Molecular Analysis and Pharmacogenetic Assessment of Calcium Signaling Pathway Variation

MISSING IMAGE

Material Information

Title:
Molecular Analysis and Pharmacogenetic Assessment of Calcium Signaling Pathway Variation
Physical Description:
1 online resource (105 p.)
Language:
english
Creator:
Davis, Heather Marie
Publisher:
University of Florida
Place of Publication:
Gainesville, Fla.
Publication Date:

Thesis/Dissertation Information

Degree:
Doctorate ( Ph.D.)
Degree Grantor:
University of Florida
Degree Disciplines:
Pharmaceutical Sciences, Pharmaceutics
Committee Chair:
Johnson, Julie A
Committee Members:
Langaee, Taimour
Zhu, Haojie
Gong, Yan
Bungert, Jorg

Subjects

Subjects / Keywords:
cardiovascular -- hypertension -- pharmacogenetics
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:
One in three American adults has hypertension. Less than half of those individuals have achieved blood pressure control despite the availability of efficacious pharmacotherapy. The current study aimed to determine whether genetic variation within Ca2+ signaling pathway genes may provide insight into the variability in responses observed to commonly used antihypertensives. Additionally, we sought to investigate the functional mechanisms of a clinically associated polymorphism found within this pathway. Blood pressure (BP) response to atenolol and hydrochlorothiazide was evaluated relative to genetic variation among Ca2+ signaling pathway candidate genes among uncomplicated hypertensives. We identified several significant and novel associations within CACNA1C (calcium channel, voltage-dependent, L type, alpha 1c subunit) and BP response to atenolol among blacks. We then evaluated the relationship between genetic variation of the Ca2+ signaling pathway and clinical susceptibility to adverse cardiovascular outcomes in a high-risk cohort of hypertensives with coronary artery disease. Herein we identified several significant and novel pharmacogenetic treatment interactions among three different racial/ethnic groups. Additionally, several polymorphisms within CACNA1C significant for BP response among uncomplicated hypertensives also demonstrated significant associations with adverse cardiovascular outcomes in our high-risk cohort. Finally, we used human vascular tissue to evaluate whether differential expression of mRNA, protein, or transcription factor binding could be observed relative to the clinically associated CACNB2 (calcium channel, voltage-dependent, beta 2 subunit) rs2357928 polymorphism. The current data show promising trends for an association with differential expression of mRNA, but definitive conclusions cannot be made due to a limiting sample size. Additional investigation is warranted, as the genetic associations reported here have yet to be replicated in an independent study population and a functional evaluation of molecular mechanisms would benefit greatly from a larger sample cohort.
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 Heather Marie Davis.
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 2014-05-31

Record Information

Source Institution:
UFRGP
Rights Management:
Applicable rights reserved.
Classification:
lcc - LD1780 2012
System ID:
UFE0043658:00001


This item is only available as the following downloads:


Full Text

PAGE 1

1 MOLECULAR ANALYSIS AND PHARMACOGENETIC ASSE SSMENT OF CALCIUM SIGNALING PATHWAY VA RIATION By HEATHER MARIE DAVIS A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT OF THE REQUI REMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY UNIVERSITY OF FLORIDA 201 2

PAGE 2

2 201 2 Heather Marie Davis

PAGE 3

3 To my parents, Collette and Robert Davis ; my sis ters Kim Cheeseman and Margo Fease r ; my aunt Eileen Larkin; and my boyfriend Justin Marquand

PAGE 4

4 ACKNOWLEDGMENTS I would like to thank my advisor, Dr. Julie Johnson for her commitment and guidance in my professional development. I would also like to thank Dr. Taimour Langaee for his gracious assistance enthusiasm for teaching, and infectious o ptimism. I owe much gratitude to Dr. Yan Gong for her expertise and advi ce I would also like to acknowledge Dr. Haojie Zhu for always t aking the time to answer my questions and making me feel welcome in his lab Additionally, I owe many thank s to Dr. J org Bungert for his advice expertise and for being available and prepared to help me with last minute questions about everything molecular biology. Members of the Vascular Surgery Department, Dr. Adam Beck, Dr. Peter Nelson, Dr. Thomas Huber, Dr. Robert Feezor, Dr. Salvatore Scali, Dr. Tim othy Flynn, Nancy Hanson, the department residents and support staff all provided invaluable assistance in the creati on and development of our vascular tissue bank, and for this I owe them much gratitude. Additionally, I would like to thank all of the patients who chose to participate in our clinical study. Current and former members of the department, including Dr. Mariellen Moore, Dr. Caitrin McDonough, Dr. Jaekyu Shin, Dr. Max imilian Lobmeyer, Dr. Elvin Price, Dr. Ju lio Duarte, Dr. Abdolreza Davoodi Semiromi, Dr. Rhonda Cooper DeHoff, Ben Burkley, Lynda Stauffer Cheryl Galloway and Azadeh Hassanzadeh all provided assistance throughout the course of this pro ject for which I have much appreciation. I would also like to thank my family for standing behind me throughout this endeavor. My mother, father, sisters and in laws have all been encouraging patient and kind. Finally, I would like to thank my boyfriend Justin and our dog Deuce for making the whole experience much more colorful

PAGE 5

5 TABLE OF CONTENTS page ACKNOWLEDGMENTS ................................ ................................ ................................ .. 4 LIST OF TABLES ................................ ................................ ................................ ............ 8 LIST OF FIGURE S ................................ ................................ ................................ ........ 10 ABSTRACT ................................ ................................ ................................ ................... 11 CHAPTER 1 BACKGROUND AND SIGNIFICANCE ................................ ................................ ... 13 Hypert ension ................................ ................................ ................................ ........... 13 Calcium Current ( I Ca ) ................................ ................................ .............................. 14 Excitation Contraction Coupling ................................ ................................ .............. 14 Calcium Signaling in Cardiovascular Disease ................................ ......................... 15 CACNA1C ................................ ................................ ................................ ........ 15 CACNA1D ................................ ................................ ................................ ........ 16 CACNB2 ................................ ................................ ................................ ........... 16 KCNMB1 ................................ ................................ ................................ .......... 18 Calcium Signaling and Antihypertensive Therapy ................................ ................... 19 Conclusion ................................ ................................ ................................ .............. 19 2 ASSOCIATION OF CALCIUM SIGNALING PATHWAY VARIATION WITH BLOOD PRESSURE RESPONSE WITHIN THE PHARMACOGENOMICS OF ANTIHYPERTENSIVE RESPONSE (PEAR) STUDY ................................ ............. 22 Introduction ................................ ................................ ................................ ............. 22 Methods ................................ ................................ ................................ .................. 23 Patient Population ................................ ................................ ............................ 23 Genotype Determination ................................ ................................ ................... 24 Statistical Analyses ................................ ................................ .......................... 25 Results ................................ ................................ ................................ .................... 27 Genotyping ................................ ................................ ................................ ....... 27 Genetic Associations with Blood Pressure Response ................................ ...... 28 Haplotype Analysis ................................ ................................ ........................... 30 Discussion ................................ ................................ ................................ .............. 31 3 ASSOCIATION OF CALCIUM SIGNALING PATHWAY VARIATION WITH CARDIOVASCULAR OUTCOMES IN THE INTERNATIONAL VERAPAMIL SR TRANDOLAPRIL STUDY GENETIC SUBSTUDY (INVEST GENES) ................. 50 Introduction ................................ ................................ ................................ ............. 50 Methods ................................ ................................ ................................ .................. 50

PAGE 6

6 INVEST GENES Clinical Cohort ................................ ................................ ...... 50 Genotype Determination ................................ ................................ ................... 51 Statistical Analyses ................................ ................................ .......................... 52 Results ................................ ................................ ................................ .................... 53 Genotyping and Quality Control ................................ ................................ ....... 54 Genetic and Pharmacogenetic Associations Wi th Adverse Cardiovascular Outcomes ................................ ................................ ................................ ...... 54 CACNA1C ................................ ................................ ................................ .. 54 CASQ2 ................................ ................................ ................................ ....... 56 KCNMB1 ................................ ................................ ................................ .... 57 RYR2 ................................ ................................ ................................ ......... 58 PLN ................................ ................................ ................................ ............ 59 Discussion ................................ ................................ ................................ .............. 59 4 INFLUENCE OF THE RS2357928 PROMOTER POLYMORPHISM ON THE EXPRESSION PROFILE OF THE CA V 1.2 BETA 2 SUBUNIT IN HUMAN VASCULAR TISSUE ................................ ................................ ............................... 70 Introduction ................................ ................................ ................................ ............. 70 Methods ................................ ................................ ................................ .................. 71 Study Protocol ................................ ................................ ................................ .. 71 DNA ................................ ................................ ................................ .................. 72 RNA ................................ ................................ ................................ .................. 72 Protein ................................ ................................ ................................ .............. 73 Nuclear extract ................................ ................................ ........................... 73 Quantitation ................................ ................................ ................................ 73 Antibodies ................................ ................................ ................................ .. 74 Western blot ................................ ................................ ............................... 74 Electrophoretic mo bility shift assay (EMSA) ................................ ............... 75 Analysis ................................ ................................ ................................ ............ 76 Results ................................ ................................ ................................ .................... 77 Discussion ................................ ................................ ................................ .............. 78 5 SUMMARY AND CONCLUSIONS ................................ ................................ .......... 85 APPENDIX A NOMINAL ASSOCIATIONS BETWEEN CALCIUM SIGNALING PATHWAY VARIATION AND BLOOD PRESSURE RE SPONSE WITHIN THE PHARMACOGENOMICS OF ANTIHYPERTENSIVE RESPONSE (PEAR) STUDY ................................ ................................ ................................ .................... 88 B NOMINAL ASSOCIATIONS BETWEEN CALCIUM SIGNALING PATHWAY VARIATION AND ADVERSE CARDIOVASCULAR OUTCOMES IN THE IN TERNATIONAL VERAPAMIL SR TRANDOLAPRIL STUDY GENETIC SUBSTUDY (INVEST GENES) ................................ ................................ .............. 92

PAGE 7

7 LIST OF REFERENCES ................................ ................................ ............................. 100 BIOGRAPHICAL SKETCH ................................ ................................ .......................... 105

PAGE 8

8 LIST OF TABLES Table page 2 1 Illumina BeadChip SNP Coverage and p value thresholds ............................... 35 2 2 Power to detect diastolic BP differences across MAF ranges by race, PEAR .... 35 2 3 PEAR Baseline Demographics ................................ ................................ ........... 36 2 4 Associations between CACNA1C rs2239101 genotype and blood pressure response to atenolol among PEAR blacks ................................ ......................... 37 2 5 Associations between CACNA1C rs12425032 genotyp e and blood pressure response to atenolol and HCTZ among PEAR blacks ................................ ....... 38 2 6 Associations between CACNA1C rs2238078 genotype and blood pressure response to atenolol among PEAR blacks ................................ ......................... 39 2 7 Associations between CACNA1C rs2238078 genotype and blood pressure response to HCTZ among PEAR whites ................................ ............................ 40 2 8 Associations between CACNA1C rs11831085 genotype and blood pressure response to atenolol among PEAR blacks ................................ ......................... 41 2 9 CACNA1C rs11831085, rs12425032 haplotype frequency among PEAR black s ................................ ................................ ................................ ................. 41 2 10 Associations between CACNA1C rs11831085, rs12425032 AT haplotype and blood pressure response among PEAR blacks ................................ ........... 42 2 11 Associations between CACNA1C rs11831085, rs12425032 GC haplotype and blood pressure response among PEAR blacks ................................ ........... 43 2 12 Summary of significant CACNA1C SNP associations wit h BP response ........... 44 3 1 Power to detect differences across MAF, INVEST GENES ............................... 64 3 2 INVEST GENES baseline characteri stics ................................ ........................... 65 3 3 Associations between SNPs of the Ca 2+ signaling pathway and adverse cardiovascular outcomes in INVEST GENES a ................................ ................... 66 3 4 Significant SNP treatment interactions a ................................ .............................. 67 4 1 Baseline demographics of vascular tissue bank participants .............................. 81 4 2 Analysis of RT PCR mean C T by CACNB2 rs2357928 genotype ..................... 81

PAGE 9

9 4 3 Net intensity of CACNB2 and GAPDH expression by CACNB2 rs2357928 genotype ................................ ................................ ................................ ............. 82 A 1 Associations between CACNA1C rs2299657 genotype and blood pressure response to HCTZ among PEAR whites ................................ ............................ 89 A 2 Associations between CACNA1C rs2299657 genotype and blood pressure response to atenolol among PEAR whites ................................ ......................... 89 A 3 Associations between CACNB2 rs7089228 genotype and blood pressure response to HCTZ among PEAR blacks ................................ ............................ 90 A 4 Associations between RYR2 rs16832052 genotype and blood pressure response to atenolol among PEAR blacks ................................ ......................... 90 A 5 Associations between CACNA1C rs2238078, rs11831085, rs12425032, rs2239101 TGCT haplotype and blood pressure response to atenolol among PEAR blacks ................................ ................................ ................................ ....... 91 A 6 Associations between CACNA1 C rs2238078, rs11831085, rs12425032, rs2239101 ATGT haplotype and blood pressure response to atenolol among PEAR blacks ................................ ................................ ................................ ....... 91 B 1 Associations between variation within CACNA1C and adverse cardiovascular outcomes in INVEST GENES a ................................ ................................ ........... 93 B 2 INVEST GENES CACNA1C SNP treatment interactions a ................................ .. 94 B 3 Associatio ns between variation within CASQ2 and adverse cardiovascular outcomes in INVEST GENES a ................................ ................................ ........... 95 B 4 INVEST GENES CASQ2 SNP treatment interactions a ................................ ...... 95 B 5 Associations between variation within KCNMB1 and adverse cardiovascular outcomes in INVEST GENES a ................................ ................................ ........... 96 B 6 INVEST GENES KCNMB1 SNP treatment interactions a ................................ .... 96 B 7 Associations between variation within RYR2 and adverse cardiovascular outcomes in INVEST GENES a ................................ ................................ ........... 97 B 8 IN VEST GENES RYR2 SNP treatment interactions a ................................ ......... 98 B 9 Associations between variation within PLN and adverse cardiovascular outcomes in INVEST GENES a ................................ ................................ ........... 99

PAGE 10

10 LIST OF FIGURES Figure page 1 1 Schematic of the calcium signaling pathway ................................ ...................... 21 2 1 LD structure of CAC NA1C significant associations within PEAR blacks, r 2 values shown ................................ ................................ ................................ ...... 44 2 2 Blood pressure response to the combined assessment of atenolol monotherapy with atenolol add on from the oppos ite treatment arm by CACNA1C rs2239101 genotype among PEAR blacks ................................ ....... 45 2 3 Blood pressure response to the combined assessment of monotherapy with add on from the opposite treatment arm by CACNA1C rs12425032 genotype among PEAR blacks ................................ ................................ ........................... 46 2 4 Blood pressure response to the combined assessment of atenolol monotherapy with atenolol add on from the opposite treatment arm by CACNA1C rs11831085 genotype ................................ ................................ ....... 47 2 5 Blood pressure response to the combined assessment of monotherapy with add on from the opposite treatment arm by CACNA1C AT haplotype among PEAR blacks ................................ ................................ ................................ ....... 48 2 6 Blood pressure response to the combined assessment of atenolol monotherapy with atenolol add on from the opposite treatment arm by CACNA1C GC haplotype among PEAR black s ................................ .................. 49 3 1 Primary outcome adjusted odds ratios (log scale shown) and 95% confidence intervals by treatment strategy among INVEST GENES case control cohort black CACNA1C rs12425032 mino r allele carriers. ................................ ............ 68 3 2 Primary outcome adjusted odds ratios and 95% confidence intervals by treatment strategy among INVEST GENES case control cohort white CASQ2 rs3811003 minor al lele carriers. ................................ ................................ ......... 69 4 1 Relative CACNB2 mRNA expression in human arterial tissue, G/G reference genotype, individual RQ values plotted. ................................ ............................. 82 4 2 Western blot analysis of CACNB2 protein expression relative to GAPDH. ......... 83 4 3 EMSA characterization of DNA and protein interactions. ................................ ... 84

PAGE 11

11 Abstract of D issertation Presented to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Doctor of Philosophy MOLECULAR ANALYSIS AND PHARMACOGENETIC ASSE SSMENT OF CALCIUM SIGNALING PATHWAY VA RI ATION By Heather Marie Davis M ay 201 2 Chair: Julie A. Johnson Major: Pharmaceutical Sciences One in three American adults has hypertension Less than half of those individuals have achieved blood pressure control despite the availability of effica cious pharmacotherapy. The current study aimed to determine whether genetic variation within Ca 2+ signaling pathway genes may provide insight into the variability in responses observed to commonly u sed antihypertensive s. Additionally, we sought to invest igate the functional mechanisms of a clinically associated polymorphism found within this pathway. Blood pressure (BP) response to atenolol and hydrochlorothiazide was evaluated relative to genetic variation among Ca 2+ signaling pathway candidate genes am ong uncomplicated hypertensives. We identified several significant and novel associations within CACNA1C (calcium channel, voltage dependent, L type, alpha 1c subunit) and BP response to atenolol among blacks We then evaluated the relationship between genetic variation of the Ca 2+ signaling pathway and clinical susceptibility to adverse cardiovascular outcomes in a high risk cohort of hypertensives with coronary artery disease. Herein we identified several significant and novel p harmacogenetic treatmen t interactions among three different

PAGE 12

12 racial/ethnic groups. Additionally several polymorphisms within CACNA1C significant for BP response among uncomplicated hypertensives also demonstrated significant associations with adverse cardiovascular outcomes in our high risk cohort. Finally, we used human vascular tissue to evaluate whether differential expression of mRNA, protein, or transcription factor binding could be observed relative to the clinically associated CACNB2 (calcium channel, voltage dependent, beta 2 subunit) rs2357928 polymorphism The current data show promising trends for an association with differential expression of mRNA but definitive conclusions cannot be made due to a limiting sample size. Additional investigation is warranted, as the genetic associations reported here have yet to be replicated in an independent study population and a functional evaluation of molecular mechanisms would benefit greatly from a larger sample cohort.

PAGE 13

13 CHAPTER 1 BACKGROUND AND SIGNIFICANCE Hypertension A n estimated 7 6 million Americans, or roughly one in three American adults has hypertension (HTN). Elevated blood pressure (BP) is a primary risk factor for myocardial infarction (MI), heart failure, stroke and kidney diseases the incidence of which incre ase proportionately with BP value. 1 Of Americans treated for HTN, as few as 4 8 % have achieved BP control in the face of widely available and efficacious antihypert ensive agents. 2 The pathophysiology of essential HTN involves numerous interrelated environmental, biologic and genetic factors that contribute to elevated BP, and t heir effects differ between individuals. 3 Successful BP control relies largely upon a and imal drug therapy as any particular antihypertensive may only be efficacious in 40 60% of patients. 4 Given the heterogeneous nature of HTN, response to any single antihypertensive agent is characterized by marked variability between individuals and traditional predictors of BP response such as race, age and gender ar e of limited usefulness in efforts to identify optimal therapy for a particular patient. 5 T he lack of rationale for drug selection and the time lag associated with identifying an effectual regimen to achieve BP control is associated with poor therapeutic outcomes 6 The principle aim of pharmacogenetic research is to identify genetic factors for use as a screening tool to aid in the selection of drug therapy that would be optimal for an individual patient. Targeting the genetic component of HTN pathophysiology may enhance efficacy and improve overall benefit from therapy. 7 This tailored approach could decrease time to and 5

PAGE 14

14 Calcium Current ( I Ca ) In its vital role as an intracellular second messenger, Ca 2+ regulates a number of diverse physiological processes including excitation contraction coupling, electrical signaling, hormone secretion and gene transcription. 8 9 Voltage gated Ca 2+ channels (Ca v ) belong to a gene superfamily of transmembrane ion channel proteins that also includes both voltage gated sodium (Na + ) and potassium (K + ) channels. 9 10 Ca v contain 1 pore forming and voltage sensing subunit is the largest. 11 1 subunits and Ca v 1 subunit they contain (Ca v 1 to Ca v 3). 9 L type Ca 2+ current ( I Ca,L ) is conducted by Ca v 1 channels and characterized by high voltage activation as wel l as sensitivity to Ca 2+ channel blockers (CCBs) widely used in the treatment of HTN and cardiac arrhythmia. 9 Ca v 1D ) is found at high levels in pacemaker cells of the heart where it controls cardiac rhythm. 9 12 Ca v 1C ), found in vascular smooth muscle and ventricular myocytes, provides the majority of I Ca,L within the myocardium, where channel activation carries the plateau phase o f the cardiac action potential. 8 10 Excitation Contraction Coupling Ca 2+ influx through depolarization activated Ca v 1.2 o n the sarcolemma produces near simultaneous activation of numerous ryanodine receptors (RyR, RYR2 ) and the coordinated release of large amounts of Ca 2+ from the sarcoplasmic reticulum (SR, Figure 1 1). In this process best known as Ca 2+ induced Ca 2+ relea se, myocardial RyRs function to increase free intracellular Ca 2+ ([Ca 2+ ] i ) within the myocyte, allowing Ca 2+ to bind and activate contractile myofilaments. 8 Relaxation arises from dec lining [Ca 2+ ] i as Ca 2+ reuptake into the SR occurs via activation of the SR Ca 2+ ATPase 2a pump

PAGE 15

15 (SERCA2a, ATP2A2 ) and cellular extrusion of Ca 2+ occurs across the sarcolemma via the Na + / Ca 2+ exchanger (NCX, SLC8A1 ). 11 In arterial smooth muscle each Ca 2+ spark also activates between 10 100 Ca 2+ activated potassium (K + ) channels (K Ca or BK) on the plasma membrane. K Ca activation causes K + efflux and subsequent membrane hyperpolarization, a process that deactivates Ca v F ound in vascular smooth muscle K Ca functions through this negative feedback mechanism and restores resting membrane potential in vascular smooth muscle following Ca 2+ induced contraction. Ca lcium Signaling in Cardiovascular Disease CACNA1C Located on chromosome 12p 13.3, CACNA1C is a large, complex gene with limited linkage disequilibrium encoding the 210 240 kDa 1C subunit of Ca v 1.2 channels. Mutations in this gene have been associated with lethal cardiac arrhythmias and sudden cardiac death. 13 Located in the coding sequence of CACNA1C the synonymous singl e nucleotide polymorphism (SNP) rs1051375 has been identified as having a significant interaction with treatment strategy in a cohort of high risk hypertensive patients with coronary artery disease (CAD). 14 Individual s homozygous for the major allele (A/A) within the INternational VErapamil SR Trandolapril STudy GENEtic Substudy (INVEST GENES) case control cohort randomized to a verapamil SR based CCB treatment strategy experienced a 45% reduced risk of primary outcome (first occurrence of all c ause mortality, nonfatal myocardial infarction (MI) or nonfatal stroke; OR 0.54, 95% CI 0.32 0.92) compared to individuals randomized to an atenolol based strategy. 14 In contrast, individuals homozygous for the minor allele (G/G) randomized to the CCB treat ment strategy experienced a 4.5 fold increase in the primary outcome

PAGE 16

16 compared to G/G homozygotes randomized to the strategy (OR 4.59, 95% CI 1.67 12.67) The evidence suggests subjects carrying the CACNA1C rs1051375 A/A genotype may experience better outcomes if treated with a CCB, whereas G/G carriers would likely benefit B t herapy. 14 Subsequen t analyses showed A/A carriers randomized to the strategy were more likely to require four or more drugs to achieve BP control than those carriers randomized to the CCB treatment strategy 14 The data suggest that differences in BP response may influence the effect observed in trea tment outcomes. 14 These data from a large, multicenter, randomized hypertensive treatment outcomes trial have incited additional investigation of the association of CACNA1C variants with cardiovascular disease phenotypes. CACNA1D Located on ch romosome 3p14.3, CACNA1D encodes the 1D subunit of Ca v 1.3 responsible for I Ca,L within pacemaker cells of the heart. Genetic variations in CACNA1D have been associated with BP response to dihydropyridine CCB in a small cohort of Japanese hypertensives. 15 CACNB2 CACNB2 encodes the 2 regulatory subunit of Ca v 1.2 involved in targeting the 1 subunit to the plasma membrane where the channel exerts its function. Via hydrophobic interactions the 2 subunit is tightly bound to a highly conserved motif in 1 subunit, interaction domain (AID). This I II loop of the 1 subunit contains a retention signal for the endoplasmic reticulum that restricts cell surface expression of the channel. Interaction with 2 subunit reverses the inhibition associated with the retention signal leading to cell surface expression of Ca v 1.2 16 As Ca v 1.2 must be on

PAGE 17

17 the cell surface for CCBs to bind, the 2 subunit may be a functionally important mediator of response to CCBs. Genetic variation in CACNB2 has been associated with B rugada syndrome, cardiac conduction disease and hypertrophic cardiomyopathy. 17 18 A genome wide association study of BP and HTN within the Cohorts for Heart and Aging Research in Genomic Epidemiology (CHARGE) Consortium found the CACNB2 rs11014166 intronic variant to be significantly associated with the phenotypes of systolic BP (SBP), diastolic BP (DBP), and HTN. 19 Within the INVEST GENES case control cohort, Hispanic carriers of the minor CACNB2 rs11014166 T allele randomized to the CCB treatment strategy had greater risk for the primary outcome than those randomized to the B treatment strategy (adjusted HR [CCB versus B], 3.13; 95% CI, 1.39 7.06; p = 0.006). 20 Additional analyses showed the pharmacogenetic effect of thi s SNP was primarily driven by all cause mortality (adjusted HR, 22.0; 95% CI, 2.63 184.17; p = 0.0043). 20 A SNP within an alternative promoter of CACN B2 rs2357928, was also found to have a significant interaction with treatment strategy in the INVEST GENES white population (p for interaction, 0.002). Minor allele homozygotes (G/G) randomized to the CCB treatment strategy were at an increased risk for adverse cardiovascular outcomes when compared to those randomized to the B treatment strategy (adjusted HR [CCB versus B] 2.35; 95% CI, 1.19 4.66; p = 0.014). 20 No similar associations were found for carriers of the major allele (A/A, A/G). 20 The rs2357928 promoter SNP of CACNB2 has also been associ ated with decreased transcriptional activity in a reporter gene assay, which lends support to the hypothesis that the observed pharmacogenetic

PAGE 18

18 interaction is driven by a functional mechanism affecting Ca v 1.2 2 subunit gene expression. 20 KCNMB1 Encoded by KCNMB1 the K Ca 1 regulatory subunit, greatly enhances Ca 2+ sensitivity of the K Ca channel complex in v ascular smooth muscle. K Ca opening restores resting cell membrane potential by coupling intracellular Ca 2+ sparks to membrane potential hyperpolarization and is thus a critical regulator of arterial tone. 21 The KCNMB1 Glu65Lys variant was evaluated within a Spanish cohort that compared normotensive subjects to those with differing levels of diastolic HTN. 22 The genotype frequency of the KCNMB1 Glu65Lys mutation decreased with increasing DBP values, an association that suggested a protective effect of the Lys65 allele and a progressively deleterious effect of the Gl u65Glu genotype. 22 1 subunit in HEK 293 cells alone or in combination with the wild 1 subunit showed enhanced K Ca channel activity, suggesting the variant provides a gain of function. 22 The KCNMB1 Glu65Lys variant was also investigated in association with variable BP response to CCB treatment within the INVEST GENES cohort A mong subjects receiving verapamil SR monotherapy KCNMB1 Lys65 vari ant carrier status was significantly associated with the need for fewer drugs to achieve BP control (OR, 0.48; 95% CI, 0.23 0.99), and shorter median time to achieve BP control compared to carriers of the Glu65Glu genotype (1.47 months v. 2.83 months; p = 0.01 ) 23 Additionally within INVEST GENES the KCNMB1 valine to leucine polymorphism at codon 110 (Val110Leu) was associated with a 33% risk reduction for the primary outcome w ithin the nested case control cohort (OR, 0.66; 95% CI, 0.43 1.01). 23 Overall, these data

PAGE 19

19 suggest genetic variation with in KCNMB1 may contribu te to the heterogeneous response to verapamil SR observed in this high risk hypertensive cohort. Calcium Signaling and Antihypertensive Therapy In addition to the CCB class of antihypertensive agents, a number of commonly prescribed antihypertensives inter cede in pathways in which Ca 2+ signaling is intimately involved. blockers inhibit the binding of endogenous catecholamines to 1 / 2 adrenergic receptors. adrenergic receptors couple to the stimulatory G protein (G s ) which initiates a signaling cascade with subsequent activation of adenylyl cyclase, increase in cy clic adenosine monophosphate, and activation of protein kinase A (PKA). 24 PKA phosphorylates a number of substrates important in intracellular Ca 2+ regulation including the Ca v 1 channels, RYR and phospholambam (PLN, Figure 1 1). 11 Additionally, a propo sed mechanism for the antihypertensive effects of thiazide and thiazide like diuretics is enhancing K Ca channel sensitivity, thereby enhancing vasodilation mediated by vascular smooth muscle. 25 Conclusion In light of evidence supporting the association of variants within the Ca 2+ signaling pathway with adverse cardiovascular outcomes in t reated hypertensives, we have developed the following hypotheses: 1) Variation present in genes comprising the activation pathway of Ca 2+ signaling and Ca v 1 are associated with BP response to antihypertensive therapy. 2) Genetic variation within the Ca 2+ signaling pathway is associated with clinical susceptibility to adverse CV outcomes in treated hypertensives with CAD.

PAGE 20

20 To address the first hypothesis, we will identify genetic associations with BP response within the P harmacogenomic E valuation of A ntih ypertensive R esponses (PEAR) study, a population of uncomplicated hypertensives, in a group of Ca 2+ pathway candidate genes: ATP2A2 CACNA1C CACNA1D CACNB2 CASQ2 KCNMB1 PLN RYR2 SLC8A1 W e will also broaden the scope of the previous investigation s w ithin INVEST GENES to test the second hypothesis that genetic variation within interrelated genes of the Ca 2+ signaling pathway ( ATP2A2, CACNA1C, CACNA1D, CACNB2, CASQ2, KCNMB1, PLN, RYR2 and SLC8A1 ) are associated with adverse cardiovascular outcomes in t reated hypertensives with CAD. A final hypothesis addresses evidence of altered transcriptional activity documented with the clinically associated rs2357928 promoter SNP of CACNB 2. 3) Quantifiable differences in gene expression may contribute to clinical ly significant genetic associations found in regulatory sequences. To investigate the underlying functional mechanism for this clinically associated promoter SNP we will utilize techniques in molecular biology to ascertain how this varia n t influences the expression profile of the Ca v 1.2 2 subunit in human vascular tissue.

PAGE 21

21 Figure 1 1. Schematic of the c a lcium s ignaling p athway

PAGE 22

22 CHAPTER 2 ASSOCIATION OF CALCI UM SIGNALING PATHWAY VARIATION WITH BLOOD PRESSURE RESPONSE WI THIN THE PHARMACOGEN OMICS OF ANTIHYPERTENSIVE RES PON SE (PEAR) STUDY Introduction Thiazide type diuretics are current Joint National Committee (JNC) guideline recommended initial therapy for hypertensives without compelling indications. 1 Despite being the third most commonly prescribed antihypertensive in the United States in 2010, the mechanism by which hydrochlorothiazide (HCTZ) produces long term BP lowering has yet to be elucidated. Current evidence suggests the v asorelaxant effects of HCTZ may be generated by opening of the K Ca channel, either directly, or via pH activation through the inhibition of carbonic anhydrase. 25 27 Add itional mechanistic evidence indicates thiazide like diuretics may cause Ca 2+ desensitization via the Rho Rho kinase pathway to moderate agonist induced vasoconstriction. 28 Also an appropriate first blockers intimately linked to the actions of Ca 2+ signaling as previously described. Gi ven the heterogeneous nature of hypertension pathophysiology and the widely variable responses observed to commonly used antihypertensives, pharmacogenetic analyses may prove valuable in informing therapeutic decisions that are currently guided by a 2+ signaling pathway genes and BP have been illuminated in recent literature. 14 19 23 29 The current study aimed to determine whether genetic variation within Ca 2+ signaling pathway genes may provide i nsight into the variability in responses to two commonly used antihypertensive drugs (Chapter 1).

PAGE 23

23 Methods Patient Population Pharmacogenomic Evaluation of Antihypertensive Responses (PEAR) was a multi center, prospective, open label, randomized study desig ned to address whether combination could be identified. 30 PEAR recruited hyper tensive individuals aged 17 65 years with DBP 90 110 mm Hg, who were without cardiovascular or renal disease, diabetes or any of the traditional exclusions to 30 Subjects were enrolled at the University of Florida (Gainesville, FL), Emory University (Atlanta, GA) and Mayo Clinic (Rochester, MN). Potential subjects were those with newly diagnosed, untreated or known hypertension treated with one or two antihyper tensive drugs. Eligible subjects currently treated with antihypertensive therapy had their therapy tapered (as necessary) and discontinued, with a minimum antihypertensive free period of 18 days, and a preferred washout period of 4 to 6 weeks. Subjects n ot meeting any exclusion criteria were further screened for BP inclusion based on untreated home and office BP. Following baseline studies that included the collection of home and 24 hour ambulatory BP data, PEAR subjects were then randomized to HCTZ 12.5 mg daily or atenolol 50 mg daily in an unblinded fashion. Subjects returned after 3 weeks on the initial dose, and proceeded through a dose titration protocol to 25 mg and 100 mg, respectively, bas ed on a threshold BP >120/70 mm Hg. Those subjects with were held at their current treatment step. Response to therapy was assessed after a minimum of 6 weeks on the target dose, after which those subjects with BP >120/70 mm Hg had the second drug added with a similar dose titration and respo nse assessment procedure. 30

PAGE 24

24 Genotype Determination PEAR genotypes for Ca 2+ signaling pathway candidate genes were obtained from the Illumina HumanCVD and HumanOmni 1 Quad BeadChips. The HumanCVD BeadChip is a custom array of over 50,000 markers designed to capture genetic diversity across approximately two thousand genes thought to be involved in a range of cardiovascular, metabolic and inflammatory syndromes. 31 Normalized DNA (50 ng/L) prepared at the University of Florida (UF) Center for Pharmacogenomics was transferred to the UF Interdisciplinary Center for Biotechnology Research (UF ICBR) where the HumanCVD BeadChip was genotyped on Illumina Infinium II Assay (Illumina San Diego, CA). Genotypes were called using GenomeStudio Software version 2011.1 and the Genotyping Module version 1.9 calling algorithm (Illumina San Diego, CA). The HumanOmni1 Quad BeadChip covers over one million loci sourced from HapMap Phases 1 3 and the 1000 Genomes Project with coverage for minor allele frequency (MAF) >0.05. HumanOmni1 Quad BeadChip genotyping was run by Illumina Inc. (San Diego, CA) coordinated by the Brown Foundation Institute of Molecular Medicine for the Prevention of Human Diseases at the University of Texas, Houston. Calcium signaling pathway candidate genes were selected for analysis as previou sly described in Chapter 1. Genotypes for SNPs within genomic regions 10 kb up and downstream of candidate genes were obtained. SNPs with MAF less than 5% in both race groups were not included in the final analysis due to the limited power to detect assoc iations at that threshold. The putative functionality of SNPs meeting criteria for significance was assessed with PupaSuite 3.1 (http://pupasuite.bioinfo.cipf.es).

PAGE 25

25 Statistical Analyses For quality control, Hardy Weinberg equilibrium was tested via 2 anal ysis performed separately by principal component analysis confirmed race, SNPs with Hardy Weinberg p values less than 10 5 had all associations ignored. Associations between genotype and BP responses were tested by linear regression following adjustment f or covariates that included age, gender and baseline BP. Baseline characteristics were compared by treatment strategy using 2 or t test as appropriate. Statistical analyses were performed in JMP Gen omic s 5 and SAS 9.2 (SAS Institute I nc Cary, NC). Demogra phic variables recorded at baseline and BP measured by home BP monitoring at the end of the baseline period and the assessment following monotherapy were used to determine BP response to monotherapy. Response to monotherapy combined with add on therapy in the opposite treatment arm was adjusted for BP at baseline in the monotherapy group and BP at the first response assessment for the add on therapy group as well as controlling for the order of drug initiation. Response to add on therapy was determined as the home BP observed at the second response assessment adjusted for the home BP value from the first (monotherapy) response assessment. All SNPs were tested for an association with monotherapy, add on therapy in the opposite treatment arm and the co mbination of monotherapy with add on therapy from the opposite treatment arm. For ATP2A2 CASQ2 KCNMB1 and PLN a prespecified level of p = 0.01 was used to identify SNPs of interest upon initial assessment of monotherapy and add on therapy. There were relatively fewer SNPs tested for these genes and thus the more liberal threshold for significance (Table 2 1) For CACNA1C

PAGE 26

26 C ACNA1D CACNB2 RYR2 and SLC8A1 given the large number of SNPs tested, a p < 0.005 was the required threshold for SNPs of interest upon initial assessment. Subsequent analysis consisted of evaluating for the consistency of the association between monothe rapy and add on therapy, requiring a nominal threshold in the alternate assessment (p < 0.05), along with testing for directional consistency when testing for the evaluations of the same drug, or an opposite effect for the other study drug (p = 0.5). Tho se SNPs achieving an initial p value of interest and surviving additional criteria for significance are considered the strongest association signals. This approach greatly reduces the probability of type I error, as demonstrated with the largest genes eva luated in this study, an initial p = 0.005 (of interest) combined with a consistent association in the alternate assessment p < 0.05, and an additional test for directional consistency yields a 1.25x10 4 (Table 2 1). SNPs previously reported in the literature for an association with BP, hypertension or antihypertensive drug response were considered if they achieved nominal significance. A strict Bonferroni correction was not used be cause of the differences in numbers of SNPs per gene (i.e. from 11 to 639), and because the Bonferroni correction assumes independence of all tests, which these were not, due to linkage disequilibrium (LD). However, the required p value to identify SNPs o f interest was reduced from the nominal p of 0.05 to try to balance too strict a p value requirement resulting in dismissal of true positives, while minimizing the false positives for which we will pursue replication in the future. Power calculations were generated using G*Power version 3.1.3, and tested the hypothesis that SNPs within candidate genes were associated with BP lowering in

PAGE 27

27 response to HCTZ or atenolol. 32 For simplicity, the power analysis is based on two groups, those subjects with at least one copy of the variant allele and the group and 0.005 performed separately by race (Table 2 2). Effect sizes were based off of diastolic BP lowering response data from PEAR. In whites (n=461), we had 80% power to detect a 4.7 he same analysis in blacks (n=304) demonstrated we had 80% power to detect a 5.9 mmHg change in diastolic BP for a SNP with MAF of 0.05. Linkage disequilibrium was assessed with PEAR genotypes in the HapMap PHASE format (http://stephenslab.uchicago.edu/so ftware.html) uploaded to Haploview 4.2. 33 Haplotypes were inferred for each racial/ethnic group using PHASE version 2.1 software and coded according to the number of copes (zero, one, or two) 33 Results Genotyping PEAR baseline demographics are included in Table 2 3. There was a significant difference in baseline BP by treatment strategy, although this is of little relevance as comparisons between treatment arms were not made. HumanCVD and HumanOmni1 Q uad data were collected on 768 and 767 PEAR subjects, respectively. Patients were excluded if sample genotype call rates were below 95% and SNPs were excluded if genotype call rates were below 90%. 81 blind duplicates were included in genotyping and had a concordance rate of 99 992 %. Gender was confirmed from X chromosome genotype data, and those who were discordant were excluded ( n =1). Cryptic relatedness was estimated by pairwise identity by descent (IBD) analysis implemented using PLINK (http://pngu.m gh.harvard.edu/purcell/plink/). 34 One pair of monozygotic

PAGE 28

28 twins was identified in this analysis and confirmed by the study coordinator, and one twin was re moved (n=1). Five pairs of samples were identified as first degree relatives, these individuals were kept for the analysis. Heterozygosity was assessed using PLINK, by estimating the inbreeding coefficient, F. One subject had F values > 4 negative stand ard deviations from the mean and was excluded. The final dataset consisted of 765 subjects. Quality control included removing monomorphic SNPs and those with a missing data rate >10%. Additional SNPs were removed from the HumanOmni1 Quad dataset with a Hardy Weinberg p < 10 6 No subjects were removed for a missing data rate >5% from either dataset. Four additional subjects identified as repeats were removed from the HumanOmni1 Quad dataset. Data were concordant for HumanOmni1 Quad BeadChip quality co ntrol repeat genotyping (n=15). Following quality control, final analyses were conducted on 761 PEAR subjects with HumanOmni1 Quad data. Genetic Associations with Blood Pressure Response Four SNPs within CACNA1C met prespecified criteria for significance to atenolol therapy among PEAR blacks. The LD structure of the significant associations is detailed in Figure 2 1 and highlights very limited LD between these SNPs The rs2239101 SNP of CACNA1C recently identified as associated with systolic BP in a can didate gene analysis of BP and hypertension in 6 European cohorts comprising the Cohorts for Heart and Aging Research in Genomic Epidemiology Consortium (CHARGE) was nominally associated with BP response to atenolol monotherapy, as well as atenolol add on, and overall significantly associated with the response to atenolol monotherapy combined with atenolol add on from the opposite treatment arm among blacks (Figure 2 2). 29 As reported in CHARGE, increasing copies of the

PAGE 29

29 CACNA1C rs2239101 minor allele were associated with a decrease in BP, within PEAR this association corresponded to a greater BP response to atenolol therapy, observed for both diastolic and systolic BP responses (Table 2 4). The association between the conserved rs12425032 intronic SNP of CACNA1C and BP response to antihypertensive therapy in PEAR (Table 2 5) is considered the strongest as BP response to atenolol therapy met criteria for significan ce among blacks, and a directionally opposite, nominal association was also observed for BP response to HCTZ in the same race group (Table 2 5). Increasing copies of the CACNA1C rs12425032 minor allele were significantly associated with a lower BP respons e to atenolol add on therapy in blacks. Directionally consistent associations were also observed for BP response to atenolol monotherapy as well as for the combined assessment of atenolol monotherapy with atenolol add on from the op posite treatment arm (F igure 2 3 ). Response to HCTZ add on therapy showed a directionally opposite association, as increasing copies of the rs12425032 minor allele were nominally associated with greater BP response (Table 2 5). A similar trend was also observed for the respons e to HCTZ monotherapy, and the combined assessment of HCTZ monotherapy with HCTZ add on from the opposite treatment arm (Figure 2 3). The association between the CACNA1C rs2238078 intronic SNP and BP response to atenolol monotherapy was consistent with ate nolol add on, where increasing copies of the minor allele were associated with greater BP response to therapy (Table 2 6). Overall, the combined assessment of atenolol monotherapy with atenolol add on from the opposite treatment arm was also significant ( Table 2 6). Also of note, CACNA1C rs2238078 was nominally associated with a directionally opposite effect on BP response

PAGE 30

30 to HCTZ monotherapy as well as the combined assessment of HCTZ monotherapy with HCTZ add on from the opposite treatment arm in whites, where increasing copies of the minor allele corresponded to lower BP response to HCTZ (Table 2 7). Finally, the rs11831085 intronic SNP of CACNA1C was also significantly associated with BP response to atenolol add on therapy in blacks. Increasing copies of the rs11831085 minor allele were associated with lower BP response to atenolol, an effect which was consistent between atenolol add on for diastolic and systolic BP responses and systolic BP response to atenolol monotherapy (Table 2 8). For the combin ed response assessment of atenolol monotherapy with atenolol add on from the opposite treatment arm, the association was highly significant for systolic BP response (p = 6.3x10 5 ), and nominally associated with diastolic BP (p = 0.017; Table 2 8, Figure 2 4 ). No associations meeting prespecified criteria for significance were observed for whites in CACNA1C or for either race group among other candidate genes included in the study analyses. A summary of the observed associations and their direction are de tailed in Table 2 12. However, additional associations showing a strong trend for significance are reported in Appendix A. Haplotype Analysis Haplotypes were imputed for PEAR blacks incorporating the 4 CACNA1C SNPs demonstrating significant associations w ith BP response. The initial haplotype analysis revealed a strong association for two relatively common 4 SNP haplotypes (Appendix A). Because the greatest amount of LD was observed between CACNA1C rs11831085 and rs12425032 haplotypes incorporating the two SNPs alone were also assigned Analysis of the 2 SNP haplotypes proved equally as informative as the original 4 SNP haplotype and more informative than the single SNP associations alone.

PAGE 31

31 Among PEAR blacks, i ncreasing copies of the common CACNA1C AT haplotype were associated with greater BP response to atenolol, whereas a directionally opposite association was observed for BP response to HCTZ (Table 2 10 ; Figure 2 5). Additionally, carriers of the alternate GC haplotype had a significantly lower BP response to atenolol (Table 2 1 1 ; Figure 2 6). Discussion The PEAR study provided a unique opportunity to assess the influence of genetic variation on BP response to two commonly used BP therapies in a racially diverse population of uncomplicated hyperte nsives. Our study demonstrated genetic variants within genes critical to the regulation of Ca 2+ signaling are significantly associated with BP response to atenolol and HCTZ. Interestingly, the strongest associations between genetic variation and BP respo nse observed in this study were among blacks, in response to atenolol therapy. A SNP previously identified as related to systolic BP in a large candidate gene cohort study was found to be nominally associated with BP response to atenolol monotherapy as wel l as atenolol monotherapy combined with atenolol add on from the opposite treatment arm in blacks. 29 The association in PEAR is directionally consistent with the allelic effect on systolic BP observed in the CHARGE cohort, where the minor allele is associated with a decrease in systolic BP. While this finding does not represent a true replication of the previously reported data, it certainly lends credence to the validity of the observed association. Three additional novel genetic associations with BP response were observed within CACNA1C the strongest of which was the intronic rs12425032. Associated with lower BP response to atenolol therapy in all response

PAGE 32

32 ass essments among blacks, CACNA1C rs12425032 was also associated with greater BP response to HCTZ in the same race group. Finally, the common CACNA1C AT haplotype comprised of the conserved rs11831085 and rs12425032 SNPs w as significantly associated with BP response to both atenolol and HCTZ in a directionally opposite manner. 2+ signaling pathway we might expect to see directionally alternate BP response associations to functionally r elevant SNPs. adrenergic stimulation enhances t he probability of functional Ca v 1.2 channel activation and stimulates a shift of channels from a null gating inactivable pool to an activable pool. 35 I Ca,L associated with adrenergic stimulation. Conversely, HCTZ may enhance the voltage dependent inactivation of Ca v 1.2 channels via K Ca channel opening which hyperpolarize s the cell membrane 25 A SNP within a conserved intronic region of CACNA1C may confer lower transcriptional efficiency or stability, or code for an alternative splice site. Such a change could manifest in a decreased I Ca,L which could contribute to a lower BP response to A decrease in I Ca,L might also exacerbate the BP response observed to HCTZ, if membrane hyperpolarization effectively blunted a Ca 2+ current that was comparatively weaker among variant carriers. The effect of genetic variation within the Ca 2+ signaling pathway on BP response in t he current study appears be to context dependent. It is generally accepted that single gene effects on antihypertensive drug response tend to be small, and it may be

PAGE 33

33 respon se to a therapy which does not directly target the underlying genetic mechanisms 5 Hence, the lack of a consistent genetic effect among PEAR whites may be attributable to an overriding BP response that is driven by a decrease in renin secretion and influe nced less by any effect on Ca 2+ signaling. The PEAR study recruited uncomplicated hypertensives, those subjects without concomitant disease states which might confound the initial response observed to antihypertensive therapies, and included a temporary wa shout period prior to randomization. 30 Reported here are novel associations between variants within genes critical to the pathway of Ca 2+ signaling and BP response to two common antihypertensive agents. Despite careful study design, we do accept the limitation that we may have failed to capture associations between genetic variation and BP response due to a limited power to detect effects at lower minor allele frequ encies. Our approach to hypothesis testing utilizing a relatively liberal p value initially was an attempt to balance the detection of SNPs with small effects while minimizing the likelihood of Type I error by testing associations between separate populat ions one receiving a drug as monotherapy and another receiving a drug as add on therapy in the opposite treatment arm Additional testing for BP associations in the opposite treatment arm as well as accounting for a directional effect yielded p values of ten more stringent than a Bonferroni corrected analysis. Overall, the associations reported here may provide some insight into the heterogeneity of BP response observed to two commonly used antihypertensive therapies in a racially diverse population with mild to moderate hypertension. Further

PAGE 34

34 study is warranted, however, as the associations reported here have yet to be replicated in an independent study population.

PAGE 35

35 Table 2 1 Illumina BeadChip SNP Coverage and p v alue thresholds Gene Human CVD Human Omni1 Quad Total Initial p value Additional p value Direction of effect Product p value ATP2A2 11 11 19 0.01 0.05 0.5 2.5x10 4 CACNA1C 258 279 428 0.005 0.05 0.5 1.25x10 4 CACNA1D 0 142 142 0.005 0.05 0.5 1.25x10 4 C ACNB2 0 242 242 0.005 0.05 0.5 1.25x10 4 CASQ2 17 58 66 0.01 0.05 0.5 2.5x10 4 KCNMB1 19 40 54 0.01 0.05 0.5 2.5x10 4 PLN 3 11 14 0.01 0.05 0.5 2.5x10 4 RYR2 324 459 639 0.005 0.05 0.5 1.25x10 4 SLC8A1 0 303 303 0.005 0.05 0.5 1.25x10 4 Direction as hypothesized for additional p value Table 2 2. Power to detect diastolic BP differences across MAF ranges by race, PEAR Black n=304 5% 10% 25% 5% 10% 25% 4 mmHg DBP 47% 80% 99% 37% 73% >99% 7 mmHg DBP 96% >99% >99% 94% 99% >99% White n=461 5% 10% 25% 5% 10% 25% 4 mmHg DBP 70% 95% >99% 61% 93% 99% 7 mmHg DBP >99% >99% >99% 96 % >99% >99%

PAGE 36

36 Table 2 3 PEAR Baseline Demographics Characteristic All (n=76 5 ) ATEN (n=385 ) HCTZ (n=380 ) p value Age 48.8 9.2 48.6 9.2 49.0 9.2 0.5 8 Sex (% female) 403 (52.7) 216 (56.1 ) 187 (49.2 ) 0.06 Race Black (%) 304 (39.7) 152 (39.5) 152 (40. 0) White (%) 461 (60.3) 233 (60.5) 228 (60.0) 0.88 Home SBP (mm Hg) 145.8 10.3 145.0 9.9 146.6 10.7 0.03 Home DBP (mm Hg) 93.7 6.0 93.3 5.9 94.2 6.0 0.03 Home heart rate (bpm) 77.5 9.5 77.8 9.4 77.2 9.7 0.37 History of hyperten sion Duration of hypertension (yr) 6.5 7.2 6.7 7.0 6.4 7.2 0. 42 Family history of hypertension (%) 586 (76.4) 298 (77.4 ) 286 (75.5 ) 0.77 Never taken an antihypertensive drug (%) 93 (12.1) 47 (12.2) 46 (12.0) 0.95 Taking antihypertensi ve drug at entry (%) 672 (87.8 ) 338 ( 87.8 ) 334 (87.8 ) 0.96 Smoking status Current smoker (%) 111 (14.5) 50 (13.0) 61 (16.0) 0.23 Ex smoker (%) 531 (82.7) 276 (84.7) 252 (80. 5 ) 0.17 BMI (kg/m 2 ) 30.8 5.5 30.8 5.9 30.8 5.1 0 .97 Waist circumference (cm) 97.8 13.3 97.6 13.1 98.1 13.5 0.57 Mean SD unless otherwise noted, p value by t test for continuous variables and 2 test for categorical variables between treatment groups.

PAGE 37

37 Table 2 4 Associations between CACNA1C rs2239101 genotype and blood pressure response to atenolol among PEAR blacks Response MAF Hmz. (mmHg) (mmHg) Minor Hmz. (mmHg) P Hmz. (mmHg) (mmHg) Minor Hmz. (mmHg) P Atenolol Monotherapy 0.05 3.8 8.0 0.04 12.0 12.1 0.026 Atenolol Add On 0.05 5.6 8.4 0.05 6.0 9.5 0.04 Ate nolol Monotherapy + Add On 0.05 4.7 8.2 0.0065 4.2 9.7 0.0036 A djusted for age, gender, and baseline BP Response to monot herapy + add on w as adjusted for BP at baseline in the monotherapy group and BP at the first response assessment for the add on therapy group as well as controlling for the order of drug initiation

PAGE 38

38 Table 2 5 Associations between CACNA1C rs12425032 genotype and blood pressure response to atenolol and HCTZ among PEAR blacks Response MAF Hmz. (mmHg) et. (mmHg) Hmz. (mmHg) P Hmz. (mmHg) (mmHg) Hmz. (mmHg) P Atenolol Monotherapy 0.25 5.2 3.2 2.1 0.046 5.1 1.2 0.0 0.03 Atenolol Add On 0.25 7.1 5.9 0.68 0.003 8.3 6.0 0.9 9.6x10 4 Atenolol Monotherap y + Add On 0.25 6.2 4.5 0.7 6.3x10 4 6.7 3.5 0.4 3.2x10 4 HCTZ Monotherapy 0.25 6.8 6.5 10.7 0.59 11.1 11.3 19.8 0.38 HCTZ Add On 0.25 6.1 8.0 11.8 0.05 10.4 14.9 20.2 0.007 HCTZ Monotherapy + Add On 0.25 6.4 7.2 11.2 0.069 10.8 13 .0 20.0 0.014 A djusted for age, gender, and baseline BP Response to monotherapy + add on w as adjusted for BP at baseline in the monotherapy group and BP at the first response assessment for the add on therapy group as well as controlling for the order of drug initiati on

PAGE 39

39 Table 2 6. Associations between CACNA1C rs2238078 genotype and blood pressure response to atenolol among PEAR blacks Response MAF Hmz. (mmHg) (mmHg) Minor Hmz. (mmHg) P Hmz. (mmHg) (mmHg) Mino r Hmz. (mmHg) P Atenolol Monotherapy 0.13 3.7 5.9 0.013 2.1 7.1 0.002 Atenolol Add On 0.13 5.5 8.0 0.016 6.3 8.6 0.018 Atenolol Monotherapy + Add On 0.13 4.6 7.1 0.001 4.1 8.0 2.8x10 4 A djusted for age, gender, and baseline BP Response to monotherapy + add on w as adjusted for BP at baseline in the monotherapy group and BP at the first response assessment for the add on therapy group as well as controlling for the order of drug initiation DBP = diastolic BP; SBP = systolic BP; Hmz. = homozygote; Het. = heterozygote.

PAGE 40

40 Table 2 7. Associations between CACNA1C rs2238078 genotype and blood pressure response to HCTZ among PEAR whites Response MAF Hmz. (mmHg) (mmHg) Minor Hmz. (mmHg) P Hmz. (mmHg) (mmHg) Minor Hmz. (mmHg) P HCTZ Monotherapy 0.39 4.9 4.1 2.6 0.01 8.0 8.1 5.6 0.12 HCTZ Add On 0.39 3.6 2.5 2.3 0.34 7.0 5.5 4.1 0.07 HC TZ Monotherapy + Add On 0.39 4.3 3.3 2.4 0.008 7.5 6.8 4.8 0.02 A djusted for age, gender, and baseline BP Response to monotherapy + add on w as adjusted for BP at baseline in the monotherapy group and BP at the first response assessment for the add on therapy group as well as controlling for the order of drug initiation DBP = diastolic BP; SBP = systolic BP; Hmz. = homozygote; Het. = heterozygote.

PAGE 41

41 Table 2 8 Associations between CACNA1C rs11831085 genotype and blood pressure response to a tenolol among PEAR blacks Response MAF Hmz. (mmHg) (mmHg) Minor Hmz. (mmHg) P Hmz. (mmHg) (mmHg) Minor Hmz. (mmHg) P Atenolol Monotherapy 0.32 4.1 4.6 2.1 0.35 4.1 3.9 5.2 0.006 Atenolol Add On 0.32 7.2 5.3 4.1 0.007 8.8 5.1 4.6 0.001 Atenolol Monotherapy + Add On 0.32 5.8 4.9 3.1 0.017 6.7 4.4 0.2 6.3x10 5 A djusted for age, gender, and baseline BP Response to monotherapy + add on w as adjusted for BP at baseline in the monotherapy group and BP at the first response asses sment for the add on therapy group as well as controlling for the order of drug initiation DBP = diastolic BP; SBP = systolic BP; Hmz. = homozygote; Het. = heterozygote. Table 2 9. CACNA1C rs11831085, rs12425032 haplotype frequency among PEAR blacks H aplotype Frequency (%) GT 15.6 GC 16.8 AT 59.7 AC 7.8

PAGE 42

42 Table 2 10 Associations between CACNA1C rs11831085, rs12425032 AT haplotype and blood pressure response among PEAR blacks Response Frequency 0 copy (mmHg) copy (mmHg) copies (mmHg) P copy (mmHg) copy (mmHg) copies (mmHg) P Atenolol Monotherapy 0.60 2.0 4.6 4.6 0.14 3.6 4.4 3.9 0.014 Atenolol Add On 0.60 2.8 5.6 8.0 1.3x10 4 3.9 5.6 9.4 3.8x10 4 Atenolol Monotherapy + Add On 0.60 2.4 5.0 6.5 3.1x10 4 0.1 4.9 7.1 6.9x10 5 HCTZ Monotherapy 0.60 8.9 7.0 6.3 0.31 15.5 12.1 10.1 0.14 HCTZ Add On 0.60 10.0 7.4 5.1 0.028 19.0 12.5 10.0 0.024 HCTZ Monotherapy + Add On 0.60 9.5 7.2 5.8 0.02 17.2 12.3 10.0 0.0095 A djusted for age, gender, and baseline BP Response to monotherapy + add on w as adjusted for BP at baseline in the monotherapy group and BP at the first response assessment for the add on therapy group as well as controlling for the order of drug initiati on

PAGE 43

43 Table 2 1 1 Associations between CACNA1C rs11831085, rs12425032 GC haplotype and blood pressure response among PEAR blacks Response Frequency 0 copy (mmHg) copy (mmHg) copies (mmHg) P copy (mmHg) copy (mmHg ) copies (mmHg) P Atenolol Monotherapy 0.17 4.8 3.1 0.08 4.8 0.0 0.0027 Atenolol Add On 0.17 6.6 5.2 0.07 7.8 4.4 0.002 Atenolol Monotherapy + Add On 0.17 5.7 4.1 0.02 6.3 2.1 4.3x10 5 A djusted for age, gender, and base line BP Response to monotherapy + add on w as adjusted for BP at baseline in the monotherapy group and BP at the first response assessment for the add on therapy group as well as controlling for the order of drug initiation

PAGE 44

44 Table 2 1 2 Summary of signi ficant CACNA1C SNP associations with BP response Treatment SNP Race MAF Allele BP response* Atenolol rs2239101 Black 0.05 T/ C rs2238078 Black 0.13 G /T rs12425032 Black 0.25 C /T rs11831085 Black 0.32 A/ G HCTZ rs12425032 Black 0.25 T/ C rs2238078 White 0.39 G /T Minor allele effect on BP response minor allele shown in bold Figure 2 1. LD structure o f CACNA1C significant associations within PEAR blacks, r 2 values shown

PAGE 45

45 Figure 2 2 Blood pressure response to the combined assessment of atenolol monotherapy with atenolol add on from the opposite treatment arm by CACNA1C rs2239101 genotype among PEAR blacks

PAGE 46

46 Figure 2 3. Blood pressure response to the combined assessment of monotherapy with add on from the opposite treatment arm by CACNA1C rs12425032 genotype among PEAR blacks

PAGE 47

47 Figure 2 4 Blood pressure response to the combined assessment of atenolol monotherapy with atenolol add on from the opposite treatment arm by CACNA1C rs11831085 genotype

PAGE 48

48 Figure 2 5. Blood pressure response to the combined assessment of monotherapy with add on from the opposite treatment arm by CACNA1C AT haplot ype among PEAR blacks

PAGE 49

49 Figure 2 6. Blood pressure response to the combined assessment of atenolol monotherapy with atenolol add on from the opposite treatment arm by CACNA1C GC haplotype among PEAR blacks

PAGE 50

50 CHAPTER 3 ASSOCIATION OF CA LCIUM SIGNALING PATH WAY VARIATION WITH CARDIOVASCULAR OUTCO MES IN THE INTERNATIONAL VERAPA MIL SR TRANDOLAPRIL STUDY GENETIC SUBSTUDY (IN VEST GENES) Introduction Genetic variation within the Ca 2+ signaling pathway has previously been associated with adver se outcomes dependent upon treatment strategy within the International Verapamil SR Trandolapril Study Genetic Substudy (INVEST GENES ) population (Chapter 1) INVEST was a large, multicenter, randomized hypertensive treatment outcomes trial utilizing both CCB and B treatment strategies. The INVEST GENES cohort provides a rare opportunity to investigate associations between variation with in genes critical to the hypertensive treatment response and adverse cardiovascular outcomes among high risk hypertensi ve patients with coronary artery disease (CAD) Methods INVEST GENES Clinical Cohort INVEST evaluated BP response and adverse outcomes occurring with either an atenolol based CCB treatment strategy in 22,576 patients with docum ented CAD and hypertension. 36 The design, protocol, and primary outcome data have been published in detail elsewhere. 36 37 To summarize the protocol required patients to be seen at baseline, 6, 12, 18 and 24 weeks, and then every 6 months thereafter until 2 years after the last patient was enrolled. At each study visit, patients had BP and heart rate measured, clinical assessment performed and additional antihypertensive medications added as needed to meet the Joint National Committee VI BP goals. 38 Blood pressure control and cardiovascular outcomes were similar between the tr eatment strategies in the main trial. 36

PAGE 51

51 The INVEST GENES cohort consisted of 5,979 patients from 184 sites in the United States and Puerto Rico who provided DNA samp les and additional written informed consent for genetic studies. Using the 5,979 patients with genetic samples, a nested case control c ohort was studied consisting of patients who experienced the primary outcome (first occurrence of all cause death, nonfa tal myocardial infarction, or nonfatal stroke) matched to patients who did not have a primary outcome event during study follow u p. Controls were matched to cases by age (in decades), sex, and race/ethnicity in a ratio of approximately 4:1. The nested ca se control approach has previously been documented to yield similar results to analyses of the entire genetic cohort. 23 39 Secondary outcome measures were defined as the individual components of the primary outcome. 36 Genotyp e Determination Buccal tissue samples were obtai ned by mouthwash and genomic DNA was isolated using the Gentra Systems PureGene kit. INVEST GENES genotypes for Ca 2+ signaling pathway candidate genes were obtained from the Illumina HumanCVD BeadCh i p (Table 2 1). 250ng to 500ng of genomic DNA from eac h patient of the INVE ST GENES case control cohort was hybridized to a HumanCVD BeadChip custom SNP array (IBC array, version 2; Illumina Inc) on Illumina according to manufacturer specified protocols. A description of the strategy for ge ne and SNP selection on the Illumina HumanCVD BeadChip is provided in detail elsewhere. 31 SNP interrogation was performed with Illumina II chemistry. Genotypes were called using Genom eStudio Software version 2011 and the Genotyping Module version 1.9 calling algorithm (Illumina San Diego, CA). The putative functionality of

PAGE 52

52 SNPs meeting criteria for significance was assessed with PupaSuite 3.1 ( http://pupasuite.bioinfo.cipf.es ). A Principal Component Analysis (PCA) was performed in all subjects using an LD pruned dataset using the EIGENSTRAT method implemented through JMP Genomic s version 5.0 (SAS Cary, NC). INVEST included primarily whi te, Hispanic, and black race/ethnic groups, determined by patient report with interaction by the study investigator. Race/ethnic groups were confirmed with PCA clustering results. If race/ethnic category disagreed strongly with PCA, patients were re cate gorized to reflect the PCA result, considered to better reflect genetic ancestry. PCA was then performed in each race/ethnic group. To adjust for ancestry in the final analyses, principal components 1 through 3 were included as covariates for whites and Hispanics and principal components 1 through 4 were included as covariates for blacks as these provided the best separation of ancestry clusters and were chosen as covariates in analysis to adjust for ancestry. Statistic al Analyses Hardy Weinberg equilib rium was tested for each PCA defined racial/ethnic group using 2 analyses. All statistical analyses were performed using SAS version 9.2 (Cary, NC ) or JMP Genomics 5 Baseline characteristics were compared by treatment strategy using 2 or analysis of variance, as appropriate. Primary and secondary outcome adjusted odds ratios (ORs) and 95% confidence intervals (CIs) were calculated using logistic regression by PCA defined race and adjusted for prespecified baseline covariates that wer e found to infl uence prognosis. These baseline covariates included age, sex, INVEST treatment strategy, previous myocardial infarction history of

PAGE 53

53 heart failure and diabetes. Significant p values for the main effect primary outcome or secondary outcome of death were p = 2.5x10 4 for CACNA1C and RYR2 and p = 0.005 for ATP2A2 CASQ2 KCNMB1 and PLN A strict Bonferroni correction was not used for reasons described in Chapter 2. For analysis of pharmacogenetic interactions, a p = 0.005 was considered significant for CA CNA1C and RYR2 and the threshold set for ATP2A2 CASQ2 KCNMB1 and PLN was p = 0.01 For those SNPs demonstrating significant associations with BP response in PEAR, a nominal association of p < 0.05 was considered significant for main effect, secondary o utcome and pharmacogenetic interaction analyses. SNPs previously represented in the literature for associations with cardiovascular outcomes had a p < 0.05 considered significant. Power calculations were performed with QUANTO version 1.2 40 Within the case control co hort we had 95 % power to detect an OR of 2. 5 0 5 and 80 % power to detect an OR of 2.5x10 4 for the primary outcome (Table 3 1) LD was assessed with INVEST GENES genotypes in the HapMap PH ASE format (h ttp://stephenslab.uchi cago.edu/software.html) uploaded to Haploview 4.2. 33 To assess LD with SNPs not included on the Human CVD BeadChip Seattle SNPs Genome Variation Server (http://gvs.gs.washington.edu/GVS/index.jsp) was utilized incorporating genotypes from HapMap populations. Results Baseline characteristics and medical history for the INVEST GENES case control cohort are shown in Table 3 2. Characteristics of the case control cohort are consistent with the original INVEST GENES cohort, consisting of an elderly and racially diverse hypertensive CAD population. Patients in the verapamil based CCB treatment strategy

PAGE 54

54 and ate nolol based B treatment strategy were similar in baseline characteristics, medical history and nonstudy medication use (Table 3 2). Genotyping and Quality Control Of the INVEST GENES subjects genotyped p atients were excluded if sample genotype call rates were below 95% and SNPs were excluded if genotype call rates were below 90%. 87 blind duplicates were included in genotyping and had a concordance rate of 99.997 %. Gender was confirmed from X chromosome genotype data, and those who were discordant were excluded. Cryptic relatedness was estimated by pairwise identity by descent (IBD) analysis implemented using PLINK ( http://pngu.mgh.harvard.edu/purcell/plink/ ). Forty one pairs of samples were identified as f irst degree relatives, these individuals were kept for the analysis. Heterozygosity was assessed using PLINK, by estimating the inbreeding coefficient, F. Six subjects had F values > 4 standard deviations from the mean. One of these subjects also had a high missing genotype rate of > 4% and this subject was excluded. Outliers in the by race/ethnic group Principal Component Analysis were also removed ( n =4). The final INVEST datas et consisted of 2,214 subjects. Genotype data were complete for 1345 patie n ts of the case control cohort, comprised of 269 patients who experienced a primary outcome event during study follow up and 1 076 patients who did not. Genetic and Pharmacogenetic Associations W ith Adverse Cardiovascular Outcomes CACNA1C The CACNA1C rs2 239101 and rs12425032 polymorphisms significantly associated with BP response to atenolol among blacks in the PEAR population were also associated with adverse cardiovascular outcomes within I NVEST GENES The

PAGE 55

55 rs2239101 SNP previously identified as associa ted with systolic BP in a candidate gene analysis of BP and hypertension from CHARGE and also associated with greater BP response to atenolol among PEAR blacks ( Table 2 4 ) was associated with an increased risk of death among whites within the INVEST GENES case control cohort (Table 3 3) 29 The association within the INVEST GENES case control cohort was consistent among whites randomized to both CCB and B treatment strategies ( main effect p = 0.004). No similar associations were observed among black or Hispanic racial/ethnic groups. The CACNA1C rs12425032 SNP significantly associated with lower BP response to atenolol and greater BP response to HCTZ a mong PEAR blacks also demonstrated a nominally significant association with the primary outcome among INVEST GENES case control cohort blacks. CACNA1C rs12425032 black minor allele carriers randomized to the CCB treatment strategy were at an almost 10 fol d greater risk of experiencing the primary outcome (OR, 9.97; 95% CI, 1.33 74.3; p = 0.025; Appendix B ). Consistent with and driven by these data, the rs12425032 SNP also demonstrated a significant pharmacogenetic association with the primary outcome amon g INVEST GENES case control cohort blacks (Figure 3 1 ; SNP*treatment interaction p = 0.005 ) Minor allele carriers randomized to the CCB treatment strategy were at a comparatively increased risk for experiencing the primary outcome relative to rs12425032 minor allele carriers randomized to the B treatment strategy (Table 3 4). No associations were observed among white or Hispanic racial/ethnic groups. Finally, the CACNA1C rs2299567 intronic SNP associated with greater BP response to HCTZ among minor allele white carriers in PEAR (Appendix A) was also

PAGE 56

56 associated also associated with a decreased risk for the primary outcome among INVEST GENES case control cohort blacks independent of treatment strategy (OR, 0.17; 95% CI, 0.05 0.57; p = 0.004 ; Table 3 3 ). The rs2299657 SNP was also associated wit h increased odds of death among INVEST GENES case control cohort whites randomized to the CCB treatment strategy (OR, 2.53; 95% CI, 1.56 4.10; p = 1.6x10 4 ; Appendix B). No associations were observed between adverse outcomes and CACNA1C rs2299657 genotype among Hispanics. CASQ2 Among INVEST GENES case control cohort whites, minor allele carriers of the conserved intronic CASQ2 rs3811003 SNP demonstrated a significant interaction with INVEST treatment strategy for both the primary outcome and death (SNP*tr eatment interaction p = 0.003, Figure 3 2). Among blacks, the rs3811003 minor allele was nominally associated with an increased risk for the outcome of death, independent of treatment strategy (OR death, 3.71; 95% CI, 1.20 11.4; p = 0.02; Table 3 3). An other intronic CASQ2 SNP in high LD with rs3811003 among INVEST GENES case control cohort whites, rs7355132 (r 2 = 0.57) also demonstrated a significant and directionally consistent pharmacogenetic interaction with treatment strategy for the outcome of deat h (p = 0.001; Table 3 4). An additional, nominally significant pharmacogenetic interaction with treatment strategy was also observed among blacks, where randomization to the CCB treatment strategy was associated with an increased risk for the primary outc ome, and a comparatively neutral effect was observed among those randomized to the B treatment strategy (p for interaction = 0.048; Table 3 4). The CASQ2 rs3811003 and rs7355132 SNPs among INVEST GENES case control cohort blacks share a lower degree of LD relative to whites (r 2 = 0.18). Notably, the

PAGE 57

57 pharmacogenetic associations observed for both CASQ2 rs3811003 and rs7355132 are directionally opposite between blacks and whites (Table 3 4) No associations between adverse outcomes and CASQ2 genotype were observed among Hispanics within the INVEST GENES case control cohort. KCNMB1 T he Lys65 variant (rs11739136) of KCNMB1 was observed to be associated with greater odds for the primary outcome among INVEST GENES case control cohort blacks independent of rand omized treatment strategy (OR, 5.58 ; 95% CI 1.6 6 18.7 ; p = 0.005 ; Table 3 3 ). An additional, nominal association for greater odds of death among Hispanics for carriers of the Lys65 variant randomized to the B treatment strategy was also observed (OR, 3. 6 5 ; 95% CI, 1.3 8 9. 67 ; p = 0.009 ; Appendix B ). No consistent associatio ns were observed among whites of the I NVEST GENES case control cohort. A nother KCNMB1 SNP, the conserved intronic rs2075612 wa s significantly associated with the outcome of death indep endent of treatment strategy among INVES T GENES case control cohort blacks (OR, 4.98 ; 95% CI 1. 67 1 4 8 ; p = 0.00 3 ; Table 3 3) There was no LD (r 2 = 0.0) observed between KCNMB1 rs11739136 and rs2075612 in this race group. Additionally, a nominally signi ficant pharmacogenetic association was observed with the primary outcome among Hispanics wherein a lower risk was observed for patients randomized to the CCB treatment strategy relative to the B treatment strategy ( p = 0.04; Appendix B). There were no a ssociation s between rs2075612 and adverse cardiovascular outcomes among whites Finally, Hispanic carriers of the KCNMB1 Leu110 variant (rs2301149) randomized to the B trea tment strategy were at an increased risk of experiencing the primary outcome (OR, 3 45 ; 95% CI, 1.2 0 9.96 ; p = 0.02 ; Appendix B ). In an analysis of

PAGE 58

58 combined race groups in the full INVEST GENES cohort, differences by genotype among patients randomized to the B treatment strategy were not observed 23 Although, when treatment strategy was not considered, Leu110 carriers had a decreased risk of ex periencing the primary outcome. No additional associations were observed among the other racial/eth nic groups for the KCNMB1 Val110Leu polymorphism in the current analyses. RYR2 The intronic RYR2 rs961121 polymorphism was significantly associated with a more than 3 fold increased risk for death among white minor allele carriers independent of treatment strategy (OR, 3.26; 95% CI, 1.89 5.61; p = 2.1 x 10 5 ; Table 3 3). Separated by treatment strategy, analyses reveal approximately equivalent risk among whites in each treatment strategy (p = 0.003 ; Appendix B ). No additional associations were observed am ong the black or Hispanic racial/ethnic groups. There were several significant pharmacogenetic interactions observed within the cardiac ryanodine receptor gene. A relatively protective effect of CC B treatment strategy against the outcome death was observ ed among white minor allele carriers of the rs2485584 SNP, whereas those randomized to the B treatment strategy experienced a neutral to increased risk for this outcome (p = 0.004; Table 3 4). No consistent associations were observed among the other raci al/ethnic groups. Among the Hispanic ethnic group, the RYR2 rs16834782 SNP was associated with an increased risk of death among carriers of the minor allele randomized to the CCB treatment strategy, whereas randomization to the B treatment strategy was as sociated with a neutral to decreased risk (p = 0.003; Table 3 4). The opposite was true for Hispanic minor allele carriers of RYR2 rs3766884, for whom the CCB treatment

PAGE 59

59 strategy was associated with an 83% reduced risk of death, whereas minor allele carrie rs randomized to the B treatment strategy were at a comparatively increased risk (p = 0.005; Table 3 4). No similar associations for either RYR 2 SNPs were observed among blacks or whites in the INVEST GENES case control cohort. PLN The intronic rs9489438 SNP within the gene encoding phospholamban demonstrated a significant pharmacogenetic interaction with treatment strategy for the primary outcome among INVEST G ENES case control cohort whites (p = 0.00 7 ; Table 3 4). Additionally, a nominal association with decreased risk for death among Hispanics randomized to the CC B treatment strategy was also observed (p = 0.0 2 ; Appendix B). The rs9489438 SNP of PLN is in complete LD (r 2 = 1.0) with a nonsynonymous SNP of chromosome 6 open reading frame 204 ( C6orf204 ) encoding an amino ac id change of proline to threonine. Found in the same chromosomal region as PLN genetic variation within C6orf204 has been identified as associated with QRS interval duration in a genome wide meta analysis in individuals of European descent. 41 Discussion CACNA1C encodes the 1C pore forming subunit of Ca v 1.2 channels the drug target of CCBs The present study has identified several CACNA1C variant alleles that associate with an increased risk for adverse cardiovascular outcomes among INVEST GENES case control cohort blacks and whites randomized to the CCB treatment strategy. The identification of the CACNA1C rs2239101 variant as a risk allele among INVEST GENES case control cohort whites is counter to previous associations indicating a phenotype of lower systolic BP. CACN A1C rs2239101 has previously been

PAGE 60

60 associated with lower systolic BP in a large cohort candidate gene study as well as nominally associated with a directionally consistent effect among PEAR blacks, whom demonstrated greater BP response to atenolol (Chapter 2) 29 The rs12425032 SNP of CACNA1C demonstrated a significant interaction with treatment strategy for the primary outcome among INVEST GENES case control cohor t blacks, where minor allele carriers randomized to the CCB treatment strategy demonstrated a significantly increased risk for the primary outcome. Although the rs12425032 minor allele was significantly associated with lower BP response to atenolol among PEAR blacks, within the INVEST GENES case control cohort, minor allele carriers randomized to the B treatment strategy had a comparatively neutral risk for experiencing the primary outcome. C alsequestrin 2, a high capacity Ca 2+ storage protein found in t he sarcoplasmic reticulum of cardiac muscle is encoded by CASQ2 Genetic variants within the coding sequence of CASQ2 have previously been associated with catecholamine induced polymorphic ventricular tachycardia, a potentially fatal arrhythmia 42 44 The CASQ2 rs3811003 minor allele associated with decreased odds of adverse cardiovascular outcomes among whites randomiz ed to the CCB treatment strategy in the INVEST GENES case control cohort is in high LD with two CASQ2 SNP s previously associated with an increased risk of sudden cardiac death among patients from a separate cohort with coronary artery disease (rs7536370 r 2 = 0.96; rs3010396 r 2 = 0.47) 45 Th e pharmacogenetic intera ction between CASQ2 rs3811003 and INVEST GENES treatment strategy for the primary outcome as well as death indicates a comparative ly increased risk for adverse outcomes among white minor allele carriers randomized to

PAGE 61

61 the B treatment strategy. Within the INVEST GENES population it appears the CCB treatment strategy modulates the risk associated with the rs3811003 minor allele in this race group. Several significant associations were observed between SNPs present in KCNMB1 and adverse cardiovascular outco mes in the INVEST GENES case control cohort A prior INVEST GENES case control cohort investigation of the KCNMB1 gene revealed no significant difference in the occurrence of the primary outcome by codon 65 genotype. 23 The current analysis performed by PCA defined race revealed an over 5 fold increased risk for the primary outcome among black rs11739136 (Glu65Lys) minor allele carriers. An additional nominally associated risk for increased odds of death was also observed among Hispanics randomized to the B treatment strategy. The KCNMB1 Lys65 allele has previously been associated with a protective effect against more severe phenotypes of diastolic hypertension, and functional data support the Lys65 variant as a gain of function to the K Ca channel, result ing in a negative feedback loop with enhanced efficiency. 22 Additional evidence indicates a Lys65 allelic dose dependent lowering of both systolic and diastolic BP in men. 46 Despite these data indicating a beneficial effect of the Lys65 allele on hypertension phenotypes, outcomes data within the black and Hispan ic racial/ethnic groups of the INVEST GENES case control cohort indicate the Lys65 may be a risk allele for adverse cardiovascular outcomes. Among INVEST GENES case control cohort Hispanics, Leu110 variant allele carriers randomized to the B treatment strategy were at an increased risk of experiencing the primary outcome. Previously genotyped in the entire INVEST GENES

PAGE 62

62 cohort (n = 5486), Leu110 variant carriers among combined racial/ethnic groups were found to have a reduced risk of the primary outcome an association which was more pronounced among those subjects randomized to the CCB treatment strategy. 23 The current finding is perhaps consistent with the previous subgroup analysis, which indicated a neutral risk among those subjects randomized to the B treatment strategy. No additional associations between Val110Leu genotype and adverse cardiovascular outcomes were observed among other racial groups of t he INVEST GENES case control cohort. Encoded by PLN phospholamban is a protein inhibitor of the cardiac sarcoplasmic reticulum Ca 2+ ATPase responsible for reuptake of Ca 2+ into the organelle. Protein kinase A phosphorylation of the phospholamban relieve s the inhibitory effect of the protein, which thereby regulates both the rate of cardiomyocyte relaxation a nd Ca 2+ storage within the sarcoplasmic reticulum. 47 The P LN rs9489438 SNP in complete LD with a non synonymous SNP of C6orf204 is also in moderate LD (r 2 = 0.37) with several additional SNPs associated with higher resting heart rate and greater LV internal diastolic dimensions in two separate genome wide associa tion analys e s performed in individuals of European descent. 48 49 The current association between PLN rs948943 8 and treatment strategy in the INVEST GENES case control cohort demonstrating lower risk of the B treatment strategy among whites adds to an accumulating wealth of associations recently reported between the chromosome 6q22 locus and cardiovascular phenot ypes. Finally, we report herein several novel associations between intronic SNPs of the c ardiac ryanodine receptor gene and adverse cardiovascular outcomes. Of interest, the

PAGE 63

63 RYR2 rs961121 minor allele was identified as a risk allele for the outcome of dea th among whites, an effect which is consistent between INVEST GENES treatment strategies. Most of the associations reported in one racial/ethnic group of the INVEST GENES case control cohort failed to reach nominal significance among the other populations represented. There may be no consistent associations within the alternate populations, or there may be a lack of power to detect the genetic effect due to differing sample sizes. Additionally, the associations reported herein may represent tagSNPs for f unctional SNPs that would not be observed due to the difference in LD between the racial/ethnic populations studied. Furthermore, despite the compelling nature of several associations reflecting the significance of previously reported yet alternate cardio vascular phenotypes, the associations reported here have yet to be replicated in a separate hypertension outcomes cohort. Overall, we have reported herein several novel associations between genetic variation within genes essential in the response to the a ntihypertensive treatment strategies utilized in this large, randomized hypertensive treatment outcomes trial and adverse cardiovascular outcomes.

PAGE 64

64 T able 3 1 Power to detect differences across MAF, INVEST GENES MAF OR INVEST GENES n=1 345 Whites n= 7 95 0.005 0.00025 0.05 0.005 0.00025 0.05 1.5 16% 3% 5 % 11% 2% 2.0 68% 35% 7% 51% 20% 2.5 95% 80% 8% 86% 58% 3.0 >99% 97% 10% 97% 85% 3.5 >99% >99% 13% >99% 97% 0.25 1.5 53% 21% 15% 37% 12% 2.0 98% 89% 36% 92% 71% 2.5 >99% >99% 57% >99% 97% 3.0 >99% >99% 74% >99% >99% 3.5 >99% >99% 85% >99% >99%

PAGE 65

65 Table 3 2. INVEST GENES baseline characteristics Characteristic INVEST GENES cohort (n=5,598) Case control cohort (n = 1, 345 ) CCB (n = 679 ) BB (n = 666 ) p value Demographics and social history Age, mean (SD), years 66.1 9.6 70.8 9.5 70.3 9.5 71.3 9.5 0.07 Women 3090 ( 55.2 ) 675 (50.2) 345 (51.1) 330 (48.9) 0.64 Race/ethnicity White 2315 (41.4) 795 (59.1) 398 (58.6) 397 (59.6) Black 657 (11.7) 170 (12.6) 82 (12.1) 88 (13.2) Hispanic 2589 (46.3) 380 (28.3) 199 (29.3) 181 (27.2) 0.62 BMI, mean (SD), kg/m2 29.4 5.5 28.6 5.4 28.6 5.3 28.7 5.4 0.77 BP, mean (SD), mmHg Systolic 148.0 18.4 148.9 18.3 148.0 19.0 149.8 17.5 0.07 Diastolic 85.4 10.7 83.3 10.5 83.1 10.5 83.5 10.4 0.43 Heart rate (beats per min) 74.8 9.6 74.9 9.3 75.1 9.3 74.7 9.3 0.48 Smoking history 2319 (41.1) 572 (42.5) 295 (43.4) 277 (41.6) 0.49 Past Medical History Myocardial infarction 1317 (23.5) 376 (27.9) 180 (26.5) 196 (29.4) 0.23 Heart Failure (NYHA class I III) 193 (3.5) 78 (5.8) 42 (6.2) 36 (5.4) 0.54 Stroke or TIA 395 (7.1) 117 (8.7) 59 (8.7) 58 (8.7) 0.99 Arrhythmia 187 (7.0) 127 (9 .4) 71 (10.5) 56 (8.4) 0.19 Left Ventricular Hypertrophy 842 (15.0) 213 (15.8) 105 (15.5) 108 (16.2) 0.70 Peripheral Vascular Disease 620 (11.1) 159 (11.8) 79 (11.6) 80 (12.0) 0.83 Diabetes 1594 (28.5) 301 (22.4) 148 (21.8) 153 (22.9) 0.60 Hypercholoe sterolemia 3068 (54.8) 819 (60.9) 426 (62.7) 393 (59.0) 0.16 Renal Insufficiency 88 (1.6) 38 (2.8) 23 (3.4) 15 (2.3) 0.21 Cancer 230 (4.1) 81 (6.0) 39 (5.7) 42 (6.3) 0.66 Medication Aspirin/other antiplatelet agent 2590 (46.3) 781 (58.1) 402 (59.2 ) 378 (56.9) 0.39 Antidiabetic medication 1392 (24.9) 260 (19.3) 126 (18.6) 134 (20.1) 0.47 Lipid Lowering Drugs 2034 (36.3) 565 (42.0) 300 (44.2) 265 (39.8) 0.10 Nitrates 1568 (28.0) 366 (27.2) 183 (26.9) 183 (27.5) 0.83 Data are presented as no. (%) or mean SD. NYHA indicates New York Heart Association; TIA, transient ischemic attack. By t test for continuous variables and 2 test for categorical variables.

PAGE 66

66 Table 3 3 A ssociations between SNPs of the Ca 2+ signaling pathway and adverse card iovas cular outcomes in INVEST GENES a Gene SNP Race MAF Outcome OR (95% CI) p value CACNA1C *rs2239101 White 0.12 Death 2.00 (1.24 3.25) 0.004 CACNA1C *r s2299657 Black 0.29 PO 0.17 (0.05 0.57) 0.004 CASQ2 Black 0.56 Death 3.71 (1.20 11.4) 0.02 KCNMB1 rs11739136 Black 0.08 PO 5.58 (1.66 18.7) 0.005 KCNMB1 rs2075612 Black 0.46 Death 4.98 (1.67 14.8) 0.003 RYR2 rs961121 White 0.05 Death 3.26 (1.89 5.6 1) 2.1x10 5 a Adjusted for age, sex, history of MI, heart failure, diabetes, ancestry ; Dominant model ; *Previous association with BP response to atenolol or HCTZ in PEAR; PO = primary outcome

PAGE 67

67 Table 3 4. Significant SNP treatment inter actions a Gene SNP Race MAF Outcome OR (95% CI) CCB OR (95% CI) B p value CACNA1C rs12425032 Black 0.22 PO 6.03 (0.83 43.9) 0.26 (0.04 1.56) 0.005 CASQ2 rs3811003 White 0.31 PO 0.44 (0.26 0.76) 1.47 (0.82 2.64) 0.003 CASQ2 rs3811003 White 0.31 Death 0. 28 (0.14 0.60) 1.23 (0.55 2.78) 0.01 CASQ2 rs7355132 White 0.25 Death 0.35 (0.16 0.75) 2.20 (0.96 5.04) 0.001 CASQ2 rs7355132 White 0.25 PO 0.51 (0.29 0.89) 1.29 (0.73 2.26) 0.02 CASQ2 rs7355132 Black 0.19 PO 32.0 (2.19 468) 1.84 (0.42 8.05) 0.048 PLN rs9489438 White 0.25 PO 1.63 (0.96 2.76) 0.49 (0.26 0.91) 0.00 7 RYR2 rs2485584 White 0.47 Death 0. 40 (0. 20 0.7 9 ) 2. 19 (0. 79 6. 04 ) 0.00 4 RYR2 rs16834782 Hisp 0.16 Death 4.74 (1.85 12.1 ) 0.2 2 (0.04 1.13 ) 0.00 3 RYR2 rs3766884 Hisp 0.21 Death 0.17 (0.05 0.5 5) 1.8 0 (0.59 5.49 ) 0.005 a Shows the dominant model a djusted for age, sex, history of MI, heart failure, diabetes, ancestry ; strongest single associations shown in bold *Previous association with BP response to atenolol or HCTZ in PEAR; PO = primary outco me

PAGE 68

68 Figure 3 1. Primary outcome adjusted odds ratios (log scale shown) and 95% confidence intervals by treatment strategy among INVEST GENES case control cohort black CACNA1C rs12425032 minor allele carriers

PAGE 69

69 Figure 3 2 Primary outcome adjusted odds ratios a nd 95% confidence intervals by treatment strategy among INVEST GENES case control cohort white CASQ2 rs3811003 minor allele carriers

PAGE 70

70 CHAPTER 4 INFLUENCE OF THE RS2 357928 PROMOTER POLY MORPHISM ON THE EXPRESSION PROFILE O F THE CA V 1.2 BETA 2 SUBUNIT IN HUMAN VAS CULAR TISSUE Introduction Ca v 1.2 channels are the fundamental machinery in maintenance of vascular smooth muscle tone and myocardial contra ction. Data from genome wide association studies of BP and hypertension have i mplicated variation ( CACNB2 ) of the Ca v 1.2 channel as related to these cardiovascular phenotypes (Chapter 1) 19 29 Additionally, w e have shown CACNB2 to contain SNPs that also associate with adverse cardiovascular outcomes in a treatment specific manner within the INVEST GENES 20 The A allele of the CACNB2 rs2357928 SNP has been associated with a significantly greater risk for experiencing a primary outcome event in patients randomized to the 2. 43, p=0.001). Th is finding was not observed among patients randomized to the CCB treatment strategy (OR 0. 90, 95% CI: 0.67 1.20, p=0.47). Furthermore, w e have previously documented the rs2357928 promoter SNP of CACNB2 to be associated with decreased tran scriptional activity in a reporter gene assay 20 T he goal of the current study is to further investigate the underlying functional mechanism of th e clin ically associated CACNB2 rs2357928 promoter SNP, utiliz ing techniques in molecular biology to ascertain how this varia n t influences the expression profile of the Ca v 1.2 2 subunit in human vascular tissue. To achieve this end, a genetic database of subjects matched to vascular tissue samples and clinical data w as developed

PAGE 71

71 Methods Study Protocol The study utilized remnant specimens of vascular tissue that would normal ly be discarded after surgery from patients undergoing lower limb amputation or other vascular procedure at UF/Shands Hospital. Men and women aged greater than 21 years requiring surgery generating vascular tissue waste were enrolled at Shands at the Univ ersity of Florida (UF) in Gainesville, FL. The study was approved by the UF Institutional Review Board (IRB#579 2010) Study participants provided written informed consent following recommendation or consultation for surgery. B aseline data including age gender, weight and self reported race/ethnic ity was obtained. Additional also collected. Exclusion criteria included documented HIV, hepatitis B or C infection, or active infection within the tissue to be collected (i.e. gangrenous). During the surgical procedure, excised tissue was placed aside in a sterile environment prior to collection or immediately placed into sterile containers containing RNA late r (Ambion ) reagent for RNA preservation. A 5 to 10 mL blood sam ple was also collected by the anesthesiologist via a central line in surgery or by the surgeon via aspiration from removed limb. If a blood sample was unavailable for collection, a mouthwash sample w as collected for the purpose of extracting genetic material. Collected samples were transported to the UF Center for Pharmacogenomics Core laboratory, where tissue sam ples were weighed to ensure submersion in 5 to 10 times greater volume of RNA later reag ent Vascular tissue samples were then i ncubated at 2 8 o C at least overnight, but no more than 3 days, then removed from RNA later reagent and placed in a freezer at 80 o C

PAGE 72

72 DNA Genomic DNA was isolated from blood lymphocytes using a commercially availabl e kit (Qiagen DNA Blood Isolation Kit) according to the manufacturer specified protocol Genotype d etermination CACNB2 rs2357928 genotype was determined by TaqMan SNP genotyping assay. Allelic discrimination was performed with a ma de to order probe f or rs2357928 ( C__2740542_10 ) Genotype accuracy was verified by duplicate genotyping for study participants, and also included additional samples from a separate study population with known CACNB2 rs2357928 genotype. RNA 50 mg of arterial tissue was hom ogenized in 1 mLTRIzol (Ambion ) reagent on ice with the Fisher Scientific TissueMiser homogenizer. 200 L of chloroform was added following room temperature incubation. Subsequent centrifugation allowed for the separation of an aqueous phase containing RNA, an interphase containing DNA and a lower phenol chloroform phase containing protein. The RNA containing aqueous phase was processed using TRIzol Plus RNA Purification Kit (Ambion ) per manufacturer specified protocol. T he interphase and phenol chl oroform phase were further processed with 300 L of 100% ethanol prior to separation and stor age at 80 o C for downstream processing. Isolated RNA was quantified in duplicate on the NanoDrop ND 1000 Spectrophotometer (Thermo Scientific Inc Rockford, IL ). RT PCR 500 ng of RNA for n = 10 samples were converted to cDNA using the Applied Biosystems High Capacity RNA to cDNA Kit. 1 L 20x TaqMan CACNB2 Gene Expression Assay (Hs00167861_m1) was combined with 2 L cDNA, 10 L 2 X TaqMan Gene Expression Mas ter Mix and 7 L RNase free water to generate a 20 L

PAGE 73

73 reaction. N = 10 samples were analyzed in triplicate along with reactions for endogenous controls to 18S (Hs03003631_g1), B2M (Hs00984320_m1) and GAPDH (Hs99999905_m1) on a n ABI 7300 Real Time PCR Syst em (Applied Biosystems) Protein Proteins were precipitated with isopropanol from the phenol ethanol supernatant previously aliquoted from tissue homogenization in TRIzol reagent. The resultant protein pellet was washed with 0.3M guanidine hydrochloride 95% ethanol and resuspended in 1% SDS buffer N uclear extract 40 mg of arterial tissue was washed with PBS prior to processing with the NE PER Nuclear and Cytoplasmic Extraction Reagents (Thermo Scientific Inc) per manufacturer specified protocol. Halt TM Protease Inhibitor Cocktail and 0.5M EDTA were added to the 400 L CER I reagent used in sample homogenization. The cytoplasmic extract was removed following subsequent incubation and centrifugation steps with 22 L CER II reagent. The remaining insol uble pellet fraction was resuspended in NER reagent, and stored at 80 o C until use. Quantitation Protein concentrations were determined using the Pierce BCA Protein Assay kit. 1% SDS diluent was used to construct the standard curve using serial dilutions of bovine serum antigen per manufacturer instructions. Absorbance was read at 562nm on the Biotek Synergy TM HT (Biotek Instruments, Inc, Winooski, VT). All concentrations were measured in duplicate.

PAGE 74

74 Antibodies Anti CACNB2 mouse monoclonal antibody (a b54920, abcam ) was used at a dilution of 1:1000. Anti GAPDH rabbit polyclonal antibody (sc 257778, Santa Cruz Biotechnology, Inc.) was used as a loading control at a dilution of 1:1000, and was a gift from Dr. Jorg Bungert (Department of Biochemistry and Molecular Biology, University of Florida). Western blot Approximately 20 g protein extract were combined in a 2:1 volume of laemmli sample buffer (62.5 mM, 25% glycerol, 2% SDS, 0.01% Bromophenol Blue, 5% mercaptoethanol) and heated to 95 o C in a water bath for 5 minutes. Samples were loaded onto a 4 20 % Mini PROTEAN TGX TM p recast gel (Bio Rad Laboratories, Inc) with 1 X Tris/Glycine/SDS buffer (25 mM Tris, 192 mM glycine, 0.1% SDS, pH 8.3) and Bio Rad Precision Plus Protein TM standards were included a s the ladder. E lectrophoresis commenced at 120 V for 60 minutes. Proteins were then transferred onto a 0.45 m nitrocellulose membrane in 1 X Towbin buffer (25 mM Tris, 192 mM glycine, 20% methanol pH 8.6 ) under ice at 15 mA overnight. Following transfer the membranes were processed according to the instructions provided in the Pierce Fast Western Blot Kit. Membranes were washed in 1X Wash Buffer and incubated in a 1:1000 dilution of primary antibody for 30 minutes at room temperature with shaking. Th e membrane was then incubated with the Optimized HRP Reagent Working Dilution for 15 minutes at room temperature with shaking, and washed 3 x 5 minutes in 1X Wash Buffer. The SuperSignal West Pico Working Solution was added, and the membrane incubated for an additional 5 minutes. Bands were then imaged on Kodak X Omat LS film with a Fisher Biotech FB XC 810 Autoradiography

PAGE 75

75 cassette. Band intensity was calculated with Quantity One data Analysis Software (Bio Rad Laboratories, Inc). Electrophoretic m obi lit y shift a ssay ( EMSA ) DNA corresponding to the sequence 24 base pairs (bp) upstream and 25 bp downstream of each CACNB2 rs2357928 allele (A/G) and reverse complement were produced by Eurofins MWG Operon. Two sequences of single stranded oligonucleotides TAACA GTGAA TCCAGGATGC TGCT(G/A)TCAGC CTGTGCATTG TGAAGAAGGC the reverse complement ATTGTCACTT AGGTCCTACG ACGA ( C /T) AGTCG GACACGTAAC ACTTCTTCCG and a GCATCATCGC TATCGGATCT ATTGTATCGA CACATGAGTC TAGACTAGAT C GTAGTAGCG ATAGCCTAGA TAACATAGCT GTGTACTCAG ATCTGATCTA concentration in pH 7.6 20 mM Tris Acetate, 5 mM MgAc 2 buffer at 95 o C for 5 min, then 70 cycles of 1 min 95 o C ( 1 o C/cycle) and subsequently held at 4 o C on a Applied Biosy stems Veriti TM Thermal Cycler. Annealing efficiency was confirmed on a 1% agarose gel stained with ethidium bromide and visualized with UV light on the Molecular Imager Gel Doc TM XR system (Bio Rad Laboratories, Inc). Nuclear protein extract was treated with the TURBO DNA free TM Kit (Ambion ) according to manufacturer instructions prior to the EMSA binding reaction. EMSA was performed with the Molecular Probes TM Electrophoretic Mobility Shift Assay (EMSA) Kit (Molecular Probes, Inc, Eugene, OR) according to manufacturer specified protocol. 30 ng of annealed oligonucleotide sequences corresponding to the CACNB2 A allele, G allele and nonconsensus sequence was incubated with a range of 9.5 19.5 g crude nuclear protein extract in 1X binding buffer conta ining 750 mM KCl, 0.5 mM

PAGE 76

76 dithiothreitol, 0.5 mM EDTA, 50 mM Tris, pH 7.4 in a final reaction volume of 20 l. Additional control reactions for DNA and protein were also prepared with deionized water and 1X binding buffer. Reactions wer e incubated at 24 o C for 35 minutes prior to loading on a 4 20% Mini PROTEAN TBE p recast gel (Bio Rad Laboratories, Inc, Hercules, CA) with 1 X TBE buffer (89 mM Tris base, 89 mM boric acid, 1 mM EDTA, pH~8.0). Electrophoresis ran for 3 hours at 4 o C, after which the gel was incubated for 20 minutes in 50 mL 1x SYBR Green EMSA staining solution. Stained nucleic acids were visualized with UV transillumination on the Gel Doc TM XR system (Bio Rad Laboratories, Inc). The gel was subsequently stained with 50 mL SYPRO Ruby EMSA protein gel stain for a minimum of 3 hours. The gel was destained in a 10% methanol, 7% acetic acid solution for 60 90 minutes prior to visualization of the stained proteins with UV transillumination on the Gel Doc TM XR system (Bio Rad Laboratories, Inc). Analysis The 2 C t method was used for the relative quantification (RQ) of CACNB2 mRNA expression between CACNB2 rs2357928 genotype groups where the ancestral homozygous G/G genotype was treated as the control RQ analysis was performed with DataAssist TM Software v3.0 (Applied Biosystems). A two test comparing C T values (Average C T endogenous control) between CACNB2 rs2357928 genotype groups was performed with SAS version 9.2 (Cary, NC ) and a p < 0.05 was considered significant For western blo test of mean band intensit y was compared between genotype groups, and a p < 0.05 was considered significant. Due to a limited sample size, a nonparametric Wilcoxon rank sum (WRS) test was also utilized to evaluate differential RNA and protein expression for both RT PCR and western blot analyses.

PAGE 77

7 7 Results Demographic and clinical information for study partici pants is outlined in Table 4 1. Patients were predominately male, young (age < 65 years), with hypertension and peripheral vas cular disease. The most common study medications were aspirin, and statins. RT PCR was successful for n = 9 samples, and n = 1 A/G heterozygote sample was eliminated prior to analysis due to a variably high concentration of cDNA transcript. B2M de monstrated the lowest average pairwise variation relative to GAPDH and 18S (calculated by DataAssist TM Software v3.0) and was selected as the endogenous C T. 50 Relative quantitation of CACNB2 expression is displayed in Figure 4 1. C T was significantly different between CACNB2 rs2357928 A/G and G/G genotype groups ( t test p = 0.02, WRS p = 0.21; Table 4 2) A similar trend for significance was also observed between A/A and G/G genotype groups ( t test p = 0.07 WRS p = 0.54 ). F or the western blot, GAPDH loading control expression was significantly different between rs2357928 G/G and A/G genotype groups ( t test p = 0.03, WRS p = 0.08 ; Table 4 3, Figure 4 2 ) Interestingly, CACNB2 protein expression in arterial tissue was not sig nificantly different b etween CACNB2 rs2357928 genotype groups (trend test for G allele carriers t test, WRS p = 0.22; Table 4 3). A post hoc power calculation was performed G*Power version 3.1.3 utilizing the means and standard deviations derived from b oth RT PCR and western blot experiments. These data indicate at least one additional rs2357928 G/G individual is required to detect significant differences in mRNA expression between A/A and G/G genotype groups, and n = 3 additional individuals of both A/ A and G/G genotypes would

PAGE 78

78 provide 86% power to detect a 30% difference in mean band intensity between the homozygote genotype groups. EMSA analysis did not reveal differences i n nucleic acid and protein binding between the alternate CACNB2 rs2357928 allel es (Figure 4 3). Bands visualized with SYBR Green nucleic acid stain do indicate an interaction between nucleic acid and nuclear protein in both CACNB2 oligonucleotide sequences that are not observed for reactions containing the nonconsensus, mutant sequ ence Unfortunately, t he current results are confounded by a high background of nonspecific interactions, due to the use of crude nuclear cell extract, and the absence of poly dIdC to prevent nonspecific protein binding. Discussion We have previously pre sented evidence of the association between the rs2357928 polymorphism within an alternative promoter of CACNB2 and increased risk for adverse cardiovascular outcomes among patients randomized to a in our INVEST GENES cohort. 20 Presently, we evaluated whether quantifiable differences in CACNB2 gene expression might be ascertained in human arterial tissue relative to rs2357928 genotyp e. Despite the limited sample size for inclusion in the molecular analysis of th e clinically associated CACNB2 rs2357928 promoter SNP significant differences in mRNA expression wer e observed between genotype groups RT PCR analysis revealed greater CA CNB2 mRNA expression within an arterial tissue sample of a CACNB2 rs2357928 G /G h omozygote compared to samples derived from A/G h eterozygou s individuals T h ese data are consistent with our previously published

PAGE 79

79 association demonstrating greater promoter ac tivity of the rs2357928 G allele in a human cell line. 20 Western blot analysis revealed inconclusive results, however, as CACNB2 protein expression amo ng G / G individuals was not found to be significantly different than the expression observed among A allele carriers Although it is difficult to draw conclusions based on o ur limited data set, there are several factors to consider that might aid the under standing of these preliminary results. Protein expression is subject to a wide range of variability in mRNA p ost transcriptional modification and degradation. Likewise, variation within the cellular machinery responsible for the nuclear export, translatio n or degradation of mRNA as well as cis acting sequences are also a potential source of differential protein expression. 51 Additionally, u se of arterial tissue from patients with ad vanced vascular disease concomitant disease states and complex pharmacotherapy may further confound these results. Finally, an EMSA analysis was used to evaluate whether differences in transcription factor binding could be observed between the alternate CACNB2 rs2357928 alleles. Although s everal putative transcription factor binding sites have been identified in the proximal region surrounding the rs2357928 SNP presently no qualitative change in protein DNA interaction could be observed between alleles Overall, the evaluation of CACNB2 mRNA expression in human arterial tissue relative to rs2357928 genotype appears promising given the consistency observed with our previously published data from a reporter gene assay However, due to a limited

PAGE 80

80 sample size a dditional experiments incorporating a larger sample cohort will be required to delineate the validity the results herein

PAGE 81

81 Table 4 1. Baseline demographics of vascular tissue bank participants Characteristic Vascular Tissue Bank (n = 12) CACNB2 genotype G/G ( n = 2 ) A/G ( n = 8 ) A/A ( n = 2 ) Demographics and Social History Age, mean (SD), years 6 0 4 13. 7 55.5 6.4 62.1 15.9 58.5 13.4 Women 4 ( 33.3 ) 1 (50.0) 2 (25.0) 1 (50.0) Race White 9 (7 5 .0) 2 (100.0) 6 (75. 0) 1 (50.0) Black 3 ( 25 .0) 0 (0.0) 2 (25.0) 1 (50.0) Weight (kg) 77.1 30 5 60.7 11.8 73.6 13.5 52.0 0.0 BP, mean (SD), mmHg Systolic 13 4 8 2 2 7 114.0 7.1 143.1 23.5 122.5 4.9 Diastolic 68. 3 10. 0 58.5 16.2 69.0 7.5 75.5 12.0 HR (beats per minute) 88.8 1 3 8 84.0 16.9 86.3 13.4 102.5 9.2 Smoking history 10 ( 83.3) 1 (50.0) 7 ( 66 7 ) 2 (100.0) Medical history Hypertension 11 (91.0) 2 (100.0) 7 (87.5) 2 (100.0) Myocardial infarction 2 (16.7) 0 (0.0) 1 (12.5) 1 (50.0 ) Heart failure 4 (33.3) 0 (0.0) 3 (37.5) 1 (50.0) Diabetes 7 (58.3) 2 (100.0) 5 (62.5) 0 (0.0) Hypercholesterolemia 7 (58.3) 2 (100.0) 4 (50.0) 1 (50.0) Peripheral vascular disease 11 (91.7) 2 (100.0) 7 (87.5) 2 (100.0) A utoimmune /genetic disord e r 2 ( 16.7 ) 1 (50.0) 1 (8.3) 0 (0.0) Medication use Aspirin 11 (91.7) 2 (100.0) 7 (66.7) 2 (100.0) 10 (83.3) 2 (100.0) 6 (75.0 ) 2 (100.0) CCB 4 (33.3) 0 (100.0) 3 (37.5) 1 (50.0) ACE inhibitor 4 (33.3) 1 (50.0) 3 (37.5) 0 (0.0) Diuretic 4 ( 33.3 ) 0 (0.0) 3 (37.5) 1 (50.0) Statin 10 ( 83.3 ) 2 (100.0) 6 (75.0) 2 (100.0) Warfarin 5 (41.7 ) 2 (100.0) 2 (25.0) 1 (50.0) Data are presented as no. (%) or mean SD ; HR = heart rate. Table 4 2. Analysis of RT PCR mean C T by CACNB2 rs2357928 genotype CACNB2 rs2357928 Sample size Mean C T SD t test p value WRS p value G/G 1 5.23 A/G 6 7.63 0.69 0.02 0.21 A/A 2 8.18 0.27 0.07 0.54 p value s represent t test or nonparametric Wilcoxon rank sum (WRS) of mean C T relative to G/G ; trend test for G allele carriers t test p = 0.31, WRS p = 0.30.

PAGE 82

82 Table 4 3. Net intensity of CACNB2 and GAPDH expressio n by CACNB2 rs2357928 genotype CACNB2 Genotype Net Intensity SD CACNB2 Net Intensity SD GAPDH t test p value CACNB2 WRS p value CACNB2 t test p value GAPDH WRS p value GAPDH G/G 473 91 599 118 A/G 552 161 364 88 0.55 0.56 0.03 0.08 A /A 675 86 510 52 0.15 0.24 0.43 0.69 t test or nonparametric Wilcoxon rank sum (WRS) of mean net intensity relative to G/G; trend test for G allele carriers CACNB2 t test WRS p = 0.22; GAPDH t test p = 0.48 WRS p = 0.33. Figure 4 1. Relative C ACNB2 mRNA expression in human arterial tissue G/G reference genotype individual RQ values plotted.

PAGE 83

83 Figure 4 2. Western blot analysis of CACNB2 protein express ion relative to GAPDH. Lanes 1 2 rs2357928 A / A genotype ; lanes 3 7 rs2357928 A/G genotype lanes 8 9 rs2357928 G / G genotype

PAGE 84

84 Figure 4 3. EMSA characterization of DNA and protein interactions 30 ng of DNA sequences corresponding to CACNB2 rs2357928 A allele (A), G allele (G), and mutant (m) were added to nuclear cell extract as describe d in Methods. ( A.) SYBR Green n ucleic stain of DNA interacting with protein. Bands highlighted in lanes 4 and 6 show interactions that are not seen with the mutant oligonucleotide or protein alone. ( B.) SYPRO Ruby protein stain.

PAGE 85

85 CHAPTER 5 SUMMARY AND CONCLUSIONS A number of medications commonly utilized in the management of hypertension are known to intercede in the pathway of Ca 2+ signaling to exert their antihypertensive effects. Perhaps not surprisingly, a n increasing amount of evidence demonstrati ng a relationship between genetic variation within the Ca 2+ signaling pathway and varied cardiovascular phenotypes has been highlighted in recent literature. 14 19 20 22 23 29 42 45 The overarching goal of the current research project was to advance our under standing of the role of variability with in genes of t he Ca 2+ signaling pathway in the treatment response s observed to commonly used antihypertensive s. We first evaluated associations between variation in genes comprising the activation pathway of Ca 2+ signaling and BP response to antihypertensive therapy in a population of uncomplicated hypertensives. Several strong associations between CACNA1C polymorphisms and BP response to atenolol were observed exclusively among PEAR blacks. N otably, the CACNA1C rs2239101 SNP that has previously been associated with s ystolic BP in a large candidate gene cohort study was also associated with greater BP response to atenolol in a manner directionally consistent with the original finding. 29 Among the novel associations reported here, we also identified a common haplotype including CACNA1C rs11831085 and rs12425032 that was significantly associated with BP re sponse to atenolol among blacks, despite the limited LD observed in this ge ne. We then evaluated the relationship between genetic variation of the Ca 2+ signaling pathway and clinical susceptibility to adverse cardiovascular outcomes in a high risk cohort of hypertensive s with coronary artery disease. To this end, several signifi cant

PAGE 86

86 associations with adverse cardiovascular outcomes among INVEST GENES case control cohort participants were observed Interestingly, the CACNA1C rs2239101 SNP was identified as a risk allele among INVEST GENES case control cohort whites a finding whi ch is counter to previous associations indicating a phenotype of lower systolic BP Additionally, the CACNA1C rs12425032 polymorphism associated with lower BP response to atenolol among PEAR blacks demonstrated a significant pharmacogenetic interaction with treatment strategy among INVEST GENES case control cohort blacks, where randomization to the CCB treatment strategy conferred an exa ggerated risk for the primary outcome. Also of considerable interest, the rs381 1 003 polymorphism within the calsequestrin 2 gene demonstrated a significant interaction with treatment strategy in the INVEST GENES case control cohort, as a considerably decr eased risk for adverse cardiovascular outcomes was observed among whites randomized to the CCB treatment strategy. Mutations within the coding sequence of CASQ2 found to affect the Ca 2+ binding affinity of the protein have previously demonstrated associat ions with fatal ventricular arrhythmias. 44 Th e intronic CASQ 2 rs3811003 was found to be in high LD with SNPs previously associated with sudden cardiac death in a separate cohort of patients with coronary artery disease. Several novel pharmacogenetic interactions with treatment strategy within the INVEST GENES case control cohort were also observed among genes encoding phospholamban and the cardiac ryanodine receptor 2. Finally, we recruited human subjects to participate in a vascular tissue ban k study for the purpose of examining whether quantifiable differences in gene expression might

PAGE 87

87 contribute to a clinically significant genetic association found within a regulatory sequence of CACNB2 CACNB2 mRNA expression levels in arterial tissue were found to be significantly d ifferent by rs2357928 genotype. C oncordant with data from a previously published reporter gene assay, G/G homozygotes demonstrated greater fold change expression than rs2357928 A allele carriers. Although examination of protein expression did not yield similar results, we accept the present sampl e size is l imited and likely confounded by the complex disease p athology of study participants. Larger sample sizes will be required to fully understand the influence of the CACNB2 promoter SNP, but these data lend additional support for this being a func tional polymorphism. Overall, we have herein documented several novel genetic associations between genes essential to the activation pathway of Ca 2+ signaling and the cardiovascular phenotypes of BP response and adv erse outcomes Additionally, significan t inroads were established for the future examination of the molecular mechanisms that produce the observed cardiovascular phenotypes of clinically meaningful pharmaco genetic associations.

PAGE 88

88 APPENDIX A NOMINAL ASSOCIATIONS BETWEEN CALCIUM SIGN ALING P ATHWAY VARIATION AND BLOOD PRESSURE R ESPONSE WITHIN THE P HARMACOGENOMICS OF ANTIHYPERTENSIVE RES PONSE (PEAR) STUDY

PAGE 89

89 Table A 1 Associations between CACNA1C rs2299657 genotype and blood pressure response to HCTZ among PEAR whites Response MAF Hmz. (mmHg) (mmHg) Minor Hmz. (mmHg) P Hmz. (mmHg) (mmHg) Minor Hmz. (mmHg) P HCTZ Monotherapy 0.26 3.3 5.7 4.3 0.005 6.1 9.9 7.7 0.002 HCTZ Add On 0.26 2.5 2.9 2.6 0.45 5.9 5.4 5.2 0.98 HCTZ Monotherapy + Add On 0.26 3.0 4.2 3.3 0.016 6.1 7.5 6.3 0.036 A djusted for age, gender, and baseline BP Response to monotherapy + add on w as adjusted for BP at baseline in the monotherapy group and BP at the first response assessment for the add on therapy group as well as controlling for the order of drug initiation DBP = diastolic BP; SBP = systolic BP; Hmz. = homozygote; Het. = heterozygote. Table A 2. Associations between CACNA1C rs2299657 genotype and blood pressure response to atenolo l among PEAR whites Response MAF Hmz. (mmHg) (mmHg) Minor Hmz. (mmHg) P Hmz. (mmHg) (mmHg) Minor Hmz. (mmHg) P Atenolol Monotherapy 0.26 9.0 10.6 11.1 0.04 9.6 12.0 13.0 0.039 Atenolol Add On 0.26 9.8 9.1 11.1 0.29 11.2 9. 4 11.2 0.73 Atenolol Monotherapy + Add On 0.26 9.3 10.0 11.1 0.03 10.4 10.9 12.4 0.15 A djusted for age, gender, and baseline BP Response to monotherapy + add on w as adjusted for BP at baseline in the monotherapy group and BP at the first respons e assessment for the add on therapy group as well as controlling for the order of drug initiation DBP = diastolic BP; SBP = systolic BP; Hmz. = homozygote; Het. = heterozygote.

PAGE 90

90 Table A 3. Associations between CACNB2 rs7089228 genotype and blood pressure response to HCTZ among PEAR blacks Response MAF Hmz. (mmHg) (mmHg) Minor Hmz. (mmHg) P Hmz. (mmHg) (mmHg) Minor Hmz. (mmHg) P HCTZ Monotherapy 0.32 8.3 6.6 4.2 0.009 14.0 11.0 7.1 0.01 HCTZ Add On 0.32 7.9 6.6 6.1 0.04 13.3 11.5 14.9 0. 37 HCTZ Monotherapy + Add On 0.32 8.1 6.6 5.2 8.1x10 4 13.7 11.3 11.3 0.015 A djusted for age, gender, and baseline BP Response to monotherapy + add on w as adjusted for BP at baseline in the monotherapy group and BP at the first response assessmen t for the add on therapy group as well as controlling for the order of drug initiation DBP = diastolic BP; SBP = systolic BP; Hmz. = homozygote; Het. = heterozygote. Table A 4. Associations between RYR2 rs16832052 genotype and blood pressure response t o atenolol among PEAR blacks Response MAF Hmz. (mmHg) (mmHg) Minor Hmz. (mmHg) P Hmz. (mmHg) (mmHg) Minor Hmz. (mmHg) P Atenolol Monotherapy 0.28 3.0 5.9 4.2 0.055 1.8 4.8 3.6 0.18 Atenolol Add On 0.28 5.2 5.6 10.1 0.004 5.7 5.9 1 1.1 0.018 Atenolol Monotherapy + Add On 0.28 4.1 5.7 7.2 0.002 3.7 5.3 7.5 0.023 A djusted for age, gender, and baseline BP Response to monotherapy + add on w as adjusted for BP at baseline in the monotherapy group and BP at the first response asse ssment for the add on therapy group as well as controlling for the order of drug initiation DBP = diastolic BP; SBP = systolic BP; Hmz. = homozygote; Het. = heterozygote.

PAGE 91

91 Table A 5. Associations between CACNA1C rs2238078, rs11831085, rs12425032, rs223910 1 TGCT haplotype and blood pressure response to atenolol among PEAR blacks Response Frequency 0 copy (mmHg) copy (mmHg) copies (mmHg) P copy (mmHg) copy (mmHg) copies (mmHg) P Atenolol Monotherapy 0.15 4.9 2.8 0.059 4.9 0.7 0.0016 Atenolol Add On 0.15 5.7 3.9 0.017 6.4 1.7 1.78x10 5 A tenolol Monotherapy + Add On 0.15 6.6 5.1 0.048 7.9 4.2 0.001 A djusted for age, gender, and baseline BP Response to monotherapy + add on w as adjusted for BP at baseline in the monotherapy group and BP at the first response assessment for the add on therapy group as well as controlling for the order of drug initiation Table A 6. Associations between CACNA1C rs2238078, rs11831085, rs12425032, rs2239101 ATGT haplotype and blood pressure response to atenolol among PEAR blacks Response Frequency DBP 0 copy (mmHg) copy (mmHg) copies (mmHg) P copy (mmHg) copy (mmHg) copies (mmHg) P Atenolol Monotherapy 0.10 3.9 5.8 0.043 2.4 6.7 0.0098 Atenolol Add On 0.10 4.7 7.0 0.0045 4.3 7.8 0.002 Aten olol Monotherapy + Add On 0.10 5.6 8.2 0.045 6.3 8.7 0.07 A djusted for age, gender, and baseline BP Response to monotherapy + add on w as adjusted for BP at baseline in the monotherapy group and BP at the first response assessment for the add on therapy group as well as controlling for the order of drug initiation

PAGE 92

92 APPENDIX B NOMINAL ASSOCIATIONS BETWEEN CALCIUM SIGN ALING PATHWAY VARIAT ION AND ADVERSE CARDIOVA SCULAR OUTCOMES IN T HE INTERNATIONAL VERAPAMIL SR TRANDOL APRIL STUDY GENETIC SUBSTUD Y (INVEST GENES)

PAGE 93

93 Table B 1. Associations between variation within CACNA1C and adverse cardiovascular outcomes in INVEST GENES a Gene SNP Race MAF Outcome Treatment strategy OR (95% CI) p value CACNA1C *rs2299657 White 0.26 Death CCB 2.53 (1.56 4.10) 1. 6x10 4 CACNA1C rs1034937 Hisp 0.09 Death 8.03 (2.23 28.9) 0.0014 CACNA1C Hisp 0.09 PO 3.80 (1.26 11.4) 0.017 CACNA1C rs1034937 Hisp 0.09 Death 4.24 (1.45 12.1) 0.0068 CACNA1C rs4765680 Hisp 0.18 Death 5.09 (1.87 13.90) 0.0015 C ACNA1C rs4765680 Hisp 0.18 PO 2.93 (1.29 6.65) 0.01 CACNA1C rs4765680 Hisp 0.18 Death Combined 1.91 (1.09 3.35) 0.02 CACNA1C rs1009281 Hisp 0.37 PO 3.27 (1.51 7.09) 0.0027 CACNA1C rs2239077 Hisp 0.39 Death 2.97 (1.22 7.23) 0.016 CACNA1C rs2239 077 White 0.47 Death CCB 1.71 (1.07 2.72) 0.024 CACNA1C rs2239077 White 0.47 Death 0.55 (0.30 0.98) 0.04 CACNA1C rs1548649 Black 0.27 PO CCB 0.008 (<0.001 0.416) 0.016 CACNA1C Hisp 0.08 Death 5.19 (1.35 19.9) 0.016 CACNA1C rs11608641 W hite 0.08 Death 2.67 (1.18 6.02) 0.018 CACNA1C *rs12425032 Black 0.22 PO CCB 9.97 (1.33 74.3) 0.025 a Adjusted for age, sex, history of MI, heart failure, diabetes, ancestry ; strongest single associations shown in bold. Dominant model ; *Previous assoc iation with BP response to atenolol or HCTZ in PEAR; PO = primary outcome

PAGE 94

94 Table B 2. INVEST GENES CACNA1C SNP treatment interactions a Gene SNP Race MAF Outcome OR (95% CI) CCB OR (95% CI) p value CACNA1C rs1034937 Hisp 0.09 Death 0.44 (0.11 1.75) 8.03 (2.23 28.9) 0.0074 CACNA1C rs1034937 Hisp 0.09 PO 0.29 (0.06 1.24) 3.80 (1.27 11.4) 0.0097 CACNA1C rs1009281 Hisp 0.37 PO 0.72 (0.32 1.62) 7.31 (1.78 29.9) 0.009 CACNA1C rs2370596 Hisp 0.35 PO 1.34 (0.59 3.04) 0.24 (0.08 0.69) 0.0095 CACNA1C rs237 0596 White 0.27 PO 0.82 (0.49 1.39) 1.78 (1.00 3.16) 0.04 CACNA1C rs4765680 Hisp 0.18 PO 0.54 (0.22 1.35) 2.89 (1.10 7.57) 0.0096 CACNA1C rs1548649 Black 0.27 PO 0.006 (<0.001 0.30) 2.89 (0.66 12.6) 0.012 CACNA1C rs3794292 Hisp 0.08 Death 0.59 (0.18 1.9 4) 5.19 (1.35 19.9) 0.014 CACNA1C rs1016388 Black 0.26 Death 0.007 (<0.001 1.05) 7.08 (1.11 45.3) 0.02 CACNA1C rs7305879 Black 0.13 PO 8.83 (1.18 66.3) 0.73 (0.14 3.77) 0.027 CACNA1C rs994901 White 0.50 Death 0.50 (0.26 0.99) 2.19 (0.72 6.64) 0.03 a S hows the dominant model a djusted for age, sex, history of MI, heart failure, diabetes, ancestry ; PO = primary outcome; strongest single associations shown in bold

PAGE 95

95 Table B 3. Associations between variation within CASQ2 and adverse cardiovascular out comes in INVEST GENES a Gene SNP Race MAF Outcome Treatment strategy OR (95% CI) p value CASQ2 White 0.31 Death CCB 0.28 (0.14 0.60) 9.8x10 4 CASQ2 rs3811003 White 0.31 PO CCB 0.44 (0.26 0.76) 0.003 CASQ2 Black 0.19 PO Combined 4.1 5 (1.5 11.43) 0.0059 CASQ2 rs7355132 White 0.25 Death CCB 0.35 (0.16 0.75) 0.006 a Adjusted for age, sex, history of MI, heart failure, diabetes, ancestry ; Dominant model ; PO = primary outcome ; strongest single associations shown in bold Table B 4. IN VEST GENES CASQ2 SNP treatment interactions a Gene SNP Race MAF Outcome OR (95% CI) CCB OR (95% CI) p value CASQ2 rs12036369 b White 0.23 Death 0.40 (0.19 0.87) 1.61 (0.73 3.55) 0.01 CASQ2 rs12036369 White 0.23 PO 0.47 (0.26 0.85) 1.17 (0.67 2.06) 0.02 CASQ2 rs2997742 c White 0.44 Death 2.54 (1.08 5.96) 0.65 (0.29 1.46) 0.017 CASQ2 rs3010396 c White 0.45 PO 2.19 (1.04 4.65) 0.68 (0.29 1.58) 0.037 a Shows the dominant model a djusted for age, sex, history of MI, heart failure, diabetes, ancestry ; strongest single association shown in bold. b LD, r 2 between rs7355132 (Chapter 3) and rs12036369 among whites = 0.81. c LD, r 2 between rs2997742 and rs3010396 among whites = 0.91. PO = primary outcome

PAGE 96

96 Table B 5. Associations between variation within KCNMB1 a nd adverse cardiovascular outcomes in INVEST GENES a Gene SNP Race MAF Outcome Treatment Strategy OR (95% CI) p value KCNMB1 rs11739136 Hisp 0.14 Death B 3.65 (1.38 9.67) 0.009 KCNMB1 rs2301149 Hisp 0.11 PO B 3.45 (1.20 9.96) 0.02 KCNMB1 Hisp 0.13 Death CCB 0.26 (0.07 0.95) 0.04 KCNMB1 White 0.12 Death CCB 0.34 (0.12 0.98) 0.04 KCNMB1 rs2075612 Hisp 0.38 PO CCB 0.49 (0.25 0.99) 0.049 a Adjusted for age, sex, history of MI, heart failure, diabetes, ancestry Dominant model Tabl e B 6. INVEST GENES KCNMB1 SNP treatment interactions a Gene SNP Race MAF Outcome OR (95% CI) CCB OR (95% CI) p value KCNMB1 rs827778 Hisp 0.13 PO 0.4 3 (0.1 4 1. 28 ) 2. 62 (0. 91 7 58 ) 0.02 KCNMB1 rs11742820 White 0.20 PO 1. 20 (0. 70 2 06 ) 0.46 (0.24 0.89) 0.03 KCNMB1 rs2075612 Hisp 0.39 PO 0.43 (0.18 1.009) 1.94 (0.67 5.60) 0.04 a Shows the dominant model a djusted for age, sex, history of MI, heart failure, diabetes, ancestry ; PO = primary outcome.

PAGE 97

97 Table B 7. Associations between variation within RYR2 and adverse cardiovascular outcomes in INVEST GENES a Gene SNP Race MAF Outcome Treatment strategy OR (95% CI) p value RYR2 rs12126370 White 0.43 Death 2.74 (1.49 5.04) 0.001 RYR2 rs12126370 White 0.43 Death Combined 1. 76 (1. 21 2. 53 ) 0.00 2 RYR2 rs2184014 Hisp 0.49 Death CCB 0.33 (0.1 6 0.6 6 ) 0.0016 RYR2 rs3766884 b Hisp 0.21 Death CCB 0.18 (0.06 0.57) 0.003 RYR2 rs10158497 White 0.25 Death 3. 62 ( 1. 51 8. 66 ) 0.00 3 RYR2 rs961121 White 0.05 Death CCB 3.14 (1.45 6.77 ) 0.00 3 RYR2 rs961121 White 0.05 Death B 3. 45 (1.52 7 86 ) 0.00 3 RYR2 rs790882 Hisp 0.22 Death CCB 0.2 2 (0. 08 0.6 2 ) 0.00 4 RYR2 rs1008956 White 0.46 Death CCB 0.38 (0.19 0.75) 0.005 RY R2 rs2253244 b Hisp 0.30 Death CCB 0.30 (0.13 0.71) 0.006 RYR2 rs2485584 White 0.47 Death CCB 0. 40 (0.20 0.7 9 ) 0.00 7 a Adjusted for age, sex, history of MI, heart failure, diabetes, ancestry ; strongest single associations shown in bold. Dominant model ; b LD r 2 between rs3766884 and rs2253244 = 0.58.

PAGE 98

98 Table B 8. INVEST GENES RYR2 SNP treatment interactions a Gene SNP Race MAF Outcome OR (95% CI) CCB OR (95% CI) p value RYR2 rs1008956 b White 0.46 Death 0.3 8 (0.19 0.7 5 ) 1.8 1 (0.7 0 4. 70 ) 0.007 RY R2 rs790882 Hisp 0.22 Death 0.21 (0.07 0.6 1 ) 1. 67 (0. 53 5. 25 ) 0.01 RYR2 rs2184014 c Hisp 0.49 Death 0.3 0 (0.13 0.7 2 ) 4.94 (0.59 41.3) 0.01 RYR2 rs12129023 Hisp 0.14 Death 0.54 (0.19 1.54) 4.13 (1.28 13.3) 0.01 RYR2 rs12129023 Hisp 0.14 PO 0.37 (0.13 1.09 ) 1.99 (0.71 5.55) 0.029 RYR2 rs2253244 Hisp 0.30 Death 0.2 7 (0. 11 0.6 9 ) 1. 67 (0. 54 5 13 ) 0.01 RYR2 rs2490373 b White 0.30 Death 0.51 (0.26 1.00) 1.71 (0.75 3.92) 0.01 RYR2 rs12038715 Hisp 0.35 PO 0.46 (0.19 1.07) 2.53 (0.87 7.29) 0.018 RYR2 rs12038715 Hisp 0.35 Death 0.7 3 (0.3 2 1.7 1 ) 5.1 6 (1. 07 24. 8 ) 0.0 35 RYR2 rs3766875 c Hisp 0.56 PO 0.47 (0.19 1.20) 8.51 (0.95 75.8) 0.02 RYR2 rs3766875 c Hisp 0.56 Death 0.3 3 (0.1 4 0. 88 ) 3. 52 (0.4 1 29.8 ) 0.04 RYR2 rs1759119 White 0.47 Death 3.38 (1.22 9.38) 0.67 (0.2 8 1.57) 0.024 RYR2 rs1759119 Hisp 0.65 Death 2.46 (0.90 6.73) 0.45 (0.13 1.51) 0.045 RYR2 rs12128519 White 0.30 Death 0.6 7 (0.3 5 1.29) 2.4 2 (1.0 1 5. 76 ) 0.02 RYR2 rs10158497 White 0.25 PO 0.87 (0.51 1.47) 1.79 (1.01 3.18) 0.04 a Shows the dominant model a djusted for age, sex, history of MI, heart failure, diabetes, ancestry ; strongest single associations shown in bold; b LD r 2 between rs1008956 and rs2490373 = 0.33; c LD r 2 between rs2184014 and rs3766875 = 0.65; PO = primary outcome.

PAGE 99

99 Table B 9. As sociations between variation within PLN and adverse cardiovascular outcomes in INVEST GENES a Gene SNP Race MAF Outcome Treatment strategy OR (95% CI) p value PLN rs9489438 Hisp 0.40 Death CCB 0.46 (0.23 0.92) 0.027 a Adjusted for age, sex, history of MI, heart failure, diabetes, ancestry

PAGE 100

100 LIST OF REFERENCES 1. Chobanian AV, Bakris GL, Black HR, et al. The Seventh Report of the Joint National Committee on Prevention, Detection, Evaluation, and Treatment of High Blood Pressure: the JNC 7 report. JAMA 2003;289:2560 72. 2. Roger VL, Go AS, Lloyd Jones DM, et al. Heart disease and stroke statistics -2011 update: a report from the American Heart Association. Circulation 2011;123:e18 e209. 3. Beevers G, Lip GY, O'Brien E. ABC of hypertensi on: The pathophysiology of hypertension. BMJ 2001;322:912 6. 4. Dickerson JE, Hingorani AD, Ashby MJ, Palmer CR, Brown MJ. Optimisation of antihypertensive treatment by crossover rotation of four major classes. Lancet 1999;353:2008 13. 5. Schwartz GL, Tu rner ST. Pharmacogenetics of antihypertensive drug responses. Am J Pharmacogenomics 2004;4:151 60. 6. Ferrari P, Bianchi G. Genetic mapping and tailored antihypertensive therapy. Cardiovasc Drugs Ther 2000;14:387 95. 7. Arnett DK, Boerwinkle E, Davis BR, Eckfeldt J, Ford CE, Black H. Pharmacogenetic approaches to hypertension therapy: design and rationale for the Genetics of Hypertension Associated Treatment (GenHAT) study. Pharmacogenomics J 2002;2:309 17. 8. Bers DM. Cardiac excitation contraction coup ling. Nature 2002;415:198 205. 9. Catterall WA, Perez Reyes E, Snutch TP, Striessnig J. International Union of Pharmacology. XLVIII. Nomenclature and structure function relationships of voltage gated calcium channels. Pharmacol Rev 2005;57:411 25. 10. Yu FH, Catterall WA. The VGL chanome: a protein superfamily specialized for electrical signaling and ionic homeostasis. Sci STKE 2004;2004:re15. 11. Bodi I, Mikala G, Koch SE, Akhter SA, Schwartz A. The L type calcium channel in the heart: the beat goes on. J Clin Invest 2005;115:3306 17. 12. Singh A, Gebhart M, Fritsch R, et al. Modulation of voltage and Ca2+ dependent gating of CaV1.3 L type calcium channels by alternative splicing of a C terminal regulatory domain. J Biol Chem 2008;283:20733 44. 13. He dley PL, Jorgensen P, Schlamowitz S, et al. The genetic basis of Brugada syndrome: a mutation update. Hum Mutat 2009;30:1256 66.

PAGE 101

101 14. Beitelshees AL, Navare H, Wang D, et al. CACNA1C gene polymorphisms, cardiovascular disease outcomes, and treatment respons e. Circ Cardiovasc Genet 2009;2:362 70. 15. Kamide K, Yang J, Matayoshi T, et al. Genetic polymorphisms of L type calcium channel alpha1C and alpha1D subunit genes are associated with sensitivity to the antihypertensive effects of L type dihydropyrid ine calcium channel blockers. Circ J 2009;73:732 40. 16. Buraei Z, Yang J. The ss subunit of voltage gated Ca2+ channels. Physiol Rev 2010;90:1461 506. 17. Hu D, Barajas Martinez H, Nesterenko VV, et al. Dual variation in SCN5A and CACNB2b underlies the development of cardiac conduction disease without Brugada syndrome. Pacing Clin Electrophysiol 2010;33:274 85. 18. Cordeiro JM, Marieb M, Pfeiffer R, Calloe K, Burashnikov E, Antzelevitch C. Accelerated inactivation of the L type calcium current due to a mutation in CACNB2b underlies Brugada syndrome. J Mol Cell Cardiol 2009;46:695 703. 19. Levy D, Ehret GB, Rice K, et al. Genome wide association study of blood pressure and hypertension. Nat Genet 2009. 20. Niu Y, Gong Y, Langaee TY, et al. Genetic varia tion 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. 21. Brenner R, Perez GJ, Bonev AD, et al. Vasoregulation by the beta1 subunit of the calcium activated potassium channel. Nature 2000;407:870 6. 22. Fernandez Fernandez JM, Tomas M, Vazquez E, et al. Gain of function mutation in the KCNMB1 potassium chann el subunit is associated with low prevalence of diastolic hypertension. J Clin Invest 2004;113:1032 9. 23. Beitelshees AL, Gong Y, Wang D, et al. KCNMB1 genotype influences response to verapamil SR and adverse outcomes in the INternational VErapamil SR/Tr andolapril STudy (INVEST). Pharmacogenet Genomics 2007;17:719 29. 24. Davis HM, Johnson JA. Heart Failure Pharmacogenetics: Past, Present, and Future. Curr Cardiol Rep 2011. 25. Duarte JD, Cooper DeHoff RM. Mechanisms for blood pressure lowering and meta bolic effects of thiazide and thiazide like diuretics. Expert Rev Cardiovasc Ther;8:793 802.

PAGE 102

102 26. Calder JA, Schachter M, Sever PS. Direct vascular actions of hydrochlorothiazide and indapamide in isolated small vessels. Eur J Pharmacol 1992;220:19 26. 27 Pickkers P, Garcha RS, Schachter M, Smits P, Hughes AD. Inhibition of carbonic anhydrase accounts for the direct vascular effects of hydrochlorothiazide. Hypertension 1999;33:1043 8. 28. Zhu Z, Zhu S, Liu D, Cao T, Wang L, Tepel M. Thiazide like diureti cs attenuate agonist induced vasoconstriction by calcium desensitization linked to Rho kinase. Hypertension 2005;45:233 9. 29. Johnson AD, Newton Cheh C, Chasman DI, et al. Association of hypertension drug target genes with blood pressure and hypertension in 86,588 individuals. Hypertension 2011;57:903 10. 30. 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 20 09;157:442 9. 31. 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. 32. Faul F, Erdfelder E, Buchner A, Lang AG. Statistical power analyses using G*Power 3.1: tests for correlation and regression analyses. Behav Res Methods 2009;41:1149 60. 33. Stephens M, Donnelly P. A comparison of bayesian methods for haplotype reconstruction from population genotype data. Am J Hum Genet 2003;73:1162 9. 34. 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. 35. Catterall WA. Structure and regulation of voltage gated Ca2 + channels. Annu Rev Cell Dev Biol 2000;16:521 55. 36. Pepine 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 trial. JAMA 2003;290:2805 16.

PAGE 103

103 37. Pepine CJ, Handberg Thurmond E, Marks RG, et al. Rationale and design of the International Verapamil SR/Trandolapril Study (INVEST): an Internet based randomized tri al in coronary artery disease patients with hypertension. J Am Coll Cardiol 1998;32:1228 37. 38. The sixth report of the Joint National Committee on prevention, detection, evaluation, and treatment of high blood pressure. Arch Intern Med 1997;157:2413 46. 39. Gerhard T, Gong Y, Beitelshees AL, et al. Alpha adducin polymorphism associated with increased risk of adverse cardiovascular outcomes: results from GENEtic Substudy of the INternational VErapamil SR trandolapril STudy (INVEST GENES). Am Heart J 2008 ;156:397 404. 40. Gauderman WJ MJ. QUANTO 1.1: A computer program for power and sample size calculations for genetic epidemiology studies,. In; 2006. 41. Sotoodehnia N, Isaacs A, de Bakker PI, et al. Common variants in 22 loci are associated with QRS dur ation and cardiac ventricular conduction. Nat Genet 2010;42:1068 76. 42. Song L, Alcalai R, Arad M, et al. Calsequestrin 2 (CASQ2) mutations increase expression of calreticulin and ryanodine receptors, causing catecholaminergic polymorphic ventricular tac hycardia. J Clin Invest 2007;117:1814 23. 43. Terentyev D, Kubalova Z, Valle G, et al. Modulation of SR Ca release by luminal Ca and calsequestrin in cardiac myocytes: effects of CASQ2 mutations linked to sudden cardiac death. Biophys J 2008;95:2037 48. 44. Chopra N, Knollmann BC. Cardiac calsequestrin: the new kid on the block in arrhythmias. J Cardiovasc Electrophysiol 2009;20:1179 85. 45. Westaway SK, Reinier K, Huertas Vazquez A, et al. Common variants in CASQ2, GPD1L, and NOS1AP are significantly as sociated with risk of sudden death in patients with coronary artery disease. Circ Cardiovasc Genet 2011;4:397 402. 46. Nielsen T, Burgdorf KS, Grarup N, et al. The KCNMB1 Glu65Lys polymorphism associates with reduced systolic and diastolic blood pressure in the Inter99 study of 5729 Danes. J Hypertens 2008;26:2142 6. 47. Asahi M, Otsu K, Nakayama H, et al. Cardiac specific overexpression of sarcolipin inhibits sarco(endo)plasmic reticulum Ca2+ ATPase (SERCA2a) activity and impairs cardiac function in mice Proc Natl Acad Sci U S A 2004;101:9199 204.

PAGE 104

104 48. Eijgelsheim M, Newton Cheh C, Sotoodehnia N, et al. Genome wide association analysis identifies multiple loci related to resting heart rate. Hum Mol Genet 2010;19:3885 94. 49. Vasan RS, Glazer NL, Felix J F, et al. Genetic variants associated with cardiac structure and function: a meta analysis and replication of genome wide association data. JAMA 2009;302:168 78. 50. Vandesompele J, De Preter K, Pattyn F, et al. Accurate normalization of real time quantit ative RT PCR data by geometric averaging of multiple internal control genes. Genome Biol 2002;3:RESEARCH0034. 51. Williams RB, Chan EK, Cowley MJ, Little PF. The influence of genetic variation on gene expression. Genome Res 2007;17:1707 16.

PAGE 105

105 BIOGRAPH ICAL SKETCH Heather Marie Davis was born in Philadelphia, Pennsylvania and raised in Longwood, Florida. She graduated from Lake Brantley High School in Altamonte Springs, Florida in 2000. She then attended the University of Central Florida in Orlando, Fl orida from 2001 2003, after which she entered the College of Pharmacy at the University of Florida in Gainesville, Florida. Dr. Davis received her Pharm.D. in 2007 a nd i s a pr a cti c ing p harm ac is t in her comm uni ty Dr Da vis re cei ved her Ph D f rom the U nive rs i ty of Fl orid a in t he spr ing of 201 2.