1.. 551.1. -
uh $1..“
,

211.7!
3......

.35..
.. S

3: n
V E 9.1.. i

59-)

 

m ~
:6 .0
.1:

5.31..
hr

((2:91
I.

siolo’s

6 5.2.5::
{£7.31 4‘
.1 8.: A! DA.

2..-! v: I)
Q..!..€ It! In
I" a$ 1.1:]:

 

1 .. I lrifTIuma . A , . , . . , . 5 .1. . .2. L51... . 3 .x.

 

 

 

THESIS

llllllllllllllllllllllllllllHllUlllllllllllllllllllllllllll

 

3 1293 01413

LIBRARY 5
Michigan State '

University

 

 

This is to certify that the

dissertation entitled

THE INTERACTION OF CALMODULIN WITH THE SARCOPLASMIC
RETICULUM CALCIUM CHANNEL PROTEIN FROM NORMAL AND
MALIGNANT HYPERTHERMIA SUSCEPTIBLE SKELETAL MUSCLE

presented by

HSIU-CHING YANG

has been accepted towards fulfillment
of the requirements for

Ph.D. degreein Food Science & Human
Nutrition

 

[1.4m

Major profc or

Date allslec

MS U i: an Affirmariw Action/Equal Opportunin Institution 0-12771

PLACE IN RETURN BOX to remove this checkout from your record.
TO AVOID FINES return on or botoro doto duo.

DATE DUE DATE DUE DATE DUE

.mmwmj
_—:J ——JL_J
—7-—— —-:l
-—|L:;——
—-J|  ——1

-—-l:_—__-:J
l—‘TV—F—j

MSU lo An Affirmative ActichEqul Opportunity IW

 

 

 

 

 

 

 

 

 

 

 

 

THE INTERACTION OF CALMODULIN WITH THE SARCOPLASMIC

RETICULUM CALCIUM CHANNEL PROTEIN FROM NORMAL AND

MALIGNANT HYPERTHERMIA SUSCEPTIBLE SKELETAL MUSCLE
By

Hsiu-ching Yang

A DISSERTATION
Submitted to
Michigan State University

in partial fulfillment of the requirements
for the degree of

DOCTOR OF PHILOSOPHY

Department of Food Science and Human Nutrition

1996

ABSTRACT

THE INTERACTION OF CALMODULIN WITH THE SARCOPLASMIC

RETICULUM CALCIUM CHANNEL PROTEIN FROM NORMAL AND

MALIGNANT HYPERTHERMIA SUSCEPTIBLE SKELETAL MUSCLE
By

Hsiu-ching Yang

The Ca2*-release channel (ryanodine receptor) Of the skeletal muscle sarcoplasmic
reticulum (SR) is modulated by various physiological and pharmacological ligands.
Calmodulin (CaM), a ubiquitous Cay-binding protein, has been demonstrated to play a role
in regulating SR Ca2+ channel activity depending on myoplasmic calcium concentration.
However, there have been no direct binding data on the interaction of CaM with the channel
protein. The major biochemical defect in malignant hypertherrnia (MH), an inherited disorder
Of skeletal muscle, is associated with a point mutation (Arg615Cys) of the Ca2*- channel
protein which is likely responsible for the abnormal Ca2+ release from SR in porcine MH
susceptible (MHS) skeletal muscle. The altered Ca2+ channel activity in MHS SR may results,
in part, from abnormal CaM regulation.

The first Objective Of this project was to define the equilibria of CaM binding to the
Ca2+-release channel in porcine skeletal muscle using fluorescence anisotropy. Our results
demonstrated that there are five CaM-binding sites per channel subunit with the affinities
depending on Ca2+ and Mg2+ concentrations. The binding of CaM to SR Caf” -channel was
modulated by caffeine, an activator Of the Cay-channel activity.

The second Objective was tO test the hypothesis that the altered Cay—channel activity

in porcine MHS SR results, in part, fi‘om abnormal CaM regulation of the Ca2"-channel. This
was examined by determining the binding equilibrium and stoichiometry ofMHS and normal
SR Ca2*-channel with CaM under defined metal ion concentrations. The stoichiometry of
CaM to the channel protein in MHS SR was significantly altered compared to normal SR in
the presence ofEGTA indicating the possibility of abnormal CaM regulation of Ca2*-channel
in MHS SR.

The third objective was to identify CaM-binding sites in Caz”-channel from rabbit
skeletal muscle SR in order to further understand stmcture—function relationship of Ca2+-
channel activity. Two CaM-binding sites, amino acid residues 1333-1508 and 2400-2515, in
the central regions of Ca2*-channel were identified by limited proteolysis combined with
photoaffinity labeling and immunoblotting analysis.

These results provide sufficient evidence that CaM plays an important role in
regulating skeletal muscle SR Cay-channel by binding to specific domains with different

affinities depending on other channel modulators concentrations.

COPyright by
HSIU-CHING YANG
1996

To my parents

ACKNOWLEDGMENTS

I would like to express my sincere gratitude to my advisor, Dr. Gale M. Strasburg, who help
me grow not only as a scientist but also in character. It is absolutely crucial to have his
guidance, advice and support throughout the course of graduate study and preparation of the
dissertation. I gratefully acknowledge the members of my guidance committee: Dr. Alfied
Haug, William Helferich, John Linz, and Robert Merkel, who critically read my manuscripts
and ofl‘ered valuable suggestions. I am indebted to Dr. Linz for generously giving up his time
and equipments to assist in teaching me molecular biology experiments. I also am indebted
to Dr. Helferich for his generosity to let me use equipments in his lab. I am gratefirlly to my
collaborators: Dr. James Michelson at Minnesota University, who provided us MHS SR
vesicles; Dr. Susan Hamilton at Baylor College of Medicine who provided us rabbit SR
vesicles and gave me a wonderfiIl opportunity to finish my last study in her lab; Dr. Yili Wu,
fiom Dr. Hamilton’s lab, who provided us calpain digestion results. Thanks to the past and
present members of Dr. Strasburg lab: Mark Reedy, Christ Burke, Dr. Li-ju Wang, Arti
Arora, Dr. Todd Byrem and Dr. Yu—chen Chang; and coworkers at Dr. Hamilton’s lab for
their friendship and support. Special thanks to Dr. Todd Byrem for reviewing and giving

valuable suggestions for part of my dissertation. I also would like to express my deepest

vi

gratitude to my family and fi'iends for their support and encouragement. Finally, I forever am
indebted to my parents, F eng-O Yang and Ai-Min Yang for their love, support, and

encouragement throughout this graduate study.

vii

TABLE OF CONTENTS

TABLE OF CONTENTS .............................................................................................. viii

LIST OF TABLES .......................................................................................................... xi

LIST OF FIGURES ........................................................................................................ xii
CHAPTER 1

INTRODUCTION AND OBJECTIVES .............................................................................. 1
CHAPTER 2

LITERATURE REVIEW .................................................................................................. 5

2.1 Sarcoplasmic Reticulum in the Skeletal Muscle ............................................................ 5

2.1.1 Sarcoplasmic reticulum and excitation-contraction coupling ......................... 5

2.1.2 Structure of the sarcoplasmic reticulum ....................................................... 7

2.1.3 Components of sarcoplasmic reticulum ......................................................... 9

2.1.4 Physiological role of the sarcoplasmic reticulum in skeletal muscle .............. 11

2.2 The Ca” Release Channel Protein/Ryanodine Receptor of Skeletal Muscle

Sarcoplasmic Reticulum ........................................................................................ 18

2.2.1 Purification and characterization of Ca2*-release channel protein/ryanodine

receptor ................................................................................................... 21

2.2.2 Structure and function of ryanodine receptor (RyR) Cay-channel

protein ..................................................................................................... 23

2.2.3 Regulation of the Ca2*-release channel ........................................................ 24

2.3 Calmodulin, a Versatile Calcium Mediator Protein .................................................... 28

2.3.1 Structure and function of calmodulin .......................................................... 28

2.3.2 The role of calmodulin in regulation of Ca2+ release from channel protein...33
2.3.3 Identification of calmodulin binding domains in Cay-release channel

protein ....................................................................................................... 36

2.4 Malignant Hyperthermia and Porcine Sress Syndromes ......................................... 39

2.4.1 Introduction .............................................................................................. 39
2.4.2 Abnormal sarcoplasmic reticulum Cay-release channel in malignant

hypertherrnia skeletal muscle ..................................................................... 40

viii

CHAPTER 3
CALMODULIN INTERACTION WITH THE SKELETAL MUSCLE SARCOPLASMIC

RETICULUM CALCIUM CHANNEL PROTEIN ......................................................... 44
3.1 Introduction ............................................................................................................ 44
3.2 Experimental procedures ......................................................................................... 47

3.2.1 Materials ................................................................................................... 47
3.2.2 Preparation of calmodulin and its derivatives .............................................. 47
3.2.3 Calmodulin cross-linking ........................................................................... 47
3.2.4 Preparation of saIOOplasmic reticulum vesicles ........................................... 48
3.2.5 Fluorescence anisotropy measurements ...................................................... 48
3.2.6 Calmodulin content of heavy SR vesicles ................................................... 50
3.2.7 Biochemical assays .................................................................................... 50
3 .3 Results .................................................................................................................... 52
3.3.1 Calmodulin content of heavy SR vesicles ................................................... 52
3.3 .2 [3H]Ryanodine binding to SR vesicles ........................................................ 52
3.3.3 Identification of calmodulin-binding proteins in SR vesicles ........................ 52
3.3.4 Titration of Rh-CaM with SR vesicles ........................................................ 57
3.3.5 Ionic strength dependence of the binding of Rh-CaM to SR vesicles .......... 60
3.3.6 [Ca2+] dependence of the binding oth-CaM to SR vesicles ...................... 61
3.3.7 [Mg2*] dependence of the binding oth-CaM to SR vesicles ..................... 66
3 .3 .8 Rh-CaWCa2+-channel protein binding equilibrium ..................................... 66
3.4 Discussion .............................................................................................................. 77
CHAPTER 4

ALTERED CALMODULIN REGULATION OF THE SARCOPLASMIC RETICULUM
RELEASE CHANNEL PROTEIN IN MALIGNANT HYPERTHERMIA-SUSCEPTIBLE

PIGS .............................................................................................................................. 84
4.1 Introduction .......................................................................................................... 84
4.2 Experimental procedures ....................................................................................... 87

4.2.1 Materials ................................................................................................... 87
4.2.2 Preparation of calmodulin and its derivatives ............................................. 87
4.2.3 Photoaflinity labeling OfMI-IS or normal SR Ca” -channel protein with
[”51]-Bz-CaM ........................................................................................... 87
4.2.4 Preparation of MHS and normal skeletal muscle SR vesicles ...................... 88
4.2.5 Fluorescence anisotropy measurements ...................................................... 88
4.2.6 Biochemical assays .................................................................................... 90
4.3 Results ................................................................................................................... 91
4.3.1 Afinity labeling of CaM-binding proteins in MHS or normal SR vesicles ...... 91
4.3.2 [3H]Ryanodine binding to MHS or normal SR vesicles ................................ 91
4.3.3 Titration of Rh-CaM with MHS or normal SR vesicles ............................... 91

4.3.4 Rh-CaWCa2+-channel protein of MHS or normal SR binding equilibrium...94
4.3.5 Ionic strength dependence of the binding of Rh-CaM to MHS or normal SR

vesicles ..................................................................................................... 103
4.3.6 Cafi‘eine dependence of the binding of Rh-CaM to MHS or normal SR

ix

vesicles .................................................................................................... 107
4.4 Discussion ............................................................................................................ 113

CHAPTER 5
LOCALIZATION OF CALMODULIN BINDING DOMAINS IN THE CAZI-RELEASE
CHANNEL (RYANODINE RECEPTOR) OF SKELETAL MUSCLE SARCOPLASMIC

RETICULUM .............................................................................................................. l 18
5.1 Introduction .......................................................................................................... 118
5.2 Experimental procedures ....................................................................................... 123

5.2.1 Materials ................................................................................................. 123
5.2.2 Preparation of rabbit skeletal heavy SR vesicles ........................................ 123
5.2.3 Preparation of CaM and its derivatives ..................................................... 123
5.2.4 Limited tryptic digestion of heavy SR vesicles .......................................... 123
5.2.5 [’”I]-Bz-CaM cross-linking with SR vesicles ........................................... 124
5.2.6 Sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE).124
5.2.7 Calpain digestion of heavy SR vesicles crosslinked with [mn-Bz-CaM ......... 125
5.2.8 Sucrose gradient purification of [”slj-Bz-CaM bound Ca2*-channel proteolytic
polypeptides ............................................................................................. 125
5.2.9 Preparation of samples for amino acid sequencing and Western blots ....... 125
5.2.10 N-tenninal sequencing ........................................................................... 126
5.2.11 Western blots ........................................................................................ 126
5.2.12 Protein Assay ......................................................................................... 126
5.3 Results ..................................................................................................................... 127
5.3.1 Aflinity labeling and purification of trypsin-treated Ca2*-release channel from
skeletal muscle heavy SR membranes ....................................................... 127

5.3.2 Identification of [”51]-Bz-CaM bound tryptic Cay-channel polypeptides from
purified 28 S complex .............................................................................. 127

5.3.3 Purification Of calpain digested Cay-channel labeled with [‘2’I]-Bz-CaM .. 140
5.3.4 Identification of [‘2‘I]-Bz-CaM bound calpain digested Ca2*-channel fragments

by immunoblots ....................................................................................... 147
5.4 Discussion ............................................................................................................... 150
CHAPTER 6
OVERALL CONCLUSIONS AND FUTURE RESEARCH .......................................... 154
CHAPTER 7
BIBLIOGRAPHY ........................................................................................................ 156

Table 3.1.

Table 4.1

Table 4.2

LIST OF TABLES

Equilibrium constants for Rh-CaM interaction with the Ca2*-channel protein
in SR vesicles ............................................................................................ '71
Equilibrium constants for Rh-CaM interaction with the Cay-channel protein
in MHS and normal SR vesicles ............................................................. 102
Stoichiometry of high afinity class of Rh-CaM binding to normal and MHS

skeletal muscle SR Cay-release channel ................................................. 106

xi

Figure 2.1

Figure 2.2

Figure 2.3

Figure 2.4

Figure 2.5

Figure 2.6

Figure 2.7

Figure 2.8

Figure 2.9

Figure 3.1

Figure 3.2

LIST OF FIGURES

Diagrammatic representation of the sarcoplasmic reticulum and T-tubules, and

their relation to the myofibrils of mammalian skeletal muscle ...................... 8
Model of the skeletal muscle triad junction ............................................ 15
Molecular coupling model for the release of Ca2*fi'om SR ......................... 16

Three-dimensional reconstruction of the calcium release channel Obtained from
negatively stained specimens ..................................................................... 21
Surface representation of the 3D structure of ice-embedded Cazi-release
channel in its closed state ......................................................................... 23
Structure of calmodulin ........................................................................... 30
Atomic resolution structures of calmodulin and its complex with skeletal
myosin light chain kinase (MLCK) peptide ............................................... 32
Helical-wheel projection of the calmodulin-binding domain ........................... 34
Schematic representation of calmodulin-binding sites in skeletal muscle SR
Cay-release channel protein ..................................................................... 37
[Mgh] and [Ca2+] dependence Ofaflinity labeling of skeletal muscle heavy and
light SR with [’251]-Bz-CaM ..................................................................... 53

Titration oth-CaM with skeletal heavy and light SR vesicles under different

xii

Figure 3 .3

Figure 3.4

Figure 3.5

Figure 3.6

Figure 3.7

Figure 3.8

Figure 4.1

Figure 4.2

Figure 4.3

Figure 4.4

Figure 4.5

Figure 4.6

Figure 4.7

divalent ion conditions ............................................................................. 58

[KCl] dependence of the binding of Rh-CaM to SR vesicles under difi'erent

divalent ion conditions .............................................................................. 62
Titration of Rh-CaM/SR vesicles with CaCl2 ........................................... 64
Titration of Rh-CaM/ SR vesicles with MgCl2 ........................................... 67

Titration of skeletal heavy SR vesicles with Rh-CaM in the presence of

EGTA ...................................................................................................... 69
Titration ofskeletal heavy SRwitth—CaMinthepresence ofCaClz ........... 73

Titration of skeletal heavy SR with Rh-CaM in the presence of CaCl2 plus
MgCl2 ...................................................................................................... 75
Affinity labeling ofMHS and normal skeletal muscle heavy SR with [mu-B2-
CaM ......................................................................................................... 92
Titration oth-CaM with MHS and normal skeletal heavy SR vesicles under
different divalent ion conditions ............................................................... 95
Titration of skeletal heavy SR vesicles with Rh-CaM in the presence of
EGTA ..................................................................................................... 98
Titration of skeletal heavy SR with Rh-CaM in the presence of CaC l2 ........ 100
Titration of skeletal heavy SR with Rh-CaM in the presence of CaCl2 plus
MgCl2 ..................................................................................................... 104
KCl dependence of the binding of Rh-CaM to SR vesicles under different
divalent ion condition ............................................................................. 108

Caffeine dependence of the binding Oth-CaM to HSR vesicles ............ 111

xiii

Figure 5.1

Figure 5.2

Figure 5.3

Figure 5.4

Figure 5.5

Figure 5.6

Figure 5.7

Figure 5.8

Figure 5.9

Autoradiography of [‘z’fl-Bz-CaM bound trypsin-treated heavy SR
membranes fi'om skeletal muscle ............................................................ 128
Sucrose gradient profile of 28 S complex of the CHAPS purified [‘Z’I]-Bz-
CaM labeled Cay-release channel ........................................................... 130
Identification of the 28 S complex of the [‘2’I]-Bz-CaM bound Cay-release
channel ................................................................................................... 132
Fragmentation map of 28 8 complex oftryptic SR CA2+ release channel from
rabbit skeletal muscle ................................................................................. 137
[‘2’I]-Bz-CaM binding tryptic fragments within C’a’ -release channel were
recognized and sequenced ...................................................................... 138
Sucrose gradient profile of calpain digested Ca2*-release channel labeled with
[‘“fl-Bz-CaM ............................................................................................ 141
SDS-PAGE and autoradiography of sucrose gradient purified calpain digested
Cay-release channel labeled with [’2’I]-Bz-CaM .................................... 143
Calpain proteolytic fragments of Ca2*-release channel identified by

immunoblots ......................................................................................... 145

CaM-binding sites in the rabbit skeletal SR Cay-release channel ............... 149

xiv

CHAPTER 1

INTRODUCTION AND OBJECTIVES

Contraction of skeletal muscle is triggered by the release of Ca2+ from the
sarcoplasmic reticulum (SR); this process is coupled to the depolarization and repolarization
Of the transverse tubular (T-tubule) membrane. Communication of voltage changes between
T-tubule and SR occurs at the triad junction where "foot” proteins, also known as ryanodine
receptor or Cay-release channel protein, span the gap between the two membrane structures.

Purified ryanodine receptor protein has four identical subunits of relative molecular
mass Of 565,000. The channel properties and subcellular distribution of the ryanodine receptor
suggest its involvement in the SR Ca2+ release that occurs in skeletal muscle excitation-
contraction (E-C) coupling. Regulation of the SR Ca2*-release channel includes numerous
physiological and pharmacological molecules such as channel activators: uM Ca”, adenine
nucleotides, cafi‘eine, nM ryanodine; and channel inhibitors: Mg”, mM Ca2+, uM ryanodine,
calmodulin (at uM [Ca2*]) which can modify channel gating properties and thereby alter the
sensitivity of the channel in response to the change of T-tubule membrane potential (Meissner
et al., 1986).

One of these channel activity modulators is calmodulin (CaM), an acidic protein

2

composed of four homologous Cay-binding domains, mediates Ca2+ stimulation of numerous
cellular processes. Initial studies on the role of CaM in Ca2+ regulation show that Ca2+ release
from heavy SR vesicles is partially inhibited by CaM in the presence of Ca2+ via a direct
interaction with the Ca2+ channel. Recently, it was reported that CaM can also activate Ca2+
release from SR when myoplasmic [Ca2+] is < 0.1 uM (Tripathy et al., 199). The Ca2+ -
channel modulators may also be involved in altering the binding of CaM to the channel
protein. In the absence of detailed CaM and channel protein binding data, it is difiicult to
assess the overall role of CaM in regulation of Ca2*-channel activity. Therefore, the first
Objective of this study was to examine the effects of some channel modulators such as Ca”,
Mg”, KCl, and cafl‘eine, on CaM/Ca 2* -channel binding interaction to define further the
physiological role of CaM in channel regulation.

Malignant hyperthermia (NIH) is an inherited myopathy in which skeletal muscle
contracture with attendant hyperrnetabolism and elevation in body temperature are triggered
by inhalation of anesthetics and skeletal muscle relaxants. In swine homozygous for the
defect, MH can also be triggered by stress; thus, the disease in the affected animals is also
referred to as porcine stress syndrome (PSS). Major economic losses in the swine industry
result from the development of pale, soft, exudative (PSE) pork that arise fiom postmortem
manifestation of the disease in MB susceptible (MHS) pigs. Because Ca2+ is the main
regulator of muscle contraction and metabolism, the defect in MH was believed to lie in Ca2+
regulation There is now considerable evidence that the primary biochemical defect in MB is
associated with an abnormal SR Cay-release mechanism (Mickelson et al., 1988). A single

point mutation (C1843 to T1843) in the porcine skeletal muscle rynodine receptor gene (ryr

3

1) has been identified (Fujii et al., 1991). This mutation, together with phenotypic indicators
demonstrate that this single amino acid alteration is the causal abnormality for MH. However,
direct evidence concerning the functional role of the mutated Cazi-channel protein and the
possible role of altered regulation by channel modulators in altering Ca2+ release properties
of the channel protein in MB is absent. To address this question, the second objective of this
study was to test the hypothesis that altered Ca2"-channel activity present in SR from MHS
swine results in part, from altered CaM binding to and regulation of the Ca2*-channel.

Cloning, sequencing and functional expression of cDNAs encoding the Ca2*-channel
have provided some clues regarding structural-fimctional relationships within the normal and
MHS Ca2*-channel protein. The localization of binding sites for various modulators including
CaM on Cali-channel has been attempted by prediction methods using the primary and
secondary structure algorithms (Takeshima et al., 1989; Zorzato et al., 1990). CaM binding
domain candidates have been identified by several groups using the ligand overlay method on
rabbit skeletal muscle channel protein fragments expressed as fusion proteins. These results
are questioned by the fact that the fusion protein containing the channel protein fragment may
not refold like that of the intact protein (Menegazzi et al., 1994; Chen et al., 1994).
Furthermore, the fact that up to nine CaM-binding sites have been identified by the two
laboratories using this method, and the fact that there is little agreement on identity of sites,
together suggest identification of sites must be done in the native protein. The roles of specific
CaM-binding sites as activators or inhibitors is important in our understanding of the

mechanism of Ca2+ regulation in health and disease.

4
In summary, the objectives of my research are as follows:

1. To determine CaM/channel protein equilibrium dissociation constant (K) and
binding capacity (Bum) under defined conditions in normal skeletal muscle.

2. To test the hypothesis that altered Ca2+ channel activity present in SR from MHS
swine results in part, from altered CaM binding to and regulation of the Ca2+-
channel.

3. To localize CaM binding domains in native skeletal muscle SR Ca2*-release

channel.
The dissertation addresses these three objectives and is organized in a series of chapters. The
first two sections include the Introduction and Literature Review for the entire dissertation.
Each study was organized as a manuscript with its specific Introduction, Experimental
Procedure, Results and Discussion. The last sections were the Overall Conclusions and

Bibliography for the entire dissertation.

CHAPTER 2

LITERATURE REVIEW

2.1 Sarcoplasmic Reticulum in the Skeletal Muscle

2.1.1 Sarcoplasmic reticulum and excitation-contraction coupling

Muscle contraction and relaxation are regulated by the myoplasmic free calcium
concentration, which in turn depends on appropriate communication between two membrane
systems, the sarcolemma/transverse-tubule (T-tubule) membranes and the sarcoplasmic
reticulum (SR) membranes in myofibrils. The actual role Of Ca2+ in the excitation-contraction
(E-C) coupling mechanism in skeletal muscle was not completely understood until discovery
of the Ca2*-dependent regulatory proteins in the late 19603 (Ebashi and Endo, 1968).

In the late 1940s, it was known that depolarization of the muscle cell membrane
induced an influx of some substances necessary to induce contraction and these substances
were recognized to be released fiom an internal source so that a lesser period of time elapsed
between excitation and contraction (For review see Entman, 1986). Several important

experiments in the 19508 implicated the involvement of the SR in muscle contraction. Marsh

6

(1951, 1952) discovered that homogenized muscle, which was normally shrunken by ATP,
could be induced to swell by addition of ATP plus a muscle extract. This swelling
corresponded to the muscle relaxation process and the relaxing effect induced by the muscle
extract was subsequently demonstrated with muscle fibers (Bendall, 1952,1953; Fujita, 1954).
The relaxing factor ('Marsh' factor) was isolated and identified as the SR Huxley (1957) and
Huxley and Taylor (195 8) demonstrated that microelectrode application of depolarizing
potassium solutions at the region of the T-tubular system resulted in only localized
contraction of the adjacent hemisarcomcres. Podolsky and Constantin (1964) ionophoretically
injected Ca” at the triad region which also resulted in hemisarcomcre contraction, suggesting
that calcium was the important link between excitation and contraction.

The role of calcium in mediating actin-myosin interaction during muscle contraction
was first demonstrated by Ebashi and Ebashi (1964) who demonstrated that a protein
component complexed to tropomyosin imparted calcium sensitivity to reconstituted actin-
myosin. In the period between 1968 and 1970, this factor was demonstrated to be troponin
(Ebashi et al., 1968); the mechanism by which troponin and tropomyosin regulate actin-
myosin interaction has been studied in considerable depth since then. The concentration of
Ca2+ required to elicit muscle contraction was consistent with the affinity of troponin for Ca”.
This strongly implicated calcium-induced conformational changes in troponin as the major

modulator of actin-myosin interaction and contractility.

7

2.1.2 Structure of the sarcoplasmic reticulum

Skeletal muscle in higher animals consists of bundles of long fibers or cells. Each
muscle fiber has numerous myofibrils which are composed of linear arrays of repeating
sarcomeres that run the length of the fiber. The cylindrical sarcomeres form the structural
units of muscle contraction. Each sarcomere is surrounded by a sarcotubular membrane
system, a sleeve-like network consisting of invaginations of the sarcolemma (plasmalemma
of muscle cells) called T-tubules, and the SR which is nestled between the T-tubules (Fig.
2.1). This SR/T-tubule membrane system is directly responsible for regulating the Ca2+
concentration in the immediate vicinity of the muscle filaments (for reviews see Franzini-
Armstrong, 1980; Martonosi, 1984).

In skeletal muscle, the total SR complement is broadly divided into two morphologic
categories: junctional SR which is closely apposed and attached to either the sarcolemma or
its derivative, the T-tubule; and nonjunctional SR which does not make a physical connection
with other membrane systems. The junctional SR region that is tightly apposed to the T-
tubule forms large dilated sacs commonly termed the terminal cistemae. Since most skeletal
muscles display a triple structure, i.e., terminal cistema-T-tubule-terrninal cistema, Porter and
Palade (1957) referred to this anatomical structure as a triad, a term which still persists. The
gap between the junctional SR and T-tubule is about 15 nm. The junctional gaps are bridged
by regularly spaced, densely staining structures termed foot structures or junctional feet. In
mammalian skeletal muscle, the SR/T-tubule junctions invariably occur at each A-I band

region. Freeze-fractured preparations of intact skeletal muscle have revealed the presence of

Myofilaments E<———-One Sarcomerc ——>-: Triad
I

 

‘ \\ >.\\V‘S‘r
“13:22:”
M ofibrils Sarcolemma
y Fenestmted Mitochondria
Terminal Transverse °° ar Longitudinal
cistemae tubule tubule

Figure 2.1 Diagrammatic representation of the sarcoplasmic
reticulum and T-tubules, and their relation to the myofibrils

of mammalian skeletal muscle (From Judge et al., 1989).

9

30 nm "dimples" or indentations usually as a single row in membranes of the tenninal
cisternae (Rayns et al., 1975; Beringer, 1976). There is a definite correlation between number
of indentations in terminal cisternae and increasing speed of muscle contraction. Evidence
suggests that charge movement and Ca2+ fluxes are related to density of indentations
(Dulhunty and Valois, 1982; Dulhunty and Valois, 1983; Dulhunty et al., 1983). Compared
to the degree ofspecialimtion in junctional SR, the nonjunctional SR is considerably simpler.
The longitudinal tubules connect medially with two terminal cisternae, forming contiguous

SR compartments.

2.1.3 Components of sarcoplasmic reticulum

A combination of differential centrifugation and isopycnic zonal ultracentrifugation
methods is used to separate SR vesicles into light and heavy SR fractions. Light SR vesicles
are obtained fiom the 30.32.5% zone of sucrose concentration, whereas heavy SR vesicles
are Obtained fi'om the 38.5-42% zone of sucrose gradients. F reeze-fracture replicas of the
light SR vesicles show an asymmetric distribution of intramembranous particles with the same
orientation and distribution as the longitudinal SR in viva (Campbell et al., 1980). Heavy SR
vesicles appear as rounded vesicles of uniform size filled with electron dense material, similar
to that seen in the terminal cisternae of the SR (Campbell et al., 1980). Biochemical
characterization of light and heavy SR vesicles demonstrated that heavy SR contains greater
than six times the calcium content of light vesicles, and the rate of passive Ca2+ efllux from

the heavy vesicles is double that of light vesicles. The biochemical and morphological data

10

strongly support the view that the light vesicles are derived from the longitudinal SR and that
the heavy vesicles are derived from the terminal cisternae (Campbell et al., 1980).

The SR comprises about one-third phospholipid and neutral lipids and two-thirds
protein (Meissner and Fleischer, 1971). There are several major proteins in the SR and
enzymatic activities that may or may not be related directly to the Ca2+ regulatory mechanisms
of the membrane. Sodium dodecyl sulfate gel electrophoresis shows that the light SR contains
predominantly (80-90%) Ca-ATPase or Ca-pump protein and 5% of a 53 kDa protein
(MacLennan and Wong, 1971). The latter is a high affinity Cazfibinding protein within the
lumen of the SR (MacLennan and Wong, 1971). The heavy SR fiaction contains 50% Ca-
ATPase, 25% calsequestrin, 5% 53 kDa protein, 3% each of 30 and 34 kDa proteins, and 2%
Ca2+ release channel protein (Campbell and Machnnan, 1981).

The Ca” and Mg" -dependent ATPase enzyme, a single polypeptide of 110 kDa,
carries out the enzymatic fimction of Ca” transport in skeletal muscle SR (de Meis and
Vianna 1979). Immunofluorescence and irnmunoferritin labeling techniques were used to
show that the CaZI-ATPase pump is localized throughout the longitudinal SR and
nonjunctional regions of the terminal cisternae, but is absent fi'om the junctional region of the
terminal cisternae, the region apposed to the T-tubule (de Meis and Vianna 1979).

Two glycoproteins of 53 kDa and 160 kDa are transmembrane proteins, intrinsic to
the SR membrane. Calsequestrin is a major SR protein, accounting for about 7% of the total
membrane protein. It binds nearly 1000 nmol of Ca2+ per mg protein with a dissociation
constant of about 1 mM (MacLennan et al., 1983). Calsequestrin is luminally located, mostly

in the terminal cisternae and it has been postulated that it acts to sequester Ca2+ in the interior

11

of the SR (MacLennan et al., 1983)

Approximately 20% of the terminal cisternae consists ofjunctional face membrane
which contains the foot structures spanning the gap between the apposed SR and T-tubules.
The feet have an unusually large size (2,000 kDa per foot protein) and characteristic shape
which allowed their direct identification with the large spanning proteins (Kawarnoto et al.,
1988; Kawamoto et al., 1986) or ryanodine receptors (Block et al., 1988; Inui et al., 1987a,b;
Kawamoto et al., 1988) that constitute the channel responsible for release of calcium fiom
the SR (for review see McPherson and Campbell, 1993). Thus, the terms "junctional foot

protein", "ryanodine receptor", and SR "Caz‘lrelease channel protein" are synonymous.
2.1.4 Physiological role of the sarcoplasmic reticulum in skeletal muscle

The SR has a central role in regulating Ca2+ homeostasis in the muscle cell. This
function may be divided into three activities: Ca2+ uptake, Ca2+ storage, and Ca2+ release
during a contraction and relaxation cycle.

1) Calcium uptake

The SR reduces the calcium concentration within the myofibrillar space to values
sufficiently low (<10'7 M) to allow and to maintain relaxation of muscle (see Weber and
Sanadi, 1966). The Cay-pump protein is responsible for translocating Cri” against the
concentration gradient fiom the myoplasm to the lumen of the SR. The energy is provided by
ATP hydrolysis; two Ca2+ ions are tramloeated per molecule of ATP hydrolyzed (MacLennan

and Holland, 1975; Tada et al., 1978;1kemoto, 1982).

12

The Ca2+ pump protein has been extensively studied in terms of structure, enzyme
kinetics, and thermodynamics by measuring Caz” uptake, by isolating SR vesicles and by

characterizing kinetic events in the purified protein (review in Fleischer and Inui, 1989).

2) Calcium Storage

The SR is the sole source of calcium responsible for muscle contraction during
activation of E—C coupling in skeletal muscle. Ca2+ taken up by the C? pump protein is
stored within the SR during relaxation until an electro chemical signal from the T-tubule
causes its release from the SR junctional terminal cisternae for contraction (Sorrrlyo et al.,
1981). The Ca2+ storage function of SR is mainly attributed to calsequestrin which is
localized in the luminal space of the terminal cisternae of skeletal muscle SR (Meissner et al.,
1973). Calsequestrin from fast-twitch skeletal muscle has molecular mass of 65,000 estimated
from primary structure (Campbell et al., 1983). The most important property of the
calsequestrin is its high capacity for binding Ca2+ with moderate affinity. Calsequestrin fi'om
skeletal muscle can bind 40-50 Ca2+ ions per molecule with a dissociation constant of ~1 mM
in isotonic salt (Maurer et al., 1985; Meissner, 1973). By binding Ca”, calsequestrin performs
the important fimction of keeping the fiee Ca2+ concentration within the SR lumen low, since
mM free Ca2+ concentrations inhibit the calcium pump. The moderate afiinity of calsequestrin
for Ca” (Kc,=mM) is also important for rapid dissociation of Ca2+ for its release fi'om the SR

lumen upon E-C coupling (MacLennan and Wong, 1971).

3) Calcium Release

13

Release of Ca2+ fiom the SR via the Caz"-release channel protein is under the control
of the membrane potential across the T-tubule. In striated muscle, rapid release of Ca2+ from
SR is initiated by a surface membrane action potential that is communicated to the SR from
the T-tubule via the "junctional feet”. The latter structures are now known as Ca2+ release
channels (for review see Meissner, 1994).

Based on previous reports, a general model for E-C coupling in striated muscle has
emerged as follows: depolarization of the T-tubule membrane by a nerve impulse triggers a
charge movement at the T-tubule/SR junctional structure, which results in the release of
calcium from the SR membrane into the myoplasm. The myoplasmic Ca2+ concentration
increases from resting levels of < 10'7 M to ~10" M. Calcium binding to troponin changes the
structure of the thin filaments and facilitates the interaction of myosin with actin, with the
subsequent activation of the cyclic cleavage of ATP and the development of contractile
tension. The general applicability of this model suggests several important prOperties of the
SR which bear directly on the control characteristics of E-C coupling. In this process, three
events must take place: detection by a voltage sensor of changes in potential across the T-
tubule membrane; transmission of the depolarization signal to the SR; and release of calcium
from the SR. All elements necessary for these events are contained within an E-C unit or
triad.

Morphology studies suggested that a specific transmission of the depolarization signal
to the SR was necessary for E-C coupling (Franzini-Armstrong, 1970). Block et a1. (1988)
suggested that there is a direct interaction between the Ca2+ channel-foot proteins in the SR

membrane and a protein component of the T-tubule membrane referred to as the

14

dihydropyridine receptor (DHPR) which serves as a voltage sensor in muscle. Biochemical
evidence demonstrates that Ca2+ channel and DHPR are present in the triad membrane
preparation in a complex, thus supporting a model of direct physical linkage ofDHPR and
Ca” channel in EC coupling (Fig. 2.2) (Marty et al., 1994b). How signal transmission occurs
at the T-SR junction is one of the major unsolved problems of muscle biology. Two major E-
C coupling mechanisms in muscle have been proposed.

1) Molecular coupling model. Based on early studies, it was proposed that in response
to an action potential, a movement of membrane-bound electrical charge, termed a gating
charge, from the cytosolic surface to the extracellular surface of the T-tubule membrane
would initiate release of accumulated Ca2+ into the myoplasm by the Cay-release channels in
the junctional region of SR (Fig. 2.3) (Chandler et al., 1976; Schneider, 1981; Schneider and
Chandler, 1973). This mechanical interaction hypothesis for E-C coupling may involve
modulation by additional interacting, coupling, or linking proteins. (Block et al., 1988;
Fleischer and Inui, 1989; Kim et al., 1990; Caswell et al., 1991; Marty et al., 1994).

2) Diffusible chemical-transmitter hypothesis. The alternative to direct mechanical
regulation of the SR Cay-release channel by the T-tubule voltage sensor is that a difiirsible
chemical transmitter is responsible for signal transmission from T-tubule to SR. The calcium
ion has long been considered as a candidate for a diflirsible transmitter. There is
morphological evidence to suggest that, at least in some skeletal muscles, only a
subpopulation of Ca” release channels are mechanically linked to T-tubule sensors (F ranzini-
Armstrong and Jorgensen, 1994). Ca2+ ions released by these channels could serve to amplify

further SR Ca2+ release by opening the remaining DHPR-unlinked channels (F ranzini-

15

 

 

 

 

 

 

 

 

Figure 2.2 HOdel of the skeletal muscle triad junction (From

MCPherson and Campbell, 1993).

 

SR lumen

 

Figure 2.3 Molecular coupling model for the release of Caz'
from SR. Voltage sensors in the T-tubular membrane move ,
outwards in response to depolarization. with sufficient
depolarization the calcium channels in the SR membrane are
open and calcium escapes into the cytoplasm and diffuses to

the myofibrils (From Jones and Round, 1990) .

17

Armstrong and Jorgensen, 1994). This would be consistent with the presence of Cay-induced
Ca2+ release (CICR) in skeletal muscle (Endo et al., 1970; Ford and Podolsky, 1970). Inositol,
1,4,5-triphosphate (1P3) has been considered as a possible candidate as the hypothetical
chemical transmitter fiom the T-tubule to the SR based on its role as a secondary messenger
for activation ofCaz“ release in smooth muscle and most non-muscle cells (Berridge, 1993).
Two early studies showing that IP3 promotes the release of Ca2+ from skinned skeletal muscle
fibers (Vergara et al., 1985) and SR vesicles (Volpe et al., 1985) suggest that IP3 may also
play a central role in the mechanism of skeletal muscle E-C coupling. However, IP3 appears
to be generated in concentrations too low and on time scale which is too slow to account for
the rapid release of Ca2+ in skeletal muscle (Walker et al., 1987).

Suficient evidence for the necessary involvement of a difl‘usible transmitter from T-
tubule to SR has not been provided and some indication against necessary participation has
been obtained. The direct molecular interaction model would thus seem to be the current
hypothesis of choice for the mechanism of T-tubule to SR signal transmission (Schneider,

1994)

18

2.2 The Ca’+ Release Channel Protein/Ryanodine Receptor of Skeletal Muscle

SarcOplasmic Reticulum

2.2.1 Purification and characterization of Ca”'-release channel protein/ryanodine

receptor

Ryanodine, a neutral alkaloid, which was isolated fiom the stems of the plant Ryam'a
speciosa Vahl, is a muscle-paralyzing agent (J enden and Fairhurst, 1969). Ryanodine induces
a progressive contracture in skeletal muscle and loss of contractile tension in cardiac muscle
by affecting the Ca” release mechanism in SR (see J enden and F airhurst, 1969, for review).
Fairhurst and Jenden (1962) and Jones et al. (1979) were the firstto show, in skeletal muscle
and cardiac muscle, respectively, that ryanodine stimulated accumulation of Ca2+ by the SR.
Sutko et al.(1985) proposed that ryanodine specifically blocks the Cay-channel of SR at high
concentration. Further studies demonstrated that the effect of ryanodine bound to the Ca”-
release channel is concentration-dependent; at concentrations in the range of 001-10 11M
ryanodine, Ca2+ release is stimulated, whereas, at concentrations in the range of 10-300 11M
ryanodine, release is inhibited (Gilchrist et al., 1992; Hasselbach and Migala, 1987; Lattanzio
et al., 1987; Meissner, 1986a; Nelson, 1987). This led to the widespread use of ryanodine as
a probe of SR function in both cardiac and skeletal muscle and provided the rationale for
using radiolabeled ryanodine to identify the presence of the Cay-release channel protein in
junctional SR Pessah et a1. (1985 and 1986) demonstrated that [’H]ryanodine binds with high

afinity in Ca2*-dependent manner in heavy SR preparations from rabbit skeletal muscle and

19

provided direct evidence for Cay-ryanodine receptor complexes involved in the release of
Ca2+ for contraction during E-C coupling. Using [H]ryanodine binding as an assay for
channel protein activity, Lai et al. (1988, 1989) purified the Ca2*-release channel and
characterized the protein as a large, 30S homotetrarner of negatively charged and
allosterically coupled polypeptides, each of M,>400,000. Purification of the Cay-release
channel has also been accomplished by immunoamnity chromatography using an anti-
ryanodine receptor monoclonal antibody (Imagawa et al., 1987) and by sequential column
chromatography on heparin-agarose and hydroxylapatite in the presence of CHAPS (Inui et
al., 1987a). Electron microscopy of the purified ryanodine receptor revealed a four-leaf clover
structure which is comparable in size and shape to the feet structures identified by F ranzini-
Arrnstrong (1970) in cisternae of SR suggesting that the ryanodine receptor is equivalent to
the feet structures (Inui et al., 1987a; Lai et al., 1988; Kawamoto et al., 1986).

The purified ryanodine receptor has been reconstituted into planar lipid bilayers where
it exhibits Ca2+ conductance properties and pharmacological modulation consistent with that
of the native Cay-release channel fiom SR vesicles (Lai et al., 1988; Lai et al., 1989;
Irnagawa et al., 1987; Smith et al., 1988; Hymel et al., 1988). The Ca2+ release channel has
a characteristically large unitary conductance and low divalent-over-monovalent ionic
selectivity, and is gated into the open state by cellular ligands (Smith et al., 1985, 1986a,

1986b)

2.2.2 Structure and Function of Ryanodine Receptor (RyR) Ca’*-Channel

20

Comparisons between freeze-fracture images of the junctional SR and rotary-
shadowed images of isolated triads and of the isolated foot protein reveal that the RyR/Caz“
channel has two domains: one is the large hydrophilic foot structure which spans the
junctional gap; the other is a hydrophobic domain buried in the membrane which presumably
forms the actual Ca2*-release channel (McPherson and Campbell, 1993). Hydropathy plots
are consistent with these morphological studies and suggest that the carboxy-tenninal pore
region, which is thought to consist of as few as four (Takeshima et al., 1989) or as many as
ten to twelve (Zorzato et al., 1990) putative transmembrane segments.

Analysis of the purified ryanodine receptor using freeze-drying and rotary shadowing
methods show a uniform population of large molecules, with four apparently identical
subunits, symmetrically disposed around a central depression. The entire large quatrefoil
structure is 26-28 nrn side-to-side and 33-36 am along the diagonal (Block, 1988). Averaged
images of negatively stained and frozen hydrated specimens have provided a more detailed
view of the structure of the rabbit skeletal muscle Cay-release channel protein (W agenknecht
et al., 1989; Radermacher et a1. 1992)). An overall dimension of 27x27x14 nm was observed
for the skeletal Ca2"-channel with an unusual ion-conduction structure composed of a central
(membrane-spanning) channel that branches into four radial channels in the cytoplasmic (foot)
region of the complex (Fig 2.4). The pathway for calcium emux is suggested by a central pore
from which four radial channels extend to each of four openings contiguous with the
myoplasm (Wegenknecht, 1989) (Fig 2.4).

Surface topology analysis combined with protease sensitivity has been used to provide

independent information identifying arrrino acid sequences with high likelihood for appearing

21

 

Figure 2.4 Three-dimensional reconstruction of the calcium
release channel obtained from negatively stained specimens.
Labeled are the central cavity (CC), the radial canal (RC),
and the peripheral vestibules (PV). (a,b) Top and bottom
faces; (c,d) reconstruction sliced open to reveal internal

structure. (From Radermacher et a1. , 1992) .

22

on the surface of the Cali-channel molecule and for predicting sequences with low surface
probability which may be buried within the hydrophobic core of the structure (Marks, et al.,
1990). Recently, Serysheva et a1. (1995) determined the low resolution three-dimensional
structure of the Cay-release channel in its closed state by exploiting the random orientations
of ice-embedded molecules imaged in an electron cryomicroscope. Their results reveal a
structure in which the transmembrane region exhibits no apparent opening on the SR lumen
side and the extended cytoplasmic region has a hollow appearance and consists, in each
monomer, of a clamp-shaped and a handle-shaped domain (Fig 2.5).

The primary structures of three mammalian RyR isoforms have been determined by
cDNA cloning and sequencing. The isoforms are encoded by three different genes coding for
skeletal muscle (ryrl) (Takeshima et al., 1989; Zorzato et al., 1990), cardiac muscle (Iyr2)
(Nakai et al., 1990; Otsu et al., 1990), and brain (Iyr3) (Hakamata et al., 1992) RyR.
Northern blot analysis of mRN A from a variety of mammalian tissues indicates that the
skeletal isofonn appears to be restricted to fast- and slow-twitch skeletal muscle (Takeshima
ct al., 1993). mRNA for ryr3 has also been found in mammalian skeletal muscle (Conti et al.,
1995)

The cDNA sequences of the skeletal muscle ryanodine receptor (ryrl) demonstrated
that the protein consists of 5,032 (from human), or 5,037 (from rabbit) amino acid residues
with molecular masses of 563,584 Da (T akashima et al., 1989) or 565,223 Da (Zorzato et al.,
1990), respectively. Primary structure predictions suggest the presence of several potential
cytoplasmic Ca2+, nucleotide, calmodulin binding, and phosphorylation sites (Takeshima et

al., 1989; Zorzato et al., 1990; Hakamata et al.,1992; Nakai et al., 1990; Otsu et al., 1990).

23

 

 

Figure 2.5 Surface representation of the 30 structure of ice-
ernbedded Ca2'-release channel in its closed state in a, bottom
or as many as ten to twelve (Zorzato et a1. , 1990) putative
transmembrane segments. The N-terminal half of the protein
view, b, top View and c, side view (From Serysheva et a1. ,

1995).

24

Two regions in the Ca2+ channel regarding regulating Cay-induced Ca2+ release in
skeletal muscle have been identified. The first was found at Arg‘” which is mutated to Cys
in swine susceptible to malignant hyperthermia. This mutation is associated with
hypersensitivity of the Ca2+ channel to Cay-induced Ca2+ release (Fujii et al., 1991). Evidence
for the second region was obtained in “Ca2+ and ruthenium red overlay studies with trpE
fusion proteins containing fragments of the Ca” channel primary structure. Three candidates
for Ca21—binding sites (amino acid residues 4246-4467, 4382-4417, and 4478-4512) have been
identified which may be involved in increasing the Ca2+ sensitivity of Ca2+ «channel protein
(Chen et al., 1992). The experimental results from Treves ct al.(1993) indicate that the Ca2+-
dependent gating domain of the Ca2+-channel lies near the junctional SR membrane at the
level of; or closely associated with, myoplasmic loop 2, corresponding to residues 43 80-4625.
This implies that the Ca21-gating domain is located far from the surface of the Ca21-channel,
in contact with the T-tubules, the membrane compartment that generates the trigger signal

for channel opening during E-C coupling.

2.2.3 Regulation of the Ca2+ release channel

The main cellular event controlled by Cay-channel in skeletal muscle is E-C coupling,
i.e. the increase in SR Ca2+ permeability triggered by muscle cell depolarization. Present
physiological and biochemical evidence suggests that the vertebrate skeletal muscle Ca2+
release channel is under the dual control of the T-tubule depolarization (discussed in Sec.

2.1.4) and various channel modulators (discussed below). Three complementary techniques

25

have been used to study regulation of the Ca2+ channel in vitro. First, the Ca2+ release rates
of passively or actively loaded triads or heavy SR vesicles have been followed by rapid
filtration or chemical quenching methods. Second, the behavior of channels incorporated fi'om
SR vesicles or of purified Ca2+ channel proteins has been studied by single channel recording
in planar lipid bilayers. Third, [3H]ryanodine binding by the channel protein has been used as
a probe to study the functional states of the Ca2+ channel.

Many compounds are known to affect SR Ca2+ release. Potentiators of SR Cd”
release include 11M Ca2+, adenine nucleotides, caffeine, halothane, ryanodine at nM
concentrations, sulflrydryl reagents, and calmodulin (at [Ca2+] < 0.1 11M). Inhibitors include
mM Mg”, Ca2+ at mM concentrations, ryanodine at 11M concentrations, ruthenium red,
procaine, dantrolene, spermine and calmodulin (at [Ca2+] > 1 11M) (McPherson and Campbell,
1993). It is likely that many of these compounds affect Ca2+ release by direct actions on the
Ca2*-channel as many of the compounds affect [’H]ryanodine binding to the purified Cd” -
channel, thus affecting the channel conductance or open state probability of purified Ca”-
channel incorporated into planar lipid bilayers (Fleischer and Inui, 1989).

A bell-shaped Ca2+ activation curve of Ca2+ efllux from heavy SR vesicles has been
obtained in the absence of other regulatory ligands such as Mg2+ and ATP; Cd‘ efllux is
maximal at 5-10 11M Ca2+ (Fill et al., 1990; Kirino et al., 1983; Meissner, 1984; Moutin and
Dupont, 1988; Nagasaki and Kasai, 1983). Ca2+ release is almost completely inhibited at 100
nM Ca2+ or > 10 mM C212+ . Such a curve suggests that the CS release channel possesses
high-aflinity activating and low-affinity inhibitory Ca” binding sites. Furthermore, there is

evidence which indicates that the aflinity and cooperativity of interaction between high- and

26

low-affinity [3H]ryanodine binding sites are dependent on Ca concentration and ionic
strength (Clm et al., 1990; Lai et al., 1989; McGrew et al., 1989; Meissner and El-Hashem,
1992; Mickelson et al., 1990; Pessah et al., 1987).

Mg"+ likely inhibits SR Ca2+ release by multiple mechanisms including (a) competition
with Ca2+ for the Ca2+ activation sites, (b) binding to the low-affinity Ca2+ inhibitory sites, or
(c) steric blocking of the channel as it binds to a site near the conduction pathway (Kirino,
1983; Meissner and Henderson, 1987; Nagasaki and Kasai, 1983).

Cay-induced Ca2+ release is greatly potentiated by physiological (mM) concentrations
of ATP (Meissner, 1984; Morii and Tonomura, 1983). Optimal channel activation was found
in the presence of M Ca2+ and mM ATP to give maximal Ca2+ release rates. Various adenine
nucleotides (AMP-PCP, ADP, AMP, cAMP, adenosine, adenine) also potentiate Ca2+ release,
which suggests that activation occurs because of binding of the nucleotide to an effector site
rather than covalent modification of the channel protein via a phosphorylation reaction.
(Meissner et al., 1986; Moutin and Dupont, 1988; Nagasaki and Kasai, 1983).

Cafl‘eine increases the sensitivity of the Ca2+ -induced Cay-release mechanism to Ca2+
and adenine nucleotides both in skinned fibers and in SR vesicles (for review see Martonosi,
1984). The increase in apparent Ca2+ affinity in C? channel in the presence of 50 mM
cafl‘eine is 20-fold. Calfeine in the millimolar range has been found to stimulate [3H]ryanodine
binding to skeletal Ca2+ release channel (Chu et al., 1990a; Zimanyi et al., 1992; Zimanyi and
Pessah, 1991). This effect resulted from an increase in the ryanodine association rate without
a change in the dissociation rate (Chu et al., 1990). In the presence of Ca2+ and Mg”, caffeine

appears to increase the affinity of the activation site for Ca2+ (Pessah et al., 1987). The effects

27

of caffeine and ATP are additive, resulting in increasing maximum rate of Ca2+ release
(Nagasaki and Kasai, 1983').

Whereas the cardiac RyR is an excellent substrate for the multifirnctional
Ca2*/calmodulin protein kinase, phosphorylation of the skeletal muscle RyR by endogenous
and exogenous CaM kinase was shown to have more variable effects (Chu et al., 1990; Strand
et al., 1993; \Vrtcher et al., 1991). In the skeletal muscle RyR, Ser2843 is a major target for
cAMP-, cGMP- and CaM-dependent kinases (Suko et al., 1993). An inactivation of Ca2+
channel activity resulting fi'om phosphorylation was observed in patch-clamp studies by Wang
and Best (1992). In contrast, an activation of single channel activities by phosphorylation,
which was ascribed to removal of block by Mg”, has been reported (Hain et al., 1993).

Calmodulin regulates skeletal Ca2+ channel by direct binding Cay-channel instead of
through a Ca21-CaM-dependent phosphorylation mechanism (Meissner, 1986; Smith et al.,
1989). Further details about the CaM binding and regulation of Ca2+ release channel protein
will be reviewed in section 2.3.2 The localization of the sites for other regulators of the Ca2+-

channel will be critical to the understanding of SR Cap-channel firnction in E-C coupling.

28

2.3 Calmodulin, a Versatile Calcium Mediator Protein

2.3.1 Structure and function of calmodulin

Calmodan (CaM) is a ubiquitous Ca2+ binding protein involved in a variety of cellular
calcium-dependent signaling pathways. The biochemical properties reveal CaM as small, heat
stable, and one of the most acidic proteins found in any tissue (Klee and Vanaman, 1982).
Based on primary structural homology, CaM is a member of a family of Ca2*-binding proteins
that includes troponin C, parvalbumin, and myosin light chains (for review see Goodman et
al., 1979). The 148 amino acid protein is a single polypeptide chain with a M, of 16,700.
Some 30% of its amino acids consist of aspartate and glutamate, accounting for the pI of 4.3.
The animal CaMs sequenced to date contain no cysteine, hydroxyproline, or tryptophan.
There is a high ratio of phenylalanine (8 residues) to tyrosine (2 residues) which results in
displaying a distinctive ultraviolet absorption pattern, a spectrum characteristic of the fine
structure of phenylalanine (Cheung, 1980). Plant CaMs typically exhibit the presence of l Cys
and only 1 Tyr (Bazari and Clarke, 1981; Corrnier, 1981).

Proteins with the functional and physicochemical properties of CaM are found in all
eukaryotic organisms. CaM is found at varying concentrations in all vertebrate tissues (Klee
and Vanaman, 1982). CaM from phylogenetically diverse sources have identical or at least
very similar biological and biochemical properties because of the highly conserved primary
structures (Cheung, 1980). Thus, CaM lacks tissue or species specificity reserved in Ca”-

binding proteins with similar physicochemical properties such as troponin C (Klee and

29

Vanaman, 1982).

Since the discovery of CaM by Cheung (1970), it has been demonstrated that this
regulatory protein mediates the cellular response to Ca2+ stimulation of at least 30 enzymes
including Ca2*-dependent protein kinases, adenylate cyclase, myosin light chain kinase,
phosphodiesterase, phosphorylase kinase, phospholipase A2, plasma membrane Ca2*—ATPase,
guanylate cyclase, and NAD kinase (Cheung, 1980). Whereas CaM will bind to some
structural proteins in the absence of Ca2+, metal ions are absolutely required for its firnction
as a regulator of most of its target enzymes in vitro (Means et al., 1991).

When Ca2+ concentration is transiently elevated in the cell in response to signal
transduction mechanisms, Ca2+ binds to CaM with 11M affinity at a stoichiometry of 4 Ca2+
atoms to 1 CaM molecule and this interaction produces a conformational change in CaM. The
Ca2*-CaM complex is now competent to interact with an acceptor or target protein. Such
binding results in a flirther conformational change in CaM as well as structural changes in the
target protein which allows expression of biological activity (Means et al., 1991).

Biophysics has revealed much about 3-dimensional structure of CaM itself,
conformational change upon binding of Ca2+, and interaction of CaM with its target proteins.
The X-ray crystallographic structure of CaM in the Ca2*-bound form shows a dumbbell-
shaped molecule with two globular domains arranged in a trans configuration that each bind
two Ca2+ ions (Fig. 2.6) (Babu et al. 1985). These domains are connected by a 26-residue,
central a-helix; the middle portion, which is highly mobile, acts as a flexible tether (Barbato
et al., 1992). Each Cay-binding domain consists of two helix-loop-helix motifs which are

commonly called EF-hands. This term is derived fi'om the structural homology of this domain

30

Calmodulin

 

’ 1

Figure 2.6. Structure of calmodulin A representation of the
three-dimensional crystal structure of vertebrate Call is
depicted. The amino and carboxyl termini are indicated by 'N '
and ‘C' respectively. The positions of the 4 bound Cay'ions

are shown by black dots (From Means et a1. , 1991) .

31

to the Ca2+-binding helix-loop-helix unit formed by the E and F helices of parvalburnin
(Kretsinger, 1980). These two pairs of helix-loop-helix are joined by a short antiparallel 8-
sheet. The carboxy-terminal lobe binds Ca2+ with high affinity (K,~10"’ M), the amino-
terminal with lower affinity (K,~10“ M).

Binding of Ca2+ ions induces a large conformational change, which makes two solvent-
exposed hydrophobic patches, one in each half of the molecule, available for target protein
interaction Each hydrophobic surface is surrounded by a polar rim which is rich in negatively
charged residues. At the center of each surface, there is a deep hydrophobic cavity that is
responsible for capturing an aromatic or a long aliphatic side chain of the target proteins
(Ikura et al., 1992; Meador et al., 1993; Meador et al., 1992) (Fig 2.7). In addition to the
extensive hydrophobic interactions, electrostatic interactions between the negatively charged
residues located at the polar rim and positively charged residues of the target proteins
contribute to the binding energy. Although the X-ray crystal structure of apo-CaM has not
been determined, the multidimensional nuclear magnetic resonance studies on the isolated
carboxy-terminal lobe have confirmed that the hydrophobic pockets are 'closed' in the absence
of Ca2+ (Finn et al. 1993). This result has been confirmed by determination of the solution
structure of Ca2*-free CaM using NMR spectroscopy (Finn et al., 1995; Zhang et al., 1995;
Kuboniwa et al., 1995). The removal of Ca2+ causes the interhelical angles of four EF-hand
motifs to increase and leads to major changes in surface properties, including the closure of
the deep hydrophobic cavity essential for target protein recognition. Various approaches have
been used to identify CaM-binding domains in its target proteins (Billingsley et al., 1990).

These include: limited proteolysis and fragment isolation by CaM affinity chromatography;

 

Figure 2.7 Atomic resolution structures of calmodulin and its
complex 'with skeletal myosin light chain kinase (MLCK)
peptide. (A) A ribbon diagram representation of the crystal
structure of unbound Ca2’-calmodulin. The four Cd‘ ions are
shown bound to the fOur helix-loop-helix EF-hands. (B) The
structure of CaӴcalmodulin.bound to the peptide representing
the calmodulin binding domain in skeletal muscle HLCK, as
determined by heteronuclear multidimensional NMR in solution

(From Torok and Whitaker, 1994) .

33

photoafiinity labeling using labeled CaM; cDNA expression and deletion mapping; and the
synthesis of peptides corresponding to putative CaM-binding regions. Most of the CaM-
binding domains are stretches of 16-35 amino acids which, in an (Jr-helical wheel
representation, show a segregation of basic and polar residues on one side and hydrophobic
amino acids on the other (O'Neil and DeGrado, 1990) (Fig. 2,8). However, the conformation
of the CaM-binding domain in the native protein remains an open question. The determination
of the structure of CaM bound to the synthetic CaM-binding domain of CaMKII, and the
refinement of the structure of the MLCK-binding domain-CaM complex indicate that the
interaction is not just hydrophobic but involves the formation of salt bridges between the basic
amino-terminal halfof the peptide and glutamic acid residues in the carboxyl terminus of CaM

(Meador et al., 1993).

2.3.2 The Role of Calmodulin in Regulation of Ca’*-Release from Channel Protein

Calmodulin was originally identified as a component of skeletal SR by Campbell and
Macme (1982) using boiled EGTA extracts of SR vesicles. These extracts were shown
to stimulate the phosphorylation of EGTA-washed SR vesicles in the same manner as
calmodulin. Chiesi and Carafoli (1982) purified CaM from skeletal muscle SR using a
Sepharose affinity column. Seiler et al. (1984) provided the first evidence that CaM bound
to high molecular mass proteins in junctional SR vesicles isolated from skeletal and cardiac
muscle which was later identified as C323 release channel.

Several lines of evidence suggest that CaM inhibits Ca” release activity via direct

34

 

Basic, highly
charged side

Figure 2.8 Helical-wheel projection of the calmodulin (CaM)-
binding domain, showing the segregation of basic (+) and
hydrophobic (O) residues to opposite sides of the helix (From

James et al., 1995).

35

interaction with Ca2*-channel instead of through a CaM-activated kinase during muscle
contraction. The inhibitory effect ofCaM on the Cay-channel is Ca” concentration dependent
and is reversible. In the absence of ATP, 2-10 11M CaM reduces “Ca” efflux rates from
passively loaded skeletal SR vesicles by a factor of 2-3 in the presence of M to mM Ca2+
concentrations (Meissner, 1986; Plank et al., 1988). Ca2+ release rates from cardiac SR are
reduced by a factor of 3-6 (Meissner and Henderson, 1987). The half-maximal inhibitory
concentration ofCaM is between 0.1 and 0.2 11M, and maximal inhibition is observed at 1-5
M (Meissner, 1986; Meissner and Henderson, 1987). CaM inhibits Cay-induced Ca2+ release
at Ca2+ concentrations between 0.1 and 100 11M without shifting the bell-shaped curve of Ca2+
dependence of release (Meissner, 1986). Single channel recordings demonstrate that 2 11M
CaM, in the absence of ATP, decreases the mean open time of Ca2*-release channels by 40%
without having an apparent effect on single channel conductance or channel permeability
properties, as measured from slope conductance and reversal potential in the presence and
absence of CaM (Smith et al., 1989; Fuentes et al., 1994). [3H]ryanodine binding studies
reveal that CaM inhibits [3H]ryanodine binding to CHAPS-solubilized and purified Cd -
channel (Fuentes et al., 1994).

Recently, it was demonstrated that CaM can also activate the Ca2+ release channel
several fold at < 0.2 11M free [Ca2+] which corresponds to the resting muscle [Ca2+] condition
(Tripathy et al., 1995; Buratti et al.,1995). Buratti et al. (1995) further demonstrated that the
central region of the Cali-channel, corresponding to residues 2937-3225 and 3546-3 655, may
contain CaM binding sites involved in the channel activation at low Ca” concentration (10"

M [Ca’1). In saponin-skinned fibers, higher concentration Of CaM (10 11M) potentiates Ca2+

36

release at low Ca2+ concentrations (< 3 11M), while it shows an inhibitory effect at high Ca2+

concentration (3 -30 11M) with 1 11M CaM (Ikemoto et al., 1995).

2.3.3 Identification of Calmodulin Binding Domains in Ca‘*-release channel

Sequence analysis of the skeletal muscle Cay-channel identified several candidate
CaM-binding sites in the C-terminal half of the protein (Fig. 2.8). Based on primary and
secondary structural analysis, Takeshima et al.(1989) predicted 2 CaM binding sites at
residues 3614-3637 and 4295-43 25, whereas Zorzato et al.(1990) predicted 3 difi‘erent sites
at residues 2807-2840, 2909-2930 and 3031-3049. Marks et al.(1990) have suggested four
putative CaM binding sites located at residues 2641-2657, 3362-3374, 3947-3965 and 4309-
43 22, by mapping the locations using limited proteolysis coupled with surface topography
analysis. Brandt et al.(1992) have suggested three other candidate sites for CaM at residues
1383-1400, 1974-1996 and 33 58-3374 by analysis of the calpain digestion pattern of the
channel protein in the presence and absence of CaM. All these regions satisfy the motif for
CaM binding sites which are basic amphiphilic helices, with the basic residues forming one
face of the helix and hydrophobic residues the other face (Fig. 2. 8).

Prediction of CaM binding domains have received some experimental support fiom
the studies by several groups who employed expression of Ca2*-channel fragments from
corresponding cDNAs (Fig 2.9). Menegazzi et al. (1994) have defined three CaM binding
regions, residues 2937-3225, 3546-3655 and 4425—4621 in rabbit skeletal muscle Ca2+-

charmel by CaM overlay on SDS PAGE of Ca2*-channel fusion proteins. The binding of CaM

37

 

0 1000 2000 3000 4000 5000
TakeshimsctaL 1989 I I
- Zorzatoetal. 1990 || |
Marksetal..l990 I I I I
Warsaw” | | I
W «$1994 - I I
amoral..l994 I I I. I I

Figure 2.9 Calmodulin-binding sites in skeletal muscle sarcoplasmic reticulum Caz*-release
channel. The linear sequence of the Caz”-channel is indicated by a horizontal line. The NI-I3+
and Coo‘ termini are marked. M’, M”, and Ml-MIO refer to predicted transmembrane
sequences (Zorzato et al., 1990). The candidates of CaM-binding sites are positioned by

verdical lines.

38

to these Cay-channel fusion proteins are Ca2+ concentration dependent. Chen et al. (1994)
have detected six Ca2*-dependent CaM-binding domains in rabbit skeletal muscle RyR by 125I-
CaM overlay on TrpE fusion proteins. Strong CaM-binding domains were localized between
amino acid residues 2063-2091, 3611-3642 and 4303-4328. Weaker CaM-binding domains
were localized between amino acid residues 921-1173, 2804-2930 and 2961-3084. The gel
overlay methods used by these groups may sometime be useful as a screening tool, but
interpretation of results raise many questions. Since different degrees of renaturation of these
firsion proteins on nitrocellulose membranes could afl‘ect measurement of CaM binding in the
overlay assay, CaM binding sites determined by the overlay method may not represent true
native binding sites in Ca” channel and this may lead to artifacts. Furthermore, although same
overlay method and fusion protein same overlay method and fusion protein system is
employed by both laboratories, there is little agreement on CaM-binding sites (Fig 2.9).
Efforts have been made to localize the CaM-binding sites in the three—dimensional
structure of the channel protein. Wagenknecht et al. (1994), using gold-cluster-labeled CaM
and electron microscopy, identified 1 CaM-binding site per subunit on the purified protein.
This CaM-binding site is at least 10 nm from the transmembrane channel of the Ca2*-channel
protein suggesting that long-range conformational changes are involved in the modulation Of

the Ca2+ channel activity by CaM.

39

2.4 Malignant Hyperthermia and Porcine Stress Syndromes

2.4.1 Introduction

Malignant hypertherrnia (MI-l) is an inherited myopathy of man and swine in which
inhalational anesthetics and skeletal muscle relaxants trigger severe skeletal muscle
contracture (Gronert, 1986). MH is associated with hyperrnetabolism and extreme elevation
in body temperature which can result in death unless promptly recognized and treated with
the skeletal muscle relaxant sodium dantrolene (Hanison, 1988). In man, MH is usually
associated with the administration of certain anesthetic agents such as halothane or
succinylcholine (Britt and Kalow, 1970). In swine homozygous for the defect, MH can also
be triggered by severe stress such as that engendered by social order fighting, exercise,
herding, hot environment, etc; thus, the disease is referred to as porcine stress syndrome
(PSS) (Cassens et al., 1980; Mitchell and Heffron, 1982).

Pale, soft, exudative (PSE) meat follows from the accelerated postmortem glycolysis,
concomitant myolactosis, and abnormally rapid postmortem fall in muscle pH peculiar to this
myopathy. This, it is proposed, alters the muscle water-holding capacity and texture by
denaturation of the sarcoplasmic proteins and contractile proteins (Lawrie et al., 195 8,
Wismer-Pedersen and Briskey, 1961). Major economic losses in the swine industry result
fiom the development of PSE meat that arise fiom postmortem manifestation of the disease
in MH susceptible (MHS) pigs. However, there are some desirable traits associated with the
presence of the gene for MB including leanness, muscle hypertrophy, and other desirable

causes traits.

40 .
2.4.2 Abnormal sarcoplasmic reticulum Ca’*-release channel in malignant

hyperthermia skeletal muscle

Because Ca2+ is the main regulator of muscle contraction and metabolism, the defect
in MH is believed to lie in abnormal Ca2+ regulation (Endo, 1977). The continued presence
of elevated Ca2+ within the skeletal muscle cell results in severe muscle contracture, enhancing
glycolytic and aerobic metabolism which depletes ATP, glucose, and oxygen, produces excess
C02, lactic acid and heat, and upsets cellular and extracellular ion. balances (Webb and
Simpson, 1986; Simpson and Webb, 1989).

The most significant abnomiality reported in MHS muscle, observed with isolated SR
fiactions, skinned muscle fibers, or muscle fiber bundles, is associated with the Ca2*-induced
calcium release mechanism of SR (Hefl‘ron and Ellis, 1985; O'Brien et al., 1986; Mickelson
et al., 1986; Mickelson et al., 1988; Carrier et al., 1991). There is now considerable evidence
that the primary biochemical defect in NIH is associated with an abnormal SR Ca2*-release
channel protein. Endo et al. (1983) observed that the Cay-induced Ca2+ release mechanism
from MI-I SR was both more sensitive to Ca”, and MH SR released Ca2+ at a greater rate than
nomial SR The data fiom Mickelson et al. (1986) suggest that there is no abnormality for the
Ca2+ uptake function of &2*-ATPm in MHS SR, that the initial phase of Ca2*-induced Ca”
release from MHS SR is enhanced, and that MHS SR has a different sensitivity to Ca”, Mg”,
ATP and cafi’eine than does normal SR.

Mickelson, et al.(1988) conelated a two-fold greater Cazi-release rate in SR isolated

fi'om MHS pigs with abnormal ryanodine binding properties. The altered Ca2+ dependence of

41

[H]ryanodine binding at the low aflinity Ca2+ site and a significantly lower K, (95 versUs 265
nM) for ryanodine in MHS Cay-release channel than that of normal SR suggested that
alterations on the SR Ca2+ channel may be responsible for the abnormalities in regulation of
Ca” release observed in MHS muscle. Mickelson et al. (1990) further utilized [3H]ryanodine
binding to the Cay-channel as a reporter Of open state of the Ca” release channel, and
further support the hypothesis that differences in the Ca2+ channel regulatory properties in
response to various channel stimulators (ATP and cafl‘eine) and inhibitors (ruthenium red and
Mg”) are responsible for the abnormal Ca2+ releasing activity of MHS SR The results from
Fill et al. (1990) showed that normal channels were inactivated by Ca” concentrations below
pCa 4, whereas MHS channels remained open at these Ca2+ concentrations for significantly
longer times. Based on these results, the hypothesis was proposed that a defect in a low-
aflinity Ca2+ binding site was responsible to the altered gating of MHS SR channel. Based on
equilibrium and kinetic evaluation of the binding of [3H]ryanodine to MHS SR Hawkes et'
al. (1992) demonstrated that the MH defect in pigs increases the apparent affinity of the SR
membranes for [’eryanodine by increasing the amount of high affinity sites relative to low
afinity binding sites. These findings suggest that the MH defect may alter the rate at which
the high afiinity form of the protein converts to the low afiinity form.

A single point mutation (T for C1843) in the porcine skeletal muscle ryanodine
receptor gene (ryr 1) has been identified which results in an alteration in amino acid sequence
from arginine at position 615 in normal RyR to a cysteine in the RyR ofMHS pigs (Fujii et
al., 1991). The following observations suggest that this single amino acid alteration is the

causal mutation for MH. Molecular genetic studies have shown that this single amino acid

4 2

alteration is cosegregated with porcine MH by comparing ryrI genotypes, as determined by
DNA-based analysis, and phenotypes, as determined by halothane challenge test which
induces signs typical of MH reaction (Fujii et al., 1991; Otsu et al., 1991). Shomer et al.
(1993) have demonstrated that the abnormalities in the MHS porcine Ca2+ release channel
activity were indeed the result of a mutation in this molecule, rather than an abnormal
membrane lipid environment or an unidentified regulatory proteins. Otsu et al. (1994) further
confirmed that this single amino acid mutation in Ca2+ channel protein was causative of MH
by showing that the mutated Ca2+ channels expressed in transfected myoblastic cells had
higher sensitivity to caffeine and halothane and resulted in higher cytosolic Ca2+ determined
by the fluorescence calcium indicator indo-l compared to wild type Ca2+ channel.

Shomer et al. (1994) reported that although Ca2+ regulation of Ca2+ release channel
activity is altered, the Arg‘” to Cys‘” mutation of the porcine Ca2+ release channel does not
affect the conductance or ion selectivity properties of the channel. Treves, et al. (1994) report
that the presence of the Arg-to-Cys point mutation in the recombinant RyR Ca2+ channel
expressed in COS-7 transfected cells causes abnormal cytosolic Ca2+ transients in response
to 4-chloro—m-cresol, an agent capable of eliciting in vitro contracture of MHS muscles. Their
results suggest that substituting Cys for Arg‘ls in the primary structure of the RyR is suflicient
to alter the intracellular Ca2+ homeostasis of eukaryotic cells.

It is apparent fi'om a number of studies that the Ca21-dependent regulation of the SR
Ca2+ release channel is altered in MH. However, it is not clear if the interaction of other
compounds which can regulate this channel such as cafi‘eine, calmodulin, Mg2+ and adenine

nucleotide, is also altered in MHS SR It has been reported that Arg ‘1’ is located on the

43

surface of the native Ca2*-release channel and is likely near important Ca2+ channel regulatory
sites (Mickelson et al., 1992). Therefore this mutated amino acid may cause the alteration of
the modulators binding to ryanodine receptor and results in the abnormal regulation of Ca2+-
release from SR channel. The sensitivity of skeletal muscle contracture by caffeine is increased
in the MHS pig (Britt, 1987; Gronert, 1986). Mickelson et al. (1990) further reported that
in the presence of optimal Ca”, MHS SR I H]ryanodine binding was more sensitive to
caffeine and ATP stimulation and less sensitive to ruthenium red or Mg2+ inhibition than was
normal SR However, the altered caffeine sensitivity of MHS muscle contracture does not
directly result from Ca2+ channel mutation in MH SR (Shomer et al., 1994). The direct
evidence concerning the functional role and interaction of other channel modulators such as
calmodulin, Mg2+ on the mutated Ca2+ channel protein in altering Ca2*-release properties of

the channel protein in MH is insufficient.

CHAPTER 3

CALMODULIN INTERACTION WITH THE SKELETAL MUSCLE

SARCOPLASMIC RETICULUM CALCIUM CHANNEL PROTEIN

3.1 Introduction

Contraction and relaxation of skeletal muscle are governed by changes in myoplasmic
[Ca2+]. Elevation Of myoplasmic [Ca2*], which triggers contraction, occurs when an action
potential induces Ca2+ release from the sarcoplasmic reticulum (SR) via a Ca21-release channel
(Martonosi, 1984; Inesi, 1985). Upon cessation of the action potential, Cay-release is
inhibited, the SR Ca2+-pump protein reduces myoplasmic [Ca2+] by pumping it back into the
SR, and the muscle relaxes. Although the Ca2+ uptake process has been described in
considerable detail, the molecular mechanisms involved in Cay-release are, by comparison,
poorly understood.

Regulation of Cay-release channel activity has been investigated by following Ca2+
efllux fiom isolated SR membrane vesicles (Nagasaki & Kasai, 1983; Ikemoto et al., 1985;
Meissner et al., 1986, Meissner, 1986) as well as by single-channel recordings of the activity
of the Ca2*-release channel protein incorporated from SR vesicles into planar lipid bilayers

(Smith et al., 1985; Smith et al., 1986a; Smith et al., 1986b). Results of these experiments

44

45

indicated the presence of a Ca2*-channel protein in SR whose activity was stimulated by uM
Ca2+, cafl‘eine, adenine nucleotides, and nM concentrations of the plant alkaloid ryanodine;
channel activity was inhibited by nM Mg", calmodulin (CaM), ruthenium red, and uM
concentrations of ryanodine.

Recently several groups have isolated a homotetrameric protein of subunit M, >
450,000 which bound ryanodine and, when incorporated into planar lipid bilayers, exhibited
Ca2+ channel activity nearly identical to that of intact SR vesicles and native SR Ca2*-channels
(Inui et al., 1987; Lai et al., 1988; Imagawa et al. 1987). The purified ryanodine
receptor/Cazl-channel protein also displayed strikingly similar morphology to the junctional
foot structure which spans the gap between the T-tubule and junctional SR (Inui et al., 1987;
Lai et al., 1988). On the basis of this evidence, the ryanodine receptor protein was proposed
to be synonymous with the Ca2*-release channel and the junctional foot of SR vesicles.

Calmodulin is a ubiquitous Caztbinding protein which regulates the activity of at least
30 known enzymes and other proteins in response to changes in cellular Ca2+ concentration
(Klee and Vanaman, 1982; Means et al., 1991) Previous studies have shown that CaM
inhibits the Ca” release rate from SR vesicles by 2-3 fold in skeletal muscle (Meissner, 1986;
Plank et al., 1988) and up to 6-fold in cardiac muscle (Meissner & Henderson, 1987).
Subsequently, Smith et al., (1989), using the planar lipid bilayer-vesicle fusion technique,
demonstrated that the inhibitory effect of CaM on both skeletal and cardiac SR channel
proteins resulted from reduction of the open time probability via direct binding of CaM to the
channel protein.

These results have led to the hypothesis that the role of CaM in regulation of the SR

46

Cay-channel activity is that of a partial feedback inhibitor of Ca2+ -release (Meissner, 1986,
Smith et al., 1989). However, in the absence of direct binding data on the interaction of CaM
with the channel protein, it is difficult to assess the physiological role of CaM in regulation
of Ca2*-release activity. These studies were initiated to define conditions under which CaM
binds to the channel protein and thus, would be capable of regulating its activity. Our results
suggest that in skeletal heavy SR, the most abunth receptor for CaM is the Cay-channel
protein. CaM binds to the channel protein with high afiinity, even in the presence of EGTA.
In the physiological range of KCl concentrations, binding of CaM is enhanced in the presence
of 0.1 mM CaClz, and further enhanced by inclusion of 1 mM MgCl . Correlation of
ryanodine binding data with CaM-binding data suggests that there are multiple CaM-binding
sites on each channel protein subunit, and that the afinities of these binding sites for CaM

change in response to metal ion concentrations.

47

3 . 2 Experimental procedures

3.2.1 Materials. Benzophenone4-maleimide and rhodamine-X-maleimide were purchased
fiom Molecular Probes (Junction City, OR). Na[n’I] and [’H]-ryanodine were obtained from
DuPont-NEN (Boston, MA). Ryanodine was purchased from Calbiochem (La Jolla, CA).

Wheat germ was a gift from International Multifoods (Minneapolis, MN).

3.2.2 Preparation of Calmodulin and its Derivatives. CaM was purified from wheat germ
using the procedure described previously (Strasburg et al., 1988). For crosslinking
experiments, CaM was iodinated with [”51] at the sole tyrosine residue (Tyr 139), followed
by site-specific modification at Cys 27 with the photoactivatable crosslinker, benzophenone-4-
maleimide (Strasburg et al., 1988). For fluorescence experiments, purified CaM was
derivatized at Cys 27 with rhodamine-X-maleimide as described for I-EDAN S (Strasburg et

al., 1988).

3.2.3 Calmodulin Crosslinking. Aflinity labeling of CaM-binding proteins in SR vesicles was
performed by incubating in darkness 0.1 uM [‘z’fl-Bz-CaM with 100 pg of SR vesicles in 100
uL of 20 mM Hepes pH 7.5, and 0.2 M NaCl. CaClz, MgCl, and EGTA were included as
indicated in Figure legends. The mixtures were placed in plastic microcentrifuge tubes on ice,
the tubes were covered with a plate of glass, and the samples were illuminated for 20 minutes
in a Stratalinker 1800 photoreactor (Stratagene Corp., La Jolla, CA) equipped with lamps

of Am, = 254 nm. After photolysis of the mixture, the samples were centrifuged in a

48

Beckman TL-100 centrifirge at 100,000 x gun, for 20 minutes. The membrane pellets were
resuspended in water, SDS was added to produce a final concentration of 1%, and the
samples were subjected to polyacrylamide gel electrophoresis using 5-20% linear gradient gels
(Laemmli, 1970). The gels were dried and placed with Kodak Omat XAR-S X-ray film in

autoradiography cassettes equipped with Dupont Lightning Plus intensifying screens.

3.2.4 Preparation of Sarcoplasmic Reticulum Vesicles. Skeletal muscle heavy SR vesicles
were isolated from longissimus muscle of pigs obtained fi'om the Michigan State University
swine farm or from the Yorkshire swine herd at the University of Minnesota using the
procedure of Mickelson et al., (1986) with the modifications that 1 mM EGTA, 0.1 mM
PMSF, 1 rig/ml of aprotinin and 1 pg/ml leupeptin were included in the homogenization
buffer. Light SR vesicles were prepared from rabbit longissimus muscles according to the

procedure of Fernandez et al. (1980).

3.2.5 Fluorescence Anisotropy Measurements. Rhodamine maleimide-labeled CaM (Rh-
CaM) binding to the channel protein in SR vesicles was monitored by fluorescence anisotropy
measurements using an SLM 4800 spectrofluorometer modified with data acquisition
hardware and operating system fi'om On-Line Instrument Systems (Bogart, GA).

Samples were held in a therrnostatted cell block maintained at 22°C. The excitation
wavelength of Rh-CaM was 580 nm, monochromator slits were set at 8 nm, and emitted light
was isolated using Schott RG-610 filters. During titrations samples were allowed to

equilibrate for 5 minutes after each addition of Rh-CaM or SR All samples included 1 rig/ml

49

aprotinin, 1 rig/ml leupeptin and 0.1mM PMSF to inhibit proteolysis during the experiment.

All fluorescence measurements were performed using semi-micro, quartz fluorescence
cuvettes (4 mm x 10 mm). Prior to fluorescence experiments the cells were rinsed with 1'
mg/ml bovine serum albumin to minimize Rh-CaM adsorption to the walls of the cuvette.
Following this treatment, the measured anisotropy value for Rh-CaM was independent of
concentration over the range of 1 nM to 1000 nM, indicating that there was negligible Rh-
CaM adsorption to the cuvettes.

(Am—21,.)
fb=
Am(1-q>+q(A,,) —A

 

(1)

f

The anisotropy of a ligand (Rh—CaM) is directly proportional to the fiaction of ligand
bound to the receptor (Ca’*-channel protein). Thus, if Af is the anisotropy of free Rh-CaM
and A, is the anisotropy of the fully bound ligand, then the fraction bound, f.,, is determined
from: where A", is the measured anisotropy for a given ligand concentration, and q, the
change in quantum yield, is the ratio of fluorescent intensity of bound species over that of the
free species. If the change in quantum yield is negligible upon binding of ligand, then equation
(1) reduces to:

__ (Am-Afr

fb______ (2)
(Ab-At)

The Man of ligand bound, and the concentrations of bound and free Rh-CaM are readily

calculated. Values of K, and BM were calculated using the computer program Enzfitter, a

50
nonlinear regression analysis program of RI . Leatherbarrow (Biosofi, Cambridge, UK). Data
were fit to a one-ligand or two-ligand binding model to obtain the best fit.

The anisotropy of unbound Rh-CaM, A,, was measured in the absence of SR. The
anisotropy of the firlly bound species, Ab, was obtained by titration of Rh-CaM with SR
vesicles, followed by curve-fitting using the Enzfitter computer program for a single class of
ligand binding sites.

Corrections for light scattering and background fluorescence were made by

application of the equation:

A=flAl+f2A2 (3)

where A, Al and A2 are the anisotropies of the sample, the blank, and the corrected sample,
respectively. The fiactional contributions, fl and £2, of these species were calculated from the
intensities measured with the excitation monochromator in the vertical position and the
emission monochromator at 55°. The corrected sample anisotropy, therefore, is that value

in the absence of background interference.

3.2.6 Calmodulin Content of Heavy SR Vesicles. Heavy SR vesicles were suspended to a
concentration of 1 mg/ml in 20 mM irnidazole buffer, pH 7.4. The suspension was incubated
for 5 minutes in a boiling water bath, and centrifuged 15 minutes at 100,000 x gum in a
Beckman TL-100 centrifuge. Aliquots of the CaM-containing supernatant were used in the
erythrocyte membrane ghost CaATPase assay described below.

Preparation of porcine erythrocyte membrane ghosts and assays for CaM-stimulable

51

activity of the CaATPase of the erythrocyte ghosts were based on the procedures described
by Thatte et al. (1987). Inorganic phosphate was determined by the method of Rockstein and
Herron (1951). A standard curve was prepared for CaATPase activity as a firnction of wheat
germ CaM concentration; CaM content of heavy SR vesicles was determined fiom the
standard curve using aliquots of the boiled SR supernatants. CaM-stimulable CaATPase
activity of the boiled supernatants was defined as the ATPase activity in the presence of 0.1

mM CaCl2 minus that of an equal aliquot in the presence of 1 mM EGTA.

3.2.7 Biochemical Assays. SR protein concentrations were determined by the Lowry method
of (Lowry et al., 1951) using bovine serum albumin as the standard. Ryanodine binding
activity of SR vesicles was determined according to the method of Mickelson et al. (1988).
Ca2+ titration of CaM/channel protein complexes were performed using an EGTA/NT A
buffer. Free Ca2+ concentrations were calculated using the computer program of Perrin and

Sayce (1967).

52

3 . 3 Results

3.3.1 Calmodulin Content of Heavy SR Vesicles. The CaM content of porcine heavy SR
vesicles was determined by heating SR samples to release CaM, centrifirgation to remove
insoluble material, and measuring CaM activity in an erythrocyte membrane ghost CaATPase
assay. The CaM content of our preparations of heavy SR ranged fiom 15-33 pmol/mg SR
protein. Prior extractions of SR vesicles with EGTA did not significantly affect these results,

indicating that the endogenous CaM was tightly bound and likely non-exchangeable.

3.3.2 Flu-Ryanodine Binding to SR Vesicles. The binding activity of [3H]-ryanodine to
heavy SR vesicles, determined by Scatchard plot analysis, show a BM value of 10.6 :t 0.9
pmol/mg for our preparations. These results were similar to those obtained by Mickelson et
al., (1988). The ryanodine-binding activity of the rabbit skeletal muscle light SR vesicles was

‘ 0.1 pmol/mg.

3.3.3 Identification of Calmodulin-binding proteins in SR vesicles. Purified heavy SR
vesicles fi'om porcine skeletal muscle and purified light SR from rabbit skeletal muscle were
incubated with the affinity-labeling derivative [‘Z’I]-Bz-CaM to identify the receptor proteins
for CaM. The autoradiograrn (Figure 3.1) of the gel electrophoretogram indicates that the
major complex formed in the heavy SR fi'action was a doublet of 114, > 450,000 which
corresponds to CaM plus the channel protein subunit (Seiler et al., 1984). This complex was

obtained in the presence of EGTA or Ca2+ at each Mg2+ concentration examined and suggests

53

Figure 3.1 Mg" and Ca2+ dependence of affinity labeling of skeletal muscle heavy and light
SR with [’2’I]-Bz-CaM [‘2’I]-Bz—CaM was incubated with skeletal muscle heavy SR vesicles
(A and B, lanes 1—6) or light SR vesicles (A and B, lanes 7-12) in 0.2 M NaCl, 20 mM Hepes
buffer (pH 7.0) plus the following components: Lanes 1,7: 1 mM EGTA; lanes 2,8: 1 mM
EGTA, 1 mM MgC12; lanes 3,9: 1 mM EGTA, 10 mM MgC12; lane 4,10: 0.1 mM CaCl,; lane
5,11: 0.1 mM CaCLb 1 mMMgClz; lane 6,12: 0.1 mM CaClz, 10 mM MgC12. A) Coomassie-
blue stained gel; B) Autoradiogram of dried gel. C) Inhibition of amnity labeling of skeletal
muscle heavy SR vesicles by [‘2’I]-Bz-CaM with unlabeled CaM. Skeletal muscle heavy SR
vesicles (1 mg/ml) were incubated with 0.1 uM [‘2’I]-Bz-CaM in the presence of 0.2 M NaCl,
20 mM Hepes buffer, pH 7.5, 0.1 mM CaCl2 and 1 mM Mng and varying amounts of
unlabeled CaM. The mixtures were separated by electrophoresis on a 5-20% acrylamide
gradient gel, and the crosslinked products were identified by autoradiography of the gel.
Lanes 1-10 represent 0, 0.05, 0.1, 0.5, 1, 2, 4, 6, 8, 10, uM of unlabeled CaM, respectively.
Arrow indicates position of the channel protein subunit in the gel, and of the channel

subunit/[mn-Bz-CaM complex in the autoradiograms.

54

A

1 23456789101112

Mb
..

 

B
-205

           

  

e.4....nt..,fif..\..fl
.J .rn. r\ls~4'~'”

e.

       

1 23456789101112

Figure 3. l--cont’

55

S6

 

57

that CaM could bind to the channel prOtein at [Ca2+] in resting muscle. In contrast, at the
same protein concentrations and under the same conditions, there was no apparent aflinity
labeling of light SR (Figure 3.1B). In order to demonstrate that the binding of CaM to the
Ca2*-channel protein was specific, aflinity labeling experiments were conducted in the
presence of increasing concentrations of unlabeled CaM. The [‘Z’I]-Bz-CaM was readily
displaced from the channel protein by the unlabeled CaM (Figure 3.1C), with crosslinking

eliminated at unlabeled CaM concentrations greater than 1 uM.

3.3.4 Titration of Rh-CaM with SR Vesicles. Since the affinity labeling experiments
indicate that Ca2+ channel protein is the most abundant receptor for CaM in our heavy SR
preparations, fluorescence anisotropy could be used to characterize CaM interaction with the
channel protein in native SR vesicles. Heavy SR vesicles were titrated into Rh-CaM under
three difl‘erent metal ion conditions: 1) +1 mM EGTA; 2) +0.1 mM CaCl,; 3) +0.1 mM CaClz,
and 1 mM MgC12. In each case, titration of SR vesicles into Rh-CaM resulted in a large
increase in fluorescence anisotropy attributableto the increased molecular mass of the Rh-
CaM/Cafi-channel protein complex (Figure 3.2). That the Rh-CaM was indeed binding to the
channel protein is supported by the affinity labeling experiments (Figure 3. 1B) and the
following control experiments. The increase in anisotropy was reversed by addition of a large
excess of unlabeled CaM (not shown). Titration of light SR vesicles into Rh-CaM resulted
in a slight increase in fluorescence only at high SR concentrations (Figure 3.2). F urthermore,
titration of Rh-CaM with SR vesicles, which were first treated with trypsin (1 :50 w/w, 1 hour

at 37°C), resulted in no change in anisotropy (not shown), indicating that there were

58

Figure 3.2 Titration of Rh-CaM with skeletal heavy and light SR vesicles under different
divalent ion conditions. The sample medium contained 10 nM Rh-CaM, 0.3 M sucrose, 0.3
M KCl, 50 mM PIPES, pH 7.0 plus one of the following conditions: 1 mM EGTA (0); 0.1
mM CaCl2 (O); 0.1 mM CaCl2 plus 1 mM MgCl2, (A) in a starting volume of 1 mL. Heavy
SR: (O, 0, A, ); light SR (A). The Rh-CaM sample was titrated with SR vesicles in parallel
with a bufl‘er blank containing the same media minus Rh-CaM. Corrections were made for

light scattering as described in "Experimental Procedures".

 

Anisotropy

0.25

0.2 3

0.2 1

0.19

0.17

6.15

59

 

 

 

 

 

1 0 , 1 00
[HSR]. uglml

1000

60

negligible non-specific Rh-CaM interactions with the membranes and that light scattering
corrections were valid.

Data from these titrations (Figure 3.2) were used to determine the anisotropy values
for the free Rh-CaM species (A,) and for the fully bound Rh-CaMCa"-channel protein
complex (A.). Normally these values are obtained fiom the limits of the titration curves.
However, values for A. could be slightly underestimated in these experiments because of
excessive light scattering at high concentrations of SR vesicles (>1 mg/ml). Instead, A. values
were calculated by extrapolation using the Enzfitter program applied for a single ligand
binding site on Rh-CaM. The A, values obtained for each metal ion condition averaged
0.2435. The A5 values obtained from 10 nM samples of Rh-CaM under the various metal ion
conditions in the absence of added SR, were 0.1808 (+ 1 mM EGTA), 0.1728 (+ 0.1 mM
CaClz), and 0.1585 (+ 0.1 mM CaClz, + 1 mM MgCl, ). There was no significant change in
fluorescence intensity upon binding of Rh-CaM to the channel protein (q = 1.0); therefore the

fraction of Rh-CaM bound was calculated using Eq. 2.

3.3.5 Ionic Strength Dependence of the Binding of Rh-CaM‘ to SR Vesicles. Our initial
experiments (not shown) were conducted in the absence of added KCl and showed that in the
presence of EGTA, there was no change in Rh-CaM anisotropy upon titration with SR
vesicles. The results thus suggested that in the absence of adding KCl there was no binding
of Rh-CaM to the channel protein at [Ca2+] < 50 nM. However, in the presence of KCl there
was an increase in fluorescence anisotropy observed upon titration of SR into Rh-CaM in the

presence of EGTA (Figure 3.2). To determine whether binding of Rh-CaM to the skeletal SR

61

Cay-channel protein could be dependent upon ionic strength, the anisotropy for fixed
concentrations of Rh-CaM and SR vesicles was measured as a function of KCl concentration.
The chosen concentrations of SR (86 pg) and Rh-CaM (10 nM) correspond to the
approximate midpoints of the titration curves in Figure 3.2. As the salt concentration was
varied, either an enhancement or inhibition of binding would be observed by an increase or
decrease in anisotropy, respectively. The KCl titration data (Figure 3 .3) clearly show that in
the presence of EGTA, the binding of Rh-CaM to the channel protein was highly ionic
strength-dependent, increasing from <10% Rh-CaM bound at 3 mM KCl to 45% bound at
0.2 M KCl. At [KCl] >03 M, the anisotropy rapidly declined, indicating decreased binding
of Rh-CaM to the Ca2+ channel protein. In the presence of 0.1 mM Ca2+ plus 1 mM Mi‘ ,
titration of KCl into the SR/Rh-CaM mixture also indicated a significantly higher aflinity of
Rh-CaM for the Ca2"-channel protein, with a maximum at about 0.3 M KCl. However, in
contrast to the EGTA conditions, binding of Rh-CaM to the channel protein did not

significantly decrease at higher KCl concentrations.

3.3.6 Ca2+ Dependence of the binding of Rh-CaM to SR vesicles. Ca2+ is required for the
binding of CaM to most of its known receptors (Klee et al., 1982), although in a few cases
the aflinity of CaM for its receptor decreases with elevated [Ca2+] (Cimler et al., 1985;
Masure et al., 1986). Therefore, to define optimal [Ca2"] for CaM/channel protein interaction,
the Ca2*-dependence of the binding of Rh—CaM was determined. The Optimal [C321] for Rh-
CaM/channel protein interaction was dependent on the MgCl2 concentration (Figure 3.4). In

the absence of Mg”, the optimal [Ca2+] for enhanced binding of Rh-CaM to the channel

62

Figure 3.3. KCl dependence of the binding of Rh-CaM to SR vesicles under different divalent
ion conditions. The sample bufl‘er contained 86 pg heavy SR vesicles, 10 nM Rh-CaM, 0.3
M sucrose, 50 mM PIPES, pH 7.0 and either 1 mM EGTA (0) or 0.1 mM CaCl2 plus 1 mM
MgC12(O) in a starting volume of 1 mL. Points represent the means t SE. of 3 preparations.
% CaM bound was calculated fi'om measured anisotropy as described in "Experimental
Procedures"; 100% bound corresponds to 229 pmol/mg in the presence of EGTA and 80

pmol/mg in the presence of calcium plus magnesium.

96 CaM Bound

100

80

60

40

20

63

 

 

l

s 1 111411

 

 

 

10

[KCI]. mM

67

Figure 3.5 Titration of Rh-CaM/ SR vesicles with MgClz. The sample medium contained 86
pg heavy SR vesicles, 10 nM Rh-CaM, 0.3 M sucrose, 0.3 M KCl, 50 mM PIPES, pH 7.0
and 0.1 mM CaCl2 (O) or 1 mM EGTA (0) in a starting volume of 1 mL. Points represent
the means :t SE. of 3 preparations. % CaM bound was calculated from measured anisotropy
as described in " Experimental Procedures"; 100% bound corresponds to 80 pmol/mg in the

presence of calcium and 229 pmol/mg in the presence of EGTA.

% CaM Bound

68

 

100

40‘

20‘

 

 

r'rrrsjrrl r

I ll!

 

o rrrrrrl r r rrrrrrl r r rrrrrrl

0.002 0.01 0.1 1

(Mean, mM

10

 

70

66
protein was approximately 1 pM, whereas in the presence of 1 mM Mg", the optimum [Ca2+]

for Rh-CaM binding to the channel protein was approximately 50 pM (Figure 3.4).

3.3.7 Mg“-dependence of the Binding of Rh-CaM to SR Vesicles. Previous results
(Meissner et al., 1986) indicated that Mg” is an antagonist of the Ca2+ release activity in
heavy SR vesicles. To determine whether Mg2+ might alter the interaction of CaM with the
Ca21- channel protein, the anisotropy of Rh-CaM was determined as a firnction of [Mgr’] in
the presence or absence of 0.1 mM Ca2+. Binding of Rh-CaM to the channel protein was
slightly enhanced as the [Mg21] was increased to 0.5 mM (+EGTA) or 1 mM («25* ) as
indicated by the increase in fluorescence anisotropy. At [Mgz‘] above 1 mM (+EGTA) or

above 7 mM (+Ca2"), binding was significantly inhibited (Figure 3.5).

3.3.8 Rh-CaM/Ca“-Channel Protein Binding Equilibrium. Having established optimal
salt and divalent metal ion concentrations, Rh-CaM was titrated into fixed concentrations of
SR vesicles to determine binding capacity and aflinity of the skeletal SR Cay—channel in SR
vesicles for Rh-CaM under the following conditions: 1) 1 mM EGTA; 2) 0.1 mM CaClz; 3)
0.1 mM CaCl2 plus 1 mM MgC12. Ar and A, values were used to calculate the fraction of Rh-
CaM bound to the channel protein in SR vesicles for each point in the titration.

Results oftitrations conducted in the presence of EGTA are indicative of a single class
of binding sites on the SR Ca2*-channel protein for CaM (Figure 3.6 and Table 3.1).
Scatchard analysis of the data yields a dissociation constant, K,,, of 8.6 :I: 0.8 nM and a binding

capacity Bm of 229 i 7 pmol/mg of SR protein. In the presence of 0.1 mM CaClz, the

69

Figure 3.6. Titration of skeletal heavy SR vesicles with Rh-CaM in the presence of EGTA
A) Anisotropy plot of titrations of heavy SR with Rh-CaM. B) Rh-CaM/SR saturation binding
curve. The inset is a Scatchard plot Rh-CaM binding to SR vesicles. The sample medium
contained 90 pg of heavy SR, 0.3 M sucrose, 0.3 M KC], 50 mM PIPES, pH 7.0 and 1 mM
EGTA in a starting volume of 1 mL. Points represent the means :t SE. of 3 preparations.
CaM bound was calculated from measured anisotropy as described in "Experimental

Procedures".

Anisotropy

Bound CaM, pmol/mg

7O

 

0.18 "

 

l

 

 

s 1 111411 1 1 suasnl 1 J 1411111

 

 

0.16
0.1

200

150

100

 

 

 

 

 

 

 

 

 

 

l 10 100 500
[CaM], nM
8
E
g
C
3
a:
0 50 100 150 200 250
Bound CaM, pmollrng
l l l l
50 100 150 200 250

Total [CaM], nM

67

Figure 3.5 Titration of Rh-CaM/SR vesicles with MgClz. The sample medium contained 86
pg heavy SR vesicles, 10 nM Rh-CaM, 0.3 M sucrose, 0.3 M KCl, 50 mM PIPES, pH 7.0
and 0.1 mM CaCl2 (O) or 1 mM EGTA (0) in a starting volume of 1 mL. Points represent
the means :t SE. of 3 preparations. % CaM bound was calculated from measured anisotropy
as described in " Experimental Procedures"; 100% bound corresponds to 80 pmol/mg in the

presence of calcium and 229 pmol/mg in the presence of EGTA.

71

Table 3.1 Equilibrium Constants for Rh-CaM Interaction with the Ca’+-channel

Protein in SR Vesicles'

 

 

 

 

 

 

 

 

 

p
Bmsxl Kn Bmsxz K42
pmol/mg nM pmol/mg nM
+1 mM EGTA 229:1:7 8.6:l:0.8 - -
+0.1 mM CaClL 54i7 3:1:l.1 166:1:28 239i102
+0.1 mM CaCl2 10.0i0.8 0. 11003 70:1:2 17:1:1
. +1 mMM l

 

 

* Data were obtained from titrations of SR vesicles with Rh-CaM in the presence of 0.3 m
KCl, 50 mM pipes, pH 7.0, and divalent ion conditions as listed below. Data are means i

S.E.of 3 preparations each.

72

titration data are consistent with two classes of ligand binding sites on the channel protein for
CaM (Figure 3.7 and Table 3.1). The high afiinity class ofsites has K,“ = 4.3 d: 1.1 nM and
BM, =54 :t 7 pmol/mg; the results of low affinity binding class site shows K,2 = 239 i 102
and BM = 166 i 28 pmol/mg. In the presence of 0.1 mM Ca2+ plus 1 mM Mg”, there is a
dramatic shift in the binding capacity of the high afiinity class of sites (Figure 3.8 and Table
3.1). The high affinity binding site has a Bml = 10.0 :t 0.8 pmol/mg and Kn = 0.10 :1: 0.03

nM; the lower afiinity class of sites has K,12 of 17 i 1 nM and a Bm2 = 70 i 2 pmol/mg.

69

Figure 3.6. Titration of skeletal heavy SR vesicles with Rh-CaM in the presence of EGTA
A) Anisotropy plot of titrations of heavy SR with Rh-CaM. B) Rh-CaM/SR saturation binding
curve. The inset is a Scatchard plot Rh-CaM binding to SR vesicles. The sample medium
contained 90 pg of heavy SR, 0.3 M sucrose, 0.3 M KC], 50 mM PIPES, pH 7.0 and 1 mM
EGTA in a starting volume of 1 mL. Points represent the means i SE. of 3 preparations.
CaM bound was calculated from measured anisotropy as described in "Experimental

Procedures".

Anisotropy

Bound CaM, pmol/mg

7O

 

0.18 '

 

 

 

 

l lllllll l J [111111 I J ljjljll l

 

 

0.16
0.1

200

150

100

 

 

 

 

 

 

 

 

 

l 10 100 500
[CaM], nM
8
I- 2 > .
$ 0
s:
8 1 . 0
m
a
0 . . .
0 50 100 150 200 250
Bound CaM, pmollrng
] l l l
50 100 150 200 250

Total [CaM], nM

73

Figure 3.7 Titration of skeletal heavy SR with Rh-CaM in the presence of CaClz. A)
Anisotropy plot of titrations of heavy SR with Rh-CaM. B) Rh-CaM/SR saturation binding
curve. The inset is a Scatchard plot of binding of Rh-CaM to SR vesicles. The sample
medium contained 150 pg heavy SR, 0.3 M sucrose, 0.3 M KC], 50 mM PIPES, pH 7.0 and
0.1 mM CaC]2 in a starting volume of 1 mL. Points represent the means :1: SE. of 3

preparations. CaM bound was calculated from measured anisotropy as described in

Experimental Procedures".

71

Table 3.1 Equilibrium Constants for Rh-CaM Interaction with the Ca”'-channel

Protein in SR Vesicles'

 

 

 

 

 

 

 

 

er Kn Burma K42
pmol/mg nM pmol/njg nM
I +1 mMEGTA 229d:7 8.6:i:O.8 - -
I +0.1 mM CaC]2 54i7 33:1.1 166i28 2391102
+0.1 mM CaCl2 10.0108 0.1:]:0.03 70i2 17d:1
+1 mM MgC12

 

 

* Data were obtained from titrations of SR vesicles with Rh-CaM in the presence of 0.3 m

KC], 50 mM pipes, pH 7.0, and divalent ion conditions as listed below. Data are means :

S.E.of 3 preparations each.

72

titration data are consistent with two classes of ligand binding sites on the channel protein for
CaM (Figure 3.7 and Table 3.1). The high aflinity class ofsites has Kd1 = 4.3 :I: 1.1 nM and
BM, =54 :t 7 pmol/mg; the results of low afiinity binding class site shows Kdz = 239 i 102
and Bud = 166 i 28 pmol/mg. In the presence of 0.1 mM Ca2+ plus 1 mM Mg”, there is a
dramatic shift in the binding capacity of the high amnity class of sites (Figure 3.8 and Table
3.1). The high aflinity binding site has a er = 10.0 i 0.8 pmol/mg and Kn = 0.10 :1: 0.03

nM; the lower aflinity class of sites has K42 of 17 i 1 nM and a Bm2 = 70 i 2 pmol/mg.

75

Figure 3.8. Titration of skeletal heavy SR with Rh-CaM in the presence of CaC]2 plus MgClz.
A) Anisotropy plot of titration of heavy SR with Rh-CaM. B) Rh-CaM/ SR saturation binding
curve. The inset is a Scatchard plot analysis of binding of Rh-CaM to SR vesicles. The sample
medium contained 50 pg of heavy SR, 0.3 M sucrose, 0.3 MKC], 50 mM PIPES, pH 7.0 and
0.1 mM CaC]2 plus 1 mM MgCl2 in a starting volume of 1 mL. Points represent the means :1:
SE. of 3 preparations. CaM bound was calculated from measured anisotropy as described

in " Experimental Procedures".

Anisotropy

Bound CaM, pmol/mg

76

 

0.10 “'

 

0.16

 

1 sjasaal

a sjsassal I

[CaM]. IIM

I 14]

 

 

40F

20

 

10

 

  

Bound/Free

J

 

 

 

“COMM"

l

 

 

 

7 14 21

Total CaM, nM

28

35

77

3.4 Discussion

It has been previously shown that CaM partially inhibits Ca2+ release fi'om SR vesicles
(Meissner, 1986; Plank et al., 1988) and lowers the open state probability of the channel
protein in planar lipid bilayers (Smith et al., 1989). The present studies were conducted to
characterize the interaction of CaM with the Ca”'-channel protein in heavy SR vesicles by
defining conditions in which CaM may bind to and thus potentially regulate channel protein
activity.

The porcine heavy SR vesicles used in this study contained a small but significant
amount of CaM. The fact that subsequent EGTA washes of the vesicles prior to the
endogenous CaM assay did not significantly reduce CaM levels suggests that the endogenous
CaM is non-exchangeable and tightly bound to proteins such as phosphorylase kinase
(Eibschutz et al., 1984). Our results for endogenous CaM are slightly higher than those
obtained by Meissner (1986) who obtained 6—12 pmol/mg of protein for rabbit skeletal muscle
heavy SR vesicles. The differences may result from species differences and in methods for
determination of endogenous CaM abundance.

Aflinity labeling experiments were conducted to identify the CaM-binding proteins in
our SR preparations. Wheat germ CaM, iodinated at Tyr 139, was derivatized at Cys 27
(located in the N-domain of CaM) with the photoaflinity label benzophenone-4-maleimide
(Strasburg et al., 1988). This probe is reactive with any methylene groups of amino acid
residues in the proximity of the label, and thus, proteins which bind within 1.0 nm of the

labeled Cys on CaM are readily crosslinked and identified by autoradiography. Our results

78

indicated that the major receptor for CaM in purified heavy SR preparations was the Ca2+-
clranne] protein and that the binding was specific (Figure 3.1). Light SR vesicles fi'om rabbit
skeletal muscle were devoid of Ca2*-channel protein as indicated by the low ryanodine-binding
activity, absence of a stained band in the gel, and absence of afiinity-labeled channel product
in the autoradiogram (Figure 3.1). The amount of crosslinked product in heavy SR was
somewhat greater in the presence of 1 mM EGTA than in the presence of 0.1 mM CaC12.
However, crosslinking should not be regarded as a quantitative method. Differences in
amount of crosslinked product could reflect actual differences in the amount of CaM bound,
but the fluorescence data (Table 3.1) indicated that this was not always the case. Formation
of transient, low aflinity complexes are quicldy locked in a covalent crosslink upon excitation
by light. Another explanation for the decreased yield in the presence of Ca2+ is decreased
efficiency of crosslinking owing to a Cay—dependent conformational change in the vicinity of
the crosslinker on CaM or a Ca2"-dependent conformational change in the channel protein.

Seiler et al. (1984) first observed that the primary CaM receptor in heavy SR was a
protein doublet corresponding to the channel protein (Lai et al., 1988). Their experiments
were done using mammalian CaM derivatized with the photoaflinity label, methyl-4-
azidobenzimidate which labels up to 4 difi‘erent lysine residues on CaM (Klevit & Vanaman,
1984)

In contrast to these studies, Vale (1988) observed one major CaM receptor (60 kDa),
and six minor CaM receptors (M, > 200, 148, 125, 41, 33, and 23 kDa) in SR membranes of
rabbit skeletal muscle. The discrepancy may be largely attributable to the fact that Vale used

a preparation of SR which was not fractionated on sucrose gradients. That preparation

79

therefore, would likely have comprised a more diverse membrane fiaction which contained
many more CaM binding proteins than our preparations or those of Seiler et a1 (1984).
Furthermore, it is possible that proteolysis of the channel protein yielded fragments in Vale's
preparations which bound CaM. Our preparation employed protease inhibitors to minimize
this possibility.

Since there was one major receptor for CaM present in our heavy SR preparations,
fluorescence spectroscopic techniques could be used to obtain quantitative data on the
interaction of CaM with channel in SR vesicles. Wheat germ CaM was chosen for
derivatization because it possesses a single sulfhydryl residue for chemical modification (Toda
et al., 1985). Rhodamine-x-maleimide was chosen for labeling of wheat germ CaM because
the rhodamine-CaM adduct retains biological activity, it has a high quantum yield, and its
emission maximum is in the red portion of the spectrum, thus providing a suitably strong
signal in the low nM concentration range of labeled CaM such that the light scattering
contribution to the total signal from the membrane vesicles is minimized (Mills et al., 1988;
Strasburg et a], 1988).

CaM is known to regulate more than 30 different proteins and enzymes in a Ca”-
dependent manner; i.e., Ca2+ first binds to CaM inducing a conformational change in CaM
which results in binding of the Ca,CaM complex to the target protein and modulation of
activity (see Klee & Vanaman, 1982, for review). An exception to this mechanism is the
interaction of CaM with the neural specific protein P-57 or neuromodulin, in which binding
of CaM to this receptor protein is enhanced in the presence of EGTA and reduced in the

presence of Ca2+ (Andreasen et al., 1983; Masure et al., 1986; Cimler et al. 1985). A novel

8O

aspect emerging from our fluorescence studies on CaM binding to the channel protein is that,
in the concentration range of 0. 1-0.3 M KC] (ie, in the range of physiological ionic strength),
Rh—CaM bound with high afiinity to the channel protein in the presence of EGTA. KC] had
a dramatic effect on the affinity of the channel protein for CaM. At low ionic strength in the
presence of EGTA, binding of CaM to the channel protein was minimal (Figure 3.3). Titration
of a mixture of Rh-CaM plus SR vesicles with KC] resulted in strong enhancement of the
affinity of the channel protein for Rh-CaM, possibly owing to an ionic strength-dependent
conformational change in the channel protein. Some enhancement of binding was also noted
when the KC] titration was conducted in the presence of 0.1 mM CaC]2 plus 1 mM MgC12,
although a greater fraction of Rh-CaM was bound initially and there was little change in
affinity at high [KCI]. These results are complemented by the affinity labeling experiments
(Figure 3.13) which show CaM crosslinking to the channel protein in the presence of EGTA.
Together, these data suggest that CaM binds to the skeletal SR Ca21-channel protein with
high affinity (K,I = 8.6 nM) at the [Ca”] present in resting muscle (< 100 nM), suggesting that
the mechanism of CaM interaction with and regulation of the skeletal SR Ca2+ channel protein
is difl‘erent from either of the classes of CaM receptors described previously. Although CaM
may bind to the skeletal SR Cay-channel in the presence of EGTA, Ca2+ is required for CaM
to exert its inhibitory effect on Ca2*-release (Smith et al., 1989).

Although it is difficult to accurately determine stoichiometry of binding of Rh-CaM
to the channel protein in the SR vesicles, a reasonable estimate may be made based on the
ryanodine-binding activity of the preparatiOns. The firnctional unit of the channel protein is

a tetramer of identical subunits (Lai et al., 1989), each of M, = 565,000 (Takeshima et al.,

81

1989; Zorzato et al., 1990). One mole of ryanodine specifically binds with high affinity per
mole of channel protein tetramer, resulting in negatively cooperative interaction between
subunits (Lai et al., 1989). Since our heavy SR preparations averaged 10.6 pmol/mg of [3H]
ryanodine-binding activity, this suggests that the channel protein subunit concentration is 42.4
pmol/mg SR protein. Scatchard analysis of the Rh-CaM binding data in the presence of
EGTA indicates a single class of CaM-binding sites with a BM of 229 pmol/mg, which is
consistent with 5-6 moles of CaM- binding sites per subunit.

In the presence of 0.1 mM Ca2+, there are two classes of binding sites with a B,mu for
the high aflinity class of CaM-binding sites of 54 pmol/mg, consistent with approximately 1
CaM binding site per subunit, and BM for the low affinity class of CaM-binding sites of 166
pmong corresponding to approximately 4 CaM binding site per subunit.

When 1 mM Mg2+ was included with 0.1 mM Ca“, there was a dramatic shift in the ’
interaction of CaM with the channel protein. The Bml for the high aflinity class of CaM-
binding sites was 10 pmol/mg. The Rh-CaM binding capacity of this class of sites was in
close agreement with the [’Ifl-ryanodine-binding data, suggesting that in the presence of both
Ca2+ and Mg”, there is approximately 1 mole of CaM bound with high affinity per tetramer.
The lower affinity class of sites showed a Bum,2 of 70 pmol/mg corresponding to
approximately 2 lower affinity Rh-CaM site per subunit. The effect of inclusion of Mg" on
CaM-binding may cause a confonnationa] change in the channel protein in high affinity CaM
binding sites such that there would be only one high affinity site per tetramer. Alternatively,
in the presence of Mg”, binding of one CaM to the channel protein may induce a cooperative,

allosteric effect on the channel protein resulting in decreased affinity of the other sites.

82

These estimates of CaM stoichiometry are based on the afinity labeling data which
suggest that the channel protein is the major receptor for CaM in heavy SR. We cannot
exclude the possibility that small amounts of other CaM-binding proteins contribute to the
total observed binding. However, the abundance of the channel protein in heavy SR, and the
limited CaM binding by light SR vesicles (Figure 3.2) suggest that most of the CaM binding
to the SR vesicles is via the channel protein.

These studies provide the first experimental data on the stoichiometry of binding of
CaM to the skeletal SR Ca2*-channe] protein. Previous studies employed sequence analysis
to locate potential CaM-binding sites in the channel protein sequence based on the predicted
requirement of a basic amphiphilic helix for CaM target proteins (O'Neil & DeGrado, 1990).
Takeshima et al., (1989) predicted 2 CaM binding sites in the vicinity of residues 3614-3637
and 4295-4325, whereas Zorzato et al. (1990) predicted 3 different sites at residues 2775-
2807, 2877-2898, and 2998-3016. Brandt et a]. (1992), reported that CaM inhibits calpain
digestion of the skeletal SR channel protein. Analysis of the digestion pattern of the channel
protein in the presence and absence of CaM, coupled with the use of computer algorithms to
identify consensus sequences for calpain digestion and CaM binding, led to the identification
of 3 more candidate sites for CaM binding: residues 1383-1400, 1974-1996, and 3358-3374.

Our experimental data suggest that there may be as many as 5-6 CaM binding sites
per subunit and that the affinity of each CaM site depends on the divalent metal ion
concentration. Our data further indicate that CaM binds to the channel protein under
conditions comparable to that of resting muscle ( < 100 nM CaC]2 ). This suggests that there

may be a difi‘erent structural element from that of a basic amphiphilic helix for CaM binding

83

in the channel protein, making predictions of CaM-binding domains based on sequence
analysis difiicult. Experiments are in progress in our laboratory to identify the CaM-binding

domains in the channel protein.

CHAPTER 4

ALTERED CALMODULIN REGULATION OF THE
SARCOPLASMIC RETICULUM CALCIUM RELEASE CHANNEL
PROTEIN IN MALIGNANT HYPERTHERMIA-SUSCEPTIBLE

PIGS

4.1 Introduction

Malignant hypertherrnia (MB) is an inherited myopathy characterized by an
accelerated skeletal muscle metabolism, muscle contracture, and rapidly elevated body
temperature in response to triggering agents such as halogenated anesthetics (for review see
Gronert, 1986). The major biochemical defect in MH is associated with alteration of the Ca2+
release mechanism via the sarcoplasmic reticulum (SR) (Kim et al., 1984; Mickelson et al.,
1986; Mickelson et al.,1988). Abnormal functions of native, detergent-purified, or
reconstituted Cay-channel protein in malignant hypertherrnia susceptible (MI-IS) SR include

altered ryanodine binding properties (Mickelson et al., 1990), altered single-channel activity

84

8 5
in planar lipid bilayers (Shomer et al., 1993 ), increased Ca2+ efllux rate by channel activators
(Ca2+, cafl‘eine, ATP, halothane) and a lower luminal C? threshold for induction of C?
release (O'Brien et al., 1986; Ohnishi et al.,l987; Carrier et al., 1991; Louis et al., 1992). The
abnormal regulation of Ca"1+ in skeletal muscle from MHS swine results from a point mutation
(Arg615 to Cys615) in the SR Cay-channel protein (Fujii et al., 1991).

SR Ca2*-release channel protein activity is modulated by numerous physiological and
pharmacological agents including Ca”, Mg”, caffeine, adenine nucleotides, etc. (Martonosi,
1984). The response of the mutant channel protein to these modulators may be altered in MH.
Caffeine increases the sensitivity of the Ca2+ induced Ca2+ -release mechanism to ca: and
adenine nucleotides both in skinned fibers and in SR vesicles (for review see Martonosi,
1984). The sensitivities of skeletal muscle contracture and Ca2+ release rate in skinned single
muscle fibers to caffeine are increased in the MHS pig (Britt, 1987; Gronert, 1986; Endo et
al., 1983; Ohta et al., 1989). Pessah et a]. (1987) have shown that caffeine dramatically
increases the affinity of Ca2+ binding sites which activate channel gating. Based on this finding
a hypothesis was proposed that the caffeine binding domain directly influences the sensitivity
of the Ca2*-regulatory site(s) (Pessah et al., 1987). Otsu et al. (1994) provided evidence that
the myoblastic cells expressing mutated Ca2+ release channel had higher sensitivity to caffeine-
induced Ca2+ release from SR Ca“ channel. However, Shomer et a]. (1994) have provided
evidence that the Arg615Cys mutation in Cay-channel in MHS pig skeletal muscle does not
appear to be directly responsible for the altered caffeine sensitivity of MHS muscle
contracture.

Calmodulin, a ubiquitous calcium-binding protein, may play dual roles in regulating

86

channel protein activity. It has been implicated as a Ca” channel inhibitor (Meissner, 1986;
Plank et al., 1988) and as an activator of the Ca2+ release process (Tripathy et al., 1995),
depending on Ca2+ concentration. The inhibitory effect of CaM results fi'om reduction of the
channelopenstateprobabilityviadirectbindingofCaMtothechannel proteininthepresence
of > 1 pM [Ca2+] (Smith et al., 1989). The channel activation efi‘ect by CaM occurs at nM
Ca2+ concentrations < 100 nM corresponding to that of resting muscle condition. Studies on
the stoichiometry of CaM-binding to the channel protein indicate that at the lower [Ca2*]
there is one class of high affinity CaM binding sites with 16-24 CaM molecules bound per
channel protein tetramer, or 4-6 CaM molecules per subunit (Yang et al., 1994; Tripathy et
al., 1995). In the presence of pM Ca2+, there are two classes of CaM binding sites: one high
affinity class of sites comprising 4 CaM‘s per tetramer and a low affinity class of sites ‘
comprising the remaining CaM molecules (Yang'et al., 1994).

The objective of this study was to test the hypothesis that the altered Ca”'-channel
activity present in SR from MHS swine results in part, from abnorrna] CaM regulation of the
Cay-channel. This was examined by determining the equilibrium binding constants and
stoichiometry of MHS and normal SR Cay-channel protein with CaM under defined metal
ion conditions. In addition, the efi‘ect of cafieine on the binding of CaM to SR vesicles from
MHS and normal skeletal muscle was examined to determine whether the altered caffeine
sensitivity of MHS SR Ca2*-induced Ca2+ release results from an allosteric effect of cafl‘eine
on the binding of CaM to MHS SR Ca2+ channel. The results suggest that the binding of CaM
to the channel protein in MHS SR is altered. However, the efi‘ect of cafl‘eine on the binding

of CaM to MHS and normal SR is not significantly different.

87

4 .2 Experimental Procedures

4.2.1 Materials. Berrzophenone-4-ma1eimide and rhodamine-X-maleimide were purchased
from Molecular Probes (Junction City, OR). Na[‘2’I] and Pin-ryanodine were obtained from
DuPont-NEN (Boston, MA). Ryanodine was purchased from Calbiochem (La Jolla, CA).
Wheat germ was a generous gift from Intemationa] Multifoods (Minneapolis, MN). Caffeine

was fi'om Sigma (St Louis, MO).

4.2.2 Preparation of calmodulin and its derivatives. CaM was purified from wheat germ
using the procedure described previously (Strasburg et al., 1988). Purified CaM was
derivatized at Cys 27 with rhodamine-X-maleimide as described for I-EDAN S (Strasburg et
al., 1988). For cross-linking experiments, CaM Was iodinated with [1251] at the sole tyrosine
residue (T yr-139), followed by site-specific modification at Cys-27 with the photoactivatable

cross-linker benzophenone-4-maleimide (Strasburg et al., 1988).

4.2.3 Photoafl'rnity labeling of MHS or normal SR Ca2+ channel protein with [mu-B2-
CaM. [ml]-Bz-CaM crosslinking with MHS or norrna] SR vesicles was performed by
incubating in darkness 0.1 pM [‘”I]-Bz-CaM with 100 pg of SR vesicles in 100 pL of 20 mM
HEPES pH 7.5, 0.15 M NaCl, and different metal ion conditions as described in figure
legend. The mixtures were placed in plastic microcentrifirge tubes on ice and the samples
were illuminated for 20 min in a Stratalinker 1800 photoreactor (Stratagene Corp, La Jolla,

CA) equipped with lamps of Am = 254 nM. After photolysis of the mixtures, the samples

88

were centrifirged in aBeckman ”IL-100 centrifirge at 100,000 gm for 20 min. The membrane
pellets were resu3pended in water, SDS was added to produce a final concentration of 1%,
and the samples were subjected to polyacrylamide gel electrophoresis using 5-20% linear
gradient gels (Laen'unli, 1970). The gels were dried and placed with Kodak Omat XAR-S X-
ray film in autoradiography cassettes equipped with Dupont Lightning Plus intensifying

screens.

4.2.4 Preparation of MHS and normal skeletal muscle SR vesicles. Longissirnus muscle
of pigs, homozygous for either the MHS (Pietrain) or normal (Yorkshire) allele of the Ca2+
release channel, were obtained from the swine genetics herd maintained by the Department
of Animal Science at the University of Minnesota Experimental Farm. Heavy SR vesicles
were prepared using the procedure of Mickelson et al., (1986) with the modifications that 1
mM EGTA, 0.1 mM PMSF, 1 pg/ml of aprotinin and 1 pg/ml leupeptin were included in the

homogenization bufl‘er and at each subsequent purification step.

4.2.5 Fluorescence anisotropy measurements. Rhodamine maleimide-labeled CaM (Rh-
CaM) binding to the channel protein in SR vesicles was monitored by fluorescence anisotropy
measurements using an SLM 4800 spectrofluorometer modified with data acquisition
hardware and operating system from On-Line Instrument Systems (Bogart, GA). Samples
were held in a therrnostatted cell block maintained at 22°C. The excitation wavelength of Rh-
CaM was 580 nm, monochromator slits were set at 8 nm, and emitted light was isolated using

Schott RGo610 filters. During titrations samples were allowed to equilibrate for 5 minutes

89

after each addition of Rh-CaM or SR All samples included 1 pg/ml aprotinin, 1 pg/ml
leupeptin and 0.1mM PMSF to inhibit proteolysis during the experiment.

All fluorescence measurements were performed using semi-micro, quartz fluorescence
cuvettes (4 mm x 10 mm). Prior to fluorescence experiments the cells were rinsed with 1
mg/ml bovine serum albumin to minimize Rh-CaM adsorption to the walls of the cuvette.
Following this treatment, the measured anisotropy value for Rh-CaM was independent of
concentration over the range of 1 nM to 1000 nM, indicating that there was negligible Rh-
CaM adsorption to the cuvettes.

The anisotropy of a ligand (Rh-CaM) is directly proportional to the fraction of ligand
bound to the receptor (Ca2*-channel protein). Thus, if A, is the anisotropy of fi'ee Rh-CaM
and A, is the anisotropy of the fully bound ligand, then the fraction bound, 1;, is determined

fi'om:

(Am-A1,)
fb= (1)
Am<1-q)+q<A,) —A,

 

where AIII is the measured anisotropy for a given ligand concentration, and q, the change in
quantum yield, is the ratio of fluorescent intensity of bound species over that of the free
species. If the change in quantum yield is negligible upon binding of ligand, then equation (1)
reduces to:

_ (Am-Af)

fb_____ (2)
(Ab‘Af)

90

The fiaction of ligand bound, and the concentrations of bound and free Rh-CaM are readily
calculated. Values of KI and B“, were calculated using the computer program Enzfitter, a
nonlinear regression analysis program of RI . Leatherbarrow (Biosoft, Cambridge, UK). Data
were fit to a one-ligand or two-ligand binding model to obtain the best fit.

The anisotropy of unbound Rh-CaM, A,, was measured in the absence of SR. The
anisotropy of the fully bound species, Ab, was obtained by titration of Rh-CaM with SR
vesicles, followed by curve-fitting using the Enzfitter computer program for a single class of
ligand binding sites.

Corrections for light scattering and background fluorescence were made by

application of the equation:

A=f1A1 +f2A2 '(3)

, where A, A1 and A2 are the anisotropies of the sample, the blank, and the corrected sample,
respectively. The fiactional contrrbutions, fl and f2, of these species were calculated from the
intensities measured with the excitation monochromator in the vertical position and the
emission monochromator at 55°. The corrected sample anisotropy, therefore, is that value in

the absence of background interference.

4.2.6 Biochemical Assays. SR protein concentrations were determined by the method of
Lowry et al., (1951) using bovine serum albumin as the standard. Ryanodine binding activity

of SR vesicles was determined according to the method of Mickelson et a]. (198 8).

91

4 . 3 Results

4.3.1 Afl'rnity labeling of CaM-binding proteins in MHS or normal SR vesicles. Heavy
SR vesicles purified fi'om MHS and normal porcine skeletal muscle were incubated with the
afinity labeling derivative [‘”l]-Bz—CaM under difl‘erent metal ion conditions to identify CaM
receptor proteins present in SR. The autoradiograrn of the gel electrophoretogram (Figure
4.1) indicates that the most abundant crosslinked product in both MHS and normal heavy SR
vesicles was a doublet of 114, > 450,000 which corresponds to CaM plus Ca2+ channel protein
subunit. This cross-linking complex was formed both in the presence of 1 mM EGTA and in
the presence of 0.1 mM CaClz. There were no major difl‘erences in the affinity labeling

patterns between normal and MH SR.

4.3.2 [’HlRyanodine binding to MHS or normal SR vesicles. The [3H]ryanodine binding
capacities (BM) for our MHS and normal heavy SR preparation, determined by Scatchard

analysis, were 14.6 d: 0.8 and 10.6 i 0.9 pmol/mg SR protein, respectively.

4.3.3 Titration of Rh-CaM with MHS or normal SR vesicles. Affinity labeling results
indicated that Cay-channel protein is the major CaM receptor in both MHS and normal heavy
SR. Although the affinity labeling experiments yield useful information on the identity of
CaM—binding components, it is difficult to obtain quantitative information on the interaction
of CaM with Ca” channel protein by this method. Therefore, fluorescence anisotropy was

used to define the affinity and stoichiometry of the CaM/channel protein complex formation

92

Figure 4.1. Affinity labeling of MHS and normal skeletal muscle heavy SR with [‘2’I]-Bz-
CaM. [‘2’I]-Bz-CaM was incubated with skeletal muscle heavy SR vesicles in the presence
of 20 mM Hepes pH 7.5, 0.15 M NaCl, and metal ion conditions as described below.
Following initiation of crosslinking, proteins were separated by electrophoresis on a 5-20%
acrylamide gradient gel, and the cross-linked products were identified by autoradiography of
the gel. Lanes 1-6: normal heavy SR; Lanes 7-12: MHS heavy SR Lanes 1,12: 0.1 mM CaC]2
+ 10 mM MgC12; Lanes 2,11: 1 mM CaClz; Lanes 3, 10: 0.1 mM CaClz; Lanes 4,9: 1 mM

EGTA + 10 mM MgC12; Lanes 5, 8: 1 mM EGTA + 1 mM MgC12; Lanes 6,7: 1 mM EGTA.

93

123 45678 9101112

2': W

94

in native SR vesicles from normal and MHS porcine skeletal muscle. Heavy SR vesicles were
titrated into Rh—CaM under three different metal ion conditions: (1) + 1 mM EGTA; (2) + 0.1
mM C30,; (3) + 0.1 mM CaC]2 and 1 mM MgC12. In each case, titration of SR vesicles into
a solution containing Rh-CaM resulted in a large increase in fluorescence anisotropy
indicating the increased molecular mass resulting fi'om Rh-CaM binding to the channel protein
(Figure 4.2).

The titration data fiom Figure 4.2 were used to determine the anisotropy values for
the fi'ee Rh-CaM species (A,) which is obtained in the absence of SR vesicles and for the firlly
bound Rh-CaM/channe] protein complex (A). The A, values obtained for each metal ion
condition in MHS SR titrations were 0.2483 (+ 1 mM EGTA), 0.2279 (+ 0.1 mM CaClz),
0.217 (+ 0.1 mM CaClz, + 1 mM MgClz). With normal muscle SR titrations, the A, values
obtained for each metal ion condition averaged 0.2435. The A, values obtained for Rh-CaM
under the three metal ion conditions in the absence of MHS or normal SR were 0.1824 and
0.1908 (+ l mMEGTA), 0.175 and 0.1728 (+ 0.1 mM CaClz), 0.1573 and 0.1585 (+ 0.1 mM
CaC], + 1 mM MgC],) respectively. The subtle difference in A, value for MB vs. normal SR

results from different labeled CaM preparations used for these experiments.

4.3.4 Rh-CaM/Ca2+ channel protein of MHS or normal SR binding equilibrium. Our
previous studies have determined the equilibrium binding constants of CaM to the channel
protein from genetically defined normal SR vesicles. In order to determine whether the altered
SR Ca2+ channel activity in MH is associated with abnormal CaM regulation of channel

activity, the aflinity and stoichiometry of CaM binding to MHS SR were determined. Rh-CaM

95

Figure 4.2. Titration of Rh-CaM with MHS and normal skeletal heavy SR vesicles under
different divalent ion conditions. The sample medium contained 10 nM Rh-CaM, 0.3 M
sucrose, 0.3 M KC], 50 mM PIPES, pH 7 .0 plus one of the following conditions: 1 mM
EGTA(A);0.1mMCaC12(O);0.1mMCaCl plus 1 mMMng,(O)inastarting
volume of 1 mL. Panel A and B represent SR vesicles purified from MHS and normal swine,
respectively. The Rh—CaM sample was titrated with SR vesicles in parallel with a buffer blank
containing the same media minus Rh-CaM. Corrections were made for light scattering as .

described in "Experimental Procedures".

>

AnisotrOpy

Anisotmpy

 

0.26

0.24

0.20

0.18

0.16

 

0.14

0.25

0.23

0.21

0.19

0.17

0.15

1 J 1111111 4 1 1 111111

1

 

j l Ill]!

 

10 100

[HSR], uglml

1000

 

l

 

1 4 1111111 1- 1 1111111

1

I 11111!

 

 

10 100

[HSR], ug/ml

1000

97

was titrated into fixed concentrations of SR vesicles under the following conditions: (1) 1 mM
EGTA; (2) 0.1 mM CaClz; (3) 0.1 mM CaC]2 plus 1 mM MgC12. The fluorescence intensity
did not change significantly upon the binding of Rh-CaM to Ca2+ channel protein. Therefore,
the fiaction of Rh-CaM bound to channel protein in SR vesicles for each titration point was
calculated by anisotropy values A, and A from Figure 4.2 using eq. 2. (Experimental
Procedures).
Analysis of fluorescence anisotropy data fi‘om Rh-CaM titration into MHS and normal
SR Ca2+ channel protein indicates that there are significant difl‘erences in affinity and
stoichiometry of CaM/Ca2+ channel interaction under these three conditions. In the presence
of EGTA there is a single class of CaM-binding sites on both MHS and normal SR Ca2+
channel protein. However, the binding capacities (IBM) and dissociation constants (K,) differ
substantially between these two proteins. IBM and Kd for MHS SR are 164 d: 4 pmol/mg and
4.2 :1: 0.3 nM, respectively, for normal SR BM = 229 :1: 7 pmol/mg and the K,I = 8.6 :1: 0.8 nM
(Figure 4.3 and Table 4.1). In the presence of 0.1 mM CaClz, the binding of CaM to
channel protein was shifted to two classes of ligand-binding sites. The high affinity class of
binding sites displayed a BM, of 45 d: 7 pmol/mg and KM of 1.6 :1: 0.5 nM for MHS SR, and
for normal SR BM, = 54 d: 7 pmol/mg, and K, = 4.3 i 1.1 nM. The low affinity class of
binding sites shows a Bm2 = 73.6 :t 24.9 pmol/mg and K,12 = 84.8 :1: 81.6 nM for MHS SR,
versus normal SR which shows a BM of 166 i: 28 pmol/mg and a IQ, of 239 i 102 nM
(Figure 4.4 and Table 4.1).
In the presence of 0.1 mM Ca2+ plus 1 mM Mg”, there was a dramatic shift of high

affinity binding-site class to low affinity binding-sites class. The BM, and Km for MHS SR

98

Figure 4.3. Titration of skeletal heavy SR vesicles with Rh-CaM in the presence of EGTA.
Rh-CaM/SR saturation binding curve. The inset is a Scatchard plot of Rh-CaM binding to SR
vesicles. The sample medium contained 90 pg of MHS ( O ) or normal ( O ) heavy SR, 0.3
M sucrose, 0.3 M KC], 50 mM PIPES, pH 7.0 and 1 mM EGTA in a starting volume of 1
mL. Points represent the means i SE. of 3 preparations. CaM bound was calculated from

measured anisotropy as described in "Experimental Procedures".

Bound CaM

250

200

150

100

50

99

 

 

 

 

 

 

 

 

 

 

Total CaM, nM

 

100

Figure 4.4. Titration of skeletal heavy SR with Rh-CaM in the presence of CaClz. Rh-
CaM] SR saturation binding curve. The inset is a Scatchard plot of Rh-CaM binding to SR
vesicles. The sample medium contained 150 pg MHS ( O ) or normal ( O ) heavy SR, 0.3
M sucrose, 0.3 M KC], 50 mM PIPES, pH 7.0 and 0.1 mM CaC]2 in a starting volume of 1
mL. Points represent the means :1: SE. of 3 preparations. CaM bound was calculated from

measured anisotropy as described in " Experimental Procedures".

 

Bound CaM, pmol/mg

101

 

150

120

90

 

60

30

 

 

 

 

 

Total CaM, nM

102

Table 4.1. Equilibrium constants for Rh-Cau Interaction with

the caa'nelease Channel Protein in.Norna1 and.uns 8R vesicles'

 

 

 

 

 

 

 

 

 

Burl K41 Bull K42
(pmollngl (n!) (plot/Is) (an)
normal
+1mn EGTA 22917 a.61o.a - -
+0.1nn Carolz 5417 4.311.1 166128 2391102
+0.1mM CaClz 10.010.8 0.1010.03 7012 1711
+1mn rigorz
ans
+1nn EGTA ‘ 16414 4.210.: - -
+0.1nx Cac1z 4517 1.61o.5 73.6124.9 84.8181.6
+0.1mM Cac12 14.91o.a o.os1o.oz 243121 57.11a.6
+1nx ugcr2

 

 

 

 

 

 

 

* Data were obtained from titrations of SR vesicles with Rh-
CaM in the presence of 0.3 M KCl, 50 mM Pipes, pH 7.0, and
divalent ion conditions as listed below. Data are 1 SE of the

means of three preparations each.

103

were 14.9 :1: 0.8 pmol/mg and 0.05 1 0.02 nM respectively and for normal SR these values
were 10.0 :1: 0.8 pmol/mg and 0.1 :t 0.03 nM respectively. The BM and K,2 for MHS were
248 :1: 24 pmol/mg and 57.4 :1: 8.6 nM, whereas normal SR shows a BM of 70 1 2 pmol/mg
and a K,2 of 17 1 1 nM (Figure 4.5 and Table 4.1).

The stoichiometry of binding of Rh-CaM to Ca2*-channel protein in the SR vesicles
may be estimated by dividing the Rh-CaM binding activity by the ryanodine binding activity.
One mole of ryanodine binds per mole of channel protein tetramer (Lai et al., 1989);
therefore, in the presence of EGTA, the CaM—binding stoichiometry of the MHS Ca2*-channe]
protein is consistent with 11.2 mole of CaM-binding sites per mole of channel tetramer or
about 3 sites per monomeric subunit (Table 4.2). In the presence of 0.1 mM CaClz, the high
afinity CaM-binding site of the MHS Ca2*-channel with a BM, of 45 pmol/mg is consistent
with 3.1 moles of CaM-binding sites per channel tetramer, or approximately one high afiinity
CaM binding site per subunit (Table 4.2). In the presence of Ca2+ plus Mg”, the high aflinity
CaM-binding sites in MHS SR Cay-channel with a er of 14.9 pmol/mg is consistent with
approximately one CaM-binding site per tetramer, which is not significantly different fi'om the

number of CaM-binding sites in normal SR Cay-channel (Table 4.2).

4.3.5 Ionic strength dependence of the binding of Rh-CaM to MHS or normal SR
vesicles. Our previous results suggested that binding of Rh-CaM to the Ca2+ channel protein
of normal skeletal muscle SR was highly ionic strength dependent (Yang et al., 1994).
Although the basis for this ionic strength effect is not clear, it is likely that changes in ionic

strength alter channel protein structure in such a way that affinity for Ca2+ is enhanced at

104

Figure 4.5. Titration of skeletal heavy SR with Rh-CaM in the presence of CaC]2 plus MgClz.
Rh-CaM/SR saturation binding curve. The inset is a Scatcth plot of Rh-CaM binding to SR
vesicles. The sample medium contained 50 pg of MHS ( O ) or normal ( O ) heavy SR, 0.3
M sucrose, 0.3 M KC], 50 mM PIPES, pH 7.0 and 0.1 mM CaC]2 plus 1 mM MgCl, in a
starting volume of 1 mL. Points represent the means :t SE. of 3 preparations. CaM bound

was calculated from measured anisotropy as described in " Experimental Procedures".

Bound CaM, pmol/mg

105

 

 

 

 

 

 

 

 

130
104 i' T
78 ' T
52 b . s F
e -‘- g L
.26 ~ . i w
0 . 1 . . “rm
0 10 20 30 40 50

Total CaM, nM

106

Table 4.2. Stoichiometry of High Affinity Class of Rh-Calt

Binding to Normal and x38 Skeletal Muscle SR Ca’*-Release

 

 

 

 

 

 

 

 

 

Channel'
High affinity Can Low affinity Can
binding class (moles binding class (moles
Can per mole of Can per mole of
channel tetra-er) channel tetra-er)
Normal
+1 n11 EGTA 21.6 -
+0.1 mu Carol2 5.5 15.7
+0.1 all! CaClz l 6.6
+1 mu HgClz
11118
+1 .14 EGTA 11.2 -
+0.1 m1! CaClz 3.1 5.1
+0.1 nu eac1’ 1 17
+1 111! HgClz

 

 

 

 

 

* The stoichiometries of high affinity Rh-CaM binding to
normal and MHS SR Ca‘2+ channel are estimated on the base of

ryanodine binding data 10.6 i 0.9 and 14.64 1 0.83.

107

physiological ionic strength Thaefore, to determine whether the Arg615Cys mutation might
have affected the ionic strength dependence of CaM binding to the channel protein, Rh-CaM
binding was determined for normal and MHS SR as a function of [KCI]. The SR
concentrations were fixed at levels corresponding to the concentrations of species at the mid-
points of the titration curves. in Figure 4.2.

In the presence of EGTA (Figure 4.6A), the binding of Rh-CaM to both normal and
MHS SR Caz“ channel protein increased as KC] concentration increased reaching an optimum
in the range of 0.1-0.3 M KC]. At [KCl] > 0.3 M, the binding of Rh-CaM to the channel
protein rapidly decreased. In the presence of 0.1 mM CaC]2 plus 1 mM MgC12, more Rh-CaM
was initially bound to Ca” channel protein in the absence of KC], and there was a gradual
increase for both normal and MHS SR as KC] concentration increased from 3 mM to 0.7 M
(Figure 4.6B). However, in contrast to the EGTA conditions, binding of Rh-CaM to the
channel protein did not significantly decrease at higher KC] concentrations in the presence of
Ca2+ and Mg”. The results fiom Figure 4.6 suggest that differences in CaM-binding between

MHS and normal SR are not a result of altered ionic strength-induced structural change.

4.3.6 Caffeine dependence of the binding of Rh-CaM to MHS or normal SR vesicles.
It has been suggested previously that the mutation in the MHS Ca2+ release channel is not
directly responsible for the altered caffeine sensitivity of MHS pig muscle contracture.
(Shomer et al., 1995). Rather, this altered cafl‘eine sensitivity may alter the response of the
channel protein to other physiological channel modulators. To determine whether caffeine

binding could affect the affinity of the normal channel protein for CaM, caffeine was titrated

108

Figure 4.6. KC] dependence of the binding of Rh—CaM to SR vesicles under different divalent
ion conditions. The sample buffer contained 100 pg MHS ( O ) and 86 pg normal ( O ) heavy
SR vesicles, 10 nM Rh-CaM, 0.3 M sucrose, 50 mM PIPES, pH 7.0 and either 0.1 mM CaC]2
plus 1 mM MgCl2 (panel A), or 1 mM EGTA (panel B) in a starting volume of 1 mL. Points
were representatives of 3 preparations. % CaM bound was calculated from measured
anisotropy as described in "Experimental Procedures"; 100% bound corresponds to 229
pmol/mg in the presence of EGTA and 80 pmol/mg in the presence of CaC]2 plus MgCl2 for
normal SR measurement. For MHS SR, 100% bound corresponds to 164 pmol/mg in the

presence of EGTA and 263 pmol/mg in the presence of CaC]2 plus MgClz.

109

 

 

 

 

 

A
100
80 "
"U
:3 60
O
a:
r— O
20 "
o 1 1 111L111 1. 1 1 111111 1 1 1 11111
1 10 100 1000
[KCl],mM
I!
100
80 b
'2 60 -
:1
O
m
°\° 4o —
20 ”

 

 

 

 

o l l IIJLJI 1 1 1111111 1 I 11111!

1 10 100 1000

[KC]], mM

110

into a fixed concentration oth-CaM and SR fi'om normal muscle in the presence of 100 pM
CaC]2 plus 1 mM MgC], These results were compared with those obtained for MHS SR to
determine whether the altered caffeine sensitivity of MHS muscle contracture might be caused
by an allosteric effect on CaM binding to and regulation of SR Ca2+ channel protein.

Figure 4.7 indicates that as the caffeine concentration was increased fiom 0.1 mM to
100 mM, the fluorescence anisotropy decreased for both MHS and normal SR indicating that
the binding of Rh-CaM to Ca2+ channel in both MHS and normal skeletal SR decreased from
90% bound to 50% bound as caffeine concentration increased. There are no significant
difference between MHS and normal SR in caffeine effect on the binding of CaM to Ca”-

channel.

111

Figure 4.7. Cafl‘eine dependence of the binding of Rh-CaM to HSR vesicles. The sample
buffer contained 100 pg of MHS ( O ) or normal ( O ) heavy SR vesicles, 10 nM Rh-CaM,
0.3 M sucrose, 50 mM Hepes (pH 7.0), 100 pM CaClz, 1 mM MgCl2 in a starting volume of
1 mL. Points were representatives of 3 preparations. % CaM bound was calculated from
measured anisotropy as described in "Experimental Procedures"; For MHS and normal SR,

100% bound corresponds to 263 and 80 pmol/mg, respectively.

% Bound

112

 

 

 

14111111] 1‘1111111J 1 1111111

 

 

l 10

[Caffeine], mM

113

4.4 Discussion

A mutation in the Ca2"-channel protein is responsible for the abnormalities in calcium
regulation observed in MHS porcine skeletal muscle, whereas other aspects of SR structure
and function are normal (Carrier et al., 1991; Fill et al., 1990; Mickelson et al., 1988;
Mickelson et al., 1989; Nelson, 1983; Ohnishi et al., 1983; Ohta et al., 1989). The altered
Ca2*-release properties of the MHS channel protein in the muscle cell could result, in part,
from aberrant regulation of channel activity by other physiological modulators of Ca2+ release.
The intracellular Cay-binding protein, CaM, binds to the channel protein and acts not only
as an inhibitor of Ca” release (at [Caf'] > 0.1 pM) (Meissner, 1986; Plank et al., 1988; Smith
et al., 1989; Fuentes et al., 1994), but may also play a role in activating the channel protein
during the resting state of muscle (Tripathy et al., 1995; Buratti et al., 1995; Ikemoto et al.,
1995). To determine whether the MH mutation alters CaM-dependent regulation of the
channel protein, we compared the stoichiometry and affinity of normal and MHS channel
proteins for CaM in SR vesicles.

The affinity labeling studies (Fig. 4.1) reveal that the Ca2+ channel protein is the major
CaM-binding protein in heavy SR, and firrther suggest that CaM binds to the channel in the
presence or absence of Ca2+ concentrations which correspond to muscle contraction or to the
resting condition, respectively. The crosslinking experiments provide valuable information on
the identity of CaM-binding proteins in SR vesicles, but the nature of the technique makes it
difficult to make interpretations of a quantitative nature. In particular, changes in structure

induced by metal ions may subtly change the orientation of the crosslinker relative to the

114

receptor protein, thus enhancing or decreasing crosslinking emciency. Weak complexes (cg.
K4~l-1000 pM) are not readily distinguished from strong complexes (Kd~1-10 nM) because
transient complexes formed during the excited state lifetime of the benzophenone probe will
be covalently joined as well as strong complexes.

Fluorescence anisotropy was therefore used to obtain quantitative data on CaM-
binding to SR vesicles. The fact that the channel protein is the major CaM receptor (Fig. 4.1)
in both normal and MHS SR vesicles suggests that increases in anisotropy are attributable
primarily to CaM binding to the channel protein. Although the autoradiographic results of
MHS and normal SR do not reveal a significant difference in the amount of CaM-channel
protein complex formed for reasons mentioned above, fluorescence anisotropy studies
demonstrated the altered CaM binding properties with Ca2+ channel in MHS skeletal muscle
SR compared to normal SR The reduction in the number of CaM binding sites in MHS SR
from 21 CaM mole/tetramer in normal SR to 11 mole/tetramer (Table 4.1), suggests that up
to half of the high affinity of CaM binding sites are lost as a result of the mutation in MHS
Ca2+ channel. However, the CaM binding amnity of the remaining sites did not differ
significantly in MHS SR compared to normal SR at low [Ca2+]. Because of differences in
stoichiometry, these results imply that regulation of the channel protein by CaM in MH is
somehow altered.

Normal skeletal muscle SR binds 4-5 CaM per channel protein subunit with high
affinity at < 0.1 pM [Ca2*] (Yang et al., 1994; Tripathy et al., 1995). The significance of CaM
binding at resting Ca2+ levels becomes apparent from the studies of Tripathy et al. (1995) and

of Buratti et al. (1995) who showed that CaM activates the skeletal muscle SR Ca” channel

115

at low Ca2+ (<0.1 pM). The reduction in number of CaM binding sites in the presence of
EGTA observed in the study presented here suggests that the CaM-dependent activation
process is defective in MH. Preliminary reports by O'Driscol] et a]. (1996) indicate that the
CaM-dependence of activation of MH SR is enhanced compared to normal as determined by
ryanodine binding. In the presence of CaM, ryanodine binding by MH SR was more than
double that of normal porcine SR. Since ryanodine binding is an indicator of the activated
state of the channel, this result suggests that the MH mutation results in hypersensitivity of
the channel protein to CaM-activation at [Ca2”] < 100 nM. The CaM binding sites lost in MH
may somehow be associated with maintaining a more stable activated state of the channel at
low [Ca”]. Alternatively, the loss of CaM sites may be an indirect consequence of the
mutation. We are presently unable to discriminate between these possibilities.

When 0.1 mM CaC]2 or CaC]2 plus MgClz were included in the titration, some of CaM
binding sites in Cay-channel protein are shifted from high-affinity class to low-afinity class
for both MHS and normal SR. In the presence of 0.1 mM [Ca2+], the stoichiometry for high
afinity CaM-binding sites is reduced from 5 CaM mole per tetramer in normal SR to 3 mole
per tetramer in MHS SR and the low affinity CaM-binding stoichiometry for MHS SR is
reduced from 16 mole to 4 mole per tetramer. However, the binding affinity for both MHS
and normal SR at either high or low affinity classes are not significantly different. Our high
aflinity CaM binding results are in agreement with the data of Tripathy et al.(1995) in which
an increase from subpM to mM [Ca2+] led to the dissociation of 12 of the 16 bound CaM
fiom Ca2+ channel complex with an appreciably slower time course (Tl,2 = 1 min). Ryanodine

binding and single-channel measurements indicate that CaM partially inhibits Ca’*-channel

116

activity in the presence of 50 pM to 1 mM [Ca’*] and vesicle-“Ca2+ emux study shows a
rapid (<2 3) CaM inhibition of "Ca” efflux fi'om SR vesicles (T ripathy et al.,1995).
Therefore, the decrease of CaM binding sites in MHS SR Cazflchannel at high [Ca2+]
apparently results from the mutation in the Caztchannel polypeptide which causes the release
of channel inhibition by CaM and led to abnormal higher myoplasmic [Ca2+].

Our data are consistent with Tripathy et a]. (1995) with respect of 1 high affinity site
and 3-4 low affinity sites per subunit in norrna] muscle channel protein. They have reported
that most of the inhibitory response of CaM at contractile [Ca2+] results fi'om the high afinity
site. These results suggest that the inhibitory response of CaM at high Ca” would not be
affected in MH. O'Driscoll et al. (1996) support this hypothesis with evidence showing that
the CaM-dependence of ryanodine binding is not altered in MH.

KC] is known to alter functional properties of the channel protein as indicated by
enhanced binding of ryanodine at higher ionic strength. The unusual efl‘ect of KC] on
increasing the affinity of CaM for channel protein in normal heavy SR has been previously
shown (Yang et al., 1994). Since the binding equilibria of CaM to channel protein were
significantly different between MHS and normal SR, we wanted to determine whether there
was a difl‘erentia] response of channel protein binding of CaM between MH and normal SR
inducible by KC]. The fact that there was no significant difference in MH vs normal SR
suggests that the structural changes induced in the channel protein by higher salt
concentration does not differentially afl‘ect CaM binding.

It has been reported previously that the cafl‘eine sensitivity of MHS skeletal muscle

fiber bundles to induce muscle contracture was altered, this difference has served as the basis

117

of the diagnostic test for MH susceptibility (Britt, 1987). However, evidence from single-
channel recording showed that there was no significant difl‘erence in the cafl‘eine sensitivities
of purified MHS and normal porcine SR Ca2+ release channels (Shomer et al., 1994).
Therefore, the increased caffeine sensitivity of the MHS skeletal muscle contracture may be
caused by other channel modulators which lead to the increase of myoplasmic [Ca2*]. CaM
has been recently reported to activate, at resting muscle [Ca2+] condition and inhibit, at higher
[Ca2+] condition, skeletal muscle SR Ca” channel activity.

In conclusion, our results provide direct evidence that the binding affinity and
stoichiometry of CaM to the mutant Cay-channel protein is altered in [C8 ],-dependent
manner in MHS SR comparing to normal SR The increase of intracellular [Cay] in MHS pig
skeletal muscle results, in part, from the altered CaM binding and regulation in SR Ca”-

release channel.

CHAPTER 5

LOCALIZATION OF CALMODULIN BINDING DOMAINS IN THE
CALCIUM RELEASE CHANNEL (RYANODINE RECEPTOR) OF

SKELETAL MUSCLE SARCOPLASMIC RETICULUM

5.1 Introduction

The release of Ca2+ from the lumen of the sarcoplasmic reticulum (SR) into the muscle
myoplasm occurs via a Cay-release channel, also known as the ryanodine receptor (RyR), in
response to transverse tubular (T-tubular) depolarization (Rios et al., 1991). This RyR/Ca2+-
channel corresponds morphologically to the 'foot' structures which span the gap at the triad
junction between the terminal cisternae of the sarcoplasmic reticulum and the T—tubule (Inui
et al., 1987a; Inui et al., 1987b; Franzini-Annstrong and Nunzi, 1983). The skeletal muscle
Ca2*-channel protein is a homotetrameric protein composed of four subunits of 5032 (human),
or 5037 (rabbit) amino acids, each with an estimated molecular weight of 565 kDa
(Takeshima et al., 1989; Zorzato et al., 1990).

The SR Cay-channel activity is modulated by numerous compounds. The activators
of SR Ca2+ release include Ca2+, adenine nucleotides, cafl‘eine, halothane and nM ryanodine,

118

119

whereas inhibitors include Mg”, mM Ca”, pM ryanodine, ruthenium red, and calmodulin
(Meissner, 1986). Recent studies suggest that the role of CaM in regulation of the channel
protein is more complex than previously thought. Four to five CaMs bind with high affinity
per channel subunit at low Ca2+ concentration (<0.1 pM) and there is only one high aflinity
CaM-binding site per channel subunit at [Ca2+] > 10 pM (Yang et al., 1994; Tripathy et al.,
1995). The combination of "Ca” efflux measurements fi'om SR vesicles, single-channel
recording data, and [3H]ryanodine binding measurements show that, at <0.2 pM Ca2+, CaM
activates the Ca2+ release channel several fold (T ripathy et al., 1995; Buratti et al., 1995;
Ikemoto et al., 1995). However, at pM to mM Ca2+ concentrations, CaM inhibits the skeletal
muscle Ca2*-release channel 2-3 fold by inhibiting the open state probability of the channel
(Meissner, 1986).

From molecular cloning and sequencing of the cDNA of the ryanodine receptor, a
crude structural model of the ryanodine receptor has emerged. Hydropathy analysis suggests
the receptor molecule has a short cytoplasmic C-terminus and either four (Takeshima et al.,
1989) or ten (Zorzato et al., 1990) transmembrane domains in the C-terminal one-fifth of the
molecule. The bulk of the molecular mass, the N-terrninal portion (80%) of this protein, is
predicted to comprise the cytoplasmic "foot" region. Three-dimensional reconstructions have
been made fi‘om electron nricrographs of negatively stained and frozen-hydrated, solubilized
Caz’fichanne]. These reconstructions support the predicted structural model of ryanodine
receptor, revealing a large cytoplasmic portion (29 x 29 X 12 nm) consisting of many
structural domains, and a smaller transmembrane assembly projecting off of the luminal side

of the SR and embedded in the bilayer membrane (Wagenknecht et al., 1989; Raderrnacher

120

et al., 1992; Radermacher et al., 1994; Serysheva et al., 1995).

Several experimental and theoretical approaches have been employed to identify the
modulator binding sites in order to characterize the relationship of structure to function within
the channel protein Using predictive algorithms to identify basic amphiphilic helices, several
putative CaM binding sites in the ryanodine receptor have been predicted through the analysis
of the deduced primary sequence (Takeshima et al., 1989; Zorzato et al., 1990). Takeshima
et al. (1989) predicted 2 CaM-binding sites at residues 3614-3637 and 4295-4325, whereas
Zorzato et al. (1990) predicted 3 different sites at residues 2807-2840, 2909-2930 and 3031-
3049. Experimental studies from various laboratories have also provided evidence for multiple
binding sites on each subunit. Marks et al. (1990) have suggested that there are four CaM
binding sites located at residues 2641-2657, 3362-3374, 3947-3965 and 4309-4322, based
on experiments utilizing limited proteolysis of the channel protein coupled with surface
topography analysis. Brandt et al. (1992) have suggested three other candidate sites for CaM
at residues 1383-1400, 1974-1996 and 3358-3374 by analysis of the calpain digestion pattern
of the channel protein in the presence and absence of CaM.

Other experimental approaches to identify and characterize CaM binding sites involve
expression of RyR cDNA fragments as firsion proteins followed by SDS-PAGE and CaM
overlay procedures. Using this approach, Menegazzi et a]. ( 1994) have defined three CaM
binding regions, residues 2937-3225, 3546-3655 and 4425-4621 in rabbit skeletal muscle
RyR Using the same approach, Chen et al. (1994) have localized three strong CaM-binding
sites in the channel protein between amino acid residues 2063 -2091, 3611-3642, and 4303-

4328, and three weaker CaM-binding sites between amino acid residues 921-1 173, 2804-2930

1 2 l
and 2961-3084 (Radermacher et al., 1994; Serysheva et al., 1995). Wagenkrrecht et a1. (1994)
have used gold-cluster-labeled CaM and electron microscopy to localize one CaM-binding
site per subunit on the purified protein. Their results suggest that this CaM-binding site is at
least 10 nm from the transmembrane channel of the receptor complex.

The susceptibility of the Ca” release channel to proteolysis has been used to study the
relationship of primary structure and firnction. Mild digestion of heavy SR membranes with
trypsin results in the rapid disappearance of the 564 kDa Ca2+ release channel protein subunit
in SDS-PAGE. However, significant alterations in the sedimentation coefiicient,
ultrastructure, high affinity [’Heranodine binding, and channel gating are observed only after
extensive proteolytic digestion (Chu et al., 1988; Shoshan-Barmatz and Zarka, 1988;
Meissner et al., 1989; Rardon et al., 1990). These results suggest that limited proteolysis of
the channel protein does not substantially alter structure or function; i.e., only after extensive
proteolysis does the channel protein lose its structural and firnctional integrity. Calcium-
activated neutral proteases (CANP), also known as calpains (calcium-dependent papain-like),
are a group of cysteine endopeptidases with neutral pH optima and are absolutely dependent
on Ca2+ for catalytic activity. They have been frequently used to generate proteolytic
fragments for structure-function studies (Murachi, 1983; Pontremoli and Melloni, 1986;
Sumki, 1987), and it has been shown that the Ca2+ release channel was the primary substrate
of calpain in heavy SR membranes (Rardon et al., 1990).

Maj or tryptic fragments and calpain proteolysis products have been identified and
positioned in the SR Cay-release channel protein sequence (Chen et al., 1993; Brandt et al.,

1992; Marks et al., 1990). Using immunoblotting analysis with site-specific antibodies, Chen

122

et al. (1993) identified 7 tryptic cleavage sites in Ca2*-channel protein, 3 of which were not
in the regions previously identified by Marks et al. (1992). Callaway et al. (1994) have
isolated and partially sequenced the polypeptides arising from limited trypsin digestion of the
Ca2*- channel complex which results in a 28 8 complex.

In this study, we have combined the approach of using limited trypsin and calpain
digestion of the native Cay-release channel in the rabbit skeletal muscle SR membranes,
coupled with crosslinking of radiolabeled CaM to the channel protein to position CaM-
binding sites in the native protein structure. Two CaM binding sites in the central regions of
Ca2*- release channel protein have been tentatively identified by immunoblotting analysis with

site-specific antibodies.

123

5 . 2 Experimental Procedures

5.2.1 Materials. Benzophenone4-maleimide was purchased fiom Molecular Probes
(Junction City, OR). NamI was obtained fi’om DuPont-NEN (Boston, MA). Wheat germ was
a generous gifl from International Multifoods (Minneapolis, MN). CHAPS, CAPS, and

MOPS were obtained fi'om Sigma.

5.2.2 Preparation of rabbit skeletal heavy SR vesicles. SR membranes were prepared from
rabbit back and hind leg white skeletal muscle by differential centrifugation and were firrther
purified using sucrose gradient centrifirgation (Hamilton and Tate, 1991; Rosemblatt, et al.,

1981).

5.2.3 Preparation of CaM and its derivatives. CaM was purified from wheat germ using
the procedure described previously (Strasburg et al., 1988). Purified CaM was iodinated with
125I at the sole tyrosine residue (Tyr-139), followed by site-specific modification at Cys-27

with the photoactivatable cross-linker benzophenone-4-maleimide (Strasburg et al., 198 8).

5.2.4 Limited tryptic digestion of heavy SR vesicles. Heavy SR was trypsinized at 37 °C
for 5 min in 0.3 M KC], 0.1 mM CaClz, and 50 mM MOPS, pH 7.0. The optimal ratio for
trypsinzheavy SR protein varied from prep to prep, but was usually 121000 to 1:500 (w/w)
conditions. The reaction was quenched with a 20-fold weight excess of soybean trypsin

inhibitor protein.

124

5.2.5 [mu-Bz-CaM cross-linking with SR vesicles. Aflinity labeling of tryptic channel
peptides were performed by incubating in darkness 1.0 pM [‘un-Bz-CaM with 10 mg trypsin
or non-trypsin treated SR vesicles in 1 mL of 20 mM Hepes pH 7.0, and 0.3 M KC] and 1
mM EGTA for 30 min. The mixtures were illuminated for 5 minutes in a Stratalinker 1800
photoreactor equipped with lamps of A“, = 254 nm. After photolysis of the mixtures, a
portion of the samples were subjected to SDS-polyacrylamide electrophoresis (Schagger and

Jagow, 1987).

5.2.6 Sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE). SDS-
PAGE was performed according to the method of Laernmfi (1970) which has been previously
described (Yang et al., 1993); and the method of Schagger and Von Jagow (1987), using 1.5
mm thick, 10% polyacrylamide separating gel (10% acrylamide, 3% bis-acrylamide), 1 M
Tris-HCl, pH 8.45, 0.1% SDS, 0.13% glycerol (w/v), 10% ammonia persulfate, 10 p]
TEMED and a stacking gel (4% acrylamide, 3% bis-acrylamide), 1 M Tris-HCl, pH 6.8, 0.1%
SDS, 10% ammonia persulfate, 10 pl TEMED. Samples were denatured for 5 min at 95 °C
in 50 mM Tris, pH 6.8, containing 4% SDS, 0.01% Serva Blue G, and 12 % glycerol. All
samples were reduced with 20 mM dithiothreitol (DTT) prior to electrophoresis. The lower
chamber running buffer contained 0.2 M Tris-HCl buffer, pH 8.9. The upper chamber running
buffer contained 0.1 M Tris-HCl buffer, 0.1 M Tricine, 0.1% SDS, pH 8.25. Electrophoresis
was conducted at 4 °C and constant voltage (80 volts). Gels were stained with 0.1%
Coomassie Brilliant Blue R-250 in 50% methanol, 10% acetic acid and destained with 40%

methanol, 10% acetic acid. The dried gels were placed with Kodak Omat XAR-S X-ray film

1 2 5
in autoradiography cassettes equipped with Dupont Lightning Plus intensifying screens. The

other portion of the samples were subjected to sucrose gradient purification.

5.2.7 Calpain digestion of heavy SR vesicles crosslinked with [mu-Bz-CaM. Calpain
digestion of [‘”I]-Bz—CaM labeled SR vesicles was performed in a digestion bufl‘er containing
0.1 mM CaClz, 50 mM NaCl, 20 mM MOPS, and 2 mM DTT (pH 7.4) at 36 °C for 1.5 hr.
The calpain/SR weight ratio was 1/50 (w/w). Digestion was terminated by the addition of
leupeptin to a final concentration of 10 pM The digested mixtures were subjected to sucrose

gradient centrifirgation for firrther purification of [‘”I]-Bz-CaM bound Cay-channel peptides.

5.2.8 Sucrose gradient purification of [mu-Bz-CaM bound Ca“-channel proteolytic
polypeptides. The proteolytic Ca2*-channel/[’2’I]-Bz-CaM complex was isolated fi'om SR
membranes by sucrose gradient centrifugation. The samples were solubilized in 2% CHAPS,
followed by density gradient centrifugation through continuous 5 - 20% sucrose for 18 hr at
110,000 x g in a Beckman SW 28 rotor as described by Callaway et a]. (1994). The fractions
from the sucrose gradients containing the peak of [”51] or [3H] radiative signal were pooled
and concentrated. The concentrated fractions were prepared for blotting and then subjected

to N-terrninal sequencing according to the method of Callaway et a]. (1994).

5.2.9 Preparation of samples for amino acid sequencing and Western blots. SDS-
polyacrylamide gels were cast 24 h in advance. Sodium thioglycolate (0.1 mM) was added

to the upper cathode running buffer to scavenge free radicals. After electrophoresis, the

126

separated bands of protein were transfered to Irnmobilon P-SQ membranes (Millipore Corp.,
Bedford, MA) for sequencing or Irnmobilon P membranes for irnmunoblot at 20 Volts in 5%
methanol, 10 mM CAPS (pH 11.0) transfer buffer at 4 °C overnight. The bands were stained
with Coomassie Brilliant Blue and cut out for sequencing according to the method of

LeGendre and Matsudaira (1989).

5.2.10 N-terminal sequencing. Sequencing was performed by Dr. Richard Cook at Baylor

College of Medicine.

5.2.1 1 Western blots. The transfer membranes were blocked with 5% Blotto (Bio-Rad),
PBS-0. 1% Tween 20 for 1 h at 37 °C or 16 h at 4 °C. Membranes were then incubated for 90
min at room temperature with the primary antibody diluted in 3% BSA, PBS-0. 1% Tween
20. After three washes with PBS-0. 1% Tween 20, membranes were incubated with alkaline
phosphatase-conjugated goat antirabbit IgG antiserum (Cappel, Durham, NC), washed five

times with PBS-0. 1% Tween 20, and developed with alkaline phosphatase substrate.

5.2.12 Protein Assay. SR protein determination was carried out by the methods of Lowry

et al. (195 1) using bovine serum albumin as a standard.

127

5 . 3 Results

5.3.1 Affinity labeling and purification of trypsin-treated Ca’+-release channel from
skeletal muscle heavy SR membranes. Previous reports have indicated that there are four
to five CaM binding sites per Caz*-release channel subunit in skeletal muscle SR (Yang et al.,
1994; Tripathy et al., 1995). This study was conducted to localize the CaM binding domains
in the channel protein. In initial studies, heavy SR membranes isolated from rabbit skeletal
muscle were partially proteolyzed for 5 min with trypsin at a proteinztrypsin ratio of 1000:1
(w/w) and affinity-labeled with [‘251]-Bz-CaM in the presence of 1 mM EGTA The
autoradiogram of SDS-PAGE indicated a significant decrease in the native Cay-channel
protein band with the appearance of five [‘ufl-Bz-CaM binding bands with apparent
molecular weights of 143 kDa, 117 kDa, 97.5 kDa, 73.2 kDa, 51.7 kDa (Fig. 5.1).
CHAPS solubilized tryptic Cay-channel complexes labeled with [‘”I]-Bz-CaM were
isolated and purified from 5-20% sucrose gradient centrifugation. This trypsin treatment
resulted in a small shift of the apparent sedimentation coefficient of the CHAPS solubilized
Cay-channel from 30 S to 28 S (Callaway et al., 1994). The fractions containing the channel

protein with bound m’I]-Bz-CaM were identified by gamma-counting (Fig. 5.2).

5.3.2 Identification of [mu-Bz-CaM bound tryptic Ca’*-channel polypeptides from the
purified 28 S complex. Two major CaM-binding polypeptides of the purified 28 S complex
labeled with [’2’I]-Bz-CaM have been identified from autoradiograms of SDS-PAGE (Fig.

5.3). The apparent molecular weights of the crosslinked products obtained in the presence of

128

Figure 5.1. Autoradiography of [‘2’I]-Bz-CaM bound trypsin-treated heavy SR membranes
from rabbit skeletal muscle. For 100 pL reaction, 0.] pM [‘2’I]-Bz-CaM was incubated with
100 pg trypsin-treated heavy SR in 0.3 M KC], 1 mM EGTA, and 50 mM Hepes bufi‘er, pH
7.0. The mixtures were separated by electrophoresis on a 5-18% gradient SDS-PAGE
(Laemmli, 1970) and the cross-linked products were identified by autoradiography of the gel.

Lane 1 and 2 represent trypsin-treated and control (no trypsin-treated) SR, respectively.

129

Rya-
1 43-1
1 17.-

se-

73-

 

52-
44-

22-

 

130

Figure 5.2. Sucrose gradient profile of 28 S complex of the CHAPS purified [’zsfl-Bz-CaM
labeled Ca2+-release channel. The 28 S complex purification process was described in
Experimental Procedures on a 16 ml linear 5-20% sucrose gradient containing 0.2% CHAPS,
300 mM KCl, 50 mM MOPS (pH 7.0). The gradient was sedimented for 18 h at 110,000 x
g in a SW 28 rotor, and 20 drop fiactions were collected from the bottom of the gradient.

Aliquots (50pl) from each fraction were counted by a gamma-counter.

 

CPM

131

Sucrose Gradient Profile of

 

 

 

 

Tryptic HSR with lZSI-Bz-CaM
1500
1200 t
900 t
600 '-
300 7
0 1 1 1 1
0 6 12 18 24 30

Fraction number

132

Figure 5.3. Identification of the 28 S complex of the [‘Z’I]-Bz-CaM bound Cd“ -release
channel. Aliquots (40 pL) of collected fractions from sucrose gradient were electrophoresed
on a 7% SDS-PAGE (Schagger and Von Jagow, 1987) described in Experimental
Procedures. Panel A and B, Coomassie-stained of gel. Panel C and D, autoradiography of
dried gels from panel A and B, respectively. Lane 1-14 in panel A were from sucrose gradient
fractions 1-14. Lane 15-26 in panel B were from fractions 15-26. Lane 27 was from the

resuspension of SR precipitate in sucrose gradient.

 

133

1234567091011121314

200
116

66
45
31

21
14

 

1234567891011121314

¢—149

c—75

4—17

 

134

Figure 5.3--cont’

 

135

15 16 17 16 16 2021 22 23 24 25 26 27

200
116

66
45
31

21
14

 

D ' .
15 16 17 16 19 2021 22 23 24 25 26 27

 

136

1 mM EGTA were 149 kDa and 75 kDa. If one CaM binds per fragment, these complexes
would correspond to CaM plus channel protein tryptic fiagments of approximately 133 kDa,
and 58 kDa Based on these apparent molecular weights, two tryptic fragment candidates are
suggested: the 135 kDa N-terrnina] fragment (residues 1-1508) and the 50 kDa fiagment
(residues 2401-2840) (Fig 5.4).

Fractions 5-9 containing the channel protein complex bound with [1251] -Bz-CaM were
pooled and subjected to SDS-PAGE (Fig. 5.3). The separated bands of protein from SDS-
PAGE were transferred to Immobilon membrane, stained with Coomassie blue and the blot
was subjected to autoradiography. Two [ml]-Bz-CaM binding bands appeared on the
autoradiogram of the membrane blot corresponding to the bands present in the autoradiogram
of the gel (Fig. 5.5). Sequencing attempts on the 149 kDa [lz’fl-Bz-CaM binding peptide
excised from the Immobilon membrane failed because the N-terminus of this peptide was
blocked. This result supports the possibility that this complex corresponds to the N-tennina]
135 kDa fragment plus CaM because sequencing results of Callaway et a]. (1994) indicated
that the 135 kDa fiagment isolated from the tryptic Cay-channel 28 S complex was blocked
at the NH2 terminus (Fig 5.4). Therefore, these results suggest that one CaM-binding peptide
is localized in the region between the N-terminus and residue 1508 in the amino acid sequence
of Ca2*-channel.

The second complex obtained fiom the 28 S purification had an apparent 114, of 75
kDa. Because of the proximity of this complex to the 76 kDa fragment, separation of these
bands was impossible with present techniques. Because the amount of 75 kDa fragment was

much less than the 76 kDa fragment in amino acid sequencing data, it is not possible to

137

 

 

 

 

 

 

 

4°00_l__ soon
4476
r
coon
n

Figure 5.4. Fragmentation map of 28 S complex of tryptic SR
Cafl'release channel from rabbit skeletal muscle (From Callaway

et al., 1994).

138

Figure 5.5. [”51]-Bz-CaM binding tryptic fragments within the ca -release channel were
recognized and sequenced. Fractions 5-9 from previous data were pooled and concentrated
by Centricon concentrators (Amicon, Inc, Beverly, MA) after dialyzing over night in 50 mM
MOPS, pH 7.0. The concentrated solution was subjected to 7% SDS-PAGE Schagger and
Von Jagow, 1987) and transferred to Immobilon-P-SQ membrane and prepared for
sequencing as described in Experimental Procedures. Lane 1 is Coomassie stain of transferred

membrane; lane 2 is autoradiogram of transferred membrane.

 

139

200
116
9

149

66 75

45

31
21
14

 

140

confirm the identity of this band fiom amino acid sequencing. However, based on the
molecular weight from the SDS gels and the likelihood that this tryptic Ca2*-channel fragment
binds 1 CaM, the site of CaM binding likely corresponds to the 50 kDa fiagment which is
positioned between amino acid 2401-2840 based on 28 S Ca2*-channel fragmentation map
(Fig. 5.4). Therefore, two tentative CaM-binding sites, amino acid residues 1-1508 and 2401-
2840, in the Ca2+-release channel have been positioned fi'om tryptic digests and photoaflinity

labeling.

5.3.3 Purification of calpain digested Ca”'-channel labeled with [mu-Bz-CaM. In order
to complement the tryptic digestion results, .we performed calpain digestion following [”51]-
Bz—CaM cross-linking with SR membranes. This work was performed in collaboration with
Dr. Hamilton and Dr. Yili-Wu at Baylor College of Medicine. Following aflinity labeling with
[‘”I]-Bz-CaM, the Caz*-channel/[‘2‘I]-Bz-CaM complex was labeled with [3H]ryanodine and
subjected to calpain digestion and the proteolytic fragments were purified on 5-20% sucrose
gradients. The sucrose gradient profile of purified proteolytic products is shown in Figure 5.6.
Aliquots from fiactions 6-10, 23 and 24 were subjected to 5% SDS-PAGE. The Coomassie
stained gel of the calpain digested channel protein and its autoradiogram are shown in Figure
5.7. Autoradiography of calpain-digested Ca2*-channel bound with” [ I]-Bz-CaM
demonstrated that there were four polypeptides which bound [‘2‘I]-Bz-CaM. The apparent
molecular weights of these complexes corresponded to 480 kDa, 365 kDa, 210 kDa, and 130
kDa (Fig. 5.8B). It has been reported that, based on the results of affinity labeling and

immunoblots, calpain-digested Ca2+-channel was initially degraded into an N-terminal 173

141

Figure 5.6. Sucrose gradient profile of calpain digested Cap-release channel labeled with
[‘Z’I]-Bz-CaM (From Yili Wu at- Baylor College of Medicine). The purification process of
calpain digested Ca2*-channel/[‘2’I]-Bz-CaM was described in Experimental Procedures.
Gradients were sedimented for 18 h at 110,000 x g in a SW 28 rotor, and 20 drop fractions
were collected from the bottom of the gradient. Aliquots (50pl) were counted for

radioactivity by scintillation counter.

 

CPM

142

The sucrose gradient profie (20%-5%)

 

12000

10000-

8000-

6000-

4000-

2000-

 

 

 

I I U l I U I I l I U I 6 I I

5 10 15
Fraction Number

20

I 1 l 6

25

 

143

Figure 5.7. SDS-PAGE and autoradiography of sucrose gradient purified calpain digested
Can-release channel labeled with [”5 I]-Bz-CaM (From Dr. Yili Wu at Baylor College of
Medicine). Panel A, Coomassie stain of 5% SDS-PAGE from collected fiactions 6-10, 23 and

24 in Figure 5.6. Panel B, autoradiography of dried gel fi'om panel A.

 

144

3'33»
i355 ‘
.4, .y.

11‘.
. 4'
n

 

(From Dr. Yili Wu at Baylor college of Medicine).

145

Figure 5.8. Calpain proteolytic fiagrnents of Ca2*-release channel identified by immunoblots
(From Dr. Yili Wu at Baylor College of Medicine). Panel A is the primary structure of Ca2+-
release channel. Site-specific antibodies against to various regions in Ca2+-channel are shown
by the arrows. Panel B is summary of calpain proteolytic Cay-channel fiagments bound to

[‘2’I]-Bz-CaM.

 

146

 

NL

 

l73kDa

fiC

1‘ i t l r l
1333- 2727- 4014- 4686-

1350 2743 Wadi-@5029-

4373 5037

 

 

3 (arokoa)
4-(13OkDa)

147

kDa fragment and a C-terminal 480 kDa fiagment which was firrther cleaved to a 365 kDa
fiagment at the N-terminus and a smaller fiagment at the C-terminus withrll, about 100 kDa
(as shown inFrg. 5.8 B) (Brandt et al., 1992; Gilchrist et al., 1992; Rardon et al., 1990). Our
results showed that there were no [‘”I]-Bz-CaM crosslinked products corresponding to the
N-terminal 173 kDa fi'agment nor the C-terminal 100 kDa fragment. These results suggest
that the CaM-binding sites are located between the C-terminal end of 173 kDa fiagment and

the N-terrninal end of 100 kDa fragment (see Fig. 5.8 B).

5.3.4 Identification of [mu-Bz-CaM bound calpain digested Ctr-channel fragments
by immunoblots. Site-specific antibodies against specific regions (as shown in Fig. 5.8 A)
of the Cay-channel were used as probes for immunoblotting to establish the identify of the
calpain digested fragments and crosslinked products. The immunoblots of [mu-Ez-
CaWCa2*-channel digested with calpain (data not shown) indicated that antibody (1333-
1350) recognized both the 210 kDa and 130 kDa bands and both bands were labeled with
['”I]-Bz-CaM (data not shown). However, antibody (2727-2743) only recognized the 210
kDa fragment and not the 130 kDa fiagment. These results imply that both the 210 kDa and
130 kDa fragments begin at the C-terminal end of 173 kDa fragment which means the 210
kDa fiagment is the degraded product of the 365 kDa fi'agment and the 210 kDa fiagment
which has been firrther cleaved to the 130 kDa fragment. Since the autoradiogram did not
show the remaining degraded product (M, = 80 kDa) from the 210 kDa fragment labeled
with [‘Z’I]-Bz-CaM, the CaM binding sites are thus suggested to localize in the 130 kDa

fragment which comprises amino acid residues 1333 to 2515 (as shown in Figure 5.8 B).

148

The calpain-digested 130 kDa fiagment is localized between residues 1333 and 2515
which comprises two [’”I]-Bz-CaM bound tryptic fragments, residues 1-1508 and residues
2400-2840 fi'om Figure 5.5. Therefore, in combining results of the trypsin and calpain digests,
CaM binding sites have been localized more precisely to amino acids 1333-1508 and 2400-

2515 in the skeletal muscle SR Cazfichanne] (Fig. 5.9).

149

 

 

 

M‘M" M1 M10
0 1000 2000 3000 4000 5000
as 1301 2m 2m 3119 «75
hyper, [7] no I 110 15013] 150 j 1‘ J
Fragment
MII?
Tsheshirnsetal..l989 I |
Zamad.,l990 H l
Mukseul, 1990 l I I l
Brandtetsl.,l992 I I l
Wail-1994 - I I
Churctal..l994 I I I. l l
Ymrgetal.,l996 I I

Figure 5.9 CaM-binding sites in skeletal muscle SR Ca2*-release channel. The linear sequence
of the Cay-channel is indicated by a horizontal line. The NH3+ and COO' termini are marked.
M’, M”, and M1-M10 refer to predicted transmembrane sequences (Zorzato et al., 1990).
The tryptic fragments are represented by open boxes with molecular masses (x1000) inside

the boxes. The candidates of CaM-binding sites are positioned by verdical lines.

150

5.4 Discussion

Limited proteolysis has previously provided insights into structure-function
relationship within the channel protein (Chu et al., 1988; Shoshan-Barrnatz and Zarka, 1988;
Meissner et al., 1989; Rardon et al., 1990; Chen et al., 1993; Callaway et al., 1994). Seven
mm'or tryptic sensitive regions within Caf-release channel have been identified and positioned
by site-specific antrbodies (Chen et al., 1992) and amino acid sequencing (Marks et al., 1990;
Callaway et al., 1994) as shown in Figure 5.4.

Calpain also catalyzes specific and limited cleavage of substrate including enzymes,
myofibrillar proteins, membrane proteins, cytoskeletal proteins, and receptor proteins (Dayton
et al., 1976; Mellgran, 1987; Puca et al., 1977; Vedeckis et al., 1980). It has been suggested
that CaM-binding proteins are good substrates for calpain which recognizes the PEST
(proline, glutamic acid, serine, threonine-rich) sequence for binding sites in substrate proteins
(Wang et al., 1989). The Ca2*-release channel protein has eight PEST sequences (Brandt et
al., 1992). One of the calpain cleavage sites in Ca”’-release channel has been suggested at
residues 1383-1400 which was predicted to be near a CaM-binding site (Brandt et al., 1992;
Shoshan-Barrnatz et al., 1994). In this work, we used combined photoamnity labeling and
immunoblotting approaches with partial trypsin and calpain proteolysis of Ca2+ release channel
in the SR membrane to localize and identify CaM binding sites in the Ca” channel.

Wheat germ CaM was iodinated at Cys-27 with the UV sensitive crosslinker
benzophenone-4-maleimide which can react with any methylene groups of amino acid residues

in the proximity of the label. Thus, polypeptide residues which are within 1.0 nm of the

151

labeled Cys-27 are readily cross-linked. Therefore, it is likely that crosslinked polypeptides
either constitute the CaM-binding sites, or are in very close proxirrrity to CaM.

When crosslinking in the presence of 1 mM EGTA, at least 5 tryptic fiagrnents fiom
SR membrane bound [‘“H-Bz-CaM (Fig 5.1). These results are consistent with previous Rh-
CaM/Caf-channel binding stoichiometry data which suggest there are 4-5 CaM binding sites
per channel subunit (Yang et al., 1994; Tripathy et al. 1995). Two of these CaM-binding
fragments in the tryptic Ca2"-channel have been firrther purified from SR membrane by
sucrose gradient centrifirgation and identified by autoradiography of SDS-PAGE. These two
purified fragments had apparent molecular weights of 149 kDa and 117 kDa which
corresponded to the 117 kDa and 73 kDa fragments, respectively in Figure 5.1. The
difference of the apparent molecular weights between these two results were due to different
SDS-PAGE systems which caused the difl‘erence of polypeptides mobility shift in the gels.

There were three ['”I]-Bz-CaM labeled fiagments missing from the sucrose gradient
purified 28 8 complex (Fig. 5.3) when compared to the results before purification (Fig 5.1).
The loss of [‘“fl-Bz-CaM bound 51.7 kDa fiagnmt in the autoradiogram of sucrose gradient
purified 28 8 complex fractions might result fi'om dissociation of this fiagment from 28 S
complex during purification process and diffusion into gradient solution. The other two
missing bands might be due to insufficient radioactive intensity of the [’z’fl-Bz-CaM in the
autoradiogram. After subtracting CaM’s molecular weight, the 149 kDa and 117 kDa
fragments labeled with [lzsfl-Bz-CaM are suggested to correspond to the 135 kDa and 50
kDa fiagments in the Ca2"-channel fiagrnentation map (Fig. 5.4) which are localized between

residues 1-1508 and 2400-2840.

152

To complement the trypsin digestion results, we further used a more selective
protease, calpain, that degrades primarily the Ca”-channel protein in the heavy SR
Photoaflinity labeling with [‘”I]-Bz-CaM and immunoblotting with site-specific antibodies
were used to identify calpain digested Ca’*-channel fiagments bound to CaM The
autoradiogram demonstrated that there were 4 calpain-digested Ca2*-channel fiagments
labeled with [‘“fl-Bz-CaM after sucrose gradient purification.

It has been reported that calpain digested the Cay-channel in an ordered sequence of
susceptible sites; that is, the first site must be cleaved before the second site becomes exposed
(Brandt et al., 1992; Gilchrist et al., 1992; Shoshan-Barmatz et al., 1994). Our results were
in agreement with these reports. Calpain initially cleaved the intact Cay-channel monomer
into peptides of 173 (N-terminal end) and 480 kDa (C-terminal end) which was subsequently
cleaved into 100 and 365 kDa fiagments. The 365 kDa peptide was cleaved to 210 kDa and
then firrther, into a 130 kDa fiagment (Figure 5A). Therefore, our calpain digested fiagments
labeled with [”"H-Bz-CaM can be traced from 480 kDa to 365 kDa and then to 210 kDa and
finally to the 130 kDa fiagment. Our immunoblots of calpain-digested Cay-channel suggested
the N-terminus of this 130 kDa fragment which labeled with [‘2’I]-Bz-CaM spanned amino
acid residues 1333-1350 and the end of C-terminus was at residue 2515. This result is in
agreement with one of the calpain cleavage sites in the Ca2*‘release channel, residues 1383-
1400, and was predicted to be near a CaM-binding site (Brandt et al., 1992; Shoshan-Barmatz
et al., 1994). In combining calpain-digested results with the results form trypsin digestion, two
CaM-binding domains were localized to residues 1333-1508 and 2400-2515.

Menegazzi et a]. (1994) and Chen et al. (1994) have tentatively identified several Ca”-

153

dependent CaM-binding sites in rabbit skeletal muscle Ca’*-channel protein with difl‘erent
experimental approaches. The CaM-binding sites in SR Ca2*-channel in our results are
difl‘erent fiom Menegazzi et al. (1994) and Chen et al. (1994), and possible reasons for these
discrepancies can be explained as follows. Both groups identified CaM binding regions in
Cay-channel protein through the use of expressed fusion proteins using cDNAs encoding
fragments of the rabbit skeletal muscle Ca2*-release channel protein and the use of gel overlay
procedures. Since difi‘erent degrees of renaturation of these fusion proteins on nitrocellulose
membranes could affect CaM binding activity in the overlay assay, CaM binding sites
determined by the overlay method may not represent true CaM binding sites in Ca2*-channel.
Furthermore, the overlay method may not have detected all types of CaM-binding sites. Thus,
it is possible that other Ca2"-dependent or Cay-independent CaM-binding sites may exist in
the native Ca2*-channel. Finally, fusion proteins only cover 90% of the length of the C? -
channel amino acid sequence (Chen et al., 1994) and CaM-binding sites located in or near the
regions that are not covered by fusion proteins would not have been detected in the overlay
study.

This study is the first to identify CaM-binding sites in the native channel protein
structure. Efforts are continuing to narrow down the CaM-binding sites in the native structure

and to correlate structure with inhibitory or activating effects on Ca21-channel activity.

CHAPTER6

OVERALL CONCLUSIONS AND FUTURE RESEARCH

1. Overall conclusions

1. There are multiple CaM-binding sites on each channel protein subunit and the
afinities of these CaM-binding sites depend on the concentration of Ca2+ and Mg”.
The binding of CaM to the SR Cay-channel is regulated by modulators of the Ca”-
channel activity itself, and this novel regulation is likely to be important in the

mechanism of excitation-contraction.

2. The equilibrium binding constants and stoichiometry of CaM to MHS SR Ca”-
release channel are altered compared to normal skeletal muscle. Our data are
consistent with our hypothesis that the altered binding equilibrium of CaM to MHS
Cay-channel results, in part, abnormal regulation of SR Ca2+ release activity.
However, the effects of ionic strength and caffeine on the binding of CaM to both

MHS and normal SR are not significantly different.

154

155

3. Two CaM-binding sites, amino acid residues 1333-1508 and 2400-2515, in the
central regions of Cay-release channel protein fiom rabbit skeletal muscle have been

identified.

H. Future research

1. Characterization of the physiological role of CaM binding to MHS skeletal SR
Caz*-release channel by different experimental approaches such as single channel

recording, ryanodine binding study and [“Ca"] efllux study.

2. Identification of CaM-binding sites in SR Ca2"-release channel fi'om MHS skeletal
muscle. Loss of specific CaM-binding sites may be associated with altered functional

activity.

3. Further examination of the physiological function of these two identified CaM-
binding sites on Ca2*-channel gated by binding to various channel modulators such

as ryanodine, cafl‘eine, Ca2+ and Mg2+ etc.

CHAPTER 7

BIBLIOGRAPHY

Andreasen, T.J., Luetje, C.W., Heidemarr, W., and Storm, DR. 1983. Purification of a novel
calmodulin binding protein from bovine cerebral cortex membranes. Biochemistry 2224615-

4618.

Babu, Y.S., Sack, J.S., Greenhough, T.J., Bugg, C.E., Means, AR and Cook, WJ. 1985.

Three-dimensional structure of calmodulin. Nature 315237-40.

Barbato, G., Ikura, M., Kay, L.E., Pastor, R.W., Bax, A. 1992. Backbone dynamics of
calmodulin studies by 15N relaxation using inverse detected two-dirnensiona] NMR

spectroscopy: the central helix is flexible. Biochemistry 31:5269-5278.

Bazari, W.L. and Clarke, M. 1981. Characterization of a novel calmoduhn from

Dictyostelium discoideum. J. Biol. Chem. 256:3 598-3603.

156

157

Bendall, IR 1952. Efl‘ect of the 'Marsh factor' on the shortening of muscle fibre models in

the presence of adenosine triphosphate. Nature 170: 105 8-1060.

Bendall, JR 1953. Further observations on a factor (the 'Marsh' factor) efl‘ecting relaxation
of ATP-shortened muscle-fibre models, and the effect of Ca and Mg ions upon it. J. Physiol.

1211232-254.

Beringer, T. 1976. A freeze-fracture study of sarcoplasmic reticulum from fast and slow

muscle of the mouse. Anat. Rec. 184:647-663.
Berridge, M.J. 1993. Inositol his-phosphate and calcium signaling. Nature 361:315-325.

Billingsley, M. L., Polli, J.W., Pennupacker, K.R. and Kincaid, RL. 1990. Identification of

. calmodulin-binding proteins. Methods Enzymol. 1842451-467.

Block BA, Imagawa, T., Campbell, K.P. and Franzini-Annstrong, C. 1988. Structural
evidence for direct interaction between the molecular components of the transverse

tubule/sarcoplasmic reticulum junction in skeletal muscle. J. Cell Biol. 10722587-2600.

Brandt, N.R., Caswell, A.H., Brandt, T. and Mellgren, R 1992. Mapping of the calpain

proteolysis products of the junctional foot protein of the skeletal muscle triad junction. J.

158

Membrane Biol. 127:35-47.

Britt, BA and Kalow, W. 1970. Maligth hypertherrnia: a statistical review. Can. Anaesth.

Soc. J. 17:293-315.
Britt, B.A (Editor) 1987. Malignant hyperthermia. Boston, MA Nrjhoff.

Buratti, R, Prestipino, G., Menegazzi, P., Treves, S. and Zorzato, F. 1995. Calcium
dependent activation of skeletal muscle Ca2+ release channel (ryanodine receptor) by

calmodulin. Biochem. Biophys. Res. Commun. 213:1082-1090.

Callaway, C., Serysheva, A., Wang, J .P., Slavik, K.J., Needleman, D.H., Cantu III, C., Wu,
Y., Jayaraman, T., Marks, AR, and Hamilton, S.L. 1994. Localization of the high and low
aflinity [’H]ryanodine binding sites on the skeletal muscle Ca2+ release channel. J. Biol. Chem.

269:. 15876-15884.

Campbell, K. P., Franzini-Arrnstrong C. and Sharnoo, AE. 1980. Further characterization of

light and heavy sarcoplasmic reticulum vesicles. Biochem. Biophys. Acta 602:97-116.

Campbell, K.P. and MacLennan, DH. 1981. Purification and characterization of the 53,000-

dalton glycoprotein from the sarcoplasmic reticulum. J. Biol. Chem. 256: 4626-4632.

159
Campbell KP. and MacLennan, D11 1982. A calmodulin-dependent protein kinase system

from skeletal muscle sarcoplasmic reticulum. J. Biol. Chem. 257: 1238-1246.

Campbell, KP, MacLennan, DH, Jorgensen, A0. and Mintzer, MC. 1983. Purification and
characterization of calsequestrin from canine cardiac sarcoplasmic reticulum and identification

of the 53,000 dalton glycoprotein. J. Biol. Chem. 258:1197-1204.

Campbell, KP, Knudson, C. M., Imagawa, T., Leung, AT., Sutko, J.L., Kah], S.D., Raab,
CR. and Madson, L. 1987. Identification and characterization of the high afinity
[H]ryanodine receptor of the junctional sarcoplasmic reticulum Ca2+ release channel. J. Biol.

Chem. 262:6460-6463.

Carrier, L., Vaillaz, M., & Dupont, Y. 1991. Abnormal rapid Ca2+ release fi'om sarcoplasmic

reticulum of malignant hyperthermia susceptible pigs. Biochirn. Biophys. Acta 1064: 175-193.

Cassens, R.G., Marple, ON, and Eikelenboom, G. 1980. Animal physiology and meat

quality. Adv. Food Sci. 24271-155.

Caswell, A.H., Brandt, N.R, Brunschwig, JP. and Purkerson, S. 1991. Localization and
partial characterization of the oligomeric disulfide-linked molecular weight 95,000 protein
(triadin) which binds the ryanodine and dihydropyridine receptors in skeletal muscle triadic

vesicles. Biochemistry. 30:7507-7513.

160

Chandler, W.K, Rakowaky, RF. and Schneider, MF. 1976. Effects of glycerol treatment and
maintained depolarization on charge movement in skeletal muscle. J. Physiol. London

254:285-316.

Chen, S.R.W., Zhang, L., MacLennan, DH. 1992. Characterization of a Ca” binding and
regulatory site in the Ca2+ release channel (ryanodine receptor) of rabbit skeletal muscle

sarcoplasrrric reticulum. J. Biol. Chem. 267:23318-23326.

Chen, S.W.R., Airey, J.A., and MacLennan, DH. 1993. Positioning of major tryptic
fiagments in the Ca2+ release channel (ryanodine receptor) resulting from partial digestion of

rabbit skeletal muscle sarcOplasmic reticulum. J. Biol. Chem. 268:22642-22649.

Chen, S.R.W., and MacLennan, DH. 1994. Identification of calmodulin-, Ca2*-, and
ruthenium red-binding domains in the Ca2+ release channel (ryanodine receptor) of rabbit

skeletal muscle sarcoplasmic reticulum. J. Biol. Chem. 269:22698-22704.

Cheung, W.Y. 1970. Cyclic 3’, 5'-nucleotide phosphodiesterase. Demonstration of an

activator. Biochem. Biophys. Res.Commun. 38:533-538.

Cheung, W.Y. 1980. Calmodulin plays a pivotal role in cellular regulation. Science 207: 19-

27.

161

Chiesi, M. and Carafoli, E. 1982. The regulation of Ca2*-transport by fast skeletal muscle

sarcoplasmic reticulum. J. Biol. Chem 257:984-991.

Chu, A, Sumbilla, C., Scales, D., Piazza, A and Inesi, G. 1988. Trypsin digestion of

junctional sarcoplasmic reticulum vesicles. Biochemistry 27 :2827-2833.

Chu, A, Diaz-Munoz, M., Hawkes, M. J ., Brush, K, Hamilton, S.L. 1990. Ryanodine as a
probe for the functional state of the skeletal muscle sarcoplasmic reticulum calcium release

channel. Mol. Pharmacol. 37:735-741.

Cimler, B.M., Andreasen, T.J., Andreasen, 11.1., and Storm, on 1985. P-57 is a neural

specific calmodulin-binding protein. J. Biol. Chem. 260: 10784-10788.

Conti, A, Tarroni, P., Rossi, D. and Sorrentino, V. 1995. Ryanodine receptorszisoforms

expression in mammals and chicken (Abstract). Biophys. J. 68:A52.

Corrnier, M.J., Charbonneau, H. and Jarrett, H.W. 1981. Plant and firngal calmodulinzCa2+-

dependent regulation of plant NAD kinase. Cell Calcium 22313-333.

Dayton, W.R, Reville, W.J., Goll, DE. and Stromer, M.A 1976. A Ca 2+-activated protease
possibly involved in myofibrillar protein turnover. Partial characterization of the purified

enzyme. Biochemistry 15:2159-2167.

162

de Meis, L. and Vianna, AL. 1979 Energy interconversion by the Ca2*-dependent ATPase

of the SR Ann Rev. Biochern 48:275.

Dulhunty, A and Valois, A 1982. A freeze-fracture study of normal and dystrophic C57BL

mouse muscle. Muscle Nerve. 5 :425-433.

Dulhunty, A and Valois, A 1983. Indentations in the terrnina] cisternae of amphibians and

mammalian skeletal muscles. J. Ultrastruct. Res. 84:34-49.

Dulhunty, A, Gage, P. and Valois, A. 1983. Indentations in the terminal cisternae of slow-

and fast-twitch muscles fiom normal and paraplegic rats. J. Ultrastruct. Res. 84:50-59.

Ebashi, S. and Endo, M. 1968. Calcium ion an dmuscle contraction. Prog. Biophys. Mol.

Biol. 18:123-183.

Ebashi, S., Kodarna, A, and Ebashi, F. 1968. Troponin, I. Preparation and physiologic

firnction. J. Biochem. 64:465-477.

Eibschutz, B., Wong, A.P., Lopaschuk, G.D., Katz, S. 1984, The presence and binding
characteristics of calmodulin in microsomal preparations enriched in sarcoplasmic reticulum

from rabbit skeletal muscle. Cell Calcium 52391-400.

163

Endo, M., Tanaka, M., Ogawa, Y. 1970. Calcium-induced release of calcium from the

sarcoplasmic reticulum of skinned muscle fibers. Nature 228:34-36.

Endo, M 1977. Calcium release from the sarwplasmic reticulum. Physiol. Rev. 57271-108.

Endo, M. 1985. Calcium release from sarcoplasmic reticulum. Curr. Top. Membr. Transp.

25:181-230.

Entrnan, M.L., Bick, R, Chu, A, Van-Winkle, W.B. and Tate, CA. 1986. The mechanism
of nucleotide induced calcium translocation across sarcoplasmic reticulum membranes:
Evidence for a non-translocated intermediate pool of calcium. J. Mol. Cell. Cardiol. 18:781-

798.

Fernandez, J.L., Rosemblatt, M., Hidalgo, C. 1980. Highly purified sarcoplasmic reticulum
vesicles are devoid of Ca 2+-independent (‘basal’) ATPase activity. Biochim. Biophys. Acta

599:552—568.

Fill, M, Coronado, R, Mickelson, JR, Vilven, J ., Ma, J .J ., et al. 1990. Abnormal ryanodine

receptor channels in malignant hyperthermia. Biophys. J. 57 :47 1-47 5.

Finn, B.E., Drakenberg, E. and Forsen, S. 1993. The structure of apo-calmodulin. A 1H

NMR examination of the carboxy-terminal domain. FEBS Lett. 336:368-3 74.

164

Finn, B.E., Evenas, J., Drakenberg, T., Waltho, J.P., Thulin, E. and Forsen, S. 1995.
Calcium-induced structural changes and domain autonomy in calmodulin. Nature Structural

Biology 2:777-783.

Fleischer, S. and Inui, M. 1989. Biochemistry and biophysics of excitation-contraction

coupling. Annu. Rev. Biophys. Biophys. Chem. 18:333-364.

Ford, LE. and Podolsky, RJ. 1970. Regenerative calcium release within muscle cells. Science

167:58-59.

Franzini-Arrnstrong, C. 1970. Purified ryanodine receptor form skeletal muscle is the Ca-

permeable pore of the calcium release channel. J. Cell Biol. 47:488-499.

F ranzini-Armstrong, C. 1980. Structure of sarcoplasmic reticulum. FASEB 39:2403-2409.

Franzini-Amrstrong, C. and Jorgensen, A0. 1994. Structure and development of E-C

coupling units in skeletal muscle Annu. Rev. Physiol. 56:509-534.

Fuentes, 0., Valdivia, C., Vaughan, D., Coronado, R and Valdivia, RH. 1994. Calcium-
dependent block of ryanodine receptor channel of swine skeletal muscle by direct binding of

calmodulin. Cell Calcium 15:305-316.

165
Fujii, J., Otsu. K, Zorzato, F., Deleon, S.,Khanna, V.K., Weiler, JE, O'Brien, RJ. and
MacLennan, DH 1991. Identification of a mutation in porcine ryanodine receptor associated

with malignant hyperthermia. Science 253 :448-451.

F ujita, K. 1954. Action of adenosine derivatives on muscle activity. 11. Glycerol treated

muscle and “relaxing factor'. Folia Pharmaco]. Japon 50:183-192.

Gilchrist, ISO 1992. Calcium-activated neutral protease effects upon skeletal muscle

sarcoplasmic reticulum protein and calcium release. J. Biol. Chem. 267 :20857-20865.

Goodman, M., Pechere, J .G., Haiech, J., and Demaille, J .G. 1979. Evolutionary
diversification of structure and firnction in the family of intracellular calcium-binding proteins.

J. Mol. Evol. 13:331-352.

Gronert, GA 1986. Malignant Hypertherrnia. In Myology (Engel, AG., and Banker, B.Q.,

eds.) pp. 1763-1784. McGraw-Hill Publications, Minneapolis, MN.

Hain, J ., Schindler, H, Nath, S. and Fleischer, S. 1993. Phosphorylation of the skeletal

muscle calcium release channel removes block by magnesium ions. Biophys. J. 64: A151.

Hakamata, Y., Nakai, J. Takeshima, H Irnoto, K. 1992. Primary structure and distribution

of a novel ryanodine receptor/calcium release channel from rabbit brain. FEBS Lett. 312:229-

166

235.

Hamilton, S.L. and Tate, CH. (1991) in Cellular Calcium (McCorrnack, J.G. and Cobbold,

P.H., eds) pp.313-343, IRL Press, Oxford.

Harrison, G.G. Malignant hyperthermia. Dantrolene-dynarnics and kinetics. 1988. Br. J.

Anaesth. 60:279-286.

Hasselbach, W. and Migala, A 1987. Activation and inhibition of the calcium gate of

sarcoplasmic reticulum by high-amnity ryanodine binding. FEBS Lett. 221: 1 19-123.
Hawkes, MJ., Nelson, T.E., & Hamilton S.L. 1992. [3H]Ryanodine as a Probe of Changes
in the Functional State of the Cay-release Channel in Malignant Hyperthermia. J. Biol. Chem.

267:6702-6709.

Hefl‘ron, J.J.A, Ellis, F .R 1985. Clinical and biochemical aspects of malignant hyperthermia.

Recent Adv. Anaesth Analg. 15:173-207.
Huxley, A. F. 1957. Muscle structure and theories of contraction. Prog. Biophys. 7:255.

Huxley, AF. and Taylor, RE. 1958. Local activation of striated muscle fibers. J. Physiol.

1442426.

167
Hymel, L., Inui, M., Fleischer, S. and Schindler, H. 1988. Purified ryanodine receptor of
skeletal muscle sarcoplasmic reticulum forms Ca21-activated oligomeric C21+ channels in

planar bilayers. Proc. Natl Acad. Sci. 85:441-445.

Ikemoto, N. 1982. Structure and function of the calcium pump protein of sarcoplasmic

reticulum. Annu. Rev. Physiol. 44:297-317.

Ikemoto, N., Antoniu, B. and Meszaros, LG. 1985. Rapid flow chemical quench studies of

calcium release from isolated sarcoplasmic reticulum. J. Biol. Chem. 260: 14096-14100.

Ikemoto, T., Iino, M. and Endo, M. 1995. Enhancing efl‘ect of calmoduhn on Ca2*-induced
Ca2+ release in the sarcoplasmic reticulum of rabbit skeletal muscle fibres. J. Physiol. 4871573-

582.

Ikura M., Clore, G.M, Gronenbom, AM., Zhu, G., Klee, CB. and Bax, A 1992. Solution
structure of a calmodulin target peptide complex by multidimensional NMR Science

256:632-638.

Imagawa, T., Smith, J.S., Coronado, R and Campbell, KP. 1987. Purified ryanodine receptor
from skeletal muscle sarcoplasmic reticulum is the Ca2+ -permeable pore of the calcium

release channel. J Biol. Chem. 262: 16636-16643.

168

Inesi, G. A. 1985. Mechanism of calcium transport. Ann. Rev. Physiol. 472573 -601.

Inui, M, Saito, A and Fleischer, S. 1987a. Purification of the ryanodine receptor and identity
with foot structure of junctional terminal cisternae of sarcoplasmic reticulum fiom fast

muscle. J. Biol. Chem. 26211740-1747.

Inui, M., Saito, A and Fleischer, S. 1987b. Isolation of the ryanodine receptor from cardiac

sarcoplasmic reticulum and identity with the feet structures. J. Biol. Chem. 262: 1563 7-15642.

James, P., Vorherr, T. and Carafoli, E. 1995. Calmodulin-binding domains: just two faced or

multi-faceted? TIBS 20:3 8-42.

Jenden, DJ. and Fairhurst, AS. 1969. The pharmacology of ryanodine. Pharmacol. Rev.

21:1-25.

Jones, DA and Round, J.M. 1990. Skeletal Muscle in Health an dEisease. pp 49. Manchester

University Press. Manchester, England.

Jones, L.R., Besch, HR, Jr, Sutko, J .L. and Willerson, J.T. 1979. Ryanodine-induced
stimulation of net Ca2+ uptake by cardiac sarcoplasmic reticulum vesicles. J .Pharmacol. Exp.

Ther. 209248-55.

169

Judge, MD, Aberle, E.D., Forrest, J .C., Hedrick, HE. and Merkel, RA 1989. Principle of

Meat Science. pp 23. Kendall/Hunt Publishing Company. Dubuque, Iowa.

Kawamoto, RM., Brunschwig, J.P., Kim, KC, and Caswell AH. 1986. Isolation,
characterization and localization of the spanning protein from skeletal muscle triads. J. Cell

Biol. 10321405-1414.

Kawamoto, RM, Bnmschwig, J .P., and Caswell AH 1988. Localization by immunoelectron
microscopy of spanning protein of triad junction in terminal cisternae/triad vesicles. J .Muscle

Res. Cell Motil. 92334-343.

Kim, DH, Sreter, F.A, Ohnishi, S.T., Ryan, J.F., Roberts, J. Allen, P. D., Meszaros, L.G.,

Antonin, B. and Ikemoto, N. 1984. Biochem. Biophys. Acta 775:320—327.

Kim, K.C., Caswell, AH., Talvenheimo, J .A., Brandt, NR 1990. Isolation of a terminal
cistema protein which may link the dihydropyridine receptor to the junctional foot protein in

skeletal muscle. Biochemistry. 29:9281-9289.

Kirchberger, MA, Tada, M., Katz, AM. 1974. Adenosine 3',S'-monophosphate—dependent
protein kinase-catalyzed phosphorylation reaction and its relationship to calcium transport in

cardiac sarcoplasmic reticulum. J. Biol. Chem. 249:6166-6173.

170

Kirino, Y., Osakabe, M. and Shimizu, H. 1983. Ca2*-induced Ca2+ release fi'om fiagmented

sarwplasmic reticulum : Ca2*-dependent passive Ca2+ efllux. J. Biochem. 94:1 1 1 1-1 118.

Klaevit, RE., Vanaman, TC. 1984. Aa'dotyrosyl calmodulin derivatives. Specific probes for

protein-binding domains. J. Biol. Chem. 259:15414-15424.

Klee, CB. and Vanaman, T. C. 1982. Calmodulin. Adv. Protein Chem. 35:213-321.

Kretsinger, RH. 1980. Structure and evolution of calcium-modulated proteins. CRC Crit.

Rev. Biochem. 8:1 19-174.

Kuboniwa, H., Tjandra, N., Grzesiek, S., Ren, H., Klee, CB. and Bax, A 1995. Solution

structure of calcium-flee calmodulin. Nature Structure Biology 2:768-776.

Laemmli, UK. 1970. Cleavage of structural proteins during the assembly of the head of

bacteriophage T4. Nature 227:680-685.

Lai, F .A, Erickson, H.P., Rousseau, B, Lin, Q.Y. and Meissner, G. 1988. Purification and

reconstitution of the calcium release channel from skeletal muscle. Nature 331:315-319.

Lai, F .A, Misra, M, Xu, Le, Smith, HA, and Meissner, G. 1989. The ryanodine receptor-Ca

super (2+) release channel complex of skeletal muscle sarcoplasmic reticulum: Evidence for

171

a cooperatively coupled, negatively charged homotetramer. J. BioL Chem. 264: 16776-16785.

Lakowicz, J .R. 1983. Principles of fluorescence spectroscopy. Plenum Press.

Lattanzio, G., Schlatterer, R, Nicar, M., Campbell, K., and Sutko, J. 1987 . The effects of
ryanodine on passive calcium fluxes across sarcoplasmic reticulum membranes. J. Biol. Chem.

262:2711-2718.

Lawrie, RA, Cahterum, DP. and Hale, HP. 1958. Abnormal low pH in pig muscle. Nature

182:807.

LeGendre, M. and Matsudaira, P. 1989. in A Practical Guide to Protein and Peptide
Purification for Microsequencing (Matsudaira, P.T., ed) pp 49-69, Academic Press, San

Diego, CA.

Louis, C.F., Zualkeman, K., Roghair, T. and Mickelson, J .R 1992. The effects of volatile
anesthetics on calcium regulation by malignant hypertherrnia-susceptible sarcoplasmic

reticulum. Anesthesiology 77:114-125.

Lowry, 0.1-I, Rosebrough, M.J., Farr, AL., & Randall, RJ. 1951. J. Biol. Chem. 193:265-

275.

1'72

MacLennan, DH and Wong, RT. 1971. Isolation of a calcium-sequestering protein from

sarcoplasmic reticulum. Proc. Natl. Acad. Sci. USA 68: 123 1-1235.

MacLennan, DH., Seeman, P., Iles, 3H. and Yip, C.C. Membrane formation by the

adenosine triphosphatase of sarcoplasmic reticulum. J. Biol. Chem. 246:2702-2710.

MacLennan, DH. and Holland, RC. 1975. Calcium transport in sarcoplasmic reticulum.

Annu. Rev. Biophys. Bioeng. 4: 377-404.

MacLennan, D.H., Campbell, K.P., and Reithmeier, RAF. 1983 Calsequestrin, in Calcium

and Cell Function, Bol. 4. Cheung, W., Ed,. Academic Press, New York, pp. 151.

Marks, AR, Fleischer, S. and Ternpst, P. 1990. Surface t0pography analysis of the ryanodine
receptor/Junctional channel complex based on proteolysis sensitivity mapping. J. Biol. Chem.
265:13143-13149.

Marsh, BB. 1951. A factor modifying muscle fibre syneresis. Nature 167: 1065-1066.

Marsh, BB. 1952. The effects of adenosine triphosphate on the fibre volume of a muscle

homogenate. Biochim. Biophys. Acta 92247-260.

173

Martonosi, AN. 1984. Mechanisms of Ca2+ release from sarcoplasmic reticulum of skeletal

muscle. Physiol. Rev. 64: 1240-1320.

Marty, 1., Robert, M., Villaz, M., De Jongh, K.S., Lai, Y. Catterall, W.A and Ronjat, M.
1994. Biochemical evidence for a complex involving dihydropyridine receptor and ryanodine

receptor in triad junctions of skeletal muscle. Proc. Natl. Acad. Sci. USA 91:2270-2274.

Masure, HR, Alexander, K.A, Wakim, ET, and Storm, DR 1986. Physicochernical and
hydrodynamic characterization of p-5 7, a neurospecific calmodulin binding protein.

Biochemistry 25:7553-7560.

Maurer, A, Tanaka, M, Ozawa, T., Fleischer, S. 1985. Purification and crystallization of the
calcium binding protein of sarcoplasmic reticulum fiom skeletal muscle. Proc. Natl. Acad. Sci.

USA 82:4036-4040.

McGrew, S.G., Wolleben, C., Siegl, P., Inui, M., and Fleischer, S. 1989. Positive
cooperativity of ryanodine binding to the calcium release channel of sarcoplasmic reticulum

fi‘om heart and skeletal muscle. Biochemistry 28: 1686-1691.

McPherson, RS. and Campbell, KR 1993. The ryanodine receptor/Ca2+ release channel. J.

Biol. Chem. 268:13765-13768.

17 4
Meador, W.E., Means, AR and Quiocho, F .A 1992. Target enzyme recognition by

calmodulin:2.4 A structure of a calmodulin-peptide complex. Science 257: 1251-1255.

Meador, W.B., Means, AR and Quiocho, FA 1993. Modulation of calmodulin plasticity

in molecular recognition on the basis of crystal structures. Science 262:1718-1721.

Means, AR, VanBerkum, M.F.A, Indrani, B., Lu, K. P. and Rasmussen, CD. 1991.

Regulatory functions of calmodulin. Phannac, Ther. 50:255-270

Meissner, G. and Fleischer, S. 1971. Characterization of sarcoplasmic reticulum from skeletal

muscle. Biochim. Biophys. Acta 241:3 56-378.

Meissner, G., Conner, GE. and Fleischer, S. 1973. Isolation of sarcoplasmic reticulum by
zonal centrifugation and purification of Ca2+ pump and C21 -binding protein. Biochim.

Biophys. Acta 298:246-269.

Meissner G. 1984. Adenine nucleotide stimulation of Cay-induced Ca 2* release in

sarcoplasmic reticulum J. Biol. Chem. 159: 1365-1374.

Meissner, G. 1986. Evidence of a role for calmodulin in the regulation of calcium release from

skeletal muscle sarcoplasmic reticulum. Biochemistry 25:244-251.

175

Meissner, G. and Henderson, J .S. 1987. Rapid calcium release fi'om cardiac sarcoplasmic
reticulum vesicles is dependent on Ca2+ and is modulated by Mg", adenine nucleotide, and

calmodulin. J. Biol. Chem. 262:3065-3073.

Meissner, G., Rousseau, E. and Lai, F.A 1989. Structural and functional correlation of the
trypsin-digested Ca” release channel of skeletal muscle of sarcoplasmic reticulum. J. Biol.

Chem. 264:1715-1722.

Meissner, G. and El-Hashem, A 1992. Ryanodine as a functional probe of the skeletal

muscle sarcoplasmic reticulum Ca2+ release channel. Mol. Cell. Biochem. 114: 1 19-123.

Meissner, G. 1994. Ryanodine receptor/Ca2+ release channels and their regulation by

endogenous efl‘ectors. Annu. Rev. Physiol. 56:485-508.

Mellgren, RL. 1987. Calcium-dependent proteases: an enzyme system active at cellular

membranes? FASEB J. 12110-115.

Menegazzi, P., Larini, F., Treves, S., Guerrini, R, Qusdroni, M., and Zorzato, F. 1994.
Identification and characterization of three calmodulin binding sites of the skeletal muscle

ryanodine receptor. Biochemistry 33 :9078-9084.

176

Mickelson, J.R, Ross, J.A, Rwd, B.K, and Louis, CF. 1986. Enhanced Ca2+ induced
calcium release by isolated sarcoplasmic reticulum vesicles fiom malignant hyperthennia

susceptible pig muscle. Biochim. Biophys. Acta 862:318-328.

Mickelson JR, Gallant, EM, Litterer, L.A, Johnson, KM., Rempel, W.B., and Louis, CF.
1988. Abnormal sarcoplasmic reticulum ryanodine receptor in malignant hyperthermia. J.

Biol. Chem. 263 :93 10-9315.

Mickelson, J.R, Litterer, L.A., Jacobson, BA and Louis, CF. 1990. Stimulation and
inhibition of [3H]ryanodine binding to sarcoplasmic reticulum from malignant hyperthermia

susceptible pigs. Arch. Biochem. Biophys. 278:251-257.

Mickelson, J. R, Knudson, C. M., Kennedy, C. F. H., Yang, D., Litterer, L. A, Rempel,
W.B., Campbell, KP. and Louis, C. F. 1992. Structural and functional correlates of a
mutation in the malignant hyperthemria-susceptible pig ryanodine receptor. FEBS Lett.

301 :49-52.

Mills, J.S., Walsh, M.P., Nemcek, K., & Johnson, J .D. 1988. Biologically active fluorescent
derivatives of spinach calmodulin that report calmodan target protein binding. Biochemistry

27:991-996.

177

Mitchell G. and Hefl‘ron. J.J.A 1982. Porcine Stress Syndromes. Adv. Food Res. 28:167-

230.

Morii H and Tonomura, Y. 1983. The gating behavior of a channel for Ca2*-induced Ca2+

release in fiagmented sarcOplasmic reticulum. J. Biochem. 93: 1271-1285.

Moutin, M.J. and Dupont, Y. 1988. Rapid filtration of Cay-induced Ca2+ release fiom skeletal

sarcoplasmic reticulum. J. Biol. Chem. 263:4228-4235.
Murachi, T. 1983. Calpain and calpastatine. Trends Biochem. Sci. 8:167-169.

Nagasaki, K. and Kasai, M. 1983. Fast release of calcium from sarcoplasmic reticulum

vesicles monitored by chlortetracycline fluorescence. J. Biochem. 94:1101-1109.

Nakai, J., Imagawa, T., Hakamata, Y., Shigekawa, M., Takeshima H. and Numa, S. 1990.
Primary structure and functional expression from cDNA of the cardiac ryanodine

receptor/calcium release channel FEBS Lett. 271: 169-177.

Nelson, TE. 1983. Abnormality in calcium release from skeletal sarcoplasmic reticulum of

pigs susceptible to malignant hyperthermia. J. Clin. Invest. 72:862-870.

178

Nelson, TE. 1987. Ryanodine: antithetical calcium channel effects in skeletal muscle

sarcoplasmic reticulum. J. Pharmacol. Exp. Ther. 242:56-61.

O'Brien, RJ. 1986. Porcine malignant hyperthermia susceptibility: hypersensitive calcium-

release mechanism of skeletal muscle sarcoplasmic reticulum. Can. J. Vet. Res. 50:318-328.

O’Driscoll, S., McCarthy, T.V., Eichinger, H.M., Erhardt, W., Lehmann-Horn, F.and
Herrmann-Frank, A Calmodulin sensitivity of the sarcoplasmic reticulum ryanodine receptor

fiom normal and malignant hypertherrnia susceptible muscle.(submitted to Biochem. J. 1996)

Ohnishi, S.T., Taylor, S. and Gronert, GA 1983. Calcium-induced Ca2+ release from
sarcOplasmic reticulum of pigs susceptible to malignant hypertherrnia: the effects of halothane

and dantrolene. FEBS Lett. 1612103-107.

Ohnishi, ST. 1987. Effects of halothane, cafl‘eine, dantrolene and tetracaine on the calcium
permeability of skeletal sarcoplasmic reticulum of malignant hyperthermic pigs. Biochim.

Biophys. Acta 897:261-268.

Ohta, T., Endo, M., Nakano, T., Morohoshi, Y., Wanikawa, K. and Ohga, A 1989. Ca2+-
induced Ca” release in malignant hyperthermia-susceptible pig skeletal muscle. Am. J.

Physiol. 256 (Cell Physiol. 25):C358-367.

179

O'Neil, K.T. and DeGrado, W.F. 1990. How calmodulin binds its targets: sequence

independent recognition of amphiphilic alpha-helices. Trends Biochem Sci. 15:59-64.

Otsu, K, erlard, HF, Khanna, V.K., Zorzato, F.,Green, NM. and MacLennan, DH. 1990.
Molecular cloning of cDNA encoding the Ca” release channel (ryanodine receptor) of rabbit

cardiac muscle sarcoplasmic reticulum J .Biol. Chem. 265: 13472-13483.

Otsu, K., Khanna, V.K., Archibald, AL. and MacLennan, DH. 1991. Cosegregation of
porcine malignant hyperthernria and a probable causal mutation in the skeletal muscle

ryanodine receptor gene in backcross families. Genomics 11:744-750.

Otsu, K., Nishida, K., Kimura, Y., Kuzuya, T., Hori, M., Karnada, T. and Tada, M. 1994.
The point mutation Arg‘” - Cysls in the Ci release channel of skeletal sarcOplasrnic
reticulum is responsible for hypersensitivity to cafi‘eine and halothane in malignant

hyperthermia. J. Biol. Chem. 269:9413-9415.
Perrin, DJ). and Sayce, LG. 1967. Talanta 14:833-842.

Pessah, I. N., Waterhouse, AL. and Casida, J. E. 1985. The calcium-ryanodine receptor

complex of skeletal and cardiac muscle. Biochem. Biophys. Res. Commun. 128:449-456.

180

Pessah, L N., Francini, AO., Scales, DJ, Waterhouse, AL. and Casida, J .E. 1986. Calcium-
ryandine receptor complex. Solubilization and partial characterization from skeletal muscle

junctional sarcoplasmic reticulum vesicles. J. Biol. Chem. 261 :8643 ~8648. '

Pessah, I.N., Starnbuk, RA and Casida, JE. 1987. Ca2+ activated ryanodine binding:
mechanisms of sensitivity and intensity modulation by Mg”, cafl‘eine and adenine nucleotides.

Mol. Pharmacol. 3 12232-238.

Plank, B.P., Wyskovsky, W. Hohenegger, M., Hellmann, G., and Suko, J. 1988. Inhibition
of calcium release from skeletal muscle sarcoplasmic reticulum by calmodulin Biochim.

Biophys. Acta 938279-88.

Podolsky, RJ. and Constantin, LL 1964. Regulation by calcium of the contraction and

relaxation of muscle fibers. Fed. Proc. Fed. Am. Soc. Exp. Biol. 23:933.

Pontremoli, S. and Melloni, E. 1986. Extra lysosomal protein degradation. Annu. Rev.

Biochem. 55:455-481.

Porter, K. R and Palade GE. 1957. Studies on the endoplasmic reticulum III. Its form and

distribution in striated muscle cells. J. Biophys. Biochem. Cytol. 3 :26.

181

Puca, G.A, Nola, E., Sica, V., and Bresciani, F. 1977. Estrogen binding proteins of calf
uterus. Molecular and functional characterization of the receptor transforming factor: A Ca2+-

activated protease. J. Biol. Chem. 252:1358-1366.

Radermacher, M, Wagenknecht, T., Grassucci, R, Frank, J. Inui, M. 1992. Cryo-EM of the
native structure of the calcium release channel/ryanodine receptor fiom sarcoplasmic

reticulum. Biophys. J. 61:936-940.

Radermacher, M., Rao, V., Grassucci, R, Frank, J., Timerman, A.P., Fleischer, S. and
Wagenknecht, T. 1994. Cryo-electron microscopy and three-dimensional reconstruction of
the calcium release channel/ryanodine receptor fi'om skeletal muscle. J. Cell Biol. 127:411-

422.

Rardon, D.P., Cefali, D.C., Mitchell, RD., Seiler, S.M., Hathaway, DR and Jones, LR
1990. Digestion of cardiac and skeletal muscle junctional sarcoplasmic reticulum vesicles with

calpain H: Efi‘ect on the Ca2+ release channel. Circ. Res. 67:84-96.

Rayns, D.G., Devine, CE. and Sutherland, CL. 1975. Freeze-fracture studies of membrane

systems in vertebrate muscle. 1. Striated muscle. J. Ultrastruct. Res. 50:306-321.

Rockstein, M. and Herron, P. 1951. Anal. Chem. 23: 1500-1501.

182

Rosemblatt, M., Hidalgo, C., Bergara, C. and Ikemoto, N. 1981. Immunological and
biochemical properties of transverse tubule membranes isolated from rabbit skeletal muscle.

J. Biol. Chem. 256:8140-8148.

Schagger, H. And J agow, G.V. 1987. Tricine-Sodium Dodecyl Sulfate-Polyacrylamide Gel
Electrophoresis for the separation of proteins in the range flour 1 to 100 kDa. Anal. Biochem.

166:368-379.

Schneider, MF. and Chandler, W.K. 1973. Voltage dependent charge movement in skeletal

muscle: a possible step in excitation-contraction coupling. Nature 242:244-246.

Schneider, MF. 1981. Membrane charge movement and depolarization-contraction coupling.

Annu. Rev. Physiol. 43:507-517.

Schneider, MF. 1994. Control of calcium release in functioning skeletal muscle fibers. Annu.

Rev. Physiol.$6:463-484.

Seiler, S., Wegener, A.D., Whang, D.D., Hathaway, DR and Jones, LR 1984. High
molecular weight proteins in cardiac and skeletal muscle junctional sarcoplasmic reticulum
vesicles bind calmodulin, are phosphorylated, and are degraded by Ca2*-activated protease.

J. Biol. Chem. 259:8550-8557.

183
Serysheva, I.I., Orlova, E.V., Chiu, W., Sherman, MB, Hamilton, S.L. and van Heel, M.
1995. Electron cryomicroscopy and angular reconstitution used to visualize the skeletal

muscle calcium release channel. Nature Structural Biology 2: 18-24.

Shomer, N.H, Louis, C.F., Fill, M., Litterer, LA and Mickelson, JR 1993. Reconstitution
of abnormalities in the malignant hyperthermia-susceptible pig ryanodine receptor. AM. J.

Physiol. 264 (Cell Physiol. 33):C125-C135.

Shomer, N.H., Mickelson, J .R. and Louis, C.F. 1994a. Caffeine stimulation of malignant
hyperthermia-susceptible sarc0plasmic reticulm Ca2+ release channel. Am. J. Physiol. 267

(Cell physiol. 36): C1253-C1261.

Shomer, N. H., Mickelson, J .R. and Louis, C.F. 1994b. Ion selectivity of porcine skeletal
muscle Ca2+ release channels is unaffected by the Arg615 to Cys615 mutation. Biophys. J.

67:641-646.

Shoshan-Barmatz,V. and Zarka, A. 1988. Trypsin destruction of the high affinity ryanodine

binding sites of the junctional sarcoplasmic reticulum. J. Biol. Chem. 263:16772-16779.

Shoshan-Barmatz,V., Weil, S., Meyer, H, Varsanyi, M. and Heilmeyer, L.M. 1994.
Endogenous, Ca (2+)-dependent cysteine-protease cleaves specifically the ryanodine

receptor/Ca 2“ release channel in skeletal muscle. J. Membr. Biol. 142:281-288.

184

Simpson, SP. and Webb, AJ. 1989. Growth and carcass performance of British Landrace

pigs heterozygous at the halothane locus. Anim Prod. 49:503.

Smith, J .S., Coronado, R and Meissner, G. 1985. Sarcoplasmic reticulum contains adenine

nucleotide-activated calcium channels. Nature 316:446-449.

Smith, J .S., Coronado, R and G. Meissner. 1986a. Single channel calcium and barium
currents of large and small conductance from sarcoplasmic reticulum. Biophysical J. 50:921-

928.

Smith, J .S., Coronado, R and G. meissner. 1986b. Single channel measurements of the
calcium release channel fiom skeletal muscle sarcoplasmic reticulum. Activation by calcium,

ATP, and modulation by magnesium. J. Gen. Physiol. 88:573-588.

Smith, J.S., Imagawa, T., Fill, M., Campbell, KP. and Coronado, R 1988. Purified ryanodine
receptor from rabbit skeletal muscle is the calcium-release channel of the sarcoplasmic

reticulum. J. Gen. Physiol. 92: 1-26.

Smith, J .S., Coronado, R. and Meissner, G. 1989. Calmodulin regulation of single
sarcoplasmic reticulum Ca2*-release channels from cardiac and skeletal muscle. Circ. Res.

64:352-359.

185

Somlyo, AV., Gonzalez-Serums, H, Shuman, H, McClellan, G. and Somlyo, AP. 1981.
Calcium release and ionic changes in the sarcoplasmic reticulum of tetanized muscle: an

electron-probe study. J. Cell Biol. 90:577-594.

Strand, MA, Louis, CF. and Mickelson, JR 1993. Phosphorylation of the porcine skeletal
and cardiac muscle sarcoplasmic reticulum ryanodine receptor. Biochem. Biophys. Biophys.

Acta 1175:319-326.

Strasburg, G.M., Hogan, M., Birrnachu, W., Thomas, DD, and Louis, CF. 1988. Site-
specific derivatives of wheat germ calmodulin interactions with troponin and sarcoplasmic

reticulum. J. Biol. Chem. 263:542-548.

Suko J, Maurer-Fogy, I., Plank, B., Bertel, 0., Wyskovsky, W., et al. 1993. Phosphorylation
of serine 2843 in ryanodine receptor-calcium release channel of skeletal muscle by cAMP-

,cGMP- and CaM-dependent protein kinase. Biochim. Bi0phys. Acta 1175: 193-206.

Sutko, J .L., Ito, K and Kenyon, J.L. 1985. Ryanodine: a modifier of sarcoplasmic reticulum

calcium release in striated muscle. Fed. Proc. 44:2984-2993.

Suzuki, K 1987. Calcium activated neutral protease: domain structure and activity regulation.

Trends Biochem. Sci. 12: 103-105.

186

Tada, M, Yamamoto, T. and Tonomura, Y. 1978. Molecular mechanism of active calcium

transport by sarcoplasmic reticulum. Physiol. Rev. 58: 1-79.

Takeshima, H, Nishimura, S., Matsumoto, R, Ishida, H, Kanawa, K, Minamrnino, N.,
Matsuo, H, Ueda M, Hanaoka, M, Hanaoka, M., Hirose, T. and Numa, S. 1989. Primary
structure and expression fi'om complementary DNA of skeletal muscle ryanodine receptor.

Nature 339:439-445.

Thatte, H.S., Addis, P.B., Thomas, D.D., Bigelow, D.B., Mickelson, J .R and Louis, CF.
1987 . Temperature-dependent abnormalities of the erythrocyte membrane in porcine

malignant hyperthermia. Biochem. Med. Metab. Biol. 38:366-377.

Toda, H, Yazawa, M., Sakiganm, F. and Yagi, K. 1985. Amino acid sequence of calmodulin

fi'om wheat germ. J. Biochem. 98: 833-842.

Torolg K and Whitaker, M. 1994. Taking a long, hard look at calmodulin’s warm embrace.

BioEssays 16:221-224.

Treves, S., Chiozzi, P. and Zorzato. F. 1993. Identification of the domain recognized by anti-
(ryanodine receptor) antibodies which affect Ca2"-induced Ca2+ release. Biochem. J. 291 :75 7-

763 .

187

Tripathy, A, Xu, L., Mann, G. and Meissner, G. 1995. Calmodulin activation and inhibition

of skeletal muscle Ca2+ release channel (ryanodine receptor). Biophy. J. 69:106-119.

Vedeckis, W.V., Freeman, MR, Schrader, WT. and O’Malley, B.W.1980. Progesterone-
binding components of chick oviduct: partial purification and characterization of a calcium-

activated protease which hydrolyzes the progesterone receptor. Biochemistry 19:33 5-3 43.

Vergara, J ., Tsien, RY., and Delay, M 1985. Inositol 1,4,5-triphosphate: A possible chemical

link in arcitation-contraction coupling in muscle. Proc. Natl. Acad. Sci. USA 82:63 52-6356.

Volpe, P., Salviati, G., DiVirgilio, F ., and Pozzan, T. 1985. Inositol 1,4,5-triphosphate

induces calcium release from sarcoplasmic reticulum of skeletal muscle. Nature 316:347-349.

Wagenknecht, T., Grassuci R Frank, J. Saito, A, Inui, M., Fleischer, S. 1989. Three-
dimensional architecture of the calcium channel/foot structure of sarcoplasmic reticulum.

Nature 338:167-170.

Wagenknecht, T., Berkowitz, J ., Grassucci, R, Tirnerrnan, A. P., and Fleischer, S. 1994.
Localization of calmodulin binding sites on the ryanodine receptor fi'om skeletal muscle by

electron microscopy. Biophys. J. 67:2286-2295.

188

Walker, J.W., Somolyo, AV., Goldman, Y.E., Somolyo, AP., and Trentharn, DR. 1987.
Kinetics of smooth and skeletal muscle activation by laser pulse photolysis of caged inositol

1,4,5 triphosphate. Nature 327:249-252.

Wang, J. and Best, RM 1992. Inactivation of the sarcoplasmic reticulum calcium channel by

protein kinase. Nature 359:73 9-741.

Wang, KKW., Villalobo, A. and Roufogalis, RD. 1989. Calmodulin-binding proteins as

calpain substrate. Biochem. J. 262:518-525.

Weber, A and Sanadi, DR ed. 1966. Current tOpics in bioenergetics. New York: Academic

Press pp.203.

Webb, AJ. and Simpson, SP. 1986. Performance of Britich Landrace pigs selected for high

and low incidence of halothane sensitivity. 11. Growth and carcass traits. Anirn. Prod. 43:493.

Vlfrsmer-Pedersen, J. and Briskey, E]. 1961. Rate of anaerobic glycolysis versus structure in

pork muscle. Nature 189:318.

Witcher, D.R, Kovacs RJ., Schulrnan, H, Cefali, DC. and Jones, LR. 1991. Unique
phosphorylation site on the cardiac ryanodine receptor regulates calcium channel activity.

J.Eiol. Chem. 266111144-11152.

189

Yang, H, Reedy, M.M., Burke, C. And Strasburg, GM. 1994. Calmoculin interaction with

the skeletal muscle sarcoplasmic retiarlum calcium channel protein Biochemistry 33 :5 18-525.

Zhang, M., Tanaka, T. and Ikura, M. 1995. Calcium-induced conformational transition

revealed by the solution structure of apo calmodulin. Nature Structure Biology 2:758-767.

Zimanyi, I. and Pessah, I. 1991. Comparison of [’H]ryanodine receptors and Ca2+ release from
rat cardiac and rabbit skeletal muscle sarcoplasmic reticulum. J. Pharmacol. Exp. Ther.

256:938-946.

Zimanyi, 1., Buck, E., Abrarnson, J., Mack, M. and Pessah, I. 1992. Ryanodine produces
persistent inactivation of the Ca2+ release channel fi'om skeletal muscle sarcoplasmic

reticulum. Mol. Pharmacol. 42: 1049-1057.

Zorzato, F ., Fujii, J ., Otsu,K, Phillips, M., Green, N.M., Lai, F .A, Meissner, G. and
MacLennan. 1990. Molecular cloning of cDNA encoding human and rabbit forms of the Ca2+
release channel (ryanodine receptor) of skeletal muscle sarcoplasmic reticulum. J. Biol. Chem.

265:2244-2256.

nrcuran srnr: UNIV. LIaRnRIEs
\IHIHINIWIIVHINWIVIWmmWINIH‘IHIHW
31293014138998