I. 0...,
.r
v

.v.:‘.
. ..

s '3: .5:

.z ‘.
1...: . 4..

...12.:X:.-V.
.. >1 {.22}. (-11.
. 1. .. .5} :

‘63:. u.
.: .

o (- that
811...} .7 .

1’51:

#1:.) u; t.

3.... .. fiat

(-
u

.0
12.31. I

 

nquIVoI ‘VI
. IJ-Iv.lh§ vii v
gilt..lr-||ԢҢ.

O! t A:
I‘vrltob'il Iii. ll

 

TH£$QL

llllllllulll

This is to certify that the

dissertation entitled

Effects of Lambert-Eaton Myasthenic Immunoglobulin on
Synaptosomal Calcium Channels: Specificity, Serum Dependence
and Subtype

presented by

Sandra Jeanne Hewett

has been accepted towards fulfillment
of the requirements for

ph,D, Pharmacology/Toxicology .

 

 

 

 

degree in
“X Z ' ~ - lit/4%
{5/ Major professor
Datefléfwfojg /??;\

.MSU is an Affirmative Action/Equal Opportunity Institution 0-12771

f a.
LIBRARY '

Michigan State

l University

‘ .J

—_

 

 

PLACE IN RETURN BOX to remove this checkom from your record.
TO AVOID FINES return on or before date due.

’féfi"

DATE DUE DATE DUE DATE DUE

#
‘1
J
—l

 

 

 

 

 

 

 

 

 

 

 

 

MSU Is An Affirmative Action/Equal Opportunity Institution
cWMS—DJ

 

 

 

 

 

 

 

 

 

 

 

EFFECTS OF LAMBERT-EATON MYASTHENIC IMMUNOGLOBULIN ON
SYNAPTOSOMAL CALCIUM CHANNELS:
SPECIFICITY, SERUM DEPENDENCE AND SUBTYPE
BY

Sandra Jeanne Hewett

A DISSERTATION

Submitted to
Michigan State University
in partial fulfillment of the requirements
for the degree of

DOCTOR OF PHILOSOPHY

Department of Pharmacology and Toxicology

1992

ABSTRACT

EFFECTS OF LAMBERT-EATON MYASTHENIC IMMUNOGLOBULIN ON
SYNAPTOSOMAL CALCIUM CHANNELS:
SPECIFICITY, SERUM DEPENDENCE AND SUBTYPE

BY

Sandra Jeanne Hewett

Lambert-Eaton Myasthenic Syndrome (LEMS) is a presynaptic disorder of
neuromuscular transmission. The reduction in nerve evoked release of transmitter
at the neuromuscular junction of patients with LEMS is thought to be caused by
an autoantibody to motor nerve terminal Ca2+ channels. The overall goal of this
dissertation was to test the hypothesis that a circulating autoantibody in the serum
of patients with LEMS interferes with the function of nerve terminal Ca2+ channels
associated with transmitter release. To test this a model system of isolated nerve
terminals from mammalian CNS (synaptosomes) was used.

Serum and plasma from several patients and IgG from a single patient with
LEMS reduce 45Ca2+ influx into synaptosomes. The antibody appeared specific for
Ca2+ channels as it was unable to disrupt 22Na+ uptake into or 86Rb" efflux from
Na+ or K+ channels, respectively. Synaptosomal membrane integrity as well as
synaptosomal membrane potential were unaltered following incubation with the
LEMS autoantibody.

In synaptosomes, the presence of human serum is important in mediating

a functional change in Ca2+ channel activity by LEMS lgG. In the absence of

human serum, LEMS lgG did not reduce uptake of 45Ca2+ into synaptosomes.
Therefore, the role of the serum component complement in the pathogenesis of
LEMS was assessed. Neither the membrane attack complex nor the alternative
pathway of complement contributed to the disruption of Ca2+ channel function by
LEMS lgG and serum in synaptosomes. In contrast, the 03 component of the
complement cascade appeared to be important in mediating Ca2+ channel
dysfunction.

The pharmacological determination of the Ca2+ channel subtype affected
by lgG from a patient with LEMS was next examined. The density of binding sites
for [3H]-verapamil but not [3H]-nitrendipine nor [‘zsl]-ro-conotoxin was reduced by
approximately 45% following incubation with LEMS lgG alone or LEMS lgG and
serum. This suggests that the IgG molecule itself interacts at this site. Verapamil
is presumed to act as an L-type Ca“ channel antagonist but recent evidence
suggests that it may not be as selective as was previously thought.

The studies in this dissertation were the first to examine specifically the
ability of antibody from LEMS patients to interact with nerve terminal Ca”+
channels. This work has strengthened the hypothesis that Ca2+ channels are the
antigenic target, provided new insight into the mechanism of Ca2+ channel
dysfunction, and explored pharmacologically the subtype of Ca2+ channels in a

nerve terminal preparation affected by the LEMS autoantibody.

To my family, with all my love
Especially my parents, Ann and Dennis Simas
and
my husband, Jim

ACKNOWLEDGEMENTS

I would like to thank my mentor, Dr. Vlfilliam Atchison, for providing me with
the opportunity to grow and develop into an independent researcher. I am grateful
for the excellent communication skills I have developed as a direct result of his
guidance. He has been an invaluable role model demonstrating that persistence
and patience invariably pay off.

I’d also like to thank the members of my thesis committee, Drs. James
Bennett, Kathy Brooks, Jim Galligan and Mary Rheuben for their guidance and
flexibility. Special thanks to Drs. Jim Galligan, David Reinke and Mary Rheuben
who were always willing to listen and lend advice. I am grateful to everyone in the
laboratories of 3331, 346 and 347 (past and present), who are too numerable to
list, for creating a most enjoyable working environment. I’d especially like to thank
Annette McLane, Marintha Heil, Lynne Ireland, Drs. Jean Ray and Cindy Hoom for
their friendship. They always seem to understand me and my moods as well as
provide a shoulder when ever necessary.

I am grateful for the excellent technical assistance of Michelle Welke, Paul
Bertrand, Kevin Borden, Becky Heydecke and Annette McLane. I hope that their
experiences in the laboratory were as worthwhile to them as their work has been
to me. Special thanks to Gretchen Humphries for her technical, artistic and
secretarial assistance.

Most importantly, I would like to acknowledge my family. I would especially
like to thank my mother and father, Ann and Dennis Simas, for always believing in
me and teaching me that I could accomplish anything with hard work, discipline
and spirit. I thank them for giving me the freedom to live, to grow and to leave
even when I know it was difficult to let go. I’d like to thank John and Eileen Hewett
for being the best second parents for which anyone could ever ask. Their support

of both me and their son have helped us to accomplish lofty goals.

Finally, I am forever grateful to my husband Jim. His strength, support, wit,
companionship, understanding, and love have helped make this endeavor not only
bearable but worthwhile. He encouraged me from beginning to end and always
believed in me even when I had trouble believing in myself. Jim, for everything I

thank you and I love you.

vi

PREFACE

Some of the data contained in this dissertation have been published previously.

Chapter Three appeared as Hewett and Atchison (1991).

vii

TABLE OF CONTENTS

viii

LIST OF TABLES ................................................................................................ xii

LIST OF FIGURES ............................................................................................. xiii

LIST OF ABBREVIATIONS ................................................................................. xvi
CHAPTER ONE: INTRODUCTION

A. General Introduction ................................. 2

B. Normal Neuromuscular Transmission ..................... 3

1. Synaptic transmission ............................ 3

2. Calcium channel subtypes ........................ 8

C. Disorders of Neuromuscular Transmrssion ................ 14

D. Lambert-5mm Myasthenic Syndrome: General Description ..... 15

1. Clinical Features ............................... 15

2. Electrophysiology .............................. 17

a. Clinical ................................. 17

b. Experimental ............................ 19

3. Biochemistry ................................. 20

4. Morphology .................................. 20

a. Morphometric analysis ..................... 20

b. Ultrastructural analysis ..................... 22

E. Lambert-Eaton Myasthenic Syndrome: Autoimmunity .......... 23

1. Clinical and experimental evidence .................. 23

2. The Antigenic Target ........................... 25

a. Calcium channel hypothesis ................. 25

b. Calcium channel subtype affected by LEMS lgG . . . 29

3. Pathophysiological Mechanisms ................... 30

a. Direct inhibition ........................... 30

b. Antigenic modulation ...................... 31

c. Complement-mediated destruction ............. 32

F. The Immune Response ............................... 36
1. Cellular interactions in the immune system ............ 37

2. Self- non-self recognition ......................... 40

a. Clonal Deletion ........................... 40

b. Clonal Anergy ........................... 41

3. Breakdown of self-tolerance ...................... 43

a. Decreased Suppression .................... 44

b. Modification of an autoantigen ................ 46

0. Self antigen plus aberrant self-encoding molecules . 46

d. Release of sequestered antigens .............. 47

e. Cross-reactivity ........................... 47

G. Autoimmune LEMSandtheroleotcancer ................. 49

CHAPTER TWO: SYNAPTOSOMES AND SYNAPTOSOMAL ION CHANNELS

A General Introduction ................................ 53
B. Synaptosomes ..................................... 54
1. General Description ............................ 54

2. Synaptosomal ion channels ....................... 55

a. Sodium channels ......................... 55

b. Potassium channels ....................... 56

c. Calcium channels ......................... 57

i. Ca2+ channel subtypes ................. 60

C. Overall Aim ....................................... 62
1. Specific Aim 1 ................................ 64

2. Specific Aim 2 ................................ 64

ix

3. Specific Aim 3 ................................ 65
4. Specific Aim 4 ................................ 65
CHAPTER THREE: EFFECT OF SERUM AND PLASMA FROM PATIENTS WITH

LAMBERT-EATON MYASTHENIC SYNDROME ON DEPOLARIZATION-
INDUCED UPTAKE OF “Ca2+ INTO RAT CORTICAL SYNAPTOSOMES

A. Summary ........................................ 68
B. Introduction ....................................... 69
C. Materials and Methods ............................... 70
D. Results .......................................... 75
E. Discussion ....................................... 78

CHAPTER FOUR: SPECIFICITY OF LAMBERT-EATON MYASTHENIC
IMMUNOGLOBULIN FOR NERVE TERMINAL CALCIUM CHANNELS

A. Summary ........................................ 87
B. Introduction ....................................... 88
C. Materials and Methods ............................... 90
D. Results .......................................... 99
E Discussion ...................................... 1 14

CHAPTER FIVE: DISRUPTION OF SYNAPTOSOMAL 032+ CHANNELS BY
LAMBERT-EATON MYASTHENIC SYNDROME IMMUNOGLOBULIN: SERUM
DEPENDENCY

A. Summary ....................................... 122
B. Introduction ...................................... 123
C. Materials and Methods .............................. 124
D. Results ......................................... 126
E. Discussion ...................................... 136

CHAPTER SIX: PHARMACOLOGICAL DETERMINATION OF SYNAPTOSOMAL
C CHANNEL SUBTYPE S AFFECTED BY LAMBERT-EATON MYASTHENIC
SYNDROME IMMUNOGL ULIN

A. Summary ....................................... 143
B. Introduction ...................................... 144
C. Materials and Methods .............................. 146
D. Results ......................................... 150
E Discussion ...................................... 157
F Note ........................................... 162

CHAPTER SEVEN: SUMMARY AND CONCLUSIONS
A. Summary and Conclusions ........................... 164

LIST OF REFERENCES ................................... 173

LIST OF TABLES

Table 1.1. Pharmacological and Biophysical Characteristics of Mammalian
Ca2+ Channel Types ............................. 10
Table 4.1. Determination of Ouantal Content via Direct Method ....... 101

Table 6.1 [3H]Nitrendipine Binding Characteristics ................ 152

xii

Figure 1.1.
Figure 1.2.
Figure 1.3.
Figure 1.4.
Figure 1.5.

Figure 3.1.

Figure 3.2.

Figure 3.3.

Figure 3.4.
Figure 3.5.
Figure 4.1.

Figure 4.2.

Figure 4.3.

Figure 4.4.

Figure 4.5.

Figure 4.6.

Figure 4.7.

LIST OF FIGURES

Neuromuscular Transmission ......................... 4
Electromyographic Characteristics of LEMS .............. 18
Complement .................................... 33
Immune Response ................................ 38
Pathogenesis of autoimmune disease .................. 45
Effect of sera from 4 patients with SCC on "‘SCaz+ uptake into

rat forebrain synaptosomes ........................ 76
Effect of plasma and sera from 5 patients with LEMS on

uptake of 45Ca2+ into rat forebrain synaptosomes ......... 77
Effect of serum from one patient with LEMS on “Ca2+ uptake

into synaptosomes. .............................. 79
Effect of LEMS sera on synaptosomal viability. ............ 80
Effect of LEMS plasma on synaptosomal viability. .......... 81
Setup for recording under current clamp ................ 94
Effects of chronic treatment of mice with LEMS plasma on

quantal content. ................................ 100

Frequency-dependent facilitation of neuromuscular transmission 102

Effect of LEMS lgG on uptake of “Ca2+ into rat forebrain

synaptosomes ................................. 103
Effect of lgG from patients with small cell carcinoma ($00) on

uptake of “Ca2+ into rat forebrain synaptosomes ........ 105
Effect of LEMS lgG on synaptosomal membrane integrity. . . . 106
Representative graph depicting typical fluorescent dye

experiment ................................... 108

xiii

Figure 4.8. Effect of LEMS lgG on synaptosomal membrane potentials . . 109
figure 4.9. Time course of 22Na+ uptake into rat forebrain synaptosomes. 110
Figure 4.10. Effect of LEMS lgG on uptake of ”Na“ into rat forebrain
synaptosomes ................................. 1 1 1
figure 4.11. Time course of ast+ loading into synaptosomes ........ 112
figure 4.12. Effect of LEMS lgG on ”RE efflux from rat forebrain
synaptosomes ................................. 1 13
Figure 5.1. Effect of LEMS lgG on uptake of 45Ca2+ into rat forebrain
synaptosomes in the presence and absence of control
human serum ................................. 127
figure 5.2. Effect of control human serum on “Ca2+ uptake into rat
forebrain synaptosomes .......................... 128
figure 5.3. Effect of BSA on “Caz+ uptake into rat forebrain
synaptosomes ................................. 129
figure 5.4. Effect of control and LEMS lgG and various concentrations of
control human serum on 45Ca2+ uptake into rat forebrain
synaptosomes ................................. 131
Figure 5.5. The effect of LEMS lgG in the presence of control or heat inactivated
serum on uptake of “Ca2+ into rat synaptosomes ....... 132
figure 5.6. The effect of 05-deficient human serum on the LEMS lgG induced
reduction of K”-stimulated 45Ca2+ uptake into rat synaptosomesl33
figure 5.7. The effect C3-deficient human serum on the LEMS IgG in the
reduction in 1 sec K+-stimulated 45Ca2+ uptake into rat
synaptosomes. ................................ 134
figure 5.8. The effect of control and LEMS lgG in the presence of Factor
B-deficient human serum 45Ca2 uptake into rat synaptosomes.135
figure 6.1. Effect of serum on nitendipine binding to synaptosomes . . . . 151
figure 6.2. Comparative effects of Control and LEMS lgG on nitrendipine
binding to synaptosomes. ............................. 153

figure 6.3. Comparative effects of control and LEMS lgG on Cng binding
to rat forebrain synaptosomes. ..................... 155

figure 6.4. Comparative effects of control and LEMS lgG on verapamil
binding to synaptosomes. ......................... 156

LIST OF ABBREVIATIONS

ACh - acetylcholine

AChE — acetylcholinesterase

ANOVA - analysis of variance

APC - alternative pathway of complement
AZPs - active zone particles

B"m - maximum number of binding sites
BSA - bovine serum albumin

BoTX - botulinum toxin

BTX - batrachatoxin

Ca2+ - calcium

[Ca2+], - extracellular concentration of calcium
ChAT - choline acetyltransferase

CMAP - compound muscle action potential
CNS - central nervous system

Cng - m -conotoxin GVIA

CVF - cobra venom factor

DHP - dihydropyridine

dis-02(5) 3,3’-diethyldicarbocyanine iodide
DMSO - dimethyl sulfoxide

DRG - dorsal root ganglion

EGTA - ethyleneglycol bis-(B-aminoethyl ether)-N,N,N,N’-tetra acetic acid
EPP - endplate potential

EtOH - ethanol

Fab - monovalent antigen binding fragment of IgG (papain-digested)
F(ab)’2 - divalent antigen binding fragment of lgG (pepsin-digested)
FIX - funnel web spider toxin

HEPES - N-2-hydroxyethylpiperazine-N’2-ethanesulfonic acid

lgG - immunoglobulin G

lgs - total immunoglobulin

K“ - potassium

[K"], - extracellular concentration of potassium

kDa- kilodalton

'Kd - equilibrium dissociation constant for ligand binding

LDH - lactate dehydrogenase

LEMS - Lambert-Eaton Myasthenic Syndrome

MAC - membrane attack complex

MEPP - miniature endplate potential

MG - myasthenia gravis

Mg2+ - magnesium

MHC - major histocompatability complex

mRNA - messenger ribonucleic acid

msec - millisecond

Na+ - sodium

NaCN - sodium cyanide

NiCI2 - nickel chloride

NMJ - neuromuscular junction

PAGE - polyacrylamide gel electrophoresis
PKC - protein kinase C

800 - small cell carcimona of the lung
SDS - sodiumdodecylsulfate

sec - seconds

SEM - standard error of the mean
synaptosomes - isolated nerve terminals from rat forebrain
TBA - tetrabutylammonium chloride

TEA - tetraethylammonium chloride

Th cells - helper T lymphocytes

T, cells - suppressor T lymphocytes

TTX - tetrodotoxin

VER - veratridine

4-AP - 4-aminopyridine

xviii

CHAPTER ONE: INTRODUCTION

A. General Introduction:

Lambert-Eaton myasthenic syndrome (LEMS) is an autoimmune disorder
characterized symptomatically by a generalized muscle weakness. The
neuromuscular junction (NMJ) was implicated as the site of the abnormality in
LEMS when no aberration could be found in the responses of the motor nerve or
muscle to direct electrical stimulation. The presynaptic terminal, specifically, was
implicated as the site of action of the LEMS autoantibody when analysis of the
NMJ in biopsy specimens from patients with the syndrome demonstrated that the
postsynaptic membrane responds normally to acetylcholine (ACh). Since a
presynaptic defect could involve an abnormality in transmitter synthesis, storage,
or release, a great deal of effort has been expended in order to determine which
of these processes is affected. Experimental evidence indicates that both synthesis
and storage of ACh are normal. Thus, the abnormality appears to be caused by
a defect in neurotransmitter release. It has been suggested that membrane
calcium (Ca2+) channels are the antigenic target for the LEMS autoantibody.

The initial sections of Chapter One will introduce normal neuromuscular
transmission. These sections provide a description of the components of the
neuromuscular junction as well as the dependence of transmission on [Ca2+]. ions.
Ca2+ channel subtypes believed to be involved in neurotransmitter release, in
general, are discussed in detail. This is followed by a detailed discussion of the
autoimmune, neuromuscular disorder, LEMS with attempts to delineate the stage

at which normal functioning is altered by a pathological process. The final sections

3

of this chapter discuss the normal immune response focussing on the mechanisms
of self- non-self recognition and how the breakdown of self-tolerance can lead to

autoimmunity.

B. Normal Neuromuscular Transmission:

1. Synaptlc transmlsslon

The synapse formed between motor nerves and muscle cells is known as
the neuromuscular junction (NMJ). The NMJ can be considered a transducing
mechanism, in which electrical signals (nerve impulses) are transduced to
chemicals signals (neurotransmitter release) and then back to electrical signals
(muscle action potentials). The transmission of an impulse from the motor nerve
to the muscle fiber involves a complex sequence of events. The structures and
steps involved in neuromuscular transmission are shown in figure 1.1.

The presynaptic part of the transducer consists of the nerve terminal. The
terminal rests in a depression in the surface of the muscle fiber, the postsynaptic
part of the transducer, and is separated from the muscle by the synaptic cleft
(Bowden and Duchen, 1976). The nerve terminal contains large numbers of
synaptic vesicles, each containing approximately 10,000 to 20,000 molecules of
ACh (Potter, 1970; Kuffler and Yoshikami, 1975). Each vesicle represents one
quantum of ACh (del Castillo and Katz, 1956; Martin, 1966; Hubbard, 1970).
Transmitter release from the nerve terminal occurs both spontaneously and as a

result of invasion of the axon terminal by an action potential. The spontaneous

 

5 - choline +
acetate

 

 

 

 

FIGURE 1.1. Neuromuscular Transmission

Following stimulation of the nerve, an action potential will be propagated down the
axon, leading to depolarization of the nerve terminal (1). Voltage-sensitive Ca2+
channels open and allow Ca2 ions to enter into the terminal (2). An increase in
intraterminal Ca2+ triggers the fusion of synaptic vesicles with the plama membrane
and the subsequent release of ACh into the synaptic cleft (3). ACh binds to
nicotinic ACh receptors in the muscle membrane resulting in an excitatory
postsynaptic potential (4). Diffusion of ACh away from the receptors and hydrolysis
of ACh by AChE rapidly terminates the ACh-receptor combination (5).

5

release of one quantum of ACh leads to tiny depolarizations of the postsynaptic
membrane known as miniature end-plate potentials (MEPPs) (Fatt and Katz, 1951;
Boyd and Martin, 1956a; Liley, 1956). Stimulated release of many quanta of ACh
leads to an excitatory postsynaptic potential which at the NMJ is termed an
endplate potential (EPP) (Fat and Katz, 1959; Takeuchi, 1963). The number of
vesicles released per nerve impulse is known as the quantal content of the EPP.

Action potentials are propagated along the axon of the motor neuron by
movement of Na+ and K+ into and out of the axon through their respective
channels (Hodgkin and Huxley, 1952). Once the action potential reaches the axon
terminal, it causes a depolarization of the terminal membrane (Brigant and Mallart,
1982; Mallart, 1985). As a result, voltage-dependent Ca2+ channels in the nerve
terminal open allowing extracellular Caz+ ions to enter down their electrochemical
gradient (Katz and Miledi, 1967a,b; Uinas et al., 1981). It is thought that these
channels are clustered in specialized areas or active zones, where vesicles
containing neurotransmitter are arranged in close apposition to the cell membrane
for subsequent release. (Heuser et al., 1974, 1979; Ceccarelli and Hurlbut, 1980;
Pumplin et al., 1981; Robitaille et al., 1990). ctat2+ entry produces the near
simultaneous exocytotic release of large numbers of quanta of ACh from the active
zone area.

The presence of [Ca2 *1. is required for any form of depolarization to release

ACh from nerve terminals. This requirement is not connected to the propagation

6

of action potentials into nerve terminals, for in conditions of Ca2+ deprivation,
action potentials can still be recorded even though no ACh release occurs (Katz
and Miledi, 1965; Dodge et al., 1969). The sequence of molecular events linking
depolarization-dependent Ca2+ influx to the release of neurotransmitters from nerve
terminals is unknown. However, interactions between Ca2+ and intracellular Ca2+-
binding proteins are thought to trigger directly or indirectly the fusion of synaptic
vesicles with the cell membrane, resulting in the release of neurotransmitter into the
synaptic cleft (Augustine et al., 1987).

One model suggests that activation of a Ca2+-binding protein in turn
activates a protein kinase which phosphorylates a protein or proteins essential for
the release of neurotransmitter (Krueger et al., 1977; DeLorenzo et al., 1979;
Dekker et al., 1989a). Evidence for this centers on studies which demonstrate that
the phosphorylation of synapsin l and B-50 (GAP-43) by Ca2*/calmodulin-
dependent protein kinase II and Protein kinase C, respectively, correlate with
neurotransmitter release. (Uinas et al., 1985; Robinson and Dunkley, 1985; Dekker
et al., 1989a; Sihra et al., 1989; Dekker et al.,1990; Nichols et al., 1992).
Furthermore, antibodies to 8-50 or injection of deposphorylated synapsin l inhibit
release from the nerve terminal (Uinas et al., 1985; Dekker et al., 1989b).

Synaptophysin, synaptotagmin, synexin, calpactin and other members of the
calelectrin family possess a cytoplasmic Caz*-binding site and have also been
implicated in the control of neurotransmitter release (Augustine et al., 1987; Smith

and Augustine, 1988; Brose et al., 1992). Depolarization of nerve terminals is

7
accompanied by the activation of a calcium-stimulated protein phosphatase, as

well. The physiological significance of the rapid dephosphorylation of the
phosphoproteins, dephosphin, P96 and P139 (Robinson et al., 1987; Sihra et al.,
1992) has yet to be determined.

Once released, ACh crosses the synaptic cleft and binds to nicotinic ACh
receptors located on the crests of the synaptic folds across from release sites
(Landau, 1978; Ceccarelli and Hurlbut, 1980). The interaction between ACh and
its receptor produces a conductance change in the postsynaptic membrane,
markedly increasing the membranes permeability to Na+ and K+ (Takeuchi and
Takeuchi, 1959; Takeuchi, 1963). This produces a local depolarization in the end-
plate (Fatt and Katz, 1951; Boyd and Martin, 1956b). EPPs of sufficient magnitude
spread from the endplate region to the surrounding muscle fiber membrane
initiating an action potential. This leads to contraction of the muscle (Hubbard,
1973)

Diffusion of ACh away from the receptors and hydrolysis of ACh by
acetylcholinesterase (AChE) rapidly terminates the ACh-receptor combination. The
choline molecules liberated by AChE action are actively taken up by nerve
terminals for resynthesis into ACh (choline + acetyl CoA) by choline
acetyltransferase (ChAT)(Collier and Macintosh, 1969; Potter, 1970; Kuher and

Murrin, 1978).

2. Calclum channel subtypes

Voltage-dependent Ca2+ channels occur in most neurons and in several
other cell types. As stated above, influx of [Ca2+], through membrane ionic
channels is the critical step for coupling the action potential to transmitter secretion
(Katz and Miledi, 1965; 1967a,b; Dodge and Rahamimoff, 1969). Evidence
indicates that more than one type of calcium channel exists (Nowycky, et al.,
1985a,b; Perney, et al., 1986). This discovery has prompted questions as to which
subtype(s) is (are) involved with the release of neurotransmitter. Wlth the exception
of some invertebrate neurons, characterization of the type of channel that exists
at the active zone is impossible as nerve terminals are too small for direct
recording techniques. Therefore, electrophysiological and pharmacological
characterizations of the various types of calcium channels were made in the cell
some with the assumption that somal events model events occurring in the nerve
terminal. Thus, the exact identity of the subtypes of calcium channels in the nerve
terminal remains unknown. It is possible that the characteristics of calcium
channels in the terminal may differ from those in the cell body (Leonard et al.,
1987; Suszkiw et al., 1989; Charlton and Augustine, 1990).

Three distinct channel subtypes have been described in a dorsal root
ganglion (DRG) preparation (Nowycky et al., 1985a,b; Fox et al., 1987). The
channel subtypes, T, L, and N are distinguishable by their differential kinetics of
activation and inactivation, as well as their pharmacological sensitivity to agonists

and antagonists. A fourth type has been defined in mammalian Purkinje cells

9

(guinea pig cerebellum) and in presynaptic terminals of the squid giant synapse
(Uinas et al., 1989; Lin et al., 1990). This channel differs, both
electrophysiologically and pharmacologically, from the previously defined T-, L- and
N-type and is termed the P-type channel (for Purkinje)(lJinas et al., 1989; Lin et al.,
1990). The biophysical and pharmacological profiles of the different Caz+ channel
subtypes are summarized in Table 1.1.

T-type (“transient”) channels activate at membrane potentials more positive
than -70 mV, have very small unitary conductances, and inactivate quickly and
completely during a maintained depolarization. L-type (“long-lasting") channels
activate at postive test potentials ( > -10 mV), have a very large unitary
conductance, and show little inactivation during a 200 msec depolarization . N-type
(“neither T- nor L-") channels activate at positive test potentials ( > -20 mV), have
a smaller unitary conductance than the L channel and decay with time constants
ranging from 50-100 msec (Fox et al., 1987). Finally, P-type channels activate at
relatively positive test potentials (-30 to -20 mV) and decay with a time constant of
650 msec (Uinas et al., 1989).

Pharmacological differences exist between the Ca2+ channel subtypes also.
The dihydropyridine (DH P) class of organic Caz+ channel agonists and antagonists
such as Bay K 8644 and nitrendipine are presumed to interact with a site directly
on or associated with L-type Ca2+ channels (Nowycky et al., 1985a) whereas N-
and T- channels are unaffected by submicromolar concentrations of these drugs.

The phenylalkylamine, verapamil and the benzothiazepine, diltiazem also appear

10
Table 1.1

Pharmacological and Biophysical

Characteristics of Mammalian Ca2+ Channel Types

 

 

T N L P
Activation range >-70 >-20 >-10 -30--20
(mV)
Inactivation fast moderate slow slow
Kinetics r=m-50ms r=95ms r=>250ms r=650ms
Divalent Cation Ni+ > ca2+ ca2+ > Ni2+ co“ > Ni2+ ca“, < 100uM
sensitivity
DHPs _ _ 'I' -
ngTx _ + _ * _
PTX _ _ _ +
Single Channel
conductance (p8) 8 13 25 13

 

 

 

* Avian and amphibian L-type C“? are sensitive to Cng (Olivera et al., 1985-;
Marqueze et al., 1988).

Data for N-, L-, T-channels modified from Tsien et al., 1988. P-channel data
compiled from Uinas et al., 1989 and Lin et al., 1990.

1 1
to act as an L-type antagonist but have separate binding sites from that of the

DHPs (Miller, 1987; Glossmann and Striessnig, 1988). It is believed that three drug
receptor sites on this one channel are linked to one another via heterotrophic
allosteric coupling mechanisms (Hofmann et al.,1987; Kemp and Miller, 1987;
Glossman and Striessnig, 1988)

In mammalian neurons, N-type channels are blocked by o-conotoxin GVIA
(Cng), a peptide purified from the venom of the marine snail Conus geographus
(McCleskey et al., 1987; Kasai et al., 1987). Both N- and L-type channels in avian
and amphibian preparations are blocked by Cng at sites which appear to be
distinct from DHP binding sites (Olivera et al., 1985; Abe et al., 1986; Dooley et al.,
1987; Marqueze et al., 1988). Cng lacks potency on the invertebrate nervous
system, however (McCleskey et al., 1987). T) channels are DHP- and Cng-
insensitive but can be blocked by amiloride and octanol (T sien et al., 1988; 1991).

The T-., L-, and N-type of Ca2+ channels can be characterized by their
sensitivity to inorganic ions as well. Inorganic heavy metal cations such as zinc,
nickel, mercury, lead, copper, and cadmium block all three subtypes of channels
(Hagiwara and Byerly, 1981 ; Augustine et al., 1987). There is considerable variation
in the sensitivity of Ca2+ channels to these ions, however. For instance, in chick
DRG neurons, cadmium blocks N- and L-type channels with a Kd of 10 uM but
about ten times more cadmium is required to block T-type channels (Nowycky et
al., 1985a; Fox et al., 1987). Conversely, T-type channels are blocked more readily

by nickel than either N- or L-channels (T sien et al., 1988).

‘I 2
The peptide, o-Agatoxin and the polyamine, FI'X are potent Ca2+ channel

blockers derived from the venom of the funnel-web spider Agelenopsis aperta
(Adams et al., 1990; Uinas et al., 1989). Subtypes Ofm -Agatoxins are distinguished
as type I, type II, and type III. All except type III toxins block Ca2+ channels in most
insect motor nerve terminals and neuronal cell bodies at nanomolar concentrations
(Adams et al., 1990; Bindokas et al., 1991). On the other hand, types II and Ill but
not I toxins, block Cng-sensitive channels in chick brain synaptosomes (Adams
et al., 1990; Venema et al., 1992). Evidence exists that Agatoxins blocks DHP-
sensitive Ca2+ currents in rat sensory neurons (Scott et al., 1990; Mintz et al.,
1991) and in cardiac muscle as well (Liebowitz et al., 1990; Mintz et al., 1991).
P-type Ca2+ channels found in Purkinje cells and in the squid giant synapse
(Uinas et al., 1989; Mintz et al., 1992) are blocked by o-Aga-lVa and FI'X
respectively (Uinas et al., 1989; Mintz etal., 1992) but not by DHPs or Cng (Uinas
et al., 1989; Charlton and Augustine, 1990). FI'X also reduces a Ca“ current
observed following injection of total rat brain mRNA into Xenopus oocytes (Lin et
al., 1990). This current is insensitive to DHP- agonists, antagonists, and Cng as
well (Leonard et al., 1987). These results suggest that the channel type expressed
from rat brain mRNA in the oocyte may be similar to the P-type channel.
Despite considerable research, the channel type(s) involved in
neurotransmitter release are still not clearly defined. Again, this is due to technical
limitations of current techniques. However, results of some neurochemical

(Miller,1987) and electrophysiological (Atchison and O’Leary, 1987; Hirning et al.,

13
1988) studies suggest that neurotransmitter release is modulated by both the L-

and N- type channel. At rat sympathetic neurons and neuromuscular junctions, L-
type Ca2+ channels can have a modulatory role in neurotransmitter release (Perney
et al., 1986; Atchison and O’Leary, 1987) but do not contribute to nerve-evoked
release under normal conditions (Perney et al., 1986; Atchison, 1989). In contrast,
L-type channels have been implicated in release of substance P from rat sensory
neurons (Perney et al., 1986) and chick sensory neurons (Reno at al., 1987; Holz
et al., 1988), growth hormone and prolactin release from rat anterior pituitary cells
(Stojilkovic et al., 1988) and 3H-norepinephrine release from undifferentiated PC12
cells (Kongsamut and Miller, 1986). N-type channels have been reported to play
a dominant role in mediating evoked-release of norepinephrine from rat
sympathetic neurons (Perney et al., 1986; Himlng et al., 1988), differentiated PC12
cells (Kongsamut and Miller, 1986) and various areas of the central nervous
system (Dooley et al., 1987).

Data exist that suggest the P-type Ca2+ channel may also mediate Ca“
entry associated with transmitter release. As discussed previously, Ca2+ channels
at the squid giant synapse have pharmacological and functional properties unlike
those of T-, L-, or N-type Ca“ channels (Charlton and Augustine, 1990). They are
DHP— and Cng-insensitive but partially sensitive to FI'X. Therefore, Ca2+ channels
in the squid giant synapse appear similar to the P-type isolated from mammalian
Purkinje cells (Uinas et al., 1989; Charlton and Augustine, 1990) and likely mediate

release in these preparations. Likewise, Cng and DHPs were unable to block K‘-

1 4
evoked release of 3H-aspartate from rat hippocampal slices or release of ACh from

murine motor nerve terminals either (Anderson and Harvey, 1987). Mammalian
neuromuscular transmission was effectively abolished by FTX, however. FTX
blocked presynaptic Ca2+ currents, nerve-evoked neurotransmitter release and
muscle contraction from a murine nerve-muscle preparation (Uchitel at al., 1992).
This suggests that channels which are neither N- or L-type but possibly P- are
involved in neurotransmitter secretion from some mammalian nerve terminals. To
date, there is no evidence indicating a role for T-type channels in neurosecretion.

It should be noted that the characteristics attributed to Ca2+ channel
subtypes do not always fit neatly into one of the above catagories. Thus, the
original classification of Ca2+ channels into T-, L-, N-, and now P- subtypes is likely
an oversimplification. It is possible that functional heterogeneity among channel
subtypes as well as additional types of channels exist. This is supported by the
isolation from rat brain of four different types of cDNA that code for putative Ca2+
channels (Snutch et al., 1990). Characterization of the primary structure suggest
that several distinct voltage-gated Caz+ channel genes exist (Snutch at al., 1990;
Mori et al., 1991; Snutch er al., 1991). Additionally, species and tissue differences
in the type of Ca2+ channel involved in neurotransmitter release are evident (Brush

et al., 1987; Nayler, 1988; Maggi et al., 1990).

C. Disorders of Neuromuscular Transmission
As the above review was meant to indicate, there are many steps in the

process of synaptic transmission at the neuromuscular junction. Thus, there are

15

many ways in which the process can be disrupted. Disorders of neuromuscular
transmission vary in etiology, frequency and clinical presentation. All produce
muscle weakness through involvement of the neuromuscular junction, with a
preference for certain muscle groups. These disorders may affect the
neuromuscular junction presynaptically (the nerve terminal), postsynaptically (the
muscle endplate) or both pro-and post-synaptically. This dissertation will focus on
the presynaptic, neuromuscular disorder Lambert-Eaton Myasthenic Syndrome and
will attempt to delineate the stage at which normal functioning is altered by a

pathological process.

0. Lambert-Eaton Myasthenic Syndrome: General Description

1. Cllnlcal Features

The principal symptoms of Lambert-Eaton Myasthenic Syndrome (LEMS)
are weakness and easy fatigability of limb muscles, which temporarily improves in
response to voluntary exercise. Tendon reflexes are reduced or absent at rest
(Eaton and Lambert, 1957; Rocks of al., 1960; Lambert et al., 1961; M86 and
McDen'not, 1962; Elmqvist and Lambert, 1968). Patients frequently complain of dry
mouth, urinary hesitancy, constipation, sexual impotence, and impaired
accomodation of the eye (Rubenstein et al., 1979; Mamdani et al., 1985; Heath et
al., 1988; Khurana et al., 1988; Manji et al., 1990). These symptoms are indicative
of a cholinergic dysautonomia and therefore demonstrate that the defect is not

limited to striated muscle but can affect smooth muscle and glands as well. Drugs

16

that increase transmitter release, such as guanidine, 4-aminopyridine and 3,4-
diaminopyridine give patients with LEMS symptomatic relief (Oh and Kim, 1973;
Lundh, 1983; Newsom-Davis and Murray, 1984; McEnvoy, et al., 1989). Serious
toxic effects, however, limit their use. AChE inhibitors have minimal benefit (Eaton
and Lambert, 1957; O’Neill et al., 1988).

LEMS occurs in two groups of patients. In both groups, male Caucasians
(median age of 54 yrs) predominate (Elmqvist and Lambert, 1968; O’Neill et al.,
1988). In one group, all have some type of malignancy, the most common (60-
7096) being small cell carcinoma of the lung (SCC)(Eaton and Lambert, 1957;
Lambert et al., 1961; Wise and MacDermot, 1962; Elmqvist and Lambert, 1968;
O’Neill et al., 1988). Symptoms of LEMS can precede the diagnosis of $00 by 5
months to 4 years (median 10 months) (O’Neill et al., 1988; Anderson et al.,1990)
and sometimes improve when the tumor is either treated or removed (Ingram et
al., 1985; Oh, 1989; Chalk et al., 1990). The second group of patients have no
detectable malignancy but present with identical clinical signs.

LEMS is generally accepted as being a paraneoplastic disease, a nervous
system dysfunction in patients with cancer that is not caused by metastases
(Vincent et al., 1989; Posner, 1991). Because the disease occurs in the absence
of tumor as well as in its presence, it follows that the tumor favors the development
of the condition but cannot in itself be the only cause (Lambert and Rocks, 1965;

Chad and Recht, 1991).

17
2. Electrophyslology

a. Cllnlcal

Electromyography is diagnostic and conventionally consists of the recording
of muscle action potentials generated following surface stimulation of nerves from
living patients (Harvey and Masland, 1941; Eaton and Lambert, 1957). An
electromyographic feature that characterizes LEMS is a significant reduction in the
amplitude of the compound muscle action potential (CMAP). Low frequencies of
nerve stimulation (1 per sec, 1H2) result in a progressive decrease in the amplitude
of the CMAP while a marked increase in the response is noted when the nerve is
stimulated at higher frequencies (10 to 50 Hz) (figure 1.2). An increase in
response amplitude can also be demonstrated after brief periods of voluntary
contraction of the muscle (Eaton and Lambert, 1957; Elmqvist and Lambert, 1968;
Swift, 1981; Jablecki, 1984).

Nerve conduction velocities in LEMS patients are within the range of normal
values and remain normal at repetitive rates of stimulation (Eaton and Lambert,
1957; McOuillen and Johns, 1967; Elmqvist and Lambert, 1968). This suggests the
defect in transmission is not due to failure of action potential generation or
propagation.

Direct stimulation of muscle fibers also reveals normal conduction velocities.

Repetitive stimulation at increasing frequencies do not produce the typical changes

18

CONTROL LEMS
Single Shock- CMAP

Stimulation /\
(NERVE) _ A

Low frequency
repetitive stimuli I I i ii
I II I

 

 

 

 

 

 

 

High frequency
repetitive stimuli l ..... I . . .. .
(> 10 Hz) m | | I

 

 

 

 

 

 

 

 

figure 1.2. Electromyographic Characteristics of LEMS

This is a schematic representation of the EMG characteristics of LEMS. As you can
see, there is a significant reduction in the CMAP. At low frequencies of stimulation
there is a progressive decrease in the CMAP while a marked increase in the
response is noted at higher frequencies of stimulation.

19

seen with indirect stimulation of the muscle via the nerve (McOuillen and Johns,
1967; Elmqvist and Lambert, 1968). Thus, the reduction in CMAP amplitude at low
frequencies of stimulation and the increase in amplitude at high frequencies of
stimulation do not arise from the muscle itself.

b. Experimental

To investigate the electrophysiological basis for the electromyographical
observations, intracellular microelectrode recordings were made from biopsied
intercostal muscle taken from patients with LEMS. The number of ACh-containing
quanta released per nerve impulse (quantal content) was markedly reduced
(Elmqvist and Lambert, 1968; Lambert and Elmqvist, 1971; Cull Candy et al.,
1980). This suggests that in vivo the EPP would fail to achieve the threshold
depolarization required for generating an action potential in the muscle membrane.
As was seen electromyographically for the CMAP, repetitive stimulation of the
nerve at 30 Hz increased quantal content.

Spontaneous release of neurotransmitter was also recorded. MEPP
amplitudes and frequencies were similar to the MEPPs recorded in normal tissue
indicating resting quantal release of ACh was normal. (Elmqvist and Lambert, 1968;
Lambert and Elmqvist, 1971). Resting, non-quantal release of ACh was decreased,

however (Lang et al., 1984; Molenaar et al., 1987).

3. Blochemlstry

Additionally, the sensitivity of LEMS muscle to ACh (Lambert and Elmqvist,
1971; Lindstrom and Lambert, 1978; Cull Candy et al., 1980) as well as the
number of ACh receptors present are normal (Lindstrom and Lambert, 1978). Also,
the reduced quantal content is not due to a decreased synthesis or storage of
neurotransmitter because the ACh and ChAT content of the nerve terminal are
normal (Molenaar et al., 1982; Molenaar et al., 1987). Thus, these potential factors
which could have contributed to the neuromuscular weakness found in LEMS
patients have been ruled out. The evidence so far strongly suggests a selective

effect on the mechanism of ACh release.

4. Morphology

a. Morphometric analysls

Electron microscopic, morphometric analysis of biopsied endplate regions
typically consists of determining the nerve terminal area, the number of synaptic
vesicles per unit nerve terminal area, the mean synaptic vesicle diameter, the
presynaptic membrane length and the postsynaptic membrane length per nerve
terminal (Loud, 1962; Engel, 1970; Engel and Santa, 1971; Hesselmans et al.,
1992). Engel and Santa (1971) reported that both methylene blue-stained
intramuscular nerves and cholinesterase-stained endplates showed no definite
abnormalities. Electron microscopy demonstrated that the mean nerve terminal

area and the mean synaptic vesicle count per unit area were not significantly

21

different from control values. When the mean synaptic vesicle diameter was
calculated for each biopsy, control and LEMS means did not differ significantly
either (Engel and Santa, 1971).

One abnormality noted, however, was an overdevelopment of the
postsynaptic region. Analysis showed a significant increase in the mean area of
clefts and folds per nerve terminal and in the mean postsynaptic to presynaptic
membrane length ratio (Engel and Santa, 1971). Multiplication of endplate areas
is a well known phenomenon in disorders of neuromuscular transmission such as
botulinum toxin (BoTx) intoxication (Duchen, 1971), myasthenia gravis (MG; in
some cases) (Coérs and Tellerman-Toppet, 1976) and congenital myasthenia with
decreases in ACh receptors (Vincent et al., 1981). The increase in contact between
muscle and nerve is likely to be compensatory to the deficiency in ACh release
(LEMS and BoTx intoxication) or ACh receptors (myasthenia gravis).

In contrast, Hesselmans and colleagues (1992) report that both postsynaptic
areas and postsynaptic membrane lengths of NMJs are significantly less in LEMS
than in controls (Hesselmans et al., 1992). This has also been observed in patients
with MG and in the chronic phase of MG’s experimental model as well (Engel and
Santa, 1971; Engel and Tsujihata, 1976). A factor which may have influenced the
differences between Hesselmans’ and Engel’s studies is the duration of the
disease before muscle biopsy was performed. In Hesselmans” study, the onset of
symptoms was at most 1.5 years before biopsy. In the study by Engel and Santa

(1971) duration of symptoms ranged from 3 to 11 years. Since newly-created

22
NMJs have small postsynaptic regions which in time increase, perhaps only short-

terrn effects of ACh release blockade were studied by Hesselmans et al., (1992).

b. Ultrastructural analysls

Ultrastructural changes in the motor nerve terminal associated with LEMS
were first reported by Fukunaga and coworkers (Fukunaga et al., 1982; 1983).
Measurements of the density of the active zones and large intramembraneous
active zone particles (AZPs) were made in freeze-fracture preparations of biopsied
human intercostal muscle. Their results indicated that both the active zones and
the AZPs were strikingly scarce and disorganized (Fukunaga et al., 1982).
Prejunctional membranes taken from patients with LEMS had fewer active
zones/u m2; fewer number of active zone particles /u m2; less particles/active zone;
and a 3.6 fold increase in the number of large particles aggregated into
clusters/u m2. All other membrane particles studied appeared normal. Since the
average decrease in the AZPs per unit of membrane area was comparable to the
average decrease in quantal content of the EPP in LEMS, Fukunaga and
colleagues concluded that a reduction in AZPs in LEMS accounted for the
diminished probability of ACh release. (Fukanaga er al., 1983). Active zones are
topographically related to the sites of synaptic vesicle exocytosis, and thus are
believed to be the sites at which neurotransmitter release occurs (Heuser et al.,
1974; 1979). Thus, electron micrographic findings in LEMS are consistent with a
presynaptic disorder of neuromuscular transmission as postulated from the results

of microelectrode studies.

23
E. Lambert-5am Myasthenic Syndrome: Autoimmunity
1. Cllnlcal and experimental evidence

Compelling evidence has indicated that the etiology and pathogenesis of
LEMS is autoimmune. Clinically, beneficial effects of plasma exchange (Newsom-
Davis and Murray, 1984) and immunosuppressive drugs (Streib and Rothner, 1981;
Newsom-Davis and Murray, 1984) have been demonstrated. Furthermore, LEMS
has frequently been shown to accompany one or more autoimmune disorders
(Gutmann et al., 1972; Lennon et al., 1982; Paris at al., 1990).

The autoimmune basis and involvement of a humoral factor was
demonstrated with certainty when the electrophysiological features of the disorder
were transferred passively to mice by injection of humoral substances from
affected patients (Denys et al., 1972; Lang et al., 1981; Newsom-Davis et al., 1982;
Lang et al., 1983a; Kim, 1985; Kim, 1986; Lambert and Lennon, 1988). Plasma
(Newsom-Davis et al., 1982; Kim, 1985), crude immunoglobulin fractions (Lang et
al., 1983a; Kim, 1986; Lambert and Lennon, 1988) and purified lgG (Denys et al.,
1972; Kim, 1986) are all effective in transferring the disease with the production
of characteristic in vitro electrOphysiologic abnormalities. Mice treated with lgG
from patients exhibiting LEMS also had a significantly lower number of active zones
and AZPs (Fukanaga et al., 1983; Nagel et al., 1988). Furthermore,
immunolocalization of LEMS lgG was demonstrated in the region of presynaptic
active zones of mice (Fukuoka et al., 1987).

It has been proposed that the AZPs correspond to voltage-sensitive calcium

24
channels of the presynaptic membrane (Pumplin et al., 1981 ; Robitaille et al., 1990;

Tarelli et al., 1991; Cohen et al., 1991). Indirect evidence to support this
assumption is as follows: (1) presynaptic voltage clamp studies in the squid giant
synapse demonstrated a very short latency between Ca2+ entry and postsynaptic
response, suggesting that Ca2+ channels are located at the site of vesicle
accumulation and neurotransmitter release (Uinas et al., 1976); (2) synaptic vesicle
exocytosis captured by quick freezing occurs near the AZPs (Heuser et al., 1979);
(3) AZPs continue to specify sites of exocytosis even when dispersed by a low
Cay-medium (Ceccarelli et al., 1979); and (4) in the squid synapse, the maximum
presynaptic Ca2+ current appears to be related to the number of AZPs (Pumplin
et al., 1981).

Direct evidence for this contention is the presynaptic localization of Cng-
sensitive Ca2+ channels at the frog neuromuscular junction (Robitaille et al., 1990;
Tarelli et al., 1991; Cohen et al., 1991). Cng- immunoreactivity revealed a pattern
composed of regularly spaced bars running parallel to each other and
perpendicularly to the major axis of the nerve terminal. This array corresponds to
the ordered arrangement of active zones in this species. A correspondence was
also found between the mean number of active zones and fluorescent bars present
within the same length of axolemma (T arelli et al., 1991). If AZPs correspond to
voltage—dependent Ca2+ channels, the reduced neurotransmitter release by nerve
impulses in LEMS patients could be related to reduced entry of calcium into the

nerve terminal.

25
2. The Antigenic Target

a. Calclum channel hypothesls

Autoantibodies that interfere with Ca2*-dependent neurotransmitter release
may do so by three distinct mechanisms. The LEMS autoantibody could act on (1)
Ca2+ entry through voltage-gated Ca2+ channels, (2) intraterminal Ca2+ metabolism
or (3) the exocytotic release process. Results of several studies indicate that LEMS
autoantibodies interact with voltage-dependent Ca2+ channels (Roberts et al.,
1985; De Azipurua et al., 1988; Kim and Neher, 1988; Peers er al., 1990). Other
studies have rendered the other possibilities unlikely (Lang et al., 1983; Kim, 1985;
1986; 1987; Wray et al., 1987; Lambert and Lennon, 1988).

Except for the well-understood requirement of Ca2+ ions for prejunctional
ACh release (Katz and Miledi, 1965; Dodge and Rahamimoff, 1969), specific
molecular events leading to exocytosis are largely unknown. Because of this, it is
difficult, if not impossible, to evaluate a possible dysfunction in the release
mechanism itself. However, indirect effects on release mechanisms can be studied
with pharmacological manipulation. Nerve evoked release of neurotransmitter from
passively transferred mice was studied over a range of [Ca2+], concentrations
(Kim, 1987; Wray et al., 1987). NMJs from LEMS lgG-treated mice showed the
characteristic decrease in quantal content at both low and physiological Caz+
concentrations when compared to control lgG-treated mice. However, at
unphysiologically high Ca2+ concentrations the number of quanta released by

news stimulation in LEMS-treated NMJs reached the same plateau value as did

26
control-treated NMJs. This suggests that the number of available quanta was not

different in passively transferred LEMS and the exocytotic release process itself
was not impaired (Kim, 1987; Wray et al., 1987).

The second possibility that intracellular mechanisms governing Ca2+
homeostasis may operate abnormally in LEMS is also not supported. An
abnormality of this type would result in a decline in both nerve~evoked and
spontaneous release of neurotransmitter. In human LEMS as well as passively
transferred murine LEMS, no appreciable reduction in either MEPP frequency or
amplitude is seen (Lang et al., 1983; Kim, 1985; 1986; Lennon and Lambert, 1988).

With the exclusion of the latter two possibilities, and with the evidence
presented thus far, the “calcium channel hypothesis” appears to be the most
plausible explanation for the pathophysiology of LEMS. Numerous studies have
helped to clarify the neuromuscular deficit associated with LEMS.

Incubation of rat cortical synaptosomes with whole serum or high molecular
weight serum fractions (> 10,000 daltons) from patients with LEMS, for as little as
three minutes, decreased the amount of ijACh released during depolarization
(Meyer et al., 1986). Although this study did not measure Ca2+ channel activity,
it was important in that it indicated that acute exposure of synaptosomes to serum
or serum fractions taken from LEMS patients could elicit responses qualitatively
similar to those obtained in chronic passive transfer experiments. Also, antigens
presumed to reside on human and rodent nerve terminals appear to share a

common epitope recognized by the LEMS autoantibody. Likewise, Login et al.,

27
(1987) exposed rat anterior pituitary cells to immunoglobulin isolated from a LEMS
patient, and found that release of growth hormone and prolactin was significantly
reduced by LEMS lgG compared with control lgG (Login et al., 1987). Again, no
assessment of ion channel activity was made but secretion from these cells is
Ca2+-dependent (Stojilkovic et al., 1988).

Roberts et al., (1985) cultured human small cell carcinoma cells (800) in the
presence and absence of LEMS IgG for seven days. Following the incubation
period, 45Ca2+ uptake induced by 30 sec of K+ depolarization was measured. The
cells treated with LEMS lgG exhibited a significant decrease in stimulated uptake
compared with untreated cells or with cells grown in control lgG (Roberts et al.,
1985). These studies were confirmed by De Azipurua et al., (1988a) and Lang et
al., (1989b) who in addition demonstrated that PC12 cells were similarly affected
(Lang et al., 1989a; 1989b).

Kim and Neher (1988) applied lgG from LEMS patients to isolated bovine
chromaffin cells from the adrenal medulla. Electrophysiological recordings
indicated a 40% reduction in the current carried through calcium channels.
Sustained inward Caz+ currents in undifferentiated mouse neuroblastoma x rat
glioma hybrid cells (NG108-15) were reduced by LEMS lgG also (Peers et al.,
1987). Taken altogether, these various results provide strong evidence for
membrane calcium channels as a target site for the autoantibody in LEMS.

However, recent evidence suggests that synaptotagmin may also be a

LEMS antigen (Leveque et al., 1992). A 58 kDa antigen, identified as

28
synaptotagmin by a partial amino acid sequence, was detected by probing
Western blots of partially purified calcium channels from rat brain with LEMS
plasma and lgG. Synaptotagmin is a synaptic vesicle protein believed to be
involved in docking synaptic vesicles to the plasma membrane at the active zone
(Augustine, 1987; Leveque et al., 1992; Brose et al., 1992). As this protein was
recognized in what was thought to be partially purified Ca2+ channels, Leveque
and colleaques (1992) believe that synaptotagmin not only binds to the plasma

a” channel as well. As such,

membrane but is tightly associated with the C
alterations in synaptotagmin function could indirectly alter Ca2+ channel activity
(Leveque et al., 1992).

However, several deficiencies in this study are evident. first, synaptotagmin
is a putative transmembrane protein with the N-terminus oriented toward the
interior of synaptic vesicles and the C-terminus at least partially inserted into the
lipid bilayer (Perin et al., 1990). In order for the circulating antibody to bind, one
must presume that this protein is exposed at the surface of the nerve terminal
following exocytosis. Secondly, synaptotagmin and calcium channels may have a
common epitope which can be recognized via SDS-PAGE followed by Western blot
analysis but which is normally concealed. finally, the possibility that LEMS sera
contains antibodies that only recognize calcium channel associated polypeptides

in their undenatured form can not be excluded.

 

 

 

29
b. Calclum channel subtype affected by LEMS lgG

The finding that LEMS lgG inhibits Caz+ entry in motor nerve terminals
(Lang et al., 1987), 806 (Roberts et al., 1985; De Azipurua et al., 1988a; Lang et
al., 1989a; 1989b), chromaffin (Kim and Neher, 1988), anterior pituitary (Login et
al., 1987), NG108-15 (Peers et al., 1987) and PC12 cell lines (Lang et al., 1989a;
1989b) indicates that the antibodies recognize an epitope that is present on
voltage-dependent Ca2+ channels from several sources. The use of these cell
lines have led to assumptions concerning the subtype of Ca“ channel putatively
affected by this antibody.

LEMS lgG was shown to reduce Ca2+ currents with characteristics of L-type
channels in NG108-15 and chromaffin cells (Kim and Neher, 1988; Peers et al.,
1987). Secretion from anterior pituitary and undifferentiated PC12 cells is believed
to be mediated by L-type channels also (Kongsamut and Miller, 1986; Login et al.,
1987; Stoljilkovic et al., 1989). Data from studies utilizing 800 cell lines (Roberts
et al., 1985; De Azipurua et al., 1988a; 1988b) suggest an effect on both L- and
N-type channels, respectively. Finally, LEMS autoantibodies have been shown to
immunoprecipitate 1’i‘I-Cng labeled voltage-gated Ca2+ channels (N-type) from a
human neuronal cell line (Sher et al., 1989).

The results from these studies would suggest that the L- and N- type Ca2+
channel are putative antigenic targets for the LEMS autoantibody. Although these
studies have proven useful in determining the antigenic target in the cell some, the

identity of Ca2+ channels in the nerve terminal remains unclear. What is clear, is

30

the types of Ca2+ channels at mammalian nerve terminals do not appear to be
either N- or L-type and are possibly of the P-type (Anderson and Harvey, 1987;

Sano et al., 1987; Charlton and Augustine, 1990; Uchitel et al., 1992).

3. Pathophysiologlcal Mechanisms

The exact mechanism by which LEMS lgG exerts its pathological effects
has not been fully established. In bovine adrenal chromaffin cells treated with
LEMS lgG, kinetic and single channel studies indicate that the pathogenic
autoantibodies neither modify the kinetics of Caz+ channel activation and
inactivation nor single channel open time nor conductance (Kim and Neher, 1988).
Also, compared with control serum, pooled LEMS sera from 5 patients reduced the
binding of the Ca2+ channel antagonist 125M -conotoxin (Cng) by 22% following
overnight incubation with the human neuronal cell line, lMR-32 (Sher et al., 1989).
Thus, it appears as if the antibody-induced dysfunction of Ca“ channels occurs
in an “all or none“ fashion. Each channel is either subject to a complete functional
block or no blockade at all. Therefore, the pathogenic action of the LEMS
autoantibody is characterized by a reduction in the number of functlonal Ca2+
channels. Functional alterations could be achieved by several mechanisms and
may act in concert.

a. Dlrect Inhlbltlon

Bound, pathogenic autoantibodies could reduce the number of functional

Ca2+ channels by direct inhibition. This does not appear to be the case in some

31

systems, however. The application of LEMS plasma to an in vitro organ culture for
up to three hr had no effect on nerve-evoked quantal release (Kim et al., 1988).
When mice were studied after various periods of IgG treatment, the t“ for the
reduction in quantal content was at least 4-6 hr (Lambert and Lennon, 1988), or
longer (Prior et al., 1985). However, It just may be that more than three hr of
incubation is needed in order to gain access to, and inhibit the function of

a2+ channels in some systems. For example, only a two hr

prejunctional C
incubation of chromaffin cells with LEMS antibodies resulted in significant
reductions in whole cell Ca2+ currents (Kim and Neher, 1988) while incubation of
synaptosomes with LEMS sera for as little as three min decreased the amount of
[3H]-ACh released during depolarization (Meyer et al., 1986).

b. Antlgenlc modulation

In view of the morphologic alterations depicting the scarcity of AZPs, it
would appear that the LEMS autoantibody eventually destroys or eliminates Ca2+
channels. Antibody-induced antigenic modulation could result in loss of channels
by initiating accelerated channel degradation. Ca2+ channel crosslinkage by LEMS
lgG could be an initial process required for final channel elimination. This process
has been observed by Nagel et al., (1988) and Lang et al., (1987). These
investigators observed that divalent F(ab)’2 (antigen-binding) fragments of LEMS
autoantibodies which retain the antigen binding capacity of normal lgG effectively

transferred the disease to mice, whereas monovalent Fab fragments failed to do

so. This phenomenon was verified by measurements of AZP density (Nagel et al.,

32
1988) and quantal content of the EPP (Lang et al., 1987). In the terminals treated

with LEMS IgG or LEMS F(ab)’2, quantal content was reduced; monovalent LEMS
Fab had no significant effect (Lang et al., 1987). There was also a significant
decrease in the density of active zones and AZPs. In the AZPs remaining,
interparticle distance was reduced (Nagel et al., 1988). This was considered to be
an early event associated with disorganization and eventual destruction of active
zones and AZPs (Fukuoka et al., 1987).

c. Complement-mediated destruction

Alternatively, loss of channels could be due to complement-mediated
channel destruction. Activation of complement plays a central role in mediating
postsynaptic membrane damage in animal models of the autoimmune
neuromuscular disorder, myasthenia gravis (MG) as well as in the human disease
state (Engel et al., 1977; Sahashi et al., 1978). It seems possible that complement
activation could be important in LEMS as well.

Complement is a complex series of proteins which form a triggered enzyme
system in the plasma. Figure 1.3 provides a summary of the complement
components and their interactions. Some of the complement components are
designated by the letter 'C" followed by a number. The complement system may
be activated by two major pathways (classical and alternative) that proceed in a
sequential fashion. Activation of individual components of complement depends
upon activation of a prior component or components in the sequence. Fragments

of native complement components are either released into the fluid phase or bind

 

(I)

Classical Pathway
Activation Signal

j _ C4
cr——> c1 / >1 iP'"
\ C2 products

1

CAT); (03 convertasc)

1

Alternative 3—9 C3-> 03a
Pathway D

 

 

 

Feedback Loop Amplification IP— C3b—-) C4b2e3b

I

CS -> 05a
Membrane Attack
Complex 9 C5b678 9n

 

figure 1.3. Complement.

Scheme of the reaction mechanisms following activation of the complement
cascade. For detailed explanation please see text (pp. 32-35).

34

to other complement components or suitable surfaces such as immune complexes
and cell membranes (Law and Levine, 1977; Capel et al., 1978; Takata et al.,
1984).

The classical pathway of complement activation is typically initiated by the
union of antibody and antigen. The antigen can be either soluble or a constituent
of a cellular or extracellular surface. Whether an antigen-antibody reaction
activates complement depends upon the antibody molecule. Most immunoglobulins
of the lgM or IgG (with the exception of lgG4) subclass activate the classical
pathway (Loos, 1987).

Antigen binds to Fab (antigen-binding) regions of immunoglobulins.
Together, these molecules undergo a conformational change that permits binding
and activation of the first component of complement, C1. Activated C1 cleaves C4
and C2, by limited proteolytic reactions. Cleavage of C4 results in the formation
of a small peptide, 04a and a large fragment, C4b, which can bind either to the
immune (antigen-antibody) complex or to the surface of the antigen. C2 is cleaved
into a small peptide, 02b and a large fragment, 02a, which binds to C4b. This
complex, C4b2a, (03 convertase) possesses new enzymatic activity that is capable
of cleaving the third component of complement, 03.

Proteolytlc cleavage of 03 yields a small fragment, 03a and a larger

fl'Elgl'nent, 03b, which possesses the ability to bind to immune complexes, as well
as to a wide variety of surfaces (Law and Levine, 1977; Capel et al., 1978; Takata

9‘ 31- . 1984). Once bound, 03b imparts a new specificity to C4b2a, enabling this

35
enzyme to function as a CS convertase. Bound 03b is also recognized by

phagocytic cells (Hostetter etal., 1984), and functions as an important intermediate
in the activation of the alternative pathway (Goetze, 1987).

Cleavage of CS by CS convertases, constitutes the final proteolytic reaction
in the classical pathway. A small fragment, 05a, is released into the fluid phase,
but the larger fragment, CSb, binds to immune complexes and to a variety of
surfaces. Whether bound or in the fluid phase, CSb forms a stable macromolecular
complex with the complement components C6, C7, and C8. This complex induces
orientation between two or more molecules of the terminal component 09. This
leads to unfolding of 09 and insertion into the lipid bilayer and polymerization to
a ”membrane attack complex" (MAC). MACs form transmembrane channels that
permit bidirectional flow of ions and ultimately macromolecules producing cellular
injury and/or lysis. (Porter and Reid, 1978; Bhakdl and Tranum-Jensen, 1978).

The alternative pathway (APC) is a non-specific, natural defense system
against pathogenic microorganisms (Joiner et al., 1984; Goetz et al., 1987). In
addition, it is an amplification system for increasing 03b depositon and 03/05
convertase activity after complement activation through the classical pathway
(Joiner et al., 1984). Once activated, the biological sequence of events of the APC
does not differ from that of the classical pathway.

Attempts to clarify the participation of complement in the pathogenesis of
LEMS have yielded negative results. Passive transfer was attempted in mice which

had received cobra venom factor (CVF) (Lennon and Lambert, 1988) and in mice

 

 

 

36
congenitally deficient in CS (Prior et al., 1985; Lennon and Lambert, 1988). CVF

is a potent complement activator, which interacts with Factor B of the alternative
pathway to form a stable C3 convertase (Grief: and Mi'iller-Eberhard, 1971). This
leads to the consumption of plasma 03 and the later components of the
complement cascade. The defect in neuromuscular transmission in mice receiving
LEMS lgG was evident in mice of normal and deficient complement status (Prior
et al., 1985; Lambert and Lennon, 1988). These results suggest that the effects of
LEMS lgG may not be due to complement activation. This unaltered response in
the absence of complement activity does not, however, mean that it is not involved
when present. Most of the effects of lgG or the alternative pathway activators on

cells are greatly enhanced in the presence of complement (Rother, 1987).

F. The Immune Response

As stated previously, the pathogenic alterations in neuromuscular
transmission in LEMS are caused by an autoantibody circulating in the plasma of
patients with the disorder. Alterations in the regulation of the normal immune
response could lead to recognition of sell-components as foreign inducing
autoimmunity. The next section will discuss the normal immune response with
emphasis on mechanisms which disrupt regulatory pathways leading to faulty self-
non-self recognition. Determinations of the mechanisms of autoimmunity in concert
with determination of the antigenic target could have important therapeutic

implications for LEMS patients or any patient with autoimmune disease.

37

1. Cellular Interactions In the Immune system

The immune system can be considered simplistically the body's defense
against foreign invaders. An intricate series of cellular events must occur in order
for the body to mount an effective immune response. Typically, an immune
response is generated by the coordinated interaction of lymphocytes and antigen-
presenting cells (Fig 1.4)(see Miller et al., 1971). All of these cells derive from the
bone marrow, from a pluripotent stem cell (Roit, 1988a).

One class of lymphocytes consists of B lymphocytes, so called because
maturation of these cells occurs in the bone marrow. B lymphocytes are the only
cells capable of producing antibodies (Nossal et al., 1968; Jacobson et al., 1970).

The second major class of lymphocytes are T lymphocytes, which arise in
the bone marrow and then migrate to and mature in the thymus. It is in the thymus
that the T lymphocyte undergoes a selection event in which 'leaming" takes place
for recognition of self-encoding molecules known as the major histocompatibility
complex (MHC). T lymphocytes are further subdivided into functionally distinct
populations, the helper (T h) cell and the suppressor/cytotoxic T (T .) cell. T" cells
promote immune responses to antigen whereas T, cells function to suppress such
a response (Schwartz, 1987; Roit, 19883).

Cells responsible for antigen presentation are usually of the phagocytic
monocyte-macrophage lineage (Adler et al., 1966; Unanue and Askonas, 1968;
Unanue and Cerottini, 1989). Other antigen-presenting cells include the

Langerhans cells in the skin, dendritic cells of the lymph nodes and spleen, and

* * antigen

macrophage

  

   

.cytokines ACtivated
*— T, cell
Clonal Expansion lCYIOKIDOS
of T, cells
B cell
Plasma Cell Clone\’Y Y antibodies

YYY

Figure 1.4. Immune Response

The immune response is initiated when antigen is processed and presented to a
Th lymphocyte by a macrOphage. The Th cell becomes activated and releases a
variety of soluble factors (cytokines) which act on both B and T lymphocytes to
augment the immune response. Once triggered by antigen and Th cell cytokines,
B cells proliferate forming a large clone of plasma cells which make antibody.

39
the interdigitating follicular cells of the thymus (Dalton and Bennett, 1992).

Normally, the interaction between the Th lymphocyte and the antigen-
presenting cell is the first event in an immune response. Helper T cells have an
unusual specificity for antigens. They recognize only peptide antigens that are
attached to MHC molecules expressed on the surfaces of antigen-presenting cells
(Lamb and Feldman, 1984). The T cell does not recognize native antigen, however.
Instead, the antigen-presenting cell goes through a series of events known as
antigen processing, whereby native antigens are internalized, partially degraded
and immunogenic fragments are returned to the cell surface (Unanue and
Askonas, 1968; Unanue and Cerottini, 1989). In addition, the antigen-presenting
cell releases protein hormones called cytokines to co-stimulate the Th cell. In
response to stimulation, Th cells also secrete cytokines, whose function is to
promote the proliferation and differentiation of the T cells themselves as well as
other cells including 8 cells (Schwartz, 1987).

Once triggered by antigen and Th cell cytokines, B cells undergo successive
waves of proliferation to build up a large clone of plasma cells which make
antibody that is antigen-specific (Jacobson et al., 1970). In addition, 8 cells are
able to recognize some antigens without any T cell help. Certain antigens, such as
bacterial lipopolysaccharides, have the ability to activate a proportion of the B cell
population by their inherent mitogenic potential (Roit, 1988a). B cells have the
ability to produce 2:10a different antibodies (Max, 1989). However, each B cell can

produce only a single heavy and a single light chain, which interact to make an

 

40
antibody with a unique specificity (T onegawa, 1983).

2. Self- non-self recognltlon

The key to effective immunity is the elimination of pathogens without
concomitant host injury. The finely tuned interaction among lymphoid cells,
cytokine molecules and foreign influences determines immune reactivity. This finely
tuned modulation of the immune system is crucial since excessive activation can
lead to allergies and autoimmune diseases and oversuppression may promote
infectious diseases and cancer (Homo-Delarche et al., 1991). Curiously enough,
autoreactive B and T cells are present normally (Dighiero et al., 1987; Rosenkrantz
et al., 1987). The body has homeostatic mechanisms to prevent them from being
triggered under normal circumstances. Immunologic tolerance to self in T and B
lymphocytes can be induced by two basic mechanisms: (1) deletion, through cell
death; and (2) functional inactivation (energy) without physical deletion.

a. Clonal Deletion

The clonal deletion theory of self-tolerance was first proposed by Bumet
(1959) to occur during fetal or early postnatal life. Autoreactive clones of
lymphocytes (both B and T) were believed to be eliminated during the stage in
development immediately after the lymphocyte expresses its antigen receptor and
before it becomes competent to respond positively to antigenic stimulation
(Lederberg, 1959). In this way autoreactive clones are filtered out.

The original theory was modified by Allison et al., (1971) and Weigle (1971)

41
who independently suggested that self-tolerance depends upon intrathymic

elimination of self-reactive T lymphocytes while B cell responsiveness to self
antigens remained intact. In this way, autoreactive T cells are unable to collaborate
with B cells to generate autoantibodies. The models put forth to explain this were
based on receptor affinity (for review see Sprent et al., 1988). T cells in the thymus
which bind with high affinity to MHC molecules on the surface of intrathymic
macrophage cells are eliminated. Those capable of reacting with MHC plus antigen
but which only bind with poor affinity to self molecules alone, survive.

The idea of clonal deletion has been now been verified for both T and B
cells using transgenic mice (Kisielow et al., 1988; She et al., 1988; Nemazee and
Bi'Jirki, 1989). However, clonal deletion may play a relatively minor role in B cell
tolerance (Nemazee and Bt'irki, 1989; Goodnow et al., 1990). Because,
autoreactive T and B cells can be recovered from normal humans and animals
(Dighiero et al., 1987; Rosenkrantz et al., 1987), the process of clonal deletion
must be incomplete. Other mechanisms must act to limit the potentially
catastrophic immune response against self.

b. Clonal Anergy

I. Induced by Suppression

Removal of anti-self effector lymphocytes may be attributable to elimination
in some cases but may be due to the presence of peripheral blood regulatory
lymphocytes in others. Self-tolerance based on the induction of T, cells has been

postulated by Rose (1988). In this model, cells are functionally inactivated without

42
being physically deleted. Self-antigens are considered normally to induce mostly

suppressor (rather than helper) T cells with the ability of suppressing that particular
immune response (for review see Cruse and Lewis, 1992). This model allows for
the existence of a reservoir of cells potentially capable of autoimmune behavior.

Clonal energy has been demonstrated in the B cell repertoire as well
(Weigle, 1973; Metcalf and Klinman, 1976; Nossal, 1989; Goodnow et al., 1990).
Studies in tissue culture demonstrate that B lymphocytes exposed to antigens in
the absence of T" lymphocytes can be induced into a state of nonresponsiveness.
Subsequent challenge of the B cells with antigen in the presence of Th cells
produces no response (Metcalf and Klinman, 1976).

ii. Induced by Signal Failure

Bretscher and Cohn (1970) suggested the need for two signals to activate
a lymphocyte: one, antigen receptor occupation; the other, a co-stimulatory signal
delivered by a second cell. In the case of B lymphocytes, the second signal comes
from activated Th cells. In the case of Th lymphocytes, the second signal comes
from antigen-presenting cells. Tolerance induction in this model was postulated to
be delivery of signal one in the absence of signal two (Bretscher and Cohn, 1970).

Inactivation of antigen-specific lymphocytes as a consequence of abortive
antigen presentation may indeed play a role in self-tolerance. In healthy organisms,
self-antigens may be presented without accessory signals (cytokines, MHC,
adhesion molecules) necessary to induce clonal proliferation and differentiation.

This would leave the T cell unchanged and possibly unable to respond later to an

43
adequate stimulus (Bretscher and Cohn, 1970; Lamb and Feldman, 1984; Parker

and Eynon, 1991). In this way, immunocompetent, autoreactive T cells could be
inactivated. Similierly, B lymphocytes whose receptors are cross-linked by certain
multivalent antigens alone are rendered non-responsive, whereas simultaneous
costimulatory signals by T cell cytokines lead to activation and differentiation

(Nossal, 1983).

3. Breakdown of self-tolerance

The state of self-tolerance may be an amalgam of many processes
including (but not limited to) clonal deletion and specific suppressor cell networks.
It lymphocyte/monocyte action is not limited by protective regulatory pathways
then these cells will either directly or indirectly attack innocent bystander cells,
disrupting tissue structure and leading to autoimmune disease. Direct attacks
include the release of biologically active agents (enzymes, peroxides, cytotoxic
cytokines etc.) from monocytes or cytotoxic lymphokines from T lymphocytes.
Indirect attacks may be mediated through antibodies released from B cells and/or
activation of the complement system (Grossman et al., 1991).

Autoimmune diseases are chronic and in some cases severe. They are
clearly multifactorial in origin with genetic, immune, endocrine and environmental
elements contributing to their development (Rose, 1988; 1989; McGregor, 1990;
Dalton and Bennett, 1992). A variety of factors have been identified that are both

indirectly and directly responsible for the inappropriate activation of immune

44
effector cells (Bottazzo ef al., 1983; Cruse and Lewis, 1987; Sarvetnick et al., 1988;

Sinhe et al., 1990; Grossman at al., 1991; Kronenberg, 1991). figure 1.5
summarizes schematically some of the initiators of the pathogenesis of
autoimmune disease. These factors are often interrelated and can act in concert
to give rise to pathogenic autoimmunity.

a. Decreased Suppression

Reversal of energy might form part of the spectrum of autoimmunity. Antigen
specific T. cells responsible for blocking the effects of autoreactive T" cells could
be rendered ineffecient and energy would be abrogated by allowing signals to
reach the B lymphocyte. Distortion in the helper/suppressor cell ratio could
produce an advantage to helper components (Krosgrud and Perkins, 1977; Tyan,
1977). Defects of T, cells have been described in systemic lupus erythemetosus,
rheumatoid arthritis, MG, and autoimmune hemolytic anemia (Shoenfield and
lsenberg, 1989a). In addition, aging alone has profound effects on thymic function
and may alter the balance of helper/suppressor cells (Serge and Serge, 1987).

b. Modlflcatlon of an autoantigen

Self antigens may be altered by a variety of chemicals. Patients receiving
chronic administration of the antihypertensive drug, a-methlydope, have a higher
incidence of developing autoimmune hemolytic anemia. This may be attributable
to a modification of the erythrocyte surface in such a way as to provide stimulation
of B cells (Rose, 1988; Holt, 1988b; Shoenfeld and lsenberg, 1989b). The addition

of a foreign determinant to native proteins may provide a mechanism for T cell

45

Decreased
Self-antigen suppression

+ Altered self-
In creased MHC l antlgen

Molecular e elf-reactlve Seq ues gel-ed
—> <-—
rnImIcry Th cell antigen

I

I cytokines l

 

 

 

 

IlgG autoantibodlesl

 

figure 1.5. Pathogenesis of autoimmune disease.

Autoimmunity may be initiated by a variety of mechanisms involving alterations in
immune regulation, changes in auto-antigens or both. These factors are often
interrelated and may act in concert to give rise to pathogenic autoimmunity.

recognition.

MG has been described in some patients on penicillamine (Bucknell et al.,
1975; Czlonkowska, 1977) and a model of penicillamine-induced MG has been
produced in animals (Bever et al., 1982; Bever and Asofsky, 1991). It is thought
that penicillamine either causes a generalized alteration in immunoregulation
(Bluestone and Goldberg, 1973; Epstein et al., 1979) and/or penicillamine
becomes covalently linked to host proteins causing antigenic changes which lead
to an immune response against the host protein, in this case, acetylcholine
receptors (Dawkins et al., 1981; Bever et al., 1982).

c. Self antigen plus aberrant self-encoding molecules

In many autoimmune diseases there is an overexpression of MHC
molecules on the target tissue (Shoenfield and lsenberg, 1989b). Because MHC
molecules are known to be important in antigen-presentation, these observations
led to the proposal that abnormal, excessive or ectopic self-encoding molecule
expression leads to the presentation of autoantigen to autoreactive Th cells
(Bottazzo et al., 1983). The basic principles of this hypothesis have been verified
experimentally using the thyroid as a model. It was shown that interferon-y induced
MHC expression in thyrocytes; that these cells could act as antigen-presenting
cells; and that, in autoimmune disease, many of the infiltrating T cells specifically

recognize these thyrocytes (Londei et al., 1984; 1985).

47
U. Release OI SOQUBSIBTBU antigens

Anatomic alterations may lead to the exposure of antigens that are normally
sequestered and concealed from the immune system. Because of this
sequestration, individuals may not be immunologically tolerant to such antigens.
Therefore, if such self antigens are released and interact with immunocompetent
lymphocytes, specific immune responses may develop (Kronenberg, 1991).
Examples of anatomically- sequestered antigens are intraocular proteins and
sperm.

9. Cross-reactivity

r. antigenic mimicry

Many examples are known in which potential autoantigenic determinants are
present on an exogenous cross-reacting antigen which provokes autoantibody
formation (Shoenfield and lsenberg, 1989b). An antiserum raised against an
antigen can cross-react with a partially related antigen which bears one or more
similar determinants. The more determinants that are similar, the greater the cross-
reactivity (Roit, 1988b).

Infection by certain microorganisms may result in antigenic mimicry. Immune
reactivity against microbial antigens may encompass the stimulation of Th
lymphocytes capable of reacting with self. Alternatively, structural mimicry between
selected antigens of microorganisms and mammalian tissue may result in cross-
reactivity of the antibody response (Rose, 1989). Well known examples of the

association of autoimmune disease with infection are the role of streptococcal

or "_I l
' —1

48
infection with the development of rheumatic fever (Cruse and Lewis, 1987; Rolf,

1988b) and the onset of diabetes mellitus type 1 following mumps (Coxsackie)
virus infection (King et al., 1983).

Ii. aging

Studies have also demonstrated a progressive decline in the immune
response associated with aging. For example, there are several changes in effector
functions of the T cell compartment (Krosgrud and Perkins, 1977; Tyan, 1977) and
there is a consistent decrease in the amount and in the average affinity of antibody
produced by B cells in aged animals (Price and Makinodan, 1972; Nordin and
Makinodan, 1974; Callard et al., 1977). Paradoxically, these changes are
accompanied by an increased incidence of autoantibodies (Goidl, 1987). Thus, the
aging immune system is poorly regulated.

The decline in antigen responsiveness associated with aging has been
characterized by the production of antibody with decreased average affinity and
decreased heterogeneity of the antibody affinity (Callard et al., 1977; Klinman,
1981). This decrease in heterogeneity in the aged is seen preferentially as the loss
of medium- to high-affinity antibody. Thus, the expressed antibody repertoire is
principally composed of low-affinity antibody (Doria et al., 1978; Goidl, 1987). The
etiology of the increase in antibodies that recognize self-components could be

explained then by inappropriate cross-reactive low-affinity antibodies.

49
G.AutoimmuneLEMSandtheroleofcancer

It is now well established that self-reactivity is a common phenomenon, but
the conversion from autoimmunity to autoimmune disease requires a variety of
additional factors. The mechanisms of autoimmunity are topics for which theories
and hypotheses outnumber facts. A major difficulty in defining the mechanisms of
human autoimmune disease has been the inability to identify the antigens that
initiate the autoimmune response. As a result the specific etiology of most
autoimmune diseases, in general, and of LEMS, in particular, is unknown.

Muscle weakness is often a feature of systemic malignancy, but is frequently
dismissed as a non-specific complaint in a chronically debilitated patient (Gomm
et al., 1990). A variety of non-metastatic neuromuscular disorders, including
cerebellar degeneration, peripheral neuropathy and myopathy may be associated
with malignant neoplasm, in particular with lung cancer (Brain at al., 1951; Hanson
etal., 1954; Gomm etal., 1990) .

As mentioned previously, an association of LEMS with SCC occurs in 60%
of LEMS patients (Eaton and Lambert, 1957; Lambert et al., 1961; Wise and
MacDermot, 1962; Elmqvist and Lambert, 1968; O'Neill et al., 1988). Early
suggestions were made that 800 tumors, produced biologically active peptides
which interferred with neuromuscular transmission (lshikawa et al., 1977). These
studies, however, where unable to be duplicated. Further studies of the association
of LEMS with neoplasms have proposed that an lgG autoantibody is produced in

response to antigens on the surface of SCC tumor cells which cross-react with

50

Ca2+ channel determinants on the presynaptic motor nerve terminal (Newsom-
Davis at al., 1982; Lennon et al., 1982; Vincent et al., 1989).

In support of this hypothesis, oat (small) cells are thought to be of
neuroectodermal origin and have many of the biochemical markers of neurons
(Gadzar et al., 1981; Moody et al., 1981; Tapia et al., 1981). Electrophysiological
recordings from, and radioactive “Ca2+ flux and Fura-2 measurements in small
cells in tissue culture indicate that SCC cells have voltage-dependent Ca2+
channels (McCann et al., 1981; Roberts et al., 1985; De Aizpurua et al., 1988; Sher
at al., 1990). Furthermore, 800 cells grown in LEMS lgG containing medium
exhibited significantly less K*-stimuleted Ca2+ influx than control lgG-treated cells
(Roberts et al., 1985; De Aizpurua et al., 1988).

More recently, the immunocytochemicel characteristics of small cell
carcinoma with and without associated LEMS was assessed by Morris et al.,
(1992). In LEMS-associated 800, there were numerous macrophages at the
periphery of the tumor and many in the tumor mass. This contrasted with the non-
LEMS cases in which there were fewer macrophages both in the periphery and in
the tumor. Lymphocyte aggregates were also seen adjacent to the tumor in LEMS
patients with fewer seen in non-LEMS tumors. Analysis of the lymphocyte
populations using subset-specific monoclonal antibodies revealed that the majority
of lymphocytes were B cells. Fewer T cells were seen and there was little
difference in proportions of Th/T, cells in either LEMS or non-LEMS tumors. The

marked macrophage and lymphocyte infiltration in LEMS-associated SCC raises

51
the possibility that it is indeed the tumor in LEMS that is provoking the immune

response (Morris et al., 1992).

Clinical evidence for neurological improvement after treatment or excision
of the tumor would be consistent with this fact. Specific tumor therapy (resection
and/or irradiation) may be followed by improvement or remission of the
neurological deficit but this has not been generally noted (Ingram et al., 1984;
Ferroir et al., 1989; Chalk et al., 1990; Oh, 1990). Furthermore, interpretation of the
data from these studies can be difficult as many of the patients receiving specific
tumor therapy are also receiving pharmacological treatment for LEMS. Therefore,
it seems likely that the tumor may be the initial antigenic target, but this can not be
stated with certainty.

The clinical and electrophysiological features of LEMS do not show any
consistent differences between those with underlying $00 and those who do not.
Thus, the disease which is homogeneous in its clinical expression has more than
one triggering factor. In about 60% of the cases, this factor appears to be 800
while in the remainder the factor(s) are unknown. One possibility is that LEMS
patients without detectable cancer had a small occult tumor which was successfully

eliminated by the immune response Mncent et al., 1989).

CHAPTER TWO

SYNAPTOSOMES AND SYNAPTOSOMAL ION CHANNELS

52

A. General IntdeJction:

The exact identity of the antigenic target of the autoantibody which results
in LEMS is unknown. Studies of the association of LEMS with $00 suggest that
an antibody is produced in response to antigens on the surface of SCC tumor cells
and that these antibodies cross-react with Ca2+ channel determinants on the
presynaptic motor nerve terminal (Newsom-Davis et al., 1982; Lennon et al., 1982;
Vincent et al., 1989). However, there is reason to believe that Ca2+ channels from
other nerve terminals may also be susceptible to the the putative blocking actions
of the LEMS autoantibody (Meyer et al., 1986). LEMS immunoglobulin (lg) appears
to interact with voltage-dependent Ca2+ channels in a number of model cell
systems (Roberts et al., 1985; Login et al., 1987; De Azipurua et al., 1988a, 1988b;
Kim and Neher, 1988; Lang et al., 1989a; Peers et al., 1990). Assumptions
concerning the subtype of Ca2+ channel affected by the LEMS autoantibody have
been made based on the subtypes found in these various cell lines. However, the
exact identity of the subtype(s) of Ca2+ channel affected in nerve terminals remains
unknown. Again, it is not only possible but probable that characteristics of Ca2+
channels in the terminal differ from those in the cell body (Leonard et al., 1987;
Suszkiw et al., 1989; Charlton and Augustine, 1990; Uchitel et al., 1992).

Passive transfer of the electrophysiological characteristics of LEMS to
rodents by chronic injection of lgG isolated from LEMS patients permits only
indirect experimental investigation of the putative site of action of the LEMS

autoantibody. This is due to the fact that motor nerve terminal Ca2+ channels are

54

not sufficieme accessible nor amenable to direct recording techniques. Since the
low density of nerve terminals in skeletal muscle makes them impraticel to isolate,
an alternative model of centrally isolated nerve terminals (synaptosomes) may be

useful to examine this problem at a more molecular level.

B.Syneptmomes

1. General Descriptlon

Synaptosomal suspensions are heterogeneous preparations of intact,
isolated nerve terminals. They are prepared from brain homogenates using density
gradient centrifugation techniques (Gray and Whittaker, 1962). Synaptosomes may
be formed by a process of pinching-off so that the cytoplasm is never at any time
exposed to the isolation medium, or conversely by resealing of the fractured
membrane after rupture of the axon with brief exposure of cytoplasm to isolation
medium (Grey and Whittaker, 1962; Bradford, 1975). Whatever the mechanism, the
results are completely sealed cytoplasmic sacs of 1-2 uM in diameter retaining all
of the cell organelles visible in the intact terminal in situ, including mitochondria,
endoplasmic reticulum and synaptic vesicles (Gray and Whittaker, 1962; Dodd et
al., 1981). Also, many metabolic and synthetic pathways remain intact in
synaptosomes. These include glycolysis, the Kreb’s cycle and neurotransmitter
synthesis (de Belleroche and Bradford, 1973; Bradford, 1975). Synaptosomes
maintain transmembrane ion gradients and therefore a membrane potential. By

using a potentiometric, fluorescent dye, investigators estimate the presence of a

55
membrane potential in the -55 to -65 mV range (Blaustein and Goldring, 1975;

Here and Atchison, 1992). High K’ concentrations and veratrum alkaloids, such
as veratridine, depolarize the synaptosomal membrane and induce Na“ and Ca“
influx, K’ efflux, and Ca2”-dependent transmitter release (Bradford et al., 1973;
Nachshen and Blaustein, 1980; Tamkun and Catterall, 1981; Draupeau and
Blaustein, 1983; Bartschart and Blaustein, 1985; Nachshen, 1985). When
synaptosomes are incubated in a glucose-containing medium at 37°C, they remain
viable for up to four hours after isolation (Bradford, 1975). Incubation at lower
temperatures or maintenance of synaptosomes on ice significantly increases their
viability.

Synaptosomes have been used widely in receptor binding assays (Turner
and Goldin, 1985; Dunn, 1988; Marqueze at al., 1988; Suszkiw et al., 1989), and
for studies of nerve terminal channel function (Nachshen and Blaustein, 1979,
1980; Bartschat and Blaustein, 1980; Tamkun and Catterall, 1981) and
neurotransmitter synthesis and release (Bradford et al., 1973; Drapeau and
Blaustein, 1983; Floor, 1983; Woodward et al., 1988). Results from such
experiments have contributed significantly to knowledge of the neurochemical
events that occur at the presynaptic nerve terminal.

2. Synaptosomal Ion channels

a. Sodium channels

Influx of Na+ through voltage-dependent channels is involved in action

potential generation and propagation. A variety of toxins from both plants and

 

 

 

56
animals specifically interact with voltage-dependent Na” channels in excitable

membranes. Saxitoxin (STX) and tetrodotoxin (TTX) specifically block these
channels while batrachatoxin (BTX) and veratridine (VER) persistently activate
voltage-sensitive Na+ channels. The latter two toxins will also depolarize excitable
cells by increasing Na+ permeability (for review see Catterall, 1980). The use of
these neurotoxins as specific pharmacological probes has made it possible to
establish biochemical methods to study Na+ channel properties in excitable tissues
and in synaptosomes.

Synaptosomal Na+ channels were indirectly demonstrated by VER-
stimulated 45032" uptake (Blaustein, 1975; Nachshen and Blaustein, 1979), VER-
stimulated neurotransmitter release (Blaustein, 1975; Cotman at al., 1976) and
VER- and BTX- induced synaptosomal depolarization (Krueger and Blaustein,
1980). Direct evidence for the presence of Na+ channels in synaptosomes was
demonstrated by Krueger and Blaustein (1980) and Tamkun and Catterall (1981),
who reported VER-stimulated uptake of 22Na+ into synaptosomes. This uptake was
TTX-sensitive. The use of these toxins has enabled these investigators to conclude
that Na+ channels in synaptosomes retain functional properties that are identical
with those in intact neurons (Krueger at al., 1980; Tamkun and Catterall, 1981).

b. Potassium channels

Potassium channels located in the presynaptic nerve terminal control
excitability and neurotransmitter release and thus have the potential to be key

regulators of synaptic transmission. Previous work has shown that synaptosomes

 

 

57
contain a variety of K“ channels whose function can be studied with “RN effiux

methods (Bartschat and Blaustein, 1985a,b). The various channels can be
distinguished on the basis of their kinetic and pharmacological characteristics
(Bartschat and Blaustein, 1985a,b; Benishen et al., 1988) and correspond to
known conductances determined in cell systems by electrophysiological
techniques.

Rat brain synaptosomes contain at least four distinct types of K” efflux
mechanisms: (1) a resting K+ permeability, seen with 5 mM K+ solutions; (2) a fast,
inactivating, voltage-dependent K+ channel sensitive to block by
tetraethylammonium (TEA), tetrabutylammonium (TBA) and 4-aminopyridine (4-AP);
(3) a slowly inactivating, voltage-dependent K+ channel which is less sensitive to
TEA, TBA and 4-AP; and (4) a Ca2*-activated K+ channel (Bartschat and Blaustein,
19853).

Rb" passes through these various types of channels as well as K+ itself. It
is accumulated in the cell by the Na”/K* ATPase and is released under
depolarizing conditions (Bartschart and Blaustein, 1985a,b). Since a“Rb“ has a
longer half-life (days) than ‘ZK" (hours), it is a more practical choice for assessing
K“ channel activity in synaptosomes or intact cells.

c. Calclum channels

Ca2+ plays an essential role in the control of a wide range of cellular
functions. Normally, cells maintain an [Ca2+], in the range of 100 nM while the

[Ca2+]. is typically in the low millimolar range. A transient rise in the [Ca2+], can

58

trigger many biological responses, including excitation-contraction coupling and

neurotransmitter release. One way in which [Ca2+],

can be rapidly increased is by
the opening of voltage-sensitive Ca2+ channels in the cell membrane, allowing Ca2+
to flow down its electrochemical gradient. Ca2+ channels can be classified into
various subtypes differing dramatically in the mechanism of inward current decay
during a maintained depolarization and in their sensitivity to pharrnacologicel
agonists and antagonists (see Chapter One).

Radiotracer flux studies, using 45Ca2+, have been useful in demonstrating
Ca2+ influx into synaptosomes. Voltage-dependent uptake of Caz+ into
synaptosomes consists of two distinct phases (Nachshen and Blaustein, 1980;
Suszkiw and O’Leary, 1983; Nachshen, 1985). The 'fast" phase of influx occurs
during the first second of depolarization. It is sensitive to block by lanthanum (<
1 it M) and nickel (40 u M) and inactivates during the the first 2-5 see of K"-induced
depolarization (Nachshen and Blaustein, 1980; Nachshen, 1985). The second or
"slow" phase of Ca2+ entry remains active for 20-90 see after depolarization and
is insensitive to block by lanthanum (Nachshen and Blaustein, 1980).

The “slow” phase of influx is believed to be mediated primarily by a reversed
Nai/Ca2+ exchanger with a residual slowly inactivating Ca2+ channel component
(Blaustein and Oborn, 1975; Nachshen et al., 1986). Removal of Na+, largely
reduces 'slow' influx (Turner and Goldin, 1985; Suszkiw et al., 1986; Shafer and
Atchison, 1989). The "fast" phase of influx is presumably mediated by voltage-

dependent Ca2+ channels (Nachshen and Blaustein, 1980; Suszkiw and O'Leary,

59
1983). Uptake in this phase is independent of external Na“ (Shafer and Atchison,

1989) and is associated with neurotransmitter release (Drapeau and Blaustein,
1983; fioor, 1983; Suszkiw and O’Leary, 1983; Nachshen, 1985; Leslie et al.,
1985).

Much like Ca2+ channels in other cell systems, voltage-dependent Ca2+
channels in synaptosomes undergo both voltage- and Ca2+-dependent inactivation
(Suszkiw and O’Leary, 1983; Suszkiw et al., 1986; 1989) and are sensitive to block
by inorganic, multivalent metal cations (Nachshen, 1984). The most potent
blockers are lanthanum, cadmium and lead. The block is rapid in onset and is not
attributed to changes in either the membrane or surface potential. Inhibition of
Caz+ flux by these cations is relieved by increasing [Ca2+], suggesting that the
inhibitory ions compete with Ca2+ and directly block passage of Ca2+ ions through
the Ca2+ channel (Nachshen, 1984). The antibiotics neomycin and polymixin B
block synaptosomal Ca2+ channels in a manner that is reversed by increasing
[Ca2+], also (Atchison et al., 1988).

In addition to inorganic metals, the organic metals methylmercury (Atchison
et al., 1986; Shafer and Atchison, 1989; Hewett and Atchison, 1992) and
ethylmercury (Hewett and Atchison, 1992) block 45Ca2+ influx into synaptosomes
in a manner that is irreversible by increasing [Ca2+]. (Atchison et al., 1986; Hewett
and Atchison, 1992). It appears that both charge and Iipophilicity are important in
determining the blocking characteristics of mercurials, as only positively charged

mercurials blocked synaptosomal Ca2+ channels and only organic mercurials do

60
so in a voltage-dependent manner (Hewett and Atchison, 1992).

finally, the organic Ca2+ channel blocker, verapamil also decreases 4‘r’Ca2“
influx into synaptosomes (Nachshen and Blaustein, 1979; Akerman and Nicholls,
1981; Norris et al., 1983; Nachshen, 1985; Carvalho et al., 1986). Single-channel
recordings of synaptosomal Ca2+ channels in lipid bilayers confirm that lanthanum,
cadmium and verapamil are potent blockers of synaptosomal Ca2+ channels
(Nelson, 1985).

i. Ca2+ channel subtypes

Synaptosomes are likely to contain more than one subtype of Ca2+ channel
but the types present have yet to be identified clearly. Block of Ca2+ channels by
verapamil suggests that L-type channels may be involved in Ca2+ influx. This is a
subject of much controversy as DHPs, also L-type antagonists, do not have
consistent effects on 4"’Ca2+ influx into synaptosomal preparations. DHP
antagonists bind to synaptosomal preparations with high affinity (Boles et al., 1984;
Greenberg et al., 1984; Turner and Goldin, 1985; Wei and Chang, 1985; Caivalho
et al., 1986; Suszkiw et al., 1986; Dunn, 1988). However, a number of investigators
have found evidence for both DHP-sensitive and -insensitive pathways mediating
synaptosomal Ca2+ influx and neurosecretion.

Middlemiss and Spedding (1985) reported that submicromolar
concentrations of the DHP agonist, Bay K 8644, could augment K*-evoked

neurotransmitter release from cortical brain slices. The augmented release was

blocked by the DHP antagonist, nifedipine (Middlemiss and Spedding, 1985).

61
Woodward and Leslie (1986) and White and Bradford (1986) independently

obtained similar results in synaptosomes (Woodward and Leslie, 1986; White and
Bradford, 1986; Woodward et al., 1988a). From the above studies, it appears that
L-type Ca2+ channels in the CNS may modulate neurotransmitter release but do
not contribute under normal circumstances. As you may recall, this was the
conclusion drawn from studies using rat sympathetic neurons as well as rat NMJs
(Perney at al., 1986; Atchison, 1989).

In contrast, various investigators report direct block of synaptosomal K*-
stimulated Ca2+ influx and neurotransmitter release by nifedipine, nimodipine and
nitrendipine (Turner and Goldin, 1985; White and Bradford, 1986; Dunn, 1988). Still,
others report no effect of DHPs on K"-stimulated Ca2+ uptake into synaptosomes
(Nachshen and Blaustein, 1979; Daniell et al., 1983; Wei and Chang, 1985;
Suszkiw at al., 1986; Reynolds 91 al., 1986; Vickroy et al., 1992). Thus, the

4..

relationship of DHP receptors to synaptosomal Ca2 channel function is still
uncertain.

Synaptosomes also contain specific high-affinity receptors for the N-type
channel antagonist, Cng. (Abe et al., 1986; Marqueze et al., 1988; Wagner at al.,
1988; Scheer, 1990). Polymixin B and several aminogycoside antibiotics potently
inhibit binding of radiolabeled Cng to guinea-pig cerebral cortex membranes
(Knaus et al., 1987) and to rat forebrain synaptosomes (Wagner et al., 1988).

Rapid “Caz+ flux measurements into rat brain synaptosomes demonstrates only

partial antagonism (a 20-40%) of voltage-dependent Ca2+ channels by Cng,

62
however (Reynolds et al., 1986; Suszkiw et al., 1987; Woodward et al., 1988b;

Suszkiw et al., 1989). Thus, N-type channels may mediate only a proportion of
Ca2+ uptake and neurotransmitter release from synaptosomes.

Because depolarization-induced Caz+ entry into rat brain synaptosomes is
largely resistant to both Cng and DHPs, it appears that synaptosomes contain
other high-voltage-activated channels which may be neither N- nor L-type (Lundy
et al.,1992). In addition, Suszkiw and coworkers (1989) demonstrated that it was
not possible to clearly classify the Ca2+ channels into the T-, L- or N-type based
on their activation and inactivation kinetics (Suszkiw et al., 1989). In some
respects, Ca2+ influx into synaptosomes has characteristics similar to currents
mediated by P-type channels (Leonard et al., 1987; Uinas et al., 1989; Suszkiw et
al., 1989). Likewise, the P-type Ca2+ channel blockers, m-agatoxin IVA and FTX,
are potent blockers of K*-induced “Ca2+ uptake into rat brain synaptosomes
(Mintz et al., 1992; Uchitel at al., 1992). These findings suggest that P-type Ca2+
channels may be involved in transmitter release from mammalian synaptosomes,

but this requires verification.

C. Overall Aim

It is reasonably clear that LEMS is due to autoimmune attack directed
against the nerve terminal in affected patients. However the exact identity of the
target remains speculative. In large part this is due to the inacceslbility of the

putative target site, the voltage-dependent Ca2+ channel in the presynaptic motor

63
nerve terminal. The overall objective of this dissertation was to test the hypothesis

that a circulating autoantibody in the plasma of patients with LEMS interferes with
nerve terminal Ca2+ channels associated with neurotranmitter release. To test this,
the model system of isolated nerve terminals from mammalian CNS was used.
Several lines of pharmacological evidence suggest close similarity between Ca“
channels in synaptosomes and those at motor nerve terminals. Cng and DHPs
are unable to block release of ACh from murine motor nerve terminals (Anderson
and Harvey, 1987; Atchison, 1989; Protti, 1991). Rapid “Ca"+ flux measurements
in rat brain synaptosomes have shown that the voltage-dependent Caz+ channels
that mediate transmitter release are, in most but not all cases, DHP insensitive
(Nachshen and Blaustein, 1979; Turner and Goldin, 1985; Suszkiw et al., 1986;
Woodward and Leslie, 1986) and at best are only partially antagonized by Cng
(Reynolds, 1986; Suszkiw et al., 1989) as well. However, mammalian
neuromuscular transmission and voltagedependent Ca2+ channels in synap-
tosomes are effectively blocked by FI'X (Uchitel et al., 1992). FTX blocked
presynaptic Ca2+ currents, nerve-evoked transmitter release and muscle
contraction from a murine nerve-muscle preparation as well as K*-stimulated
45Ca2+ influx into rat brain synaptosomes (Uchitel, 1992). Thus, synaptosomes may
be an useful model system in which to study the potential interaction of the

autoantibody in LEMS with Ca2+ channels involved in neurotransmitter release.

1. Specific Alm 1

Incubation of rat cortical synaptosomes with serum or high molecular weight
serum fractions from patients with LEMS decreased the amount of [3Hj-ACh
released during depolarization (Mayer et al., 1986). This study was important in
that it indicated that acute exposure of synaptosomes to the serum or serum
fractions from LEMS patients could elicit responses qualitatively similar to those
obtained in chronic passive experiments in mice. It did not measure Ca2+ channel
activity, however. Therefore, studies were performed to determine whether acute
application of plasma and serum from patients with LEMS could reduce
depolarization-dependent uptake of Ca2+ into rat forebrain synaptosomes. The

results from these studies are presented in Chapter Three.

2. Speclflc Alm 2

Chronic injection of mice with plasma, crude lgs or purified lgG from
patients with LEMS was effective in transferring both the electrophysiological and
morphological characteristics of the disease (Denys et al., 1972; Lang et al., 1981;
Newsom-Davis eteI., 1982; Fukunaga etal., 1983; Ifim, 1985; 1986; Lennon and
Lambert, 1988; Nagel et al., 1988). This led to the conclusion that the abnormality
in LEMS was due to the production of lgG autoantibodies. The objective of the
studies in Chapter Four was to test whether acute application of IgG isolated from
a patient with LEMS could disrupt synaptosomal Ca2+ channels, and if so, whether

other ion channels in nerve terminals were similarly affected. In addition, studies

65

were performed to determine if non-specific membrane damage contributed to the

pathogenesis of LEMS.

3. Speclflc Aim 3

Activation of the serum component complement plays a central role in
mediating postsynaptic membrane damage in animal models of the autoimmune
neuromuscular disorder MG, as well as in the human disease state (Engel at al.,
1977; Sahashi et al., 1978). Attempts to clarify the participation of complement in
the pathogenesis of LEMS have so far yielded negative results (Prior et al., 1985;
Lennon and Lambert, 1988). As it is possible that complement activation may be
important in LEMS, studies were undertaken to assess the role of serum in the
pathogenesis of Ca2+ channel dysfunction induced by LEMS lgG in isolated nerve
terminals from mammalian CNS. The results from these studies are presented in

Chapter Five.

4. Speclflc Alm 4

LEMS immunoglobulin (lg) appears to interact with voltage-dependent Ca2+
channels in a number of model cell systems (Roberts et al., 1985; Login et al.,
1987; Peers et al., 1987; De Azipurua et al., 1988a, 1988b; Kim and Neher, 1988;
Lang et al., 1989a). Although, these studies have proven useful in the
characterization of the subtype of Ca2+ channel affected by the LEMS autoantibody

in the cell some, the exact identity of the subtype(s) of Ca2+ channel affected in

66

nerve terminals remains unknown. Therefore, attempts to characterize
pharmacologically the subtype of Ca2+ channel in synaptosomes that is affected

by the LEMS autoantibody are presented in Chapter Six.

CHAPTER THREE

EFFECT OF SERUM AND PLASMA FROM PATIENTS WITH LAMBERT-

EATON MYASTHENIC SYNDROME ON DEPOLARIZATION-INDUCED

UPTAKE or 45cit2+ INTO RAT CORTICAL SYNAPTOSOMES

67

A. Summary

The reduction in nerve-evoked release of transmitter at the neuromuscular
junction of patients with LEMS is thought to be caused by a circulating
autoantibody to calcium channels of presynaptic motor nerve terminals. Studies
were undertaken to determine whether acute application of plasma and serum from
patients with LEMS or SCC would reduce depolarization-dependent uptake of
“scat+ into isolated nerve terminals of the CNS. Net potassium-stimulated influx
was reduced by sera and plasma from five patients with LEMS but not by sera
from four patients with 900. Lactate dehydrogenase (LDH) release from
synaptosomes incubated with plasma or serum from patients with LEMS was not
increased over control indicating the reduction in “Ca2+ uptake was not a result
of disruption of synaptosomal membrane integrity. These results are significant
because: (1) they demonstrate that acute exposure to a circulating factor in
sore/plasma from a patient with LEMS is sufficient to inhibit Ca2+ channel activity
in isolated nerve terminals, as opposed to chronic regimens used in other models
for the disease; (2) they indicate that the existence of SCC alone in some patients
is insufficient to trigger a LEMS-like autoimmune response; (3) they suggest that
Ca2+ channels of nerve terminals secreting different transmitters may share
common epitopes recognized by the LEMS autoantibody; and (4) they suggest
that synaptosomes will be useful in determining the neurochemical site and

specificity of the LEMS autoantibody.

B. Introduction

LEMS is a presynaptic disorder of neuromuscular transmission frequently
associated with 800 (Eaton and Lambert, 1957; Wise and MacDermot, 1962;
Elmqvist and Lambert, 1968). The disease is characterized by a reduction in Ca2+-
dependent quantal release of ACh in response to nerve stimuli (Lambert and
Elmqvist, 1971; Cull-Candy et al., 1980). There is no evidence of a defect in ACh
synthesis nor of decreased sensitivity of the postsynaptic membrane to ACh
(Lambert and Elmqvist, 1971; Lindstrom and Lambert, 1978; Cull-Candy et al.,
1980). Muscle fiber excitability and nerve conduction appear normal (Pascuzzi and
Kim, 1978). Thus, LEMS is presumed to result from defective transmitter release.

Evidence for an autoimmune origin of LEMS exists (Lang et al., 1981;
Lennon at al., 1982; Newsom-Davis and Murray, 1984). The ability to transfer the
human neuromuscular defect to mice by injections of either plasma or
immunoglobulin derived from LEMS patients (Denys et al., 1982; Newsom-Davis
etel., 1982; Lang etel., 1983a; Lang etel., 1984; Kim, 1985; Kim, 1986; Lambert
and Lennon, 1988) supports the hypothesis that circulating autoantibodies to some
presynaptic component are present in patient serum and mediate the pathogenesis
of this disease. Evidence suggests that these autoantibodies interact with voltage-
dependent Ca2+ channels since they have the ability to alter Caz+ flux in several
cell lines of diverse origin (Roberts et al., 1985; Login et al.,1987; Peers et al.,
1987; Kim and Neher, 1988; De Azipurua et al., 1988). These studies have proven

useful in the characterization of the LEMS autoantibody's antigenic target in the cell

70
some. However, studies which specifically examine the ability of the LEMS

autoantibody to interact with Caz+ channels in nerve terminals are lacking. This is
important as it is possible that the type of Ca2+ channel in the terminal may differ
from that in the cell body (Suszkiw et al., 1989; Charlton and Augustine, 1990;
Uchitel et al., 1992).

In the present investigation, we sought to determine whether acute
application of plasma and serum from LEMS patients could disrupt Ca2+ channel
activity in presynaptic nerve terminals from the CNS. Synaptosomes maintain the
morphological and functional properties of intact news terminals and several
studies suggest that Ca2+ channels in the CNS resemble those in the periphery
(Anderson and Harvey, 1987; Nachshen and Blaustein, 1979; Reynolds et al.,
1986; Suszkiw etel., 1986;1989; Atchison, 1989; Protti etel., 1991; Uchitel etel.,
1992). LEMS patients do not, however, normally exhibit CNS symptomology
because lgG is a high molecular weight protein (1:150 kDa) that does not readily
pass the blood brain barrier. If the autoantibody in the sera of patients with LEMS
could gain access to the calcium channels of the CNS by incubation in vitro, it
might be possible to induce a passive transfer of the nerve terminal dysfunction to

central neurons.

C. Materials and Methods
1. Anlmals
Male Sprague-Dawley rats (Harlan; 175-2009) were maintained on a 12 hr

light—dark cycle in a temperature- and humidity-controlled room for at least 3 days

71
prior to use. Standard laboratory rodent chow and water were provided ed Iibitum.

2. Serum/Plasma Samples

Sara and plasma samples from patients diagnosed as having LEMS were
kindly provided by the following persons: Andrew Massey, M.D., Dept of
Neurology, University of Kentucky Medical Center, Lexington,KY; Mark Glasberg,
M.D., Henry Ford Hospital, Dept of Neurology, Detroit, MI; David Johnson, MD,
Vanderbilt University, Nashville, TN; Donald Sanders, MD, Division of Neurology,
Duke University Medical Center, Durham, NC; Daniel Watson, M.D., Dept of
Neurology, Johns Hopkins School of Med, Baltimore, MD; Eva Feldman, M.D.,
Dept. of Neurology, University of Michigan, Ann Arbor, MI and Daniel Drachman
M.D., Dept of Neurology, John Hopkins School of Med., Baltimore, MD.

Plasma and serum samples from patients with 800 were provided by David
Johnson, MD, Vanderbilt University, Nashville, TN and purchased from the
Cooperative Tissue Network, Columbus, OH. Control plasma from patients with no
history of neurological or immunological disease was purchased from the American

Red Cross, Lansing MI.

3. Preparation of Synaptosomes

Synaptosomes were prepared from whole brain (minus the brainstem and
cerebellum) of male Spraque-Dawley rats according to a modified method of Gray
and Whittaker (1962). Forebrains were homogenized in 10 ml of 0.32 M unbuffered

sucrose and centrifuged at 1000xg, 4°C, for 10 min. The supernatant was

72
decanted and the pellet discarded. The supernatant was next centrifuged at

10,000xg for 20 min and the resulting pellet resuspended in fresh 0.32 M
unbuffered sucrose. The synaptosomal fraction was further purified by layering
onto a sucrose density gradient consisting of 0.8 and 1.2 M unbuffered sucrose,
and centrifuged at 69,000xg for 2 hr. The synaptosome-enriched fraction was
subsequently removed from the interface between the 0.8 and the 1.2 M layers of
sucrose, and diluted 1 :1 (v/v) with a physiological saline buffer. The synaptosomal
suspension was then washed free of sucrose by centrifugation at 10,000xg for 10
min and resuspended in the physiological saline to an approximate concentration
of 15-25mg protein/ml as determined by the method of Lowry et al., (1951).
Following isolation, equal aliquots of the synaptosomal suspension were
incubated at 37°C for one hr, while shaking, in a pure O2 atmosphere with varying
concentrations of either plasma or sera (expressed as " % of total incubation
volume”) from patients with no known neurological or immunological illness
(control), SCC in which no neurological involvement was observed, or LEMS. The
proper volume of physiologic saline buffer was added to maintain a constant

reaction volume.

4. Solutions

The solutions used in the 4"’Ca2+ uptake experiments contained the
following: physiological saline buffer (mM): NaCl, 145; KCI, 5; MgClz, 1 ; d-glucose,
10; and HEPES, 10; depolarizing K: solutions (mM): NaCI, 72.5; KCI, 77.5; MgCIz,

1; CaClz, 0.04; d-glucose, 10; and HEPES, 10; and quench solution (mM): KCI, 5;

73
MgClz, 2; EGTA, 1; d—glucose, 10; HEPES, 10; and N-methyl-d-glucamine, 145.

Non-depolarizing K“ solutions were identical to depolarizing K“ solutions except
that [K"], was 5 mM, and 72.5 mM choline chloride was added to maintain iso-
osmolarity.

The LDH reaction mixture contained (mM): KzHPO,, 41.2; KH2P04, 6.8; Na+

pyruvate, 0.6; NADH, 0.2 ; and NaHCOa, 1.1.

5. Measurement of “Ca2+ Uptake

Uptake of “Ca“ was measured as described previously (Atchison et al.,
1986). fifty ill of the synaptosomal suspension was added to 50 ill of either
depolarizing or non-depolarizing K*-buffer containing 1 llCi of “Ca2+ (New
England Nuclear, Boston, MA) and uptake was measured for 10 sec. The uptake
of 4ice“ associated with transmitter release occurs within the first 2 sec and
accounts for 80% of the uptake measured over a 10 sec period (Nachshen and
Blaustein, 1980; Drapeau and Blaustein, 1983). Therefore, these data accurately
reflect the uptake of “"‘Ca2+ associated with voltage-dependent Ca”+ channels in
the synaptosomes as reported previously (Nachshen and Blaustein, 1980; Drapeau
and Blaustein, 1983; Suszkiw et al., 1986; Suszkiw et al., 1989; Shafer and
Atchison, 1989).

The reaction was stopped by the addition of 2 ml of cold quenching solution
followed by suction-filtration of the sample through 0.45 itM HVLP Millipore filters
(Millipore, Bedford, MA). Radioactivity remaining on the filters in low [K‘], solution

was subtracted from that obtained in high [K‘]. solutions as depolarization-

74
independent background. Experiments were replicated five to seven times. Values

obtained for any particular experiment were the average of three replicates.
Readings of dpm were corrected back to the original volume of synaptosomes

filtered and expressed as nmole “Cazv 119 protein (Lowry et al., 1951).

6. Measurement of Lactate Dehydrogenase (LDH) Activity

LDH release was quantified by the method of Bergmeyer and Bernt (1974).
Synaptosomes were incubated with varying concentrations of control and LEMS
lgG as described above. Immediately after the incubation period, synaptosomes
were pelleted by centrifugation. The supernatant was decanted into a separate test
tube. A 30 ul aliquot of supernatant was added to cuvettes containing 3 ml of the
reaction mixture. LDH activity was quantified by the rate of oxidation of
nicotinamide adenine dinucleotide (NADH), which was followed
spectrophotometrically at 340 nM for 3 min. Activity was expressed as the
percentage of LDH released into the supernatant. Total LDH activity (100 %) was
determined by lysing an equal volume of synaptosomes in 0.1 % (v/v) Triton-X-100

followed by sonication.

7. Statlslcal Analysls
Data were analyzed via a randomized complete block analysis of variance
followed by Dunnett’s t-test or by Student’s t-test for paired samples. Statistical

significance was assessed at p < 0.05.

75
D. Results

1. Uptake of “Ca”

Many LEMS patients, including patients used in this study, have SCC.
Uptake of “Ca2+ was determined following incubation of synaptosomes with SCC
serum. Serum from four patients with SCC in which no neurological involvement
was observed produced no substantial reduction in depolarization-dependent
uptake of “Ca2+ into rat forebrain synaptosomes (fig. 3.1).

The effect of both control and LEMS patient sera on “Ca2+ uptake into
synaptosomes during 10 sec K*-stimulation is depicted in figure 3.2.
Concentrations of 6 and 12% (v/v) control sera and plasma, had little effect on
depolarization-induced uptake of “Ca2+ into synaptosomes when compared to
uptake into synaptosomes with no added sera/plasma. However, incubation with
24% (v/v) control plasma/sera significantly reduced uptake compared to
incubations with no sera/plasma (fig. 3.2).

Conversely, incubation of synaptosomes with plasma or sera from five
LEMS patients at concentrations of 6 and 12% of the total incubation volume
resulted in significant reductions in net K*-stimulated influx (fig 3.2). Inhibition of
4"’Ca'“ uptake ranged from 0 to 20% (10.6 t 4.2%) and 14 to 30% (20.4 t 2.6%)
at concentrations of 6 and 12%, (v/v) respectively. At a concentration of 24% (v/v),
LEMS sera/plasma reduced influx by 0 to 61.8% (22.4 1 10.8%), but this was not
significantly less than its collective paired control. Incubation of synaptosomes with

serum from a sixth patient with LEMS reduced uptake of “Ca” by 20.2 t 18.1%,

76

 

 

4. —-con —scc
s " '
0
iii. a
5% —"-e- i. "
“-3 ' At T °
Dazp—‘r— . 8 —-—-_
I}: O A D
83:“ i'
‘2
5
o I I I In
0 6 12 24

SERUM CONCENTRATION

(% of total lncubatlon volume)

Figure 3.1. Effect of sera from 4 patients with SCC and no evidence of
neurological involvement on “Ca2+ uptake into rat forebrain synaptosomes during
10 sec of depolarization with 42.5 mM KCI. Prior to uptake, synaptosomes were
incubated for 1 hr with SCC or control serum. Uptake in 5 mM KCI was subtracted
from that in 42.5 mM KCI as depolarization-independent background. Each symbol
represents the mean of at least 3 experiments per patient. Paired experiments are
denoted by identical symbols in the control and SCC group. The horizontal bar
represents the mean of all 4 patients tested.

77

 

 

A6” ——con —LEMS

3 v

£5 .0 V V

5% 1 , _. e v

33.3". .a 8b.

a s o t — "“0

8; El U E —

:2 '3' 15

° El
E

5 O
0 i l 1 #4

0 6 12 24

SERA/PLASMA CONCENTRATION

(% of total Incubation volume)

Figure 3.2. Effect of plasma and sera from 5 patients with LEMS on uptake of
45Ca2+ into rat forebrain synaptosomes during a 10 sec depolarization.
Synaptosomes were incubated for 1 hr with either LEMS serum or plasma or
control serum or plasma, respectively. Uptake was measured as described in
figure 1.1. Each symbol represents the mean of at least three experiments per
patient. Paired experiments are denoted by identical symbols in both the control
and LEMS-treated group. The horizontal bar denotes the mean of all 5 patients
tested. ‘ Indicates values which are significantly less than its corresponding paired
control value (p < 0.05). b Indicates a value that is significantly less than uptake
into synaptosomes that received no serum or plasma (p < 0.05).

78
52.2 2 18.4% and 43.8 1 9.4% at 10, 30 and 50% of the total incubation volume,

respectively (fig 3.3).

2. Synaptosomal Membrane Integrity

To rule out the possibility that the effect of LEMS sera/plasma was a result
of disruption of the synaptosomal membrane integrity, synaptosomal viability was
assessed by measuring lactate dehydrogenase (LDH) released into the
extrasynaptosomal fluid following incubation with plasma and sore. Neither
incubation with LEMS serum nor LEMS plasma increased LDH release compared

to incubation with control serum or plasma (figs. 3.4, 3.5).

E. Discussion

The aim of the present study was to test whether acute application of
plasma and serum from LEMS patients could disrupt synaptosomal Ca2+ channel
activity. Our results indicate that serum and plasma from patients with LEMS act
on presynaptic nerve terminal Ca2+ channels in the CNS to reduce depolarization-
induced uptake of 45Ca“. These deficits are not due to a decrease in
synaptosomal viability as LDH release was not increased over that released by
control sera/plasma. These studies, to my knowledge, are the first to examine
specifically the ability of a circulating humoral factor in serum from patients with
SCC and LEMS to interact with nerve terminal Ca2+ channels. This is important as
the putative site of action for the LEMS autoantibody is believed to be Ca2+

channels on motor nerve terminals.

79

I: con
LEMS

A

figure 3. 3. Effect of serum from one patient with LEMS on 45Ca2+ uptake into
synaptosomes. Synaptosomes were incubated with serum for 1 hr and uptake
measured as described in Fig. 1.1. The asterisk, , indicates uptake that is
significantly smaller than its respective control (p < 0. 05).

O)
]——r
|——I
|__.

3W

“Ca” UPTAKE
(nmoI/pg protein E-OS)
o)

k\\\\\\\\\\\\l—”

 

 

 

 

 

 

 

 

 

 

 

 

CD

30 5
SERUM CONCENTRATION

(% total incubation volume)

0

 

1OO -
_ —A- CONTROL
a
fi 3 —o-- LEMS
O .
=3 5° I
:I: 32 T I 1.--...“

d
T/ l l l l J

0 1O 20 30 40 50

0

SERUM CONCENTRATION

( % of total Incubation volume )

figure 3.4. Effect of LEMS sera on synaptosomal viability. LDH activity was
measured from the supernatant of synaptosomes treated for 1 hr with either LEMS
or control sera. LDH activity was quantified by the rate of oxidation of NADH at 340
nm. The quantity of enzyme is expressed as the percent release of the total
enzyme content of the synaptosomes, which was determined by lysis of an equal
volume of synaptosomes with 0.1% (v/v) Triton-X-100 followed by sonication.
Values are the mean 1 SEM of 3 experiments.

81

 

100 '
-A- CONTROL
Ill ‘3
m 13 ’ —o- LEMS
< 8
m I-
—l o
I.“ 3 50 '
o: . i
iii 0 I
-' 2 ’ OT”~~T»”’A;> ‘~ 3
s r‘” 1
o l l l 1 I

 

O 10 2O 30 40 50

PLASMA CONCENTRATION

(% of total Incubation volume)

Figure 3.5. Effect of LEMS plasma on synaptosomal viability. LDH activity was
measured from the supernatant of synaptosomes treated for 1 hr with either LEMS
plasma, control plasma or physiological saline. LDH activity was quantified by the
rate of oxidation of NADH at 340 nm. The quantity of LDH is expressed as the
percent over release in physiological saline only. All values are the mean 1 SEM
of 5 experiments.

82
Serum from SCC patients lacking neurological symptoms did not reduce

depolarization-dependent uptake of “Ca2+ into synaptosomes. This study
controlled for possible effects associated with neoplastic disease, since
approximately 60% of patients with LEMS also have SCC of the lung (Eaton and
Lambert, 1957; Wise and MacDermot, 1962; Elmqvist and Lambert, 1968). Oat
cells (small cells) possess several neuronal properties and are possibly of neural
origin (Gadzar et al., 1981; Moody at al., 1981; Tapia et al., 1981). They are
electrically excitable, and action potentials in cat cells may be calcium channel-
mediated as they are inhibited by Co2+ and Mn“, known calcium channel blockers
(McCann et al.,1981; Roberts et al., 1985). It has been suggested that LEMS may
arise as an immune response to SCC and that cholinergic neurons may have
similar epitopes leading to cross-reactivity between tumor and neuron (Newsom-
Davis, 1982; Lennon et al., 1982; Vincent et al., 1989). LEMS, however, also exists
in the non-neoplastic form and autoantibodies antagonistic to SCC Caz+ channel
activity were found in patients with LEMS independent of cancer status (Roberts
et al., 1985; De Azipurua et al., 1988). Since sera from SCC patients failed to inhibit
calcium flux at any concentration tested, this suggests that the existence of SCC
neoplastic disease is insufficient to trigger an autoimmune response in all patients.
These data also demonstrate that interpretation of data derived from LEMS patients
with SCC would not be complicated by the presence of carcinoma.

Incubation of synaptosomes with control plasma/sera at a concentration of

24% (v/v) significantly reduced “Ca2+ uptake compared to uptake in

83
synaptosomes which were incubated without sera/ plasma. This may be explained

simply by increased serum binding of 45Ca2+ thereby effectively decreasing the
concentration of radioactivity available for uptake by synaptosomes. This needs
to be tested further.

It is evident from the variability in the data that the circulating factor found
in different LEMS patients sera/plasma is not equieffective in disrupting “Ca2+
influx into synaptosomes. This is most likely due to differences in antibody titer
between patients. Nevertheless, a modest yet significant decrease in uptake
mediated largely by voltage—dependent Ca2+ channels was observed in
synaptosomes treated with 6 and 12% sera or plasma from five LEMS patients as
compared to their paired controls. A striking decrease in synaptosomal 45Ca2+
uptake was observed in one patient studied in isolation.

Neither incubation with LEMS plasma nor LEMS serum increased LDH
release compared to incubation with control serum or plasma. Thus, it is unlikely
that the decrease in calcium uptake is due to decreased synaptosomal viability, but
rather suggests block or disruption of presynaptic voltage-dependent calcium
channels in the nerve terminal.

It is well established that nerve-evoked quantal release of neurotransmitter
is dependent upon the influx of extracellular Caz+ through membrane ionic
channels (Katz and Miledi, 1965; Katz and Miledi, 1967a; Dodge 91 al., 1969). In
the absence of 0a“, the nerve action potential invades the nerve terminal but fails

to evoke transmitter release (Katz and Miledi, 1965; Dodge and Rahamimoff, 1967).

84

Transmission at many synapses has a steep dependence on extracellular Ca2+
concentration (Dodge and Rahamimoff, 1967; Katz and Miledi, 1970; Drapeau and
Blaustein, 1983; Smith et al., 1985; Zucker and Fogelson, 1986). These
observations have led to the hypothesis that Ca2+ ions act in a cooperative,
exponential manner at intracellular sites to trigger secretion of transmitter quanta.
The power index appears to vary considerably at different synapses and it has
been reported that transmitter release following a nerve impulse may increase with
as much as the fourth or even higher power of [Ca2+]. (Dodge and Rahamimoff,
1967; Katz and Miledi, 1970; Zucker and Fogelson, 1986). This power relationship
may be of functional importance in controlling transmitter release. Changes in the
magnitude of presynaptic Ca2+ entry could have an amplifying effect on the
amount of neurotransmitter secreted (Smith et al., 1985). Thus, if transmitter
secretion at the neuromuscular junction and/or in synaptosomes follows this
power relationship then modest changes in Ca2+ influx, such as those in this study,
could conceivably result in large changes in neurotransmitter secretion.

Rat cortical synaptosomes are neurochemically heterogeneous with only
10% percent of the suspension thought to be cholinergic (Gray and Whittaker,
1962). Incubation with LEMS sera/plasma resulted in decreases of “Ca2+ uptake
which ranged from 0 to 60%. These decreases suggest that the autoantibody must
interact with nerve terminal Ca2+ channels which secrete different
neurotransmitters. These data also provide further evidence for the similarity

between somatic motor nerve terminal and central nerve terminal Ca2+ channels.

85
Reduction in 10 sec depolarization-dependent “Ca2+ uptake into

synaptosomes was evident following 1 hr of incubation of plasma/sera. Thus,
acute exposure to LEMS autoantibody was sufficient to induce a change in activity
associated with Ca2+ channels. This is in contrast to other models of the disease
which have utilized chronic exposure. Using this acute, in vitro system may allow
for more efficient use of the isolated immunoglobulin fraction from patients sera,
in subsequent studies. This is desirable as patient sera is often difficult to acquire
as the incidence of the disorder is low.

In conclusion, incubation of rat cortical synaptosomes with sera/plasma
from patients with LEMS but not SCC reduced depolarization-dependent uptake
of 4"’Ca'“ This was not due to decreased synaptosomal viability. The results from
this study are significant as they demonstrate that acute rather than chronic
exposure to LEMS sera/plasma is sufficient to inhibit Ca2+ channel activity in
isolated nerve terminals; that data obtained from patients with LEMS with and
without SCC can be analyzed identically; and that Ca“ channels of nerve terminals
secreting different transmitters may share common epitopes recognized by the
LEMS autoantibody. Thus, the synaptosomal preparation may a useful model in
which to study, in more detail, the potential interaction of the autoantibody in LEMS

with calcium channels involved in transmitter release.

CHAPTER FOUR

SPECIFICITY OF LAMBERT-EATON MYASTHENIC SYNDROME

IMMUOGLOBULIN FOR NERVE TERMINAL CALCIUM CHANNELS

86

87
A. Summary
Data from Chapter Three provide evidence which suggests that synapto-

somal Ca2+ channel function is altered by incubation with serum /plasma from
patients with LEMS. The goal of the present study was to examine the specificity
of LEMS autoantibodies for nerve terminal Caz+ channels compared to other
voltage-sensitive ion channels in nerve terminals, and to determine if non-specific
membrane damage contributed to the pathogenesis of LEMS. The ion channel
specificity of LEMS autoantibody was assessed by comparing the ability of acute
application of lgG isolated from the plasma of a patient with LEMS to reduce depo-
larization-dependent uptake of “Ca“ and ”Na” into or efflux of “RV from rat
forebrain synaptosomes. The clinical diagnosis of LEMS was confirmed electro-
physiologically by treatment of mice for 30 days with plasma (1.5 mI/day) taken
from this patient. Characteristic reduction of quantal content elicited at 1 Hz and
facilitation at 20 Hz was observed in mice treated with LEMS plasma compared to
those treated with control plasma. One sec, K*-stimulated 45Ca2+ uptake was
inhibited 36.5 1 14.5% and 44.5 1 9.8% by acute application of 2 and 4 mg/ml
LEMS lgG, respectively; lgG from patients with SCC had no effect on 45Ca2+ entry.
The same concentrations of LEMS lgG affected neither voltage-dependent uptake
of 22Na* into veratridine-depolarized synaptosomes nor a“Rb" efflux from K"-
depolarized synaptosomes. LDH release from synaptosomes incubated with
varying concentrations LEMS IgG was not increased over control indicating the

reduction in 45Ca2+ uptake was not a result of disruption of synaptosomal

88

membrane integrity. Likewise, normal activation of Ca2+ channels was not pre-
vented by IgG-induced alterations in synaptosomal membrane potentials as the
fluorescence of a membrane potential sensitive carbocyanine dye (diS-C2(5)) was
unaltered following acute application of 2 mg/ml LEMS or control lgG. These
results indicate that immunoglobulins isolated from a patient with LEMS alter
presynaptic nerve terminal Ca2+ channel activity. These deficits are not due to
decreases in synaptosomal viability nor to lgG-induced alterations in the
synaptosomal membrane potential but appear to be a direct block or modulation
of voltage-dependent Ca2+ channels. The antibody from this patient appears to be
specific for nerve terminal Ca2+ channels as it did not affect flow through voltage-

dependent Na+ or K+ channels.

B. Infmducfion

The electrophysiological features of LEMS include reduced quantal content
at low frequencies of nerve stimulation with marked facilitation at high frequencies
of stimulation (Elmqvist and Lambert, 1968; Lambert and Elmqvist, 1971; Cull-
Candy et al., 1980). Morphologic data suggest that AZPs represent the target of
the pathogenic autoantibodies in LEMS (Fukunaga et al., 1982; 1983; Fukuoka et
al., 1987). Mice injected with either plasma or immunoglobulin derived from LEMS
patients (Kim, 1985:1986; Lang et al., 1983) reproduce both the electTOphy-
siological and morphological features of LEMS.

It has been proposed that AZPs correspond to voltage-sensitive Ca2+

89
channels of the presynaptic membrane (Pumplin etel., 1981; Robitaille at al., 1990;

Tarelli at al., 1991; Cohen et al., 1991). Thus, the reduced transmitter release in
LEMS could be related to a reduced entry of Ca2+ into the nerve terminal. In
support of this contention, LEMS autoantibodies appear to interact with voltage-
dependent Ca2+ channels in several cell lines of diverse origin (Roberts et al.,
1985; Login et al., 1987; De Azipurua etel., 1988a, 1988b; Kim and Neher, 1988;
Lang et al., 1989a; Peers et al., 1990). Although these studies have proven useful,
several important deficiencies still remain. first, the identity of Ca2+ channels in
motor nerve terminals remains unclear. Ca2+ channel subtypes in nerve terminals
or populations of nerve terminals may differ from those in the cell body (Suszkiw
et al., 1986; Charlton and Augustine, 1990). Second, the function and number of
ion channels could be affected by damage to the membrane or membrane
depolarization. Third, no assessment has been made of the Ca2+ channel
specificity of the LEMS autoantibody as opposed to other voltage-gated channels
present in nerve terminal. An antibody directed at an epitope on voltage-dependent
Ca2+ channels might also affect other voltage-dependent ion channels if the lgG
recognizes portions of the channel proteins common to all ion channels.

Data from the previous chapter indicates that serum and plasma from
patients with LEMS reduces depolarization-dependent uptake of “Ca2+ into
synaptosomes (Hewett and Atchison, 1991). As an lgG autoantibody is presumed
to mediate the pathogenesis of LEMS, the present study was designed to answer

the following three questions: does lgG isolated from a patient with LEMS reduce

90
depolarization-dependent “Ca2+ into nerve terminal Ca2+ channels?; if so, does

the lgG cause Caz+ channel dysfunction secondary to nonspecific membrane
damage or alterations of membrane potential?; and are other voltage-dependent

ion channels in the terminal also affected?

C. Materials and Methods

1. Materials

Dimethyl sulfoxide (DMSO) was obtained from EM Science (Gibbstown,
NJ). Tetraethylammonium chloride (TEA), tetrabutylammonium chloride (TBA),
tetrodotoxin (TTX), bovine serum albumin (BSA) cyclophosphamide, oligomycin,
nickel chloride (NiCIz), ethyleneglycol-bisw-aminoethylether) N,N,N',N’-tetraacetic
acid (EGTA), and ouabain were purchased from Sigma Chemical Co. (St. Louis,
MO.) Sodium cyanide (NaCN) was obtained from Mallinckrodt Chemical Works (St.
Louis, MO). Triton-X-100 was obtained from Research Products Intemetional Corp.
(Elk Grove, IL). The radiolabels 4“Caz" (15.50 mCi/mg), at’Rb“ (1-10 mCi/mg), and
”Na“ (800-1000 mCi/mg) were purchased from New England Nuclear (Boston,
MA). 3,3’-Diethyldicarbocyanine iodide [dis-02(5)] was obtained from Molecular
Probes (Eugene, OR). Disease-free control human plasma was purchased from the
American Red Cross (Lansing, MI). Control human serum was donated by a
healthy volunteer. Plasma from patients with the clinical and electromyographic
characteristics of LEMS was provided by Drs. Mark Glasberg, Eva Feldman,
Andrew Massey, and James Albers. Serum from patients with SCC was provided

by Dr. David Johnson.

91
2. Solutlons

For electrophysiological measurements of neuromuscular function, solutions
used for both bathing and recording contained (mM): NaCl, 137.5; KCI, 2.5; MgCIz,
1; CaClz, 2; d—glucose, 11; and HEPES, 14. The solutions used in the “Ca2+
uptake experiments were identical to those described in Chapter Three.

For experiments using ”Na“, incubation buffer contained (mM): KCI, S;
MgClz, 1; choline chloride, 145; d-glucose, 10; HEPES, 10; veratridine (VER), 0.15;
ouabain, 0.1; BSA, 1 mg/ml; 1 TTX, 0.001. Uptake solutions were identical to the
incubation buffer except that 2 mM NaCl and ”Na” (2 irCi/ml) were added and the
ouabain concentration was increased to 1 mM. Cold quench solution was identical
to that described in Chapter Three. Veratridine was used in lieu of K*-depolar-
ization because voltage-dependent Na+ channels inactivate rapidly in response to
normal depolarizing stimuli (T amkun and Catterall, 1981).

The wash buffer used in the E"’Rb‘ efflux experiments contained (mM): NaCl,
145; KCI, 5; MgCIz, 2; RbCl, 0.1; d-glucose, 10; and HEPES, 10. Nondepolarizing
efflwr buffer was identical to the wash buffer. In the depolarizing efflux buffer, NaCl
was replaced by KCI for a final [K‘], of 150 mM. Cold stop solution contained
(mM); TEA, 70; TBA, 10; NiCIz, 10; RbCl, 5; MgCIz, 1; d-glucose, 10; HEPES, 10;
and N-methyI-d-glucamine, 50.

Stock solutions of dis-02(5) for fluorescence measurements were dissolved
in DMSO. Working solution were made daily by dilution of the stock into physiologi-

cal saline buffer (final concentration, 0.5 uM). The final concentration of DMSO

92

after addition of dye to synaptosomes was 0.05% (v/v).

3. Immunoglobulln Isolation

Control, SCC and LEMS sera and/or plasma were stored at -20°C until
used. After thawing, fibrin was removed from the plasma samples by clotting which
was facilitated by the addition of 1 M CaClz. Total immunoglobulin (Ig’s) were
isolated by serial precipitations of serum in 30% and 50% saturated ammonium
sulfate at 4°C as described by Heide and Schwick (1978). Each precipitate was
pelleted by centrifugation at 1000xg for 15 min and redissolved in 0.85% (w/v)
saline. The final precipitate was redissolved to 1/2 the original volume and
dialyzed at 4°C against saline using Spectrapor 2 dialysis tubing (15-25000 m.w.
cutoff, Spectrum Medical lnd., Los Angelese, CA) until the sulfate ions were no
longer detectable (at 72 hr) as discerned by the barium chloride test (Garvey et
al., 1977). fiocculent material formed during dialysis was removed by
centrifugation at 1000xg for 15 min at 4°C. Ig’s were quantitated by bicinchoninic
acid protein assay (Pierce Chemical, Rockford, IL). The lgG concentration was
presumed to be 80% of the total protein (Kimbal, 1986).

The LEMS lgG used in the synaptosomal studies was isolated from a patient
with SCC. We show here that this patients’ plasma caused the characteristic
reduction of evoked release of ACh at the neuromuscular junction (Patient 3; fig.

4.2).

93

4. Measurement of Ouantal Content (m)

Plasma was used from three patients with typical clinical and
electromyographic features of LEMS. One LEMS patient (Patient 3) has SCC.
Plasma fractions (1.5 ml) from these patients and from disease-free human
controls were injected daily intraperitoneally into Swiss Webster female mice (15-20
g) (Kim, 1985). In order to induce Immune tolerance to the foreign protein, animals
were given cyclophosphamide, 300 mg /kg Up), one day following the first plasma
injection (Lang et al., 1983). On day 30, the diaphragm with phrenic news intact
was isolated and was perfused continuously with a low K” (2.5 mM) saline solu-
tion continually aerated with 100% 02. Muscle fibers were transected approximately
4 mm on either side of the main intramuscular nerve branch to prevent
contractions due to stimulation of the phrenic nerve (Barstad and Lilleheil, 1968;
Glavinovic, 1979). Preparations were pinned out in a Sylgerd-coated plexiglass
chamber and perfused with low [K*] saline. All experiments were performed at
room temperature (~25°C) and replicated in a minimum of four animals.

The setup for recording under current clamp is depicted schematically in
figure 4.1 . Intracellular microelectrode recordings of synaptic potentials were made
using conventional techniques and borosilicate glass microelectrodes (1.0 mm,
WPI, New Haven, CT) having impedences of 15-30 meg it when filled with 3 M KCI.
Prior to nerve stimulation, an average MEPP amplitude was obtained by recording
5 min of MEPPs on magnetic tape using an FM instrumentation tape recorder.

EPPs were elicited by single shock rectangular voltage stimuli of 0.07 msec

 

 

 

 

 

 

 

 

 

 

 

 

figure 4.1. Setup for recording under current clamp. Hemidiaphragm preparations
with phrenic nerve intact were removed from control- and LEMS-treated mice.
Preparations were transected to prevent contraction following stimulation of the
nerve. Muscles were pinned down in a Sylgerd-coated plexiglas chamber and
perfused with Low K“ buffer. End-plate regions were impaled with a KCL-filled
microelectrode and EPPs and MEPPs recorded.

figure modified from that contained in the dissertation of John M. Spitsbergen

95
duration applied to the phrenic nerve at a frequency of 1, 20, or 40 Hz (200

stimuli) allowing 5 min of rest between pulse protocols for redistribution of Ca2+.
Ouantal content was determined off-line from the amplitudes and variances of the
EPPs (variance method) and by dividing the average amplitude of EPPs by the
average MEPP amplitude (mean value method). EPP amplitudes were standardized
to a membrane potential of -40 mV (Katz and Theseleff, 1967). Only those EPPs

and MEPPs with rise time s 1 msec were used for subsequent analysis.

5. Preparatlon of Synaptosomes

Synaptosomal preparation was identical to that previously described in
Chapter Three. In all experiments except “RN efflux, synaptosomes were
incubated at 4°C under 100% 02 for 60 min with 10% (v/v) control human serum
and varying concentrations of IgG from patients with no known neurological or
immunological illness (control), SCC or LEMS. The rationale for using control
human serum will be discussed in Chapter five. The proper volume of physiologic
saline buffer was added to maintain a constant incubation volume. Prior to uptake

measurements, synaptosomes were allowed to equilibrate for 30 min at 37°C.

6. Measurement of “Ca“ Uptake
Uptake of “Ca2+ was measured as described in detail in Chapter Three.
Readings of dpm were corrected back to to the original volume of synaptosomes

filtered and expressed as nmole “Ca“ /u g protein or as % of disease-free control.

96
7. Measurement of 22m“ Uptake

The time course of uptake of 22Na“ into rat forebrain synaptosomes was
measured by the addition of 50 ul of synaptosomes to 50 ul of incubation buffer
containing 150 uM VER with or without TTX (1 uM) for varying periods of time.
VER is used to activate synaptosomal Na+ channels as these channels inactivate
rapidly in response to K*-depolarization (Kreuger and Blaustein, 1980; Tankum and
Catterall, 1981). NaCl was replaced isotonically with choline chloride in all solutions.
Ouabain was present during both the incubation and uptake periods to prevent
22Na+ extrusion by blocking the Na”/K” ATPase. Following this incubation, a 100
pl of uptake solution containing 22Na+ (2 uCi/ml) was added. After 5 sec of incuba-
tion, 2 ml of cold quench solution was added followed by suction-filtration. Uptake
of ”Na" due to activation of voltage-dependent Na+ channels was determined as
the difference of uptake in the presence and absence of TTX. Readings of dpm
were corrected back to the original volume of synaptosomes filtered and
expressed as nmole 22Na“/ug protein.

22Na‘ uptake following incubation with control and LEMS lgG (2 and 4
mg/ml) was measured as described above except that the synaptosomal
suspension was added to the incubation buffer for 10 min only prior to the addition

of uptake solution.

8. Measurement of l"’Flb“ Loadlng and Efflux
Loading of synaptosomes with ”Rb“ and efflux under depolarizing
conditions were done using a modification of the method of Bartschat and

97
Blaustein (1985a). Synaptosomes were first incubated for 45 min at 4°C with the

following: (1) ”Rb“ (20 uCi/ml) alone; (2) a"Rb“ plus 10% (v/v) control human
serum; (3) ”Rb", 10% (v/v) control human serum and control lgG (4 mg/ml); or
(4) 13‘int)“, 10% (v/v) control human serum and LEMS IgG (4 mg/ml). Prior to the
initiation of m5Rb+ release, synaptosomes were incubated for an additional 45 min
at 37°C. It is during this time that the synaptosomes became loaded with the
radiolabel. Next, 100 pl of labeled synaptosomes were placed on a filtration
apparatus containing 2 ml of a wash solution and suction applied. Synaptosomes
were washed in this manner with a total of 10 ml of washing solution to remove
any extracellular “RV. Following the final wash, the filter was removed to a
second filtration apparatus and 2 ml of depolarizing (77.5 mM) or resting (5 mM)
K+ solution was added. Five sec later, 2 ml of stop solution was added and suction
applied. The filtrate was collected directly. Radioactivity in the filtrate and that
remaining on the filter was estimated by gamma counting. Efflux in low [K‘]. was
subtracted from that in high [K], solutions as depolarization-independent back—
ground. In order to control for possible differences in ”RV loading, efflux is

expressed as a % of total.

9. Measurement of Lactate Dehydrogenase (LDH) Actlvlty
LDH release was quantified by the method of Bergmeyer and Bernt (1974),
as described in detail in Chapter Three. Activity was expressed as the percentage
of LDH released into the supernatant. Total LDH activity (100%) was determined

by lysing an equal volume of synaptosomes in 0.1% (v/v) Triton-X-100.

98

1o. Fluorescence Measurements

Measurements of dis-02(5) fluorescence as an index of synaptosomal
membrane potential were made as described by Hare and Atchison (1992). Briefly,
a 50 ul aliquot of prewarmed synaptosomal suspension was added to 2 ml
physiologic saline buffer containing 0.5 uM dis-02(5) and maintained at 37°C for
10 min. Determination of diS-Cz(S)-mediated fluorescence in synaptosomes was
made using a SPEX fluorolog spectrofluorometer (SPEX Industries, Edison, NJ) at
excitation and emission wavelengths of 649 and 682 nm respectively. In order to
measure plasma membrane potential fluorescence independent of mitochondrial
fluorescence, NaCN (2 mM) and oligomycin (4 ug/ml) were added to the
suspension to collapse the mitochondrial membrane potential(Heinonen er al.,
1985; Hare and Atchison, 1992). Once baseline fluorescence was judged to be
stable, KCI (40 mM) was added to depolarize the plasma membrane. Data are

expressed as the percent change from baseline fluorescence.

11. Statlstlcal analysls

Differences in quantal content in animals receiving control or LEMS plasma
were analyzed by one way analysis of variance (ANOVA) followed by Dunnett’s t-
test. Uptake of 4"’Ca“ and 22Na+ were analyzed using Student’s t-test for paired
samples. Efflux of “RV and fluorescence measurements were analyzed by a
completely blocked ANOVA. The LDH experiments were analyzed by split-plot

design ANOVA (2X5). Difference were considered significant at p < 0.05.

D. Results

1. Quantal Content

To verify experimentally the clinical diagnosis of LEMS, and thus correlate

potential actions on synaptosomal Ca2+ channels with clinical neuromuscular
effects, the ability of LEMS patient plasma to reduce quantal content when given
chronically was examined. Quantal content of nerve-evoked EPPs was reduced
significantly in mice receiving chronic injection of plasma from three patients
clinically diagnosed with LEMS. Mean quantal content as determined by the
variance method was reduced from s: 70 in diaphragms from control mice to 28-19
in diaphragms from mice receiving LEMS plasma (Fig. 4.2). This represents a
reduction of approximately 60 to 70%. Similar percent reductions were observed
using the mean value method (Table 4.1). MEPP amplitude was unaffected by
treatment with LEMS plasma (Table 4.1). Cyclophosphamide alone had no effect
on quantal content (results not shown).

Facilitation of quantal content at 20 Hz was observed in diaphragms from
mice treated with LEMS plasma from all three patients. Quantal content in mice
treated with plasma from two of three patients facilitated at 40 Hz. Facilitation was
not observed in mice treated with control plasma (Fig. 4.3).

2. “Ca” Uptake

The effect of LEMS lgG from Patient 3 on uptake of “Ca2+ during 1 sec of
K*-depolarization into rat forebrain synaptosomes is depicted in Figure 4.4.

Depolarization-induced uptake was inhibited 36.5 2 14.5%, 44.5 i 9.8% and 34.6

100

vvvvvvv

100
'—
z
In
'—
z
8
_, so
I5 * * *
z
<
D
O

 

 

   

 

Control LEMST LEM82 LElvlss
TREATMENT

Figure 4.2. Effects of chronic treatment of mice with LEMS plasma on quantal
content. Mice were treated for 30 days with 1.5 ml/day 0p.) of plasma from
control or one of 3 different patients with the clinical diagnosis of LEMS.
Neuromuscular transmission was examined using conventional intracellular
recordings of MEPPs and EPPS elicited at 1 Hz from isolated hemidiaphragm
preparations. Contractions evoked by phrenic nerve stimulation were prevented by
transsection of the myofibers. Quantal content (m) was determined from the
amplitudes and variances of the EPPs. At least 200 EPPs were used in the
calculation of m. Values are the mean 2 SEM of at least four animals. Asterisks
indicate values of quantal content which are significantly lower than control (p <
0.05).

101

Table 4.1
Determination of Quantal Content via Direct Method

l

 

m MEPP (mV)
Control 9.75 z 2.2 0.353 t 0.05
Patient 1 3.92 2 1.7‘ 0.335 a 0.14
Patient 2 5.78 a 1.2‘ 0.328 .+. 0.04
Patient 3 3.33 .+. 0.51' 0.350 a 0.06

 

Mice were treated for 30 days with 1.5 ml/day 0p.) of plasma from control
or one of 3 different patients with the clinical diagnosis of LEMS. Neuromuscular
transmission was examined using conventional intracellular recordings of MEPPs
and EPPS elicited at 1 Hz from isolated hemidiaphragm preparations. Quantal
content (m) was determined by dividing the average amplitude of EPPs by the
average amplitude of the MEPPs. Values are the mean 1 SEM of at least four
animals. Asterisks indicate values of quantal content which are significantly lower
than control (p < 0.05).

102

- Control. LEMS1[--] LEMS 2|: LEMS 3

 

 

 

 

 

 

 

 

 

 

 

  

 

 

 

 

€ aoo~
5
C
o ..
o zoor
ax°
v nF--__
5 too-
"" V
2 ’/
': 0
d
o
g -1oo-
20 Hz 40 Hz

STIMULATIONFREQUENCY

Figure 4.3. Frequency-dependent facilitation of neuromuscular transmission in mice
treated chronically with LEMS plasma. Treatments and recording procedures were
as described in Figure 4.2. EPPs were elicited at the indicated frequencies of
stimulation. Facilitation is defined as an increase in quantal content compared to
that obtained at 1 Hz.

103

125 '

N'—

‘5 Ca” UPTAKE
of control)
\I
01

 

 

 

 

 

 

 

 

 

 

O

.xmfi
°° “N—

"S

2 4
[lgG]
(mg/ml synaptosomal suspension)

Figure 4.4. Effect of LEMS lgG on uptake of “Ca2+ into rat forebrain synapto-
somes. Synaptosomes were incubated for 1 hr at 4°C with varying concentrations
of LEMS or control lgG and 10 96 control human serum. Prior to initiation of
uptake, synaptosomal suspensions were equilibrated at 37°C for 30 min. One sec
uptake in 5 mM K+ was subtracted from that in 42.5 mM K+ as depolarization-
independent background. “Ca2+ uptake is expressed as 96 of control. All values
are the mean 1- SEM of 7 experiments with each individual experiment being the
average of 3 measurements. Asterisks indicate values which are significantly lower
than their respective paired control (p < 0.05).

104
x 19.8% by 2, 4 and 8 mg/ml LEMS lgG respectively, as compared to

synaptosomes incubated with equivalent concentrations of control lgG. Depolariza-
tion-independent uptake was unaffected (results not shown).

A number of LEMS patients, including the patient used in the previous and
proceeding studies have 300. Uptake of “Ca“ was determined following
incubation with SCC lgG pooled from a number of patients in which no
neurological involvement was observed to control for possible effects associated
with neoplastic disease. SCC lgG (2 and 4 mg/ml) produced no reduction in

depolarization-dependent uptake of “Ca2+ (Fig. 4.5).

3. Synaptosomal Membrane lntegrlty

To rule out the possibility that the effect of LEMS lgG was a result of
nonspecific disruption of the synaptosomal membrane, LDH release following
incubation with varying concentrations of LEMS and control IgG was measured.
LDH released into the extrasynaptosomal fluid following incubation with LEMS lgG
was not increased over that which occurred with control lgG (Fig. 4.6). Thus, it is

+

unlikely that decreased Ca2 uptake is due to disruption of synaptosomal

membrane integrity.

4. Potentiometrlc Dye Fluorescence
Alterations in the synaptosomal membrane potential by lgG could prevent
normal activation of the Ca2+ channels. We tested for this possibility using the

fluorescent, potentiometric carbocyanine dye dis-Cz(5). NaCN (2 mM) and

105

'3‘

con sec

4 i

" Ca ‘* UPTAKE
(nmollug proteln 505)
.°
8

 

 

 

 

 

 

 

 

 

 

W
“m

0 2
I G
(ITIMM syn-small WWO")

Figure 4.5. Effect of lgG from patients with small cell carcinoma (SCC) on uptake
of “Ca“ into rat forebrain synaptosomes during 1 sec of depolarization by 41.25
mM KCI. Synaptosomes were incubated for 1 hr at 4°C with 10% control human
serum and control or SCC lgG. Prior to initiation of uptake, synaptosomal
suspensions were equilibrated for 30 min at 37°C. Uptake in 5 mM K+ was
subtracted from that in 42.5 mM K” as depolarization-independent background.
Data are expressed as nmol 45Ca/lig protein. All values are the mean 1 SEM of
4 experiments with each individual experiment being the average of 3 measurements.

106

100 r

-0— CONTROL
"I“ LEMS

50

 

LDH RELEASE
(% of total)

0 1 2 4 8

(mg/ml synaptosomal Mien)

Figure 4.6. Effect of LEMS lgG on synaptosomal membrane integrity. LDH release
was measured from the extrasynaptosomal fluid following treatment for 90 min with
10 % control human serum and varying concentrations of LEMS or control lgG.
LDH activity was quantified by the rate of NADH oxidation at 340 nM. The quantity
of the enzyme is expressed as the percent of release of the total enzyme content
of the terminals, which was determined by lysis of equal volume of synaptosomes
with 0.1% triton-X-100 followed by sonication. Values are expressed as the mean
i SEM of 4 experiments.

107
oligomycin (4 lug/ml) were first added to the suspension to collapse the

mitochondrial membrane potential (Fig. 4.7). LEMS lgG caused no alterations in

dye fluorescence following addition of KCI (40 mM) (Fig. 4.8).

5. ”Na“ Uptake

In order to determine the length of time the synaptosomes should be
incubated with VER, a time course of ”Na" uptake was performed. The reults are
depicted in Figure 4.9. Five sec ”Na" uptake into synaptosomes is maximal
following 10 min incubation with incubation buffer containing 150 AM VER. These
results are similar to those of Tankum and Catterall (1981).

Incubation of synaptosomes with concentrations of LEMS lgG which
significantly altered uptake of “Ca2+ during K*-evoked depolarization had no effect
on depolarization-induced uptake of 22Na+ into VER-treated synaptosomes (Fig.

4.10).

e. “30* Loading and Efflux

Treatment of synaptosomes with 10 % (v/v) control human serum or serum
and either control or LEMS lgG (4 mg/ml) did not alter loading of a"Rb“ into rat
forebrain synaptosomes (Fig. 4.11). Likewise, incubation of synaptosomes with
concentrations of LEMS lgG which significantly altered uptake of 45Ca2+ during K”-
evoked depolarization had no effect on “RE efflux from K*-depolarized synapto-

somes (Fig. 4.12).

108

 

 

 

 

E 1.50 r

3 E _

3‘

2 71‘

3 r: -s

f £5 0.75 f t

2 5 K’

3 v . l

8 cu-

; ollgo

E 0.00 ‘ ' ‘ '
0 200 400

Time (sec)

Figure 4.7. Representative graph depicting typical fluorescent dye experiment. An
aliquot of untreated synaptosomal suspension was added to 2 ml of physiologic
saline buffer containing 0.5 uM diS-C2(5) and maintained at 37°C for 10 min. In
order to measure fluorescence derived from the plasma membrane independent
of mitochondrial fluorescence, NaCN (2 mM) and oligomycin (4 pg /ml) were added
to the synaptosomal suspension to collapse the mitochondrial membrane potential.
Once baseline was judged to be stable (visibly), KCI was added to depolarize the
plasma membrane. Data expressed in counts per second.

 

 

109

100

 

 

 

FLUORESCENCE
(% charge from baseline}
01
O

 

 

 

 

 

O

PLASMA MEMBRANE

Figure 4.8. Effect of LEMS lgG on synaptosomal membrane potentials.
Synaptosomes were incubated with 2 mg/ml of control or LEMS lgG, or no lgG
for 90 min. Following incubation, synaptosomes were combined with a fluorescent,
potentiometric dye, dis-C2(5). Initially, NaCN (2 mM) and oligomycin (4 ug/ml)
were added to collapse the mitochondrial membrane potential. Addition of 40 mM
KCI depolarized the plasma membrane potential. Data are expressed as % change
in baseline fluorescence. Values are the mean 2 SEM of 3 experiments.

110

12l

22 Na * LPTAKE
(W90 m 3'00)
O)

 

‘o ‘ s 10‘ 20
Time (min)

Figure 4.9. Time course of 22Na+ uptake into rat forebrain synaptosomes.
Synaptosomes were added to incubation buffer which contained no added Na‘,
150 MM VER, 100 0M 1 TTX 1 uM) for varying periods of time. After incubation,
uptake solution containing Na+ was added and allowed to mix for 5 sec
subsequent to addition of cold quenching solution. Uptake is expressed as nmol
22Na"/ug protein. Values are the mean 2 SEM of 4 experiments.

111

CZICON -LEMS

2"’ Na * UPTAKE
W m m
.p.
. |-'
|—_.
“—1

 

 

 

 

 

 

 

 

 

0

im

0 2
[I96]

(mg/ml synaptosomal suspension)

Figure 4.10. Effect of LEMS lgG on uptake of 22Na+ into rat forebrain synapto-
somes depolarized with veratridine. Synaptosomes were incubated with lgG and
serum as in Fig. 4.4 after which they were combined with veratridine (150 uM) for
10 min in a solution in which the NaCl had been replaced with choline chloride.
Synaptosomes were then incubated with 22N3“ for 5 sec. Uptake of 22Na+ due to
activation of voltage-dependent Na channels was determined as the difference of
uptake in the presence and absence of tetrodotoxin (TTX; 1 14M). Values are the
mean 1 SEM of 4 separate experiments.

112

fi

[:1 0011 tails

 

l

 

"Na’ UPTAKE
(nmollug protein E-OB)
r.

 

 

 

 

 

W

 

 

 

0 2 4

['99]
(I'IIII synaptosomal suspension)

figure 4.1 1. Time course of “RV loading into synaptosomes. Synaptosomes were
treated for 1 hr as described in methods. Prior to addition of radiolabel,
synaptosomes were allowed to equilibrate at 37°C for 30 min. ”RH (20 uCi/ml)
was added to the synaptosomal suspension and loading allowed to proceed for
varying periods of time. After the specified time, 2 ml of cold quench solution was
added to terminate loading. Uptake is expressed as nmol m3Rb“/ug protein.

113

 

    

  

 

 

 

 

 

 

E Con lgG
o l s ‘
t‘ 39"“
“- 15 ' Serum
0 None only LEMS+lgG
98 ' __, Serum
§ 10 ' ‘

'O'O'OV
“E l % 335:3:
LU . /
+ 5 /
.0 /
‘I z
3 ° 2 3

 

Treatment

figue 4.12. Effect of LEMS lgG on ”Rb“ efflux from rat forebrain synaptosomes.
Following incubation and removal of residual extracellular ”RN, efflux of 1”Pb“
was measured by addition of depolarizing and nondepolarizing K+ solutions to the
filter well. After 5 sec, cold quenching solution was added followed by suction
filtration. Radioactivity in the filtrate and that remaining on the filter was
determined. Efflux in low [K‘], solutions was subtracted from that in high [K‘].
solutions as depolarization-independent background. Effiux is expressed as % of
total. Values are the mean i SEM of 4 separate experiments.

114
E. Discussion

The aim of the present study was to test explicitly whether LEMS
autoantibody affects nerve terminal Ca2+ channels and whether this action was
direct and/or specific. In particular, we sought to determine if effects of LEMS lgG
on nerve terminal function were related to a generalized membrane disturbance,
and if not, whether the lgG recognized other voltage-dependent ion channels in
axon terminals. This is important as nonspecific membrane alterations and/or
alteration in function of other ion channels could result indirectly in Ca2+ channel
dysfunction. Our results indicate that lgG from a patient with LEMS acts on
presynaptic nerve terminal Ca2+ channels in the CNS to reduce depolarization-
dependent uptake of 45Ca2+. The antibody from this patient appears to be specific
for Ca2+ channels as it is unable to disrupt ion flow through voltage-dependent
Na“ or K+ channels. These deficits are not due to decreases in synaptosomal
viability nor to lgG-induced alterations in the synaptosomal membrane potential but
appear to be a direct block of or alteration in voltage-dependent Ca2+ channel
function.

Passive transfer experiments demonstrated that the plasma from the three
patients studied produced the electrophysiological characteristics associated with
LEMS. Quantal content of the EPP was decreased when the nerve was stimulated
at low frequencies of stimulation, but was facilitated at high frequencies of stimu-
lation. MEPP amplitude was unaffected. These transmission characteristics are

qualitatively similar whether the disease is transferred passively by plasma, crude

115
lg’s or purified lgG (Newsom-Davis et al., 1982; Lang et al., 1983; Kim, 1985; 1986;

Lennon and Lambert, 1988). Passive transfer experiments were useful in
determining that a presynaptic abnormality existed in LEMS. However, they
represent an indirect measurement of nerve terminal Ca2+ channel activity as the
postsynaptic response measured results from many processes ongoing in the
presynaptic nerve terminal.

Likewise, Meyer et al. (1986) and Login et al.(1987) showed that serum
and serum fractions from patients with LEMS reduced ACh release from
synaptosomes and growth hormone and prolactin release from rat anterior pituitary
cells respectively. However, no measurement of ion channel activity was made
(Meyer et al., 1986; Login et al., 1987). Still, others have shown that LEMS lgG
reduced the function of somal Ca2+ channels in human soc (Roberts et al., 1985;
De Azipurua et al., 1988a,b), rodent neuroblastoma X glioma hybrid (Peers et al.,
1987) , bovine adrenal chromaffin (Kim and Neher, 1988), and rat pheochro-
mocytoma (PC12) cells (Lang et al., 19893,b). It is quite possible, however, that
the type of Ca2+ channel in the terminal may differ from that in the cell body
(Nachshen and Blaustein, 1979; Suszkiw et al., 1989; Charlton and Augustine,
1990; Utchitel et al., 1992). Also, the binding of 01-0ng to somal Ca”+ channels
has been used as an index of localization of LEMS autoantibodies (Sher et al.,
1989), although «.1me does not block mammalian neuromuscular transmission
(Anderson and Harvey, 1987; Protti et al., 1991) whereas the LEMS autoantibody

clearly does (Lambert and Elmqvist, 1971; Lang et al., 1983; Kim, 1985:1986).

116
We demonstrated previously that serum and plasma from patients with

LEMS reduced depolarization-dependent uptake of “Ca“ (10 sec) into
synaptosomes (Hewett and Atchison, 1991; Chapter Three). In the present study
the interaction of isolated LEMS lgG with nerve terminal Ca2+ channels thought to
be involved in neurotransmitter release was examined directly. Reduction in
depolarization-dependent uptake of 45Ca2+ during 1 sec was evident following 90
min incubation with LEMS lgG. This 'fast' phase of uptake (1 sec) into synapto-
somes has been consistently correlated with neurotransmitter release from
synaptosomes (Drapeau and Blaustein, 1983; fioor, 1983; Leslie et al., 1985).

lgG from SCC patients lacking neurological symptoms did not reduce
depolarization-dependent uptake of “Ca” into synaptosomes. Thus, the presence
of a tumor alone is not sufficient to result in nerve terminal Ca2+ channel
dysfunction. This is not unexpected, as many patients with SCC do not have the
manifestation of LEMS. Nonetheless, interpretation of data obtained by using lgG
from a LEMS patient with SCC is not complicated by actions of lg’s in the plasma
of patients with SCC.

+

The decrease in “Ca2 uptake into synaptosomes was not a result of
disruption of synaptosomal membrane intergrity. LDH is a cytosolic enzyme that
can be released into the extrasynaptosomal fluid if the plasma membrane is per-
meabilized. Its release following incubation with LEMS lgG was not increased over
that of paired controls, so it is unlikely that the decrease in Ca2+ uptake was due

to nonspecific membrane damage. It remains possible that the LEMS lgG causes

117

membrane damage in only a fraction of the terminals and the release of LDH from
the whole population may not be sufficiently sensitive to detect this. However, in
view of the lack of discrimination of the LEMS lgG for L- and N-type channels in
other systems, and the magnitude of decreases seen with both serum/plasma
(Hewett and Atchison, 1991) and lgG (present study) in synaptosomes, it seems
unlikely that a specific population of terminals would be targeted.

lgG-induced alterations in the synaptosomal membrane potential could
reduce “Ca2+ uptake indirectly. lf lgG depolarized the synaptosomal membrane,
a greater fraction of Ca2+ channels might be inactivated and unable to respond to
subsequent K*-depolarization. Alternatively, hyperpolarization of the synaptosomal
membrane would decrease uptake of Ca2+ by reducing the probability of opening
of channels for a given depolarizing pulse of K“. Changes in the fluorescence of
carbocyanine dyes as an indicator of alterations in membrane potentials have been
studied in a variety of cell types and in synaptosomes (Blaustein and Goldring,
1975; Heinonen et al., 1985; Hare and Atchison, 1992). Multiple dye compart-
mentation is avoided by elimination of the mitochondrial membrane potential by
depolarization (Heinonen et al., 1985; Hare and Atchison, 1992). Neither plasma
membrane depolarization nor hyperpolarization is involved in the reduction of
4‘r’Cafl uptake since LEMS lgG incubation caused no alterations in dye
fluorescence following addition of KCI (40 mM). Thus, LEMS lgG induces a direct
functional disruption of nerve terminal Ca2+ channels rather than indirectly by

altering the membrane potential.

113
To determine if the LEMS antibody was specific for Ca2+ channels, we

examined its potential to interact with other voltage-dependent channels. If the
antibody were not specific for Caz+ channels, it may indicate the lgG could
recognize a portion of the channel molecule common to all ion channels. This
possibility had not been previously tested directly.

Synaptosomes contain voltage-dependent Na+ channels. Maximum uptake
of 22Na+ into synaptosomes occurs following 10 min incubation with 150 uM VER.
VER is used to activate synaptosomal Na+ channels as these channels inactivate
rapidly in response to K*—depolarization (Kreuger and Blaustein, 1980; Tankum and
Catterall, 1981).

There is considerable homology between the a subunit of the voltage-
dependent Na+ channel and or1 subunit of the dihydropyridine (DHP)-sensitive (L-
type) voltage-dependent Caz+ channel (Ellis et al., 1988). LEMS lgG affects L-type
Ca2+ channels in several cell lines (Login et al., 1987; Peers et al., 1987; Kim and
Neher, 1988). Since the 0 subunits contain the binding site for pharmacological
antagonists, are the pore forming subunits and are indispensable for the function
of the channeI(Catterall, 1988), they are potential candidates as antigenic
determinants for autoantibodies. Therefore, LEMS antibodies directed against the
a, subunit of the Ca2+ channel might cross-react with the a subunit of Na“
channels as well. Incubation of synaptosomes with concentrations of LEMS lgG
that significantly altered uptake of “Ca2+ into K"-depolarized synaptosomes had

no effect on uptake of ”Na“ into veratridine-depolarized synaptosomes. This is

1 19
consistent with the work of Pascuzzi and Kim (1988) in which functional in vivo

measurements of nerve conduction velocity made from animals chronically treated
with LEMS lgG demonstrated that conduction was normal.

In some cells, Ca2+ and K“ channels may be structurally similar (Hume,
1985; Nerbonne et al., 1987). Also, some patients with MG, another autoimmune,
neuromuscular disease, have antibodies against a p-bungarotoxin binding protein
(Lu et al., 1991). This protein has been demonstrated to belong to a class of
voltage-gated K+ channels (Rehn and Betz, 1984). If patients with LEMS have a
similar antibody, alterations in K+ channel conductance could indirectly alter Ca2+
channel conductance. Treatment of synaptosomes with serum or serum and lgG
did not alter the loading of “RN into synaptosomes by the Na“/K+ ATPase.
Likewise, concentrations of LEMS lgG that significantly altered uptake of 4‘SCaz+
into K"—depolarized synaptosomes had no effect on efflux of “RN following K‘-
depolarization. This provides further evidence that the LEMS autoantibody
recognizes an epitope which is specific for voltage-dependent Ca2+ channels.

In summary, LEMS IgG from Patient 3 acts on central nerve terminal Ca2+
channels to reduce depolarization-dependent uptake of 45Ca“. The antibody
appears specific for Ca2+ channels as it is unable to disrupt ion flow through
voltage-dependent Na+ or K“ channels. These deficits are not due to indirect
effects on membrane integrity or synaptosomal membrane potential but appear to
result from a direct block of or alteration in Ca2+ channel function. The preferential

action of the LEMS antibody for nerve terminal Ca2+ channels suggests that there

120

t

is no common determinant or epitope on different voltage—gated channel proteins
to which the antibody can bind. We cannot exclude, however, the possibility that
some patients with LEMS may synthesize antibodies with different specificities.
These data also suggest antigenic similarities exist between somatic motor nerve
terminal and central nerve terminal Ca2+ channels. These findings are important
as the existence of autoantibodies which are specific for a component of the
neurosecretory process may serve as molecular tools for studying the steps
involved in quantal transmitter release both in the peripheral and central nervous

system.

CHAPTER FIVE

DISRUPTION OF SYNAPTOSOMAL 662+ CHANNELS BY LAMBERT-EATON
MYATHENIC SYNDROME IMMUNOGLOBULIN: SERUM DEPENDENCY

121

1 22
A. Summary

Chapter Four demonstrated that in the presence of control human serum,

a2+ into

immunoglobulins isolated from a patient with LEMS reduced flux of C
isolated nerve terminals during depolarization. The objective of the present study
was to determine the role of serum in reducing uptake of “Ca2+ into rat brain
synaptosomes by LEMS lgG. Depolarization-dependent uptake of “0a“ through
voltage-gated Ca2+ channels was determined using synaptosomes incubated with
control (disease-free) and LEMS lgG with or without control human serum. In the
absence of human serum, LEMS lgG did not reduce uptake of "“Ca2+ into
synaptosomes. However, in the presence of control human serum (10% of total
incubation volume), “Ca2+ uptake was reduced significantly by LEMS lgG (2 and
4 mg /ml), but not by lgG from disease-free patients or by 10% (v/v) control human
serum alone. This concentration of serum was found to be optimal; higher
concentrations produced significant reductions in Caz+ uptake, whereas at lower
concentrations the serum / lgG combination was ineffective. The depressant effect
of serum alone on 45C3“ uptake was mimicked by bovine serum albumin (BSA)
suggesting that deficits in “Ca2+ uptake produced by serum were the result of
increased protein binding of the radiolabel. Heat-inactivating the serum abolished
its ability to interact with LEMS lgG to depress 45Ca2+ uptake. This suggested a
role for complement in this effect. To test whether the membrane attack complex

(MAC) or the alternative pathway of complement (APO) contributed to the alteration

in Ca” channel function, “Ca2+ uptake into synaptosomes was measured

123
following incubation with LEMS IgG and C5- or Factor B- deficient serum,

respectively. LEMS lgG reduced K*-stimulated uptake in both cases. Thus, neither
the MAC nor the APC contributed to disruption of Ca2+ channel function by LEMS
lgG in synaptosomes. In contrast, incubation of synaptosomes with LEMS IgG and
CS-deficient serum prevented the reduction in “Ca2+ uptake suggesting that the
C3 component of the complement pathway contributes to Ca2+ channel

dysfunction in synaptosomes following incubation with serum and LEMS lgG.

B. Introduction

Autoantibodies, such as those in LEMS, that interfere with the release
process may either interrupt Ca2+ flux into the terminal or some intraterminal event
critical to transmitter release. Results from several studies indicate that LEMS
autoantibodies interact with voltage-dependent Ca2+ channels in cell lines of
diverse origin (Peers et al., 1987; De Azipurua, 1988a; Kim and Neher, 1988).
LEMS serum also reduces depolarization-dependent Ca2+ entry into and ACh
release from isolated nerve terminals of rat brain (Meyer et al., 1986; Hewett and
Atchison, 1991). Data in Chapter Four demonstrate that in the presence of control
human serum, lgG isolated from a patient with LEMS reduced Ca2+ flux into
synaptosomes as well.

Activation of the serum component complement plays a central role in
mediating postsynaptic membrane damage in animal models of the autoimmune
neuromuscular disorder MG, as well as in the human disease state (Engel et al.,

1977; Sahashi et al., 1978). Attempts to clarify the participation of complement in

124
the pathogenesis of LEMS has so far yielded negative results (Prior er al., 1987;

Lennon and Lambert, 1988). As it is still possible that complement activation may
be important in LEMS, the present study was undertaken to assess the role of
serum in the pathogenesis of Ca2+ channel dysfunction induced by LEMS lgG in

isolated nerve terminals from mammalian CNS.

C. Materials and Methods

1.Materlals

N-methyI-d-glucamine, 05-, C3-and Factor B-deficient human sera were
purchased from Sigma Chemical Co. (St. Louis, MO). HVLP (0.45 0M pore size;
25 mm diameter) filters were purchased from Millipore (Bedford, MA). HEPES was
purchased from United States Biochemical Corp. (Cleveland, OH). The radiolabel
“Ca" (15-50 mCi/mg) was purchased from New England Nuclear (Boston, MA).
Disease-free control human plasma was purchased from the American Red Cross
(Lansing, MI). Control human serum was donated by a healthy volunteer. Plasma
from a patient with the clinical and electromyographic features of LEMS was

donated by Mark Glasberg, M.D., Henry Ford Hospital, Detroit, MI.

2. Immunoglobulln Isolation
Immunoglobulln from control and LEMS patients was isolated and
quantitated as described in detail in Chapter Four. The LEMS lgG used in these

studies was identical to that used in Chapter Four as well.

125

3.Preparatlon of Synaptosomes

Synaptosomes were prepared according to a modification of the method of
Gray and Whittaker (1962), as described in detail in Chapter Three. Equal aliquots
of the synaptosomal suspension were incubated at 4°C under 100% 02 for 60 min
with either no serum or 10% (v/v) control, 05-, 03- or Factor B-deficient serum
and varying concentrations of lgG from patients with no known neurological or
immunological illness (control) or LEMS. Alternatively, one concentration of lgG
and varying concentrations of human serum were added to the suspension to test
effects of serum on 4"’Ca2+ uptake into synaptosomes. Prior to uptake measure-

ments, synaptosomes were allowed to equilibrate for 30 min at 37°C.

4. Solutions and Measurement of “Ca“ Uptake
The solutions for and the measurement of “Ca“ are described in detail in
Chapter Three. Uptake is expressed as nmole 45Ca2+/ttg protein or as a

percentage of disease-free control.

5. Statistical Analysis
Data were analyzed by randomized complete block analysis of variance
followed by Dunnett’s t-test or Students t-test for paired samples. Differences were

considered significant at p < 0.05.

126
D. Results

1. LEMS lgG In the Presence and Absence of Control Human Serum

Uptake of 4”Ca” into synaptosomes during 1 sec of K+-depolarization was
unaffected by 90 min incubation with LEMS lgG alone (fig. 5.1). However, in the
presence of control human serum (10% of the total incubation volume) 2 and 4
mg/ml LEMS lgG inhibited uptake by 28.3 t 10.0% and 39.1 t 21.3%,
respectively, as compared to their paired controls incubated with equivalent
concentrations of control (disease-free) lgG (fig. 5.1). This concentration of serum

(10 %; v/v) is the concentration used in standard tissue culture procedures.

2. Control Human Serum

Since the presence of serum with LEMS lgG is important in reducing “Ca“
uptake, the effect of control (disease-free) human serum on uptake of “Ca“ into
synaptosomes was examined (fig. 5.2). Synaptosomes were incubated with 2
mg/ml IgG from disease-free patients and varying concentrations of control
(disease-free) human serum. A concentration-dependent suppression of uptake is
evident, with uptake at concentrations 2 15 % (v/v) being significantly lower than
those in the absence of serum. No statistically significant reduction in uptake was
observed with a serum concentration of 10 % (v/v). Similar results were obtained
when bovine serum albumin was substituted for serum (fig. 5.3).

Synaptosomes were incubated with 2 mg/ml control or LEMS lgG and
varying concentrations of disease-free human serum to determine the optimal

concentration of serum needed in the reaction mixture. Again, suppression of

12

‘1

No Control
serum

‘
1

 

 

 

 

 

 

 

 

g ‘5 . —Jfi * _i_. *
t’g’ °°f
2:3 .0. / %
0' A //
2 [LEMSIgG] 4
(molml synaptosomal suspenslon)

figure 5.1. Effect of LEMS lgG in the presence and absence of control human
serum on uptake of 45Ca2+ into rat forebrain synaptosomes. Synaptosomes were
incubated for 60 min at 4°C with 2 and 4 mg/ml LEMS or control lgG 1- control
human serum. Prior to the initiation of uptake, synaptosomal suspensions were
equilibrated at 37°C for 30 min. Uptake in 5 mM K” was subtracted from that in
42.5 mM K+ as depolarization-independent background. 45Ca2+ uptake is
expressed as % of paired control, control being synaptosomes incubated with
equivalent concentrations of control lgG. All values are the mean 1 SEM of 5
experiments. Asterisks indicate values that are significantly lower than their
respective paired control (p < 0.05).

128

 

 

 

 

 

1 5 P
g;
X III
‘::{E 10 . , ‘*
v v ‘
i- 4
‘%~9 ‘3?
”as 5%” ,.- l
. 9 9 ,
:3 Q”! wuw' ”3% iVVV
E: ‘Vfl’ 0%0' Qflfll fifi’
5 r so. 1941 poor 040
:3 cos flfifii fififi "fit
13 c} (oat flfifii one: ””5
90-.940 -oos too
sot 1404 too« 449
E5 - .avi flfifli §V5t ”V5
3 so. 900 on, too
~n ‘Qfl’ 0%”! fifi? QR?
”MP ififlfi 29%5 «Q9
0 A A A . .A.A. ’A’A’A‘ .A.A.A‘ ’A’A’

0 5 10 15 20 25
PERCENT SERUM

figure 5.2. Effect of control human serum on “"’Ca2+ uptake into rat forebrain
synaptosomes during 1 sec depolarization by 41.25 mM KCI. Synaptosomes were
incubated with 2 mg/ml control lgG and varying concentrations of control human
serum for 60 min at 4°C. Prior to the initiation of uptake, synaptosomes were
allowed to equilibrate for 30 min at 37°C. Uptake in 5 mM K+ was subtracted from
uptake in 42.5 mM K“ as depolarization-independent background. 45Ca2+ uptake
is expressed as nmol/pg protein. All values are the mean : SEM of 3 experiments.
Asterisks indicate values that are significantly lower than uptake into synaptosomes
which had been incubated with no serum.

129

*

a

10 15 20 25
ERCENT BSA

""Ca‘a UPTAKE

 

 

 

s\\\\\\\\\\\\l~
W
W”

 

 

 

 

 

 

 

 

 

(nmollug protein E-OS)
o W '

'0‘”

figure 5.3. Effect of BSA on “Ca2+ uptake into rat forebrain synaptosomes during
1 sec of depolarization. A BSA solution containing the approximate protein
concentration of serum was made in deionized water. Synaptosomes were
incubated with 2 mg/ml control lgG and varying concentrations of BSA as in
figure 5.2. All values are the mean 1 SEM of 5 experiments. Asterisks indicate
values that are significantly lower than uptake into synaptosomes which had been
incubated with no BSA (p < 0.05).

130

uptake by serum alone is evident (fig. 5.4). “Ca2+ uptake was reduced
significantly only by the combination of 2mg/ml LEMS lgG and 10% (v/v) serum

when compared to its paired control (2 mg/ml control lgG and 10% serum).

3. LEMS lgG and Complement
Control human serum was incubated for 45 min at 57°C to inactivate

complement. This prevents the reduction in uptake of “Ca2+

into the synap-
tosomes by LEMS lgG (2 and 4 mg /ml) and control human serum (10%) (fig. 5.5).

Two crucial components of the complement cascade are 03 and Cs.
Removal of CS prevents the assembly of the membrane attack complex (MAC).
The ability of LEMS lgG and control human serum to reduce “Ca2+ uptake into
synaptosomes was unaltered when control human serum was replaced with C5-
deficient human serum (Fig. 5.6). Conversely, removal of the C3 component of
complement from control human serum prevented the action of the LEMS lgG (fig.
5.7). Incubation of synaptosomes with 2 and 4 mg/ml of LEMS lgG and C3-
deflcient serum (10 %; v/v) abolished the reduction in uptake produced by the
same concentrations of LEMS lgG in the presence of control human serum or OS-
deficient serum.

To determine if the alternative pathway of complement (APC) played a role
in mediating the dysfunction of synaptosomal Ca2+ channels by LEMS lgG and
control human serum, synaptosomes were incubated with 2 and 4 mg/ml control

and LEMS lgG in the presence of either control human serum or Factor B-deficient

serum (fig. 5.8). Factor B is essential for activation of the APO. In the presence of

131

A

 

'1— CONTROL lgG

* --o-- LEMS igc

“Ca '* UPTAKE
(nmol/pg protsln E-06)

 

 

° 0 b 1.0 I 20
PERCENT SERUM

Figure 5.4. Effect of control and LEMS £962 and various concentrations of control
human serum on 1 sec K*-stimulated Caz“ uptake into rat forebrain synapto-
somes. Synaptosomes were incubated with 2 mg/ml control and LEMS IgG and
varying concentrations of control human serum as described in Figure 5. 2. All
values are the mean + SEM of 5 experiments. ‘;", indicates values that are
significantly different from their paired control. (b), indicates values that are
significantly different from uptake into synaptosomes that were incubated with no
serum. (p < 0.05)

132

1:1 Control m Heated
seruri serLI'n

120-
LU _

K:

<g90' -x- -x-
Es --‘- —Ln
3°60-

00‘s

033d

0v

1'

 

 

 

 

 

 

 

 

4
[LEMS lgG]
ling/ml synaptosomal suspension)

figure 5.5. The effect of LEMS lgG in the presence of control or heat inactivated
serum on uptake of “Ca2+ into rat synaptosomes. Synaptosomes were incubated
for 60 min at 4°C with 2 & 4 mg/ml control or LEMS lgG and heat inactivated
serum or control human serum. Prior to uptake, synaptosomal suspensions
equilibrated for 30 min at 37°C. Synaptosomes were then incubated for one sec
in a solution containing “Ca2+ and 42.5 mM KCI. Results are the mean 1.- the SEM
of 8 experiments Asterisks indicate values that are significantly lower than their
paired control, control being synaptosomes incubated with equivalent concentra-
tions of control lgG (p < 0.05).

133

CS deficient sera

6 . I: con LEMS

“L ‘L .1.

“Ca” U PTAKE
(mom mm E-osi
Q

 

 

 

 

 

 

 

 

 

 

0

_N\\\\\\\\\\\\ 4
Ts\\\\\\\\\\‘1~*

['9 G

(mg/ml synaptosomal suspension)

figure 5. 6. The effect of CS-deficient human serum on the LEMS lgG induced
reduction of K -stimulated MCa uptake into rat synaptosomes. Synaptosomes
were incubated with 2 and 4 mg/ml control and LEMS lgG in the presence of 05-
deficient serum for 60 min at 4°C. Thirty min prior to the initiation of uptake,
synaptosomes were allowed to equilibrate at 37°C. Uptake in 5 mM K was
subtracted from uptake in 42. 545 mM 2K as depolarization-independent background.
Data are expressed as nmol 452+Ca uptake/lug protein. Values are the mean +
SEM of 6 experiments. Asterisks indicate values that are significantly lower than
their paired control, control being synaptosomes incubatedWlth equivalent
concentrations of control lgG (p < 0. 05).

134

C3 deficient sera
[:1 con LEMS

i

a:

' J.

 

 

 

W

 

 

 

 

 

 

 

4

“Ca” UPTAKE
(nmolluo 9min E—oo)
p...
.. N W

0
['93

(mg/ml synaptosomal suspension)

‘ F gure 5. 7. The effect of C3-defici ent zhuman serum on the LEMS lgGi induced

reduction in 1 sec K -stimulated 452+Ca uptake into rat synaptosomes. Synap-
tosomes werei incubated as describedi in figure 5. 6. Values are the mean- SEM
of 4 experiments.

13

0|

 

1:: Control VIA B-deficient

 

serun serum
UJ 120'
X: l
{990. *
l—E’ L_I__ * -x-
%860
<15
8g 30'
to" . /
v /A 2

 

 

 

 

 

 

 

4
2 [LEMS lgG]

(mg/ml synaptosomal suspension)

figure 5.8. The effect of control and LEMS lgG in the presence of Factor B-
deflcient human serum on 1 sec K+-stimulated 45Ca2+ uptake into rat synap-
tosomes. Synaptosomes were incubated with 2 and 4 mg/ml control and LEMS
IgG in the presence of control human serum or B-deficient serum as described in
Figure 5.6. Values are the mean 1 SEM of 5 experiments. Asterisks indicate values
that are significantly lower than their paired control, control being synaptosomes

incubated with equivalent concentrations of control lgG (p < 0.05).

136
control human serum, uptake was reduced 20.4 1 8.4% and 34.1 1 5.1% by 2

and 4 mg/ml LEMS lgG when compared to synaptosomes incubated equivalent
concentrations of control lgG. Uptake in the presence of B-deficient human serum
was inhibited 8.8 1 13.0% and 29.9 1 6.0% by 2 and 4 mg/ml LEMS lgG, when
compared to the respective paired controls. At 4 mg/ml LEMS lgG this reduction

was statistically significant.

E. Discussion

LEMS lgG in the presence of control human serum is able to reduce
depolarization—dependent uptake of “Ca2+ into isolated central nerve terminals
(Chapter Four). The aim of the present study was to assess the role of serum, in
particular complement, in the pathogenesis of Ca2+ channel dysfunction in
synaptosomes following incubation with LEMS lgG and human serum. Comple-
ment has been shown to be important in mediating other autoimmune, neuromus-
cular diseases (Engel et al., 1987; Sahashi et al., 1978; Toyka et al., 1988).

In this study, the lgG from the LEMS patient that was studied in Chapter
Four was used. The plasma preparation from this patient has been shown to
transfer the electrophysiological features of LEMS to mice (Patient 3; Chapter
Four), thus providing us with a positive functional control. IgG isolated from this
patient only alters 45Ca2+ flux into rat forebrain synaptosomes in the presence of
control human serum. Incubation with lgG alone or heat-inactivating the serum

abolishes this effect. The optimal concentration of serum appears to be 10% (v/v)

137

of the total incubation volume. At this concentration, serum alone caused no
significant decreases in “Ca2+ uptake into synaptosomes. The MAC of comple-
ment does not appear to be involved in the reduction of uptake produced by
LEMS lgG and serum since uptake is reduced when LEMS lgG is incubated with
CS-deficient serum. Incubation of synaptosomes with 4 mg/ml LEMS and B-
deficient serum showed similar reductions as those incubated with the same
concentration of lgG but with control human serum. This suggests that at least at
the higher concentration of lgG the alternative pathway does not appear to be
involved. Our data suggests that it may play a role at lower lgG concentrations,
however. In contrast, incubation of synaptosomes with C3-deficient serum and
either 2 or 4 mg/ml LEMS lgG did not reduce K*-stimulated 4“Ca2+ uptake.

LEMS lgG alone did not reduce Ca2+ flux into isolated nerve terminals
following 90 min of incubation. Incubation of synaptosomes with LEMS lgG in the
presence of control human serum, however, resulted in significant reductions in
stimulated 45Ca2+ uptake. In other studies, cells cultured with LEMS lgG for one
to seven days reduced Ca2+ flux (Roberts et al., 1985; Peers et al., 1987).
However the tissue culture media used in the above studies contained 10% (v/v)
fetal calf or horse serum. As this is standard cell culture procedure, the importance
of serum components with LEMS lgG may not have been apparent.

Serum alone decreases depolarization-dependent uptake of 45Ca2+ in a
concentration-dependent manner. This is likely due to increased binding of “Ca2+

to serum proteins thereby effectively decreasing the concentration of radiolabel

138
available for uptake, since equivalent concentrations of BSA alone reduce uptake.

More notable is the observation mat 10% (v/v) control human serum does not
significantly alter uptake as compared to uptake measured in the absence of
serum. This is important since 10% of the total incubation volume is the concentra-
tion of serum used in our experimental paradigm.

This concentration of serum appears to be optimal. It was the only serum
concentration tested, which when combined with 2 mg/ml LEMS lgG resulted in
reduced depolarization-dependent uptake of “Ca2+ into synaptosomes. There
were no alterations in uptake between control and LEMS lgG with lower
concentrations of serum. Higher concentrations of serum suppressed uptake to
such an extent that any effects of the LEMS lgG on uptake may have been
masked. Thus, the concentration of control serum should be 10% of the total
incubation volume.

The reduction in uptake of “Ca2+ by LEMS IgG and serum was abolished
by heat inactivating the serum. Heating serum will inactivate proteins or proteases
present in the serum, such as complement, as CS, the pivotal component in both
the classical and alternative pathway, is heat labile (Bokisch et al., 1975; Stossel
et al., 1975).

Two of the most important components of complement are 03 and CS.
The formation of the C3 convertase is the essential step of the complement
reaction in generating activated C3 (C3b). C3b facilitates phagocytosis and

generates CS convertase (Hostteter et al., 1984). It also functions as an important

139
intermediate in the activation of the alternative pathway (GOetz, 1987). Cleavage

of C5 by CS convertases constitutes the final proteolytic reaction in both the
classical and alternative pathway. Once present, 05b induces the assembly of the
cytotoxic CSb-9 complex (MAC), which forms transmembrane channels that permit
bidirectional flow of ions and ultimately macromolecules. This produces cellular
injury and/or lysis (Bhakdl and Tranum-Jensen, 1978; Porter and Reid, 1978).
In the preceeding chapter, it was demonstrated that the decrease in “Ca2+
uptake by LEMS IgG was not a result of non-specific membrane damage. LDH
release from synaptosomes incubated with LEMS lgG and control human serum
was not increased over that of paired controls. Although membrane lysis did not
occur, the present study sought to rule out the possibility that the MAC may cause
discrete lesions that alter Ca“ channel permeability (Bhakdl and Tranum-Jensen,
197s). Deficits in 45Ca2+ uptake following incubation with LEMS lgG and co-
deficient serum indicate that the MAC does not play a role in the disruption of Ca2+
channel function in our system. This is consistent with results of Prior etel., (1987)
and Lambert and Lennon (1988) who demonstrated that the latter steps in the
complement cascade (MAC) are not essential for the passive transfer of LEMS.
Genetically CS-deficient mice are susceptible to the LEMS lgG as are mice with
normal complement.

Incubation of LEMS lgG with C3—deficient serum prevented the reduction of

synaptosomal “Ca2+ entry seen with control human serum. These findings are in

contrast to those of Lambert and Lennon (1988) who report that the electro-

140
physiological characteristics of LEMS could be transferred passively to mice that

had been depleted of C3 by cobra venom factor (CVF). Their results suggest that
in vivo the effects of LEMS lgG may not be due to complement activation. This
unaltered response in the absence of complement activity does not, however,
mean that it is not involved when present. Many of the effects of lgG or alternative
pathway activators on cells are greatly enhanced in the presence of complement
(Rother, 1987). Furthermore, CVF interacts with Factor B of the APC to form a
stable C3 convertase (Goetz and M0ller- Eberhard, 1971) thereby depleting C3 by
continually activating it. In normal animals the CS depletion induced by CVF is
relatively short-lived (Klaus, 1987) and C3 depression must be maintained by
repeated injections of CVF (Lennon and Lambert, 1988). Continual activation of 03
may result in binding of 03b to a non-specific acceptor, such as the plasma
membrane (Law and Levine, 1977) or to the antigen-antibody complex itself (Capel
et al., 1978). Thus, there is still a possibility that activated 03 (C3b) may be
important in mediating the LEMS pathogenesis in vivo. An alternative explanation,
however, is that our in vitro model may not mimic precisely events occurring In
vivo. As this is an artificial system, this possibility can not be excluded. Nonethe-
less, our results suggest a role for CS in the action of LEMS lgG on synaptosomal
Ca2+ channels and perhaps suggests that further experimentation is needed to
clarify the participation of complement in the pathogenesis of LEMS.

The APC is a non-specific, natural defense system against pathogenic

microorganisms (Joiner et al., 1984; Hostteter et al., 1984; Goetz, 1987). In

141
addition, it is an amplification system for increasing C3b deposition and 03/05

convertase activity after complement activation through the classical pathway
(Hostetter et al., 1984). Without Factor B, the APC can not function. The fact that
a reduction in uptake was seen after 1 sec of K*-induced depolarization following
incubation of synaptosomes with 4 mg/ml LEMS IgG and Factor B-deficient serum
suggests that at least at higher concentrations of LEMS lgG the APC is not
contributing to the alterations in “Ca2+ uptake seen with the LEMS autoantibody
and human serum. It may play a role at lower concentrations, however.

In summary, in synaptosomes, the presence of serum is important in
mediating a functional change in Ca2+ channel activity by LEMS lgG. The optimal
concentration of serum appears to be 10% (v/v) of the total incubation volume.
This is a concentration of serum that alone does not reduce stimulated “Ca2+
uptake into synaptosomes, significantly. The MAC and APC do not appear to be
involved. In contrast, the heat-labile, complement component, 03 is needed to
disrupt 45Ca2+ uptake into isolated nerve terminals following incubation of
synaptosomes with LEMS IgG and control human serum. Formation of the MAC
is not involved in the pathogenesis of LEMS in vivo (Prior et al., 1987; Lennon and
Lambert, 1988) or In vitro (present study). Enhanced phagocytosis of the antigen-
antibody complex by macrophage infiltration cannot occur in our system. Perhaps,
deposition of cab (z 180 kDa) adjacent to the autoantibody (z 150 kDa) sterically

hinders the gating process of Ca2+ channels.

CHAPTER SIX

PHARMACOLOGICAL DETERMINATION OF SYNAPTOSOMAL Ca2+

CHANNEL SUBTYPE(S) AFFECTED BY LAMBERT-EATON MYASTHENIC

SYNDROME IMMUNOGLOBULIN

142

143

A. Summary

The data presented thus far support the hypothesis that a circulating
autoantibody in the plasma of patients with LEMS interferes with nerve terminal
Ca2+ channels. In the following study, the equilibrium binding of [am-nitrendipine,
[mu-to »Cng and [am-verapamil was measured to determine whether binding sites
reputed to be associated with pharmacologically identitified classes of Ca2+
channels are reduced or altered by acute exposure to LEMS lgG alone and/or
LEMS IgG in the presence of serum. [am-nitrendipine binds to synaptosomes with
high affinity in both normal and elevated K1 solutions (Shafer et al., 1990). As no
difference in the binding characteristics exists between the two solutions, all
experiments were performed in non-depolarizing K+ solutions. Scatchard analysis
of binding of [am-nitrendipine to synaptosomes in normal K+ solution resulted in
a ligand dissociation constant (Kd) value of 214.0 1 84.0 pM and a maximum
number of binding sites (BM) value of 62.0 1 12.0 frnoI/mg protein. In the
presence of 10% (v/v) control human serum, the values for Kd and 8m“ were
152.8 1 48.0 pM and 62.5 1 12.0 fmoI/mg protein, respectively. Incubation of
synaptosomes with control or LEMS lgG at 4 mg/ml prior to initiation of binding
assays did not significantly alter binding of [am-nitrendipine. Likewise, incubation
with control or LEMS IgG in the presence of 10% (v/v) control human serum did
not alter the binding characteristics of [am-nitrendipine to synaptosomes. [WI]-
Cng binds to a single high affinity binding site on synaptosomes, with half-

saturation occurring in the mid-picomolar range. LEMS lgG at 4 mg/ml did not

144
alter the binding of [‘z‘lj-Cng as compared to synaptosomes incubated with an

equivalent concentration of control IgG. The combination of LEMS lgG and control
human serum failed to alter the binding characteristics as well. finally, the binding
characteristics of [am-verapamil to synaptosomes were examined. LEMS lgG both
in the presence and absence of control human serum, significantly decreased the
maximum number of binding sites for verapamil by approximately 45%. The affinity
of verapamil for the remaining receptors was not significantly altered, however.
Thus, lgG from this patient with LEMS can interact with a Ca2+ channel binding site

for verapamil but not for DHPs or Cng.

B. Introduction

The observation that multiple types of Ca2+ channels exists in nerve
terminals (Penner and Dryer, 1986; Atchison and O’Leary, 1987; Miller, 1987)
raises the question of the identity of the subtype of Ca“ channels affected by
LEMS lgG. Experimental evidence suggests that LEMS lgG inhibits Caz+ entry into
voltage-dependent 032+ channels from several sources, including motor nerve
terminals (Lang et al., 1987), SCC cells (Roberts et al., 1985; De Azipurua et al.,
19883; Lang et al., 19893; 1989b), chromaffin cells (Kim and Neher, 1988), anterior
pituitary cells (Login et al., 1987), NG108-15 cells (Peers et al., 1987), PC12 cells
(Lang et al., 19893; 19890) and synaptosomes (Chapters 4,5). Due to the
inaccessibility of the putative target site of the LEMS lgG (motor nerve terminals),

elucidation of the subtype(s) of Ca2+ channel affected by LEMS lgG has been

145

attempted in cell lines with the assumption that somal Ca2+ channels are similar
to nerve terminal Ca2+ channels.

LEMS lgG was shown to reduce Ca2+ currents with the electrophysiological
characteristics of L-type channels in NG108-15 and chromaffin cells. (Kim and
Neher, 1988; Peers et al., 1990). Data from studies using SCC cell lines (Roberts
et al., 1985; De Azipurua et al., 19883; 1988b) suggest an effect on both L- and
N-type channels. finally, LEMS autoantibodies have been shown to
immunoprecipitate an [‘”I]-Cng labeled voltage-gated Ca2+ channel (N-type) from
a human neuroblastoma cell line (Sher et al., 1989).

The results from these studies imply that the L- and N- type Ca2+ channel
are putative antigenic targets for the LEMS autoantibody. However, the types of
Ca2+ channels that exist at the nerve terminal is unclear. In fact, experimental
evidence suggests that Ca2+ channels at motor nerve terminals are neither N- nor
L-, and are possibly of the P- (Anderson and Harvey, 1987; Sano et al., 1987;
Protti ef al., 1991; Uchitel et al., 1992) or some as yet undetermined type.

Synaptosomes maintain the morphological and functional properties of intact
nerve terminals and several pharmacological studies suggest that Ca”+ channels
in the CNS resemble those in the periphery (Anderson and Harvey, 1987;
Nachshen and Blaustein, 1979; Reynolds 3131., 1986; Suszkiw et al., 1986;1989;
Atchison, 1989; Protti et al., 1991; Uchitel et al., 1992). The DHP nitrendipine has
been used in numerous binding and flux studies in synaptosomes to examine the

properties of binding sites which are presumed to be associated with L-type Ca2+

146
channels (Boles et al., 1984; Greenberg et al., 1984; Turner and Goldin, 1985; Wei

and Chang, 1985; Carvalho et al., 1986; Suszkiw et al., 1986; Dunn, 1988). The
phenylalkylamine, verapamil also appears to act as an L-type antagonist but has
a separate binding site from that of the DHPs (Carvalho et al., 1986; Miller, 1987;
Glossmann and Striessnig, 1988). finally, synaptosomes contain specific high-
affinity receptors for the N-type channel antagonist, Cng, as well (Abe efal., 1986;
Marqueze et al., 1988; Wagner et al., 1988; Scheer, 1990).

Thus, the goal of the present study was to determine whether binding sites
reputed to be associated with Ca2+ channels are reduced or altered by acute
exposure to LEMS lgG. The equilibrium binding of the Ca2+ channel antagonists
[am-nitrendipine, ['alj-Cng and [am-verapamil to synaptosomal membranes
following incubation with control and LEMS lgG, both in the presence and absence

of serum was measured.

C. Maten'als and MeMods

1. Materials

The radioisotopes [am-nitrendipine (74 Ci/mmol), [am-verapamil (66
Ci/mmol), and [‘25I1-Cng (2200 Ci/mmol) were purchased from New England
Nuclear (Boston, MA). Unlabelled Cng was purchased from Penninsula
Laboratories (Belmont, CA). Polyethylenimine, verapamil and nifedipine were
purchased from Sigma Chemical Co. (St. Louis, MO). Disease-free control human
plasma was purchased from the American Red Cross (Lansing, MI). Control

human serum was donated by a healthy volunteer. Plasma from a patient with the

147
clinical and electromyographic features of LEMS was donated by Mark Glasberg,

M.D., Henry Ford Hospital, Detroit, MI.

2. Immunoglobulln Isolatlon
Immunoglobulin from control and LEMS patients was isolated and
quantitated as described in detail in Chapter Four. The LEMS lgG used in these

studies was identical to that used in Chapter Four and five as well.

3. Preparatlon of Synaptosomes

Synaptosomes were prepared according to 3 modification of the method of
Gray and Whittaker (1962), as described in detail in Chapter Three. The final pellet
was resuspended in physiologic saline buffer. Equal aliquots of synaptosomal
suspension were incubated at 4°C under 100% 02 for 60 min with either
physiologic saline buffer, 10% (v/v) serum, 4 mg/ml control or LEMS lgG alone or
lgG plus 10% (v/v) serum. The proper volume of physiologic saline buffer was
added to maintain a constant incubation volume. Prior to the initiation of binding
experiments, synaptosomes were allowed to equilibrate for 30 min at 37°C.

Equilibrium binding assays were carried out at room temperature.

4. Solutions
Stock solutions of the radiolabels were made in non-depolarizing K+ buffer.
The incubations were terminated and the filters washed by the addition of cold

quench solution. Contents of these solutions are described in Chapter Three.

148
5. Binding Experiments

Binding of [am-nitrendipine was initiated by addition of 40 ul of
synaptosomal suspension (200 -300 pg protein) to 400 pl of non-depolarizing K+
buffer containing 10—1500 pmol of [3H]-nitrendipine in ethanol (EtOH) for 60 min at
25°C in a room illuminated by a sodium lamp. The final concentration of EtOH was
always less than 0.1%. Non-specific binding was measured in the presence of 5
11M unlabelled nifedipine. Incubation was terminated by addition of five ml of cold
quench solution and rapid filtration through glass fiber filters (Whaman GF\B) that
had been presoaked in quench solution containing 0.1% polyethylenimine. The
filters were rinsed twice with five ml of cold quench solution. Scintillation cocktail
was added 12 hr before the radioactivity trapped on the filters was estimated in a
liquid scintillation counter having an approximate efficiency of 45% for 3H. Protein
content was determined by the method of Lowry et al., (1951) using BSA as a
standard. To determine total ligand concentration, a 40 pl aliquot was taken from
the stock solution and the radioactivity was estimated. The average of triplicate
values was used for Scatchard Analysis (Scatchard, 1949) to determine the
equilibrium dissociation constant (Kd) for binding and the maximum density of
binding sites (Bmax)'

Initiation of [‘alj-Cng binding was essentially identical to that described
above except that 40 ul of synaptosomal suspension was added to 400 pl of non-
depolarizing K+ buffer containing 30-480 pmol of [‘25I]-Cng in deionized H20 and

incubated for 60 min at 25°C. In addition, both the stock solutions and quench

149
buffer contained 0.5 mg/ml BSA to limit non-specific binding to of Cng test tubes

and filters. Non-specific binding was measured in the presence of 1 pM unlabelled
Cng. Radioactivity remaining on the filters was estimated using a gamma counter
with an efficiency of 80% for ”5|. Since binding of Cng is essentially irreversible,
scatchard analysis of the data could not be performed. Thus, the average of
duplicate values were used for estimation of Bmam by saturation analysis.
[am-verapamil binding was again essentially identical to the methods
described for [3H]-nitrendipine binding. Binding of [3H]-verapamil was initiated by
addition of 40 pl of synaptosomal suspension to 400 pl of non-depolarizing K+
buffer containing 0.2-128 nmol of [am-verapamil in EtOH. They were next
incubated for 60 min at 25°C in a room illuminated by sodium lamp. The final
concentration of EtOH was always less than 0.1%. Both the stock solutions and
quench buffer contained 0.5 mg/ml BSA to limit non-specific binding of verapamil
to test tubes and filters. Non-specific binding was measured in the presence of 25
mM unlabelled verapamil. Radioactivity trapped on the filters was estimated in a
liquid scintillation counter having an approximate efficiency of 45% for 3H. The
average of duplicate values were used for Scatchard Analysis (Scatchard, 1949)
to determine the equilibrium dissociation constant (Kd) for binding and the densitity

of binding sites (BM).

6. Statistlcal Analysis
Data from the equilibrium binding of [am-nitrendipine and [am-verapamil

were analyzed by mixed design ANOVA followed by Student’s t-test for paired

150

samples. Data from Cng binding experiments were analyzed using a randomized

complete block ANOVA. Differences were considered significant at p < 0.05.

D. Results

1. Nltrendlplne Blndlng

A single high affinity binding site for [am-nitrendipine (10-1500 pmol) was
observed (Fig. 6.1). Non-specific binding accounted for approximately 25% of total
binding. Scatchard analysis of the data (n = 3) yielded a Kd value of 214 1 84 pM
and an apparent 8mx value of 62.0 1 12.0 frnol/mg protein. Hill slope was
approximately equal to one. These values are consistent with those reported by
others for [°Hj-nitrendipine binding in synaptosomes (Boles et al., 1984; Turner
and Goldin, 1985; Suszkiw et al., 1986). Addition of control human serum (10%
v/v) did not significantly alter the linearity of [am-nitrendipine binding or the K, and
8m" values which were 153 1 48 and 62.5 1 12.0, respectively (fig. 6.1).

Comparative effects of LEMS IgG and control IgG (4 mg/ml) on [3H]-
nitrendipine binding are depicted in figure 6.2A. The calculated BM, and Kd values
for binding of nitrendipine in the presence of control and LEMS lgG were not
significantly different from one another, nor were they different from values
obtained following incubation in buffer or serum alone (Table 6.1). Likewise,
incubation of synaptosomes with LEMS IgG in the presence of 10% (v/v) control
human serum did not alter either the Kd or BM values for binidng of nitrendipine
as compared to those observed using equivalent concentrations of control lgG and

serum (fig. 6.28; Table 6.1).

151

NITREBDIPIBE

BOUND/FREE
(L/mg x 10")

 

 

 

80
BOUND (fmol/mg protein)

figure 6..1 Effect of serum on nitendipine binding to synaptosomes. Specific
binding of [3 Hj-nitrendipine to synaptosomes in the presence and absence of
control human serum (10%, v/v) in non-depolarizing K buffer. Values are the
mean 1 SEM of three experiments. When SE bars are not shown the SE' is smaller
than the size of the symbol.

152
Table 6.1

[3H]Nitrendipine Binding Characteristics

 

 

Bmax Kd n
Buffer 62.0 1 12.0 214.1 1 84.0 3
Serum alone 62.5 1 12.0 152.8 1 48.0 3
Control lgG 75.2 1 10.2 236.0 1 49.0 6
LEMS lgG 72.4 1 18.5 289.6 1 77.0 6
Control lgG 75.1 1 14.5 256.0 1 76.4 6
+ serum
LEMS lgG 65.2 1 9.9 190.6 1 33.0 6
+ serum

 

Equal aliquots of synaptosomal suspension were incubated at 4°C under
100% 02 for 60 min with either physiologic saline buffer, 10% (v/v) serum, control
or LEMS lgG alone (4 mg/ml) or IgG ( 4 mg/ml) plus 10% (v/v) serum. Prior to
the initiation of binding experiments, synaptosomes were allowed to equilibrate for
30 min at 37°C. Equilibrium binding assays were carried out at room temperature.
Synaptosomes were added to non-depolarizing K+ buffer containing 10-1500 pmol
of [am-nitrendipine for 60 min at 25°C. Incubation was terminated by addition of
five ml of cold quench solution followed by suction-filtration. The average of
triplicate values was used for Scatchard Analysis to determine the equilibrium
dissociation constant (Kd; pM) for binding and the maximum density of binding
sites. (BM; frnol/mg protein).

153

NlTFiEhDIPIhE A

‘CCNIfi
OLENSqu

BOW/FREE
Limo x 1 O“)

 

 

0' - 1 - s - 1 ;+ s
O 14 28 42 56 70

8010 “ml/mo protein)

BOLhiD/FREE
(L/mo x 1 0“)

 

 

O 15 30 45 60 75
m (firtol/mo proteiri

figure 6.2. Comparative effects of Control and LEMS IgG in the absence (A) and
presence (8) of control human serum on nitrendipine binding to synaptosomes.
Values are the mean 1 SEM of 6 individual experiments.

154
2. m-Conotoxln Blndlng

The specific binding of [1”l]—Cng to synaptosomes in the presence of
control and LEMS IgG (4 mg/ml) is depicted in figure 6.3A. Non-specific binding
accounted for less than 10% of total binding. Cng bound to synaptosomes with
a half-saturation of between 0.1 to 0.2 nM. Since binding of Cng is essentially
irreversible, the maximum density of binding sites was estimated by extrapolation
of the saturated component to the Y-axis. BM values for Cng binding in
synaptosomes treated with control or LEMS IgG were 260.6 1 6.7 and 259.2 1
1.0. These values are consistent with those reported by others (Cruz and Olivera,
1986; Wagner et al.,1988; Marqueze et al., 1988). Almost identical binding curves
were obtained when synaptosomes were incubated with control or LEMS IgG (4
mg/ml) in the presence of 10% (v/v) control human serum prior to initiation of

Cng binding (fig. 6.38).

3. Verapamil Binding

Over a concentration range of 0.2-128 nmol of [am-verapamil, a single high-
affinity binding site was observed when synaptosomes were incubated with any
IgG or lgG plus serum (fig. 6.4A,B). Non-specific binding accounted for 40-50 %
of total binding. Scatchard analysis of the data derived from incubation of
synaptosomes with control IgG alone (4 mg/ml; n=3) yielded a K,I value of 19.9
1 7.6 nM and an apparent BM of 106.1 1 12.2 frnol/mg of protein. Following

incubation with LEMS lgG, the value for K, was unchanged at 19.9 1 8.7 nM, but

155

 

 

 

 

 

 

 

m, mm A
l
.3
3200'
O
3.
www'O'LBEW
(5:100 B".-
0 - . 1 . - . - . - .
000 0.10 0.20 0.30 0.40 0.50
[Cng] M
B
300r $611111
4.4
i
3200-
0
§.
X? WN+LBSU3
F\‘w' B.-

 

 

 

J

0 - . - . - . - . -
000 0.10 020 0.30 0.40 0.50
[Cng] lrM

Figure 6.3. Comparative effects of control and LEMS lgG on Cng binding to rat
forebrain synaptosomes. Equal aliquots of synaptosomes were incubated with
control or LEMS lgG (4 mg/ml) alone (A) or control and LEMS IgG (4 mg/ml) in
the presence of 10%2gv/v) human serum (8) 90 min prior to initiation of binding.
Specific binding of [‘ I]-Cng was determined as described in methods. Values
shown are the mean 1 SEM of three experiments. When SE bars are not shown
the SE is smaller than the size of the symbol.

10

 

BOW/FREE
(Limo x 10“)

 

0 30 60 90 120
BOLND (fmol/mg protein)

 

 

 

B
10' ‘Wfl-I-senm
III
"M“
E9
OX
is
in
150

 

BOUD (frnollmg protein)

Figure 6.4. Comparative effects of control and LEMS lgG on verapamil binding to
synaptosomes. Equal aliquots of synaptosomes were incubated with control or
LEMS lgG (4 mg/ml) alone (A) or control and LEMS IgG (4 mg/ml) in the
presence of 10% gV/V) human serum (B) 90 min prior to initiation of binding.
Specific binding of [ H]-verapamil was determined as described in methods. Values
shown are the mean 1 SEM of six experiments. When SE bars are not shown the
SE is smaller than the size of the symbol.

157
the 8mam was decreased significantly to 48.8 1 14.0 frnol/mg protein (n --- 4) (Fig.

6.4A). Similar results were obtained following incubation of synptosomes with
control and LEMS lgG in the presence of control human serum (fig. 6.48). The Kd
values for binding of [aHj-verapamil in the presence of control and LEMS lgG plus
serum were 17.1 1 4.2 and 14.6 1 9.1 nM respectively. The BM value was
reduced from 94.7 1 15.0 to 37.3 1 3.8 fmol/mg of protein (n = 4) by incubation
with LEMS lgG and serum as compared to incubation with control lgG plus serum

(Fig. 8.48).

E. Discussion

The aim of the present study was to to determine whether
pharmacologically distinct binding sites reputed to be associated with Ca“
channels are reduced in number or altered in their affinity for antagonist by acute
exposure to LEMS lgG. Specifically, we sought to determine if the equilibrium
binding of the 032+ channel antagonists [am-nitrendipine, [125']-(t)-CQTX and [3H]-
verapamil to synaptosomal membranes was altered following incubation with
control and LEMS lgG. These studies were performed both in the presence and
absence of control human serum because functional alterations in uptake of 4"’Ca'“
into synaptosomes by LEMS lgG are seen only in the presence of serum (Chapter
Five). It could be that the activated C3 component of complement alters the
binding of Ca“ channel antagonists alone or in conjunction with the LEMS lgG.

Our results indicate that IgG from a patient with LEMS does not alter the

equilibrium binding characteristics of [am-nitrendipine nor [‘25I]-Cng either in the

158

presence or absence of control human serum. Conversely, LEMS lgG both in the
presence and absence of control human serum significantly decreased the B"W
for [3Hj-verapamil while having no effect on its Kd. These results indicate that LEMS
lgG alone interacts with a binding site on synaptosomal membranes that is specific
for verapamil.

Synaptosomes are likely to contain more than one subtype of Ca2+ channel
but the types present have yet to be identified clearly. Synaptosomal Ca2+ influx
is blocked by the phenylalkylamine verapamil (IC50 z 20 pM) (Nachshen and
Blaustein, 1979; Norris et al.,1983) suggesting that L-type channels may be
involved in Ca2+ influx. However, DHPs such as nitrendipine, which are also L-type
antagonists, do not have consistent effects on “Ca“ influx into synaptosomal
preparations. Various investigators report direct block of synaptosomal K*-
stimulated Ca2+ influx and neurotransmitter release by nifedipine, nimodipine and
nitrendipine (Turner and Goldin, 1985; White and Bradford, 1986; Dunn, 1988). Still,
others report no effect of DHPs on K"-stimulated Ca“ uptake into synaptosomes
(Nachshen and Blaustein, 1979; Daniell et al., 1983; Wei and Chang, 1985;
Suszkiw et al., 1986; Reynolds et al., 1986; \fickroy et al., 1992) even though a
high affinity binding site for DHPs exists on synaptosomes (Boles et al., 1984;
Greenberg et al., 1984; Turner and Goldin, 1985; Wei and Chang, 1985; Carvalho
et al., 1986; Suszkiw et al., 1986; Dunn, 1988). In either case, the concentration
of DHP required to block “032* influx into synaptosomes is much greater than the

concentrations which bind specifically to synaptosomes.

159

Previously, LEMS IgG was shown to reduce Ca2+ currents with the
electrophysiological and pharmacological characteristics of L-type channels in
undifferentiated NG108-15 cells (Peers et al., 1990). L-type channels contributing
to Ca2+ currents in adrenal medullary chromaffin cells were reduced by LEMS lgG
as well (Kim and Neher, 1988). In the present study, [aHj-nitrendipine bound to a
single class of high affinity sites in rat forebrain synaptosomes. The values for Kd
and BM correlate well with those observed by others (Boles et al., 1984; Turner
and Goldin, 1985; Suszkiw et al., 1986). The affinity of binding as well as the
maximum number of binding sites were unaltered by prior incubation of
synaptosomes with LEMS lgG in the presence or absence of serum. If L-type
channels do exist in isolated nerve terminals, this would suggest that they are not
altered at the DHP binding site by LEMS lgG nor an activated serum component.
The possibility that the antibody binds to an alternate site on the L-type channel
can not be excluded. These results do correlate with the fact that L-type channels
are not normally involved in transmitter release at motor nerve terminals, however
(Atchison, 1989). '

Synaptosomes contain specific high-affinity receptors for the N-type channel
antagonist, Cng, as well (Abe et al., 1986; Marqueze et al., 1988; Wagner et al.,
1988; Scheer, 1990). Rapid 445Ca2+ flux measurements into rat brain synaptosomes
demonstrates partial antagonism (s20-40%) of voltage-dependent Ca2+ channels
by Cng (Reynolds et al., 1986; Suszkiw et al., 1987; Woodward et al., 1988b;

Suszkiw et al., 1989) suggesting N-type channels may mediate a proportion of

160

Ca2+ uptake and neurotransmitter release from synaptosomes.

In the present study, Cng bound to a single class of high affinity sites on
rat forebrain synaptosomes as did nitrendipine. Like nitrendipine, Cng binding was
unaltered by incubation of synaptosomes with LEMS IgG in the presence or
absence of serum. This suggests that N-type channels in isolated nerve terminals
are not altered at the Cng binding site by LEMS lgG nor an activated serum
component either. This is in agreement with the work of Sher and colleaques
(1989) who demonstrated that LEMS IgG was unable to directly inhibit the binding
of [125']wa to neuroblastoma cell membranes. LEMS IgG was able to
immunoprecipitate voltage-dependent Ca2+ channels prepared from the same
human neuroblastoma cell line which were prelabeled with radiolabeled Cng,
however (Sher et al., 1989). Therefore, the antibody must recognize antigenic
determinants on the N-type Ca2+ channel in the cell line which are separate from
the site occupied by the toxin. This possibility in synaptosomes can not be
excluded. However, it should be noted that, in either case, release of ACh from
mammalian motor nerve terminals is unaltered by Cng (Anderson and Harvey,
1987; Protti et al., 1991).

32+ influx is blocked by the phenylalkylamine

finally, synaptosomal C
verapamil (Nachshen and Blaustein, 1979; Norris et 3!..1983). Likewise, single-
channel recordings of synaptosomal Ca2+ channels in lipid bilayers confirm that
verapamil is a potent blocker of synaptosomal Ca2+ channels (Nelson, 1985). A

verapamil-sensitive Ca2+ current exists at mouse motor nerve terminals as well

161
(Penner and Dryer, 1986; Anderson and Harvey, 1987).

The BM for verapamil binding to synaptosomal membranes was decreased
significantly by prior incubation of synaptosomes with LEMS lgG both in the
presence and absence of serum as compared to synaptosomes incubated with
equivalent concentrations of control IgG with and without serum. The Kd was
unchanged. Verapamil is presumed to act as an L-type antagonist but it is believed
to have a separate binding site from that of the DHPs (Carvalho et al., 1986; Miller,
1987; Glossmann and Striessnig, 1988). More recent evidence suggests that
verapamil may not be as selective for L-type channels as was recently thought. In
cerebellar granule cells (Carboni and Wojcik, 1988) and neocortical neurons
(Mangano et al., 1991) in primary culture, Ca2+ influx resulting from cell
depolarization is only partially inhibited by nitrendipine. Verapamil, in contrast,
completely blocked the depolarization-induced influx suggesting that verapamil
either blocks all subtypes of voltage-sensitive Ca2+ channels or blocks the activity
of a novel Ca2+ channel (Mangano et al.,1991). In any case, these results indicate
that LEMS lgG interacts with a verapamil-binding site on rat brain synaptosomes.
Further experiments will be required to determine whether this binding site
represents an L-type channel only, all subtypes of channels, or a novel voltage-
dependent Ca2+ channel.

Caution, however must be taken when interpreting these data since the
correlation of DHP, Cng or verapamil binding sites correspond to functional Ca2+

channels is unclear. For each compound, the concentration required to block

162

“C3“ influx into synaptosomes is in excess of that which specifically binds to
synaptosomal membranes. Although the pharmacological binding potencies of
Ca2+ channel antagonists do not necessarily correlate with their functional effects,
binding assays nonetheless provide a measure of the ability of LEMS lgG to
compete with known Ca2+ channel antagonists for their binding sites on nerve
terminal membranes. The results from this study demonstrate that lgG from a
patient with LEMS does not alter the equilibrium binding characteristics of [3H]-
nitrendipine nor [mu-Cng but does decrease significantly the BM of [3H]-
verapamil binding. Interaction with the verapamil binding site takes place both in
the presence and absence of serum suggesting that the IgG molecule itself
interacts at this site. Interaction of LEMS lgG at this site may be necessary but is
not sufficient to alter synaptosomal Ca2+ channel function as reductions of “Ca”
uptake are evident only in the presence of both the IgG and the serum (Chapter
5).
F. Note

ro-Agatoxin and FTX, are potent blockers of K*-induced 45Ca2+ uptake into
rat brain synaptosomes suggesting that a P-type channel may also mediate
neurotransmitter release from these terminals (Mintz er al., 1992; Uchitel et al.,
1992). As Agatoxin and FIX are not available commercially and attempts to
procure it elsewhere were unsuccessful, it was not possible in this dissertation to
examine whether LEMS lgG interacted with a binding site specific for either one of

the toxins.

CHAPTER SEVEN

SUMMARY AND CONCLUSIONS

163

164

Summary and Conclusions

The overall aim of this dissertation was to test the hypothesis that a
circulating autoantibody in the plasma of patients with the autoimmune,
neuromuscular disorder Lambert-Eaton Myasthenic Syndrome interferes with nerve
terminal Ca2+ channels associated with neurotransmitter release. The reduction in
nerve-evoked release of transmitter at the NMJ of patients with LEMS is thought
to be caused by a circulating autoantibody to Ca2+ channels of presynaptic motor
nerve terminals. The small size and close anatomical association of motor nerve
terminals with muscle fibers restrict direct measurements of Ca2+ channel function
in intact NMJ preparations. In fact, data from such experiments represent indirect
measurements of Ca2+ channel activity as the postsynaptic response measured
results from many processes ongoing in the presynaptic nerve terminal. Therefore,
studies in this dissertation were performed on isolated nerve terminals derived from
rat brain. Although LEMS patients do not normally exhibit CNS symptomology, it
was our contention that if the LEMS autoantibody could gain access to the calcium
channels of the CNS by incubation in vitro, it might be possible to induce a passive
transfer of the nerve terminal dysfunction to central neurons.

Specfically, the ability of serum, plasma and lgG from patients with LEMS
to interact with synaptosomal Ca2+ channels was examined. In most of the studies,
radiotracer flux measurements using “Ca“ were used to determine Ca2+ channel

activity. In addition, studies were performed to assess the ion channel specificity

165
of the LEMS immunoglobulin as well as to determine the mechanism by which this

antibody may block or alter the function of synaptosomal Ca2+ channels. The final
studies in this dissertation attempted to characterize pharmacologically the
subtype(s) of Ca2+ channel in synaptosomes that is affected by LEMS lgG.

Preliminary experiments characterized the ability of serum and plasma from
several patients with LEMS to block Ca2+ influx into isolated nerve terminals.
Incubation of rat cortical synaptosomes with LEMS serum or plasma reduced
depolarization-induced uptake of “Ca“ as compared to uptake in synaptosomes
that had been treated with control serum or plasma. These deficits were not due
to a decrease in synaptosomal viability as lactate dehydrogenase release from
synaptosomes was not increased over that released by control sera/plasma
(Chapter Three). Lactate dehydrogenase is a cytosolic enzyme that can be
released into the extrasynaptosomal fluid if the membrane is permeabilized. As
LDH is rather large, discrete changes in the synaptosomal membrane would not
be detected by this assay.

As an lgG autoantibody is presumed to mediate the pathogenesis of LEMS,
the effects of isolated lgG derived from the plasma of a single patient with LEMS
on synaptosomal Ca2+ channels, in particular, and synaptosomal ion channels, in
general, were assessed. Although a stronger conclusion could be drawn after
studying multiple patient sera, we feel that the examination of one patient allows
for a more detailed, in depth, investigation. Furthermore, there is precedence in the

literature for detailed analysis of one or two patients.

166
Incubation of synaptosomes with lgG from a patient with LEMS in the

presence of control human serum resulted in a reduction in one sec depolarization-
dependent uptake of “Ca“. This phase of influx is mediated by voltage-dependent
Ca2+ channels (Nachshen and Blaustein, 1980; Suszkiw and O’Leary, 1983) and
has been consistently correlated with neurotransmitter release (Drapeau and
Blaustein, 1983; fioor, 1983; Suszkiw and O’Leary, 1983; Nachshen, 1985; Leslie
31 al., 1985). The antibody appeared specific for Ca2+ channels as it was unable
to disrupt ”Na+ uptake into or 86Rb+ efflux from synaptosomal Na“ or K+
channels, respectively. These deficits, again, were not due to gross alterations of
synaptosomal membrane integrity, as assessed by LDH release, or to alterations
in synaptosomal membrane potential, as measured by potentiometric,
carbocyanine dye fluorescence, but appeared to result from a direct block of or
alteration in Ca2+ channel function (Chapter Four).

In synaptosomes, the presence of serum is important in mediating a
functional change in Ca2+ channel activity by LEMS lgG. In the absence of human
serum, LEMS lgG did not reduce uptake of “Ca“ into synaptosomes. Heat-
inactivating the serum abolished this effect as well. This suggested 3 role for
complement in the pathogenesis of synaptosomal Ca2+ channel dysfunction by
LEMS lgG. Neither the membrane attack complex nor the alternative pathway of
complement contributed to disruption of Ca2+ channel function by LEMS IgG in
synaptosomes. In contrast, the C3 component of the complement pathway

appears to be important in mediating Ca2+ channel dysfunction in synaptosomes

167
following incubation with serum and LEMS lgG (Chapter 5).

The data presented support the hypothesis that a circulating autoantibody
in the plasma of patients with LEMS interferes with nerve terminal Ca2+ channels.
This raised the question of the identity of the subtype(s) of Ca2+ channels affected
by LEMS lgG. To date, there are three well characterized Ca2+ channel types in
DRG neurons: N, L,and T (Nowycky et al., 19853; Tsien, 1988); and one in
mammalian Purkinje cells: P (Uinas et al., 1989). The N- and L-type channels have
been proposed to regulate neurotransmitter release from a variety of central,
peripheral and clonal cell lines while the P-type channel has been implicated in
release of transmitter at the squid giant synapse, mammalian neuromuscular
junction and in synaptosomes. The results from several studies using primary cell
or clonal cell cultures suggest that the L- and N-type Ca2+ channel are putative
antigenic targets for the LEMS autoantibody (Roberts et al., 1985; Login et al.,
1987; Kim and Neher, 1988; De Azipurua et al., 19883,b; Sher 3131., 1989; Peers
et 3!. 1990). However, the type(s) of Ca2+ channel affected by LEMS lgG in the
motor nerve terminal is unclear. The goal of the final study in this dissertation was
to determine whether pharmacological binding sites reputed to be associated with
Ca2+ channels in a nerve terminal preparation were reduced in number or altered
in affinity following acute exposure to LEMS lgG. IgG from a patient with LEMS
does not alter the equilibrium binding characteristics of the L-type antagonist, [3H]-
nitrendipine nor the N-type antagonist, [125']_0ng but does decrease significantly

the maximum number of binding sites of [aHj-verapamil binding. LEMS lgG

168

reduced verapamil binding both in the presence and absence of serum suggesting
the IgG molecule itself interacts at this site. Interaction of LEMS lgG at this site may
be necessary for alterations of Ca“ channel activity but is not sufficient as
reductions in 45Ca2+ uptake are evident only in the presence of both lgG and
serum.

The studies presented in this dissertation were the first to examine
specifically the ability of a circulating humoral factor in serum from patients with
LEMS to interact with nerve terminal Ca2+ channels. A thorough characterization
of the autoantibodies effects at the nerve terminal has strengthened the hypothesis
that Ca2+ channels are the antigenic target of LEMS lgG. In addition, results
described provide new insight into the mechanism of Ca2+ channel dysfunction by
LEMS lgG showing that complement may indeed be important. Furthermore, the
type(s) of nerve terminal 032* channel(s) affected by LEMS lgG have been
explored.

The ability of sera/plasma and lgG derived from LEMS patients to block
Ca2+ influx into a hetereogeneous preparation may represent a common action of
the LEMS autoantibody at all nerve terminals. This may be an unexpected benefit
of these studies as it suggests that Ca2+ channels of nerve terminals secreting
different transmitters may share common epitopes recognized by the LEMS
autoantibody and perhaps share the same type(s) of Ca2+ channel reponsible for
neurotransmitter release. Antigenic similarities between peripheral and central nerve

terminal C32+ channels must exist as an autoantibody whose primary target is the

169

somatic motor nerve terminal 03“ channel can interact and alter activity of central
nerve terminal Ca2+ channels as well.

The preferential action of the LEMS autoantibody for nerve terminal Ca2+
channels as opposed to Na“ or K+ channels suggests that there is no common
determinant or epitope on different voltage-gated channel proteins to which the
antibody can bind. However, the possibility that some patients with LEMS may
synthesize antibodies with different specificities can not be excluded. These
findings are important as the existence of autoantibodies which are specific for
Ca2+ channels involved in the neurosecretory process may serve as molecular
tools for studying the steps involved in quantal transmitter release both in the
peripheral and central nervous system.

The complement component C3 appears to be necessary for mediating the
functional alteration in Ca2+ channel activity following incubation of synaptosomes
with LEMS IgG and control human serum. This finding is in contrast to that of
Lambert and Lennon (1988) who reported that the electrophysiological
characteristics of LEMS could be transferred passively to mice that had been
depleted of CS by cobra venom factor (CVF). It is also inconsistent with the work
of Lang and colleagues (1987) who found that quantal content of the EPPs was
reduced in a murine nerve-muscle preparation following chronic injection of LEMS
F(ab)’2 fragments. F(ab)’2 fragments lack the Fc portion but retain the binding
capacity of the full immunoglobulin molecule. The Fc portion of immunoglobulin is

important in activation of the complement cascade, however.

170

There can be several possible explanations for this discrepancy. first, our
In vitro model may not mimic precisely events occurring In vivo. As this is an
artificial system, this possibility can not be excluded. Next, the possibility that
activated C3 (C3b) may be important in mediating the LEMS pathogenesis In vivo
may exist. C3 depletion induced by CVF in normal animals is relatively short-lived
(Klaus, 1987) and C3 depression must be maintained by repeated injections of
CVF (Lennon and Lambert, 1988). Continual activation of C3 may result in binding
of 03b to a non-specific acceptor, such as the plasma membrane (Law and
Levine, 1977) or to the antigen-antibody complex itself (Capel at al., 1978). finally,
the characteristic electrophysiological abnormalities produced by lgG and F(ab)’2
fragments or the reduction in 4"’Ca“ uptake into synaptosomes following
incubation with IgG and serum may be patient specific. In each study, the IgG or
the F(ab)’2 fragments of lgG from a single patient with LEMS was examined. It may
be that like the triggering factor for LEMS, the mechanism of Ca2+ channel
dysfunction differs between patients. Complement, like small cell carcinoma of the
lung, may be important in mediating the pathogensis of LEMS in some patients but
not in others.

The density of binding sites for [am-verapamil but not [am-nitrendipine nor
[ml1-co-conotoxin was reduced by approximately 45% following incubation with
LEMS lgG alone or LEMS lgG with serum. The affinity of verapamil for the
remaining receptors was not significame altered, however. The reduction of

verapamil binding corresponds nicely with the reduction of depolarization-induced

1 71
“Ca“ uptake observed following incubation of synaptosomes with LEMS lgG and

serum (Chapter 4,5).

Verapamil is presumed to act as an L-type antagonist with a separate
binding site from that of the DHPs (Carvalho et al., 1986; Miller, 1987; Glossmann
and Striessnig, 1988). Although verapamil functions as a potent selective inhibitor
of Ca2+ channels in cardiac, smooth and skeletal muscle, it is capable of blocking
Na+ channels as well as 032* channels in neuronal preparations (Nachshen and
Blaustein, 1979; Norris, 1983; Carvalho et al., 1986). However, the concentration
of verapamil needed to block active Na+ channels (50 to 100 p M) is greater than
that needed to block Ca2+ channels selectively (1 to 20 pM) (Nachshen and
Blaustein, 1979; Norris etaL, 1983; Nelson, 1985). Furthermore, the concentration
range (0.2-128 nM) used in chapter 6 for binding of [am-verapamil is consistent
with that of Carvalho et al.,(1986) demonstrating specific, saturable, high affinity
binding to Ca2+ channels in synaptosomal membranes.

Recent evidence suggests that verapamil may not be as selective for just
L-type Ca2+ channels as was recently thought (Carboni and Wojcik, 1988;
Mangano et al., 1991). In cerebellar granule cells (Carboni and Wojcik, 1988) and
neocortical neurons (Mangano et al., 1991) in primary culture, Ca2+ influx resulting
from cell depolarization is only partially inhibited by nitrendipine but completely
blocked by verapamil suggesting that verapamil may block all subtypes of voltage-
sensitive Ca2+ channels or may block the activity of a novel Ca2+ channel. Further

experiments will be required to determine whether this binding site represents an

172

L-type channel only, all subtypes of channels, or a novel voltage-dependent Ca2+
channel.

It is important to characterize all the effects of LEMS IgG and try to
synthesize a sequence of events that contribute to their pathogenesis. The results
described in this dissertation along with the data which preceded this work indicate
that LEMS autoantibodies are potentially myasthenogenic if they can fix
complement (Chapter 5), cross-link neighboring Ca2+ channels (Lang et al.,1987;
Nagel et al., 1988), or interfere with Ca2+ channels opening or C3“ fluxes (Kim
and Neher, 1988; Hewett and Atchison, 1991; Chapter 4,5). Reduction of the
density of functional channels in the presynaptic membrane of nerve terminals
appears to be the ultimate cause of neuromuscular transmission failure in LEMS
(Fukunaga et al., 1982; 1983; Nagel et al., 1988).

Further progress in clarifying the molecular mechanisms of LEMS will come
by elucidating the autoimmune mechanism at the molecular level. This will involve
the identification of the voltage-dependent Ca2+ channel subunit(s) that are the
targets for the autoantibody and the mapping of B and T cell epitopes with
recombinant and synthetic peptides as soon as the human cDNA's are available.
Detailed knowledge of the amino acid sequences that form these epitopes may

eventually allow for the development of specific immunotherapy.

LIST OF REFEERENCES

174

LIST OF REFERENCES

Abe, T., Koyano, K. Saisu, H., Nishiuchi, Y., and Sakakibara, S. (1986) Binding of
to -conotoxin to receptor sites associated with the voltage-sensitive calcium
channel. Neurosci. Left. 71: 203-208.

Adams, M.E., Bindokas, V.P., Hasegawa, L., and Venema, V.J. (1990) to -Agatoxins:
novel calcium channel antagonists of two subtypes from funnel web spider
toxin (Agelenopsis aperta) venom. J. Biol. Chem. 265: 861 -867.

Adler, F.L., fishman, M., and Dray, S. (1966) Antibody formation initiated in vitro.
J. Immunol. 97: 554-558.

Akerman, K.E.O., and Nicholls, D.G. (1981) Ca2+ transport by intact
synaptosomes: the voltage-dependent Ca2+ channel and reevaluation of the
role of N3"/Ca2+ exchange. Eur. J. Biochem. 117: 491-497.

Allison, A.C., Denman, A.M., and Barnes, RD. (1971) Cooperating and controlling
function of thymus-derived lymphocytes in relation to autoimmunity. Lancet
ii: 135-140.

Anderson, T...l., Carroll, G.J., Avery, S.J., MacDonaldson, LG, and Parkin, P.J.
(1990) Eaton-Lambert myasthenic syndrome: two cases with typical and
atypical features. New Zealand Med. J. 103: 240-242.

Anderson, A.J., and Harvey, M. (1987) to -Conotoxin does not block the verapamil-
sensitive calcium channels at mouse motor nerve terminals. Neurosci. Lett.
82: 1 77-1 80.

Atchison, W.D., Joshi, U., and Thomburg, J.E. (1986) Irreversible suppression of
calcium entry into nerve terminals by methylmercury. J. Pharmacol. Exp.
Ther. 238: 618-624.

Atchison, W.D., and SM. O’Leary (1987) BAY K 8644 increases release of
acetylcholine at the murine neuromuscular junction. Brain Res. 419:315-3-
19.

Atchison, W.D., Adgate, L, and Beaman, CM. (1988) Effects of antibiotics on
uptake of calcium into isolated nerve terminals. J. Pharmacol. Exp. Ther.
245: 394-401.

175

Atchison, W.D. (1989) Dihydropyridine -sensitive and -insensitive components of
acetylcholine release from rat motor nerve terinals. J. Pharmacol. Exp. Ther
251: 672-678.

Augustine, G.J., Charlton, M.J., and Smith, SJ. (1987) Calcium action in synaptic
transmitter release. Ann Rev. Neurosci. 10: 633-693.

Barstad, J.A.B., and G. Lilleheil (1968) Transversely cut diaphragm preparation
from rat. Arch. Int. Pharmacodyn. Ther. 175: 373-390.

Bartschart, D.K., and MP. Blaustein (19853) Potassium channels in isolated
presynaptic nerve terminals from rat brain. J. Physiol. (Land) 361: 419-440.

Bartschart, D.K., and MP. Blaustein (1985b) Calcium-activated potassium channels
in isolated presynaptic nerve terminals from rat brain. J. Physiol. (Land)
361: 441-457.

Benishen, C.G., Sorenson, R.G., Brown, W.E., Krueger, B.K., and Blaustein, MP.
(1988) Four polypeptide components of green mamba venom selectively
block certain potassium channels in rat brain synaptosomes. Mol.
Pharmacol. 34: 152-159.

Bergmeyer, H.U., and Bernt, E., (1974) Lactate dehydrogenase UV assay with
pyruvate and NADH. In Methods In Enzymatic Analysis, vol 2, (H.U.
Bergmeyer ed.), Academic Press, New York, pp. 574-579.

Bever, C.T., Chang, H.W., Penn, A.S., Jaffe, I.A., and Bock, E. (1982) Penicillamine
induced myasthenia gravis: effects of penicillamine on the acetylcholine
receptor. Neurology 32: 1077-1082.

Bever C.T., and Asofsky, R. (1991) Augmented lgG anti-acetylcholine receptor
response following chronic penicillamine administration. J. Neuroimmunol.
35: 131-137.

Bhakdl, S., and Tranum-Jensen, J. (1978) Molecular nature of the complement
lesion. Proc. Natl. Acad. Sci. (USA) 75: 5655-5659.

Bindokas, V.P., Venema, V.J., and Adams, ME. (1991) Differential antagonism of
transmitter release by subtypes of to-Agatoxins. J. Neurophysiol. 66: 590-
601.

Blaustein, MP. (1975) Effects of potassium, veratridine and scorpion venom on
calcium accumulation and transmitter release by nerve terminals in vitro. J.
Physiol. (Land) 247: 617-655.

176

Blaustein, MP, and Goldring, J.M. (1975) Membrane potentials in pinched-off
presynaptic nerve terminals monitored with a fluorescent probe: Evidence
that synaptosomes have potassium diffusion potentials. J. Physiol. (Lond.)
247: 589-615.

Bluestone, R., and Goldberg, LS. (1973) Effect of d-penicillamine on serum
immunoglobulins and rheumatoid factor. Ann. Rheum. Dis. 32: 50-52.

Bokisch, V.A., Dietrich, MP, and MOIler-Eberhard, H.J. (1975) Third component
of Complement (C3): Structural properties in relation to functions. Proc.
Natl. Acad. Sci (USA) 72: 1989-1993.

Boles, R.G., Yamamura, H.I., Shoemaker, H., and Roeske, W.R. (1984)
Temperature-dependent modulation of [3H]-nitrendipine binding by the
calcium antagonists verapamil and diltiazem in rat brain synaptosomes. J.
Pharmacol. Exp. Ther. 229: 333-339.

Bottazzo, G.F., Pujol-Burrel, R., and Hanafusam T. (1983) Role of aberrant HLA-
DR expression and antigen presentation in induction of endocrine
autoimmunity. Lancet ii 1 1 15-1 1 18.

Bowden, REM, and Duchen, LW. (1976) The anatomy and pathology of the
neuromuscular junction. In Handbook of Experimental Pharmacology.
Neuromuscular Junction. (Springer, Z.E., Ed.) Berlin, West Germany pp 23-
97.

Boyd, I.A., and Martin, AR. (19563) Spontaneous subthreshold activity at
mammalian neuromuscular junctions. J. Physiol (Lond.) 132: 61-73.

Boyd, IA, and Martin, A.R. (1956b) The end-plate potential in mammalian muscle.
J. Physiol. (Lond.) 132: 74-91.

Bradford, H.F., Bennett, G.W., and Thomas, A.J. (1973) Depolarizing stimuli and
the release of physiologically active amino acids from suspensions of
mammalian synaptosomes. J. Neurochem. 21: 495-505.

Bradford, HF. (1975) Isolated nerve terminals as an in vitro preparation for the
study of dynamic aspects of transmitter metabolism and release. In
Handbook of Psychopharmacology vol 1, (LL lversen, S.W. lversen, S.H.
Snyder, eds), Plenum Press, New York. pp. 191-252.

177

Brain, W.R., Daniel, PM, and Greenfield, JG (1951) Subacute cortical cerebellar
degeneration and its relation to carcinoma. J. Neurol. Neurosurg. Psychiat.
14: 59-75.

Bretscher, P., and Cohn, M. (1970) A theory of self-nonself discrimination. Science
169: 1042-1049.

Brigant, J.L, and Mallart, A. (1982) Presynaptic currents in mouse motor nerve
endings. J. Physiol. (Lond.) 333: 619-636.

Brose, N., Petrenko, A.G., St'idhof, C., Jahn, R. (1992) Synaptotagmin: a calcium
sensor on the synaptic vesicle surface. Science 256: 1021-1025.

Brush, K.L., Perez, M., Hawkes, M., Pratt, D.R., Hamilton, S.L (1987) Low affinity
binding sites for 1,4-dihydropyridines in mitochondria and in guinea pig
ventricular membranes. Biochem. Pharmacol. 36: 4153-4161.

Bumet, FM. (1959) The clonal selection theory of acquired immunity. Vanderbilt
University Press, Nashville.

Bucknall, R.C., Dixon, A.S., Glick, E.N., Woodland, J., and Sutshi, W.E. (1975)
Myasthenia gravis associated with penicillamine treatment for rheumatoid
arthritis. Br. J. Med. 1: 600-602.

Byerly, L, and S. Hagiwara (1981) Calcium channel. Ann. Rev. Neurosci.
4:69-125.

Callard, R.E., Basten, A., and Waters, LK (1977) Immune function in aged mice:
B cell function. Cell. Immunol. 31: 26-36.

Capel, P.J.A., Groenboer, O., Grosveld, G., and Pondman, K.W. (1978) The
binding of activated C3 to polysaccharides and immunoglobulins. J.
Immunol. 121: 2566-2572.

Carboni, E., and Wojcik, W.J. (1988) Dihydropyridine binding sites regulate calcium
influx through specific voltage-sensitive calcium channels in cerebellar
granule cells. J. Neurochem. 50: 1279-1286.

Carvalho, C.M., Santos, S.V., and Carvalho, AP. (1986 y-aminobutyric acid
release from synaptosomes as influenced by Ca * and Ca"+ channel
blockers. Eur. J. Pharmacol. 131: 1-12.

178

Catterall, WA (1980) Neurotoxins that act on voltage-sensitive sodium channels
in excitable membranes. Annu. Rev. Pharmacol. Toxicol. 20: 14-43.

Catterall, WA (1988) Structure and function of voltage-sensitive ion channels.
Science 242: 50-61.

Ceccarelli, B., Grohovaz, F., and Hurlbut, WP. (1979) Freeze-fracture studies of
the frog neuromuscular junction during intense release of neurotransmitter:
Effects of black widow spider venom and C32*-free solutions on the
structure of the active zone. J. Cell Biol. 81: 163-177.

Ceccarelli, B., and Hurlbut, WP. (1980) Vesicle hypothesis of the release of quanta
acetylcholine. Physiol. Rev. 60: 396-441.

Chad, DA, and Recht, L0. (1991) Neuromuscular Complications of Systemic
Cancer. Neurologic Clinics. 9: 901 -918.

Chalk, C.H., Murray, N.M.F., Newsom-Davis, J., O’Neill, J.H., and Spiro, 8.6.
(1990) Response of the Lambert-Eaton myasthenic syndrome to treatment
of the associated small-cell lung carcinoma. Neurology 40: 1552-1556.

Charlton, MP, and Augustine, G.J. (1990) Classification of presynaptic calcium
channels at squid giant synapse: neither T-, L-, nor N-type. Brain Res. 525:
133-139.

Coérs, C., and Tellerman-Toppet, N. (1976) Morphological and histochemical
changes of motor units in myasthenia. Ann. NY. Acad. Sci. 274: 6-19.

Collier, B., and Macintosh, EC. (1969) The source of choline for acetylcholine
synthesis in a sympathetic ganglion. Can. J. Physiol. Pharmacol. 47: 127-
135.

Cohen, M.W., Jones, OT, and Angelides, KJ. (1991) Distribution of Ca2+
channels on frog motor nerve terminals revealed by fluorescent to-
Conotoxin. J. Neurosci. 11: 1032-1039.

Cotman, C.W., Haycock, J.W., and White, F. (1976) Stimulus-secretion coupling
processes in brain: analysis of noradrenaline and gamma-aminobutyric acid
release. J. Physiol. (Lond.) 254: 475-505.

Cruse, J.M., and Lewis, R.E., Jr. (1987) Regulation of immune reactivity to self.
Concepts Immunopathol. 4: 1-23.

179

Cruse, J.M., and Lewis, R.E., Jr. (1992) Autoimmunity in the age of molecular
biology. Concepts lmmunopathol 8: 1-17.

Cruz, LJ., and Olivera, BM. (1986) Calcium channel antagonists: to-conotoxin
defines a new high affinity site. J. Biol. Chem. 261: 6230-6233.

Cull-Candy, S.G., Miledi, R., Trautmann, A, and Uchitel, 0.0. (1980) On the
release of transmitter at normal, myasthenia gravis and
myasthenic-syndrome affected human endplates. J. Physiol. (Lond)
299:621-638.

Czlonkowska, A. (1977) Myasthenia syndrome during penicillamine treatment. Br.
J. Med. 2: $726-$727.

Dalton, TA, and Bennett, J.C. (1992) Autoimmune disease and the major
histocompatibility complex: Therapeutic implications. Am. J. Med 92: 183-
188.

Daniell, L, Barr, EM, and Leslie, SW. (1983) “C3“ uptake into rat whole brain
synaptosomes unaltered by dihydropyridine calcium antagonist. J.
Neurochem. 41: 1455-1459.

Dawkins, R.L, Zilko, P.J., Carrano, J., Garlepp, J.J., and McDonald, BL (1981)
lmmunobiology of d-penicillamine. J. Rheumatol. 8: $56-$61.

De Aizpurua, H.J., Griesman G.E., Lambert, E.H., and Lennon, VA (19883)
Voltage-dependent Ca * channels in small cell carcinoma are blocked by
autoantibodies from patients with Lambert-Eaton myasthenic syndrome.
Ann. NY. Acad. Sci. 540: 369-371.

De Aizpurua, H.J., Lambert, E.H., Greismann, G.E., Olivera, B.M., and Lennon,
V.A. (1988b) Antagonism of voltage-gated calcium channels in small cell
carcinoma of patients with and without Lambert-Eaton myasthenic
syndrome by autoantibodies, o-conotoxin and adenosine. Cancer Res. 48:
4719-4724.

de Belleroche, J.J., and Bradford, HF. (1973) The synaptosome: An isolated
working neuronal compartment. In Progress In Neurobiology, vol 1,
Pergamon Press, Oxford, pp. 275-298

Dekker, LV., De Graan, Versteeg, D.H.G., Oestreicher, AB, and Gisper, W.R.
(19893) Phosphorylation of B-50 (GAP-43) is correlated with
neurotransmitter release in rat hippocampal slices. J. Neurochem. 52: 24-30.

180

Dekker, LV., DeGraan, P.N.E., Oestreicher, A.B., Versteeg, D.H.G., and Gispen,
W.H. (1989b) Inhibition of noradrenaline release by antibodies to B-50 (GAP-
43) Nature 342: 74-76.

Dekker, LV., De Graan, P.N.E., De Wit, M., Hens, J.J.H., and Gispen, W.H. (1990)
Depolarization-induced phosphorylation of the protein kinase C substrate 8-
50 (GAP-43) in rat cortical synaptosomes. J. Neurochem. 54: 1645-1652.

del Castillo, J. and Katz, 8. (1956) Biophysical aspects of neuromuscular
transmission. Progr. Biophys. 6: 121-170.

DeLorenzo, R.J., Freedman, S.D., Yohe, W.B., and Maurer, SC. (1979) Stimulation
of C32+-dependent neurotransmitter release and presynaptic nerve terminal
protein phosphorylation by calmodulin and a calmodulin-like protein isolated
from synaptic vesicles. Proc. Natl. Acad. Sci. (USA) 76: 1838-1842.

Denys, E.H., Dau, PC, and Hofrnann, WW. (1972) Passive transfer of the Eaton-
Lambert syndrome: A preliminary report. Acta Neurol. Scand. Suppl. 60:
205.

Dighiero, G., Kaushik, A., Poncet, P., and Ge, X.R. (1987) Origin and significance
of autoantibodies. Concepts lmmunopathol. 4: 42-76.

Dodd, P.R., Hardy, J.A., Oakley, A. E., Edwardson, J.A., Perry, E.K., and Delaunoy,
JP (1981) A rapid method for preparing synaptosomes: comparison with
alternative procedures. Brain Res. 226: 107-118.

Dodge, F.A., Miledi, R. and Rahamimoff, R. (1969) Strontium and quantal release
of transmitter at the neuromuscular junction. J. Physiol. (Land) 200: 267-
283.

Dodge, F.A., and Rahamimoff, R. (1967) Co-operative action of calcium ions in
transmitter release at the neuromuscular junction J. Physiol. (Land) 193:
419-432.

Dooley, D.J., Lupp, A., and Herting, G. (1987) Inhibition of central neurotransmitter
release by ot-conotoxin GVIA, a peptide modulator of the N-type
voltage-sensitive calcium channel. Naunyn-Schmied. Arch. Pharmacol.
336:467-470.

Doria, G., D’Agostaro, G., and Poretti, A. (1978) Age-dependent variations in
antibody avidity. Immunology 35: 601-611.

181

Drapeau, P., and Blaustein, MP. (1983) Initial release of [3H]-dopamine from rat
striatal synaptosomes: correlation with calcium entry. J. Neurosci. 3:703-
713.

Duchen, LW. (1970) Changes in motor innervation and cholinesterase localization
induced by botulinum toxin in skeletal muscle of the mouse: differences
between fast and slow muscles. J. Neurol. Neurosurg. Psychiatry. 33: 40-54.

Dunn, S.M.J. (1988) Multiple calcium channels in synaptosomes: Voltage-
dependence of 1,4-dihydropyridine binding and effects on function.
Biochemistry 27: 5275-5281.

Eaton, LM., and Lambert, EH. (1957) Electromyography and electric stimulation
of nerves in diseases of the motor unit: observations on myasthenic
syndrome associated with malignant tumors. J. Amer. Med. Assoc. 163:
1117-1124.

Ellis, S.D., Vlfilliams, M.E., Ways, N.R.,Brenner, R., Sharp, A.H., Leung, A.T.,
Campbell, K.P., McKenna, E., Koch, N.J., Hui, A.,Schwartz, A., and Harpold,
MM. (1988) Sequence and expression of mRNAs encoding the or1 and or2
subunits of a DHP-sensitive calcium channel. Science 241: 1661-1663.

Elmqvist, D., and Lambert, EH. (1968) Detailed analysis of neuromuscular
transmission in a patient with the myasthenic syndrome sometimes
associated with bronchogenic carcinoma. Mayo Clin. Proc., 43: 689-713.

Engel, AG. (1970) Locating motor endplates for electron microscopy. Mayo. Clin.
Proc. 45: 450-458.

Engel, AG, and Santa, T. (1971) Histometric analysis of the ultrastructure of the
neuromuscular junction in myasthenia gravis and in the myasthenic
syndrome. Ann. NY. Acad. Sci. 183: 46-63.

Engel, AE., and Tsujihata, M. (1976) The motor endplate in myasthenia gravis and
in experimental autoimmune myasthenia gravis. A quantitative Ultrastructural
study. Ann. NY. Acad. Sci. 274: 60-79.

Engel, AG., Lindstrom, J.M., Lambert, E.H., and Lennon, VA (1977)
Ultrastructural localization of the acetylcholine receptor on myasthenia gravis
and its experimental autoimmune model. Neurology 27, 307-315.

Engel, A. (1984) Myasthenia gravis and myasthenic syndromes. Ann. Neurol.
16:519-534.

182

Engel, A.G., Fukuoka, T., Lang, 8., Newsom-Davis, J. Vincent, A., and Wray, D.W.
(1987) Lambert-Eaton myasthenic syndrome lgG: Early morphological
effects of IgG and immunolocalization at the motor endplate. Ann. NY.
Acad. Sci. 505: 333-345.

Epstein, O., DeVeiIliers, D., Jain, S., Potter, 8.J., Thomas, H.C., and Sherlock, S.
(1979) Reduction of immune complexes and immunoglobulins induced by
d-penicillamine in primary biliary cirrhosis. New England J. Med. 300: 274-
278.

Fambrough, D., Drachman, 0.8., and Satyamurti, S.(1973) Neuromuscular
transmission in myasthenia gravis: Decreased acetylcholine receptors.
Science 182:293-295.

Fatt, P., and Katz, 8. (1951) An analysis of the end-plate potential recorded with
an intracellular electrode. J. Physiol. (Lond.) 115: 320-331.

Ferroir, J.P., Hervie-Caen, C., Reignier, A., Nicolle, M.H., and Guillard, A. (1989)
Lambert-Eaton’s syndrome: remission, recurrence and satisfactory course
after excision of a laryngeal cancer, without evidence of a tumor recurrence
at autopsy. Rev. Neurol. (Paris) 145: 851 -852.

Floor, E. (1983) Substance P release from K*-depolarized synaptosomes at one
second resolution. Brain Res. 279: 321-324.

Fosset, M., Jaimovich, E., Delpont, E., and Lazdunski, M. (1983) [aHjNitrendipine
receptors in skeletal muscle. Properties and preferential localization in
transverse tubules. J. Biol. Chem. 258: 6068-6092.

Fox, A.P., Nowycky, MC, and Tsien, R.W. (1987) Kinetic and pharmacological
properties distinguishing three types of calcium currents in chick sensory
neurones. J. Physiol.(Lond) 394: 149-172.

Fukunaga, H., Engel, A.G., Osame, M., and Lambert, EH. (1982) Paucity and
disorganization of presynaptic membrane active zones in the Lambert-Eaton
Myasthenic Syndrome. Muscle Nerve 5: 686-697.

Fukunaga, H., Engel, A.G., Lang, 8., Newsom-Davis, J., and Vincent, A. (1983)
Passive transfer of Lambert-Eaton Myasthenic syndrome with lgG from man
to mouse depletes the presynaptic membrane active zones. Proc. Natl.
Acad. Sci. (USA) 80: 7636-7640.

183

Fukuoka, T., Engel, A.G., Lang, 8., Newsom-Davis, J., Prior, C., and Wray. D.W.
(1987) Lambert-Eaton myasthenic syndrome: I. Early morphologic effects
of lgG in the presynaptic active zones. Ann. Neurol. 22: 193-199.

Gadzar, AF., Zweig, M.H., Carney, D.N., Van Steirteghen, A.C., Baylin, 8.8., and
Minna, JD. (1981) Levels of creatine kinase and its 88 isoenzyme in lung
cancer specimens and cultures. Cancer Res. 41: 2773-2777.

Garvey, J.S., Cremer, NE, and Sussdorf, DH. (1977) Methods in Immunology.
Benjamin, W.A. Ed. Reading. Ch. 26.

Glossmann, H., and Striessnig, J. (1988) Structure and pharmacology of voltage-
dependent calcium channels. ISI Atlas Pharmacol. 202:208.

Glavinovic, MI. (1979) Voltage-clamping of unparalyzed cut diaphragm for study
of transmitter release. J. Physiol. (Lond.) 290: 467-480.

Goetz, O. (1987) The Alternative Pathway of Activation. In The Complement
System. (Rother, K and Till, G. eds.) Springer-Verlag Berlin Heidelberg New
York pp. 154-168.

Goetz, O., and Mailer-Eberhard, H.J. (1971) The C3-activator system: an alternative
pathway of complement activation. J Exp. Med. 134, 905-108s.

Goidl, E.A. Aging and autoimmunity. In Aging and the Immune response: Cellular
and humoral aspects. (E.A. Goidl ed.), Marcel Dekker, Inc. New York. pp.
345-358.

Gomm, S.A., Thatcher, N, Barber, P.V., and Cumming, W.J.K. (1990) A
clinicopathological study of the paraneoplastic neuromuscular syndromes
associated with lung cancer. Ouaterly J. Med. 75: 577-595.

Goodnow, C.S., Adelstein, S., and Basten, A. (1990) The need for central and
peripheral tolerence in the B cell repetoire. Science 248: 1373-1378.

Gray, E.G., and Whittaker, VP. (1962) The isolation of nerve endings from brain:
An electron microscopic study of cell fragments derived by homogenization
and centrifugation. J.Anat. 96: 79-96.

Greenberg, D.A., Cooper, EC, and Carpenter, CL (1985) Reversible
dihydropyridine isothiocyanate binding to brain calcium channels. J.
Neurochem. 44: 319-321.

184

Grossman, C.J., Roselle, GA, and Mendenhall, CL (1991) Sex steroid regulation
of autoimmunity. J. Steroid Biochem. Molec. Biol. 40: 649-659.

Guttman, L, Crosby, T.W., Takemori, M., and Martin, FD. (1972) The Eaton-
Lambert syndrome and autoimmune disorders. J. Amer. Med. Assoc. 53:
354-356.

Hagiwara, S., and Byerly, L (1981) Calcium Channel. Annu. Rev. Neurosci. 4: 69-
125.

Hare, M.F., and Atchison, W.D. (1992) Differentiation between alterations in plasma
and mitochondrial membrane potentials in synaptosomes using a
carbocyanine dye. J. Neurochem. 58: 1321-1329.

Harvey, A.M., and Masland, R.L (1941) A method for study of neuromuscular
transmission in human subjects. Bull. Johns Hopkins Hosp. 68: 81-93.

Heath, JR, Ewing, D.J., and Cull, RE. (1988) Abnormalities of autonomic function
in the Lambert-Eaton myasthenic syndrome. J. Neurol., Neurosurg.,
Psychiat. 51: 436-439.

Heide, K., and Schwick, H.G. (1978) Salt fractionation of immunoglobulins. In:
Handbook of Experimental Immunology, vol. 1, Immunochemistry (Weir,
D.M. ed.) Blackwell Scientific Publications, Oxford, pp. 7.1-7.11.

Heinonen, E., Akerman, KE.O., Kaila, K, and Scott, LG. (1985) Dependence of
cytOpIasmic calcium transients on the membrane potential in isolated nerve
endings of the guinea pig. Biochim. Biophys. Acta. 815: 203-208.

Henson, R.A., Russell, 0.8., and Wilkinson, M. (1954) Carcinomatous neuropathy
and myopathy: a clinical and pathological study. Brain 77: 82-121.

Hesselmans, LF.G.M., Jennekens, F.G.I., Kartman, J., Wokke, J.H.J., de Weser,
M., KIaver-Krol, E.G., DeBaets, M., Spaans, F., and Veldman, H. (1992)
Secondary changes of the motor endplate in Lambert-Eaton myathenic
syndrome: a quantitative study. Acta Neuropathol. 83: 202-206.

Heuser, J.E., Reese, T.S., Dennis, M.J., Jan, Y., Jan, L, and Evans, L. (1979)
Synaptic vesicle exocytosis captured by quick freezing and correlated with
quantal transmitter release. J. Cell Biol. 81:275-300.

Heuser, J.E., Reese, TS, and Landis, D.M.D. (1974) Functional changes in frog
neuromuscular junction studied with freeze fracture. J. Neurocyt. 3: 109-131 .

185

Hewett, S. J., and Atchison, W. D. (1991) Serum and plasma from patients with
Lambert-Eaton2 Myasthenic Syndrome reduce depolarization-dependent
uptake of “Ca into rat cortical synaptosomes. Brain Res. 566: 320-324.

Hewett, S. J., and Atchison, W. D. (1992)2 Effect of charge and Iipophilicity on
mercurial-induced disruption of MCa uptake into isolated nerve terminals.
Tox. Appl. Pharmacol. 113: 267-273.

Hirning, LD., Fox, A.P., McCleskey, E.W., Olivera, B.M., Thayer,S.A., Miller, R.J.,
and Tsien, R.W. (1988) Dominant role of N-type Ca channels in evoked
release of norepinephrine from sympathetic neurons. Science 239: 57-61.

Hodgkin, AL, and Huxley, AF. (1952) A quantitative description of membrane
current and its application to conduction and excitation in the nerve. J.
Physiol. (Land) 117: 500-544.

Hoffman, F., Nastainczyk, W., Rohrkasten, A., Schneider, T., and Sieber, M. (1987)
Regulation of the L-type calcium channel. Trends Pharmacol. Sci. 8: 393-
398.

Holz, G.G., Dunlap, K., and Kream, RM. (1988) Characterization of the electrically-
evoked release of substance P form dorsal root ganglion neurons: Methods
and dihydropyridine sensitivity. J. Neurosci. 8: 463-471.

Homo-Delarche, F., Fitzpatrick, F., Christeff, N., Nunez, E.A., Bach, J.F., and
Dardenne, M. (1991) Sex steroids, glucocorticoids, stress and
autoimmunity. J. Steroid Biochem. Molec. Biol. 40: 619-637.

Hostetter, MK, Krueger, R.A., and Schmeling, DJ. (1984) The biochemistry of
opsonization: central role of the reactive thioester of the third component of
complement. J. Infect. Diseases 150: 653-661.

Hubbard, J.I. (1970) Mechanisms of transmitter release. Progr. Biophys. Mol. Biol.
21: 33-124.

Hubbard, J.I. (1973) Microphysiology of the vertebrate neuromuscular junction.
Physiol. Rev. 53: 674-723.

Hubbard, J.I., and Wilson, DP. (1973) Neuromuscular transmission in a
mammalian preparation in the absence of blocking drugs and the effects of
d-tubocurarine. J. Physiol. (Lond.) 228: 307-325.

186

Hume, JR. (1985) Comparative interactions of organic Ca2+ channel antagonists
with myocardial Ca2+ and K+ channels. .1. Pharmacol. Exp. Ther. 234: 134-
140.

Ingram, D.A., Davis, G.R., Schwartz, M.S., Traub, M., Newland, A.C., and Swash,
M. (1984) Cancer-associated myasthenic (Eaton-Lambert) syndrome:
distribution of abnormality and effect of treatment. J. Neurol., Neurosurg.,
Psychiat. 47: 806-812.

lshikawa, K., Engelhardt, J.T., Fujisawa, T., Okamoto, T., and Katsuki, H. (1977)
A neuromuscular transmission block produced by a cancer tissue extract
derived from a patient with the myasthenic syndrome. Neurology 27: 140-
143.

Jablecki, C. (1984) Lambert-Eaton Myasthenic Syndrome. Muscle and Nerve 7:
250-257.

Jacobson, E.B., L’age-Stehr, J., and Herzenberg, LA. (1970) Immunological
memory in mice. J. Exp. Med. 131: 1109-1120.

Joiner, K.A., Brown, E.J., and Frank, MM. (1984) Complement and bacteria:
chemistry and biology of host defense. Annu. Rev. Immunol. 2: 461-472.

Kamp, T.J., and Miller, R.J. (1987) voltage-sensitive calcium channels and calcium
antagonists. ISI Atlas of Sci.: Pharmacol. 133-138

Kasai, H., Aosake, T., and Fukuda, J. (1987) Presynaptic calcium antagonist m-
oonotoxin irreversibly blocks N-type Ca channels in chick sensory neurons.
Neurosci. Res. 4: 228-235.

Kass, RS, and Tsien, R.W. (1975) Multiple effects of calcium antagonists in
plateau currents on cardiac Purkinge fibers. J. Gen. Physiol. 66: 169-192.

Katz, 8., and Miledi, R. (1965) The effect of calcium on acetylcholine release from
motor nerve endings, Proc. Roy. Soc. [Biol] 161: 453-481.

Katz, 8., and Miledi, R. (19673) The timing of calcium action during neuromuscular
transmission. J Physiol. (Lond.) 189: 535-544.

Katz, 8., and Miledi, R. (1967b) A study of synaptic transmission in the absence
of nerve impulses J. Physiol. (Land) 192: 407-436.

Katz, 8., and Miledi, R.(1970) Further study of the role of calcium in synaptic
transmission. J. Physiol. (Land) 207: 789-801.

187

Katz, 8., and Thesleff, S. (1957) On the factors which determine the amplitude of
the miniature endplate potential. J. Physiol. (Lond.) 137: 267-278.

Khurana, R.K., Koski, CL, and Mayer, RF. (1988) Autonomic dysfunction in
Lambert-Eaton myasthenic syndrome. J. Neurol. Sci. 85: 7786.

Kim, Y.I. (1986) Passively transferred Lambert-Eaton Syndrome in mice receiving
purified lgG. Muscle Nerve 9: 523-530.

Kim, Y.I. (1985) Passive transfer of the Lambert-Eaton Myasthenic Syndrome:
neuromuscular transmission in mice injected with plasma. Muscle Nerve 8:
162-172.

Kim, Y.I. (1987) Lambert-Eaton myasthenic syndrome: evidence for calcium
channel blockade. Ann. NY. Acad. Sci. 505: 377-379.

Kim, Y.I., and Neher, E.(1988) IgG from patients with Lambert-Eaton syndrome
blocks voltage-dependent calcium channels. Science 239: 405-408.

Kim, Y.I., Sanders, D.B., Johns, T.R., Phillips, LH., and Smith, RE. (1988)
Lambert-Eaton myasthenic syndrome: lack of short-term in vitro effects of
serum factors on neuromuscular transmission. J. Neurol. Sci. 87: 1-13.

IGmbal, J.W. (1986) Introduction to Immunology. Macmillan Publishing Company,
New York, p. 201.

King, M.L, Bildwell, D., and Shaikh, A. (1983) Coxsackie B virus specific lgM
response in children with IDDM. Lancet i: 1392-1399.

Kisielow, P., Bluthmann, H., Staertz, U.D., Steinmetz, M., and von Boehmer, H.
(1988) Tolerance in T cell receptor transgenic mice involves deletion of
nonmature CD4+CD8+ thymocytes. Nature (Land) 333: 742-746.

Klaus, G.G.B. (1987) The role of complement in the induction of antibody
responses, in The Complement System (Rother, K. and TI", 0., eds) pp.
327-337. Springer-Verlag, New York.

Knaus, H.G., Stressnig, J., Koza, A., and Glossman, H. (1987) Neurotoxic
aminoglycosides are potent inhibitors of [l‘zsj-omega-conotoxin GVIA
binding to guinea-pig cerebral cortex membranes. Naunyn-Schmiedeberg's
Arch. Pharmacol. 336: 583-586.

188

Kongsmut, S., and Miller, R.J. (1986) Nerve growth factor modulates the drug
sensitivity of neurotransmitter release from PC12 cells. Proc. Natl. Acad. Sci.
(USA) 83: 2243-2247.

Krosgrud, R.L, and Perkins, EH. (1977) Age-related changes in T cell function. J.
Immunol. 118: 1607-1611.

Krueger, B.K., and Blaustein, MP. (1980) Sodium channels in presynaptic nerve
terminals. J. Gen. Physiol. 76: 287-313.

Krueger, B.K., Forn, J,. and Greengard, P. (1977) Depolarization-induced
phosphorylation of specific proteins mediated by calcium ion influx in rat
brain synaptosomes. J. Biol. Chem. 252: 2764-2777.

Kuffler, SW, and Yoshikami, D. (1975) The number of transmitter molecules in a
quantum: an estimate from iontophoretic applications of acetylcholine at the
neuromuscular synapse. J. Physiol. (Land) 251: 465-482.

Kuher, M.J., and LC. Murrin (1978) Sodium-dependent, high-affinity choline
uptake. J. Neurochem. 30: 15-21.

Lambert, E.H. and Rooke, ED. (1965) Myasthenic state and lung cancer. In The
remote effects of cancer on the nervous system. (Brian, L and Norris, F.H.,
Eds.) Grune and Statton, Inc. New York pp. 67-80.

Lambert, E.H., and Elmqvist, D. (1971) Quantal components of endplate potentials
in the myasthenic syndrome. Ann. NY. Acad.Sci. 183:183-199.

Lambert, E.H., and Lennon, VA (1988) Selected lgG rapidly induces Lambert-
Eaton Myasthenic Syndrome in mice: complement independence and EMG
abnormalities. Muscle and Nerve 11, 1133-1145.

Lambert, E.H., Rooke, E.D., Eaton, L.M., and Hodgeson, OH. (1961) Myasthenic
syndrome ocassionally associated with bronchial neoplasm:
neurophysiologic studies. In Myasthenia Gravis: The 2nd International
Symposium Proceedings. (Viets, H.R. ed), Charles C. Thomas, Springfield
IL pp. 362-410.

Landau, EM. (1978) Structure and function of the ACh receptor at the muscle end-
plate. Prog. NeurobioI. 10: 253-288.

Lang, 8., Newsom-Davis, J., Wray, D., and Wncent, A.(1981) Autoimmune
aetiology for myasthenic (Eaton-Lambert) syndrome. The Lancet II 224-226.

189

Lang, 8., Newsom-Davis, J., Prior, C., and Wray, D. (19833) Antibodies to motor
nerve terminals: An electrophysiological study of a human myasthenic
syndrome transferred to mouse. J. Physiol. (London) 344: 335-345.

Lang, 8., Molenaar, P.C., Newsom-Davis, J., and Vincent, A. (1984) Passive
transfer of Lambert-Eaton Myasthenic Syndrome in mice: Decrease rates
of resting and evoked release of acetylcholine from skeletal muscle. J.
Neurochem. 42:658-662.

Lang, 8., Newsom-Davis, J., Peers, C., and Wray, D.W. (1987) The action of
myasthenic syndrome antibody fragments on transmitter release in the
mouse. J. Physiol. (Lond.) 390: 173P.

Lang, 8., Vincent, A., Murray, N.M.F., and Newsom-Davis, J. (19893) Lambert-
Eaton myasthenic syndrome : Immunoglobulin G inhibition of Ca2+ flux in
tumor cells correlates with disease severity. Ann. Neurol. 25: 265-271.

Lang, 8., Leys, K., Wncent, A., and Newsom-Davis, J.(1989b) K*-stimulated Ca2+
influx in cell lines derived from small cell lung cancer and neuronal tumors.
Ann. NY. Acad. Sci. 560: 294-296.

Lamb, JR, and Feldman, M. (1984) Essential requirement for major
histocompatibility complex recognition in T cell tolerance induction. Nature
(Land) 308: 72-74.

Law, S.K., Uchtenberg, NA, and Levine RP. (1980) Covalent binding and
hemolytic activity of complement proteins. Proc. Natl. Acad. Sci. (USA) 77:
7194-7198.

Law, SK, and Levine, RP. (1977) Interaction between the third complement
protein and cell surface macromolecules. Proc. Natl. Acad. Sci. (USA) 74:
2701-2705.

Lederberg, J. (1959) Genes and antibodies. Science 129: 1649-1653.

Lennon, V.A., Lambert, E.H., Whittingham, S., and Fairbanks, V. (1982)
Autoimmunity in the Lambert-Eaton Myasthenic Syndrome. Muscle and
Nerve 5: $21-$25.

Leonard, J.P., Nargeot, J., Snutch, T.P., Davidson, N., and Lester, HA (1987) Ca
channels induced in Xenopus oocytes by rat brain mRNA. J. Neurosci. 7:
875-881.

190

Leslie, S.W., Woodward, J.J., and Wilcox, RE. (1985) Correlation of rates of
calcium entry and endogenous dopamine release in mouse striatal
synaptosomes. Brain Res. 325: 99-105.

Leveque, C., Hoshino, T., David, P., Shoji-Kasai, Y., Leys, K., Omori, A., Lang, 8.,
El Far, 0., Sato, K., Martin-Moutot, N., Newsom-Davis, J., Takahashi, M.,
and Seager, M.J. (1992) The synaptic vesicle protein synaptotagmin
associates with calcium channels and is a putative Lambert-Eaton
myasthenic syndrome antigen. Proc. Natl. Acad. Sci. (USA) 89: 3625-3629.

Liebowitz, M.D., Bale, T., Adams, M.E., Venema, V.J., and Cohen, C.J. (1990)
Selective block of atrial L-type Ca2+ channels by the spider toxin to -Aga-IIIA.
Soc. Neurosci. Abstr. 16: 956.

Liley, AW. (1956) The effects of presynaptic polarization on the spontaneous
activity at the mammalian neuromuscular junction. J. Physiol. (Lond.) 134:
427-443.

Lin, J.-W., Rudy, 8., and Uinas, R. (1990) Funnel-web spider venom and a toxin
fraction block calcium current expressed from rat brain mRNA in Xenopus
oocytes. Proc. Natl. Acad. Sci. (USA) 87: 4538-4542.

Lindstrom, J., and Lambert, EH. (1978) Content of acetylcholine receptor in
myasthenia gravis, experimental autoimmune myasthenia gravis and Eaton-
Lambert syndrome. Neurology 8: 130-138.

Lindstrom, J., and Dau, P. (1980) Biology of myasthenia gravis. Ann. Rev.
Pharmacol. Toxicol. 20:337-362.

Uinas, R., Steinberg, I.Z., and Walton, K. (1976) Presynaptic calcium currents and
their relation to synaptic transmission: Voltage clamp study in squid giant
synapse and theoretical model for the calcium gate. Proc. Natl. Acad. Sci.
(USA) 73: 2918-2922.

Uinas, R., Steinberg, I.Z., and Walton, K. (1981) Presynaptic calcium currents in
squid giant synapse. Biophys. J. 33: 289-322.

Uinas, R., McGuinness, T.L, Leonard, C.S., Sugimori, M. and Greengard, P. (1985)
lntraterminal injection of synapsin I or calcium /calmodulin-dependent protein
Kinase ll alters neurotransmitter release at the squid giant synapse. Proc.
Natl. Acad. Sci. (USA) 82: 3035-3039.

191

Uinas, R., Sugimori, M., Lin, J.-W., and Cherksey, B. (1989) Blocking and isolation
of a calcium channel from neruons in mammals and cephalopods utilizing
a toxin fraction (FTX) from funnel-web spider poison. Proc. Natl. Acad. Sci.
(USA) 86: 1689-1693.

Login, I.S., Kim, Y.I., Judd, A.M., Spangelo, BL, and Macleod, RM. (1987)
Immunoglobulins of Lambert-Eaton Myasthenic syndrome inhibit rat pituitary
hormone release. Ann. Neurol. 22: 610-614.

Londei, M., Lamb, J.R., Bottazzo, G.F., and Feldman, M. (1984) Epithelial cells
expressing aberrant MHC class II determinants can present antigens to
cloned human T cells. Nature (Land) 312: 639-641.

Londei, M., Botazzo, G.F., and Feldman, M. (1985) Human T cell clones form
autoimmune thyroid glands: specific recognition of autologous thyroid cells.
Science 228: 85-89.

Loos, M. (1987) “Classical“ Pathway of Activation. In The Complement System.
(Rother, K. and Till, O. eds.), Springer-Verlag Berlin Heidelberg New York
pp. 136-141.

Loud, A.V. (1962) A method for the quantitative estimation of cytoplasmic
structures. J. Cell Biol. 15: 181-197.

Lowry, O.H., Rosebrough, N.J., Farr, AL, and Randall, R.J. (1951) Protein
measurements with the folin phenol reagent. J. Biol. Chem. 193: 265-275.

Lu, C-Z., Link, H., Mo, X-A., Xiao, 8-G., Zhang, Y-L, and Oin, Z. (1991) Anti-
presynaptic membrane receptor antibodies in myasthenia gravis. J. Neurol.
Sci. 102: 39-45.

Lundh, H., Nilsson, O., and Rosen, T. (1983) Novel drug of choice in Eaton-
Lambert syndrome. J. Neurol., Neurosurg. and Psychiat. 46:684.

Lundy, P.M., Hong, A., and Frew, R. (1992) Inhibition of a dihydropyridine, w-
conotoxin insensitive Ca2+ channel in rat brain synaptosomes by venom of
the spider Hololena curta. Euro. J. Pharmacol. 225: 51 -56.

Maggi, C.A., Tramontana, M., Cecconi, R., and Santicioli, P. (1990) Neurochemical
evidence for the involvement of N-type channels in transmitter secretion
from peripheral endings of sensory nerve in guinea pigs. Neurosci. Letts.
114: 203-206.

192

Mallart, A. (1985) Electric current flow inside perineural sheaths of mouse motor
nerves. J. Physiol. (Lond.) 368: 565-575.

Mamdani, M.B., Walsh, R.L, Rubino, F.A., Brannegan, R.T., and Hwang, M.H.
(1985) Autonomic dysfunction and Eaton Lambert syndrome. J. Autonomic
Nerv. Syst. 12: 315-320.

Mangano, T.J., Patel, J., Salama, AL, and Keith, RA (1991) Inhibition of K‘-
evoked [3H]-aspartate release and neuronal calcium influx by verapamil,
diltiazem and dextromethorphan: evidence for non-L/non-N voltage-
sensitive calcium channels. Eur. J. of Pharmacol. 192: 9-17.

Manji, H., Schwartz, MS, and McKeran, RD. (1990) Lambert Eaton syndrome:
autonomic neuropathy and inappropriate antidiuretic hormone secretion in
a patient with small cell carcinoma of the lung. J. Neurol. 237: 324-325.

Marqueze, 8., Martin-Moutot, N. Leveque, C., and Couraud, F. (1988)
Characterization of the to-conotoxin binding molecule in rat brain
synaptosomes and cultured neurons. Mol. Pharmacol. 34: 87-90.

Martin, AR. (1966) Quantal nature of synaptic transmission. Physiol. Rev. 46: 51-
66.

Max, E.E. (1989) Immunoglobulins: Molecular genetics. In W.E. Paul, (Ed.)
Fundamental Immunology 2nd Ed. Raven Press, New York, p. 235.

McCann, F.V., Pettengill, O.S., Cole, J.J., Russell, J.A.G., and Sorenson, GD.
(1981) Calcium spike electrogenesis and other electrical activity in
continuously cultured small cell carcinoma of the lung, Science 212: 1155-
1157.

McCIeskey, E.W., Fox, A.P., Feldman, D.H., Cruz, L.J., Olivera, B.M., Tsien, R.W.,
and Yokishami, D. (1987) o-contoxin: direct and persitant blockade of
specific types of calcium channels in neurons but not muscle. Proc. Natl.
Acad. Sci. (USA) 84: 4327-4331.

McEnvoy, K.M., Vlflndebank, A.J.; Daube, JR, and Low P.A. 3,4-Diaminopyridine
in the treatment of Lambert-Eaton Myasthenic Syndrome. (1989) New Eng.
J. Med. 321: 1567-1571.

McGregor, AM. (1990) Immunoendocrine interactions and autoimmunity. New
Engl. J. Med. 322: 1739-1741 .

193

McOuiIlian, MP, and Johns, R.J. (1967) The nature of the defect in Eaton-Lambert
Syndrome. Neurology 17: 527- 536.

Metcalf, ES, and Klinman, NR. (1976) In vitro tolerance induction in neonatal
murine spleen cells. J. Exp. Med. 143: 1840-1847.

Meunier, F. (1984) Relationship between presynaptic membrane potentials and
acetylcholine release in synaptosomes from Torpedo electric organ. J.
Physiol. (Lond.) 354: 121-137.

Meyer, E.M., Momol, A.E., Kramer, 88., Ross, W.E., and Scott, G. (1986) Effects
of serum fractions from patients with Eaton-Lambert syndrome on rat
cortical synaptosomal [aHjacetyIchoIine release. Biochem. Pharmacol. 35:
3412-3414.

Middlemiss, D.N., and Spedding, M. (1985) A functional correlate for the
dihydropyridine binding site in rat brain. Nature (Lond.) 314: 94-96.

Miller, R.J. (1987) Multiple calcium channels and neuronal function. Science
235:46-52.

Miller, J.F.A.P., Basten, A., Sprent, J., and Cheers, C. (1971) Interaction between
lymphocytes in immune responses. Cell Immunol. 2: 469-495.

Mintz, l.M., Venema, V.J, Adams, ME, and Bean, 8. (1991) Inhibition of N-and L-
type Ca2+ channels by the spider venom toxin 0 -Aga-IllA. Proc. Natl. Acad.
SCI. 88: 6628-6631.

Mintz, l.M., Venema, V.J, Swiderek, K.M., Lee, T.D., Bean, BR, and Adams, ME.
(1992) P-type calcium channels blocked by the spider toxin to-Aga-IVA.
Nature 355: 827-829.

Molenaar, P.C., Newsom-Davis, J., Lang, 8., and Vincent, A. (1987) The
metabolism of acetylcholine in the Lambert-Eaton myasthenic syndrome. In
Cellular and Molecular Basis of Cholinergic Function (Powdall, M.J., and
Hawthorne. J.N. eds) Oxford University Press, Oxford. pp. 757-763.

Molenaar, P.C., Newsom-Davis, J., Polack, R.L, and Vincent, A. (1982) Eaton-
Lambert syndrome: Acetylcholine and choline acetyltransferase content in
skeletal muscle. Neurology 32: 1062-1065.

Moody, T.W., Pert, C.B. Gazdar, A.F., Carney, D.N., and Minna, JD. (1981) High
levels of intracellular bombesin characterize human small-cell lung
carcinoma. Science 214: 1246-1248.

194

Mori, Y., Friedrich, T., Kim, M.-S., Mikami, A., Nakai, J., Ruth, P., Bosse, E.,
Hoffmann, F., fiockerzi, V., Furuichi, T., Mikoshiba, K, lmoto., K, Tanabe,
T. and Numa, S. (1991) Primary structure and functional expression from
complementary DNA of a brain calcium channel. Nature (Land) 350: 398-
402.

Morris, C.S., Esiri, M.M., Marx, A., and Newsom-Davis, J. (1992)
lmmunocytochemical characterizations of small cell lung carcinoma
associated with the Lambert-Eaton myasthenic syndrome. Am. J. Pathol.
140: 839-845.

Nachshen, DA (1984) Selectivity of the Ca binding site in synaptosome Ca
channels. Inhibition of Ca influx by multivalent metal cation. J. Gen.
Physiol. 83: 941-967.

Nachshen, DA (1985) The early time course of potassium-stimulated calcium
uptake in presynaptic terminals isolated from rat brain. J. Physiol. (Lond.)
361: 251-268.

Nachshen, DA, and Oborn, C.J. (1975) The influence of sodium on calcium fluxes
in pinched-off nerve terminals in vitro. J. Physiol. (Lond.) 247: 657-686.

Nachshen, DA, and Blaustein, MP. (1979) The effects of some organic “calcium
antagonists" on calcium influx in presynaptic nerve terminals. Mol.
Pharmacol. 16: 579-586.

Nachshen, DA, and Blaustein, MP. (1980) Some properties of potassium
stimulated calcium influx in presynaptic nerve endings. J. Gen. Physiol. 76:
709-728.

Nachshen, D.A., Sanchez-Armass, S., and Weinstein, AM. (1986) The regulation
of cytosolic calcium in rat brain synaptosomes by sodium-dependent
calcium efflux. J. Physiol. (Lond.) 381: 17-28.

Nagel, A., Engel, A.G., Lang, 8. Newsome-Davis, J., and Fukuoka, T. (1988)
Lambert-Eaton Myasthenic Syndrome lgG depletes presynaptic membrane
active zone particles by antigenic modulation. Ann. Neurol. 24: 552-558.

Nayler, W.G. (1988) The calcium antagonist drugs. Therapeutics 149: 682-686.

Nelson, MT. (1985) Effects of blockers and permeant ions on the properties of
single calcium channels from brain. In: Proteins of Excitable Membranes.
(Hille, 8., and Fambrough, D., eds.) John Wiley and Sons, New York. pp.
215-230.

195

Nelson, MT (1986) Interactions of divalent cations with single calcium channels
from rat brain synaptosomes. J. Gen. Physiol. 87: 201-222.

Nelson-Davis, ME, and Patrick, R.L (1990) DiacylglyceroI-induced stimulation of
neurotransmitter release from rat brain striatal synaptosomes. J.
Neurochem. 54: 662-668.

Nemazee, DA, and Barki, K (1989) Clonal deletion of B lymphocytes in a
transgenic mouse bearing anti-MHC class I antibody genes. Nature (Land)
337: 562-566.

Nerbonne, J.M., and Gurney, AM. (1987) Blockade of Ca2+ and K+ currents in
Bag cell neurons oprIysIa califomica by dihydropyridine Ca2+ antagonists.
J. Neurosci. 7: 882-893.

Newsom-Davis, J., and Murray, N.M.F. (1984) Plasma exchange and immunosup-
pressive drug treatment in the Lambert-Eaton Myasthenic Syndrome.
Neurology 34: 480-485.

Newsom-Davis, J. (1985) Lambert-Eaton Myasthenic Syndrome. Springer SemIn.
lmmunopathol. 8:129-140.

Newsom-Davis, J., Murray, N., Wray, D., Lang, 8., Prior, C., and Vincent, A. (1982)
Lambert-Eaton Myasthenic Syndrome: Electrophysiological evidience for a
humoral factor. Muscle and Nerve 5: $17-$20.

Newsom-Davis, J., Leys, K. Vincent, A., Ferguson, l., Modi, G. and Mills, K. (1991)
Immunological evidence for the co-existence of the Lambert-Eaton
myasthenic syndrome and myasthenia gravis in two patients. J. Neurol.,
Neurosurg., and Psychiat. 54: 452-453.

Nichols, R.A., Chilcote, T.J., Czemik, A.J., and Greengard, P. (1992) Synapsin I
regulates glutamate release from rat brain synaptosomes. J. Neurochem.
58: 783-785.

Nordin, AA, and Makinodan, T. (1974) Humoral immunity in aging. Fed. Proc. 33:
2033-2035.

Norris, P.J., Dhaliwal, D.K, Druce, DP, and Bradford, H. F. (1983) The
suppression of stimulus-evoked release of amino acid neurotransmitters
from synaptosomes by verapamil. J. Neurochem. 40: 514-521.

196

Nossal, G.J.V., Cunningham, A., Mitchell, G.F., and Miller, G.F.A.P. (1968)
Chromasomal markers analysis of single antibody-forming cells in
reconstituted, irradiated or thymectomized mice. J. Exp. Med. 128: 839-850.

Nossal, G.J.V. (1983) Cellular mechanisms of immunologic tolerance. Annu. Rev.
Immunol. 1: 33-62.

Nossal, G.J.V. (1989) Immunologic tolerance: collaboration between antigen and
lymphokines. Science 245: 147-153.

Nowycky, M.C., Fox, A.P., and Tsien, R.W. (19853) Three types of neuronal
calcium channel with different calcium agonist sensitivity. Nature (Lond.)
316: 440-443.

Nowycky, M.C., Fox, A.P., and Tsien, R.W. (1985b) Long-opening mode of gating
of neuronal calcium channels and its promotion by the dihydropyridine
calcium agonist Bay K 8644. Proc. Nat. Acad. Sci. (USA) 82: 2178-2182.

O’Neill, J.H., Murray, N.M.F., and Newsom-Davis, J. (1988) The Lambert-Eaton
Myasthenic Syndrome: A review of 50 cases. Brain 111: 577-596.

Oh, S.J., Dwyer, OS, and Bradley, R.J. (1987) Overlap myasthenic syndrome:
combined myasthenia gravis and Eaton-Lambert syndrome. Neurology 37:
1411-1414.

Oh, S.J., and Kim, KW. (1973) Guanidine hydrochloride in the Eaton-Lambert
Syndrome. Neurology 23: 1084-1090.

Oh, SJ. (1990) SFEMG improvement with remission in cancer-associated Lambert-
Eaton Myasthenic Syndrome. Muscle and Nerve 12: 844-848.

Olivera, B.M., Gray, W.R., Ziekus, R., McIntosh, J.M., WCtoria di Santos, J.R., and
Cruz, LJ. (1985) Peptide neurotoxins from fish hunting cone snails. Science
230: 1338-1343.

Olivera, B.M., McIntosh, J.M., Cruz, LJ., Luque, FA, and Gray, W.R. (1984)
Purification and sequence of a presynaptic toxin from Conus geographus
venom. Biochemistry 23:5087-5090.

Parker, DC, and Eynon, E.E. (1991) Antigen presentation in acquired
immunological tolerance. FASEB J. 5: 2777-2784.

197

Pascuzzi, RM, and Kim, Y.I. (1988) In vivo nerve conduction studies in Swiss-
Webster mice with passively transferred Lambert-Eaton Syndrome. Ann. NY.
Acad. Sci. 540: 784-787.

Peers, C., Lang, 8., Newsom-Davis, J., and Wray, D.W. (1987) Selective action of
Lambert-Eaton myasthenic syndrome antibodies on calcium channels in a
rodent neuroblastoma x glioma hybrid cell line. J. Physiol. (Lond.) 421: 293-
308.

Penner, R., and Dreyer, F. (1986) Two different presynaptic calcium currents in
mouse motor nerve terminals. Pflugers Arch. 406: 190-197.

Perin, M.S., Fried, V.A, Mignery, G.A., Jahn, R., and Sudhof, TC. (1990)
Phospholipid binding by a synaptic vesicle protein homologous to the
regulatory region of protein kinase C. Nature (Land) 345: 260-263.

Peris, P., Del Olme, J., Gratacos, J., and Munoz, J. (1990) The Lambert-Eaton
Syndrome in association with rheumatoid arthritis. Brit. J. Rheum. 29: 75-77.

Perney, T.M., Hirning, LD., Leeman, SE, and Miller, R.J. (1986) Multiple calcium
channels mediate transmitter release from peripheral neurones. Proc. Natl.
Acad. Sci. (USA) 83:6656- 6659.

Porter, RR, and Reid, KB.M. (1978) The biochemistry of complement. Nature 275:
699-704.

Posner, J.B. (1991) Paraneoplastic syndromes. Neurologic Clinics 9: 919-936.

Potter, LT. (1970) Synthesis, storage and release of 1‘c-acetylcholine in isolated
rat diaphragm muscles. J. Physiol. (Land) 206: 145-166.

Price, 68., and Makinodan, T. (1972) Immunologic deficiencies in senescence. I
Characterization of intrinsic deficiencies. J. Immunol. 108: 403-412.

Prior, 0., Lang, 8., Wray, D., and Newsome-Davis, J. (1985) Action of
Lambert-Eaton Myasthenic Syndrome lgG at mouse motor nerve terminals.
Ann. Neurol. 17: 587-592.

Protti, D.A., Szczupak, L, Scornik, PS, and Uchitel, 0D. (1991) Effect of to-
conotoxin GVIA on neurotransmitter release at the mouse neuromuscular
junction. Brain Res. 557: 336-339.

Pumplin, D.W., Reese, TS, and Uinas, R. (1981) Are the presynaptic membrane
particles the calcium channels. Proc. Natl. Acad. Sci (USA) 78: 7210-7213.

198

Reno, S.G., Holz, LV., and Dunlap, K (1987) Dihydropyridine inhibition of neuronal
calcium current and Substance P release. Pflt'igers Arch. 409: 361-366.

Rehn, S., and Betz, H. (1984) Solubilization and characterization of the 6-
bungarotoxin binding protein of chick brain membranes. J. Biochem. 259:
6865-6869.

Reynolds, I.J., Wagner, J., Snyder, S.H., Thayer, S.A., Olivera, B.M., and Miller,
R.J. (1986) Brain voltage-sensitive calcium channel subtypes differntiated by
to -conotoxin fraction GVIA. Proc. Natl. Acad. Sci. (USA) 83: 8804-8807.

Rivier, J., Galgean, R., Gray, W.R., Azimi-Zonooz, A., McIntosh, J.M., Cruz, LJ.,
and Olivera, 8. (1987) Neuronal calcium channel inhibitors. J. Biol. Chem.
262: 1194-1198.

Roberts, A., Perrera, S., Lang, 8., Vincent, A., and Newsom-Davis, J. (1985)
Paraneoplastic myasthenic syndrome lgG inhibit “Ca“ flux in a human
small cell carcinoma line. Nature (London) 317: 737-739.

Robinson, P.J., and Dunkley, RR. (1985) Depolarization-dependent protein
phosphorylation and dephosphorylation in rat cortical synaptosomes is
modulated by calcium. J. Neurochem. 44: 338-348.

Robinson, P.J., Hauptschein, R., Lovenberg, W., and Dunkley, PR (1987)
Dephosphorylation of synaptosomal proteins P96 and P139 is regulated by
both depolarization and calcium, but not by a rise in cytosolic calcium
alone. J. Neurochem. 48: 187-195.

Robitaille, R.R., Adler, EM, and Charlton, MP. (1990) Strategic location of calcium
channels at transmitter release sites of frog neuromuscular synapses.
Neuron 5: 773-779.

Roit, l.M. (19883) The acquired immune response: development. In Essential
Immunology. Blackwell Scientific Publications, Boston, MA pp. 134-137.

Roit, l.M. (1988b) Autoimmune Diseases. In Essential Immunology. Blackwell
Scientific Publications, Boston, MA. pp. 238-251.

Rooke, E.D., Eaton, LM., Lambert, E.H., and Hodgeson, OH. (1960) Myasthenia
and malignant intrathoracic tumor. Med. Clin. N. Amer. 44: 977-988.

Rose, NR. (1988) Current concepts of autoimmune diseases. Transpl. Proc. 20:
3-10.

199

Rose, NR. (1989) Pathogenic mechanisms in autoimmune diseases. Clin. Immun.
lmmunopath. 53: S7-S16.

Rosenkrantz, K, Dupont, 8., and Flomenberg, N, (1987) Relevance of
autocytotoxic and autoregulatory lymphocytes in the maintenance of self
tolerance. Concepts lmmunopathol. 4: 24-41.

Rother, U. (1987) lmmunopathological Aspects of the Complement System. In The
ComMement System.(Rother, K and Till, G. eds.) Springer-Variag Berlin
Heidelberg New York pp. 504-509.

Rubenstein, A.E., Horowitz, SH, and Bender, AN. (1979) Cholinergic
dysautonomia and Eaton-Lambert syndrome. Neurology 29: 720-723.

Sahashi, K, Engel, A.G., Undstrom, J.M., Lambert, E.H., and Lennon, VA (1978)
Ultrastructural localization of immune complexes (lgG & C3) at the endplate
in experimental autoimmune myasthenia gravis. J. Neuropath. Exp.
Neurology 37: 212-223.

Sano, K, Enomoto, K, and Maeno, T. (1987) Effects of synthetic to-conotoxin., a
new type of Ca2+ antagonist on frog and mouse neuromuscuIar
transmission. Eur. J. Pharmacol. 141:235-241.

Scheer, H.W. (1990) Interactions between the presynaptically active neurotoxins
a-latrotoxin and to-conotoxin GVIA: studies on calcium fluxes and binding
parameters in rat and chick synaptosomes. Can. J. Physiol. Pharmacol. 68:
1049-1054.

Schwartz, MS, and Stalberg, E. (1975) Myasthenia gravis with features of the
myasthenic syndrome. Neurology 25: 80-84.

Schwartz, RH. (1987) Basic Features of cellular interactions in the immune system.
Ann. NY. Acad. Sci. 505: 1-8.

Schwartz, RH. (1989) Acquisiton of immunologic self-tolerance. Cell 57: 1073-
1081.

Scott, R. H., Dolphin, AC, Bindokas, V. P., and Adams, M. E. (1990) Inhibition of
neuronal Ca2+ channels by the funnel web spider toxon, o-Aga-IA. Mol.
Pharmacol. 38: 711-718.

Seabrook, GR, and Adams, DJ. (1989) Inhibition of neurally-evoked transmitter
release by calcium channel antagonists in rat parasympathetic ganglia. Brit.
J. Pharmacol. 97: 1125-1136.

200

Serge, D., and Serge, M. (1987) Suppressor cells in aging. In Aging and the
Immune response: Cellular and humoral aspects. (Goidl, E.A., ed.) Marcel
Dekker, Inc. New York. pp. 271-278.

Sha, W.C., Nelson, C.A., Nev/perry, R.D., Kranz, D.M., Russell, J.H., and Loh, D.Y.
(1988) Positive and negative selection of an antigen receptor on T cells in
transgenic mice. Nature (Land) 336: 73-76.

Shafer, T.J., and Atchison, W.D. (1989) Block of “Ca uptake into synaptosomes
by methylmercury: Ca- and Na-dependency J. Pharmacol. Exp. Ther. 248:
696-702.

Shafer, T.J., Contreras, ML, and Atchison W.D. (1990) Characterization of
interactions of methylmercury with Ca + channels in synaptosomes and
pheochromocytoma cells: radiotracer flux and binding studies. Mol.
Pharmacol. 38: 102-113.

Sher, E., Pandiello, A., and Clementi, F. (1990) Voltage-gated calcium channels in
small cell lung carcinoma cell lines: Pharmacological, functional and
immunological properties. Cancer Res. 50: 3892-3896.

Sher, E., Gotti, G., Canal, N., Scopetta, C., Piccolo, G., Evoli, A., and Clementi, F.
(1989) Specificity of calcium channel autoantibodies in Lambert-Eaton
Myasthenic Syndrome. The Lancet ii 640-643.

Shoenfield, Y., and lsenberg, D. (19893) The mosaic of autoimmunity. Elsevier,
Amsterdam. pp 144-147.

Shoenfield, Y., and lsenberg, D. (1989b) The mosaic of autoimmunity. Elsevier,
Amsterdam. pp 321-366.

Sihra, T.S., Wang, J.K.T., Gorelick, PS, and Greengard, P. (1989) Translocation
of synapsin I in response to depolarization of isolated nerve terminals. Proc.
Natl. Acad. Sci. (USA) 86: 8108-8112.

Sihra, T.S., Bogonez, E., and Nicholls, D.G. (1992) Localized Ca2+ entry
preferentially effects protein dephosphorylation, phosphorylation, and
glutamate release. J. Biol. Chem. 267: 1983-1989.

Silinsky, EM. (1981) On the calcium receptor that mediates depolarization-
secretion coupling at cholinergic motor nerve terminals. Br. J. Pharm. 73:
413-429.

201

Sinha, AA, Lopez, M.T., and McDevit, HQ. (1990) Autoimmune diseases: the
failure of self-tolerance. Science 248: 1380-1388.

Sim, R.B., Twose, T.M., Paterson, OS, and Sim, E. (1981) The covalent-binding
reaction of complement component C3. Biochem J. 193: 115-127.

Smith, S.J., and Augustine, G.J. (1988) Calcium ions, active zones and synaptic
transmitter release. Trends In Neurosci. 11: 458-464.

Smith, S.J., Augustine, G.J., and Charlton, MP. (1985) Transmission at voltage-
clamped giant synapse of the squid: Evidence for cooperativity of
presynaptic calcium action. Proc. Natl. Acad. Sci. (USA), 82: 622-625.

Snutch, T.P., Leonard, J.P., Gilbert, M.M., Lester, HA and Davidson, N. (1990) Rat
brain expresses a heterogeneous family of calcium channels. Proc. Natl.
Acad. Sci. (USA) 87: 3391-3395.

Snutch, T.P., Tomlinson, W.J., Leonard, J.P., and Gilbert, MM. (1991) Distinct
calcium channels are generated by alternative splicing and are differentially
expressed in mammalian CNS. Neuron 7: 45-57.

Sprent, J., Lo, D., Gao, E.-K, and Ron, Y. (1988) T cell selection in the thymus.
Immunol. Rev. 101: 173-190.

Stojilkovic, S.S.; Izumi, S-i, and Catt, KJ. (1988) Participation of voltage-sensitive
calcium channels in pituitary hormone release. J. Biologic. Chem. 263:
13054-13061.

Stossel, T.P, Field, R.J., Gitlin, J.D., Alper, CA, and Rosen, ES. (1975) The
opsonic fragment of the third component of human complement (C3). J.
Exp. Med. 141: 1329-1347.

Streib, E.W., and Rothner, AD. (1981) Eaton-Lambert Syndrome: Long term
treatment of three patients with prednisone. Ann. Neurol. 10:448-453.

Suszkiw, J8, and O'Leary, ME. (1983) Temporal characteristics of potassium
-stimulated acetylcholine release and inactivation of calcium influx in rat
brain synaptosomes. J. Neurochem. 41:868-873.

Suszkiw, J.B., O’Leary, M.E., Murawsky, MM, and Wang, T. (1986) Presynaptic
calcium channels in rat cortical synaptosomes: Fast kinetics of phasic
calcium influx in rat brain synaptosomes. J. Neurosci., 6: 1349-1357.

202

Suszkiw, J.B., Murawsky, MM, and Fortner, RC. (1987) Heterogeneity of
presynaptic calcium channel revealed by species differences in the
sensitivity of synaptosomal “Ca entry to to -conotoxin. Biochem. Biophys.
Res. Comm. 145: 1283-1286.

Suszkiw, J.B., Murawsky, MM, and Shi, M. (1989) Further characterization of
phasic influx in rat cerebrocortical synaptosomes: inferences regarding
calcium channel type(s) in nerve endings. J. Neurochem. 52: 1260-1269.

Swift, TR. (1981) Disorders of neuromuscular transmission other than Myasthenia
Gravis. Muscle and Nerve 4: 334-353.

Takata, Y., Tamura, N., and Fujita, T. (1984) Interaction of 03 with antigen-antibody
complexes in the process of solubilization of immune precipitates. J.
Immunol. 132: 2531-2537.

Takeuchi, A., and Takeuchi, N. (1959) Active phase of frog’s end-plate potential J.
Neurophysiol. 22: 395-411 .

Takeuchi, N. (1963) Some properties of the conductance change at the endplate
during the action of acetylcholine. J. Physiol. (Lond.) 167: 128-139.

Tamkun, MM, and Catterall, WA (1981) Ion flux studies of voltage-sensitive
sodium channels in synaptic nerve ending particles. Mol. Pharmacol. 19: 78-
86.

Taphoorn, M.J.B., Vanduijn, H., and Wolters, E.C.H. (1988) A neuromuscular
transmission disorder: combined myasthenia gravis and Lambert-Eaton
myasthenic syndrome in one patient. J. Neurol. Neurosurg. Psychiat. 51,
880-882.

Tapia, F.J., Barbosa, A.J.A., Marangos, P.J., Polak, J.M., Bloom, S.R., Derrnondy,
C., and Pearse, A.G.E. (1981) Neuron-specific enolase is produced by
neuroendocrine tumours. The Lancet ii 808-811.

Tarelli, F.T., Passafaro, M., Clementi, E, and Sher, E. (1991) Presynaptic
localization of o-conotoxin-sensitive calcium channels at at the frog
neuromuscular junction. Brain Res. 547: 331-334.

Tonegawa, S. (1983) Somatic generation of antibody diversity. Nature (Lond.)
302: 575-581.

203

Tsien, R.W., Lipscombe, D., Madison, D.V., Bley, K.R., and Fox, AP. (1988)
Multiple types of neuronal calcium channels and their selective modulation.
Trends In Neurosci. 10: 431 -438.

Tsien, R.W., Ellinor, PT, and Home, WA (1991) Molecular diversity of voltage-
dependent Ca2+ channels. Trends Pharmacol. Sci. 12: 349-354.

Turner, T.J., and Goldin, SM. (1985) Calcium channels in rat brain synaptosomes:
Identification and pharmacological characterization. J. Neurosci. 5(3):841-
849.

Tyan, ML (1977) Age-related decrease in mouse T cell progentiors. J. Immunol.
118: 846-851.

Uchitel, O.D., Protti, D.A, Sanchez, V., Cherksey, 8.D., Sugimori, M., and Uinas,
R. (1992) P-type voltage-dependent calcium channel mediates presynaptic
calcium influx and transmitter release in mammalian synapses. Proc. Natl.
Acad. Sci. (USA) 89: 3330-3333.

Unanue, ER. and Askonas, BA (1968) Immune response of mice to antigen in
macrophages. Immunology 15: 287-296.

Unanue, ER, and Cerottini, J-C. (1989) Antigen presentation. FASEB J. 3: 2496-
2502.

Venema, V.J., Swiderek, KM., Lee, T.D., Hathaway, GM, and Adams, ME. (1992)
Antagonism of synaptosomal calcium channels by subtypes oft.) -Agatoxins.
J. Biol. Chem. 267: 2610-2615.

chroy, T.W., Schneider, C.J., Hildreth, J.M. (1992) Pharmacological heterogeneity
among calcium channels that subserve acetylcholine release in vertebrate
forebrain. Neuropharmacol. 31: 307-309.

\hncent, A., Lang, 8., and Newsom-Davis, J. (1989) Autoimmunity to the voltage-
gated calcium channel underlies the Lambert-Eaton myasthenic syndrome,
a paraneoplastic disorder. Trends Neurosci. 12: 496-502.

Vincent, A, Cull-Candy, S.G., Newsom-Davis, J., Trautmann, A., Molenaar, PC,
and Polack, R.L (1981) Congenital myasthenia: endplate acetylcholine
receptors and electrophysiology in five cases. Muscle and Nerve 4: 306-
318.

204

Wagner, J.A., Snowman, A.M., Buswas, A., Olivera, B.M., and Snyder, SH. (1988)
to-conotoxin binding to a high affinity receptor in brain: characterization,
calcium sensitivity, and solubilization. J. Neurosci. 8: 3354-3359.

Wei, J.-W., and Chang, D.-H. (1985) Studies of [3H]nitrendipine binding and KCL-
induced calcium uptake in rat cortical synaptosomes. Gen. Pharmacol. 16:
211-216.

Weigle, W. (1971) Recent observations and concepts in immunological
unresponsiveness and autoimmunity. Clin. Exp. Immunol. 9: 437-447.

Weigle, W. (1973) Immunological unresponsiveness. Adv. Immunol. 16: 61-12.

White, E.J., and Bradford, HP. (1986) Enhancement of depolarization-induced
synaptosomal calcium uptake and neurotransmitter release by Bay K 8644.
Biochem. Pharmacol. 35: 2193-2197.

\Mse, RP, and MacDermot, V. (1962) A myasthenic syndrome associated with
bronchial carcinoma. J. Neurol. Neurosurg. Psychiat. 25: 31 -39.

Woodward, J.J., and Leslie, SW. (1986) Bay K 8644 stimulation of calcium entry
and endogenous dopamine release in rat striatal synaptosomes
antagonized by nimodipine. Brain Res. 370: 397-400.

Woodward, J.J., Cook, ME, and Leslie, S.W. (19883) Characterization of
dihydropyridine-sensitive calcium channels in rat brain synaptosomes. Proc.
Natl. Acad. Sci. 85: 7389-7393.

Woodward, J.J., Rezazadeh, SM, and Leslie, S.W. (1988b) Differential sensitivity
of synaptosomal calcium entry and endogenous dopamine release to to-
conotoxin. Brain Res. 475: 141-145.

Wray, D.W., Peers, 0., Lang, 8., Lande, S., and Newsom-Davis, J. (1987)
Interference with calcium channels by Lambert-Eaton myasthenic syndrome
antibody. Ann. NY Acad Sci. 505: 368-376.

Wray, D.W., Lang, 8., Newsom-Davis, J., and C. Peers (1989) Antibodies against
calcium channels in Lambert-Eaton Myasthenic Syndrome. Ann. NY Acad
Sci. 560: 269-277.

Zucker, RS, and Fogelson, AL. (1986) Relationship between transmitter release
and presynaptic calcium influx when calcium enters through discrete
channels. Proc. Natl. Acad. Sci. (USA) 83: 3032-3036.

"71111111111101IS