Depression of neurotransmitter release during repetitive stimulation at the frog neuromuscular junction

MISSING IMAGE

Material Information

Title:
Depression of neurotransmitter release during repetitive stimulation at the frog neuromuscular junction
Physical Description:
x, 149 leaves : ill. ; 29 cm.
Language:
English
Creator:
Sosa, Maria A., 1965-
Publication Date:

Subjects

Subjects / Keywords:
Research   ( mesh )
Rana pipiens -- physiology   ( mesh )
Neuromuscular Junction -- physiology   ( mesh )
Calcium -- metabolism   ( mesh )
Muscle, Skeletal -- physiology   ( mesh )
Neurotransmitters -- metabolism   ( mesh )
Curare -- pharmacology   ( mesh )
Glycerol -- pharmacology   ( mesh )
Peptides, Cyclic -- pharmacology   ( mesh )
Department of Neuroscience thesis Ph.D   ( mesh )
Dissertations, Academic -- College of Medicine -- Department of Neuroscience -- UF   ( mesh )
Genre:
bibliography   ( marcgt )
non-fiction   ( marcgt )

Notes

Thesis:
Thesis (Ph.D.)--University of Florida, 1993.
Bibliography:
Bibliography: leaves 134-148.
Statement of Responsibility:
by Maria A. Sosa.
General Note:
Typescript.
General Note:
Vita.

Record Information

Source Institution:
University of Florida
Rights Management:
All applicable rights reserved by the source institution and holding location.
Resource Identifier:
aleph - 002294565
oclc - 49802567
notis - ALP7773
System ID:
AA00011801:00001


This item is only available as the following downloads:


Full Text













DEPRESSION OF NEUROTRANSMITTER RELEASE DURING
REPETITIVE STIMULATION AT THE FROG
NEUROMUSCULAR JUNCTION










By

MARIA A. SOSA


A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL
OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT
OF THE REQUIREMENTS FOR THE DEGREE OF
DOCTOR OF PHILOSOPHY

UNIVERSITY OF FLORIDA


1993













ACKNOWLEDGEMENTS


I would like to thank my mentor and advisor, Dr. Janet E. Zengel,

for all her guidance and unwavering support. It has been through her

dedication and example that I have learned what the essential

characteristics of a good scientist are. The confidence she has always

fostered in me and the knowledge that I have obtained from her, both of

concepts and principles of physiology as well as of the skills

essential for survival as a woman scientist in the world of academic

research today, will serve as invaluable tools in the development of my

academic career. I also wish to thank the other members of my

committee, Dr. Peter A. V. Anderson, Dr. Richard D. Johnson, Dr. Bruce

E. Hunter and Dr. Susan Suarez, for their support and helpful

suggestions.

Special thanks are also due to the professors at the Catholic

University of Puerto Rico who first introduced me to the field of

neuroscience and encouraged me to pursue a career in scientific

research.

Finally, and most importantly, I wish to thank my family, for

always expecting the best from me, for always believing in me and for

all the support and encouragement they have given me throughout all

these years. I would not be where I am today had it not been for

having them by my side every step of the way.














TABLE OF CONTENTS

page

ACKNOWLEDGEMENTS....... ......................... .... ..... ....... ii

LIST OF TABLES.................. .. ........ ................ v

LIST OF FIGURES................................................. vi

ABSTRACT............................ ........................... ix

CHAPTERS

1 INTRODUCTION........................ .. .. .. ................ 1

Chemical Synaptic Transmission.............................. 1
Stimulation-Induced Changes in Synaptic Efficacy............ 2

2 EFFECT OF D-TUBOCURARINE ON STIMULATION-INDUCED CHANGES
IN SYNAPTIC TRANSMISSION.......... ...................... 8

Introduction............................ ....... .............. 8
Materials and Methods...................................... 9
Results .................................................. 19
Discussion ............................................... 33
Notes .................................................... 36

3 EFFECT OF GLYCEROL TREATMENT ON STIMULATION-INDUCED
CHANGES IN SYNAPTIC TRANSMISSION ......................... 37

Introduction ................................................ 37
Materials and Methods..................................... 38
Results .................................................. 39
Discussion ................................................ 52

4 USE OF MU-CONOTOXIN GIIIA FOR THE STUDY OF SYNAPTIC
TRANSMISSION................ ........ .. ................. 56

Introduction .............................................. 56
Materials and Methods..................................... 57
Results ................................................... 57
Discussion ................................................ 66








5 CALCIUM AND DEPRESSION OF TRANSMITTER RELEASE............... 68

Introduction........................ ..... ................. 8 68
Materials and Methods.................................. ..... 70
Results.................................................... 75
Discussion......................................... .. 99
Notes........................... ........................... 101

6 CHANGES IN MINIATURE ENDPLATE POTENTIAL FREQUENCY
DURING REPETITIVE STIMULATION UNDER CONDITIONS OF
DEPRESSION OF EVOKED TRANSMITTER RELEASE................. 102

Introduction ............................................... 102
Materials and Methods ................................. ... 104
Results........................ ............................ 106
Discussion .. .............................................. 126

7 CONCLUSIONS................................................. 130

REFERENCES................. ................................ 134

BIOGRAPHICAL SKETCH .................. .... .................... 149














LIST OF TABLES

TABLE 9a9e

4-1. EFFECT OF MU-CONOTOXIN GIIIA ON EXTRACELLULARLY
RECORDED NERVE ACTION POTENTIALS........................ 61

4-2. EFFECT OF MU-CONOTOXIN GIIIA ON PARAMETERS OF
TRANSMITTER RELEASE..................................... 62

6-1. TIME CONSTANTS OF DECAY OF THE COMPONENTS OF
INCREASED TRANSMITTER RELEASE.......................... 121













LIST OF FIGURES


FIGURE page

2-1. Plot of V(t), the fractional change in EPP amplitude,
during repetitive stimulation under conditions of
high levels of release................................. 15

2-2. Tracings of EPPs recorded extracellularly during
conditioning trains of impulses under conditions
of high levels of release in the absence and
presence of curare.................................. 20

2-3. Effect of curare on stimulation-induced changes in EPP
amplitude under conditions of high levels of release.... 22

2-4. Concentration dependence of the effect of curare on
peak V(t) and D(t) during 200-impulse trains of
stimulation ................................. .......... 26

2-5. Concentration dependence of the effect of curare on peak
V(t) and D(t) during 100-impulse trains of
stimulation in the presence of 5 pM mu-conotoxin........ 29

2-6. Effect of curare on stimulation-induced changes in EPP
amplitude under conditions of low levels of release..... 32

3-1. Effect of glycerol treatment on stimulation-induced
changes in EPP amplitude under conditions of high
levels of release................................... 41

3-2. Effect of glycerol treatment on peak V(t), final V(t),
and D(t)........... ......................... ........ 43

3-3. Effect of time and exposure to hyperosmotic sucrose
solution on stimulation-induced changes in EPP
amplitude under conditions of high levels of release.... 45

3-4. Effect of glycerol treatment on stimulation-induced
changes in EPP amplitude under conditions of low
levels of release.................................. 48

3-5. Effect of glycerol treatment on stimulation-induced
changes in EPP amplitude and quantal content............ 50








FIGURE page9

4-1. Tracings of extracellularly recorded responses of a
muscle to nerve stimulation in the presence and
absence of mu-conotoxin GIIIA......................... 59

4-2. Effect of 5 pM mu-conotoxin GIIIA on stimulation-induced
changes in EPP amplitude under conditions of low and
high levels of release.... ......................... 65

5-1. Plot of fractional change in EPP amplitude (V(t))
during repetitive stimulation at 20 impulses/sec
under conditions of high levels of release.............. 73

5-2. Effect of [Ca2+]o on stimulation-induced changes in EPP
amplitude under conditions of high levels of release.... 77

5-3. Effect of [Ca2+]o on peak V(t), final V(t), D(t),
01 and D2........................................... 79

5-4. Effect of [Ca2+]o on stimulation-induced changes in EPP
amplitude under conditions of high levels of release
in the absence of curare.............................. 82

5-5. Effect of Cd2+ on stimulation-induced changes in EPP
amplitude under conditions of high levels of release.... 86

5-6. Effect of Zn2+ on stimulation-induced changes in EPP
amplitude under conditions of high levels of release.... 89

5-7. Concentration dependence of the effects of Cd2+, Co2+,
Zn2+ and Ni2+ on D(t) during 200-impulse trains of
stimulation................................... ............. 90

5-8. Effect of omega-conotoxin (cZ-CgTx) on stimulation-
induced changes in EPP amplitude under conditions
of high levels of release.............................. 93

5-9. Effect of ionomycin and cyclopiazonic acid (CPA)
on stimulation-induced changes in EPP amplitude
under conditions of high levels of release.............. 97

6-1. Effect of repetitive stimulation on EPP amplitude
and MEPP frequency under conditions of high levels
of release ................................... ..... 107

6-2. Effect of repetitive stimulation on EPP amplitude,
MEPP frequency and MEPP amplitude under conditions
of high levels of release............................... 109

6-3. Effect of repetitive stimulation on MEPP amplitude
distribution under conditions of high levels of
release................................... ........... 113








FIGURE


6-4. Analysis of the time course of the decay of increased
MEPP frequency following 400-impulse conditioning
trains under conditions of high levels of
release............................................ 117

6-5. Effect of Ca2+ concentration on stimulation-induced
increases in MEPP frequency ............................ 122

6-6. Comparison of the effects of repetitive stimulation on
EPP amplitude and MEPP frequency under conditions of
low and high levels of release.......................... 125













Abstract of Dissertation Presented to the Graduate School
of the University of Florida in Partial Fulfillment of the
Requirements for the Degree of Doctor of Philosophy

DEPRESSION OF NEUROTRANSMITTER RELEASE DURING
REPETITIVE STIMULATION AT THE FROG
NEUROMUSCULAR JUNCTION

By

Marfa A. Sosa

August 1993


Chair: Janet E. Zengel
Major Department: Neuroscience

At the frog neuromuscular junction, repetitive stimulation of the

nerve can lead to a progressive decrease in endplate potential (EPP)

amplitude. Because the magnitude of this depression appears to depend

on the amount of transmitter released by conditioning impulses,

depression has traditionally been attributed to depletion of the store

of transmitter available for release. Nevertheless, it is also

possible that this decrease in release is due to a reduction or

inactivation of some other essential component of the release

mechanism. The purpose of this study was to obtain a better

understanding of the effects of depression and of the role played by

calcium in the mechanisms underlying this phenomenon. Experiments were

carried out using the sartorius muscle and nerve of the leopard frog

(Rana pipiens).

It is shown that two methods commonly used to block muscle

contraction under conditions of high levels of release, curare and

ix







glycerol treatment, affect stimulation-induced changes in release,

whereas a third method that has become available recently, addition of

mu-conotoxin, does not affect release in the presence or absence of

repetitive stimulation.

Two phases of depression have been observed during trains of

stimulation: an initial rapidly developing decrease in EPP amplitude,

followed by a further decrease that progresses more slowly. It is

shown that these two phases are affected differently by agents that

affect Ca2+ levels inside the nerve terminal. Block of Ca2+ influx

through voltage-dependent calcium channels by Cd2+ can affect the

second phase selectively, while directly increasing Ca2+ inside the

terminal primarily affects the first phase. These results suggest that

the two phases of depression arise from different mechanisms and that

in addition to its effects on control levels of transmitter release,

Ca2+ can affect depression directly.

It is also shown that asynchronous release is not affected by

depression. This suggests that either depression of evoked release is

not due to depletion of available transmitter or that there are

differences in the mechanisms underlying evoked and asynchronous

release. In either case, the results indicate that the phenomenon of

depression must involve an element exclusive to the mechanism of evoked

transmitter release.













CHAPTER 1


INTRODUCTION


Chemical Synaptic Transmission

Nerve cells communicate with each other and with effector organs

through the process of chemical synaptic transmission. Chemical

synapses operate by releasing chemical neurotransmitters from the nerve

terminal into the synaptic cleft; the released transmitter then acts

upon the postsynaptic cell. Although the basic steps of chemical

neurotransmission are now well established, the specific mechanisms

underlying the process of transmitter release itself are poorly

understood. It is known that Ca2+ influx through voltage-dependent

calcium channels in response to depolarization of the nerve terminal

leads to the release of transmitter (Katz and Miledi, 1967a; Llinas and

Nicholson, 1975) and that this transmitter is released in quanta, or

packets of a discrete number of molecules (del Castillo and Katz,

1954b; Fatt and Katz, 1952). The anatomical correlates of these

quantal packets of transmitter are thought to be the synaptic vesicles

found inside the nerve terminal (e.g., del Castillo and Katz, 1955;

Palay, 1956; Atwood et al., 1972; Heuser and Reese, 1973; Birks, 1974;

Heuser et al., 1979; Dickenson-Nelson and Reese, 1983). Postsynaptic

potentials (PSPs) are produced when multiple quanta of transmitter are

released synchronously in response to nerve terminal depolarization in

what is commonly referred to as evoked or synchronous quantal release.








Individual quanta are also released at random in the absence of nerve

stimulation, giving rise to miniature PSPs (Fatt and Katz, 1952; Dudel

and Kuffler, 1961; Katz and Miledi, 1963; also see review by Martin,

1966). This form of quantal release is known as spontaneous or

asynchronous release.

Notwithstanding this general understanding of the process of

release, many fundamental questions remain unanswered. For instance,

it is not known how Ca2+ entering the nerve terminal triggers the

release of quanta, nor is it known what leads to the synchronization of

release of multiple quanta in response to nerve stimulation. It is

also not known what mechanisms control the number of quanta released in

response to stimulation.

Stimulation-Induced Changes in Svnaptic Efficacy


The coding and transfer of information in both the central and

peripheral nervous systems most often involves some pattern of

repetitive stimulation (trains or bursts of impulses). One of the

properties of chemical synapses is that the amplitude of the

postsynaptic response to nerve stimulation is not constant but is a

function of previous synaptic activity (e.g., Feng, 1941; Liley,

1956b). It is important to understand the changes in synaptic

transmission that occur as a result of repetitive stimulation, not only

because these processes may be involved in the mechanisms underlying

learning and memory (e.g., Magleby, 1987; Hawkins et al., 1993), but

also because understanding the phenomena of stimulation-induced changes

in synaptic transmission can help us better understand the mechanism of

transmitter release itself.








Stimulation-induced changes in synaptic efficacy occur in various

forms at synapses of both the central and peripheral nervous systems

(e.g., Martin and Pilar, 1964; Magleby, 1973; Andersen et al., 1977;

Hirst et al., 1981; Ito and Kano, 1982; Kandel et al., 1983). The time

courses of these various phenomena span a range from milliseconds to

days. In this dissertation, I will focus on short-term (milliseconds

to minutes) stimulation-induced changes in synaptic efficacy.

The nature of the short-term phenomena of repetitive stimulation

depends on the levels of transmitter released from the nerve terminal

in response to depolarization. When the levels of release are lower

than normal, repetitive stimulation often leads to a progressive

increase in the amplitude of the postsynaptic response resulting from

an increase in the amount of transmitter being released (e.g., del

Castillo and Katz, 1954c; Curtis and Eccles, 1960; Kuno, 1964). When

the levels of release are normal or high, repetitive stimulation can

lead to a decrease in the amplitude of the postsynaptic response that

results from a reduction in the amount of transmitter being released

(e.g., del Castillo and Katz, 1954c; Brooks and Thies, 1962). The

mechanisms underlying these stimulation-induced changes in release are

not well understood.


Stimulation-Induced Increases in Transmitter Release

Short-term stimulation-induced changes in release have been studied

extensively at the neuromuscular junction of the frog, where the

progressive increase in the amplitude of the postsynaptic response, the

endplate potential (EPP), that is observed under conditions of reduced

levels of release can be described by four different components: first








and second components of facilitation (Mallart and Martin, 1967),

augmentation (Magleby and Zengel, 1976a; Erulkar and Rahamimoff, 1978)

and potentiation (Hubbard, 1963; Rosenthal, 1969; Magleby and Zengel,

1975ab). These components are distinguished from one another by

differences in their time constants of decay and pharmacological

sensitivities (Magleby and Zengel, 1975ab, 1976a; Zengel and Magleby,

1980, 1982). Under conditions of reduced levels of release, repetitive

stimulation also leads to an increase in the frequency of miniature

EPPs (MEPPs) that can be described by four components with similar time

courses and pharmacological sensitivities as the four components of

increased EPP amplitude (Zengel and Magleby, 1981).

The mechanisms underlying these four components of increased

release are not known, but it has been suggested that an increased

entry of Ca2+ or an accumulation of Ca2+ inside the nerve terminal

during repetitive stimulation may be responsible for one or more of

these processes (e.g., Katz and Miledi, 1968; Rosenthal, 1969; Miledi

and Thies, 1971; Weinreich, 1971; Erulkar and Rahamimoff, 1978; Zengel,

Sosa and Poage, 1993).

Depression of Transmitter Release

Repetitive stimulation at the frog neuromuscular junction under

conditions of normal or higher levels of release typically leads first

to a brief increase in EPP amplitude, probably due to the components of

increased release mentioned above, followed by a progressive decrease

in EPP amplitude for the remaining duration of stimulation. This

stimulation-induced decrease in EPP amplitude, known as depression, is

the topic of this dissertation.








Because the magnitude of depression appears to be dependent on the

amount of transmitter released by conditioning impulses, depression has

traditionally been attributed to a depletion of the store of

transmitter available for release (Thies, 1965). Nevertheless, this is

not the only possibility since this observed relationship between

depression and the total amount of transmitter being released during

stimulation can be indirect. The amount of transmitter released

following each nerve impulse is itself dependent on other factors such

as the amount of Ca2+ entering the nerve terminal or already present in

the terminal.

Both rapidly and slowly developing phases of depression have been

described (Takeuchi, 1958; Brooks and Thies, 1962; Elmqvist and

Quastel, 1965b; Mallart and Martin, 1968; Betz, 1970). However, not

much attention has been given to the study of the mechanisms underlying

these two phases. It is not known, for instance, whether they have a

common mechanism or if they are due to different processes. It is

important to study depression of transmitter release because it is most

likely the predominant phenomenon of changes in release during periods

of repetitive synaptic activity under physiological conditions in vivo,

at both central and peripheral synapses, and because understanding

depression of transmitter release can help us better understand how the

process of release can be controlled or modulated. Also, from a

clinical point of view, depression may play a part in exacerbating the

symptoms of conditions such as myasthenia gravis, in which the

amplitudes of EPPs are smaller than normal (e.g., Fambrough et al.,

1973; Lindstrom and Lambert, 1978) and can thus quickly reach levels








below the threshold for the generation of muscle action potentials

during periods of repetitive stimulation (Rowland, 1985).

The major goal of my research project was to obtain a better

understanding of the mechanisms underlying depression of transmitter

release and of the role played by Ca2+ in these mechanisms.

Experiments were carried out using the sartorius muscle and nerve of

leopard frogs (Rana opiiens). Since depression is observed under

conditions in which activation of the nerve leads to the generation of

muscle action potentials, some means of preventing these action

potentials must be used to measure the underlying EPPs. Thus, the

first specific aim of the project was to determine whether the methods

available for preventing muscle action potentials affect the processes

of stimulation-induced changes in release. The second specific aim was

to assess the effects of changes in extracellular and intracellular

Ca2+ and of blocking influx of Ca2+ through voltage-dependent calcium

channels on depression of EPP amplitude during trains of repetitive

stimulation. The third specific aim was to examine the effects of

depression on the asynchronous form of quantal release.

I found that the two methods most commonly used to block muscle

contraction under conditions of normal or higher levels of release,

addition of curare (Chapter 2) and glycerol treatment (Chapter 3), both

affect stimulation-induced changes in release. A third method that has

become available recently, addition of mu-conotoxin (Chapter 4), was

found to be effective for blocking muscle action potentials without

affecting transmitter release, in the presence or absence of repetitive

stimulation.






7

I also found that Ca2+ influx through voltage-dependent calcium

channels and changes in the levels of Ca2+ inside the nerve terminal

may play a direct role in the mechanisms underlying the two observed

phases of depression (Chapter 5). The results also suggest that the

two phases of depression indeed arise from two distinct processes that

are mediated through different mechanisms.

Finally, I found that while asynchronous release is increased by

the phenomena of stimulation-induced increases in release, it is not

affected by depression (Chapter 6). This suggests that the mechanisms

underlying depression of release must involve some element that is

exclusive to the mechanism of evoked neurotransmission.













CHAPTER 2


EFFECT OF D-TUBOCURARINE ON STIMULATION-INDUCED
CHANGES IN SYNAPTIC TRANSMISSION

Introduction

Depression of transmitter release is normally observed during and

following repetitive stimulation under conditions of normal or higher

levels of release. Under these conditions, stimulation of the motor

nerve results in an EPP that is large enough to depolarize the muscle

fiber to threshold, thus leading to the generation of an action

potential and muscle contraction (Kuffler, 1942). This muscle action

potential and contraction must be blocked to be able to measure the

underlying EPP. This has frequently been accomplished through the use

of curare which blocks the postsynaptic acetylcholine receptors

(Jenkinson, 1960; Adams, 1975; Colquhoun et al., 1979) and thus, at

appropriate doses, reduces the amplitude of the EPP below the threshold

for the generation of an action potential at the muscle fiber (e.g.,

Eccles et al., 1941; Fatt and Katz, 1951). There have been, however,

numerous reports indicating that in addition to the well characterized

effects of this drug on the postsynaptic acetylcholine receptor, curare

can have presynaptic effects as well (e.g., Lilleheil and Naess, 1961;

Hubbard et al., 1969; Hubbard and Wilson, 1973; Magleby et al., 1981;

Gibb and Marshall, 1984). In particular, a number of studies have

shown that curare increases the progressive decline in EPP amplitude








that occurs during high frequency (>50 impulses/sec) stimulation at

peripheral synapses (e.g., Hubbard et al., 1969; Hubbard and Wilson,

1973; Blaber, 1973; Glavinovic, 1979; Magleby et al., 1981; Gibb and

Marshall, 1984; Chang et al., 1988).

Before using curare in experiments designed to study depression of

transmitter release, I decided to examine in more detail if and how

this drug affects release during trains of stimulation applied at a

lower frequency (<50 impulses/sec; see Note 1 at the end of this

chapter). Experiments were done to determine the effect of various

concentrations of curare on changes in EPP amplitude observed during

trains of stimuli applied at a frequency of 20 impulses/sec (see Note 1

at the end of this chapter) under conditions of low and high levels of

release at the frog neuromuscular junction. It was found that even at

this low frequency of stimulation curare increases depression of

transmitter release as well as the facilitatory processes that are

observed both in the presence and absence of depression. These effects

of curare were concentration-dependent and reached a plateau level at

concentrations above 4-5 pg/ml.

Material and Methods


Preparation and Solutions

The sartorius nerve-muscle preparation from small to medium size

(2.5-3.0 inches in length) leopard frogs (Rana pipiens) of either sex,

obtained from Charles Sullivan Inc. (Nashua, TN), were used in all

experiments. The frogs were housed in basins (9-12 frogs per basin) at

20-220C and were fed crickets, usually once every two weeks. Frogs

were decapitated after pithing and the sartorius muscle with the








attached nerve was dissected in a normal Ringer solution (see below).

The nerve to the sartorius muscle was cut at the point of its

separation from the sciatic nerve, approximately 10-15 mm from the

entry site of the sartorius nerve into the muscle. The muscle was

stretched slightly by pinning the connective tissue at its sides and

ends to wax in the bottom of a 50 mm diameter Pyrex petri dish which

served as the recording chamber.

The standard bathing (normal Ringer) solution had the composition

(mM): NaCl, 116; KC1, 2; CaC12, 1.8; HEPES, 2; glucose, 5. This

solution was modified by either lowering the concentration of Ca2+ to

0.5-0.8 mM and adding 5 mM Mg2+ (low Ca Ringer) in experiments

requiring low levels of release, or by raising the concentration of

Ca2+ to 3.6-7.2 mM (high Ca Ringer) in experiments requiring high

levels of release. In some experiments, 5-15 PM mu-conotoxin GIIIA, a

toxin that selectively blocks the muscle's voltage-dependent Na

channels (see Chapter 4), was also added to the high Ca Ringer to

prevent muscle contraction. The effects of 1-8 yg/ml curare were

examined by adding appropriate amounts from a concentrated stock

solution directly to the Ringer in the bathing solution. The

osmolarity of the solutions was maintained by making appropriate

changes in the concentration of NaCl. All solutions were adjusted to

pH 7.2-7.4 before use. Salts for the Ringer solutions and

d-tubocurarine were obtained from Sigma Chemical (St. Louis, MO);

mu-conotoxin GIIIA was obtained from BACHEM (Torrance, CA). Solution

changes were carried out with Pasteur pipettes and were repeated 3-4

times for each solution to ensure thorough removal of the previous

solution. All experiments were conducted at room temperature

(20-220C).









Data Collection

A fluid suction electrode (Dudel and Kuffler, 1961) was used to

stimulate the nerve. One end of a PE-60 polyethylene tube of 1.22 mm

outside diameter and 0.8 mm inside diameter (Becton, Dickinson & Co.,

Parsippancy, NJ) was drawn out after gentle heating to an inside

diameter of approximately 0.5 mm. The other end of the tube was

connected to a 10 cc syringe, which was used to draw the nerve and

bathing solution into the tapered end. Stimulus pulses of 0.01-0.10

msec duration were delivered using a Grass S48 stimulator through a

Grass SIU5 stimulus isolation unit (Grass Instrument Co., Quincy, MA)

to the nerve through the suction electrode and adjusted in amplitude to

be clearly suprathreshold.

For most experiments, extracellular recordings of EPPs (see Note 2

at end of this chapter) were obtained with surface electrodes from

endplate regions of the frog sartorius muscle. Although surface

recording does not give any information concerning absolute levels of

transmitter being released, it gives a valid measure of the average

intracellular activity and of relative changes in release.

Furthermore, it has the advantage over intracellular recording of

summing the response from several endplates, thus reducing the amount

of data required to obtain estimates of the average response (Mallart

and Martin, 1967; Magleby, 1973). The recording electrodes were

constructed from polyethylene tubing similar to that used for the

stimulating electrode. The recording end of the tubing was heated

slightly to form a smooth 1 mm flange around the opening of the tube

and a Ag-AgCl wire was inserted into the tube from the other end.








Potentials were measured between the chlorided silver wire in the tube

and a similar wire in the bath (recording chamber). The muscle

remained immersed in the bathing solution during recording. The

surface recording electrode was filled with bathing solution of

identical composition to that used in the experiment and positioned

just above the muscle surface at an endplate region to obtain the EPP

with maximum amplitude and minimum rise time. The responses were

amplified with a Grass P511 A.C. preamplifier and displayed on a

Tektronix model 5113 dual beam storage oscilloscope (Tektronix,

Beaverton, OR). Amplified responses were filtered at 10 KHz. An

estimate of the number of synapses under the recording electrode was

obtained by gradually increasing the strength of the stimulus and

counting the step increases in the amplitude of the EPP. On average,

data were recorded from 3-5 synapses simultaneously. In some

experiments, with careful placement of the extracellular electrode,

presynaptic action potentials from nerve could also be recorded.

For some experiments, EPPs and MEPPs were recorded intracellularly

using standard microelectrode techniques. Recordings were performed

using a World Precision Instruments Microprobe system (WPI, Inc., New

Haven, CT) and glass microelectrodes filled with 3 M KC1, with tip

resistances of 5-20 Mohms. The glass micropipettes for the electrodes

were pulled on a vertical pipette puller (model 720, David Kopf

Instruments, Tujunga, CA) using glass capillary tubing obtained from

World Precision Instruments (WPI #TW150F). Endplate localization was

determined by inserting the electrode along the muscle fiber to find

the position of maximum EPP amplitude and minimum rise time. Average

MEPP amplitudes varied between 0.3 and 1.0 mV. The noise level in the








recording system was low because of the low resistance electrodes. It

was usually possible to hold cells for 1-6 hours with resting

potentials more negative than -70 to -80 mV.

In all experiments, the nerve was stimulated with conditioning

trains of 100-200 impulses applied at a frequency of 20 impulses/sec

(see Note 1 at end of this chapter). Sufficient time was allowed

between trains to ensure that the release level had recovered to the

preconditioning level (15 min after trains of impulses in experiments

in which quantal content was normal or higher, and 5 min after trains

of impulses in experiments in which quantal content was low; Magleby

and Zengel, 1975a, 1976b). It was usually necessary to average the

response from several identical trials in order to obtain reliable

estimates of the stimulation-induced changes in EPP amplitude.

A MINC-11 computer (Digital Equipment Corp., Marlboro, MA) was used

to generate the stimulation patterns, measure and store EPP amplitudes

during the experiment, and analyze the data. Data were collected and

analyzed in terms of trials. In a typical high quantal content

experiment, the nerve was stimulated once every 30 sec for 2-3 min to

establish a control EPP amplitude. The nerve was then stimulated

100-200 times at a frequency of 20 impulses/sec to condition the nerve.

The effect of the conditioning stimulation was then followed by testing

with single impulses applied once every 2 sec for three impulses to

test for fast decaying changes in EPP amplitude and then once every

30-60 sec for 15 min to test for the more slowly decaying changes in

EPP amplitude. Data obtained from intracellular recording experiments

(and occasionally from extracellular recording experiments as well)

were also recorded on videotape using a PCM Recording Adaptor (A.R.








Vetter Co., Pacer Scientific Instruments, Los Angeles, CA) and a

standard VCR. Three separate channels were used to record the data,

one to save a trigger pulse, and the other two to save the signal at

both a high and low gain to be able to observe MEPPs and EPPs,

respectively. The basic sampling rate of each channel was 88.2 KHz and

the channel rise time was 50 psec; the AD/DA resolution was 14 bit.

The data saved on videotape were later transferred for analysis to a

386 Zenith computer equipped with 12 bit A/D acquisition hardware and

the AxoTape software (Axon Instruments, Inc., Foster City, CA). In

some instances, videotaped data were averaged for analysis using a

Nicolet 1170 signal average (Nicolet Instrument Co., Madison, WI).

Data Analysis

Estimates of the fractional change in EPP amplitude resulting from

repetitive stimulation were expressed as:

V(t) = EPPt/EPPc 1 (eqn. 2-1)


where EPPt is the EPP amplitude at time t, and EPPc is the control EPP

amplitude in the absence of repetitive stimulation. Positive values of

V(t) indicate an EPP amplitude larger than control while negative

values indicate an EPP amplitude smaller than control. When the EPP

amplitude is the same as control, the value of V(t) is zero.

Figure 2-1 presents typical data obtained during repetitive

stimulation under conditions of high levels of release (inset), and

illustrates how the various measures of changes in EPP amplitude were

obtained. Estimates of D(t), the magnitude of depression, were

obtained from the difference between peak V(t), usually reached during













1.0I
0.8
0.6
0.4 peak

0.2-1. t V(t) in(t)

a ho o--------- V--(t)




012345678910








obtained (see Materials and Methods). Inset: tracing of
extracellularly recorded EPPs during a train of 20 impulses delivered
at 20 impulses/sec in 1.8 mM Ca Ringer. Vertical scale: 100 fV;
horizontal scale: 100 msec.
horizontal scale: 100 msec.








the first second of the conditioning train, and final V(t), the value

of V(t) at the end of the train.

For intracellular recording experiments, quantal content was

estimated using at least two of the three methods available: the direct

method, the method of coefficient of variation and the method of

failures (del Castillo and Katz, 1954a; Martin, 1966). The direct

method is based on the assumption that the spontaneous MEPP is the

basic unit of transmitter action and that the EPP that results from a

nerve impulse is made up of an integral multiple of such unit

components (del Castillo and Katz, 1954b). Thus, quantal content can

be calculated directly using the equation

m= EPP/MEPP (eqn. 2-2)

where m is the quantal content and EPP and MEPP are the mean EPP and

MEPP amplitudes, respectively.

When the calcium concentration is lowered and magnesium is added to

the bathing solution, the amount of acetylcholine released by an

impulse can be reduced to a very low level. Under these experimental

conditions, EPP amplitude is not only greatly reduced but it also

fluctuates at random between multiples of the size of the individual

quanta, with occasional failures of response (Fatt and Katz, 1952).

The indirect methods of estimating quantal content are based on the

assumption that under such conditions of low levels of release, the

distribution of EPP amplitude is described by Poisson's law (del

Castillo and Katz, 1954b). Indirect estimates of quantal content (m)

can thus be obtained using the method of coefficient of variation and

the method of failures as shown in the equations below:








m = 1/(CV)2 (eqn. 2-3)

m = In(N/F) (eqn. 2-4)

where CV is the coefficient of variation, N is the total number of

stimulations and F is the number of times stimulation failed to evoke

an EPP (del Castillo and Katz, 1954b).

Averaged data are presented as mean + standard error. In most

experiments, differences between means were tested for statistical

significance using Student's t-tests. In experiments in which the

variability of the control and experimental groups were significantly

different, a nonparametric test, the Wilcoxon rank-sum test, was used

to test for significant differences between population distributions

(Marks, 1982).

EPP amplitude in experiments conducted under conditions of high

quantal content could be quite large, particularly in those experiments

in which very little or no curare was added. It has been demonstrated

that when EPP amplitudes exceed 10-15 mV, the relationship between the

amount of transmitter released and EPP amplitude is no longer linear

(Martin, 1955). As the membrane potential of the muscle fiber

approaches the reversal potential, the driving force for cations

flowing into the muscle fiber is reduced, resulting in a reduction of

the amplitude of the EPP. This non-linearity could result in an

underestimate of the magnitude of changes in V(t) observed during

conditioning trains when the concentrations of curare are low (when EPP

amplitudes are largest), compared to the changes observed when the

concentrations of curare are higher (smaller EPP amplitudes). To take

into account the possibility of underestimating the changes in V(t)

observed in the experiments presented here, estimates of what the








corrected values of EPP amplitude would be were made assuming the

extreme situation that in all experiments the amplitudes of EPPs at the

lower concentrations of curare were as large as possible (see below).

Note that only estimates (instead of exact calculations) of the

correction for non-linear summation could be made since in most of the

high quantal content experiments EPPs were recorded extracelullarly,

and thus it was not known what their actual amplitudes were. The

corrections were made using the equation proposed by McLachlan and

Martin (1981):

EPP' = EPP / (1-(0.55 x (EPP/Vm))) (eqn. 2-6)

where EPP' is the corrected EPP amplitude, EPP is the uncorrected EPP

amplitude and Vm is the resting membrane potential. To make these

estimates of corrected EPP amplitude, the following assumptions were

made: (1) the resting membrane potential was -85 mV; (2) in experiments

in which curare had been added in a sufficient concentration to block

muscle action potentials, the largest EPP amplitude observed during

conditioning trains in an experiment was assumed to be 44 mV, the

largest value possible before reaching threshold for the generation of

an action potential (see Table 11 in Fatt and Katz, 1951); the

amplitudes of all other EPPs during the train were adjusted or

normalized accordingly; (3) in experiments in which muscle action

potentials were blocked using mu-conotoxin, the amplitude of control or

unconditioned EPPs (EPPs before the trains of repetitive stimulation)

were considered to be 44 mV, the mean amplitude of EPPs recorded

intracellularly in 3.6 mM Ca Ringer solutions containing 5 pM

mu-conotoxin (data obtained from other experiments done in this








laboratory). Again, the amplitudes of all other EPPs during the train

were adjusted or normalized accordingly.

Results

Effect of Addition of Curare on Stimulation-Induced Changes in EPP
Amplitude under Conditions of High Levels of Release

Figure 2-2 shows tracings of EPPs recorded under conditions of high

levels of release during trains of stimulation applied at a frequency

of 20 impulses/sec prior to and following the addition of 3 pg/ml

curare. The top trace shows the typical response observed during

repetitive stimulation in the absence of curare. There is a small

increase in EPP amplitude at the beginning of the conditioning train

that is quickly followed by a decrease that progresses during the

remainder of the train. When curare is added (bottom trace), not only

is the amplitude of the control EPP (the first EPP of the train)

greatly reduced (notice different vertical scales), but the initial

increase in EPP amplitude observed at the beginning of the train as

well as the depression that follows are both greater than they were

before curare had been added.

These effects of curare can be seen more clearly in Figure 2-3A,

where the fractional change in EPP amplitude, V(t), is plotted as a

function of time during 200-impulse conditioning trains of stimulation

recorded from another experiment in normal (1.8 mM) Ca Ringer. The

lines plot V(t) during trains before (continuous line), during

(short-dashed line) and after (long-dashed line) exposure to 3 Yg/ml

curare. It can be clearly observed that in the presence of curare the

initial increase in V(t) normally seen at the beginning of the trains














0 curare


7


+ 3 pgg/ml curare


Figure 2-2. Tracings of EPPs recorded extracellularly during
conditioning trains of impulses under conditions of high levels of
release in the absence and presence of curare. In this and all other
figures in this dissertation, the frequency of stimulation during
conditioning trains was 20 impulses/sec. Data were collected in Ringer
solution containing 3.6 mM Ca2+ and 10 pM mu-conotoxin. Trains of EPPs
were recorded first in the absence of curare and 20 min after the
addition of 3 jg/ml curare. Horizontal bar: 400 msec. Vertical bar:
25 pV in the absence of curare, 9.7 yV after addition of 3 pg/ml
curare.


1 I1h


























Figure 2-3. Effect of curare on stimulation-induced changes in EPP
amplitude under conditions of high levels of release. (A): Data
collected from a single preparation in Ringer solution containing 1.8
mM Ca2+ and 5 pM mu-conotoxin. The lines plot changes in V(t) as a
function of time during 200-impulse trains of stimulation at 20
impulses/sec before (continuous line), during (short-dashed line) and
after (long-dashed line) exposure to 3 pg/ml curare. Each line
represents the average of from 2 to 14 consecutive trials. (B): Effect
of curare on mean peak V(t), final V(t) and D(t) (see Materials and
Methods). Data obtained from 3 experiments in normal (1.8 mM) Ca
Ringer solution; nerves conditioned with 200-impulse trains of
stimulation. Standard errors indicated by vertical lines. Asterisks
indicate statistically significant differences between mean values
obtained in the absence and presence of curare (p<0.03, paired t-test).









A 0.6
S-- 0 curare (control)
0.4 + ------ + 3 pg/ml curare
---- 0 curare (recovery)
0.2

0.0

-0.2
O .4 .- """,-....,
2+ ".'*.- .
1.8 mM Ca 2
-0.6 1
0 1 2 3 4 5 6 7 8 9 10

Time during train (sec)


B
1.0 1.0
0.8 0.8
0.6 0.6
0.4 0.4
final
0.2 nalv() 0.2
0.0 0.0
-0.2 pek D(t) -0.2
-0. V(t) <0.03 -0
-0.4 -0.4
p<0.03


E0cr + 3-5 pg/ml curare


I 0 curare








of stimulation is increased further and that the decrease in V(t) that

follows is much greater. These effects of curare are readily reversed

upon removal of the drug (long-dashed line). Similar effects of added

curare (2-5,mg/ml) were observed during 200-impulse conditioning trains

in 5 additional experiments done in normal (1.8 mM) or high (3.6 mM) Ca

Ringer.

Figure 2-3B summarizes results of 3 experiments done in normal (1.8

mM) Ca Ringer. In these experiments estimates of peak V(t), final V(t)

and D(t) were obtained as illustrated in Figure 2-1 in the Materials

and Methods section of this chapter. Adding curare (3-5 pg/ml)

resulted in a significant increase in peak V(t) and D(t) (p<0.03;

paired t-test). Even though the value of V(t) at the end of the trains

(final V(t)) was more negative in the presence of curare in all 3

experiments, the average change was not significant presumably due to

the large variability in the data obtained from the 3 preparations.

Similar effects of curare were also observed in experiments done in

Ringer solution containing 3.6 mM Ca2+ (data not shown).

The increases in peak V(t) and D(t) following addition of curare

were still statistically significant when these values were calculated

using estimates of EPP amplitudes corrected for non-linear summation

(see Materials and Methods for details of this correction): peak V(t),

0.27 + 0.03 vs. 0.69 + 0.13, and D(t), 0.35 + 0.15 vs. 0.89 0.16, in

the absence and presence of curare, respectively (p<0.05, paired

t-test).








Concentration Dependence of the Effect of Curare on Peak V(t) and D(t)
under Conditions of High Levels of Release

In the experiments described in the previous section it was

possible to compare changes in V(t) during conditioning trains in the

absence and presence of curare by blocking muscle action potentials

with mu-conotoxin. Prior to availability of mu-conotoxin, since it was

not possible to record EPPs in the absence of curare, I compared the

effects on V(t) of changing the concentration of curare. Results from

experiments of this kind are summarized in Figure 2-4. This figure
shows plots of mean values of peak V(t) (A) and D(t) (B) as a function

of the concentration of curare. The greatest effects of curare on

these parameters of stimulation-induced changes in release occurred at

the lower end of the range of concentrations tested, particularly from

0 to 2 yg/ml. At the higher end of the range of concentrations tested

(4-8 pg/ml), the magnitudes of peak V(t) and D(t) appeared to plateau.

(The effects of higher concentrations of curare could not be examined

because at concentrations above 8 yg/ml EPPs were completely blocked).

Similar results were observed in experiments done under conditions of

normal levels of release (1.8 mM Ca2+; data not shown).

The data included in Figure 2-4 were also analyzed following

correction of the EPP amplitudes for non-linear summation (see

Materials and Methods). The only difference in the concentration

dependence of the effect of curare that could be observed when the
corrected data were used to calculate peak V(t) and D(t) was that the

plateau level of D(t) appeared to be reached at concentrations of
curare slightly lower (3-4 yg/ml) than those at which plateau levels of

peak V(t) and D(t) of the uncorrected data were reached (data not

shown).




























Figure 2-4. Concentration dependence of the effect of curare on peak
V(t) (A) and D(t) (B) during 200-impulse trains of stimulation. Data
obtained in 3.6 mM Ca2+ Ringer solution. Symbols plot mean values at
each concentration of curare. Standard errors indicated by vertical
lines. In the 3 experiments in which data were obtained in 0 and 2
pjg/ml curare, 5 ,M mu-conotoxin was present throughout the experiment.
The number of preparations from which data were obtained are indicated
in parentheses.










A
0.5
(3) (13) (7) (12)
(3)
0.4 -

>0.3 -

S 0.2 -

0.1 3

0.0 -" I I I
0 1 2 3 4 5 6 7 8

curare (pg/ml)

B 1.2 ( 7) ( 12) (3)
(13) -
1.0 (3)
0.8 -
0.6 -
0.4
(3)
0.2 -
0.0 -I '
0 1 2 3 4 5 6 7 8
curare (Ag/ml)








The Concentration-Dependence of the Effect of Curare on Peak V(t) and
D(t) Is Not Affected by mu-Conotoxin

In the 3 experiments included in Figure 2-4 in which data were
obtained in 0 and 2 pg/ml curare it was necessary to add 5 pM

mu-conotoxin to the Ringer solution to block muscle action potentials.

To ensure that the shape of the concentration dependence curve (Figure

2-4) is not related to the presence of mu-conotoxin in these few

experiments, the concentration dependence of the effects of curare on

peak V(t) and D(t) was also determined in experiments in which this

toxin was present at all concentrations of curare being tested.

Results from these experiments are summarized in Figure 2-5, which

shows plots of mean peak V(t) (A) and D(t) (B) as a function of curare

concentration. In these experiments, the nerves were conditioned with

100-impulse trains of stimulation. As in the data obtained in the

absence of toxin (see Figure 2-4), peak V(t) and D(t) were increased as

the concentration of curare was increased, again reaching plateau

levels at around 4-5 pg/ml curare.

Effect of Curare on Stimulation-Induced Changes in EPP Amplitude under
Conditions of Low Levels of Release

The effect of curare on the initial increase in V(t) observed at

the beginning of trains of stimulation (peak V(t)) may be an indication
that the phenomena of stimulation-induced increases in release may also

be affected by curare. Since these facilitatory processes are studied
more easily under conditions of low levels of release in which

depression is absent, experiments were done to look at the effect of

curare on changes in V(t) during trains of repetitive stimulation under

conditions of reduced levels of release (0.8 mM Ca2+, 5 mM Mg2+).





























Figure 2-5. Concentration dependence of the effect of curare on peak
V(t) (A) and D(t) (B) during 100-impulse trains of stimulation in the
presence of 5 pM mu-conotoxin. Data obtained in 3.6 mM Ca2+ Ringer
solution. Standard errors indicated by vertical lines. The numbers of
preparations from which data were obtained are indicated in
parentheses.











1.50
1.25
t 1.00
0.75
o (2)
a 0.50 -
a-
0.25 -4)
0.00 -
0 1


B
0.6
0.5
0.4 (2)
0.4
K 0.3
0.2 -(
0.1
0.0


(2)
*1'


I I I I I I
2 5 4 5 6 7 8
curare (Ag/ml)


() 1)
EU U


0 1 2 3 4 5 6 7 8
curare (ug/ml)








Figure 2-6A shows results from one such experiment. The lines plot

changes in V(t) during 200-impulse conditioning trains before
(continuous line), during (short-dashed line) and after (long-dashed

line) exposure to 1 pg/ml curare. It can be clearly observed that
following the addition of curare, V(t) during the conditioning train

was increased; this effect was readily reversed upon the removal of

curare. Similar increases in V(t) following the addition of curare

were observed in 10 additional experiments. Figure 2-6B plots data

averaged from 5 preparations in which the effect of 1 pg/ml curare was
examined during 200-impulse trains. The observed increase in V(t)

during the trains of stimulation following the addition of curare was

statistically significant (p<0.04, paired t-test). (The concentration

dependence of the effect of curare on V(t) under conditions of low

quantal content was not examined because the highest concentration of

curare that could be added while still being able to measure EPPs

accurately was only 2 pg/ml [one experiment; data not shown]).

To determine if the effect of curare on V(t) during trains of

stimulation results from an effect on quantal release, in 4 experiments
EPPs were recorded intracellularly (see Materials and Methods) under

conditions of reduced levels of release (0.6 mM Ca2+, 5 mM Mg2+). For

these experiments, it was necessary to use very low (0.375-0.500 ig/ml)

concentrations of curare so that MEPPs could still be detected and

measured following addition of the drug. Adding curare reduced the

mean amplitudes of both MEPPs and control (unconditioned) EPPs to the
same extent (an average 41% reduction in the amplitudes of both types
of responses), suggesting that these effects are mediated through the

well-characterized postsynaptic blocking action of curare. In these



























Figure 2-6. Effect of curare on stimulation-induced changes in EPP
amplitude under conditions of low levels of release. Data collected in
Ringer solution containing 0.6 mm Ca2+, 5 mM Mg2+. (A): Data collected
from a single preparation. The lines plot changes in V(t) as a
function of time during 200-impulse trains of stimulation before
(continuous line), during (short-dashed line) and after (long-dashed
line) exposure to 1 pg/ml curare. Each line represents the average of
from 14 to 20 consecutive trials. (B): Averages of data collected in
the absence (continuous line) and presence (short-dashed line) of 1
pg/ml curare, obtained from 5 different preparations. Standard errors
indicated by vertical lines. Asterisks indicate statistically
significant differences between mean values obtained in the absence and
presence of curare (p<0.04, paired t-test).
























0 1 2 3 4 5 6 7 8 9 10

- 0 curare ------- +1 pg/ml curare 0 curare
(control) (recovery)


*


* *


0 0.6 mM Ca

0 1 2 3 4 5 6 7 8 9 10

Time during train (sec)








experiments, addition of curare did not significantly change mean

control quantal content (0.97 + 0.33 vs. 0.96 + 0.36, before and after

addition of curare, respectively) or mean control MEPP frequency

(0.42 0.13 vs. 0.37 + 0.13 sec-1, before and after addition of

curare, respectively). However, V(t) of quantal content at the end of

the 200-impulse trains was increased from 4.26 + 0.78 to 5.59 + 0.85

following the addition of curare. Although this increase in V(t) was

not statistically significant, probably because of the large

variability between preparations, the effect was seen in all 4

experiments. These results indicate that the effect of curare on V(t)

under conditions of low levels of release is at least in part

presynaptic.


Discussion

The results presented here show that curare increases depression of

EPP amplitude during repetitive stimulation, even at a frequency of 20

impulses/sec, indicating that this effect of curare is not limited to

the conditions of high frequency stimulation used by previous

investigators (e.g., Blaber, 1973; Hubbard and Wilson, 1973; Magleby et

al., 1981; Gibb and Marshall, 1984). Curare also increased the initial

facilitation observed at the beginning of trains of stimulation under

conditions of normal or higher levels of release as well as the

stimulation-induced increases in EPP amplitude observed under

conditions of low levels of release (in which depression is absent).

These results are consistent with previous reports indicating that

curare increases paired pulse facilitation at the frog neuromuscular

junction (Matzner et al., 1988). The effect of curare on V(t) is due,








at least in part, to an effect on the amount of transmitter being

released from the nerve terminal.

Curare at low concentrations had no apparent effect on control

quantal content, in contrast to the results of Matzner and coworkers

(1988) who previously reported that a similarly low concentration of

curare (0.367 pg/ml) decreased quantal content. This discrepancy could

be due to differences in species of frogs used or perhaps to the

presence of a higher concentration of Mg2+ in the Ringer solutions used

in the experiments described here; it had been observed previously that

increasing the concentration of Mg2+ in the millimolar range can reduce

the affinity constant of curare (Jenkinson, 1960). While the

possibility that control quantal content was indeed affected when

higher concentrations of curare were used cannot be ruled out, the

results presented here indicate that the effect of curare on V(t)

during conditioning trains is not dependent on an effect on control

quantal content.

The mechanism by which curare affects stimulation-induced changes

in transmitter release is not clear. The concentrations of curare

shown here to have an effect on depression are in a range similar to

the affinity constant for binding of curare to the postsynaptic

receptors at the frog neuromuscular junction (1.7 pg/ml; Jenkinson,

1960). The effect of curare on depression appeared to reach a maximal

or plateau level at concentrations of between 4-5 pg/ml. These

observations suggest that the effect of curare on stimulation-induced

changes in release may involve binding to a specific site, such as a

presynaptic nicotinic receptor. In support of this, previous

investigators have reported that in addition to its well-known








postsynaptic actions, curare can also affect transmitter release

through a presynaptic action (e.g., Magleby et al., 1981; Matzner et

al., 1988), presumably by acting on a receptor that is involved in the

mediation of some form of feedback mechanism. There is some

controversy, however, concerning the question of whether such feedback

is positive (Bowman, 1980) or negative (Wilson, 1982; Wilson and

Thomsen, 1991). The results presented here indicate that if a feedback

mechanism exists at the frog motor nerve terminal, it must involve an

aspect of the mechanism of release that can result in increases in both

facilitation and depression of transmitter release. Since it is known

that both of these processes are affected by calcium (e.g., Lundberg

and Quilisch, 1953b; Katz and Miledi, 1968; Rosenthal, 1969; Zengel and

Magleby, 1977, 1980, 1981; also see Chapter 5), one possibility would

be that the binding of acetylcholine to presynaptic nicotinic receptors

leads to a reduction either in Ca2+ influx or in levels of

intraterminal free Ca2+. When this binding is blocked by curare, Ca2+

levels inside the nerve terminal would be increased. This increase in

intraterminal Ca2+ could, in turn, result in a greater depression of

transmitter release during repetitive stimulation under conditions of

normal levels of release, or in an increase in the facilitatory

processes observed under conditions of low levels of release.

Addition of curare remains one of the most practical methods of

blocking muscle action potentials at the neuromuscular junction,

primarily because of its low cost, ease of use and reversibility.

Since the mechanisms underlying the effects of this drug on transmitter

release during repetitive stimulation are not yet completely

understood, in experiments in which curare is used to block muscle








contraction it is important to consider possible interactions of the

effects of curare with other agents being tested. Certainly, if curare

is used, care should be taken to maintain its concentration constant

throughout the study, since varying concentrations of curare would add

another source of variability.

Notes

1 The reason for choosing a frequency of 20 impulses/sec for these
experiments is two-fold. First, while the rate of firing of motor
nerve fibers during brief reflexes may approximate frequencies as high
as 50 impulses/sec, the rate of discharge of motor units during
sustained contractions is believed to be lower, in the range of 5-25
impulses/sec (e.g., Adrian and Bronk, 1929; Fischbach and Robbins,
1969; Grimby and Hannerz, 1977). Thus, studying the phenomena of
changes in transmitter release observed during repetitive stimulation
at low frequencies may be a better approximation of the normal patterns
of activity of motoneurons. Second, while the various components of
the stimulation-induced increases in release can be studied in the
absence of depression, the opposite is probably not true. These
components of increased release are also thought to be present under
the conditions of normal or elevated quantal content in which
depression of transmitter release is observed (e.g., Lundberg and
Quilish, 1953a; Curtis and Eccles, 1960; Thies, 1965; Pallotta and
Magleby, 1979; Magleby and Zengel, 1982). Thus, it is important to
consider how these different processes act together to modulate
transmitter release and how they may interact with one another. Since
many of the previous studies of the phenomena of stimulation-induced
increases in release at the frog neuromuscular junction have been done
using conditioning trains applied at a frequency of 20 impulses/sec
(e.g., Magleby and Zengel, 1976c, 1982; Zengel and Magleby, 1980, 1981,
1982; Zengel, Lee, Sosa and Mosier, 1993; Zengel, Sosa and Poage,
1993), I decided to use this same frequency of stimulation in
experiments designed to study depression of transmitter release. In
this manner, I will be able to more easily compare the results of my
experiments with those of these previous studies.
2 The potentials being measured represent the potential difference
between the extracellular recording electrode and the reference
(ground) electrode and not the actual potential that develops across
the membrane at the endplate. The change in potential recorded between
the two electrodes closely reflects the time course of the current
flowing across the cell membrane. Nevertheless, this potential
difference has by convention been referred to as an endplate
potential (EPP; e.g., Eccles et al., 1941; Feng, 1941; Liley and North,
1953; Lundberg and Quilisch, 1953; Takeuchi, 1958; Katz, 1962; Thies,
1965; Mallart and Martin, 1967; Betz, 1970; Magleby, 1973; Magleby and
Zengel, 1975; Zengel and Magleby, 1982).













CHAPTER 3


EFFECT OF GLYCEROL TREATMENT ON STIMULATION-INDUCED
CHANGES IN SYNAPTIC TRANSMISSION


Introduction

Another method commonly used to prevent muscle contraction for

studies of neuromuscular transmission consists of immersing the muscle

for 1 hour in a solution that has been made hypertonic with glycerol

(400-1000 mM), followed by a solution of normal osmolarity (Howell and

Jenden, 1967; Howell, 1969). The drastic changes in tonicity result in

the disruption of the t-tubule system of the muscle, thus disabling the

pathway for excitation-contraction coupling (Howell and Jenden, 1967;

Howell, 1969). It is thought that the glycerol treatment affects the

t-tubules exclusively because the membrane of the t-tubules is more

permeable or accessible to glycerol than the membrane of other

structures and/or organelles in the muscle (Howell, 1969). Even though

muscle action potentials can still be generated following glycerol

treatment (Fujino et al., 1961; Gage and Eisenberg, 1967), it is

possible, for some reason not yet understood, to record EPPs in some

areas of the muscle without interference from action potentials

(Miyamoto, 1975).

It has been assumed that glycerol treatment does not affect the

nerve terminal (e.g., Furshpan, 1956; Gage and Eisenberg, 1967; Lermer

and Rahamimoff, 1970; Miyamoto, 1975). However, no studies have








been done in which release is compared before and after exposure to

glycerol treatment. In this chapter, evidence is presented indicating

that exposure of the frog neuromuscular junction to glycerol treatment

to prevent muscle contraction has an effect on evoked release of

transmitter as well as on the phenomena of stimulation-induced changes

in release.


Material and Methods

Details of the preparation, solutions, and the recording, data

collection and analysis techniques were described in Chapter 2.

Glycerol Treatment

Isolated sartorius muscles were equilibrated in a control Ringer

solution for at least 60 min. This solution was then removed and

replaced by a modified Ringer solution of similar composition but also

containing glycerol (60-1000 mM). The muscle remained immersed in this

solution for 60 min, after which the glycerol solution was removed and

replaced with the control Ringer solution (Howell and Jenden, 1967).

This procedure will be referred to as "glycerol treatment". In several

experiments, the preparation was exposed to a Ringer solution made

hyperosmotic with sucrose instead of glycerol; the concentration of

sucrose was adjusted so that the osmolarity of the solution was similar

to that glycerol solution to which it was being compared. Glycerol and

sucrose were obtained from Sigma Chemical (St. Louis, MO).








Results

Effect of Glycerol Treatment on Stimulation-Induced Changes in EPP
Amplitude under Conditions of High Levels of Release

Figure 3-1 shows the effect of glycerol treatment on changes in EPP

amplitude during repetitive stimulation under conditions of high levels

of release. The continuous line in Figure 3-1A plots V(t), the

fractional change in EPP amplitude (Equation 2-1 in Materials and

Methods, Chapter 2), during a 200-impulse conditioning train in normal

Ca (1.8 mM) Ringer. As is typical under these experimental conditions,

there was an initial increase in EPP amplitude at the start of the

conditioning train, followed by a depression of EPP amplitude that

progressed during the remainder of the train. A very similar response

was obtained during 1 hour exposure to a Ringer solution containing 500

mM glycerol (long-dashed line). However, following removal of glycerol

and return of the preparation to the control normal Ringer solution

(short-dashed lines), an increase in V(t) that progressed with time

following removal of glycerol was observed.

Figure 3-1B presents data from another experiment in which a

similar effect of glycerol was observed following exposure to 100 mM

glycerol. Notice again that there was little change in V(t) during

exposure to glycerol, but following removal of glycerol the magnitude

of V(t) was increased throughout the 200-impulse train. In this

experiment the magnitude of this effect continued to increase for more

than 8 hours following removal of glycerol.

Results like those illustrated in Figure 3-1 were obtained in a

total of 16 experiments in which the effects of from 60 mM to 1 M

glycerol on V(t) during repetitive stimulation under conditions of high

























Figure 3-1. Effect of glycerol treatment on stimulation-induced
changes in EPP amplitude under conditions of high levels of release.
The lines plot V(t), the fractional change in the amplitude of
extracellularly recorded EPPs (Equation 2-1 in Materials and Methods,
Chapter 2), as a function of time during 200-impulse trains of
stimulation applied at a frequency of 20 impulses/sec. Data were first
obtained in control Ringer solution (continuous lines), then in the
presence of glycerol (long-dashed lines), and finally, at various times
following removal of glycerol and return to the control Ringer solution
(short-dashed lines). The times in this and following figures indicate
the midpoint of the time following glycerol removal during which data
were averaged. Each line represents the average of from 2 to 10
consecutive trains applied at 15 min intervals. A: Data collected in
Ringer solution containing 1.8 mM Ca2+ and 3.5 pg/ml curare. B: Data
(from another experimental preparation) collected in Ringer solution
containing 3.6 mM Ca2+ and 5 pg/ml curare.









A 1.0

0.8
0.6
0.4
4. 0.2
> 0.0
-0.2
-0.4
-0.6
-0.8




B1.0

0.8
0.6
0.4
2 0.2
> 0.0
-0.2
-0.4
-0.6
-0.8


-- control
S---- +500 mM glycerol
S---- .--- post-glycerol


.... ........................ ........... 150 m in

.~...-s._.. '. -- 90 min


2+
1.8 mM Co

0 1 2 3 4 5 6 7 8 9 10

Time during train (sec)


control
---- +100 mM glycerol
------- post-glycerol

-* *

... 50

... :: -- -... -- --..... .. .
.. 44

91
2+
3.6 mM Ca
1 1 1 1 1 1 1 1 1 1 1
0 1 2 3 4 5 6 7 8 9 10

Time during train (sec)


3 min

9 min
8 min
9 min








levels of release were examined. The results of these experiments are

summarized in Figure 3-2, which presents mean values of peak V(t),

final V(t) and D(t) (see Materials and Methods, Chapter 2) before and

following glycerol treatment.

Effect of Time on Stimulation-Induced Changes in EPP Amplitude

To eliminate the possibility that the apparent effects of glycerol

might arise simply from time alone, a number of experiments similar to

those described above but in which the preparation was allowed to

remain in the control Ringer solution throughout the experiment were

carried out. Figure 3-3A presents results from one such experiment in

which data were collected for 9 hours. It can be seen that there was

no change in V(t) from the first to the last hour of conditioning

stimulation. Similar results were obtained in 12 additional

experiments in which changes in V(t) were monitored for up to 11 hours.

The results of these experiments, summarized in Figure 3-3B, indicate

that the apparent glycerol-induced changes in V(t) do not arise as a

result of time alone.


Effect of Exposure to Hyperosmotic Sucrose Solution


To determine whether the effects of glycerol on stimulation-induced

changes in EPP amplitude were due solely to changes in osmolarity or

whether they were specific to glycerol, experiments were done to look

at the effect of exposing the preparation for 1 hour to a solution made

hyperosmotic with sucrose instead of glycerol. Figure 3-3C shows

results from one such experiment. The continuous line plots changes in

V(t) during 200-impulse conditioning stimulation in 3.6 mM Ca Ringer
















1.0 1.0
0.8 0.8
0.6 0.6
0.4 final 0.4

0.2 V(t) 0.2
0.0 0.0
peak D(t) -0.2
-0.2 v(t) ** .2
-0.4 -0.4
-0.6 -0.6
I I control m post-glycerol
p<.001 ** p<.01


Figure 3-2. Effect of glycerol treatment on peak V(t), final V(t), and
D(t). Data averaged from 16 experimental preparations exposed to
Ringer solution containing 60-1000 mM glycerol. Standard errors
indicated by vertical lines. Asterisks indicate statistically
significant differences between control and post-glycerol data.























Figure 3-3. Effect of time and exposure to hyperosmotic sucrose
solution on stimulation-induced changes in EPP amplitude under
conditions of high levels of release. (A): Changes in V(t) during
200-impulse trains obtained in control (3.6 mM Ca2+) Ringer solution
during the first (continuous line) and last hour (dashed line) of a 9
hour experiment. 5 jg/ml curare present throughout experiment. (B):
Peak V(t), final V(t) and D(t) during the first and last hour of
experiments lasting from 3 to 11 hours. Data averaged from the
experiment presented in (A) and 12 additional experiments carried out
in control Ringer containing 1.8-7.2 mM Ca2+ and 5 pg/ml curare.
Standard errors indicated by vertical lines. (C): Changes in V(t)
during 200-impulse trains obtained in control (3.6 mM Ca2+) Ringer
solution prior to (continuous line; average of 2 trials) and following
(dashed line; average of 12 trials) a 1 hour exposure to a Ringer
solution made hyperosmotic with 448 mM sucrose (similar osmolarity as
that of a 500 mM glycerol solution). 3 pg/ml curare present throughout
experiment. (D): Peak V(t), final V(t) and D(t) before and following
sucrose treatment. Data averaged from the experiment presented in (C)
and 4 additional experiments. Standard errors indicated by vertical
lines.
















1.2
1.0
0.8
0.6
0.4
S0.2
S0.0
-0.2
-0.4
-0.6
-0.8
-1.0


1.2
1.0
0.8
0.6
0.4
0.2
0.0
-0.2
-0.4
-0.6
-0.8
-1.0


1.2
1.0
0.8
0.6
0.4
0.2
0.0
-0.2
-0.4
-0.6
-0.8
-1.0


I-- start M and


start (0-1 hr)
....... end (8-9 hrs)









0 1 2 3 45 6 7 8 910

Time during train (sec)




C

control
....... post-sucrose
(180 min)








0 1 2 3 4 5 6 7 8 910

Time during train (sec)


U 1.0
0.8
0.6
0.4
final 0.2
0.2
V(t)
peak D(t) -0.2
V(t) -0.4
-0.4
-0.6
-0.8
J control M post-sucrose


1.0
0.8
0.6
0.4
0.2
0.0
-0.2
-0.4
-0.6
-0.8


1.0
0.8
0.6
0.4
0.2
> 0.0
-0.2
-0.4
-0.6
-0.8








prior to the addition of sucrose. The dashed line plots changes in

V(t) following return of the preparation to the control Ringer

solution. In this and 4 additional experiments no change in V(t)

following exposure to solutions made hyperosmotic with sucrose were

observed. The results of these experiments are summarized in Figure

3-3D. It can be seen that, in contrast to the marked effects of

glycerol treatment on peak and final V(t) (Figure 3-2), sucrose

treatment had no effect on these parameters of stimulation-induced

changes in EPP amplitude.

To verify that the sucrose-exposed preparations were capable of

responding to glycerol treatment, and to further test the specificity

of glycerol in generating the effects on V(t), in 4 experiments the

preparations were exposed to glycerol 1 to 2.5 hours following removal

of sucrose. After 1 hour in the glycerol Ringer solution, the

preparation was then placed in a sucrose Ringer solution of the same

osmolarity as the glycerol solution. In all four experiments, effects

on V(t) similar to those illustrated in Figure 3-1 were observed; in

each case V(t) was increased following the glycerol treatment. Again,

these effects of glycerol treatment progressed with time following

removal of glycerol. Thus, it appears that the effect of glycerol

removal on stimulation-induced changes in V(t) cannot be

attributed simply to a change in osmolarity.


The Effect of Glycerol Treatment on V(t) Appears to Result from an
Increase in the Magnitude of Underlying Stimulation-Induced
Increases in EPP Amplitude

The above described effects of glycerol treatment on V(t) could

result from either a decrease in the level of depression or from an








increase in the stimulation-induced increases in release that are

thought to be present under conditions of normal levels of release, but

masked by depression (Thies, 1965; Pallotta and Magleby, 1979; Magleby

and Zengel, 1982). As shown in Figure 3-2, although there were

statistically significant effects of glycerol treatment on both peak

V(t) and final V(t), there was no apparent effect of glycerol treatment

on D(t), which represents the net decrease in EPP amplitude from the

initial peak value to the value at the end of the conditioning train

(see Figure 2-1). These results suggest that the increase in V(t)

observed after removal of glycerol may be due to an increase in the

magnitudes of one or more of the components of stimulation-induced

increases in EPP amplitude, rather than a decrease in depression.

To investigate this possibility, experiments were carried out under

conditions of low levels of release (0.6-0.8 mM Ca2+, 5 mM Mg2+), where

the components of increased release can be observed without

interference from depression. Figure 3-4 shows results from one such

experiment. The continuous line plots changes in V(t) during a

200-impulse conditioning train prior to the addition of glycerol-

containing low Ca Ringer. The long-dashed line plots changes in V(t)

during exposure to 500 mM glycerol. As in the experiments done under

high levels of release, there was little change in V(t) during exposure

to glycerol. However, following removal of glycerol there was an

increase in V(t) that became progressively greater during the

conditioning train. This effect of glycerol treatment on V(t) also

increased with time following glycerol removal, in this experiment for

the 3 hours in which data continued to be collected.

















-- control
---- +500 mM glycerol
........ post-glycerol


.170 min


30 min


C0 m 2+
0.6 mM Ca


0 1 2 3 4 5 6 7 8 9 10

Time during train (sec)



Figure 3-4. Effect of glycerol treatment on stimulation-induced
changes in EPP amplitude under conditions of low levels of release.
The lines plot changes in V(t) during 200-impulse trains in a low (0.6
mM) Ca Ringer prior to (continuous line), during (long-dashed line),
and following (short-dashed lines) a 1 hour exposure to 500 mM
glycerol. Each line represents the average of from 6 to 14 consecutive
trials.


--








Figure 3-5A summarizes the results of 3 experiments in which the

effects of 500 mM glycerol on stimulation-induced changes in EPP

amplitude during 200-impulse trains under conditions of low levels of

release were examined. In these experiments final V(t), the magnitude

of V(t) at the end of the 200-impulse conditioning trains, was

increased from 3.60 + 0.14 prior to glycerol treatment to 5.07 + 0.02

following removal of glycerol (p<0.01, paired t-test). Thus it appears

that glycerol treatment results in an increase in the stimulation-

induced increase in EPP amplitude observed under conditions of low

levels of release. This effect can account for the increase in V(t)

observed under conditions of high levels of release.

Effect of Glycerol Treatment on Stimulation-Induced Changes in Quantal
Release

To verify that the observed effects of glycerol treatment on

stimulation-induced changes in EPP amplitude resulted from presynaptic

changes in transmitter release, intracellular recording techniques were

used to obtain direct estimates of quantal content before and during

200-impulse trains of stimulation. Because quantal content can be

estimated most reliably under low quantal conditions, these experiments

were carried out under conditions of reduced quantal content (0.6 mM

Ca2+, 5 mM Mg2+).

Figure 3-5B summarizes the results of 10 experiments in which

the effects of 500 mM glycerol on stimulation-induced changes in

quantal content during 200-impulse trains were examined. In these

experiments, V(t) of quantal content at the end of the trains was

increased significantly from 4.16 + 0.54, prior to the addition of

glycerol, to 5.47 0.71, following the removal of glycerol (p<0.02;











A B
A (t) of B v(t) of
8 EPP quantal 8
7 amplitude content 7
6 6 -
5 5
4 4
3 3
2 2
1 1
0 L0
[I control post-glycerol
p<.01 ** p<.02

Figure 3-5. Effect of glycerol treatment on stimulation-induced
changes in EPP amplitude and quantal content. Data were obtained in a
low Ca Ringer containing 0.6-0.8 mM Ca2+ and 5 mM Mg2+. Values of V(t)
represent the final V(t) obtained at the end of 200-impulse
conditioning trains before and following exposure to 500 mM glycerol.
The vertical lines indicate standard errors. Asterisks indicate
statistically significant differences between control and post-glycerol
data. (A): Effect of glycerol treatment on V(t) of EPP amplitude.
Data averaged from 3 experiments, including the experiment presented in
Figure 3-4. (B): Effect of glycerol treatment on V(t) of quantal
content. Data averaged from 10 experiments.








paired t-test). These results are similar to the effect of glycerol on

stimulation-induced changes in EPP amplitude under similar conditions

of low levels of release (low quantal content; Figure 3-5A), suggesting

that glycerol is mediating its effects presynaptically.

In 6 of the 10 intracellular experiments summarized in Figure 3-5B,

final V(t) was increased more than 25% following the removal of

glycerol. In these same experiments, control (unconditioned) quantal

content was also decreased significantly following glycerol treatment

(1.84 + 0.05 vs. 0.68 + 0.25; p<0.05, paired t-test). In the other 4

experiments there was little or no effect of glycerol on either control

quantal content or on final V(t). These results suggest that not all

nerve terminals are equally affected by glycerol treatment, a not

altogether surprising finding considering the previous observations of

similar variability in the effects of glycerol on muscle t-tubules from

muscle to muscle, and even from fiber to fiber within a single muscle

(Eisenberg and Eisenberg, 1968; Howell, 1969). Since the responses

recorded with surface electrodes represent averaged responses of

several synapses (see Materials and Methods, Chapter 2), one would

expect to more consistently observe a glycerol effect with

extracellular recording, even if not all synapses are equally affected.

(In 4 of the 6 experiments in which large effects of glycerol-treatment

on V(t) and quantal content were observed, measures of MEPP amplitude

were also obtained and no significant changes were found, suggesting

that in these experiments glycerol had little or no effect on

postsynaptic sensitivity.)








Discussion

Fujino and coworkers first reported in 1961 that the return of frog

skeletal muscle to an isosmotic Ringer solution following hyperosmotic

glycerol treatment resulted in an irreversible loss of muscle twitch

which was not accompanied by any effect on resting or action potentials

(Fujino et al., 1961). This treatment was later shown to selectively

disrupt the muscle's t-tubule system (e.g., see Howell and Jenden,

1967; Eisenberg and Eisenberg, 1968; Howell, 1969). Despite the rather

drastic effects of glycerol treatment on muscle function, results of

several studies indicated that this treatment had no effect on either

spontaneous transmitter release or on the properties of unconditioned

evoked release (Furshpan, 1956; Gage and Eisenberg, 1967; Lermer and

Rahamimoff, 1970; Miyamoto, 1975). Thus, it has been assumed that

glycerol has no effect on the process of transmitter release. However,

the results presented here indicate that glycerol treatment can indeed

affect release when the nerve terminal is stimulated repetitively. The

effect of glycerol treatment on the phenomena of stimulation-induced

changes in release progresses with time for hours following the removal

of the glycerol solution.

The mechanism by which glycerol might alter stimulation-induced

changes in release is not clear. However, the results of the sucrose

experiments suggest that the effects are specific to glycerol and not

due simply to changes in osmolarity. Glycerol has been shown to

penetrate into cells (e.g., Jenkinson, 1960; Bozler, 1961), and thus is

most likely acting at some site within the nerve terminal, perhaps by

causing a disruption of some subcellular organelle, much like the known








glycerol-induced disruption of muscle t-tubules (e.g., see Howell and

Jenden, 1967; Howell, 1969).

Stimulation-induced increases in release have been attributed to an

increase in intraterminal Ca2+ levels (e.g., see Elmqvist and Quastel,

1965b; Gage and Hubbard, 1966; Rosenthal, 1969; Erulkar and Rahamimoff,

1978; Zengel and Magleby, 1982). It is possible that the effect of

glycerol may involve the release of calcium ions from an intraterminal

calcium store, such as the mitochondria or smooth endoplasmic

reticulum, or alternatively, an impairment in a mechanism involved in

the buffering of calcium at the nerve terminal. In support of this

idea, Bianchi and Bolton (1974) observed a marked increase in total

intracellular calcium in frog skeletal muscle following glycerol

treatment. The fact that the effect of glycerol treatment on V(t)

progresses with time is consistent with a reduced capacity to buffer
Ca2+ inside the terminal; if the buffering capacity is somehow

compromised, Ca2+ would tend to build up with continuous or repetitive

stimulation.

Interestingly, the levels of unconditioned quantal content were

reduced following removal of the glycerol solution. This effect is the

opposite of what might be expected if intraterminal Ca2+ levels were

increased following glycerol solution. Perhaps different pools of

calcium ions within the nerve terminal are involved in the initiation

and the modulation of release (see Zengel and Magleby, 1980; Swandulla

et al., 1991). If so, then one possible explanation for this apparent

discrepancy is that the effects of glycerol treatment on unconditioned

release and on stimulation-induced changes in release may be mediated

through different mechanisms, with the latter effect involving a








specific calcium pool that is primarily involved in the modulation of

transmitter release, rather than in its initiation.

The lack of understanding of the actual mechanism of the effects of

glycerol on release, combined with the fact that the effect of glycerol

on stimulation-induced changes in release progresses with time, poses a

real problem when interpreting data from experiments aimed at examining

changes in transmitter release at the neuromuscular junction.

Therefore, it may be better to use some other technique for blocking

muscle contraction and/or action potentials. The method most commonly

used to prevent the generation of muscle action potentials is the

addition of curare to partially block the muscle's acetylcholine

receptors (Jenkinson, 1960). This method, however, has also been shown

to affect stimulation-induced changes in transmitter release at the

frog neuromuscular junction (Magleby et al., 1981; Matzner et al.,

1988; Sosa and Zengel, 1992; also, see Chapter 2). Del Castillo and

Escalona de Motta (1978) found that exposing a nerve-muscle preparation

to a formamide-containing Ringer solution was effective in blocking

excitation-contraction coupling. However, these investigators also

reported that prolonged exposure to formamide led to a block of

neuromuscular transmission, possibly as a result of effects on calcium

movement and storage within the nerve terminal (del Castillo and

Escalona de Motta, 1978).

Other methods for preventing muscle contraction that may be better

suited for experiments on neurotransmitter release include the

cut-muscle preparation (Barstad, 1962; Barstad and Lilleheil, 1968;

Hubbard and Wilson, 1973) or the addition of mu-conotoxin GIIIA, a

toxin isolated from the marine snail Conus geographus that








selectively blocks the sodium channels of the muscle (Cruz et al.,

1985; Hong and Chang, 1989a; also see Chapter 4). Results of

experiments presented in the next chapter indicate that this latter

technique is effective in preventing muscle action potentials, while

having little or no effect on stimulation-induced changes in release.

The neuromuscular junction remains one of the models best suited

for the study of stimulation-induced changes in release. In order to

use this preparation to study release under normal conditions, care

must be taken in selecting a method for preventing generation of muscle

action potentials and/or contractions. Although the method of glycerol

treatment has been used successfully in the past for a number of

different studies, the effects of this treatment on transmitter release

reported here should be taken into consideration before choosing to use

it in experiments concerned with the study of nerve terminal

physiology.













CHAPTER 4


USE OF MU-CONOTOXIN GIIIA FOR THE STUDY OF
SYNAPTIC TRANSMISSION


Introduction

As discussed previously, block of muscle action potentials and

muscle contraction has traditionally been achieved by adding

d-tubocurarine to block postsynaptic acetylcholine receptors and reduce

the EPP amplitude to subthreshold levels (Eccles et al., 1941;

Jenkinson, 1960), or by exposing the preparation to glycerol treatment

to disrupt the muscle t-tubule system and disable excitation-

contraction coupling (Howell, 1969). Both of these methods of

preventing muscle contraction, however, can also affect transmitter

release (see Chapters 2 and 3). An alternative method that has

recently become available consists of blocking the generation of muscle

action potentials by adding mu-conotoxin GIIIA to the bathing solution.

This toxin, which comes from the venom of the marine snail Conus

geographus, selectively blocks the voltage-dependent sodium channels of

skeletal muscle without affecting those of motor nerves (Cruz et al.,

1985; Moczydlowski et al., 1986). This ability to preferentially block

muscle but not axonal sodium channels makes this toxin a convenient

tool for studying neuromuscular transmission without interference from

muscle action potentials.








Previous studies of the effects of this toxin on neuromuscular

transmission concentrated mainly on its effect on nerve action

potentials (Cruz et al., 1985). Thus, little was known about the

effects of mu-conotoxin GIIIA on transmitter release. This chapter

presents results of experiments designed to look at the effects of the

synthetic form of mu-conotoxin GIIIA on various parameters of synaptic

transmission at the frog neuromuscular junction. It was found that 5

PM mu-conotoxin consistently blocked muscle action potentials, but had
no effect on nerve action potentials. The toxin also had no effect on

the amplitude or frequency of MEPPs, on the amplitude or time course of

EPPs, or on stimulation-induced changes in EPP amplitude.


Material and Methods

Details of the preparation, solutions, and the recording, data

collection and analysis techniques were described in Chapter 2.

A stock solution of the synthetic mu-conotoxin was made by

dissolving it in deionized water in a concentration of 1 mM. Aliquots

of this solution were then added directly to the bathing solution

to achieve the desired final concentrations. Synthetic mu-conotoxin

GIIIA was obtained from BACHEM (Torrance, CA).


Results

Effective Concentration of mu-Conotoxin Required to Block Muscle Action
Potentials at the Frog Neuromuscular Junction

Cruz and coworkers (1985) were able to paralyze cutaneous

pectoralis muscle from frog using 3 yM mu-conotoxin GIIIA purified from

crude venom. I found that, under the conditions of my experiments, a








minimum of 5 yM synthetic mu-conotoxin GIIIA was required to

consistently block muscle action potentials and contraction at the frog

sartorius muscle when the levels of release were normal or high (1.8 or

3.6 mM Ca2+, respectively). EPP amplitudes as large as 55 mV were

recorded (in 3.6 mM Ca2+ Ringer) at this concentration of toxin. (On

two occasions, higher concentrations of between 10 and 15 uM

mu-conotoxin GIIIA were needed to paralyze the muscle, but this was

attributed to partial loss of the biological activity of the particular

batch of toxin being used. It was interesting to note that in these

two particular instances, the muscles started to twitch weakly after

being in the toxin for about 2 hours).

Figure 4-1 shows muscle responses to nerve stimulation recorded

extracellularly under conditions of high levels of release. Prior to

addition of the toxin (control), a muscle action potential was

generated. Ten min after adding 5 PM mu-conotoxin GIIIA, the muscle no

longer generated an action potential and the underlying EPP could be

recorded. The muscle action potential returned 33 min after removing

the toxin. On average, muscle contraction was completely blocked 10 to

20 min after the addition of 5 PM toxin. The effect of the toxin was

reversible and muscles started to twitch again 20 to 35 min after

removal of toxin.

Effect of mu-Conotoxin on the Nerve Action Potential

Once the minimum concentration of synthetic mu-conotoxin GIIIA

needed to consistently block muscle contraction was determined,

experiments were done to determine what effects, if any, this

concentration of toxin had on the nerve action potential. Results from




























5 IpM -conotoxin (10 min)
5 |iM li-conotoxin (10 min)


wash (33 min)


Figure 4-1. Tracings of extracellularly recorded responses of a muscle
to nerve stimulation in the presence and absence of mu-conotoxin GIIIA.
Data were obtained in a high Ca Ringer containing 3.6 mM Ca2+.
Vertical scale: 400 )V; horizontal scale: 4 msec.


control








these experiments are presented in Table 4-1 and are described below.

Nerve action potentials were recorded extracellularly. The results of

an experiment carried out under conditions of low levels of release are

presented in Table 4-1A. In this experiment the preparation was first

exposed to 5 pM toxin for 10 min; the concentration was then increased

to 50 pM for another 20 min. There was no significant difference in

either amplitude or duration of the nerve action potential after

exposure to the toxin, even at a concentration 10 times that required

to block muscle contraction. Similar results were obtained when the

concentration of toxin was increased from 5 to 25 PM under conditions

of high quantal content (Table 4-B1; there seemed to be a trend towards

an increase in action potential duration, but this increase was not

statistically significant).

Effect of mu-Conotoxin on Synaptic Transmission

Experiments were also done to look at the effect of 5 PM

mu-conotoxin GIIIA on various parameters of synaptic transmission. The

results of these experiments are presented in Table 4-2. No

significant effects of the toxin on the amplitude and time course of

EPPs, on quantal content, or on the frequency of MEPPs were observed,

suggesting that mu-conotoxin GIIIA has no effect on quantal transmitter

release. The toxin also appears to have no effect on postsynaptic

sensitivity to acetylcholine, since there was no significant change in

the amplitude of MEPPs. Also no appreciable effects of the toxin on

the amplitude of extracellularly recorded EPPs were observed, even at

concentrations as high as 50 pM (data not shown).























TABLE 4-1. EFFECT OF MU-CONOTOXIN GIIIA ON EXTRACELLULARLY RECORDED
NERVE ACTION POTENTIALS

mu-conotoxin amplitude duration
GIIIA (pM) (mV) (msec)

A. 0.6 mM Ca2+ 0 0.29 + 0.01 0.84 + 0.03 (3)
5 0.30 + 0.01 0.76 + 0.01 (3)
50 0.30 0.01 0.88 0.03 (4)
B. 3.6 mM Ca2+ 5 0.23 + 0.01 0.85 + 0.02 (2)
25 0.22 + 0.01 1.02 + 0.04 (3)
Values are mean + standard error; number of observations in brackets.
Each observation was obtained from averages of between 32 and 80
trials. Measurements obtained using a Nicolet 1170 signal average.






















TABLE 4-2. EFFECT OF MU-CONOTOXIN GIIIA ON PARAMETERS OF TRANSMITTER
RELEASE

0 toxin 5 pM toxin

EPP amplitude (mV) 0.69 + 0.15 0.75 + 0.18 (4)
EPP 10-90% rise time (msec) 0.60 + 0.03 0.61 + 0.05 (4)
EPP 1/2 decay time (msec) 2.48 + 0.24 2.57 + 0.50 (4)
quantal content 2.62 + 0.41 2.38 + 0.46 (23)
MEPP frequency (sec-1) 1.51 + 0.31 2.16 + 0.38 (24)
MEPP amplitude (mV) 0.48 + 0.04 0.47 + 0.05 (24)

Experiments were carried out under conditions of low levels of release
using intracellular recording techniques. Estimates of mean MEPP
amplitude were obtained from averages of between 200 and 400 MEPPs
measured at each end-plate. MEPP frequency was calculated by counting
MEPPs recorded during time periods of 10 minutes. A Nicolet 1170
signal average was used to measure mean amplitude, 10-90% rise time
and 1/2 decay time of averages of 100 EPPs. Values are mean + standard
error; number of experiments indicated in brackets.








Effect of mu-Conotoxin on Stimulation-Induced Changes in Transmitter
Release

One important feature of the mechanism of release is that the

amount of transmitter released following each nerve impulse changes as

a function of previous synaptic activity. At the frog neuromuscular

junction, repetitive stimulation under conditions of low levels of

release leads to a progressive increase in EPP amplitude that has been

attributed to an increase in the amount of transmitter being released

(e.g., Magleby and Zengel, 1976a). Experiments were done to

investigate whether mu-conotoxin GIIIA affects stimulation-induced

increases in release. Nerves were conditioned with trains of 100

impulses (20 impulses/sec). These trains were applied continuously

throughout the experiment, allowing sufficient time between trains to

ensure that release levels had recovered to preconditioning levels (see

Materials and Methods in Chapter 2 for further details). Figure 4-2A

shows results from one experiment done under conditions of low levels

of release. The lines plot changes in EPP amplitude (expressed as % of

the first EPP of the train) during conditioning trains before

(continuous line) and after (dashed line) adding 5 pM toxin. There was

virtually no difference in the changes in EPP amplitude observed during

the conditioning trains following addition of mu-conotoxin. Similar

results were observed in 5 other experiments carried out under

conditions of low levels of release.

Since mu-conotoxin GIIIA promises to be most valuable in the study

of neuromuscular transmission under conditions where the amount of

transmitter released is great enough to lead to muscle contraction,

experiments were also done to investigate whether the toxin affects
























Figure 4-2. Effect of 5 pM mu-conotoxin GIIIA on stimulation-induced
changes in EPP amplitude under conditions of low (A) and high (B)
levels of release. The lines plot changes in EPP amplitude as a
function of time during 100-impulse trains applied at a frequency of 20
impulses/sec. EPP amplitude is expressed as a percent of the amplitude
of the first EPP of the conditioning train. Each line represents the
average of from 4 to 16 consecutive trials. (A): Data collected in
Ringer solution containing 0.7 mM Ca2+ and 5 mM Mg2+. Conditioning
trains were applied at intervals of 7 min to allow for the recovery of
EPP amplitude to preconditioning levels. The dashed line represents
the average of data collected between 10 and 120 min following addition
of toxin. B: Data collected in Ringer solution containing 3.6 mM Ca2+
and 3 pg/ml curare. Conditioning trains were applied at intervals of
17 min to allow for the recovery of EPP amplitude to preconditioning
levels. The dashed line represents the average of data collected
between 10 and 150 min following addition of toxin.










450
400
350
300
250
200
150
100
0 1

--control

140
120
100
80
60
40
20
0


0.7 mM Ca2+


2 3 4 5

--...---- +5 /M a/-conotoxin


3.6 mM Ca2+


4 5


0 1 2 3
Time (sec)








stimulation-induced changes in release observed under such conditions.

Under conditions of normal or higher levels of release, repetitive

stimulation typically leads to a brief initial increase in EPP

amplitude that is immediately followed by a decrease that continues for

the duration of stimulation. This decrease in EPP amplitude,

attributed to a decrease in the amount of transmitter being released,

is known as depression (del Castillo and Katz, 1954c; Takeuchi, 1958;

Brooks and Thies, 1962). Experiments were done to look at the effect

of the toxin on depression of transmitter release under conditions of

high levels of release (3.6 mM Ca2+ Ringer; to compare changes in

release before and after addition of the toxin, curare was present at a

constant concentration throughout the experiments). The stimulation

paradigm used was the same as that described above for the experiments

under conditions of low levels of release. Figure 4-2B shows results

from one experiment. The lines plot changes in EPP amplitude during

conditioning trains before (continuous line) and after (dashed line)

adding 5 yM toxin. As in the experiments done under conditions of low

levels of release, there was no difference in the changes in EPP

amplitude observed during the conditioning trains following addition of

toxin. Similar results were observed in 5 other experiments carried

out under conditions of high levels of release.


Discussion

It has been shown that a concentration of 5 pM synthetic

mu-conotoxin GIIIA is effective in blocking muscle action potentials at

the frog neuromuscular junction, while having no effects on the

presynaptic nerve action potential. Even at toxin concentrations as








high as 50 PM no effect on the amplitude or duration of the nerve

action potential was observed, indicating that this toxin is highly

selective for muscle voltage-dependent sodium channels in this

preparation (Cruz et al., 1985).

Previous investigators (Nakamura et al., 1983; Kobayashi et al.,

1986; Ohizumi et al., 1986; Hong and Chang, 1989, 1991) have reported

that a different form of mu-conotoxin, mu-conotoxin GIIIB

(geographutoxin II), effectively blocked skeletal muscle contraction in

mammalian preparations at lower concentrations (0.2-1.5 pM) than those

required in the experiments described here. It has also been reported

that EPPs at mouse neuromuscular junctions were abolished by

mu-conotoxin GIIIB at concentrations as low as 3 pM (Hong and Chang,

1989). These differences could arise from either differences in the

biological activities of the two forms of toxin (Sato et al., 1983;

Moczydlowski et al., 1986), or to species differences (Olivera et al.,

1985; Ohizumi et al., 1986).

The lack of any apparent effect of this toxin on the presynaptic

and postsynaptic mechanisms of synaptic transmission, combined with the

complete reversibility of its effect on the muscle action potential,

makes the synthetic form of mu-conotoxin GIIIA an invaluable tool for

the study of synaptic transmission at the neuromuscular junction under

conditions of normal levels of release.













CHAPTER 5


CALCIUM AND DEPRESSION OF TRANSMITTER RELEASE


Introduction

At the neuromuscular junction, repetitive stimulation under

conditions of normal or higher levels of release leads to a decrease in

EPP amplitude that continues to develop for the duration of stimulation

(del Castillo and Katz, 1954c; Takeuchi, 1958; Betz, 1970). Several

lines of evidence suggest that this decrease in EPP amplitude, termed

depression, results from a decrease in the number of quanta that are

released from the nerve terminal. Del Castillo and Katz (1954c) showed

that following prolonged periods of stimulation, EPP amplitude was

reduced while the size of the miniature EPPs (MEPPs) remained

unchanged. Brooks and Thies (1962) observed that the decrease of EPP

amplitude during repetitive stimulation was linearly related to a

reduction in the quantal content of the response. It has also been

shown at the frog neuromuscular junction that the sensitivity of the

postsynaptic membrane to iontophoresed acetylcholine remains unchanged

during depression of EPP amplitude (Otsuka et al., 1962). In addition,

depression can be observed even when the frog nerve terminal is

stimulated electrotonically (Katz and Miledi, 1967b; Betz, 1970),

suggesting that this phenomenon is not dependent on changes in the

nerve action potential. At the squid synapse, depression develops








under conditions in which the presynaptic action potential indeed

remains unchanged (Takeuchi and Takeuchi, 1962).

Because the magnitude of depression appears to be dependent on the

amount of transmitter released by conditioning impulses (Thies, 1965),

depression has traditionally been attributed to a depletion of the

store of transmitter available for release (Takeuchi, 1958; Thies,

1965; Charlton et al., 1982). However, this is not the only

possibility since this observed relationship between depression and the

total amount of transmitter being released can be indirect. The amount

of transmitter released following each nerve impulse is itself

dependent on other factors, including the amount of Ca2+ entering the

nerve terminal and/or already present inside the terminal (Katz and

Miledi, 1967a). Since depression of transmitter release is most

prominent under conditions of normal levels of release, in which

intraterminal Ca2+ levels are expected to be high, it is possible

that these high levels of Ca2+ could in some way reduce release. In

support of this, in several studies experimental manipulations which

would be expected to increase intracellular Ca2+ levels have been

observed to dramatically decrease release (Adams et al., 1985; Molgo

and Pecot-Dechavassine, 1988; Fatatis et al., 1992).

Both rapid and slow developing phases of depression have been

described (Takeuchi, 1958; Brooks and Thies, 1962; Elmqvist and

Quastel, 1965b; Mallart and Martin, 1968; Betz, 1970). In many studies

of depression of transmitter release, the conditions of frequency and

duration of stimulation used are such that only the rapidly developing

phase is clearly observed (e.g., Lundberg and Quilish, 1953ab; Thies,

1965; Magleby et al., 1981). Thus, not much attention has been given








to the study of the mechanisms underlying the two apparent phases of

depression. For instance, it is not known whether the two phases of

depression have a common mechanism or if they are due to different

processes. If the mechanisms underlying each of these two phases are

different, it may be possible to selectively affect one phase, or

alternatively, the two phases may be affected differently by the same

experimental manipulation.

The purpose of this study was to examine the role of Ca2+ in

depression of transmitter release by determining how manipulation of

some of the processes that affect intraterminal levels of calcium

affect the changes in EPP amplitude observed during repetitive

stimulation under conditions of normal or higher levels of release at

the neuromuscular junction of the frog. Experiments were done using a

variety of agents known to affect calcium entry and intracellular

levels of free calcium. It was found that while changes in the levels

of Ca2+ inside the nerve terminal affected primarily the initial

rapidly developing phase of depression, the entry of Ca2+ through

voltage-dependent calcium channels played a more significant role in

the second slower phase. These results indicate that the two phases of

depression are most likely mediated through different mechanisms and

that changes in calcium associated with different compartments inside

the nerve terminal appear to be involved in these mechanisms.


Materials and Methods

Preparation and Solutions

The preparation and bathing solutions used for these experiments

are the same as those described in the Materials and Methods section in

Chapter 2.








Concentrated stock solutions of divalent cation calcium channel

blockers (1-100 mM) and omega-conotoxin GVIA (0.01 mM) were made up in

deionized water and appropriate amounts added directly to the bathing

solution in the recording chamber to achieve the desired final

concentration. Concentrated stock solutions (10 mM) of ionomycin and

cyclopiazonic acid (CPA) were made up using dimethyl sulfoxide (DMSO)

as the solvent, and appropriate amounts of these solutions were then

added to the bathing solution to obtain the desired final

concentration. The final concentration of DMSO in the bathing solution

was 0.1% (v/v). Although previous observations in this laboratory as

well as those of others (e.g., McLarnon et al., 1986) indicate that

DMSO at this concentration has no effects on synaptic transmission or

on stimulation-induced changes in release, control data for all

ionomycin and CPA experiments were obtained in Ringer solutions

containing DMSO.

Omega-conotoxin GVIA (lot numbers 801429, 902334, 902789),

ionomycin (free acid form; lot number 902918) and CPA (lot number

472292) were all obtained from Calbiochem (San Diego, CA). All salts

of the divalent cations and the solvent DMSO were obtained from Sigma

Chemical (St. Louis, MO).

Data Collection

For all experiments, standard extracellular recording techniques

were used to record EPPs before, during and after 100-200 impulse

trains of stimulation delivered at a frequency of 20 impulses/sec. A

MINC-11 computer was used to generate the stimulation patterns, measure

and store EPP amplitudes during the experiments and analyze the data.








For further details on the recording and data collection techniques,

refer to the Materials and Methods section in Chapter 2.

Data Analysis

Estimates of the fractional change in EPP amplitude resulting from

repetitive stimulation were expressed as V(t) (equation 2-1, Materials

and Methods, Chapter 2). Figure 5-1 presents a typical plot of V(t) as

a function of time during and following a 200-impulse (20 impulses/sec)
conditioning train (filled circles). It can be seen that immediately

after the onset of the conditioning stimulation (time=O) there is an

initial increase in V(t) that is quickly followed by a decrease. It

can also be noticed that there appear to be two distinct phases of

depression, an initial rapidly developing phase during the first second

of stimulation, followed by a more slowly developing phase that
continues for the remainder of the conditioning train. To be able to

make comparisons of the effects of experimental manipulations on the

two phases of depression, it is important to be able to describe each

phase in a quantitative manner that is both consistent and

reproducible. The magnitude of each phase of depression in this study

is measured as an absolute value of fractional change in EPP amplitude

(V(t)). The change in V(t) during the first phase of depression (01)
is measured from the peak (maximal value) V(t) reached during the

initial increase observed at the beginning of the train to the point

where the slower phase of depression (D2) begins (arrow in Figure 5-1).
The change in V(t) during the second phase (D2) is measured from the

end of the first phase to the end of the conditioning train (see Figure

5-1).












slope
~-----------------------------------


* data
- regression line




D2


I I


0 1 2 3 4 5 6 7 8 9 10

Time during train (sec)


Figure 5-1. Plot of fractional change in EPP amplitude (V(t), Equation
2-1), during repetitive stimulation at 20 impulses/sec under conditions
of high levels of release. This plot illustrates how the point at
which the first phase of depression ends and the second phase begins is
determined (see Data Analysis in the Materials and Methods section).


0.4


0.2


0.0


-0.2


-0.4


-0.6


-0.8


I I I








The point at which the changes in V(t) during the first phase end

and those of the second phase begin is determined using the computer

graphics software SigmaPlot (Jandel Scientific, San Rafael, CA) to

calculate the point in time at which the change in slope of the plot of

V(t) vs. time is greatest. This point is determined in the following

manner:

Using the computer, a regression line is fitted to the data, from

the peak value of V(t) to the end of the conditioning train.

This regression line (solid line in Figure 5-1) is described by a

simple polynomial equation of order 10 (r value of 0.999).

Using the parameters of the equation that describes the

regression line, the computer calculates (for each time point)

the first derivative of the mathematical function that describes

V(t). The derivative is equivalent to the slope of a tangent

line drawn at each point. The computer also calculates the

change in slope observed from one point to the next. (The dotted

line on Figure 5-1 is a plot of the values of the slope at each

point as a function of time; the axis showing the values of slope

is not shown).

The time at which the slope is 30% of the slope at peak V(t) is

determined. The change in slope between this point and the next

is noted. The time at which D1 ends and D2 begins (arrow in

Figure 5-1) is then marked as that at which the change in slope

is half that observed at 30% of the slope of peak V(t). This is

the middle point of the time period during which the greater

change in slope of the plot of V(t) vs. time is observed.








Other parameters used to describe changes in V(t) under conditions

of depression of EPP amplitude are total depression (D(t)), which

represents the sum of D1 and D2, and final V(t), the value of V(t) at

the end of the train (see Figure 2-1 in the Materials and Methods

section of Chapter 2).

Results

The actual amount of free Ca2+ inside the nerve terminal at any

given moment is determined by a balance of several processes, including
Ca2+ entry, release from intraterminal stores, and intracellular Ca2+

buffering, sequestering and extrusion. The manner in which these

processes interact with each other determines the extent and

distribution of Ca2+ gradients in various compartments within the

presynaptic nerve terminal. If depression of transmitter release
reflects a change in Ca2+ in one or more compartments inside the

terminal, then manipulation of the processes involved in the regulation
of the intracellular concentration of Ca2+ ([Ca2+]i) might be expected

to have an effect on depression during repetitive stimulation.

Effect of Changes in the Concentration of Extracellular Calcium on
Depression of EPP amplitude in the Presence of Curare

Figure 5-2 shows the effect of increasing the concentration of Ca2+

in the bathing solution ([Ca2+]o) on changes in EPP amplitude during
repetitive stimulation under conditions of high levels of release. The

continuous line in Figure 5-2A plots V(t), the fractional change in EPP
amplitude (equation 2-1, Materials and Methods, Chapter 2), during a
200-impulse conditioning train in normal Ca (1.8 mM) Ringer. When the
[Ca2+]o was doubled to 3.6 mM (long-dashed line), V(t) during the train




























Figure 5-2. Effect of [Ca2+]o on stimulation-induced changes in EPP
amplitude under conditions of high levels of release. The lines plot
changes in V(t) as a function of time during 200-impulse trains of
stimulation. Data were obtained in Ringer solutions containing 5 ug/ml
curare and 1.8 mM (continuous line), 3.6 mM (long-dashed line) or 7.2
mM (short-dashed line) Ca2+. Each line represents the average of from
2 to 6 consecutive trials. (A): Data collected from a single
preparation. (B): Averages of data collected from different
preparations. In this and all other figures, n indicates the number of
preparations from which the data were averaged. Standard errors
indicated by vertical lines.










A 0.6

0.4
0.2

0.0
- -0.2
-0.4
-0.6
-0.8
-1.0


2+
- 1.8 mM Ca
2+
--- 3.6 mM Ca +
....... 7.2 mM Co


3-.,. -- -J- t'--l n=12
*-I-::- -"-_
." ---. -.. .... .
"" I--.I ... n=39
-... --. n=15



1 2 3 4 5 6 7 8 9 10

Time during train (sec)


2+
1.8 mM Co
---- 3.6 mM Ca
.......------ 7.2 mM Ca









I I I I I I I I I I

0 1 2 3 4 5 6 7 8 9 10

Time during train (sec)


0.6
0.4
0.2

0.0
-0.2
-0.4

-0.6
-0.8
-1.0








was decreased. The initial increase in V(t) normally observed at the

beginning of the train was reduced and the value of V(t) at the end of
the train (final V(t)) was more negative, indicating that the decrease

in the amplitude of successive EPPs (relative to EPP amplitude before
the conditioning train) was more pronounced in the higher [Ca2+]o. A

further decrease in V(t) during the conditioning train was observed

when [Ca2+1] was doubled once more to 7.2 mM (short-dashed line).

Similar results were observed in a total of 8 experiments in which
nerves were conditioned with 200-impulse trains of stimulation applied

at a frequency of 20 impulses/sec and [Ca2+]o was varied between 1.8,

3.6 and/or 7.2 mM, regardless of the order in which the preparations

were exposed to each concentration of Ca2+. These results are

consistent with the hypothesis that Ca2+ plays a role in the mechanisms

underlying depression of transmitter release and with previous

observations of the effects of changes in extracellular Ca2+ on
stimulation-induced changes in EPP amplitude (e.g., Lundberg and

Quilisch, 1953b; Thies, 1965).
Figure 5-2B plots V(t) during 200-impulse conditioning trains

averaged from different experimental preparations in which the effects

of one or more concentrations of [Ca2+]o were examined. With the

exception of the points of maximal initial increase (peak V(t)) in 3.6

and 7.2 mM Ca, each point marked by standard error bars during the
trains in the three different [Ca2+]o were significantly different from
the corresponding point in the other two concentrations (p<0.015;

unpaired t-test). Results from these experiments are summarized in

Figure 5-3. Raising [Ca2+]o from the normal concentration of 1.8 mM to

3.6 or 7.2 mM significantly reduced peak V(t) and final V(t), the value













1.0




0.5




0.0
peak
V(t)

-0.5




-1.0

1 1.8 mM Ca2+
(n-12)


1.0




0.5

nal
(t)
0.0
D(t) D1 D2


-0.5



p<0.05
-1.0

3.6 mM Ca2+ 7.2 mM Co2+
(n-3) (n-15)


Figure 5-3. Effect of [Ca2+]o on peak V(t), final V(t), D(t), Dl and
D2. Data averaged from the same experiments included in Figure 5-2B.
Standard errors indicated by vertical lines. Asterisks indicate
statistically significant differences between mean values obtained in
3.6 and 7.2 mM Ca2+ and those obtained in 1.8 mM Ca2+.








of V(t) at the end of the conditioning train. D(t), the difference

between peak V(t) and final V(t) (see Figure 2-1), was increased,

although not significantly, at the higher [Ca2+]o. Also shown in this

figure are estimates of the magnitudes of the two observed phases of
depression, Dl and D2 (see Materials and Methods). Raising the [Ca2+]o

significantly increased the magnitude of D1, while not having an

appreciable effect on the magnitude of D2.

Effects of Changes in the Concentration of Extracellular Calcium on
Depression of EPP Amplitude in the Absence of Curare

All experiments described above were done in the presence of 5

pg/ml curare to reduce the amplitude of EPPs to levels that were below

threshold for the generation of a muscle action potential. As was

shown in Chapter 2 of this dissertation, depression of EPP amplitude is

dramatically increased in the presence of curare (see Figure 2-3A).

The apparent lack of an effect of increased [Ca2+]o on D2 in the
experiments described above could be an indication that a maximal level

has already been reached in normal (1.8 mM) Ca2+. To determine if this

was the case, and to rule out the possibility that the observed effects

on V(t) of changing [Ca2+]o are in some way related to the presence of

curare, I decided to examine these effects in the absence of curare,

using mu-conotoxin to block muscle action potentials (see Chapter 4).

Results from these experiments are shown in Figure 5-4.

As expected, the levels of depression observed in the three
concentrations of Ca2+ examined were smaller in the absence of curare
(compare Figure 5-4A with Figure 5-2). Nevertheless, increasing
[Ca2+]o in the absence of curare had a similar effect on changes in

V(t) during conditioning trains as that observed in the presence of



























Figure 5-4. Effect of [Ca2+]o on stimulation-induced changes in EPP
amplitude under conditions of high levels of release in the absence of
curare. (A): Data collected from a single preparation. The lines plot
changes in V(t) as a function of time during 200-impulse trains of
stimulation in Ringer solutions containing 1.8 mM (continuous line),
3.6 mM (long-dashed line) and 7.2 mM (short-dashed line) Ca2+. Muscle
action potentials were blocked with 10 pM mu-conotoxin. Each line
represents the average of from 2 to 3 consecutive trials. (B): Effect
of [Ca2+]o on mean peak V(t), final V(t), D(t), DI and D2. Standard
errors indicated by vertical lines. Asterisks indicate statistically
significant differences between mean values obtained in 1.8 mM Ca2+ and
those obtained in 3.6 and 7.2 mM Ca2+.









A 1.8 mM Ca2
---- 3.6 mM Ca
0.4 .--.....- 7.2 mM Ca

0.2

> 0.0 ----
.-....... .. .
0.2 "........................

-0.4

0 1 2 3 4 5 6 7 8 9 10
Time during train (sec)

B 0.4 0.4
final p*
0.2 V(t) 0.2


0.0 0.0
peak D(t) D1 D2
V(t)
-0.2 -v -0.2


-0.4 p<.025 -0.4

I] 1.8 mM Ca2+ 7 3.6 mM Ca2+ M 7.2 mM Ca2+
(n-s) (n-6) (n-3)








curare. This is illustrated in Figure 5-4, which presents results from

a single experiment (A) and averaged data from a number of experiments

(B) carried out in the absence of curare. Raising [Ca2+]o from 1.8 mM

to 3.6-7.2 mM in the absence of curare resulted in significant changes

in peak V(t), final V(t), and the magnitudes of D(t) and D2 (p<0.025;

unpaired t-test). Dl was increased but not significantly, probably due

to the variability amongst the small number of preparations studied.

Final V(t) and D(t) were also changed significantly when [Ca2+]o was
increased from 3.6 to 7.2 mM.

As expected, both D(t) and D2 were increased as [Ca2+]o was raised

in the absence of curare. It is possible then, that the levels of

depression observed during conditioning trains in the presence of high

concentrations of curare are closely approaching or have already

reached a maximal level that cannot be increased much further by other

experimental manipulations such as increasing [Ca2+]o. Regardless of
whether this is the case or not, results from these experiments clearly
indicate that the effects of increasing [Ca2+]o on depression of EPP

amplitude are not dependent on the presence of curare in the bathing

solution.

Effect of Calcium Channel Blockers on Depression of EPP Amplitude

Increases in [Ca2+]o can increase the resting level of

intracellular calcium in the absence of stimulation, as well as
increase the amount of calcium that enters the nerve terminal through

voltage-dependent calcium channels in response to depolarization
(Requena et al., 1977). For example, Suszkiw and coworkers (1987) have

reported that increasing [Ca2+]o from 0 to 1 mM leads to an almost








3-fold increase in resting Ca2+ levels in rat brain synaptosomes.

Thus, the observed effects of increasing [Ca2+]o on V(t) described

above could be due to an effect of an increase in intracellular

calcium levels and/or an increase in calcium entry through the

voltage-dependent calcium channel. If Ca2+ entering through voltage-

dependent calcium channels contributes to depression then one would

expect to see a reduction in depression in the presence of blockers of

these channels. To test this hypothesis, I decided to look at the

effect of various calcium channel blockers on depression of EPP
amplitude.

Effect of divalent cations

As expected, addition to the bathing solution of 10-30 pM Cd2+, a

potent calcium channel blocker (e.g., Llinas et al., 1981; Cooper and

Manalis, 1984; Narahashi et al., 1987), led to a reduction in the

magnitude of depression (n=17). This reduction in depression was

observed almost immediately following the addition of Cd2+ to the

bathing solution and was readily reversed following its removal.

Figure 5-5A shows the effect of 20 pM Cd2+ on V(t) during 200-impulse

conditioning trains in one experiment. The continuous line plots V(t)

as a function of time during the train in the absence of Cd2+.

Following the addition of Cd2+ (short-dashed line), the initial

increase in V(t) was reduced and the second more slowly developing

phase of depression was almost completely eliminated. Figure 5-5B
summarizes results from various experiments in which the effects of

10 and 20 pM Cd2+ on depression of EPP amplitude were examined. Peak

V(t), final V(t) (absolute value) and D(t), as well as D1 and D2, were
reduced, in a concentration-dependent manner, following addition


























Figure 5-5. Effect of Cd2+ on stimulation-induced changes in EPP
amplitude under conditions of high levels of release. (A): Data
collected from a single preparation. The lines plot changes in V(t) as
a function of time during 200-impulse trains of stimulation in the
absence (continuous line) and presence (short-dashed line) of 20 pM
Cd2+. Data collected in Ringer solution containing 3.6 mM Ca2+ and 2.5
pg/ml curare. Each line represents the average of 7 consecutive
trials. (B): Effect of Cd2+ on mean peak V(t), final V(t), D(t), Dl
and D2. Standard errors indicated by vertical lines. Asterisks
indicate statistically significant differences between values obtained
in the absence and presence of Cd2+. (Statistical significance was
determined using Students unpaired t-test except when evaluating the
effect of 20 pM Cd2+ on Dl, in which case the variability between
control and experimental groups was different and the Wilcoxon rank-sum
test was used instead).













- 0 Cd2+
.......... + 20 LM Cd +


3.6 mM Ca2+


0.2


0.0


-0.2


-0.4


-0.6


1.0



0.5



0.0



-0.5


S 0 /M Cd2+ 77 +10 uM Cd2+ M +20 pM Cd2+
(n-13) (n-11) (n-8)


I I I I 1 I I I I I
0 1 2 3 4 5 6 7 8 9

Time during train (sec)


0.5



0.0



-0.5








of Cd2+. These effects were significant (p<0.025, unpaired t-test or

Wilcoxon rank-sum test; see figure legend), with the exception of the

reduction in peak V(t) and final V(t) following addition of 10 pM Cd2+.

It has been shown in previous studies that Cd2+ also reduces the

stimulation-induced increases in release that normally occur during
repetitive stimulation under conditions of low levels of release, when

depression is absent (Zengel, Lee, Sosa and Mosier, 1993). The
Cd2+-induced reduction in peak V(t) observed in the experiments

presented here most likely results from a reduction of the underlying

components of increased release that are thought to be present

simultaneously with depression. It is also likely that the observed
reduction in the magnitude of the first phase of depression is largely

due to this reduction in the initial increase in V(t). When the effect

of Cd2+ on peak V(t) was subtracted from its effect on DI, this latter

effect was no longer evident. On average, the value of V(t) reached at

the point at which one phase of depression ends and the other one
begins (see arrow in Figure 5-5A) was not changed significantly

following addition of Cd2+ (-0.29 0.03 in the absence of Cd2+,

-0.29 + 0.03 in 10 M Cd2+, -0.33 0.05 in 20 pM Cd2+; also, see

Figure 5-5A). Thus, it appears that Cd2+ primarily affects the second,
more slowly developing phase of depression. Similar results were

observed after addition of 20 pM Cd2+ in 3 other experiments in which
muscle action potentials were blocked using 10-20 PM mu-conotoxin
instead of curare.

The effects of other divalent cations, Co2+, Zn2+ and Ni2+, on

depression of EPP amplitude were also examined. Like Cd2+, these other

divalent cations also block voltage-dependent calcium channels








(Hagiwara and Takahashi, 1967; Weakly, 1973). Figure 5-6 shows the
effect of adding 300 pM Zn2+ on changes in V(t) during 200-impulse

conditioning trains. The continuous line plots V(t) as a function of
time during the train in the absence of Zn2+. Following the addition
of Zn2+ (short-dashed line), V(t) was increased for the whole duration
of the conditioning train. This effect was reversed following removal
of Zn2+ (long-dashed line). Unlike Cd2+, addition of Zn2+ resulted in
a decrease in both phases of depression instead of only in the second
phase. Similar effects were observed following the addition of Co2+
(100-250 pM) or Ni2+ (500-750 pM; data not shown).
The effect of the divalent cations on depression of EPP amplitude
was concentration-dependent. This concentration-dependence is clearly
seen in Figure 5-7, which plots changes in D(t) as a function of cation
concentration. It can also be seen that Cd2+ was effective at reducing
D(t) at much lower concentrations (10-30 uM) than the other cations.
Co2+ and Zn2+ were effective only at concentrations of 100 uM or
greater, while Ni2+ was the weakest blocker, producing an effect only

at concentrations greater than 500 pM. Thus, the order of potency in
reducing D(t) was: Cd2+>>Co2+,Zn2+>Ni2+. This order of potency is

similar to that reported for blocking some types of Ca2+ currents

(e.g., Byerly et al., 1985; Penner and Dreyer, 1986; Narahashi et al.,
1987; Dudel, 1990; Lentzner et al., 1992). It is also similar to the
order of potency reported for reducing control EPP amplitude and V(t)
at the end of short trains of stimulation under conditions of low
levels or release (Zengel, Lee, Sosa and Mosier, 1993).















0.4
--- 0 Zn2" (control)

S"2+
S:' -- 0 Zn (recovery)
0 .0 'I, ...




-0.4 V '.


2+ J ^^^^^s^^.
3.6 mM Co r
-0.8 -
-0. 8 I I I I I I I I

0 1 2 3 4 5 6 7 8 9 10

Time during train (sec)




Figure 5-6. Effect of Zn2+ on stimulation-induced changes in EPP
amplitude under conditions of high levels of release. Data collected
from a single preparation. The lines plot changes in V(t) as a
function of time during 200-impulse trains of stimulation before
(continuous line), during (short-dashed line) and after exposure
(long-dashed line) to 300 jM Zn2+. Data collected in Ringer solution
containing 3.6 mM Ca2+ and 5 pg/ml curare. Each line represents the
average of from 2 to 5 consecutive trials.














I ''' "I "' "" I '" "I .
100 ----------------- ------- ................. ... .......... ... ...
90 %-
--' 80
CD 70
- 60 -
L-
4. 50
0 40 2+
C 30 Co (2)
S 20 V Zn2+
A Ni
10
O I ,,,,I ,,,, ,,,,, .. I
0.1 1 10 100 1000

Cation concentration (j/M)



Figure 5-7. Concentration dependence of the effects of Cd2+, Co2+,
Zn2+ and Ni2+ on D(t) during 200-impulse trains of stimulation.
Control values of D(t) were first obtained in Ringer solution
containing 3.6 mM Ca2+ and 3-5 pg/ml curare. Values of D(t) following
the addition of varying concentrations of divalent cation blockers were
then expressed as a percent of these control values. Numbers in
parentheses indicate the number of preparations from which the data
were averaged. Standard errors indicated by vertical bars (points
without vertical bars were obtained from a single preparation).




Full Text
xml version 1.0 encoding UTF-8
REPORT xmlns http:www.fcla.edudlsmddaitss xmlns:xsi http:www.w3.org2001XMLSchema-instance xsi:schemaLocation http:www.fcla.edudlsmddaitssdaitssReport.xsd
INGEST IEID EFA3XJT84_UXHG6P INGEST_TIME 2012-09-24T14:29:30Z PACKAGE AA00011801_00001
AGREEMENT_INFO ACCOUNT UF PROJECT UFDC
FILES