<%BANNER%>

Alpha 1A- and Beta 2-Adrenoceptor Gene-Expression Differences in Hypertensive and Normotensive Persons by Race


PAGE 1

ALPHA 1AAND BETA 2-ADRENOCEPTOR GENE EXPRESSION DIFFERENCES IN HYPERTENSIVE AND NORMOTENSIVE PERSONS BY RACE By JENNIFER RENE’ DUNGAN A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLORI DA IN PARTIAL FULFILLMENT\ OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY UNIVERSITY OF FLORIDA 2006

PAGE 2

Copyright 2006 by Jennifer Rene’ Dungan

PAGE 3

This document is dedicated to my husband, Crai g for his endless love and support, to my Great-grandmother, Margaret Gray for in spiring me to become a nurse, to my Grandmother, Julia Chodzinski, who has show n me the power of perseverance and the importance in believing in my dreams, and fi nally, to my mother, Cheryl Crossland, who instilled in me a strong sense of hard-wor k and determination, and who has given me confidence when I needed it most.

PAGE 4

iv ACKNOWLEDGMENTS I gratefully acknowledge the contributions guidance, and enc ouragement of my dissertation committee Chair, Carolyn Yucha, PhD, and members, Julie Johnson, Pharm D, Yvette Conley, PhD, Shawn Kneipp, PhD, and Taimour Langaee, PhD. I also extend my apprecia tion to the numerous people who assisted me with the completion of this project in many important ways: The UF and VA TCV Surgery Departments; the staff at the UF ICBR fac ilities; and my colleagues and peers in the College of Nursing. In addition, I extend my warmest thanks to my family and friends for their support throughout this process. Special thanks go to Mandy Elliott for living this experience with me and being the most s upportive friend anyone c ould have throughout this process. Furthermore, I would like to acknowledge the funding agencies that financially supported this project: The Nati onal Institute of Nursing Re search, the American Heart Association, and the Alpha Theta Chapter of Sigma Theta Tau International Nursing Honor Society. Finally, I gratefully acknowle dge the participants of this study for their important contributions to the success of this project.

PAGE 5

v TABLE OF CONTENTS page ACKNOWLEDGMENTS.................................................................................................iv LIST OF TABLES...........................................................................................................viii LIST OF FIGURES.............................................................................................................x ABSTRACT......................................................................................................................x ii CHAPTER 1 INTRODUCTION........................................................................................................1 Background and Problem Statement............................................................................1 Purpose of the Study.....................................................................................................6 Hypotheses....................................................................................................................7 Definitions of Terms.....................................................................................................8 Assumptions.................................................................................................................9 Significance of the Study............................................................................................10 2 REVIEW OF LITERATURE.....................................................................................16 Introduction.................................................................................................................16 Genetic Influences on E ssential Hypertension...........................................................16 Genetic Models of Hypertension................................................................................17 Cardiovascular Reactivity, HTN and Adrenergic Receptors......................................18 Alpha 1Aand Beta 2-Adrenergic Receptors.............................................................19 Pharmacotherapeutic Aspects of Alpha-1 A and Beta-2 Adrenergic Receptors.........21 Population Differences in HTN..................................................................................22 Recommendations for the Collecti on of Racial and Ethnic Data........................25 Implications for Studying Disease by Race.........................................................28 Positive Inotrope Administration................................................................................33 Genetic Analysis Techniques.....................................................................................33 Collection of Human Tissues fo r Gene Expression Analyses....................................38 Summary.....................................................................................................................39 3 PROCEDURES AND METHODS............................................................................40 Introduction.................................................................................................................40

PAGE 6

vi Design......................................................................................................................... 40 Subject Recruitment....................................................................................................41 Consenting Process and HIPAA Regulations......................................................43 Setting..................................................................................................................44 Research Variables.....................................................................................................44 Study Protocol............................................................................................................46 Data Collection and Laboratory Methods...........................................................46 Blood and tissue collection..........................................................................46 Genomic DNA analyses...............................................................................47 RNA isolation and reverse-transcription......................................................49 Real-time polymerase chain reaction...........................................................52 Positive inotrope data collection..................................................................53 Calculations for Relative Gene Expr ession and Selecti on of Calibrator.............54 Methods for Statistical Analyses................................................................................55 4 RESULTS...................................................................................................................57 Introduction.................................................................................................................57 Descriptive Results.....................................................................................................57 Subject Demographics.........................................................................................57 Assessment of GAPDH for Relative Quantitation..............................................61 Assumptions of Normality..................................................................................64 Analytic Results for Hypotheses.........................................................................64 Exploratory Aims................................................................................................70 Effect Sizes and Power Calculations...................................................................76 5 DISCUSSION AND RESULTS.................................................................................78 Introduction.................................................................................................................78 Discussion of Results..................................................................................................78 Demographics......................................................................................................78 Gene Expression Measures of Central Tendency and Variance.........................81 Discussions for Choice of GA PDH for Normalization Gene..............................82 Assessment of the Performance of GAPDH as a Normalizer.............................83 Aims....................................................................................................................87 Exploratory Aims................................................................................................90 Sample versus population alle le frequency comparisons.............................90 Sample versus population genotype frequency comparisons.......................91 Limitations..................................................................................................................94 Normalization with GAPDH...............................................................................94 Power...................................................................................................................94 Internal Validity...................................................................................................95 Construct Validity of the Variable, Normotension..............................................96 External Validity.................................................................................................96 Minimal Sample Template..................................................................................96 Confounding Variables........................................................................................97 Nursing Relevance......................................................................................................99

PAGE 7

vii Practiceand Care-Related Relevance..............................................................100 Nurse-Directed Research and Qualitative Findings..........................................102 Case Study.........................................................................................................103 Summary............................................................................................................104 Recommendations for Future Research....................................................................105 Conclusions...............................................................................................................106 APPENDIX A SUBJECT ENROLLMENT /DEMOGRAPHIC FORM...........................................109 B UF IRB-01 INFORMED CONSENT FORM...........................................................112 C VA SCI INFORMED CONSENT FORM................................................................122 D TAQMAN REAL-TIME PCR AMPLIFICATION PLOTS....................................132 LIST OF REFERENCES.................................................................................................133 BIOGRAPHICAL SKETCH...........................................................................................143

PAGE 8

viii LIST OF TABLES Table page 2-1 Adrenergic cardiovascular stress patterns................................................................23 3-1 Inclusion and exclusion criteria with rationale........................................................42 3-2 Genotyping primers..................................................................................................49 3-3 Target gene assay information.................................................................................53 3-4 Single-plex plate set-up, one sample........................................................................53 4-1 Demographics of all enrolled subjects.....................................................................58 4-2 Clinical characteristics of all enrolled subjects........................................................58 4-3 Demographics for ge ne expression subset...............................................................60 4-4 Clinical characteristics for gene expression subset..................................................60 4-5 Gene expression medians and IQRs for total sample...............................................64 4-6 Gene expression medians and IQRs for subjects by diagnosis................................65 4-7 Median fold differences in gene expression between normotensive and hypertensive subjects and Mann-Whitney U tests...................................................65 4-8 Gene expression medians, IQRs, and minimum and maximum values for White/Caucasian subjects.........................................................................................66 4-9 Median fold differences in ge ne expression between White/Caucasian normotensive and hypertensive subjects and Mann-Whitney U tests......................67 4-10 Gene expression medians and IQRs for Black/AA subjects....................................67 4-11 Median fold differences in ge ne expression between White/Caucasian hypertensive and Black/AA hypertensive s ubjects and Mann-Whitney U tests......68 4-12 Median, IQR, minimum and maximum values for 1A-ADR and 2-ADR fold difference in gene expression and need for post-operative positive inotrope medication................................................................................................................69

PAGE 9

ix 4-13 Fold differences in gene expression between non-inotrope an d inotrope subjects and Mann-Whitney U tests.......................................................................................70 4-14 Allele frequencies for populati on versus sample, by SIR/ancestry..........................71 4-15 Genotype frequencies for populait on versus sample, by SIR/ancestry....................71 4-16 Association between genotype a nd diagnoses of NT and HTN for the 1A-ADR and 2-ADR genes....................................................................................................72 4-17 Fisher’s Exact for genotype differenc es in White/Caucasian hypertensive vs. normotensive subjects..............................................................................................74 4-18 Chi-square for allele counts by diagnosis for the 1A-ADR and 2-ADR genes in all subjects................................................................................................................74 4-19 Chi-square for alleles by diagnosis for the 1A-ADR and 2-ADR genes in White/Caucasian subjects.........................................................................................75 4-20 Kruskal–Wallis tests for genotype counts by gene expression 1A-ADR and 2ADR genes...............................................................................................................76 4-21 Power and effect sizes by aim..................................................................................76

PAGE 10

x LIST OF FIGURES Figure page 1-1 Abbreviations used.....................................................................................................1 2-1 Chromosome (Ensembl human map view) showing th e locations of both ADR genes.........................................................................................................................2 0 2-2 Amplification of gene expression using TaqMan Real Time RT PCR....................37 2-3 TaqMan RT-PCR steps (adapted from Bustin, 2000)..............................................38 3-1 1A -ADR gene with promoter, intron and exon boundaries and investigated polymorphism...........................................................................................................45 3-2 2-ADR gene with promoter, exon bounda ry and investigated polymorphisms......46 3-3 Tissue pieces immersed in RNAlater preservation solution....................................47 3-4 Blotting tissue on Kimwipe......................................................................................51 3-5 Grinding tissue in mortar and pestle on liquid nitrogen...........................................51 3-6 Powdered tissue in mortar........................................................................................51 3-7 Homogenizing tissue slush w ith rotar-stator homogenizer......................................52 4-1 Range of average duplicate Ct values of GAPDH per sample number...................61 4-2 Range of average duplicate GAPDH Ct measurements grouped by plate number..62 4-3 Boxplot for average duplicate GAPDH by diagnosis..............................................63 4-4 Boxplot for all gene ex pression raw Ct values.........................................................63 4-5 Boxplots for both gene’s expression by diagnosis...................................................65 4-6 Boxplots for White/Caucasian subjects, for both gene’s expression by diagnosis..67 4-7a Boxplots for 1A-ADR gene expression for White/Caucasian HTN versus Black/AA HTN subjects...........................................................................................68

PAGE 11

xi 4-8 Boxplots for 2-ADR gene expression for White/Caucasian HTN versus Black/AA HTN subjects...........................................................................................68 4-9 Boxplots for both genes’ expressi on by need for post-operative positive inotrope.....................................................................................................................69 4-10 Bar chart of 1A-ADR, codon 347 by diagnosis.......................................................72 4-11 Bar chart of 2-ADR, codon 16 by diagnosis...........................................................73 4-12 Bar chart of 2-ADR, codon 27 by diagnosis...........................................................73 D-1 TaqMan Real-time amplification plot for each gene.............................................132

PAGE 12

xii Abstract of Dissertation Pres ented to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Doctor of Philosophy ALPHA 1AAND BETA 2-ADRENOCEPTOR GENE EXPRESSION DIFFERENCES IN HYPERTENSIVE AND NORMOTENSIVE PERSONS BY RACE By Jennifer Rene’ Dungan May 2006 Chair: Carolyn B. Yucha Major Department: Nursing It has been hypothesized that genes of the adrenergic receptor (ADR) system contribute to hypertension (HTN). This notion is supported by genetic (gene-knockout and association), physiological and pharmacological studies of ADRs in animal and human models. The 1Aand 2-subtypes are two of nine ADRs. Briefly, vascular contraction is mediated by the 1A-ADR, whereas vascular dilation is mediated by the 2ADR. Gene expression studies of these subtypes in animal strains (particularly between normotensive and hypertensive strains) suggest an important role in the development of HTN; however, studies of this nature have not been conduc ted with human tissues in a between-groups design. This st udy explores the feasibility of conducting such a study in humans and the relative ge ne expression differences of the two aforementioned ADR genes in people with and without HTN and e xplores the impact of self-identified race. Gene expression refers to tr anscription of ribonucleic aci d (RNA) from deoxyribonucleic

PAGE 13

xiii acid (DNA). This process is a necessary step in the making of proteins. Gene expression is influenced both by genetic and envir onmental factors durin g transcription. Relative levels of RNA of the 1Aand 2-ADR genes were measured in arterial tissue samples obtained from 41 subjects who had coronary artery bypass surgery at either Shands at Alachua General Hospital or the Malcom V. Randall Veterans Hospital. Subjects were grouped according to the diagnosis of HTN ( n = 24) or NT ( n =17), as defined by national guidelines. During surger y, a small amount (10-30 mg) of normallydiscarded internal mammary artery tissue wa s provided to the rese archer, processed, and analyzed with Real-Time, reverse-transcription polymerase chain reaction to obtain relative quantitation of gene expression. Hypertensive subjects showed 3.92a nd 2.05-fold differences in relative 1Aand 2-ADR gene expression (respectively) co mpared to normotensives (statistically significant with alpha of 0.05), with hypertensi ves demonstrating reduced expression of both genes. Fold differences for both ADR subtypes remained significant when comparing White/Caucasian hypertensive versus normotensive subjects. Further exploratory aims produced some signifi cant findings. This study experienced methodological issues with the reference gene thereby affecting accuracy of relative gene expression quantitation; therefore, interpretation of re sults is cautioned.

PAGE 14

1 CHAPTER 1 INTRODUCTION This chapter will introduce the main research problem and background and delineate the hypotheses to be tested. The definitions of variables, major terms, assumptions, and significance of th e study will also be presented. Figure 1-1. Abbreviations used. Background and Problem Statement Just over 26% of adults worldwide (approximately 972 million adults) have hypertension (HTN). Essential HTN is synonymous with “primary” HTN, in which the cause of the high blood pressure (BP) is unknow n. This type represents 90-95% of all cases of HTN (American Heart Associati on [AHA], 2005). Alternately, secondary HTN accounts for the other 5-10% of all cases. Fo r these cases, the cause is known and often AA = African American ADR = adrenergic receptor, or adrenoceptor 1A-ADR = adrenergic receptor, Alpha 1A-subtype 2-ADR = adrenergic receptor, Beta2-subtype BP = blood pressure CABG = coronary artery bypass graft cDNA = complementary DNA CVR = cardiovascular reactivity DBP = diastolic blood pressure HTN = hypertension IMA = left internal mammary artery mRNA = messenger RNA NT = normotension PCR = polymerase chain reaction RNA = ribonucleic acid RT = reverse transcription SBP = systolic blood pressure SIR = self-identified race T2DM = type 2 diabetes TCV = Thoracic and Cardiovascular Surgery TPR = total peripheral resistance VAMC = Veterans Administration Medical Center

PAGE 15

2 correctable. Renal dise ase (of various types) is the most common cause of secondary HTN. Secondary HTN can also arise from si ngle-gene disorders such as glucocorticoid remediable aldosteronism, the most common au tosomal dominant form of inherited HTN. Secondary causes can also be conditions (s uch as pregnancy or stress) that, when corrected, bring BP back to normal levels. Heritability (h2) is the ratio of additive geneti c variance to total phenotypic variance. It can be thought of as the amount of variation in high BP attributable to the variation in our genetic make up. Pedigree, twin and sibling studies have di scerned that the heritable portion of essential HTN is approximately 30% (Ambler & Brown, 1999). More recent data suggest that BP traits such as SBP and DBP have high estimated heritability at about 72% and 63%, respectively (Zeegers, Rijs dijk, Sham, Fagard, Gielen, de Leeuw, et al., 2004). This could mean that the estimated heritability of HTN is actually higher than we previously thought; or that this particul ar study had inflated values due to study design and/or analyses. Furthermore, the recurrence risk of HTN increases as the number of parents with HTN increases, so that an offspring has approximately 4% chance of developing HT N with no hypertensive parents, a 10-20% risk with one hypertensive parent and the risk increases to 25-45% when both parents are hypertensive (Lucassen, 1999). These findings po int to an obvious link to genetics in explaining some of the variance in HBP. HTN is known as the “silent killer,” as few, if any, symptoms are noted by its sufferers. It is widely accepted that essentia l HTN is multifactorial, developing as a result of multiple genes and multiple environmental f actors, their interactions producing altered homeostasis of BP regulation in th e body. Furthermore, HTN has a complex

PAGE 16

3 pathophysiology involving the cardiovascular renal, endocrine, neurohumoral, and immune systems. Within these systems are subsystems that contribute to the grand schema of developing HTN, each mechanism having a number of genetic components. These include the renin-angiotensin-aldos terone (RAA) system, the angiotensin converting enzyme (ACE), the sodium bala nce either by the kidneys or by hormonal influences, or the vascular system, to name a few. Nearly every mechanism has its own candidate gene(s) for HTN. Some examples are those coding for the following proteins: renin, angiotensinogen, angiotensin I and II, angiotensin-conver ting enzyme, atrial natriuretic peptide (B and C t ypes), nitric oxide (inducible an d endothelial), endothelins, dopamine, kallikrein, adducing -subunit, and adrenergic receptors (ADRs). The ADRs are particularly important in re gulating BP. They are the main binding sites for the catecholamines epinephrine and norepinephrine, which work in delicate balance to regulate vasodilation and vasoc onstriction. These vasomotor reactions can influence the rising and falling of BP levels in the body. The ADRs can also mediate BP regulation through renal sodium excretion and release of renin from the juxtaglomerular cells in the kidneys (DiBona, 1989). Decades of research have shown that ADRs are important in the regulation of BP in humans and animals, a nd that alterations in ADRs at the cellular and genetic levels may lead to HTN. Various functional differences in ADRs have been reported between normotensive a nd hypertensive humans and animals. Of novel interest are recent anim al studies of gene expressi on differences in two specific ADRs in HTN: the 1Aand 2-subtypes. Both subtypes are involved in vasomotor tone via expression in the arteries and veins and both are implicated in the grand schema of HTN. Each has a gene that codes for its re ceptor protein. How thes e genes are expressed

PAGE 17

4 in the tissues can lead us to important information about their role in HTN. Gene expression of these ADR subtypes can be m easured by their messenger ribonucleic acid (mRNA) levels found in the tissues where they are present. The mRNA levels provide us with direct information about the level of transcri ption of the genes th at code for these subtypes. Current technology allows us to pres erve tissue samples in such a way that we can accurately measure this mRNA (or level of transcription) and compare these levels between groups of people (for example, be tween people with normotension (NT) versus HTN). (More detailed background information on these subtypes and gene expression is provided in Chapter 2: Revi ew of the Literature.) The previous paragraph explained that differences in ADR exist between hypertensive and normotensive humans and animal strains; that two specific subtypes have been implicated in the pathophysio logy of HTN; and that examining gene expression of these genes may provide a novel insight into one aspect of the disease process. ADR differences in HTN have also been reported among se lf-identified races and ethnicities. While it is accepted that ther e is great interindiv idual variability among people with regard to ADR function, expression, physiologic response and pharmacologic response (Small, McGraw, & Li ggett, 2003), potential racial and ethnic differences attract attention because of the disproportionate statistics regarding hypertensive disease in racial and et hnic subpopulations. Disease prevalence, management, morbidity and mortality among the black or African American (AA) population are particularly problematic because AAs exhibit the highest rate of HTN and the worst health outcomes in regards to morb idity and mortality in the U.S. The possible explanations of race-specific differences in health and disease outcomes are at the center

PAGE 18

5 of great debate. Possible variab les include socioeconomic factor s as well as differences in pathophysiologic mechanisms, pharmacologic responses, and recently genetic variability. In the ADR literature, there are reports of adrenergic-specific differences in cardiovascular reactivity (C VR) within black/AA populat ions (McAdoo, Weinberger, Miller, Fineberg, & Grim, 1990; St ein, Lang, Singh, He, & Wood, 2000; Knox, Hausdorff, & Markovitz, 2002) and adrenergic-s pecific diversity in medication response in AAs (Humphreys & Delvin, 1968; Jenni ngs & Parsons, 1976; Seedat, 1980; Cushman, Reda, Perry, Williams, Abdellatif, & Materson, 2000). The estimated heritability for HTN in pe ople of sub-Saharan African descent is 4568% (Rotimi, Cooper, Cao, Ogunbiyi, Ladipo et al., 1999; Gu, Borecki, Gagnon, Bouchard, Leon, Skinner, et al., 1998). While this heritability estimate is specific to people having origins of the sub-Saharan region of Africa, which is not generalizable to anyor everyone having origin s in Africa, it warrants furt her investigation into genetic sub-population differences. Genetic studies of as sociation have used self-identified race (SIR) as a variable. Reports of racial differences in allele frequency of ADR polymorphisms exist (Hindorff, Heckbert, Ps aty, Lumley, Siscovic k, Herrington, et al., 2005; Xie, Kim, Stein, Gainer, Brown, & Wo od, 1999). If all of these aforementioned ADR differences (cellular, f unctional, pathophysiologic, a nd pharmacologic) truly exist between HTN and NT people and among SIRs and ethnicities of hypertensives, can genetics explain these differences? Relatedl y, do differences exist regarding how these ADR genes are expressed in the vascular tissu e? These are some expl oratory issues that will be addressed by this study.

PAGE 19

6 Purpose of the Study The purpose of this study is to examin e the relationships among HTN and gene expression of the 1Aand 2-adrenergic receptors (ADRs) in the human population, and to explore if SIR helps to explain some of the differences. The study will address three specific aims: Specific aim 1: To quantify differences in gene expression of 1A-ADR and 2ADR in the internal mammary artery (IMA) between subjects with normotension (NT) and HTN. To quantify relative differences in 1A-ADR gene expression between study groups with NT and HTN. To quantify relative differences in 2-ADR gene expression between study groups with NT and HTN. Specific aim 2: To explore relative differen ces in gene expression of 1A-ADR and 2-ADR in the IMA between subjects with NT and HTN by SIR. To explore relative differences in 1Aand 2-ADR gene expression between White/Caucasian subjects with NT and HTN. To explore relative differences in the 1Aand 2-ADR gene expression between White/Caucasians with HTN versus Black/AAs with HTN. Specific aim 3: To explore the relationship between level of 1Aand 2-ADR gene expression and need for post-operative positiv e inotropic medication administration. Exploratory aim 1 (E1): To explore the associati on between diagnosis of HTN and 1Aand 2-ADR genotypes. Three genotypes will be explored: 1A-ADR (Codon 347, refSNP ID:1048101), 2-ADR (Codon 16, refSNP ID:1042713 and Codon 27, refSNP ID: 1042714) for their association with HTN. E1-1: To explore the impact of SIR on genotype by diagnosis association.

PAGE 20

7 Exploratory aim 2 (E2): To explore the association between genotype and gene expression. E2-1. To explore the association between 1A-ADR (Codon 347) single nucleotide polymorphism (SNP) and 1A-ADR gene expression. E2-2. To explore the association between 2-ADR (Codon 16) SNP and 2-ADR gene expression. E2-3. To explore the association between 2-ADR (Codon 27) SNP and 2-ADR gene expression. Hypotheses For all specific aims, the null hypothesis that no statistically signi ficant differences exist between groups will be tested. In keepi ng with the neurohumoral model of HTN, if 1A-ADRs contribute to vasonconstriction and 2-ADRs contribute to vasodilation, it could be hypothesized that subjects with HTN would display greater levels of 1A-ADRs and lower levels of 2-ADR gene expression than NT s ubjects; however, as the reverse phenomenon can also lead to HTN via regu latory feedback loops (Anderson, McNeilly, & Myers, 1992), a unidirectional hypothesi s is not appropriate Similarly, racial differences in cardiovascular reactivity patterns could possibl y support directional hypotheses of AAs showing greater 1A-ADR gene expression than Caucasians, and Caucasians showing greater 2-ADR gene expression than AAs; however, there is not enough evidence in the l iterature to support a unidirectional hypothesi s at this time. For E1, HTN is expected to be positively associ ated with each of the three ADR SNPs, based both on previous findings and the compelling ev idence supporting the ro le of these SNPs in the disease process of HTN (Small, Mc Graw, Liggett, 2003). For E2 (E2-1 through E2-3), it is hypothesized that gene expression may be affected by the variants in SNPs in

PAGE 21

8 the gene, so that variations in genotypes may affect gene expression; however, the direction of this relationshi p is not established enough fo r unidirectional hypotheses. Definitions of Terms Terms discussed in this study are defined as below. Hypertension – Defined by the Seventh Report of the Joint National Committee on Prevention, Detection, Evaluati on, and Treatment of High Blood Pressure (JNC VII) as having three consecutive BP readings of 140/90 mmHg or greater, the diagnosis of high BP by a health care practitioner, or taking antihypertensive medications specifically for BP control (JNC VII, 2003). Normotension – Having BP below 130/90 mmHg, having never been diagnosed with high BP, and not taking antihypertensive medications specifically for BP control, also defined by the JNC VII (2003). Adrenergic receptors A group of nine G-protein rece ptors from the super family of cell-surface receptors that signal the symp athetic nervous system in response to the need for BP homeostasis. 1A-ADRs contribute to vasoconstriction. 2-ADRs contribute to vasodilatation. Cardiovascular reactivity A complex cardiovascular trait in which individual cardiac and vascular responses to physiological and psycholog ical stressors may lead to changes in systolic blood pressure (SBP), di astolic blood pressure (D BP), total peripheral resistance (TPR), and other hemodynamic meas urements that represent vascular response to maintain cardiovascular homeostasis. Gene expression Process by which genes are “e xpressed” in the body. DNA is a double-stranded sequence of nucleotides th at codes for proteins. DNA strands are transcribed (or copied), making si ngle-strand messenger RNA (mRNA). The

PAGE 22

9 transcription process begins at the promot er region of the gene. The mRNA template produced by transcription is then translated into proteins The levels of mRNA found in a biological specimen are indicators of the le vel of transcription. How the genes are expressed (or how much transcription is ta king place) can be informative of how much the gene is functioning or how much is be ing copied to produce specific proteins. Race – Used in this study as a self-identificatio n of one of 5 catego ries that reflects their geographic origin based on the corresponding regions and populations listed in the groups. Generally, cultural aspects of affiliatio ns with these groups of origin is implied. These categories are set by the U.S. Office of Management and Budget (OMB) and are meant to reflect population-specific self-identif ication, not skin colo r. (See Appendix A.) This variable is referred to as “self-identifie d race,” or SIR for shor t. See sections titled “Recommendations for the collection of raci al and ethnic data” a nd “Implications for studying disease by race” in Chapter 2. Assumptions This study focuses only on essential HTN. The first major assumption is that essential HTN is a multifactorial disease process with multiple genetic and environmental factors that are likely to in teract. While many models of HT N exist, this study focuses on the neurohumoral model of HTN. The second major assumption is that the neurohumoral model plays a major role in the pathophysiology of HTN. Th is model concentrates on the importance of activation of the sympathetic nervous system and neurohumoral substances (namely epinephrine and norepinephrin e). When released, these endogenous catecholamines interact with and -ADRs to elicit a cascade of cellular membrane and intracellular events (Berecek & Carey, 2003) that affect the cardiovascular system. A third assumption is that mRNA levels are in dicative of receptor regulation and a fourth

PAGE 23

10 assumption is that receptor regulation impacts disease mechanisms at the receptor and/or cellular level. It is already known that the “transcripti on rate and steady-state level of ADR messenger RNA” is modified when the receptors are stimulated (2003, p. 3). This is yet to be confir med in regards to -ADRs, but induction of tr anscription is likely to play a similar role in regulating these recep tors. Finally, a fifth assumption is that RTPCR only quantitates steady state mRNA levels and therefore only a “snapshot in time” (Bustin, 2002). Furthermore, th ese levels may not reflect le vels of protein produced by the cell (Gygi, Rochon, Franza, & Aebersold, 1999). Significance of the Study HTN is one of the most prevalent chronic diseases in the United States (U. S.). The AHA reports there are an estimated 65,000,000 Amer icans over age 6 and 1 in 3 adults that have HTN (2005). Although HTN is easily detected and usually controllable, the cause of 90-95% of cases is unknown (AHA, 2005). Economic costs of hypertension in the U.S. are estimated to total $59.7 bill ion in 2005 (AHA, 2005). Vascular-related comorbidites of HTN include diabetes, peripheral vascular disease, and stroke. In short, HTN and its vascular consequences have major impacts on our society's health and economy. HTN’s complex pathophysiology lead s to a complex phenotype with many clinical variations. Its silent nature and disease complexity often result in poor rates of diagnosis and control. This is evident by a cont rol rate of merely 34% in all known hypertensives (JNC VII, 2003). Overall, the complexity of the disease process makes HTN difficult to manage and study. Much promise has been placed in the study of genetics, particularly in regards to popular gene association studies where associating the frequency of a particular allele and/or haplotype (combination of alleles) with a disease is the focus. As summarized by Small and collea gues (2003), some allele-based association

PAGE 24

11 studies report positive associations between ADR alleles and HTN, while others report no significance of these alleles. Specifically, the 1AADR has been hypothesized to play a role in HTN due to its role in vasomoti on; however, only one study has associated a polymorphism of the 1AADR gene to HTN (Jiang, Mao, Zhang, Hong, Tang, and Li, et al., 2005). Polymorphisms of the 2-ADR have been positively associated with HTN (Timmermann, Mo, Luft, Gerdts, and Busjahn, et al.,1998.) Gene asso ciation studies are numerous, but many are inconclusive, inc onsistent, and poorly powered. Animal and human studies focusing on physiologic, pharm acologic, genomic, and genetic factors have shown promise in providing evidence for and -ADR mechanisms in mediating CVR in HTN, as will be delineated in Chapte r 2: Review of Litera ture. Still, little attention has been paid to th e role of gene expression of and -ADRs in vascular tissue in HTN and CVR in human studies. Gene expression stud ies have been conducted to identify the role of and -ADRs, but predominantly in non-human models (Gaballa, Peppel, Lefkowitz, Aguirre, Dober, and Pennoc k, et al., 1998) or specifically to focus on the effect of medication (Wang & Brown, 2001; Nishio & Watanabe, 1999) or aging (Miller, Hu, Okazaki, Fujinaga, & Hoffman, 1996) on gene expression of these ADRs. Few studies have been found in the literatur e regarding gene expression analysis of ADRs in humans. Only one study demonstrated the differences in gene expression of these ADRs between persons with and without HTN. This group of researchers examined the presence of three 1-ADRs in peripheral blood lymphocytes of human NT and hypertensive subjects (Veglio, Tayebati, Schiav one, Ricci, Mulatero, and Bronzetti, et al., 2001). They studied gene expression of the ADR genes located in the blood. They also compared the human blood sample gene expre ssion data to that of NT and hypertensive

PAGE 25

12 strains of rats, finding similar 1-ADR densities in human blood and animal tissues. Also, significant differences in expression of certain 1-ADR subtypes were found both between humans with HTN and their normote nsive controls, as well as between the normotensive and hypertensive strains of rats (Veglio et al., 2001). This study provided important information about th e use of peripheral blood lymphoc ytes in the analysis of gene expression of 1-ADR subtypes, as well as relati on of human to animal models. Their findings supported the link between 1-ADR subtypes in HTN at the gene expression level (Veglio et al., 2001). Some lim itations of this study include isolation of the 1Asubtype in rat vas deferens tissue and not arterial, venous, aortic, or myocardial tissue. Finally, measuring mRNA levels via peripheral ly mphocytes is an indirect measure of transcription because the meas urement is not occurring in the tissue of interest (or the tissue th ought to be directly involved in the disease pathway). This is less reliable than direct methods, where mRNAs are examined in the tissue. Veglio and colleagues (2001) reported chosing this met hod because human tissue was not possible to obtain. Theoretically, there are concerns with using blood to examine gene expression. One major issue is the source of mRNAs in circ ulation; the origin of the mRNAs that are found in the bloodstream is lymphocytes. It is not clear if mRNAs expressed in the blood have differential expression than those expresse d directly in the tissu es. While Veglio and colleagues (2001) were able to show similari ties in expression between vas deferens tissue and blood lymphocytes, further inve stigation is needed to compare blood lymphocytes with other tissues. Many consider it less reliable to use circulating blood mRNAs to examine the direct relationship be tween transcriptional processes and disease

PAGE 26

13 mechanisms because other factors circulat ing in the blood could potentially vary the expression at any given time and thus make the research less replicable. Other human gene expression studies i nvolving ADR subtypes in human tissues have focused on other disease processes such as congestive heart failure, and have obtained samples from human myocardial tissue, most commonly obtained from endomyocardial-biopsy specimens (Lowes, G ilbert, Abraham, Minobe, Larrabee, and Ferguson, et al., 2001; Moniotte, Vaerman, Kockx, Larrouy, Langin, and Noihomme, et al., 2001). Gene expression studies examining the differences in mRNA level present in the actual tissue could help to explain if the actual expression of the proteins that make up the ADRs play a role in HTN, rather than me rely the presence of a particular allele. Researching the mechanisms that account fo r these differences has the potential to increase our understanding of the impact of gene expression on phe notypic variance in HTN and adrenergically-d riven vascular tone. The first aim in this study investigat es whether or not gene expression of 1Aand 2-ADRs are related to the di agnosis of HTN by examining their expression in human arterial tissue between subj ects with and without a dia gnosis of HTN. This method provides direct measures of steady-state levels of mRNA and insight into the genetic picture of real-time transcrip tion activity in each subject’s individual environment. To my knowledge, an investigation of this nature ha s not been previously reported in human arterial tissue, comparing hypertensive a nd normotensive subjects. This study provides preliminary results about the role of 1Aand 2-ADR gene transcrip tion levels and their relationship to HTN, and could lead to larg er-scale studies in the future. On a more

PAGE 27

14 innovative note, this study could further s upport gene therapy involving the ADR system in the management of HTN. The second aim of the study explored if differences in 1Aand 2-ADR gene expression levels exist when SIR was taken into account. This aim provides information about whether or not similar ADR gene expression trends exis t in self-identified racial groups. This study may contribute to the ove rall goal of reducing health disparities related to SIR and HTN that may be due to genetic variation, as identified by Healthy People 2010 (National Heart, Lung, and Blood Ins titute [NHLBI], 2003). The health disparities among hypertensive black and AA groups are real. What is uncertain is whether or not a genetic basis exists to help explain those dispari ties. This study provides preliminary data to begin to answer that question. The third specific aim of the study is to explore the relationship between 1Aand 2-ADR gene expression levels and the need for post-operative positive inotrope medication administration. The need for pos itive inotrope administration in the postoperative stage of recovery from coronary artery bypass surgery (C ABG) is most often the result of a negative card iac event (such as acute c ongestive heart failure, cardiac arrest, hypovolemia, or arrhythmias) that ne cessitates increased va scular resistance to correct the problem. Since 1Aand 2-ADRs are involved in va scular resistance, and inotropes increase vascular resistance, this aim provides indirect information about subjects’ cardiovascular reac tivity. In addition, it provides in formation about whether or not gene expression of the 1Aand 2-ADRs is related to poorer cardiovascular outcomes, as measured by their need for positive inotropes, in subjects undergoing CABG.

PAGE 28

15 The global health and economic consequences of this disease are incredible. HTN is a disease that affects every continent and population, some more disproportionately than others. Any novel information regarding the impact of 1Aand 2-ADR gene expression on people with HTN could prove to be a valuable building block for future studies. This study bridges the gap between bench and the bedside using advanced genetic technology to examine if and how gene expression of 1Aand 2-ADRs relate to HTN. In addition, the study seeks to explore the impact of SIR on gene expression of the two chosen ADR genes, and the impact of gene expression on the need for emergency cardiac medication in the post-ope rative stage of recovery.

PAGE 29

16 CHAPTER 2 REVIEW OF LITERATURE Introduction This chapter will present a review of the l iterature regarding all relevant topics of the study, including: genetics, hypertension, cardiovascular reactivity, adrenergic receptors, the 1Aand 2-ADR subtypes, pharmacotherapeutic aspects of the 1Aand 2ADRs, population differences in HTN devel opment and management, recommendations for the collection of racial and ethnic data implications for studying disease by SIR, positive inotrope administration, and genetic analysis techniques. Genetic Influences on Essential Hypertension Early heritability studies car ried out with essential hypert ensive twins estimated at least 63% of the variability in BP was due to genetic factors and re ported “little evidence for environmental influence on the familial aggregation of BP” (Grim et al., 1984, p. 453). The most cited estimate of essentia l HTN heritability is approximately 30% (Ambler & Brown, 1999). Some recent research ers examined the heritability of BP traits In a classical twin study of 173 dizygotic (DZ) and 251 monozygotic (MZ) twin pairs aged 18-34 years, randomly selected from th e East Flanders Prospective Twin Survey, heritability estimates were: SBP 74% (95% CI: 0.68-0.79) and DBP 63% (95% CI: 0.550.59). These heritability estimates were not confounded by the following potential risk factors: body mass index (BMI), cholesterol ratio, birthweigh t, physical activity, sex, and cigarette smoking (Zeegers, Rijsdijk, Sham, Faga rd, Gielen, and de Leeuw, et al., 2004).

PAGE 30

17 Genetic Models of Hypertension HTN has two distinct genetic classes: monogenic (meaning caused by one gene) and polygenic (caused by multiple genes). A number of monogenic forms of HTN have been identified, such as glucocorticoid re mediable aldosteronism, Liddle’s syndrome, Gordon’s syndrome, and Bilginturan s yndrome (all autosomal dominant gene abnormalities), and apparent mineralocorticoid excess, caused by an autosomal recessive gene abnormality (Beevers, Lip & O’Brien, 2001). These are rare in the general population. Essential HTN is considered polyg enic. This “polygenic model” of disease stems from R. A. Fischer’s “quantitativ e genetics” theory proposed in 1918. He postulated that a phenotype (observable expres sion of a genotype as a trait or disease) “was influenced by a large number of genes, each behaving accordi ng to basic Mendelian rules, but each having only a small indi vidual effect on the phenotype” (McClearn, Vogler, & Plomin, 1996, p. 96). Not long after, scientists realized that the environment could influence a phenotype, and the debate be gan concerning ‘nature versus nurture’. This led to introduction of the gene -environment interaction model. The earliest research designs that suppor ted these models of familial aggregation were pedigree studies involving twins, siblings, and families. MZ twins share as much as 99% of genetic information; DZ twins and non-tw in siblings share as much as 50% of the same genetic information. Large-scale pedigree studies in which families are observed longitudinally created the basis for the common linkage analysis that now uses pedigrees and genetic testing to link specific genes to phenotypes of disease. From these studies, a great number of candidate genes were discove red for high BP. Association studies take this concept one step further. These studies correlate candidate genes to hypertensive

PAGE 31

18 phenotypes in the general popul ation, looking for associa tions between alleles (or variants of the gene) and HTN. Another current model of HTN is the an imal model. Animal models ultimately serve as models for human disease. Anim al researchers use sophisticated breeding methods and more recently, gene knock-out and knock-in techniques to manipulate and control genetic, environmental and even phenot ype variables. Using this basic research model, researchers can investig ate numerous theories in HTN that could not otherwise be studied in humans. Cardiovascular Reactivity, HTN and Adrenergic Receptors As previously defined, CVR is a complex cardiovascular trait in which individual cardiac and vascular responses to physiological and psycholog ical stressors may lead to changes in SBP, DBP, TPR, and other he modynamic measurements that represent vascular response to maintain homeostasis These changes differ between NTs and hypertensives, in both human (de Visser, va n Hooft, van Doornen, Hofman, Orlebeke, & Grobbee, 1995) and non-human animal models (McDougall, Paull, Widdop, & Lawrence, 2000). Furthermore, these res ponses have been categorized as being predominantly and -adrenergic in nature (L inden, Gerin, & Davidson, 2003). and -ADRs are members of the super family of cell surface receptors that carry out signaling via coupling to guanine nucleotide binding pr oteins (G-proteins) (Small, McGraw, & Liggett 2003). They are critical components in the sympathetic nervous system's response to disease and maintenance of homeostasis, as they are the target receptors for epinephrine and norepinephrine (Small et al., 2003). Theoretical ly, alterations in peripheral vascular mechanisms are the proposed basis for the and -ADR sensitivitymodulated CVR, in which exaggerated res ponses to a stressor produce differing CVR

PAGE 32

19 (Lovallo & Gerin, 2003). These exaggerated resp onses are proposed to be a result of preclinical alterations in vasc ular resistance that can cause a disproportionate rise in BP relative to an otherwise normal demand for blood flow (Lovallo & Gerin, 2003). Fredrikson, Tuomisto, and Sundin (1990) re port vascular dysr egulation to both conditioned and unconditioned va soconstriction in their comparison study of vascular response to classical conditioning in mild hypertensives versus NT. Miller and Ditto (1991) report patterns of incr easing vascular resistance in response to an active-coping psychological stressor, which we re purported to be due to -adrenergic activity, and not neurohumorally independent autoregulation. Alpha 1Aand Beta 2-Adrenergic Receptors There are nine subtypes in the family of human ADRs; the 1Aand 2-subtypes are specifically located in th e vasculature (Small, McGr aw, & Liggett, 2003). The 1A-ADR gene is located on chromosome 8 at location 8p21 (OMIM # 104221, 2002). The 2-ADR gene is located on chromosome 5 at location 5q32-34 (OMIM # 109690, 2003). Figure 21 shows the approximate locations of each ADR gene. Vascular contraction is controlled primarily by 1-ADRs, and their importance in BP regulation is emphasized by the efficacy of 1-AR antagonists in human HTN (Rokosh & Simspon, 2002; Guthrie & Siegel, 1999; ALLHAT Collabora tive Research Group; 2000). The 1A-ADR receptor gene product is required to maintain arterial BP, as evidenced by a recent mouse gene knockout study (Rokosh & Simspon, 2002). Leech & Faber (1996) reported that constriction of rat skeletal muscle arte rioles is mediated predominantly by an 1D-ADR. However, Reja, Goodchild, and Pilowsky (2002) reported 1A-receptor messenger ribonucleic acid (mRNA) expression was significantly greater in spontaneously hypertensive (SHT) rat tissue samples compared with NT rats, and was positively

PAGE 33

20 correlated with SBP in all central tissue investigated. The 1A-ADR mRNA expression level appears to be an important determinant of SPB, and is one of the genetic markers examined in this study. A B Figure 2-1. Chromosome (Ens embl human map view) show ing the locations of both ADR genes. A) Chromosome 8, with br acket indicating approximate location of the 1A -ADR gene. B) Chromosome 5, with bracket indicating approximate location of the 2-ADR gene. During activity or stress, -AR signaling is responsible for regulating changes in heart rate, BP, and contractility (Reja, Goodchild, & Pilowsky, 2002). After selective ADR receptor activation, both 1and 2-ADR elicited dilation of large coronary arteries (Young, Vatner, & Vatner, 1990). Monopoli and co lleagues (Monopoli, Conti, Forlani, & Ongini, 1993) reported that human coronary artery contains equimolar amounts of 1and 2-receptor subtypes and that 2-ADR specifically mediates vasodilation in vascular smooth muscle. Likewise, anothe r study reported predominantly 2-mediated relaxation in human IMA exposed to both epinephrin e and norepinephrine in vitro (Ferro, Kaumann, & Brown, 1993). Polymorphisms of the 2-adrenoroceptor gene have been

PAGE 34

21 associated with: 1) Interindi vidual variability in resting SBP and DBP in response to mental challenge (McCaffery, Pogue-Geile Ferrell, Petro, & Manuck, 2002); 2) The level of resting and stress-re lated BP (Li, Faulhaber, Rosenthal, Schuster, Jordan, Timmermann, and Hoehe, et al., 2001); and 3) Vascular reactivity as indicated by lower basal blood flow and attenuated increases in forearm blood flow in hypertensive adults (Cockroft, Gazis, Cross, Wheatley, Dewar, a nd Hall, et al., 2000). A polymorphism of the 2ADR gene, the Gln27Glu (glutamine, c odon 27, glutamate) which causes a point mutation of cytosine (C) -to-guanine (G), was examined by Bray and colleagues (2000). They reported an occurrence of HTN with the Glu27 allele was 1.8 times higher than with one or two copies of the Gln27 a llele (95% confidence interval, 1.08 to 3.00, p = 0.023). Chruscinski and colleagues (2001) demonstrated a positive role for 2-ADR in mediating vascular dilation when BP respons e to was blunted in a mouse gene knockout model. Knowing that the vasc ular system is rich with 2ADRs and that they mediate vasodilatation, Iaccarino and colleag ues (2002) chose to overexpress 2-ADRs via adenoviral-mediated gene transfer in nor motensive Wystar-Kyoto and spontaneously hypertensive rats. They reported su ccessful gene transfer of the 2ADR gene and enhanced vasorelaxation in the carotid arteries of hypertensive strain of rats versus the NT strain ( n = 8 to 10 per group) after 2ADR overexpression ( F = 3.088, P < 0.05). 2ADR appears to be an important determinan t of BP, and was the second genetic marker examined in the study. Pharmacotherapeutic Aspects of Alpha -1A and Beta-2 Adrenergic Receptors The primary indication for both and -blocking drugs is HTN. In the case of blockers, cardioselective types ar e preferred to reduce side e ffects caused by blockade of multiple ADR subtypes. Selective peripheral 1 blockers such as prazosin and terazosin

PAGE 35

22 induce vasodilation by blocking the 1 receptors in vascular smooth muscle (arterioles and veins). Their selectivity to 1 causes less reflex tachycardia than drugs that inhibit 2 (Kalkanis, Sloane, Stri chartz & Lilly, 1998). More commonly used in the treatment of HTN are blockers, like metoprolol, propanolol, and atenolol. The JNC VII (2003) suppor ts their use in various populations and they have been shown to reduce morbid ity and mortality in randomized controlled trials. Cardioselective -blockers principally block 1 receptors and partially block 2 receptors. This reduces side effects of blocking all 2 receptors in the lungs and blood vessels. Although ADRs, when stimulated cause vasodilation, -blockers also reduce renin release from the juxtaglomerular ce lls of the kidney, thus reducing the renin angiotensin system’s effect on increasing BP. In addition, -blockers interfere with sympathetic vasoconstrictor nerve activity a nd block the effects of catecholamine surges (Khan, 1999). blockers also reduce heart rate plasma norepinephrine, muscle sympathetic nerve traffic and systemic norepinep hrine spillover, all indices of adrenergic activity in essential HTN (Grassi, 2004). Population Differences in HTN The prevalence of HTN in AAs in the U. S. is among the highest of all groups. AAs tend to have worse clinical sequelae than their White, non-H ispanic counterparts (Cooper & Rotimi, 1997). Americans who self-identify with African descent have a 1.5-2 fold increase in prevalence of HTN compared to Americans who self-report descent from Europe; comparing women in these two groups leads to the highest prevalence differences (Eberhardt et al., 2001). Dis ease management is often particularly problematic, with AAs often requiring a more than one (and often multiple) antihypertensive medications to effectively manage their hi gh BPs, many of whom cannot

PAGE 36

23 afford them. HTN is a particular problem in h ealth care in the Southeastern U. S., as the prevalence of HTN among blacks and whites is greater, and death rates from stroke are higher in this region th an others (AHA, 2003). In regards to developing HTN, examples of racial differences in physiologic mechanisms and pharmacologic responses involving ADRs exist in the literature. Differences in CVR are noted in relati on to BP and heart rate, and include and adrenergic patterns that are associated w ith racial cohorts--particularly AAs versus Caucasians. Anderson, McNeilly, & Meyers (1992) explain tw o dichotomous CVR patterns that are reported to be associated with SIR: the myocardial and the vascular. These patterns are summarized in Table 2-1. Table 2-1. Adrenergic cardi ovascular stress patterns. Myocardial Reactivity Pattern ( -adrenergically driven) Vascular Reactivity Pattern ( -adrenergically driven) in BP associated with: Cardiac Output (C.O) Stroke Volume Heart Rate Epinephrine and Norepinephrine in Total Peripheral Resistance in BP associated with: Norepinephrine Total Peripheral Resistance Characteristic of Caucasian reactivity pattern Characteristic of Black/AA reactivity pattern There are reported differences in drug dis position and responsiveness in relation to adrenergic-agonists and an tagonists (Wood & Zhou, 1991), implying ADR differences across SIRs. Some support the above model, a nd some do not. Sentinel research shows that -blockade and combined and -blockade via pharmacologica l agents appear to be less efficacious in AAs, South Africans, Jama icans, and West Indians as compared to

PAGE 37

24 Caucasians (Humphreys & Delvin, 1968; Jennings & Parsons, 1976; Seedat, 1980). Another seminal study in 1977 by the Hypert ension Detection and Follow-up Program Cooperative Group reported -blockers to be less efficacious in African Americans. Gibbs, Beevers, and Lip (1999) purported that this may be due to decreased cardiac output and renin release, causi ng increased total peripheral re sistance. In support, Wood (2002) reported a marked impairment of 2AR-mediated vasodilation in blacks, accompanied by increased -adrenergically mediated vasoco nstriction, as well as racial differences in response to endogenous and e xogenous agonists. Al so, vasoconstrictor response to endogenously stimulated norepin ephrine is higher in blacks than whites (Stein, Lang, Singh, He & Wood, 2000). Male, bl ack Veterans residing inside the ‘Stroke Belt’ (southeast U.S.) are repor ted to have lower treatment success rates with captopril ( p = 0.07); and, regardless of region, blacks in the study were less likely than whites to achieve successful lowering of their BP with atenolol ( p = 0.02), prazosin ( p = 0.03), and more likely with diltiazem ( p = 0.05) (Cushman et al., 2000). Jamerson and DeQuattro (1996) disagree, explaining that while obs erved responses of blacks to both ACE inhibitors and -blockers in the treatment of HTN ar e less favorable than is seen in whites, the responses are still clinically significant. Litera ture on nonpharmacologic differences by SIR is also present. Ferro and Walton (2001) report significant differences in short-term BP responses to a 10-w eek regimen of nonpharmacological treatments (dietary, activity, stress reduc tion, and education sessions) in HTN for African/Caribbean blacks compared to whites. Blacks and the c ontrol group experienced no change in either systolic or diastolic BP at 10 weeks, and st atistically significant decline in systolic ( p < 0.005) and diastolic BP ( p < 0.05) were seen in the Ca ucasian group. In a 2005 meta-

PAGE 38

25 analysis of 137 monotherapy clin ical trials and 28 combina tion therapy trials (totaling 11,739 participants), Wu and colleagues reporte d that AAs had better BP reduction with calcium channel blockers th an their non-AA counterparts ( p = 0.001); and, that non-AAs responded better than AAs to 1-blockers, 1-blockers, and angiotensin converting enzyme inhibitors ( p = 0.0001) (Wu, Kraja, Oberman, Le wis, Curtis, Ellison, and Arnett; 2005). The estimated heritability for HTN in pe ople of sub-Saharan African descent is 4568% (Rotimi, Cooper, Cao, Ogunbiyi, and Ladipo et al., 1999; Gu, Borecki, Gagnon, Bouchard, Leon, and Skinner, et al., 1998). Genetic studies of association have reported racial differences in allele frequency of ADR polymorphism s (Hindorff, Heckbert, Psaty, Lumley, Siscovick, and Herringt on, et al., 2005; Small, McGr aw, & Liggett, 2003; Xie, Kim, Stein, Gainer, Brown, & Wood, 1999). This review of the literature highlights the many inconsistencies across studies in regards to the influence of race on health a nd disease. A number of things can explain these inconsistencies: study design issues ; differing measurement/report of race; inadequate power; and different medications studied. Another plau sible explanation is that the differences are really attributable to non race-based, interindividual variability. Recommendations for the Collection of Racial and Ethnic Data Much like its original use for classi fications of groups, some mainstream definitions of race today infer major biological underpinnings. The online MerriamWebster dictionary (2003) defines race as “a division of mankind possessing traits that are transmissible by descent and sufficient to characterize it as a distinct human type”. This definition inherently includes genetics as a factor in race by use of the phrase ‘transmissible by descent’. Numerous similari ties in other definitions exist, involving

PAGE 39

26 such phrases as: ‘physically distinguishable,’ ‘having comm on ancestries,’ and ‘having certain biological characteristic s that set them apart from ot her groups.’ It is easy to see why and how the paradigm of linking genetics to race exists even now, as these are definitions from no longer than a century a go. Very recently, there has been an upsurge of efforts to discard notions of race as biologically-associated. Some researchers and ‘civilians’ wish the term to be recognized so lely as an antiquated system of skin-colorbased classification that inherently carries with it socio-politically charged notions of racism. Others argue for less biologically-based definitions of race that incorporate social beliefs about language, history, and culture (s uch as Witzig, 1996). Definitions like these seem very similar to those of ethnicity, where language, history, culture, and sociopolitical factors are main constructs of this term. To further complicate the matter, some utilize the term ethnicity with the presum ption that they are avoiding any biological undertones inherent in the term ‘race’; howev er, definitions of race and ethnicity are markedly similar and most people use the two terms interchangeab ly (Sankar & Cho, 2002). In a meta-analyses of articles published in Nursing Research authors Drevdahl, Taylor, and Phillips (2001) present this case well, comparing three op erational definitions of race and three of ethnicity used (some c ited from other sources) by nursing researchers within the 1990s and 2000. Race and/or racial group was defined as: “Imply[ing] biological characteristics…that are geneti cally transmitted from one generation to another” (Schubert & Lionberger, 1999, p. 116); “Concept that signif ies and symbolizes sociopolitical conflicts and in terests in reference to diffe rent types of human bodies” (Winant, 2000, p. 172); and, “Group that is soci ally defined as having certain biological characteristics that set them apart from othe r groups, often in invi dious ways (Pincus &

PAGE 40

27 Ehrlich, 1999, p12). The three ethnicity and/or ethnic group definitions were: “Contains information about the history of the populat ion, and hence the genetics of the group, as well as sociocultural information” (C rews & Bindon, 1991, p. 45); “Group that has certain cultural characteristic s that set them off from other groups and whose members see themselves as having a common past” (P incus & Ehrlich, 1999); and, “ Segment of a larger society whose members are thought…t o have a common origin and to share important segments of a common culture a nd who…participate in sh ared activities in which the common origin and culture are si gnificant ingredients” (Yinger, 1994, p. 3). While debate continues over the construct of race and ethnicity, it is generally agreed upon that: Whatever definition is used, it s hould be clearly delineated (Sankar & Cho, 2002); Racial classifications should be critically evalua ted for their usefulness and contribution to the testable theories (Duste r, 2001); and, The methods of capturing race and/or ethnicity should be carefully outlined (Williams, & Jackson, 2000). Standardization of definitions is a major issue and is recommended for consistency in reporting results in research (U.S. De partment of Health and Human Services [USDHHS], 2003). The OMB Race and Ethnicity Classification system can be utilized for standardization. It includes both race and ethnicity categories, and defines race by context of geographical origin. Cultural associations are a ccounted for in the Ethnicity self-report section. Users have the option of selecting more than one self-affiliated category of race; or, the option to not an swer the question at all. Based on this knowledge, this study will utilize the OMB Classf ication system. This self-report system allows the user to choose more than one r ace, if applicable. It utilizes standardized

PAGE 41

28 constructs of race and ethnicity that are rooted in ancestral and geographic origins of their predecessors and is widely used in research. Implications for Studying Disease by Race An abundance of health disparities literatur e exists on the social, ethical, and legal ramifications of studying disease by SIR. Hi storically speaking, some experiments that target racial groups have re sulted in serious social a nd ethical problems for that population (for example, the Tuskegee syphilis ex periments). Current debate in regards to identifying genetic differences by race ha s identified many concerns, including the potential to send the message that resear chers are trying to find clear biological differences that would imply certain races are 'unequal' to others. People fear that finding biological differences will justify certain social inequalities. The key opposition to studying diseases by race are the following points: 1) There may be many inherent problems in measuring and grouping races in a multi-racial society for the purpose of genetic clustering (Wilson, Weale, Smith, Gr atrix, Fletcher, and Thomas, et al., 2001; Williams & Jackson, 2000); 2) The sociopolitical context of race is an important variable that is often disregarded in research (Burchard, Ziv Coyle, Gomez, Tang, and Karter, et al., 2003) ; 3) Fear of justif ying inequality (Bonham, 2003); and 4) Racial boundaries are not likely to be equally useful in all kinds of genetic research (Sankar & Cho, 2002). Others argue for research that examines race carefully making sure to address the above concerns. Duster (2001) purports that race s hould not be discarde d as a variable in research just because the categories do not biologically map exactly, and that race remains alive in the context of practical application. Fu llilove (1994) warns against a priori consideration of race as important in medi cine without question, but also states that

PAGE 42

29 little is truly done to explain the meanings in associations between health outcomes and race. In certain disorders and diseases, race is very much a risk-associated variable. Sickle-cell anemia (SSA) maintains a significa ntly higher prevalence in people of African and Mediterranean descent. Likewise, Cystic Fi brosis (CF) is more likely to affect those from Western European descent. That is not to say that whites have not been diagnosed with SSA, or that cases of CF have not been seen in blacks or other non-whites. It is only to say that there is a higher risk associ ated with particular diseases among certain populations. Population genetics has ascertain ed that greater genetic differentiation occurs between continentally separated groups (Burchard et al., 2003), and that more variation is present within racially-stratifie d populations than betw een them. Nonetheless, others have reported great ge netic variation among the five racial groups (five groups as categorized by the OMB classification sy stem) (Risch, Burchard, Ziv, & Tang, 2002). Genetically speaking, some racial groups po ssess low frequencies of certain alleles associated with disease, (2003). Whether or not the low frequency of alleles is truly associated with race or simply a result of sampling methods or sta tistical errors in the research is currently under debate. While it is more easily seen in “simple” diseases like SSA and CF, the genetic influence of race is much more difficult to ascertain in complex diseases like type 2 diabetes (T2DM), asthma, HTN, and Alzheimer’s disease. Nonetheless, specific susceptibility gene vari ants for chronic diseases have been found in specific populations. Phimister (2003) explains that a vari ant of the calpain-10 gene (associated with T2DM) is specific to a population of Mexican Americans in Texas. Moreover, a variation of the ETS family of genes that predispose carriers to asthma has

PAGE 43

30 been discovered in a population inhabiting th e island of Tristan da Cunha of the South Atlantic (2003). In recent edit orial, Fine, Ibrahim & Thomas (2005) cite three studies of complex genetic disorders (Cr ohn’s disease and Factor V Le iden) where genetic variation by race was reported (Ridker, Miletich, Henne kens, & Buring 1997; Shen, Lin & Tsay, 1997; Hugot, Chamaillard, Zouali, Lesage, Cezard, and Belaiche, et al., 2001). For some, results of these studies are not convincing enough to conclude that race is an appropriate variable for use in res earch. Cooper, Kaufman, and Ward (2003) argue against studying disease by race, first asserti ng that race-specific findings in research are better explained by environmentally-determined socioeconomic factors. They (2003) also report examples of inconsiste ncies in research in which Type I error affected the outcomes and interpretations, such as with the reported race-specific effect of ACE inhibitors in the Antihyperte nsive and Lipid-Lowering Treatme nt to Prevent Heart Attack Trial. Cooper and colleagues also argue that true race-specific genetic results have not been found, and are “mathematically and bi ologically implausi ble” (p. 1167, 2003). In 2001, Wilson and colleagues performed a unique study designed specifically to test the validity of race and ethnicity as genetic research variables. They studied eight populations of varying origin, some of whic h were extremely specific (for example, Amharicand Oromo-speaking Ethiopians from Shewa and Wollo provinces collected in Addis Ababa). Using a model-based cluste ring method known as STRUCTURE, they were able to estimate the proportion of each individual’s genome having ancestry in each cluster. One underpinning in this model is the idea that admixture plays an important role in the variability of race. Genetic admixture reflects multiplicity or variation in race and ethnicity (Burchard et al., 2003). Wilson and his co-authors ascertained the

PAGE 44

31 apportionment of individuals (average per-i ndividual proportion of ancestry) by using the STRUCTURE model to characterize ‘clusters’ based on a set of allele frequencies at each locus. From there, the researchers matched the clusters with specific geographical areas (in this case, four broad regions), and interp ret the similarities seen in each cluster. Because 62% of the Ethiopians fell into the cluster with the Jews, Norwegians, and Armenians, they concluded that identifying th ese people as “black” in race would be an “inaccurate reflection of the genetic stru cture” (2003, p. 266). They concluded similar results with subjects from China and New Guinea in which “Asian” race grouping would have been inappropriate (2003). While al most truly convincing, this study requires replication, and use of “presumably neutral micr osatellite markers” (p. 265) needs to be validated as truly neutral. Based on these a nd other data, many researchers have moved toward the use of ancestry, rather than race or ethnicity in their research models. In fact, self-reported ethnicity and ancestry constructs have been related in biological models. Helgadottir and colleagues (2006) found that self-reported ethnicity was highly correlated with genetic determination of estimated indi vidual ancestry (via ancestry informative markers) and even group ancestry (determine d by weighted least squares). They also reported ethnicity-base d differences in risk for myocardial infarction in African Americans who had European admixture. This implies that self-reported ethnicity can be informative and useful in genetic resear ch as a means to group individuals for comparison. The PI agrees that numerous other factors may impact the poor health outcomes seen in the AA population, incl uding various socioeconomic indicators and factors related to health care access and delivery. It is important to reiterate that the literature

PAGE 45

32 presented clearly identifies specific pa thophysiologic and biologic differences among racial groups involving HTN that direct this line of research In further support of this venture, at least one other (K rieger, 2005) suggests that in the case of gene expression, observed phenotype differences seen among pop ulations could possibly reflect variation in gene expression (more so than simple gene frequency) because of the nature of gene expression patterns occurring in the context of certain (perhaps shared) environmental conditions. Moreover, one charge of the Ta sk Force of the American Sociological Association was to comment on the further in vestigation of race and ethnicity in the contribution of disparate outcomes (2003). Th ey presented the example of AA health disparities in regards to HTN and affirm that research needs to continue in this line of research to distinguish between social and biological forces at play (2003). Properly designed research in this ar ea could provide answers regard ing molecular differences and poor health outcomes, a better understandi ng of the disease process in certain populations, and tailoring of me dications or gene therapy. Fu rthermore, proper statistical analyses and careful interpreta tion of findings can strengthen results and help to reduce potential ‘racial profiling.’ The ethical principle of social justice, in its simplest form, states that all people should be afforded equal benefits, (such as goods and services) re gardless of their personal characteristics, choices, or beliefs. Current research that examines race as a variable has also been criticized for limiting social justice. For example, research that excludes subjects based on their race (a practice th at is declining) prev ents certain groups from reaping the benefits of a study. Simila rly, certain socioecono mically limited groups tend not to benefit from research because of the inherent cost associated with new

PAGE 46

33 technology that comes from it. As certain races have been associated with lower socioeconomic status (SES), social justice is compromised for these groups. It is from research such as this that we learn of the social and ethical rami fications that can be incurred if research is not properly conducted involving race and diseas e. In the research community we must be most careful in our interpretation of da ta involving race, ethnicity, ancestry, and/or soci oeconomic status so that mi sleading conclusions are not made that adversely affect so cial justice, policy, and practi ce. We should be, at minimum, cognizant of how the design, implementation, analyses, and interpretation of results involving health disparities can affect th e overall well-being of socioeconomically disparate groups. Positive Inotrope Administration Short-term positive inotropes are often administered post-operatively to CABG patients to increase vascular resistance in ca ses of acute congestive heart failure, cardiac arrest, hypovolemia, and arrhythmias. The th ree subclasses of positive inotropes are: cardiac glycosides (digitalis, digoxin); -adrenergic agonists (dopamine, dobutamine, epinephrine); and phosphodiesterase inhib itors (milrinone, amrinone, enoximone). The need for positive inotropes in the post-operative phase of recovery from bypass surgery can be informative as an indirect assessment of the subjects’ cardiov ascular reactivity. As ADRs are involved in vascular tone, knowledge about whethe r or not a subject needed this medication could help to explain some of the differences in gene expression of the 1Aand 2-ADR genes. Genetic Analysis Techniques There are many types of genetic methodologi es used in research today. Linkage studies identify regions of the genome that contain putative candi date genes that are

PAGE 47

34 proposed to be related to a phenotype or disease process based on their location on chromosomes. Association studies investigate the prevalence of certain gene alleles and their association with a phenotype, or diseas e process. Both of these methods examine deoxyribonucleic acid (DNA). Unlike these, gene expression studies examine the relationship between the leve l of ribonucleic acid (RNA) and disease. While DNA is a double-stranded nucleic acid made up of nucleot ides, RNA is single-stranded, and is the result of transcription of genetic information; the information encoded in DNA is transcribed into mRNA, which is an intermed iate and one of the re gulatory steps in the synthesis of new proteins. Many cellular char acteristics concerning survival, growth and differentiation are reflected in altered patte rns of gene expressi on and the ability to quantify transcription levels of specific genes is central to research into gene function (Zamorano, Mahesh, & Brann, 1996). Common gene expression methods (quantific ation of steady-state transcription) are northern blotting, in-situ hybridization, RNAse protection assays, cDNA arrays, and RTPCR (Bustin, 2000). RT-PCR is a type of PCR that allows one to compare the levels of a specific mRNA in different sample populati ons, to characterize patterns of mRNA expression, to discriminate between closely related mR NAs, and to analyze RNA structure (Bustin, 2000) This method invol ves isolating mRNA from the biological sample (in this case, IMA) and performi ng reverse-transcription (RT) to make complimentary DNA (cDNA) The cDNA represen ts only the mRNA component of the total RNA, which can then be analyzed by ge ne expression equipment. TaqMan is a type of RT-quantitative PCR method that continuous ly measures (in real time) accumulated PCR product. The PCR product reflects the origin al level of mRNA template (See Figure

PAGE 48

35 2-2). This is measured using a TaqMan pr obe. The probe is a dual-labeled fluorogenic oligonucleotide. The dual-labels are a repor ter dye and a quenching dye. The ABI Prism (Applied Biosystems HT 7900) equipment and software examines the fluorescence intensity of the reporter and quencher dyes a nd calculates the increase in normalized reporter emission intensity over the course of the PCR amplifi cation (Genomics and Proteomics Core Laboratories, 2003) within a ll samples located on the 96-well plate. The normalized reporter is known as a housekeeping gene. This is a type of gene in which there is a known and predicte d level of expression. Its purpose in gene expression analysis is to act as an internal reference or control, whereby all samples are normalized by this same gene. (Practical ly speaking, since the level is known, the housekeeping gene values are subtracted from the sample values to obtain the normalized levels.) This allows relative expression to be estab lished, instead of absolute quan tification of the data, which is tedious and impractical (Bustin, 2000). A spectrum of standard housekeeping genes are available. The protocol for this st udy will use glyceraldehydes-3-phosphate dehydrogenase (GAPDH), a housekeeping gene th at has been previously used with successful results in normalizing data for ADR gene expression in arterial tissues (Wang & Brown, 2001; Reja, Goodchild & Pilowski, 20 02; Peuster, Fink, Reckers, Beerbaum & von Schnakenburg, 2004). Northern blotting, in-situ hybridization, RNAse prot ection assays, and cDNA arrays are all alternate methods of measuring gene expression Northern blotting isolates RNA by elecrophoretic separation on an agaros e gel, ‘blotting’ or transferring RNA fragments from the gel onto a membrane (usu ally nitrocellulose), adding a labeded probe to the membrane and detecting the band w ith the probe bound to the target sequence

PAGE 49

36 (CRISP Thesaurus, accessed 8/2/05 ). In situ-hybridization dete rmines the presence of an RNA sequence of interest by hybridizing a prob e to the target sequence, and visualizing on a microscope the location of the bound R NA target in the chromosome or cell (cytoplasm, for example) (Human Genome Pr oject, Talking Glossar y, accessed 8/2/05). RNAse protection assays examine gene expression by hybridiz ing antisense RNA corresponding to known genes with an unknown sample. Next, the sample is digested with a single-strand specific RNAse and any surviving RNA left is presumed to be complimentary to the antisense and therefore, transcribed from the gene of interest (CRISP Thesaurus, accessed 8/2/05). Fina lly, cDNA arrays (more commonly known as microarrays) utilize a microarray ‘chip’ or platform with many small spots that correspond to a different gene on each spot. Th e spots are pre-treated with cDNA that is the only coding part of the sequence of intere st that corresponds to an mRNA transcript. These cDNA spots are hybridized with a pr obe. The chips are incubated in solution containing the genetic material being investigated. Messenger RNA transcripts floating in the solution hybridize to their cDNA already on the chip. When the chip is exposed to ultraviolet light, the fluores cent probes emit light at va rying intensities, allowing qualitative comparison of e xpression between the different genes on each chip and between subjects (Rice Univ ersity Connections webpage accessed 8/2/05). These techniques can be limited in their sensitivity (Melton, Kreig, Rebagliati, Maniatis, Zinn & Green, 1984) and in their cost-effectiven ess (Bustin, 2000). RT-PCR is the optimal method when evaluating a limited number of ge nes and starting mRNA template is low. There are additional advantages in using TaqMan Real Time RT-PCR gene expression analysis. Unlike other forms of quantitative RT-PCR, this method quantitates the initial

PAGE 50

37 amount of the mRNA template (the geometric phase), rather than the final amplified product (the plateau phase), allo wing detection of a 2-fold ve rsus a 10-fold change. This improves the sensitivity, specificity, and reproducibility of the method (Dorak, 2003; Dawson, 2003). Real Time RT-PCR also involv es only three major steps, whereas other conventional RT-PCR methods involve nine st eps. Reducing the number of steps in the gene expression process assists in minimizi ng error in sample analysis. Also, once the ABI Prism equipment has completed its fluorescence phases, the data are fed into a computer linked to the equipment, elimina ting the need for post-PCR processing (Bustin, 2000). The only clear disadvantage of this method is the cost of the predeveloped reagents and the ABI Prism, but it is s till more cost-efficient than cDNA array (microarray) methods when evaluating a sm all number of genes. The TaqMan RT-PCR principle steps are represented in Figure 2-3. Figure 2-2. Amplification of gene expre ssion using TaqMan Real Time RT PCR.

PAGE 51

38 Figure 2-3. TaqMan RT-PCR steps (adapted from Bustin, 2000). Collection of Human Tissues for Gene Expression Analyses Tissue samples and/or biopsies are the s ource of choice for analysis of gene expression/transcription, simply because it is the direct site at which to examine the mRNA levels mediating protein production in the body. It is widely accepted that mRNAs found in circulating blood lymphocytes provide indirect evidence of this process. During CABG, it is common for por tions of surgical remnants of IMA to be discarded. The IMA branches off of the left subclavian artery and supplies the thoracic cavity with oxygenated blood. It is most typi cally used for bypass of th e left anterior descending coronary artery. As previously described, 1Aand 2-ADRs are specifically located in the vasculature (Small, McGraw, & Liggett, 2003). Alpha1A mRNA has been detected in the IMA (Gow, Mitchell, & Wait, 2003). 2-AR has been detected in the human coronary arteries (Monopoli, Conti, Forlani & Ongini, 1993) a nd in the human IMA (Ferro, Kaumann, & Brown, 1993). Very minute amounts of arterial tissue (about 30 mg) are necessary for this type of analysis, whic h involves isolating messenger RNA from the tissue sample and performing re verse-transcription to make cDNA, which can then be used for TaqMan (Real Time) gene expression analysis.

PAGE 52

39 Summary HTN has a genetic component, as supporte d by empirical and experimental data. The 1A-ADR and 2-ADR subtypes of the ADR receptor genes are hypothesized to play a role in mediating the disease process of HTN via gene expression differences. Race and/or ethnicity may also contribute to th e variance seen in HTN and adrenergicallydriven vascular tone, as suppor ted by previous studies descri bed. While it is controversial to use race/ethnicity as variab les in genetic research, the use of a standardized measure and careful interpretation can be informativ e and help reduce the potential for social injustice. The use of Real-Time PCR for analysis of gene expression is an ideal method for examining steady-state transcri ption levels and comparing relative fold-differences between groups. This analysis can provide direct informati on about the function of these genes in the given environment at that par ticular moment. Tissue samples are the desired source for examining mRNA levels; however, human studies of the 1A-ADR and 2ADR genes expression in human tissue are not currently reported. This study attempts to fill the gap in this knowledge by using human arterial tissue for analyses of the 1A-ADR and 2-ADR genes expression levels between h ypertensive and normotensive adults.

PAGE 53

40 CHAPTER 3 PROCEDURES AND METHODS Introduction This chapter presents study design and pr otocols. Details regarding recruitment techniques, research setti ngs, variables, and methods are discussed. The following methods are thoroughly explicated: collection and storage of samples, processing of tissue and blood for isolation of genomic DNA and RNA, a nd post-isolation processing of RNA for gene e xpression analyses. Design This study used an exploratory, quantita tive design to meet the goals of the aforementioned specific aims. This was a twoarm gene expression study to compare NT versus hypertensive persons primarily, then to explore differences between two selfreported race categories: Bl ack/AAs and White/Caucasians. Recruitment goals were set at a total of 60 subjects w ith 15 subjects in each arm. This sample size was based on a formulation of 82% power, an effect size of 0.25 (medium), and a significance of 0.05 for a two-tailed test. Gpower comput er software was used to calculate the required sample size (Erdfelder, Faul, & Buchner, 1996). A medi um effect size was consistent with other gene expression studies of adrenergic mechanisms in HTN (Wang & Brown, 2001). Following consent, pencil and paper data co llection was used to obtain some basic demographic and medical history informati on. During surgery, a small amount of IMA, (normally discarded during CABG surgery) was collected and later analyzed for 1Aand 2-ADR gene expression via TaqMan (Real-T ime) RTPCR. Add itionally, about 10 cc

PAGE 54

41 of blood was collected intraoperatively for genotyping. A post-operative chart review was completed to determine the subjects’ need for positive inotrope pharmacotherapy while in intensive care. Subject Recruitment Prior to initiation of the study protocol, Institutional Re view Board (IRB) approval was obtained from the UF IRB-01 and the VA Subcommittee for Clinical Investigations. Adult subjects between the ages of 21 and 70 who were scheduled for CABG surgery were recruited from those admitted to the University of Florida Thoracic and Cardiovascular Surgery (TCV) team. These s ubjects included patients from the following facilities: Shands at Alachua General Hospital and the Malcolm Randall Veterans Administration Medical Center (VAMC), both located in Gainesville, Florida. These facilities mainly serve Alachua County, but of ten include patients from the entire North Central Florida and surrounding areas. HIPAA Waivers of Authorization were obtained from the UF and VAMC IRBs. This IRB-approved waiver allowed the applican t to review charts of the scheduled CABG patients to determine which patients met th e inclusion/exclusion criteria of the study. Inclusion and exclusion criter ia with rationale are presen ted in Table 3-1. Prior to beginning the study, all surgeons agreed to ha ve their patients screened for this study. These individuals were contacted by tele phone or via face-to-face meeting during preoperative appointments for recruitment. This recruitment process was continued until the planned group allotments were filled. Subjects were consid ered hypertensive if their medical chart indicated: a) di agnosis of HTN by a practitioner, b) three consecutive office BPs > 140/90 mmHg, or c) prescription of antihyp ertensive medications specifically for high BP. Subjects were also as ked to verify that they were diagnosed with high BP.

PAGE 55

42 Subjects were not excluded on the basis of ra ce, religion, ethnicity, socioeconomic status, or level of education. Women and minoritie s were recruited for this study. The AHA (2003) reports that more men than women have high BP until age 55. From age 55 and older, the percentage of women with high BP continues to incr ease (2003). Moreover, HTN is primarily an underlying cause of death for more women than men (2003). As indicated in the TCV surgery databa se, the average percent of women undergoing CABG surgeries by the TCV surgery department is 21%. This means that for every one female undergoing the procedure by this department, there are approximately 5 males. U.S. Census data for Alachua Count y indicates that approximate ly 73.5% of the population is Caucasian, 19.3% is AA, and 7.2% is ‘other’. The JNC VII (2003) reports that health disparities exist in minority populations with HTN. Inclus ion of women and minorities will create a study population that is representa tive of the entire population of those undergoing CABG procedure. Children were not included in this study. While 50 million Americans age 6 and older have high BP ( AHA, 2003), it is rare for children to have CABG surgeries. While it is important to study childhood HTN and its long-term consequences, it is not feasible to recru it such a minimally re presented population (children with HTN who undergo CABG). Table 3-1. Inclusion and exclus ion criteria with rationale. Inclusion Rationale age 21-70 -children excluded: rare CABG -There is evidence that genetics still plays a role in HTN even into late 70's, without conf ounding of isolated systolic hypertension (ISH) past the age of 70; however, ISH results in a clinically and pat hophysiologically differe nt phenotype from essential HTN (Sleight, 2004). This difference in phenotype could lead to a difference in gene expression that would confound the data.

PAGE 56

43 Table 3-1 Continued. Inclusion Rationale undergoing scheduled CABG surgery -prime surgery to obtain arterial tissue that is often discarded (Wang & Brown, 2001) -improper data collection could occur with unscheduled cases read/write English -unable to provide interpre ter for multiple languages Exclusion Rationale undergoing heart surgery that does not include the internal mammary artery (IMA) -the IMA is most frequent artery used in bypass surgery, usually grafted to the left anterior descending artery low cardiac output syndrome (LCOS)* -subjects with this post-oper ative hemodynamic diagnosis may exhibit increased total peripheral resistance secondary to the diagnosis and confound the inotrope-related data Consenting Process and HIPAA Regulations The PI obtained approval from the UF and VAMC IRBs for human subjects’ research. Subjects signed an informed cons ent to participate in the study, including informed consent for chart review and rese arch use of normally discarded surgical remnants of IMA (see Appendices B & C). Subjects were not compensated for their participation in the study. Subjects were inform ed they could withdraw from the study at any time. There was no anticipated direct benef it to the subjects: they did not receive any information concerning their hemodynamic res ponses to positive inotrope administration, nor did they receive results of their gene expression of 1A-and 2-ADRs. This eliminated the need for another informed consent that was designed specifically for disclosure of genetic information, and also eliminated the n eed for genetic counseli ng related to testing and/or results. It is uncertain at this time what the expression of these genes in human tissue actually means in terms of health bene fits; therefore, lay in terpretation would be difficult at this time.

PAGE 57

44 Data collection containers were labeled w ith a subject ID barcode sticker. A handwritten table containing the coding system for the subjects was kept in a locked filing cabinet with only the PI having access. This was the only source of data that matched the subject to the ID code. Setting This study was completed in Gainesville, Florida at the Gainesville VAMC and Shands at Alachua General Hospital facilities Screening of patient charts for eligibility occurred at the VAMC and the TCV surgery office in the UF Health Science Center. Subjects were approached fo r recruitment and consented at the VAMC, AGH, or at Cardiology Associates of Gainesville, all pl aces where subjects we re either undergoing pre-operative assessments or were admitted. Data collection occurred in the VAMC and AGH ‘heart’ surgical suites. Bl ood and tissue samples were stor ed at the UF Center for Pharmacogenomics. Laboratory analysis occu rred at the UF Pharmacogenomics Core facility and the UF Interdis ciplinary Center for Biotechno logy Research (ICBR). Subject folders with consents and hard-data were st ored in locked filing cabinets in the PI’s student office space at the UF College of Nursing. Research Variables The independent variables were diagnosis of HTN/NT and self-identified OMB racial category. Both were categorical, nomina l variables. SIR (self-identified race) was determined by self-report of one or more of the five categories as defined by the OMB (see Appendix A). Diagnosis was a dichotomous variable with either HTN or NT as variable choices. The dependent va riables were gene expression of 1Aand 2-ADR (continuous), genotype (categorical/nominal ), and post-operative positive inotrope administration (dichotomous, nominal). Gene expression was determined using the ABI

PAGE 58

45 Prism 7900 and ABI Assays on Demand for 1A-ADR (Hs00169124_m1) and 2-ADR (Hs00240532_s1) (Applied Biosystems, Foster City, CA). Relative gene expression of 1Aand 2-ADRs were utilized to determine the values of the gene expression variables, as outlined by Livak and Schmittgen (2001). Three ADR single nucleotide polymor phisms (SNPs) were examined. The 1A-ADR (Codon 347, refSNP ID:1048101) SNP is located on chromosome 8 at location 8p21, in the second exon, or coding region. The 2-ADR (Codon 16, refSNP ID:1042713 & Codon 27, refSNP ID: 1042714) SNPs are located on chromosome 5q32-34, both in the first (and only) exon of the gene. Figures 3.1 and 3.2 show the loci of investigated polymorphisms with relation to the 1Aand 2-ADR genes. Genotypes for the 1A-ADR (Codon 347), 2-ADR (Codon 16 & Codon 27) were determined by PCR followed by pyrosequencing (PSQHS96A System, Biotag e, Uppsala, Sweden ) on genomic DNA isolated from blood samples (PSQHS 96 Syst em, Uppsala, Sweden). Chart review was conducted to determine subjects’ need for post-operative positive inotrope medication. Figure 3-1. 1A -ADR gene with promoter, intron and exon boundaries and investigated polymorphism. 5 3 Key Promoter region Exon (coding) regions Intron (noncoding) regions 1A –ADR, Codon 347 polymorphism

PAGE 59

46 Figure 3-2. 2-ADR gene with promoter, exon boundary and investigated polymorphisms. Study Protocol Data Collection and Laboratory Methods Subjects completed a pencil and paper dem ographic form that provided information about their race, age, past medical history, medication use, height, weight, and income. As recommended by the Federal Drug Administration ([FD A], 2003), race was determined by self-report using the OM B revised race and et hnicity categories. Blood and tissue collection All blood and tissue samples were collected in the surgical suites. Afte r the patient was anesthetized for surgery, approximately 5-10 cc of arterial blood obtained from the central arterial line was placed in a purple -top tube containing EDTA and placed in a cooler with ice. The surgical remnants of IMA pedicle were cleaned by either the PI or the surgeon after removal from the patient and then placed in a sterile specimen container by the PI. The PI quickly (in a sterile field) cut the tissue to pieces smaller than 0.5 cm, and transferred the pieces immediately to a microtubule containing 100 microliters of RNAlater solution (Qiagen , Valencia, CA, USA) (see Fi gure 3-3). RNAlater is a nontoxic, aqueous tissue and cell storage reagent that protects cellular RNA in intact and unfrozen samples. It stabilizes RNA and preserves its integrity by halting mRNA degradation upon its infusion into the sample At this point, according to Qiagen’s 5 3 Key Promoter region Exon (coding) region untranslated regions 2-ADR Codons 16 & 27 polymorphism

PAGE 60

47 instructions, the RNA is protected from degr adation for 24 hours at 37 C, one month at 4 C, and indefinitely at 20 C. Completing these steps in a very quick manner minimizes RNA degradation and any changes in the mR NA expression level; thus, the PI worked very quickly to complete this process, wh ich often took less than 1 minute to complete. The samples were then placed in a cooler with ice and transported to the UF Center for Pharmacogenomics Core Laboratory, where they were incubated at 2-8 C at least overnight, but no more than 12 days, th en placed in a freezer at -80 C. Figure 3-3. Tissue pieces immersed in RNAlater preservation solution. Genomic DNA analyses Genomic DNA was isolated from blood lymphocytes using a commercially available kit (Qiagen DNA Blood Isolation Kit (Qiagen, Valencia, CA, USA). Genotype was determined by polymerase chain reaction (PCR), followed by pyrosequencing (Pyrosequencing, Uppsala, Sweden) (La ngaee & Ronaghi, 2005) using a PSQ HS96A single nucleotide polymorphism (SNP) reag ent kit according to the manufacturer’s protocol (Biotage AG, Upspsala, Sweden). In summary, 10 l of biotinylated PCR product was immobilized to streptavidin-coated Sepharose beads (Amersham

PAGE 61

48 Biosciences, Piscataway, NJ). After incubati on, the beads were isolated and treated with 70% ethanol, denaturation buffer, and wash buf fer. The beads then were released into designated wells containing annealing buffer and 10 pmol of sequencing primer, followed by a 2-minute incubation at 80C (Langaee & Ronaghi, 2005). The 1A-ADR (Arg347Cys) polymerase chain reaction (PCR) amplification was determined by using the primers listed in Tabl e 3-2.The PCR mixture consisted of 6.25 ul HotStarTaq Master Mix (Qiagen GmbH, Hilden, Germany), 0.75 l of dimethylsulfoxide (Sigma-Aldrich, St. Loui s, MO), 10 pmol of each primer (Operon Biotechnologies, Huntsville, AL), 1.5 l of water, and 50-100 ng of genomic DNA. The PCR amplification was performed under the fo llowing conditions: init ial denaturation at 95C for 15 minutes, 45 cycles of denaturation at 95C for 30 seconds, annealing at 56C for 30 seconds, and extension at 72C for 1 mi nute, followed by a final extension step at 72C for 7 minutes. The 2-ADR (Arg16Gly and Glu27Gln) polymerase chain reaction (PCR) amplifications were determined by using the primers listed in Table 4. Note that the same forward and reverse biotinylated primers were used; however, two different forward sequencing primers were used. The PCR mixture consisted of 12.5 ul HotStarTaq Master Mix (Qiagen GmbH, Hilden, Germany), 1.5 l of dimethylsulfoxide (Sigma-Aldrich, St. Louis, MO), 10 pmol of each primer (Operon Biotechnologies Huntsville, AL), 7 l of water, and 50100 ng of genomic DNA. The PCR amplification was performed under the following conditions: initial denaturation at 95C for 15 minutes, 40 cycles of denaturation at 95C for 30 seconds, annealing at 63C for 30 seconds, and extension at 72C for 1 minute, followed by a final extension step at 72C for 7 minutes.

PAGE 62

49 Table 3-2. Genotyping primers. Gene Primers Amplicon length 1A-ADR (Codon 347) Forward: CCCCATCATATACCCATGCT Biotinilated Reverse: GTAGCCCAGGGCATGTTTG Forward Sequencing Primer: TGTCTTGAGAATCCAGTGT Sequence to analyze: CTCT/CGCAGAAAGCAGTCT 109 2-ADR (Codon 16) Forward: CGAGTCCCCACCACACCC Biotinilated Reverse 5': AGCACATTGCCAAACACGATG Forward Sequencing Primer: CGGACCACGACGTCAC Sequence to analyz e: G/AGAAGCCATGCG 297 2-ADR (Codon 27) Forward 3': CGAGTCCCCACCACACCC Biotinilated Reverse 5': AGCACATTGCCAAACACGATG Forward Sequencing Primer: TGGCTGGCACCCAAT Sequence to analyze: GCAGC/GAAAGGGACGA 297 RNA isolation and reverse-transcription Once all tissue samples were collecte d, tissue processing for RNA extraction began. To avoid any degradation of RNA by RNAse, all surfaces and tools were thoroughly cleaned with either RNAZap (Ambion, Inc., Austin, TX, USA) or RNase AWAY (Molecular BioProducts, Inc., SanD iego, CA) and rinsed with diethylpyrocarbonate (DEPC) water. Tissues were removed from RNALater solution (Qiagen Valencia, CA, USA), gently blotted on kimwip es to remove excess solution (see Figure 3-4), weighed, quickly sliced into smaller pi eces, then transferred to a ceramic mortar. After the addition of a small amount of li quid nitrogen, the frozen samples were ground into a fine powder with a ceramic pestle (see Figures 3-5 & 3-6). The powdered tissue was then combined with 500 ul of propr ietary Lysis/Bindi ng solution from the RNAqueous Kit (Ambion, Inc., Austin, TX). The slush was then homogenized with a PowerGen 125 electric rotor-stator homogenizer (Fisher Scientific, Pittsburgh, PA) and Omni-Tips™ Plastic Disposable Generator Pr obes (Fisher Scientific, Pittsburgh, PA).

PAGE 63

50 (see Figure 3-7). After a 30-s econd centrifugation to remove large debris, the supernatant was removed from the lysate and processed per the manufacturer’s protocol. All samples were eluted in 50 ul total volume of proprie tary Elution Solution, included in the kit. Total RNA was quantified by Nanodrop (Nanodrop Technologies, Wilmington, DE). The Nanodrop determined seven samples to have concentrations less than 10 ng/ul. These seven samples were placed in a Cenrivap Console speed-vacuum (Labconco, Kansas City, MO) on the no-heat setting for approxi mately 20 minutes. These samples were then reconstituted in 20 ul of R NAqueous Kit’s Elution Solution. All 260/280 ratios were above 1.7. UF ICBR Core staff evaluated qua lity of 18s and 28s peaks generated by a 2100 Bioanalyzer (Agilent Technologies, Palo Alt o, CA). Quality of peaks were consistent across all samples, indicating little RNA degradation. Next, RNA aliquots were made to equal 10 ng/ul and brought up to 50 ul total volume with RNAse-free water. Samples were then reverse-tr anscribed with the cDNA Archive Kit (Applied Biosystems, Foster City, CA) at 25C for 10 minutes, followed by 2 hours at 37C in a thermal cycler. All samples were stored at –20C.

PAGE 64

51 Figure 3-4. Blotting tissue on Kimwipe. Figure 3-5. Grinding tissue in mort ar and pestle on liquid nitrogen. Figure 3-6. Powdered tissue in mortar.

PAGE 65

52 Figure 3-7. Homogenizing tissue slush with rotar-stator homogenizer. Real-time polymerase chain reaction Twenty microliter (ul) reactions were pr epared for single-plex Real-Time PCR with the ABI PRISM 7900 system (Applied Biosystems, Foster City, CA), located in the UF Interdisciplinary Center for Biotechnology Research (ICBR). Primers and probes for each of the three assays ( 1A-ADR, 2-ADR, and GAPDH) are listed in Table 3-3. The housekeeping gene GAPDH was used for normali zation of gene expression data, as described in the section titled “Genetic Analysis Techniques” in Chapter 2. For each reaction, 10.0 ul of TaqMan Universal PCR Master Mix (2X) with AmpErase UNG was prepared with 1.0 ul of each respective TaqMan Gene Expression Assays on Demand (20X) (Table 3-3). Eleven microliters of each master mix and 9.0 ul of cDNA template was added to each well of the 96-well plate. Triplicate samples were run, as recommended by Bustin (2000) to in crease accuracy of the methodology. Table 34 shows the plate set-up for si ngle-plexing of these three assays. PCR conditions were 50C for 2 minutes, 95C for 10 minutes, follo wed by 40 cycles of 95C for 15 seconds, and 60C for 1 minute. Fluorescence data we re processed and analyzed with the ABI

PAGE 66

53 PRISM Sequence Detection Software (Applied Biosystems, Foster City, CA). Results were expressed as Ct number (number of cycl es needed to generate a fluorescent signal above a predetermined threshold) or Ct (target Ct ( 1A-ADR or 2-ADR) minus normalizer Ct (GAPDH)). The Ct value was determined with the ABI 7900 software. The software determines the baseline automatical ly by assessing the normalized fluorescence signal versus cycle data, per plate. From this baseline, each sample’s Ct value is obtained. Table 3-3. Target gene assay information. Gene Assay # Probes/Quenchers Amplicons (base pairs) 1AADR Hs00169124_m1Probe: FAM 112 2-ADR Hs00240532_s1 Probe: FAM 65 GAPDH 4310884E (ABI product #) Probe: VIC Quencher: TAMRA 226 Table 3-4. Single-plex plate set-up, one sample. Sample 1 + GAPDH Sample 1 + GAPDH Sample 1 + GAPDH Sample 1 + 1A-ADR Sample 1 + 1A-ADR Sample 1 + 1A-ADR Sample 1 + 2-ADR Sample 1 + 2-ADR Sample 1 + 2-ADR Positive inotrope data collection The PI examined the subjects’ need fo r standard-of-care positive inotrope administration in the immediate post-operativ e period. Chart review was conducted in the post-operative phase for intensive care unit (ICU) documentation of administration of positive inotropic medications. The pharmaceutical agent and dosage were documented. ICU chart was reviewed for presence of the exclusion criteria, diagnosis of low cardiac output syndrome. The need for positive inot rope administration (dichotomized) was

PAGE 67

54 tested for relationship with 1Aand 2-ADRs gene expression of hypertensive versus NT groups. Calculations for Relative Gene Expression and Selection of Calibrator All gene expression data were imported into EXCEL for relative gene expression analyses. As previously stated, samples were run in triplicates for determination of the threshold cycle (Ct) in TaqMan RT-PCR. To co ntrol for outlier Ct values, the largest Ct value from each triplicate was removed a nd the duplicate values were averaged to determine the ‘average Ct’. The largest Ct values were chosen for removal because for the majority of the samples, one of the three raw Ct values was greater than 2 Ct’s away from the next closest value, indicating, in esse nce, an outlier. To maintain consistency in this process, each triplicate had the highest value removed. Next, the Ct averages were used to calculate the following: a) Delta Ct = average target Ct – average endogenous c ontrol (GAPDH) Ct; b) Delta Delta Ct = (Delta Ct (sample x) – (Delta Ct (calibrator)); c) 2^-(Delta Delta Ct) = two to the negative power of the DDCt; gene e xpression relative to the calibrator. A requirement of the DDCt method of relati ve quantitation requi res selection of a calibrator. One subject (#006142) was chosen as the calibrator. This subject was a normotensive, White/Caucasian male who was not taking any medicatio ns at the time of the study, and reported no other cardiac diagnos es. His BMI was comparable to the mean (30.9 versus mean of 29.0). He also reported not having ever been a smoker and did not drink or exercise. The determination of “fold difference” betw een groups is expressed as a ratio of the measures of central tendency for the groups compared. Said another way, the fold difference is a ratio of the one measure of cen tral tendency to another, so that if the

PAGE 68

55 median 2^-DDCt of group A was 25.0 and the median 2^-DDCt of group B was 5.0, then the ratio is 25.0: 5.0, indicating a 5-fold diffe rence between groups; or similarly, a 5-fold decrease in gene expression in grou p A versus B. All fold-diffe rence data were calculated in this fashion. The relationship between the Ct value and gene expres sion is indirect, in that the lower the Ct, the highe r the gene expression and vice versa. This same principle applies even after normalizati on, so that the 2^-DDCt value holds the same interpretation. Briefly stated, during amplification in the Real-Time RT-PCR system, the earlier the mRNA’s amplification is detected (thus, th e lower the Ct), the more abundant the mRNA. Conversely, if it takes longer fo r the amplification to be de tected (producing a higher Ct value), then the mRNA is less abundant. Methods for Statistical Analyses Data were analyzed using Microsoft Ex cel (Microsoft Corporation, Redmond, WA) and SPSS Version 14 (SPSS Inc., Chicago, IL). Desc riptive statistics were used to obtain summary measures for the data. Tests of norma lity for the gene expression data indicated non-normal distributions, necessitating use of nonparametric tests. To quantify the differences in gene expression of 1Aand 2-ADRs in the IMA by diagnosis of HTN versus NT (specific aim 1), the Mann-Whitn ey U test was performed. The Mann-Whitney U test was also used to explore relative differences in gene expression of the 1Aand 2ADRs genes and diagnosis, by SIR (specific aim 2), and to explore the relationship between level of 1Aand 2-ADR gene expression and the need for post-operative positive inotrope administration (specific aim 3). To test the association of diagnosis (HTN vs. NT) and the 1AArg347Cys C T genotype, Pearson Chi-square, and where necessary for nonparametric data, Fisher’s Exac t tests were used. To test the association

PAGE 69

56 of diagnosis (HTN vs. NT) and the 2-ADR Arg16Gly G A, and the 2-ADR Glu27Gln C G genotypes, Fisher’s exact test was use d. The Kruskall-Wallis test was used for testing the association between: a) 1A-ADR relative gene expression and the 1A-ADR Arg347Cys C T genotype; b) 2-ADR relative gene expression and each of the 2-ADR Arg16Gly G Aand Glu27Gln C G genotypes. All hypotheses were two-tailed and tested with alpha set at 0.05.

PAGE 70

57 CHAPTER 4 RESULTS Introduction The primary aim of this exploratory, pilot study was to determine relative differences in gene expression of the 1Aand 2-ADR genes between people with and without high blood pressure. The secondary ai m was to determine the influence of race on differences in gene expression. A tertiary aim was to examine the impact of gene expression of the 1A-ADR and 2 subtypes on the need for post-operative positive inotrope pharmacotherapy. This chapter will fi rst present descriptiv e results, including means, standard deviations, and frequency da ta for all variables investigated. The three hypotheses posed in Chapter 1 will be addr essed using the Mann-Whitney U, Pearson Chi-square, and Fisher’s exact tests. Cohen’s d effect sizes will also be provided. For SIR, the OMB Classification tool allowed fo r multiple choices of race. Three subjects self-identified as having tw o races: White/Caucasian and American Indian or Alaska Native. When statistical analyses included th e variable SIR, these three subjects were recoded as “White”. Explanatory aims E1 and E2 involving genotype by diagnoses associations and genotype by gene expre ssion associations are also included. Descriptive Results Subject Demographics Fifty one subjects were enrolled in the study between August 2004 and July 2005. Four subjects were excluded because bl ood and tissue samples were unable to be collected due to surgery scheduling changes. As a result, 47 subjects were included in the

PAGE 71

58 data analyses. This sample consisted of 37 males and 10 females with an overall mean age of 56.5 years (range 44-70). Thirty-seven of the subjects were recruited from Shands at Alachua General Hospital and 10 from the VAMC. Twenty subjects were normotensive and 27 subjects were hypertensi ve. The normotensive group ranged in age from 44-67 with a mean of 55.8 years. The age of the hypertensive group ranged from 44-70 with a mean of 57.3 years. These s ubjects are included in analyses involving genotyping. Table 4-1 shows the de mographic characteristics of this data set, expressed in numbers and percentage. Table 4-2 presents subjects’ clinical characteristics, including height, weight, and BMI, expressed as mean and standard deviation, and prescribed blocker/dose, concomitant diagnosis of di abetes mellitus-Type2 and surgery facility, expressed as number and percentage. Table 4-1. Demographics of all enrolled subjects. All enrolled N = 47 Normotensive n = 20 Hypertensive n = 27 N % n % n % Gender Male Female 37 10 78.7 21.3 18 2 90.0 10.0 19 8 70.4 29.6 Race White/Caucasian Black/AA White/Caucasian & Native American 34 10 3 72.3 21.3 6.4 17 2 1 85.0 10.0 5.0 17 8 2 63.0 29.6 7.4 Ethnicity Non-Hispanic Hispanic Did not know 45 1 1 95.7 2.1 2.1 20 0 0 100 0 0 25 1 1 92.6 3.7 3.7 Table 4-2. Clinical characteristic s of all enrolled subjects. All N = 47 Normotensive n = 20 Hypertensive n = 27 Height (in) 68.8 + 4.1 69.8 + 3.6 68.1 + 4.3 Weight (lbs) 197.4 + 40.0 192.6 + 36.9 199.7 + 42.8 BMI (kg/m2) 29.2 + 5.9 27.4 + 5.3 30.3 + 6.1

PAGE 72

59 Table 4-2 Continued. All N = 47 Normotensive n = 20 Hypertensive n = 27 -locker (Rx & dose) Not prescribed Metoprolol 12.5mg BID Metoprolol 25mg BID Metoprolol 50mg BID Metoprolol 75mg BID Metoprolol 100mg BID Labetalol 100mg TID Missing data 16 (34.0%) 5 (10.6%) 12 (25.5%) 10 (21.3%) 1 (2.1%) 1 (2.1%) 1 (2.1%) 1 (2.1%) 10 (52.6%) 3(15.8%) 3 (15.8%) 2 (10.5%) 0 (0%) 0 (0%) 0 (0%) 1 (5%) 6 (22.2%) 2 (7.4%) 8 (29.6%) 8 (29.6%) 1 (3.7%) 1 (3.7%) 1 (3.7%) 0 (0)% T2DM No Yes Pre-DM 30 (63.9%) 16 (34.0%) 1 (2.1%) 19 (95%) 1 (5%) 0 (0%) 11 (40.7%) 15 (55.6%) 1 (3.7%) Surgery facility AGH VA 37 (78.7%) 10 (21.3%) 15 (75 %) 5 (25%) 22 (81.5%) 5 (18.5%) BMI = body mass index; T2DM = Diabetes mellitus-Type 2; AGH = Alachua General Hospital; VA = Veterans Administra tion Hospital, BID = twice per day Six additional subjects were excluded from analyses involving gene expression for the following reasons: 1) The PI could not co llect tissue from one subject due to change in surgery schedule; and 2) Tissues from five subjects did not yield sufficient RNA material to complete the an alyses or had “undetermined” readings in the TaqMan RTPCR gene expression output. The final sample size for gene expression analyses was 41 subjects. This subset of 32 males and 9 fe males had a mean age of 57.3 (range 44-70). Of these 41 subjects, 17 were normotensive a nd 24 were hypertensive. The normotensive group ranged in age from 44-67 with a mean of 56.7 years. The age of the hypertensive group ranged from 45-70 with a mean of 57.7 y ears. See table 4-3 for the demographic summary by groups, expressed in numbers a nd percentage. Table 4-4 presents these subjects’ clinical characteris tics, including height, weight, and BMI, expressed as mean and standard deviation, and prescribed -blocker/dose, concomitant diagnosis of diabetes

PAGE 73

60 mellitus-Type2 and surgery facility, expresse d as number and percentage. Student’s ttests confirmed that hypertensive and normote nsive groups did not significantly differ in age ( t = -0.803, df = 45, p = 0.426), height ( t = 1.374, df = 44, p = 0.177), weight ( t = 0.463, df = 45, p = 0.646), or BMI ( t = -1.500, df = 44, p = 0.141). Table 4-3. Demographics fo r gene expression subset. Total subset N = 41 Normotensive n = 17 Hypertensive n = 24 N % N % N % Gender Male Female 32 9 78.0 22.0 15 2 88.2 11.8 17 7 70.8 29.2 Race White/Caucasian Black/AA White/Caucasian & Native American 29 9 3 70.7 22.0 7.3 14 2 1 82.5 11.8 5.9 15 7 2 62.5 29.2 8.3 Ethnicity Non-Hispanic Hispanic Did not know 39 1 1 95.1 2.4 2.4 17 0 0 100 0 0 22 1 1 91.7 4.2 4.2 Table 4-4. Clinical characteristic s for gene expression subset. Subset n = 41 Normotensive n = 17 Hypertensive n = 24 Height (in) 68.7 + 4.3 69.8 + 3.7 67.9 + 4.5 Weight (lbs) 195.8 + 41.5 193.1 + 39.1 197.7 + 43.9 Subset n = 41 Normotensive n = 17 Hypertensive n = 24 BMI (kg/m2) 29.0 + 6.0 27.5 + 5.5 30.2 + 6.2 -blocker (Rx & dose) Not prescribed Metoprolol 12.5mg BID Metoprolol 25mg BID Metoprolol 50mg BID Metoprolol 75mg BID Metoprolol 100mg BID Labetalol 100mg TID Missing data 15 (36.6%) 4 (9.8%) 10 (24.4%) 8 (19.5%) 1 (2.4%) 1 (2.4%) 1 (2.4%) 1 (2.4%) 9 (52.9%) 2 (11.8%) 3 (17.6%) 2 (11.8%) 0 (0%) 0 (0%) 0 (0%) 1 (5.9%) 6 (25.0%) 2 (8.3%) 7 (29.2%) 6 (25.0%) 1 (4.2%) 1 (4.2%) 1 (4.2%) 0 (0%) T2DM No Yes Pre-DM 27 (65.9%) 13 (31.7%) 1 (2.4%) 17 (100%) 0 (0%) 0 (0%) 10 (41.7%) 13 (54.2%) 1 (4.2%) Surgery facility AGH VA 32 (78.0%) 9 (22.0%) 12 (79.6%) 5 (29.4%) 20 (83.3%) 4 (16.7%) BMI = body mass index; T2DM = Diabetes mellitus-Type 2; AGH = Alachua General Hospital; VA = Veterans Administra tion Hospital, BID = twice per day

PAGE 74

61 Assessment of GAPDH for Relative Quantitation The duplicate GAPDH Ct values had a m ean and standard deviation of 27.53 + 2.80 and median of 27.4 with values rangi ng from 22.1-34.2. Figure 4-1 displays the average duplicate Ct values for each sample showing this 12-point range. Not only should the raw triplicate Ct values be close (no more than one-half Ct different), the averages should show little variation across samples. Figure 4-2 shows these data again, grouped by 96-well plate number. Evaluation of these graphical data shows that this variance was not plate-specific, meaning each pl ate showed variation in Ct values for the GAPDH. Figure 4-1. Range of average duplicate Ct va lues of GAPDH per sample number. Note: Arrow indicates calibrator sample. Range in GAPDH Duplicate Ct Measurements20 22 24 26 28 30 32 34 36 1 Sample #Raw Ct Value 1 3 4 5 6 7 8 9 11 12 13 14 17 18 19 20 22 23 24 25 26 27 28 30 3 1 32 33 34 35 37 38 40 41 42 43 44 45 46 48 49 50

PAGE 75

62 0 5 10 15 20 25 30 35 plate 1plate 2plate 3plate 4plate 5plate 6 Plate #GAPDH Average Duplicate Ct Values by Plate # Figure 4-2. Range of average duplicate GAPDH Ct measurements grouped by plate number. Note: Arrow indicates calibrator sample. Nonparametric Spearman’s rho was perfor med to test correlations between the duplicate GAPDH Ct and each of the 1Aand 2-ADR duplicate Ct variables. GAPDH was significantly correlated with 1A-ADR duplicate Ct ( R = 0.628, p < 0.05). Conversely, GAPDH was not signif icantly correlated with 2-ADR duplicate Ct ( R = 0.247, p = 0.120). In addition, Student’s t-test confirmed that GAPDH duplicate Ct differed significantly between subjects with HTN versus NT ( t = -2.634, df = 39, p < 0.05). Figure 4-3’s boxplot represents the groupwise difference in GAPDH between hypertensive and normotensive subjects. Each boxplot contains a box with a bisecting line and two “whiskers” extending from either end. The upper and lower ends of the box represent the upper and lower qua rtiles, respectively; or, the cutoffs for the 75th and 25th percentiles, respectively. The line that bisects this box re presents the median, or middle value. The whiskers extend to the minimum and maximum values for the data. Figure 44’s boxplot shows the groupwise differences in GAPDH, 1Aand 2-ADR raw Ct values. These figures illustrate the variability in raw Ct values between groups and are not Average Ct Values

PAGE 76

63 informative of relative gene expression diffe rences. Further interpretation of these data and discussion of their importance is thor oughly presented in Chapter 5: Discussion. Figure 4-3. Boxplot for averag e duplicate GAPDH by diagnosis. Figure 4-4. Boxplot for all gene expression raw Ct values.

PAGE 77

64 Assumptions of Normality Data were assessed for normality with skew ness, kurtosis, and the Kolmogorov-Smirnof test. These measures of normality indicated th at relative gene expression variables (2^DDCt) for both 1Aand 2-ADR were non-normally distributed (skewness = 5.447, kurtosis = 32.127, p < 0.05; and, skewness = 4.302, kurtosis = 20.722, p < 0.05, for 1Aand 2-ADR, respectively). Due to these violat ions of normality, non-parametric tests were used for data analyses invol ving relative gene expression. Analytic Results for Hypotheses As these data were non-normally distri buted, medians and inter quartile ranges (IQRs) are presented for measures of central tendency and variance. These values for the total sample are listed in Table 4-5. Further analytic results are presented by aim. In addition, amplification plots for each gene, as expressed in the TaqMan Real-Time PCR system (ABI Prism 7900) are presented in Appendix D. Table 4-5. Gene expression median s and IQRs for total sample Total sample n = 41 Gene Min 25% Med 75% max 1A-ADR 2^-DDCt 0.004 0.336 0.63 1.766 30.484 2-ADR 2^-DDCt 0.009 0.105 0.32 1.000 40.224 Specific aim 1: To quantify differences in gene expression of 1Aand 2-ADR in the IMA between subjects with NT and HTN. a. To quantify relative differences in 1A-ADR gene expression between study groups with NT and HTN. b. To quantify relative differences in 2-ADR gene expression between study groups with NT and HTN. A summary of the median and IQR for each gene by diagnosis is presented in Table 4-6. Figure 4-5 shows boxplots of these data. For specific aim 1, the relative differences in 1A(aim 1a) and 2-ADR (aim 1b) gene expression between subjects with NT and

PAGE 78

65 HTN was examined using the Mann-Whitney U test. For this nonparame tric test, the null hypothesis is that the two variables compar ed have identical distributions. More specifically, it tests that the mean ranks of the 2^-DDCt values do not differ from the sum of the ranks (mean of ranks not to be confused with mean of data). The results for these tests are presented in Table 4-7. Median fold difference in gene expression of 1A-ADR and 2-ADR between subjects with NT and HTN were significant for p < 0.05. Folddifferences are expressed as a ratio of HTN to NT subjects. Table 4-6. Gene expression medians a nd IQRs for subjects by diagnosis. HTN n = 24 NT n = 17 Gene Min 25% Med 75% Max Min 25% Med 75% Max 1A-ADR 2^-DDCt 0.15 0.53 1.41 2.68 30.48 0.004 0.13 0.36 0.84 1.72 2-ADR 2^-DDCt 0.02 0.05 0.45 1. 70 40.22 0.008 0.15 0.22 0.45 14.1 A B Figure 4-5. Boxplots for both gene’s expressi on by diagnosis. A) Unadjusted scale with black box showing selection for resca ling. B) Y-axis rescaled for better visualization. Table 4-7. Median fold differences in gene expression between normotensive and hypertensive subjects and Mann-Whitney U tests. alpha < 0.05 Gene expression Ratio of subjects with HTN:NT median-fold-difference p 1A-ADR 2^-DDCt 3.92 0.01* 2-ADR 2^-DDCt 2.05 0.02* Hypertensive Normotensive Diagnosis NT vs. HTN 42 40 38 36 34 32 30 28 26 24 22 20 18 16 14 12 10 8 6 4 2 0 -2 18 45 4 24 38 13 45 30 4 B2 duplicate 2^ (DDCt) A 1A duplicate 2^ (DDCt)Gene expression differences by diagnoses

PAGE 79

66 Specific aim 2: To explore relative differen ces in gene expression of 1A-ADR and 2-ADR in the IMA between subjects with NT and HTN by race. a. To explore relative differences in 1Aand 2-ADR gene expression between White/Caucasian subjects with NT and HTN. b. To explore relative differences in the 1Aand 2-ADR gene expression between White/Caucasians with HTN versus Black/AAs with HTN. For aim 2a, a summary of the median and IQR for each gene in White/Caucasian subjects is presented in Table 4-8. Figure 4.6 shows boxplots of thes e data. To test the hypothesis that the relative fold-differe nces in gene expression of the 1A-ADR and 2ADR may be due, in part, to race (Aim 2a), Mann Whitney U test was performed to compare gene expression differences between White/Caucasians with and without HTN. When Caucasian hypertensive versus normote nsive subjects were compared, ranks of relative difference remained si gnificant between median fold -differences in each gene’s expression. The fold-difference is expressed as a ratio of Wh ite/Caucasian subjects with HTN to NT. These data are presented in Table 4-10. Table 4-8. Gene expression medians, IQRs and minimum and maximum values for White/Caucasian subjects. White/Caucasian n = 32 Gene Min 25% Med 75% max 1A-ADR 2^-DDCt Total Hypertensive ( n = 17) Normotensive ( n =15) 0.004 0.146 0.004 0.261 0.381 0.103 0.63 1.45 0.36 1.71 3.19 1.00 30.484 30.484 1.717 2-ADR 2^-DDCt Total Hypertensive ( n = 17) Normotensive ( n =15) 0.009 0.021 0.009 0.079 0.223 0.042 0.37 0.66 0.22 1.25 2.82 0.42 40.224 40.224 14.026

PAGE 80

67 A B Figure 4-6. Boxplots for White/Caucasian subjects, for both gene’s expression by diagnosis. A) Unadjusted scale wi th black box showing selection for rescaling. B) Y-axis rescaled for better visualization. Table 4-9. Median fold differences in gene expression between White/Caucasian normotensive and hypertensive subjects and Mann-Whitney U tests. Gene expression Ratio of subjects with HTN:NT Median fold-difference P 1A-ADR 2^-DDCt 4.03 0.01* 2-ADR 2^-DDCt 5.27 0.02* *alpha < 0.05 For aim 2b, a summary of the median and IQR for hypertensive subjects by SIR is presented in Table 4-11. Figur es 4-7 and 4-8 show boxplots of these data, by gene. For aim 2b, the Mann-Whitney U test was performed to compare Caucasian HTN versus Black/AA HTN. This comparison did not show significance, as presented in Table 4-11. This table also shows the median fold-differences in each gene’s expression, expressed as a ratio of White/Caucasian HTN to Black/AA HTN subjects. Table 4-10. Gene expression medians and IQRs for Black/AA subjects. Hypertensives n = 27 Gene Min 25% Med 75% max 1A-ADR 2^-DDCt White/Caucasian ( n = 15) Black/AA ( n =9) 0.146 0.507 0.381 0.620 1.45 0.99 3.19 2.53 30.484 2.732 2-ADR 2^-DDCt White/Caucasian ( n = 15) Black/AA ( n =9) 0.021 0.055 0.223 0.095 0.66 0.17 2.82 1.00 40.224 11.004 Hypertensive Normotensive Diagnosis NT vs. HTN 42 40 38 36 34 32 30 28 26 24 22 20 18 16 14 12 10 8 6 4 2 0 -2 18 45 4 38 13 45 4 B2 duplicate 2^ (DDCt) A1A duplicate 2^ (DDCt)

PAGE 81

68 white black or AA Recoded SIR 35.000 30.000 25.000 20.000 15.000 10.000 5.000 0.000 A1A duplicate 2^ -(DDCt) 45 4 Hypertensive Normotensive Diagnosis NT vs. HTN A B Figure 4-7a. Boxplots for 1A-ADR gene expression for White/Caucasian HTN versus Black/AA HTN subjects. A) Unadjust ed scale with black box showing selection for rescaling. B) Boxplot rescaled. A B Figure 4-8. Boxplots for 2-ADR gene expression for White/Caucasian HTN versus Black/AA HTN subjects. A) Unadjust ed scale with black box showing selection for rescaling. B) Boxplot rescaled. Table 4-11. Median fold differences in gene expression between White/Caucasian hypertensive and Black/AA hypertensive s ubjects and Mann-Whitney U tests. Gene expression Ratio of Cauc/White HTN:Black/AA HTN median-fold-difference p 1A-ADR 2^-DDCt 1.47 0.55 2-ADR 2^-DDCt 3.88 0.28 alpha < 0.05 Specific aim 3: To explore the relationship between level of 1Aand 2-ADR gene expression and need for post-operative posit ive inotropic medicat ion administration. white black or AA Recoded SIR 42.000 40.000 38.000 36.000 34.000 32.000 30.000 28.000 26.000 24.000 22.000 20.000 18.000 16.000 14.000 12.000 10.000 8.000 6.000 4.000 2.000 0.000 -2.000 B2 duplicate 2^ (DDCt) 24 45 4 13 18 38 Hypertensive Normotensiv e Diagnosis NT vs. H T

PAGE 82

69 Table 4-12 shows the median, IQR, minimum and maximum values for both genes by need for positive inotropes. Figure 4-9 show s the boxplots of these data. To test the hypothesis that fold-differences in gene e xpression exist between subjects who required post-operative positive inotrope administrati on and those who did not, the Mann-Whitney U test was performed (Table 4-13 ). Median fold-difference of 1A-ADR and 2-ADR gene expression between those who did a nd did not require po st operative positive inotropes is also shown in Table 4-13. Fold difference is expressed as a ratio of subjects who received inotrope treatment to those who did not. Table 4-12. Median, IQR, minimum and maximum values for 1A-ADR and 2-ADR fold difference in gene expression and need for post-operative positive inotrope medication. Inotropes n = 41 Gene Min 25% Med 75% Max 1A-ADR 2^-DDCt No inotropes ( n = 34) Yes inotropes ( n = 7) 0.004 0.339 0.301 0.507 0.625 0.727 1.95 1.59 30.484 1.670 2-ADR 2^-DDCt No inotropes ( n = 34) Yes inotropes ( n = 7) 0.009 0.021 0.089 0.248 0.273 0.451 0.93 4.47 40.224 14.026 A B Figure 4-9. Boxplots for both genes’ expr ession by need for post-operative positive inotrope. A) Unadjusted scale with bl ack box showing selection for rescaling. B) Boxplot rescaled. yes NO Dichotomous: inotrope administered 40 20 0 45 4 24 22 18 45 4 B2 duplicate 2^ (DDCt) A 1A duplicate 2^ (DDCt)

PAGE 83

70 Table 4-13. Fold differences in gene expr ession between non-inotrope and inotrope subjects and Mann-Whitney U tests. Gene expression Ratio of inotrope:non-inotrope Median fold-difference p 1A-ADR 2^-DDCt 1.18 0.73 2-ADR 2^-DDCt 1.67 0.36 alpha < 0.05 Exploratory Aims Following statistical analyses of the th ree specific aims, further exploratory analyses were completed. For exploratory ai m A, genotype data were examined for the 1A-ADR Arg347Cys C T, the 2-ADR Arg16Gly G A, and the 2-ADR Glu27Gln C G polymorphisms. Tables 4-14 and 4-15 indi cate allele and genotype frequencies for 47 subjects who had blood collected, separate d by racial/ancestral groups. Population values for the 1A-ADR Arg347Cys C T polymorphism were obtained from the Ensembl database (http://www.ensembl.or g/index.html). Population estimates for European Americans were from 24 samples and estimates for African Americans were from 23 samples form the Coriell Cell re pository (Ensembl, 2005). Population estimates for the 2-ADR Arg16Gly G A, and the 2-ADR Glu27Gln C G polymorphisms were obtained from the Pharmacogenetics and Pharmacogenomics Knowledge Base (Pharm GKB) database of the INVEST study ( IN ternational VE rapamil SR and Trandolapril St udy, unpublished data), whereby 325 African American and 1,100 Cacuasian/European American subjects were gentoyped (PharmGKB, 2006) All genotypes were determined to be in Hardy-Weinberg Equilibrium (data not shown), indicating that the gene frequencies and genotype rati os remained constant from generation to generation in a randomly-breeding population.

PAGE 84

71 Table 4-14. Allele frequencies for popul ation versus sample, by SIR/ancestry. Amino Acid/ Allele Population Allele Frequency Sample Allele Frequency Gene/ Ref ID # Codon Major Minor Black/AA Cauc/ Eur-Amer Black/AA n = 10 Cauc/ Eur-Amer n = 37 1A1048101 347 Cys T Arg* C T .28 C .72 T .56 C .44 T .20 C .80 T .49 C .51 21042713 16 Gly A Arg G G .48 A .52 G .40 A .60 G .55 A .45 G .64 A .36 21042714 27 Gln C Glu G G .18 C .82 G .40 C .60 G .35 C .65 G .40 C .60 *In the AA population, Arg is the major allele Table 4-15. Genotype frequencies for popula iton versus sample, by SIR/ancestry. Amino Acid/ Allele Population Genotype Frequency Sample Genotype Frequency Gene/ Ref ID # Codon Major Minor Black/AA Cauc/ Eur-Amer Black/AA Cauc/ Eur-Amer 1A1048101 347 Cys T Arg* C T/T .04 T/C .48 C/C .48 T/T .17 T/C .54 C/C .29 T/T .10 T/C .20 C/C .70 T/T .32 T/C .41 C/C .35 21042713 16 Gly A Arg G G/G .27 A/G .50 A/A .23 G/G .37 A/G .46 A/A .17 G/G .30 A/G .50 A/A .20 G/G .41 A/G .60 A/A .12 21042714 27 Gln C Glu G G/G .03 G/C .30 C/C .67 G/G .16 G/C .49 C/C .35 G/G .10 G/C .50 C/C .40 G/G .27 G/C .38 C/C .44 *In the AA population, Arg is the major allele. Chi-square analyses were performed to examine the association between each of the genes by diagnosis (HTN vs. NT). When cells contained values less than 5, Fisher’s Exact tests were used for nonparametric da ta. Table 4-16 shows these results. All Chisquare/Fisher’s Exact test s were nonsignificant for p < 0.05.

PAGE 85

72 Table 4-16. Association between genotype and diagnoses of NT and HTN for the 1AADR and 2-ADR genes. Gene NT n = 20 HTN n = 27 p Count % within diagnosis Count % within diagnosis 1A-ADR Codon 347 C/C C/T T/T 5 8 7 25.0 40.0 35.0 14 8 5 51.9 29.6 18.5 0.18† 2-ADR, Codon 16 G/G G/A A/A 8 10 2 40.0 50.0 10.0 8 14 5 29.6 51.9 18.5 0.67§ 2-ADR, Codon 27 C/C C/G G/G 3 7 10 15.0 35.0 50.0 7 11 9 25.9 40.7 33.3 0.52§ † = Chi-square § = Fisher’s Exact Figure 4-10. Bar chart of 1A-ADR, codon 347 by diagnosis.

PAGE 86

73 Figure 4-11. Bar chart of 2-ADR, codon 16 by diagnosis. Figure 4-12. Bar chart of 2-ADR, codon 27 by diagnosis. To examine the effect of confounding or population stratification of race on genotype differences in HTN versus NT subj ects, Fisher’s exact was performed, but only on White/Caucasian subjects. Tables 4-17 show th ese data and indicate these associations

PAGE 87

74 are not significant. Given the allele a nd genotype frequency differences between White/Caucasian and Black/AA subjects (for both sample and population estimates), further analyses comparing associations st ratified by SIR are warranted. However, very low cell counts for the Black/AA group in this sample prevent any meaningful analyses with data analyzed by genotype. When data we re analyzed by allele for each gene with Chi-square, cell counts were sufficient to ex amine allele by diagnos is associations for both total sample and the White/Caucasian group, but were still too low in the Black/AA group to warrant meaningful analyses. These data are presented in Tables 4-18 and 4-19. Table 4-17. Fisher’s Exact for genotype differe nces in White/Caucasian hypertensive vs. normotensive subjects. White/Caucasian n = 37 Gene NT n = 18 HTN n = 19 Counts Count % within diagnosis Count % within diagnosis P 1A-ADR, Codon 347 C/C C/T T/T 3 8 7 16.7 44.4 38.9 9 6 4 47.4 41.6 21.1 0.13§ 2-ADR, Codon 16 G/G G/A A/A 8 8 2 44.4 44.4 11.1 6 11 2 31.6 57.9 10.5 0.69§ 2-ADR, Codon 27 C/C C/G G/G 3 7 8 16.7 38.9 44.4 6 6 7 31.6 31.6 36.8 0.57§ Table 4-18. Chi-square for allele counts by diagnosis for the 1A-ADR and 2-ADR genes in all subjects. All subjects N = 47 Gene NT n = 20 HTN n = 27 Counts Allele frequency % count Allele frequency % count p

PAGE 88

75 Table 4-18 Continued. All subjects N = 47 Gene NT n = 20 HTN n = 27 Counts Allele frequency % count Allele frequency % count p 1A-ADR, Codon 347 C T .45 .55 33.0 55.0 .67 .33 66.0 45.0 0.03* 2-ADR, Codon 16 G A .65 .35 46.0 37.0 .56 .44 54.0 63.0 0.04 2-ADR, Codon 27 C G .33 .67 34.0 48.0 .46 .54 66.0 52.0 0.18 alpha < 0.05. Note: Percent coun t equals row count. Table 4-19. Chi-square for al leles by diagnosis for the 1A-ADR and 2-ADR genes in White/Caucasian subjects. White/Caucasian n = 37 Gene NT n = 18 HTN n = 19 Counts Allele frequency % count Allele frequency % count p 1A-ADR, Codon 347 C T .39 .61 38.9 61.1 .63 .37 63.1 36.8 0.04* 2-ADR, Codon 16 G A .67 .33 66.7 33.3 .61 .39 60.5 39.5 0.58 2-ADR, Codon 27 C G .36 .64 36.1 63.9 .47 .53 47.4 52.6 0.33 alpha < 0.05 To explore if genotype differences we re correlated with gene expression differences, Kruskal-Wallis tests were performe d. This test is a nonparametric alternative to a One-Way Analysis of Variance that extends the Mann-Whitney U test to more than two groups. This is necessary, as we need to examine three groups for the genotypes. No significant differences were found for genot ype by gene expression. These data are presented in Table 4-20.

PAGE 89

76 Table 4-20. Kruskal–Wallis tests for genotype counts by gene expression 1A-ADR and 2-ADR genes. n = 41 Genotype X Gene expression Count Median gene expression p 1A-ADR, Codon 347 X 1A-ADR 2^-DDCt C/C C/T T/T 17 13 11 0.99 0.58 0.42 0.49 2-ADR, Codon 16 X 2-ADR 2^-DDCt G/G A/G A/A 14 21 6 0.55 0.67 1.15 0.80 2-ADR, Codon 27 x 2-ADR 2^-DDCt C/C C/G G/G 9 14 18 1.00 0.63 0.57 0.33 Effect Sizes and Power Calculations Cohen’s d (actual effect sizes) were calculated for all aims and are presented in Table 4-21. Actual effect si zes varied by aims. The original power calculations anticipated a medium effect size, based on the literature. Therefore, these Cohen’s d values indicate a greater sample size was n eeded to power these aims. With such small effect sizes, the group sizes would need to be greater than what was sampled to achieve a power of 80%. Table 4-21. Power and effect sizes by aim. Calculations Aim Statistical test Power Effect Size Number neededa 1a* Total sample by diagnosis: 1A-ADR Mann-Whitney U 0.66 0.67 34 1b* Total sample by diagnosis: 2-ADR Mann-Whitney U 0.35 0.35 72 2a* White/Caucasians by diagnosis: 1A-ADR Mann-Whitney U 0.56 0.56 30 2a White/Caucasians by diagnosis: 2-ADR Mann-Whitney U 0.30 0.40 64 2b Hypertensives by SIR: 1A-ADR Mann-Whitney U 0.31 0.35 58

PAGE 90

77 Table 4-21 Continued. Calculations Aim Statistical test Power Effect Size Number neededa 2b Hypertensives by SIR: 2-ADR Mann-Whitney U 0.17 0.27 100 Indicates these tests were significant in analyses. a The number needed in each group to ach ieve a power of 0.80. This assumes equal numbers in each group.

PAGE 91

78 CHAPTER 5 DISCUSSION AND RESULTS Introduction All descriptive and analytic results for th e proposed aims and exploratory analyses will be discussed in this chapter. Conclusi ons and implications for nursing as well as recommendations for future research will also be provided. Discussion of Results Demographics In this study, 42.6% of a ll enrolled subjects were normotensive and 57.4% were hypertensive. When examining those includ ed in the gene expression subset, the percentages were similar with 41.5% normo tensive and 58.5% hypertensive subjects. These percentages indicate that there ar e a number of normotensive patients undergoing bypass surgery, a phenomenon surprising to so me researchers. These normotensive subjects were comprised predominantly of self-identified White/Caucasians ( n = 17, 85%), with only 2 subjects (10%) who self-ide ntified as Black/AA. This study anticipated 15 subjects who were Black/AA with NT which, based on the UF TCV Surgery department’s database of patients from 2001-2002, 15 subjects would have amounted to roughly 60% of their total Black/AA NT population for that fiscal year. Only 8 subjects who self-identified as Black/AA and were di agnosed with HTN were recruited for the study. This made a total of 10 (21.3%) self-i dentified Black/AA subjects for the entire study. (This number declined by one for th e gene expression subset.) The 2000 U.S. Census for Alachua County estimates the Black/AA population to be just 19.3%. Based

PAGE 92

79 solely on this, this study attained a representa tive sample of self-i dentified Black/AA in Alachua County. However, the UF TCV Su rgery department’s database for 2001-2002 indicated approximately 100 of 389 patient s who had bypass surgery were listed as Black/AA. This led us to believe that sa mpling 30 Black/AA subjects should not be a problem. In addition, sampling occurred at both a community-based general hospital (Shands at AGH) and a regional Veteran’s me dical center, which we felt would possibly lead to a greater Black/AA population from whic h to sample. Despite these points, the PI was unable to recruit a sufficient cohort of Black/AAs. Very few NT Black/AA patients were identified for potential recruitment. Some possible reasons for this are that many of these patients could have been emergent cases (and unable to be consented 24 hours prior to surgery, as required by IRB), many of these patients may have refused surgical intervention. Finally, any normotensive patients, not just normotensive AAs, may have been referred for interventional procedures su ch as percutaneous coronary intervention (for example, stenting, atherect omy, or balloon catheter angiopl asty). This is perhaps the most plausible explanation for the reduced number of normotensive patients who undergo bypass surgery, in that they may have less se vere comorbidities and are recommended to interventional cardiology rather than to thoracic surgery for bypass. Only a small number of Black/AAs (regardless of diagnosis) appr oached to participate in the study chose not to enroll. Major reasons cited for not wishing to participate were: not wanting to be “bothered with anything else” and “not feeling comfortable with the study”. Only three subjects (6.4%) self-identified as having more than one race; all three considered themselves both White/Caucasian and American Indian/Alaska Native. This number could not be compared to US Census data, as the Census does not specifically

PAGE 93

80 report combinations of dual-identification. The percentage of self-ide ntified Hispanic and non-Hispanic were 2.1% and 95.7%, respectively. Compared to US Census reports for Alachua County, approximately 5.7% report Hispanic (of any race) and 94.3% report non-Hispanic ethnicities. Just over 21% of enrolled subjects were female and nearly 79% were male (22% and 78% female and male, respectivel y, in the gene expression subset). As indicated in the TCV Surgery department’s database, the average percent of women undergoing CABG surgeries by the TCV surgery department is 21%. This means that for every one female undergoing the procedure by this depart ment, there are approximately four males. Therefore, the enrolled percentage of female s met the expected percentage. This indicates that this sample is representative of th e population of females undergoing bypass surgery. Interpreted collectively, the demographic da ta of this sample suggest it to be moderately representative of the populati on of bypass patients who undergo surgery in Alachua County, but is not completely representa tive of the entire ra cial, ethnic or bypass populations. Clinical characteristics for subjects are pr esented in Tables 4-2 and 4-4 for enrolled and gene expression subsets, respectively. H ypertensive and normotensive groups did not significantly differ in age, height, weight, or BMI (refer to page 59). The concomitant diagnosis of T2DM was seen in 34% of the overall sample, with 2.1% ( n = 1) having a diagnosis of pre-diabetes. Only 5% ( n = 1) of all NT subject s were diagnosed with T2DM. Among all hypertensive subjects, 55.6% were Type 2 diabetic and 3.7% were pre-diabetic. In regards to prescribing of -blockers, a first-line class of drugs for both coronary artery disease and hypertensi on (AHA, 2005; JNC VII, 2003), a surprising

PAGE 94

81 percentage of subjects (34% overall) were not prescribed this medication. Twenty two percent of hypertensive and 53% of normotensive subjects were not on -blocker medication. Of subjects with NT, 47.4% were prescribed -blockers, but none were prescribed more than 50 mg, twice a day. Of subjects with HTN, 77.8% were prescribed this therapy with only three subjects (11.1%) taking more th an 50 mg, twice a day. Table 4.2 shows that the majority of subjects who were prescribed -blocker therapy (57.4% of the overall sample) were taking between 12.5mg to 50.0 mg, twice a day. Gene Expression Measures of Central Tendency and Variance When examining measures of central tende ncy in these data, a few things warrant consideration. First, the means are not the best representation of centrality of these data because multiple outliers skew these data and bi as the means. While the mean is typically considered more stable over time (and with repeated random selection), the median is considered a middle point, an index of averag e position, that is not affected by skewed data with outliers (P ortney & Watkins, 2000). When comparing the means and medians (data not shown), nearly every mean value is visibly inflated by the outliers and the medians appear to better represent the centr ality of these data. Furthermore, evaluating the standard deviations (SDs) causes more concern. The SDs are fairly large (data not shown), especially compared to the means and medians. Since the nature of the SD is to represent the variability in the data, it is ty pically expressed as the spread from each end of the mean (for example, ‘plus or minus’). If we were to subtract some of these SDs from their corresponding means, we would act ually be left with a negative number. For example, the mean and SD for the 1A-ADR relative gene expression for the total sample is 1.97 and 4.83, respectively. This would i ndicate a range of values from -2.86 to 6.8. From the point of view of the gene expre ssion biological assay, it is impossible to have

PAGE 95

82 negative numbers. This phenomenon occurs for nearly every set of values in this study. Therefore the SD is not the best measure of va riability for these data. Taken together, this information indicates the median and IQRs ar e the best representation of centrality and spread in these data and are presented for th ese data throughout the chapters (in tables and boxplots). Furthermore, because the medi an-derived 2^-DDCt values are similarly less influenced by outliers, median fold-d ifferences between groups was presented. Discussions for Choice of GAPDH for Normalization Gene During the initial planning phase of th is project, the normalization (or, housekeeping)gene anticipated for use was cyclo philin A, as it was previously reported as a successful normalizer in arte rial tissue (Lieu, Withycom be, Walker, Rong, Walzem, and Wong, et al., 2003; Trogan, Choudhury, Dans ky, Rong, Breslow, & Fisher, 2002). Prior to purchasing this housekeeping gene, another review of the litera ture was conducted to see if any new information had been reported about this gene’s use in normalization. In fact, a recent publication by Escobales and Cr espo reported evidence that reactive oxygen species appeared to be mediated by a number of factors, including cyclophilin A (2005). As the reactive oxygen pathway is implicat ed in HTN, the use of cyclophilin A to normalize samples of hypertensive subjects made this gene inappropriate for use as a housekeeping gene in this study. Another review of the literature re vealed support for the use of GAPDH in similar tissue types. Pe uster, Fink, Reckers, Beerbaum, and von Schnakenburg reported consistent amplificati on of GAPDH among samples in a study of unstented coronary arteries in pigs (2004). Wang and Brow n showed successful use of GAPDH for normalization in their study of -ADR subtypes in atrial appendages (2001). Furthermore, preliminary analyses with a small selection of the PI’s sample in singleplexed reactions yielded triplicate GAPDH values that were less than one-half a Ct from

PAGE 96

83 one another and Ct values between samples that were very close. Technical experts in the UF ICBR Gene Expression Core facility viewed these preliminary data and supported the decision to use GAPDH in the final experiments. These aforementioned references and preliminary results provided the foundati on for the decision to choose GAPDH as the housekeeping gene in this study. Assessment of the Performance of GAPDH as a Normalizer To reiterate, one of the major assumpti ons of performing the relative quantitation method of gene expression analyses is that the housekeeping, or normalization gene (here, GAPDH) expresses similarly across subj ects and/or experimental conditions. This is typically assessed by examining the Ct valu es of the GAPDH wherein the Ct values are expected to show very little variability ( no more than -2 Cts difference) and the standard deviation of the mean should be small. If the GAPDH gene expression shows greater variability than this, it is theoretically a poor housek eeping gene for the data and is cautioned for normalization use. In this particular study, the duplicate GAPDH Ct values had a mean and standard deviation of 27.56 + 2.80 and median of 27.4 with values ranging from 22.1 to 34.2. According to Dorak (2003), the endogenous control (housekeeping gene) should be more abundant (or, have smaller Ct values) than the target genes. This was true for the 1A-ADR (median 29.45, SD 2.50), but not the 2-ADR (median 26.10, SD 0.78). Possible explanations for this greater-than-expected Ct for GAPDH are poor PCR efficien cy or low copy numbers (Dorak, 2003). The 12-point range of Ct values for GAPDH indicates the GAPDH did not, in fact, express consistently across subjects. While poor PCR efficiency, low copy number, and/or pipetting errors may contribute to this variation in GAPDH another plausible explanation is the occurrence of RNA degradation. As previously stated, RNA degradation was determined

PAGE 97

84 via a 2100 Bioanalyzer (Agilent Technologies, Palo Alto, CA), and concluded that overall, the level of RNA degradation for thes e samples was low. However, to examine if RNA degradation may have played a role in GAPDH variation, the 18s and 28s graphical data were examined and compared to those samp les that had poor triplicate Ct values for GAPDH (meaning, those with more than 2 Cts difference). Qualitatively speaking this review of the data concluded that, at least for a handful of the samples, RNA degradation could help to explain some of the variance in GAPDH. The most extreme case of this was with one sample, whereby raw Ct values were 35.9, 30.9, and 27.9 and the graphical data from the bioanalyzer indicated more RNA de gradation in this particular sample as compared to other samples. Perhaps results for the GAPDH Ct values may have been less variable for some of the samples where RNA degradation was a potential issue. Figure 4-2 shows that variation in GAPDH was not plate-specific, meaning certain 96-well plates did not show more or less varia tion than others. This refutes the notion that specific plates may have been outliers due to the order of preparation, time lapse between preparation and initializa tion of Real Time RT-PCR, or othe r sources of external error. In addition, a stem and leaf plot (data not shown) indicated 3 “extremes” greater than 32.9; however, these values are not greater than 2 SDs from the mean. Similar variance has been reported in the literature for GAPDH in various species and tissues. Schmid, Cohen, Henger, Irrang, Schlndorff, and Kretzler (2003 ) showed variation by tissue source in their study, reporting GAPDH Ct median a nd standard deviation of 22.57 + 2.61 in tubulointerstitial compartments and 28.96 + 2.38 in glomeruli, both from human renal biopsies. They did not report a range of Ct values. Peuster, Fink, Reckers, Beerbaum and von Schnakenburg (2004) reported a median of 22.2 (range 19.8-26.9) of GAPDH Ct

PAGE 98

85 values in their examination of porcine left coronary artery. Despite this 7-point range, they reported consistent amplification acr oss all samples, as determined by serial dilutions of GAPDH. They used the Delta Delta Ct method, normalizing with GAPDH. Lennmyr, Ternt, Svynen, and Barbany ( 2005) discuss instability in GAPDH gene expression in their samples of middle cerebral ar tery in rats, but repor ted that because the changes were not statistically significant, th ey used the gene to normalize. Tricarico, Pinzani, Bianchi, Paglierani, Distante, and P azzagli, et al. (2002) graphically showed raw GAPDH Cts between approximately 22-23 in hu man breast tissue. They correlated their target genes with the GAPDH by Spearman’s rank (nonparametric), showing significance. This supported their decision not to use the GAPDH as a normalizer. The authors suggest normalization to total RNA concen tration as an alternative in this type of situation. The methodology behind this solution was not delin eated by the authors. The following arguments were discussed in pers onal communication with Y. Conley (2006) regarding this possible solution: First and fore most, if we assume that the expression of a gene is altered during a dis ease state (here expression of the ADR genes in HTN), then this could theoretically alter the total RNA. This would seem inappropriate, then, to normalize both diseased and non-diseased samp les with a total RNA value. Whatever value is used to normalize, it must be c onstant. This same principle applies when considering taking the average of all GAPDH values across all samples (or even by plate) and normalizing in this fashion, maki ng these options also undesirable. Barber, Harmer, Coleman, and Clark ( 2005) performed a thorough evaluation of GAPDH as a housekeeping gene, examining its expression in 72 human tissue types. One tissue type examined was coronary artery, although the specific artery was not noted.

PAGE 99

86 They reported (graphically) a mean Ct value of approximately 20, but no range. Among the 72 tissues examined, GAPDH mRNA gene e xpression varied 15-fold between tissue types, further supporting previously publis hed variability. The authors also reported GAPDH Ct outliers below 13 and above 32.761; in these cases, they removed these data points. These studies highlight the variability in reports of GAPDH in different tissues and species as well as variation in how re searchers handle the nor malization dilemma and outliers. No studies were found that repor ted expected raw Ct values for GAPDH in human IMA tissue, thus providi ng little evidence of an exp ected Ct value for GAPDH in the IMA tissues used in this study. Given the literature presented above, GAP DH was analyzed statistically for its correlation with the target ge nes and differences between ex perimental groups. There was a significant correlation be tween GAPDH duplicate Ct and 1A-ADR duplicate Ct but not between GAPDH duplicate Ct and 2-ADR duplicate Ct (data shown on page 62). The first glance indicates that GAPDH should not be used for normalization of the 1A-ADR gene expression data, but could be used for normalization of the 2-ADR. The reason for this is that a direct linear relationship should not exist be tween a target and housekeeping gene, theoretically sp eaking. The housekeeping gene should remain constant at any given value of the target gene expression. Additio nally, the housekeeping gene should not show differential expression between experimental groups. In fact, the GAPDH housekeeping gene expression in this study was signifi cantly different between hypertensive and normotensive subjects (see data page 65 and Fi gure 4-3). These data collectively indicate that the GAPDH used in this study did not optimally perform as a housekeeping gene. However, the data were normalized to GAPDH because no other options were available.

PAGE 100

87 Raw Ct values are unable to be used for comparison, as this is an exponential value determined from a log-linear plot of PCR signal versus cycle number (Livak & Schmittgen, 2001). No other housekeeping ge nes were used. For this reason, interpretation of statistically significant results is extremely cautioned and further inferential analyses would be inappropriate Boxplots of group differences and GAPDH values were shown (Figures 43 and 4-4) to allow for visual comparison of group data (as recommended in personal communication with N. Chegini, 2005). On a final note, use of single-plexing (loading targets and GAPDH in separate wells) instead of multiplexing (loading targets and GAPDH in same wells) at the very least, makes us confident that the values we obtained for the expression of the targets and GAPDH are more valid and reliable. This is because they were amplifie d separately and did not have to compete for reagents during cycling and de tection. A pitfall in single-plexing is that this less accurately controls fo r pipetting errors. Aims Hypertensive subjects showed a 3.92-fold difference in relative 1A-ADR gene expression compared to normotensive subjec ts, a difference that was statistically significant (aim 1a, Table 4-7). Examinati on of the median 2^-DDCt values for the 1AADR informs us that hypertensive subjects s howed nearly 4 times lower expression of the 1AADR gene in the arterial tissue investig ated, perhaps suggesting the possibility of blunted vasoconstriction in this group as compared to normotensives (note: the higher 2^DDCt median in the hypertensive group indi cates lower gene expression and thus, downregulation of the gene). It ha s previously been described that 1A-ADR downregulation could be explained as a conse quence of enhanced sympathetic tone (for example, increased vascular resistance) in HTN (Jacobs, Lenders, Willemsen & Thien,

PAGE 101

88 1997; Kinugawa, Endo, Kato, Kat o, Ahmmed, Omodani, 1997). This is likely the result of a negative feedback loop. This downregulation of the 1AADR gene in hypertensive subjects most similarly supports the work of Xu, Lu, Wei, Zhang and Han (1997), who reported a decrease in 1AADR gene expression in 12-month old systemic hypertensive rats (SHRs). Conversely, Reja and colleagues (2002) reported 1A-mRNA expression was significantly greater in SHT rat tissue samp les. Furthermore, Veglio and colleagues (2001) reported no difference in 1AADR gene expression between SHRs and WistarKyoto (normotensive strain) rats in either blood lymphocytes or aortas. Our results for the 1AADR gene expression appear to be both supported by and contradictory to other reports; however, no exact study design was available for true comparison. For the 2-ADR gene, hypertensive subjects show ed 2.05-fold difference in relative gene expression compared to normotensives, a difference that was also statistically significant (aim 1b, Table 4-7). Once agai n, hypertensive subjects showed reduced expression of the 2-ADR gene in the arterial tissue compared to normotensives. No true comparisons of this data to other reports in the literature regarding 2-ADR expression in HTN. Gaballa and colleagues (1998) reported enhanced 2-ADR mediated vasorelaxation in large artery walls of rats after adenovirally-mediated delivery of 2-ADRs. Similarly, Iaccarino and colleagues (2002) overexpressed 2-ADR in SHR and WKY rats and reported enhanced 2-ADR-mediated vasorelaxation in both rat strains, with a lessened vasodilatory effect in the hypertensive stra in. Based on further aims, the authors (2002) also concluded this vasorelaxation was directly related to 2-ADR signaling dysfunction and not endothelium-dependent nitric oxide metabolism. These studies support our data and provide a plausible explanation for th e downregulation in hypertensives; that

PAGE 102

89 decreased expression of 2-ADRs implies impaired vasorela xation in HTN. However, as explained by Bustin (2002), these values re present only steady-state mRNA levels and do not allow inference of these numbers on pos t-transcriptional f actors. In addition, normalization with GAPDH could be causing artifact in this data, as discussed previously. These differences in 1Aand 2-ADR relative gene expression remained significant when comparing onl y Caucasian NT versus HTN subjects (Aims 2a and b, Table 4-8). For the 1A-ADR gene, White/Caucasian hypertensives showed 4.03-fold difference in gene expression compared to nor motensives (Table 4-9) This same pattern of significantly reduced gene expression in hypertensives compared to normotensvies continued with the 2-ADR gene; there was a 5.27-fold difference in relative gene expression in hypertensives compared to normote nsive subjects. These data indicate that, in part, fold-differences in gene expressi on of the two genes may be related to the diagnosis of HTN in self-ide ntified White/Caucasian subjec ts. To see if SIR further attributed to these differences (when di agnosis is accounted fo r), self-identified White/Caucasian hypertensives were compared to self-identified Black/AA hypertensives (additional aim not previously listed). For the 1A-ADR, hypertensives who selfidentified as Black/AAs and White/Caucasia ns showed 1.47 fold-difference in gene expression, and a 3.88 fold difference for 2-ADR, with Black/AAs showing greater expression of the gene; however, these diffe rences were not signi ficant (see Tables 4-10 and 4-11). Given this trend of greater gene e xpression for both genes in the self-identified Black/AA group, larger group size s (note only 7 subjects for Black/AAs) may have the power to detect actual differences by SI R. Additionally, the GAPDH may be causing

PAGE 103

90 artifact in the data here. It is important to note that the additional between-group comparisons were planned (such as compar ing Black/AA HTN versus NT subjects); however, small sample sizes ( n = 7 and 2, respectively) fo r these groups impeded this goal. When examining the gene expression differences by need for inotrope pharmacotherapy, no significant fold-differences were found between those who did and did not require positive inotropes in the post-operative phase of recovery and gene expression of both genes (Aims 3a and b, Tabl es 4-12 and 4-13). Once again, these gene expression values were influenced by nor malization issues and should be cautiously interpreted. Exploratory Aims Sample versus population allele frequency comparisons Some minor differences were noted in allele frequencies between sample and population estimates (refer to Table 4-14). As previously explained, allele and genotype population estimates for the 1A-ADR (Arg347Cys) polymorphi sm were obtained from the Ensembl database (http://www.ensembl.org /index.html). No other allele or genotype frequency reports with larger samples we re found for this polymorphism. Population estimates for the 2-ADR Arg16Gly G A, and the 2-ADR Glu27Gln C G polymorphisms were obtained from the INVEST study (PharmGKB, 2006), as aforementioned. It is initially important to cl arify that the Arg allele is considered the major (wild-type) allele for the Black/AA population in the 1A-ADR (Arg347Cys) polymorphism, but is the minor allele for the European-descent populations (White/Caucasian groups) (Small, McGraw, & Liggett, 2003). The allele frequencies in this sample are mainly comparable to the population frequency reports, with the Arg (C)

PAGE 104

91 allele being slightly more frequent in the White/Caucasian sample than in the general population. However, it is important to note that these are subtle disp arities, especially due to the fact that all estimates we re obtained from small cohorts. For the 2-ADR (Arg16Gly) polymorphism, allele frequency comparisons had some distinctions. The Black/AA cohort had a slightly increased fre quency of the Arg (G) allele versus the population frequencies, but this is subtle considering the small sample size. For the White/Caucasian cohort, the same Arg (G) alle le was much more frequent in the sample as compared to the population estimates, with the frequencies being nearly inverted. This finding is most likely due to the small sample size of the subjects in this group; a larger sample may bring the sample values cl oser to the population estimates. For the 2-ADR (Gln27Glu) polymorphism,. The Glu (G) allele is much more frequent in the sample of Black/AAs than what is expected in the population estimate (35% versus 18%); however, this result is once again likely a reflection of small sample size. In the White/Caucasian group, the sample allele frequency estimates are an exact match to the population estimates indicated in the INVE ST database (PharmGKB, 2006). Sample versus population genotype frequency comparisons Similarly, minor disparities were note d between sample and population genotype frequencies (refer to Table 4-15). For the 1A-ADR (Arg347Cys) polymorphism, Black/AA homozygotes for Arg (C/C genotype) had a higher frequency in the sample than in the population (70% versus 48%). Th is was the likely cause of the reduction in heterozygote frequency in the sample (20% heterozygotes in sample versus 48% heterozygotes in population). Caucasians th at were homozygous for both Arg/Arg (C/C genotype) and Cys/Cys (T/T genotype) were slightly inflated compared to population estimates, but were mostly comparable. Once again, these subtle differences between

PAGE 105

92 sample and population frequencies are likely due to small sample sizes in both the population samples and the study samples. In regards to the 2-ADR (Arg16Gly) genotypes, frequencies were very sim ilar for both Black/AA and White/Caucasian sample versus population estimates. For the 2-ADR (Gln27Glu) genotype frequencies, Black/AA heterozygotes (Gln/Glu, G/C genot ype) were much more frequent in the sample than in the population (50% versus 30%). In addition, Bl ack/AA homozygotes for the Gln (C) allele were less fr equent (40% in the sample versus 67% in the population). Inversely, White/Caucasian subjects actu ally had a higher frequency of Gln (C) homozygotes (44% versus 35%) and lower fr equency of heterozygotes (Gln/Glu, G/C genotype) (38% versus 49%) in comparison of sample versus population estimates. All of these nuances are likely due to the small number of subjects obt ained in each group. A larger sample size would probably yield more analogous allele and genotype frequencies between sample and population estimates. N onetheless, all genotypes examined were determined to be in Hardy-Weinbe rg equilibrium (data not shown). None of the ADR genotypes studied were positively associated with HTN (Table 416). When accounting for racial differences in these associations by only analyzing White/Caucasians, statistical significance was still not achieved (see Table 4-17). With genotype association analyses, our small samp le size is the most likely culprit of the insignificant findings, as previ ous literature has reported ge notype associations for the 2ADR polymorphisms (McCaffery, Pogue-Ge ile, Ferrell, Petro, & Manuck, 2002; Li, Faulhaber, Rosenthal, Schuster, Jordan, Ti mmermann, and Hoehe, et al., 2001; Cockroft, Gazis, Cross, Wheatley, Dewar, and Hall, et al., 2000; Bray, Kr ushkal, Li, Ferrell, Kardia, Sing, and Turner, et al., 2000).

PAGE 106

93 An additional exploration was made into the association between individual alleles and diagnosis of HTN, since the aforementi oned genotype analyses were insignificant. Allele counts were determined by weighti ng the alleles based on genotype, so that heterozygotes had half of the weight of homo zygotes for each allele. Chi-square analyses were conducted for all subjects with HTN ve rsus NT by allele for each of the three polymorphisms and are presented in Table 4-18. For the 1A-ADR (Arg347Cys) the C (Arg) allele is more frequent in HTN than the T (Cys) allele ( p = 0.03), suggesting it is more important in HTN. This association remained significant for the C allele in this gene when only White/Caucasians were examined ( p = 0.04). This may be the first report of positive association of the Arg allele of the 1A-ADR (Arg347Cys) polymorphism with HTN. Only one other study reported significant association with the 1A-ADR (Arg347Cys) polymorphism and HTN, but reporte d greater association of the Cys allele, which is inconsistent with our findings. (Jiang, Mao, Zhang, Hong, Tang, and Li, et al., 2005). Functional studies of these alleles are limited to one by Shibata and colleagues, where no difference in agonist or antagoni st binding, receptor-mediated calcium signaling, or extent of receptor desensiti zation following agonist exposure was found between either the Arg347 or Cys347 allele of this polymorphism in transfected Chinese hamster ovary cells (Shibata, Hirasa wa, Moriyama,Kawabe, Ogawa, and Tsujimoto,1996). No other ADR alleles were posi tively associated with HTN (see Tables 4-18 and 4-19). The lack of associ ation between the alleles of the 2-ADR polymorphisms at codons 16 and 27 is suppor ted by Castellano and colleagues (2002), Herrmann and colleagues (2002), and To maszewski and colleagues (2002).

PAGE 107

94 Finally, no significant associations were f ound in the analyses of genotypes by gene expression (see Table 4-20). While small sample size may have also affected these analyses, this finding could be explained by th e fact that all thr ee polymorphic variants are located in the coding regi on of the gene, whereby their ne utral variants could impact the function of the resulting protein but not necessarily the level of gene expression. If any of the variants were locat ed in the promoter region of the gene (see Figures 3.1 and 3.2), they may be more likely to affect RNA stability and thus possibly produce a detectable change in gene expression. Limitations Normalization with GAPDH In this study of human arte rial tissue samples in subjects with coronary artery disease and other comorbid diagnoses, GAP DH did not perform well as a normalizer gene even though preliminary an alyses indicated it should have At the very least, these GAPDH data support others’ findings of inc onsistent expression with GAPDH as a housekeeping gene (Barber, Harmer, Colema n, & Clark 2005; Tricarico, Pinzani, Bianchi, Paglierani, Distante, and Pazzagli, et al., 2002), especially in regards to human clinical samples (personal communication with N. Chegini, 2005). As no other acceptable options existed for normalization of these da ta, the GAPDH was used for this purpose. Interpretation of results is cautioned due to this factor. Power For all aims, low power presented a probl em. The power analyses conducted during the planning phase of the study factored in a medium effect size, as reported in the literature (Wang & Brown, 2001). The actual e ffect sizes were variable and produced different power. All of these aims would have required larger sample sizes per group to

PAGE 108

95 facilitate 80% power (see Table 4-21). This l ack of power creates a greater possibility for Type II error for these aims, wherein we fail to reject th e null hypotheses when it is in fact false. This is another limitation of this study. In regards to the sample sizes necessary to obtain this power given the variable effect sizes, some aims remain feasible, while others (examining 2-ADR by SIR) would take a tremen dous amount of effort, time, and expense to recruit 200 subject s as done for this study. Internal Validity This study only attempted to compare s ubjects with and without hypertension in regards to their levels of ge ne expression of two specific ADR genes. As the design of this study was exploratory in nature, we ar e unable to infer the causes of the gene expression differences found. We can only say that differences exist. We can not even be certain of the true significance of the differe nces reported, as the normalization of data may be inaccurate. Therefore, the housekeeping gene, GAPDH, contributes to reducing the internal validity of this study. We are unc ertain even if these differences in GAPDH expression are due to researcher error and/or subject-related genetic or environmental factors. Likewise, the PI responsibly admits to the possibility of pipetting errors in the laboratory that could affect inte rnal validity and overall reliability of triplicate data points for Ct values. Additionally, other factors not studied may contri bute to expression differences in these genes, eroding our c onfidence in internal validity. There are transcription factors related to recruitmen t and processing of relevant RNA polymerase that affect binding of transcri ption factors to the promoter region of the gene. Strachen & Read (1999) further explain that other regulat ory regions (flanking the gene or within introns) can interact wi th other protein factors to ame nd basal levels of transcription.

PAGE 109

96 Construct Validity of the Variable, Normotension The JNC VII criteria for HTN were used to identify hypertensives in the study. This tool considers that ‘absence’ of th e criteria for HTN constitutes NT. While this seems logical, it is important to consider so me key points regarding this variable. First, some false negatives could have existed, wh erein there was poor documentation of past medical history/current diagnoses, BP values, and prescribed medicatio ns that indicated a subject was not diagnosed with HTN, when in fact they were. Second, we are uncertain of which subjects who are currently normote nsive will eventually develop HTN. This potential to develop HTN coul d set them apart both biolog ically and environmentally, decreasing our confidence in these subjects as being true controls. Construct validity regarding this variable is a difficult problem to address in research involving BP and HTN and should be considered when interpreting results. External Validity As briefly discussed previously, this samp le is only moderately representative of the population. These data may only be generaliz ed to other IMA data in which the same normalization issues occurred with similar hous ekeeping gene results, for similar sample characteristics. It would be ir responsible to generalize these da ta to clinical patients with and without HTN. Minimal Sample Template Another limitation of this study was the small amount of starting template. The tissue samples ranged in weight from 10 mg to 34 mg. While no more than 30 mg can be used for each mRNA extraction, multiple extrac tions could have been possible if more tissue were available. Having more tissue would have allowed more opportunity for optimization of reagents and, possibly, of th e housekeeping gene as well. Four subjects

PAGE 110

97 were excluded from gene expression an alyses due to insufficient cDNA and/or “undetermined” readings in th e Real Time RT-PCR process. If more template had been available, additional RT could have been performed to make more cDNA, which would have allowed additional reactions to be redone, allowing inclusi on of these subjects. It is likely that some steps in the tissue prepara tion and RNA extraction processes could have resulted in lower yields of RNA; however, even 10 mg mo re tissue per subject would have provided ‘backup’ sources when necessa ry. It would be prudent to discuss the feasibility of this with the cardiothoracic surgeons for future studies. Confounding Variables Given that -blockers block -ADR receptors (and some -ADR receptors) at the cellular level, it was hypothesi zed that concurrent use of -blockers could impact gene expression and act as a confounder for gene expr ession of the two target genes studied. A simple correlation between the variable ‘type/dose of -blocker’ and each of the target genes’ expression showed no significant correlations ( R = 0.146, p = 0.367 for 1A-ADR; R = 0.095, p = 0.562 for 2-ADR). As most subjects in th is study were on metoprolol, a selective 1-ADR blockade medication, confounding with the 2-ADR receptor system is probably unlikely. On the other hand, one subj ect in this study was taking Labetalol, a 2 antagonist that is non cardioselective with 1-blocking activity. While the presence of this medication could confound the indivi dual subject’s data for both the 1A-and 2-ADR, it would unlikely affect the enti rety of the data. At least one other study supports that pharmacologic 1-ADR blockade does not affect gene expression (Wang & Brown, 2001). They (2001) utilized TaqMan gene expr ession analysis to study gene expression response to 1-ADR blockade in adenylat e cyclase subtypes and in -ADR kinase within human atrium. The authors (2001) found no di fferences in absolute gene expression

PAGE 111

98 between groups receiving pharmacologic 1-ADR blockade, indicating that pharmaceutical 1 blockade did not affect gene expression of the -ADR isoforms. This study was important in examining the environmen tal factors that may affect expression of the ADR subtypes in human tissue. The aut hor is aware of no literature to support -ADR expression or protein diffe rences as a result of blockade. Based on the literature described below, the co-diagnosis of T2DM was also hypothesized to confound gene expression of th e ADR target genes investigated. Recent studies have examined T2DM and vascular ch anges in regards to adrenergic activity. A study conducted on rats with experimentally -induced diabetes demonstrated that vasomotor responses are impaired in diab etes (Kamata, Satoh, Tanaka, & Shigenobu, 1997). However, others have repo rted no significant differences in response to inotropes in human diabetics versus non-diabetics (Din cer, Onay, Ari, Ozcelikay, & Altan, 1998). One group studied humans with T2DM and reported enhanced CVR in this population, and indicated a possible alte ration at the receptor level as the proposed mechanism (Cipolla, Harker, & Porter, 1996). One particular study lin ked HTN and T2DM genetically, implicating the Arg16 allele of the 2-adrenergic receptor gene with an increased risk for HTN in subjects with T2DM (Bengtsson, OrhoMelander, Melander, Lindblad, Ranstam, and Ranstam, et al., 2001) Despite the abundance of association and linkage studies, no expression studies were found that examined the differences in human 1A-and 2ADR gene expression in diabetics versus non-diabetics; therefore, it can only be hypothesized that T2DM could confound AD R-specific gene expression. In this study, T2DM did not significantly correlate with either ta rget gene expression ( R = -0.052, p = 0.745 for 1A-ADR; R = -0.069, p = 0.669 for 2-ADR). The absence of a correlation

PAGE 112

99 between these variables provides some co nfidence that T2DM was not a significant confounder of the gene expression data. Nursing Relevance The approach for this study and its desi gn were borne out of the investigator’s clinical experiences as a nurse and from trai ning received at the Na tional Institutes of Health, National Institute of Nursing Research Summer Genetics Institute. Having experience on a cardiac-telemetry unit made th e investigator aware that normotensive bypass patients do, in fact, exist and could pot entially serve as “normal” controls. In addition, being aware that gene expression re search was at the forefront of genomic scientific inquiry, and having learned of its potential to be highly discriminatory of genomic differences in rat models, the inves tigator saw the potential of this methodology being informative in the clinical phenotype of HTN. This investigator saw great opportunity in approaching the clinical phenotype of HTN in a way that would create a foundation for investigating these particular genes in HTN at the genomic level in the clinical setting. This study wa s approached and designed perh aps differently than a pure bench researcher may have. At some level, this approach may yield more clinically applicable information to those researcher s seeking knowledge about the expression of these receptors in the human phenotype of HTN. In general, gene expression studies of th is nature have the potential to inform practitioners of the impact of gene expression on disease stat es. In particular, studies of this nature are important in that they focus on human disease patterns and biological differences that may contribute to the overall picture of HT N. As this was a pilot study, direct clinical application to nursing or health care can not be extrapolated; however, important knowledge was generated from th is study. This pilot study has shown the

PAGE 113

100 feasibility of collecting human arterial tissu e for the purpose of gene expression research and provided a basis for what is necessary methodologically to develop future gene expression studies in HTN. Ju st as pathophysiological rese arch has informed us of disease mechanisms, gene expression rese arch stands to take us further into understanding those pathways of disease. This type of research can be more informative than just group differences between cases and controls, but of individual patterns whereby environmental factors are accounted fo r in the presence of susceptibility genes. This methodology examines how the genes are expressed in a given environment. This gene by environment interaction is the pivotal crux that makes studying diseases in this way more amenable to nursing questions. The fo cus is not just on the genetic component, but on how the genes are expressed in the environment (a milieu created by the individual’s life choices) a nd how certain people are more susceptible to disease than others with these genetic and environmenta l factors. Nurse researchers can and should consider this area of research to look at ways to benef it patients holistically through scientific research that examines both ge netic and environmenta l factors and their interplay in health and disease. Recent a dvances in genomic research are leading us toward more individualized medicine, es pecially with microarray technology and the optimization of blood-based gene expression tools. Not only will practitioners benefit from knowledge and tools gained from this type of research for screening and treatment, patients stand to benefit from tailored health car e that results from this type of research. Practiceand Care-Related Relevance As genetics becomes an integral part of health care and treatment considerations, nurses will be increasingly required to have more sophisticated understanding of these genetic concepts and current research. For d ecades, nurses have taken on roles in genetics

PAGE 114

101 as patient educators, advocates, and counselors (especially with regard to prenatal testing and genetics-related reproductive issues). Now, these previous roles are becoming more sophisticated and new roles are being develope d in which nurses are needed (and desired) to help translate genetics and genomics research into practice. The fu ture of health care will require a contingent of health care provide rs to deliver the information accurately for appropriate use in decision-ma king by patients. The optimal source for these provisions is genetic counselors (GCs); however, the lack of and competitive nature of GC programs, and issues with reimbursement of GCs for th eir services has resulted in a small GC workforce. It is becoming increasingly apparent within the nursing profession (and among GCs) that these current GC professional issues are putting nurse s in the position to assume GC-like roles; to pr ovide patients with knowledge interpretation, counseling, and assistance in decision-making in regards to genetic inform ation, screening, testing, and participation in research. Furthermore, as the pace of biot echnology exponentially accelerates, so does the incorporation of bi otechnology into practice and care. Nurses must be active in continuing education regard ing genetics and genomics and be evermore aware of the research in these fields that wi ll likely affect practice in the near future. With specific regard to research, nursing research in genetics through interdisciplinary teams is a charge of the NIH Roadmap (Huerta, Farber, Wilder, Kleinman, Grady, and Schwartz et al., 2005) and is supporte d for individual research endeavors by the National Institutes of Health (Jenkins, Grady, & Collins, 2005), National Institute of Nursing Research (Grady & Collins, 2003; Sigmon, Grady, & Amende, 1997), and the International Soci ety of Nurses in Genetics (ISONG; organization goals, http://www.isong.org/about/index.cfm 2005). Loescher and Merkle

PAGE 115

102 (2005) published a review of recently-funded NI NR studies using genetic and/or genomic methods where nurses were primary investigator s. At least thirteen nurse-driven studies were cited in which advanced biological, ge netic and/or genomic methods were utilized, from linkage analyses to microarray, reinfo rcing the notion that wh ere there is funding occurring, there is desire, need, and support for the research. Lashley, a geneticist and nursing leader in genetics, lists ‘disease markers’ and ‘disease mechanisms’ as two of the top research directions fo r nursing research in genetic s (2001). Dr. Janet Williams, another nurse leader in genetics, developed the first NIH-funded pos tdoctoral program in Nursing Genetics at the University of Iowa, wher e scientific research that uses biological, molecular, and genetic methods and technol ogy is fostered and encouraged among nurse scientists in the program. In 2000, Hill pub lished her research on comprehensive HTN care in young urban black men, a study that integrated genetic science, clinical interventions, and patient outcomes. Feet ham (2000) summarized Hill’s recommendation that “nursing research must address resear ch questions and hypotheses that integrate genetic science and issues if nur sing practice is to be based on science that is current, relevant, and leveraged effectively” (p. 258). Nurse-Directed Research and Qualitative Findings Historically, nurses have recognized the impor tance of the patient as a whole. This philosophy has transcended prac tice and has permeated nurse-directed research. During the completion of this study, the PI had a number of experiences with potential and enrolled subjects that reflected the notion that it is both prudent and important to maintain a philosophy of holism when c onducting clinically-bas ed research. These next paragraphs present some of the qualitative findings r ealized during the comp letion of the study.

PAGE 116

103 This study involved recruitment and cons ent of subjects who are preparing to undergo a moderateto high-risk surgical cardiac intervention. This brought about some valuable interactions between the researcher and potential/enrolled subjects. Some of these interactions should be highlighted in order to gain the most out of the research experience. For one, while some participants expressed apprehension about enrolling in genetics studies, particularly among Black/AAs, a majority of the participants in this study reported a strong desire to contribute to research for HTN and genetics. Subject concerns regarding the study re volved around protection of data from insurers, protection of DNA and tissue, and having any unwanted 3rd party from obtaining the information. A majority of potential subj ects and participants expre ssed feeling stressed about their upcoming bypass surgery. These expressions often occurred during the recruitment and consenting process. The researcher ma intained awareness of this during these procedures and often decided to postpone di scussion of the study w ith potential subjects in order to talk about their fears and con cerns. Many concerns stemmed from fear of death or complications, being out of work for surgery and re covery, feeling like a burden to family members, and concerns with insurance and payment of hospital bills. Once patient education was provided in regards to the surgery, the researcher felt more comfortable proposing participation in the study. The researcher made it clear that potential subjects were not, in any way obligated to participate given the ‘counseling’. To emphasize this, the researcher provided a dditional time for their review of the study materials prior to consenting subjects. Case Study One particular participant sta nds out in regards to appreh ension in participating in the study. He was a self-identified Black/AA male, 44 years old. He was concerned with

PAGE 117

104 the aims involving race and voiced mistrust of research. A conversation regarding the studies’ background and significance of HTN with regards to race occu rred. After further discussion, he explained that his sensitivity to the issue stemmed from feeling like a victim of racism in the hospital. He was also feeling a great deal of stress from financial insecurity, a poor support system, fear of poor recovery, and lack of resources after discharge. In addition, he felt that he would continue to receive le ss than standard-of-care because of his race and financial status. The researcher initiated an in-depth discussion about his feelings and concer ns and provided him with the name and contact information for the hospital’s patient advocate/liaison. Th e researcher felt that continuing with the recruitment of this subject for the study would be unethical; however, the subject expressed interest. The ethical issues with obligati on of reciprocation were explained and the researcher expressed not wanting to a dd any further undue stress in his life. The subject requested an opport unity to review the paperw ork until the next day. The following day, he expressed sincer e desire to participate, stat ing [paraphrasing] that he wasn’t sure if he would ever be able to c ontribute to science and that if he could, he would want it to be something that would possibly contribute to people in his population with similar health problems. This patient was enrolled in the study. Summary These interactions provided the researcher with a great deal of insight into the qualitative issues surrounding subject particip ation in research a nd the stress invoked by patients facing bypass surgery. Thes e interactions may not be en tirely unique to nursing, but are important to nursing rese arch and in the developing of future research involving genetics and bypass patients. The case study particularly highlight s the importance of nurses’ being well-versed on issues of ge netics and race in research and practice.

PAGE 118

105 Recommendations for Future Research First and foremost, it would be most prude nt to test and optimize more than one housekeeping gene when relying on an e ndogenous control for normalization. When financially feasible, a housekeeping gene “pla te” can be purchased (such as the one sold by Applied Biosystems), where a panel of common housekeeping genes can be tested with the sample of interest in order to select the one that meets all of the criteria for a proper endogenous control. Second, attempting to obtain larger tissu e samples would likely lead to greater success in optimizing the reactions, perhaps ev en for multi-plexing (which reduces costs and saves template). Other non-bypass arterial tissue sources should also be considered, as not every bypass patient requ ires a bypass using the IMA. Future studies that examine these genes and their differential expression in HTN could be much more informative if actual BP values could be obtained. While there are numerous issues with the co llection and use of BP (inc luding, but not limited to, measurement issues, interindivi dual variability in values, ci rcadian varia tion, the Whitecoat phenomenon, and most importantly, the confounding of BP values by concomitant use of antihypertensive medications), having a continuous variable by which to compare the gene expression, rather than a simple dichotomous variable, could lead to more informative and inferential statistical anal yses (where the aforementioned limitations could be allayed). For example, one could po ssibly predict gene expression levels based on BP levels with regression analyses. Future studies that examine the im pact of self-ide ntified race should very carefully consider the current debate in this area. If a greater (and more re presentative) sample were obtained, there would be sufficient power to analyze other health-related variables

PAGE 119

106 linked to race such as indicators of socio economic status (wealth, income, poverty level, access to health care), educa tion, and indices of racism. Mo reover, the use of ancestry informative markers to account for the biolog ical variance in geographic ancestry could improve further inquiries and provide a more quantitative measure of ancestral origins, rather than with the indirect self-report of race as a proxy. Qualitative inquiries that focused on a) th e nature of participation in research involving race and/or genetics; and b) perceived stress by patients preparing for bypass surgery could be important. This information c ould ameliorate studies of this nature in the future. Finally, future studies should examine more HTN candidate genes and additional environmental factors so that more complex que stions can be answered that may lead to better prevention, screening, and treatment of HTN through the use of biomarkers. Eventually such information can inform pr actitioners of individualized biophysiological patterns and treatment. The PI recognizes that the feasibility of using human tissue samples for future clinical sc reening would be an impractic al endeavor. The use of bloodbased gene expression methods is now bei ng refined and will likely be adapted for clinical screening tools. The us e of tissue in research, however, still stands to provide the best direct measurement of ge ne expression and should continue to be utilized in diseasebased, gene expression research. Conclusions A great deal of literature has supported th e potential for gene expression to unlock many of the mysteries of complex diseases. Th is is likely due to the nature of gene expression being affected by both genetic and environmenta l factors. While animal

PAGE 120

107 models are invaluable in this line of res earch, gene expression studies that extend to human models of disease will serve us greatly in truly understanding human disease. To the author’s knowledge, this study is the first to explore 1Aand 2-ADR gene expression differences in hu man arterial tissue between normotensive and hypertensive adults. This study demonstrated that colle cting, preserving, and pr eparing human tissue for the purpose of performing Real Time RT-PCR for relative gene expression was possible and moderately feasible, given some training and adequate funding. This study outlines some of the technical challenges in performing these types of analyses but shows the potential to optimize this pr ocess for more internally valid results. This study also provided estimates of effect sizes for use in future research. In addition, this study obtained valuable qualitative information re garding subject recr uitment, consent, perceived stress, and concerns regarding partic ipation in research i nvolving genetics and race. All statistical findings from this study should be cautiously interpreted. To summarize: 1) Significant fold-differences in 1Aand 2-ADR gene expression were found between people with and without HTN; 2) These differences remained significant when examining only self-identified White/C aucasian hypertensive versus normotensive subjects; 3) Significant di fferences were not found between self-identified White/Caucasians and self-identified Black /AAs with HTN; 4) The need for postoperative positive inotrope medication did not significantly correlate to either 1Aor 2ADR gene expression; 5) ADR genotypes were not significantly associated with the diagnosis of HTN; 6) The C (Arg) allele of the 1AADR 347 polymorphism was significantly associated with HTN; 7) There was no signi ficant relationship between ADR genotypes and ADR gene expression; 8) Unsuccessful normalization with GAPDH

PAGE 121

108 in this study supports other fi ndings and contributes largely to the lack of confidence in results. Replication of this research with careful consideration of the points discussed would lead to important information about the nature of clinical differences in ADR gene expression in the disease process of HTN.

PAGE 122

109 APPENDIX A SUBJECT ENROLLMENT/DEMOGRAPHIC FORM Name ________________________________ Date _______________ Phone ________________________________ A. Age : _____ Date of Birth:____/____/____ B. Sex : 1. Male 2. Female C. ETHNICITY Do you consider yourself to be Hispanic or Latino? (See definition below.) Select one or more. Hispanic or Latino (A person of Mexican, Puerto Rican, Cuban, South or Central American, or other Spanish cu lture or origin, regardless of race.) Not Hispanic or Latino D. RACE What race do you consider yourself to be? Select one or more of the following. American Indian or Alaska Native. A person having origins in any of the original peoples of North, Central, or South America, and who maintains tribal affiliation or community attachment. Asian. A person having origins in any of the original peoples of the Far East, Southeast Asia, or the Indian subcontinent, including, for example, Cambodia, China, India, Japan, Korea, Malaysia, Pakistan, the Philippine Islands, Thailand, and Vietnam. (Note: Individuals from the Philippine Islands have been recorded as Pacific Islanders in previous data collection strategies.) Black or African American. A person having origins in any of the black racial groups of Africa. Terms such as “Haitian” or “Negro” can be used in addition to “Black” or African American.” Native Hawaiian or Other Pacific Islander. A person having origins in any of the original peoples of Hawaii, Gu am, Samoa, or other Pacific Islands. White. A person having origins in any of the original peoples of Europe, the Middle East, or North Africa. Check here if you do not wish to provide some or all of the above information.

PAGE 123

110 E. Your highest level of completed education is: ____Less than 9th grade in high school ____Less than 12th grade in high school ____High school diploma ____GED ____Some college or technical training ____Associate degree ____Bachelor’s degree ____Master’s degree ____Doctoral degree (M.D., Ph.D., or J.D.) F. Body Height: ____________ ft Weight: ____________ lbs G. Did you smoke? YES NO If yes, do you smoke now? YES NO How long have you (or did you) smoke? _________ years H. Do you drink alcohol? YES NO If yes, how often do you drink alcohol? _______________ How many drinks do you usua lly have in one week? _________________ I. Do you exercise? YES NO What types of activities do you do when you exercise?_____________________ How many times during the week do you usually exercise? _________________ J. Marital status: (please circle) Married Single Divorced Separated Other K. Please circle the size of your family un it (number of people in household) and circle what your approximate yearly household income is (this information is confidential). Size of Family Unit < 200% 200-40 0% > 400% 1 $0 – 17,999 $18,000 – 34,999 $35,000 + 2 $0 – 23,999 $24,000 – 46,999 $47,000 + 3 $0 – 29,999 $30,000 – 58,999 $59,000 + 4 $0 – 35,999 $36,000 – 70,999 $71,000 + 5 $0 – 41,999 $42,000 – 82,999 $83,000 + 6 $0 – 47,999 $48,000 – 94,999 $95,000 + 7 $0 – 53,999 $54,000 – 106,999 $107,000 + 8 $0 – 59,999 $60,000 – 118,999 $119, 000 +

PAGE 124

111 Past Medical History ________________________________________________________________________ ________________________________________________________________________ ________________________________________________________________________ ________________________________________________________________________ Medications Prescribed: ________________________________________________________________________ ________________________________________________________________________ ________________________________________________________________________ Over the counter: ________________________________________________________________________ ________________________________________________________________________ ________________________________________________________________________ Herbal Remedies ________________________________________________________________________ ________________________________________________________________________ N. Would you like me to let your primary care provider know that you are participating in this study? If so, please provi de your PCP’s name and address. PCP Name ________________________________ Address ________________________________ ________________________________ Phone ________________________________

PAGE 125

112 APPENDIX B UF IRB-01 INFORMED CONSENT FORM IRB# 33-2004 Informed Consent to Pa rticipate in Research and Authorization for Collection, Use, and Disclosure of Protected Health Information You are being asked to take part in a res earch study. This form provides you with information about the study and seeks your authorization for the collection, use and disclosure of your protected health informa tion necessary for the study. The Principal Investigator (the person in charge of this research) or a representative of the Principal Investigator will also describe this study to you an d answer all of your questions. Your participation is entirely voluntar y. Before you decide whether or not to take part, read the information below and ask questions about a nything you do not unders tand. If you choose not to participate in this study you will not be penalized or lose any benefits to which you would otherwise be entitled. 1. Name of Participant ("Study Subject") ____________________________________________________________________ 2. Title of Research Study Alpha 1Aand Beta 2-Adrenoceptor Gene E xpression Differences In Hypertensive and Normotensive Persons By Race Subtitle 1: Race, HTN, and Vascular Adrenoceptor Gene Expression

PAGE 126

113 3. Principal Investigator and Telephone Number(s) Jennifer R. Dungan, MSN, ARNP PI Office: (352) 273-6512 PI Cell: (352) 256-7487 Co PI: Ann Horgas, RN, PhD Co-PI Office: (352) 273-6318 4. Source of Funding or Other Material Support Material Support: University of Florida Funding: American Heart A ssociation, # 0415124B (partial) NIH/National Institutes of Nursing Research, # 1 F31 NR009148-01 Sigma Theta Tau, Alpha Theta Chapter (no grant #) 5. What is the purpose of this research study? The purpose of this research study is to look at differences in the gene expression of two genes (the alpha 1Aand beta 2-adrene rgic receptor genes) between persons with and without high blood pressure. Gene e xpression refers to the way your genes work. This is done by looking at the me ssenger ribonucleic acid (mRNA) levels found in your tissue sample. These levels tell us if the gene is functioning too much or not enough. By collecti ng a tissue sample from you a nd others in the study, we may be able to determine if function of certain genes is important in high blood pressure. 6. What will be done if you ta ke part in this research study? If you decide to take part in this resear ch study, three things will happen. First, we will collect a teaspoon of blood from a vein in one of your arms and store it in a locked freezer. This blood will be used to genotype—or characterize—certain genes that are important in cardi ovascular disease and genes that may be important in statistically determining similarity in gr oups within the population. Second, the PI will be given a piece of tissue from your ches t that is normally thrown away during your coronary artery bypass surgery. This piece of tissue is leftover from surgery, and will come from an artery near your hear t. This piece of tissue will be analyzed for gene expression of two genes. Once the tissue is analyzed, it will be thrown away. Third, the PI will re view your medical chart to s ee if you received a certain type of heart medication. If you did receive the type of medicati on, the PI will collect information about your response to the medication. NOTE: You will not purposefully be given this drug as a result of being in the study. Small amounts of your DNA may be sent to outside laboratories for this and future analyses. Any future analyses of your DNA would be for research purposes only. Tests for clinical markers that would affect your clinical care (for example, testing for a specific disease) will not be performed w ithout your signed informed consent.

PAGE 127

114 The choice to let Jennifer Dungan draw a nd keep your blood for doing research is entirely up to you. No matter what you decide to do, it will not a ffect your care. If you decide that your blood can be kept fo r research but you later change your mind, tell Jennifer Dungan who will remove and de stroy any of your blood she still has. Otherwise, the samples may be kept up to 15 years, until they are used up, or until Jennifer Dungan decides to destroy them. Please review the following four statements carefully. If you disagree with any of these statements, you should not participate in the study. 1. I understand that my samples will be stor ed for up to 15 years, coded to protect my identity, and that my identity will not be disclosed to anyone without my permission, except when required by law. 2. I understand that some excess blood may be kept by Jennifer Dungan for use in future research to learn about, prev ent, treat, or cure hypertension and cardiovascular disorders. 3. I understand that my blood (but not tissue) may be used for research to answer other medical questions th at are not necessarily re lated to hypertension and cardiovascular disorders. 4. I understand that Jennifer D ungan (or someone she chooses) can contact me in the future to ask me to take part in more research. If you have any questions now or at any time during the study, you may contact the Principal Investigator lis ted in #3 of this form. 7. If you choose to participate in this study, how long will yo u be expected to participate in the research? The active participation in the study to collect blood a nd tissue is approximately 1 hour. Passive participation, which involve s reviewing your chart after surgery, is expected to take one month or less. Fo llowing chart review, y our participation is complete in the study. 8. How many people are expected to participate in this research? For this study, we expect to have between 6072 participants. We ai m to have at least 30 people with high blood pressure a nd 30 people with normal blood pressure enrolled in the study.

PAGE 128

115 9. What are the possible discomforts and risks? The only potential risk is that associated with drawing blood. The risks of drawing blood from a vein include di scomfort at the site of puncture; possible bruising and swelling around the puncture si te; rarely an infection; and, uncommonly, faintness from the procedure. This study may include risks th at are unknown at this time. Participation in more than one research st udy or project may furthe r increase the risks to you. Please inform the Principal Investigat or (listed in #3 of this consent form) or the person reviewing this consent with you before enrolling in this or any other research study or project. Throughout the study, the researchers will notify you of new in formation that may become available and might affect your decision to remain in the study. If you wish to discuss the information a bove or any discomforts you may experience, you may ask questions now or call the Principa l Investigator or contact person listed on the front page of this form. 10a. What are the possible benefits to you? There are no possible benefits to you if you participate in the study. You will not be compensated for the study. You will also no t be given any information regarding the results of this study. 10b. What are the possible benefits to others? Even though the research that is done on your tissue will not be used to help you, it may be helpful to others with high blood pr essure in the future. Because of this study, it may be possible in the future to de termine the importance of gene expression in high blood pressure. 11. If you choose to take pa rt in this research study, will it cost you anything? There will be no cost to your for any specimens that are collected and stored, or any other materials used in th is research project. Costs for routine medical care procedures that are not being done only for the study will be charged to you or your insurance. These costs may not be charged if you are a veteran and you are being treated at the Nort h Florida/South Georgi a Veterans Health System (NF/SG VHS).

PAGE 129

116 12. Will you receive compensation for ta king part in this research study? No, you will not receive compensati on for taking part in this study. 13. What if you are injure d because of the study? If you experience an injury that is dire ctly caused by this study, only professional consultative care that you receive at the Univ ersity of Florida Health Science Center will be provided without charge. However, hospital expenses will have to be paid by you or your insurance provider No other co mpensation is offered. Please contact the Principal Investigator listed in Item 3 of this form if you e xperience an injury or have any questions about any discomforts that yo u experience while participating in this study. 14. What other options or trea tments are available if you do not want to be in this study? Participation in this study is entirely voluntar y. You are free to refuse to be in this study. If you refuse to participate, it will not effect the treatment to which you are otherwise entitled. 15a. Can you withdraw from this research study? You are free to withdraw your consent and to st op participating in this research study at any time. If you do withdraw your consent, there will be no penalty, and you will not lose any benefits you are entitled to. If you decide to withdraw y our consent to participate in this research study for any reason, you should contact Jennife r R. Dungan at (352) 273-6512. If you have any questions regarding your ri ghts as a research su bject, you may phone the Institutional Review Board (IR B) office at (352) 846-1494. 15b. If you withdraw, can information abou t you still be used and/or collected? If you decide to withdraw from the study, da ta already collected may be used in the completion of the study’s analysis, but no furthe r data will be collected and/or used. 15c. Can the Principal Investigator wit hdraw you from this research study? You may be withdrawn from the study w ithout your consent for the following reasons: It is discovered that you no longer meet the inclusion/exclusion criteria.

PAGE 130

117 16. If you agree to participate in this res earch study, the Princi pal Investigator will create, collect, and use private information about you and your health. Once this information is collected, how w ill it be kept secret (confide ntial) in order to protect your privacy? Information collected about y ou and your health (called prot ected health information), will be stored in locked fili ng cabinets or in comp uters with security passwords. Only certain people have the legal right to revi ew these research re cords, and they will protect the secrecy (confidentiality) of these re cords as much as th e law allows. These people include the researchers fo r this study, certain University of Florida officials, the hospital or clinic (if any) involved in this research, and the Institutional Review Board (IRB; an IRB is a group of people who are re sponsible for looking af ter the rights and welfare of people taking part in research). Otherwise your research records will not be released without your permission unl ess required by law or a court order. Results of genetic testing will not be included in your medi cal record and will be kept in a secure electronic database that is availa ble only to the study investigators. Your records will not be released to any person or agency unles s you provide written consent to Jennifer Dungan requesting the release of the results of the study. Jennifer Dungan will be responsible for making sure that any stored blood sample is protected in the specimen bank and that your medical information is kept confidential. Your sample will not be stored with your name or other identifying information, but instead will be given a code number to protect your identity These samples and this code number will only be gi ven to researchers whose re search is approved by an Institutional Review Board (IRB). (An IRB is a group of people who are responsible for looking after the rights and welfare of people taking part in research.) The researchers will not be told who you are or given any id entifying information about you. Because the nature and va lue of any future research cannot be known at this time, any results obtained from using your tissue will not be given to you or your doctor. If the results of this research are publishe d or presented at scientific meetings, your identity will not be disclosed. If you participate in this research study, th e researchers will collect use, and share your protected health information with others. Items 17 to 26 below describe how this information will be collect ed, used, and shared.

PAGE 131

118 17. If you agree to participate in this research study, what protected health information about you may be collect ed, used and shared with others? To determine your eligibility for the study and as part of your participation in the study, your protected health informa tion may be collected, used, an d shared with others to determine if you can participate in the study, and then as part of your participation in the study. This information can be gathered from you or your past, current or future health records, from procedur es such as physical examina tions, x-rays, blood or urine tests or from other procedur es or tests. This info rmation will be created by participating in study procedures, or from your study visits and telephone calls. More specifically, the following info rmation may be collected, used, and shared with others: Complete past medical history, current me dications, race, age, and records of previous blood pressures obtained duri ng physical examinations to determine eligibility criteria. Also, as part of the demographic informati on obtained from you to be used in data analysis, the following information will be obtained: name, age, phone number, race, ethnicity, current medications records of previous blood pressure, income, education, job, marriage status, number of people in your household, first degree relative information and information about your sm oking, drinking, and exer cise habits. If you agree to be in this research study, it is possible that some of the information collected might be copied into a "limited da ta set" to be used for other research purposes. If so, the limited data set ma y only include information that does not directly identify you. For ex ample, the limited data set cannot include your name, address, telephone number, social secu rity number, or any other photographs, numbers, codes, or so forth that link you to th e information in the li mited data set. If used, limited data sets have legal agr eements to protect your identity and confidentiality and privacy. 18. For what study-related purposes will your protected health information be collected, used, and sh ared with others? Your protected health inform ation may be collected, used, and shared with others to make sure you can participate in the re search, through your participation in the research, and to evaluate the results of th e research study. More specifically, your protected health information may be collect ed, used, and shared with others for the following study-related purpose(s): 1) to determine the impact of gene expr ession on high blood pre ssure and response to certain medications 19. Who will be allowed to coll ect, use, and share your protected health information? Your protected health inform ation may be collected, used, and shared with others by:

PAGE 132

119 the study Principal Investigator Je nnifer R. Dungan and her staff other professionals at the University of Florida or Shands Ho spital that provide study-related treatment or procedures the University of Florida Institutional Review Board 20. Once collected or used, who may your pr otected health information be shared with? Your protected health inform ation may be shared with: the study sponsors: The American Heart Association, The National Institutes of Nursing Research (of the National Ins titutes of Health) an d Sigma Theta Tau, Alpha Theta Chapter United States and foreign governmental agencies who are responsible for overseeing research, such as the Food and Drug Administration, the Department of Health and Human Servi ces, and the Office of Human Research Protections Government agencies wh o are responsible for ove rseeing public health concerns such as the Centers for Diseas e Control and Federal, State and local health departments Malcom Randall VA Medi cal center (Gainesville) 21. If you agree to participate in this resea rch, how long will you r protected health information be used and shared with others? Your protected health information will be used until the end of th e study. Thereafter, your protected health informatio n will be stripped from the research database and the database containing the unidentifiable info rmation will be kept under the protection of the principal investigator forever. 22. Why are you being asked to allow the colle ction, use and sharing of your protected health information? Under a new Federal Law, researchers cannot collect, use, or share with others any of your protected health informa tion for research unless you a llow them to by signing this consent and authorization. 23. Are you required to sign th is consent and authorizatio n and allow the researchers to collect, use and share with others your protected health information? No, and your refusal to sign will not affect your treatment payment, enrollment, or eligibility for any benefits out side this research study. However, you cannot participate in this research unless you allow the coll ection, use and sharing of your protected health information by signing this consent/authorization.

PAGE 133

120 24. Can you review or co py your protected health inform ation that has been collected, used or shared with others under this authorization? You have the right to review and copy your protected health information. However, you will not be allowed to do so until after the study is finished. 25. Is there a risk that your protected health information could be given to others beyond your authorization? Yes. There is a risk that information recei ved by authorized persons could be given to others beyond your authorizati on and not covered by the law. 26. Can you revoke (cancel) your authorizatio n for collection, use and sharing with others of your protected health information? Yes. You can revoke your authorization at any time before, during, or after your participation in the research. If you revoke, no new information will be collected about you. However, information th at was already collected ma y still be used and shared with others if the researchers have relied on it to complete and protect the validity of the research. You can revoke your authoriza tion by giving a written request with your signature on it to the Principal Investigator. 27. How will the researcher(s) benefi t from your being in this study? In general, presenting research results helps the career of a scien tist. Therefore, the Principal Investigator may benefit if the results of this study are presented at scientific meetings or in scientific journals. 28. Signatures As a representative of this study, I have ex plained to the participant the purpose, the procedures, the possible benefits, and the risks of this research study; the alternatives to being in the study; and how the participan t’s protected health information will be collected, used, and shared with others: _________________ _________________ ________________ _________________ Signature of Person Obtaining Consent and Authorization Date You have been informed about this study’s purpose, procedures, possible benefits, and risks; the alternatives to being in the st udy; and how your protected health information will be collected, used and shared with others. You have rece ived a copy of this Form. You have been given the opportunity to ask questions before you sign, and you have been told that you can ask other questions at any time.

PAGE 134

121 You voluntarily agree to participate in this study. You hereby authorize the collection, use and sharing of your protected health in formation as described in sections 17-26 above. By signing this fo rm, you are not waiving a ny of your legal rights. _________________ _________________ ________________ _________________ Signature of Person Consenting and Authorizing Date

PAGE 135

122 APPENDIX C VA SCI INFORMED CONSENT FORM IRB # 33-2004 Informed Consent to Pa rticipate in Research and Authorization for Collection, Use, and Disclosure of Protected Health Information You are being asked to take part in a res earch study. This form provides you with information about the study and seeks your auth orization for the collection, use and sharing of your protected health information necessary for the study. The Principal Investigator (the person in charge of this research) or a representative of the Prin cipal Investigator will also describe this study to you and answer all of your questions. Your participation is entirely voluntary. Before you decide whether or not to take part, read the information below and ask questions about anything you do not understand. If you choose not to participate in this study you will not be penali zed or lose any benef its to which you would otherwise be entitled. 1. Name of Participant ("Study Subject") _____________________________________________________________________ 2. Title of Research Study Alpha 1Aand Beta 2-Adrenoceptor Gene Expression Differences In Hypertensive and Normotensive Persons By Race Subtitle 1: Race, HTN, and Vascular Adrenoceptor Gene Expression 3. Principal Investigator and Telephone Number(s)

PAGE 136

123 Jennifer R. Dungan, MSN, ARNP PI Office: (352) 273-6512 PI Cell: (352) 256-7487 Co PI: Ann Horgas, RN, PhD (352) 273-6318 Sub-PI (VA PI): Philip J. Hess, M.D. (352) 413-0143 (pager) 4. Source of Funding or Other Material Support Material Support: University of Florida Funding: American Heart Asso ciation, # 0415124B (partial) NIH/National Institutes of Nursing Research, # 1 F31 NR009148-01 Sigma Theta Tau, Alpha Theta Chapter (no grant #) 5. What is the purpose of this research study? The purpose of this research study is to look at differences in the gene expression of two genes (the alpha 1Aand beta 2-adrene rgic receptor genes) between persons with and without high blood pressure. Gene expression refers to the way your genes work. This is done by looking at the messenger ribonucleic acid (mRNA) levels found in your tissue sample. These levels tell us if the gene is functioning too much or not enough. By collecting a tissue sample from you and others in the study, we may be able to determine if functi on of certain genes is importa nt in high blood pressure. 6. What will be done if you ta ke part in this research study? If you decide to take part in this research study, three things will happen. First, we will collect a teaspoon of blood from a vein in one of your arms and store it in a locked freezer. This blood will be used to genotype—or characterize—certain genes that are important in cardi ovascular disease and genes that may be important in statistically determining similarity in gr oups within the population. Second, the PI will be given a piece of tissue from your ches t that is normally thrown away during your coronary artery bypass surgery. This piece of tissue is leftover from surgery, and will come from an artery near your hear t. This piece of tissue will be analyzed for gene expression of two genes. Once the tissue is analyzed, it will be thrown away. Third, the PI will re view your medical chart to s ee if you receiv ed a certain type of heart medication. If you did receive the type of medicati on, the PI will collect information about your response to the medication. NOTE: You will not purposefully be given this drug as a result of being in the study. The data we analyze and use from this study (not the tissue) may be used in future studies that address similar questions. You may choose not to allo w us to use your data in future studies Small amounts of your DNA may be sent to outside laboratories for this and future analyses. Any future analyses of your DNA would be for research purposes only. Tests for clinical markers that would affect your clinical care (for example, testing for a specific disease) will not be performed w ithout your signed informed consent.

PAGE 137

124 The choice to let Jennifer Dungan draw a nd keep your blood for doing research is entirely up to you. No matter what you decide to do, it will not a ffect your care. If you decide that your blood can be kept fo r research but you later change your mind, tell Jennifer Dungan who will remove and de stroy any of your blood she still has. Otherwise, the samples may be kept up to 15 years, until they are used up, or until Jennifer Dungan decides to destroy them. Please review the following four statements carefully. If you disagree with any of these statements, you should not participate in the study. 5. I understand that my samples will be stor ed for up to 15 years, coded to protect my identity, and that my identity will not be disclosed to anyone without my permission, except when required by law. 6. I understand that some excess blood may be kept by Jennifer Dungan for use in future research to learn about, prev ent, treat, or cure hypertension and cardiovascular disorders. 7. I understand that my blood (but not tissue) may be used for research to answer other medical questions th at are not necessarily re lated to hypertension and cardiovascular disorders. 8. I understand that Jennifer D ungan (or someone she chooses) can contact me in the future to ask me to take part in more research. If you have any questions now or at any time during the study, you may contact the Principal Investigator lis ted in #3 of this form. 7. If you choose to participate in this study, how long will yo u be expected to participate in the research? The active participation in the study to collect blood a nd tissue is approximately 1 hour. Passive participation, which involve s reviewing your chart after surgery, is expected to take one month or less. Fo llowing chart review, y our participation is complete in the study. 8. How many people are expected to participate in this research? For this study, we expect to have between 6072 participants. We ai m to have at least 30 people with high blood pressure a nd 30 people with normal blood pressure enrolled in the study.

PAGE 138

125 9. What are the possible discomforts and risks? The only potential risk is that associated with drawing blood. The risks of drawing blood from a vein include di scomfort at the site of puncture; possible bruising and swelling around the puncture si te; rarely an infection; and, uncommonly, faintness from the procedure. This study may include risks th at are unknown at this time. Participation in more than one research st udy or project may furthe r increase the risks to you. Please inform the Principal Investigat or (listed in #3 of this consent form) or the person reviewing this consent with you before enrolling in this or any other research study or project. Throughout the study, the researchers will notify you of new in formation that may become available and might affect your decision to remain in the study. If you wish to discuss the information a bove or any discomforts you may experience, you may ask questions now or call the Principa l Investigator or contact person listed on the front page of this form. 10a. What are the possible benefits to you? There are no possible benefits to you if you participate in the study. You will not be compensated for the study. You will also no t be given any information regarding the results of this study. 10b. What are the possible benefits to others? Even though the research that is done on your tissue will not be used to help you, it may be helpful to others with high blood pr essure in the future. Because of this study, it may be possible in the future to de termine the importance of gene expression in high blood pressure. 11. If you choose to take pa rt in this research study, will it cost you anything? There will be no cost to your for any specimens that are collected and stored, or any other materials used in th is research project. Costs for routine medical care procedures that are not being done only for the study will be charged to you or your insurance. These costs may not be charged if you are a veteran and you are being treated at the Nort h Florida/South Georgi a Veterans Health System (NF/SG VHS). 12. Will you receive compensation for ta king part in this research study? No, you will not receive compensati on for taking part in this study.

PAGE 139

126 13. What if you are injure d because of the study? If you experience an injury that is dire ctly caused by this study, only professional consultative care that you receive at the Univ ersity of Florida Health Science Center will be provided without charge. However, hospital expenses will have to be paid by you or your insurance provider No other co mpensation is offered. Please contact the Principal Investigator listed in Item 3 of this form if you e xperience an injury or have any questions about any discomforts that yo u experience while participating in this study. You will not have to pay hospital expenses if you are being treated at the North Florida/South Georgia Veterans Health System (NF/SG VHS) and experience any physical injury during participation in a Veterans Health System-approved study. 14. What other options or trea tments are available if you do not want to be in this study? Participation in this study is entirely voluntar y. You are free to refuse to be in this study. If you refuse to participate, it will not effect the treatment to which you are otherwise entitled. 15a. Can you withdraw from this research study? You are free to withdraw your consent and to st op participating in this research study at any time. If you do withdraw your consent, there will be no penalty, and you will not lose any benefits you are entitled to. If you decide to withdraw y our consent to participate in this research study for any reason, you should contact Jennife r R. Dungan at (352) 273-6512. If you have any questions regarding your ri ghts as a research su bject, you may phone the Institutional Review Board (IR B) office at (352) 846-1494. 15b. If you withdraw, can information abou t you still be used and/or collected? If you decide to withdraw from the study, da ta already collected may be used in the completion of the study’s analysis, but no furthe r data will be collected and/or used. 15c. Can the Principal Investigator wit hdraw you from this research study? You may be withdrawn from the study w ithout your consent for the following reasons: It is discovered that you no longer meet the inclusion/exclusion criteria.

PAGE 140

127 16. If you agree to participate in this research study, the Principal Investigator will create, collect, and use private informatio n about you and your health. Once this information is collected, how w ill it be kept secret (confide ntial) in order to protect your privacy? Information collected about y ou and your health (called prot ected health information), will be stored in locked fili ng cabinets or in comp uters with security passwords. Only certain people have the legal right to revi ew these research re cords, and they will protect the secrecy (confidentiality) of these re cords as much as th e law allows. These people include the researchers fo r this study, certain University of Florida officials, the hospital or clinic (if any) involved in this research, and the Institutional Review Board (IRB; an IRB is a group of people who are re sponsible for looking af ter the rights and welfare of people taking part in research). Otherwise your research records will not be released without your permission unl ess required by law or a court order. Results of genetic testing will not be included in your medi cal record and will be kept in a secure electronic database that is availa ble only to the study investigators. Your records will not be released to any person or agency unles s you provide written consent to Jennifer Dungan requesting the release of the results of the study. Jennifer Dungan will be responsible for making sure that any stored blood sample is protected in the specimen bank and that your medical information is kept confidential. Your sample will not be stored with your name or other identifying information, but instead will be given a code number to protect your identity These samples and this code number will only be gi ven to researchers whose re search is approved by an Institutional Review Board (IRB). (An IRB is a group of people who are responsible for looking after the rights and welfare of people taking part in research.) The researchers will not be told who you are or given any id entifying information about you. Because the nature and va lue of any future research cannot be known at this time, any results obtained from using your tissue will not be given to you or your doctor. If the results of this research are publishe d or presented at scientific meetings, your identity will not be disclosed. If you participate in this research study, th e researchers will collect use, and share your protected health information with others. Items 17 to 26 below describe how this information will be collect ed, used, and shared.

PAGE 141

128 17. If you agree to participate in this research study, what protected health information about you may be collect ed, used and shared with others? To determine your eligibility for the study and as part of your participation in the study, your protected health informa tion may be collected, used, an d shared with others to determine if you can participate in the study, and then as part of your participation in the study. This information can be gathered from you or your past, current or future health records, from procedur es such as physical examina tions, x-rays, blood or urine tests or from other procedur es or tests. This info rmation will be created by participating in study procedures, or from your study visits and telephone calls. More specifically, the following info rmation may be collected, used, and shared with others: Complete past medical history, current me dications, race, age, and records of previous blood pressures obtained duri ng physical examinations to determine eligibility criteria. Also, as part of the demographic informati on obtained from you to be used in data analysis, the following information will be obtained: name, age, phone number, race, ethnicity, current medications records of previous blood pressure, income, education, job, marriage status, number of people in your household, first degree relative information and information about your sm oking, drinking, and exer cise habits. If you agree to be in this research study, it is possible that some of the information collected might be copied into a "limited da ta set" to be used for other research purposes. If so, the limited data set ma y only include information that does not directly identify you. For ex ample, the limited data set cannot include your name, address, telephone number, social secu rity number, or any other photographs, numbers, codes, or so forth that link you to th e information in the li mited data set. If used, limited data sets have legal agr eements to protect your identity and confidentiality and privacy. 18. For what study-related purposes will your protected health information be collected, used, and sh ared with others? Your protected health inform ation may be collected, used, and shared with others to make sure you can participate in the re search, through your participation in the research, and to evaluate the results of th e research study. Mo re specifically, your protected health information may be collect ed, used, and shared with others for the following study-related purpose(s): 1) to determine the impact of gene expr ession on high blood pre ssure and response to certain medications 19. Who will be allowed to coll ect, use, and share your protected health information? Your protected health inform ation may be collected, used, and shared with others by:

PAGE 142

129 the study Principal Investigator Je nnifer R. Dungan and her staff the University of Florida Institutional Review Board Malcom Randall VA Medi cal center (Gainesville) 20. Once collected or used, who may your pr otected health information be shared with? Your protected health inform ation may be shared with: the study sponsors: The American Heart Association, The National Institutes of Nursing Research (of the National Ins titutes of Health) an d Sigma Theta Tau, Alpha Theta Chapter United States and foreign governmental agencies who are responsible for overseeing research, such as the Food and Drug Administration, the Department of Health and Human Servi ces, and the Office of Human Research Protections Government agencies wh o are responsible for ove rseeing public health concerns such as the Centers for Diseas e Control and Federal, State and local health departments other professionals at the University of Florida or Shands Ho spital that provide study-related treatment or procedures Malcom Randall VA Medi cal center (Gainesville) 21. If you agree to participate in this rese arch, how long will you r protected health information be used and shared with others? Your protected health information will be used until the end of th e study. Thereafter, your protected health informatio n will be stripped from the research database and the database containing the unidentifiable info rmation will be kept under the protection of the principal investigator forever. 22. Why are you being asked to allow the colle ction, use and sharing of your protected health information? Under a new Federal Law, researchers cannot co llect, use, or share with others any of your protected health informa tion for research unless you a llow them to by signing this consent and authorization. 23. Are you required to sign th is consent and authorizatio n and allow the researchers to collect, use and share with others your protected health information? No, and your refusal to sign will not affect your treatment payment, enrollment, or eligibility for any benefits out side this research study. However, you cannot participate in this research unless you allow the coll ection, use and sharing of your protected health information by signing this consent/authorization.

PAGE 143

130 24. Can you review or co py your protected health inform ation that has been collected, used or shared with others under this authorization? You have the right to review and copy your protected health information. However, you will not be allowed to do so until after the study is finished. 25. Is there a risk that your protected health information could be given to others beyond your authorization? Yes. There is a risk that information recei ved by authorized persons could be given to others beyond your authorizati on and not covered by the law. 26. Can you revoke (cancel) your authorizatio n for collection, use and sharing with others of your protected health information? Yes. You can revoke your authorization at any time before, during, or after your participation in the research. If you revoke, no new information will be collected about you. However, information th at was already collected ma y still be used and shared with others if the researchers have relied on it to complete and protect the validity of the research. You can revoke your authoriza tion by giving a written request with your signature on it to the Principal Investigator. 27. How will the researcher(s) benefi t from your being in this study? In general, presenting research results helps the career of a scien tist. Therefore, the Principal Investigator may benefit if the results of this study are presented at scientific meetings or in scientific journals. 28. Signatures As a representative of this study, I have ex plained to the participant the purpose, the procedures, the possible benefits, and the risks of this research study; the alternatives to being in the study; and how the participan t’s protected health information will be collected, used, and shared with others: _________________ _________________ ________________ _________________ Signature of Person Obtaining Consent and Authorization Date You have been informed about this study’s purpose, procedures, possible benefits, and risks; the alternatives to being in the st udy; and how your protected health information will be collected, used and shared with others. You have rece ived a copy of this Form. You have been given the opportunity to ask questions before you sign, and you have been told that you can ask other questions at any time.

PAGE 144

131 You voluntarily agree to participate in this study. You hereby authorize the collection, use and sharing of your protected health in formation as described in sections 17-26 above. By signing this fo rm, you are not waiving a ny of your legal rights. _________________ _________________ ________________ _________________ Signature of Person Consenting and Authorizing Date VA regulations require a witness for all of the signatures provided above. _________________ _________________ ________________ _________________ Signature of Witness Date

PAGE 145

132 APPENDIX D TAQMAN REAL-TIME PCR AMPLIFICATION PLOTS A B C Figure D-1. TaqMan Real-time amplif ication plot for each gene. A) 1A-ADR gene. B) 2-ADR gene. C) GAPDH normalizer gene Note: Red line in dicates threshold line and plots are for duplicate analyses, adjusted to the chosen calibrator and normalizer.

PAGE 146

133 LIST OF REFERENCES ALLHAT Collaborative Research Group ( 2000). Major cardiovascular events in hypertensive patients randomized to doxazosin vs. chlorthalidone: The antihypertensive and lipid-lowering treatme nt to prevent hear t attack trial. Journal of the American Medical Association 283, 1967-75. Ambler, S. K., & Brown, R. D. (1999). Genetic determinants of blood pressure regulation. Journal of Cardiovascular Nursing 13(4), 59-77. American Heart Association (AHA) (2003). Heart and stroke statistical update Dallas, TX: AHA, 2002. American Heart Association (AHA) (2005). Heart Disease and Stroke Statistics — 2005 Update Dallas, TX.: American Heart Association; 2004. American Sociological Association (2003). The importance of collecting data and doing social scientific research on race. Washington, DC: American Scociological Association. Anderson, N. B., McNeilly, M., & Myers, H. (1992). Toward understanding difference in autonomic reactivity: A proposed contextual model (Ch. 7). In J. R. Turner (Ed.), Individual differences in cardiovascular response to stress. New York: Plenum Press, pp. 125-45. Beevers, G., Lip, G. Y., & O’Brien, E. (2001). ABC of hypertension: The pathophysiology of hypertension. British Medical Journal 322, 912-16. Bengtsson, K., Orho-Melander, M., Melander, O., Lindblad, U., Ranstam, J., and Ranstam, L., et al. (2001). Beta(2)-adre nergic receptor gene variation and hypertension in subjects with type 2 diabetes. Hypertension 37(5),1303-8. Berecek, K. H., & Carey, R. (1999). Adrenerg ic and dopaminergic r eceptors and actions. In Hypertension Primer J. L. Izzo and H. R. Black, (Eds.). American Heart Association, pp 3-6. Bonham, V. L. (2003). Race, genetics and h ealth disparities: W hy we all must be engaged. Presentation given at the NIH-NINR Summer Genetics Institute, Bethesda, MD, July 23, 2003.

PAGE 147

134 Bray, M. S., Krushkal, J., Li, L., Ferrell, R., Kardia, S., Sing, C. F., and Turner, S. T., et al. (2000). Positional genomic analysis identifies the 2-adrenergic receptor gene as a susceptibility locus for human hypertension. Circulation 101, 2877-82. Burchard, E. G., Ziv, E., Coyle, N., Gomez, S. L., Tang, H., Karter, A. J., Mountain, J. L., Perez-Stable, E. J., Sheppard, D., Ri sch, N. (2003). The importance of race and ethnic background in biomedical re search and clinical practice. New England Journal of Medicine, 348(12), 1170-5. Bustin, S. A. (2002). Quantification of mRNA using real-time reverse transcription PCR (RT-PCR): Trends and problems. Journal of Molecular Endocrinology 29(1):2339. Bustin, S. A. (2000). Absolute quantifi cation of mRNA using real-time reverse transcription polymerase chain reaction assays. Journal of Molecular Endocrinology 25, 169-93. Chruscinski, A., Brede, M. E., Meinel, L., Lohse, M. J., Kobilka, B. K., & Hein, L. (2001). Differential distribution of -adrenergic receptor subtypes in blood vessels of knockout mice lacking 1or 2-adrenergic receptors. Molecular Pharmacology 60, 955-62. Cipolla, M. J., Harker, C.T., & Porter, J. M. (1996). Endothelial function and adrenergic reactivity in human type-II di abetic resistance arteries. Journal of Vascular Surgery 23, 940-9. Cockroft, J. R., Gazis, A. G., Cross, D. J., Wheatley, A., Dewar, J., and Hall, I. P., et al. (2000). 2-adrenoceptor polymorphism determines vascular reactivity in humans. Hypertension 36, 371-5. Cooper, R., & Rotimi, C. (1997). Hypertension in blacks. American Journal of Hypertension 10, 804-12. Cooper, R., Kaufman, J., & Ward, R. (2003). Race and genomics. New England Journal of Medicine 348 (12), 1166-70. Crews, D. E., & Bindon, J. R. (1991). Ethnici ty as a taxonomic tool in biomedical and biosocial research. Ethnicity & Disease 1, 42-9. CRISP Thesaurus (2005). Retrieved August 2, 2005, from www.crisp.cit.nih.gov/Thesaurus/index Cushman, W. C., Reda, D. J., Perry, H. M., Williams, D., Abdellatif, M., & Materson, B. J. (2000). Regional and racial differen ces in response to antihypertensive medication use in a randomized controlled trial of men with hypertension in the United States. Departmen of Vetera ns Affairs Cooperative Study Group on Antihypertensive Agents. Archives of Internal Medicine 160 (6), 825-31.

PAGE 148

135 Dawson, H. D. (2003). Real Time PCR Biotechnology training course attended on 7/16/03 at National Institutes of Health, Bethesda, MD. DiBona, G. F. (1989). Sympathe tic nervous system influences on the kidney: Role in hypertension. American Journal of Hypertension 2, 119s-124s. Dincer, U. D., Onay, A., Ari, N., Ozcelikay, A. T., & Altan, V. M (1998). The effects of diabetes on beta-adrenoceptor mediated response.veness of human and rat atria. Diabetes Research and Clinical Practice 40(2), 113-22. Dorak, M. T. (2003). Real-Time PCR Retrieved June 13, 2003, from http://droakmt.tripod.com /genetics/realtime.html Drevdahl, D., Taylor, J. Y., & Phillips, D. A. (2001). Race and ethnic ity as variables in nursing research, 1952-2000. Nursing Research 50(5), 305-13. Duster, T. (2001). Buried alive: The concept of race in science. The Chronicle of Higher Education 48 (3), Retrieved November 2, 2003 from http://chronicle.com/weekly/v48/i03/03b01101.htm Eberhardt, M. S., Ingram, D. D., Makuc, D. M, et al. (2001). Urban and Rural Health Chartbook. Health, United States, 2001. Hyattsville, Maryland: National Center for Health Statistics. Erdfelder E, Faul F & Buchner A (1996). GP OWER: A general power analysis program. Behavior Research Methods, Instruments, and Computers 28,1-11. Federal Drug Administration (FDA) (2003). FDA issues guida nce for collection of race and ethnicity data in clinical trials for FDA regulated products. FDA Talk Paper Retrieved June 15, 2003 from http://www.fda.gov/bbs/topi cs/ANSWERS/2003/ANS01193.html Ferro, A., Kaumann, A. J., & Brown, M. J. (1993). Beta 1and beta-2ADR-mediated relaxation in human internal mammary ar tery and saphenous vein: Unchanged betaand alpha-ADR responsiveness afte r chronic beta 1-ADR blockade. British Journal of Pharmacology 109 (4), 1053-8. Ferro, A., & Walton, R. (2001). Racial di fferences in the effectiveness of nonpharmacologic treatment of hypertension. Hypertension 39 (4), E24. Fine, M. J., Ibrahim, S. A., & Thomas, S. B. (2005). Editorial: The role of race and genetics in health disparities research. American Journal of Public Health 95 (12), 2125-28. Fredrikson, M., Tuomisto, M., & Sundin, O. (1990). Classical conditioning of vascular responses in mild hypertensives and normotensives. Journal of Hypertension 8 (12), 1105-9.

PAGE 149

136 Fullilove, M. T. (1994). Deconstr ucting race in me dical research. Archives of Pediatrics and Adolescent Medicine 148 (10), 1014-15. Gaballa, M.A., Peppe, K., Lefkowitz, R. J., Aguirre, M., Dober, P. C., and Pennock, G. D.,et al.(1998). Enhanced vasorelaxati on by overexpression of beta-2-adrenergic receptors in large arteries. Journal of Molecular and Cellular Cardiology 30, 1037-45. Gow, I. F., Mitchell, E., & Wait, M. ( 2003). Adrenergic receptors in the bovine mammary artery. Biochemical pharmacology 65, 1747-53. Gu, C. Borecki, I., Gagnon, J., Bouchard, C., Leon, A. S., Skinner, J. S., et al. (1998). Familial resemblance fro resting blood pressu re with particular reference to racial differences: Preliminary analyses from the HERITGE Family Study. Human Biology 70. 77-90. Guthrie, R. M. & Siegel, R. L. (1999). A multicenter, community-based study of doxazosin in the treatment of concomita nt hypertensioin and symptomatic benign prostatic hyperplasia: The Hypertensi on and BPH Intervention Trial (HABIT). Clinical Therapeutics 21, 1732-48. Gygi, S. P., Rochon, Y., Franza, B. R., & Aebersold, R. (1999). Correlation between protein and mRNA abundance in yeast. Molecular and Cell Biology 19, 1720-30. Helgadottir, A., Manolescu, A., Helgason, A., Thorleifsson, G., Thorsteinsdottir, U., and Gudbjartsson, D. F., et al., (2006). A varian t of the gene encoding leukotriene A4 hydrolase confers ethnicity-specific risk of myocardial infarction. Nature Genetics 38(1), 68-74. Herrmann, S. M., Nicaud, V., Tiret, L. Evans, A. ,Kee, F., & Ruidavets, J. B., (2002) Polymorphisms of the 2-adrenoceptor (ADRB2) gene and essential hypertension: the ECTIM and PEGASE studies. Journal of Hypertension 20, 229–35 Hill, M. N. (2000). Comprehensive hypert ension care in young urban black men: an example of a program of nursing research th at integrates genetic science, clinical interventions, and patient outcomes. Nursing Clinics of North America 35(3), 77393. Hindorff, L. A., Heckbert, S. R., Psaty, B. M., Lumley, T., Siscovick, D. S., and Herrington, D. M., et al. (2005). Beta(2 )-Adrenergic recepto r polymorphisms and determinants of cardiovascular risk : the Cardiovascular Health Study. American Journal of Hypertension 18(3):392-7. Hugot, J. P., Chamaillard, M., Zouali, H., Lesa ge, S., Cezard, J. P., and Belaiche, J., et al., (2001). Association of NOD2 leucine-rich repeat varian ts with susceptibility to Crohn's disease. Nature 411(6837), 599-603.

PAGE 150

137 Human Genome Project (2005). Talking gl ossary. Retrieved August 2, 2005, from http://www.genome.gov/glossary.cfm Humphreys, G. S., & Delvin, D. G. (1968). In effectiveness of propanolol in hypertensive Jamaicans. British Medical Journal 2, 601-3. Hypertension Detection and Followup Program Cooperative Group (1977). Blood pressure studies in 14 communities: a two-stage screen for hypertension. Journal of the American Medical Association ; 237:2385–91. Iaccarino, G., Cipolletta, E., Fior illo, A., Annecchiarico, M., Ciccarelli, M., Cimini, V., Koch, W. J., & Trimarco, B. (2002). 2-Adrenergic receptor gene delivery to the endothelium corrects impaired adrenerg ic vasorelaxation in hypertension. Circulation, 106, 349-55. International Society of Nurses in Genetics (ISONG) (2005). About ISONG. Retrieved January 20, 2006, from: http://www.isong.org/about/index.cfm Jacobs, M. C., Lenders, J. W., Willemsen, J. J., & Thien, T. (1997). Adrenomedullary secretion of epinephrine is increas ed in mild essential hypertension. Hypertension 29, 1303-08. Jamerson, K., & DeQuattro, V. (1996). The impact of ethnicity on response to antihypertensive therapy. American Journal of Medicine 101 (3A), 22S-32S. Jenkins, J., Grady, P. A., & Collins, F. S. (2005). Nurses and the genomic revolution. Journal of Nursing Scholarship, 37(2), 98-101. Jennings, K., & Parsons, V. (1976). A study of labetalol in patients of European, West Indian, and West African origin. British Journal of Clinical Pharmacology 3, 773S-5S. Jiang, S., Mao, G., Zhang, S., Hong, S., Tang, G., and Li, X., et al. ( 2005). Individual and joint association of 1A-adrenergic receptor Arg347Cys polymorphism and plasma irbesartan concentration with blood pressure therapeutic response in Chinese hypertensive subjects. Clinical Pharmacology and Therapeutics, 78 (3), 239-48. Kamata, K., Satoh, T., Tanaka, H., & Shigenobu, K. (1997). Changes in electrophysiological and mechanical responses of the rat papillary muscle to alphaand beta-agonist in strept ozotocin-induced diabetes. Canadian Journal of Physiology and Pharmacology 75(7),781-8. Kinugawa, T., Endo, A., Kato, M., Kato, T., Ahmmed, G. U., and Omodani, H., et al. (1997). Responses of plasma catecholam ines, renin-angiotensin-aldosterone system, and atrial natriuretic peptide to exercise in patients with essential hypertension. Cardiology 88(3),238-45.

PAGE 151

138 Knox, S. S., Hausdorf, J., & Markovitz, J. H. (2002). Reactivity as a predictor of subsequent blood pressure: Racial differences in th e Coronary Artery Risk Development in Young Adults (CARDIA) Study. Hypertension 40, 914-19. Krieger, N. (2005). Stormy weather: Race, gene expression, and the science of health disparities. American Journal of Public Health 95 (12), 2155-60. Langaee T., & Ronaghi, M. (2005). Genetic variation analyses by Pyrosequencing. Mutation Research, 3;573(1-2),96-102. Leech, C. J., & Faber, .J. E. (1996). Different -adrenoceptor subtypes mediate constriction of arte rioles and venules. American Journal of Physiology, 270, H71022. Li, G., Faulhaber, H., Rosenthal, M., Schuste r, H., Jordan, J., and Timmermann, B., et al. (2001). -2 adrenergic receptor ge ne variations and blood pr essure under stress in normal twins. Psychophysiology 38(3),485-9. Lieu, H. D., Withycombe, S. K., Walker, Q ., Rong, J. X., Walzem, R. L., and Wong, J. S., et al. (2003). Eliminating athero genesis in mice by switching off hepatic lipoprotein secretion. Circulation 107, 1315-21. Linden, W., Gerin, W., & Davidson, K. (2003). Cardiovascular reactivity: Status quo and a research agenda for the new millennium. Psychosomatic Medicine 65, 5-8. Livak, K. J., & Schmittgen, T. D. (2001). Analys is of relative gene expression data using real-time quantitative PCR a nd the 2 (-delta delta C(T)). Methods 25, 402-8. Lovallo, W. R., & Gerin, W. (2003). Psyc hophysiological reactivity: Mechanisms and pathways to cardiovascular disease. Psychosomatic Medicine 65, 36-45. Lowes, B. D., Gilbert, E. M., Abraham, W. T., Minobe, W. A., Larrabee, P., and Ferguson, D., et al. (2001). Myocardial gene expressi on in dilated cardiomyopathy treated with beta-blocking agents. New England Journal of Medicine 346 (18), 1357-65. Lucassen, A. (1999). Genetics of multifactorial diseases. In Practical Genetics for Primary Care by P.Rose and A.Lucassen. Oxford: Oxford University Press pp.145165. McAdoo, W. G., Weinberger, M. H., Miller, J. Z., Fineberg, N. S., & Grim, G. E. (1990). Race and gender influence hemodynamic responses to psychological and physical stimuli. Journal of Hypertension 8(10), 961-7. McCaffery, J. M., Pogue-Geile, M. F., Ferrell R. E., Petro, N., & Manuck, S. B (2002). Variability within and -adrenoceptor genes as a pr edictor of cardiovascular function at rest and in res ponse to mental challenge. Journal of Hypertension 20, 1105-14.

PAGE 152

139 McDougall, S. J., Paull, J. R. A., Widdop, R. E., & Lawrence, A. J. (2000). Restraint stress: Differential cardiovascular res ponses in Wistar-Kyoto and spontaneously hypertensive rats. Hypertension 35, 126-9. Melton, D. A., Kreig, P. A., Rebagliati, M. R., Maniatis, T., Zinn, K. & Green, M. R. (1984). Efficient in vitro synthesis of biologically ac tive RNA and RNA hypbridization probes from plasmids cont aining a bacteriophage SP6 promoter. Nucleic Acids Research 12, 7035-56. Merriam-Webster Dictionary, Online ( 2003). Retrieved February 4, 2006, from http://www.m-w.com/cgi-bin/dictionary Miller, J. W., Hu, Z. W., Okazaki, M., Fujinaga, M., & Hoffman, B. B. (1996). Expression of alpha-1 adrenergic receptor subtype mRNAs in the rat cardiovascular system with aging. Mechanisms of Ageing and Development 87, 75-89. Miller, S. B., & Ditto, B. (1991). Exaggerate d sympathetic nervous system response to extended psychological stress in offspring of hypertensives. Psychophysiology 28 (1), 103-13. Moniotte, S., Vaerman, J.-L., Kockx, M. M ., Larrouy, D., Langin, D., Noirhomme, P., & Balligand, J.-L. (2001). Real-time RT-PCR fo r the detection of beta-adrenoceptor messenger RNAs in small human endomyocardial biopsies. Journal of Molecular and Cellular Cardiology 33, 2121-33. Monopoli, A., Conti, A., Forlan i, A., & Ongini, E. (1993). 1 and 2 adrenoceptors are involved in mediating vasodilation in the human coronary artery. Pharmacological Research 27 (3), 273-9. National Heart, Lung, and Blood Instittute (NHLBI) Healthy People 2010 (2003). NHLBI Cardiovascular hea lth for all performance goals. Retrieved November 12, 2003, from http://hin.nhlbi.nih.gov/m inority/aa.frameset.htm Nishio, E., & Watanabe, Y. (1999). Troglitaz one inhibits alpha-1-adrenoceptor-induced DNA synthesis in vascular smooth muscle cells. European Journal of Pharmacology 374, 127-35. Online Mendelian Inheritance in Man, (O MIM) (TM). Johns Hopkins University, Baltimore, MD. 1A MIM Number: {104221}: {2/7/ 2002}. Retrieved June 25, 2003, from http://www.ncbi.nlm.nih.gov/omim/ Online Mendelian Inheritance in Man, (O MIM) (TM). Johns Hopkins University, Baltimore, MD. 2 MIM Number: {109690}: {2/10/ 2003}. Retrieved June 25, 2003 from http://www.ncbi.nlm.nih.gov/omim/ Pharmacogenetics and Pharmacogenomics Knowledge Base (Pharm GKB), (2006) Retrieved January 17, 2006, from http://www.pharmgkb.org/index.jsp

PAGE 153

140 Phimister, E. G. (2003). Medi cine and the racial divide. New England Journal of Medicine 348 (12), 1081-82. Pincus, F. L., & Ehrlich, H. J. (1999). The study of race and ethnic relations. In F. L. Pincus & H. J. Ehrlich (Eds.), Race and ethnic conflict (2nd ed.) (p. 11-13). Boulder, CO: Westview Press. Portney, L. G., & Watkins, M. P. (2000). Foundations of clinical research: Applications to practice (2nd ed.). Upper Saddle Rive r, NJ: Prentice Hall. Reja, V., Goodchild, A. K., & Pilowsky, P. M. (2002). Catecholamine-related gene expression correlates with blood pressure in SHR. Hypertension 40, 342-7. Rice University (2005). Connections: Shar ing knowledge and building communities. Reteived August 2, 2005, from http://cnx.rice.edu/ Ridker, P. M., Miletich, J. P., Henneke ns, C. H., & Buring, J. E.. (1997). Ethnic distribution of Factor V Leiden in 40 47 men and women. Implications for venous thromboembolism screening. Journal of the Americ an Medical Association, 277(16),1305-7. Risch, N., Burchard, E., Ziv, E., & Tang, H. (2002). Categorization of humans in biological research: genes, race and disease. Genome Biology 3 (7), 2007.12007.12. Rokosh, D. G., & Simspon, P. C. (2002). Knockout of the -1A/C-adrenergic receptor subtype: The 1A/C is expressed in resistance arte ries and is required to maintain arterial blood pressure. Proceedings of the National Academy of Science 99 (14), 9474-9. Rotimi, C. N., Cooper, R. S., Cao, G., Ogunbi yi, O., Ladipo, M. et al. (1999). Maximumliklihood generalized heritability estimates for blood pressure in Nigerian families. Hypertension 35, 1141-47. Sankar, P., & Cho, M. (2002). Genetics. To ward a new vocabulary of human genetic variation. Science 298, (5597), 1337-8. Schubert, P. E., & Lionberger, H. J. (1999). Cu ltural and spiritual pers pectives. In J. E. Hitchock, P. E. Schubert, & S. A. Thomas (Eds.), Community health nursing: Caring in action (p. 111-35). Albany, NY: Delmar. Seedat, Y. K. (1980). Trial of atenolol and chlorthalidone for hypertension in black South Africans. British Medical Journal 281, 1241-3. Seventh Report of the Joint National Comm ittee on Prevention, Detection, Evaluation, and Treatment of High Blood Pressure (JNC VII Express) (2003). Retrieved March 6, 2003, from http://www.nhlbi.nih.gov/guidelines/ hypertension/jncintro.htm

PAGE 154

141 Shen, M. C., Lin, J. S., Tsay, W. (1997). Hi gh prevalence of antithrombin III, protein C and protein S deficiency, but no Factor V Leiden mutation in venous thrombophilic Chinese patients in Taiwan. Thrombosis Research 87(4):377-85. Shibata, K., Hirasawa, A., Moriyama, N., Kawabe, K., Ogawa, S., & Tsujimoto, G. (1996). Alpha1a-adrenoceptor polymorphi sm:pharmacological characterizationand association with beni gn prostatic hypertrophy. British Journal of Pharmacology, 118,1403–8. Sleight, P. (2004). Fact sheet : Isolated hypertension (ISH). World Hypertension League. Retreived October 14, 2004, from: http://www.mco.edu/whl/isyshype.html Small, K. M., McGraw, D. W., & Liggett S. B. (2003). Pharmacology and physiology of human adrenoceptor polymorphisms. Annual Review of Pharmacology and Toxicology 43, 381-411. Stein, C. M., Lang, C. C., Singh, I., He, H. B., & Wood, A. J. (2000). Increased vascular adrenergic vasoconstriction and decreased vasodilation in blacks. Additive mechanisms leading to enha nced vascular reactivity. Hypertension 36(6), 945-51. Timmermann, B., Mo, R., Luft, F. C., Gerdts E., Busjahn, A., and Omvik, P., et al. (1998). Beta-2 adrenoceptor genetic variation is a ssociated with genetic predisposition to essential hypertensi on: The Bergen Blood Pressure Study. Kidney International 53(6), 1455-60. Tomaszewski, M., Brain, N. J. R., Charchar F. J., Wang, W. Y. S., Lacka, B., and Padmanabahn, S., et al. (2002) Essential hypertension and 2-adrenergicreceptor gene: Linkage and association analysis. Hypertension 40, 286–91. Trogan, E., Choudhury, R. P., Dansky, H. M., Rong, J. X., Breslow, J. L., & Fisher, E. A. (2002). Laser capture microdi ssection analysis of gene expression in macrophages from atherosclerotic lesions of apolipoprotein E-deficient mice. Proceedings of the National Academy of Sciences 99 (4), 2234-9. United States Department of Heath and Hu man Services (USDHHS) (2003). Guidance for industry: Collection of race and ethnicity data in clinical tr ials: draft guidance. World Wide Web URL: http: //www.fda.gov/cder/guidance Veglio, F., Tayebati, S. K., Schiavone, D., Ricci A., Mulatero, P., and Bronzetti, E., et al. (2001). Alpha-1-adrenergic receptor subtypes in periphe ral blood lymphocytes of essential hypertensives. Journal of Hypertension 19, 1847-54. Wang, T., & Brown, M. J. (2001). Influence of 1-adrenoceptor blockade on the gene expression of adenylate cyclase subtypes and -adrenoceptor kinase in human atrium. Clinical Science 101, 211-14.

PAGE 155

142 Williams, D. R. & Jackson, J. S. (2000). Race/ethnicity and the 2000 Census: Recommendations for African Amerian a nd other black populations in the United States. American Journal of Public Health 90 (11), 1728-30. Wilson, J. F., Weale, M. E., Smith, A. C., Gratrix, F., Fletcher, B., Thomas, M. F., Bradman, N., Goldstein, D. B. (2001). Popul ation genetic structure of variable drug response. Nature Genetics 29, 265-9. Winant, H. (2000). Race and race theory. Annual Review of Sociology 26, 169-185. Witzig, R. (1996). The medicalization of race: Sc ientific legitimization of a flawed social construct. Annals of Internal Medicine 125 (8), 675-9. Wood, A. J., & Zhou, H. H. (1991). Ethnic differences in drug disposition and responsiveness. Clinical Pharmacokinetics 20(5), 350-73. Wood, A. J. (2002). Variability in -adrenergic receptor response in the vasculature: Role of receptor polymorphism. Journal of Allergy and Clinical Immunology 110, S318-21. Wu, J., Kraja, A. T., Oberman, A., Lewis, C. E., Ellison, R. C., and Arnett, D. K., et al. (2005). A summary of the effects of an tihypertensive medications on measured blood pressure. American Journal of Hypertension 18, 935-42. Xie, H. G., Kim, R. B., Stein, C. M., Gain er, J. V., Brown, N. J., Wood, A. J. (1999). Alpha1A-adrenergic receptor polymorphism : Association with ethnicity but not essential hypertension. Pharmacogenetics 9(5):651-6. Xu, K., Lu, Z., Wei, H., Zhang, Y., & Han, C. (1998). Alteration of -adrenoceptor subtypes in aortas of 12-month-ol d spontaneously hypertensive rats. European Journal of Pharmacology 344, 31-6. Yinger, J. M. (1994). Ethnicity: Source of strength? Source of conflict? Albany, NY: SUNY Press. Young, M. A., Vatner, D. E., & Vatner, S. F. (1990). Alphaand beta-adrenergic control of large coronary arteri es in conscious calves. Basic Research in Cardiology 85 (Suppl 1), 97-109. Zamorano, P. L., Mahesh, V. B. & Bra nn, D. W. (1996). Quantitative RT-PCR for neuroendocrine studie s: A minireview. Neuroendocrinology 63, 397-407. Zeegers, M. P., Rijsdijk, F., Sham, P., Fagard, R., Gielen, M., and de Leeuw, P. W., et al., (2004).The contribution of risk factors to blood pressure heritability estimates in young adults: The East Flanders prospective twin study. Twin Research 7(3),24553.

PAGE 156

143 BIOGRAPHICAL SKETCH Jennifer Dungan was born Jennifer Rene’ Elizabeth Chodzinski in Melbourne, Florida. Having an interest in the medical field early, she became a certified nursing assistant through the dual-enrollment program at the community college and worked in two sub-acute/nursing home facilities in Melbourne. She completed her bachelor’s degree in nursing at the Univers ity of Florida in 2001. During her undergraduate e ducation, she was one of th ree nursing students chosen to participate in the UF University Scholar s Undergraduate Research Program, where she was mentored by Carolyn Yucha, then Associ ate Dean for Research in Nursing. Always having an interest in cardiova scular disease, she focused her mini-research project on a non-pharmacological intervention to reduce bl ood pressure in a dults, incorporating biofeedback as a research tool. She conti nued this project for her Honors project in nursing, and was able to present her research fi ndings as a student at local, regional, and national conferences. She gra duated with high honors and wa s the recipient of the UF College of Nursing Research Award. Following graduation, she worked full-time on the Cardiac-Telemetry Unit at Shands at Alachua General Hospital (AGH) and was accepted into the UF College of Nursing’s Accelerated Bachelor’s to PhD Program the following Fall semester. During this rigorous program, she continued to wo rk part-time at AGH and as a research assistant in the UF CON Office for Research Support. She completed her master’s in adult health nursing in December 2002. Her pers onal family history of cardiac disease

PAGE 157

144 and her experiences on the cardiac-telemetry un it led her to become interested in the genetics of cardiovascular disease. In the summer of 2003, she was one of twenty nurses accepted into the National Institutes of Hea lth (NIH), National Institute of Nursing Research’s (NINR) Summer Genetics Institu te (SGI). Through this intense summer program, she obtained didactic and laboratory training in mol ecular and clinical genetics, and developed her proposal for her dissertat ion study. This program provided her with her minor in genetics, as granted by Georgeto wn University in affiliation with the NIHNINR SGI program. Upon her return in the Fall, she worked with her advisor, Carolyn Yucha’s R01 grant, “Prediction of Biofeedback Success, ” where she genotyped subject samples and examined adrenergic receptor genotypes as pr edictors of biofeedback success in lowering blood pressure in adults with hypertension. Fo r her presentation on th is project, she won Best Student Poster at the annual m eeting of the Associ ation of Applied Psychophysiology and Biofeedback. This contin ued to foster her goals in genetics research in cardiova scular disease. Ms. Dungan secured partial funding for he r dissertation study from the American Heart Association. She was then awarded th e Ruth L. Kirschstein National Research Service Award from the NIH-NINR, which pr ovided additional training and funding for her study. In addition, Jennifer obtained a sma ll research grant from the Alpha Theta Chapter of Sigma Theta Tau Intern ational Nursing Honor Society. Ms. Dungan’s long-term career goal is to develop a successful, well-funded research program in translational resear ch involving cardiovasc ular genetics. Her underlying objective is to better serve patient s through science. Her particular area of

PAGE 158

145 interest is hypertension and how susceptibility genes and e nvironmental factors interplay to facilitate poor outcomes. After defending her dissertation in February, she will be entering a Postdoctoral Fellowship in Aging at the Duke University Center for the Study of Aging and Human Development. She will focus her postdoctoral training in the areas of gerontology, genomics, genetics and cardiovascular disease.


Permanent Link: http://ufdc.ufl.edu/UFE0013487/00001

Material Information

Title: Alpha 1A- and Beta 2-Adrenoceptor Gene-Expression Differences in Hypertensive and Normotensive Persons by Race
Physical Description: Mixed Material
Copyright Date: 2008

Record Information

Source Institution: University of Florida
Holding Location: University of Florida
Rights Management: All rights reserved by the source institution and holding location.
System ID: UFE0013487:00001

Permanent Link: http://ufdc.ufl.edu/UFE0013487/00001

Material Information

Title: Alpha 1A- and Beta 2-Adrenoceptor Gene-Expression Differences in Hypertensive and Normotensive Persons by Race
Physical Description: Mixed Material
Copyright Date: 2008

Record Information

Source Institution: University of Florida
Holding Location: University of Florida
Rights Management: All rights reserved by the source institution and holding location.
System ID: UFE0013487:00001


This item has the following downloads:


Full Text












ALPHA 1A- AND BETA 2-ADRENOCEPTOR GENE EXPRESSION DIFFERENCES
IN HYPERTENSIVE AND NORMOTENSIVE PERSONS BY RACE













By

JENNIFER RENE' DUNGAN


A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL
OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT\
OF THE REQUIREMENTS FOR THE DEGREE OF
DOCTOR OF PHILOSOPHY

UNIVERSITY OF FLORIDA


2006

































Copyright 2006

by

Jennifer Rene' Dungan

































This document is dedicated to my husband, Craig for his endless love and support, to my
Great-grandmother, Margaret Gray for inspiring me to become a nurse, to my
Grandmother, Julia Chodzinski, who has shown me the power of perseverance and the
importance in believing in my dreams, and finally, to my mother, Cheryl Crossland, who
instilled in me a strong sense of hard-work and determination, and who has given me
confidence when I needed it most.















ACKNOWLEDGMENTS

I gratefully acknowledge the contributions, guidance, and encouragement of my

dissertation committee Chair, Carolyn Yucha, PhD, and members, Julie Johnson, Pharm

D, Yvette Conley, PhD, Shawn Kneipp, PhD, and Taimour Langaee, PhD.

I also extend my appreciation to the numerous people who assisted me with the

completion of this project in many important ways: The UF and VA TCV Surgery

Departments; the staff at the UF ICBR facilities; and my colleagues and peers in the

College of Nursing. In addition, I extend my warmest thanks to my family and friends for

their support throughout this process. Special thanks go to Mandy Elliott for living this

experience with me and being the most supportive friend anyone could have throughout

this process.

Furthermore, I would like to acknowledge the funding agencies that financially

supported this project: The National Institute of Nursing Research, the American Heart

Association, and the Alpha Theta Chapter of Sigma Theta Tau International Nursing

Honor Society.

Finally, I gratefully acknowledge the participants of this study for their important

contributions to the success of this project.
















TABLE OF CONTENTS



A C K N O W L E D G M E N T S ................................................................................................. iv

LIST OF TABLES .................................................... ........ .. .............. viii

LIST OF FIGURES ............................... ... ...... ... ................. .x

ABSTRACT .............. ..................... .......... .............. xii

CHAPTER

1 IN TR OD U CTION ............................................... .. ......................... ..

Background and Problem Statem ent ........................................ ........................ 1
Purpose of the Study ............... ............... .................................. .6
H ypotheses ................................................ 7
D definitions of Term s ............... ................. ........... ................... .... 8
A ssum options ............................................................ 9
Significance of the Study ............................................................................ .... .......10

2 REVIEW OF LITERATURE ........................................................ .............. 16

Introduction .................................................... ........ ............................. 16
Genetic Influences on Essential Hypertension ....................................................16
G enetic M odels of H ypertension................... ........................................... .......... 17
Cardiovascular Reactivity, HTN and Adrenergic Receptors................................ 18
Alpha 1A- and Beta 2-Adrenergic Receptors ................... ................19
Pharmacotherapeutic Aspects of Alpha-1A and Beta-2 Adrenergic Receptors .........21
Population D differences in H TN ............................................................. ................... 22
Recommendations for the Collection of Racial and Ethnic Data........................25
Implications for Studying Disease by Race............................................. 28
P positive Inotrope A dm inistration..................................................... .....................33
G enetic A analysis Techniques ........................ ... ................. ............... ....33
Collection of Human Tissues for Gene Expression Analyses.................................38
S u m m ary ...................................... .................................................. 3 9

3 PROCEDURES AND M ETHODS ........................................ ........................ 40

Intro du action ...................................... ................................................ 4 0









D e sig n ................................................................................................................... 4 0
Subject R ecruitm ent ............................. .................... .............. .. 41
Consenting Process and HIPAA Regulations................... .................................43
S e ttin g .......................................................................................4 4
Research V ariables .................................... ..... .......... .............. .. 44
Study Protocol ............................................................... 46
Data Collection and Laboratory M ethods ................................. ............... 46
B lood and tissue collection ........................................ ....... ............... 46
G enom ic DN A analyses ................................................................ ......... 47
RNA isolation and reverse-transcription............. ...... ..............49
Real-tim e polym erase chain reaction ................................. ............... 52
Positive inotrope data collection ........................ ............................ .... 53
Calculations for Relative Gene Expression and Selection of Calibrator ............54
M ethods for Statistical A nalyses ........................................ .......................... 55

4 R E S U L T S .......................................................................... 5 7

Introduction ..................................................................57
D escriptiv e R esu lts .................................................................. ........ ...... .. 57
Subject D em ographics......... ................. ................................ ............... 57
Assessment of GAPDH for Relative Quantitation ...........................................61
Assumptions of Normality ................................. ......... .................. 64
A nalytic R results for H ypotheses ........................................ ...... ............... 64
Exploratory A im s ............................ .. ...................... ............ ........ 70
Effect Sizes and Power Calculations .............. ...........................................76

5 DISCU SSION AND RESULTS ......................................... ............................ 78

Intro du action ...................................... ................................................ 7 8
D iscu ssion of R esults......... ............................................................ .. .... ...... 78
Dem graphics ......................................... .............. 78
Gene Expression Measures of Central Tendency and Variance .........................81
Discussions for Choice of GAPDH for Normalization Gene ..............................82
Assessment of the Performance of GAPDH as a Normalizer .............................83
A im s .............................................................................................................. 8 7
Exploratory A im s ...................................................... ..... .... 90
Sample versus population allele frequency comparisons ..........................90
Sample versus population genotype frequency comparisons....................91
L im itatio n s .... ............... ...................... .. .. .............. ...... ............... 9 4
N orm alization w ith GAPDH ..................................................... ...... ......... 94
P o w e r ............................................................................................................. 9 4
Internal Validity.................................................. ........ 95
Construct Validity of the Variable, Normotension ....................................96
External V validity .................................. .. .. ...... ...............96
M inim al Sam ple Tem plate ............................................................................ 96
C onfounding V ariables............................................... ............................. 97
N using R elevance .......... ..... .......... .... ......... ............. .... ........99









Practice- and Care-Related Relevance ................................... ............... 100
Nurse-Directed Research and Qualitative Findings .......................................102
C ase Stu dy ............................................................... .... .............. 10 3
Sum m ary................................ .. ................ ........... ......... 104
Recommendations for Future Research............. .................................... 105
C o n clu sio n s.................................................... ................ 10 6

APPENDIX

A SUBJECT ENROLLMENT/DEMOGRAPHIC FORM.......................................109

B UF IRB-01 INFORMED CONSENT FORM ......................................................... 112

C VA SCI INFORMED CONSENT FORM.................................... .....................122

D TAQMAN REAL-TIME PCR AMPLIFICATION PLOTS ...................................132

L IST O F R E F E R E N C E S ...................................................................... ..................... 133

BIOGRAPHICAL SKETCH ............................................................. ............... 143
















LIST OF TABLES

Table page

2-1 Adrenergic cardiovascular stress patterns. ............. .......................... ............... 23

3-1 Inclusion and exclusion criteria with rationale ....................................................... 42

3-2 G enotyping prim ers .................. ...................................... .. ............ 49

3-3 Target gene assay inform ation. ........................................ .......................... 53

3-4 Single-plex plate set-up, one sam ple ................................................ ............... 53

4-1 Demographics of all enrolled subjects. ..........................................................58

4-2 Clinical characteristics of all enrolled subjects. ...................................................58

4-3 Demographics for gene expression subset. ................................... ............... 60

4-4 Clinical characteristics for gene expression subset. ............................................60

4-5 Gene expression medians and IQRs for total sample.............................................64

4-6 Gene expression medians and IQRs for subjects by diagnosis.............................65

4-7 Median fold differences in gene expression between normotensive and
hypertensive subjects and Mann-Whitney U tests. ................................................65

4-8 Gene expression medians, IQRs, and minimum and maximum values for
W hite/Caucasian subjects ............... ....................... .. .... ............... 66

4-9 Median fold differences in gene expression between White/Caucasian
normotensive and hypertensive subjects and Mann-Whitney U tests....................67

4-10 Gene expression medians and IQRs for Black/AA subjects.............. ................67

4-11 Median fold differences in gene expression between White/Caucasian
hypertensive and Black/AA hypertensive subjects and Mann-Whitney U tests......68

4-12 Median, IQR, minimum and maximum values for alA-ADR and 32-ADR fold
difference in gene expression and need for post-operative positive inotrope
m education ............................................................................69









4-13 Fold differences in gene expression between non-inotrope and inotrope subjects
and M ann-W hitney U tests .......................................................................... .. ..... 70

4-14 Allele frequencies for population versus sample, by SIR/ancestry..........................71

4-15 Genotype frequencies for populaiton versus sample, by SIR/ancestry ..................71

4-16 Association between genotype and diagnoses of NT and HTN for the alA-ADR
and 32-ADR genes ........................ ........................ ......... 72

4-17 Fisher's Exact for genotype differences in White/Caucasian hypertensive vs.
norm otensive subjects. ....................................... ............. ..........74

4-18 Chi-square for allele counts by diagnosis for the alA-ADR and 32-ADR genes in
all subjects. ............................................................................ 74

4-19 Chi-square for alleles by diagnosis for the alA-ADR and 32-ADR genes in
W hite/Caucasian subjects ............... ....................... .. .... ............... 75

4-20 Kruskal-Wallis tests for genotype counts by gene expression alA-ADR and 12-
A D R genes. .......................................... ............................ 76

4-21 Pow er and effect sizes by aim ........................................................................... 76
















LIST OF FIGURES


Figure p

1-1 Abbreviations used. .......................................................... .. ........ .. 1

2-1 Chromosome (Ensembl human map view) showing the locations of both ADR
g e n e s ................................................................................................................... 2 0

2-2 Amplification of gene expression using TaqMan Real Time RT PCR ..................37

2-3 TaqMan RT-PCR steps (adapted from Bustin, 2000). ..................... ...............38

3-1 alA -ADR gene with promoter, intron and exon boundaries and investigated
p oly m orp h ism ...................................... ............................................. 4 5

3-2 02-ADR gene with promoter, exon boundary and investigated polymorphisms...... 46

3-3 Tissue pieces immersed in RNAlater preservation solution. ..................................47

3-4 B lotting tissue on K im w ipe ......................................................................... ....... 51

3-5 Grinding tissue in mortar and pestle on liquid nitrogen .......................................51

3-6 Pow dered tissue in m ortar. .............................................. .............................. 51

3-7 Homogenizing tissue slush with rotar-stator homogenizer................... ................52

4-1 Range of average duplicate Ct values of GAPDH per sample number...................61

4-2 Range of average duplicate GAPDH Ct measurements grouped by plate number..62

4-3 Boxplot for average duplicate GAPDH by diagnosis. ...........................................63

4-4 Boxplot for all gene expression raw Ct values............... ................................63

4-5 Boxplots for both gene's expression by diagnosis. ...............................................65

4-6 Boxplots for White/Caucasian subjects, for both gene's expression by diagnosis. .67

4-7a Boxplots for GaA-ADR gene expression for White/Caucasian HTN versus
Black/A A H TN subjects.......................................................... ............... 68









4-8 Boxplots for 32-ADR gene expression for White/Caucasian HTN versus
Black/A A H TN subjects.......................................................... ............... 68

4-9 Boxplots for both genes' expression by need for post-operative positive
inotrope..................................... .................. ................ ......... 69

4-10 Bar chart of GaA-ADR, codon 347 by diagnosis......... ................ ................72

4-11 Bar chart of 32-ADR, codon 16 by diagnosis.......... ...................................73

4-12 Bar chart of P2-ADR, codon 27 by diagnosis...................................73

D-1 TaqMan Real-time amplification plot for each gene.. ........................................132















Abstract of Dissertation Presented to the Graduate School
of the University of Florida in Partial Fulfillment of the
Requirements for the Degree of Doctor of Philosophy

ALPHA 1A- AND BETA 2-ADRENOCEPTOR GENE EXPRESSION DIFFERENCES
IN HYPERTENSIVE AND NORMOTENSIVE PERSONS BY RACE
By

Jennifer Rene' Dungan

May 2006

Chair: Carolyn B. Yucha
Major Department: Nursing

It has been hypothesized that genes of the adrenergic receptor (ADR) system

contribute to hypertension (HTN). This notion is supported by genetic (gene-knockout

and association), physiological, and pharmacological studies of ADRs in animal and

human models. The alA- and 32-subtypes are two of nine ADRs. Briefly, vascular

contraction is mediated by the alA-ADR, whereas vascular dilation is mediated by the 32-

ADR. Gene expression studies of these subtypes in animal strains (particularly between

normotensive and hypertensive strains) suggest an important role in the development of

HTN; however, studies of this nature have not been conducted with human tissues in a

between-groups design. This study explores the feasibility of conducting such a study in

humans and the relative gene expression differences of the two aforementioned ADR

genes in people with and without HTN and explores the impact of self-identified race.

Gene expression refers to transcription of ribonucleic acid (RNA) from deoxyribonucleic









acid (DNA). This process is a necessary step in the making of proteins. Gene expression

is influenced both by genetic and environmental factors during transcription.

Relative levels of RNA of the alA- and 32-ADR genes were measured in arterial

tissue samples obtained from 41 subjects who had coronary artery bypass surgery at

either Shands at Alachua General Hospital or the Malcom V. Randall Veterans Hospital.

Subjects were grouped according to the diagnosis of HTN (n = 24) or NT (n =17), as

defined by national guidelines. During surgery, a small amount (10-30 mg) of normally-

discarded internal mammary artery tissue was provided to the researcher, processed, and

analyzed with Real-Time, reverse-transcription polymerase chain reaction to obtain

relative quantitation of gene expression.

Hypertensive subjects showed 3.92- and 2.05-fold differences in relative alA- and

P2-ADR gene expression (respectively) compared to normotensives (statistically

significant with alpha of 0.05), with hypertensives demonstrating reduced expression of

both genes. Fold differences for both ADR subtypes remained significant when

comparing White/Caucasian hypertensive versus normotensive subjects. Further

exploratory aims produced some significant findings. This study experienced

methodological issues with the reference gene, thereby affecting accuracy of relative

gene expression quantitation; therefore, interpretation of results is cautioned.
















CHAPTER 1
INTRODUCTION

This chapter will introduce the main research problem and background and

delineate the hypotheses to be tested. The definitions of variables, major terms,

assumptions, and significance of the study will also be presented.


Figure 1-1. Abbreviations used.


Background and Problem Statement

Just over 26% of adults worldwide (approximately 972 million adults) have

hypertension (HTN). Essential HTN is synonymous with "primary" HTN, in which the

cause of the high blood pressure (BP) is unknown. This type represents 90-95% of all

cases of HTN (American Heart Association [AHA], 2005). Alternately, secondary HTN

accounts for the other 5-10% of all cases. For these cases, the cause is known and often


AA= African American
ADR = adrenergic receptor, or adrenoceptor
alA-ADR = adrenergic receptor, Alpha 1A-subtype
P2-ADR = adrenergic receptor, Beta2-subtype
BP = blood pressure
CABG = coronary artery bypass graft
cDNA = complementary DNA
CVR = cardiovascular reactivity
DBP = diastolic blood pressure
HTN= hypertension
IMA = left internal mammary artery
mRNA = messenger RNA
NT = normotension
PCR = polymerase chain reaction
RNA = ribonucleic acid
RT = reverse transcription
SBP = systolic blood pressure
SIR = self-identified race
T2DM = type 2 diabetes
TCV = Thoracic and Cardiovascular Surgery
TPR = total peripheral resistance
VAMC = Veterans Administration Medical Center









correctable. Renal disease (of various types) is the most common cause of secondary

HTN. Secondary HTN can also arise from single-gene disorders such as glucocorticoid

remediable aldosteronism, the most common autosomal dominant form of inherited HTN.

Secondary causes can also be conditions (such as pregnancy or stress) that, when

corrected, bring BP back to normal levels.

Heritability (h2) is the ratio of additive genetic variance to total phenotypic

variance. It can be thought of as the amount of variation in high BP attributable to the

variation in our genetic makeup. Pedigree, twin and sibling studies have discerned that

the heritable portion of essential HTN is approximately 30% (Ambler & Brown, 1999).

More recent data suggest that BP traits such as SBP and DBP have high estimated

heritability at about 72% and 63%, respectively (Zeegers, Rijsdijk, Sham, Fagard, Gielen,

de Leeuw, et al., 2004). This could mean that the estimated heritability of HTN is

actually higher than we previously thought; or that this particular study had inflated

values due to study design and/or analyses. Furthermore, the recurrence risk of HTN

increases as the number of parents with HTN increases, so that an offspring has

approximately 4% chance of developing HTN with no hypertensive parents, a 10-20%

risk with one hypertensive parent and the risk increases to 25-45% when both parents are

hypertensive (Lucassen, 1999). These findings point to an obvious link to genetics in

explaining some of the variance in HBP.

HTN is known as the "silent killer," as few, if any, symptoms are noted by its

sufferers. It is widely accepted that essential HTN is multifactorial, developing as a result

of multiple genes and multiple environmental factors, their interactions producing altered

homeostasis of BP regulation in the body. Furthermore, HTN has a complex









pathophysiology involving the cardiovascular, renal, endocrine, neurohumoral, and

immune systems. Within these systems are subsystems that contribute to the grand

schema of developing HTN, each mechanism having a number of genetic components.

These include the renin-angiotensin-aldosterone (RAA) system, the angiotensin

converting enzyme (ACE), the sodium balance either by the kidneys or by hormonal

influences, or the vascular system, to name a few. Nearly every mechanism has its own

candidate gene(s) for HTN. Some examples are those coding for the following proteins:

renin, angiotensinogen, angiotensin I and II, angiotensin-converting enzyme, atrial

natriuretic peptide (B and C types), nitric oxide induciblee and endothelial), endothelins,

dopamine, kallikrein, adducing a-subunit, and adrenergic receptors (ADRs).

The ADRs are particularly important in regulating BP. They are the main binding

sites for the catecholamines epinephrine and norepinephrine, which work in delicate

balance to regulate vasodilation and vasoconstriction. These vasomotor reactions can

influence the rising and falling of BP levels in the body. The ADRs can also mediate BP

regulation through renal sodium excretion and release of renin from the juxtaglomerular

cells in the kidneys (DiBona, 1989). Decades of research have shown that ADRs are

important in the regulation of BP in humans and animals, and that alterations in ADRs at

the cellular and genetic levels may lead to HTN. Various functional differences in ADRs

have been reported between normotensive and hypertensive humans and animals. Of

novel interest are recent animal studies of gene expression differences in two specific

ADRs in HTN: the alA- and p2-subtypes. Both subtypes are involved in vasomotor tone

via expression in the arteries and veins and both are implicated in the grand schema of

HTN. Each has a gene that codes for its receptor protein. How these genes are expressed









in the tissues can lead us to important information about their role in HTN. Gene

expression of these ADR subtypes can be measured by their messenger ribonucleic acid

(mRNA) levels found in the tissues where they are present. The mRNA levels provide us

with direct information about the level of transcription of the genes that code for these

subtypes. Current technology allows us to preserve tissue samples in such a way that we

can accurately measure this mRNA (or level of transcription) and compare these levels

between groups of people (for example, between people with normotension (NT) versus

HTN). (More detailed background information on these subtypes and gene expression is

provided in Chapter 2: Review of the Literature.)

The previous paragraph explained that differences in ADR exist between

hypertensive and normotensive humans and animal strains; that two specific subtypes

have been implicated in the pathophysiology of HTN; and that examining gene

expression of these genes may provide a novel insight into one aspect of the disease

process. ADR differences in HTN have also been reported among self-identified races

and ethnicities. While it is accepted that there is great interindividual variability among

people with regard to ADR function, expression, physiologic response and

pharmacologic response (Small, McGraw, & Liggett, 2003), potential racial and ethnic

differences attract attention because of the disproportionate statistics regarding

hypertensive disease in racial and ethnic subpopulations. Disease prevalence,

management, morbidity and mortality among the black or African American (AA)

population are particularly problematic because AAs exhibit the highest rate of HTN and

the worst health outcomes in regards to morbidity and mortality in the U.S. The possible

explanations of race-specific differences in health and disease outcomes are at the center









of great debate. Possible variables include socioeconomic factors as well as differences in

pathophysiologic mechanisms, pharmacologic responses, and recently genetic variability.

In the ADR literature, there are reports of adrenergic-specific differences in

cardiovascular reactivity (CVR) within black/AA populations (McAdoo, Weinberger,

Miller, Fineberg, & Grim, 1990; Stein, Lang, Singh, He, & Wood, 2000; Knox,

Hausdorff, & Markovitz, 2002) and adrenergic-specific diversity in medication response

in AAs (Humphreys & Delvin, 1968; Jennings & Parsons, 1976; Seedat, 1980; Cushman,

Reda, Perry, Williams, Abdellatif, & Materson, 2000).

The estimated heritability for HTN in people of sub-Saharan African descent is 45-

68% (Rotimi, Cooper, Cao, Ogunbiyi, Ladipo, et al., 1999; Gu, Borecki, Gagnon,

Bouchard, Leon, Skinner, et al., 1998). While this heritability estimate is specific to

people having origins of the sub-Saharan region of Africa, which is not generalizable to

any- or everyone having origins in Africa, it warrants further investigation into genetic

sub-population differences. Genetic studies of association have used self-identified race

(SIR) as a variable. Reports of racial differences in allele frequency of ADR

polymorphisms exist (Hindorff, Heckbert, Psaty, Lumley, Siscovick, Herrington, et al.,

2005; Xie, Kim, Stein, Gainer, Brown, & Wood, 1999). If all of these aforementioned

ADR differences (cellular, functional, pathophysiologic, and pharmacologic) truly exist

between HTN and NT people and among SIRs and ethnicities of hypertensives, can

genetics explain these differences? Relatedly, do differences exist regarding how these

ADR genes are expressed in the vascular tissue? These are some exploratory issues that

will be addressed by this study.









Purpose of the Study

The purpose of this study is to examine the relationships among HTN and gene

expression of the alA- and 32-adrenergic receptors (ADRs) in the human population, and

to explore if SIR helps to explain some of the differences. The study will address three

specific aims:

Specific aim 1: To quantify differences in gene expression of alA-ADR and 32-

ADR in the internal mammary artery (IMA) between subjects with normotension (NT)

and HTN.

* To quantify relative differences in alA-ADR gene expression between study groups
with NT and HTN.

* To quantify relative differences in 32-ADR gene expression between study groups
with NT and HTN.

Specific aim 2: To explore relative differences in gene expression of alA-ADR and

P2-ADR in the IMA between subjects with NT and HTN by SIR.

* To explore relative differences in alA- and 32-ADR gene expression between
White/Caucasian subjects with NT and HTN.

* To explore relative differences in the alA- and 32-ADR gene expression between
White/Caucasians with HTN versus Black/AAs with HTN.

Specific aim 3: To explore the relationship between level of alA- and 32-ADR gene

expression and need for post-operative positive inotropic medication administration.

Exploratory aim 1 (El): To explore the association between diagnosis of HTN

and Xal- and 32-ADR genotypes. Three genotypes will be explored: alA-ADR (Codon

347, refSNP ID:1048101), 32-ADR (Codon 16, refSNP ID:1042713 and Codon 27,

refSNP ID: 1042714) for their association with HTN.

* El-1: To explore the impact of SIR on genotype by diagnosis association.









Exploratory aim 2 (E2): To explore the association between genotype and gene

expression.

* E2-1. To explore the association between alA-ADR (Codon 347) single nucleotide
polymorphism (SNP) and alA-ADR gene expression.

* E2-2. To explore the association between 32-ADR (Codon 16) SNP and 32-ADR
gene expression.

* E2-3. To explore the association between 32-ADR (Codon 27) SNP and 32-ADR
gene expression.

Hypotheses

For all specific aims, the null hypothesis that no statistically significant differences

exist between groups will be tested. In keeping with the neurohumoral model of HTN, if

alA-ADRs contribute to vasonconstriction and 32-ADRs contribute to vasodilation, it

could be hypothesized that subjects with HTN would display greater levels of alA-ADRs

and lower levels of p2-ADR gene expression than NT subjects; however, as the reverse

phenomenon can also lead to HTN via regulatory feedback loops (Anderson, McNeilly,

& Myers, 1992), a unidirectional hypothesis is not appropriate. Similarly, racial

differences in cardiovascular reactivity patterns could possibly support directional

hypotheses of AAs showing greater alA-ADR gene expression than Caucasians, and

Caucasians showing greater 32-ADR gene expression than AAs; however, there is not

enough evidence in the literature to support a unidirectional hypothesis at this time. For

El, HTN is expected to be positively associated with each of the three ADR SNPs, based

both on previous findings and the compelling evidence supporting the role of these SNPs

in the disease process of HTN (Small, McGraw, Liggett, 2003). For E2 (E2-1 through

E2-3), it is hypothesized that gene expression may be affected by the variants in SNPs in









the gene, so that variations in genotypes may affect gene expression; however, the

direction of this relationship is not established enough for unidirectional hypotheses.

Definitions of Terms

Terms discussed in this study are defined as below.

Hypertension Defined by the Seventh Report of the Joint National Committee on

Prevention, Detection, Evaluation, and Treatment of High Blood Pressure (JNC VII) as

having three consecutive BP readings of 140/90 mmHg or greater, the diagnosis of high

BP by a health care practitioner, or taking antihypertensive medications specifically for

BP control (JNC VII, 2003).

Normotension Having BP below 130/90 mmHg, having never been diagnosed

with high BP, and not taking antihypertensive medications specifically for BP control,

also defined by the JNC VII (2003).

Adrenergic receptors- A group of nine G-protein receptors from the super family

of cell-surface receptors that signal the sympathetic nervous system in response to the

need for BP homeostasis. clA-ADRs contribute to vasoconstriction. p2-ADRs contribute

to vasodilatation.

Cardiovascular reactivity- A complex cardiovascular trait in which individual

cardiac and vascular responses to physiological and psychological stressors may lead to

changes in systolic blood pressure (SBP), diastolic blood pressure (DBP), total peripheral

resistance (TPR), and other hemodynamic measurements that represent vascular response

to maintain cardiovascular homeostasis.

Gene expression- Process by which genes are "expressed" in the body. DNA is a

double-stranded sequence of nucleotides that codes for proteins. DNA strands are

transcribed (or copied), making single-strand messenger RNA (mRNA). The









transcription process begins at the promoter region of the gene. The mRNA template

produced by transcription is then translated into proteins. The levels of mRNA found in a

biological specimen are indicators of the level of transcription. How the genes are

expressed (or how much transcription is taking place) can be informative of how much

the gene is functioning or how much is being copied to produce specific proteins.

Race Used in this study as a self-identification of one of 5 categories that reflects

their geographic origin based on the corresponding regions and populations listed in the

groups. Generally, cultural aspects of affiliations with these groups of origin is implied.

These categories are set by the U.S. Office of Management and Budget (OMB) and are

meant to reflect population-specific self-identification, not skin color. (See Appendix A.)

This variable is referred to as "self-identified race," or SIR for short. See sections titled

"Recommendations for the collection of racial and ethnic data" and "Implications for

studying disease by race" in Chapter 2.

Assumptions

This study focuses only on essential HTN. The first major assumption is that

essential HTN is a multifactorial disease process with multiple genetic and environmental

factors that are likely to interact. While many models of HTN exist, this study focuses on

the neurohumoral model of HTN. The second major assumption is that the neurohumoral

model plays a major role in the pathophysiology of HTN. This model concentrates on the

importance of activation of the sympathetic nervous system and neurohumoral substances

(namely epinephrine and norepinephrine). When released, these endogenous

catecholamines interact with a- and P-ADRs to elicit a cascade of cellular membrane and

intracellular events (Berecek & Carey, 2003) that affect the cardiovascular system. A

third assumption is that mRNA levels are indicative of receptor regulation and a fourth









assumption is that receptor regulation impacts disease mechanisms at the receptor and/or

cellular level. It is already known that the "transcription rate and steady-state level of 0-

ADR messenger RNA" is modified when the 0 receptors are stimulated (2003, p. 3). This

is yet to be confirmed in regards to a-ADRs, but induction of transcription is likely to

play a similar role in regulating these receptors. Finally, a fifth assumption is that RT-

PCR only quantitates steady state mRNA levels, and therefore only a "snapshot in time"

(Bustin, 2002). Furthermore, these levels may not reflect levels of protein produced by

the cell (Gygi, Rochon, Franza, & Aebersold, 1999).

Significance of the Study

HTN is one of the most prevalent chronic diseases in the United States (U. S.). The

AHA reports there are an estimated 65,000,000 Americans over age 6 and 1 in 3 adults

that have HTN (2005). Although HTN is easily detected and usually controllable, the

cause of 90-95% of cases is unknown (AHA, 2005). Economic costs of hypertension in

the U.S. are estimated to total $59.7 billion in 2005 (AHA, 2005). Vascular-related

comorbidites of HTN include diabetes, peripheral vascular disease, and stroke. In short,

HTN and its vascular consequences have major impacts on our society's health and

economy. HTN's complex pathophysiology leads to a complex phenotype with many

clinical variations. Its silent nature and disease complexity often result in poor rates of

diagnosis and control. This is evident by a control rate of merely 34% in all known

hypertensives (JNC VII, 2003). Overall, the complexity of the disease process makes

HTN difficult to manage and study. Much promise has been placed in the study of

genetics, particularly in regards to popular gene association studies, where associating the

frequency of a particular allele and/or haplotype (combination of alleles) with a disease is

the focus. As summarized by Small and colleagues (2003), some allele-based association









studies report positive associations between ADR alleles and HTN, while others report no

significance of these alleles. Specifically, the alA- ADR has been hypothesized to play a

role in HTN due to its role in vasomotion; however, only one study has associated a

polymorphism of the alA- ADR gene to HTN (Jiang, Mao, Zhang, Hong, Tang, and Li, et

al., 2005). Polymorphisms of the 32-ADR have been positively associated with HTN

(Timmermann, Mo, Luft, Gerdts, and Busjahn, et al.,1998.) Gene association studies are

numerous, but many are inconclusive, inconsistent, and poorly powered. Animal and

human studies focusing on physiologic, pharmacologic, genomic, and genetic factors

have shown promise in providing evidence for a- and 3-ADR mechanisms in mediating

CVR in HTN, as will be delineated in Chapter 2: Review of Literature. Still, little

attention has been paid to the role of gene expression of a- and P-ADRs in vascular tissue

in HTN and CVR in human studies. Gene expression studies have been conducted to

identify the role of a- and P-ADRs, but predominantly in non-human models (Gaballa,

Peppel, Lefkowitz, Aguirre, Dober, and Pennock, et al., 1998) or specifically to focus on

the effect of medication (Wang & Brown, 2001; Nishio & Watanabe, 1999) or aging

(Miller, Hu, Okazaki, Fujinaga, & Hoffman, 1996) on gene expression of these ADRs.

Few studies have been found in the literature regarding gene expression analysis of

ADRs in humans. Only one study demonstrated the differences in gene expression of

these ADRs between persons with and without HTN. This group of researchers examined

the presence of three al-ADRs in peripheral blood lymphocytes of human NT and

hypertensive subjects (Veglio, Tayebati, Schiavone, Ricci, Mulatero, and Bronzetti, et al.,

2001). They studied gene expression of the ADR genes located in the blood. They also

compared the human blood sample gene expression data to that of NT and hypertensive









strains of rats, finding similar al-ADR densities in human blood and animal tissues. Also,

significant differences in expression of certain al-ADR subtypes were found both

between humans with HTN and their normotensive controls, as well as between the

normotensive and hypertensive strains of rats (Veglio et al., 2001). This study provided

important information about the use of peripheral blood lymphocytes in the analysis of

gene expression of al-ADR subtypes, as well as relation of human to animal models.

Their findings supported the link between al-ADR subtypes in HTN at the gene

expression level (Veglio et al., 2001). Some limitations of this study include isolation of

the alA- subtype in rat vas deferens tissue and not arterial, venous, aortic, or myocardial

tissue. Finally, measuring mRNA levels via peripheral lymphocytes is an indirect

measure of transcription because the measurement is not occurring in the tissue of

interest (or the tissue thought to be directly involved in the disease pathway). This is less

reliable than direct methods, where mRNAs are examined in the tissue. Veglio and

colleagues (2001) reported choosing this method because human tissue was not possible to

obtain. Theoretically, there are concerns with using blood to examine gene expression.

One major issue is the source of mRNAs in circulation; the origin of the mRNAs that are

found in the bloodstream is lymphocytes. It is not clear if mRNAs expressed in the blood

have differential expression than those expressed directly in the tissues. While Veglio and

colleagues (2001) were able to show similarities in expression between vas deferens

tissue and blood lymphocytes, further investigation is needed to compare blood

lymphocytes with other tissues. Many consider it less reliable to use circulating blood

mRNAs to examine the direct relationship between transcriptional processes and disease









mechanisms because other factors circulating in the blood could potentially vary the

expression at any given time and thus make the research less replicable.

Other human gene expression studies involving ADR subtypes in human tissues

have focused on other disease processes such as congestive heart failure, and have

obtained samples from human myocardial tissue, most commonly obtained from

endomyocardial-biopsy specimens (Lowes, Gilbert, Abraham, Minobe, Larrabee, and

Ferguson, et al., 2001; Moniotte, Vaerman, Kockx, Larrouy, Langin, and Noihomme, et

al., 2001). Gene expression studies examining the differences in mRNA level present in

the actual tissue could help to explain if the actual expression of the proteins that make up

the ADRs play a role in HTN, rather than merely the presence of a particular allele.

Researching the mechanisms that account for these differences has the potential to

increase our understanding of the impact of gene expression on phenotypic variance in

HTN and adrenergically-driven vascular tone.

The first aim in this study investigates whether or not gene expression of alA- and

32-ADRs are related to the diagnosis of HTN by examining their expression in human

arterial tissue between subjects with and without a diagnosis of HTN. This method

provides direct measures of steady-state levels of mRNA and insight into the genetic

picture of real-time transcription activity in each subject's individual environment. To my

knowledge, an investigation of this nature has not been previously reported in human

arterial tissue, comparing hypertensive and normotensive subjects. This study provides

preliminary results about the role of alA- and 32-ADR gene transcription levels and their

relationship to HTN, and could lead to larger-scale studies in the future. On a more









innovative note, this study could further support gene therapy involving the ADR system

in the management of HTN.

The second aim of the study explored if differences in alA- and 32-ADR gene

expression levels exist when SIR was taken into account. This aim provides information

about whether or not similar ADR gene expression trends exist in self-identified racial

groups. This study may contribute to the overall goal of reducing health disparities

related to SIR and HTN that may be due to genetic variation, as identified by Healthy

People 2010 (National Heart, Lung, and Blood Institute [NHLBI], 2003). The health

disparities among hypertensive black and AA groups are real. What is uncertain is

whether or not a genetic basis exists to help explain those disparities. This study provides

preliminary data to begin to answer that question.

The third specific aim of the study is to explore the relationship between alA- and

32-ADR gene expression levels and the need for post-operative positive inotrope

medication administration. The need for positive inotrope administration in the post-

operative stage of recovery from coronary artery bypass surgery (CABG) is most often

the result of a negative cardiac event (such as acute congestive heart failure, cardiac

arrest, hypovolemia, or arrhythmias) that necessitates increased vascular resistance to

correct the problem. Since G1A- and p2-ADRs are involved in vascular resistance, and

inotropes increase vascular resistance, this aim provides indirect information about

subjects' cardiovascular reactivity. In addition, it provides information about whether or

not gene expression of the alA- and 32-ADRs is related to poorer cardiovascular

outcomes, as measured by their need for positive inotropes, in subjects undergoing

CABG.









The global health and economic consequences of this disease are incredible. HTN

is a disease that affects every continent and population, some more disproportionately

than others. Any novel information regarding the impact of G1A- and 32-ADR gene

expression on people with HTN could prove to be a valuable building block for future

studies. This study bridges the gap between bench and the bedside using advanced

genetic technology to examine if and how gene expression of alA- and 32-ADRs relate to

HTN. In addition, the study seeks to explore the impact of SIR on gene expression of the

two chosen ADR genes, and the impact of gene expression on the need for emergency

cardiac medication in the post-operative stage of recovery.














CHAPTER 2
REVIEW OF LITERATURE

Introduction

This chapter will present a review of the literature regarding all relevant topics of

the study, including: genetics, hypertension, cardiovascular reactivity, adrenergic

receptors, the alA- and 32-ADR subtypes, pharmacotherapeutic aspects of the alA- and 12-

ADRs, population differences in HTN development and management, recommendations

for the collection of racial and ethnic data, implications for studying disease by SIR,

positive inotrope administration, and genetic analysis techniques.

Genetic Influences on Essential Hypertension

Early heritability studies carried out with essential hypertensive twins estimated at

least 63% of the variability in BP was due to genetic factors and reported "little evidence

for environmental influence on the familial aggregation of BP" (Grim et al., 1984, p.

453). The most cited estimate of essential HTN heritability is approximately 30%

(Ambler & Brown, 1999). Some recent researchers examined the heritability of BP traits.

In a classical twin study of 173 dizygotic (DZ) and 251 monozygotic (MZ) twin pairs

aged 18-34 years, randomly selected from the East Flanders Prospective Twin Survey,

heritability estimates were: SBP 74% (95% CI: 0.68-0.79) and DBP 63% (95% CI: 0.55-

0.59). These heritability estimates were not confounded by the following potential risk

factors: body mass index (BMI), cholesterol ratio, birthweight, physical activity, sex, and

cigarette smoking (Zeegers, Rijsdijk, Sham, Fagard, Gielen, and de Leeuw, et al., 2004).









Genetic Models of Hypertension

HTN has two distinct genetic classes: monogenic (meaning caused by one gene)

and polygenic (caused by multiple genes). A number of monogenic forms of HTN have

been identified, such as glucocorticoid remediable aldosteronism, Liddle's syndrome,

Gordon's syndrome, and Bilginturan syndrome (all autosomal dominant gene

abnormalities), and apparent mineralocorticoid excess, caused by an autosomal recessive

gene abnormality (Beevers, Lip & O'Brien, 2001). These are rare in the general

population. Essential HTN is considered polygenic. This polygenicc model" of disease

stems from R. A. Fischer's "quantitative genetics" theory proposed in 1918. He

postulated that a phenotype (observable expression of a genotype as a trait or disease)

"was influenced by a large number of genes, each behaving according to basic Mendelian

rules, but each having only a small individual effect on the phenotype" (McClearn,

Vogler, & Plomin, 1996, p. 96). Not long after, scientists realized that the environment

could influence a phenotype, and the debate began concerning 'nature versus nurture'.

This led to introduction of the gene-environment interaction model.

The earliest research designs that supported these models of familial aggregation

were pedigree studies involving twins, siblings, and families. MZ twins share as much as

99% of genetic information; DZ twins and non-twin siblings share as much as 50% of the

same genetic information. Large-scale pedigree studies in which families are observed

longitudinally created the basis for the common linkage analysis that now uses pedigrees

and genetic testing to link specific genes to phenotypes of disease. From these studies, a

great number of candidate genes were discovered for high BP. Association studies take

this concept one step further. These studies correlate candidate genes to hypertensive









phenotypes in the general population, looking for associations between alleles (or

variants of the gene) and HTN.

Another current model of HTN is the animal model. Animal models ultimately

serve as models for human disease. Animal researchers use sophisticated breeding

methods and more recently, gene knock-out and knock-in techniques to manipulate and

control genetic, environmental and even phenotype variables. Using this basic research

model, researchers can investigate numerous theories in HTN that could not otherwise be

studied in humans.

Cardiovascular Reactivity, HTN and Adrenergic Receptors

As previously defined, CVR is a complex cardiovascular trait in which individual

cardiac and vascular responses to physiological and psychological stressors may lead to

changes in SBP, DBP, TPR, and other hemodynamic measurements that represent

vascular response to maintain homeostasis. These changes differ between NTs and

hypertensives, in both human (de Visser, van Hooft, van Doornen, Hofman, Orlebeke, &

Grobbee, 1995) and non-human animal models (McDougall, Paull, Widdop, &

Lawrence, 2000). Furthermore, these responses have been categorized as being

predominantly a- and P-adrenergic in nature (Linden, Gerin, & Davidson, 2003). a- and P

-ADRs are members of the super family of cell surface receptors that carry out signaling

via coupling to guanine nucleotide binding proteins (G-proteins) (Small, McGraw, &

Liggett 2003). They are critical components in the sympathetic nervous system's response

to disease and maintenance of homeostasis, as they are the target receptors for

epinephrine and norepinephrine (Small et al., 2003). Theoretically, alterations in

peripheral vascular mechanisms are the proposed basis for the a- and P-ADR sensitivity-

modulated CVR, in which exaggerated responses to a stressor produce differing CVR









(Lovallo & Gerin, 2003). These exaggerated responses are proposed to be a result of

preclinical alterations in vascular resistance that can cause a disproportionate rise in BP

relative to an otherwise normal demand for blood flow (Lovallo & Gerin, 2003).

Fredrikson, Tuomisto, and Sundin (1990) report vascular dysregulation to both

conditioned and unconditioned vasoconstriction in their comparison study of vascular

response to classical conditioning in mild hypertensives versus NT. Miller and Ditto

(1991) report patterns of increasing vascular resistance in response to an active-coping

psychological stressor, which were purported to be due to a -adrenergic activity, and not

neurohumorally independent autoregulation.

Alpha 1A- and Beta 2-Adrenergic Receptors

There are nine subtypes in the family of human ADRs; the alA- and p2-subtypes are

specifically located in the vasculature (Small, McGraw, & Liggett, 2003). The alA-ADR

gene is located on chromosome 8 at location 8p21 (OMIM # 104221, 2002). The 32-ADR

gene is located on chromosome 5 at location 5q32-34 (OMIM # 109690, 2003). Figure 2-

1 shows the approximate locations of each ADR gene. Vascular contraction is controlled

primarily by al-ADRs, and their importance in BP regulation is emphasized by the

efficacy of al-AR antagonists in human HTN (Rokosh & Simspon, 2002; Guthrie &

Siegel, 1999; ALLHAT Collaborative Research Group; 2000). The alA-ADR receptor

gene product is required to maintain arterial BP, as evidenced by a recent mouse gene

knockout study (Rokosh & Simspon, 2002). Leech & Faber (1996) reported that

constriction of rat skeletal muscle arterioles is mediated predominantly by an alD-ADR.

However, Reja, Goodchild, and Pilowsky (2002) reported G1A-receptor messenger

ribonucleic acid (mRNA) expression was significantly greater in spontaneously

hypertensive (SHT) rat tissue samples compared with NT rats, and was positively













correlated with SBP in all central tissue investigated. The alA-ADR mRNA expression


level appears to be an important determinant of SPB, and is one of the genetic markers


examined in this study.


Known Genes % GC SNPs Chromosome 5
Genes Repeats
Known Genes X GC SNPs Chromosome 8 p1i.33
Genes Repeats 1p31
p15.2
p23.2 p15.1
p23.1 p14.
14.1
pp13.3


SP12 1
'11.2
pil.21 1 2.1

411.23 413.3
ar. 199) alt 1n c ( &
112 1 14.1
q1313 a I

g21.11 1 l n 21.








A2D 1g sub.nes. A Ch 8 wit bckt i 3diate ap oi lca
q2- q23. 1
q22 1 a 23 .2
22.223
n. 23.3
122.3,
a1.3
q23.1
qa2








Figure 2-1. Chromosome (Ensembl human map view) showing the locations of both

ADR genes. A) Chromosome 8, with bracket indicating approximate location

of the alA -ADR gene. B) Chromosome 5, with bracket indicating

approximate location of the 32-ADR gene.


During activity or stress, 3-AR signaling is responsible for regulating changes in


heart rate, BP, and contractility (Reja, Goodchild, & Pilowsky, 2002). After selective 3-


ADR receptor activation, both 031- and 32-ADR elicited dilation of large coronary arteries


(Young, Vatner, & Vatner, 1990). Monopoli and colleagues (Monopoli, Conti, Forlani, &


Ongini, 1993) reported that human coronary artery contains equimolar amounts of 3l-


and 132-receptor subtypes and that 32-ADR specifically mediates vasodilation in vascular


smooth muscle. Likewise, another study reported predominantly 132-mediated relaxation


in human IMA exposed to both epinephrine and norepinephrine in vitro (Ferro,


Kaumann, & Brown, 1993). Polymorphisms of the 32-adrenoroceptor gene have been









associated with: 1) Interindividual variability in resting SBP and DBP in response to

mental challenge (McCaffery, Pogue-Geile, Ferrell, Petro, & Manuck, 2002); 2) The

level of resting and stress-related BP (Li, Faulhaber, Rosenthal, Schuster, Jordan,

Timmermann, and Hoehe, et al., 2001); and 3) Vascular reactivity as indicated by lower

basal blood flow and attenuated increases in forearm blood flow in hypertensive adults

(Cockroft, Gazis, Cross, Wheatley, Dewar, and Hall, et al., 2000). A polymorphism of the

32- ADR gene, the Gln27Glu (glutamine, codon 27, glutamate) which causes a point

mutation of cytosine (C) -to-guanine (G), was examined by Bray and colleagues (2000).

They reported an occurrence of HTN with the Glu27 allele was 1.8 times higher than

with one or two copies of the Gln27 allele (95% confidence interval, 1.08 to 3.00, p =

0.023). Chruscinski and colleagues (2001) demonstrated a positive role for 32-ADR in

mediating vascular dilation when BP response to was blunted in a mouse gene knockout

model. Knowing that the vascular system is rich with 32- ADRs and that they mediate

vasodilatation, laccarino and colleagues (2002) chose to overexpress p2-ADRs via

adenoviral-mediated gene transfer in normotensive Wystar-Kyoto and spontaneously

hypertensive rats. They reported successful gene transfer of the 32- ADR gene and

enhanced vasorelaxation in the carotid arteries of hypertensive strain of rats versus the

NT strain (n = 8 to 10 per group) after 32- ADR overexpression (F = 3.088, P < 0.05). 02-

ADR appears to be an important determinant of BP, and was the second genetic marker

examined in the study.

Pharmacotherapeutic Aspects of Alpha-lA and Beta-2 Adrenergic Receptors

The primary indication for both a- and P-blocking drugs is HTN. In the case of P-

blockers, cardioselective types are preferred to reduce side effects caused by blockade of

multiple ADR subtypes. Selective peripheral al blockers such as prazosin and terazosin









induce vasodilation by blocking the al receptors in vascular smooth muscle arterioless

and veins). Their selectivity to al causes less reflex tachycardia than drugs that inhibit a2

(Kalkanis, Sloane, Strichartz & Lilly, 1998).

More commonly used in the treatment of HTN are 0 blockers, like metoprolol,

propanolol, and atenolol. The JNC VII (2003) supports their use in various populations

and they have been shown to reduce morbidity and mortality in randomized controlled

trials. Cardioselective P-blockers principally block 31 receptors and partially block 32

receptors. This reduces side effects of blocking all p2 receptors in the lungs and blood

vessels. Although P ADRs, when stimulated cause vasodilation, P-blockers also reduce

renin release from the juxtaglomerular cells of the kidney, thus reducing the renin

angiotensin system's effect on increasing BP. In addition, P-blockers interfere with

sympathetic vasoconstrictor nerve activity and block the effects of catecholamine surges

(Khan, 1999). p blockers also reduce heart rate, plasma norepinephrine, muscle

sympathetic nerve traffic and systemic norepinephrine spillover, all indices of adrenergic

activity in essential HTN (Grassi, 2004).

Population Differences in HTN

The prevalence of HTN in AAs in the U. S. is among the highest of all groups. AAs

tend to have worse clinical sequelae than their White, non-Hispanic counterparts (Cooper

& Rotimi, 1997). Americans who self-identify with African descent have a 1.5-2 fold

increase in prevalence of HTN compared to Americans who self-report descent from

Europe; comparing women in these two groups leads to the highest prevalence

differences (Eberhardt et al., 2001). Disease management is often particularly

problematic, with AAs often requiring a more than one (and often multiple) anti-

hypertensive medications to effectively manage their high BPs, many of whom cannot









afford them. HTN is a particular problem in health care in the Southeastern U. S., as the

prevalence of HTN among blacks and whites is greater, and death rates from stroke are

higher in this region than others (AHA, 2003).

In regards to developing HTN, examples of racial differences in physiologic

mechanisms and pharmacologic responses involving ADRs exist in the literature.

Differences in CVR are noted in relation to BP and heart rate, and include a- and P-

adrenergic patterns that are associated with racial cohorts--particularly AAs versus

Caucasians. Anderson, McNeilly, & Meyers (1992) explain two dichotomous CVR

patterns that are reported to be associated with SIR: the myocardial and the vascular.

These patterns are summarized in Table 2-1.

Table 2-1. Adrenergic cardiovascular stress patterns.
Myocardial Reactivity Pattern Vascular Reactivity Pattern
(P-adrenergically driven) (a-adrenergically driven)
Sin BP associated i iilh. Pin BP associated i iilh.
1 Cardiac Output (C.O) 1 Norepinephrine
1 Stroke Volume Total Peripheral Resistance
1 Heart Rate

1 Epinephrine and Norepinephrine
[ in Total Peripheral Resistance
Characteristic of Caucasian reactivity Characteristic of Black/AA reactivity
pattern pattern

There are reported differences in drug disposition and responsiveness in relation to

adrenergic-agonists and antagonists (Wood & Zhou, 1991), implying ADR differences

across SIRs. Some support the above model, and some do not. Sentinel research shows

that P-blockade and combined a- and P-blockade via pharmacological agents appear to be

less efficacious in AAs, South Africans, Jamaicans, and West Indians as compared to









Caucasians (Humphreys & Delvin, 1968; Jennings & Parsons, 1976; Seedat, 1980).

Another seminal study in 1977 by the Hypertension Detection and Follow-up Program

Cooperative Group reported P-blockers to be less efficacious in African Americans.

Gibbs, Beevers, and Lip (1999) purported that this may be due to decreased cardiac

output and renin release, causing increased total peripheral resistance. In support, Wood

(2002) reported a marked impairment of 32AR-mediated vasodilation in blacks,

accompanied by increased a-adrenergically mediated vasoconstriction, as well as racial

differences in response to endogenous and exogenous agonists. Also, vasoconstrictor

response to endogenously stimulated norepinephrine is higher in blacks than whites

(Stein, Lang, Singh, He & Wood, 2000). Male, black Veterans residing inside the 'Stroke

Belt' (southeast U.S.) are reported to have lower treatment success rates with captopril (p

= 0.07); and, regardless of region, blacks in the study were less likely than whites to

achieve successful lowering of their BP with atenolol (p = 0.02), prazosin (p = 0.03), and

more likely with diltiazem (p = 0.05) (Cushman et al., 2000). Jamerson and DeQuattro

(1996) disagree, explaining that while observed responses of blacks to both ACE

inhibitors and P-blockers in the treatment of HTN are less favorable than is seen in

whites, the responses are still clinically significant. Literature on nonpharmacologic

differences by SIR is also present. Ferro and Walton (2001) report significant differences

in short-term BP responses to a 10-week regimen of nonpharmacological treatments

(dietary, activity, stress reduction, and education sessions) in HTN for African/Caribbean

blacks compared to whites. Blacks and the control group experienced no change in either

systolic or diastolic BP at 10 weeks, and statistically significant decline in systolic (p <

0.005) and diastolic BP (p < 0.05) were seen in the Caucasian group. In a 2005 meta-









analysis of 137 monotherapy clinical trials and 28 combination therapy trials (totaling

11,739 participants), Wu and colleagues reported that AAs had better BP reduction with

calcium channel blockers than their non-AA counterparts (p = 0.001); and, that non-AAs

responded better than AAs to al-blockers, 31-blockers, and angiotensin converting

enzyme inhibitors (p = 0.0001) (Wu, Kraja, Oberman, Lewis, Curtis, Ellison, and Arnett;

2005).

The estimated heritability for HTN in people of sub-Saharan African descent is 45-

68% (Rotimi, Cooper, Cao, Ogunbiyi, and Ladipo, et al., 1999; Gu, Borecki, Gagnon,

Bouchard, Leon, and Skinner, et al., 1998). Genetic studies of association have reported

racial differences in allele frequency of ADR polymorphisms (Hindorff, Heckbert, Psaty,

Lumley, Siscovick, and Herrington, et al., 2005; Small, McGraw, & Liggett, 2003; Xie,

Kim, Stein, Gainer, Brown, & Wood, 1999).

This review of the literature highlights the many inconsistencies across studies in

regards to the influence of race on health and disease. A number of things can explain

these inconsistencies: study design issues; differing measurement/report of race;

inadequate power; and different medications studied. Another plausible explanation is

that the differences are really attributable to non race-based, interindividual variability.

Recommendations for the Collection of Racial and Ethnic Data

Much like its original use for classifications of groups, some mainstream

definitions of race today infer major biological underpinnings. The online Merriam-

Webster dictionary (2003) defines race as "a division of mankind possessing traits that

are transmissible by descent and sufficient to characterize it as a distinct human type".

This definition inherently includes genetics as a factor in race by use of the phrase

'transmissible by descent'. Numerous similarities in other definitions exist, involving









such phrases as: 'physically distinguishable,' 'having common ancestries,' and 'having

certain biological characteristics that set them apart from other groups.' It is easy to see

why and how the paradigm of linking genetics to race exists even now, as these are

definitions from no longer than a century ago. Very recently, there has been an upsurge

of efforts to discard notions of race as biologically-associated. Some researchers and

'civilians' wish the term to be recognized solely as an antiquated system of skin-color-

based classification that inherently carries with it socio-politically charged notions of

racism. Others argue for less biologically-based definitions of race that incorporate social

beliefs about language, history, and culture (such as Witzig, 1996). Definitions like these

seem very similar to those of ethnicity, where language, history, culture, and socio-

political factors are main constructs of this term. To further complicate the matter, some

utilize the term ethnicity with the presumption that they are avoiding any biological

undertones inherent in the term 'race'; however, definitions of race and ethnicity are

markedly similar and most people use the two terms interchangeably (Sankar & Cho,

2002). In a meta-analyses of articles published in Nursing Research, authors Drevdahl,

Taylor, and Phillips (2001) present this case well, comparing three operational definitions

of race and three of ethnicity used (some cited from other sources) by nursing researchers

within the 1990s and 2000. Race and/or racial group was defined as: "Imply[ing]

biological characteristics.. .that are genetically transmitted from one generation to

another" (Schubert & Lionberger, 1999, p. 116); "Concept that signifies and symbolizes

sociopolitical conflicts and interests in reference to different types of human bodies"

(Winant, 2000, p. 172); and, "Group that is socially defined as having certain biological

characteristics that set them apart from other groups, often in invidious ways (Pincus &









Ehrlich, 1999, p12). The three ethnicity and/or ethnic group definitions were: "Contains

information about the history of the population, and hence the genetics of the group, as

well as sociocultural information" (Crews & Bindon, 1991, p. 45); "Group that has

certain cultural characteristics that set them off from other groups and whose members

see themselves as having a common past" (Pincus & Ehrlich, 1999); and, Segment of a

larger society whose members are thought...to have a common origin and to share

important segments of a common culture and who... participate in shared activities in

which the common origin and culture are significant ingredients" (Yinger, 1994, p. 3).

While debate continues over the construct of race and ethnicity, it is generally agreed

upon that: Whatever definition is used, it should be clearly delineated (Sankar & Cho,

2002); Racial classifications should be critically evaluated for their usefulness and

contribution to the testable theories (Duster, 2001); and, The methods of capturing race

and/or ethnicity should be carefully outlined (Williams, & Jackson, 2000).

Standardization of definitions is a major issue and is recommended for consistency

in reporting results in research (U.S. Department of Health and Human Services

[USDHHS], 2003). The OMB Race and Ethnicity Classification system can be utilized

for standardization. It includes both race and ethnicity categories, and defines race by

context of geographical origin. Cultural associations are accounted for in the Ethnicity

self-report section. Users have the option of selecting more than one self-affiliated

category of race; or, the option to not answer the question at all. Based on this

knowledge, this study will utilize the OMB Classfication system. This self-report system

allows the user to choose more than one race, if applicable. It utilizes standardized









constructs of race and ethnicity that are rooted in ancestral and geographic origins of their

predecessors and is widely used in research.

Implications for Studying Disease by Race

An abundance of health disparities literature exists on the social, ethical, and legal

ramifications of studying disease by SIR. Historically speaking, some experiments that

target racial groups have resulted in serious social and ethical problems for that

population (for example, the Tuskegee syphilis experiments). Current debate in regards to

identifying genetic differences by race has identified many concerns, including the

potential to send the message that researchers are trying to find clear biological

differences that would imply certain races are 'unequal' to others. People fear that finding

biological differences will justify certain social inequalities. The key opposition to

studying diseases by race are the following points: 1) There may be many inherent

problems in measuring and grouping races in a multi-racial society for the purpose of

genetic clustering (Wilson, Weale, Smith, Gratrix, Fletcher, and Thomas, et al., 2001;

Williams & Jackson, 2000); 2) The sociopolitical context of race is an important variable

that is often disregarded in research (Burchard, Ziv, Coyle, Gomez, Tang, and Karter, et

al., 2003) ; 3) Fear of justifying inequality (Bonham, 2003); and 4) Racial boundaries are

not likely to be equally useful in all kinds of genetic research (Sankar & Cho, 2002).

Others argue for research that examines race carefully, making sure to address the above

concerns. Duster (2001) purports that race should not be discarded as a variable in

research just because the categories do not biologically map exactly, and that race

remains alive in the context of practical application. Fullilove (1994) warns against a

priori consideration of race as important in medicine without question, but also states that









little is truly done to explain the meanings in associations between health outcomes and

race.

In certain disorders and diseases, race is very much a risk-associated variable.

Sickle-cell anemia (SSA) maintains a significantly higher prevalence in people of African

and Mediterranean descent. Likewise, Cystic Fibrosis (CF) is more likely to affect those

from Western European descent. That is not to say that whites have not been diagnosed

with SSA, or that cases of CF have not been seen in blacks or other non-whites. It is only

to say that there is a higher risk associated with particular diseases among certain

populations. Population genetics has ascertained that greater genetic differentiation

occurs between continentally separated groups (Burchard et al., 2003), and that more

variation is present within racially-stratified populations than between them. Nonetheless,

others have reported great genetic variation among the five racial groups (five groups as

categorized by the OMB classification system) (Risch, Burchard, Ziv, & Tang, 2002).

Genetically speaking, some racial groups possess low frequencies of certain alleles

associated with disease, (2003). Whether or not the low frequency of alleles is truly

associated with race or simply a result of sampling methods or statistical errors in the

research is currently under debate. While it is more easily seen in "simple" diseases like

SSA and CF, the genetic influence of race is much more difficult to ascertain in complex

diseases like type 2 diabetes (T2DM), asthma, HTN, and Alzheimer's disease.

Nonetheless, specific susceptibility gene variants for chronic diseases have been found in

specific populations. Phimister (2003) explains that a variant of the calpain-10 gene

(associated with T2DM) is specific to a population of Mexican Americans in Texas.

Moreover, a variation of the ETS family of genes that predispose carriers to asthma has









been discovered in a population inhabiting the island of Tristan da Cunha of the South

Atlantic (2003). In recent editorial, Fine, Ibrahim & Thomas (2005) cite three studies of

complex genetic disorders (Crohn's disease and Factor V Leiden) where genetic variation

by race was reported (Ridker, Miletich, Hennekens, & Buring 1997; Shen, Lin & Tsay,

1997; Hugot, Chamaillard, Zouali, Lesage, Cezard, and Belaiche, et al., 2001).

For some, results of these studies are not convincing enough to conclude that race

is an appropriate variable for use in research. Cooper, Kaufman, and Ward (2003) argue

against studying disease by race, first asserting that race-specific findings in research are

better explained by environmentally-determined socioeconomic factors. They (2003) also

report examples of inconsistencies in research in which Type I error affected the

outcomes and interpretations, such as with the reported race-specific effect of ACE

inhibitors in the Antihypertensive and Lipid-Lowering Treatment to Prevent Heart Attack

Trial. Cooper and colleagues also argue that true race-specific genetic results have not

been found, and are "mathematically and biologically implausible" (p. 1167, 2003). In

2001, Wilson and colleagues performed a unique study designed specifically to test the

validity of race and ethnicity as genetic research variables. They studied eight

populations of varying origin, some of which were extremely specific (for example,

Amharic- and Oromo-speaking Ethiopians from Shewa and Wollo provinces collected in

Addis Ababa). Using a model-based clustering method known as STRUCTURE, they

were able to estimate the proportion of each individual's genome having ancestry in each

cluster. One underpinning in this model is the idea that admixture plays an important role

in the variability of race. Genetic admixture reflects multiplicity or variation in race and

ethnicity (Burchard et al., 2003). Wilson and his co-authors ascertained the









apportionment of individuals (average per-individual proportion of ancestry) by using the

STRUCTURE model to characterize 'clusters' based on a set of allele frequencies at each

locus. From there, the researchers matched the clusters with specific geographical areas

(in this case, four broad regions), and interpret the similarities seen in each cluster.

Because 62% of the Ethiopians fell into the cluster with the Jews, Norwegians, and

Armenians, they concluded that identifying these people as "black" in race would be an

"inaccurate reflection of the genetic structure" (2003, p. 266). They concluded similar

results with subjects from China and New Guinea in which "Asian" race grouping would

have been inappropriate (2003). While almost truly convincing, this study requires

replication, and use of "presumably neutral microsatellite markers" (p. 265) needs to be

validated as truly neutral. Based on these and other data, many researchers have moved

toward the use of ancestry, rather than race or ethnicity in their research models. In fact,

self-reported ethnicity and ancestry constructs have been related in biological models.

Helgadottir and colleagues (2006) found that self-reported ethnicity was highly correlated

with genetic determination of estimated individual ancestry (via ancestry informative

markers) and even group ancestry (determined by weighted least squares). They also

reported ethnicity-based differences in risk for myocardial infarction in African

Americans who had European admixture. This implies that self-reported ethnicity can be

informative and useful in genetic research as a means to group individuals for

comparison.

The PI agrees that numerous other factors may impact the poor health outcomes

seen in the AA population, including various socioeconomic indicators and factors

related to health care access and delivery. It is important to reiterate that the literature









presented clearly identifies specific pathophysiologic and biologic differences among

racial groups involving HTN that direct this line of research. In further support of this

venture, at least one other (Krieger, 2005) suggests that in the case of gene expression,

observed phenotype differences seen among populations could possibly reflect variation

in gene expression (more so than simple gene frequency) because of the nature of gene

expression patterns occurring in the context of certain (perhaps shared) environmental

conditions. Moreover, one charge of the Task Force of the American Sociological

Association was to comment on the further investigation of race and ethnicity in the

contribution of disparate outcomes (2003). They presented the example of AA health

disparities in regards to HTN and affirm that research needs to continue in this line of

research to distinguish between social and biological forces at play (2003). Properly

designed research in this area could provide answers regarding molecular differences and

poor health outcomes, a better understanding of the disease process in certain

populations, and tailoring of medications or gene therapy. Furthermore, proper statistical

analyses and careful interpretation of findings can strengthen results and help to reduce

potential 'racial profiling.'

The ethical principle of social justice, in its simplest form, states that all people

should be afforded equal benefits, (such as goods and services) regardless of their

personal characteristics, choices, or beliefs. Current research that examines race as a

variable has also been criticized for limiting social justice. For example, research that

excludes subjects based on their race (a practice that is declining) prevents certain groups

from reaping the benefits of a study. Similarly, certain socioeconomically limited groups

tend not to benefit from research because of the inherent cost associated with new









technology that comes from it. As certain races have been associated with lower

socioeconomic status (SES), social justice is compromised for these groups. It is from

research such as this that we learn of the social and ethical ramifications that can be

incurred if research is not properly conducted involving race and disease. In the research

community we must be most careful in our interpretation of data involving race,

ethnicity, ancestry, and/or socioeconomic status so that misleading conclusions are not

made that adversely affect social justice, policy, and practice. We should be, at minimum,

cognizant of how the design, implementation, analyses, and interpretation of results

involving health disparities can affect the overall well-being of socioeconomically

disparate groups.

Positive Inotrope Administration

Short-term positive inotropes are often administered post-operatively to CABG

patients to increase vascular resistance in cases of acute congestive heart failure, cardiac

arrest, hypovolemia, and arrhythmias. The three subclasses of positive inotropes are:

cardiac glycosides (digitalis, digoxin); P-adrenergic agonists (dopamine, dobutamine,

epinephrine); and phosphodiesterase inhibitors (milrinone, amrinone, enoximone). The

need for positive inotropes in the post-operative phase of recovery from bypass surgery

can be informative as an indirect assessment of the subjects' cardiovascular reactivity. As

ADRs are involved in vascular tone, knowledge about whether or not a subject needed

this medication could help to explain some of the differences in gene expression of the

alA- and P2-ADR genes.

Genetic Analysis Techniques

There are many types of genetic methodologies used in research today. Linkage

studies identify regions of the genome that contain putative candidate genes that are









proposed to be related to a phenotype or disease process based on their location on

chromosomes. Association studies investigate the prevalence of certain gene alleles and

their association with a phenotype, or disease process. Both of these methods examine

deoxyribonucleic acid (DNA). Unlike these, gene expression studies examine the

relationship between the level of ribonucleic acid (RNA) and disease. While DNA is a

double-stranded nucleic acid made up of nucleotides, RNA is single-stranded, and is the

result of transcription of genetic information; the information encoded in DNA is

transcribed into mRNA, which is an intermediate and one of the regulatory steps in the

synthesis of new proteins. Many cellular characteristics concerning survival, growth and

differentiation are reflected in altered patterns of gene expression and the ability to

quantify transcription levels of specific genes is central to research into gene function

(Zamorano, Mahesh, & Brann, 1996).

Common gene expression methods (quantification of steady-state transcription) are

northern blotting, in-situ hybridization, RNAse protection assays, cDNA arrays, and RT-

PCR (Bustin, 2000). RT-PCR is a type of PCR that allows one to compare the levels of a

specific mRNA in different sample populations, to characterize patterns of mRNA

expression, to discriminate between closely related mRNAs, and to analyze RNA

structure (Bustin, 2000). This method involves isolating mRNA from the biological

sample (in this case, IMA) and performing reverse-transcription (RT) to make

complimentary DNA (cDNA) The cDNA represents only the mRNA component of the

total RNA, which can then be analyzed by gene expression equipment. TaqMan is a type

of RT-quantitative PCR method that continuously measures (in real time) accumulated

PCR product. The PCR product reflects the original level of mRNA template (See Figure









2-2). This is measured using a TaqMan probe. The probe is a dual-labeled fluorogenic

oligonucleotide. The dual-labels are a reporter dye and a quenching dye. The ABI Prism

(Applied Biosystems HT 7900) equipment and software examines the fluorescence

intensity of the reporter and quencher dyes and calculates the increase in normalized

reporter emission intensity over the course of the PCR amplification (Genomics and

Proteomics Core Laboratories, 2003) within all samples located on the 96-well plate. The

normalized reporter is known as a housekeeping gene. This is a type of gene in which

there is a known and predicted level of expression. Its purpose in gene expression

analysis is to act as an internal reference or control, whereby all samples are normalized

by this same gene. (Practically speaking, since the level is known, the housekeeping gene

values are subtracted from the sample values to obtain the normalized levels.) This allows

relative expression to be established, instead of absolute quantification of the data, which

is tedious and impractical (Bustin, 2000). A spectrum of standard housekeeping genes are

available. The protocol for this study will use glyceraldehydes-3-phosphate

dehydrogenase (GAPDH), a housekeeping gene that has been previously used with

successful results in normalizing data for ADR gene expression in arterial tissues (Wang

& Brown, 2001; Reja, Goodchild & Pilowski, 2002; Peuster, Fink, Reckers, Beerbaum &

von Schnakenburg, 2004).

Northern blotting, in-situ hybridization, RNAse protection assays, and cDNA

arrays are all alternate methods of measuring gene expression. Northern blotting isolates

RNA by elecrophoretic separation on an agarose gel, 'blotting' or transferring RNA

fragments from the gel onto a membrane (usually nitrocellulose), adding a labeded probe

to the membrane and detecting the band with the probe bound to the target sequence









(CRISP Thesaurus, accessed 8/2/05). In situ-hybridization determines the presence of an

RNA sequence of interest by hybridizing a probe to the target sequence, and visualizing

on a microscope the location of the bound RNA target in the chromosome or cell

(cytoplasm, for example) (Human Genome Project, Talking Glossary, accessed 8/2/05).

RNAse protection assays examine gene expression by hybridizing antisense RNA

corresponding to known genes with an unknown sample. Next, the sample is digested

with a single-strand specific RNAse and any surviving RNA left is presumed to be

complimentary to the antisense and therefore, transcribed from the gene of interest

(CRISP Thesaurus, accessed 8/2/05). Finally, cDNA arrays (more commonly known as

microarrays) utilize a microarray 'chip' or platform with many small spots that

correspond to a different gene on each spot. The spots are pre-treated with cDNA that is

the only coding part of the sequence of interest that corresponds to an mRNA transcript.

These cDNA spots are hybridized with a probe. The chips are incubated in solution

containing the genetic material being investigated. Messenger RNA transcripts floating in

the solution hybridize to their cDNA already on the chip. When the chip is exposed to

ultraviolet light, the fluorescent probes emit light at varying intensities, allowing

qualitative comparison of expression between the different genes on each chip and

between subjects (Rice University Connections webpage, accessed 8/2/05). These

techniques can be limited in their sensitivity (Melton, Kreig, Rebagliati, Maniatis, Zinn &

Green, 1984) and in their cost-effectiveness (Bustin, 2000). RT-PCR is the optimal

method when evaluating a limited number of genes and starting mRNA template is low.

There are additional advantages in using TaqMan Real Time RT-PCR gene expression

analysis. Unlike other forms of quantitative RT-PCR, this method quantitates the initial











amount of the mRNA template (the geometric phase), rather than the final amplified


product (the plateau phase), allowing detection of a 2-fold versus a 10-fold change. This


improves the sensitivity, specificity, and reproducibility of the method (Dorak, 2003;


Dawson, 2003). Real Time RT-PCR also involves only three major steps, whereas other


conventional RT-PCR methods involve nine steps. Reducing the number of steps in the


gene expression process assists in minimizing error in sample analysis. Also, once the


ABI Prism equipment has completed its fluorescence phases, the data are fed into a


computer linked to the equipment, eliminating the need for post-PCR processing (Bustin,


2000). The only clear disadvantage of this method is the cost of the predeveloped


reagents and the ABI Prism, but it is still more cost-efficient than cDNA array


(microarray) methods when evaluating a small number of genes. The TaqMan RT-PCR


principle steps are represented in Figure 2-3.









ARn I
I -- --- -- --- --- -- -- -







*Geometric Exponential Plateau
Cycle
A Rn= change in fluorescence label detected by TaqMan equipment
Cycle = number of amplifications over time
*Geometric phase = "exponential" phase indicating that the exact doubling of product is
accumulating at every cycle; this phase is very specific and is the phase that threshold cycle is
determined for relative gene expression analysis
Exponential phase = phase in which reaction products are consumed and the reaction slows and
degrades; highly variable
Plateau phase = signifies complete halting of the reaction


Figure 2-2. Amplification of gene expression using TaqMan Real Time RT PCR.
-* "' "



!-
!-


*Gemeri Exoeta lta
Cycl
/ =cag nfursec abldtce yTqa qimn
Cyl f u/ e faloiiain vrtft














Figure 2-2. Amplification of gene expression using TaqMan Real Time RT PCR.












Sample Trascripti


ABI PRISM 7900 Thermal cycler/detector Quantification via
ABI software


Figure 2-3. TaqMan RT-PCR steps (adapted from Bustin, 2000).

Collection of Human Tissues for Gene Expression Analyses

Tissue samples and/or biopsies are the source of choice for analysis of gene

expression/transcription, simply because it is the direct site at which to examine the

mRNA levels mediating protein production in the body. It is widely accepted that

mRNAs found in circulating blood lymphocytes provide indirect evidence of this process.

During CABG, it is common for portions of surgical remnants of IMA to be discarded.

The IMA branches off of the left subclavian artery and supplies the thoracic cavity with

oxygenated blood. It is most typically used for bypass of the left anterior descending

coronary artery. As previously described, X1A- and 02-ADRs are specifically located in

the vasculature (Small, McGraw, & Liggett, 2003). AlphalA mRNA has been detected in

the IMA (Gow, Mitchell, & Wait, 2003). 02-AR has been detected in the human coronary

arteries (Monopoli, Conti, Forlani & Ongini, 1993) and in the human IMA (Ferro,

Kaumann, & Brown, 1993). Very minute amounts of arterial tissue (about 30 mg) are

necessary for this type of analysis, which involves isolating messenger RNA from the

tissue sample and performing reverse-transcription to make cDNA, which can then be

used for TaqMan (Real Time) gene expression analysis.









Summary

HTN has a genetic component, as supported by empirical and experimental data.

The alA-ADR and 32-ADR subtypes of the ADR receptor genes are hypothesized to play

a role in mediating the disease process of HTN via gene expression differences. Race

and/or ethnicity may also contribute to the variance seen in HTN and adrenergically-

driven vascular tone, as supported by previous studies described. While it is controversial

to use race/ethnicity as variables in genetic research, the use of a standardized measure

and careful interpretation can be informative and help reduce the potential for social

injustice.

The use of Real-Time PCR for analysis of gene expression is an ideal method for

examining steady-state transcription levels and comparing relative fold-differences

between groups. This analysis can provide direct information about the function of these

genes in the given environment at that particular moment. Tissue samples are the desired

source for examining mRNA levels; however, human studies of the alA-ADR and 12-

ADR genes expression in human tissue are not currently reported. This study attempts to

fill the gap in this knowledge by using human arterial tissue for analyses of the alA-ADR

and 32-ADR genes expression levels between hypertensive and normotensive adults.














CHAPTER 3
PROCEDURES AND METHODS

Introduction

This chapter presents study design and protocols. Details regarding recruitment

techniques, research settings, variables, and methods are discussed. The following

methods are thoroughly explicated: collection and storage of samples, processing of

tissue and blood for isolation of genomic DNA and RNA, and post-isolation processing

of RNA for gene expression analyses.

Design

This study used an exploratory, quantitative design to meet the goals of the

aforementioned specific aims. This was a two-arm gene expression study to compare NT

versus hypertensive persons primarily, then to explore differences between two self-

reported race categories: Black/AAs and White/Caucasians. Recruitment goals were set

at a total of 60 subjects with 15 subjects in each arm. This sample size was based on a

formulation of 82% power, an effect size of 0.25 (medium), and a significance of 0.05 for

a two-tailed test. Gpower computer software was used to calculate the required sample

size (Erdfelder, Faul, & Buchner, 1996). A medium effect size was consistent with other

gene expression studies of adrenergic mechanisms in HTN (Wang & Brown, 2001).

Following consent, pencil and paper data collection was used to obtain some basic

demographic and medical history information. During surgery, a small amount of IMA,

(normally discarded during CABG surgery) was collected and later analyzed for alA- and

32-ADR gene expression via TaqMan (Real-Time) RT- PCR. Additionally, about 10 cc









of blood was collected intra-operatively for genotyping. A post-operative chart review

was completed to determine the subjects' need for positive inotrope pharmacotherapy

while in intensive care.

Subject Recruitment

Prior to initiation of the study protocol, Institutional Review Board (IRB) approval

was obtained from the UF IRB-01 and the VA Subcommittee for Clinical Investigations.

Adult subjects between the ages of 21 and 70 who were scheduled for CABG surgery

were recruited from those admitted to the University of Florida Thoracic and

Cardiovascular Surgery (TCV) team. These subjects included patients from the following

facilities: Shands at Alachua General Hospital and the Malcolm Randall Veterans

Administration Medical Center (VAMC), both located in Gainesville, Florida. These

facilities mainly serve Alachua County, but often include patients from the entire North

Central Florida and surrounding areas.

HIPAA Waivers of Authorization were obtained from the UF and VAMC IRBs.

This IRB-approved waiver allowed the applicant to review charts of the scheduled CABG

patients to determine which patients met the inclusion/exclusion criteria of the study.

Inclusion and exclusion criteria with rationale are presented in Table 3-1. Prior to

beginning the study, all surgeons agreed to have their patients screened for this study.

These individuals were contacted by telephone or via face-to-face meeting during pre-

operative appointments for recruitment. This recruitment process was continued until the

planned group allotments were filled. Subjects were considered hypertensive if their

medical chart indicated: a) diagnosis of HTN by a practitioner, b) three consecutive office

BPs > 140/90 mmHg, or c) prescription of antihypertensive medications specifically for

high BP. Subjects were also asked to verify that they were diagnosed with high BP.









Subjects were not excluded on the basis of race, religion, ethnicity, socioeconomic status,

or level of education. Women and minorities were recruited for this study. The AHA

(2003) reports that more men than women have high BP until age 55. From age 55 and

older, the percentage of women with high BP continues to increase (2003). Moreover,

HTN is primarily an underlying cause of death for more women than men (2003). As

indicated in the TCV surgery database, the average percent of women undergoing CABG

surgeries by the TCV surgery department is 21%. This means that for every one female

undergoing the procedure by this department, there are approximately 5 males. U.S.

Census data for Alachua County indicates that approximately 73.5% of the population is

Caucasian, 19.3% is AA, and 7.2% is 'other'. The JNC VII (2003) reports that health

disparities exist in minority populations with HTN. Inclusion of women and minorities

will create a study population that is representative of the entire population of those

undergoing CABG procedure. Children were not included in this study. While 50 million

Americans age 6 and older have high BP (AHA, 2003), it is rare for children to have

CABG surgeries. While it is important to study childhood HTN and its long-term

consequences, it is not feasible to recruit such a minimally represented population

(children with HTN who undergo CABG).

Table 3-1. Inclusion and exclusion criteria with rationale.
Inclusion Rationale
age 21-70 -children excluded: rare CABG
-There is evidence that genetics still plays a role in HTN even
into late 70's, without confounding of isolated systolic
hypertension (ISH) past the age of 70; however, ISH results in
a clinically and pathophysiologically different phenotype from
essential HTN (Sleight, 2004). This difference in phenotype
could lead to a difference in gene expression that would
confound the data.









Table 3-1 Continued.
Inclusion Rationale
undergoing -prime surgery to obtain arterial tissue that is often discarded
scheduled CABG (Wang & Brown, 2001)
surgery -improper data collection could occur with unscheduled cases
read/write -unable to provide interpreter for multiple languages
English
Exclusion Rationale
undergoing heart -the IMA is most frequent artery used in bypass surgery,
surgery that does usually grafted to the left anterior descending artery
not include the
internal
mammary artery
(IMA)
low cardiac -subjects with this post-operative hemodynamic diagnosis may
output syndrome exhibit increased total peripheral resistance secondary to the
(LCOS)* diagnosis and confound the inotrope-related data

Consenting Process and HIPAA Regulations

The PI obtained approval from the UF and VAMC IRBs for human subjects'

research. Subjects signed an informed consent to participate in the study, including

informed consent for chart review and research use of normally discarded surgical

remnants of IMA (see Appendices B & C). Subjects were not compensated for their

participation in the study. Subjects were informed they could withdraw from the study at

any time. There was no anticipated direct benefit to the subjects: they did not receive any

information concerning their hemodynamic responses to positive inotrope administration,

nor did they receive results of their gene expression of alA-and p2-ADRs. This eliminated

the need for another informed consent that was designed specifically for disclosure of

genetic information, and also eliminated the need for genetic counseling related to testing

and/or results. It is uncertain at this time what the expression of these genes in human

tissue actually means in terms of health benefits; therefore, lay interpretation would be

difficult at this time.









Data collection containers were labeled with a subject ID barcode sticker. A hand-

written table containing the coding system for the subjects was kept in a locked filing

cabinet with only the PI having access. This was the only source of data that matched the

subject to the ID code.

Setting

This study was completed in Gainesville, Florida at the Gainesville VAMC and

Shands at Alachua General Hospital facilities. Screening of patient charts for eligibility

occurred at the VAMC and the TCV surgery office in the UF Health Science Center.

Subjects were approached for recruitment and consented at the VAMC, AGH, or at

Cardiology Associates of Gainesville, all places where subjects were either undergoing

pre-operative assessments or were admitted. Data collection occurred in the VAMC and

AGH 'heart' surgical suites. Blood and tissue samples were stored at the UF Center for

Pharmacogenomics. Laboratory analysis occurred at the UF Pharmacogenomics Core

facility and the UF Interdisciplinary Center for Biotechnology Research (ICBR). Subject

folders with consents and hard-data were stored in locked filing cabinets in the PI's

student office space at the UF College of Nursing.

Research Variables

The independent variables were diagnosis of HTN/NT and self-identified OMB

racial category. Both were categorical, nominal variables. SIR (self-identified race) was

determined by self-report of one or more of the five categories as defined by the OMB

(see Appendix A). Diagnosis was a dichotomous variable with either HTN or NT as

variable choices. The dependent variables were gene expression of alA- and 32-ADR

(continuous), genotype (categorical/nominal), and post-operative positive inotrope

administration dichotomouss, nominal). Gene expression was determined using the ABI










Prism 7900 and ABI Assays on Demand for GaA-ADR (Hs00169124_ml) and 02-ADR

(Hs00240532_sl) (Applied Biosystems, Foster City, CA).

Relative gene expression of alA- and 32-ADRs were utilized to determine the

values of the gene expression variables, as outlined by Livak and Schmittgen (2001).

Three ADR single nucleotide polymorphisms (SNPs) were examined. The alA-ADR

(Codon 347, refSNP ID:1048101) SNP is located on chromosome 8 at location 8p21, in

the second exon, or coding region. The 02-ADR (Codon 16, refSNP ID:1042713 &

Codon 27, refSNP ID: 1042714) SNPs are located on chromosome 5q32-34, both in the

first (and only) exon of the gene. Figures 3.1 and 3.2 show the loci of investigated

polymorphisms with relation to the alA- and 32-ADR genes. Genotypes for the alA-ADR

(Codon 347), 02-ADR (Codon 16 & Codon 27) were determined by PCR followed by

pyrosequencing (PSQHS96A System, Biotage, Uppsala, Sweden) on genomic DNA

isolated from blood samples (PSQaHS 96 System, Uppsala, Sweden). Chart review was

conducted to determine subjects' need for post-operative positive inotrope medication.


5' 3' Key
.............. Promoter region
...... .... ,',',ft ,','" .\ w I Exon (coding) regions
SIntron (noncoding)
regions
Sal -ADR, Codon 347
polymorphism

Figure 3-1. alA -ADR gene with promoter, intron and exon boundaries and investigated
polymorphism.










Key
5' 3' .............. Promoter region

............... Exon (coding) region
Suntranslated regions

1 32-ADR Codons 16 &
27 polymorphism


Figure 3-2. 32-ADR gene with promoter, exon boundary and investigated
polymorphisms.

Study Protocol

Data Collection and Laboratory Methods

Subjects completed a pencil and paper demographic form that provided information

about their race, age, past medical history, medication use, height, weight, and income.

As recommended by the Federal Drug Administration ([FDA], 2003), race was

determined by self-report using the OMB revised race and ethnicity categories.

Blood and tissue collection

All blood and tissue samples were collected in the surgical suites. After the patient

was anesthetized for surgery, approximately 5-10 cc of arterial blood obtained from the

central arterial line was placed in a purple-top tube containing EDTA and placed in a

cooler with ice. The surgical remnants of IMA pedicle were cleaned by either the PI or

the surgeon after removal from the patient and then placed in a sterile specimen container

by the PI. The PI quickly (in a sterile field), cut the tissue to pieces smaller than 0.5 cm,

and transferred the pieces immediately to a microtubule containing 100 microliters of

RNAlater solution (Qiagen Valencia, CA, USA) (see Figure 3-3). RNAlater is a

nontoxic, aqueous tissue and cell storage reagent that protects cellular RNA in intact and

unfrozen samples. It stabilizes RNA and preserves its integrity by halting mRNA

degradation upon its infusion into the sample. At this point, according to Qiagen's









instructions, the RNA is protected from degradation for 24 hours at 370 C, one month at

4 C, and indefinitely at -200 C. Completing these steps in a very quick manner minimizes

RNA degradation and any changes in the mRNA expression level; thus, the PI worked

very quickly to complete this process, which often took less than 1 minute to complete.

The samples were then placed in a cooler with ice and transported to the UF Center for

Pharmacogenomics Core Laboratory, where they were incubated at 2-80C at least

overnight, but no more than 12 days, then placed in a freezer at -80o C.


















Figure 3-3. Tissue pieces immersed in RNAlater preservation solution.

Genomic DNA analyses

Genomic DNA was isolated from blood lymphocytes using a commercially

available kit (Qiagen DNA Blood Isolation Kit (Qiagen, Valencia, CA, USA). Genotype

was determined by polymerase chain reaction (PCR), followed by pyrosequencing

(Pyrosequencing, Uppsala, Sweden) (Langaee & Ronaghi, 2005) using a PSQ HS96A

single nucleotide polymorphism (SNP) reagent kit according to the manufacturer's

protocol (Biotage AG, Upspsala, Sweden). In summary, 10 pl ofbiotinylated PCR

product was immobilized to streptavidin-coated Sepharose beads (Amersham









Biosciences, Piscataway, NJ). After incubation, the beads were isolated and treated with

70% ethanol, denaturation buffer, and wash buffer. The beads then were released into

designated wells containing annealing buffer and 10 pmol of sequencing primer, followed

by a 2-minute incubation at 800C (Langaee & Ronaghi, 2005).

The alA-ADR (Arg347Cys) polymerase chain reaction (PCR) amplification was

determined by using the primers listed in Table 3-2.The PCR mixture consisted of 6.25 ul

HotStarTaq Master Mix (Qiagen GmbH, Hilden, Germany), 0.75 pl of

dimethylsulfoxide (Sigma-Aldrich, St. Louis, MO), 10 pmol of each primer (Operon

Biotechnologies, Huntsville, AL), 1.5 [tl of water, and 50-100 ng of genomic DNA. The

PCR amplification was performed under the following conditions: initial denaturation at

95C for 15 minutes, 45 cycles of denaturation at 950C for 30 seconds, annealing at 56C

for 30 seconds, and extension at 720C for 1 minute, followed by a final extension step at

72C for 7 minutes. The 32-ADR (Argl6Gly and Glu27Gln) polymerase chain reaction

(PCR) amplifications were determined by using the primers listed in Table 4. Note that

the same forward and reverse biotinylated primers were used; however, two different

forward sequencing primers were used. The PCR mixture consisted of 12.5 ul

HotStarTaq Master Mix (Qiagen GmbH, Hilden, Germany), 1.5 pl of dimethylsulfoxide

(Sigma-Aldrich, St. Louis, MO), 10 pmol of each primer (Operon Biotechnologies,

Huntsville, AL), 7 ptl of water, and 50-100 ng of genomic DNA. The PCR amplification

was performed under the following conditions: initial denaturation at 950C for 15

minutes, 40 cycles of denaturation at 950C for 30 seconds, annealing at 630C for 30

seconds, and extension at 720C for 1 minute, followed by a final extension step at 72C

for 7 minutes.









Table 3-2. Genotyping primers.
Gene Primers Amplicon
length
alA-ADR Forward: CCCCATCATATACCCATGCT 109
(Codon Biotinilated Reverse: GTAGCCCAGGGCATGTTTG
347) Forward Sequencing Primer:
TGTCTTGAGAATCCAGTGT
Sequence to analyze: CTCT/CGCAGAAAGCAGTCT
02-ADR Forward: CGAGTCCCCACCACACCC 297
(Codon Biotinilated Reverse 5':
16) AGCACATTGCCAAACACGATG
Forward Sequencing Primer:
CGGACCACGACGTCAC
Sequence to analyze: G/AGAAGCCATGCG
02-ADR Forward 3': CGAGTCCCCACCACACCC 297
(Codon Biotinilated Reverse 5':
27) AGCACATTGCCAAACACGATG
Forward Sequencing Primer: TGGCTGGCACCCAAT
Sequence to analyze: GCAGC/GAAAGGGACGA

RNA isolation and reverse-transcription

Once all tissue samples were collected, tissue processing for RNA extraction

began. To avoid any degradation of RNA by RNAse, all surfaces and tools were

thoroughly cleaned with either RNAZap (Ambion, Inc., Austin, TX, USA) or RNase

AWAY (Molecular BioProducts, Inc., SanDiego, CA) and rinsed with diethyl-

pyrocarbonate (DEPC) water. Tissues were removed from RNALater solution (Qiagen,

Valencia, CA, USA), gently blotted on kimwipes to remove excess solution (see Figure

3-4), weighed, quickly sliced into smaller pieces, then transferred to a ceramic mortar.

After the addition of a small amount of liquid nitrogen, the frozen samples were ground

into a fine powder with a ceramic pestle (see Figures 3-5 & 3-6). The powdered tissue

was then combined with 500 ul of proprietary Lysis/Binding solution from the

RNAqueous Kit (Ambion, Inc., Austin, TX). The slush was then homogenized with a

PowerGen 125 electric rotor-stator homogenizer (Fisher Scientific, Pittsburgh, PA) and

Omni-TipsTM Plastic Disposable Generator Probes (Fisher Scientific, Pittsburgh, PA).









(see Figure 3-7). After a 30-second centrifugation to remove large debris, the supernatant

was removed from the lysate and processed per the manufacturer's protocol. All samples

were eluted in 50 ul total volume of proprietary Elution Solution, included in the kit.

Total RNA was quantified by Nanodrop (Nanodrop Technologies, Wilmington, DE).

The Nanodrop determined seven samples to have concentrations less than 10 ng/ul. These

seven samples were placed in a Cenrivap Console speed-vacuum (Labconco, Kansas

City, MO) on the no-heat setting for approximately 20 minutes. These samples were then

reconstituted in 20 ul of RNAqueous Kit's Elution Solution. All 260/280 ratios were

above 1.7. UF ICBR Core staff evaluated quality of 18s and 28s peaks generated by a

2100 Bioanalyzer (Agilent Technologies, Palo Alto, CA). Quality of peaks were

consistent across all samples, indicating little RNA degradation. Next, RNA aliquots

were made to equal 10 ng/ul and brought up to 50 ul total volume with RNAse-free

water. Samples were then reverse-transcribed with the cDNA Archive Kit (Applied

Biosystems, Foster City, CA) at 250C for 10 minutes, followed by 2 hours at 370C in a

thermal cycler. All samples were stored at -200C.






51


Figure 3-4. Blotting tissue on Kimwipe.










I I


Figure 3-5. Grinding tissue in mortar and pestle on liquid nitrogen.


Figure 3-6. Powdered tissue in mortar.


























Figure 3-7. Homogenizing tissue slush with rotar-stator homogenizer.

Real-time polymerase chain reaction

Twenty microliter (ul) reactions were prepared for single-plex Real-Time PCR with

the ABI PRISM 7900 system (Applied Biosystems, Foster City, CA), located in the UF

Interdisciplinary Center for Biotechnology Research (ICBR). Primers and probes for each

of the three assays (alA-ADR, 02-ADR, and GAPDH) are listed in Table 3-3. The

housekeeping gene GAPDH was used for normalization of gene expression data, as

described in the section titled "Genetic Analysis Techniques" in Chapter 2.

For each reaction, 10.0 ul of TaqMan Universal PCR Master Mix (2X) with

AmpErase UNG was prepared with 1.0 ul of each respective TaqMan Gene Expression

Assays on Demand (20X) (Table 3-3). Eleven microliters of each master mix and 9.0 ul

of cDNA template was added to each well of the 96-well plate. Triplicate samples were

run, as recommended by Bustin (2000) to increase accuracy of the methodology. Table 3-

4 shows the plate set-up for single-plexing of these three assays. PCR conditions were

500C for 2 minutes, 95C for 10 minutes, followed by 40 cycles of 95C for 15 seconds,

and 600C for 1 minute. Fluorescence data were processed and analyzed with the ABI









PRISM Sequence Detection Software (Applied Biosystems, Foster City, CA). Results

were expressed as Ct number (number of cycles needed to generate a fluorescent signal

above a predetermined threshold) or ACt (target Ct (alA-ADR or 32-ADR) minus

normalizer Ct (GAPDH)). The Ct value was determined with the ABI 7900 software. The

software determines the baseline automatically by assessing the normalized fluorescence

signal versus cycle data, per plate. From this baseline, each sample's Ct value is obtained.

Table 3-3. Target gene assay information.
Gene Assay # Probes/Quenchers Amplicons (base pairs)
alA- Hs00169124 ml Probe: FAM 112
ADR
P2-ADR Hs00240532 sl Probe: FAM 65
GAPDH 4310884E (ABI Probe: VIC 226
product #) Quencher: TAMRA

Table 3-4. Single-plex plate set-up, one sample.
Sample 1 Sample 1 Sample 1
+ + +
GAPDH GAPDH GAPDH

Sample 1 Sample 1 Sample 1
+ + +
GaA-ADR a 1A-ADR alA-ADR

Sample 1 Sample 1 Sample 1
+ + +
32-ADR 32-ADR 32-ADR

Positive inotrope data collection

The PI examined the subjects' need for standard-of-care positive inotrope

administration in the immediate post-operative period. Chart review was conducted in the

post-operative phase for intensive care unit (ICU) documentation of administration of

positive inotropic medications. The pharmaceutical agent and dosage were documented.

ICU chart was reviewed for presence of the exclusion criteria, diagnosis of low cardiac

output syndrome. The need for positive inotrope administration dichotomizedd) was









tested for relationship with XIlA- and 32-ADRs gene expression of hypertensive versus NT

groups.

Calculations for Relative Gene Expression and Selection of Calibrator

All gene expression data were imported into EXCEL for relative gene expression

analyses. As previously stated, samples were run in triplicates for determination of the

threshold cycle (Ct) in TaqMan RT-PCR. To control for outlier Ct values, the largest Ct

value from each triplicate was removed and the duplicate values were averaged to

determine the 'average Ct'. The largest Ct values were chosen for removal because for

the majority of the samples, one of the three raw Ct values was greater than 2 Ct's away

from the next closest value, indicating, in essence, an outlier. To maintain consistency in

this process, each triplicate had the highest value removed. Next, the Ct averages were

used to calculate the following:

a) Delta Ct = average target Ct average endogenous control (GAPDH) Ct; b) Delta

Delta Ct = (Delta Ct (sample x) (Delta Ct calibratorr)); c) 2A-(Delta Delta Ct) = two to

the negative power of the DDCt; gene expression relative to the calibrator.

A requirement of the DDCt method of relative quantitation requires selection of a

calibrator. One subject (#006142) was chosen as the calibrator. This subject was a

normotensive, White/Caucasian male who was not taking any medications at the time of

the study, and reported no other cardiac diagnoses. His BMI was comparable to the mean

(30.9 versus mean of 29.0). He also reported not having ever been a smoker and did not

drink or exercise.

The determination of "fold difference" between groups is expressed as a ratio of the

measures of central tendency for the groups compared. Said another way, the fold

difference is a ratio of the one measure of central tendency to another, so that if the









median 2A-DDCt of group A was 25.0 and the median 2A-DDCt of group B was 5.0, then

the ratio is 25.0: 5.0, indicating a 5-fold difference between groups; or similarly, a 5-fold

decrease in gene expression in group A versus B. All fold-difference data were calculated

in this fashion. The relationship between the Ct value and gene expression is indirect, in

that the lower the Ct, the higher the gene expression and vice versa. This same principle

applies even after normalization, so that the 2A-DDCt value holds the same interpretation.

Briefly stated, during amplification in the Real-Time RT-PCR system, the earlier the

mRNA's amplification is detected (thus, the lower the Ct), the more abundant the mRNA.

Conversely, if it takes longer for the amplification to be detected (producing a higher Ct

value), then the mRNA is less abundant.


Methods for Statistical Analyses

Data were analyzed using Microsoft Excel (Microsoft Corporation, Redmond, WA)

and SPSS Version 14 (SPSS Inc., Chicago, IL). Descriptive statistics were used to obtain

summary measures for the data. Tests of normality for the gene expression data indicated

non-normal distributions, necessitating use of nonparametric tests. To quantify the

differences in gene expression of alA- and p2-ADRs in the IMA by diagnosis of HTN

versus NT (specific aim 1), the Mann-Whitney U test was performed. The Mann-Whitney

U test was also used to explore relative differences in gene expression of the alA- and p2-

ADRs genes and diagnosis, by SIR (specific aim 2), and to explore the relationship

between level of alA- and 02-ADR gene expression and the need for post-operative

positive inotrope administration (specific aim 3). To test the association of diagnosis

(HTN vs. NT) and the alA- Arg347Cys C--T genotype, Pearson Chi-square, and where

necessary for nonparametric data, Fisher's Exact tests were used. To test the association






56


of diagnosis (HTN vs. NT) and the 32-ADR Argl6Gly G-A, and the 32-ADR Glu27Gln

C--G genotypes, Fisher's exact test was used. The Kruskall-Wallis test was used for

testing the association between: a) alA-ADR relative gene expression and the alA-ADR

Arg347Cys C--T genotype; b) 32-ADR relative gene expression and each of the 32-ADR

Argl6Gly G--Aand Glu27Gln C--G genotypes. All hypotheses were two-tailed and

tested with alpha set at 0.05.














CHAPTER 4
RESULTS

Introduction

The primary aim of this exploratory, pilot study was to determine relative

differences in gene expression of the alA- and 32-ADR genes between people with and

without high blood pressure. The secondary aim was to determine the influence of race

on differences in gene expression. A tertiary aim was to examine the impact of gene

expression of the alA-ADR and 12 subtypes on the need for post-operative positive

inotrope pharmacotherapy. This chapter will first present descriptive results, including

means, standard deviations, and frequency data for all variables investigated. The three

hypotheses posed in Chapter 1 will be addressed using the Mann-Whitney U, Pearson

Chi-square, and Fisher's exact tests. Cohen's d effect sizes will also be provided. For

SIR, the OMB Classification tool allowed for multiple choices of race. Three subjects

self-identified as having two races: White/Caucasian and American Indian or Alaska

Native. When statistical analyses included the variable SIR, these three subjects were

recorded as "White". Explanatory aims El and E2 involving genotype by diagnoses

associations and genotype by gene expression associations are also included.

Descriptive Results

Subject Demographics

Fifty one subjects were enrolled in the study between August 2004 and July 2005.

Four subjects were excluded because blood and tissue samples were unable to be

collected due to surgery scheduling changes. As a result, 47 subjects were included in the









data analyses. This sample consisted of 37 males and 10 females with an overall mean

age of 56.5 years (range 44-70). Thirty-seven of the subjects were recruited from Shands

at Alachua General Hospital and 10 from the VAMC. Twenty subjects were

normotensive and 27 subjects were hypertensive. The normotensive group ranged in age

from 44-67 with a mean of 55.8 years. The age of the hypertensive group ranged from

44-70 with a mean of 57.3 years. These subjects are included in analyses involving

genotyping. Table 4-1 shows the demographic characteristics of this data set, expressed

in numbers and percentage. Table 4-2 presents subjects' clinical characteristics, including

height, weight, and BMI, expressed as mean and standard deviation, and prescribed P-

blocker/dose, concomitant diagnosis of diabetes mellitus-Type2 and surgery facility,

expressed as number and percentage.

Table 4-1. Demographics of all enrolled subjects.
All enrolled Normotensive Hypertensive
N= 47 n = 20 n = 27
N % n % n %
Gender
Male 37 78.7 18 90.0 19 70.4
Female 10 21.3 2 10.0 8 29.6
Race
White/Caucasian 34 72.3 17 85.0 17 63.0
Black/AA 10 21.3 2 10.0 8 29.6
White/Caucasian & 3 6.4 1 5.0 2 7.4
Native American
Ethnicity
Non-Hispanic 45 95.7 20 100 25 92.6
Hispanic 1 2.1 0 0 1 3.7
Did not know 1 2.1 0 0 1 3.7


Table 4-2. Clinical characteristics of all enrolled subjects.
All Normotensive Hypertensive
N= 47 n =20 n= 27
Height (in) 68.8 +4.1 69.8 +3.6 68.1 +4.3
Weight (lbs) 197.4 + 40.0 192.6 + 36.9 199.7 + 42.8
BMI (kg/m2) 29.2 + 5.9 27.4 + 5.3 30.3 + 6.1









Table 4-2 Continued.
All Normotensive Hypertensive
N= 47 n = 20 n = 27
P-locker (Rx & dose)
Not prescribed 16 (34.0%) 10 (52.6%) 6 (22.2%)
Metoprolol 12.5mg BID 5 (10.6%) 3(15.8%) 2 (7.4%)
Metoprolol 25mg BID 12 (25.5%) 3 (15.8%) 8 (29.6%)
Metoprolol 50mg BID 10(21.3%) 2 (10.5%) 8 (29.6%)
Metoprolol 75mg BID 1 (2.1%) 0 (0%) 1 (3.7%)
Metoprolol 100mg BID 1 (2.1%) 0 (0%) 1 (3.7%)
Labetalol 100mg TID 1 (2.1%) 0 (0%) 1 (3.7%)
Missing data 1 (2.1%) 1(5%) 0 (0)%
T2DM
No 30 (63.9%) 19(95%) 11(40.7%)
Yes 16(34.0%) 1(5%) 15 (55.6%)
Pre-DM 1 (2.1%) 0 (0%) 1 (3.7%)
Surgery facility
AGH 37(78.7%) 15 (75 %) 22(81.5%)
VA 10(21.3%) 5 (25%) 5 (18.5%)
BMI = body mass index; T2DM= Diabetes mellitus-Type 2; AGH= Alachua General
Hospital; VA = Veterans Administration Hospital, BID = twice per day

Six additional subjects were excluded from analyses involving gene expression for

the following reasons: 1) The PI could not collect tissue from one subject due to change

in surgery schedule; and 2) Tissues from five subjects did not yield sufficient RNA

material to complete the analyses or had "undetermined" readings in the TaqMan RT-

PCR gene expression output. The final sample size for gene expression analyses was 41

subjects. This subset of 32 males and 9 females had a mean age of 57.3 (range 44-70). Of

these 41 subjects, 17 were normotensive and 24 were hypertensive. The normotensive

group ranged in age from 44-67 with a mean of 56.7 years. The age of the hypertensive

group ranged from 45-70 with a mean of 57.7 years. See table 4-3 for the demographic

summary by groups, expressed in numbers and percentage. Table 4-4 presents these

subjects' clinical characteristics, including height, weight, and BMI, expressed as mean

and standard deviation, and prescribed P-blocker/dose, concomitant diagnosis of diabetes










mellitus-Type2 and surgery facility, expressed as number and percentage. Student's t-

tests confirmed that hypertensive and normotensive groups did not significantly differ in


age (t = -0.803, df= 45, p = 0.426), height (t = 1.374, df= 44, p


0.177), weight (t


0.463, df= 45, p = 0.646), orBMI (t = -1.500, df= 44, p = 0.141).


Table 4-3. Demographics for gene expression subset.
Total subset Normotensive Hypertensive
N= 41 n = 17 n = 24
N % N % N %
Gender
Male 32 78.0 15 88.2 17 70.8
Female 9 22.0 2 11.8 7 29.2
Race
White/Caucasian 29 70.7 14 82.5 15 62.5
Black/AA 9 22.0 2 11.8 7 29.2
White/Caucasian & 3 7.3 1 5.9 2 8.3
Native American
Ethnicity
Non-Hispanic 39 95.1 17 100 22 91.7
Hispanic 1 2.4 0 0 1 4.2
Did not know 1 2.4 0 0 1 4.2



Table 4-4. Clinical characteristics for gene expression subset.
Subset Normotensive Hypertensive
n =41 n = 17 n = 24
Height (in) 68.7 + 4.3 69.8 + 3.7 67.9 + 4.5
Weight (lbs) 195.8 + 41.5 193.1 + 39.1 197.7 + 43.9
Subset Normotensive Hypertensive
n =41 n = 17 n = 24
BMI (kg/m2) 29.0 + 6.0 27.5 + 5.5 30.2 + 6.2
P-blocker (Rx & dose)
Not prescribed 15 (36.6%) 9 (52.9%) 6 (25.0%)
Metoprolol 12.5mg BID 4 (9.8%) 2 (11.8%) 2 (8.3%)
Metoprolol 25mg BID 10 (24.4%) 3 (17.6%) 7 (29.2%)
Metoprolol 50mg BID 8 (19.5%) 2 (11.8%) 6 (25.0%)
Metoprolol 75mg BID 1(2.4%) 0 (0%) 1(4.2%)
Metoprolol 100mg BID 1(2.4%) 0 (0%) 1(4.2%)
Labetalol 100mg TID 1(2.4%) 0 (0%) 1(4.2%)
Missing data 1 (2.4%) 1 (5.9%) 0 (0%)
T2DM
No 27(65.9%) 17 (100%) 10(41.7%)
Yes 13 (31.7%) 0 (0%) 13 (54.2%)
Pre-DM 1 (2.4%) 0 (0%) 1 (4.2%)
Surgery facility
AGH 32 (78.0%) 12 (79.6%) 20 (83.3%)
VA 9 (22.0%0) 5 (29.4%) 4 (16.7%)


BMI = body mass index; T2DM = Diabetes mellitus-Type 2; AGH = Al
Hospital; VA = Veterans Administration Hospital, BID = twice per day


achua General







61


Assessment of GAPDH for Relative Quantitation

The duplicate GAPDH Ct values had a mean and standard deviation of 27.53 +

2.80 and median of 27.4 with values ranging from 22.1-34.2. Figure 4-1 displays the

average duplicate Ct values for each sample, showing this 12-point range. Not only

should the raw triplicate Ct values be close (no more than one-half Ct different), the

averages should show little variation across samples. Figure 4-2 shows these data again,

grouped by 96-well plate number. Evaluation of these graphical data shows that this

variance was not plate-specific, meaning each plate showed variation in Ct values for the

GAPDH.


Range in GAPDH Duplicate Ct Measurements

36
34
32
30 -
28
S26 -
24 -
22 -
20
1 3 4 5 6 7 8 9 11 12 13 14 17 18 19 20 22 23 24 25 26 27 28 30 31 32 33 34 35 37 38 40 41 42 43 44 45 46 48 49 50

Sample #


Figure 4-1. Range of average duplicate Ct values of GAPDH per sample number. Note:
Arrow indicates calibrator sample.










GAPDH Average Duplicate Ct Values by Plate #


M 25
20





plate 1 plate 2 plate 3 plate 4 plate 5 plate 6
Plate #


Figure 4-2. Range of average duplicate GAPDH Ct measurements grouped by plate
number. Note: Arrow indicates calibrator sample.

Nonparametric Spearman's rho was performed to test correlations between the

duplicate GAPDH Ct and each of the alA- and 32-ADR duplicate Ct variables. GAPDH

was significantly correlated with GaA-ADR duplicate Ct (R = 0.628, p < 0.05).

Conversely, GAPDH was not significantly correlated with 02-ADR duplicate Ct (R =

0.247, p = 0.120). In addition, Student's t-test confirmed that GAPDH duplicate Ct

differed significantly between subjects with HTN versus NT (t = -2.634, df= 39, p <

0.05). Figure 4-3's boxplot represents the groupwise difference in GAPDH between

hypertensive and normotensive subjects. Each boxplot contains a box with a bisecting

line and two "whiskers" extending from either end. The upper and lower ends of the box

represent the upper and lower quartiles, respectively; or, the cutoffs for the 75th and 25th

percentiles, respectively. The line that bisects this box represents the median, or middle

value. The whiskers extend to the minimum and maximum values for the data. Figure 4-

4's boxplot shows the groupwise differences in GAPDH, alA- and 32-ADR raw Ct values.

These figures illustrate the variability in raw Ct values between groups and are not








informative of relative gene expression differences. Further interpretation of these data
and discussion of their importance is thoroughly presented in Chapter 5: Discussion.


Z sl -


NamokrdiM


Hyierkrelue


Diagnosis NT us. HTN
Figure 4-3. Boxplot for average duplicate GAPDH by diagnosis.


Is


BGAPDH (BC)
diplcat aue age
- Al A di plical
auerageCt
EB2 dip alk age E
Ct


Diagnosis NT vs. HTN


Figure 4-4. Boxplot for all gene expression raw Ct values.


I


iri









Assumptions of Normality

Data were assessed for normality with skewness, kurtosis, and the Kolmogorov-Smimof

test. These measures of normality indicated that relative gene expression variables (2A-

DDCt) for both alA- and 32-ADR were non-normally distributed skewnesss = 5.447,

kurtosis = 32.127, p < 0.05; and, skewness = 4.302, kurtosis = 20.722, p < 0.05, for alA-

and P2-ADR, respectively). Due to these violations of normality, non-parametric tests

were used for data analyses involving relative gene expression.

Analytic Results for Hypotheses

As these data were non-normally distributed, medians and inter quartile ranges

(IQRs) are presented for measures of central tendency and variance. These values for the

total sample are listed in Table 4-5. Further analytic results are presented by aim. In

addition, amplification plots for each gene, as expressed in the TaqMan Real-Time PCR

system (ABI Prism 7900) are presented in Appendix D.

Table 4-5. Gene expression medians and IQRs for total sam le
Gene Total sample
n= 41
Min 25% Med 75% max
alA-ADR 2A-DDCt 0.004 0.336 0.63 1.766 30.484
P2-ADR 2^-DDCt 0.009 0.105 0.32 1.000 40.224

Specific aim 1: To quantify differences in gene expression of alA- and 32-ADR in

the IMA between subjects with NT and HTN.

* a. To quantify relative differences in alA-ADR gene expression between study
groups with NT and HTN.
* b. To quantify relative differences in 32-ADR gene expression between study
groups with NT and HTN.

A summary of the median and IQR for each gene by diagnosis is presented in Table

4-6. Figure 4-5 shows boxplots of these data. For specific aim 1, the relative differences

in alA- (aim la) and 32-ADR (aim lb) gene expression between subjects with NT and











HTN was examined using the Mann-Whitney U test. For this nonparametric test, the null


hypothesis is that the two variables compared have identical distributions. More


specifically, it tests that the mean ranks of the 2A-DDCt values do not differ from the sum


of the ranks (mean of ranks not to be confused with mean of data). The results for these


tests are presented in Table 4-7. Median fold difference in gene expression of alA-ADR


and 32-ADR between subjects with NT and HTN were significant forp < 0.05. Fold-


differences are expressed as a ratio of HTN to NT subjects.


Table 4-6. Gene expression medians and IQRs for subjects by diagnosis.
Gene HTN NT
n=24 n=17
Min 25% Med 75% Max Min 25% Med 75% Max
c1A-ADR 2^-DDCt 0.15 0.53 1.41 2.68 30.48 0.004 0.13 0.36 0.84 1.72
P2-ADR 2^-DDCt 0.02 0.05 0.45 1.70 40.22 0.008 0.15 0.22 0.45 14.1



Gene expression differences by diagnoses
A .lA dIpilat2 -
B2 duplicate 2^ (DDC:
(DDCt) I-
:* r:l




A B






visualization.
13



030




Diagnosis NT vs. HTN

Diagnosis NT vs. HTN
A B
Figure 4-5. Boxplots for both gene's expression by diagnosis. A) Unadjusted scale with
black box showing selection for rescaling. B) Y-axis rescaled for better
visualization.

Table 4-7. Median fold differences in gene expression between normotensive and
hypertensive subjects and Mann-Whitney U tests.
Gene expression Ratio of subjects with HTN:NT
median-fold-difference P
alA-ADR 2^-DDCt 3.92 0.01*
P2-ADR 2^-DDCt 2.05 0.02*
*alpha < 0.05










Specific aim 2: To explore relative differences in gene expression of alA-ADR and

P2-ADR in the IMA between subjects with NT and HTN by race.

* a. To explore relative differences in alA- and 32-ADR gene expression between
White/Caucasian subjects with NT and HTN.

* b. To explore relative differences in the alA- and 32-ADR gene expression between
White/Caucasians with HTN versus Black/AAs with HTN.

For aim 2a, a summary of the median and IQR for each gene in White/Caucasian

subjects is presented in Table 4-8. Figure 4.6 shows boxplots of these data. To test the

hypothesis that the relative fold-differences in gene expression of the alA-ADR and 12-

ADR may be due, in part, to race (Aim 2a), Mann Whitney U test was performed to

compare gene expression differences between White/Caucasians with and without HTN.

When Caucasian hypertensive versus normotensive subjects were compared, ranks of

relative difference remained significant between median fold-differences in each gene's

expression. The fold-difference is expressed as a ratio of White/Caucasian subjects with

HTN to NT. These data are presented in Table 4-10.

Table 4-8. Gene expression medians, IQRs, and minimum and maximum values for
White/Caucasian subjects.
Gene White/Caucasian
n = 32
Min 25% Med 75% max
alA-ADR 2A-DDCt
Total 0.004 0.261 0.63 1.71 30.484
Hypertensive (n = 17) 0.146 0.381 1.45 3.19 30.484
Normotensive (n=15) 0.004 0.103 0.36 1.00 1.717
32-ADR 2A-DDCt
Total 0.009 0.079 0.37 1.25 40.224
Hypertensive (n = 17) 0.021 0.223 0.66 2.82 40.224
Normotensive (n=15) 0.009 0.042 0.22 0.42 14.026











S A--A duplicate 2' 7 oDct)2"
B2(DD plicat 22 dtplicate 2^-
(DDCt)

2-- 13
:2--
2 --


-l4



Normenslve Hypeenlve lo lm e HerImel
Diagnosis NT vs. HTN Diagrois NT s. HTN


A B
Figure 4-6. Boxplots for White/Caucasian subjects, for both gene's expression by
diagnosis. A) Unadjusted scale with black box showing selection for
rescaling. B) Y-axis rescaled for better visualization.

Table 4-9. Median fold differences in gene expression between White/Caucasian
normotensive and hypertensive subjects and Mann-Whitney U tests.
Gene expression Ratio of subjects with HTN:NT
Median fold-difference P
XlA-ADR 2^-DDCt 4.03 0.01*
P2-ADR 2A-DDCt 5.27 0.02*
*alpha < 0.05

For aim 2b, a summary of the median and IQR for hypertensive subjects by SIR is


presented in Table 4-11. Figures 4-7 and 4-8 show boxplots of these data, by gene. For


aim 2b, the Mann-Whitney U test was performed to compare Caucasian HTN versus


Black/AA HTN. This comparison did not show significance, as presented in Table 4-11.


This table also shows the median fold-differences in each gene's expression, expressed as


a ratio of White/Caucasian HTN to Black/AA HTN subjects.


Table 4-10. Gene expression medians and IQRs for Black/AA subjects.
Gene Hypertensives
n = 27
Min 25% Med 75% max
alA-ADR 2A-DDCt
White/Caucasian(n = 15) 0.146 0.381 1.45 3.19 30.484
Black/AA (n =9) 0.507 0.620 0.99 2.53 2.732
32-ADR 2A-DDCt
White/Caucasian (n = 15) 0.021 0.223 0.66 2.82 40.224
Black/AA (n =9) 0.055 0.095 0.17 1.00 11.004













Diagnosis NTvs HTN
SNormotensive
SHypertensive


DiqlostNTsu.m1
OD NT*I, H
SNHyperiUMe


Recorded SIR


Figure 4-7a. Boxplots for alA-ADR gene expression for White/Caucasian HTN versus
Black/AA HTN subjects. A) Unadjusted scale with black box showing
selection for rescaling. B) Boxplot rescaled.


Diagnosis NT vs H
E Normotensive
a Hypertensive


Dhgi~Os NTUg. HT
E Nonmlt -IN-1
B RHpe teitbw


Rc d SI13





Recorded SIR


Figure 4-8. Boxplots for 32-ADR gene expression for White/Caucasian HTN versus
Black/AA HTN subjects. A) Unadjusted scale with black box showing
selection for rescaling. B) Boxplot rescaled.


Table 4-11. Median fold differences in gene expression between White/Caucasian

hypertensive and Black/AA hypertensive subjects and Mann-Whitney U tests.
Gene expression Ratio of Cauc/White HTN:Black/AA HTN
niediaI-hold-dithcreIce P

ClA-ADR 2A-DDCt 1.47 0.55
P2-ADR 2A-DDCt 3.88 0.28
*alpha < 0.05


Specific aim 3: To explore the relationship between level of alA- and 32-ADR gene


expression and need for post-operative positive inotropic medication administration.


5 000 T

0000-
blackorM white
Recoded SIR


70D-









Table 4-12 shows the median, IQR, minimum and maximum values for both genes by

need for positive inotropes. Figure 4-9 shows the boxplots of these data. To test the

hypothesis that fold-differences in gene expression exist between subjects who required

post-operative positive inotrope administration and those who did not, the Mann-Whitney

U test was performed (Table 4-13). Median fold-difference of GaA-ADR and 32-ADR

gene expression between those who did and did not require post operative positive

inotropes is also shown in Table 4-13. Fold difference is expressed as a ratio of subjects

who received inotrope treatment to those who did not.

Table 4-12. Median, IQR, minimum and maximum values for alA-ADR and 32-ADR fold
difference in gene expression and need for post-operative positive inotrope
medication.
Gene Inotropes
n=41
Min 25% Med 75% Max
alA-ADR 2^-DDCt
No inotropes (n = 34) 0.004 0.301 0.625 1.95 30.484
Yes inotropes (n = 7) 0.339 0.507 0.727 1.59 1.670
V2-ADR 2^-DDCt
No inotropes (n = 34) 0.009 0.089 0.273 0.93 40.224
Yes inotropes (n = 7) 0.021 0.248 0.451 4.47 14.026


]AI dupltate 2"
(DDCt)


Dichotomous: inotrope administered Dichotomous inotrope ad mini stored

A B
Figure 4-9. Boxplots for both genes' expression by need for post-operative positive
inotrope. A) Unadjusted scale with black box showing selection for rescaling.
B) Boxplot rescaled.


p

-2





QI









Table 4-13. Fold differences in gene expression between non-inotrope and inotrope
subjects and Mann-Whitney U tests.
Gene expression Ratio of inotrope:non-inotrope
Median Ptld-dirt'ereince
alA-ADR 2^-DDCt 1.18 0.73
P2-ADR 2A-DDCt 1.67 0.36
*alpha < 0.05


Exploratory Aims

Following statistical analyses of the three specific aims, further exploratory

analyses were completed. For exploratory aim A, genotype data were examined for the

alA-ADR Arg347Cys C-T, the 32-ADR Argl6Gly G-A, and the 32-ADR Glu27Gln

C--G polymorphisms. Tables 4-14 and 4-15 indicate allele and genotype frequencies for

47 subjects who had blood collected, separated by racial/ancestral groups. Population

values for the alA-ADR Arg347Cys C--T polymorphism were obtained from the

Ensembl database (http://www.ensembl.org/index.html). Population estimates for

European Americans were from 24 samples and estimates for African Americans were

from 23 samples form the Coriell Cell repository (Ensembl, 2005). Population estimates

for the P2-ADR Argl6Gly G-A, and the 32-ADR Glu27Gln C-G polymorphisms were

obtained from the Pharmacogenetics and Pharmacogenomics Knowledge Base (Pharm

GKB) database of the INVEST study (INternational VErapamil SR and Trandolapril

Study, unpublished data), whereby 325 African American and 1,100 Cacuasian/European

American subjects were gentoyped (PharmGKB, 2006)

All genotypes were determined to be in Hardy-Weinberg Equilibrium (data not

shown), indicating that the gene frequencies and genotype ratios remained constant from

generation to generation in a randomly-breeding population.










Table 4-14. Allele frequencies for population versus sample, by SIR/ancestry.
Gene/ Codon Amino Acid/ Population Sample
Ref ID # Allele Allele Frequency Allele
Frequency
Major Minor Black/AA Cauc/ Black/AA Cauc/
Eur-Amer n = 10 Eur-Amer
n =37
1A- 347 Cys Arg* T .28 T .56 T .20 T .49
1048101 T C C .72 C .44 C .80 C .51
P2- 16 Gly Arg G .48 G .40 G .55 G .64
1042713 A G A .52 A .60 A .45 A .36
P2- 27 Gin Glu G .18 G .40 G .35 G .40
1042714 C G C .82 C .60 C .65 C .60

*In the AA population, Arg is the major allele

Table 4-15. Genotype frequencies for populaiton versus sample, by SIR/ancestry.
Gene/ Codon Amino Acid/ Population Sample
Ref ID # Allele Genotype Genotype
Frequency Frequency
Major Minor Black/AA Cauc/ Black/AA Cauc/
Eur-Amer Eur-Amer

alA- 347 Cys Arg* T/T .04 T/T .17 T/T .10 T/T .32
1048101 T C T/C .48 T/C .54 T/C .20 T/C .41
C/C .48 C/C .29 C/C .70 C/C .35



132- 16 Gly Arg G/G .27 G/G .37 G/G .30 G/G .41
1042713 A G A/G .50 A/G .46 A/G .50 A/G .60
A/A .23 A/A .17 A/A .20 A/A .12


132- 27 Gin Glu G/G .03 G/G .16 G/G .10 G/G .27
1042714 C G G/C .30 G/C .49 G/C .50 G/C .38
C/C .67 C/C .35 C/C .40 C/C .44


*In the AA population, Arg is the major allele.


Chi-square analyses were performed to examine the association between each of

the genes by diagnosis (HTN vs. NT). When cells contained values less than 5, Fisher's

Exact tests were used for nonparametric data. Table 4-16 shows these results. All Chi-


square/Fisher's Exact tests were nonsignificant forp < 0.05.











Table 4-16. Association between genotype and diagnoses of NT and HTN for the alA-
ADR and P2-ADR genes.
Gene NT HTN p
n = 20 n = 27
Count % within Count % within
diagnosis diagnosis
alA-ADR, Codon 347 0.18
C/C 5 25.0 14 51.9
C/T 8 40.0 8 29.6
T/T 7 35.0 5 18.5
P2-ADR, Codon 16 0.671
G/G 8 40.0 8 29.6
G/A 10 50.0 14 51.9
A/A 2 10.0 5 18.5
P2-ADR, Codon 27 0.521
C/C 3 15.0 7 25.9
C/G 7 35.0 11 40.7
G/G 10 50.0 9 33.3
S= Chi-square
S= Fisher's Exact


CTr Let r1rygote
Genotype: ADR A1A


Diagnosis NTvs. HTN
F Normoters ive
M Hypertensive


TTr I om oygote


Figure 4-10. Bar chart of alA-ADR, codon 347 by diagnosis.


60.0. -


310.0% -


C/C mozo
CIC hom0o49goi


i. .,-



,, ,












6Diagnosis NT s. HTN
[] Nornotensive
SHypere nsive








120 -










A honmozygole A he eroygole Gi0 hoIozmgoie
Genotype: ADR B2 Codon 16
Figure 4-11. Bar chart of p2-ADR, codon 16 by diagnosis.


Diagnosis NT %s. HTN
50.
















00- Normotensi-

SHyertensie


30A -


CIC hotm ygo CGoe herIoygole GiG h0m1 cyo e
Genotype: ADR B2 Codon 27

Figure 4-12. Bar chart of p2-ADR, codon 27 by diagnosis.


To examine the effect of confounding or population stratification of race on


genotype differences in HTN versus NT subjects, Fisher's exact was performed, but only


on White/Caucasian subjects. Tables 4-17 show these data and indicate these associations










are not significant. Given the allele and genotype frequency differences between

White/Caucasian and Black/AA subjects (for both sample and population estimates),

further analyses comparing associations stratified by SIR are warranted. However, very

low cell counts for the Black/AA group in this sample prevent any meaningful analyses

with data analyzed by genotype. When data were analyzed by allele for each gene with

Chi-square, cell counts were sufficient to examine allele by diagnosis associations for

both total sample and the White/Caucasian group, but were still too low in the Black/AA

group to warrant meaningful analyses. These data are presented in Tables 4-18 and 4-19.

Table 4-17. Fisher's Exact for genotype differences in White/Caucasian hypertensive vs.
normotensive subjects.
Gene White/Caucasian
n = 37
NT HTN
n =18 n =19
Counts Count % within Count % within P
diagnosis diagnosis
alA-ADR, Codon 347 0.131
C/C 3 16.7 9 47.4
C/T 8 44.4 6 41.6
T/T 7 38.9 4 21.1

32-ADR, Codon 16 0.69
G/G 8 44.4 6 31.6
G/A 8 44.4 11 57.9
A/A 2 11.1 2 10.5

32-ADR, Codon 27 0.57
C/C 3 16.7 6 31.6
C/G 7 38.9 6 31.6
G/G 8 44.4 7 36.8




Table 4-18. Chi-square for allele counts by diagnosis for the alA-ADR and 32-ADR genes
in all subjects.
Gene All subjects
N= 47
NT HTN
n = 20 n = 27
Counts Allele % count Allele % count P
frequency frequency_










Table 4-18 Continued.
Gene All subjects
N= 47
NT HTN
n = 20 n = 27
Counts Allele % count Allele % count R
frequency frequency
(XIA-ADR, Codon 347 0.03
C .45 33.0 .67 66.0
T .55 55.0 .33 45.0
32-ADR, Codon 16 0.04
G .65 46.0 .56 54.0
A .35 37.0 .44 63.0
32-ADR, Codon 27 0.18
C .33 34.0 .46 66.0
G .67 48.0 .54 52.0
*alpha < 0.05. Note: Percent count equals row count.


Table 4-19. Chi-square for alleles by diagnosis for the alA-ADR and 32-ADR genes in
White/Caucasian subjects.
Gene White/Caucasian
n = 37
NT HTN
n = 18 n = 19
Counts Allele % count Allele % count R
frequency frequency
IA-ADR, Codon 347 0.04
C .39 38.9 .63 63.1
T .61 61.1 .37 36.8
32-ADR, Codon 16 0.58
G .67 66.7 .61 60.5
A .33 33.3 .39 39.5
32-ADR, Codon 27 0.33
C .36 36.1 .47 47.4
G .64 63.9 .53 52.6
*alpha < 0.05

To explore if genotype differences were correlated with gene expression

differences, Kruskal-Wallis tests were performed. This test is a nonparametric alternative

to a One-Way Analysis of Variance that extends the Mann-Whitney U test to more than

two groups. This is necessary, as we need to examine three groups for the genotypes. No

significant differences were found for genotype by gene expression. These data are

presented in Table 4-20.










Table 4-20. Kruskal-Wallis tests for genotype counts by gene expression alA-ADR and
P2-ADR genes.
Genotype X Gene n = 41 p
expression
expression Count Median gene
expression
,IA-ADR, Codon 347 0.49
X clA-ADR 2^-DDCt
C/C 17 0.99
C/T 13 0.58
T/T 11 0.42
P2-ADR, Codon 16 X 0.80
P2-ADR 2^-DDCt
G/G 14 0.55
A/G 21 0.67
A/A 6 1.15
32-ADR, Codon 27 x 0.33
32-ADR 2^-DDCt
C/C 9 1.00
C/G 14 0.63
G/G 18 0.57

Effect Sizes and Power Calculations

Cohen's d (actual effect sizes) were calculated for all aims and are presented in

Table 4-21. Actual effect sizes varied by aims. The original power calculations

anticipated a medium effect size, based on the literature. Therefore, these Cohen's d

values indicate a greater sample size was needed to power these aims. With such small

effect sizes, the group sizes would need to be greater than what was sampled to achieve a

power of 80%.


Table 4-21. Power and effect sizes by aim.
Calculations
Aim Power Effect Size Number
Statistical test needed
la* Total sample by diagnosis: (alA-ADR 0.66 0.67 34
Mann-Whitney U
lb* Total sample by diagnosis: 32-ADR 0.35 0.35 72
Mann-Whitney U
2a* White/Caucasians by diagnosis: clA-ADR 0.56 0.56 30
Mann-Whitney U
2a White/Caucasians by diagnosis: 32-ADR 0.30 0.40 64
Mann-Whitney U
2b Hypertensives by SIR: (alA-ADR 0.31 0.35 58
Mann-Whitney U







77


Table 4-21 Continued.
Calculations
Aim Power Effect Size Number
Statistical test needed
2b Hypertensives by SIR: (32-ADR 0.17 0.27 100
Mann-Whitney U
* Indicates these tests were significant in analyses.
a The number needed in each group to achieve a power of 0.80. This assumes equal
numbers in each group.














CHAPTER 5
DISCUSSION AND RESULTS

Introduction

All descriptive and analytic results for the proposed aims and exploratory analyses

will be discussed in this chapter. Conclusions and implications for nursing as well as

recommendations for future research will also be provided.

Discussion of Results

Demographics

In this study, 42.6% of all enrolled subjects were normotensive and 57.4% were

hypertensive. When examining those included in the gene expression subset, the

percentages were similar with 41.5% normotensive and 58.5% hypertensive subjects.

These percentages indicate that there are a number of normotensive patients undergoing

bypass surgery, a phenomenon surprising to some researchers. These normotensive

subjects were comprised predominantly of self-identified White/Caucasians (n = 17,

85%), with only 2 subjects (10%) who self-identified as Black/AA. This study anticipated

15 subjects who were Black/AA with NT, which, based on the UF TCV Surgery

department's database of patients from 2001-2002, 15 subjects would have amounted to

roughly 60% of their total Black/AA NT population for that fiscal year. Only 8 subjects

who self-identified as Black/AA and were diagnosed with HTN were recruited for the

study. This made a total of 10 (21.3%) self-identified Black/AA subjects for the entire

study. (This number declined by one for the gene expression subset.) The 2000 U.S.

Census for Alachua County estimates the Black/AA population to be just 19.3%. Based









solely on this, this study attained a representative sample of self-identified Black/AA in

Alachua County. However, the UF TCV Surgery department's database for 2001-2002

indicated approximately 100 of 389 patients who had bypass surgery were listed as

Black/AA. This led us to believe that sampling 30 Black/AA subjects should not be a

problem. In addition, sampling occurred at both a community-based general hospital

(Shands at AGH) and a regional Veteran's medical center, which we felt would possibly

lead to a greater Black/AA population from which to sample. Despite these points, the PI

was unable to recruit a sufficient cohort of Black/AAs. Very few NT Black/AA patients

were identified for potential recruitment. Some possible reasons for this are that many of

these patients could have been emergent cases (and unable to be consented 24 hours prior

to surgery, as required by IRB), many of these patients may have refused surgical

intervention. Finally, any normotensive patients, not just normotensive AAs, may have

been referred for interventional procedures such as percutaneous coronary intervention

(for example, stenting, atherectomy, or balloon catheter angioplasty). This is perhaps the

most plausible explanation for the reduced number of normotensive patients who undergo

bypass surgery, in that they may have less severe comorbidities and are recommended to

interventional cardiology rather than to thoracic surgery for bypass. Only a small number

of Black/AAs (regardless of diagnosis) approached to participate in the study chose not to

enroll. Major reasons cited for not wishing to participate were: not wanting to be

"bothered with anything else" and "not feeling comfortable with the study".

Only three subjects (6.4%) self-identified as having more than one race; all three

considered themselves both White/Caucasian and American Indian/Alaska Native. This

number could not be compared to US Census data, as the Census does not specifically









report combinations of dual-identification. The percentage of self-identified Hispanic and

non-Hispanic were 2.1% and 95.7%, respectively. Compared to US Census reports for

Alachua County, approximately 5.7% report Hispanic (of any race) and 94.3% report

non-Hispanic ethnicities.

Just over 21% of enrolled subjects were female and nearly 79% were male (22%

and 78% female and male, respectively, in the gene expression subset). As indicated in

the TCV Surgery department's database, the average percent of women undergoing

CABG surgeries by the TCV surgery department is 21%. This means that for every one

female undergoing the procedure by this department, there are approximately four males.

Therefore, the enrolled percentage of females met the expected percentage. This indicates

that this sample is representative of the population of females undergoing bypass surgery.

Interpreted collectively, the demographic data of this sample suggest it to be

moderately representative of the population of bypass patients who undergo surgery in

Alachua County, but is not completely representative of the entire racial, ethnic or bypass

populations.

Clinical characteristics for subjects are presented in Tables 4-2 and 4-4 for enrolled

and gene expression subsets, respectively. Hypertensive and normotensive groups did not

significantly differ in age, height, weight, or BMI (refer to page 59). The concomitant

diagnosis of T2DM was seen in 34% of the overall sample, with 2.1% (n = 1) having a

diagnosis of pre-diabetes. Only 5% (n = 1) of all NT subjects were diagnosed with

T2DM. Among all hypertensive subjects, 55.6% were Type 2 diabetic and 3.7% were

pre-diabetic. In regards to prescribing of P-blockers, a first-line class of drugs for both

coronary artery disease and hypertension (AHA, 2005; JNC VII, 2003), a surprising









percentage of subjects (34% overall) were not prescribed this medication. Twenty two

percent of hypertensive and 53% of normotensive subjects were not on 3-blocker

medication. Of subjects with NT, 47.4% were prescribed P-blockers, but none were

prescribed more than 50 mg, twice a day. Of subjects with HTN, 77.8% were prescribed

this therapy with only three subjects (11.1%) taking more than 50 mg, twice a day. Table

4.2 shows that the majority of subjects who were prescribed P-blocker therapy (57.4% of

the overall sample) were taking between 12.5mg to 50.0 mg, twice a day.

Gene Expression Measures of Central Tendency and Variance

When examining measures of central tendency in these data, a few things warrant

consideration. First, the means are not the best representation of centrality of these data

because multiple outliers skew these data and bias the means. While the mean is typically

considered more stable over time (and with repeated random selection), the median is

considered a middle point, an index of average position, that is not affected by skewed

data with outliers (Portney & Watkins, 2000). When comparing the means and medians

(data not shown), nearly every mean value is visibly inflated by the outliers and the

medians appear to better represent the centrality of these data. Furthermore, evaluating

the standard deviations (SDs) causes more concern. The SDs are fairly large (data not

shown), especially compared to the means and medians. Since the nature of the SD is to

represent the variability in the data, it is typically expressed as the spread from each end

of the mean (for example, 'plus or minus'). If we were to subtract some of these SDs

from their corresponding means, we would actually be left with a negative number. For

example, the mean and SD for the alA-ADR relative gene expression for the total sample

is 1.97 and 4.83, respectively. This would indicate a range of values from -2.86 to 6.8.

From the point of view of the gene expression biological assay, it is impossible to have









negative numbers. This phenomenon occurs for nearly every set of values in this study.

Therefore the SD is not the best measure of variability for these data. Taken together, this

information indicates the median and IQRs are the best representation of centrality and

spread in these data and are presented for these data throughout the chapters (in tables

and boxplots). Furthermore, because the median-derived 2A-DDCt values are similarly

less influenced by outliers, median fold-differences between groups was presented.

Discussions for Choice of GAPDH for Normalization Gene

During the initial planning phase of this project, the normalization (or,

housekeeping)gene anticipated for use was cyclophilin A, as it was previously reported as

a successful normalizer in arterial tissue (Lieu, Withycombe, Walker, Rong, Walzem, and

Wong, et al., 2003; Trogan, Choudhury, Dansky, Rong, Breslow, & Fisher, 2002). Prior

to purchasing this housekeeping gene, another review of the literature was conducted to

see if any new information had been reported about this gene's use in normalization. In

fact, a recent publication by Escobales and Crespo reported evidence that reactive oxygen

species appeared to be mediated by a number of factors, including cyclophilin A (2005).

As the reactive oxygen pathway is implicated in HTN, the use of cyclophilin A to

normalize samples of hypertensive subjects made this gene inappropriate for use as a

housekeeping gene in this study. Another review of the literature revealed support for the

use of GAPDH in similar tissue types. Peuster, Fink, Reckers, Beerbaum, and von

Schnakenburg reported consistent amplification of GAPDH among samples in a study of

unstented coronary arteries in pigs (2004). Wang and Brown showed successful use of

GAPDH for normalization in their study of P-ADR subtypes in atrial appendages (2001).

Furthermore, preliminary analyses with a small selection of the PI's sample in single-

plexed reactions yielded triplicate GAPDH values that were less than one-half a Ct from









one another and Ct values between samples that were very close. Technical experts in the

UF ICBR Gene Expression Core facility viewed these preliminary data and supported the

decision to use GAPDH in the final experiments. These aforementioned references and

preliminary results provided the foundation for the decision to choose GAPDH as the

housekeeping gene in this study.

Assessment of the Performance of GAPDH as a Normalizer

To reiterate, one of the major assumptions of performing the relative quantitation

method of gene expression analyses is that the housekeeping, or normalization gene

(here, GAPDH) expresses similarly across subjects and/or experimental conditions. This

is typically assessed by examining the Ct values of the GAPDH wherein the Ct values are

expected to show very little variability (no more than 1/2 -2 Cts difference) and the

standard deviation of the mean should be small. If the GAPDH gene expression shows

greater variability than this, it is theoretically a poor housekeeping gene for the data and

is cautioned for normalization use. In this particular study, the duplicate GAPDH Ct

values had a mean and standard deviation of 27.56 + 2.80 and median of 27.4 with values

ranging from 22.1 to 34.2. According to Dorak (2003), the endogenous control

(housekeeping gene) should be more abundant (or, have smaller Ct values) than the target

genes. This was true for the alA-ADR (median 29.45, SD 2.50), but not the 32-ADR

(median 26.10, SD 0.78). Possible explanations for this greater-than-expected Ct for

GAPDH are poor PCR efficiency or low copy numbers (Dorak, 2003). The 12-point

range of Ct values for GAPDH indicates the GAPDH did not, in fact, express consistently

across subjects. While poor PCR efficiency, low copy number, and/or pipetting errors

may contribute to this variation in GAPDH, another plausible explanation is the

occurrence of RNA degradation. As previously stated, RNA degradation was determined









via a 2100 Bioanalyzer (Agilent Technologies, Palo Alto, CA), and concluded that

overall, the level of RNA degradation for these samples was low. However, to examine if

RNA degradation may have played a role in GAPDH variation, the 18s and 28s graphical

data were examined and compared to those samples that had poor triplicate Ct values for

GAPDH (meaning, those with more than 2 Cts difference). Qualitatively speaking this

review of the data concluded that, at least for a handful of the samples, RNA degradation

could help to explain some of the variance in GAPDH. The most extreme case of this was

with one sample, whereby raw Ct values were 35.9, 30.9, and 27.9 and the graphical data

from the bioanalyzer indicated more RNA degradation in this particular sample as

compared to other samples. Perhaps results for the GAPDH Ct values may have been less

variable for some of the samples where RNA degradation was a potential issue.

Figure 4-2 shows that variation in GAPDH was not plate-specific, meaning certain

96-well plates did not show more or less variation than others. This refutes the notion that

specific plates may have been outliers due to the order of preparation, time lapse between

preparation and initialization of Real Time RT-PCR, or other sources of external error. In

addition, a stem and leaf plot (data not shown) indicated 3 "extremes" greater than 32.9;

however, these values are not greater than 2 SDs from the mean. Similar variance has

been reported in the literature for GAPDH in various species and tissues. Schmid, Cohen,

Henger, Irrang, Schlondorff, and Kretzler (2003) showed variation by tissue source in

their study, reporting GAPDH Ct median and standard deviation of 22.57 + 2.61 in

tubulointerstitial compartments and 28.96 + 2.38 in glomeruli, both from human renal

biopsies. They did not report a range of Ct values. Peuster, Fink, Reckers, Beerbaum and

von Schnakenburg (2004) reported a median of 22.2 (range 19.8-26.9) of GAPDH Ct









values in their examination of porcine left coronary artery. Despite this 7-point range,

they reported consistent amplification across all samples, as determined by serial

dilutions of GAPDH. They used the Delta Delta Ct method, normalizing with GAPDH.

Lennmyr, Terent, Svyanen, and Barbany (2005) discuss instability in GAPDH gene

expression in their samples of middle cerebral artery in rats, but reported that because the

changes were not statistically significant, they used the gene to normalize. Tricarico,

Pinzani, Bianchi, Paglierani, Distante, and Pazzagli, et al. (2002) graphically showed raw

GAPDH Cts between approximately 22-23 in human breast tissue. They correlated their

target genes with the GAPDH by Spearman's rank (nonparametric), showing

significance. This supported their decision not to use the GAPDH as a normalizer. The

authors suggest normalization to total RNA concentration as an alternative in this type of

situation. The methodology behind this solution was not delineated by the authors. The

following arguments were discussed in personal communication with Y. Conley (2006)

regarding this possible solution: First and foremost, if we assume that the expression of a

gene is altered during a disease state (here expression of the ADR genes in HTN), then

this could theoretically alter the total RNA. This would seem inappropriate, then, to

normalize both diseased and non-diseased samples with a total RNA value. Whatever

value is used to normalize, it must be constant. This same principle applies when

considering taking the average of all GAPDH values across all samples (or even by plate)

and normalizing in this fashion, making these options also undesirable.

Barber, Harmer, Coleman, and Clark (2005) performed a thorough evaluation of

GAPDH as a housekeeping gene, examining its expression in 72 human tissue types. One

tissue type examined was coronary artery, although the specific artery was not noted.









They reported (graphically) a mean Ct value of approximately 20, but no range. Among

the 72 tissues examined, GAPDH mRNA gene expression varied 15-fold between tissue

types, further supporting previously published variability. The authors also reported

GAPDH Ct outliers below 13 and above 32.761; in these cases, they removed these data

points. These studies highlight the variability in reports of GAPDH in different tissues

and species as well as variation in how researchers handle the normalization dilemma and

outliers. No studies were found that reported expected raw Ct values for GAPDH in

human IMA tissue, thus providing little evidence of an expected Ct value for GAPDH in

the IMA tissues used in this study.

Given the literature presented above, GAPDH was analyzed statistically for its

correlation with the target genes and differences between experimental groups. There was

a significant correlation between GAPDH duplicate Ct and GaA-ADR duplicate Ct but not

between GAPDH duplicate Ct and 32-ADR duplicate Ct (data shown on page 62). The

first glance indicates that GAPDH should not be used for normalization of the alA-ADR

gene expression data, but could be used for normalization of the 32-ADR. The reason for

this is that a direct linear relationship should not exist between a target and housekeeping

gene, theoretically speaking. The housekeeping gene should remain constant at any given

value of the target gene expression. Additionally, the housekeeping gene should not show

differential expression between experimental groups. In fact, the GAPDH housekeeping

gene expression in this study was significantly different between hypertensive and

normotensive subjects (see data page 65 and Figure 4-3). These data collectively indicate

that the GAPDH used in this study did not optimally perform as a housekeeping gene.

However, the data were normalized to GAPDH because no other options were available.









Raw Ct values are unable to be used for comparison, as this is an exponential value

determined from a log-linear plot of PCR signal versus cycle number (Livak &

Schmittgen, 2001). No other housekeeping genes were used. For this reason,

interpretation of statistically significant results is extremely cautioned and further

inferential analyses would be inappropriate. Boxplots of group differences and GAPDH

values were shown (Figures 4-3 and 4-4) to allow for visual comparison of group data (as

recommended in personal communication with N. Chegini, 2005). On a final note, use of

single-plexing (loading targets and GAPDH in separate wells) instead of multiplexing

(loading targets and GAPDH in same wells) at the very least, makes us confident that the

values we obtained for the expression of the targets and GAPDH are more valid and

reliable. This is because they were amplified separately and did not have to compete for

reagents during cycling and detection. A pitfall in single-plexing is that this less

accurately controls for pipetting errors.

Aims

Hypertensive subjects showed a 3.92-fold difference in relative alA-ADR gene

expression compared to normotensive subjects, a difference that was statistically

significant (aim la, Table 4-7). Examination of the median 2A-DDCt values for the alA-

ADR informs us that hypertensive subjects showed nearly 4 times lower expression of

the alA- ADR gene in the arterial tissue investigated, perhaps suggesting the possibility of

blunted vasoconstriction in this group as compared to normotensives (note: the higher 2A-

DDCt median in the hypertensive group indicates lower gene expression and thus,

downregulation of the gene). It has previously been described that alA-ADR

downregulation could be explained as a consequence of enhanced sympathetic tone (for

example, increased vascular resistance) in HTN (Jacobs, Lenders, Willemsen & Thien,