PLACE IN RETURN BOX to remove this checkout from your record.
TO AVOID FINES return on or before date due.
MAY BE RECALLED with earlier due date if requested.

 

DATE DUE

DATE DUE

DATE DUE

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

6/01 cJClRC/DatoDuepGS-p. 15

 

RELATIONSHIP BETWEEN METHYLMERCURY-INDUCED DISRUPTION OF
INTRACELLULAR CALCIUM AND NEURONAL DEATH

By

Tobi Leigh Limke

A DISSERTATION

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

DOCTOR OF PHILOSOPHY

Department of Pharmacology and Toxicology
and
Institute for Environmental Toxicology

2001

ABSTRACT

RELATIONSHIP BETWEEN METHYLMERCURY—INDUCED DISRUPTION OF
INTRACELLULAR CALCIUM AND NEURONAL DEATH

By

Tobi Leigh Limke

The mechanism by which the environmental neurotoxicant methylmercury
(MeHg) causes elevations of intracellular Ca2+ ([Ca2+]i) and subsequent neuronal death
was examined using single cell microfluorimetry of rat cerebellar granule neurons in
primary culture. In granule cells loaded with fura-2 to monitor changes in [Ca2+]i, 0.2-1.0
uM MeHg causes early release of Ca2+ from at least one intracellular source; the possible
sources examined were the smooth endoplasmic reticulum (SER) and the mitochondria.
The non-specific muscarinic acetylcholine (ACh) receptor antagonist atropine delayed
MeHg-induced elevations of [Ca2+]i, while down-regulation of the muscarinic receptors
and the inositol-1,4,5-triphosphate (1P3) receptor with 24 hr application of the muscarinic
agonist bethanechol decreased the amplitude of the MeHg-induced release of Ca2+i by
approximately 30-40%. Depletion of SER Ca2+ stores with thapsigargin also reduced the
amplitude of the MeHg-induced release of Ca2+i by approximately 30-40%. Inhibition of
Ca2+ release through the ryanodine receptors had minimal effect. Removal of
mitochondrial Ca2+ (Ca2+m) content prior to MeHg exposure using carbonyl cyanide m-
chlorophenylhydrazone (CCCP) and oligomycin decreased the amplitude of the MeHg-
induced release of Ca2+i by approximately 70%. Additionally, inhibition of the
mitochondrial permeability transition pore (MTP) using cyclosporin A (CsA) delayed the

increase in [Ca2+]i. In granule cells loaded with tetramethylrhodamine ethyl ester

(TMRE) to monitor changes in mitochondrial membrane potential, CsA delayed the
irreversible loss of membrane potential caused by 0.5 uM MeHg. In granule cells loaded
with rhod-2 to monitor changes in [CaZ+]m, MeHg caused an early increase in [Cazjm
followed several minutes later by release of dye from the mitochondria into the cytosol.
The initial increase in [Ca2+]m occurred independently of extracellular Ca2+, while the
release of dye from the mitochondria was delayed by prior application of thapsigargin or
CsA. Use of a calcein AM-ethidium homodimer cell viability assay revealed that
increasing concentrations of MeHg (0.2-1.0 uM) caused a corresponding increase in cell
death at 24 hr post-exposure. Atropine and ryanodine did not protect against MeHg-
induced cell death, while 24 hr BCh pretreatment significantly protected against cell
killing. The BCh-mediated protection was reversed by atropine and the M3 muscarinic
ACh receptor antagonist 4-diphenylacetoxyl-N-methylpiperidine methiodide (4-DAMP)
but not by the M2 receptor antagonist methoctramine or the nicotinic ACh receptor
antagonist dihydro-B-erythroidine hydrobromide (DHE). Thapsigargin itself was toxic,
highlighting the sensitivity of granule cells to disruption of SER Ca2+. CsA also provided
significant protection against cell death at 24 hr post-MeHg exposure. These results
suggest that MeHg: 1) acts at M3 muscarinic ACh receptors to cause production of 1P3,
which releases Ca2+ from the SER and 2) causes Ca 2+ uptake into mitochondria which is
then released into the cytosol via opening of the MTP. Additionally, these disruptions of
[Ca2+]; contribute to MeHg-induced neuronal death, and may underlie the specific

neurotoxicity of MeHg within granule neurons of the cerebellar cortex.

For the family and friends who never gave up hope, and passed that hope to me.

iv

ACKNOWLEDGMENTS

I would like to thank my parents, grandparents and extended family for their
support and encouragement through the insanity of the past 4% years. Their emotional
support and love will save me many years of therapy down the road.

I would like to thank Dr. William Atchison and my guidance committee for their
ideas and suggestions throughout my thesis, and their willingness to share their time to
serve on the committee. Their input was invaluable.

I would also like to thank my adopted family in the Atchison lab who were
always willing to teach me something new, help out when there was a problem, and
provided the most memorable conversations imaginable. Their patience and sense of
humor were appreciated every day, and will not be forgotten. This dissertation would not
have been possible without the excellent technical assistance of Erica Lange Fritz,
Aizhen Yao, Steve Lundback, Dr. Ravindra Hajela and Dawn Autio. In particular, the
amazing efforts of several undergraduates, including Scott A. Loiselle, Ilyana C.
Martinez, Bermily Maldonado, Maria Sanchez and Rebecca Hall, and one seriously
talented veterinary student, Jeremy J. Bearrs, allowed for more experiments to be
performed than should be humanly possible. Input and guidance from Dr. Sue Marty was

greatly appreciated.

PREFACE

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

Portions of Chapter 5 appeared as Limke and Atchison (2002).

vi

TABLE OF CONTENTS

LIST OF TABLES .................................................................................. x
LIST OF FIGURES ................................................................................ xi
LIST OF SYMBOLS AND ABBREVIATIONS ............................................ xvi
CHAPTER ONE: INTRODUCTION ........................................................... 1
A) GENERAL INTRODUCTION ....................................................... 2
B) BACKGROUND ........................................................................ 9
a. Humans and Methylmercury: Routes and Biological Consequences of
Exposure ........................................................................ 9
b. Cellular and Molecular Effects of Methylmercury ....................... 15
i. MeHg-Induced Cell Death: Apoptosis or Necrosis? ........... 16
ii. Calcium and Cell Death ............................................. 18
iii. MeHg and Cholinergic Neurotransmission ...................... 20
iv. Intracellular Calcium Stores ....................................... 23
v. MeHg and the Smooth Endoplasmic Reticulum ................ 29
vi. MeHg and Mitochondria ........................................... 33
vii. MeHg-Induced Interactions Between SER and Mitochondria
Ca2+ Stores ............................................................ 35
viii. Non-Ca2+ Divalent Cations ........................................ 36

C) RAT CEREBELLAR GRANULE NEURONS IN PRIMARY CULTURE AS
A MODEL FOR THE SELECTIVE NEUROTOXICITY OF MeHg .......... 37

vii

CHAPTER TWO: INTERACTION OF METHYLMERCURY WITH
MUSCARINIC RECEPTORS CAUSES CALCIUM RELEASE FROM THE
SMOOTH ENDPLASMIC RETICULUM INOSITOL-l,4,5-TRIPHOSPHATE

RECEPTORS IN RAT CEREBELLAR GRANULE NEURONS ....................... 40
A) ABSTRACT ............................................................................ 41
B) INTRODUCTION ..................................................................... 42
C) MATERIALS AND METHODS ................................................... 44
D) RESULTS .............................................................................. 49
E) DISCUSSION .......................................................................... 71

CHAPTER THREE: ACUTE METHYLMERCURY EXPOSURE CAUSES
NEURONAL DEATH IN RAT CEREBELLAR GRANULE CELLS THROUGH

INTERACTION WITH INOSITOL-l,4,5-TRIPHOSPHATE RECEPTORS ....... 76
A) ABSTRACT ........................................................................... 77
B) INTRODUCTION .................................................................... 78
C) MATERIALS AND METHODS ................................................... 81
D) RESULTS ............................................................................... 86
E) DISCUSSION .......................................................................... 98

CHAPTER FOUR: DISRUPTION OF MITOCHONDRIAL CALCIUM
SIGNIFICANTLY CONTRIBUTES TO METHYLMERCURY-INDUCED

ELEVATIONS OF [Ca2+], ...................................................................... 103
A) ABSTRACT ........................................................................... 104
B) INTRODUCTION ................................................................... 105
C) MATERIALS AND METHODS .................................................. 108
D) RESULTS ............................................................................. 111
E) DISCUSSION ........................................................................ 129

viii

CHAPTER FIVE: ACUTE EXPOSURE TO METHYLMERCURY OPENS THE
MITOCHONDRIAL PERMEABILITY TRANSITION PORE IN RAT

CEREBELAR GRANULE CELLS .......................................................... 134
A) ABSTRACT .......................................................................... 134
B) INTRODUCTION .................................................................... 136
C) MATERIALS AND METHODS .................................................. 139
D) RESULTS ............................................................................. 144
E) DISCUSSION ........................................................................ 159
CHAPTER SIX: SUMMARY AND CONCLUSIONS .................................. 165
A) SUMMARY AND CONCLUSIONS ............................................. 166
REFERENCES .................................................................................... 178

ix

LIST OF TABLES

Table 2.1 Effect of pharmacological inhibitors on peak fura-2 ratio induced by

1.5 min 1 mM BCh in cerebellar granule cells ................................ 51

Figure 1.1

Figure 1.2

Figure 1.3

Figure 1.4

Figure 2.1

Figure 2.2

Figure 2.3

Figure 2.4

Figure 2.5

Figure 2.6

LIST OF FIGURES

MeHg causes a biphasic increase in [Ca2+], in cerebellar granule cells in
primary culture ...................................................................... 5
Elevations of fuIa-2 fluorescence in a cerebellar granule cell in culture
exposed to 0.5 uM MeHg in CaZI-fi'ee buffer .................................. 6

Hypothesis regarding MeHg-induced elevations of {Cap} in rat cerebellar
granule neurons ..................................................................... 8
Elevations of fula-2 fluorescence in a cerebellar granule cell in culture
exposed to 0.5 uM MeHg in Cari-flee buffer ................................. 14

Alternating exposures to 40 mM K+ and 1 mM BCh cause increases in
[Ca2+], in cerebellar granule cells ............................................... 50

Thapsigargin does not affect the time-to-onset of MeHg-induced increases
in [Ca2+], ............................................................................ 52
Thapsigargin does not affect the time-to-onset of MeHg-induced [Ca2+]i
increases in the absence of Ca2+e ................................................ 53

Thapsigargin decreases the amplitude of MeHg-induced fura-2 ratio
increases, as measured in the absence of Caz: ................................ 54
Pretreatment with BCh hastens the MeHg-induced release of [Ca2+],
without affecting the onset of MeHg-induced influx of Ca2+: ............... 56
Pretreatment with BCh hastens the MeHg-induced release of [Ca2 i, as

measured in the absence of Ca2+c ................................................ 57

xi

Figure 2.7

Figure 2.8

Figure 2.9

Figure 2.10

Figure 2.11

Figure 2.12

Figure 2.13

Figure 2.14

Figure 2.15

Figure 2.16

Figure 3.1

Figure 3.2

Figure 3.3

Pretreatment with BCh (1 mM, 24 hr) does not increase the [Ca2+] within
mitochondria ....................................................................... 58
Pretreatment with BCh decreases the amplitude of the fura-2 ratio increase
caused by MeHg, as measured in the absence of Ca”... ...................... 59
Atropine (“AP”) delays the first-phase release of Car} without affecting
the second-phase influx of Ca2+e caused by MeHg ............................ 61
Atropine (“AP) delays the time-to-onset of the MeHg-induced first-phase
release of Ca2+;, as measured in the absence of Ca2+¢ ........................ 62
Atropine (“AP”) decreases the amplitude of the fura-2 ratio increase
caused by MeHg, as measured in the absence of Ca”c ...................... 63
Ryanodine (“Ry”) does not affect the time-to-onset of MeHg-induced
increases in [Ca2+], ................................................................ 66
Ryanodine (“Ry”) does not affect the time-tO-onset of the MeHg-induced
first-phase release of Ca2+i, as measured in the absence of Ca2+e ........... 67
Ryanodine does not affect the amplitude of MeHg-induced increases in
[Ca2+], as measured in the absence of Ca2+e .................................... 68
Depletion of SER Ca2+ triggers SOCE in rat cerebellar granule cells. . ....69
SK&F 96365 inhibits depolarization-stimulated Ca2+ influx through
VOCCS .............................................................................. 7O
MeHg-induced cell death at 30 min and 24 hr post-exposure ............... 87
24 hr pretreatment with 1 mM BCh protects against MeHg-induced cell
death ................................................................................. 89

Atropine plus BCh does not protect against MeHg-induced cell death. . ..90

xii

Figure 3.4

Figure 3.5

Figure 3.6

Figure 3.7

Figure 3.8

Figure 4.1

Figure 4.2

Figure 4.3

Figure 4.4

Figure 4.5

Figure 4.6

Figure 4.7

Figure 4.3

4-DAMP plus BCh does not protect against MeHg-induced cell death. . .92
The nicotinic receptor antagonist DHE does not affect BCh-mediated
protection against MeHg-induced cell death .................................. 93

Atropine alone does not protect against MeHg-induced cell death. . . . . ....94
Xestospongin C does not protect against MeHg-induced cell death ....... 95

Ryanodine does not protect against MeHg-induced cell death .............. 97
CCCP and oligomycin cause a rapid decrease in TMRE fluorescence with
a concomitant increase in fura—2 fluorescence ............................... 112
Effect of oligomycin on TMRE fluorescence, as well as effect of CCCP on
TMRE and fura—2 fluorescence ................................................ 113
Effect of mitochondrial inhibitors on first-phase increase in [Ca2+]i caused
by 0.5 11M MeHg ................................................................ 115

Effect of CCCP and oligomycin on first-phase time-tO-onset and ratio
amplitude induced by 0.5 11M MeHg in the absence of Ca2+: ............. 116

Comparison of time-to—onset of Ca2+c influx caused by 0.5 11M MeHg in
the presence or absence of mitochondrial inhibitors ........................ 117
Changes in fura-2 and rhod-2 fluorescence in response to increasing
concentrations of extracellular K+ ............................................. 120

MeHg causes a biphasic increase in rhod-2 fluorescence in granule cells.
...................................................................................... 122
Effect of CCCP and oligomycin on MeHg-induced elevations of rhod-2

fluorescence ...................................................................... 123

xiii

Figure 4.9

Figure 4.10

Figure 4.11

Figure 4.12

Figure 5.1

Figure 5.2

Figure 5.3

Figure 5.4

Figure 5.5

Figure 5.6

Figure 5.7

Figure 5.8

Mitochondrial membrane depolarization abolished the initial elevation of
rhod-2 fluorescence without affecting the time-to-onset of the Spreading of
rhOd-2 throughout the cytosol .................................................. 124
Thapsigargin delays the spreading of rhod-2 from punctate to whole-cell
staining as caused by 0.5 11M MeHg .......................................... 126
CSA delays the spreading of rhod-2 from punctate to whole-cell staining
as caused by 0.5 11M MeHg .................................................... 127

Effect of CsA and thapsigargin on interval between onset of first and
second rhOd-2 fluorescence increases ......................................... 128
Increasing concentrations of CSA cause increasing delays in the time-to-
onset of [Ca2+]: elevations caused by 0.5 11M MeHg ........................ 145
5 11M CSA significantly delays the time-to-onset of MeHg-induced [Cay],
increases ........................................................................... 146
5 11M CsA delays the time-to-onset of MeHg-induced release of Ca2+i in
the absence of Ca2+c ............................................................. 149

Effect of calcineurin inhibitors on MeHg-induced [Ca2+], elevations. ...150

PhAsO causes a rapid decrease in TMRE fluorescence .................... 151

Exposure to 0.5 11M MeHg causes irreversible loss of TMRE
fluorescence ...................................................................... 152

CSA, but not CN inhibitors, delay the time-tO-onset of mitochondrial
membrane depolarization caused by 0.5 11M MeHg ........................ 153

CSA provides limited protection against MeHg-induced cell death at 24 hr

post-exposure ..................................................................... 156

xiv

Figure 5.9

Figure 5.10

Figure 6.1

Comparison of protection against MeHg-induced cell death by
pretreatment or Simultaneous treatment with 5 11M CSA ................... 157
Pretreatment with 20 11M W7 significantly protects against cell death
caused by 0.5 11M MeHg at 24 hr post-exposure ............................ 158

Summary of findings ............................................................ 167

XV

LIST OF SYMBOLS AND ABBREVIATIONS

‘I’m — inner mitochondrial membrane potential

ACh — acetylcholine

AMPA - alpha-aminO-3-hydroxy-5-methyl-5-isoxazole propionate

AN OVA - analysis of variance

Ara—C — cytosine-B—arabinofilranoside

ATP -— adenosine 5'-triphosphate

BAPTA - 1,2- bis(2-aminophenoxy)ethane-N,N,N',N'-tetraacetic acid tetrakis
(acetoxymethyl)ester

BCh — carbamyl-B-methylcholine chloride (bethanechol)

Calcein AM — calcein acetoxymethylester

[Ca2+]c — extracellular Ca2+ concentration

{Cay} — intracellular Ca” concentration

Ca2+... — Ca2+ contained within the mitochondrial lumen

CCh - carbamylcholine

CCCP - carbonyl cyanide m-chlorophenylhydrazone

CICR — Ca2+-induced Ca2+ release

CMF-HBSS — CaZI-Mg2+ fiee Hank’s Buffered Saline Solution

CN — calcineurin (protein phosphatase ZB)

CNS - central nervous system

CsA — cyclosporin A

4-DAMP - 4-dipheny1acetoxyl-N-methylpiperidine methiodide

xvi

DHE — dihydro-B—erythroidine hydrobromide
DMEM — Dulbecco’s Modified Eagle Medium
DNase I — deoxyribonuclease I

EGTA — ethyleneglycol bis-(B-aminoethyl ether)-N,N,N',N',-tetraacetic acid
EtOH -— ethanol

FBS — fetal bovine serum

F ura-2 AM - fura-2 acetoxymethylester

GABA — gamma-aminobutyric acid

GSH - glutathione

HBS — HEPES-buffered saline solution

HEPES — N-2-hydroxyethylpiperazine-N'-2-ethanesulfonic acid
hr — hour(s)

1P3 — inositOl-l,4,5-triphosphaste

MeHg — methylmercury

min — minute(s)

MTP — mitochondrial permeability transition pore
NMDA - N-methyl-D-aspartate

PhAsO — phenylarsine oxide

FLA; - phospholipase A2

PLC — phospholipase C

PNS — peripheral nervous system

ROS - reactive oxygen species

SEM — standard error of the mean

xvii

SER — smooth endoplasmic reticulum

SERCA — smooth endoplasmic reticulum Ca2+ ATPase

SOCC — store—operated Ca2+ channel

SOCE — store-operated Ca2+ entry

TMRE — tetramethylrhodamine ethyl ester

TPEN — N,N,N‘,N‘-tetrakis-(2-pyridy1methyl)ethylenediamine
W7 — N-(6-aminohexyl)-5-chloro—1-napthalenesu1fonamide

XeC - xestospongin C

xviii

CHAPTER ONE

INTRODUCTION

A) GENERAL INTRODUCTION

Methylmercury (MeHg) is a highly lipophilic environmental contaminant which
easily crosses both the blood-brain barrier and placental barrier, resulting in
neurotoxicity in adults and fetuses alike (Friberg and Mottet, 1989; Aschner and
Aschner, 1990; Myers and Davidson, 1998). Despite its ability to bind thiols
indiscriminately with high affinity (Roberts et al., 1980), making any protein or peptide
with a cysteine group a potential target, MeHg causes a surprisingly Specific pattern of
neuronal damage in both the central and peripheral nervous systems (CNS and PNS,
respectively). Over the past few decades, there has been considerable effort to elucidate
the specific mechanism(s) by which MeHg causes neurotoxicity within discrete areas of
the CNS by focusing on specific molecular targets of the compound. By identifying the
molecular targets of MeHg within the CNS, particularly proximate targets in cells which
exhibit heightened sensitivity, it is hoped that more reliable risk assessment can be
performed in which results from in vitro assays can be correlated with in vivo effects
observed at environmentally relevant exposure levels.

The primary pathological result of both acute and chronic MeHg poisoning is
neurotoxicity, with clinical signs ranging from ataxia and visual disturbances to paralysis
and death (Rustam et al., 1975; Chang, 1980). Several areas of the brain exhibit marked
pathological damage upon MeHg exposure, including the cerebral and cerebellar cortices
(Hunter and Russell, 1954; Leyshon—Sorland et al. , 1994). In both humans and rats, the
granule cell layer of the cerebellum is particularly sensitive to MeHg-induced cell death
(Hunter and Russell, 1954) through an as yet undefined mechanism. In contrast, the

neighboring Purkinje cells are less sensitive to MeHg-induced cell death despite

accumulating equal or greater MeHg concentrations (Hunter and Russell, 1954). The
mechanism underlying the heightened susceptibility of these neurons is not yet known,
but is hypothesized to be due to interactions with specific molecular targets which have
not yet been identified (Castoldi et al., 1996; Robertson and Orrenius, 2000; Castoldi et
al. , 2001).

Given the large number of potential molecular targets containing cysteine
moieties, it is not surprising that MeHg has a number of effects which could contribute to
the observed toxicity. At the cellular level, MeHg interacts with a number of Sites to
cause disruption of divalent cation homeostasis, particularly Ca2+ and Zn2+ (Denny et a1. ,
1993; Hare et al. , 1993), elevation of spontaneous release of neurotransmitter (Juang,
1976; Atchison and Narahashi, 1982; Atchison, 1986), generation of reactive oxygen
species (Sarafian and Verity, 1991; Yee and Choi, 1994) and cell death (Kunimoto, 1994;
Marty and Atchison, 1998). The disruption of intracellular Ca2+ (Ca2+i) has received
much attention in recent years, with both plasma membrane Ca2+ channels and
intracellular Ca2+ stores being investigated as potentially critical targets during MeHg
exposure (Tan et al., 1993; Oyama et al., 1994; Sakamoto et al., 1996; Marty and
Atchison, 1997; Marty and Atchison, 1998; Sirois and Atchison, 2000). Cell viability
assays in which Ca2+; was buffered by the Ca2+i chelator 1,2- bis(2-
aminophenoxy)ethane-N,N,N',N'—tetraacetic acid tetrakis (acetoxymethyl)ester (BAPTA)
indicated that removal of Ca2+: provides significant protection against neuronal death at 3
hr post-exposure, suggesting an important role for loss of Ca2+ homeostasis in MeHg-

induced neurotoxicity (Marty and Atchison, 1998).

In the highly susceptible cerebellar granule cell, MeHg affects multiple subtypes
of voltage-gated Ca2+ channels on the plasma membrane (Marty and Atchison, 1997;
Sirois and Atchison, 2000). However, very little is known about the intracellular Ca2+
targets in this cell type. Previous studies demonstrated that MeHg disrupts intracellular
Ca2+ (Ca2+i) homeostasis in cultures of rat cerebellar granule cells, resulting in a biphasic
increase in [Ca2+]i: the first phase results from release of Ca2+ from intracellular pools,
and the second phase results from influx of extracellular Ca2+ (C322) (Hare et al., 1993;
Marty and Atchison, 1997) (Figure 1.1). The initial increase in [Caz+]t is intracellular in
origin, as removal of Ca2+c abolishes the secondary fura-2 fluorescence elevation (Hare et
al., 1993; Marty and Atchison, 1997) (Figure 1.2). In cerebellar granule cells, the MeHg-
induced influx of Ca2+c occurs through nifedipine- and w-conotoxin—MVIIC-sensitive
pathways, suggesting that L, N and/or P/Q-type voltage-gated Ca2+ channels participate
in MeHg-induced Ca2+ entry; however, the intracellular Ca2+ source has not yet been
identified (Marty and Atchison, 1997). Two intracellular Ca2+ stores could contribute to
the early-onset elevation of [Cay], during MeHg exposure: the smooth endoplasmic
reticulum (SER) and the mitochondria. The SER provides high-affinity, low-capacity
Ca2+ storage while the mitochondria are generally considered to be a low-affinity, high-
capacity pool (Fohrman et al., 1993; Gunter and Gunter, 1994; Simpson et al., 1995;
Budd and Nicholls, 1996b; Masgrau et al. , 2000). The SER contains ryanodine- and IP3-
sensitive Ca2+ pools which can release sequestered Ca2+ into the cytosol (Irving et al.,
1992). In NGlO8—15 cells, part of the MeHg-induced first-phase is due to release of Ca2+

from the IP3-sensitive pool in the SER; however, emptying

0.5 1.1M MeHg
3 ’ ' :

E First ; Second
:Phase : Phase

 

x .
+

 

----
p-----—--

l l l i l

 

0 l l n l l. l i
0 10 20 30 40 50 60

Time (min)

Figure 1.1. MeHg causes a biphasic increase in [Ca2+]i in cerebellar granule cells in
primary culture. Representative tracing of the fura-2 fluorescence fiom a cerebellar
granule cell exposed to 40 mM K.+ for 1 min (as a viability test), followed by HBS
washout and 0.5 11M MeHg. The ratio of fluorescence emitted at excitation wavelengths
of 340 nm and 380 nm gives the approximate level of [Ca2+]i, which increases in a
biphasic manner. Time-tO-onset of fluorescence increases is measured from the
beginning of MeHg treatment.

Modified from: Marty and Atchison, Toxicol. Appl. Pharmacol. 147 :319-3 30 (1997).

0.5 uM MeHg-EGTA

 

First Phase

   

l-I-c- --------------

 

i l l l l

40 50 60

 

O
y—a
O
N
O
DJ
0

Time (min)

Figure 1.2. Elevations of fun-2 fluorescence in a cerebellar granule cell in culture
exposed to 0.5 M MeHg in Ca2+-free buffer. Under these conditions, the first [Ca2+]:
phase is evident while the second phase (due to influx of [Ca2+].,) is absent.

Modified from: Marty and Atchison, Toxicol. Appl. Pharmacol. 147:319-330 (1997).

this pool does not eliminate the first phase, suggesting that at least one additional source
contributes to the first phase Ca2+ increase (Hare and Atchison, 1995). Mitochondria also
store and release Ca2+, and are also affected by MeHg. Ca2+ release can occur via
reversal of the uniporter, activation of the Na‘I/Ca2+ exchanger, or opening of the
mitochondrial permeability transition pore (MTP). MeHg has a number of effects on
mitochondria, including decreased Ca2+ uptake and increased Ca2+ efflux from preloaded
rat brain mitochondria, as well as inhibition of respiration, increased permeability of the
inner membrane, and depolarization of the inner mitochondrial membrane (Sone et al. ,
1977; Levesque and Atchison, 1991; Hare and Atchison, 1992), any of which could
contribute to the first phase Ca2+ elevation via disruption of mitochondrial Ca2+ buffering.
Several lines of evidence suggest a critical relationship between intracellular Ca2+
pools and cell death, thus necessitating inquiry in this area to gain filll understanding of
neuronal death induced by MeHg. AS such, the goal of this dissertation is to identify the
intracellular Ca2+ pools targeted by MeHg in cerebellar granule neurons and determine
their relative importance in MeHg-induced cytotoxicity of cerebellar granule neurons.
This dissertation reports results of experiments designed to investigate the hypothesis that
MeHg initiates Ca2+ release fi'om the SER which is buffered by mitochondria; however,
MeHg impairs the ability of mitochondria to hold Ca2+ (by causing loss of mitochondrial
membrane potential and inhibiting respiration), leading to opening of the mitochondrial

permeability transition pore and cell death (Figure 1.3).

MeHg

  

Ryanodine
Receptors

 

Figure 1.3. Hypothesis regarding MeHg-induced elevations of [Ca2+]i in rat
cerebellar granule neurons. MeHg binds M3 muscarinic ACh receptors, causing
production of 1P3 and release of CaZ+ through the 1P3 receptors. This elevation of [Ca2+],
triggers Ca2+-induced Ca2+ release through the ryanodine receptors. Ca2+ released from
the SER is buffered by mitochondria. Excess uptake of Ca2+ into the mitochondria,
combined with direct effects of MeHg on the mitochondrial respiratory chain, leads to
opening of the mitochondrial permeability transition pore (MTP) and promotion of cell
death.

B) BACKGROUND

a) Humans and MeHg: Routes and Biological Consequences of Exposure
MeHg is an organic form of mercury which enters the environment through two primary
routes: release of MeHg itselfas a by-product of human activities, and conversion of
inorganic mercury to MeHg by biological organisms. Organic mercury has been used as
a fungicide and in preservative solutions for wood, paper pulp and leather. Additionally,
inorganic mercury has been used as a catalyst in several industrial processes, allowing for
the release of both inorganic and organic mercury (as a by-product of the reaction) into
the environment through discharge of untreated waste into the water and air (Kurland,
1973; Gerstner and Huff, 1977). Organic mercury is also produced by methylation of
mercury by aquatic microorganisms in the sediment of river and lake beds (Choi and
Bartha, 1994; Baldi, 1997; Guimaraes et al., 2000). MeHg is a highly lipophilic
substance which, once in the environment, readily accumulates in the food chain
(Rabenstein and Evans, 1978). The most common mode of MeHg intoxication for
humans is through ingestion of contaminated foodstuffs (Hansen, 1990; Renzoni et al.,
1998). Once ingested, MeHg readily enters the blood and is transported in the plasma
throughout the body, where it easily crosses the blood-brain barrier (Steinwall and
Klatzo, 1966; Chang and Hartrnann, 1972; Chang and Hartmann, 1972; Aschner and
Aschner, 1990). In rats exposed to radioactively labeled MeHg, the greatest
accumulation of the compound occurs in the spinal dorsal root ganglia, with the cerebral
cortex and cerebellum also exhibiting high concentrations following exposure (Somjen et

al. , 1973). MeHg is excreted from the body at a slow rate, with a half-life of

approximately 70 days in humans (Clarkson, 1972; Neathery and Miller, 1975). While
the rate of MeHg elimination via metabolism and excretion is sufficient that
accumulation within the body is limited (Clarkson, 1972), there is evidence suggesting
that the brain half-life is considerably longer than that of the whole body during chronic
exposure (Rice, 1989; Carrier et al. , 2001).

MeHg poisoning was first examined in the 19508 following the study of four
patients poisoned via inhalation of MeHg from fungicidal dusts in a cereal factory. The
predominant symptoms in these patients were neurological, including generalized gross
ataxia, impaired speech, and severely constricted visual fields. Histological analysis of
one of the four patients in autopsy revealed gross atrophy of the cerebrum as well as
marked degeneration of the granule cell layer of the cerebellum and neuronal destruction
in the visual area of the calcarine (visual) cortex (Hunter and Russell, 1954). The authors
noted that the observed ataxia was probably due to gross alterations in the cerebellar
cortex because there was no damage to the cerebellar motor pathways other than the loss
of granule cells. The cerebellar damage itself was very selective, with loss of granule
cells occurring in the absence of damage to nearby Purkinje cells.

The two most widely known incidents of MeHg poisoning occurred in Japan in
the 1950s and 19608 (Takeuchi et al., 1962; Harada, 1995) and Iraq in 1972 (Bakir et
al., 1973; Clarkson er al., 1976). There were epidemics in Japan in both Minamata
(from 1953 to 1960) and Niigata (from 1960 to 1965) as a result of eating contaminated
seafood (Chang et al., 1977). In Minamata, both organic and inorganic mercury were
part of a chemical plant’s waste effluent discharged into Minamata Bay. Once in the

waterway, MeHg accumulated in the food chain, resulting in 46 deaths and

10

approximately 2200 people poisoned through consumption of contaminated seafood
(Harada, 1995). In addition, prenatal exposure to MeHg resulted in many infants being
born with nervous system damage. This fetal neurological syndrome is now referred to
as “Minamata syndrome.” During the Iraqi poisoning incident, approximately 6500
hospitalizations and 450 deaths were reported as a result of MeHg poisoning from
ingestion of contaminated seed grain in bread (Clarkson et al., 1976). In both the
Minimata and Iraqi incidents, the primary sign of exposure was neurological dysfunction
as characterized by cerebellar ataxia, generalized extremity weakness and sensory
disturbances including impairments of speech, vision and hearing. Correlations between
body burden of MeHg and fiequency of adverse symptoms (ataxia, paresthesia, etc.) in
adult victims of the Iraqi poisoning indicated that the threshold for adverse effects was at
a body burden level of 25 to 40 mg of mercury (Bakir et al., 1973). Additionally, Iraqi
children exposed in utero, through breastfeeding, or both, exhibited marked
developmental deficiencies, including delayed onset of speech and impaired motor,
sensory and autonomic function (Bakir et al. , 1980). In the more severe cases, children
were born blind and deaf (Bakir et al., 1980). Dose-response analysis suggests that fetal
effects are observed when maternal hair mercury concentrations exceed 10 ppm (Crump
et al., 2000).

In adults, MeHg causes a very specific pattern of damage within the PNS and
CNS that correlates with the observed symptoms of poisoning (Chang and Hartrnann,
1972; Chang et al., 1977). In the PNS, there is axonal degeneration characterized by
secondary loss of the myelin sheath of the sensory branch (Hunter and Russell, 1954;

Takeuchi, 1982). The disruption of motor and sensory nerve function correlates with the

11

observed extremity weakness and impairment of sensory function (Chang, 1980). Early
studies focused on effects of MeHg at the neuromuscular junction, in part due to the
observed increased incidence of neuromuscular weakness during the Iraqi incident
(Rustam et al., 1975) and in part because the ACh release process is well characterized
both biochemically and physiologically. At the neuromuscular junction, the primary
effect is to decrease nerve-evoked release of acetylcholine (ACh) and alter its
spontaneous quantal release by causing it to first increase, then decrease (J uang, 1976;
Atchison and Narahashi, 1982). Interestingly, the elevated spontaneous release of
neurotransmitter may be due to increased [Ca2+], because MeHg increases [Cay], in the
absence of Ca2+,- in isolated nerve terminals (Komulainen and Bondy, 1987). Further,
MeHg increases the frequency of spontaneous ACh release in the absence of Ca2+c in the
neuromuscular junction (Atchison, 1986).

While MeHg can damage the PNS, its effects within the CNS are of greater
concern, especially in the developing organism. During chronic MeHg exposure, MeHg
accumulates in the cerebellum, with the highest concentration in the Purkinje and Golgi
cells of the granular layer, and a Slightly lower concentration in the granule cells, stellate
cells, and basket cells (Chang et al., 1977; Moller-Madsen, 1990; Moller-Madsen, 1991)
(Leyshon-Sorland et al. , 1994) (Figure 1.4). However, despite the lower accumulation
of MeHg in the granule cell layer, these neurons are most affected during both acute and
chronic MeHg poisoning, resulting in a characteristic loss of the granule cell layer
(Hunter and Russell, 1954; Chang et al., 1977; Leyshon and Morgan, 1991). The
resultant cerebellar ataxia observed in both acute and chronic MeHg poisoning is

attributed to the loss of these neurons. Additionally, MeHg causes specific loss of

12

neurons from the second through fourth layers of the visual area of the calcarine cortex,
resulting in constriction of the visual field (Hunter and Russell, 1954; Takeuchi, 1982).

The fetal CNS exhibits greater sensitivity to MeHg than does the adult CNS, as
children exposed in utero may be affected in the absence of maternal toxicity (Bakir et
al., 1980; Takeuchi, 1982). Unlike the toxicity in adults, fetal poisoning produces a less
specific pattern of damage that is dependent on the concentration and duration of
exposure, as well as the stage of development during exposure (Rodier, 1995). Prenatal
exposure interferes with neuron migration (Choi et al. , 1978) and decreases the mitotic
activity of CNS neurons (Rodier et al. , 1984). Following the Iraqi incident, infants born
to highly exposed mothers demonstrated permanent damage to the CNS, with symptoms
including delayed achievement of developmental milestones, seizures and abnormal
reflexes (Marsh et al., 1987). In several cases involving Japanese children, there was a
marked decrease in the number of neurons in all lobes of the cereme cortex, as well as
widespread disorganization of the cerebral cortex architecture resulting from neuronal
degeneration (Choi, 1989). Similar to the adult symptoms, severely affected children
exhibit severe atrophy of the cerebellar cortex such that the granule cell layer
degenerates and virtually disappears (Choi, 1989). Unlike the adult cases, in which the
remaining neurons of the cerebellar cortex are relatively unaffected, children exhibit a
narrowing of the cortical molecular layer and a reduction in the number of Purkinje
neurons (Choi, 1989).

More recent studies have focused on the more subtle neurological effects resulting
fiom low-concentration exposure to MeHg in the developing organism. Currently, the

most common source of MeHg exposure is consumption of contaminated seafood

13

  
  

J ‘0 [
'Efi‘lb‘ui"
‘F‘zi'. €11.75;
"if‘fifil‘?
\ .\ Q's '

80,
BC /

2
CF

    

 

   

 

  
 

 

 

 

 

 

Seams:

Figure 1.4. Schematic diagram of cerebellar cortex. The cerebellar cortex is divided
into three distinct layers which lie outside the white matter (WM); these layers are the
outermost molecular layer (ML), the Purkinje layer (PL), and the innermost granular
layer (GL). The granular layer contains not only the cell bodies of granule cells (Gr) but
also contains Golgi cell (Gol) bodies, axons and dendrites, mossy fiber (MF) terminals,
and the axons of Purkinje cells with their accompanying climbing fibers (CF). The
Purkinje layer contains only the cell bodies of Purkinje cells (PC). The molecular layer
consists of the stellate cells (SC) and basket cells (BC), as well as the dendritic arbors of
Purkinje cells and Golgi cells, and the parallel fiber (PF) branches of granule cell axons.
Diagram courtesy of Dr. Yukun Yuan.

l4

(Hansen, 1990; Renzoni et al., 1998); thus, MeHg exposure is of concern in regions of
the world where indigenous populations heavily consume fish as part of their diet. At-
risk populations include those in the Seychelles (Myers et al., 1995; Myers et al., 1997;
Palumbo et al., 2000) and Faroe Islands (Weihe et al., 1996; Grandjean et al., 1999), as
well as Greenland’s Inuits (Hansen, 1990) and certain tribes of North American Native
Americans (Dellinger et al., 1996). While there were no adverse effects observed in a
longitudinal study of children in the Seychelles Islands (Davidson et al., 2000), studies in
the Faroe Islands (Grandjean et al., 1999; Steuerwald et al., 2000) and Brazil (Grandjean
et al., 1999) found that long-term exposure to low concentrations of MeHg was
correlated with deficiencies in motor fimction, attention, memory, and language
development. The growing recognition that the greatest threat is to the developing
organism is reflected in the recent lowering of the US EPA reference dose fiom 0.3 to 0.1
rig/kg per day (Rice et al. , 2000). The lower reference dose is derived from a study of
children exposed in utero during the Iraqi poisoning, in which increased maternal MeHg
exposure was associated with delayed neurodevelopment in the child (Marsh et al. ,
1987). However, there is still considerable debate over the lowest exposure needed to

produce adverse effects in the fetus.

b) Cellular and Molecular Effects of Methylmercury

One of the issues complicating the risk assessment of MeHg is the large number
of potential molecular targets within the body. Because MeHg binds thiol groups with
high affinity, any protein or peptide containing a cysteine is a potential target. Given the

wide range of possible targets during MeHg exposure, the question arises whether there is

15

a Specific target, or set of targets, which make specific areas of the CNS vulnerable to
MeHg-induced disruption of neuronal development and viability. This dissertation will
address the question of whether MeHg targets specific receptors and ion channels
involved in Ca2+ signaling in the CNS, including plasma membrane Cholinergic receptors
and intracellular Ca2+ channels associated with Cay-sequestering organelles, which may

be involved in the pathogenesis of MeHg.

i) MeHg-Induced Cell Death: Apoptosis or necrosis?

To understand whether MeHg has specific targets during neurotoxicity, it is
critical to understand the mechanisms by which MeHg causes cell death. The cell death
process can be divided into two morphologically and biochemically distinguishable
categories: necrosis and apoptosis. Necrosis is generally characterized by cell swelling,
organelle damage, loss of plasma membrane integrity and cell lysis, as well as rapid
adenosine 5'-t1iphophate (ATP) depletion and severe alteration of mitochondrial
ultrastructure, such that an inflammatory response is initiated. Apoptosis, on the other
hand, is characterized by cell shrinkage, chromatin aggregation, plasma membrane
blebbing, and DNA condensation accompanied by maintenance of ATP levels, activation
of caspases, and loss of plasma membrane asymmetry (Kroemer, 1997; Mattson, 2000).
However, the distinction between apoptosis and necrosis is not clear-cut, as both share
several critical characteristics. For example, both modes of cell death are accompanied
by elevated [Ca2+]; levels, and each can result fiom exposure to the same compound
depending on the severity of the insult (Kruman and Mattson, 1999). Removal of Ca2+:

with buffering agents (such as BAPTA) can protect cells from cell death in several

16

different systems, including neurons (Marty and Atchison, 1998). Both mechanisms also
involve plasma membrane blebbing and mitochondrial damage, making the distinction
between necrosis and apoptosis diffith in some cases. There is also evidence that the
antiapoptotic gene bcl-2 can protect cells from both apoptosis and necrosis (Kane et al. ,
1995; Kroemer, 1997). Indeed, although necrosis and apoptosis are considered to be
different forms of cell death, there is now debate over whether they are actually part of a
continuum of overlapping mechanisms of cell death (Rafliay and Cohen, 1997).

In MeHg neurotoxicity, there is increasing evidence both in vivo and in vitro that
MeHg causes apoptosis at lower concentrations and necrosis at higher concentrations
(Nakada and Imura, 1983; Miura and Imura, 1987; Kunimoto, 1994; Nagashima et al.,
1996). When studied in vivo, the cerebellar damage tends to be apoptotic, as assessed
using conventional techniques such as the terminal deoxynucleotidyl transferase-
mediated dUTP nick end labeling (TUNEL) assay, DNA gel electrophoresis and
ultrastructural examination (Nagashima et al., 1996). When studied in vitro, cerebellar
granule cells undergo apoptosis following 18 hr of exposure to lower MeHg
concentrations (0.5-1.0 11M), and necrosis within 1 hr of exposure to higher
concentrations (5 and 10 11M) (Castoldi et al., 2000). When granule cells in primary
culture are studied after 72 hr of exposure, apoptosis occurs at 0.1-0.3 11M, with the
switch to necrosis occurring at higher MeHg concentrations (Kunimoto, 1994). Hence,
the mode of cell death in vitro depends on both the duration and intensity of exposure.
However, there is no clear-cut threshold for which death pathway is followed, as both

modes of cell death can be observed in granule cells in primary culture in response to 1

l7

11M MeHg within an hour of exposure, as determined using a combination of fluorescent

microscopy and morphometric analysis, and the TUNEL assay (Castoldi er al., 2000).

ii) Calcium and Cell Death

The central role of Ca2+ in the cell death process is a common thread between a
wide range of neurotoxic insults, including the neurotoxicity of lead and organic mercury,
as well as glutamate excitotoxicity (White and Reynolds, 1996). Neurons contain a
variety of mechanisms for transporting Ca2+ across both internal and external membranes.
The concentration of Ca2+: is tightly regulated to maintain a relatively low [Ca2+]; (in the
100-200 nm range) compared to the [Ca2+]c (in the 1—2 mM range) (Kass and Orrenius,
1999). When the [Ca2+]; becomes elevated for sustained periods of time, the
physiological signal becomes pathological as Ca2+ begins to promote cell death.
Disruption of Ca2+; homeostasis results in activation of multiple enzymes which are
involved in the catabolism of proteins, phospholipids and nucleic acids, thereby leading
to degradation of molecules needed for cell survival (Kass and Orrenius, 1999). For
example, calpains are Ca2+-activated cytosolic proteases which, when activated, are
capable of promoting cell death, as indicated by the effectiveness of calpain inhibitors in
preventing Ca2+-mediated cell death (Chan and Mattson, 1999). Several phosholipases,
including PLC and phospholipase A2 (PLAZ) are also activated by Ca2+ either directly
(PLC) or indirectly through CaZI-calmodulin (PLAz). PLC generates 1P3 which releases
Ca2+; from the SER, while PLA; breaks down phospholipids to form arachidonic acid
(Exton, 1990). Sustained elevations of Ca2+, can lead to sustained activation of PLA;,

thereby leading to extensive membrane degradation and the generation of toxic

18

metabolites (Exton, 1990). Further, there are Ca2+-activated endonucleases which
mediate DNA strand cleavage at internucelosomal linker regions to produce
approximately 200 base pair long fragments (Arends et al., 1990). Although influx of
Ca2+c is often associated with apoptosis, it is not necessarily an essential requirement as
elevated Ca2+ originating from intracellular stores can be sufficient to induce cell death
(Wei and Perry, 1996; Wei et al., 1998; Szalai et al., 1999). However, there is also
evidence for a lack of correlation between changes in [Cam]: and apoptosis caused by
specific insults, most notably in thymocytes and lymphocytes (Alnemri and Litwack,
1990; Beaver and Waring, 1994).

MeHg causes elevated [Ca2+]; in a number of cell types and preparations,
including neuroblastoma-glioma cells (Hare et al., 1993), human lymphocytes and
monocytes (Shenker et al. , 1992; Shenker et al., 1993; Tan et al. , 1993), mouse
thymocytes (Oyama et al., 1995), rat forebrain synaptosomes (Denny et al., 1993) and rat
cerebellar neurons (Sarafian, 1993; Oyama et al. , 1994; Marty and Atchison, 1997;
Mundy and Freudenrich, 2000). In cerebellar granule cells in primary culture, inhibition
of MeHg-induced Ca2+c influx through voltage-operated Ca2+ channels with the Ca2+
channel blockers co-conotoxin MVIIC and nifedipine reduced the number of dead
neurons at 3.5 hr post-exposure (Marty and Atchison, 1998). Additionally, chelating
Ca2+i with BAPT A protected against cell death at 3.5 hr post—exposure to 0.5-1.0 11M
MeHg; however, BAPT A was not protective at 24.5 hr post-exposure (Marty and
Atchison, 1998). In granule cells in vitro, MeHg-induced cell death is not due to
excitotoxic insult, as inhibitors of excitatory amino acid channels do not prevent MeHg-

induced elevations of [Ca2+]: (Marty and Atchison, 1997) or MeHg-induced cell death

19

(Marty and Atchison, 1998; Castoldi et al., 2000). In vivo studies support the in vitro
experiments, as voltage operated Ca2+ channel blockers prevented the appearance of
neurological disorders and mortality in rats treated with MeHg (Sakamoto et al., 1996).
Thus, MeHg neurotoxicity has a Ca2+-dependent component which, when inhibited,
prevents some of the toxic effects of the compounds, including cell death.

MeHg potentially causes elevations of [Ca2+]; through non-specific damage to
lipid membranes on intracellular organelles and/or the plasma membrane, or through
specific interactions with receptors and/or ion channels associated with Ca2+ signaling.
MeHg causes non-specific membrane damage through lipoperoxidation in cerebellar
granule cells in suspension (Sarafian and Verity, 1991); however, this membrane damage
does not appear to contribute to MeHg-induced cell death, as prevention of
lipoperoxidation is not neuroprotective (Sarafian and Verity, 1991). Alternatively, there
is substantial evidence that interactions between MeHg and specific Ca2+ signaling
receptors and ion channels may be important in MeHg-induced neurotoxicity, as outlined

in the following sections.

iii) MeHg and Cholinergic Neurotransmission

MeHg alters Cholinergic transmission both pre- and post-synaptically within the
CNS and PNS alike. One of the early experimental observations regarding MeHg
neurotoxicity was its disruption of neurotransmitter release at the neuromuscular
junction, which was studied in part because of the observed increased incidence of
neuromuscular weakness during the Iraqi incident (Rustam et al., 1975) and in part

because the ACh release process is well characterized both biochemically and

20

physiologically. At the neuromuscular junction, the primary presynaptic effect is to
decrease nerve-evoked release of ACh and alter its spontaneous quantal release in a
mechanism apparently dependent on disruption of [Ca2+], (Juang, 1976; Atchison and
Narahashi, 1982). However, it is the post-synaptic effects of MeHg on Cholinergic
transmission which are of interest in the study of MeHg-induced neuronal death in the
CNS.

MeHg binds to both nicotinic and muscarinic Cholinergic receptors in a variety of
neuronal preparations. In Torpedo, MeHg inhibits binding of ACh to the muscle-type
nicotinic ACh receptors (Sharnoo et al., 1976; Eldefrawi et al., 1977). In vitro, 10-30 11M
MeHg inhibits the fast depolarizing response of nicotinic receptors in NIB-115
neuroblastoma cells (Quandt et al., 1982). Both muscarinic and nicotinic receptors
contain a conserved pair of extracellular cysteine residues that are predicted to from a
disulfide bond between two of the extracellular loops of the receptor (Sato et al., 1976;
long et al., 1999). These thiol groups are critical in agonist and antagonist binding by
muscarinic receptors (Hedlund and Bartfai, 1979), and agents which modify these thiol
groups alter muscarinic receptor activity (Sato et al., 1976; Berstein et al., 1988). Given
that MeHg binds thiol groups with high affinity, it is not surprising that MeHg interacts
with several subtypes of muscarinic receptors both in vivo and in vitro. MeHg inhibits
agonist binding to both M1 and M2 muscarinic receptors in rat brain cortical membranes,
although its affinity for M1 receptors is much higher than for M2 receptors (IC50 values
of 3.4 11M and 149 11M for M1 and M2, respectively) (Castoldi et al., 1996). In this
preparation, MeHg inhibited agonist binding to the receptors in a competitive manner and

decreased the affinity of the agonist carbachol for the receptor, suggesting that MeHg

21

prevents agonist binding to the receptors, thereby disrupting neurotransmission post-
synaptically (Castoldi et al., 1996).

MeHg also affects muscarinic receptors in vivo. In rats treated with 0.5 or 2
mg/kg/day MeHg for 16 days, there was a Significant increase in the density of
muscarinic receptors in the cerebellum and hippocampus with no effect on receptor
affinity (Coccini et al., 2000). This effect was region-specific, as there was no
observable change in muscarinic receptor density in the cerebral cortex. Further, the
increased receptor density was not observed until 2 weeks after the cessation of
treatment, indicating a delayed up-regulation in the number of muscarinic receptors in
these brain regions (Coccini et al. , 2000).

The specific location of muscarinic receptors within the CNS may explain, in part,
the specific pattern of neurotoxicity observed following MeHg exposure. For example,
cerebellar granule cells express a higher density of muscarinic receptors than do any
other neurons within the cerebellar cortex as assessed by [3H]]quinuclidinyl benzilate
binding (Neustadt et al., 1988). The up-regulation of muscarinic receptors in the
cerebellum and hippocampus is of particular interest, as MeHg damages cells and
disrupts neurotransmission in these two regions (Yuan and Atchison, 1993; Yuan and
Atchison, 1994; Yuan and Atchison, 1995; Yuan and Atchison, 1997; Coccini et al.,
2000; Gasso et al., 2000). In cerebellar granule cells, muscarinic receptors modulate cell
death induced by several toxicants, ranging from non-depolarizing growth medium to the
wasp-venom peptide mastoporan (Galli et al. , 1995; Yan et al., 1995; de Luca et al. ,
1996; Lin et al., 1997). Apoptosis caused by mastoparan is mediated by the M3

muscarinic receptor and not the M2 or nicotinic receptor, suggesting that activation of a

22

Specific, PLC-linked muscarinic receptor can be important in toxicant-induced death of

cerebellar granule cells (Yan et al. , 1995).

iv) Intracellular Calcium Stores

Given the importance of maintaining low cytosolic levels of [Ca2+], and the role
of Ca2+ as an intracellular signal , it is not surprising that in addition to mechanisms for
Ca2+ removal fi'om the cell, there are also mechanisms for the intracellular storage of
Ca2+. Within neurons, the two primary organelles which store Caz+ are the mitochondria
and the SER. Mitochondria represent a low-affinity, high-capacity Ca2+ store, whereas
the SER is a high-affinity, low-capacity Ca2+ pool (Somlyo et al., 1985). While the SER
will move Ca2+ fi'om the cytosol into the SER lumen at low cytosolic [Ca2+]i, the
mitochondria require a more powerful stimulus, with a local extra-mitochondrial Ca2+
concentration of approximately 0.5 11M needed for neuronal mitochondria (N icholls,
1978; Nicholls and Scott, 1980). Disruption of Ca2+ regulation in either of these stores
can pose a serious challenge to neuronal survival. As discussed below, mitochondrial
effects are crucial in Ca2+-mediated cell death, particularly apoptosis. In fact, there is
substantial evidence to support the hypothesis that mitochondrial Ca2+ uptake is crucial
for both apoptotic and necrotic neuronal death (lchas and Mazat, 1998; Lemasters et al. ,
1998; Andreyev and Fiskum, 1999; Kruman and Mattson, 1999; Duchen, 2000).
Additionally, disruption of SER Ca2+ content can be lethal to neurons (Kass and
Orrenius, 1999; Mattson et al., 2000), as will also be discussed below.

Mitochondria accumulate Ca2+ based on the extremely negative membrane

potential within the inner mitochondrial membrane (140-180 mV, negative inside)

23

(Gunter and Gunter, 1994), making Ca2+ buffering an energetically favorable process.
Uptake occurs via the ruthenium red-sensitive uniporter as well as by the so-called rapid
uptake mode (Gunter and Gunter, 1994; Sparagna et al., 1995; Buntinas et al., 2001).
However, Ca2+ buffering comes at a price: uptake of a large amount of Ca2+ occurs at the
expense of the proton motive force, and can thus cause uncoupling of ATP synthesis
from respiration. Under physiological conditions, mitochondrial Ca2+ buffering is
important in regulating cytosolic [Cap] elevations which result from either influx of
Ca2+, or release of Ca2+ fiom the SER, and may play a role in modulating cellular Ca2+
oscillations (Simpson and Russell, 1996; Rizzuto et al. , 1998). At low concentrations,
mitochondrial matrix Ca2+ (Ca2+m) can activate several mitochondrial enzymes associated
with the tricarboxylic acid (TCA) cycle, as well as stimulate the adenine nucleotide
translocase to cause increased movement of ATP from the mitochondria into the cell.
Further, mitochondrial buffering of physiological Ca2+ transients results in parallel
depolarization of the inner membrane, regardless of whether the Ca2+ source is
intracellular or extracellular (Loew et al., 1994). However, high [Ca2+]m can become
hazardous to the cell via inhibition of the TCA cycle, the electron transport chain, the
mitochondrial ATPase, and dissipation of the proton gradient, all of which can contribute
to loss of mitochondrial ATP synthesis and cell death (Simpson and Russell, 1998b).
Mitochondria are capable of releasing matrix Ca2+ by three different mechanisms
(see Gunter and Pfeiffer, 1990, Gunter and Gunter, 1994, and Bemardi, 1996, for reviews
of these mechanisms). The first is reversal of the uniporter, which normally extrudes
cytosolic Ca2+ into the mitochondrial matrix. The second is a NaI-Ca2+ exchanger, which

is the primary mechanism for mitochondrial Ca2+ efflux in mitochondria isolated fiom

24

brain (Simpson and Russell, 1998b). The third is opening of the mitochondrial
permeability transition pore (MTP). The MTP is a megapore formed on the inner
mitochondrial membrane in response to adverse conditions including oxidative stress,
high mitochondrial matrix Ca2+ and anoxia (Dubinsky and Rothman, 1991; Petronilli et
al., 1994; Chernyak et al., 1995; Byrne et al., 1999). When open, the MTP allows the
passage of any molecule less than 1500 Da, including Ca2+, Zn2+ and the pro-apoptotic
cytochrome c, across the usually impermeable inner mitochondrial membrane (Szabo and
Zoratti, 1992; Chemyak and Bemardi, 1996; Pastorino et al., 1999; Szalai et al., 1999;
He et al., 2000; Petronilli et al., 2001). Because opening of the MTP can dissipate the
proton gradient within the inner membrane, a corresponding loss of 111m (depolarization) is
often observed and can be used as a diagnostic tool for observing a permeability
transition (White and Reynolds, 1996; Trost and Lemasters, 1997). Induction of the MTP
has been correlated with conditions such as ischemia, stroke and apoptosis, although its
precise physiological role is not yet understood (see Bemardi and Petronilli, 1996 for
review). Sustained opening of the MTP has also been linked to cell death caused by a
variety of compounds and conditions, including glutamate excitotoxicity in cerebellar
granule cells (N ieminen et al., 1996; Schinder et al., 1996; White and Reynolds, 1996).
Opening of the MTP can be inhibited by the immunosuppressant cyclosporin A (CSA),
which acts to keep the pore closed or delay its opening in the presence of inducing agents
(Broekemeier et a1. , 1989; Bemardi et al., 1993; Nienrinen et al., 1996). Because
opening of the MTP can release significant amounts of Ca2+ into the cytosol, MTP
induction is a possible mechanism by which MeHg causes elevated [Ca2+]i during the

early stages of MeHg exposure.

25

In addition to mitochondrial Ca2+, neurons can sequester Ca2+ within the SER.
The SER contains ryanodine and IP3-sensitive Ca2+ pools which are thought to be distinct
in cerebellar granule cells (Irving et al., 1992). The ryanodine receptors are tetrarneric
complexes formed from 550-565 kDa monomers, forming one of the largest proteins
present within cells (Coronado et al. , 1994). Ryanodine receptors are characterized by
the ability of the plant alkaloid ryanodine to bind with nanomolar affinity, causing Ca2+
release at low concentrations and inhibition of Ca2+ release at higher concentrations.
Ca2+ release fiom the SER ryanodine-sensitive Ca2+ pool is elicited by a variety of
agonists, including caffeine, low concentrations of ryanodine, Ca2+, and cyclic ADP-
ribose, and can be blocked by antagonists such as high concentrations of ryanodine,
dantrolene, and ruthenium red (Coronado et al., 1994). While originally identified in
skeletal and cardiac muscle as a Ca2+ release mechanism during depolarization-mediated
contraction, ryanodine receptors have now been identified in many tissues including
neurons. Ryanodine receptors are involved in the process of Ca2+-induced Cay-release
(CICR), in which cytosolic Ca2+ activates ryanodine receptors to cause Ca2+ release fiom
the SER CICR has been identified in several neuronal preparations as participating in
the Observed increase in Ca2+; following application of a depolarizing stimulus, and in
cerebellar granule cells has been shown to be a major component of total Ca2+ signal
detected following entry of Ca2+c (Simpson et al., 1993). Immunocytochemistry in rat
cerebellar granule cells indicates the presence of Type II ryanodine receptors (although at
very low levels), whereas neighboring Purkinje neurons contain Type I receptors
(Simpson et al., 1996). Cerebellar granule cells grown in culture possess a ryanodine-

sensitive SER Ca2+ pool; however, studies in this system are limited in that after 4 days in

26

culture the ryanodine receptor agonist caffeine no longer produces Ca2+ release from the
SER (Irving et al., 1992).

The SER IP3-releasable Ca2+ pool also plays a role in elevations of Ca2+; in a
variety of cells, including neurons. The homotetrameric 1P3 receptors, while related
structurally to ryanodine receptors, differ in that their subunits are smaller (305—3 13 kDa
per monomer) and have a 4-10 times smaller conductance than ryanodine receptors
(Marks, 1997). Of the 3 known receptor subtypes, both cerebellar granule and Purkinje
neurons possess Type I 1P3 receptors, with particularly high levels in cerebellar Purkinje
cells (Sharp et al., 1999). While 1P3 is the primary agonist at these receptors, their
function is also antagonized by heparin and decavanadate. Activation of
phosphoinositide-specific phospholipase C (PLC) causes the hydrolyis of
phosphatidylinositol 4,5-bisphosphate to yield inositol 1,4,5-triphosphate (1P3) and 1,2-
diacylglycerol. The IP3 then acts at 1P3 receptors on the surface of the SER membrane to
cause release of Ca2+ fiom the SER These receptors are also sensitive to local cytosolic
Ca2+ levels in a bell-shaped fashion, with activation at low [Ca2+], inactivation at higher
[Ca2+], and peak activity at approximately 300 nM (Bezprozvanny et al., 1991). Their
activity is complicated further by their interaction with nearby ryanodine receptors such
that release of Ca2+ fi'om the 1P3 pool can elicit CICR from ryanodine receptors, thus
magnifying the Ca2+ Signal (Irving et al., 1992).

Both the 1P3 and ryanodine-sensitive Ca2+ pools are filled by SER Ca2+-ATPases
(SERCAS) which allow for rapid uptake of cytoplasmic Ca2+. Although there are 5
subtypes known to date, they all possess varying degrees of sensitivity to the inhibitors

thapsigargin and 2,5-Di(tert-butyl)-l ,4-benzohydroquinone (BHQ). Thapsigargin causes

27

an elevation of [Ca2+]: by inhibiting the SERCA, which leads to emptying the SER Ca2+
stores (Thastrup et al., 1990). In cerebellar granule cells, there is some distinction
between the SERCAS filling the 1P3 and ryanodine pools in that BHQ will affect the IP3
but not the ryanodine pool (Simpson et al. , 1996), suggesting that a separate filling
mechanism exists for the 1P3 pool. Under physiological conditions, release of SER Ca2+
does not harm the cell; however, several lines of evidence suggest that prolonged loss of
SER Ca2+ levels can have deleterious effects on normal cell function. For example,
depletion of SER Ca2+ can disrupt normal protein synthesis pathways, as evidenced by
the inhibition of general protein synthesis following rapid release of SER Ca2+ (Brostrom
and Brostrom, 1990). Depletion of SER Ca2+ can also cause the release of a Specific
deoxyribonuclease involved in DNA degradation, which could contribute to apoptotic
cell death (Peitsch et al. , 1993). Further, depletion of SER Ca2+ content with thapsigargin
has been shown to cause cell death in a variety of cell types, including neurons (Levick et
al., 1995; Wei and Perry, 1996; Wei et al., 1998). It is thus not surprising that many cells
possess a mechanism for refilling the SER Ca2+ stores via store-operated (or capacitative)
Ca2+ entry (SOCE) (Putney, 1986). When the SER becomes depleted of Ca2+, there is a
sensing mechanism in the SER that detects the Ca2+ decrease and, through an as yet
unidentified signal, sends a signal to plasma membrane Ca2+ channels to activate Ca2+
influx to replenish the stores (Putney, 1986; Ma et al., 2000). The affected Ca2+ channels
are neither the classic receptor-operated nor voltage-gated channels and are thus termed
store-operated Ca2+ channels (SOCCs) (Kiselyov et al., 1998). Although this system was
originally identified in non-excitable cells, it is also found in neuronal preparations,

including cerebellar granule neurons (Chavis et al., 1995; Simpson et al., 1995).

28

Interestingly, recent evidence suggests that opening of SOCCS may be due to physical
interactions between 1P3 receptors (Kiselyov et al., 1998; Ma et al., 2000) and/or
ryanodine receptors (U sachev and Thayer, 1999; Kiselyov et al. , 2000) with the SOCCS
themselves.

In addition to interactions between the SER and the plasma membrane, there is
also increasing evidence for interaction between SER Ca2+ pools and mitochondria.
Release of Ca2+ fi'om the 1P3 pool is followed closely by increased mitochondrial matrix
Ca2+ in response to the microdomain of high [Ca2+] generated near the mouth of the IP3
receptor channels (Miyata et al., 1991). Use of CaZI-Sensitive photoproteins
simultaneously targeted to the mitochondrial matrix and SER indicated that in living
HeLa cells there is a close relationship between release of Ca2+ fi'om the SER and
increased matrix Ca2+ of mitochondria which were in close apposition to 1P3 receptors
(Rizzuto et al., 1998). In fact, Ca2+ released from the IF; pool caused a greater increase
in mitochondrial matrix [Cay] than in the bulk cytosol, whereas intracellular perfusion of
a buffered Ca2+ solution increased [Ca2+], of the SER and mitochondria to the same
extent, suggesting preferential Ca2+ uptake of 1P3 Ca2+ by mitochondria. It has been
hypothesized that mitochondria located near 1P3 receptors may function to buffer released
Ca2+ to prevent receptor inactivation by high [Ca2+], thereby potentiating 1P3 receptor

activity (Rizzuto et al., 1998; Simpson and Russell, 1998a).
v) MeHg and the Smooth Endoplasmic Reticulum

In disrupting [Ca2+]; homeostasis, MeHg causes Ca2+ release from intracellular

stores into the cytosol in cerebellar granule cells (Sarafian, 1993; Marty and Atchison,

29

1997), synaptosomes (Kauppinen et al., 1989; Denny et al., 1993) and NG108-15 cells
(Hare er al., 1993; Hare and Atchison, 1995). In T lymphocytes (Tan et al., 1993) and
NG108-15 cells (Hare and Atchison, 1995), most of this Ca2+ originates in the SER.
Both organic and inorganic mercurials cause rapid release of Ca2+ fi'om muscle SER-
derived vesicles (Abramson et al., 1983). The release of SER Ca2+ may result fiorn the
observed inhibition of the SER CaZI-ATPase by MeHg, presumably resulting from its
binding to critical thiols in the Ca2+ pump (Abramson et al., 1983; Chiu et al., 1983), or
from the ability of MeHg to cause non-specific damage to lipid membranes through lipid
peroxidation (Sarafian and Verity, 1991; Verity et al., 1994; InSug et al., 1997).
However, the Ca2+ release may also occur by more specific mechanisms mediated by
intracellular Ca2+ channels located on the SER membrane, such as the 1P3 and ryanodine
receptors.

There are three subtypes of 1P3 receptor (Types 1, II and III) which are
differentially expressed throughout the CNS. In rat brain, Type I is the primary neuronal
type, expressed in large quantities in the cerebellar Purkinje neurons as well as in the
neighboring granule cells (which also express Type III receptors) (Sharp et al., 1999).
Type 11 receptors are found only in glia, while Type III receptors are enriched in lirnbic
and basal forebrain regions, particularly in the nerve terminals (Sharp et al., 1999).
Additionally, there is a developmental time-source for the expression of receptor
subtypes within the CNS, such that during embryonic development, only the Type III
receptor is expressed in large quantities, while the Type I receptor is expressed highly
only postnatally (Sharp et al., 1999). Expression of the Type III receptor is hypothesized

to play a role in the normal onset of programmed cell death in the brain during

30

development (Sharp et al., 1999; Blackshaw et al., 2000). T cells that are deficient in 1P3
Type I receptors are resistant to multiple inducers of apoptosis, suggesting that these
receptors may play a role in Ca2+-mediated cell death (Jayaraman and Marks, 1997;
Marks, 1997). Type III IP3 receptor expression is augmented in lymphocytes undergoing
cell death, and Type HI 1P3 receptors are expressed at high levels during developmental
apoptosis in postnatal cerebellar granule cells, chick dorsal root ganglia and embryonic
hair follicles (Blackshaw et al., 2000). Introduction of a Type III 1P3 receptor antisense
oligonucleotide prevents cell death caused by removal of nerve growth factor in chick
dorsal root ganglia, implicating the Type III receptor in neuronal death processes
(Blackshaw et al., 2000). Thus, 1P3 receptors are involved not only in Ca2+: signaling but
also in neuronal development and death.

The differential expression of these receptors becomes important when addressing
MeHg neurotoxicity, as the second messenger IP3 and its receptor have been implicated
as playing a role in MeHg-induced alterations of [Ca2+]i. In rat cerebellar granule cells in
culture, exposure to 5 11M MeHg for 30 min resulted in a twofold increase in cellular
inositol levels which potentially contributes to the observed elevation of [Cay], observed
in this system (Sarafian, 1993). Presumably, this increased [1P3] contributes to the
release of intracellular Ca2+ that was observed within the same time frame (Sarafian,
1993). In rat cerebellar membrane preparations, MeHg stimulates binding of [3 H] 1,4,5-
triphosphate and [3H]inositol l,3,4,5-tetrakisphosphate binding to 1P3 receptors (Chetty et
al., 1996). Additionally, rats treated with 2.5 mg/kg body wt/day MeHg for 7 days
exhibited an increase in binding of [3H]IP3 to cerebellar membranes; this increased

binding was not observed following treatment with inorganic mercury or lead (Chetty et

31

al. , 1996). MeHg-induced alterations in 1P3 production are not limited to the nervous
system but are also found on the immune system. In rat T lymphocytes, MeHg causes
sustained elevations in [Ca2+]: which, like those observed in cerebellar granule cells,
consist of an intracellular and extracellular component (Tan et al., 1993). The release of
Ca2+, was largely blocked by the 1P3 receptor antagonist heparin (Tan et al., 1993). In
NG108-15 cells, depletion of the IP3-sensitive Ca2+ pool reduced, but did not abolish, the
MeHg-induced increase in [Ca2+]; (Hare and Atchison, 1995). Interestingly, in these
cells, MeHg did not cause an increase in [1P3], suggesting that MeHg may interact
directly with the receptor to cause Ca2+ release. Finally, in mouse lymphocytes, exposure
to 1 11M MeHg for 1 hr resulted in a 50% reduction in the rate of inositol metabolism,
suggesting that once [1P3] becomes elevated, MeHg maintains [1P3] at higher-than-normal
levels (Nakatsuru et al., 1985). Thus, MeHg causes elevated [1P3] in a variety of neurons
and other cells.

Ca2+ release from the SER can also occur by opening of ryanodine receptor-linked
channels localized to the SER membrane. Ryanodine receptors, like 1P3 receptors, are
expressed as three different isoforms (1, II and HI) which are expressed differentially in
the CNS. Ryanodine receptors are involved in the process of CICR, in which cytosolic
Ca2+ activates ryanodine receptors to cause Ca2+ release fiom the SER (Irving et al.,
1992), and have also been implicated as mediators of neuronal death (Marks, 1997).
There is also limited evidence that MeHg targets ryanodine receptors to cause Ca2+
release fi'om the SER. Inhibition of Ca2+ release through ryanodine receptors causes

minor reductions in the amplitude of the MeHg-induced [Ca2+], increase in NG108-15

32

cells (Hare and Atchison, 1995). However, the role of ryanodine receptors in MeHg-

induced cell death has not yet been investigated.

vi) MeHg and Mitochondria

A number of studies indicate that MeHg targets several parameters of
mitochondrial fimction, including respiration and oxidative phosphorylation.
Additionally, MeHg has serious effects on the regulation of mitochondrial Ca2+. In
mitochondria derived from rat forebrain, MeHg causes release of ”Ca2+ from preloaded
mitochondria while preventing further Ca2+ uptake (Levesque and Atchison, 1991).
MeHg-induced alterations in neurotransmitter release appear to result primarily from
perturbation of mitochondrial Ca2+ regulation, with minimal or no contribution of Ca2+
from the SER in hippocampal slices (Gasso et al. , 2000) and the rat neuromuscular
junction (Levesque and Atchison, 1988).

More recent evidence suggests that MeHg causes opening of the MTP, which has
been implicated in apoptotic cell death caused by a wide variety of toxic insults (Ichas
and Mazat, 1998; Lemasters et al., 1998). Excess accumulation of Ca“ within the
mitochondrial lumen, production of reactive oxygen species, and depletion of ATP are all
activators of the MTP both in cells and in isolated mitochondria (see Bemardi et al.,
1998, for review). In MeHg toxicity, there is substantial evidence for the induction of all
three of these conditions; thus it is not surprising that in recent studies with human
monocytes (InSug et al., 1997) and lymphoid cells (Shenker et al., 2000), there is
evidence of MeHg-induced Opening of the MTP. Additionally, transfection of the murine

GT1-7 hypothalamic neuronal cell line with the MTP inhibitor bcl-Z reduces cell death

33

caused by MeHg, suggesting an important role for the MTP in MeHg-induced neuronal
death (Sarafian et al., 1994).

MeHg is also widely recognized as causing oxidative stress. MeHg-induced
elevations of reactive oxygen Species (ROS), such as superoxide anion, hydroxyl radicals
and hydrogen peroxide, are observed in human monocytes (InSug et al. , 1997), human T
cells (Shenker et al., 1998), and murine and rat brain (Ali et al., 1992; Yee and Choi,
1994). MeHg also causes lipid peroxidation in rat cerebellar granule cells in suspension
(Sarafian and Verity, 1991). The antioxidant vitamin B provides some protection
against neuronal damage in vivo (Chang et al., 1978; Usuki et al., 2001), suggesting that
ROS contribute to cellular damage caused by MeHg. The ability of a cell to handle ROS
may be correlated with its sensitivity to MeHg. MeHg-induced increases in fi'ee radicals
occur in brain regions specifically sensitive to the effects of MeHg, including the
cerebellum (LeBel et al., 1990; LeBel et al., 1992). Additionally, cells expressing
increased glutathione (GSH) exhibit decreased sensitivity to MeHg-induced damage
(Miura et al., 1994). However, there is considerable evidence that MeHg-induced
oxidative stress may not be the determinant factor in MeHg-induced neuronal death in
specific brain regions. Vitamin E prevents MeHg-induced increases in ROS and lipid
peroxidation but does not protect against cell death in rat cerebellar granule cells
(Sarafian and Verity, 1991). Additionally, modulation of intracellular GSH levels using
N-acetylcysteine (to increase GSH) or L-buthioniner(S,R)-sulfoximine (BSO) (to
decrease GSH) did not alter the neurotoxic effects of MeHg in primary cultures of rat

embryonic CNS cells (Du et al., 1999). Taken together, these studies suggest that while

34

MeHg causes increased levels of ROS in both adult and developing brains, the resultant

damage is not a determinant factor in the selective neurotoxicity of this compound.

vii) MeHg-Induced Interaction Between SER and Mitochondria Ca2+ Stores

There is growing evidence that interactions between SER and mitochondria are
important in determining cell survival. In permeabilized RBL-2H3 cells (Csordas et al.,
1999), permeabilized hepatocytes (Hajnoczky et al., 1999) and rat pulmonary smooth
muscle cells (Drummond and Tuft, 1999), Ca2+ release from the 1P3 receptor results in a
strong Ca2+ signal at the mitochondria. In HeLa cells transfected with specifically
targeted, Ca2+ sensitive green fluorescent proteins, Ca2+ release fiom 1P3 receptors results
in a higher localized concentration of Ca2+ at the outer surface of nearby mitochondria
than is observed in the cytosol (Rizzuto et al., 1998), revealing the preferential Signal
received by mitochondria from the SER. This Ca2+ signal between the SER and
mitochondria can be translated into a pro-apoptotic Signal; in a human tumor cell line,
activation of 1P3 receptors releases Ca2+ which causes opening of the MTP in
mitochondria, with subsequent release of cytochome c and promotion of apoptotic cell
death (Szalai et al., 1999). Mitochondria also communicate with the SER in a negative
feedback system which alters SER Ca2+ uptake and release (Landolfi et al., 1998),
indicating the complex nature of this Signaling pathway. Thus, any MeHg-induced
changes in the [Ca2+] of one intracellular store must be considered in the context of its
effects on the other store, and vice versa, in order to account for the complicated pattern

of Ca2+ signaling which occurs within neurons.

35

viii) Non-Ca2+ Divalent Cations

Elevation of intracellular concentrations of non-Ca2+ divalent cations is also
observed during MeHg exposure in NG108-15 cells (Hare et al., 1993; Denny and
Atchison, 1994), rat cerebellar granule cells (Marty and Atchison, 1997), and
synaptosomes (Denny and Atchison, 1994). In synaptosomes, the non-Ca2+ cation was
identified using 19F-NMR as Zn2+ (Denny and Atchison, 1994). The elevated [lay]; was
possibly due to the displacement of endogenous Zn2+ by MeH g fiom the numerous
proteins which bind Zn2+ with cysteine sulfhydryl groups (Hunt et al., 1984; Vallee and
Auld, 1990). Zn2+ itself can be neurotoxic (Manev et al., 1997; Cookson and Shaw,
1999), and has recently been shown to induce opening of the MTP in cortical neurons
(Sensi et al., 2000). In rats exposed to MeHg there was an elevated [ZnZI] in the brain
which was not found in control or inorganic Hg-exposed animals (Muto et al., 1991).
MeHg also caused elevated [Zn2+] in the kidney, another sensitive target organ during
MeHg toxicity (Muto et al., 1991). In spite of the intriguing possibility that disruption of
[Zn2+]i is a decisive factor in the selective toxicity of MeHg, several lines of evidence
argue against this hypothesis. First, the non-Ca2+ component of the MeHg-induced
increase in cerebellar granule cells is much smaller in amplitude than the Ca2+ component
(Marty and Atchison, 1997). Second, chelation of non-Ca2+ divalent cations with the
cell-permeant heavy metal chelator N,N,N',N'-tetrakis-(2-pyridylmethyl)ethylenediamine
(TPEN) does not protect against cell death caused by low micromolar concentrations of
MeHg at 3.5 hr post-exposure (Marty and Atchison, 1998). Finally, in cerebellar granule
neurons in culture, Zn2+-mediated cell death was found to depend on protein synthesis

(Manev et al., 1997). MeHg inhibits protein synthesis in cerebellar neurons (Sarafian et

36

al., 1984; Sarafian and Verity, 1985), cerebral cortex synaptosomes (Verity et al., 1977),
and brain cortex Slices (Yoshino et al., 1966), suggesting that the mechanism by which
Zn2+ exerts its toxic effects is non-functional during MeHg exposure. Thus, the

contribution of elevated [Zn2+]i to MeHg-induced neurotoxicity is minimal at best.

C) Rat Cerebellar Granule Neurons in Primary Culture as a Model for the

Selective Neurotoxicity of MeHg

As described above, MeHg causes preferential degeneration of the granule cell
layer of the cerebellar cortex in humans and experimental animals alike (Hunter and
Russell, 1954; Syversen et al., 1981; Leyshon and Morgan, 1991). While studies
performed in vivo provide the closest parallel to human exposures, they do not provide
enough information regarding the mechanisms underlying the selective neurotoxicity of
MeHg. Additionally, experiments performed in vivo are complicated by the large number
of interacting variables between different types of neurons within the cerebellar cortex.
Thus, the experiments in this dissertation were performed in rat cerebellar granule cells in
primary culture to allow for in-depth exploration of the actions of MeHg in these cells.
Because cerebellar granule cells are the most populous neuronal type in the cerebellum,
and are affected by several diseases and toxicants, they have been well characterized in
vivo. When isolated fiom neonatal rats (approximately 6-7 days old), cerebellar granule
cells grown in culture continue to develop and mature; they extend processes which
bifurcate into fibers resembling the parallel fibers found in the cerebellar cortex, and
make functional synapses with neighboring neurons. The presence and activity of many

plasma membrane and intracellular ion channels, receptors and proteins have been well

37

characterized for the first 14 days post-isolation, with developmental profiles compiled
for many cellular proteins (Gonzalez et al. , 1992; Aronica et al. , 1993; Aronica et al. ,
1993; Aronica et al., 1993; Resink et al. , 1994; Xia et al. , 1995; de Luca et al., 1996). In
the experiments performed for this dissertation, all cells were used at days 6-8 post-
isolation to allow for the maturation of cells. At this time in culture, granule cells express
L, N, P/Q and R types of voltage-gated Ca2+ channels; M2 and M3 muscarinic receptors;
metabotropic glutamate receptors; excitatory amino acid (glutamate) receptors (N -
methyl-D-aspartate (NMDA), alpha-amino-3 -hydroxy-5-methyl-5-isoxazole propionate
(AMPA), and kainate); GABAA receptors; Type I 1P3 receptors; and Type II ryanodine
receptors, among other receptors and channels (Meier et al. , 1984; Alonso et al., 1990;
Pizzi et al., 1991; Aronica et al., 1993; Aronica et al., 1993; Randall and Tsien, 1995;
Simpson et al., 1996; Randall and Tsien, 1997; Mhyre et al. , 2000; Mogensen and
Jorgensen, 2000). Thus, the cells express many of the channels and receptors which
characterize functionally mature cerebellar granule cells, making granule cells is primary
culture a useful model for the study of MeHg neurotoxicity.

The use of granule cells in culture is limited, however, by the lack of normal
interactions with other neurons that are present in the intact animal. In particular, there is
a relative lack of astrocytes in the culture, which comprise less than 10% of the cells in
the culture (Marty and Atchison, 1997). Astrocytes have been intensively studied as a
potential mediator of MeHg neurotoxicity (Aschner, 1996). Astrocytes concentrate
MeHg and, when exposed to MeHg, release the neurotransmitter glutamate (Oyake et al. ,
1966; Garman et al., 1975; Aschner et al., 1994; Aschner et al., 2000). Excess glutamate

causes excitotoxic death in cerebellar granule cells by causing influx of Ca”, through

38

NMDA receptors (Ankarcrona et al. , 1995; Budd and Nicholls, 1996a; Nicholls et al.,
1999). Thus, MeHg-induced release of glutamate from astrocytes may contribute to the
toxicity of MeHg in vivo. In granule cells in primary culture, inhibition Of ion flux
through NMDA, kainite and AMPA receptors does not alter MeHg-induced elevations of
[Ca2 1 (Marty and Atchison, 1997); however, these cells are grown in the relative
absence of astrocytes, which would be present in vivo. The lack of astrocytes in the
culture is a disadvantage when comparing the results obtained in the culture system to
those obtained in the whole animal. However, in examining the relative sensitivity of
granule cells to MeHg-induced damage, the lack of astrocytes is advantageous, as it
removes a potentially contributing factor to MeHg-induced neurotoxicity which is
probably not unique to the cerebellar cortex. Additionally, granule cells in culture
undergo cell death at much lower concentrations of MeHg than is observed in other less-
sensitive cell types, suggesting that granule cells are inherently more sensitive to MeHg-
induced cell death in the absence of any external contributing factors (Park et al. , 1996;
Marty and Atchison, 1998). Thus, the use of granule cells in primary culture provides an
opportunity to identify potential molecular targets which confer the unique sensitivity of

this cell type to the effects of MeHg.

39

CHAPTER TWO

INTERACTION OF METHYLMERCURY WITH MUSCARINIC
RECEPTORS CAUSES CALCIUM RELEASE FROM THE SMOOTH
ENDPLASMIC RETICULUM INOSITOL-l,4,5-TRIPHOSPHATE RECEPTORS

IN RAT CEREBELLAR GRANULE NEURONS

40

ABSTRACT

The contribution of Ca2+ release from the SER to MeHg-induced cation
dysregulation was studied using Single-cell microfluorimetry of granule cells in vitro
loaded with fura-2. Depletion Of SER Ca2+ with thapsigargin (10 uM) prior to MeHg
(0.2-1.0 11M) exposure decreased the amplitude of MeHg-induced release of [Ca2+]; stores
by approximately 30%, indicating that SER Ca2+ is released by MeHg. However, the
time-to-onset of [Ca2+]; elevations by MeHg was unaffected by thapsigargin treatment.
When the muscarinic receptors and inositol-1,4,5-triphosphate (1P3) receptors were
down-regulated by 24 hr pretreatment with bethanechol (BCh, 1 mM), MeHg-induced
release of Ca2+, was again reduced by 30-40%. BCh pretreatment significantly hastened
the time-to-onset of the MeHg-induced release of Ca2+i without altering the influx of
Caz} Atropine (10 11M) Significantly delayed the time-to-onset and reduced the amount
of [Ca2+]; released into the cytosol by MeHg, suggesting that MeHg interacts with
muscarinic receptors and produces 1P3 , which subsequently releases Ca2+ from the SER.
Ryanodine (10 11M) did not affect the time-to-onset or amplitude of MeHg-induced
increases in [Ca2+]i. Thus, MeHg interacts with muscarinic receptors with the end result
of Ca2+ release from the SER, which contributes to the MeHg-induced loss of [Ca2+];

regulation.

41

INTRODUCTION

As discussed in Chapter One, MeHg causes an early release of Ca2+ fiom at least
one intracellular source in cerebellar granule cells in culture (Marty and Atchison, 1997).
One possible source is the SER, which is a high-affinity, low-capacity Ca2+ pool filled by
the SERCA (Irving et al. , 1992; Gunter and Gunter, 1994). Under normal conditions,
Ca2+ release fiom the SER into the cytosol occurs via the [P3 receptor and/or the
ryanodine receptor. In NG108-15 cells, most of the MeHg-induced first-phase is due to
release of Ca2+ fiom the IP3-Sensitive pool in the SER, with some contribution from the
ryanodine-sensitive pool; however, emptying this pool with the SERCA inhibitor
thapsigargin does not eliminate the first phase, suggesting that at least one additional
source contributes to the first phase Ca2+ increase (Hare and Atchison, 1995). In granule
cells, the contribution of the SER to MeHg-induced elevations of [Ca2+]; is examined for
the first time in the experiments presented in this chapter.

At the SER, both organic and inorganic mercurials cause rapid release of Ca2+
from muscle SER-derived vesicles (Abramson et al., 1983). In granule cells in culture,
MeHg causes an increase in [1P3] at low micromolar concentrations (Sarafian, 1993),
although this increase in [1P3] is not observed in all neuronal models (Hare and Atchison,
1995) and has only been observed in granule cells after exposure to 5 uM MeHg for 30
min, which does not shed light on whether there is an early increase in [1P3] (Sarafian,
1993). The elevation of [1P3] could be due to binding of MeHg to muscarinic receptors,
which has been Observed in several neuronal systems (Eldefiawi et al., 1977; Von Burg
et al., 1980; Abd-Elfattah and Shamoo, 1981; Castoldi et al., 1996; Candura et al. , 1997).

Granule cells express M2 and M3 muscarinic receptors, with the M3 receptors linked to

42

PLC (Fukamauchi et al., 1991; Fohrrnan et al. , 1993). MeHg is also reported to stimulate
1P3 binding to its receptor in rat cerebellar membrane preparations, suggesting that MeHg
interacts with the 1P3 receptor itself (Chetty et al., 1996).

The experimental approach taken to study the contribution of SER Ca2+ was
similar to tint used by Hare and Atchison (1994) and Marty and Atchison (1997), in
which different receptor agonists and antagonists were used to inhibit different portions
of Ca2+ signaling pathways during MeHg exposure. First, the SERCA inhibitor
thapsigargin was applied prior to MeHg application to deplete the SER of its Ca2+ content
(Thastrup et al., 1990; Irving et al., 1992). Second, the muscarinic and 1P3 receptors
were down-regulated prior to MeHg exposure using the muscarinic agonist bethanechol
(BCh); this approach was taken to observe the contribution of the entire M3-PLC-IP3
receptors pathway to MeHg-induced increases in [Ca2+]; (F ohrrnan et al. , 1993;

F ukamauchi et al., 1993; Wojcikiewicz et al. , 1994; Simpson et al., 1996; F ukamauchi et
al., 2000). Third, the nonspecific muscarinic antagonist atropine was used to determine if
MeHg interacts with these receptors to causes elevations of [Ca2+]; (Yan et al., 1995).
Finally, the ryanodine receptor antagonist ryanodine was used to inhibit Ca2+
mobilization through these channels (Irving et al., 1992). The results presented here
indicate that MeHg interacts with muscarinic receptors to cause Ca2+ release from the
SER IP3-sensitive Ca2+ pool, and that this Ca2+ contributes to the whole-cell elevation of

[Ca2+], observed during MeHg neurotoxicity.

43

MATERIALS AND METHODS

Materials and Experimental Solutions. Deoxyribonuclease I (DNase I) and
type II trypsin were purchased from Worthington Biochemicals (Freehold, NJ). Cell
culture supplies, including Dulbecco’s Modified Eagle Medium (DMEM) and fetal
bovine serum were purchased from Gibco BRL (Grand Island, NY). SK&F 96365 was
purchased from Calbiochem (La J olla, CA). F ura-2 acetoxymethylester (fura-2 AM) was
purchased from Molecular Probes (Eugene, OR). Thapsigargin, carbamyl-B-
methylcholine chloride (bethanechol), carbamylcholine (carbachol), atropine, ryanodine,
poly-D-lysine and cytosine-B-arabinofirranoside were purchased fiom Sigma (St. Louis,
MO). Methyl mercuric chloride (MeHg) was purchased fi'om ICN Biochemicals Inc.
(Aurora, OH) and dissolved in deionized water.

Granule cell isolation. Cerebellar granule cells are isolated as described
previously (Marty and Atchison, 1997). Approximately 5-10 7-day-old Sprague Dawley
rats were used per isolation. Each rat was decapitated, then the skin was cut down the
dorsal midline from the neck to the nose. Using the flat end of the scissors, the skull was
carefully peeled back to reveal the brain. The cerebellum was cut along each side using
scissors, then gently lifted away fiom the rest of the brain. Each extracted cerebellum
was placed in an ice-cold beaker containing Cay-Mg2+ fi'ee Hank’s Buffered Saline
Solution (CMF-HBSS). CMF-HBSS contains (mM): 5.36 KCl, 0.44 KH2P04, 136.89
NaCl, 20.02 HEPES, 0.59 EDTA, 5.5 d-glucose, and 4.17 NaI-lCO; (pH 7.3, 37°C).
Once all cerebella had been dissected, the meninges were removed and the tissue finely

minced using sterile scissors. The tissue was then added to a beaker containing 0.25%

trypsin (w/v) and digested at 37°C in a Shaking water bath for 15 min. Cold (4°C)
DMEM with 5% fetal bovine serum and 0.04% DNase I (w/v) was added to inactivate the
trypsin. The cell suspension was centrifuged at 134g for 5 min (4°C), then the pellet was
resuspended in DMEM with 5% F BS and 0.04% DNase I and triturated approximately 15
times with a wide-bore fire-polished Pasteur pipette. The cell suspension was allowed to
sit for 4 min on ice, then the supernatant (which contained the dissociated cells) was
transferred to a centrifuge tube on ice. The addition of PBS-DNase followed by
trituration, settling of debris and transfer of supernatant was performed two more times.
The collected supernatant was gently underlaid with 4% BSA and 3.82% MgSOa
(dissolved in CMF-l-IBSS) and the tube was centrifuged at 134g at 4°C for 5 min. The
resulting pellet was resuspended in the plating medium that consists of DMEM with 10%
PBS, 25 mM KCl, SOuM GABA, 50uM kainate, 5 rig/ml insulin, 100 U/ml penicillin and
50 ug/ml streptomycin. Cells were plated at a density of 2.0-2.2x106 cells/35 mm dish,
each containing a 25 mm glass coverslip coated with 0.1 mg/ml poly-D-lysine. After 21
hours, 10 11M cytosine-B-arabinofirranoside was added to inhibit glial cell proliferation.
Penicillin-streptomycin was not added at this or subsequent medium changes because
streptomycin alters Ca2+ channel function (Atchison et al. , 1988; Redman and Silinsky,
1994). Cells for all experiments were used at 6-8 days in vitro to allow for cell
attachment and maturation.

Measuring changes in [Ca2+]r. Unless otherwise noted, the standard buffer used for
experimental solutions is HEPES Buffered Saline (HBS) which contains (mM): 150
NaCl, 5.4 KCl, 1.8 CaClz, 0.8 MgSO4, 20 d-glucose, and 20 HEPES (fi'ee acid) (pH 7.3 at

room temperature of 23-25°C). The 40 mM K+ solution contains the same components

45

as HBS except with 40 mM K+ and 115.4 mM NaCl. The low-[Ca2+], EGTA-containing
bufl‘er (EGTA-HBS) has the same components as HBS minus CaClz and plus 20 11M
EGTA. For all experiments, pharmacological agents are dissolved in either HBS or
EGTA-HBS, with appropriate controls for any additional solvents used (DMSO or
ethanol, maximum final solvent of 0.01% (v/v), for all experiments). In a cell-free
system, fura-2 did not interact with MeHg or any of the agents used (BCh, atropine,
thapsigargin and ryanodine) at concentrations used in the present study (results not
shown).

To measure changes in [Ca2+]i, cells were loaded with 3-4 11M fura-2 AM in HBS
for 1 hr at 37°C, followed by perfusion with HBS for 30 min. Digital fluorescent images
were obtained using a Nikon Diaphot microscope (Nikon, Tokyo, Japan) or an Olympus
D(70 microscope (Olympus Optical Co., Tokyo, Japan), coupled to an IonOptix system
(Milton, MA) with an Open Perfusion Micro-Incubator (Medical Systems Corp.),
maintained at 37°C. For each experiment, changes in emitted fluorescence at excitation
wavelengths of 340 and 380 nm were monitored simultaneously in multiple soma (3-10
cells per experiment) within the same microscopic field. The fluorescence ratio (Fm/330)
indicated the approximate amount of [Ca2+]i, however the data were not converted to
[Ca2 i due to fura-2 interactions with other divalent cations, such as Zn”, a cation known
to play a role in the MeHg response (Hare et al., 1993; Denny and Atchison, 1994).

All experiments began with a l min wash with HBS (to establish baseline
fluorescence), followed by l min exposure to 40 mM K+ solution; only cells which
exhibited a reversible increase in [Ca2+], following the KI exposure were considered to be

viable for the experiment. The K" solution was washed out for 3 min with HBS, then one

46

of the experimental protocols was followed. Pharmacological agents were used either to
empty the SER of its Ca2+ content or block Ca2+ release fiom receptors on the SER
membrane. To empty the SER prior to addition of MeHg, the SERCA inhibitor
thapsigargin (10 11M) was applied for 5 min immediately prior to MeHg (0.2-1 .0 uM) in
HBS or EGTA-HBS (Irving et al., 1992; Simpson et al., 1996). To block the muscarinic
receptors, the nonspecific muscarinic receptor antagonist atropine (10 11M) was applied
for 10 min prior to, as well as during, exposure to 0.2-1.0 uM MeHg in HBS or EGTA-
HBS (Xu and Chuang, 1987; Yan et al., 1995). To inhibit Ca2+ release from the
ryanodine receptors, 10 uM ryanodine was applied for 10 min prior to, as well as during,
exposure to 0.2-1.0 11M MeHg in HBS or EGTA-HBS (Irving et al., 1992). To down-
regulate the M3 muscarinic receptors and thus subsequently the IP3 receptors, 1 mM
bethanechol (BCh) was added directly to the growth media 24-30 hr prior to loading cells
with firra-2 (Fohrman et al., 1993; F ukarnauchi et al. , 1993; Wojcikiewicz et al., 1994;
Simpson et al., 1996; Fukamauchi et al. , 2000). For BCh experiments, the fura-2 AM
was added directly to the growth media, thus cells were in BCh-fiee solution for the 30
min I-IBS rinse and the 5 min viability test with 40 mM KL In these experiments, MeHg
(0.2-1.0 11M) in HBS or EGTA-HBS was perfused immediately following the 3 min K+
solution wash period.

Following exposure to MeHg, the time-to-onset of the first and second CaZ+i phases
was determined: the first phase (from release of Ca2+: pools) was measured from the
point at which the ratioed fluorescence irreversibly leaves baseline, and the second phase
began at the point at which the 380 nm fluorescence sharply drops. For experiments

performed in the absence of Ca2+,” the amplitude of the fura-2 ratio increase was

47

determined as the fluorescence increase above baseline, and was normalized to the peak
fura-2 ratio caused by the 1 min exposure to 40 mM K" (“normalized ratio”). Ratio data
was normalized to remove Slight variations in fluorescence intensity resulting from
differences in cell size and fura-2 loading efficiency. For both time-to-onset and
normalized ratio experiments, data from each cell observed in an experiment were
averaged to provide mean time-to-onset for that dish of cells (n=1). In order to minimize
differences between cell isolates, experiments using MeHg alone and MeHg with the
pharmacological agent were run on the same day, and experiments using the same agents
were performed in at least two separate cell isolates.

Statistics. Comparisons of mean time-to-onset for “MeHg” vs the corresponding
“MeHg plus inhibitor” cells are made using Student’s paired t-test, with values of p<0.05

considered to be statistically significant.

48

RESULTS

The SER contributes Ca2+ to the MeHg-induced release of Car's. To
determine whether the SER contributed to the first-phase [Ca2+]t elevation, the SER was
depleted of its Ca2+ content with the SERCA inhibitor thapsigargin. To verify that
thapsigargin abolished agonist-induced release of Ca2+ fi'om this organelle, experiments
were performed using the nonspecific muscarinic agonist bethanechol (BCh, 1 mM),
which activates M3 muscarinic receptors to cause production Of 1P3 and subsequent
release of Ca2+ fiom the 1P3 receptor (Figure 2.1). A 5 min application of 10 11M
thapsigargin abolished the BCh-activated increase in [Ca2+]i in 13 of 15 cells examined
(from 3 dishes per treatment and two separate cell isolations), indicating the effectiveness
of thapsigargin in reducing the SER Ca2+ pool (Table 2.1). Thapsigargin (10 11M) was
then applied for 5 min immediately prior to MeHg (0.2-1.0 11M). In the presence of
Ca2+“ thapsigargin did not alter the time-to-onset of the frrst-phase release of [Caz+]i or
the second-phase influx of [Ca2+]; at any MeHg concentration tested (Figure 2.2). The
time-to-onset for the first-phase increase in fura-2 ratio between neurons treated with
MeHg only and MeHg plus thapsigargin was not different for 0.2, 0.5 or 1.0 uM MeHg.
Additionally, thapsigargin did not change the time-tO-onset of the second-phase Ca2+
influx for 0.2, 0.5 or 1.0 uM MeHg. In nominally Ca2+,-fiee solution (“EGTA-HBS”),
thapsigargin again did not alter the time-to-onset of the first-phase [Cay], increase caused

by 0.2, 0.5 or 1.0 11M MeHg (Figure 2.3).

49

 

 

 

 

080 - g l_(_“
A
§ B_Ch {Eh
s? 0.70 -
('1
LL.
v
0
-~ 0.60 -
‘5 al
ed
0.50 L L L l
0 5 10 15 20
Time (min)

Figure 2.1. Alternating exposures to 40 mM K+ and 1 mM BCh cause increases in
[Ca2+]. in cerebellar granule cells. Representative changes in fura-2 fluorescence
(“Ratio”) caused by alternating exposure to 40 mM K+ (1 min) and 1 mM BCh (1.5 min),
in a single granule cell. The amplitude of the fluorescence change in response to the BCh
is normalized to the response to the preceding K+ exposure as a relative measure of the
amount of Ca2+ released from the IP3-sensitive Ca2+ pool. This protocol was used to test
the effect of pharmacological agents on release of Ca2+ fiom the IP3-Sensitive Ca2+ pool
(see Table 2.1).

50

 

 

 

 

 

 

INHIBITOR NORMALIZED NORMALIZED PERCENT
RATIO (N O RATIO (WITH DECREASE
INHIBITOR) INHIBITOR)

No inhibitor 0.71:1:0.11 0.56i0.1 1 21.1%

(control)

Thapsigargin 0.59d:0.08 0.042t0.03* 93.2%

(10 11M)

BCh (1 mM) 0.13d:0.01 0.03zt0.01* 76.9%

Atropine 0.42:1:0.03 0.004zt0.004* 99.3%

(10 11M)

Ryanodine 0.51i0.09 0.242t0.04* 53.0%

( 10 11M)

 

 

 

 

 

 

Table 2.1. Effect of pharmacological inhibitors on peak fura-2 ratio induced by 1.5
min 1 mM BCh in cerebellar granule cells. Cells were exposed to alternating
challenges of 40 mM K+ (1 min) and 1 mM BCh (1.5 min), as shown in Figure 2.1. The
pharmacological inhibitor was then applied, and the KI-BCh protocol repeated. The peak
fura-2 ratio response to the BCh treatment was normalized to the preceding response to
the K+ depolarization (“normalized ratio”). The peak BCh response was compared in the
same cell in the absence or presence of inhibitor. Because not all cells responded to BCh,
comparisons were made only in cells which responded to the initial BCh exposure. For
the 24 hr 1 mM BCh experiments, comparisons were made between cells receiving 24 hr
BCh pretreatment and cells receiving 24 hr vehicle pretreatment. Because the
experimental protocol for the 24 hr BCh experiments prevented comparisons within the
same cell, all cells were included in this data, including control cells that did not respond
to the BCh treatment, to prevent any experimental bias. Thus, the normalized ratios '
under control conditions are lower than those observed for the other pharmacological
agents due to the inclusion of non-responding cells in the 24 hr BCh data. For the
ryanodine experiments, 1 mM carbachol was used instead of 1 mM BCh. Results are
presented as meaniSEM (n = 3). The asterisk (*) indicates a value significantly different
from the response prior to addition of the pharmacological agent (p<0.05).

51

A - - Thaps - + Thaps

   

     

A 15

.S

E:

8

G

O

5':

c5
.§

[-1

0.5 1.0
[MeHg] (11M)
B - -Thaps - +Thaps

.S

E 50

8

1::

0

5'3

ti»

.§

[-1

0.5
[MeHg] (FM)

Figure 2.2. Thapsigargin does not affect the time-to-onset of MeHg-induced
increases in [Ca2+]i. A) Pretreatment with 10 uM thapsigargin (“thaps”) for 5 min did
not alter the time-to-onset of the MeHg-induced first-phase [Ca +]i increase caused by
0.2, 0.5, or 1.0 uM MeHg. (B) In the same cells, pretreatment with 10 11M thapsigargin
for 5 min also did not alter the time-to-onset of the MeHg-induced second-phase influx of
Ca2+, caused by 0.2, 0.5, or 1.0 11M MeHg. Results are presented as mean i SEM (n=3-
5).

52

- - Thaps - + Thaps

N
O
I

 

Time-to-Onset (min)
25

 

O

   

0.5
[MeHg] (11M)

Figure 2.3. Thapsigargin does not affect the time-to-onset of MeHg-induced [Ca2+].
increases in the absence of Ca2+” Pretreatment with 10 M thapsigargin (“thaps”) for 5
min did not alter the time-to-onset of the MeHg-induced first-phase 2[Caz’TJi increase

caused by 0. 2, 0. 5, or 1.0 M MeHg, measured In the absence of Ca2 c. Results are
presented as mean i SEM (n=3- 5).

53

- - Thaps - + Thaps

N
1

Normalized Ratio

 

 

O

0.2 0.5 1.0
[MeHg] (11M)

Figure 2.4. Thapsigargin decreases the amplitude of MeHg-induced fura-2 ratio
increases, as measured in the absence of Ca2+” Pretreatment with 10 11M thapsigargin
(“thaps”) for 5 min significantly decreased the amplitude of the MeHg-induced first-
phase [Ca2+]; increase caused by 0.2, 0.5, and 1.0 1.1M MeHg. The firm-2 ratio increase
was normalized to the amplitude Of the 40 mM K+ response (“Normalized Ratio”).
Results are presented as mean i SEM (n=3-5). The asterisk (*) indicates a value
significantly different from the corresponding MeHg-only data (p < 0.05).

54

However, there was a significant decrease in the amplitude of the first phase caused by
0.2 uM MeHg (from a normalized ratio of 1.11i0.2 to 0.602t0.08), 0.5 uM MeHg (from
1.38:1:024 to 0.86:1:0.08), and 1.0 uM MeHg (from 1.13i0.08 to 0.85:1:009) (Figure 2.4).
This represents a decrease in the first-phase ratio amplitude of approximately 46%, 38%
and 25% caused by 0.2, 0.5 and 1.0 11M MeHg, respectively.

Down-regulation and desensitization of the M3-IP3 receptor pathway
decreases the MeHg-induced increase in [Carin A 24 hr application of 1 mM
carbachol causes down-regulation and desensitization of M2 and M3 muscarinic
receptors, and IP3 receptors, in cerebellar granule cells in culture (F ohrman et al., 1993;
Simpson et al., 1994). BCh, like carbachol, is a non-specific muscarinic agonist and thus
would be expected to have similar effects; BCh, however, has the advantage of being
more selective for muscarinic receptors with minimal activity at nicotinic ACh receptors.
Pretreatment with 1 mM BCh for 24 hr significantly inhibited the Ca2+ increase in
response to a 1.5 min exposure to 1 mM BCh from 0.13i0.01 to 0.032h0.01, a 77% loss of
the IF; response (Table 2.1). The 24 hr BCh treatment was then tested for its effect on
MeHg-induced increases of [Ca2+]. BCh pretreatment hastened the time-to-onset of the
first-phase increase in fura-2 ratio for 0.2 and 0.5uM MeHg, but did not change the time-
to-onset of [Cay]; increases caused by 1.0 11M MeHg (Figure 2.5A). However, the time-
to-onset of the second-phase influx of Ca2+, was not changed for MeHg as compared to

MeHg plus 24 hr BCh pretreatment for 0.2 uM, 0.5 GT 1.0 11M MeHg (Figure 2.53).

55

- -BCh - +BCh

N
O
l

Time-to-Onset (min)
5

 

 

 

 

0
0.2 0.5 1.0
[MeHg] (PM)
B
- -BCh - +BCh
A 30 -
.E
E
a 20
2
C?
$ 10
0)
E
r—
o
0.2 0.5 1.0
[MeHg] (MM)

Figure 2.5. Pretreatment with BCh hastens the MeHg-induced release of [Ca2+]i
without affecting the onset of MeHg-induced influx of Ca2+, A) Pretreatment with 1
mM BCh for 24 hr significantly hastened the time-to-onset of the first-phase MeHg-
induced release of Ca2+, caused by 0.2 and 0.5, but not 1.0 11M, MeHg. B) In the same
cells, 1 mM BCh pretreatment did not alter the time-to-onset of the second-phase influx
of Ca2+, at any MeHg concentration examined. Results are presented as mean :h SEM
(n=4-6). The asterisk (*) indicates a value significantly different from the corresponding
MeHg-only data (p < 0.05).

56

   

 

 

A 15 ~
.5
E
r; 10
‘8
C?
$ 5
0)
.§
F-

o

0.2 0.5 1.0
[MeHg] (11M)

Figure 2.6. Pretreatment with BCh hastens the MeHg-induced release of [Ca2+]i, as
measured '- lh absence of Ca2+, Pretreatment with 1 mM BCh for 24 hr significantly
hastened tin: tim-to-onset of the first-phase increase in [Ca2+]i caused by 0.2 and 0.5, but
not 1.0, uM, MeHg. Results are presented as mean :1: SEM (n=4-5). The asterisk (*)
indicates a value significantly different from the corresponding MeHg-only data (p <
0.05).

57

CCCP + Oligomycin

 

 

 

 

 

 

A 0.80 -
A
§
§ 070 - -BCh
LL?
v
o
g 0.60
a:
0.50 ' L ‘
0 5 10 15
Time (min)
CCCP + Oligomycin
0.80 —
B A
a3 +BCh
,5, 0.70 *-
LL.
v
o
g 0.60 '
ad
0.50 4 ‘ ‘
0 5 10 15
Time (min)

Figure 2.7. Pretreatment with BCh (1 mM, 24 hr) does not increase the [Ca2+]
within mitochondria. Representative changes in fura-2 fluorescence in response to
mitochondrial depolarization with carbonyl cyanide m-chlorophenylhydrazone (CCCP)
(5 11M) and oligomycin (10 11M). Granule cells were treated with vehicle (“-BCh”) or 1
mM BCh (“+BCh”) for 24 hr. CCCP and oligomycin were then applied to depolarize the
empty the mitochondria of their Ca2+ content (see Chapter Four) to determine the amount
of Ca2+ in these organelles following BCh treatment.

58

- -BCh - +BCh

Normalized Ratio

 

 

0.2 0.5 1.0
[MeHg] (11M)

Figure 2.8. Pretreatment with BCh decreases the amplitude of the fura-2 ratio
increase caused by MeHg, as measured in the absence of Ca2+, Granule cells were
treated with 1 mM BCh for 24 hr prior to 0.2-1.0 11M MeHg in EGTA-HBS. Results are
presented as mean :1: SEM (n=4-5). The asterisk (*) indicates a value significantly
different fi‘om the corresponding MeHg-only data (p < 0.05).

59

In the absence of Cab}, 24 hr pretreatment with BCh again hastened the time-to-onset of
the frrst-phase ratio increase for 0.2 and 0.5 11M MeHg but not 1.0 11M MeHg (Figure
2.6). The hastened onset of the first phase was not due to an increased amount of Ca2+ in
the mitochondria, as mitochondrial membrane depolarization with 5 uM carbonyl
cyanide m-chlorophenylhydrazone (CCCP) and 10 11M oligomycin indicated no
difference in the amount of mitochondrial Ca2+ content in cells treated with BCh as
compared to untreated cells (Figure 2.7). Like thapsigargin, BCh pretreatment also
decreased the amplitude of the first-phase fura-2 ratio increase caused by 0.2 11M MeHg
(fiom 1.11i0.13 to 0.65i0.06), 0.5 11M MeHg (from 1.14:1:0.17 to 0.70:1:0.15), and 1.0
11M MeHg (from 1.00:1:0. 12 to 0.69d:0.09) (Figure 2.8). This represents a decrease in the
first-phase ratio amplitude of approximately 41%, 39% and 31% caused by 0.2, 0.5 and
1.0 uM MeHg, respectively.

Blocking muscarinic receptors with atropine also decreases the MeHg-
induced increase in [C32+]|. To determine whether MeHg causes Ca2+ release from the
SER via binding to the M3 muscarinic receptor, experiments were performed using the
nonspecific muscarinic receptor antagonist atropine. Atropine blocked the response to a
brief (1 .5 min) 1 mM challenge with BCh, indicating the effectiveness of this compound
in blocking muscarinic receptors (Table 2.1). Application of 10 11M atropine for 10 min
prior to, as well as during, exposure to MeHg in the presence of Ca2+, indicated that
atropine significantly delayed the time-to-onset of the first-phase ratio increase caused by
0.2 and 0.5 11M MeHg, but not did not change the time-to-onset of elevations of [Ca2+];
caused by 1.0 11M MeHg (Figure 2.9A). Atropine did not alter the time-to-onset of the

second-phase influx of Ca2+c caused by 0.2, 0.5 or 1.0 uM MeHg (Figure 2.93).

60

- MeHg only - MeHg + AP

60

Time-TO-Onset (min)

 

0.2 0.5 1.0
[MeHg] (11M)

- MeHg only - MeHg + AP

 

 

Time-TO-Onset (min)

0.2 0.5 1.0
[MeHg] (FM)

Figure 2. 9. Atropine (“AP”) delays the first-phase release of Ca2+ .without affecting
the second-phase influx of Ca2+ e caused by MeHg. A) Granule cells were treated with
atropine (10 11M) for 10 min prior to, as well as during, exposure to MeHg. Atropine
significantly delayed the time-to-onset of the first- phase MeHg- induced release of Ca2 .
caused by 0. 2 and 0. 5, but not 1.0 11M, MeHg. B) In the +same cells, atropine did not alter
the time-tO-onset Of the second- phase influx of Ca2+ c at any MeHg concentration
examined. Results are presented as mean :1: SEM (n=4). The asterisk (") indicates a
value significantly different from the corresponding MeHg—only data (p < 0.05).

61

 

- MeHg only - MeHg + AP

.5.
o
I

U)
Q

p—i
O

 

 

Time-To-Onset (min)
N
O

O

0.2 0.5 1.0
[MeHg] (HM)

Figure 2.10. Atropine (“AP”) delays the time-to-onset of the MeHg-induced first-
phase release of Ca2+” as measured in the absence of Ca2+, Treatment with atropine
(10 11M) for 10 min prior to, as well as during, exposure to MeHg Significantly delayed
the time-to-onset of the [Ca2+]: increase caused by 0.2 and 0.5, but not 1.0, 11M MeHg.
Results are presented as mean i SEM (n=3-6). The asterisk (*) indicates a value
significantly different from the corresponding MeHg-only data (p < 0.05).

62

--AP -+AP

 

1.50 -
.9.
a
,6 1.00
O)
.521.
"a
g 0.50
0
Z

0.00

 

0.2 0.5 1.0
[MeHg] (11M)

Figure 2.11. Atropine (“AP”) decreases the amplitude of the fura-2 ratio increase
caused by MeHg, as measured in the absence of Ca2+” Granule cells were treated with
atropine (10 11M) for 10 min prior to, as well as during, exposure to MeHg. Results are
presented as mean i SEM (n=3-6). The asterisk (‘) indicates a value significantly
different fiom the corresponding MeHg-only data (p < 0.05).

63

In cells treated with MeHg in EGTA-HBS, atropine again delayed the time-to-onset of
the first-phase ratio increase caused by 0.2 and 0.5 11M MeHg but did not change the
time-to-onset Of the first phase caused by 1.0 11M MeHg (Figure 2.10). Additionally,
atropine caused a Significant decrease in the first-phase amplitude caused by 0.2 uM
MeHg (fi'om a normalized ratio of 1.21:1:0.06 to 0.73d:0.05), 0.5 uM MeHg (fi'om
1.042t0.10 to 0.67i0.09), and 1.0 11M MeHg (from 08810.05 to 0.54i0.06) (Figure 2.11).
This represents a decrease in the first-phase ratio amplitude of approximately 40%, 36%
and 39% caused by 0.2, 0.5 and 1.0 uM MeHg, respectively.

Effect of ryanodine on MeHg-induced [Ca2+]. elevations. Granule cells in
culture exhibit a developmental profile in the ability of their ryanodine receptors to
respond to agonists, such that after 4 days in culture, they lose the ability to respond to
ryanodine receptor agonists such as caffeine (Irving et al., 1992); unpublished
observation). However, these receptors are still functional and participate in CICR, as
evidenced by the ability of 10 11M ryanodine to inhibit CICR initiated fi'om 1P; receptors
in granule cells in culture (Irving et al., 1992). Because all experiments in this
dissertation were used at days 6-8 in vitro, it was impossible to test directly the ability of
ryanodine, and thus experiments were performed to test the ability of 10 uM ryanodine to
inhibit CICR initiated by release of Ca2+ fi'om the 1P3 receptors when activated with 1
mM carbachol for 1 min. In agreement with Irving et al., (1992) ryanodine inhibited
elevations of [Ca2+], initiated by release of Ca2+ through the 1P3 receptor, as evoked by
the muscarinic agonist BCh (Table 2.1). Treatment with 10 11M ryanodine for 10 min
prior to, as well as during, treatment with 0.2-1.0 11M MeHg, did not change the time-to-

onset of the first-phase ratio increase for 0.2, 0.5 or 1.0 11M MeHg (Figure 2.12A).

Ryanodine also did not delay the time-to-onset of the second-phase Ca2+ influx caused by
0.2, 0.5 or 1.0 pM MeHg (Figure 2.123). In the nominally Ca2+-f1’ee buffer, ryanodine
again did not change the time-to-onset of the first-phase ratio increase caused by 0.2, 0.5
or 1.0 11M MeHg (Figure 2.13). Ryanodine also failed to decrease amplitude of the first-
phase ratio increase caused by MeHg at any concentration examined (Figure 2.14).
Contribution of store-operated Ca2+ entry (SOCE). Experiments were
performed to determine whether MeHg-induced emptying of the SER triggers opening of
store-Operated Caz+.channels (SOCC), thereby contributing to the second-phase influx of
Ca2+c. As shown in Figure 2.15, treatment with 10 11M thapsigargin for 10 min, followed
by 1 mM EGTA dissolved in HBS, causes a noticeable decline in the baseline fura-2
ratio. Immediately following replacement of the EGTA solution with HBS (which
contains 1.8 mM Ca2+), the baseline firra-2 ratio returns to near-pretreatment levels,
which is consistent with opening of SOCCS. Experiments were then performed to
determine whether the return to normal [Ca2+]; was due to opening of SOCCS. The
socc inhibitor SK&F-96365 was tested for its ability to inhibit Ca2+ entry through
SOCCS. SK&F-96365 is a known inhibitor of SOCCS (Lo and Thayer, 1995); however,
SK&F-96365 is reported to also interact with voltage-operated Ca2+ channels (V OCCS),
inhibiting Ca2+ movement through these channels (Koch et al. , 1994). For this reason,
the effect of 10 uM SK&F-96365 on VOCCS was examined by adding the compound
during plasma membrane depolarization-induced influx of Ca2+c caused by l min
exposure to 40 mM K‘“. As seen in Figure 2.16, SK&F-96365 significantly decreased
Ca2+ entry through VOCCS by approximately 50%; thus, this compound could not be

used to study the contribution of SOCE to MeHg-induced influx of Ca2+e.

65

A - MeHg only - MeHg + Ry

 

 

75‘ 60

55, F

8

a

o

6

1'.‘

0

.§

i-

0.2 0.5 1.0
[MeHg] (11M)

- MeHg only - MeHg + Ry

 

 

E
E
a
t:
0
c5
[1‘
0
.§
[-1
0.2 0.5 1.0
[MeHg] (FM)

Figure 2.12. Ryanodine (“Ry”) does not affect the time-to-onset of MeHg-induced
increases in [Ca2+]r. A) Treatment with 10 11M ryanodine for 10 min prior to, as well as
during MeHg exposure, did not alter the time-to-onset of the MeHg-induced first-phase
[Ca2+]: increase caused by 0.2, 0.5, or 1.0 uM MeHg. (B) In the same cells, treatment
with 10 uM ryanodine also did not alter the time-to-onset of the MeHg-induced second-

phase influx of Ca2+.= caused by 0.2, 0.5, or 1.0 11M MeHg. Results are presented as
mean :1: SEM (n=6-7).

66

 

- MeHg only - MeHg + Ry

L»
O
1

N
O
1

   

p—d
O
I

 

 

Time-To-Onset (min)

O

 

0.5 1 .0
[MeHg] (HM)

Figure 2.13. Ryanodine (“Ry”) does not affect the time-to-onset of the MeHg-
induced first-phase release of Ca2+” as measured in the absence of Ca2+, Treatment
with ryanodine (10 uM) for 10 min prior to, as well as during, exposure to MeHg
significantly does not alter the time-to-onset of the {Cay} increase caused by 0.2, 0.5, or
1.0 11M MeHg. Results are presented as mean i SEM (n=4-7).

67

--Ry -+Ry

1.50 —
.2
a
,6 1.00 -
a)
.u
To
g 0.50
0
Z

0.00

 

   

0.2 0.5
[MeHg] (11M)

Figure 2.14. Ryanodine does not affect the amplitude of MeHg-induced increases in
[Ca2+]i as measured in the absence of Ca2+” Ryanodine caused slight decreases in the
amplitude of the MeHg-induced first-phase [Ca2+]i increase caused by 0.2, 0.5, and 1.0
uM MeHg; however, the differences were not statistically significant. Results are
presented as mean i SEM (n=4-7).

68

 

 

 

 

 

 

 

 

0.45 E 13
ii BCh Th HBS BCh
A ‘— a si ar in —
§ 0.40 - p g g EGTA ll
a
3.
El 035 -
o
.5 M) l
9‘2 0.30 -
0.25 L l l l l i l L J

0 5 10 15 20 25 30 35 40 45

Time (min)

Figure 2.15. Depletion of SER Ca2+ triggers SOCE in rat cerebellar granule cells.
Granule cells were exposed to 40 mM K+ (1 min) followed by 1 mM BCH (1.5 min) to
release Ca2+ from the IPg-Sensitive Ca2+ pool. Thapsigargin (10 11M, 10 min) followed by
a hi concentration of EGTA (1 mM, 5 min) caused a noticeable decline in the baseline
[Ca +]i. Returning the cells to a Cay-containing medium (HBS) caused an immediate
return to baseline [Ca2+]i, which is characteristic of SOCE. The loss of SER Ca2+ content
with this treatment is confirmed by the loss of response to a second exposure to 1 mM
BCh (1.5 min).

69

1.00 '

 

 

g 0.80
C6
Gd
3 0.60
.52].
7'9 0.40
E
z 0.20
0.00
- +
SKF-96365 (10 11M)

Figure 2.16. SK&F 96365 inhibits depolarization-stimulated Ca2+ influx through
VOCCs. Granule cells were exposed to 40 mM K+ for 1 min, in the presence or absence
of 10 11M SK&F-96365, and the amplitude of the peak [Cay]; response to the K+
depolarization was measured. Results are presented as mean :1: SEM (n=4). The asterisk
(*) indicates a value significantly different from the corresponding vehicle control data (p
< 0.05).

70

DISCUSSION

MeHg increases [Ca2+]; in rat cerebellar granule neurons, which contributes to
MeHg-induced neuronal death (Marty and Atchison, 1997; Marty and Atchison, 1998).
Because loss of Ca2+ regulation is a pivotal step in both apoptosis and necrosis, these
experiments sought to identify the Ca2+ source in granule cells to understand the
mechanism by which MeHg causes elevations of [Cam]; in cerebellar granule cells.
MeHg interacts with muscarinic ACh receptors (Castoldi et al. , 1996; Coccini et al.,
2000), causes generation of 1P3 in rat cerebellar granule cells in culture (Sarafian, 1993),
and causes Ca2+ release through the 1P3 receptor in NG108-15 cells and rat T
lymphocytes (Tan et al., 1993; Hare and Atchison, 1995); thus, these experiments
examined whether MeHg interacts with these receptors to cause Ca2+ release fiom the
SER.

Data presented in this chapter reveals that Ca2+ release from the SER by MeHg
contributes approximately 30-40% to the observed increase in fura-2 ratio during the
initial phase of MeHg exposure. This differs fi'om results Obtained in NG108-15 cells, in
which emptying the SER with thapsigargin and bradykinin prior to MeHg treatment
reduced the MeHg-induced frrst-phase ratio increase by 68% (Hare and Atchison, 1995).
As was found in NG108-15 cells, the ryanodine receptor Ca2+ pool in granule cells
contributes minimal Ca2+ during MeHg exposure (Hare and Atchison, 1995). All four
drug treatments (thapsigargin, BCh, atropine and ryanodine) decreased the first-phase
ratio amplitude to a Similar extent for all MeHg concentrations examined, suggesting that
MeHg activates a common signaling pathway which involves muscarinic receptors, 1P3

receptors and ryanodine receptors.

71

While pharmacological manipulation of the SER with different compounds
caused similar decreases in the amount of Ca2+ released fi'om within the cell by MeHg,
there was a complicated pattern of effects on the time-to-onset of the initial increase in
[Ca2+]i. Atropine delayed the time-to-onset of the initial [Ca2+]; elevation, suggesting that
the MeHg-induced elevation of [Ca2+]; is initiated by hiding of MeHg to muscarinic
receptors. Because granule cell M3 receptors are coupled to Gq, which leads to
generation of 1P3 through PLC, while M2 receptors are linked to G, which inhibits
adenylate cyclase and thus is not involved in Ca2+ regulation, the effect of atropine is
likely due to inhibition of M3 receptors (Whitham et al., 1991; Doble et al., 1992;
Contrera et al. , 1993; F ohrman et al. , 1993). These results support the observations that
MeHg binds multiple subtypes of muscarinic receptors in both the central and peripheral
nervous systems (Abd-Elfattah and Shamoo, 1981; Castoldi et al., 1996; Coccini et al.,
2000) and causes increased [IP3] in cerebellar granule neurons in vitro (Sarafian, 1993).
The binding of MeHg to muscarinic receptors may partially explain the selective
neurotoxicity of specific cell types within the central nervous system (CNS). In rats,
granule cells express the highest density of muscarinic receptors in the cerebellar cortex,
as assessed by [3H]quinuclidinyl benzilate binding (Neustadt et al., 1988). In vivo studies
indicate that MeHg causes an up-regulation of muscarinic receptors in both the
cerebellum and hippocampus, which tend to accumulate MeHg and exhibit neuronal
death following MeHg exposure (Coccini et al., 2000). In granule cells, inhibition of
muscarinic receptors protects against apoptosis induced by growth in non-depolarizing
conditions (Y an et al., 1995), suggesting that muscarinic receptor activation can activate

cell death in granule cells. Because the primary downstream target of M3 receptors is the

72

SER-localized 1P3 receptor, the functional consequence of MeHg activation of M3
receptors would be Ca2+ release from the SER, which would then contribute to the whole-
cell [Cazji elevation observed during MeHg exposure.
Depletion of the SER Ca2+ store with the smooth endoplasmic reticulum Ca2+
ATPase (SERCA) inhibitor thapsigargin, while decreasing the amount of Ca2+; released
into the cytosol, did not alter the time-to-onset of the initial Ca+ increase. Complete
emptying of the SER by thapsigargin would be expected to delay or prevent the MeHg-
induced first-phase increase of [Ca2+], if it originated from the SER, as is suggested by
the atropine data. However, this would only be the case if thapsigargin completely
er3.311315% the SER of its CaZ+ content. In these experiments, it was found that a 5 min
trea-tment with 10 11M thapsigargin abolished the response to treatment with 1 min 1 mM
Bell, indicating loss of responsiveness of the M3 muscarinic receptor-1P3 receptor
pat—I'lway. This was presumably due to store depletion. However, in experiments
Dex-formed with rat cerebellar granule cells in primary culture, 20 min treatment with 10
MM thapsigargin only reduced the SER Ca2+ store by approximately 50% in spite of the
cOmplete abolishment of response to 1 mM carbachol (Masgrau et al., 2000). The
authors concluded that granule cells may contain a thapsigargin-insensitive SER Ca2+
pool; this Ca2+ pool could be the initial target of MeHg, thereby contributing to the
Observed [Ca2+]; increases irrespective of thapsigargin treatment. The total SER Ca2+
pool in granule cells can be subdivided into pharrnacologically, and possibly physically,
distinct subsets, making the existence of a thapsigargin-insensitive pool plausible
(Simpson et al., 1996). Thus, the ability of thapsigargin treatment to decrease the

amplitude of the first-phase [Ca2+]; increase without affecting the time-to-onset of the

73

initial [Ca2+]; elevation suggests that the thapsigargin-sensitive store does contribute to
the overall [Ca2+]i increase without necessarily being the initial target.
In order to examine the contribution of Ca2+ released fi'om the 1P3 receptor, cells
were treated with BCh (1 mM, 24 hr) to down-regulate the SER 1P3 receptor. In granule
cells in vitro, 24 hr treatment with 0.1-1 mM carbachol is well documented as causing
down-regulation and desensitization of M2, M3 and Type I IP3 receptors (F ohrman et al.,
1 993; Fukamauchi et al., 1993; Simpson et al., 1994; Simpson et al., 1996), thus a 24 hr
pretreatment with 1 mM BCh should cause a similar down-regulation of these receptors.
me 24 hr BCh treatment caused a 77% decrease in the peak Ca2+ response to a brief 1
mm BCh treatment (Table 2.1), which agreed with the 77i14% loss of carbachol-induced
I-esponse caused by 24 hr 1 mM carbachol exposure (Simpson et al., 1996). These same
authors found that while 24 hr carbachol treatment caused a 65% reduction in agonist-
induced accumulation of 1P3, it did not abolish the response (Simpson et al. , 1996),
suggesting that 24 hr BCh treatment might inhibit but not abolish the MeHg-induced
rel ease of Caz+i if it occurs through the 1P; receptor. Accordingly, exposure to 1 mM
B Ch decreased the amplitude of the first-phase Ca2+, elevation, indicating that loss of [P3
receptors corresponded with a decreased response to MeHg. Surprisingly, BCh treatment
hastened the release of Ca2+; during MeHg exposure. The BCh treatment does not alter
the armount of Ca2+ release by preloaded mitochondria, thus the BCh treatment does not
Shi fl (3a2+ from the SER to the mitochondria. Because 24 hr carbachol treatment reduces
me number of 1P3 receptors by approximately 80% (Simpson et al., 1994) and reduces

but does not abolish the 1P3 accumulation following muscarinic receptor activation

74

(Simpson et al. , 1996), the remaining 1P3 receptors could exhibit increased activity to
compensate for the lost receptors.
Depletion Of Ca2+; stores can trigger entry of Ca2+.: via store-Operated Ca2+ entry
( SOCE) in a variety of cells, including neurons. Previous experiments indicated that
MeHg causes influx of Ca2+.: which cannot be completely blocked by voltage-gated Ca2+
channel blockers (Marty and Atchison, 1997), leading to the hypothesis that at least part
of this Ca2+ influx was through activation of SOCE. Rat cerebellar granule cells do
undergo SOCE when the SER is depleted of Ca2+ by a combination of thapsigargin and
high-EGTA medium; however, direct attempts to block SOCE with SK&F 96365 were
unsuccessful, as this compound did not inhibit SOCE while causing significant inhibition
of C 32+: entry through voltage-gated Ca2+ channels. MeHg-induced activation of SOCE
in granule cells is doubtful, however, because of the lack of effect any SER targeting
compound on the MeHg-induced influx of Cab}. Because none of the pharmacological
agents used in this study altered the MeHg-induced Ca2+ influx, this suggests (but

remains untested) that SOCE does not play a role in the observed loss of Ca2+, regulation.

75

CHAPTER THREE

ACUTE METHYLMERCURY EXPOSURE CAUSES NEURONAL DEATH IN
RAT CEREBELLAR GRAN ULE CELLS THROUGH INTERACTION WITH

INOSITOL-l,4,5-TRIPHOSPHATE RECEPTORS

76

ABSTRACT
Low concentrations of methylrnercury (MeHg) cause Ca2+-dependent cell death in rat
cerebellar granule cells in vitro. As demonstrated in Chapter Two, MeHg interacts with
muscarinic ACh receptors to cause the release of Ca2+ fiom the SER, primarily through
the inositol-1,4,5-triphosphate (1P3) receptor, in cerebellar granule cells. Experiments
presented in this chapter examined whether this Ca2+ release plays a role in MeHg-
induced cell death in granule cells. Acute exposure to increasing concentrations of MeHg
(0-2-1.0 11M) caused a corresponding increase in cell death at 24.5 hr post-exposure.
Pfior down-regulation of the M3 muscarinic receptor-1P3 receptor pathway with the
muscarinic receptor agonist bethanechol (BCh, 1 mM) provided significant protection
against MeHg-induced cell death. This protection was reversed by the non-specific
muscarinic receptor antagonist atropine (10 11M) and the M3 receptor antagonist 4-
diphenylacetoxyl-N-methylpiperidine methiodide (10 uM), but not the M2 receptor
antagonist methoctramine (10 uM) or neuronal nicotinic receptor antagonist dihydro-B-
e I'3’11111‘oidine hydrobromide (10 11M). Atropine alone (10 uM) was not protective against
MeHg-induced neuronal death. Acute or 24 hr pre-treatrnent with the [Pg-receptor
antagonist xestospongin C (XeC) (1 llM) was also not neuroprotective, with 24 hr XeC
itself causing significant cell death. Inhibition of the SER ryanodine receptors with 10
”M I'D’arrodine was also not protective, indicating that MeHg-induced Ca2+ release from
tIfiS store does not play a Significant role in cell death. These results indicate that

‘nhibition of MeHg-induced Ca2+ release from the SER protects against MeHg-induced
Cell (1

77

INTRODUCTION
Results of experiments described in Chapter Two indicate that a significant
portion of the MeHg-induced increase of [Cay]; in granule cells originates from the SER,
as released through the 1P3 receptor. This release of Ca2 from the IP3-sensitive Ca2+ pool
appears to be due to interaction of MeHg with muscarinic receptors, although the
possibility of a direct interaction between MeHg and intracellular Ca2+ channels is not
ruled out. The next set of experiments was designed to test the hypothesis that MeHg-
induced Ca2+ release from the SER plays a role in MeHg-induced cell death in cerebellar
granule cells. To test this hypothesis, pharmacological inhibitors of specific receptors
and ion channels involved in Ca2+i signaling pathways were tested for their ability to
pI-event MeHg-induced cell death. The receptors and channels examined in these
experiments include muscarinic ACh receptors, 1P; receptor-linked channels and
ryanodine receptor-linked channels.
Granule cells express a high density of M2 and M3 muscarinic receptors;
aetivation of the M3 receptor leads to generation of 1P3 by activation of PLC (Alonso et
0L , l 990; Fohrman et al., 1993; Fukamauchi et al., 1993; Simpson et al., 1994).
M ‘JScarinic receptors of all subtypes contain two critical thiol groups within the
r<=C‘P'ID1ior’s active site, suggesting that MeHg could potentially bind to, and activate, these
receli>t<ns (Castoldi et al., 1996; Coccini et al. , 2000). This interaction between MeHg
mid the muscarinic receptors may affect MeHg-induced cell death, as granule cell death
ca‘lsed by a number of toxicants is mediated by the M3 receptors (Galli et al., 1995; Yan

et
01- , 1 995; de Luca et al., 1996; Lin et al., 1997).

78

The functional consequence of activating M3 muscarinic receptors is the production

of IP; and subsequent release of Ca2+ release from the IP3—sensitive store in the SER.

There is growing evidence that the release of Ca2+ through these channels plays a critical
role in neuronal death caused by a variety of toxicants (Marks, 1997; Szalai et al., 1999;
Blackshaw et al., 2000). T cells lacking Type I 1P3 receptors are resistant to apoptosis

induced by dexamethasone, T-cell receptor stimulation, ionizing radiation, and F as
(Jayaraman and Marks, 1997). Resistance to apoptosis caused by these agents is reversed
by elevating [Ca2+]5, suggesting that the 1P3 receptor’s importance may be in the release
of Ca2+ into the cytosol rather than activation of a protein-mediated mechanism
(Jayaraman and Marks, 1997). In permeabilized HepG2 cells, Ca2+ signals from the 1P3
receptor are directly translated into elevations of mitochondrial [Ca2+], opening of the

M TP and subsequent release of the pro-apoptotic cytochrome c into the cytosol (Szalai et
al. , l 999). Thus, MeHg-induced Ca2+ release through the 1P3 receptors could be a critical

step in MeHg-induced cell death.
In addition to SER-localized 1P3 receptors, ryanodine receptors also influence

neuronal viability (Marks, 1997) and contribute a small amount of Ca2+ to the whole-cell
[Ca2+]. increase caused by MeHg (previous chapter). Inhibition of Ca2+ release fiom the
I 3 anodine receptor-activated store is protective in several in vitro and in vivo models of

net-Ironal damage (Wei and Perry, 1996; Marks, 1997; Wei et al., 2000). Thus these

I‘ece'IDtors could potentially be important in MeHg-induced neuronal death.
The purpose of these experiments was to determine whether inhibition of specific

c
olztlI><>tlents of the SER-linked Ca2+ signaling pathways altered MeHg-induced cell death

in
rat Cerebellar granule neurons in primary culture. The first set of experiments

79

examined the effect of SER store depletion on MeHg-induced cell death using the
SERCA inhibitor thapsigargin. The second set of experiments examined the effect of
down-regulation of both the 1P3 receptor and the muscarinic receptors using the
nonspecific muscarinic agonist bethanechol (BCh). Specific portions of this signaling
pathway were also examined using specific muscarinic receptor subtype antagonists and
the IF; receptor inhibitor XeC. The final set of experiments examined the role of the

Iyanodine receptors in MeHg-induced cytotoxicity using the ryanodine receptor inhibitor

ryanodine.

80

MATERIALS AND METHODS

The materials and solutions for isolation of cerebellar granule cells are described
in detail in Chapter Two. Unless specifically stated, all solutions were identical to those

described in Chapter Two. The methods and solutions described below are specific to

this chapter.

Materials and experimental solutions. The Live/Dead Eukolight viability/cytotoxicity
kit was purchased from Molecular Probes (Eugene, OR). 4-diphenylacetoxyl-N-
methylpiperidine methiodide (4-DAMP), dihydro-B-erythroidine hydrobromide (DHE),
and methoctramine hydrochloride (methoctramine) were purchased fiom Sigma (St.
Louis, MO). Xestospongin C (XeC) was purchased fiom Calbiochem (La Jolla, CA).
Cell Laocate coverslips were purchased fiom Eppendorf Scientific (Madison, WI).
Experimental solutions were prepared in sterile-filtered HBS. MeHg was
prepared as a 5 mM stock solution in deionized water and diluted to working
concentrations. Thapsigargin was prepared as a 1 mM stock in ethanol. All other
p ha-rlnacological inhibitors were prepared as stock solution in deionized water. Control

solutions contained an equal volume of vehicle in sterile HBS.
Rat cerebellar granule cell isolation procedure. Cerebellar granule cells were isolated

as described in Chapter Two and were plated at a density of 1.8x108 cells/35 mm2 dish,

ch c()ntaining a 13 mm CellLocate coverslip coated with 0.1 mg/ml poly-D-lysine.

81

C ells were maintained at 37°C in 5% C02 for 6-8 days in vitro to allow for cell

maturation (Aronica et al. , 1993).

Measurement of granule cell viability. The cell viability protocol was designed to
parallel experimental conditions in previous experiments examining MeHg-induced
e levations of [Ca2+]i in granule cells in primary culture (Marty and Atchison, 1998). To
replicate these conditions, all cells were incubated for 90 min at 37°C in HBS to mimic
the fura—2 AM incubation (1 hr) and HBS perfusion rinse (30 min) of prior [Ca2+]i
measurement experiments. Cells were then exposed to HBS (76 min), 0.2 uM MeHg (76
min), 0.5 uM MeHg (45 min) or 1.0 uM MeHg (38 min). MeHg exposure duration and
concentrations were based on previous experiments by Marty and Atchison (1997), which
determined these times sufficient for approximately 97% of cells to undergo MeHg-
induced release of intracellular Ca2+ stores and influx of Caz} For all experiments,
appropriate solvent and drug controls were performed at the same time for 76 min each,
as this was the maximum MeHg exposure interval. Following MeHg exposure, cells
were returned to conditioned medium (to bind excess MeHg) for 24 hr. Previous
ex periments indicated no significant loss of cell viability immediately following exposure
atld significant cell death at 24 hr post-exposure, thus cells were examined at 24 hr in
these experiments. At 24 hr after treatment, cells were removed from conditioned
medium, rinsed twice with HBS and overlaid with Live/Dead Eukolight
Viability/Cytotoxicity reagents (Molecular Probes, Eugene, OR) for 30 min at 37°C. As
per tile kit instructions, the reagent buffer contains 2 uM calcein acetoxymethylester

(caICQin-AM) and 4 uM ethidium homodimer which labels healthy cells green (calcein-

82

AM) and dead cells red (ethidium homodirner), both dissolved in HBS. Following the 30
min exposure, cells were examined using a Nikon Diaphot epifluorescence microscope
and the number of live and dead cells in 8 randomly selected CellLocate grids were
counted in order to determine percent viability for that coverslip. Because of the 30 min
period in Live/Dead reagent, cells were counted at 24.5 hr after the cessation of MeHg
treatment.

To determine whether emptying of the SER Ca2+ pool inhibited MeHg-induced
neuronal death, cells were treated with 10 pM thapsigargin for 5 min immediately prior
to 0-1.0 pM MeHg. Previous work demonstrated that this protocol empties the
thapsigargin-sensitive SER Ca2+ store (Simpson et al., 1996); previous chapter).

Following MeHg treatment, cells were returned to conditioned medium for 24 hr, and
counted at 24.5 h post-exposure.

To determine if BCh treatment protected cells from MeHg-induced neuronal

death, BCh was added directly to the growth medium to give a final concentration of 1
mM BCh. Similar exposure to 1 mM carbachol down-regulates the M3 receptor in
granule cells as a result of continued receptor stimulation, thus 24 hr of 1 mM BCh would
be expected to cause a similar down-regulation (Fohrman et al., 1993; Fukamauchi et al. ,
1 993 ; Wojcikiewicz et al., 1994; Simpson et al., 1996; Fukamauchi et al., 2000). At 24
hr afier application of BCh, cells were treated with MeHg alone or MeHg plus 1 mM
BCh- I" or each set of cells exposed to MeHg there were corresponding HBS, 1 mM BCh,

and Vehicle control cells which were treated and counted at the same time as the MeHg-
treated cells. For cells treated with BCh, all HBS washes contained 1 mM BCh;

additiOnally, the conditioned medium to which they were returned after MeHg treatment

83

contained 1 mM BCh for the entire 24 hr post-treatment period. In experiments in which
B Ch was added with a specific Cholinergic receptor antagonist, the antagonist was added
for 10 min prior to the initial application of 1 mM BCh, as well as during BCh treatment
and HBS wash. The antagonists tested were atropine (nonspecific muscarinic receptor
alltagonist), methoctramine (M2 antagonist), 4-DAMP (M3 antagonist) and DHE
(neuronal nicotinic ACh receptor antagonist), all at a final concentration of 10 pM. The
concentrations of each antagonist were based on the results of Yan et al. (1995) and Lin
et al. (1997), who used these compounds to inhibit cell death induced by non-
depolarizing growth medium and mastoparan, respectively, in rat cerebellar granule
neurons in primary culture. In a similar set of experiments, atropine was added directly
to the conditioned medium to a final concentration of 10 pM 10 min prior to, as well as
dming, exposure to 0-1.0 uM MeHg.

In the next experiment, the reversible IP3-receptor inhibitor XeC (1 uM) was
added for 10 min prior to, as well as during, treatment with 0-1.0 pM MeHg. This
concentration of XeC was chosen based on the results of Netzeband etal., 1999, who
found 1 uM XeC sufficient to prevent Ca2+ release from the 1P3 receptor in rat cerebellar

gratlule neurons in primary culture. Finally, cells were treated with the ryanodine

l‘e<=€=l:>t0r inhibitor ryanodine (10 pM) for 10 min prior to, as well as during, exposure to
0‘ 1 - 0 1.4M MeHg. This concentration of ryanodine has previously been shown to be
e1i‘eC-‘etive at inhibiting Ca2+-induced Ca2+ release from ryanodine receptors in cerebellar

granule cells (Irving et al., 1992).

84

Statistics. To account for variability between cell isolates, each set of experiments was
performed in at least two separate isolates. Further, all experiments within a set (HBS
only, drug only, and MeHg with or without drug) were performed on the same day (n=l),
allowing the data to be paired for statistical comparison. For each exposure group, the
percent viability (number of live cells/total number of cells x 100) was determined, and
these percentages were normalized using an angular transformation. Values are
presented as the mean i SEM. Comparisons of cell viability were made using a repeated
measures analysis of variance (AN OVA) followed by Tukey’s procedure for post-hoe
comparisons to compare MeHg to MeHg + inhibitor, with p < 0.05 considered to be

statistically significant.

85

 

RESULTS
These experiments were designed to examine the role of SER Ca2+ in MeHg-

induced death of cerebellar granule neurons. In the first set of experiments, the effect of
MeHg on granule cell viability was determined both immediately and 24 hr after
exposure to 0-2 pM MeHg. Cell viability is similar to control immediately following

MeHg treatment; however, at 24 hr postoexposure MeHg causes increasing cell death in a
concentration-dependent manner (Figure 3.1). These results confirm earlier findings by
Marty and Atchison (1998). Because MeHg does not immediately cause neuronal death,

all cell viability assays were performed at 24 hr post-exposure to MeHg.

The next experiment sought to determine whether emptying of the SER Caz+ pool
with the SERCA inhibitor thapsigargin was protective against cell death at 24.5 h post-
exposure to 0-1 uM MeHg. Cells were treated with 10 pM thapsigargin for 5 min
immediately prior to MeHg; this exposure was based on earlier experiments in which this
treatment depleted the SER of its Ca2+ content (previous chapter). However, this brief

exposure to thapsigargin was itself toxic, causing cell death at 24.5 h post-exposure
(results not shown).

In the next set of experiments, the nonspecific muscarinic agonist BCh (1 mM)

was applied for 24 hr prior to MeHg treatment to down-regulate the muscarinic-1P3
I'e<=el:>tor pathway. Application of 1 mM carbachol is well documented as causing down-
regulation and desensitization of M2, M3 and subsequently, 1P3 receptors in granule

nemons (Fohrman et al., 1993; Fukamauchi et al. , 1993; Wojcikiewicz et al., 1994;

Simpson et al., 1996; Fukamauchi et al., 2000). Treatment with 1 mM BCh significantly

86

- Immediate % 24hr

% Viability

 

 

MeHg (PM)

Figure 3.1. MeHg-induced cell death at 30 min and 24 hr post-exposure.
Immediately following exposure to 0.2-2.0 uM MeHg, there is no apparent loss of
g1‘ar1ule cell viability. However, at 24 hr post-exposure, there is a concentration-
dependent increase in cell death. Results are presented as mean :t SEM (n = 4).
( * )significantly different from the corresponding 0 uM MeHg data (p < 0.05).

87

inhibits agonist-induced release of Ca2+ through granule cell 1P3 receptors (previous
chapter), suggesting that BCh, like carbachol, also causes down—regulation of the
receptors in this pathway.
In the first experiment, cells were treated with 1 mM BCh for 24 hr, followed by
O— 1 uM MeHg for the durations described earlier. As seen in Figure 3.2, increasing
concentrations of MeHg killed an increasing percentage of cells at 24.5 h post-exposure.
'I'he loss of cell viability was consistent with that observed in previous experiments
(Marty and Atchison, 1998). Treatment with 1 mM BCh significantly protected cells
fiom MeHg-induced cell death at all concentrations examined. Furthermore, at 0.2 and
O - 5 pM MeHg the incidence of cell death was equal to that of untreated cells. However,
at 1 uM MeHg, the protection provided by 1 mM BCh was not as great as at the lower
M eI-Ig concentrations. To determine whether the protection provided by BCh was due to
action at muscarinic receptors, the nonspecific muscarinic antagonist atropine (10 uM)
was applied for 10 min prior to BCh, as well as during the 24 hr BCh treatment for 24 hr
prior to exposure to MeHg. Cell viability was again measured at 24 hr post-MeHg
exposure. As seen in Figure 3.3, addition of atropine prevented the protection afforded
by 1 mM BCh (Figure 3.2), resulting in no significant difference in the number of cells
surviving MeHg alone or MeHg plus BCh and atropine.
Granule cells express M2 and M3 receptors (Alonso et al., 1990), thus the
prOtection provided by BCh could be through either or both of these receptors. In a
Similar protocol to the BCh plus atropine protocol, the M2 antagonist methoctramine (10

RM) Was applied for 10 min prior to, as well as during, treatment with 1 mM BCh.

88

- MeHg - MeHg + BCl

 

 

 

100 —
80
.23?
=1 60
7%
> 40
°\°
20
o
HBS o 0.2 0.5 1.(
lMeHglmM)

Figure 3.2. 24 hr pretreatment with 1 mM BCh protects against MeHg-i1

d ea th. Granule cells exhibit a concentration-dependent decrease in survival a
post-exposure to increasing concentrations of MeHg. Cell death is attenuated
pre—exposure to 1 mM BCh. Results are presented as meaniSEM (n=4). (*)5
difl‘erent fi'om MeHg-only pair (p < 0.05).

89

- MeHg - MeHg+BCh+AP

% Vi bili

 

 

HBS 0 0.2 0.5 1.0

 

[MeHg] (uM)

Figure 3.3. Atropine plus BCh does not protect against MeHg-induced cell death.
In contrast to the protection afforded by BCh alone (Figure 3.2), application of 10 uM
atI-opine for 10 min prior to, as well as during, the 1 mM BCh pre-exposure period is not
protective against cell death caused by 0.2-1.0 uM MeHg. Results are presented as
meamiSEM (n = 6).

9O

However, application of BCh and methoctramine alone for 24 hr resulted in a significant
increase in the incidence of cell death, thus preventing examination of whether the
protective effect of BCh was due to action at the M2 receptors (data not shown).
Application of the M3 receptor antagonist 4-DAMP (10 uM) did prevent any protection
by 1 mM BCh in a manner similar to atropine (Figure 3.4). Again, the lack of protection
was complete at lower MeHg concentrations (0.2 and 0.5 uM) and partial at the higher
MeHg concentration (1 uM). The actions of BCh were through muscarinic ACh
receptors and not nicotinic ACh receptors, as the nicotinic receptor antagonist DHE (10
uM) did not affect the neuroprotective effects of BCh (Figure 3.5).

The 24 hr BCh treatment down-regulated both the M3 receptor and the IP3-
receptor, thus the next experiment sought to determine whether preventing MeHg from
interacting with the M3 receptor would protect the cells fiom dying. To this end, cells
were exposed to atropine alone (10 uM) for 10 min prior to, as well as during, exposure

to MeHg. Cell viability was again measured at 24.5 h post-exposure. Unlike the 24 hr
BCh pretreatment, atropine alone did not protect against MeHg-induced cell death
(Figure 3.6). This is in contrast to its ability to decrease significantly MeHg-induced
elevations in [Ca2+]; cerebellar granule neurons (previous chapter).

Because atropine alone was not neuroprotective, the next step was to block the
1P 3-receptor during MeHg exposure to determine whether MeHg interacts directly with
this receptor in granule cells. The reversible IP3-receptor inhibitor XeC (1 uM) was
applied for 10 min prior to, as well as during, exposure to MeHg. As seen in Figure 3.7,
acute application of XeC also did not prevent MeHg-induced cell death. Pretreatment for

24 hr with 1 uM XeC, which would presumably down-regulate

91

- MeHg - MeHg+BCh+4D

 

 

100 -
80 -
A?
a 60
23
> 40
g
20
0

HBS O 0.2 0.5 1.0

 

[MeHg] (HM)

Figure 3.4. 4-DAMP plus BCh does not protect against MeHg-induced cell death.
In contrast to the protection afforded by BCh alone (Figure 3.2), application of the M3-
specific antagonist 4-DAMP for 10 min prior to, as well as during, the 1 mM BCh pre-
exposure period does not protect against MeHg-induced cell death at 24 hr post-exposure.
Results are presented as meaniSEM (n=4).

92

- MeHg - MeHg+BCh+DHE
100 ‘

% Viability

 

 

HBS O 0.2 0.5 1.0

 

[MeHg] (HM)

Figure 3.5. The nicotinic receptor antagonist DHE does not affect BCh-mediated
protection against MeHg-induced cell death. The protective effect of 1 mM BCh was
not due to action at nicotinic ACh receptors, because inclusion of the nicotinic receptor
antagonist DHE (10 uM) for 10 min prior to, as well as during, MeHg exposure did not
alter the effect of BCh. Results are presented as meaniSEM (n=8). (*)Significantly
different from MeHg-only control (p < 0.05).

93

- MeHg - MeHg+Atropine
100 ‘

% Viability
A O\ OO
O O O

N
O

 

 

O

HBS O 0.2 0.5 1.0

 

[MeHg] (HM)

Figure 3.6. Atropine alone does not protect against MeHg-induced cell death. A
brief pretreatment with atropine (10 min, 10 uM), as well as treatment with atropine
during MeHg exposure, did not affect MeHg-induced cell death at 24.5 h post-exposure.
Results are presented as meaniSEM (n=6).

94

- MeHg - MeHg+XeC

100 ‘
80
6O
40
20

O

% Viabili

 

 

HBS 0 0.2 0.5 1.0

 

[MeHg] (HM)

Figure 3.7. Xestospongin C does not protect against MeHg-induced cell death.
Pretreatment with xestosponginc C (XeC) (10 min, 1 uM), as well as inclusion of XeC in
the MeHg solution, did not affect MeHg-induced cell death at 24.5 h post-exposure.
Results are presented as meaniSEM (n=4).

95

the 1P3 receptors, was also not neuroprotective and was itself neurotoxic (data not
shown).

Finally, the participation of ryanodine receptors in MeHg-induced neuronal death
was examined. Previous results indicated that inhibition of Ca2+ release through the
ryanodine receptors did not affect MeHg-induced increases of [Ca2+]i in granule cells
(previous chapter). To test whether ryanodine receptors are involved in MeHg-induced
cell death, granule cells were exposed to 10 pM ryanodine for 10 min prior to, as well as
during, exposure to MeHg, and again cell viability was determined at 24.5 h post-
exposure. In results similar to those of the Ca2+ imaging experiments, ryanodine did not

protect against MeHg-induced neuronal death (Figure 3 .8).

96

- MeHg - MeHg + Ry

 

   

100-

3‘ 80

:3 60

.53

> 40

O

°\ 20
0

HBS O 0.2 0.5 1.0
MeHg (11M)

 

Figure 3.8. Ryanodine does not protect against MeHg-induced cell death.
Application of 10 uM ryanodine for 10 min prior to, as well as during, exposure to MeHg
did not protect against cell death at 24.5 hr post-exposure to 0.2, 0.5 or 1.0 uM MeHg.
Results are presented as meaniSEM (n=4).

97

DISCUSSION

The purpose of these experiments was to elucidate the mechanism by which
MeHg causes neuronal death in cerebellar granule cells which are highly sensitive to
MeHg. The experiments focused on Ca2+ signaling pathways which converged at the
SER Ca2+ pool, as this Ca2+i store has recently been implicated in MeHg-induced loss of
[Ca2+]; homeostasis in granule cells (previous chapter). While there has been evidence
for MeHg interacting with various components of this signaling network, including
muscarinic receptors, 1P3 receptors and ryanodine receptors (Hare and Atchison, 1995;
Castoldi et al., 1996; Candura et al., 1997; Coccini et al., 2000), this is the first report in
which cytotoxicity in cerebellar granule cells resulting fi'om specific interactions of
MeHg with these receptors is examined.

In the first set of experiments, the SERCA inhibitor thapsigargin was used to
examine the effect of depleting the SER of its Ca2+ content prior to MeHg exposure. Not
surprisingly, thapsigargin treatment was itself neurotoxic, suggesting that disruption of
SER Ca2+ regulation can induce neuronal death in granule cells. Thapsigargin-induced
cell death could be due to a thapsigargin-induced increase in cytosolic [Ca2+] or induction
of a SER “stress” response (Mattson et al., 2000); however, the mechanism of
thapsigargin-induced cell death was not explored further.

In the next set of experiments, 24 hr BCh was used to down-regulate the
muscarinic and 1P3 receptors for 24 hr prior to MeHg exposure (F ohrrnan et al., 1993;

F ukamauchi et al., 1993; Simpson et al., 1994). Similar treatment with 1 mM carbachol
is well documented as causing down-regulation and desensitization of both the M3 and

1P3 receptors in rat cerebellar granule neurons (Fohrman et al., 1993; F ukamauchi et al. ,

98

1993; Wojcikiewicz et al., 1994; Simpson et al. , 1996; Fukamauchi et al., 2000). As
described in the previous chapter, 24 hr treatment with 1 mM BCh caused loss of granule
cell responsiveness to subsequent muscarinic receptor stimulation by 1 mM BCh,
suggesting that BCh has the same effect as carbachol. BCh was chosen based on its
higher affinity for muscarinic receptors and lesser interaction with nicotinic ACh
receptors than carbachol. In this study, 24 hr pretreatment with 1 mM BCh provided
significant protection against MeHg-induced neuronal death at 24.5 hr post-exposure,
indicating the importance of this pathway in the neurotoxicity of MeH g. Experiments
using muscarinic receptor-specific antagonists suggest that the protective effect of BCh is
mediated through the M3 receptor, asthe protection provided by BCh is reversed by the
M3 -prefering antagonist 4-DAMP. The protection is also not mediated through an effect
on nicotinic ACh receptors, as it is not reversed by DHE. However, the contribution of
M2 receptors was not determined due to the toxicity of the M2 antagonist methoctramine
itself and cannot be ruled out by these experiments.

Experiments were then performed in an attempt to differentiate between MeHg
interactions with muscarinic receptors and with the 1P3 receptor. Atropine alone did not
protect against MeHg-induced neurotoxicity, which was surprising given the known
interactions of MeHg with muscarinic receptors. As described in Chapter Two, atropine
delayed the onset of MeHg-induced elevations of [Ca2+]i. Additionally, both in vivo and
in vitro studies suggest that MeHg binds to multiple subtypes of Cholinergic receptors,
inhibiting radioligand binding to both nicotinic and muscarinic receptors in a competitive
fashion (Zhang and Wu, 1994; Castoldi et al., 1996; Coccini et al., 2000). At two weeks

after a 16 day exposure to low concentrations of MeHg (0.5-2.0 mg/kg/day), rats exhibit

99

an increase in the number of muscarinic receptors without a change in receptor affinity in
both the cerebellum and hippocampus, suggesting receptor up-regulation to compensate
for loss of functional receptors via irreversible binding by MeHg (Coccini et al., 2000).
Autoradiography of MeHg accumulation in mice indicates that MeHg accumulates
preferentially in these two brain regions (Berlin and Ullberg, 1963), suggesting a possible
correlation between the selective neurotoxicity of MeHg within specific areas of the
CNS, and location of specific muscarinic receptor subtypes. However, the results
presented here suggest that the binding of MeHg to muscarinic receptors does not explain
the selective targeting of neurons during toxicity, as block of these receptors with
atropine did not protect against cell death. Alternatively, MeHg may have higher affinity
for muscarinic receptors than does atropine, resulting in displacement of atropine by
MeHg and subsequent promotion of neurotoxicity. MeHg competes with M1 - and M2-
prefering antagonists in rat cortical membranes, and demonstrates higher affinity for the
M1 receptor (ICso = 3.4 uM) vs. the M2 receptor (IC50 = 149 uM) (Castoldi et al., 1996).
MeHg also binds to cholinergic (presumably nicotinic) receptors isolated fiom Torpedo
with high affinity and blocks binding of [3H]acetylcholine to the receptor (Eldefrawi et
al., 1977). In guinea pig intestinal smooth muscle, MeHg inhibits contractions induced
by stimulation of cholinergic nerves or by external application of ACh, suggesting a high
affmity for muscarinic receptors (Fukushi and Wakui, 1985). Because MeHg cannot be
removed from cholinergic receptors by mere wash with a MeHg-free solution, it is very
likely that MeHg has higher affinity for muscarinic receptors than does atropine;
however, the ability of MeHg to replace atropine on the muscarinic receptor has not yet

been examined directly.

100

Attempts to block the 1P3 receptor with XeC also did not prevent MeHg-induced
cell death. Given the protective effect of BCh and its reversibility by atropine, this result
was also surprising. The 1P3 receptor has been implicated in MeHg-induced loss of
[Ca2+], homeostasis in NGlO8-15 neuroblastoma cells (Hare and Atchison, 1995), T cells
(Tan et al., 1993), and cerebellar granule neurons (Sarafian, 1993); previous chapter.
Indeed, T cells that are deficient in 1P3 receptors are resistant to apoptosis induced by a
number of agents (Jayaraman and Marks, 1997; Marks, 1997). In cerebellar granule cells
in culture, neurons which have a relative deficiency in 1P3 receptors are resistant to low-
[K+]- induced apoptosis (Oberdorf et a1. , 1997), suggesting that this receptor plays a
critical role in granule cell death. Given that MeHg does increase [IP3], and that down-
regulation of the M3 and 1P3 receptors protects against MeHg-induced neuronal death,
the lack of protection afforded by xestospongin C may reflect a higher affinity for the 1P3
receptor by MeHg. Alternatively, the results suggest that MeHg-induced Ca2+ release
through the 1P3 receptor does not contribute to MeHg-induced neurotoxicity, although
this conclusion is not supported by the 24 hr BCh data.

Inhibition of Ca2+ release through the ryanodine receptor did not protect against
MeHg-induced cell death. This was not surprising, given that ryanodine receptors
contribute only a fraction of the [Ca2+]i increase observed in granule cells following
MeHg exposure (previous chapter). Ryanodine receptors also do not contribute
significantly to MeHg-induced [Ca2+]t increases in NG108-15 cells (Hare and Atchison,
l 995), suggesting that these receptors do not play a significant role in MeHg-induced

neurotoxicity. Most likely, their primary contribution is through participation in Ca2+-

induced Cay-release originating in IP3 receptors and/or mitochondria.

101

In summary, the experiments described in Chapters Two and Three indicate that
MeHg causes Ca2+ release from the SER which contributes to MeHg-induced cell death
of rat cerebellar granule cells. While the SER does not contribute the majority of the
Ca2+i released during exposure, inhibition of this Ca2+ release is neuroprotective. As will
be discussed in greater detail in Chapter Pour, the real significance of the release of SER
Ca2+ may be in its effects on mitochondria by contributing to mitochondrial Ca2+

dysregulation and thereby promoting of cell death through mitochondrial dysfrmction.

102

CHAPTER FOUR

DISRUPTION OF MITOCHONDRIAL CALCIUM SIGNIFICANTLY

CONTRIBUTES TO METHYLMERCURY-INDUCED ELEVATIONS OF [Cay]:

103

ABSTRACT

The contribution of Ca2+.“ to whole-cell [Ca2+]i elevations caused by acute
exposure to low concentrations of methylrnercury (MeHg) was examined in rat cerebellar
granule cells. In granule cells loaded with fura-2 to observe changes in whole-cell
[Ca2+]i, prior depletion of Ca2+... using CCCP and oligomycin significantly decreased the
amplitude of [Ca2+]i release from intracellular stores, as well as delayed the time-to-onset
of whole-cell [Ca2+]i elevations, caused by 0.5 uM MeHg. Oligomycin delayed the
MeHg-induced influx of Ca2+c, suggesting that ATP depletion triggers Ca2+ influx. In
granule cells loaded with rhod-2, MeHg exposure caused a biphasic increase in
fluorescence. The initial increase in fluorescence was not dependent on Ca2+¢ and was
abolished by mitochondrial depolarization, indicating uptake of Ca2+ fi'om a non-
mitochondrial Ca”; source. The secondary increase was associated with spreading of the
dye fi'om a punctate staining to whole-cell distribution, and was significantly delayed by
thapsigargin and CsA, suggesting that Ca2+ released from the SER contributes to MeHg-
induced opening of the MTP. It is concluded that MeHg causes an initial release of Ca
from the SER which is buffered by mitochondria, leading to elevated [Ca2+]m and

opening of the MTP in rat cerebellar granule neurons.

104

INTRODUCTION

As discussed in the previous chapters, acute exposure to low concentrations of
MeHg causes release of Ca2+ from at least one intracellular store into the cytosol.
Depletion of SER Ca2+ content reduced the amplitude of the MeHg-induced Ca2+i release
by less than half; thus, at least one other intracellular store contributes to the MeHg-
induced [Ca2*]i elevations. The next two chapters discuss experiments designed to
examine the contribution of Ca2+.“ to MeHg-induced increases in [Ca2+]i and cell death.

Mitochondria are “excitable organelles” that can accumulate a large amount of
Ca2+ within the lumen and, upon excess accumulation, release Ca2+m into the cytosol via
reversal of the uniporter and/or opening of the MTP (N icholls, 1978; Thayer and Miller,
1990; Budd and Nicholls, 1996b). Not only does disruption of Ca2+In contribute to
opening of the MTP and release of pro-apoptotic factors (such as cytochrome c) into the
cytosol, but release of sequestered Ca2+ contributes to cytosolic elevations of {Cay}
which promotes cell death through activation of Cay-dependent proteases (calpains and
caspases), phospholipases and endonucleases which degrade cellular protein and DNA
(Cohen and Duke, 1984; Orrenius and Nicotera, 1994; Waring and Sjaarda, 1995; Moran
et al., 1999). Recent studies indicate that elevation of Ca2+“. is a critical step during a
variety of neurotoxic insults, including glutamate excitotoxicity, generation of reactive
oxygen species, and exposure to heavy metals, including manganese and mercury
(Chavez and Holguin, 1988; Chavez et al., 1989; Richter and Kass, 1991; Nicholls and
Budd, 1998; Gavin et al., 1999; Kruman and Mattson, 1999; Verity, 1999)

Several parameters of mitochondrial fimction are affected by MeHg. MeHg binds

to mitochondria with high affinity, both in vivo and in vitro (Yoshino et al., 1966; Chang

105

and Hartmann, 1972; O'Kusky, 1983). Further, MeHg causes irreversible loss of
mitochondrial membrane potential in human monocytes (InSug et al. , 1997), isolated
nerve terminals (Bondy and McKee, 1991), NG108-15 neuroblastoma cells (Hare et al.,
1993), and guinea pig cerebral cortex synaptosomes (Kauppinen et al., 1989). Inhibition
of Ca2+... uptake and release inhibits MeHg-induced increases in the frequency of
spontaneous neurotransmitter release in hemidiaphragm nerve terminals (Levesque and
Atchison, 1988), and attenuates the uptake of 45Ca and release of 45Ca2+ fi'om preloaded,
isolated mitochondria Ca2+ uptake (Levesque and Atchison, 1991). MeHg is also well
documented as inhibiting mitochondrial respiration, decreasing cellular ATP content and
causing production of reactive oxygen species, all of which can contribute to
mitochondria-triggered cell death (Kauppinen et al., 1989; Sarafian et al., 1989;
Levesque and Atchison, 1991; Sarafian and Verity, 1991; Yee and Choi, 1996). Because
the Ca2+; chelator BAPTA protects against MeHg-induced cell death in rat granule cells
in vitro, suggesting a Ca2+-dependent process (Marty and Atchison, 1998), there is a
strong possibility that MeHg-induced changes in Ca2+m may be critical in MeHg-induced
neuronal death.

The experiments described in this chapter examined whether mitochondria
contribute Ca2+m to the first-phase [Ca2+]i increase during acute exposure to low
concentrations of MeHg. Because alterations in cellular and mitochondrial Ca2+
regulation are critical events in both apoptosis and necrosis, the experiments were
designed to characterize the effects of MeHg on Ca2+", in the intact neuron. The first
approach was to determine the relative contribution of Ca2+m to whole-cell elevations of

[Caz+]t to determine whether mitochondria are a significant target in these cells. The next

106

experiments examined the effect of MeHg on regulation of Ca”... in the intact cell,
including interactions between intracellular Ca2+ stores in the mitochondria and SER
during MeHg exposure. The results presented here suggest that MeHg targets both the
mitochondria and SER during MeHg exposure, with Ca2+ release from the SER
contributing to Ca2+ release from the mitochondria via opening of the MTP. Further,
these experiments indicate that Ca2+m contributes the bulk of the cytosolic [Ca2+]i
observed during MeHg exposure, highlighting the importance of mitochondria during

MeHg neurotoxicity.

107

on;

MATERIALS AND METHODS

The materials and solutions for isolation of cerebellar granule cells and
measurement of changes in [Ca2+]i are described in detail in Chapter Two. Unless
specifically stated, all solutions were identical to those described in Chapter Two. The

methods and solutions described below are specific to this chapter.

Materials and solutions. Rhod-2 acetoxymethylester (rhod-2 AM) and
tetramethylrhodamine ethyl ester (TMRE) were purchased fiom Molecular Probes
(Eugene, OR). Cyclosporin A (CsA), CCCP and oligomycin were purchased from Sigma
(St. Louis, MO). For all experiments, pharmacological agents are dissolved in either
HBS or EGTA-HBS, with appropriate controls for any additional solvents used (DMSO
or ethanol, with maximum solvent of 0.01% (w/v) for all experiments). In a cell-free

system, none of the pharmacological agents altered fura-2 fluorescence.

Measurement of [Ca2+]i. Cells were loaded with fura-2 for 1 hr, and rinsed in HBS for
30 min, as described in Chapter Two. Changes in cellular fluorescence were measured
using the IonOptix system described in Chapter Two. As before, following exposure to
MeHg, the times-to-onset of the first and second phases of increased Ca2+, were
determined. Additionally, in experiments performed in the nominally Ca2+-free buffer
(EGTA-HBS), the amplitude of [Caz] release caused by MeHg was normalized to the
peak fluorescence ratio amplitude of the 1 min exposure to 40 mM K+ to provide a
“normalized ratio” relative to K+-induced depolarization response. For both time-to-

onset and normalized fluorescence ratio, the data fiom each trial were averaged to

108

provide a mean time-to-onset or mean normalized ratio for that dish of cells (n=1). In
order to minimize differences between cell isolates, experiments using MeHg alone and
MeHg with a given pharmacological agent were run on the same day, and experiments

using the same agents were replicated using at least two separate cell isolates.

Measurement of [Ca2+]... To measure changes in [Ca2+]m, cells were loaded with 1 uM
rhod—2 AM for 1 hr at room temperature (23-25°), followed by overnight incubation in
normal culture medium (18-24 hr) (Trost and Lemasters, 1997; Peng et al., 1998;
Simpson and Russell, 1998a). Because rhod-2 will fluoresce in both the cytosol and
mitochondria, this protocol was used to ensure that mitochondrial loading occurred and
provide time for the cytosolic dye to leak out of the cell. Cells were imaged as in the
fura-2 experiments, except that the excitation wavelength was 540 nm and the emission
wavelength was 590 nm. Efficiency of mitochondrial loading was confirmed using an 8
min exposure to 5 uM CCCP to uncouple oxidative phosphorylation, and 10 pM
oligomycin to dissipate \ym; this combination prevents mitochondrial Ca2+ uptake during
Ca2+ influx caused by K+ depolarization (40 mM), thereby abolishing changes in rhod-2
fluorescence. Like fura-2, rhod-2 data are analyzed for time-to-onset of fluorescence
changes, with the time-to-onset of the initial elevation determined as the point at which
fluorescence rose above baseline, and the secondary elevation (from spreading of rhod-2
fluorescence fi'om a punctate stain to whole-cell fluorescence) as the point at which there
is a sharp increase in rhod-2 fluorescence. The times-to-onset of each phase of
fluorescence were calculated for all cells monitored within an experiment and were

averaged to provide mean time-to-onset for that dish of cells (n=1).

109

Measurement of ‘1'... The fluorescent dye tetramethylrhodamine ethyl ester (TMRE)
was used to observe changes in ‘I’m. TMRE preferentially loads into mitochondria with
normal membrane potentials (around —160 mV) and loses fluorescence when ‘Pm
becomes depolarized (F arkas et al. , 1989). Granule cells were loaded with 100 nM
TMRE in HBS for 30 min at 37°C, then rinsed in 10 nM TMRE in HBS for 30 rrrin.
Cellular fluorescence was then measured continuously using the IonOptix system. In
these experiments, perfusion solutions contained 10 nM TMRE to maintain dye
availability for the mitochondria. Fluorescence was measured relative to an initial
fluorescence of 100%. Control experiments indicated that TMRE fluorescence remained

stable for at least 1 hr of observation (data not shown).

Statistics
Comparisons of mean time-to-onset for “MeHg” vs. the corresponding “MeHg
plus inhibitor” cells were made using Student’s paired t-test, with values of p<0.05

considered to be statistically significant.

110

RESULTS

Neonatal granule cells in culture exposed to a low concentration of MeHg
undergo a characteristic biphasic increase in [Ca2+]i, with the first phase due to release of
Ca2+ fi'om at least one intracellular store, and the second phase due to influx of Ca2+:
(Marty and Atchison, 1997). To examine the contribution of the mitochondrial Ca2+
store, the mitochondria were depolarized with 5 uM CCCP and 10 uM oligomycin for 8
min immediately prior to exposure to 0.5 uM MeHg. Because mitochondria accumulate
Ca2+ only when their inner membranes are sufiiciently hyperpolarized, this depolarization
causes both release of Ca2+“. and prevention of further Ca2+ uptake (Budd and Nicholls,
1996b). Monitoring cells using TMRE fluorescence confirmed that this protocol was
sufficient to depolarize completely the mitochondria (Figure 4.1A). The brief increase in
TMRE fluorescence intensity is probably due to mitochondrial swelling, rather than
mitochondrial membrane hyperpolarization, which accompanies the increased
permeability of the inner mitochondrial membrane and subsequent increase in lumen
volume (White and Reynolds 1996; Trost and Lemasters 1997). Further, experiments
using cells loaded with fura-2 indicated that this protocol caused a large increase in
[Ca2+]; (Figure 4.1B). Following 8 min exposure to CCCP plus oligomycin, the cells did
not respond to another application of CCCP plus oligomycin, indicating the irreversible
loss of mitochondrial membrane potential, and successful depletion of Ca2+m, with this
protocol (Figure 4.1C). Treatment with 10 uM oligomycin alone caused a more gradual
loss of mitochondrial membrane potential, with a 75-85% loss of TMRE fluorescence at
20 min of exposure; TMRE fluorescence could be lowered further with 5 uM CCCP,

indicating that oligomycin alone did not cause complete mitochondrial membrane

111

CCCP+Oligo

 

 

 

 

 

 

 

 

 

A a
5
e’
(U
v
0
O
a
d)
O
U)
Q)
l—
O
.2
LL
Time (min)
B o 60
fé Ki
ED 0 40 CCCP+Olrgo
.9.
H
N
at
0.20 a 1 a 4
o 5 to 15 20 25
Time (min)
C 0.60 -
as CCCP+OIigo CCCP+Oligo
'3 0.50 —
3.
K W
v
o
E. 0.40
0.30 l l l
0 8 16 24 32
Time (min)

Figure 4.1. CCCP and oligomycin cause a rapid decrease in TMRE fluorescence
with a concomitant increase in fora-2 fluorescence. A) Changes in TMRE
fluorescence in a single cerebellar granule neuron in culture exposed to 5 uM CCCP and
10 uM oligomycin for 8 min. CCCP and oligomycin cause a rapid decrease in TMRE
fluorescence (as the mitochondrial membrane becomes depolarized). In granule cells, the
rapid loss of Van caused by CCCP and oligomycin is irreversible for up to 60 min post-
treatrnent. TMRE fluorescence is measured relative to the initial fluorescence intensity.
B) Changes in fura-2 fluorescence (“Ratio (F 340330)”) in a single cerebellar granule
neuron exposed to 5 pM CCCP and 10 uM oligomycin for 8 min. Mitochondrial
membrane depolarization causes a large-amplitude increase in [Ca2+]i which recovers to
baseline [Ca2+]. by the end of the 8 min exposure. C) Changes in fura-2 fluorescence
(“Ratio (F 340330)”) in a single cerebellar granule neuron exposed to 5 uM CCCP and 10
pM oligomycin for 8 min, followed by 10 min HBS and another 8 min application of 5
uM CCCP and 10 uM oligomycin. The second exposure to these compounds does not
result in additional increases in fura-2 ratio, indicating that a single application is
sufficient to deplete Ca2+m.

112

'v.
o

Oligomycin

 

é

.0
m
o

Fluorescence (arb. units)

8

 

 

 

20

10

 

 

 

 

 

 

o 30 40 so 60
Time (min)
A
g 1.50 — CCCP
C‘.
:3
.23
3 1.00
v
8
5
o 0.50 t
(D
23.
S
_ 000 r r r r
‘1‘ o 5 10 15 20
Time (min)
C 0.60 r
A K‘
§
§ 0.40 ‘ CCCP
‘1‘!“
v
0
0m 1 L L I
o s 10 15 20
Time (min)

Figure 4.2 Effect of oligomycin on TMRE fluorescence, as well as effect of CCCP on
TMRE and fora-2 fluorescence. A) Changes in TMRE fluorescence in a single
cerebellar granule neuron in culture exposed to 10 pM oligomycin for 45 min.
Oligomycin causes a gradual loss of TMRE fluorescence which is further lowered by the
addition of 5 uM CCCP, indicating that oligomycin alone does not completely depolarize
the mitochondrial membrane. TMRE fluorescence is measured relative to the initial
fluorescence intensity. B) Changes in TMRE fluorescence in a single cerebellar granule
neuron in culture exposed to 5 uM CCCP for 8 min. CCCP alone causes a rapid decrease
in TMRE fluorescence which does not recover during wash with CCCP-free HBS.
TMRE fluorescence is measured relative to the initial fluorescence intensity. C)

Changes in fora-2 fluorescence (“Ratio (F34o,3go)”) in a single cerebellar granule neuron
exposed to 5 uM CCCP for 8 min. CCCP alone causes a much smaller increase in [Ca2+];
than CCCP and oligomycin together (see Figure 4.1B).

113

depolarization (Figure 4.2A). Application of 5 uM CCCP for 8 min did cause rapid,
complete loss of TMRE fluorescence which did not recover after removal of CCCP and
wash with HBS, indicating that CCCP alone causes rapid depolarization of the
mitochondrial membrane (Figure 4.2B). However, unlike CCCP plus oligomycin, CCCP
alone caused a much smaller increase in cytosolic [Ca2+] with an amplitude
approximately half that caused by CCCP plus oligomycin (Figure 4.2C).

Next, the effect of depleting mitochondrial Ca2+ stores on MeHg-induced
elevations of [Ca2+]; was examined. Complete removal of Ca2+m using 5 uM CCCP plus
10 uM oligomycin for 8 min delayed the time-to-onset of increases in [Ca2+]; caused by
0.5 uM MeHg (from 12.6 d: 2.5 min to 37.7 i 3.0 min) (Figure 4.3). The time-to-onset of
the first-phase increase in [Ca2+]; was not significantly altered by CCCP alone (11.5 i 1.4
min vs. 10.7 :t 1.7 min) or oligomycin alone (9.8 :t 1.9 rrrin vs. 17.4 :1: 5.8 min) (Figure
4.3). Similarly, emptying of the mitochondria with CCCP and oligomycin prior to
exposure to 0.5 uM MeHg in EGTA-HBS revealed a similar delay in the increased
[Ca2+]i, delaying the onset of the first-phase [Ca2+]; increase fiom 15 .3i3.2 to 45.4:l:3.9
min (Figure 4.4A). Additionally, CCCP and oligomycin treatment significantly reduced
the amplitude of the fura-2 ratio increase from a normalized ratio of 1.252t0.26 to
0.41:1:0.13, representing a 67% decrease in the ratio amplitude (Figure 4.4B). Thus
complete removal of Ca2+,“ is required to delay significantly the MeHg-induced release of
Ca2+ from intracellular pools, but does not prevent an eventual increase in [Ca2+]i.
Additionally, this data indicates that most of the Cay, released by MeHg originated in the
mitochondria.

The mitochondrial inhibitors had differing effects on the tirne-to-onset of the

114

 

- MeHg only - Inhibitor + MeHg

 

 

E 50 "

E 40

ES

Cg) 3O

é. 20

a5 10

.§

F-i . .
CCCP Olrgo CCCP+ Olrgo

Treatment

Figure 4.3. Effect of mitochondrial inhibitors on first-phase increase in [01”];
caused by 0.5 pM MeHg. Granule cells were pretreated with 5 uM CCCP (8 min), 10
11M oligomycin (8 min), or 5 uM CCCP plus 10 pM oligomycin (8 min) immediately
prior to 0.5 uM MeHg. Time-to-onset of the [Ca2+]; increase was compared between
MeHg alone and MeHg plus inhibitor for each treatment. Results are presented as
meaniSEM (n=5-8). The asterisk (*) indicates a value significantly different fi'om
MeHg-only control (p<0.05).

115

 

 

- VH i; if CCCP + oligo

A 50 *

.S -

E

3

U)

c

C?

,9.

a')

.§

H

lst Phase
B - MeHg only -a C+O+MeHg

Normalized Ratio
~ N

O

 

Figure 4.4. Effect of CCCP and oligomycin on first-phase time-to-onset and ratio
amplitude induced by 0.5 pM MeHg in the absence of Cat“, A) Measurement of
time-to-onset of [Cay]; elevations caused by 0.5 uM MeHg in the absence of Ca2+,
Cells were preheated with vehicle control (“VH”) or 5 uM CCCP plus 10 uM
oligomycin (“CCCP+oligo”) for 8 min immediately prior to 0.5 uM MeHg. B)
Measurement of the first-phase ratio amplitude in the absence of C322. Cells were
exposed to 0.5 pM MeHg alone (“MeHg”) or 5 uM CCCP plus 10 uM oligomycin (8
min) immediately prior to 0.5 uM MeHg (“C+O”). The amplitude of the first-phase ratio
increase was normalized to the peak fura-2 response to the 1 min 40 mM K+ test (at the
beginning of each experiment). Results are presented as mearttSEM. (n=5). The
asterisk (*) indicates a value significantly different from MeHg-only control (p<0.05).

116

- MeHg only - Inhibitor + MeHg

Q
o
1

*

4.}.
(ll

t-a
1J1

 

Time-to-Onset (min)
DJ
O

 

CCCP Oligo CCCP+ Oligo

Treatment

Figure 4.5. Comparison of time-to-onset of Ca2+c influx caused by 0.5 uM MeHg in
the presence or absence of mitochondrial inhibitors. Granule cells were pretreated
with 5 uM CCCP (8 min), 10 pM oligomycin (8 min), or 5 uM CCCP plus 10 uM
oligomycin (8 min) immediately prior to 0.5 uM MeHg. Comparisons are made between
MeHg alone and MeHg plus inhibitor for each treatment. Results are presented as
meaniSEM (n=5-8). The asterisk (*) indicates a value significantly different from
MeHg-only control (p<0.05).

117

influx of Ca2+': caused by 0.5 uM MeHg. Mitochondrial membrane depolarization
without maintenance of cellular ATP levels using CCCP alone hastened the influx of
Caz“c caused by MeHg, from 22.0 d: 3.3 min to 15.6 i: 2.7 min (Figure 4.5). This Ca2+
influx may occur independently of any effects of MeHg, as mitochondrial membrane
depolarization causes reversal of the mitochondrial ATPase, leading to conversion of
ATP to ADP and depletion of ATP, thereby causing Ca2+ influx through loss of ATP for
plasma membrane Ca2+ pmnps (N icholls and Scott, 1980). Oligomycin alone had the
opposite effect, causing a significant delay in the time-to—onset of MeHg-induced Ca2+.
influx, fi'om 17.8 i 3.7 min to 42.7 :1: 9.7 min (Figure 4.5). Because oligomycin does not
completely depolarize the mitochondria by 8 min but does maintain cellular ATP levels,
this delayed influx of Ca2+. suggests that MeHg-induced depletion of cellular ATP
triggers onset of the second-phase increase in [Ca2+]i. Application of CCCP plus
oligomycin caused a similar delay in the time-to-onset of the MeHg-induced Ca2+ influx
as that caused by oligomycin alone, from 22 :1: 3.4 min to 36.3 at 8.3 min (Figure 4.5).
Thus, MeHg-induced influx of Ca2+e appears to be dependent on loss of cellular ATP
content as a result of MeHg-induced alteration of mitochondrial function.

The next step was to examine the route by which MeHg causes uptake of Ca2+
into the mitochondria, and subsequent release of Ca2+", from the mitochondria into the
cytosol. As described in Chapter Two, previous experiments suggested that MeHg
causes release of Ca2+ from the SER via activation of muscarinic receptors linked to
production of 1P3. Based on these results, it was hypothesized that MeHg caused Ca2+ to
be released from the SER, which was then buffered by mitochondria. To test this

hypothesis, the effect of MeHg on Ca2+,“ was examined using the fluorescent Ca2+,“

118

indicator rhod—2. Cells were first loaded with fura-2 and exposed to increasing
concentration of K+ (IO-100 mM) for 1 min each, to demonstrate that this protocol causes
reversible increase in [Caz] which exhibit increasing fura-2 ratio amplitude as the [K+]
increases (Figure 4.6A). Granule cells were then loaded with rhod-2 to monitor changes
in [Ca2+]m. Control experiments using 1 min depolarizations with increasing
concentrations of K+ (10- 1 00 mM) indicated that rhod-2 fluorescence increases during an
increase in [Ca2+]; and returns to baseline after removal of the K+ solution (Figure 4.68).
Further, this increase in fluorescence was due to a change in Ca”... and not due to
cytosolic loading of the dye as the increase in rhod-2 fluorescence caused by 1 min
treatment with 40 mM K+ was abolished by 8 min exposure to CCCP and oligomycin
(Figure 4.6B).

As seen in Figure 4.7, granule cells loaded with rhod-2 exhibit a biphasic increase
in rhod-2 fluorescence during exposure to 0.5 uM MeHg. Rhod-2 fluorescence increases
to a similar intensity as that caused by 1 min depolarization with 40 mM K“ (Figure 7A);
this initial increase is abolished by 8 rrrin exposure to CCCP and oligomycin, indicating
that the initial increase in fluorescence is due to an increase in Ca2+“, (Figure 7B). The
secondary increase in fluorescence is associated with a spreading of the dye from the
mitochondria to the cytosol which was apparent both in the digital image recorded by the
IonOptix software, and in the fluorescence emission recording. The dye movement was
associated with an increase in rhod-2 fluorescence intensity above that caused by the 1
min 40 mM K+ depolarization; thus we measured the secondary fluorescence increase as
the point at which rhod-2 fluorescence became elevated above that of the 40 K+-induced

depolarization. Interestingly, treatment with CCCP and oligomycin abolished the initial

119

 

 

 

 

 

 

 

1.40 -
r-xo i
00
g 100
3‘ 50
E"; 070 . 40 l
.g 10 20 l
e
ed
0.00 l l L J 1
0 10 20 30 40 50
Time (min)
”a?
B E 6 - 100 40
D
. 50 l c+o
.o 40 ——
2 2° 1
<1) 4 r i '
O
c:
Q)
0
U)
<1)
H
o l L l
2 2 L l 1
LL. 0 10 20 30 40 50 60

Time (min)

Figure 4.6. Changes in fura-2 and rhod-2 fluorescence in response to increasing
concentrations of extracellular 1C. A) Representative example of changes in fura-2
fluorescence (“Ratio (Fm/330)”) in a single cerebellar granule neuron exposed to
increasing concentrations of K+ (IO-100 mM). Each K+ solution was applied for 1 min
starting at the arrow, with the [K] (mM) indicated by the number above the arrow. B)
Representative example of changes in rhod-2 fluorescence (“Fluorescence (arb. units)”)
in a single cerebellar granule neuron exposed to increasing concentrations of K+ (IO-100
mM). Each K+ solution was applied for 1 min starting at the arrow, with the [K] (mM)
indicated by the number above the arrow. At the end of the experiment, 5 uM CCCP and
10 uM oligomycin were applied for 8 min, followed by another 1 min treatment with 40
mM K". CCCP and oligomycin abolished rhod-2 fluorescence changes in response to the
K4” depolarization, confirming that rhod-2 was localized within the mitochondria.

120

elevation of fluorescence but did not alter the time-to-onset of the secondary fluorescence
increase associated with the dye spreading throughout the cell (Figure 4.9A). Because
the initial increase of fluorescence did not occur until the spreading of dye throughout the
cell occurs, the “first-phase” increase in rhod-2 fluorescence occurred simultaneously
with the “second-phase” release of dye into the cytosol (Figure 4.8). When granule cells
were exposed to 0.5 uM MeHg in the nominally Cay-free EGTA-HBS, the pattern of
rhod-2 fluorescence changes was similar, as MeHg again caused an initial fluorescence
increase which was abolished by CCCP and oligomycin, and a secondary fluorescence
increase from dye movement into the cytosol (Figure 4.9B).

Because MeHg caused an increase in Ca2+,“ in the absence of C322, it was
hypothesized that the mitochondria were buffering Ca2+ that was released from another
intracellular Ca2+ store, possibly the SER; thus, the next experiment examined the effect
of SER Ca2+ depletion on MeHg-induced elevations of Ca2+m using the SERCA inhibitor
thapsigargin (10 uM for 5 min). Thapsigargin treatment immediately prior to exposure to
0.5 uM MeHg did not affect the time-to-onset of the initial rhod-2 fluorescence increase,
but did significantly delay the dye movement into the cytosol (Figure 4.10).

Finally, experiments were performed to test whether the release of rhod-2 from
the mitochondria into the cytosol was due to opening of the MTP. Related experiments
indicated that the MTP inhibitor CsA delays the MeHg-induced increase in [Ca2+]; caused
by 0.2 and 0.5, but not 1.0 uM MeHg (see Chapter Five). In granule cells loaded with
rhod-2, application of 5 uM CsA for 10 min immediately before 0.5 uM MeHg did not

alter the time-to-onset of the initial fluorescence increase but did significantly delay the

121

0.5 uM MeHg

 

  
 

Onset of Initial
Fluorescence Increase

 

 

 

0 10 20 30 4O 50

Relative Fluorescence (arb. units)
U»)

Time (min)

Figure 4.7. MeHg causes a biphasic increase in rhod-2 fluorescence in granule cells.
Changes in rhod-2 fluorescence in a single neuron exposed to 0.5 uM MeHg. Cells were
initially exposed to 1 mM K+ to verify cell viability (“K”). During MeHg exposure,
there is a biphasic increase in rhod-2 fluorescence: the initial rise plateaus at a similar
fluorescence intensity as that caused by the 1 min 40 mM K+; the secondary increase is
associated with the dye spreading throughout the entire cytosol.

122

 

 

 

4 .—
3; 0.5 uM MeHg + oligomycin
.6
E: 3
8
§ 2
i
E
1
33'.
0
m 0 iilrlrirrlrrlrlnnaalnrnrlrrrll

 

 

0 10 20 30 40 50 60

Time (min)

Figure 4.8. Effect of CCCP and oligomycin on MeHg-induced elevations of rhod-2
fluorescence. Changes in rhod-2 fluorescence in a single neuron exposed to 5 pM CCCP
and 10 uM oligomycin (“C+O”) for 8 min immediately before 0.5 11M MeHg. In these
cells, the initial rise in rhod-2 fluorescence is absent; however, the secondary
fluorescence increase from the dye spreading throughout the cytosol is still present.

123

A - MeHg only - C+O+MeHg

40 '
30
20

10

Time-to-Onset (min)

 

 

lst Phase 2nd Phase

- MeHg only - C+O+MeHg

20"

10‘

 

 

Time-to-Onset (min)

1st Phase 2nd Phase

Figure 4.9. Mitochondrial membrane depolarization abolished the initial elevation of
rhod-2 fluorescence without affecting the time-to-onset of the spreading of rhod-2
throughout the cytosol. A) Granule cells were pretreated with 5 uM CCCP plus 10 uM
oligomycin (8 min) immediately prior to 0.5 uM MeHg. Time-to—onset of the [Ca2+]i
increase was compared between MeHg alone and MeHg plus CCCP and oligomycin.
Results are presented as meaniSEM (n=4). B) Granule cells were pretreated with 5 uM
CCCP plus 10 uM oligomycin (8 min) immediately prior to 0.5 uM MeHg in EGTA-
HBS. Results are presented as mean i SEM (n=3). The asterisk (*) indicates a value
significantly different from the corresponding MeHg-only data (p < 0.05).

124

spreading of rhod-2 throughout the cell (Figure 4.11). A comparison of the effect of CsA
and thapsigargin on the interval between onset of first and second rhod-2 fluorescence

increases is shown in Figure 4.12.

125

- MeHg only - Thaps + MeHg

 

 

AZOP
.S.
:3. _
H *
0)
E’
010-
5'2.
Q5 .
.§
E...

0

   

lst Phase 2nd Phase

Figure 4.10. Thapsigargin delays the spreading of rhod-Z from punctate to whole-
cell staining as caused by 0.5 uM MeHg. Granule cells were pretreated with 10 uM
thapsigargin for 5 min immediately prior to 0.5 pM MeHg. The first-phase elevation is
measured where rhod-2 fluorescence rises above baseline; the second-phase elevation is
measured where the fluorescence rises above the fluorescence increase caused by the 1
min 40 mM 1C test. Results are presented as meaniSEM (n=5). The asterisk (’)
indicates a value significantly different from MeHg-only control (p<0.05).

126

 

 

- MeHg only - CsA + MeHg
30 "

N
O
l

I

Time-to-Onset (min)
8

 

   

lst Phase 2nd Phase

Figure 4.11. CsA delays the spreading of rhod-2 from punctate to whole-cell
staining as caused by 0.5 pM MeHg. Granule cells were pretreated with 5 uM CsA
immediately prior to 0.5 uM MeHg. The first-phase elevation is measured where rhod-2
fluorescence rises above baseline; the second-phase elevation is measured where the
fluorescence rises above the fluorescence increase caused by the l min 40 mM K+ test.
Results are presented as meaniSEM (n=7). The asterisk (*) indicates a value
significantly different fiom MeHg-only control (p<0.05).

127

 

 

- MeHg only - Drug + MeHg

20'

Time (min)
8

 

 

CsA Thapsigargin

Figure 4.12. Effect of CsA and thapsigargin on interval between onset of first and
second rhod-2 fluorescence increases. The time between the first- and second-phase
increases in rhod-2 fluorescence indicates the amount of time between the initial uptake
of Ca2+ into the mitochondrial lumen and the release of rhod-2 into the cytosol. Results
are presented as mean d: SEM (n=5-7). The asterisk (*) indicates a value significantly
different from the corresponding MeHg-only data (p < 0.05).

128

DISCUSSION

Disruption of [Ca2+]i is an early step in MeHg neurotoxicity which apparently
contributes to MeHg-induced cell death in rat cerebellar granule neurons (Marty and
Atchison, 1998). These experiments indicate that the bulk of this [Cay]; originates in the
mitochondria. When the mitochondrial Ca2+ store is emptied following mitochondrial
membrane depolarization with CCCP plus oligomycin, the amplitude of the whole-cell
[Caz+]i elevation caused by MeHg is significantly reduced by approximately 65%.
Removal of Ca2+". does not abolish the MeHg-induced release of {Cap}, indicating the
participation of more than one intracellular Ca2+ source. The participation of more than
one intracellular store agrees with results of experiments in T lymphocytes, in which
MeHg causes release of Ca2+ from both mitochondria and SER (Tan et al., 1993).
However, the first-phase increase in firm-2 ratio is not comprised solely of Ca2+, as there
is at least one other divalent cation whose intracellular levels increase in response to
MeHg (Denny et al., 1993; Hare et al., 1993; Marty and Atchison, 1997). In rat cortical
synaptosomes, this cation was identified as Zn2+ by l9F-NMR (Denny and Atchison,
1994). Zn2+ has recently been identified as a toxic cation capable of opening the MTP
during neurotoxic insult (W udarczyk et al. , 1999). Although non-Ca2+ divalent cations
do not appear to contribute to MeHg-induced cell death in granule cells (Marty and
Atchison, 1997), elevated Zn2+ could contribute to the opening of the MTP observed in
this and other studies (Limke and Atchison, 2002). Thus, the non-mitochondrial
component of the first-phase increase in fora-2 ratio is likely comprised of a combination
of Ca2+ from another intracellular store, such as the SER, and the non-Ca2+ divalent

cation, most likely Zn”.

129

The large contribution of Ca2+“. to the first-phase increase in [Ca2+]. in cerebellar
granule cells is in contrast to the elevations of [Ca2+]; in NG108-15 cells, in which the
primary source of Ca2+; was the SER through the 1P3 receptor, with minimal contribution
fi'om the mitochondria (Hare and Atchison, 1995). The preference for Ca2+ release fi'om
mitochondria as Opposed to release from another Ca2+ pool parallels experiments in the
neuromuscular junction, in which MeHg-induced increases in spontaneous release of
neurotransmitter were decreased to less than 10% of control by inhibition of Ca2+ release
from the mitochondria and were not decreased by inhibitors of SER Ca2+ release
(Levesque and Atchison, 1988). Motor nerve terminals may be one of the targets of
MeHg during in vivo exposure, as a neuromuscular weakness resembling myasthenia
gravis is observed (Rustam et al., 1975). The heightened susceptibility of neurons
exhibiting significant changes in Ca2+". suggests a correlation between MeHg interaction
with mitochondria and subsequent neuronal damage. Given the relatively small
mitochondrial contribution to whole-cell {Caz} elevations in less sensitive neuronal
models, such as NG108-15 cells (Hare et al., 1993; Hare and Atchison, 1995), these
results suggest that disruption of mitochondrial Ca2+ regulation may be an important
factor in determining cellular sensitivity to MeHg.

The differential effects of mitochondrial inhibitors highlight the pivotal role of
Ca2+,“ regulation in determining the extent of whole-cell increases of [Cay]; induced by
MeHg. Mitochondrial membrane depolarization with CCCP alone did not cause
significant release of Ca2+... into the cytosol. These results agree with previous
experiments in granule cells (Budd and Nicholls, 1996b), in which mitochondrial

membrane depolarization alone did not cause significant release of Ca2+m. Interestingly,

130

 

mitochondrial membrane depolarization alone did not alter the time-to-onset of the
MeHg-induced release of Ca2+t. The smaller release of Ca2+,“ into the cytosol induced by
CCCP alone suggests that enough Ca2+.“ remains to be released by MeHg.

Application of oligomycin alone resulted in partial depolarization of
mitochondrial membrane and, as an ATP synthase inhibitor, maintenance of cellular ATP
levels (Budd and Nicholls, 1996b). Oligomycin alone did not cause an increase in
cytosolic [Ca2+], indicating that partial mitochondrial membrane depolarization is
insufficient to empty mitochondrial CaZ+ stores. This is supported by the lack of effect of
oligomycin on the time-to—onset of the initial increase in [Ca2+]i during MeHg exposure.
However, oligomycin alone did delay significantly the onset of MeHg-induced influx of
Caz} [Ca2+]t is maintained, in part, by plasma membrane Ca2+ pumps which require
ATP to remove Ca2+ from the cytosol (Werth et al., 1996; Morgans et al., 1998), thus
depletion of cellular ATP content inhibits ATP-dependent extrusion of Ca2+ from the cell.
This is supported by the hastening of the MeHg-induced, second-phase influx of Ca2+.
caused by CCCP alone. When the mitochondrial membrane becomes depolarized, the
mitochondrial ATP synthase reverses and begins hydrolyzing ATP into ADP in an
attempt to maintain \th (N icholls and Scott, 1980). Experiments in cerebellar granule
neurons in primary culture indicate that 5 min exposure to 2 uM CCCP rapidly depletes
cellular ATP levels (Budd and Nicholls, 1996b), thus in our experiments, 8 nrin exposure
to 5 uM CCCP would similarly be expected to deplete ATP. Under these conditions,
CCCP-induced depletion of cellular ATP hastens the MeHg-induced depletion of ATP

(Sarafian et al., 1989) and causes an earlier onset of Ca2+.3 influx.

131

 

Depletion of Ca2+m with CCCP and oligomycin delays and reduces the first-phase
elevation of [Caz‘] but does not prevent an eventual rise in [Ca2+]t. The delayed
movement of rhod-2 into the cytosol by thapsigargin indicates that MeHg causes Ca2+
release from the SER, with this Ca2+ subsequently buffered by mitochondria. There is
substantial evidence that release of Ca2+ fiom the SER causes a rapid, reversible rise in
Ca2+... which is a normal component of intracellular signaling (Landolfi et al., 1998;
Pinton et al. , 1998; Rizzuto et al., 1998). In HeLa cells transfected with a Ca2+-sensitive
photoprotein localized to the outer face of the inner mitochondrial membrane, Ca2+
released by activation of 1P3 receptors resulted in a high [Caz+] within the mitochondria
(Rizzuto et al., 1998). This [Ca2+] was much higher than that observed in the bulk
cytosol as a result of the close spatial relationship between mitochondria and SER in
these cells (Pinton et al., 1998; Rizzuto et al., 1998).

These experiments also provide evidence that MeHg causes opening of the MTP,
because the rhod-2 movement from punctate to diffuse staining is delayed by the MTP
inhibitor CsA. Because the rhod-2 fluorescence increase associated with opening of the
MTP is also delayed by thapsigargin, this suggests that release of Ca2+ from the SER
contributes to the MeHg-induced loss of Ca2+m regulation via opening of the MTP. This
shifting of intracellular Ca2+ from the SER to the mitochondria explains the large amount
of Ca2+“. released by MeHg in granule cells. Mitochondria are a low-affinity Ca2+ store
which do not contain much Ca2+ at rest; high [Ca2+].m is energetically unfavorable and
interferes with the respiratory chain and the tricarboxylic acid cycle (Budd and Nicholls,
1996b; Wang and Thayer, 1996). Because mitochondria normally contain little (332+, this

movement of Ca2+ from the SER to the mitochondria underlies the large amplitude

132

 

release of Ca2+", during MeHg exposure. Additionally, it suggests in granule cells that
the SER is in fact the initial Ca2+. target, with mitochondria being affected at a later time
point. The sensitivity of mitochondria to MeHg-induced increases in Ca2+m may be
dependent on the spatial relationship between SER Ca2+ release channels and nearby
mitochondria. In NG108-15 cells, the primary source of Ca2+i during MeHg exposure
originates in the IP3-sensitive Ca2+ pool in the SER (Hare and Atchison, 1995). The
greater proportion of Ca2+ fi'om the SER in NG108-15 cells, as compared to granule cells,
may be due to a lack of rrritochondrial buffering rather than less release of Ca2+ from this
store. Alternatively, the apparent sensitivity of granule cell mitochondria may be due to
the Ca2+ uptake and release characteristics of the mitochondria themselves. Mitochondria
in different brain regions have differential sensitivities to induction of the MTP in
different brain regions (F riberg et a1. , 1999). Because mitochondria are such a prominent
target during MeHg exposure, the preferential sensitivity of the granule cells to MeHg-
induced cell death may be related to the Ca2+ regulating characteristics of their

mitochondria.

133

 

CHAPTERS

ACUTE EXPOSURE TO METHYLMERCURY OPENS THE

MITOCHONDRIAL PERMEABILITY TRANSITION PORE IN RAT

CEREBELLAR GRANULE CELLS

 

134

ABSTRACT

Following acute MeHg exposure, granule cells in culture undergo an increase in
[Ca2+]; that apparently contributes to cell death. This effect consists of several temporally
and kinetically distinct phases. The initial elevation arises fiom release of Ca2+; stores;
the second phase results from entry of Ca2+e. These experiments tested the hypothesis
that release of mitochondrial Ca2+ through the MTP contributes to the initial release of
Ca2+i. Neonatal rat cerebellar granule cells in culture and single cell microfluorimetry
were used to examine MeHg-induced changes in [Caz+]i and mitochondrial membrane
potential (Wm). Pretreatment with the MTP inhibitor CsA (5 uM) delayed the initial
phase of increased [Ca2+], induced by 0.2 and 0.5 uM MeHg, but not 1.0 uM MeHg.
CsA (5 uM) also delayed the irreversible loss of ‘1’“, induced by 0.5 uM MeHg. Ca2+.
was not required for either effect, because the effect of CsA on the first phase increase in
[Ca2+]; and loss of ‘I’m was not altered in nominally Cay-flee buffer. Increasing
concentrations of MeHg (0.2-2.0 uM) caused increasing incidence of cell death at 24 hr
post-exposure. Treatment with CsA provided protection against cytotoxicity at lower
MeHg concentrations (0.2-0.5 uM), but not at higher MeHg concentrations (1.0-2.0 pM).
Thus the MTP appears to play a significant role in the cellular effects following acute

exposure of cerebellar granule neurons to MeHg.

135

INTRODUCTION
At the cellular level, MeHg interacts with a number of processes to disrupt
divalent cation homeostasis (Komulainen and Bondy, 1987; Hare et al., 1993; Marty and
Atchison, 1997), elevate spontaneous release of neurotransmitter (J uang, 1976; Atchison
and Narahashi, 1982; Atchison et al., 1986), generate reactive oxygen species (ROS)
(LeBel et al., 1990; Sarafian and Verity, 1991; Yee and Choi, 1996) and induce cell death F

(Sarafian et al., 1989; Marty and Atchison, 1998). As described in previous chapters,

 

low concentrations of MeHg (0.2-2.0 uM) disrupt the ability of rat cerebellar granule
cells in primary culture to maintain [Ca2+]i, causing a biphasic increase in [Caz+]i. The I
initial phase (“first phase”) involves release of Ca2+i, followed several minutes later by a
secondary (“second phase”) influx of Ca2+,3 (Marty and Atchison, 1997). MeHg-induced
disruption of [Ca2+]i is critical in determining neuronal survival, because agents which
buffer [Ca2+]; protect against MeHg-induced granule cell cytotoxicity (Marty and
Atchison, 1998), and Ca2+ channel blockers afford some protection against MeHg-
induced toxicity in viva (Sakamoto et al., 1996).

One possible source of the early-onset Ca2+i elevation during MeHg exposure is
the mitochondria. MeHg causes a number of effects in mitochondria, including inhibition
of respiration, increased permeability of the inner membrane, and depolarization of the
membrane potential (w...) (Verity et al., 1975; Sone et al., 1977). Additionally, MeHg
exposure disrupts Ca2+ regulation in mitochondria derived from rat brain by decreasing
Ca2+ uptake and inducing Ca2+ release (Levesque and Atchison, 1991). Inhibition of
mitochondrial Ca2+ release reduces MeHg-induced release of ACh from rat cortical

synaptosomes, supporting the idea that mitochondrial Ca2+ release plays a key role in

136

 

MeHg-induced neurotoxicity (Levesque et al., 1992). In granule neurons, mitochondria
can store and release large amounts of Ca2+ under both physiological and pathological
conditions (Budd and Nicholls, 1996b). Given that MeHg targets mitochondria in other
systems, it was hypothesized that MeHg affects mitochondrial Ca2+ regulation in rat
cerebellar granule neurons, causing release of mitochondrial Ca2+ into the cytosol during
MeHg exposure.

One route by which mitochondria release Ca2+ is the MTP (Bemardi et al., 1992;

Petronilli et al. , 1993). The MTP is a megapore formed on the inner mitochondrial

 

membrane in response to adverse conditions including oxidative stress, elevation of
mitochondrial matrix Ca2+ and anoxia (Dubinsky and Rothman, 1991; Bemardi et al. ,
1993). When open, the MTP allows the passage of molecules 51.5 KDa, including Ca2+,
across the usually impermeable inner mitochondrial membrane. Because MTP opening
dissipates the proton gradient across the inner mitochondrial membrane, there is a
corresponding loss of rum which can be used as a diagnostic tool for observing a
permeability transition (White and Reynolds, 1996; Trost and Lemasters, 1997). Both in
vitro and in vivo, MTP induction is linked to neuronal death caused by a variety of
compounds and conditions, including ischemia, stroke and glutamate excitotoxicity
(Bemardi and Petronilli, 1996; Schinder et al. , 1996; White and Reynolds, 1996).
Opening of the MTP can be inhibited by the immunosuppressant CsA, even in the
presence of inducing agents (Bemardi et al., 1993); (N ieminen et a1. , 1996).

While there is evidence for induction of a permeability transition in MeHg-
induced death of non-neuronal cells (InSug et al., 1997; Shenker et al., 1998), there is no

direct experimental evidence linking MeHg to MTP induction in granule neurons.

137

Because MeHg creates conditions which favor opening of the MTP (generation of ROS,
elevation of [Ca2+]i, etc.), it seemed likely that MTP induction plays a role in MeHg-
induced neurotoxicity. These experiments tested the hypothesis that MeHg induces
opening of the MTP (through direct or indirect action at the mitochondria), causing it to
release Ca2+ into the cytoplasm, thereby contributing to the elevated [Caz+]i observed
during the first, Ca2+t-dependent phase of MeHg neurotoxicity. MTP induction was
measured by examining the ability of CsA to prevent changes in [Ca2+]i, rum, and cell
death following acute exposure to MeHg. Because CsA also inhibits the protein
phosphatase calcineurin (CN) (Bemardi and Petronilli, 1996), the effects of the CN
inhibitors F K506 and W7 were examined to determine if effects on CN function

contributed to the observed actions of CsA.

138

 

MATERIALS AND METHODS

The materials and solutions for isolation of cerebellar granule cells and
measurement of changes in [Ca2+]; are described in detail in Chapter Two. The procedure
for the Eukolight cell viability assay is described in detail in Chapter Three. The
purchase and handling of CsA, and the use of TMRE to measure changes in ‘I’m, are
described in Chapter Four. Unless specifically stated, all solutions were identical to those
described in Chapter Two. The methods and solutions described below are specific to

this chapter.

Materials and Experimental Solutions. Phenylarsine oxide (PhAsO) was purchased
from Sigma (St. Louis, MO). F K506 and N-(6-aminohexyl)-5-chloro-1-
napthalenesulfonamide (W7) were purchased fi'om Calbiochem (La Jolla, CA). PhAsO
was prepared as a 100 mM stock solution in 100% DMSO. F K506 was dissolved in
DMSO as a 1 mM stock solution. W7 was dissolved into distilled water as a 1 mM stock
solution. Control solutions contained corresponding concentrations of solvent, with all

final solutions containing no more than 0.1% (v/v) solvent.

Measurement of [Ca2+]. Changes Using Fura-Z. To examine the role of a permeability
transition in MeHg-induced changes in [Ca2+]i, the potent MTP inhibitor CsA was
utilized (Bemardi et al., 1993). Because CsA also inhibits CN function, the CN
inhibitors W7 [which inhibits calmodulin (Ankarcrona et al. , 1996)] and F K506 [which

inhibits FK506 binding protein (Marcaida et al. , 1996)] were also used. Granule cells

139

were loaded with fora-2 for 1 hr, then rinsed with HBS as described in Chapter Two. All
experiments began with a 1 min baseline recording, followed by a 1 min exposure to 40
mM KL and a 3 min washout with HBS. For a cell to be considered viable, fura-2
fluorescence must increase transiently in response to the K+ depolarization challenge and
subsequently return to baseline during the washout period. In the first set of experiments,
the K+ depolarization was followed by a 10 min exposure to one of the inhibitors (0.25-5
pM CsA, 20 uM W7 or 1 uM F K506) or an appropriate solvent control. Next, the
perfirsion solution was replaced with either 0.5 uM MeHg (experiment) or HBS (control)
for 45 min. In the second set of experiments, the K+ depolarization was followed by
simultaneous application of 5 uM CsA and 0.5 pM MeHg for 45 min; corresponding
buffer and vehicle controls were also performed on the same day. Finally, both the 5 pM
CsA pretreatment and simultaneous treatment experiments were repeated in Ca2+-free
buffer to verify the intracellular origins of the first Ca2+ phase. For the 10 min
pretreatment experiments in Ca2+-free buffer, the data were also analyzed for fura—2 ratio
amplitude as a measure of [Ca2+]; released. In this analysis, the fora-2 fluorescence ratio
was normalized to the peak fluorescence increase caused by 1 min 40 mM K“ to account

for differences in dye loading.

Visualization of Changes in ‘1’... Mitochondrial depolarization is a key feature of MTP
opening; thus, prevention or attenuation of mitochondrial depolarization by CsA is
indicative of MTP opening (N ieminen et al., 1996; White and Reynolds, 1996; Trost and
Lemasters, 1997). To measure changes in ‘I’m, cells were loaded with TMRE as

described in Chapter Four. As before, all perfusion solutions contained 10 nM TMRE to

140

 

 

 

 

maintain dye availability for the mitochondria. Fluorescence was measured relative to an
initial fluorescence of 100%. Control experiments to verify the ability of cerebellar
granule cell mitochondria to undergo a permeability transition utilized the known MTP-
inducing agent PhAsO (5-15 uM) (Kowaltowski and Castilho, 1997). Experiments with
MeHg were performed in a similar manner to the [Ca2+]; experiments, with application of
40 mM K+ for 1 min (to verify cell viability), followed by 3 min rinse with HBS, a 10
min pretreatment with 5 uM CsA, and 45 min exposure to 0.5 uM MeHg. Similar

experiments were performed in the absence of Ca2+... and in the absence of the l min 40

 

mM K+ treatment. For all MeHg experiments using TMRE, the parameter measured was
time-to-onset of the mitochondrial depolarization, determined at the point at which
TMRE fluorescence dropped precipitously, indicating loss of Wm. For TMRE data,
fluorescence was measured relative to 100% at the beginning of the experiment. The
times-to-onset fiom all cells in an experiment were averaged to provide a mean time-to-

onset of mitochondrial membrane depolarization for that dish of cells (n=1).

Cell Viability Assays. To determine whether inhibition of MTP activation would
prevent MeHg-induced cell death, a series of experiments was performed which
paralleled the fura—2 imaging experimental protocol. All cells were incubated for 90 min
at 37°C in HBS to mimic the fura-2 incubation (1 hr) and HBS perfusion rinse (30 min)
of the [Cay], measurement experiments. This was followed by a treatment protocol (see
below), after which the cells were returned to conditioned medium (to bind excess
MeHg) until the Live/Dead assay was performed (as detailed in Chapter Three).

In the first set of experiments, the effect of CsA treatment on MeHg-induced cell

death was examined. Cells were pre-exposed to HBS for 90 min, followed by incubation

141

in O, 1 or 5 uM CsA for 10 min prior to application of 0-2 uM MeHg at the following
concentrations and durations: 0 MeHg (control) for 60 min; 0.2 uM MeHg for 76 min;
0.5 pM MeHg for 45 min; 1 pM MeHg for 38 min; or 2 uM MeHg for 30 min. These
exposure times were based on the time needed for both the first and second phase [Ca2+];
increases to occur in 97.5% of the cells at each concentration (Marty and Atchison,
1998). After MeHg exposure, cells were washed with conditioned medium and returned
to the incubator for 24 hr. At 24 hr post-exposure, cells were washed twice with HBS,
then incubated in cytotoxicity assay reagents for 30 min and assayed for. fluorescence.
Viability was also determined 24 hr after simultaneous exposure to 5 uM CsA and 0-2
uM MeHg for the given time periods.

In the second set of experiments, viability was detemrined 24 hr after
simultaneous exposure to 5 uM CsA and 0-2 uM MeHg for the given time periods, using
the same MeHg concentrations and durations listed above. In the final set of
experiments, the effect of CN inhibition on MeHg-induced cell death was examined
using 20 uM W7 and 1 uM FK506. Cells were pre-exposed to HBS for 90 min, followed
by 10 min treatment with either W7 or F K506 and 45 min of 0.5 uM MeHg; viability was
determined at 24 hr post-exposure. For all cytotoxicity assays, appropriate solvent and

buffer controls were performed simultaneously.

Statistics. For fura-2 and TMRE experiments, comparisons of mean time-to-onset for
“MeHg” vs the corresponding “MeHg with inhibitor” were made using Student’s paired
t-test. For the cell viability experiments, the percent viability data were transformed

using an angular transformation and analyzed per experimental day using a repeated

142

 

measures one-way analysis of variance. For all experiments, a p<0.05 was considered to

be statistically significant.

I43

RESULTS

As noted in previous chapters, neonatal rat cerebellar granule cells in culture
undergo a biphasic increase in [Ca2+]i in response to acute exposure to MeHg. Exposure
to CsA for 10 min immediately prior to MeHg exposure delayed the time-to-onset of both
the first and second phase of [Ca2+]i elevations in a concentration-dependent manner.

That is, lower concentrations of CsA (0.25 and 0.5 uM) did not delay the time-to-onset of
either phase, while higher concentrations (1.0 and 5.0 uM) significantly delayed the time-
to-onset of the first-phase [Ca2+]; elevation (Figure 5.1). The higher CsA concentrations
also delayed the time-to—onset of the second phase; however, these results were not quite
statistically significant (p=0.05). Interestingly, both phases were delayed to a similar
extent, in that the first phase was delayed from 13.6:1:1 .4 min (0.5 11M MeHg alone) to
26.9i5.3 min and 27.7:t:3 .4 min for 1.0 and 5.0 uM CsA + 0.5 uM MeHg, respectively.
The second phase was also delayed to a similar extent, from 21.6:b2.5 min (0.5 uM MeHg
alone) to 33.13255 and 37.1d:1.7 min for 1.0 and 5.0 uM CsA + 0.5 uM MeHg,
respectively.

The next set of experiments tested the effect of a single CsA concentration on the
time-to-onset of [Ca2+]; elevations caused by increasing concentrations of MeHg. A 10
min pretreatment with 5 uM CsA significantly delayed the first-phase release of Ca2+;
caused by 0.2 uM MeHg (fi'om 18.0i1.6 min to 345i 3.3 min) and 0.5 uM MeHg (from
12.7:l:1.2 min to 25.6:t3.7 min), but not 1.0 uM MeHg (1.8:t:0.4 min for MeHg only vs.
1.6i0.5 min for CsA pretreatment) (Figure 5.2A). The same treatment also significantly
delayed the influx of Ca2+. caused by 0.5 uM MeHg (from 20.3i2.4 min to 33:89 min)

and delayed, although not significantly, the influx of Ca2+c caused by 0.2 uM MeHg

144

 

Time to Onset (min)

 

 

0 0.25 0.5 1.0 5.0
[Cyclosporin A] (uM)

Figure 5.1. Increasing concentrations of CsA cause increasing delays in the time-to-
onset of [Ca2+]t elevations caused by 0.5 pM MeHg. Granule cells pretreated with
increasing concentrations of CsA exhibited increasing delays in the time-to-onset of both
the first and second phase increases in [Ca2+]; caused by 0.5 uM MeHg. The time-to-
onset was measured from the time of initial MeHg exposure. Although the times-to-onset
of the second phases for pretreatment with 1 and 5 uM CsA were not considered to be
statistically different from the 0 uM CsA cells, the results were remarkably close (p =
0.05 for each). Results are presented as mean :1: SEM (n = 6-10). The asterisk (*)
indicates a value significantly different from control (p < 0.05).

145

- MeHg only - CsA + MeHg

50 '
4O
3O
20
10
0

 

 

Time-to-Onset (min)

0.2 0.5 1 .0
[MeHg] (MM)

- MeHg only - CsA + MeHg

70 T
60
50
40
30
20
10

0

 

Time-to-Onset (min)

 

0.2 0.5 1.0
[MeHg] (luM)

Figure 5.2. 5 pM CsA significantly delays the time-to-onset of MeHg-induced
[Ca2+]: increases. (A) Granule cells pretreated with 5 pM CsA exhibited delayed the
time-to-onset of the MeHg-induced first-phase [Ca2+]i increase caused by 0.2 and 0.5, but
not 1.0 uM MeHg (n = 3-9). (B) 5 uM CsA pretreatment also delayed the time-to-onset
of Ca2+ influx (second phase) caused by 0.2 and 0.5, but not 1.0 uM MeHg. Results are
presented as mean :1: SEM (n = 3-9). The asterisk (*) indicates a value significantly
different from control (p < 0.05).

146

 

(from 39.7:1z7.2 min to 56:1:5.5 min). CsA pretreatment did not affect the Ca2+c influx
caused by 1.0 pM MeHg (Figure 5.2B).

The effect of CsA is reported to be transient such that inhibition of the MTP lasts
only 10-20 min (N ieminen et al., 1996). As such, continuous exposure to both CsA and

MeHg might increase the delay in time-to-onset of both phases to a greater extent than

 

was seen in the pre-exposure experiments. However, there was no difference between a m
10 min pretreatment with 5 uM CsA or 45 min simultaneous treatment with 5 uM CsA in

terms of time-to-onset for the first and second phase [Ca2+]; elevations (data not shown).

Simultaneous treatment with CsA delayed the effects of MeHg for both the first phase I

(from 15.5i2.6 min to 27.7:t3 .4 min) and second phase (from 24.2iZ.5 min to 37.1:t:1.7
min). Attempts to potentiate the effect of CsA using trifluoperazine were unsuccessful
because granule cells respond to trifluoperazine alone with elevated [Ca2+]; (data not
shown).

In EGTA-containing buffer, pretreatment with 5 uM CsA caused similar delays in
the time-to-onset of the first-phase [Ca2+]i elevation as seen in Ca2+-containing buffer. 5
pM CsA delayed the time-to-onset of the first phase for 0.2 pM MeHg (fiom 14.7d:1.5
min to 24.2:tl.6 min) and 0.5 uM MeHg (fiom 12.3:t:2.4 min to 2233.8 min), but not
for 1.0 uM MeHg (from 2.1-1:0.2 min to 2.6i0.7 min) (Figure 5.3). Measurement of the
first phase ratio amplitude in these cells indicated that CsA did not alter the amount of
Ca2+i released during MeHg exposure at any MeHg concentration examined (data not
shown). Simultaneous exposure to 5 uM CsA and 0.5 uM MeHg in the EGTA buffer
had a similar effect, delaying the first phase to the same extent (data not shown). These

results are very similar to those obtained in Cay-containing medium, and implicate the

147

contribution of an intracellular Ca2+ store sensitive to CsA in the first phase elevation of
[Ca2+]; caused by MeHg.
Several CN inhibitors were used in an experimental protocol similar to that with
CsA to test whether CN contributes to the MeHg-induced changes in [Cay]; and mm.
Cells were pretreated with either 20 uM W7 or 1 pM F K506 for 10 min, followed by
treatment with 0.5 uM MeHg. These concentrations were chosen based on previous E

findings by Marcaida et al. (1996) and Ankarcrona et al. (1996). Neither F K506 nor W7

altered the time-to-onset of either the first or second MeHg-induced [Ca2+]Ii phases L.

 

(Figure 5.4), indicating that CN does not play a significant role in changes in [Ca2+]i
induced by MeHg.

The effect of CsA on MeHg-induced changes in rpm was monitored using the
potentiometric dye TMRE. Control experiments using 5-15 uM of the MTP-inducing
agent PhAsO (Kowaltowski and Castilho, 1997) indicated that granule cell mitochondria
can undergo a permeability transition, which is visualized as a complete loss of TMRE
fluorescence due to depolarization of the inner mitochondrial membrane (Figure 5.5). As
seen in Figure 5.6, MeHg causes loss of TMRE fluorescence consistent with a loss of rum.
This is consistent with previous findings in intact monocytes and T cells (InSug et al. ,
1997; Shenker et al. , 1998), as well as with mitochondria—containing synaptosomes (Hare
and Atchison, 1992). Exposure to 0.5 uM MeHg alone caused loss of TMRE
fluorescence at 25.0i2.5 min (Figure 5.7). The loss of rum was irreversible even when the
MeHg was replaced with HBS, as the TMRE fluorescence did not recover during a 60
min period following MeHg exposure (figure not shown). However, this loss of

fluorescence was delayed by a 10 min pretreatment with 5 uM CsA, such that neurons

148

- MeHg only - CsA + MeHg

 

 

 

Ta“ 30' a a

E

a; 20-

c:

q

3 10 -

o

E O

H 0.5 1.0
[MeHg](uM)

Figure 5.3. 5 uM CsA delays the time-to-onset of MeHg-induced release of Ca”. in
the absence of Ca2+” In the absence of C322, 5 pM CsA delayed the time-to—onset of
[Ca2+]; elevations caused by 0.2 and 0.5, but not 1.0 uM MeHg. Results are presented as
mean as SEM (n = 3-7). The asterisk (‘) indicates a value significantly different from
control (p < 0.05).

149

Time-to-Onset (min)

Time-to-Onset (min)

- MeHg only - FK506 + MeHg
40 ’
30

20

 

 

1st Phase 2nd Phase

- MeHg only - W7 + MeHg

40
30

20

T‘—I_—'_—l_‘

   

lst Phase 2nd Phase

Figure 5.4. Effect of calcineurin inhibitors on MeHg-induced [Ca2+]i elevations. A)
A 10 min pretreatment with l LuM FK506 did not alter the time-to-onset of [Ca2+];
increases induced by 0.5 uM MeHg Sn = 6). B) A 10 min pretreatment with 20 11M W7
did not alter the time-to—onset of [Ca +]i increases induced by 0.5 uM MeHg (n = 6).

150

1.50 _ 15 M PhAsO

 

 

 

 

(D

O

c:

CD

0

U)

g 1.00 1
a

[-1-4

E 0.50
E

0)

m 0.00 l l J l l

 

O 10 20 30 40 50 60

Time (min)

Figure 5.5. PhAsO causes a rapid decrease in TMRE fluorescence. Representative
example of changes in TMRE fluorescence in a single granule cell. The known MTP-
inducing compound PhAsO (15 [AM ) causes mitochondrial membrane depolarization, as
indicated by the loss of TMRE fluorescence as measured relative to an initial
fluorescence of 100%.

151

 

1.50

1

0.5 pM MeHg

 

 

1.00

0.50

Relative Fluorescence

 

 

 

 

 

0.00 A I l l ? l 1 ‘
0 10 20 30 4O 50 60

Time (min)

Figure 5.6. Exposure to 0.5 uM MeHg causes irreversible loss of TMRE
fluorescence. In this representative example of a single granule cell loaded with TMRE,
exposure to 0.5 11M MeHg causes rapid, irreversible loss of TMRE fluorescence,
indicating mitochondrial membrane depolarization (dissipation of Wm). Time-to—onset of
mitochondrial membrane depolarization is measured at the point at which TMRE
fluorescence drops below baseline (dashed line). TMRE fluorescence is measured
relative to an initial fluorescence of 100%.

 

152

 

- MeHg only - Treatment + MeHg

A40~ *
.S
530-
‘5
a
020"
52
£210-
.§
[—1

0

 

 

CsA W7 FK506

Figure 5.7. CsA, but not CN inhibitors, delay the time-to-onset of mitochondrial
membrane depolarization caused by 0.5 pM MeHg. Pretreatment with 5 uM CsA
significantly delayed the loss of wm; however, pretreatment with 1 uM FK506 or 20 uM
W7 did not. Results are presented as mean d: SEM (n = 5-9). The asterisk (*) indicates a
value significantly different from the corresponding 0.5 uM MeHg data (p < 0.05).

153

receiving CsA did not lose TMRE fluorescence until 34.6:l:2.8 min. Mitochondrial
membrane depolarization was not delayed by 10 min pretreatment with 1 pM FK506 or
20 nM W7 (Figure 5.7). Removal of Ca2+, with EGTA did not alter the time-to-onset of
the depolarization caused by 0.5 pM MeHg (data not shown). The presence or absence

of the initial K+ challenge also did not affect the MeHg-induced loss of rpm (data not
shown).

CsA prevents cell death caused by anoxia and glutamate excitotoxicity,
presumably by inhibiting the MTP (Pastorino et al., 1995; White and Reynolds, 1996).
Because CsA altered changes in [Ca2+]; and rum caused by MeHg, the ability of CsA to
prevent MeHg-induced cell death was examined. Pretreatment for 10 min with 1 uM
CsA protected cells against subsequent exposure to 0.2 uM MeHg, but not against higher
concentrations of MeHg (Figure 5.8). Increasing the CsA concentration to 5 uM was
more effective, and protected against cell death caused by 0.5 1.1M MeHg but again, not
against that caused by higher MeHg concentrations. Thus CsA is cytoprotective against
acute exposure to low concentrations of MeHg.

Viability assays were also performed comparing the effect of 10 min pretreatment
with 5 uM CsA vs. 5 uM CsA given with MeHg on the MeHg-induced cell death
observed at 24 hr post-exposure (Figure 5.9). As with the comparison of the effect of
pretreatment and simultaneous treatment with CsA on MeHg-induced increases in
[Ca2+]i, there was no difference between these two protocols in their effectiveness in
preventing MeHg-induced cell death.

The ability of 20 uM W7 and 1 uM FK506 to prevent 0.5 uM MeHg-induced cell

death was examined using a parallel pretreatment regimen. Treatment with 1 uM F K506

154

for 10 nrin prior to 0.5 pM MeHg did not improve cell viability at 24 hr post-exposure
(data not shown). Pretreatment with 20 11M W7 for 10 min significantly protected
against MeHg-induced cell death, increasing the percent viability from 41.3 i 5.6% in
cells treated with 0.5 uM MeHg only, to 61.6 i 4.6% in cells treated with W7 just prior
to 0.5 uM MeHg (Figure 5.10). However, W7 was unable to reverse completely the

MeHg-induced cell death and did not protect cells to the extent CsA did.

155

 

 

- 5pMCsA W luMCsA l:l OpMCsA

%Viability

 

m\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\‘<

 

 

 

 

MeHgGlM)

Figure 5.8. CsA provides limited protection against MeHg-induced cell death at 24
hr post-exposure. Granule cell viability following exposure to 0, 1 or 5 uM CsA and
increasing concentrations of MeHg at 24 hr post-MeHg exposure. Increasing
concentrations of CsA resulted in increasing protection against MeHg-induced cell death.
At 1 uM CsA, cells were protected from death caused by 0.2 uM MeHg, whereas 5 uM
CsA also protected cells from death from MeHg at concentrations up to 0.5 11M.
However, protection was lost at higher MeHg concentrations. Results are presented as
mean i SEM (n = 4). The asterisk (*) indicates a value significantly different from the
corresponding 0 uM MeHg data (p < 0.05).

156

- CsA pretreat - CsA + MeH [iii MeHg only

0 0.2

 
 

LL **

"f a”;

'1 l

~fl<
1.0 2.0

% Viability

80 * if
60 ‘
40 ;
20 L
l

O .

solv 0.5

 

MeHg (PM)

Figure 5.9. Comparison of protection against MeHg-induced cell death by
pretreatment or simultaneous treatment with 5 uM CsA. Granule cell were
pretreated with 5 uM CsA for 10 min (“CsA pretreat”) prior to exposure to increasing
concentrations of MeHg, or with 5 uM CsA simultaneously with increasing
concentrations of MeHg (“CsA+MeHg”). Results are presented as mean :t SEM (n = 4).
The asterisk (*) indicates a value significantly different from the corresponding 0 uM
MeHg data (p < 0.05).

157

% Viability

 

 

solv 0 0.5 0.5 + W7
[MeHg] (MM)

Figure 5.10. Pretreatment with 20 uM W7 significantly protects against cell death
caused by 0.5 pM MeHg at 24 hr post-exposure. Results are presented as mean i
SEM (n = 4). The asterisk (*) indicates a value significantly different from the
corresponding 0.5 11M MeHg data (p < 0.05).

158

DISCUSSION

These experiments examined the role of MTP induction in cerebellar granule cell
neurotoxicity induced by acute exposure to MeHg. The MTP inhibitor CsA was used to

define the role of the MTP in MeHg-induced changes in Ca2+ homeostasis. Assuming

that CsA prevents mitochondrial Ca2+ release through the MTP, it was anticipated that the
first phase increase of Ca2+; caused by MeHg would be affected by CsA; surprisingly,
both the first and second phases were delayed, suggesting that the two phases of

increased [Ca2+]i are interrelated with the first phase perhaps triggering or contributing in

some way to the onset of the second phase. The CsA-mediated delay was not dependent
on Ca2+¢, as indicated by the similar first phase times-to-onset in the presence and
absence of Cafe. Pretreatment with CsA delayed, but did not prevent, MeHg-induced

changes in [Ca2+]i, agreeing with the short-lived protection of CsA found in other systems

(N ieminen et al., 1996). The CN inhibitors W7 and F K506 did not change the time-to-
onset of increased Capri, indicating that CN does not contribute substantially to the
delayed MeHg response induced by CsA. Thus, it appears that the MTP participates in
the initial increase in [Ca2+]; by releasing Ca2+ into the cytosol upon its opening. Whether
the elevated [Ca2+]i is a causative factor or a reaction to a permeability transition cannot

be determined fiom these experiments.

Previous studies indicate that the second phase of increased [Ca2+]; is due to entry
of Ca2+e, with a possible contribution of L, N and/or P/Q-type voltage-gated Ca2+

channels (Marty and Atchison, 1997). To date, no inhibitor has been able to block

completely this second phase. It was hypothesized that initiation of the second phase

159

 

 

may be dependent on the first phase, either though pathways to replenish Ca2+; stores

(depleted by release of Ca2+ from the SER, for example) or through another as yet
undefined mechanism. These experiments provide evidence of a link between the two

phases, as delay of the first phase with CsA is accompanied by delay of the second phase.

Thus, induction of the MTP may link the first and second Ca2+ phases as well as be a
critical factor in determining cell survival following MeHg exposure.
Using TMRE, it was found that mitochondria depolarize irreversibly in response

to a low concentration (0.5 uM) of MeHg. Because an identical depolarization occurs in

the absence of Ca2+c, it is concluded that this depolarization is dependent on an

intracellular factor (such as elevated [Ca2+]i), on MeHg alone, or on a combination of

direct damage by MeHg and other cellular factors. Neither CsA treatment nor
replacement of the MeHg with normal medium reversed the loss of Wm, suggesting that
MeHg binds at least one critical site with high affinity. MeHg exhibits high-affinity
binding to intracellular organelles, especially mitochondria (Chang and Hartmann, 1972),
thereby altering protein sulflrydryl groups and disrupting cellular functions, including
respiration, ion transport and regulation, and protein synthesis (Verity et al., 1977;
Kauppinen et al., 1989; Levesque and Atchison, 1991; Yee and Choi, 1996). Because
MTP induction depends on critical, voltage-sensitive thiol groups (Scorrano et aI. , 1997),
it is possible that direct binding of MeHg to a site on the MTP could induce a
permeability transition. MeHg binding would likely be irreversible, which could explain
the irreversible mitochondrial depolarization observed with MeHg exposure. This is in
contrast to other MTP-inducing agents; for example, exposure to low concentrations of

NMDA in cultures of rat hippocampal neurons causes a mitochondrial depolarization

160

 

which recovers to near normal levels when the neurons also receive CsA (Schinder et al.,
1996). MeHg exposure, on the other hand, does not permit recovery of 01.... This
suggests a more severe, and possibly more direct insult to the mitochondria which may be
exacerbated by elevated cytosolic Ca2+, thereby overwhelming the cell and leading to
neuronal death.

Despite its inability to prevent mitochondrial depolarization, CsA was effective at
protecting cerebellar granule cells fiom cell death at 24 hr following exposure to low
concentrations of MeHg. Cytotoxicity assays performed at 24 hr post-exposure indicate
that a significant fraction of the cultured cells undergo cell death; this fi'action increases
with increasing MeHg concentrations. These results agree with earlier findings by Marty
and Atchison, 1998. In previous experiments, treatment with the Ca2+; chelator BAPTA
protected cerebellar granule cells from death 3.5 h after exposure to low concentrations of

MeHg (Marty and Atchison, 1998). Because onset of a permeability transition is favored

by elevated Ca2+i levels, and cell death at low MeHg concentrations is attenuated by

chelating [Ca2+]i with BAPTA, these experiments suggest that elevated [Ca2+]i

contributes to a CsA-sensitive permeability transition which leads to cell death. As
higher MeHg concentrations are used, less protection is provided by CsA, indicating a
limit to its effectiveness. The CN inhibitor W7 also reduced the percentage of cells killed
by 0.5 uM MeHg; however, W7 did not significantly alter the MeHg-induced changes in
[Ca2+]i or rpm. Thus, its effects on cell viability may not be due to inhibition of CN but
rather to some other consequence of calmodulin inhibition. The inhibition of MeHg-
induced cell death by CsA is interesting, as CsA delays but does not prevent the onset of

the MTP in granule cells. The observed inhibition of cell death in these experiments may

161

 

simply reflect a delay in the onset of cell death in the presence of CsA, rather than actual
prevention of cytotoxicity.

These experiments emphasize the importance of mitochondrial Ca2+ homeostasis
in MeHg-induced neurotoxicity. In recent years, the importance of mitochondrial Ca2+
regulation has become evident not only in neuronal damage caused by a variety of agents,
including glutamate excitotoxicity in cerebellar granule cells (Budd and Nicholls, 1996a),
but also in regulating normal cell physiology (see Duchen, 2000, for review). Loss of
mitochondrial Ca2+ buffering is crucial in both apoptotic and necrotic Cay-mediated
neuronal death (Kruman and Mattson, 1999). Mitochondria accumulate Ca2+ based on
the extremely negative membrane potential within the inner mitochondrial membrane
(140-180 mV, negative inside) (Gunter and Gunter, 1994), making Ca2+ buffering an
energetically favorable process. However, Ca2+ buffering comes at a price: uptake of a
large amount of Ca2+ occurs at the expense of the proton motive force, and can thus cause
uncoupling of ATP synthesis from respiration. Under physiological conditions,
mitochondrial Ca2+ buffering is important in regulating cytosolic [Ca2+] elevations which
result fiom either influx of Ca2+: or release of Ca2+ from the SER, and may play a role in
modulating cellular Ca2+ oscillations (Simpson and Russell, 1996; Rizzuto et al., 1998).
However, high [Ca2+]... can become hazardous to the cell via inhibition of the electron
transport chain and the mitochondrial ATPase, as well as dissipation of the proton
gradient, all of which can contribute to loss of mitochondrial ATP synthesis, opening of
the MTP and cell death (Gunter and Gunter, 1994; Simpson and Russell, 1998b). Loss of
mitochondrial Ca2+ regulation appears to be a central feature of MeHg-induced loss of

neuronal Ca2+ regulation (Levesque and Atchison, 1991). In fact, emptying of

162

 

 

mitochondrial Ca2+ stores prior to exposure to 0.5 11M MeHg decreases the amplitude of
the MeHg-induced first-phase fura-2 ratio increase in granule cells by approximately
65% (see Figure 4.3B), indicating the significant contribution of mitochondrial Ca2+
stores to MeHg-induced [Ca2+]; elevations in these highly sensitive cells.

By examining the changes which occur in cerebellar granule cells during MeHg
exposure, several possible reasons for their heightened sensitivity to toxicity are revealed.
For example, mitochondria are not all identical, and brain mitochondria exhibit
differential sensitivity by region (F riberg et al. , 1999). Because these experiments
indicate that MeHg induces a permeability transition in granule cells, it is possible that
their mitochondria are more sensitive to conditions which favor MTP induction, such as

elevated [Ca2+]i and production of ROS. Cerebellar granule neurons in culture have a

relatively weak capacity to up-regulate GSH levels in response to MeHg as compared to
cerebellar glia; this may contribute to a heightened sensitivity to onset of the MTP

(Sarafian and Verity, 1991). Further, the difference in sensitivities may be related to

different Ca2+ buffering capacities. Granule cells lack certain Ca2+ buffering proteins
present in neighboring cells which are less sensitive to MeHg. For example, Purkinje
cells contain high concentrations of the calcium binding proteins parvalbumin and
calbindin D-zgx (Celio, 1990). Because MTP induction relies on increased
extramitochondrial Ca2+ levels for Ca2+ loading, it is possible that a lack of sufficient
Ca2+ binding proteins puts more of a burden on mitochondria to buffer sustained
increases of [Ca2+]; in cerebellar granule cells.

MeHg exposure has also been linked to elevated levels of ROS, which are

believed to contribute to MeHg-induced neurotoxicity (Sarafian and Verity, 1991).

163

However, elevated levels of ROS may not be sufficient for MeHg to be lethal. In

 

experiments using isolated rat liver mitochondria, sensitivity of the MTP to ROS was

shown to be highly dependent on the mitochondrial Ca2+ load (Scorrano et al. , 1997).

Moreover, mitochondrial Ca2+ uptake is an early, critical event in both apoptosis and

necrosis in the neuronal model PC12 cells (Kruman and Mattson, 1999). In these cells,
an initial increase in cytosolic Ca2+ was followed by a delayed increase in mitochondrial

Ca2+, then accumulation of ROS and mitochondrial depolarization, caspase release and

cell death. Both the mitochondrial depolarization and cell death were CsA-sensitive,

 

implicating the permeability transition as a key event. This is remarkably similar the

effect of MeHg in granule cells, in which an initial increase in [Ca2+]i (resulting fiom

release of Ca2+i stores) is followed by mitochondrial depolarization and cell death, with

all three components exhibiting sensitivity to CsA. Thus, it appears that the disruption of
[Ca2+]; homeostasis induced in granule neurons by MeHg is the critical factor in MeHg-

induced cell death, with the MTP playing a critical role.

164

 

CHAPTER SIX

SUMMARY AND CONCLUSIONS

165

SUMMARY AND CONCLUSIONS

This dissertation presents the results of experiments designed to examine the
hypothesis that MeHg causes Ca2+ dysregulation in cerebellar granule cells by interacting
with specific intracellular Ca2+ stores. Specifically, it was hypothesized that MeHg
initiates Ca2+ release from the SER which is buffered by mitochondria; however, MeHg
also impairs the ability of mitochondria to hold Ca2+ (by causing loss of mitochondrial
membrane potential and inhibiting respiration), resulting in opening of the mitochondrial
permeability transition pore and subsequent cell death. The results presented in this
dissertation not only support this hypothesis but also provide additional information
regarding the mechanisms underlying MeHg-induced elevations of [Ca2+]; and
subsequent neuronal death. The summary of the results delineated in these experiments
is presented in Figure 6.1.

In the first experiments, the role of SER Ca2+ in MeHg-induced changes in [Ca2+]i
and cell viability was examined. Overall, the results suggest that the SER plays a small
role in the whole-cell changes in [Ca2+]; caused by MeHg in rat cerebellar granule cells.
This is not surprising, given the small capacity of this Ca2+ store (F ohrman et al., 1993;
Masgrau et al., 2000). The Ca2+ was released primarily from 1P3 receptors, with a
minimal contribution of Ca2+ fiom the ryanodine receptors. Atropine delayed the time-
to—onset and reduced the amplitude of MeHg-induced [Ca2+]; increases, indicating that
this occurred as a result of activation of muscarinic ACh receptors. However, the
potential interaction between MeHg and the 1P3 receptors themselves is not ruled out by
these experiments. SER store depletion does not appear to trigger Ca2+ influx through

SOCE, as none of the inhibitors of SER Ca2+ release affected

166

 

 

    

 
   

lCO" In: {If

\

ti

SERCA

can

Figure 6.1. Summary of findings. MeHg binds M3 muscarinic ACh receptors, causing
production of 1P3 and release of Ca2+ through the 1P3 receptors. This elevation of [Ca2+],
triggers Ca2+-induced Ca2+ release through the ryanodine receptors. Ca2+ released fiom
the SER is buffered by mitochondria. Excess uptake of Ca2+ into the mitochondria,
combined with direct effects of MeHg on the mitochondrial respiratory chain, leads to
opening of the mitochondrial permeability transition pore (MTP) and promotion of cell
death. However, there may be direct effects of MeHg on mitochondria which also
contribute to the observed changes in cellular function (dashed arrow).

1P,
Receptors

RYanodtne
Receptors

 

ism

   

167

the MeHg-induced influx of Cafe.

Despite the small amount of Ca2+ release fi'om the SER store, inhibition of this
Ca2+ release was a determinant factor in later cell death caused by MeHg. Cell death was
prevented by down-regulation of muscarinic and IF; receptors using a 24 hr treatment
with BCh. This protection was likely mediated by the M3 receptor and not the nicotinic
ACh receptor, as protection was not afforded by BCh when 4-DAMP or atropine were
included, but was not affected by the addition of DHE. The M3 receptor is coupled to P3
production in granule cells (Whitham et al., 1991; F ohrman et al., 1993), thus inhibition
of M3 receptors presumably prevents a MeHg-induced increase in [1P3]. The potential
contribution of the M2 receptor to MeHg-induced neurotoxicity was not examined as a
result of the toxicity of methoctramine alone. However, typically activation of M2
muscarinic receptors is coupled to activation of potassium conductance (Pan and
Williams, 1994), which would not be expected to facilitate Ca2+ release or entry.
Indiscriminate release of Ca2+ from the SER is insufiicient to contribute to MeHg-
induced cell death, as indicated by the lack of protection afforded by ryanodine.
Surprisingly, atropine alone and XeC were not neuroprotective, in contrast to the ability
of atropine to delay MeHg-induced elevations of [Ca2+]i. In sum, these experiments
provide evidence that the SER participates in MeHg-induced elevations of [Ca2+]; and
subsequent neuronal death.

In the second set of experiments, the role of mitochondrial Ca2+ in MeHg-induced
elevations of [Ca2+]; and cell death was examined. Unlike the small contribution of SER
Ca2+, mitochondria contribute a large portion of the [Ca2+]; observed in cerebellar granule

cells during exposure to MeHg. Removal of mitochondrial Ca2+ prior to MeHg exposure

168

decreased the amplitude of the Ca2+ release from intracellular sources by more than 50%.
Observation of cells loaded with rhod-2 to monitor changes in [Ca2+]... indicated that
MeHg causes an early increase in [Ca2+]... fiom Ca2+ originating within the cell. This
increase in [Ca2+],fl is followed several minutes later by spreading of the dye from a
punctate stain, presumably mitochondrial in origin, to more generalized whole-cell
localization. The spreading of the dye beyond the mitochondria was delayed by
thapsigargin, suggesting that SER Ca2+ contributes to the elevated [Ca2+]m. The inhibitor
of MTP opening CsA also delayed the spreading of rhod-2 beyond the punctate pattern,
suggesting that the dye movement results from opening of the MTP to allow the dye to
cross rapidly through the usually impermeable inner mitochondrial membrane. The
participation of the MTP in MeHg neurotoxicity is implicated further by the ability of
CsA to attenuate MeHg-induced elevations of [Ca2+]i, loss of mitochondrial membrane
potential, and cell death. Calcineurin inhibitors did not affect the MeHg-induced increase
in [Ca2+]; or loss of mitochondrial membrane potential. However, the calmodan
inhibitor W7 did protect against MeHg-induced cell death at 24 hr post-exposure. This is
not surprising, because calmodulin is known to play a role in Calf-mediated cell death
(Orrenius and Nicotera, 1994; Urushitani et al., 2001).

The impact of MeHg-induced alterations on [Ca2+]... has widespread implications
for the viability of affected neurons, leading to a proposed scenario in which MeHg-
induced disruption of mitochondrial firnction is a focal point for many of the observed
changes which occur in cerebellar granule neurons. As demonstrated in this dissertation,
MeHg causes an early increase in [Ca2+]In as a result of Ca2+ uptake into the

mitochondrial lumen. Initially, this uptake of Ca2+ into the mitochondria by the Ca2+

169

 

 

uniporter would be driven by the mitochondrial membrane potential (140-180 mV,
negative inside) (N icholls, 1974); however, as the [Ca2+]m increases, the inner membrane
potential becomes more positive, thereby making uptake of Ca2+ less favorable
energetically (N icholls and Scott, 1980). As the transmembrane potential dissipates,
mitochondria buffer Ca2+ using ATP-dependent mechanisms (N icholls and Scott, 1980),
which diverts a portion of the neuron’s ATP away fiem other cellular processes. The
inner membrane depolarization reduces the proton motive force which drives ATP
mitochondrial ATP synthesis, thus elevated [Ca2+]... can effectively uncouple ATP
synthesis from respiration (Hillered et al., 1983). Additionally, high [03*]... can inhibit
the electron transport chain and the mitochondrial ATP synthase, which results in
inhibition of mitochondrial ATP synthesis (see Gunter and Gunter, 1994, and Simpson
and Russell, 1998b, for review). In an effort to maintain trim, the ATP synthase can
reverse, and begin hydrolyzing ATP to ADP to produce the protons necessary to maintain
the negative inner membrane potential (Leyssens et al., 1996). The end result is
depletion of cellular ATP, which has a wide range of effects within the cell. Loss of ATP
deprives plasma membrane Ca2+ extrusion proteins of the energy necessary to maintain
the steep [Cay] gradient across the plasma membrane, and reduces the amount of ATP
available for Ca2+ pumps on storage organelles, such as the SERCA pump of the SER.
Additionally, loss of ATP inhibits the Nai’-K+ ATPase, which maintains the membrane
potential across the plasma membrane. Inhibition of the Na+-K+ ATPase would lead to
plasma membrane depolarization, thereby leading to activation of plasma membrane
voltage-gated Ca2+ channels and influx of Ca2+c into a cell now compromised in its ability

to extrude Ca2+ fiom the cytoplasm. ATP depletion, loss of rpm and elevated [Ca2+]... all

170

contribute to induction of the MTP and promotion of cell death (Bemardi and Petronilli,
1996), and each of these actions has been shown to occur with MeHg (Levesque and
Atchison, 1991; Hare and Atchison, 1992; present thesis). The far-reaching
consequences of mitochondrial effects are supported by experiments presented in this
dissertation and elsewhere. For example, the ability of oligomycin to delay the time-to-
onset of MeHg-induced influx of Ca2+c (Figure 4.5) suggests that MeHg-induced
depletion of ATP triggers influx of Ca2+, in cerebellar granule cells. This influx occurs
primarily through voltage-gated Ca2+ channels (Marty and Atchison, 1997) and occurs at
approximately the same time as the onset of plasma membrane depolarization
(Stringfellow et al. , 2000). Thus, the mitochondria are not only affected directly by
MeHg, but are also potentially responsible for a large number of the observed cellular
responses which lead to neuronal damage and death.

The results presented in this dissertation support the idea that MeHg neurotoxicity
results fiom disruption of an intricate signaling network in which disruption of one
portion of the network affects the functioning of another portion such that studying each
component separately does not provide an adequate picture of the mechanism underlying
MeHg neurotoxicity. As suggested by results presented here, the importance of the SER
as a target in MeHg neurotoxicity may lie in the effect of Ca2+ released from the SER on
mitochondria. Under normal conditions, granule cell mitochondria do not store Ca2+;
however, when they are exposed to a high extra-mitochondrial [Caz+], they usually
respond by buffering this Ca2+ via several uptake mechanisms (the uniporter, the rapid
active mode transporter, etc.) (Budd and Nicholls, 1996a; Budd and Nicholls, 1996b).

Experiments using specifically-targeted, Cay-sensitive proteins demonstrate that opening

171

 

 

of the IP3 receptor-activated channels results in higher [Ca2+] at the mitochondrial surface
than in the general cytosol (Rizzuto et al. , 1998), suggesting a preferential signaling
pathway in which Ca2+ stored within the SER lumen is released to communicate a pro-
apoptotic signal to mitochondria. Excess mitochondrial matrix Ca2+ contributes to cell
death by promoting release of cytochrome C, enhancing production of reactive oxygen
species, inhibiting mitochondrial ATP synthesis, and opening the MTP, all of which can
contribute to cellular damage and death (Bemardi and Petronilli, 1996; Szalai et al. ,
1999). In HeLa cells, overexpression of calreticulirr, an SER luminal protein which binds
Ca2+, causes an increased sensitivity to induction of apoptosis and increase in release of
cytochrome C fiom the mitochondria, highlighting the role of SER-mitochondria
communication in apoptosis (N akamura et al. , 2000). By causing pathological release of
Ca2+ from the SER which contributes to the mitochondrial dysfunction, MeHg causes a
complicated pattern of Ca2+; changes which contribute to cell death.

In interpreting the data presented in this dissertation, there are several
inconsistencies which complicate the process of delineating a precise, mechanistic model
for the actions of MeHg in cerebellar granule cells. The model (as shown in Figure 6.1)
focuses only on effects of MeHg on Ca2+ signaling systems within the cell. In this model,
there is an assumption that any effects on Ca2+ signaling are due to a direct interaction
between MeHg and the receptor signaling system being examined. In reality, MeHg-
induced disruption of [Ca2+]i may reflect disruption of a non-Ca2+ system which in turn
affects [Ca2+]i. For example, MeHg increases free [Zn2+] within the cytosol in rat cortical
synaptosomes (Denny et al., 1993; Denny and Atchison, 1994) and possibly within

NG108-15 neuroblastoma cells (Hare et al., 1993) and rat cerebellar granule cells (Marty

172

and Atchison, 1997). MeHg also causes an elevation of ROS within rat cerebellar
granule cells and human T lymphocytes (Sarafian and Verity, 1991; Oyama et al., 1994;
Yee and Choi, 1996; Shenker et al., 1999). Because both Zn2+ and ROS are known
inducers of the MTP (Wudarczyk et al. , 1999), the observed MeHg-induced elevation of
[Ca2+]; that is CsA-sensitive (see Chapter Five) may be an indirect result of MeHg-
induced elevations of Zn2+ and/or ROS. Similarly, MeHg depletes cellular
concentrations of ATP (Kuznetsov et al., 1987; Sarafian et al. , 1989), which are critical
for maintenance of [Ca2+]; by providing energy to both intracellular Ca2+ pumps (such as
the SERCA) and plasma membrane extrusion proteins (such as the plasma membrane
Ca2+ pump) (Orrenius and Nicotera, 1994; Guerini et al., 1998). Like Zn2+ and ROS,
mitochondrial ATP depletion is a contributing factor to opening of the MTP (Simbula et
al., 1997), which could then contribute to elevation of [Cay]; through opening of the
MTP. Thus, the observed elevation of [Cay]; in cerebellar granule cells may be
symptomatic of other, non-Ca2+ effects of MeHg within the cell.

Additionally, the non-Ca2+ effects may contribute to the direct effects of MeHg on
Ca2+ signaling pathways that are shown in Figure 6.1. For example, MeHg causes
mitochondrial membrane depolarization associated with opening of the MTP (see
Chapters Four and Five). However, mitochondrial depolarization may not be sufficient to
cause opening of the MTP without contributing factors, such as elevations of Ca2+... or
elevation of ROS. In hepatocytes, mitochondrial membrane depolarization with CCCP is
not suflicient to open the MTP; however, CCCP plus the pro-oxidant t-
butylhydroperoxide causes rapid opening of the MTP which is accompanied by ATP

depletion and subsequent cell death (N ieminen et al., 1995). This phenomenon of

173

 

 

mitochondrial membrane depolarization without opening of the MTP may underlie the
effect of CCCP and oligomycin on MeHg-induced elevations of rhod-2 fluorescence
(Figure 4.8). The combination of CCCP and oligomycin causes rapid loss of rum but does
not prevent the movement of rhod-2 fiom localized to diffuse staining when cells are
exposed to 0.5 uM MeHg. This suggests that there is another factor contributing to
opening of the MTP, such as production of ROS or elevation of cytosolic [Zn2+], which
occurs during MeHg exposure and which was not considered in the experiments
undertaken for this dissertation.

Another inconsistency in this model is the assumption that MeHg has specific
interactions which occur in the same temporal order during every MeHg exposure in
cerebellar granule cells. The model presented in Figure 6.1 indicates an initial interaction
with muscarinic receptors, followed by release of Ca2+ fi'om the SER and subsequent
disruption of Ca2+... regulation. The problem with this assumption is apparent in the
inconsistencies observed in the time-to—onset data for the first-phase increase of [Ca2+];
caused by MeHg. For example, both atropine, which indirectly affects release of Ca2+
fi-om the SER, and CsA, which inhibits opening of the MTP, delay the time-to-onset of
MeHg-induced release of [Ca2+]t. These data contradict the proposed model because they
suggest that MeHg is targeting both the SER and mitochondria simultaneously rather than
sequentially, as suggested by the thapsigargin-sensitive nature of MeHg-induced
increases of C322... Another discrepancy is apparent in the lack of effect of thapsigargin
on the time-to-onset of MeHg-induced elevations of [Ca2+]i. Both BCh and atropine
changed the time-to-onset, but thapsigargin had no effect on the timing of the [Ca2+];

increase. While this can be explained by the presence of a thapsigargin-insensitive store

174

in cerebellar granule cells, as has been suggested by others (Masgrau et al., 2000), this

hypothesis was not tested. Thus, while the data presented in this dissertation provide

 

strong evidence for participation of M3 muscarinic receptors, 1P3 receptors and the MTP
in MeHg-induced elevations of [Ca2+]; and subsequent neuronal death, they do not yet
provide a clear picture of the temporal order of the events which take place during the
initial exposure period in cerebellar granule cells.

Regardless of the source(s) of the initial increase of [Ca2+]i, in all experiments
there is a latent period between exposure to MeHg and the increase in [Ca2+]; which has
been observed in this dissertation and elsewhere (Hare et al., 1993; Marty and Atchison,
1997). This latent period decreases as the concentration of MeHg increases, suggesting a
threshold for the Ca2+ response. The physiological basis for this threshold is unknown.
Hypothetically, the threshold could depend on saturation of binding of MeHg to proteins
not directly involved in the observed elevation of [Ca2+]i. For example, MeHg binds
thiols contained in cytosolic GSH (Sarafian and Verity, 1991; Sanfeliu et al., 2001).
Saturation of intracellular GSH would not only increase the concentration of non-bound
MeHg in the cytosol but would also deplete the GSH reserve, thereby interfering with the
cell’s ability to manage oxidative stress (Sanfeliu et al., 2001). As discussed above,
oxidative stress contributes to opening of the MTP, thus MeHg-induced elevations of
[Ca2+]i could potentially depend on depletion of GSH. The latent period could also result
from saturation of extracellular, non-specific binding sites. Many plasma membrane
proteins, including muscarinic and nicotinic ACh receptors, as well as NMDA receptors,
contain cysteine residues which regulate receptor activity and which could potentially

participate in non-specific binding of MeHg (Sato et al., 1976; Janaky et al., 1993; Zeng

175

et al. , 1999). Whereas none of these possibilities was tested directly in experiments
presented in this dissertation, MeHg has been shown to impair function of neuronal
nicotinic and muscarinic receptor-mediated responses in NIB-115 neuroblastoma cells
(Quandt et al., 1982) and of glutamate receptors (Yuan and Atchison, 1997).

The model presented in this dissertation is based on pharmacological
manipulations which did not provide sufficient evidence to answer unequivocally the
question of whether MeHg interacts with specific, potentially unique molecular targets in
cerebellar granule cells. An alternative approach which could potentially provide this
answer would be based on genetic manipulation of biological systems to add and/or
remove specific proteins fiom cells and, when possible, experimental animals.
Theoretically, a better approach would use transgenic methodology to create, for
example, 1P3 knockout mice to remove the 1P3 receptor fi'om mouse cells and observe
whether its removal alters the effects of MeHg in the whole animal and/or cells derived
from these animals. However, experiments of this nature are limited by cost, the time
needed to engineer genetically-altered animals, and the limited viability of some strains
of knockout mice. Additionally, animals often compensate by substituting a different
isoform of the receptor, which can complicate interpretation of data gathered fi'om this
type of experiment. Thus, genetic manipulation provides a potentially enlightening
experimental approach to determining the specific molecular target(s) of MeHg, but as
observed with the pharmacological approaches, is also constrained by potential problems.

In conclusion, this dissertation provides evidence that MeHg alters [Ca2+];
regulation by causing early, irreversible release of Ca2+ from both the SER and

mitochondria to cause cell death in cerebellar granule cells. In cerebellar granule cells,

176

MeHg-induced neuronal death is prevented by inhibition of opening of the MTP or by
inhibition of Ca2+ release through the 1P3 receptor, with protection falling off as the
MeHg concentration exceeds 1 uM. It is interesting to note that, while MeHg has many
effects which can be harmful (i. e., generation of reactive oxygen species, disruption of
protein and ATP synthesis, etc.), the highly sensitive granule cells can be protected by
interfering with the MeHg—induced release of Ca2+ fi'om intracellular stores (Sarafian et
al., 1984; Sarafian, 1999). Because there is great diversity in the location and abundance
of intracellular Ca2+-release channels, particularly the IP3 receptor, and the location of
their upstream receptors, such as the muscarinic receptors, within different areas of the
nervous system, this may contribute to the selective neurotoxicity of MeHg within the

cerebellum.

177

REFERENCES

ABD-ELFATTAH, A. S. and SHAMOO, A. E. (1981). Regeneration of a functionally
active rat brain muscarinic receptor by D-penicillarnine after inhibition with
methylmercury and mercuric chloride. Mol. Pharmacol. 20(3): 492-497.

ABRAMSON, J. J., TRIMM, J. L., WEDEN, L. and SALAMA, G. (1983). Heavy metals
induce rapid calcium release from sarcoplasmic reticulum vesicles isolated from
skeletal muscle. Proc. Natl. Acad Sci. U. S. A. 80(6): 1526-1530.

ALI, S. F., LEBEL, C. P. and BONDY, S. C. (1992). Reactive oxygen species formation
as a biomarker of methylmercury and trimethyltin neurotoxicity.
NeuroToxicoIogy 13(3): 637-648.

ALNEMRI, E. S. and LITWACK, G. (1990). Activation of intemucleosomal DNA
cleavage in human CEM lymphocytes by glucocorticoid and novobiocin.
Evidence for a non-Ca2m-requiring mechanism(s). J. Biol. Chem. 265(28):
17323-17333.

ALONSO, R., DIDIER, M. and SOUBRIE, P. (1990). [3H]N-methylscopolamine binding
studies reveal M2 and M3 muscarinic receptor subtypes on cerebellar granule
cells in primary culture. J. Neurochem. 55(1): 334-337.

ANDREYEV, A. and F ISKUM, G. (1999). Calcium induced release of mitochondrial
cytochrome c by different mechanisms selective for brain versus liver. Cell Death
Differ. 6(9): 825-832. -

ANKARCRONA, M, DYPBUKT, J. M, BONFOCO, E., ZHIVOTOVSKY, B,
ORRENIUS, s, LIPTON, s. A. and NICOTERA, P. (1995). Glutamate-induced

neuronal death: a succession of necrosis or apoptosis depending on mitochondrial
frmction. Neuron 15(4): 961-973.

ANKARCRONA, M., DYPBUKT, J. M, ORRENIUS, S. and NICOTERA, P. (1996).
Calcineurin and mitochondrial function in glutamate-induced neuronal cell death.
FEBS Lett. 394(3): 321-324.

ARENDS, M. J ., MORRIS, R. G. and WYLLIE, A. H. (1990). Apoptosis. The role of the
endonuclease. Am. J. Pathol. 136(3): 593-608.

ARONICA, E., CONDORELLI, D. F., NICOLETTI, F., DELL'ALBANI, P., AMICO, C.

and BALAZS, R. (1993). Metabotropic glutamate receptors in cultured cerebellar
granule cells: developmental profile. J. Neurochem. 60(2): 559-565.

178

ARONICA, E., DELL'ALBANI, P, CONDORELLI, D. F., NICOLETI‘I, F., HACK, N.
and BALAZS, R. (1993). Mechanisms underlying developmental changes in the
expression of metabotropic glutamate receptors in cultured cerebellar granule

cells: homologous desensitization and interactive effects involving N-methyl-D-
aspartate receptors. Mol. Pharmacol. 44(5): 981-989.

ARONICA, E., NICOLETII, F., CONDORELLI, D. F. and BALAZS, R. (1993).
Pharmacological characterization of metabotropic glutamate receptors in cultured
cerebellar granule cells. Neurochem. Res. 18(5): 605-612.

ASCHNER, M. (1996). Astrocytes as modulators of mercury-induced neurotoxicity.
NeuroToxicology 17(3-4): 663-669.

ASCHNER, M. and ASCHNER, J. L. (1990). Mercury neurotoxicity: mechanisms of
blood-brain barrier transport. Neurosci. Biobehav. Rev. 14(2): 169-176.

ASCHNER, M, MULLANEY, K. J ., WAGONER, D, LASH, L. H. and KIMELBERG,
H. K. (1994). Intracellular glutathione (GSH) levels modulate mercuric chloride
(MC)- and methylmercuric chloride (MeHgCl)-induced amino acid release from
neonatal rat primary astrocytes cultures. Brain Res. 664(1-2): 133-140.

ASCHNER, M, YAO, C. P, ALLEN, J. W. and TAN, K. H. (2000). Methylmercury
alters glutamate transport in astrocytes. Neurochem. Int. 37(2-3): 199-206.

ATCHISON, W. D. (1986). Extracellular calcium-dependent and -independent effects of
methylmercury on spontaneous and potassium-evoked release of acetylcholine at
the neuromuscular junction. J. Pharmacol. Exp. Ther. 237(2): 672-680.

ATCHISON, W. D, ADGATE, L. and BEAMAN, C. M. (1988). Effects of antibiotics
on uptake of calcium into isolated nerve terminals. J. Pharmacol. Exp. Ther.
245(2): 394-401.

ATCHISON, W. D, JOSHI, U. and THORNBURG, J. E. (1986). Irreversible
suppression of calcium entry into nerve terminals by methylmercury. J.
Pharmacol. Exp. Ther. 238(2): 618-624.

ATCHISON, W. D. and NARAHASHI, T. (1982). Methylmercury-induced depression of
neuromuscular transmission in the rat. Neuro Toxicology 3(3): 37-50.

BAKIR, F., DAMLUJI, S. F., AMIN-ZAKI, L, MURTADHA, M, KHALIDI, A, AL-
RAWI, N. Y, TIKRITI, S, DAHAHIR, H. I., CLARKSON, T. W, SMITH, J. C.
and DOHERTY, R. A. (1973). Methylmercury poisoning in Iraq. Science
181(96): 230-241.

179

BAKIR, F., RUSTAM, H, TIKRITI, S, AL-DAMLUJI, S. F. and SHIHRISTANI, H.
(1980). Clinical and epidemiological aspects of methylmercury poisoning.
Postgrad Med. J. 56(651): 1-10.

BALDI, F. (1997). Microbial transformation of mercury species and their importance in
the biogeochemical cycle of mercury. Met. Ions Biol. syst. 34: 213-257.

BEAVER, J. P. and WARIN G, P. (1994). Lack of correlation between early intracellular
calcium ion rises and the onset of apoptosis in thymocytes. Immunol. Cell Biol.
72(6): 489-499.

BERLIN, M. and ULLBERG, S. (1963). Accumulation and retention of mercury in the
mouse. III: An autoradiographic comparison of methyl mercuric dicyandiamide
with inorganic mercury. Arch Environ. Health 6: 610-616.

BERNARDI, P. (1996). The permeability transition pore. Control points of a cyclosporin
A- sensitive mitochondrial channel involved in cell death. Biochim. Biophys. Acta
1275(1-2): 5-9.

BERNARDI, P, COLONNA, R., COSTANTINI, P, ERIKSSON, 0., FONTAINE, E.,
ICHAS, F ., MASSARI, S, NICOLLI, A, PETRONILLI, V. and SCORRANO,
L. (1998). The mitochondrial permeability transition. Biofactors 8(3-4): 273-281.

BERNARDI, P. and PETRONILLI, V. (1996). The permeability transition pore as a
mitochondrial calcium release channel: a critical appraisal. J. Bioenerg.
Biomembr. 28(2): 13 1 -l 38.

BERNARDI, P, VASSANELLI, S, VERONESE, P, COLONNA, R., SZABO, I. and
ZORATTI, M. (1992). Modulation of the mitochondrial permeability transition
pore. Effect of protons and divalent cations. J. Biol. Chem. 267(5): 2934-2939.

BERNARDI, P, VERONESE, P. and PETRONILLI, V. (1993). Modulation of the
mitochondrial cyclosporin A-sensitive permeability transition pore. 1. Evidence
for two separate Me2+ binding sites with opposing effects on the pore open
probability. J. Biol. Chem. 268(2): 1005—1010.

BERSTEIN, G, HAGA, K, HAGA, T. and ICHIYAMA, A. (1988). Agonist and
antagonist binding of muscarinic acetylcholine receptors purified from porcine
brain: interconversion of high- and low-affinity sites by sulfhydryl reagents. J.
Neurochem. 50(6): 1687-1694.

BEZPROZVANN Y, I, WATRAS, J. and EHRLICH, B. E. (1991). Bell-shaped calcium-

response curves of Ins(l,4,5)P3- and calcium-gated channels from endoplasmic
reticulum of cerebellum. Nature 351(6329): 751-754.

180

BLACKSHAW, S, SAWA, A, SHARP, A. H, ROSS, C. A, SNYDER, S. H. and
KHAN, A. A. (2000). Type 3 inositol 1,4,5-trisphosphate receptor modulates cell
death. FASEBJ. 14(10): 1375-1379.

BONDY, S. C. and MCKBE, M. (1991). Disruption of the potential across the
synaptosomal plasma membrane and mitochondria by neurotoxic agents. Toxicol.
Lett. 58(1): 13-21.

BROEKEMEIER, K. M, DEMPSEY, M. E. and PFEIFFER, D. R. (1989). Cyclosporin
A is a potent inhibitor of the inner membrane permeability transition in liver
mitochondria. J. Biol. Chem. 264(14): 7826-7830.

 

BROSTROM, C. O. and BROSTROM, M. A. (1990). Calcium-dependent regulation of
protein synthesis in intact mammalian cells. Annu. Rev. Physiol. 52: 577-590.

BUDD, S. L. and NICHOLLS, D. G. (1996a). Mitochondria, calcium regulation, and
acute glutamate excitotoxicity in cultured cerebellar granule cells. J. Neurochem.
67(6): 2282-2291.

 

BUDD, S. L. and NICHOLLS, D. G. (1996b). A reevaluation of the role of mitochondria
in neuronal Ca2+ homeostasis. J. Neurochem. 66(1): 403-411.

BUNTINAS, L, GUNTER, K. K, SPARAGNA, G. C. and GUNTER, T. E. (2001). The
rapid mode of calcium uptake into heart mitochondria (RaM): comparison to RaM
in liver mitochondria. Biochim. Biophys. Acta 1504(2-3): 248-261.

BYRNE, A. M, LEMASTERS, J. J. and NIEMINEN, A. L. (1999). Contribution of
increased mitochondrial free Ca2+ to the mitochondrial permeability transition
induced by tert-butylhydroperoxide in rat hepatocytes. Hepatology 29(5): 1523-
1531.

CANDURA, S. M, D'AGOSTINO, G, CASTOLDI, A. F., MESSORI, E., LIUZZI, M,
MANZO, L. and TONINI, M. (1997). Effects of mercuric chloride and methyl

mercury on cholinergic neuromuscular transmission in the guinea-pig ileum.
Pharmacol. T oxicol. 80(5): 218-224.

 

CARRIER, G, BRUNET, R. C, CAZA, M. and BOUCHARD, M. (2001). A
toxicokinetic model for predicting the tissue distribution and elimination of
organic and inorganic mercury following exposure to methyl mercury in animals
and humans. 1. Development and validation of the model using experimental data
in rats. T oxicol. Appl. Pharmacol. 171(1): 38-49.

CASTOLDI, A. F., BARNI, S, TURIN, I, GANDINI, C. and MANZO, L. (2000). Early
acute necrosis, delayed apoptosis and cytoskeletal breakdown in cultured

cerebellar granule neurons exposed to methylmercury. J. Neurosci. Res. 59(6): ;
775-7 87 . ‘

181

CASTOLDI, A. F., CANDURA, S. M, COSTA, P, MANZO, L. and COSTA, L. G.
(1996). Interaction of mercury compounds with muscarinic receptor subtypes in
the rat brain. NeuroToxicology 17(3-4): 735-741.

CASTOLDI, A. F., COCCINI, T, CECCATELLI, S. and MANZO, L. (2001).
Neurotoxicity and molecular effects of methylmercury. Brain Res. Bull. 55(2):
197-203.

CELIO, M. R. (1990). Calbindin D-28k and parvalbumin in the rat nervous system.
Neuroscience 35(2): 375—475.

CHAN, S. L. and MATTSON, M. P. (1999). Caspase and calpain substrates: roles in
synaptic plasticity and cell death. J. Neurosci. Res. 58(1): 167-190.

CHANG, L. W. (1980). Mercury. In: Experimental and Clinical Neurotoxicology
(Spencer, P. S. and Schaumburg, H. H, eds.). The Williams and Wilkins Co.,
Baltimore, pp. 508-526.

CHANG, L. W, GILBERT, M. and SPRECHER, J. (1978). Modification of
methylmercury neurotoxicity by vitamin E. Environ. Res. 17(3): 356-366.

CHANG, L. W. and HARTMANN, H. A. (1972). Blood-brain barrier dysfunction in
experimental mercury intoxication. Acta Neuropathol. 21(3): 179-184.

CHANG, L. W. and HARTMANN, H. A. (1972). Electron microscopic histochemical
study on the localization and distribution of mercury in the nervous system after
mercury intoxication. Exp. Neurol. 35(1): 122-137.

CHANG, L. W, REUHL, K. R. and SPYKER, J. M. (1977). Ultrastructural study of the
latent effects of methyl mercury on the nervous system after prenatal exposure.
Environ. Res. 13(2): 171-185.

CHAVEZ, E. and HOLGUIN, J. A. (1988). Mitochondrial calcium release as induced by
Hg2+. J Biol Chem 263(8): 3582-3587.

CHAVEZ, E., ZAZUETA, C, DIAZ, E. and HOLQUIN, J. A. 1989). Characterization
by Hg2+ of two different pathways for mitochondrial Ca + release. Biochim.
Biophys. Acta 986(1): 27-32.

CHAVIS, P, NOONEY, J. M, BOCKAERT, J ., FAGNI, L, FELTZ, A. and BOSSU, J.
L. (1995). Facilitatory coupling between a glutamate metabotropic receptor and

dihydropyridine-sensitive calcium channels in cultured cerebellar granule cells. J.
Neurosci. 15(1 Pt 1): 135-143.

182

CHERNYAK, B. V. and BERNARDI, P. (1996). The mitochondrial permeability
transition pore is modulated by oxidative agents through both pyridine nucleotides
and glutathione at two separate sites. Eur. J. Biochem. 238(3): 623-630.

CHERNYAK, B. V, DEDOV, V. N. and CHERNYAK, V. (1995). Ca(2+)-tr'iggered
membrane permeability transition in deenergized mitochondria fiom rat liver.
FEBS Lett. 365(1): 75-78.

CHE'ITY, C. S, RAJANNA, S, HALL, E., YALLAPRAGADA, P. R. and RAJANNA,
B. (1996). In vitro and in vivo effects of lead, methyl mercury and mercury on
inositol 1,4,5-trisphosphate and l,3,4,5-tetrakisphosphate receptor bindings in rat
brain. Toxicol. Lett. 87(1): 11-17.

CHIU, V. C, MOURING, D. and HAYNES, D. H. (1983). Action of mercurials on the
active and passive transport properties of sarcoplasmic reticulum. J. Bioenerg.
Biomembr. 15(1): 13-25.

CHOI, B. H. (1989). The effects of methylmercury on the developing brain. Prog.
Neurobiol. 32(6): 447-470.

CHOI, B. H, LAPHAM, L. W, AMIN-ZAKI, L. and SALEEM, T. (1978). Abnormal
neuronal migration, deranged cerebral cortical organization, and diffuse white
matter astrocytosis of human fetal brain: a major effect of methylmercury
poisoning in utero. J. Neuropathol. Exp. Neurol. 37(6): 719-733.

CHOI, S. C. and BARTHA, R. (1994). Environmental factors affecting mercury
methylation in estuarine sediments. Bull. Environ. Contam. T oxicol. 53(6): 805-
812.

CLARKSON, T. W. (1972). The biological properties and distribution of mercury.
Biochem. J. 130(2): 61P-63P.

CLARKSON, T. W, AMIN-ZAKI, L. and AL-TIKRITI, S. K. (1976). An outbreak of
methylmercury poisoning due to consumption of contaminated grain. Fed. Proc.
35(12): 2395-2399.

COCCINI, T, RANDINE, G, CANDURA, S. M, NAPPI, R. E., PROCKOP, L. D. and
MANZO, L. (2000). Low-level exposure to methylmercury modifies muscarinic
cholinergic receptor binding characteristics in rat brain and lymphocytes:
physiologic implications and new opportunities in biologic monitoring. Environ.
Health Perspect. 108(1): 29-33.

COHEN, J. J. and DUKE, R. C. (1984). Glucocorticoid activation of a calcium-

dependent endonuclease in thymocyte nuclei leads to cell death. J Immunol.
132(1): 38-42.

183

 

CONTRERA, J. G, MCLESKEY, S. W, HOLOPAINEN, I, XU, J. and WOJCIK, W. J.
(1993). Muscarinic m2 receptors in cerebellar granule cell cultures from rat:
mechanism of short-term desensitization. J. Pharmacol. Exp. Ther. 265(1): 433-
440.

COOKSON, M. R and SHAW, P. J. (1999). Oxidative stress and motor neurone disease.
Brain Pathol. 9(1): 165-186.

CORONADO, R., MORRISSETTE, J ., SUKHAREVA, M. and VAUGHAN, D. M.
(1994). Structure and function of ryanodine receptors. Am. J. Physiol. 266(6 Pt 1):
C1485-1504.

CRUMP, K. S, VAN LANDINGHAM, C, SHAMLAYE, C, COX, C, DAVIDSON, P.
W, MYERS, G. J. and CLARKSON, T. W. (2000). Benchmark concentrations
for methylmercury obtained from the Seychelles Child Development Study.
Environ. Health Perspect. 108(3): 257-263.

CSORDAS, G, THOMAS, A. P. and HAJNOCZKY, G. (1999). Quasi-synaptic calcium
signal transmission between endoplasmic reticulum and mitochondria. EMBO J.
18(1): 96-108.

DAVIDSON, P. W, PALUMBO, D, MYERS, G. J ., COX, C, SHAMLAYE, C. F.,
SLOANE-REEVES, J ., CERNICHIARI, E., WILDING, G. E. and CLARKSON,
T. W. (2000). Neurodevelopmental outcomes of Seychellois children from the
pilot cohort at 108 months following prenatal exposure to methylmercury fi'om a
maternal fish diet. Environ. Res. 84(1): 1-11.

DE LUCA, A, WELLER, M, FREI, K. and FONTANA, A. (1996). Maturation-
dependent modulation of apoptosis in cultured cerebellar granule neurons by
cytokines and neurotrophins. Eur. J. Neurosci. 8(9): 1994-2005.

DELLINGER, J. A, MEYERS, R. M, GEBHARDT, K. J. and HANSEN, L. K. (1996).
The Ojibwa Health Study: fish residue comparisons for Lakes Superior,
Michigan, and Huron. Toxicol. Ind Health 12(3-4): 393-402.

DENNY, M. F. and ATCHISON, W. D. (1994). Methylmercury-induced elevations in
intrasynaptosomal zinc concentrations: an 19F-NMR study. J. Neurochem. 63(1):
383-386.

DENNY, M. F., HARE, M. F. and ATCHISON, W. D. (1993). Methylmercury alters

intrasynaptosomal concentrations of endogenous polyvalent cations. T oxicol.
Appl. Pharmacol. 122(2): 222-232.

184

DOBLE, A, HUBERT, J. P. and BLANCHARD, J. C. (1992). Pertussis toxin
pretreatment abolishes the inhibitory effect of riluzole and carbachol on D-
[3H]aspartate release from cultured cerebellar granule cells. Neurosci. Lett.
140(2): 251-254.

DRUMMOND, R. M. and TUFT, R. A. (1999). Release of Ca2+ fi'om the sarcoplasmic

reticulum increases mitochondrial [Ca2+] in rat pulmonary artery smooth muscle

cells. J. Physiol. (Land) 516(Pt 1): 139-147.

DUBINSKY, J. M. and ROTHMAN, S. M. (1991). Intracellular calcium concentrations
during "chemical hypoxia" and excitotoxic neuronal injury. J. Neurosci. 11(8):
2545-2551.

DUCHEN, M. R. (2000). Mitochondria and calcium: from cell signalling to cell death. J.
Physiol. 529 Pt 1: 57-68.

ELDEFRAWI, M. E., MANSOUR, N. A. and ELDEFRAWI, A. T. (1977). Interactions
of acetylcholine receptors with organic mercury compounds. Adv. Exp. Med Biol.
84: 449-463.

EXTON, J. H. (1990). Signaling through phosphatidylcholine breakdown. J. Biol. Chem.
265(1): 14.

FARKAS, D. L, WEI, M. D, FEBBRORIELLO, P, CARSON, J. H. and LOEW, L. M.
(1989). Simultaneous imaging of cell and mitochondrial membrane potentials.
Biophys. J. 56(6): 1053-1069.

FOHRMAN, E. B, DE ERAUSQUIN, G, COSTA, E. and WOJCIK, W. J. (1993).
Muscarinic m3 receptors and dynamics of intracellular Ca2+ cerebellar granule
neurons. Eur. J. Pharmacol. 245(3): 263-271.

FRIBERG, H, CONNERN, C, HALESTRAP, A. P. and WIELOCH, T. (1999).
Differences in the activation of the mitochondrial permeability transition among

brain regions in the rat correlate with selective vulnerability. J. Neurochem. 72(6):
2488-2497.

F RIBERG, L. and MOTTET, N. K. (1989). Accumulation of methylmercury and
inorganic mercury in the brain. Biol. Trace Elem. Res. 21: 201-206.

FUKAMAUCHI, F., AIHARA, O. and KUSAKABE, M. (2000). Intemalization and
down-regulation of muscarinic acetylcholine receptors in cerebellar granule cells
of tenascin-gene deficient mice. Neurochem. Int. 36(2): 153-158.

FUKAMAUCHI, F., HOUGH, C. and CHUANG, D. M. (1991). Expression and agonist-

induced down-regulation of mRNAs of m2- and m3- muscarinic acetylcholine
receptors in cultured cerebellar granule cells. J. Neurochem. 56(2): 716-719.

185

 

FUKAMAUCHI, F., SAUNDERS, P. A, HOUGH, C. and CHUANG, D. M. (1993).
Agonist-induced down-regulation and antagonist-induced up-regulation of m2-
and m3-muscarinic acetylcholine receptor mRN A and protein in cultured
cerebellar granule cells. Mol. Pharmacol. 44(5): 940-949.

GALLI, C, MEUCCI, O, SCORZIELLO, A, WERGB, T. M, CALISSANO, P. and
SCHETTINI, G. (1995). Apoptosis in cerebellar granule cells is blocked by high
KCl, forskolin, and IGF-1 through distinct mechanisms of action: the involvement
of intracellular calcium and RNA synthesis. J. Neurosci. 15(2): 1172-1179.

GARMAN, R. H, WEISS, B. and EVANS, H. L. (1975). Alkylrnercurial encephalopathy
in the monkey (Saimiri sciureus and Macaca arctoides): a histopathologic and
autoradiographic study. Acta Neuropathol. 32(1): 61-74.

GASSO, S, SUNOL, C, SANFELIU, C, RODRIGUEZ-PARKE, E. and CRISTOFOL,
R M. (2000). Pharmacological characterization of the effects of methylmercury

and mercuric chloride on spontaneous noradrenaline release from rat hippocampal
slices. Life Sci. 67(10): 1219-1231.

 

GAVIN, C. E., GUNTER, K. K. and GUNTER, T. E. (1999). Manganese and calcium
transport in mitochondria: implications for manganese toxicity. NeuroToxicology
20(2-3): 445-453.

GERSTNER, H. B. and HUFF, J. E. (1977). Selected case histories and epidemiologic
examples of human mercury poisoning. Clin. Toxicol. 11(2): 131-150.

GONZALEZ, B, LEROUX, P, LAMACZ, M, BODENANT, C, BALAZS, R. and
VAUDRY, H. (1992). Somatostatin receptors are expressed by immature

cerebellar granule cells: evidence for a direct inhibitory effect of somatostatin on
neuroblast activity. Proc. Natl. Acad. Sci. U. S. A. 89(20): 9627-9631.

GRANDJEAN, P, BUDTZ-JORGENSEN, E., WHITE, IL F., JORGENSEN, P. J.,
WEIHE, P, DEBES, F. and KEIDING, N. (1999). Methylmercury exposure

biomarkers as indicators of neurotoxicity in children aged 7 years. Am. J.
Epidemiol. 150(3): 301-305.

GRANDJEAN, P, WHITE, R. F., NIELSEN, A, CLEARY, D. and DE OLIVEIRA
SANTOS, E. C. (1999). Methylmercury neurotoxicity in Amazonian children
downstream from gold mining. Environ. Health Perspect. 107(7): 587-591.

GUERINI, D, GARCIA-MARTIN, E., ZECCA, A, GUIDI, F. and CARAFOLI, E.
(1998). The calcium pump of the plasma membrane: membrane targeting, calcium

binding sites, tissue-specific isoform expression. Acta Physiol. Scand. Suppl. 643:
265-273.

186

GUIMARAES, J. R., MEILI, M, HYLANDER, L. D, DE CASTRO E SILVA, E.,
ROULET, M, MAURO, J. B. and DE LEMOS, R. (2000). Mercury net

methylation in five tropical flood plain regions of Brazil. Sci. Total Environ.
261(1-3): 99-107.

GUNTER, K. K. and GUNTER, T. E. (1994). Transport of calcium by mitochondria. J.
Bioenerg. Biomembr. 26(5): 471-485.

GUNTER, T. E. and PFEIFFER, D. R. (1990). Mechanisms by which mitochondria
transport calcium. Am. J. Physiol. 258(5 Pt 1): C755-786.

HAJNOCZKY, G, HAGER, R. and THOMAS, A. P. (1999). Mitochondria suppress
local feedback activation of inositol 1,4, 5- trisphosphate receptors by Ca”. J.
Biol. Chem. 274(20): 14157-14162.

HANSEN, J. C. (1990). Human exposure to heavy metals through consumption of marine
foods: a case study of exceptionally high intake among Greenlanders. In: Heavy
Metals in the Marine Environment (Furnow, A. W. and Rainbow, P. S, eds).
CRC Press, New York, pp. 227-243.

HARADA, M. (1995). Minarnata disease: methylmercury poisoning in Japan caused by
environmental pollution Crit. Rev. Toxicol. 25(1): 1-24.

HARE, M. F. and ATCHISON, W. D. (1992). Comparative action of methylmercury and
divalent inorganic mercury on nerve terminal and intraterrninal mitochondrial
membrane potentials. J. Pharmacol. Exp. Ther. 261(1): 166-172.

HARE, M. F. and ATCHISON, w. D. (1995). Methylmercury mobilizes Ca” from
intracellular stores sensitive to inositol 1,4,5-trisphosphate in NGlO8-15 cells. J.
Pharmacol. Exp. Ther. 272(3): 1016-1023.

HARE, M. F., MCGINNIS, K. M. and ATCHISON, W. D. (1993). Methylmercury
increases intracellular concentrations of CaH and heavy metals in NGlO8-15
cells. J. Pharmacol. Exp. Ther. 266(3): 1626-1635.

HE, L, POBLENZ, A. T, MEDRANO, C. J. and FOX, D. A. (2000). Lead and calcium
produce rod photoreceptor cell apoptosis by opening the mitochondrial
permeability transition pore. J. Biol. Chem. 275(16): 12175-12184.

HEDLUND, B. and BARTFAI, T. (1979). The importance of thiol- and disulfide groups
in agonist and antagonist binding to the muscarinic receptor. Mol. Pharmacol.
15(3): 531-544.

HILLERED, L, MUCHIRI, P. M, NORDENBRAND, K. and ERNSTER, L. (1983).

Mn2+ prevents the Ca2+-induced inhibition of ATP synthesis in brain
mitochondria. F EBS Lett. 154(2): 247-250.

187

HUNT, J. B, NEECE, S. H, SCHACHMAN, H. K. and GINSBURG, A. (1984).
Mercurial-promoted Zn2+ release from Escherichia coli aspartate
transcarbamoylase. J. Biol. Chem. 259(23): 14793—14803.

HUNTER, D. and RUSSELL, D. S. (1954). Focal cerebral and cerebellar atrophy in a
human subject due to organic mercury compounds. Neurol. Neurosurg. Psychiat.
17: 169-235.

ICHAS, F. and MAZAT, J. P. (1998). From calcium signaling to cell death: two
conformations for the mitochondrial permeability transition pore. Switching from
low- to high- conductance state. Biochim. Biophys. Acta 1366(1-2): 33-50.

INSUG, O, DATAR, S, KOCH, C. J., SHAPIRO, I. M. and SHENKER, B. J. (1997).
Mercuric compounds inhibit human monocyte function by inducing apoptosis:
evidence for formation of reactive oxygen species, development of mitochondrial

membrane permeability transition and loss of reductive reserve. Toxicology
124(3): 211-224.

IRVING, A. J., COLLINGRIDGE, G. L. and SCHOFIELD, J. o. (1992). Interactions
between Ca2+ mobilizing mechanisms in cultured rat cerebellar granule cells. J.
Physiol. (Lond) 456: 667-680.

IRVING, A. J ., COLLINGRIDGE, G. L. and SCHOFIELD, J. G. (1992). L-glutarnate
and acetylcholine mobilise Ca2+ fi'om the same intracellular pool in cerebellar

granule cells using transduction mechanisms with different Ca2+ sensitivities. Cell
Calcium 13(5): 293-301.

JANAKY, R., VARGA, V, SARANSAARI, P. and OJA, S. S. (1993). Glutathione
modulates the N-methyl-D-aspartate receptor-activated calcium influx into
cultured rat cerebellar granule cells. Neurosci. Lett. 156(1-2): 153-157.

JAYARAMAN, T. and MARKS, A. R. (1997). T cells deficient in inositol 1,4,5-
trisphosphate receptor are resistant to apoptosis. Mol. Cell Biol. 17(6): 3005-3012.

JU AN G, M. S. (1976). Depression of frog muscle contraction by methylmercuric
chloride and mercuric chloride. T oxicol. Appl. Pharmacol. 35(1): 183-185.

JUAN G, M. S. (1976). An electrophysiological study of the action of methylmercuric
chloride and mercru'ic chloride on the sciatic nerve-sartorius muscle preparation
of the frog. Toxicol. Appl. Pharmacol. 37(2): 339-348.

KANE, D. J ., ORD, T, ANTON, R. and BREDESEN, D. E. (1995). Expression of bcl-2
inhibits necrotic neural cell death. J. Neurosci. Res. 40(2): 269-275.

KASS, G. E. and ORRENIUS, S. (1999). Calcium signaling and cytotoxicity. Environ.
Health Perspect. 107 Suppl 1: 25-35.

188

KAUPPINEN, R. A, KOMULAINEN, H. and TAIPALE, H. (1989). Cellular
mechanisms underlying the increase in cytosolic free calcium concentration

induced by methylmercury in cerebrocortical synaptosomes from guinea pig. J.
Pharmacol. Exp. Ther. 248(3): 1248-1254.

KISELYOV, K, XU, X, MOZHAYEVA, G, KUO, T, PESSAH, I, MIGNERY, G,
ZHU, X, BIRNBAUMER, L. and MUALLEM, S. (1998). Functional interaction
between InsP3 receptors and store-operated Htrp3 channels. Nature 396(6710):
478-482.

KISELYOV, K. I, SHIN, D. M, WANG, Y, PESSAH, I. N, ALLEN, P. D. and
MUALLEM, S. (2000). Gating of store-operated channels by conformational
coupling to ryanodine receptors. Mol. Cell 6(2): 421-431.

KOCH, B. D, FAUROT, G. F., KOPANITSA, M. V. and SWINNEY, D. C. (1994).
Pharmacology of a Ca(2+)-influx pathway activated by emptying the intracellular
Ca2+ stores in HL-60 cells: evidence that a cytochrome P- 450 is not involved.
Biochem. J. 302(Pt 1): 187-190.

KOMULAINEN, H. and BONDY, S. C. (1987). Increased fiee intrasynaptosomal Ca2+
by neurotoxic organometals: distinctive mechanisms. Toxicol. Appl. Pharmacol.
88(1): 77-86.

KOWALTOWSKI, A. J. and CASTILHO, R. F. (1997). Ca2+ acting at the external side
of the inner mitochondrial membrane can stimulate mitochondrial permeability
transition induced by phenylarsine oxide. Biochim. Biophys. Acta 1322(2-3): 221-
229.

KROEMER, G. (1997). The proto-oncogene Bel-2 and its role in regulating apoptosis.
Nat. Med 3(6): 614-620.

KRUMAN, I. I. and MATTSON, M. P. (1999). Pivotal role of mitochondrial calcium
uptake in neural cell apoptosis and necrosis. J. Neurochem. 72(2): 529-540.

KUNIMOTO, M. (1994). Methylmercury induces apoptosis of rat cerebellar neurons in
primary culture. Biochem. Biophys. Res. Commun. 204(1): 310-317.

KURLAND, L. T. (1973). Methyl-mercury sources, mode of action and clinical and

pathological effects on the developing nervous system. Res. Publ. Assoc. Res.
Nerv. Ment. Dis. 51(10): 283-297.

189

KUZNETSOV, D. A, ZAVIJALOV, N. V, GOVORKOV, A. V. and SIBILEVA, T. M.
(1987). Methyl mercury-induced nonselective blocking of phosphorylation

processes as a possible cause of protein synthesis inhibition in vitro and in vivo.
Toxicol. Lett. 36(2): 1 53-160.

LANDOLFI, B, CURCI, S, DEBELLIS, L, POZZAN, T. and HOFER, A. M. (1998).
Ca2+ homeostasis in the agonist-sensitive internal store: functional interactions
between mitochondria and the ER measured In situ in intact cells. J. Cell Biol.
142(5): 1235-1243.

LEBEL, C. P, ALI, S. F. and BONDY, S. C. (1992). Deferoxarnine inhibits methyl
mercury-induced increases in reactive oxygen species formation in rat brain.
T oxicol. Appl. Pharmacol. 112(1): 161-165.

LEBEL, C. P, ALI, S. F., MCKEE, M. and BONDY, S. C. (1990). Organometal-induced
increases in oxygen reactive species: the potential of 2',7'-dichlorofluorescin
diacetate as an index of neurotoxic damage. Toxicol. Appl. Pharmacol. 104(1):
17-24.

LEMASTERS, J. J ., NIEMINEN, A. L, QIAN, T, TROST, L. C, ELMORE, S. P,
NISHIMURA, Y, CROWE, R. A, CASCIO, W. E., BRADHAM, C. A,
BRENNER, D. A. and HERMAN, B. (1998). The mitochondrial permeability
transition in cell death: a common mechanism in necrosis, apoptosis and
autophagy. Biochim. Biophys. Acta 1366(1-2): 177-196.

LEVESQUE, P. C. and ATCHISON, W. D. (1988). Effect of alteration of nerve terminal
Ca2+ regulation on increased spontaneous quantal release of acetylcholine by
methyl mercury. Toxicol. Appl. Pharmacol. 94(1): 55-65.

LEVESQUE, P. C. and ATCHISON, W. D. (1991). Disruption of brain mitochondrial
calcium sequestration by methylmercury. J. Pharmacol. Exp. Ther. 256(1): 236-
242.

LEVESQUE, P. C, HARE, M. F. and ATCHISON, W. D. (1992). Inhibition of
mitochondrial Ca2+ release diminishes the effectiveness of methyl mercury to

release acetylcholine fiom synaptosomes. Toxicol. Appl. Pharmacol. 115(1): 11-
20.

LEVICK, V, COFFEY, H. and D'MELLO, S. R. (1995). Opposing effects of
thapsigargin on the survival of developing cerebellar granule neurons in culture.
Brain Res. 676(2): 325-335.

LEYSHON, K. and MORGAN, A. J. (1991). An integrated study of the morphological

and gross-elemental consequences of methyl mercury intoxication in rats, with
particular attention on the cerebellum. Scanning Microsc. 5(3): 895-904.

190

 

LEYSHON-SORLAND, K, JASANI, B. and MORGAN, A. J. (1994). The localization
of mercury and metallothionein in the cerebellum of rats experimentally exposed
to methylmercury. Histochem. J. 26(2): 161-169.

LEYSSENS, A, NOWICKY, A. V, PATTERSON, L, CROMPTON, M. and
DUCHEN, M. R. (1996). The relationship between mitochondrial state, ATP
hydrolysis, [Mgzji and [Ca2+], studied in isolated rat cardiomyocytes. J. Physiol.
496 (Pt 1): 111-128.

LIMKE, T. L. and ATCHISON, W. D. (2002). Acute exposure to methylmercury opens
the mitochondrial permeability transition pore in rat cerebellar granule cells.
Toxicol. Appl. Pharmacol. (in press).

LIN, S. Z, YAN, G. M, KOCH, K. E., PAUL, S. M. and IRWIN, R. P. (1997).
Mastoparan-induced apoptosis of cultured cerebellar granule neurons is initiated
by calcium release from intracellular stores. Brain Res. 771(2): 184-195.

LOEW, L. M, CARRINGTON, W, TUFT, R. A. and FAY, F. S. (1994). Physiological
cytosolic Ca” transients evoke concurrent mitochondrial depolarizations. Proc.
Natl. Acad. Sci. U. S. A. 91(26): 12579-12583.

MA, H. T, PATTERSON, R. L, VAN ROSSUM, D. B, BIRNBAUMER, L,
MIKOSHIBA, K. and GILL, D. L. (2000). Requirement of the inositol

trisphosphate receptor for activation of store-operated Ca2+ channels. Science
287(5458): 1647-1651.

MANEV, H, KHARLAMOV, E., UZ, T, MASON, R. P. and CAGNOLI, C. M. (1997).
Characterization of zinc-induced neuronal death in primary cultures of rat
cerebellar granule cells. Exp. Neurol. 146(1): 171-178.

MARCAIDA, G, KOSENKO, E., MINANA, M. D, GRISOLIA, s. and FELIPO, v.
(1996). Glutamate induces a calcineurin-mediated dephosphorylation of Na+,K(+’-

ATPase that results in its activation in cerebellar neurons in culture. J.
Neurochem. 66(1): 99-104.

MARKS, A. R. (1997). Intracellular calcium-release channels: regulators of cell life and
death. Am. J. Physiol. 272(2 Pt 2): H597-605.

MARSH, D. O, CLARKSON, T. W, COX, C, MYERS, G. J ., AMIN-ZAKI, L. and
AL-TIKRITI, S. (1987). Fetal methylmercury poisoning. Relationship between
concentration in single strands of maternal hair and child effects. Arch. Neurol.
44(10): 1017-1022.

MARTY, M. S. and ATCHISON, W. D. (1997). Pathways mediating Ca2+ entry in rat

cerebellar granule cells following in vitro exposure to methyl mercury. T oxicoI.
Appl. Pharmacol. 147(2): 319-330.

191

 

MARTY, M. S. and ATCHISON, W. D. (1998). Elevations of intracellular Ca2+ as a
probable contributor to decreased viability in cerebellar granule cells following
acute exposure to methylmercury. Toxicol. Appl. Pharmacol. 150(1): 98-105.

MASGRAU, R., SERVITJA, J. M, SARRI, E., YOUNG, K. W, NAHORSKI, S. R. and
PICATOSTE, F. (2000). Intracellular Ca2+ stores regulate muscarinic receptor

stimulation of phospholipase C in cerebellar granule cells. J. Neurochem. 74(2):
818-826.

MATTSON, M. P. (2000). Apoptosis in neurodegenerative disorders. Nat. Rev. Mol.
Cell. Biol. 1(2): 120-129.

MATTSON, M. P, LAFERLA, F. M, CHAN, S. L, LEISSRING, M. A, SHEPEL, P.
N. and GEIGER, J. D. (2000). Calcium signaling in the ER: its role in neuronal
plasticity and neurodegenerative disorders. Trends Neurosci. 23(5): 222-229.

MEIER, E., DREJER, J. and SCHOUSBOE, A. (1984). GABA induces functionally
active low-affinity GABA receptors on cultured cerebellar granule cells. J.
Neurochem. 43(6): 1737-1744.

MHYRE, T. R., MAINE, D. N. and HOLLIDAY, J. (2000). Calcium-induced calcium
release from intracellular stores is developmentally regulated in primary cultures
of cerebellar granule neurons. J. Neurobiol. 42(1): 134-147.

MIURA, K, CLARKSON, T. W, IKEDA, K, NAGANUMA, A. and IMURA, N.
(1994). Establishment and characterization of methylmercury-resistant PC12 cell
line. Environ. Health Perspect. 102 Suppl 3: 313-315.

MIURA, K. and IMURA, N. (1987). Mechanism of methylmercury cytotoxicity. Crit.
Rev. Toxicol. 18(3): 161-188.

MIYATA, H, SILVERMAN, H. s, SOLLOTI‘, s. J., LAKATTA, E. G, STERN, M. D.
and HANSFORD, R. G. (1991). Measurement of mitochondrial free Ca2+

concentration in living single rat cardiac myocytes. Am J. Physiol. 261(4 Pt 2):
H1123-1134.

MOGENSEN, H. S. and JORGENSEN, O. S. (2000). AMPA receptor subunit mRNAs
and intracellular [Ca(2+)] in cultured mouse and rat cerebellar granule cells. Int. J.
Dev. Neurosci. 18(1): 61-68.

MOLLER-MADSEN, B. (1990). Localization of mercury in CNS of the rat. II.

Intraperitoneal injection of methyhnercuric chloride (CH3HgCl) and mercuric
chloride (I-IgClz). Toxicol. Appl. Pharmacol. 103(2): 303-323.

192

MOLLER-MADSEN, B. (1991). Localization of mercury in CNS of the rat. III. Oral
administration of methylmercuric chloride (CH3HgCl). F undam. Appl. T oxicol.
16(1): 172-187.

MORAN, J , ITOH, T, REDDY, U. R., CHEN, M, ALNEMRI, E. S. and PLEASURE,
D. (1999). Caspase-3 expression by cerebellar granule neurons is regulated by
calcium and cyclic AMP. J. Neurochem. 73(2): 568-577.

MORGANS, C. W, EL FAR, O, BERNTSON, A, WASSLE, H. and TAYLOR, W. R.
(1998). Calcium extrusion from mammalian photoreceptor terminals. J. Neurosci.
18(7): 2467-2474.

MUNDY, W. R. and FREUDENRICH, T. M. (2000). Sensitivity of immature neurons in
culture to metal-induced changes in reactive oxygen species and intracellular free
calcium. NeuroToxicology 21(6): 1 135-1 144.

MUTO, H, SHINADA, M, TOKUTA, K. and TAKIZAWA, Y. (1991). Rapid changes
in concentrations of essential elements in organs of rats exposed to

methylmercury chloride and mercuric chloride as shown by simultaneous
multielemental analysis. Br. J. Ind. Med. 48(6): 382-388.

MYERS, G. J. and DAVIDSON, P. W. (1998). Prenatal methylmercury exposure and
children: neurologic, developmental, and behavioral research. Environ Health
Perspect. 106 Suppl 3: 841-847.

MYERS, G. J ., DAVIDSON, P. W, SHAMLAYE, C. F., AXTELL, C. D,
CERNICHIARI, E., CHOISY, O, CHOI, A, COX, C. and CLARKSON, T. W.
(1997). Effects of prenatal methylmercury exposure from a high fish diet on
developmental milestones in the Seychelles Child Development Study.
NeuroToxicology 18(3): 819-829.

MYERS, G. J., MARSH, D. O, COX, C, DAVIDSON, P. W, SHAMLAYE, C. F.,
TANNER, M. A, CHOI, A, CERNICHIARI, E., CHOISY, O. and CLARKSON,
T. W. (1995). A pilot neurodevelopmental study of Seychellois children following

in utero exposure to methylmercury from a maternal fish diet. Neuro Toxicology
16(4): 629-638.

NAGASHIMA, K, FUJII, Y, TSUKAMOTO, T, NUKUZUMA, S, SATOH, M,
FUJITA, M, FUJIOKA, Y. and AKAGI, H. (1996). Apoptotic process of

cerebellar degeneration in experimental methylmercury intoxication of rats. Acta
Neuropathol. 91(1): 72-77.

NAKADA, S. and IMURA, N. (1983). Susceptibility of lipids to mercurials. J. Appl.
T oxicol. 3(3): 131-134.

193

NAKAMURA, K, BOSSY-WETZEL, E., BURNS, K, FADEL, M. P, LOZYK, M,
GOPING, I. S, OPAS, M, BLEACKLEY, R. C, GREEN, D. R. and
MICHALAK, M. (2000). Changes in endoplasmic reticulum luminal environment
affect cell sensitivity to apoptosis. J. Cell Biol. 150(4): 731-740.

NAKATSURU, S, OOHASHI, J ., NOZAKI, H, NAKADA, S. and IMURA, N. (1985).
Effect of mercurials on lymphocyte functions in vitro. Toxicology 36(4): 297-305.

NEATHERY, M. W. and MILLER, W. J. (1975). Metabolism and toxicity of cadmium,
mercury, and lead in animals: a review. J. Dairy Sci. 58(12): 1767-1781.

NEUSTADT, A, FROSTHOLM, A. and ROTTER, A. (1988). Topographical
distribution of muscarinic cholinergic receptors in the cerebellar cortex of the

mouse, rat, guinea pig, and rabbit: a species comparison. .1. Comp. Neurol.
272(3): 317-330.

NICHOLLS, D. G. (1974).The influence of respiration and ATP hydrolysis on the
protonelectrochemical gradient across the inner membrane of rat liver
mitochondria as determined by ion distribution. Eur. J. Biochem. 50: 305-315.

NICHOLLS, D. G. (1978). The regulation of extramitochondrial free calcium ion
concentration by rat liver mitochondria. Biochem. J. 176(2): 463 -474.

NICHOLLS, D. G. and BUDD, S. L. (1998). Mitochondria and neuronal glutamate
excitotoxicity. Biochim. Biophys. Acta 1366(1-2): 97-112.

NICHOLLS, D. G, BUDD, S. L, CASTILHO, R. F. and WARD, M. W. (1999).
Glutamate excitotoxicity and neuronal energy metabolism. Ann. N. Y. Acad Sci.
893: 1-12.

NICHOLLS, D. G. and SCO'I'I‘, I. D. (1980). The regulation of brain mitochondrial
calcium-ion transport. The role of ATP in the discrimination between kinetic and

membrane-potential- dependent calcium-ion efflux mechanisms. Biochem. J.
186(3): 833-839.

NIEMINEN, A. L, PETRIE, T. G, LEMASTERS, J. J. and SELMAN, W. R. (1996).
Cyclosporin A delays mitochondrial depolarization induced by N-methyl-D-
aspartate in cortical neurons: evidence of the mitochondrial permeability
transition. Neuroscience 75(4): 993-997.

NIEMINEN, A. L, SAYLOR, A. K, TESFAI, S. A, HERMAN, B. and LEMASTERS,
J. J. (1995). Contribution of the mitochondrial permeability transition to lethal

injury after exposure of hepatocytes to t-butylhydroperoxide. Biochem. J. 307(Pt
1 ): 99-1 06.

194

OBERDORF, J ., VALLANO, M. L. and WOJCIKIEWICZ, R J. (1997). Expression and
regulation of types I and H inositol 1,4,5- trisphosphate receptors in rat cerebellar
granule cell preparations. J. Neurochem. 69(5): 1897-1903.

O'KUSKY, J. (1983). Methylmercury poisoning of the developing nervous system:
morphological changes in neuronal mitochondria. Acta Neuropathol. 61(2): 116-
122.

ORRENIUS, S. and NICOTERA, P. (1994). The calcium ion and cell death. J. Neural
T ransm. Suppl. 43: 1-11.

OU, Y. C, WHITE, C. C, KREJSA, C. M, PONCE, R. A, KAVANAGH, T. J. and
FAUSTMAN, E. M. (1999). The role of intracellular glutathione in

methyhnercury-induced toxicity in embryonic neuronal cells. Neuro Toxicology
20(5): 793-804.

OYAKE, Y, TANAKA, M, KUBO, H. and CHICHIBU, M. (1966). [Neuropathological
studies on organic mercury poisoning with special reference to the staining and
distribution of mercury granules]. Shinlcei Kenkyu No Shimpo 10(4): 744-750.

OYAMA, Y, CARPENTER, D. O, UENO, S, HAYASHI, H. and TOMIYOSHI, F.
(1995). Methylmercury induces Ca(2+)'dependent hyperpolarization of mouse

thymocytes: a flow cytometric study using fluorescent dyes. Eur. J. Pharmacol.
293(2): 101-107.

OYAMA, Y, TOMIYOSHI, F ., UENO, S, FURUKAWA, K. and CHIKAHISA, L.
(1994). Methylmercury-induced augmentation of oxidative metabolism in
cerebellar neurons dissociated from the rats: its dependence on intracellular Ca2+.
Brain Res. 660(1): 154-157.

PALUMBO, D. R., COX, C, DAVIDSON, P. W, MYERS, G. J ., CHOI, A,
SHAMLAYE, C, SLOANE-REEVES, J ., CERNICHIARI, E. and CLARKSON,
T. W. (2000). Association between prenatal exposure to methylmercury and
cognitive functioning in Seychellois children: a reanalysis of the McCarthy Scales
of Children's Ability fiom the main cohort study. Environ. Res. 84(2): 81-88.

PAN, Z. Z. and WILLIAMS, J. T. (1994) Muscarine hyperpolarizes a subpopulation of
neurons by activating an M2 muscarinic receptor in rat nucleus raphe magnus in
vitro. J. Neurosci. 14: 1332-1338.

PARK, S. T, LIM, K. T, CHUNG, Y. T. and KIM, S. U. (1996). Methylmercury-

induced neurotoxicity in cerebral neuron culture is blocked by antioxidants and
NMDA receptor antagonists. NeuroToxicology 17(1): 37-45.

195

PASTORINO, J. G, SNYDER, J. W, HOEK, J. B. and FARBER, J. L. (1995). Ca2+
depletion prevents anoxic death of hepatocytes by inhibiting mitochondrial
permeability transition. Am. J. Physiol. 268(3 Pt 1): C676-685.

PASTORINO, J. G, TAFANI, M, ROTHMAN, R. J ., MARCINKEVICIUTE, A,
HOEK, J. B, FARBER, J. L. and MARCINEVICIUTE, A. (1999). Functional
consequences of the sustained or transient activation by Bax of the mitochondrial
permeability transition pore. J. Biol. Chem. 274(44): 31734-31739.

PEITSCH, M. C, POLZAR, B, STEPHAN, H, CROMPTON, T, MACDONALD, H.
R., MANNHERZ, H. G. and TSCHOPP, J. (1993). Characterization of the
endogenous deoxyribonuclease involved in nuclear DNA degradation during
apoptosis (programmed cell death). EMBO J. 12(1): 371-377.

PENG, T. 1., JOU, M. J., SHEU, S. S. and GREENAMYRE, J. T. (1998). Visualization
of NMDA receptor-induced mitochondrial calcium accumulation in striatal
neurons. Exp. Neurol. 149(1): 1-12.

PETRONILLI, V, COLA, C. and BERNARDI, P. (1993). Modulation of the
mitochondrial cyclosporin A-sensitive permeability transition pore. II. The
minimal requirements for pore induction underscore a key role for transmembrane
electrical potential, matrix pH, and matrix Ca2+. J. Biol. Chem. 268(2): 1011-

101 6.

PETRONILLI, V, COSTANTINI, P, SCORRANO, L, COLONNA, R.,
PASSAMONTI, S. and BERNARDI, P. (1994). The voltage sensor of the
mitochondrial permeability transition pore is tuned by the oxidation-reduction
state of vicinal thiols. Increase of the gating potential by oxidants and its reversal
by reducing agents. J. Biol. Chem. 269(24): 16638-16642.

PETRONILLI, V, PENZO, D, SCORRANO, L, BERNARDI, P. and DI LISA, F.
(2001). The mitochondrial permeability transition, release of cytochrome c and

cell death. Correlation with the duration of pore openings in situ. J. Biol. Chem.
276(15): 12030-12034.

PINTON, P, BRINI, M, BASTIANU'ITO, C, TUFT, R. A, POZZAN, T. and
RIZZUTO, R. (1998). New light on mitochondrial calcium. Biofactors 8(3-4):
243-253.

PIZZI, M, RIBOLA, M, VALERIO, A, MEMO, M. and SPANO, P. (1991). Various
Ca2+ entry blockers prevent glutamate-induced neurotoxicity. Eur. J. Pharmacol.
209(3): 169-173.

PUTNEY, J. W, JR. (1986). A model for receptor-regulated calcium entry. Cell Calcium
7(1): 1-12.

196

QUANDT, F. N, KATO, E. and NARAHASHI, T. (1982). Effects of methylmercury on
electrical responses of neuroblastoma cells. NeuroToxicology 3(4): 205-220.

RABENSTEIN, D. L. and EVANS, C. A. (1978). The mobility of methylmercury in
biological systems. Bioinorg. Chem. 8(2): 107-101.

RAFFRAY, M. and COHEN, G. M. (1997). Apoptosis and necrosis in toxicology: a
continuum or distinct modes of cell death? Pharmacol. Ther. 75(3): 153-177.

RANDALL, A. and TSIEN, R. W. (1995). Pharmacological dissection of multiple types
of Ca2+ channel currents in rat cerebellar granule neurons. J. Neurosci. 15(4):
2995-3012.

RANDALL, A. D. and TSIEN, R. W. (1997). Contrasting biophysical and
pharmacological properties of T-type and R- type calcium channels.
Neuropharmacology 36(7): 879-893.

REDMAN, R. S. and SILINSKY, E. M. (1994). Decrease in calcium currents induced by
aminoglycoside antibiotics in frog motor nerve endings. Br. J. Pharmacol.
113(2): 375-378.

RENZONI, A, ZINO, F. and FRANCHI, E. (1998). Mercury levels along the food chain
and risk for exposed populations. Environ. Res. 77(2): 68-72.

RESINK, A, HACK, N, BOER, G. J. and BALAZS, R. (1994). Growth conditions
differentially modulate the vulnerability of developing cerebellar granule cells to
excitatory amino acids. Brain Res. 655(1-2): 222-232.

RICE, D. C. (1989). Brain and tissue levels of mercury after chronic methylmercury
exposure in the monkey. J. T oxicol. Environ. Health 27(2): 189-198.

RICE, G, SWARTOUT, J ., MAHAFFEY, K. and SCHOENY, R. (2000). Derivation of
US. EPA's oral Reference Dose (RfD) for methylmercury. Drug Chem. T oxicol.
23(1): 41-54.

RICHTER, C. and KASS, G. E. (1991). Oxidative stress in mitochondria: its relationship
to cellular Ca2+ homeostasis, cell death, proliferation, and differentiation. Chem.
Biol. Interact. 77(1): 1-23.

RIZZUTO, R., PINTON, P, CARRINGTON, W, FAY, F. S, FOGARTY, K. E.,
LIFSHITZ, L. M, I UP I, R. A. and POZZAN, T. (1998). Close contacts with the

endoplasmic reticulum as determinants of mitochondrial Ca2+ responses. Science
280(5370): 1763-1766.

ROBERTS, L. R, STEINRAUF, L. K. and BLICKENSTAF F, R. T. (1980). New
evaluation of potential methylmercury scavengers. J. Pharm. Sci. 69(8): 964-967.

197

ROBERTSON, J. D. and ORRENIUS, S. (2000). Molecular mechanisms of apoptosis
induced by cytotoxic chemicals. Crit. Rev. T oxicol. 30(5): 609-627.

RODIER, P. M. (1995). Developing brain as a target of toxicity. Environ. Health
Perspect. 103 Suppl 6: 73-76.

RODIER, P. M, ASCHNER, M. and SAGER, P. R. (1984). Mitotic arrest in the
developing CNS after prenatal exposure to methylmercury. Neurobehav. T oxicol.
T eratol. 6(5): 379-3 85.

RUSTAM, H, VON BURG, R., AMIN-ZAKI, L. and EL HASSANI, S. (1975).
Evidence for a neuromuscular disorder in methylmercury poisoning. Arch.
Environ. Health 30(4): 190-195.

SAKAMOTO, M, IKEGAMI, N. and NAKANO, A. (1996). Protective effects of Ca2+
channel blockers against methyl mercury toxicity. Pharmacol. Toxicol. 78(3):
193-199.

SANFELIU, C, SEBASTIA, J. and K1, S. U. (2001). Methylmercury neurotoxicity in
cultures of human neurons, astrocytes, neuroblastoma cells. NeuroToxicology
22(3): 317-327.

SARAFIAN, T, HAGLER, J ., VARTAVARIAN, L. and VERITY, M. A. (1989). Rapid
cell death induced by methyl mercury in suspension of cerebellar granule neurons.
J. Neuropathol. Exp. Neurol. 48(1): 1-10.

SARAFIAN, T. and VERITY, M. A. (1985). Inhibition of RNA and protein synthesis in
isolated cerebellar cells by in vitro and in vivo methyl mercury. Neurochem.
Pathol. 3(1): 27-39.

SARAF IAN, T. and VERITY, M. A. (1991). Oxidative mechanisms underlying methyl
mercury neurotoxicity. Int. J. Dev. Neurosci. 9(2): 147-153.

SARAF IAN , T. A. (1993). Methyl mercury increases intracellular Ca2+ and inositol
phosphate levels in cultured cerebellar granule neurons. J. Neurochem. 61(2):
648-657.

SARAF IAN, T. A. (1999). Methylmercury-induced generation of free radicals: biological
implications. Met. Ions Biol. Syst. 36: 415-444.

SARAFIAN, T. A, CHEUNG, M. K. and VERITY, M. A. (1984). In vitro methyl

mercury inhibition of protein synthesis in neonatal cerebellar perikarya.
Neuropathol. Appl. Neurobiol. 10(2): 85-100.

198

SARAFIAN, T. A, VARTAVARIAN, L, KANE, D. J., BREDESEN, D. E. and
VERITY, M. A. (1994). bcl-2 expression decreases methyl mercury-induced free-
radical generation and cell killing in a neural cell line. Toxicol. Lett. 74(2): 149-
155.

SATO, T, SATO, M. and SAWADA, M. (1976). Effects of disulfide bond reduction on
the excitatory and inhibitory postsynaptic responses of Aplysia ganglion cells.
Jpn. J. Physiol. 26(5): 471-485.

SCHINDER, A. F., OLSON, E. C, SPITZER, N. C. and MONTAL, M. (1996).
Mitochondrial dysfunction is a primary event in glutamate neurotoxicity. J.
Neurosci. 16(19): 6125-6133.

SCORRANO, L, PETRONILLI, V. and BERNARDI, P. (1997). On the voltage
dependence of the mitochondrial permeability transition pore. A critical appraisal.
J. Biol. Chem. 272(19): 12295-12299.

SENSI, S. L, YIN, H. Z. and WEISS, J. H. (2000). AMPA/kainate receptor-triggered
Zn” entry into cortical neurons induces mitochondrial Zn2+ uptake and persistent
mitochondrial dysfunction. Eur. J Neurosci. 12(10): 3813-3818.

SHAMOO, A. E., MACLENNAN, D. H. and ELDERFRAWI, M. E. (1976). Differential
effects of mercurial compounds on excitable tissues. Chem. Biol. Interact. 12(1):
41-52.

SHARP, A. H, NUCIFORA, F. C, JR, BLONDEL, O, SHEPPARD, C. A, ZHANG,
C, SNYDER, S. H, RUSSELL, J. T, RYUGO, D. K. and ROSS, C. A. (1999).
Differential cellular expression of isoforms of inositol 1,4,5- triphosphate
receptors in neurons and glia in brain. J. Comp. Neurol. 406(2): 207-220.

SHENKER, B. J., BERTHOLD, P, DECKER, S, MAYRO, J., ROONEY, C, VITALE,
L. and SHAPIRO, I. M. (1992). Immunotoxic effects of mercuric compounds on

human lymphocytes and monocytes. II. Alterations in cell viability.
Immunopharmacol. Immunotoxicol. 14(3): 555-577.

SHENKER, B. J ., BERTHOLD, P, ROONEY, C, VITALE, L, DEBOLT, K. and
SHAPIRO, I. M. (1993). Immunotoxic effects of mercuric compounds on human
lymphocytes and monocytes. III. Alterations in B-cell function and viability.
Immunopharmacol. Immunotoxicol. 15(1): 87-112.

SHENKER, B. J ., GUO, T. L, O, I. and SHAPIRO, I. M. (1999). Induction of apoptosis
in human T-cells by methyl mercury: temporal relationship between

mitochondrial dysfunction and loss of reductive reserve. Toxicol. Appl.
Pharmacol. 157(1): 23-35.

199

SHENKER, B. J ., GUO, T. L. and SHAPIRO, I. M. (1998). Low-level methylmercury
exposure causes human T-cells to undergo apoptosis: evidence of mitochondrial
dysfunction. Environ. Res. 77(2): 149-159.

SHENKER, B. J ., GUO, T. L. and SHAPIRO, I. M. (2000). Mercury-induced apoptosis
in human lymphoid cells: evidence that the apoptotic pathway is mercurial species
dependent. Environ. Res. 84(2): 89-99.

SIMBULA, G, GLASCOTT, P. A, JR, AKIT A, S, HOEK, J. B. and FARBER, J. L.
(1997). Two mechanisms by which ATP depletion potentiates induction of the
mitochondrial permeability transition. Am. J. Physiol. 273(2 Pt 1): C479-488.

SIMPSON, P. B, CHALLISS, R A. and NAHORSKI, S. R. (1993). Involvement of
intracellular stores in the Ca2+ responses to N-Methyl-D- aspartate and
depolarization in cerebellar granule cells. J. Neurochem. 61(2): 760-763.

SIMPSON, P. B, CHALLISS, R. A. and NAHORSKI, S. R. (1994). Chronic activation
of muscarinic and metabotropic glutamate receptors down-regulates type I

inositol 1,4,5-trisphosphate receptor expression in cerebellar granule cells. J.
Neurochem. 63(6): 2369-2372.

SIMPSON, P. B, CHALLISS, R A. and NAHORSKI, S. R. (1995). Divalent cation
entry in cultured rat cerebellar granule cells measured using Mn2+ quench of fura
2 fluorescence. Eur. J Neurosci. 7(5): 831-840.

SIMPSON, P. B, CHALLISS, R. A. and NAHORSKI, S. R. (1995). Neuronal Ca2+
stores: activation and function. Trends Neurosci. 18(7): 299-306.

SIMPSON, P. B, NAHORSKI, S. R. and CHALLISS, R. A. (1996). Agonist-evoked
Ca2+ mobilization from stores expressing inositol 1,4,5- trisphosphate receptors
and ryanodine receptors in cerebellar granule neurones. J. Neurochem. 67(1):
364-373.

SIMPSON, P. B. and RUSSELL, J. T. (1996). Mitochondria support inositol 1,4,5-
trisphosphate-mediated Ca2+ waves in cultured oligodendrocytes. J. Biol. Chem.
271(52): 33493-33501.

SIMPSON, P. B. and RUSSELL, J. T. (1998a). Mitochondrial Ca2+ uptake and release
influence metabotropic and ionotropic cytosolic Ca2+ responses in rat
oligodendrocyte progenitors. J. Physiol. (Land) 508(Pt 2): 413-426.

SIMPSON, P. B. and RUSSELL, J. T. (1998b). Role of mitochondrial Ca2+ regulation in
neuronal and glial cell signalling. Brain Res. Brain Res. Rev. 26(1): 72-81.

200

 

 

SIROIS, J. E. and ATCHISON, W. D. (2000). Methylmercury affects multiple subtypes
of calcium channels in rat cerebellar granule cells. T oxicol. Appl. Pharmacol.
167(1): 1-11.

SOMJEN, G. G, HERMAN, S. P, KLEIN, R, BRUBAKER, P. E., BRINER, W. H,
GOODRICH, J. K, KRIGMAN, M. R. and HASEMAN, J. K. (1973). The uptake

of methyl mercury (203 Hg) in different tissues related to its neurotoxic effects. J.
Pharmacol. Exp. Ther. 187(3): 602-611.

SOMLYO, A. P, BOND, M. and SOMLYO, A. V. (1985). Calcium content of
mitochondria and endoplasmic reticulum in liver frozen rapidly in viva. Nature
314(6012): 622-625.

SONE, N, LARSS'I'U VOLD, M. K. and KAGAWA, Y. (1977). Effect of methyl
mercury on phosphorylation, transport, and oxidation in mammalian
mitochondria. J. Biochem. (Tokyo) 82(3): 859-868.

SPARAGNA, G. C, GUNTER, K. K, SHEU, S. S. and GUNTER, T. E. (1995).
Mitochondrial calcium uptake fi'om physiological-type pulses of calcium. A
description of the rapid uptake mode. J. Biol. Chem. 270(46): 27510-27515.

STEINWALL, O. and KLATZO, I. (1966). Selective vulnerability of the blood-brain
barrier in chemically induced lesions. J. Neuropathol. Exp. Neurol. 25(4): 542-
559.

STEUERWALD, U, WEIHE, P, JORGENSEN, P. J , BJERVE, K, BROCK, J .,
HEINZOW, B, BUDTZ-JORGEN SEN, E. and GRANDJEAN, P. (2000).

Maternal seafood diet, methylmercury exposure, and neonatal neurologic
function. J. Pediatr. 136(5): 599-605.

STRINGFELLOW, T. L, MARTY, M. S, FRITZ, E. L. AND ATCHISON, W. D.
(2000). K+ depolarization delays methylmercury-induced elevations of cerebellar
granule cell intracellular calcium independently of changes in plasma membrane
and mitochondrial membrane potentials. Soc. for Neurosci. (Abstracts) 26: 598.

SYVERSEN, T. L, TOTLAND, G. and FLOOD, P. R. (1981). Early morphological
changes in rat cerebellum caused by a single dose of methyhnercury. Arch.
Toxicol. 47(2): 101-111.

SZABO, I. and ZORATTI, M. (1992). The mitochondrial megachannel is the
permeability transition pore. J. Bioenerg. Biomembr. 24(1): 111-117.

SZALAI, G, KRISHNAMURTHY, R. and HAJNOCZKY, G. (1999). Apoptosis driven
by IP(3)-linked mitochondrial calcium signals. EMBO J. 18(22): 6349-6361.

201

TAKEUCHI, T. (1982). Pathology of Minamata disease. With special reference to its
pathogenesis. Acta Pathol. Jpn. 32(Suppl. 1): 73-99.

TAKEUCHI, T, MORIKAWA, N, MATSUMOTO, H. and SHIRAISHI, Y. (1962). A
pathological study of Minimata disease in Japan. Acta Neuropathol. 2: 40-57.

TAN, X. X, TANG, C, CASTOLDI, A. F., MANZO, L. and COSTA, L. G. (1993).
Effects of inorganic and organic mercury on intracellular calcium levels in rat T
lymphocytes. J. T oxicol. Environ. Health 38(2): 159-170.

THASTRUP, O, CULLEN, P. J ., DROBAK, B. K, HANLEY, M. R. and DAWSON, A.
P. (1990). Thapsigargin, a tumor promoter, discharges intracellular Ca2+ stores by
specific inhibition of the endoplasmic reticulum Ca w-ATPase. Proc. Natl. Acad
Sci. U. S. A. 87(7): 2466-2470.

THAYER, S. A. and MILLER, R. J. (1990). Regulation of the intracellular free calcium
concentration in single rat dorsal root ganglion neurones in vitro. J. Physiol.
(Land) 425: 85-115.

TROST, L. C. and LEMASTERS, J. J. (1997). Role of the mitochondrial permeability
transition in salicylate toxicity to cultured rat hepatocytes: implications for the
pathogenesis of Reye's syndrome. T oxicol. Appl. Pharmacol. 147(2): 431-441.

URUSHITANI, M, NAKAMIZO, T, INOUE, R., SAWADA, H, KIHARA, T,
HONDA, K, AKAIKE, A. and SHIMOHAMA, S. (2001). N-methyl-D-aspartate
receptor-mediated mitochondrial Cam) overload in acute excitotoxic motor
neuron death: a mechanism distinct from chronic neurotoxicity after Ca0+) influx.
J. Neurosci. Res. 63(5): 377-387.

USACHEV, Y. M. and THAYER, s. A. (1999). Ca2+ influx in resting rat sensory
neurones that regulates and is regulated by ryanodine-sensitive Ca2+ stores. J.
Physiol. (Land) 519 Pt 1: 115-130.

USUKI, F., YASUTAKE, A, UMEHARA, F., TOKUNAGA, H, MATSUMOTO, M,
ETO, K, ISHIURA, S. and HIGUCHI, I. (2001). In vivo protection of a water-

soluble derivative of vitamin E, Trolox, against methylmercury-intoxication in the
rat. Neurosci. Lett. 304(3): 199-203.

VALLEE, B. L. and AULD, D. S. (1990). Active-site zinc ligands and activated H20 of
zinc enzymes. Proc. Natl. Acad Sci. U. S. A. 87(1): 220-224.

VERITY, M. A. (1999). Manganese neurotoxicity: a mechanistic hypothesis.
NeuroToxicalogy 20(2-3): 489-497.

202

VERITY, M. A, BROWN, W. J. and CHEUNG, M. (1975). Organic mercurial
encephalopathy: in vivo and in vitro effects of methyl mercury on synaptosomal
respiration. J. Neurochem. 25(6): 759-766.

VERITY, M. A, BROWN, W. J., CHEUNG, M. and CZER, G. (1977). Methyl mercury
inhibition of synaptosome and brain slice protein synthesis: in vivo and in vitro
studies. J. Neurochem. 29(4): 673-679.

VERITY, M. A, SARAF IAN , T, PACIFICI, E. H. and SEVANIAN, A. (1994).
Phospholipase A2 stimulation by methyl mercury in neuron culture. J.
Neurochem. 62(2): 705-714.

VON BURG, R., NORTHINGTON, F. K. and SHAMOO, A. (1980). Methylmercury
inhibition of rat brain muscarinic receptors. T oxicol. Appl. Pharmacol. 53(2):
285-292.

WANG, G. J. and THAYER, S. A. (1996). Sequestration of glutamate-induced Ca2+ loads
by mitochondria in cultured rat hippocampal neurons. J. Neurophysiol. 76(3):
1 61 l -1 62 1 .

WARING, P. and SJAARDA, A. (1995). Extracellular calcium is not required for
gliotoxin or dexamethasone- induced DNA fragmentation: a reappraisal of the use
of EGTA. Int. J. Immunopharmacol. 17(5): 403-410.

WEI, H, LEEDS, P, CHEN, R. W, WEI, W, LENG, Y, BREDESEN, D. E. and
CHUAN G, D. M. (2000). Neuronal apoptosis induced by pharmacological
concentrations of 3- hydroxykynurenine: characterization and protection by
dantrolene and Bel-2 overexpression. J. Neurochem. 75(1): 81-90.

WEI, H. and PERRY, D. C. (1996). Dantrolene is cytoprotective in two models of
neuronal cell death. J. Neurochem. 67(6): 2390-2398.

WEI, H, WEI, W, BREDESEN, D. E. and PERRY, D. C. (1998). Bel-2 protects against
apoptosis in neuronal cell line caused by thapsigargin-induced depletion of
intracellular calcium stores. J. Neurochem. 70(6): 2305-2314.

WEIHE, P, GRANDJEAN, P, DEBES, F. and WHITE, R. (1996). Health implications
for Faroe islanders of heavy metals and PCBs from pilot whales. Sci. Total
Environ. 186(1-2): 141-148.

WERTH, J. L, USACHEV, Y. M. and THAYER, S. A. (1996). Modulation of calcium

efflux fi'om cultured rat dorsal root ganglion neurons. J. Neurosci. 16(3): 1008-
1015.

203

WHITE, R. J. and REYNOLDS, I. J. (1996). Mitochondrial depolarization in glutamate-
stirnulated neurons: an early signal specific to excitotoxin exposure. J. Neurosci.
16(18): 5688-5697.

WHITHAM, E. M, CHALLISS, R. A. and NAHORSKI, S. R (1991). M3 muscarinic
cholinoceptors are linked to phosphoinositide metabolism in rat cerebellar granule
cells. Eur. J. Pharmacol. 206(3): 181-189.

WOJCIKIEWICZ, R. J ., FURUICHI, T, NAKADE, S, MIKOSHIBA, K. and
NAHORSKI, S. R. (1994). Muscarinic receptor activation down-regulates the
type I inositol 1,4,5- trisphosphate receptor by accelerating its degradation. J.
Biol. Chem. 269(11): 7963-7969.

WUDARCZYK, J ., DEBSKA, G. and LENARTOWICZ, E. (1999). Zinc as an inducer of
the membrane permeability transition in rat liver mitochondria. Arch Biochem.
Biophys. 363(1): 1-8.

XIA, Y, RAGAN, R. E., SEAH, E. E., MICHAELIS, M. L. and MICHAELIS, E. K.
(1995). Developmental expression of N-methyl-D-aspartate (NMDA)-induced
neurotoxicity, NMDA receptor function, and the NMDARI and glutarnate-
binding protein subtmits in cerebellar granule cells in primary cultures.
Neurochem. Res. 20(5): 617-629.

XU, J. and CHUANG, D. M. (1987). Muscarinic acetylcholine receptor-mediated
phosphoinositide turnover in cultured cerebellar granule cells: desensitization by
receptor agonists. J. Pharmacol. Exp. Ther. 242(1): 238-244.

YAN, G. M, LIN, S. Z, IRWIN, R. P. and PAUL, S. M. (1995). Activation of G proteins
bidirectionally affects apoptosis of cultured cerebellar granule neurons. J
Neurochem. 65(6): 2425-2431.

YAN, G. M, LIN, S. Z, IRWIN, R. P. and PAUL, S. M. (1995). Activation of
muscarinic cholinergic receptors blocks apoptosis of cultured cerebellar granule
neurons. Mol. Pharmacol. 47(2): 248-257.

YEE, S. and CHOI, B. H. (1994). Methylmercury poisoning induces oxidative stress in
the mouse brain. Exp. Mol. Pathol. 60(3): 188-196.

YEE, S. and CHOI, B. H. (1996). Oxidative stress in neurotoxic effects of
methylmercury poisoning. NeuroToxicology 17(1): 17-26.

YOSHINO, Y, MOZAI, T. and NAKAO, K. (1966). Distribution of mercury in the brain

and its subcellular units in experimental organic mercury poisoning. J.
Neurochem. 13: 397-406.

204

 

YUAN, Y. and ATCHISON, W. D. (1993). Disruption by methylmercury of membrane
excitability and synaptic transmission of CA1 neurons in hippocampal slices of
the rat. Toxicol. Appl. Pharmacol. 120(2): 203-215.

YUAN, Y. and ATCHISON, W. D. (1994). Comparative effects of inorganic divalent
mercury, methylmercury and phenylrnercury on membrane excitability and
synaptic transmission of CA1 neurons in hippocampal slices of the rat.
NeuroToxicology 15(2): 403-41 1.

YUAN, Y. and ATCHISON, W. D. (1995). Methylmercury acts at multiple sites to block
hippocampal synaptic transmission. J. Pharmacol. Exp. Ther. 275(3): 1308-1316.

YUAN, Y. and ATCHISON, W. D. (1997). Action of methylmercury on GABA(A)
receptor-mediated inhibitory synaptic transmission is primarily responsible for its
early stimulatory effects on hippocampal CA1 excitatory synaptic transmission. J.
Pharmacol. Exp. Ther. 282(1): 64-73.

ZENG, F. Y, SOLDNER, A, SCHONEBERG, T. and WESS, J. (1999). Conserved
extracellular cysteine pair in the M3 muscarinic acetylcholine receptor is essential
for proper receptor cell surface localization but not for G protein coupling. J.
Neurochem. 72(6): 2404-2414.

ZHAN G, Z. and WU, D. (1994). [Effect of methylmercury chloride on brain muscarinic
receptor in rat]. Hua Xi Yi Ke Da Xue Xue Bao 25(4): 388-392.

205

 

 

 

   

lullliltiflll‘rfll’llfillll