Fibroblast growth factor 2-mediated cardioprotection: the mediators and downstream targets of FGF2-induced protection from ischemia and reperfusion injury.

A dissertation submitted to the

Division of Graduate Studies of the University of Cincinnati

In partial fulfillment of the requirements for the degree of

Doctor of Philosophy

In the Department of Pharmacology and Cell Biophysics

2012

Janet R. Bodmer Manning BA, Xavier University, 2002 Committee Chair: Jo El J. Schultz

ABSTRACT

Although heart disease is the primary cause of death in several industrialized nations,

there are no widely-used therapies targeting the ischemic heart muscle; current therapies focus

on rapid restoration of blood flow, which produces its own set of injuries. Fibroblast growth

factor 2 (FGF2) has been shown to protect the heart from ischemia and reperfusion (I/R) injury,

reducing infarct size and postischemic dysfunction. However, it is unclear by what mechanism

the two classes of FGF2 expressed in the cardiomyocyte, high molecular weight (HMW) FGF2,

and low molecular weight (LMW) FGF2, exert protective action on the ischemic heart. It has

been established that LMW FGF2 protects the heart from postischemic dysfunction, while

endogenously expressed HMW FGF2 reduces contractile function and relaxation after I/R injury.

The mechanisms by which this occurs are not well understood.

The purpose of this dissertation was to investigate the mechanisms of these differential effects, including elucidating the role of a known cardioprotective kinase, kinase C

(PKC), in the pathways initiated by FGF2 isoforms, as well as the downstream targets of this and other . Of particular interest was determining which isoforms of PKC are mediating LMW FGF2-induced protection from I/R injury, and investigating known and novel targets of these PKC isoforms at the and that may modulate contractile function.

Using mice that only express the LMW FGF2 isoform, it was determined that LMW

FGF2 differentially activates PKCε and α, and that these isoforms of PKC were necessary for

LMW FGF2-mediated protection. Expression of only LMW FGF2 was also found to increase

I and T phosphorylation during ischemia, as well as the activity of actomyosin ATPase,

in a manner that was dependent on PKCα. ATPase. Additionally, differences in calcium cycling

ii

were seen in hearts only expressing LMW FGF2, although no changes in basal levels of calcium

cycling were observed. However, a significant elevation of phosphorylated pThr-17

phospholamban was seen at early ischemia, suggesting that this phosphorylation may play a role

in LMW FGF2 mediated protection from postischemic dysfunction.

Also of interest is the mechanism by which HMW FGF2 reduces postischemic function.

It was hypothesized that HMW FGF2 produces its detrimental effects by interfering with

protective LMW FGF2 signaling. It was found that after I/R injury, the HMW FGF2

overexpresion results in lowered FGF receptor 1 (FGFR1) activation, which is necessary for

LMW FGF2-mediated protection, as well as lowered activity of downstream kinases of FGFR1.

Finally, novel targets of both HMW and LMW FGF2 were investigated. It was found

using pharmacological methodologies that overexpression of HMW and LMW FGF2 result in a

PKC- and MAPK-dependent increase in nitric oxide (NO) production, suggesting that NO

synthase (NOS) is a target of FGF2 signaling. In addition, it was found using a microarray that

HMW and LMW differentially regulate the expression of that may play a potentially protective role in I/R injury. These results elucidate a novel mechanism for a potentially

therapeutic molecule to protect the heart from ischemia and reperfusion injury.

iii

iv

ACKNOWLEDGEMENTS

This dissertation work would not have been possible without the help of many friends

and colleagues, to whom I am greatly indebted. First and foremost, I would like to gratefully

acknowledge my advisor and mentor, Dr. Jo El Schultz, for her unwavering support throughout

my time as a graduate student. Her careful direction and advice were invaluable to my growth as

a scientist, and her method of challenging me to think independently and to defend my research

has taught me to become a meticulous, confident experimentalist. I am thankful for her guidance

over the years, her advocacy, and her friendship.

I would also like to extend my gratitude to the members of my dissertation committee,

Drs. W. Keith Jones, Terence Kirley, Evangelia Kranias, Abdul Matlib, and Jeffrey Molkentin,

for their encouragement, collaboration, and constructive criticism. The submission of a sound

and thorough dissertation project is the result of their meticulous attention to the progress of my

research, and my own progress as a graduate student.

I am grateful as well to the past and present members of the Schultz laboratory, Gilbert

Newman, Dr. Craig Bolte, Adeola Adeyemo, Dan Pietras, Yu Zhang, Angel Whitaker, and

Colleen York, for their insights and suggestions, and for joining me in celebration when an

experiment worked and in commiseration when it didn’t. I owe a particular debt of gratitude to

my student mentor, Dr. Siyun Liao, who spent many patient hours at the bench with me as a new and inexperienced graduate student, and was always available for advice or suggestions, from my first day in the lab to long after her own graduation.

I am fortunate to have benefited from the assistance of many collaborators at the

University of Cincinnati, including Stela Florea and Wen Zhao in the laboratory of Dr. Kranias,

who generously lent their expertise in isolating murine cardiomyocytes and measuring calcium

v

cycling and cell contraction, and Drs. Aruna Wijeratne and Ken Greis, who put a great deal of

hard work into developing a mass spectrometry protocol for determining the relative levels of

phosphoproteins. In addition, the collaboration of Dr. W. Glen Pyle and Sarah Parker at the

University of Guelph in isolating and analyzing fractions of the hearts I collected

was essential for uncovering the effects of FGF2 isoforms at the myofibril. I would also like to

express my gratitude to Dr. Muthu Periasamy and Meghna Pant at the Ohio State University, for

their insights and collaboration examining the role of in our heart models.

I would also like to thank the medical, graduate, and undergraduate students who have

rotated through or volunteered in the Schultz lab, and made significant contributions towards this

dissertation, including Brian Oloizia, Greg Carpenter, Laura Moon, Edward Wright, Kristin

Luther, Arial Rydeen, Xiaoqian Gao, and Chi Keung Lam. Their enthusiasm brought fresh energy to a variety of projects that were essential for completing the ideas put forth in this manuscript.

I owe a tremendous debt of gratitude to the past and current members of the Department of Pharmacology, including Nancy Thyberg, George Sfyris, Carol Ross, Damita Harris, Mark

Spanyer, and Donna Gering, for ensuring that deadlines were met, software was working, supplies were ordered, and paperwork was filed.

Finally, I would like to thank my friends and family for their love, encouragement, and

patience. I offer my gratitude to my parents, for their infectious belief from the very beginning that I could succeed in whatever I wanted to do, and their unconditional support that allowed me to do so under sometimes discouraging circumstances. I would also like to show appreciation for my sister Lisa, for drawing on her own experiences to help talk me through some of the more

difficult parts of graduate school, and offering always-welcome advice, and to my brother Bill,

vi

for helping me to remember to leave the lab occasionally to have fun. I would like to extend my gratitude to my extended family, and in particular my uncle John, for showing a curious fourteen-year-old how a sphygmomanometer worked many years ago. I would also like to thank many of my friends for their patience and encouragement, especially Cate and Solange, who went above and beyond during some of the disheartening moments of the past few years to keep me fed, sheltered, sane, and sufficiently caffeinated to complete my degree. Finally, I’d like to thank my boyfriend Mike, for his unwavering encouragement and love.

vii

TABLE OF CONTENTS Page

Abstract ii

Acknowledgments v

List of Tables and Figures xiv

List of Abbreviations xix

Introduction and Background 1

Cardiac Ischemia-Reperfusion Injury 1

Fibroblast Growth Factors and FGF2 5

FGF2 and Cardioprotection 10

High and low molecular weight FGF2 12

Protein Kinase C 15

PKCs and cardioprotection 18

FGF2 and PKCs 20

Downstream targets of FGF2 and PKCs: Contractile proteins 22

Downstream targets of FGF2 and PKCs: Calcium-handling proteins 25

Other targets of FGF2 in the heart during I/R 28

Statement of Purpose 31

Material and Methods 36

Animals and Exclusion Criteria 36

Mouse Generation and Breeding 37

Generation of Fgf2 KO mice 38

viii

Generation of high molecular weight FGF2 knockout (HMWKO) mice 39

Generation of low molecular weight FGF2 knockout (LMWKO) mice 42

Generation of PKC alpha KO (PKCαKO) mice 43

Generation of human 24 kDa high molecular weight FGF2 transgenic

(HMW Tg) mice 46

Generation of cardiac-specific overexpression of FGF2 transgenic

(FGF2 Tg) mice 46

Isolated Working Mouse Heart Ischemia/Reperfusion Studies 49

Time course I/R studies 52

Pharmacological studies 53

Immunoblotting 54

Analysis of FGF2 protein content in heart homogenate 55

Immunoblotting and detection of proteins involved in kinase signaling,

FGF2 signaling, calcium cycling/handling 56

Analysis of protein and phospho-protein levels related to FGF2

signaling and calcium handling 56

Analysis of PKC translocation to the membrane 58

Evaluation of Nitrite Levels in Coronary Effluent 59

Assessment of Cardiac Myofilament Activity 60

Actomyosin Mg2+ ATPase activity in ischemic/reperfused mouse hearts 60

Contractile protein phosphorylation 61

Translocation of PKC isoforms to 61

ix

Myocyte Isolation and Calcium Transient Evaluation 62

Mouse cardiomyocyte isolation 62

Evaluation of isolated cardiomyocyte function 62

Cardiomyocyte calcium transients and sarcoplasmic reticulum load 63

Microarray Assessment of Expression in Mouse Hearts with only

LMW or HMW FGF2 64

RNA preparation 64

RNA Microarray 65

Real-time reverse transcriptase PCR (qRTPCR) validation 67

Immunoblot validation 68

Statistical Analysis 69

Results and Discussion

Chapter 1: Low molecular weight FGF2 activation of protein kinase Cs in ischemia/reperfusion

Results I. 71

Expression of only LMW FGF2 results in differential activation of

PKC isoforms epsilon and alpha. 71

Expression of only LMW FGF2 improves post-ischemic

function in isolated mouse hearts. 76

Ablation of PKCα expression abrogates the improvement in

post-ischemic contractility seen when only LMW FGF2 is

expressed. 78

x

PKCε RACK binding peptide, εV-1,prevents PKCε

translocation, while not interfering with PKCα or PKCδ 80

Selective inhibition of PKCε translocation attenuates

the improvement in post-ischemic contractility seen when

only LMW FGF2 is expressed. 83

PKCε and α activation in heart only expressing

LMW FGF2 are not interdependent. 85

Discussion I. 87

Chapter 2: LMW FGF2 and myofibrillar proteins

Results II. 93

PKC isoform translocation to the myofibril is altered during

I/R injury in hearts only expressing LMW FGF2 compared

to wildtype. 93

Troponin phosphorylation is altered in myofibrils of hearts only

expressing LMW FGF2 during ischemia. 96

Changes in troponin phosphorylation in myofibrils of hearts only

expressing LMW FGF2 is abrogated by the in the absence of PKCα 98

Actomyosin ATPase activity is altered during I/R injury in myofibrils

of hearts only expressing LMW FGF2. 100

Actomyosin ATPase activity in myofibrils of hearts only expressing

LMW FGF2 is altered in the absence of PKCα expression but not

in the absence of PKCε activity 102

xi

Discussion II. 105

Chapter 3: LMW FGF2 and calcium handling

Results III. 111

Cardiomyocytes only expressing LMW FGF2 have

depressed calcium cycling. 111

Hearts only expressing LMW FGF2 do not have altered

expression of candidate calcium-handling proteins. 115

Phospholamban phosphorylation during I/R is higher in

hearts expressing LMW FGF2. 117

Phospholamban phosphorylation during I/R in hearts only

expressing LMW FGF2 is not altered in the absence of PKCα 119

CamKII phosphorylation during ischemia is higher in hearts only

expressing LMW FGF2 compared to wildtype 121

Discussion III. 123

Chapter 4: LMW FGF2 and HMW FGF2 crosstalk during ischemia-reperfusion injury.

Results IV. 129

Contractile and relaxation recovery of hearts overexpressing

HMW FGF2 is depressed compared to non-transgenic cohorts 129

FGFR activation during reperfusion is lowered in HMW FGF2

overexpressing hearts 131

xii

PKCs and MAPKs, kinases downstream of FGFR1, have lowered

activation in hearts overexpressing HMW FGF2 during reperfusion 133

Post-ischemic recovery of cardiac function in 24 kDa HMW FGF2

transgenic hearts treated with the selective FGFR1 inhibitor

PD173074 is elevated compared to wildtype cohorts 136

Discussion IV. 138

Chapter 5: LMW and HMW FGF2-induced NO production during I/R

Results V. 141

NO release is increased during I/R in treated hearts overexpressing

all isoforms of FGF2, and dependent on PKC and MAPK activity 141

Discussion V. 146

Chapter 6: LMW and HMW FGF2-regulated gene transcription

Results VI. 147

LMW and HMW FGF2 differentially regulate gene transcription

in non-ischemic hearts 147

Discussion VI. 173

Conclusions and Significance 179

List of Publications and Abstracts 196

References 198

xiii

LIST OF TABLES AND FIGURES Page

FIGURES

Figure 1: Human FGF2 isoform expression 13

Figure 2: FGF2 isoform protein expression in Fgf2 isoform knockout

hearts 41

Figure 3: PKC and FGF2 isoform expression in PKCα and Fgf2 isoform

knockout hearts 45

Figure 4: The isolated working mouse heart 48

Figure 5: Ischemia/reperfusion experimental protocols 51

Figure 6: Microarray study design 66

Figure 7: PKC isoform phosphorylation in HMWKO hearts during I/R injury 73

Figure 8: Functional recovery of hearts only expressing LMW FGF2 77

Figure 9: Functional recovery of HMWKO hearts in the absence of PKCα 79

Figure 10: Translocation of PKC isoforms in the presence of increasing

concentrations of a PKCε inhibitor 81

Figure 11: Translocation of PKC isoforms in the presence of a PKCε inhibitor

during I/R 82

Figure 12: Functional recovery of HMWKO hearts in the presence of a

PKCε inhibitor 84

xiv

Figure 13: PKCε and α isoform crosstalk during I/R in HMWKO hearts 86

Figure 14: Schematic of PKC isoforms’ role in HMWKO hearts during I/R 92

Figure 15: Translocation of PKC isoforms to the myofibril during I/R in

HMWKO hearts 95

Figure 16: Phosphorylation of and in HMWKO hearts 97

Figure 17: Phosphorylation of troponin I and troponin T in HMWKO

hearts in the absence of PKCα 99

Figure 18: Actomyosin ATPase activity, EC50, and Hill coefficient in

HMWKO hearts in the absence of PKCα 103

Figure 19: Actomyosin ATPase activity, EC50, and Hill coefficient in

HMWKO hearts in the presence of PKCε inhibition 104

Figure 20: Schematic representing the role of myofibrillar proteins in

LMW FGF2-mediated protection from postischemic dysfunction 110

Figure 21: Calcium cycling in non-ischemic HMWKO isolated cardiomyocytes 113

Figure 22: Function of non-ischemic HMWKO isolated cardiomyocytes 114

Figure 23: Levels of calcium-handling proteins in non-ischemic HMWKO

hearts 116

Figure 24: Phosphorylation of phospholamban at 17 during I/R in

HMWKO hearts 118

xv

Figure 25: Phospholamban phosphorylation during I/R in the absence of

PKCα in HMWKO hearts 120

Figure 26: CaMKII phosphorylation during I/R in HMWKO hearts 122

Figure 27: Schematic representing LMW FGF2 modulation of calcium-handling

proteins in the heart during I/R 128

Figure 28: Functional recovery of hearts overexpressing HMW FGF2 130

Figure 29: FGFR1 phosphorylation during reperfusion in HMW Tg hearts 132

Figure 30: PKC isoform phosphorylation after I/R injury in HMW Tg hearts 134

Figure 31: MAPK phosphorylation after I/R injury in HMW Tg hearts 135

Figure 32: Functional recovery of hearts overexpressing HMW FGF2

in the presence of FGFR1 inhibitor 137

Figure 33: Schematic representing the influcend of HMW FGF2 on LMW

FGF2 protective signaling during I/R in the heart 140

Figure 34: NO release from hearts overexpressing all isoforms of human

FGF2 in the presence of a PKC inhibitor 143

Figure 35: NO release from hearts overexpressing all isoforms of human

FGF2 in the presence of an ERK1/2 inhibitor 144

Figure 36: NO release from hearts overexpressing all isoforms of human

FGF2 in the presence of a p38 inhibitor 145

xvi

Figure 37: Schematic representing genes altered by FGF2 isoform expression 164

Figure 38: Expression of sarcolipin mRNA in HMWKO hearts 167

Figure 39: Expression of mRNA in HMWKO and LMWKO hearts 168

Figure 40: Expression of calreticulin protein in HMWKO and LMWKO hearts 169

Figure 41: Expression of light chains 7 and 4 mRNA in HMWKO hearts 170

Figure 42: Expression of myosin light chains protein in HMWKO hearts 171

Figure 43: Schematic representing the mechanism of action of the HMW and 180

LMW FGF2 isoforms on postischemic function

TABLES

Table 1: List of genotypes and primers used to identify genetically modified

mice by PCR 37

Table 2: , sources, and concentrations used in immunoblotting 54

Table 3: Actomyosin ATPase activity, maximum efficacy, EC50, and Hill

coefficient in HMWKO hearts during I/R 101

Table 4: Genes altered between HMKWO and Fgf2 KO hearts,

identified by microarray 148

Table 5: Genes altered between LMKWO and Fgf2 KO hearts,

identified by microarray 156

Table 6: Genes altered between LMKWO and HMKWO hearts,

identified by microarray 160

Table 7: Genes identified by the microarray common to both

xvi

LMW-only and HMW-only expression 165

Table 8: Biological categories of FGF2 isoform-regulated genes selected

for further testing 166

Table 9: Targets of microRNA identified by the microarray 172

xvi

LIST OF ABBREVIATIONS

CRT calreticulin

CSQ

FRS FGF receptor substrate

MYL myosin light chain

NLS nuclear localization signal

NCX sodium calcium exchanger

NHE sodium hydrogen exchanger

SH2 src homology domain 2

AU arbitrary units

DAG diacylglyceride

FRS2 FGFR substrate 2

HSPG heparan sulfate proteoglycan

CRF rich FGF receptor

HMW high molecular weight

LMW low molecular weight

PKA

PKG protein kinase G

CaMK kinase

RACK receptor for activated C kinase

IPC ischemic preconditioning

MHC myosin heavy chain

GAPDH glyceraldehyde-3-phosphate dehydrogenase

xix

I-1 protein phosphatase inhibitor 1

MLC myosin light chain

MMP matrix metalloproteinase

RYR2 2

NOS nitric oxide synthase

PCR polymerase chain reaction

PGK phosphoglycerate kinase

PP1 protein phosphatase 1

SLN sarcolipin

SR sarcoplasmic reticulum

ATP adenosine triphosphate

FGF fibroblast growth factor

PKC

MAPK activated protein kinase

ERK extracellular-regulated kinase

JNK c-Jun NH2-terminal kinase

FGFR fibroblast growth factor receptor

I/R ischemia/reperfusion

SERCA sarco- calcium ATPase

TnI troponin I

TnT troponin T

PLB phospholamban

MI myocardial infarction

xx

PIP2 phosphatidylinositol 4,5-bisphosphate

IP3 inositol 1,4,5 triphosphate

PLC phospholipase C

Kd dissociation constant

xxi

INTRODUCTION AND BACKGROUND

1. Cardiac Ischemia-Reperfusion Injury

Since 1919, cardiovascular disease has remained the highest cause of death every year

among the population of the United States (Lloyd-Jones et al., 2010), and is currently the leading cause of death worldwide (World Health Organization, 2008). The largest portion of these deaths is caused by ischemic heart disease, or myocardial infarction (MI). MI is precipitated by a sharp

reduction (ischemia), followed by a subsequent restoration (reperfusion), of blood flow to the heart. It is estimated that ischemic heart disease kills nearly 7.3 million people every year (World

Health Organization, 2008) throughout the world. Current therapies, such as thrombolytics, anticoagulants, vasodilators, or beta adrenergic antagonists, focus on restoring blood flow or slowing the demand of the heart for oxygen and nutrients, but do not treat the injured heart muscle itself (Kloner & Schwartz Longacre, 2011). Despite the clear and profound need for pharmacological intervention, there are few clinically available therapeutics that directly target the ischemic heart.

In the healthy heart, contraction is initiated in the right atrium by the depolarization of

specialized heart muscle cells (Irisawa, 1987). A periodic inward sodium ‘pacemaker’ current in

these cells results in a slight depolarization of the cells, which, after reaching a sufficient

threshold, opens voltage-gated sodium channels, allowing sodium to flood the cell and initiating

excitation/contraction (EC) coupling (Irisawa, 1987). The inward flux of positively charged

sodium ions further depolarizes the cells to a sufficient degree both to initiate depolarization in

neighboring myocytes (allowing the to propagate through the heart from base to

apex), as well as to open voltage-gated calcium channels, allowing cytosolic calcium

concentrations to sharply increase (Amin, Tan, & Wilde, 2010). This effect is positively

reinforced by calcium-activated calcium channels at the membrane of the sarcoplasmic

1

reticulum, which allow intracellular calcium stores to be released and further increase

intracellular calcium concentrations (Besch & Watanabe, 1975; Fabiato, 1983). It is this increase

in cytosolic calcium that activates the contraction machinery of the cell, binding to and activating

actomyosin ATPase at the thick filament (Bozler, 1952), and relieving the inhibition of

on myosin binding of through troponin binding at the thin filament (Greene &

Eisenberg, 1988; Schaub, Hartshorne, & Perry, 1967). These effects allow crossbridge cycling to

take place, which consumes high levels of ATP to generate force, pulling the thick filament

along the thin filament towards the Z-disk and contracting the cell (Solaro, 1975). Repolarization

of the cell is achieved through opening of voltage-gated potassium channels during

depolarization, which release positively charged potassium from the cell (Amin et al., 2010).

During this period, voltage-gated calcium channels are inactivated (Kohlhardt, Krause, Kubler,

& Herdey, 1975), and intracellular calcium concentrations rapidly drop as calcium is removed

from the cell, or resequestered back in the sarcoplasmic reticulum via the sarco(endo)plasmic

reticulum calcium ATPase (SERCA) (Barry & Bridge, 1993). This drop in calcium prevents

crossbridge cycling from taking place and the heart relaxes as the cell expands back to its

uncontracted length (Fanburg, Finkel, & Martonosi, 1964).

Cardiac ischemia/reperfusion (I/R) is characterized by a unique set of intracellular conditions in the cardiomyocyte, ultimately leading to decreased function, arrhythmia, and tissue death in the heart (Jennings & Reimer, 1991). At the onset of ischemia, the region of the heart that is deprived of blood flow quickly loses oxygen and nutrients (Horwitz, Sayen, Sheldon, &

Kuo, 1950). Within these ischemic cells, ATP is rapidly depleted in the absence of F0/F1-

mediated ATP generation, which is dependent on the presence of oxygen (Arnold M., 1973). As

a result, the heart switches from oxidative phosphorylation as a primary source of energy to

anerobic glycolysis to meet the ATP demand (Kubler & Spieckermann, 1970). However, as a

2

byproduct of glycolysis are hydrogen ions, the pH of the cell drops (Arnold M., 1973), activating

sodium hydrogen (NHE) exchangers in the (Tani & Neely, 1989). This, in turn, results in an increase in the intracellular sodium concentration (Pike, Kitakaze, & Marban, 1990),

which triggers the reversal of the sarcolemmal sodium-calcium exchanger (NCX) (Tani & Neely,

1989). NCX channels in normal cardiomyocytes remove calcium from the and bring in sodium, but during ischemia, the high intracellular sodium levels initiate the extrusion of sodium and the intake of calcium (Grinwald, 1982). The intracellular calcium concentration subsequently increases rapidly after the onset of ischemia (Akutsu et al., 2010; Shen & Jennings,

1972).

Calcium overload is a primary effector of I/R injury, with a sharp rise in intracellular

calcium correlating to the degree of injury (Steenbergen, Fralix, & Murphy, 1993). Calcium

overload results in the hypercontracture of the myofibril during ischemia and the activation of

calcium-dependent (Bolli, 1990; Dong Gao, Liu, Mellgren, & Marban, 1996; Opie,

1989; Piper, 1989). Calcium-induced activation of has been associated with the

degradation of fodrin (Yoshida et al. 1995) and ankyrin R (Yoshida & Harada, 1997), which

leads to a decrease in contractile function (Tsuji et al. 2001), as well as with the degradation of

ryanodine receptors (Pedrozo et al, 2010). In addition, calcium primes several metabolic

in the mitochondria, resulting in the rapid production of reactive oxygen species at the

onset of reperfusion that further result in intracellular oxidative damage to lipid membranes and

proteins, as well as mitochondrial dysfunction, ultimately leading to the apoptosis of the

cardiomyocyte (Farber, Chien, & Mittnacht, 1981; Garcia-Rivas & Torre-Amione, 2009; Nayler,

1983).

At the level of the heart, these intracellular responses can manifest themselves in several

ways. The first of these is a prolonged but reversible mechanical dysfunction of the myocardium,

3

also known as stunning. First characterized in reperfused dog hearts by Heyndrickx and

colleagues in 1975 (Heyndrickx, Millard, McRitchie, Maroko, & Vatner, 1975), myocardial stunning is thought to be the result of decreased myofilament responsiveness to intracellular calcium cycling, due both to calcium overload and reactive oxygen species generation (Bolli &

Marban, 1999). In the stunned myofibril, decreased maximal calcium-induced force (Carrozza et al., 1992; Gao, Atar, Backx, & Marban, 1995; Kusuoka, Koretsune, Chacko, Weisfeldt, &

Marban, 1990) and a decreased sensitivity to calcium, a rightward shift in activity in response to intracellular calcium concentration (Gao et al., 1995; Miller, McDonald, & Moss, 1996), have

been implicated as mechanisms for cardiac dysfunction. Stunning has a clear clinical role in

morbidity and mortality of MI patients, resulting in possible hemodynamic instability and

cardiogenic shock, particularly in patients suffering from repeated undetected ischemic episodes,

which may result in chronically depressed left ventricular dysfunction (Bolli, 1990).

In addition to reversible functional injury, ischemic areas of the heart may also develop

an irreversible infarct, or region of dead tissue. Tissue death may occur as a result of apoptosis,

necrosis, or autophagy (Y. Dong, Undyala, Gottlieb, Mentzer, & Przyklenk, 2010; Machado,

Alves, Carvalho, & Oliveira, 2009; Whelan, Kaplinskiy, & Kitsis, 2010). Reperfusion injury is associated with an increase in the production of ROS at the mitochondria (Ambrosio et al.,

1993), which may trigger opening of the mitochondrial permeability transition pore (MPTP) and necrosis (Baines et al., 2005) as well as cytochrome C release and caspase 3 activation,

ultimately leading to initiation of the intrinsic apoptosis cascade (Freude et al., 2000). This area

of tissue death may result in irreversible impaired cardiac function. Additionally, it may present a

conduction block to the propagation of the action potential, leading to the final manifestation of

I/R injury, which is the development of arryhthmia (Mont et al., 1996). Myocardial ischemia has

been identified as having a strong correlation with ventricular arrhythmia and sudden cardiac

4

arrest (European Heart Rhythm Association et al., 2006). Arrhythmias, which develope during

ischemia, have profound implications for the morbidity and mortality of ischemic heart disease

patients (Hatzinikolaou-Kotsakou et al., 2007).

As myocardial ischemia and reperfusion have such an extensive impact on public health and quality of life, substantial resources have been directed towards developing therapeutic interventions that may improve the outcome of patients with ischemic heart disease. Any pharmacological agent that results in a reduction in stunning, infarct development, and/or arrhythmia has the potential to significantly reduce the morbidity and mortality of patients suffering from MI. One potential therapeutic molecule is fibroblast growth factor 2 (FGF2).

2. Fibroblast Growth Factors and FGF2

Fibroblast growth factor 2 (FGF2) is one of 22 FGFs identified. FGFs were first

discovered as a mitogen for 3T3 fibroblasts (Gospodarowicz, 1974), and have since been found

in a number of organisms, including mammals, birds, fish, amphibians, fruit flies, and worms.

FGFs in vertebrates range from 17 to 34 kDa in size and share a conserved sequence of 120

amino acids with roughly 30-60% (Itoh & Ornitz, 2011). Despite this homology, the FGFs produce a wide range of biological effects. All FGFs are broadly classified

according to their mechanisms of action as traditional, hormonal (hFGFs), or intracellular

(iFGFs) (Itoh & Ornitz, 2008). Traditional FGFs have biological activities that are largely

receptor mediated, and have paracrine/autocrine effects (Itoh & Ornitz, 2008). Hormonal FGFs

also produce biological effects via receptors, but are transmitted through the bloodstream and

therefore act as endocrine factors (Itoh & Ornitz, 2008). Intracellular FGFs are not released from

the cell, and their effects are not receptor-mediated (Itoh & Ornitz, 2008). FGF2, or basic FGF,

falls within the traditional classification of FGF, and primarily produces its effects via both high

5

affinity FGF receptor (FGFR) binding, and low-affinity heparan sulphate proteoglycan (HSPG) binding (Soulet et al., 1994).

FGF2 was first identified as a brain and pitutary-derived mitogen for fibroblasts

(Gospodarowicz, 1975). FGF2 has been shown to modulate cell proliferation, migration, motility, differentiation, adhesion, and death signaling in a number of cell types (Black, Logan,

Davis, & Sheppard, 1990; Fei, Xiao, Doetschman, Coffin, & Hurley, 2011; Folkman et al., 1988;

Grunz et al., 1988; Iwamoto, Shimazu, Nakashima, Suzuki, & Kato, 1991; Jackson & Reidy,

1993; Kondo, Matsuda, & Yonezawa, 1993; Mignatti, Morimoto, & Rifkin, 1991; Miller-Davis,

McKeehan, & Carpenter, 1988; Reidy, 1993; Sato, Abe, & Takaki, 1990; Taylor, Greenberg,

Turley, & Wright, 1993; Westermann, Grothe, & Unsicker, 1990; Yayon & Klagsbrun, 1990). In addition, FGF2 has been implicated in development, wound healing, vasculogenesis, angiogenesis, and arteriogenesis (Cox & Poole, 2000; Davidson & Broadley, 1991; Fiddes et al.,

1991; Gendron, Tsai, Paradis, & Arceci, 1996; Hebda, Klingbeil, Abraham, & Fiddes, 1990;

Knighton, Phillips, & Fiegel, 1990; Lyons, Anderson, & Meyer, 1991; Reidy & Lindner, 1991;

Weckbecker, Liu, & Tolcsvai, 1992). FGF2 is expressed in cardiomyocytes (Speir et al., 1992;

Spirito, Fu, Yu, Epstein, & Casscells, 1991), and has been shown to play a number of roles in the heart, including the regulation of the development of the heart, the remodeling response of the heart to injury, and the acute response of the heart to ischemia/reperfusion (Detillieux, Sheikh,

Kardami, & Cattini, 2003; Liao et al., 2009). FGF2 is found in developing embryonic heart valves, particularly the ateriovenous and outflow tract mesenchymal nuclei (Liao et al., 2009). In addition, FGF2 is important for epicardial and cardiac vasculature development (including aortic arch smooth development) (Lavine et al., 2005; Liao et al., 2009; Merki et al., 2005), as well as epithelial-mesenchymal cell transformation (Morabito, Dettman, Kattan, Collier, &

Bristow, 2001). Overexpression of FGF2 in cardiomyocytes has been demonstrated to exacerbate

6

the inflammatory response to injurious administration of isoproterenol (Meij et al., 2002).

Hypertrophy of the heart due to increased hemodynamic load or hypertension has been shown to

upregulate both FGF2 and FGFR1 (Hellman et al., 2008; Spruill et al., 2008; Spruill, Baicu, Zile,

& McDermott, 2008), and ablation of FGF2 expression reduces the cardiac hypertrophic

response to transverse aortic coarctation (Schultz et al., 1999) as well as to angiotensin II

treatment (Pellieux et al., 2001).

FGF2 also affects the development of blood vessels, both during development

(vasculogenesis) as well as adaptive growth of blood vessels in the adult organism (angiogenesis

and arteriogenesis). The migration and formation of tube-like structures by endothelial cells in

vitro is induced by the application of FGF2 (Montesano, Vassalli, Baird, Guillemin, & Orci,

1986). Disruption of FGF2 expression alters vascular development in murine cells as well

(Leconte, Fox, Baldwin, Buck, & Swain, 1998), although Sullivan and colleagues found no

alteration in the regrowth of blood vessels in the ischemic hindlimbs of mice lacking FGF2

(Sullivan, Doetschman, & Hoying, 2002). In addition to increasing blood flow via the development of new blood vessels, FGF2 has been shown to have vasodilatory effects that are mediated by nitric oxide (NO) production and ATP-sensitive potassium channels (Kajita,

Takayasu, Yoshida, Dietrich, & Dacey, 2001; Tiefenbacher & Chilian, 1997; Unger et al., 1994).

FGFs, including FGF2, bind to two types of receptors, with high affinity to the fibroblast

growth factor receptor (FGFR1-4), and with lower affinity to heparan sulfate proteoglycans

(HSPGs) (Klagsbrun & Baird, 1991). Recent evidence has shown that some of FGF2’s effects

require binding to both kinds of receptors (Guimond & Turnbull, 1999; Schlessinger et al., 2000;

X. Xu et al., 2007). In adult humans, FGFR1 and 4 are found on cardiomyocytes, while in

rodents only FGFR1 is expresssed (Cool, Sayer, van Heumen, Pickles, & Nurcombe, 2002;

Hughes, 1997; Kardami, Liu, Pasumarthi, Doble, & Cattini, 1995). FGF2 binds with high affinity

7

and activates both FGFR1 (Kd: 0.24nM) and FGFR4 (Kd: 2.5nM) (Vainikka et al., 1992). FGFR

is a typical receptor tyrosine kinase (RTK), with a ligand binding extracellular domain, a single

helical transmembrane domain, and intracellular tyrosine kinase region (Plotnikov, Schlessinger,

Hubbard, & Mohammadi, 1999; Schlessinger et al., 2000). As FGF2 binds to the FGFR, the

receptor dimerizes and autophosphorylates at several tyrosine residues, resulting in receptor

activation (Plotnikov et al., 1999; Schlessinger et al., 2000). Mohammadi and colleagues (M.

Mohammadi et al., 1996) have mapped seven phosphorylatable tyrosine residues to FGFR1;

among these, Y653 and Y654 lie in the kinase domain and are phosphorylated first, which

induce autophosphorylation of subsequent tyrosine residues. The stimulated receptor then

recruits and activates several docking proteins containing src homology (SH-2) domains or

phosphotyrosine binding domains (Dhalluin et al., 2000; Karlsson et al., 1995; Landgren, Klint,

Yokote, & Claesson-Welsh, 1998; Larsson, Klint, Landgren, & Claesson-Welsh, 1999).

Phospholipase C (PLC) γ and Shb are two of the former, while SHC and FRS2 (FGFR substrate

2) are two of the latter.

PLCγ is activated from binding to the phosphorylated tyrosine Y766 on the active

FGFR1 (M. Mohammadi, Schlessinger, & Hubbard, 1996). Active PLCγ then cleaves PIP2 into

inositol 1,4,5 triphosphate (IP3) and diacylglycerides (DAG) (Low & Finean, 1976). DAGs may

then activate PKCs, and IP3 regulates calcium release from intracellular organelles (Williamson,

1986). Shb also binds to Y766, which in turn phosphorylates and activates FRS (Cross et al.,

2002), that has its own juxtamembrane on the receptor (H. Xu, Lee, & Goldfarb,

1998). SHC binds to Y653 and Y654, and both Shb and FRS function as scaffolding proteins that

recruit GRB2-SOS to the membrane, a complex which in turn binds to RAS (Curto, Frankel,

Carrero, & Foster, 1998; Klint, Kanda, & Claesson-Welsh, 1995; M. Mohammadi et al., 1996;

Ryan, Paterno, & Gillespie, 1998; Schuller et al., 2008). Membrane-associated RAS then binds

8

to the /threonine kinase RAF-1, which initiates a kinase cascade leading to the activation of

mitogen-activated protein kinase (MAPK) (N. G. Williams et al., 1993). RAF-1 acts as a MAPK

kinase kinase (MAPKKK), which phosphorylates and activates MAPK kinases (MAPKKs), such

as MEK (Johnson & Vaillancourt, 1994). MAPKKs, in turn, phosphorylate and activate MAPKs,

such as ERK1/2, p38, and JNK, which phosphorylate a number of intracellular targets with a

wide range of biological activities (Johnson & Vaillancourt, 1994).

In addition to FGFRs, most FGFs bind to both serum heparin and HSPGs on the cell

surface. Several kinds of HSPGs have been shown to be necessary for mediating the effects of

FGFs, including syndecans, glypican, and perlecan, (Chuang et al., 2010; Filla, Dam, &

Rapraeger, 1998; Horowitz, Tkachenko, & Simons, 2002; Iwabuchi & Goetinck, 2006; Iwabuchi

& Goetinck, 2006; Su et al., 2006; Yan & Lin, 2007). There is evidence that various FGFs may

bind HSPGs with different affinities; for example, glypican reduces the response to FGF7, while

promoting the mitogenic activities of FGF2 (Bonneh-Barkay et al., 1997). Recent evidence

shows that FGF2 binds with higher affinity to highly sulfated HSPGs compared to HSPGs with low sulfation ((Naimy, Buczek-Thomas, Nugent, Leymarie, & Zaia, 2011). HSPG may have a

number of biological roles in FGF signaling, including determining the distribution of the growth

factor, concentration of the growth factor at the cell surface, and protection from degradation

(Faham, Linhardt, & Rees, 1998; Rahmoune, Chen, Gallagher, Rudland, & Fernig, 1998;

Sasisekharan, Ernst, & Venkataraman, 1997; Vlodavsky, Miao, Medalion, Danagher, & Ron,

1996). Perhaps most importantly, HSPGs have been shown to facilitate the binding of FGF to

FGFR in a ternary complex and inducing FGFR dimerization (McKeehan, Wu, & Kan, 1999;

Schlessinger et al., 2000).

FGFs may also bind to transmembrane sialoglycoproteins known as cysteine-rich FGF

receptors (CRF), although the biological activities of these proteins are not well understood.

9

Primarily located at the Golgi apparatus, recent studies suggest that CRF may be cleaved by

furin-like proprotein convertases and subsequently secreted (Antoine et al., 2009). The FGF

binding site of these proteins overlaps with that for FGFR, suggesting that CRF may compete

with and diminish FGFR signal transduction. Consistent with this hypothesis, overproduction of

CRF in Chinese hamster ovary cells reduced intracellular FGF1 and FGF2 levels (Zuber, Zhou,

Burrus, & Olwin, 1997). Also, some hormonal FGFs, such as FGF23, FGF15/19, and FGF21

have been shown to elicit their biological actions via klotho, a transmembrane , in

conjunction with traditional FGFRs (Kurosu et al., 2006; Kurosu et al., 2007; Sinha et al., 2008;

Urakawa et al., 2006); disruption of the klotho-FGF interaction is associated with renal and cardiovascular disease (Bernheim & Benchetrit, 2011), and alterations in and calcium homeostasis in bone (Wohrle et al., 2011).

In the heart, both HSPGs and FGFR1 and 4 are expressed (Cool et al., 2002; Hughes,

1997; Kardami et al., 1995), allowing FGF2 to exert biological functions on the myocardium

(Strunz et al., 2011). FGF2, binding to these receptors, has been shown to have a profound

impact on the heart during I/R injury (Jiang et al., 2002; Liao et al., 2010).

3. FGF2 and Cardioprotection

The first indication that FGF2 plays a role in I/R injury can be found by examining the

response to the heart after an ischemic attack. FGF2 is released from the heart during ischemia

and reperfusion, resulting in increased serum levels (Cuevas et al., 1997; Hasdai et al., 1997;

Weihrauch, Tessmer, Warltier, & Chilian, 1998), suggesting that FGF2 may be act as a protector

of the heart during I/R. FGF2 has the potential to protect the heart in the long term by inducing

arteriogenesis and angiogenesis, increasing blood flow to ischemic organs. In a canine model of

permanent ischemia, administration of recombinant FGF2 was shown to increase blood flow and

10

collateral vessel formation (Unger et al., 1994), and increased capillary density in the

myocardium (Horrigan, Malycky, Ellis, Topol, & Nicolini, 1999). FGF1, a homologue of FGF2

which binds to FGFR1 with similar affinity, has been shown to increase vascular density and

increase cardiac blood flow after I/R injury. In addition to its vascular effects, it has been well

established that, FGF2 can protect the heart from acute I/R injury independent of increasing

blood flow, reducing infarct size and preventing dysfunction both when administered

exogenously or overexpressed (Cuevas, Carceller, Martinez-Coso, Asin-Cardiel, & Gimenez-

Gallego, 2000; House et al., 2003; Padua, Sethi, Dhalla, & Kardami, 1995; Sheikh, Sontag,

Fandrich, Kardami, & Cattini, 2001). Importantly, FGF2 has been shown to protect the heart

when administered after the onset of ischemia (Horrigan et al., 1999; Jiang et al., 2002; Jiang,

Srisakuldee, Soulet, Bouche, & Kardami, 2004), making it a viable candidate for treating MI

patients in a clinical setting where one cannot predict when the onset of ischemia may occur. The

cardioprotective effects of FGF2 have been shown to be mediated through FGFR (Jiang et al.,

2002; Liao et al., 2010), and FGF1, which also binds FGFR1 with high affinity, produces the

same protective phenotype independent of its angiogenic effects (Cuevas et al., 1997). Potential

downstream mediators of FGF2/FGFR1 protective signaling include PKCs and MAPKs, such as

p38, JNK, and ERK1/2 (House, Branch, Newman, Doetschman, & Schultz Jel, 2005; House,

Melhorn, Newman, Doetschman, & Schultz Jel, 2007; Jiang et al., 2002; Liao et al., 2007), as

well as other downstream pathways that have not been evaluated in this dissertation, such as

phosphoinositide-3-kinase/AKT (PI3K/AKT), which can be activated by phospholipase C (G.

W. Dorn 2nd & Force, 2005) and Ras (Mendoza, Er, & Blenis, 2011).

Elucidation of these mechanisms of action of FGF2 is necessary for the development and use of FGF2 as a therapeutic agent for MI patients. The first step in understanding the signaling

11

pathways involved require insight into the different classes of protein isoforms, high molecular weight and low molecular weight, that make up FGF2.

4. High and low molecular weight FGF2

There are two classes of FGF2 protein isoforms: high molecular weight (HMW) and low molecular weight (LMW), which are produced from the same gene and the same mRNA (Figure

1) (Florkiewicz & Sommer, 1989). HMW and LMW FGF2 have nearly identical primary structure with the exception of an additional N-terminus sequence possessed by all

HMW FGF2 isoforms (Florkiewicz & Sommer, 1989). LMW FGF2 is translated from a traditional AUG start site, and in both humans and mice consists of a single 18 kDa isoform

(Florkiewicz & Sommer, 1989). Alternately, HMW FGF2 is comprised of multiple protein isoforms, two isoforms (21.5 and 22 kDa) in mice and four (21.5, 22, 24, and 34 kDa) in human

(Florkiewicz & Sommer, 1989). These isoforms are all the result of CUG translational start sites upstream of the AUG site that produces LMW FGF2 (Florkiewicz & Sommer, 1989). These isoforms contain a nuclear localization sequence (NLS) that initiates translocation of these isoforms to the nucleus of the cell (Bugler, Amalric, & Prats, 1991; Renko, Quarto, Morimoto, &

Rifkin, 1990). Conversely, LMW FGF2 is found primarily in the cytosol (although recent evidence demonstrates that it is in the nucleus) (Liao et al., 2010; Renko et al., 1990) and is released to the extracellular space in the heart (Liao et al., 2010). As FGF2 lacks a traditional

ER/Golgi export signal (Mignatti, Morimoto, & Rifkin, 1992), the mechanism of the release of

FGF2 from the cell is not well characterized, although recent publications by Nickel and investigators suggest that Tec kinase and phosphatidylinositol 4,5 bisphosphate (PIP2)

12

Human FGF2 mRNA CUG1 (319) CUG0 (86) CUG2 (346) NLS AUG (486) Stop(951) CUG3 (361) 5’ NLS 3 Protein 34 kDa HMW 24 kDa nuclear 22 kDa 21 kDa

LMW 18 kDa cytosolic

Figure 1. Human FGF2 mRNA and protein isoform expression. Nuclear localization signals

(NLSs) and stop codon are indicated. All isoforms of FGF2 are expressed from a single mRNA. 4 HMW isoforms result from upstream translational start cites (CUG0, CUG1,

CUG2, CUG3), while a single LMW isoform is translated from the AUG start site.

13

play an important role (Ebert et al., 2010). Until recently, most research did not distinguish

between the two classes of FGF2 protein isoforms, evidence suggests that LMW and HMW

FGF2 have unique biological roles, resulting in distinct (Quarto, Fong, &

Longaker, 2005), and may even regulate each other’s biological activities (Estival et al., 1996).

Until recently, the role of the different protein isoforms of FGF2 in mediating protection

from I/R injury was largely unknown, and the effects of LMW and HMW FGF2 were not

distinguished, although many of the studies performed in the previous two decades using

recombinant FGF2 were likely evaluating only the LMW FGF2 isoform. However, in view of the fact that HMW and LMW FGF2 can have differential effects on various cell types (Bikfalvi et al., 1995), the biological effects of LMW FGF2 independent of HMW FGF2, and vice versa, in the cardiovascular system are critical to delineate. Our laboratory has shown that endogenous expression of only the LMW isoform of FGF2 significantly enhances (>80% recovery) cardiac function after I/R, while endogenous expression of only the HMW isoform, as well as overexpression of the HMW isoform worsen post-ischemic cardiac dysfunction (Liao et al.,

2007; Liao et al., 2010). In addition, investigation by Kardami and group (Hirst, Herlyn, Cattini,

& Kardami, 2003) has shown that expression of HMW FGF2 results in chromatin compaction and cytotoxicity in cardiomyocytes. Conversely, the cardiac effects of exogenously added HMW

FGF2 have a response to I/R injury that closely resembles that of LMW (Jiang et al., 2009), suggesting that the localization of HMW FGF2 is important to its effects. Studies from our laboratory have shown that, while LMW FGF2 is released, via an unknown mechanism, by the myocyte, endogenously expressed HMW FGF2 is retained in the cell (Liao et al., 2010). This prevents HMW FGF2 from binding to its high affinity receptor on the extracellular side of the sarcolemma, which is necessary for producing the protective effects of LMW FGF2 (Liao et al.,

2010).

14

A critical element in the delineation of the roles of LMW and HMW FGF2 in the heart

during I/R, is the elucidation of downstream signaling pathways triggered during and after I/R

injury. An important candidate protein to consider for this role is protein kinase C (PKC).

5. Protein Kinase C

Protein kinase C, first described by Takai and colleagues (Takai, Yamamoto, Inoue,

Kishimoto, & Nishizuka, 1977), is a family of serine/threonine kinases that are expressed throughout the body. PKCs have an AGC kinase structure also found in PKA and PKG, which

all share a highly conserved ATP binding catalytic domain and an inactivating regulatory domain

(Steinberg, 2008). PKCs are comprised of a single polypeptide that contains both domains joined

by a hinge region, with the regulatory sequence at the N-terminus, and the catalytic domain at the

C-terminus (Steinberg, 2008). The C-terminal catalytic domain recognizes a number of potential

phosphorylation consensus sites, including (R/K1-3, X2-0)-S*/T*- (X2-0, R/K1-3), S*/T*-(X2-0,

R/K1-3), and ( R/K1-3, X2-0)-S*/T* (Kennelly & Krebs, 1991). The regulatory domain of PKCs contains a pseudosubstrate sequence that binds to its catalytic domain and prevents it from phosphorylating its targets. This regulatory subunit also contains a number of domains that bind selectively to activators of the kinase, which stabilize the regulatory region in an “open” conformation and free the catalytic subunit for activity. In particular, the PKC regulatory region may contain one or more C1 domains, which contain a HX12CX2CX(13-14)CX2CX4HX2CX7C sequence (H, histidine, C, cysteine) bind to phorbol esters/DAGs (Hurley, Newton, Parker,

Blumberg, & Nishizuka, 1997; Steinberg, 2008; R. X. Xu, Pawelczyk, Xia, & Brown, 1997).

Additionally, they may contain a calcium-binding , which contain several highly conserved Asp residues; Asp187, Asp193, Asp246, Asp248, and Asp254 have been identified in

PKCα (Luo & Weinstein, 1993; Medkova & Cho, 1998).

15

PKCs are divided into three subfamilies, depending on which of these regulatory subunits

they contain. Conventional PKCs (cPKCs), which include PKCα, β1, β2, and γ, contain both C1

and C2 domains; as such, they are activated by both DAGs and calcium (Medkova & Cho, 1998;

Rotenberg et al., 1998; Zhu, Hansen, Su, Shieh, & Riedel, 1994). Novel PKCs (nPKCs), PKCδ,

ε, η, and θ, lack the C2 domain and are therefore, calcium independent (Chang, Xu,

Raychowdhury, & Ware, 1993). However, nPKCs contain two C1 domains, and are strongly

activated by DAGs (Chang et al., 1993). Atypical PKCs (aPKCs), PKCι/λ and ζ contain no C2

domain, and an altered C1 domain that does not bind DAGs, and hence are activated by neither

calcium nor DAGs (C. C. Chen, 1993; Johannes, Prestle, Eis, Oberhagemann, & Pfizenmaier,

1994). aPKCs instead become activated by recruitment by binding partners to the membrane,

where phosphatidylserine (PS) stabilize the enzyme in an active form by binding to the C1

domain (Puceat & Vassort, 1996).

In addition to these allosteric activators, PKCs can become activated by phosphorylation.

PKCs are first phosphorylated at a threonine in the activation loop, which is a highly conserved

peptide sequence in the region of the substrate binding domain (Cazaubon, Bornancin, & Parker,

1994). This phosphorylation site is analogous to T197 in PKA, which is necessary for aligning

the catalytic site of the enzyme (Knighton et al., 1991). The T197 site may be phosphorylated by

3-phosphoinositide-dependent protein kinase 1 (PDK-1), or through an autocatalytic mechanism

(Moore, Kanter, Jones, & Taylor, 2002; M. R. Williams et al., 2000). Once the PKC is phosphorylated at this priming motif, the kinase undergoes two autophosphorylations on the kinase domain at the C-terminus, on the turn motif and hydrophobic motif of the V5 domain, stabilizing the active enzyme (Behn-Krappa & Newton, 1999; Stensman & Larsson, 2007; J.

Zhang, Wang, Petrin, Bishop, & Bond, 1993).

Upon activation, PKC isoforms translocate from the cytosol to membranous structures in

16

the cell, allowing the C1 regions of the activated kinases to bind to PS, and placing them in close

proximity to membrane-bound targets (R. M. Bell, Hannun, & Loomis, 1986). The translocation

of PKC is dependent on both the isoform and the stimulus, suggesting that the localization of the

activated kinase may impart a degree of selectivity to the target (Puceat & Vassort, 1996).

Studies by the laboratory of Mochly-Rosen (D. Schechtman & Mochly-Rosen, 2001) have

focused on membrane associated receptors for activated C kinase (RACKs), which bind to the

variable region of the amino-terminus of the kinase. Thus far, RACKs for PKCβ and PKCε have

been identified, which colocalize with, induce translocation to, and increase the activity of their

respective kinases without acting as a substrate (Csukai, Chen, De Matteis, & Mochly-Rosen,

1997; Ron et al., 1994; D. Schechtman et al., 2004). In addition, a RACK for PKCδ has been

described, although it should be noted that this RACK has been shown to bind to PKCθ in both

the inactive and active conformation (Robles-Flores et al., 2002).

While allosteric activator binding, phosphorylation, and translocation are the most well

understood mechanisms of PKC activation, alternate non-canonical activation pathways have been described. Caspase may cleave PKC during apoptosis, leading to separation of the regulatory and kinase domains and resulting in an activated kinase (Basu et al., 2002). In addition, reactive oxygen species may result in the activation or inactivation of PKCs, by oxidation of amino acids in the C1 regulatory domain in the former case (Gopalakrishna, Chen,

& Gundimeda, 1997), or targeting the catalytic region in the latter (Gopalakrishna et al., 1997).

Finally, tyrosine nitration of several PKC isoforms has been shown to inhibit kinase activity

(Knapp, Kanterewicz, Hayes, & Klann, 2001).

The ubiquity of PKCs results in a vast number of intracellular roles, and research from

the past two decades has implicated several PKC isoforms as key players in the normal

regulation of the heart, as well as the response of the heart to an ischemic insult.

17

6. PKCs and cardioprotection

The role of PKCs in I/R injury has been the subject of numerous studies. Several PKCs

have been found to become activated by I/R, and have been found to contribute or mitigate the

injurious effects of ischemia (Budas, Churchill, & Mochly-Rosen, 2007; Murphy & Frishman,

2005). During ischemia, PKCα, δ, and ε isoforms translocate to the membrane fraction in rat

hearts (Yoshida et al., 1996). Administration of an activator of PKCε prior to ischemia protects

the heart from I/R injury, while addition of an inhibitor of PKCδ is protective (G. W. Dorn et al.,

1999; K. Inagaki et al., 2003). Co-administration of a PKCε pseudo-RACK peptide activator and

a PKCδ RACK blocker produces an additively protective effect (K. Inagaki, Hahn, Dorn, &

Mochly-Rosen, 2003).

In addition to their role in ischemia/reperfusion injury, several PKCs have been shown to mediate the effects of other cardioprotective stimuli (Bouwman, Musters, van Beek-Harmsen, de

Lange, & Boer, 2004; C. H. Chen, Gray, & Mochly-Rosen, 1999; Das, Ockaili, Salloum, &

Kukreja, 2004; Fryer, Wang, Hsu, & Gross, 2001; Hassouna, Matata, & Galinanes, 2004; House et al., 2007; K. Inagaki et al., 2003; Jiang et al., 2002; Kanaya, Gable, Murray, & Damron, 2003;

Kaneda et al., 2008; Kudo, Wang, Xu, Ayub, & Ashraf, 2002; Maslov et al., 2009; Melling,

Thorp, Milne, & Noble, 2009; Miki, Cohen, & Downey, 1998; Oestreich et al., 2009; Okusa et al., 2009; T. T. Pan, Neo, Hu, Yong, & Bian, 2008; P. Ping et al., 1997; Qiu et al., 1998; Saurin et al., 2002; Takahashi et al., 2007; Turrell, Rodrigo, Norman, Dickens, & Standen, 2011;

Uecker et al., 2003; Wickley, Ding, Murray, & Damron, 2006; H. Y. Zhang et al., 2002; H. Z.

Zhou, Karliner, & Gray, 2002). Importantly, the PKC isoforms involved are stimulus-dependent, and the isoform necessary for mitigating I/R injury under one type of stimulus may not be involved, or even may be detrimental to the cardiac outcome under another stimulus. An

18

example of this is demonstrated by one of the most-studied cardioprotective interventions, ischemic preconditioning (IPC), which is the protection of the heart from I/R injury through the

administration of short periods of ischemia prior to a longer ischemic event. First described by

Murry and colleagues. in 1986 (Murry, Jennings, & Reimer, 1986), IPC has since been

determined to occur in two phases, early and late phase preconditioning (Guo et al., 1998; Qiu et

al., 1997). Early phase preconditioning refers to the window of protection offered acutely (1-2

hours) after the initial protective short ischemic events (Murry et al., 1986). This phase is not

dependent on the synthesis of new proteins, and is thought to be the result of rapid ,

i.e. phosphorylative and nitrosylative events (Thornton et al., 1990). Late phase preconditioning occurs 24 hours after the initial ischemic events, and is thought to be the result of transcriptional

regulation and the production of protective proteins (Bolli, 2000; Yamashita, Hoshida,

Taniguchi, Kuzuya, & Hori, 1998). PKC isoforms have been implicated in both phases of IPC.

Classical ischemic preconditioning is mediated by adenosine, bradykinin, and opioid

receptor stimulation, in a manner that is PKC-dependent (Goto et al., 1995; Miki et al., 1998;

Sakamoto, Miura, Goto, & Iimura, 1995; Schultz, Rose, Yao, & Gross, 1995). In particular,

PKCε appears to play the most important role in IPC. PKCε has been shown to translocate to

membrane fractions after IPC, along with PKCη in rabbits (P. Ping et al., 1997) and PKCδ in rats

(Uecker et al., 2003). IPC is dependent on the activation of PKCε as IPC produced no effect in

PKCε knockout mice (Saurin et al., 2002). Additionally, administration of a PKCε inhibitor

abrogated the acute effects of IPC in neonatal (Gray, Karliner, & Mochly-Rosen, 1997) and adult

(G. S. Liu, Cohen, Mochly-Rosen, & Downey, 1999) cardiomyocytes. Late phase

preconditioning also requires activation of PKCε (Qiu et al., 1998), as well as endothelial nitric

oxide synthase (eNOS) appears to be instrumental in its activation to produce these late stages of

cardioprotection (Xuan et al., 2007). In addition to PKCε, PKCα has been demonstrated by

19

Schulz and colleagues to be necessary for IPC, acting downstream of PKCε (Schulz et al., 2003)

and phosphorylating 43 (Doble, Ping, & Kardami, 2000).

Although IPC significantly protects the heart from I/R injury, its clinical relevance is

limited by the fact that it is an invasive technique, and must be administered prior to the onset of

ischemia. Several laboratories have therefore undertaken the investigation of other pharmacological interventions that may utilize similar protective signaling pathways as IPC, such as pharmacological manipulation of pathways involved in IPC, in an effort to produce a non-invasive therapy that might protect the heart from I/R injury after the onset of ischemia.

Opioid receptor agonists have been found to reduce I/R injury in a manner that is dependent on

PKCδ (Fryer, Wang et al., 2001). Landiolol also protects the rat heart from I/R injury in a

manner that is PKC dependent, and is associated with increased PKCε activity (Takahashi et al.,

2007). Similarly, ethanol at physiological levels prevents the injurious effects of I/R injury in the

heart via activation of PKCε (C. H. Chen et al., 1999). PKCα has been implicated in the

cardioprotection mediated by sevofluorane and sildenafil (Das et al., 2004; Okusa et al., 2009).

Phenylephrine requires active PKCδ to promote the survival of ischemic myocytes, which

modulates the activity of sarcolemmal ATP-sensitive potassium channels (KATP) (Turrell et al.,

2011). In addition to these interventions, FGF2 has been shown to require PKC activity to

protect the heart, as discussed in the following section.

7. FGF2 and PKCs

FGF2 can activate PKCs in a number of cell types (Albuquerque, Akiyama, & Schnaper,

1998; Debiais et al., 2001; Doble et al., 2000; Haimovitz-Friedman et al., 1994; Hrzenjak &

Shain, 1997; Hsu, Nicholson, & Hajjar, 1994; Kent et al., 1995; Kozawa, Suzuki, & Uematsu,

1997; Louis, Magal, Gerdes, & Seifert, 1993; Lynch, Fernandez, Pappalardo, & Peluso, 2000; L.

20

Y. Oh, Goodyer, Olivier, & Yong, 1997; Padua et al., 1998; Peluso, Pappalardo, & Fernandez,

2001; Presta, Tiberio, Rusnati, Dell'Era, & Ragnotti, 1991; Skaletz-Rorowski et al., 1999;

Yamamura, Nelson, & Kent, 1996). FGF2 requires endothelial cell-localized PKCα to elicit tube

formation during angiogenesis (Maffucci et al., 2009), and to mediate increased NO production

to induce vasodilation and increased blood flow (Im et al., 2007; Partovian et al., 2005).

Additionally, FGF2 has been shown to produce biological effects through PKC in differentiating

nerve cells (Kiraly et al., 2009), renal fibroblasts (Vasko et al., 2009), trophoblasts (Yang,

Johnson, & Ealy, 2011), and proliferating chondrocytes (Im et al., 2007). While many of these

studies implicate FGF2-FGFR signaling in PKC activation, it should be noted that non-canonical

FGF2 signaling may also activate PKCs. For example, during angiogenesis, the FGF2 binding

partner HSPG, syndecan 4, has been shown to regulate the localization and activity of PKCα in

endothelial cells (Keum et al., 2004).

The question of whether endogenously expressed HMW or LMW activate the same PKC

isoforms in various cell types remains unknown. Work by Gaubert and group (Gaubert et al.,

2001) suggest that this may not be the case, using pancreatic cells transfected with either LMW

or HMW FGF2 to demonstrate that LMW FGF2 expression results in increased PKCε, while

HMW FGF2 results in increased PKCδ expression and activation; interestingly, this increase in

PKCδ activation was not FGFR-mediated. Overexpression of LMW FGF2 has been associated

with an increase in membrane-associated PKCα and ε in non-ischemic mouse hearts (Sheikh et

al., 2001). Additionally, it has been shown that exogenously administered HMW FGF2 activates

similar kinases to LMW FGF2 in ischemic hearts in an FGFR-dependent manner, with a slight

increase in PKCζ activation in HMW FGF2 compared to LMW FGF2 (Jiang et al., 2009),

although it should be noted that work from our laboratory demonstrated that HMW FGF2 does

not leave the cardiomyocyte at any point during I/R and is therefore unlikely to act as an FGFR

21

ligand under physiological conditions (Liao et al., 2010).

In the mouse heart during I/R, our laboratory has previously demonstrated that

overexpression of all isoforms of FGF2 induce differential activation of several PKC isoforms

compared to wildtype hearts, modifying the translocation and phosphorylation of PKCα, δ, ε and

ζ (House et al., 2007). PKCα, δ and ζ decrease their translocation at early ischemia and/or early

(α) or late (δ) reperfusion in FGF2 overexpressing hearts, while PKCε shows increased

translocation at early reperfusion. In addition, blocking multiple isoforms of PKC with the

nonselective PKC inhibitor bisindolylmaleimide (GFX) can abrogate both the improvement in

cardiac function and the reduction in infarct size observed when mouse hearts overexpress all

human isoforms of FGF2 (House et al., 2007). Similar effects were seen when cardioprotection

was induced by exogenous administration of FGF2 on rat hearts; the protective effects of FGF2

were abolished by the administration of chelerythrine, an inhibitor of several PKC isoforms

(Jiang et al., 2002). In this model, PKCα, ε, and ζ showed movement from the cytosol to the

membrane or particulate fraction after 30 minutes of ischemia and 30 minutes of reperfusion

(Jiang et al., 2002). However, it remains to be seen whether endogenously expressed LMW and

HMW FGF2 activate the same PKC isoforms in the ischemic heart, or whether the different

biological activities of these two families of FGF2 are mediated by differential PKC isoform

activation.

8. Downstream targets of FGF2 and PKCs: Contractile proteins

As the protective effects against myocardial dysfunction that are mediated by LMW

FGF2 are acutely seen after the onset of reperfusion, the targets of FGF2-PKC are hypothesized

to be phosphorylatable proteins that produce an immediate effect on contractile function. One

22

such class of proteins is the myofibrillar proteins that directly impact crossbridge cycling and

force production between the thick myosin filament and the thin actin filament

Myosin, the key effector of cross-bridge cycling, consists of dimers of myosin heavy

chain (MHC) associated with essential or regulatory light chains (MLC) in a one-to-one ratio

(Rayment et al., 1993). In the adult ventricle, the dominant isoforms of MHC are alpha in mice and beta in humans (Bouvagnet, Leger, Pons, Dechesne, & Leger, 1984). The primary effector of

contraction in the heart is actomyosin ATPase, the force-generating, magnesium- and calcium-

dependent, ATP consuming element in the head domain of the myosin filament. In order to

produce contraction, the head domain first binds ATP, which in turn allows it to bind only

weakly to actin (Brenner, 1988). ATP is hydrolyzed, resulting in a strong bond to (Dantzig,

Goldman, Millar, Lacktis, & Homsher, 1992). Subsequent to actin binding, the ATP hydrolysis

products inorganic phosphate and ADP are released, allowing the myosin head to bend to a 45

degree angle and generating a power stroke that pulls the bound actin toward the M-band of the

myofibril (Huxley & Simmons, 1971; Lymn & Taylor, 1971). Finally, ATP binds to the myosin

head, returning the angle to slightly less than 90 degrees and dissociating myosin and actin

(Lombardi et al., 1995), allowing the head to release the actin filament and restart the cycle.

Several proteins regulate this cycle; among the best-characterized is tropomodulin, which

is an actin-associated protein that blocks the actomyosin-binding site (Zot & Potter, 1987).

Tropomodulin itself is regulated by a complex of , which, on binding calcium,

allosterically alter tropomodulin to reveal the myosin binding site and allow crossbridge cycling

(B. S. Pan & Solaro, 1987). The troponins form a trinary complex, consisting of troponins T, C,

and I (Filatov, Katrukha, Bulargina, & Gusev, 1999). Troponin T contains the tropomyosin

binding site which directly interacts with tropomyosin (Ueno, 1978). binds to

calcium and troponin T and transduces alterations in calcium concentration to a conformational

23

change in troponin T (Iio, 1985). Troponins C and T, as well as tropomyosin, are bound to and

regulated by troponin I, which is a phosphorylatable inactivator of strong actin-myosin binding

(Galinska et al., 2010).

Troponins are known targets for post-translational modification by kinases, including

PKCs, with phosphorylatable residues having been identified on troponin I and T. Troponin T is

known to be phosphorylated at several different residues, including threonine 190, 199, and 280,

and serine 201, all of which have the effect of reducing the maximum rate of actomyosin ATPase activity (Noland, Raynor, & Kuo, 1989; Noland & Kuo, 1993; Sumandea, Pyle, Kobayashi, de

Tombe, & Solaro, 2003). Troponin I may also be phosphorylated at several residues, and the effects of these various phosphorylations is site-dependent. Serine 22/23 is a target of PKA or

PKCδ, and has been shown to alter calcium sensitivity and increase , resulting in a

higher rate of relaxation and subsequent contraction of the cardiomyocyte (Cole & Perry, 1975;

Engel, Hinken, & Solaro, 2009). However, serine 41/43, a target of PKCα, has been shown to

reduce calcium-induced contractility by reducing the affinity of myosin for the thin filament and reducing the ability of calcium bound to TnC to relieve TnI inhibition (Noland et al., 1995;

Noland et al., 1996). PKCα also targets threonine 144, which results in reduced calcium

sensitivity (Engel et al., 2009; Noland et al., 1996).

The role of troponins in I/R injury is the subject of much study. Mutation of the PKC

sites of troponin I, serine 41/43 and threonine 144, to pseudophosphorylated residues results in

enhanced inhibition of ATPase activity and reduced calcium binding under acidic conditions, as

are seen during ischemia (Engel et al., 2009). Phosphorylation of troponins by PKC has been

associated with a slowing of ATP depletion and protection of the heart from stunning during I/R

(Pyle, Chen, & Hofmann, 2003). Compelling evidence suggests that κ-opioid agonist mediated

cardioprotection relies on this mechanism to protect the heart from ischemia (Pyle, Smith, &

24

Hofmann, 2000; Pyle, Lester, & Hofmann, 2001). Given that acute administration of

recombinant FGF2 results in increased levels of ATP after I/R injury (Jiang et al., 2002), the

question of whether the cardioprotection induced by endogenously expressed FGF2 is dependent

on a similar mechanism, i.e. on troponin phosphorylation and modulation of actomyosin ATPase, is an important one to address.

9. Downstream targets of FGF2 and PKCs: Calcium-handling proteins

In addition to contractile proteins, potential targets of FGF2-activated PKC may be

proteins directly responsible for calcium cycling in the cardiomyocyte. It has been shown that

PKC isoforms may protect the heart by altering calcium cycling on stimulation with H2S during

I/R (T. T. Pan et al., 2008). Proteins that may affect calcium cycling include the proteins of the

sarcoplasmic reticulum (SR) that are necessary for the release of calcium from intracellular

stores into the cytosol during systole, and the reuptake of calcium during diastole. The release of

calcium from the SR is mediated by the ryanodine receptor, a calcium-activated

(Fill & Copello, 2002) during the course of the action potential, L-type calcium channels are

opened by the depolarization of the membrane, resulting in an increase of calcium, which then

triggers ryanodine receptor opening and a further rise in intracellular calcium concentration. The

ryanodine receptor (RYR) is a homotetrameric protein that may be regulated by phosphorylation

(Fill & Copello, 2002) and may be phosphorylated by PKA, PKG, and CaM kinase (Huke &

Bers, 2007; Rodriguez, Bhogal, & Colyer, 2003; Wehrens, Lehnart, Reiken, & Marks, 2004;

Xiao et al., 2006). Three isoforms of RYR have been identified, with RYR2 the form

predominantly expressed in the heart (Nakai et al., 1990). Inactivation of the ryanodine receptor

is associated with improved post-ischemic contractility, reduced infarct development, and

lowered occurrence of arrhythmia (Akita, et al., 1993; Meldrum et al., 1996; Thandroyen,

25

McCarthy, Burton, & Opie, 1988; Zucchi et al., 1995).

The reuptake of calcium into the SR is primarily mediated by the sarco(endo)plamsic reticulum calcium ATPase (SERCA), with SERCA2a the isoform predominantly expressed in the adult heart (Lytton & MacLennan, 1988). The main regulator of SERCA in the human adult ventricle is phospholamban, which reduces the affinity of SERCA for calcium when bound to it

(Kirchberber, Tada, & Katz, 1975; Verboomen, et al, 1992). Phospholamban may exist in the monomeric state, which is more likely to bind to SERCA, or pentameric state, which has decreased association with SERCA (MacLennan & Kranias, 2003). A determining factor for the affinity of phospholamban to SERCA is phosphorylation, which causes a disordering of phospholamban and prevents it from binding SERCA (Metcalfe, Traaseth, & Veglia, 2005).

Phospholamban may be phosphorylated on serine 16 by PKA, or on threonine 17, a calmodulin kinase II (CaMKII) target, and phosphorylation of both residues reduce the affinity of phospholamban for SERCA (Davis, Schwartz, Samaha, & Kranias, 1983; Tada et al., 1980; W.

Zhao et al., 2004). While phospholamban may not be directly phosphorylated by PKCs, there is evidence that PKCs may act upstream of PKA or CaMKII to impact phospholamban activity.

Work done by Braz and colleagues (Braz et al., 2004) has shown that PKCα phosphorylates inhibitor-1 (I-1) in the heart, which reduces its inhibition of protein phosphatase-1, promoting dephosphorylation of phospholamban at its PKA site. This leads to decreased SERCA activity, and depressed contractility of the heart (Braz et al., 2004). PKCα has also recently been shown to regulate phospholamban during reperfusion through activation of calpain, which dephosphorylates both serine-16 and threonine-17 (Shintani-Ishida & Yoshida, 2011). PKCε may also play a role in phospholamban regulation through phosphorylation and activation of

CaMKII (Oestreich et al., 2009).

The regulation of phospholamban has a significant impact on the calcium cycling of the

26

cardiomyocyte and the recovery of the heart from I/R injury. Both serine-16 and threonine-17

show increased phosphorylation during ischemia and reperfusion (Kimura-Kurosawa et al.,

2007; Said et al., 2003). Studies examining the hearts of mice expressing only phospholamban

mutated to be unphosphorylatable have demonstrated that phosphorylation at both threonine-17

and serine-16 are necessary for full recovery from reversible functional injury after I/R

(Mattiazzi, Mundina-Weilenmann, Vittone, & Said, 2004; Mattiazzi, Mundina-Weilenmann,

Vittone, Said, & Kranias, 2006; Said et al., 2003), and are required for a return to normal

calcium cycling (Valverde et al., 2006). Landiolol-treated isolated guinea pig hearts show a

marked decrease in serine-16 phosphorylation of phospholamban after reperfusion compared to

untreated hearts (Kimura-Kurosawa et al., 2007).

Another protein which can regulate calcium cycling in the heart is calsequestrin, a highly acidic calcium-binding protein located inside lumen of the SR (MacLennan & Wong,

1971).Calsequestrin is found associated to the membrane of the SR and may regulate RYR2 activity, stabilizing it in the open conformation through unknown mechanisms (Györke, Stevens,

& Terentyev, 2009). Calsequestrin was found to be upregulated in neonatal ischemic rabbit cardiomyocytes, compared to non-ischemic hearts (Seehase et al., 2006). Conversely, in adult rat heart cells, calsequestrin expression was found to be downregulated after ischemia, but restored to normal levels by ischemic preconditioning (Temsah, Kawabata, Chapman, & Dhalla, 2002), suggesting that calsequestrin may play a key role in resolving altered calcium homeostasis after ischemia. In addition to keeping calcium localized near SR calcium channels, calsequestrin has been shown to play a role in the endoplasmic reticulum (ER) stress response (Hunter, Mitchell-

Felton, Essig, & Kandarian, 2001). The role of the ER stress response is somewhat controversial in I/R injury, associated with both increased apoptosis and pathogenesis (Azfer, Niu, Rogers,

Adamski, & Kolattukudy, 2006; Szegezdi et al., 2006; Thuerauf et al., 2006), as well as

27

cardioprotective autophagy (Petrovski et al., 2011).

The question of whether FGF2 expression in the heart can modulate the levels and/or

activity of SERCA, phospholamban, or calsequestrin, has yet to be determined. Given that these

proteins may impact the recovery of the heart from I/R injury, it is necessary to determine the

effect of FGF2 on these proteins prior to and during I/R in order to fully delineate the

cardioprotective pathways of FGF2.

10. Other targets of FGF2 in the heart during I/R

In addition to kinase-regulated alterations in myofibrillar and calcium-handling proteins,

another potential mechanism for FGF2-mediated protection makes use of the nitric oxide (NO)

signaling pathway. Several biological actions of FGFs have been shown to signal via NO, and

NO synthase (NOS), in a number of tissue types, including the heart as well as vasculature, where FGF2-induced NO production mediates vasodilation and angiogenesis (Cuevas et al.,

1999; Hampton et al., 2000; Huang, Chen, Huang, Finklestein, & Moskowitz, 1997; Kajita et al.,

2001; Sellke et al., 1994; Tiefenbacher & Chilian, 1997). FGF2 has also been shown to regulate

the expression of NOS isoforms (Iwai-Kanai et al., 2002; Kostyk et al., 1995). This is

particularly relevent as NOS has been implicated as a key player in cardioprotection and

ischemia/reperfusion injury (Jones & Bolli, 2006).

Three isoforms of NOS have been identified; NOS1 or neuronal NOS (nNOS), NOS2 or

inducible NOS (iNOS) and NOS3 or endothelial NOS (eNOS). All three isoforms are found in

the myocardium (Forstermann et al., 1994). nNOS and eNOS are expressed constitutively in the

heart, while iNOS is expressed after the heart is challenged with a stressor, such as I/R injury or hypoxia (Jones & Bolli, 2006; Jung, Palmer, Zhou, & Johns, 2000; Kelly, Balligand, & Smith,

1996; Wildhirt, Dudek, Suzuki, & Bing, 1995). nNOS has been shown to regulate cardiac

28

function in healthy hearts by interactions with sarcoplasmic reticulum calcium-handling proteins,

including the ryanodine receptor (Sears et al., 2003; K. Y. Xu, Huso, Dawson, Bredt, & Becker,

1999; Y. H. Zhang et al., 2008), and plays a role in post-ischemic cardiac function and

development of infarct (Burkard et al., 2010; Dawson et al., 2005; Saraiva et al., 2005). iNOS

has also been implicated in ischemic injury, with overexpression protecting the heart from infarct

development, and ablation exacerbating reperfusion injury (West et al., 2008; Zingarelli et al.,

2002). iNOS has been demonstrated to be a key effector of late phase preconditioning (Bolli et

al., 1998), and overexpression of iNOS is associated with a protected phenotype (Q. Li et al.,

2003). eNOS is localized to the sarcolemma and caveolae in the cardiomyocyte, where it is

proposed to interact with L-type calcium channels and regulate calcium influx (Barouch et al.,

2002). The role of eNOS in the heart during I/R is somewhat controversial, due in part to

conflicting data generated by two different genetic mouse models of eNOS ablation (R. M. Bell

& Yellon, 2001; Sharp, Jones, Rimmer, & Lefer, 2002), although recent, careful analysis by Guo

and investigators (Guo et al., 2008) utilizing both mouse models suggests that eNOS may also

play a role in I/R injury and ischemic preconditioning.

A number of pharmacological cardioprotective stimuli have been shown to require NOS.

Among these are sildenafil (Das, Salloum, Xi, Rao, & Kukreja, 2009), norepinephrine (Imani,

Faghihi, Sadr, Niaraki, & Alizadeh, 2011), sulfaphenazole (Khan et al., 2009), δ2 opioid receptors (Maslov et al., 2009), MAP kinase homologues (Cuenda et al., 1995), G-protein coupled receptor kinases (GRK) (Brinks et al., 2010), propofol (Sun et al., 2009), and bradykinin

(Yeh, Chen, Wang, Lin, & Fang, 2010). Hampton and colleagues (Hampton et al., 2000) have demonstrated that the application of exogenous FGF2 protected the heart from post-ischemic stunning via an iNOS-dependent mechanism , although the role of NOS isoforms in cardioprotection mediated by endogenously produced FGF2 remains to be identified.

29

NOS isoform expression and activity may be regulated by phosphorylation by PKCs and

MAPKs, which are downstream of FGFs. A number of phosphorylatable residues have been identified for nNOS, with differing effects on activity. nNOS phosphorylation at S847 and S1412 have been shown to increase the activity of the enzyme (Adak et al., 2001; G. A. Rameau et al.,

2007; G. A. Rameau, Chiu, & Ziff, 2004). Similarly, phosphorylation of eNOS at S1179, a site analogous to S1412, by FGF2-activated PKCα has been shown to upregulate NO production in endothelial tissue (Partovian et al., 2005). Ping and group (P. Ping et al., 1999) have also shown that eNOS is phosphorylatable by PKCε and this is associated with increased activity in preconditioned hearts. In addition to PKC isoforms, MAPKs have also been implicated in the regulation of NOS activity in the heart (Das et al., 2009; Singh, Balligand, Fischer, Smith, &

Kelly, 1996; T. C. Zhao, Taher, Valerie, & Kukreja, 2001). The ERK pathway, in particular, has been shown to mediate FGF2-induced cardioprotection from iNOS-induced apoptosis in cardiomyocytes (Iwai-Kanai et al., 2002).

Rapid signaling mediated by kinase cascades or nitric oxide-activated pathways are not the only targets of FGF2 that have the potential to impact the recovery of the myocardium after

I/R as FGF2 has been shown to have long term genomic effects as well. FGF2 may regulate gene transcription in the heart (Schneider, McLellan, Black, & Parker, 1992). Genes that are regulated by FGF2 include those that affect growth and metabolism, survival, growth and differentiation, and migration (Diecke, Quiroga-Negreira, Redmer, & Besser, 2008; Ito, Sawada, Fujiwara, &

Tsuchiya, 2008; Polnaszek et al., 2003; Quarto et al., 2005; Turner, Clinton, Thompson, Watson,

& Akil, 2011). In addition to activating MAPKs and PKCs which alter gene transcription, FGFs may affect gene transcription in kinase-independent mechanisms (Diecke et al., 2008), and evidence suggests that FGF2 may enter the nucleus as well to activate transcription (Bouche et al., 1987; Zhan, Hu, Friesel, & Maciag, 1993).

30

The roles of HMW and LMW FGF2 in the regulation of gene transcription in the heart

remain to be elucidated. Work published by Quarto and colleagues (Quarto et al., 2005) using

3T3 fibroblasts has demonstrated differential transcriptional regulation of a number of genes between cells expressing LMW and HMW FGF2, suggesting that each family of isoform has distinct genomic activity. In particular, given the opposing roles of each isoform in the heart, as well as the differential localization of HMW and LMW FGF2 in the cardiomyocyte, it is expected that mice absent of a particular FGF2 protein isoform will demonstrate distinctly altered mRNA and protein production.

11. Statement of Purpose

The overall objective of this dissertation research is to identify the mechanisms by which

FGF2 protects the heart from ischemia/reperfusion injury. As LMW FGF2 has been implicated as the acute effector of improved post-ischemic cardiac function, this dissertation research is undertaken to determine the kinases, and, particularly, the isoforms of protein kinase C (PKC) and their phosphorylation targets at the myofibril and sarcoplasmic reticulum, that are responsible for the protective effects of LMW FGF2. In addition, this project seeks to evaluate novel effectors that contribute to the cardioprotective effects of both isoforms of FGF2, through priming the heart via differential regulation of potentially protective or injurious genes. Finally, this project is intended to begin to explore the role of HMW FGF2 on the ischemic heart, both in terms of its impact on LMW FGF2 signaling, as well as the impact that overexpression of HMW and LMW FGF2 together may have on the as-yet unclear role of NO production on infarct size.

Specific Aim 1: Determine the role of isoforms of PKC in LMW FGF2-mediated protection from I/R-induced cardiac dysfunction, and identify potential downstream targets.

31

As previously described, PKC has been identified as a necessary second messenger for

FGF2-induced cardioprotection, with both the reduction of infarct size and the improvement in

contractile function seen after I/R, when FGF2 is exogenously administered or overexpressed,

ablated in the presence of inhibitors of PKC (House et al., 2007; Jiang et al., 2002). However, the

PKC isoforms that mediate this protection have not been identified. Previous work has demonstrated that different isoforms of PKC can have varying effects on subcellular signaling in the heart, and therefore, identification of these PKC isoforms is critical for elucidating the mechanisms by which LMW FGF2 protects the heart from post-ischemic dysfunction. In addition, identifying the potential targets of these PKC isoforms allows for a more complete understanding of these mechanisms. It is hypothesized that LMW FGF2 will differentially

activate one or more of these isoforms of PKC, leading to the phosphorylation of unique

downstream targets that will directly impact the post-ischemic function of the heart. To

determine which isoforms of PKC are required for LMW FGF2 to produce in an improvement in

post-ischemic function, immunoblotting will be used to evaluate the activation (indicated by

phosphorylation) of the α, δ, and ε isoforms of PKC in hearts that only express LMW FGF2

(HMWKO), at key time points during I/R. Isoforms that are shown to be differentially activated

in HMWKO hearts will then be tested by either genetic ablation or pharmacological inhibition of

the isoform of interest, to determine if removing the activity of the PKC isoform suppresses the

cardioprotective effects of LMW FGF2. Finally, candidate downstream targets of these isoforms

will be evaluated by SDS-PAGE, immunoblot, or by fractionation and activity assays to

determine changes in phosphorylation state and/or activity, to provide insight into the

mechanism of action of LMW FGF2/PKC isoform mediated protection from post-ischemic

cardiac dysfunction.

32

Specific Aim 2: Determine the influence of HMW FGF2 on LMW FGF2 signal

transduction.

HMW FGF2 and LMW FGF2 have been shown to have distinct biological roles, localize

to distinct intracellular areas (HMW FGF2 to the nucleus, LMW FGF2 to the cytosol), and are

capable of regulating one another’s biological function (Bikfalvi et al., 1995; Estival et al., 1996;

Quarto et al., 2005). Previous work from our laboratory has demonstrated that expression of only

HMW FGF2 results in lowered post-ischemic function (Liao et al., 2007), while expression of

only LMW FGF2 significantly improves cardiac function following I/R injury (Liao et al., 2010),

but the mechanism by which these differential effects are achieved is currently unknown. An

important step in understanding the effects of FGF2 on the ischemic heart is to determine the relationship that HMW and LMW FGF2 have on each other. It has been demonstrated by our laboratory and others that LMW FGF2 protects the heart by activation of the FGFR1 receptor in the mouse and rat heart (Jiang et al., 2002; Liao et al., 2010). It is hypothesized that HMW FGF2

interferes with FGFR activation and kinase signaling involved in the recovery of post-ischemic

cardiac contractility to attenuate LMW FGF2’s protective effects on the heart. To test this

hypothesis, hearts overexpressing the human 24 kDa isoform of HMW FGF2 will be subjected to

I/R, and the activation of the FGFR receptor and PKC and MAPKs activity will be evaluated.

Specific Aim 3: Identify novel targets of HMW and LMW FGF2 responses to cardiac

ischemia-reperfusion injury and cardioprotection.

In addition to PKC and MAPK isoform activation, which our laboratory has shown are

necessary for FGF2-mediated protection from both infarct development and functional injury

(House et al., 2005; House et al., 2007), this dissertation project also seeks to identify other

protective pathways involved in the actions of LMW and HMW in the ischemic heart. One

33

possible effector of FGF2-mediated protection is NOS, which produces NO, a molecule with a

complex but potentially protective effect on the ischemic heart (Jones & Bolli, 2006). FGF2 has

been shown to signal though NOS in both cardiac and other tissue types, but the effect of

endogenously overexpressed FGF2 on the activity of NOS in the heart during I/R is not well

understood. In particular, the crosstalk between the known protective pathways of PKC/MAPKs and NOS has not been explored in FGF2-overexpressing hearts. To examine this crosstalk, NO production in hearts overexpressing human FGF2 will be subjected to I/R in the absence and presence of inhibitors of PKC, ERK, p38, and JNK. It is hypothesized that the protective kinase

pathways utilized by FGF2 will increase NO production, indicating regulation of the activity of

NOS isoforms.

In addition to testing candidate signaling pathways such as PKC, MAPK, and NOS, novel

targets of LMW and HMW FGF2 will also be explored. FGF2 is known to alter transcriptional

regulation of genes in a number of cell types, via FGFR activation (Bikfalvi et al., 1995;

Moscatelli & Quarto, 1989), and HMW FGF2 has been shown to translocate to the nucleus of

cells, indicating that it may have a unique role to play in transcriptional regulation (Chlebova et

al., 2009; Renko et al., 1990). In fibroblasts, HMW and LMW FGF2 have been demonstrated to

uniquely and differentially modulate the expression of a number of genes (Quarto et al., 2005).

However, the role of HMW and LMW FGF2 in the transcriptional regulation of genes in the

heart has not been evaluated. Since the effects of I/R injury on the heart may be the result of

having been ‘primed’ by the up- or down-regulation of protective or injurious genes by both

HMW and LMW FGF2, it is hypothesized that hearts only expressing LMW FGF2 will show an

upregulation of genes that protect the heart from contractile injury, and a downregulation of

genes that are injurious, while HMW FGF2 is expected to show the inverse with respect to genes

that affect contractile injury. However, since a reduction in infarct is only seen when the LMW

34

and HMW isoforms of FGF2 are overexpressed (House et al., 2003), it is also hypothesized that

HMW FGF2 expression will result in regulation of genes that favor cell survival during I/R. To test this hypothesis, RNA transcripts via microarray analysis from mice lacking expression of all

FGF2 isoforms will be compared to mice only expressing a particular FGF2 protein isoform.

35

MATERIALS AND METHODS

Animals and Exclusion Criteria

All mice were housed in a pathogen-free facility and handled in accordance with standard

use protocols, animal welfare regulations, and the NIH Guide for the Care and Use of

Laboratory Animals. All protocols were approved by the University of Cincinnati Institutional

Animal Care and Use Committee. Animals were excluded from the ischemia-reperfusion study if

it was determined that an aortic or pulmonary leak had occurred during the procedure. An aortic

leak was defined by an aortic pressure of<60 mmHg during retrograde perfusion (Langendorff

perfusion). A pulmonary vein leak was defined as a low aortic flow (<2.0 mL/min) during

normal anterograde perfusion, in the presence of low atrial pressure (<4 mmHg) and high

perfusate pO2 (>380 mmHg). In addition, studies were excluded if a leak was clearly visible, as in the case where a hole was apparent in the ventricle or atrium (i.e., perfusate was “squirting out” of ruptured atrium due to high atrial pressure during reperfusion). Additionally, hearts that

were unable to reach a stable baseline after 1 hour on the isolated heart apparatus were excluded

as well. A total of 21 mice were excluded from all studies.

36

Mouse Generation and Breeding

Genotype Primers Fgf2 KO Wildtype:

5’- CGA GAA GAG CGA CCC ACA C

5’- CCA GTT CGG GGA CCC TAT T

Knockout:

5’- AGG AGG CAA GTG GAA AAC GAA

5’- CCC AGA AAG CGA AGG AAC AAA

HMWKO 5’-CCC AAG AGC TGC CAC AG

5’-CGC CGT TCT TGC AGT AGA G

LMWKO 5’- CCC GCA CCC TAT CCT TAC ACA

5’- GCC GCT TGG GGT CCT TG

PKCαKO 5' -CG CAT CGC CTT CTA TCG C

5'-AGC TAG GTC CTG TTG GTA AC

5'-CCA AGT GTG AAG TGT GTG AG

24 kDa HMW FGF2 Tg 5’- CTT CAA AAG CGC ACG TCT GC C

5’- GCC TGC CAC ACC TCA AGC TT

α-MHC FGF2 Tg 5’-TGT GTT ACG GAT GAG TGT TTC TTT T

5’-GGG AGG TGT GGG AGG TTT TTT

Table 1: List of genotypes and primers used to identify genetically-modified mice by PCR.

37

Generation of Fgf2 KO mice

Fgf2 KO mice were generated on a mixed 50% 129 and 50% Black Swiss background by

Ming Zhou in the laboratory of Dr. Thomas Doetschman using a tag and exchange construct, as

previously described (M. Zhou et al., 1998). Briefly, the promoter region and the entire exon 1 of

the Fgf2 gene was replaced with an Hprt (hypoxanthine-guanine phosphoribosyltransferase)

minigene in E14TG2a ES cell lines from strain 129 mice, using a herpes simplex virus - thymidine kinase (HSV-tk) construct outside of the 3’ homologous region to select for homologous recombination. Embryonic stem (ES) cells that had incorportated the construct were

selected for with hypoxanthine-aminopterin-thymidine (HAT) medium, and random insertion of

the construct was selected against using gancyclovir. Double resistant cells were confirmed to

have correctly incorporated the construct with PCR and Southern blotting. These cells were then

injected into a C57/BL6 blastocyst, generating chimeric mice, which were then bred to Black

Swiss mice to determine germ line transmission of the knockout. Pups resulting form this and

subsequent breeding were screened by PCR analysis to reveal +/+, +/-, and -/- genotypes. Loss of

FGF2 expression was also confirmed with Southern, Northern, and Western blotting.

For PCR analysis, DNA was isolated from a tail clip by digestion at 60° for 2 hours with

a buffer of 100 mM NaCl, 10 mM Tris-Cl pH 8, 25 mM EDTA, 0.5% SDS, and 5 µg of

proteinase K, and the homogenate was centrifuged to remove cell debris. DNA was precipitated

with 100% isopropanol, and washed once with 70% ethanol, and then dissolved in TE buffer.

PCR was carried out using Roche Taq DNA polymerase and the following primers: Wildtype

allele, 5’- CGA GAA GAG CGA CCC ACA C, 5’- CCA GTT CGG GGA CCC TAT T;

Knockout allele, 5’- AGG AGG CAA GTG GAA AAC GAA, 5’- CCC AGA AAG CGA AGG

AAC AAA. Polymerase was added prior to temperature cycling, and DNA was amplified during

35 cycles of 90° for 30 seconds (denaturation), 58° for 50 seconds (annealing), and 72° for 90

38

seconds, with a final 72° incubation for 10 minutes (elongation). PCR products were run on a

1% agarose gel at 130 mV for 30-40 minutes. A product band at 185bp indicates the presence of

a wildtype allele, while a band at 1299 bp indicates the FGF2 knockout allele.

Generation of high molecular weight FGF2 knockout (HMWKO) mice

FGF2 HMWKO mice were generated on a mixed 50% 129/Black Swiss background by

Dr. Mohammed Azhar in the laboratory of Dr. Thomas Doetschman using a tag and exchange

construct, as previously described (Azhar et al., 2009). Briefly, ES cells containing the tagged

construct used to generate Fgf2 KO mice described above were subjected to a targeting

‘exchange’ vector containing a 14 bp insertion (containing a XbaI endonuclease site) after the

final CUG translational start site, in order to induce a frame shift in all translational products

upstream of the LMW FGF2 AUG translational start site. Loss of the Hprt gene in ES cells from

strain 129 mice was selected for by 6-thioguanine (6-Tg). Resistant cells were confirmed to have

the mutated Fgf2 gene with PCR, Southern blotting, and XbaI digest. These cells were then injected into a C57/BL6 blastocyst, generating chimeric mice, which were then bred to Black

Swiss mice to determine germ line transmission of the knockout. Pups resulting from this and subsequent breeding were screened by PCR analysis to reaveal +/+, +/-, and -/- genotypes. Loss of HMW FGF2 expression was confirmed with Southern, Northern, and Western blotting.

Western blot analysis confirmed loss of HMW FGF2 protein, with a slight increase in LMW

FGF2 protein expression (Figure 2).

For PCR analysis, DNA was isolated from a tail clip by digestion at 60° for 2 hours with

Qiagen’s PureGene kit Cell Lysis Buffer and 5 µg of proteinase K, and the homogenate was

centrifuged with Qiagen’s PureGene kit Protein Precipitation Buffer to remove cell debris. DNA

39

was precipitated with 100% isopropanol, and washed once with 70% ethanol, and dissolved in

Qiagen PureGene hydration buffer. PCR was carried out using Epicentre Tfl polymerase and the following primers, which amplified both wildtype and knockout alleles: 5’-CCC AAG AGC

TGC CAC AG, and 5’-CGC CGT TCT TGC AGT AGA G. Polymerase was added prior to temperature cycling. After one incubation at 97° for 5 minutes (denaturation), DNA was amplified by 35 cycles at 97° for 60 seconds (denaturation), 58° for 60 seconds (annealing), and

72° for 2 minutes, with a final 72° incubation for 10 minutes (elongation). PCR ran on a 4% agarose gel at 70 mV for three hours. A band at 152 bp indicates the presence of a wildtype allele, while a band at 166 bp indicates the HMW FGF2 knockout allele.

40

LMW FGF2 expression 9000 Wt * 8000 FGF2 HMWKO

7000

6000

5000

4000

Arbitrary Units 3000

2000

1000

0

HMW 22 kDa 21 kDa LMW 18 kDa

Wt Fgf2 KO

Figure 2: FGF2 isoform expression in genetically modified mouse models. Expression of the isoforms of FGF2 protein in wildtype, Fgf2 KO, and HMWKO hearts. All isoforms of

FGF2 are absent in the Fgf2 KO, while only the LMW isoform of FGF2 is expressed in the

HMWKO. *p<0.05 (n=5 per group)

41

Generation of low molecular weight FGF2 knockout (LMWKO) mice

FGF2 LMWKO mice were generated on a mixed 50% 129/Black Swiss background by

Ming Zhou in the laboratory of Dr. Thomas Doetschman using a tag and exchange construct, as

previously described (M. Zhou et al., 1998). Briefly, ES cells from strain 129 mice containing

the tagged construct used to generate Fgf2 KO mice described above were subjected to a

targeting ‘exchange’ vector that replaced the ATG start codon of LMW FGF2 (5’-GG GGC

CGC GGA AGG GCC ATG) with a diagnostic Pst I (5’-GG GGC CGC GGA AGG GCT GCA) site. Loss of the Hprt gene was selected for by 6-thioguanine (6-Tg). Resistant cells were confirmed to have the mutated Fgf2 gene missing the LMW FGF2 start codon with PCR and subsequent Pst I digest, as well as Southern blotting, and. These cells were then injected into a

C57/BL6 blastocyst, generating chimeric mice, which were then bred to Black Swiss mice to determine germ line transmission of the knockout. Pups resulting from this and subsequent breeding were screened by PCR and digest analysis to reaveal +/+, +/-, and -/- genotypes. Loss of LMW FGF2 expression was confirmed with Southern, Northern, and Western blotting.

For PCR analysis, DNA was isolated from a tail clip by digestion at 60° for 2 hours with

Qiagen’s PureGene kit Cell Lysis Buffer and 5 µg of proteinase K, and the homogenate was centrifuged with Qiagen’s PureGene kit Protein Precipitation Buffer to remove cell debris. DNA was precipitated with 100% isopropanol, and washed once with 70% ethanol, and dissolved in

Qiagen PureGene hydration buffer. PCR was carried out using Epicentre Tfl polymerase and the following primers, which amplified both wildtype and knockout alleles: 5’- CCC GCA CCC

TAT CCT TAC ACA, and 5’- GCC GCT TGG GGT CCT TG. Polymerase was added prior to temperature cycling. After one incubation at 97° for 5 minutes (denaturation), DNA was amplified by 35 cycles of 97° for 60 seconds (denaturation), 58° for 60 seconds (annealing), and

72° for 2 minutes, with a final 72° incubation for 10 minutes (elongation). PCR products were

42

incubated at 37° with Pst I, Buffer 3 and BSA (New England Biolabs), and the digests were run on a 1% agarose gel at 130 mV for 30-40 minutes. A band at 566 bp indicates the presence of a wildtype allele, while a band at 476 bp and 90 bp indicates the LMW FGF2 knockout allele.

Generation of PKC alpha KO (PKCαKO) mice

PKCαKO mice were generated by the laboratory of Dr. Jeffrey Molkentin as previously described (Braz et al., 2004). Briefly, the exon encoding the ATP binding cassette of the protein kinase C alpha isoform (Prkca) gene was replaced via homologous recombination with a neomycin resistance gene, using a HSV-tk construct outside of the 3’ homologous region to select for homologous recombination. Neomycin and gancylcovir were used as positive and negative selection. Cells surviving both selection protocols were injected into a blastocyst, generating chimeric mice that were subsequently bred to wildtype mice to confirm germline transmission. Loss of PKC alpha was confirmed with Southern and Western blotting. Pups resulting from this and subsequent breeding were screened by PCR analysis to reaveal +/+, +/- and -/- genotypes. To determine the role of PKCα in LMW FGF2-mediated protection of the heart from I/R induced cardiac dysfuction, these mice were crossed to HMWKO mice and bred for homozygous loss of both HMW FGF2 and PKCα. Pups resulting from this and subsequent breeding were screened by PCR analysis to reveal +/+ HMWKO, +/+ PKCαKO; +/- HMWKO,

+/+ PKCαKO; +/+ HMWKO, +/- PKCαKO; +/- HMWKO, +/- PKCαKO; -/- HMWKO, +/+

PKCαKO; -/- HMWKO, +/- PKCαKO; +/+ HMWKO, -/-PKCαKO; +/-HMWKO, -/-PKCαKO;

-/- HMWKO, -/- PKCαKO. The loss of expression of both genes was confirmed with Western blotting, and other PKCs were examined to ensure that, as previously described, loss of PKCα did not affect the expression of PKCδ or ε (Figure 3).

43

For PCR analysis, DNA was isolated from a tail clip by digestion at 60° for 2 hours with

Qiagen’s PureGene kit Cell Lysis Buffer and 5 µg of proteinase K, and the homogenate was centrifuged with Qiagen’s PureGene kit Protein Precipitation Buffer to remove cell debris. DNA was precipitated with 100% isopropanol, and washed once with 70% ethanol, and dissolved in

Qiagen PureGene hydration buffer. PCR for the PKCαKO allele was carried out using Epicentre

Tfl polymerase and the following primers: knockout allele: 5' -CG CAT CGC CTT CTA TCG C, knockout and wildtype allele 5'-AGC TAG GTC CTG TTG GTA AC, and wildtype allele 5'-

CCA AGT GTG AAG TGT GTG AG. Polymerase was added prior to temperature cycling. DNA was amplified with 30 cycles of 94° for 30 seconds (denaturation), 57° for 60 seconds

(annealing), and 72° for 60 seconds (elongation). PCR product were run on a 1% agarose gel at

130 mV, and a band at 250 bp indicates the presence of a wildtype allele, while a band at 400 bp indicates the knockout allele. For the HMWKO allele, PCR was carried out using Tfl polymerase

(Epicentre) and the following primers, which amplified both wildtype and knockout alleles: 5’-

CCC AAG AGC TGC CAC AG, and 5’-CGC CGT TCT TGC AGT AGA G. Polymerase was added prior to temperature cycling. After one incubation at 97° for 5 minutes (denaturation).

DNA was amplified by 35 cycles of 97° for 60 seconds (denaturation), 58° for 60 seconds

(annealing), and 72° for 2 minutes, with a final 72° incubation for 10 minutes (elongation).

PCR product were run on a 4% agarose gel at 70 mV, and a band at 152 bp indicates the presence of a wildtype allele, while a band at 166 bp indicates the HMWKO allele.

44

HMW FGF2 - + - + - LMW FGF2 - + + + + PKC α + + + -- FGF2

PKC α

PKC ε PKC δ

Figure 3: The expression of FGF2, and PKCα, ε, and δ protein in FGF2 KO, wildtype,

HMWKO, PKCαKO, and PKCαKOxHMWKO hearts as determined by Western blot.

45

Generation of human 24 kDa high molecular weight FGF2 transgenic (HMW Tg) mice

HMW Tg mice were generated on an FVB/N background by Dr. Douglas Coffin in the

laboratory of Dr. Thomas Doetschman (M. G. Davis et al., 1997). Briefly, the AUG and two

CUG codons initiating translation of the 18, 21.5, and 22 kDa isoforms of FGF2 were point

mutated in the human FGF2 gene, and expression was driven with a phosphoglycerate kinase

(PGK) promoter. This construct, which can only express the 24 kDa isoform of FGF2, was

injected into the pronucleus of FVB/N mouse oocyte to generate founder mice. Pups

incorporating the transgene were identified with PCR, and transgene expression was confirmed

with Western blotting. Two lines were generated in this way, 24IP20 and 24IP28, to confirm that

effects seen in transgenic animals were not due to insertional mutagenesis.

For PCR analysis, DNA was isolated from a tail clip by digestion at 60° for 2 hours with

a buffer of 100 mM NaCl, 10 mM Tris-Cl pH 8, 25 mM EDTA, 0.5% SDS, and 5 µg of

proteinase K, and the homogenate was centrifuged to remove cell debris. DNA was precipitated

with 100% isopropanol, and washed once with 70% ethanol, and then dissolved in TE buffer.

PCR was carried out using Roche Taq DNA polymerase and the following primers, which

amplified only the transgenic allele: 5’- CTT CAA AAG CGC ACG TCT GC C, 5’- GCC TGC

CAC ACC TCA AGC TT. Polymerase was added prior to temperature cycling. After an initial incubation at 94° for 5 minutes (denaturation), DNA was amplified with 36 cycles at 94° for 30 seconds (denaturation), 59° for 30 seconds (annealing), and 72° for 60 seconds, with a final 72° incubation for 10 minutes (elongation). A product band at 300 bp indicates the presence the 24 kDa HMW transgene.

Generation of cardiac-specific overexpression of FGF2 transgenic (FGF2 Tg) mice

FGF2 Tg mice were generated on an FVB/N background by Ming Zhou in the laboratory

46

of Dr. Thomas Doetschman, as previously described (House et al., 2003; M. Zhou, 1997).

Briefly, the human FGF2 gene was ligated onto an α-myosin heavy chain (α-MHC) promoter,

which drives expression in the adult mouse heart (Subramaniam et al., 1991), and this construct

was injected into the pronuclei of FVB/N mouse oocytes to generate founder mice. Pups

incorporating the transgene were identified with PCR and Southern blotting, and transgene

expression was confirmed with Western blotting.

For PCR analysis, DNA was isolated from a tail clip by digestion at 60° for 2 hours with

a buffer of 0.1 M TRIS-HCl (pH 8.0), 0.005 M EDTA (pH 8.0), 0.2 M NaCl 0.2% SDS and 5 µg

of proteinase K, and the homogenate was centrifuged to remove cell debris. DNA was

precipitated with 100% isopropanol, dried, and dissolved in water. PCR was carried out using

Epicentre Tfl DNA polymerase and the following primers, which amplify only the transgenic

allele: 5’-TGT GTT ACG GAT GAG TGT TTC TTT T, and 5’-GGG AGG TGT GGG AGG

TTT TTT. After an initial incubation of 95° for 3 minutes, after which 1 µL of polymerase was

added, DNA was amplified with 30 cycles of 95° for 60 seconds (denaturation), 58° for 2

minutes (annealing), and 72° for 3 minutes (elongation). A product band at 453 base pairs indicates the presence the human FGF2 transgene.

47

Figure 4. The isolated working heart. A) Schematic depicting the flow of Krebs-

Henseleit buffer during anterograde perfusion B) Photograph of isolated working mouse heart.

48

Isolated Working Mouse Heart Ischemia/Reperfusion Studies

Age- (10 weeks-14 weeks) and sex-matched mice were evaluated using an isolated

working mouse heart (Figure 4). The mice were anesthetized with an intraperitoneal injection of

sodium pentobarbital (80 mg/kg). After mice were unconscious and unresponsive to painful

stimulus, the thoracic cavity was opened and the heart was carefully and quickly excised, after

which it was placed in a bath of warm, oxygenated, heparinized Krebs-Henseleit buffer (118 mM

NaCl, 25 mM NaHCO3, 0.5 mM Na-EDTA, 5 mM KCl, 1.2 mM KH2PO4, 1.2 mM MgSO4, 2.5

mM CaCl2 and 11 mM glucose). The aorta was immediately cannulated with a 20 gauge stainless

steel cannula, secured with 5-0 silk suture, preserving the aortic valve and coronary artery ostia.

The heart was subjected to retrograde perfusion (Langendorff perfusion) with warm (37.7°C),

oxygenated Krebs-Henseleit buffer at a hydrostatic pressure of approximately 60 mmHg. To

measure the left ventricular pressure, an open PE-50 catheter with one flared end was inserted

into the left atrium through the pulmonary vein, advanced past the mitral valve into the left

ventricle, and while supporting the heart, carefully pulled through the apex of the left ventricle.

Next, the pulmonary vein was cannulated while carefully preserving the atrium and atrial septum, and secured with 5-0 suture, forming a tight seal. The heart was switched from retrograde to anterograde perfusion with a cardiac output of 5 mL/min and a perfusion pressure of 50 mm Hg, resulting in a constant workload of 250 mm Hg*mL/min. During this time, the coronary flow was approximately 2 mL/min. A small hole was created in the pulmonary artery

proximal to the outflow tract, allowing for sampling of the coronary sinus effluent with a

capillary tube to determine venous perfusate gas. The pO2 and pCO2 of arterial and venous

perfusate samples were measured using an automated blood gas analyzer (model 248, Ciba

Corning Diagnostics Corp). Oxygen consumption (MVO2) by the perfused hearts was

determined as the product of the coronary flow and the difference between arterial and venous

49

oxygen content (normalized by the mass of the heart). The heart was paced at the atrium at a rate

10-20 beats/min higher than its intrinsic rate (below the force-frequency relationship) to maintain

a constant during the course of ischemia and reperfusion. The heart was permitted to

stabilize at this workload for at least 30 minutes, during which time parameters of aortic

pressure, left atrial pressure, left ventricular pressure, positive and negative left ventricular dP/dt

were recorded and measured via COBE pressure transducers and a custom-made data acquisition

system along with a Grass polygraph (Figure 5). In addition, coronary flow, aortic flow, and

perfusate pH, pO2, and pCO2 levels were assessed. Coronary perfusate was collected for every two minutes for the last 10 minutes of this baseline equilibration, immediately placed on ice, and subsequently frozen for further analysis (Figure 5).

50

A. Treatment (drug or Treatment (drug or vehicle) vehicle) Heart snap frozen for proteomic Reperfusion Baseline Ischemia analysis

B. Treatment (drug or Treatment (drug or vehicle) vehicle)

Baseline Ischemia Reperfusion

Treatment (drug or vehicle) Heart snap frozen for proteomic

Baseline Ischemia analysis

Treatment (drug or vehicle)

Baseline

Figure 5: Schematic representing ischemia reperfusion experimental protocols using the isolated working mouse heart. A) For functional studies and analysis of the heart after ischemia and reperfusion injury, hearts equilibrated for 30 minutes (baseline), then underwent 60 minutes of ischemia followed by 120 minutes of reperfusion. B) For proteomic analysis of the heart during I/R injury, hearts equilibrated for 30 minutes, and/or were subjected to 15 minutes of ischemia and snap frozen, or 60 minutes of ischemia and 5 minutes of reperfusion, and snap frozen. Arrows indicate times when functional parameters were measured, coronary effluent was collected, and perfusate gasses were measured.

51

After 30 minutes of equilibration (baseline), global low-flow ischemia was induced by

quickly (<60 seconds) reducing the venous return/cardiac output from 5 mL/min to 1 mL/min by

1 mL increments, resulting in a severely (90-95%) reduced coronary flow of approximately 0.1

mL/min for one hour. Cardiac parameters were measured after each 1 mL/min reduction in

venous return/cardiac output, and for each minute of the first five minutes of ischemia, as well as

at 15 minutes, 30 minutes, 45 minutes, and 60 minutes of ischemia. Coronary perfusate was

collected continuously for the first 30 minutes of ischemia, and for the last 15 minutes of

ischemia. After 60 minutes of low-flow ischemia, venous return was quickly restored via 1 mL

increments to 5 mL/min similarly in a stepwise fashion, with cardiac parameters measured at

each 1 mL/min increase in flow and for the first 5 minutes of reperfusion. Coronary effluent was

collected in 2 minute increments for the first 14 minutes of reperfusion as well as the final 10 minutes of reperfusion. The heart was reperfused for 120 minutes, during which cardiac parameters were measured at during the first 5 minutes as described above, and then at 15 minutes, 30 minutes, 60 minutes, 90 minutes, and 120 minutes. At the end of the study, the heart was arrested and snap frozen in liquid nitrogen for further study (Figure 3A).

Time course I/R studies

Mouse hearts were subjected to the ischemia/reperfusion protocol as described above,

and hearts were arrested at designated points: 1) 30 minutes of equilibration and no ischemia

(sham), 2) 5 minutes of ischemia, 3) 60 minutes of ischemia + 5 minutes of reperfusion, and 4)

60 minutes of ischemia +120 minutes of reperfusion, and snap frozen in liquid nitrogen for

further analysis (Figure 3).

52

Pharmacological studies

Protein kinase C (PKC) epsilon isoform translocation inhibitor (εV1-2), TAT vehicle, PD

173074, an FGFR inhibitor, GFX, an inhibitor of PKCα, β, , δ, ε, and ζ isoforms, U0125, an inhibitor of ERK1/2, and SB203580, an inhibitor of p38, were employed in the ischemia/reperfusion protocol as described above (Figure 3). εV1-2 and TAT vehicle (KAI pharmaceuticals) were dissolved in water, and immediately aliquotted and frozen. εV1-2 was used at a concentration of 100 nM, in Krebs-Henseleit buffer, a concentration that was determined by existing literature (L. Chen et al., 2001), and confirmed with a concentration- response curve (Figure 10) to establish that the inhibitor concentration used was sufficient to prevent translocation of the PKC isoform from cytosol to membrane fraction (a measure of activity), without affecting the translocation of other isoforms of PKC (α, δ, ζ) that signal through FGF2. PD 173074 was a gift from Pfizer, and was first dissolved in DMSO to improve solubility in Krebs-Henseleit buffer. PD 173074 was used at a concentration of 25 nM, which has been previously established to be sufficient for inhibition of FGFR without off-target effects

(Liao et al., 2010). GFX was used at a concentration of 100µM, a concentration at which PKC isoforms are effectively inhibited (Toullec et al., 1991). U0125 was used at at a concentration of

2.5µM, and SB203580 was used at a concentration of 2µM, concentrations at which ERK1/2 or p38 are inhibited (Cuenda et al., 1995; Favata et al., 1998) Inhibitors were added to a separate reservoir containing Krebs-Henseleit buffer, and flow of perfusate was switched from the reservoir containing Krebs-Henseleit perfusate only to that of Krebs-Henseleit perfusate with the drug. The pharmacological agents were administered for the last 15 minutes of baseline and the first 15 minutes ischemia, and again for the last 15 minutes of ischemia and the first 15 minutes of reperfusion (Figure 3).

53

Immunoblotting

Antibody Source Concentration

FGF2 Santa Cruz 1: 500 PKCε Santa Cruz 1: 500 phospho-PKCε Santa Cruz 1: 500 PKCα Santa Cruz 1: 500 phospho-PKCα Santa Cruz 1: 500 PKCδ Santa Cruz 1: 500 phospho-PKCδ Santa Cruz 1: 500 troponin (total) AbCAM 1: 1000 phospho-FGFR AbCAM 1: 100 FGFR AbCAM 1: 100 FGF12 Santa Cruz 1: 100 Tmod4 Santa Cruz 1: 100 Calreticulin Sigma 1: 300 Myosin light chain Sigma 1: 800 phospho-CAMKIIa SCB 1: 500 CAMKII SCB 1: 500 Affinity SERCA Bioreagents 1: 1000 Affinity Calsequestrin Bioreagents 1: 2500 Affinity phospholamban Bioreagents 1: 1000 phospho-thr17 phospholamban Badrilla 1: 1000 phospho-thr16 phospholamban Badrilla 1: 1000 p38 Santa Cruz 1: 1000 phospho-p38 Santa Cruz 1: 1000 phospho-JNK Promega 1: 1000 JNK Santa Cruz 1: 1000 phospho-ERK 1/2 Santa Cruz 1: 1000 ERK 1/2 Signal Transduction 1: 1000 GAPDH Santa Cruz 1: 500

Table 2. Antibodies, sources, and concentrations used in immunoblotting.

54

Analysis of FGF2 protein content in heart homogenate

To evaluate FGF2 protein content in heart homogenate, FGF2 was first concentrated

using a heparin-sepharose precipitation as previously described (Liao et al., 2010). Briefly, hearts

were pulverized and homogenized on ice with a teflon pestle and glass homogenization tube, in a

buffer containing 20 mM Tris, 2 mM EDTA, 2 M NaCl, 1% NP40, Roche complete mini EDTA-

free inhibitor cocktail (1 tablet/10 mL), and 1 mM PMSF, and spun down at 15,000 g

for 15 minutes. Supernatant was collected, and the protein concentration was determined using a

Biorad DC protein assay, and 1-2 mg of protein was added to homogenization buffer to generate

a volume of 1 mL. 4 mL of Tris-EDTA (TE) buffer was added to the protein to lower the salt

concentration to promote electrostatic and hydrophillic interactions between FGF2 and heparin,

and then incubated with 100 µL of heparin-sepharose bead slurry (Amersham) for 1-2 hours at

4°C in a 15 mL conical tube. After incubation, the beads were spun down for 3 minutes at 4,000 g, and washed three times with a washing buffer (0.6 M NaCl, 10 mM Tris-HCl pH 7.4Roche complete mini EDTA-free protease inhibitor cocktail [1 tablet/10 mL], and 1 mM PMSF), spinning down after each wash for 3 minutes at 4,000 g. Beads were collected after the final wash in a 1.5 mL eppendorf tube and carefully dried by pipetting wash fluid with a 10 mL pipette. Beads were resuspended in 30 µL of loading buffer and boiled for 15 minutes to elute the bound protein. Elutant was run on a 15% SDS-PAGE gel at 130 volts for 45 minutes (or until the loading dye was run off the gel) and then gels were transferred onto a nitrocellulose membrane for 90-120 minutes at 240 mA. Membranes were then briefly stained with a Ponceau reagent to confirm transfer and check for equal protein loading, then immediately placed in a blocking solution of 5% milk in phosphate-buffered saline with 0.1% Tween 20 (PBST), for one hour. Membranes were subsequently placed in primary FGF2 (1:500, Santa Cruz) which detects both HMW and LMW FGF2, diluted in blocking solution and incubated overnight

55

at 4°C. After primary incubation, membranes were washed three times for five minutes with

PBST, and then incubated at room temperature in secondary antibody (rabbit, 1:5000, Santa

Cruz), diluted in blocking solution, for two hours. Following three washes for five minutes each

in PBST, membranes were incubated with ECL reagent for one minute, lightly blotted, and

exposed to film. Densitometry of bands was evaluated by measuring the average Integrated

Density Value (IDV) per unit area using an Alpha Innotech imager and software.

Immunoblotting and detection of proteins involved in kinase signaling, FGF2 signaling, calcium

cycling/handling

Analysis of protein and phospho-protein levels related to FGF2 signaling and calcium handling

To detect changes in total and/or phosphorylated levels of PKC isoforms, MAPKs,

FGFRs, SERCA, calsequestrin, phospholamban, or CaMKIIa, snap-frozen hearts were halved,

rapidly pulverized, and homogenized at 4°C in a glass homogenizer with a teflon pestle in a

buffer of containing 25 mM Hepes, 150 mM NaCl, 1% Triton X-100, 5 mM EDTA, 1%

glycerol, 1 mM sodium orthovanadate, 25 mM β–glycerolphosphate, 50 mM sodium fluoride,

0.5 mM okadaic acid, 100 mM calpain inhibitor, Pefabloc Stock 1 and 2 (Roche), Roche

phosphatase inhibitor (1 tablet/10 mL), Roche complete mini EDTA-free protease inhibitor

cocktail (1 tablet/10 mL), and 1 mM PMSF. Cellular debris was subsequently spun down at

15,000 g for 15 minutes, and the supernatant was pipetted into a clean eppendorff tube.

Homogenate protein concentration was established using a Biorad DC protein assay, and equal amounts of protein were loaded onto a sodium-dodecyl sulfate-polyacrylamide gel (PKC

isoforms, MAPK isoforms, CaMKIIa: 100 μg of protein on a 10% acrylamide gel; FGFR: 150 μg

of protein on an 8% acrylamide gel; SERCA, calsequestrin, and phospholamban: 5 μg of protein

on a 15% acrylamide gel; calreticulin and myosin light chains: 50 μg of protein on a 12%

56

acrylamide gel). The gels were subjected to electrophoresis at 130 volts until the protein loading

dye was run off the gel, and then gels were transferred onto a nitrocellulose membrane for 90-

120 minutes at 240 mA. Membranes were then briefly stained with a Ponceau reagent to confirm

transfer and check for equal protein loading, then immediately placed in a blocking solution of

5% milk in phosphate buffered saline with 0.1% Tween 20 (PBST), for one hour. Membranes

were subsequently placed in primary antibody diluted in blocking solution and incubated

overnight at 4°C. Primary antibodies used were PKCα (1:500, Santa Cruz), phospho-T657 PKCα

(1:500, Santa Cruz), PKCε (1:500, Santa Cruz), phospho-S729 PKCε (1:500, Santa Cruz),

PKCδ (1:500, Santa Cruz), phospho-T507 PKCδ (1:500, Santa Cruz), PKCζ (1:100, Santa Cruz),

phospho-T410 PKCζ (1:100, Santa Cruz), phospho-p38 (1:1000, Santa Cruz), p38 (1:1000, Santa

Cruz), phospho-JNK2 (1:1000, Promega), JNK2 (1:1000, Santa Cruz), phospho-ERK (1:1000,

Santa Cruz), ERK (1:1000 Signal Transduction), SERCA (1:1000, ABR), calsequestrin (1:2500,

ABR), phospholamban (1:1000, ABR), phospho-Thr17 phospholamban (1:1000, Badrilla), phospho-Ser16 phospholamban (1:1000, Badrilla), FGFR1 (1:100, Abcam), phospho-FGFR1

(1:100, Abcam), phospho-CaMKIIa (1:500, Santa Cruz), and CaMKIIa (1:500, Santa Cruz).

After primary incubation, membranes were washed three times for five minutes with PBST, and

then incubated at room temperature in secondary antibody (rabbit, 1:5000, Santa Cruz for all

PKC isoforms, all phosphorylated PKC isoforms, p38, phospho-JNK2, phospho-ERK1/2,

SERCA, calsequestrin, phospho-Thr17 and phosphor-Ser16 phospholamban, FGFR1, CaMKIIa, and GAPDH,or mouse, 1:10,000, Biorad for phosphor-p38, JNK2, ERK1/2, phospholamban,

phosphor-FGFR1, phospho-CaMKIIa), diluted in blocking solution, for two hours. Following

three washes for five minutes each in PBST, membranes were incubated with ECL reagent for

one minute, lightly blotted, and exposed to film. Densitometry of bands was evaluated by

measuring the average Integrated Density Value (IDV) per unit area using an Alpha Innotech

57

imager and software.

Analysis of PKC translocation to the membrane

The degree of PKC isoform activation by evaluating translocation to the membrane was performed using centrifugal fractionation to separate soluble (cytosolic) and particulate

(membrane) fractions. Snap-frozen halved hearts were rapidly pulverized, and homogenized at

4°C in a glass homogenizing tube with a teflon pestle in a buffer of containing 25 mM Tris, 4 mM EGTA, 2 mM EDTA, 5 mM DTT, Roche complete mini EDTA-free protease inhibitor cocktail (1 tablet/10 mL), and 1 mM PMSF. Homogenate was subjected to centrifugation at

100,000 g, at 4°C, for 30 minutes, and the supernatant (soluble fraction) was collected. The pellet, containing the particulate fraction and cellular debris, was resuspended in the homogenization buffer with 0.1% Triton, and incubated on ice for 30 minutes, after which the rehomogenized pellet was subjected to centrifugation at 100,000 g, at 4°C, for 30 minutes. The supernatant (the particulate fraction) was collected and the pellet was discarded. Membrane and cytosol fractions were run side-by-side on a 10% SDS-PAGE at 130 volts until the protein loading dye was run off the gel, and then gels were transferred onto a nitrocellulose membrane for 120 minutes at 240 mA. Membranes were then briefly stained with a Ponceau reagent to confirm transfer and check for equal protein loading, then immediately placed in a blocking solution of 5% milk in phosphate buffered saline with 0.1% Tween 20 (PBST), for one hour.

Membranes were subsequently placed in primary antibody diluted in blocking solution and incubated overnight at 4°C. Antibodies used were PKC α (1:500, Santa Cruz), PKC ε (1:500,

Santa Cruz), PKC δ (1:500, Santa Cruz), and PKC ζ (1:100, Santa Cruz). After primary incubation, membranes were washed three times for five minutes with PBST, and then incubated at room temperature in secondary antibody (rabbit, 1:5000, Santa Cruz), diluted in blocking

58

solution, for two hours. Following three washes for five minutes each in PBST, membranes were incubated with ECL reagent for one minute, lightly blotted, and exposed to film. Densitometry of bands was evaluated by measuring the average Integrated Density Value (IDV) per unit area using an Alpha Innotech imager and software.

Evaluation of Nitrite Levels in Coronary Effluent

Activity of nitric oxide synthases (NOSs) during ischemia and reperfusion was

- determined by evaluating the concentration of nitrite (NO2 ) in coronary effluent collected for

two -minute intervals at 4 and 2 minutes before ischemia, for the first 30 minutes and last 15

minutes of ischemia, for two -minute intervals after the onset of reperfusion for the first 14

- minutes, and the last two minutes of reperfusion (see protocol bar, Figure 5). Levels of NO2 , the result of rapidly oxidized NO, the product of NOS, was determined using a fluorometric nitric oxide assay kit (Calbiochem). The assay utilizes the reaction of nitrite with 2,3- diaminonapthalene (DAN) to a fluorescent 1-(H)-napthotriazole at a rate relative to the amount

of nitrite present. 200 μL of nitrite standard curve reagent or effluent was pipetted into the wells of a transparent 96-well plate, with 50 μl of assay buffer. 10 μL of a fluorometric reagent, DAN was added to each well, and the plate was covered and incubated away from light for 10 minutes.

During this incubation, DAN was reduced to the fluorescent product napthotriazole at a rate determined by the concentration of nitrite. This reaction was stopped by adding 20 μL of 2.8 M

NaOH to each well, and immediately the fluorescence was read with an excitation wavelength of

360 nm and emission wavelength at 450 nm using a GENios fluorometer plate reader (Tecan). A standard slope was used to convert fluorescence readings to nitrite concentrations, and concentrations were normalized to coronary flow and heart weight.

59

Assessment of Cardiac Myofilament Activity

Cardiac were prepared and assessed for actomyosin ATPase activity,

myofibril protein phosphorylation, and PKC isoform translocation in the laboratory of Dr. W.

Glen Pyle. Hearts were kept on ice and homogenized in a buffer of 60 mM KCl, 2 mM MgCl2,

30 mM imdazole (pH 7.0), 0.2 mM PMSF, 0.1 mM Leupeptin, 0.01 mM Pepstatin A.

Homogenate was centrifuged at 12,000g for 15 minutes at 4°C, and the pellets were resuspended

in skinning buffer (1% Triton X-100, 10 mM EGTA, 8.2 mM MgCl2, 14.4 mM KCl, 60 mM

imidazole (pH 7.0), 5.5 mM ATP, 12 mM creatine phosphate, 10 U/mL creatine phosphokinase,

0.2 mM PMSF, 0.1 mM Leupeptin, 0.01 mM Pepstatin A), mixed at 4°C for 45 minutes, and

subsequently centrifuged at 1,100g for 15 minutes at 4°C. Supernatant was discarded and the

pellets were washed 3 times in cold homogenization buffer. Protein concentration was measured

using a Bradford assay.

Actomyosin Mg2+ ATPase activity in ischemic/reperfused mouse hearts

Myofilaments isolated from hearts were analyzed for actomyosin Mg2+ATPase activity as

previously described (Pyle et al., 2003). Purified myofilaments (50 µg) were incubated at 32°C

for 15 minutes in buffers containing increasing amounts of Ca2+, which were made by mixing maximally activating and relaxing ATPase buffers. Maximally activating ATPase buffer contained 23.5 mM KCl, 5 mM MgCl2, 3.2 mM ATP, 2 mM EGTA, 20 mM imidazole, 2.2 mM

CaCl2, 0.2 mM PMSF, 0.1 mM Leupeptin, and 0.01 mM Pepstatin A (pH 7.0). The relaxing

ATPase buffer contained 26 mM KCl, 5.1 mM MgCl2, 3.2 mMATP, 2 mM EGTA, 20

mMimidazole, 4.9 µM CaCl2, 0.2 mM PMSF, 0.1 mM Leupeptin, and 0.01 mM Pepstatin A (pH

7.0). Free Ca2+ was calculated using the program of Patton and colleagues (Patton, Thompson, &

Epel, 2004).10% trichloroacetic acid was used to halt the reaction, and the samples were

60

subsequently centrifuged (14,100 g for 3 minutes). The supernatant was evaluated for the levels

of inorganic phosphate, which was determined colorimetrically with the addition of an equal

amount of developing solution (0.5% FeSO4: 0.5% ammonium molybdate in 0.5 M H2SO4) to the supernatant. The absorbance was then measured at 630 nm.

Contractile protein phosphorylation

Contractile proteins (40 μg) were run on 12.5% SDS-PAGE gels and fixed in 50%

methanol-10% acetic acid overnight at room temperature. The fixed gels were stained for

phosphorylated proteins with Pro-Q Diamond (Molecular Probes, Eugene, OR), according to the

manufacturer's instructions. Stained gels were then imaged using a Typhoon gel scanner (GE

Healthcare, Baie d’Urfé, PQ), and analyzed with Image J software (NIH, Bethesda, MD). After

imaging, Coomassie stain was used to assess protein loading.

Translocation of PKC isoforms to myofibrils

On 10% polyacrylamide gels, myofilament proteins (75 μg) were separated using

electrophoresis, and transferred to a nitrocellulose membrane. Membranes were probed with

antibodies for PKCα, PKCδ, and PKCε (1:1,000, BD Biosciences, Mississauga, Ontario, Canada,

and Upstate, Mississauga, Ontario, Canada). Anti-actin antibody (1:25,000, Millipore,

Etobicoke, Ontario, Canada) was used to control for equal loading. Secondary antibodies

(Transduction Labs) conjugated to horseradish peroxidase were used at 1:5,000 dilution. PKC

bands were detected using Western Lightning (PerkinElmer Life and Analytical Sciences,

Woodbridge, ON). Bands were analyzed using Image J software, and band density was

normalized to actin and expressed as a percent change in band density.

61

Myocyte Isolation and Calcium Transient Evaluation

Mouse cardiomyocyte isolation

Myocytes were isolated in collaboration with the laboratory of Dr. E.G. Kranias, as previously described (DeSantiago, Maier, & Bers, 2002). Age (10-14 week old)- and sex- matched mice were anesthetized with sodium pentobarbital (150 μL, i.p.), and heparinized.

Hearts were excised and the aorta was rapidly cannulated on a Langendorff perfusion apparatus.

Hearts were immediately subjected to retrograde perfusion with 37°C oxygenated working

Tyrode solution (0.118 M NaCl, 5.4 mM KCl, 10 mM HEPES, 33 mM NaH2PO4, 2.0 mM

MgCl2, 10 mM dextrose, 30 mM taurine, and 10 mM 2,3-Butanedione monoxime [BDM], pH adjusted to 7.4), and equilibrated for 10 minutes. Perfused hearts were then digested for 8-10 minutes by perfusion with 0.1% collagenase/albumin in the working Tyrode solution, until tissue was spongy and soft, and then removed from the cannula and gently broken apart with forceps and pipette. Cell mixture was filtered through a cheese cloth to remove proteins and, after confirming the presence of healthy (striated and rectangular) cells under a microscope, allowed to settle into a pellet at room temperature for 5 minutes. Supernatant was pipetted off and cells were resuspended in warmed oxygenated working Tyrode solution for with increasing concentrations of calcium (20 μM,100 μM,200 μM,1 mM), for 5 minutes at each concentration. At 1 mM calcium, cells were evaluated once again under a microscope to confirm a high number of cells, with 30-50% healthy cells. Cells were then incubated on cover slips preloaded with laminin for 15 minutes, after which myocyte function and calcium transient properties were evaluated.

Evaluation of isolated cardiomyocyte function

Isolated cardiomyocytes loaded onto coverslips were gently perfused in a Plexiglass

62

chamber with a constant wash of 1.8 mM calcium in working Tyrode solution, and evaluated at

20X magnification on an inverted epifluorescence microscope (Nikon Diaphot 200).

Cardiomyocytes were stimulated with a Grass S5 stimulator in an electrical field at 10-12 mV at

0.5 Hz, and 10-15 healthy (rectangular and striated) cells were digitally recorded on a computer.

Video edge detection using FELIX software (Photon Technology International) was used to

evaluate the function of the cardiomyocyte. Total length, rate of contraction and relaxation

(+dL/dt and -dL/dt, respectively) and fractional shortening of 10-15 cardiomyocytes were

measured over 6 beats, and analyzed using Ionwizard (Ionoptix) software.

Cardiomyocyte calcium transients and sarcoplasmic reticulum load

Coverslips containing cardiomyocytes were incubated with 1 mM Fura-2AM (a cell permeable calcium dye) in 1.0 mM calcium in working Tyrode solution away from light for 45 minutes. The supernatant was gently pipetted off, and coverslips were incubated with a wash of

1.8 mM calcium in working Tyrode solution for 30 minutes. Cells were gently perfused with a

constant wash of 1.8 mM calcium in working Tyrode solution, and evaluated at 40X

magnification using FELIX software. Cardiomyocytes were stimulated in an electrical field at

10-12 mV at 0.5 Hz, and 10-15 healthy (rectangular and striated) cells were measured.

Cardiomyocytes were excited by a Delta Scan dual-beam spectrophotofluorometer (Photon

Technology International), fluorescence at 340 nm and 380 nm was measured over the course of

6-10 beats, and then stimulation was removed. The cardiomyocyte was then perfused with a 50

mM caffeine solution to stimulate the release of calcium from the sarcoplasmic reticulum, and

fluorescence at 340 nm and 380 nm was recorded using using FELIX software (Photon

Technology International). Changes in cytosolic calcium levels were taken as the change in

340/380 ratio from baseline fluorescence, and peak amplitude, relaxation time and tau were

63

analyzed using Ionwizard (Ionoptix) software.

Microarray Assessment of Gene Expression in Mouse Hearts with only LMW or HMW

FGF2

RNA preparation

Mouse hearts snap-frozen in liquid nitrogen were sliced into four sections and submerged

in RNAice (Ambian) at -80°C at least 24 hours prior to RNA extraction. 30 μg of tissue was

minced and homogenized at room temperature in Qiagen RLT Lysis buffer and incubated with

proteinase K (Roche) at 55°C for 10 minutes. Homogenate was centrifuged at 8,000 g for 3

minutes, and the supernatant was transferred to a clean tube with 100% ethanol to precipitate

RNA. Precipitated RNA was filtered out with a column (Qiagen) and washed with Qiagen RW1

Wash buffer, then subjected to DNAse treatment for 15 minutes. Column was washed once more with RW1 and twice with Qiagen RPE Wash buffer + 30% ethanol, discarding flow-through for each wash. Column was dried by centrifuging at 8,000 g for 1 minute. RNA was solubilized in

30 µL RNAse free water, and collected by spinning the column at 8,000 g for 1 minute. Flow- through was pipetted back through the column and spun again to increase RNA yield. Purity of

RNA was determined by reading the absorbance at 260 nm and 280 nm. An absorbance of at least 0.1500 at 280 nm and a 260/280 ratio of >1.7 was used for microarray and RT-qPCR studies. Additionally, RNA integrity was evaluated by electrophoresis of 1 μL of sample on a 1% agarose gel stained with ethidium bromide. The presence of two discrete ribosomal bands at 5 kilobases (28S) and 2 kilobases (18S), indicated intact RNA; a smear or the absence of bands indicated degraded RNA, and the sample was discarded.

64

RNA Microarray

RNA was further analyzed for purity at the University of Cincinnati Genomics and

Microarray Laboratory using an Agilent nanodrop spectrophotometer. RNA that was confirmed to have 260/280 ratio of 1.7 or greater was then subjected to a GeneChip Mouse Gene 1.0 ST

Array, testing each of the 28,853 mouse genes with approximately 27 probes for each gene. RNA from three biological sets each of Fgf2 KO, HMWKO and LMWKO were compared to determine difference in regulation due to expression each class of isoforms of FGF2 (Figure 6).

Genes that were found to have a change of 1.25-fold or greater with a p-value of 0.005 or less were returned for further analysis. A total of 401 genes were identified, which were further analyzed according to their biological role; genes known to affect the function of the contractile apparatus, mitochondria, and calcium handling, as well as survival were identified by the

Database for Annotation, Visualization and Integrated Discovery (DAVID). Changes in these proteins were then confirmed using RT-qPCR and immunoblotting.

65

Overlap: 35 Genes LMWKO HMWKO

FGFKO

Genes regulated by Genes regulated HMW FGF2 only by LMW FGF2 only Genes regulated by both isoforms of FGF2

Figure 6. Microarray study design. mRNA from non-ischemic hearts of mice expressing only HMW FGF2 (LMWKO), LMW FGF2 (HMWKO) or with no isoforms of FGF2 expressed (FGFKO) was analyzed, and overlap between each group was analyzed, as well as genes with altered expression common to all three.

66

Real-time reverse transcriptase PCR (qRTPCR) validation

RNA isolated as described above was reverse transcribed using SuperScript II Reverse

Transcriptase kit (Inivtrogen). 200 ng of RNA was incubated with random hexamers and mixed

dNTPs at 65°C for 5 minutes and placed on ice. The RNA mix was then added to 1X RT buffer,

5 mM MgCl2, 0.01 M DTT and 1 μL RNAse OUT and incubated for 2 minutes at room

temperature. One μL of SuperScript II reverse transcriptase was added and the mixture was

incubated at room temperature for 10 minutes, followed by 50 minutes at 42°C, 15 minutes at

70°C, and 1 minute at 4°C. At 4°C, 1 μL of RNAse H was added to each sample to degrade the

template RNA, leaving only cDNA. cDNA generated from control RNA was evaluated with a

control PCR prior to qPCR. The control PCR tested four 10-fold dilutions of each cDNA

(1:1000, 1:10,000, 1:100,000, 1:1,000,000), using 15.4 μL DEPC-treated water, 2 μL 10X PCR

buffer (Roche), 1 μL 2.5 mM dNTPs, 0.5 μL cDNA, 0.5 μL Primer A (5'-GAC ATG GAA GCC

ATC ACA GAC), 0.5 μL Primer B (5'-AGA CCG TTC AGC TGG ATA TTA C), and 0.1 μL

Taq (Roche).

For sarcolipin, cDNA was evaluated with via real time PCR (qPCR) using SYBR green

reagent (Sigma). qPCR conditions, input RNA levels, and input cDNA levels were optimized to

ensure that amplification increased linearly and to minimize primer dimer formation (as determined by post-amplification melt curve). 18S was used as an internal control, after confirmation that alteration of FGF2 expression did not change 18S expression between genotypes evaluated. Primers used for sarcolipin amplification were: 5’-TGT GCC CCT GCT

CCT CTT C and 5’-TGA TTG CAC ACC AAG GCT TG (Vasu et al., 2009), input RNA was

200 ng and cDNA levels were diluted 1:10. 2 μL of diluted cDNA was added to 12.5 μL SYBR green mix (SA Biosciences), 1.5 μL 10 mM primer mix (SA Biosciences), and 9 μL DEPC water. Sarcolipin qPCR conditions were: 95°C for 10 min followed by 40 cycles, each at 95°C

67

for 15 s (denaturing) and 60°C for 1 min (annealing and extension). qPCR product was

subsequently subjected to electrophoresis to confirm a PCR product of the 80 base pairs.

Immunoblot validation

To detect changes in total calreticulin, myosin light chains, tropomodulin 4 and FGF12,

snap-frozen hearts were halved, rapidly pulverized, and homogenized at 4°C in a glass

homogenizer with a teflon pestle in a buffer of containing 25 mM Hepes, 150 mM NaCl, 1%

Triton X-100, 5 mM EDTA, 1% glycerol, 1 mM sodium orthovanadate, 25 mM β–

glycerolphosphate, 50 mM sodium fluoride, 0.5 mM okadaic acid, 100 mM calpain inhibitor,

Pefabloc Stock 1 and 2 (Roche), Roche phosphatase inhibitor (1 tablet/10 mL), Roche complete

mini EDTA-free protease inhibitor cocktail (1 tablet/10 mL), and 1 mM PMSF. Cellular debris

was subsequently spun down at 15,000 g for 15 minutes, and the supernatant was pipetted into a

clean eppendorff tube.

Homogenate protein concentration was established using a Biorad DC protein assay, and equal amounts of protein were loaded onto a 12% (calreticulin and tropomodulin 4) or 15%

(FGF12 and myosin light chains) sodium-dodecyl sulfate-polyacrylamide gel . The gels were subjected to electrophoresis at 130 volts until the protein loading dye was run off the gel, and then gels were transferred onto a nitrocellulose membrane for 90-120 minutes at 240 mA.

Membranes were then briefly stained with a Ponceau reagent to confirm transfer and check for equal protein loading, then immediately placed in a blocking solution of 5% milk in phosphate buffered saline with 0.1% Tween 20 (PBST), for one hour. Membranes were subsequently placed in primary antibody diluted in blocking solution and incubated overnight at 4°C. Primary antibodies used were: calreticulin (1:300, Sigma), myosin light chains (1:800, Sigma), tropomodulin 4 (1:100, Santa Cruz) and FGF12 (1:100, Santa Cruz). After primary incubation,

68

membranes were washed three times for five minutes with PBST, and then incubated at room temperature in secondary antibody (rabbit, 1:5000, Santa Cruz for calreticulin, goat 1:5000,

Santa Cruz for FGF12 and tropomodulin 4, or mouse, 1:10,000, Biorad for myosin light chains), diluted in blocking solution, for two hours. Following three washes for five minutes each in

PBST, membranes were incubated with ECL reagent for one minute, lightly blotted, and exposed to film. Densitometry of bands was evaluated by measuring the average Integrated Density Value

(IDV) per unit area using an Alpha Innotech imager and software.

Statistical Analysis

All data given in figures and text is represented as the mean value ± standard error of the mean (SEM). Statistical significance, assigned to where probability is less than or equal to 5 percent (p<0.05), is indicated where present. Comparison between single data points in two groups was done with a Student’s t-test; this includes differences in baseline or post-ischemic contraction and relaxation, and differences in protein levels for untreated or untreated hearts between genetically-modified and wildtype hearts. Time course measurements between untreated hearts of two different genotypes, including functional data over the course of the I/R experiments, kinase translocation and activation, and protein phosphorylation evaluated by SDS-

PAGE, as well as nitric oxide release evaluated by nitrite assay, were analyzed with a one-way analysis of variance (ANOVA) followed by a post-hoc Student’s t-test. Hearts treated with inhibitors, or double knockouts, including functional data over the course of the I/R experiments, kinase translocation and activation, and protein phosphorylation evaluated by SDS-PAGE, as well as nitric oxide release evaluated by nitrite assay were analyzed by a two-way ANOVA followed by a post-hoc Student’s t-test.

The microarray was evaluated according to the method of Smyth and group (Smyth,

69

2004), and performed with R statistical software on the Bioconductor platfrom. Each gene was evaluated by fitting the following model, based off of analysis of variance (ANOVA): Yijk = m

+ Ai + Sj + Ck+ eijk, where Yijk is the normalized log-intensity on the array corresponding to i and the treatment condition responding to j, labeled with the dye number corresponding to k (k=1

Cy5: k=1, Cy3: k=2). M is defined as the mean log-intensity. A is the effect of the array, S is the effect of the treatment, and Ck is the effect of the dye. Following each comparison, T-statistics were calculated using an empirical Bayesian moderated T-method, using estimates of variance from all genes to improve the individual gene estimates. False discovery rates were less than

1.00 and a significance of p<0.005 were used to return genes of interest with a change of 1.25- fold or greater.

70

Chapter 1: A role of protein kinase C isoforms alpha and epsilon in low molecular weight

fibroblast growth factor-mediated protection from post-ischemic cardiac dysfunction

RESULTS I.

Expression of only LMW FGF2 results in differential activation of PKC isoforms epsilon

and alpha.

Expression of only LMW FGF2 has been shown by our laboratory to prevent post-

ischemic dysfunction, compared to wildtype hearts and hearts only expressing HMW FGF2

(Liao, 2008; Liao et al., 2010). Additionally, overexpression of all human isoforms of FGF2 has

been shown to improve the contractility and relaxation of the heart after I/R injury, in a manner

that is PKC-dependent (House et al., 2007). However, it is not yet known which isoforms of

PKC, if any, are involved in LMW-FGF2 mediated protection from post-ischemic dysfunction.

To examine the effects of LMW FGF2 expression on PKC isoform activation during I/R injury, immunoblotting was used to determine the phosphorylation of PKC isoforms at baseline, early ischemia (5 minutes ischemia), early reperfusion (60 minutes ischemia + 5 minutes reperfusion), and late reperfusion (60 minutes ischemia + 120 minutes reperfusion) in wildtype and FGF2

HMWKO hearts (only expressing LMW FGF2). The isoforms examined were α, ε, and δ, as

these isoforms have been previously determined to be activated by FGF2 in the heart, and play a

role in the modulation of cardiac dysfunction (Bouwman et al., 2004; Budas et al., 2007; C. H.

Chen et al., 1999; Churchill & Mochly-Rosen, 2007; G. W. Dorn et al., 1999; Fryer, Wang et al.,

2001; Hassouna et al., 2004; House et al., 2003; House et al., 2007; K. Inagaki et al., 2003; K.

Inagaki et al., 2003; Jiang et al., 2002; Kanaya et al., 2003; Kudo et al., 2002; Lochner et al.,

2009; Maslov et al., 2009; Melling et al., 2009; Miki et al., 1998; Padua et al., 1998; P. Ping et

al., 1997; Pyle et al., 2003; Sheikh et al., 2001; Turrell et al., 2011; Uecker et al., 2003; Wickley

71

et al., 2006; Yoshida et al., 1996; Yoshida, Kawamura, Mizukami, & Kitakaze, 1997; H. Y.

Zhang et al., 2002; H. Z. Zhou et al., 2002). It was determined that none of the isoforms examined showed a difference in phosphorylation at baseline. In HMWKO hearts, PKCα showed significantly increased activation at early ischemia (p<0.05), and significantly decreased activation at early reperfusion (p<0.05) compared to wildtype, with no difference between the genotypes at late reperfusion (Figure 7). PKCε showed no difference in activation between

HMWKO and WT at early ischemia, but significantly increased activation at early reperfusion

(p<0.05), and significantly decreased activation at late reperfusion (p<0.05), compared to wildtype (Figure 7). PKCδ showed no significant difference in phosphorylation at any of the timepoints measured during ischemia or reperfusion (Figure 7). No significant differences were seen in the activation of any of the isoforms within the same genotype across the measured timepoints during I/R. These findings are similar to those found by Jiang and colleagues (Jiang et al., 2002), who determined that 30 minutes of ischemia, followed by 30 minutes of reperfusion resulted in an increased translocation, in the presence of recombinant FGF2, of PKCα to membrane fractions, and of PKCε and ζ to the particulate fraction, suggesting increased activation of these PKC isoforms in the heart. Similarly, Padua and group found that sarcolemma-associated PKCε increased significantly, and PKCα increased slightly, after treating non-ischemic rat hearts with recombinant FGF2 (Padua et al., 1998). Alternately, Sheikh and colleagues (Sheikh et al., 2001), who examined the effect of the overexpression of LMW FGF2 in mouse hearts, found higher levels of membrane-associated PKCα in transgenic hearts compared to wildtype, suggesting higher activity of this isoform at basal levels when this FGF2 isoform is overexpressed; this group additionally found that there was higher cytosol-associated

PKCε, suggesting lowered basal activity. Finally, our laboratory has previously examined the translocation of PKC isoforms during I/R when all isoforms of FGF2 are overexpressed (House

72

et al., 2007). It was determined that, as seen here when only LMW FGF2 is expressed, PKCε translocation increased during early reperfusion. However, PKCα translocation was decreased during I/R injury, suggesting that the presence of HMW FGF2 alters the translocation of this isoform; further examination of this question is given in chapter 4 of this dissertation.

A.

Baseline 5’I 60’I + 5’R 60’I + 120’R WT KO WT KO WT KO WT KO pPKCα

Total PKCα

pPKCε

Total PKCε

pPKCδ

Total PKCδ

73

B. *p<0.05 vs. WT 1.4 Wildtype 1.2 HMWKO 1 * 0.8 * 0.6 (arbitrary units) (arbitrary 0.4 0.2

Phosphorylated/Totalalpha PKC * 0 ShamBaseline 5’I5'I 60’I60'I + +5’R 5'R60’I 60'I + + 120’R120'R

C. 1 Wildtype 0.8 HMWKO * 0.6 * 0.4 (arbitrary units) (arbitrary 0.2

Phosphorylated/Totalepsilon PKC 0 ShamBaseline 5’I5'I 60’I60'I + +5’R 5'R60’I 60'I + +120’R 120'R

D. 0.6 Wildtype 0.5 HMWKO 0.4 0.3 0.2 0.1 0 (arbitrary units) (arbitrary ShamSham 5’I 5'I 60’I +60'I 5’R + 5'R60’I 60'I + 120’R + 120'R Phosphorylated/Total delta PKC

74

Figure 7. Representative immunoblot (A) and quantitation (B-D) for the phosphorylation of PKC isoforms during ischemia and reperfusion as a measure of kinase activation for

PKCα (B), ε (C), and δ (D) at baseline (sham), early ischemia (5’I), early reperfusion (60’I

+ 5’R) and late reperfusion (60’I + 120’R) in wildtype (diamond) and FGF2 HMWKO

(square) hearts. PKCα showed significantly higher phosphorylation in HMWKO hearts compared to wildtype at early ischemia, and significantly lowered phosphorylation at early reperfusion (B); PKCε showed significantly higher phosphorylation at early reperfusion, but significantly lowered phosphorylation at late reperfusion in HMWKO compared to wildtype hearts. PKCδ was not significantly different between HMWKO and wildtype hearts at any time point during I/R. p<0.05 vs. wildtype (n=5-8 per group).

75

Expression of only LMW FGF2 improves post-ischemic function in isolated mouse hearts.

Hearts from 10-12 week old, sex-matched mice only expressing LMW FGF2 and

wildtype littermates were subjected to 60 minutes of global, low-flow ischemia and 120 minutes

of reperfusion on an isolated working heart apparatus. Our laboratory (Liao et al., 2010) has

previously reported that expression of only LMW FGF2 results in improved contractility and

relaxation after I/R injury. This was confirmed, as hearts expressing only LMW FGF2 showed a

significantly higher percentage of functional recovery after I/R injury compared to their wildtype

cohorts (HMWKO: 73% ±3contractile recovery, 59%±5; WT: 41% ± 12contactile recovery,

41%±7 relaxation recovery, Figure 8). FGF2 HMWKO hearts showed a significant increase in both post-ischemic contractility (+dP/dt) and relaxation (-dP/dt) compared to wildtype (p<0.05,

Figure 8).

76

A.

100

80 *p<0.05 vs. WT *

60

40

20

0 WTWT HMWKOHMWKO % Contractile Recovery (+dP/dt) Recovery Contractile %

B. 100 *p<0.05 vs. WT 80 * 60

40

20

0 WT HMWKOHMWKO

% Relaxation (-dP/dt)Recovery % WT

Figure 8: The percent recovery of post-ischemic contractility (+dP/dt) and relaxation

(-dP/dt) for wildtype (black bar) or FGF2 HMWKO (striped bar) hearts that have undergone 60 minutes ischemia and 120 minutes reperfusion. Post-ischemic functional recovery is expressed as a percentage of baseline function. HMWKO hearts had significantly higher post-ischemic contractility (A) and relaxation (B) compared to wildtype hearts. *p<0.05 vs. wildtype (n=5-10 per group).

77

Ablation of PKCα expression abrogates the improvement in post-ischemic contractility seen when only LMW FGF2 is expressed.

As hearts only expressing LMW FGF2 showed a higher level of PKCα phosphorylation at early ischemia than wildtype cohorts (Figure 9), it was hypothesized that the increase in activation of this PKC isoform may contribute to the cardioprotective phenotype of these hearts.

To test this, mice expressing only LMW FGF2 were bred to mice with no PKCα expression.

Age- and sex-matched wildtype, FGF2 HMWKO, PKCα knockout (KO), and FGF2 HMWKO crossed to PKCα KO (HMWKOxPKCαKO) mice were subjected to 60 minutes of global, low- flow ischemia and 120 minutes of reperfusion on an isolated working heart apparatus. As previously observed (Liao et al., 2010) ( and see Figure 9), FGF2 HMWKO hearts recovered their post-ischemic cardiac function to a significantly (p<0.05) greater extent than wildtype cohorts, with a marked increase in both post-ischemic contractility (+dP/dt) and relaxation (- dP/dt). Interestingly, hearts with an absence of both HMW FGF2 and PKCα expression showed a significantly poorer post-ischemic recovery of cardiac function compared to hearts with only

HMW FGF2 ablated (Figure 9, p<0.05), suggesting that the improvement in post-ischemic function mediated by LMW FGF2 requires PKCα. There was no significant difference in post- ischemic recovery of cardiac function between PKCα knockout and wildtype hearts, suggesting that PKCα plays a role in the recovery of the heart from I/R only when LMW FGF2 alone is expressed.

78

*p<0.05 vs. WT †p<0.05 vs. HMWKO A. 100 * HMW WT 80 HMWKO

60 † 40

20

0 PKCPKC alphaα present PKC PKCaKOα absent % Contractile Recovery (+dP/dt) Recovery Contractile %

B. *p<0.05 vs. WT

) 100

/d t †p<0.05 vs. HMWKO Wildtype p -d ( HMWKO 80

erfusion erfusion * p 60 † 40

20

% 120m at re Relaxation in. 0 PKCPKC alphaα present present PKCPKCaKO α absent % Relaxation Recovery (-dP/dt) Recovery Relaxation %

Figure 9. The percent recovery of post-ischemic contractility (+dP/dt) and relaxation

(-dP/dt) of wildtype (black bar) or FGF2 HMWKO (striped bar) hearts that have undergone 60 minutes ischemia and 120 minutes reperfusion, in the presence or absence of

PKCα . Recovery is expressed as a percentage of baseline function. HMWKO hearts had significantly higher post-ischemic contractility (A) and relaxation (B) compared to wildtype hearts; this recovery was abrogated in the absence of PKCα. *p<0.05 vs. wildtype (n=5-10

per group).

79

PKCε RACK binding peptide, εV-1, prevents PKCε translocation, while not interfering with PKCα or PKCδ

Given that hearts only expressing LMW FGF2 showed a significantly higher level of

PKCε phosphorylation at early reperfusion compared to wildtype cohorts (Figure 7), the effect of

PKCε inhibition on these hearts was explored to further elucidate the role of this PKC isoform in

LMW FGF2-mediated protection from post-ischemic myocardial dysfunction. A recently available translocation inhibitor for PKCε, εV-1, was first evaluated in a concentration-response curve, to determine the effective concentration that would prevent translocation of PKCε without affecting the translocation of other PKC isoforms in the mouse heart during I/R injury. The concentration response curve was first performed in hearts overexpressing all isoforms of FGF2 and confirmed in FGF2 HMWKO hearts. Ten-fold varying concentrations of εV-1 or TAT vehicle were given 15 minutes prior to the onset of ischemia continuously until 15 minutes of ischemia, and again at 45 minutes of ischemia through reperfusion (see protocol, Figure 5).

Hearts were flash-frozen after 5 minutes of reperfusion, as this was the timepoint at which PKCε has been previously shown to translocate to membrane fractions in an isolated working heart model of I/R (House et al., 2007), and the translocation of PKCε was evaluated via immunoblotting of soluble and particulate fractions. It was determined in hearts overexpressing all isoforms of FGF2 that concentrations above 63 nM selectively blocked PKCε, without altering the translocation of PKCα or δ (Figure 10). Similarly, in FGF2 HMWKO hearts, 100 nM selectively blocked translocation of PKCε at early reperfusion, without altering the translocation of PKCα at 5 minutes of reperfusion and δ at 30 minutes of ischemia, both times when these

PKC isoforms show the highest translocation (House et al., 2007) (Figure 11). Therefore, further evaluations of the effects of PKCε in FGF2 HMWKO hearts used a 100 nM concentration of εV-

1.

80

A. PKC epsilon Tg 5 ** NTg 4 *p<0.05 vs. TAT 3 2

fraction 1 0

Cytosol/membrane 630 63 6 TAT PKC epsilon inhibitor concentration (nM)

B. PKC alpha Tg 8 NTg 6 4 fraction 2 0 Cytosol/membrane 630 63 6 TAT PKC epsilon inhibitor concentration (nM)

C. 16 PKC delta Tg NTg 14 12 10 8 6 4 fraction 2 0

Cytosol/membrane 630 63 6 TAT PKC epsilon inhibitor concentration (nM)

Figure 10. Quantitation of the translocation of PKC isoforms at 60 minutes ischemia+ 5 minutes reperfusion (when PKCε is activated; see Figure 7), using the PKCε inhibitor (εV-

1) at 630 nM, 63 nM, 6 nM , and vehicle treatment (TAT). Less PKCε is translocated to the membrane at 60 minutes ischemia + 5 minutes reperfusion with εV-1 (A), while translocation of PKCα and δ are not affected by the inhibitor (B and C). Cy: cytosol fraction, M: membrane fraction (n=3 hearts per genotype, per treatment group).

81

PKCε inhibitor 100 nM PKCε inhibitor 100 nM (30 min ischemia) (60 min Ischemia + 5 min reperfusion) Vehicle PKCε inhibitor Vehicle PKCε inhibitor C MMC C MMC y y y y PKC ε

PKC δ PKC α

Figure 11. Translocation of PKC isoforms in the presence of 100 nM PKCε inhibition.

Representative immunoblots for PKCε, δ, and α at 30 minutes ischemia, and 60 minutes ischemia + 5 minutes reperfusion (when PKCε is activated; see Figure 7). Less PKCε is translocated to the membrane at 60 minutes ischemia+ 5 minutes reperfusion, while translocation of PKCα and δ are not affected by the inhibitor. Cy: cytosol fraction, M: membrane fraction (n=3 hearts per treatment group).

82

Selective inhibition of PKCε translocation attenuates the improvement in post-ischemic contractility seen when only LMW FGF2 is expressed.

To test the hypothesis that LMW FGF2 relies on PKCε activity to protect the heart from

I/R-induced myocardial dysfunction, mouse hearts only expressing LMW FGF2 and wildtype cohorts were subjected to 60 minutes of global, low-flow ischemia and 120 minutes of reperfusion. As previously seen (Figure 8), FGF2 HMWKO hearts had a significant increase in both post-ischemic contractility (+dP/dt) and relaxation (-dP/dt) compared to their wildtype cohorts (Figure 12, p<0.05). Administration of the PKC epsilon inhibitor, εV-1 (100 nM) significantly reduced the post-ischemic recovery of cardiac function in FGF2 HMWKO hearts compared to TAT vehicle-treated cohorts or wildtype hearts (Figure 12, p<0.05). This suggests that the improvement in post-ischemic myocardial function mediated by the presence of only

LMW FGF2 requires the activation of PKCε. Administration of the TAT vehicle peptide did not produce any change in recovery of HMWKO hearts compared to wildtype (Figure 12).

83

A. *p<0.05 vs. WT 100 Wildtype †p<0.05 vs. HMWKO HMWKO 80 * * † 60

40 (+dP/dt)

20

% Contractile Recovery Recovery Contractile % 0 Untreated TAT vehicle (100nM) PKC epsilon inhibitor (100nM)

B.

100 *p<0.05 vs. WT Wildtype HMWKO 80 * 60 *

40

(-dP/dt) 20

0

% Relaxation at 120 min. reperfusion (-dp/dt) min. 120 reperfusion at % Relaxation Untreated TAT vehicle PKC epsilon (100nM) inhibitor (100nM) % Relaxation% Recovery

Figure 12. The percent recovery of post-ischemic contractility (+dP/dt) and relaxation (-

dP/dt) of wildtype or HMWKO hearts that have undergone 60 minutes ischemia and 120

minutes reperfusion, in the absence or presence of a selective inhibitor of PKCε (εV-1,

100nM) or its vehicle (TAT) . Recovery is expressed as a percentage of baseline function.

The significantly higher postischemic contractility (A) and relaxation (B) seen in HMWKO hearts was significantly reduced after I/R. *p<0.05 vs. wildtype, †p<0.05 vs. HMWKO

untreated or TAT vehicle (n = 5-8 per group).

84

PKCε and α activation in heart only expressing LMW FGF2 are not interdependent.

Since both PKCα and PKCε are required for LMW FGF2 to protect the heart from post- ischemic dysfunction, whether crosstalk occurs between the two PKC isoforms in the signaling cascade of LMW FGF2-induced cardioprotection was investigated. As shown previously (Figure

7), PKCα had an elevated activation at early ischemia, whereas PKCε activation occurred at early reperfusion in hearts with only the presence of the LMW isoform of FGF2. Therefore, it was hypothesized that PKCα may act upstream of PKCε activation in this cardioprotective effect.

To test this, the phosphorylation of PKCε was evaluated in hearts only expressing LMW FGF2, in the presence and absence of PKCα. If PKCα was acting upstream of PKCε, it would be expected that the activation of PKCε would be reduced by the absence of PKCα; however, this was not the case, and in fact, PKCε activation was slightly elevated in the absence of PKCα expression, indicating that PKCα is not upstream of PKCε (Figure 13A). Additionally, to determine if PKCε is upstream of PKCα activation at early ischemia, the phosphorylation of

PKCα was evaluated in the presence of 100nM TAT vehicle or 100nM PKCε inhibitor in

HMWKO hearts. No difference was seen in the phosphorylation of PKCα when PKCε was inhibited (Figure 13B), suggesting that crosstalk between the two isoforms does not contribute to the activation of either during I/R injury.

85

A. *p<0.05 vs. HMWKO 1.4 *

ε 1.2 1 PKCα PKCα 0.8 present KO 0.6 pPKCε 0.4 (arbitrary units) (arbitrary PKCε

Phospho/total PKC Phospho/total 0.2 0 HMWKOHMWKO HMWKOHMWKOxPKCaKO x PKCaKO B. 1.5 TAT PKCε α 1.3 vehicle inhibited 1.1 pPKCα 0.9 0.7 PKCα 0.5

(arbitrary units) (arbitrary 0.3 Phospho/total PKC Phospho/total 0.1 -0.1 HMWKOTAT + HMWKOPKCe inh + TAT vehicle PKCε inhibitor

Figure 13. The amount of phosphorylated PKCε at early reperfusion in hearts only expressing LMW FGF2 in the presence (HMWKO) and absence (HMWKOxPKCαKO) of

PKCα expression (A), and phosphorylated PKCα at early ischemia in hearts only expressing LMW FGF2 in the presence of 100nM TAT vehicle or 100nM PKCε inhibitor

(B). Phospho-PKCε is normalized to total PKCε levels, and phosphor-PKCα is normalized to total PKCα levels. *p<0.05 vs. HMWKO (n=4-8 per group).

86

DISCUSSION I.

The data presented in this dissertation demonstrate that expression of only the LMW

isoform of FGF2 results in differential activation of the α and ε isoforms of PKC during global

low-flow I/R injury (Figure 14). PKCα is activated to a significantly greater extent (p<0.05) in

HMWKO hearts during early ischemia compared to wildtype cohorts, and PKCε is activated to a

significantly greater extent (p<0.05) in HMWKO hearts during early reperfusion. To determine if

these differences in activation translate to the improvement in post-ischemic cardiac function

seen when only LMW FGF2 is expressed, the recovery of hearts either lacking PKCα

expression, or with PKCε activity blocked, was evaluated (Figures 9 and 12). Removing or

inhibiting either PKC isoform resulted in a decrease in the recovery of cardiac function seen after

I/R injury, suggesting that both isoforms contribute to LMW FGF2-mediated cardioprotection

against myocardial dysfunction (Figures 9 and 12).

The increase in PKC activation is determined in these experiments by evaluating the

levels of phosphorylated PKC isoforms via immunoblotting techniques. This is not a direct

measure of PKC activity, which is best determined by in vitro assays (Steinberg, 2008).

Unfortunately, given the relative promiscuity of various PKC isoforms for substrates in vitro, determining the activity of a specific PKC isoform in the heart requires a more selective measure of activation, which is provided by the specificity of antibodies for each isoform of PKC used in immunoblotting. It has been shown that both PKCα and PKCε require phosphorylation of kinases at their autophosphorylation sites for full enzymatic activity (Orr & Newton, 1994;

Parekh, Ziegler, & Parker, 2000; Rybin et al., 2003). Nevertheless, it should be noted that, in some instances, PKC phosphorylation may occur independent of PKC activity (Steinberg, 2008).

A further measure of activation is the translocation of these isoforms from the cytosol to particulate or membrane fractions (R. M. Bell et al., 1986), offering further insight in the

87

selectivity of the activated kinase for its substrate or specific subcellular organelles. The

translocation of these isoforms to specific subcellular regions where PKC targets are expected to

be found will be explored in chapter 2 to confirm these changes in activity.

As indicated in Figure 7, the differences in phosphorylation between HMWKO and

wildtype hearts are significant. Both PKCα and PKCε show about a 20% difference at ischemia

and reperfusion, respectively. This is analogous to changes in PKC phosphorylation or activation

seen in other forms of PKC-dependent cardioprotection (P. Ping et al., 1997), including PKCα- dependent landiolol-mediated protection (Takahashi et al., 2007), demonstrating that a change in activation to this degree is sufficient to mediate an attenuation of I/R.

The data presented here further support the hypothesis that FGF2 activates PKC isoforms

the heart. The PKC isoform activation in hearts only expressing LMW FGF2 mirror that of Jiang

and group (Jiang et al., 2002), who observed PKCα and ε activation following administration of

exogenous FGF2 during I/R injury. Additionally, studies examining the activation of PKCs in

non-ischemic hearts have observed that the addition of exogenous recombinant FGF2 (Padua et

al., 1998) or overexpression of LMW FGF2 (Sheikh et al., 2001) resulted in an increased

translocation of PKCε and PKCα, although changes in the basal activation of these isoforms

when only expressing endogenous LMW FGF2 (Figure 7) were not seen in this dissertation.

Previous work from our laboratory has examined the activation of these PKC isoforms in hearts

overexpressing all human isoforms (low and high molecular protein forms) of FGF2, a model that also resists post-ischemic cardiac dysfunction as well as the development of infarct (House

et al., 2003). PKCε showed an increase in translocation and activation at early reperfusion in that

model (House et al., 2007), similar to that observed in this dissertation. Interestingly, when all

isoforms of FGF2 are overexpressed, PKCα translocation and phosphorylation is reduced during

I/R injury (House et al., 2007), which is different from what is shown here when only the LMW

88

isoform of FGF2 is expressed (Figure 7). This indicates that the activation of PKCα in the heart may be dependent on the isoforms of FGF2 that are being expressed. The differential regulation and activation of PKC isoforms by HMW FGF2 has been established in pancreatic cells, with cells only expressing HMW FGF2 showing altered expression of PKCδ and ε, and altered phosphorylation and translocation efficiency of PKCδ (Gaubert et al., 2001). Further evidence for a role of HMW FGF2 in the regulation of PKCα in the I/R-subjected heart is given in chapter

4 of this dissertation.

The role of PKCα in LMW FGF2-mediated protection from I/R injury is evaluated in this dissertation using a genetically-modified mouse model with prkca gene expression ablated, due to the fact that pharmacological inhibitors that are specific for PKCα are unavailable. Many inhibitors for PKCα have been shown to target other PKCs or kinases; for example, Go6976 has been shown to target PKCβ (Martiny-Baron et al., 1993), and LY333531 may also target PDK1

(Komander et al., 2004). While genetic manipulation, therefore, presents the best solution, there is the risk that this model may have unknown compensatory changes that upregulate other kinases or proteins in the PKCα signaling pathway. To partially address this problem, Braz and colleagues (Braz et al., 2004) examined the levels and translocation efficiencies of other PKC isoforms (β, δ, and ε) in the hearts of these mice and found no changes. In addition, it has been determined that the isoforms of FGF2 are also not altered by PKCα ablation, and the expression levels of PKCδ and ε are not changed by the ablation of both HMW FGF2 and PKCα (Figure 3).

However, the data presented in this dissertation must be interpreted with the caveat that other unknown proteins, such as signaling kinases or downstream targets that modulate cardiac function, that have not yet been evaluated, may impact post-ischemic recovery of cardiac function in these PKCα knockout mice; one such target that has not been investigated yet is inhibitor-1, which has been shown to be regulated by PKCα (Braz et al., 2004).

89

In this dissertation, ablation of PKCα results in complete abrogation of the protective

phenotype in hearts expressing LMW, producing a decrease in post-ischemic function compared

to hearts with LMW and PKCα expression intact. Paradoxically, PKCα inhibition or knockout

has been shown to produce an increase in contractility in mouse models, and overexpression of PKCα results in depressed contractility (Braz et al., 2004; Hambleton et al.,

2006; Q. Liu et al., 2009). The divergent effects of this isoform on cardiac function can be explained as a result of the distinct injury pathways initiated by each stimulus; a number of pathways have similarly been shown to be detrimental in heart failure while beneficial during ischemia (or vice versa), including ERK1/2 (Bueno et al., 2000; Fryer, Pratt, Hsu, & Gross,

2001), CamKII (Anderson, 2009; Xie et al., 2005; Z. Yu, Wang, & Yang, 2009), and TGF-β (W.

Chen & Frangogiannis, 2010; Lefer, Tsao, Aoki, & Palladino, 1990). In the I/R injury model,

PKCα is transiently activated only at the onset of ischemia; whereas, in heart failure PKCα is

chronically activated (Bayer et al., 2003; Bowling et al., 1999; Wang, Liu, Sentex, Takeda, &

Dhalla, 2003), resulting in long-term suppression of SERCA via phospholamban

dephosphorylation (Braz et al., 2004). In fact, a transient phosphorylation of a number of PKCα targets, including troponin (Pyle et al., 2000; Pyle et al., 2003), has been associated with improved function under the unique intracellular conditions initiated by ischemia. This may be due to PKCα depressing the function of the cell during I/R, slowing ATP consumption.

In the investigation of the role of PKCε, a selective and specific translocation inhibitor

has been used, which was developed based off of the highly selective RACK receptor for PKCε, and competitively blocks the PKC binding site to prevent PKCε from binding to its RACK and

translocating to the membrane (Liron, Chen, Khaner, Vallentin, & Mochly-Rosen, 2007). This inhibitor has been shown, and confirmed in this dissertation, to prevent the translocation of

PKCε from cytosol to membrane fractions without altering the translocation efficiencies of other

90

PKC isoforms (Liron et al., 2007); Figure 10 and 11). While this PKCε inhibitor does not block

its substrate- or ATP-binding site, translocation of PKCs in vivo is a critical step in its activation process and preventing translocation has been shown to prevent PKCs from phosphorylating their targets (Liron et al., 2007).

Another important consideration for these experiments is the I/R model used. An isolated

working heart model was determined to be the best option for examining the effects of LMW

FGF2 and downstream signaling proteins on post-ischemic cardiac function, because it allows

myocardial function to be evaluated in the absence of neurohumoral effects. FGF2 and PKCs

both have biological activity in a number of different organ systems, including the central

nervous system and peripheral vasculature, which may confound the effects of gene/protein

ablation or pharmacological inhibition on cardiac function in in vivo mouse models. Also, the

importance of certain hemodynamic parameters at the organ level, such as afterload or preload,

on the autoregulation of cardiac function in the heart in the absence and presence of I/R, allow

the isolated working heart to be a better model for post-ischemic functional injury than isolated

cardiomyocytes (Barr & Lopaschuk, 2000; Vidavalur, Swarnakar, Thirunavukkarasu, Samuel, &

Maulik, 2008).

These studies are the first to demonstrate that the protective effects of LMW FGF2

against post-ischemic cardiac dysfunction correspond to differential activation of PKC isoforms.

Furthermore, this dissertation is the first to demonstrate that endogenous LMW FGF2-mediated

protection requires the activity of PKCα and PKCε, forming the first step in the protective

pathway of a potentially therapeutic molecule.

91

FGFR 1 LMW FGF2

PKCε PKCα

Improved postischemic function

Figure 14. Schematic representing the hypothesized role of PKC isoforms in LMW FGF2- mediated protection from post-ischemic dysfunction. LMW FGF2 signals through both

PKCα and ε to protect the heart from I/R-mediated functional injury.

92

Chapter 2: LMW FGF2 and myofibrillar proteins

RESULTS II.

PKC isoform translocation to the myofibril is altered during I/R injury in hearts only expressing LMW FGF2 compared to wildtype.

To fully understand the mechanism by which LMW FGF2 protects the heart from post- ischemic dysfunction, it is necessary to determine the downstream targets that are phosphorylated by the signaling kinases, PKCα and ε, that mediate this protection. As the protective effects of LMW FGF2 are manifested as changes in contractility and relaxation after

I/R injury, the primary candidates to be tested are proteins that are known to modulate function in the heart, and have been shown to have an effect on the recovery of the heart after I/R injury.

These include contractile proteins located at the myofibril (Arnold M., 1973; Bolli & Marban,

1999; Carrozza et al., 1992; Dong Gao et al., 1996; Engel et al., 2009; Gao et al., 1995; Miller et al., 1996; Pyle et al., 2000; Pyle et al., 2003; Solaro, 1975; Takimoto et al., 2004). To investigate the role of the myofibril in LMW FGF2/PKC-mediated protection from post-ischemic dysfunction, the translocation of PKC isoforms to the myofibril was determined. The isoforms examined were α, ε, and δ, as these isoforms have been previously determined to be activated by

FGF2 in the heart, and play a role in the modulation of cardiac dysfunction (Bouwman et al.,

2004; Budas et al., 2007; C. H. Chen et al., 1999; Churchill & Mochly-Rosen, 2007; G. W. Dorn et al., 1999; Fryer, Wang et al., 2001; Hassouna et al., 2004; House et al., 2003; House et al.,

2007; K. Inagaki et al., 2003; K. Inagaki et al., 2003; Jiang et al., 2002; Kanaya et al., 2003;

Kudo et al., 2002; Lochner et al., 2009; Maslov et al., 2009; Melling et al., 2009; Miki et al.,

1998; Padua et al., 1998; P. Ping et al., 1997; Pyle et al., 2003; Sheikh et al., 2001; Turrell et al.,

2011; Uecker et al., 2003; Wickley et al., 2006; Yoshida et al., 1996; Yoshida et al., 1997; H. Y.

93

Zhang et al., 2002; H. Z. Zhou et al., 2002). Levels of PKCα, δ, and ε in cardiac homogenates

enriched for myofibrillar proteins were evaluated using immunoblotting. All three isoforms demonstrated altered basal levels at the myofibril prior to I/R in hearts that were only expressing

LMW FGF2 (Figure 15). Basally, levels of PKCα were significantly lowered, while levels of

PKCδ and ε were significantly raised in HMWKO hearts compared to wildtype cohorts (p<0.05).

At the onset of ischemia, PKCα translocation to the myofibril significantly increased in hearts only expressing LMW FGF2 (Figure 15A, p<0.05); this corresponds to the whole heart phosphorylation of PKCα during early ischemia (Figure 7, p<0.05). Translocation of PKCδ decreased during early ischemia in hearts only expressing LMW FGF2, suggesting that the activity of this isoform is suppressed at the myofibril at ischemia (Figure 15C, p<0.05). Levels of

PKCε to the myofibril were similar basally as well as during ischemia (Figure 15B). At the onset of reperfusion, however, PKCε translocation to the myofibril increased significantly in hearts only expressing LMW FGF2 compared to wildtype (Figure 15B, p<0.05) ); this corresponds to the whole heart phosphorylation of PKCε during early reperfusion (Figure 7, p<0.05). Similarly,

PKCα translocation to the myofibril was significantly elevated at early reperfusion (Figure 15,

p<0.05). Levels of PKCδ at early reperfusion to the myofibril was significantly less than observed basally (Figure 15C, p<0.05). After two hours of reperfusion, localization of PKCα and

PKCε at the myofibril drops back to basal levels (Figure 15A and B), while levels of PKCδ at the myofibril increase in HMWKO hearts compared to wildtype or its basal cohorts (Figure 15C, p<0.05).

94

*p<0.05 vs. WT PKCα A. †p<0.05 vs. Sham †* 2 1.8 WT 1.6 † 1.4 HMWKO 1.2 1 0.8 0.6 * 0.4 0.2 0 ShamSham 5’I5'I 60’I60'I + +5'R 5’R 60’I60'I + 120'R 120’R PKC (arbritraryPKC units) Myofilament associated

PKCε B.

† 4 WT * 3.5 HMWKO 3 * 2.5 * 2 1.5 1 0.5 0 ShamSham 5’I5'I 60’I60'I + +5'R 5’R 60’I 60'I+ + 120’R120'R PKC (arbritraryPKC units) Myofilament associated

PKCδ C. 3 † WT * 2.5 HMWKO

2 * † † 1.5

1

0.5

0 ShamSham 5’I5'I 60’I60'I + +5'R 5’R 60’I 60'I+ + 120’R120'R PKC (arbritraryPKC units)

Myofilament associated

Figure 15. Translocation of PKC isoforms is altered in LMW FGF2 hearts (striped bar) compared to wildtype (black bar). Localization of A) PKCα, B) PKC δ, and C) PKCε at baseline (sham), early ischemia (5 minutes ischemia), early reperfusion (60 minutes ischemia + 5 minutes reperfusion) and late reperfusion (60 minutes ischemia + 120 minutes reperfusion). *p<0.05 vs. wildtype, †p<0.05 vs. sham (n=3-16 per group).

95

Troponin phosphorylation is altered in myofibrils of hearts only expressing LMW FGF2

during ischemia.

Given that PKCs translocated differentially to the contractile apparatus during I/R in hearts that only express LMW FGF2, it was next evaluated whether troponin I and T showed

differences in phosphorylation during these same time points, as these are known targets of PKC

isoforms at the myofibril (W. J. Dong et al., 1997; Engel et al., 2009; Finley et al., 1999; Kanaya

et al., 2003; Kooij et al., 2009; M. X. Li et al., 2003; Molnar et al., 2009; Noland et al., 1989;

Noland et al., 1995; Noland et al., 1996; Pyle et al., 2003; Ramirez-Correa, Cortassa, Stanley,

Gao, & Murphy, 2010; Shaw et al., 2009; Sumandea et al., 2003; Walker, Walker, Ambler, &

Buttrick, 2010); therefore, the phosphorylation state of troponin I and T was evaluated at

baseline, early ischemia, early reperfusion, and late reperfusion, to determine if changes in

troponin regulation correspond to the protection from post-ischemic dysfunction seen when only

LMW FGF2 is expressed. In wildtype hearts, troponin T phosphorylation was unchanged

throughout I/R, while troponin I showed a gradual decrease in phosphorylation (Figure 16). In

hearts only expressing LMW FGF2, both troponin I and T show a significant increase in

phosphorylation at early ischemia, which remains through early reperfusion (Figure 16, p<0.05).

96

A. Troponin I 1.8 *p<0.05 vs. WT 1.6 * WT 1.4 HMWKO 1.2 1 * 0.8 0.6 (arbritrary units) 0.4

Phosphoroylated troponin 0.2 0 ShamSham 5’I5'I 60’I60'I + +5'R 5’R 60’I60'I + + 120'R120’R

B. Troponin T 2.5 † *p<0.05 vs. WT WT * 2 † p<0.05 vs. sham HMWKO 1.5 * * 1 (arbritrary units) 0.5 Phosphoroylated troponin 0 ShamSham 5’I5'I 60’I +60'I 5’R +5'R 60’I60'I + 120'R 120’R

Figure 16. The ratio of phosphorylated/total troponin I (A) and troponin T (B) in hearts only expressing LMW FGF2 at baseline (sham), early ischemia (5 minutes ischemia), early reperfusion (60 minutes ischemia + 5 minutes reperfusion) and late reperfusion (60 minutes ischemia + 120 minutes reperfusion). Phosphorylation of both troponin I and T were increased in hearts expressing LMW (HMWKO, striped bar) compared to wildtype (black bar) at early ischemia and early reperfusion, and troponin T remained phosphorylated to a higher degree at late reperfusion. *p<0.05 vs. wildtype, †p<0.05 vs. sham (n=3-16 per group).

97

Changes in troponin phosphorylation in myofibrils of hearts only expressing LMW FGF2

is abrogated in the absence of PKCα

Given that PKCα shows increased activation and localization to the myofibril during

early ischemia (Figures 7 and 15), at the same time point that troponins first show increased

phosphorylation (Figure 16), troponin phosphorylation at this time point was assessed in the

absence of PKCα to determine if these post-translational modifications were dependent on PKC

activity. The troponin isoforms T and I were investigated, as these are known targets of PKC that

modulate the function of the heart (Cole & Perry, 1975; W. J. Dong et al., 1997; Engel et al.,

2009; Filatov et al., 1999; Finley et al., 1999; Kooij et al., 2009; M. X. Li et al., 2003; Noland et

al., 1989; Noland & Kuo, 1993; Noland et al., 1995; Ramirez-Correa et al., 2010; Sumandea,

Burkart, Kobayashi, De Tombe, & Solaro, 2004), and have been implicated in I/R injury as well

as cardioprotection (K. Inagaki et al., 2003; Kanaya et al., 2003; Pyle et al., 2000; Pyle et al.,

2003; Sawicki et al., 2005; Walker et al., 2010). Hearts with PKCα ablated were evaluated for

troponin phosphorylation in wildtype hearts or those expressing only the LMW isoform of FGF2.

Ablation of PKCα in hearts expressing only LMW FGF2 reduced both troponin I and troponin T

phosphorylation to sham levels at early ischemia (Figure 17, p<0.05), suggesting that the increase in phosphorylation of both of these proteins involves PKCα signaling.

98

A. Troponin I

2 WT *p<0.05 vs. WT * HMWKO 1.5 DKO

1

0.5 (arbritrary units)

0 Sham 5'I Phosphoroylated troponin Sham 5’I

B. Troponin T 2.5 * † WT HMWKO 2 *p<0.05 vs. WT † p<0.05 vs. sham DKO 1.5

1 (arbritrary units) 0.5 Phosphoroylated troponin 0 ShamSham 5’I5'I

Figure 17. The ratio of phosphorylated/total troponin I (A) and troponin T (B) in wildtype hearts (black bar), hearts only expressing LMW FGF2 (striped bar), or hearts only expressing LMW FGF2 with PKCα expression ablated (white bar) at baseline (sham), early ischemia (5 minutes ischemia) in the presence or absence of PKCα. Increased hosphorylation of both troponin I and T in HMWKO hearts compared to wildtype at early ischemia was ablated by the absence of PKCα. *p<0.05 vs. wildtype, †p<0.05 vs. sham

(n=3-16 per group).

99

Actomyosin ATPase activity is altered during I/R injury in myofibrils of hearts only

expressing LMW FGF2.

As changes were seen in the phosphorylation of troponin I and T during I/R in hearts only

expressing LMW FGF2, it was next investigated whether these changes resulted in alterations of

the activity of actomyosin ATPase, which is modulated by troponin I and T phosphorylation

(Greene & Eisenberg, 1988; Noland & Kuo, 1993; Noland et al., 1996; Pyle et al., 2003). This

question is particularly relevant towards elucidating the cardioprotective mechanisms of LMW

FGF2, as lowering the activity of actomyosin ATPase during ischemia has been shown to protect

the heart from injury (Kanaya et al., 2003; Pyle et al., 2000; Pyle et al., 2001; Pyle et al., 2003).

To determine if expression of only LMW FGF2 results in alterations of the actomyosin ATPase

during I/R injury leading to the observed cardioprotective phenotype, activity was evaluated in

HMWKO and WT hearts at baseline, early ischemia, early reperfusion, and late reperfusion.

Maximal activity of actoymyosin ATPase was significantly higher in HMWKO hearts during reperfusion compared to wildtype (Table 3, p<0.05). In addition, hearts only expressing the

LMW isoform of FGF2 show markedly elevated EC50 (the concentration of calcium required to

elicit 50% of the maximal response) at baseline, early ischemia, and early reperfusion; at these

time points (Table 3), higher amounts of calcium are necessary to achieve half of the maximum

response, indicating a loss of calcium sensitivity in the HMWKO compared to wildtype hearts.

This data provides evidence that expression of only LMW FGF2 affects the response of the

myofibril to changes in calcium concentrations before and during I/R injury. Additionally, no

significant difference was seen in the Hill coefficient, which is a measure of the of

calcium binding and activation of actoymyosin ATPase, suggesting that expression of only

LMW FGF2 does not affect cooperativity of calcium activation of actomyosin ATPase.

100

60’I + Sham 5'I 60’I +5'R 120'R WT HMWKO WT HMWKO WT HMWKO WT HMWKO Maximum (nM Pi/min/mg protein) 271±10 280±12 284±10 285±26 236±12 *288±14 290±13 285±11 Sub-maximal (nM Pi/min/mg protein) 176±10 156±9 186±12 155±17 160±6 165±8 190±9 188±12 EC50 (nM Calcium) 0.67±0.05 *0.81±0.05 0.73±0.04 *0.92±0.07 0.58±0.08 *0.8±0.03 0.7±0.03 0.67±0.05 1.88±0.1

Table 3. Actomyosin ATPase activity, maximum efficacy, EC50, and Hill coefficient in

wildtype hearts and hearts only expressing the LMW isoform of FGF2 at baseline (sham),

early ischemia (5’I), early reperfusion (60’I + 5’R) and late reperfusion (60’I + 120’R).

Significant differences were seen in the EC50 at baseline, early ischemia, and early

reperfusion, and a significant difference in maximal activity was seen at early reperfusion.

*p<0.05 vs wildtype (n=3-16 per group).

101

Actomyosin ATPase activity in myofibrils of hearts only expressing LMW FGF2 is altered in the absence of PKCα expression but not in the absence of PKCε activity

As PKC isoforms are phosphorylated (Figure 7) and induced to translocate to the myofibril at the onset of ischemia (Figure 15), and are required for the protection from post- ischemic dysfunction (Figures 9 and 12), it was determined whether the changes in actomyosin

ATPase seen during I/R in the presence of only LMW FGF2 expression are mediated by PKCα or ε. The significant increase in the EC50 of actomyosin ATPase calcium-induced activity seen during baseline and ischemia, but not early reperfusion, in the HMWKO hearts (Table 3) was inhibited in the absence of PKCα (Figure 18F). Inhibition of PKCε had no effect on calcium sensitivity (EC50) at early reperfusion, the timepoint when PKCε was activated and translocated to the myofibril (Figure 19F). The increase in the maximum activity of actomyosin ATPase at early reperfusion in hearts only expressing LMW FGF2 was not changed by PKCα ablation or

PKCε inhibition, suggesting that this increase is not dependent on either isoform (Figure 18D and 19D). There was also a significant increase in the Hill coefficient of calcium-activated actomyosin ATPase in the presence of PKCε inhibition (Figure 19E) that was independent of genotype, suggesting that PKCε affects the cooperativity of calcium binding and activation of actomyosin ATPase independent of FGF2. This data indicates that the decreased sensitivity to calcium in hearts only expressing LMW FGF2 during baseline and early ischemia requires the presence of PKCα, but not PKCε.

102 A. Sham B. 5’ I C. 60’ I + 5’ R

ein) ein) ein) activity (nM activity activity (nM activity activity (nM activity Pi/min/mg prot Pi/min/mg prot Pi/min/mg prot coysnATPase Actomyosin coysnATPase Actomyosin coysnATPase Actomyosin

Free Calcium (μM) Free Calcium (μM) Free Calcium (μM) D. Maximum Efficacy WT 320 * HKO 300 DKO

280 *p<0.05 vs. WT 260 †p<0.05 vs. 240 HMWKO pPi/min/mg protein)c 220

Actomyosin ATPase activity (nM 200 ShamSham 5’I 5'I 60’I +5'R 5’R EC F. 50 E. Hill Coefficient 1.2 WT 2.5 WT HKO HKO 1 * DKO DKO 2 * † 0.8 * † 1.5 0.6 units) mMCalcium 1 0.4

0.5 Hill coefficient (arbitrary (arbitrary coefficient Hill 0.2

0 0 ShamSham 5’I5'I 60’I5'R + 5’R ShamSham 5’I5'I 60’I5'R + 5’R

Figure 18. Actomyosin ATPase activity (A-C), maximum efficacy (D), Hill coefficient (E),

and EC50 (F) in wildtype hearts (black bar) and hearts only expressing the LMW isoform

of FGF2, in the presence (striped bar) and absence (white bar) of PKCα. The significant

increase seen in the EC50 of HMWKO hearts at baseline, early ischemia was abrogated by

the ablation of PKCα, while there was no change in the maximum activity seen at early reperfusion in only LMW FGF2-expressing hearts compared to wildtype when PKCa was ablated. *p<0.05 vs. wildtype (n=3-16 per group).

103

A. Sham B. 60’ I + 5’ R Pi/min/mg protein) Pi/min/mg protein) Pi/min/mg Actomyosin ATPase activity (nM activity ATPase Actomyosin Actomyosin ATPase activity (nM activity ATPase Actomyosin

Free Calcium (μM) C. Free Calcium (μM) Maximum Efficacy WT 350 HKO * WT e inh 300 HKO e inh *p<0.05 vs. WT 250

pPi/min/mg protein) 200 ShamSham 60’I +5'R 5’R Actomyosin ATPase activity (nM

Hill Coefficient EC 50 D. WT E. 1 * WT 3 * HKO * * HKO 2.5 * WT e inh 0.8 WT e inh HKO e inh HKO e inh 2 0.6 1.5 0.4 units)

1 mMCalcium 0.5 0.2 Hill coefficient (arbitrary 0 0 ShamSham 60’I +5'R 5’R ShamSham 60’I +5'R 5’R

Figure 19. Actomyosin ATPase activity (A-C), maximum efficacy (D), Hill coefficient (E), and EC50 (F) in untreated wildtype hearts (black bar and dotted bar) and hearts only

expressing the LMW isoform of FGF2 (shaded bar and white bar), and in the presence of

100 nM TAT vehicle or 100 nM PKCε inhibitor. There were no changes in the maximum

efficacy or EC50 in the presence of the inhibitor. There was a significant increase in the Hill

coefficient of calcium-activated actomyosin ATPase in the presence of the inhibitor that

was independent of genotype. *p<0.05 vs. wildtype (n=3-16 per group).

104

DISCUSSION II.

The studies presented in this dissertation give evidence that PKCα, activated in the

presence of only LMW FGF2 expression, targets to myofibrillar proteins during ischemia

(Figure 20). Expression of only LMW FGF2 results in movement of PKCα and ε to the

myofibril, and movement of PKCδ away from the myofibril, during I/R injury (Figure 15).

Troponin I and T both show increased phosphorylation at early ischemia (Figure 16), when

PKCα in hearts expressing LMW first initiates movement to the myofibril (Figure 15A).

Furthermore, this phosphorylation state of troponin I and T was significantly decreased in the absence of PKCα expression, but not PKCε inhibition (Figure 16). Finally, these dissertation

results demonstrate that the expression of only LMW FGF2 enhances maxiumum actomyosin

ATPase activity during early reperfusion, as well as decreased calcium sensitivity before and

during I/R injury (Table 3), and that this decreased sensitivity is restored by the absence of PKCα expression, but not the inhibition of PKCε (Figures 18 and 19).

It is interesting to note that the change in localization (translocation) of PKC isoforms to

the myofibril during I/R injury mirrors the pattern of phosphorylation in the whole heart,

suggesting that PKCα and ε are becoming activated and moving to the myofibril at various

points during ischemia and reperfusion in the presence of only LMW FGF2. PKCα moves to the

myofibril at early ischemia (Figure 15A), and this timeframe is similar to that observed when the

phosphorylation of PKCα is increased in the HMWKO hearts compared to wildtype (Figure 7A), which peaks shortly after the onset of ischemia. These data support the supposition that the expression of only LMW FGF2 results in increased PKCα activity at the myofibril rapidly after ischemia is induced, and is consistent with the idea that PKCα is an early, and therefore potentially, a key regulator of signal transduction in HMWKO hearts during I/R. Similarly, the change in PKCε localization at the myofibril in hearts only expressing LMW FGF2 also occurs

105

at early reperfusion (Figure 15B), the same timeframe as the increase in PKCε phosphorylation was observed (Figure 7B), once again suggesting that LMW FGF2 expression results in increased activity of this PKC isoform at the myofibril shortly after the restoration of blood flow to the heart. These data together support the hypothesis that LMW FGF2-activated PKC isoforms target myofilament proteins during I/R.

However, an important consideration to be addressed is the difference seen in the basal

levels of the PKC isoforms at the myofibril in hearts only expressing LMW FGF2. While the

translocation of PKC isoforms to the contractile apparatus during I/R agree with the

phosphorylation data, the levels of PKC at the myofibril prior to the onset of ischemia indicate

that other roles for these isoforms cannot be ruled out. For example, PKCα, while showing an

increase in overall movement to the myofibril during I/R, is present at lower levels at the

myofibril in non-ischemic hearts. It may be, therefore, that the movement of PKCα to the

myofibril is simply restoring PKCα to levels observed in the wildtype during ischemia, and is an

effect, not a cause, of LMW FGF2-mediated cardioprotective signal transduction. Alternately, it

may be that lowered levels of PKCα at the myofibril, together with higher levels of PKCε and δ,

may be acting to precondition the heart, by modulating the basal phosphorylation of unknown

proteins that have an impact on the function of the heart during or after ischemia. This concern is

somewhat mitigated by the fact that complete ablation of PKCα removes, not enhances, LMW

FGF2-mediated protection against ischemia-induced cardiac dysfunction; loss of PKCα

expression would be expected to improve post-ischemic myocardial function, if it were the case

that loss of PKCα activity at the myofibril is cardioprotective. In addition, known

cardioprotective targets of PKC isoforms, troponin I and T (Engel et al., 2009; Pyle et al., 2001;

Pyle et al., 2003), show increased PKC-dependent phosphorylation only on the onset of

ischemia, suggesting that a preconditioned state is not achieved at these myofibrillar proteins.

106

However, as all regulatable myofilament proteins are not evaluated in these studies, it is possible

that lowered basal levels of PKCα, or elevated basal levels of PKCε and δ, may affect other

contractile proteins that are responsible for LMW FGF2-mediated cardioprotection. These

include ’s PEVK region (Hidalgo et al., 2009), myosin binding protein C (Kooij et al., 2009;

Walker et al., 2010), myosin light chain 2 (Kanaya et al., 2003), and cardiac actin capping

protein (Hartman, Martin, Solaro, Samarel, & Russell, 2009), as well as those that have not yet

been discovered.

An interesting observation that has not been explored in this dissertation is the

localization of PKCδ in LMW FGF2-mediated protection from post-ischemic cardiac dysfunction. While PKCδ is not activated differentially in hearts only expressing LMW FGF2, this isoform is present at higher levels at the myofibril in non-ischemic hearts and translocates

away from the myofibril at the onset of ischemia (Figure 15C). Finally, this isoform localizes to

the myofibril to a much higher degree in HMWKO hearts compared to wildtype at late

reperfusion (Figure 15C). The role of PKCδ in I/R injury and cardioprotection has been shown to be a complex one, acting both as a mediator of injury and as a cardioprotective molecule when activated by different stimuli (Churchill & Mochly-Rosen, 2007; Fryer, Wang et al., 2001; K.

Inagaki et al., 2003; Kudo et al., 2002). Given that PKCδ may phosphorylate troponin I at S22/23

(Ramirez-Correa et al., 2010), further study of the presence or absence of this isoform at the myofibril may shed additional light onto the means by which LMW FGF2 fine-tunes troponin regulation of contractility.

This dissertation demonstrates that hearts only expressing LMW FGF2 show increased

phosphorylation of both troponin I and T and alter actomyosin ATPase activity, decreasing the

EC50 of calcium-activated ATPase prior to and during ischemia, and increasing the maximum

activity of the ATPase soon after reperfusion (Figure 16 and Table 3). These effects can be

107

expected to protect the heart from the development of I/R injury. As the heart undergoes ischemia, injury is precipitated by a loss of ATP and an increase in average intracellular calcium concentration (Grinwald, 1982). This increase in calcium, among other things, causes further activation of actomyosin ATPase, resulting in a feedback mechanism that further reduces intracellular ATP stores (Grinwald & Nayler, 1981); it has been shown that by simply removing extracellular calcium, ATP depletion and ischemic injury are slowed (Shine & Douglas, 1983).

Additionally, slowing the rate of actomyosin ATPase activity has also been shown to protect the heart from ischemic injury (Pyle et al., 2003). By reducing the EC50 of the actomyosin ATPase activity, more calcium is necessary to produce the same degree of activity. This would suggest that, in hearts only expressing LMW FGF2, the increase in intracellular calcium seen during ischemia would result in a dampened ATP activation, and would be expected to slow ATP consumption and therefore mitigate I/R injury. In fact, slowed depletion of intracellular ATP during I/R was seen in rat hearts exposed to exogenously applied FGF2 (Jiang et al., 2002), which is expected to have similar effects on the cardiomyocyte as endogenously produced LMW

FGF2. Alternatively, after the onset of reperfusion, oxygen and nutrients are restored to the heart and ATP depletion is no longer a concern. At this point, increasing the activity of the actomyosin

ATPase is expected to improve contractility, reversing lowered post-ischemic cardiac dysfunction and mitigating myocardial stunning.

It should be noted that the actomyosin ATPase studies presented here are conducted on skinned myofibrils. While this allows for evaluation of the myofilament outside of the influences of other intracellular structures, it removes the cytosolic environment that may play a role in regulating actomyosin ATPase activity in vivo. For example, during ischemia, the pH of the cytosol is lowered, which has been demonstrated to have a dampening effect on actomyosin

ATPase activity (Pyle et al., 2003). The effects of using this method have been minimized by

108

comparing treated and untreated hearts, and genetically-modified and wildtype hearts, under similar conditions, although differential effects of altered cytosolic environment on each group may occur.

The data presented in this dissertation provides a link between LMW FGF2, PKCα, and the alteration of myofilament proteins that are responsible for regulating contractility and ATP consumption at the myofibril. Alteration of the phosphorylation of troponin I and T and the activity of actomyosin ATPase in the presence of LMW FGF2 in a manner that is dependent on the presence of PKCα give evidence for a mechanism of protection from post-ischemic dysfunction that relies on modulation of myofilament proteins in the ischemic heart.

109

FGFR 1 LMW FGF2

PKCε PKCα

Myofibril

Calcium sensitivity during ischemia

Improved postischemic function

Figure 20. Schematic representing the hypothesized role of PKC α and ε at the myofibril in

LMW FGF2-mediated protection from post-ischemic dysfunction. LMW FGF2 activates

PKCα, which phosphorylated troponin I and T at the myofibril and reduces calcium sensitivity (EC50), to protect the heart during ischemia.

110

Chapter 3: LMW FGF2 and calcium handling proteins

RESULTS III.

Cardiomyocytes only expressing LMW FGF2 have depressed calcium cycling.

The improvement in post-ischemic function seen when only LMW FGF2 is expressed is associated with changes in the phosphorylation of troponin I and T at the myofibril, as demonstrated in chapter 2 of this dissertation. However, other proteins may also play a role in protecting the heart from post-ischemic cardiac dysfunction. Among these are proteins that modulate the cycling of calcium in the myocyte, which have been shown to have a significant effect on the function of the heart after I/R injury (Akita et al., 1993; Kimura-Kurosawa et al.,

2007; Mattiazzi et al., 2004; Mattiazzi et al., 2006; Meldrum et al., 1996; Said et al., 2003;

Temsah et al., 2002; Thandroyen et al., 1988; Valverde et al., 2006; Zucchi et al., 1995). This is a particularly important question to address, since FGF2 has been shown to produce its biological effects via the modulation of calcium in a number of cell types (Boxer, Moreno, Rudy, & Ziff,

1999; Browaeys-Poly, Cailliau, & Vilain, 1998; Distasi, Torre, Antoniotti, Munaron, &

Lovisolo, 1998; El Idrissi & Trenkner, 1999; Kessler, Budde, Gekle, Fabian, & Schwab, 2008;

Kuhlmann et al., 2004; X. Liu, Wu, Cai, & Sun, 2008; Lynch et al., 2000; Malo, Browaeys-Poly,

Fournier, Cailliau, & Vilain, 1997; Merle, Usson, Robert-Nicoud, & Verdetti, 1997; Miyamoto,

Leconte, Swain, & Fox, 1998; Munaron & Fiorio Pla, 2000; D. Y. Oh et al., 2008; Peluso et al.,

2001; Peluso, 2003; Qu & Zhang, 2004; Samain et al., 2000; Wiecha et al., 1998; Yagami et al.,

2010). To examine the effect of LMW FGF2 on calcium cycling in the cardiomyocyte, cardiomyocytes from mice only expressing the LMW isoform of FGF2 were isolated, and myocyte contractility and calcium cycling was evaluated. Sarcoplasmic load was also evaluated by measuring the increase in calcium concentration on exposure to caffeine. Cardiomyoctes only

111

expressing LMW FGF2 demonstrated significantly decreased calcium transients compared to cells isolated from wildtype mice (Figure 21, p<0.05), as well as significantly decreased caffeine-induced calcium release, indicative of a decrease in SR calcium load (Figure 21, p<0.05). Additionally, there was a trend towards decreased contractility (+dL/dt) and relaxation

(-dL/dt) in HMWKO hearts compared to wildtype (Figure 22), although no alterations in fractional shortening (the % reduction in length of the contracting myocyte) were observed. This suggests that expression of only LMW FGF2 has a depressive effect on basal calcium cycling in cardiomyocytes.

112

Basal Calcium Transient A. Amplitude 340/380 Transient Amplitude

0.36 *p<0.05 vs. WT 0.34 0.32 0.3 0.28 * 0.26 0.24 0.22 340/380 Ratio (arbitrary units) (arbitrary Ratio 340/380 0.2 WT HMWKO

340/380 Caffeine Peak Amplitude B. Basal SR Calcium 0.55 Load 0.5 0.45 0.4 * 0.35

0.3 0.25

340/380 Ratio (arbitrary units) (arbitrary Ratio 340/380 0.2 WT HMWKO

Figure 21. Changes in the calcium levels of isolated myocytes from wildtype hearts (black bar) or hearts only expressing LMW FGF2 (striped bar). Calcium concentration is indicated as the ratio of fura-2 fluorescence at 340nm to 380nm, and levels are expressed as arbitrary units. Lowered average changes in calcium concentrations from beat-to-beat

(calcium transients) were seen in hearts only expressing LMW FGF2 (A), as well as lowered calcium peak on exposure to caffeine, an indication of the total SR calcium load

(B). p<0.05 (n=10-15 cells per heart, 3 hearts per group).

113

Contractility A. and Relaxation WT 80 HMWKO 70 60 50 40

uM /sec 30 20 10 0 pdL/dt NdL/dt

B. Fractional Shortening 7 WT HMWKO 6 5 4 3 % baseline 2 1

0 WT HMWKO

Figure 22. The function of isolated myocytes from wildtype hearts (black bar) or hearts only expressing LMW FGF2 (striped bar). Contractility is expressed as the pdP/dt, and relaxation is expressed as NdP/dt. There was no significant change in contractility or relaxation (A) or fractional shortening (B) between wildtype and HMWKO hearts. (n=10-

15 cells per heart, 3 hearts per group).

114

Hearts only expressing LMW FGF2 do not have altered expression of candidate calcium- handling proteins.

To determine if these changes in calcium cycling seen when only LMW FGF2 is expressed are due to the regulation of the levels of calcium handling proteins, basal levels of calcium handling proteins were evaluated in hearts with only the LMW isoform of FGF2. Levels of sarco(endo)plasmic reticulum calcium ATPase (SERCA), calsequestrin, and phospholamban were evaluated in non-ischemic HMWKO (LMW expressed), FGF2 knockout (FGF2 KO, where all FGF2 protein isoforms are absent) and wildtype hearts. No differences in the levels of these proteins were seen in HMWKO mice compared to wildtype or FGF2 knockout mice (Figure 23), indicating that LMW FGF2 does not regulate expression of these calcium handling/cycling proteins.

115

A. SERCA 2.5 WT FGFKO 2 HMWKO SERCA 1.5 GAPDH 1 0.5 to (AU) GAPDH

SERCA, normalizedSERCA, 0 WTWT FGFKOFGFKO HMWKOHMWKO

B. Phospholamban WT

1 FGFKO HMWKO 0.8 PLB 0.6 GAPDH 0.4 0.2 to (AU) GAPDH 0

Total PLB, normalized normalized PLB, Total WTWT FGFKOFGFKO HMWKOHMWKO C. 1.8 Calsequestrin 1.6 1.4 1.2 WT FGFKO HMWKO 1 CSQ 0.8 0.6 GAPDH

to GAPDH (AU) 0.4 0.2 Calsequestrin Normalized to GAPDH (AU) GAPDH to Normalized Calsequestrin 0

Calsequestrin, normalized WTWT FGFKO HMWKO FGFKO HMWKO

Figure 23. Levels of calcium handling proteins in non-ischemic wildtype hearts (WT, black

bar), hearts with FGF2 expression ablated (FGFKO, white bar) and hearts only expressing

LMW FGF2 (HMWKO, striped bar). No significant changes were observed in the

expression of SERCA (A), phospholamban (B), or calsequestrin (C). Hearts were probed

for SERCA, phospholamban, or calsequestrin via immunoblotting and normalized to

GAPDH (n=6-9 per group).

116

Phospholamban phosphorylation during I/R is higher in hearts expressing LMW FGF2.

Since the effects of LMW FGF2 are observed acutely after the onset of reperfusion, it

was hypothesized that these actions may be related to the post-translational modifications of

phospholamban during I/R injury. The phosphorylation of phospholamban at threonine-17 was

examined due to the importance of this site in the development of ischemic injury in the heart

(Mattiazzi et al., 2004; Mattiazzi et al., 2006; Mundiña-Weilenmann et al., 2005) and its role as a

target for indirect regulation by PKC isoforms (Oestreich et al., 2009; Z. Yu, Wang, & Yang,

August 2009). In hearts only expressing the LMW isoform of FGF2, there were no differences

in basal levels of phospholamban phosphorylation at threonine-17 (Figure 24). However, levels

of phosphorylated Thr17 phospholamban were significantly elevated at early ischemia and early

reperfusion compared to FGF2 knockout hearts (Figure 24, p<0.05).

117

pThr17 PLB/total PLB

) 3.5 * WT units 3

y *p<0.05 vs. FGFKO FGFKO 2.5 HMWKO arbitrar ( 2 * 1.5

1 ho/Total PLB ho/Total

p 0.5

Phos 0 Baseline Early Ischemia (5 min) Early Reperfusion (60 Late Reperfusion (60 min min isch + 5 min rep) isch + 120 min rep) Time point

Figure 24. Phospholamban phosphorylation at baseline (sham), early ischemia (5 minutes ischemia), early reperfusion (60 minutes ischemia + 5 minutes reperfusion) and late reperfusion (60 minutes ischemia + 120 minutes reperfusion). Phosphorylation was measured as the ratio of phosphorylated phospholamban to total phospholamban in wildtype hearts (circle with solid line), hearts only expressing LMW FGF2 (HMWKO, triangle with dashed line), or hearts with expression of all isoforms of FGF2 ablated (FGF2

KO, square with dotted line). PLB phosphorylation significantly increased in WT and

HMWKO hearts at early ischemia compared to FGF2 KO hearts. *p<0.05 vs. FGF2 KO

(n=6-9 per group).

118

Phospholamban phosphorylation during I/R in hearts only expressing LMW FGF2 is not altered in the absence of PKCα

Since both the phosphorylation of phospholamban and the activation of PKCα increases at early ischemia, it was next examined whether this increase in the phosphorylation of phospholamban at threonine-17 observed at early ischemia or early reperfusion was dependent on the presence of PKCα in hearts only expressing LMW FGF2. The level of phosphorylated phospholamban was evaluated in HMWKO hearts in the absence and presence of PKCα at early ischemia and early reperfusion. There was no significant difference in the ratio of phosphorylated phospholamban to total phospholamban when PKCα was ablated (p<0.05, Figure

25), suggesting that PKCα is not involved in the phosphorylation of phospholamban at threonine-17 during I/R.

119

HM WKO )

ts HM WKOxPKCaKO

ni 1.2 u

y r

a 1 r t i b

r 0.8 (a 0.6 l P L B

a 0.4 ot

T 0.2 o/ h

p 0 os 5'I 5'R Ph

Figure 25. Phospholamban phosphorylation during ischemia and reperfusion in hearts expressing only LMW FGF2 at early ischemia (5 minutes ischemia), early reperfusion (60 minutes ischemia + 5 minutes reperfusion), in the absence and presence of PKCα.

Phosphorylation was measured as the ratio of phosphorylated phospholamban to total phospholamban. n=4-6 hearts per group

120

CamKII phosphorylation during ischemia is higher in hearts only expressing LMW FGF2

compared to wildtype

As the threonine-17 residue of PLB is known to be phosphorylated by CamKII, it was hypothesized that the difference in phosphorylation at this site during early ischemia was due to an increase in CamKII activation (i.e., phosphorylation) in hearts only expressing LMW FGF2.

In wildtype hearts, levels of phosphorylated CamKII drop significantly at early ischemia

(p<0.05, Figure 26). In hearts only expressing LMW FGF2, there was a significantly higher ratio of phosphorylated/total CamKII, compared to wildtype at this same time point which corresponds to the time point at which higher levels of phosphorylated Thr17PLB were seen as

well (p<0.05, Figure 24).

121

60’I+ 60’I+ Sham 5’I 5’R 120’R WT WT HMWKO WT WT HMWKO HMWKO HMWKO Phospho-CaMKIIa CaMKIIa

Phospho/Total CamKIIa

1.2

1 HMWKO WT s 0.8 *

0.6 †

0.4 Arbitrary Unit Arbitrary *p<0.05 vs. WT 0.2 †p<0.05 vs. Sham

0 Sham 5'I 5'R 120'R

Figure 26. CamKII phosphorylation in hearts expressing only LMW FGF2 (HMWKO, diamond with dashed line) and wildtype (WT, square with solid line) hearts at baseline

(sham), early ischemia (5 minutes ischemia), early reperfusion (60 minutes ischemia + 5 minutes reperfusion) and late reperfusion (60 minutes ischemia + 120 minutes reperfusion).

Phosphorylation was measured as the ratio of phosphorylated CamKII to total CamKII.

CamKII showed a higher degree of phosphorylation at early ischemia in HMWKO hearts compared to wildtype. *p<0.05 vs. WT (n=5-10 per group).

122

DISCUSSION III.

The data presented in this dissertation demonstrate that LMW FGF2 directly modulates calcium cycling in non-ischemic cardiomyocytes. These differences do not appear to be the result of basal changes in candidate calcium handling proteins, including SERCA, phospholamban, or calsequestrin (Figure 23). It was found, however, that during ischemia and reperfusion, the phosphorylation of phospholamban at threonine-17 was increased in hearts expressing LMW FGF2 compared to hearts not expressing FGF2 (Figure 25), suggesting that

LMW FGF2 may modulate calcium cycling during I/R injury by altering the phosphorylation state of phospholamban, and subsequent regulation of SERCA (Figure 27). It was found that, although PKCα is activated simultaneously to this increase in phospholamban phosphorylation, the increase in phosphorylation is not dependent on the presence of PKCα (Figure 25).

It has been well established that FGF2 can modulate calcium-handling proteins in a number of cell types (Boxer et al., 1999; Browaeys-Poly et al., 1998; Distasi et al., 1998; El

Idrissi & Trenkner, 1999; Kessler et al., 2008; Kuhlmann et al., 2004; X. Liu et al., 2008; Lynch et al., 2000; Malo et al., 1997; Merle et al., 1997; Miyamoto et al., 1998; Munaron & Fiorio Pla,

2000; D. Y. Oh et al., 2008; Peluso et al., 2001; Peluso, 2003; Qu & Zhang, 2004; Samain et al.,

2000; Wiecha et al., 1998; Yagami et al., 2010) but these are the first studies to directly indicate that endogenously produced LMW FGF2 reduces calcium cycling in the cardiomyocyte. This finding allows the possibility that one of the means by which LMW FGF2 may protect the heart from post-ischemic cardiac dysfunction is the modulation of intracellular calcium concentrations.

No significant difference in basal contractility was observed between the groups (Figure 22), confirming previous findings from our laboratory (Liao et al., 2010), suggesting that this change in calcium transient amplitude and SR calcium load does not translate to a change in function.

There are multiple reasons why this may be the case: first, the difference in peak calcium may

123

not be sufficient to result in a functional change in non-ischemic heart cells. A non-significant

trend towards a decrease in contractility (+dP/dt) and relaxation (–dP/dt) in isolated

cardiomyocytes (Figure 22) supports this hypothesis. It should be noted that, while this

difference in intracellular calcium concentrations does not result in a change in function for the

non-ischemic cardiomyocyte, this may not be the case under conditions of ischemia or

reperfusion, where elevated of calcium levels result in significant injury. Second, there may be

other compensatory factors in HMWKO heart cells that may have an impact on the influence of

intracellular calcium concentration on contraction, such as differences in the calcium sensitivity

of myofilament proteins to changes in calcium concentrations.

To account for the difference in calcium cycling in cardiomyocytes between wildtype and

FGF2 HMWKO mice, the protein levels of known regulators of calcium cycling were evaluated,

focusing on proteins known to mediate calcium reuptake into the SR. Proteins examined were

SERCA, which uses ATP to the SR from the cytosol; phospholamban, which regulates the activity of SERCA, and calsequestrin, which binds to calcium and keeps it localized near SR calcium channels (MacLennan & Wong, 1971). Regulation of all of these proteins has been shown to modulate intracellular calcium handling and alter cardiac contractility and relaxation

(Bluhm et al., 2000; Loukianov et al., 1998; Periasamy & Huke, 2001; Schmidt et al., 2000;

Schwinger et al., 2000; W. Zhao et al., 2003). However, the levels of these proteins were not found to be significantly changed between non-ischemic hearts from wildtype mice and those only expressing the LMW isoform of FGF2, or lacking expression of all isoforms of FGF2.

While none of the candidate proteins showed changes in expression, the identification of novel calcium-handling proteins that may be regulated by LMW FGF2 expression will be discussed in chapter 5.

The phosphorylation of phospholamban was also evaluated, and no changes were

124

observed in sham hearts evaluated (Figure 25). However, after ischemia was induced, hearts

expressing LMW FGF2 showed significantly higher phosphorylation at threonine-17 compared

to hearts lacking FGF2 expression, suggesting that LMW FGF2 expression promotes the

phosphorylation of phospholamban at this residue. Work by Mattiazzi and colleagues (Mattiazzi

et al., 2004; Mattiazzi et al., 2006) suggest that phosphorylation of the threonine-17 residue on

phospholamban during the first few minutes of ischemia is critical for recovery of the heart from

I/R injury, and hearts expressing phospholamban that has been mutated to be unphosphorylatable do not recover to the same degree as their wildtype counterparts. In view of this data, it is interesting to speculate whether phospholamban phosphorylation may contribute to LMW FGF2

-mediated cardioprotection. Work is ongoing to examine whether hearts only expressing LMW

FGF2 are still protected from cardiac dysfunction when crossbred to a mouse expressing only

T17A phospholamban with threonine-17 mutated to an unphosphorylatable alanine (T17A), a mouse model with alterations in the positive relationship between contraction and frequency in cardiomyocytes (Zhao et al., 2004).

The intermediary steps by which LMW FGF2 expression results in increased

phosphorylation of phospholamban at threonine-17 are still unclear. This site has been identified

as a target of CamKII (Simmerman et al., 1986), which may be modulated by PKC (Oestreich et

al., 2009). However, the PKC isoform that is activated early enough to impact phospholamban

phosphorylation at this time point, PKCα, does not appear to regulate phosphorylation of

phospholamban at early ischemia or early reperfusion. In fact, the activation of this PKC isoform

at this timepoint may be more of an effect than a cause of phospholamban regulation;

phosphorylation at threonine-17 is expected to relieve phospholamban’s inhibition on SERCA,

promoting increased calcium uptake and higher amplitude of calcium release during systole

(Mattiazzi et al., 2006), which may in turn result in higher activation of PKCα, a conventional

125

calcium-activated kinase (Kikkawa et al., 1987).

Alternatively, CamKII shows a higher degree of activation in hearts only expressing

LMW FGF2 compared to wildtype (Figure 26), suggesting that this increase in activity could be

responsible for the increase in PLB phosphorylation. The data presented here suggest that the

levels of phosphorylated CamKII drop in wildtype hearts during I/R injury, which is supported

by Zhuo and colleagues (Zhuo et al., 2009), who demonstrate decreased CamKII activity under

conditions of hypoxia in rat hearts. However, in the presence of only LMW FGF2 expression, the

phosphorylation state of CamKII remains constant through ischemia, and is significantly higher

than wildtype cohorts during early ischemia (Figure 26), suggesting that LMW FGF2 may be

positively regulating CamKII’s activity. The question of how an increase in the phosphorylation of a CamKII target can correspond to no change in CamKII activity as determined by phosphorylation, as seen when hearts only express LMW FGF2, remains to be determined; it may be that CamKII activity is finely regulated during ischemia, and LMW FGF2 expression alters the balance in favor of higher activation, or it may be the case that CamKII already is

maximally activated in HMWKO mice during ischemia. CamKII has been observed to be

activated by acidosis in the heart (Mundiña-Weilenmann et al., 2005) suggesting that its activity

should be increased during ischemia. Future studies are necessary to determine if this is the case.

Although the data presented in this dissertation give evidence for a role for LMW FGF2 regulation of calcium cycling in mice, a limitation of this work may be its relevance to humans.

Calcium cycling in mouse and human ventricular myocytes is regulated differently, with a much larger role for the SR in mice compared to humans (Hovnanian, 2007). Nevertheless, the importance of phospholamban regulation in cardiac disease has been well established in human patients, in spite of the SR’s diminished role in normal cardiac function (Haghighi et al., 2008;

Landstrom et al., 2011).

126

The research presented in this dissertation reveals a new perspective by which LMW

FGF2 may have an impact on post-ischemic cardiac function. The understanding that LMW

FGF2 may regulate intracellular , and may result in regulatory changes in a

protein that is essential for recovery from I/R injury, opens up the possibility that LMW FGF2

may directly affect calcium cycling in the myocyte to protect the heart from post-ischemic cardiac dysfunction.

127

FGFR 1 LMW FGF2

HMW FGF2

CaMKII

SR PLB

SERCA

Improved postischemic function

Figure 27. Schematic representing the hypothesized role of calcium handling protiens in

LMW FGF2-mediated protection from post-ischemic dysfunction. LMW FGF2 expression

results in increased activity of CaMKII and increased phosphorylation of phospholamban at threonine 17, which reduces phospholamban’s inhibition on SERCA during I/R injury.

128

Chapter 4: LMW FGF2 and HMW FGF2 crosstalk during ischemia-reperfusion injury.

RESULTS IV.

Contractile and relaxation recovery of hearts overexpressing HMW FGF2 is depressed compared to non-transgenic cohorts

To fully understand the effect of FGF2 isoforms in the heart, it is necessary to determine the role of HMW FGF2 in the recovery of the heart from I/R injury, in addition to LMW FGF2 as described in previous chapters. As previously demonstrated by our laboratory, mouse hearts only expressing the HMW isoform of FGF2 have poorer recovery after I/R injury (Liao et al.,

2010). To further investigate the role of HMW FGF2 on post-ischemic function, hearts from mice overexpressing the human 24 kDa HMW isoform of FGF2 were subjected to I/R injury.

The contractility and relaxation of these hearts was significantly reduced after reperfusion

(Figure 28), confirming that HMW FGF2 has an injurious effect on cardiac function after I/R.

129

A. 100

80 *p<0.05 vs. WTNTg

60

40 * * 20

% Contractile recovery (+dP/dt) recovery Contractile % 0 NTgNTg FGF2FGF2 HMW HMWTg Tg 24IP20FGF2FGF2 FGF2 HMW HMWTg HMWTgTg 24IP28 24IP20 24IP2824IP28 B.

100 80 *p<0.05 vs. NTg 60 40 * * 20 0 NTgNTg FGF2 HMW Tg 24IP20 FGF2FGF2 HMW HMWTg Tg 24IP28

% Relaxation recovery (-dP/dt) recovery Relaxation % FGF2 HMWTg 24IP20 24IP28

Figure 28. The percent recovery of contractility (+dP/dt) (A) and relaxation (-dP/dt) (B) of non-transgenic (NTg, black bar) or HMW FGF2 overexpressing (FGF2 HMW Tg 24IP20 and FGF2 HMW Tg 24IP28, striped bars) hearts that have undergone 60 minutes ischemia and 120 minutes reperfusion. HMW Tg hearts performed significantly worse after I/R injury. Post-ischemic functional recovery is expressed as a percentage of baseline function.*p<0.05 vs. NTg (n=4-6 hearts per group).

130

FGFR activation during reperfusion is lowered in HMW FGF2 overexpressing hearts

The protective effects of LMW FGF2 on post-ischemic cardiac function are mediated via

the FGFR1 receptor (Liao et al., 2010), and HMW FGF2 shares sequence homology that may

allow it, if given exogenously, to bind to FGFR1 as well (Florkiewicz & Sommer, 1989;

Plotnikov et al., 2000). However, it is unknown if endogenously expressed HMW FGF2

exacerbates I/R-induced post-ischemic dysfunction via interaction with the FGFR1 receptor.

Therefore, whether the effects of HMW FGF2 overexpression were mediated by the FGF

receptor was addressed. Hearts overexpressing the human 24 kDa HMW isoform of FGF2 were

subjected to I/R injury, and the activation (i.e., phosphorylation) of FGFR was evaluated.

Overexpression of HMW FGF2 resulted in a decrease in the phosphorylation of FGFR, with a

trend in decreased FGFR activation immediately at the onset of reperfusion after 60 minutes of ischemia (Figure 29A), and a statistically significant decrease after 120 minutes of reperfusion

(Figure 29B), suggesting that not only does HMW FGF2 not activate FGFR1, but overexpression prevents normal FGFR activation seen in wildtype cohorts. Both non-transgenic (NTg) and

HMW FGF2 overexpressing hearts were subjected to the selective FGFR1 inhibitor PD173074

(Skaper et al., 2000), which resulted in reduced FGFR1 phosphorylation of the non-transgenic hearts, but there was no further decrease in FGFR1 phosphorylation in HMW Tg hearts after 60 minutes of ischemia and 120 minutes of reperfusion (Figure 29). Importantly, the FGFR1 inhibitor reduced phosphorylated FGFR1 levels in the wildtype hearts to that observed in hearts with an overexpression of HMW FGF2, which suggests that this decrease in phosphorylation induced by HMW FGF2 is sufficient to be physiologically relevant.

131

A. FGFR1 activation (Early Reperfusion) 2 DMSO PD173074 1.5 *p<0.05 vs. NTg vehicle treated 1 * 0.5

Phospho/Total FGFR1 Phospho/Total 0 NTgNTg HMWTg Tg FGFR1 activation (Late Reperfusion) B. 4 *p<0.05 vs NTg vehicle treated

3 DMSO PD173074

2 * * *

Phospho FGFR1/Total FGFR1 Phospho FGFR1/Total 1 Phospho/Total FGFR1 Phospho/Total 0 NTgNTG HMW24 kDa Tg

Figure 29. Ratio of phosphorylated/total FGFR1 in non-transgenic (NTg) and 24 kDa

HMW FGF2 overexpressing (Tg) hearts subjected to 60 minutes of ischemia and 15 minutes of reperfusion (A), or 60 minutes of ischemia and 120 minutes of reperfusion (B) in the absence (black bar) or presence (white bar) of the selective FGFR1 inhibitor

PD173073. *p<0.05 vs. vehicle-treated NTg. (n=4-6 hearts).

132

PKCs and MAPKs, kinases downstream of FGFR1, have lowered activation in hearts overexpressing HMW FGF2 during reperfusion

Given that overexpression of HMW FGF2 appears to interfere with FGFR1 signaling, the activity of the kinases downstream of FGFR1 in transgenic and non-transgenic mice was assessed. The kinases evaluated were PKCα, ε, and δ, and MAPKs ERK1/2, JNK, and p38, as these kinases were found to be involved in the signaling pathways activated by FGF2 during I/R to protect the heart from cell death and functional injury (House et al., 2005; House et al., 2007;

Jiang et al., 2002; Liao et al., 2007; Padua et al., 1998; Sheikh et al., 2001). Only one isoform of

PKC, PKCα, showed a lesser degree of phosphorylation in hearts with an overexpression of the

24 kDa HMW isoform of FGF2compared to NTg hearts (Figure 30A). In addition, JNK activity was lower with overexpression of human 24kDa HMW FGF2, while p38 and ERK1/2 showed no change compared to NTg hearts (Figure 31). The selective FGFR inhibitor PD173074 did not affect the activation of any PKCs, but it was shown that the FGFR blockade resulted in an increase in ERK activation, only in non-trangenic hearts (Figures 30 and 31); HMW FGF2 overexpressing cohorts did not show this increase with exposure to the FGFR inhibitor.

Similarly, p38 phosphorylation was increased in the presence of the FGFR inhbition as well in

HMW FGF2 overexpressing hearts (Figure 31C), although no statistically significant difference was found in non-transgenic hearts.

133

A. *p<0.05 vs. NTg

/ PKC alpha activation

α 2 vehicle treated α 1.5 DMSO 1 PD173074 * 0.5

Total PKC 0

Phospho-PKC NTg HMWTg Tg

PKC epsilon activation B. DMSO / ε PD173074 ε 3

2

1

Total PKC Total 0 Phospho-PKC NTgNTg HMWTg Tg

PKC delta activation C. DMSO /

δ PD173074 1.5 δ 1

0.5

0 Total PKC Total

Phospho-PKC NTgNTg HMWTg Tg

Figure 30. PKC isoform activation in the absence (black bar) or presence (white bar) of the

FGFR1 inhibitor PD173074 (25 nM) after 60 minutes of ischemia and 5 minutes of reperfusion in non-transgenic (NTg) and 24 kDa HMW FGF2 overexpressing (Tg) hearts.

A significant decrease in the activation of PKCα was seen Tg hearts in the absence of drug treatment (A). No significant difference was seen in the activation of PKC epsilon (B) or

PKC delta (C). *p<0.05 vs. vehicle-treated NTg. (n=5-8 hearts per group).

134

*p<0.05 vs. NTg vehicle treated JNK activation DMSO A. 1.2 PD173074 1 0.8 0.6 * 0.4 0.2 Total JNK Total 0 Phospho-JNK/ Phospho-JNK/ NTgNTg HMWTg Tg

p38 activation B. 1.2 DMSO * 1 PD173074 0.8 0.6 0.4 Total p38 0.2 Phospho-p38/ Phospho-p38/ 0 NTgNTg HMWTg Tg

C. ERK1/2 activation

0.5 0.4 * DM SO 0.3 PD173074 0.2 0.1

Total ERK1/2 0

Phospho-ERK1/2/ Phospho-ERK1/2/ NTg HMWTg Tg NTg

Figure 31. MAPK activation in the absence (black bar) or presence (white bar) of the

FGFR1 inhibitor after 60 minutes of ischemia and 5 minutes of reperfusion in non- transgenic (NTg) and 24 kDa HMW FGF2 overexpressing (Tg) hearts. A. JNK phosphorylation was decreased in untreated hearts (A), while p38 was increased in transgenic hearts in the presence of the FGFR inhibitor (B), and ERK1/2 showed a higher degree of phosphorylation in the presence of FGFR inhibition in NTg hearts. *p<0.05 vs. vehicle-treated NTg. (n=5-8 hearts).

135

Post-ischemic recovery of cardiac function in 24 kDa HMW FGF2 transgenic hearts

treated with the selective FGFR1 inhibitor PD173074 is elevated compared to wildtype

cohorts.

To determine if HMW FGF2-mediated functional injury after I/R is dependent on the

FGFR receptor, hearts from mice overexpressing the human 24 kDa HMW isoform of FGF2

were exposed to 60 minutes of ischemia and 120 minutes of reperfusion in the presence of the

selective FGFR inhibitor, PD173074. As previously observed (Liao et al., 2010) (Figure 28),

overexpression of this HMW isoform resulted in lowered post-ischemic contractile and

relaxation recovery. Surprisingly, the administration of PD173074 resulted in a clear increase in post-ischemic function in two independent mouse lines overexpressing the HMW isoform of

FGF2 to levels higher than in nontransgenics (Figure 32).

136

A. *p<0.05 vs. NTg vehicle treated

100 #p<0.05 vs. NTg PD173074 treated # * # 80 DMSO * PD173074

60

40 %Recovery (+dp/dt) * * 20 %Contractile Recovery (+dP/dt) %Contractile Recovery 0 NTgNTg FGF2FGF2 HMW HMW Tg 24IP20 Tg FGF2FGF2 HMW HMW Tg 24IP28 Tg 24IP20 24IP28

B. *p<0.05 vs. NTg vehicle treated #p<0.05 vs. NTg PD173074 treated

100 DMSO # 80 * PD173074 * # 60 40 * * 20

% Relaxation% Recovery (-dP/dt) 0 NTgNTg FGF2FGF2 HMW HMW Tg 24IP20 Tg FGF2FGF2 HMW HMW Tg 24IP28 Tg 24IP20 24IP28

Figure 32. The percent recovery of contractility (+dP/dt) (A) and relaxation (B) of non- transgenic (NTg) or HMW FGF2 overexpressing (FGF2 HMW Tg 24IP20 and FGF2

HMW Tg 24IP28) hearts that have undergone 60 minutes ischemia and 120 minutes reperfusion in the presence of DMSO (black bar) or 25 nM FGFR1 inhibitor, PD173074

(white bar). Transgenic hearts of both lines recovered to a significantly higher degree in the presence of the inhibitor after I/R injury. *p<0.05 vs. vehicle-treated NTg. #p<0.05 vs.

PD-treated Ntg (n=4-6 hearts).

137

DISCUSSION IV.

These data present evidence for a detrimental role of HMW FGF2 in cardiac function

following ischemia and reperfusion injury. Mouse lines overexpressing the human 24 kDa HMW

isoform of FGF2 show reduced post-ischemic recovery of cardiac function, suggesting that

HMW FGF2 is injurious to the contractile recovery of the heart (Figure 28). It was found that

this decrease in contractility corresponds to a decrease in FGFR1 activation, reducing phosphorylation of FGFR1 to levels seen on exposure to an FGFR1 inhibitor (Figure 29).

Additionally, downstream targets of FGFR1 were also found to have lowered activation,

including PKCα and JNK (Figures 30 and 31), suggesting that HMW FGF2 may elicit increased

postischemic dysfunction by interfering with the protective pathways activated by LMW FGF2

(Figure 33).

The role of FGFR in I/R injury has been extensively studied by our laboratory (Liao et al., 2010) and others (Jiang et al., 2002). FGFR plays a protective role in the presence of LMW

FGF2 expression (Liao et al., 2010), or when exogenous FGF2 is administered to the heart (Jiang et al., 2002). Both of these instances involve FGF2 available in the extracellular matrix, where it is free to bind to HSPGs and the extracellular domain of FGFR, inducing downstream signal cascades that can protect the heart. However, this dissertation shows a decrease in the activation of FGFR1, demonstrated by significantly less phosphorylation of the receptor at the first tyrosine residues, which undergo autophosphorylation and receptor dimerization, by overexpression of human 24 kDa HMW FGF2 (Figure 29), which remains intracellular (Liao et al., 2010). This suggests that HMW FGF2 may interfere with protective FGF2 signaling. Several mechanisms are speculated as how this may occur: FGF2 may bind to LMW FGF2 within the cell and prevent its release (and subsequent activation of the receptor); HMW FGF2 may bind directly to the

intracellular domain of the receptor, or scaffolding proteins at the receptor, and prevent FGFR

138

dimerization or activation or, HMW FGF2 may bind to intracellular FGFR, preventing localization of the receptor at the sarcolemma and rendering it unavailable for ligand binding and activation. Future research will determine which, if any, of these mechanisms are occurring to produce the effects on FGFR1 observed in this dissertation.

The downstream effectors that are implicated in HMW FGF2 Tg hearts after I/R are

PKCα and JNK, both of which show depressed phosphorylation compared to NTg cohorts. The

importance of PKCα activation in LMW FGF2-mediated protection from I/R-induced

dysfunction is discussed in depth in chapter 1 and 2 of this dissertation, where it was found that

PKCα is activated by LMW FGF2 in early ischemia and is necessary for the improvement in

post-ischemic cardiac function, as well as the increase in the phosphorylation of regulatory

myofilament proteins during early ischemia. This lends credence to the theory that HMW FGF2

produces its deleterious effects by interfering, at some level, with LMW FGF2’s protective

signaling cascade. In addition, previous work from our lab examining the role of MAPKs in

hearts only expressing HMW FGF2, with LMW FGF2 expression ablated, have implicated a

role for p38 and JNK (Liao et al., 2007). Interestingly, hearts that have LMW FGF2 expression

ablated show a higher degree of JNK activation. This suggests that LMW FGF2 and HMW

FGF2 both finely regulate the activity of JNK, with the expression of only HMW FGF2 resulting

in JNK activation, but with both LMW FGF2 expressed and HMW FGF2 overexpressed

resulting in decreased JNK activation compared to wildtype hearts.

An explanation for the effects of FGFR1 inhibitor on HMW FGF2 remains unclear based

on the data presented here. While there is no detectable difference in FGFR1 phosphorylation after I/R in HMW FGF2 transgenic hearts between vehicle-treated and PD173074-treated hearts, the FGFR1 inhibitor, PD173074, has demonstrated cardioprotective effects against cardiac dysfunction in HMW FGF2 overexpressing hearts, suggesting that the mechanism by which this

139

occurs is not due to the inhibitor’s effects on its intended targets. While these results are intriguing, they appear to be due to off-target effects of the inhibitor.

FGFR 1 LMW FGF2

HMW FGF2

JNK PKCα

Improved postischemic function

Figure 33. Schematic representing the hypothesized role of HMW FGF2 in LMW FGF2 protective signaling during I/R injury. LMW FGF2 protects the heart via activation of

FGFR1 and PKCα, which are inhibited by overexpression of HMW FGF2.

140

Chapter 5: LMW and HMW FGF2-induced NO production during I/R

RESULTS V.

NO release is increased during I/R in treated hearts overexpressing all isoforms of FGF2,

and dependent on PKC activity.

Previous studies have established that protection from post-ischemic cardiac functional

injury and infarct development seen when all isoforms of FGF2 are overexpressed is dependent

on the activity of PKC and MAPKs (House et al., 2005; House et al., 2007). A potential target of these kinases are nitric oxide synthases (NOSs), which are regulated by both PKCs (Das et al.,

2004; Kaneda et al., 2008; Partovian et al., 2005; H. Y. Zhang et al., 2002) and MAPKs ERK and p38 (Das et al., 2009; Khan et al., 2009; Singh et al., 1996; T. C. Zhao et al., 2001). Our laboratory has demonstrated that nitric oxide signaling is important for FGF2-induced cardioprotection against myocardial infarction (Manning et al., 2012 (in press)) as well as with other modes of cardioprotection or ischemic preconditioning (R. M. Bell & Yellon, 2001; Bolli et al., 1998; Bouwman et al., 2004; Burkard et al., 2010; Cuevas et al., 1999; Das et al., 2009;

Dawson et al., 2005; Downey, Davis, & Cohen, 2007; Guo et al., 2008; Hampton et al., 2000;

Kaneda et al., 2008; Khan et al., 2009; Maslov et al., 2009; Saraiva et al., 2005; Sears et al.,

2003; Sharp et al., 2002; Sun et al., 2009; West et al., 2008; Xuan et al., 2007). To determine if the kinases activated by overexpression of all isoforms of FGF2 alter nitric oxide synthesis during I/R, the release of NO from FGF2 transgenic and nontransgenic hearts was evaluated in the presence and absence of an inhibitor of PKCα, β, , δ, ε, and ζ isoforms (GFX) as well as

MAPKs ERK1/2 (U0126), and p38 (SB203580). Hearts that overexpress all isoforms of FGF2 showed higher levels of NO oxidation product (nitrite, NO2-) in the coronary effluent during

reperfusion. However, in the presence of PKC inhibitor, GFX (Figure 34), the MEK/ERK

141

inhibitor, U0126 (Figure 35), and the p38 inhibitor, SB203580 (Figure 36), nitrite levels were significantly reduced (p<0.05), suggesting that PKCs and MAPKs are upstream of FGF2- induced activation of NOS.

142

*p<0.05 vs. NTg from same treatment #p<0.05 vs. FGF2 Tg PKC inhibitor treatment †p<0.05 vs. NTg PKC inhibitor *# treatment *# *# *# *# *# *# † *# † † † † *# † *#

Ischemia (min) Reperfusion (min)

Figure 34. Nitric oxide (NO) oxidation product (nitrite) released from NTg (triangle) and

FGF2 Tg (square) hearts. Hearts overexpressing all isoforms of FGF2 and non-transgenic cohorts were subjected to vehicle (black) or bisindolylmaleimide (GFX) (grey), an inhibitor of multiple isoforms of PKC. Coronary effluent was analyzed for the concentration of NO2-,

an oxidation product of the nitric oxide synthase product NO. Overexpression of all

isoforms of FGF2 resulted in significantly higher levels of NO2- released into the coronary

effluent during reperfusion. Treatment with GFX reduced NO2- back to non-transgenic

levels. Gray region indicates timepoints of kinase inhibitor treatment. *p<0.05 vs. NTg.

#p<0.05 vs. FGF2 Tg + GFX. †p<0.05 vs. NTg + GFX (n=6-11 per group).

143

*p<0.05 vs. NTg from same treatment #p<0.05 vs. FGF2 Tg MEK/ERK inhibitor treatment †p<0.05 vs. NTg MEK/ERK inhibitor treatment *# *# *# *# *# *# *# *# *# † † † † † † † †

Ischemia Reperfusion (min) (min)

Figure 35. Nitric oxide (NO) oxidation product (nitrite) released from NTg (triangle) and

FGF2 Tg (square) hearts. Hearts overexpressing all isoforms of FGF2 and non-transgenic cohorts were subjected to vehicle (black) or 2.5µM U0126 (grey), an inhibitor of

MEK/ERK. Coronary effluent was analyzed for the concentration of NO2-, an oxidation

product of the nitric oxide synthase product NO. Overexpression of all isoforms of FGF2

resulted in significantly higher levels of NO2- released into the coronary effluent during

reperfusion. Treatment with U0126 reduced NO2- back to non-transgenic levels. Gray

region indicates timepoints of kinase inhibitor treatment. *p<0.05 vs. NTg. #p<0.05 vs.

FGF2 Tg + U0126. †p<0.05 vs. NTg + U0126. (n=6-11 per group).

144

*p<0.05 vs. NTg from same treatment #p<0.05 vs. FGF2 Tg p38 inhibitor treatment †p<0.05 vs. NTg p38 inhibitor *# treatment *# *# *# *# *# *# *# *# † † † † † †

Ischemia Reperfusion (min) (min)

Figure 36. Nitric oxide (NO) oxidation product (nitrite) released from NTg (triangle) and

FGF2 Tg (square) hearts. Hearts overexpressing all isoforms of FGF2 and non-transgenic

cohorts were subjected to vehicle (black) or 2µM SB203580 (grey), an inhibitor of p38.

Coronary effluent was analyzed for the concentration of NO2-, an oxidation product of the

nitric oxide synthase product NO. Overexpression of all isoforms of FGF2 resulted in significantly higher levels of NO2- released into the coronary effluent during reperfusion.

Treatment with SB203580 reduced NO2- back to non-transgenic levels. *p<0.05 vs. NTg.

#p<0.05 vs. FGF2 Tg + SB203580. †p<0.05 vs. NTg + SB203580. (n=6-11 per group).

145

DISCUSSION V.

In addition to initiating kinase cascades in the ischemic heart, which target the myofibril

and calcium handling proteins, FGF2 isoforms may also use other signal transduction

mechanisms to affect post-ischemic cardiac dysfunction. FGF2 activates NOS during reperfusion in a manner that is dependent on the activity of PKC.

The data presented in this dissertation show that overexpression of all isoforms of FGF2

results in an increase in NO oxidation products, suggesting that NOS activity is upregulated by

one or more isoforms of FGF2. Further, our laboratory has shown that blocking all isoforms of

NOS, or just nNOS or iNOS, results in abrogation of the protection from infarct development

seen in all-isoform overexpressoring mouse hearts (Manning et al., 2012 (in press)). This

dissertation and manuscript show that this increase in NOS activity seems to be dependent on

PKC, with an inhibitor of multiple isoforms of PKC removing the increase in NO oxidation products during reperfusion. PKCs have been shown to regulate the function of isoforms of

NOS, including eNOS and nNOS (Adak et al., 2001; Partovian et al., 2005; G. A. Rameau et al.,

2007; G. A. Rameau et al., 2004). Phosphorylation of these NOS isoforms may have differing

effects on their activity, depending on the residues phosphorylated. nNOS may be

phosphorylated at S847 and S1412, both of which increase its activity (G. A. Rameau et al.,

2007; G. A. Rameau et al., 2004). Similarly, eNOS has been shown to be phosphorylated by

PKCα at S1179 (Partovian et al., 2005), a site analogous to S1412, which has been shown to

increase NO production in endothelial tissue (Adak et al., 2001). The finding that overexpression

of all human isoforms of FGF2 results in a PKC- and MAPK-dependent increase in NO2- during

reperfusion sheds light on a new signaling pathway for HMW and LMW FGF2, and may lead to

a better understanding of the role of NOS in FGF2-mediated cardioprotection.

146

Chapter 6: LMW and HMW FGF2-regulated gene transcription

RESULTS VI.

LMW and HMW FGF2 differentially regulate gene transcription in non-ischemic hearts.

In addition to rapid signaling events such as kinase cascades or NO signaling, it was also hypothesized that modulation of the FGF2 isoform expression may impact the regulation of proteins at the genomic level, altering the transcription of genes that may affect the functional recovery of the heart after I/R injury. To gain insight into the transcriptional regulation of each isoform of FGF2 in the ventricle, RNA was isolated from hearts that express only LMW FGF2, only HMW FGF2, or do not express either isoform. This RNA was then analyzed using an

Affymetrix GeneST mouse array, evaluating all known mouse genes to identify up or downregulation of the genes by either isoform. Genes that showed a significant change in expression were identified, resulting in approximately 500 genes that were altered between

HMWKO (expressing LMW), Fgf2 KO (no FGF2 isoforms), and LMWKO (expressing HMW) mouse hearts with a p value of less than 0.005 and FDR < 1 (Tables 4, 5, and 6). From these

genes that were initially documented, genes impacting survival, ion homeostasis, calcium

handling, or cytoskeletal/contractile roles are depicted in Table 8 and were further validated.

Several candidate genes were identified, and changes in expression level were confirmed via

qRT-PCR and/or immunoblotting. Genes with altered expression that were confirmed with either

qRT-PCR or immunoblotting were sarcolipin (Figure 38) and myosin light chain (Figures 41 and

42), respectively. Calreticulin was also predicted by the microarray (Figure 39), but no

differences were seen at the protein level using immunoblot (Figure 40). Identified by the

microarray but not confirmed were FGF12, tropomodulin 4, and targets of several microRNAs

147

that have been demonstrated to affect the outcome of hearts that have undergone I/R injury,

including Mir-15 (Porrello et al., 2011; van Rooij et al., 2006).

Genes differentially regulated in HMWKO vs. Fgf2 KO False Discovery Abbreviation Gene Name Fold Change P Value Rate (FDR) Myl7 myosin, light polypeptide 7, regulatory 31.12418683 7.87E-07 0.01668477 Mybphl myosin binding protein H-like 5.385087369 1.46E-05 0.118256157 Myl4 myosin, light polypeptide 4 27.0730983 1.67E-05 0.118256157 Sln sarcolipin 13.55439298 4.04E-05 0.202807027 Fmn1 formin 1 -1.675573797 4.78E-05 0.202807027

Hprt hypoxanthine guanine phosphoribosyl -2.024429969 7.83E-05 0.25237465 Higd1a HIG1 domain family, member 1A -2.999052135 9.26E-05 0.25237465 Egln3 EGL nine homolog 3 (C. elegans) -1.715069004 9.52E-05 0.25237465 Sh3bgr SH3-binding domain glutamic acid-rich protein 1.394881837 0.000112 0.264999416 Fnip1 folliculin interacting protein 1 -1.338251551 0.000226 0.394218957 Msi2 Musashi homolog 2 (Drosophila) -1.293672708 0.000227 0.394218957 Fah fumarylacetoacetate -2.435338686 0.000228 0.394218957 Fam59a family with sequence similarity 59, member A -1.334663757 0.000259 0.394218957 Gm9625 predicted gene 9625 2.188391024 0.000273 0.394218957

Fsd2 fibronectin type III and SPRY domain containing 2 -1.631638351 0.000279 0.394218957 Midn midnolin -1.545892499 0.000359 0.442880353 Tmx4 thioredoxin-related 4 -1.230505766 0.000361 0.442880353 Alad aminolevulinate, delta-, dehydratase 2.054208912 0.000376 0.442880353 acyl-CoA synthetase medium-chain family member Acsm5 5 -1.758091669 0.000427 0.459919986 Zfp281 zinc finger protein 281 -1.233882551 0.000434 0.459919986 NADH dehydrogenase (ubiquinone) 1 alpha Ndufaf4 subcomplex, assembly factor 4 -1.219934997 0.000474 0.46593234 Gm5617 predicted gene 5617 -1.716997795 0.000483 0.46593234 Zbtb37 zinc finger and BTB domain containing 37 -1.311761301 0.000515 0.474679905 2310061C15Rik RIKEN cDNA 2310061C15 gene -1.221317883 0.000545 0.481242572 gamma-aminobutyric acid (GABA) C receptor, Gabrr2 subunit rho 2 -1.49022082 0.000593 0.493917112 Gm5093 predicted gene 5093 1.794075961 0.000606 0.493917112

148

Gm2701 predicted gene 2701 -1.252111444 0.000734 0.544084124 Gm1078 predicted gene 1078 1.899270684 0.000752 0.544084124 Fgf2 fibroblast growth factor 2 1.90985098 0.000764 0.544084124 solute carrier family 16 (monocarboxylic acid Slc16a1 transporters), member 1 -1.313311022 0.00077 0.544084124 elongation factor Tu GTP binding domain Eftud2 containing 2 1.355510976 0.000828 0.566582324

Hey1 hairy/enhancer-of-split related with YRPW motif 1 -1.368175968 0.000897 0.571342421 asparagine-linked glycosylation 10 homolog B (yeast, Alg10b alpha-1,2-glucosyltransferase) -1.175690839 0.000898 0.571342421 Fgf12 fibroblast growth factor 12 5.134264575 0.000921 0.571342421 Vmn2r4 vomeronasal 2, receptor 4 -1.660852699 0.000963 0.571342421 Supv3l1 suppressor of var1, 3-like 1 (S. cerevisiae) -1.307477871 0.001083 0.571342421 Narf nuclear prelamin A recognition factor -1.250676912 0.001125 0.571342421 Rfk riboflavin kinase -1.171541791 0.001134 0.571342421 Alport syndrome, mental retardation, midface hypoplasia and elliptocytosis chromosomal region Ammecr1 gene 1 homolog (human) -1.60931826 0.001148 0.571342421 Gata6 GATA binding protein 6 -1.283043946 0.001194 0.571342421 ATP-binding cassette, sub-family D (ALD), member Abcd2 2 -1.439548787 0.001217 0.571342421 4833439L19Rik RIKEN cDNA 4833439L19 gene -1.181252048 0.00122 0.571342421 Ifltd1 intermediate filament tail domain containing 1 2.888030678 0.001251 0.571342421 Lsm12 LSM12 homolog (S. cerevisiae) -1.27600088 0.001256 0.571342421 Gm6843 predicted gene 6843 1.315540358 0.001396 0.571342421 Gm6155 predicted gene 6155 1.346669932 0.001435 0.571342421 dehydrogenase/reductase (SDR family) member Dhrs7c 7C 1.290394955 0.00146 0.571342421 twinfilin, actin-binding protein, homolog 2 Twf2 (Drosophila) 1.384324714 0.001461 0.571342421 Robld3 roadblock domain containing 3 1.238987174 0.001495 0.571342421 STT3, subunit of the oligosaccharyltransferase Stt3b complex, homolog B (S. cerevisiae) -1.250144511 0.001501 0.571342421 Zfp831 zinc finger protein 831 1.183081039 0.001536 0.571342421 Rab30 RAB30, member RAS oncogene family -1.474958076 0.00157 0.571342421 Rps25 ribosomal protein S25 1.437685201 0.001586 0.571342421 Rapgef1 Rap guanine nucleotide exchange factor (GEF) 1 -1.347137372 0.001602 0.571342421 Mcam melanoma 1.403997089 0.001636 0.571342421 Ubiad1 UbiA prenyltransferase domain containing 1 -1.435130155 0.001638 0.571342421 Pcdh11x protocadherin 11 X-linked 1.165559679 0.001703 0.571342421 glyceraldehyde-3-phosphate dehydrogenase Gm12537 pseudogene 1.370417329 0.001722 0.571342421 Ptp4a2 protein tyrosine phosphatase 4a2 -1.133881907 0.001728 0.571342421 Gm5934 predicted gene 5934 -1.247936088 0.001754 0.571342421 Asb2 ankyrin repeat and SOCS box-containing 2 -1.353564619 0.001789 0.571342421 Gm10361 predicted gene 10361 1.327954903 0.001824 0.571342421 Lrrc2 leucine rich repeat containing 2 1.250918202 0.00183 0.571342421 Pkig protein kinase inhibitor, gamma -1.326088272 0.001893 0.571342421 Gca grancalcin -1.3960236 0.001902 0.571342421

149

Stx4a syntaxin 4A (placental) 1.319481089 0.001906 0.571342421 Frmd6 FERM domain containing 6 -1.307570547 0.001962 0.571342421 5330426P16Rik RIKEN cDNA 5330426P16 gene 1.924609414 0.001981 0.571342421 Pgam5 phosphoglycerate mutase family member 5 -1.28231175 0.001993 0.571342421 1600029D21Rik RIKEN cDNA 1600029D21 gene 1.395907862 0.002014 0.571342421 9230105E10Rik RIKEN cDNA 9230105E10 gene -1.501778451 0.002037 0.571342421

Cpeb4 cytoplasmic polyadenylation element binding protein 4 -1.190644992 0.002038 0.571342421 acidic (leucine-rich) nuclear phosphoprotein 32 Anp32b family, member B 1.43107291 0.00206 0.571342421 Gm5866 predicted gene 5866 -1.361750492 0.002079 0.571342421 P2ry1 purinergic receptor P2Y, G-protein coupled 1 -1.646029313 0.00208 0.571342421 Fam20b family with sequence similarity 20, member B -1.631731963 0.00212 0.571342421 Tbl1x transducin (beta)-like 1 X-linked -1.161999503 0.002164 0.571342421 protein associated with topoisomerase II homolog 1 Patl1 (yeast) -1.226580184 0.002213 0.571342421 4930402E16Rik RIKEN cDNA 4930402E16 gene -1.453585977 0.002224 0.571342421 Snrpd3 small nuclear ribonucleoprotein D3 1.370262492 0.002239 0.571342421 1500003O03Ri k RIKEN cDNA 1500003O03 gene -1.285411749 0.002249 0.571342421 Gm5113 predicted gene 5113 -1.258525766 0.002259 0.571342421 Gm5785 predicted gene 5785 -1.269278061 0.002282 0.571342421 Fam36a family with sequence similarity 36, member A 1.154755158 0.002285 0.571342421 1110020G09Ri k RIKEN cDNA 1110020G09 gene -1.264935628 0.002317 0.571342421 Hddc3 HD domain containing 3 -1.904713879 0.002337 0.571342421

Gnb5 guanine nucleotide binding protein (G protein), beta 5 -1.237728968 0.002344 0.571342421 Lrg1 leucine-rich alpha-2-glycoprotein 1 -1.61080769 0.002543 0.612890772 Ang angiogenin, ribonuclease, RNase A family, 5 2.080137111 0.00258 0.614826079 Psg28 pregnancy-specific glycoprotein 28 -1.21487339 0.002635 0.614941437 Hamp hepcidin antimicrobial peptide 20.99420895 0.002657 0.614941437 1700019E19Rik RIKEN cDNA 1700019E19 gene 1.170251363 0.00268 0.614941437 Vmn2r28 vomeronasal 2, receptor 28 -1.464493046 0.002875 0.614941437 Tecrl trans-2,3-enoyl-CoA reductase-like -1.173309262 0.002877 0.614941437 protease (prosome, macropain) 26S subunit, ATPase Psmc5 5 1.143952533 0.002918 0.614941437 Gm13226 predicted gene 13226 1.518433923 0.003067 0.614941437 Mphosph6 M phase phosphoprotein 6 1.377590385 0.00309 0.614941437 myelin and lymphocyte protein, T-cell Mal differentiation protein -1.486514073 0.003107 0.614941437 Pm20d1 peptidase M20 domain containing 1 -1.775341988 0.00312 0.614941437 Sva seminal vesicle antigen -1.19045682 0.003122 0.614941437 Pla2g2d phospholipase A2, group IID 1.183502205 0.003213 0.614941437 Commd6 COMM domain containing 6 -1.414566279 0.003213 0.614941437 Gm6124 predicted gene 6124 -3.02440824 0.003215 0.614941437 Gm5437 ribosomal protein L31 pseudogene 2.09017162 0.003216 0.614941437 Igkv1-110 immunoglobulin kappa chain variable 1-110 -1.513774648 0.003226 0.614941437 processing of precursor 1, ribonuclease P/MRP Pop1 family, (S. cerevisiae) 1.368352385 0.003254 0.614941437

150

Csdc2 cold shock domain containing C2, RNA binding -1.402843203 0.003299 0.614941437

Gfod1 glucose-fructose domain containing 1 -1.242747725 0.003314 0.614941437 Gm7618 predicted gene 7618 1.127584981 0.003325 0.614941437 Bzw2 basic leucine zipper and W2 domains 2 -1.181692854 0.003333 0.614941437 1810030N24Rik RIKEN cDNA 1810030N24 gene -1.357067724 0.003415 0.614941437 protein tyrosine phosphatase, non-receptor type Ptpn11 11 -1.554102137 0.003497 0.614941437 Got1 glutamate oxaloacetate transaminase 1, soluble -1.152151375 0.003504 0.614941437 Zbtb16 zinc finger and BTB domain containing 16 -1.662144804 0.00354 0.614941437 Gpr19 G protein-coupled receptor 19 1.263778387 0.003565 0.614941437 Tmem14c transmembrane protein 14C -1.229429591 0.003569 0.614941437 Pank1 pantothenate kinase 1 -1.362763957 0.003576 0.614941437 Cds1 CDP-diacylglycerol synthase 1 -1.420611904 0.003613 0.614941437 Lyrm7 LYR motif containing 7 -1.348360744 0.003617 0.614941437 Clic5 chloride intracellular channel 5 -1.376424302 0.003617 0.614941437 Slc25a40 solute carrier family 25, member 40 -1.602153762 0.003633 0.614941437 Ank progressive ankylosis -1.222247558 0.003698 0.614941437 AI593442 expressed sequence AI593442 -1.918831571 0.003808 0.614941437 Vgll3 vestigial like 3 (Drosophila) 1.2963203 0.003822 0.614941437 Shisa3 shisa homolog 3 (Xenopus laevis) -1.195025097 0.003823 0.614941437 Gde1 glycerophosphodiester phosphodiesterase 1 1.365307188 0.003875 0.614941437 Rptn repetin 1.11961022 0.003901 0.614941437 Trak2 trafficking protein, kinesin binding 2 -1.422433692 0.00392 0.614941437 2310002L09Rik RIKEN cDNA 2310002L09 gene -1.214309628 0.003927 0.614941437 Lims2 LIM and senescent cell antigen like domains 2 -1.378608449 0.003932 0.614941437 Grm1 glutamate receptor, metabotropic 1 -1.698738512 0.003963 0.614941437 Rbpms2 RNA binding protein with multiple splicing 2 -1.555103123 0.003974 0.614941437 CCR4 carbon catabolite repression 4-like (S. Ccrn4l cerevisiae) -1.305997507 0.003977 0.614941437 Gm16528 predicted gene, 16528 -1.61581123 0.004011 0.614941437 P2ry2 purinergic receptor P2Y, G-protein coupled 2 -1.420140173 0.004058 0.614941437 Crybg3 beta-gamma crystallin domain containing 3 -1.27081882 0.004071 0.614941437 Ssbp2 single-stranded DNA binding protein 2 -1.203282952 0.00408 0.614941437 Ipmk inositol polyphosphate multikinase -1.233276895 0.004093 0.614941437 Pfkp phosphofructokinase, platelet -2.061544472 0.004094 0.614941437 Prkacb protein kinase, cAMP dependent, catalytic, beta -1.206449209 0.004095 0.614941437 Gpr155 G protein-coupled receptor 155 -1.373970327 0.004145 0.614941437 Efcab7 EF-hand calcium binding domain 7 1.266315827 0.004149 0.614941437 Apip APAF1 interacting protein 1.228031354 0.004161 0.614941437 Jph1 junctophilin 1 -1.167148823 0.004176 0.614941437 Gm10448 predicted gene 10448 -1.788233432 0.004345 0.617484102 Wdr20a WD repeat domain 20A -1.253044709 0.004371 0.617484102 Gm6245 predicted gene 6245 -1.082016713 0.004398 0.617484102 Hipk3 homeodomain interacting protein kinase 3 -1.169388682 0.004407 0.617484102 Gm5823 predicted gene 5823 -1.240592483 0.00444 0.617484102 0610009O20Rik RIKEN cDNA 0610009O20 gene -1.231590453 0.004455 0.617484102 Gm6651 predicted gene 6651 1.24350331 0.004461 0.617484102 4833413D08Rik RIKEN cDNA 4833413D08 gene -1.411093856 0.00448 0.617484102 Abhd6 abhydrolase domain containing 6 -1.569084105 0.004501 0.617484102

151

Dag1 1 -1.307204491 0.004539 0.617484102 Fam65a family with sequence similarity 65, member A -1.156695714 0.004633 0.617484102 Esd esterase D/formylglutathione hydrolase 1.356347441 0.004653 0.617484102 Ppara peroxisome proliferator activated receptor alpha -1.387511855 0.004653 0.617484102 Lrrc52 leucine rich repeat containing 52 -1.294598263 0.00466 0.617484102 Gm6177 predicted gene 6177 1.434365117 0.004666 0.617484102

Acss1 acyl-CoA synthetase short-chain family member 1 -1.156430822 0.004671 0.617484102 Plcg1 phospholipase C, gamma 1 -1.234599283 0.004688 0.617484102 Gm5746 predicted gene 5746 1.211650038 0.004743 0.61814212 transmembrane and tetratricopeptide repeat Tmtc1 containing 1 -1.316735014 0.004786 0.61814212 Heatr5a HEAT repeat containing 5A -1.394187727 0.004789 0.61814212 Cep68 centrosomal protein 68 -1.269423738 0.004809 0.61814212 Ube3b ubiquitin protein E3B -1.211297613 0.00491 0.623910626 Gm5532 predicted gene 5532 -1.433548572 0.00494 0.623910626 Thsd7a thrombospondin, type I, domain containing 7A -1.377583438 0.004968 0.623910626 Gm5514 predicted gene 5514 -3.581312612 0.004979 0.623910626 Arhgap26 Rho GTPase activating protein 26 -1.137589085 0.005021 0.623910626 protein phosphatase 1, regulatory (inhibitor) subunit Ppp1r12b 12B -1.247231423 0.005032 0.623910626 Rbp7 retinol binding protein 7, cellular -1.317388577 0.005087 0.623910626 Fkbp4 FK506 binding protein 4 1.274706489 0.00509 0.623910626 N4bp2l2 NEDD4 binding protein 2-like 2 -1.296169587 0.005133 0.625555556

Abcd3 ATP-binding cassette, sub-family D (ALD), member 3 -1.200480629 0.005194 0.627251221 Kpna6 karyopherin (importin) alpha 6 -1.148852298 0.005206 0.627251221 Epas1 endothelial PAS domain protein 1 -1.303355438 0.005311 0.63421211 Gm6128 H3 histone, family 3A pseudogene -1.214258425 0.005323 0.63421211 Bckdk branched chain ketoacid dehydrogenase kinase -1.231146365 0.005389 0.635215042 Lin37 lin-37 homolog (C. elegans) 1.333303343 0.005392 0.635215042 Fbxo9 f-box protein 9 -1.160079063 0.005481 0.637386969 acidic (leucine-rich) nuclear phosphoprotein 32 family, Anp32a member A 1.188274352 0.005496 0.637386969 RNA binding motif, single stranded interacting Rbms2 protein 2 -1.372658896 0.0055 0.637386969 Mavs mitochondrial antiviral signaling protein -1.415920292 0.005595 0.644842117 6720456H20Rik RIKEN cDNA 6720456H20 gene 1.276323567 0.00563 0.645390654 Lrrc48 leucine rich repeat containing 48 -1.507454564 0.005688 0.648480934 Fads3 fatty acid desaturase 3 -1.187094431 0.005737 0.650638499 1700066M21Rik RIKEN cDNA 1700066M21 gene 1.179193999 0.005874 0.654752584 Gm11263 predicted gene 11263 1.416648636 0.006016 0.654752584 Tbc1d23 TBC1 domain family, member 23 -1.381551864 0.006079 0.654752584 2410129H14Rik RIKEN cDNA 2410129H14 gene -1.222932809 0.006179 0.654752584 Nkx2-1 NK2 homeobox 1 -1.141261072 0.006179 0.654752584 Mfsd4 major facilitator superfamily domain containing 4 -1.183360482 0.00618 0.654752584 Rasl10b RAS-like, family 10, member B -1.577164221 0.006196 0.654752584 Ttc17 tetratricopeptide repeat domain 17 -1.175769039 0.00628 0.654752584 Gm10196 predicted gene 10196 1.785604734 0.006283 0.654752584 Mapk14 mitogen-activated protein kinase 14 -1.275672772 0.00633 0.654752584

152

Xlr X-linked lymphocyte-regulated complex -1.87615402 0.006351 0.654752584 2410091C18Rik RIKEN cDNA 2410091C18 gene 1.17336119 0.006356 0.654752584 Nsun5 NOL1/NOP2/Sun domain family, member 5 -1.135700302 0.006378 0.654752584 Tmprss13 transmembrane protease, serine 13 -1.785172719 0.006389 0.654752584 Pfdn5 prefoldin 5 1.380456459 0.006455 0.654752584 Bves blood vessel epicardial substance -1.169515361 0.006456 0.654752584 a disintegrin and metallopeptidase domain 9 (meltrin Adam9 gamma) -1.23427036 0.006551 0.654752584 Tmcc3 transmembrane and coiled coil domains 3 -1.174964126 0.006606 0.654752584 Cetn3 centrin 3 1.222346334 0.006617 0.654752584 heart and neural crest derivatives expressed Hand2 transcript 2 -1.304007845 0.006617 0.654752584 protein phosphatase 1, regulatory (inhibitor) Ppp1r13l subunit 13 like -1.302341982 0.006625 0.654752584 Tcerg1 transcription elongation regulator 1 (CA150) -1.152823799 0.00663 0.654752584 Galt galactose-1-phosphate uridyl transferase -1.629430471 0.00664 0.654752584 Larp7 La ribonucleoprotein domain family, member 7 1.137421794 0.006641 0.654752584 Rab11fip5 RAB11 family interacting protein 5 (class I) -1.198922574 0.006655 0.654752584 Slc35a4 solute carrier family 35, member A4 -1.293058606 0.006668 0.654752584 Nisch nischarin -1.193115489 0.006678 0.654752584 Zdhhc5 zinc finger, DHHC domain containing 5 -1.205386555 0.006685 0.654752584 Gm10054 predicted gene 10054 1.434197637 0.006722 0.654752584 Cdk11b cyclin-dependent kinase 11B 1.152935986 0.006724 0.654752584 Gm3550 predicted gene 3550 -1.696137499 0.006735 0.654752584 Srf serum response factor -1.280382109 0.006785 0.654752584 Defb23 defensin beta 23 1.398197425 0.00682 0.654752584 Dym dymeclin -1.177226042 0.006849 0.654752584 Fryl furry homolog-like (Drosophila) -1.214510979 0.006917 0.654752584 Gm7606 predicted gene 7606 1.316330192 0.006931 0.654752584 potassium voltage-gated channel, Shal-related Kcnd2 family, member 2 -1.286524569 0.006935 0.654752584

Xpot exportin, tRNA (nuclear export receptor for tRNAs) -1.225454554 0.006975 0.654752584 Eda ectodysplasin-A -1.289134984 0.006978 0.654752584 Zfp238 zinc finger protein 238 -1.152891187 0.007112 0.663262563 Nfia nuclear factor I/A -1.18342123 0.007231 0.663262563 Clec2f C-type lectin domain family 2, member f 1.159058641 0.007245 0.663262563 Olfr1076 olfactory receptor 1076 1.276127333 0.007257 0.663262563 Bdh1 3-hydroxybutyrate dehydrogenase, type 1 -1.942987613 0.007346 0.663262563 Nfe2l1 nuclear factor, erythroid derived 2,-like 1 -1.188868108 0.007406 0.663262563 Mettl2 methyltransferase like 2 -1.265838745 0.007445 0.663262563 Mtch1 mitochondrial carrier homolog 1 (C. elegans) -1.164432981 0.007446 0.663262563 Git2 G protein-coupled receptor kinase-interactor 2 -1.184805169 0.007524 0.663262563 Zdhhc8 zinc finger, DHHC domain containing 8 -1.354673897 0.007545 0.663262563 Mrpl16 mitochondrial ribosomal protein L16 -1.303818948 0.007559 0.663262563 Gm7867 predicted gene 7867 -1.406956077 0.007559 0.663262563 nuclear protein localization 4 homolog (S. Nploc4 cerevisiae) -1.259731794 0.007631 0.663262563 solute carrier family 6 (neurotransmitter transporter, Slc6a8 creatine), member 8 -1.214496849 0.007641 0.663262563

153

CTD (carboxy-terminal domain, RNA polymerase II, Ctdsp1 polypeptide A) small phosphatase 1 -1.170744492 0.007664 0.663262563 Phactr1 phosphatase and actin regulator 1 1.35233176 0.007673 0.663262563 Zfp318 zinc finger protein 318 -1.18877786 0.007708 0.663262563

Coq2 coenzyme Q2 homolog, prenyltransferase (yeast) -1.275238269 0.007716 0.663262563 Cct4 chaperonin containing Tcp1, subunit 4 (delta) 1.163622556 0.007717 0.663262563

Ubr2 ubiquitin protein ligase E3 component n-recognin 2 -1.155629778 0.007785 0.663262563 4933417A18Rik RIKEN cDNA 4933417A18 gene -1.127794151 0.007848 0.663262563

Eif3f eukaryotic translation initiation factor 3, subunit F 1.117319651 0.007857 0.663262563 Mtap9 microtubule-associated protein 9 1.161981414 0.007875 0.663262563 Tcra-V8 T-cell receptor alpha, variable 8 -1.140997752 0.007911 0.663262563 NADH dehydrogenase (ubiquinone) 1 alpha Ndufa1 subcomplex, 1 1.294728553 0.007915 0.663262563 Stk40 serine/threonine kinase 40 -1.364637556 0.007936 0.663262563 Atrnl1 attractin like 1 -1.266814104 0.007945 0.663262563 4933403F05Rik RIKEN cDNA 4933403F05 gene -1.25491501 0.008058 0.663262563 Ankrd16 ankyrin repeat domain 16 -1.191932268 0.008114 0.663262563 Habp4 hyaluronic acid binding protein 4 -1.175228755 0.008139 0.663262563 Nkiras2 NFKB inhibitor interacting Ras-like protein 2 -1.30574026 0.008142 0.663262563 Calr calreticulin 1.291440848 0.008146 0.663262563 Tmem41a transmembrane protein 41a -1.448373987 0.008157 0.663262563 Dirc2 disrupted in renal carcinoma 2 (human) -1.161626849 0.008163 0.663262563 Nr3c2 nuclear receptor subfamily 3, group C, member 2 -1.140216253 0.00825 0.663464908 Olfr1303 olfactory receptor 1303 1.404173797 0.008253 0.663464908 Nt5dc2 5'-nucleotidase domain containing 2 -1.325845618 0.008261 0.663464908

Ppm1k protein phosphatase 1K (PP2C domain containing) -1.213118407 0.008291 0.663464908 Gm7730 predicted gene 7730 1.16806136 0.008365 0.664973397 Gm15623 predicted gene 15623 1.811239178 0.008412 0.664973397 solute carrier family 25 (mitochondrial Slc25a20 carnitine/acylcarnitine ), member 20 -1.573847228 0.008414 0.664973397 Rnf139 ring finger protein 139 -1.361719319 0.008435 0.664973397 Ndrg4 N-myc downstream regulated gene 4 -2.143777941 0.008551 0.668849889 1700063H04Rik RIKEN cDNA 1700063H04 gene 1.147476467 0.008595 0.668849889 disease (autosomal recessive, early onset) Park7 7 1.140640886 0.008597 0.668849889 Gm6728 predicted gene 6728 -1.285475447 0.008657 0.668849889 Gm1943 WD repeat domain 70 pseudogene -1.530658209 0.008762 0.668849889 Fv1 Friend virus susceptibility 1 -1.662389989 0.008762 0.668849889 Pla2g4b phospholipase A2, group IVB (cytosolic) 1.199093966 0.008781 0.668849889 Tuba8 tubulin, alpha 8 -1.312287347 0.008784 0.668849889 Dnmt3a DNA methyltransferase 3A -1.151931053 0.008788 0.668849889 Churc1 churchill domain containing 1 1.335550422 0.008799 0.668849889 Churc1 churchill domain containing 1 1.335550422 0.008799 0.668849889 Ranbp6 RAN binding protein 6 -1.522158138 0.008872 0.670582038 1700112E06Rik RIKEN cDNA 1700112E06 gene -1.21859446 0.008915 0.670582038 Ttll1 tubulin tyrosine ligase-like 1 -1.131360828 0.008917 0.670582038

154

Mbnl2 muscleblind-like 2 -1.222025922 0.008994 0.671332439 Lpcat3 lysophosphatidylcholine acyltransferase 3 -1.206206939 0.008995 0.671332439 Lmtk2 lemur tyrosine kinase 2 -1.271139354 0.009093 0.671332439 Lars2 leucyl-tRNA synthetase, mitochondrial -1.682606516 0.009105 0.671332439 Smyd1 SET and MYND domain containing 1 -1.133183372 0.009128 0.671332439 Fem1c fem-1 homolog c (C.elegans) -1.194580233 0.009162 0.671332439 neural precursor cell expressed, developmentally Nedd8 down-regulated gene 8 1.235747739 0.009173 0.671332439 Defb14 defensin beta 14 1.191320318 0.00918 0.671332439 transforming growth factor, beta receptor Tgfbrap1 associated protein 1 -1.554282492 0.009299 0.675215203 Sec62 SEC62 homolog (S. cerevisiae) 1.400514722 0.009395 0.675215203 Iqsec1 IQ motif and Sec7 domain 1 -1.278595306 0.009413 0.675215203 Srd5a1 steroid 5 alpha-reductase 1 1.166401427 0.009437 0.675215203 Cnn1 calponin 1 1.402450546 0.009439 0.675215203

Adipoq adiponectin, C1Q and collagen domain containing 1.54351398 0.00945 0.675215203 Agpat3 1-acylglycerol-3-phosphate O-acyltransferase 3 -1.180357192 0.009456 0.675215203 2610507B11Rik RIKEN cDNA 2610507B11 gene -1.209788162 0.009511 0.676176787 Ptdss1 phosphatidylserine synthase 1 -1.19929526 0.009548 0.676176787 Cyr61 cysteine rich protein 61 1.357549119 0.009639 0.676176787 Gm15456 predicted gene 15456 -1.317739551 0.009658 0.676176787 Morf4l1 mortality factor 4 like 1 1.178725035 0.009677 0.676176787 Gm11868 predicted gene 11868 -1.19240268 0.009696 0.676176787 Acat1 acetyl-Coenzyme A acetyltransferase 1 -1.075447117 0.009721 0.676176787 4930562F07Rik RIKEN cDNA 4930562F07 gene -1.532311249 0.00974 0.676176787 Git1 G protein-coupled receptor kinase-interactor 1 -1.228949775 0.009798 0.676176787 Clcc1 CLIC-like 1 -1.215324256 0.009944 0.676176787 Fam78a family with sequence similarity 78, member A -1.318145694 0.00995 0.676176787 Hdgf hepatoma-derived growth factor -1.135153277 0.009983 0.676176787

Table 4. All genes identified by the microarray with significant up- or down-regulation in

HMWKO hearts compared to Fgf2 KO (n=3 hearts per group). Genes with changes

greater than 1.25-fold are bolded.

155

Genes differentially regulated in LMWKO vs. Fgf2 KO False Discovery Abbreviation Gene Name Fold Change P Value Rate (FDR) Rps11 ribosomal protein S11 -4.484652376 9.51E-08 0.002017657 Sun3 Sad1 and UNC84 domain containing 3 5.50144906 9.07E-06 0.096162765 hypoxanthine guanine phosphoribosyl Hprt transferase -2.023017705 8.96E-05 0.542042931 Acyp2 acylphosphatase 2, muscle type -1.437578209 0.000102 0.542042931 D3Ertd751e DNA segment, Chr 3, ERATO Doi 751, expressed -2.16837808 0.00015 0.638331279 Fgf2 fibroblast growth factor 2 1.799731866 0.000183 0.647656398 Iqgap2 IQ motif containing GTPase activating protein 2 2.49247374 0.000263 0.795532178 branched chain aminotransferase 2, Bcat2 mitochondrial 1.536562203 0.000307 0.813881996 processing of precursor 1, ribonuclease P/MRP Pop1 family, (S. cerevisiae) 1.325335403 0.000442 0.897381466 Gm5599 predicted gene 5599 -1.585377955 0.000523 0.897381466 Ddah1 dimethylarginine dimethylaminohydrolase 1 -1.277850903 0.000544 0.897381466 Olfr19 olfactory receptor 19 -1.22957763 0.000633 0.897381466 Mcrs1 microspherule protein 1 -2.011551144 0.000914 0.897381466 Egln3 EGL nine homolog 3 (C. elegans) -1.475149559 0.000963 0.897381466 Fcgrt Fc receptor, IgG, alpha chain transporter -1.272991945 0.00102 0.897381466 OTTMUSG00000018964 predicted gene, OTTMUSG00000018964 1.412864871 0.001036 0.897381466 Tuft1 tuftelin 1 -1.355337645 0.001105 0.897381466 Nup62 nucleoporin 62 -1.505763725 0.001349 0.897381466 Slc6a16 solute carrier family 6, member 16 1.999284373 0.001351 0.897381466 Gm11868 predicted gene 11868 -1.277742246 0.001562 0.897381466 Rtp4 receptor transporter protein 4 1.222817227 0.001845 0.897381466 Gm6124 predicted gene 6124 -3.571177278 0.001905 0.897381466 Epm2aip1 EPM2A (laforin) interacting protein 1 -1.397088721 0.00199 0.897381466 Ugdh UDP-glucose dehydrogenase 1.235699557 0.002032 0.897381466 Olfr993 olfactory receptor 993 1.147287542 0.002186 0.897381466 Hist2h3c1 histone cluster 2, H3c1 2.87988259 0.002239 0.897381466 Vmn2r54 vomeronasal 2, receptor 54 -1.419367618 0.002296 0.897381466 Srf serum response factor -1.200911237 0.002317 0.897381466 Gm6905 predicted gene 6905 1.379375654 0.002398 0.897381466 Fam82a1 family with sequence similarity 82, member A1 -1.273788764 0.00256 0.897381466 Gm5415 predicted gene 5415 -1.329908744 0.00265 0.897381466 Olfr685 olfactory receptor 685 -1.159367367 0.002811 0.897381466 Mup20 major urinary protein 20 -1.770287044 0.002884 0.897381466 Gspt1 G1 to S phase transition 1 -1.212110894 0.002987 0.897381466 Snapin SNAP-associated protein -1.27630921 0.003059 0.897381466 Rbp7 retinol binding protein 7, cellular -1.350575717 0.003144 0.897381466 Cpxm2 carboxypeptidase X 2 (M14 family) 1.497374071 0.003247 0.897381466 Vmn2r4 vomeronasal 2, receptor 4 -1.240039108 0.003251 0.897381466 0610009B22Rik RIKEN cDNA 0610009B22 gene -1.15749688 0.003339 0.897381466 Tmx4 thioredoxin-related transmembrane protein 4 -1.148836777 0.003375 0.897381466 BC080695 cDNA sequence BC080695 -1.390841152 0.003453 0.897381466

156

STT3, subunit of the oligosaccharyltransferase Stt3b complex, homolog B (S. cerevisiae) -1.25232092 0.003505 0.897381466 guanine nucleotide binding protein (G protein), beta Gnb5 5 -1.20622492 0.003534 0.897381466 Fam59a family with sequence similarity 59, member A -1.313002327 0.003698 0.897381466 Tmprss13 transmembrane protease, serine 13 -1.504801259 0.003738 0.897381466 Rbpms2 RNA binding protein with multiple splicing 2 -1.341288737 0.00374 0.897381466 CCR4 carbon catabolite repression 4-like (S. Ccrn4l cerevisiae) -1.226256638 0.003962 0.897381466 Gm4841 predicted gene 4841 1.315796058 0.00399 0.897381466 Crybg3 beta-gamma crystallin domain containing 3 -1.442601152 0.004025 0.897381466 Mcam melanoma cell adhesion molecule 1.333644829 0.004055 0.897381466 potassium voltage-gated channel, subfamily H Kcnh2 (eag-related), member 2 -1.262463629 0.004115 0.897381466 Zbtb37 zinc finger and BTB domain containing 37 -1.239573753 0.004118 0.897381466 solute carrier family 16 (monocarboxylic acid Slc16a1 transporters), member 1 -1.20838543 0.004262 0.897381466 Gm5617 predicted gene 5617 -1.201351719 0.004346 0.897381466 Ethe1 ethylmalonic encephalopathy 1 1.168334361 0.00437 0.897381466 1700112E06Rik RIKEN cDNA 1700112E06 gene -1.221399287 0.004397 0.897381466 Gca grancalcin -1.285631429 0.004406 0.897381466 Gde1 glycerophosphodiester phosphodiesterase 1 1.343552851 0.004421 0.897381466 non imprinted in Prader-Willi/Angelman Nipa2 syndrome 2 homolog (human) -1.314796708 0.004433 0.897381466 Dnase1l3 deoxyribonuclease 1-like 3 1.200116168 0.0045 0.897381466 Sms spermine synthase -1.164950639 0.004628 0.897381466 Ak3l1 adenylate kinase 3-like 1 -1.191919381 0.004735 0.897381466 Nkx2-1 NK2 homeobox 1 -1.231363266 0.004803 0.897381466 Olfr1013 olfactory receptor 1013 -1.228859853 0.004857 0.897381466 Bend6 BEN domain containing 6 -1.530161304 0.004863 0.897381466 Btbd1 BTB (POZ) domain containing 1 -1.302957443 0.004938 0.897381466 Lrg1 leucine-rich alpha-2-glycoprotein 1 -1.53650583 0.004959 0.897381466 Krt73 keratin 73 -1.207454456 0.005099 0.897381466 Gm4977 predicted gene 4977 1.4837636 0.005456 0.897381466 Gm5453 predicted gene 5453 1.380126628 0.005506 0.897381466 Olfr125 olfactory receptor 125 -1.177065255 0.005566 0.897381466 4933427G23Rik RIKEN cDNA 4933427G23 gene -1.167669021 0.005578 0.897381466 Rnf139 ring finger protein 139 -1.278420505 0.005593 0.897381466 Klk1 kallikrein 1 -1.180340571 0.00565 0.897381466 Higd1a HIG1 domain family, member 1A -2.141198271 0.005678 0.897381466 Lrrc48 leucine rich repeat containing 48 -1.447162344 0.00569 0.897381466 Etl4 enhancer trap 4 -1.485527669 0.005869 0.897381466 Adrm1 adhesion regulating molecule 1 1.249994825 0.005889 0.897381466 Igkv1-110 immunoglobulin kappa chain variable 1-110 -1.571158274 0.005948 0.897381466 Lce1h late cornified envelope 1H -1.259222981 0.006039 0.897381466 Tm9sf2 transmembrane 9 superfamily member 2 -1.107156706 0.006071 0.897381466 Gm2095 predicted gene 2095 -1.685428042 0.006113 0.897381466 Tiparp TCDD-inducible poly(ADP-ribose) polymerase -1.266369896 0.006145 0.897381466 Ptp4a2 protein tyrosine phosphatase 4a2 -1.113621646 0.006245 0.897381466 Tmem56 transmembrane protein 56 1.61868596 0.006312 0.897381466

157

Cyr61 cysteine rich protein 61 1.209434056 0.006344 0.897381466 Cbln3 cerebellin 3 precursor protein 1.168286023 0.006432 0.897381466 glycerol-3-phosphate acyltransferase 2, Gpat2 mitochondrial -1.12429359 0.006435 0.897381466 Klk1b22 kallikrein 1-related peptidase b22 -1.278693683 0.006437 0.897381466 Nanos3 nanos homolog 3 (Drosophila) -1.168826408 0.00645 0.897381466 glucose-fructose oxidoreductase domain containing Gfod1 1 -1.209393735 0.006509 0.897381466 Calr calreticulin 1.286831909 0.006678 0.897381466 Pnrc1 proline-rich nuclear receptor coactivator 1 1.216047674 0.006681 0.897381466 E2f6 E2F transcription factor 6 -1.167293751 0.006687 0.897381466 a disintegrin-like and metallopeptidase (reprolysin type) with thrombospondin type 1 Adamts6 motif, 6 1.466530267 0.006697 0.897381466 spermatogenesis associated glutamate (E)-rich Speer3 protein 3 -1.188539733 0.00677 0.897381466 Pcdhb1 protocadherin beta 1 -1.208019909 0.006909 0.897381466 Gm3756 predicted gene 3756 -1.934665659 0.006916 0.897381466 Fam20b family with sequence similarity 20, member B -1.448743932 0.006967 0.897381466 Dfnb59 deafness, autosomal recessive 59 (human) -1.156828852 0.007018 0.897381466 C330024D12Rik RIKEN cDNA C330024D12 gene 1.154385264 0.007062 0.897381466 Tmprss12 transmembrane protease, serine 12 -1.137405083 0.00707 0.897381466 Rnmtl1 RNA methyltransferase like 1 1.205398849 0.007086 0.897381466 Cela2a chymotrypsin-like elastase family, member 2A -1.214395444 0.007102 0.897381466 Ngf nerve growth factor 1.405213051 0.007227 0.897381466 Il6ra interleukin 6 receptor, alpha 1.141040515 0.007643 0.897381466 Slc38a10 solute carrier family 38, member 10 1.208012728 0.00769 0.897381466 Fkbp5 FK506 binding protein 5 -1.394827267 0.007802 0.897381466 CWC22 spliceosome-associated protein homolog Cwc22 (S. cerevisiae) 1.441868188 0.007813 0.897381466 Onecut3 one cut domain, family member 3 -1.139440728 0.007848 0.897381466 Gm5963 predicted gene 5963 1.268843304 0.007856 0.897381466 2310047M10Rik RIKEN cDNA 2310047M10 gene -1.260445046 0.008079 0.897381466 C3 and PZP-like, alpha-2-macroglobulin domain Cpamd8 containing 8 -1.216518491 0.008115 0.897381466 Gm2524 predicted gene 2524 1.366668961 0.008116 0.897381466 Rnps1 ribonucleic acid binding protein S1 -1.157906945 0.008208 0.897381466 solute carrier family 17 (sodium-dependent Slc17a7 inorganic phosphate cotransporter), member 7 1.286588788 0.008362 0.897381466 spindle and kinetochore associated complex subunit Ska2l 2-like 1.162939952 0.008514 0.897381466 Eef2 eukaryotic translation elongation factor 2 1.096060902 0.00852 0.897381466 Cenpf centromere protein F 2.03102514 0.008663 0.897381466 Gm614 predicted gene 614 -1.302823802 0.008882 0.897381466 Tm7sf3 transmembrane 7 superfamily member 3 1.212973297 0.008903 0.897381466 A530082C11Rik RIKEN cDNA A530082C11 gene 1.259965627 0.008948 0.897381466 Mtap1a microtubule-associated protein 1 A 1.373507451 0.008955 0.897381466 Olfr68 olfactory receptor 68 -1.223635709 0.009062 0.897381466 Itpr2 inositol 1,4,5-triphosphate receptor 2 1.183635127 0.009291 0.897381466 Olfr1162 olfactory receptor 1162 -1.86100182 0.009739 0.897381466

158

ST6 (alpha-N-acetyl-neuraminyl-2,3-beta- galactosyl-1,3)-N-acetylgalactosaminide alpha- St6galnac2 2,6-sialyltransferase 2 1.395283052 0.009949 0.897381466 Luzp2 leucine zipper protein 2 1.193491562 0.009971 0.897381466

Table 5. All genes identified by the microarray with significant up- or down-regulation in

LMWKO hearts compared to Fgf2 KO (n=3 hearts per group). Genes with changes greater

than 1.25-fold are bolded.

159

Genes differentially regulated in HMWKO vs. LMWKO False Discovery Abbreviation Gene Name Fold Change P Value Rate (FDR) glycerophosphodiester Gdpd3 phosphodiesterase domain containing 3 11.43419269 1.39E-05 0.162299983 Rps11 ribosomal protein S11 -4.415094212 1.53E-05 0.162299983 Btbd1 BTB (POZ) domain containing 1 -1.368399806 0.000183 0.821048529 Ank progressive ankylosis 1.301304267 0.000196 0.821048529 Clic5 chloride intracellular channel 5 1.284218772 0.00021 0.821048529 Rhox4f reproductive homeobox 4F -1.450795598 0.000232 0.821048529 Klk1 kallikrein 1 -1.219648398 0.000352 0.858212681 Gm5437 ribosomal protein L31 pseudogene -1.752413397 0.00065 0.858212681 carcinoembryonic antigen-related cell Ceacam2 adhesion molecule 2 1.6153357 0.000729 0.858212681 Trak2 trafficking protein, kinesin binding 2 1.306216683 0.000792 0.858212681 Hddc3 HD domain containing 3 1.562337306 0.000888 0.858212681 Gm13226 predicted gene 13226 -1.432546132 0.000905 0.858212681 protein tyrosine phosphatase, non- Ptpn11 receptor type 11 1.263244496 0.000915 0.858212681 Stk40 serine/threonine kinase 40 1.255119088 0.000971 0.858212681 Fah fumarylacetoacetate hydrolase 1.840029836 0.001052 0.858212681 1700018F24Rik RIKEN cDNA 1700018F24 gene -1.255359838 0.001078 0.858212681 Gm5599 predicted gene 5599 -1.378239596 0.001171 0.858212681 Zfp30 zinc finger protein 30 -2.552286702 0.001206 0.858212681 Epas1 endothelial PAS domain protein 1 1.314550606 0.001386 0.858212681 Wolf-Hirschhorn syndrome candidate 2 Whsc2 (human) 1.256623466 0.00143 0.858212681 Alport syndrome, mental retardation, midface hypoplasia and elliptocytosis chromosomal region gene 1 homolog Ammecr1 (human) 1.318808453 0.001451 0.858212681 Ephb1 Eph receptor B1 2.238505809 0.001469 0.858212681 Itpr2 inositol 1,4,5-triphosphate receptor 2 1.566678905 0.001646 0.858212681 Nceh1 arylacetamide deacetylase-like 1 1.404514374 0.001673 0.858212681 acyl-CoA synthetase family member 2 Gm5540 pseudogene -1.722266447 0.002097 0.858212681 Fnip1 folliculin interacting protein 1 1.255695411 0.002262 0.858212681 Filip1l filamin A interacting protein 1-like -1.385701918 0.002372 0.858212681 1700019E19Rik RIKEN cDNA 1700019E19 gene -1.170134873 0.002425 0.858212681 DNA segment, Chr 3, ERATO Doi 751, D3Ertd751e expressed -1.89458324 0.002427 0.858212681 Ssbp2 single-stranded DNA binding protein 2 1.199740987 0.002493 0.858212681 Pm20d1 peptidase M20 domain containing 1 1.427796202 0.0025 0.858212681 Stfa2l1 stefin A2 like 1 1.205407936 0.002562 0.858212681 Gm5617 predicted gene 5617 1.429221574 0.002589 0.858212681 Zfp592 zinc finger protein 592 1.239597009 0.002692 0.858212681 glycerol-3-phosphate acyltransferase, Gpam mitochondrial 1.34915218 0.002732 0.858212681 mediator of RNA polymerase II transcription, Med12 subunit 12 homolog (yeast) 1.174629896 0.002816 0.858212681

160

Heatr5a HEAT repeat containing 5A 1.36605341 0.002838 0.858212681 Fem1a feminization 1 homolog a (C. elegans) 1.26332972 0.003108 0.858212681 Slc6a16 solute carrier family 6, member 16 1.8848089 0.003235 0.858212681 4833439L19Rik RIKEN cDNA 4833439L19 gene 1.19433843 0.003261 0.858212681 Acer3 alkaline ceramidase 3 1.30720551 0.003373 0.858212681 angiogenin, ribonuclease, RNase A Ang family, 5 -1.932700924 0.003458 0.858212681 Lcn2 lipocalin 2 1.215220632 0.003562 0.858212681 Defb12 defensin beta 12 -1.741940933 0.003642 0.858212681 N-deacetylase/N-sulfotransferase (heparan Ndst1 glucosaminyl) 1 1.183162984 0.00366 0.858212681 Serf1 small EDRK-rich factor 1 -1.398993605 0.003703 0.858212681 Fcgrt Fc receptor, IgG, alpha chain transporter -1.326491129 0.003721 0.858212681 ubiquitin protein ligase E3 component n- Ubr2 recognin 2 1.179481942 0.00374 0.858212681 H2-T22 histocompatibility 2, T region locus 22 -1.521812851 0.003766 0.858212681 Rxra retinoid X receptor alpha 1.165090408 0.003832 0.858212681 Nisch nischarin 1.12730081 0.003851 0.858212681 Map2k3 mitogen-activated protein kinase kinase 3 1.216991167 0.004008 0.858212681 xeroderma pigmentosum, complementation Xpa group A 1.18221605 0.004055 0.858212681 Gm3744 predicted gene 3744 1.20616376 0.004058 0.858212681 Dag1 dystroglycan 1 1.24846917 0.004102 0.858212681 Cpxm2 carboxypeptidase X 2 (M14 family) 1.958499913 0.00416 0.858212681 Gm6843 predicted gene 6843 -1.327645185 0.004255 0.858212681 Il6ra interleukin 6 receptor, alpha 1.220963078 0.004284 0.858212681 TAF10 RNA polymerase II, TATA box Taf10 binding protein (TBP)-associated factor 1.530327299 0.004364 0.858212681 Asb2 ankyrin repeat and SOCS box-containing 2 1.232599252 0.004503 0.858212681 2410017I17Rik RIKEN cDNA 2410017I17 gene -1.611708593 0.004543 0.858212681 Rptn repetin -1.207262351 0.004622 0.858212681 adaptor-related protein complex AP-1, Gm8532 sigma 3 pseudogene -1.582101122 0.004629 0.858212681 Onecut3 one cut domain, family member 3 -1.195063702 0.004636 0.858212681 Lmtk2 lemur tyrosine kinase 2 1.211720719 0.004863 0.858212681 cytoplasmic polyadenylation element binding Cpeb4 protein 4 1.22622374 0.00501 0.858212681 UDP-Gal:betaGlcNAc beta 1,3- B3galt4 galactosyltransferase, polypeptide 4 -1.455064151 0.005023 0.858212681 Exoc5 exocyst complex component 5 1.293098987 0.005067 0.858212681 Fryl furry homolog-like (Drosophila) 1.147357194 0.005196 0.858212681 Gm9625 predicted gene 9625 -1.686411915 0.005212 0.858212681 Olfr1221 olfactory receptor 1221 1.147948875 0.005316 0.858212681 Lrrc52 leucine rich repeat containing 52 1.222333155 0.005359 0.858212681 Zfp445 zinc finger protein 445 1.258254673 0.005447 0.858212681 Pla2g2d phospholipase A2, group IID -1.188823276 0.005499 0.858212681 Rnf38 ring finger protein 38 1.234669072 0.005511 0.858212681 hairy/enhancer-of-split related with YRPW Hey1 motif 1 1.406255261 0.005532 0.858212681 4430402I18Rik RIKEN cDNA 4430402I18 gene -1.297762973 0.005637 0.858212681 ATPase, Na+/K+ transporting, alpha 2 Atp1a2 polypeptide 1.184262414 0.005709 0.858212681 161

Pcdhb1 protocadherin beta 1 -1.26936259 0.005749 0.858212681 succinate-Coenzyme A ligase, ADP-forming, Sucla2 beta subunit -1.206666655 0.005772 0.858212681 Glrx glutaredoxin 1.154335301 0.005799 0.858212681 1600029D21Rik RIKEN cDNA 1600029D21 gene -1.451493622 0.005802 0.858212681 Tmod4 tropomodulin 4 -1.406236547 0.005813 0.858212681 Nup62 nucleoporin 62 -1.304997728 0.005815 0.858212681 protein phosphatase 1, regulatory (inhibitor) Ppp1r12b subunit 12B 1.177155918 0.005828 0.858212681 Ethe1 ethylmalonic encephalopathy 1 1.244744579 0.005875 0.858212681 Tbc1d23 TBC1 domain family, member 23 1.322597844 0.005927 0.858212681 Pfkp phosphofructokinase, platelet 1.850536796 0.005932 0.858212681 A630052C17Ri k RIKEN cDNA A630052C17 gene 1.251130619 0.005936 0.858212681 Sec24 related gene family, member C (S. Sec24c cerevisiae) 1.153787669 0.005944 0.858212681 Rho guanine nucleotide exchange factor Arhgef7 (GEF7) 1.147091893 0.006101 0.858212681 ATP-binding cassette, sub-family B Abcb4 (MDR/TAP), member 4 1.198595352 0.006125 0.858212681 acyl-Coenzyme A dehydrogenase family, Acad11 member 11 1.209725645 0.006164 0.858212681 Ccnb1ip1 cyclin B1 interacting protein 1 1.301973877 0.006184 0.858212681 Dym dymeclin 1.1653212 0.006226 0.858212681 branched chain aminotransferase 2, Bcat2 mitochondrial 1.535179967 0.006271 0.858212681 RNA binding motif, single stranded Rbms2 interacting protein 2 1.212483372 0.006372 0.858212681 Rplp1 ribosomal protein, large, P1 1.339826915 0.0064 0.858212681 Cpne1 copine I 1.284209548 0.006429 0.858212681 Phf11 PHD finger protein 11 -1.264965616 0.006451 0.858212681 purinergic receptor P2Y, G-protein P2ry1 coupled 1 1.381184662 0.006501 0.858212681 BC080695 cDNA sequence BC080695 -1.347752143 0.006502 0.858212681 MTOR associated protein, LST8 homolog Mlst8 (S. cerevisiae) 1.324998776 0.006561 0.858212681 Lifr leukemia inhibitory factor receptor 1.197790244 0.006699 0.858212681 Dync2h1 dynein cytoplasmic 2 heavy chain 1 1.165029483 0.006718 0.858212681 4932416K20Rik RIKEN cDNA 4932416K20 gene -1.174403728 0.006766 0.858212681 Slc25a40 solute carrier family 25, member 40 1.430345317 0.006859 0.858212681 Prep prolyl endopeptidase 1.214131906 0.006888 0.858212681 1110018H23Rik RIKEN cDNA 1110018H23 gene -1.19778266 0.006916 0.858212681 Olfr627 olfactory receptor 627 1.225428115 0.007067 0.858212681 1700069L16Rik RIKEN cDNA 1700069L16 gene -1.478598472 0.007076 0.858212681 Gm10448 predicted gene 10448 1.808838031 0.007198 0.858212681 Sf1 splicing factor 1 1.180963794 0.007229 0.858212681 Gm2701 predicted gene 2701 1.262097904 0.007292 0.858212681 transmembrane and tetratricopeptide repeat Tmtc1 containing 1 1.245996167 0.007429 0.858212681 Cnih4 cornichon homolog 4 (Drosophila) -1.490290808 0.007526 0.858212681 Trim12 tripartite motif-containing 12 -1.98326074 0.007585 0.858212681 Bckdk branched chain ketoacid dehydrogenase 1.272982288 0.007586 0.858212681

162

kinase a disintegrin and metallopeptidase domain 9 Adam9 (meltrin gamma) 1.23124514 0.007646 0.858212681 twinfilin, actin-binding protein, homolog 2 Twf2 (Drosophila) -1.253284052 0.007886 0.858212681 Gm5611 predicted gene 5611 -1.997279121 0.007901 0.858212681 Plcg1 phospholipase C, gamma 1 1.237208115 0.007924 0.858212681 zinc finger and BTB domain containing Zbtb16 16 1.495141623 0.007987 0.858212681 Atf7 activating transcription factor 7 1.14857627 0.007994 0.858212681 Olfr446 olfactory receptor 446 -1.197659746 0.007999 0.858212681 nuclear protein localization 4 homolog (S. Nploc4 cerevisiae) 1.309289176 0.008047 0.858212681 2410006H16Rik RIKEN cDNA 2410006H16 gene -1.452867866 0.008125 0.858212681 Bves blood vessel epicardial substance 1.123866468 0.008158 0.858212681 Olfr68 olfactory receptor 68 -1.223999953 0.008211 0.858212681 Man2a2 mannosidase 2, alpha 2 1.258071659 0.008356 0.858212681 Phactr1 phosphatase and actin regulator 1 -1.289620763 0.008421 0.858212681 Slc38a10 solute carrier family 38, member 10 1.230139718 0.008474 0.858212681 Ighv1-26 immunoglobulin heavy variable V1-26 1.189294333 0.008522 0.858212681 Zfp281 zinc finger protein 281 1.115717724 0.008573 0.858212681 Zdhhc8 zinc finger, DHHC domain containing 8 1.265638986 0.00862 0.858212681 Vac14 Vac14 homolog (S. cerevisiae) 1.171236001 0.008715 0.858212681 Cenpv centromere protein V 1.222321339 0.008852 0.858212681 Gm16505 predicted gene 16505 -1.245665002 0.008993 0.858212681 1700066M21Ri k RIKEN cDNA 1700066M21 gene -1.330492521 0.009017 0.858212681 proteasome (prosome, macropain) Psma3 subunit, alpha type 3 1.360410886 0.009039 0.858212681 Gm889 predicted gene 889 -1.275269125 0.009155 0.858212681 Ptchd3 patched domain containing 3 1.513664617 0.009391 0.858212681 Vmn2r4 vomeronasal 2, receptor 4 1.339355096 0.009523 0.858212681 Pvr poliovirus receptor 1.398894656 0.009559 0.858212681 G protein-coupled receptor kinase- Git1 interactor 1 1.254761958 0.009586 0.858212681 carboxymethylenebutenolidase-like Cmbl (Pseudomonas) 2.003339718 0.009597 0.858212681 Gm7774 predicted gene 7774 -1.264138763 0.00961 0.858212681 Pah hydroxylase 1.3893744 0.00964 0.858212681 Dfnb59 deafness, autosomal recessive 59 (human) -1.185537757 0.00975 0.858212681 Ccdc71 coiled-coil domain containing 71 1.181932232 0.009765 0.858212681 Ttc17 tetratricopeptide repeat domain 17 1.188358133 0.009786 0.858212681 Midn midnolin 1.397447975 0.009935 0.858212681 Myot myotilin -1.387167603 0.009942 0.858212681

Table 6. All genes identified by the microarray with significant up- or down-regulation) in

LMWKO hearts compared to HMWKO (n=3 hearts per group). Genes with changes

greater than 1.25-fold are bolded.

163

164

Figure 37. Schematic representing the genes identified by the microarray. A total of 100 genes were altered between LMWKO and HMWKO hearts, 308 between HMWKO and

FGF2 KO hearts, and 128 between LMWKO and FGF2 KO. A total of 35 genes were common to all three comparisons.

165

HMWKO fold LMWKO fold Symbol Gene name change change Hprt hypoxanthine guanine phosphoribosyl transferase -2.024429969 -2.023017705 Fgf2 fibroblast growth factor 2 1.90985098 1.799731866

Pop1 processing of precursor 1, ribonuclease P/MRP family, (S. cerevisiae) 1.368352385 1.325335403 Egln3 EGL nine homolog 3 (C. elegans) -1.715069004 -1.475149559 Gm11868 predicted gene 11868 -1.19240268 -1.277742246 Gm6124 predicted gene 6124 -3.02440824 -3.571177278 Srf serum response factor -1.280382109 -1.200911237 Rbp7 retinol binding protein 7, cellular -1.317388577 -1.350575717 Vmn2r4 vomeronasal 2, receptor 4 -1.660852699 -1.240039108 Tmx4 thioredoxin-related transmembrane protein 4 -1.230505766 -1.148836777 STT3, subunit of the oligosaccharyltransferase complex, homolog B (S. Stt3b cerevisiae) -1.250144511 -1.25232092 Gnb5 guanine nucleotide binding protein (G protein), beta 5 -1.237728968 -1.20622492 Fam59a family with sequence similarity 59, member A -1.334663757 -1.313002327 Tmprss13 transmembrane protease, serine 13 -1.785172719 -1.504801259 Rbpms2 RNA binding protein with multiple splicing 2 -1.555103123 -1.341288737 Ccrn4l CCR4 carbon catabolite repression 4-like (S. cerevisiae) -1.305997507 -1.226256638 Crybg3 beta-gamma crystallin domain containing 3 -1.27081882 -1.442601152 Mcam melanoma cell adhesion molecule 1.403997089 1.333644829 Zbtb37 zinc finger and BTB domain containing 37 -1.311761301 -1.239573753 Slc16a1 solute carrier family 16 (monocarboxylic acid transporters), member 1 -1.313311022 -1.20838543 Gm5617 predicted gene 5617 -1.716997795 -1.201351719 1700112E06Rik RIKEN cDNA 1700112E06 gene -1.21859446 -1.221399287 Gca grancalcin -1.3960236 -1.285631429 Gde1 glycerophosphodiester phosphodiesterase 1 1.365307188 1.343552851 Nkx2-1 NK2 homeobox 1 -1.141261072 -1.231363266 Lrg1 leucine-rich alpha-2-glycoprotein 1 -1.61080769 -1.53650583 Rnf139 ring finger protein 139 -1.361719319 -1.278420505 Higd1a HIG1 domain family, member 1A -2.999052135 -2.141198271 Lrrc48 leucine rich repeat containing 48 -1.507454564 -1.447162344 Igkv1-110 immunoglobulin kappa chain variable 1-110 -1.513774648 -1.571158274 Ptp4a2 protein tyrosine phosphatase 4a2 -1.133881907 -1.113621646 Cyr61 cysteine rich protein 61 1.357549119 1.209434056 Gfod1 glucose-fructose oxidoreductase domain containing 1 -1.242747725 -1.209393735 Calr calreticulin 1.291440848 1.286831909 Fam20b family with sequence similarity 20, member B -1.631731963 -1.448743932

Table 7. Genes identified by the microarray common to both LMW and HMW expression

only, when compared to Fgf2 KO. Evaluation of non-ischemic Fgf2 KO (no expression of

FGF2 isoforms), HMWKO (expression of only LMW isoform), and LMWKO (expression

of only HMW isoform) (n=3 hearts per group).

166

False Discovery Abbreviation Gene Name Comparison Fold Change P Value Rate (FDR) Category HMWKO vs. Sln sarcolipin FGFKO 13.55439298 4.04E-05 0.202807027 Calcium handling myosin, light polypeptide 7, HMWKO vs. Myl7 regulatory FGFKO 31.12418683 7.87E-07 0.01668477 Contractile apparatus myosin, light HMWKO vs. Myl4 polypeptide 4 FGFKO 27.0730983 1.67E-05 0.118256157 Contractile apparatus LMWKO vs. Tmod4 tropomodulin 4 HMWKO -1.406236547 0.005813 0.858212681 Contractile apparatus fibroblast growth HMWKO vs. Fgf12 factor 12 FGFKO 5.134264575 0.000921 0.571342421 Ion homeostasis HMWKO vs. FGFKO 1.291440848 0.008146 0.663262563 LMWKO vs. Calcium handling, Calr calreticulin FGFKO 1.286831909 0.006678 0.897381466 survival

Table 8. Genes identified by the microarray as having different expression between Fgf2

KO, LMWKO, and HMWKO, which fall into the biological categories likely to impact

cardiac function during I/R injury.

167

A. Sarcolipin mRNA expression (microarray) * 16 *p<0.05 vs. FDR = .897 14 FGFKO 12 10 8 6 4 2

Relative mRNA expression mRNA Relative 0 FGFKOFGFKO HMWKOHMWKO B. Sarcolipin mRNA expression (qRT-PCR) 2500 * *p<0.05 vs. FGFKO 2000

1500

1000

500

0 FGFKOFGFKO HMWKOHM WKO

mRNA expression (arbitrary units) (arbitrary expression mRNA

Figure 38. Sarcolipin mRNA expression as determined by microarray (A) and qRT-PCR

(B) in non-ischemic hearts only expressing LMW FGF2 compared to Fgf2 KO hearts. FDR is given as a measure of error (FDR = false discovery rate). *p<0.05. (n=3 hearts per group).

168

Calreticulin mRNA expression (microarray)

** 1.4 *p<0.05 vs. FDR = .663 FDR = .897 FGFKO FDR= 1.2

1

0.8

0.6

0.4

Relative expression mRNA 0.2

0 FGFKOFGFKO HMWKOHMWKO LMWKOLMWKO

Figure 39. Calreticulin mRNA expression as determined by in non-ischemic hearts only expressing LMW FGF2 or HMW FGF2 compared to Fgf2 KO hearts. FDR is given as a measure of error (FDR = false discovery rate). *p<0.05. (n=3 hearts per group).

169

A. Calreticulin protein expression

2 FGFKO 1.8 FGFKO HMWKO HMWKO 1.6 Calreticulin

1.4

1.2 CSQ

1

0.8

0.6

0.4

0.2 Normalized calsequestrin (arbtrary units) 0 B. 1 FGFKOFGFKO HMWKOHM WKO LMWKO LMWKO FGFKO 0.9 FGFKO

0.8 Calreticulin 0.7 CSQ 0.6

0.5

0.4

0.3

0.2

Normalized calsequestrin (arbtrary units) (arbtrary calsequestrin Normalized 0.1

0 FGFKOFGFKO LMWKOLMWKO

Figure 40. Representative immunoblot and quantitation of calreticulin protein expression in non-ischemic hearts expressing no isoforms of FGF2 (Fgf2 KO), compared to hearts only expressing LMW FGF2 (HMWKO) (A) or HMW FGF2 (LMWKO) (B). Calreticulin was normalized to calsquestrin. (n=3).

170

Myosin light chain 4 and 7 mRNA expression (microarray)

* 35 FGFKO FDR = .0167 *p<0.05 vs. HMWKO FGFKO * 30 FDR = .118 25

20

15

10

5

0 FGFKO HMWKO

Relative mRNA expression Myl 7 Myl 4

Figure 41. Myosin light chains 7 and 4 mRNA expression as determined by microarray in non-ischemic hearts only expressing LMW FGF2 compared to Fgf2 KO hearts. FDR is given as a measure of error (FDR = false discovery rate). *p<0.05. (n=3 hearts per group).

171

FGFKO HMWKO

Myl CSQ

Myosin light chains protein expression 1.4 *p<0.05 vs. FGFKO *

1.2

1

0.8

0.6

0.4 Normalized Myl (arbitrary units) 0.2

0 FGFKO HM WKO

Figure 42. Myosin light chains protein expression (corresponding to clone MY-21), as determined by immunoblot, in non-ischemic hearts only expressing LMW FGF2

(HMWKO) or no FGF2 isoforms (Fgf2 KO)*p<0.05 vs Fgf2 KO. (n=7-9 per group).

172

Targets of mRNAs Biological Role in the identified nGenes zScore pValue heart Change Contributes to cell death and infarct development Downregulated in HMWKO MIR-320 230 2.9398 0.001642 (Ren et al., 2009) vs. FGFKO Upregulated in viable tissue in I/R (Dong et al., Downregulated in HMWKO MIR-21 113 2.9059 0.001831 2009) vs. FGFKO MIR-15A, MIR-16, Fetal development and MIR-15B, MIR- heart failure (Porrello et 195, MIR- al., 2011; van Rooij et al., Downregulated in HMWKO 424,MIR-497 552 4.1123 1.96E-05 2006) vs. FGFKO

Protective in IPC, HIF- MIR-199A, MIR- 1α regulator (Rane et al., Upregulated in LMWKO vs. 199B 149 3.0244 0.001246 2009) FGFKO

Table 9. microRNA targets identified by the microarray in non-ischemic hearts expressing

only HMW FGF2 or LMW FGF2 compare to hearts with FGF2 ablated. (n=3 hearts per

group).

173

DISCUSSION VI.

While FGF2 has been shown to protect the heart via the activation of rapid signaling

mechanisms such as kinase cascades and NO production, there is also the possibility that LMW

and HMW FGF2 may impact the recovery of the heart by regulating the transcription of genes

that modulate the response of the heart to ischemia and reperfusion. Both isoforms of FGF2 also

have distinct genomic effects in cardiomyocytes, which may protect the heart by priming it

against stressors, such as I/R injury.

Downstream targets of HMW and LMW FGF2 were revealed by the microarray data,

which allow new insight into previously undescribed targets of HMW and LMW FGF2 in the

heart. FGF2 has been shown to regulate the transcription of genes in various cell types, and work by Quarto and group suggests that HMW and LMW FGF2 differentially regulate various genes in 3T3 fibroblasts (Quarto et al., 2005). These investigators published that HMW FGF2

upregulated genes that promoted growth arrest and tumor suppression, such as nuclear factor X

(NfI-X) and nuclear protein 1 (Nupr1), as well as tumor suppressor St5, while downregulating

genes that promote growth, such as Egr-1, which promotes proliferation (Quarto et al., 2005). On

the other hand, it was found that LMW FGF2 tended to upregulate genes that promote growth

and angiogenesis, including Angptl-4, and ribosomal protein Rps5 (Quarto et al., 2005).

Interestingly, HMW FGF2 was found to upregulate cardiac troponin T2, and downregulate S100

calcium binding protein A13 (Quarto et al., 2005), which may parallel findings in our own study

demonstrating that calcium-handling proteins and myofilament protein expression may be altered

(Table 8). Our microarray identified 589 genes with differential regulation in the ventricles of

heart that either express only LMW FGF2, only HMW FGF2, or do not express FGF2 (Figure

37, Tables 4, 5, and 6). The threshold for identifying genes was set at a 1.10-fold change in

expression; while this is low, it allows for a greater sensitivity and may identify slight changes in

174

genes that could result in a significant physiological effect. Due to the low threshold used, validation of identified genes of interest by qRT-PCR or immunoblot is an important consideration. The genes that were identified were further analyzed via Database for Annotation,

Visualization and Integrated Discovery (DAVID) program for functional significance, and overlap between genes regulated by HMW and LMW FGF2 was also analyzed. 33 genes were identified that were modified by both HMW and LMW FGF2 (not counting two genes that were directly manipulated to generate the knockouts, Hprt and Fgf2 (Table 7)). Among these were several genes known to affect calcium handling, ion homeostasis, contraction, or survival (Table

8), which were subsequently further analyzed.

One of the strongest differences in gene transcription was found in the expression of sarcolipin, which was upregulated 13-fold in the HMWKO hearts vs. Fgf2 KO hearts as determined by the microarray. qRT-PCR confirmed this upregulation, and immunoblotting is currently underway to verify that this increase is seen at the protein level. Sarcolipin is a regulator of SERCA with homology to phospholamban, which reduces the activity of SERCA when bound to it (Babu et al., 2006; Babu et al., 2007). Like phospholamban, sarcolipin has a phosphorylatable residue near the N-terminus, Thr5, which is targeted by CamKII and reduces its affinity for SERCA (Bhupathy, Babu, Ito, & Periasamy, 2009). In our mouse model, it was seen that expression of only LMW FGF2 resulted in decrease calcium transient amplitude in unchallenged cardiomyocytes (Figure 21), without any change in several candidate calcium handling proteins, including SERCA, phospholamban, or calsequestrin (Figure 23). An increase in sarcolipin production would be expected to result in decreased SERCA activity, and may explain the difference in calcium cycling seen in myocytes only expressing LMW FGF2; overexpression of sarcolipin has been associated with reduced calcium transients (Babu et al.,

2005). In addition, this increase in sarcolipin may also contribute towards the priming of the

175

heart, by preventing the uptake of calcium into the SR, reducing SR calcium load and therefore

reducing the contribution of SR to calcium overload at the onset of ischemia. This would agree

with other studies demonstrating that sarcolipin may have a role to play in I/R injury; primarily

expressed in the atrium in healthy tissue, sarcolipin upregulation in the heart has been associated

with exercise-induced cardioprotection from I/R in dog hearts (Babu, Bhupathy, Carnes,

Billman, & Periasamy, 2007).

A second gene identified by microarray to be upregulated both by HMW and LMW

FGF2 is calreticulin, a calcium-binding protein present in the ER/SR. This protein is of particular

interest due to its implication in the ER stress response (Park et al., 2001), a known mediator of

I/R injury and infarct development (X. H. Liu et al., 2011). As our laboratory has previously

shown, overexpression of both classes of isoforms of FGF2 reduce infarct development after I/R

(House et al., 2003), and the implications of a stress response gene upregulated by both isoforms

might explain this phenomenon. However, immunoblotting revealed that there were no

detectable differences in the levels of protein between HMWKO and Fgf2 KO hearts, or

LMWKO and Fgf2 KO hearts, suggesting that the expression of this isoform does not contribute

to an increase in protein level; this may be due to regulation of calreticulin at the protein level,

either by translational regulation or modulation of calreticulin degradation. qRT-PCR is currently

being performed to address this issue.

The two genes identified with the highest degree of difference between the groups evaluated were myosin light chain 4 (Myl4) and myosin light chain 7 (Myl7), each of which was determined to have nearly 30-fold increased expression in HMWKO hearts compared to

FGFKO. This LMW FGF2-mediated increase in myosin light chains was subsequently confirmed with immunoblotting (Figure 42). Myosin light chains bind to the force-generating myosin heavy chains to form the thick filament (Kuwayama & Yagi, 1980) and are necessary for

176

myofibrillar integrity (Cardinaud, 1987; Moretti et al., 2002). There is evidence that matrix-

metalloproteinases, which exacerbate I/R injury, target these proteins during the development of

post-ischemic dysfunction (Sawicki et al., 2005). An upregulation of these proteins may result in

less severe functional injury due to proteolytic degradation in hearts only expressing LMW

FGF2.

Several genes relating to ion homeostasis, myofibril integrity, or microRNA-mediated

modulation of survival and function were identified that have not yet been validated by qRT-

PCR or immunoblotting, but may serve as a starting point for further examination of the role of

HMW and LMW FGF2 on transcriptional regulation in the heart. Among these are FGF12, a

member of the FGF family that is not secreted from the cell, and is thought to mediate its effects

intracellularly (Goldfarb et al., 2007; C. Liu, Dib-Hajj, & Waxman, 2001; Nakayama et al.,

2008). The microarray data demonstrated that FGF12 was upregulated in hearts only expressing

LMW FGF2 compared to Fgf2 KO hearts. While little is known about the role of FGF12 in the

heart, in neuronal tissue, FGF12 has been shown to colocalize with and depress the conductivity of voltage-gated sodium channels (Goldfarb et al., 2007). Whether they have a similar effect in heart tissue remains to be seen, but interference with sodium channels would be expected to affect the occurrence of calcium overload, and slow the onset of I/R injury. Another gene of interest that was identified by microarray but not confirmed is tropomodulin 4. Tropomodulins

are tropomyosin binding proteins that cap the actin filament and stabilize it (Mudry, Perry,

Richards, Fowler, & Gregorio, 2003). Interestingly, this protein appears to be differentially

regulated in LMWKO and HMWKO hearts, with lowered expression in hearts only expressing

HMW FGF2 (LMWKO) compared to those expressing LMW FGF2 (HMWKO). Like the

myosin light chains discussed above, tropomodulin 4 may act to stabilize the myofilament

against proteolytic degradation during I/R injury.

177

A final class of genes was found to be regulated by HMW and LMW FGF2, comprised of

genes targeted by certain microRNAs known to play a role in I/R injury. Targets of mir-320, mir-

21, mir-15 family, and mir-199A were all altered in isoform-specific knockout mouse ventricles,

with targets of mir-320, mir-21, mir-15 family downregulated by expression of only LMW

FGF2, and targets of mir-199A upregulated by expression of only HMW FGF2. Mir-320 has

been shown to contribute to cell death and larger infarct development in hearts subjected to I/R

injury (Ren et al., 2009). Mir-21 has been identified as being upregulated by I/R injury (Ren et

al., 2009), and is upregulated in healthy at-risk tissue in a model of ischemic preconditioning (S.

Dong et al., 2009). The mir-15 family has been implicated in neonatal cardiac development

(Porrello et al., 2011), and may contribute to a heart failure phenotype (van Rooij et al., 2006).

Mir-199A has been shown to affect hypoxia-inducible factor 1, and reduced the expression of

several pro-apoptotic proteins during hypoxia (Rane et al., 2009). In addition, mir-199A has been

implicated in ischemic preconditioning pathways (Rane et al., 2009). The effects of LMW and

HMW FGF2 on the targets of these mircoRNAs suggest that both isoforms FGF2 may regulate

protein expression postranscriptionally, and opens a new field of study for LMW FGF2

signaling.

Several genes identified were regulated by both LMW and HMW FGF2 (Table 7), which

opens up a new avenue of research into the synergistic actions of both classes of isoforms. As

shown previously by our laboratory, the overexpression of both LMW and HMW FGF2 results

in a decrease in infarct size after I/R injury (House et al., 2003), while manipulation of the

expression of only one class of isoform only affects post-ischemic function, not infarct size (Liao

et al., 2007; Liao et al., 2010); this suggests that the FGF2-mediated reduction in infarct size

requires the presence of both isoforms. The discovery of genes similarly regulated by both

provides a starting point for the further investigation of the role of transcriptional regulation in

178

the protection of the heart from infarct development in the presence of both classes of isoforms.

Interestingly, the genes common to both data sets were found to be similarly regulated (i.e. genes upregulated by LMW FGF2 were also upregulated by HMW FGF2). This suggests that, while

LMW FGF2 and HMW FGF2 may produce different biological effects in the heart by targeting different transcriptional pathways, they do not have opposing roles in regulating the same genes.

In summary, the microarray has identified several genes regulated by HMW and LMW

FGF2 that have the potential to have an impact on the recovery of the heart after I/R injury.

Among the genes confirmed to be altered by FGF2 isoform expression include genes that affect calcium handling and myofibrillar integrity. This suggests that HMW and LMW FGF2 may target these classes of proteins not only through rapid kinase signaling, but through modulation of their levels even before ischemia is induced.

179

CONCLUSIONS AND SIGNIFICANCE

The goal of this dissertation is to elucidate the role of the isoforms of FGF2 in the heart

during ischemia/reperfusion injury. As it has been demonstrated, the complexity of FGF2’s

effects on cardiac function of the ischemic heart must take into account the differential activation

of kinases as well as downstream targets at both the myofibril and the sarcoplasmic reticulum, of

HMW and LMW FGF2. In addition, the data presented in this dissertation suggest that crosstalk

between FGF2 isoforms may occur, with HMW FGF2 interfering with the protective signaling of

LMW FGF2. Taken together, the data presented here suggest a finely regulated and multifaceted

mechanism by which LMW and HMW FGF2 impact post-ischemic contractility of the heart

after I/R (Figure 43). Additionally, the effects of HMW and LMW FGF2 on non-kinase signal transduction, including NO signaling and transcriptional regulation of calcium handling and contractile genes, may also factor into the effects of these isoforms on the heart before, during, and after I/R injury.

This dissertation provides significant advances in the understanding of FGF2’s biological

actions on the heart, in particular the actions of the protein isoforms of FGF2 on the activation of

PKC isoforms and their downstream targets. While it has been shown that PKC is involved in

FGF2-mediated cardioprotection (House et al., 2003; Jiang et al., 2002; Padua et al., 1998;

Sheikh et al., 2001), data is provided in this dissertation, for the first time, that LMW FGF2,

which is the only isoform that mediates a protective response against post-ischemic cardiac

dysfunction (Liao et al., 2007; Liao et al., 2010), activates PKCα and ε isoforms to protect the

heart (Figure 8).

180

FGFR 1 LMW FGF2

HMW FGF2

PKCε PKCα CaMKII SR Myofibril PLB SERCA

Calcium sensitivity during ischemia ?

Improved postischemic function

Figure 43: Schematic representing the hypothetical pathways activated and/or inhibited by

LMW and HMW FGF2 to impact post-ischemic contractility in the heart.

181

The identification of the isoforms of PKC responsible for this protection, and the demonstration that the ablation or pharmacological inhibition of these PKCs results in a loss of

LMW FGF2-mediated protection from post-ischemic cardiac dysfunction, offers important

insight, as it has been shown that different PKCs may have profoundly different actions on the

cardiomyocyte upon stress. For example, opioid receptor agonists depend on the activity of

PKCδ to reduce I/R injury in the heart (Fryer, Wang et al., 2001), and phenylephrine has also

been shown to require active PKCd to prevent ischemia-induced cell death via activation of

sarcolemmal ATP-sensitive potassium channels (Turrell et al., 2011), while administration of a

PKCδ inhibitor has been shown to be detrimental to cell survival (Churchill & Mochly-Rosen,

2007). Conversely, PKCε has been shown to be required for landiolol-mediated cardioprotection,

(Takahashi et al., 2007), ethanol-induced cardioprotection (C. H. Chen et al., 1999), and

ischemic preconditioning (Ping et al., 1997; Qiu et al., 1998; Saurin et al., 2002; Yoshida,

Kawamura, Mizukami, & Kitakaze, 1997), while not playing a role in sildenafil-mediated

protection, which is associated with increased PKCα activation instead (Das et al., 2004). Thus,

since different stimuli produce their effects through different PKC isoforms, it is important to

identify which isoform or isoforms of PKC are activated by LMW FGF2 in order to produce a

clear picture of the signaling pathways activated during I/R injury.

Previous studies conducted by our laboratory have determined that overexpressing all

isoforms of FGF2 results in cardioprotection that may be blocked by the administration of

bisindolylmaleimide, which inhibits the α, β, γ, δ, ε, and ζ isoforms of PKC (House et al.,

2003), suggesting that one or more of these isoforms may be the likely candidate by which FGF2 protects the heart from I/R injury. Based on this evidence, a candidate approach was therefore used to identify, within a practical timeframe, which of these isoforms was necessary for LMW

FGF2-mediated protection from post-ischemic dysfunction. The decision to examine PKCα, δ,

182

and ε isoforms was based on evidence that FGF2 has been previously shown to activate these

isoforms of PKC in the heart (House et al., 2003; House et al., 2007; Jiang et al., 2002; Padua et al., 1998; Sheikh et al., 2001) and that these isoforms are relevant for modulating the response of the cardiomyocyte to I/R injury (Bouwman et al., 2004; Budas et al., 2007; C. H. Chen et al.,

1999; Churchill & Mochly-Rosen, 2007; G. W. Dorn et al., 1999; Fryer, Wang et al., 2001;

Hassouna et al., 2004; House et al., 2003; House et al., 2007; K. Inagaki et al., 2003; K. Inagaki

et al., 2003; Jiang et al., 2002; Kanaya et al., 2003; Kudo et al., 2002; Lochner et al., 2009;

Maslov et al., 2009; Melling et al., 2009; Miki et al., 1998; Padua et al., 1998; P. Ping et al.,

1997; Pyle et al., 2003; Sheikh et al., 2001; Turrell et al., 2011; Uecker et al., 2003; Wickley et al., 2006; Yoshida et al., 1996; Yoshida, Kawamura, Mizukami, & Kitakaze, 1997; H. Y. Zhang et al., 2002; H. Z. Zhou et al., 2002). It should, therefore, be noted that while it is demonstrated here that PKCα and PKCε play a necessary role in LMW FGF2-mediated protection from post-

ischemic dysfunction, these studies do not exclude the possibility that other isoforms of PKC may also contribute to this protection. While examining all the isoforms of PKC expressed in the heart is outside of the scope and resources of a dissertation project, other isoforms that may be

interesting to examine include PKCβΙΙ, which, while it has not been implicated in the recovery

of the heart acutely after I/R, and has been shown to play a role in the development of

hypertrophy (Ferreira, Brum, & Mochly-Rosen, 2011). Blocking this isoform during I/R

attenuates heart failure and prevents fibrosis (Palaniyandi, Ferreira, Brum, & Mochly-Rosen,

2011). Additionally, it would be interesting to examine the role of PKCζ, which phosphorylates

myofibrillar proteins troponin I, troponin T, and in the heart to modulate cardiac

contractility (Wu & Solaro, 2007).

The observation that LMW FGF2-mediated protection from post-ischemic dysfunction

requires PKCα and ε describes a unique mechanism of cardioprotection. These isoforms are also

183

activated by ischemic preconditioning (Chen & Frangogiannis, 2010; Fryer, Wang, Hsu, &

Gross, 2001; Hassouna, Matata, & Galinanes, 2004; Okusa et al., 2009; Schulz et al., 2003;

Takahashi et al., 2007; Wickley, Ding, Murray, & Damron, 2006; Yoshida et al., 1997), although

work by both Schulz and colleagues (Schulz et al., 2003) and Hassouna and colleagues

(Hassouna et al., 2004) have determined that in IPC, PKCα is downstream of PKCε. It is

demonstrated here that this is not the case in LMW FGF2-mediated protection, where PKCα and

PKCε are activated independently of one another, and in fact, PKCα is activated earlier at the

onset of ischemia, while PKCε is activated during early reperfusion. It is worth noting that the

cardioprotection mediated by LMW FGF2 is, therefore, distinct from IPC, one of the most well-

characterized cardioprotective stimuli studied. Additionally, it has been determined that

sevoflurane-induced preconditioning also corresponds to the translocation of the α and ε

isoforms of PKC, although whether this activation is causal has yet to be addressed (Okusa et al.,

2009).

Both the phosphorylation and translocation of PKC isoforms during ischemia-reperfusion

are transient, which suggests that these isoforms are activated at very specific timepoints during

I/R. This indicates careful regulation of PKC isoforms during ischemia and reperfusion, which is consistent with the observation that the protective effects of PKCα and ε are seen acutely, after only two hours of reperfusion. PKC-mediated protection in this case is not due to sustained activation, but rapid activation followed by rapid deactivation. The observation that PKCα is activated at early ischemia in hearts only expressing LMW FGF2 suggests that there is a role for

PKCα at this time point, and only this time point; by early reperfusion the activation of this isoform has fallen below the level of wildtype hearts. This is expected, as I/R injury is a dynamic

184

process, and what may be protective during early ischemia might fail to have an effect, or even

exacerbate injury, during reperfusion.

A particularly interesting phenomenon that is relevant for the characterization of the

effects of FGF2 in the ischemic heart is the observation that overexpression of both classes of

FGF2 isoforms are necessary to produce a decrease in infarct size (House et al., 2003); while

expression of only LMW FGF2 produces an improvement in post-ischemic cardiac function

(Liao et al., 2010), and expression of only HMW FGF2 resulted in a poor recovery of function

following I/R injury (Liao et al., 2007). The studies presented in this dissertation suggest one

possible explanation for this apparent paradox. It is demonstrated here that expression of only

LMW FGF2 results in an increase in PKCα activation during early ischemia, while I/R injury in

hearts overexpressing the HMW 24kDa isoform of FGF2 results in reduced phosphorylation of

PKCα. Further, it is demonstrated that LMW FGF2 produces an increase in the recovery of the

heart after I/R injury that is dependent on the expression of PKCα. Given that increased

phosphorylation of troponin isoforms when only LMW FGF2 is expressed is also dependent on

the expression of PKCα, it is reasonable to hypothesize that PKCα phosphorylation of troponin I

and T, and subsequent regulation of the actomyosin ATPase at the myofibril, may be one

possible mechanism by which different isoforms of FGF2 produce opposing effects on the

function of the heart after I/R. By directly altering the phosphorylation of myofibrillar proteins,

LMW and HMW FGF2 would be able to exert differential effects on the contractility and

relaxation of the cardiomyocyte, without altering other pathways that might influence survival. It

would be expected, based on this hypothesis, that overexpression of both classes of isoforms

would protect the heart from infarct development after I/R in a PKCα-independent mechanism.

Based on the novel and significant findings of this dissertation, the effects of LMW FGF2 expression at the myofibril are, therefore, of great interest as part of the mechanism of action of

185

FGF2 on myocardial dysfunction. It is demonstrated here that the expression of only LMW

FGF2 results in the increased phosphorylation of both troponin T and I during early ischemia.

Troponin T may be phosphorylated at one of four sites by PKCs, and the phosphorylation of any of these residues has been reported to reduce cardiac contractility (Noland, Raynor, & Kuo,

1989; Noland & Kuo, 1993; Sumandea, Pyle, Kobayashi, de Tombe, & Solaro, 2003). Thus, the increase in phosphorylation of troponin T in hearts only expressing LMW FGF2 would be expected to decrease contractility during early ischemia. The results of troponin I phorphorylation are more complex, with different residues producing different effects on the affinity of troponin for calcium, and the affinity of myosin for actin during crossbridge cycling

(Cole & Perry, 1975; Engel, Hinken, & Solaro, 2009; Noland et al., 1996). Serine 23 of troponin

I is phosphorylated by PKA or PKCδ producing an increase in lusitropy (and a subsequent increase in inotropy) (Cole & Perry, 1975; Engel, Hinken, & Solaro, 2009), which would be expected to increase the activity of actomyosin ATPase. Conversely, troponin I may also be phosphorylated on serine 43 or threonine 144 by PKC isoforms, which results in depressed calcium sensitivity, and reduced actomyosin ATPase activity (Engel et al., 2009; Noland et al.,

1996). In view of the fact that the increase in troponin phosphorylation is associated with the expression of PKCα, as opposed to PKCδ, it would be expected that the residues that are contributing to this phosphorylation are Ser43 or Thr144; however, attempts to test this hypothesis with antibodies that recognize single phosphorylated residues have not been successful, due to the lack of specificity and high background of commercially available antibodies.

This increase in troponin isoform phosphorylation at the onset of ischemia in hearts only expressing LMW FGF2, combined with the observation that actomyosin ATPase displays decreased calcium sensitivity, suggests that LMW FGF2 may protect the heart in a manner

186

similar to κ-opioid agonism, which induces similar changes at the myofibril at the onset of

ischemia (Pyle, Smith, & Hofmann, 2000; Pyle, Lester, & Hofmann, 2001). It is hypothesized

that these changes protect the heart via the slowing of ATP consumption at the myofibril during

ischemia, and a delay of the onset of the ATP depletion that triggers acidosis and calcium

overload. This hypothesis is supported by studies conducted by Jiang and colleagues (Jiang et al.,

2002), who examined the levels of intracellular ATP after the onset of ischemia in rat hearts

exposed to exogenously added LMW FGF2. It was determined that administration of ATP to

ischemic hearts resulted in an increase in the levels of ATP during ischemia in LMW FGF2-

treated hearts compared to vehicle-treated hearts, suggesting that one of the mechanisms by

which LMW FGF2 may protect the heart is a slowing of ATP consumption. In this instance,

however, the caveat should be made that there is no evidence to indicate whether the increase in

ATP levels is due to an increase in ATP generation, or a decrease in ATP hydrolysis. It must also

be considered that the administration of FGF2 prior to, or immediately after, reperfusion, has

protective effects as well (Horrigan et al., 1999; Jiang et al., 2002; Jiang, Srisakuldee, Soulet,

Bouche, & Kardami, 2004), suggesting that ATP conservation during ischemia may not be the

only means by which FGF2 protects the heart.

An interesting alternate mechanism of protecting the heart from post-ischemic

dysfunction may be suggested by the differential regulation of myosin light chains (MLCs) by

hearts only expressing LMW FGF2. Myosin light chains have a regulatory effect on both the

structure and function of the thick filament, stabilizing the myosin head and affecting stiffness

(Eden & Highsmith, 1997). MLCs can be subdivided into essential or regulatory peptides, with

the primary difference between the two being that the regulatory light chains may affect the interaction between the myosin head and actin via phophorylation near the N-terminus (Eden &

Highsmith, 1997; Sweeney, Bowman, & Stull, 1993). Conversely, the essential light chains

187

affect the direct binding of actin and myosin through direct interactions with actin (Timson,

2003). Our microarray data, which was also validated at the protein level, demonstrates that both myl4 (an essential light chain) and myl7 (a regulatory light chain) mRNA and MLC protein

(antibody not isoform-specific) are significantly increased in non-ischemic hearts expressing

only LMW FGF2, which may be acting to prime the heart for I/R injury. One way in which this

might prime the heart is by ameliorating the damage to cardiac dysfunction seen when

proteolytic degradation of the myofibril occurs, which is a significant contributor to post-

ischemic left ventricular dysfunction (Phatharajaree, Phrommintikul, & Chattipakorn, 2007).

Sawicki and colleagues (Sawicki et al., 2005) have determined that matrix metalloproteinases

(MMPs), which exacerbate I/R injury (Cheung et al., 2000), target these proteins during the

development of post-ischemic dysfunction, and that the degradation of a myosin light chain 1

contributes to increased dysfunction after I/R (Polewicz et al., 2011). Therefore, it may be that

production of more of these proteins allows for less MMP-mediated proteolytic damage in the

heart during an ischemic insult. Further studies are needed to determine if MMP-induced

degradation of myosin light chains is prevented in hearts only expressing LMW FGF2, due to

increased substrate (i.e. myosin light chains 4 and 7) expression.

The ability of FGF2 to modulate calcium signaling in various cell types has been

investigated in the past by several laboratories. FGF2 can modulate calcium currents via the

NMDA receptor in hippocampal neurons (Boxer, Moreno, Rudy, & Ziff, 1999), and depress

glutamate-receptor mediated calcium intake in cerebellar granule cells (El Idrissi & Trenkner,

1999). FGF2-induced hippocampal neurite outgrowth has been shown to be dependent on the calcium-binding protein (Oh et al., 2008) and FGF2 has been shown to initiate calcium influx in chick ciliary ganglion neurons (Distasi, Torre, Antoniotti, Munaron, &

Lovisolo, 1998). In addition, FGF2 has been shown to alter calcium concentration oscillation in

188

Xenopus oocytes (Browaeys-Poly, Cailliau, & Vilain, 1998; Malo, Browaeys-Poly, Fournier,

Cailliau, & Vilain, 1997). FGF2-induced tumor cell and vascular cell migration requires

calcium-sensitive potassium channels (Kessler, Budde, Gekle, Fabian, & Schwab, 2008; Wiecha

et al., 1998), which are activated via a g-protein-dependent mechanism (Kuhlmann et al., 2004),

and FGF2 has been shown to regulate intracellular calcium levels in spontaneously immortalized

granulosa cells via PKCδ (Lynch et al., 2000). Similarly, FGF2 also stimulates calcium efflux in

ovarian granulosa cells via activation of plasma membrane calcium ATPase (PMCA) and PKCδ

(Peluso et al., 2001; Peluso, 2003). Calcium influx is induced by FGF2 receptors in arterial

endothelial cells (Munaron & Fiorio Pla, 2000) and lens epithelial calls (Qu & Zhang, 2004). In

the heart, FGF2 has also been shown to increase angiogenesis in ischemic myocardium via

upregulation of calcineurin, a calcium-binding protein (Liu, Wu, Cai, & Sun, 2008), and

enhances the opening of Ca2+-permeable channels, altering calcium concentrations, in rat

ventricular myocytes (Merle, Usson, Robert-Nicoud, & Verdetti, 1997). However, the effect of

endogenously expressing only LMW FGF2 on calcium cycling in the heart has not been described in the literature. Post-translational modification of calcium-handling proteins and subsequent changes in calcium cycling have been shown to be important in a number of pathological states of the heart, including I/R injury (Akutsu et al., 2010; Gao et al., 1995;

Garcia-Rivas & Torre-Amione, 2009; Landstrom et al., 2011; Liu et al., 2011; Periasamy,

Bhupathy, & Babu, 2008; Shintani-Ishida & Yoshida, 2011; Steenbergen et al., 1993; Zhao et al.,

2003). In this dissertation, there is evidence that expression of only LMW FGF2 results in depressed calcium cycling in non-ischemic isolated myocytes. The mechanism for this is unclear, although one possible explanation may come from the data uncovered by the microarray described in this dissertation. One of the genes strongly upregulated by LMW FGF2 is sarcolipin, a modulator of the activity of SERCA. Sarcolipin is primarily expressed in the atrium

189

in normal healthy hearts, but has been found to be upregulated in the ventricle by ischemia

(Babu, Bhupathy, Carnes, Billman, & Periasamy, 2007). Sarcolipin acts in a manner similar to

phospholamban, binding to and decreasing the affinity of SERCA for calcium (Babu et al., 2006;

Babu et al., 2007), which would be expected to attenuate the reuptake of calcium into the SR and

decrease the calcium transient amplitude, as observed in this dissertation. Therefore, it may be

that LMW FGF2 is depressing calcium cycling in the cardiomyocyte by the increased expression

of this regulatory protein. This depression in calcium cycling may be relevant for understanding

LMW FGF2's protective effects on postischemic cardiac contractility, as inhibition of SR

calcium cycling during ischemia has been associated with improved contractility in ventricular

myocytes (Koyama, Boston, Ikenouchi, & Barry, 1996) and reduces stunning in the isolated

heart (du Toit & Opie, 1994; Mitchell, Winter, Banerjee, & Harken, 1993). Although the

mechanism for this is not well understood, it may be expected that lower peak calcium transient

amplitudes may result in a slowing of the calcium overload that is responsible for I/R injury.

Further studies are needed to determine if this is the case in hearts only expressing LMW FGF2.

Interestingly, the changes in calcium cycling in the cardiomyocyte were not found to

result in a significant difference in the contractility of the cell, in terms of the ±dL/dt or fractional

shortening. This mirrors what is seen in the ex vivo model, where HMWKO hearts (expressing

only LMW) do not show a difference in basal contractility (Liao et al., 2010). However, it

remains unclear how a decrease in the amplitude of calcium transients in the cardiomyocyte fail

to produce corresponding decreases in the contractility of the cell, as these two parameters are

often closely related. A similar phenomenon was seen by Qiujing Song (Song, 2004), examining the effect of phospholamban ablation on the heart, where an increase in calcium cycling did not

translate to an increase in in vivo contractility (although in this case, an increase in contractility

at the cellular level was observed). It is possible that there are unknown mechanisms by which

190

the expression of only LMW FGF2 regulates the responsiveness of the non-ischemic myofibril to

intracellular calcium concentrations, perhaps through regulation of myofibrillar proteins that

have not yet been tested, or via calcium sequestration. Futher studies are needed to address this

hypothesis.

It is demonstrated in this dissertation that expression of only LMW FGF2 results in an

increase in the phosphorylation of phospholamban at threonine 17 at the onset of ischemia.

However, it is also observed that PKCα phosphorylation is increased at this same time. Braz and

colleagues (Braz et al., 2004) have demonstrated that PKCα phosphorylates inhibitor-1 at serine

67, resulting in the increased activity of protein phosphatase-1, which dephosphorylates both the

serine 16 and threonine 17 residues of phospholamban and reduces cardiac contractility. For this

reason, it is surprising that both an increase in the activity of PKCα and phospho-threonine 17

phospholamban is observed at the same time point. This suggests that the expression of LMW

FGF2 results in the increased activation of another regulatory agent of phospholamban that is

sufficient to overcome the enhanced phosphatase activity of PP1 induced by PKCα at this same

time point. Regardless, the alterations in calcium cycling seen during ischemia when only LMW

FGF2 is expressed may be expected to be independent of PKCα activation. It has been well

established that the Thr17 residue of phospholamban is phosphorylated by CamKII. There is

some evidence that FGF2 activates CamKII in vascular cells to induce migration

in vasculogenesis (Bilato et al., 1995). Similarly, LMW FGF2 has been shown here to activate

PKCε, which can phosphorylate and activate CamKII to result in the increased phosphorylation

of phospholamban at threonine 17 (Oestreich et al., 2009).

CamKII activation may be one way that PKCε affects post-ischemic contractility in hearts that only express LMW FGF2. While there do not appear to be other effects for PKCε at

191

the myofibril in terms of troponin phosphorylation or actomyosin ATPase acitivity for this

isoform, other targets have been implicated for PKCε in during I/R injury, including mitochondrial targets such as the ATP-sensitive potassium (KATP) channel (Aizawa, Turner,

Weihrauch, Bosnjak, & Kwok, 2004; Edwards et al., 2009; Kim et al., 2006; Quinlan et al.,

2008; Wang & Ashraf, 1999) and the mitochondrial permeability transition pore (MPTP) (Miura,

Tanno, & Sato, 2010; Pravdic et al., 2009). It has been shown that PKCε is imported into the mitochondria in a heat shock protein-90 dependent manner after I/R injury (Budas et al. 2010).

Pravdic and colleagues (Pravdic et al., 2009) have demonstrated that anesthesia-mediated preconditioning is induced by a PKCε-mediated delay in the MPTP opening, which contributes

to cardioprotection by maintaining mitochondrial integrity. In addition, PKCε has been shown to

play a role in the activation of KATP channels at the sarcolemma, reducing the magnitude of ischemia-induced injury in females (Edwards et al., 2009). Whether any of these proteins are the targets of PKCε in the presence of only LMW FGF2 remains to be seen.

The question of how HMW and LMW FGF2 interact with one another is an important

consideration, particularly given that the two appear to have opposing effects on the recovery of

post-ischemic myocardial function (Liao et al., 2007; Liao et al., 2009; Liao et al., 2010), and yet

only when both are expressed is the protection from infarct development seen (House et al.,

2003). The data presented in this dissertation demonstrate, for the first time, that endogenously

expressed HMW FGF2 results in lowered activation of both the FGF receptor-1 and kinases that

are activated by LMW FGF2 to protect the heart from functional injury. While the mechanism

for this interaction is still unclear, there are a number of reasons that this may occur.

Unpublished in silico data generated in collaboration with Dr. Jerek Meller has suggested that the isoforms of FGF2 may bind to one another to form heterodimers. As HMW FGF2 is not released from the cell, binding to LMW FGF2 may effectively sequester it inside the cell and

192

prevent LMW FGF2 from binding to and activating the receptor. HMW FGF2 may also bind to

the intracellular domain of the receptor and interfere with the autophosphorylation or trafficking

of the receptor to the cellular membrane, rendering it unavailable for activation by LMW FGF2.

Despite the opposing effects of endogenously expressed LMW and HMW FGF2 on post-

ischemic contractile recovery, both classes of isoforms appear to be necessary for reduction of

infarct size, and this appears to be reliant on the modulation of the activity of NOS (Manning et

al., 2012). The means by which NOS protects the heart when all isoforms of FGF2 are overexpressed remains unclear. Other models of cardioprotection rely on the ability of NOS to

activate KATP channels, including studies utilizing levosimendan (Das & Sarkar, 2007), and

cilostazol (Bai et al., 2011). There is evidence that FGF2-mediated protection relies on the

activity of KATP channels (Manning et al., 2012), although whether this activity is directly

mediated by NOS has yet to be tested. Nitric oxide (NO) produced by NOS can directly activate

KATP channels by S-nitrosylation of a cysteine residue on the SUR2A regulatory subunit

(Kawano et al., 2009). In chronically hypoxic hearts, NO activates sarcolemmal KATP channels

through a mechanism dependent on cGMP activity (Baker et al., 2001). KATP channels, in turn,

are thought to protect the heart by shortening the duration of the action potential, resulting in a

reduction of calcium influx on a beat-to-beat basis and ameliorating calcium overload (Zingman

et al., 2002). In addition, NO may activate PKCε through direct nitrosylation of the regulatory

subunit, which has been shown to be a mechanism of cardioprotection in ischemic

postconditioning, where PKCε phosphorylates tyrosine kinases SRC and LCK and the

transcription factor NF-κB (Dawn & Bolli 2002). Further studies are needed to determine if this

is the means by which NOS protects the heart from infarct development in hearts overexpressing

all isoforms of FGF2

193

These findings shed light on several previously unknown biological actions of both the

LMW and HMW FGF2 in the ischemic heart, which is important for the development of clinical

therapies making use of, or targeting, FGF2 signaling. This dissertation indicates that LMW, and

not HMW FGF2, protects the heart from post-ischemic dysfunction, and therefore, the most

effective therapies for ischemic heart disease patients suffering from ventricular dysfunction

would make use of LMW FGF2 exclusively. Additionally, this research has revealed that PKC

isoforms α and ε are integral for LMW FGF2’s cardioprotective response, which has implications for combination therapies; often patients with ischemic heart disease have co- morbidities which require several different therapeutic strategies, and the elucidation of the protective signaling pathways of LMW FGF2 allows for greater understanding and prevention of potential adverse interactions. In this case, drugs that interfere with the protective PKC isoforms,

or alter the function of the myofibrillar or calcium-handling proteins identified by this research,

should be carefully evaluated for their impact on endogenously expressed LMW FGF2 protective

signaling, as well as LMW FGF2 that may be given therapeutically. There is a clear need for

novel pharmacological therapies that protect the heart from both functional injury and infarct

development. The research presented in this dissertation sheds light on an important protective

pathway that has the potential to significantly increase the quality of life of patients suffering

from ischemic heart disease.

FUTURE DIRECTIONS

This dissertation research provides evidence, for the first time, that LMW FGF2 requires

PKCα and ε to protect the heart from post-ischemic dysfunction, and that increases in troponin I

and T phosphorylation at early ischemia in hearts that only express LMW FGF2 are mediated by

PKCα, resulting in a decrease in myofilament calcium sensitivity. Given that previous

194

investigators have linked the phosphorylation of troponin I and T to cardioprotection (Pyle et al.,

2001; Pyle, Chen, & Hofmann, 2003; Shaw et al., 2009), it is reasonable to hypothesize that this contributes to LMW FGF2-mediated protection from post-ischemic dysfunction, but it should be noted that this has not yet been determined experimentally. The most thorough way to test this hypothesis would be to cross HMWKO mice to mice that express mutant forms of troponin I or

T that are not phosphorylatable, and to determine whether these are protected from I/R-induced contractile dysfunction. While these experiments are currently outside of the scope of this dissertation project, they would effectively demonstrate a causal link between the increase in troponin phosphorylation and LMW FGF2-mediated protection from postischemic dysfunction.

Similarly, while it has been noted that expression of LMW FGF2 results in an increase in the phosphorylation of phospholamban at threonine 17 during early ischemia, it has not yet been determined whether this increase is responsible for the cardioprotection seen in HMWKO hearts.

Work done by Matiazzi and colleagues (Mattiazzi, Mundina-Weilenmann, Vittone, & Said,

2004; Mattiazzi, Mundina-Weilenmann, Vittone, Said, & Kranias, 2006) has demonstrated that increased phosphorylation at this site is associated with an improvement in post-ischemic function, and this hypothesis is currently being tested in our model, where genetically-modified mice lacking the threonine 17 phosphorylation site have been bred to HMWKO mice, to determine if the improvement in post-ischemic function is still observed in the absence of a phosphorylatable Thr17 residue.

Another interesting question raised by this dissertation research is whether the acute improvement in cardiac function after I/R observed when only LMW FGF2 is expressed translates to improved long-term function. Work done by Kardami and group (Kardami et al.,

2004) suggests that the LMW and HWM isoforms of FGF2 play a differential role in the development of maladaptive heart failure, with only HMW FGF2 isoforms promoting

195

remodeling in pressure overload- and angiotensin-induced hypertrophy. As I/R injury often leads

to heart failure, it would be of great clinical relevance to determine if chronic left ventricular dysfunction induced by MI in hearts only expressing LMW FG2 is ameliorated. Work is currently underway to examine the impact of temporary occlusion of the left anterior decending

(LAD) artery in HMWKO mice on cardiac function.

As previously mentioned, a candidate approach was taken to identify the isoforms of

PKC, and their downstream targets, activated by expression of only LMW FGF2; the proteins

that were the most likely to be activated by FGF2 and impact post-ischemic contractility were

analyzed first. However, a weakness of this approach is that other, less well-characterized, PKC

isoforms or target proteins may be affecting the ischemic HMWKO heart as well. In particular, it

would be interesting to examine the effects of PKC isoforms ζ and β in hearts that only express

LMW FGF2, as these have been implicated in the phosphorylation of myofibrillar proteins (Wu

& Solaro, 2007), and heart failure progression (Ferreira et al., 2011; Palaniyandi et al., 2011),

respectively. As part of a broader approach to identifying phosphorylated proteins that may slip

through the cracks, a protocol utilizing mass spectrometry is currently being developed in

collaboration with Dr. Kenneth Greis and Dr. Aruna Wijeratne, making use of isotope-tagging of

proteins from homogenates derived from HMWKO hearts subjected to ischemia and reperfusion,

to screen for proteins with significant differences in phosphorylation when hearts express only

LMW FGF2. Preliminary results have identified over a hundred proteins that may show

differences in phosphorylation, and efforts are currently underway to confirm and validate these

results.

196

LIST OF PUBLICATIONS AND ABSTRACTS

Abstracts

Bodmer J, Liao S, Newman G Schultz JJ. The role of Protein Kinase C Epsilon in Ischemic

Cardioprotection Mediated by the Low Molecular Weight Isoform of Fibroblast Growth Factor

2, UCCOM Pharmacology Research Retreat, 2008

Bodmer J, Wright T, Newman G, Schultz JJ. High molecular weight Fibroblast Growth Factor 2

(FGF2) requires an active Fibroblast Growth Factor Receptor 1 (FGFR1) to depress post-

ischemic cardiac function, Experimental Biology 2009

Bodmer J, Newman G Schultz JJ. The role of PKC alpha in FGF2-mediated cardioprotection,

UCCOM Pharmacology Research Retreat, 2009

Oloizia B, Bodmer J, Newman G, Schultz JJ. SR Ca2+-Handling Proteins and PKCε: A Role in

Low Molecular Weight FGF2-Mediated Cardioprotection?(abstract), Experimental Biology 2010

Bodmer J, Oloizia B, Newman G, Schultz JJ. The alpha isoform of PKC is necessary for low molecular weight fibroblast growth factor-2 mediated cardioprotection (abstract), Gordon

Research Conference 2010

Bodmer J, Gao X, S, Pyle WG , Schultz JJ. A mechanism for LMW FGF2-mediated

protection from postischemic cardiac dysfunction: protein kinase C isoforms and myofibrillar

proteins (abstract). Cincinnati Children’s Research Foundation Postdoctoral recruitment forum

2011

Wijeratne AB, Bodmer JR, Haffeya WD, Shua H, Schultz JJ, Greis KD Profiling Strategy for

Relative Quantification of Phophoproteome from Tissue Extracts: Method Validation and

Application to a Cardiovascular Disease Model System (Abstract) ASMS June 2011

Bodmer JR, Gao X, Perkins SO, Pyle WG, Schultz JJ A mechanism for LMW FGF2-mediated

protection from postischemic cardiac dysfunction: protein kinase C isoforms and myofibrillar

197

proteins (abstract) Cincinnati Children’s Research Foundation Postdoctoral Recruitment

Symposium, April 2011

Manuscripts

Liao S, Bodmer J, Pietras D, Azhar M, Doetschman T, Schultz JJ. Biological functions of the

low and high molecular weight protein isoforms of fibroblast growth factor-2 in cardiovascular development and disease. Dev Dyn. 2009 Feb

Liao S, Bodmer J, Newman G, Doetschman T, Schultz JJ. The influence of FGF2 high molecular weight (HMW) isoforms in the development of cardiac ischemia-reperfusion injury.J

Mol Cell Cardiol 2010 Jan

Manning JR, Carpenter G, Porter D, PietrasDA, House SL, Doetschman T, Schultz JJ.

Fibroblast growth factor-2-induced cardioprotection against myocardial infarction occurs via the interplay between nitric oxide, protein kinase signaling, and ATP-sensitive potassium channels

Growth Factor 2012 (in press)

Manning JR, Gao X, Newman G, Perkins SO, Pyle WG, Schultz JJ. The low molecular weight isoform of fibroblast growth factor 2 protects the heart from ischemic dysfunction through

activation of protein kinase C alpha and epsilon isoforms and myofibrillar protein

phosphorylation.(in preparation, to be submitted March 2012)

Manning JR, Oloizia B, Newman G, Wang HS, Schultz JJ. Low molecular weight fibroblast

growth factor 2-mediated protection from postischemic function is associated with changes in

calcium cycling (in preparation).

Manning JR, Wright T, Newman G, Schultz JJ. A novel role for fibroblast growth factor

receptor 1 in high molecular weight fibroblast growth factor 2 mediated injury in cardiac

ischemia (in preparation)

198

REFERENCES

Adak, S., Santolini, J., Tikunova, S., Wang, Q., Johnson, J. D., & Stuehr, D. J. (2001). Neuronal

nitric-oxide synthase mutant (ser-1412 → asp) demonstrates surprising connections between

heme reduction, NO complex formation, and catalysis. Journal of Biological Chemistry,

276(2), 1244-1252. doi:10.1074/jbc.M006857200

Aizawa, K., Turner, L. A., Weihrauch, D., Bosnjak, Z. J., & Kwok, W. M. (2004). Protein kinase

C-epsilon primes the cardiac sarcolemmal adenosine triphosphate-sensitive potassium

channel to modulation by isoflurane. Anesthesiology, 101(2), 381-389.

Akita, T., Abe, T., Kato, S., Kodama, I., & Toyama, J. (1993). Protective effects of diltiazem and

ryanodine against ischemia-reperfusion injury in neonatal rabbit hearts. The Journal of

Thoracic and Cardiovascular Surgery, 106(1), 55-66.

Akutsu, Y., Hamazaki, Y., Kaneko, K., Kodama, Y., Li, H., Suyama, J., . . . Kobayashi, Y. (2010).

Visualization of excessive intracellular calcium ion overload caused by the occurrence of

reperfusion injury. Cardiovascular Revascularization Medicine, 11(4), 267-268.

doi:10.1016/j.carrev.2010.01.004

Albuquerque, M. L., Akiyama, S. K., & Schnaper, H. W. (1998). Basic fibroblast growth factor

release by human coronary artery endothelial cells is enhanced by matrix proteins, 17beta-

estradiol, and a PKC signaling pathway. Experimental Cell Research, 245(1), 163-169.

doi:10.1006/excr.1998.4243

Ambrosio, G., Zweier, J. L., Duilio, C., Kuppusamy, P., Santoro, G., Elia, P. P., . . . Chiariello, M.

(1993). Evidence that mitochondrial respiration is a source of potentially toxic oxygen free

radicals in intact rabbit hearts subjected to ischemia and reflow. The Journal of Biological

Chemistry, 268(25), 18532-18541.

199

Amin, A. S., Tan, H. L., & Wilde, A. A. M. (2010). Cardiac ion channels in health and disease.

Heart Rhythm, 7(1), 117-126. doi:10.1016/j.hrthm.2009.08.005

Anderson, M. E. (2009). CaMKII and a failing strategy for growth in heart. The Journal of

Clinical Investigation, 119(5), 1082-1085.

Antoine, M., Köhl, R., Tag, C. G., Gressner, A. M., Hellerbrand, C., & Kiefer, P. (2009). Secreted

cysteine-rich FGF receptor derives from posttranslational processing by furin-like

prohormone convertases. Biochemical and Biophysical Research Communications, 382(2),

359-364. doi:DOI: 10.1016/j.bbrc.2009.03.026

Arnold M., K. (1973). Effects of ischemia on the contractile processes of heart muscle. The

American Journal of Cardiology, 32(4), 456-460. doi:10.1016/S0002-9149(73)80036-0

Azfer, A., Niu, J., Rogers, L. M., Adamski, F. M., & Kolattukudy, P. E. (2006). Activation of

endoplasmic reticulum stress response during the development of ischemic heart disease.

American Journal of Physiology.Heart and Circulatory Physiology, 291(3), H1411-20.

doi:10.1152/ajpheart.01378.2005

Azhar, M., Yin, M., Zhou, M., Li, H., Mustafa, M., Nusayr, E., . . . Doetschman, T. (2009). Gene

targeted ablation of high molecular weight fibroblast growth factor-2. Developmental

Dynamics : An Official Publication of the American Association of Anatomists, 238(2), 351-

357. doi:10.1002/dvdy.21835

Babu, G. J., Bhupathy, P., Carnes, C. A., Billman, G. E., & Periasamy, M. (2007). Differential

expression of sarcolipin protein during muscle development and cardiac pathophysiology.

Journal of Molecular and Cellular Cardiology, 43(2), 215-222.

doi:10.1016/j.yjmcc.2007.05.009

Babu, G. J., Bhupathy, P., Petrashevskaya, N. N., Wang, H., Raman, S., Wheeler, D., . . .

Periasamy, M. (2006). Targeted overexpression of sarcolipin in the mouse heart decreases

200

sarcoplasmic reticulum calcium transport and cardiac contractility. The Journal of Biological

Chemistry, 281(7), 3972-3979. doi:10.1074/jbc.M508998200

Babu, G. J., Bhupathy, P., Timofeyev, V., Petrashevskaya, N. N., Reiser, P. J., Chiamvimonvat,

N., & Periasamy, M. (2007). Ablation of sarcolipin enhances sarcoplasmic reticulum

calcium transport and atrial contractility. Proceedings of the National Academy of Sciences

of the United States of America, 104(45), 17867-17872. doi:10.1073/pnas.0707722104

Babu, G. J., Zheng, Z., Natarajan, P., Wheeler, D., Janssen, P. M., & Periasamy, M. (2005).

Overexpression of sarcolipin decreases myocyte contractility and calcium transient.

Cardiovascular Research, 65(1), 177-186. doi:10.1016/j.cardiores.2004.08.012

Bai, Y., Muqier, Murakami, H., Iwasa, M., Sumi, S., Yamada, Y., et al. (2011). Cilostazol

protects the heart against ischaemia reperfusion injury in a rabbit model of myocardial

infarction: Focus on adenosine, nitric oxide and mitochondrial ATP-sensitive potassium

channels. Clinical and Experimental Pharmacology & Physiology, 38(10), 658-665.

Baines, C. P., Kaiser, R. A., Purcell, N. H., Blair, N. S., Osinska, H., Hambleton, M. A., . . .

Molkentin, J. D. (2005). Loss of cyclophilin D reveals a critical role for mitochondrial

permeability transition in cell death. Nature, 434(7033), 658-662. doi:10.1038/nature03434

Baker, J. E., Contney, S. J., Singh, R., Kalyanaraman, B., Gross, G. J., & Bosnjak, Z. J. (2001).

Nitric oxide activates the sarcolemmal K(ATP) channel in normoxic and chronically

hypoxic hearts by a cyclic GMP-dependent mechanism. Journal of Molecular and Cellular

Cardiology, 33(2), 331-341.

Barouch, L. A., Harrison, R. W., Skaf, M. W., Rosas, G. O., Cappola, T. P., Kobeissi, Z. A., . . .

Hare, J. M. (2002). Nitric oxide regulates the heart by spatial confinement of nitric oxide

synthase isoforms. Nature, 416(6878), 337-339. doi:10.1038/416005a

Barr, R. L., & Lopaschuk, G. D. (2000). Methodology for measuring in vitro/ex vivo cardiac

201

energy metabolism. Journal of Pharmacological and Toxicological Methods, 43(2), 141-

152.

Barry, W. H., & Bridge, J. H. (1993). Intracellular calcium homeostasis in cardiac myocytes.

Circulation, 87(6), 1806-1815.

Basu, A., Lu, D., Sun, B., Moor, A. N., Akkaraju, G. R., & Huang, J. (2002). Proteolytic

activation of protein kinase C-ε by caspase-mediated processing and transduction of

antiapoptotic signals. Journal of Biological Chemistry, 277(44), 41850-41856.

doi:10.1074/jbc.M205997200

Bayer, A. L., Heidkamp, M. C., Patel, N., Porter, M., Engman, S., & Samarel, A. M. (2003).

Alterations in protein kinase C isoenzyme expression and autophosphorylation during the

progression of pressure overload-induced left ventricular hypertrophy. Molecular and

Cellular Biochemistry, 242(1-2), 145-152.

Behn-Krappa, A., & Newton, A. C. (1999). The hydrophobic phosphorylation motif of

conventional protein kinase C is regulated by autophosphorylation. Current Biology : CB,

9(14), 728-737.

Bell, R. M., Hannun, Y. A., & Loomis, C. R. (1986). Mechanism of regulation of protein kinase

C by lipid second messengers. Symposium on Fundamental Cancer Research, 39, 145-156.

Bell, R. M., & Yellon, D. M. (2001). The contribution of endothelial nitric oxide synthase to

early ischaemic preconditioning: The lowering of the preconditioning threshold. an

investigation in eNOS knockout mice. Cardiovascular Research, 52(2), 274-280.

doi:10.1016/S0008-6363(01)00394-7

Bernheim, J., & Benchetrit, S. (2011). The potential roles of FGF23 and klotho in the prognosis

of renal and cardiovascular diseases. Nephrology, Dialysis, Transplantation : Official

Publication of the European Dialysis and Transplant Association - European Renal

202

Association, 26(8), 2433-2438. doi:10.1093/ndt/gfr208

Besch, H. R.,Jr, & Watanabe, A. M. (1975). Role of free calcium and ATP in calcium release

from cardiac sarcoplasmic reticulum fragments. Recent Advances in Studies on Cardiac

Structure and Metabolism, 5, 143-149.

Bhupathy, P., Babu, G. J., Ito, M., & Periasamy, M. (2009). Threonine-5 at the N-terminus can

modulate sarcolipin function in cardiac myocytes. Journal of Molecular and Cellular

Cardiology, 47(5), 723-729. doi:10.1016/j.yjmcc.2009.07.014

Bikfalvi, A., Klein, S., Pintucci, G., Quarto, N., Mignatti, P., & Rifkin, D. B. (1995). Differential

modulation of cell phenotype by different molecular weight forms of basic fibroblast growth

factor: Possible intracellular signaling by the high molecular weight forms. The Journal of

Cell Biology, 129(1), 233-243.

Bilato, C., Pauly, R. R., Melillo, G., Monticone, R., Gorelick-Feldman, D., Gluzband, Y. A., et

al. (1995). Intracellular signaling pathways required for rat cell

migration. interactions between basic fibroblast growth factor and platelet-derived growth

factor. The Journal of Clinical Investigation, 96(4), 1905-1915.

Black, E. G., Logan, A., Davis, J. R., & Sheppard, M. C. (1990). Basic fibroblast growth factor

affects DNA synthesis and cell function and activates multiple signalling pathways in rat

thyroid FRTL-5 and pituitary GH3 cells. The Journal of Endocrinology, 127(1), 39-46.

Bluhm, W. F., Kranias, E. G., Dillmann, W. H., & Meyer, M. (2000). Phospholamban: A major

determinant of the cardiac force-frequency relationship. American Journal of

Physiology.Heart and Circulatory Physiology, 278(1), H249-55.

Bolli, R. (1990). Mechanism of myocardial "stunning". Circulation, 82(3), 723-738.

Bolli, R. (2000). The late phase of preconditioning. Circulation Research, 87(11), 972-983.

Bolli, R., Dawn, B., Tang, X. L., Qiu, Y., Ping, P., Xuan, Y. T., . . . Zhang, J. (1998). The nitric

203

oxide hypothesis of late preconditioning. Basic Research in Cardiology, 93(5), 325-338.

Bolli, R., & Marban, E. (1999). Molecular and cellular mechanisms of myocardial stunning.

Physiological Reviews, 79(2), 609-634.

Bonneh-Barkay, D., Shlissel, M., Berman, B., Shaoul, E., Admon, A., Vlodavsky, I., . . . Ron, D.

(1997). Identification of glypican as a dual modulator of the biological activity of fibroblast

growth factors. Journal of Biological Chemistry, 272(19), 12415-12421.

doi:10.1074/jbc.272.19.12415

Bouche, G., Gas, N., Prats, H., Baldin, V., Tauber, J. P., Teissie, J., & Amalric, F. (1987). Basic

fibroblast growth factor enters the nucleolus and stimulates the transcription of ribosomal

genes in ABAE cells undergoing G0----G1 transition. Proceedings of the National Academy

of Sciences of the United States of America, 84(19), 6770-6774.

Bouvagnet, P., Leger, J., Pons, F., Dechesne, C., & Leger, J. J. (1984). Fiber types and myosin

types in human atrial and ventricular myocardium. an anatomical description. Circulation

Research, 55(6), 794-804.

Bouwman, R. A., Musters, R. J., van Beek-Harmsen, B. J., de Lange, J. J., & Boer, C. (2004).

Reactive oxygen species precede protein kinase C-delta activation independent of adenosine

triphosphate-sensitive mitochondrial channel opening in sevoflurane-induced

cardioprotection. Anesthesiology, 100(3), 506-514.

Bowling, N., Walsh, R. A., Song, G., Estridge, T., Sandusky, G. E., Fouts, R. L., . . . Vlahos, C. J.

(1999). Increased protein kinase C activity and expression of Ca2+-sensitive isoforms in the

failing human heart. Circulation, 99(3), 384-391. doi:10.1161/01.CIR.99.3.384

Boxer, A. L., Moreno, H., Rudy, B., & Ziff, E. B. (1999). FGF-2 potentiates ca(2+)-dependent

inactivation of NMDA receptor currents in hippocampal neurons. Journal of

Neurophysiology, 82(6), 3367-3377.

204

Bozler, E. (1952). Evidence of an ATP-actomyosin complex in relaxed muscle and its response to

calcium ions. The American Journal of Physiology, 168(3), 760-765.

Braz, J. C., Gregory, K., Pathak, A., Zhao, W., Sahin, B., Klevitsky, R., . . . Molkentin, J. D.

(2004). PKC-alpha regulates cardiac contractility and propensity toward heart failure.

Nature Medicine, 10(3), 248-254. doi:10.1038/nm1000

Brenner, B. (1988). Effect of Ca2+ on cross-bridge turnover kinetics in skinned single rabbit

psoas fibers: Implications for regulation of . Proceedings of the National

Academy of Sciences of the United States of America, 85(9), 3265-3269.

Brinks, H., Boucher, M., Gao, E., Chuprun, J. K., Pesant, S., Raake, P. W., . . . Koch, W. J.

(2010). Level of G protein–Coupled receptor kinase-2 determines myocardial

Ischemia/Reperfusion injury via pro- and anti-apoptotic mechanisms / novelty and

significance. Circulation Research, 107(9), 1140-1149.

doi:10.1161/CIRCRESAHA.110.221010

Browaeys-Poly, E., Cailliau, K., & Vilain, J. P. (1998). Fibroblast and epidermal growth factor

receptor expression in xenopus oocytes displays distinct calcium oscillatory patterns.

Biochimica Et Biophysica Acta, 1404(3), 484-489.

Budas, G. R., Churchill, E. N., & Mochly-Rosen, D. (2007). Cardioprotective mechanisms of

PKC isozyme-selective activators and inhibitors in the treatment of ischemia-reperfusion

injury. Pharmacological Research, 55(6), 523-536. doi:DOI: 10.1016/j.phrs.2007.04.005

Bueno, O. F., De Windt, L. J., Tymitz, K. M., Witt, S. A., Kimball, T. R., Klevitsky, R., . . .

Molkentin, J. D. (2000). The MEK1-ERK1/2 signaling pathway promotes compensated

cardiac hypertrophy in transgenic mice. The EMBO Journal, 19(23), 6341-6350.

doi:10.1093/emboj/19.23.6341

Bugler, B., Amalric, F., & Prats, H. (1991). Alternative initiation of translation determines

205

cytoplasmic or nuclear localization of basic fibroblast growth factor. Molecular and Cellular

Biology, 11(1), 573-577.

Burkard, N., Williams, T., Czolbe, M., Blömer, N., Panther, F., Link, M., . . . Ritter, O. (2010).

Conditional overexpression of neuronal nitric oxide synthase is cardioprotective in

Ischemia/Reperfusion / clinical perspective. Circulation, 122(16), 1588-1603.

doi:10.1161/CIRCULATIONAHA.109.933630

Cardinaud, R. (1987). Proteolysis rates of a myosin heavy chain site with papain. evidence for a

combined LC2-filament-mediated mechanism. FEBS Letters, 220(2), 376-382.

Carrozza, J., Bentivegna, L., Williams, C., Kuntz, R., Grossman, W., & Morgan, J. (1992).

Decreased myofilament responsiveness in myocardial stunning follows transient calcium

overload during ischemia and reperfusion. Circulation Research, 71(6), 1334-1340.

Cazaubon, S., Bornancin, F., & Parker, P. J. (1994). Threonine-497 is a critical site for permissive

activation of protein kinase C alpha. The Biochemical Journal, 301 ( Pt 2)(Pt 2), 443-448.

Chang, J. D., Xu, Y., Raychowdhury, M. K., & Ware, J. A. (1993). Molecular cloning and

expression of a cDNA encoding a novel isoenzyme of protein kinase C (nPKC). A new

member of the nPKC family expressed in , megakaryoblastic cells, and

platelets. The Journal of Biological Chemistry, 268(19), 14208-14214.

Chen, C. C. (1993). Protein kinase C alpha, delta, epsilon and zeta in C6 glioma cells. TPA

induces translocation and down-regulation of conventional and new PKC isoforms but not

atypical PKC zeta. FEBS Letters, 332(1-2), 169-173.

Chen, C. H., Gray, M. O., & Mochly-Rosen, D. (1999). Cardioprotection from ischemia by a

brief exposure to physiological levels of ethanol: Role of epsilon protein kinase C.

Proceedings of the National Academy of Sciences of the United States of America, 96(22),

12784-12789.

206

Chen, L., Wright, L. R., Chen, C. H., Oliver, S. F., Wender, P. A., & Mochly-Rosen, D. (2001).

Molecular transporters for peptides: Delivery of a cardioprotective epsilonPKC agonist

peptide into cells and intact ischemic heart using a transport system, R(7). Chemistry &

Biology, 8(12), 1123-1129.

Chen, W., & Frangogiannis, N. G. (2010). The role of inflammatory and fibrogenic pathways in

heart failure associated with aging. Heart Failure Reviews, 15(5), 415-422.

doi:10.1007/s10741-010-9161-y

Cheung, P., Sawicki, G., Wozniak, M., Wang, W., Radomski, M. W., & Schulz, R. (2000).

Matrix metalloproteinase-2 contributes to ischemia-reperfusion injury in the heart.

Circulation, 101(15), 1833-1839.

Chlebova, K., Bryja, V., Dvorak, P., Kozubik, A., Wilcox, W. R., & Krejci, P. (2009). High

molecular weight FGF2: The biology of a nuclear growth factor. Cellular and Molecular

Life Sciences : CMLS, 66(2), 225-235. doi:10.1007/s00018-008-8440-4

Chuang, C. Y., Lord, M. S., Melrose, J., Rees, M. D., Knox, S. M., Freeman, C., . . . Whitelock,

J. M. (2010). Heparan sulfate-dependent signaling of fibroblast growth factor 18 by

chondrocyte-derived perlecan. Biochemistry, 49(26), 5524-5532. doi:10.1021/bi1005199

Churchill, E. N., & Mochly-Rosen, D. (2007). The roles of PKCdelta and epsilon isoenzymes in

the regulation of myocardial ischaemia/reperfusion injury. Biochemical Society

Transactions, 35(Pt 5), 1040-1042. doi:10.1042/BST0351040

Cole, H. A., & Perry, S. V. (1975). The phosphorylation of troponin I from . The

Biochemical Journal, 149(3), 525-533.

Cool, S. M., Sayer, R. E., van Heumen, W. R., Pickles, J. O., & Nurcombe, V. (2002). Temporal

and spatial expression of fibroblast growth factor receptor 4 isoforms in murine tissues. The

Histochemical Journal, 34(6-7), 291-297.

207

Cox, C. M., & Poole, T. J. (2000). Angioblast differentiation is influenced by the local

environment: FGF-2 induces angioblasts and patterns vessel formation in the quail embryo.

Developmental Dynamics : An Official Publication of the American Association of

Anatomists, 218(2), 371-382. doi:2-Z

Cross, M. J., Lu, L., Magnusson, P., Nyqvist, D., Holmqvist, K., Welsh, M., & Claesson-Welsh,

L. (2002). The shb adaptor protein binds to tyrosine 766 in the FGFR-1 and regulates the

Ras/MEK/MAPK pathway via FRS2 phosphorylation in endothelial cells. Molecular

Biology of the Cell, 13(8), 2881-2893. doi:10.1091/mbc.E02-02-0103

Csukai, M., Chen, C., De Matteis, M. A., & Mochly-Rosen, D. (1997). The coatomer protein β′-

COP, a selective binding protein (RACK) for protein kinase Cε. Journal of Biological

Chemistry, 272(46), 29200-29206. doi:10.1074/jbc.272.46.29200

Cuenda, A., Rouse, J., Doza, Y. N., Meier, R., Cohen, P., Gallagher, T. F., . . . Lee, J. C. (1995).

SB 203580 is a specific inhibitor of a MAP kinase homologue which is stimulated by

cellular stresses and interleukin-1. FEBS Letters, 364(2), 229-233. doi:DOI: 10.1016/0014-

5793(95)00357-F

Cuevas, P., Barrios, V., Gimenez-Gallego, G., Martinez-Coso, V., Cuevas, B., Benavides, J., . . .

Asin-Cardiel, E. (1997). Serum levels of basic fibroblast growth factor in acute myocardial

infarction. European Journal of Medical Research, 2(7), 282-284.

Cuevas, P., Carceller, F., Martinez-Coso, V., Asin-Cardiel, E., & Gimenez-Gallego, G. (2000).

Fibroblast growth factor cardioprotection against ischemia-reperfusion injury may involve

K+ ATP channels. European Journal of Medical Research, 5(4), 145-149.

Cuevas, P., Carceller, F., Martinez-Coso, V., Cuevas, B., Fernandez-Ayerdi, A., Reimers, D., . . .

Gimenez-Gallego, G. (1999). Cardioprotection from ischemia by fibroblast growth factor:

Role of inducible nitric oxide synthase. European Journal of Medical Research, 4(12), 517-

208

524.

Cuevas, P., Reimers, D., Carceller, F., Martinez-Coso, V., Redondo-Horcajo, M., Saenz de

Tejada, I., & Gimenez-Gallego, G. (1997). Fibroblast growth factor-1 prevents myocardial

apoptosis triggered by ischemia reperfusion injury. European Journal of Medical Research,

2(11), 465-468.

Curto, M., Frankel, P., Carrero, A., & Foster, D. A. (1998). Novel recruitment of shc, Grb2, and

sos by fibroblast growth factor receptor-1 in v-src-transformed cells. Biochemical and

Biophysical Research Communications, 243(2), 555-560. doi:10.1006/bbrc.1997.7982

Dantzig, J. A., Goldman, Y. E., Millar, N. C., Lacktis, J., & Homsher, E. (1992). Reversal of the

cross-bridge force-generating transition by photogeneration of phosphate in rabbit psoas

muscle fibres. The Journal of Physiology, 451, 247-278.

Das, A., Ockaili, R., Salloum, F., & Kukreja, R. C. (2004). Protein kinase C plays an essential

role in sildenafil-induced cardioprotection in rabbits. American Journal of Physiology.Heart

and Circulatory Physiology, 286(4), H1455-60. doi:10.1152/ajpheart.01040.2003

Das, B., & Sarkar, C. (2007). Pharmacological preconditioning by levosimendan is mediated by

inducible nitric oxide synthase and mitochondrial KATP channel activation in the in vivo

anesthetized rabbit heart model. Vascular Pharmacology, 47(4), 248-256.

Das, A., Salloum, F. N., Xi, L., Rao, Y. J., & Kukreja, R. C. (2009). ERK phosphorylation

mediates sildenafil-induced myocardial protection against ischemia-reperfusion injury in

mice. American Journal of Physiology.Heart and Circulatory Physiology, 296(5), H1236-43.

doi:10.1152/ajpheart.00100.2009

Davidson, J. M., & Broadley, K. N. (1991). Manipulation of the wound-healing process with

basic fibroblast growth factor. Annals of the New York Academy of Sciences, 638, 306-315.

Davis, B. A., Schwartz, A., Samaha, F. J., & Kranias, E. G. (1983). Regulation of cardiac

209

sarcoplasmic reticulum calcium transport by calcium-calmodulin-dependent

phosphorylation. The Journal of Biological Chemistry, 258(22), 13587-13591.

Davis, M. G., Zhou, M., Ali, S., Coffin, J. D., Doetschman, T., & Dorn, I.,Gerald W. (1997).

Intracrine and autocrine effects of basic fibroblast growth factor in vascular smooth muscle

cells. Journal of Molecular and Cellular Cardiology, 29(4), 1061-1072. doi:DOI:

10.1006/jmcc.1997.0383

Dawson, D., Lygate, C. A., Zhang, M., Hulbert, K., Neubauer, S., & Casadei, B. (2005). nNOS

gene deletion exacerbates pathological left ventricular remodeling and functional

deterioration after myocardial infarction. Circulation, 112(24), 3729-3737.

doi:10.1161/CIRCULATIONAHA.105.539437

Debiais, F., Lemonnier, J., Hay, E., Delannoy, P., Caverzasio, J., & Marie, P. J. (2001). Fibroblast

growth factor-2 (FGF-2) increases N- expression through protein kinase C and src-

kinase pathways in human calvaria osteoblasts. Journal of Cellular Biochemistry, 81(1), 68-

81.

DeSantiago, J., Maier, L. S., & Bers, D. M. (2002). Frequency-dependent acceleration of

relaxation in the heart depends on CaMKII, but not phospholamban. Journal of Molecular

and Cellular Cardiology, 34(8), 975-984. doi:10.1006/jmcc.2002.2034

Detillieux, K. A., Sheikh, F., Kardami, E., & Cattini, P. A. (2003). Biological activities of

fibroblast growth factor-2 in the adult myocardium. Cardiovascular Research, 57(1), 8-19.

Dhalluin, C., Yan, K. S., Plotnikova, O., Lee, K. W., Zeng, L., Kuti, M., . . . Zhou, M. M. (2000).

Structural basis of SNT PTB domain interactions with distinct neurotrophic receptors.

Molecular Cell, 6(4), 921-929.

Diecke, S., Quiroga-Negreira, A., Redmer, T., & Besser, D. (2008). FGF2 signaling in mouse

embryonic fibroblasts is crucial for self-renewal of embryonic stem cells. Cells, Tissues,

210

Organs, 188(1-2), 52-61. doi:10.1159/000121282

Distasi, C., Torre, M., Antoniotti, S., Munaron, L., & Lovisolo, D. (1998). Neuronal survival and

calcium influx induced by basic fibroblast growth factor in chick ciliary ganglion neurons.

The European Journal of Neuroscience, 10(7), 2276-2286.

Doble, B. W., Ping, P., & Kardami, E. (2000). The epsilon subtype of protein kinase C is required

for cardiomyocyte connexin-43 phosphorylation. Circulation Research, 86(3), 293-301.

Dong Gao, W., Liu, Y., Mellgren, R., & Marban, E. (1996). Intrinsic myofilament alterations

underlying the decreased contractility of stunned myocardium : A consequence of Ca2+-

dependent proteolysis? Circulation Research, 78(3), 455-465. doi:10.1161/01.RES.78.3.455

Dong, S., Cheng, Y., Yang, J., Li, J., Liu, X., Wang, X., . . . Zhang, C. (2009). MicroRNA

expression signature and the role of MicroRNA-21 in the early phase of acute myocardial

infarction. Journal of Biological Chemistry, 284(43), 29514-29525.

doi:10.1074/jbc.M109.027896

Dong, W. J., Chandra, M., Xing, J., She, M., Solaro, R. J., & Cheung, H. C. (1997).

Phosphorylation-induced distance change in a cardiac muscle troponin I mutant.

Biochemistry, 36(22), 6754-6761. doi:10.1021/bi9622276

Dong, Y., Undyala, V. V., Gottlieb, R. A., Mentzer, R. M.,Jr, & Przyklenk, K. (2010). Autophagy:

Definition, molecular machinery, and potential role in myocardial ischemia-reperfusion

injury. Journal of Cardiovascular Pharmacology and Therapeutics, 15(3), 220-230.

doi:10.1177/1074248410370327

Dorn, G. W.,2nd, & Force, T. (2005). Protein kinase cascades in the regulation of cardiac

hypertrophy. The Journal of Clinical Investigation, 115(3), 527-537. doi:10.1172/JCI24178

Dorn, G. W., Souroujon, M. C., Liron, T., Chen, C., Gray, M. O., Zhou, H. Z., . . . Mochly-Rosen,

D. (1999). Sustained in vivo cardiac protection by a rationally designed peptide that causes

211

protein kinase C translocation. Proceedings of the National Academy of Sciences, 96(22),

12798-12803. doi:10.1073/pnas.96.22.12798

Downey, J. M., Davis, A. M., & Cohen, M. V. (2007). Signaling pathways in ischemic

preconditioning. Heart Failure Reviews, 12(3-4), 181-188. doi:10.1007/s10741-007-9025-2 du Toit, E. F., & Opie, L. H. (1994). Inhibitors of Ca2+ ATPase pump of sarcoplasmic reticulum

attenuate reperfusion stunning in isolated rat heart. Journal of Cardiovascular

Pharmacology, 24(4), 678-684.

Ebert, A. D., Laußmann, M., Wegehingel, S., Kaderali, L., Erfle, H., Reichert, J., . . . Nickel, W.

(2010). Tec-kinase-mediated phosphorylation of fibroblast growth factor 2 is essential for

unconventional secretion. Traffic, 11(6), 813-826. doi:10.1111/j.1600-0854.2010.01059.x

Eden, D., & Highsmith, S. (1997). Light chain-dependent myosin structural dynamics in solution

investigated by transient electrical birefringence. Biophysical Journal, 73(2), 952-958.

Edwards, A. G., Rees, M. L., Gioscia, R. A., Zachman, D. K., Lynch, J. M., Browder, J. C., et al.

(2009). PKC-permitted elevation of sarcolemmal KATP concentration may explain female-

specific resistance to myocardial infarction. The Journal of Physiology, 587(Pt 23), 5723-

5737.

El Idrissi, A., & Trenkner, E. (1999). Growth factors and taurine protect against excitotoxicity by

stabilizing calcium homeostasis and energy metabolism. The Journal of Neuroscience : The

Official Journal of the Society for Neuroscience, 19(21), 9459-9468.

Engel, P. L., Hinken, A., & Solaro, R. J. (2009). Differential effects of phosphorylation of

regions of troponin I in modifying cooperative activation of cardiac thin filaments. Journal

of Molecular and Cellular Cardiology, 47(3), 359-364. doi:10.1016/j.yjmcc.2009.04.016

Estival, A., Monzat, V., Miquel, K., Gaubert, F., Hollande, E., Korc, M., . . . Clemente, F. (1996).

Differential regulation of fibroblast growth factor (FGF) receptor-1 mRNA and protein by

212

two molecular forms of basic FGF. modulation of FGFR-1 mRNA stability. The Journal of

Biological Chemistry, 271(10), 5663-5670.

European Heart Rhythm Association, Heart Rhythm Society, Zipes, D. P., Camm, A. J.,

Borggrefe, M., Buxton, A. E., . . . European Society of Cardiology Committee for Practice

Guidelines. (2006). ACC/AHA/ESC 2006 guidelines for management of patients with

ventricular arrhythmias and the prevention of sudden cardiac death: A report of the american

college of Cardiology/American heart association task force and the european society of

cardiology committee for practice guidelines (writing committee to develop guidelines for

management of patients with ventricular arrhythmias and the prevention of sudden cardiac

death). Journal of the American College of Cardiology, 48(5), e247-346.

doi:10.1016/j.jacc.2006.07.010

Fabiato, A. (1983). Calcium-induced release of calcium from the cardiac sarcoplasmic reticulum.

The American Journal of Physiology, 245(1), C1-14.

Faham, S., Linhardt, R. J., & Rees, D. C. (1998). Diversity does make a difference: Fibroblast

growth factor-heparin interactions. Current Opinion in Structural Biology, 8(5), 578-586.

Fanburg, B., Finkel, R. M., & Martonosi, A. (1964). The role of calcium in the mechanism of

relaxation of cardiac muscle. The Journal of Biological Chemistry, 239, 2298-2305.

Farber, J. L., Chien, K. R., & Mittnacht, S.,Jr. (1981). Myocardial ischemia: The pathogenesis of

irreversible cell injury in ischemia. The American Journal of Pathology, 102(2), 271-281.

Favata, M. F., Horiuchi, K. Y., Manos, E. J., Daulerio, A. J., Stradley, D. A., Feeser, W. S., . . .

Trzaskos, J. M. (1998). Identification of a novel inhibitor of mitogen-activated protein

kinase kinase. Journal of Biological Chemistry, 273(29), 18623-18632.

doi:10.1074/jbc.273.29.18623

Fei, Y., Xiao, L., Doetschman, T., Coffin, D. J., & Hurley, M. M. (2011). Fibroblast growth factor

213

2 stimulation of osteoblast differentiation and bone formation is mediated by modulation of

the wnt signaling pathway. Journal of Biological Chemistry, 286(47), 40575-40583.

doi:10.1074/jbc.M111.274910

Ferreira, J. C., Brum, P. C., & Mochly-Rosen, D. (2011). betaIIPKC and epsilonPKC isozymes

as potential pharmacological targets in cardiac hypertrophy and heart failure. Journal of

Molecular and Cellular Cardiology, 51(4), 479-484.

Fiddes, J. C., Hebda, P. A., Hayward, P., Robson, M. C., Abraham, J. A., & Klingbeil, C. K.

(1991). Preclinical wound-healing studies with recombinant human basic fibroblast growth

factor. Annals of the New York Academy of Sciences, 638, 316-328.

Filatov, V. L., Katrukha, A. G., Bulargina, T. V., & Gusev, N. B. (1999). Troponin: Structure,

properties, and mechanism of functioning. Biochemistry.Biokhimiia, 64(9), 969-985.

Fill, M., & Copello, J. A. (2002). Ryanodine receptor calcium release channels. Physiological

Reviews, 82(4), 893-922. doi:10.1152/physrev.00013.2002

Filla, M. S., Dam, P., & Rapraeger, A. C. (1998). The cell surface proteoglycan syndecan-1

mediates fibroblast growth factor-2 binding and activity. Journal of Cellular Physiology,

174(3), 310-321. doi:10.1002/(SICI)1097-4652(199803)174:3<310::AID-JCP5>3.0.CO;2-R

Finley, N., Abbott, M. B., Abusamhadneh, E., Gaponenko, V., Dong, W., Gasmi-Seabrook, G., . .

. Rosevear, P. R. (1999). NMR analysis of cardiac troponin C-troponin I complexes: Effects

of phosphorylation. FEBS Letters, 453(1-2), 107-112.

Florkiewicz, R. Z., & Sommer, A. (1989). Human basic fibroblast growth factor gene encodes

four polypeptides: Three initiate translation from non-AUG codons. Proceedings of the

National Academy of Sciences of the United States of America, 86(11), 3978-3981.

Folkman, J., Klagsbrun, M., Sasse, J., Wadzinski, M., Ingber, D., & Vlodavsky, I. (1988). A

heparin-binding angiogenic protein--basic fibroblast growth factor--is stored within

214

basement membrane. The American Journal of Pathology, 130(2), 393-400.

Forstermann, U., Closs, E., Pollock, J., Nakane, M., Schwarz, P., Gath, I., & Kleinert, H. (1994).

Nitric oxide synthase isozymes. characterization, purification, molecular cloning, and

functions. Hypertension, 23(6), 1121-1131.

Freude, B., Masters, T. N., Robicsek, F., Fokin, A., Kostin, S., Zimmermann, R., . . . Schaper, J.

(2000). Apoptosis is initiated by myocardial ischemia and executed during reperfusion.

Journal of Molecular and Cellular Cardiology, 32(2), 197-208. doi:10.1006/jmcc.1999.1066

Fryer, R. M., Pratt, P. F., Hsu, A. K., & Gross, G. J. (2001). Differential activation of extracellular

signal regulated kinase isoforms in preconditioning and opioid-induced cardioprotection.

The Journal of Pharmacology and Experimental Therapeutics, 296(2), 642-649.

Fryer, R. M., Wang, Y., Hsu, A. K., & Gross, G. J. (2001). Essential activation of PKC-delta in

opioid-initiated cardioprotection. American Journal of Physiology.Heart and Circulatory

Physiology, 280(3), H1346-53.

Galinska, A., Hatch, V., Craig, R., Murphy, A. M., Van Eyk, J. E., Wang, C. L., . . . Foster, D. B.

(2010). The C terminus of cardiac troponin I stabilizes the Ca2+-activated state of

tropomyosin on actin filaments. Circulation Research, 106(4), 705-711.

doi:10.1161/CIRCRESAHA.109.210047

Gao, W. D., Atar, D., Backx, P. H., & Marban, E. (1995). Relationship between intracellular

calcium and contractile force in stunned myocardium. direct evidence for decreased

myofilament Ca2+ responsiveness and altered diastolic function in intact ventricular muscle.

Circulation Research, 76(6), 1036-1048.

Garcia-Rivas, G. J., & Torre-Amione, G. (2009). Abnormal mitochondrial function during

ischemia reperfusion provides targets for pharmacological therapy. Methodist DeBakey

Cardiovascular Journal, 5(3), 2-7.

215

Gaubert, F., Escaffit, F., Bertrand, C., Korc, M., Pradayrol, L., Clemente, F., & Estival, A.

(2001). Expression of the high molecular weight fibroblast growth factor-2 isoform of 210

amino acids is associated with modulation of protein kinases C delta and epsilon and ERK

activation. The Journal of Biological Chemistry, 276(2), 1545-1554.

doi:10.1074/jbc.M001184200

Gendron, R. L., Tsai, F. Y., Paradis, H., & Arceci, R. J. (1996). Induction of embryonic

vasculogenesis by bFGF and LIF in vitro and in vivo. Developmental Biology, 177(1), 332-

346. doi:10.1006/dbio.1996.0167

Goldfarb, M., Schoorlemmer, J., Williams, A., Diwakar, S., Wang, Q., Huang, X., . . . D'Angelo,

E. (2007). Fibroblast growth factor homologous factors control neuronal excitability through

modulation of voltage-gated sodium channels. Neuron, 55(3), 449-463.

doi:10.1016/j.neuron.2007.07.006

Gopalakrishna, R., Chen, Z., & Gundimeda, U. (1997). Selenocompounds induce a redox

modulation of protein kinase C in the cell, compartmentally independent from cytosolic

glutathione: Its role in inhibition of tumor promotion, Archives of Biochemistry and

Biophysics, 348(1), 37-48. doi:DOI: 10.1006/abbi.1997.0335

Gospodarowicz, D. (1974). Localisation of a fibroblast growth factor and its effect alone and

with hydrocortisone on 3T3 . Nature, 249(453), 123-127.

Gospodarowicz, D. (1975). Purification of a fibroblast growth factor from bovine pituitary.

Journal of Biological Chemistry, 250(7), 2515-2520.

Goto, M., Liu, Y., Yang, X., Ardell, J. L., Cohen, M. V., & Downey, J. M. (1995). Role of

bradykinin in protection of ischemic preconditioning in rabbit hearts. Circulation Research,

77(3), 611-621.

Gray, M. O., Karliner, J. S., & Mochly-Rosen, D. (1997). A selective ε-protein kinase C

216

antagonist inhibits protection of cardiac myocytes from hypoxia-induced cell death. Journal

of Biological Chemistry, 272(49), 30945-30951. doi:10.1074/jbc.272.49.30945

Greene, L. E., & Eisenberg, E. (1988). Relationship between regulated actomyosin ATPase

activity and cooperative binding of myosin to regulated actin. Cell Biophysics, 12, 59-71.

Grinwald, P. M. (1982). Calcium uptake during post-ischemic reperfusion in the isolated rat

heart: Influence of extracellular sodium. Journal of Molecular and Cellular Cardiology,

14(6), 359-365. doi:DOI: 10.1016/0022-2828(82)90251-6

Grinwald, P. M., & Nayler, W. G. (1981). Calcium entry in the calcium paradox. Journal of

Molecular and Cellular Cardiology, 13(10), 867-880.

Grunz, H., McKeehan, W. L., Knochel, W., Born, J., Tiedemann, H., & Tiedemann, H. (1988).

Induction of mesodermal tissues by acidic and basic heparin binding growth factors. Cell

Differentiation, 22(3), 183-189.

Guimond, S. E., & Turnbull, J. E. (1999). Fibroblast growth factor receptor signalling is dictated

by specific heparan sulphate saccharides. Current Biology : CB, 9(22), 1343-1346.

Guo, Y., Li, Q., Wu, W. J., Tan, W., Zhu, X., Mu, J., & Bolli, R. (2008). Endothelial nitric oxide

synthase is not necessary for the early phase of ischemic preconditioning in the mouse.

Journal of Molecular and Cellular Cardiology, 44(3), 496-501.

doi:10.1016/j.yjmcc.2008.01.003

Guo, Y., Wu, W. J., Qiu, Y., Tang, X. L., Yang, Z., & Bolli, R. (1998). Demonstration of an early

and a late phase of ischemic preconditioning in mice. The American Journal of Physiology,

275(4 Pt 2), H1375-87.

Györke, S., Stevens, S. C. W., & Terentyev, D. (2009). Cardiac calsequestrin: Quest inside the

SR. The Journal of Physiology, 587(13), 3091-3094. doi:10.1113/jphysiol.2009.172049

Haghighi, K., Chen, G., Sato, Y., Fan, G. C., He, S., Kolokathis, F., . . . Kranias, E. G. (2008). A

217

human phospholamban promoter polymorphism in alters

transcriptional regulation by glucocorticoids. Human Mutation, 29(5), 640-647.

doi:10.1002/humu.20692

Haimovitz-Friedman, A., Balaban, N., McLoughlin, M., Ehleiter, D., Michaeli, J., Vlodavsky, I.,

& Fuks, Z. (1994). Protein kinase C mediates basic fibroblast growth factor protection of

endothelial cells against radiation-induced apoptosis. Cancer Research, 54(10), 2591-2597.

Hambleton, M., Hahn, H., Pleger, S. T., Kuhn, M. C., Klevitsky, R., Carr, A. N., . . . Molkentin,

J. D. (2006). Pharmacological- and gene therapy-based inhibition of protein kinase

Calpha/beta enhances cardiac contractility and attenuates heart failure. Circulation, 114(6),

574-582. doi:10.1161/CIRCULATIONAHA.105.592550

Hampton, T. G., Amende, I., Fong, J., Laubach, V. E., Li, J., Metais, C., & Simons, M. (2000).

Basic FGF reduces stunning via a NOS2-dependent pathway in coronary-perfused mouse

hearts. American Journal of Physiology.Heart and Circulatory Physiology, 279(1), H260-8.

Hartman, T. J., Martin, J. L., Solaro, R. J., Samarel, A. M., & Russell, B. (2009). CapZ dynamics

are altered by endothelin-1 and phenylephrine via PIP2- and PKC-dependent mechanisms.

American Journal of Physiology.Cell Physiology, 296(5), C1034-9.

doi:10.1152/ajpcell.00544.2008

Hasdai, D., Barak, V., Leibovitz, E., Herz, I., Sclarovsky, S., Eldar, M., & Scheinowitz, M.

(1997). Serum basic fibroblast growth factor levels in patients with ischemic heart disease.

International Journal of Cardiology, 59(2), 133-138.

Hassouna, A., Matata, B. M., & Galinanes, M. (2004). PKC-epsilon is upstream and PKC-alpha

is downstream of mitoKATP channels in the signal transduction pathway of ischemic

preconditioning of human myocardium. American Journal of Physiology.Cell Physiology,

287(5), C1418-25. doi:10.1152/ajpcell.00144.2004

218

Hatzinikolaou-Kotsakou, E., Tziakas, D., Hotidis, A., Stakos, D., Floros, D., Mavridis, A., . . .

Hatseras, D. I. (2007). Could sustained monomorphic ventricular tachycardia in the early

phase of a prime acute myocardial infarction affect patient outcome? Journal of

Electrocardiology, 40(1), 72-77. doi:10.1016/j.jelectrocard.2006.02.004

Hebda, P. A., Klingbeil, C. K., Abraham, J. A., & Fiddes, J. C. (1990). Basic fibroblast growth

factor stimulation of epidermal wound healing in pigs. The Journal of Investigative

Dermatology, 95(6), 626-631.

Hellman, U., Hellstrom, M., Morner, S., Engstrom-Laurent, A., Aberg, A. M., Oliviero, P., . . .

Waldenstrom, A. (2008). Parallel up-regulation of FGF-2 and hyaluronan during

development of cardiac hypertrophy in rat. Cell and Tissue Research, 332(1), 49-56.

doi:10.1007/s00441-007-0562-8

Heyndrickx, G. R., Millard, R. W., McRitchie, R. J., Maroko, P. R., & Vatner, S. F. (1975).

Regional myocardial functional and electrophysiological alterations after brief coronary

artery occlusion in conscious dogs. The Journal of Clinical Investigation, 56(4), 978-985.

doi:10.1172/JCI108178

Hidalgo, C., Hudson, B., Bogomolovas, J., Zhu, Y., Anderson, B., Greaser, M., . . . Granzier, H.

(2009). PKC phosphorylation of titin's PEVK element: A novel and conserved pathway for

modulating myocardial stiffness. Circulation Research, 105(7), 631-8, 17 p following 638.

doi:10.1161/CIRCRESAHA.109.198465

Hirst, C. J., Herlyn, M., Cattini, P. A., & Kardami, E. (2003). High levels of CUG-initiated FGF-

2 expression cause chromatin compaction, decreased cardiomyocyte mitosis, and cell death.

Molecular and Cellular Biochemistry, 246(1-2), 111-116.

Horowitz, A., Tkachenko, E., & Simons, M. (2002). Fibroblast growth factor-specific

modulation of cellular response by syndecan-4. The Journal of Cell Biology, 157(4), 715-

219

725. doi:10.1083/jcb.200112145

Horrigan, M. C., Malycky, J. L., Ellis, S. G., Topol, E. J., & Nicolini, F. A. (1999). Reduction in

myocardial infarct size by basic fibroblast growth factor following coronary occlusion in a

canine model. International Journal of Cardiology, 68 Suppl 1, S85-91.

Horwitz, O., Sayen, J. J., Sheldon, W. F., & Kuo, P. T. (1950). Experimental studies of

intramyocardial oxygen tension: Increases consequent on breathing pure oxygen in normal

hearts and at the borders of ischaemic areas. The Journal of Clinical Investigation, 29(6),

823-824.

House, S. L., Bolte, C., Zhou, M., Doetschman, T., Klevitsky, R., Newman, G., & Schultz Jel, J.

(2003). Cardiac-specific overexpression of fibroblast growth factor-2 protects against

myocardial dysfunction and infarction in a murine model of low-flow ischemia. Circulation,

108(25), 3140-3148. doi:10.1161/01.CIR.0000105723.91637.1C

House, S. L., Branch, K., Newman, G., Doetschman, T., & Schultz Jel, J. (2005).

Cardioprotection induced by cardiac-specific overexpression of fibroblast growth factor-2 is

mediated by the MAPK cascade. American Journal of Physiology.Heart and Circulatory

Physiology, 289(5), H2167-75. doi:10.1152/ajpheart.00392.2005

House, S. L., Melhorn, S. J., Newman, G., Doetschman, T., & Schultz Jel, J. (2007). The protein

kinase C pathway mediates cardioprotection induced by cardiac-specific overexpression of

fibroblast growth factor-2. American Journal of Physiology.Heart and Circulatory

Physiology, 293(1), H354-65. doi:10.1152/ajpheart.00804.2006

Hovnanian, A. (2007). SERCA pumps and human diseases. Sub-Cellular Biochemistry, 45, 337-

363.

Hrzenjak, M., & Shain, S. A. (1997). Fibroblast growth factor-2 and TPA enhance prostate-

cancer-cell proliferation and activate members of the ras and PKC signal transduction

220

pathways. Receptors & Signal Transduction, 7(4), 207-219.

Hsu, H. Y., Nicholson, A. C., & Hajjar, D. P. (1994). Basic fibroblast growth factor-induced low

density lipoprotein receptor transcription and surface expression. signal transduction

pathways mediated by the bFGF receptor tyrosine kinase. The Journal of Biological

Chemistry, 269(12), 9213-9220.

Huang, Z., Chen, K., Huang, P. L., Finklestein, S. P., & Moskowitz, M. A. (1997). bFGF

ameliorates focal ischemic injury by blood flow-independent mechanisms in eNOS mutant

mice. The American Journal of Physiology, 272(3 Pt 2), H1401-5.

Hughes, S. E. (1997). Differential expression of the fibroblast growth factor receptor (FGFR)

multigene family in normal human adult tissues. The Journal of Histochemistry and

Cytochemistry : Official Journal of the Histochemistry Society, 45(7), 1005-1019.

Huke, S., & Bers, D. M. (2007). Temporal dissociation of frequency-dependent acceleration of

relaxation and protein phosphorylation by CaMKII. Journal of Molecular and Cellular

Cardiology, 42(3), 590-599. doi:10.1016/j.yjmcc.2006.12.007

Hunter, R. B., Mitchell-Felton, H., Essig, D. A., & Kandarian, S. C. (2001). Expression of

endoplasmic reticulum stress proteins during skeletal muscle disuse atrophy. American

Journal of Physiology.Cell Physiology, 281(4), C1285-90.

Hurley, J. H., Newton, A. C., Parker, P. J., Blumberg, P. M., & Nishizuka, Y. (1997). Taxonomy

and function of C1 protein kinase C homology domains. Protein Science : A Publication of

the Protein Society, 6(2), 477-480. doi:10.1002/pro.5560060228

Huxley, A. F., & Simmons, R. M. (1971). Proposed mechanism of force generation in striated

muscle. Nature, 233(5321), 533-538.

Iio, T. (1985). Conformational change of troponin T induced by calcium binding to troponin C.

Journal of Biochemistry, 98(1), 261-263.

221

Im, H. J., Muddasani, P., Natarajan, V., Schmid, T. M., Block, J. A., Davis, F., . . . Loeser, R. F.

(2007). Basic fibroblast growth factor stimulates matrix metalloproteinase-13 via the

molecular cross-talk between the mitogen-activated protein kinases and protein kinase cdelta

pathways in human adult articular chondrocytes. The Journal of Biological Chemistry,

282(15), 11110-11121. doi:10.1074/jbc.M609040200

Imani, A., Faghihi, M., Sadr, S. S., Niaraki, S. S., & Alizadeh, A. M. (2011). Noradrenaline

protects in vivo rat heart against infarction and ventricular arrhythmias via nitric oxide and

reactive oxygen species. Journal of Surgical Research, 169(1), 9-15. doi:DOI:

10.1016/j.jss.2009.10.025

Inagaki, K., Hahn, H. S., Dorn, G. W.,2nd, & Mochly-Rosen, D. (2003). Additive protection of

the ischemic heart ex vivo by combined treatment with delta-protein kinase C inhibitor and

epsilon-protein kinase C activator. Circulation, 108(7), 869-875.

doi:10.1161/01.CIR.0000081943.93653.73

Inagaki, K., Chen, L., Ikeno, F., Lee, F. H., Imahashi, K., Bouley, D. M., . . . Mochly-Rosen, D.

(2003). Inhibition of δ-protein kinase C protects against reperfusion injury of the ischemic

heart in vivo. Circulation, 108(19), 2304-2307. doi:10.1161/01.CIR.0000101682.24138.36

Irisawa, H. (1987). Membrane currents in cardiac pacemaker tissue. Experientia, 43(11-12),

1131-1135.

Ito, T., Sawada, R., Fujiwara, Y., & Tsuchiya, T. (2008). FGF-2 increases osteogenic and

chondrogenic differentiation potentials of human mesenchymal stem cells by inactivation of

TGF-beta signaling. Cytotechnology, 56(1), 1-7. doi:10.1007/s10616-007-9092-1

Itoh, N., & Ornitz, D. M. (2008). Functional evolutionary history of the mouse fgf gene family.

Developmental Dynamics : An Official Publication of the American Association of

Anatomists, 237(1), 18-27. doi:10.1002/dvdy.21388

222

Itoh, N., & Ornitz, D. M. (2011). Fibroblast growth factors: From molecular evolution to roles in

development, metabolism and disease. Journal of Biochemistry, 149(2), 121-130.

doi:10.1093/jb/mvq121

Iwabuchi, T., & Goetinck, P. F. (2006). Syndecan-4 dependent FGF stimulation of mouse

vibrissae growth. Mechanisms of Development, 123(11), 831-841. doi:DOI:

10.1016/j.mod.2006.08.003

Iwai-Kanai, E., Hasegawa, K., Fujita, M., Araki, M., Yanazume, T., Adachi, S., & Sasayama, S.

(2002). Basic fibroblast growth factor protects cardiac myocytes from iNOS-mediated

apoptosis. Journal of Cellular Physiology, 190(1), 54-62. doi:10.1002/jcp.10036

Iwamoto, M., Shimazu, A., Nakashima, K., Suzuki, F., & Kato, Y. (1991). Reduction of basic

fibroblasts growth factor receptor is coupled with terminal differentiation of chondrocytes.

The Journal of Biological Chemistry, 266(1), 461-467.

Jackson, C. L., & Reidy, M. A. (1993). Basic fibroblast growth factor: Its role in the control of

smooth muscle cell migration. The American Journal of Pathology, 143(4), 1024-1031.

Jennings, R. B., & Reimer, K. A. (1991). The cell biology of acute myocardial ischemia. Annual

Review of Medicine, 42, 225-246. doi:10.1146/annurev.me.42.020191.001301

Jiang, Z. S., Padua, R. R., Ju, H., Doble, B. W., Jin, Y., Hao, J., . . . Kardami, E. (2002). Acute

protection of ischemic heart by FGF-2: Involvement of FGF-2 receptors and protein kinase

C. American Journal of Physiology.Heart and Circulatory Physiology, 282(3), H1071-80.

doi:10.1152/ajpheart.00290.2001

Jiang, Z. S., Srisakuldee, W., Soulet, F., Bouche, G., & Kardami, E. (2004). Non-angiogenic

FGF-2 protects the ischemic heart from injury, in the presence or absence of reperfusion.

Cardiovascular Research, 62(1), 154-166. doi:10.1016/j.cardiores.2004.01.009

Jiang, Z. S., Wen, G. B., Tang, Z. H., Srisakuldee, W., Fandrich, R. R., & Kardami, E. (2009).

223

High molecular weight FGF-2 promotes postconditioning-like cardioprotection linked to

activation of the protein kinase C isoforms akt and p70 S6 kinase. Canadian Journal of

Physiology and Pharmacology, 87(10), 798-804. doi:10.1139/y09-049

Johannes, F. J., Prestle, J., Eis, S., Oberhagemann, P., & Pfizenmaier, K. (1994). PKCu is a

novel, atypical member of the protein kinase C family. The Journal of Biological Chemistry,

269(8), 6140-6148.

Johnson, G. L., & Vaillancourt, R. R. (1994). Sequential protein kinase reactions controlling cell

growth and differentiation. Current Opinion in Cell Biology, 6(2), 230-238.

Jones, S. P., & Bolli, R. (2006). The ubiquitous role of nitric oxide in cardioprotection. Journal

of Molecular and Cellular Cardiology, 40(1), 16-23. doi:DOI: 10.1016/j.yjmcc.2005.09.011

Jung, F., Palmer, L. A., Zhou, N., & Johns, R. A. (2000). Hypoxic regulation of inducible nitric

oxide synthase via hypoxia inducible factor-1 in cardiac myocytes. Circulation Research,

86(3), 319-325.

Kajita, Y., Takayasu, M., Yoshida, J., Dietrich, H. H., & Dacey, R. G.,Jr. (2001). Vasodilatory

effect of basic fibroblast growth factor in isolated rat cerebral arterioles: Mechanisms

involving nitric oxide and membrane hyperpolarization. Neurologia Medico-Chirurgica,

41(4), 177-85; discussion 185-6.

Kanaya, N., Gable, B., Murray, P. A., & Damron, D. S. (2003). Propofol increases

phosphorylation of troponin I and myosin light chain 2 via protein kinase C activation in

cardiomyocytes. Anesthesiology, 98(6), 1363-1371.

Kaneda, K., Miyamae, M., Sugioka, S., Okusa, C., Inamura, Y., Domae, N., . . . Figueredo, V. M.

(2008). Sevoflurane enhances ethanol-induced cardiac preconditioning through modulation

of protein kinase C, mitochondrial KATP channels, and nitric oxide synthase, in guinea pig

hearts. Anesthesia and Analgesia, 106(1), 9-16, table of contents.

224

doi:10.1213/01.ane.0000297298.93627.36

Kardami, E., Liu, L., Pasumarthi, S. K., Doble, B. W., & Cattini, P. A. (1995). Regulation of

basic fibroblast growth factor (bFGF) and FGF receptors in the heart. Annals of the New

York Academy of Sciences, 752, 353-369.

Kardami, E., Jiang, Z. S., Jimenez, S. K., Hirst, C. J., Sheikh, F., Zahradka, P., et al. (2004).

Fibroblast growth factor 2 isoforms and cardiac hypertrophy. Cardiovascular Research,

63(3), 458-466.

Karlsson, T., Songyang, Z., Landgren, E., Lavergne, C., Di Fiore, P. P., Anafi, M., . . . Welsh, M.

(1995). Molecular interactions of the src homology 2 domain protein shb with

phosphotyrosine residues, tyrosine kinase receptors and src homology 3 domain proteins.

Oncogene, 10(8), 1475-1483.

Kawano, T., Zoga, V., Kimura, M., Liang, M. Y., Wu, H. E., Gemes, G., et al. (2009). Nitric

oxide activates ATP-sensitive potassium channels in mammalian sensory neurons: Action

by direct S-nitrosylation. Molecular Pain, 5, 12.

Kelly, R. A., Balligand, J., & Smith, T. W. (1996). Nitric oxide and cardiac function. Circulation

Research, 79(3), 363-380.

Kennelly, P. J., & Krebs, E. G. (1991). Consensus sequences as substrate specificity determinants

for protein kinases and protein phosphatases. Journal of Biological Chemistry, 266(24),

15555-15558.

Kent, K. C., Mii, S., Harrington, E. O., Chang, J. D., Mallette, S., & Ware, J. A. (1995).

Requirement for protein kinase C activation in basic fibroblast growth factor-induced human

endothelial cell proliferation. Circulation Research, 77(2), 231-238.

Kessler, W., Budde, T., Gekle, M., Fabian, A., & Schwab, A. (2008). Activation of cell migration

with fibroblast growth factor-2 requires calcium-sensitive potassium channels. Pflugers

225

Archiv : European Journal of Physiology, 456(5), 813-823. doi:10.1007/s00424-008-0452-2

Keum, E., Kim, Y., Kim, J., Kwon, S., Lim, Y., Han, I., & Oh, E. S. (2004). Syndecan-4 regulates

localization, activity and stability of protein kinase C-alpha. The Biochemical Journal,

378(Pt 3), 1007-1014. doi:10.1042/BJ20031734

Khan, M., Mohan, I. K., Kutala, V. K., Kotha, S. R., Parinandi, N. L., Hamlin, R. L., &

Kuppusamy, P. (2009). Sulfaphenazole protects heart against ischemia-reperfusion injury

and cardiac dysfunction by overexpression of iNOS, leading to enhancement of nitric oxide

bioavailability and tissue oxygenation. Antioxidants & Redox Signaling, 11(4), 725-738.

doi:10.1089/ARS.2008.2155

Kikkawa, U., Ogita, K., Ono, Y., Asaoka, Y., Shearman, M. S., Fujii, T., . . . Nishizuka, Y. (1987).

The common structure and activities of four subspecies of rat brain protein kinase C family.

FEBS Letters, 223(2), 212-216. doi:10.1016/0014-5793(87)80291-0

Kim, M. Y., Kim, M. J., Yoon, I. S., Ahn, J. H., Lee, S. H., Baik, E. J., et al. (2006). Diazoxide

acts more as a PKC-epsilon activator, and indirectly activates the mitochondrial K(ATP)

channel conferring cardioprotection against hypoxic injury. British Journal of

Pharmacology, 149(8), 1059-1070.

Kimura-Kurosawa, S., Kanaya, N., Kamada, N., Hirata, N., Nakayama, M., & Namiki, A.

(2007). Cardioprotective effect and mechanism of action of landiolol on the ischemic

reperfused heart. Journal of Anesthesia, 21(4), 480-489. doi:10.1007/s00540-007-0558-2

Kiraly, M., Porcsalmy, B., Pataki, A., Kadar, K., Jelitai, M., Molnar, B., . . . Varga, G. (2009).

Simultaneous PKC and cAMP activation induces differentiation of human dental pulp stem

cells into functionally active neurons. Neurochemistry International, 55(5), 323-332.

doi:10.1016/j.neuint.2009.03.017

Kirchberber, M. A., Tada, M., & Katz, A. M. (1975). Phospholamban: A regulatory protein of the

226

cardiac sarcoplasmic reticulum. Recent Advances in Studies on Cardiac Structure and

Metabolism, 5, 103-115.

Klagsbrun, M., & Baird, A. (1991). A dual receptor system is required for basic fibroblast growth

factor activity. Cell, 67(2), 229-231.

Klint, P., Kanda, S., & Claesson-Welsh, L. (1995). Shc and a novel 89-kDa component couple to

the Grb2-sos complex in fibroblast growth factor-2-stimulated cells. The Journal of

Biological Chemistry, 270(40), 23337-23344.

Kloner, R. A., & Schwartz Longacre, L. (2011). State of the science of cardioprotection:

Challenges and opportunities-- proceedings of the 2010 NHLBI workshop on

cardioprotection. Journal of Cardiovascular Pharmacology and Therapeutics, 16(3-4), 223-

232. doi:10.1177/1074248411402501

Knapp, L. T., Kanterewicz, B. I., Hayes, E. L., & Klann, E. (2001). Peroxynitrite-induced

tyrosine nitration and inhibition of protein kinase C. Biochemical and Biophysical Research

Communications, 286(4), 764-770. doi:DOI: 10.1006/bbrc.2001.5448

Knighton, D. R., Phillips, G. D., & Fiegel, V. D. (1990). Wound healing angiogenesis: Indirect

stimulation by basic fibroblast growth factor. The Journal of Trauma, 30(12 Suppl), S134-

44.

Knighton, D. R., Zheng, J. H., Ten Eyck, L. F., Ashford, V. A., Xuong, N. H., Taylor, S. S., &

Sowadski, J. M. (1991). Crystal structure of the catalytic subunit of cyclic adenosine

monophosphate-dependent protein kinase. Science (New York, N.Y.), 253(5018), 407-414.

Kohlhardt, M., Krause, H., Kubler, M., & Herdey, A. (1975). Kinetics of inactivation and

recovery of the slow inward current in the mammalian ventricular myocardium. Pflugers

Archiv : European Journal of Physiology, 355(1), 1-17.

Komander, D., Kular, G. S., Schuttelkopf, A. W., Deak, M., Prakash, K. R., Bain, J., . . . van

227

Aalten, D. M. (2004). Interactions of LY333531 and other bisindolyl maleimide inhibitors

with PDK1. Structure (London, England : 1993), 12(2), 215-226.

doi:10.1016/j.str.2004.01.005

Kondo, H., Matsuda, R., & Yonezawa, Y. (1993). Autonomous migration of human fetal skin

fibroblasts into a denuded area in a cell monolayer is mediated by basic fibroblast growth

factor and collagen. In Vitro Cellular & Developmental Biology.Animal, 29A(12), 929-935.

Kooij, V., Boontje, N., Zaremba, R., Jaquet, K., Dos Remedios, C., Stienen, G. J., & van der

Velden, J. (2009). Protein kinase C alpha and epsilon phosphorylation of troponin and

myosin binding protein C reduce ca(2+) sensitivity in human myocardium. Basic Research

in Cardiology, doi:10.1007/s00395-009-0053-z

Kostyk, S. K., Kourembanas, S., Wheeler, E. L., Medeiros, D., McQuillan, L. P., D'Amore, P. A.,

& Braunhut, S. J. (1995). Basic fibroblast growth factor increases nitric oxide synthase

production in bovine endothelial cells. American Journal of Physiology - Heart and

Circulatory Physiology, 269(5), H1583-H1589.

Koyama, T., Boston, D., Ikenouchi, H., & Barry, W. H. (1996). Survival of metabolically

inhibited ventricular myocytes is enhanced by inhibition of rigor and SR Ca2+ cycling. The

American Journal of Physiology, 271(2 Pt 2), H643-50.

Kozawa, O., Suzuki, A., & Uematsu, T. (1997). Basic fibroblast growth factor induces

interleukin-6 synthesis in osteoblasts: Autoregulation by protein kinase C. Cellular

Signalling, 9(6), 463-468.

Kubler, W., & Spieckermann, P. G. (1970). Regulation of glycolysis in the ischemic and the

anoxic myocardium. Journal of Molecular and Cellular Cardiology, 1(4), 351-377.

Kudo, M., Wang, Y., Xu, M., Ayub, A., & Ashraf, M. (2002). Adenosine A(1) receptor mediates

late preconditioning via activation of PKC-delta signaling pathway. American Journal of

228

Physiology.Heart and Circulatory Physiology, 283(1), H296-301.

doi:10.1152/ajpheart.01087.2001

Kuhlmann, C. R., Wu, Y., Li, F., Munz, B. M., Tillmanns, H., Waldecker, B., & Wiecha, J.

(2004). bFGF activates endothelial Ca2+-activated K+ channels involving G-proteins and

tyrosine kinases. Vascular Pharmacology, 41(6), 181-186. doi:10.1016/j.vph.2004.10.003

Kurosu, H., Choi, M., Ogawa, Y., Dickson, A. S., Goetz, R., Eliseenkova, A. V., . . . Kuro-o, M.

(2007). Tissue-specific expression of betaKlotho and fibroblast growth factor (FGF) receptor

isoforms determines metabolic activity of FGF19 and FGF21. The Journal of Biological

Chemistry, 282(37), 26687-26695. doi:10.1074/jbc.M704165200

Kurosu, H., Ogawa, Y., Miyoshi, M., Yamamoto, M., Nandi, A., Rosenblatt, K. P., . . . Kuro-o,

M. (2006). Regulation of fibroblast growth factor-23 signaling by klotho. The Journal of

Biological Chemistry, 281(10), 6120-6123. doi:10.1074/jbc.C500457200

Kusuoka, H., Koretsune, Y., Chacko, V., Weisfeldt, M., & Marban, E. (1990). Excitation-

contraction coupling in postischemic myocardium. does failure of activator Ca2+ transients

underlie stunning? Circulation Research, 66(5), 1268-1276.

Kuwayama, H., & Yagi, K. (1980). Localization of g2 light chain in the link between the heads

and tail of cardiac myosin. Journal of Biochemistry, 87(6), 1603-1607.

Landgren, E., Klint, P., Yokote, K., & Claesson-Welsh, L. (1998). Fibroblast growth factor

receptor-1 mediates chemotaxis independently of direct SH2-domain protein binding.

Oncogene, 17(3), 283-291. doi:10.1038/sj.onc.1201936

Landstrom, A. P., Adekola, B. A., Bos, J. M., Ommen, S. R., & Ackerman, M. J. (2011). PLN-

encoded phospholamban mutation in a large cohort of hypertrophic cardiomyopathy cases:

Summary of the literature and implications for genetic testing. American Heart Journal,

161(1), 165-171. doi:10.1016/j.ahj.2010.08.001

229

Larsson, H., Klint, P., Landgren, E., & Claesson-Welsh, L. (1999). Fibroblast growth factor

receptor-1-mediated endothelial cell proliferation is dependent on the src homology (SH)

2/SH3 domain-containing adaptor protein crk. The Journal of Biological Chemistry,

274(36), 25726-25734.

Lavine, K. J., Yu, K., White, A. C., Zhang, X., Smith, C., Partanen, J., & Ornitz, D. M. (2005).

Endocardial and epicardial derived FGF signals regulate myocardial proliferation and

differentiation in vivo. Developmental Cell, 8(1), 85-95. doi:10.1016/j.devcel.2004.12.002

Leconte, I., Fox, J. C., Baldwin, H. S., Buck, C. A., & Swain, J. L. (1998). Adenoviral-mediated

expression of antisense RNA to fibroblast growth factors disrupts murine vascular

development. Developmental Dynamics, 213(4), 421-430. doi:10.1002/(SICI)1097-

0177(199812)213:4<421::AID-AJA7>3.0.CO;2-B

Lefer, A. M., Tsao, P., Aoki, N., & Palladino, M. A.,Jr. (1990). Mediation of cardioprotection by

transforming growth factor-beta. Science (New York, N.Y.), 249(4964), 61-64.

Li, M. X., Wang, X., Lindhout, D. A., Buscemi, N., Van Eyk, J. E., & Sykes, B. D. (2003).

Phosphorylation and mutation of human cardiac troponin I deferentially destabilize the

interaction of the functional regions of troponin I with troponin C. Biochemistry, 42(49),

14460-14468. doi:10.1021/bi035408y

Li, Q., Guo, Y., Xuan, Y., Lowenstein, C. J., Stevenson, S. C., Prabhu, S. D., . . . Bolli, R. (2003).

Gene therapy with inducible nitric oxide synthase protects against myocardial infarction via

a cyclooxygenase-2–Dependent mechanism. Circulation Research, 92(7), 741-748.

doi:10.1161/01.RES.0000065441.72685.29

Liao, S. (2008). The role of fibroblast growth factor-2 isoforms in ischemia-reperfusion injury

and cardioprotection.

Liao, S., Bodmer, J., Pietras, D., Azhar, M., Doetschman, T., & Schultz Jel, J. (2009). Biological

230

functions of the low and high molecular weight protein isoforms of fibroblast growth factor-

2 in cardiovascular development and disease. Developmental Dynamics : An Official

Publication of the American Association of Anatomists, 238(2), 249-264.

doi:10.1002/dvdy.21677

Liao, S., Bodmer, J. R., Azhar, M., Newman, G., Coffin, J. D., Doetschman, T., & Schultz Jel, J.

(2010). The influence of FGF2 high molecular weight (HMW) isoforms in the development

of cardiac ischemia-reperfusion injury. Journal of Molecular and Cellular Cardiology,

48(6), 1245-1254. doi:10.1016/j.yjmcc.2010.01.014

Liao, S., Porter, D., Scott, A., Newman, G., Doetschman, T., & Schultz Jel, J. (2007). The

cardioprotective effect of the low molecular weight isoform of fibroblast growth factor-2:

The role of JNK signaling. Journal of Molecular and Cellular Cardiology, 42(1), 106-120.

doi:10.1016/j.yjmcc.2006.10.005

Liron, T., Chen, L. E., Khaner, H., Vallentin, A., & Mochly-Rosen, D. (2007). Rational design of

a selective antagonist of epsilon protein kinase C derived from the selective allosteric

agonist, pseudo-RACK peptide. Journal of Molecular and Cellular Cardiology, 42(4), 835-

841. doi:10.1016/j.yjmcc.2007.01.007

Liu, C., Dib-Hajj, S. D., & Waxman, S. G. (2001). Fibroblast growth factor homologous factor

1B binds to the C terminus of the tetrodotoxin-resistant rNav1.9a (NaN).

The Journal of Biological Chemistry, 276(22), 18925-18933. doi:10.1074/jbc.M101606200

Liu, G. S., Cohen, M. V., Mochly-Rosen, D., & Downey, J. M. (1999). Protein kinase C-ε is

responsible for the protection of preconditioning in rabbit cardiomyocytes. Journal of

Molecular and Cellular Cardiology, 31(10), 1937-1948.

Liu, Q., Chen, X., Macdonnell, S. M., Kranias, E. G., Lorenz, J. N., Leitges, M., . . . Molkentin,

J. D. (2009). Protein kinase C{alpha}, but not PKC{beta} or PKC{gamma}, regulates

231

contractility and heart failure susceptibility: Implications for ruboxistaurin as a novel

therapeutic approach. Circulation Research, 105(2), 194-200.

doi:10.1161/CIRCRESAHA.109.195313

Liu, X., Wu, X., Cai, L., & Sun, S. (2008). Calreticulin downregulation is associated with FGF-

2-induced angiogenesis through calcineurin pathway in ischemic myocardium. Shock

(Augusta, Ga.), 29(1), 140-148. doi:10.1097/shk.0b013e318123e822

Liu, X. H., Zhang, Z. Y., Andersson, K. B., Husberg, C., Enger, U. H., Ræder, M. G., . . . Louch,

W. E. (2011). Cardiomyocyte-specific disruption of Serca2 in adult mice causes

sarco(endo)plasmic reticulum stress and apoptosis. Cell Calcium, 49(4), 201-207. doi:DOI:

10.1016/j.ceca.2010.09.009

Lloyd-Jones, D., Adams, R. J., Brown, T. M., Carnethon, M., Dai, S., De Simone, G., . . .

American Heart Association Statistics Committee and Stroke Statistics Subcommittee.

(2010). Executive summary: Heart disease and stroke statistics--2010 update: A report from

the american heart association. Circulation, 121(7), 948-954.

doi:10.1161/CIRCULATIONAHA.109.192666

Lochner, A., Marais, E., Genade, S., Huisamen, B., du Toit, E. F., & Moolman, J. A. (2009).

Protection of the ischaemic heart: Investigations into the phenomenon of ischaemic

preconditioning. Cardiovascular Journal of Africa, 20(1), 43-51.

Lombardi, V., Piazzesi, G., Ferenczi, M. A., Thirlwell, H., Dobbie, I., & Irving, M. (1995).

Elastic distortion of myosin heads and repriming of the working stroke in muscle. Nature,

374(6522), 553-555. doi:10.1038/374553a0

Louis, J. C., Magal, E., Gerdes, W., & Seifert, W. (1993). Survival-promoting and protein kinase

C-regulating roles of basic FGF for hippocampal neurons exposed to phorbol ester,

glutamate and ischaemia-like conditions. The European Journal of Neuroscience, 5(12),

232

1610-1621.

Loukianov, E., Ji, Y., Grupp, I. L., Kirkpatrick, D. L., Baker, D. L., Loukianova, T., . . .

Periasamy, M. (1998). Enhanced and increased Ca2+ transport

function in transgenic hearts expressing the fast-twitch skeletal muscle sarcoplasmic

reticulum Ca2+-ATPase. Circulation Research, 83(9), 889-897.

Low, M. G., & Finean, J. B. (1976). The action of phosphatidylinositol-specific phospholipases C

on membranes. The Biochemical Journal, 154(1), 203-208.

Luo, J. H., & Weinstein, I. B. (1993). Calcium-dependent activation of protein kinase C. the role

of the C2 domain in divalent cation selectivity. The Journal of Biological Chemistry,

268(31), 23580-23584.

Lymn, R. W., & Taylor, E. W. (1971). Mechanism of adenosine triphosphate hydrolysis by

actomyosin. Biochemistry, 10(25), 4617-4624.

Lynch, K., Fernandez, G., Pappalardo, A., & Peluso, J. J. (2000). Basic fibroblast growth factor

inhibits apoptosis of spontaneously immortalized granulosa cells by regulating intracellular

free calcium levels through a protein kinase cdelta-dependent pathway. Endocrinology,

141(11), 4209-4217.

Lyons, M. K., Anderson, R. E., & Meyer, F. B. (1991). Basic fibroblast growth factor promotes

in vivo cerebral angiogenesis in chronic forebrain ischemia. Brain Research, 558(2), 315-

320.

Lytton, J., & MacLennan, D. H. (1988). Molecular cloning of cDNAs from human kidney coding

for two alternatively spliced products of the cardiac Ca2+-ATPase gene. The Journal of

Biological Chemistry, 263(29), 15024-15031.

Machado, N. G., Alves, M. G., Carvalho, R. A., & Oliveira, P. J. (2009). Mitochondrial

involvement in cardiac apoptosis during ischemia and reperfusion: Can we close the box?

233

Cardiovascular Toxicology, 9(4), 211-227. doi:10.1007/s12012-009-9055-1

MacLennan, D. H., & Kranias, E. G. (2003). Phospholamban: A crucial regulator of cardiac

contractility. Nature Reviews.Molecular Cell Biology, 4(7), 566-577. doi:10.1038/nrm1151

MacLennan, D. H., & Wong, P. T. (1971). Isolation of a calcium-sequestering protein from

sarcoplasmic reticulum. Proceedings of the National Academy of Sciences of the United

States of America, 68(6), 1231-1235.

Maffucci, T., Raimondi, C., Abu-Hayyeh, S., Dominguez, V., Sala, G., Zachary, I., & Falasca, M.

(2009). A phosphoinositide 3-kinase/phospholipase Cgamma1 pathway regulates fibroblast

growth factor-induced capillary tube formation. PloS One, 4(12), e8285.

doi:10.1371/journal.pone.0008285

Malo, M., Browaeys-Poly, E., Fournier, F., Cailliau, K., & Vilain, J. P. (1997). Ca2+ oscillations

induced by fibroblast growth factor 2 in xenopus oocytes expressing fibroblast growth factor

receptors. Molecular Membrane Biology, 14(4), 205-210.

Manning, J. R., Carpenter, G., Porter, D. R., House, S. L., Pietras, D. A., Doetschman, T., et al.

(2012). Fibroblast growth factor-2-induced cardioprotection against myocardial infarction

occurs via the interplay between nitric oxide, protein kinase signaling, and ATP-sensitive

potassium channels. Growth Factors (Chur, Switzerland),

Martiny-Baron, G., Kazanietz, M. G., Mischak, H., Blumberg, P. M., Kochs, G., Hug, H., . . .

Schachtele, C. (1993). Selective inhibition of protein kinase C isozymes by the

indolocarbazole go 6976. The Journal of Biological Chemistry, 268(13), 9194-9197.

Maslov, L. N., Lishmanov, Y. B., Oeltgen, P. R., Barzakh, E. I., Krylatov, A. V., Govindaswami,

M., & Brown, S. A. (2009). Activation of peripheral δ2 opioid receptors increases cardiac

tolerance to ischemia/reperfusion injury: Involvement of protein kinase C, NO-synthase,

KATP channels and the autonomic nervous system. Life Sciences, 84(19-20), 657-663.

234

doi:DOI: 10.1016/j.lfs.2009.02.016

Mattiazzi, A., Mundina-Weilenmann, C., Vittone, L., & Said, M. (2004). Phosphorylation of

phospholamban in ischemia-reperfusion injury: Functional role of Thr17 residue. Molecular

and Cellular Biochemistry, 263(1-2), 131-136.

Mattiazzi, A., Mundina-Weilenmann, C., Vittone, L., Said, M., & Kranias, E. G. (2006). The

importance of the Thr17 residue of phospholamban as a phosphorylation site under

physiological and pathological conditions. Brazilian Journal of Medical and Biological

Research = Revista Brasileira De Pesquisas Medicas e Biologicas / Sociedade Brasileira

De Biofisica ...[Et Al.], 39(5), 563-572. doi:/S0100-879X2006000500001

McKeehan, W. L., Wu, X., & Kan, M. (1999). Requirement for anticoagulant heparan sulfate in

the fibroblast growth factor receptor complex. The Journal of Biological Chemistry, 274(31),

21511-21514.

Medkova, M., & Cho, W. (1998). Mutagenesis of the C2 domain of protein kinase C-α. Journal

of Biological Chemistry, 273(28), 17544-17552. doi:10.1074/jbc.273.28.17544

Meij, J. T. A., Sheikh, F., Jimenez, S. K., Nickerson, P. W., Kardami, E., & Cattini, P. A. (2002).

Exacerbation of myocardial injury in transgenic mice overexpressing FGF-2 is T cell

dependent. American Journal of Physiology - Heart and Circulatory Physiology, 282(2),

H547-H555. doi:10.1152/ajpheart.01019.2000

Meldrum, D. R., Cleveland, J. C.,Jr, Mitchell, M. B., Rowland, R. T., Banerjee, A., & Harken, A.

H. (1996). Constructive priming of myocardium against ischemia-reperfusion injury. Shock

(Augusta, Ga.), 6(4), 238-242.

Melling, C. W., Thorp, D. B., Milne, K. J., & Noble, E. G. (2009). Myocardial Hsp70

phosphorylation and PKC-mediated cardioprotection following exercise. Cell Stress &

Chaperones, 14(2), 141-150. doi:10.1007/s12192-008-0065-x

235

Mendoza, M. C., Er, E. E., & Blenis, J. (2011). The ras-ERK and PI3K-mTOR pathways: Cross-

talk and compensation. Trends in Biochemical Sciences, 36(6), 320-328.

doi:10.1016/j.tibs.2011.03.006; 10.1016/j.tibs.2011.03.006

Merki, E., Zamora, M., Raya, A., Kawakami, Y., Wang, J., Zhang, X., . . . Ruiz-Lozano, P.

(2005). Epicardial retinoid X receptor α is required for myocardial growth and coronary

artery formation. Proceedings of the National Academy of Sciences of the United States of

America, 102(51), 18455-18460. doi:10.1073/pnas.0504343102

Merle, P. L., Usson, Y., Robert-Nicoud, M., & Verdetti, J. (1997). Basic FGF enhances calcium

permeable channel openings in adult rat cardiac myocytes: Implication in the bFGF-induced

increase of free Ca2+ content. Journal of Molecular and Cellular Cardiology, 29(10), 2687-

2698. doi:10.1006/jmcc.1997.0500

Metcalfe, E. E., Traaseth, N. J., & Veglia, G. (2005). Serine 16 phosphorylation induces an order-

to-disorder transition in monomeric phospholamban. Biochemistry, 44(11), 4386-4396.

doi:10.1021/bi047571e

Mignatti, P., Morimoto, T., & Rifkin, D. B. (1991). Basic fibroblast growth factor released by

single, isolated cells stimulates their migration in an autocrine manner. Proceedings of the

National Academy of Sciences of the United States of America, 88(24), 11007-11011.

Mignatti, P., Morimoto, T., & Rifkin, D. B. (1992). Basic fibroblast growth factor, a protein

devoid of secretory signal sequence, is released by cells via a pathway independent of the

endoplasmic reticulum-golgi complex. Journal of Cellular Physiology, 151(1), 81-93.

doi:10.1002/jcp.1041510113

Miki, T., Cohen, M. V., & Downey, J. M. (1998). Opioid receptor contributes to ischemic

preconditioning through protein kinase C activation in rabbits. Molecular and Cellular

Biochemistry, 186(1-2), 3-12.

236

Miller, W. P., McDonald, K. S., & Moss, R. L. (1996). Onset of reduced Ca2+ sensitivity of

tension during stunning in porcine myocardium. Journal of Molecular and Cellular

Cardiology, 28(4), 689-697. doi:10.1006/jmcc.1996.0064

Miller-Davis, S., McKeehan, W., & Carpenter, G. (1988). Prostatropin and acidic FGF also

support proliferation of an EGF-dependent keratinocyte cell line. Experimental Cell

Research, 179(2), 595-599.

Mitchell, M. B., Winter, C. B., Banerjee, A., & Harken, A. H. (1993). Inhibition of sarcoplasmic

reticulum calcium release reduces myocardial stunning. The Journal of Surgical Research,

54(5), 411-417.

Miura, T., Tanno, M., & Sato, T. (2010). Mitochondrial kinase signalling pathways in

myocardial protection from ischaemia/reperfusion-induced necrosis. Cardiovascular

Research, 88(1), 7-15.

Miyamoto, T., Leconte, I., Swain, J. L., & Fox, J. C. (1998). Autocrine FGF signaling is required

for vascular smooth muscle cell survival in vitro. Journal of Cellular Physiology, 177(1),

58-67. doi:2-D

Mohammadi, M., Dikic, I., Sorokin, A., Burgess, W. H., Jaye, M., & Schlessinger, J. (1996).

Identification of six novel autophosphorylation sites on fibroblast growth factor receptor 1

and elucidation of their importance in receptor activation and signal transduction. Molecular

and Cellular Biology, 16(3), 977-989.

Mohammadi, M., Schlessinger, J., & Hubbard, S. R. (1996). Structure of the FGF receptor

tyrosine kinase domain reveals a novel autoinhibitory mechanism. Cell, 86(4), 577-587.

doi:DOI: 10.1016/S0092-8674(00)80131-2

Molnar, A., Borbely, A., Czuriga, D., Ivetta, S. M., Szilagyi, S., Hertelendi, Z., . . . Toth, A.

(2009). Protein kinase C contributes to the maintenance of contractile force in human

237

ventricular cardiomyocytes. The Journal of Biological Chemistry, 284(2), 1031-1039.

doi:10.1074/jbc.M807600200

Mont, L., Cinca, J., Blanch, P., Blanco, J., Figueras, J., Brotons, C., & Soler-Soler, J. (1996).

Predisposing factors and prognostic value of sustained monomorphic ventricular tachycardia

in the early phase of acute myocardial infarction. Journal of the American College of

Cardiology, 28(7), 1670-1676. doi:10.1016/S0735-1097(96)00383-X

Montesano, R., Vassalli, J. D., Baird, A., Guillemin, R., & Orci, L. (1986). Basic fibroblast

growth factor induces angiogenesis in vitro. Proceedings of the National Academy of

Sciences of the United States of America, 83(19), 7297-7301.

Moore, M. J., Kanter, J. R., Jones, K. C., & Taylor, S. S. (2002). Phosphorylation of the catalytic

subunit of protein kinase A. Journal of Biological Chemistry, 277(49), 47878-47884.

doi:10.1074/jbc.M204970200

Morabito, C. J., Dettman, R. W., Kattan, J., Collier, J. M., & Bristow, J. (2001). Positive and

negative regulation of epicardial-mesenchymal transformation during avian heart

development. Developmental Biology, 234(1), 204-215. doi:10.1006/dbio.2001.0254

Moretti, A., Weig, H. J., Ott, T., Seyfarth, M., Holthoff, H. P., Grewe, D., . . . Laugwitz, K. L.

(2002). Essential myosin light chain as a target for caspase-3 in failing myocardium.

Proceedings of the National Academy of Sciences of the United States of America, 99(18),

11860-11865. doi:10.1073/pnas.182373099

Moscatelli, D., & Quarto, N. (1989). Transformation of NIH 3T3 cells with basic fibroblast

growth factor or the hst/K-fgf oncogene causes downregulation of the fibroblast growth

factor receptor: Reversal of morphological transformation and restoration of receptor

number by suramin. The Journal of Cell Biology, 109(5), 2519-2527.

Mudry, R. E., Perry, C. N., Richards, M., Fowler, V. M., & Gregorio, C. C. (2003). The

238

interaction of tropomodulin with tropomyosin stabilizes thin filaments in cardiac myocytes.

The Journal of Cell Biology, 162(6), 1057-1068. doi:10.1083/jcb.200305031

Munaron, L., & Fiorio Pla, A. (2000). Calcium influx induced by activation of tyrosine kinase

receptors in cultured bovine aortic endothelial cells. Journal of Cellular Physiology, 185(3),

454-463. doi:2-A

Mundiña-Weilenmann, C., Ferrero, P., Said, M., Vittone, L., Kranias, E. G., & Mattiazzi, A.

(2005). Role of phosphorylation of Thr17 residue of phospholamban in mechanical recovery

during hypercapnic acidosis. Cardiovascular Research, 66(1), 114-122.

doi:10.1016/j.cardiores.2004.12.028

Murphy, S., & Frishman, W. H. (2005). Protein kinase C in cardiac disease and as a potential

therapeutic target. Cardiology in Review, 13(1), 3-12.

doi:10.1097/01.crd.0000124914.59755.8d

Murry, C. E., Jennings, R. B., & Reimer, K. A. (1986). Preconditioning with ischemia: A delay of

lethal cell injury in ischemic myocardium. Circulation, 74(5), 1124-1136.

Naimy, H., Buczek-Thomas, J. A., Nugent, M. A., Leymarie, N., & Zaia, J. (2011). Highly

sulfated nonreducing end-derived heparan sulfate domains bind fibroblast growth factor-2

with high affinity and are enriched in biologically active fractions. Journal of Biological

Chemistry, 286(22), 19311-19319. doi:10.1074/jbc.M110.204693

Nakai, J., Imagawa, T., Hakamat, Y., Shigekawa, M., Takeshima, H., & Numa, S. (1990).

Primary structure and functional expression from cDNA of the cardiac ryanodine

receptor/calcium release channel. FEBS Letters, 271(1-2), 169-177.

Nakayama, F., Muller, K., Hagiwara, A., Ridi, R., Akashi, M., & Meineke, V. (2008).

Involvement of intracellular expression of FGF12 in radiation-induced apoptosis in mast

cells. Journal of Radiation Research, 49(5), 491-501.

239

Nayler, W. G. (1983). Calcium and cell death. European Heart Journal, 4 Suppl C, 33-41.

Noland, T. A.,Jr, Guo, X., Raynor, R. L., Jideama, N. M., Averyhart-Fullard, V., Solaro, R. J., &

Kuo, J. F. (1995). Cardiac troponin I mutants. phosphorylation by protein kinases C and A

and regulation of ca(2+)-stimulated MgATPase of reconstituted actomyosin S-1. The

Journal of Biological Chemistry, 270(43), 25445-25454.

Noland, T. A.,Jr, & Kuo, J. F. (1993). Protein kinase C phosphorylation of cardiac troponin I and

troponin T inhibits ca(2+)-stimulated MgATPase activity in reconstituted actomyosin and

isolated myofibrils, and decreases actin-myosin interactions. Journal of Molecular and

Cellular Cardiology, 25(1), 53-65. doi:10.1006/jmcc.1993.1007

Noland, T. A.,Jr, Raynor, R. L., Jideama, N. M., Guo, X., Kazanietz, M. G., Blumberg, P. M., . . .

Kuo, J. F. (1996). Differential regulation of cardiac actomyosin S-1 MgATPase by protein

kinase C isozyme-specific phosphorylation of specific sites in cardiac troponin I and its

phosphorylation site mutants. Biochemistry, 35(47), 14923-14931. doi:10.1021/bi9616357

Noland, T. A.,Jr, Raynor, R. L., & Kuo, J. F. (1989). Identification of sites phosphorylated in

bovine cardiac troponin I and troponin T by protein kinase C and comparative substrate

activity of synthetic peptides containing the phosphorylation sites. The Journal of Biological

Chemistry, 264(34), 20778-20785.

Oestreich, E. A., Malik, S., Goonasekera, S. A., Blaxall, B. C., Kelley, G. G., Dirksen, R. T., &

Smrcka, A. V. (2009). Epac and phospholipase cepsilon regulate Ca2+ release in the heart by

activation of protein kinase cepsilon and calcium-calmodulin kinase II. The Journal of

Biological Chemistry, 284(3), 1514-1522. doi:10.1074/jbc.M806994200

Oh, D. Y., Cho, J. H., Park, S. Y., Kim, Y. S., Yoon, Y. J., Yoon, S. H., . . . Han, J. S. (2008). A

novel role of hippocalcin in bFGF-induced neurite outgrowth of H19-7 cells. Journal of

Neuroscience Research, 86(7), 1557-1565. doi:10.1002/jnr.21602

240

Oh, L. Y., Goodyer, C. G., Olivier, A., & Yong, V. W. (1997). The promoting effects of bFGF and

astrocyte extracellular matrix on process outgrowth by adult human oligodendrocytes are

mediated by protein kinase C. Brain Research, 757(2), 236-244.

Okusa, C., Miyamae, M., Sugioka, S., Kaneda, K., Inamura, Y., Onishi, A., . . . Figueredo, V. M.

(2009). Acute memory phase of sevoflurane preconditioning is associated with sustained

translocation of protein kinase C-alpha and epsilon, but not delta, in isolated guinea pig

hearts. European Journal of Anaesthesiology, 26(7), 582-588.

doi:10.1097/EJA.0b013e32832a22c2

Opie, L. H. (1989). Reperfusion injury and its pharmacologic modification. Circulation, 80(4),

1049-1062.

Orr, J. W., & Newton, A. C. (1994). Requirement for negative charge on "activation loop" of

protein kinase C. The Journal of Biological Chemistry, 269(44), 27715-27718.

Padua, R. R., Merle, P. L., Doble, B. W., Yu, C. H., Zahradka, P., , G. N., . . . Kardami, E.

(1998). FGF-2-induced negative inotropism and cardioprotection are inhibited by

chelerythrine: Involvement of sarcolemmal calcium-independent protein kinase C. Journal

of Molecular and Cellular Cardiology, 30(12), 2695-2709. doi:10.1006/jmcc.1998.0832

Padua, R. R., Sethi, R., Dhalla, N. S., & Kardami, E. (1995). Basic fibroblast growth factor is

cardioprotective in ischemia-reperfusion injury. Molecular and Cellular Biochemistry,

143(2), 129-135.

Palaniyandi, S. S., Ferreira, J. C., Brum, P. C., & Mochly-Rosen, D. (2011). PKCbetaII

inhibition attenuates myocardial infarction induced heart failure and is associated with a

reduction of fibrosis and pro-inflammatory responses. Journal of Cellular and Molecular

Medicine, 15(8), 1769-1777.

Pan, B. S., & Solaro, R. J. (1987). Calcium-binding properties of troponin C in detergent-skinned

241

heart muscle fibers. The Journal of Biological Chemistry, 262(16), 7839-7849.

Pan, T. T., Neo, K. L., Hu, L. F., Yong, Q. C., & Bian, J. S. (2008). H2S preconditioning-induced

PKC activation regulates intracellular calcium handling in rat cardiomyocytes. American

Journal of Physiology.Cell Physiology, 294(1), C169-77. doi:10.1152/ajpcell.00282.2007

Parekh, D. B., Ziegler, W., & Parker, P. J. (2000). Multiple pathways control protein kinase C

phosphorylation. The EMBO Journal, 19(4), 496-503. doi:10.1093/emboj/19.4.496

Park, B. J., Lee, D. G., Yu, J. R., Jung, S. K., Choi, K., Lee, J., Ahnn, J. (2001). Calreticulin, a

calcium-binding molecular , is required for stress response and fertility in

caenorhabditis elegans. Molecular Biology of the Cell, 12(9), 2835-2845.

Partovian, C., Zhuang, Z., Moodie, K., Lin, M., Ouchi, N., Sessa, W. C., Simons, M. (2005).

PKCalpha activates eNOS and increases arterial blood flow in vivo. Circulation Research,

97(5), 482-487. doi:10.1161/01.RES.0000179775.04114.45

Patton, C., Thompson, S., & Epel, D. (2004). Some precautions in using chelators to buffer

metals in biological solutions. Cell Calcium, 35(5), 427-431. doi:10.1016/j.ceca.2003.10.006

Pellieux, C., Foletti, A., Peduto, G., Aubert, J. F., Nussberger, J., Beermann, F., , T.

(2001). Dilated cardiomyopathy and impaired cardiac hypertrophic response to angiotensin

II in mice lacking FGF-2. The Journal of Clinical Investigation, 108(12), 1843-1851.

doi:10.1172/JCI13627

Pedrozo Z, Sánchez G, Torrealba N, Valenzuela R, Fernández C, Hidalgo C,

Lavandero S, Donoso P. and proteasomes mediate degradation of ryanodine

receptors in a model of cardiac ischemic reperfusion. (2010) Biochim Biophys Acta.

Mar;1802(3):356-62.

Peluso, J. J. (2003). Basic fibroblast growth factor (bFGF) regulation of the plasma membrane

calcium ATPase (PMCA) as part of an anti-apoptotic mechanism of action. Biochemical

242

Pharmacology, 66(8), 1363-1369.

Peluso, J. J., Pappalardo, A., & Fernandez, G. (2001). Basic fibroblast growth factor maintains

calcium homeostasis and granulosa cell viability by stimulating calcium efflux via a PKC

delta-dependent pathway. Endocrinology, 142(10), 4203-4211.

Periasamy, M., Bhupathy, P., & Babu, G. J. (2008). Regulation of sarcoplasmic reticulum Ca2+

ATPase pump expression and its relevance to cardiac muscle physiology and pathology.

Cardiovascular Research, 77(2), 265-273. doi:10.1093/cvr/cvm056

Periasamy, M., & Huke, S. (2001). SERCA pump level is a critical determinant of

ca(2+)homeostasis and cardiac contractility. Journal of Molecular and Cellular Cardiology,

33(6), 1053-1063. doi:10.1006/jmcc.2001.1366

Petrovski, G., Das, S., Juhasz, B., Kertesz, A., Tosaki, A., & Das, D. K. (2011). Cardioprotection

by endoplasmic reticulum stress-induced autophagy. Antioxidants & Redox Signaling,

14(11), 2191-2200. doi:10.1089/ars.2010.3486

Phatharajaree, W., Phrommintikul, A., & Chattipakorn, N. (2007). Matrix metalloproteinases and

myocardial infarction. The Canadian Journal of Cardiology, 23(9), 727-733.

Pike, M. M., Kitakaze, M., & Marban, E. (1990). 23Na-NMR measurements of intracellular

sodium in intact perfused ferret hearts during ischemia and reperfusion. American Journal of

Physiology - Heart and Circulatory Physiology, 259(6), H1767-H1773.

Ping, P., Zhang, J., Qiu, Y., Tang, X. L., Manchikalapudi, S., Cao, X., & Bolli, R. (1997).

Ischemic preconditioning induces selective translocation of protein kinase C isoforms

epsilon and eta in the heart of conscious rabbits without subcellular redistribution of total

protein kinase C activity. Circulation Research, 81(3), 404-414.

Ping, P., Takano, H., Zhang, J., Tang, X., Qiu, Y., Li, R. C. X., Bolli, R. (1999). Isoform-selective

activation of protein kinase C by nitric oxide in the heart of conscious rabbits : A signaling

243

mechanism for both nitric Oxide–Induced and ischemia-induced preconditioning.

Circulation Research, 84(5), 587-604.

Piper, H. M. (1989). Energy deficiency, calcium overload or oxidative stress: Possible causes of

irreversible ischemic myocardial injury. Klinische Wochenschrift, 67(9), 465-476.

Plotnikov, A. N., Hubbard, S. R., Schlessinger, J., & Mohammadi, M. (2000). Crystal structures

of two FGF-FGFR complexes reveal the determinants of ligand-receptor specificity. Cell,

101(4), 413-424.

Plotnikov, A. N., Schlessinger, J., Hubbard, S. R., & Mohammadi, M. (1999). Structural basis for

FGF receptor dimerization and activation. Cell, 98(5), 641-650.

Polewicz, D., Cadete, V. J. J., Doroszko, A., Hunter, B. E., Sawicka, J., Szczesna-Cordary, D., et

al. (2011). Ischemia induced peroxynitrite dependent modifications of cardiomyocyte MLC1

increases its degradation by MMP-2 leading to contractile dysfunction. Journal of Cellular

and Molecular Medicine, 15(5), 1136-1147.

Polnaszek, N., Kwabi-Addo, B., , L. E., Ozen, M., Greenberg, N. M., Ortega, S.,

Ittmann, M. (2003). Fibroblast growth factor 2 promotes tumor progression in an

autochthonous mouse model of prostate cancer. Cancer Research, 63(18), 5754-5760.

Porrello, E. R., Johnson, B. A., Aurora, A. B., Simpson, E., Nam, Y., Matkovich, S. J., Olson, E.

N. (2011). miR-15 family regulates postnatal mitotic arrest of cardiomyocytes / novelty and

significance. Circulation Research, 109(6), 670-679.

doi:10.1161/CIRCRESAHA.111.248880

Pravdic, D., Sedlic, F., Mio, Y., Vladic, N., Bienengraeber, M., & Bosnjak, Z. J. (2009).

Anesthetic-induced preconditioning delays opening of mitochondrial permeability transition

pore via protein kinase C-epsilon-mediated pathway. Anesthesiology, 111(2), 267-274.

Presta, M., Tiberio, L., Rusnati, M., Dell'Era, P., & Ragnotti, G. (1991). Basic fibroblast growth

244

factor requires a long-lasting activation of protein kinase C to induce cell proliferation in

transformed fetal bovine aortic endothelial cells. Cell Regulation, 2(9), 719-726.

Puceat, M., & Vassort, G. (1996). Signalling by protein kinase C isoforms in the heart. Molecular

and Cellular Biochemistry, 157(1-2), 65-72.

Pyle, W. G., Chen, Y., & Hofmann, P. A. (2003). Cardioprotection through a PKC-dependent

decrease in myofilament ATPase. American Journal of Physiology.Heart and Circulatory

Physiology, 285(3), H1220-8. doi:10.1152/ajpheart.00076.2003

Pyle, W. G., Lester, J. W., & Hofmann, P. A. (2001). Effects of kappa-opioid receptor activation

on myocardium. American Journal of Physiology.Heart and Circulatory Physiology, 281(2),

H669-78.

Pyle, W. G., Smith, T. D., & Hofmann, P. A. (2000). Cardioprotection with kappa-opioid receptor

stimulation is associated with a slowing of cross-bridge cycling. American Journal of

Physiology.Heart and Circulatory Physiology, 279(4), H1941-8.

Qiu, Y., Ping, P., Tang, X. L., Manchikalapudi, S., Rizvi, A., Zhang, J., Bolli, R. (1998). Direct

evidence that protein kinase C plays an essential role in the development of late

preconditioning against myocardial stunning in conscious rabbits and that epsilon is the

isoform involved. The Journal of Clinical Investigation, 101(10), 2182-2198.

doi:10.1172/JCI1258

Qiu, Y., Tang, X. L., Park, S. W., Sun, J. Z., Kalya, A., & Bolli, R. (1997). The early and late

phases of ischemic preconditioning: A comparative analysis of their effects on infarct size,

myocardial stunning, and arrhythmias in conscious pigs undergoing a 40-minute coronary

occlusion. Circulation Research, 80(5), 730-742.

Qu, B., & Zhang, J. S. (2004). The pathway of bFGF induced changes of calcium in cultured

human lens epithelial cells. [Zhonghua Yan Ke Za Zhi] Chinese Journal of Ophthalmology,

245

40(12), 832-835.

Quarto, N., Fong, K. D., & Longaker, M. T. (2005). Gene profiling of cells expressing different

FGF-2 forms. Gene, 356, 49-68. doi:10.1016/j.gene.2005.05.014

Quinlan, C. L., Costa, A. D., Costa, C. L., Pierre, S. V., Dos Santos, P., & Garlid, K. D. (2008).

Conditioning the heart induces formation of signalosomes that interact with mitochondria to

open mitoKATP channels. American Journal of Physiology.Heart and Circulatory

Physiology, 295(3), H953-H961

Rahmoune, H., Chen, H. L., Gallagher, J. T., Rudland, P. S., & Fernig, D. G. (1998). Interaction

of heparan sulfate from mammary cells with acidic fibroblast growth factor (FGF) and basic

FGF. regulation of the activity of basic FGF by high and low affinity binding sites in

heparan sulfate. The Journal of Biological Chemistry, 273(13), 7303-7310.

Rameau, G. A., Tukey, D. S., Garcin-Hosfield, E. D., Titcombe, R. F., Misra, C., Khatri, L., Ziff,

E. B. (2007). Biphasic coupling of neuronal nitric oxide synthase phosphorylation to the

NMDA receptor regulates AMPA receptor trafficking and neuronal cell death. The Journal

of Neuroscience : The Official Journal of the Society for Neuroscience, 27(13), 3445-3455.

doi:10.1523/JNEUROSCI.4799-06.2007

Rameau, G. A., Chiu, L., & Ziff, E. B. (2004). Bidirectional regulation of neuronal nitric-oxide

synthase phosphorylation at serine 847 by the N-methyl-d-aspartate receptor. Journal of

Biological Chemistry, 279(14), 14307-14314. doi:10.1074/jbc.M311103200

Ramirez-Correa, G. A., Cortassa, S., Stanley, B., Gao, W. D., & Murphy, A. M. (2010). Calcium

sensitivity, force frequency relationship and cardiac troponin I: Critical role of PKA and

PKC phosphorylation sites. Journal of Molecular and Cellular Cardiology, 48(5), 943-953.

doi:10.1016/j.yjmcc.2010.01.004

Rane, S., He, M., Sayed, D., Vashistha, H., Malhotra, A., Sadoshima, J., Abdellatif, M. (2009).

246

Downregulation of MiR-199a derepresses hypoxia-inducible factor-1α and sirtuin 1 and

recapitulates hypoxia preconditioning in cardiac myocytes. Circulation Research, 104(7),

879-886. doi:10.1161/CIRCRESAHA.108.193102

Rayment, I., Rypniewski, W. R., Schmidt-Base, K., Smith, R., Tomchick, D. R., Benning, M. M.,

. . . Holden, H. M. (1993). Three-dimensional structure of myosin subfragment-1: A

molecular motor. Science (New York, N.Y.), 261(5117), 50-58.

Reidy, M. A. (1993). Neointimal proliferation: The role of basic FGF on vascular smooth muscle

cell proliferation. Thrombosis and Haemostasis, 70(1), 172-176.

Reidy, M. A., & Lindner, V. (1991). Basic FGF and growth of arterial cells. Annals of the New

York Academy of Sciences, 638, 290-299.

Ren, X. P., Wu, J., Wang, X., Sartor, M. A., Qian, J., Jones, K., Fan, G. C. (2009). MicroRNA-

320 is involved in the regulation of cardiac ischemia/reperfusion injury by targeting heat-

shock protein 20. Circulation, 119(17), 2357-2366.

doi:10.1161/CIRCULATIONAHA.108.814145

Renko, M., Quarto, N., Morimoto, T., & Rifkin, D. B. (1990). Nuclear and cytoplasmic

localization of different basic fibroblast growth factor species. Journal of Cellular

Physiology, 144(1), 108-114. doi:10.1002/jcp.1041440114

Robles-Flores, M., Rendón-Huerta, E., González-Aguilar, H., Mendoza-Hernández, G., Islas, S.,

Mendoza, V., . . . López-Casillas, F. (2002). p32 (gC1qBP) is a general protein kinase C

(PKC)-binding protein. Journal of Biological Chemistry, 277(7), 5247-5255.

doi:10.1074/jbc.M109333200

Rodriguez, P., Bhogal, M. S., & Colyer, J. (2003). Stoichiometric phosphorylation of cardiac

ryanodine receptor on serine 2809 by calmodulin-dependent kinase II and protein kinase A.

Journal of Biological Chemistry, 278(40), 38593-38600. doi:10.1074/jbc.C301180200

247

Ron, D., Chen, C. H., Caldwell, J., Jamieson, L., Orr, E., & Mochly-Rosen, D. (1994). Cloning

of an intracellular receptor for protein kinase C: A homolog of the beta subunit of G

proteins. Proceedings of the National Academy of Sciences of the United States of America,

91(3), 839-843.

Rotenberg, S. A., Zhu, J., Hansen, H., Li, X. D., Sun, X. G., Michels, C. A., & Riedel, H. (1998).

Deletion analysis of protein kinase calpha reveals a novel regulatory segment. Journal of

Biochemistry, 124(4), 756-763.

Ryan, P. J., Paterno, G. D., & Gillespie, L. L. (1998). Identification of phosphorylated proteins

associated with the fibroblast growth factor receptor type I during early xenopus

development. Biochemical and Biophysical Research Communications, 244(3), 763-767.

doi:10.1006/bbrc.1998.8326

Rybin, V. O., Sabri, A., Short, J., Braz, J. C., Molkentin, J. D., & Steinberg, S. F. (2003). Cross-

regulation of novel protein kinase C (PKC) isoform function in cardiomyocytes. Journal of

Biological Chemistry, 278(16), 14555-14564. doi:10.1074/jbc.M212644200

Said, M., Vittone, L., Mundina-Weilenmann, C., Ferrero, P., Kranias, E. G., & Mattiazzi, A.

(2003). Role of dual-site phospholamban phosphorylation in the stunned heart: Insights from

phospholamban site-specific mutants. American Journal of Physiology.Heart and

Circulatory Physiology, 285(3), H1198-205. doi:10.1152/ajpheart.00209.2003

Sakamoto, J., Miura, T., Goto, M., & Iimura, O. (1995). Limitation of myocardial infarct size by

adenosine A1 receptor activation is abolished by protein kinase C inhibitors in the rabbit.

Cardiovascular Research, 29(5), 682-688.

Samain, E., Bouillier, H., Miserey, S., Perret, C., Renaud, J. F., Safar, M., & Dagher, G. (2000).

Extracellular signal-regulated kinase pathway is involved in basic fibroblast growth factor

effect on angiotensin II-induced ca(2+) transient in vascular smooth muscle cell from wistar-

248

kyoto and spontaneously hypertensive rats. Hypertension, 35(1 Pt 1), 61-67.

Saraiva, R. M., Minhas, K. M., Raju, S. V. Y., Barouch, L. A., Pitz, E., Schuleri, K. H., Hare, J.

M. (2005). Deficiency of neuronal nitric oxide synthase increases mortality and cardiac

remodeling after myocardial infarction. Circulation, 112(22), 3415-3422.

doi:10.1161/CIRCULATIONAHA.105.557892

Sasisekharan, R., Ernst, S., & Venkataraman, G. (1997). On the regulation of fibroblast growth

factor activity by heparin-like glycosaminoglycans. Angiogenesis, 1(1), 45-54.

Sato, Y., Abe, M., & Takaki, R. (1990). Platelet factor 4 blocks the binding of basic fibroblast

growth factor to the receptor and inhibits the spontaneous migration of vascular endothelial

cells. Biochemical and Biophysical Research Communications, 172(2), 595-600.

doi:10.1016/0006-291X(90)90715-Y

Saurin, A. T., Pennington, D. J., Raat, N. J. H., Latchman, D. S., Owen, M. J., & Marber, M. S.

(2002). Targeted disruption of the protein kinase C epsilon gene abolishes the infarct size

reduction that follows ischaemic preconditioning of isolated buffer-perfused mouse hearts.

Cardiovascular Research, 55(3), 672-680. doi:10.1016/S0008-6363(02)00325-5

Sawicki, G., Leon, H., Sawicka, J., Sariahmetoglu, M., Schulze, C. J., Scott, P. G., Schulz, R.

(2005). Degradation of myosin light chain in isolated rat hearts subjected to ischemia-

reperfusion injury. Circulation, 112(4), 544-552.

doi:10.1161/CIRCULATIONAHA.104.531616

Schaub, M. C., Hartshorne, D. J., & Perry, S. V. (1967). Effect of tropomyosin on the calcium-

activated adenosine triphosphatase of actomyosin. Nature, 215(5101), 635-636.

Schechtman, D., & Mochly-Rosen, D. (2001). Adaptor proteins in protein kinase C-mediated

signal transduction. Oncogene, 20(44), 6339-6347. doi:10.1038/sj.onc.1204778

Schechtman, D., Craske, M. L., Kheifets, V., Meyer, T., Schechtman, J., & Mochly-Rosen, D.

249

(2004). A critical intramolecular interaction for protein kinase Cϵ translocation. Journal of

Biological Chemistry, 279(16), 15831-15840. doi:10.1074/jbc.M310696200

Schlessinger, J., Plotnikov, A. N., Ibrahimi, O. A., Eliseenkova, A. V., Yeh, B. K., Yayon, A., . . .

Mohammadi, M. (2000). Crystal structure of a ternary FGF-FGFR-heparin complex reveals

a dual role for heparin in FGFR binding and dimerization. Molecular Cell, 6(3), 743-750.

Schmidt, A. G., Kadambi, V. J., Ball, N., Sato, Y., Walsh, R. A., Kranias, E. G., & Hoit, B. D.

(2000). Cardiac-specific overexpression of calsequestrin results in left ventricular

hypertrophy, depressed force-frequency relation and pulsus alternans in vivo. Journal of

Molecular and Cellular Cardiology, 32(9), 1735-1744. doi:10.1006/jmcc.2000.1209

Schneider, M. D., McLellan, W. R., Black, F. M., & Parker, T. G. (1992). Growth factors, growth

factor response elements, and the cardiac phenotype. Basic Research in Cardiology, 87

Suppl 2, 33-48.

Schuller, A. C., Ahmed, Z., Levitt, J. A., Suen, K. M., Suhling, K., & Ladbury, J. E. (2008).

Indirect recruitment of the signalling adaptor shc to the fibroblast growth factor receptor 2

(FGFR2). The Biochemical Journal, 416(2), 189-199. doi:10.1042/BJ20080887

Schultz, J. E., Rose, E., Yao, Z., & Gross, G. J. (1995). Evidence for involvement of opioid

receptors in ischemic preconditioning in rat hearts. American Journal of Physiology - Heart

and Circulatory Physiology, 268(5), H2157-H2161.

Schultz, J. E., Witt, S. A., Nieman, M. L., Reiser, P. J., Engle, S. J., Zhou, M., Doetschman, T.

(1999). Fibroblast growth factor-2 mediates pressure-induced hypertrophic response. The

Journal of Clinical Investigation, 104(6), 709-719. doi:10.1172/JCI7315

Schulz, R., Gres, P., Skyschally, A., Duschin, A., Belosjorow, S., Konietzka, I., & Heusch, G.

(2003). Ischemic preconditioning preserves connexin 43 phosphorylation during sustained

ischemia in pig hearts in vivo. The FASEB Journal, doi:10.1096/fj.02-0975fje

250

Schwinger, R. H., Brixius, K., Savvidou-Zaroti, P., Bolck, B., Zobel, C., Frank, K., Erdmann, E.

(2000). The enhanced contractility in phospholamban deficient mouse hearts is not

associated with alterations in (Ca2+)-sensitivity or myosin ATPase-activity of the contractile

proteins. Basic Research in Cardiology, 95(1), 12-18.

Sears, C. E., Bryant, S. M., Ashley, E. A., Lygate, C. A., Rakovic, S., Wallis, H. L., Casadei, B.

(2003). Cardiac neuronal nitric oxide synthase isoform regulates myocardial contraction and

calcium handling. Circulation Research, 92(5), e52-e59.

doi:10.1161/01.RES.0000064585.95749.6D

Seehase, M., Quentin, T., Wiludda, E., Hellige, G., Paul, T., & Schiffmann, H. (2006). Gene

expression of the na-Ca2+ exchanger, SERCA2a and calsequestrin after myocardial

ischemia in the neonatal rabbit heart. Biology of the Neonate, 90(3), 174-184.

doi:10.1159/000092888

Sellke, F. W., Wang, S. Y., Friedman, M., Harada, K., Edelman, E. R., Grossman, W., & Simons,

M. (1994). Basic FGF enhances endothelium-dependent relaxation of the collateral-perfused

coronary microcirculation. American Journal of Physiology - Heart and Circulatory

Physiology, 267(4), H1303-H1311.

Sharp, B. R., Jones, S. P., Rimmer, D. M., & Lefer, D. J. (2002). Differential response to

myocardial reperfusion injury in eNOS-deficient mice. American Journal of Physiology -

Heart and Circulatory Physiology, 282(6), H2422-H2426. doi:10.1152/ajpheart.00855.2001

Shaw, E. E., Wood, P., Kulpa, J., Yang, F. H., Summerlee, A. J., & Pyle, W. G. (2009). Relaxin

alters cardiac myofilament function through a PKC-dependent pathway. American Journal

of Physiology.Heart and Circulatory Physiology, 297(1), H29-36.

doi:10.1152/ajpheart.00482.2008

Sheikh, F., Sontag, D. P., Fandrich, R. R., Kardami, E., & Cattini, P. A. (2001). Overexpression

251

of FGF-2 increases cardiac myocyte viability after injury in isolated mouse hearts. American

Journal of Physiology.Heart and Circulatory Physiology, 280(3), H1039-50.

Shen, A. C., & Jennings, R. B. (1972). Myocardial calcium and magnesium in acute ischemic

injury. The American Journal of Pathology, 67(3), 417-440.

Shine, K. I., & Douglas, A. M. (1983). Low calcium reperfusion of ischemic myocardium.

Journal of Molecular and Cellular Cardiology, 15(4), 251-260. doi:10.1016/0022-

2828(83)90280-8

Shintani-Ishida, K., & Yoshida, K. (2011). Ischemia induces phospholamban dephosphorylation

via activation of calcineurin, PKC-alpha, and protein phosphatase 1, thereby inducing

calcium overload in reperfusion. Biochimica Et Biophysica Acta, 1812(7), 743-751.

doi:10.1016/j.bbadis.2011.03.014

Simmerman, H. K., Collins, J. H., Theibert, J. L., Wegener, A. D., & Jones, L. R. (1986).

Sequence analysis of phospholamban. identification of phosphorylation sites and two major

structural domains. The Journal of Biological Chemistry, 261(28), 13333-13341.

Singh, K., Balligand, J., Fischer, T. A., Smith, T. W., & Kelly, R. A. (1996). Regulation of

cytokine-inducible nitric oxide synthase in cardiac myocytes and microvascular endothelial

cells. Journal of Biological Chemistry, 271(2), 1111-1117.

Sinha, J., Chen, F., Miloh, T., Burns, R. C., Yu, Z., & Shneider, B. L. (2008). Beta-klotho and

FGF-15/19 inhibit the apical sodium-dependent bile acid transporter in enterocytes and

cholangiocytes. American Journal of Physiology.Gastrointestinal and Liver Physiology,

295(5), G996-G1003. doi:10.1152/ajpgi.90343.2008

Skaletz-Rorowski, A., Waltenberger, J., Muller, J. G., Pawlus, E., Pinkernell, K., & Breithardt, G.

(1999). Protein kinase C mediates basic fibroblast growth factor-induced proliferation

through mitogen-activated protein kinase in coronary smooth muscle cells. Arteriosclerosis,

252

Thrombosis, and Vascular Biology, 19(7), 1608-1614.

Skaper, S. D., Kee, W. J., Facci, L., Macdonald, G., Doherty, P., & Walsh, F. S. (2000). The

FGFR1 inhibitor PD 173074 selectively and potently antagonizes FGF-2 neurotrophic and

neurotropic effects. Journal of Neurochemistry, 75(4), 1520-1527. doi:10.1046/j.1471-

4159.2000.0751520.x

Smyth, G. K. (2004). Linear models and empirical bayes methods for assessing differential

expression in microarray experiments. Statistical Applications in Genetics and Molecular

Biology, 3, Article3. doi:10.2202/1544-6115.1027

Solaro, R. J. (1975). Calcium regulation of cardiac myofibrillar activation: Effects of MgATP.

Journal of Supramolecular Structure, 3(4), 368-375. doi:10.1002/jss.400030409

Song, Q. (2004). Effects of genetic manipulation of phospholamban protein levels on contractile

function and remodeling in murine cardiac and slow-twitch skeletal muscles. Dissertation,

University of Cincinnati.

Soulet, L., Chevet, E., Lemaitre, G., Blanquaert, F., Meddahi, A., & Barritault, D. (1994). FGFs

and their receptors, in vitro and in vivo studies: New FGF receptor in the brain, FGF-1 in

muscle, and the use of functional analogues of low-affinity heparin-binding growth factor

receptors in tissue repair. Molecular Reproduction and Development, 39(1), 49-54;

discussion 54-5. doi:10.1002/mrd.1080390109

Speir, E., Tanner, V., Gonzalez, A. M., Farris, J., Baird, A., & Casscells, W. (1992). Acidic and

basic fibroblast growth factors in adult rat heart myocytes. localization, regulation in culture,

and effects on DNA synthesis. Circulation Research, 71(2), 251-259.

Spirito, P., Fu, Y., Yu, Z., Epstein, S., & Casscells, W. (1991). Immunohistochemical localization

of basic and acidic fibroblast growth factors in the developing rat heart. Circulation, 84(1),

322-332. doi:10.1161/01.CIR.84.1.322

253

Spruill, L. S., Baicu, C. F., Zile, M. R., & McDermott, P. J. (2008). Selective translation of

mRNAs in the left ventricular myocardium of the mouse in response to acute pressure

overload. Journal of Molecular and Cellular Cardiology, 44(1), 69-75.

doi:10.1016/j.yjmcc.2007.10.011

Steenbergen, C., Fralix, T. A., & Murphy, E. (1993). Role of increased cytosolic free calcium

concentration in myocardial ischemic injury. Basic Research in Cardiology, 88(5), 456-470.

Steinberg, S. F. (2008). Structural basis of protein kinase C isoform function. Physiological

Reviews, 88(4), 1341-1378. doi:10.1152/physrev.00034.2007

Stensman, H., & Larsson, C. (2007). Identification of acidic amino acid residues in the protein

kinase C alpha V5 domain that contribute to its insensitivity to diacylglycerol. The Journal

of Biological Chemistry, 282(39), 28627-28638. doi:10.1074/jbc.M702248200

Strunz, C. M., Matsuda, M., Salemi, V. M., Nogueira, A., Mansur, A. P., Cestari, I. N., &

Marquezini, M. V. (2011). Changes in cardiac heparan sulfate proteoglycan expression and

streptozotocin-induced diastolic dysfunction in rats. Cardiovascular Diabetology, 10, 35.

doi:10.1186/1475-2840-10-35

Su, G., Meyer, K., Nandini, C. D., Qiao, D., Salamat, S., & Friedl, A. (2006). Glypican-1 is

frequently overexpressed in human gliomas and enhances FGF-2 signaling in glioma cells.

The American Journal of Pathology, 168(6), 2014-2026. doi:10.2353/ajpath.2006.050800

Subramaniam, A., Jones, W. K., Gulick, J., Wert, S., Neumann, J., & Robbins, J. (1991). Tissue-

specific regulation of the alpha-myosin heavy chain gene promoter in transgenic mice.

Journal of Biological Chemistry, 266(36), 24613-24620.

Sullivan, C. J., Doetschman, T., & Hoying, J. B. (2002). Targeted disruption of the Fgf2 gene

does not affect vascular growth in the mouse ischemic hindlimb. Journal of Applied

Physiology, 93(6), 2009-2017. doi:10.1152/japplphysiol.00451.2002

254

Sumandea, M. P., Burkart, E. M., Kobayashi, T., De Tombe, P. P., & Solaro, R. J. (2004).

Molecular and integrated biology of thin filament protein phosphorylation in heart muscle.

Annals of the New York Academy of Sciences, 1015, 39-52. doi:10.1196/annals.1302.004

Sumandea, M. P., Pyle, W. G., Kobayashi, T., de Tombe, P. P., & Solaro, R. J. (2003).

Identification of a functionally critical protein kinase C phosphorylation residue of cardiac

troponin T. The Journal of Biological Chemistry, 278(37), 35135-35144.

doi:10.1074/jbc.M306325200

Sun, H. Y., Xue, F. S., Xu, Y. C., Li, C. W., Xiong, J., Liao, X., & Zhang, Y. M. (2009). Propofol

improves cardiac functional recovery after ischemia-reperfusion by upregulating nitric oxide

synthase activity in the isolated rat hearts. Chinese Medical Journal, 122(24), 3048-3054.

Sweeney, H. L., Bowman, B. F., & Stull, J. T. (1993). Myosin light chain phosphorylation in

vertebrate striated muscle: Regulation and function. The American Journal of Physiology,

264(5 Pt 1), C1085-95.

Szegezdi, E., Duffy, A., O’Mahoney, M. E., Logue, S. E., Mylotte, L. A., O’Brien, T., & Samali,

A. (2006). ER stress contributes to ischemia-induced cardiomyocyte apoptosis. Biochemical

and Biophysical Research Communications, 349(4), 1406-1411. doi:DOI:

10.1016/j.bbrc.2006.09.009

Tada, M., Yamada, M., Ohmori, F., Kuzuya, T., Inui, M., & Abe, H. (1980). Transient state

kinetic studies of Ca2+-dependent ATPase and calcium transport by cardiac sarcoplasmic

reticulum. effect of cyclic AMP-dependent protein kinase-catalyzed phosphorylation of

phospholamban. The Journal of Biological Chemistry, 255(5), 1985-1992.

Takahashi, Y., Takemura, S., Minamiyama, Y., Shibata, T., Hirai, H., Sasaki, Y., Suehiro, S.

(2007). Landiolol has cardioprotective effects against reperfusion injury in the rat heart via

the PKCepsilon signaling pathway. Free Radical Research, 41(7), 757-769.

255

doi:10.1080/10715760701338810

Takai, Y., Yamamoto, M., Inoue, M., Kishimoto, A., & Nishizuka, Y. (1977). A proenzyme of

cyclic nucleotide-independent protein kinase and its activation by calcium-dependent neutral

protease from rat liver. Biochemical and Biophysical Research Communications, 77(2), 542-

550. doi:DOI: 10.1016/S0006-291X(77)80013-2

Takimoto, E., Soergel, D. G., Janssen, P. M., Stull, L. B., Kass, D. A., & Murphy, A. M. (2004).

Frequency- and afterload-dependent cardiac modulation in vivo by troponin I with

constitutively active protein kinase A phosphorylation sites. Circulation Research, 94(4),

496-504. doi:10.1161/01.RES.0000117307.57798.F5

Tani, M., & Neely, J. R. (1989). Role of intracellular na+ in Ca2+ overload and depressed

recovery of ventricular function of reperfused ischemic rat hearts. possible involvement of

H+-na+ and na+-Ca2+ exchange. Circulation Research, 65(4), 1045-1056.

Taylor, W. R., Greenberg, A. H., Turley, E. A., & Wright, J. A. (1993). Cell motility, invasion,

and malignancy induced by overexpression of K-FGF or bFGF. Experimental Cell Research,

204(2), 295-301. doi:10.1006/excr.1993.1036

Temsah, R. M., Kawabata, K., Chapman, D., & Dhalla, N. S. (2002). Preconditioning prevents

alterations in cardiac SR gene expression due to ischemia-reperfusion. American Journal of

Physiology.Heart and Circulatory Physiology, 282(4), H1461-6.

doi:10.1152/ajpheart.00447.2001

Thandroyen, F. T., McCarthy, J., Burton, K. P., & Opie, L. H. (1988). Ryanodine and caffeine

prevent ventricular arrhythmias during acute myocardial ischemia and reperfusion in rat

heart. Circulation Research, 62(2), 306-314.

Thingholm, T. E., Jensen, O. N., & Larsen, M. R. (2009). Analytical strategies for

phosphoproteomics. Proteomics, 9(6), 1451-1468. doi:10.1002/pmic.200800454

256

Thornton, J., Striplin, S., Liu, G. S., Swafford, A., Stanley, A. W., Van Winkle, D. M., & Downey,

J. M. (1990). Inhibition of protein synthesis does not block myocardial protection afforded

by preconditioning. The American Journal of Physiology, 259(6 Pt 2), H1822-5.

Thuerauf, D. J., Marcinko, M., Gude, N., Rubio, M., Sussman, M. A., & Glembotski, C. C.

(2006). Activation of the unfolded protein response in infarcted mouse heart and hypoxic

cultured cardiac myocytes. Circulation Research, 99(3), 275-282.

doi:10.1161/01.RES.0000233317.70421.03

Tiefenbacher, C. P., & Chilian, W. M. (1997). Basic fibroblast growth factor and heparin

influence coronary arteriolar tone by causing endothelium-dependent dilation.

Cardiovascular Research, 34(2), 411-417. doi:10.1016/S0008-6363(97)00029-1

Timson, D. J. (2003). Fine tuning the myosin motor: The role of the essential light chain in

striated muscle myosin. Biochimie, 85(7), 639-645.

Toullec, D., Pianetti, P., Coste, H., Bellevergue, P., Grand-Perret, T., Ajakane, M., Loriolle, F.

(1991). The bisindolylmaleimide GF 109203X is a potent and selective inhibitor of protein

kinase C. The Journal of Biological Chemistry, 266(24), 15771-15781.

Tsuji T, Ohga Y, Yoshikawa Y, Sakata S, Abe T, Tabayashi N, Kobayashi S,

Kohzuki H, Yoshida KI, Suga H, Kitamura S, Taniguchi S, Takaki M. Rat cardiac

contractile dysfunction induced by Ca2+ overload: possible link to the

proteolysis of alpha-fodrin. (2001) Am J Physiol Heart Circ Physiol.

281(3):H1286-94.

Turner, C. A., Clinton, S. M., Thompson, R. C., Watson, S. J., & Akil, H. (2011). Fibroblast

growth factor-2 (FGF2) augmentation early in life alters hippocampal development and

rescues the anxiety phenotype in vulnerable animals. Proceedings of the National Academy

of Sciences, doi:10.1073/pnas.1103732108

257

Turrell, H. E., Rodrigo, G. C., Norman, R. I., Dickens, M., & Standen, N. B. (2011).

Phenylephrine preconditioning involves modulation of cardiac sarcolemmal K(ATP) current

by PKC delta, AMPK and p38 MAPK. Journal of Molecular and Cellular Cardiology,

51(3), 370-380. doi:10.1016/j.yjmcc.2011.06.015

Uecker, M., Da Silva, R., Grampp, T., Pasch, T., Schaub, M. C., & Zaugg, M. (2003).

Translocation of protein kinase C isoforms to subcellular targets in ischemic and anesthetic

preconditioning. Anesthesiology, 99(1), 138-147.

Ueno, H. (1978). Binding of troponin components to tropomyosin fragments. Journal of

Biochemistry, 84(4), 1009-1012.

Unger, E. F., Banai, S., Shou, M., Lazarous, D. F., Jaklitsch, M. T., Scheinowitz, M., Epstein, S.

E. (1994). Basic fibroblast growth factor enhances myocardial collateral flow in a canine

model. The American Journal of Physiology, 266(4 Pt 2), H1588-95.

Urakawa, I., Yamazaki, Y., Shimada, T., Iijima, K., Hasegawa, H., Okawa, K., Yamashita, T.

(2006). Klotho converts canonical FGF receptor into a specific receptor for FGF23. Nature,

444(7120), 770-774. doi:10.1038/nature05315

Vainikka, S., Partanen, J., Bellosta, P., Coulier, F., Birnbaum, D., Basilico, C., Alitalo, K. (1992).

Fibroblast growth factor receptor-4 shows novel features in genomic structure, ligand

binding and signal transduction. The EMBO Journal, 11(12), 4273-4280.

Valverde, C. A., Mundina-Weilenmann, C., Reyes, M., Kranias, E. G., Escobar, A. L., &

Mattiazzi, A. (2006). Phospholamban phosphorylation sites enhance the recovery of

intracellular Ca2+ after perfusion arrest in isolated, perfused mouse heart. Cardiovascular

Research, 70(2), 335-345. doi:10.1016/j.cardiores.2006.01.018 van Rooij, E., Sutherland, L. B., Liu, N., Williams, A. H., McAnally, J., Gerard, R. D., Olson, E.

N. (2006). A signature pattern of stress-responsive microRNAs that can evoke cardiac

258

hypertrophy and heart failure. Proceedings of the National Academy of Sciences, 103(48),

18255-18260. doi:10.1073/pnas.0608791103

Vasko, R., Koziolek, M., Ikehata, M., Rastaldi, M. P., Jung, K., Schmid, H., Strutz, F. (2009).

Role of basic fibroblast growth factor (FGF-2) in diabetic nephropathy and mechanisms of

its induction by hyperglycemia in human renal fibroblasts. American Journal of

Physiology.Renal Physiology, 296(6), F1452-63. doi:10.1152/ajprenal.90352.2008

Vasu, V. T., Ott, S., Hobson, B., Rashidi, V., Oommen, S., Cross, C. E., & Gohil, K. (2009).

Sarcolipin and ubiquitin carboxy-terminal hydrolase 1 mRNAs are over-expressed in

skeletal muscles of alpha-tocopherol deficient mice. Free Radical Research, 43(2), 106-116.

doi:10.1080/10715760802616676

Verboomen, H., Wuytack, F., De Smedt, H., Himpens, B., & Casteels, R. (1992). Functional

difference between SERCA2a and SERCA2b Ca2+ pumps and their modulation by

phospholamban. The Biochemical Journal, 286 ( Pt 2)(Pt 2), 591-595.

Vidavalur, R., Swarnakar, S., Thirunavukkarasu, M., Samuel, S. M., & Maulik, N. (2008). Ex

vivo and in vivo approaches to study mechanisms of cardioprotection targeting

ischemia/reperfusion (i/r) injury: Useful techniques for cardiovascular drug discovery.

Current Drug Discovery Technologies, 5(4), 269-278.

Vlodavsky, I., Miao, H. Q., Medalion, B., Danagher, P., & Ron, D. (1996). Involvement of

heparan sulfate and related molecules in sequestration and growth promoting activity of

fibroblast growth factor. Cancer Metastasis Reviews, 15(2), 177-186.

Walker, L. A., Walker, J. S., Ambler, S. K., & Buttrick, P. M. (2010). Stage-specific changes in

myofilament protein phosphorylation following myocardial infarction in mice. Journal of

Molecular and Cellular Cardiology, 48(6), 1180-1186. doi:10.1016/j.yjmcc.2009.09.010

259

Wang, J., Liu, X., Sentex, E., Takeda, N., & Dhalla, N. S. (2003). Increased expression of protein

kinase C isoforms in heart failure due to myocardial infarction. American Journal of

Physiology - Heart and Circulatory Physiology, 284(6), H2277-H2287.

doi:10.1152/ajpheart.00142.2002

Wang, Y., & Ashraf, M. (1999). Role of protein kinase C in mitochondrial KATP channel-

mediated protection against Ca2+ overload injury in rat myocardium. Circulation Research,

84(10), 1156-1165.

Weckbecker, G., Liu, R., & Tolcsvai, L. (1992). Intradermal angiogenesis in nude mice induced

by human tumor cells or b-FGF. EXS, 61, 296-301.

Wehrens, X. H. T., Lehnart, S. E., Reiken, S. R., & Marks, A. R. (2004). Ca2+/Calmodulin-

dependent protein kinase II phosphorylation regulates the cardiac ryanodine receptor.

Circulation Research, 94(6), e61-e70. doi:10.1161/01.RES.0000125626.33738.E2

Weihrauch, D., Tessmer, J., Warltier, D. C., & Chilian, W. M. (1998). Repetitive coronary artery

occlusions induce release of growth factors into the myocardial interstitium. American

Journal of Physiology - Heart and Circulatory Physiology, 275(3), H969-H976.

West, M. B., Rokosh, G., Obal, D., Velayutham, M., Xuan, Y. T., Hill, B. G., Bhatnagar, A.

(2008). Cardiac myocyte-specific expression of inducible nitric oxide synthase protects

against ischemia/reperfusion injury by preventing mitochondrial permeability transition.

Circulation, 118(19), 1970-1978. doi:10.1161/CIRCULATIONAHA.108.791533

Westermann, R., Grothe, C., & Unsicker, K. (1990). Basic fibroblast growth factor (bFGF), a

multifunctional growth factor for neuroectodermal cells. Journal of Cell

Science.Supplement, 13, 97-117.

Whelan, R. S., Kaplinskiy, V., & Kitsis, R. N. (2010). Cell death in the pathogenesis of heart

disease: Mechanisms and significance. Annual Review of Physiology, 72, 19-44.

260

doi:10.1146/annurev.physiol.010908.163111

Wickley, P. J., Ding, X., Murray, P. A., & Damron, D. S. (2006). Propofol-induced activation of

protein kinase C isoforms in adult rat ventricular myocytes. Anesthesiology, 104(5), 970-

977.

Wiecha, J., Reineker, K., Reitmayer, M., Voisard, R., Hannekum, A., Mattfeldt, T., Hombach, V.

(1998). Modulation of Ca2+-activated K+ channels in human vascular cells by insulin and

basic fibroblast growth factor. Growth & IGF Research : Official Journal of the

Growth Hormone Research Society and the International IGF Research Society, 8(2), 175-

181.

Wildhirt, S. M., Dudek, R. R., Suzuki, H., & Bing, R. J. (1995). Involvement of inducible nitric

oxide synthase in the inflammatory process of myocardial infarction. International Journal

of Cardiology, 50(3), 253-261. doi:DOI: 10.1016/0167-5273(95)02385-A

Williams, M. R., Arthur, J. S. C., Balendran, A., van der Kaay, J., Poli, V., Cohen, P., & Alessi, D.

R. (2000). The role of 3-phosphoinositide-dependent protein kinase 1 in activating AGC

kinases defined in embryonic stem cells. Current Biology, 10(8), 439-448. doi:DOI:

10.1016/S0960-9822(00)00441-3

Williams, N. G., Paradis, H., Agarwal, S., Charest, D. L., Pelech, S. L., & Roberts, T. M. (1993).

Raf-1 and p21v-ras cooperate in the activation of mitogen-activated protein kinase.

Proceedings of the National Academy of Sciences of the United States of America, 90(12),

5772-5776.

Williamson, J. R. (1986). Role of inositol lipid breakdown in the generation of intracellular

signals. state of the art lecture. Hypertension, 8(6 Pt 2), II140-56.

Wohrle, S., Bonny, O., Beluch, N., Gaulis, S., Stamm, C., Scheibler, M., Graus-Porta, D. (2011).

FGF receptors control vitamin D and phosphate homeostasis by mediating renal FGF-23

261

signaling and regulating FGF-23 expression in bone. Journal of Bone and Mineral Research

: The Official Journal of the American Society for Bone and Mineral Research, 26(10),

2486-2497. doi:10.1002/jbmr.478; 10.1002/jbmr.478

World Health Organization. (2008). World health organization. the global burden of disease:

2008 update. geneva, world health organization. available at

http://www.who.int/evidence/bod.

Wu, S. C., & Solaro, R. J. (2007). Protein kinase C zeta. A novel regulator of both

phosphorylation and de-phosphorylation of cardiac sarcomeric proteins. The Journal of

Biological Chemistry, 282(42), 30691-30698.

Xiao, B., Zhong, G., Obayashi, M., Yang, D., Chen, K., Walsh, M. P., Chen, S. R. (2006). Ser-

2030, but not ser-2808, is the major phosphorylation site in cardiac ryanodine receptors

responding to protein kinase A activation upon beta-adrenergic stimulation in normal and

failing hearts. The Biochemical Journal, 396(1), 7-16. doi:10.1042/BJ20060116

Xie, Y., Zhu, Y., Zhu, W. Z., Chen, L., Zhou, Z. N., Yuan, W. J., & Yang, H. T. (2005). Role of

dual-site phospholamban phosphorylation in intermittent hypoxia-induced cardioprotection

against ischemia-reperfusion injury. American Journal of Physiology.Heart and Circulatory

Physiology, 288(6), H2594-602. doi:10.1152/ajpheart.00926.2004

Xu, H., Lee, K. W., & Goldfarb, M. (1998). Novel recognition motif on fibroblast growth factor

receptor mediates direct association and activation of SNT adapter proteins. Journal of

Biological Chemistry, 273(29), 17987-17990. doi:10.1074/jbc.273.29.17987

Xu, K. Y., Huso, D. L., Dawson, T. M., Bredt, D. S., & Becker, L. C. (1999). Nitric oxide

synthase in cardiac sarcoplasmic reticulum. Proceedings of the National Academy of

Sciences of the United States of America, 96(2), 657-662.

Xu, R. X., Pawelczyk, T., Xia, T. H., & Brown, S. C. (1997). NMR structure of a protein kinase

262

C-gamma phorbol-binding domain and study of protein-lipid micelle interactions.

Biochemistry, 36(35), 10709-10717. doi:10.1021/bi970833a

Xu, X., Rao, G., Quiros, R. M., Kim, A. W., Miao, H. Q., Brunn, G. J., . . . Prinz, R. A. (2007). In

vivo and in vitro degradation of heparan sulfate (HS) proteoglycans by HPR1 in pancreatic

adenocarcinomas. loss of cell surface HS suppresses fibroblast growth factor 2-mediated cell

signaling and proliferation. The Journal of Biological Chemistry, 282(4), 2363-2373.

doi:10.1074/jbc.M604218200

Xuan, Y., Guo, Y., Zhu, Y., Wang, O., Rokosh, G., & Bolli, R. (2007). Endothelial nitric oxide

synthase plays an obligatory role in the late phase of ischemic preconditioning by activating

the protein kinase Cε–p44/42 mitogen-activated protein Kinase–pSer-signal transducers and

activators of Transcription1/3 pathway. Circulation, 116(5), 535-544.

doi:10.1161/CIRCULATIONAHA.107.689471

Yagami, T., Takase, K., Yamamoto, Y., Ueda, K., Takasu, N., Okamura, N., . . . Fujimoto, M.

(2010). Fibroblast growth factor 2 induces apoptosis in the early primary culture of rat

cortical neurons. Experimental Cell Research, 316(14), 2278-2290.

doi:10.1016/j.yexcr.2010.03.023

Yamamura, S., Nelson, P. R., & Kent, K. C. (1996). Role of protein kinase C in attachment,

spreading, and migration of human endothelial cells. The Journal of Surgical Research,

63(1), 349-354. doi:10.1006/jsre.1996.0274

Yamashita, N., Hoshida, S., Taniguchi, N., Kuzuya, T., & Hori, M. (1998). A "second window of

protection" occurs 24 h after ischemic preconditioning in the rat heart. Journal of Molecular

and Cellular Cardiology, 30(6), 1181-1189. doi:10.1006/jmcc.1998.0682

Yan, D., & Lin, X. (2007). Drosophila glypican dally-like acts in FGF-receiving cells to

modulate FGF signaling during tracheal morphogenesis. Developmental Biology, 312(1),

263

203-216. doi:10.1016/j.ydbio.2007.09.015

Yang, Q. E., Johnson, S. E., & Ealy, A. D. (2011). Protein kinase C delta mediates fibroblast

growth factor-2-induced interferon-tau expression in bovine trophoblast. Biology of

Reproduction, 84(5), 933-943. doi:10.1095/biolreprod.110.087916

Yayon, A., & Klagsbrun, M. (1990). Autocrine regulation of cell growth and transformation by

basic fibroblast growth factor. Cancer Metastasis Reviews, 9(3), 191-202.

Yeh, C., Chen, T., Wang, Y., Lin, Y., & Fang, S. (2010). Cardiomyocytic apoptosis limited by

bradykinin via restoration of nitric oxide after cardioplegic arrest. Journal of Surgical

Research, 163(1), e1-e9. doi:DOI: 10.1016/j.jss.2010.04.005

Yoshida K, Inui M, Harada K, Saido TC, Sorimachi Y, Ishihara T, Kawashima S, Sobue K.

Reperfusion of rat heart after brief ischemia induces proteolysis of calspectin (nonerythroid

spectrin or fodrin) by calpain. (1995) Circulation Research 77(3):603-10.

Yoshida, K., Hirata, T., Akita, Y., Mizukami, Y., Yamaguchi, K., Sorimachi, Y., Kawashiama, S.

(1996). Translocation of protein kinase C-alpha, delta and epsilon isoforms in ischemic rat

heart. Biochimica Et Biophysica Acta, 1317(1), 36-44.

Yoshida, K., Kawamura, S., Mizukami, Y., & Kitakaze, M. (1997). Implication of protein kinase

C-alpha, delta, and epsilon isoforms in ischemic preconditioning in perfused rat hearts.

Journal of Biochemistry, 122(3), 506-511.

Yoshida K, Harada K. Proteolysis of erythrocyte-type and brain-type ankyrins

in rat heart after postischemic reperfusion. (1997) Journal of Biochemistry 122(2):279-85.

Yu, Z., Wang, Z. H., & Yang, H. T. (2009). Calcium/calmodulin-dependent protein kinase II

mediates cardioprotection of intermittent hypoxia against ischemic-reperfusion-induced

cardiac dysfunction. American Journal of Physiology.Heart and Circulatory Physiology,

297(2), H735-42. doi:10.1152/ajpheart.01164.2008

264

Yu, Z., Wang, Z., & Yang, H. (August 2009). Calcium/calmodulin-dependent protein kinase II

mediates cardioprotection of intermittent hypoxia against ischemic-reperfusion-induced

cardiac dysfunction. American Journal of Physiology - Heart and Circulatory Physiology,

297(2), H735-H742. doi:10.1152/ajpheart.01164.2008

Zhan, X., Hu, X., Friesel, R., & Maciag, T. (1993). Long term growth factor exposure and

differential tyrosine phosphorylation are required for DNA synthesis in BALB/c 3T3 cells.

The Journal of Biological Chemistry, 268(13), 9611-9620.

Zhang, H. Y., McPherson, B. C., Liu, H., Baman, T., McPherson, S. S., Rock, P., & Yao, Z.

(2002). Role of nitric-oxide synthase, free radicals, and protein kinase C δ in opioid-induced

cardioprotection. Journal of Pharmacology and Experimental Therapeutics, 301(3), 1012-

1019. doi:10.1124/jpet.301.3.1012

Zhang, J., Wang, L., Petrin, J., Bishop, W. R., & Bond, R. W. (1993). Characterization of site-

specific mutants altered at protein kinase C beta 1 isozyme autophosphorylation sites.

Proceedings of the National Academy of Sciences of the United States of America, 90(13),

6130-6134.

Zhang, Y. H., Zhang, M. H., Sears, C. E., Emanuel, K., Redwood, C., El-Armouche, A., . . .

Casadei, B. (2008). Reduced phospholamban phosphorylation is associated with impaired

relaxation in left ventricular myocytes from neuronal NO Synthase–Deficient mice.

Circulation Research, 102(2), 242-249. doi:10.1161/CIRCRESAHA.107.164798

Zhao, T. C., Taher, M. M., Valerie, K. C., & Kukreja, R. C. (2001). p38 triggers late

preconditioning elicited by anisomycin in heart. Circulation Research, 89(10), 915-922.

doi:10.1161/hh2201.099452

Zhao, W., Frank, K. F., Chu, G., Gerst, M. J., Schmidt, A. G., Ji, Y., . . . Kranias, E. G. (2003).

Combined phospholamban ablation and SERCA1a overexpression result in a new

265

hyperdynamic cardiac state. Cardiovascular Research, 57(1), 71-81.

Zhao, W., Uehara, Y., Chu, G., Song, Q., Qian, J., Young, K., & Kranias, E. G. (2004). Threonine-

17 phosphorylation of phospholamban: A key determinant of frequency-dependent increase

of cardiac contractility. Journal of Molecular and Cellular Cardiology, 37(2), 607-612.

doi:10.1016/j.yjmcc.2004.05.013

Zhou, M. (1997). Role of fibroblast growth factor 2 in cardiovascular development and function.

(Doctor of Philosophy, University of Cincinnati).

Zhou, H. Z., Karliner, J. S., & Gray, M. O. (2002). Moderate alcohol consumption induces

sustained cardiac protection by activating PKC-epsilon and akt. American Journal of

Physiology.Heart and Circulatory Physiology, 283(1), H165-74.

doi:10.1152/ajpheart.00408.2001

Zhou, M., Sutliff, R. L., Paul, R. J., Lorenz, J. N., Hoying, J. B., Haudenschild, C. C., . . .

Doetschman, T. (1998). Fibroblast growth factor 2 control of vascular tone. Nature

Medicine, 4(2), 201-207.

Zhu, J., Hansen, H., Su, L., Shieh, H. L., & Riedel, H. (1994). Ligand regulation of bovine

protein kinase C alpha response via either cysteine-rich repeat of conserved region C1.

Journal of Biochemistry, 115(5), 1000-1009.

Zingarelli, B., Hake, P. W., Yang Zequan, O'Connor, M., Denenberg, A., & Wong, H. R. (2002).

Absence of inducible nitric oxide synthase modulates early reperfusion-induced NF-κB and

AP-1 activation and enhances myocardial damage. The FASEB Journal, 16(3), 327-342.

doi:10.1096/fj.01-0533com

Zingman, L. V., Hodgson, D. M., Bast, P. H., Kane, G. C., Perez-Terzic, C., Gumina, R. J., et al.

(2002). Kir6.2 is required for adaptation to stress. Proceedings of the National Academy of

Sciences of the United States of America, 99(20), 13278-13283.

266

Zot, A. S., & Potter, J. D. (1987). Structural aspects of troponin-tropomyosin regulation of

skeletal muscle contraction. Annual Review of Biophysics and Biophysical Chemistry, 16,

535-559. doi:10.1146/annurev.bb.16.060187.002535

Zuber, M. E., Zhou, Z., Burrus, L. W., & Olwin, B. B. (1997). Cysteine-rich FGF receptor

regulates intracellular FGF-1 and FGF-2 levels. Journal of Cellular Physiology, 170(3), 217-

227. doi:10.1002/(SICI)1097-4652(199703)170:3<217::AID-JCP1>3.0.CO;2-R

Zucchi, R., Ronca-Testoni, S., Yu, G., Galbani, P., Ronca, G., & Mariani, M. (1995).

Postischemic changes in cardiac sarcoplasmic reticulum Ca2+ channels: A possible

mechanism of ischemic preconditioning. Circulation Research, 76(6), 1049-1056.

267