<%BANNER%>

Clinical and Genetic Predictors of the Dysglycemic Effects of Antihypertensive Medications

Permanent Link: http://ufdc.ufl.edu/UFE0044461/00001

Material Information

Title: Clinical and Genetic Predictors of the Dysglycemic Effects of Antihypertensive Medications
Physical Description: 1 online resource (236 p.)
Language: english
Creator: Moore, Mariellen J
Publisher: University of Florida
Place of Publication: Gainesville, Fla.
Publication Date: 2012

Subjects

Subjects / Keywords: antihypertensive -- atenolol -- hydrochlorothiazide -- prediabetes
Pharmaceutics -- Dissertations, Academic -- UF
Genre: Pharmaceutical Sciences thesis, Ph.D.
bibliography   ( marcgt )
theses   ( marcgt )
government publication (state, provincial, terriorial, dependent)   ( marcgt )
born-digital   ( sobekcm )
Electronic Thesis or Dissertation

Notes

Abstract: The top five classes of guideline recommended antihypertensive agents promote markedly dissimilar effects on glucose and development of type 2 diabetes mellitus. While ACE inhibitors and ARBs may exert beneficial glycemic effects, beta-blockers and thiazide diuretics are notable for their propensity to elevate glucose and are associated with diabetes development. In a pilot study, we investigated two methods for evaluating pre-diabetes, fasting glucose and a 2-hour glucose level after an oral glucose tolerance test, in untreated hypertensives prior to and after treatment with a beta-blocker (atenolol), thiazide diuretic (hydrochlorothiazideHCTZ), and the combination. We found fasting glucose and an oral glucose tolerance test similar in detecting medication associated pre-diabetes. In a larger cohort, we then assessed the clinical and laboratory parameters associated with elevations in glucose after treatment with atenolol or HCTZ. We found that for both drugs baseline glucose was the strongest predictor for medication associated elevation in glucose, in both derivation and validation cohorts. To evaluate possible genetic factors involved in medication associated dysglycemia, we studied candidate genes important for drug response and insulin signaling. We found significant associations in blacks treated with HCTZ with the blood pressure response genes SCNN1Gand NOS3. Further, we found additional associations in both drugs in an insulin signaling gene (atenolol and HCTZ monotherapy, non-blacks in IRS1). To our knowledge, this study addressed many clinical concerns not previously answered in the literature. The current data suggest fasting glucose is an appropriate phenotype to measure antihypertensive-associated adverse glycemic changes, baseline glucose is the strongest predictor for antihypertensive-associated elevations in glucose and of the genes selected, both blood pressure response genes and insulin signaling genes may play a role in antihypertensive-associated change in glucose.However, further research is necessary to fully understand the mechanism and implications of antihypertensive medication-associated dysglycemia.
General Note: In the series University of Florida Digital Collections.
General Note: Includes vita.
Bibliography: Includes bibliographical references.
Source of Description: Description based on online resource; title from PDF title page.
Source of Description: This bibliographic record is available under the Creative Commons CC0 public domain dedication. The University of Florida Libraries, as creator of this bibliographic record, has waived all rights to it worldwide under copyright law, including all related and neighboring rights, to the extent allowed by law.
Statement of Responsibility: by Mariellen J Moore.
Thesis: Thesis (Ph.D.)--University of Florida, 2012.
Local: Adviser: Johnson, Julie A.
Electronic Access: RESTRICTED TO UF STUDENTS, STAFF, FACULTY, AND ON-CAMPUS USE UNTIL 2014-12-31

Record Information

Source Institution: UFRGP
Rights Management: Applicable rights reserved.
Classification: lcc - LD1780 2012
System ID: UFE0044461:00001

Permanent Link: http://ufdc.ufl.edu/UFE0044461/00001

Material Information

Title: Clinical and Genetic Predictors of the Dysglycemic Effects of Antihypertensive Medications
Physical Description: 1 online resource (236 p.)
Language: english
Creator: Moore, Mariellen J
Publisher: University of Florida
Place of Publication: Gainesville, Fla.
Publication Date: 2012

Subjects

Subjects / Keywords: antihypertensive -- atenolol -- hydrochlorothiazide -- prediabetes
Pharmaceutics -- Dissertations, Academic -- UF
Genre: Pharmaceutical Sciences thesis, Ph.D.
bibliography   ( marcgt )
theses   ( marcgt )
government publication (state, provincial, terriorial, dependent)   ( marcgt )
born-digital   ( sobekcm )
Electronic Thesis or Dissertation

Notes

Abstract: The top five classes of guideline recommended antihypertensive agents promote markedly dissimilar effects on glucose and development of type 2 diabetes mellitus. While ACE inhibitors and ARBs may exert beneficial glycemic effects, beta-blockers and thiazide diuretics are notable for their propensity to elevate glucose and are associated with diabetes development. In a pilot study, we investigated two methods for evaluating pre-diabetes, fasting glucose and a 2-hour glucose level after an oral glucose tolerance test, in untreated hypertensives prior to and after treatment with a beta-blocker (atenolol), thiazide diuretic (hydrochlorothiazideHCTZ), and the combination. We found fasting glucose and an oral glucose tolerance test similar in detecting medication associated pre-diabetes. In a larger cohort, we then assessed the clinical and laboratory parameters associated with elevations in glucose after treatment with atenolol or HCTZ. We found that for both drugs baseline glucose was the strongest predictor for medication associated elevation in glucose, in both derivation and validation cohorts. To evaluate possible genetic factors involved in medication associated dysglycemia, we studied candidate genes important for drug response and insulin signaling. We found significant associations in blacks treated with HCTZ with the blood pressure response genes SCNN1Gand NOS3. Further, we found additional associations in both drugs in an insulin signaling gene (atenolol and HCTZ monotherapy, non-blacks in IRS1). To our knowledge, this study addressed many clinical concerns not previously answered in the literature. The current data suggest fasting glucose is an appropriate phenotype to measure antihypertensive-associated adverse glycemic changes, baseline glucose is the strongest predictor for antihypertensive-associated elevations in glucose and of the genes selected, both blood pressure response genes and insulin signaling genes may play a role in antihypertensive-associated change in glucose.However, further research is necessary to fully understand the mechanism and implications of antihypertensive medication-associated dysglycemia.
General Note: In the series University of Florida Digital Collections.
General Note: Includes vita.
Bibliography: Includes bibliographical references.
Source of Description: Description based on online resource; title from PDF title page.
Source of Description: This bibliographic record is available under the Creative Commons CC0 public domain dedication. The University of Florida Libraries, as creator of this bibliographic record, has waived all rights to it worldwide under copyright law, including all related and neighboring rights, to the extent allowed by law.
Statement of Responsibility: by Mariellen J Moore.
Thesis: Thesis (Ph.D.)--University of Florida, 2012.
Local: Adviser: Johnson, Julie A.
Electronic Access: RESTRICTED TO UF STUDENTS, STAFF, FACULTY, AND ON-CAMPUS USE UNTIL 2014-12-31

Record Information

Source Institution: UFRGP
Rights Management: Applicable rights reserved.
Classification: lcc - LD1780 2012
System ID: UFE0044461:00001


This item has the following downloads:


Full Text

PAGE 1

1 CLINICAL AND GENETIC PREDICTORS OF THE DYSGLYCEMIC EFFECTS OF ANTIHYPERTENSIVE MEDICATIONS By MARIELLEN J. MOORE A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT OF THE R EQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY UNIVERSITY OF FLORIDA 201 2

PAGE 2

2 201 2 Mariellen J. Moore

PAGE 3

3 To my family: my mother Gloria Moore my father, Dan Moore my brother, Matthew Moore my grandmothers, Eleanor Zaremba & Betty Moore my aunts and uncles Maria & Ricardo Menendez; Elizabeth & Richard Henderson; J. Thomas Moore; & Walter Roshetski

PAGE 4

4 ACKNOWLEDGMENTS I would like to express my sincere gratitude to my mentor Dr. Julie Johnson for providing me the opportunity to conduct my studies i n her laboratory. I greatly apprecia te her many years of support and guidance in my professional development I value the numerous opportunities I ha ve had for enrichment under her mentorship I have learned much from her passionate and creative approach t o science. I would also like to thank the members of my graduate supervisory committee, Dr. Rhonda Cooper DeHoff, Dr. Yan Gong, and Dr. Chris Baylis, for their expertise and advice Dr. Cooper extensive experience with clinical trials and IRBs pro ved invaluable in developing and conducting my clinical study. Dr. Gong was always willing to offer insight when I was stumped and provide guidance, no matter how simple my questions se background proved thought provoking and helped me co nsider ideas beyond my topic. The completion of this dissertation would not be possible without the commitment and patience of my mentor and graduate committee. I am grateful to current and former faculty, staff, and colleagues of the department of Pharma cotheray and Translational Research for their support, adv ice and teaching. In particular I would like to thank Dr s Melonie Stanton, Heather Davis, Julio Duarte, Caitrin McDonough, Mohamed E. Mohamed, Maximilian Lobmeyer, Elv in Price, Anzeela Schentrup Gr eg Welder, Amber Beitelshees, Jaekyu Shin, Taimour Langaee and Reginald Frye I would also like to thank the graduate students and post doctoral fellows I have had the privilege to learn from over the years Further, I would like to thank the students in t he college of pharmacy I have been honored to work with. I would like to acknowledge Dr. Taimour Langaee, Dr. Greg Welder, Ms. Cheryl Galloway, Mr. Benjamin Burkley, and Ms. Lynda Stauffer for their significant support in laboratory

PAGE 5

5 efforts. I would also l ike to thank Mrs. Pamela Connolly, Ms. Delores Buffington, Mrs. Danielle Taylor, and Ms. Tamithy Morris for their assistance with the clinical study. I am grateful to the PEAR steering committee for allowing me to recruit PEAR patients into the clinical su b study. Importantly, I would like to thank all PEAR patients that participated in the clinical sub study. Additionally I must acknowledge NIH grant T32 HL083810 04 and the Department of Physiology & Functional Genomics, College of Medicine and UF Hyperte nsion Center for support of my graduate studies as well as Dr. laboratory efforts My heartfelt gratitude goes to my family and friends for their unconditional love, encouragement, s upport and laughter. Importantly, my love and abundant appreciation go to my Mom. Her snail mail, calls, and trips to Gainesville meant more than I can express.

PAGE 6

6 TABLE OF CONTENTS page ACKNOWLEDGMENTS ................................ ................................ ................................ .. 4 LIST OF TABLES ................................ ................................ ................................ ............ 8 LIST OF FIGURES ................................ ................................ ................................ ........ 12 LIST OF ABBREVIATIONS ................................ ................................ ........................... 14 ABSTRACT ................................ ................................ ................................ ................... 17 CHAPTER 1 INTRODUCTION AND BACKGROUND ................................ ................................ 19 Cardiovascular Disease, Hyperte nsion and Diabetes ................................ ............. 19 Interrelationship of Hypertension, Diabetes, and Antihypertensive Agents ............. 22 Use of Beta blockers and Diure tics in Hypertension and Dysglycemia ................... 28 Lack of Clinical Predictors Associated with Beta blocker and Diuretic Associated Adverse Metabolic Effects ................................ .......................... 29 Clinical Predictors of Risk for Antihypertensive Medication Associated Dysglycemia ................................ ................................ ................................ .. 32 Genetic Determinants of Type 2 Diabetes Mellitus and Risk for Drug Associated Type 2 Diabetes Mellitus ................................ ............................ 34 Gaps in Knowledge and Study Objectives: Adverse Metabolic Effects Associated with Antihypertensive Medications ................................ .................... 37 2 ANTIHYPERTENSIVE MEDICATION EXPOSURE AS RISK FOR IMPAIRED GLUCOSE TOLERANCE: A PEAR SUB STUDY ................................ ................... 46 Introduction ................................ ................................ ................................ ............. 46 Metho ds ................................ ................................ ................................ .................. 48 Study Population ................................ ................................ .............................. 48 PEAR Protocol ................................ ................................ ................................ 48 PEAR Sub Study Protoc ol ................................ ................................ ................ 49 Statistical Methods ................................ ................................ ........................... 52 Results ................................ ................................ ................................ .................... 53 Primary Findings: Fastin g Glucose Versus OGTT data ................................ .... 53 Secondary Findings: Surrogate Indexes for Insulin Resistance ....................... 54 Discussion ................................ ................................ ................................ .............. 55 3 CLINICAL PREDICTORS OF ELEVATION IN GLUCOSE ASSOCIATED WITH USE OF BETA BLOCKERS AND THIAZIDE DIURETICS ................................ ..... 68 Introduction ................................ ................................ ................................ ............. 68

PAGE 7

7 Methods ................................ ................................ ................................ .................. 69 Protocol ................................ ................................ ................................ ............ 69 Laboratory Measurements ................................ ................................ ................ 70 Biochemical Assays ................................ ................................ ......................... 70 Anthropometric Measurements ................................ ................................ ........ 71 Statistical Methods ................................ ................................ ........................... 71 Results ................................ ................................ ................................ .................... 73 Discussion ................................ ................................ ................................ .............. 75 4 EFFECT OF GENETIC VARIATION IN PHARMACOLOGICAL TARGET AND INSU LIN SIGNALING GENES ON ALTERATIONS IN GLUCOSE AFTER TREATMENT WITH BETA BLOCKERS AND THIAZIDE DIURETICS ................... 91 Introduction ................................ ................................ ................................ ............. 91 Metho ds ................................ ................................ ................................ .................. 94 Study Population ................................ ................................ .............................. 94 Glucose Measurement ................................ ................................ ..................... 94 DNA Collection and Genotyping ................................ ................................ ....... 94 Statistics ................................ ................................ ................................ ........... 96 Results ................................ ................................ ................................ .................... 99 Genetic Associations with Change in Glucose: Primary Analysis of Known Functional SNPs in Pharmacological Candidate Genes ............................. 100 Genetic Associations with Change in Glucose: Secondary Analysis in Pharmacological Cand idate Genes ................................ ............................. 101 Genetic Associations with Change in Glucose: Insulin Signaling Genes ........ 101 Discussion ................................ ................................ ................................ ............ 102 5 SUMMARY AND CONCLUSION ................................ ................................ .......... 118 APPENDIX A DATA AND FIGURES ASSOCIATED WITH CHAPTER 2 ................................ .... 123 B DATA AND FIGURES ASSOCIATED WITH CHAPTER 3 ................................ .... 13 7 C DATA AND FIGURES ASSOCIATED WITH CHAPTER 4 ................................ .... 153 D EFFECT OF ATENOLOL ON EXPRESS ION OF THE INSULIN RECEPTOR BY GENOTYPE GROUP FOR THE RS7508679 SINGLE NUCLEOTIDE POLYMORPHISM ................................ ................................ ................................ 200 LIST OF REFERENCES ................................ ................................ ............................. 214 BIOGRA PHICAL SKETCH ................................ ................................ .......................... 236

PAGE 8

8 LIST OF TABLES Table page 1 1 Definition of key terms. ................................ ................................ ....................... 42 1 2 Odd s ratio for incident diabetes in antihypertensive clinical trials ....................... 43 1 3 American Diabetes Association glucose thresholds for normal gluc ose, pre diabetes, and diabetes ................................ ................................ ....................... 44 1 4 Formulas for surrogate indexes of insulin resistance ................................ .......... 45 2 1 Baseline demographics ................................ ................................ ...................... 62 2 2 Number of pati ents classified as pre diabetic by fasting glucose and/or 2 hour post load of 75 gram glucose solution ................................ ........................ 63 2 3 Effect of nine weeks of antihypertensive monotherapy, then additional nine weeks with second antihypertensive agent, on glucose and insulin. .................. 64 2 4 Surrogate markers for insulin resistance indexes at each visit based on calculatio ns from fasting glucose values or OGTT values ................................ .. 67 3 1 Baseline demographics, clinical and laboratory parameters. Mean + standard devi a tion or percent as appropriate ................................ ................................ .... 81 3 2 Change in glucose and length of days of media tion therapy. Mean + standard deviation. ................................ ................................ ................................ ............ 82 3 3 Change in glucose, by assignment and medication order. Mean + standard deviation, full cohort. ................................ ................................ ........................... 83 3 4 Univariate analysis for medication associated change in glucose, derivation cohort. ................................ ................................ ................................ ................ 84 3 5 Predictors for atenolol associated change in glucose in derivation cohort and evaluation of predictors in validation cohort. ................................ ....................... 85 3 6 Predictors for hydrochlorothiazide associated change in glucose in derivation cohort and evaluation of predictors in validation cohort. ................................ ..... 86 3 7 Correlation between model predicted atenolol associated change in glucose, in validation cohort, to actual atenolol associated change in glucose. ................ 87 3 8 Correlation between model predicted HCTZ associated change in glucose, in validation cohort, to actual drug associated change in glucose. ......................... 89

PAGE 9

9 4 1 Candidate genes for adverse glycemic effects associated with use of thiazides diuretics and beta blockers. ................................ ............................... 109 4 3 Power to detect glucose differences across MAF ranges by race. ................... 111 4 4 Baseline demographics. ................................ ................................ ................... 112 4 5 Association between SCNN1G rs13306654 and rs4499239 genotype and change in glucose after treatment with H CTZ among PEAR blacks. ................ 114 4 6 Association between NOS3 rs3800787 genotype and change in glucose after treatment with HCTZ among PEAR blacks. ................................ ...................... 115 4 7 Association between IRS1 rs1801278 genotype and change in glucose after treatment with ATEN or HCTZ in PEAR non blacks. ................................ ........ 116 A 1 Summary data for glucose, log glucose, insulin and log i nsulin area under the curve during a two hour oral glucose tol erance tes t ................................ ......... 124 A 2 Fasting glucose and insulin co llected during OGTT & fasting glucose and insulin collected for core laboratory values, restricted to sub study patients. ... 126 A 3 Fasting glucose by treatment arm in PEAR, UF PEAR, UF PEAR OG TT main study collected labs, UF PEA R OGTT sub study collected labs ....................... 127 A 4 Fasting glucose in PEAR, UF PEAR, UF PEAR OGTT main study collected labs, UF PEAR OGTT sub study collected labs ................................ ............... 128 A 5 Absolute change between study visits in glucose in PEAR, UF PEAR, UF PEAR OGTT main study labs, UF PEA R OGTT sub study labs ....................... 129 A 6 Change in glucose in PEAR, UF PEAR, UF PEAR OGTT main study collected labs, UF PEA R OGTT sub study collected labs ................................ 130 A 7 Blood pressure in PEAR, UF PEAR, UF PEA R OGTT sub study ..................... 131 A 8 Heart rate in PEAR, UF PEAR, UF PEAR OGTT sub study ............................. 132 A 9 UF PEAR OGTT sub study stratified by a change in glucose of greater than or equal to 10mg/dL. ................................ ................................ ......................... 133 A 10 Log transformed glucose and insulin AUC at baseline, after nine weeks of monotherapy, then after a dd on therap y ................................ .......................... 136 B 1 Univariate analysis for body size parameters, for medications associated change in glucose, derivation cohort. ................................ ............................... 139 B 2 Stepwise regression, (model generated) predictors for atenolol associated change in glucose in de rivation cohort and evaluation in validation cohort. ..... 140

PAGE 10

10 B 3 Stepwise regression, (model generated) predictors for H CTZ associated change in glucose in derivation cohort and evaluation in validation cohor t ...... 141 B 4 Correlation between model predicted atenolol associated change in glucose, in validation cohort, to actual drug associated change in glucose. .................. 142 B 5 Correlation between model predicted HCTZ associated change in glucose, in validation cohort, to actual drug associated change in glucose ................... 144 B 6 Full model, all variables forced, for medication associated change in glucose, derivation cohort. ................................ ................................ .............................. 146 B 7 Predictors for atenolol associated change in glucose in derivation cohort and eval uation of predictors in validation cohort ................................ ...................... 147 B 8 Predictors for hydrochlorothiazide associated change in glucose in derivation cohort and e valuation of predictors in validation cohort ................................ .... 148 B 9 Correlation between model predicted atenolol associated change in glucose, in validation cohort, to actual atenolol associated change in glucose. .............. 149 B 10 Correlation between model predicted HCTZ associated change in glucose, in validation cohort, to act ual drug associated change in glucose. ....................... 151 C 1 Baseline demographics with p values. ................................ ............................. 154 C 2 Pharmacological candidate genes an d SNPs. ................................ .................. 155 C 3 Insulin signaling genes and SNPs. ................................ ................................ ... 157 C 4 Results from pharmacological candidate genes, functional SNPs. ................... 159 C 5 Results from SNPs in beta blocker pharmacological candidate genes. ............ 160 C 6 Results from SNPs in thiazide diuretic pharmacolog ical candidate genes. ...... 164 C 7 Results from SNPs in insulin signaling candidate genes. ................................ 170 C 8 Initial findings with and withou t PC1 and PC2 as covariates. All findings controlled for baseline glucose. ................................ ................................ ........ 192 C 9 Association between NOS3 rs3800787 genotype and change in glucose after H CTZ treatment among PEA R blacks ................................ .............................. 193 C 10 Association between IRS1 rs1801278 genotype and change in glucose after treatment with ATEN or HCTZ in PEAR non blacks ................................ ......... 194 C 12 Association between NOS3 rs3800787 genotype and change in glucose after treatment with HCTZ among PEAR blacks. ................................ ...................... 196

PAGE 11

11 C 13 Association between IRS1 rs1801278 genotype and change in glucose after treatment with ATEN or HCTZ in PEAR non blacks. ................................ ........ 197 C 14 Association between INSR rs7508679 genotype and change in glucose after treatment with ATEN monotherapy. ................................ ................................ .. 198 C 15 Association between INSR rs7508679 genotype and change in glucose after ATEN treatment among PEAR blacks. ................................ ............................. 199 D 1 Association between INSR rs7508679 genotype and change in glucose after treatment with ATEN monotherapy. ................................ ................................ .. 207 D 2 Change in glucose of PEAR patients selected for INSR expression study. ...... 209 D 3 Analysis of RT T by INSR rs7508679 genotype. ....................... 210

PAGE 12

12 LIST OF FIGURE S Figure page 1 1 Progression along the glucose continuum to type 2 diabetes mellitus. IFG, impaired fasting glucose; IGT, impaired glucose to lerance. ............................... 40 2 1 PEAR sub study design. OGTTs were incorporated at visit 1, 2 and 3 for the sub study if participants followed the normal progression ................................ 60 2 2 PEAR sub study enrollment. ................................ ................................ ............... 61 2 3 Glucose and insulin median, interquartile ran ge and outliers at each time point ................................ ................................ ................................ .................... 65 2 4 Mean and standard deviation of log transformed AUC ................................ ....... 66 3 1 Final PEAR dataset for evaluating clinical predictors of medication associated increase in glucose. ................................ ................................ .......... 80 3 2 Correlation between model predicted atenolol associated change in glucose, in validation cohort, to actual drug associated change in glucose. ..................... 88 3 3 Correlation between model predicted HCTZ associated change in glucose, in validation cohort, to actual HCTZ associated change in glucose. ....................... 90 4 1 Many genes are involved in insulin signaling and maintaining glucose homeostasis ................................ ................................ ................................ ..... 108 4 2 Final PEAR dataset for evaluating the impact of genetic variation on medication associated dysglycemia. ................................ ................................ 113 4 3 Change in glucose after atenolol monotherapy and hydrochlorothiazide monotherapy by IRS1 rs1801278 genotype in PEAR non blacks ..................... 117 A 1 Glucos e at each time point for each participant ................................ ................ 134 A 2 Insulin at each time point for each participant ................................ ................... 135 B 1 Correlation between model predicted atelolol associated change in glucose, in validation cohort, to actual drug associated change in glucose. .................. 143 B 2 Correlation between model predicted HCTZ associated change in glucose, in validation cohort, to actual drug associated change in glucose. ...................... 145 B 3 Correlation between model predicted atenolol associated change in glucose, in validation cohort, to actual drug associated change in glucose. ................... 150

PAGE 13

13 B 4 Correlation between model predicted HCTZ associated change in glucose, in validation cohort, to actual HCTZ associated change in glucose. ..................... 152 D 1 Change in glucose after atenolol monotherapy by INSR rs7508679 genotype in PEAR blacks and non blacks. ................................ ................................ ....... 208 D 2 Relative INSR expression by rs7508679 genotype in leukocytes, in 16 black PEAR patients ran domized to receive atenolol ................................ ................. 211 D 3 Relative INSR expression by rs7508679 genotype in leukocytes, in 15 non black PEAR patients randomized to receive atenolol ................................ ....... 212 D 4 Relative INSR expression by rs7508679 genotype in leukocytes, in 31 PEAR patients rand omized to receive atenolol ................................ .......................... 213

PAGE 14

14 LIST OF ABBREVIATION S ACE inhibitor Angiotensin converting enzyme inhibitor ADA American Diabetes Association ALLHAT Antihypertensive and Lipid Lowering Treatment to Prevent Heart Attac k Trial AMEs A dverse metabolic effects ARB Angiotensin receptor blocker ASCOT BPLA Anglo Scandinavian Cardiac Outcomes Trial Blood Pressure Lowering Arm ATEN Atenolol AUC Area under the curve BL Baseline BMI Body mass index BP Blood pressure BSA Body su rface area bpm B eats per minute CCB Calcium channel blocker CI Confidence interval CVD Cardiovascular disease DBP Diastolic blood pressure DIAGRAM Diabetes Genetics Replication and Metaanalysis DGI Diabetes Genetics Initiative DREAM Diabetes Reduction Assessment with Ramipril and Rosiglitazone trial eGFR estimated glomerular filtration rate FG Fasting glucose

PAGE 15

15 FUSION Finland United States Investigation of NIDDM Genetics GenHAT Genetics of Hypertension Association Treatments GERA Genetic Epidemiology of Responses to Antihypertensives study GWAS Genome wide association study HbA1c Glycated hemoglobin (type A1c) HCTZ Hydrochlorothiazide HDL High density lipoprotein HOMA Homeostatic model assessment HR Hazard ratio HTN Hypertension IFG Impaired fasting gl ucose IGT Impaired glucose tolerance INVEST International Verapamil Slow Release (SR) Trandolapril study JNC Joint National Committee on Prevention, Detection, Evaluation, and Treatment of High Blood Pressure LD Linkage disequilibrium LDL Low density lipo protein MAF Minor allele frequency MDRD Modification of Diet in Renal Disease equation for estimating renal function mg/dL Milligram per deciliter mmHg Millimeters of mercury NAVIGATOR Nateglinide and Valsartan in Impaired Glucose Tolerance Outcomes Re search study NHLBI National Heart, Lung and Blood Institute OGTT Oral glucose tolerance test

PAGE 16

16 OR Odds ratio PC Principal component PCA Principal component analysis PE Parameter estimate PEAR Pharmacogenomic Evaluation of Antihypertensive Responses trial P PARG Peroxisome proliferator activated receptor gamma QUICKI Quantitative insulin sensitivity check index RR Relative risk SBP Systolic blood pressure SD Standard deviation SNP Single nucleotide polymorphism TC Total cholesterol TG Triglycerides Tx Trea tment WT CCC Wellcome Trust Case Control Consortium

PAGE 17

17 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 CLINICAL AND GENE TIC PREDICTORS OF THE DYSGLYCEMIC EFFECTS OF ANTIHYPERTENSIVE MEDICATION S By Mariellen J. Moore December 2012 Chair: Julie A. Johnson Major: Pharmaceutical Sciences The top five classes of guideline recommended antihypertensive agents promote markedly dissimilar effects on glucose and development of type 2 diabetes mellitus While ACE inhibitors and ARBs may exert beneficial glycemic effects, beta blockers and thiazide diuretics are notable for their propensity to elevate glucose and are associated wit h diabetes development. In a pilot study, we investigated two methods for evaluating pre diabetes fasting glucose and a 2 hour glucose level after an oral glucose tolerance test, in untreated hypertensives prior to and after treatment with a beta blocker (atenolol), thiazide diuretic ( hydrochlorothiazide [ HCTZ ] ), and the combination. We found fasting glucose and an oral glucose tolerance test similar in detecting medication associated pre diabetes. In a larger cohort, we then assessed the clinical and lab oratory parameters associated with elevations in glucose after treatment with atenolol or HCTZ. We found that for both drugs baseline glucose was the strongest predictor for medication associated elevation in glucose in both derivation and validation coho rt s To evaluate possible genetic factors involved in medication associated dysglycemia, we studied candidate genes important for drug response and insulin

PAGE 18

18 signaling. We found significant associations in blacks treated with HCTZ with the blood pressure re sponse genes SCNN1G and NOS3 Further, we found additional associations in both drugs in an insulin signaling gene (atenolol and HCTZ monotherapy non blacks in IRS1) To our knowledge, this study addressed many clinical concerns not previously answered i n the literature The current data suggest fasting glucose is an appropriate phenotype to measure antihypertensive associated adverse glycemic changes baseline glucose is the strongest predictor for antihypertensive associated elevations in glucose and of the genes selected, both blood pressure response genes and insulin signaling genes may play a role in antihypertensive associated change in glucose. However, further research is necessary to fully understand the mechanism and implications of antihypertens ive medication associated dysglycemia.

PAGE 19

19 CHAPTER 1 INTRODUCTION AND BACKGROUND Cardiovascular Disease, Hypertension and Diabetes Cardiovascular disease is endemic in the United States, currently affecting m ore than 82 million adults in America 1 Furthermore, c ardiovascular disease is the primary cause of mortality in the United States, responsible for one in three deaths in 200 8 1 Cardiovascular diseases include the most common medical conditions in industrialized nations and are becoming an increasing problem in developing nations. Prior to 1900, malnutrition and infectious d isease were the most common cause of death worldwide, with less than 10% of all deaths attributed to cardiovascular disease. Currently, cardiovascular disease accounts for approximately 30% of deaths worldwide, including nearly 40% in high income countries and about 28% in low and middle income countries. 2 4 Cardiovascular disease s include coronary heart disease (including acute coronary syndromes and angina ) heart failure stroke and congenital cardiovascular defects. While all of these diseases lead to significant morbidity and mortality hypertension, coronary heart disease, and stroke are among the 15 leading conditions that cause disabilities in the United States Hype rtension defined as systolic blood pressure > 140 mmHg or diastolic blood pressure > 90 mmHg, affects one third of adults in America or more than 76 million adults > 20 year s of age 1 (Definitions of key terms are listed in Table 1 1.) Hypertension is frequently associated with other cardiovascular disease risk fa ctors and the quantity of risk factors increases cardiovascular disease risk. 5 The presence of hypertension doubles the risk of other cardiovascular diseases such as stroke, heart failure and coronary heart disease In 200 8 mortality attributed to

PAGE 20

20 hypertension was ap proximately 61 ,000 and was the underlying cause of mortality in more than 347 ,000 deaths 1 While t he mortality rate directly associated with hypertension may appear low considering the prevalence, the aftermath associated with this disease is significant. Almost 70% of people experiencing their first myocardial infarction, 77 % of those experiencing their first stroke and 74 % of those with heart failure have blood pre ssure greater than 140mmHg/90mmHg 1 The presence of hypertension is associated with a shorter life expectancy and more years lived with cardiovascular disease. 1 As compared to hypertensive 50 year old men, men with untreated blood pressure under 140mmHg/90mmHg survived an average of 7.2 years longer without cardiovascular dise ase and spent 2.1 fewer years of life with cardiovascular disease 6 Hypertension occurs frequently with disorders such as dyslipidemias, insulin resistance and obesity. When these disorders cluster together, the risk for stroke, diabetes, coronary heart disease, and card iovascular disease are further elevated. 1 Furthermore, insulin resistance is also associated with elevation of detrimental endothelial mediators that regulate coagu lation, fibrinolysis, platelet aggregation and vessel tone. 7 The term metabolic syndrome is used to classify the grouping of insulin resistance, hypertension, dy slipidemia and obesity. 8 The c ardiovascular risk associated with diabetes (type 2 diabetes mellitus) is significant. In fact, the impact of diabetes on atherosclerotic disease is so strong that the National Cholesterol E ducation Program, which provided the most recent guidelines for cholesterol testing and management, considers diabetes a coronary hear t disease risk equivalent. 8 Furthermore cardiovascular risk increases wi th increasing blood glucose.

PAGE 21

21 Pre diabetes, when an individual has blood glucose levels higher than normal (fasting glucose > 99mg/dL) but not yet diabetic (fasting glucose <126mg/dL) increases risk of future development of diabetes, heart disease and strok e. 9 Approximately 8.3% of the U.S. population has diabetes, or m ore than 25 million Americans. D iabetes affects about a quarter of the United States population > 65 years old. 9 Pre diabetes is even more prevalent, affecting 35% of Americans > 20 years old, or 50% of Americans > 65 years old. D iabetes imparts significant morbidity and mortality. In the United States, diabetes was the seventh leading c ause of death in 2007 listed as an underlying cause of death in more than 71,000 cases. 9 Importantly, the risk of death among diabetics is double that of non diabetics of the same age. Significant mor bidity is also associated with diabetes, including nervous system diseases, kidney disease and blindness. Diabetes is the leading cause of kidney failure and new cases of blindness in adults in the United States. 9 Up to 70% of diabetics have some form of nervous system damage, which is a major contributor to lower extremity amputations. Heart disease, stroke and hypertension are common comorbidities in diabetics. From 2005 to 2008, 67% of diabetics wer e hypertensive. In people > 65 years old, almost 70% of diabetes related deaths in 2004 were also associated with heart disease. The occurrence of heart disease and risk of stroke in diabetics is two to four times higher than in non diabetics. 9 Considering the impact of both hypertension and type 2 diabetes mellitus on cardiovascular morbidity and mortality appropriate treatment for both diseases is critical Yet for hypertension and diabetes, the disease state and pharmacotherapy selection are interrelated. It is well established that p ersons with hypertension have a high prevalence of diabetes. 9 W hile there are many options w hen considering blood

PAGE 22

22 pressure lowering medications, b eta blockers and diuretics have a well established role in treating hypertension sp anning many decades and are among guideline recommended first line agents. 5 However, g rowing evidence suggests beta blockers and diuretics have harmful metabolic effects specifically on increasing glucose or incident diabetes. The very c linical trials that strongly support diuretic and beta blocker use for blood pressure control and cardiovascular risk reduction 10 13 also report findings of adverse metabolic effects (AMEs) on glucose, insulin, and lipids 14 16 as well as an increased risk for the development of diabetes 17 19 Yet AMEs are only seen in a portion of diuretic and beta blocker treated patients and risk factors for development of AMEs are poorly understood. Interr elationship of Hypertension, Diabetes and Antihypertensive Agents The relationship between insulin resistance and primary hypertension is complex. Metabo lic syndrome, the clustering of hypertension central adiposity, d yslipidemia and dysglycemia, is increasingly being viewed as a disease of insulin resistance with many contributing factors 8 ( Figure 1 1 depicts progression from euglycemia to overt diabetes, and includes predisposing factors. ) Furthermore, persons with hy pertension have a high prevalence of insulin resistance, metabolic syndrome and ensuing diabetes Data from observational studies have shown that hypertension increases risk for diabetes compared to those without hypertension 18 Studies have shown that not only does hypertension worsen insulin resistance 20 but insulin resistance and hyperinsulinemia can predispose a person to become hypertensive 21 22 The glycemic effects of the five most commonly prescribed antihypertensive drug classes are of growing interest to the medical community. Data suggest that ACE inhibitors and ARBs protect against diabetes development, calcium channel blockers

PAGE 23

23 (CCBs) appear to be neutral or slightly protective, and thiazide diuretics and beta blockers may pose a risk. 23 Concern over both beta blockers and thiazide diuretics causing AMEs is well documented and since the last hypertension guideline revision, large scale studies and meta analyses have published compelling data that suggest be ta blockers and thiazide diuretics are not without risk. 17 23 26 More than 40 years of clinical trial d ata support the widespread use of thiazide diuretics for the treatment of hypertension Yet, concern regarding thiazide induced dysglycemia prompted the National Heart, Lung and Blood Institute (NHLBI) of the National Institutes of Health to issue a call f or additional research on the link between potassium and thiazide associated dysglycemia, the different methods to correct for diuretic associated dysglycemia, and whether populations at greatest risk for thiazide associated dysglycemia can be identified 27 Many mechanisms may be important in thiazide associated dysglycemia. Some have hypothesized that diuretic associated hyperglycemia is related to potassium deplet ion 28 decreased insulin secretion 29 upregulation of the renin angiotensin system, 30 31 and/or decreased peripheral insulin sensitivity. 32 (Mechanisms and site of action where beta blockers and thiazide diuretics may increase glucose are depicted in Figure 1 2. ) The importance o f potassium depletion in thiazide associated dysregulation has been proposed 33 36 and an inverse relationship between potassium and glucose following thiazide t reatment has been observed in study level data 30 37 A possible mechanism involves hyperpolarization of pancre atic beta cells by the thiazide induced opening of calcium activated potassium channels. 38 Considering thiazide diuretics can open calcium a ctivated potassium channels in vascular smooth muscle, 39 the same is likely for pancreatic beta cells. When

PAGE 24

24 hyperpolari zed, calcium influx is inhibited and subsequently insulin release is also limited. 38 It has been suggested that hyperglycemia may be improve d with potassium supplementation 40 however this theory has not been prospectively proven. 41 While the potassium mediated dysglycemia theory is still under debate in the scientific community, this theory is based on small, retros pective, meta analyses or observational studies. 30 33 36 42 Furthermore, a PEAR 43 trial (Pharmacogenomic Evaluation of Antihypertensive Responses) subgroup analysis in 202 hypertensives receiving HCTZ monotherapy for nine weeks does not support a correlation between serum potassium and serum glucose or insulin. 44 Considering the sympathetic nervous system and renin angiotensin system are simulated by diuretic induced lowering in blood pressure, additional mechanism may also be important in elevations in glucose. 45 However, whether hyperglycemia is due to direct diuretic actions on blood volume, decreasing perfusion to skeletal muscle, or activation of compensatory systems, is unknown. 27 Thus the mechanism for thiazide associated increase in glucose remains controversial and unclear. In the Antihypertensive and Lipid Lowering Treatment to Prevent Heart Attack Trial (ALLHAT), diabetes developed in 4 9 % of patient s after two years depending on treatment arm with a rate of diabetes development of 9.3% in the chlorthalidone group, 7.2% in the amlodipine group and 5.6% in the lisinopril group 10 46 Thiazide diuretic treatment in ALLHAT resulted in a two to three mg/dL higher fasting glucose level compared to other medication classes after two years of antihypertensive treatment W hile subsequent analysis in ALLHAT did not find an increased risk for cardiovascular outcomes associated with incident thiazide associated diabetes, 14 47 it is possible that a

PAGE 25

25 three year follow up is insufficient. In a hypertensive cohort followed for up to 16 years, incident diabetes associated with thiazide diuretic use was linked with increased cardiovascular di sease risk. 48 Even after adjustment for blood pressure, the relative risk of events in incident diabetics was 2.92 (95% CI: 1.33 6.41), which was similar to th e risk in those with diabetes at study entry (RR 3.57, 95% CI: 1.65 7.73). 48 Furthermore, significant cardiac morbidity such as heart failure, has been associ ated with new onset diabetes that develops while treated with antihypertensive agents. 49 Whether incident diabetes imparts an equivalent CV risk as prevalent diab etes, 48 or a more intermediate risk 49 has not been established, i t is clear antihypertensive associated diabetes development is not benign. A network meta analysis of 22 long term clinical trials, analyzing 143,153 patients, assessed the risk of incident diabetes across the five major drug classes. 23 The risk for incident diabetes in patients treated with a beta blocker verses patients treated with a diuretic was not found to be significantly different ( OR 0.93, 95% CI: 0.78 1.11 p=0. 43 ). Importantly, treatment with plac ebo posed a significantly lower risk for incident diabetes compared to diuretic treatment (OR 0.75, 95% CI: 0.63 0.89, Table 1 2 ). As expected, ARBs and ACE inhibitors showed a decreased risk for incident diabetes compared to diuretic (OR 0. 62 95% CI: 0.5 1 0.77 ; OR 0. 67 95% CI: 0. 57 0.79 respectively). 23 Considering beta blockers, the mechanism of the h yperglycemic effects of beta blockers may be explained through diminished pancreatic beta cell insulin release increased body weight, or reduced peripheral glucose utilization (Figure 1 2) 50 52 Beta blockers also modify metabolism of carbohydrates and lipids, potentially via blocking glycogenolysis and attenuating release of fatty acids prompting additional harmful

PAGE 26

26 downstream changes such as weight gain and dyslip idemia. 53 54 Additionally, b eta blockers may promote hyperglycemia by inhibit insulin secretion by beta 2 bl ockade on the pancreas 25 or through promoting weight gain. 55 56 While non selective beta blockers or higher doses of selective beta blockers result in the largest adverse metabolic changes, agents that block both beta and alpha receptors ( such as carvedilol) minimally induce glycemic changes. 54 A likely mechanism involves increased peripheral vasoconstriction second ary to unopposed alpha activity with beta blocker use, decreasing blood flow and thus glucose available to skeletal muscle. 57 Various meta analyses have evaluated incident diabetes in beta blocker trials, 23 25 general ly reporting similar findings on the propensity of beta blockers to increase risk for diabetes development. One meta analysis by Bangalore and colleagues found that beta blockers impart a 22% increased risk for diabetes development. 25 This meta analysis reported a greater risk for developing diabetes in those with a higher body mass index, baseline fasting glucose, final blood pressure difference and those > 60 ye ar of age. The authors conclude d that if 1,000 patients were treated with beta blockers for 4.4 years, approximately fourteen excess cases of diabetes will result. The authors further stated that given that 65 million Americans have hypertension, this woul d result in 910,000 excess cases of diabetes. Based on updated data on the incidence of hypertension and the calculations, with 82 million hypertensive Americans 1 almost 1.2 million excess cases of diabetes would result. Renin angiotensin system antagonists, including angiotensin converting enzyme (ACE) inhibitors and angiotensin receptor blockers (ARBs), have been shown to reduce diabetes development in n umerous cardiovascular outcome and blood pressure studies

PAGE 27

27 and confirmed in multiple meta analyses. 58 60 On top of standard cardiovascular therapy, treatment with an ACE inhibitor or angiotensin receptor blocker reduced new onset diabetes by approximately 20%. 61 Inhibition of the renin angiotensin system improves insulin sensitivity and pancreatic beta cell function. 57 The Diabetes Reduction Assessment with Ramipril and Rosiglitazone (DREAM) trial sought to evaluate the ability o f an ACE inhibitor to prevent diabetes in people with cardiova scular disease or hypertension. 62 More than 5,000 patients without CV disease but considered pre diabetic were randomized to receive ramipril or placebo. The investigators fo und those receiving ramipril were significantly more likely to regress to normoglycemia. The Nateglinide and Valsartan in Impaired Glucose Tolerance Outco mes Research (NAVIGATOR) study 63 in pre diabetics with established cardiovascular disease or cardiovascular risk factors randomly assigned patients to receive valsartan or placebo More than 9,000 patients were followed for a median of five years for incident d iabetes. In conjunction with lifestyle modification, valsartan reduced the incidence of diabetes by 14%. Calcium channel blockers are generally considered metabolically neutral, either having slightly beneficial effects or not significantly altering gluco se homeostasis. 23 These effects may be due to vasodilation and enhanced blood flow to the muscle as wel l as decreasing the cytosolic free calcium concentrations. 64 The International Verapamil Slow Release (SR) Trandolapril Study (INVEST) found that in hypertensive patients with coronary artery disease a CCB/ACE inhibitor strategy significantly reduced risk of new onset diabetes compared to a beta blocker/thiazide diuretic strategy. These effects in the calcium channel blocker treated patients were redu ced

PAGE 28

28 with the addition of a thiazide diuretic. 65 In the Anglo Scandinavian Cardiac Outcomes Trial Blood Pressure Lowering Arm (ASCOT BPLA) 26 the com bination of an ACE inhibitor/CCB no t only lowered incident diabetes by 30%, but also conferred greater glycemic control versus those treated with a beta blocker/diuretic regimen. Accordantly, while data are still amassing about the glycemic effects of the five major antihypertensive classes, much is still unknown about the mechanisms involved Use of Beta blockers a nd Diuretics in Hypertension a nd Dysglycemia Of availabl e blood pressure medication classes, beta blockers and diuretics are guideline (JNC V II ) recommended initial therapy 5 based on comorbidities. Thiazide type diur etics have proven efficacious in preventing cardiovascular complications of hypertension and as such are recommended by JNC VII as initial therapy for most patients, depending on compelling indications. 5 Considering the myriad of antihypertensive options, choice is often based on expected blood pressure reduction without regard to possible AMEs. However, c omplicating drug selection, growing evidence exists that th e five first line antihypertensive agents exhibit widely differing effects on glucose homeostasis 23 50 66 JNC VII recommendations concerning dysgycemia note d the importance of hypertension control in persons with diabetes and established lower bloo d pressure goals for these patients. Additionally, JNC VII identifi ed diabetes as a compelling indication for specific medication classes for diabetes. While the propensity for beta blockers and thiazide diuretics to worsen glucose homeostasis is addressed, JNC VII states the effects tend to be small and are not an absolu te contraindication. Additionally, t he American Diabetes Association revised the hypertension section in their annually published recommendations 67 in treating diabetic patients to emphasize the importance of achieving blood pressure

PAGE 29

29 go als and individualizing therapy. However, the glycemic effects of various antihypertensive therapies were not addr essed and no distinction was made in antihypertensive therapy recommendations for patients at high risk for developing diabetes Based on either guideline recommendation, practitioners might not fully appreciate the potential glycemic consequences of these antihypertensive medications. Lack of Clinical Predictors Associated with Beta blocker and Diuretic Associated Adverse Metabolic Effects Identifying easily measurable clinical predictors is the initial step in evaluating patients at risk for developing A MEs, insulin resistance and diabetes from thiazide diuretics and beta blockers Diabetes and pre diabetes may be evaluated through multiple methods; current guidelines from the American Diabetes Association for diagnosing diabetes and pre diabetes are list ed in Table 1 3 P rior to diabetes develo pment, impaired fasting glucose and impaired glucose intolerance (both considered pre diabetic classifications) may be present, though not always easily detectable in typically collected fasting laboratory values. 68 71 The insulin resistant, pre diabetic phenotype may be difficult to detect until the patient has progressed to overt, symptomatic diabetes. Many individuals that are diagnosed as diabetic through one method, such as having impaired glucose tolerance after an oral glucose tolerance test ( OGTT ) are euglycemic in their daily lives. Furthermore, persons with impaired glucose tolerance may have normal or near nor mal glycated hemoglobin levels and often manifest hyperglycemia only when challenged with an oral glucose load used in an OGTT. 67 Early in the spectrum of dysglycemia, insulin secretion is upregulated to maintain eu glycemia. After increasing periods of hyperinsulinemia target organs, namely adipose tissue, skeletal muscle and the liver, will eventually cease to dispose of

PAGE 30

30 glucose appropri ately. It is at this point, when hepatic insulin sensitivity appears normal, but peripheral insulin resistance is escalating t hat fasting glucose is within normal range, but resp onse to a n OGTT is abnormal 72 The OGTT is considered more sensitive and mo derately more specific than fasting glucose in diagnosing diabetes. 67 The OGTT is relatively simple to incorporate into clinical practic e, yet not routinely performed in the clinic due to time limitations, inconvenience and cost. 67 Furthermore, t he pre diabetic phenotype detected via an OGTT, impaired glucose tolerance, has been associated with increased CV risk and endothelial dysfunction, a marker of macrovascular disease. 73 74 Thus, OGTT data potentially provide different insight than fasting glucose, classifying a greater proportion of patients as hav ing pre diabetes Many methods to evaluate insulin resistance through surrogat e indexes have been developed, using glucose and insulin from fasting or OGTT results. 75 While the most frequently used methods correlate well with the gold stan dard the euglycemic hyperinsulinemic clamp, for d etermining insulin sensitivity, each method has limitations. 72 The homeostatic model assessment (HOMA 76 ) eva luates insulin resistance using fasting glucose and insulin plasma concentrations (T able 1 4 ) HOMA is a simple calculation and is a frequently used assessment of insulin sensitivity clinically. 77 80 Yet HOMA may classify a person with pre diabetes as normal (based on fasting glucose), whereas if evaluated by an OGTT, impaired glucose tolerance may be evident 72 81 The quantitative insulin sensitivity check index (QUICKI 82 ) is simil ar to HOMA in that fasting glucose and insulin values are incorporated but takes the logarithm and reciprocal of the fasting glucose insulin product. This method is considered more accurate than

PAGE 31

31 HOMA over a broad range of insulin sensitivities, has been u sed in large population studies, but still only incorporates fasting glucose and insulin concentrations into the calculation Finally, the Matsuda 83 calculation uses OGTT data and is a method for evaluating whole body insulin sensitivity that correlates well with the clamp method and provides a comprehensive assessment, yet requires more time and effort for both patie nts and practitioners. These indices are among the most commonly incorporated surrogate markers for insulin resistance in clinical studies, with a strong correlation to the euglycemic hyperinsulinemic clamp. Yet, while all these calculations are attempting to predict insulin resistance, they are not equally precise and each is based on underlying assumptions. Calculations based on fasting data (HOMA and QUICKI) are only able to provide an estimation based on hepatic insulin sensitivity, assuming that hepati c and peripheral insulin sensitivity are equal, which is inaccurate. 72 While the QUICKI calculation is considered more accurate than HOMA, the accuracy of QUICKI to evaluate mild insulin resis tance in those with normal glucose tolerance is poor. While limitations exist in using OGTT data for the Matsuda index, this me thod posses ses both greater accuracy over fasting glucose and greater acceptability over the euglycemic hyperinsulinic clamp. 72 Though no i nsulin sensitivity calculations is comparable to the clamp method, data collected from an OGTT provides a detailed assessment of glucose tolerance and insulin sensitivity and as such is an excellent method to incorporate into clinical investigations. Further though no equation is perfect in evaluating insulin resistance, all are useful and provide insight beyond fasti ng glucose or insulin levels alone. In addition, an OGTT offers valuable data in that the information can be used to evaluate patients with the various indexes for assessing insulin sensitivity.

PAGE 32

32 Despite available measures to evaluate insulin resistance in the clinic and laboratory, such as fasting glucose, OGTT data, and HOMA, QUICKI, and Matsuda calculations studies are lacking on predictive parameters to assess risk for medication associated AMEs in a population with primary hypertension Using a descri ptive phenotype such as information obtained from an OGTT, coupled with baseline clinical parameters and demographics may lead to improved characterization of clinical parameter s associated with antihypertensive medication associated adverse glycemic effe cts Clinical Predictors of Risk for Antihypertensive Medication Associated Dysglycemia Data is growing evaluating the relationship of clinical characteristic and dysglycemic effects of antihypertensive mediations Abdominal obesity age, female sex, mino rity ethnicity, elevated fasting and postprandial glucose, low levels of high density lipop rotein, left ventricular hypertrophy, and the degree of elevation of systolic and diastolic blood pressure have been associated with adverse metabolic effects accomp anying antihypertensive therapy. 17 23 46 84 85 An analysis in the INVEST trial sought to determine clinical characteristics of incident diabetics among individuals with hypertension and coronary arte ry disease. Including more than 16,000 individuals, the investigators found that United States residence (HR 1.62, 95% CI:1.37 1.91), left ventricular hypertrophy (HR 1.27, 95% CI: 1.10 1.46), previous stroke/transient ischemic attack (HR 1.26, 95% CI: 1.0 3 1.56), Hispanic ethnicity (HR 1.21, 95% CI: 1.05 1.39), coronary revascularization (HR 1.18, 95% CI: 1.03 1.35), hypercholesterolemia (HR 1.17, 95% CI: 1.04 1.31), and greater body mass index ( HR 1.05, 95% CI: 1.04 1.06 ) were associated with an increased diabetes risk. 17 A post hoc

PAGE 33

33 analysis in ALLHAT sought to further evaluate the risk of new onset diabetes associated with cholorthalidone. 46 The investigators found body mass index, higher baseline glucose, and higher systolic blood pressure increased th e odds of diabetes development where as a less consistent relatio n ship with sex and age over time was observed 46 A t midpoint in the PEAR study, a cohort of 395 middle aged patients without diabetes, considered low risk hyper tensive patients, was assessed for AMEs attributable to atenolol and hydrochlorothiazide. 84 The population was categorized based on the presence or absence of abdominal obesity and evaluated for new cases of impaired fasting g lucose, elevated triglycerides, and new occurrence of diabetes. The investigators found that individuals with abdominal obesity had a 20% occurrence of impaired fasting glucose at basel ine, which elevated to a 40% occurrence of impaired fasting glucose at study conclusion (p<0.0001) The proportion of individuals with triglycerides > 150 mg/dL increased from 33% to 46% by study conclusion (p<0.01). New onset diabetes occurred in 6% of abd ominally obese individuals and 2% of those not abdominally obese. Predictors for incident diabetes in abdominally obese individuals after monotherapy included randomization to receive initial hydrochlorothiazide treatment, female sex, and uric acid. Baseli ne glucose, t r i glycerides, and high density lipoprotein were predictive of adverse metabolic changes 84 Thus, while data on clinical parameters associated with medication associated AMEs is slowly growing, much remains unknown. C lear identification of patients at increased risk for AMEs to these medications would be invaluable for clinicians as these patients are at risk to develop diabetes. 86 87

PAGE 34

34 Genetic Determinants of Type 2 Diabetes Mellitus and Risk for Drug Associated Type 2 Diabetes Mellitus Numerous genome wid e, pathway driven, and candidate gene association studies support type 2 diabetes to be a multifaceted, polygenic disease, 88 90 with a heritability of 16 37% in d izygotic twins. 91 93 Importantly, diabetes is one of the most successful examples where well replicated, disease associated candidate genes were subsequentl y validated by genome wide association studies (GWAS s ) 94 97 GWAS s have led to the discovery of at least 38 single nucleotide polymorphisms ( SNPs ) associated with diabe tes and at least 20 associated with various dysglycemic phenotype s The first diabetes GWAS was conducted in 661 cases (body mass index <30 kg/m 2 first degree family history of type 2 diabetes) and 614 non diabetic controls. Although other genes were not subsequently replicated, novel signals in SLC30A8 (solute carrier, family 30 [zinc transporter], member 8; this gene is highly expressed in the pancreas, particularly in the islets of Langerhans) and HHEX (hematopoietically expressed homeobox; this gene en codes a transcription factor that may be involved in the differentiation and maintenance of hepatocytes ) were later replicated, and the importance of TCF7L2 (transcription factor 7 like 2; this gene encodes a transcription factor that may regulate progluca gon in enteroendocrine cells ) confirmed, thus supporting pre GWAS disease genetics findings. 98 99 Then comb ining the Wellcome Trust Case Control Consortium (WTCCC), the Finland United States Investigation of NIDDM Genetics (FUSION), and the Diabetes Genetics Initiative (DGI) formed the Diabetes Genetics Replication and Metaanalysis (DIAGRAM) consortium, led to a greater power to detect common variants with low effect size and further significant contributions 99 The current consortium, DIAGRAM+, includes most diabetes genetics

PAGE 35

35 cohorts, with a sampl e size of more than 22,000 persons of European ancestry. In the most recent meta analysis, nine novel signals were reported and signals previously reported in smaller cohorts were confirmed, including IRS1 (insulin receptor substrate 1, encodes IRS1 protei n, a docking protein important in insulin signaling that is activated by the insulin receptor) MTNR1B (melatonin receptor 1B, MTNR1B is a subtype of melatonin receptor and is expressed in pancreatic islets) and KCNQ1 ( potassium voltage ga ted channel, KQT like subfamily member 1 encodes a potassium channel important for repolarization in cardiac muscle, but is also expressed in adipose tissue and the pancreas ) 1 00 105 As GWAS diabete s data continued to grow, assessment of more detailed physiological parameters l ed to both discovery of new variants and further support that genes previously associated with diabetes were also associated with related traits such as fasting glucose, glucose 2 hours post OGTT, HOMA calculations, and fasting insulin. 99 102 103 106 Considering candidate gene studies, w ell replicated pharmacogenomic studies indicate there are genetic markers that are associated with metabolism, respons e or adverse effects of diabetic medications. 106 110 Genetic variation in the organic cation transporter 1 ( SLC22A1 solute carrier family 22 member 1, this gene enco des the organic cation transporter OCT1, an influx transporter that interacts with many drugs ) was found to influence the pharmacokinetics of metformin 107 In 93 no n diabetic Hispanic women, variants in peroxisome proliferator activated receptor gamma ( PPARG peroxisome proliferator activated receptor gamma, encodes a member of the PPAR subfamily of nuclear receptors, regulates adipocyte differentiation ) were associa ted with response to thiazolidinediones. 108 Variation in renin, endothelin 1, and

PAGE 36

36 genes in the epithelial sodium channel pathway may be important in edema assoc iated with PPARG agonist. 109 110 Thus, many genes with a strong biological basis whether important for phar macokinetics, pharmacodynamics, or glucose regulation pathways, have shown to be significant in diabetes pharmacogenomics. As may be expected, there are few studies evaluating the genetic determinants of medication associated dysglycemia. However, a handf ul of studies have evaluated this phenotype retrospectively in large hypertension trials. The Genetic Epidemiology of Responses to Antihypertensives (GERA) study, a n ambulatory practice cohort of more than 500 adults with hypertension, underwent monotherap y with HCTZ for four weeks; change in plasma total cholesterol, triglycerides, and glucose were evaluated before and after HCTZ treatment A total of 19 thiazide response candidate genes were selected which included 45 polymorphisms. Glucose increased by an average 3.5 + 9.5 mg/dL in the population. Ethnicity, baseline glucose, weight, urine sodium excretion and two polymorphisms in KCNJ1 ( potassium inwardly rectifying channel, subfamily J, member 1 ) and ADRB2 (beta 2 adrenergic receptor) were significant predictors for change in glucose. The study concluded that demographic, environmental and genetic factors predict 11% of the variation in glucose response to HCTZ of which genetic predictors accounte d for only 2% of the variation. 111 The GenHAT study included a large proportion of ALLHAT participants with available DNA, of which more than 9,000 were included in this analysis. This group also evaluate d hypertension candidate genes and fasting glucose, analyzing 24 polymorphisms in 11 candidate genes. Only participants without diabetes at baseline were included. After two years of treatment, fasting glucose increased by 6.8mg/dL with chlorthalidone, 4.8 mg/dL with amlodipine, and 3.0mg/dL

PAGE 37

37 with lisinopril. Variation in SCNN1A ( rs2228576 ) and ACE I/D (rs1799725 and rs4291) were found to be associated with altered fasting glucose, but having relatively small effects on glucose variant carriers for rs4291 had an average 1.72mg/dL lower fasting glucose p=0.0014) O nly the signal in ACE (rs4291) met criteria for significance after multiple comparisons 112 While these g enes have been studied for other phenotypes, this is the first to evaluate these genes on glucose phenotypes. Thus, while some investigators have begun evaluating the genetic contribution to antihypertensive associated dysglycemia, much is yet unknown. Th ough researchers have contributed significantly over the preceding decade s in detailing the genetic basis for diabetes, there remains a large portion of unexplained heritability and significant gaps in knowledge To date, genetic variants explain approxima tely 10% of type 2 diabetes heritability. 99 111 112 Additionally, it is unknown whether the genes important in diabetes development versus drug associated diabetes development are the same Potentially, drug associated diabetes or drug associated change in glucose may involve a differ ent group of genes than those for diabetes. Further study on the genetic basis of antihypertensive associated dysglycemia is warranted. Gaps in Knowledge and Study Objectives : Advers e Metabolic Effects Associated w ith Antihypertensive Medications Evidence exists that individuals respond differently to drug therapies. 113 118 based on genetic makeup, coupled with traditional clinical approaches, offers more specialized treatment options for the clinician. 119 While many researchers have studied the ability of beta blockers and diuretics to contribute to AMEs individually, few have

PAGE 38

38 studied the medication classes head to head, or in combination, using contemporary dose s and an OGTT as the phenotype for glucose tolerance. 120 123 We sought to determine if fasting glucose compared to OGTT provide s similar insight in evaluating dy sglycemia associated with these drugs. To further characterize persons most at risk for beta blocker and diuretic associated AMEs, a three step approach was undertaken to further describe patient specific factors linked to changes in glucose. Specifically, phenotypic, clinical, and genetic factors associated with adverse glycemic responses to beta blockers and thiazides diuretics were identified. Patients with essential hypertension are frequently prescribed beta blockers and thiazide diuretics. Considerin g the wide spread use of these medications, the in ability to predict those at risk for medication associated AMEs further complicates drug selection Additional antihypertensive medication associated c hange in glucose We address various aspects of medication associated dysglicemia in this dissertation project. Thus, we have three hypotheses: In a sub population of hypertensive patients treated with a thiazide diuretic or beta blocker, alone or combin ed, evaluate changes in glucose. Data from an OGTT will detect increases in glucose which are not detectable by fasting glucose alone. In a population of hypertensive patients treated with a thiazide diuretic or beta blocker, alone or combined, clinical p arameters are able to predict elevations in glucose resulting from medication exposure In a population of hypertensive patients treated with a thiazide diuretic or beta blocker, alone or combined, variation in pharmacological response genes and insulin sig naling genes are associated with dysglycemia resulting from medication exposure To address o ur first hypothesis, a population randomized to HCTZ or atenolol was evaluated by an OGTT at three time points, prior to drug therapy, after monotherapy,

PAGE 39

39 and after combined therapy. Our study will help evaluate whether data from an OGTT is more sensitive than fasting data in early detection of AMEs in a hypertensive population with normal fasting glucose at baseline. Next to evaluate the second hypothesis, clinical parameters associated with elevations in glucose in an atenolol and HCTZ treated cohort was evaluated. Clinical characterization of patients that develop elevations in glucose associated with these medications may contribute to medication selection in spec ific patients. Lastly, to address the third hypothesis, a candidate gene study was conducted in the same cohort to determine genetic factors associated with dysglycemia from these drugs This genetic association study will not only further mechanistic know ledge of medication associated dysglycemia but in the future may be important in medication selection when genetic information is r eadily available in the clinic. s to expand on the currently sparse literature regarding the dysglycemic effects of antihypertensive medications and potentially support more appropriate and informed hypertension treatment decisions

PAGE 40

40 Figure 1 1. Progression along the glucose continuu m to type 2 dia betes mellitus. IFG, impaired fasting glucose; IGT, impaired glucose tolerance. Figure adapted from LeSalle 124

PAGE 41

41 Figure 1 2. B eta blockers (BB) and thiazide diuretics (TD) impact glucose through many mechanisms. Organ cartoons from Basic & Clinical Pharmacology text, 125 126 pathway information from the literature. 27 30 53 57 127

PAGE 42

42 Table 1 1. Definition of key terms. Term Definition Diabetes For purposes of the present study, type 2 diabetes mellitus is referred to as diabetes This is the most common form of diabetes (90 95% of diabetics) and ranges in phenotype from predominantly in sulin resistant with relative insulin deficiency to predominantly issues with insulin secretion concurrent with insulin resistance. Diabetes is primarily defined by the level of glucose associated with an elevation in risk of microvascular complications. 67 128 Hypertension Blood pressure > 140 mmHg systolic / 90 mmHg diastolic 1 5 Impaired fasting glucose A type of pre diabetes, identified by a fasting glucose level 100 to 125mg/dL. Fasting is defined as no caloric intake f or at leas t 8 hours. 67 Impaired glucose tolerance A type of pre diabetes, identified by a glucose level 2 hours after an OGTT of 140 to 199 mg/ dL. 67 129 Insulin resistance When the body is unable to respond to and use the insulin it produces. 72 130 OGTT A provocation test to determine the efficiency of the body to metabolize glucose and provide information on latent diabete s states. The test is initiated on fasting adult patients. Plasma glucose is drawn prior to and 2 hours after a 75 gram glucose load. 128 129 Pre diabetes An intermediate group of persons with glucose levels higher than normal, but not overtly diabetic. These persons are at an elevated risk of future diabetes development. 67

PAGE 43

43 Table 1 2 Odds ratio for incident diabetes in antihypertensive clinical trials 23 Medication class Odds ratio Standard Deviation ARB 0.62 0.51 0.77 ACE inhibitor 0.67 0.57 0.79 Placebo 0.75 0.63 0.89 CCB 0.79 0.67 0.92 Beta blocker 0.93 0.78 1.11 Diuretic 1.0 --

PAGE 44

44 Table 1 3 American Diabetes Association glucose thresholds for normal gluco se, pre diabetes, and diabetes. 67 Normal Pre Diabetic Diabetic Fasting glucose (mg/dL) < 99 100 125 > 126 2 hour 75 gram OGTT glucose (mg/dL) < 139 140 199 ** > 200 HbA1c (%) < 5.6 5.7 6.4 > 6.5 Random glucose (mg/dL) --> 200 A pre diabetic f asting glucose of 100 125mg/dL is also termed impaired fasting glucose. **A pre diabetic glucose of 140 199mg/dL 2 hours after a 75 OGTT is also termed impaired glucose tolerance.

PAGE 45

45 Table 1 4 F ormulas for surrogate indexes of insulin resistance. Glucose un its mg/dL, IR, HOmeostatic Model Assessment Insulin Resistance; QUICKI, QUantitative Insulin sensitivity ChecK Index; G 0 fasting glucose; I 0 fasting insulin. HOMA IR 76 QUICKI 82 Matsuda 83

PAGE 46

46 CHAPTER 2 ANTIHYPERTENSIVE MEDICATION EXPOSURE AS RISK FOR IMPAIRED GLUCOSE TOLERANCE: A PEAR SUB STUDY Introduction B eta blockers and diuretics are guideline (JNC 7) 5 recommended initial therapy for hypertension along with three other classes, based on comorbidities. Considering the myriad of antihypertensive options, selection of a specific agent is either empiric or based on anticipated blood pressure reduction However, concern over beta blockers and diuretics causing adverse glycemic effects is well documented. 14 19 48 131 Since the publication of JNC 7 large scale studies and meta analyses have summarized compelling da ta that suggest beta blockers and thiazide diuretics drug classes should be used with caution in those already at risk for developing diabetes. 17 23 26 Data from a network meta analysis of 22 long term clinical trials found that placebo pose s a significant ly lower risk for incident diabetes mellitus compared to use of a diuretic (OR 0.75 p=0.00 1 ). Additiona lly, it was shown that calcium channel blockers (OR 0.79 p=0. 004 ) angiotensin converting enzyme inhibitors (OR 0.67, p<0.0001) and angiotensin receptor blockers (OR 0.62, p<0.0001.) decrease risk of incident diabetes versus initial trea tment with a diur etic Furthermore, the same meta analysis reported that the risk for incident diabetes in beta blocker versus diuretic treated patients was not significantly different 23 P rior to diabetes development, gluc ose intolerance and other metabolic abnormalities are apparent. 68 71 The insulin resistant, pre diabetic phenotype is often not detected until the patient has progressed to diabetes 132 Fasting glucose, a di agnostic criterion for metabolic syndrome and diabetes often appears normal in persons with impaired glucose tolerance. 67 Additionally, persons with impaired glucose

PAGE 47

47 tolerance may also have nor mal glycated hemoglobin (HbA1c) levels with hyperglycemia only detectable when challenged with an OGTT 67 133 Impaired glucose tolerance has been strongly associated with all cause, cardiovascular disease and coronary heart disease mortality 134 136 Furthermore, postprandial hyperglycemia has been associated with factors related to atherosclerosis, such as oxidative stress 137 and the atherosclerotic process, such as carotid intima media thickness 138 and end othelial dysfunction 74 139 Yet there exists a lack of studies comparing the ability of OGTT data and fasting gl ucose data with regard to detection of adverse changes in glucose resulting from antihypertensive medications Thus, it is unknown whether a 2 hour OGTT has greater sensitivity versus fasting glucose to detect antihypertensive associated metabolic changes after therapy with beta blockers and thiazide diuretics. C onsidering the greater sensitivity of the OGTT to detect pre diabetes and diabetes versus fasting glucose, 67 133 it is possible that the OGTT is also more sensitive in predicting antihypertensive associated changes in glucose tolerance The OGTT is more commonly performed in research studies than clinical p ractice primarily due to time limitations, inconvenience and cost. 67 While t he hyperinsuli nemic euglycemic clamp remains the gold standard for studies on insulin sensitivity in vivo, this method is not used clinically due to the length of time and significant risk involved. 83 An additional method used to evaluate insulin resistance is th e homeostatic model assessment (HOMA) which incorporates fasting glucose and insulin The HOMA calculation correlates well with insulin sensitivity determined by the hyperinsulinemic euglycemic clamp method 76 yet the calculation has poor sensitivity in evaluating glycemic abnormalities as only fasting values are incorporated, thu s only providing an

PAGE 48

48 estimate of hepatic insulin sensitivity 72 A n OGTT can be highly useful clinically considering the test is minimally invasive and may be performed in the clinic, and the test provides data on both a fasting and stimulated state. Despite available resources to evaluate insulin resistance in the clin ic and laboratory, studies are lacking on predictive biochemical parameters to assess risk for medication associated dysglycemia After exposure to beta blockers and thiazide diuretics, OGTT, fasting glucose or a mathematical model may provide disparate r esults, with one or more method classifying a greater proportion of patients as dysglycemic than the others. Considering the risk of elevations in glucose associated with use of beta blockers and thiazide diuretics, clear and early identification of patien ts most at risk would be invaluable for clinicians. We hypothesize d that, i n a subset of patients treated with beta blockers and thiazide diuretics in the PEAR study adverse changes in glucose would be detectable through an OGTT but not have be en evident from fasting glucose alone Methods Study Population Individuals who participated in this PEAR Sub Study were recruited from the PEAR study at the University of Florida enrollment site PEAR was a prospective, open label, randomized study of atenolol and HCTZ as monotherapy then combination therapy to determine the genetic associations of both antihypertensive response and adverse metabolic effects. The details o f the PEAR study and this PEAR sub study are as follows. PEAR Protocol PEAR (clinicaltrials.g ov identifier: NCT00246519,) is a recently concluded hypertension pharmacogenomics study Details of the PEAR study design have been

PAGE 49

49 described previously 43 Brie fly, recruitment into PEAR began in 2005 and included enrollment of hypertensive participants at the University of Florida (Gainesville, FL), Mayo Clinic (Rochester, MN), and Emory University (Atlanta, GA). A total of 768 patients completed the trial. Elig ibility criteria included patients with mild to moderate hypertension between the ages of 17 and 65, of any race, gender or ethnicity. Participants had newly diagnosed, untreated or known hypertension treated with less than three hyperte nsive medications. Participants with s econdary forms of hypertension isolated systolic hypertension heart rate under 55 beats per minute, known cardiovascular disease known type 1 or type 2 diabetes mellitus, screening fasting blood glucose above 126mg/dL, screening non fasting blood glucose above 200mg/dL, renal insufficiency, primary renal disease, pregnancy or lactation, elevated liver enzymes, or treatment with confounding medications were excluded. Participants treated with antihypertensive medication at study entry had their medication s withdrawn using an appropriate taper as necessary A minimum antihypertensive medication free period of 18 days was required, with a preferred washout of at least four weeks, to give sufficient time for blood pressure values to return to pre treatment levels. After this period, potential study participants were eligible for inclusion in PEAR if their average seated home diastolic blood pressure was greater than 85 mmHg for the preceding week, and average seated office diastolic blood p ressure was greater than 90 mmHg at the study visit. Study participants with a diastolic blood pressure greater than 110 mmHg or systolic blood pressure greater than 180 mmHg were excluded from enrolling. PEAR Sub Study Protocol All participants enrolle d in PEAR after May 2009 at the Gainesville, FL study site were asked by PEAR study personnel about their interest in parti cipation in the PEAR

PAGE 50

50 sub study (clinicaltrials.gov identifier: NCT01099397). If interested, PEAR participants were then contacted by a PEAR sub study investigator and enrolled during a PEAR clinic visit after the participant was determined to meet eligibility criteria The University of Florida Institutional Review Board approved the PEAR sub study protocol and all participants provided written informed consent. PEAR sub study visits were conducted during normally scheduled PEAR study visits A PEAR sub study visit s chedule is depicted in Figure 2 1. The baseline OGTT (visit 1) took place at the PEAR baseline visit, prior to beginning a ny PEAR study medication. For those patients who underwent successful titration of study drug, w hen the participant returned to the clinic for visit 2 after nine weeks of single drug therapy, a second OGTT (visit 2) was administered. The final OGTT (visit 3) was administered approximately 18 weeks after initiation of study medication. If a participant did not undergo dose titration of either study medication to a dose of 25mg HCTZ and/or 100mg atenolol, as part of the PEA R protocol (see dashed line in F igu re 2 1) the participant was excluded from further participation in this PEAR sub study to minimize variance between subjects As part of each PEAR sub study visit, participants reported to the clinic in a fasting state, not having consumed food or bevera ges other than water since midnight prior to the visit. Participants underwent a two hour OGTT at the three study visits whereby they were given a 75 gram glucose solution (Trutol 100, product 401207, Lot 4410, Nerl Diagnostics, East Providence, RI) to dr ink. Bl ood samples were drawn prior to, and one and two hours after the glucose solution was consumed to measure glucose and insulin a total of 30 mL s of blood per visit Glucose and insulin samples were measured by the

PAGE 51

51 clinical reference laboratory Quest Diagnostics in Tampa, FL (insulin test code 6695, glucose test code 23475) Glucose concentrations from plasma were determined spectrophotometrically by the Olympus Analyzer 5400 ( Olympus Beckman Coulter, CA ) with a clinically reportable range of 10 800mg /dL A spectrophotometer measures the amount of light that reaches the detector after passing through the analyte sample. 140 The intraassay precision (coefficient of variation) was within 1.43% (0.79mg/dL) at low (55.1mg/dL), 0.95% (1.24mg/dL) at medium (130.5mg/dL) and 0.93% (3.86mg/dL) at high (415.5mg/dL) glucose concentrations. Insulin concentrations from serum were determined by the Siemens Immulite 2000 Insulin reagent kit on S iemens IMMULITE 2000 Chemiluminescent Immunoassay System (Malvern, PA) with a clinically reportable range of 2 300uIU/mL An immunoassay identifies and measures an analyte based on the binding between antigen and antibody. 141 The intraassay precision (coefficient of variation) was within 8.3% (1uIU/mL) at low, medium, and high levels (6 13, and 60uIU/mL). Insulin values considered below the quantification limit of the assay (<2uIU/mL) were included and given the value of half of the lower limit of detection ( 1uIU/mL ) Of possible methods to address values below the limit of detection for an assay, this method is well established and best fits the data set. 142 A more specific assay previously employed in this lab oratory for OGTT insulin data found that a lower limit of detection of 0.798uIU/mL yielded 30 values between 1 and 1.8uIU/mL with only three values <0.798uIU/mL. Those three samples, w of detection, were detectable at 0.7 15 uIU/mL. 143 Thus, for purposes of the present study, considering values below the limit of detection as equal to 1uIU/mL is reasonable.

PAGE 52

52 Statistical Methods Descriptive statistics were used to determine frequency distributions, means and standard deviations. Baseline characteristics were compared by treatment assignment using 2 test or t test, as appropriate. For the endpoint t o describe medication associated change in OGTT at 2 hours, we required 24 total subjects for 80% power to detect an effect size of 0. 6 using a paired t test, with 12 subjects in each arm. To address our primary hypothesis, b ased on fasting glucose and OG TT response, each patien t was classified as normal or having elevated glucose at each time point. Consistent with the ADA classification for pre diabetes, a fasting glucose <100mg/dL was considered normal, > 100mg/dL was considered impaired fasting glucose and a 2 hour post glucose bolus <140mg/dL was considered normal and > 140 was considered glucose intolerant 67 T he difference between the classifications of patients based on fasting values or OGTT values along the continuum of diabetes development was variable (yes or no, pre diabetes) on repeated measures data ( fasting versus OGTT at 2 hours ). Secondarily we evaluated drug associated c hange in fasting glucose and drug associated change in OGTT area under the curve (AUC) to further evaluate sensitivity of the OGTT. A paired t test was used to compare change fro m baseline to mono and combination therapy Based on the Shapiro Wilk test glucose and insulin data were not normally distributed. Thus glucose and insulin data were evaluated by multiple methods including log 10 transforming and conducting nonparametric statistical tests Three validated surrogate indexes for insulin resistance were used to evaluate change in insulin resistance HOMA 76 QUICKI 82 and Matsuda 83 The equations for each index are shown in Table 1 4 Data are expressed a s mean values + standard deviation.

PAGE 53

53 A paired t test was used to compare change from baseline to mono and combination therapy for the surrogate indexes. Given the exploratory nature of the study, nominal statistical significance is reported and a p value o f less than 0.05 was considered statistically significant. All statistical analyses were performed using SAS (version 9.2, Cary, NC). P harmacokinetic parameters were calculate d by noncompartmental analysis, the linear trapezoidal method, using WinNonlin (v ersion 5.3, Pharsight Corporation, Mountain View, CA). Results Of 46 participants who provided consent, 39 completed the first study visit, 26 participants completed at least the second visit and were included in the a nalysis. Reasons for withdrawal from the study included non compliance in the main PEAR trial meeting study exclusion in the sub study (dose of second medication was not increased) lost to follow up, or an adverse event associated with the PEAR trial Figure 2 2 details enrollment and reason s for withdrawal. The study sample of 26 participants was comprised of fourteen Caucasians, nine African Americans, one Hispanic and two of multiracial ancestry. The mean age of participants was 46.7 + 11.1 years with 21 males and five women. Table 2 1 lists b aseline characteristics for sub study subjects that completed to study m idpoint and for PEAR subjects at the UF site. Of note, the sub study subjects were generally similar at baseline to UF subjects in terms of age, race, and blood pressure. However, a greater proportion of men were enrolled and higher baseline glucose was observed in the sub study. Primary Findings: Fasting Glucose Versus OGTT data This study did not detect a difference in the number of patients diagnosed with pre diabetes based on fa sting glucose criteria or 2 hour OGTT criteria (Table 2 3 ) In

PAGE 54

54 contrast with our hypothesis, a few people were classified as pre diabetic based on fasting glucose compared to 2 hour glucose, though these findings were not statistically significant. Further based on the proportion of discordance after monotherapy ( data in Table 2 2) between fasting glucose (8 patients pre diabetic based on fasting glucose) and 2 hour glucose (6 patients pre diabetic based on 2 hour OGTT) of 54 % that was observed ( 14 patient s with discordant results of 26 total patients ) approximately 370 patients would have to be enrolled to detect a statistically significant difference. Secondarily, no significant change in glucose or insulin between visits was detected (Table 2 3 ) Data a t each study visit, according to treatment arm, at each time point during the OGTT is depicted in T able 2 3 No significant changes were found across study visits after monotherapy or combination therapy, in glucose or insulin. Figure 2 3 depicts the media n, interquartile r ange and outliers for glucose and insulin at each study visit No significant changes were found in OGTT log glucose AUC or log insulin AUC between study visits ( see Figure 2 4 for log transformed glucose and insulin AUC across study visi ts; data additional analysis of log transformed glucose and insulin are reported in the appendix, Table A 10 ) Secondary Findings: Surrogate Indexes for Insulin Resistance Three validated surrogate markers for insulin resistance were also evaluated ; two c alculations for evaluating insulin resistance use fasting values (HOMA and QUICKI) the third incorporates OGTT values (Matsuda) Considering these methods are widely used and correlate well with the gold standard, the euglycemic hyperinsulinemic clamp ( HO MA, r=0.71 ; QUICKI, r=0.71; Matsuda, r=0.77), 75 these calculations were used to quantify insulin resistance across study visits None of these formulas detected significant changes over the study visits (Table 2 4)

PAGE 55

55 Discussio n We hypothesized that medication asso ciated changes in glucose are difficult to detect in fasting glucose levels, but an oral glucose tolerance test would prove more sensitive. Though multiple methods were incorporated to compare OGTT to fasting glucose findings, including pre diabetes classi fication, AUC analysis, and glucose/insulin inclusive formulas for insulin resistance, no greater utility was observed with OGTT over fasting glucose to detect medication associated changes in the first four to six months of antihypertensive therapy Howev er, t here were no significant differences in either glucose or insulin at any time point during the OGTT, after addition of mono therapy or combination therapy, in either randomized arm. Considering our sample size, we did not expect nor were we powered t o detect a difference in fasting glucose The change in fasting glucose in the atenolol arm across study visits was similar between the sub study and overall PEAR study (Table A 3) where the change in fasting glucose was significantly different in the ove rall PEAR study 84 The change in fasting glucose in the HCTZ arm across study visits was not similar between the sub study and overall PEAR study (Table A 3). As may be the case with small sample sizes, this difference between a subset and larger p opulation may be primarily due to chance. I n contrast to our primary hypothesis, we were not able to prove that an OGTT was more sensitive in detecting medication associated changes in glucose. As our power estimate was based on the OGTT glucose phenotype (not fasting glucose) we expected to detect a difference, if a difference existed. Thus, these data suggest that, in this subset of PEAR patients, during the early periods of antihypertensive therapy, when clinicians are potentially more likely to consider dysglycemic effects of medication selection, there may be no benefit in evaluating

PAGE 56

56 patients using the more intensive OGTT compared to fasting glucose. This study did not evaluate glycemic changes resulting from antihypertensive medications beyond 18 weeks, so it is possible that glycemic changes evaluated by OGTT or fasting glucose may prove dissimilar with increasing length of therapy. To our knowledge, this stud y is unique in phenotype (fasting glucose versus 2hr OGTT glucose) and population ( a subset of a prospective, open label, randomized study of HCTZ and /or atenolol treatment ) While previous studies have found alterations in glucose or insulin after treatme nt with these medication classes, phenotypes vary from the more intensive and impractical intravenous glucose tolerance test using the eug lycemic hyperinsulinemic clamp 144 146 to the less invasive but easier to collect fasting glucose 46 121 In addition, beta blocker s and thiazide diuretics are often studied in combination with glycemically neutral or beneficial medications 123 further complicating glycemic findings specific for either drug class. A separate sub study was previously conducted in atenolol treated PEAR patients, but with the objective of evaluating the relationship between atenolol exposure and metabolic effects at a single time point. 147 The single study evaluating glucose phenotypes associated with a dysglycemic antihypertensive medication (atenolol ) in hypertensives, included patients not treated with antihypertensiv e medication in the previous six months and also treated with monotherapy over a longer time period of 26 weeks 120 O ur study included patients more likely to be represented in the clinic, including new, never treated, and previously treated hypertensives. Additionally, our study included 9 weeks of monotherapy, so findings are not truly comparable and again, it is possible that our study duration was insufficient

PAGE 57

57 Several limitations in the study may have impacted our inability to detect significant findings. First, the sample size was small, particularly when considering the variability inherent in OGTTs. It is possible that a larger sample size would have yield ed significant results. However, the data do not trend toward the OGTT having a greater sensitivity to detect medication associated dysglycemia over fasting glucose, thus it is unlikely that a larger sample size would result in different findings. In many cases a potentially clinically relevant change in 2 hour glucose was seen, but did not reach the threshold for significance considering the large variability In both randomized treatment arms the mean 2 hour glucose decreased after monotherapy. After co mbination therapy, the mean 2 hour glucose increased to values similar to or exceeding baseline 2 hour glucose. These trend s were also seen in the mean 1 hour glucose values for both arms and in 0 hour (fasting) glucose values in the hydrochlorothiazide ra ndomized arm. Mean insulin values show a less consistent temporal pattern, but in a greater sample size would likely correlate changes in insulin with the changes seen in glucose. These trends may be explained by exhaustion of the pancreas over time. 72 Potentially, after nine weeks of antihypertensive therapy the panc reas is still able to compensate for antihypertensive associated physiological changes, resulting i n lower glucose than baseline. While no changes were found significant, a n additional nine weeks with a second antihypertensive medication may further pressu re the pancreas to respond, promoting increased pancreatic stress with glucose increasing back to baseline levels but insulin levels not rising substantially further Secondly this group of hypertensives may be too early in the dysglycemic disease proces s to detect significant changes in glucose response to an OGTT As seen in

PAGE 58

58 mean change in glucose and insulin in this cohort of participants the pancreas maintained an ability to compensate for physiologic changes over the first four to six months of anti hypertensive treatment Third, while participants were randomized by medication arm in the main study, in a small sub study groups may be imbalanced and not closely representative of the UF study population See Table 2 1 for the sub study and UF site demo graphics Not only was there imbalance between medication arm (slightly more participants in the atenolol arm), but there was also an imbalance in gender compared to enrollment at the UF site In the UF study population, females accounted for approximatel y half of the participants while our study was more than 80% male We felt justified in studying a much smaller sample population, considering the exploratory nature of the study and potentially greater sensitivity of the OGTT phenotype. While the OGTT di d not prove to be more specific we accept that our findings may have been different in a larger population. Finally, it is known that OGTT s are not highly reproducible 148 149 ADA criteria for diagnosing pre diabetes or diabetes recommend confirming results by repeat testing utilizing the same test, on a subsequent day. Though relatively noninvasive, OGTTs r equire more time and venipuncture compared to a fasting glucose test. While replication of each OGTT at each visit would strengthen the data set, it is likely that participants would have been reluctant to participate or complete the study protocol, thus p ossibly lowering the sample size further Given that the study was a subset of a larger hypertension trial, certain aspects where not able to be controlled by the sub study. Participants reported for study visits at least eight hours fasting, but at variab le times throughout the day. Some participants started the 2 hour OGTT visit early in the

PAGE 59

59 morning after approximately eight hours fasting, whereas other started the visit in the early afternoon approaching sixteen hours fasting. R esponse to an OGTT after varying lengths of time fasting may differ causing variable hepatic glucose release throughout the time period, which could impact OGTT results. Furthermore, smoking, alcohol consumption and physical activity within 24 hours of an OGTT can affect results by lowering insulin sensitivity, impairing gluconeogenesis, or altering peripheral glucose disposal, respectively. 72 Knowing that many factors may acutely impact insulin secretion and glucose disposal, the difficulty in sufficiently controlling the study conditions for each participant is apparent. While the OGTT mai ntains clinical utility in ruli ng out a diagnosis of diabetes the variability is well known and diagnosing diabetes using an OGTT requires a repeated test 67 which was not possible in this study. Considering the primary phenotype of the main study was antihypertensive response in a community based practice setting, we created as controlled an environment as possible. Fasting status was ev aluated at each visit, glucose beverage was measured precisely for each bolus, precise times were collected for sample collection and samples were promptly chilled and analyzed. Summarily, we were unable to detect greater sensitivity in evaluating metabol ic changes through an OGTT versus fasting glucose. These findings suggest that when evaluating drug associated adverse metabolic effects in the first six months of antihypertensive therapy, the simpler and more commonly used fasting glucose is equally as sensitive as an OGTT.

PAGE 60

60 Figure 2 1. PEAR sub study design. OGTTs were incorporated at visit 1, 2 and 3 for the sub study if participants followed the normal progression, not meeting the blood pressure exclusion for dose titration.

PAGE 61

61 Figure 2 2 PEAR s ub study enrollment

PAGE 62

62 Table 2 1. Baseline demographics. Sub study: ATEN Sub study: HCTZ P value Sub study o verall UF PEAR s ite P value ** N 15 11 26 255 Age (years) 49.8 + 8.3 42.3 + 42.3 0.09 46.711.2 47.89.9 0.51 Sex (% male) 80.0 81.8 0.91 80.8 51 .0 0.001 ATEN (%) 57.7 50.2 Race (%) 0.29 0.68 African American 26.7 45.5 34.6 38.4 Caucasian 60.0 45.5 53.9 56.1 Other 13.3 9.1 11.5 5.5 BMI (kg/m 2) 31.3 + 5.6 31.1 + 3.5 0.91 31.24.8 31.76.6 0.59 Systolic CBP (mmHg) 149.3 + 12.5 149.5 + 11.7 0.98 149.411.9 150.212.8 0.72 Diastolic CBP (mmHg) 99.8 + 5.6 96.4 + 4.6 0.11 98.35.4 97.65.3 0.48 Glucose (mg/dL) 97.6 + 15.4 98.2 + 16.7 0.92 97.915.6 92.815.1 0.02 Insulin (uIU/mL) 8.8 + 5.0 12.1 + 12.0 0.41 10.28.6 13.215.9 0.22 Trigly cerides (mg/dL) 119.4 + 65.9 147.6 + 89.7 0.36 131.376.5 135.897.5 0.72 HDL (mg/dL) 47.3 + 12.7 41.2 + 6.8 0.16 44.710.9 46.914.8 0.43 P value* = ATEN vs. HCTZ; P value**= Sub study vs UF PEAR site; Data are expressed as mean SD unless otherwise noted. All laboratory data were collected while participant was fasting. ATEN= atenolol, BMI= body mass index; CBP= clinic blood pressure.

PAGE 63

63 Table 2 2. Number of patients classified as pre diabetic by fasting glucose and/ or 2 hour post load of 75 gram glucose sol ution. Normal glucose Pre diabetic FG & 2hr OGTT FG only 2hour OGTT only P value Combined N Visit 1 26 16 (61.5) 4 (15.4) 4 (15.4) 2 (7.7) 0.4142 Visit 2 26 11 (42.3) 1 (3.9) 8 (30.8) 6 (23.1) 0.5930 Visit 3 22 11 (50.0) 5 (22.7) 4 (18.2) 2 (9.1) 0.4142 ATEN arm Visit 1 15 10 (66.7) 3 (20.0) 2 (13.33) 0 0.1573 Visit 2 15 5 (33.3) 1 (6.7) 6 (40.0) 3 (20.0) 0.3173 Visit 3 13 6 (46.2) 4 (30.8) 2 (15.38) 1 (7.69) 0.5637 HCTZ arm Visit 1 11 6 (54.6) 1 (9.1) 2 (18.18) 2 (18.1 8) 1.0 Visit 2 11 6 (54.6) 0 2 (18.18) 3 (27.27) 0.6547 Visit 3 9 5 (55.6) 1 (11.1) 2 (22.22) 1 (11.11) 0.5637 Number (percent) of patients is depicted. P Pre diabetes was defined as > 100mg/dL fasting plasma glucose and > 140m g/dL 120minutes post load glucose. FG=fasting glucose

PAGE 64

64 Table 2 3. Effect of nine weeks of antihypertensive monotherapy, then additional nine weeks with second antihypertensive agent on glucose and insulin Time Visit 1 Visit 2 Visit 3 P* P** P*** Gluc ose (mg/dL) ATEN (n=15) Baseline ATEN +HCTZ 0 95.9 + 13.2 98.6 + 14.9 99.6 + 11.7 0.7484 0.3845 0.7087 60 172.3 + 37.6 167.9 + 42.5 186.1 + 38.3 0.7093 0.6226 0.5147 120 123.5 + 38.9 115.4 + 30.1 139.8 + 40.4 0.3665 0.2659 0.3593 HCTZ (n=11) Baseline HC TZ +ATEN 0 94.8 + 14.7 90.6 + 10.5 95.4 + 9.1 0.3057 0.3281 0.6361 60 177.7 + 46.1 169.2 + 44.1 173.1 + 62.8 0.333 0.8203 0.5857 120 113.6 + 25.8 107.4 + 34.6 111.1 + 25.4 0.3096 0.4883 0.7739 Insulin (uIU/mL) ATEN (n=15) Baseline ATEN +HCTZ 0 3.3 + 2. 9 7.2 + 15.3 6.0 + 10.1 0.3665 0.2659 0.8208 60 34.9 + 24.3 56.1 + 45.5 48.9 + 38.8 0.7093 0.6226 0.3494 120 28.7 + 15.5 35.7 + 32.8 41.0 + 38.4 0.3665 0.2659 0.7541 HCTZ (n=11) Baseline HCTZ +ATEN 0 7.8 + 10.2 6.7 + 6.3 8.6 + 7.8 0.3096 0.4883 0.8976 60 70 .1 + 59.5 83.7 + 82.9 74.8 + 64.6 0.3330 0.8203 0.9923 120 61.6 + 54.4 60.9 + 52.4 72.4 + 68.2 0.3096 0.4883 0.9635 Days since baseline ATEN -70.7 + 12.6 141.9 + 20.7 HCTZ -74.5 + 15.6 147.4 + 20.2 P values generated using nonparametric stati stics. P* value compares change in variable from baseline to monotherapy. P** value compares change from baseline to combination therapy. P*** value compares the change in variable across baseline, monotherapy, and combination therapy time points. Time, mi nutes; g lucose, mg/dL, insulin, uIU/mL.

PAGE 65

65 Figure 2 3. Glucose and insulin median, interquartile range and outliers at each time point. A) Glucose, the atenolol treated arm. B) Glucose, hydrochlorothiazide treated arm. C) Ins ulin, atenolol treated arm D) Insulin, hydrochlorothiazide treated arm. A B C D

PAGE 66

66 A. C. B. D. Figure 2 4. Mean and standard deviation of log transformed AUC. A) Log glucose AUC at each visit, in the atenolol arm. B) Log insulin AUC at each visit, in the atenolol arm. C) Log glucose AUC at each visit, in the hydrochlorothiazide arm. D) Log insulin AUC at each visit, in the hydrochlorothiazide arm. Glucose units, min*mg/dL; insulin units, min*IU/mL p= 0.7348 p=0.5770 p=0.9864 p=0.9692

PAGE 67

67 Table 2 4. Surrogate markers for insulin resistan ce indexes at each visit based on calculations from fasting glucose values (HOMA and QUICKI) or OGTT values (Matsuda). Visit 1 Visit 2 Visit 3 P* P** Atenolol (n=15) Baseline ATEN +HCTZ HOMA 0.920.74 2.274.97 1.72.88 0.3019 0.2898 QUICKI 0.420.07 0.430.09 0.410.08 0.3019 0.2898 Matsuda 15.4411.32 15.09 11.11 12.87 10.26 0. 9028 0. 2841 Hydrochlorothiazide (n=11) Baseline HCTZ +ATEN HOMA 2.052.74 1.61.41 2.141.85 0.5375 0.9260 QUICKI 0.40.09 0.40.0 8 0.390.09 0.9172 0.4275 Matsuda 11.82 10.74 13.20 12.69 13.34 14.89 0. 6195 0. 9707 Mean + standard deviation. P* value compares change from baseline to monotherapy. P** value compares change from baseline to combination therapy. For HOMA, incre asing values indicate elevated insulin resistance. For QUICKI and Matsuda, decreasing values indicate elevated insulin resistance.

PAGE 68

68 CHAPTER 3 C LINICAL PREDICTORS OF ELEVATION IN GLUCOSE ASSOCIATED WITH USE OF BETA BLOCKERS AND THIAZIDE DIURETICS Introduc tion There exists a growing interest in the medical community in discerning which patients are most at risk for antihypertensive medication associated elevations in glucose or development of diabetes 25 27 150 152 Previous studies have found that minority ethnicity, BMI, left ventricular hypertrophy, higher foll ow up systolic blood pressure 17 elevated glucose at entry 25 48 uric acid 84 female sex 84 and age 25 are associated with diabetes development after antihypertensive therapy Concerning thiazide diuretic associated elev ations in glucose, African ancestry, lower baseline glucose, and urinary sodium excretion were found to be significantly associated with change in glucose after hydrochlorothiazide. 111 However, these findings are based on a single study that evaluated glucose changes after only one month of hydrochlorothiazide. 111 Not only is it is unknown whether the predictors for antihypertensive associated diabetes are the same as antihypertensive associated elevations in glucose, but only a single study has evaluated drug associated increase in glucose asso ciated with use of antihypertensive medications. Many of the previously mentioned patient characteristics associated with new onset diabetes have a logical basis considering the v arious mechanisms that have been hypothesized for thiazide and beta blocker associated elevations in glucose Diuretic associated diabetes may be related to potassium depletion 28 decreased insulin secretion 29 upregulation of the renin angiotensin system, 30 31 and/or decreased peripheral insulin sensitivity. 32 Beta blocker diabetogenic effects may resu lt from diminished pancreatic beta cell insulin release and reduced peripheral glucose

PAGE 69

69 utilization; 50 52 altered carbohydrate and lipid metabolism and attenuated release of fatty acids promoting harmful downstream changes such as weight gain and dyslipidemia. 53 54 Addi tional detail on the mechanisms involved is provided in Chapter 1. W hile data on the predictive potential of demographics, medical history, comorbidities, and serum biomarkers to detect medication associated diabetes are slowly growing, much remains unkn own specifically regarding early elevations in glucose prior to overt diabetes development G iven that the elevation in glucose associated with use of thiazide diuretics and beta blockers may offset the clinical benefit, 48 49 153 154 it is important to ide ntify the clinical factors that place a patient at risk T he current study aimed to determine the clinical parameters associated with increase in glucose resulting from atenolol or HCTZ exposure. Methods PEAR was a prospective, open label, multi center, r andomized study of atenolol and HCTZ alone and combined, to determine the genetic associations of antihypertensive and adverse metabolic responses Additional detai ls are provided in Chapter 2. Protocol Baseline studies included collection of home, office and ambulatory blood pressure data as well as biological samples for laboratory measurements. Participants were then randomized to receive either atenolol 50 mg daily or HCTZ 12.5 mg daily After three weeks on the initial dose, participants with an aver age home or office systolic blood pressure >120 mmHg or diastolic blood pressure >70 mmHg had the ir dose doubled. After at least 6 additional weeks at the target dose, comprehensive blood pressure and biological samples were again assessed (first response assessment or

PAGE 70

70 monotherapy assessment). Participants with blood pressure >120/70 mm Hg at monotherapy assessment then had the alternative study medication added, with the same dose titration and response assessment (second response assessment or combination therapy assessment). Laboratory Measurements At baseline, monotherapy, and combination therapy assessment study visits, fasting blood samples were collected for glucose, insulin, potassium, uric acid, total cholesterol, creatinine, high density lipoprote in (HDL), low density lipoprotein (LDL), and triglycerides. Fasting was defined as not having consumed food or beverages other than water in at least the previous 8 hours. Estimated glomerular filtration rate was calculated using the Modification of Diet i n Renal Disease equation (MDRD) 155 I f a change in glucose between study visits was above or below the population m ean by three standard deviations, the data were considered to be outlier s (potentially non fasting) and excluded from analysis. Biochemical Assays Serum glucose, HDL, uric acid, potassium, and triglycerides were quantified on a Hitachi 911 Chemistry Analyzer (Roche Diagnostics) at the central laborato ry at the Mayo Clinic. Serum g lucose, HDL, uric acid, potassium, and triglycerides concentrations were determined spectrophotometrically by automated enzymatic assays and potassium concentrations were determined by an ion selective electrode. LDL cholester ol was calculated. The Access Ultrasensitive Insulin immunoassay system (Beckman Instruments) measured plasma insulin levels. Samples were tested in duplicate

PAGE 71

71 Anthropometric Measuremen ts Weight was measured to the nearest 0.1kg; height was measured to the nearest 0.1cm. Waist circumference was measured to the nearest 0.5cm. Waist circumference was quantified by trained study personnel whereby the tape measure w as placed snuggly around t he abdomen, level with the umbilicus, slightly above the uppermost lateral border of the right iliac crest. Waist circumference was measured while the participant was standing, with hands at the side, and normal to minimal respirations. Body surface area was calculated using the DuBois formula. 156 157 Statisti cal Methods For purposes of this a nalysis, PEAR was randomly divided into derivation and validation cohorts using the surveyselect procedure in SAS in order that we could develop and then test a prediction model Proc surveyselect is a procedure for selec ting random probability based samples. The specific method for sampling was simple random sampling (SRS) which means each unit has an equal probability of selection and sampling is without replacement. All analyse s were initially conducted in the derivatio n cohort (n= 367 ) Findings were then confirmed in the validation cohort (n= 368 ) Analyse s for atenolol and HCTZ associated change in glucose were conducted separately. Atenolol analysis included the change in glucose from baseline to first response assessm ent in patients randomized to receive atenolol first (atenolol monotherapy) and the change from first response assessment to second response assessment in those randomized to receive HCTZ first (atenolol add on) HCTZ analysis included the change in glucos e from baseline to first response assessment in patients randomized to receive HCTZ first (HCTZ monotherapy) and the change from

PAGE 72

72 first response assessment to second response assessment in those randomized to receive atenolol first (HCTZ add on) Descriptiv e statistics were used for demographic and laboratory parameters. In the derivation cohort, univariate linear regression was conducted to evaluate each variabl e for inclusion in later models; v ariables with a p value <0.2 in univariate analysis were consid ered for inclusion in subsequent models. Variables tested in the univariate model included age, gender, drug assignment, race (black or non black), number of alcoholic beverages per week, smoking status, waist circumference, estimated glomerular filtration rate, days treated with study medication, and baseline values for home systolic blood pressure, home diastolic blood pressure, home heart rate, glucose, insulin, uric acid, potassium, total cholesterol, triglycerides, HDL, and LDL. The race variable, blac k or non black, was based on self identified race and confirmed by principal component clustering with African ancestry or non African ancestry based on genome wide genotype data from Illumina Human Omni 1M chip Waist circumference was selected as the bod y size parameter based on the detrimental physiological effects of abdominal obesity, such as secretion of free fatty acids hormones, inflammatory markers 158 163 as well as the associat ed elevated cardiovascular risk, independent of other body size paramete rs 8 164 169 After univariate analysis, a stepwise linear regression selection procedure was used; variables with a p value <0.05 were considered significant predictors for change in fasting glucose. To validate the regression model from the derivation cohort the correlati on between the predicted (based on the regression equation from the derivation cohort) and the observed drug associated change in

PAGE 73

73 glucose was evaluated in the validation cohort S tatistical analyses were conducted using SAS version 9.2 (SAS Institute Car y, NC). Results A total of 768 patients were considered in the initial analysis. Five patients were excluded due to lack of glucose data at monotherapy assessment ; 4 patients were excluded due lack of waist circumference measurement An additional 2 1 patie nts were considered outliers based on change in glucose between visits and were dropped from the analyses. Considering the findings from this analysis are important in evaluating the genetics involved in medication associate d dysglycemia in C hapter 4 thre e patients were excluded based on genetic sample concerns. (See C hapter 4 for further details .) This resulted in a total of 735 patients included in this analysis. Baseline demographics, clinical and laboratory parameters, and change in glucose over the co urse of the study are summarized in Table 3 1 and 3 2 Patients studied were on average 49 years old, with slightly fewer men enrolled (47%) than women, and approximately 39 % o f the population was of African ancestry. The mean baseline glucose in the deriv ation cohort was 91.5 mg/dL and mean baseline glucose in the validation cohort was 90.7 mg/dL which was not significantly different (p=0. 37 ) No baseline demographics, clinical or laboratory parameters were significantly different between the derivation and validation cohorts (Table 3 1). To justify combining medication associated change in glucose from monotherapy with add on therapy change in glucose by treatment and order was evaluated (Table 3 3). Following atenolol monotherapy glucose was elevated by 1.90mg/dL; when atenolol was added to existing HCTZ treatment, glucose was elevated by 2.29mg/dL (p=0. 60 atenolol monotherapy versus atenolol add on). Following HCTZ monotherapy glucose was elevated by 1.94mg/dL; after HCTZ was added to existing

PAGE 74

74 atenolol treatment, glucose was elevated by 2.36mg/dL (p=0. 55 HCTZ monotherapy versus HCTZ add on). Considering the mean change in glucose in monotherapy versus add on therapy was not statistically significant for either medication we determined that combining m onotherapy response and add on therapy response was appropriate. After treatment with either study medication glucose was elevated (Table 3 2 Table 3 3 ). For both antihypertensive agents studied, baseline glucose was the strongest predictor of change in glucose in univariate analysis (p<0.0001 Table 3 4 ) In univariate analysis for atenolol, baseline glucose (p<0.0001) baseline insulin (p=0.0009) and baseline uric acid (p=0.1096) were pulled forward for multivariate regression (p<0.2 Table 3 4 ). Of va riables included in the multivariate stepwise regression model in the derivation cohort for atenolol associated increase in glucose, baseline glucose (p<0.0001) was the only significant predictor (Table 3 5 ) T he atenolol model explained 13.06 % of the va ri ability in change in glucose in the derivation cohort The regression model for atenolol associated increase in glucose was defined by the following equation: Y= 30.22812 0. 30479 x baseline glucose Baseline glucose maintained strong significance when ev aluated in the atenolol validation cohort (p<0.0001, Table 3 5). Baseline glucose explained 22.60 % of the variability in change in glucose in the atenolol validation cohort. Based on the univariate analysis for HCTZ, bas eline glucose (P<0.0001) baseline insulin (p=0.0132) baseline HDL (p=0.1933), lengthy of therapy (p=0.1731) and gender (p=0.0072) were included in multivariate regression analysis (p<0.2 Table 3 4 ) After multivariate stepwise regression model for HCTZ associated increase in glucos e,

PAGE 75

75 b aseline glucose (p<0.0001) was the only variable to maintain significance (Table 3 6 ). The HCTZ model explained 12.11 % of the va riability in change in glucose in the derivation cohort The regression model for HCTZ associated increase in glucose was define d by the following equation: Y= 27.64829 0. 27860 x baseline glucose Baseline glucose in the HCTZ validatio n cohort retained significance ( p<0.0001, Table 3 6 ). In the HCTZ model validation cohort, baseline glucose explained 8.44 % of the variability in c hange in gl ucose. The correlation between the model predicted versus observed glucose change in the validation cohort was significant for atenolol and HCTZ ( p<0.0001 for both models; r= 0.47536 for atenolol model; r= 0.29051 for HCTZ model; Table s 3 7 and 3 8 ; Figure s 3 2 and 3 3) For both antihypertensive associated treatment elevations in glucose, the average predicted change in glucose was within 0. 3 mg/dL the observed average change in glucose (Table s 3 8 and 3 9 ). Discussion We confirmed b aseline glucos e to be an important predictor of antihypertensive medication associated rise in glucose during treatment with either atenolol or HCTZ. No other clinical variables were validated as predictors after treatment with atenolol or HCTZ However, overall, baseli ne glucose has low predictive value for drug associated glucose elevation It is notable that we found a lower baseline glucose value was associated with a greater change in glucose after drug therapy. While directionally inconsistent with studies predict ing incident diabetes, 25 48 a study similar to the present study also found lower baseline glucose to be significantly predictive of greater glucose change in

PAGE 76

76 response to HCTZ. 111 Maitland van der Zee and colleagues 111 sought to evaluate clinical and genetic predictors of adverse metabolic effects after HCTZ treatment. The researchers found a mean increase in glucose of 3.5 + 9.5mg/dL, with approxi mately 11% of the variation explained by clinical predictors including African ancestry, lower plasma glucose, and lower urinary sodium excretion. In our study, approximately 12 13% of the variability in change in glucose after antihypertensive treatment w as explained by a lower baseline glucose value. A previous analysis in the PEAR population at midpoint included 395 patients assessed for AMEs attributable to atenolol and hydrochlorothiazide. 84 The population was divided based on the presence or absence of abdominal obesity and evaluated for new cases of impaired fasting glucose and incident diabetes. This analysis established that predictors for new onset diabetes (defined as fasting glucose > 126mg/dL) in the entire population after monotherapy included higher baseline glucose (OR 1.10, 95% CI 1.03 1.18), randomization to receive HCTZ first (OR 13.37, 95% CI 1.45 123), and female sex (OR 10.00, 95% CI 1.63 62.5). Further, while higher baseline glucose was the most consistent predictor for glucose > 100mg/dL, gender and stratification by abdominal obesity were also important predictors. 84 While similar initial predictors for medication associated elevations in glucose were identified in the present study, where female gender (p=0.0072) and length of therapy with HCTZ (p=0.1731) in the HCTZ response phenotype an d uric acid (p=0.1096) in the atenolol response phenotype met the threshold for inclusion in the multivariate model (p<0.02), none of these parameters remained significant in the multivariate analysis Thus, while the current and previous PEAR analyses mai ntained different study aims, baseline glucose was a significant

PAGE 77

77 predictor for change in glucose in the current study and pre diabetes (glucose > 100mg/dL) or new onset diabetes (glucose > 126mg/dL) in the previous study. However, of note, lower baseline gluco se predicted a greater cha nge in glucose in PEAR whereas higher baseline glucose predicted an increased risk of new onset diabetes or new onset impaired fasting glucose. F or both medication associated change in glucose and medication associated diabetes d evelopment, glucose is increasing. In relatively young and healthy hypertensive patients evaluated in the present PEAR analysis after treatment with medications that adversely affect glucose, a lower glucose led to a greater increase in glucose In compli cated hypertensive patients with complex comorbidities 17 25 addition of medications that elevate gluc ose pushed patients across the threshold for diabetes diagnosis. I n the previously mentioned PEAR analysis, 84 elevated baseline glucose predicted new onset diabetes and impaired fasting glucose. Hence, it is likely that though a difference in underlying disease severity may be important in comparing the present analysis to prior studies, it is more likely that a difference in phenotype assessment is the reas on for seemingly inconsistent findings. These discrepancies emphasize the difference between change in glucose as a continuous variable, where the amount of increase may be more strongly associated with lower values compared to rate of diabetes or impaire d fasting glucose development as a dichotomous variable, where movement across the dichotomous cut point is greatest the closer the value is to a threshold. Thus, predictors of new onset diabetes or new onset impaired fasting glucose represent distinct res earch questions versus evaluation of the magnitude of drug associated glucose change. T hese findings consistently suggest that baseline

PAGE 78

78 glucose is important in predicting adverse glycemic changes, whether an elevation in glucose or overt diabetes developme nt, resulting fro m antihypertensive medications. There exist relevant limitations that should be considered. An important aspect of our analysis included combining atenolol or HCTZ monotherapy with add on therapy (+atenolol add on or +HCTZ add on respect ively) associated responses, irrespective of the order the medications were given. While possible that the order of medications administered may impact the degree of resulting elevation in glucose this did not appear to be the case in this population wher e the change in glucose was very similar between monotherapy and add on therapy ( 1.90mg/dL after atenolol monotherapy versus 2.29mg/dL after atenolol add on, p=0.5990; 1.94mg/dL after HCTZ monotherapy versus 2.36mg/dL after HCTZ add on, p=0.5489; Table 3 3 ) and to our knowledge there are no data suggesting otherwise. Further, since initial univariate analysis (Table 3 4) did not suggest drug assignment influenced medication associated change in glucose, drug assignment was not pulled forward into multivaria te analysis. Additionally, the PEAR trial only measured fasting glucose, not oral glucose tolerance based on an oral glucose tolerance test. Though oral glucose tolerance is an excellent phenotype to detect diabet es, per our previous findings (C hapter 2) f asting glucose is sufficiently sensitive to detect antihypertensive medication associated dysglycemia during the study period available Finally, though our sample size may have been smaller than previous studies evaluating elevations in glucose after anti hypertensive medications 46 we feel that use of derivation and validation cohort s stren gthen our findings. Summarily, we found baseline glucose was the strongest and most consistent predictor of atenolol or HCTZ associated elevation in glucose with lower glucose levels

PAGE 79

79 at baseline leading to larger drug associated changes Though additional parameters were significant in initial univariate analysis, they did not r emain significant in subsequent multivariate analysis Further study is warranted to establish additional clinical parameters that may predict antihypertensive medication elevations in glucose in diverse populations.

PAGE 80

80 Figure 3 1 Final PEAR dataset for evaluating clinical predictors of medication associated increase in glucose

PAGE 81

81 Table 3 1. Baseline demographics, clinical and l a boratory parameters. Mean + standard deviation or percent as appropriate. PEAR Derivation cohort Validation cohort P value N= 735 N= 367 N= 368 Age (year) 48.9 3 9.1 9 49.0 2 9.3 8 48.85 9.0 1 0. 8101 Male (%) 47.08 47 14 47.01 0. 9722 Black (%) 39. 32 37. 33 41. 30 0. 2700 Waist circumference (cm) 97.75 13.1 0 97. 61 12 .70 97.89 13.5 1 0. 7746 Drinks per week 1.9 3 4.0 3 2.0 5 4.2 9 1.8 0 3.7 5 0.41 00 Current smoker (%) 14.42 13.90 14.9 5 0. 6856 e GFR (mL/min per 1.73m 2 ) 98.0 7 21.4 4 98. 69 23. 38 97. 46 19. 31 0. 4374 Home SBP (mm Hg) 145.8 3 10.3 4 145.8 0 9.72 145. 85 10.9 4 0. 9503 Home DBP (mm Hg) 93.7 8 5.95 94.0 4 5. 86 93. 5 2 6.04 0. 2 3 09 Pulse (bpm) 77.45 9.4 0 77. 48 8.99 77. 41 9.8 0 0. 9221 Glucose (mg/dL) 91.1 2 11.3 8 91.4 9 11.61 90. 7 4 11. 16 0.3 693 Insulin ( I U/mL) 9.1 6 8.0 3 9.01 8.26 9.3 1 7. 80 0. 6134 Uric Acid (mg/dL) 5.59 1.4 4 5. 59 1.45 5.59 1.43 0. 9918 Potassium (mEq/L) 4.26 0.4 3 4.25 0.4 2 4.27 0.4 4 0. 5029 Total cholesterol (mg/dL) 194.2 8 35. 64 193. 37 35. 61 195. 19 35. 69 0. 4888 Triglyceride (mg/dL) 126. 24 91. 29 131. 90 102. 60 120. 58 78.1 2 0.09 30 HDL (mg/dL) 49.3 7 14.1 5 48.6 3 13. 19 50.1 0 15.0 4 0. 1585 LDL (mg/dL) 120.2 8 30.6 3 119.4 4 30. 89 121. 10 30. 39 0. 4663 HTN = hypertension, eGFR = estimated glomerular filtration rate, SBP = systolic blood pressure, DBP = diastolic blood pressure. *P value compares between derivation and validation cohorts.

PAGE 82

82 Table 3 2. Change in glucose and length of days o f mediation therapy. Mean + standard deviation. PEAR Derivation Validation P value* ATEN Glucose (mg/dL) 2. 12 9.82 2.09 10.15 2.14 9.51 0.9494 Days treated 67. 3 2 16. 9 5 67 87 19.54 66.78 13.90 0.3884 HCTZ Glucose (mg/dL) 2.1 0 9.46 1.77 9.39 2.42 9.54 0.3606 Days treated 66. 18 13. 28 67.91 14.23 65.46 12.26 0. 1456 *P value compares between derivation and validation cohorts.

PAGE 83

83 Table 3 3. Ch ange in glucose, by assignment and medication order. Mean + standard deviation full cohort Monotherapy Add on therapy P value ATEN 1.90 9.31 2.29 10.35 0. 5990 HCTZ 1.94 9.75 2.36 9.15 0. 5489 Glucose in mg/dL ; Monotherapy and add on therapy are defined as change in glucose after 9 weeks of one study medication.

PAGE 84

84 Table 3 4 Univariate analysis for medication associated change in glucose derivation c ohort Atenolol Hydrochlorothiazide Predictor measured + SE P value + SE P value Age (years) 0. 0 1 0.0 6 0. 8837 0.05 0. 05 0. 3606 Sex (male) 0. 45 1.0 8 0. 6776 2.68 0. 99 0. 0072 Race (black) 0.93 1.11 0. 4056 1.16 1.03 0. 2616 Assignment (atenolol) 0. 88 1 .0 8 0. 4121 0.01 1.00 0. 9888 Length of therapy (days) 0. 0 1 0.03 0. 73 17 0. 0 5 0. 04 0. 1731 Waist circumference (cm) 0.01 0.04 0.8215 0.0 3 0.04 0.4308 Drinks per week 0.01 0.12 0. 9532 0.1 1 0.12 0. 3248 Current smoker (%) 1.3 0 1.5 8 0. 4098 0. 72 1.45 0. 618 3 e GFR (mL/min per 1.73m 2 ) 0.02 0.02 0. 2945 0.01 0.02 0. 7345 Home SBP (mm Hg) 0.0 4 0.05 0. 4292 0.01 0.04 0. 8099 Home DBP (mm Hg) 0.01 0.08 0. 9880 0.03 0.06 0. 5620 Pulse (bpm) 0.01 0.06 0. 9408 0.04 0.05 0. 4973 Glucose (mg/dL) 0.31 0.04 <0.0001 0.28 0.04 <0.0001 Insulin ( I U/mL) 0.17 0.05 0.0009 0.17 0.07 0. 0132 Uric Acid (mg/dL) 0.5 1 0.3 2 0. 1096 0. 16 0. 35 0. 6529 Potassium (mEq/L) 0. 44 1.1 5 0. 7049 0. 11 1.18 0. 9281 Total cholesterol (mg/dL) 0.01 0.01 0. 5871 0.01 0.01 0. 4990 Triglyceride (mg/ dL) 0.01 0.01 0. 4669 0.01 0.01 0. 6544 HDL (mg/dL) 0.03 0.04 0. 3916 0.05 0.04 0.19 33 LDL (mg/dL) 0.01 0.02 0. 9265 0.02 0.02 0. 2903 HTN = hypertension, eGFR = estimated glomerular filtration rate, SBP = systolic blood pressure, DBP = diastolic blood pre ssure.

PAGE 85

85 Table 3 5 Predictors for atenolol associated change i n glucose in derivation cohort and evalu ation of predictors in validation cohort. Partial R2 Model R2 Parameter Estimate P value Derivation Intercept 30.22812 <0.0001 BL Glucose 0.1 306 0.1306 0.30479 <0.0001 Validation Intercept 37.13384 <0.0001 BL Glucose 0.2260 0.2260 0.38047 <0.0001 Derivation regression equation BL= baseline.

PAGE 86

86 Table 3 6 Pre dictors for hydrochlorothiazide associated change in glucose in derivation cohort and evaluati on of predictors in validation cohort. Partial R2 Model R2 Parameter Es timate P value Derivation Intercept 27.64829 <0.0001 BL Glucose 0. 1211 0.1211 0. 27860 <0.0001 Validation Intercept 26.23189 <0.0001 BL Glucose 0.0844 0. 0844 0. 26035 <0.0001 Derivation regression equation BL= baseline.

PAGE 87

87 Table 3 7 Correlation between model predicted atenolol associated change in glucose, in validation cohort, to actual atenolol associated change in glucose. Mean Standard Deviation Minimum Maximum Predicted (mg/dL) 2.20 3.62 11.22 12.86 (mg/dL) 2.14 9.51 24 .00 36.50 R 0.47 P value <0.0001

PAGE 88

88 Figure 3 2 Correlation between model predicted atenolol associated change in glucose, in validation cohort, to actual drug associated change in glucose.

PAGE 89

89 Table 3 8 Correlation between model predicted HCTZ associated change in glucose, in validation cohort, to actual drug associated change in glucose. Mean Standard Deviation Minimum Maximum Predicted glucose (mg/dL) 2. 1 7 2.96 11.36 11.49 Actual glucose (mg/dL) 2.42 9.54 32.00 40.00 R 0.29 P value <0.0001

PAGE 90

90 Figure 3 3 Correlation between model predicted HCTZ associated change in glucose, in va lidation cohort, to actual HCTZ associated change in glu cose.

PAGE 91

91 CHAPTER 4 EFFECT OF GENETIC VARIATION IN PHARMACOLOGICAL TARGET AND INSULIN SIGNALING GENES ON ALTERATIONS IN GLUCOSE AFTER TREATMENT WITH BETA BLOCKERS AND THIAZIDE DIURETICS Introduction Many decades of clinical trial data strongly support diure tic and beta blocker use for blood pressure control and cardiovascular risk reduction, 10 13 yet these studies also report finding s of adverse metabolic effects on glucose insulin, and lipids 14 16 as well as an increased risk for diabetes development. 17 19 M edication associated alterations in glucose are only seen in a portion of diuretic and beta blocker treated patients and the impact of genetic variation on development of antihypertensive associated glucose change is poorly underst ood. Pharmacogenomic studies are valuable in understanding the role of genet ics in both the desired response to medications as well as understanding the adverse effects A handful of pharmacogenomic studies suggest various genes may be associated with gluc ose r esponse to diabetes medications. 106 108 110 170 173 (Refer to Chapter 1 for additional details.) Further, w hile genes such as ADRB1 (beta 1 adrenergic receptor) and CACNB2 (calcium channel, voltage dependent, beta 2 subunit) have proved important in assessing blood pressure response to antihypertensive me dications 174 179 glucose changes associated with antihypertensive medication use is not well understood A candidate gene approach is well suited to evaluate an tihypertensive associated dysglycemia, allowing focus on both strong pharmacological target genes as well as genes important in insulin signaling. Pharmacological candidate genes were selected for this research project on. Polymorphisms in genes encoding beta receptors have shown altered response to beta

PAGE 92

92 blockers for cardiovascular disease associated endpoints, observed in multiple populations. 180 182 Further, SNPs in the beta adrenergic receptor genes have been linked to hypertension ( ADRB2 ) 174 coronary artery disease ( ADRB1 ) 180 obesity ( ADRB1, ADRB2, ADRB3 ) 183 184 acute coronary syndrome ( ADRB2) 185 and heart failure ( ADRB1 186 ADRB2 187 ). Thiazide response genes were selected to i nclude genes with previously identified associations with blood pressure phenotypes. These genes are important in either sodium transport systems ( SCNN1G, CLCNKB, SLC12A1, SLC12A3 ) or modulate vasoconstriction ( NOS3 ), and are frequently associated with blood pressure levels ( CLCNKB, 188 SLC12A1 189 SLC12A3 175 190 192 SCNN1G 193 195 ) Further, SNPs in the endothelial nitric oxide synthase ( NOS3 ) and sodium channel gamma subunit promoter ( SCNN1G ) genes have been assoc iated with an tihy pertensive response to HCTZ. 111 It is possible that the genes important for blood pressure response may also be important for dysglycemia associated wi th blood pressure medications. Further, it is likely that genetic variation in signaling pathways independent of direct drug action, may also have an important influence on medication response Evaluating genes important in insulin signaling, irrespec tive of drug mechanism, will represent what are thought to be eventual common pathways for dysglycemia. This is particularly appropriate in the present study, given that we are studying a thiazide diuretic and beta blocker. Considering these classes of med ications have very distinct mechanisms of action, yet both have the potential to produce dysglycemic effects suggests the potential for a common role in the insulin signaling pathway. Pancreatic insulin signaling involves synthesis, storage and secretion of insulin by beta cells,

PAGE 93

93 followed by degradation at the target site. 72 Insulin resistance and glucose intolerance precede diabetes and are characterized by altered cellular response in the pancreas, liver and peripheral tissue. 72 As the pancreas compensates for higher levels of glucose, pancreatic beta cell dysfunction develops transiently then permanently, leading to impaired i nsulin secretion. 72 196 Receptors, signaling proteins, and nuclear response elements work in concert for regul ation or dysregulation. 197 glucose regulation, including liver, adipose and muscle cells, result in complex downstream events. Insulin ( INS ) binds to the ins ulin receptor ( INSR) which phosphorylates insulin receptor substrates ( IRS1 and IRS2) Insulin receptor substrates then initiate a signaling cascade to activate a serine kinase Akt, which activates glycogen synthase ( GYS1) and ultimately glycogen synthesi s. As this cascade is initiated, Akt inhibits transcription factors ( FOXO1) from initiating glycolysis. Akt activation also initiates translocation of GLUT4 vesicles ( SLC2A4) from the intracellular area to the plasma membrane. Once in the membrane GLUT4 f acilitates uptake of glucose into adipose and muscle cell s Glucose transporter 2 ( SLC2A2) is specific to the liver, islet beta cells, intestine and kidney epithelium, facilitating glucose transport. Multiple steps in this process converge to a set of rate limiting genes in the cascade which when polymorphic may result in alterations in glucose homeostasis (Figure 4 1). Pathway based tools were used to identify additional biologically plausible candidate genes including KEGG 197 and Biocarta ( www.biocarta.com ). The complete list of candidate genes is listed in Table 4 1. To date, there exist few genetic studies evaluating drug associated alterat ions in glucose While much is unknown about the genetics of drug associated alterations in

PAGE 94

94 glucose incorporating knowledge of both insulin signaling pathways and drug targets (Table 4 1 ) may provide insight This study aimed to test candidate genes, both pharmacological target and insulin signaling genes, and glycemic change after antihypertensive therapies. Methods Study Population The PEAR trial was a prospective, open label, multi center, randomized study designed to evaluate the genetic determinants o f antihypertensive and adverse metabolic responses to HCTZ a tenolol, and their combination. 43 Addi tional details are provided in C hapter s 2 and 3. Glucose Meas urement At baseline, monotherapy, and combination therapy assessment study visits, fasting blood samples were collected for glucose, electrolytes and lipoproteins. Additional details are provided in C hapter 3. DNA Collection and Genotyping Ge notypes for th e candidate genes of interest were obtained from the Illumina HumanCVD BeadChip. 198 The HumanCVD BeadChip is a custom array of over 2,200 genes and 50,000 markers selected from published literature, cardiovascular disease pathway analysis, and w hole genome analysis data sets The University of Florida Center for Pharmacogenomics n ormalized DNA (50ng/ L) that w as genotyped by the University of Florida Interdisciplinary Center for Biotechnology Research on the (Illumina, San Diego, CA). GenomeStudio Software version 2011.1 and Genotyping

PAGE 95

95 Mo dule version 1.9 calling algorithm (Illumina, San Diego, CA) were used to call genotype. A total of 16 genes were selected for this study, eight strong pharmacological candidates and eight insulin signaling pathway genes. For the pharmacologically based ge nes, SNPs were chosen based on previous drug response associations or were nonsynonymous SNPs ( 27 SNPs included in the primary analysis ) as well as comprehensive gene coverage (224 tagSNPs Table C 2 in appendix ). A tagSNP approach was also incorporated fo r coverage of the insulin signaling genes (212 SNPs, T able C 3 in appendix ). The minimum threshold for selection of tagSNPs was a reported minor allele frequency of at least 5% and an r 2 of 50%. SNPs five kilobases up and downstream of candidate genes were included. SNPs with a minor allele frequency of less than 3% in both race groups were excluded from initial analysis due to limited power to detect associations at that threshold. PupaSuite 3.1 ( http://pupas uite.bioinfo.cipf.es ) 199 Fast SNP ( http://fastsnp.ibms.sinica.edu.tw/ ) 200 and F SNP ( http://compbio.cs.queensu.ca/F SNP/ ) 201 were use d to evaluate putative functionality of SNPs meeting the criteria for significance. HumanCVD chip genotype data were available for 768 PEAR participants. Samples were excluded if genotype call rates (fraction of potential genotypes called only passing SN Ps counted ) were below 95% and SNPs were excluded if genotype call rates were below 90%. A total of 81 blind duplicates were included in genotyping, with a concordance rate of 99.992%. Gender was confirmed by X chromosome genotype data; one sample was excl uded for discordance. PLINK ( http://pngu.mgh.harvard.edu/purcell/plink/ ) 202 was used to estimate cryptic relatedness

PAGE 96

96 by pairwise identity by descent analysis. A pair of monozygotic twins was identified, confirmed by the study coordinator, and one twin was removed from data set. Five pairs of sam ples were identified as first degree relatives and kept in the data set. PLINK was used to assess heterozygosity by estimating the inbreeding coefficient (a measure of relatedness) Only one participant sample had inbreeding coefficient values greater than 4 negative standard deviations from the mean; this sample was excluded. Principal component analysis was co nducted in all subjects using the EIGENSTRAT method 203 Race groups were confirmed with principle component (PC) clustering results. Principal components 1 and 2 provided the best separation of ancestry clusters and were selected as covariates in analysis to adjust for ancestry. S tatistics Hardy Weinberg equilibrium was tested separately by PC confirmed race for quality control purposes. SNPs with a Hardy Weinberg p value less than 0.0 00 1 (Bonferroni p value for 430 SNPs) in both race groups were ignored. Data were stratified by ra ce ( PC confirmed black or non black) and treatment (a t enolol and hydrochlorothiazide associated change in glucose ) for analysis. The phenotype of interest, medication associated change in glucose, was defined as the glucose difference (delta) between base line and end of monotherapy (glucose at monotherapy minus glucose at baseline) as well as the difference in glucose between monotherapy and combination therapy (glucose at combination therapy minus glucose at monotherapy) A linear regression model was use d to evaluate associations between genotype and change in glucose. The model was adjusted for pre treatment glucose and principal components 1 and 2. We adjusted for principal components to ad dress population stratification. Population stratification is a concern when studying complex

PAGE 97

97 disorders in an admixed population. In association studies, such as case control studies, an identified significant association may be incorrect due to underlying sub population heterogeneity and different minor allele frequen cies between cases and controls, where cases have a higher degree of relatedness. If not addressed population stratification, not a response or disease locus, could cause a false positive association 203 204 (The parameter estimates and p values for the primary findings described in the c hapter are detailed in Table C 8 of the appendix, adjusted for baseline g lucose, PC1 and PC1 and baseline glucose only. Findings were similar whether or not analysis was adjusted for PC1 and PC2 .) Delta glucose after monotherapy was adjusted for baseline glucose. Delta glucose after combination therapy was adjusted for monother apy glucose. (See C hapter 3 for rationale for adjusting for baseline glucose.) Statistical analyses were performed in JMP Genomics 5 and SAS 9.2 (SAS Institute, Cary NC). All SNPs were tested for an association with glucose after monotherapy and after add on therapy. For primary analysis of SNP s in pharmacological candidate genes a prespecified p < 0.05 was used to identify SNPs of interest for initial assessment (after monotherapy or add on therapy) considering all SNPs were selected based on previous fin dings and/or functional significance. Subsequent analysis compared for consistency in the direction of effect (probability by chance 0.5) between monotherapy and add on therapy, within PC confirmed race group. SNPs that maintained directional consistency i n monotherapy and add on therapy for the same medication, as well as nominal significance (p<0.05), were considered the strongest signals (0.05 initial p value 0.5 direction 0.05 same drug p value = 0.0013; See Table 4 2.). For tagSNP

PAGE 98

98 analysis in both candidate gene lists, p<0.005 was th e initial threshold for significance in conjunction with the product p value approach detailed above and in Table 4 2 For insulin signaling genes directional consistency was evaluated irrespective of medication or ra ce group This was justified in that variation in genes selected for inclusion in the insulin signaling cascade may affect glucose independent of specific drug action. Further, considering the ex ploratory nature of this gene/phenotype combination and lack of information in the literature in evaluating insulin signaling genes in a n antihypertensive treated pharmacogenomic study, it was felt that directional consistency across drug or race would support the possible importance of these genes in antihypertensi ve ass ociated alterations in glucose. As many statistical tests were conducted, it was necessary to adjust for multiple comparisons to avoid false positives. A Bonferroni adjusted p value for 27 SNPs (primary analysis in pharmacological candidate genes) i s 0.0019 (0.05/27) and for the tag SNPs (secondary analysis in pharmacological candidate genes and analysis in insulin signaling genes) studied is 0.00012 (0.05/4 03 ). A Bonferroni correction was not considered appropriate due to the Bonferroni assumption th at all sta tistical tests are independent In candidate gene studies where genes were selected based on a strong biological plausibility and tagSNPs were selected for gene coverage this approach may be too stringent considering SNPs are not inherited in a completely independent manner. 205 Ideally, our approach d id not dismiss true positives because of an overly strict initial p value and identification of false positives was minimized after incorporation of d irectional consistency in a second group.

PAGE 99

99 Power calculations (T able 4 3 ) to evaluate medication associated change in glucose were generated using G*Power 3.1.2 (Franz Faul, Germany http://www.psycho.uni duesseldorf.de/aap/projects/gpower/ ). Power was calculated stratified by PC defined race group (black or non black), for minor allele frequencies 0.05 to 0.25 tested for significance at an (primary analysis in known functional SNPs in pharmacological candidate genes) and 0.005 (tagSNPs in pharmacological candidate and insulin signaling genes) For non blacks at an of 0.05 we had >80% power at a minor allele frequency of 0.25 to d etect a 3mg/dL change in glucose. For minor allele frequencies above 0.25 in both race groups at an of 0.05 we had > 9 0% power to detect a 4mg/dL change in glucose. For non blacks at an of 0.005 we had > 9 0% power at a minor allele frequency of 0.25 to d etect a 4mg/dL change in glucose For SNPs in either race with a min or allele frequency of 0.25 at an >90% power to detect a change in glucose of 5mg/dL. Linkage disequilibrium was determined using Haploview 4.2 206 Results Baseline demographics were similar between medication assignment groups within PEAR (Table 4 4 ; Table C 1 in appendix with p values ). A total of 26 samples were excluded due to missing data or deviation from the m ean delta glucose between study visits of greater than three standard deviations ; three samples were excluded in quality control evaluation of the genetic data The final dataset consisted of 739 participants (Figure 4 2)

PAGE 100

100 Genetic Associations with Change in Glucose: Primary Analysis of Known Functional SNPs in Pharmacological Candidate Genes The strongest associations with directional consistency are shown in T able s 4 5 to 4 8 Of 27 SNP s in the primary analysis for pharmacological candidate genes SNPs ch osen based on strong functional data or previous drug response associations, one SNP met the thr eshold for significance of p<0.05 For rs13306654 in SCNN1G the minor allele was highly asso ciated with a greater elevation in glucose after treatment with HCT Z monotherapy in black participants ( those homozygous for the major allele 0.02mg/dL change in glucose, heterozygous +5.86mg/dL change in glucose homozygous for the minor allele +4.17mg/dL change in glucose ; p=0. 0020 Table 4 5 ). Further blacks treated with HCTZ add on therapy also showed a marginally significant increase in glucose ( those homozygous for the major allele +1.72mg/dL change in glucose heterozygous +4.15mg/dL change in glucose homozygous for the minor allele +8.80mg/dL change in glucose ; p=0. 0596 Table 4 5 ). A similar association was found in rs4499239 in SCNN1G ( those homozygous for the major allele +0.5 0 mg/dL change in glucose heterozygous +5.24mg/dL change in glucose homozygous for the minor allele +6.9 0 mg/dL change in glucose; p=0. 0 025 ) in the HCTZ monotherapy group, and (those homozygous for the major allele +1.85mg/dL change in glucose, heterozygous +3.89mg/dL change in glucose, homozygous for the minor allele +11.25mg/dL change in glucose; p=0. 0751 ) in the HCTZ add on group; Table 4 5 In non blacks, neither rs13306654 nor rs4499239 were directionally consistent or marginally significant (rs13306654, p=0.341 in HCTZ monotherapy, p=0.893 in HCTZ add on; rs4499239 p=0.245 in HCTZ monotherapy, p=0.924). SNPs rs13306654 and rs4499239 are in high linkage disequilibrium in blacks (r2= 0.85 ). No other SNPs were statistically significant in

PAGE 101

101 primary analysis for the pharmacological candidate genes. (Data is presented collectively for all analyses in Tables C 4 and C 5 in the appendix. An ast erisk denotes SNPs in the primary analysis.) Genetic Associations with Change in Glucose: Secondary Analysis in Pharmacological Candidate Genes Looking further into the pharmacological candidate genes, the only tag SNP that was significant at p<0.005 in in itial analysis was rs3800787 close to NOS3 In the black cohort, t he minor allele was associated with less of a change in glucose, after HCTZ monotherapy ( those homozygous for the major allele +3.7mg/dL change in glucose, heterozygous 0.41mg/dL change in glucose, homozygous for the minor allele 24mg/dL change in glucose; p=0. 0025 ) or HCTZ add on ( those homozygous for the major allele +3.53mg/dL change in glucose, heterozygous 1.30mg/dL change in glucose, homozygous for the minor allele 0.50mg/dL change in glucose; p=0. 0359 Table 4 6 ). In non blacks, rs3800787 was not significant after HCTZ monotherapy (p=0.142 and directionally inconsistent) or HCTZ add on therapy (p=0.663). Genetic Associations with Change in Glucose: Insulin Signaling Genes Finally, i n evaluating tagSNPs in insulin signaling candidate genes, IRS1 showed a strong signal for monotherapy response, irrespective of drug In non blacks rs1801278 in IRS1 resulted in a substantial increase in glucose after treatment with either ATEN or HCTZ. A fter ATEN monotherapy, rs1801278 minor allele carriers showed a significant elevation in glucose ( those homozygous for the major allele +1.65mg/dL change in glucose and minor allele carriers showed a +6.30mg/dL change in glucose; p= 0 0041 ). After HCTZ mon otherapy the direction was consistent and significant ( those homozygous for the major allele +1.10mg/dL change in glucose and

PAGE 102

102 minor allele carriers showed a +5.79mg/dL change in glucose; p=0. 0348 ), with minor allele carriers having a greater elevation in g lucose (Table 4 7 Figure 4 3 ) These findings suggest that the impact of rs1801278 is irrespective of medication treatment. No SNP associations met prespecified criteria for significance in other insulin signaling or pharmacological candidate genes. (Dat a is presented collectively for all analyses in Tables C 4 and C 5 in the appendix.) Discussion The PEAR population offers many opportunities to study the genetics of blood pressure response as well as glucose change after antihypertensive medication use i n uncomplicated hypertensive patients Our study incorporated two methods for candidate gene selection to evaluate antihypertensive associated change in glucose, a group of candidate genes chosen based on pharmacological action as well as a group of genes important in insulin signaling Of pharmacological candidate genes selected, i n blacks, we found genetic variation in SCNN1G (rs13306654 and rs4499239) and NOS3 (rs3800787) were associated with change in glucose after HCTZ monotherapy and add on therapy Further our most interesting finding in insulin signaling genes was irrespective of drug treatment ( IRS1 ) IRS1 rs1801278 was associated with change in glucose in non blacks, after either HCTZ or ATEN monotherapy. SCNN1G encodes the gamma subunit of the amiloride sensitive epithelial sodium channel (ENaC). 207 Mutations in SCNN1G or other ENaC subunits may result in altered sodium reclamation in the distal nephron and cause salt sensitive hypertension. 208 Variants in this gene have been associated with monogenic forms of hypertension 209 211 as well as blood pressure variation in the general population 194 212 213 Furt her, SNPs in strong linkage disequilibrium (rs5723, rs5729 rs11649420 ) with our SNPs in SCNN1G

PAGE 103

103 (rs13306654 and rs4499239) were associated with blood pressure response to HCTZ. 193 214 215 Relatively recently, t he importance of ENaC subunits for diuretic associated glucose phenotypes was of intere st to the GenHAT researchers, an ancillary study of ALLHAT. 112 T he nonsynonymous SNP rs2228576 in the alpha subunit of ENaC ( SCNN1A ) was significant in the amlodi pine vs. chlorthalidone comparison (p=0.001) for fasting glucose Generally, treatment with chlorthalidone resulted in a higher glucose at year six compared to amlodipine, with the highest glucose seen in those homozygous for the major allele in rs2228576 (106mg/dL after chlorthalidone vs. 100mg/dL after amlodipine) Though the alpha and gamma ENaC subunits are on different chromosomes (12 and 16, respectively), these findings suggest ENaC subunits may impact glucose changes after treatment with a thiazide diuretic. Though ENaCs are often thought to be primarily expressed in the distal nephron of the kidney and linked to disorders involving blood pressure regulation, 207 this superfamily of channels are of increasing interest to researchers and are expressed throughout the body. Researchers have found ENaC expressed in the bladder, brain, esophagus, gastrointestinal tract, lung, lymphocytes, retina, skin, taste buds and uterus. 216 Of relevance for the present study, expression in the pancreas was evaluated based on the literature and web based gene expression resources. While some researchers report ENaC expression in the pancreas or pacinian corpuscles in the pancreas 217 others have not proved that the pancreas, or pancreatic ducts, express functional ENaC. 218 Web resource s report variable results, from moderate ( SCNN1G), to low (SCNN1A and SCNN1B) expression of the subunits, 219 to moderate to high expression of all subunits, depending on the expression array. 220 As more research is

PAGE 104

104 conducted on ENaCs and where they are expressed the mechanism fo r variation within SCNN1G possibly promoting HCTZ associated dysglycemia may be further elucidated. We evaluated 17 SNPs surrounding rs13306654 and rs4499239 in SCNN1G in high linkage disequilibrium ( > 0.80) us ing FastSNP to predict function and found two SNPs with predicted altered transcription factor binding. SNP rs13306653 (r2 with rs13306654 of 0.947) is predicted to alter splicing and rs5723 ( r2 with rs13306654 of 0.895) is predicted to a ffect splice regulation. As other researchers have noted, it is likely that in this region of high linkage disequilibrium, the association is due to an unknown functional mutation decreasing the function of ENaC after thiazide exposure, 214 not the SNPs identified in the present study. Though the p valu e for the SCNN1G SNPs rs13306654 and rs4499239 in our study did not reach Bonferroni significance or the level of significance required for secondary analysis in our p harmacological candidate genes, variation in ENaC subunits and this genomic region of SCN N1G could prove important for changes in glucose after treatment with HCTZ In the black race group, carriers for the minor allele in rs3800787 less than 2kb from endothelial nitric oxide synthase ( NOS3 ) showed a smaller increase in glucose after treatme nt with HCTZ. NOS3 has previously been associated with diseases such as hypertension 117 221 ischemic heart disease 222 and ischemic stroke, 223 as well as blood pressure response to antihypertensives 214 The SNP rs3800787 has a reported minor allele frequency of 3 4% in the Hapmap Yoruban and Coriell African American populations. Though our black population had a higher minor allele frequency than expected (approximately 10% ) this SNP was also chosen for gene coverage based on minor allele frequencies in Caucasian and Asian populations of 41 48%. This SNP is

PAGE 105

105 not in significant linkage disequilibrium (r2<0.30) with neighboring SNPs and has unknown in silico function in both pupasuite and FastSNP Thus, while rs3800787 in NOS3 is a strong pharmacological candidate SNP in the present s tudy, little is known about the possible basis for the observed association with this SNP Among insulin signaling genes, a var ian t in IRS1 ( rs1801278 ) was associated with antihypertensive associated change in glucose Insulin binds the insulin receptor ( INSR) signaling insulin receptor tyrosine kinases to phosphorylate the cytosolic protein insulin receptor substrate 1, encoded by IRS1 224 V ariants in IRS1 have been linked to coronary artery disease 225 endothelial dysfunction 226 weight loss 227 insulin resistance 228 and rare mutations associated with diabetes mellitus. 229 230 To our knowledge, this study is t he first to evaluate IRS1 in antihyper tensive associated dysglycemia. IRS1 rs1801278 is predicted to have possibly damaging effects per F SNP and FastSNP IRS1 rs1801278 is a nonsynonymous SNP (glycine 971 to arginine) likely leading altered insulin signal ing involving phosphatidylinositol 3 kinase (PI3K) through altered binding to the p85 subunit 230 231 The rs1801278 SNP has been previously investigated in a diabetic Finnish population, though this SNP was not associated with impaired insulin secretion capacity or insulin action (evaluated by OGTT and eugly cemic clamp methods). 232 Considering IRS1 is the first insulin signaling protein in the cascade of insulin action, is widely expressed, highly conserved across s pecies and tissues 233 defects in IRS1 could lead to significant defects in glucose metabolism. Further, the intergenic region between IRS1 and NYAP2 (or KIAA1486 ) has been linked to numerous metabolic phenotypes in various genome wide association studies

PAGE 106

106 including adiponectin levels (rs925735) 234 adiposity (rs2943650) 235 HDL (rs2972146) 236 triglycerides (rs2972146) 236 and type 2 diabetes (rs7578326 and rs2943641) 88 89 However, the SNPs identified in the aforementioned genome wide association studies are approximately 500kb from IRS1 with no LD to the SNPs in the present study. It is possible that SNPs in and around IRS1 may impact glyc emic changes after treatment with antihypertensive therapy. In preliminary analysis of insulin signaling genes, a weak signal for atenolol monotherapy associated dysglycemia was identified in INSR (rs7508679, data depicted in Appendix D) Over the years, t he insulin receptor has been studied extensively and rare mutations within INSR have been associated with diabetes 237 238 and insulin resistance 239 241 More recently GWASs have identified common intronic INSR poly morphisms are associated with diabetic retinopathy 242 and height. 243 The insulin receptor gene ( INSR ) is more than 180 kilobases long, containing 22 exons. INSR rs7508679 is between exon 20 and 21, in an intron of more than 82 kilobases. F SNP identifies rs7508679 as impacting transcriptional regulation and FastSNP predicts the SNP may be a splicing enhancer. Previous GWA Ss or candidate gene studies have not identified the rs7508679 varian t as associated with glucose or diabetes phenotypes. While this SNP did not maintain significance in final analysis, variation in INSR and other insulin signaling genes may warrant further study in evaluating their impact on medication associated dysglycem ia. T here are several limitations that should be mentioned. Of note, we had limited power to detect associations at lower minor allele frequencies. For the SNPs studied, a minor allele frequency of 0.25 was required to detect a change in glucose of 3 5mg/ dL

PAGE 107

107 thus our study was suited to detect more common variation Though our approach of using a more liberal p value initially, followed by replication in a separate population, allowed for greater confidence in our findings, it is notable that none of our f indings met a true Bonferroni correction. Further, while our initial threshold for significance included either monotherapy or add on therapy associated change in glucose, it is interesting that all associations were first detected in monotherapy. However, t he order the g lucose (see Chapter 3) Summarily, we found multiple significant associations between common variation in SCNN1G, NOS3, and IRS1 and antihypertensive associated alterations in glucose in uncomplicated hypertensive patients. While many of the genes studied have been studied extensively for various phenotypes, the dysglycemia associations identified in our study are novel. Further study is warranted to replicate an d validated these findings.

PAGE 108

108 Figure 4 1. Many genes are involved in insulin signaling and maintaining glucose homeostasis. Multiple arrows indicated many steps are involved in signaling. INSR and GLUT4 are membrane bound receptors. GLUT4 facilitates gl ucose uptake.

PAGE 109

109 Table 4 1. Candidate genes for adverse glycemic effects associated with use of thiazides diuretics and beta blockers. Gene Symbol Chromosome Gene name Pharmacological Targets ADRB1 244 10 Beta 1 adrenergic receptor ADRB2 245 5 Beta 2 adrenergic receptor ADRB3 246 247 8 Beta 3 adrenergic receptor CLCNKB 214 1 Chloride channel Kb NOS3 214 7 Nitric oxide synthase 3 (endothelial) SCNN1G 214 16 Sodium channel, non voltage gated 1, gamma SLC12A1 214 15 Na + /K + /Cl transporter SLC12A3 214 16 Na + /Cl cotransporter, thiazide sensitive Insulin Signaling FOXO1 248 249 13 Forkhead Box O1 GYS1 250 19 Glycogen synthase 1 (muscle) INS 251 252 11 Insulin IRS1 253 2 Insulin receptor substrate 1 IRS2 254 13 Insulin receptor substrate 2 INSR 255 256 19 Insulin receptor SLC2A2 257 258 3 Glucose transporter 2 SLC2A4 259 17 Glucose transporter 4

PAGE 110

110 Table 4 2. P value thresholds. Numbe r of SNPs Initial p value Direction of effect Additive p value Product p value Bonferroni p value Primary analysis, pharmacological candidate genes 27 0.05 0.5 0.05 0.001 25 0.001 85 tag SNP analysis pharmacological and insulin signaling candidate genes 403 0.005 0.5 0.05 0.000125 0.00012 4

PAGE 111

111 Table 4 3 Power to detect glucose diffe rences across MAF ranges by race =0.05 =0.005 MAF 0.05 0.10 0.25 0.05 0.10 0.25 Black (n=292) 3 mg/dL 32% 51% 72% 10% 21% 39% 4 mg/dL 52% 76% 92% 21% 44% 71% 5 mg/dL 71% 91% 99% 38% 70% 92% Non Black (n=447) 3 mg/dL 47% 70% 87% 17% 37% 63% 4 mg/dL 71% 91% 99% 38% 69% 91% 5 mg/dL 88% 99% >99% 63% 91% 99%

PAGE 112

112 Table 4 4 Baseline demographics. All (n= 739 ) ATEN (n=371 ) HCTZ (n=3 68 ) Age 49.0 9.2 48.6 9.2 49.19.3 Sex (% female) 392 (53.0) 210 (56.6) 182 (49.5) Race Black (%) 292 (39.5) 143 (38.5) 149 (40.5) Non black (%) 447 (60.5) 228 (61.5) 219 (59.5) Home SBP (mm Hg) 145.8 10.3 145.09.9 146.7 10.7 Home DB P (mm Hg) 93.8 6.0 93.36.0 94.2 6.0 Home heart rate (bpm) 77.4 9.5 77.79.3 77.19.6 History of hypertension Duration of hypertension (yr) 6.67.1 6.97.0 6.37.1 Family history of hypertension (%) 564 (76.4) 288 (77.6) 276 (75.2) Nev er taken an antihypertensive drug (%) 85 (11.5) 42 (11.3) 43 (11.7) Taking antihypertensive drug at entry (%) 551 (74.6) 280 (75.5 ) 271 (73.6) Smoking status Current smoker (%) 106 (14.3) 47 (12.7) 59 (16.0) Ex smoker (%) 176 (23.8 ) 92 (24.8) 84 (22.8) BMI (kg/m 2 ) 30.7 5.5 30.75.9 30.8 5.1 Waist circumference (cm) 97.813.1 97.413.0 98.113.3 Glucose (mg/dL) 91.111.4 90.811.2 91.511.5 Insulin ( I U/mL) 9.18.0 9.28.5 9.17.5 Data are expressed as mean SD unless otherwise noted.

PAGE 113

113 Figure 4 2 Final PEAR dataset for evaluating the impact of genetic variation on medication associated dysglycemia

PAGE 114

114 Table 4 5 Association between SCNN1G rs13306654 and rs4499239 ( r2=0.85 ) genotype and change in glucose after treatment with HCTZ among PEAR blacks. Response MAF Major Hmz. Het. Minor Hmz. P value rs13306654 0.24 N 81 61 6 HCTZ monotherapy 0.02 5.86 4.17 0. 0020 N 83 46 5 HCTZ add on 1.72 4.15 8.80 0.05 96 rs4499239 0.22 N 85 58 5 HCTZ monotherapy 0.5 5.24 6.9 0. 0025 N 85 45 4 HCTZ add on 1.85 3.89 11.25 0. 0751 Glucose, mg/dL ; P value adjusted for base line glucose an d principal component 1 and 2 ; linear regression model An additive model was used.

PAGE 115

115 Table 4 6 Association between NOS3 rs3 800787 genotype and change in glucose after treatment with HCTZ among PEAR blacks. Response MAF Major Hmz. Het. Minor Hmz. P value rs3800787 0.10 N 119 27 2 HCTZ monotherapy 3.7 0.41 24 0. 0025 N 112 20 1 HCTZ add on 3.53 1.3 0.5 0. 0359 Glucose, mg/dL ; P value adjusted for base line glucose an d principal com ponent 1 and 2; linear regression model. An additive model was used. Results from a dominant model are included in Table C 9 in the appendix.

PAGE 116

116 Table 4 7 Association between IRS1 rs1801278 genotype and change in glucose after treatment with ATEN or HCTZ in PEAR non blacks. Response MAF Major Hmz. Minor allele carriers P value rs1801278 0.049 N 204 22 ATEN monotherapy 1.65 6.30 0. 0041 N 200 19 HCTZ monotherapy 1.10 5.79 0.0348 Glucose, mg/dL ; P value adjusted for base line glucose an d principal component 1 and 2 ; linear regression model. A dominant model was used. Results from an additive model are included in Table C 10 in the appendix.

PAGE 117

117 Change in glucose by IRS1 genotype Figure 4 3 Change in glucose after atenolol monotherapy and hydrochlorothiazide monotherapy by IRS1 rs1801278 genotype in PEAR non blacks. A dominant model was used.

PAGE 118

118 CHAPTER 5 SUMMARY AND CONCLUSION Hypertension is a common disease in the U.S., affecting more than 76 million American adults. 1 B eta blockers and thiazide diuretics are frequently prescribed antihypertensives yet both may adversely alter glucose and influence progression to d iabetes. Concern over beta blocker and diuretic associ ated glycemic effects is well documented. 14 19 48 131 To improve our understanding and identification of persons most at risk for antihypertensive associated adverse glycemic effects, a three step approach was undertaken to describe patient specific factors lin ked to medication associated changes in glucose. We first evaluated whether two of the ADA 67 thresholds for identifying pre diabetes, impaired fasting glucose ( fasting glucose >99mg/dL) and impaired glucose tolerance ( 2 hour OGTT glucose >139mg/dL) were equally sensitive in detecting pre diabetes before and after treatment with atenolol, HCTZ an d the combination. Through multiple methods for analysis, such as pre diabetes classification and AUC analysis, we found fasting and 2 hour OGTT similar in their ability to detect antihypertensive associated changes in glucose. Though our sample size in th e pilot study was small, the data do not trend toward the OGTT having greater sensitivity over fasting glucose in detecting medication associated dysglycemia. Further, we evaluated three surrogate indexes for insulin resistance, HOMA, QUICKI and Matsuda to quantify insulin resistance across visits. Not surprisingly none of these formulas detected significant changes in insulin resistance across study visits. Importantly, our findings suggest that during initiation of antihypertensive therapy when clinician s are possibly more likely to consider adverse glycemic effects of medication choice, there may be no benefit in

PAGE 119

119 evaluating patients using the more time consuming OGTT compared to a simple fasting glucose value. Further, our study included a diverse group of patients, likely to be representative of the clinical, including new, never treated, and previously treated hypertensive patients. Next we investigated the clinical predictors associated with elevations in glucose after antihypertensive use Though ma ny large trials have evaluated predictors for antihypertensive associated diabetes development, 17 25 48 84 less is known regarding the clinical predictors for antihypertensive associated elevations in glucose Through univariate and multivariat e analysis in a derivation cohort then replication in a validation cohort w e evaluated clinical and laboratory parameters for atenolol and HCTZ associated elevations in glucose. I n univariate analysis for atenolol baseline glucose, insulin and uric acid reached the entry threshold p value for further analysis. I n univariate analysis for HCTZ, baseline glucose, insulin, HDL, leng th of therapy and female gender met the entry threshold p value for further analysis Though univariate analysis appears to conf irm findings for clinical predictors for new onset diabetes and new onset impaired fasting glucose from a previous PEAR analysis, 84 most of these predictor s did not survive multivariate modeling. For both medications, only baseline glucose remained significant in multivariate analysis in the derivation cohort. Further, baseline glucose was also a significant predictor for medication associated change in gluc ose in the validation cohort. Interestingly, a lower glucose level at baseline led to larger drug associated changes. Of note, o ur finding was consistent with a similar study evaluating clinical predictors for HCTZ a ssociated elevations in glucose. 111 While overall baseline glucose has low predictive value for drug associated glucose elevations, it is

PAGE 120

120 clear that baseline glucose is a consistent predic tor of atenolol or HCTZ associated elevations in glucose. Additional research is warranted to investigate additional clinical parameters that may predict antihypertensive medication associated elevations in glucose in diverse populations. Finally, we sough t to evaluate the pharmacogenomics of antihypertensive medication associated dysglycemia and evaluated pharmacological target and insulin signaling candidate genes Though a handful of researchers have evaluated this phenotype in their blood pressure respo nse cohorts, 111 112 many of our genes of interest had not previously been studied. We found nomi nally significant associations for dysglycemia for both HCTZ and atenolol, in both blacks and non blacks. In blacks treated with HCTZ, variation in SNPs in SCNN1G ( rs13306654 and rs4499239 r2=0.85 ) was associated with a greater elevation in glucose. Varia tion in t his region of SCNN1G has previously been associated with poorer blood pressure response to HCTZ, 172 173 205 though this is the first report of an association with HCTZ associated elevation in glucose. Also in blacks treated with HCTZ, variation in NOS3 ( rs3800787 ) was associated with less of a change in glucose after HCTZ treatm ent While NOS3 has been linked to multiple diseases, 117 221 223 little is known about this SNP and m ultiple web based tools do not offer insight into possible functional properties. In non blacks treated with either atenolol or HCTZ, variant carriers for rs1801278 in IRS1 had a much greater elevation in glucose after medication treatment. IRS1 rs1801278 is a nonsynonymous SNP, which potentially results in altered insulin signaling involving phosphatidylinositol 3 kinase. 230 231 W e report findings for antihypertensive associated dysglycemia in both pharmacological target genes ( SCNN1G and NOS3 ) and an insulin

PAGE 121

121 signaling gene ( IRS1 ). While in the case of SNPs in SCNN1G and IRS1 previous studies support these SNP s are important in related phenotypes, the SNP in NOS3 has not been previously characterized. Thus, our pharmacogenetic findings for medication associated dysglycemia offer novel findings in previously well characterized genes. However, replication is nece ssary to validate these findings. Overall we have incorporated several methods to describe antihypertensive associated dysglycemia. Through findings from a pilot study focused on phenotype assessment, statistical modeling in a large population, and a pha rmacogenetic study in a diverse population, we have garnered insight in evaluating patients most at risk. O ur findings suggest that fasting glucose is an appropriately sensitive test for detecting antihypertensive medication associated elevation s in glucos e. Further, baseline glucose is a replicated predictor for elevation in glucose associated with antihypertensive therapy. Finally, many genes may be important in antihypertensive associated change in glucose whether associated with the pharmacological age nt, or insulin signaling pathways Yet, m ore research is needed on this topic to fully examine antihype rtensive associated dysglycemia associated with other antihypertensive medications, in dive rse populations, for possibly longer treatment duration s Thou gh the PEAR population is an excellent cohort to initially evaluate clinical and genetic predictors of antihypertensive adverse glycemic effects many of our finds should certainly be considered exploratory. The ongoing PEAR 2 study will offer many opportu nities to f urther evaluate beta blocker and thiazide diuretic associated dysglycemia. Fasting glucose and HbA1c levels will be collected over the course of treatment with metoprolol monotherapy and chlorthalidone monotherapy. Though an

PAGE 122

122 HbA1c threshold is a recent addition to the ADA guideline definitions for pre diabetes, it would be interesting to evaluate drug associated pre diabetes development using HbA1c. Further, PEAR 2 is ideal for evaluating clinical predictors for dysglycemia development particula rly those suggested interesting in univariate analysis in the present study. Though our drug associated dysglycemia genetic finds were nominally significant, PEAR 2 may serve as a population to evaluate these findings as well as other pharmacological or in sulin signaling candidate genes. Finally, as the literature continues to expand on medication associated dysglycemia, we aspire toward improved identification of individuals at the highest risk for this adverse effect and promote a more personalize d approa ch to medication selection.

PAGE 123

123 APPENDIX A DATA AND FIGURES ASS OCIATED WITH CHAPTER 2 Additional tables and figures in Appendix A include further detail related to Chapter 2 that was too detailed, not significant or an alternate analysis. If not specificall y referenced in Chapter 2, this information was included in Appendix A to aid in underst anding of analysis undertaken. Summary and p a tient level data (Table A 1; Figures A 1 and A 2 ), comparisons of data from different laboratory sites (Table A 2 ) and diff erent sub sets (Tables A 3, A 4, A 5 and A 6) or phenotypes (Table s A 7, A 8 and A 9; Figure A 10) is depicted.

PAGE 124

124 Table A 1 Summary data for g lucose, log glucose, insulin and log insulin area under the curve (AUC 0 2 ) during a two hour oral glucose toler ance test. Glucose units, min*mg/dL; insulin units, min* u IU/mL Glucose AUC 0 1 2 0 (min*mg/dL) Insulin AUC 0 120 (min* u IU /mL) Participant Log Log ATEN visit 1 Mean 16844.52 254.24 3067.43 134.20 Median 16793.00 253.73 3083.50 149.71 SD 3209.34 10.62 1634.26 40.09 Minimum 12990.00 238.85 855.50 36.91 Maximum 25960.00 274.07 5133.00 180.28 HCTZ visit 1 Mean 16833.86 252.28 6266.64 168.07 Median 18464.50 256.75 4539.50 171.96 SD 3773.07 21.25 5271.29 49.38 Minimum 1 0211.50 203.01 1078.00 91.36 Maximum 20372.50 281.72 19559.00 248.36 ATEN visit 2 Mean 16725.17 257.07 4665.92 150.91 Median 16756.00 256.46 3846.00 147.92 SD 3440.40 16.06 3402.03 46.78 Minimum 10921.00 231.32 1147.50 38.38 Maximum 21410.00 280.71 14120.00 223.73 HCTZ visit 2 Mean 16298.05 254.99 6814.00 170.97 Median 16869.00 249.28 5077.50 190.56 SD 3541.23 16.06 5335.43 58.15 Minimum 11315.00 231.57 678.50 94.92 Maximum 23026.50 292.91 19258.00 248.98

PAGE 125

125 T able A 1. Continued Glucose AUC 0 120 (min*mg/dL) Insulin AUC 0 120 ( min* u IU /mL) Participant Visit Log Log ATEN visit 3 Mean 18094.12 255.57 4291.81 148.07 Median 17010.00 255.79 3262.00 160.08 SD 3619.20 12.83 2956.72 45.09 Minimum 12853.00 225.42 1320.00 38.13 Maximum 24671.50 274.28 14722 .00 210.04 HCTZ visit 3 Mean 16592.61 251.85 6814.00 169.59 Median 18323.50 255.02 5011.50 184.88 SD 4800.58 21.82 5335.43 56.63 Minimum 8316.00 216.10 678.50 76.24 Maximu m 22450.00 288.34 14722.00 230.35

PAGE 126

126 Table A 2. Fasting glucose and insulin collected during OGTT & fasting glucose and insulin collected for core laboratory values, restricted to sub study patients Sub study measured PEAR core measured Glucose I nsulin Glucose Insulin N Mean SD Mean SD Mean SD Mean SD Combined Visit 1 26 95.42 13.57 5.54 6.96 97.87 15.61 10.23 6.64 Visit 2 26 95.23 13.61 7.38 11.90 95.50 14.63 11.15 11.81 Visit 3 22 97.91 10.65 7.41 8.87 98.87 11.51 11.62 8.90 ATEN Visit 1 15 95.87 13.22 3.67 2.53 97.60 15.37 8.84 5.04 Visit 2 15 98.60 14.93 7.67 15.06 100.43 15.32 12.53 14.79 Visit 3 13 99.62 11.68 6.38 9.90 100.27 12.63 10.11 8.98 HCTZ Visit 1 11 94.82 14.65 8.09 9.99 9 8.23 16.69 12.13 12.00 Visit 2 11 90.64 10.51 7.00 6.00 88.77 10.96 9.27 5.99 Visit 3 9 95.44 9.04 8.89 7.44 97.05 10.22 13.59 8.86

PAGE 127

127 Table A 3 Fasting glucose by treatment arm in PEAR, UF PEAR, UF PEAR OGTT main study collected labs, UF PEAR OGTT su b study collected labs. Glucose in mg/dL. Visit N Mean SD Median Range Min Max ATEN PEAR 1 386 91.10 11.86 89.25 90.50 57.50 148.00 2 385 93.45 12.91 92.00 119.50 55.00 174.50 3 357 95.90 13.30 94.50 101.00 72.50 173.50 PEAR UF 1 128 92.48 15.10 89.25 90.50 57.50 148.00 2 127 95.04 16.87 92.50 116.50 58.00 174.50 3 111 97.59 13.98 96.00 82.00 72.50 154.50 PEAR UF OGTT 1 15 97.60 15.37 92.50 53.50 79.00 132.50 2 15 100.43 15.32 99.50 57.50 81.50 139.00 3 13 100.27 12.63 98.00 43.00 76.00 119.00 PEAR UF OGTT (Sub Study) 1 15 95.87 13.22 96.00 42.00 82.00 124.00 2 15 98.60 14.93 98.00 56.00 79.00 135.00 3 13 99.62 11.68 95.00 36.00 81.00 117.00 HCTZ PEAR 1 381 92.03 12.50 91.00 100.50 55.00 155.50 2 378 94.19 14.87 92 .00 177.00 56.50 233.50 3 352 96.21 12.94 94.50 73.00 67.50 140.50 PEAR UF 1 126 93.18 15.05 92.00 100.50 55.00 155.50 2 125 94.89 18.98 92.00 177.00 56.50 233.50 3 111 97.23 14.22 96.50 71.50 68.50 140.00 PEAR UF OGTT 1 11 98.23 16.69 95.50 57.50 77.00 134.50 2 11 88.77 10.96 88.50 39.00 74.50 113.50 3 9 96.83 10.82 96.00 30.00 82.00 112.00 PEAR UF OGTT (Sub Study) 1 11 94.82 14.65 93.00 52.00 73.00 125.00 2 11 90.64 10.51 92.00 32.00 72.00 104.00 3 9 95.44 9.04 97.00 31.00 77.00 108.00

PAGE 128

128 Table A 4 Fasting glucose in PEAR, UF PEAR, UF PEAR OGTT main study collected labs, UF PEAR OGTT sub study collected labs Visit N Mean SD Median Range Min Max PEAR 1 767 91.56 12.18 90.00 100.50 55.00 155.50 2 763 93.82 13.91 92.00 178.50 55.00 233. 50 3 709 96.06 13.11 94.50 106.00 67.50 173.50 PEAR UF 1 254 92.83 15.05 90.75 100.50 55.00 155.50 2 252 94.97 17.91 92.00 177.00 56.50 233.50 3 222 97.41 14.07 96.25 86.00 68.50 154.50 PEAR UF OGTT 1 26 97.87 15.61 94.00 57.50 77.00 134.50 2 26 95.50 14.63 92.00 64.50 74.50 139.00 3 22 98.86 11.78 97.5 43.00 76.00 119.00 PEAR UF OGTT (Sub study) 1 26 95.42 13.57 94.50 52.00 73.00 125.00 2 26 95.23 13.61 95.50 63.00 72.00 135.00 3 22 97.91 10.65 96.00 40.00 77.00 117.00

PAGE 129

129 Table A 5. Absol ute change between study visits in glucose in PEAR, UF PEAR, UF PEAR OGTT main study collected labs, UF PEAR OGTT sub study collected labs. Glucose in mg/dL. Visit N Mean SD Median Range Min Max PEAR 2 1 762 2.26 12.77 2.50 167.50 57.00 110.50 3 2 709 2.04 12.45 2.00 193.50 124.50 69.00 3 1 708 4.39 11.69 3.50 125.50 54.00 71.20 PEAR UF 2 1 251 2.12 18.13 2.00 167.50 57.00 110.50 3 2 222 2.12 15.92 2.50 163.00 124.50 38.50 3 1 221 4.47 14.22 3.50 107.00 54.00 53.00 PEAR UF OGTT 2 1 26 2.3 7 13.97 3.25 71.50 32.00 39.50 3 2 22 2.80 12.35 3.50 50.00 25.50 24.50 3 1 22 1.30 10.84 3.25 37.50 18.50 19.00 PEAR UF OGTT (Sub Study) 2 1 26 0.19 12.05 1.00 60.00 21.00 39.00 3 2 22 2.00 8.87 4.50 40.00 20.00 20.00 3 1 22 3.00 8.99 4.5 0 32.00 11.00 21.00

PAGE 130

130 Table A 6. Change in glucose in PEAR, UF PEAR, UF PEAR OGTT main study collected labs, UF PEAR OGTT sub study collected labs. Glucose in mg/dL. Visit N Mean SD Median Range Min Max ATEN PEAR 2 1 385 2.33 12.87 2.50 163.5 0 57.00 106.50 3 2 357 2.10 13.01 2.00 163.66 94.66 69.00 3 1 357 4.46 12.77 3.50 159.66 88.16 71.50 PEAR UF 2 1 127 2.47 18.82 2.00 163.50 57.00 106.50 3 2 111 2.18 13.16 2.00 114.50 76.00 38.50 3 1 111 5.04 12.77 4.00 86.00 33.00 53.00 PE AR UF OGTT 2 1 15 2.83 13.80 0.50 55.00 15.50 39.50 3 2 13 2.96 10.95 5.00 40.50 25.50 15.00 3 1 13 2.23 10.38 1.00 30.50 16.50 14.00 PEAR UF OGTT (Sub Study) 2 1 15 2.73 12.45 0.0 54.00 15.00 39.00 3 2 13 1.46 8.97 2.00 30.00 20.00 10.00 3 1 13 2.23 8.61 3.00 28.00 9.00 19.00 HCTZ 2 1 3 2 3 1 2 1 3 2 3 1 2 1 3 2 3 1 2 1 3 2 3 1

PAGE 131

131 Table A 7 Blood pressure in PEAR, UF PEAR, UF PEAR OGTT sub study. SBP, systolic blood pressure; DBP, diastolic blood pressu re. Blood pressure in mmHg. SBP DBP ATEN Visit N Mean SD Median N Mean SD Median PEAR 1 378 144.97 9.92 144.46 378 93.31 5.95 92.64 2 381 136.94 13.05 135.44 381 85.53 8.38 85.07 3 357 128.93 11.25 125.55 357 81.17 7.67 81.00 PEAR UF 1 121 144.85 10.27 143.88 121 92.80 5.54 92.17 2 124 136.62 13.05 134.88 124 85.07 8.58 84.62 3 111 128.89 11.76 128.00 111 80.61 7.91 80.44 PEAR UF OGTT 1 15 144.24 9.71 144.77 15 93.66 5.23 94.38 2 15 140.79 13.41 140.08 14 88.53 6.98 87.38 3 13 130.02 13.26 131.81 13 83.33 8.53 83.00 HCTZ PEAR 1 377 94.20 6.00 93.60 2 376 88.89 7.55 89.16 3 355 80.79 7.24 80.20 PEAR UF 1 122 93.63 5.5 7 93.08 2 123 88.39 7.23 88.00 3 113 81.23 7.20 80.71 PEAR UF OGTT 1 11 91.04 5.54 91.75 2 11 88.86 7.68 88.00 3 10 88.17 5.44 89.0 0

PAGE 132

132 Table A 8. Heart rate in PEAR, UF PEAR, UF PEAR OGTT sub study. Heart rate in beats per minute. ATEN Visit N Mean SD Median PEAR 1 2 3 PEAR UF 1 2 3 PEAR UF OGTT 1 2 3 HCTZ PEAR 1 2 3 PEAR UF 1 2 3 PEAR UF OGTT 1 2 3

PAGE 133

133 Table A 9. UF PEAR OGTT sub study stratified by a change in glucose of greater than or equal to 10mg/dL. <10 > 10 Visit N Mean SD Median N Mean SD Median All 2 1 21 5.24 8.04 4.50 3 13.67 2.25 13.50 3 2 16 0.38 7.76 0.25 5 18.60 4.80 20.50 3 1 14 3.79 8.26 3.25 7 13.29 2.93 13.00 ATEN 2 1 11 3.45 7.25 2.00 3 13.67 2.25 13.50 3 2 11 2.55 7.79 5.00 1 1 5.00 3 1 9 4.94 7.03 4.50 4 13.00 1.41 13.50 HCTZ 2 1 10 7.2 8.78 5.75 3 2 5 4.40 5.72 7.50 4 19.50 5.03 20.50 3 1 5 1.70 10.69 3.50 3 13.67 4.73 12.00

PAGE 134

134 Figure A 1 Glucose at each time point for each participant. A) Atenolol treated arm, baseline ( visit 1 ) B) Atenolol treated a rm, ATEN monotherapy ( visit 2 ) C) Atenolol treated arm, ATEN+HCTZ ( visit 3 ) D) Hydrochlorothiazide treated arm, baseline ( visit 1 ) E) Hydrochlorothiazide treated arm HCTZ monotherapy ( visit 2 ) F) Atenolol Arm Hydrochlorothiazide Arm A. Baseline (visit 1) D. Baseline ( visit 1) B. ATEN Monotherapy (visit 2) E. HCTZ Monotherapy (visit 2) C. ATEN+HCTZ (visit 3) F. HCTZ+ATEN (visit 3)

PAGE 135

135 Figure A 2 Insulin at each time point for each participant. A) Atenolol treated a rm, baseline ( visit 1 ) B) Atenolol treated arm, ATEN monotherapy ( visit 2 ) C) Atenolol treated arm, ATEN + HCTZ ( visit 3 ) D) Hydrochlorothiazide treated arm, baseline ( visit 1 ) E) Hydrochlorothiazide treated arm, HCTZ monotherapy ( visit 2 ) F) Hydrochl orothiazide treated arm, HCTZ + ATEN ( visit 3 ) Atenolol Arm A. Baseline (visit 1) B. ATEN monotherapy (visit 2) C. ATEN + HCTZ (visit 3) Hydrochlorothiazide Arm D. Baseline (visit 1) E. HCTZ monotherapy (visit 2) F. HCTZ + ATEN (visit 3)

PAGE 136

136 Table A 10. Log transformed glucose and insulin AUC at baseline, after nine weeks of antihypertensive monotherapy, then after an additional nine weeks wit h a second antihypertensive agent Bar graphs of the data below are depicted in Figure 2 4. Visit 1 Visit 2 Visit 3 P* P** P*** Glucose ATEN (n=15) Baseline ATEN +HCTZ 254.24 + 10.62 257.07 + 16.06 255.57 + 12.83 0. 5036 0. 9569 0. 7348 HCTZ (n=11) Baseline HCTZ +ATEN 252.28 + 21.25 254.99 + 1 6.06 251.85 + 21.82 0. 6664 0. 7475 0. 9692 Insulin ATEN (n=15) Baseline ATEN +HCTZ 136.20 + 40.09 150.91 + 46.78 148.07 + 45.09 0. 1953 0. 6857 0. 5770 HCTZ (n=11) Baseline HCTZ +ATEN 168.07 + 49.38 170.97 + 58.15 16 9 .59 + 56.63 0. 8183 0. 7920 0. 9864 P* value compares baseline to monotherapy. P** value compares monotherapy to combination therapy. P*** value compares across baseline, monotherapy, and combination therapy time points. Glucose units, min*mg/dL; insulin units, min*IU/mL

PAGE 137

137 APPENDIX B DATA AND FIGURES ASSOCIATED WITH CHAPTER 3 Additional tables and figures in Appendix B include further detail related to Chapter 3 that was too detailed, not significant or an alternate analysis. If not specifically referenced in Chapter 3, this information was in cluded in Appendix B to aid in understanding of analysis undertaken. Additional analysis ( Table B 1 ), alternate analysis (Tables B 2, B 3, B 4, B 5; B 6, B 7, B 8, B 9 and B 10; Figures B 1, B 2, B 3 and B 4) are detailed. Body S ize P arameter S election Mu ltiple measures to evaluate body size were evaluated including waist circumference, body mass index (BMI), body surface area (BSA), BSA 2 and waist to hip ratio. Univariate analysis for each variable was evaluated (Table B 1 ) BMI and waist circumference w ere selected for further evaluation based on previous literature. S ensitivity analyses were conducted using only waist circumference and bo th waist circumference and BMI Ultimately, findings were similar in the validation analysis, whether BMI or waist ci rcumference was selected. Cohort Selection and Model Building Two sample sets were selected for evaluating medication associated dysglycemia. Initial analysis was conducted in the first half of PEAR participants (derivation cohort, approximately 418 possib le patients), then the second half of patients (validation cohort, approximately 350 possible patients). The second sample set was randomly generated from the full PEAR population using proc surveyselect, the simple random sampling method and seed 1234567 Tables B 2, B 3, B 4, B 5, and B 6, Figures B 1 and B 2 are findings from the first half (derivation) and second half (validation) division of PEAR

PAGE 138

138 patients. All of the findings from the first and second half of pear involved stepwise selection in the de rivation cohort followed by validation. Tables B 1, B 7, B 8, B 9, and B 10, Figures B 3 and B 4 are findings from the randomly generated derivation and validation cohorts. The findings from the randomly generated cohorts follow the same statistical method s outlined in Chapter 3 with the exception of the variables age, race, gender, and waist circumference forced into the model. This analysis was originally justified in that these variables are commonly associated with new onset diabetes associated with ant ihypertensive medication use. However, since these variables have not been proven to be associated with elevations in glucose associated with antihyper tensive medication use and were not significant in univariate analysis this analysis was revised as desc ribed in Chapter 3.

PAGE 139

139 Table B 1 Univariate analysis for body size parameters, for medicatio ns associated change in glucose, derivation cohort. Atenolol Hydrochlorothiazide Predictor measured + SE P value + SE P value Waist circumference (cm) 0.009 0.0 4 0.8215 0.0 3 0.04 0.4308 Body mass index (kg/m 2 ) 0.05 0.09 0.5638 0.08 0.09 0.3863 Body surface area (m 2 ) 1.10 2.38 0.6449 2.90 2.22 0.1922 Body surface area 2 0.18 0.59 0.7596 0.70 0.55 0.2043 Waist to hip ratio 3.64 6.51 0.5760 0.97 6.10 0.8743

PAGE 140

140 Table B 2 Stepwise regression, ( model generated ) p redictors for atenolol associated change i n glucose in derivation cohort and evalua tion of predictors in validation cohort. Waist circumference included. Partial R2 Model R2 Parameter Estimate P va lue Derivation BL Glucose (mg/dL) 0.1534 0. 1534 0. 3207 <0.0001 Male gender 0.02 82 0. 1816 3.0522 0.00 04 Validation BL Glucose (mg/dL) 0. 2144 0. 2144 0. 4071 <0.0001 Male gender 0. 0016 0. 2160 0.8791 0. 4188 Derivation regression equation BL= baseline.

PAGE 141

141 Table B 3 Stepwise regression, ( model generated ) p redictors for hydrochlorothiazide associated change i n glucose in derivation cohort and evaluatio n of predictors in validation cohort. Waist circumference included. Partial R2 Model R2 Parameter Estimate P value Derivation BL Glucose (mg/dL) 0. 1330 0. 1330 0.3175 <0.0001 Waist circumference (cm) 0. 0161 0. 1491 0.1142 0.0092 Male gender 0. 0091 0. 1 582 1.9581 0. 0485 Validation BL Glucose (mg/dL) 0.0936 0.0936 0.2845 <0.0001 Waist circumference (cm) 0.0280 0.1217 0.1203 0.0016 Male gender 0.000 0.1217 0.0142 0.98 Derivation regression equation BL= baseline.

PAGE 142

142 Table B 4 Correlation between model (Table B 2) predicted atenolol associated change in glucose, i n validation cohort, to actual drug associated change in glucose. Mean Standard Deviation Minimum Maximum (mg/dL) 3. 06 3. 91 16.99 15.25 (mg/dL) 2. 08 10. 62 3 3 .00 39.50 R 0.44 P value <0.0001

PAGE 143

143 Figu re B 1 Correlation between model (Table B 4) predicted atelolol associated change in glucose, in validation cohort, to actual drug associated change in glucose

PAGE 144

144 Table B 5 Correlation between model (Table B 3) predicted HCTZ associated change in gluc ose, in validation cohort, to actual drug associated change in glucose Mean Standard Deviation Minimum Maximum Predicted glucose (mg/dL) 3.31 3.74 12.18 16.73 Actual glucose (mg/dL) 1.26 9.18 32.00 29.00 R 0.34 P value <0.0001

PAGE 145

145 Figure B 2 Correlation between model (Table B 5) predicted HCTZ associated change in glucose, in validation cohort, to actu al drug associated change in glucose.

PAGE 146

146 Table B 6 Full model all variables forced, for medication associated change in glucose, derivation cohort. Atenolol Hydrochlorothiazide Predictor measured + SE P value + SE P value Intercept 28.59 9.85 0.0039 7.06 11.08 0.5244 Age (years) 0.05 0.05 0.3660 0.05 0.06 0.3864 Sex (male) 3.19 1.13 0.0051 2.43 1.28 0.06 Race (black) 2.26 1.13 0.0454 0.25 1.27 0.8431 Assignment (atenolol) 1.96 1.05 0.0637 1.97 1.28 0.1240 Length of therapy (days) 0.06 0.04 0 .0804 0.05 0.04 0.1727 Waist circumference (cm) 0.07 0.04 0.0687 0.11 0.04 0.0113 Drinks per week 0.07 0.14 0.6284 0.19 0.17 0.3091 Current smoker (%) 1.13 1.51 0.4528 3.66 1.64 0.0264 e GFR (mL/min per 1.73m 2 ) 0.02 0.02 0.4630 0.05 0.02 0.0469 Ho me SBP (mm Hg) 0.02 0.06 0.7498 0.05 0.06 0.3714 Home DBP (mm Hg) 0.04 0.10 0.6983 0.02 0.10 0.8464 Pulse (bpm) 0.02 0.06 0.7472 0.10 0.06 0.1160 Glucose (mg/dL) 0.33 0.04 <0.0001 0.33 0.05 <0.0001 Insulin ( I U/mL) 0.05 0.06 0.3883 0.06 0.08 0.411 4 Uric Acid (mg/dL) 0.73 0.34 0.0329 0.29 0.43 0.4961 Potassium (mEq/L) 0.70 1.02 0.4926 2.46 1.12 0.0284 Total cholesterol (mg/dL) 0.01 0.04 0.4215 0.03 0.02 0.0736 Triglyceride (mg/dL) 0.003 0.01 0.6715 0.01 0.01 0.0989 HDL (mg/dL) 0.06 0.04 0 .1656 0.10 0.05 0.0303 LDL (mg/dL) 0.10 0.05 0.0303 0.05 0.05 0.3637 BMI = body mass index, HTN = hypertension, eGFR = estimated glomerular filtration rate, SBP = systolic blood pressure, DBP = diastolic blood pressure.

PAGE 147

147 Table B 7 Predictors for at enolol associated change i n glucose in derivation cohort and evalu ation of predictors in validation cohort. Age, gender, race and waist circumference forced. Partial R2 Model R2 Parameter Estimate P value Derivation Intercept 24.46383 <0.0001 Age 0.0022 0.0022 0.05187 0.3386 Male gender 0.0024 0.0046 1.07289 0.3202 Black race 0.0014 0.0060 0.79935 0.4580 Waist circumference 0.0049 0.0109 0.05988 0.1592 BL Glucose 0.1436 0.1545 0.33556 <0.0001 Validation Intercept 34.04164 <0 .0001 Age 0.0001 0.0001 0.00343 0.9466 Male gender 0.0137 0.0138 1.81979 0.0544 Black race 0.0015 0.0153 0.78741 0.3988 Waist circumference 0.0066 0.0219 0.06181 0.0814 BL Glucose 0.2260 0.2479 0.42007 <0.0001 Derivation regression equation BL= baseline.

PAGE 148

148 Table B 8 Pre dictors for hydrochlorothiazide associated cha nge in glucose in derivation cohort and evaluati on of predictors in validation cohort. Age, gender, race and waist circumference forced. Partial R2 Model R2 Parameter Estimate P value Derivation Intercept 13.74351 0.0083 Age 0.0107 0.0107 0.10736 0 .0369 Male gender 0.0068 0.0175 1.69238 0.0964 Black race 0.0007 0.0182 0.55316 0.5791 Waist circumference 0.0216 0.0398 0.11828 0.0031 BL Glucose 0.1268 0.1666 0.30360 <0.0001 Validation Intercept 13.42001 0.0145 Age 0.0091 0.0091 0.10624 0.0513 Male gender 0.0001 0.0092 0.42442 0.6800 Black race 0.0002 0.0094 0.22142 0.8268 Waist circumference 0.0254 0.0348 0.11815 0.0016 BL Glucose 0.0844 0.1192 0.30263 <0.0001 Derivation regression equation BL= baseline.

PAGE 149

149 Table B 9 Correlation between model (Table B 7) predicted atenolol associated change in glucose, in validation cohort, to actual atenolol associated change in glucose. Mean Standard Deviation Minimum Maximum (mg/dL) 2.16 3.72 12.90 13.37 (mg/dL) 2.14 9.51 24 .00 36.50 R 0.49 P value <0.0 001

PAGE 150

150 Figure B 3 Correlation between model (Table B 7) predicted atenolol associated change in glucose, in validation cohort, to actual drug associated change in glucose

PAGE 151

151 Table B 10. Correlation between model (Table B 8) predicted HCTZ associated change in glucose, in validation cohort, to actual drug associated change in glucose. Mean Standard Deviation Minimum Maximum Predicted glucose (mg/dL) 2.27 3.49 12.80 11.94 Actual glucose (mg/dL) 2.42 9.54 32.00 40.00 R 0.34 P value <0.0001

PAGE 152

152 Figure B 4 Correlation between model (Table B 8) predicted HCTZ associated change in glucose, in va lidation cohort, to act ual HCTZ associated change in glucose.

PAGE 153

153 APPENDIX C DATA AND FIGURES ASSOCIATED WITH CHAPTER 4 Additional tables and figures in Appendix C include further detail related to Chapter 4 that was too detailed, not significant or an alternate analysis. If not specifically referenced in Chapter 4, this information was included in Appendix C to aid in understanding of analysis undertaken. Additional analysis (Tables C 1 and C 8), SNP lists (Tables C 2 and C 3), comprehensive SNP level data (Tables C 4, C 5, C 6, and C 7), and alternate analysis or models (Tables C 9, C 10, C 11, C 12, C 13, C 14 and C 15 ) is included.

PAGE 154

154 Table C 1 Baseline demographics with p values All (n= 739 ) ATEN (n=371 ) HCTZ (n=368 ) P value Age 49.0 9.2 48.6 9.2 49.19.3 0.61 Sex (% fema le) 392 (53.0) 210 (56.6) 182 (49.5) 0.05 Race Black (%) 292 (39.5) 143 (38.5) 149 (40.5) Non black (%) 447 (60.5) 228 (61.5) 219 (59.5) 0.59 Home SBP (mm Hg) 145.8 10.3 145.09.9 146.7 10.7 0.03 Home DBP (mm Hg) 93.8 6.0 93.36.0 94.2 6.0 0.06 Home heart rate (bpm) 77.4 9.5 77.79.3 77.19.6 0.36 History of hypertension Duration of hypertension (yr) 6.67.1 6.97.0 6.37.1 0.33 Family history of hypertension (%) 564 (76.4) 288 (77.6) 276 (75.2) 0.71 Never taken an antihyp ertensive drug (%) 85 (11.5) 42 (11.3) 43 (11.7) 0.88 Taking antihypertensive drug at entry (%) 551 (74.6) 280 (75.5 ) 271 (73.6) 0.57 Smoking status Current smoker (%) 106 (14.3) 47 (12.7) 59 (16.0) 0.19 Ex smoker (%) 176 (23.8) 92 (24. 8) 84 (22.8) 0.53 BMI (kg/m 2 ) 30.7 5.5 30.75.9 30.8 5.1 0.8 4 Waist circumference (cm) 97.813.1 97.413.0 98.113.3 0. 46 Glucose (mg/dL) 91.111.4 90.811.2 91.511.5 0.40 Insulin ( I U/mL) 9.18.0 9.28.5 9.17.5 0.78 Data are expressed as mean SD unless otherwise noted.

PAGE 155

155 Table C 2 Pharmacological candidate genes and SNPs Gene Priority* Primary analysis Secondary analysis ADRB1 1 rs1801252, rs1801253 rs10787516, rs10885531, rs11196612, rs11196613, rs12357112, rs17875421, rs17875422, rs17875425 rs17875434, rs17875435, rs17875437, rs17875440, rs17875441, rs17875445, rs17875447, rs17875449, rs17875451, rs17875453, rs17875467, rs17875473, rs17875474,rs17875475, rs17875477, rs17875478, rs17875479, rs17875480, rs17875482, rs17875483, rs2050395, rs24 29511, rs2773469, rs3813720, rs4917675, rs7074495, rs7076938, rs7093444, rs728422, rs7894700, rs7901971, rs7917912, rs7921133 ADRB2 1 rs1042713, rs1042714, rs1800888, rs3729943 rs1042718, rs1042719, rs1042720, rs1432622, rs17778257, rs1801704, rs2895795 rs33942282, rs33947624, rs33968470, rs34056949, rs34385707, rs34417021, rs34421087, rs34556816, rs34589808, rs34708630, rs34927547, rs34952098, rs34989676, rs35118767, rs35140776, rs35372244, rs35636011, rs35731905, rs35933628, rs3729943, rs3777124, rs77 02861 ADRB3 1 rs4994, rs36031925, ** rs28364012 ** rs28364013 rs10481461, rs28434339, rs34031968, rs34424060, rs34434657, rs34444065, rs34619057, rs34659602, rs35138363, rs35155257, rs35361594, rs35381288, rs35646917, rs35656808, rs35730538, rs3586114 6, rs35875779, rs35925053, rs35937033, rs36031925, rs4999, rs802162, rs9693898, rs9694197 CLCNKB 2 ** rs2015352 rs11588392, rs5253 rs10803410, rs10803412, rs10803414, rs10927894, rs11588392, rs12015135, rs2863435, rs5253

PAGE 156

156 Table C 2 Continued Gene Priority* Primary analysis Secondary analysis NOS3 1 rs3918196 rs3918166 rs3918232 rs3918201 rs3918234 rs3918235 rs1008140, rs10276930, rs12703107, rs1800780, rs1800782, rs1808593, rs2853792, rs2853795, rs2853796, rs3730003, rs3730009, rs3730 011, rs3730305, rs3793342, rs3918156, rs3918157, rs3918158, rs3918161, rs3918164, rs3918169, rs3918170, rs3918171, rs3918175, rs3918177, rs3918185, rs3918186, rs3918187, rs3918193, rs3918196, rs3918198,rs3918199, rs3918203, rs3918207, rs3918211, rs3918213, rs3918214, rs3918222, rs3918227, rs3918228, rs3918229, rs3918233, rs3918235, rs3918238, rs3918239, rs41377046, rs743506, rs7830, rs867225, rs891511, rs891512 SCNN1G 2 rs7200952 rs13306654 rs4499239 rs5736 rs5738 rs13306654, rs4309398, rs4341 748, rs4401050, rs4499239, rs4967986, rs5736, rs5737, rs5738, rs7200952, rs7404408 SLC12A1 2 ** rs1552311 rs10851463, rs11636073, rs12438818, rs12593807, rs12907018, rs1320052, rs1484552, rs1531916, rs16960661, rs16960679, rs16960680, rs16960681, rs169606 82, rs16960698, rs17350938, rs2279366, rs2279369, rs3784614, rs3825960, rs4082950, rs6493317, rs7165179, rs7179027, rs8025278, rs8028035, rs8032941, rs8033546, rs9806541, rs9806747, rs9920281 SLC12A3 3 rs11643718 rs12708965 rs11076172, rs1138429, rs 11643718, rs12446689, rs12449275, rs12708965, rs12932041, rs13306673, rs13306677, rs1436424, rs16963619, rs17367381, rs2099107, rs2304478, rs2304482, rs2304483, rs35762591, rs4567697, rs4784732, rs6499858, rs6499860, rs711746, rs711747, rs711748, rs7184451 rs7185859, rs7204044, rs8043560, rs8046857, rs8056954, rs8063406 *Priority level = 1 indicates priority 1 genes per HumanCVD chip, minor allele frequency (MAF) >0.02, r2 >0.8. Coverage = 2, MAF >0.05, r2>0.5. Coverage = 3, lower priority gene on HumanC VD chip, only non synonymous SNPs and known functional variants of MAF >0.01 were captured. **SNP did not pass QC, excluded from analysis.

PAGE 157

157 Table C 3 Insulin signaling genes and SNPs Gene Priority* TagSNPs FOXO1A 2 rs12876443, rs17446614, rs2701858, r s2701859, rs2755211, rs2984121, rs2995991, rs3858869, rs3908774, rs4943795, rs7139990, rs7325210, rs7981045, rs9603776 GYS1 2 rs1042265, rs12610125, rs12980441, rs2270936, rs2287755, rs5460, rs5464, rs8104760 INS 1 rs10743149, rs10770141, rs10840489, rs10840491, rs2070762, rs3842768, rs4320932, rs689, rs7119275, rs7924316, rs7950050 IRS1 1 rs10170579, rs10181778, rs10182336, rs10498212, rs1078533, rs13018009, rs13423855, rs16822574, rs16822579, rs16822601, rs16822604, rs16822626, rs16822630, rs1682 2638, rs17208470, rs1801118, rs1801123, rs1801278, rs2229613, rs2234931, rs2288586, rs2435182, rs2435185, rs3731594, rs3731596, rs3769647, rs4675094, rs4675096, rs6725330, rs6725556, rs7567312 IRS2 1 rs1044364, rs12583454, rs1865434, rs4771646, rs477308 7, rs4773088, rs4773092, rs4773094, rs7323191, rs7981705, rs7982446, rs7997595, rs7999797, rs913949, rs9515120, rs9521510, rs9559646, rs9559656, rs9583423, rs9587980

PAGE 158

158 Table C 3. Continued Gene Priority* TagSNPs INSR 1 rs1035939, rs10402346, rs1040416 6, rs10404318, rs10408111, rs10408374, rs10410272, rs10411667, rs10411676, rs10415205, rs10415589, rs10416539, rs10417205, rs10420008, rs10420382, rs10424224, rs10426094, rs10500204, rs11667110, rs11668751, rs11671297, rs11672739, rs12150997, rs12459488, r s12460089, rs12460755, rs12609995, rs12610022, rs12971499, rs12979424, rs13306446 rs1346490, rs1549616, rs16990074, rs16994200, rs16994220, rs16994221, rs16994316, rs17175790, rs17175860, rs17254521, rs1896639, rs2042902, rs2059807, rs2115386, rs2229428, r s2303672, rs2860172, rs2860173, rs2860175, rs2860177, rs2860183, rs2860184, rs3745544, rs3745545, rs3745550, rs3745551, rs3815901, rs3852876, rs4247374, rs4499341, rs4804195, rs4804219, rs4804366, rs4804377, rs4804404, rs4804414, rs4804418, rs4804428, rs48 04433, rs6417197, rs6510947, rs6510949, rs6510950, rs6510952, rs6510955, rs6510959, rs6510960, rs6510976, rs7245562, rs7252268, rs7254060, rs7254921, rs7255710, rs7258382, rs7507911, rs7508679, rs8100109, rs8101064, rs8102612, rs8103483, rs8104314, rs81054 06, rs8108622, rs8109559, rs8111710, rs8112883, rs890859, rs890860, rs890861, rs891087, rs891088, rs919275 SLC2A2 2 rs10513684, rs10513685, rs11711437, rs11924032, rs11925298, rs12488694, rs16855638, rs2229608, rs5396, rs5397, rs5400, rs7610064, rs98283 78 SLC2A4 2 rs16956647, rs2073476, rs222847, rs2654185, rs35198331, rs5411, rs5412, rs5415, rs5435, rs5436, rs8076649, rs9902838 *Coverage = 1 indicates priority 1 genes per HumanCVD chip, minor allele frequency (MAF) >0.02, r2 >0.8. Coverage = 2, MAF >0.05, r2>0.5. Coverage = 3, lower priority gene on HumanCVD chip, only non synonymous SNPs and known functional variants of MAF >0.01 were captured.

PAGE 159

159 Table C 4 Results from pharmacological candidate genes functional SNPs Gene SNP Race MAF Tx PE P val ue Tx PE P value ADRB1 rs1801252 black 0.229 ATEN 0.884 0.131 +ATEN 1.578 0.214 ADRB1 rs1801253 black 0.416 ATEN 0.035 0.922 +ATEN 2.329 0.046 ADRB2 rs1042713 black 0.500 ATEN 0.130 0.741 +ATEN 1.526 0.224 ADRB2 rs1042714 black 0.149 ATEN 0.702 0.198 +ATEN 0.410 0.812 ADRB2 rs1800888 black 0.008 ATEN 0.132 0.737 +ATEN 4.328 0.539 ADRB2 rs3729943 black 0.074 ATEN 0.776 0.167 +ATEN 3.602 0.108 ADRB3 rs36031925 black 0.027 ATEN 0.599 0.252 +ATEN 5.893 0.164 ADRB3 rs4994 black 0.114 ATEN 0.090 0.813 + ATEN 0.876 0.611 ADRB1 rs1801252 non black 0.124 ATEN 0.050 0.891 +ATEN 1.043 0.457 ADRB1 rs1801253 non black 0.257 ATEN 0.159 0.693 +ATEN 1.253 0.277 ADRB2 rs1042713 non black 0.381 ATEN 0.432 0.370 +ATEN 0.553 0.597 ADRB2 rs1042714 non black 0.435 A TEN 1.251 0.056 +ATEN 0.958 0.341 ADRB2 rs1800888 non black 0.007 ATEN 0.046 0.899 +ATEN 3.132 0.742 ADRB2 rs3729943 non black 0.000 ATEN 0.000 1.000 +ATEN 0.000 1.000 ADRB3 rs36031925 non black 0.000 ATEN 0.000 1.000 +ATEN 0.000 1.000 ADRB3 rs4994 non black 0.090 ATEN 0.084 0.824 +ATEN 1.971 0.200 CLCNKB rs11588392 black 0.247 HCTZ 0.958 0.501 +HCTZ 0.545 0.690 CLCNKB rs5253 black 0.331 HCTZ 1.863 0.135 +HCTZ 0.864 0.484 NOS3 rs3918166 black 0.065 HCTZ 0.224 0.929 +HCTZ 1.081 0.622 NOS3 rs39181 96 black 0.054 HCTZ 1.732 0.532 +HCTZ 1.358 0.624 NOS3 rs3918232 black 0.000 HCTZ 0.000 1.000 +HCTZ 0.000 1.000 NOS3 rs3918201 black 0.030 HCTZ 4.166 0.251 +HCTZ 2.513 0.461 NOS3 rs3918234 black 0.000 HCTZ 0.000 1.000 +HCTZ 0.000 1.000 NOS3 rs391823 5 black 0.000 HCTZ 0.000 1.000 +HCTZ 0.000 1.000 SCNN1G rs5736 black 0.034 HCTZ 1.492 0.656 +HCTZ 3.714 0.136 SCNN1G rs5738 black 0.003 HCTZ 0.000 1.000 +HCTZ 6.821 0.147 SCNN1G rs7200952 black 0.279 HCTZ 2.367 0.076 +HCTZ 0.282 0.833 SCNN1G rs44 99239 black 0.221 HCTZ 4.422 0.002 +HCTZ 2.632 0.075 SCNN1G rs13306654 black 0.238 HCTZ 4.404 0.002 +HCTZ 2.704 0.060 SLC12A3 rs11643718 black 0.027 HCTZ 2.183 0.556 +HCTZ 5.720 0.143 SLC12A3 rs12708965 black 0.074 HCTZ 1.956 0.443 +HCTZ 0.410 0.843 CLCNKB rs11588392 non black 0.036 HCTZ 1.357 0.528 +HCTZ 1.769 0.461 CLCNKB rs5253 non black 0.017 HCTZ 6.832 0.051 +HCTZ 6.038 0.038 NOS3 rs3918166 non black 0.001 HCTZ 0.000 1.000 +HCTZ 6.720 0.447 NOS3 rs3918196 non black 0.070 HCTZ 3.355 0.059 +H CTZ 1.882 0.236 NOS3 rs3918232 non black 0.002 HCTZ 0.700 0.874 +HCTZ 0.590 0.888 NOS3 rs3918201 non black 0.001 HCTZ 0.000 1.000 +HCTZ 2.758 0.762 NOS3 rs3918234 non black 0.003 HCTZ 5.352 0.306 +HCTZ 0.000 1.000 NOS3 rs3918235 non black 0.003 HC TZ 0.000 1.000 +HCTZ 5.947 0.222 SCNN1G rs5736 non black 0.003 HCTZ 0.000 1.000 +HCTZ 1.828 0.628 SCNN1G rs5738 non black 0.009 HCTZ 7.545 0.086 +HCTZ 4.649 0.055 SCNN1G rs7200952 non black 0.361 HCTZ 0.533 0.534 +HCTZ 0.651 0.411 SCNN1G rs449 9239 non black 0.183 HCTZ 1.219 0.245 +HCTZ 0.097 0.924 SCNN1G rs13306654 non black 0.187 HCTZ 1.009 0.341 +HCTZ 0.135 0.893 SLC12A3 rs11643718 non black 0.083 HCTZ 0.748 0.620 +HCTZ 1.286 0.423 SLC12A3 rs12708965 non black 0.017 HCTZ 0.332 0.924 +HCTZ 3.705 0.218 MAF= minor allele frequency; PE= parameter estimate; Tx= treatment

PAGE 160

160 Table C 5 Results from SNPs in beta blocker pharmacological candidate genes. Gene SNP Race MAF Tx PE P value Tx PE P value ADRB1 rs7076938 black 0.413 ATEN 0.092 0.809 +ATEN 3.082 0.011 ADRB1 rs7894700 black 0.025 ATEN 1.423 0.038 +ATEN 4.854 0.201 ADRB1 rs4917675 black 0.302 ATEN 0.304 0.496 +ATEN 1.282 0.278 ADRB1 rs17875473 black 0.012 ATEN 0.566 0.272 +ATEN 0.169 0.977 ADRB1 rs17875434 black 0.039 ATEN 0.089 0 .815 +ATEN 1.826 0.498 ADRB1 rs17875435 black 0.060 ATEN 0.117 0.763 +ATEN 0.920 0.667 ADRB1 rs17875474 black 0.032 ATEN 0.238 0.578 +ATEN 7.219 0.044 ADRB1 rs2429511 black 0.272 ATEN 0.413 0.386 +ATEN 2.031 0.139 ADRB1 rs2050395 black 0.272 ATEN 0.8 55 0.140 +ATEN 1.251 0.309 ADRB1 rs17875440 black 0.018 ATEN 0.792 0.161 +ATEN 2.173 0.597 ADRB1 rs17875475 black 0.000 ATEN 0.000 1.000 +ATEN 0.000 1.000 ADRB1 rs12357112 black 0.084 ATEN 0.141 0.722 +ATEN 3.005 0.168 ADRB1 rs17875441 black 0.042 ATE N 1.209 0.062 +ATEN 5.688 0.057 ADRB1 rs7921133 black 0.131 ATEN 0.480 0.331 +ATEN 3.057 0.097 ADRB1 rs17875477 black 0.000 ATEN 0.000 1.000 +ATEN 0.000 1.000 ADRB1 rs17875447 black 0.008 ATEN 0.418 0.382 +ATEN 12.136 0.034 ADRB1 rs17875449 black 0.0 02 ATEN 0.000 1.000 +ATEN 2.982 0.764 ADRB1 rs3813720 black 0.478 ATEN 0.221 0.602 +ATEN 2.742 0.018 ADRB1 rs17875451 black 0.064 ATEN 0.576 0.266 +ATEN 4.215 0.078 ADRB1 rs17875453 black 0.012 ATEN 0.054 0.884 +ATEN 2.074 0.722 ADRB1 rs7074495 blac k 0.231 ATEN 1.167 0.068 +ATEN 1.546 0.210 ADRB1 rs11196612 black 0.087 ATEN 0.114 0.769 +ATEN 2.265 0.238 ADRB1 rs17875479 black 0.023 ATEN 0.086 0.820 +ATEN 5.902 0.156 ADRB1 rs11196613 black 0.199 ATEN 0.446 0.358 +ATEN 2.138 0.126 ADRB1 rs1787548 0 black 0.007 ATEN 0.007 0.984 +ATEN 0.296 0.967 ADRB1 rs17875467 black 0.042 ATEN 0.334 0.464 +ATEN 5.294 0.105 ADRB1 rs17875482 black 0.002 ATEN 0.079 0.834 +ATEN 0.000 1.000 ADRB1 rs728422 black 0.101 ATEN 0.252 0.560 +ATEN 3.525 0.070 ADRB1 rs1787 5483 black 0.000 ATEN 0.000 1.000 +ATEN 0.000 1.000 ADRB1 rs7901971 black 0.032 ATEN 0.120 0.758 +ATEN 2.474 0.413 ADRB1 rs10787516 black 0.134 ATEN 0.150 0.709 +ATEN 0.923 0.599 ADRB1 rs7093444 black 0.227 ATEN 1.000 0.100 +ATEN 1.481 0.304 ADRB1 rs1 0885531 black 0.193 ATEN 0.121 0.756 +ATEN 1.271 0.396 ADRB1 rs17875421 black 0.082 ATEN 0.214 0.611 +ATEN 0.557 0.790 ADRB1 rs17875422 black 0.020 ATEN 0.208 0.619 +ATEN 7.752 0.121 ADRB1 rs7917912 black 0.065 ATEN 0.032 0.928 +ATEN 2.846 0.245 ADRB 1 rs17875425 black 0.045 ATEN 0.106 0.783 +ATEN 1.167 0.700 ADRB1 rs2773469 black 0.408 ATEN 0.109 0.778 +ATEN 2.455 0.053 ADRB2 rs1432622 black 0.412 ATEN 0.209 0.619 +ATEN 0.116 0.926 ADRB2 rs17778257 black 0.232 ATEN 0.065 0.862 +ATEN 1.698 0.209 A DRB2 rs35140776 black 0.000 ATEN 0.000 1.000 +ATEN 0.000 1.000 ADRB2 rs2895795 black 0.352 ATEN 0.302 0.499 +ATEN 1.427 0.292 ADRB2 rs34927547 black 0.010 ATEN 0.059 0.873 +ATEN 6.652 0.186 ADRB2 rs33947624 black 0.057 ATEN 0.640 0.229 +ATEN 0.035 0.99 0 ADRB2 rs1801704 black 0.150 ATEN 0.702 0.198 +ATEN 0.517 0.764

PAGE 161

161 Table C 5 Continued. Gene SNP Race MAF Tx PE P value Tx PE P value ADRB2 rs1042718 black 0.347 ATEN 0.302 0.499 +ATEN 1.348 0.317 ADRB2 rs35933628 black 0.002 ATEN 0.000 1.000 +ATEN 2.178 0.827 ADRB2 rs1042719 black 0.364 ATEN 0.119 0.761 +ATEN 1.626 0.215 ADRB2 rs1042720 black 0.433 ATEN 0.332 0.465 +ATEN 1.120 0.393 ADRB2 rs3777124 black 0.002 ATEN 0.607 0.247 +ATEN 0.000 1.000 ADRB2 rs33942282 black 0.002 ATEN 0.017 0.961 +ATE N 0.000 1.000 ADRB2 rs33968470 black 0.002 ATEN 0.033 0.926 +ATEN 0.000 1.000 ADRB2 rs34056949 black 0.032 ATEN 0.123 0.753 +ATEN 5.229 0.125 ADRB2 rs34589808 black 0.159 ATEN 0.097 0.799 +ATEN 0.400 0.811 ADRB2 rs34708630 black 0.109 ATEN 0.641 0.229 +ATEN 0.297 0.876 ADRB2 rs34385707 black 0.000 ATEN 0.000 1.000 +ATEN 0.000 1.000 ADRB2 rs34556816 black 0.018 ATEN 0.851 0.141 +ATEN 5.750 0.199 ADRB2 rs34989676 black 0.039 ATEN 0.785 0.164 +ATEN 0.680 0.820 ADRB2 rs35636011 black 0.025 ATEN 0.065 0.862 +ATEN 7.204 0.108 ADRB2 rs34952098 black 0.010 ATEN 0.274 0.532 +ATEN 0.731 0.900 ADRB2 rs35731905 black 0.034 ATEN 0.362 0.435 +ATEN 0.289 0.927 ADRB2 rs34421087 black 0.000 ATEN 0.000 1.000 +ATEN 0.000 1.000 ADRB3 rs9694197 black 0.319 ATEN 0.178 0.663 +ATEN 1.588 0.241 ADRB3 rs34619057 black 0.082 ATEN 0.406 0.393 +ATEN 1.247 0.552 ADRB3 rs35730538 black 0.060 ATEN 0.366 0.431 +ATEN 0.166 0.944 ADRB3 rs34434657 black 0.081 ATEN 0.113 0.771 +ATEN 1.113 0.607 ADRB3 rs35646917 black 0.045 A TEN 0.128 0.744 +ATEN 0.805 0.795 ADRB3 rs4999 black 0.122 ATEN 0.078 0.836 +ATEN 2.604 0.162 ADRB3 rs35155257 black 0.007 ATEN 0.143 0.720 +ATEN 0.418 0.953 ADRB3 rs35138363 black 0.012 ATEN 0.092 0.809 +ATEN 3.614 0.537 ADRB3 rs35381288 black 0.022 ATEN 0.105 0.786 +ATEN 2.358 0.490 ADRB3 rs35361594 black 0.000 ATEN 0.000 1.000 +ATEN 0.000 1.000 ADRB1 rs7076938 non black 0.263 ATEN 0.135 0.732 +ATEN 1.421 0.209 ADRB1 rs7894700 non black 0.000 ATEN 0.000 1.000 +ATEN 0.000 1.000 ADRB1 rs4917675 non black 0.261 ATEN 0.127 0.746 +ATEN 0.052 0.960 ADRB1 rs17875473 non black 0.093 ATEN 0.137 0.729 +ATEN 1.341 0.388 ADRB1 rs17875434 non black 0.000 ATEN 0.000 1.000 +ATEN 0.000 1.000 ADRB1 rs17875435 non black 0.000 ATEN 0.000 1.000 +ATEN 0.000 1. 000 ADRB1 rs17875474 non black 0.042 ATEN 0.225 0.595 +ATEN 0.601 0.826 ADRB1 rs2429511 non black 0.480 ATEN 0.188 0.649 +ATEN 1.088 0.261 ADRB1 rs2050395 non black 0.122 ATEN 0.020 0.956 +ATEN 1.159 0.403 ADRB1 rs17875440 non black 0.000 ATEN 0.000 1. 000 +ATEN 0.000 1.000 ADRB1 rs17875475 non black 0.004 ATEN 0.264 0.544 +ATEN 1.272 0.893 ADRB1 rs12357112 non black 0.124 ATEN 0.028 0.938 +ATEN 1.172 0.395 ADRB1 rs17875441 non black 0.000 ATEN 0.000 1.000 +ATEN 0.000 1.000 ADRB1 rs7921133 non black 0.001 ATEN 0.000 1.000 +ATEN 5.731 0.554 ADRB1 rs17875477 non black 0.005 ATEN 0.131 0.739 +ATEN 2.874 0.670 ADRB1 rs17875447 non black 0.000 ATEN 0.000 1.000 +ATEN 0.000 1.000 ADRB1 rs17875449 non black 0.008 ATEN 1.644 0.023 +ATEN 1.472 0.788 ADRB1 rs3813720 non black 0.373 ATEN 0.364 0.432 +ATEN 1.755 0.085 ADRB1 rs17875451 non black 0.000 ATEN 0.000 1.000 +ATEN 0.000 1.000 ADRB1 rs17875453 non black 0.001 ATEN 0.000 1.000 +ATEN 1.706 0.867

PAGE 162

162 Table C 5 Continued. Gene SNP Race MAF Tx PE P val ue Tx PE P value ADRB1 rs7074495 non black 0.161 ATEN 0.250 0.562 +ATEN 0.453 0.677 ADRB1 rs11196612 non black 0.124 ATEN 0.007 0.985 +ATEN 1.652 0.228 ADRB1 rs17875479 non black 0.031 ATEN 0.427 0.374 +ATEN 0.521 0.886 ADRB1 rs11196613 non black 0.1 24 ATEN 0.020 0.956 +ATEN 1.652 0.228 ADRB1 rs17875480 non black 0.049 ATEN 0.568 0.270 +ATEN 0.414 0.824 ADRB1 rs17875467 non black 0.000 ATEN 0.000 1.000 +ATEN 0.000 1.000 ADRB1 rs17875482 non black 0.008 ATEN 0.397 0.401 +ATEN 3.345 0.545 ADRB1 rs 728422 non black 0.000 ATEN 0.000 1.000 +ATEN 0.000 1.000 ADRB1 rs17875483 non black 0.012 ATEN 1.226 0.059 +ATEN 0.167 0.963 ADRB1 rs7901971 non black 0.000 ATEN 0.000 1.000 +ATEN 0.000 1.000 ADRB1 rs10787516 non black 0.498 ATEN 0.314 0.485 +ATEN 0. 683 0.477 ADRB1 rs7093444 non black 0.124 ATEN 0.028 0.938 +ATEN 1.172 0.395 ADRB1 rs10885531 non black 0.497 ATEN 0.314 0.485 +ATEN 0.751 0.438 ADRB1 rs17875421 non black 0.000 ATEN 0.000 1.000 +ATEN 0.000 1.000 ADRB1 rs17875422 non black 0.036 ATEN 0.414 0.385 +ATEN 0.456 0.857 ADRB1 rs7917912 non black 0.144 ATEN 0.037 0.918 +ATEN 1.276 0.280 ADRB1 rs17875425 non black 0.000 ATEN 0.000 1.000 +ATEN 0.000 1.000 ADRB1 rs2773469 non black 0.263 ATEN 0.000 0.999 +ATEN 0.691 0.558 ADRB2 rs1432622 non black 0.438 ATEN 1.316 0.048 +ATEN 1.051 0.299 ADRB2 rs17778257 non black 0.382 ATEN 1.122 0.075 +ATEN 0.574 0.578 ADRB2 rs35140776 non black 0.001 ATEN 0.724 0.189 +ATEN 0.000 1.000 ADRB2 rs2895795 non black 0.180 ATEN 0.120 0.758 +ATEN 0.585 0.635 ADRB2 rs34927547 non black 0.000 ATEN 0.000 1.000 +ATEN 0.000 1.000 ADRB2 rs33947624 non black 0.000 ATEN 0.000 1.000 +ATEN 0.000 1.000 ADRB2 rs1801704 non black 0.435 ATEN 1.192 0.064 +ATEN 0.958 0.341 ADRB2 rs1042718 non black 0.149 ATEN 0.160 0.692 +ATEN 1.933 0.163 ADRB2 rs35933628 non black 0.000 ATEN 0.000 1.000 +ATEN 0.000 1.000 ADRB2 rs1042719 non black 0.279 ATEN 0.272 0.534 +ATEN 0.354 0.738 ADRB2 rs1042720 non black 0.316 ATEN 0.655 0.221 +ATEN 1.371 0.201 ADRB2 rs3777124 non black 0.00 2 ATEN 0.027 0.940 +ATEN 8.411 0.367 ADRB2 rs33942282 non black 0.030 ATEN 0.011 0.974 +ATEN 0.089 0.973 ADRB2 rs33968470 non black 0.008 ATEN 0.373 0.423 +ATEN 1.274 0.817 ADRB2 rs34056949 non black 0.000 ATEN 0.000 1.000 +ATEN 0.000 1.000 ADRB2 rs3 4589808 non black 0.000 ATEN 0.000 1.000 +ATEN 0.000 1.000 ADRB2 rs34708630 non black 0.000 ATEN 0.000 1.000 +ATEN 0.000 1.000 ADRB2 rs34385707 non black 0.000 ATEN 0.000 1.000 +ATEN 0.000 1.000 ADRB2 rs34556816 non black 0.000 ATEN 0.000 1.000 +ATEN 0. 000 1.000 ADRB2 rs34989676 non black 0.000 ATEN 0.000 1.000 +ATEN 0.000 1.000 ADRB2 rs35636011 non black 0.000 ATEN 0.000 1.000 +ATEN 0.000 1.000 ADRB2 rs34952098 non black 0.000 ATEN 0.000 1.000 +ATEN 0.000 1.000 ADRB2 rs35731905 non black 0.000 ATEN 0.000 1.000 +ATEN 0.000 1.000 ADRB2 rs34421087 non black 0.000 ATEN 0.000 1.000 +ATEN 0.000 1.000 ADRB3 rs9694197 non black 0.093 ATEN 0.084 0.824 +ATEN 2.051 0.174 ADRB3 rs34619057 non black 0.090 ATEN 0.084 0.824 +ATEN 1.971 0.200 ADRB3 rs35730538 non black 0.001 ATEN 0.000 1.000 +ATEN 4.494 0.635 ADRB3 rs34434657 non black 0.000 ATEN 0.000 1.000 +ATEN 0.000 1.000 ADRB3 rs35646917 non black 0.000 ATEN 0.000 1.000 +ATEN 0.000 1.000

PAGE 163

163 Table C 5 Continued. Gene SNP Race MAF Tx PE P value Tx PE P va lue ADRB3 rs4999 non black 0.000 ATEN 0.000 1.000 +ATEN 0.000 1.000 ADRB3 rs35155257 non black 0.000 ATEN 0.000 1.000 +ATEN 0.000 1.000 ADRB3 rs35138363 non black 0.000 ATEN 0.000 1.000 +ATEN 0.000 1.000 ADRB3 rs35381288 non black 0.000 ATEN 0.000 1.00 0 +ATEN 0.000 1.000 ADRB3 rs35361594 non black 0.003 ATEN 1.007 0.098 +ATEN 0.000 1.000 MAF= minor allele frequency; PE= parameter estimate; Tx= treatment

PAGE 164

164 Table C 6 Results from SNPs in thiazide diuretic pharmacological candidate genes. Gene SNP Race MAF Tx PE P value Tx PE P value CLCNKB rs10803410 black 0.415 HCTZ 1.042 0.346 +HCTZ 0.254 0.824 CLCNKB rs2863435 black 0.389 HCTZ 0.168 0.890 +HCTZ 0.644 0.576 CLCNKB rs10927894 black 0.134 HCTZ 1.078 0.573 +HCTZ 0.419 0.789 CLCNKB rs10803412 black 0.359 HCTZ 0.479 0.700 +HCTZ 0.487 0.697 CLCNKB rs12015135 black 0.252 HCTZ 0.486 0.722 +HCTZ 1.276 0.314 CLCNKB rs10803414 black 0.227 HCTZ 0.146 0.914 +HCTZ 1.918 0.176 NOS3 rs3918169 black 0.445 HCTZ 0.983 0.400 +HCTZ 0.107 0.924 NOS3 rs39181 70 black 0.023 HCTZ 2.455 0.508 +HCTZ 3.773 0.381 NOS3 rs3793342 black 0.089 HCTZ 1.343 0.453 +HCTZ 0.332 0.884 NOS3 rs1549758 black 0.111 HCTZ 0.501 0.790 +HCTZ 0.381 0.831 NOS3 rs3918171 black 0.081 HCTZ 2.535 0.189 +HCTZ 3.034 0.146 NOS3 rs154186 1 black 0.107 HCTZ 1.381 0.459 +HCTZ 1.614 0.414 NOS3 rs3918175 black 0.007 HCTZ 2.137 0.718 +HCTZ 1.269 0.893 NOS3 rs1800780 black 0.427 HCTZ 0.433 0.710 +HCTZ 0.298 0.791 NOS3 rs3730003 black 0.049 HCTZ 0.552 0.833 +HCTZ 4.208 0.171 NOS3 rs1800 782 black 0.190 HCTZ 0.537 0.708 +HCTZ 0.330 0.813 NOS3 rs2853792 black 0.164 HCTZ 0.146 0.935 +HCTZ 0.103 0.946 NOS3 rs3918177 black 0.143 HCTZ 0.654 0.709 +HCTZ 1.783 0.275 NOS3 rs3918227 black 0.010 HCTZ 1.542 0.740 +HCTZ 13.641 0.147 NOS3 rs39182 28 black 0.000 HCTZ 0.000 1.000 +HCTZ 0.000 1.000 NOS3 rs3918185 black 0.030 HCTZ 5.981 0.158 +HCTZ 3.507 0.228 NOS3 rs3918186 black 0.225 HCTZ 0.562 0.688 +HCTZ 1.062 0.415 NOS3 rs3918187 black 0.086 HCTZ 0.662 0.770 +HCTZ 0.510 0.813 NOS3 rs391818 8 black 0.386 HCTZ 0.867 0.504 +HCTZ 0.915 0.447 NOS3 rs3918229 black 0.000 HCTZ 0.000 1.000 +HCTZ 0.000 1.000 NOS3 rs2853795 black 0.052 HCTZ 5.728 0.035 +HCTZ 2.965 0.267 NOS3 rs2853796 black 0.451 HCTZ 1.376 0.289 +HCTZ 0.306 0.791 NOS3 rs3730305 black 0.235 HCTZ 0.969 0.466 +HCTZ 0.764 0.560 NOS3 rs3918193 black 0.039 HCTZ 3.650 0.174 +HCTZ 0.498 0.892 NOS3 rs891511 black 0.419 HCTZ 0.776 0.534 +HCTZ 0.287 0.809 NOS3 rs1008140 black 0.193 HCTZ 1.136 0.456 +HCTZ 2.584 0.065 NOS3 rs3918198 black 0.084 HCTZ 2.356 0.276 +HCTZ 1.444 0.466 NOS3 rs3918199 black 0.003 HCTZ 2.622 0.797 +HCTZ 5.348 0.567 NOS3 rs743506 black 0.421 HCTZ 1.962 0.104 +HCTZ 0.322 0.778 NOS3 rs3730009 black 0.177 HCTZ 1.984 0.190 +HCTZ 1.973 0.196 NOS3 rs867225 b lack 0.112 HCTZ 2.424 0.206 +HCTZ 3.195 0.058 NOS3 rs891512 black 0.027 HCTZ 3.305 0.313 +HCTZ 1.927 0.596 NOS3 rs3918203 black 0.039 HCTZ 2.752 0.428 +HCTZ 0.747 0.819 NOS3 rs1808593 black 0.242 HCTZ 1.066 0.477 +HCTZ 1.508 0.245 NOS3 rs7830 black 0.218 HCTZ 1.822 0.205 +HCTZ 1.313 0.348 NOS3 rs3918207 black 0.015 HCTZ 6.553 0.164 +HCTZ 1.574 0.713 NOS3 rs3730011 black 0.010 HCTZ 1.877 0.752 +HCTZ 2.815 0.672 NOS3 rs3918211 black 0.154 HCTZ 0.756 0.662 +HCTZ 0.688 0.644

PAGE 165

165 Table C 6 Contin ued. Gene SNP Race MAF Tx PE P value Tx PE P value NOS3 rs3918238 black 0.000 HCTZ 0.000 1.000 +HCTZ 0.000 1.000 NOS3 rs3918213 black 0.055 HCTZ 0.286 0.913 +HCTZ 1.717 0.537 NOS3 rs3918214 black 0.012 HCTZ 8.049 0.117 +HCTZ 10.115 0.128 NOS3 rs3918239 black 0.000 HCTZ 0.000 1.000 +HCTZ 0.000 1.000 NOS3 rs3800787 black 0.102 HCTZ 5.602 0.003 +HCTZ 4.392 0.036 NOS3 rs2373885 black 0.487 HCTZ 1.410 0.233 +HCTZ 0.766 0.508 NOS3 rs10276930 black 0.007 HCTZ 3.937 0.506 +HCTZ 7.269 0.440 NOS3 rs12703 107 black 0.470 HCTZ 0.742 0.532 +HCTZ 0.491 0.662 NOS3 rs1800783 black 0.453 HCTZ 1.209 0.291 +HCTZ 0.002 0.999 NOS3 rs3918156 black 0.000 HCTZ 0.000 1.000 +HCTZ 0.000 1.000 NOS3 rs3918158 black 0.067 HCTZ 1.747 0.474 +HCTZ 2.171 0.346 NOS3 rs18007 79 black 0.134 HCTZ 0.998 0.557 +HCTZ 3.080 0.083 NOS3 rs3918161 black 0.000 HCTZ 0.000 1.000 +HCTZ 0.000 1.000 NOS3 rs3918226 black 0.008 HCTZ 1.130 0.850 +HCTZ 6.028 0.521 NOS3 rs3918164 black 0.024 HCTZ 5.062 0.237 +HCTZ 1.236 0.671 NOS3 rs180078 1 black 0.025 HCTZ 4.286 0.225 +HCTZ 0.130 0.976 SCNN1G rs7404408 black 0.117 HCTZ 0.113 0.955 +HCTZ 2.594 0.128 SCNN1G rs5737 black 0.067 HCTZ 2.212 0.377 +HCTZ 0.353 0.874 SCNN1G rs13331086 black 0.208 HCTZ 3.559 0.016 +HCTZ 2.740 0.063 SCN N1G rs11074553 black 0.421 HCTZ 2.654 0.033 +HCTZ 0.403 0.726 SCNN1G rs4299163 black 0.362 HCTZ 0.245 0.848 +HCTZ 2.188 0.068 SCNN1G rs4401050 black 0.109 HCTZ 1.489 0.447 +HCTZ 0.531 0.749 SCNN1G rs4309398 black 0.354 HCTZ 0.268 0.835 +HCTZ 2 .613 0.033 SCNN1G rs4341748 black 0.355 HCTZ 0.327 0.800 +HCTZ 2.613 0.033 SCNN1G rs4967986 black 0.169 HCTZ 0.818 0.625 +HCTZ 2.018 0.162 SLC12A1 rs1320052 black 0.468 HCTZ 0.381 0.764 +HCTZ 0.594 0.641 SLC12A1 rs4082950 black 0.104 HCTZ 4.004 0.082 +HCTZ 0.066 0.969 SLC12A1 rs16960661 black 0.275 HCTZ 2.454 0.088 +HCTZ 0.057 0.968 SLC12A1 rs9920281 black 0.344 HCTZ 0.172 0.895 +HCTZ 0.035 0.979 SLC12A1 rs8032941 black 0.149 HCTZ 2.057 0.216 +HCTZ 1.534 0.407 SLC12A1 rs1484552 black 0.0 90 HCTZ 1.109 0.553 +HCTZ 2.145 0.377 SLC12A1 rs16960679 black 0.285 HCTZ 0.924 0.467 +HCTZ 0.840 0.500 SLC12A1 rs16960680 black 0.094 HCTZ 2.177 0.218 +HCTZ 1.460 0.496 SLC12A1 rs16960681 black 0.064 HCTZ 2.014 0.355 +HCTZ 2.154 0.333 SLC12A1 r s16960682 black 0.015 HCTZ 1.621 0.786 +HCTZ 4.201 0.388 SLC12A1 rs12907018 black 0.446 HCTZ 1.400 0.271 +HCTZ 0.608 0.608 SLC12A1 rs3825960 black 0.000 HCTZ 0.000 1.000 +HCTZ 0.000 1.000 SLC12A1 rs3784614 black 0.182 HCTZ 2.526 0.086 +HCTZ 0.311 0.8 42 SLC12A1 rs16960698 black 0.206 HCTZ 1.105 0.411 +HCTZ 0.255 0.858 SLC12A1 rs1531916 black 0.406 HCTZ 1.221 0.305 +HCTZ 0.442 0.693 SLC12A1 rs9806541 black 0.121 HCTZ 1.021 0.506 +HCTZ 0.362 0.862 SLC12A1 rs11636073 black 0.341 HCTZ 0.104 0.934 + HCTZ 0.708 0.552 SLC12A1 rs8028035 black 0.163 HCTZ 0.777 0.633 +HCTZ 1.727 0.276 SLC12A1 rs8033546 black 0.141 HCTZ 3.244 0.043 +HCTZ 1.833 0.223

PAGE 166

166 Table C 6 Continued. Gene SNP Race MAF Tx PE P value Tx PE P value SLC12A1 rs2279366 black 0.435 H CTZ 1.414 0.237 +HCTZ 1.227 0.262 SLC12A1 rs2279369 black 0.091 HCTZ 0.535 0.808 +HCTZ 1.744 0.304 SLC12A1 rs12593807 black 0.461 HCTZ 0.717 0.547 +HCTZ 0.760 0.485 SLC12A1 rs9806747 black 0.079 HCTZ 1.127 0.583 +HCTZ 1.045 0.676 SLC12A1 rs6493317 black 0.403 HCTZ 1.064 0.376 +HCTZ 0.665 0.527 SLC12A1 rs10851463 black 0.148 HCTZ 0.445 0.789 +HCTZ 1.592 0.341 SLC12A1 rs7179027 black 0.433 HCTZ 0.096 0.935 +HCTZ 0.865 0.416 SLC12A1 rs17350938 black 0.087 HCTZ 0.227 0.923 +HCTZ 1.712 0.334 SLC12A 1 rs8025278 black 0.438 HCTZ 0.907 0.448 +HCTZ 1.945 0.074 SLC12A1 rs7165179 black 0.102 HCTZ 2.090 0.324 +HCTZ 2.693 0.135 SLC12A1 rs12438818 black 0.266 HCTZ 2.878 0.033 +HCTZ 1.690 0.182 SLC12A3 rs12932041 black 0.169 HCTZ 1.257 0.463 +HCTZ 0.6 18 0.655 SLC12A3 rs4784732 black 0.087 HCTZ 0.296 0.883 +HCTZ 1.546 0.488 SLC12A3 rs1436424 black 0.386 HCTZ 1.286 0.285 +HCTZ 1.154 0.312 SLC12A3 rs2304478 black 0.002 HCTZ 9.529 0.359 +HCTZ 0.000 1.000 SLC12A3 rs13306673 black 0.211 HCTZ 0.141 0.928 +HCTZ 3.232 0.012 SLC12A3 rs8043560 black 0.195 HCTZ 1.492 0.316 +HCTZ 1.659 0.254 SLC12A3 rs8063406 black 0.492 HCTZ 0.167 0.887 +HCTZ 0.103 0.929 SLC12A3 rs2304483 black 0.346 HCTZ 0.373 0.756 +HCTZ 1.013 0.390 SLC12A3 rs11076172 black 0.102 HCTZ 2.842 0.096 +HCTZ 1.137 0.618 SLC12A3 rs2304482 black 0.072 HCTZ 0.971 0.681 +HCTZ 4.734 0.074 SLC12A3 rs13306677 black 0.082 HCTZ 0.043 0.983 +HCTZ 2.305 0.299 SLC12A3 rs8056954 black 0.326 HCTZ 0.090 0.942 +HCTZ 1.200 0.364 SLC12A3 rs8046857 black 0.119 HCTZ 1.511 0.375 +HCTZ 2.497 0.195 SLC12A3 rs6499858 black 0.315 HCTZ 0.068 0.959 +HCTZ 0.118 0.923 SLC12A3 rs12449275 black 0.248 HCTZ 0.864 0.528 +HCTZ 2.020 0.123 SLC12A3 rs12446689 black 0.099 HCTZ 0.457 0.820 +HCTZ 0.463 0.818 SLC 12A3 rs17367381 black 0.002 HCTZ 0.000 1.000 +HCTZ 5.261 0.574 SLC12A3 rs7184451 black 0.186 HCTZ 0.973 0.551 +HCTZ 3.422 0.022 SLC12A3 rs16963619 black 0.064 HCTZ 0.910 0.721 +HCTZ 0.259 0.905 SLC12A3 rs7185859 black 0.081 HCTZ 0.215 0.927 +HCTZ 3. 073 0.100 SLC12A3 rs7204044 black 0.044 HCTZ 3.674 0.155 +HCTZ 0.979 0.763 SLC12A3 rs1138429 black 0.020 HCTZ 0.935 0.840 +HCTZ 0.125 0.974 SLC12A3 rs2099107 black 0.178 HCTZ 0.931 0.596 +HCTZ 1.302 0.369 SLC12A3 rs6499860 black 0.238 HCTZ 2.175 0.1 12 +HCTZ 2.467 0.069 SLC12A3 rs4567697 black 0.169 HCTZ 1.294 0.352 +HCTZ 1.579 0.371 SLC12A3 rs711748 black 0.329 HCTZ 1.308 0.311 +HCTZ 0.051 0.967 SLC12A3 rs711747 black 0.379 HCTZ 1.249 0.307 +HCTZ 1.081 0.336 SLC12A3 rs711746 black 0.237 HCTZ 2.418 0.090 +HCTZ 0.730 0.577 CLCNKB rs10803410 non black 0.443 HCTZ 0.014 0.987 +HCTZ 2.236 0.006 CLCNKB rs2863435 non black 0.443 HCTZ 0.284 0.757 +HCTZ 2.481 0.002 CLCNKB rs10927894 non black 0.236 HCTZ 1.312 0.196 +HCTZ 1.873 0.046 CLCNKB rs10 803412 non black 0.180 HCTZ 0.267 0.798 +HCTZ 0.599 0.575 CLCNKB rs12015135 non black 0.126 HCTZ 0.038 0.977 +HCTZ 0.044 0.970

PAGE 167

167 Table C 6 Continued. Gene SNP Race MAF Tx PE P value Tx PE P value CLCNKB rs10803414 non black 0.466 HCTZ 0.420 0.623 +H CTZ 1.125 0.173 NOS3 rs3918169 non black 0.144 HCTZ 1.314 0.282 +HCTZ 0.427 0.698 NOS3 rs3918170 non black 0.000 HCTZ 0.000 1.000 +HCTZ 0.000 1.000 NOS3 rs3793342 non black 0.139 HCTZ 1.432 0.256 +HCTZ 0.306 0.785 NOS3 rs1549758 non black 0.320 HCTZ 1.697 0.053 +HCTZ 0.140 0.872 NOS3 rs3918171 non black 0.002 HCTZ 0.845 0.896 +HCTZ 11.765 0.170 NOS3 rs1541861 non black 0.394 HCTZ 2.028 0.013 +HCTZ 0.728 0.371 NOS3 rs3918175 non black 0.000 HCTZ 0.000 1.000 +HCTZ 0.000 1.000 NOS3 rs1800780 non black 0.462 HCTZ 1.559 0.058 +HCTZ 0.140 0.863 NOS3 rs3730003 non black 0.000 HCTZ 0.000 1.000 +HCTZ 0.000 1.000 NOS3 rs1800782 non black 0.072 HCTZ 1.904 0.243 +HCTZ 2.450 0.113 NOS3 rs2853792 non black 0.397 HCTZ 1.878 0.023 +HCTZ 0.264 0.753 NOS3 r s3918177 non black 0.001 HCTZ 0.000 1.000 +HCTZ 2.758 0.762 NOS3 rs3918227 non black 0.098 HCTZ 1.191 0.391 +HCTZ 1.786 0.209 NOS3 rs3918228 non black 0.017 HCTZ 3.972 0.162 +HCTZ 1.517 0.719 NOS3 rs3918185 non black 0.000 HCTZ 0.000 1.000 +HCTZ 0.000 1.000 NOS3 rs3918186 non black 0.074 HCTZ 1.422 0.382 +HCTZ 1.846 0.235 NOS3 rs3918187 non black 0.000 HCTZ 0.000 1.000 +HCTZ 0.000 1.000 NOS3 rs3918188 non black 0.362 HCTZ 1.069 0.226 +HCTZ 0.096 0.910 NOS3 rs3918229 non black 0.005 HCTZ 2.628 0.609 +HCTZ 2.601 0.756 NOS3 rs2853795 non black 0.170 HCTZ 1.375 0.230 +HCTZ 0.644 0.532 NOS3 rs2853796 non black 0.473 HCTZ 1.643 0.048 +HCTZ 0.310 0.705 NOS3 rs3730305 non black 0.074 HCTZ 1.422 0.382 +HCTZ 1.803 0.260 NOS3 rs3918193 non black 0.000 HCT Z 0.000 1.000 +HCTZ 0.000 1.000 NOS3 rs891511 non black 0.330 HCTZ 1.263 0.153 +HCTZ 0.863 0.332 NOS3 rs1008140 non black 0.019 HCTZ 1.310 0.688 +HCTZ 0.020 0.994 NOS3 rs3918198 non black 0.000 HCTZ 0.000 1.000 +HCTZ 0.000 1.000 NOS3 rs3918199 non bl ack 0.000 HCTZ 0.000 1.000 +HCTZ 0.000 1.000 NOS3 rs743506 non black 0.260 HCTZ 2.679 0.003 +HCTZ 0.373 0.694 NOS3 rs3730009 non black 0.000 HCTZ 0.000 1.000 +HCTZ 0.000 1.000 NOS3 rs867225 non black 0.000 HCTZ 0.000 1.000 +HCTZ 0.000 1.000 NOS3 rs8915 12 non black 0.226 HCTZ 2.753 0.003 +HCTZ 0.080 0.934 NOS3 rs3918203 non black 0.001 HCTZ 0.000 1.000 +HCTZ 4.700 0.575 NOS3 rs1808593 non black 0.256 HCTZ 2.633 0.003 +HCTZ 0.589 0.533 NOS3 rs7830 non black 0.339 HCTZ 1.415 0.103 +HCTZ 0.247 0.782 NOS3 rs3918207 non black 0.000 HCTZ 0.000 1.000 +HCTZ 0.000 1.000 NOS3 rs3730011 non black 0.000 HCTZ 0.000 1.000 +HCTZ 0.000 1.000 NOS3 rs3918211 non black 0.001 HCTZ 0.000 1.000 +HCTZ 2.758 0.762 NOS3 rs3918238 non black 0.001 HCTZ 0.000 1.000 +HCTZ 8.579 0.306 NOS3 rs3918213 non black 0.000 HCTZ 0.000 1.000 +HCTZ 0.000 1.000 NOS3 rs3918214 non black 0.001 HCTZ 7.571 0.396 +HCTZ 0.000 1.000 NOS3 rs3918239 non black 0.008 HCTZ 16.367 0.009 +HCTZ 0.001 1.000 NOS3 rs3800787 non black 0.373 HCTZ 1.2 29 0.142 +HCTZ 0.374 0.663 NOS3 rs2373885 non black 0.247 HCTZ 0.470 0.623 +HCTZ 0.261 0.787

PAGE 168

168 Table C 6 Continued. Gene SNP Race MAF Tx PE P value Tx PE P value NOS3 rs10276930 non black 0.000 HCTZ 0.000 1.000 +HCTZ 0.000 1.000 NOS3 rs12703107 non black 0.240 HCTZ 0.502 0.600 +HCTZ 0.194 0.845 NOS3 rs1800783 non black 0.376 HCTZ 1.359 0.104 +HCTZ 0.522 0.530 NOS3 rs3918156 non black 0.002 HCTZ 0.000 1.000 +HCTZ 5.357 0.367 NOS3 rs3918158 non black 0.000 HCTZ 0.000 1.000 +HCTZ 0.000 1.000 NOS3 r s1800779 non black 0.374 HCTZ 1.339 0.110 +HCTZ 0.575 0.490 NOS3 rs3918161 non black 0.001 HCTZ 0.000 1.000 +HCTZ 10.321 0.218 NOS3 rs3918226 non black 0.084 HCTZ 2.981 0.072 +HCTZ 2.192 0.079 NOS3 rs3918164 non black 0.000 HCTZ 0.000 1.000 +HCTZ 0.0 00 1.000 NOS3 rs1800781 non black 0.142 HCTZ 1.508 0.227 +HCTZ 0.489 0.650 SCNN1G rs7404408 non black 0.252 HCTZ 1.473 0.142 +HCTZ 1.313 0.141 SCNN1G rs5737 non black 0.046 HCTZ 1.237 0.511 +HCTZ 1.581 0.424 SCNN1G rs13331086 non black 0.193 HCTZ 1.118 0.291 +HCTZ 0.346 0.722 SCNN1G rs11074553 non black 0.408 HCTZ 0.167 0.850 +HCTZ 0.222 0.781 SCNN1G rs4299163 non black 0.215 HCTZ 0.811 0.424 +HCTZ 0.721 0.473 SCNN1G rs4401050 non black 0.225 HCTZ 1.062 0.293 +HCTZ 0.024 0.981 SCNN1G rs43 09398 non black 0.206 HCTZ 0.800 0.432 +HCTZ 0.671 0.505 SCNN1G rs4341748 non black 0.207 HCTZ 0.743 0.470 +HCTZ 0.671 0.505 SCNN1G rs4967986 non black 0.253 HCTZ 1.473 0.142 +HCTZ 1.249 0.154 SLC12A1 rs1320052 non black 0.016 HCTZ 0.985 0.766 +HCTZ 8.104 0.192 SLC12A1 rs4082950 non black 0.001 HCTZ 0.000 1.000 +HCTZ 4.585 0.594 SLC12A1 rs16960661 non black 0.015 HCTZ 0.985 0.766 +HCTZ 6.854 0.315 SLC12A1 rs9920281 non black 0.019 HCTZ 2.804 0.458 +HCTZ 1.659 0.762 SLC12A1 rs8032941 non black 0.006 HCTZ 4.852 0.276 +HCTZ 27.697 0.002 SLC12A1 rs1484552 non black 0.033 HCTZ 1.502 0.532 +HCTZ 0.569 0.828 SLC12A1 rs16960679 non black 0.010 HCTZ 1.644 0.703 +HCTZ 6.454 0.286 SLC12A1 rs16960680 non black 0.000 HCTZ 0.000 1.000 +HCTZ 0.000 1.000 SLC12A1 rs16960681 non black 0.000 HCTZ 0.000 1.000 +HCTZ 0.000 1.000 SLC12A1 rs16960682 non black 0.016 HCTZ 1.017 0.796 +HCTZ 2.540 0.399 SLC12A1 rs12907018 non black 0.115 HCTZ 0.870 0.493 +HCTZ 1.013 0.435 SLC12A1 rs3825960 non black 0.001 HCTZ 0 .000 1.000 +HCTZ 3.932 0.647 SLC12A1 rs3784614 non black 0.009 HCTZ 0.121 0.977 +HCTZ 10.476 0.014 SLC12A1 rs16960698 non black 0.022 HCTZ 1.087 0.751 +HCTZ 4.148 0.097 SLC12A1 rs1531916 non black 0.179 HCTZ 1.515 0.157 +HCTZ 0.502 0.631 SLC12A1 rs98 06541 non black 0.000 HCTZ 0.000 1.000 +HCTZ 0.000 1.000 SLC12A1 rs11636073 non black 0.145 HCTZ 2.599 0.053 +HCTZ 0.109 0.924 SLC12A1 rs8028035 non black 0.002 HCTZ 5.081 0.457 +HCTZ 0.000 1.000 SLC12A1 rs8033546 non black 0.022 HCTZ 4.521 0.182 +HC TZ 0.981 0.693 SLC12A1 rs2279366 non black 0.187 HCTZ 1.878 0.090 +HCTZ 0.223 0.830 SLC12A1 rs2279369 non black 0.137 HCTZ 1.012 0.417 +HCTZ 0.096 0.932 SLC12A1 rs12593807 non black 0.155 HCTZ 1.305 0.262 +HCTZ 0.305 0.784 SLC12A1 rs9806747 non blac k 0.000 HCTZ 0.000 1.000 +HCTZ 0.000 1.000 SLC12A1 rs6493317 non black 0.204 HCTZ 2.326 0.031 +HCTZ 0.737 0.455 SLC12A1 rs10851463 non black 0.080 HCTZ 3.820 0.018 +HCTZ 1.152 0.398

PAGE 169

169 Table C 6 Continued. Gene SNP Race MAF Tx PE P value Tx PE P value SLC12A1 rs7179027 non black 0.200 HCTZ 2.366 0.028 +HCTZ 0.713 0.464 SLC12A1 rs17350938 non black 0.138 HCTZ 1.104 0.364 +HCTZ 0.604 0.595 SLC12A1 rs8025278 non black 0.047 HCTZ 4.628 0.025 +HCTZ 1.974 0.299 SLC12A1 rs7165179 non black 0.001 HCTZ 8.134 0.364 +HCTZ 0.000 1.000 SLC12A1 rs12438818 non black 0.024 HCTZ 4.429 0.149 +HCTZ 1.232 0.622 SLC12A3 rs12932041 non black 0.265 HCTZ 0.000 1.000 +HCTZ 0.146 0.887 SLC12A3 rs4784732 non black 0.220 HCTZ 0.043 0.966 +HCTZ 0.712 0.475 SLC12A3 rs1 436424 non black 0.413 HCTZ 0.815 0.342 +HCTZ 0.576 0.509 SLC12A3 rs2304478 non black 0.015 HCTZ 1.448 0.660 +HCTZ 0.826 0.831 SLC12A3 rs13306673 non black 0.089 HCTZ 2.047 0.145 +HCTZ 0.275 0.845 SLC12A3 rs8043560 non black 0.403 HCTZ 0.044 0.960 + HCTZ 0.554 0.495 SLC12A3 rs8063406 non black 0.324 HCTZ 0.315 0.733 +HCTZ 1.243 0.158 SLC12A3 rs2304483 non black 0.342 HCTZ 0.350 0.702 +HCTZ 1.155 0.184 SLC12A3 rs11076172 non black 0.199 HCTZ 0.176 0.868 +HCTZ 1.138 0.272 SLC12A3 rs2304482 non black 0.082 HCTZ 0.492 0.746 +HCTZ 0.923 0.579 SLC12A3 rs13306677 non black 0.079 HCTZ 1.310 0.405 +HCTZ 1.920 0.251 SLC12A3 rs8056954 non black 0.182 HCTZ 0.291 0.796 +HCTZ 0.824 0.395 SLC12A3 rs8046857 non black 0.040 HCTZ 0.476 0.842 +HCTZ 1.194 0.560 SLC12A3 rs6499858 non black 0.279 HCTZ 0.128 0.892 +HCTZ 0.899 0.297 SLC12A3 rs12449275 non black 0.220 HCTZ 0.010 0.992 +HCTZ 1.226 0.201 SLC12A3 rs12446689 non black 0.155 HCTZ 0.466 0.711 +HCTZ 0.034 0.975 SLC12A3 rs17367381 non black 0.04 5 HCTZ 2.215 0.308 +HCTZ 1.297 0.500 SLC12A3 rs7184451 non black 0.001 HCTZ 2.300 0.801 +HCTZ 0.000 1.000 SLC12A3 rs16963619 non black 0.000 HCTZ 0.000 1.000 +HCTZ 0.000 1.000 SLC12A3 rs7185859 non black 0.004 HCTZ 0.161 0.985 +HCTZ 6.865 0.123 SLC12 A3 rs7204044 non black 0.243 HCTZ 0.980 0.330 +HCTZ 1.298 0.174 SLC12A3 rs1138429 non black 0.119 HCTZ 1.493 0.248 +HCTZ 0.891 0.431 SLC12A3 rs2099107 non black 0.120 HCTZ 0.063 0.961 +HCTZ 0.914 0.493 SLC12A3 rs6499860 non black 0.459 HCTZ 0.342 0.69 2 +HCTZ 0.737 0.368 SLC12A3 rs4567697 non black 0.469 HCTZ 0.536 0.534 +HCTZ 0.625 0.453 SLC12A3 rs711748 non black 0.467 HCTZ 0.733 0.387 +HCTZ 0.691 0.392 SLC12A3 rs711747 non black 0.473 HCTZ 0.894 0.293 +HCTZ 0.678 0.401 SLC12A3 rs711746 non blac k 0.467 HCTZ 0.733 0.387 +HCTZ 0.688 0.393 MAF= minor allele frequency; PE= parameter estimate; Tx= treatment

PAGE 170

170 Table C 7 Results from SNPs in insulin signaling candidate genes. Gene SNP Race MAF Tx PE P value Tx PE P value FOXO1 rs7325210 black 0.045 ATEN 4.343 0.056 +ATEN 0.636 0.845 FOXO1 rs2701858 black 0.069 ATEN 0.457 0.838 +ATEN 4.045 0.064 FOXO1 rs17446614 black 0.161 ATEN 0.071 0.958 +ATEN 1.534 0.362 FOXO1 rs2701859 black 0.435 ATEN 0.305 0.772 +ATEN 0.638 0.570 FOXO1 rs2755211 black 0.4 35 ATEN 0.305 0.772 +ATEN 0.638 0.570 FOXO1 rs2995991 black 0.069 ATEN 3.988 0.060 +ATEN 1.951 0.435 FOXO1 rs2984121 black 0.029 ATEN 4.990 0.098 +ATEN 4.748 0.223 FOXO1 rs3858869 black 0.060 ATEN 0.031 0.990 +ATEN 1.268 0.623 FOXO1 rs4943795 black 0.436 ATEN 0.570 0.584 +ATEN 0.326 0.776 FOXO1 rs3908774 black 0.057 ATEN 0.318 0.902 +ATEN 1.631 0.505 FOXO1 rs12876443 black 0.005 ATEN 0.000 1.000 +ATEN 3.185 0.591 FOXO1 rs7981045 black 0.096 ATEN 4.062 0.026 +ATEN 5.364 0.009 FOXO1 rs7139990 bl ack 0.154 ATEN 0.029 0.985 +ATEN 1.663 0.327 FOXO1 rs9603776 black 0.005 ATEN 11.844 0.055 +ATEN 16.041 0.109 GYS1 rs2387583 black 0.258 ATEN 0.231 0.850 +ATEN 0.584 0.674 GYS1 rs4645894 black 0.104 ATEN 6.655 0.000 +ATEN 0.999 0.597 GYS1 rs4645900 black 0.000 ATEN 0.000 1.000 +ATEN 0.000 1.000 GYS1 rs905238 black 0.384 ATEN 0.085 0.940 +ATEN 2.472 0.057 GYS1 rs4645903 black 0.154 ATEN 2.885 0.048 +ATEN 0.215 0.898 GYS1 rs4645905 black 0.306 ATEN 0.178 0.871 +ATEN 1.483 0.272 GYS1 rs4645908 bla ck 0.000 ATEN 0.000 1.000 +ATEN 0.000 1.000 GYS1 rs1042265 black 0.086 ATEN 3.447 0.057 +ATEN 1.963 0.395 GYS1 rs8104760 black 0.005 ATEN 3.088 0.727 +ATEN 4.572 0.515 GYS1 rs13306418 black 0.015 ATEN 1.240 0.700 +ATEN 3.867 0.699 GYS1 rs12980441 blac k 0.305 ATEN 0.207 0.852 +ATEN 1.458 0.257 GYS1 rs5464 black 0.289 ATEN 0.210 0.857 +ATEN 0.211 0.865 GYS1 rs5460 black 0.062 ATEN 1.980 0.422 +ATEN 4.568 0.046 GYS1 rs2287755 black 0.111 ATEN 1.717 0.293 +ATEN 1.718 0.354 GYS1 rs12610125 black 0.03 2 ATEN 0.763 0.836 +ATEN 1.965 0.564 INS rs4320932 black 0.112 ATEN 0.958 0.568 +ATEN 0.551 0.774 INS rs7924316 black 0.357 ATEN 0.617 0.595 +ATEN 1.655 0.177 INS rs3842768 black 0.023 ATEN 7.642 0.061 +ATEN 2.098 0.564 INS rs689 black 0.247 ATEN 0.725 0.544 +ATEN 0.941 0.510 INS rs2070762 black 0.208 ATEN 1.603 0.202 +ATEN 2.356 0.132 INS rs6356 black 0.128 ATEN 1.873 0.210 +ATEN 1.627 0.411 INS rs7950050 black 0.128 ATEN 0.673 0.618 +ATEN 2.906 0.128 INSR rs12150997 black 0.379 ATEN 1.376 0.203 +ATEN 0.027 0.982 INSR rs3745551 black 0.229 ATEN 0.151 0.905 +ATEN 1.490 0.248

PAGE 171

171 Table C 7 Continued. Gene SNP Race MAF Tx PE P value Tx PE P value INSR rs3745550 black 0.466 ATEN 0.428 0.698 +ATEN 1.479 0.186 INSR rs6510947 black 0.064 ATE N 0.650 0.737 +ATEN 3.914 0.151 INSR rs10408374 black 0.114 ATEN 1.893 0.258 +ATEN 3.258 0.069 INSR rs2860172 black 0.230 ATEN 2.764 0.027 +ATEN 1.961 0.147 INSR rs13306446 black 0.046 ATEN 0.201 0.936 +ATEN 3.419 0.198 INSR rs2860173 black 0.023 ATEN 0.195 0.952 +ATEN 7.642 0.098 INSR rs2860175 black 0.262 ATEN 0.131 0.912 +ATEN 1.774 0.191 INSR rs1549616 black 0.050 ATEN 0.631 0.774 +ATEN 4.025 0.220 INSR rs10420382 black 0.302 ATEN 0.329 0.769 +ATEN 1.172 0.357 INSR rs11672739 black 0.17 4 ATEN 0.250 0.853 +ATEN 0.618 0.704 INSR rs6510949 black 0.034 ATEN 2.597 0.350 +ATEN 3.111 0.363 INSR rs12610022 black 0.166 ATEN 1.676 0.248 +ATEN 1.312 0.409 INSR rs6510950 black 0.344 ATEN 1.692 0.157 +ATEN 0.768 0.535 INSR rs11667110 black 0 .138 ATEN 0.771 0.622 +ATEN 0.549 0.735 INSR rs16990074 black 0.017 ATEN 1.902 0.607 +ATEN 2.331 0.642 INSR rs16994200 black 0.029 ATEN 1.837 0.507 +ATEN 0.819 0.843 INSR rs8103483 black 0.257 ATEN 0.366 0.769 +ATEN 0.308 0.824 INSR rs16994220 b lack 0.003 ATEN 3.339 0.593 +ATEN 0.000 1.000 INSR rs16994221 black 0.138 ATEN 0.624 0.697 +ATEN 0.937 0.571 INSR rs6510952 black 0.029 ATEN 2.828 0.479 +ATEN 3.645 0.241 INSR rs8102612 black 0.366 ATEN 1.511 0.214 +ATEN 2.782 0.038 INSR rs8100109 black 0.446 ATEN 1.787 0.101 +ATEN 1.373 0.288 INSR rs2059807 black 0.257 ATEN 1.053 0.391 +ATEN 1.243 0.367 INSR rs3815901 black 0.391 ATEN 0.048 0.964 +ATEN 1.282 0.325 INSR rs2860177 black 0.184 ATEN 0.408 0.764 +ATEN 0.664 0.654 INSR rs651095 5 black 0.263 ATEN 1.475 0.237 +ATEN 0.959 0.478 INSR rs7252268 black 0.376 ATEN 1.488 0.188 +ATEN 1.804 0.159 INSR rs10415589 black 0.087 ATEN 3.932 0.090 +ATEN 2.886 0.153 INSR rs8109559 black 0.466 ATEN 1.029 0.366 +ATEN 0.944 0.413 INSR rs10411667 black 0.386 ATEN 1.621 0.149 +ATEN 2.055 0.109 INSR rs8112883 black 0.182 ATEN 0.032 0.980 +ATEN 0.184 0.904 INSR rs8105406 black 0.029 ATEN 5.082 0.111 +ATEN 8.237 0.032 INSR rs16994316 black 0.030 ATEN 5.300 0.122 +ATEN 1.729 0.595 INSR rs8108622 black 0.208 ATEN 1.322 0.332 +ATEN 1.007 0.515 INSR rs10500204 black 0.190 ATEN 1.385 0.313 +ATEN 0.902 0.579 INSR rs6510959 black 0.241 ATEN 0.263 0.840 +ATEN 2.374 0.070 INSR rs891087 black 0.181 ATEN 0.784 0.559 +ATEN 0.021 0.989 INSR rs891088 bl ack 0.423 ATEN 1.120 0.296 +ATEN 0.024 0.983 INSR rs4804366 black 0.298 ATEN 0.268 0.803 +ATEN 1.694 0.227

PAGE 172

172 Table C 7 Continued. Gene SNP Race MAF Tx PE P value Tx PE P value INSR rs10416539 black 0.260 ATEN 0.098 0.934 +ATEN 0.885 0.534 INSR rs1 035939 black 0.383 ATEN 1.300 0.229 +ATEN 1.330 0.239 INSR rs2860183 black 0.305 ATEN 1.279 0.293 +ATEN 0.773 0.555 INSR rs12460089 black 0.461 ATEN 0.645 0.534 +ATEN 0.324 0.763 INSR rs4804377 black 0.208 ATEN 1.188 0.324 +ATEN 0.201 0.894 INSR r s2115386 black 0.428 ATEN 2.515 0.020 +ATEN 2.389 0.040 INSR rs4499341 black 0.371 ATEN 1.165 0.271 +ATEN 1.389 0.243 INSR rs6510960 black 0.428 ATEN 1.676 0.133 +ATEN 0.196 0.872 INSR rs1896639 black 0.328 ATEN 0.139 0.871 +ATEN 0.931 0.381 INSR r s2042902 black 0.218 ATEN 0.207 0.864 +ATEN 0.397 0.796 INSR rs10426094 black 0.072 ATEN 0.941 0.617 +ATEN 3.680 0.118 INSR rs12609995 black 0.144 ATEN 0.168 0.906 +ATEN 1.831 0.287 INSR rs12459488 black 0.305 ATEN 1.604 0.151 +ATEN 0.230 0.859 IN SR rs3745545 black 0.144 ATEN 0.921 0.562 +ATEN 0.761 0.666 INSR rs8104314 black 0.154 ATEN 1.481 0.321 +ATEN 2.549 0.107 INSR rs12971499 black 0.446 ATEN 0.036 0.974 +ATEN 0.192 0.866 INSR rs7245562 black 0.175 ATEN 3.233 0.009 +ATEN 1.088 0.503 INSR rs4804404 black 0.205 ATEN 2.759 0.017 +ATEN 1.078 0.487 INSR rs7508679 black 0.339 ATEN 3.178 0.002 +ATEN 0.098 0.938 INSR rs890859 black 0.087 ATEN 4.198 0.041 +ATEN 0.213 0.903 INSR rs4804414 black 0.243 ATEN 2.862 0.022 +ATEN 0.655 0.650 I NSR rs4804418 black 0.136 ATEN 2.369 0.126 +ATEN 2.598 0.123 INSR rs890860 black 0.285 ATEN 1.391 0.270 +ATEN 0.833 0.514 INSR rs890861 black 0.037 ATEN 1.850 0.434 +ATEN 2.385 0.563 INSR rs7255710 black 0.079 ATEN 2.145 0.295 +ATEN 0.030 0.990 INS R rs12460755 black 0.050 ATEN 1.076 0.642 +ATEN 4.461 0.127 INSR rs17175790 black 0.029 ATEN 1.569 0.558 +ATEN 2.027 0.653 INSR rs10420008 black 0.218 ATEN 2.331 0.060 +ATEN 0.674 0.629 INSR rs17175860 black 0.121 ATEN 1.689 0.324 +ATEN 2.235 0.196 INSR rs4804428 black 0.356 ATEN 0.392 0.714 +ATEN 0.254 0.825 INSR rs17254521 black 0.319 ATEN 1.696 0.144 +ATEN 1.544 0.223 INSR rs3852876 black 0.483 ATEN 1.764 0.097 +ATEN 3.438 0.003 INSR rs10424224 black 0.487 ATEN 0.206 0.852 +ATEN 0.697 0.5 45 INSR rs4804433 black 0.374 ATEN 1.967 0.069 +ATEN 0.185 0.871 INSR rs1346490 black 0.401 ATEN 2.024 0.047 +ATEN 2.333 0.055 INSR rs10404318 black 0.161 ATEN 0.768 0.613 +ATEN 0.553 0.717 INSR rs6417197 black 0.000 ATEN 0.000 1.000 +ATEN 0.000 1.000 INSR rs11671297 black 0.441 ATEN 0.129 0.903 +ATEN 0.138 0.912 INSR rs10415205 black 0.312 ATEN 1.379 0.225 +ATEN 2.305 0.068

PAGE 173

173 Table C 7 Continued. Gene SNP Race MAF Tx PE P value Tx PE P value INSR rs11668751 black 0.372 ATEN 1.047 0.364 +ATEN 1.744 0.165 INSR rs4247374 black 0.023 ATEN 5.719 0.188 +ATEN 5.148 0.135 INSR rs10402346 black 0.471 ATEN 0.622 0.584 +ATEN 0.187 0.878 INSR rs10417205 black 0.326 ATEN 0.260 0.814 +ATEN 0.733 0.586 INSR rs4804195 black 0.253 ATEN 0.210 0.870 +AT EN 0.104 0.938 INSR rs919275 black 0.270 ATEN 1.679 0.148 +ATEN 2.268 0.111 INSR rs7258382 black 0.443 ATEN 0.648 0.530 +ATEN 0.220 0.854 INSR rs4804219 black 0.154 ATEN 0.238 0.875 +ATEN 0.668 0.678 INSR rs6510976 black 0.357 ATEN 0.957 0.382 +A TEN 0.463 0.733 INSR rs3745544 black 0.000 ATEN 1.000 +ATEN 1.000 INSR rs10408111 black 0.092 ATEN 0.011 0.995 +ATEN 0.623 0.733 INSR rs12979424 black 0.062 ATEN 3.471 0.085 +ATEN 5.453 0.064 INSR rs10411676 black 0.365 ATEN 0.279 0.796 +ATEN 0.3 16 0.797 INSR rs7254921 black 0.354 ATEN 0.122 0.909 +ATEN 1.252 0.290 INSR rs10404166 black 0.253 ATEN 0.514 0.689 +ATEN 1.585 0.250 INSR rs7254060 black 0.478 ATEN 0.690 0.527 +ATEN 1.083 0.352 INSR rs2860184 black 0.133 ATEN 0.794 0.639 +ATEN 2. 499 0.110 INSR rs10410272 black 0.030 ATEN 3.317 0.370 +ATEN 3.371 0.278 INSR rs8111710 black 0.168 ATEN 0.964 0.560 +ATEN 0.383 0.809 INSR rs8101064 black 0.191 ATEN 0.499 0.723 +ATEN 1.843 0.153 INSR rs7507911 black 0.409 ATEN 0.401 0.709 +ATEN 0 .432 0.712 IRS1 rs16822570 black 0.054 ATEN 2.135 0.404 +ATEN 2.642 0.361 IRS1 rs16822573 black 0.054 ATEN 2.135 0.404 +ATEN 2.642 0.361 IRS1 rs16822574 black 0.086 ATEN 1.382 0.460 +ATEN 0.312 0.897 IRS1 rs16822579 black 0.035 ATEN 0.038 0.990 +AT EN 0.682 0.834 IRS1 rs17208470 black 0.013 ATEN 0.496 0.924 +ATEN 3.928 0.398 IRS1 rs10182336 black 0.359 ATEN 0.084 0.938 +ATEN 1.668 0.181 IRS1 rs10181778 black 0.128 ATEN 0.582 0.729 +ATEN 0.817 0.644 IRS1 rs16822601 black 0.262 ATEN 0.603 0.612 +ATEN 1.140 0.422 IRS1 rs16822604 black 0.273 ATEN 0.481 0.677 +ATEN 1.397 0.306 IRS1 rs7567312 black 0.027 ATEN 0.515 0.828 +ATEN 3.971 0.431 IRS1 rs1078533 black 0.223 ATEN 0.734 0.544 +ATEN 2.064 0.138 IRS1 rs2435185 black 0.003 ATEN 2.963 0.744 +ATEN 6.532 0.343 IRS1 rs16822626 black 0.091 ATEN 0.801 0.692 +ATEN 1.268 0.545 IRS1 rs16822630 black 0.284 ATEN 0.401 0.714 +ATEN 1.296 0.320 IRS1 rs16822638 black 0.153 ATEN 0.332 0.807 +ATEN 1.088 0.508 IRS1 rs10498212 black 0.140 ATEN 0.634 0 .645 +ATEN 0.967 0.582 IRS1 rs4675094 black 0.245 ATEN 0.486 0.687 +ATEN 2.866 0.028 IRS1 rs10170579 black 0.030 ATEN 1.740 0.638 +ATEN 0.546 0.887

PAGE 174

174 Table C 7 Continued. Gene SNP Race MAF Tx PE P value Tx PE P value IRS1 rs2288586 black 0.144 ATEN 1.224 0.382 +ATEN 0.786 0.645 IRS1 rs3769647 black 0.017 ATEN 4.938 0.177 +ATEN 2.887 0.565 IRS1 rs1801278 black 0.059 ATEN 3.074 0.232 +ATEN 1.063 0.642 IRS1 rs2229613 black 0.102 ATEN 0.185 0.919 +ATEN 1.344 0.481 IRS1 rs1801123 black 0.470 ATEN 0 .093 0.932 +ATEN 0.639 0.581 IRS1 rs2234931 black 0.060 ATEN 3.074 0.232 +ATEN 0.551 0.806 IRS1 rs6725330 black 0.223 ATEN 0.084 0.945 +ATEN 0.629 0.637 IRS1 rs6725556 black 0.233 ATEN 0.473 0.700 +ATEN 1.428 0.301 IRS1 rs13423855 black 0.076 ATEN 3.064 0.191 +ATEN 1.521 0.512 IRS1 rs13018009 black 0.002 ATEN 0.559 0.949 +ATEN 0.000 1.000 IRS1 rs4675096 black 0.480 ATEN 0.817 0.471 +ATEN 0.148 0.901 IRS2 rs1044364 black 0.174 ATEN 0.792 0.551 +ATEN 0.218 0.892 IRS2 rs1865434 black 0.185 ATEN 1.437 0.303 +ATEN 1.228 0.383 IRS2 rs7982446 black 0.294 ATEN 1.017 0.357 +ATEN 1.662 0.202 IRS2 rs913949 black 0.308 ATEN 0.588 0.647 +ATEN 0.143 0.905 IRS2 rs9587980 black 0.133 ATEN 1.711 0.243 +ATEN 0.398 0.847 IRS2 rs12583454 black 0.002 ATE N 0.000 1.000 +ATEN 4.236 0.671 IRS2 rs9559646 black 0.220 ATEN 1.028 0.465 +ATEN 0.207 0.871 IRS2 rs4773087 black 0.218 ATEN 0.233 0.857 +ATEN 2.369 0.087 IRS2 rs4773088 black 0.297 ATEN 1.035 0.432 +ATEN 0.166 0.893 IRS2 rs9515120 black 0.010 AT EN 4.459 0.317 +ATEN 3.613 0.612 IRS2 rs7323191 black 0.334 ATEN 0.227 0.844 +ATEN 1.742 0.173 IRS2 rs7999797 black 0.166 ATEN 0.024 0.987 +ATEN 4.104 0.007 IRS2 rs9521510 black 0.052 ATEN 0.870 0.778 +ATEN 3.082 0.178 IRS2 rs4771646 black 0.317 ATE N 1.155 0.348 +ATEN 0.407 0.743 IRS2 rs9559656 black 0.277 ATEN 1.144 0.342 +ATEN 0.680 0.584 IRS2 rs9583423 black 0.116 ATEN 1.459 0.379 +ATEN 0.748 0.711 IRS2 rs7997595 black 0.092 ATEN 0.242 0.908 +ATEN 3.670 0.056 IRS2 rs7981705 black 0.393 ATEN 0.084 0.939 +ATEN 1.674 0.168 IRS2 rs4773092 black 0.388 ATEN 0.002 0.998 +ATEN 0.119 0.924 IRS2 rs4773094 black 0.208 ATEN 0.273 0.831 +ATEN 1.069 0.432 SLC2A2 rs7610064 black 0.045 ATEN 6.441 0.011 +ATEN 0.151 0.957 SLC2A2 rs5397 black 0.000 ATEN 0.000 1.000 +ATEN 0.000 1.000 SLC2A2 rs16855638 black 0.112 ATEN 1.304 0.334 +ATEN 2.649 0.137 SLC2A2 rs10513684 black 0.034 ATEN 1.067 0.705 +ATEN 4.422 0.147 SLC2A2 rs10513685 black 0.381 ATEN 1.495 0.159 +ATEN 0.482 0.655 SLC2A2 rs11711437 b lack 0.490 ATEN 1.587 0.112 +ATEN 0.456 0.664 SLC2A2 rs5400 black 0.460 ATEN 1.615 0.103 +ATEN 0.274 0.794 SLC2A2 rs11924032 black 0.393 ATEN 1.332 0.240 +ATEN 0.048 0.967

PAGE 175

175 Table C 7 Continued. Gene SNP Race MAF Tx PE P value Tx PE P value SLC2A2 rs11925298 black 0.111 ATEN 1.706 0.355 +ATEN 0.764 0.670 SLC2A2 rs9828378 black 0.094 ATEN 1.016 0.559 +ATEN 2.835 0.135 SLC2A2 rs12488694 black 0.064 ATEN 2.232 0.350 +ATEN 0.120 0.956 SLC2A2 rs5396 black 0.357 ATEN 0.910 0.429 +ATEN 0.968 0.398 SLC2A4 rs8076649 black 0.146 ATEN 0.756 0.613 +ATEN 0.054 0.973 SLC2A4 rs2654185 black 0.367 ATEN 0.285 0.803 +ATEN 0.580 0.634 SLC2A4 rs9902838 black 0.366 ATEN 0.743 0.479 +ATEN 1.319 0.295 SLC2A4 rs2073476 black 0.002 ATEN 0.000 1.000 +ATEN 0.00 0 1.000 SLC2A4 rs5411 black 0.106 ATEN 1.813 0.263 +ATEN 1.466 0.408 SLC2A4 rs5412 black 0.025 ATEN 1.338 0.690 +ATEN 0.311 0.937 SLC2A4 rs5415 black 0.076 ATEN 0.069 0.975 +ATEN 0.593 0.801 SLC2A4 rs222847 black 0.012 ATEN 2.049 0.690 +ATEN 1.569 0.757 SLC2A4 rs16956647 black 0.045 ATEN 2.112 0.445 +ATEN 4.920 0.076 SLC2A4 rs5435 black 0.122 ATEN 0.868 0.627 +ATEN 2.344 0.199 SLC2A4 rs5436 black 0.064 ATEN 1.606 0.406 +ATEN 0.344 0.894 FOXO1 rs7325210 non black 0.001 ATEN 9.655 0.228 +ATEN 0.000 1.000 FOXO1 rs2701858 non black 0.048 ATEN 0.706 0.671 +ATEN 0.866 0.688 FOXO1 rs17446614 non black 0.145 ATEN 0.763 0.491 +ATEN 0.135 0.917 FOXO1 rs2701859 non black 0.244 ATEN 0.419 0.655 +ATEN 1.011 0.342 FOXO1 rs2755211 non black 0.244 ATEN 0.419 0.655 +ATEN 1.011 0.342 FOXO1 rs2995991 non black 0.426 ATEN 0.882 0.238 +ATEN 0.629 0.496 FOXO1 rs2984121 non black 0.181 ATEN 0.823 0.388 +ATEN 1.417 0.221 FOXO1 rs3858869 non black 0.000 ATEN 0.000 1.000 +ATEN 0.000 1.000 FOXO1 rs4943795 n on black 0.298 ATEN 0.988 0.255 +ATEN 1.048 0.307 FOXO1 rs3908774 non black 0.000 ATEN 0.000 1.000 +ATEN 0.000 1.000 FOXO1 rs12876443 non black 0.097 ATEN 0.316 0.803 +ATEN 3.998 0.015 FOXO1 rs7981045 non black 0.251 ATEN 0.304 0.726 +ATEN 1.509 0.181 FOXO1 rs7139990 non black 0.051 ATEN 0.288 0.867 +ATEN 0.165 0.937 FOXO1 rs9603776 non black 0.026 ATEN 0.150 0.952 +ATEN 0.665 0.791 GYS1 rs2387583 non black 0.137 ATEN 2.773 0.009 +ATEN 1.982 0.155 GYS1 rs4645894 non black 0.010 ATEN 4.197 0.458 +ATEN 1.093 0.782 GYS1 rs4645900 non black 0.039 ATEN 1.413 0.463 +ATEN 3.784 0.136 GYS1 rs905238 non black 0.485 ATEN 0.433 0.567 +ATEN 1.053 0.268 GYS1 rs4645903 non black 0.103 ATEN 1.692 0.149 +ATEN 2.129 0.196 GYS1 rs4645905 non black 0.477 ATE N 0.434 0.568 +ATEN 1.533 0.109 GYS1 rs4645908 non black 0.001 ATEN 0.000 1.000 +ATEN 1.706 0.867 GYS1 rs1042265 non black 0.113 ATEN 1.243 0.278 +ATEN 0.871 0.579 GYS1 rs8104760 non black 0.039 ATEN 2.672 0.210 +ATEN 0.910 0.684 GYS1 rs13306418 non black 0.002 ATEN 0.800 0.887 +ATEN 0.000 1.000

PAGE 176

176 Table C 7 Continued. Gene SNP Race MAF Tx PE P value Tx PE P value GYS1 rs12980441 non black 0.383 ATEN 0.766 0.330 +ATEN 1.665 0.068 GYS1 rs5464 non black 0.299 ATEN 0.554 0.525 +ATEN 1.509 0.125 G YS1 rs5460 non black 0.000 ATEN 0.000 1.000 +ATEN 0.000 1.000 GYS1 rs2287755 non black 0.083 ATEN 0.328 0.809 +ATEN 0.145 0.935 GYS1 rs12610125 non black 0.052 ATEN 0.677 0.693 +ATEN 1.687 0.429 INS rs4320932 non black 0.226 ATEN 0.614 0.516 +ATEN 0.579 0.604 INS rs7924316 non black 0.479 ATEN 0.595 0.457 +ATEN 0.646 0.499 INS rs3842768 non black 0.000 ATEN 0.000 1.000 +ATEN 0.000 1.000 INS rs689 non black 0.278 ATEN 0.326 0.704 +ATEN 0.461 0.668 INS rs2070762 non black 0.491 ATEN 0.626 0.451 + ATEN 1.910 0.046 INS rs6356 non black 0.389 ATEN 0.085 0.913 +ATEN 0.012 0.991 INS rs7950050 non black 0.000 ATEN 0.000 1.000 +ATEN 0.000 1.000 INSR rs12150997 non black 0.181 ATEN 0.969 0.309 +ATEN 1.158 0.386 INSR rs3745551 non black 0.383 ATEN 0. 304 0.684 +ATEN 0.107 0.917 INSR rs3745550 non black 0.194 ATEN 0.658 0.460 +ATEN 1.638 0.174 INSR rs6510947 non black 0.102 ATEN 0.982 0.438 +ATEN 1.527 0.311 INSR rs10408374 non black 0.136 ATEN 1.347 0.212 +ATEN 2.815 0.065 INSR rs2860172 non bla ck 0.181 ATEN 0.785 0.415 +ATEN 1.071 0.399 INSR rs13306446 non black 0.115 ATEN 0.004 0.997 +ATEN 0.321 0.837 INSR rs2860173 non black 0.076 ATEN 0.245 0.861 +ATEN 2.273 0.181 INSR rs2860175 non black 0.134 ATEN 1.327 0.224 +ATEN 2.628 0.086 INSR rs1549616 non black 0.079 ATEN 0.319 0.818 +ATEN 2.594 0.122 INSR rs10420382 non black 0.090 ATEN 0.484 0.719 +ATEN 0.888 0.632 INSR rs11672739 non black 0.108 ATEN 0.789 0.523 +ATEN 0.788 0.601 INSR rs6510949 non black 0.062 ATEN 1.439 0.379 +ATEN 0.854 0.673 INSR rs12610022 non black 0.063 ATEN 0.132 0.933 +ATEN 1.559 0.475 INSR rs6510950 non black 0.072 ATEN 0.140 0.925 +ATEN 1.476 0.475 INSR rs11667110 non black 0.297 ATEN 0.519 0.533 +ATEN 0.886 0.410 INSR rs16990074 non black 0.006 ATEN 4.503 0.348 +ATEN 2.440 0.717 INSR rs16994200 non black 0.006 ATEN 4.503 0.348 +ATEN 2.440 0.717 INSR rs8103483 non black 0.479 ATEN 0.430 0.598 +ATEN 1.238 0.215 INSR rs16994220 non black 0.002 ATEN 3.475 0.530 +ATEN 0.000 1.000 INSR rs16994221 non black 0.049 ATEN 0.043 0.982 +ATEN 2.605 0.247 INSR rs6510952 non black 0.026 ATEN 1.262 0.625 +ATEN 0.290 0.918 INSR rs8102612 non black 0.001 ATEN 0.000 1.000 +ATEN 15.646 0.105 INSR rs8100109 non black 0.012 ATEN 3.681 0.301 +ATEN 2.876 0.478 I NSR rs2059807 non black 0.372 ATEN 0.318 0.692 +ATEN 0.011 0.991 INSR rs3815901 non black 0.469 ATEN 0.943 0.203 +ATEN 0.340 0.717 INSR rs2860177 non black 0.260 ATEN 0.452 0.621 +ATEN 0.582 0.578

PAGE 177

177 Table C 7 Continued. Gene SNP Race MAF Tx PE P va lue Tx PE P value INSR rs6510955 non black 0.262 ATEN 0.400 0.661 +ATEN 0.541 0.605 INSR rs7252268 non black 0.204 ATEN 0.943 0.284 +ATEN 0.902 0.452 INSR rs10415589 non black 0.001 ATEN 4.296 0.620 +ATEN 0.000 1.000 INSR rs8109559 non black 0.202 AT EN 0.795 0.367 +ATEN 0.621 0.606 INSR rs10411667 non black 0.204 ATEN 1.002 0.263 +ATEN 1.057 0.378 INSR rs8112883 non black 0.296 ATEN 0.341 0.684 +ATEN 1.411 0.167 INSR rs8105406 non black 0.000 ATEN 0.000 1.000 +ATEN 0.000 1.000 INSR rs16994316 no n black 0.002 ATEN 12.314 0.122 +ATEN 17.560 0.062 INSR rs8108622 non black 0.234 ATEN 0.425 0.644 +ATEN 0.800 0.451 INSR rs10500204 non black 0.290 ATEN 0.817 0.343 +ATEN 0.969 0.330 INSR rs6510959 non black 0.174 ATEN 0.282 0.771 +ATEN 1.035 0.39 5 INSR rs891087 non black 0.113 ATEN 0.858 0.441 +ATEN 1.892 0.183 INSR rs891088 non black 0.291 ATEN 0.528 0.503 +ATEN 0.166 0.873 INSR rs4804366 non black 0.113 ATEN 0.693 0.529 +ATEN 1.881 0.187 INSR rs10416539 non black 0.115 ATEN 0.740 0.499 +A TEN 1.926 0.179 INSR rs1035939 non black 0.281 ATEN 0.408 0.627 +ATEN 0.533 0.613 INSR rs2860183 non black 0.396 ATEN 0.095 0.900 +ATEN 1.042 0.274 INSR rs12460089 non black 0.390 ATEN 0.034 0.964 +ATEN 0.933 0.328 INSR rs4804377 non black 0.283 ATEN 0.428 0.610 +ATEN 0.256 0.808 INSR rs2115386 non black 0.490 ATEN 0.786 0.274 +ATEN 0.193 0.830 INSR rs4499341 non black 0.389 ATEN 0.150 0.840 +ATEN 0.957 0.308 INSR rs6510960 non black 0.125 ATEN 0.559 0.604 +ATEN 0.629 0.644 INSR rs1896639 non black 0.446 ATEN 0.821 0.208 +ATEN 1.036 0.222 INSR rs2042902 non black 0.484 ATEN 0.786 0.276 +ATEN 0.628 0.490 INSR rs10426094 non black 0.224 ATEN 0.747 0.385 +ATEN 0.208 0.850 INSR rs12609995 non black 0.261 ATEN 0.690 0.420 +ATEN 1.211 0.265 INSR rs12459488 non black 0.284 ATEN 0.857 0.301 +ATEN 0.439 0.661 INSR rs3745545 non black 0.173 ATEN 1.000 0.289 +ATEN 1.119 0.367 INSR rs8104314 non black 0.001 ATEN 9.242 0.247 +ATEN 0.000 1.000 INSR rs12971499 non black 0.424 ATEN 1.224 0.097 +A TEN 0.115 0.903 INSR rs7245562 non black 0.076 ATEN 0.023 0.987 +ATEN 1.492 0.385 INSR rs4804404 non black 0.152 ATEN 0.866 0.396 +ATEN 1.435 0.300 INSR rs7508679 non black 0.413 ATEN 1.259 0.116 +ATEN 0.402 0.683 INSR rs890859 non black 0.001 ATEN 9.242 0.247 +ATEN 0.000 1.000 INSR rs4804414 non black 0.409 ATEN 0.232 0.767 +ATEN 0.862 0.362 INSR rs4804418 non black 0.227 ATEN 0.180 0.849 +ATEN 1.999 0.077 INSR rs890860 non black 0.198 ATEN 1.052 0.263 +ATEN 0.242 0.834 INSR rs890861 non black 0.087 ATEN 0.111 0.930 +ATEN 0.375 0.841 INSR rs7255710 non black 0.011 ATEN 5.327 0.145 +ATEN 1.145 0.810

PAGE 178

178 Table C 7 Continued. Gene SNP Race MAF Tx PE P value Tx PE P value INSR rs12460755 non black 0.296 ATEN 1.068 0.170 +ATEN 0.496 0.619 INS R rs17175790 non black 0.079 ATEN 0.312 0.819 +ATEN 1.376 0.475 INSR rs10420008 non black 0.229 ATEN 0.066 0.941 +ATEN 1.049 0.372 INSR rs17175860 non black 0.197 ATEN 0.356 0.711 +ATEN 1.878 0.100 INSR rs4804428 non black 0.188 ATEN 1.286 0.156 +A TEN 0.049 0.968 INSR rs17254521 non black 0.488 ATEN 0.882 0.244 +ATEN 0.930 0.327 INSR rs3852876 non black 0.308 ATEN 0.067 0.932 +ATEN 1.160 0.231 INSR rs10424224 non black 0.390 ATEN 1.142 0.148 +ATEN 1.575 0.109 INSR rs4804433 non black 0.222 AT EN 0.237 0.797 +ATEN 0.370 0.761 INSR rs1346490 non black 0.396 ATEN 0.606 0.400 +ATEN 1.169 0.230 INSR rs10404318 non black 0.013 ATEN 2.803 0.437 +ATEN 9.704 0.056 INSR rs6417197 non black 0.001 ATEN 6.497 0.422 +ATEN 0.000 1.000 INSR rs11671297 non black 0.440 ATEN 2.391 0.003 +ATEN 2.106 0.029 INSR rs10415205 non black 0.282 ATEN 0.309 0.724 +ATEN 1.161 0.254 INSR rs11668751 non black 0.290 ATEN 0.261 0.764 +ATEN 1.313 0.200 INSR rs4247374 non black 0.134 ATEN 2.477 0.018 +ATEN 0.203 0.890 INSR rs10402346 non black 0.291 ATEN 0.294 0.735 +ATEN 1.313 0.200 INSR rs10417205 non black 0.328 ATEN 0.760 0.356 +ATEN 3.648 0.000 INSR rs4804195 non black 0.431 ATEN 1.292 0.100 +ATEN 1.842 0.055 INSR rs919275 non black 0.396 ATEN 0.050 0.947 +A TEN 0.882 0.392 INSR rs7258382 non black 0.157 ATEN 0.114 0.916 +ATEN 1.998 0.131 INSR rs4804219 non black 0.045 ATEN 2.478 0.246 +ATEN 3.903 0.061 INSR rs6510976 non black 0.471 ATEN 0.324 0.655 +ATEN 0.076 0.936 INSR rs3745544 non black 0.000 ATE N 1.000 +ATEN 1.000 INSR rs10408111 non black 0.000 ATEN 0.000 1.000 +ATEN 0.000 1.000 INSR rs12979424 non black 0.329 ATEN 0.339 0.686 +ATEN 1.575 0.124 INSR rs10411676 non black 0.418 ATEN 0.183 0.813 +ATEN 1.166 0.228 INSR rs7254921 non black 0. 490 ATEN 0.407 0.599 +ATEN 1.082 0.273 INSR rs10404166 non black 0.002 ATEN 9.204 0.251 +ATEN 0.000 1.000 INSR rs7254060 non black 0.076 ATEN 2.068 0.166 +ATEN 1.563 0.368 INSR rs2860184 non black 0.366 ATEN 0.164 0.841 +ATEN 1.546 0.125 INSR rs1041 0272 non black 0.000 ATEN 0.000 1.000 +ATEN 0.000 1.000 INSR rs8111710 non black 0.252 ATEN 0.362 0.683 +ATEN 0.863 0.459 INSR rs8101064 non black 0.053 ATEN 1.046 0.581 +ATEN 0.839 0.642 INSR rs7507911 non black 0.165 ATEN 1.898 0.074 +ATEN 2.180 0.0 85 IRS1 rs16822570 non black 0.060 ATEN 0.300 0.845 +ATEN 2.411 0.228 IRS1 rs16822573 non black 0.058 ATEN 0.787 0.629 +ATEN 2.411 0.228 IRS1 rs16822574 non black 0.058 ATEN 0.867 0.601 +ATEN 2.411 0.228 IRS1 rs16822579 non black 0.000 ATEN 0.000 1. 000 +ATEN 0.000 1.000

PAGE 179

179 Table C 7 Continued. Gene SNP Race MAF Tx PE P value Tx PE P value IRS1 rs17208470 non black 0.107 ATEN 2.275 0.069 +ATEN 0.140 0.928 IRS1 rs10182336 non black 0.084 ATEN 0.161 0.904 +ATEN 2.573 0.141 IRS1 rs10181778 non blac k 0.058 ATEN 0.126 0.936 +ATEN 2.903 0.164 IRS1 rs16822601 non black 0.026 ATEN 0.732 0.789 +ATEN 2.725 0.361 IRS1 rs16822604 non black 0.026 ATEN 0.057 0.984 +ATEN 3.125 0.279 IRS1 rs7567312 non black 0.000 ATEN 0.000 1.000 +ATEN 0.000 1.000 IRS1 rs 1078533 non black 0.024 ATEN 0.907 0.757 +ATEN 3.125 0.279 IRS1 rs2435185 non black 0.021 ATEN 0.258 0.917 +ATEN 2.753 0.481 IRS1 rs16822626 non black 0.000 ATEN 0.000 1.000 +ATEN 0.000 1.000 IRS1 rs16822630 non black 0.028 ATEN 0.722 0.765 +ATEN 2. 487 0.369 IRS1 rs16822638 non black 0.083 ATEN 0.024 0.986 +ATEN 3.596 0.040 IRS1 rs10498212 non black 0.081 ATEN 0.165 0.904 +ATEN 3.596 0.040 IRS1 rs4675094 non black 0.098 ATEN 0.130 0.920 +ATEN 3.008 0.048 IRS1 rs10170579 non black 0.000 ATEN 0.0 00 1.000 +ATEN 0.000 1.000 IRS1 rs2288586 non black 0.081 ATEN 0.160 0.907 +ATEN 3.596 0.040 IRS1 rs3769647 non black 0.073 ATEN 0.157 0.913 +ATEN 3.175 0.080 IRS1 rs1801278 non black 0.050 ATEN 5.524 0.001 +ATEN 2.263 0.346 IRS1 rs2229613 non black 0.001 ATEN 0.000 1.000 +ATEN 4.494 0.635 IRS1 rs1801123 non black 0.101 ATEN 0.036 0.978 +ATEN 2.845 0.060 IRS1 rs2234931 non black 0.051 ATEN 5.052 0.002 +ATEN 2.263 0.346 IRS1 rs6725330 non black 0.115 ATEN 0.627 0.617 +ATEN 1.831 0.167 IRS1 rs6725 556 non black 0.062 ATEN 1.662 0.364 +ATEN 1.438 0.405 IRS1 rs13423855 non black 0.000 ATEN 0.000 1.000 +ATEN 0.000 1.000 IRS1 rs13018009 non black 0.024 ATEN 3.879 0.176 +ATEN 1.327 0.637 IRS1 rs4675096 non black 0.080 ATEN 0.056 0.968 +ATEN 2.302 0 .217 IRS2 rs1044364 non black 0.052 ATEN 1.501 0.411 +ATEN 5.962 0.004 IRS2 rs1865434 non black 0.153 ATEN 0.741 0.464 +ATEN 1.076 0.415 IRS2 rs7982446 non black 0.052 ATEN 1.201 0.522 +ATEN 5.956 0.004 IRS2 rs913949 non black 0.193 ATEN 0.719 0.45 9 +ATEN 2.499 0.033 IRS2 rs9587980 non black 0.001 ATEN 0.602 0.944 +ATEN 0.000 1.000 IRS2 rs12583454 non black 0.017 ATEN 2.198 0.474 +ATEN 6.689 0.065 IRS2 rs9559646 non black 0.498 ATEN 0.486 0.498 +ATEN 0.876 0.368 IRS2 rs4773087 non black 0.49 9 ATEN 0.001 0.999 +ATEN 0.936 0.349 IRS2 rs4773088 non black 0.308 ATEN 0.123 0.877 +ATEN 0.230 0.833 IRS2 rs9515120 non black 0.062 ATEN 1.533 0.341 +ATEN 2.584 0.201 IRS2 rs7323191 non black 0.142 ATEN 0.503 0.627 +ATEN 0.462 0.749 IRS2 rs7999797 non black 0.482 ATEN 0.279 0.701 +ATEN 0.715 0.485 IRS2 rs9521510 non black 0.356 ATEN 0.095 0.904 +ATEN 1.516 0.133 IRS2 rs4771646 non black 0.319 ATEN 0.130 0.869 +ATEN 0.760 0.472

PAGE 180

180 Table C 7 Continued. Gene SNP Race MAF Tx PE P value Tx PE P v alue IRS2 rs9559656 non black 0.283 ATEN 0.675 0.397 +ATEN 0.526 0.636 IRS2 rs9583423 non black 0.001 ATEN 0.602 0.944 +ATEN 0.000 1.000 IRS2 rs7997595 non black 0.167 ATEN 0.967 0.328 +ATEN 1.201 0.352 IRS2 rs7981705 non black 0.148 ATEN 0.435 0.6 79 +ATEN 0.273 0.847 IRS2 rs4773092 non black 0.383 ATEN 0.425 0.592 +ATEN 0.686 0.491 IRS2 rs4773094 non black 0.222 ATEN 0.913 0.291 +ATEN 0.564 0.656 SLC2A2 rs7610064 non black 0.001 ATEN 0.000 1.000 +ATEN 3.327 0.620 SLC2A2 rs5397 non black 0.00 3 ATEN 0.000 1.000 +ATEN 2.598 0.698 SLC2A2 rs16855638 non black 0.002 ATEN 5.048 0.395 +ATEN 0.000 1.000 SLC2A2 rs10513684 non black 0.046 ATEN 0.546 0.756 +ATEN 1.804 0.469 SLC2A2 rs10513685 non black 0.135 ATEN 1.522 0.152 +ATEN 0.628 0.653 SLC 2A2 rs11711437 non black 0.140 ATEN 1.811 0.084 +ATEN 0.438 0.746 SLC2A2 rs5400 non black 0.140 ATEN 1.836 0.077 +ATEN 0.438 0.746 SLC2A2 rs11924032 non black 0.264 ATEN 1.222 0.138 +ATEN 1.339 0.253 SLC2A2 rs11925298 non black 0.020 ATEN 4.360 0. 130 +ATEN 2.680 0.381 SLC2A2 rs9828378 non black 0.155 ATEN 0.997 0.315 +ATEN 2.714 0.052 SLC2A2 rs12488694 non black 0.020 ATEN 4.360 0.130 +ATEN 2.680 0.381 SLC2A2 rs5396 non black 0.288 ATEN 1.570 0.050 +ATEN 0.954 0.377 SLC2A4 rs8076649 non bl ack 0.001 ATEN 3.673 0.645 +ATEN 0.000 1.000 SLC2A4 rs2654185 non black 0.381 ATEN 0.236 0.765 +ATEN 1.324 0.191 SLC2A4 rs9902838 non black 0.013 ATEN 4.030 0.153 +ATEN 1.681 0.802 SLC2A4 rs2073476 non black 0.002 ATEN 5.092 0.457 +ATEN 0.000 1.000 SL C2A4 rs5411 non black 0.001 ATEN 3.673 0.645 +ATEN 0.000 1.000 SLC2A4 rs5412 non black 0.173 ATEN 0.554 0.584 +ATEN 0.794 0.543 SLC2A4 rs5415 non black 0.309 ATEN 0.254 0.765 +ATEN 1.322 0.178 SLC2A4 rs222847 non black 0.042 ATEN 0.478 0.824 +ATEN 1. 748 0.424 SLC2A4 rs16956647 non black 0.011 ATEN 4.080 0.202 +ATEN 1.681 0.802 SLC2A4 rs5435 non black 0.375 ATEN 0.264 0.728 +ATEN 1.268 0.216 SLC2A4 rs5436 non black 0.000 ATEN 0.000 1.000 +ATEN 0.000 1.000 FOXO1 rs7325210 black 0.045 HCTZ 1.020 0 .760 +HCTZ 0.912 0.707 FOXO1 rs2701858 black 0.069 HCTZ 0.698 0.753 +HCTZ 2.083 0.381 FOXO1 rs17446614 black 0.161 HCTZ 2.793 0.092 +HCTZ 2.248 0.116 FOXO1 rs2701859 black 0.435 HCTZ 1.484 0.184 +HCTZ 1.403 0.212 FOXO1 rs2755211 black 0.435 HCTZ 1. 484 0.184 +HCTZ 1.403 0.212 FOXO1 rs2995991 black 0.069 HCTZ 3.546 0.154 +HCTZ 2.948 0.209 FOXO1 rs2984121 black 0.029 HCTZ 0.048 0.990 +HCTZ 1.643 0.614 FOXO1 rs3858869 black 0.060 HCTZ 2.182 0.394 +HCTZ 1.501 0.572 FOXO1 rs4943795 black 0.436 HCTZ 0.115 0.919 +HCTZ 0.332 0.772 FOXO1 rs3908774 black 0.057 HCTZ 0.620 0.801 +HCTZ 1.907 0.538

PAGE 181

181 Table C 7 Continued. Gene SNP Race MAF Tx PE P value Tx PE P value FOXO1 rs12876443 black 0.005 HCTZ 9.233 0.125 +HCTZ 0.000 1.000 FOXO1 rs7981045 black 0.096 HCTZ 5.420 0.010 +HCTZ 3.405 0.086 FOXO1 rs7139990 black 0.154 HCTZ 0.416 0.807 +HCTZ 1.068 0.515 FOXO1 rs9603776 black 0.005 HCTZ 0.841 0.933 +HCTZ 4.577 0.496 GYS1 rs2387583 black 0.258 HCTZ 0.670 0.626 +HCTZ 0.727 0.584 GYS1 rs4645894 bla ck 0.104 HCTZ 3.272 0.083 +HCTZ 1.304 0.483 GYS1 rs4645900 black 0.000 HCTZ 0.000 1.000 +HCTZ 0.000 1.000 GYS1 rs905238 black 0.384 HCTZ 1.963 0.135 +HCTZ 1.610 0.197 GYS1 rs4645903 black 0.154 HCTZ 1.738 0.299 +HCTZ 1.715 0.282 GYS1 rs4645905 black 0.306 HCTZ 0.636 0.623 +HCTZ 0.943 0.438 GYS1 rs4645908 black 0.000 HCTZ 0.000 1.000 +HCTZ 0.000 1.000 GYS1 rs1042265 black 0.086 HCTZ 0.262 0.906 +HCTZ 1.440 0.461 GYS1 rs8104760 black 0.005 HCTZ 3.366 0.641 +HCTZ 2.858 0.761 GYS1 rs13306418 black 0.015 HCTZ 4.484 0.662 +HCTZ 3.017 0.379 GYS1 rs12980441 black 0.305 HCTZ 0.633 0.629 +HCTZ 0.207 0.861 GYS1 rs5464 black 0.289 HCTZ 1.172 0.349 +HCTZ 0.239 0.850 GYS1 rs5460 black 0.062 HCTZ 1.440 0.529 +HCTZ 3.093 0.240 GYS1 rs2287755 black 0.111 H CTZ 1.134 0.521 +HCTZ 1.543 0.383 GYS1 rs12610125 black 0.032 HCTZ 2.354 0.461 +HCTZ 0.779 0.855 INS rs4320932 black 0.112 HCTZ 1.094 0.571 +HCTZ 0.324 0.859 INS rs7924316 black 0.357 HCTZ 0.221 0.859 +HCTZ 0.247 0.850 INS rs3842768 black 0.023 HCT Z 3.380 0.334 +HCTZ 4.062 0.356 INS rs689 black 0.247 HCTZ 1.269 0.360 +HCTZ 2.127 0.105 INS rs2070762 black 0.208 HCTZ 1.317 0.409 +HCTZ 2.469 0.078 INS rs6356 black 0.128 HCTZ 2.243 0.234 +HCTZ 0.134 0.936 INS rs7950050 black 0.128 HCTZ 0.184 0.924 +HCTZ 1.573 0.276 INSR rs12150997 black 0.379 HCTZ 0.998 0.430 +HCTZ 0.521 0.652 INSR rs3745551 black 0.229 HCTZ 0.338 0.800 +HCTZ 0.257 0.849 INSR rs3745550 black 0.466 HCTZ 0.052 0.963 +HCTZ 0.504 0.679 INSR rs6510947 black 0.064 HCTZ 1.799 0.497 +HCTZ 0.261 0.902 INSR rs10408374 black 0.114 HCTZ 0.224 0.904 +HCTZ 0.511 0.778 INSR rs2860172 black 0.230 HCTZ 0.758 0.583 +HCTZ 0.460 0.736 INSR rs13306446 black 0.046 HCTZ 3.107 0.253 +HCTZ 2.850 0.295 INSR rs2860173 black 0.023 HCTZ 0.95 3 0.826 +HCTZ 3.850 0.262 INSR rs2860175 black 0.262 HCTZ 0.573 0.678 +HCTZ 1.893 0.155 INSR rs1549616 black 0.050 HCTZ 1.044 0.746 +HCTZ 3.105 0.203 INSR rs10420382 black 0.302 HCTZ 0.628 0.613 +HCTZ 2.345 0.054 INSR rs11672739 black 0.174 HCTZ 3.3 34 0.040 +HCTZ 0.332 0.827 INSR rs6510949 black 0.034 HCTZ 1.293 0.715 +HCTZ 5.956 0.041

PAGE 182

182 Table C 7 Continued. Gene SNP Race MAF Tx PE P value Tx PE P value INSR rs12610022 black 0.166 HCTZ 1.371 0.391 +HCTZ 5.090 0.002 INSR rs6510950 black 0.344 HC TZ 2.249 0.066 +HCTZ 4.165 0.002 INSR rs11667110 black 0.138 HCTZ 0.266 0.872 +HCTZ 0.335 0.844 INSR rs16990074 black 0.017 HCTZ 1.937 0.706 +HCTZ 4.650 0.280 INSR rs16994200 black 0.029 HCTZ 1.981 0.640 +HCTZ 1.155 0.709 INSR rs8103483 black 0.257 HCTZ 2.055 0.127 +HCTZ 2.985 0.026 INSR rs16994220 black 0.003 HCTZ 0.000 1.000 +HCTZ 3.448 0.604 INSR rs16994221 black 0.138 HCTZ 0.296 0.859 +HCTZ 1.564 0.361 INSR rs6510952 black 0.029 HCTZ 0.544 0.860 +HCTZ 4.294 0.312 INSR rs8102612 black 0.3 66 HCTZ 0.246 0.858 +HCTZ 1.388 0.307 INSR rs8100109 black 0.446 HCTZ 0.174 0.894 +HCTZ 0.269 0.820 INSR rs2059807 black 0.257 HCTZ 1.075 0.443 +HCTZ 0.161 0.904 INSR rs3815901 black 0.391 HCTZ 0.392 0.766 +HCTZ 1.780 0.124 INSR rs2860177 black 0.1 84 HCTZ 1.469 0.326 +HCTZ 1.094 0.461 INSR rs6510955 black 0.263 HCTZ 0.387 0.775 +HCTZ 1.091 0.420 INSR rs7252268 black 0.376 HCTZ 0.089 0.944 +HCTZ 1.650 0.197 INSR rs10415589 black 0.087 HCTZ 2.407 0.233 +HCTZ 0.608 0.807 INSR rs8109559 black 0. 466 HCTZ 0.513 0.658 +HCTZ 1.948 0.120 INSR rs10411667 black 0.386 HCTZ 0.606 0.633 +HCTZ 1.606 0.208 INSR rs8112883 black 0.182 HCTZ 1.431 0.353 +HCTZ 0.788 0.568 INSR rs8105406 black 0.029 HCTZ 2.729 0.464 +HCTZ 0.412 0.910 INSR rs16994316 black 0. 030 HCTZ 2.869 0.370 +HCTZ 2.708 0.491 INSR rs8108622 black 0.208 HCTZ 1.038 0.486 +HCTZ 0.639 0.666 INSR rs10500204 black 0.190 HCTZ 0.400 0.807 +HCTZ 0.439 0.768 INSR rs6510959 black 0.241 HCTZ 0.394 0.762 +HCTZ 0.418 0.769 INSR rs891087 black 0. 181 HCTZ 0.509 0.741 +HCTZ 0.021 0.989 INSR rs891088 black 0.423 HCTZ 0.726 0.525 +HCTZ 0.460 0.699 INSR rs4804366 black 0.298 HCTZ 1.393 0.327 +HCTZ 0.485 0.678 INSR rs10416539 black 0.260 HCTZ 0.913 0.526 +HCTZ 0.349 0.785 INSR rs1035939 black 0 .383 HCTZ 0.734 0.520 +HCTZ 0.974 0.403 INSR rs2860183 black 0.305 HCTZ 0.314 0.810 +HCTZ 1.170 0.376 INSR rs12460089 black 0.461 HCTZ 0.614 0.574 +HCTZ 0.437 0.698 INSR rs4804377 black 0.208 HCTZ 0.350 0.818 +HCTZ 0.733 0.573 INSR rs2115386 bla ck 0.428 HCTZ 2.156 0.063 +HCTZ 0.061 0.959 INSR rs4499341 black 0.371 HCTZ 0.780 0.521 +HCTZ 1.269 0.271 INSR rs6510960 black 0.428 HCTZ 0.160 0.892 +HCTZ 0.132 0.916 INSR rs1896639 black 0.328 HCTZ 1.065 0.261 +HCTZ 0.293 0.752 INSR rs2042902 black 0.218 HCTZ 1.034 0.493 +HCTZ 0.227 0.862 INSR rs10426094 black 0.072 HCTZ 0.661 0.782 +HCTZ 2.035 0.326

PAGE 183

183 Table C 7 Continued. Gene SNP Race MAF Tx PE P value Tx PE P value INSR rs12609995 black 0.144 HCTZ 0.230 0.895 +HCTZ 0.654 0.670 INSR rs124 59488 black 0.305 HCTZ 1.387 0.285 +HCTZ 0.193 0.876 INSR rs3745545 black 0.144 HCTZ 1.593 0.332 +HCTZ 0.507 0.766 INSR rs8104314 black 0.154 HCTZ 0.236 0.887 +HCTZ 0.528 0.747 INSR rs12971499 black 0.446 HCTZ 0.589 0.607 +HCTZ 2.593 0.027 INSR rs7 245562 black 0.175 HCTZ 3.152 0.053 +HCTZ 3.129 0.022 INSR rs4804404 black 0.205 HCTZ 2.853 0.067 +HCTZ 3.496 0.005 INSR rs7508679 black 0.339 HCTZ 0.594 0.641 +HCTZ 2.857 0.013 INSR rs890859 black 0.087 HCTZ 0.558 0.754 +HCTZ 0.350 0.882 INSR rs480 4414 black 0.243 HCTZ 0.267 0.855 +HCTZ 1.711 0.212 INSR rs4804418 black 0.136 HCTZ 0.212 0.901 +HCTZ 2.615 0.122 INSR rs890860 black 0.285 HCTZ 1.196 0.340 +HCTZ 1.377 0.323 INSR rs890861 black 0.037 HCTZ 4.419 0.296 +HCTZ 0.313 0.904 INSR rs725 5710 black 0.079 HCTZ 0.584 0.804 +HCTZ 0.876 0.685 INSR rs12460755 black 0.050 HCTZ 2.003 0.475 +HCTZ 2.965 0.243 INSR rs17175790 black 0.029 HCTZ 0.565 0.903 +HCTZ 2.896 0.310 INSR rs10420008 black 0.218 HCTZ 1.736 0.217 +HCTZ 1.199 0.378 INSR rs1 7175860 black 0.121 HCTZ 0.058 0.974 +HCTZ 2.663 0.148 INSR rs4804428 black 0.356 HCTZ 0.671 0.571 +HCTZ 1.441 0.217 INSR rs17254521 black 0.319 HCTZ 1.968 0.125 +HCTZ 2.594 0.040 INSR rs3852876 black 0.483 HCTZ 0.641 0.589 +HCTZ 2.152 0.072 INSR rs 10424224 black 0.487 HCTZ 1.764 0.127 +HCTZ 1.713 0.153 INSR rs4804433 black 0.374 HCTZ 0.224 0.848 +HCTZ 0.807 0.489 INSR rs1346490 black 0.401 HCTZ 0.525 0.673 +HCTZ 0.908 0.427 INSR rs10404318 black 0.161 HCTZ 2.224 0.149 +HCTZ 1.180 0.479 INSR rs6417197 black 0.000 HCTZ 0.000 1.000 +HCTZ 0.000 1.000 INSR rs11671297 black 0.441 HCTZ 0.187 0.882 +HCTZ 1.534 0.191 INSR rs10415205 black 0.312 HCTZ 0.103 0.937 +HCTZ 0.584 0.644 INSR rs11668751 black 0.372 HCTZ 0.584 0.650 +HCTZ 0.067 0.958 IN SR rs4247374 black 0.023 HCTZ 0.569 0.872 +HCTZ 5.190 0.264 INSR rs10402346 black 0.471 HCTZ 1.270 0.291 +HCTZ 0.126 0.919 INSR rs10417205 black 0.326 HCTZ 0.570 0.669 +HCTZ 0.655 0.589 INSR rs4804195 black 0.253 HCTZ 0.847 0.524 +HCTZ 2.943 0.034 I NSR rs919275 black 0.270 HCTZ 1.254 0.345 +HCTZ 0.100 0.938 INSR rs7258382 black 0.443 HCTZ 0.388 0.745 +HCTZ 0.708 0.528 INSR rs4804219 black 0.154 HCTZ 2.568 0.112 +HCTZ 3.317 0.044 INSR rs6510976 black 0.357 HCTZ 0.614 0.647 +HCTZ 1.287 0.295 INS R rs3745544 black 0.000 HCTZ 1.000 +HCTZ 1.000 INSR rs10408111 black 0.092 HCTZ 2.697 0.141 +HCTZ 1.580 0.472

PAGE 184

184 Table C 7 Continued. Gene SNP Race MAF Tx PE P value Tx PE P value INSR rs12979424 black 0.062 HCTZ 4.160 0.136 +HCTZ 1.184 0.585 INSR r s10411676 black 0.365 HCTZ 0.739 0.553 +HCTZ 0.480 0.686 INSR rs7254921 black 0.354 HCTZ 0.254 0.834 +HCTZ 0.921 0.430 INSR rs10404166 black 0.253 HCTZ 1.615 0.245 +HCTZ 0.332 0.816 INSR rs7254060 black 0.478 HCTZ 0.056 0.962 +HCTZ 0.601 0.614 IN SR rs2860184 black 0.133 HCTZ 0.381 0.810 +HCTZ 0.265 0.887 INSR rs10410272 black 0.030 HCTZ 2.530 0.395 +HCTZ 1.976 0.647 INSR rs8111710 black 0.168 HCTZ 0.940 0.556 +HCTZ 2.150 0.232 INSR rs8101064 black 0.191 HCTZ 0.524 0.690 +HCTZ 0.703 0.646 I NSR rs7507911 black 0.409 HCTZ 0.095 0.936 +HCTZ 0.708 0.551 IRS1 rs16822570 black 0.054 HCTZ 0.358 0.895 +HCTZ 5.808 0.038 IRS1 rs16822573 black 0.054 HCTZ 0.358 0.895 +HCTZ 5.808 0.038 IRS1 rs16822574 black 0.086 HCTZ 0.410 0.860 +HCTZ 3.700 0.077 IRS1 rs16822579 black 0.035 HCTZ 1.571 0.639 +HCTZ 0.045 0.988 IRS1 rs17208470 black 0.013 HCTZ 7.218 0.125 +HCTZ 5.540 0.317 IRS1 rs10182336 black 0.359 HCTZ 1.484 0.230 +HCTZ 0.493 0.670 IRS1 rs10181778 black 0.128 HCTZ 0.009 0.996 +HCTZ 4.516 0 .013 IRS1 rs16822601 black 0.262 HCTZ 1.761 0.208 +HCTZ 0.129 0.920 IRS1 rs16822604 black 0.273 HCTZ 1.768 0.189 +HCTZ 0.503 0.682 IRS1 rs7567312 black 0.027 HCTZ 9.115 0.076 +HCTZ 0.700 0.781 IRS1 rs1078533 black 0.223 HCTZ 2.077 0.137 +HCTZ 1.304 0.315 IRS1 rs2435185 black 0.003 HCTZ 4.565 0.529 +HCTZ 2.781 0.771 IRS1 rs16822626 black 0.091 HCTZ 0.971 0.650 +HCTZ 1.733 0.419 IRS1 rs16822630 black 0.284 HCTZ 1.375 0.288 +HCTZ 0.071 0.952 IRS1 rs16822638 black 0.153 HCTZ 1.604 0.314 +HCTZ 1.4 66 0.314 IRS1 rs10498212 black 0.140 HCTZ 1.753 0.298 +HCTZ 0.223 0.880 IRS1 rs4675094 black 0.245 HCTZ 1.957 0.140 +HCTZ 0.285 0.826 IRS1 rs10170579 black 0.030 HCTZ 2.784 0.377 +HCTZ 4.332 0.309 IRS1 rs2288586 black 0.144 HCTZ 1.820 0.286 +HCTZ 0.4 10 0.788 IRS1 rs3769647 black 0.017 HCTZ 5.286 0.303 +HCTZ 8.779 0.024 IRS1 rs1801278 black 0.059 HCTZ 2.496 0.297 +HCTZ 3.616 0.189 IRS1 rs2229613 black 0.102 HCTZ 2.663 0.169 +HCTZ 0.058 0.977 IRS1 rs1801123 black 0.470 HCTZ 0.000 1.000 +HCTZ 0.71 1 0.544 IRS1 rs2234931 black 0.060 HCTZ 2.393 0.306 +HCTZ 3.616 0.189 IRS1 rs6725330 black 0.223 HCTZ 0.058 0.966 +HCTZ 0.959 0.463 IRS1 rs6725556 black 0.233 HCTZ 0.278 0.840 +HCTZ 0.050 0.970 IRS1 rs13423855 black 0.076 HCTZ 0.137 0.953 +HCTZ 0.601 0.816 IRS1 rs13018009 black 0.002 HCTZ 0.000 1.000 +HCTZ 5.261 0.574 IRS1 rs4675096 black 0.480 HCTZ 1.264 0.285 +HCTZ 0.089 0.942

PAGE 185

185 Table C 7 Continued. Gene SNP Race MAF Tx PE P value Tx PE P value IRS2 rs1044364 black 0.174 HCTZ 0.763 0.640 +HCTZ 0.182 0.902 IRS2 rs1865434 black 0.185 HCTZ 0.427 0.766 +HCTZ 1.659 0.274 IRS2 rs7982446 black 0.294 HCTZ 2.079 0.101 +HCTZ 0.346 0.773 IRS2 rs913949 black 0.308 HCTZ 0.382 0.755 +HCTZ 2.049 0.139 IRS2 rs9587980 black 0.133 HCTZ 2.748 0.171 +H CTZ 2.049 0.202 IRS2 rs12583454 black 0.002 HCTZ 19.657 0.067 +HCTZ 0.000 1.000 IRS2 rs9559646 black 0.220 HCTZ 0.668 0.607 +HCTZ 2.291 0.138 IRS2 rs4773087 black 0.218 HCTZ 3.321 0.015 +HCTZ 1.098 0.463 IRS2 rs4773088 black 0.297 HCTZ 1.619 0.1 93 +HCTZ 0.470 0.749 IRS2 rs9515120 black 0.010 HCTZ 6.002 0.411 +HCTZ 3.339 0.543 IRS2 rs7323191 black 0.334 HCTZ 2.584 0.041 +HCTZ 1.426 0.263 IRS2 rs7999797 black 0.166 HCTZ 2.245 0.149 +HCTZ 1.744 0.285 IRS2 rs9521510 black 0.052 HCTZ 5.512 0. 017 +HCTZ 1.429 0.677 IRS2 rs4771646 black 0.317 HCTZ 1.167 0.350 +HCTZ 1.036 0.446 IRS2 rs9559656 black 0.277 HCTZ 0.771 0.540 +HCTZ 1.020 0.443 IRS2 rs9583423 black 0.116 HCTZ 2.196 0.266 +HCTZ 2.111 0.240 IRS2 rs7997595 black 0.092 HCTZ 1.12 9 0.549 +HCTZ 5.066 0.031 IRS2 rs7981705 black 0.393 HCTZ 1.151 0.340 +HCTZ 1.376 0.251 IRS2 rs4773092 black 0.388 HCTZ 0.329 0.791 +HCTZ 0.443 0.718 IRS2 rs4773094 black 0.208 HCTZ 0.405 0.770 +HCTZ 1.797 0.196 SLC2A2 rs7610064 black 0.045 HCTZ 2 .303 0.416 +HCTZ 7.329 0.008 SLC2A2 rs5397 black 0.000 HCTZ 0.000 1.000 +HCTZ 0.000 1.000 SLC2A2 rs16855638 black 0.112 HCTZ 0.013 0.994 +HCTZ 0.326 0.845 SLC2A2 rs10513684 black 0.034 HCTZ 1.573 0.617 +HCTZ 0.708 0.813 SLC2A2 rs10513685 black 0.38 1 HCTZ 0.816 0.451 +HCTZ 0.743 0.519 SLC2A2 rs11711437 black 0.490 HCTZ 0.751 0.479 +HCTZ 1.111 0.306 SLC2A2 rs5400 black 0.460 HCTZ 0.059 0.956 +HCTZ 1.616 0.135 SLC2A2 rs11924032 black 0.393 HCTZ 0.351 0.768 +HCTZ 1.214 0.339 SLC2A2 rs11925298 bla ck 0.111 HCTZ 3.552 0.047 +HCTZ 0.826 0.686 SLC2A2 rs9828378 black 0.094 HCTZ 1.168 0.545 +HCTZ 4.390 0.026 SLC2A2 rs12488694 black 0.064 HCTZ 2.875 0.192 +HCTZ 0.928 0.716 SLC2A2 rs5396 black 0.357 HCTZ 0.958 0.409 +HCTZ 0.440 0.730 SLC2A4 rs80766 49 black 0.146 HCTZ 0.198 0.899 +HCTZ 1.608 0.326 SLC2A4 rs2654185 black 0.367 HCTZ 0.145 0.905 +HCTZ 1.302 0.302 SLC2A4 rs9902838 black 0.366 HCTZ 1.910 0.134 +HCTZ 0.024 0.983 SLC2A4 rs2073476 black 0.002 HCTZ 0.000 1.000 +HCTZ 0.000 1.000 SLC2A4 rs5411 black 0.106 HCTZ 0.549 0.750 +HCTZ 0.601 0.735 SLC2A4 rs5412 black 0.025 HCTZ 2.073 0.607 +HCTZ 0.755 0.832 SLC2A4 rs5415 black 0.076 HCTZ 4.767 0.037 +HCTZ 3.015 0.218

PAGE 186

1 86 Table C 7 Continued. Gene SNP Race MAF Tx PE P value Tx PE P value SLC2 A4 rs222847 black 0.012 HCTZ 0.995 0.848 +HCTZ 5.947 0.276 SLC2A4 rs16956647 black 0.045 HCTZ 3.561 0.219 +HCTZ 3.434 0.243 SLC2A4 rs5435 black 0.122 HCTZ 4.518 0.011 +HCTZ 0.830 0.672 SLC2A4 rs5436 black 0.064 HCTZ 1.581 0.507 +HCTZ 0.120 0.954 F OXO1 rs7325210 non black 0.001 HCTZ 0.000 1.000 +HCTZ 0.073 0.993 FOXO1 rs2701858 non black 0.048 HCTZ 0.531 0.791 +HCTZ 0.446 0.803 FOXO1 rs17446614 non black 0.145 HCTZ 2.666 0.021 +HCTZ 0.122 0.919 FOXO1 rs2701859 non black 0.244 HCTZ 0.905 0.348 + HCTZ 0.757 0.455 FOXO1 rs2755211 non black 0.244 HCTZ 0.905 0.348 +HCTZ 0.757 0.455 FOXO1 rs2995991 non black 0.426 HCTZ 0.574 0.496 +HCTZ 0.290 0.723 FOXO1 rs2984121 non black 0.181 HCTZ 0.976 0.351 +HCTZ 0.497 0.636 FOXO1 rs3858869 non black 0.0 00 HCTZ 0.000 1.000 +HCTZ 0.000 1.000 FOXO1 rs4943795 non black 0.298 HCTZ 1.073 0.251 +HCTZ 0.995 0.287 FOXO1 rs3908774 non black 0.000 HCTZ 0.000 1.000 +HCTZ 0.000 1.000 FOXO1 rs12876443 non black 0.097 HCTZ 1.335 0.345 +HCTZ 0.606 0.658 FOXO1 rs798 1045 non black 0.251 HCTZ 0.079 0.937 +HCTZ 1.232 0.187 FOXO1 rs7139990 non black 0.051 HCTZ 0.452 0.815 +HCTZ 0.718 0.698 FOXO1 rs9603776 non black 0.026 HCTZ 0.300 0.898 +HCTZ 0.143 0.956 GYS1 rs2387583 non black 0.137 HCTZ 0.200 0.875 +HCTZ 0.029 0 .980 GYS1 rs4645894 non black 0.010 HCTZ 3.633 0.285 +HCTZ 10.283 0.082 GYS1 rs4645900 non black 0.039 HCTZ 2.336 0.306 +HCTZ 1.582 0.444 GYS1 rs905238 non black 0.485 HCTZ 0.727 0.387 +HCTZ 0.501 0.542 GYS1 rs4645903 non black 0.103 HCTZ 1.599 0.29 1 +HCTZ 0.678 0.591 GYS1 rs4645905 non black 0.477 HCTZ 0.144 0.865 +HCTZ 0.622 0.454 GYS1 rs4645908 non black 0.001 HCTZ 5.475 0.563 +HCTZ 0.000 1.000 GYS1 rs1042265 non black 0.113 HCTZ 0.607 0.675 +HCTZ 0.977 0.427 GYS1 rs8104760 non black 0.039 H CTZ 4.524 0.028 +HCTZ 1.119 0.630 GYS1 rs13306418 non black 0.002 HCTZ 0.000 1.000 +HCTZ 9.447 0.110 GYS1 rs12980441 non black 0.383 HCTZ 0.140 0.866 +HCTZ 1.015 0.234 GYS1 rs5464 non black 0.299 HCTZ 0.666 0.448 +HCTZ 0.514 0.588 GYS1 rs5460 non bl ack 0.000 HCTZ 0.000 1.000 +HCTZ 0.000 1.000 GYS1 rs2287755 non black 0.083 HCTZ 3.793 0.022 +HCTZ 0.015 0.992 GYS1 rs12610125 non black 0.052 HCTZ 0.746 0.690 +HCTZ 1.781 0.344 INS rs4320932 non black 0.226 HCTZ 0.265 0.793 +HCTZ 0.148 0.884 INS rs7 924316 non black 0.479 HCTZ 0.071 0.935 +HCTZ 0.350 0.688 INS rs3842768 non black 0.000 HCTZ 0.000 1.000 +HCTZ 0.000 1.000 INS rs689 non black 0.278 HCTZ 1.022 0.285 +HCTZ 0.284 0.761 INS rs2070762 non black 0.491 HCTZ 0.903 0.299 +HCTZ 0.705 0.426 I NS rs6356 non black 0.389 HCTZ 1.061 0.244 +HCTZ 0.707 0.401

PAGE 187

187 Table C 7 Continued. Gene SNP Race MAF Tx PE P value Tx PE P value INS rs7950050 non black 0.000 HCTZ 0.000 1.000 +HCTZ 0.000 1.000 INSR rs12150997 non black 0.181 HCTZ 0.504 0.658 +HCTZ 0.353 0.737 INSR rs3745551 non black 0.383 HCTZ 0.762 0.406 +HCTZ 0.352 0.663 INSR rs3745550 non black 0.194 HCTZ 1.873 0.085 +HCTZ 0.287 0.764 INSR rs6510947 non black 0.102 HCTZ 1.106 0.415 +HCTZ 1.448 0.289 INSR rs10408374 non black 0.136 HCTZ 2 .265 0.095 +HCTZ 1.418 0.230 INSR rs2860172 non black 0.181 HCTZ 0.342 0.760 +HCTZ 0.730 0.480 INSR rs13306446 non black 0.115 HCTZ 1.719 0.215 +HCTZ 1.602 0.206 INSR rs2860173 non black 0.076 HCTZ 0.727 0.640 +HCTZ 0.356 0.812 INSR rs2860175 non black 0.134 HCTZ 2.267 0.095 +HCTZ 1.634 0.170 INSR rs1549616 non black 0.079 HCTZ 0.791 0.607 +HCTZ 0.663 0.655 INSR rs10420382 non black 0.090 HCTZ 2.152 0.189 +HCTZ 0.861 0.563 INSR rs11672739 non black 0.108 HCTZ 0.352 0.795 +HCTZ 1.414 0.290 INSR rs6510949 non black 0.062 HCTZ 0.009 0.996 +HCTZ 1.026 0.566 INSR rs12610022 non black 0.063 HCTZ 0.418 0.830 +HCTZ 2.840 0.089 INSR rs6510950 non black 0.072 HCTZ 0.848 0.647 +HCTZ 2.708 0.087 INSR rs11667110 non black 0.297 HCTZ 0.239 0.807 + HCTZ 0.187 0.834 INSR rs16990074 non black 0.006 HCTZ 1.685 0.788 +HCTZ 3.088 0.540 INSR rs16994200 non black 0.006 HCTZ 1.685 0.788 +HCTZ 3.088 0.540 INSR rs8103483 non black 0.479 HCTZ 0.388 0.661 +HCTZ 0.164 0.853 INSR rs16994220 non black 0.002 HCTZ 0.000 1.000 +HCTZ 9.163 0.114 INSR rs16994221 non black 0.049 HCTZ 1.988 0.319 +HCTZ 0.514 0.821 INSR rs6510952 non black 0.026 HCTZ 0.168 0.947 +HCTZ 3.530 0.214 INSR rs8102612 non black 0.001 HCTZ 25.074 0.004 +HCTZ 0.000 1.000 INSR rs8100109 no n black 0.012 HCTZ 1.450 0.701 +HCTZ 1.201 0.748 INSR rs2059807 non black 0.372 HCTZ 0.652 0.459 +HCTZ 0.551 0.525 INSR rs3815901 non black 0.469 HCTZ 0.326 0.699 +HCTZ 0.666 0.408 INSR rs2860177 non black 0.260 HCTZ 0.150 0.870 +HCTZ 1.399 0.157 INSR rs6510955 non black 0.262 HCTZ 0.071 0.938 +HCTZ 1.351 0.170 INSR rs7252268 non black 0.204 HCTZ 0.992 0.361 +HCTZ 0.548 0.577 INSR rs10415589 non black 0.001 HCTZ 0.000 1.000 +HCTZ 2.758 0.762 INSR rs8109559 non black 0.202 HCTZ 0.969 0.372 + HCTZ 0.486 0.621 INSR rs10411667 non black 0.204 HCTZ 0.980 0.369 +HCTZ 0.563 0.574 INSR rs8112883 non black 0.296 HCTZ 0.035 0.969 +HCTZ 0.966 0.287 INSR rs8105406 non black 0.000 HCTZ 0.000 1.000 +HCTZ 0.000 1.000 INSR rs16994316 non black 0.002 HC TZ 1.696 0.848 +HCTZ 4.590 0.584 INSR rs8108622 non black 0.234 HCTZ 0.480 0.605 +HCTZ 0.825 0.405 INSR rs10500204 non black 0.290 HCTZ 0.354 0.687 +HCTZ 0.953 0.310 INSR rs6510959 non black 0.174 HCTZ 0.871 0.417 +HCTZ 0.639 0.538

PAGE 188

188 Table C 7 Con tinued. Gene SNP Race MAF Tx PE P value Tx PE P value INSR rs891087 non black 0.113 HCTZ 0.255 0.844 +HCTZ 0.379 0.751 INSR rs891088 non black 0.291 HCTZ 0.495 0.597 +HCTZ 0.486 0.565 INSR rs4804366 non black 0.113 HCTZ 0.175 0.894 +HCTZ 0.656 0.579 INSR rs10416539 non black 0.115 HCTZ 0.204 0.876 +HCTZ 0.618 0.599 INSR rs1035939 non black 0.281 HCTZ 0.207 0.827 +HCTZ 0.206 0.820 INSR rs2860183 non black 0.396 HCTZ 0.466 0.589 +HCTZ 0.088 0.914 INSR rs12460089 non black 0.390 HCTZ 0.151 0.861 +HC TZ 0.203 0.803 INSR rs4804377 non black 0.283 HCTZ 0.329 0.728 +HCTZ 0.302 0.738 INSR rs2115386 non black 0.490 HCTZ 0.266 0.744 +HCTZ 0.274 0.729 INSR rs4499341 non black 0.389 HCTZ 0.445 0.601 +HCTZ 0.183 0.820 INSR rs6510960 non black 0.125 HCTZ 1.397 0.260 +HCTZ 0.570 0.623 INSR rs1896639 non black 0.446 HCTZ 0.461 0.548 +HCTZ 0.407 0.570 INSR rs2042902 non black 0.484 HCTZ 0.714 0.385 +HCTZ 0.413 0.603 INSR rs10426094 non black 0.224 HCTZ 1.200 0.218 +HCTZ 0.680 0.462 INSR rs12609995 n on black 0.261 HCTZ 0.035 0.971 +HCTZ 0.171 0.853 INSR rs12459488 non black 0.284 HCTZ 0.220 0.807 +HCTZ 1.557 0.086 INSR rs3745545 non black 0.173 HCTZ 0.062 0.956 +HCTZ 0.693 0.492 INSR rs8104314 non black 0.001 HCTZ 0.000 1.000 +HCTZ 0.929 0.912 INSR rs12971499 non black 0.424 HCTZ 0.724 0.395 +HCTZ 0.600 0.455 INSR rs7245562 non black 0.076 HCTZ 0.658 0.680 +HCTZ 0.700 0.648 INSR rs4804404 non black 0.152 HCTZ 1.269 0.313 +HCTZ 1.048 0.356 INSR rs7508679 non black 0.413 HCTZ 0.534 0.546 +HCTZ 0.264 0.762 INSR rs890859 non black 0.001 HCTZ 0.000 1.000 +HCTZ 0.929 0.912 INSR rs4804414 non black 0.409 HCTZ 0.155 0.856 +HCTZ 0.428 0.619 INSR rs4804418 non black 0.227 HCTZ 0.782 0.441 +HCTZ 0.049 0.963 INSR rs890860 non black 0.198 HCTZ 1.521 0.137 +HCTZ 1.170 0.252 INSR rs890861 non black 0.087 HCTZ 0.928 0.578 +HCTZ 0.656 0.639 INSR rs7255710 non black 0.011 HCTZ 2.261 0.610 +HCTZ 4.589 0.232 INSR rs12460755 non black 0.296 HCTZ 0.982 0.269 +HCTZ 0.005 0.996 INSR rs17175790 no n black 0.079 HCTZ 0.356 0.835 +HCTZ 1.428 0.346 INSR rs10420008 non black 0.229 HCTZ 1.447 0.164 +HCTZ 0.469 0.621 INSR rs17175860 non black 0.197 HCTZ 0.846 0.418 +HCTZ 0.157 0.882 INSR rs4804428 non black 0.188 HCTZ 0.592 0.587 +HCTZ 0.332 0.734 INSR rs17254521 non black 0.488 HCTZ 0.001 0.999 +HCTZ 0.356 0.666 INSR rs3852876 non black 0.308 HCTZ 0.112 0.898 +HCTZ 0.048 0.956 INSR rs10424224 non black 0.390 HCTZ 0.294 0.737 +HCTZ 0.534 0.535 INSR rs4804433 non black 0.222 HCTZ 1.522 0.158 +HCTZ 0.237 0.811 INSR rs1346490 non black 0.396 HCTZ 1.363 0.118 +HCTZ 0.172 0.824 INSR rs10404318 non black 0.013 HCTZ 0.400 0.933 +HCTZ 0.777 0.855

PAGE 189

189 Table C 7 Continued. Gene SNP Race MAF Tx PE P value Tx PE P value INSR rs6417197 non black 0.0 01 HCTZ 0.000 1.000 +HCTZ 7.196 0.397 INSR rs11671297 non black 0.440 HCTZ 0.218 0.802 +HCTZ 0.302 0.728 INSR rs10415205 non black 0.282 HCTZ 1.244 0.169 +HCTZ 0.765 0.424 INSR rs11668751 non black 0.290 HCTZ 1.361 0.134 +HCTZ 0.708 0.455 INSR rs42 47374 non black 0.134 HCTZ 1.574 0.233 +HCTZ 0.387 0.734 INSR rs10402346 non black 0.291 HCTZ 1.361 0.134 +HCTZ 0.548 0.564 INSR rs10417205 non black 0.328 HCTZ 0.375 0.668 +HCTZ 0.719 0.422 INSR rs4804195 non black 0.431 HCTZ 0.128 0.883 +HCTZ 0.118 0.890 INSR rs919275 non black 0.396 HCTZ 1.322 0.148 +HCTZ 0.732 0.356 INSR rs7258382 non black 0.157 HCTZ 1.776 0.128 +HCTZ 0.015 0.990 INSR rs4804219 non black 0.045 HCTZ 1.566 0.402 +HCTZ 0.787 0.765 INSR rs6510976 non black 0.471 HCTZ 1.560 0 .065 +HCTZ 0.559 0.479 INSR rs3745544 non black 0.000 HCTZ 1.000 +HCTZ 1.000 INSR rs10408111 non black 0.000 HCTZ 0.000 1.000 +HCTZ 0.000 1.000 INSR rs12979424 non black 0.329 HCTZ 0.125 0.893 +HCTZ 0.668 0.459 INSR rs10411676 non black 0.418 HCTZ 0.742 0.395 +HCTZ 0.386 0.647 INSR rs7254921 non black 0.490 HCTZ 1.691 0.054 +HCTZ 0.261 0.753 INSR rs10404166 non black 0.002 HCTZ 0.000 1.000 +HCTZ 0.924 0.913 INSR rs7254060 non black 0.076 HCTZ 0.832 0.604 +HCTZ 1.931 0.230 INSR rs2860184 non b lack 0.366 HCTZ 1.367 0.128 +HCTZ 0.339 0.700 INSR rs10410272 non black 0.000 HCTZ 0.000 1.000 +HCTZ 0.000 1.000 INSR rs8111710 non black 0.252 HCTZ 0.644 0.539 +HCTZ 0.647 0.500 INSR rs8101064 non black 0.053 HCTZ 0.852 0.600 +HCTZ 1.379 0.500 INSR rs7507911 non black 0.165 HCTZ 1.785 0.114 +HCTZ 0.704 0.560 IRS1 rs16822570 non black 0.060 HCTZ 0.509 0.777 +HCTZ 0.786 0.633 IRS1 rs16822573 non black 0.058 HCTZ 0.509 0.777 +HCTZ 0.994 0.569 IRS1 rs16822574 non black 0.058 HCTZ 0.509 0.777 +HCTZ 1.358 0.444 IRS1 rs16822579 non black 0.000 HCTZ 0.000 1.000 +HCTZ 0.000 1.000 IRS1 rs17208470 non black 0.107 HCTZ 1.208 0.387 +HCTZ 0.778 0.568 IRS1 rs10182336 non black 0.084 HCTZ 1.650 0.283 +HCTZ 0.698 0.628 IRS1 rs10181778 non black 0.058 HCTZ 0.104 0.955 +HCTZ 0.583 0.728 IRS1 rs16822601 non black 0.026 HCTZ 6.617 0.006 +HCTZ 4.444 0.118 IRS1 rs16822604 non black 0.026 HCTZ 6.283 0.008 +HCTZ 1.897 0.515 IRS1 rs7567312 non black 0.000 HCTZ 0.000 1.000 +HCTZ 0.000 1.000 IRS1 rs1078533 non black 0.024 HCTZ 5.968 0.015 +HCTZ 3.016 0.324 IRS1 rs2435185 non black 0.021 HCTZ 0.297 0.930 +HCTZ 0.093 0.973 IRS1 rs16822626 non black 0.000 HCTZ 0.000 1.000 +HCTZ 0.000 1.000 IRS1 rs16822630 non black 0.028 HCTZ 6.209 0.006 +HCTZ 0.783 0.757 IR S1 rs16822638 non black 0.083 HCTZ 1.086 0.469 +HCTZ 0.649 0.655

PAGE 190

190 Table C 7 Continued. Gene SNP Race MAF Tx PE P value Tx PE P value IRS1 rs10498212 non black 0.081 HCTZ 0.847 0.586 +HCTZ 0.453 0.757 IRS1 rs4675094 non black 0.098 HCTZ 1.820 0.177 +H CTZ 0.990 0.475 IRS1 rs10170579 non black 0.000 HCTZ 0.000 1.000 +HCTZ 0.000 1.000 IRS1 rs2288586 non black 0.081 HCTZ 0.847 0.586 +HCTZ 0.448 0.759 IRS1 rs3769647 non black 0.073 HCTZ 0.145 0.928 +HCTZ 0.304 0.843 IRS1 rs1801278 non black 0.050 HCT Z 4.433 0.035 +HCTZ 0.257 0.888 IRS1 rs2229613 non black 0.001 HCTZ 3.366 0.703 +HCTZ 0.000 1.000 IRS1 rs1801123 non black 0.101 HCTZ 1.999 0.125 +HCTZ 0.798 0.564 IRS1 rs2234931 non black 0.051 HCTZ 4.433 0.035 +HCTZ 0.255 0.886 IRS1 rs6725330 non black 0.115 HCTZ 0.970 0.419 +HCTZ 2.097 0.114 IRS1 rs6725556 non black 0.062 HCTZ 2.658 0.083 +HCTZ 3.655 0.062 IRS1 rs13423855 non black 0.000 HCTZ 0.000 1.000 +HCTZ 0.000 1.000 IRS1 rs13018009 non black 0.024 HCTZ 2.813 0.281 +HCTZ 4.944 0.153 I RS1 rs4675096 non black 0.080 HCTZ 1.665 0.320 +HCTZ 1.156 0.444 IRS2 rs1044364 non black 0.052 HCTZ 1.802 0.328 +HCTZ 0.352 0.858 IRS2 rs1865434 non black 0.153 HCTZ 0.096 0.936 +HCTZ 0.148 0.891 IRS2 rs7982446 non black 0.052 HCTZ 1.804 0.329 +HCTZ 1.214 0.546 IRS2 rs913949 non black 0.193 HCTZ 0.524 0.622 +HCTZ 0.391 0.709 IRS2 rs9587980 non black 0.001 HCTZ 0.000 1.000 +HCTZ 6.189 0.495 IRS2 rs12583454 non black 0.017 HCTZ 2.371 0.455 +HCTZ 0.398 0.901 IRS2 rs9559646 non black 0.498 HCTZ 0. 202 0.819 +HCTZ 0.534 0.500 IRS2 rs4773087 non black 0.499 HCTZ 0.491 0.580 +HCTZ 0.081 0.918 IRS2 rs4773088 non black 0.308 HCTZ 1.295 0.169 +HCTZ 0.036 0.968 IRS2 rs9515120 non black 0.062 HCTZ 0.988 0.586 +HCTZ 0.361 0.839 IRS2 rs7323191 non blac k 0.142 HCTZ 0.509 0.696 +HCTZ 0.371 0.750 IRS2 rs7999797 non black 0.482 HCTZ 0.250 0.784 +HCTZ 0.368 0.646 IRS2 rs9521510 non black 0.356 HCTZ 0.522 0.559 +HCTZ 0.386 0.656 IRS2 rs4771646 non black 0.319 HCTZ 1.118 0.230 +HCTZ 0.260 0.765 IRS2 rs955 9656 non black 0.283 HCTZ 1.315 0.188 +HCTZ 0.420 0.634 IRS2 rs9583423 non black 0.001 HCTZ 0.000 1.000 +HCTZ 6.189 0.495 IRS2 rs7997595 non black 0.167 HCTZ 0.921 0.419 +HCTZ 0.102 0.924 IRS2 rs7981705 non black 0.148 HCTZ 0.570 0.655 +HCTZ 0.487 0.6 82 IRS2 rs4773092 non black 0.383 HCTZ 1.620 0.059 +HCTZ 0.275 0.750 IRS2 rs4773094 non black 0.222 HCTZ 2.092 0.062 +HCTZ 0.288 0.764 SLC2A2 rs7610064 non black 0.001 HCTZ 1.147 0.855 +HCTZ 0.000 1.000 SLC2A2 rs5397 non black 0.003 HCTZ 1.409 0.784 + HCTZ 0.000 1.000 SLC2A2 rs16855638 non black 0.002 HCTZ 0.000 1.000 +HCTZ 3.382 0.588 SLC2A2 rs10513684 non black 0.046 HCTZ 1.160 0.617 +HCTZ 0.225 0.907 SLC2A2 rs10513685 non black 0.135 HCTZ 3.068 0.012 +HCTZ 1.338 0.273

PAGE 191

191 Table C 7 Continued. G ene SNP Race MAF Tx PE P value Tx PE P value SLC2A2 rs11711437 non black 0.140 HCTZ 2.789 0.019 +HCTZ 0.325 0.788 SLC2A2 rs5400 non black 0.140 HCTZ 2.789 0.019 +HCTZ 0.834 0.495 SLC2A2 rs11924032 non black 0.264 HCTZ 0.149 0.883 +HCTZ 0.359 0.690 S LC2A2 rs11925298 non black 0.020 HCTZ 0.939 0.742 +HCTZ 1.136 0.766 SLC2A2 rs9828378 non black 0.155 HCTZ 1.892 0.120 +HCTZ 0.190 0.862 SLC2A2 rs12488694 non black 0.020 HCTZ 0.939 0.742 +HCTZ 1.136 0.766 SLC2A2 rs5396 non black 0.288 HCTZ 0.203 0. 830 +HCTZ 0.159 0.859 SLC2A4 rs8076649 non black 0.001 HCTZ 0.000 1.000 +HCTZ 2.437 0.771 SLC2A4 rs2654185 non black 0.381 HCTZ 1.178 0.186 +HCTZ 1.006 0.248 SLC2A4 rs9902838 non black 0.013 HCTZ 3.865 0.537 +HCTZ 2.221 0.480 SLC2A4 rs2073476 non b lack 0.002 HCTZ 0.000 1.000 +HCTZ 10.823 0.130 SLC2A4 rs5411 non black 0.001 HCTZ 0.000 1.000 +HCTZ 2.437 0.771 SLC2A4 rs5412 non black 0.173 HCTZ 0.806 0.480 +HCTZ 0.634 0.565 SLC2A4 rs5415 non black 0.309 HCTZ 0.790 0.364 +HCTZ 1.102 0.234 SLC2A4 r s222847 non black 0.042 HCTZ 1.285 0.518 +HCTZ 0.310 0.891 SLC2A4 rs16956647 non black 0.011 HCTZ 3.865 0.537 +HCTZ 1.704 0.640 SLC2A4 rs5435 non black 0.375 HCTZ 1.084 0.226 +HCTZ 0.899 0.280 SLC2A4 rs5436 non black 0.000 HCTZ 0.000 1.000 +HCTZ 0.00 0 1.000 MAF= minor allele frequency; PE= parameter estimate; Tx= treatment

PAGE 192

192 Table C 8 Initial findings with and without PC1 and PC2 as covariates. All findings controlled for baseline glucose. Gene SNP Treatment Race PE P value Baseline glucose, PC1 a nd PC2 as covariates SCNN1G rs13306654 HCTZ Black 4.404 0.00199 SCNN1G rs4499239 HCTZ Black 4.422 0.00247 NOS3 rs3800787 HCTZ Black 5.602 0.00304 IRS1 rs1801278 ATEN Non black 5.524 0.00087 Only baseline glucose as covariate SCNN1G rs13306654 H CTZ Black 4.4613 0.00213 SCNN1G rs4499239 HCTZ Black 4.4352 0.00175 NOS3 rs3800787 HCTZ Black 5.7063 0.00227 IRS1 rs1801278 ATEN Non black 5.4255 0.00107 PE= parameter estimate

PAGE 193

193 Table C 9 Association between NOS3 rs3800787 genotype and change in gluc ose after HCTZ treatment among PEAR blacks. See Chapter 4 for additive model. Response MAF Hmz. Allele Carrier P value rs3800787 0.10 N 119 29 HCTZ monotherapy 3. 53 1.34 0.0099 N 112 21 HCTZ add on 3.48 0.97 0.0663 Glucose, mg/dL ; P value adjusted for base line glucose an d principal component 1 and 2; linear regression model. Dominant model depicted.

PAGE 194

194 Table C 1 0 Association between IRS1 rs1801278 genotype and change in glucose after treatment with ATEN or H CTZ in PEAR non blacks. See Chapter 4 for dominant model. Response MAF Major Hmz. Het. Minor Hmz. P value rs1801278 0.049 N 204 21 1 ATEN monotherapy 1. 60 5. 76 26.13 0.0009 N 200 19 HCTZ monotherapy 1.1 2 5.55 0.0348 Glucose, mg/dL ; P value adjusted for base line glucose an d principal component 1 and 2; linear regression model. Additive model depicted.

PAGE 195

195 Table C 11. Association between SCNN1G rs13306654 and rs4499239 genotype and change in glucose after treatment with HCTZ among PEAR blacks. Response MAF Major Hmz. Het. Minor Hmz. P value rs13306654 0.24 N 81 6 0 6 HCTZ monotherapy 0.0 1 5.36 1.91 0.0078 N 8 2 45 5 HCTZ add on 1. 68 3.88 8.80 0.0544 rs4499239 0.22 N 85 5 7 5 HCTZ monotherapy 0. 27 5.07 3.61 0.0089 N 85 45 4 HCTZ add on 0.93 2.79 10.09 0.0658 Glucose, mg/dL ; P value adjusted for age, gender, waist circumference, baseline glucose, baseline insulin and principal component 1 and 2.

PAGE 196

196 Table C 12. Association between NOS3 rs3800787 genotype and change in gl ucose after treatment with HCTZ among PEAR blacks. Response MAF Hmz. Het. Minor Hmz. P value rs3800787 0.10 N 118 27 2 HCTZ monotherapy 3.21 0.60 20.96 0.0014 N 110 20 1 HCTZ add on 2.56 2.03 4.35 0.0376 Glucose, mg/dL ; P value adjusted for age, gender, waist circumfe rence, baseline glucose, baseline insulin an d principal component 1 and 2.

PAGE 197

197 Table C 13. Association between IRS1 rs1801278 genotype and change in glucose after treatment with ATEN or HCTZ in PEAR non blacks. Response MAF Major Hmz. Het. Glu Minor Hmz. P value rs1801278 0.049 N 203 21 1 ATEN monotherapy 1.6 0 5.01 23.47 0.0046 N 200 19 HCTZ monotherapy 1.05 5.41 0.0354 Glucose, mg/dL ; P value adjusted for age, gender, waist circumference, baseline glucose, baseline insulin and principal component 1 and 2.

PAGE 198

198 Table C 14. Association between INSR rs7508679 genotype and change in glucose after treatment with ATEN monotherapy. Race group MAF Major Hmz. Het. Minor Hmz. P value rs7508679 N 0.339 59 63 21 Black 4.02 1.25 4.07 0.0034 N 0.413 78 117 33 Non black 4.49 0.4 2.32 0.0295 Glucose, mg/dL ; P value adjusted for age, gender, waist circumference, base line glucose, baseline insulin and principal component 1 and 2.

PAGE 199

199 Table C 1 5 Association between INSR rs7508679 genotype and change in glucose after ATEN treatment among PEAR blacks. See Appendix D for additive model. Race group MAF Hmz. Allele Carrier P value rs7508679 N 0.339 59 84 Black 4.02 0.08 0.0066 N 0.413 78 150 Non black 4.49 0.82 0.0180 Glucose, mg/dL ; P value adjusted for base line glucose an d principal component 1 and 2; linear regression model.

PAGE 200

200 APPENDIX D EFFECT OF ATENOLOL ON EXPRESSION OF THE INSULIN RECEPTOR BY GENOTYPE GROUP FOR THE RS7508679 SINGLE NUCLEOTIDE POLYMORPHISM Introduction Lymphocyte expression studies are an efficient means of analyzing gene expression without performing tis sue biopsies. 260 Many genes of interest in this project are moderately expressed in lymphocytes 219 220 Metabolic syndrome and insulin resistance are complex syndromes for w hich molecular level studies in patients treated with these drugs classes may provide insight. Linking altered expression to a clinical phenotype is important to fully evaluate and understand this complicated relationship and relate this information to cli nical practice. Limited data exists on the effects of antihypertensive therapy on altered expression, either in human research studies or in cell culture studies While the mechanism of antihypertensive effects for these agents has been well studied, the m echanisms of dysglycemia associated with these classes of medications are less well characterized. Based on findings from Table D 1 (and Figure D 1; see Chapter 4 for analysis methods) persons with one or two copies of the minor allele of the INSR rs7508 679 SNP experienced a lower elevation in glucose after atenolol monotherapy as compared to persons homozygous for the major allele. The change in glucose after atenolol treatment in rs7508679 minor allele carriers ranged from a moderate increase to a decre ase in glucose (from +2mg/dL to 4mg/dL change in glucose from baseline ) P ersons homozygous for the major allele showed an elevation in glucose (from +4 to +4.5mg/dL change in glucose from baseline ). Though we found the greatest change in the black race g roup, findings were directionally consistent within the non black race group. (See Chapter 4 for further detail on the complexity of insulin signaling, Figure 4 1

PAGE 201

201 for a relevant figure, and the role of INSR.) In blacks and non blacks, the minor allele was associated with a lower increase in glucose after ATEN monotherapy (p=0.0022 in blacks, p=0.1163 in non blacks; Table D 1 Figure D 1 ) compared to persons homozygous for the major allele Although in non blacks the p value was not significant, based on dir ectional consistency this SNP was considered moderately interesting. Though historically a bias existed in studying intronic SNPs, such as ident ified in the present study, GWAS methods have contributed significantly to our understanding. 261 Further, an evaluation of trait/disease associated SNPs collected from GWASs reported that nonsynonymous SNPs accounte d for 9%, intergenic SNPs accounted for 43% and intronic SNPs accounted for 45%. The remaining few percent were untranslated regions or synonymous SNPs. 262 Genetic variants in intronic splice enhancer or suppressor sites may result in aberrant splicing. Though the frequency of polymorphisms that affect splicing is uncertain some researchers have suggested more than 50% of all diseases are associated with aberrant splicing. While this seems an overestimate, it could also be an underestimate since studies are sometimes not carried out in the relevant target tissue in a suffic ient number of samples. Further, examples in pharmacogenomics exist (CYP2D6*41, CYP1A2*1F, DRD2, HMGCR) where an intronic polymorphism resulted in clinically meaningful phenotypic differences. 261 The aim of the present study was to further evaluate a pharmacogenomic signal in INSR which was associated with a differential change in glucose by genotype group. We assessed whether expression differences existed by INSR rs7508679 homozygous

PAGE 202

202 major allele versus minor allele groups, in PEAR patients that were treated with atenolol monotherapy. Methods Study Population The PEAR trial was a prospective, open label, m ulti center, randomized study designed to evaluate the genetic determinants of antihypertensive and adverse metabolic responses to HCTZ, atenolol, and their combination. 43 Additional details of the PEAR trial are provided in Chapter 3. INSR rs7508679 was selected for expression studies based on differential change in glucose after atenolol monotherapy, by PC confirmed race group. After atenolol treatment in black s, across genotype groups, addition of a minor allele(s) lessened the elevation in glucose (+4.02mg/dL for C/C, +1.25mg/dL for C/T and 4.07mg/dL for T/T, p=0.002). Though not as strong a signal, the same trend was observed in non blacks. Persons homozygou s for the major allele experienced the greatest elevation in glucose after atenolol treatment (C/C, +4.49mg/dL), and other groups experienced less of an elevation (C/T +0.4mg/dL, T/T +2.32mg/dL; p=0.1163 across genotype groups). While notable that the aten olol add on group did not exhibit the same trend (p=0.9378 for blacks; p=0.6835 for non blacks) these patients were treated with HCTZ for 9 weeks prior to 9 weeks of atenolol treatment, potentially modifying atenolol associated glucose effects. Patients we re selected for inclusion in the present study by rs7508679 genotype group. All patients homozygous for the minor allele with RNA samples available were

PAGE 203

203 included. Approximately the same number of patients homozygous for the major allele, with the most extr eme glucose response, was also selected. Genotyping Ge notypes for INSR were obtained from the Illumina HumanCVD BeadChip. Additional details on genotyping methods and quality control procedures are provided in Chapter 4. Sample Collection In both race g roups, all available patient samples homozygous for the minor allele (T/T) were included. The patients homozygous for the major allele (C/C) with the most extreme change in glucose after treatment with atenolol were included. Expression of INSR was measure d in 16 PEAR patients of PC confirmed African ancestry, 8 homozygous for the major allele (C/C) in rs7508679 and 8 homozygous for the minor allele (T/T). A total of 15 PEAR PC confirmed non African ancestry were also analyzed, 8 homozygous for the major al lele in rs7508679 and 7 homozygous for the minor allele. RNA was isolated from whole blood using the PAXgene Blood RNA kit #762165 (PreAnalytiX GmbH, Hombrechtikon, Switzerland). Considering the rapid degradation of intracellular RNA after collection, an a dditive that reduces RNA degradation and minimizes gene induction is included in the PAXgene Blood RNA Tube (2.5mL whole blood collected), which stabilizes the in vivo transcription profile. PAXgene Blood RNA Tubes were collected pre protocol, including im mediate inversion and storage at room temperature for at least two hour but not more than 72 hours before processing or freezing. Gene Expression Measurement

PAGE 204

204 A total of 1.1 g of RNA was used per 20 L reaction in conversion to cDNA using the Applied Biosys tems RNA to cDNA kit. For the expression assay, 1 L 20X TaqMan Gene Expression Assay was combined with 10 L 2X TaqMan Gene Expression Master Mix, cDNA template (1 100ng), and RNase free water (9 L minus X L of cDNA template), for a total reaction volume of 20 L per sample, each sample in triplicate. Four reference genes ( 18s, Hs03003631_g1; B2M 2 microglobulin Hs00984320_m1; GAPDH Hs99999905_m1; ACTB or actin, Hs01060665_g1) were evaluated to determine amplification efficiency compared to INSR (Hs00961554_m1) at various amounts of total input cDNA (10ng, 30ng, 50ng, and 100ng) in triplic ate. Based on the validation reaction, 2 microglobulin exhibited the least pairwise variation across differing amounts of cDNA ( C T values measured in triplicate) and was selected for the reference gene Based on C T levels for INSR at cDNA amounts betwee n10ng and 100ng, 30ng of cDNA was selected for the reaction. Expression measurement via quantitative real time RT PCR used TaqMan Gene Expression Assays and the TaqMan 7900HT Real Time PCR System (Applied Biosystems, Foster City, CA, USA). Statistical Anal ysis Expression was calculated via the comparative 2 Ct method 263 for relative quantification of INSR expression between rs7508679 genotype groups, where the major allele group was c onsidered the reference group. Relative quantification analysis was evaluated using DataAssist Software v3.01 (Applied Biosystems). A paired t test T (average C T for target gene average C T for reference gene) in IN SR expression after atenolol treatment. An unpaired t test was

PAGE 205

205 used to compare expression between INSR rs7508679 genotyped groups. Analysis was conducted in SAS 9.2 (SAS Institute, Cary, NC) with a p value < 0.05 considered significant. Results The change in glucose after treatment with atenolol for patients included in the INSR expression study is depicted in Table D 2 Though RNA was not available for all PEAR patients, samples were available and RT PCR was successful in a total of 31 patient samples. Me an C T values (Table D 3 ) were not different after treatment with atenolol, by race and genotype group. Though not significant, prior to atenolol treatment, blacks that were homozygous for the minor allele (T/T) had 10% higher relative INSR expression com pared to those homozygous for the major allele, C/C (Figure D 2 p=0.67); non blacks that were T/T had a 13% higher INSR expression compared to C/C ( Figure D 3 p=0.67). In persons homozygous for the major allele (C/C) in either race group, there was a non significant increase in expression ( Figure D 2 4% in blacks, Figure D 2, p=0.83; 5% in non blacks, p=0.87). However, in persons homozygous for the minor allele (T/T) results by race groups were directionally inconsistent. In T/T blacks, INSR expression ap peared to decrease after atenolol treatment (Figure D 2 ) and in T/T non blacks INSR expression appeared to increase after atenolol treatment (Figure D 3 ) Discussion To offer mechanistic insight into our findings in a SNP in INSR we evaluated expression o f INSR in a subset of PEAR patients by genotype groups for rs7508679. Though we previously found a differential change in glucose after atenolol treatment by

PAGE 206

206 rs7508679 genotype, expression was not significantly different between INSR rs7508679 T/T and C/C genotype groups by race group. Considering the stronger signal was in the black race group, though directionally consistent in the non black race group, analysis was conducted by race. However, based on expression work in both races, there were neither si gnificant findings by rs7508679 genotype groups nor consistent trends between races. The lack of consistency between race groups is inconsistent with previous findings that this polymorphism may impact glucose after atenolol treatment. While possible that problems inherent in conducting expression studies (RNA degredation, incorrect tissue, variation between replicates) negatively impacted our findings, these issues were addressed or justified previously. An important strength of our study was the availabil ity of both genotype information and samples for expression work in a large portion of PEAR patients before and after medication treatment. While our findings in the expression work did not meet statistical significance, the ability to confirm or refute fi ndings from genetic studies with expression work allows greater confidence in our overall results. For INSR expression, we did not detect a significant change by rs7508679 genotype group, before or after treatment with atenolol, by race group. Based on th ese findings, we are less confident that rs7508679 has a significant impact on glucose after treatment with atenolol. While INSR expression does not appear to vary by the intronic rs7508679 SNP, our data does not rule out the potential importance of other SNPs within INSR to impact change in glucose after antihypertensive treatment.

PAGE 207

207 Table D 1. Association between INSR rs7508679 genotype and change in glucose after treatment with ATEN monotherapy. Race group MAF Major Hmz. Het. Minor H mz. P value rs7508679 N 0.339 59 63 21 Black 4.02 1.25 4.07 0.0022 N 0.413 78 117 33 Non black 4.49 0.4 2.32 0.1163 Glucose, mg/dL; P value adjusted for base line glucose an d principal component 1 and 2; linear regression model. An add itive model was used. Results from a dominant model are included in Table C 13 in the appendix.

PAGE 208

208 Change in glucose by INSR genotype Figure D 1 Change in glucose after atenolol monotherapy by INSR rs7508679 genotype in PEAR blacks and non blacks

PAGE 209

209 Table D 2 Change in glucose of PEAR patients selected for INSR expression study Combined INSR rs7508679 genotype C/C T/T C/C T/T Black Non black N 31 8 8 8 7 after atenolol 7.9 (12.3) 16.3 (10.8) 2.7 (11.2) 14.3 (6.5) 3 .14 (10.1) Data are presented as no. (%) or mean + SD.

PAGE 210

210 Table D 3 Analysis of RT T by INSR rs7508679 genotype. INSR rs7508679 N T SD *P value Black C/C Baseline 8 12.29 0.66 After ATEN 8 12.21 0.58 0.8038 T/ T Baseline 8 12.14 0.62 After ATEN 8 12.60 0.92 0.3197 Non black C/C Baseline 8 11.90 0.69 After ATEN 8 11.89 0.90 0.9651 T/T Baseline 7 11.72 0.71 After ATEN 7 11.45 1.27 0.3981 *P value represents paired t test of me T

PAGE 211

211 Figure D 2 Relative INSR expression by rs7508679 genotype in leukocytes, in 16 black PEAR patients randomized to receive atenolol. Reference genotype is C/C. Expression is normalized to reference 2 microglobulin. C/C baseline versus at enolol treated p=0.8260; C/C baseline versus T/T baseline p=0.6588.

PAGE 212

212 Figure D 3 Relative INSR expression by rs7508679 genotype in leukocytes, in 15 non black PEAR patients randomized to receive atenolol. Reference genotype is C/C. Expression is normal 2 microglobulin. C/C baseline versus atenolol treated p=0.8717; C/C baseline versus T/T baseline p=0.6724.

PAGE 213

213 Figure D 4 Relative INSR expression by rs7508679 genotype in leukocytes, in 31 PEAR patients randomized to receive ate nolol. Reference genotype is C/C. 2 microglobulin.

PAGE 214

214 LIST OF REFERENCES 1. Roger VL, Go AS, Lloyd Jones DM, et al. Heart disease and stroke statistics -2012 update: a report from the America n Heart Association. Circulation 2012;125:e2 e220. 2. Jamison DT et al (eds): Disease Control Priorities in Developing Countries, 2nd ed. Washington, DC, Oxford University Press, 2006. 3. Lawes CM, Vander Hoorn S, Rodgers A. Global burden of blood pressure related disease, 2001. Lancet 2008;371:1513 8. 4. Lopez AD et al (eds): Global Burden of Disease and Risk Factors. Washington, DC, Oxford University Press, 2006. 5. Chobanian AV, Bakris GL, Black HR, et al. The Seventh Report of the Joint National Committ ee on Prevention, Detection, Evaluation, and Treatment of High Blood Pressure: the JNC 7 report. JAMA 2003;289:2560 72. 6. Franco OH, Peeters A, Bonneux L, de Laet C. Blood pressure in adulthood and life expectancy with cardiovascular disease in men and wo men: life course analysis. Hypertension 2005;46:280 6. 7. DeFronzo RA, Abdul Ghani M. Assessment and treatment of cardiovascular risk in prediabetes: impaired glucose tolerance and impaired fasting glucose. Am J Cardiol 2011;108:3B 24B. 8. Executive Summar y of The Third Report of The National Cholesterol Education Program (NCEP) Expert Panel on Detection, Evaluation, And Treatment of High Blood Cholesterol In Adults (Adult Treatment Panel III). JAMA 2001;285:2486 97. 9. Centers for Disease Control and Preve ntion. National diabetes fact sheet: national estimates and general information on diabetes and prediabetes in the United States, 2011. Atlanta, GA: U.S. Department of Health and Human Services, Centers for Disease Control and Prevention, 2011. 10. Major o utcomes in high risk hypertensive patients randomized to angiotensin converting enzyme inhibitor or calcium channel blocker vs diuretic: The Antihypertensive and Lipid Lowering Treatment to Prevent Heart Attack Trial (ALLHAT). JAMA 2002;288:2981 97. 11. Pe pine CJ, Handberg EM, Cooper DeHoff RM, et al. A calcium antagonist vs a non calcium antagonist hypertension treatment strategy for patients with coronary artery disease. The International Verapamil Trandolapril Study (INVEST): a randomized controlled tria l. JAMA 2003;290:2805 16.

PAGE 215

215 12. Prevention of stroke by antihypertensive drug treatment in older persons with isolated systolic hypertension. Final results of the Systolic Hypertension in the Elderly Program (SHEP). SHEP Cooperative Research Group. JAMA 1991 ;265:3255 64. 13. Ostman J, Asplund K, Bystedt T, et al. Comparison of effects of quinapril and metoprolol on glycaemic control, serum lipids, blood pressure, albuminuria and quality of life in non insulin dependent diabetes mellitus patients with hyperten sion. Swedish Quinapril Group. J Intern Med 1998;244:95 107. 14. Wright JT, Jr., Harris Haywood S, Pressel S, et al. Clinical outcomes by race in hypertensive patients with and without the metabolic syndrome: Antihypertensive and Lipid Lowering Treatment t o Prevent Heart Attack Trial (ALLHAT). Arch Intern Med 2008;168:207 17. 15. Black HR, Davis B, Barzilay J, et al. Metabolic and clinical outcomes in nondiabetic individuals with the metabolic syndrome assigned to chlorthalidone, amlodipine, or lisinopril a s initial treatment for hypertension: a report from the Antihypertensive and Lipid Lowering Treatment to Prevent Heart Attack Trial (ALLHAT). Diabetes Care 2008;31:353 60. 16. Kostis JB, Wilson AC, Freudenberger RS, Cosgrove NM, Pressel SL, Davis BR. Long term effect of diuretic based therapy on fatal outcomes in subjects with isolated systolic hypertension with and without diabetes. Am J Cardiol 2005;95:29 35. 17. Cooper Dehoff R, Cohen JD, Bakris GL, et al. Predictors of development of diabetes mellitus i n patients with coronary artery disease taking antihypertensive medications (findings from the INternational VErapamil SR Trandolapril STudy [INVEST]). Am J Cardiol 2006;98:890 4. 18. Gress TW, Nieto FJ, Shahar E, Wofford MR, Brancati FL. Hypertension and antihypertensive therapy as risk factors for type 2 diabetes mellitus. Atherosclerosis Risk in Communities Study. N Engl J Med 2000;342:905 12. 19. Holzgreve H, Nakov R, Beck K, Janka HU. Antihypertensive therapy with verapamil SR plus trandolapril versus atenolol plus chlorthalidone on glycemic control. Am J Hypertens 2003;16:381 6. 20. Ferrannini E, Buzzigoli G, Bonadonna R, et al. Insulin resistance in essential hypertension. N Engl J Med 1987;317:350 7. 21. Lissner L, Bengtsson C, Lapidus L, Kristjansso n K, Wedel H. Fasting insulin in relation to subsequent blood pressure changes and hypertension in women. Hypertension 1992;20:797 801.

PAGE 216

216 22. Bonora E, Capaldo B, Perin PC, et al. Hyperinsulinemia and insulin resistance are independently associated with plas ma lipids, uric acid and blood pressure in non diabetic subjects. The GISIR database. Nutr Metab Cardiovasc Dis 2008;18:624 31. 23. Elliott WJ, Meyer PM. Incident diabetes in clinical trials of antihypertensive drugs: a network meta analysis. Lancet 2007;3 69:201 7. 24. Dahlof B, Devereux RB, Kjeldsen SE, et al. Cardiovascular morbidity and mortality in the Losartan Intervention For Endpoint reduction in hypertension study (LIFE): a randomised trial against atenolol. Lancet 2002;359:995 1003. 25. Bangalore S Parkar S, Grossman E, Messerli FH. A meta analysis of 94,492 patients with hypertension treated with beta blockers to determine the risk of new onset diabetes mellitus. Am J Cardiol 2007;100:1254 62. 26. Dahlof B, Sever PS, Poulter NR, et al. Prevention of cardiovascular events with an antihypertensive regimen of amlodipine adding perindopril as required versus atenolol adding bendroflumethiazide as required, in the Anglo Scandinavian Cardiac Outcomes Trial Blood Pressure Lowering Arm (ASCOT BPLA): a mult icentre randomised controlled trial. Lancet 2005;366:895 906. 27. Carter BL, Einhorn PT, Brands M, et al. Thiazide induced dysglycemia: call for research from a working group from the national heart, lung, and blood institute. Hypertension 2008;52:30 6. 28 Morgan DB, Davidson C. Hypokalaemia and diuretics: an analysis of publications. Br Med J 1980;280:905 8. 29. Sandstrom PE. Inhibition by hydrochlorothiazide of insulin release and calcium influx in mouse pancreatic beta cells. Br J Pharmacol 1993;110:135 9 62. 30. Zillich AJ, Garg J, Basu S, Bakris GL, Carter BL. Thiazide diuretics, potassium, and the development of diabetes: a quantitative review. Hypertension 2006;48:219 24. 31. Duarte JD, Cooper DeHoff RM. Mechanisms for blood pressure lowering and meta bolic effects of thiazide and thiazide like diuretics. Expert Rev Cardiovasc Ther 2010;8:793 802. 32. Zhou MS, Schulman IH, Jaimes EA, Raij L. Thiazide diuretics, endothelial function, and vascular oxidative stress. J Hypertens 2008;26:494 500. 33. Langfor d HG, Cutter G, Oberman A, Kansal P, Russell G. The effect of thiazide therapy on glucose, insulin and cholesterol metabolism and of glucose on potassium: results of a cross sectional study in patients from the Hypertension Detection and Follow up Program. J Hum Hypertens 1990;4:491 500.

PAGE 217

217 34. Wilkins RW. New drugs for the treatment of hypertension. Ann Intern Med 1959;50:1 10. 35. Rapoport MI, Hurd HF. Thiazide induced glucose intolerance treated with potassium. Arch Intern Med 1964;113:405 8. 36. Goldner MG Zarowitz H, Akgun S. Hyperglycemia and glycosuria due to thiazide derivatives administered in diabetes mellitus. N Engl J Med 1960;262:403 5. 37. Shafi T, Appel LJ, Miller ER, Klag MJ, Parekh RS. Changes in serum potassium mediate thiazide induced diabet es. Hypertension 2008;52:1022 9. 38. Pickkers P, Schachter M, Hughes AD, Feher MD, Sever PS. Thiazide induced hyperglycaemia: a role for calcium activated potassium channels? Diabetologia 1996;39:861 4. 39. Calder JA, Schachter M, Sever PS. Potassium chann el opening properties of thiazide diuretics in isolated guinea pig resistance arteries. J Cardiovasc Pharmacol 1994;24:158 64. 40. Helderman JH, Elahi D, Andersen DK, et al. Prevention of the glucose intolerance of thiazide diuretics by maintenance of body potassium. Diabetes 1983;32:106 11. 41. Luna B, Feinglos MN. Drug induced hyperglycemia. JAMA 2001;286:1945 8. 42. Sagild U, Andersen V, Andreasen PB. Glucose tolerance and insulin responsiveness in experimental potassium depletion. Acta Med Scand 1961;16 9:243 51. 43. Johnson JA, Boerwinkle E, Zineh I, et al. Pharmacogenomics of antihypertensive drugs: rationale and design of the Pharmacogenomic Evaluation of Antihypertensive Responses (PEAR) study. Am Heart J 2009;157:442 9. 44. Smith SM, Anderson SD, Wen S, et al. Lack of correlation between thiazide induced hyperglycemia and hypokalemia: subgroup analysis of results from the pharmacogenomic evaluation of antihypertensive responses (PEAR) study. Pharmacotherapy 2009;29:1157 65. 45. DeFronzo RA, Bia M, Bir khead G. Epinephrine and potassium homeostasis. Kidney Int 1981;20:83 91. 46. Barzilay JI, Davis BR, Cutler JA, et al. Fasting glucose levels and incident diabetes mellitus in older nondiabetic adults randomized to receive 3 different classes of antihypert ensive treatment: a report from the Antihypertensive and Lipid Lowering Treatment to Prevent Heart Attack Trial (ALLHAT). Arch Intern Med 2006;166:2191 201.

PAGE 218

218 47. Wright JT, Jr., Probstfield JL, Cushman WC, et al. ALLHAT findings revisited in the context of subsequent analyses, other trials, and meta analyses. Arch Intern Med 2009;169:832 42. 48. Verdecchia P, Reboldi G, Angeli F, et al. Adverse prognostic significance of new diabetes in treated hypertensive subjects. Hypertension 2004;43:963 9. 49. Aksnes TA Kjeldsen SE, Rostrup M, Omvik P, Hua TA, Julius S. Impact of new onset diabetes mellitus on cardiac outcomes in the Valsartan Antihypertensive Long term Use Evaluation (VALUE) trial population. Hypertension 2007;50:467 73. 50. Padwal R, Laupacis A. Antih ypertensive therapy and incidence of type 2 diabetes: a systematic review. Diabetes Care 2004;27:247 55. 51. Wicklmayr M, Rett K, Dietze G, Mehnert H. Effects of beta blocking agents on insulin secretion and glucose disposal. Horm Metab Res Suppl 1990;22:2 9 33. 52. Austin C, Chess Williams R. The in vitro effects of insulin and the effects of acute fasting on cardiac beta adrenoceptor responses in the short term streptozotocin diabetic rat. J Pharm Pharmacol 1994;46:326 31. 53. Manrique C, Giles TD, Ferdina nd KC, Sowers JR. Realities of newer beta blockers for the management of hypertension. J Clin Hypertens (Greenwich) 2009;11:369 75. 54. Bakris GL, Fonseca V, Katholi RE, et al. Metabolic effects of carvedilol vs metoprolol in patients with type 2 diabetes mellitus and hypertension: a randomized controlled trial. JAMA 2004;292:2227 36. 55. Sharma AM, Pischon T, Hardt S, Kunz I, Luft FC. Hypothesis: Beta adrenergic receptor blockers and weight gain: A systematic analysis. Hypertension 2001;37:250 4. 56. Polla re T, Lithell H, Selinus I, Berne C. Sensitivity to insulin during treatment with atenolol and metoprolol: a randomised, double blind study of effects on carbohydrate and lipoprotein metabolism in hypertensive patients. BMJ 1989;298:1152 7. 57. Macfarlane DP, Paterson KR, Fisher M. Cardiovascular drugs as antidiabetic agents: evidence for the prevention of type 2 diabetes. Diabetes Obes Metab 2008;10:533 44. 58. Scheen AJ. Renin angiotensin system inhibition prevents type 2 diabetes mellitus. Part 1. A meta analysis of randomised clinical trials. Diabetes Metab 2004;30:487 96.

PAGE 219

219 59. Abuissa H, Jones PG, Marso SP, O'Keefe JH. Angiotensin converting enzyme inhibitors or angiotensin receptor blockers for prevention of type 2 diabetes: a meta analysis of randomize d clinical trials. J Am Coll Cardiol 2005;46:821 6. 60. Gillespie EL, White CM, Kardas M, Lindberg M, Coleman CI. The impact of ACE inhibitors or angiotensin II type 1 receptor blockers on the development of new onset type 2 diabetes. Diabetes Care 2005;28 :2261 6. 61. Tocci G, Paneni F, Palano F, et al. Angiotensin Converting Enzyme Inhibitors, Angiotensin II Receptor Blockers and Diabetes: A Meta Analysis of Placebo Controlled Clinical Trials. Am J Hypertens 2011. 62. Bosch J, Yusuf S, Gerstein HC, et al. Effect of ramipril on the incidence of diabetes. N Engl J Med 2006;355:1551 62. 63. McMurray JJ, Holman RR, Haffner SM, et al. Effect of valsartan on the incidence of diabetes and cardiovascular events. N Engl J Med 2010;362:1477 90. 64. Baron AD, Brechtel G, Wallace P, Edelman SV. Rates and tissue sites of non insulin and insulin mediated glucose uptake in humans. Am J Physiol 1988;255:E769 74. 65. Cooper DeHoff RM, Pacanowski MA, Pepine CJ. Cardiovascular therapies and associated glucose homeostasis: imp lications across the dysglycemia continuum. J Am Coll Cardiol 2009;53:S28 34. 66. Opie LH, Schall R. Old antihypertensives and new diabetes. J Hypertens 2004;22:1453 8. 67. Standards of medical care in diabetes -2012. Diabetes Care 2012;35 Suppl 1:S11 63. 68. Retnakaran R, Qi Y, Sermer M, Connelly PW, Hanley AJ, Zinman B. Glucose intolerance in pregnancy and future risk of pre diabetes or diabetes. Diabetes Care 2008;31:2026 31. 69. Lyssenko V, Almgren P, Anevski D, et al. Predictors of and longitudinal cha nges in insulin sensitivity and secretion preceding onset of type 2 diabetes. Diabetes 2005;54:166 74. 70. Hanley AJ, Karter AJ, Williams K, et al. Prediction of type 2 diabetes mellitus with alternative definitions of the metabolic syndrome: the Insulin R esistance Atherosclerosis Study. Circulation 2005;112:3713 21. 71. Wilson PW, D'Agostino RB, Parise H, Sullivan L, Meigs JB. Metabolic syndrome as a precursor of cardiovascular disease and type 2 diabetes mellitus. Circulation 2005;112:3066 72.

PAGE 220

220 72. Borai A Livingstone C, Ferns GA. The biochemical assessment of insulin resistance. Ann Clin Biochem 2007;44:324 42. 73. Schmoelzer I, Wascher TC. Effect of repaglinide on endothelial dysfunction during a glucose tolerance test in subjects with impaired glucose t olerance. Cardiovasc Diabetol 2006;5:9. 74. Su Y, Liu XM, Sun YM, Wang YY, Luan Y, Wu Y. Endothelial dysfunction in impaired fasting glycemia, impaired glucose tolerance, and type 2 diabetes mellitus. Am J Cardiol 2008;102:497 8. 75. Lorenzo C, Haffner SM, Stanckov A, Laakso M. Relation of direct and surrogate measures of insulin resistance to cardiovascular risk factors in nondiabetic finnish offspring of type 2 diabetic individuals. J Clin Endocrinol Metab 2010;95:5082 90. 76. Matthews DR, Hosker JP, Ru denski AS, Naylor BA, Treacher DF, Turner RC. Homeostasis model assessment: insulin resistance and beta cell function from fasting plasma glucose and insulin concentrations in man. Diabetologia 1985;28:412 9. 77. Rallidis LS, Lekakis J, Kolomvotsou A, et a l. Close adherence to a Mediterranean diet improves endothelial function in subjects with abdominal obesity. Am J Clin Nutr 2009;90:263 8. 78. Schauer PR, Kashyap SR, Wolski K, et al. Bariatric surgery versus intensive medical therapy in obese patients wit h diabetes. N Engl J Med 2012;366:1567 76. 79. Sathyapalan T, Kilpatrick ES, Coady AM, Atkin SL. The effect of atorvastatin in patients with polycystic ovary syndrome: a randomized double blind placebo controlled study. J Clin Endocrinol Metab 2009;94:103 8. 80. Messerli FH, Bell DS, Fonseca V, et al. Body weight changes with beta blocker use: results from GEMINI. Am J Med 2007;120:610 5. 81. Olufadi R, Byrne CD. Clinical and laboratory diagnosis of the metabolic syndrome. J Clin Pathol 2008;61:697 706. 82. Katz A, Nambi SS, Mather K, et al. Quantitative insulin sensitivity check index: a simple, accurate method for assessing insulin sensitivity in humans. J Clin Endocrinol Metab 2000;85:2402 10. 83. Matsuda M, DeFronzo RA. Insulin sensitivity indices obtain ed from oral glucose tolerance testing: comparison with the euglycemic insulin clamp. Diabetes Care 1999;22:1462 70. 84. Cooper DeHoff RM, Wen S, Beitelshees AL, et al. Impact of abdominal obesity on incidence of adverse metabolic effects associated with a ntihypertensive medications. Hypertension 2010;55:61 8.

PAGE 221

221 85. Ford ES, Li C, Sattar N. Metabolic syndrome and incident diabetes: current state of the evidence. Diabetes Care 2008;31:1898 904. 86. Daly CA, Hildebrandt P, Bertrand M, et al. Adverse prognosis a ssociated with the metabolic syndrome in established coronary artery disease: data from the EUROPA trial. Heart 2007;93:1406 11. 87. Huptas S, Geiss HC, Otto C, Parhofer KG. Effect of atorvastatin (10 mg/day) on glucose metabolism in patients with the meta bolic syndrome. Am J Cardiol 2006;98:66 9. 88. Voight BF, Scott LJ, Steinthorsdottir V, et al. Twelve type 2 diabetes susceptibility loci identified through large scale association analysis. Nat Genet 2010;42:579 89. 89. Rung J, Cauchi S, Albrechtsen A, et al. Genetic variant near IRS1 is associated with type 2 diabetes, insulin resistance and hyperinsulinemia. Nat Genet 2009;41:1110 5. 90. Dupuis J, Langenberg C, Prokopenko I, et al. New genetic loci implicated in fasting glucose homeostasis and their impa ct on type 2 diabetes risk. Nat Genet 2010;42:105 16. 91. Beck Nielsen H, Vaag A, Poulsen P, Gaster M. Metabolic and genetic influence on glucose metabolism in type 2 diabetic subjects -experiences from relatives and twin studies. Best Pract Res Clin Endoc rinol Metab 2003;17:445 67. 92. Poulsen P, Kyvik KO, Vaag A, Beck Nielsen H. Heritability of type II (non insulin dependent) diabetes mellitus and abnormal glucose tolerance -a population based twin study. Diabetologia 1999;42:139 45. 93. Kaprio J, Tuomile hto J, Koskenvuo M, et al. Concordance for type 1 (insulin dependent) and type 2 (non insulin dependent) diabetes mellitus in a population based cohort of twins in Finland. Diabetologia 1992;35:1060 7. 94. Genome wide association study of 14,000 cases of s even common diseases and 3,000 shared controls. Nature 2007;447:661 78. 95. Maeda S, Osawa N, Hayashi T, Tsukada S, Kobayashi M, Kikkawa R. Genetic variations associated with diabetic nephropathy and type II diabetes in a Japanese population. Kidney Int Su ppl 2007:S43 8. 96. Salonen JT, Uimari P, Aalto JM, et al. Type 2 diabetes whole genome association study in four populations: the DiaGen consortium. Am J Hum Genet 2007;81:338 45. 97. Scott LJ, Mohlke KL, Bonnycastle LL, et al. A genome wide association s tudy of type 2 diabetes in Finns detects multiple susceptibility variants. Science 2007;316:1341 5.

PAGE 222

222 98. Sladek R, Rocheleau G, Rung J, et al. A genome wide association study identifies novel risk loci for type 2 diabetes. Nature 2007;445:881 5. 99. Billing s LK, Florez JC. The genetics of type 2 diabetes: what have we learned from GWAS? Ann N Y Acad Sci 2010;1212:59 77. 100. Egefjord L, Jensen JL, Bang Berthelsen CH, et al. Zinc transporter gene expression is regulated by pro inflammatory cytokines: a potent ial role for zinc transporters in beta cell apoptosis? BMC Endocr Disord 2009;9:7. 101. Prokopenko I, Langenberg C, Florez JC, et al. Variants in MTNR1B influence fasting glucose levels. Nat Genet 2009;41:77 81. 102. Bouatia Naji N, Bonnefond A, Cavalcanti Proena C, et al. A variant near MTNR1B is associated with increased fasting plasma glucose levels and type 2 diabetes risk. Nat Genet 2009;41:89 94. 103. Lyssenko V, Nagorny CL, Erdos MR, et al. Common variant in MTNR1B associated with increased risk of type 2 diabetes and impaired early insulin secretion. Nat Genet 2009;41:82 8. 104. Yasuda K, Miyake K, Horikawa Y, et al. Variants in KCNQ1 are associated with susceptibility to type 2 diabetes mellitus. Nat Genet 2008;40:1092 7. 105. Unoki H, Takahashi A, Kawaguchi T, et al. SNPs in KCNQ1 are associated with susceptibility to type 2 diabetes in East Asian and European populations. Nat Genet 2008;40:1098 102. 106. Zhou K, Bellenguez C, Spencer CC, et al. Common variants near ATM are associated with glycemic response to metformin in type 2 diabetes. Nat Genet;43:117 20. 107. Shu Y, Sheardown SA, Brown C, et al. Effect of genetic variation in the organic cation transporter 1 (OCT1) on metformin action. J Clin Invest 2007;117:1422 31. 108. Wolford JK, Yeatts KA Dhanjal SK, et al. Sequence variation in PPARG may underlie differential response to troglitazone. Diabetes 2005;54:3319 25. 109. Geese WJ, Achanzar W, Rubin C, et al. Genetic and gene expression studies implicate renin and endothelin 1 in edema caused b y peroxisome proliferator activated receptor gamma agonists. Pharmacogenet Genomics 2008;18:903 10. 110. Spraggs C, McCarthy A, McCarthy L, et al. Genetic variants in the epithelial sodium channel associate with oedema in type 2 diabetic patients receiving the peroxisome proliferator activated receptor gamma agonist farglitazar. Pharmacogenet Genomics 2007;17:1065 76.

PAGE 223

223 111. Maitland van der Zee AH, Turner ST, Schwartz GL, Chapman AB, Klungel OH, Boerwinkle E. Demographic, environmental, and genetic predictor s of metabolic side effects of hydrochlorothiazide treatment in hypertensive subjects. Am J Hypertens 2005;18:1077 83. 112. Irvin MR, Lynch AI, Kabagambe EK, et al. Pharmacogenetic association of hypertension candidate genes with fasting glucose in the Gen HAT Study. J Hypertens 2010;28:2076 83. 113. Roden DM, Altman RB, Benowitz NL, et al. Pharmacogenomics: challenges and opportunities. Ann Intern Med 2006;145:749 57. 114. Woodcock J. The prospects for "personalized medicine" in drug development and drug th erapy. Clin Pharmacol Ther 2007;81:164 9. 115. Reitman ML, Schadt EE. Pharmacogenetics of metformin response: a step in the path toward personalized medicine. J Clin Invest 2007;117:1226 9. 116. Sun B, Williams JS, Svetkey LP, Kolatkar NS, Conlin PR. Beta2 adrenergic receptor genotype affects the renin angiotensin aldosterone system response to the Dietary Approaches to Stop Hypertension (DASH) dietary pattern. Am J Clin Nutr 2010;92:444 9. 117. Pacanowski MA, Zineh I, Cooper Dehoff RM, Pepine CJ, Johnson J A. Genetic and pharmacogenetic associations between NOS3 polymorphisms, blood pressure, and cardiovascular events in hypertension. Am J Hypertens 2009;22:748 53. 118. Brunner M, Cooper DeHoff RM, Gong Y, et al. Factors influencing blood pressure response t o trandolapril add on therapy in patients taking verapamil SR (from the International Verapamil SR/Trandolapril [INVEST] Study). Am J Cardiol 2007;99:1549 54. 119. Levy H, Young JH. Perspectives from the clinic: will the average physician embrace personali zed medicine? Clin Pharmacol Ther 2008;83:492 3. 120. Konrady AO, Kasherininov YR, Shavarov AA, et al. How can we block sympathetic overactivity? Effects of rilmenidine and atenolol in overweight hypertensive patients. J Hum Hypertens 2006;20:398 406. 121. Lindholm LH, Persson M, Alaupovic P, Carlberg B, Svensson A, Samuelsson O. Metabolic outcome during 1 year in newly detected hypertensives: results of the Antihypertensive Treatment and Lipid Profile in a North of Sweden Efficacy Evaluation (ALPINE study) J Hypertens 2003;21:1563 74. 122. Maritz FJ, Weich HF, Schoeman HS. A comparison of the metabolic effects of captopril and atenolol on glucose, insulin and lipoproteins in patients with mild to moderate essential hypertension. S Afr Med J 1995;85:1342 5.

PAGE 224

224 123. Bakris G, Molitch M, Hewkin A, et al. Differences in glucose tolerance between fixed dose antihypertensive drug combinations in people with metabolic syndrome. Diabetes Care 2006;29:2592 7. 124. LaSalle JR. Management of Type 2 Diabetes: Focus on the Thiazolidinediones. Hospital Physician 2005;46:37 42. 125. Nolte Kennedy MS. Chapter 41. Pancreatic Hormones & Antidiabetic Drugs. In: Katzung BG, Masters SB, Trevor AJ, eds. Basic & Clinical Pharmacology. 12nd ed. New York: McGraw Hill; 2012. http://www.accesspharmacy.com/content.aspx?aID=55828662 Accessed August 9, 2012. 126. Benowitz NL. Chapter 11. Antihypertensive Agents. In: Katzung BG, Masters SB, Trevor AJ, eds. Basic & Clinic al Pharmacology. 12nd ed. New York: McGraw Hill; 2012. http://www.accesspharmacy.com/content.aspx?aID=55821618 Accessed August 9, 2012. 127. Manrique C, Johnson M, Sowers JR. Thiazide diuretics alone or with beta blockers impair glucose metabolism in hypertensive patients with abdominal obesity. Hypertension 2010;55:15 7. 128. Definition and diagnosis of diabetes mellitus and intermediate hyperglycemia: report of a WHO/IDF consultation 2006. Geneva, Switzerland. World Health Organization and International Diabetes Federation, 2006. 129. Laboratory diagnosis and monitoring of diabetes mellitus, 2002. Geneva, Switzerland. World Health Organization, 2002. 130. National Institute of Diabet es and Digestive and Kidney Diseases. Diabetes dictionary, 2012. Bethesda, MD. The National Diabetes Information Clearinghouse, National Institute of Diabetes and Digestive and Kidney Diseases, 2012. http://diabetes.niddk.nih.gov/dm/pubs/dictionary/ 131. Gupta AK, Dahlof B, Dobson J, Sever PS, Wedel H, Poulter NR. Determinants of new onset diabetes among 19,257 hypertensive patients randomized in the Anglo Scandinavian Cardiac Outcomes Trial -Blood Pressure Lowering Arm and the relative influence of antihypertensive medication. Diabetes Care 2008;31:982 8. 132. Drzewoski J, Czupryniak L. Concordance between fasting and 2 h post glucose challenge criteria for the diagnosis of diabetes mellitus and glucose intolerance in high risk individuals. Diabet Med 2001;18:29 31. 133. Abdul Ghani MA, Tripathy D, DeFronzo RA. Contributions of beta cell dysfunction and insulin resistance to the pathogenesis of impaired glucose tolerance and impaired fasting glucose. Diabetes Care 2006;29:1130 9.

PAGE 225

225 134. DECODE Study Group ErDEG. Is the current definition for diabetes relevant to mortality risk from all causes and cardiovascular and noncardiovascular diseases? Diabetes Care 2003;26:688 96. 135. Coutinho M, Gerste in HC, Wang Y, Yusuf S. The relationship between glucose and incident cardiovascular events. A metaregression analysis of published data from 20 studies of 95,783 individuals followed for 12.4 years. Diabetes Care 1999;22:233 40. 136. DECODE Study Group tE DEG. Glucose tolerance and cardiovascular mortality: comparison of fasting and 2 hour diagnostic criteria. Arch Intern Med 2001;161:397 405. 137. Ceriello A, Quagliaro L, Piconi L, et al. Effect of postprandial hypertriglyceridemia and hyperglycemia on cir culating adhesion molecules and oxidative stress generation and the possible role of simvastatin treatment. Diabetes 2004;53:701 10. 138. Esposito K, Ciotola M, Carleo D, et al. Post meal glucose peaks at home associate with carotid intima media thickness in type 2 diabetes. J Clin Endocrinol Metab 2008;93:1345 50. 139. Ceriello A, Esposito K, Piconi L, et al. Glucose "peak" and glucose "spike": Impact on endothelial function and oxidative stress. Diabetes Res Clin Pract 2008;82:262 7. 140. Kennelly PJ, Rod welxcl VW. Chapter 7. Enzymes: Mechanism of Action. In: Murray RK, Kennelly PJ, Rodwell VW, Botham KM, Bender DA, Weil PA, eds. Harper's Illustrated Biochemistry. 29th ed. New York: McGraw Hill; 2011. http://www.accesspharmacy.com/content.aspx?aID=55881558 Accessed July 1, 2012. 141. Rainey PM. Chapter 6. Laboratory Principles. In: Rainey PM, ed. Goldfrank's Toxicologic Emergencies. 9th ed. New York: McGraw Hill; 2011. http://www.accesspharmacy.com.lp.hscl.ufl.edu/content.aspx?aID=6502905 Accessed July 1, 2012. 142. Beal SL. Ways to fit a PK model with some data below the quantification limit. J Phar macokinet Pharmacodyn 2001;28:481 504. 143. Le MT, Frye RF, Rivard CJ, et al. Effects of high fructose corn syrup and sucrose on the pharmacokinetics of fructose and acute metabolic and hemodynamic responses in healthy subjects. Metabolism 2011. 144. Eriks son JW, Jansson PA, Carlberg B, et al. Hydrochlorothiazide, but not Candesartan, aggravates insulin resistance and causes visceral and hepatic fat accumulation: the mechanisms for the diabetes preventing effect of Candesartan (MEDICA) Study. Hypertension 2 008;52:1030 7.

PAGE 226

226 145. Poirier L, Cleroux J, Nadeau A, Lacourciere Y. Effects of nebivolol and atenolol on insulin sensitivity and haemodynamics in hypertensive patients. J Hypertens 2001;19:1429 35. 146. Bonner G, Schmieder R, Chrosch R, Weidinger G. Effect of bunazosin and atenolol on glucose metabolism in obese, nondiabetic patients with primary hypertension. Cardiovasc Drugs Ther 1997;11:21 6. 147. Navare HA, Frye RF, Cooper Dehoff RM, et al. Atenolol exposure and risk for development of adverse metabolic effects: a pilot study. Pharmacotherapy 2010;30:872 8. 148. Catalano PM, Avallone DA, Drago NM, Amini SB. Reproducibility of the oral glucose tolerance test in pregnant women. Am J Obstet Gynecol 1993;169:874 81. 149. Ganda OP, Day JL, Soeldner JS, Connon JJ, Gleason RE. Reproducibility and comparative analysis of repeated intravenous and oral glucose tolerance tests. Diabetes 1978;27:715 25. 150. Stump CS, Hamilton MT, Sowers JR. Effect of antihypertensive agents on the development of type 2 diabetes melli tus. Mayo Clin Proc 2006;81:796 806. 151. Cooper DeHoff RM. Thiazide induced dysglycemia: it's time to take notice. Expert Rev Cardiovasc Ther 2008;6:1291 4. 152. Chrysant SG, Chrysant GS, Dimas B. Current and future status of beta blockers in the treatmen t of hypertension. Clin Cardiol 2008;31:249 52. 153. Almgren T, Wilhelmsen L, Samuelsson O, Himmelmann A, Rosengren A, Andersson OK. Diabetes in treated hypertension is common and carries a high cardiovascular risk: results from a 28 year follow up. J Hype rtens 2007;25:1311 7. 154. Dunder K, Lind L, Zethelius B, Berglund L, Lithell H. Increase in blood glucose concentration during antihypertensive treatment as a predictor of myocardial infarction: population based cohort study. BMJ 2003;326:681. 155. Levey AS, Bosch JP, Lewis JB, Greene T, Rogers N, Roth D. A more accurate method to estimate glomerular filtration rate from serum creatinine: a new prediction equation. Modification of Diet in Renal Disease Study Group. Ann Intern Med 1999;130:461 70. 156. DuBo is D, EF D. A formula to estimate the approximate surface area if height and weight be known. Arch Intern Med 1916;17:863 71. 157. Wang Y, Moss J, Thisted R. Predictors of body surface area. J Clin Anesth 1992;4:4 10.

PAGE 227

227 158. Sepe A, Tchkonia T, Thomou T, Zam boni M, Kirkland JL. Aging and regional differences in fat cell progenitors a mini review. Gerontology 2011;57:66 75. 159. Poledne R, Lorenzova A, Stavek P, et al. Proinflammatory status, genetics and atherosclerosis. Physiol Res 2009;58 Suppl 2:S111 8. 160. Boden G, Shulman GI. Free fatty acids in obesity and type 2 diabetes: defining their role in the development of insulin resistance and beta cell dysfunction. Eur J Clin Invest 2002;32 Suppl 3:14 23. 161. Lewis GF, Carpentier A, Adeli K, Giacca A. Diso rdered fat storage and mobilization in the pathogenesis of insulin resistance and type 2 diabetes. Endocr Rev 2002;23:201 29. 162. Bjorntorp P. Metabolic implications of body fat distribution. Diabetes Care 1991;14:1132 43. 163. Berg AH, Scherer PE. Adipos e tissue, inflammation, and cardiovascular disease. Circ Res 2005;96:939 49. 164. Schneider HJ, Friedrich N, Klotsche J, et al. The predictive value of different measures of obesity for incident cardiovascular events and mortality. J Clin Endocrinol Metab 2010;95:1777 85. 165. Ghandehari H, Le V, Kamal Bahl S, Bassin SL, Wong ND. Abdominal obesity and the spectrum of global cardiometabolic risks in US adults. Int J Obes (Lond) 2009;33:239 48. 166. Yusuf S, Hawken S, Ounpuu S, et al. Obesity and the risk of myocardial infarction in 27,000 participants from 52 countries: a case control study. Lancet 2005;366:1640 9. 167. Dagenais GR, Yi Q, Mann JF, Bosch J, Pogue J, Yusuf S. Prognostic impact of body weight and abdominal obesity in women and men with cardiovas cular disease. Am Heart J 2005;149:54 60. 168. Leitzmann MF, Moore SC, Koster A, et al. Waist circumference as compared with body mass index in predicting mortality from specific causes. PLoS One 2011;6:e18582. 169. Testa G, Cacciatore F, Galizia G, et al. Waist circumference but not body mass index predicts long term mortality in elderly subjects with chronic heart failure. J Am Geriatr Soc 2010;58:1433 40. 170. Kirchheiner J, Roots I, Goldammer M, Rosenkranz B, Brockmoller J. Effect of genetic polymorphis ms in cytochrome p450 (CYP) 2C9 and CYP2C8 on the pharmacokinetics of oral antidiabetic drugs: clinical relevance. Clin Pharmacokinet 2005;44:1209 25.

PAGE 228

228 171. Makino H, Shimizu I, Murao S, et al. A pilot study suggests that the G/G genotype of resistin single nucleotide polymorphism at 420 may be an independent predictor of a reduction in fasting plasma glucose and insulin resistance by pioglitazone in type 2 diabetes. Endocr J 2009;56:1049 58. 172. Yu W, Hu C, Zhang R, et al. Effects of KCNQ1 polymorphisms o n the therapeutic efficacy of oral antidiabetic drugs in Chinese patients with type 2 diabetes. Clin Pharmacol Ther 2011;89:437 42. 173. Schroner Z, Dobrikova M, Klimcakova L, et al. Variation in KCNQ1 is associated with therapeutic response to sulphonylur eas. Med Sci Monit 2011;17:CR392 6. 174. Johnson JA, Zineh I, Puckett BJ, McGorray SP, Yarandi HN, Pauly DF. Beta 1 adrenergic receptor polymorphisms and antihypertensive response to metoprolol. Clin Pharmacol Ther 2003;74:44 52. 175. van Rijn Bikker PC, M airuhu G, van Montfrans GA, et al. Genetic factors are relevant and independent determinants of antihypertensive drug effects in a multiracial population. Am J Hypertens 2009;22:1295 302. 176. Liu J, Liu ZQ, Yu BN, et al. beta1 Adrenergic receptor polymorp hisms influence the response to metoprolol monotherapy in patients with essential hypertension. Clin Pharmacol Ther 2006;80:23 32. 177. Ehret GB, Munroe PB, Rice KM, et al. Genetic variants in novel pathways influence blood pressure and cardiovascular dise ase risk. Nature 2011;478:103 9. 178. Levy D, Ehret GB, Rice K, et al. Genome wide association study of blood pressure and hypertension. Nat Genet 2009;41:677 87. 179. Niu Y, Gong Y, Langaee TY, et al. Genetic variation in the beta2 subunit of the voltage gated calcium channel and pharmacogenetic association with adverse cardiovascular outcomes in the INternational VErapamil SR Trandolapril STudy GENEtic Substudy (INVEST GENES). Circ Cardiovasc Genet 2010;3:548 55. 180. Pacanowski MA, Gong Y, Cooper Dehoff RM, et al. beta adrenergic receptor gene polymorphisms and beta blocker treatment outcomes in hypertension. Clin Pharmacol Ther 2008;84:715 21. 181. Cresci S, Kelly RJ, Cappola TP, et al. Clinical and genetic modifiers of long term survival in heart failur e. J Am Coll Cardiol 2009;54:432 44. 182. Dorn Ii GW, Liggett SB. Pharmacogenomics of beta adrenergic receptors and their accessory signaling proteins in heart failure. Clin Transl Sci 2008;1:255 62. 183. Dionne IJ, Garant MJ, Nolan AA, et al. Association between obesity and a polymorphism in the beta(1) adrenoceptor gene (Gly389Arg ADRB1) in Caucasian women. Int J Obes Relat Metab Disord 2002;26:633 9.

PAGE 229

229 184. Ellsworth DL, Coady SA, Chen W, Srinivasan SR, Boerwinkle E, Berenson GS. Interactive effects betwee n polymorphisms in the beta adrenergic receptors and longitudinal changes in obesity. Obes Res 2005;13:519 26. 185. Lanfear DE, Jones PG, Marsh S, Cresci S, McLeod HL, Spertus JA. Beta2 adrenergic receptor genotype and survival among patients receiving bet a blocker therapy after an acute coronary syndrome. JAMA 2005;294:1526 33. 186. Mialet Perez J, Rathz DA, Petrashevskaya NN, et al. Beta 1 adrenergic receptor polymorphisms confer differential function and predisposition to heart failure. Nat Med 2003;9:13 00 5. 187. Wagoner LE, Craft LL, Singh B, et al. Polymorphisms of the beta(2) adrenergic receptor determine exercise capacity in patients with heart failure. Circ Res 2000;86:834 40. 188. Enriquez R, Adam V, Sirvent AE, Garcia Garcia AB, Millan I, Amoros F Gitelman syndrome due to p.A204T mutation in CLCNKB gene. Int Urol Nephrol 2010;42:1099 102. 189. Monette MY, Rinehart J, Lifton RP, Forbush B. Rare mutations in the human Na K Cl cotransporter (NKCC2) associated with lower blood pressure exhibit impaire d processing and transport function. Am J Physiol Renal Physiol 2011;300:F840 7. 190. Tsutsui H, Hamano T, Kawaura Y, et al. A case of Gitelman syndrome associated with idiopathic intracranial hypertension. Intern Med 2011;50:1493 6. 191. Balavoine AS, Bat aille P, Vanhille P, et al. Phenotype genotype correlation and follow up in adult patients with hypokalaemia of renal origin suggesting Gitelman syndrome. Eur J Endocrinol 2011;165:665 73. 192. Kamdem LK, Hamilton L, Cheng C, et al. Genetic predictors of g lucocorticoid induced hypertension in children with acute lymphoblastic leukemia. Pharmacogenet Genomics 2008;18:507 14. 193. Turner ST, Bailey KR, Schwartz GL, Chapman AB, Chai HS, Boerwinkle E. Genomic Association Analysis Identifies Multiple Loci Influe ncing Antihypertensive Response to an Angiotensin II Receptor Blocker. Hypertension 2012;59:1204 11. 194. Busst CJ, Bloomer LD, Scurrah KJ, et al. The epithelial sodium channel gamma subunit gene and blood pressure: family based association, renal gene exp ression, and physiological analyses. Hypertension 2011;58:1073 8. 195. Zhao Q, Gu D, Hixson JE, et al. Common variants in epithelial sodium channel genes contribute to salt sensitivity of blood pressure: The GenSalt study. Circ Cardiovasc Genet 2011;4:375 80.

PAGE 230

230 196. Brunton LL, Lazo JS, Parker KL. Goodman & Gilman's The Pharmacological Basis of Therapeutics. In: The McGraw Hill Companies Inc; 2006. 197. Kanehisa M, Araki M, Goto S, et al. KEGG for linking genomes to life and the environment. Nucleic Acids Res 2008;36:D480 4. 198. Keating BJ, Tischfield S, Murray SS, et al. Concept, design and implementation of a cardiovascular gene centric 50 k SNP array for large scale genomic association studies. PLoS One 2008;3:e3583. 199. Conde L, Vaquerizas JM, Dopazo H, et al. PupaSuite: finding functional single nucleotide polymorphisms for large scale genotyping purposes. Nucleic Acids Res 2006;34:W621 5. 200. Yuan HY, Chiou JJ, Tseng WH, et al. FASTSNP: an always up to date and extendable service for SNP function analy sis and prioritization. Nucleic Acids Res 2006;34:W635 41. 201. Lee PH, Shatkay H. An integrative scoring system for ranking SNPs by their potential deleterious effects. Bioinformatics 2009;25:1048 55. 202. Purcell S, Neale B, Todd Brown K, et al. PLINK: a tool set for whole genome association and population based linkage analyses. Am J Hum Genet 2007;81:559 75. 203. Price AL, Patterson NJ, Plenge RM, Weinblatt ME, Shadick NA, Reich D. Principal components analysis corrects for stratification in genome wide association studies. Nat Genet 2006;38:904 9. 204. Freedman ML, Reich D, Penney KL, et al. Assessing the impact of population stratification on genetic association studies. Nat Genet 2004;36:388 93. 205. Newton Cheh C, Hirschhorn JN. Genetic association s tudies of complex traits: design and analysis issues. Mutat Res 2005;573:54 69. 206. Barrett JC, Fry B, Maller J, Daly MJ. Haploview: analysis and visualization of LD and haplotype maps. Bioinformatics 2005;21:263 5. 207. Canessa CM, Schild L, Buell G, et al. Amiloride sensitive epithelial Na+ channel is made of three homologous subunits. Nature 1994;367:463 7. 208. Iwai N, Baba S, Mannami T, et al. Association of sodium channel gamma subunit promoter variant with blood pressure. Hypertension 2001;38:86 9. 209. Tobin MD, Tomaszewski M, Braund PS, et al. Common variants in genes underlying monogenic hypertension and hypotension and blood pressure in the general population. Hypertension 2008;51:1658 64.

PAGE 231

231 210. Hansson JH, Nelson Williams C, Suzuki H, et al. Hype rtension caused by a truncated epithelial sodium channel gamma subunit: genetic heterogeneity of Liddle syndrome. Nat Genet 1995;11:76 82. 211. Strautnieks SS, Thompson RJ, Gardiner RM, Chung E. A novel splice site mutation in the gamma subunit of the epit helial sodium channel gene in three pseudohypoaldosteronism type 1 families. Nat Genet 1996;13:248 50. 212. Wong ZY, Stebbing M, Ellis JA, Lamantia A, Harrap SB. Genetic linkage of beta and gamma subunits of epithelial sodium channel to systolic blood pres sure. Lancet 1999;353:1222 5. 213. Jin HS, Hong KW, Lim JE, et al. Genetic variations in the sodium balance regulating genes ENaC, NEDD4L, NDFIP2 and USP2 influence blood pressure and hypertension. Kidney Blood Press Res 2010;33:15 23. 214. Maitland van de r Zee AH, Turner ST, Schwartz GL, Chapman AB, Klungel OH, Boerwinkle E. A multilocus approach to the antihypertensive pharmacogenetics of hydrochlorothiazide. Pharmacogenet Genomics 2005;15:287 93. 215. Turner ST, Schwartz GL, Chapman AB, Boerwinkle E. WNK 1 kinase polymorphism and blood pressure response to a thiazide diuretic. Hypertension 2005;46:758 65. 216. Qadri YJ, Rooj AK, Fuller CM. ENaCs and ASICs as therapeutic targets. Am J Physiol Cell Physiol 2012;302:C943 65. 217. Garcia Suarez O, Calavia MG, Perez Molto FJ, et al. Immunohistochemical profile of human pancreatic pacinian corpuscles. Pancreas 2010;39:403 10. 218. Novak I, Hansen MR. Where have all the Na+ channels gone? In search of functional ENaC in exocrine pancreas. Biochim Biophys Acta 2002 ;1566:162 8. 219. Wu C, Orozco C, Boyer J, et al. BioGPS: an extensible and customizable portal for querying and organizing gene annotation resources. Genome Biol 2009;10:R130. 220. Kent WJ, Sugnet CW, Furey TS, et al. The human genome browser at UCSC. Gen ome Res 2002;12:996 1006. 221. Jachymova M, Horky K, Bultas J, et al. Association of the Glu298Asp polymorphism in the endothelial nitric oxide synthase gene with essential hypertension resistant to conventional therapy. Biochem Biophys Res Commun 2001;284 :426 30. 222. Casas JP, Bautista LE, Humphries SE, Hingorani AD. Endothelial nitric oxide synthase genotype and ischemic heart disease: meta analysis of 26 studies involving 23028 subjects. Circulation 2004;109:1359 65.

PAGE 232

232 223. Berger K, Stogbauer F, Stoll M, et al. The glu298asp polymorphism in the nitric oxide synthase 3 gene is associated with the risk of ischemic stroke in two large independent case control studies. Hum Genet 2007;121:169 78. 224. Sun XJ, Rothenberg P, Kahn CR, et al. Structure of the insu lin receptor substrate IRS 1 defines a unique signal transduction protein. Nature 1991;352:73 7. 225. Baroni MG, D'Andrea MP, Montali A, et al. A common mutation of the insulin receptor substrate 1 gene is a risk factor for coronary artery disease. Arterio scler Thromb Vasc Biol 1999;19:2975 80. 226. Perticone F, Sciacqua A, Scozzafava A, et al. Impaired endothelial function in never treated hypertensive subjects carrying the Arg972 polymorphism in the insulin receptor substrate 1 gene. J Clin Endocrinol Met ab 2004;89:3606 9. 227. Qi Q, Bray GA, Smith SR, Hu FB, Sacks FM, Qi L. Insulin receptor substrate 1 gene variation modifies insulin resistance response to weight loss diets in a 2 year randomized trial: the Preventing Overweight Using Novel Dietary Strate gies (POUNDS LOST) trial. Circulation 2011;124:563 71. 228. Marini MA, Frontoni S, Mineo D, et al. The Arg972 variant in insulin receptor substrate 1 is associated with an atherogenic profile in offspring of type 2 diabetic patients. J Clin Endocrinol Meta b 2003;88:3368 71. 229. Esposito DL, Mammarella S, Ranieri A, et al. Deletion of Gly723 in the insulin receptor substrate 1 of a patient with noninsulin dependent diabetes mellitus. Hum Mutat 1996;7:364 6. 230. Esposito DL, Li Y, Vanni C, et al. A novel T6 08R missense mutation in insulin receptor substrate 1 identified in a subject with type 2 diabetes impairs metabolic insulin signaling. J Clin Endocrinol Metab 2003;88:1468 75. 231. Drakas R, Tu X, Baserga R. Control of cell size through phosphorylation of upstream binding factor 1 by nuclear phosphatidylinositol 3 kinase. Proc Natl Acad Sci U S A 2004;101:9272 6. 232. Laakso M, Malkki M, Kekalainen P, Kuusisto J, Deeb SS. Insulin receptor substrate 1 variants in non insulin dependent diabetes. J Clin Inves t 1994;94:1141 6. 233. Araki E, Sun XJ, Haag BL, 3rd, et al. Human skeletal muscle insulin receptor substrate 1. Characterization of the cDNA, gene, and chromosomal localization. Diabetes 1993;42:1041 54. 234. Dastani Z, Hivert MF, Timpson N, et al. Novel loci for adiponectin levels and their influence on type 2 diabetes and metabolic traits: a multi ethnic meta analysis of 45,891 individuals. PLoS Genet 2012;8:e1002607.

PAGE 233

233 235. Kilpelainen TO, Zillikens MC, Stancakova A, et al. Genetic variation near IRS1 ass ociates with reduced adiposity and an impaired metabolic profile. Nat Genet 2011;43:753 60. 236. Teslovich TM, Musunuru K, Smith AV, et al. Biological, clinical and population relevance of 95 loci for blood lipids. Nature 2010;466:707 13. 237. Odawara M, K adowaki T, Yamamoto R, et al. Human diabetes associated with a mutation in the tyrosine kinase domain of the insulin receptor. Science 1989;245:66 8. 238. Elbein SC, Sorensen LK, Schumacher MC. Methionine for valine substitution in exon 17 of the insulin r eceptor gene in a pedigree with familial NIDDM. Diabetes 1993;42:429 34. 239. Kakehi T, Hisatomi A, Kuzuya H, et al. Defective processing of insulin receptor precursor in cultured lymphocytes from a patient with extreme insulin resistance. J Clin Invest 19 88;81:2020 2. 240. Yoshimasa Y, Seino S, Whittaker J, et al. Insulin resistant diabetes due to a point mutation that prevents insulin proreceptor processing. Science 1988;240:784 7. 241. Moller DE, Flier JS. Detection of an alteration in the insulin recept or gene in a patient with insulin resistance, acanthosis nigricans, and the polycystic ovary syndrome (type A insulin resistance). N Engl J Med 1988;319:1526 9. 242. Grassi MA, Tikhomirov A, Ramalingam S, Below JE, Cox NJ, Nicolae DL. Genome wide meta anal ysis for severe diabetic retinopathy. Hum Mol Genet 2011;20:2472 81. 243. Lango Allen H, Estrada K, Lettre G, et al. Hundreds of variants clustered in genomic loci and biological pathways affect human height. Nature 2010;467:832 8. 244. Sandilands AJ, O'Sh aughnessy KM. The functional significance of genetic variation within the beta adrenoceptor. Br J Clin Pharmacol 2005;60:235 43. 245. Lemaitre RN, Heckbert SR, Sotoodehnia N, et al. beta1 and beta2 adrenergic receptor gene variation, beta blocker use and risk of myocardial infarction and stroke. Am J Hypertens 2008;21:290 6. 246. Hoffstedt J, Poirier O, Thorne A, et al. Polymorphism of the human beta3 adrenoceptor gene forms a well conserved haplotype that is associated with moderate obesity and altered re ceptor function. Diabetes 1999;48:203 5.

PAGE 234

234 247. Festa A, Krugluger W, Shnawa N, Hopmeier P, Haffner SM, Schernthaner G. Trp64Arg polymorphism of the beta3 adrenergic receptor gene in pregnancy: association with mild gestational diabetes mellitus. J Clin Endo crinol Metab 1999;84:1695 9. 248. Puigserver P, Rhee J, Donovan J, et al. Insulin regulated hepatic gluconeogenesis through FOXO1 PGC 1alpha interaction. Nature 2003;423:550 5. 249. Nakae J, Kitamura T, Kitamura Y, Biggs WH, 3rd, Arden KC, Accili D. The fo rkhead transcription factor Foxo1 regulates adipocyte differentiation. Dev Cell 2003;4:119 29. 250. Orho Melander M, Almgren P, Kanninen T, Forsblom C, Groop LC. A paired sibling analysis of the XbaI polymorphism in the muscle glycogen synthase gene. Diabe tologia 1999;42:1138 45. 251. Santoro N, Cirillo G, Amato A, et al. Insulin gene variable number of tandem repeats (INS VNTR) genotype and metabolic syndrome in childhood obesity. J Clin Endocrinol Metab 2006;91:4641 4. 252. Meigs JB, Dupuis J, Herbert AG, Liu C, Wilson PW, Cupples LA. The insulin gene variable number tandem repeat and risk of type 2 diabetes in a population based sample of families and unrelated men and women. J Clin Endocrinol Metab 2005;90:1137 43. 253. Abe H, Yamada N, Kamata K, et al. Hypertension, hypertriglyceridemia, and impaired endothelium dependent vascular relaxation in mice lacking insulin receptor substrate 1. J Clin Invest 1998;101:1784 8. 254. Mammarella S, Romano F, Di Valerio A, et al. Interaction between the G1057D variant of IRS 2 and overweight in the pathogenesis of type 2 diabetes. Hum Mol Genet 2000;9:2517 21. 255. Hart LM, Stolk RP, Heine RJ, Grobbee DE, van der Does FE, Maassen JA. Association of the insulin receptor variant Met 985 with hyperglycemia and non insulin dependent diabetes mellitus in the Netherlands: a population based study. Am J Hum Genet 1996;59:1119 25. 256. Hart LM, Stolk RP, Dekker JM, et al. Prevalence of variants in candidate genes for type 2 diabetes mellitus in The Netherlands: the Rotterdam st udy and the Hoorn study. J Clin Endocrinol Metab 1999;84:1002 6. 257. Tanizawa Y, Riggs AC, Chiu KC, et al. Variability of the pancreatic islet beta cell/liver (GLUT 2) glucose transporter gene in NIDDM patients. Diabetologia 1994;37:420 7.

PAGE 235

235 258. Eny KM, Wo lever TM, Fontaine Bisson B, El Sohemy A. Genetic variant in the glucose transporter type 2 is associated with higher intakes of sugars in two distinct populations. Physiol Genomics 2008;33:355 60. 259. Jeyaraj S, Boehmer C, Lang F, Palmada M. Role of SGK1 kinase in regulating glucose transport via glucose transporter GLUT4. Biochem Biophys Res Commun 2007;356:629 35. 260. Mohr S, Liew CC. The peripheral blood transcriptome: new insights into disease and risk assessment. Trends Mol Med 2007;13:422 32. 261. Sadee W, Wang D, Papp AC, et al. Pharmacogenomics of the RNA world: structural RNA polymorphisms in drug therapy. Clin Pharmacol Ther 2011;89:355 65. 262. Hindorff LA, Sethupathy P, Junkins HA, et al. Potential etiologic and functional implications of geno me wide association loci for human diseases and traits. Proc Natl Acad Sci U S A 2009;106:9362 7. 263. Livak KJ, Schmittgen TD. Analysis of relative gene expression data using real time quantitative PCR and the 2( Delta Delta C(T)) Method. Methods 2001;25: 402 8.

PAGE 236

236 BIOGRAPHICAL SKETCH Mariellen J. Moore was born in Bristol, Tennessee and lived in various cities in the southeast US. Upon admission to the University of Florida and she was invited to participate in the UF honors program. After completing her Doctor of Pharmacy degree in 2007, she entered the Clinical Pharmaceutical Sciences graduate program in the is supported by a NIH T32 training grant focused on hypertensio n, through the department of Physiology and Functional Genomics in the College of Medicine. She is a licensed pharmacist in the state of Florida and currently practices at the Gainesville became a Board Certified Pharmacotherapy Specialist (BCPS) in 2010. She works under the supervision of Dr. Julie A. Johnson studying the glycemic effects of antihypertensive medications.