Serum-free media was harvested from cells expressing the mutant enzymes for protein purification as described in the methods section. The recombinant mutant protein enzymes were renumbered to reflect the mature protein numbering by subtracting the 31 amino acid leader sequence as follows: R34Q Æ

R3Q; D165H Æ D134H; and H353N Æ H322N. All AChE wildtype and mutant proteins purified yielded approximately 20 – 40 mgs of protein except in the case of the D134H recombinant protein. Problems encountered will be discussed below.

Table III.8 Vector pcDNA3 + hAChE cDNA (7224 bp)

Name Primer Position Restriction RE Position Band Vector on Vector Enzyme (RE) on Vector Size Size

wtT547 2630 - 2654 Not1; XbaI 2546; 2761 215 bp 7009 bp

R34Q 998 - 1022 HindIII; BstEII 889; 1574 685 bp 6539 bp

D165H 1336 - 1364 HindIII; BstEII 889; 1574 685 bp 6539 bp

H353N 1948 - 1975 BstEII; NotI 1574; 2546 972 bp 6252 bp

Restriction enzymes used in excising cDNA fragment with incorporated mutation. Band sizes indicate length of cDNA fragment cut out for each specific mutation and subcloned back into parental vector. Vector sizes indicate the appropriate length of the parental vector.

Temperature Sensitivity of the hAChE D134H Mutant Enzyme: Initially cultured cells transfected with the D165H construct expressed such a low quantity of gene product that protein production for purification was impossible. It was unclear whether poor expression was due to a problem at the transcription and/or at the translational or mature protein level. To determine if it was at the transcriptional level experiments using Real Time (RT)-PCR were performed. RT-PCR experiments were used to quantitate relative gene expression from cells expressing mutant D165H mRNA and cells expressing wtT547 mRNA (Fig. III.5). The mean relative gene expression for the wtT547 was 1.1 with a SEM of 0.013 and for the D165H mutant the mean was 0.95 with a SEM of 0.069. Comparison of the means by using Student t test analysis gave a p value of 0.227 indicating there was no significant difference between the wtT547 and the mutant D134H gene expression levels.

Previous work performed in our lab [178] had shown a temperature sensitivity for expression of a mutant mBChE recombinant protein enzyme. In decreasing the temperature to 31 °C the relative expression of the wild type product was diminished, but espression for the mutant was enhanced. The temperature dependence experiments were conducted at 31 °C and 37 °C. Transient and stably transfected

HEK 293 cells that expressed either the wtT547 construct or the D134H mutant construct were incubated at 37 °C or 31 °C for a total of 60 hours. Serum-free medium was collected and tested by Ellman assay for relative enzyme activity at 24,

48, and 60 hour intervals. Results were normalized to the wtT547 protein at 37 °C for each time interval. A representative graph illustrating enzyme activity for wtT547 and

D134H at a 48 hour time interval is shown (Fig.III.6).

Figure III.5 Comparison of wtT547 and D134H Gene Expression: WtT547 and mutant D134H cDNA was transcribed from mRNA to quantitate relative gene expression using RT(Real Time)-PCR. Template cDNA from the wildtype and the mutant cDNA were diluted either 1:100 or 1:1000. A t test comparison of the means using GraphPad Prism software found no significant difference with a p value of 0.227

Figure III.6 D134H Temperature Sensitivity: Stably transfected cells in serum- free medium expressing either the wildtype enzyme or the D134H mutant were incubated at 37 °C and 31 °C for 48 hours. Serum-free media was collected and tested by Ellman assay for enzymatic activity. Results shown here have been normalized to the wildtype protein at 37 °C.

Relative enzyme activity of media collected from cultured cells expressing the wtT547 protein at 31 °C reflected a significant decrease at each time interval when compared to measured activity from cells expressing the wtT547 protein at 37 °C. For the

D134H mutant media collected from cells at 37 °C showed virtually no activity while media collected from cells incubated at 31 °C expressed activity levels that were from

3 – 10 fold higher although still significantly lower then for the wtT547 expression.

Notably by dropping the temperature to 31 °C cells expressing the D134H mutant protein were able to express sufficient quantities of enzyme product for subsequent purification. A total of 2 mgs was purified which was enough to allow for protein characterization.

Kinetic Characterization of the hAChE Mutants: To ascertain whether the nonsynonymous cSNPs affected enzymatic function of the mutant ChE proteins (R3Q,

D134H, H322N) catalytic parameters (Km, Kss, kcat and b factor) for all the hAChE recombinant protein enzymes with the incorporated cSNPs were compared to the wtT547 recombinant protein enzyme using the following Scheme and Equation III.1

[82]:

Scheme and Equation III.1

Here E stands for free enzyme and S stands for substrate molecules; ES is the

Michaelis-Menten complex; SE is substrate bound to an allosteric, peripheral site on the enzyme; and SES is the ternary complex with two substrate molecules one bound to the active site and one to peripheral site. Kss is the dissociation constant for substrate bound to the peripheral site; kcat is the turnover number for the Michaelis-

Menten complex and bkcat is the turnover number for the ternary complex [94]. From this equation when the enzyme turnover by the factor b = 1, the reaction is the same as from Michaelis-Menten kinetics and binding of substrate to the peripheral site is kinetically inactive. When b < 1, the enzyme is inhibited by excess substrate and when b > 1, the enzyme activity is increased.

Examining substrate dependence using acetylthiocholine as the substrate and evaluating the number of active sites in titration experiments with SpDMB, an irreversible organophosphate, the obtained steady-state kinetics were employed to determine the catalytic parameters for substrate hydrolysis and inhibition for the wtT547 and mutant recombinant proteins. All substrate dependence experiments were performed a minimum of 3 times and normalized (v/Vmax) to accommodate for potential differences in protein concentrations. pS Curves were generated using normalized data from the wtT547 and mutant recombinant proteins and are represented in Fig. III.7. Specific enzyme activity (kcat) for each recombinant protein was expressed as a ratio of the quantity of hydrolyzed substrate per quantity of enzyme as obtained from the titration experiments. Titration experiments were performed on all the recombinant proteins a minimum of 3 times and averaged.

Specific activity was then used to re-plot the pS curves for the recombinant proteins.

A B

C D

Figure III.7 pS Curves: A = wtT547 B = R3Q C = D134H D = H322N All curves were generated using the Ellman assay over a range of acetylthiocholine concentrations as described by Radić et al.[82]. Colored symbols indicate individual experiments for each protein. Averaged activities are indicated by black boxes. Enzyme activities are expressed in relative units as fraction of corresponding maximal activity (v/Vmax).

proteins (Fig. III.8). Kinetic parameters (Km, Kss, b factor and kcat) are tabulated in

Table III.9 and reflect similar results for all proteins analyzed.

wtT547 R3Q D134H H322N

Figure III.8 Comparison of pS Curves for wtT547 AChE and Mutants: All curves were generated using the Ellman assay over a range of acetylthiocholine concentrations as described by Radić et al.[82]. Specific enzyme activities expressed as a ratio of the quantity of hydrolyzed substrate per quantity of enzyme from titration experiments using the irreversible organophosphate inhibitor SpDMB. For comparison all curves are overlaid in this graph and demonstrate no evident significant differences.

100

Table III.9 Kinetic Parameters for wtT547 and Mutant Protein Enzymes

5 Enzyme Km (mM) Kss (mM) b kcat (10 /min)

wtT547 0.13 ± 0.01 11.1 ± 2.0 0.036 ± 0.03 1.9 ± 0.2 R3Q 0.15 ± 0.02 8.6 ± 1.6 0.086 ± 0.03 1.8 ± 0.1 D134H 0.17 ± 0.02 10.9 ± 2.7 0.11 ± 0.04 2.0 ± 0.4 H322N 0.14 ± 0.02 8.4 ± 2.4 0.096 ± 0.04 1.9 ± 0.2 D134H/R136Q 0.14 ± 0.02 16.4 ± 3.7 0.062 ± 0.05 1.8 ± 0.1

Values for Km, Kss, and b were calculated using nonlinear computer fitting according to Scheme III.1 [82]. kcat was evaluated using SpDMB titrations with specific enzyme activity expressed as a ratio of the quantity of hydrolyzed substrate per quantity of enzyme. When there is no substrate inhibition, b = 1. If substrate inhibition is complete, b Æ 0.

101

Thermal and Chemical Stability Analysis of the hAChE Mutants: To delineate the stability parameters of the mutant enzymes (R3Q, D134H, H322N) thermal and chemical denaturation assays were conducted alongside the wildtype enzyme for comparison. After an initial screen of activity as a function of temperature thermal denaturation assays were carried out at 50 °C and chemical denaturation assays using 3M urea were conducted at room temperature. An aliquot of enzyme taken for each time point was measured using the Ellman method for remaining relative enzyme activity over a time course of one hour. All assays were performed a minimum of 3 times for all recombinant protein enzymes. Using GraphPad Prism nonlinear regression analysis of remaining relative enzyme activity calculated for all time points was used to generate one-phase exponential decay curves. Representative decay curves are depicted in Figure III.9.

One-Way ANOVA analysis was performed using GraphPad Prism to determine if there were significant differences in the stability parameters of the mutant protein enzymes when compared to the wild type profile. In the stability assays the

R3Q andthe H322N mutants showed a small but not statistically significant difference

(p > 0.05) in stability when compared to the wild type species. The D134H mutant, however, showed significant differences from the wtT547 enzyme in stability both in the thermal assay (p < 0.01) and in the chemical assay (p < 0.05).

Rescue of the D134H Single Mutant: In reviewing the experimental results for the D134H mutant it was noted that the catalytic kinetic parameters were equivalent to the wtT547 but the stability parameters of the D134H mutant showed

102

0.6

0.5

0.4

50 °C abs/min) Δ

0.3

0.2

ActivityAChE ( 0.1

0.0 0 10 20 30 40 50 60 Time (min) 0.6

0.5

wtT547

3M Urea abs/min) 0.4 T-R3Q Δ T-D134H T-H322N 0.3

0.2

( AChE Activity 0.1

0.0 0 10 20 30 40 50 60 Time (min)

Figure III.9 Exponential Decay Curves from Stability Assays: Representative curves comparing stability parameters for all purified recombinant enzymes are shown here. The top graph displays denaturation curves at 50 °C and the bottom graph represents chemical denaturation using 3M urea for all recombinant protein enzymes. Only the D134H mutant shows significant difference in stability when compared to the wtT547 form.

103 significant differences when compared to the wild type form. Instability of catalytic activity found for the D134H mutant may be related to local alterations in protein structure leading to a metastable conformational state. To examine this possibility the human AChE crystal structure [76] obtained from the RCSB PDB database

(http://www.pdb.org/pdb/explore/explore.do?structureId=1B41) was used to model the substituted histidine at residue 134 (Fig. III.10). The modeled mutation was found to be in close proximity to an arginine (Arg) at residue 136. The change in charge from negative to positive or neutral and the close proximity of the histidine (His) to another positively charged amino acid could potentially have an adverse effect on proper folding of the mutant enzyme. Possibly, this may be due to charge repulsion and steric hindrance since the molecular mass for the His side chain (MW: 81) is greater then that for Asp (MW: 59). Analyzing the model it was noted that the Asp at residue 134 was in proximity to the backbone structure leading to the oxyanion hole. It was then hypothesized that electrostatic forces from the change in charge created by the His substitution may be displacing the Asp residue which is close to the carbonyl of the surface loop (see Fig. III.10) resulting in less favorable metastable conformation.

To test this hypothesis, by using the single D134H cDNA as a template a double mutant was constructed in an attempt to rescue the D134H mutant. To alleviate the pressure put on the enzyme by the close proximity of two charged amino acids the uncharged neutral amino acid glutamine (Gln or Q) was substituted for the

Arg at position 136 (R136 Æ Q). Theoretically this would eliminate both the charge repulsion and steric hindrance since Gln is uncharged with a smaller side chain then the Arg at this position (Fig. III.11).

104

wt hAChE:

Asp134 in the wildtype form is close proximity to the Arg at 136; note also the close proximity of the backbone carbonyl from the surface loop.

D134H:

Substitution of the Asp134 to a His places two cationic residues in proximity.

Figure III.10 Human AChE Structure: Top panel with the Asp residue; Bottom panel with the His cSNP. Yellow Ribbon indicates amide backbone trace to the oxyanion hole. Light blue residues indicate catalytic triad. Note the close proximity of two positive charges with the His134 to the Arg136 in the mutant. Visualized by WebLab Viewer (Accelrys, San Diego)

105

Figure III.11 Double Mutant (D134H/R136Q): The double mutant modeled on the human AChE crystal structure in relation to the oxyanion hole and the catalytic triad. Note the steric hindrance constraint by the Arg next to the carbonyl of the surface loop is lifted with the Gln substitution. Visualized by WebLab Viewer (Accelrys, San Diego).

106

Upon successful incorporation and verification of the substitution R136 Æ Q by site-directed mutagenesis and sequencing, HEK cells were stably transfected and allowed to grow at 37 °C. Notably cells were able to produce the double mutant enzyme with expressed protein yields from harvested media equivalent to wild type production. This demonstrated at least a partial rescue by the R136Q substitution in that temperature sensitivity of the single D134H mutant in cultured cells was presumably corrected (Fig III.12).

Characterization of the double mutant was conducted as outlined above for the recombinant protein enzymes. Kinetic parameters (Km, Kss, b factor and kcat) determined for the D134H/R136Q mutant enzyme (Fig. III.13) did not differ significantly from the wtT547 or other single mutant protein enzymes (Table III.9).

Thermal and chemical stability assays were performed as described above at 50 °C and in 3M Urea on the D134H/R136Q mutant protein enzyme. Results were analyzed and compared against the wtT547 profile and the other single mutant protein enzymes.

Stability was characterized by the half-life (t50) parameter (time at which 50% of relative enzyme activity is still present). Comparison of the stability parameter (t50) tested for all the recombinant protein enzymes were analzyed by running a one-way

ANOVA using GraphPad Prism. Results from this analysis demonstrated a significant difference in the half-life of the mutant enzymes when compared to the half-life of the wtT547 enzyme, but again, particularly significant difference for the D134H mutant.

In depth examination of the single D134H recombinant mutant enzyme and the

D134H/R136Q double mutant using the Student t test verified that in fact the double mutant did reverse a significant degree of instability in both the temperature and

107 chemical denaturation assays with significant p values of 0.035 and 0.0001

(Fig.III.14).

Figure III.12 Protein Expression Normalized to the wtT547: Serum-free media from cells expressing wtT547, D134H/R136Q, or D134H protein was collected and tested by Ellman assay for enzymatic activity. Results shown here have been normalized to the wildtype protein. In this representative graph the double mutant protein expression is equivalent to the wildtype profile.

108

wtT547 R3Q D134H H322N D134H/R136Q

Figure III.13 D134H/R136Q pS Curve (Top Panel) and Summary of pS Curves for all Recombinant Proteins (Bottom Panel): pS curve was obtained as described by Radić et al. over a range of acetylthiocholine concentrations. Colored symbols indicate individual experiments with averaged activities indicated by black boxes. Specific enzyme activities expressed as a ratio of the quantity of hydrolyzed substrate per quantity of enzyme from titration experiments using the irreversible organophosphate inhibitor SpDMB. To compare results curves for all recombinant proteins are overlaid in this graph. No significant differences are evident.

109

52 °C Assay

3M Urea Assay

Figure III.14 Half-Life (t50) Parameter Normalized to the wtT547: Top panel shows the t50 profiles for thermal denaturation at 50 °C. Bottom panel represent profiles for chemical (3M urea) denaturation. Half-lives for all mutant protein enzymes were significantly shorter (p < 0.01) when compared to the wildtype for both assays but in particular for the D134H mutation. The D134H/R136Q double mutant rescue of the D134H single mutant is significant with a p value of 0.0347 in the thermal denaturation assay and a p value of 0.0001 in the chemical denaturation assay.

110

Further detailed analysis of the stability parameter for the mutant recombinant proteins was performed by calculating the t50 ratio (t50 mutant/t50 wild type) for each denaturation agent (Table III.10). The single D134H mutant enzyme half-life for both denaturants was found to be significantly shorter (< 20%) when compared to the wtT547 enzyme.

Table III.10 Relative Stability of the Mutant Enzymes (*t50 mutant / t50 wt)

50 °C Decay 3M Urea Decay t t Enzyme Rate ± SD 50 Rate ± SD 50 Ratio* Ratio*

wtT547 0.011 ± 0.0022 0.040 ± 0.0028 R3Q 0.024 ± 0.0079 0.46 0.049 ± 0.0052 0.82 D134H 0.300 ± 0.077 0.036 0.250 ± 0.042 0.17 H322N 0.022 ± 0.0044 0.48 0.064 ± 0.0098 0.63 D134H/R136Q 0.065 ± 0.023 0.18 0.096 ± 0.0056 0.42

Relative stability of the mutated hAChE enzymes are significantly less then for the wild type form. Ratio of the half-life for the single mutant D134H reflects a significant difference from wild type with less then 20% of the wild type half-life for both denaturants.

111

Organophosphate Inhibition of Cholinesterases: As described in my introductory chapter and summarized here OPs are progressive irreversible cholinesterase inhibitors. They are considered irreversible because they behave like substrate in that OPs bind to the ChE active serine, but unlike substrate once bound the process of deacylation is extremely slow thus inhibiting AChE from binding ACh, its natural substrate. To determine if the D134H mutant protein behaved differently than the wtT547 when exposed to an OP, I exposed both enzymes to paraoxon an OP pesticide. The process of progressive inhibition of AChE by an OP inhibitor can be defined by Scheme and Equation III.2 [179]:

Scheme and Equation III.2

Enzyme (E) and the organophosphate (OP) form a reversible Michaelis-type complex

([E][OP]) where the covalently phosphorylated enzyme (EP) and the leaving group arise; k+2 is the first-order inhibition rate constant and ki is the overall second-order

rate constant of inhibition and can be defined by equation III.2 where v0 and vi is the enzyme activity in the absence and in the presence of OP inhibitor at the time (t) of inhibition. Ki approximates to the dissociation constant of the [E][OP] complex, and kobs is the first-order rate constant of inhibition determined at the given inhibitor concentration [179].

112

Three independent experiments were run and rates were obtained using

SigmaPlot. Taking an average of the calculated results the rates are as follows: wtT547 kobs = 0.20; D134H kobs = 0.09. The inhibition rate of the wtT547 and the

D134H protein results indicate that the mutant enzyme is inhibited at least 2 to 2.5 times slower then the wildtype enzyme (Fig. III.15).

D134H wtT547

Figure III.15 Paraoxon Inhibition Curves for wtT547 and D134H Proteins: A representative graph of inhibition curves obtained for the wtT547 and D134H recombinant enzyme proteins are shown. Curves were generated using SigmaPlot (StyStat Software, Inc.) Individual experiments (n=3) were run in parallel with uninhibited enzymes as a control. The mutant D134H recombinant protein exhibited a slower rate of inhibition (2 – 2.5 times slower).

113

Oxime-Assisted Reactivation of Phosphorylated Cholinesterases: Oxime- assisted reactivation of inhibited enzyme is more efficient than water because oximes are stronger nucleophiles. The oxime-catalyzed reactivation proceeds via a Michaelis- type complex between the phosphorylated enzyme and the reactivator. This results in the inhibited enzyme becoming a free enzyme with the concomitant phosphorylation of the oxime [180]. To compare reactivation rates for the paraoxon inhibited wtT547 protein and the D134H mutant protein 2-PAM was used as the oxime in the reactivation process. Oxime reactivation of phosphorylated cholinesterases can be described using Scheme and Equation III.3 [111].

Scheme and Equation III.3

In this scheme EP is the phosphorylated enzyme, [EP][OX] is the Michaelis-type complex between EP and the oxime (OX), E is the active enzyme and P-OX the phosphorylated oxime, k+2 is the maximum first-order rate constant, and kr is the overall second-order rate constant of reactivation. Scheme III.3 is defined by the equation III.3 where [EP]0 and [EP]t are the concentrations of the phosphorylated enzyme at time zero and at time t. Kox is equal to the ratio, (k-1 + k+2)/k+1, and typically approximates to the dissociation constant of the [EP][OX] complex. kobs is the first-order rate constant of reactivation at any given oxime concentration [111].

114

Four independent experiments were run; two were run at 37 °C and two were performed at 25 °C, and rates were obtained using SigmaPlot. The experiments were run at different temperatures to account for the temperature sensitivity of the D134H mutant protein. Here again, unexpectedly the D134H mutant protein had a rate of reactivation that was at least 4 times faster than the rate of reactivation for the wtT547 protein. Taking an average of the calculated results from experiments performed at 25

°C the rates are as follows: wtT547 kobs = 0.0.027; D134H kobs = 0.11 (Fig. III.16).

D134H wtT547

Figure III.16 Oxime-Assisted Reactivation Curves for the wtT547 and D134H Proteins: This is a representative graph of reactivation curves obtained for the wtT547 and D134H recombinant proteins. Curves were generated using SigmaPlot (StyStat Software, Inc.) Individual experiments (n=4) were run in parallel with uninhibited enzymes as a control. The mutant D134H protein exhibited reactivation rates that were at least 4 times faster than the reactivation rates for the wtT547 protein.

115

E. Discussion

Results outlined in this study illustrate the importance and relevance of ongoing research in the field of pharmacogenomics. Polymorphisms in a person’s genetic makeup can significantly affect the individual response to physiological stress and/or environmental exposure. Differences in genetic associations and in physiologic responses were found in individuals with different ethnic backgrounds highlighting the importance of continued investigation of characterizing variants and their frequency levels in different populations. Although adverse consequences of cSNPs found in this study may only appear under conditions of stress and chemical assaults in-vivo, expression of SNPs in a homozygous state may potentially create severe consequences both in protein structural stability and in potential protein-protein interactions. These considerations highlighting the importance of continuing in-depth studies related to functional consequences of variable outcomes in disease and drug therapy in the general population. Alternatively noncoding SNPs, especially in reference to those in highly conserved regulatory regions (see Fig. III.3 SNPs 1-3), may not alter protein structure but can affect protein expression through influence on transcription and stability of the gene message (mRNA) and should be further investigated.

In particular stability testing in this study demonstrated how a single nucleotide change may produce alterations in a protein’s structural conformation that can alter its catalytic properties under stress-related conditions possibly due to local misfolding that results in a metastable conformation. In vivo and in physiological conditions individuals with this mutation could suffer consequences such as hypersensitivity to

116 agents aimed at the cholinesterase function or in responses to physiological stress especially if an individual is homozygous for a base substitution. Nerve agent OPs have been used in chemical terrorism. They and insecticide OPs react covalently with the active site serine of AChE. Individuals with different rates of inhibition reaction or reactivation by spontaneous hydrolysis may be more or less sucesptible to these agents.

Position of an amino acid encoded by a nonsynonymous cSNP on the AChE surface may also affect interaction of monomers in association with other molecules and could possibly produce antigenic responses. Again functional consequences of nonsynonymous cSNPs need further investigation. As mentioned in Chapter I another consideration is the tissue location of AChE. AChE is readily accessible in whole blood and serum samples whereas the tissues where activity most likely affects physiologic function are typically not accessible. However, my studies should reveal intrinsic differences in generalized expression parameters so that the AChE analyzed from these samples should be comparable to functional consequences in the CNS, skeletal muscle or autonomic nervous system.

The mechanism involved in the instability of the D134H mutant does not seem to affect its kinetic parameters probably due to its position on the surface of the mature protein as this mutation is far from the catalytic triad (Ser-203, His-447, Glu-334)

[100] that is located deep within the aromatic gorge of the enzyme structure. On the other hand the amino acid residue 134 is in the proximity of the oxyanion hole (Gly-

121, Gly-122, Ala-204) connected to it by the regional protein backbone that does not have a well-defined secondary structure (i.e. an α-helix or β-sheet) making this region

117 less stable and therefore more sensitive to changes that can be caused by polymorphisms in this region. This substitution from a negative residue (Asp134) to a positive residue (His134) places the positive His in close proximity to Arg136 that also contains a positive charge. In addition, Arg136 is close to the surface loop near the N-terminus. Placing two positive residues in close proximity could create an electrostatic field of charge repulsion making the protein with the histidine mutation less stable even under normal conditions, but especially under stress related (thermal or chemical) conditions. Under stress related conditions the histidine may alter the protein’s conformation making it more labile and by putting additional pressure on the surface loop causing it to be in an unstable position with more of the enzyme exposed to the surrounding environment. These events may play a significant role in a person’s susceptibility to disease and as we have shown for this cSNP substitution a greater sensitivity to environmental insults. On the other hand, if the D134H recombinant protein was stable at 37 °C, it could confer an advantage to a carrier due to its slower process of inhibition by an OP and by its faster rate of oxime-assisted reactivation. This demonstrates the importance of continuing efforts in using recombinant DNA methods for the development of new novel methods both for the detection of OP poisoning and in new antidotes for the treatment of OP toxicity [181].

In summary, with the structure of ChE enzymes and genes delineated this research is especially relevant today in analyzing disease susceptibility particularly in the realm of cardiovascular diseases, hypersensitivity to pesticides, and also individual risk associated with chemical terrorism agents aimed at the cholinesterase function.

118

F. Acknowledgements

The work presented in this chapter was a collaborative effort and I would like to thank all those involved including Drs. Dan O’Connor, Dr. Brinda Rana, and Dr. Nik Schork for assistance and helpful discussions and Dr. Phil Bourne for the PDB database structures. I especially want to thank Dr. Zoran Radić for his many hours of assistance and patient discussions relating to protein structure and enzyme kinetics. I would also like to thank Shelley Camp for her assistance in molecular biology, general lab protocols, and helpful discussions. This work was supported by NIH Grant R37-

GM18360 and GM07752 (Pharmacological Sciences Training Grant).

Chapter IV

Summary and Closing Remarks

My dissertation research project was part of a large collaborative effort at

UCSD to understand genetic mechanisms involved in cardiovascular function and to identify polymorphisms in candidate genes that may contribute to the hypertensive disease states. As the cholinergic nervous system is critical in the homeostatic maintenance of blood pressure the cholinesterases were selected as the initial candidate genes of interest for several reasons. In contrast to the great diversity in cholinergic receptors, a single gene, acetylcholinesterase, subserves the entire cholinergic nervous system as a key modulator of cholinergic neurotransmission, and though a natural substrate for butyrylcholinesterase has not been identified, it can also catalyze the hydrolysis of the cholinergic neurotransmitter, acetylcholine, as efficiently as acetylcholinesterase. To this end I assessed cholinesterase activity to study BChE activity covariation with cardiovascular risk factors and components of the metabolic syndrome, and I investigated the role of naturally occurring AChE polymorphisms on the gene and gene product. AChE and BChE is readily accessible in whole blood and serum samples however the tissues where activity most likely affects physiologic function are typically not accessible. However, my studies should reveal intrinsic differences in generalized expression parameters so that the AChE analyzed from these samples should be comparable to functional consequences in the

CNS, skeletal muscle or autonomic nervous system. Following is a summary of my research objectives: 1) Assessing BChE enzyme activity to ascertain covariation with

119 120 cardiovascular risk factors and components of the metabolic syndrome; 2) Linkage analysis to look for quantitative trait loci affecting BChE activity; 3) Engineering the nonsynonymous AChE cSNPs into a plasmid vector for expression, purification and subsequent characterization; 4) Statistical analysis of AChE polymorphisms in relation to their effect on cardiovascular physiological endpoints.

A: The Butyrylcholinesterase Story

In summary, results obtained from this study emphasize the relevance of cholinesterase activity to cardiovascular risk and the metabolic syndrome. Results confirm and extend the connection between heritability of cholinesterase activity and relevant physiological endpoints that can contribute to the hypertensive disease state and related cardiovascular conditions [36]. In Chapter II statistical analysis of BChE heritability and BChE activity correlation with CV risk factors were obtained using the

Mx software. Parallel analysis was also run using the SOLAR software that gave similar results. This further validates and confirms the accuracy and significance of our findings. Sources of variation of cholinesterase activity located with the linkage analysis showed a significant peak at the BChE locus on chromosome 3 and given the high heritability of enzyme activity further investigation into variants in this area is warranted. Detailed analysis of known variants in or near the BChE locus might help to elucidate effects of gene expression and enzyme activity to determine whether cholinesterase activity should be considered only as a biological marker of CV risk where low activity leads to lower values for risk or whether there are genetic causes that contribute to high enzyme activity leading to increased risk of disease.

121

Detailed examination of the genes located at the peak on chromosome 5 may result in a candidate gene or genes being recognized that could potentially play a role in CV risk. Running a quick analysis of ~28 mb (14 mb on each side of the peak) from the UCSC Genome Browser (http://genome.ucsc.edu/) identification of a gene peptidylglycine α-amidating monooxygenase (PAM) [182] with a potential nicotine dependence risk was located that warrants further study (Fig. IV.1 and Table IV.1).

Detailed analysis of genes in this area and a search using single-nucleotide variations and haplotype analysis will help to identify potential candidates.

An alternative consideration might relate to BChE being involved in the metabolic control of endogenous esters that affect CV function. This phenomenon might be further studied in individuals devoid of BChE or individuals, as were identified in this study, on the high end of the frequency distribution curve for BChE activity. However such studies which link metabolomics to pharmacogenomics will require large populations for study. Genetic control of cardiovascular function is likely to be richly polygenic and these gene products may control hydrolysis of several endogenous esters.

122

Figure IV.I Ideogram of Genes Located Around GATA12G02: A pictorial list of genes obtained for the area around the GATA12G02 peak identified in Chapter II.

123

Table IV.1 Links to Identified Genes around GATA12G02 on the UCSC Genome Browser GENE GENE PRODUCT AP3B1 adaptor-related protein complex 3 beta 1 SCAMP1 secretory carrier membrane protein 1 LHFPL2 lipoma HMGIC fusion partner-like 2 ARSB arylsulfatase B isoform 1 precursor arylsulfatase B isoform 2 precursor DMGDH dimethylglycine dehydrogenase precursor BHMT2 betaine-homocysteine methyltransferase 2 BHMT betaine-homocysteine methyltransferase JMY junction-mediating and regulatory protein HOMER1 homer 1 PAPD4 PAP associated domain containing 4 CMYA5 cardiomyopathy associated 5 MTX3 metaxin 3 THBS4 thrombospondin 4 precursor SERINC5 developmentally regulated protein TPO1 SPZ1 spermatogenic leucine zipper 1 ZFYVE16 zinc finger FYVE domain containing 16 UNQ9217 hypothetical protein LOC167555 DP58 cytosolic phosphoprotein DP58 DHFR dihydrofolate reductase MSH3 mutS homolog 3 RASGRF2 Ras protein-specific guanine CKMT2 sarcomeric mitochondrial creatine kinase ZCCHC9 zinc finger CCHC domain containing 9 ACOT12 acyl-CoA thioesterase 12 SSBP2 single-stranded DNA binding protein 2 ATG10 APG10 autophagy 10-like RPS23 ribosomal protein S23 LOC92270 hypothetical protein LOC92270 TMEM167 hypothetical protein LOC153339 XRCC4 X-ray repair cross complementing protein 4 VCAN versican HAPLN1 hyaluronan and proteoglycan link protein 1 EDIL3 EGF-like repeats and discoidin I-like COX7C cytochrome c oxidase subunit VIIc precursor RASA1 RAS p21 protein activator 1 isoform 1 RAS p21 protein activator 1 isoform 2 LOC644285 hypothetical protein LOC644285 CCNH cyclin H TMEM161B hypothetical protein LOC153396 MEF2C myocyte enhancer factor 2C CETN3 centrin 3

124

Table IV.1 (continued) Links to Identified Genes around GATA12G02 GENE GENE PRODUCT LOC153364 similar to metallo-beta-lactamase superfamily POLR3G polymerase (RNA) III (DNA directed) polypeptide LYSMD3 LysM putative peptidoglycan-binding domain GPR98 G protein-coupled receptor 98 precursor ARRDC3 arrestin domain containing 3 NR2F1 nuclear receptor subfamily 2 group F member 1 C5orf21 hypothetical protein LOC83989 FLJ25680 hypothetical protein LOC134187 C5orf36 hypothetical protein LOC285600 ANKRD32 ankyrin repeat domain 32 MCTP1 multiple C2 domains transmembrane 1 isoform S multiple C2 domains transmembrane 1 isoform L FAM81B hypothetical protein LOC153643 KIAA0372 hypothetical protein LOC9652 ARSK arylsulfatase K GPR150 G protein-coupled receptor 150 RFESD Rieske (Fe-S) domain containing SPATA9 spermatogenesis associated 9 RHOBTB3 rho-related BTB domain containing 3 GLRX glutaredoxin (thioltransferase) ELL2 elongation factor RNA polymerase II 2 PCSK1 proprotein convertase subtilisin/kexin type 1 CAST calpastatin isoform f 11 isoforms: f, g, h, a, e, i, j, k, l, b, and c ARTS-1 type 1 tumor necrosis factor receptor shedding (may be involved in BP) LRAP leukocyte-derived arginine aminopeptidase LNPEP leucyl/cystinyl aminopeptidase isoform 1 leucyl/cystinyl aminopeptidase isoform 2 (cleaves vasopressin) LIX1 limb expression 1 RIOK2 RIO kinase 2 RGMB RGM domain family member B isoform 2 precursor RGM domain family member B isoform 1 precursor CHD1 chromodomain helicase DNA binding protein 1 TMEM157 hypothetical protein LOC345757 ST8SIA4 ST8 alpha-N-acetyl-neuraminide SLCO4C1 solute carrier organic anion transporter family SLCO6A1 solute carrier organic anion transporter family PAM peptidylglycine alpha-amidating monooxygenase * GIN1 zinc finger H2C2 domain containing HISPPD1 Histidine acid phosphatase domain containing 1 C5orf30 hypothetical protein LOC90355 NUDT12 nudix -type motif 12 *This gene was identified in a separate genomewide linkage scan as a gene that affects risk for nicotine dependence [182]

125

B. The Acetylcholinesterase Study

Successful engineering of three of the nonsynonymous AChE cSNPs into a plasmid vector for expression and purification allowed for detailed characterization of the recombinant mutant proteins in comparison with the wildtype protein and in relation to their effect on enzymatic function. In summary results obtained from stability assays revealed a significant difference in stability between the wildtype enzyme and the mutant enzymes but most significantly for the D134H substitution.

Statistical analysis also revealed significant correlation between some of the identified

SNPs and cardiovascular physiological endpoints and interestingly ethnicity differences were found when analyzing haplotype frequencies between African-

Americans and Americans of northern and western European ancestry.

Previous work in our lab has identified a highly conserved regulatory region in

Intron 1 at the 5’ end [183] which warrants further investigation and detailed characterization of SNPs 1-3 found in this region (see Fig. III.3). Deletion of 255 bp of the critical region in Intron 1 in knockout mice showed a trembling phenotype and small size compared to their wild type litter mates. Detailed examination of the KO mice showed an absence of AChE activity in skeletal muscle tissue and provided in vivo evidence of the critical involvement of the 5’ intronic region in AChE expression during myogenesis making this a critical area for investigation.

Initial work done on the inhibition and reactivation of the D134H recombinant mutant enzyme revealed a slower inhibition rate by the organophosphate paraoxon and a faster rate of reactivation by the oxime 2-PAM when compared to the wtT547 recombinant enzyme. This preliminary finding warrants further study of all the

126 mutant recombinant enzymes with a range of inhibitor concentrations and the use of different OPs and also warrants a futher detailed look into the oximes tested for reactivation analysis. As outlined in the introductory chapter new methods for designing novel antidotes and detection of OP poisoning are especially relevant in today’s political climate of chemical agents used against the cholinesterases [112].

C. Closing Remarks

Taken together results from these studies reveal the relevance of future pharmacogenomics research to identify and characterize natural variants in human populations especially in candidate genes suspected to be involved in polygenic disorders. With the advent of new molecular and diagnostic technology and bioinformatics databases it is now feasible to conduct genome-wide scans to look for and identify variants that may adversely affect a gene’s expression or the gene product and potentially lead to a disease state. Results of this dissertation project in finding genetic loci attributable to cholinesterase activity and polymorphisms that exhibit protein structural differences and correlation with cholinesterase polymorphisms and

CV risk factors shows the relevance of continuing pharmacogenomic studies.

References

1. Kalow, W., Pharmacogenetics: A Historical Perspective. 2004: p. 251-272.

2. Garrod, A.E., The incidence of alkaptonuria. A study in chemical individuality. Lancet, 1902: p. 1616-1620.

3. Garrod, A.E., The inborn factors of disease. 1931.

4. Synder, L.H., Studies in human inheriance IX. The inheritance of taste deficiency in man. Ohio J. Sci., 1932. 32: p. 436-468.

5. Alving, A.S., et al., Enzymatic deficiency in primaquine-sensitive erythrocytes. Science, 1956. 124(3220): p. 484-5.

6. Bonicke, R. and W. Reif, Enzymatische inaktivierung von isonicotinsaure hydrazide im menschlichen und tierischen Organismus. Arch. Exp. Pathol. Pharmakol, 1953. 220: p. 321-333.

7. Hughes, H.B., On the metabolic fate of isoniazid. J Pharmacol Exp Ther, 1953. 109(4): p. 444-52.

8. Hughes, H.B., et al., Metabolism of isoniazid in man as related to the occurrence of peripheral neuritis. Am Rev Tuberc, 1954. 70(2): p. 266-73.

9. Kalow, W. and N. Staron, On distribution and inheritance of atypical forms of human serum cholinesterase, as indicated by dibucaine numbers. Can J Biochem Physiol, 1957. 35(12): p. 1305-20.

10. Motulsky, A.G., Drug reactions enzymes, and biochemical genetics. J Am Med Assoc, 1957. 165(7): p. 835-7.

11. Vogel, F., Moderne Probleme der Humangenetik. . Ergebnisse der inneren Medizin und Kinderheilkunde, 1959. 12: p. 52-125.

12. Meyer, U.A., Pharmacogenetics - five decades of therapeutic lessons from genetic diversity. Nat Rev Genet, 2004. 5(9): p. 669-76.

13. Alexanderson, B., D.A. Evans, and F. Sjoqvist, Steady-state plasma levels of nortriptyline in twins: influence of genetic factors and drug therapy. Br Med J, 1969. 4(5686): p. 764-8.

14. Vesell, E.S., Twin studies in pharmacogenetics. Hum Genet Suppl, 1978(1): p. 19-30.

127 128

15. Wilke, R.A., et al., Identifying genetic risk factors for serious adverse drug reactions: current progress and challenges. Nat Rev Drug Discov, 2007. 6(11): p. 904-16.

16. Tanaka, T., [International HapMap project]. Nippon Rinsho, 2005. 63 Suppl 12: p. 29-34.

17. Thorisson, G.A., et al., The International HapMap Project Web site. Genome Res, 2005. 15(11): p. 1592-3.

18. Galton, F., The history of twins as a criterion of the relative powers of nature and nurture. J.R. Anthropol. Inst. Gt Br. Ireland, 1875. 5: p. 391-406

19. Siemens, H.W., Twin Pathology: Its Importance, Its Methodology, Its Previous Results. 1924, Berlin: Springer.

20. MacGregor, A.J., et al., Twins. Novel uses to study complex traits and genetic diseases. Trends Genet, 2000. 16(3): p. 131-4.

21. Boomsma, D., A. Busjahn, and L. Peltonen, Classical twin studies and beyond. Nat Rev Genet, 2002. 3(11): p. 872-82.

22. Almasy, L. and J. Blangero, Multipoint quantitative-trait linkage analysis in general pedigrees. Am J Hum Genet, 1998. 62(5): p. 1198-211.

23. Neale, M., Mx: Statistical Modeling. 5th ed. 1999: Department of Psychiatry. Richmond, VA: Medical College of Virginia.

24. Ijzerman, R.G., C.D. Stehouwer, and D.I. Boomsma, Evidence for genetic factors explaining the birth weight-blood pressure relation. Analysis in twins. Hypertension, 2000. 36(6): p. 1008-12.

25. Roy, M.A., et al., A twin study of generalized anxiety disorder and major depression. Psychol Med, 1995. 25(5): p. 1037-49.

26. Snieder, H., L.J. van Doornen, and D.I. Boomsma, The age dependency of gene expression for plasma lipids, lipoproteins, and apolipoproteins. Am J Hum Genet, 1997. 60(3): p. 638-50.

27. Martin, N.G., et al., Co-twin control studies: vitamin C and the common cold. Prog Clin Biol Res, 1982. 103 Pt A: p. 365-73.

28. Forsti, A., et al., Use of monozygotic twins in search for breast cancer susceptibility loci. Twin Res, 2001. 4(4): p. 251-9.

129

29. Mack, T.M., et al., Heritable breast cancer in twins. Br J Cancer, 2002. 87(3): p. 294-300.

30. Eaves, L., et al., Comparing the biological and cultural inheritance of personality and social attitudes in the Virginia 30,000 study of twins and their relatives. Twin Res, 1999. 2(2): p. 62-80.

31. Truett, K.R., et al., A model system for analysis of family resemblance in extended kinships of twins. Behav Genet, 1994. 24(1): p. 35-49.

32. Weksberg, R., et al., Discordant KCNQ1OT1 imprinting in sets of monozygotic twins discordant for Beckwith-Wiedemann syndrome. Hum Mol Genet, 2002. 11(11): p. 1317-25.

33. Petronis, A., Epigenetics and twins: three variations on the theme. Trends Genet, 2006. 22(7): p. 347-50.

34. Cannon, T.D., et al., The inheritance of neuropsychological dysfunction in twins discordant for schizophrenia. Am J Hum Genet, 2000. 67(2): p. 369-82.

35. Knoblauch, H., et al., A cholesterol-lowering gene maps to chromosome 13q. Am J Hum Genet, 2000. 66(1): p. 157-66.

36. Valle, A., et al., Butyrylcholinesterase: association with the metabolic syndrome and identification of 2 gene loci affecting activity. Clin Chem, 2006. 52(6): p. 1014-20.

37. Hoffman, B.B.a.T., Palmer, Neurotransmission The Autonomic and Somatic Motor Nervous Systems, in Goodman & Gilman's The Pharmacological Basis of Therapeutics, J.G.a.L. Hardman, Lee E., Editor. 2001, McGraw-Hill: New York. p. 115-153.

38. Purves, D., et al., Neuroscience. Second ed, ed. D. Purves. 2001, Sunderland: Sinauer Associates, Inc.

39. Raffa, R.B., S.M. Rawls, and E.P. Beyzarov, Netter's Illustrated Pharmacology. First ed, ed. P. Kelly. 2005, Teterboro: Icon Learning Systems LLC.

40. Collins, R., et al., Blood pressure, stroke, and coronary heart disease. Part 2, Short-term reductions in blood pressure: overview of randomised drug trials in their epidemiological context. Lancet, 1990. 335(8693): p. 827-38.

41. Cruickshank, J.M., J.M. Thorp, and F.J. Zacharias, Benefits and potential harm of lowering high blood pressure. Lancet, 1987. 1(8533): p. 581-4.

130

42. Cutler, J.A., S.W. MacMahon, and C.D. Furberg, Controlled clinical trials of drug treatment for hypertension. A review. Hypertension, 1989. 13(5 Suppl): p. I36-44.

43. Machado, M., et al., Sensitivity of patient outcomes to pharmacist interventions. Part II: Systematic review and meta-analysis in hypertension management. Ann Pharmacother, 2007. 41(11): p. 1770-81.

44. MacMahon, S., et al., Blood pressure, stroke, and coronary heart disease. Part 1, Prolonged differences in blood pressure: prospective observational studies corrected for the regression dilution bias. Lancet, 1990. 335(8692): p. 765-74.

45. Buccafusco, J.J., The role of central cholinergic neurons in the regulation of blood pressure and in experimental hypertension. Pharmacol Rev, 1996. 48(2): p. 179-211.

46. Alves, F.H., et al., Cardiovascular effects of carbachol microinjected into the bed nucleus of the stria terminalis of the rat brain. Brain Res, 2007. 1143: p. 161-8.

47. Angelone, T., et al., The emerging cardioinhibitory role of the hippocampal cholinergic neurostimulating peptide. J Pharmacol Exp Ther, 2006. 318(1): p. 336-44.

48. Hagiwara, Y. and T. Kubo, Cholinergic systems in the posterior hypothalamic nucleus are involved in blood pressure decrease-induced excitation of anterior hypothalamic area neurons in rats. Neurosci Lett, 2005. 390(2): p. 61-5.

49. Soares, P.P., et al., Cholinergic stimulation with pyridostigmine increases heart rate variability and baroreflex sensitivity in rats. Auton Neurosci, 2004. 113(1-2): p. 24-31.

50. Kubo, T., et al., Activation of hypothalamic angiotensin receptors produces pressor responses via cholinergic inputs to the rostral ventrolateral medulla in normotensive and hypertensive rats. Brain Res, 2002. 953(1-2): p. 232-45.

51. Nobrega, A.C., et al., Enhancement of heart rate variability by cholinergic stimulation with pyridostigmine in healthy subjects. Clin Auton Res, 2001. 11(1): p. 11-7.

52. Raj, S.R., et al., Acetylcholinesterase inhibition improves tachycardia in postural tachycardia syndrome. Circulation, 2005. 111(21): p. 2734-40.

131

53. Singer, W., et al., Acetylcholinesterase inhibition: a novel approach in the treatment of neurogenic orthostatic hypotension. J Neurol Neurosurg Psychiatry, 2003. 74(9): p. 1294-8.

54. Thayer, J.F. and R.D. Lane, The role of vagal function in the risk for cardiovascular disease and mortality. Biol Psychol, 2007. 74(2): p. 224-42.

55. Tomimatsu, T., et al., Effects of neonatal hypoxic-ischemic brain injury on skilled motor tasks and brainstem function in adult rats. Brain Res, 2002. 926(1-2): p. 108-17.

56. Kubo, T., Cholinergic mechanism and blood pressure regulation in the central nervous system. Brain Res Bull, 1998. 46(6): p. 475-81.

57. Schumacher, M., et al., Primary structure of Torpedo californica acetylcholinesterase deduced from its cDNA sequence. Nature, 1986. 319(6052): p. 407-9.

58. Getman, D.K., et al., The human gene encoding acetylcholinesterase is located on the long arm of chromosome 7. Am J Hum Genet, 1992. 51(1): p. 170-7.

59. Li, Y., et al., Gene structure of mammalian acetylcholinesterase. Alternative exons dictate tissue-specific expression. J Biol Chem, 1991. 266(34): p. 23083- 90.

60. Li, Y., S. Camp, and P. Taylor, Tissue-specific expression and alternative mRNA processing of the mammalian acetylcholinesterase gene. J Biol Chem, 1993. 268(8): p. 5790-7.

61. Allderdice, P.W., et al., The cloned butyrylcholinesterase (BCHE) gene maps to a single chromosome site, 3q26. Genomics, 1991. 11(2): p. 452-4.

62. Gaughan, G., et al., Refinement of the localization of human butyrylcholinesterase to chromosome 3q26.1-q26.2 using a PCR-derived probe. Genomics, 1991. 11(2): p. 455-8.

63. Lockridge, O., et al., Complete amino acid sequence of human serum cholinesterase. J Biol Chem, 1987. 262(2): p. 549-57.

64. MacPhee-Quigley, K., P. Taylor, and S. Taylor, Primary structures of the catalytic subunits from two molecular forms of acetylcholinesterase. A comparison of NH2-terminal and active center sequences. J Biol Chem, 1985. 260(22): p. 12185-9.

132

65. Massoulie, J., et al., Molecular and cellular biology of cholinesterases. Prog Neurobiol, 1993. 41(1): p. 31-91.

66. Bauld, H.W., et al., Aetiology of prolonged apnoea after suxamethonium. Br J Anaesth, 1974. 46(4): p. 273-81.

67. Hotelier, T., et al., ESTHER, the database of the alpha/beta-hydrolase fold superfamily of proteins. Nucleic Acids Res, 2004. 32(Database issue): p. D145-7.

68. Neville, L.F., et al., Anionic site interactions in human butyrylcholinesterase disrupted by two single point mutations. J Biol Chem, 1990. 265(34): p. 20735-8.

69. Bartels, C.F., K. James, and B.N. La Du, DNA mutations associated with the human butyrylcholinesterase J-variant. Am J Hum Genet, 1992. 50(5): p. 1104-14.

70. Loewenstein-Lichtenstein, Y., et al., Genetic predisposition to adverse consequences of anti-cholinesterases in 'atypical' BCHE carriers. Nat Med, 1995. 1(10): p. 1082-5.

71. Prody, C.A., et al., De novo amplification within a "silent" human cholinesterase gene in a family subjected to prolonged exposure to organophosphorous insecticides. Proc Natl Acad Sci U S A, 1989. 86(2): p. 690-4.

72. Bartels, C.F., T. Zelinski, and O. Lockridge, Mutation at codon 322 in the human acetylcholinesterase (ACHE) gene accounts for YT blood group polymorphism. Am J Hum Genet, 1993. 52(5): p. 928-36.

73. Soreq, H. and S. Seidman, Acetylcholinesterase--new roles for an old actor. Nat Rev Neurosci, 2001. 2(4): p. 294-302.

74. Hasin, Y., et al., A paradigm for single nucleotide polymorphism analysis: the case of the acetylcholinesterase gene. Hum Mutat, 2004. 24(5): p. 408-16.

75. Ollis, D.L., et al., The alpha/beta hydrolase fold. Protein Eng, 1992. 5(3): p. 197-211.

76. Kryger, G., et al., Structures of recombinant native and E202Q mutant human acetylcholinesterase complexed with the snake-venom toxin fasciculin-II. Acta Crystallogr D Biol Crystallogr, 2000. 56(Pt 11): p. 1385-94.

133

77. Nachon, F., et al., Engineering of a monomeric and low-glycosylated form of human butyrylcholinesterase: expression, purification, characterization and crystallization. Eur J Biochem, 2002. 269(2): p. 630-7.

78. Nicolet, Y., et al., Crystal structure of human butyrylcholinesterase and of its complexes with substrate and products. J Biol Chem, 2003. 278(42): p. 41141- 7.

79. Ngamelue, M.N., et al., Crystallization and X-ray structure of full-length recombinant human butyrylcholinesterase. Acta Crystallogr Sect F Struct Biol Cryst Commun, 2007. 63(Pt 9): p. 723-7.

80. Aldridge, W.N., Enzyme Inhibitors as Substrates. 1972, Amsterdam: Elsevier.

81. Augustinsson, K.B., Cholinesterases: a study in comparative enzymology. Acta Physiol Scand, 1948. 52(Suppl 15): p. 1-182.

82. Radic, Z., et al., Three distinct domains in the cholinesterase molecule confer selectivity for acetyl- and butyrylcholinesterase inhibitors. Biochemistry, 1993. 32(45): p. 12074-84.

83. Silver, A., The Biology of Cholinesterases. 1974, Amsterdam: Elsevier.

84. Taylor, P., et al., Structural bases for the specificity of cholinesterase catalysis and inhibition. Toxicol Lett, 1995. 82-83: p. 453-8.

85. Ceveransky, C., et al., Snake Toxins, A.L. Harvey, Editor. 1991, Pergamon: New York. p. 303-321.

86. Eastman, J., et al., Fasciculin 2 binds to the peripheral site on acetylcholinesterase and inhibits substrate hydrolysis by slowing a step involving proton transfer during enzyme acylation. J Biol Chem, 1995. 270(34): p. 19694-701.

87. Radic, Z., et al., Site of fasciculin interaction with acetylcholinesterase. J Biol Chem, 1994. 269(15): p. 11233-9.

88. Sussman, J.L., et al., Atomic structure of acetylcholinesterase from Torpedo californica: a prototypic acetylcholine-binding protein. Science, 1991. 253(5022): p. 872-9.

89. Gibney, G., et al., Mutagenesis of essential functional residues in acetylcholinesterase. Proc Natl Acad Sci U S A, 1990. 87(19): p. 7546-50.

134

90. Harel, M., et al., Quaternary ligand binding to aromatic residues in the active- site gorge of acetylcholinesterase. Proc Natl Acad Sci U S A, 1993. 90(19): p. 9031-5.

91. Ordentlich, A., et al., Dissection of the human acetylcholinesterase active center determinants of substrate specificity. Identification of residues constituting the anionic site, the hydrophobic site, and the acyl pocket. J Biol Chem, 1993. 268(23): p. 17083-95.

92. Ordentlich, A., et al., Functional characteristics of the oxyanion hole in human acetylcholinesterase. J Biol Chem, 1998. 273(31): p. 19509-17.

93. Taylor, P., Anticholinesterase Agents, in Goodman & Gilman's The Pharmacological Basis of Therapeutics, J.G. Hardman Ph.D. and L.E. Limbird Ph.D., Editors. 2001, McGraw-Hill Medical Publishing Division: New York. p. 175 - 191.

94. Radic, Z. and P. Taylor, Structure and Function of Cholinesterases, in Toxicology of Organophosphate and Carbamate Compounds. 2006, Elsevier, Inc.

95. Changeux, J.P., Responses of acetylcholinesterase from Torpedo marmorata to salts and curarizing drugs. Mol Pharmacol, 1966. 2(5): p. 369-92.

96. Rosenberry, T.L. and S.A. Bernhard, Studies of catalysis by acetylcholinesterase. Synergistic effects of inhibitors during the hydrolysis of acetic acid esters. Biochemistry, 1972. 11(23): p. 4308-21.

97. Roufogalis, B.D. and E.E. Quist, Relative binding sites of pharmacologically active ligands on bovine erythrocyte acetylcholinesterase. Mol Pharmacol, 1972. 8(1): p. 41-9.

98. Taylor, P. and S. Lappi, Interaction of fluorescence probes with acetylcholinesterase. The site and specificity of propidium binding. Biochemistry, 1975. 14(9): p. 1989-97.

99. Taylor, P. and Z. Radic, The cholinesterases: from genes to proteins. Annu Rev Pharmacol Toxicol, 1994. 34: p. 281-320.

100. Shafferman, A., et al., Mutagenesis of human acetylcholinesterase. Identification of residues involved in catalytic activity and in polypeptide folding. J Biol Chem, 1992. 267(25): p. 17640-8.

101. Masson, P., et al., Asp7O in the peripheral anionic site of human butyrylcholinesterase. Eur J Biochem, 1996. 235(1-2): p. 36-48.

135

102. Saxena, A., et al., Aromatic amino-acid residues at the active and peripheral anionic sites control the binding of E2020 (Aricept) to cholinesterases. Eur J Biochem, 2003. 270(22): p. 4447-58.

103. Bourne, Y., et al., Freeze-frame inhibitor captures acetylcholinesterase in a unique conformation. Proc Natl Acad Sci U S A, 2004. 101(6): p. 1449-54.

104. Bourne, Y., P. Taylor, and P. Marchot, Acetylcholinesterase inhibition by fasciculin: crystal structure of the complex. Cell, 1995. 83(3): p. 503-12.

105. Lewis, W.G., et al., Click chemistry in situ: acetylcholinesterase as a reaction vessel for the selective assembly of a femtomolar inhibitor from an array of building blocks. Angew Chem Int Ed Engl, 2002. 41(6): p. 1053-7.

106. Mileson, B.E., et al., Common mechanism of toxicity: a case study of organophosphorus pesticides. Toxicol Sci, 1998. 41(1): p. 8-20.

107. Aldridge, W.N. and E. Reiner, Acetylcholinesterase. Two types of inhibition by an organophosphorus compound: one the formation of phosphorylated enzyme and the other analogous to inhibition by substrate. Biochem J, 1969. 115(2): p. 147-62.

108. Fukuto, T.R., Mechanism of action of organophosphorus and carbamate insecticides. Environ Health Perspect, 1990. 87: p. 245-54.

109. Radic, Z., E. Reiner, and V. Simeon, Binding sites on acetylcholinesterase for reversible ligands and phosphorylating agents. A theoretical model tested on haloxon and phosphostigmine. Biochem Pharmacol, 1984. 33(4): p. 671-7.

110. Wilson, I.B. and B. Ginsburg, A powerful reactivator of alkylphosphate- inhibited acetylcholinesterase. Biochim Biophys Acta, 1955. 18(1): p. 168-70.

111. Kovarik, Z., et al., Mutant cholinesterases possessing enhanced capacity for reactivation of their phosphonylated conjugates. Biochemistry, 2004. 43(11): p. 3222-9.

112. Taylor, P., et al., Acetylcholinesterase: converting a vulnerable target to a template for antidotes and detection of inhibitor exposure. Toxicology, 2007. 233(1-3): p. 70-8.

113. Hemingway, J., et al., The molecular basis of insecticide resistance in mosquitoes. Insect Biochem Mol Biol, 2004. 34(7): p. 653-65.

136

114. Ehrlich, G., et al., Population diversity and distinct haplotype frequencies associated with ACHE and BCHE genes of Israeli Jews from trans-Caucasian Georgia and from Europe. Genomics, 1994. 22(2): p. 288-95.

115. Evans, F.T., et al., Sensitivity to succinylcholine in relation to serum- cholinesterase. Lancet, 1952. 1(25): p. 1229-30.

116. Harris, H., Enzymes and Drug Sensitivity. the Genetics of Serum Cholinesterase 'Deficiency' in Relation to Suxamethonium Apnoea. Proc R Soc Med, 1964. 57: p. 503-6.

117. Yen, T., et al., Butyrylcholinesterase (BCHE) genotyping for postsuccinylcholine apnea in an Australian population. Clin Chem, 2003. 49(8): p. 1297-308.

118. George, S.T., et al., Aryl acylamidase activity in human erythrocyte, plasma and blood in pesticide (organophosphates and carbamates) poisoning. Clin Chim Acta, 1985. 145(1): p. 1-7.

119. Abbott, C.A., et al., Relationship between serum butyrylcholinesterase activity, hypertriglyceridaemia and insulin sensitivity in diabetes mellitus. Clin Sci (Lond), 1993. 85(1): p. 77-81.

120. Alcantara, V.M., et al., Butyrylcholinesterase activity and metabolic syndrome in obese patients. Clin Chem Lab Med, 2005. 43(3): p. 285-8.

121. Annapurna, V., et al., Relationship between serum pseudocholinesterase and triglycerides in experimentally induced diabetes mellitus in rats. Diabetologia, 1991. 34(5): p. 320-4.

122. Kutty, K.M. and R.H. Payne, Serum pseudocholinesterase and very-low- density lipoprotein metabolism. J Clin Lab Anal, 1994. 8(4): p. 247-50.

123. Rustemeijer, C., et al., Is pseudocholinesterase activity related to markers of triacylglycerol synthesis in Type II diabetes mellitus? Clin Sci (Lond), 2001. 101(1): p. 29-35.

124. Randell, E.W., et al., Relationship between serum butyrylcholinesterase and the metabolic syndrome. Clin Biochem, 2005. 38(9): p. 799-805.

125. Hashim, Y., et al., Butyrylcholinesterase K variant on chromosome 3 q is associated with Type II diabetes in white Caucasian subjects. Diabetologia, 2001. 44(12): p. 2227-30.

137

126. Johansen, A., et al., Large-scale studies of the functional K variant of the butyrylcholinesterase gene in relation to Type 2 diabetes and insulin secretion. Diabetologia, 2004. 47(8): p. 1437-41.

127. Crawford, F., et al., The butyrylcholinesterase gene is neither independently nor synergistically associated with late-onset AD in clinic- and community- based populations. Neurosci Lett, 1998. 249(2-3): p. 115-8.

128. Ghebremedhin, E., et al., Age-dependent association between butyrylcholinesterase K-variant and Alzheimer disease-related neuropathology in human brains. Neurosci Lett, 2002. 320(1-2): p. 25-8.

129. Lehmann, D.J., C. Johnston, and A.D. Smith, Synergy between the genes for butyrylcholinesterase K variant and apolipoprotein E4 in late-onset confirmed Alzheimer's disease. Hum Mol Genet, 1997. 6(11): p. 1933-6.

130. McIlroy, S.P., et al., Butyrylcholinesterase K variant is genetically associated with late onset Alzheimer's disease in Northern Ireland. J Med Genet, 2000. 37(3): p. 182-5.

131. Panegyres, P.K., et al., Butyrycholinesterase K variant and Alzheimer's disease. J Neurol, 1999. 246(5): p. 369-70.

132. Singleton, A.B., et al., No association between the K variant of the butyrylcholinesterase gene and pathologically confirmed Alzheimer's disease. Hum Mol Genet, 1998. 7(5): p. 937-9.

133. Wiebusch, H., et al., Further evidence for a synergistic association between APOE epsilon4 and BCHE-K in confirmed Alzheimer's disease. Hum Genet, 1999. 104(2): p. 158-63.

134. Yamamoto, Y., et al., Failure to confirm a synergistic effect between the K- variant of the butyrylcholinesterase gene and the epsilon4 allele of the apolipoprotein gene in Japanese patients with Alzheimer's disease. J Neurol Neurosurg Psychiatry, 1999. 67(1): p. 94-6.

135. Online Mendelan Inheritance in Man. Butyrylcholinesterase; BCHE. .

136. Steegmuller, H., On the geographical distribution of pseudocholinesterase variants. Humangenetik, 1975. 26(3): p. 167-85.

137. Online Mendelian Inheritance in Man. Cholinesterase, serum, 2; ChE2. .

138

138. Akizuki, S., et al., Genetic and immunological analyses of patients with increased serum butyrylcholinesterase activity and its C5 variant form. Clin Chem Lab Med, 2004. 42(9): p. 991-6.

139. Whitfield, J.B., et al., Smoking, obesity, and hypertension alter the dose- response curve and test sensitivity of carbohydrate-deficient transferrin as a marker of alcohol intake. Clin Chem, 1998. 44(12): p. 2480-9.

140. Ellman, G.L., et al., A new and rapid colorimetric determination of acetylcholinesterase activity. Biochem Pharmacol, 1961. 7: p. 88-95.

141. Miller, S.A., D.D. Dykes, and H.F. Polesky, A simple salting out procedure for extracting DNA from human nucleated cells. Nucleic Acids Res, 1988. 16(3): p. 1215.

142. Beekman, M., et al., Heritabilities of apolipoprotein and lipid levels in three countries. Twin Res, 2002. 5(2): p. 87-97.

143. Nyholt, D.R., et al., Genomewide significant linkage to migrainous headache on chromosome 5q21. Am J Hum Genet, 2005. 77(3): p. 500-12.

144. Abecasis, G.R., et al., Genomewide scan in families with schizophrenia from the founder population of Afrikaners reveals evidence for linkage and uniparental disomy on chromosome 1. Am J Hum Genet, 2004. 74(3): p. 403- 17.

145. Lander, E. and L. Kruglyak, Genetic dissection of complex traits: guidelines for interpreting and reporting linkage results. Nat Genet, 1995. 11(3): p. 241- 7.

146. Shete, S., et al., Effect of winsorization on power and type 1 error of variance components and related methods of QTL detection. Behav Genet, 2004. 34(2): p. 153-9.

147. Whitby, L.G., F.L. Mitchell, and D.W. Moss, Quality control in routine clinical chemistry. Adv Clin Chem, 1967. 10: p. 65-156.

148. Thamer, C., et al., Elevated serum GGT concentrations predict reduced insulin sensitivity and increased intrahepatic lipids. Horm Metab Res, 2005. 37(4): p. 246-51.

149. Rantala, A.O., et al., Gamma-glutamyl transpeptidase and the metabolic syndrome. J Intern Med, 2000. 248(3): p. 230-8.

139

150. Kutty, K.M., S.N. Huang, and K.T. Kean, Pseudocholinesterase in obesity: hypercaloric diet induced changes in experimental obese mice. Experientia, 1981. 37(11): p. 1141-2.

151. Popescu, T.A., et al., Serum pseudocholinesterase activity during experimental fattening. Med Interne, 1976. 14(1): p. 71-3.

152. Higashi, Y. and M. Yoshizumi, New methods to evaluate endothelial function: method for assessing endothelial function in humans using a strain-gauge plethysmography: nitric oxide-dependent and -independent vasodilation. J Pharmacol Sci, 2003. 93(4): p. 399-404.

153. McQueen, M.B., et al., A QTL genome scan of the metabolic syndrome and its component traits. BMC Genet, 2003. 4 Suppl 1: p. S96.

154. Olswold, C. and M. de Andrade, Localization of genes involved in the metabolic syndrome using multivariate linkage analysis. BMC Genet, 2003. 4 Suppl 1: p. S57.

155. Heijmans, B.T., et al., Meta-analysis of four new genome scans for lipid parameters and analysis of positional candidates in positive linkage regions. Eur J Hum Genet, 2005. 13(10): p. 1143-53.

156. Suh, T.H., Wang, C.H., and Lim, R.K.S., The effect of intracisternal application of acetylcholine and the localization of the pressor centre and tract. Chin. J. Physiol., 1936. 10: p. 61-79.

157. Brezenoff, H.E. and Y.F. Xiao, Acetylcholine in the posterior hypothalamic nucleus is involved in the elevated blood pressure in the spontaneously hypertensive rat. Life Sci, 1989. 45(13): p. 1163-70.

158. Kubo, T., et al., Enhanced release of acetylcholine in the rostral ventrolateral medulla of spontaneously hypertensive rats. Brain Res, 1995. 686(1): p. 1-9.

159. Vargas, H.M. and H.E. Brezenoff, Suppression of hypertension during chronic reduction of brain acetylcholine in spontaneously hypertensive rats. J Hypertens, 1988. 6(9): p. 739-45.

160. Khan, I.M., et al., Nicotinic receptor gene cluster on rat chromosome 8 in nociceptive and blood pressure hyperresponsiveness. Physiol Genomics, 2002. 11(2): p. 65-72.

161. Irisawa, H., H.F. Brown, and W. Giles, Cardiac pacemaking in the sinoatrial node. Physiol Rev, 1993. 73(1): p. 197-227.

140

162. Nakahara, T., et al., Cholinesterase affects dynamic transduction properties from vagal stimulation to heart rate. Am J Physiol, 1998. 275(2 Pt 2): p. R541- 7.

163. Cockburn, M., et al., The occurrence of chronic disease and other conditions in a large population-based cohort of native Californian twins. Twin Res, 2002. 5(5): p. 460-7.

164. Zhang, L., et al., Functional allelic heterogeneity and pleiotropy of a repeat polymorphism in tyrosine hydroxylase: prediction of catecholamines and response to stress in twins. Physiol Genomics, 2004. 19(3): p. 277-91.

165. Ferraris, A., et al., Pyrosequencing for detection of mutations in the connexin 26 (GJB2) and mitochondrial 12S RNA (MTRNR1) genes associated with hereditary hearing loss. Hum Mutat, 2002. 20(4): p. 312-20.

166. Ronaghi, M., Pyrosequencing sheds light on DNA sequencing. Genome Res, 2001. 11(1): p. 3-11.

167. Stephens, M., N.J. Smith, and P. Donnelly, A new statistical method for haplotype reconstruction from population data. Am J Hum Genet, 2001. 68(4): p. 978-89.

168. Stephens, M. and P. Donnelly, A comparison of bayesian methods for haplotype reconstruction from population genotype data. Am J Hum Genet, 2003. 73(5): p. 1162-9.

169. Worek, F., et al., Improved determination of acetylcholinesterase activity in human whole blood. Clin Chim Acta, 1999. 288(1-2): p. 73-90.

170. Marchot, P., et al., Soluble monomeric acetylcholinesterase from mouse: expression, purification, and crystallization in complex with fasciculin. Protein Sci, 1996. 5(4): p. 672-9.

171. Wang, W. and B.A. Malcolm, Two-stage PCR protocol allowing introduction of multiple mutations, deletions and insertions using QuikChange Site- Directed Mutagenesis. Biotechniques, 1999. 26(4): p. 680-2.

172. Livak, K.J. and T.D. Schmittgen, Analysis of relative gene expression data using real-time quantitative PCR and the 2(-Delta Delta C(T)) Method. Methods, 2001. 25(4): p. 402-8.

173. Wang, X. and B. Seed, A PCR primer bank for quantitative gene expression analysis. Nucleic Acids Res, 2003. 31(24): p. e154.

141

174. Berman, J.D. and M. Young, Rapid and complete purification of acetylcholinesterases of electric eel and erythrocyte by affinity chromatography. Proc Natl Acad Sci U S A, 1971. 68(2): p. 395-8.

175. Camp, S., et al., Acetylcholinesterase (AChE) gene modification in transgenic animals: functional consequences of selected exon and regulatory region deletion. Chem Biol Interact, 2005. 157-158: p. 79-86.

176. Bourne, Y., et al., Structural insights into ligand interactions at the acetylcholinesterase peripheral anionic site. Embo J, 2003. 22(1): p. 1-12.

177. Berman, H.M., et al., The Protein Data Bank. Nucleic Acids Res, 2000. 28(1): p. 235-42.

178. De Jaco, A., et al., A mutation linked with autism reveals a common mechanism of endoplasmic reticulum retention for the alpha,beta-hydrolase fold protein family. J Biol Chem, 2006. 281(14): p. 9667-76.

179. Kovarik, Z., et al., Acetylcholinesterase active centre and gorge conformations analysed by combinatorial mutations and enantiomeric phosphonates. Biochem J, 2003. 373(Pt 1): p. 33-40.

180. Reiner, E. and Z. Radic, Mechanism of action of cholinesterase inhibitors, in Cholinsterases and Cholinesterase Inhibitors, P. Ezio Giacobini Md, Editor. 2000, Martin Dunitz Ltd.: London. p. 103-119.

181. Taylor, P., et al., Application of recombinant DNA methods for production of cholinesterases as organophosphate antidotes and detectors. Arh Hig Rada Toksikol, 2007. 58(3): p. 339-45.

182. Gelernter, J., et al., Genomewide linkage scan for nicotine dependence: identification of a chromosome 5 risk locus. Biol Psychiatry, 2007. 61(1): p. 119-26.

183. De Jaco, A., S. Cam, and P. Taylor, Influence of the 5' intron in the control of acetylcholinesterase gene expression during myogenesis. Chemico-Biological Interactions, 2005. 157-158: p. 372-373.