The Role of Troponin C in the Heart DISSERTATION

Home , Desmin, Myofilament, Troponin C

The Role of Troponin C in the Heart

DISSERTATION

Presented in Partial Fulfillment of the Requirements for the Degree Doctor of Philosophy

in the Graduate School of The Ohio State University

Sean Carl Little

Graduate Program in Integrated Biomedical Science Program

The Ohio State University

2012

Dissertation Committee:

Dr. Jonathan P. Davis, PhD (Advisor),

Dr. Philip F. Binkley, MD

Dr. Paul M.L. Janssen, PhD

Dr. Mark T. Ziolo, PhD

Sean Carl Little

2012

Abstract

Heart disease is a broad category composed of coronary artery disease, heart failure, arrhythmias and many other miscellaneous heart problems. A common characteristic for many of these diseases is the impairment of regulatory mechanisms believed to control the rate and extent of relaxation in the heart. An important component of myocardial relaxation is the inactivation of the myofilament, for which the rate- limiting steps are not completely understood. Depending on how fast the myofilament inactivates and how long it remains inactivated can influence the relaxation of the heart.

Thus, the faster the myofilament is able to inactivate the more time the heart muscle is able relax to allow the heart to fill with blood. If the rate of myofilament inactivation is decreased, the heart relaxes slower and the heart is not able to relax to the same extent as before, resulting in inadequate filling of blood.

The work described within focused on studying the molecular mechanisms of myofilament inactivation to contribute to the overall knowledge of heart relaxation. The rate-limiting step of cardiac muscle relaxation has been proposed to reside in the myofilament. Both the rates of cross-bridge detachment and Ca2+ dissociation from

Troponin C (TnC) have been hypothesized to rate-limit myofilament inactivation. To study these inactivation events we used a fluorescent TnC to measure both the rate of

Ca2+ dissociation from TnC and the rate of cross-bridge detachment from several different species of ventricular myofibrils. The fluorescently labeled TnC was sensitive ii to both Ca2+ dissociation and cross-bridge detachment, which allowed for a direct

comparison between the two proposed rates of myofilament inactivation. Unlike Ca2+

dissociation from TnC, cross-bridge detachment varied in myofibrils from different

species and was rate-limited by ADP release. At sub-physiological temperatures (<

20oC), the rate of Ca2+ dissociation from TnC was faster than the rate of cross-bridge detachment in the presence of ADP. These results support the hypothesis that cross- bridge detachment rate-limits relaxation. However, Ca2+ dissociation from TnC was not

as temperature sensitive as cross-bridge detachment. At a near physiological temperature

(35oC) and [ADP], the rate of cross-bridge detachment may actually be faster than the

rate of Ca2+ dissociation. This provides evidence that there may not be a simple, single

rate-limiting step of myofilament inactivation and TnC may play a vital role in rate-

limiting relaxation. One way to further study the role of TnC in muscle physiology is to

alter the function of the protein to determine its affect on Ca2+ binding.

Cardiac TnCs that had altered Ca2+ binding properties in the Tn and reconstituted

thin filament systems, were incorporated into ventricular myofibrils. These specifically designed TnCs had increased or decreased Ca2+ sensitivities, as well as modified Ca2+ dissociation rates when measured from the ventricular myofibrils. In addition, Ca2+ sensitized fluorescent TnCs were able to report the rate of Ca2+ dissociation from the

neighboring, unlabeled, endogenous TnC. The modified TnC exposed the cross-talk that

occurs between neighboring Tn’s in the myofilament that had different Ca2+ sensitivities.

In addition, a Ca2+ desensitized TnC increased the rate of cross-bridge detachment in the

presence of ADP, providing evidence that TnC may be able to influence the rate of thick

iii

filament inactivation in the dynamic myofilament. Thus, the TnCs with altered Ca2+

binding provided evidence that the rate of Ca2+ dissociation can be slowed and brought to

light interactions that may occur in the dynamic and highly interactive myofilament.

The specifically designed TnCs allowed us to test whether a direct increase or

decrease in the Ca2+ sensitivity of the myofilament leads to a diseased phenotype. Upon

incorporation of engineered TnCs into an animal model, the Ca2+ desensitized TnC,

D73N, was able to recapitulate the diseased cardiac phenotype of a dilated

cardiomyopathy as measured by echocardiography. This was accomplished by intra-

peritoneal injection of adeno-associated virus serotype 9 (AAV-9) containing the D73N

TnC into neonatal mice 1-2 days after birth. On the other hand, AAV-9 containing the

Ca2+ sensitized, L48Q TnC, was unable to recapitulate the restrictive or hypertrophic

cardiomyopathy phenotypes commonly associated with an increased myofilament Ca2+

sensitivity. AAV-9 containing GFP was used to verify that the AAV-9 was able to target the heart, as well as skeletal muscle, due to the use of a cyctomegalovirus promoter. This was a proof of principle study designed to test the feasibility of using AAV-9 to transduce the cardiomyocytes with modified TnCs. The work described within examined the effects that the modified TnCs had on the in vivo contractile properties of the heart.

In addition, the results showed that an alteration in the Ca2+ sensitivity of the myofilament

does not always lead to a diseased heart. In fact these engineered TnCs may be used as a

treatment strategy against various cardiac diseases.

Dedication

DEDICATED TO THE MEMORIES OF JANE MARKOVICH AND CAROL MUIR

Acknowledgments

I would like to thank all of those that have helped me throughout the years with

everything from collaborations, insightful discussions, keeping me mentally focused

during stressing times, and the many opportunities and interactions that I have had. To

begin I would like to thank my advisor, friend, and colleague, Dr. Jonathan Davis. Jon has taught me how to critically think and to question everything in regards to experimental design and scientific thinking. In addition to scientific discussions, there were also many heated games of racquetball played throughout the years in which Jon gave me lessons in how not to control ones temper. I would like to thank my dissertation advisory committee: Drs. Philip Binkly, Paul Janssen, and Mark Ziolo for the invaluable insight into experimental design, collaboration, interpretation of the data, and guidance during my journey. I would like to thank Dr. Brandon Biesiadecki for his help in teaching me the ins and outs of gel chromatography in addition to the numerous spirited conversations. I would like to thank fellow graduate student Steve Roof for his tireless effort in the collection of the isolated cardiomyocyte data. I would like to thank Dr

Jianchao Zhang, also known as Remington, for the production of the different viral constructs that were vital to the completion of my project. Lastly, I would like to thank my family who has always supported me in whatever endeavor it was that I chose and I am truly blessed to have them in my life. My loving wife, Jenna, has always supported me and always finds ways to pick me up when I stumbled. Without everyone’s help I would not be the scientist, and more importantly, the man I have become.

Vita

May 29th, 1984 ...... Born, Trumbull Hospital, Warren, Ohio

2002-2006 ...... B.A. Chemistry, Hiram College

2007-2012 ...... Graduate Research Associate, Department

of Physiology and Cell Biology, The Ohio State University

2009-2011 ...... Graduate Research Fellow, Department of

Physiology and Cell Biology, The Ohio State University

Publications

1. Little SC, Biesiadecki BJ, Kilic A, Higgins RS, Janssen PM, Davis JP, The rates

of Ca2+ dissociation and cross-bridge detachment from ventricular myofibrils as

reported by a fluorescent cardiac troponin C. J Biol Chem 2012 Jun 20. [Epub

ahead of print]

2. Nixon BR, Thawornkaiwong A, Jin J, Brundage EA, Little SC, Davis JP, Solaro

RJ, Biesiadecki BJ, AMP activated protein kinase phosphorylates cardiac

troponin I at Ser-150 to increase myofilament calcium sensitivity and blunt PKA

dependent function. J Biol Chem 2012 Apr 6. [Epub ahead of print]

vii

3. Little SC, Tikunova SB, Norman C, Swartz DR, Davis JP, Measurement of

calcium dissociation rates from troponin C in rigor skeletal myofibrils. Front

Physiology 2011 Oct 11; 2:70

4. Tikunova SB, Liu B, Swindle N, Little SC, Gomes AV, Swartz DR, Davis JP,

Effect of calcium-sensitizing mutations on calcium binding and exchange with

troponin C in increasingly complex biochemical systems. Biochemistry 2010 Mar

9; 49(9):1975-84

Fields of Study

Major Field: Integrated Biomedical Science Program

1. Emphasis: Cardiac Muscle Physiology

2. Emphasis: Translational Research

viii

Table of Contents

The Role of Troponin C in the Heart ...... 1

DISSERTATION ...... 1

Abstract ...... ii

Dedication ...... v

Acknowledgments...... vi

Publications ...... vii

Fields of Study ...... viii

Table of Contents ...... ix

List of Tables ...... xvi

List of Figures ...... xvii

Chapter 1: Introduction ...... 1

1.1 General Introduction ...... 1

1.2 Basic Cardiac Cycle ...... 3

1.3 Contraction of the Heart (Systole) ...... 4

1.4 The Contractile Apparatus ...... 6

1.4.1 Thin Filament Regulation ...... 7

1.4.2 Thick Filament Regulation ...... 9

1.5 Relaxation of the Heart (Diastole) ...... 12

1.5.1 Myofilament Inactivation...... 13

1.5.2 Proposed Biochemical Rate-limiting Steps of Myofilament Relaxation ...... 13

1.6 Physiological Modulation of Contraction and Relaxation ...... 14

1.7 Cardiomyopathies Resulting in Altered Ca2+ Binding Properties ...... 16

1.8 Measuring the Rates of Ca2+Dissociation from Troponin C ...... 17

1.9 Utilization of Troponin C’s with Altered Ca2+ Binding Properties ...... 18

1.10 Overall Goal and Significance ...... 19

Chapter 2: The Rates of Ca2+ Dissociation and Cross-Bridge Detachment from

Ventricular Myofibrils ...... 21

2.1 Introduction ...... 21

2.1 Methods ...... 24

2.1.1 Materials ...... 24

2.1.2 Protein Purification ...... 25

2.1.3 Troponin C labeling with IAANS or IANBD ...... 25

2.1.4 Troponin Complex Formation ...... 26

2.1.5 Myofibril Preparation...... 27

2.1.6 Determining the Myofibril Concentration ...... 28

2.1.7 Exchange of Labeled Tn into the Myofibrils ...... 29

2.1.8 Cross-linking of Myofibrils ...... 30

2.1.9 Stopped-Flow Kinetic Measurements from Myofibrils ...... 31

2.1.10 Formation of the Reconstituted Thin Filaments ...... 31

2.1.11 Determination of the Myosin S1 Concentration and Detachment Rate from

Reconstituted Thin Filaments ...... 32

2.1.12 Statistical Analysis ...... 33

2.2 Results ...... 33

2.2.1 Measuring Ca2+ Dissociation ...... 33

2.2.2 Effect of Myofibril Species of the Rigor Ca2+ Dissociation Rate ...... 37

2.2.3 Effect of ATP on the Observed Fluorescence Changes of TnC ...... 38

2.2.4 Sensing Cross-bridge Detachment Through TnC ...... 43

2.2.5 Species Dependence on the Cross-Bridge Detachment Rate ...... 50

2.2.6 Recovery of the Ca2+ Dissociation Signal with Slowed Cross-Bridge Detachment 52

2.2.7 Physiological Significance of Ca2+ Dissociation and Cross-Bridge Detachment .... 56

2.3 Discussion ...... 59

2.3.1 Summary of Findings ...... 59

2.3.2 Proposed Mechanism for How TnC is Able to Report Cross-Bridge Detachment . 60

2.3.3 Comparison Between Myofibrils and Previous Systems Used to Measure Cross-

Bridge Detachment ...... 62

2.3.4 Physiological Significance of the Myofibril Biochemical System ...... 64

2.4 Limitations and Pitfalls ...... 65

Chapter 3: Characterization of the Myofibril Ca2+ Binding Properties using Modified

TnC’s...... 67

3.1 General Introduction ...... 67

3.2 Methods ...... 70

3.2.1 Determining the Steady-State Ca2+ Binding Affinities ...... 70

3.2.4 Statistical Analysis ...... 71

3.3 Results ...... 72

3.3.1 Steady-State Ca2+Binding Affinities ...... 72

3.3.2 Ca2+ Dissociation Kinetics of Modified TnCs ...... 74

3.3.3 Investigating the Origin of the Biphasic L48Q Ca2+ Dissociation Curve ...... 76

3.3.4 Stopped-Flow Titration of L48Q Myofibrils ...... 81

3.3.5 Effect of Cycling Cross-Bridges on the Ca2+ Dissociation Rate of the Modified

TnCs ...... 83

3.3.6 Cross-Bridge Detachment Reported by Modified TnCs ...... 85

3.3.7 Temperature Dependence on the Ca2+ Dissociation Rate of Modified TnCs ...... 90

xii

3.4 Discussion ...... 93

3.4.1 Summary of Findings ...... 93

3.4.2 Effect of TnC Modifications on Ca2+ Dissociation...... 94

3.4.3 Mechanism of Cross-talk Between Neighboring Tn ...... 95

3.4.4 Physiological Significance ...... 97

3.4 Limitations of the Myofibrils ...... 99

3.5 Future Directions ...... 102

Chapter 4: Modulating heart function through the use of TnC’s with modified Ca2+ binding properties ...... 105

4.1.1 Approaches to Studying TnC in the Heart...... 105

4.1.2 AAV as a Transfer Vehicle for Gene Therapy ...... 106

4.2 Methods ...... 108

4.2.1 Design of the TnC Viral Vector Construct ...... 109

4.2.2 Induction of Hypothermia and Injection of Neonatal Pups ...... 112

4.2.3 Echocardiography ...... 114

4.2.4 Cardiomyocyte Isolation and Measurement of Myocyte Function ...... 115

4.2.5 Preparing AAV-9 Injected Tissue for Western Blot Analysis ...... 117

4.2.6 SDS Gel and Western Blot Analysis ...... 118

4.2.7 Tissue Fixation and Sectioning for GFP Expression ...... 119

xiii

4.2.8 Statistical Analysis ...... 120

4.3 Results ...... 120

4.3.1 Effect of FLAG-tag on TnC’s Ca2+ Dissociation Kinetics ...... 120

4.3.2 AAV-9 GFP Targets both Cardiac and Skeletal Muscle ...... 124

4.3.3 In Vivo Incorporation of FLAG-tag TnC into the Myofilament ...... 128

4.3.4 Death Rates of AAV-9 Injected Mice ...... 131

4.3.5 In Vivo Heart Function of AAV-9 TnC Injected Mice ...... 133

4.3.6 EKG Abnormalities of AAV-9 D73N TnC Injected Mice ...... 140

4.3.7 Determination of the Contractile Properties of Isolated Cardiomyocytes ...... 141

4.3.8 Frequency Dependency of Isolated Myocyte Ca2+ Transient and Shortening ...... 145

4.3.9 Maternal Transmission of AAV-9 D73N TnC to Offspring ...... 149

4.4 Discussion ...... 153

4.4.1 Transduction of TnC into Cardiomyocytes ...... 153

4.4.2 Possible Routes of AAV-9 Transduction to the Heart after IP Injection ...... 154

4.4.3 Functional Alteration of the Intact Heart Due to a Desensitized TnC ...... 157

4.4.4 Insight into the Effect of a Desensitized TnC on Cardiomyocyte Performance .... 158

4.4.5 Maternal Transmission of AAV-9 and Clinical Implications ...... 159

4.5 Limitations and Obstacles ...... 159

4.6 Future Directions ...... 162

xiv

Appendix A: Protein Purification Protocols ...... 167

Appendix B: Purification of Ventricular Myofibrils ...... 183

Appendix C: Tissue Fixation and Imaging for GFP Expression ...... 189

References ...... 193

List of Tables

Table 1. Temperature Dependence on the Ca2+ Dissociation Rate of Modified TnCs. .... 92

Table 2. Echocardiographic Parameters and Values for Wild-type Mice...... 135

xvi

List of Figures

Figure 1. Ca2+ Dissociation from IANBD Labeled TnC in Tn and Tn Exchanged Rabbit

Rigor Myofibrils...... 35

Figure 2 ...... 38

Figure 3. Effect of ATP on Rabbit Ventricular Myofibrils and Reconstituted Thin

Filaments + S1...... 40

Figure 4. Dose-dependency of ATP and ADP on the Rate of Cross-bridge Detachment 46

Figure 5. Cross-bridge Detachment Rates from Varying Species of Myofibrils...... 51

Figure 6. Effect of Slowed Cross-bridge Detachment on the Rate of Ca2+ Dissociation from Rabbit Myofibrils ...... 54

Figure 7. Effect of Temperature on the Rates of Ca2+ Dissociation and Cross-bridge

Detachment as Reported by TnC in Rabbit Myofibrils ...... 58

Figure 8. Steady-State Ca2+ Sensitivities of Tn Exchanged Myofibrils ...... 73

Figure 9. Ca2+ Dissociation from Rigor Myofibrils with Modified TnC's ...... 75

Figure 10. Cross-talk Between Neighboring Tn on the Reconstituted Thin Filament. ... 78

Figure 11. Stopped Flow Titration of L48Q Myofibrils to Differentiate Fast and Slow

Phase Ca2+ Sensitivity ...... 83

Figure 12. ATP Dependent Changes in the Observed Fluorescence of TnCs with

Modified Ca2+ Binding Properties ...... 85

Figure 13. Cross-bridge Detachment as Reported by Modified TnCs ...... 88 xvii

Figure 14. DNA Sequence of Control TnC Gene ...... 110

Figure 15. Ca2+ Dissociation from Flag-tagged Control TnC ...... 122

Figure 16. Tissue from Mice Transduced with AAV-9 GFP...... 125

Figure 17. FLAG-tag Western Blot of AAV-9 D73N TnC Injected Mice...... 130

Figure 18. Survival Rates of AAV Injected Mice ...... 132

Figure 19. Echocardiograph and EKG Characterization of AAV Injected Mice ...... 136

Figure 20. Representative M-mode Short-Axis Echocardiographic and EKG Images . 137

Figure 21. Comparison Between Heart Weight and Myocyte Size of AAV-9 Injected

Mice ...... 143

Figure 22. Contractile Parameters of Isolated Myocytes from AAV Injected Mice. .... 144

Figure 23. Frequency Dependency of the Ca2+ Transient and Shortening Amplitudes. . 147

Figure 24. Internal Dimensions of Mice that were Maternally Transfected with AAV-9

D73N TnC ...... 151

Figure 25. FLAG-tag Western Blot of Maternally Transfected AAV-9 D73N Heart

Tissue ...... 152

xviii

Chapter 1: Introduction

1.1 General Introduction

At the most basic level, the function of the heart is to pump blood throughout the

body to supply the organs with freshly oxygenated and nutrient rich blood. When the

pumping action of the heart becomes insufficient, not only does the heart suffer but the vital organs such as the brain and kidneys do as well. There are many ways the pumping of the heart can be disrupted, which can ultimately lead to heart disease. Heart disease is a broad category that is composed of coronary artery disease, heart failure, arrhythmias and many other miscellaneous heart problems. Cardiomyopathy is a general term used to describe a disease of the heart muscle that leads to disturbed functionality. There are four main phenotypes that classify the different cardiomyopathies: dilated cardiomyopathy

(DCM), hypertrophic cardiomyopathy (HCM), restrictive cardiomyopathy (RCM) and unclassified or idiopathic [1] (a more detailed description will be presented later). Most of the time the causes of these cardiomyopathies are unknown but throughout the years many contributing factors have been identified. Cardiomyopathies have been associated with mutations in genes encoding practically every protein of the heart, as well as other non-cardiac related causes (metabolic disorders, high blood pressure, pregnancy, etc.) [1-

7]. The number of cardiomyopathy associated deaths is ~10,000 out of the ~600,000

total deaths caused by heart disease in the United States each year [8]. Although a

relatively small group, the study of these cardiomyopathies provides vital information to

better understand the contractile and relaxation properties of the heart. Understanding the

intrinsic properties of the beating heart may bring new light to potential therapeutic

targets for improving systolic and diastolic dysfunction, not only for the aforementioned

cardiomyopathies but for a wide range of heart diseases as well.

The difficulty in diagnosing purely diastolic dysfunction is that the signs and symptoms of diastolic heart failure are similar to those of systolic heart failure [9].

Furthermore, most of the treatments for diastolic dysfunction treat the symptoms, and not the cause, by using angiotensin-converting-enzyme inhibitors, Ca2+ channel blockers, diuretics and vasodilators rather than directly improving relaxation [9, 10]. The origins of altered relaxation have been implicated in all of the major processes required to keep the heart beating throughout its lifespan. These physiological processes include but are not limited to the electrical conduction system, the Ca2+ handling system, the

myofilament, the mitochondria (energetic), and the cytoskeleton or scaffolding network

[1]. The work described within focused on studying the molecular mechanisms of myofilament inactivation to contribute to the overall knowledge of heart relaxation. To accomplish this task, a physiologically relevant biochemical system (ventricular myofibrils) was used to elucidate the rate-limiting steps of myofilament inactivation. In

addition, a previously engineered myofilament protein (troponin C) with altered function

was incorporated into an animal model. This was a proof of principle study designed to

examine the effects that the modified troponin C (TnC) has on the in-vivo contractile properties of the heart and the potential development of a new therapy for improving cardiac relaxation.

1.2 Basic Cardiac Cycle

Every time the heart beats there is a contraction (systole) and relaxation (diastole) phase. The signal that initiates heart contraction arises from within the sino-atrial node

(SA) and generates an action potential (AP) that spreads throughout all of the myocytes in the heart. This electrical signal passes to and through the cells of the atria, via gap junctions, and reaches the atrio-ventricular node (AV), where the electrical signal is delayed to allow for atrial contraction. This delay is necessary to allow the ventricles to adequately fill with blood before they contract. Once past the AV, the electrical signal is conducted down the bundle of His, through the bundle branches and down to the terminal endings of the purkinje fibers. The AP is then transmitted throughout all of the ventricular cells to depolarize the myocytes in a nearly synchronous manner to cause the ventricles to contract and eject blood out of the heart. Relaxation of the ventricles soon follows the repolarization of the myocytes and allows the myocytes to recover for the next contraction. Thus, the cardiac cycle is composed of a contraction (systole) and relaxation (diastole) phase. Of particular importance is the fact that the electrical signal or action potential does not cause contraction but rather controls the release of intracellular Ca2+. In this regard Ca2+ acts as a second messenger to set in motion the

additional biochemical and mechanical process responsible for contraction of the heart.

A more in depth physiological description of the cardiac contraction and relaxation

phases will be described below.

1.3 Contraction of the Heart (Systole)

The process responsible for the regulation of the heart on a beat to beat basis is

known as excitation-contraction coupling[11]. Contraction of the myocyte is initiated

when the cell is electrically excited and becomes depolarized. The AP causes the

myocytes to depolarize due to the rapid opening of Na+ channels in the t-tubules and

2+ leads to the opening of the voltage sensitive Ca channel ICa,L [11]. The opening of the

Ca2+ channel allows a small amount of Ca2+ (~ 10 μM) to enter the myocyte and bind to the ryanodine receptors (RyRs) on the sarcoplasmic reticulum (SR) [12]. The binding of the extracellular Ca2+ to the RyRs results in an ~ 3 fold larger increase in Ca2+ release from the SR, the main intracellular Ca2+ storage site [11]. This large amount of Ca2+

release is required to raise the free intracellular concentration from the diastolic level of

100 nM to a systolic level of 600 nM due to the high Ca2+ buffering capacity of the

myocyte [13]. Under basal conditions, the amount of Ca2+ released in the myocytes is not

thought to be enough to saturate TnC. The magnitude of the SR Ca2+ release is

dependent on the degree to which the SR is filled with Ca2+, which is dependent upon the

duration between contractions [14]. Approximately 30% of the total Ca2+ transient is due

to the entry of extracellular Ca2+ into the myocyte and ~ 70% is from the intracellular release from the SR in the human [11]. The ratio of the intra/extracellular Ca2+ that

contributes to total Ca2+ transient varies between species [11, 15]. The process of calcium-induced-calcium release is responsible for the rise of the intracellular Ca2+ transient. The release of additional Ca2+ is an important physiological regulator of

contraction and will be discussed in more detail later.

The decline of the intracellular Ca2+ transient is due to the uptake of the

intracellular Ca2+ by the sarcoplasmic-endoplasmic reticulum Ca2+ ATPase (SERCA)/

phospholamban (PLB) complex back into the SR [11]. Once the Ca2+ transient reaches

SERCA, the sequestration of Ca2+ back into the SR begins. SERCA pumps Ca2+ back

into the SR at the expense of one ATP molecule per two Ca2+ ions [16] and the rate at which Ca2+ is pumped into the SR is regulated by the level of PLB phosphorylation [17].

In the dephosphorylated state, PLB reduces SERCA’s affinity for Ca2+, thus reducing the

influx of Ca2+ into the SR. Once PLB is phosphorylated, its inhibition of SERCA is

reduced and the rate at which the intracellular Ca2+ is pumped back into the SR is

increased [17]. To maintain Ca2+ homeostasis, the extracellular Ca2+ that entered into the

cell through the ICa,L channel is pumped back out of the cell through the sodium-calcium exchanger (NCX). NCX is an anti-porter membrane channel that pumps one Ca2+ out of the cell at the expense of the entry of three Na+ ions into the cell [18]. The decline of the

intracellular Ca2+ transient removes the Ca2+ from the myofilament, which ultimately

contributes to myocyte relaxation. The Ca2+ transient can be modulated to increase or

decrease the magnitude of the contraction. Although Ca2+ plays a vital role in heart

contraction, it is only a messenger and as previously described, Ca2+ needs a receptor to

bind and deliver its message. The message that Ca2+ is delivering on a beat to beat basis

is to initiate contraction and the recipient of that message is the contractile machinery of

the heart, the myofilament.

1.4 The Contractile Apparatus

The contractile apparatus or the myofilament is composed of thick and thin

filaments that are arranged in a parallel orientation to one another to form the sarcomere.

The sarcomere is the repetitive structural unit of the myofilament used to generate the force required for contraction through the sliding of the thick and thin filament proteins past one another (for review see [19]). The regulation and composition of the contractile apparatus will be described in more detail below.

The thin filament consists of troponin (Tn), tropomyosin (Tm) and actin. The thin filament maintains a strict protein stoichiometry that consists of one Tn complex and one

Tm dimer for every seven actin monomers. The Tn complex consists of three protein subunits, Troponin C (TnC), Troponin I (TnI) and Troponin T (TnT). There are approximately 25,000 TnCs per half sarcomere for each myofibril in a myocyte [20, 21]

and roughly 1014 sarcomeres in the entire left ventricle [22]. The sarcomeres are arranged

in series and in parallel and packed into myofibrils. Myofibrils are then packed in

parallel to form the contractile apparatus. TnT binds TnC and TnI to anchor the Tn complex to actin-tropomyosin, while TnI interacts with TnC, TnT, and actin to inhibit force production at resting Ca2+ levels (for review see [23]). TnC is the Ca2+ sensor and

upon the binding of Ca2+, a series of structural changes spreads throughout the contractile

apparatus (described in more detail later). The cardiac thick filament is composed of

myosin II, the motor protein that binds actin to produce force and muscle contraction.

There are approximately 90,000 myosin heads per half sarcomere in the ventricular

myocyte [24]. When the thin filament is activated by Ca2+, the binding site for myosin is

revealed and allows myosin to bind and interact with actin. Myosin will bind to actin in

the presence of the ATP hydrolysis products (ADP + Pi). When ADP + Pi dissociate

from myosin, the myosin cross-bridges slide the thin filament past the thick filament in an anti-parallel fashion which results in contraction and force production. A more in depth discussion of the myofilament proteins and their molecular interactions will follow.

For a more thorough review of thin filament structure and the different protein-protein

interactions the reader should refer to the following reviews [19, 23]

1.4.1 Thin Filament Regulation

The myofilament is classically thought to exist in one of three states [25, 26]. The three states are in reference to the level of thin filament activation and the position that

Tm occupies as it sits on top of actin (blocked, closed, and open). Which state is occupied is determined by the availability of Ca2+-binding to TnC and the state of myosin

in regards to its relationship with actin. The processes of thin filament regulation will be

described below.

As mentioned before, TnC is the Ca2+ sensor of the myofilament and is

responsible for the initial activation steps of thin filament activation. Structurally, TnC

resembles a dumbbell with the N-terminal and C-terminal making up the two globular

ends [27]. There are three divalent metal ion binding sites in cardiac TnC [19] that

contain the helix-loop-helix motif that is typically found across all EF hand Ca2+ binding

proteins [28]. The N-terminal domain of TnC (site II) is the regulatory domain of TnC

and is responsible for binding the intracellular Ca2+ released with each beat of the heart 7

[29]. Site I of cardiac TnC is unable to bind Ca2+ due to specific amino acid substitutions

that render it incapable of binding the metal ions. Having an N-terminal with only one

Ca2+/Mg2+ binding domain is a distinguishing characteristic between cardiac TnC (also known as the slow skeletal TnC isoform) from the fast skeletal TnC, which has two functional Ca2+ binding sites [19]. Site II of cardiac TnC binds Ca2+ with an affinity of ~

10-5 M and a Mg2+ affinity of 10-3 M [30]. The N-terminal domain of TnC is able to quickly respond to the rise of intracellular Ca2+ released during the electrical stimulation of the myocytes. The C-terminal domain of cardiac TnC is considered the structural domain since it is thought to be always bound with either Ca2+ or Mg2+ at sites III and IV under all physiological conditions due to increased Ca2+ and Mg2+ affinities (10-7 M and

10-4 M, respectively [30]). Being constantly bound with Ca2+/Mg2+ keeps the structural

domain of TnC in contact with N-terminal domain of TnI, thus anchoring TnC into the

Tn complex. The binding of Ca2+ to the regulatory domain of TnC changes its structure and results in a cascade of structural changes within the thin filament that subsequently

causes myocyte contraction [29]. The Ca2+ regulation of the myofilament protein

interactions will be described below.

Prior to TnC binding the activating Ca2+, the thin filament is considered to be in

the blocked state. The actin filament is composed of actin monomers that from a

longitudinal double stranded helix. The coiled coil dimer of Tm is positioned on top of

actin in such a way that it appears to wrap around actin from one side to the other [19].

The Tm dimer gains stability from the hydrophobic residues of the non-polar amino acids

of each monomer that interacts with each other in a head to tail fashion [31]. Each Tm

dimer sits atop the actin and blocks the myosin binding sites on actin. Upon the binding

of Ca2+ to the N-terminal of TnC, a conformational change in TnC occurs and results in the subsequent binding to the switch region of the C-terminal of TnI [29]. The binding of

TnI to TnC further opens the N-terminal Ca2+ binding domain of TnC and leads to the activation of the thin filament. The binding of the TnI switch peptide to TnC weakens the interaction between TnI’s inhibitory peptide (IP) and actin, leading to IP dissociation from actin. Once the IP region of TnI is removed from the actin-Tm complex, the azimuthal or lateral movement of Tm across actin is permitted [32]. Tm moves across

actin and towards the groove in the actin helix created by the actin monomers [33]. This

movement begins to expose the myosin binding sites and allows myosin to weakly bind

actin in what is considered to be the closed state of muscle activation [33]. The final step

of muscle activation involves the transition of myosin from a weakly bound state to a

strongly bound state [19]. Strongly bound myosin continues to further expose additional

myosin binding sites on actin in a cooperative manner [19]. In doing so, the thin filament

is then considered to be in the open state and capable of producing force through the anti-

parallel sliding of the thin and thick filaments past one another .

1.4.2 Thick Filament Regulation

The thick filament is composed of myosin and the cardiac myosin binding protein

C (cMyBP-C) [19]. Myosin II is composed of two heavy chains (MHC) and four light

chains (an essential and a regulatory light chain). Myosin can be divided into different

regions and is composed of a tail, neck, and head region. The C-terminal tails of two 9

MHCs dimerize to form a two chain coiled coil structure that comprises the myosin

molecule [19]. Each tail of the myosin molecule comes together to form the backbone of

the thick filament. Extending from the thick filament is the myosin neck. The neck

region extends from the thick filament and connects the myosin heads (N-terminal

globular structure) to the thick filament. The neck projects the head region outwards

from the thick filament to bind actin. The neck and head region, termed sub-fragment

1(S1), is commonly used in many biochemical studies to mimic the presence of the entire

myosin structure [32, 34, 35]. Each cardiac MHC can be expressed as two different

isoforms (i.e. MHC-α and MHC-β) [19]. The isoforms of MHC are differentiated from

one another based on the rate of their intrinsic ATPase activity and thus their maximum

velocity of shortening that they can produce [36, 37]. In humans, the predominate MHC

isoform is MHC-β expressed at around ~ 90% and can be even higher during heart failure

[36]. The last component of the thick filament is cMyBP-C. cMyBP-C is believed to be involved in the modulation of the Ca2+ activation of the thick filament and the regulation of the myosin cross-bridge cycle (for review see [38]).

The structure and position of the myosin head is dependent upon its nucleotide bound state. ATP is hydrolyzed to produce ADP and Pi and in the process allows myosin to transition through the cross-bridge cycle to generate the force needed for contraction of the myocyte [33]. For simplicity, the scheme of the cross-bridge cycle will be explained starting with the myofibrils in the rigor state (nucleotide free). In all reality, there is not an actual start point to the cross-bridge cycle. Starting from a rigor state, the cycle begins with the binding of ATP to the S1 head (step 1). The binding of ATP to myosin results in

10 the detachment of S1 from actin (step 2). After S1 dissociates from actin, ATP is hydrolyzed by myosin to produce the products of ADP and Pi, which remain bound to myosin (step 3). The hydrolysis of ATP puts energy into myosin, much like the cocking of a pistol. The likelihood that ADP and Pi will dissociate from myosin when it is unbound to actin is prevented or reduced due to detached myosin’s intrinsically high affinity for the hydrolysis products. The hydrolysis of ATP allows myosin to weakly bind to the thin filament as long as Ca2+ is bound to TnC (step 4). If actin is available, myosin transitions from a weakly bound to a strongly bound state (step 5). The next step

(6) is the release of the Pi from myosin that results in the power stroke or the force generating step of myosin activation. During the power stroke, the neck region swings relative to the position of the actin bound myosin head and causes the sliding of the thin filament past the thick filament. For each ATP molecule, the myosin head moves ~ 10 nm along the thin filament [39]. Step 7 is the proposed isomerization step that occurs within the actin bound myosin before ADP can dissociate from myosin. Finally, ADP is released from myosin (step 8) to allow the next ATP molecule to bind to myosin and dissociate the thick and thin filaments from one another. For a more in-depth review of the cross-bridge cycle consult [33].

The actual rate-limiting step of the cross-bridge cycle in cardiac muscle appears to be dependent upon the biochemical system used to measure the rates of the cross-bridge cycle [24, 40-42]. These proposed steps are thought to be the isomerization step, the release of Pi, or the release of ADP from myosin [24, 41-45]. The reason why it is important to know the rate-limiting step of the cross-bridge cycle is because this rate is

also the proposed rate-limiting step of cardiac muscle relaxation [24, 45]. Thus, the rate of cross-bridge detachment will be further studied in the ventricular myofibrils across various species.

1.5 Relaxation of the Heart (Diastole)

In the physiological context of the heart, the relaxation phase or diastole is a vital

process in maintaining normal function on a beat to beat basis. Relaxation of heart involves complex feedback interactions between the intracellular processes of the myocytes and the extrinsic loading conditions [46]. Cardiac relaxation can be divided into two phases; isovolumetric relaxation followed by isotonic relaxation [47]. During isovolumetric relaxation, the pressure of the LV is decreased while maintaining a fixed volume. A reduced pressure in the ventricles before they fill with blood results in an almost passive filling of the ventricles during isotonic relaxation. In addition, the rapid drop in pressure that occurs before the filling of the ventricles allows the coronary arteries to be perfused with freshly oxygenated blood. This pressure reduction is favorable for the passive filling of the ventricle and lengthening of the myocytes during isotonic relaxation. The following will describe more specifically, the mechanical events that occur during the inactivation of the myofilament.

1.5.1 Myofilament Inactivation

Inactivation of the myofilament, in the most basic sense, can be thought of as the reversal of the intracellular steps which resulted in contraction. These intracellular events are highly modulated in the healthy heart to meet the metabolic demands of the body as discussed below [47]. Relaxation begins with the active removal of the cytosolic Ca2+ as

it is pumped back into the SR by SERCA and pumped out of the cell by NCX [18]. As

the intracellular Ca2+ concentration is reduced, Ca2+ dissociates from TnC to inactivate the

thin filament. In the absence of Ca2+, the interaction between TnC and TnI is drastically

reduced and allows TnI to rebind to actin. The rebinding of the TnI inhibitory peptide to

actin inhibits the movement of Tm and keeps Tm in the blocked state [19]. Until the thin filament becomes inactivated, myosin continues to progress through the cross-bridge cycle to maintain pressure [19]. Once the cross-bridge has dissociated from actin and the intracellular Ca2+ concentration is reduced, the myosin cross-bridge cycle is temporarily

halted due to the inability of myosin to bind actin. Thus, myosin binding to actin is

inhibited by the inactivation of the thin filament.

1.5.2 Proposed Biochemical Rate-limiting Steps of Myofilament Relaxation

There are multiple proposed biochemical rate-limiting steps of myocardial

relaxation. These intracellular biochemical and mechanical changes include the decline

of the Ca2+ transient, thin filament inactivation controlled by Ca2+ dissociation from TnC, and cross-bridge detachment from actin [17, 23, 24, 43]. The exact rate-limiting mechanism is currently unknown. Modifications in any of the three proposed rate- 13 limiting steps of relaxation have been shown to alter the properties of contraction and relaxation in-vitro and in-vivo [48-50]. This work will focus on studying the rates of myofilament inactivation through the measurement of the rates of Ca2+ dissociation from

TnC as well as the measurement of cross-bridge dissociation. Since these rates cannot be currently measured in the heart they need to be measured in the next best system, the ventricular myofibrils. Additional evidence for the role that TnC plays in modulating the rate of cardiac muscle relaxation will be described in subsequent chapters.

1.6 Physiological Modulation of Contraction and Relaxation

Each proposed rate-limiting step of myocardial relaxation can be modulated, whether it is through the phosphorylation of proteins, activation or deactivation of ion channels or direct modulation of Ca2+ cycling [51-53]. To meet the increased demands of the body, the contractile proteins of the heart are regulated on a beat to beat basis. One example of physiological regulation is beta-adrenergic stimulation, which leads to an increase in the rate and strength of contraction as well as an increase in the rate of relaxation [11]. The release of catecholamines (nor-epinephrine, epinephrine) binds to and activates the beta-adrenergic receptors (β-AR), primarily β1, on the myocytes and leads to the activation of a stimulatory G-protein (Gs). Once activated, Gs activates adenylate cyclase, which leads to an increase in cyclic AMP (cAMP) levels. The increase in cAMP results in the activation of protein kinase A (PKA). PKA is then

2+ responsible for phosphorylation of a wide range of Ca handling proteins (ICa,L and PLB) as well as TnI in the myofilament . For a thorough review on the β-AR pathway please 14

see [54]. The overall effect of β-AR stimulation at a mechanical level is to increase the cardiac output of the heart through the modulation of specific proteins described below.

Contractility of the heart is increased by increasing the cytosolic level of Ca2+

released during systole. The Ca2+ transient amplitude can be increased by the phosphorylation of ICa,L, RyR, and PLB. PLB phosphorylation reduces its inhibitory

effect on SERCA to increase the rate of Ca2+ cycling and the SR Ca2+ load [55]. The increased SR Ca2+ load allows more Ca2+ to be released into the cytosol and thus, more

Ca2+ available to activate the myofilament [54]. The extra release of Ca2+ is able to bind

and activate additional TnC in the myofilament and increase the strength of contraction.

To adjust to the increase in the heart rate (HR) and rate of contraction, the rate of

relaxation is also increased to keep the heart functioning properly. If relaxation was not

also increased, the heart would not be able to fill with an adequate volume of blood

before the next contraction. To increase the rate of relaxation, Ca2+ uptake back into the

SR is increased as well as a decrease in the myofilament Ca2+ sensitivity, which is

brought about by the phosphorylation of TnI. The phosphorylation of TnI alters the

interaction between TnC and TnI to allow Ca2+ to dissociate from TnC at a faster rate and

leads to quicker myofilament inactivation [56]. Thus, the modulation of TnC’s Ca2+ binding properties under conditions of physiological regulation exemplifies the importance of Ca2+ dissociation in modulating the rate of myocardial relaxation.

1.7 Cardiomyopathies Resulting in Altered Ca2+ Binding Properties

Similar to physiological regulation, there are a multitude of mutations in myofilament proteins that are known to alter Ca2+ sensitivity of the heart, both in-vitro

and in-vivo [1]. Of particular interest are the mutations that cause dilated (DCM) and

hypertrophic (HCM) cardiomyopathies. DCMs are characterized by the dilation of the

ventricles, weakened or impaired contractility, and a decreased Ca2+ sensitivity of force

production [1]. On the other hand, HCMs present with increased ventricle thickness,

reduced diastolic volume, and a reduced relaxation rate that eventually leads to a declined

cardiac output [1]. Unlike DCM, HCMs are typically characterized by an increased Ca2+

sensitivity [57]. The similarity between DCM and HCM is that the various myofilament

mutations resulted in altered myofilament Ca2+ sensitivities.

Although there are a few mutations in TnC that directly alter the Ca2+ sensitivity

of the myofilament [58, 59], most of the myofilament protein mutations affect TnI, TnT,

Tm, actin, MyBP-C, and myosin [1]. Although TnC is the Ca2+ sensor of the

myofilament, mutations in these proteins also result in cardiomyopathies with altered

Ca2+ sensitivities, as previously stated. The myofilament mutations seem to feed back

onto TnC, which acts as a central hub in determining the overall myofilament Ca2+

sensitivity and rate of thin filament inactivation [5, 7, 60-64]. Cardiomyopathies presenting with altered Ca2+ binding properties provide additional evidence for the role of

TnC in the heart. To further study thin filament inactivation during cardiac muscle

relaxation, the Ca2+ dissociation kinetics from TnC need to be measured. The following section will briefly describe how the Ca2+ dissociation kinetics are measured from TnC in

a physiologically relevant biochemical system to study the proposed rate-limiting steps of myofilament inactivation.

1.8 Measuring the Rates of Ca2+Dissociation from Troponin C

An important tool for studying the Ca2+ binding properties of the myofilament is the previous development of a fluorescently labeled cardiac TnC [65, 66]. The fluorescently labeled TnC allows the Ca2+ binding properties to be studied in multiple

biochemical systems that includes isolated TnC, the Tn complex, reconstituted thin

filaments, and ventricular myofibrils [34]. A fluorescent probe (IAANS or IANBD) is

covalently attached to TnC that does not alter the intrinsic function of TnC and its ability to incorporate into the thin filament, myofibrils, or muscle [34, 67, 68]. Because the fluorescent probe is environmentally sensitive it is able to report the structural changes in

TnC as Ca2+ binds and dissociates. In the next chapter, I will describe that it is also

sensitive to other structural changes in the myofilament (cross-bridge detachment).

The simple biochemical systems have their pros and cons for studying TnC, but

are most helpful in reducing the system to the elementary myofilament proteins to

elucidate the potential mechanisms of relaxation [35]. Since TnC works in concert with the other proteins of the thin and thick filament, it is imperative to incorporate TnC into systems more complex than isolated TnC, Tn and the reconstituted thin filaments. The myofibrils are the biochemical system of choice because they maintain the geometric structure of the contractile proteins at stoichiometric protein ratios [69]. The myofibril consists of longitudinal stacks of sarcomeres and multiple myofibrils combine together in 17

parallel to form the contractile machinery of the myocyte. The contractile proteins of

myofibrils are arranged to maximize thin/thick filament interactions while maintaining

the sarcomeric units, which the simple biochemical systems lack [42]. Using myofibrils,

the kinetic mechanisms of myofilament inactivation (Ca2+ dissociation and cross-bridge

detachment) can be studied from a single biochemical system.

1.9 Utilization of Troponin C’s with Altered Ca2+ Binding Properties

We believe that TnC plays a significant role in modulating the rate of

myofilament inactivation. Evidence for this is supported by the fact that TnC Ca2+ sensitivity is modulated by β-AR stimulation, direct mutation in TnC can result in cardiomyopathies with altered Ca2+ sensitivities, and mutations in the additional

myofilament proteins feedback onto TnC to alter the Ca2+ sensitivity [70]. Thus, we hypothesize that the rate at which Ca2+ dissociates from TnC plays an important role in

modulating the rate of cardiac muscle relaxation. To test this hypothesis we have

previously developed TnCs with specific amino acid mutations that have an array of Ca2+

sensitivities as well as Ca2+ dissociation rates in many different biochemical systems [30,

34, 35, 66, 71]. The effect that TnCs with modified Ca2+ binding properties have on the

proposed steps of myofilament inactivation will be further studied in the myofibrils in

Chapter 3. To study the physiological influence of these TnCs on the in-vivo function

they need to be incorporated into the beating heart. An emerging technique for

incorporating proteins into animal models has been through the in-vivo transfection with

adeno-associated virus (AAV) containing the gene of interest. AAV’s containing the modified TnCs will be studied in the mouse model as described in Chapter 4

1.10 Overall Goal and Significance

The goal of our research is to study the potential rate-limiting steps of myocardial relaxation and bring to light any new potential therapeutic targets for treating diastolic dysfunction. One target that we are specifically interested in is TnC and the role it plays in modulating contraction and relaxation of the heart. To accomplish this we have thoroughly characterized the properties of TnC in the myofibril and how the alteration of its Ca2+ binding properties can affect the rest of the myofilament. Furthermore, the end goal of our research is to correct the altered Ca2+ sensitivity and the disease related phenotypes caused by the wide range of mutations in the contractile proteins of the heart.

The idea would be to incorporate a specifically designed TnC into the heart that has a

Ca2+ sensitivity that would compensate for that of the diseased tissue. For example, the

TnI R192H mutation has an increased Ca2+ sensitivity of force production with prolonged

relaxation [48, 72, 73]. The incorporation of a TnC with a decreased Ca2+ sensitivity into

the myofilament would correct the diseased phenotype caused by TnI to increase the rate

of Ca2+ dissociation and thus the rate of relaxation. Our lab has been able to correct the disease-related altered myofilament Ca2+ sensitivities caused by both TnI and TnT

cardiomyopathies in multiple biochemical systems by using these specifically designed

TnCs [70].

Before TnCs with specifically designed Ca2+ sensitivities can be used as a therapeutic to correct the aberrant Ca2+ sensitivities for the wide range of diseased animal

models, the feasibility of the technique needs to be tested. To accomplish this task,

TnC’s with either drastically increased or decreased Ca2+ sensitivities were incorporated

into the AAV-9 viral vector to target the heart. Once injected we were able to examine

the effect that TnCs with altered Ca2+ binding properties had on the overall function of

the heart in an animal model. Thus, AAV-9 containing the TnC gene of choice could be used as a treatment for correcting the diastolic dysfunction caused by the numerous familial cardiomyopathies. Furthermore, if TnC is able to specifically treat diastolic dysfunction it could be used treat the many other forms of heart disease presenting with impaired relaxation.

Chapter 2: The Rates of Ca2+ Dissociation and Cross-Bridge Detachment from Ventricular Myofibrils

2.1 Introduction

Cardiac function is dynamically regulated to meet the demands of the human

body [26, 74]. However, in many cardiovascular diseases relaxation becomes impaired,

which can lead to diastolic dysfunction of the heart [75, 76]. A more complete

understanding of the molecular mechanisms that govern cardiac muscle relaxation will

help to develop better treatment strategies for diastolic dysfunction. Biochemically, there

are two main factors thought to control the rate of relaxation; the decline of the

intracellular Ca2+ transient and the inactivation of the myofilament [43, 59, 65].

Although the removal of intracellular Ca2+ is a vital component to relaxation, it would

appear that myofilament inactivation is equally important, if not rate-limiting [43, 77]. In order for the myofilament to relax, Ca2+ must dissociate from Troponin C (TnC) to

inactivate the thin filament and cross-bridges must detach from actin to alleviate the force

[59]. It is generally thought that the rate of cross-bridge detachment is substantially slower than the rate of Ca2+ dissociation from TnC and thus rate-limits myofilament

inactivation [42, 59]

ATP binding to the acto-myosin complex detaches myosin from actin. However,

this ATP cannot complete the cross-bridge cycle and detach myosin until the previous hydrolysis products of ATP (phosphate (Pi) and ADP) are released [40]. It is believed that Pi release is associated with either the transition of myosin from a weakly bound

state to a strongly bound state or with the force producing power stroke [78-80]. In

cardiac muscle, ADP release is thought to follow the power stroke and must dissociate

before another ATP can bind and detach myosin [33, 81]. The rate of ADP dissociation from myosin has been proposed to rate-limit cross-bridge detachment and potentially relaxation [41, 45]. In support of this idea, increasing concentrations of ADP slowed the rate of cross-bridge detachment from acto-myosin [45] and slowed the rate of relaxation of skeletal myofibrils and skinned cardiac muscle [82, 83]. Furthermore, the rate of ADP dissociation from acto-myosin and the rate of cardiac muscle relaxation were substantially slower when each system utilized β-myosin as compared to α-myosin [49].

However, ADP dissociation has only been measured from un-regulated acto-myosin and was quantitatively an order of magnitude faster than the rate of cardiac muscle relaxation

[45, 83-85]. Either the rate of ADP dissociation does not rate-limit cardiac muscle relaxation or the actual rate of ADP dissociation is slower when measured from within the confines of the sarcomere.

The other component of myofilament relaxation is the inactivation of the thin filament, which is controlled by Ca2+ dissociation from TnC. Ca2+ dissociates from

isolated TnC at least two orders of magnitude faster than the rate of myocardial relaxation

[66, 86, 87]. Due to this fact, Ca2+ dissociation from TnC was deemed an insignificant

regulator of relaxation. However, TnC does not function in isolation, but as an integral

component of the thin filament system. Incorporation of TnC into the thin filament

slowed the rate of Ca2+ dissociation by at least an order of magnitude [34]. The binding

of myosin-S1 to the thin filament further slowed the rate of Ca2+ dissociation from TnC

an additional order of magnitude [34]. These biochemical studies suggest that the rate of

Ca2+ dissociation can be slowed to a rate that could influence the rate of relaxation. A recent study increased the complexity of the biochemical system even further by using ventricular myofibrils [85]. This study concluded that the rate of Ca2+ dissociation from

TnC in isolated cardiac myofibrils was still too rapid to rate-limit myofibril relaxation at

10oC. However, the rate of Ca2+ dissociation from cardiac TnC is little affected by

temperature [30], whereas cross-bridge function is highly temperature sensitive [84]. The

actual rate of Ca2+ dissociation from TnC and the rate of cross-bridge detachment in

ventricular myofibrils at physiological temperature are unknown.

To further investigate the rate-limiting steps of myofilament inactivation we

utilized a fluorescent TnC that was able to report the rate of Ca2+ dissociation from TnC

and cross-bridge detachment in ventricular myofibrils over a wide range of temperatures.

Myofibrils provided a unique biochemical system in which both Ca2+ dissociation and cross-bridge detachment could be measured from within the confines of the regulated sarcomere, which contained the myofilament proteins at a physiologically relevant geometry and stoichiometry.

Cross-bridge detachment in the absence of ADP was always faster than the rate of

Ca2+ dissociation from TnC. However, the addition of ADP substantially slowed the rate

of cross-bridge detachment across all temperatures. This would suggest that cross-bridge

detachment is rate-limited by ADP dissociation. At temperatures less than ~20oC, the

rate of cross-bridge detachment in the presence of ADP was slower than the rate of Ca2+

dissociation. This would suggest that cross-bridge detachment rate-limits myofilament

inactivation and potentially relaxation at cold temperatures. However, the rate of cross-

bridge detachment was more sensitive to changes in temperature than was the rate of Ca2+

dissociation. At a more physiological temperature (35oC) and [ADP], the rate of cross- bridge detachment may actually be faster than the rate of Ca2+ dissociation. This

provides evidence that there may not be a simple, single rate-limiting step of myofilament

inactivation. Thus, Ca2+ dissociation and cross-bridge detachment may both influence the rate of relaxation and may be potential therapeutic targets for improving relaxation across a broad range of cardiomyopathies.

2.1 Methods

The following section includes an in-depth description of the methods that were used to complete the work described in this chapter. Many of the methods described are commonly used and will include the basic experimental procedures required to complete the specific assay. Others, such as the myofibril preparation will include more in-depth

details that will hopefully allow the reader to complete the preparation as easily as

possible. These same methods were utilized in the next chapter as well.

2.1.1 Materials

Phenyl-Sepharose, Tween 20, EGTA, ATP and ADP were purchased from Sigma

Chemical Co. (St. Louis, MO). 4-(N-(iodoacetoxy)ethyl-N-methyl)amino-7-nitrobenz-2- oxa-1,3-diazole (IANBD) was purchased from Invitrogen (Carlsbad, CA). N-(3-

Dimethylaminopropyl)-N-ethyl-carbodiimide hydrochloride (EDC) and N- hydroxysulfosuccinimide sodium salt (NHS) were purchased from Fluka Analytical. All other chemicals were of analytical grade.

2.1.2 Protein Purification

The human cardiac TnC with the C35S, T53C, and C84S mutations (herein designated TnCT53C) was expressed and purified as previously described [34]. The D65A mutation was inserted into the TnCT53C expression vector using techniques previously described [34]. The mutations were confirmed by DNA sequence analysis. Expression and purification of D65A TnCT53C was conducted as previously described [35]. Human

cardiac TnI and TnT (isoform 3) were expressed and purified as previously described

[34]. Detailed protein purification protocols are presented in Appendix A.

2.1.3 Troponin C labeling with IAANS or IANBD

TnC was labeled with either the IAANS or IANBD fluorescent probe at the Cys that

strategically replaced Thr 53. The two other endogenous Cys at amino acids 35 and 84

were point mutated to Thr. TnC containing the T53C modification that had been

previously reduced with 1mM DTT was dialyzed in 2 L of 50 mM Tris, 90 mM KCl, 1

mM EGTA, 6 M Urea, at pH 7.5 overnight. The concentration was determined after

dialysis (extinction coefficient of TnC = 4470). Based on the concentration of TnC,

IAANS or IANBD were added to the TnC at a concentration that was 5x the

concentration of the TnC. A 10 mM stock of the fluorescent probe (IAANS = 0.5

mg/100 μl, IANBD = 0.4 mg/ 100 μl) was made in dimethylformamide, making it fresh

each time used. The fluorescent label was slowly added drop wise to the TnC and the

tube containing TnC was capped and inverted or swirled in between drops. The tube was

mixed well, wrapped with foil and rocked for 5-8 hours at 4oC or 3 hours at room

temperature. 5 mM DTT was added to the labeled TnC to stop the reaction. The TnC

was dialyzed overnight in the same buffer that TnC was initially dialyzed in, with an additional 0.5 mM DTT added to the buffer. The TnC was removed from dialysis and the concentration was measured based on the fluorescent probe used. TnC labeled with

IAANS was scanned from 250-400 nm and absorbance read at 325 nm. The concentrations was then determined using Beers Law A=εcl, where extinction coefficient

= 24,900. TnC labeled with IANBD was scanned from 450-550 and absorbance read at

480 nm, extinction coefficient = 25,000.

2.1.4 Troponin Complex Formation

The labeled TnC was dialyzed individually in dissolving buffer (10 mM MOPS,

4.6 M Urea, 0.01 % NaN3, 0.5 mM DTT, at pH 7.0) prior to complex formation. The Tn

was reconstituted using TnT, TnI, and TnC at 12 μM, 12 μM, and 8 μM, respectively.

The volume needed for each protein was determined, the values were summed, and then subtracted from the total volume to determine how much buffer would be needed. The

Tn complex buffer (10 mM MOPS, 4.6 M Urea, 0.01% NaN3, 10 mM DTT at pH 7.0)

was added to a centrifuge tube and the proteins were slowly added to the buffer in the 26

order of TnT, TnI and TnC. After the addition of TnT and TnI the solution had a cloudy

appearance that disappeared after the addition of TnC. The Tn complex was placed into a

6,000-8,000 MW dialysis tube and dialyzed against 5 buffers (4 L each) for at least 8

hours a piece. Buffer 1: 10 mM MOPS, 2 M Urea, 1 M KCl, 0.01 % NaN3, 0.5 mM

DTT, 3 mM MgCl2, pH 7.0. Buffer 2: 10 mM MOPS, 1 M KCl, 0.01 % NaN3, 0.5 mM

DTT, 3 mM MgCl2, pH 7.0. Buffer 3, 4, 5: 10 mM MOPS, 150 mM KCl, 0.01 % NaN3,

0.5 mM DTT, 3 mM MgCl2, pH 7.0. After dialysis the Tn complex was removed from

the dialysis bags and placed into 1.5 ml micro centrifuge tubes and centrifuged at 14,000

x g for 20 minutes at 4oC to remove any precipitated proteins. The complexes were

stored at -80oC.

2.1.5 Myofibril Preparation

Ventricular cardiac muscle was obtained from male New Zealand White rabbits

(2-3-month old, ~ 2 kg weight [88]), male LBN-F1 rats (175-225 g), heartworm free

mongrel dogs (weighing 19.0 ± 0.4 kg; 2–3 year old [88]), and de-identified failing

human tissue from ongoing studies with Dr. Paul Janssen. All ventricular muscle was

dissected in a Krebs-Henseleit solution containing 137 mM NaCl, 5 mM KCl, 1.2 mM

MgSO4, 1.2 mM NaH2PO4, 20 mM NaHCO3, 10 mM glucose, 0.25 mM CaCl2 and 20 mM 2, 3-butanedione monoxime (BDM) to prevent contraction and damage to the cut

tissue. The ventricular myofibrils were prepared by the method previously described

[89]. An in-depth description can be found in Appendix B. Briefly, the isolated ventricular tissue was minced with scissors in Krebs-Henseleit solution and then 27 homogenized with 10s bursts of a polytron homogenizer. The suspension was filtered through cheesecloth to remove connective tissue and then further Dounce homogenized.

The myofibrils were collected by centrifugation and washed in rigor buffer before resuspension in glycerol rigor buffer. The myofibril stocks were stored at -20oC. All animals and tissues were handled in accordance with the National Institutes of Health

Guidelines and approved by the Institutional Laboratory Animal Care and Use

Committee at The Ohio State University. All human tissue was obtained in accordance with a tissue acquisition protocol approved by the Biomedical Sciences Institutional

Review Board of The Ohio State University.

2.1.6 Determining the Myofibril Concentration

0.5 ml of myofibrils from the glycerol stock were washed 3x with water and pelleted at 2000 x g in a pre-weighed 1.5 ml micro-centrifuge tube to remove the buffer components. After the last wash the supernatant was removed from the myofibril pellet.

The myofibrils were vacuum centrifuged for two hours to allow the proteins to dry.

Three aliquots of myofibrils were prepared and the final dry weight of the myofibrils was determined and averaged. The stock contained ~ 5.6 mg/ml of myosin and was based on the fact that myosin composes ~ 43% of the total myofilament weight. The concentration of myosin in the final diluted volume of buffer to be used for the stopped- flow experiments was ~ 1 μM (calculated from myosin having a molecular weight of

500,000 g/mol).

2.1.7 Exchange of Labeled Tn into the Myofibrils

Many of the stopped-flow and steady state experiments required Tn to be

exchanged into the thin filament of the myofibrils to be able to observe the apparent rates

of Ca2+ dissociation and cross-bridge detachment. To accomplish this, a 500 μl aliquot

of myofibrils from the glycerol stock was placed into a 1.5 ml centrifuge tube and washed

in Working Buffer ((WB), 10 mM MOPS, 150 mM KCl, 3 mM MgCl2, 1 mM DTT and

0.02% Tween 20 at pH 7.0) to remove the cryogenic glycerol solution. Briefly, 1 ml of

WB was added to the myofibrils and pelleted at 2000 x g for one minute, supernatant was

removed, and the myofibril pellet was re-suspended in another 1 ml of WB by pipetting

up and down (5-7 times). This wash step was repeated two more times and after the final centrifugation and removal of supernatant, the myofibrils were re-suspended in 1 ml of the fluorescently labeled Tn (~ 7 μM stock concentration). The tubes were wrapped in aluminum foil and placed in 4oC overnight (~ 12-18 hrs). The myofibrils were initially

placed on a shaker overnight but that resulted in the clumping of the myofibrils without

increasing the amount of Tn incorporated into the myofibrils.

The next morning the Tn exchanged myofibrils were pelleted at 2000 x g and

washed three times in WB as mentioned above. When measuring Ca2+ dissociation

kinetics from the myofibrils, 200 μM Ca2+ was added to the WB before washing the

myofibrils or if measuring cross-bridge detachment, 5 mM EGTA was added to the WB.

After the final centrifugation of the washed myofibrils they were re-suspended in the appropriated WB and then diluted to 4.5 ml with the same WB (Ca2+ or EGTA) that was 29

required for each specific experiment. The diluted myofibrils were filtered through a

nylon mesh that had pores with an internal diameter of 100 μm. The nylon mesh filter

was wrapped around the top of a 15 ml centrifuge tube and a fine tip transfer pipette was

used to slowly drop the myofibrils into the center of the filter. Care was taken to slowly

filter the myofibrils or unfiltered myofibrils would clump together on top of the filter and

flow over the edge. The myofibrils were initially vacuum filtered but that resulted in a

greater loss of myofibrils and it would pull through a greater proportion of large/chunky

myofibrils through the filter. However, if there was a large amount of lipofuscin in the

myofibrils (dog or human) the myofibrils were very sticky and made it nearly impossible

to gravity filter through the nylon mesh (> 50% loss of myofibrils). Instead, the

myofibrils had to be vacuum filtered through the nylon mesh into a small Erlenmeyer

flask. After filtering, the myofibrils were ready to be used in the respecitive WB (± Ca2+

or EGTA).

2.1.8 Cross-linking of Myofibrils

After Tn exchange, the myofibrils were washed three times in WB without DTT.

5mM NHS and 2mM EDC were added to the myofibrils, inverted multiple times and set

in 4oC. After 15, 30 or 60 minutes, the reaction was quenched with the addition of 25mM

glycine and 10mM DTT for 15 minutes at 4oC. After the allotted time the myofibrils

were washed three times with WB, re-suspended in 5ml buffer and filtered through 100

micron nylon mesh the same way the Tn exchanged myofibrils were prepared.

2.1.9 Stopped-Flow Kinetic Measurements from Myofibrils

The kinetic values were measured using an Applied Photophysics Ltd.

(Leatherhead, U.K.) model SX.18 MV stopped-flow instrument with a dead time of 1.4

ms at 15oC. IANBD fluorescence was excited at 470 nm and monitored using a 500 nm

long-pass interference filter from Newport (Irvine, CA). IAANS fluorescence was

excited at 330 nm and monitored using a 510 nm broad spectrum filter from Newport

(Irvine, CA). The experimental conditions for the Ca2+ dissociation and cross-bridge

detachment kinetics are described in the figure legends. Both ATP and ADP contained

an equimolar concentration of Mg2+ so as to maintain a constant concentration of free

Mg2+. The data were fit using a program (by P. J. King, Applied Photophysics Ltd.) that utilizes the nonlinear Levenberg-Marquardt algorithm. Each Ca2+ dissociation and cross- bridge detachment event represents an average of at least three separate experiments, each averaging at least five traces fit with either a single or double-exponential equation.

2.1.10 Formation of the Reconstituted Thin Filaments

The reconstituted thin filaments were prepared using rabbit skeletal actin, bovine cardiac Tm and reconstituted Tn complex. 4 μM rabbit skeletal actin and 0.57 μM bovine cardiac Tm were added to the appropriate volume of the thin filament buffer

((TFB) 10 mM MOPS, 150 mM KCl, 3 mM Mg2+, 1 mM DTT at pH 7.0). Because the

purified actin was very viscous, the tube containing the actin stock was inverted a couple

times to produce a homogenous solution before the aliquot of actin was removed. After each protein addition the mixture was slowly inverted a couple times, without generating 31 bubbles. The actin and Tm were incubated at room temperature for 20 minutes. 0.3 μM of the appropriate Tn was added to the actin and Tm, slowly inverted to mix and incubated at room temperature an additional 15 minutes. The reconstituted thin filaments were put on ice until ready to use.

2.1.11 Determination of the Myosin S1 Concentration and Detachment Rate from Reconstituted Thin Filaments

To prepare myosin S1, an aliquot from the stock was dialyzed in 10 mM MOPS,

150 mM KCl, 3 mM Mg2+, 1 mM DTT at pH 7.0 overnight. The S1 was recovered by pelleting at 14,000 x g for 5 minutes at 4oC. Spectrophotometer was blanked with dialysis buffer and absorbance read at 280 nm and 400 nm. Myosin S1 concentration was calculated ((A280 – A400)/0.75) /110,000), with an expected concentration of ~ 20 μM.

The regulated thin filaments were prepared and reconstituted as described above.

IAANS fluorescence was excited at 330 nm and monitored using a 510 nm broad spectrum filter from Newport (Irvine, CA). Varying concentrations of rabbit skeletal myosin S1 were added to the reconstituted thin filament and then rapidly mixed with 2 mM ATP to detach the S1 from the regulated actin. The experimental conditions are described in the figure legend.

2.1.12 Statistical Analysis

All data are expressed as mean (± SE). Statistical significance of the data was determined by using 1-way ANOVA followed by a post-hoc least significance difference test using GraphPad Prism 4 (La Jolla, CA). Statistical significance was defined as P <

0.05.

2.2 Results

2.2.1 Measuring Ca2+ Dissociation

We have previously shown that TnCT53C labeled with IAANS reported the Ca2+

binding properties of reconstituted Tn and thin filaments with high fidelity [34].

Although TnCT53C IAANS reported similar Ca2+ dissociation and cross-bridge detachment kinetics from myofibrils (data not shown), the signal to noise ratio was at least half of that reported by TnCT53C labeled with IANBD (TnC ). Figure 3A shows T53C IANBD the apparent rate of Ca2+ dissociation from TnC reconstituted into the Tn complex T53C IANBD (40.8 ± 0.5/s) at 15oC, which was nearly identical to that previously reported by the

IAANS probe [34]. Unlike the fluorescence signal from Tn labeled with IANBD at Cys

84 in TnC [85], the complete fluorescence change was observed between the Ca2+

saturated and Ca2+ free baselines of TnC Tn. Similarly, upon exchange of the Tn T53C IANBD complex containing TnC into rabbit ventricular myofibrils in rigor, the complete T53C IANBD fluorescence change also occurred between the Ca2+ saturated and Ca2+ free baselines,

33 with an apparent rate of Ca2+ dissociation at 25.3 ± 0.7/s at 15oC (Figure 3B). The inset of Figure 3B shows that TnC Tn incorporated into the thin filaments of the T53C IANBD myofibrils and replaced ~ 60% of the endogenous Tn as determined by the Western blot ratio of endogenous to exogenous TnI (data not shown). A similar rate of Ca2+ dissociation was observed from the myofibrils when endogenous TnC was extracted and reconstituted with TnC (21.1 ± 0.9/s; data not shown), however the signal to noise T53C IANBD ratio was approximately half of that observed with Tn exchanged myofibrils. Thus,

2+ TnC was capable of measuring the apparent rate of Ca dissociation from the Tn T53C IANBD complex, which slowed ~ 2-fold upon incorporation into the rigor cardiac myofibrils.

Figure 1. Ca2+ Dissociation from IANBD Labled TnC in Tn and Tn Exchanged Rabbit Rigor Myofibrils. Panel A shows the time course of IANBD fluorescence as Ca2+ was chelated by EGTA and removed from the regulatory binding site of TnC reconstituted into Tn. The Tn (0.3 μM) in Buffer A + 200 μM Ca2+ was T53C o 2+ rapidlyIANBD mixed with an equal volume of Buffer A + 10 mM EGTA at 15 C (Ca off Tn trace, 40.8 ± 0.5/s). The Ca2+ saturated Tn baseline was collected by mixing the Ca2+ saturated Tn with Buffer A + 200 μM Ca2+. The Ca2+ free Tn baseline was acquired by rapidly mixing Tn in Buffer A + 5 mM EGTA against equal volumes of Buffer A + 5 mM EGTA. Panel B shows the time course of the IANBD fluorescence as Ca2+ was removed by EGTA from TnC Tn T53C exchanged rabbit rigor ventricular myofibrils. TnC myofibrilsIANBD in Buffer A + 2+ T53C 200 μM Ca were rapidly mixed with an equal volumeIANBD of the Buffer A + 10 mM EGTA at 15oC (Ca2+ off Myofibrils trace, 25.3 ± 0.7/s). The Ca2+ saturated and Ca2+ free myofibril baselines were acquired from TnC myofibrils under T53C same buffer conditions as described for TnC Tn inIANBD panel A. T53C IANBD

(continued)

Figure 1 continued.

To verify that the observed rate of Ca2+ dissociation occurred from the regulatory

domain of TnC, Ca2+ binding to the N-terminal domain of TnCT53C was abolished by the

D65A mutation [90]. The fluorescence intensity of reconstituted D65A TnC Tn and T53C IANBD Tn exchanged myofibrils was insensitive to changes in [Ca2+] (data not shown). Thus, the observed kinetics of TnC Tn and TnC myofibrils can be attributed to the T53C T53C IANBD IANBD apparent rate of Ca2+ dissociation from the N-terminal, regulatory domain of TnC.

2.2.2 Effect of Myofibril Species of the Rigor Ca2+ Dissociation Rate

It is well-established that isoforms of the thick and thin filament proteins vary

across large and small species of mammals [36, 37, 40, 49]. It is currently unknown

whether these different protein isoforms can affect the rate of Ca2+ dissociation from

TnC. Figure 4 shows the apparent rates of Ca2+ dissociation from ventricular myofibrils

in rigor from small, medium and large sized mammals at 15oC. Compared to the Ca2+

dissociation rate from rabbit myofibrils, the rates from rigor rat (16.7 ± 0.5/s) and dog

(21.6 ± 0.7/s) were moderately, but statistically slower while the rate from failing human

ventricular myofibrils (22.9 ± 0.9/s) was not significantly different. Thus, the rigor Ca2+ dissociation rate from human cardiac TnC in different species of myofibrils was T53C IANBD similar.

Figure 2. Effect of Myofibril Species on the Rate of Ca2+ Dissociation from Tn Exchanged Rigor Myofibrils. TnC Tn was exchanged into rat, dog, and T53C 2+ failing human ventricular myofibrilsIANBD to acquire the apparent rates of Ca dissociation for the respective myofibrils. The rigor Ca2+ dissociation rates for the different species were; rat (16.7 ± 0.5/s), dog (21.6 ± 0.7/s), and failing human ventricular myofibrils (22.9 ± 0.9/s). Buffer conditions were identical to those described for panel B of Figure 1. All traces were acquired at 15oC. The data traces were normalized and staggered for clarity.

2.2.3 Effect of ATP on the Observed Fluorescence Changes of TnC

In a normal relaxing heart, Ca2+ dissociates from TnC while cross-bridges are actively cycling, rather than in rigor. Unfortunately, the fluorescence intensity remained at the Ca2+ saturated baseline and did not change when Ca2+ was removed from the rabbit myofibrils by EGTA that were also simultaneously mixed with 2 mM ATP, to allow cross-bridge cycling (Figure 5A, Ca2+ Saturated Control vs. EGTA + ATP trace). This

phenomenon was also observed when ATP was added to the Ca2+ saturated myofibrils

prior to mixing with EGTA. However, when D65A TnC myofibrils were rapidly T53C IANBD mixed with 2 mM ATP, a rapid decrease in fluorescence intensity was observed at a rate

of 203 ± 11/s at 15oC (Figure 3A, D65A vs. EGTA + ATP trace). This ATP dependent

fluorescence decrease occurred from D65A TnC myofibrils in the absence or T53C IANBD presence of Ca2+ and originated from the rigor baseline fluorescence level (Figure 5A,

D65A Rigor trace). This data suggests that there are at least two processes influencing the fluorescence level of TnC myofibrils during Ca2+ dissociation in the presence of T53C IANBD cycling cross-bridges.

Figure 3. Effect of ATP on Rabbit Ventricular Myofibrils and Reconstituted Thin Filaments + S1. Panel A shows the time course of IANBD fluorescence as Ca2+ was removed from TnC and D65A TnC rabbit myofibrils in the T53C T53C presence of ATP. TnC IANBD rabbit myofibrilsIANBD (Ca2+ Saturated Control vs. EGTA T53C + ATP trace, no observedIANBD rate) and D65A TnC rabbit myofibrils (D65A vs. 2+ EGTA + ATP trace, 203 ± 11/s) in Buffer A + T20053C μM Ca were mixed with an equal volume of the Buffer A + 10 mM EGTAIANBD and 2 mM ATP at 15oC. The D65A TnC Ca2+ free myofibril baseline was acquired by rapidly mixing the T53C D65A myofibrilsIANBD in Buffer A + 5 mM EGTA against equal volumes of Buffer A + 5 mM EGTA (D65A Rigor). Panel B shows the time course of IANBD fluorescence decay from Ca2+ free TnC rabbit myofibrils by mixing with T53C ATP in the presence or absence of ADP.IANBD TnC myofibrils in Buffer A + 5 T53C mM EGTA ± 2 mM ADP were rapidly mixed withIANBD equal volumes of the Buffer A + 5 mM EGTA + 2 mM ATP (Ca2+ Free Myofibrils vs. ATP trace, 163 ± 8/s, and Ca2+ Free Myofibrils + ADP vs. ATP trace, 6.9 ± 0.6/s, respectively) . To examine the effect of ATP on the Ca2+ saturated state, TnC myofibrils in 2+ T53C Buffer A + 200 μM Ca were mixed with an equal volume IANBDof Buffer A + 200 μM Ca2+ + 2 mM ATP (Ca2+ Saturated vs. Ca2+ + ATP trace, no observed rate). The Ca2+ free baseline (Ca2+ Free Control trace) from Figure 1B was included as a reference point for the fluorescence decay induced by ATP. Panel C shows the time course of IAANS fluorescence decay from Ca2+ free reconstituted thin filaments. (continued) 40

Figure 3 continued.

containing TnC Tn upon the removal of myosin-S1. TnC reconstituted T53C T53C thin filaments withIAANS a myosin S1 to actin subunit ratio of 1:7, 2:7IAANS and 4:7 in Buffer A (without Tween20) + 5 mM EGTA were rapidly mixed with equal volumes of the Buffer A (without Tween20) + 5 mM EGTA + 2 mM ATP. The rates of S1 detachment reconstituted at the ratios of 1:7, 2:7, and 4:7 concentrations were ~ 300/s, 268 ± 11/s, and 265 ± 10/s, respectively. All data were collected at 15oC.

(continued)

Figure 3 continued.

2.2.4 Sensing Cross-bridge Detachment Through TnC

Unlike D65A TnC myofibrils, the fluorescence intensity did not change when T53C IANBD Ca2+ saturated TnC myofibrils were mixed with 2 mM ATP (Figure 5B, Ca2+ T53C IANBD Saturated vs. Ca2+ + ATP). However, the addition of 2 mM ATP to Ca2+ free TnC T53C IANBD myofibrils decreased the fluorescence at a rate of 163 ± 8/s at 15oC (Ca2+ Free Control vs.

ATP trace). Similar to D65A TnC myofibrils, the ATP dependent fluorescence T53C IANBD decrease originated from the Ca2+ free rigor baseline (Ca2+ Free Control trace). If this

ATP dependent decrease in fluorescence was related to cross-bridge detachment, then the

addition of ADP to the myofibrils would be expected to slow the rate [45]. As expected,

the addition of 2 mM ADP to the Ca2+ free TnC myofibrils substantially slowed the T53C IANBD rate of the ATP dependent fluorescence decrease to 6.9 ± 0.6/s (Ca2+ Free Control + ADP vs. ATP trace). 2 mM ADP also slowed the ATP dependent fluorescence decrease from

D65A TnC myofibrils to 6.4 ± 0.2/s (data not shown). Thus, TnC T53C T53C IANBD IANBD fluorescence was sensitive to both Ca2+ binding and cross-bridge dissociation, each of

which decreased IANBD fluorescence.

Although changes in Trp fluorescence, light scatter and various fluorescent nucleotides have been utilized to more directly observe cross-bridge detachment in other systems [45, 84, 91], none of these approaches were suitable with the ventricular myofibrils (data not shown). In order to further test the idea that the fluorescent TnC was sensitive to cross-bridge detachment, reconstituted thin filaments bound by myosin S1 were utilized. A fluorescent skeletal Tn was previously shown to be sensitive to myosin

S1 binding and dissociation in reconstituted thin filaments [92]. Similarly, Figure 5C

shows the decrease in TnC fluorescence as increasing amounts of myosin S1 were T53C IAANS detached from the reconstituted thin filaments by 2 mM ATP. At a myosin S1 to actin

subunit ratio of 1:7, the fluorescence decreased at a rate of ~ 300/s. This rate, although

substantially slower than the true rate of myosin S1 detachment from nucleotide free

reconstituted thin filaments, was similar to that observed by a fluorescent skeletal Tn [92]

and that observed in the ventricular myofibrils of this study. Increasing the myosin S1 to

actin ratio to 2:7 and then 4:7 had little effect on the apparent detachment rate (268 ± 11/s

and 265 ± 10/s, respectively), but linearly increased the amplitude of the signal. Similar

to the ventricular myofibrils, in the presence of Ca2+, the fluorescence intensity of

TnC on the reconstituted thin filament was not sensitive to myosin S1 detachment T53C IAANS (data not shown). Thus, it would appear that the fluorescent TnC in the ventricular

myofibrils and reconstituted thin filament are reporting similar ATP dependent decreases

in fluorescence that are related to myosin detachment. Unfortunately, the addition of

ADP to the reconstituted thin filaments caused myosin S1 detachment, complicating any further studies with ATP.

To further characterize the influence of ATP on the rabbit ventricular myofibrils,

Figure 6 shows the dependency of the concentration of ATP and ADP on the apparent rate of cross-bridge detachment from the Ca2+ free TnC myofibrils at 15oC. When T53C IANBD the myofibrils were mixed with 5, 10, 50, and 100 µM ATP (concentration before

mixing), the apparent rate of cross-bridge detachment occurred at 0.51 ± 0.02/s, 1.1±

0.2/s, 5.5 ± 0.6/s, and 16 ± 2/s, respectively (Figure 6A). Over a longer time, there was an additional slow decrease in fluorescence that occurred at ~ 0.02/s for all

concentrations of ATP (Figure 6B). Presumably as the ATP was depleted, the

fluorescence slowly increased at ~ 0.03/s back towards the Ca2+ free rigor baseline as the

cross-bridges went back into rigor (Figure 4B). Increasing amounts of ATP delayed the

reverse in fluorescence and rebinding of rigor cross-bridges (Figure 6B). Figure 4C

shows the [ATP] dependence on the rate of the initial fast phase of cross-bridge

detachment from Ca2+ free TnC myofibrils. Previously, it was shown that increasing T53C IANBD concentrations of ATP linearly accelerated the rate of rabbit cardiac myosin-S1

detachment from unregulated actin with an apparent second order rate constant of 2 x 106

M-1 s-1 at 15oC [84]. The apparent rate of cross-bridge detachment in Ca2+ free TnC T53C IANBD myofibrils increased linearly at an apparent second order rate constant of 2.8 x 105 ± 0.01

M-1 s-1 for concentrations of ATP between 0 and 250 μM (Figure 4C). At higher

concentrations of ATP, the cross-bridge detachment rate hyperbolically plateaued at ~

330/s (Figure 6C). In order to show that these ATP dependent fluorescent changes were

not due to myofibril shortening, the insets of Figure 6C show the transmitted light

microscopy images of Ca2+ free rabbit myofibrils taken from the stopped-flow after

mixing with 50 μM ATP (bottom right) and 2 mM ATP (top left, enlarged myofibril).

Thus, similar to skeletal myofibrils [69], the addition of ATP to Ca2+ free ventricular

myofibrils did not cause shortening of the myofibrils.

Figure 4. Dose-dependency of ATP and ADP on the Rate of Cross-bridge Detachment. Panel A shows the effect of increasing [ATP] on the apparent rate (fast phase) of cross-bridge detachment from Ca2+ free TnC rabbit T53C myofibrils. TnC rabbit myofibrils in Buffer A + 5 mMIANBD EGTA were mixed T53C against equal volumesIANBD of Buffer A + 5 (0.51 ± 0.02/s), 10 (1.1 ± 0.2/s), 50 (5.5 ± 0.6/s) or 100 (16 ± 2/s) μM ATP (traces shown top to bottom respectively). Panel B shows the tri-phasic fluorescence change that occurs upon ATP addition to the Ca2+ free TnC rabbit myofibrils with increasing [ATP] from 5 to 100 μM T53C 2+ over an extendedIANBD period of time. The Ca free baseline over these long times displayed a linear decrease in fluorescence that was due to either the bleaching of the signal or myofibril settling, which was subtracted from the experimental traces. Buffer conditions were identical to those described for panel A. Panel C shows the dose-dependent effect that increasing ATP had on the apparent fast rate of cross-bridge detachment from Ca2+ free TnC rabbit myofibrils. The T53C bottom right inset shows the transmitted light microscopyIANBD image of the myofibrils after being mixed with 50 μM ATP in the stopped flow. The top left inset is an enlarged image of a myofibril after being mixed with 2 mM ATP in the stopped flow. Panel D shows the effect that increasing [ADP] had on the apparent fast rate of cross-bridge detachment of Ca2+ free TnC rabbit myofibrils. T53C 2+ Increasing concentrations of ADP (0 to 2000 μM)IANBD were added to the Ca free TnC myofibrils in Buffer A + 5 mM EGTA and rapidly mixed against T53C o BufferIANBD A + 5 mM EGTA + 2 mM ATP at 15 C. Both the concentrations of ATP and ADP shown in panels C and D were the final concentration after mixing in the stopped-flow. (continued) 46

Figure 4 continued.

(continued)

Figure 4 continued.

(continued) 48

Figure 4 continued.

In muscle, ATP cannot dissociate myosin from actin until ADP dissociates from

the acto-myosin complex (for review [33] and [81]). As previously shown, the ADP dissociation rate limits rabbit cardiac myosin-S1 detachment from actin at ~ 115/s at

o 2+ 15 C [84]. Increasing concentrations of ADP in Ca free TnC myofibrils T53C IANBD hyperbolically decreased the apparent rate of cross-bridge detachment when the

myofibrils were mixed with 2 mM ATP (Figure 6D). The rate of cross-bridge

detachment began to plateau at ~ 25 µM ADP, signifying an ADP dissociation rate of 13

± 1/s with an apparent affinity of 1.3 ± 0.1 µM. Further increases in [ADP] resulted in a

mild linear decrease in the apparent rate of cross-bridge dissociation. At these high

[ADP], increasing the [ATP] did not accelerate the apparent rate of cross-bridge 49

detachment beyond the plateau rate of ~ 13/s (data not shown). Although increasing the

concentration of Pi can increase the rate of relaxation [82], increasing the [Pi] up to 20

mM in the myofibrils had no effect on the rate of rigor Ca2+ dissociation or the rate of cross-bridge detachment with or without ADP (data not shown).

2.2.5 Species Dependence on the Cross-Bridge Detachment Rate

It is well-established that larger mammals have a higher percentage of the slow β- myosin, whereas smaller mammals have a greater percentage of the fast α-myosin [93,

94]. Thus, the rate of cross-bridge detachment in myofibrils from large and small mammals would be expected to differ [45, 84]. In the absence of ADP, only the failing human myofibrils (41 ± 4/s) had a rate of cross-bridge detachment (induced by 2 mM

ATP) that was significantly slower than that of the rat (159 ± 6/s), dog (125 ± 17/s), and rabbit myofibrils (163 ± 8/s, Figure 7A). However, compared to the rabbit (6.9 ± 0.6/s), the rate of cross-bridge detachment in the presence of 2 mM ADP was significantly faster for rat myofibrils (42 ± 3/s) and slower for the failing human myofibrils (2.3 ± 0.3/s), but similar to dog myofibrils (6.3 ± 0.6/s) (Figure 7B). This data supports the hypothesis that

TnC fluorescence is sensitive to cross-bridge detachment in ventricular myofibrils T53C IANBD and that different myosin isoforms have different rates of cross-bridge detachment that

appear rate-limited by ADP [95].

Figure 5. Cross-bridge Detachment Rates from Varying Species of Myofibrils. Panel A displays the rates of cross-bridge detachment in the absence of ADP at 15oC from failing human (41 ± 4/s), dog (125 ± 17/s), rabbit (163 ± 8/s), and rat ventricular myofibrils (159 ± 6/s) that were exchanged with TnC Tn. The T53C TnC myofibrils in Buffer A + 5 mM EGTA were rapidly mixedIANBD with equal T53C volumesIANBD of the Buffer A + 5 mM EGTA + 2 mM ATP. Panel B displays the rates of cross-bridge detachment in the presence of 2 mM ADP from failing human (2.3 ± 0.3/s), dog (6.3 ± 0.6/s), rabbit (6.9 ± 0.6/s), and rat ventricular myofibrils (42 ± 3/s) that were exchanged with TnC Tn at 15oC. TnC myofibrils in T53C T53C Buffer A + 5 mM EGTA + 2 mM ADPIANBD were rapidly mixedIANBD with equal volumes of the Buffer A + 5 mM EGTA + 2 mM ATP.

(continued)

Figure 5 continued.

2.2.6 Recovery of the Ca2+ Dissociation Signal with Slowed Cross-Bridge

Detachment

In the presence of rapidly detaching cross-bridges, the Ca2+ dissociation event

from TnC myofibrils could not be observed (Figure 5A, Control vs. EGTA + ATP T53C IANBD 2+ trace). However, when Ca was rapidly chelated from TnC rabbit myofibrils with T53C IANBD slowly detaching cross-bridges due to 2 mM ADP, a biphasic fluorescence signal was

observed (Figure 8A, Myofibrils + ADP vs. EGTA + ATP trace). The signal originated

from the Ca2+ saturated baseline and then decayed back to the Ca2+ saturated baseline.

Under these conditions, the Ca2+ dissociation event (increase in fluorescence) and cross-

bridge detachment event (decrease in fluorescence) were both observed. A mono-phasic

increase in fluorescence (Figure 6B, Subtraction trace) resulted from subtracting the

cross-bridge detachment event (Figure 6A, Ca2+ free myofibrils + ADP vs. ATP trace)

from the biphasic fluorescence signal. The rate of this fluorescence increase (~ 25/s) was

nearly identical to that of rigor Ca2+ dissociation. A biphasic fluorescence signal was also observed when Ca2+ was dissociated from TnC rabbit myofibrils while cross- T53C IANBD bridge detachment was slowed by decreasing the temperature (5oC, data not shown).

Similarly, mixing the Ca2+ saturated myofibrils with EGTA and a reduced amount of

ATP (200 µM, before mixing) at 15oC also caused a biphasic signal (data not shown).

Figure 6. Effect of Slowed Cross-bridge Detachment on the Rate of Ca2+ Dissociation from Rabbit Myofibrils. Panel A represents the biphasic kinetic trace for TnC myofibrils in Buffer A + 200 μM Ca2+ + 2 mM ADP when T53C o mixed with anIANBD equal amount of Buffer A + 10 mM EGTA and 2 mM ATP at 15 C (Myofibrils + ADP vs. EGTA + ATP trace, biphasic). The Ca2+ Free Myofibrils + ADP vs. ATP trace from Figure 3, panel B (6.9 ± 0.6/s) was shown for a graphical comparison. Panel B shows the effect that ADP and cross-linking had on the rate of Ca2+ dissociation at 15oC. The Ca2+ Free Myofibrils + ADP vs. ATP trace (panel A) was subtracted from the biphasic Myofibrils + ADP vs. EGTA + ATP trace (panel A) to recover the Ca2+ dissociation signal (Subtraction, ~ 25/s). TnC myofibrils in Buffer A + 200 μM Ca2+ + 2 mM ADP were T53C o rapidly mixedIANBD with an equal volume of the Buffer A + 10 mM EGTA at 15 C (Myofibril + ADP vs. EGTA trace, 24 ± 1/s). Cross-linked TnC myofibrils 2+ T53C in Buffer A + 200 μM Ca were rapidly mixed with an equal volumeIANBD of the Buffer A + 10 mM EGTA + 2 mM ATP (Cross-linked Myofibrils vs. EGTA + ATP trace, 22 ± 1/s).

(continued)

Figure 6 continued.

Additional studies were performed to examine the effect of ADP and cross-

linking on the apparent rates of Ca2+ dissociation and cross-bridge detachment. The addition of 2 mM ADP to Ca2+ saturated TnC rigor rabbit myofibrils did not alter T53C IANBD the apparent rate of Ca2+ dissociation (24 ± 1/s; Figure 6B, Myofibril + ADP vs. EGTA)

compared to that of nucleotide free rigor myofibrils. Although cross-linked myofibrils were no longer sensitive to ATP dependent cross-bridge detachment events (data not shown), the rate of Ca2+ dissociation from cross-linked myofibrils in the absence (21.3 ±

0.2/s; data not shown) or presence of ATP (22 ± 1/s; Figure 6B, Cross-linked Myofibril vs. EGTA + ATP) was also similar to that of rigor myofibrils. Thus, myofibrils with slowed or non-dissociating cross-bridges had similar apparent rates of Ca2+ dissociation.

Furthermore, cross-bridge detachment could be prevented by cross-linking the myofibrils.

2.2.7 Physiological Significance of Ca2+ Dissociation and Cross-Bridge

Detachment

Controversy abounds in the literature regarding the rate-limiting step of cardiac

muscle relaxation. It is commonly thought that the rate of Ca2+ dissociation from TnC is

too rapid and therefore cross-bridge detachment would rate-limit relaxation [19, 41, 85,

96]. Consistent with this hypothesis, Figure 7A shows that the apparent rate of cross-

bridge detachment in the presence of 2 mM ADP was significantly slower than the rate of

Ca2+ dissociation from TnC rabbit myofibrils at temperatures less than 25oC. T53C IANBD However, at a more physiological temperature (35oC), the two rates were nearly identical.

Physiologically, the concentration of cytosolic ADP in the cardiac myocyte is thought to

be ~ 30 µM and can increase to mM levels during ischemia [97]. Thus, the [ADP]

dependence on the apparent rate of cross-bridge detachment in TnC myofibrils was T53C IANBD determined at 35oC. Similar to the results at 15oC, at 35oC (in Figure 7B) ADP

hyperbolically decreased the apparent rate of cross-bridge detachment when the

myofibrils were mixed with 2 mM ATP. The rate of cross-bridge detachment began to

plateau at ~ 100 µM ADP, signifying an ADP dissociation rate of 104 ± 5/s with an

apparent affinity of 4 ± 1µM. At concentrations of ADP that were supra-physiological

there was a more pronounced linear drop-off in the rate of cross-bridge detachment at

35oC (Figure 7B) than was observed at 15oC (Figure 6D). At these supra-physiological 56

[ADP], increasing the [ATP] increased the apparent rate of cross-bridge detachment to that comparable to the plateau rate of ~ 100/s (data not shown). We suggest that this plateau rate is the rate of ADP dissociation (Figure 7A, ‘Maximal Proposed Rate of ADP

Dissociation’), which rate-limits cross-bridge detachment. Thus, the apparent rate of cross-bridge detachment may actually be slightly faster than the rate of Ca2+ dissociation

at 35oC in the presence of ADP and saturating ATP, but not at temperatures below ~

20oC (Figure 7A). The data suggests that there may not be a simple, single rate-limiting step for myofilament inactivation. Furthermore, both the rates of Ca2+ dissociation and

cross-bridge detachment may be slow enough to influence ventricular muscle relaxation.

Figure 7. Effect of Temperature on the Rates of Ca2+ Dissociation and Cross- bridge Detachment as Reported by TnC in Rabbit Myofibrils. Panel A shows the comparison between the rates of rigor Ca2+ dissociation (Δ), cross-bridge detachment + 2 mM ADP (□), and maximal proposed rate of ADP dissociation (○) at increasing temperatures. Rigor Ca2+ dissociation (Δ) was determined by mixing TnC myofibrils in Buffer A + 200 μM Ca2+ with an equal volume of T53C the Buffer AIANBD + 10 mM EGTA. The ‘Cross-Bridge Detachment (No ADP)’ plot (inset, □) represents cross-bridge detachment from ADP free myofibrils. Cross bridge detachment in presence of ADP was determined by mixing TnC T53C myofibrils in Buffer A + 5 mM EGTA + 2 mM ADP (□) with an equal volumeIANBD of the Buffer A + 5 mM EGTA + 2 mM ATP. The shaded area is used to highlight the difference between the maximal proposed rate of ADP dissociation to that with 2 mM ADP. Panel B shows the effect that increasing ADP had on the apparent fast rate of cross-bridge detachment of Ca2+ free TnC rabbit o T53C myofibrils at 35 C. Increasing concentrations of ADP (0 to 2,000IANBD μM) were added to the Ca2+ free TnC myofibrils in Buffer A + 5 mM EGTA and o rapidly mixed against BufferT53 AC + 2 mM ATP at 35 C. The concentrations of ADP shown were the final concentrationIANBD after mixing in the stopped-flow.

(continued)

Figure 7 continued.

2.3 Discussion

2.3.1 Summary of Findings

The fluorescent TnC was able to report Ca2+ dissociation, as well as ATP T53C IANBD dependent events. We hypothesized that the fluorescent TnC was capable of measuring the rate of cross-bridge detachment. This hypothesis was supported by several lines of evidence in regards to the measured rate of cross-bridge detachment: 1) increasing [ATP] hyperbolically increased the apparent rate (Figure 6C); 2) upon ATP depletion the cross- bridges rebound and reversed the fluorescence signal, which took longer for increased concentrations of ATP; 3) in the absence of ADP (Figure 6C), the maximal rate was

similar to the rate of rigor cardiac muscle relaxation induced by ATP [98]; 4) increasing concentrations of ADP hyperbolically decreased the apparent rate (Figure 6D); 5) in the presence of ADP, the rate varied in an manner expected for different species of myofibrils (i.e. fastest in rat and slowest with the failing human, Figure 7B); 6) TnC fluorescence was sensitive to S1 detachment from the reconstituted thin filaments; 7) the addition of ADP slowed the cross-bridge detachment rate in ventricular myofibrils to that comparable to the rate of relaxation for different cardiac muscles at similar temperatures

[98-100]; and 8) cross-linking the myofibrils abolished the rate.

2.3.2 Proposed Mechanism for How TnC is Able to Report Cross-Bridge

Detachment

In the absence of ADP, ATP induced detachment of myosin from reconstituted skeletal and cardiac acto-myosin is extremely fast (>500/s) [32, 45]. This may suggest that cross-bridge detachment observed by TnC is being sensed and limited by T53C IANBD movement of another myofilament protein (tropomyosin (Tm), TnI or TnT) or that the

rate of cross-bridge detachment from the nucleotide free myofibrils is slower than that

measured from reconstituted acto-myosin. The idea that Tm movement may limit the observed rate of cross-bridge detachment is consistent with skeletal thin filament studies where myosin detachment was > 500/s as measured by light scatter but sensed at ~ 200/s by a fluorescent Tm [32]. However, by following a change in Trp fluorescence, ATP also hyperbolically accelerated the rate of cross-bridge detachment in skeletal myofibrils

(in the absence of ADP) [91]. The maximal rate observed in skeletal myofibrils was 60

similar to the maximal rate of cross-bridge detachment observed by TnC in cardiac T53C IANBD myofibrils. Furthermore, the rate of rigor relaxation induced by caged ATP (in the

absence of ADP, [98]) was also similar to the maximal rate of cross-bridge detachment

sensed by TnC . Thus, cross-bridge detachment in the sarcomere may be slower T53C IANBD than that from a reconstituted filament. In any regard, it would appear that the other myofilament proteins (Tm, TnI and TnT) can move on the thin filament at a rate that is significantly faster than either Ca2+ dissociation from TnC or cross-bridge detachment in

the presence of ADP.

The actual mechanism behind the ability of TnC to sense cross-bridge T53C IANBD detachment is unknown. However, there is ample evidence that the binding of myosin to actin in reconstituted thin filaments, myofibrils, and muscle can influence the fluorescence of multiple TnC constructs [85, 101-103]. It has been shown that cross- bridges can alter the orientation of various helices within TnC exchanged into cardiac trabeculae [104]. These changes in TnC structure are not thought to be caused by a direct interaction of myosin with TnC, but through myosin’s ability to alter Tm’s position on actin [33]. Thus, the structure of TnC could be directly altered by the movement of Tm or indirectly through the movement modifying TnC’s interactions with TnI and TnT.

Regardless, TnC was only able to report cross-bridge detachment in the absence of T53C IANBD Ca2+. The simplest explanation for this phenomenon is that Ca2+ bound TnC exists in a

structural conformation that makes the fluorescent probe insensitive to cross-bridge

detachment.

2.3.3 Comparison Between Myofibrils and Previous Systems Used to Measure

Cross-Bridge Detachment

Our data is consistent with the hypothesis that the rate of cross-bridge detachment

is limited by the rate of ADP dissociation. Previously, the rates of ADP dissociation

from different cardiac myosins were inferred from the study of unregulated acto-myosin in solution [84], laser trap experiments [98], and sinusoidal analysis on skinned cardiac muscle [105-107]. The apparent rate of ADP dissociation measured by our fluorescent

TnC is nearly an order of magnitude slower than that measured from acto-myosin in

solution for the different fast and slow myosins [84]. This result may not be surprising

considering that previous studies have shown that myosin S1 and single headed myosin

have different properties than even heavy meromyosin [108]. In agreement with our

results, the apparent rates of ADP dissociation measured from the rabbit, rat, and human

myofibrils were similar to the calculated rates of ADP dissociation measured by the laser

trap assay (rabbit and rat) [95, 98] and from the sinusoidal length perturbation analysis of

cross-bridge detachment (human, rabbit and rat) [105-107].

There is evidence that the rate of ADP dissociation from various non-muscle and smooth muscle myosins is load-dependent [109-112]. This phenomenon has recently been suggested to be true for cardiac myosin [3]. It may be that the cross-bridges in the myofibrils are under some sort of load that would affect the rate of ADP dissociation.

Consistent with this idea, it is striking that the apparent rate of ADP dissociation from the myofibrils from different species in this study was similar to the reported rates of cardiac muscle relaxation from the respective species Again, this data would suggest that in more

complex physiological systems the rate of ADP dissociation is slowed and that cross-

bridge detachment, rate-limits relaxation at sub-physiological temperatures.

Previous data from skinned skeletal muscle and cardiac myofibrils suggest that

the rate of Ca2+ dissociation from TnC can influence the duration and rate of relaxation

[67, 113]. This was primarily determined by utilizing TnC mutants that possessed slower

rates of Ca2+ dissociation. To date, the rate of Ca2+ dissociation from TnC in intact

muscle has not been measured. Current evidence suggests that the rate of Ca2+

dissociation from TnC may not be constant and is dependent upon strongly bound cross-

bridges. For instance, the rate of Ca2+ dissociation is slowed nearly an order of

magnitude upon myosin-S1 binding to the thin filament [34]. Consistent with these

studies, extra bursts of Ca2+ were released from the myofilament when myosin was

forced to rapidly detach from skinned and intact cardiac muscle [114, 115]. It is believed that the rapid detachment of myosin from actin weakened the affinity of TnC for Ca2+

and accelerated the rate of Ca2+ dissociation. Interestingly, the rate of Ca2+ dissociation

from TnC in guinea pig ventricular myofibrils with detached cross-bridges was nearly an

order of magnitude faster than the rate of relaxation [85]. This study concluded that the rate of Ca2+ dissociation from TnC could not rate-limit relaxation since it was substantially faster than that of relaxation. However, the initiation of relaxation occurs in the presence of strongly bound cross-bridges. Thus, we set out to measure the rate of Ca2+

dissociation from TnC in Tn exchanged ventricular myofibrils under conditions of

strongly-bound cross-bridges.

2+ In rigor, the rate of Ca dissociation from TnC myofibrils was slightly T53C IANBD influenced by the species of myofibrils (Figure 2). The addition of ADP to the rigor

myofibrils had little effect on the rate of Ca2+ dissociation from TnC (Figure 6B). In the

presence of ATP, the Ca2+ dissociation signal could not be observed unless ADP was

present, the temperature was decreased, a minimal amount of ATP was used, or the

myofibrils were cross-linked. This data suggests that the rates of Ca2+ dissociation from

TnC in myofibrils with different strongly-bound cross-bridge states are similar.

However, even with a slower rate of cross-bridge detachment, the rate of Ca2+

dissociation appears too fast to rate-limit myofilament inactivation and relaxation at sub-

physiological temperatures.

2.3.4 Physiological Significance of the Myofibril Biochemical System

It has been suggested that the myofilament rather than Ca2+ dynamics rate-limits relaxation at physiological temperature [43]. It is clear that cross-bridge dynamics are highly temperature sensitive with a Q10 as high as 5 [116]. On the other hand, the rate of

Ca2+ dissociation from cardiac TnC has been shown to have negligible temperature

sensitivity [30]. Consistent with these studies, the rate of cross-bridge detachment was

substantially more affected by temperature than the rate of Ca2+ dissociation from TnC in

the cardiac myofibrils. Interestingly, at physiological temperature, the rate of Ca2+

dissociation and cross-bridge detachment in the ventricular myofibrils were equivalent at

1 mM ADP (after mixing). However, at more physiological concentrations of ADP, the

rate of cross-bridge detachment (~ 100/s) was actually faster than the rate of Ca2+

dissociation (~ 60/s). Thus, under physiological conditions the rate of Ca2+ dissociation may actually rate-limit myofilament inactivation. The maximal proposed rate of ADP dissociation (Figure 7A) may be an overestimation of the rate since the myofibrils may be an unloaded system. Thus, in cardiac muscle the two proposed rate-limiting mechanisms for myofilament inactivation may actually be kinetically tuned with one another. This would suggest that slowing one of these two rates would slow relaxation, but in order to accelerate relaxation both rates would have to be increased. Therefore, therapeutic strategies designed to increase the rate of relaxation may need to target both the thin and thick filament.

2.4 Limitations and Pitfalls

Although we provided extensive experimental data to show that we were able to

measure cross-bridge detachment by monitoring the fluorescence change in TnC, we

were unable to directly measure cross-bridge detachment in the myofibrils. We

attempted to use multiple techniques to directly observe cross-bridge detachment that

included light scatter, the fluorescent nucleotides mant-ADP, mant-ATP, deacATP and deacADP but to no avail. The light scatter produced similar rates of cross-bridge detachment ± ADP but the signal to noise ratio was very poor. Similar to light scatter, the mant-ADP and mant-ATP fluorescent nucleotides had a low signal to noise ratio and numerous additional kinetics unrelated to cross-bridge detachment. For instance, the baseline shots for Ca2+ dissociation produced multiple rates of fluorescence change at a 65

time when cross-bridges were not being detached. Finally, the deacADP and deacATP probes provided a great signal to noise ratio that was better than any of the previous fluorescent probes. The problem was that the fluorescent signal had at least 4 different

rates, which made it impossible to determine which one was related to cross-bridge

detachment. The deacADP and ATP fluorescent nucleotides show promise for measuring the cross-bridge detachment rate in myofibrils but future experiments will need to

decipher the various rates to determine which rate is related to cross-bridge detachment.

Furthermore, the different rates may be due to different steps of the cross-bridge cycle.

Thus, the myofibril biochemical system along with the deac fluorescent nucleotides may

be used to verify the various steps of the cross-bridge cycle in relation to the binding and dissociation of ATP and ADP to and from myosin, respectively.

Chapter 3: Characterization of the Myofibril Ca2+ Binding Properties using Modified TnC’s

3.1 General Introduction

There are a large number of myofilament related cardiomyopathies that appear to

strongly correlate with an altered myofilament Ca2+ sensitivity [7, 117]. DCMs typically

present with a decreased Ca2+ sensitivity and HCM and RCMs have an increased Ca2+

sensitivity. As stated before, TnC is the Ca2+ binding sensor and is part of the myofilament system that determines the overall myofilament Ca2+ sensitivity. In order for the myofilament to relax, Ca2+ must dissociate from TnC to inactivate the thin filament and cross-bridges must detach from actin to alleviate the force [59]. In the previous chapter we showed that we were able to measure both the rate of Ca2+

dissociation from TnC and cross-bridge detachment in the myofibrils. The rate of Ca2+

dissociation from TnC was on par to be a rate-limiting step of myofilament inactivation at

physiological temperature. Thus, in the diseased state, an alteration in the Ca2+ binding of

TnC may directly contribute to altered relaxation of the heart.

Work was previously done in our lab to develop TnCs that had altered Ca2+

binding properties to test the hypothesis that the rates of Ca2+ dissociation from TnC can

be modified [34, 35, 66]. Furthermore, these TnCs were designed to test the hypothesis

that an alteration in the Ca2+ sensitivity of TnCs may alter the normal physiological

processes of the heart and provide a more direct link between impaired Ca2+ binding

properties of TnC to the overall performance of the heart. Our lab has previously 67

designed cardiac TnCs that have a wide range of Ca2+ binding properties that have an

increased or decreased Ca2+ sensitivity, as well as modified Ca2+ association and

dissociation rates [34, 35]. The modified TnCs have been studied in the Tn complex,

reconstituted thin filaments +/- S1, and even studies in skinned trabeculae [67]. Each system was able to contribute to the general characterization of the TnC’s Ca2+ binding

properties, but the Ca2+ dissociation kinetics were not able to be measured from the

skinned trabeculae due to their size and incompatibility with the stopped-flow apparatus.

It became clear during these studies that the complexity of the system and the influence

of the additional myofilament proteins had a large influence on the Ca2+ binding properties of TnC. Although the Ca2+ dissociation rates were measured from Tn and

reconstituted thin filament systems, the rates have not been measured from within the

confines of a more physiologically relevant biochemical system.

To overcome this pitfall we used myofibrils to measure the Ca2+ binding

properties of the modified TnCs from within the cardiac myofilament. As mentioned

before, myofibrils provide a unique biochemical system in which both Ca2+ dissociation

and cross-bridge detachment can be measured from the regulated sarcomere, which

contain the myofilament proteins at a physiologically relevant geometry and

stoichiometry (Chapter 2). Myofibrils were used to measure the Ca2+ dissociation rates

and steady-state Ca2+ sensitivities from the increased and decreased Ca2+ sensitivity TnCs

(L48Q TnC and D73N TnC, respectively). This allowed us to directly show that the

modified Ca2+ binding properties of TnC observed in the Tn complex and reconstituted

thin filaments remained altered in the presence of additional myofilament proteins and at

physiological temperature. If the Ca2+ binding properties of TnC were able to be

modified within the myofibril system it would give confidence that these properties may

lead to altered heart function in-vivo, as will be discussed in the next chapter. In

addition, we wanted to determine whether modified Ca2+ binding affected the cross-

bridge detachment kinetics and vice versa.

The Ca2+ binding properties of D73N and L48Q TnC remained altered as measured from within the myofibrils. D73N TnC had decreased steady-state Ca2+

sensitivity and increased rate of Ca2+ dissociation and also reported an increase in the rate

of cross-bridge detachment in the presence of ADP. Unlike D73N TnC, L48Q TnC had

increased steady-state Ca2+ sensitivity and a decreased rate of Ca2+ dissociation, with

cross-bridge detachment rate similar to that reported by Control TnC in the previous

chapter. In addition, L48Q TnC was able to report the rate of Ca2+ dissociation of the un-

exchanged, endogenous TnC in the rabbit myofibrils. This TnC was used as a tool to

provide insight into additional protein-protein interactions that may occur within the

myofilament of the intact heart. Thus, by examining how TnC affects the Ca2+ binding

properties in increasingly complex biochemical systems, it enhances our understanding

for how the Ca2+ sensitivity of the myofilament may alter myocardial relaxation. These

studies have also given us confidence to test the role of impaired TnC Ca2+ binding in the animal model.

3.2 Methods

The methods describing the purification and labeling of proteins, Tn reconstitution,

reconstitution of the thin filaments, handling of the myofibrils, and how the kinetics were

measured in the stopped flow were described in detail in Chapter 2 Methods.

3.2.1 Determining the Steady-State Ca2+ Binding Affinities

The steady-state fluorescence measurements were performed using a Perkin-

Elmer LS55 spectrofluorimeter at 15oC. IANBD was excited at 470 nm and was

monitored at 522 nm as μl amounts of CaCl2 were added to the 2 ml total volume of

rabbit myofibrils with constant stirring. The myofibrils were prepared from 1 ml of

filtered Tn exchanged myofibrils in Buffer A composed of 10 mM MOPS, 150 mM KCl,

3mM MgCl2, 1 mM DTT, 0.02% Tween 20 (pH 7.0) and then mixed with 1 ml of

titration buffer (TB), composed of 390 mM MOPS, 150 mM KCl, 4 mM EGTA, 3mM

2+ MgCl2, 1 mM DTT, 0.02% Tween 20 (pH 7.0). The free Ca concentration was

calculated using EGCA02 developed by Robertson and Potter[118]. The Ca2+

sensitivities for the Tn exchanged myofibrils were reported as Kd (pCa50), representing a

mean of three separate titrations ± standard error (S.E.). The data were fit with a logistic

sigmoid function as previously described[71].

3.2.3 Stopped-Flow Titration

The kinetic values were measured using an Applied Photophysics Ltd.

(Leatherhead, U.K.) model SX.18 MV stopped-flow instrument with a dead time of 1.4

ms at 15oC. IANBD fluorescence was excited at 470 nm and monitored using a 500 nm long-pass interference filter from Newport (Irvine, CA). The myofibrils were prepared in

200 mM MOPS, 2 mM EGTA, 150 mM KCl, 3mM MgCl2, 1 mM DTT, 0.02% Tween

20 (pH 7.0). This buffer had the same composition as that used for the steady-state Ca2+

titration. Increasing amounts of CaCl2 were added to the prepared myofibrils as calculated using EGCA02 developed by Robertson and Potter[118] and then mixed against 10 mM EGTA to remove the Ca2+ from TnC. The data were fit using a program

(by P. J. King, Applied Photophysics Ltd.) that utilizes the nonlinear Levenberg-

Marquardt algorithm. Each Ca2+ dissociation trace was an average of at least 10 traces

and fit with either a single or double-exponential equation. For each trace, the rate of

Ca2+ dissociation and the respective amplitude of fluorescence change was recorded for both single or double exponential curves.

3.2.4 Statistical Analysis

0.05.

3.3 Results

3.3.1 Steady-State Ca2+Binding Affinities

The steady-state Ca2+ binding properties of the L48Q TnC and D73N T53C IAANS TnC were previously measured in the reconstituted thin filament system and were T53C IAANS shown to have altered Ca2+ sensitivities, as compared to Control TnC [35, 70]. T53C IAANS Instead of using the IAANS (TnC ) fluorescent probe, IANBD (TnC ) was used T53C T53C IAANS IANBD instead as it had a greater signal-to-noise ratio upon incorporation into the myofibrils.

L48Q TnC and D73N TnC were reconstituted with hcTnI and hcTnT to form T53C T53C IANBD IANBD the Tn complex and then exchanged into rabbit ventricular myofibrils. Compared to

Control TnC myofibrils (pCa50 = 6.18 ± 0.08 with a hill coefficient (nH) 1.4 ± 0.2), T53C IANBD 2+ L48Q TnC myofibrils had an increased Ca affinity (pCa50 = 6.566 ± 0.004 and nH T53C IANBD 2+ = 0.94 ± 0.02), while D73N TnC myofibrils had a decreased Ca affinity (pCa50 = T53C IANBD 5.69 ± 0.06 with a nH of 1.4 ± 0.2), as seen in Figure 8. In addition, there was no change

in fluorescence observed with increasing concentrations of Ca2+ for myofibrils exchanged

with D65A TnC Tn. The D65A mutation in TnC knocks out the ability of the N- T53C IANBD terminal, regulatory domain of TnC to bind Ca2+. Thus, the specifically modified TnCs

(L48Q and D73N TnC) were able to maintain their altered Ca2+ binding properties upon

incorporation into rabbit myofibrils and the fluorescence change was attributed to the binding of Ca2+ to the N-terminal domain of TnC.

Figure 8. Steady-State Ca2+ Sensitivities of Tn Exchanged Myofibrils. This panel shows the steady-state Ca2+ sensitivities measured from Control TnC (Control), D73N TnC (D73N), L48Q TnC (L48Q), and T53C T53C T53C D65AIANBD TnC (D65A) Tn exchangedIANBD myofibrils. The IANBDIANBD fluorescence T53C change of theIANBD Tn exchanged myofibrils in 1 ml of Buffer A and 1 ml of Titration Buffer were recorded at increasing amounts of Ca2+ (pCa 9-4) at 15oC. L48Q myofibrils had an increased steady-state Ca2+ sensitivity and D73N myofibrils had decreased Ca2+ sensitivity as compared to Control myofibrils. When the Ca2+ binding to the N-terminal domain of TnC was abolished (D65A myofibrils), the Ca2+ dependent fluorescence change was not observed. The decrease in the fluorescence signal with increasing amounts Ca2+ was consistent with the increase in fluorescence that was observed when Ca2+ was dissociated from Control rigor myofibrils.

3.3.2 Ca2+ Dissociation Kinetics of Modified TnCs

In addition to the steady-state properties, myofibrils exchanged with L48Q and

D73N TnC had modulated Ca2+ dissociation kinetics. Figure 9 displays the T53C IANBD apparent rates of Ca2+ dissociation observed from rabbit myofibrils when exchanged with

L48Q TnC and D73N TnC Tn. Consistent with its decreased Ca2+ affinity, T53C T53C IANBD IANBD D73N TnC myofibrils had an increase in the apparent rate of Ca2+ dissociation (43 ± T53C IANBD 4/s), as compared to Control myofibrils (25.3 ± 0.7/s, from Figure 9A) at 15oC. Ca2+

dissociation from L48Q TnC was complicated by the fact that the fluorescence T53C IANBD change was no longer a single exponential increase. When fit with a single exponential,

the apparent rate of Ca2+ dissociation from L48Q TnC myofibrils was ~ 12/s but the T53C IANBD data was not fit well, as represented by the residuals of the single exponential fit (Figure

9B, blue trace). The data was better fit with a double exponential with rates of 17.3 ±

0.5/s and 4.73 ± 0.07/s, with respective amplitudes of 40 and 60% of the total

fluorescence change. The residuals from the double exponential fit were ~ half of those

observed for the single exponential curve and were substantially more flat (Figure 9B, red trace). Similar to Control TnC myofibrils, the fluorescence change that occurred T53C IANBD upon Ca2+ dissociation from both L48Q TnC and D73N TnC myofibrils T53C T53C IANBD IANBD originated at the ‘Ca2+ Saturated Control Myofibrils’ baseline and increased in

fluorescence until it reached the Ca2+ free baseline (data not shown). This would indicate

that the fluorescence change observed upon the removal of Ca2+ from the myofibrils is

the apparent rate of Ca2+ dissociation from TnC, which can be altered by TnC mutations.

Figure 9. Ca2+ Dissociation from Rigor Myofibrils with Modified TnCs. Panel A shows the time course of IANBD fluorescence as Ca2+ was chelated by EGTA and removed from the regulatory binding site of Control, D73N, or L48Q TnC T53C after Tn exchange into rabbit myofibrils. The myofibrils in Buffer A + 200IANBD μM Ca2+ were rapidly mixed with an equal volume of Buffer A + 10 mM EGTA at 15oC. The ‘Ca2+ Saturated Control Myofibrils’ baseline was collected by mixing the Ca2+ saturated Control Myofibrils with Buffer A + 200 μM Ca2+. The data points for Control and D73N myofibrils were fit with a single exponential and a double exponential for the L48Q myofibrils to calculate the apparent rates of rigor Ca2+ dissociation. Traces have been normalized for ease of comparison. Panel B shows the residuals from the single (blue) and double exponential fits (red) to the L48Q Ca2+ dissociation data points from Panel A. After examining the residuals it became clear that the double exponential better fit the data.

(continued) 75

Figure 9 continued.

3.3.3 Investigating the Origin of the Biphasic L48Q Ca2+ Dissociation Curve

The fast phase of the biphasic change in fluorescence from L48Q TnC T53C IANBD myofibrils upon Ca2+ dissociation had a rate that was similar to the ~ 25/s measured from

Control TnC myofibrils. We hypothesized that the biphasic characteristic of L48Q T53C IANBD TnC myofibrils was due to the ability of the fluorescent TnC to report the rate of T53C IANBD Ca2+ dissociation from the neighboring TnC present within the myofibrils. As stated in

Chapter 2, Tn exchange in the myofibrils was ~ 60%. This would mean that ~ 40 % of

Tn in the L48Q TnC myofibrils was un-exchanged, non-fluorescent endogenous Tn. T53C IANBD To test the hypothesis that L48Q TnC can report or be influenced by a T53C IANBD neighboring TnC, the reconstituted thin filament system was utilized to control the ratios 76

of the Tn species present within the thin filament. Initially, the thin filaments were

reconstituted with L48Q TnC Tn alone, which had a mono-phasic Ca2+ dissociation T53C IAANS rate of 16/s (Figure 10A, 100% L48Q IAANS). When the thin filaments were

reconstituted with 50% L48Q TnC Tn and 50% unlabeled Control Tn a biphasic T53C IAANS Ca2+ dissociation curve appeared (Figure 10A, 50% L48Q IAANS + 50% Control). The

fast and slow rates were ~ 97/s (55% total fluorescence amplitude) and 16/s (45% total

fluorescence amplitude), respectively. When the thin filaments were reconstituted with

25% L48Q TnC Tn and 75% unlabeled Control Tn the rates remained the same but T53C IAANS the slow phase amplitude decreased to 36% of the total fluorescence change (Figure 10

C). The rate of Ca2+ dissociation from Control TnC in the reconstituted thin T53C IAANS filaments alone was monophasic with a rate of ~103/s (Figure 10A, 100% Control

IAANS). When the thin filaments were reconstituted with 50% Control TnC Tn and T53C IAANS 50% unlabeled L48Q Tn there was also a biphasic Ca2+ dissociation curve with the rates

of 110/s (85% of total fluorescence change) and 25/s (15% of total fluorescence change,

Figure 10A, 50% Control IAANS + 50% L48Q). The exponential fits for the previous

reconstituted Ca2+ dissociation curves were normalized to more easily visualize the differences in Figure 10B (the color coding of the traces are described in the figure

legend).

Figure 10. Cross-talk Between Neighboring Tn on the Reconstituted Thin Filament. Panel A shows the time course of IAANS fluorescence decay as Ca2+ was chelated by EGTA and removed from the regulatory binding site of reconstituted thin filaments containing different mixtures of Control and L48Q TnC Tn. The reconstituted thin filaments in Buffer A (without Tween20) + T53C 2+ 200 μMIAANS Ca were rapidly mixed with an equal volume of Buffer A (without Tween20) + 10 mM EGTA at 15oC. The apparent rate of Ca2+ dissociation for reconstituted thin filaments containing 100% L48Q TnC Tn (‘100% L48Q T53C IAANS’) or Control TnC Tn (‘100% Control IAANS’)IAANS were calculated from T53C a single exponential curve.IAANS Reconstituted thin filaments composed of 50% L48Q TnC Tn + 50% unlabeled Control TnC T53C Tn (‘50% L48Q IAANS + 50% T53C Control’)IAANS and 50% Control TnC Tn + 50% unlabeled L48Q TnC T53C Tn T53C (‘50% Control IAANS + 50% L48Q’)IAANS were fit with a double exponential curve. Panel B shows the single and double exponential curve fits for the respective traces from Panel A. The total amplitude of fluorescence changes for the different fits were normalized to better compare the differences in the rates of Ca2+ dissociation. The fits are color coded for the following: 100% L48Q IAANS (Red), 50% L48Q IAANS + 50% Control (Magenta), 50% Control IAANS + 50% L48Q (Gray), and 100% Control IAANS (Black). Panel C shows the % fluorescence amplitude changes for the fast and slow phases of the curves fit in Panel A. In addition to the curves of Panel A, the fast and slow amplitudes for reconstituted thin filaments containing 25% L48Q TnC Tn + 75% unlabeled T53C Control TnC T53C Tn is also shown. IAANS (continued)

Figure 10 continued.

The amplitudes of the fast and slow phases of fluorescence change that occurs

during Ca2+ dissociation from reconstituted thin filaments with varying ratios of either

L48Q TnC Tn and unlabeled Control Tn or Control TnC Tn and unlabeled T53C T53C IAANS IAANS L48Q Tn are displayed in Figure 10C. The amplitudes of the fast and slow phases were

similar to the amount of L48Q TnC Tn and unlabeled Control Tn present within the T53C IAANS reconstituted thin filaments, respectively. It would appear that increasing the amount of

unlabeled Control Tn increased the fast phase of the fluorescence change, supporting the

idea that L48Q TnC is able to sense the neighboring TnC. The biphasic Ca2+ T53C IAANS dissociation curve was also reported by another sensitized TnC exchanged into the

myofibrils and reconstituted in the thin filament. V44Q TnC is another TnC modification

that has been shown to increase the Ca2+ sensitivity of the Tn and reconstituted thin filaments [119]. Ca2+ dissociation from V44Q TnC myofibrils was also biphasic T53C IANBD with rates of ~ 25/s and 8/s, with respective fluorescence amplitudes of 40 and 60% (data

not shown). A similar biphasic phenomenon was observed with V44Q TnC Tn and T53C IAANS unlabeled Control Tn within the reconstituted thin filament but the distinction between

the fast and slow phases was more difficult to determine in this system (data not shown).

However, when the reconstituted thin filaments are composed of Control

TnC Tn and unlabeled L48Q Tn the biphasic fluorescence change was not as T53C IAANS prominent. One explanation for this is that the fluorescent TnC is only able to observe the neighboring TnC when it is bound with Ca2+. For instance, the L48Q TnC has a greater Ca2+ sensitivity and will be bound with Ca2+ longer than the neighboring Control

or endogenous TnC. This allows it to report the rate of Ca2+ dissociation from its

neighboring TnC with a faster dissociation rate. However, when the fluorescent TnC has

a faster Ca2+ dissociation rate than its neighboring unlabeled TnC, its ability to report the

Ca2+ dissociation rate from the neighboring TnC is reduced once Ca2+ dissociates from the

fluorescent TnC.

3.3.4 Stopped-Flow Titration of L48Q Myofibrils

If the fluorescent L48Q TnC is able to sense Ca2+ dissociation from the

neighboring TnC, it would be expected that the Ca2+ sensitivities would also be different.

To further investigate the biphasic fluorescence change during Ca2+ dissociation from

L48Q TnC myofibrils, the apparent Ca2+ dissociation rates and percentage of total T53C IANBD fluorescence change for the slow and fast phases were measured at increasing [Ca2+] in the stopped-flow. The fluorescence change was initially mono-phasic and slow (~4/s) at low Ca2+ and became biphasic at higher Ca2+. Interestingly, what emerged after plotting

the % change of fluorescence for the slow (Blue trace) and fast phases (Green trace) at

their respective pCa levels were two distinct curves. These traces overlapped with the

steady-state Ca2+ titrations of L48Q TnC myofibrils (Red trace) and Control T53C IANBD TnC myofibrils (Black trace), respectively (Figure 11A). The change in T53C IANBD fluorescence of the slow phase for L48Q TnC myofibrils had an apparent calculated T53C IAANS sensitivity (pCa50 = 6.66 ± 0.07 and nH of 1.1 ± 0.1) which was similar to the steady-state

titration of the L48Q TnC myofibrils. In addition, the apparent calculated T53C IANBD sensitivity of the fast phase of L48Q TnC myofibrils (pCa50 = 6.2 ± 0.1 and nH of T53C IANBD 1.01 ± 0.04) was similar to the steady-state titration of Control TnC myofibrils. A T53C 81 IANBD

similar trend was observed for V44Q TnC myofibrils. The slow phase overlapped T53C IANBD with the V44Q steady-state Ca2+ titrations of V44Q TnC myofibrils and the fast T53C IANBD phase overlapped with the steady-state Ca2+ titrations of Control TnC myofibrils T53C IANBD (data not shown). Analogous to the V44Q in the reconstituted thin filament, the error of

these curves was increased in the myofibrils as compared to the L48Q TnC T53C IANBD myofibrils. Thus, the L48Q TnC myofibril Ca2+ dissociation signal is composed of T53C IANBD two components with varying Ca2+ sensitivities that were undetectable under the steady- state conditions where the two phases blended together. This provides additional evidence that L48Q TnC TnC is able to report the Ca2+ dependent activity of its T53C IANBD neighboring TnC.

Figure 11. Stopped Flow Titration of L48Q Myofibrils to Differentiate Fast and Slow Phase Ca2+ Sensitivity. Panel A shows the % change in fluorescence amplitudes for both the slow phase (blue trace) and fast phase (green trace) plotted at their respective pCa and overlaid on the steady-state Ca2+ sensitivity curves (Panel A). The L48Q TnC myofibrils were prepared for the stopped- 2+ T53C flow Ca titration in a buffer that containedIANBD equal amounts of Buffer A and Titration Buffer, which replicated the final buffer conditions during the steady state experiments at 15oC (200 mM MOPS, 2 mM EGTA, 150 mM KCl, 3 mM Mg2+, 1 mM DTT, and 0.02% Tween20 at pH 7.0). The L48Q TnC 2+ T53C myofibrils containing controlled amounts of Ca were rapidly mixedIANBD with excess EGTA (10 mM) in the stopped-flow apparatus, as was done before to acquire the Ca2+ dissociation kinetics.

3.3.5 Effect of Cycling Cross-Bridges on the Ca2+ Dissociation Rate of the

Modified TnCs

In Chapter 2 it was shown that in presence of cycling cross-bridges (addition of

ATP to myofibrils), the Ca2+ dependent increase in fluorescence was no longer observed

for Control TnC myofibrils. Similar to these results, when Ca2+ was dissociated T53C IANBD from D73N TnC myofibrils in the presence of ATP no fluorescence change was T53C IANBD observed (Figure 12A, D73N ATP Ca2+ Dissociation). Unlike Control or D73N

myofibrils, the Ca2+ dependent change in fluorescence for L48Q TnC myofibrils T53C IANBD was still observed in the presence of ATP (Figure 12A, L48Q ATP Ca2+ Dissociation).

The biphasic rigor Ca2+ dissociation rates of ~ 17/s & 5/s from L48Q TnC T53C IANBD myofibrils (Figure 12A, L48Q Rigor Ca2+ Dissociation), increased to 36 ± 1/s and 10.9 ±

0.6/s (Figure 12A, L48Q ATP Ca2+ Dissociation) in the presence of ATP. ATP doubled

the rates for both the slow and fast phase of fluorescence change, where each phase

composed ~ 50% of the total amplitude. A similar phenomenon was observed for V44Q

TnC myofibrils. Ca2+ dissociation from V44Q TnC myofibrils in the presence T53C T53C IANBD IANBD of ATP increased the rates of both phases ~ 2 fold (~ 60/s and 18/s, data not shown) and as stated in previous experiments, the % amplitude of each phase was much more variable. For instance, the slow phase amplitude varied from 25 to 80% of the total fluorescence change. It is unclear why the biphasic characteristic of V44Q TnC T53C IANBD myofibrils was not as consistent as L48Q myofibrils but may be due to the 2-fold

increase in the rates of Ca2+ dissociation over the L48Q myofibrils, which made it harder

to separate out the rates. Interestingly, the increase in the apparent rate of Ca2+

dissociation for both V44Q and L48Q myofibrils in the presence of cycling cross-bridges

was abolished upon the addition of 2 mM ADP to the myofibrils, prior to mixing with

ATP (data not shown). In the presence of ADP both TnCs had Ca2+ dissociation rates

similar to the rigor rates. Thus, when the cross-bridges are strongly bound to actin in the myofibrils they feedback onto TnC to slow the rate of Ca2+ dissociation.

Figure 12. ATP Dependent Changes in the Observed Fluorescence of TnCs with Modified Ca2+ Binding Properties. Panel A shows the time course of IANBD fluorescence as Ca2+ was removed from Control TnC myofibrils (from T53 C chapter 2) D73N TnC myofibrils, and L48Q TnCIANBD rabbit myofibrils in T53C T53C 2+ the presence of ATP. IANBDThe respective myofibrils in BufferIANBD A + 200 μM Ca were mixed with an equal volume of the Buffer A + 10 mM EGTA and 2 mM ATP at 15oC. The ‘L48Q Rigor Ca2+ Dissociation’ curve from Figure 2 was shown as a reference to the increased rate of Ca dissociation from L48Q TnC myofibrils 2+ T53C in the presence of ATP (‘L48Q ATP Ca Dissociation’). IANBD

3.3.6 Cross-Bridge Detachment Reported by Modified TnCs

Previously it was shown that Control TnC was able to report the rate of T53C IANBD cross-bridge detachment in the ventricular myofibrils upon mixing with ATP in the 85

absence of Ca2+. Similar to Control myofibrils, the ATP dependent event of cross-bridge

detachment was also observed by the modified TnCs in the absence of Ca2+. Figure 13A

displays the fluorescence change for Control TnC myofibrils (Blue trace) as T53C IANBD previously shown in Chapter 2 for a graphical comparison to the modified TnCs. When

Ca2+ free D73N TnC myofibrils were mixed against 2 mM ATP, a slightly increased T53C IANBD rate of cross-bridge detachment (193 ± 6/s, Figure 13A, D73N) was observed. In contrast to Control and D73N TnC myofibrils, L48Q TnC myofibrils had a reduced rate T53C T53C IANBD IANBD of detachment (104 ± 7/s, Figure 13A, L48Q). To further support the idea that this

fluorescence change was related to cross-bridge detachment, ADP was added to the

myofibrils to slow the rate of cross-bridge detachment as done in Chapter 2 for Control

myofibrils. The addition of 2 mM ADP to the myofibrils prior to mixing with 2 mM

ATP drastically slowed the rate of cross-bridge detachment for both D73N TnC T53C IANBD myofibrils (9 ± 2/s) and L48Q TnC myofibrils (11 ± 1/s, data not shown). The rate T53C IANBD of detachment in the presence of 2 mM ADP is not believed to be the actual rate of ADP

dissociation due to a competition effect of the excess ADP. The actual rate of cross-

bridge detachment or ADP dissociation rate was calculated from the plateau phase of the

ADP dose response of the Control myofibrils in Chapter 2. Compared to the Control

myofibrils, L48Q TnC myofibrils had a similar plateau rate of cross-bridge T53C IANBD detachment (~ 15/s) whereas D73N TnC myofibrils had an increased rate (~55/s) as T53C IANBD shown in Figure 13B.

In the absence of ADP it would appear that L48Q myofibrils have a decreased

rate of cross-bridge detachment. However, since the rate of cross-bridge detachment

observed by Control TnC in ventricular myofibrils in the absence of ADP is thought T53C IANBD to be limited by the movement of Tm or another thin filament protein (Chapter 2), L48Q

TnC may affect these proteins (movement of Tm across actin), rather than the actual rate of detachment being reduced. The plateau rate of the ADP dose response curve is believed to be the actual rate of ADP dissociation and compared to Control myofibrils, the rate reported by L48Q TnC was similar. Conversely, D73N TnC T53C T53C IANBD IANBD myofibrils had an increased rate of ADP dissociation as determined from the plateau

phase of the ADP dose response curve (Figure 13 B). Since this rate is determined by

cross-bridge detachment and not the mechanism by which cross-bridge detachment is

observed, this shows the desensitized D73N TnC may actually increase the rate of cross-

bridge detachment.

Figure 13. Cross-bridge Detachment as Reported by Modified TnCs. Panel A shows the time course of IANBD fluorescence decay from Ca2+ free Control TnC (‘Control’ (blue trace)), D73N TnC (‘D73N’ (green trace)), and T53C T53C L48QIANBD TnC (‘L48Q’ (red trace)) rabbit myofibrilsIANBD by mixing with ATP at o T53C 15 C. The IANBDATP dependent decrease in fluorescence for Control myofibrils from chapter 2 was shown as a reference for D73N and L48Q myofibrils. The respective TnC myofibrils in Buffer A + 5 mM EGTA were rapidly mixed T53C with equal volumesIANBD of the Buffer A + 5 mM EGTA + 2 mM ATP (Control, D73N, and L48Q). Panel B shows the effect that increasing ADP had on the apparent fast rate of cross-bridge detachment of Ca2+ free Control, D73N, and L48Q TnC rabbit myofibrils at 35oC. Increasing concentrations of ADP (0 T53C 2+ to 2,000 μM)IANBD were added to the Ca free TnC myofibrils in Buffer A + 5 T53C o mM EGTA and rapidly mixed against Buffer AIANBD + 2 mM ATP at 15 C. The concentrations of ADP shown were the final concentration after mixing in the stopped-flow. The Control trace from chapter 2 was shown as a reference.

(continued)

Figure 13 continued.

3.3.7 Temperature Dependence on the Ca2+ Dissociation Rate of Modified TnCs

A goal for creating TnCs with modifications was to determine whether or not the

Ca2+ binding properties could be altered and in future studies, how that affects heart

function. Comparing the L48Q and D73N TnC to Control TnC it was shown that the

Ca2+ dissociation kinetics could be decreased and increased at 15oC, respectively.

However, as shown in Chapter 2, temperature had a significant effect on cross-bridge

detachment, more so than on Ca2+ dissociation from TnC at the more physiological

temperature of 35oC. Thus, it was important to determine if the modified TnCs remained

altered when compared to Control TnC in the myofibrils across a wide range of

temperatures.

The Ca2+ dissociation rates of Control, L48Q and D73N TnC myofibrils T53C IANBD were measured at 5, 15, 25 and 35oC. Table 1 displays the Ca2+ dissociation kinetics and

the effect that increasing temperature had on the rate of rigor Ca2+ dissociation from

rabbit myofibrils. Compared to Control TnC myofibrils, the fast phase of L48Q T53C IANBD TnC myofibrils had Ca2+ dissociation rates similar to Control TnC myofibrils. T53C T53C IANBD IANBD Although the fast rate of Ca2+ dissociation was statistically different than the Control

myofibrils at 5, 15, and 35oC, the values were similar for all intents and purposes. At

o 25 C the two rates were nearly identical. The Q10 for the fast phase of L48Q TnC myofibrils was ~ 2.2 compared to a Q10 of ~ 1.5 for Control TnC myofibrils. The slow phase of Ca2+ dissociation from L48Q TnC myofibrils was slower than the Control TnC

2+ myofibrils at all temperatures and had a Q10 of ~ 2.6. Ca dissociation from D73N

myofibrils at temperatures beyond 15oC were difficult to measure as the signal/noise ratio

drastically increased. This could have been due to the rate being too fast to measure. If

o D73N TnC had a Q10 similar to Control TnC its rate at 35 C would be ~ 100/s. However,

if the Q10 were closer to L48Q TnCs slow phase the extrapolated rate would be closer to ~

350/s. Reducing the temperature to 5oC did not slow the apparent rate of Ca2+

dissociation for D73N TnC myofibrils. Thus, at 35oC, the rate of Ca2+ dissociation T53C IANBD due to the increased Ca2+ sensitivity L48Q TnC remained slower than Control TnC and

may be much faster for the D73N TnC.

Double Mono Exponential Exponential % Fast Rate Rate % Slow Phase Phase Temperature Control D73N L48Q L48Q L48Q L48Q 5oC 19.6 ± 0.5/s 54 ± 2/s # 9.3 ± 0.3/s # 2.5 ± 0.1/s # 26 74 15oC 25.3 ± 0.7/s 43 ± 4/s # 17.3 ± 0.5/s # 4.73 ± 0.07/s # 39 61 25oC 37. ± 3/s * * 37 ± 2/s 12.1 ± 0.2/s # 43 57 35oC 61/s ± 7 * * 95 ± 6/s # 37 ± 2/s # 50 50 Table 1. Temperature Dependence on the Ca2+ Dissociation Rate of Modified TnCs. The apparent rates of Ca2+

92 dissociation from D73N, and L48Q TnC myofibrils were compared to the rates previously measured from Control T2+53 C TnC myofibrils (chapter 2). The CaIANBDdissociation rates were measured in the stopped from at increasing T53C temperaturesIANBD with buffer conditions identical to those of Figure 9. In addition, the % of slow phase amplitude and % of fast phase amplitude change were recorded as well. * * denotes the inability of the Ca2+ dissociation rate to measured due to the lack of fluorescence change. # denotes a statistical significance as compared to Control myofibrils at the same temperature with a P-value < 0.05.

3.4 Discussion

3.4.1 Summary of Findings

The rates of Ca2+ dissociation, as well as the steady-state Ca2+ binding properties

were measured from modified TnCs (D73N, L48Q, and V44Q mutations) after Tn

exchange into ventricular myofibrils. These specific amino acid modifications were able

to modify the Ca2+ dependent changes in the myofilament to increase (D73N TnC) or

decrease (L48Q TnC) the rate at which Ca2+ dissociates from TnC. In agreement with the

Ca2+ dissociation kinetics, the Ca2+ sensitivities decreased and increased, respectively.

Ca2+ dissociation from L48Q TnC was biphasic due to it reporting the rate of Ca2+

dissociation from its non-fluorescent, neighboring TnC in both the reconstituted thin

filament and Tn exchanged myofibrils. Although L48Q TnC had a greater Q10 for the

rate Ca2+ dissociation than Control TnC, the rate remained slower than Control TnC at

o 35 C. The increased Q10 for L48Q TnC was previously observed in the isolated TnC

[66]. The amplitude of the fluorescence change due to the L48Q TnC (slow phase) was proportional to the amount of Tn reconstituted into the thin filaments and exchanged into

the myofibrils. The Ca2+ dissociation rate from the endogenous, unlabeled rabbit TnC, as

reported by L48Q, had a similar Ca2+ dissociation rate and Ca2+ sensitivity to the

fluorescently labeled Control TnC. Unlike Control TnC, the sensitized L48Q and V44Q

TnCs were able to report Ca2+ dissociation in the presence of cycling cross-bridges and the rate was increased ~ 2-3 fold over the rigor rate which was slowed by ADP. In addition to Ca2+ binding, the modified TnCs were also able to report the rate of cross-

bridge detachment, which was rate-limited by ADP dissociation. In the absence of ADP,

L48Q TnC had a slower rate of cross-bridge detachment than either Control or D73N

TnC (which were similar). However, in the presence of ADP, Control and L48Q TnC

had a similar rate of cross-bridge detachment and D73N TnC had an increased rate.

These results show that Ca2+ dissociation from TnC can be directly modified and the TnC

modifications can be used as tools to further probe protein-protein interactions within the

myofilament.

3.4.2 Effect of TnC Modifications on Ca2+ Dissociation.

The original goal of the myofibril studies was to compare the rate of Ca2+

dissociation from rigor myofibrils to those with cycling cross-bridges. Rigor and cycling

cross-bridges have been shown to exert different effects on the Ca2+ dependent changes

in the Tn structure [104]. It was previously shown that the rate of Ca2+ dissociation was

increased in the reconstituted thin filament system upon the removal of myosin from

actin [34]. However, as shown in the previous chapter, Ca2+ dissociation was not able to

be measured from Control myofibrils with cycling cross-bridges. Similar to Control

TnC, there was also no Ca2+ dependent change in fluorescence for D73N myofibrils in the

presence of ATP. In contrast to these two TnCs, a Ca2+ dependent fluorescence signal

was still observed in the presence of ATP for the Ca2+ sensitized L48Q and V44Q

myofibrils. In the presence of cycling or detached cross-bridges the biphasic rate of Ca2+

dissociation was increased ~ 2 fold for both phases of the L48Q myofibrils and ~ 3 fold

for V44Q myofibrils. This finding is in agreement with previous work, in which the rate

of Ca2+ dissociation was increased (~ 10 fold) in the reconstituted thin filaments

containing L48Q TnC, upon the removal S1 [35]. However, the fold-increase in the rate

of Ca2+ dissociation from myofibrils was not as drastic as that reported with the

reconstituted thin filament. This provides further support that Ca2+ dissociation from

TnC is system dependent and is modified by increasing the complexity of the system as well as the ATP dependent state of that system. It is unclear why the increased

sensitivity TnCs were still able to report the Ca2+ dependent fluorescence change in the

presence of ATP. One possible explanation is that the Ca2+ sensitized L48Q TnC is in a

different conformation that allowed the fluorescent probe to still observe Ca2+

dissociation. Thus, a single hydrophobic amino acid change in TnC can change the

structure of Tn and feedback to alter multiple properties of the myofilament and will be

discussed in further detail later.

3.4.3 Mechanism of Cross-talk Between Neighboring Tn

There has been substantial evidence for the ability of cross-bridges to influence the conformational state of TnC as reported by various fluorescence probes attached to

TnC [101, 103, 120-122]. In addition to cross-bridges, we have shown that neighboring

TnCs can feedback upon one another to influence TnCs fluorescence level. The exact mechanism for how the fluorescently labeled TnC is able to observe Ca2+ dissociation

from the unlabeled TnC is not known. It is also unclear if the fluorescent TnC is

observing the neighboring TnC along the length of the actin filament or on the back side

of actin. If it were along the filament the signal might be transmitted through the

movement of Tm and if on the back side, it could be possible that TnT might be involved.

For the latter, TnT would be implicated due to the ability of the linker domain to reach

around to the other side of the the actin filament, allowing the N-terminal domain of TnT

to bind to the opposite Tm [123]. Through Tm, the two Tn subunits are connected and

able to cross-talk and feedback upon each other. It would appear that the fluorescent TnC

was able to observe or report the unlabeled neighboring TnCs activity only when it was

bound with Ca2+. L48Q TnC seemed to report, rather than be influenced by the

neighboring TnC, due to the fluorescent probe reporting two distinct rates. Having the ability to report both the Ca2+ dissociation kinetics of the endogenous TnC as well as the

cross-bridge kinetics, L48Q TnC could be used as a tool for determining myofilament

function in different diseased heart tissue. Through Tn exchange into myofibrils

prepared from tissue from various species, L48Q could provide insight into the

myofilament state of the diseased tissue, and potentially expose targets for improving

heart function.

When Ca2+ dissociation was measured from D73N Tn exchanged myofibrils there

was a mono-exponential increase in fluorescence. However, its rate varied from batch to

batch of Tn exchanged myofibrils more so than either L48Q or Control TnC. This could

have been due to the amount of Tn exchanged into the myofibrils and could have been

more or less influenced by the more Ca2+ sensitive, endogenous Tn. Thus, when there

was greater D73N exchange the Ca2+ dissociation rate would be expected to increase.

When D73N exchange was poor, the endogenous TnC would have exerted a greater influence on the D73N TnC and resulted in a slower rate of Ca2+ dissociation. In this case, the increased sensitivity TnC was able to influence the neighboring D73N TnC to alter its rate. Cross-talk between TnCs within the myofilament may be an intrinsic property of thin filament activation or may be involved in the cooperativity of activation for the entire myofilament. It’s long been known that the strong binding of myosin to actin is able to increase the Ca2+ sensitivity and cooperativity of myofilament activation

[19, 124]. Our work shows that TnC-TnC cross-talk could be another myofilament

communication mechanism similar to that observed with the binding of cross-bridges to

actin to produce more homogenous thin filament activation.

3.4.4 Physiological Significance

There are cardiomyopathies that result from mutations in almost every myofilament protein and altered Ca2+ sensitivity has been implicated in most of them [2,

125-127]. DCM has decreased Ca2+ sensitivity whereas RCM and HCM present with increased Ca2+ sensitivity. In the previous chapter we were investigating the two possible

rate-limiting steps of myofilament inactivation; Ca2+ dissociation from TnC and cross-

bridge detachment. We have shown that specific mutations in TnC can alter the

sensitivity as well as the Ca2+ dissociation rate from rigor myofibrils at 15oC. More

importantly, Ca2+ dissociation from L48Q TnC remained slower than endogenous and

Control TnC at 35oC, similar to the temperature at which the human heart works. Thus,

when TnC has an increased Ca2+ sensitivity it may be able to rate-limit myocardial relaxation and lead to the diseased phenotypes of human RCM and HCM. In studies focused on examining the role of L48Q in muscle relaxation, they found that L48Q TnC delayed relaxation [113]. Our biochemical studies that utilized L48Q TnC determined

that the rate of Ca2+ dissociation was decreased but the rate of cross-bridge detachment in

the presence of ADP was similar to Control myofibrils. These results are also consistent

with the increased Ca2+ sensitivity of force production found in many RCM and HCM

cardiac muscle [7, 48, 125] and shows that direct alteration of TnC was able to

recapitulate these biochemical phenotypes. Thus, the delayed relaxation of L48Q TnC in

muscle may be due to the decreased rate of Ca2+ dissociation from TnC.

Contrary to the increased Ca2+ sensitivity of L48Q TnC is the decreased Ca2+

sensitivity D73N TnC mutation. Although the D73N TnC was unable to give us Ca2+

binding information at physiological temperature, the studies at 15oC provided additional

mechanistic insight into myofilament deactivation. D73N TnC was unable to report the

TnC-TnC cross-talk in the myofibrils but introduced a link between the decreased Ca2+

sensitivity TnC and an increased rate of cross-bridge detachment rate in the presence of

ADP. The decreased Ca2+ dependent activation of D73N TnC may directly feedback on

the decreased amount of thick filament activation. A proposed mechanism may be the

altered interaction between the N-terminal of TnC with the switch peptide of TnI that is

unable to fully mobilize the inhibitory domain of TnI to dissociate from actin during thin

filament activation, thereby keeping Tm less mobile. Thus, with the thin filament unable

to fully activate, the cross-bridges are unable to strongly bind to actin with the same

affinity, which resulted in an increased rate of cross-bridge detachment from D73N Tn

exchanged myofibrils. These findings are consistent with the phenotypic presentation of

reduced Ca2+ sensitivity of force production found in many DCM cardiomyopathies [128-

130]. In conclusion, examining the effects of the TnC mutations on Ca2+ binding and

cross-bridge dissociation kinetics in increasingly complex biochemical systems enhances

understanding of how TnC responds to Ca2+ to regulate muscle function and may expose new targets for therapeutic intervention.

3.4 Limitations of the Myofibrils

The myofibrils are a unique biochemical system in which both the rate of Ca2+ dissociation from TnC and cross-bridge detachment can be measured from a fluorescently labeled TnC. This is a result of the structural change that occurs in TnC upon Ca2+ dissociation from regulatory domain as well as the structural change in TnC as

a result of cross-bridge detachment. There are however a few limitations for using the

myofibrils to measure the rate-limiting steps of myocardial relaxation and inactivation.

In the simpler biochemical systems (isolated TnC and Tn complex), the apparent rate of

Ca2+ dissociation reported by the fluorescent TnC was able to be verified by using an

extrinsic Ca2+ chelating fluorescent probe (Quin-2) [34]. This fluorescence change was a

more direct measurement of the rate of Ca2+ dissociation from TnC and the Tn complex.

Quin-2 was not able to be used in the myofibrils to verify that the fluorescence change

reported by the fluorescent TnC was related to Ca2+ dissociation due to the large amount

of contaminating Ca2+ in the myofibrils and other Ca2+ binding proteins such as actin.

Another limitation of the myofibril system was the inability of the Control and D73N

TnC to report the rate of Ca2+ dissociation in the presence of cycling cross-bridges. This

was due to the fact that the TnC fluorescence change was affected by Ca2+ dissociation and cross-bridge detachment, resulting in neither event being reported. One possible way to circumvent this downfall would be to attach a fluorescent probe at either native Cys residue (C35 or C84), which was previously mutated to a Ser residue to allow us to attach the fluorescent probe exclusively to T53C. Stehle et al. showed that by using these positions to attach the fluorescence label they were able to measure Ca2+ dissociation

from TnC in the myofibrils and also observed a cross-bridge detachment influence on the fluorescence level but did not investigate it further [85]. However, attaching the fluorescent label to these sites influences Ca2+ binding. Another alternative may be

through the use of a different reactive fluorescent dye. One observation made during the

removal of Ca2+ from the various TnC mutations in the myofibrils was that the increased sensitivity mutations had a drastically increased signal-to-noise ratio and may be a reason

for their ability to still observe Ca2+ dissociation in the presence of ATP. One florescent

probe of interest is 5 – Iodoacetamidofluorescein (5-IAF), which can be attached to the reactive thiol in the Cys residue under the same conditions as IAANS and IANBD were attached to TnC. Lastly, the myofibrils are an unloaded system and the actual rates of

Ca2+ dissociation and cross-bridge detachment from intact muscle may be altered in the

100

presence of force. It is currently not technically possible to measure the rate of Ca2+ dissociation from intact muscle under load. Alternatively, the attachment of streptavidin agarose beads (Thermo Scientific, Rockford, IL) to the ends of the myofibrils could create a drag when the myofibrils shorten and may mimic shortening in a loaded system.

To accomplish this, the myofibrils will need to be bound with Biotin-XX phalloidin

(Invitrogen, Carlsbad, CA). The biotin conjugated phalloidin will bind to the filamentous actin and then the biotinylated streptavidin beads can bind to the biotin. Due to the large size of the beads the strepavidin will only bind to the ends of the myofibrils where actin is exposed since accessibility will not be hindered by the additional myofilament proteins.

An ideal biochemical system to study the rate-limiting steps of myofilament inactivation would be a single myofibril under loaded conditions. This could theoretically be accomplished by using a fluorescent microscope that had a PMV attached to the objective that would record the fluorescence change of TnC from a single Tn exchanged myofibril. The single myofibril would also be mounted on both ends to a force transducer to be able to record the amount of force produced during contraction as well as the rate of force decline during myofilament inactivation. The myofibril would need to be in a flow chamber that would allow for the switching of solutions on the millisecond time scale to dissociate Ca2+ from TnC, detach the cross-bridges with ATP or

both at the same time. Thus, this system would allow the study for both Ca2+ dissociation and cross-bridge detachment from a regulated and load bearing myofibril.

101

3.5 Future Directions

To further elucidate the mechanism by which the signal of cross-bridge

detachment is observed by the fluorescent TnC, additional experiments will need to be

performed. It is possible that the mechanism by which TnC was able to sense cross-

bridge detachment was through the movement of Tm feeding back onto the structure of

the Tn complex. Future experiments to test this hypothesis might involve the use of

fluorescent moieties attached to Tm [131] to report the rate at which Tm movement

occurs within the myofibrils during cross-bridge detachment. This may also be

accomplished through the use of a fluorescent TnT or TnI [132, 133]. Miki et. al. showed

that the fluorescence change for both proteins were sensitive to the detachment of myosin-S1from actin in the reconstituted thin filament, similar to our fluorescent TnC.

We have preliminary data that utilized a fluorescent TnI labeled with IANBD at S150C.

The fluorescent TnI was sensitive to both Ca2+ dissociation and cross-bridge detachment

in the myofibrils. It was difficult to determine the actual rates for each event because the

fluorescence change was composed of at least three different phases. Additional work

will need to be done to tease out the actual rates but preliminary results support the

hypothesis that TnI may be involved in the transmission of the cross-bridge detachment

signal to TnC due to TnI reporting a similar rate as that observed by TnC.

It would also be of importance to measure the fluorescence change from these

proteins during the dissociation of Ca2+ from TnC in the reconstituted thin filaments and

102

Tn exchanged myofibrils. Furthermore, incorporation of the fluorescent TnC, TnT, and

TnI with disease mutated TnI and TnT proteins (truncated TnI R192, TnI S166F, TnI

D190H, TnT 203E, TnT ΔK210, TnT R144W) may uncouple the fluorescence change

related to cross-bridge detachment and provide mechanistic insight into the structural

changes that occur within the Tn complex during cross-bridge detachment in the

myofibril system. Preliminary results showed that the HCM related S166F TnI slowed

the rate of rigor Ca2+ dissociation in the rabbit myofibrils, consistent with the diseased

phenotype with a rate similar to the L48Q TnC. In addition, the rate of cross-bridge detachment was studied for TnI S166F and TnI D190H in the myofibrils and both slowed the rate of detachment as observed by IANBD labeled TnC in the myofibrils. Future studies might incorporate the fluorescent TnC into the myofibrils of different transgenic animal models to determine if there are functional differences in myofilament inactivation as related to Ca2+ dissociation and cross-bridge detachment. As briefly mentioned in the previous discussion, Control, D73N and L48Q TnCs could be used as tools to determine the overall state of the myofilament in myofibrils from human tissue samples. One of the reoccurring themes in research is the applicability of the research to the human and how it can be used to improve the characterization, diagnosis, and treatment of human disease. To address these areas of interest, our fluorescently labeled

TnCs could be incorporated into myofibrils prepared from human cardiac tissue samples.

A common procedure performed in the diseased human heart is the insertion of a left ventricle assist device (LVAD). When the device is inserted a sample of tissue is

103 removed from the heart. This tissue is then added to the human tissue sample core at the

Ohio State University, where the tissue is genotyped, and run through a gamut of experiments done to determine what normal properties of cardiac function have been altered or impaired (i.e. phosphorylation status of the contractile proteins). Thus, our study of the Ca2+ dissociation and cross-bridge detachment kinetics through the use the fluorescent TnC will be able to further characterize the diseased state of the patient’s myocardium and could potentially be used to develop a more direct therapeutic approach for treating heart dysfunction. Our experimental approach along with the results from genotyping and additional physiological measurements of the tissue samples may lead to a more personalized approach to translational medicine.

104

Chapter 4: Modulating heart function through the use of TnC’s with modified Ca2+ binding properties

4.1.1 Approaches to Studying TnC in the Heart.

In the previous chapters we have demonstrated the importance of TnC during the

inactivation of the myofilament in the cardiac myofibrils. Even though the myofibrils are

a complex system that contains the contractile machinery necessary for contraction and relaxation, they are not complete. They were devoid of the extensive Ca2+ handling

system, as well as the other chemical, hormonal, and neural factors that regulate heart

function on a beat to beat basis. The ideal biological system to study TnC and the effects

that its altered Ca2+ binding properties have on the physiological processes of the contracting and relaxing myocyte is in-vivo. The work done in this chapter describes the steps taken to incorporate our modified TnCs into the beating heart to better understand

TnC’s role in cardiac physiology.

The many different genetic approaches used for manipulating the heart have expanded the field of molecular cardiology [134]. A majority of the proteins in the heart have been targeted for genetic manipulation, whether it was through the creation of transgenic animal models or the use of viral vectors [135]. Virus constructs developed from retroviruses, lentiviruses, adenoviruses, and adeno-associated viruses (AAV) have been used to target the heart and, more specifically, the cardiomyocytes [135]. This viral technology has allowed many different modified proteins and their effect on whole organ,

105

as well as whole animal function, to be studied. For each specific viral vector there are

positive and negative aspects in regards to their ability to safely and efficiently transduce

the heart cells. For instance, lentiviruses and adenoviruses are highly efficient at

transducing the isolated myocytes with a large gene cassette (~ 8-10 kb). The downside

to these viral vectors is that they can induce a potent cellular and humoral immune

response, which can be toxic and/or lethal for the animal [136-138]. A more promising approach to genetically modifying the myocardial tissue has been the use of AAV vectors

[139, 140].

4.1.2 AAV as a Transfer Vehicle for Gene Therapy

AAV has been shown to be a successful candidate for the transfection of genetic constructs into the myocyte and has multiple serotypes with different tissue specificity

[141]. Recombinant AAV (rAAV) contains single stranded DNA (ssDNA) and is incapable of replicating without the help of a helper virus (for instance, herpes virus).

The host cell must convert the ssDNA vector genome into double stranded DNA

(dsDNA) before gene expression can occur [142]. To increase protein transduction efficiency, self-complimentary serotype 9 AAV (scAAV-9,) was used. The scAAV already contains dsDNA, which bypasses the de novo synthesis of ssDNA by the host cell. Although the efficiency of protein transduction is drastically increased, one of the limitations of scAAV is the size of the vector construct that can be packaged inside the capsid. When rAAV is used the maximal vector size is ~ 4700 bp and is reduced to ~

106

2500 bp for scAAV [142]. One of the most beneficial aspects of using AAV for gene

therapy is little or no immune response brought on by the induction of AAV titers into systemic circulation, unlike the adenovirus and lentivirus [140]. However, there may be recent evidence that AAV is able to induce an immune response in a certain human population [143], thereby eliciting the need for screening before gene therapy could be utilized. Another concern of gene therapy is the insertion of the viral vector into the host’s genome. Although integration into the genome is rare, it can occur with the episomal DNA of AAV and usually integrates into transcriptionally active or open chromatin [144, 145]. Removal of the replication proteins leads to a much lower integration [140, 146]. In addition to lower genomic integration, AAV serotype-9 has been shown to produce high levels of protein expression in striated muscle [147, 148].

Finally, in studying models of chronic disease, such as heart disease, it is imperative to maintain the expression levels for long periods of time. AAV has been shown to express the transfected genes for periods of at least a year [149].

AAV gene therapy has been able to target the Ca2+ handling protein, SERCA

[150], but there have been no published reports that used in vivo AAV transfection of

TnC to study its role in cardiac physiology. The following sections describe the preliminary work undertaken to show that TnC gene transfer into the heart using scAAV is an effective method for studying TnCs physiological role in the heart. Through the use of TnCs with modified Ca2+ binding properties we can begin to recapitulate the diseased

phenotype of DCM in the animal model. As shown in the previous chapter, the D73N

107

TnC mutation resulted in a decreased Ca2+ sensitivity and increased rate of Ca2+

dissociation as measured from Tn exchanged myofibrils. Similar to these results, upon

insertion of the D73N TnC into the animal model, the contractile parameters of the heart were similar to a DCM model phenotype. On the other hand, the L48Q TnC had an

increased Ca2+ sensitivity and decreased rate of Ca2+ dissociation in the myofibrils that would be characteristic of an HCM heart model. However, upon insertion of L48Q TnC into the heart, the contractile properties of the heart were not impaired and did not result in an HCM model as measured by echocardiography. In fact, the L48Q TnC may actually improve cardiac function. Thus, the studies examined the effect of altered TnC

Ca2+ sensitivity and Ca2+ dissociation kinetics on intact heart function and also examined

whether or not the D73N and L48Q were able to recapitulate the respective DCM and

HCM phenotypes. The data also provided the groundwork for the design and

implementation for future work to examine whether or not modified TnCs can correct the

diseased phenotypes for a broad range of cardiomyopathies.

4.2 Methods

This section describes the methods used to create the viral transfer vector

containing TnCs with modified Ca2+ binding properties. The methods also describe how

the scAAV-9 containing the TnC genetic constructs were injected into neonatal mice and

how the in vivo heart function was monitored by echocardiography. The scAAV-9

serotype was used for all vectors and will be identified as AAV-9 throughout the rest of the thesis. The production of the AAV-9 containing the different TnC genes (Control,

108

D73N, and L48Q) and green fluorescent protein (GFP) were performed by Dr. Jiancho

Zhang in the Davis lab. All of the experiments involving the use of isolated cardiac myocytes were performed by Steve Roof, a GRA in Dr. Mark Ziolo’s lab.

4.2.1 Design of the TnC Viral Vector Construct

In order to produce the TnC AAV-9, our TnC gene had to be modified and then transferred from the bacterial expression vector into the B54 viral transfer vector. The modifications of the bacterial plasmid involved the insertion of two NotI restrictions sites, the addition of a Kozak sequence before the start codon, and a FLAG-tag attached to the C-terminal domain of the TnC gene. Additionally, to increase the chances that the

TnC protein would incorporate into the myofilament of the intact myocytes, the remaining Cysteine (Cys) at the 53 amino acid position was point-mutated back to the native Threonine (Thr). This eliminates the translated and over-expressed TnC protein inside the myocytes from forming disulfide bonds with one another, which would lead to their degradation. The NotI sites were critical to allow the gene to be cut from the bacterial plasmid (pet3a) and ligated into the viral vector (B54). The NotI restriction site is recognized by a restriction enzyme which binds to the DNA and cuts the fragment out at all specified sites. The Kozak sequence is important for the translation of the mRNA in eukaryotic cells and can vary from gene to gene. Depending on the Kozak sequence present, the amount of protein translated can be effected. This is another way for the cell to regulate gene expression. The Kozak sequence that was inserted in front of the TnC gene is believed to produce maximal gene expression [151]. The addition of the FLAG-

109 tag polypeptide sequence (N-DYKDDDDK-C, or nucleotide sequence GAT TAC AAG

GAT GAC GAC GAT AAG) at the C-terminal domain of the TnC before the stop codon allowed for the detection of our TnC through the use of a FLAG-tag specific antibody

(Sigma Aldrich, St. Louis, MO). The final Control TnC gene sequence before insertion into the viral transfer vector is displayed in Figure 14. All of the described modifications were put into the Control TnC, and the two modified TnCs, D73N TnC and L48Q TnC.

All gene modifications (point mutations and addition of Kozak, FLAG-tag, etc.) were done using primer-based mutagenesis as previously described [34]. Briefly, each program started with the lid heated to 100oC and then 30s at 95oC. The following loop cycle consisted of 30 seconds at 95oC, 1 minute at 55oC, 7 minutes at 68oC, which was repeated 17 times. Upon completion, the reaction was held at 4oC.

GCGGCCGCCATGCCACCATGGATGACATCTACAAGGCTGCGGTAGAGCAGCT GACAGAAGAGCAGAAAAATGAGTTCAAGGCAGCCTTCGACATCTTCGTGCTG GGCGCTGAGGATGGCTCCATCAGCACCAAGGAGCTGGGCAAGGTGATGAGG ATGCTGGGCCAGAACCCCACCCCTGAGGAGCTGCAGGAGATGATCGATGAG GTGGACGAGGACGGCAGCGGCACGGTGGACTTTGATGAGTTCCTGGTCATGA TGGTTCGGTCCATGAAGGACGACAGCAAAGGGAAATCTGAGGAGGAGCTGT CTGACCTCTTCCGCATGTTTGACAAAAATGCTGATGGCTACATCGACCTGGAT GAGCTGAAGATAATGCTGCAGGCTACAGGCGAGACCATCACGGAGGACGAC ATCGAGGAGCTCATGAAGGACGGAGACAAGAACAACGACGGCCGCATCGAC TATGATGAGTTCCTGGAGTTCATGAAGGGTGTGGAGGATTACAAGGATGACG ACGATAAGTGAGCGGCCGCAA

Figure 14. DNA Sequence of Control TnC Gene. The colors highlight the different regions of the plasmid containing the Control TnC gene:Not1 Restriction Site, Kozak, TnC, FLAG-tag, Stop Codon

110

After the insertion of the NotI sites into the TnC gene, the gene fragment was digested using the NotI restriction enzyme (New England BioLabs, Ipswich, MA) and ran on an agarose gel. A 100 ml 1 % agarose gel was made by dissolving the agarose in 0.5x

TBE (Invitrogen, Grand Island, NY) with a microwave for ~ 1 minute. The solution was cooled on the bench for ~ 5 minutes before adding 2 μl of ethidium bromide (10mg/ml) and slowly swirled to mix. The solution was poured into the gel caster without creating bubbles and the lane comb was inserted and sat for 1 hour with the lid on to solidify.

The gel was loaded with a lane marker and DNA samples and was run at a constant 50 V in 0.5x TBE buffer for 10 minutes and then increased to 100V until the dye front was ¾ the length of the gel. The gel was examined under UV light and the band at ~ 500 kb was cut and extracted following the QIAquick Gel Extraction Kit protocol (Qiagen, Valencia,

CA). The band intensity of the TnC gene fragment was compared to the lane marker at

500 bp (42 ng) to determine the approximate concentration of DNA (typically ~ 14 ng/μl).

The TnC gene fragment was ligated into the previously digested and calf intestinal phosphatase treated B54 viral vector that had complimentary NotI restriction sites (kindly produced by Dr. Reed Clark, Children’s Hospital). For the ligation procedure, 133 ng of vector and 67 ng of the TnC gene fragment were required for optimal ligation. This reduced the chances that the TnC gene fragment would ligate to itself instead of into the vector. The T4 DNA Ligase protocol (New England BioLabs,

Ipswich, MA) was followed and upon completion, 5 μl of vector was transfected into Xl-

111

1 Blue competent cells (Invitrogen, Grand Island, NY), plated on LB plates and grown

overnight at 37oC. Multiple colonies were picked and grown up in 5 ml of LB with

ampicillin and then the QIAprep Spin Miniprep Kit (Qiagen, Valencia, CA) was used to

collect the ligated vector with the gene. Finally, the B54 vector containing the respective

TnC genes were NotI digested and then ran on an agarose gel to check for the appropriate

size fragments (527 bp, Figure 14). The ligated vectors were sent away for DNA sequence analysis (Plant Microbe Genomics Facility, The Ohio State University), to verify our Control, D73N and L48Q TnC gene.

4.2.2 Induction of Hypothermia and Injection of Neonatal Pups

Neonatal mice were injected between 1 and 2 days after birth. Mice were anesthetized by induction of hypothermia prior to viral injection. The ideal age for injection was day 1-2 (the first 24 hours after birth is considered day 0), because they have had at least two days to gain strength before being injected. Before inducing hypothermia, the parents were removed from the pups and placed in a separate cage. The bedding from the pups’ cage was rubbed over gloved hands to remove any unnatural or suspicious scents. Three pups at a time were placed on a glass Petri dish that was placed on top of icy water. The induction time was approximately 2-5 minutes or until they were unresponsive to touch. It was important to make sure the mice did not move during or right after the injection to reduce the amount of virus that leaked out of the injection site. The anesthetized neonatal pups were injected using a modified approach to intra- 112

peritoneal injections to minimize the leakage of the virus after the injection. To inject, a

30-gauge needle was inserted into the skin, parallel to the right or left femoral vessels and

then advanced subcutaneously until the abdominal cavity was punctured. If the

abdominal cavity was not penetrated then a subcutaneous bolus would appear during the

injection. A maximum volume of 50 μl was delivered into the peritoneal cavity that

contained ~ 5 x 1011 vector genomes or DNase resistant particles (drp). After the injection, the pup was tattooed on their paws to distinguish the different AAV-9 viruses, as well as mice that had leaky injections. This was done by dipping the tip of a 30-gauge needle into green tattoo ink and then piercing the paw all the way through and into a

Styrofoam block that the mouse was placed atop. This prevented sticking oneself with the needle while tattooing the mouse. The mouse paw was wiped to remove any excess

tattoo ink or blood. The pups were placed on a heating pad (either a bag filled with warm

water or an ice pack that had been equilibrated with warm water) that was covered with a

paper towel for approximately 5 minutes or until they regained their pink/red color and

were capable of spontaneous movement. The liquid filled heating pad was heated by

soaking in hot tap water or using a bag filled with warm water. This prevented the pad

from getting too hot and potentially burning or overheating the pups. Once warmed, the

pups were gently rubbed with the old bedding and returned to their cage. The parents

were then returned to the cage after all of the pups were warmed and returned to the cage.

113

4.2.3 Echocardiography

Echocardiography was performed on WT and AAV-9 injected mice to measure the in vivo function of the heart using the Vevo 2100 (Visualsonics, Toronto, Ontario,

Canada). One mouse at a time was placed into the anesthesia chamber and anesthetized using 2.0 % Isoflurane that was added to 95% O2 / 5% CO2 at a rate of ~ 0.8 L/min. Once it was sufficiently anesthetized (did not move while being touched), the mouse was removed from the chamber and placed in the supine position on the heated handling table. Care was taken to avoid over anesthetizing the mouse, which can drastically reduce the respiratory rate (< 50-90 respirations/minute), as well as the heart rate and its function. Anesthesia was maintained through the use of a nose cone that provided oxygen and 1% Isoflurane. A small amount of electrode gel was placed on the four EKG leads on the platform and the four mouse limbs were taped to the platform. The EKG was monitored while positioning the limbs to make sure there was adequate electrical conduction for data collection. A temperature probe was inserted into the rectum of the mouse to monitor the core temperature which was maintained at ~ 35oC. The heated

platform temperature was set at 42oC and a heat lamp was placed over top of the mouse

to regulate and maintain the temperature. Nair (Church & Dwight Co., Princeton, NJ)

containing baby oil was used to remove the hair from the chest, which was vital for a

strong signal during the echocardiography.

Temperature, respiration rate, and the EKG with heart rate output were collected

while simultaneously performing the echocardiography. The MS-400 transducer was

114

used to collect the long and short axis, B and M mode echocardiographs for each of the

mice. Basic measurements collected from the M-mode included the internal dimensions

of the left ventricle during diastole and systole of the heart. To collect the data from the

long axis M-mode the probe was positioned to have the apex, the aorta, and the papillary

muscle all in the same field of view and then the cursor was placed slightly inferior to the

papillary muscle. For the short axis data collection the cursor was positioned superior of the apex so that two papillary muscles were in the field of view about half way up the heart. The cursor to collect the measurements was then placed just to the side of the papillary muscles to measure the anterior and posterior wall movement. From these values, the computer was able to calculate diastolic and systolic internal dimensions, diastolic and systolic chamber volumes, stroke volume, ejection fraction, fractional

shortening, and mass of left ventricle.

4.2.4 Cardiomyocyte Isolation and Measurement of Myocyte Function

Ventricular myocytes were isolated from mice as previously described [152].

Briefly, the heart was cannulated and hung on a Langendorff apparatus. It was then

perfused with Ca2+ free tyrode solution for 4 min. The solution was then switched to tyrode solution containing Liberase Blendzyme II (0.077 mg/ml) (Roche Applied

Science, Indianapolis, IN). After 3-5 min, the heart was taken down, the ventricles minced, and myocytes were dissociated by trituration. Subsequently the myocytes were filtered, centrifuged, and resuspended in Tyrode solution containing 200 μM Ca2+. 115

Myocytes were used within 4 hours of isolation. All the animal protocols and procedures

were performed in accordance with National Institutes of Health guidelines and approved

by the Institutional Laboratory Animal Care and Use Committee at The Ohio State

University.

Ca2+ transient measurements were performed as previously described [152].

Briefly, myocytes were loaded at room temperature with Fluo-4 AM (10 μM, Molecular

Probes, Eugene, OR) for 30 min, and then another 30 min were allowed for intracellular de-esterification. The solution for de-esterification was Tyrode solution containing 200

μM Ca2+. The instrumentation used for cell fluorescence measurements was a Cairn

2+ Research Limited (Faversham, UK) epi-fluorescence system. [Ca ]i was measured by

Fluo-4 epi-fluorescence with excitation at 480±20 nm and emission at 535±25 nm. The

illumination field was restricted to collect the emission of a single cell. Data were

expressed as ΔF/F0, where F is the fluorescence intensity and F0 is the intensity at rest.

Myocytes were stimulated at a range from 0.5 to 2 Hz via platinum electrodes connected

to a Grass Telefactor S48 stimulator (West Warwick, RI). Measurements were

performed at room temperature. Normal Tyrode solution consisted of (in mM): 140

NaCl, 4 KCl, 1 MgCl2, 1 CaCl2, 10 glucose, 5 HEPES, pH 7.4 adjusted with NaOH or

HCl. Isoproterenol (ISO, 1 μM, a non-selective β-AR agonist) was prepared fresh each day.

116

4.2.5 Preparing AAV-9 Injected Tissue for Western Blot Analysis

The heart tissue was weighed (~ 50 mg) and then homogenized in 10x

volume/weight (~ 500 μl) of SRB-TX containing proteases in a 2 ml glass dounce tissue

homogenizer using a loose pestle and then a tight pestle. They type of tissue (i.e. LV,

RV, Atria) was specified in the figure legends. The homogenate was pelleted at 16000 x

g at 4oC for 10 minutes. The supernatant was removed from the myofibrils and the

procedure to precipitate the proteins out is described in the next paragraph. The

myofibril pellet was re-suspended in 500 μl of SRB containing proteases using a loose

pestle to homogenize the sample and then re-pelleted. The last myofibril pellet was re-

suspended in Sample Buffer (SB) containing; 50 mM Tris-HCl pH 6.8, 2% SDS, 0.1%

bromophenol blue, 10% glycerol, 67 mM DTT.

To determine if FLAG-tag TnC was present in the cytosol of the myocyte, the supernatant was saved from the previous steps and acetone precipitated. Acetone, cooled to -20oC (4 times the supernatant volume, ~ 1 ml) was added to the supernatant. The tube

was vortexed and incubated at -20oC for 1 hour. The precipitated protein was pelleted at

14,000 x g at 4oC for 10 minutes, supernatant decanted, and re-suspended in SB (~ 100

μl). Both the myofibril and cytosolic proteins were run on SDS gel for Western blot

analysis.

117

4.2.6 SDS Gel and Western Blot Analysis

The samples in SB were loaded onto a 15% SDS resolving gel (1 mm) with a 29:1 ratio of acrylamide/bisacrylamide with a 4% SDS stacking gel. Gels were run at a constant 20 mA per gel at 10oC for ~ 1 hour in SDS running buffer (25 mM Tris, 192

mM glycine, 0.1% SDS). While running the SDS gel, the 0.2 μm PVDF membrane was

cut to size of the resolving gel (58 x 80 mm) and activated in 100% methanol for 30s and

then placed in transfer buffer (TB) containing; 25 mM Tris, 192 mM glycine, and 20%

methanol for 15 minutes. The filters and filter papers were soaked in TB for 15 minutes.

The SDS gel was soaked in TB for 10 minutes. To transfer the proteins from the gel to

the PVDF membrane the gel and membrane were placed in between two layers of filter

paper and the two filters on the outside. The entire transfer cassette was submerged in TB

and the proteins transferred at constant 90 V for 90 minutes at 10oC.

After the transfer, the PVDF membrane with transferred proteins was washed

three times in Tris Buffered Saline (TBS) containing 50 mM Tris-HCl, pH 7.4, 150 mM

NaCl with 0.1% Tween20 (TBST). The membrane was blocked with 5 % milk in TBST for 30 minutes at room temperature. When probing for the FLAG-tag TnC, a monoclonal anti-FLAG-tag antibody that was conjugated with horseradish peroxidase (HRP, Sigma,

St. Louis, MO) was used at 1:1000 in 5% milk TBS and lightly rocked overnight at 4oC.

The membrane was washed 3x in TBS for 5 minutes and then developed using enhanced chemi-luminescence (ECL +, GE Healthcare, Waukesha, WI).

118

To probe the total TnC present in the myofibrils, the membrane containing the

anti-FLAG-tag antibody was stripped with Western blot stripping buffer (Thermo

Scientific, Rockford, IL) for 20 minutes at room temperature. The membrane was

washed 3x with TBST for 5 minutes, blocked 30 minutes with 5% milk in TBST. The primary antibody, anti-mouse cardiac TnC clone 7B9 (Fitzgerald, Acton, MA) at 1:7500 in 5% milk in TBST, was added to membrane for 2 hours at room temperature. The membrane was washed 3x with TBS (5 minutes) and was incubated with a secondary anti-mouse IgG-peroxidase (Sigma, St. Louis, MO) at 1:25,000 in 5% milk in TBST for 1 hour at room temperature. The membrane was washed 3x with TBS (5 minutes) and then

developed using ECL+.

4.2.7 Tissue Fixation and Sectioning for GFP Expression

Detailed methods for the fixation and sectioning of the heart tissue can be found

in Appendix B. Briefly, after fixation in 4% para-formaldehyde, the tissue is cryoprotected by infiltration in 20% sucrose in buffer with agitation overnight at 4oC. After

24 hrs the tissue block will sink in the solution, indicating they are infiltrated. Liquid

Isopentane was added to a glass beaker and then cooled in liquid nitrogen. The piece of

embedded tissue was placed at the bottom of an aluminum foil cup that was filled

halfway with Tissue Tek OCT (Fisher Scientific, Rockville, MD). The aluminum cup

with tissue was placed half way into the chilled Isopentane using large forceps. The

sample was kept in the liquid Isopentane until the OCT turned completely white. The

119

sample was removed and stored in the -80oC freezer in individual small tubes. The frozen tissue samples that were embedded in OCT were cut on a cryostat to 20 μm thick

sections and GFP was observed on the Olympus Filter Confocal Microscope (FV1000) using a 10x objective. The excitation wavelength was 488 nm with an emission wavelength of 520 nm. For a rough estimate of GFP expression in the various tissues, pieces of tissue were sliced as thinly as possible with a razor blade, placed onto a glass microscope slide, covered with a glass cover slip, and examined under a fluorescent inverted microscope. A more detailed description is in the figure legend from the collected images (Figure 16).

4.2.8 Statistical Analysis

All data are expressed as mean (± SE). Statistical significance was determined by using 1-way ANOVA followed by a post-hoc least significance difference test using

GraphPad Prism 4 (La Jolla, CA). Statistical significance was defined as P < 0.05.

4.3 Results

4.3.1 Effect of FLAG-tag on TnC’s Ca2+ Dissociation Kinetics

A FLAG-tag was attached to the C-terminal domain of Control T53C TnC using

insertional mutagenesis into the recombinant DNA. Depending on what biochemical system was used to measure the Ca2+ dissociation kinetics, the Control T53C TnC was

120

fluorescently labeled with either IAANS or IANBD. Reconstituted thin filaments were

formed using Control TnC and the myofibrils were Tn exchanged with Control T53C IAANS TnC Tn as described in Figure 15. Panel A of Figure 15 shows the Ca2+ T53C IANBD dissociation kinetics from FLAG-tagged Control TnC reconstituted thin filaments T53C IAANS (Flag-tagged Control TnC, (98 ± 3/s) and non-FLAG-tagged Control TnC T53C IAANS reconstituted thin filaments (Control TnC, 107 ± 2/s). Similar to the reconstituted thin filaments, Flag-tagged TnC did not affect the Ca2+ dissociation kinetics from FLAG-

tagged Control TnC rabbit ventricular myofibrils (Panel B, Flag-tagged Control T53C IANBD TnC (28 ±1/s)) as compared to the non-FLAG-tagged Control TnC myofibrils T53C IANBD (Panel B, Control TnC (25.3 ± 0.7/s)). Thus, the FLAG-tag attached to the C-terminal

domain of TnC did not alter the apparent rate of Ca2+ dissociation as measured from

multiple biochemical systems. For the remaining experiments in this chapter, Cys 53 was

modified back to Thr for Control, L48Q, and D73N TnC before the TnC genes were

inserted into the viral transfer vector. This was done to prevent the formation of disulfide

bonds between the Cys residues of the TnCs upon transduction of the TnC gene into an

animal model.

121

Figure 15. Ca2+ Dissociation from Flag-tagged Control TnC. Panel A shows the time course of IAANS fluorescence decay as Ca2+ was chelated by EGTA and removed from the regulatory binding site of reconstituted thin filaments containing either Control or Flag-tagged Control TnC Tn. The reconstituted T53C 2+ thin filaments in Buffer A (without Tween20) + 200 μMIAANS Ca were rapidly mixed with an equal volume of Buffer A (without Tween20) + 10 mM EGTA at 15oC. Panel B shows the time course of IANBD fluorescence as Ca2+ was chelated by EGTA and removed from the regulatory binding site of Control and Flag-tagged Control TnC after Tn exchange into rabbit myofibrils. The myofibrils in T53C 2+ Buffer A + 200IANBD μM Ca were rapidly mixed with an equal volume of Buffer A + 10 mM EGTA at 15oC.

(continued)

122

Figure 15 continued.

123

4.3.2 AAV-9 GFP Targets both Cardiac and Skeletal Muscle

There were four AAV-9 vectors prepared in the lab by Dr. Jiancho Zhang which

contained either GFP, Control TnC, D73N TnC, or L48Q TnC. AAV-9-GFP was used to

determine what tissue the virus transfected and whether or not the gene was being

transcribed. Figure 16 shows the GFP expression within the different cardiac and skeletal

muscles at 8 weeks post neonatal AAV-9 GFP injection. Panels A-D show the GFP expression in the atria, left ventricle, diaphragm, and psoas muscle, respectively, as observed from a 10x objective on an inverted fluorescence microscope. Panel E shows the GFP fluorescence from a 20 μm thick section of left ventricular tissue as observed from a 10x objective lens on an Olympus Filter confocal microscope. Panel F shows the

GFP fluorescence of Panel E overlaid on the reflected light differential interference contrast (DIC) image. For a comparison to the GFP images, Panel G shows the background fluorescence of the left ventricle of the WT mouse that was overlaid on the

DIC image in Panel H. The percent of GFP cells present in Panel E was ~ 50%. In other areas of the heart the transfection percentage was much lower (~ 5-10%) and varied from region to region, making the total quantification difficult (data not shown). The brain, kidney, lungs, stomach, intestines, liver, and spleen had very low levels of GFP expression (< 5 GFP positive cells in tissue examined under 10x objective (data not shown)), in agreement with a previous study that used AAV-9 containing LacZ [147].

Thus, the AAV-9 GFP was able to target the heart after IP injection and transduced the cells to produce the GFP protein.

124

Panel A Panel B

Figure 16. Tissue from Mice Transduced with AAV-9 GFP. This figure shows AAV-9 GFP expression in different tissues from 8 week old mice C57BL/6 mice that were IP injected at day 2-3 after birth. Panels A-D show the GFP expression in tissue from various striated muscles (atria, ventricle, diaphragm, psoas, respectively) using a 10x objective on a Nikon Inverted Fluorescence Microscope. A piece of tissue was sliced with a razor blade (~200-400 μm thick) and was placed on a glass slide and coverslipped. The FITC fluorescence filter was utilized and had a bandpass filter with excitation wavelength of 460-500 nm and emission wavelength of 510-560 nm. The images were taken through the eyepiece with a camera phone. Panels E-H show images from tissue that was fixed in paraformaldehyde and sucrose, embedded in OTC, sliced in 20 μm thick sections on a cryostat, placed on glass-slide and measured with the Olympus Filter Confocal Microscope (FV1000) using a 10x objective. The excitation wavelength was 488 nm with an emission wavelength of 520 nm. Panel E shows the GFP expression in the AAV-9 GFP transduced left ventricular tissue and was overlaid on the DIC microscopy image in Panel F. Panel G shows the background fluorescence of the left ventricle from a WT mouse at the same intensity as that used to acquire the image in Panel E. Panel H shows the WT background fluorescence of Panel G overlaid on its DIC microscopic image.

(continued)

125

Figure 16 continued.

Panel C Panel D

Panel E Panel F

(continued)

126

Figure 16 continued.

Panel G Panel H

127

4.3.3 In Vivo Incorporation of FLAG-tag TnC into the Myofilament

To determine whether or not the AAV-9 containing the FLAG-tagged TnC was able to transduce the mouse hearts, a Western blot was done using the anti-FLAG antibody. Initially, the whole heart tissue homogenate was probed with an antibody for the FLAG-tag, which was positive for the D73N TnC heart tissue (data not shown). To further probe the location of D73N TnC inside the myocyte, the myofilament was separated from the cytosol by centrifugation. This was done to determine whether the

TnC incorporated into the myofilament or was simply free floating in the cytosol. Figure

17 compares the myofilament preparation to that of the cytosolic fraction. The FLAG-tag

TnC stained positive in the myofilament fraction and was not present in the supernatant, or at least not at levels that were sensed by the antibody. The positive FLAG-tagged TnC was circled in red. The myofilament bands above the red circle are non-specific antibody binding and are not the FLAG-tagged TnC because this band was also present in the non- injected WT myofilament. The non-specific FLAG-tag stained bands in the supernatant ran at a lower molecular weight than the FLAG-tagged TnC of the myofilament circled in red. This contaminating, non-specific FLAG-tag binding band also showed up in the non-injected WT mouse and provided additional evidence that this band is unrelated to the FLAG-tagged TnC. This would indicate that the TnC was transfected into the heart and there was no free FLAG-tagged TnC in the cytosol of the heart tissue of the TnC injected mice. As can be seen by the non-specific protein binding of the FLAG-tag antibody, it was not the best for distinguishing the FLAG-tagged TnC from the tissue

128 preparations. A western blot that probed for total TnC levels utilized a cardiac TnC antibody (data not shown). This antibody stained positive for TnC in the same region as the bands circled in red but the two non-specific bands were not able to be identified by the TnC specific antibody.

129

Figure 17. FLAG-tag Western Blot of AAV-9 D73N TnC Injected Mice. Anti- FLAG (DYKDDDDK) antibody was used to detect the FLAG-tagged D73N TnC in the hearts of the AAV-9 D73N TnC transduced mice as compared to WT. The myofilament proteins were separated from the cytosolic proteins to determine the location of the FLAG-tagged TnC and whether or not it incorporated into the myofilament. The red circle highlights the regions that stained positive for FLAG-tagged TnC. The bands in the supernatant resulted from non-specific antibody binding and ran lower that the FLAG-tag positive bands.

130

4.3.4 Death Rates of AAV-9 Injected Mice

As seen in Figure 16, the AAV-9 is able to target the heart as well as skeletal

muscle and transduce the cardiac and skeletal tissue with GFP. In addition to AAV-9

GFP, neonatal mice were also injected with AAV-9 Control TnC, AAV-9 D73N TnC,

and AAV-9 L48Q TnC. The survival rates for all of the AAV-9 injected mice can be

seen in Figure 18. At 4 weeks, 47% of the D73N TnC injected mice (25/53) died and/or

were eaten at 4 weeks after injection. Unlike D73N TnC, only 1/32 of the GFP, Control

TnC, and L48Q TnC mice had died. One GFP mouse had to be euthanized at 3.5 because the mouse was sick and not eating for an unknown reason. At 8 weeks the death rate of the D73N TnC injected mice increased to 60% (32/53) and one additional death for GFP,

Control TnC, and L48Q TnC injected mice. An L48Q mouse died and/or was eaten two days after injection. Due to the D73N TnC mice being eaten, the other mice (GFP,

Control TnC and L48Q TnC) may have been contaminated due to the transmission of the

D73N TnC and will be discussed in more detail later. It is possible that the two mice that died in the GFP, Control TnC, and L48Q TnC group may have been due to D73N TnC contamination. It would appear that the D73N TnC virus can be passed on from parents that have eaten their young, as will be discussed later.

131

Panel A

4 Week Survival

100 Survival 47% 80 Death

60 3% 25/53

40 (1/32) % of Mice 20

0 Control, GFP, L48Q D73N Treatment

Panel B

8 Week Survival

100 Survival 80 60% Death

60 6% (2/32) (32/53) 40 % of Mice 20

0 Control, GFP, L48Q D73N Treatment

Figure 18. Survival Rates of AAV Injected Mice. The death rate of AAV-9 (containing Control TnC, GFP, L48Q TnC, and D73N TnC) were compared at 4 and 8 weeks after neonatal injection. The 4 week death rate of AAV-9 containing either Control, GFP, and L48Q injected mice (3%, n=32) were drastically reduced compared to AAV-9 D73N TnC injected mice (47%, n = 53). At 8 weeks the death rate for the Control, GFP, and L48Q TnC injected mice increased to 6% compared to the D73N TnC mice which increased to 60%. 132

4.3.5 In Vivo Heart Function of AAV-9 TnC Injected Mice

The baseline short-axis echocardiography values were measured at 4 weeks for

WT, D73N, and L48Q mice. The values collected are presented in Table 2. The values presented were collected and averaged from 5 WT, 11 D73N TnC, and 9 L48Q TnC injected mice. The systolic and diastolic dimensions were used to compare the WT,

D73N, and L48Q TnC injected mice. Of the D73N TnC injected mice that survived beyond 4 weeks, 50% (14/28) had dilated hearts as determined during the echoardiography measurements at 4 weeks (defined by being more than two standard deviations larger than the WT mice). Both the end systolic and diastolic diameters for these mice were greater than the WT mice. Even more intriguing is the presence of EKG abnormalities in 43% of the D73N TnC mice that had a dilated left ventricle (6/14, EKG

Beyond 4 Weeks, Figure 19) and will be discussed in more detail later.

A representative short axis M-mode view of a WT mouse can be seen in Figure

20A. Figure 20B shows how the wall chambers of a dilated D73N TnC mouse heart were outlined to allow the computer software to calculate the contractile parameters.

There are two lines that outline the internal anterior and posterior wall of the heart.

Systole is the point at which the walls are the closest together and diastole occurs when the heart relaxes (farthest distance between the interior walls). From these dimensions, the software calculates all of the aforementioned parameters. The lines outlining the outside of the heart are used to calculate the mass of the heart based on the thickness of the walls, which the software uses for the extrapolation of the LV mass calculation.

133

Compared to the WT mice (1.80 ± 0.05 & 3.09 ± 0.03 mm), D73N TnC mice (2.41 ±

0.06 & 3.62 ± 0.08 mm) had an increase in both systolic and diastolic diameter,

respectively. L48Q TnC had an increase in only the systolic diameter (2.09 ± 0.03 mm).

In addition to diameters, the heart rate, systolic and diastolic volumes, stroke volume, ejection fraction, fractional shortening, cardiac output, LV mass, and corrected LV mass were collected. Both systolic and diastolic volumes, stroke volume, and LV mass were increased in the D73N TnC mice, whereas ejection fraction and fractional shortening were decreased for the D73N TnC mice as compared to WT mice. The L48Q TnC mice

had a slightly increased systolic volume and LV mass with a reduced ejection fraction

and fractional shortening as compared to WT. Overall, the D73N TnC mice had a

phenotype consistent with DCM, even though some mice had a much greater DCM

presentation than others. The L48Q mice had echocardiograph values that would indicate

a slightly enlarged heart with an increased systolic volume. This increased volume could

have accounted for why there was a reduced ejection fraction and fractional shortening.

Once again, these differences could have been due to the transmission of AAV-9 D73N

TnC and will be discussed further in section 4.3.8. The L48Q mice seemed to be healthy

and did not have any EKG abnormalities during the time span of 8 weeks.

134

LV 4 Week Heart Diameter; Diameter; Volume; Volume; Stroke Ejection Fractional Cardiac LV Mass Old Rate s d s d Volume Fraction Shortening Output Mass Cor BPM mm mm μl μl μl % % ml/min mg mg WT Mean 504 1.80 3.09 9.9 37.7 27.8 74 42 14.0 48 39 n = 5 ± SE 9 0.05 0.03 0.8 0.9 0.5 1 1 0.5 2 2 ± SD 36 0.21 0.12 3.0 3.6 2.0 6 5 1.8 8 6

D73N Mean 479 2.41 # 3.62 # 21 # 56 # 36 * 63.1 # 33.5 # 19 71 # 57 # n = 11 ± SE 22 0.06 0.08 1 3 2 0.8 0.6 2 3 2 ± SD 122 0.31 0.45 7 19 13 4.6 3.3 10 10 8

135 L48Q Mean 503 2.09 # 3.29 14.3 * 44.1 29.7 67.3 # 36.5 # 15.0 60 ^ 48 ^ n = 9 ± SE 6 0.03 0.03 0.4 0.9 0.8 0.9 0.7 0.5 2 2 ± SD 29 0.13 0.15 2.3 4.9 4.2 4.9 3.7 2.6 11 9 Table 2. Echocardiographic Parameters and Values for Wild-type Mice. The table displays the parameters measured by the Vevo 2100 Ultrasound machine for wild-type C57BL/6 mice at 4 weeks. The mice were under ~ 1% Isoflurane anesthesia and values reported are from the short-axis measurements of the heart. 1-way ANOVA followed by a post-hoc least significance difference test was performed using GraphPad Prism 4 (La Jolla, CA). Statistical significance was defined as P < 0.05. * denotes P-values significance of < 0.05, ^ < 0.01, # < 0.001 as compared to WT mice.

135

Panel A

Internal Dimensions (Beyond 4 weeks)

100 Normal 80 50% Dilated

60 (14/28)

% of Mice 20 5 0 WT D73N Treatment

Panel B

EKG (Beyond 4 weeks)

100 Normal 43% 80 Abnormality (6/14) 60

40 % of Mice 20 5 0 WT D73N Treatment

Figure 19. Echocardiograph and EKG Characterization of AAV Injected Mice. The internal dimensions of mice injected with AAV-9 D73N TnC that survived beyond 4 weeks were compared to the WT mice. 50% of the D73N mice (14/28), as compared to WT dimensions, had an increase in the systolic and diastolic internal diameters (> two SD). Of the D73N mice that were dilated, 43% (6/14) had abnormal EKG patterns as recorded during the collection of the echocardiograph data. 136

Panel A

Figure 20. Representative M-mode Short-Axis Echocardiographic and EKG Images. Images from the short-axis M-mode and EKG traces collected on the Vevo 2100 are shown in Panels A-D. Panel A image is from a 4-week old wild- type mouse and is shown as a baseline for the AAV-9 D73N TnC injected mice. Panel B short-axis image is from an AAV-9 D73N TnC injected mouse at 4 weeks that was dilated with a relatively normal EKG trace. Panel C image is of an AAV-9 D73N TnC injected mouse at 4 weeks that was dilated but had an erratic and irregular heart beat. Panel D shows additional EKG traces from the mouse presented in Panel C.

(continued) 137

Figure 20 continued.

Panel B

Panel C

(continued) 138

Figure 20 continued.

Panel D

139

4.3.6 EKG Abnormalities of AAV-9 D73N TnC Injected Mice

During the collection of the echocardiograph data the electrocardiogram was

simultaneously collected from the limb leads of the mice. Below the image of the short

axis M-mode view of a WT mouse left ventricle wall tracing in Figure 20 A is the EKG.

The P-wave (atrial contraction) and QRS wave (ventricular depolarization) preceded the ventricular wall contraction, which occurs under normal physiological conditions. Panel

B is a representative short axis M-mode view of a D73N TnC mouse that was dilated but maintained a relatively normal EKG. The P-wave is not as pronounced but still occurs before each and every QRS wave. Panel C is the representative short axis M-mode view of a D73N TnC mouse that was dilated and also had EKG abnormalities. The QRS wave

is arrhythmic and can be further seen in wall motion. There does not appear to be a P-

wave for each ventricular contraction or it may be hidden,which initally presented as a

SA node conduction problem. However, Panel D shows more EKG strips from the same

mouse over a longer time frame that were collected throughout the Echocardiograph data

collection procedure. In this panel it becomes more apparent that P-waves still occur at a normal mouse heart rate under anesthesia (~450/s) but there is not a QRS wave that follows each P-wave. At first this indicated an AV-nodal heart block of the 3rd degree or

a bundle branch block, but there were missing QRS complexes that did not occur for 1-7

P-waves. The arrhythmia was also not regular and the shape of the QRS wave changed

throughout the recording. It was unclear at this point what was causing the EKG

abnormalities but they seemed to progress with time (4 to 8 weeks). Each D73N TnC

140

injected mouse that had an arrhythmia (6/53) presented with different waveform

abnormalities. This phenomenon is not like the various EKG abnormalities observed in

humans with DCM. In addition, all of the arrhythmias occurred in hearts that were

dilated at 4 weeks. None of the other mice (WT, GFP, Control, or L48Q TnC mice, 32

total) indicated any kind of arrhythmias.

4.3.7 Determination of the Contractile Properties of Isolated Cardiomyocytes

Additional functional studies were measured from the myocytes of the different

AAV-9 injected mice (performed by Steve Roof in Dr. Mark Ziolo’s Lab). In agreement with the echo data, the D73N TnC hearts had an increased heart weight/tibia length ratio compared to WT, Control TnC, and L48Q TnC injected mice (Figure 21, Panel A). Panel

B shows that the isolated myocytes from the D73N TnC injected mice had an increased length/width ratio as compared to the WT, Control TnC, and L48Q TnC mice.

Quantitatively, the Ca2+ transient amplitudes were similar between all of the mice

at basal levels of 1 Hz stimulation at room temperature (Figure 22, Panel A). The

shortening amplitudes were increased in L48Q TnC myocytes (5.4 ± 0.6), whereas the

D73N TnC myocytes (2.4 ± 0.2) trended towards a decrease in shortening amplitude

(Panel B) as compared to WT and Control TnC myocytes (3.5 ± 0.4). These results

would be consistent with the increased and decreased Ca2+ sensitivities of L48Q TnC and

D73N TnC as previously measured in the myofibrils (Chapter 3). Under conditions of β-

AR stimulation (addition of 1μM Isoproterenol), the Ca2+ transient and shortening were

141

expected to increase for all of the mice. The Ca2+ transient of the D73N TnC (3.3 ± 0.3)

myocytes did not increase to the extent that the Ca2+ transient of the WT (4.8 ± 0.3),

Control TnC (4.7 ± 0.2), or L48Q TnC myocytes did (4.3 ± 0.2, Panel C). This was indicative of a reduced β-AR response for the D73N TnC isolated myocytes. The shortening amplitudes under conditions of Isoproterenol stimulation trended towards an increase for L48Q TnC myocytes (10 ± 1) and a decrease for D73N TnC myocytes (4.9 ±

0.7) as compared to Control (8 ± 1) and WT (7.4 ± 0.9, Panel D).

142

Panel A Panel B

Heart/Tibia Ratio Cardiomyocyte length/width ratio 12.5 10.0 5.5 7.5 4.5 5.0 mg/mm 2.5 3.5 0.0 0

WT WT

Cont AAVTnC L48QTnC D73N Cont AAVTnC L48QTnC D73N

Figure 21. Comparison between Heart Weight and Myocyte Size of AAV-9 Injected Mice. Panel A shows the heart weight (mg) to tibia length (mm) of the WT and AAV injected mice. Panel B shows the isolated cardiomyocyte length (mm) to width (mm) ratio for the WT and AAV injected mice.

143

Panel A Panel B

2+ Ca Transient Amplitude Shortening Amplitude 2.0 6 * 1.5 5 0 4

F/F 1.0

∆ 3 % RCL % 0.5 2 1 0.0 0

WT WT

Cont AAVTnC L48QTnC D73N Cont AAVTnC L48QTnC D73N

Panel C Panel D

2+ Ca Transient Amplitude Shortening Amplitude

5 10 4 0 8 3 * F/F 6 ∆

2 RCL % 4 1 2 0 0

WT WT

Cont AAVTnC L48QTnC D73N Cont AAVTnC L48QTnC D73N

Figure 22. Contractile Parameters of Isolated Myocytes from AAV Injected 2+ Mice. Panel A shows the intracellular Ca transient amplitude ([Ca]i) as 2+ measured using Fluo-4 in the presence of 1 mM extracellular Ca ([Ca]o). Panel B shows the % shortening as compared to the resting cell length (% RCL). Panel C and D show the [Ca]i and % RCL (respectively) for the cardiomyocytes in the presence of isoproterenol (1 μM) to induce β-AR stimulation. The data are presented as mean ± SE for the pooled myocytes from the respective AAV injected mouse hearts.

144

4.3.8 Frequency Dependency of Isolated Myocyte Ca2+ Transient and

Shortening

The frequency dependency of the Ca2+ transient, shortening amplitudes, Ca2+

2+ transient time to peak (CaTTP), 50% decline of the Ca transient (CaRT50), shortening

time to peak (sTTP), and 50% relaxation time or relengthening (sRT50) for the various myocytes is shown in Figure 23. There was no difference in Ca2+ transient amplitudes

for WT, Control TnC, D73N TnC, and L48Q TnC at 0.25 Hz, 0.5 Hz, 1 Hz, and 2 Hz

(Panel A). The shortening amplitudes for the modified TnCs were drastically different at the extreme frequencies (0.25 Hz and 2 Hz). Panel B shows that at 0.25 Hz the shortening amplitude of D73N TnC myocytes was drastically reduced. On the other hand, at 2 Hz the shortening amplitude of L48Q TnC myocytes was increased compared

to WT, Control TnC, and D73N TnC. Panel C shows a decreased Ca2+ transient TTP for

D73N TnC at 0.25 Hz as compared to WT, Control, and L48Q TnC. There were no

2+ differences in the CaRT50 for the Ca transient decline for any of the groups (Panel D).

There was however, a decrease in the sTTP (0.25, 0.5, and 1 Hz, Panel E) and sRT50 (all

frequencies, Panel F) for D73N TnC as compared to WT, Control, and L48Q TnC. The

decreased shortening amplitude would be expected for a myofilament which had a

decreased sensitivity and a faster rate of Ca2+ dissociation from TnC. At similar Ca2+

transient amplitudes, the reduced shortening of the D73N TnC myocytes was consistent with the DCM phenotype, in which the heart is unable to produce an adequate amount of force. The D73N TnC myocytes also had an increased rate of shortening and

145

relengthening, but this may be due to its drastically reduced shortening amplitude. With

a smaller shortening amplitude there was not as much distance to travel for the D73N

TnC myocytes, which may result in decreased times of shortening and relengthening. If

this were the case then L48Q TnC would be expected to have increased times of

relengthening, but this did not occur. The L48Q TnC myocytes had increased shortening

amplitude at higher frequencies but with the same kinetics as WT. Thus the faster rates

of shortening and relaxation may be due to the increased rates of Ca2+ dissociation and

cross-bridge detachment observed in the D73N TnC myofibrils.

A trend that occurs with increasing frequency is that the Ca2+ transient amplitude

decreased for all of the WT and TnC injected myocytes. At a low frequency the Ca2+

transient was the highest level and all of the mice had their greatest shortening

amplitudes. It could be that the D73N mice were unable to produce a similar maximal

shortening due to the decreased Ca2+ sensitivity TnC. At high frequency where the Ca2+

transient was most decreased, the increased sensitivity L48Q TnC was able to

compensate for the decreased Ca2+ transient to maintain its shortening amplitude. In total, this would provide additional evidence that the phenotype observed in the echocardiograph of the D73N mice was due to the myofilament properties, rather than changes in the Ca2+ levels, but does not explain why there was not a global change for the

L48Q TnC hearts. If anything, it would appear that the L48Q myocytes have enhanced

contractile function.

146

Panel A Panel B

Ca2+ Transient Amplitude Shortening Amplitude 4 WT 8 Cont AAV 3 TnC D73N 6 0 TnC L48Q

F/F 2 4 ∆ % RCL % 1 2

0 0 0.0 0.5 1.0 1.5 2.0 0.0 0.5 1.0 1.5 2.0 Frequency (Hz) Frequency (Hz)

Figure 23. Frequency Dependency of the Ca2+ Transient and Shortening Amplitudes. For each panel the isolated cardiomyocytes from the respective WT and AAV injected mice were stimulated at 0.2, 0.5, 1.0 and 2.0 Hz as shown on the x-axis. Panel A shows the [Ca]i transient amplitude. Panel B shows the % of shortening based on the resting cell length RCL. Panel C shows the [Ca]i transient time to peak CaTTP and Panel D shows time taken for the [Ca]i transient to decline by 50% (CaRT50). Panel E shows the time required for cell shortening (sTTP) and Panel F shows the time for the cardiomyocytes to re- lengthen to 50% of resting cell length (sRT50). The data are presented as mean ± SE for the pooled myocytes from the respective AAV injected mouse hearts (n=2 for each group).

(continued)

147

Figure 23 continued.

Panel C Panel D

2+ 2+ Ca Transients Time to Peak Ca Transients RT50 500 150 400 100 300 WT 200 Time (ms) Time (ms) 50 TnC D73N TnC L48Q 100 0 0 0.0 0.5 1.0 1.5 2.0 0.0 0.5 1.0 1.5 2.0 Frequency (Hz) Frequency (Hz)

Panel E Panel F

Shortening Time to Peak Relengthening RT50

500 500 400 400 300 300 200 200 Time (ms) Time (ms) 100 100 0 0 0.0 0.5 1.0 1.5 2.0 0.0 0.5 1.0 1.5 2.0 Frequency (Hz) Frequency (Hz)

148

4.3.9 Maternal Transmission of AAV-9 D73N TnC to Offspring

An interesting finding that occurred during these studies involved the transmission of the AAV-9 D73N TnC from one litter to another or within a litter. One of the first signs that something strange was occurring was the death of a breeder pair, which was initially chalked up to be a random occurrence. Then un-injected mice from the breeder pairs that previously had litters injected with the D73N TnC began to shown echocardiograph abnormalities. Initially, neonatal pups were injected with the AAV-9

D73N TnC. For some litters, after injection the pups either died or the parents chose to eat them. The mother then gave birth to another litter and these pups were not injected.

Upon doing echoes on these mice at 4 (4 week ■) and 7 (7 week ■) weeks, the systolic and diastolic diameters of the left ventricle were increased as compared to the WT values at 4, 6, and 8 weeks (Figure 24, Panel A and Panel B, respectively). These mice (n=4) did not have any EKG abnormalities, with the internal diameters of the left ventricle being the only indication of anything abnormal. Validation that the dilated ventricles were due to the presence of a desensitized TnC was done by a Western blot with an antibody against the FLAG-tag in the tissue of the heart. The Flag-tag antibody was used to probe for the presence of FLAG-tagged D73N TnC in the atria, left ventricle and right ventricle. Figure 25 shows the presence of the FLAG-tag TnC in the left and right ventricular tissue with a very faint amount in the atria. Purified FLAG-tag TnC that was reconstituted into Tn was used as a positive control (loaded at two different

149 concentrations), as well as the Flag-tag D73N TnC positive myofilament preparation that was shown in Figure 17. There was no indication that FLAG-tag was in the WT mouse tissue from a breeder pair that had never been injected. The mechanisms for how the neonates were transfected with the D73N TnC is not known. These results allude to the possibility that the AAV-9 injected virus may be able to transmit the injected gene of interest from one mouse to another either through genetic insertion into the gametes, through the mother’s milk, or blood to blood transmission. Either way, this finding may need to be considered in current AAV-9 clinical trials as well as future gene therapy studies.

150

Panel A Panel B

Systolic Diameter Diastolic Diameter 4.5 3.25 3.00 4.0 2.75 2.50 3.5

2.25 (mm) (mm) 2.00 3.0 1.75

Diastolic Diameter Diastolic 2.5 Systolic Diameter Systolic 1.50    

4 Week 7 week 4 Week 7 week WT 4 Week WT 6 Week WT 8 Week WT 4 Week WT 6 Week WT 8 Week

Figure 24. Internal Dimensions of Mice that were Maternally Transfected with AAV-9 D73N TnC. The systolic and diastolic dimensions (Panel A and B, respectively) of the heart were determined by echocardiography in mice believed to have been transfected by maternal transmission of the AAV-9 D73N TnC. The dimensions of the unknown mice (believed to be maternally transfected) were measured at 4 and 7 weeks (4 week ■ and 7 week ■) and plotted against the WT values at 4, 6 and 8 weeks.

151

Figure 25. FLAG-tag Western Blot of Maternally Transfected AAV-9 D73N Heart Tissue. Anti-FLAG (DYKDDDDK) antibody was used to detect the Flag- tagged D73N TnC in the whole tissue homogenates of the hearts believed to have been transfected by maternal transmission of the AAV-9 D73N TnC. The TnC FLAG (positive) was from recombinant bacterial purified TnC. The D73N TnC mouse sample was from the myofilament fraction seen in Figure 22. The 8 Week ■ Mouse is whole tissue homogenate from the mice believed to be transfected and was taken from different parts of the heart (LV = left ventricular tissue, RV = right ventricular tissue, Atria = atrial tissue). Gel was ran by Elizabeth Brundage in Dr. Brandon Biesiadecki’s Lab.

152

4.4 Discussion

4.4.1 Transduction of TnC into Cardiomyocytes

To our knowledge, there is no current published literature that has used any virus

or transgenic mouse model to transfect the heart with a TnC containing modified Ca2+ binding properties to study its regulatory role. Lim et al. showed that the function of isolated cardiomyocytes can be altered using adenovirus to transfect the cultured cells with a desensitized TnC [153]. To increase the complexity beyond isolated myocytes, we utilized AAV to transduce the heart with TnCs that had specifically modified Ca2+

binding properties. The AAV-9 serotype has been shown to be a good vehicle for the

transduction of the in-vivo heart [147, 148, 154]. Compared to other serotypes, AAV-9

has a remarkably delayed blood clearance time [154], and may allow for better cardiac

transduction. The mechanism for AAV-9 transduction of cardiac cells is still unknown,

but recent evidence points towards the AAV capsid binding to galactose [155, 156].

There is conflicting data in the literature regarding the exact cardiac transduction

efficiency of the AAV-9 serotype. As seen in most of the previous studies using AAV-9

GFP, the percent of cardiomyocyte transduction varied from ~ 50% at the high end and ~

5% at the lower limit [147, 148, 154, 157, 158]. This variability was dependent on the

route of injection, promoter, titer amount of virus, species of animal, and age. Based on

X-gal (5-bromo-4-chloro-indolyl-β-D-galactopyranoside) staining, rAAV-9 had ~ 50%

transduction efficiency in the hearts of adult mice injected via the tail vein and when

injected subcutaneously there was ~ 25% efficiency (~ 1x1011 drp/animal) [154]. The

153

greatest transduction efficiency (~ 90%, as observed by the expression of LacZ) was

observed with the pseudo-typed vector AAV-2/9, which was injected directly into the

pericardial cavity [147]. Based on our GFP images, we had at most ~ 50% transduction

efficiency but it was not uniform and varied from region to region in the heart. The

regions of higher transfection seemed to be in the myocardium nearer to the apex of the

ventricular chamber. It is unclear why certain regions of the heart were transfected better

than others. It could have been due to the blood flow of one coronary artery being greater

than another, in which case, the transduced area is related to the amount of blood

perfusion. In our study we utilized the CMV promoter. The cardiac specificity of AAV

transduction to the heart can be increased by using a cardiac specific promoter, such as

the myosin light chain promoter fused to the CMV enhancer region [159]. Others have used the chicken β-actin promoter [155, 158] and have fused it with the CMV enhancer region for a more robust cardiac protein expression [147]. To be more cardiac specific with future experiments these methods may need to be employed.

4.4.2 Possible Routes of AAV-9 Transduction to the Heart after IP

Injection

The route of AAV-9 transmission to the heart after intraperitoneal (IP) injection is unclear. Studies looking at the absorption of drugs into the blood stream after IP

injection determined that most of the drugs were reabsorbed through the rich vascular

network of the intestinal tract [160]. This is a possible route for which AAV-9 TnC may

154

enter the blood stream and then target the heart. For this to occur, AAV-9 TnC would have to cross the mesenteric vasculature to enter the blood stream of the hepatic portal vein. The blood would then enter the liver, where the AAV-9 can be cleared from circulation before the venous blood reaches the heart. A more plausible transmission route of AAV-9 to the heart after IP injection is through the lymphatic system. Once injected into the peritoneal cavity, AAV-9 TnC would enter the lymphatic system through the cisterna chyli and into the thoracic duct. The thoracic duct travels upward

and enters the left brachiocephalic vein that empties into the superior vena cava. This

venous blood then enters into the right atrium. This proposed route of transmission

bypasses the initial event of hepatic circulation and the filtering effect of the liver, thereby increasing the chances of myocardial transduction. Throughout this study there were multiple batches of AAV-9 D73N TnC used to inject the mice. The second batch was of a higher titer amount as compared to the first batch. To maintain a consistent titer amount of 5 x 1011 the volume of the second batch was reduced from 100 μl to 50 μl.

Interestingly, the first batch of virus produced the most dilated and arrhythmic hearts out

of all the mice injected. The mice injected from the second batch of virus had dilated

hearts but not to the same degree. One of the variables that came to mind was the volume

of AAV-9 D73N TnC injected into the IP cavity of the neonatal mouse. There is a direct

correlation between the IP pressure and the amount of fluid removed from the IP cavity

by the lymphatic system [161]. Even though the first batch of AAV-9 D73N TnC had a

lower stock titer amount, the volume injected was larger and would have increased the IP

155

pressure to a greater extent than the smaller volume of the second AAV-9 D73N TnC batch. Therefore, the pressure differences could have increased the rate of AAV-9 D73N

TnC uptake and resulted in better transduction of the heart to produce a more dominant

DCM phenotype.

The targets of AAV-9 transduction include the heart, skeletal muscle [148], neuronal cells of the CNS [162], and the epithelium lining the airways and lungs [155].

Through the use of AAV-9 GFP we have shown the ability of the AAV to transduce both

the cardiac and skeletal muscle following IP injection in neonatal mice. Expression of

GFP was more uniform in the skeletal muscle than the cardiac muscle and may be a result

of the route of administration or AAV-9 tissue specificity. AAV-9 has been shown to

also efficiently transduce the outer retinal layer of the eye [157]. Another observation

was that in a couple of the D73N TnC mice there was cloudiness in the eyes with the

possibility of blindness. This was briefly tested by using a pencil to see if the mice

responded to bringing the pencil close to each side of their face. In the eyes that were

cloudy the mice did not move in response to the pencil as compared to the side with a

normal looking eye.

Another concern regarding the injection of the AAV-9 TnCs into the animal

model was whether or not the exogenous TnC would be transcribed, as well as

incorporate into the myofilament. Western blot analysis using α-FLAG-tag, determined

that the exogenous D73N TnC gene was transcribed and incorporated into the

myofilament of the ventricular tissue. Quantification of the amount of FLAG-tagged

156

TnC transcribed inside the myocytes was difficult to measure because of the non-linearity

of the FLAG-tag antibody staining intensity at increasing concentrations of FLAG-tagged

TnC. Future studies may need to use real-time PCR to quantify the amount of the modified TnC mRNA present within the transduced heart.

4.4.3 Functional Alteration of the Intact Heart Due to a Desensitized TnC

Functional studies of the AAV-9 TnC transduced hearts included both

echocardiograph and isolated cardiomyocyte data. Based on the data from the

echocardiograph and EKG, the D73N mice displayed many characteristics of a DCM

animal model. There were large differences in the functional data between the various

D73N TnC injected mice and may be due to the amount of D73N TnC transduced within

the heart tissue. Based on the LV mass calculation determined by the echocardiograph

software there was also an increase in the mass compared to the WT mice. Another

characteristic of DCM is the presence of unexplained ventricular arrhythmias and sudden

cardiac death [128]. Similar to the variable presentation of DCM in humans, some of the

D73N TnC mice displayed irregular and unexplainable ventricular arrhythmias while

others maintained a relatively normal sinus rhythm. In addition, a couple of the D73N

that did not display an abnormal EKG suddenly died without a known cause. This

phenotype was consistent with many different DCM models in human patients. Thus,

some of the D73N mice were not affected to the same degree and maintained normal

157

systolic dysfunction, while others continued to progressively worsen as determined by echocardiography.

4.4.4 Insight into the Effect of a Desensitized TnC on Cardiomyocyte

Performance

The DCM characteristic of enlarged cardiomyocytes was observed during the studies of the isolated D73N TnC cardiomyocytes. In addition to the cardiomyocyte size, the isolated myocyte data provided interesting insight into the effect that altered TnC

Ca2+ sensitivity and dissociation kinetics had on the normal processes of cardiomyocyte function. Similar to a previously studied DCM related desensitized TnC in isolated cardiomyocytes [153], the D73N TnC cardiomyocytes also displayed a marked decrease in the cardiomyocyte shortening, independent of a change in the intracellular Ca2+

handling at lower frequencies (0.25 Hz 0.5 Hz, and 1.0 Hz). This data supports the

hypothesis that the change in shortening amplitude is related to changes in the

myofilament sensitivity, rather than changes in the Ca2+ handling properties of the

myocytes under basal conditions. However, under ISO conditions, the β-AR response as

measured by Ca2+ transient amplitude appears to be altered for the D73N mice. In

addition, the sensitized L48Q TnC cardiomyocytes trended towards an increase in

shortening amplitudes independent of the intracellular Ca2+ handling at higher

frequencies (1.0 Hz and 2.0 Hz). Although these data provided compelling results there

was only an n=2 for each mouse and the experiments will need to be repeated.

158

4.4.5 Maternal Transmission of AAV-9 and Clinical Implications

The transmission of the AAV-9 from litter to parents and then back to a new litter has not been shown before in mice, or any species for that matter. How the AAV-9 was able to pass from one offspring to the next is currently unclear but was due to the parents eating the neonatal mice. The possible modes of AAV-9 TnC transmission could include the passage of the AAV from the neonatal blood through the mucus membranes of the parents. Another proposed route is that once the injected mice were eaten, the AAV was absorbed through the GI tract. Once out of the digestive system, it is unclear whether the

AAV affected the gametes, the in-utero embryos of the new litter, or if the AAV somehow entered the mother’s milk and was passed to the next offspring during feeding of the neonatal mice. If this project is pursued, it would provide important and relevant information about the AAV due to its utilization in many current human clinical trials.

The results might change how AAV-9 is utilized in future clinical trials and provide additional knowledge of how AAV-9 works to transduce the various tissues in the body.

4.5 Limitations and Obstacles

As with any study there were limitations and obstacles to our study. When the

titer of the AAV genomic particles was determined, the real-time PCR machine

determined the number of genomic particles present. The titer amount does not take into

account the number of empty viral capsids that may be present. Thus, when the mice are

injected with a known titer amount, they may also be injected with a large amount of

159

empty capsids that may induce a systemic response [143, 163]. In our AAV-9 TnC injected mice there was no sign of an immune response nor did the animals look lethargic or sick throughout the study. Although this did not seem to be a problem in our study, this may be of importance for future studies and is definitely an area of concern for clinical trials. To circumvent this altogether and reduce the variables, the viral purification method that results in the cleanest preparation will need to be used. Another limitation of this study was inconsistencies in the reproducibility of the animal model.

This was due to the many variables that existed from the point of injection to the measurement of the physiological function. Because the mice were injected IP, the route of transmission to the heart was unclear and was not known if the same amount of virus was able to reach the heart for each and every animal. With each injection there were also varying degrees of leakage from the injection site that could have altered the titer of

AAV-9 particles that were present within the mouse. Alternative injection sites such as the jugular vein, craniofacial vein, injection of the thoracic cavity or direct heart injection may need to be implemented to more directly target the heart [139, 147, 148, 157]. These alternative methods of injection introduce variables of their own and may not be advantageous.

The question initially proposed in the previous chapters was whether or not TnC played a role in rate-limiting relaxation of the heart. To study the role of TnC during relaxation of the intact heart the physiological measurements need to directly measure the rate of relaxation. This was difficult to do in the mouse model when using

160 echocardiography as a tool to measure relaxation. M-mode echocardiography did not directly measure the rates of myocardial relaxation. In addition, it was difficult to detect slight differences in heart function using echocardiography in the mouse due to the sensitivity and variability of the echocardiography technique as well as differences from mouse to mouse. New techniques will need to be implemented to directly measure the rates of myocardial relaxation in the intact heart. Preliminary data collected from LV

Doppler blood flow attempted to measure the isovolumetric relaxtion time. This was done on the same ultrasound machine used to collect the M-mode measurements. Blood flow was monitored as it entered into the LV through the mitral valve and exited the LV through the aortic valve. The isovolumetric relaxation time was calculated from the time that the aortic valve closed and the mitral valve opened. Preliminary results showed that the isovolumetric relaxation time was decreased for the D73N TnC mice as compared to

WT mice. This would indicate an increased rate of relaxation brought about by the desensitized D73N TnC after incorporation into the myofilament. However, the results were variable and highly dependent on the heart rate. Therefore, the heart rate of the mice would need to be controlled in future studies and the measurements repeated.

Finally, maternal transmission of the AAV-9 TnC from one liter to another was an unexpected finding and was not accounted for in the initial experimental design. In our studies each litter was injected with different AAV-9 TnCs as a way to have self- contained internal controls. Future studies will need to separate each litter, and only inject with one type of TnC virus. If the parents eat the pups at any point after injection

161

of AAV-9 TnCs, the breeder pair should not be used for any additional liters. This will

prevent the spread of AAV-9 containing TnCs with drastically different Ca2+ sensitivities.

4.6 Future Directions

To gain a more comprehensive understanding for how a TnC with altered Ca2+

sensitivity affects the physiology of the heart, additional studies will need to be

performed in addition to the cardiac functional measurements. Histopathology studies

will need to examine the tissue using the Hematoxylin and Eosin stain (H&E) [164]. The staining of the cell structures (cytoplasm, nucleus, organelles and extracellular components) will produce diagnostic information related to a disease diagnosis based on the structural organization of the myocytes. Masson’s Trichrome stain can be used to determine if there is a change in the collagenous connective tissue commonly seen with increased fibrosis [165]. Periodic Acid Schiff stain may be used to examine the amount of carbohydrates present [166]. This would be used if the hearts are believed to have metabolic disruption that might lead to an increase in glycogen storage.

To determine the effect of TnC on whole body function a complete blood count

and chemistry profile will need to be performed [162]. The complete blood count determines the number of red blood cells, white blood cells, hemoglobin, hematocrit, platelets, and a total white blood cell count. Information regarding the white blood cells and the specific types (neutrophils, lymphocytes, monocytes, eosinophils, and basophils) can determine if there is an immune response [167]. The blood chemistry profile

162

includes the measurement of sodium, potassium, bicarbonate, creatine kinase, glucose,

calcium, blood urea nitrogen, creatinine, AST, GGT, albumin, globulins, and total protein

[162]. The implications for altered baseline values can determine kidney function, liver function, muscle enzymes due to injury, and levels of inflammation.

Silencing the native TnC gene could be accomplished with short hairpin RNA

(shRNA) [168] if the designed shRNA is able to distinguish between the native TnC and our inserted TnC. Two differences between the two TnCs is that the native TnC contains two Cys residues (C35, C84) and does not have a Flag Tag attached to the C-terminal domain. These differences may be enough for the shRNA to distinguish between the two, and reduce the level of native TnC expression to allow for greater incorporation of the modified TnCs into the myofilament. To better quantify the TnC gene transduction efficiency, a southern blot will need to be performed [169]. In addition, the percent of mutant protein expression will need to be determined, which was unsuccessful with the current FLAG-tag antibody. The quantification of the FLAG-tag TnC could possibly be determined by 2D gel analysis. This method could allow for the separation between the endogenous and exogenous transduced TnC based on the charge differences contributed by the FLAG-tag.

Studying the force-frequency in intact trabeculae while also examining the Ca2+ transients will add an additional level of complexity and understanding to the study.

Using the trabeculae versus the isolated cardiomyocytes will study the cardiomyocytes under a loaded versus an unloaded system, respectively. Using the different AAV-9

163

TnCs, it would be important to see if the altered Ca2+ binding properties of TnC have an

effect on the Ca2+ levels and force production in intact muscle preparations. It will also be important to determine the phosphorylation status of the many different contractile and

Ca2+ handling proteins. The phosphorylation status of TnI, PLB, RyR, IcaL, and MyBP-

C will be important in determining the overall effect that TnC has on the Ca2+ handling

and regulatory mechanisms of heart contraction.

Future echocardiography measurements will need to more directly study

myocardial relaxation. To accomplish this task, Color Doppler or Pulsed Doppler will

need to be used to measure the rates of blood flow in the heart. For example, as blood

flows into the left ventricle during the filling or diastolic phase of the cardiac cycle, the rate at which the blood flows into the ventricle can give important information regarding the relaxation properties of the heart. Measuring the rate at which blood enters the left ventricle (E and A waves) after leaving the left atrium and passing through the mitral valve is an important determining factor of myocardial relaxation. The E wave is formed from the initial passive entry of blood into the left ventricle upon the opening of the

mitral valve and the A wave is additional blood flow due to atrial contraction. The ratio

of these two waves can determine the diastolic properties of the heart and whether or not

there is diastolic dysfunction. If the ventricle is stiff or unable to adequately relax the

E/A wave ratio may change, depending on what degree of diastolic dysfunction is present or what stage of failure the heart is in [9]. One of the difficulties in measuring and comparing the E/A ratio as well as isovolumetric relaxation time is the dependency of

164

these values on the heart rate. The heart rate values will need to be controlled to be

similar between different animals to make a comparison.

An additional method to more directly measure the relaxation properties of the

intact heart is the P-V loop catheter. This method is currently an end-stage/non-survival

surgery where a pressure/volume catheter is inserted through the apex and into the left

ventricle of the heart. This procedure allows the basic cardiac contractile parameters to

be measured (stroke volume, cardiac output, and ejection fraction) as well as dP/dt. From

the changes in pressure the contractility of the heart as well as Tau, the isovolumetric

relaxation constant.

As noted earlier, the GFP expression throughout the heart was variable from

region to region. It would be interesting to see if the contractile parameters for the

different regions are also different after injection of AAV-9 D73N TnC. This could be accomplished through speckle tracking imaging of the heart wall that is able to also be collected on the ultrasound machine [170]. This method will determine whether there are different radial and circumferential strains for different ventricular wall segments. By including the GFP in the D73N viral vector a more direct determination for the location of D73N TnC could be made. The location of the wall strain abnormalities could then be correlated with different regions of the GFP stained heart tissue.

The end goal of this study would be the use of specifically designed TnCs that have a wide range of Ca2+ binding properties to correct the diseased myocardium due to

altered Ca2+ sensitivity. Our lab has previously shown that the altered myofilament Ca2+

165

sensitivity caused my mutations in TnI and TnT can be corrected through the use of a

specifically designed TnCs [70]. Even though TnC is the Ca2+ sensor, many cardiomyopathies are not due to mutations in TnC, but rather the proteins it interacts with. The R192H TnI mutation that leads to RCM was discovered in humans and has since been translated into the mouse animal model with the R193H TnI mutation. The

RCM mouse model has an increased Ca2+ sensitivity of force development and a

decreased relaxation rate, ultimately reducing the cardiac output [48, 125]. The incomplete mechanical relaxation is not due to altered Ca2+ cycling but rather the actin- myosin interaction that is regulated by the Ca2+ binding properties of the thin filament

[48]. Incorporation of a desensitized TnC into the Tn complex containing the TnI

mutation was able to correct the steady-state Ca2+ sensitivity in the thin filament, the

reconstituted thin filament Ca2+ dependent ATPase activity, as well as the Ca2+ sensitivity

of force production in the skinned cardiac trabeculae [70]. This was also done for

additional TnI and TnT mutations that presented with the DCM phenotype (decreased

Ca2+ sensitivity) through the use of a TnC that had an increased Ca2+ sensitivity.

Our hypothesis is that replacing the endogenous TnC in the diseased models with a TnC

construct that has Ca2+ sensitivity opposite to the diseased state may correct the overall

myofilament Ca2+ sensitivity. Thus, by injecting a specific AAV-9 containing the respective modified TnCs it would potentially correct the diseased myofilament Ca2+

sensitivity to improve the contraction and relaxation of the heart.

166

Appendix A: Protein Purification Protocols

HcTnC purification protocol

Day 1

1) Start growing a single colony in 2 ml of LB media + Ampicillin (Amp) at around 11

am.

2) At 4-5 pm, transfer 0.5 ml of the bacteria culture to 4 conical tubes each containing

20 ml of LB media+ Amp, grow overnight

Day 2

1) Transfer 2 tubes of the 20ml of bacteria culture to 1 L of LB media+ Amp. Induce the

expression of protein with IPTG until OD 600 is between 0.6-0.8

2) Grow for 3 more hours.

3) Spin bacteria culture down, 15mins at 8K rpm @ 10oC.

4) Resuspend cells in 35ml resuspension buffer, freeze cells at –80oCs

Day 3

167

1) Thaw the bacteria cells. Sonicate cells, sonicator is set at output 6, duty cycle 60%.

Sonicate 4 times, 2 mins each time.

2) Spin cells down, 30mins, 11.5k rpm @ 4oC.

3) Slowly add 35% solid ammonium sulfate to the supernatant, with constant stirring @

4oC for 1 hour.

4) Spin cells down, 30 mins, 11.5k rpm @ 4oC.

5) Collect the supernatant, freeze at –80oC.

Day 4

1) Thaw the supernatant, add 5 mM Ca2+ to the supernatant, and load on phenyl- sepharose column, wash with buffer A (50mM Tris, pH7.5, 0.35 M ammonium sulfate,

0.5 mM CaCl2) until the elutant is clean.

2) Elute with buffer B (50mM Tris, pH7.5, 0.35M NaCl, 5mM EDTA), collect the fractions and store the concentrated fractions @ -20oC.

Resuspension buffer:

50mM Tris, 2mM EDTA, 1mM DTT, 1mM PMSF, pH7.5.

Wash buffer A:

50mM Tris, 0.3 M ammonium sulfate, 0.5 mM CaCl2, 1 mM DTT, pH7.5.

Elution Buffer B:

50mM Tris, 5mM EDTA, 1mM DTT, 0.35M NaCl, pH 7.5.

168

HcTnI purification protocol

Day 1

1) Start growing a single colony in 2 ml of LB media+Amp (around 11 am)

2) At 4-5 pm, transfer 0.5 ml of the bacteria culture to 4 conical tubes each containing

20ml of LB media+ Amp, grow overnight

Day 2 1) Transfer 2 tubes of the 20ml of bacteria culture to 1 L of LB media+ Amp. Induce the expression of protein with IPTG until OD 600 is between 0.6-0.8

2) Grow for 3 more hours.

3) Spin bacteria culture down, 15mins at 8K rpm @ 10oC.

4) Resuspend cells in 35ml sucrose resuspension buffer, add one pellet of protease

inhibitor and freeze cells at –80oC.

Day 3

1) Thaw and put on ice immediately.

2) Add 1.75ml Triton X-100 to the resuspended cells to break the cell membrane.

3) Add 0.35ml lysozyme (0.1g/ml) to break the cell membrane.

4) Set the tubes on ice for 30 mins.

5) Sonicate cells at 60%, 6 output, 6 times for 20 secs.

169

6) Spin cells 19k rpm, 4oC, 30mins.

7) Dump the supernatant, resuspend in 35ml sucrose resuspension buffer.

8) Sonicate cells at 60%, 6 output, 6 times 20 secs.

9) Spin cells 19k rpm, 4oC, 30 mins.

10) Dump the supernatant, resuspend in 40ml Urea resuspension buffer .

11) Sonicate cells at 60%, 6 output, 6 times 20 secs.

12) Spin cells 11.5k rpm, 4oC, 30 mins.

13) Collect the supernatant; add one pellet of protease inhibitor.

14) Freeze cells at –80oC.

Day 4

1) Load the supernatant onto DEAE-sepharose column, collect the flow through. Besides

the supernatant, load an extra of 50ml of Buffer 3 Loading Buffer .

2) Load the collected flow through the CM-sepharose column, after loading; wash with

the Buffer 3 Loading Buffer loading buffer long enough to get rid of the contaminating

proteins.

3) Starting eluting with Buffer4 Elution Buffer, when the flow through is clean.

4) Combine the high concentration fractions and store at -80oC.

Day 5

170

1) Thaw the fractions from previous day. Dilute the combined fractions 6 fold with

Buffer1 TnC Column Loading and add 5mM Ca2+.

2) Load the fractions onto the TnC affinity column, after loading, wash the column with

Buffer1 TnC Column Loading long enough.

3) When the flow through is clean, start eluting with Buffer2 TnC column elution.

Buffers

TnI (sucrose) resuspension buffer :

50mM Tris HCl, 5mM EDTA, 8% sucrose, 1mM PMSF, 1mM DTT, pH8.0.

Urea- TnI resuspension:

20mM Tris, 1mM EDTA, 1mM PMSF, 1mM DTT,6M Urea, pH8.0.

Buffer 3 Loading Buffer:

20mM Tris, 1mM EDTA, 1mM DTT, 6M Urea, pH8.0.

Buffer4 Elution Buffer:

20mM Tris, 6M Urea, 1mM DTT, 1mM EDTA, 0.3 M NaCl, pH8.0.

Buffer1 TnC Column Loading:

50mM Tris, 6M Urea, 1mM CaCl2, 1mM DTT,pH8.0.

Buffer2 TnC column elution:

50mM Tris, 6M Urea, 5mM EDTA, 1M NaCl, 1mM DTT, pH 8.0. 189

171

HcTnT purification protocol

Day 1

1) Thaw and sonicate 60%, 6 output, 20 sec each until clear.

2) Spin @ 11500 rpm, 4oC, 30 min, and collect supernatant.

3) Add ammonium sulfate slowly from 0% to 30% at 4oC.

4) Let sit 30 min in cold room.

5) Spin @ 11500 rpm, 4oC, 30 min, and collect supernatant.

6) Add ammonium sulfate slowly from 30% to 70% at 4oC.

7) Spin @ 11500 rpm, 4oC, 30 min.

8) Resuspend pellet using ~40 mL DEAE Wash Buffer

50 mM Tris, 6 M Urea, 1 mM EDTA, 1 mM DTT, 0.2 mM PMSF, pH 8.0.

9) Add one pellet of protease inhibitor.

10) Freeze at -80oC.

Day 2

11) Thaw and spin down protein ~10 min.

12) Dilute the conical vial (~40mL) with 200mL buffer.

13) Load the TnT+buffer (~280mL) onto DEAE Sepharose Fast Flow column at ~1.3

mL/min.

14) Wash with DEAE Wash Buffer and set a ~500-700 min gradient.

172

Day 3

15) Measure OD280 and collect good fractions.

Note: Should come out in the second peak, not first.

16) Run gel to confirm TnT.

17) Dialyze overnight against TnT Dialysis Buffer (same as SP Wash Buffer) 50 mM

Na Acetate, 6 M Urea, 1 mM EDTA, 1 mM DTT, pH 5.3.

Day 4

18) Load fractions onto SP Sepharose column.

19) Wash with SP Wash Buffer and set a ~500-700 min gradient, flow rate ~1.3 mL/min.

Day 5

20) Measure OD280, calculate concentration (Ex280=16,500).

21) Dialyze goods fractions against 4.6M Dialysis.

22) Can confirm clean TnT by running a gel.

173

174

Actin purification protocol

Get prepared: Make G-Actin buffer without ATP,

Actin prep- from rabbit white acetone powder

1) Make 4L G-buffer following the table below, and put the buffer in the cold room.

(Make sure the buffer has NO ATP at this step)

2) In the morning start extracting actin, add 100 micro molar ATP (pH 8.0) to the 4 L G-

buffer.

3) Weigh 5.5g (around 5 g) acetone powder in 400ml beaker and add 140 ml G buffer

(~20 ml G buffer/g powder)

4) Stir the acetone powder within the G buffer in the cold room for 1 hour, with occasional mixing of the contents with plastic rod. (Every 15min, because the connective tissue may stick to the stirring bar and impedes stirring, if not stirring well, the powder does not dissolve well but gets very thick and gooey)

5) Filter the extract through 8 layers of cheesecloth into (squeezing hard to remove all liquid) a clean beaker.

6) Add the solution equally to 4 JA20 centrifuge tubes and centrifuge the solution with

JA20 rotor at 4C, for ½ hour at 18,250 RPM~40,000*g . Collect supernatant (sometimes some of the pellet became loose and cloudy-pellet did not pack nicely)

7) Filter through o.45um syringe tip filters using 60ml capacity syringe.

175

8) Total volume was ~75 ml collected into 250ml beaker.

9) Around 1:00pm, take the solution to cold room and add slowly with stirring in order listed (very important) a) 50mM KCl

b) 1mM ATP c) 2mM MgCl2

10) Let sit for ~2 hrs at 4C (cold room) with NO stirring so that F-actin can be formed.

11) As we can see the solution is viscous by watching bubbles slowly rising

12) Add 0.6M solid KCl, adding slowly and mixing with stirring bar in cold room for ½ hour (helps to dissociate actin binding protein such as TM)

13) Place the solution into clean ultracentrifuge tubes, spin at 4C, 35,000 RPM

(150,000*g) for 1 hour.

14) Obtain clear pellets as is expected for F-actin

15) Resuspend the pellets with 50ml G buffer, (~10-15* starting powder weight) with addition of 0.6M KCl, 1mM ATP and 2mM MgCl2.

16) Put slurry into dounce and dounced 25 times with tight pestle to homogenize the actin.

17) Place the solution into clean ultracentrifuge tubes (balanced), spin at 4C, 35,000RPM

(150,000g) for 1hour.

18) Resuspend pellets with 20mL G buffer (~5* starting powder weight).

176

19) Put slurry into dounce and dounce 25 times with tight pestle to homogenize the actin.

20) Put suspension in dialysis bag and place into 1 L G buffer. Exchange two more times

over the next day in 1 L G buffer.

21) On day three, place the actin prep into ultracentrifuge tubes and spin at 4C, 35000

RPM for 1 hr in ultracentrifuge

22) Keep the SUPERNATANT

23) Add to the supernatant in the order listed below, slowly in a conical tube by gentle

twirling as added and then invert to complete mixing: 50mM KCl, 1mM ATP, 2mM

MgCl2.

24) After the contents were mixed, place in refrigerator for ~1.5hrs. NO Stirring.

25) Solution became viscous which means we got F actin.

26) Place solution into the ultracentrifuge and spin at 4oC, 35000RPM for 1hr.

27) Pellets were clear and big.

28) Resuspend the pellet in 7ml of buffer 10mM Mops, 90 mM KCl, 1mM MgCl2,

0.5mM DTT, pH7.0.

29) Place the slurry into a dounce and dounced ~40 times with the tight pestle.

30) Put the suspended actin filaments into a dialysis bag and exchange 3 times in 4L

10mM Mops, 90 mM KCl, 1mM MgCl2, 0.5mM DTT, pH7.0 @ 4oC.

G-actin Buffer: 2mM Tris, 1mM NaN3, 0.1mM ATP, 0.25mM CaCl2, 0.25mM D

177

Tropomyosin purification protocol

Part 1: preparation of Acetone Powder from muscle tissue

(Adapted from Methods in Enzymology V85 1982; L.B. Smillie)

Put the ethanol and acetone in the cold room to cool them down the day before the

experiment.

Steps 1-13 require 2 people working.

1) Mince 500 g of fresh tissue from either skeletal or cardiac muscle in a pre-cooled

stainless steel grinder at 4oC.

Do the following steps (2-12) in cold room.

2) Stir the meat in 500 ml of water for 2-3mins.

3) Let the mixture stand for 20 mins

4) Remove the liquid through 2 layers of cheesecloth.

5) Stir the remaining solid material for 2-3mins in 500 ml of 100% ethanol.

6) Remove the liquid through cheesecloth, residue extracted in 2 L of 50% ethanol, stirred for 2-3 mins. (1st time)

7) Liquid expressed again through cheesecloth, residue extracted in 2L 50% ethanol, stirred for 2-3 mins. (2nd time)

8) Liquid expressed again, residue extracted in the 2 L 50% ethanol for the (3rd time), stirred for 2-3 mins.

9) Liquid expressed, residue extracted in 2 L of 100% ethanol, stirred for 2-3 mins.

178

10) Liquid expressed, residue extracted in residue extracted in 2 L of 100% ethanol,

stirred for 2-3 mins (2nd time)

11) Liquid expressed, residue extracted in residue extracted in 2 L of acetone, stirred for

2-3 mins (1st time)

12) Liquid expressed, residue extracted in residue extracted in 2 L of acetone, stirred for

2-3 mins (2nd time)

13) The liquid is expressed through cheesecloth. The acetone powder (grayish white for

skeletal while light brown for heart tissue) is then spread thinly over a large sheet of filter

paper in a fume hood and allowed to air dry at room temperature.

The dried powder can be stored indefinitely at –20oC. Yield is approximately 60-75 g

from skeletal muscle and 55-60 g from cardiac muscle. If more acetone powder is to be

made, scale up as per their recipe.

Part 2: Purification of tropomyosin from Acetone Powder

1) The acetone powder (100 g) is extracted with 1 L of 1.0 M KCl, 0.5 mM DTT pH

7.0 with 10mM MOPS for 16 hours at room temperature with slow overhead stirring.

2) Express the liquid through 2 layers of cheesecloth, the residue is extracted with 700

ml of the same solution for 2 hours.

The following steps are carried out 4oC,

179

3) The combined extracts are put in an ice water bath, add acetic acid to adjust pH to

4.6, @ 4oC. Let it stir for 30 mins. Initially the solution is reddish brown, and then when

pH is dropping, the color turned milky.

4) Centrifuge at 6000 x g (6.5k RPM JA14 rotor) for 20 mins, should get big reddish

brown precipitate, try to break the precipitate as much as possible using the transfer

pipette, remaining chunks were filtered through 300 μm mesh to break it further. Then let it stir for 20 mins. pH it to make sure the pH is 7.0 with KOH.

5) The precipitate is dissolved in 1 Liter of extraction buffer (pH maintained at 7) by stirring for 20 mins, and insoluble material is removed by centrifugation at 6000 g for 10 mins. Collect the supernatant.

6) The isoelectric precipitation at pH 4.6 and dissolution in extraction buffer at pH 7.0 are repeated twice more, except that the final pH 4.6 precipitate is dissolved in 1.5 L of

0.5 mM DTT in water (maintained pH at 7.0).

7) Solid ammonium sulfate is then added to 53% saturation at 0 degrees, (31.2 g/100 ml) while maintaining the pH at 7.0.

8) After standing for 30 mins, the precipitate is removed by centrifugation at 11,000 x g

(30 min) (9K RPM JA14 Rotor)

9) The supernatant is brought to 65% saturation at 4oC with solid (NH4)SO4 (7.34

g/100 ml), keeping the pH at 7.0 with KOH.

10) The precipitate is collected by centrifugation at 11,000 x g for 30 min and dissolved in 0.1 L of 0.5 mM DTT.

180

11) Dialyzed extensively against 2 mM 2-mercaptoehanol or DTT in water.

12) The yield at this stage is about 1.0-1.2g skeletal Tm and 0.9 g cTm per 100 g of

acetone powder, and the protein can be stored indefinitely in the freeze-dried state.

13) The product at this stage should be checked for purity by SDS-PAGE at several

sample loadings (10-60 μg) in the presence and in the absence of 6M Urea.

Part 3: Chromatography on Hydroxyapatite to remove troponin contamination

1) The protein is dialyzed against the starting buffer: 10mM sodium phosphate or

potassium phosphate, 1M KCl, 0.25 mM DTT, 0.01% NaN3, pH 7.0.

2) A column (2.5*50 cm) is packed with hydroxyapatite (DNA Grade BioGel HTP, Bio-

Rad laboratories), previously rehydrated with starting buffer as described by the supplier.

3) The column is equilibrated with 2 column volumes (450 ml) of starting buffer: 10mM

sodium or potassium phosphate, 1 M KCl, 0.25 mM DTT, 0.01% NaN3, pH 7.0. The

protein, previously dialyzed, is applied to the column as a solution of 10mg/ml.

4) A linear gradient (total volume 2.2 L) to a final concentration of 300 mM sodium

phosphate in starting buffer is applied at a flow rate of 30ml/hr.

5) Tm is eluted at about 150-200 mM phosphate with the troponin components eluted at

100 mM or lower. Up to 600 mg of protein have been routinely purified in a single column run using this procedure.

181

6) The column can be regenerated with 1 column volume of starting buffer made 0.4 M with respect to sodium phosphate and then re-equilibrated with 2-3 column volumes of starting buffer.

182

Appendix B: Purification of Ventricular Myofibrils

Ventricular cardiac muscle was obtained from male New Zealand White rabbits

(2-3-month old, ~ 2 kg weight [88]), male LBN-F1 rats (175-225 g), heartworm free

mongrel dogs (weighing 19.0 ± 0.4 kg; 2–3 year old [88]), and de-identified failing human tissue from ongoing studies with collaborators. All animals and tissues were handled in accordance with the National Institutes of Health Guidelines and approved by the Institutional Laboratory Animal Care and Use Committee at The Ohio State

University. All human tissue was obtained in accordance with a tissue acquisition protocol approved by the Biomedical Sciences Institutional Review Board of The Ohio

State University. All of the heart tissue was procured after the trabeculae were removed from the inside of the ventricular chambers. The hearts were dissected in Krebs

Henseleit solution that contained 137 mM NaCl, 5 mM KCl, 1.2 mM MgSO4, 1.2 mM

NaH2PO4, 20 mM NaHCO3, 10 mM glucose, 0.25 mM CaCl2 and 20 mM 2, 3-

butanedione monoxime (BDM) and had gas (composed of 95% O2 and 5% CO2) bubbled into the solution. The solution contained BDM to prevent contraction and damage to the tissue.

Once the tissue was harvested, the myofibrils were prepared from the fresh tissue

that same day or the tissue was frozen in liquid nitrogen and then stored in a – 80oC

freezer for later use. A standard prep contained ~ 10 to 25g of tissue (4-5 rat hearts, 2-3

183

rabbit hearts, and pieces of tissue from the dog and human). The myofibrils were prepared from the ventricular tissue similar to a protocol previously published (Swartz,

D.R., 1999). 150 ml of Standard Rigor Buffer + Triton (SRB-TX) was made prior to cutting the tissue to allow for enough time for the Triton-X to dissolve and then ~ 25 ml of SRB-TX was placed into the centrifuge tube on ice. The SRB-TX was composed of

75 mM KCl, 10 mM Imidazole, 2 mM MgCl2, 2 mM EGTA, 1 mM NaN3, 1% Triton X

with 0.1 mM PMSF and 1 mM Benzamidine-HCL for the protease inhibitors at pH 7.2.

The tissue was cut into ~ 2 mm long cubes using either a pair of scissors, or a razorblade

and a pair of tweezers. This was done inside of a disposable polyethylene petri dish that

was set on top of ice to keep the tissue cool. If frozen tissue was used then the razorblade

was used instead of the scissors to cut the chunks of tissue to size. Care was taken to

exclude as much fat and connective tissue as possible to minimize the difficulty of

douncing the myofibrils during the later steps of purification. The dissected cubes were

subsequently placed into 25 ml of SRB-TX that was in a 50 ml centrifuge tube on ice.

This volume of SRB-TX allowed the tissue to be added without overflowing the tube and

left enough room for the homogenization.

The cubed tissue was placed on ice for ~ 1hr to allow the Triton-X to dissolve the

myocyte cell membranes. After an hour, the chunks were homogenized on a polytron at

10,000 RPM for 3 bursts of 5s each. In between each 5s burst the myofibrils were placed

back on the ice to prevent the preparation from over-heating due to the violent mixing.

Pieces of tissue that remained in the polytron mixer were added back to the centrifuge

184

tube that contained the myofibrils to help increase yield. Any foam created from the

polytron was removed with a transfer pipette and care was taken to not to suck up any

myofibrils. At the completion of the homogenization the white connective tissue was

removed and thrown away.

The homogenized myofibrils were transferred to a 25 ml dounce and were

homogenized with the loose pestle 20x on ice. The initial dounces were difficult to do

because of the white connective tissue. The myofibrils were homogenized in a bucket of

ice to keep them cold and provide stability for the dounce during the initially difficult

steps. The myofibrils were filtered through a single layer of cheese cloth to remove the

connective tissue, rinsed with ~ 2-3 ml of SRB-TX , and then transferred to into a clean

dounce. The cheese cloth removed a large majority of the connective tissue and made the

following douncing steps easier. This was also the step where the yield of myofibrils was most affected. The myofibril sample was homogenized an additional 20x with the loose pestle and was easier after the connective tissue was removed. The myofibrils were filtered through another single layer of cheese cloth, rinsed with 2-3 ml of SRB-TX, and homogenized with the loose pestle an additional 40x. The myofibrils were transferred to a 50 ml centrifuge tube and pelleted at 2,000 x g for 10 minutes at 4oC.

After centrifugation the supernatant was discarded, the myofibril pellet was

resuspended in 20 ml of SRB-TX using a transfer pipette and then transferred to the

dounce. The myofibrils were homogenized 10x with loose pestle and filtered through

two layers of cheese cloth, rinsed with 2-3 ml of SRB-TX, and transferred to a clean

185 dounce. The myofibrils were homogenized 80x with a tight pestle. A 10 μl aliquot of homogenized myofibrils were placed on a glass slide, covered with glass cover slip, and examined under a 40x objective to examine if the myofibrils were still in chunks or single strands. The myofibrils were mostly single or double stranded. Once the myofibrils were mostly single stranded they were transferred to 50 ml centrifuge tube and collected at

2,000 x g spin for 10 minutes at 4oC.

The supernatant was removed from the myofibrils. The myofibril pellet was re- suspended in 30 ml Standard Rigor Buffer without Triton X (SRB) with a transfer pipette and transferred to a dounce and homogenized 15x with a tight pestle. The myofibril homogenate was transferred to a 50 ml centrifuge tube and collected at same speed, time and temperature as before. The re-suspension of the myofibrils with SRB, homogenization, and centrifuge collection was repeated an additional two times to remove the Triton X from the myofibrils. Before the last spin down, the weight of the 50 ml centrifuge tube was determined. The weight of the empty tube was subtracted from the centrifuged myofibrils after the supernatant was removed to determine the weight of the myofibrils. The myofibrils were re-suspended in three times the volume/weight of

Glycerol Rigor Buffer (GRB). The GRB contained 10 mM MOPS, 75 mM KCL, 2 mM

MgCl2, 2 mM EGTA, 1 mM NaN3, 75% glycerol, 1 mM DTT, 0.1 mM PMSF and 1 mM

Benzamidine-HCL at pH 7.2. The myofibrils in GRB were transferred to a dounce and homogenized 15x with a loose pestle and transferred back into the 50 ml centrifuge tube and stored at -20oC for later use.

186

To check the viability of the myofibrils after purification, the myofibrils were

washed three times in Working Buffer ((WB), 10 mM MOPS, 150 mM KCl, 3 mM

2+ MgCl2, 1 mM DTT and 0.02% Tween 20 at pH 7.0) + 200 μM Ca to remove the

cryogenic glycerol solution. Briefly, 200 μl of glycerol myofibrils were pelleted at 2000 x g for one minute, supernatant removed, and re-suspended in 750 μl of WB by gently pipetting. A 100 μl aliquot of myofibrils was diluted with 100 μl of WB then mixed well by pipetting. 10 μl of diluted myofibrils were placed onto a glass slide and then cover slipped. The myofibrils were examined on a 40x objective to examine the striations of the sarcomeres. The images were captured and printed off with a scale attached (12 mm to 10 μM) to measure the sarcomere length with an average length of 1.7 ± 0.1 μM. 10 mM ATP was added to the edge of the cover slip and allowed to passively diffuse throughout the myofibrils. The viable myofibrils rapidly shortened upon the addition

ATP and formed myofibril balls that were barely visible.

When myofibrils were prepared from dog or human tissue there was a brown layer on top of the myofibrils after centrifugation. This layer was believed to be lipofuscin, which is composed of pigments and oxidized unsaturated fatty acids that have been collected and stored throughout the years in the heart tissue [171]. Due to the age of the dog and human tissue, there was a greater accumulation in this tissue as compared to the rabbit or rat tissue. This layer was removed after each myofibril pelleting by gently scraping off of the top layer with a transfer pipette. It was important to remove this layer

187 before the final homogenization in the glycerol buffer because if present, it made the myofibrils much more sticky and difficult to work with.

188

Appendix C: Tissue Fixation and Imaging for GFP Expression

Drop in fixation:

1) Anesthetize the animal and dissect out the tissue of interest.

2) Place the tissue in 4% paraformaldehyde fixative. In order to get adequate fixation it

is important to, at a minimum, cut the tissue of interest into pieces < 3 mm thick

3) Incubate the pieces in 4% paraformaldehyde fixative for 2-3 hrs with agitation, long

for thicker pieces (3-6 hrs).

4) Rinse in PBS

Cryoprotection before freezing:

After fixation in 4% paraformaldehyde, the tissue is cryoprotected by infiltration in 20% sucrose in buffer with agitation overnight at 4oC. After 24 hrs the tissue block will sink in the solution, indicating they are infiltrated. This infiltration step is critical. If skipped, the tissue will freeze with damaged cells and holes from ice crystals, resulting in tissue sections that will be brittle and might crack.

Freezing:

189

In a beaker that is cooled in liquid nitrogen, add liquid isopentane. The isopentane will

freeze solid white in the beaker that is submerged in liquid nitrogen. The sucrose

embedded tissue is placed into an aluminum cup that is filled halfway with O.C.T. All

aluminum cups were formed with aluminum foil wrapped around a large nut or endcap

(10-15 cm in diameter) and labeled before adding O.C.T. compound. It is important to

submerge the tissue completely on the bottom of the O.C.T. compound so that it is in

contact with the bottom surface. This makes it easier for tissue sectioning on the cryostat. Once the Isopentane is cooled, submerge the aluminum foil cup with O.C.T and tissue sample half way into the Isopentane using large forceps. Keep in liquid until the

O.C.T. turns completely white. Remove samples and store in -80oC freezer in individual

small tubes.

Solutions:

Paraformaldehyde Stock Solution (15%); makes 200ml

1) In a 500ml Erlenmeyer beaker with 100 ml of distilled water, add 30g of

paraformaldehyde powder.

2) In a fume hood, place beaker on hot plate, add stir bar, and warm to 70oC.

3) Add drop-wise 1 M NaOH slowly and the solution will begin to clear. Add up to 1

ml or until the solution dose not clear further. This solution will not be totally clear.

4) Let it cool to room tempertature in the hood. Add 99 ml of distilled water.

5) Filter through Whatman No. 1 filter paper in the hood. The solution should be clear

190

6) Store at 4oC for no more than 1 month.

Phosphate Buffered Saline (PBS); add the following in order to a volumetric flask

beginning with 800 ml of distilled water.

KCl 0.2 g, KH2PO4 0.24 g, NaCl 8.0 g, Na2HPO4 and bring volume to 1000 ml.

Working Fixative Solution 4% Paraformaldehyde in PBS (100 ml):

1) Measure out 73.3 ml of PBS into a beaker.

2) Add 25.7 ml of paraformaldehyde stock solution

3) Check that the pH is 7.3. If the pH is higher adjust with HCl.

4) Use the fixative within 1 day.

0.2M Phosphate Buffer pH 7.3; add in order.

1) Monobasic sodium phosphate (mol wt. 137.9) 2.70 g.

2) Dibasic sodium phosphate 7-hydrate (mol wt. 268.6) 21.45 g. Dibasic sodium phosphate is available in both anhydrous and 7-hydrate, which have very different molecular weights.

3) 500 ml distilled water.

4) Do not adjust the pH; it should be 7.3.

191

Working Fixative Solution 4% Paraformaldehyde, 0.12 M Phosphate Buffer pH 7.3,

Sucrose 60 mM; makes 500 ml.

1) Add 300 ml of 0.2 M Phosphate Buffer (pH 7.3) to a beaker.

2) Add 10.27 g of sucrose.

3) Add 133.5 ml of 15 % paraformaldehyde stock solution

4) Add 41.5 ml of distilled water.

5) Stir until all of the sucrose is dissolved.

6) Check the pH, it should be 7.3. In necessary, adjust the pH.

7) Use the fixative within 1 day.

192

References

1. Maisch, B., et al., Cardiomyopathies: classification, diagnosis, and treatment. Heart Fail Clin, 2012. 8(1): p. 53-78. 2. Chang, A.N. and J.D. Potter, Sarcomeric protein mutations in dilated cardiomyopathy. Heart Fail Rev, 2005. 10(3): p. 225-35. 3. Greenberg, M.J., et al., Cardiomyopathy-linked myosin regulatory light chain mutations disrupt myosin strain-dependent biochemistry. Proc Natl Acad Sci U S A, 2010. 107(40): p. 17403-8. 4. Kushwaha, S.S., J.T. Fallon, and V. Fuster, Restrictive cardiomyopathy. N Engl J Med, 1997. 336(4): p. 267-76. 5. Robinson, P., et al., Dilated and hypertrophic cardiomyopathy mutations in troponin and alpha-tropomyosin have opposing effects on the calcium affinity of cardiac thin filaments. Circ Res, 2007. 101(12): p. 1266-73. 6. Tardiff, J.C., Thin filament mutations: developing an integrative approach to a complex disorder. Circ Res, 2011. 108(6): p. 765-82. 7. Willott, R.H., et al., Mutations in Troponin that cause HCM, DCM AND RCM: what can we learn about thin filament function? J Mol Cell Cardiol, 2010. 48(5): p. 882-92. 8. Towbin, J.A. and N.E. Bowles, The failing heart. Nature, 2002. 415(6868): p. 227-33. 9. Mandinov, L., et al., Diastolic heart failure. Cardiovasc Res, 2000. 45(4): p. 813-25. 10. Lalande, S. and B.D. Johnson, Diastolic dysfunction: a link between hypertension and heart failure. Drugs Today (Barc), 2008. 44(7): p. 503-13. 11. Bers, D.M., Cardiac excitation-contraction coupling. Nature, 2002. 415(6868): p. 198- 205. 12. Delbridge, L.M., J.W. Bassani, and D.M. Bers, Steady-state twitch Ca2+ fluxes and cytosolic Ca2+ buffering in rabbit ventricular myocytes. Am J Physiol, 1996. 270(1 Pt 1): p. C192-9. 13. Bers, D.M., Calcium fluxes involved in control of cardiac myocyte contraction. Circ Res, 2000. 87(4): p. 275-81. 14. Banijamali, H.S., et al., Force-interval relations of twitches and cold contractures in rat cardiac trabeculae. Effect of ryanodine. Circ Res, 1991. 69(4): p. 937-48. 15. Bassani, J.W., R.A. Bassani, and D.M. Bers, Relaxation in rabbit and rat cardiac cells: species-dependent differences in cellular mechanisms. J Physiol, 1994. 476(2): p. 279-93. 16. Brini, M. and E. Carafoli, Calcium pumps in health and disease. Physiol Rev, 2009. 89(4): p. 1341-78. 17. Brittsan, A.G. and E.G. Kranias, Phospholamban and cardiac contractile function. J Mol Cell Cardiol, 2000. 32(12): p. 2131-9. 18. Bers, D.M. and K.S. Ginsburg, Na:Ca stoichiometry and cytosolic Ca-dependent activation of NCX in intact cardiomyocytes. Ann N Y Acad Sci, 2007. 1099: p. 326-38. 19. Gordon, A.M., E. Homsher, and M. Regnier, Regulation of contraction in striated muscle. Physiol Rev, 2000. 80(2): p. 853-924.

193

20. Solaro, R.J., et al., Calcium requirements for cardiac myofibrillar activation. Circ Res, 1974. 34(4): p. 525-30. 21. Endoh, M., Cardiac Ca2+ signaling and Ca2+ sensitizers. Circ J, 2008. 72(12): p. 1915-25. 22. Hunter, P.J., A.D. McCulloch, and H.E. ter Keurs, Modelling the mechanical properties of cardiac muscle. Prog Biophys Mol Biol, 1998. 69(2-3): p. 289-331. 23. Davis, J.P. and S.B. Tikunova, Ca(2+) exchange with troponin C and cardiac muscle dynamics. Cardiovasc Res, 2008. 77(4): p. 619-26. 24. Stehle, R. and B. Iorga, Kinetics of cardiac sarcomeric processes and rate-limiting steps in contraction and relaxation. J Mol Cell Cardiol, 2010. 48(5): p. 843-50. 25. McKillop, D.F. and M.A. Geeves, Regulation of the interaction between actin and myosin subfragment 1: evidence for three states of the thin filament. Biophys J, 1993. 65(2): p. 693-701. 26. Tobacman, L.S., Thin filament-mediated regulation of cardiac contraction. Annu Rev Physiol, 1996. 58: p. 447-81. 27. Spyracopoulos, L., et al., Calcium-induced structural transition in the regulatory domain of human cardiac troponin C. Biochemistry, 1997. 36(40): p. 12138-46. 28. Kawasaki, H., S. Nakayama, and R.H. Kretsinger, Classification and evolution of EF-hand proteins. Biometals, 1998. 11(4): p. 277-95. 29. Li, M.X., X. Wang, and B.D. Sykes, Structural based insights into the role of troponin in cardiac muscle pathophysiology. J Muscle Res Cell Motil, 2004. 25(7): p. 559-79. 30. Tikunova, S.B. and J.P. Davis, Designing calcium-sensitizing mutations in the regulatory domain of cardiac troponin C. J Biol Chem, 2004. 279(34): p. 35341-52. 31. Hitchcock-DeGregori, S.E., Y. Song, and N.J. Greenfield, Functions of tropomyosin's periodic repeats. Biochemistry, 2002. 41(50): p. 15036-44. 32. Borrego-Diaz, E. and J.M. Chalovich, Kinetics of regulated actin transitions measured by probes on tropomyosin. Biophys J, 2010. 98(11): p. 2601-9. 33. Gordon, A.M., M. Regnier, and E. Homsher, Skeletal and cardiac muscle contractile activation: tropomyosin "rocks and rolls". News Physiol Sci, 2001. 16: p. 49-55. 34. Davis, J.P., et al., Effects of thin and thick filament proteins on calcium binding and exchange with cardiac troponin C. Biophys J, 2007. 92(9): p. 3195-206. 35. Tikunova, S.B., et al., Effect of calcium-sensitizing mutations on calcium binding and exchange with troponin C in increasingly complex biochemical systems. Biochemistry, 2010. 49(9): p. 1975-84. 36. James, J., et al., Effects of myosin heavy chain manipulation in experimental heart failure. J Mol Cell Cardiol, 2010. 48(5): p. 999-1006. 37. VanBuren, P., et al., Cardiac V1 and V3 myosins differ in their hydrolytic and mechanical activities in vitro. Circ Res, 1995. 77(2): p. 439-44. 38. Flashman, E., et al., Cardiac myosin binding protein C: its role in physiology and disease. Circ Res, 2004. 94(10): p. 1279-89. 39. Huxley, A.F., Muscle structure and theories of contraction. Prog Biophys Biophys Chem, 1957. 7: p. 255-318. 40. Bloemink, M.J. and M.A. Geeves, Shaking the myosin family tree: Biochemical kinetics defines four types of myosin motor. Semin Cell Dev Biol, 2011.

194

41. Poggesi, C., C. Tesi, and R. Stehle, Sarcomeric determinants of striated muscle relaxation kinetics. Pflugers Arch, 2005. 449(6): p. 505-17. 42. Stehle, R., et al., Insights into the kinetics of Ca2+-regulated contraction and relaxation from myofibril studies. Pflugers Arch, 2009. 458(2): p. 337-57. 43. Janssen, P.M., L.B. Stull, and E. Marban, Myofilament properties comprise the rate- limiting step for cardiac relaxation at body temperature in the rat. Am J Physiol Heart Circ Physiol, 2002. 282(2): p. H499-507. 44. Lionne, C., et al., Evidence that phosphate release is the rate-limiting step on the overall ATPase of psoas myofibrils prevented from shortening by chemical cross-linking. Biochemistry, 2002. 41(44): p. 13297-308. 45. Siemankowski, R.F. and H.D. White, Kinetics of the interaction between actin, ADP, and cardiac myosin-S1. J Biol Chem, 1984. 259(8): p. 5045-53. 46. Brutsaert, D.L. and S.U. Sys, Relaxation and diastole of the heart. Physiol Rev, 1989. 69(4): p. 1228-315. 47. Chemla, D., et al., Mechanics of Relaxation of the Human Heart. News Physiol Sci, 2000. 15: p. 78-83. 48. Du, J., et al., Impaired relaxation is the main manifestation in transgenic mice expressing a restrictive cardiomyopathy mutation, R193H, in cardiac TnI. Am J Physiol Heart Circ Physiol, 2008. 294(6): p. H2604-13. 49. Fitzsimons, D.P., J.R. Patel, and R.L. Moss, Role of myosin heavy chain composition in kinetics of force development and relaxation in rat myocardium. J Physiol, 1998. 513 ( Pt 1): p. 171-83. 50. Gwathmey, J.K., A.I. Yerevanian, and R.J. Hajjar, Cardiac gene therapy with SERCA2a: from bench to bedside. J Mol Cell Cardiol, 2011. 50(5): p. 803-12. 51. Kentish, J.C., et al., Phosphorylation of troponin I by protein kinase A accelerates relaxation and crossbridge cycle kinetics in mouse ventricular muscle. Circ Res, 2001. 88(10): p. 1059-65. 52. Fitzsimons, D.P. and R.L. Moss, Cooperativity in the regulation of force and the kinetics of force development in heart and skeletal muscles: cross-bridge activation of force. Adv Exp Med Biol, 2007. 592: p. 177-89. 53. Massion, P.B. and J.L. Balligand, Modulation of cardiac contraction, relaxation and rate by the endothelial nitric oxide synthase (eNOS): lessons from genetically modified mice. J Physiol, 2003. 546(Pt 1): p. 63-75. 54. Salazar, N.C., J. Chen, and H.A. Rockman, Cardiac GPCRs: GPCR signaling in healthy and failing hearts. Biochim Biophys Acta, 2007. 1768(4): p. 1006-18. 55. Magne, S., et al., Beta(2)-adrenergic receptor agonists increase intracellular free Ca(2+) concentration cycling in ventricular cardiomyocytes through p38 and p42/44 MAPK- mediated cytosolic phospholipase A(2) activation. J Biol Chem, 2001. 276(43): p. 39539- 48. 56. Robertson, S.P., et al., The effect of troponin I phosphorylation on the Ca2+-binding properties of the Ca2+-regulatory site of bovine cardiac troponin. J Biol Chem, 1982. 257(1): p. 260-3. 57. Brouwer, W.P., et al., The development of familial hypertrophic cardiomyopathy: from mutation to bedside. Eur J Clin Invest, 2011. 41(5): p. 568-78. 195

58. Biesiadecki, B.J., et al., The troponin C G159D mutation blunts myofilament desensitization induced by troponin I Ser23/24 phosphorylation. Circ Res, 2007. 100(10): p. 1486-93. 59. Liang, B., et al., Familial hypertrophic cardiomyopathy-related cardiac troponin C mutation L29Q affects Ca2+ binding and myofilament contractility. Physiol Genomics, 2008. 33(2): p. 257-66. 60. Baudenbacher, F., et al., Myofilament Ca2+ sensitization causes susceptibility to cardiac arrhythmia in mice. J Clin Invest, 2008. 118(12): p. 3893-903. 61. Du, C.K., et al., Knock-in mouse model of dilated cardiomyopathy caused by troponin mutation. Circ Res, 2007. 101(2): p. 185-94. 62. Gomes, A.V., et al., Cardiac troponin T isoforms affect the Ca(2+) sensitivity of force development in the presence of slow skeletal troponin I: insights into the role of troponin T isoforms in the fetal heart. J Biol Chem, 2004. 279(48): p. 49579-87. 63. Iorga, B., et al., Lys184 deletion in troponin I impairs relaxation kinetics and induces hypercontractility in murine cardiac myofibrils. Cardiovasc Res, 2008. 77(4): p. 676-86. 64. Kruger, M., et al., Effects of the mutation R145G in human cardiac troponin I on the kinetics of the contraction-relaxation cycle in isolated cardiac myofibrils. J Physiol, 2005. 564(Pt 2): p. 347-57. 65. Davis, J.P., et al., Engineering competitive magnesium binding into the first EF-hand of skeletal troponin C. J Biol Chem, 2002. 277(51): p. 49716-26. 66. Davis, J.P., et al., Mutations of hydrophobic residues in the N-terminal domain of troponin C affect calcium binding and exchange with the troponin C-troponin I96-148 complex and muscle force production. J Biol Chem, 2004. 279(17): p. 17348-60. 67. Lee, R.S., et al., Effect of Ca2+ binding properties of troponin C on rate of skeletal muscle force redevelopment. Am J Physiol Cell Physiol, 2010. 299(5): p. C1091-9. 68. Little, S.C., et al., Measurement of calcium dissociation rates from troponin C in rigor skeletal myofibrils. Front Physiol, 2011. 2: p. 70. 69. Lionne, C., et al., Why choose myofibrils to study muscle myosin ATPase? J Muscle Res Cell Motil, 2003. 24(2-3): p. 139-48. 70. Liu, B., et al., Engineered troponin C corrects disease-related cardiac myofilament calcium sensitivity. J Biol Chem, 2012. 71. Tikunova, S.B., J.A. Rall, and J.P. Davis, Effect of hydrophobic residue substitutions with glutamine on Ca(2+) binding and exchange with the N-domain of troponin C. Biochemistry, 2002. 41(21): p. 6697-705. 72. Gomes, A.V., J. Liang, and J.D. Potter, Mutations in human cardiac troponin I that are associated with restrictive cardiomyopathy affect basal ATPase activity and the calcium sensitivity of force development. J Biol Chem, 2005. 280(35): p. 30909-15. 73. Kobayashi, T. and R.J. Solaro, Increased Ca2+ affinity of cardiac thin filaments reconstituted with cardiomyopathy-related mutant cardiac troponin I. J Biol Chem, 2006. 281(19): p. 13471-7. 74. Bers, D.M., Altered cardiac myocyte Ca regulation in heart failure. Physiology (Bethesda), 2006. 21: p. 380-7. 75. Janssen, P.M. and M. Periasamy, Determinants of frequency-dependent contraction and relaxation of mammalian myocardium. J Mol Cell Cardiol, 2007. 43(5): p. 523-31. 196

76. Grossman, W., Diastolic function and heart failure: an overview. Eur Heart J, 1990. 11 Suppl C: p. 2-7. 77. Backx, P.H., et al., The relationship between contractile force and intracellular [Ca2+] in intact rat cardiac trabeculae. J Gen Physiol, 1995. 105(1): p. 1-19. 78. Brenner, B., J.M. Chalovich, and L.C. Yu, Distinct molecular processes associated with isometric force generation and rapid tension recovery after quick release. Biophys J, 1995. 68(4 Suppl): p. 106S-111S. 79. Eisenberg, E. and T.L. Hill, Muscle contraction and free energy transduction in biological systems. Science, 1985. 227(4690): p. 999-1006. 80. Hibberd, M.G., et al., Phosphate release and force generation in skeletal muscle fibers. Science, 1985. 228(4705): p. 1317-9. 81. Nyitrai, M. and M.A. Geeves, Adenosine diphosphate and strain sensitivity in myosin motors. Philos Trans R Soc Lond B Biol Sci, 2004. 359(1452): p. 1867-77. 82. Simnett, S.J., et al., Effect of pH, phosphate, and ADP on relaxation of myocardium after photolysis of diazo 2. Am J Physiol, 1998. 275(3 Pt 2): p. H951-60. 83. Tesi, C., et al., Relaxation kinetics following sudden Ca(2+) reduction in single myofibrils from skeletal muscle. Biophys J, 2002. 83(4): p. 2142-51. 84. Siemankowski, R.F., M.O. Wiseman, and H.D. White, ADP dissociation from actomyosin subfragment 1 is sufficiently slow to limit the unloaded shortening velocity in vertebrate muscle. Proc Natl Acad Sci U S A, 1985. 82(3): p. 658-62. 85. Solzin, J., et al., Kinetic mechanism of the Ca2+-dependent switch-on and switch-off of cardiac troponin in myofibrils. Biophys J, 2007. 93(11): p. 3917-31. 86. Dong, W., et al., Kinetic studies of calcium binding to the regulatory site of troponin C from cardiac muscle. J Biol Chem, 1996. 271(2): p. 688-94. 87. Hazard, A.L., et al., The kinetic cycle of cardiac troponin C: calcium binding and dissociation at site II trigger slow conformational rearrangements. Protein Sci, 1998. 7(11): p. 2451-9. 88. Xu, Y., et al., Effect of Twitch Interval Duration on the Contractile Function of Subsequent Twitches in Isolated Rat, Rabbit, and Dog Myocardium under Physiological Conditions. J Appl Physiol, 2011. 89. Swartz, D.R., D. Zhang, and K.W. Yancey, Cross bridge-dependent activation of contraction in cardiac myofibrils at low pH. Am J Physiol, 1999. 276(5 Pt 2): p. H1460-7. 90. Gillis, T.E., et al., Investigation of thin filament near-neighbour regulatory unit interactions during force development in skinned cardiac and skeletal muscle. J Physiol, 2007. 580(Pt. 2): p. 561-76. 91. Candau, R. and M. Kawai, Correlation between cross-bridge kinetics obtained from Trp fluorescence of myofibril suspensions and mechanical studies of single muscle fibers in rabbit psoas. J Muscle Res Cell Motil, 2011. 32(4-5): p. 315-26. 92. Trybus, K.M. and E.W. Taylor, Kinetic studies of the cooperative binding of subfragment 1 to regulated actin. Proc Natl Acad Sci U S A, 1980. 77(12): p. 7209-13. 93. Liu, J.X., et al., Myonuclear domain size and myosin isoform expression in muscle fibres from mammals representing a 100,000-fold difference in body size. Exp Physiol, 2009. 94(1): p. 117-29.

197

94. Hamilton, N. and C.D. Ianuzzo, Contractile and calcium regulating capacities of myocardia of different sized mammals scale with resting heart rate. Mol Cell Biochem, 1991. 106(2): p. 133-41. 95. Palmiter, K.A., et al., Kinetic differences at the single molecule level account for the functional diversity of rabbit cardiac myosin isoforms. J Physiol, 1999. 519 Pt 3: p. 669- 78. 96. Belus, A., N. Piroddi, and C. Tesi, Mechanism of cross-bridge detachment in isometric force relaxation of skeletal and cardiac myofibrils. J Muscle Res Cell Motil, 2003. 24(4-6): p. 261-7. 97. Ingwall, J.S., ATP and the Heart2002: Kluwer Academic Publishers. 98. Han, Y.S. and O. Ogut, Force relaxation and thin filament protein phosphorylation during acute myocardial ischemia. Cytoskeleton (Hoboken), 2011. 68(1): p. 18-31. 99. Belus, A., et al., The familial hypertrophic cardiomyopathy-associated myosin mutation R403Q accelerates tension generation and relaxation of human cardiac myofibrils. J Physiol, 2008. 586(Pt 15): p. 3639-44. 100. Martin, H. and R.J. Barsotti, Relaxation from rigor of skinned trabeculae of the guinea pig induced by laser photolysis of caged ATP. Biophys J, 1994. 66(4): p. 1115-28. 101. Hannon, J.D., D.A. Martyn, and A.M. Gordon, Effects of cycling and rigor crossbridges on the conformation of cardiac troponin C. Circ Res, 1992. 71(4): p. 984-91. 102. Martyn, D.A., et al., Ca2+ and cross-bridge-induced changes in troponin C in skinned skeletal muscle fibers: effects of force inhibition. Biophys J, 1999. 76(3): p. 1480-93. 103. Zot, A.S. and J.D. Potter, Reciprocal coupling between troponin C and myosin crossbridge attachment. Biochemistry, 1989. 28(16): p. 6751-6. 104. Sun, Y.B., F. Lou, and M. Irving, Calcium- and myosin-dependent changes in troponin structure during activation of heart muscle. J Physiol, 2009. 587(Pt 1): p. 155-63. 105. Suzuki, T., et al., Effects of cardiac myosin isoform variation on myofilament function and crossbridge kinetics in transgenic rabbits. Circ Heart Fail, 2009. 2(4): p. 334-41. 106. Maughan, D., et al., Work production and work absorption in muscle strips from vertebrate cardiac and insect flight muscle fibers. Adv Exp Med Biol, 1998. 453: p. 471- 80. 107. McCurdy, D.T., et al., Myocardial cross-bridge kinetics in transition to failure in Dahl salt- sensitive rats. Am J Physiol Heart Circ Physiol, 2001. 281(3): p. H1390-6. 108. Kad, N.M., et al., A mutant heterodimeric myosin with one inactive head generates maximal displacement. J Cell Biol, 2003. 162(3): p. 481-8. 109. Kovacs, M., et al., Load-dependent mechanism of nonmuscle myosin 2. Proc Natl Acad Sci U S A, 2007. 104(24): p. 9994-9. 110. Laakso, J.M., et al., Myosin I can act as a molecular force sensor. Science, 2008. 321(5885): p. 133-6. 111. Veigel, C., et al., Load-dependent kinetics of myosin-V can explain its high processivity. Nat Cell Biol, 2005. 7(9): p. 861-9. 112. Veigel, C., et al., Load-dependent kinetics of force production by smooth muscle myosin measured with optical tweezers. Nat Cell Biol, 2003. 5(11): p. 980-6. 113. Kreutziger, K.L., et al., Calcium binding kinetics of troponin C strongly modulate cooperative activation and tension kinetics in cardiac muscle. J Mol Cell Cardiol, 2010. 198

114. Allen, D.G. and J.C. Kentish, Calcium concentration in the myoplasm of skinned ferret ventricular muscle following changes in muscle length. J Physiol, 1988. 407: p. 489-503. 115. Housmans, P.R., N.K. Lee, and J.R. Blinks, Active shortening retards the decline of the intracellular calcium transient in mammalian heart muscle. Science, 1983. 221(4606): p. 159-61. 116. de Tombe, P.P. and G.J. Stienen, Impact of temperature on cross-bridge cycling kinetics in rat myocardium. J Physiol, 2007. 584(Pt 2): p. 591-600. 117. Maron, B.J., et al., Contemporary definitions and classification of the cardiomyopathies: an American Heart Association Scientific Statement from the Council on Clinical Cardiology, Heart Failure and Transplantation Committee; Quality of Care and Outcomes Research and Functional Genomics and Translational Biology Interdisciplinary Working Groups; and Council on Epidemiology and Prevention. Circulation, 2006. 113(14): p. 1807-16. 118. Robertson, S.P. and J.D. Potter, The regulation of free calcium ion concentration by metal chelators. Methods Pharmacol, 1984. 5: p. 63-75. 119. Norman, C., et al., Modulation of the rate of cardiac muscle contraction by troponin C constructs with various calcium binding affinities. Am J Physiol Heart Circ Physiol, 2007. 293(4): p. H2580-7. 120. Guth, K. and J.D. Potter, Effect of rigor and cycling cross-bridges on the structure of troponin C and on the Ca2+ affinity of the Ca2+-specific regulatory sites in skinned rabbit psoas fibers. J Biol Chem, 1987. 262(28): p. 13627-35. 121. Morano, I. and J.C. Ruegg, What does TnCDANZ fluorescence reveal about the thin filament state? Pflugers Arch, 1991. 418(4): p. 333-7. 122. Putkey, J.A., et al., Fluorescent probes attached to Cys 35 or Cys 84 in cardiac troponin C are differentially sensitive to Ca(2+)-dependent events in vitro and in situ. Biochemistry, 1997. 36(4): p. 970-8. 123. Manning, E.P., J.C. Tardiff, and S.D. Schwartz, A model of calcium activation of the cardiac thin filament. Biochemistry, 2011. 50(34): p. 7405-13. 124. Campbell, K., Rate constant of muscle force redevelopment reflects cooperative activation as well as cross-bridge kinetics. Biophys J, 1997. 72(1): p. 254-62. 125. Davis, J., et al., Thin filament disinhibition by restrictive cardiomyopathy mutant R193H troponin I induces Ca2+-independent mechanical tone and acute myocyte remodeling. Circ Res, 2007. 100(10): p. 1494-502. 126. Dong, W.J., et al., Structural kinetics of cardiac troponin C mutants linked to familial hypertrophic and dilated cardiomyopathy in troponin complexes. J Biol Chem, 2008. 283(6): p. 3424-32. 127. Gomes, A.V. and J.D. Potter, Molecular and cellular aspects of troponin cardiomyopathies. Ann N Y Acad Sci, 2004. 1015: p. 214-24. 128. Luk, A., et al., Dilated cardiomyopathy: a review. J Clin Pathol, 2009. 62(3): p. 219-25. 129. Mirza, M., et al., Dilated cardiomyopathy mutations in three thin filament regulatory proteins result in a common functional phenotype. J Biol Chem, 2005. 280(31): p. 28498- 506. 130. Venkatraman, G., et al., Different functional properties of troponin T mutants that cause dilated cardiomyopathy. J Biol Chem, 2003. 278(43): p. 41670-6. 199

131. Chandy, I.K., J.C. Lo, and R.D. Ludescher, Differential mobility of skeletal and cardiac tropomyosin on the surface of F-actin. Biochemistry, 1999. 38(29): p. 9286-94. 132. Shitaka, Y., et al., Kinetics of the structural transition of muscle thin filaments observed by fluorescence resonance energy transfer. Biochemistry, 2004. 43(33): p. 10739-47. 133. Shitaka, Y., C. Kimura, and M. Miki, The rates of switching movement of troponin T between three states of skeletal muscle thin filaments determined by fluorescence resonance energy transfer. J Biol Chem, 2005. 280(4): p. 2613-9. 134. Turner, I., et al., Molecular cardiology in translation: gene, cell and chemical-based experimental therapeutics for the failing heart. J Cardiovasc Transl Res, 2008. 1(4): p. 317-27. 135. Davis, J., et al., Designing heart performance by gene transfer. Physiol Rev, 2008. 88(4): p. 1567-651. 136. Jooss, K., et al., Transduction of dendritic cells by DNA viral vectors directs the immune response to transgene products in muscle fibers. J Virol, 1998. 72(5): p. 4212-23. 137. Negroni, M. and H. Buc, Mechanisms of retroviral recombination. Annu Rev Genet, 2001. 35: p. 275-302. 138. Yang, Y., et al., Cellular immunity to viral antigens limits E1-deleted adenoviruses for gene therapy. Proc Natl Acad Sci U S A, 1994. 91(10): p. 4407-11. 139. Bish, L.T., et al., Adeno-associated virus vector delivery to the heart. Methods Mol Biol, 2011. 807: p. 219-37. 140. Gray, S.J. and R.J. Samulski, Optimizing gene delivery vectors for the treatment of heart disease. Expert Opin Biol Ther, 2008. 8(7): p. 911-22. 141. Wright, M.J., et al., In vivo myocardial gene transfer: optimization and evaluation of intracoronary gene delivery in vivo. Gene Ther, 2001. 8(24): p. 1833-9. 142. McCarty, D.M., Self-complementary AAV vectors; advances and applications. Mol Ther, 2008. 16(10): p. 1648-56. 143. Veron, P., et al., Humoral and cellular capsid-specific immune responses to adeno- associated virus type 1 in randomized healthy donors. J Immunol, 2012. 188(12): p. 6418-24. 144. Nakai, H., et al., AAV serotype 2 vectors preferentially integrate into active genes in mice. Nat Genet, 2003. 34(3): p. 297-302. 145. Nakai, H., et al., Helper-independent and AAV-ITR-independent chromosomal integration of double-stranded linear DNA vectors in mice. Mol Ther, 2003. 7(1): p. 101-11. 146. Samulski, R.J., et al., Targeted integration of adeno-associated virus (AAV) into human chromosome 19. EMBO J, 1991. 10(12): p. 3941-50. 147. Bish, L.T., et al., Adeno-associated virus (AAV) serotype 9 provides global cardiac gene transfer superior to AAV1, AAV6, AAV7, and AAV8 in the mouse and rat. Hum Gene Ther, 2008. 19(12): p. 1359-68. 148. Pacak, C.A., et al., Recombinant adeno-associated virus serotype 9 leads to preferential cardiac transduction in vivo. Circ Res, 2006. 99(4): p. e3-9. 149. Woo, Y.J., et al., One year transgene expression with adeno-associated virus cardiac gene transfer. Int J Cardiol, 2005. 100(3): p. 421-6. 150. Zhu, X., et al., Immunosuppression decreases inflammation and increases AAV6- hSERCA2a mediated SERCA2a expression. Hum Gene Ther, 2012. 200

151. Kozak, M., Point mutations define a sequence flanking the AUG initiator codon that modulates translation by eukaryotic ribosomes. Cell, 1986. 44(2): p. 283-92. 152. Roof, S.R., et al., Effects of increased systolic Ca(2)(+) and phospholamban phosphorylation during beta-adrenergic stimulation on Ca(2)(+) transient kinetics in cardiac myocytes. Am J Physiol Heart Circ Physiol, 2011. 301(4): p. H1570-8. 153. Lim, C.C., et al., A novel mutant cardiac troponin C disrupts molecular motions critical for calcium binding affinity and cardiomyocyte contractility. Biophys J, 2008. 94(9): p. 3577- 89. 154. Kotchey, N.M., et al., A potential role of distinctively delayed blood clearance of recombinant adeno-associated virus serotype 9 in robust cardiac transduction. Mol Ther, 2011. 19(6): p. 1079-89. 155. Bell, C.L., et al., Identification of the galactose binding domain of the AAV9 capsid. J Virol, 2012. 156. Shen, S., et al., Terminal N-linked galactose is the primary receptor for adeno-associated virus 9. J Biol Chem, 2011. 286(15): p. 13532-40. 157. Lei, B., et al., Adeno-associated virus serotype-9 efficiently transduces the retinal outer plexiform layer. Mol Vis, 2009. 15: p. 1374-82. 158. Qi, Y., et al., Selective tropism of the recombinant adeno-associated virus 9 serotype for rat cardiac tissue. J Gene Med, 2010. 12(1): p. 22-34. 159. Muller, O.J., et al., Augmentation of AAV-mediated cardiac gene transfer after systemic administration in adult rats. Gene Ther, 2008. 15(23): p. 1558-65. 160. Weiner, I.M., Mechanisms of drug absorption and excretion. The renal excretion of drugs and related compounds. Annu Rev Pharmacol, 1967. 7: p. 39-56. 161. Stachowska-Pietka, J., et al., Distributed model of peritoneal fluid absorption. Am J Physiol Heart Circ Physiol, 2006. 291(4): p. H1862-74. 162. Foust, K.D., et al., Intravascular AAV9 preferentially targets neonatal neurons and adult astrocytes. Nat Biotechnol, 2009. 27(1): p. 59-65. 163. Foust, K.D., et al., Neonatal intraperitoneal or intravenous injections of recombinant adeno-associated virus type 8 transduce dorsal root ganglia and lower motor neurons. Hum Gene Ther, 2008. 19(1): p. 61-70. 164. Yamamoto, C., et al., Protective effects of statin on cardiac fibrosis and apoptosis in adrenomedullin-knockout mice treated with angiotensin II and high salt loading. Hypertens Res, 2011. 34(3): p. 348-53. 165. Ouyang, J., et al., Utility of desmin and a Masson's trichrome method to detect early acute myocardial infarction in autopsy tissues. Int J Clin Exp Pathol, 2009. 3(1): p. 98- 105. 166. Liang, J.L., et al., Effect of impaired glucose tolerance on cardiac dysfunction in a rat model of prediabetes. Chin Med J (Engl), 2011. 124(5): p. 734-9. 167. Yue, Y., et al., A single intravenous injection of adeno-associated virus serotype-9 leads to whole body skeletal muscle transduction in dogs. Mol Ther, 2008. 16(12): p. 1944-52. 168. Zhu, C., et al., ShRNA-mediated gene silencing of AHR promotes the differentiation of P19 mouse embryonic carcinoma cells into cardiomyocytes. Mol Med Report, 2012. 6(3): p. 513-8.

201

169. Juan, F., et al., The changes of the cardiac structure and function in cTnTR141W transgenic mice. Int J Cardiol, 2008. 128(1): p. 83-90. 170. Du, W.H., et al., Usefulness of Speckle Tracking Imaging to Assess Myocardial Contractility in Intra-Abdominal Hypertension: Study in a Mini-Pig Model. Cell Biochem Biophys, 2012. 171. Jung, T., N. Bader, and T. Grune, Lipofuscin: formation, distribution, and metabolic consequences. Ann N Y Acad Sci, 2007. 1119: p. 97-111.

202