AN ABSTRACT OF THE THESIS OF
Xiaoqin S. Zhao for the degree of Master of Science in Pharmacy presented on
December 20. 1994
Title: Purine Nucleotide-induced Seizure in Rat Prepiriform Cortex
Redacted for Privacy
Abstract Approved:
Dr. Thoma
.
Murray
ATP has recently been shown to function as a fast neurotransmitter in guinea
pig coeliac neurons and in the rat medial habenula. ATP and related P2 purinoceptor
agonists generally evoke excitation of peripheral and central neurons. In an attempt
to further assess the functional roles of neuronal ATP receptors, we have
pharmacologically characterized a P2 purinoceptor which mediates a convulsant
response evoked from the rat prepiriform cortex (PPC). All nucleotides and test
compounds were focally administered by unilateral microinjection in the rat PPC.
Test compounds were dissolved in ACSF or equimolar CaC12 solution and injected in
a volume of 120-500 nl. The rank order of potency as convulsants following
unilateral focal injection of nucleotides was 13,1`-methylene-L-ATP > a,B-methylene-
ATP > > ATP. The most potent analog 13,11-methylene-L-ATP had an ED50 value of
4.3 nmol. The convulsant dose-response curve of 13,r-methylene-L-ATP was shifted
to the right by prior focal injection of the putative P2 purinoceptor antagonist
suramin. Unlike the other nucleotide analogs tested, 2-methylthio-ATP and ADPBS
did not elicit behavioral seizures when administered alone. 2-Methylthio-ATP was
also distinct from the other nucleotides tested in that it induced an anticonvulsant
response when focally administered prior to bicuculline methiodide. This
anticonvulsant response is likely the result of degradation of 2-methylthio-ATP to 2­
methylthioadenosine. 2-Methylthioadenosine was shown to be an anticonvulsant
against bicuculline methiodide in the PPC with an ED50 value of 4.1 nmol; a value
similar to that of 2-methylthio-ATP (5.3 nmol). The anticonvulsant effect of 2­
methylthio-ATP is therefore likely to be due to its breakdown to 2-methylthio­
adenosine which is an agonist at adenosine Al receptors. To further establish the role
of adenosine Al receptors in the anticonvulsant actions of 2-methylthioadenosine, the
affinity of 2-methylthio-adenosine for the Al receptor was determined. Equilibrium
competition analysis was performed using [3H]DPCPX as ligand for the Al receptor.
The order of potency of an array of adenosine analogs including 2-methylthio­
adenosine as inhibitors of [3H]DPCPX binding to brain membrane was significantly
correlated with their respective potencies as anticonvulsants (r = 0.91). Thus, the
pharmacology of the P2 purinoceptor mediating the convulsant effect of nucleotides
displays an agonist profile suggestive of P2, receptor existence in cerebral cortical
tissue, whereas the anticonvulsant effect of 2-methylthio-ATP appears to be mediated
through the activation of an adenosine Al receptor.
Purine Nucleotide-induced Seizures in Rat Prepiriform Cortex
by
Xiaoqin S. Zhao
A THESIS
submitted to
Oregon State University
in partial fulfillment of
the requirements for the
degree of
Master of Science
Completed December 20, 1994
Commencement June 1995
Master of Science thesis of Xiaoqin S. Zhao presented on December 20, 1994
APPROVED:
Redacted for Privacy
Major Professor, representing Pharnjology
Redacted for Privacy
Chair of College of Pharmacy
Redacted for Privacy
Dean of Grad
e School
I understand that my thesis will become part of the permanent collection of Oregon
State University libraries. My signature below authorizes release of my thesis to any
reader upon request.
Redacted for Privacy
'Xiaoqin S. Zhao, Author
ACKNOWLEDGEMENTS
I wish to express the sincerest appreciation to my major professor, Dr.
Thomas F. Murray for his guidance, assistance, encouragement and financial support
in the course of this research investigation and the preparation of this manuscript.
I would also like to acknowledge Dr. Paul Franklin for his advice, help and
encouragement through my graduate studies.
Special thanks to our team members and my fellow graduate students for their
friendship, encouragement and help through my graduate study. I am proud I was
ever a member of this great team.
I am deeply grateful to my parents, Wen-ji Lin and Shi-jian Zhao, for their
love, encouragement and support to accomplish this study.
TABLE OF CONTENTS
Page
1. INTRODUCTION
1.1 Historical Review of Purine Physiology
1.1.1 Initial Observation of Purine Sensitivity
1.1.2 Purinergic Nerve Hypothesis
1.1.3 Formation, Storage, Release and Inactivation of ATP
1.1.4 Structure of Purinergic Nerves
1
1
1
2
4
7
1.2 Pharmacological Classification of Purinergic Receptors
8
1.3 Neuronal ATP Receptors and Their Transmembrane
Signalling Systems
15
1.4 Purinergic Receptor Radioligand Binding Studies and Receptor
Molecular Biology
20
1.5 Epileptogenic Site in Rat Prepiriform Cortex
2. MATERIALS AND METHODS
25
26
2.1 Animals and Stereotaxic Surgery
26
2.2 Drugs and Drug Administration
27
2.3 Experimental Protocol and Behavioral Assay
29
2.3.1
2.3.2
2.3.3
2.3.4
Verification of Cannula Placement
Evaluation of the Effects of Test Compounds
Coadministration
Desensitization
29
29
30
30
2.4 Membrane Preparation and Radioligand Binding Assay
31
2.5 Statistical Analysis
32
3. RESULTS
3.1 Nucleotide-evoked Convulsant Response in Rat
Prepiriform cortex
32
32
3.1.1 ACSF as vehicle
3.1.2 CaC12 as vehicle
32
34
3.2 Pharmacological Characterization of ATP Receptor
Mediating Convulsant Effects in Rat Prepiriform Cortex
36
3.3 Investigation of Anticonvulsant Effect of 2-MeSATP
in Rat Prepiriform Cortex
37
3.4 Correlation Between 2-MeSAdo as Anticonvulsant and
its Affinity for Adenosine Al Receptors.
38
3.5 8-pSPT on ATP Response in Presence of CaC12.
39
3.6 Desensitization
40
4. DISCUSSION
41
5. BIBLIOGRAPHY
65
LIST OF FIGURES
FIGURE
1.
2.
3.
4.
5.
6.
7.
8.
PAGE
Structures of ATP agonists and antagonists used in
this study
57
Seizures induced by focal injection of
B,r-methylene-L-ATP, a,B-methylene-ATP, ATP, AOPCP,
ADPBS, and UTP in rat prepiriform cortex (CaC12 vehicle)
58
Suramin antagonism of B,I'- methylene -L- ATP induced
seizures in rat prepiriform cortex
59
Anticonvulsant effect of 2-methylthio-ATP and
2-methylthio-adenosine against bicuculline methiodide­
induced seizures in rat prepiriform cortex
60
Agonist competition of [3H]DPCPX binding in a rat brain
cerebral cortical membrane preparation
61
Correlation between potencies of adenosine agonists as
anticonvulsants and their affinity for adenosine Al receptors
62
Influence of the P1 receptor antagonist 8-(p-sulfopheny1)­
theophylline (8-pSPT) on ATP responses in the presence of
CaC12
63
Comparison of a,B-methylene-ATP-induced seizure responses
on repeated application to rat prepiriform cortex
64
LIST OF TABLES
TABLE
1.
2.
3.
4.
5.
6.
7.
8.
9.
Page
Seizures induced by focal injection of suramin
B,r-methylene-L-ATP, a,B-methylene-ATP,
B,r-methylene-ATP, ADPBS, a,13-methylene-ADP,
UTP, and ATP in rat prepiriform cortex
(ACSF vehicle).
47
Sodium citrate and sodium pyrophosphate-induced
seizures in rat prepiriform cortex
49
Seizures induced by focal injection of
B,r-methylene-L-ATP, a,B-methylene-ATP, ATP,
ADPBS, AOPCP and UTP in rat prepiriform
cortex (CaC12 vehicle)
50
Potencies of nucleotides as convulsants in rat
prepiriform cortex
51
Anticonvulsant effect of 2-methylthio-ATP
and 2-methylthio-adenosine against seizures
induced by bicuculline methiodide in rat
prepiriform cortex
52
Anticonvulsant potency of 2-methylthio-ATP
and 2-methylthio-adenosine against
seizures induced by bicuculline methiodide
in rat prepiriform cortex
53
Effect of P1 receptor antagonist,
8- (p- sulfophenyl)- theophylline (8-pSPT) on ATP
response in rat prepiriform cortex
54
Potencies of adenosine analogs in protection
against BMI seizures in the rat PPC
55
The affinity of adenosine analogs as
inhibitors of [3H]DPCPX binding to rat
brain membranes.
56
PURINE NUCLEOTIDE-INDUCED SEIZURES
IN RAT PREPIRIFORM CORTEX
1. INTRODUCTION
1.1
Historical review of ATP and purine physiology.
1.1.1 Initial observations of purine sensitivity
Intracellular nucleotides play fundamental roles in energy metabolism, nucleic acid
synthesis, and enzyme regulation.
However, during the past sixth years purine
nucleotides and nucleosides have been shown to have potent extracellular actions on
excitable membranes which may be involved in physiological regulatory processes.
In a study of the pharmacology of simple extracts from heart muscle, brain, kidney and
spleen, Drury and Szent-Gyorgyi (1929) found that these crude preparations had potent
effects upon the mammalian heart (Drury and Gyorgyi, 1929).
These effects were
attributed to the adenosine monophosphate (AMP) content of the extracts. Adenosine,
prepared from yeast,
was found to have an identical action and, upon intravenous
administration in whole animals, it caused a slowing of heart rate, general lowering of
arterial pressure, dilation of coronary vessels and inhibition of intestinal movements.
During 1950's, the clinical effects of ATP administration in man were being widely
explored, especially in geriatric patients with cardiovascular disorders.
The potent vasodilatory actions of adenyl compounds led Holton and Holton
(1954) to suggest that ATP might be the vasodilatory substance which was released on
2
antidromic stimulation of sensory nerves (Holton and Holton, 1954). Since ATP was
considerably more potent than adenosine in producing vasodilatation of coronary vessels,
the possibility that ATP makes a more significant contribution to the physiological
regulation of coronary blood vessels than adenosine has been considered (Paddle and
Burnstock, 1974).
1.1.2 Purinergic Nerve Hypothesis
Classically the autonomic nervous system consists of two components, cholinergic
and adrenergic nerves. Hints of the existence of autonomic nerves other than those of
the two classical components can be found in the early literature.
In the early 1960's
transmural stimulation of the guinea-pig taenia coli with a single pulse of short duration
was shown to produce large hyperpolarisations or inhibitory junction potentials in smooth
muscle cells, which were associated with relaxation, and which persisted in the presence
of both atropine and guanethidine (Burnstock et al., 1963; Burnstock et al., 1964). It
was established that these responses were nerve-mediated since they were abolished by
low concentrations of tetrodotoxin. In addition, the possibility that the non-cholinergic,
non-adrenergic excitatory nerves shown to supply some organs (e.g. urinary bladder,
intestine) may be purinergic was considered.
The pelvic nerve provides excitatory
control of the urinary bladder and it was assumed for many years that this was mediated
by classical cholinergic nerves, even though most of the response was resistant to
blockade by atropine. "Atropine resistance" to the excitatory autonomic nerve supplying
the bladder appears to be a characteristic feature throughout vertebrates including
placental mammals (Mantegazza and Naimzada, 1967). The explanation for atropine
3
resistance is that the majority of excitatory nerves supplying the bladder are non­
cholinergic.
ATP closely mimics the response to non-cholinergic excitatory nerve
stimulation and has been proposed as the transmitter released by these nerves (Burnstock
et al., 1972a). The evidence for a third nerve component in the autonomic nervous
system, which was neither adrenergic nor cholinergic, was therefore advanced
(Burnstock, 1972b). Burnstock proposed that the principal active substance released from
at least some of these nerves was the purine nucleotide, adenosine 5'-triphosphate (ATP)
(Burnstock, 1972b). These non-adrenergic and non-cholinergic nerves were tentatively
termed "purinergic".
The evidence that ATP is the transmitter released from non­
adrenergic and non-cholinergic nerves in the gut includes the following:
a) synthesis and storage of ATP in nerve terminals such that it is available for
release upon stimulation of the nerves
b) the release of endogenous purine nucleotides, and of tritium-labeled compounds
during stimulation of purinergic nerves after exposure to [3H]- adenosine
c) the parallel effects of low concentrations of ATP and purinergic nerve
stimulation
d) the presence in purinergically-innervated tissues of enzymes that degrade ATP,
including ATPase, 5'-nucleotidase and adenosine deaminase
e) the demonstration that various drugs produce parallel blocking and potentiating
actions on the responses to purinergic nerve stimulation and directly applied
ATP.
4
1.1.3 Formation, storage, release and inactivation of ATP
Studies with radioactive tracers indicate that ATP is stored in terminal axons in
such a way that it can be released during nerve stimulation. Strips of taenia coli were
shown to take up large amounts of tritium-labeled adenosine, adenosine monophosphate
(AMP), adenosine diphosphate (ADP) and ATP. Adenosine was rapidly converted into
and retained largely as [3H] -ATP. More of the labeled ATP was located in nerves than
in muscle or other tissue components.
This was demonstrated by analyzing the
radioactivity in frozen sections cut serially at different levels through preparations of the
caecal wall and overlying taenia muscle layer (Burnstock, 1972b); peaks of activity were
located in Auerbach's plexus and in the outer, serosal region of the taenia where the neural
component consists largely of terminal varicose fibers (Bernett and Rogers, 1967).
The recently established evidence for ATP and other nucleotides as neural
messengers has called for a close examination of nucleotide contents of synaptic vesicles.
Co-storage of ACh and ATP in vesicles was demonstrated for a variety of peripheral and
central synaptic vesicles (Zimmermann, 1994). Similarly, noradrenaline and ATP are
co-stored in vesicles of sympathetic nerve terminals, and in granules from the related
chromaffin cells. In either case, vesicular nucleotides are out-numbered by their co­
transmitter. While there is some uncertainty concerning the exact number of transmitter
substances inside vesicles, it is clear that the nucleotides are stored in the millimolar range.
The molar ratios of ACh or catecholamine compared with ATP are 3/1 to 10/1 and 18/1
to 60/1 respectively in cholinergic or catecholaminergic systems.
Molar ratios of released ACh and ATP close to those of synaptic-vesicle storage
5
have been observed for synaptosomes from the electric organ of the electric ray, and pure
cholinergic synaptosomes from rat caudate nucleus (Richardsom, et al., 1987). There is
much evidence for a Ca'-dependent neurogenic release of ATP from nerve terminals.
This has been demonstrated directly for synaptosomes isolated from the electric organ
of the electric ray and mammalian brain, and for peripheral tissues including bladder, vas
deferens and the rat neuromuscular junction (Smith et al., 1991). Both the fast electrical
response to sympathetic-nerve stimulation and the twitch component of neurogenic
contraction are mediated by ATP. Examples of this are the rat tail artery and vas deferens
where the first, and rapid, phase of the neurogenic contractile response is mediated by
ATP via P2 purinoceptors, whereas noradrenaline mediates a second and sustained tonic
contraction via al-adrenoceptors (Bao et al. , 1993; Vizi et al. , 1992). In addition, release
of noradrenaline and ATP is inhibited via prejunctional purinoceptors. This could be a
result of the presynaptic action of ATP and its degradation product adenosine.
Postsynaptic ATP release can be several fold higher than presynaptic release and might
contribute to the muscle contraction (Vizi et al., 1992). It has even been suggested that
ATP can itself elicit further ATP release and, thus, additionally potentiate the response
(Vizi et al., 1992). In parasympathetic neurotransmission, ATP might act as a co­
transmitter with ACh.
The effect of ATP at the neuromuscular junction is puzzling and not yet fully
understood (Bean, 1992). Because ATP is co-stored with acetylcholine in some synaptic
vesicles, attention has focussed on the effects of ATP at nicotinic synapses where ATP
has been found to potentiate the effects of acetylcholine on skeletal muscle (Eward, 1976).
6
There are conflicting data concerning whether ATP by itself can activate nicotinic
cholinergic receptor channels. In cell-attached patch recording from skeletal muscle ATP
induces activity of nicotinic acetylcholine receptor channels (Kolb, 1983).
However,
when outside-out patches are formed from skeletal muscle, application of ATP does not
activate channels. Furthermore, micromolar concentrations of ATP were reported to
produce a two- to threefold enhancement of channel activity in response to 0.1nM
acetylcholine in Xenopus skeletal muscle (Igusa, 1988). Therefore, ATP' s reported ability
to enhance synergistically the effect of low acetylcholine concentrations may be more
important than any agonist activity of its own.
The elucidation of the non-exocytotic release of ATP remains a challenge.
Abraham et al have described very interesting observations which suggest that the
multidrug resistance (mdrl) gene product (or P glycoprotein) can function as a channel
for ATP. Overexpression of P glycoprotein in transfected Chinese hamster ovary cells
was accompanied by a three-fold increase in the steady-state release of ATP to the
extracellular medium. The rate of ATP release in other cell lines known to express mdrl
showed a clear correlation with the relative amount of P glycoprotein present in the plasma
membrane. Significantly, patch-clamp electrophysiological measurements indicated that
ATP may directly permeate the P glycoprotein channel which may function as a charge
carrier. (Abraham et al., 1993).
Transmission using ATP as a messenger requires mechanisms for inactivation of
the nucleotide. This is particularly apparent when ATP acts as a fast transmitter, and
impulses of high frequency need to be transmitted. Surface-located enzymes for hydrolysis
7
of ATP have been found at sites of ATP release (Zimmermann, 1994). The end-product
of ATP hydrolysis is generally adenosine, a messenger in its own right and a salvage
product of cellular purine metabolism. The ubiquitous ecto-ATPase is a surface-located,
membrane-bound enzyme that presumably occurs in neural and glial plasma membrane,
as well as on the surface of endothelial cells. The activity of ecto-ATPase is generally
higher than that of ecto-ADPase. The enzyme ecto-ATPase has a broad substrate
specificity for purine and pyrimidine 5'-trinucleotides.
Thus, it is also capable of
hydrolyzing GTP or UTP. The enzyme responsible for the hydrolysis of extracellular
AMP (and also of other 5'-mononucleotides), 5'-nucleotidase, resides largely on glia and
endothelium and is also present on peripheral Schwann cells. Thus, it is closely associated
with sites of neurally elicited nucleotide release (Zimmermann, 1994). In a number of
cases, for example, in cholinergic synaptosomes from striatum or hippocampal mossy
fibres, 5'nucleotidase is located directly on nerve terminals (Zimmermann, 1994).
Furthermore, it is associated transiently with developing synapses (Zimmermann, 1994).
1.1.4 Structure of Purinergic Nerve
It is generally accepted that autonomic nerve profiles which contain predominantly
agranular vesicles are cholinergic, while those containing predominantly small (300-500
A) granular vesicles are adrenergic, at least in mammals (Burnstock, 1972). In addition
both cholinergic and adrenergic profiles contain a small number(about 3 -4 %) of large (600­
1500 A) granular vesicles which have also been called type I granular vesicles or
neurosecretory vesicles. A third profile, which has been characterized by large numbers
of mitochondria, usually with only two cristae, has been identified in smooth muscle
8
systems. This profile probably represents sensory nerves. In addition to the three types
of profiles for cholinergic, adrenergic and sensory nerves, a fourth nerve profile in
association with smooth muscle has been described.
This is characterized by a
predominance of large granular vesicles, which differ from those seen in small numbers
in adrenergic or cholinergic nerves in that they are larger (800-2000 A) and do not have
a prominent electro-transparent halo between the vesicle membrane and its granular core.
In order to distinguish these vesicles from the large granular vesicle (LGV) seen in
adrenergic and cholinergic nerves, the term large opaque vesicle (LOV) has been
introduced (Burnstock and Iwayama, 1971; Burnstock, 1972b)
The strongest evidence that nerve profiles containing large opaque vesicles represent
purinergic nerves comes from studies of amphibian lung and mammalian gut. It has been
shown that when the adrenergic nerves supplying the toad lung are destroyed with 6­
hydroxydopamine (6 -OHDA) both the purinergic inhibitory response to vagal stimulation
and the neuronal profile with a predominance of large opaque vesicles remain unchanged
(Robinson et al., 1971). These results are supported by the observation that treatment
with a catecholamine-depleting drug, reserpine, while strongly reducing the amount of
catecholamine detectable in the lung by histochemical and chemical assay techniques, did
not deplete the cores of large opaque vesicles.
1.2
Pharmacological Classification of Purinoceptor Receptor
In 1978, Burnstock proposed a basis for distinguishing two types of purinergic
receptors P1 and P2, which relied largely on an analysis of the voluminous literature
regarding the actions of purine nucleotides and nucleosides on a wide variety of tissue
9
(Burnstock, 1978). The original classification was based on four criteria:
a)
relative potency of the adenine compounds in a wide variety of tissues;
b)
selective action of the antagonists, methylxanthines (such as
theophylline and caffeine);
c)
activation of the adenylyl cyclase system by adenosine, but not ATP;
d)
induction of prostaglandin synthesis by ATP and ADP, but not
adenosine.
The first major receptor type activated by adenosine, was termed P1- purinergic
receptor and exhibited a potency order of adenosine > ADP > AMP; those activated
by ATP, were termed P2-purinergic receptors, and exhibited a potency order of ATP >
ADP > adenosine. P1 purinoceptors have subsequently been subclassified as A1, A2A ,
A2B and A3 receptors.
Al receptors are preferentially activated by N6-substituted adenosine
analogues, whereas A2 receptors are preferentially activated by ribose 5'-substituted
compounds. All adenosine receptors are coupled to adenylyl cyclase, with the Al and
A3 receptor inhibiting and A2 receptors stimulating cyclic AMP synthesis.
The A3
adenosine receptor is present in the heart and nerve endings. It has been found that this
receptor is inhibitory to adenylyl cyclase and couples to phospholipase C (Ramkumar et
al., 1993; Linden, 1991).
The development of relatively non-hydrolyzable ATP analogues has been utilized
in the characterization of P2 receptors, and the observation of differential effects of these
ATP and ADP analogues led Burnstock and Kennedy to propose that P2-purinergic receptor
subtypes exist. Burnstock and Kennedy provided the first in-depth review of the functional
10
effects of various ATP analogs and suggested a P2,, and P2), receptor nomenclature to
differentiate excitatory and relaxant effects mediated via P2 receptor activation (Burnstock
and Kennedy, 1985). The P2, purinoceptor mediates the contraction of the vas deferens
and urinary bladder of guinea-pig and rat, as well as the rabbit ear artery smooth muscle;
the P2), purinoceptor mediates relaxation of guinea-pig taenia coli and rabbit portal vein.
Nucleotides can be structurally modified in three major domains: the base moiety, the
ribose (or deoxyribose) group, and the polyphosphate chain. Particularly useful ATP
analogues are those containing substitutions at the 2'-hydroxyl of the ribose or substitutions
of methylene bridges for the normal oxygen bridges of the polyphosphate chain. The
agonist potency order at the P2, purinoceptor is )3,1'-methylene-L-ATP > a,B-methylene-
ATP (aB- MeATP) > ATP = 2-methylthio-ATP (William and Cusack, 1990), while the
potency order at the P2), purinoceptor is 2-methylthio-ATP > ADPBS > ATP > aBMeATP (Cooper et al., 1989; Dubyak et al., 1993). Substitution of a methylene bridge
in the polyphosphate chain increases smooth muscle stimulant or contractile potency
(Cusack and Hourani, 1991; Magurire and Satchell, 1979), but reduces potency on smooth
muscles that are relaxed by ATP (Magurire and Satchell, 1979). 2-Methylthio-substitution
greatly increases the potency of ATP in some tissues, e.g. rat taenia coli (Magurire and
Satchell, 1979) with little or no effected in others, e.g. guinea pig urinary bladder
(Burnstock et al., 1983).
Signal transduction via P2 and P2-purinergic receptors for extracellular ATP have
been widely studied. The P2,, receptors are members of the transmitter-gated ion channel
superfamily of receptors (Barnard et al., 1992) and have been detected by the direct
11
recording of rapid responses to ATP and the specific agonist a,B-methylene-ATP, both
at peripheral and central synapses (Bean, 1992). The examples include smooth muscle
of vas deferens, bladder , ear artery, other vasculature and some neurons (Fedan et al.,
1982; Cusack et al., 1987; O'Connor et al. ,1990; Bean, 1992). A common feature noted
in all of these cells is the ability of a,B-methylene-ATP, a selective agonist for P2.-type
receptors, to induce rapid activation and inactivation of these receptors and thereby
desensitize the cells to subsequent stimulation by ATP.
Purinoceptor transduction
mechanisms for the excitatory actions of ATP via P2x-purinoceptors on vascular and
visceral smooth muscle cells appear to 1) be associated with the opening of nonselective
ligand-gated cation channels, resulting in depolarization and subsequent opening of voltage-
dependent Ca2+ channels (Freil et al., 1988) or 2) be associated with a receptor-operated
Ca2+-permeable channel (Benham and Tsien, 1987). The effects of ATP in neuronal cells
are complex, but one direct effect is a rapid depolarization caused by increased cation
conductance (Krishtal et al., 1983; Krishtal, 1988; Bean, 1990).
The Pty receptors appear to belong to the G-protein coupled superfamily of
receptors.
Extracellular ATP at low (micromolar) concentrations stimulates, via G-
proteins, inositol 1,4,5-triphosphate production and/or phospholipase C (PI-PLC system)
and effects a mobilization of intracellular Ca2+ in hepatocytes, adrenal medullary cells
and vascular endothelial cells (Charest et al., 1985; Pearson and Carter, 1990).
P2y­
purinoceptors coupled to phospholipase C activation and intracellular Ca2+ mobilization
have also been demonstrated in turkey erythrocyte membranes (Harden et al., 1987;
Cooper et al., 1989; Boyer et al., 1989).
12
ATP effects in a variety of tissue systems were further delineated when the Plc
(platelet) and P2z (mast cell) subtypes were defined (Gordon, 1986). Platelets express
a nucleotide receptor (P2) which is activated when occupied by ADP. This receptor is
distinct from other P2-purinergic receptors which recognize ADP as an agonist, in that
ATP acts as an antagonist of ADP-induced signaling in platelets (Greco et al., 1992).
The P2z
receptor is defined as that entity mediating the response to the
tetraphosphate form of ATP (ATP-4).
In 1977, Rozenggurt et al. reported that
extracellular ATP could alter the membrane permeability properties of certain transformed
(but not untransformed) murine fibroblasts and rapidly release their endogenous nucleotides
and other normally impermeable metabolites to the extracellular medium. During the
intervening 15 years, other investigators established that this ATP-induced loss of
endogenous metabolites reflects the formation of nonselective pores that mediate a
generalized bidirectional increase in plasma membrane permeability to molecules as large
as 900 Da (Rozengurt, 1977).
Thus responsive cells not only release endogenous
molecules and ions but also become hyperpermeable to extracellular molecules such as
Ca2+, other ions, and exogenous organic molecules. This type of ATP-induced response
has been well-characterized in macrophages (Beyer et al., 1991) . The prolonged opening
of these ATP-induced pores or channels will lead not only to equilibration of
transmembrane ionic gradients (and hence to the increase in the intracellular Ca2+) but
also to loss of the critical intracellular metabolites (e.g., nucleotides) and eventual cell
death. P2z or pore-forming ATP receptors exhibit a very high selective for ATP over other
adenine and nonadenine nucleotides (Rozergurt, 1977). ADP, AMP, and all nonadenine
13
nucleotide triphosphates produce no response or only modest responses even when tested
as high concentration. Synthetic ATP analogues that are modified on the adenine or
triphosphate moieties (e.g. a,11- methylene -ATP) are generally inactive. Although the
receptor is highly selective for ATP, it is important to stress that maximal activation of
the receptor requires the presence of near-millimolar concentrations of extracellular ATP
when target cells are incubated in saline containing physiological concentrations of divalent
cations. This contrasts with the ability of ATP, in the micromolar to hundred micromolar
range, to activate G protein-coupled ATP receptors or the ATP-gated ion channel
receptors, even when cells are incubated in medium containing standard concentrations
of extracellular Mg' and Ca". However, when divalent cations are removed from the
extracellular medium, P2z-receptor effects on ion flux and pore formation can be maximally
activated by micromolar levels of ATP (Buisman et al., 1988). Thus ATP', rather than
Mg-ATP, is the actual ligand that activates these receptors.
Further characterization of the effects of nucleotide analogues led to the
identification of a nucleotide receptor (P2u) based on agonist studies (O'Connor et al.,
1991).
P2u receptors represent a subclass of P2 receptors sensitive to the pyrimidine,
uridine triphosphate (UTP) (Needham et al., 1987). When prostaglandin PGI2 production
in piglet aorta endothelial cells is measured, 2-MeSATP produces a lower maximum
response than ATP, suggesting that it could be a partial agonist. Similarly, in rat aorta,
in which 2-MeSATP is a potent endothelial-dependent relaxant but produces a maximum
response 41% of that shown by ATP. A partial agonist will act as a competitive antagonist
of a full agonist acting at the same receptor when the full agonist is present at a
14
concentration that produces a high level of receptor occupancy. Surprisingly, 2-MeSATP
failed to act in the predicted fashion, subsequent responses to the full agonists were not
inhibited. The fact that 2-MeSATP does not interact with ATP suggests that it acts by
a different mechanism (Dainty et al., 1990; Needham el at., 1987). The equivalence of
UTP with ATP suggests that "purinoceptor" may be an inappropriate description for this
putative subpopulation of P2- receptors. Indeed, appreciation of this point has also led
Davidson and colleagues to introduce the term "nucleotide" receptor for the ATP/UTP­
sensitive site on sheep pituitary cells (Davidson and Wakefield, 1990). Therefore, this
"nucleotide" receptor was proposed and characterized by the following agonist potency
order: UTP = ATP > ADP > a(3-MeATP, 2-MeSATP (O'Connor, 1991). The rat aorta
may contain a heterogeneous population of receptor types (possibly both Pty and P2,
receptors) activation of which results in the same functional response
in this case
endothelial-dependent relaxation (Needham et al., 1987). The lack of interaction observed
between partial and full agonists as described above strongly supports the concept of
distinct receptors, rather than different transducers linked to the same receptor.
Later
studies have shown that this receptor subtype, like the Pty receptor, couples to a Gprotein and activates phospholipid-specific phospholipase C (PLC) and Ca2+ mobilization
(Sternweis et al., 1992).
Among the five main classes of P2 receptors, the P2, and Pty receptors are best
characterized. Two key reference compounds typically used to discriminate between
P2,,
and Pty receptors are «03-methylene-ATP and 2-methylthio-ATP selective agonists at the
P2x
and Pty receptors, respectively. Owing to the lack of specific, potent and reversible
15
antagonists, current pharmacological subclassification P2 purinoceptors is base mainly on
agonist potencies. The best-characterized antagonist is the aromatic polysulfonic acid
suramin. Suramin is a trypanocidal drug which was introduced into therapy in 1920.
It was reported that suramin could antagonize the response of mouse vas deferens to
aflMeATP and it was suggested as a specific antagonist for P2-purinoceptor (Dunn and
Blake ley, 1988). However, pharmacological results have shown that suramin is also an
antagonist of the Pty-purinoceptor, with an affinity equivalent to the P2, receptor (Hoyle
et al., 1990). Thus the action of suramin dose not appear to be selective. More recently,
pyridoxalphosphate- 6- azophenyl -2 -4- disulfonic acid (PPADS) has been introduced as a
novel antagonist of P2 purinoceptors which was found to interact specifically with Pa
receptors in the rabbit vas deferens and other smooth muscles (Lambrecht et al. , 1994;
Zigamshin et al. , 1993; Zigamshin et al. , 1994).
1.3
Neuronal ATP Receptors and Their Transmembrane Signalling
Systems
The above subclassification has been based on non-neural tissues and does not apply
to all tissues and species investigated. It might not match functional responses in nervous
tissue (Ines et al. , 1993). ATP activates at least three types of ATP receptors in the CNS
(P2., P2y, P2J Purinoceptors are present not only at nerve terminals and effector organs
of peripheral neurons, but also at the cell bodies of both peripheral and central neurons.
The first characterization of an ATP-activated conductance under voltage-clamp
was reported by Krishtal et al. (1983) in mammalian sensory neurons (dorsal root ganglion
neuron: DRG). Notably, all the conductances are highly selective for ATP as a ligand
16
and show prominent desensitization with continued presence of ATP in these sensory
neurons.
In all cases, ATP activates a cation-selective conductance that has a reversal
potential near OmV and shows strong inward rectification. The kinetics of activation are
very rapid in all the cells, suggesting that ATP may gate the conductance by directly
binding to the channel without an intervening second-messenger pathway (Bean, 1990).
This direct activation of non-selective cationic channels by P2 purinoceptors was observed
in about 40% of DRG neurons from rats and bullfrogs (Ines and Norenberg, 1993).
P2-purinoceptor activation appears to depolarize locus coeruleus neurons, a major
group of noradrenergic neurons that is concentrated in a pontine nucleus. It was shown
in this study that ATP itself did not produce consistent changes in the membrane potential
or input resistance in these neurones. However, in the presence of the P1- purinoceptor
antagonist 8- cyclopentyl- 1,3- dipropylxanthine (DPCPX), ATP consistently increased the
firing rate and evoked an inward current. The rank order of agonist potency was
aj3MeATP > 2-MeSATP > ATP in these rat locus coeruleus neurones. Thus, afiMeATP
was the most active of the three P2 -agonists tested, suggesting that the receptor belongs
to the P2x -subtype (Tschopl, 1992). However, there was no desensitization to afi-MeATP
on repeated or continuous application, suggesting that locus coeruleus neurons may possess
a P2 purinoceptor with somewhat atypical characteristics.
This corroborates earlier
observations with dorsal horn neurones that had shown that ATP elicits inward currents
and neuron depolarization (Jahr and Jessell, 1983). It has been reported that rat superior
cervical ganglion (SCG) contains P2x purinoceptors (Connolly et al., 1993).
a,B­
Methylene-ATP depolarized the SCG and suramin antagonized the depolarization.
17
Prolonged contact of SCG with a13MeATP selectively depressed subsequent responses
to ATP and a13MeATP (Connolly, 1994). This selective desensitization by aliMeATP
provides a means of identifying P2x-purinoceptors (Burnstock and Kennedy, 1985).
It
has also been reported that SCG neurons and PC12 cells are insensitive to «13MeATP (Li,
et al. 1993; Brake, et al., 1994; Nawazaka, et al., 1990).
The reasons for this
discrepancy are presently unclear, however, the possibility of multiple P2, receptor
subtypes in neural tissue must be considered.
There is strong evidence that ATP can act as a fast excitatory neurotransmitter
in both the CNS and PNS. In cultured coeliac ganglia neurons of guinea-pigs, electrical
stimulation of the fibre network evoked fast excitatory postsynaptic potentials (Evans et
al., 1992; Silinsky et al., 1992). These potentials were mimicked by ATP and blocked
by the P2 purinoceptor antagonist suramin.
ATP (10 iiM) and ai3MeATP (10 gM)
produced responses of similar amplitude. The importance of the study is that it shows
for the first time that ATP is a fast transmitter at a neuro-neuronal synapse.
Additional
experiments in the CNS showed that electrical stimulation of inputs to medial habenula
neurons evoked an excitatory postsynaptic current. ATP itself was ineffective (probably
due to its rapid degradation by ecto-nucleotidases), but 0/13-MeATP caused a small inward
current which was also inhibited by suramin.
Therefore, ATP appears to be a fast
transmitter in this area of the CNS and P2 purinoceptors located in these synapses belong
to the P2 family of ligand-gated ion channels (Edwards et al., 1992)
Furthermore, in a large proportion of the neurons in parasympathetic ganglia of
the cat urinary bladder wall as well as in many guinea-pig and rat parasympathetic
18
intracardiac neurons, ATP induces depolarization by opening the same type of nonse­
lective channels as in dorsal root ganglia (Fieber and Adams, 1991; Krishtal et al. , 1988).
However, in cardiac ganglia of rat and bullfrog, the order of agonist potency was 2­
MeSATP = ATP > af3MeATP, a profile consistent with the presence of a Pty receptor
subtype. The lack of desensitization of the response in rat cardiac neurones resembles
that observed in dorsal root ganglion (DRG) cells (Bean et al., 1990). This type of P2
receptor was defined as P2ya ( Illes and Norenberg, 1993), since the pharmacologic profile
is similar to Pty. It was also reported that this type of P2y,, receptor existed in locus
coeruleus neurons (Shen and North, 1993). The rank order of potency in evoking the
inward current was 2-MeSATP > ATP > a ,BMeATP.
An alternative pathway has been described in bullfrog DRG neurons (Tokimasa
and Akasu, 1990). These DRG neurons are characterized by a round shape and a diameter
larger than 50 pm, P2 purinoceptors indirectly, via a guanine-nucleotide binding (G)
protein, inhibited the hyperpolarizing M current and thereby caused depolarization. This
current is a particular type of voltage-sensitive, non-inactivating K+ conductance, which
contributes to the resting membrane potential and is blocked by acetylcholine via
muscarinic acetylcholine receptor activation (M for muscarine).
The absence of
desensitization during a long-lasting application of ATP may also suggest a Pty
purinoceptor-mediated effect. This neuronal P2 receptor was defined as P23,6 (Illes and
Norenberg, 1993).
From the above electrophysiological investigations, two alternative ionic transduc­
tion mechanisms may be activated by neuronal P2 receptors. One mechanism is a ligand­
19
gated ion channel.
Its activation by ATP leads to opening of nonselective cationic
channels, which were described originally in neurons of dorsal root ganglia (Krishtal et
al., 1988). Extracellular Na+ and Ca 2+ enter the cell, while intracellular K+ leaves the
cell via these channels; hence, ATP evokes a net inward current and depolarization. The
reversal potential of ATP-induced currents is near OmV.
This suggests an equal
permeability for Na+ and K+, while the ratio of Na+ and Ca2+ permeabilities is 0.3
(Bean, 1990b). Another representative of the neuronal P2 receptor family may be coupled
to a G protein. This G protein closes a K+ channel, thereby, preventing intracellular K+
from leaving the cell, and causes depolarization. Such a transduction mechanism was
described in a subset of DRG neurons (larger than 50 pm in diameter); in these cells a
particular type of K+ current, inhibited also by muscarinic acetylcholine receptor activa­
tion, the M current, was blocked by the P2 purinoceptors (Tokimasa et al., 1990). The
reversal potential of the ATP-induced current is identical to the reversal potential of the
M current and depends on the ratio of extra- and intracellular K+ only.
In rat pheochromocytoma (PC12) cells ATP and selected analogues activated an
inward current, which probably passes through similar non-selective cationic channels
as found in sensory neurons (Nakazawa et al., 1990b; Inoue and Nakazawa, 1992). ATP-
activated inward current in PC12 pheochromocytoma cells was characterized using the
whole-cell voltage-clamp technique. The order of potency in activating the inward current
was ATP > ATPTS > ADP; AMP, adenosine and a,13-methylene-ATP were inactive
at concentrations up to 1mM ( Nakazawa et al., 1990b). This inward current was blocked
by suramin ( Nakazawa et al., 1990a). In addition, ATP-activated Ca' influx directly
20
triggers catecholamine release. Therefore, ATP can play a role as neurotoxicant such
as glutamate which increases intracellular Ca2+ concentration to a toxic level (Inoue and
Nakazawa, 1992).
Furthermore, in PC12 cells where responses to ATP and UTP with
similar potency are mediated by phospholipase C activation, 2-MeSATP has little or no
activity, suggesting the receptor is distinct from the Pty receptor (Boarder et al., 1992).
This is the first evidence for the existence of a P2u type receptor on neuronal
cells.
In summary, all above P2,, and P2),,, neuronal receptors in the locus coeruleus,
superior cervical ganglia, coeliac ganglia, medial habenula and parasympathetic neurons
appear to open ligand-gated ion channel by opening the same type of nonselective cation
channel. This type of channel is similar to the ion channel on sensory neurons which was
first described by Krishtal et al. (1983). However, receptors on a subset of DRG neurons
appear to be G protein-coupled. ATP may indirectly via the activation of a G protein
inhibit a resting K+ conductance in this subset of DRG neurons.
1.4
Purinoceptor Radioligand Binding Studies and Receptor Molecular Biology
[311]«13-MeATP (Bo and Burnstock, 1989) and [35S]ADPj3S (Cooper et al., 1989)
have been used to label P2 and P2), receptors, respectively. The displacement experiments
demonstrated that the pharmacological specificity of [3H]«(3 -MeATP binding is : 00­
MeATP > Or-MeATP > suramin > ATP > ADP > 2-MeSATP > > adenosine, which
is in good agreement with the potency order of these ligands in pharmacological actions
at the P2x-purinoceptor (Burnstock & Kennedy, 1985).
This study has also shown
magnesium ions had profound effects on [31-I]co3-MeATP binding, with magnesium
21
producing a decrease in both the high- and low-affinity maximum binding of [3li]a13­
MeATP. However, magnesium ions had only a slight effect on the Kd values for highaffinity binding, while the Kd values for the low-affinity binding showed larger increases
at high magnesium concentration. In contrast the study of P2y-purinoceptor binding using
[35S]ADPj3S
indicated that without magnesium the high-affinity binding was almost
completely lost (Cooper et al., 1989). The potency order was: 2-MeSATP > ADPj3D
> ATP > ADP > a43 -MeATP > fir-MeATP for displacement of [35H]ADPJIS binding
to turkey erythrocyte membranes (Cooper et al., 1989).
The cloning of P2 receptors led Barnard and co-worker to develop a new systematic
nomenclature for P2 G protein-coupled receptors based on the molecular cloning of these
receptors (Barnard et al., 1994). A cDNA encoding a novel member of a G-protein­
coupled receptor (GCR) superfamily, an ATP receptor, has been isolated from an
embryonic chick whole brain cDNA library. The encoded protein has a sequence of 362
amino acids (41 kDa) and shares no more than 27% amino acid identity with any known
GCR.
When injected with a complementary RNA (cRNA) of this receptor, Xenopus
oocytes display a slowly-developing inward current (the Ca24--dependent Cl- channel) in
response to application of ATP. The pharmacology of the expressed protein defines it
as a Pty purinoceptor (Webb et al., 1993). This receptor responded to agonists with the
potency order 2-MeSATP > ATP > ADP, but did not respond to UTP and the P2,-­
receptor selective ligand «,13-methylene-ATP.
This brain subtype has been designated
as subtype, P2y1 (Barnard et al., 1994). In contrast, a recombinant receptor characterized
by Lustig et al. when expressed in Xenopus oocytes, responded to agonists with the
22
potency order ATP =UTP > ATPPS > > 2-methylene-ATP and was designated as a
P2. receptor, a subtype was identified in NG108-15 cells. In the scheme of Barnard, this
subtype was named P2y2 (Barnard et al., 1994).
This cloned receptor resembles a
metabotropic P2, receptor; activation by either ATP or UTP elicits the mobilization of
intracellular calcium (Lustig et al., 1993). Furthermore, Webb et al have identified a
second recombinant brain-derived receptor that responds to agonists with another activity
series, having a strong preference for ADP (Barnard et al., 1994). This series is similar,
but not identical to that of the Pet receptor of mammalian platelets (which is activated by
ADP, but not ATP). Since it is distinct from the platelet receptor, it has been designated
P2y3 (Barnard et al., 1994).
Recently Valera and North group have cloned a complementary DNA encoding
a P2, receptor from rat vas deferens and expressed it in Xenopus oocytes and mammalian
cells.
ATP activates a cation-selective ion channel with relatively high calcium
permeability.
Structural predictions suggest that the protein (399 amino acids long) is
mostly extracellular and contains only two transmembrane domains plus a pore-forming
motif which resembles that of potassium channels. The P2x receptor thus defines a new
family of ligand-gated ion channels.
ATP evoked inward currents in oocytes injected
with poly(A)+ RNA from rat vas deferens, but not uninjected oocytes. 2-methylthio-ATP,
ATP, a,B-methylene-ATP and ADP evoked inward currents with a rank order of potency:
2-methylthio-ATP > ATP > a,B-methylene-ATP > > ADP. UTP and GTP were
ineffective. Currents evoked by ATP, a,13- methylene -ATP, 2-methylthio-ATP and ADP
were reversibly blocked by suramin (Valera et al., 1994).
These pharmacological
23
properties define the cloned receptor as a P2 purinoceptor.
Simultaneously with the report of Valera et al. (1994), Julius and coworkers
reported the cloning of a P2 receptor from PC12 cells (Brake et al., 1994).
A
complementary DNA expression library constructed from PC12 messenger RNA was
screened for the production of ATP-evoked inward current responses in Xenopus oocytes,
and a single cDNA clone was isolated. Prolonged application of ATP elicited a large
inward current with a rapid onset and slow rate of desensitization.
ATP, ATP-r-S and
2-methylthio-ATP were found to be roughly equipotent as agonists, whereas cc,fi­
methylene-ATP and 13,r-methylene-ATP were inactive as agonists or antagonists at the
P2x receptor cloned from PC12 cells. This contrasts with the rank order of potency of
2-methylthio-ATP > ATP > a,13-methylene-ATP > > ADP at the P2 receptor cloned
from vas deferens. The non-subtype selective P2 receptor antagonists suramin and reactive
blue-2 reversibly antagonized ATP-evoked responses by over 95 % at the PC12 derived
P2x receptor (Brake et al., 1994).
Dose-response analysis revealed a half-maximal effective concentration (ED50) for
ATP of 60AM at the PC12 P2 receptor clone (Brake et al., 1994). Extracellular Zn2+
has been shown to potentiate ATP-activated currents in rat sensory and sympathetic
neurons possibly by increasing the affinity of the P2 receptor for agonists (Cloues et al.,
1993). Addition of 10AM Zn' to the bathing solution had a similar effect on the cloned
PC12 receptor shifting the EC50 from 60 AM to 15 AM.
The cloned PC12 P2x receptor cDNA P2xR1 contains a 1,416 nucleotide open
reading frame encoding a 472-amino-acid protein with a predicted MW of 52,557. This
24
P2xR1 clone contains two hydrophobic putative transmembrane domains (Ml and M2)
with an intervening hydrophilic loop of 270 residues. Biochemical and genetic data
suggest that within this family of cation channels the M2 domain, together with an adjacent
hydrophobic segment (H5), forms the ion pore.
The current-voltage relationship of the cloned PC12 P2x receptor shows a marked
inward rectification with a reversal potential of about -5mV.
Replacement of all
extracellular monovalent cations by the large organic cation Tris, shifted the reversal
potential to -45 mV and reduced the current amplitude significantly. Replacement of all
extracellular Na + by K+ did not affect the reversal potential. These results suggest that,
like native P2 receptors (Fieber et al., 1991), the cloned receptor incorporates an inwardly
rectifying, nonspecific cation channel equally permeable to Na+ and K+, allowing
conduction of even large cations. Finally, replacement of all extracellular Ca2+ by Ba2+
had no effect on the observed current indicating that there is no significant contribution
from endogenous Ca2+-activated Cl- channels in Xenopus oocytes.
The existence of multiple P2x subtypes has been inferred from the differential
actions of agonists such as a,13-methylene-ATP and 2-methylthio-ATP (Bean, 1992). In
neurons and PC12 cells, the agonist selectivity does not fall neatly into any of the
subclassifications of P2 receptors that have been previously proposed (Burnstock et al,
1985). Unlike the agonist selectivity for the P2x receptor, a,13-methylene-ATP and 13,I'­
methylene -ATP are not more potent than ATP for activating the channels studied. For
example, ATP and ATPI'S activated an inward current in PC12 cells, but a13MeATP was
inactive in the these cells.
25
1.5
Epileptogenic Site in Rat Prepiriform Cortex
The deep prepiriform cortex has been described as a crucial epileptogenic site
(Piredda and Gale, 1985). This represents the first identification of a site in the forebrain
from which generalized convulsions can be elicited by a focal injection of low doses of
bicuculline in low doses. It was found that picomole amounts of bicuculline, kainic acid,
or carbachol, and micromole amounts of glutamate injected unilaterally into this area
resulted in contralateral (followed by bilateral) clonus of forelimbs, rearing and falling,
accompanied by epileptiform activity in EEG recordings from cortical and subcortical sites.
According to the studies of Gale and coworkers, the most consistently effective
site in the rat prepiriform cortex (PPC) for obtaining motor seizures corresponds to the
following stereotaxic coordinates when the incisor bar is 5.0 mm above the interaural line:
4.0 mm anterior to bregma, 3.0 mm lateral to midline, and 6.5 mm below dura. It is
therefore possible that seizures obtained following high doses of convulsants placed
elsewhere in the forebrain may, in fact, be caused by diffusion of small amounts of the
drug to the prepiriform cortex. The doses of convulsants effective for producing seizures
following focal injections in the prepiriform cortex are considerably lower than those that
have been previously applied to other forebrain areas in attempts to produce seizures. Even
with doses more than 20 times higher than those used by Gale's laboratory, neither
bicuculline nor carbachol caused seizures when placed in the amygdala or hippocampus
when placed in the cortex these high doses have produced, at most, only focal seizure
activity (Piredda and Gale, 1985). The authors concluded that this site in the forebrain
was distinct from other sites in three important ways: (1) it was considerably more
26
sensitive; (2) it did not require repeated stimulation in order to elicit bilateral motor
seizures; and (3) it was responsive to chemoconvulsants with a variety of mechanisms
of action, including manipulation of GABAergic, cholinergic, and excitatory amino acid
receptors.
The fact that bilaterally synchronous seizure activity consistently and rapidly
occurred following unilateral application of the chemoconvulsants into the deep prepiriform
cortex suggests that this area may be a source of an excitatory drive to limbic and cortical
structures of both hemispheres.
The goals of this study were (1) to determine whether the rat prepiriform cortex
is a neuroanatomical substrate for responses to focal injections of ATP receptor agonists
; and (2) to pharmacologically characterize the ATP receptor mediating the response in
this site.
2. MATERIALS AND METHODS
2.1
Animals and Stereotaxic Surgery
Male Sprague-Dawley rats (350-375g) maintained at 22°C on a standard 12 hour
light/dark schedule, with ad libitum
access to food and water. Stereotaxic surgery,
microinjection technique and drug treatment protocol were performed using methods
described in detail by Franklin et al. , (1989). Animals were anaesthetized with Equithesin
(2.7m1/kg, i.p.) for stereotaxic surgery. The head of the rat was immobilized in a
stereotaxic frame with an incisor bar positioned 3.5 mm below the interaural line. In each
animal a paired stainless-steel 22-gauge guide and 28-gauge injection cannulas (Plastic
27
One, Inc., Roanoke, VA) was implanted unilaterally in the prepiriform cortex at
coordinates 6.5 mm below the dura, and 3.3 mm right of the midline, and 2.0 mm anterior
to the bregma (Piredda and Gale,1985; Franklin et al., 1988).
The tips of the
corresponding 28 gauge injection cannulas always extended from the guide lumen a
minimum of 1.0 mm. After implantation, the guide cannula were fixed to the cranium
and one stainless-steel anchoring screw threaded into adjacent bone by means of dental
acrylic (Kerr Manufacturing Co., Romulas, MI).
2.2
Drugs and Drug Administration
(-)-Bicuculline methiodide (BMI), 8-(p-sulfophenyl) theophylline (8-pSPT),
ADPBS, 2-methylthio-ATP (2-MeSATP), 2-methylthio-adenosine (2-MeSAdo) and B,11­
methylene-L-ATP (B,r-Me-L-ATP) were purchased from Research Biochemicals, Inc.
a,B-methylene-ATP (a,B-Me-ATP), B,r-methylene-ATP (B,11-Me-ATP), ATP, UTP,
a,13-methylene-ADP (AOPCP), sodium citrate and sodium pyrophosphate were purchased
from Sigma Chemicals. Suramin was purchased from Miles Laboratories (West Haven,
CT).
All drugs listed in Table 1 were dissolved in artificial cerebrospinal fluid (ACSF),
in mM: Na 150, K 3.0, Ca 1.4, Mg 0.8, P 1.0, Cl 155) at the desired concentrations.
The 2-MeSAdo solution was prepared by dissolving the compound in a small volume
of 0.1 N HC1 and adjusting the pH to neutral with 0.1N NaOH and adding ACSF to reach
the desired concentration. In order to avoid forming a precipitate, this solution required
gentle heating and sonication to achieve complete dissolution. The pH values of solution
of 13,11-methylene-L-ATP, ADPBS, a,13-methylene-ATP, 13,11-methylene-ATP and ATP
28
were adjusted to neutrality with 0.1N NaOH.
Intracerebral microinjections were performed using a modification of the method
described by Franklin et al. (1989). A 28 gauge injection cannula was connected by a
PE 20 tubing to a Hamilton microsyringe (1 Al) mounted in a Harvard infusion pump.
All drugs solutions in the ACSF vehicle were injected at a constant rate of 0.9n1/sec in
a volume of 120 n1 over a 2 min and 7.2 sec period; except 2-MeSAdo which was injected
at a constant rate of 2.0 nl/sec in a volume of 500 nl over a 4 min and 10 sec period.
The injection cannula was left in place for 1 min after termination of injections. The drugs
listed in Table 2 were administered in a CaC12 vehicle (equimolar concentration ATP
analogs and CaC12) and were injected at a rate of 0.9n1/sec and volume of 120 nl over
2 min and 7.2 sec or at a rate of 2.0 nl/sec in a volume of 500 n1 over a 4 min and 10
sec. An injection volume of 1/21 was used in experiments involving coadiministration of
ATP and 8-pSPT in the CaC12 vehicle (2.0 nl/sec), inasmuch as 8-pSPT was somewhat
insoluble in the CaC12 solution requiring a more dilute solution.
Suramin, sym-bis (m-aminobenzoyl-m-amino-p-methylbenoy1-1-naphthylamino-4,6
trisulphonate) carbamide, was used as the hexasodium salt. This compound has six sulfide
groups. The ability of sulfite and sulfate groups to form complexes which tissue Ca2+
(Labella, 1979; Smith and Martell, 1975c) is a potential confounding variable in
pharmacological studies employing microinjections.
To eliminate the role of Ca2+
chelation in the pharmacological responses to suramin injection, a CaC12 vehicle was
employed (CaC12:suramin molar ratio, 3:1).
29
2.3
Experimental Protocol and Behavioral Assay
2.3.1 Verification of cannula placement
Animals were allowed a 24-48 hour period to recover from stereotaxic surgery
prior to testing. Animals were challenged with a dose of 118 pmol BMI injected into the
PPC and were then placed in a 40 x 40 x 30 cm plexiglass testing chamber for observation
of seizure activity.
The observation period for seizure activity was 30 min beginning
at the time of infusion of the drug into the rat PPC.
5 were used for subsequent studies.
Rats with seizure scores of 4 or
In most cases, the seizures began with mouth and
jaw clonus and myoclonic jerks which rapidly progressed to bilateral clonic movements
of the forelimb, and finally with rearing and falling (stage 4 and 5) within 10
40
seconds. Severity of generalized motor seizures induced by BMI was scored as follows:
0, no seizure; 1, myoclonic jerks of the contralateral forelimb; 2, mild forelimb clonus
(mouth and facial movements, clonus of jaw, myoclonic jerks and head nodding) lasting
more than 5 sec.; 3, severe forelimb clonus lasting more than 15 sec.; 4, rearing, in
addition to forelimb clonus; 5, loss of balance and for falling in addition to rearing and
forelimb clonus.
On day two, 24 hours after the pre-test, animals were tested with
experimental compounds and the severity of seizures was recorded.
2.3.2 Evaluation of effects of test compounds
In order to test whether PPC is a neuroanatomical substrate for responses to focal
injection of ATP, animals were unilaterally microinjected by a series of ATP analogues
on day 2. The severity of generalized motor seizures induced by the series of ATP
analogues was scored using the same system as described for BMI.
30
In the experiments assessing the interaction of nucleotides or nucleosides with BMI,
animals were pretreated with various doses of 2-MeSATP or 2-MeSAdo which were
injected into the PPC 15 min prior to the administration of a challenge dose of BMI. In
addition, in the experiments assessing the interaction of the P2 antagonist suramin with
the P2,, agonist13,r-Me-L-ATP, animals were pretreated with 1.0 nmol of suramin, which
was injected into the PPC 15 min prior to the administration of various doses of B,T-Me-L­
ATP. If a sign of seizure behavior was not observed or if seizure score was less than
the control BMI response following any anticonvulsant drug treatment, anticonvulsant
effects were confirmed by assessment of the level response to BMI challenge 24 hour later
(post-test day 3). If animals post-tested with BMI did not maintain the same magnitude
of sensitivity to BMI as displayed on Day 1, they were excluded from data analysis.
2.3.3 Coadministration of Compounds
On day 1, animals were challenged with a dose of 118 pmol BMI injected into
the PPC, and only rats with seizure scores of 4 or 5 were used for coadministration
studies. On day 2, the focal injection of ATP alone (100 nmol), 8-pSPT alone (1.0 nmol)
or co-injection of ATP (100nmol) with 8pSPT (1.0nmol) was performed. All compounds
were injected in a CaC12 vehicle. The injection volume was 1A1.
2.3.4 Desensitization
This experiment was designed to investigate possible desensitization to «,13-MeATP
in rat prepiriform cortex.
Two sequential injections of a,13-methylene-ATP were
administered 30 min apart to assess the presence of desensitization.
31
2.4
Membrane Preparation and Binding Assay
Male Sprague-Dawley rats were decapitated and the frontal, ventral area of the
cerebral cortex was quickly dissected on ice.
A total particulate fraction was prepared
by homogenizing tissue with a Dounce homogenizer in 100 volumes of 10 mM Tris buffer
(pH 7.7) containing 10mM EDTA, and centrifuging at 37,000 x g at 4° C for 10 min.
The pellet was resuspended and incubated in the same buffer containing 2.5 IU/ml
adenosine deaminase, 150 mM NaC1 and 100/1M GTP at 37° C for 30 min in order to
eliminate endogenous adenosine from membranes. Following centrifugation as described
above, this pellet was washed 3 additional times in 50 mM Tris buffer, and the final pellet
was resuspended in a 50 mM Tris buffer containing 2.5 mM MgC12. The membrane
preparation was used immediately in binding assays.
Equilibrium competition experiments utilized [3H]DPCPX as an antagonist
radioligand to selectively label adenosine Al receptors. A series of adenosine analogous
including 2-methylthio-adenosine were used as displacers in these assays. The order of
addition for the radioligand binding assay was displacer or vehicle (50 til), membranes
(400 gl, 100-130/4 protein), 50 mM Tris buffer (pH 7.7) containing 2.5 mM MgC12
(525/21) and [311] DPCPX (25 gl). Incubation were carried out at 22°C for 90 min. The
final [3H] -DPCPX concentration was 0.15-0.2 nM. Nonspecific binding was defined as
that occurring in the presence of 100gM 2-chloroadenosine and was typically < 7.5 %
of total binding. Incubation was terminated by rapid filtration using a Brandel Cell
Harvester (Brandel Instruments, Gaithershurg, MD) over polyethylenimine (0.5 %)-soaked
GF/B filters, which were washed 4 times with 4 ml each of ice-cold 50 mM Tris buffer.
32
The filters were placed in scintillation vials and 5m1 of CytoScint scintillation cocktail
(ICN Biochemicals, Irvine, CA) was added. Filter disks were allowed to elute overnight
in scintillation fluid, and then counted in a Beckman LS6000 scintillation counter with
a counting efficiency of 48% to 55%. Analysis of competition data was performed by
a nonlinear regression fit of a logistic equation to the data using Graph Pad Inplot
software (Version 4.0).
2.5
Statistical Analysis
Linear regression analysis of modified Hill plots were used to quantitatively analyze
dose-response data. For each compound, the log [E/(100-E)] was plotted against the log
dose (nmol) where E represents the effect of the drug as a percentage of the maximum
response.
Protective responses from administration of compounds prior to BMI (118pmol)
injection were expressed as percent protection, which was calculated as percent reduction
of mean control seizure score. Comparison of seizure scores for various drug treatments
was accomplished using a Rank Sum test (Devore et al., 1986), which is a nonparametric
test for the difference of 2 population means.
3. RESULTS
3.1
Nucleotide-Evoked Convulsant Response in Rat Prepiriform Cortex
3.1.1 ACSF vehicle
The ACSF vehicle administered alone was without effect in the rat PPC. To
33
determine whether the rat prepiriform cortex is responsive to purine nucleotides, a
series of ATP analogues were focally injected into this brain area. All of ATP
analogues were injected in a ACSF vehicle.
Following injection a 30 min continuous
behavioral record was compiled and the highest mean seizure score obtained for each
animal with each treatment was recorded.
Of nucleotides tested, the potencies of
B,r-methylene-L-ATP (ED50 ± SE of 2.53 ± 0.4 nmol), «,13-methylene-ATP
(ED50 ± SE: 4.15 ± 0.64 nmol) and B,r-methylene-ATP (ED50 ± SE : 5.09 ±
0.52 nmol) were similar. These stable ATP analogs were much more potent than the
endogenous pyrimidine nucleotide UTP (ED50 ± SE : 56.17 ± 0.82 nmol) and the
purine nucleotide ATP (ED50 ± SE : 120.32 ± 0.45 nmol) as convulsants in the rat
prepiriform cortex (Table 1). The lower potencies of UTP and ATP are probably due
to UTP and ATP breakdown to UMP, uridine, AMP and adenosine respectively. In
addition, a,B-methylene-ADP (AOPCP), an ectonucleotidase inhibitor, also evoked
seizures with an ED50 value (ED50 ± SE: 10.75 ± 0.67 nmol) similar to that of
a,B-MeATP. Furthermore, a P2,, agonist, ADPBS, when applied to this brain area
showed a convulsant effect with an ED50 ± SE of 6.71 ± 0.91 nmol. Another P23,
agonist, 2-MeSATP, did not have any convulsant effects following focal injection into
this brain area even at doses as high as 100 nmol. The dose-dependent effects of the
nucleotides on behavioral seizure scores are shown Table 1.
The behavioral
characteristics of seizures elicited by nucleotides were qualitatively identical to those
of BMI-induced seizures in the rat PPC. The rank order of potency of nucleotides in
inducing convulsant effects in rat PPC was: B,r-methylene-L-ATP > = «,13-MeATP
34
> = 13,r-MeATP > = ADM > = AOPCP > UTP > ATP, although the potencies
of all the synthetic nucleotide derivatives varied by only a factor of five-fold.
3.1.2 CaC12 vehicle
Many studies had shown that ATP exerts excitatory effects in neurophysiological
studies. This excitation in many large part be related to the Ca' chelating action of ATP,
as other Ca' chelators are also quite effective (Galindo at el., 1969; Curtis R. at el.,
1960). The affinity of ATP for Ca2+ is comparable to that of citric acid and somewhat
less than that of pyrophosphoric acid (Bjerrum at el., 1957; curtis et al., 1960).
Naturally, the extracellular compartment is the ultimate source of all intracellular
ions, and whenever Ca channels open in the plasma membrane, extracellular Ca2+ ions
enter and stimulate intracellular events. However, there is another, strictly extracellular,
'stabilizing' action of calcium (raises the threshold for electrical excitation of nerve and
muscle). There are two theories that have been proposed to explain these effects (Hille,
1984).
The first is the divalent gating-particle theory, and the second, the surface-
potential theory. Both theories have validity, but the surface-potential contribution seems
to be the more important one. The first theory proposes that Ca2+ ions are an essential
component in all voltage-dependent gating: they bind to a channel component, closing
the pore, and they are pulled off the channel by membrane depolarization, opening the
pore. The second explanation is the surface potential theory. In this theory [Ca21,, closes
gates of voltage-dependent channels, raises the resting membrane resistance, and raises
the threshold for electrical excitation of nerve and muscle to give a stabilizing action.
It has been proposed that the external face of the membrane bears negatively charged sites,
35
ionized acids, to which Ca2+ ions bind. In the presence of high [Ca 21o, all the charges
might be neutralized by bound ions such that the electric field in the membrane would
simply be due to the resting membrane potential. In the absence of Ca2+, however, the
outer surface bears a net negative charge, setting up a local negative potential and altering
the electric field within the membrane. An intramembrane voltage sensor would see a
change in field equivalent to a membrane depolarization, thus Na+ channels, K+ channels,
and Ca2+ channels would tend to open (Bille, 1984).
Since all the test solutions in this study contained nucleotides at very high
concentration (8.3mM
100mM), it was necessary to investigate whether convulsant
effects were the result of a specific action at the P2 receptor or rather due to Ca2+
chelation. To establish the potential for Ca2+ chelation to effect a convulsant response,
citric acid and pyrophosphate were injected in the rat prepiriform cortex. The reason for
using these two compounds is that they have similar affinity to that of ATP for Ca2+
chelation (Curtis et al., 1960). The results of sodium citrate and sodium pyrophosphate
injection are summarized in Table 2. Both compounds were effective convulsants at a
dose of 24 nmol. It was therefore necessary to investigate of the effects of ATP analogues
using equimolar CaC12 as a vehicle.
To determine whether the CaC12 vehicle alone can
alter excitability, BMI (118pmol) and CaC12 (100mM) were coadministered. The results
indicated that CaC12 did not modify BMI-induced seizures in PPC (data not shown).
Using equimolar CaC12 as a vehicle, the most potent agonist was 11,1"-methylene-L­
ATP (ED50 ± SE: 4.32 ± 1.38 nmol), which is a P2 selective compound. The potency
of a,13-methylene-ATP was somewhat less (ED50 ± SE: 15.85 ± 1.29 nmol).
In
36
contrast to the results with the ACSF as vehicle, ADPBS, AOPCP and UTP have did
not any convulsant effects following the application into the PPC. The results indicated
the convulsant effects of these latter compounds may have been due to their ability to
chelate extracellular Ca', thereby changing the membrane surface charge and causing
membrane depolarization. The potencies of the compounds tested in the CaC12 vehicle
are summarized in Table 3. The dose response curves for the convulsant effects of B,r­
methylene-L-ATP and a,B-methylene-ATP are depicted in Figure 3. The rank order of
the potency was B,r-Me-L-ATP > a ,B-MeATP > > ATP.
3.2
Pharmacological Characterization of ATP Receptor Mediating Convulsant
Effects in Rat Prepiriform Cortex
In order to determine whether the excitatory effect of ATP analogues in rat PPC
is mediated by a P2 receptor, the P2 selective antagonist, suramin, was used in this study.
Suramin has been reported to antagonize the response of mouse vas deferens
to a,B-MeATP and therefore is considered to be a specific antagonist of the P2­
purinoceptor (Dunn and Blake ley, 1988). More recent results have confirmed its affinity
for P2, purinoceptors in vas deferens (Bo and Burnstock, 1992).
However,
pharmacological results have shown that suramin is also an antagonist of Pty-purinoceptors,
with equal affinity in guinea pig taenia coli (Hoyle et al., 1990). Therefore, suramin does
not discriminate between Pa and P23, receptors. When focally injected alone in the PPC
using ACSF as a vehicle, suramin had very potent seizure inducing effects (Table 1) with
an ED50 of 0.29 ± 1.23 nmol. To avoid the potential for complex formation between
the sulfite groups of suramin and extracellular calcium ions (Labella et al., 1979; Smith
37
and Martell, 1975), we used CaC12 (molar ratio Ca': suramin, 3:1) as a vehicle for
suramin injections. The molar ratio of 3:1 was based on the potential for a suramin
molecule to chelate 3 Ca2+ ions. As predicted this compound did not show any convulsant
effects even with the highest dose of 3 nmol when using CaC12 as a vehicle. This
demonstrates that Ca' chelation is the underlying mechanism for the convulsant actions
of suramin in an ACSF vehicle.
In order to investigate the interaction P2,, selective compound, 13,T-methylene-L­
ATP and P2 antagonist suramin, suramin was focally injected in a CaC12 vehicle into rat
PPC 15 min before the application 13,r-methylene-L-ATP. As shown in Figure 4, suramin
(lnmol) produced an antagonism of the B,P-methylene-L-ATP convulsant effect. The
dose response curve of B ,r-methylene-L-ATP was shifted to the right in a parallel
manner by suramin (Figure 3).
3.3
Investigation of Anticonvulsant Effects of 2-MeSATP in Rat
Prepiriform Cortex
2-MeSATP has been considered as a potent Per agonist in many tissue (Burnstock
et al. , 1985). This observation prompted us to study the responses of 2-MeSATP which
represents a reference compound for Pty receptors. However, 2-MeSATP was devoid
of convulsant effects in the rat PPC. In addition, 2-MeSATP has also demonstrated to
be the most unstable Per agonist (Cusack and Williams, 1990) and it can be rapidly
degraded to 2-MeSadenosine by ectonucleotidase (Cusack et al., 1987; Magurire et al.,
1979). Given the absence of a convulsant effect of 2-MeSATP, we assessed the potential
for this compound to exert an anticonvulsant response. In contrast to other nucleotides,
38
2-MeSATP produced an anticonvulsant effect against seizures induced by BMI. To
explore the potential involvement of the 2-MeSATP metabolite, 2-MeSadenosine in this
response, the effects of 2-MeSadenosine were assessed. 2-MeSadenosine produced an
anticonvulsant response which was similar to 2-MeSATP (ED50 = 5.3 nmol for 2­
MeSATP and ED50 = 4.1 nmol for 2-MeSAdo). The anticonvulsant effect of 2-MeSATP
may therefore be due to its rapid breakdown to 2-MeSadenosine.
The anticonvulsant
effect of 2-MeSadenosine may be due to the ability of this compound to activate adenosine
Al receptors. To determine the role of Pty receptors in the observed anticonvulsant effect
of 2-MeSATP, the effects of ADPBS on BMI seizures was determined. However, ADPBS
did not afford protection against BMI induced seizure (data not shown).
3.4
Correlation Between Adenosine Agonists as Anticonvulsant and
their Affinity for Adenosine Al Receptors.
To document the interaction of 2-methylthio-adenosine with the Al receptor, an
equilibrium competition binding experiment was performed using [3H]DPCPX to label
Al receptors. The equilibrium titration curves of adenosine analog were best
described by a two-site model. The 1050 values for the high affinity sites representing
adenosine analog-bound to Al receptors coupled to G-proteins to form a ternary
complex were used in subsequent correlation analysis. The agonist rank order of
potency as inhibitors of the specific binding [3H] DPCPX was R-PIA > CHA =
CPA > 2 -CIA > NECA > > S-PIA > > 2-MeSAdo > > CGS21680 > >
adenosine (Table 8).
Unilateral focal injection of a series of adenosine agonists in rat prepiriform cortex
39
elicited a dose-dependent reduction in the seizures induced by bicuculline methiodide.
The rank order of potency of these compounds in suppressing seizures is as follow: NECA
> cyclohexyladenosine > cyclophetyladenosine > R-PIA > 2-Chloroadenosine > S-PIA
> > 2-phenylaminoadenosine > > adenosine. These data suggest that the antiseizure
activity of these compounds in the PPC results from activation of Al adenosine receptors.
To explore correlations between Al receptor affinity and behavioral potency, linear
regression analysis was performed on inhibition of [3H]DPCPX binding by adenosine
analogs and their anticonvulsant potency as depicted in figure 6. The rank order potency
of adenosine analogs in vivo was significantly correlated with to that of their respective
IC50 values as inhibitors of [3H]DPCPX binding in vitro with a correlation coefficient
(r) value of 0.91 (P < 0.01, Figure 6).
3.5
8-pSPT on ATP Response in Presence of CaC12.
The ectonucleotidase activity of brain membranes may underlie the relative low
order of potency of ATP as a convulsant. The end product of ATP hydrolysis is generally
adenosine. The ubiquitous ecto-ATPase is a surface located, membrane-bound enzyme.
Previous studies have shown that ATP produces initial excitation followed by depression
(Phillis and Wu, 1981) or has excitatory effect only in the presence of the Al receptor
antagonist, DPCPX (Tschopl et al., 1992). In addition, adenosine is an inhibitory
neuromodulator in the CNS and its analogs have extremely potent anticonvulsant properties
when injected focally to block seizures induced by administration of bicuculline methiodide
into the prepiriform cortex (Zhang et al., 1992). Therefore, the lower potency of ATP
as a convulsant may be due to its degradation to adenosine which in turn may activate
40
Al receptors (Table 3). To further confirm this hypothesis, the co-administration of ATP
with the A,- antagonist, 8pSPT, was performed in the rat prepiriform cortex. Both 8-pSPT
(1 nmol) and ATP (100nmol) were each tested alone. The result showed that ATP
(100nmol) administered alone produced 5% of the maximum seizure response following
the focal injection of this compound in rat PPC, while the P1-antagonist 8pSPT (1 nmol)
was inactive when given alone. The coadministration of 8-pSPT (1 nmol) and ATP (100
nmol) potentiated the seizure activity of ATP (100 nmol). Coadministration of 8-pSPT
and ATP at these doses produced a mean seizure score of 43 % of maximum response.
This response was statistically different from the effect of ATP administered (P < 0.01
Table 7, Figure 7). These results indicated that the seizure induced by ATP administered
alone is attenuated by its metabolite adenosine through activation of Al receptors
3.6
.
Desensitization
Desensitization of P2x-purinoceptors by prolonged addition of a,B-MeATP has been
used to identify Pa-purinoceptors-mediated components of neurogenic vasoconstriction
in isolated blood vessels (Von Kugelgen et al., 1985). In addition, the Pa purinoceptor
mediating contraction of the rat femoral artery, rabbit ear artery and rabbit portal vein
can also be desensitized by a,B-MeATP (Burnstock and Kennedy, 1985).
An experiment was designed to investigate possibility of desensitization to ce,B-
MeATP in rat prepiriform cortex.
The seizure responses produced by a,13 -MeATP
(48nmo1/0.48/41) applied 30 min after the initial injection was of a similar magnitude to
the initial effect (Figure 8). Although a, B-MeATP is known to strongly desensitize the
P2, purinoceptor (Burnstock and Kennedy, 1985), the lack of evidence for desensitization
41
to a,13-MeATP in PPC neurones following repeated application may indicate that PPC
neurones possess P2x receptors which are distinct from peripheral P2x sites.
4. DISCUSSION
The results of this study provide the first demonstration of a purine nucleotide
induced convulsant effect following focal injection in rat brain. Although numerous reports
have documented excitatory actions of ATP on neural and smooth muscle preparations,
it has been suggested that in some cases the action of adenine nucleotides is not a direct
action on the cell membrane, but rather due to an alteration of the calcium concentration
by chelation (Falck, 1956; Bernies et al., 1969; Smith and Martell 1975a). It has also
been shown that the affinity of ATP for Ca'
(stability constant log K = 3.7) is
comparable to that of citric acid (log K = 3.2 ) and somewhat less than that of
pyrophosphoric acid (log K = 3.7) (Galindo et al., 1969; Curtis et al., 1960; Bjerrum
et al., 1957, Smith and Martell, 1975). Dorsal horn interneurons and cuneate cells were
strongly excited by ATP, pyrophosphoric acid and citric acid, but weak chelators were
ineffective. Thus Ca2+ chelation may be responsible for the excitatory for the excitatory
effects of ATP in these preparations (Galindo et al., 1969; Curtis et al., 1960). The
underlying mechanism of the excitatory effect of calcium chelators has been explained
by both a divalent gating-particle theory and a surface-potential theory (Hi lle, 1984).
Based on these observations it was necessary to investigate whether the ATP
analog-induced convulsant effects were due to the specific action at P2 purinoceptors or
due to a chelation of extracellular Ca'. Two calcium chelators, sodium pyrophosphate
42
and sodium citrate at dose of 24 nmol / 0.12/11 were focally injected into rat prepiriform
cortex. These two compounds induced convulsant responses in the rat PPC. The effects
of these two compounds on behavioral seizures are summarized in Table 2. We also
observed that P2 agonists (B,T-methylene-L-ATP, B,T-methylene-ATP and a,B-methylene-
ATP), a P2y agonist (ADPBS) and an ectonucleotidase inhibitor (a,B-methylene-ADP) all
elicited convulsant responses with similar potencies in rat PPC when injected in an ACSF
vehicle. These rank order potencies were not consonant with either a P2,, or a P2, receptor
profile (Table 1). Moreover, a P2 receptor antagonist, suramin, elicited a much more
potent convulsant response than all of the P2 receptor agonists. The requirement for small
volumes in microinjection studies dictated that drugs solutions be used in high
concentration (8.3mM
100mM).
Therefore, it was tentatively concluded that the
convulsant responses observed were due to calcium chelation.
These experiments were followed by application of a series of ATP analogues
focally injected into rat PPC using an equimolar CaC12 vehicle. Of all the nucleotides
tested with the CaC12 vehicle, the most potent agonist was B,T-methylene-L-ATP (ED50
± SE: 4.32 ± 1.38 nmol), while the potencies of a,B-methylene-ATP (ED50 ± SE: 15.85
± 1.29 nmol) and ATP (ED50 > 100mM) were less (Table 4). Moreover, in contrast
to previous results, ADPBS, a,B-methylene-ADP, and UTP did not elicit any convulsant
effects following unilateral injection of these compounds into the rat PPC (Table 3, Figure
2). B,T-methylene-L-ATP has been considered to be a potent and extremely selective P2,
receptor agonist, being devoid of activity at other P2 receptor subtypes (Hourani et al. ,
1985; Tatham et al. , 1988). This compound is completely resistant to dephosphorylation
43
and, in the absence of a D-adenosine moiety, eliminates the possibility of generation of
L-adenosine with activity at the Al receptor (Williams et al., 1990). Both the rank order
of potency (13,T-methylene-L-ATP > «,13-methylene-ATP > ATP) and the lack of a
convulsant effect of UTP, ADP13S and «,13-methylene-ADP suggest that the convulsant
effects of these P2 agonists were mediated by a P2,, purinoceptor in the rat prepiriform
cortex. Although there is evidence that UTP binds to a P2,, receptor in rat urinary bladder,
the affinity of UTP as a displacer of [311]«13MeATP binding to urinary bladder smooth
muscle homogenates is lower than that of ATP (Bo and Burnstock, 1993).
To further characterize the pharmacology of the ATP receptor mediating the
convulsant responses in the PPC, a P2 receptor antagonist, suramin, was used. Suramin
is a polysulphonic heterocyclic compound. The ability of drugs with sulfate and sulfite
groups to form complexes with tissue Ca2+ represents a potential confounding variable
in pharmacological studies employing microinjections (LaBella et al., 1979; Smith and
Martell, 1975c).
To eliminate the role of Ca' chelation following suramin injection,
a CaC12 vehicle was employed (CaC12 : suramin molar ratio, 3:1).
When focally
administered in a CaC12 vehicle, suramin did not evoke seizures administered alone.
Seizures induced by B,I1-methylene-L-ATP were inhibited by prior injection of suramin.
A suramin dose of 1 nmol produced a parallel rightward shift of theiLT-methylene-L-ATP
dose response curve, consistent with the involvement of a P2 receptor in the observed
response (Figure 3).
In contrast to other ATP analogs, 2-methylthio-ATP was found to have a potent
anticonvulsant effect against seizures induced by bicuculline methiodide in the prepiriform
44
cortex (Table 5, Figure 4). 2-Methylthio-ATP is a metabolically unstable P2y agonist
which is degraded by ectonucleotidase at same rate as ATP in guinea pig urinary bladder
(Welford et al., 1987).
In order to investigate the underlying mechanisms of the
anticonvulsant effect, the metabolite of 2-methylthio-ATP, 2-methylthio-adenosine, was
tested by focal injection of this compound into the prepiriform cortex. As expected, 2­
methylthio-adenosine (ED50 = 4.1 nmol) exerted an anticonvulsant effect against seizures
induce by bicuculline methiodide in this brain area.
Moreover, the potency of 2­
methylthio-adenosine was similar to that of 2-methylthio-ATP (ED50 = 5.32 nmol) (Table
6). These results suggest that the anticonvulsant effect of 2-methylthio-ATP may be due
to its rapid breakdown to 2-methylthio-adenosine which in turn may activate adenosine
Al receptors and produce an anticonvulsant effect. To further establish the role of
adenosine Al receptors in the anticonvulsant actions of 2-methylthio adenosine, the affinity
of 2-methylthio-adenosine for the Al receptor was determined. Equilibrium competition
analysis was performed, using [3H]DPCPX as ligand for the Al receptor. The rank order
of potency of an array of adenosine analogs including 2-methylthio-adenosine to brain
membrane was significantly correlated with their respective potencies as anticonvulsants.
(r = 0.91).
Previous studies in various brain preparations have shown that ATP produces an
initial excitation followed by depression (Phillis and Wu, 1981), or has an excitatory effect
only in the presence of the P1 receptor antagonist, DPCPX (Tschopl et al., 1992). The
degradation of ATP to adenosine in brain slice preparations may be responsible for these
observations inasmuch as adenosine is an inhibitory neuromodulator in the CNS and has
45
anticonvulsant properties when injected focally into the prepiriform cortex (Murray et al.,
1992; Franklin et al. , 1988; Zhang et al. , 1992). Therefore, the lower order of potent
of ATP as a convulsant in the rat prepiriform cortex may be related to degradation to
adenosine which in turn would activate the Al receptor (Table 3). To confirm this
hypothesis, the co-administration of ATP with a P,- antagonist, 8-pSPT, was performed
in the rat prepiriform cortex. The coadministration of 8-pSPT significantly increased the
convulsant response to a 100 nmol dose of ATP (P < 0.05; Figure 7 and Table 7).
Desensitization of P2x-purinoceptors by prolonged application of a,13-methylene-ATP
has been used to identify P2x-purinoceptor-mediated components of neurogenic
vasoconstriction (Von Kugelgen et al. , 1985). In the rat and frog dorsal root ganglion
neuron, fast desensitization to ATP-activated current was shown by application of high
concentration of ATP (Bean, 1992a). In contrast to the P2x receptor of both vasculature
smooth muscle and sensory neurons, we were unable to demonstrate reduced convulsant
response to a,B-methylene-ATP (48 nmol /0.48 AD when applied 30 min following the
initial injection.
In summary, the rat prepiriform cortex appears to be a neuroanatomical site which
displays sensitivity to the convulsant effect of ATP analogs. Based on the evidence of
a rank order of potency of B,I'- methylene -L -ATP > a,13-methylene-ATP > > ATP and
the apparently competitive antagonism of suramin, these results suggest the involvement
of a P2 like receptor in the observed behavioral responses to ATP analogs. Based on the
recent cloning reports of P2x receptors, it appears that the PC12 P2x is expressed at low
levels in the brain, whereas the vas deferens P2x receptor is not present in brain. The
46
identification of additional Pa receptors in neuronal tissue is likely to follow in the near
future, inasmuch as the pharmacologic profile of the PC12 P2 receptor is not consistent
with that observed in the medial habenula or locus coeruleus (Edwards et al., 1992;
Tschopl et al., 1992)
47
Table 1
Seizures induced by focal injection of suramin, 13,1'-methylene-L-ATP, a,13­
methylene -ATP, 13,r-methylene-ATP, ADPBS, AOPCP, UTP and ATP in rat
prepiriform cortex (ACSF vehicle)
Nucleotide
Distribution of Seizure Score`
(nmol/.141)
(n)b
0
2
4
5
4
1
6
2
(5)
(9)
2 (3)
4 (4)
4.7
5.0
(5)
(4)
(4)
0.6
3.0
4.3
(5)
(11)
(4)
1.8
3.1
1
3
70c
Mean
seizure
score
Suramin
0.1
0.3
1.0
1
1
2.4
13,r-Me-L-ATP
1.0
3
2.4
7.2
a,B-Me-ATP
2.4
7.2
24.0
1
1
1
1
1
1
1
1
1
4
3
1
3
3
3
2
2
0.8
16
1.4
29
94
100
12
60
85
4.8
36
62
96
13,r-Me-ATP
2.4
7.2
24.0
3
5
2.4
7.2
24.0
4
1
3
2
1
1
5
1
3
(5)
(10)
(4)
1.4
28
2.5
4.8
50
(5)
(5)
(5)
1.0
2.0
4.6
20
40
92
2.1
43
3.0
4.6
60
92
1.7
34
48
96
ADPBS
2
3
AOPCP
7.2
24.0
48.0
4
2
24.0
100.0
300.0
4
30.0
5
3
1
2
2
2
3
(8)
(6)
(5)
UTP
1
2
1
1
(6)
(5)
(6)
2.4
4.7
(5)
(5)
4 (5)
0.0
2.2
4.6
1
1
2
4
1
1
93
ATP
100.0
300.0(0.5/11)
2
1
1
0
44
92
48
Nucleotides were microinjected at the indicated doses unilaterally into the
prepiriform cortex as described in the text (ACSF vehicle). Animals were then observed
over a 30 min period and the mean highest seizure score attained for each group of animals
for each treatment was determined.
a.
b.
c.
Below each seizure score, the number of rats reaching that score is shown.
n indicates number of animals in each group
Percent of maximum seizure response for groups of animals receiving doses of
each nucleotide compared to control (BMI). The mean seizure score of control
responses was score 5.
49
Table 2
Sodium citrate and sodium pyrophosphate induced seizures in rat prepiriform
cortex.
Compounds
Distribution of Seizure Scores
0
1
1
2
2
3
4
5 (n)b
Mean
seizure
score
1
1 (5)
3.4
68
6 (6)
5.0
100
%
24nmol/.12/11
Sodium
citrate
Sodium
pyrophosphate
Sodium citrate and sodium pyrophosphate were microinjected at the indicated dose
unilaterally into the prepiriform cortex as described in the text. Animals were then
observed over a 30 min period and the mean highest seizure score attained for each
group of animals for each treatment was determined.
a.
b.
c.
Below each seizure score, the number of rats reaching that score is shown.
n indicates number of animals in each group.
Percent of maximum seizure response for groups of animals receiving doses of
each nucleotide compared to control (BMI). The mean seizure score of control
responses was score 5.
50
Table 3
Seizures induced by focal injection of 13,r-methylene-L-ATP, ot,11-methylene-ATP,
ATP, ADPBS, AOPCP and UTP in rat prepiriform cortex (CaC12 vehicle)
Nucleotide
(nmol)
Distribution
0
1
2
of
Seizure Score°
3
4
5
Mean
seizure
score
%c
(n)"
B,r-Me-L-ATP
2.4
3
7.2
2
24.0
(4)
1.0
20
3
(5)
3.0
60
8
(10)
4.2
84
1
(4)
4.3
85
(3)
0.0
0
(6)
1.3
27
1
1
1
36.0
3
a,B-Me-ATP
2.4
3
7.2
3
2
24.0
2
1
1
2
(6)
2.7
53
2
1
(2)
4.3
87
(4)
0.3
5
5
(5)
0.0
0
4
(4)
0.0
0
4
(4)
0.0
0
48.0
ATP
100.0
1
3
1
ADPBS
24.0
AOPCP
48.0
UTP
100.0
Nucleotides were microinjected at the indicated doses unilaterally into the
prepiriform cortex as described in the text (CaC12 vehicle). Animals were then observed
over a 30 min period and the mean highest seizure score attained for each group of
animals for each treatment was determined.
a.
b.
c.
Below each seizure score, the number of tats reaching that score is shown.
n indicates number of animals in each group.
Percent of maximum seizure response for groups of animals receiving doses of
each nucleotide compared to control (BMI). The mean seizure score of control
responses was score 5.
51
Table 4
Potencies of nucleotides as convulsants in rat prepiriform cortex.
Nucleotides
B, T -Me -L -ATP
a,B-MeATP
ATP
ED50 ± S.E. (nmol)
4.3 ± 1.38
15.9 ± 1.29
> 100.0
ED50 in this table represents the dose of intracerebrally administered
nucleotides required to elicit a 50 % maximum seizure response in the prepiriform
cortex. S.E. indicates standard error. The values in this table were calculated
from data in Table 3.
52
Table 5
Anticonvulsant effect of 2-methylthio-ATP and 2-methylthio-adenosine against
seizures induced by bicuculline methiodide in rat prepiriform cortex.
Treatment
0
Drug
Distribution of seizure scores
1
2
3
4
5
Dose
(nmol)
Control
2MeSATP
1.0
Control
2MeSATP
2.4
Control
2MeSATP
Mean
%`
seizure
score protection
4
1
3
2
2
4
1
1
5
7.2
3
2
1
2MeSATP 24.0
4
Control
3
4
Control
2MeSAdo
1.25
Control
2MeSAdo
2.5
1
Control
2MeSAdo
5.0
3
Control
2MeSAdo 10.0
5
Control
2MeSAdo 15.0
5
a.
b.
c.
(n)b
(4)
(4)
5.0
4.7
6
(6)
(6)
4.7
3.8
19
(6)
(6)
4.8
2.1
57*
(4)
(4)
5.0
0
100*
(5)
(5)
4.8
4.4
8
1
4
3
2
1
4
1
3
(5)
(5)
4.8
3.8
21
5
2
(5)
(5)
5.0
2.0
60
6
(6)
(6)
5.0
0.7
86*
(5)
(5)
5.0
1
5
0
100*
Below each seizure score, teh number of rats raching that score is shown.
n indicates number of animals in each group.
Percent protection is given as the percent maximum anticonvulsant response: the
percent reduction in the mean seizure score of animals pretreated with the
indicated dose of 2-methylthio-ATP or 2-methylthio-adenosine compared to
control (BMI). Statistical lly significant differences between control and treated
responses were assessed by one tailed Rank Sum test: * < 0.05
53
Table 6
Anticonvulsant potency of 2-methylthio-ATP and 2-methylthio-adenosine against
seizures indcued by bicuculline methiodide in rat prepiriform cortex.
Compounds
ED50 + S.E. (nmol)
95% C.L. (nmol)
2MeSATP
2MeSAdo
5.32 + 1.14
4.09 + 1.06
3.04
3.35
9.28
4.98
ED50 values in this table represents the dose of 2-methylthio-ATP and 2­
methylthio-adenosine required to produce a 50 % maximum anticonvulsant effect against
seizures induced by BMI in the prepiriform cortex. S.E. indicates standard error.
The values in this table were calculated from data in Table 5.
54
Table 7
Effect of pretreatment with the P1 receptor antagonist, 8- (p- sulfophenyl)- theophylline
(8-pSPT), on ATP-induced convulsant response in rat prepiriform cortex.
Treatment
Distribution of seizure score°
0
1
2
3
4
5
(a)"
Mean
seizure
score
(4)
0.00
(4)
0.20°
(4)
0.00'
(4)
2.17``'"
(CaC12, 100mM/1.0/11)
CaC12
100.0nmol
4
ATP
100.0nmol
3
8-pSPT 1.0nmol
ATP
4
100.0nmol
+8-pSPT 1.0nmol
a.
b.
c.
d.
1
2
1
2
1
Below each seizure score, the number of rats reaching that socre is shown.
n indicates number of animals in each group.
Mean seizure score for groups of animals receiving doses of each compound.
Mean seizure score for groups of animals receiving doses of each compound by
co-injection with 8-pSPT into the prepiriform cortex. Statistically significant
difference from control response (* P < 0.05, one tail Rank Sum test)
55
Table 8
Potencies of adenosine analogs in protection against BMI-induced seizures in the rat
PPC.
ED50
NECA
CHA
CPA
RPIA
2CLA
SPIA
CGS21680
2MeSAdo
Adenosine
4.0
7.9
17.9
20.3
32.2
208.0
605.3
4073.8
47863.0
±
SE (pmol)
3.0
0.07
0.8
7.0
0.3
0.1
46.7
106.0
8400.0
ED50 values were determined by nonlinear regression analysis as described under
"methods". ED50 values for NECA, CHA, CPA, RPIA, 2C1A, SPIA, CGS21680,
adenosine are from Zhang et al. (1992)
Table 9
The affinity of adenosine analogs as inhibitors of [3H]DPCPX binding to rat brain membranes.
Compound
two-site
F
site 1
%
RPIA
CHA
CPA
2CLA
NECA
SPIA
2MeSAdo
CGS21680
Ado
53.0
51.4
47.6
47.0
53.4
35.1
21.9
31.5
IC50(nM)
1.1± 0.1
1.7+ 0.1
2.0± 0.1
7.3+ 0.0
16.4± 0.8
127.2+ 16.0
953.1± 27.0
4556.0± 178.2
P
site 2
%
47.0
48.6
52.4
53.0
46.6
64.1
70.9
64.9
100.0
IC50(nM)
49± 1.1
49± 0.8
63± 0.9
550± 5.4
870± 12.0
11660± 866.0
16920± 1456.0
459400± 43167.6
114600± 9517.0
80.7
99.6
76.3
176.9
141.8
53.9
51.5
75.2
0.0001
0.0001
0.0002
0.0001
0.0001
0.0001
0.0001
0.0001
IC50 valuces and percentages of binding sites in the high- and low-affinity states of the At adenosine receptor
were derived by nonlinear regression analysis of the data.
57
Figure 1
Structures of ATP agonist and antagonist used in this study
TP
OK OK
OH 04{
OK OK
CK
oK
0
II
0
oK OK
fir-Me-L-ATP
0
0
1
_o
_o
o
t
_0
OK OK
Suramin
58
Figure 2
Seizures induced by focal injection of 13,r-methylene-ATP, «,13-methylene-ATP,
ADM, AOPCP, ATP, UTP in rat prepiriform cortex
100
/1,11MeLATP
O
L)
(r)
0 a,f3--MeATP
80
60
U)
40
M
20
ADPgS AOPCP
V
1
LOG DOSE (nmol)
ATP
th
2
U P
59
Figure 3
Suramin antagonism of 13,11-methylene-L-ATP induced seizures in rat Prepiriform
cortex.
0.6
0.3
F-T1
0.0
0
1-1
0.3
0.6
LOG DOSE (nmol)
60
Figure 4
Protection by 2-methylthio-ATP and 2-methylthio-adenosine against bicucullin-induced
seizures in rat prepiriform cortex
120
100
80
60
40
20
0
1
0
1
LOG DOSE (nmol)
2
61
Figure 5
Agonist competition of [3H]DPCPX binding in a rat brain cerebral cortical membrane
preparation.
100
z
80
riq
t0
X
P--i
i24
60
1,4
RPIA
(-)
CHA
CPA
0
crp
40.
2CLA
NECA
SPIA
20
2MeSAdo
CGS21680
adenosine
0
10
9
8
7
6
-LOG [AGONIST] (M)
5
4
62
Figure 6
Correlation between potencies of adenosine agonists as anticonvulsants and their
affinity for adenosine Al receptors
5
4
cp
3
10
It
tc
0
2
1
0
0
1
2
3
4
Log ED50 (pmol)
ANTICONVULSANT POTENCY
5
63
Figure 7
Influence of the P1 receptor antagonist 8- (p- sulfophenyl)- theophylline (8-pSPT) on
ATP responses in rat prepiriform cortex
3
O
U
cn
**
2
1
0
ATP
100nmol
ATP(100nmol)+
1.0nmol 8pSPT(1.0n.mol)
BpSPT
64
Figure 8
Comparison of «,13-methylene-ATP-induced seizure responses on repeated application
to rat prepiriform cortex.
5
4
3
2
1
0
agMeATP
aieMeATP
48 nmo1/0.48,u,1
48 nmo1/0.48,a1
initial injection
30 min after
initial injection
65
5. BIBLIOGRAPHY
Abraham, E.H., Prat, A.G., Gerweek, L., Seneveatne, T. Arceci, R.J., Kramer, R.,
Guidotti, G., and Cantiello, H.F. (1993) The multidrug resistance (mdrl) gene product
functions as an ATP channel. Biochem. 90,312-316
Bao, J.X. and Stjarne, L. (1993) Dual contractile effects of ATP released by field
stimulation revealed by effects of a,13-methylene ATP and suramin in rat tail artery. Br.
J. Pharmacol. 110,1421-1428
Barnard, E.A. (1992) Receptor classes and the transmitter-gated ion channels. Trends
Biochem. Sci. 17,368-374
Barnard, E.A., Burnstock, G. and Webb, T.E. (1994) G protein-coupled receptors for
ATP and other nucleotides: a new receptor family Trends Pharmacol. 15,67-70
Bean, B.P. (1990a)
ATP-activated channels in rat and bullfrog sensory neurons:
concentration dependence and kinetics. J.Neurosci. 10,1-10
Bean, B.P. (1992) Pharmacology and electrophysiology of ATP-activated ion channels
Trends Pharmacol. 13,87-90
Bean, B.P., William, C.A. and Ceelen, P.W. (1990b) ATP-activated channels in rat and
bullfrog sensory neurons: current-voltage relation and single-channel behavior J. Neurosci.
10,11-19
Benham, C.D. and Tsien, R.W. (1987) A novel receptor-operated Ca'-permeable
channel activated by ATP in smooth muscle. Nature 328,275-278
Bennett, M.R. and Rogers, D.C.(1967) A study of the innervation of the taenia coli
J.Cell.Biol. 33,573-596
Bernie, K.H., Pletscher, A. and DaPrada, M. (1969) Metal-dependent aggregation of
biogenic amines: a hypothesis for their storage and release. Nature 224,281-283
Beyer, E.C. and Steiberg, T.H. (1991) Evidence that the gap junction protein
connexin-43 is the ATP-induced pore of mouse macrophages. J.Biol.Chem.
266,7971-7974
Bjerrum, J., Schwarzenbach, G. and Sillen, L.G. (1957). Stability constants. Part 1.
organic Ligands. Spec.Publs Chem.Soc. pp 6.
Bo, X. and Burnstock, G. (1993) Triphosphate, the key structure of the ATP molecule
responsible for interaction with Pa-purinoceptors. Gen.Pharmacol. 24,637-640
66
Bo, X., Simon, J., Burnstock, G. and Barnard, E.A. (1992) Solubilization and molecular
size determination of the P2,, -purinoceptor from rat vas deferens. J.Biol. Chem., 267(25)
17581-17587
Boarder, M.R. and Richard, J.A. (1992). Neuronal "nucleotide" receptor linked to
phospholipase C and phospholipase D? stimulation of PC12 cells by ATP analogs and
UTP. Mol.Phannacol. 41:561-568.
Brake, A.J. , Wageboch, M.J. and Julius, D. (1994) New structural motif for ligand-gated
ion channels defined by an ionotropic ATP receptor. Nature 317,519-523
Buisman H. P. Steinberg T. H., Fischbarg J., Silverstein S. C., Vogelzang S. A., Ince
C., Ypey, P. and Leijh C. j., (1988) Extracellular ATP induces a large nonselective
conductance in macrophage plasma membranes. Proc.Natl.Acad.Sci. USA 85,7988-7992.
Burnstock, G. (1972) Purinergic nerves. Pharmac.Rev. 24, 509-581.
Burnstock, G. (1976) Do some nerve cells release more than one transmitter? Neurosci.
1, 239-248.
Burnstock, G. (1978) A basic for distinguishing two types of purinergic receptor. In
Straub RW, Bolis L (eds): Cell membrane receptors for drugs and hormones: a multi­
disciplinary approach. Raven Press, New York, pp 107-118.
Burnstock, G. (1990) Overview purinergic mechanisms. Ann.N.Y. Acad.Sci. 603,1-17
Burnstock, G, Campbell, G. and Bennett, M. (1964) Innervation of the guinea-pig taenia
coli: Are there intrinsic inhibitory nerves which are distincted from sympathetic nerves
in the gut? Br.J.Pharmacol. 40,668-688.
Burnstock, G, Campbell, G., Bennett, M. and Holman, M.H. (1963) The effects of drugs
on the transmission of inhibition from autonomic nerves to the smooth muscle of the
guinea-pig taenia coli. Biochem. Pharmacol. 12:suppl., 134
Burnstock, G. Cusack, N.J., Hills, J.M. , Mackenzie, I. and Meghji, P. (1983) Studies
on the stereoselectivity of the P2-purinoceptor. Br.J.Phannac. 79,907-913
Burnstock, G. Dunsday, B. and Smythe, A. (1972) Atropine resistant excitation of the
urinary bladder the possibility of transmission via nerves releasing a purine nucleotide,
Br.J.Phannacol. Chemother. 44,451-461
Burnstock, G. and Iwayama, T (1971) Five structural identification of autonomic nerves
and their relation to smooth muscle.
In symposium on histochemistry of nervous
transmission. Progr.Brain Res. 34,389-404
67
Burnstock, G. and Kennedy, C.(1985). Is there a basis for
distinguishing two types of P2-purinoceptor? Gen.Pharmaco. 16,433-440.
Charest R., Pepic V, Extom J.H. Blackmore P.F. (1985): Stimulation of inositol
triphosphate formation in hepatocytes by vasopressin, adrenaline and angiotensin II and
its relationship to changes in cytosolic free Ca2+. Biochem.J. 227,79-89
Cloues, R., Susan, J. and Brown D. (1993) Zn2+ potentiates ATP-activated currents in
rat sympathetic neurons. Pfugers Arch. 424,152-158
Connolly, G.P. (1994) Evidence from desensitization studies for distinct receptors for ATP
and UTP on the rat superior cervical ganglion Br.J.Pharmacol. 112,357-359
Connolly, G.P., Harrison, P.J. and Stone, T.W. (1993) Action of purine and pyrimidine
nucleotides on the rat superior cervical ganglion. Br.J.Pharmacol. 110,1297-1304
Cooper, C.L., Morris, A.J. and Harden, T.K. (1989) Guanine nucleotide-sensitive
interaction of a radiolabeled agonist with a phospholipase C-linked P-purinergic receptor.
J.Biol.Chem. 264,6202-6206
Curtis, D.R., Perrin, D.D. and Watkins, J.C. (1960) The excitation of Spinal neurones
by the ionophoretic application of agents which chelate calcium. J.Neurochem. Vol 6,
pp1-20
Cusack, N.J. and Hourani, M.O. (1991)
Structure-activity relationships of ATP
receptors. Role of adenosine and adenine nucleotides in the biological system. (Shoichi
Imai and Mikio Nakazawa, eds. 17-25
Cusack, N.J., Hourani, M.O., Loizou, G.D. and Welford, L.A. (1987)
Pharmacological effects of isopolar phosphonate analogues of ATP on P2-purinoceptors
in guinea pig taenia coli and urinary bladder. Br.J. Pharmacol. 90,791-795
Dainty, I.A., Steptoe, J.E. , O'Connor, S.E. and Leff, P. (1990) Is 2-methylthio-ATP
an appropriate tool for the identification of P2y-purinoceptor? Br.J.Pharmacol. 101, 507P
Davidson, J.S. , Wakefield, I.K., Sohnius, U. and Millar, R.P. (1990). A novel extracel­
lular nucleotide receptor coupled to phosphoinositidase-C in pituitar cells. Endocrinology
126, 80-87.
Devore, J. and Peck, R. (1986) The exploration and analysis of data. In Statistics, pp393,
West Publishing Company, St.Paul
68
Drury, A. N. and Szent-Gyorgyi A. (1929). The physiological activity of adenine
compounds with special reference to their action upon the mammalian heart. J. Physiol.
72, 213-237.
Dubyak, G. R. and Moatassim C. E. (1993) Signal transduction via P2-purinergic
receptors for extracellular ATP and other nucleotides. Am.J.Physiol. 265, C577-606
Dunn P.M. and Blake ley, A.G.H. (1988) Suramin: a reversible P2-purinoceptor
antagonist in the mouse was deferens. Br.J.Pharmacol. 93,243-245
Edwards, F. and Gibb, A. (1992) ATP receptor-mediated synaptic currents in the central
nervous system. Nature 359,144-147.
Evans, R.L., Derkach, V. and Surprenant, A. (1992) ATP mediates fast synaptic
transmission in mammalian neurons. Nature 357,503-505
Ewald, D.A. (1976) Potentiation of postjunctional cholinergic sensitivity of rat diaphragm
muscle by high-energy phosphate adenine nucleotides J.Membr.Biol. 29,47-65
Falck, G. (1956) Nonspecificity of ATP-contraction of living muscle. Science 123,632­
633
Fedan, J.S., Hogaboom, K. G. Westfall, D.P. and O'Donnell, J.P. (1982) Comparison
of contractions of smooth muscle of the guinea-pig vas deferens induced by ATP and
related nucleotides. Eur.J.Pharmacol. 81, 193-204
Feiber, L.A. and Adams, D.J. (1991). Adenosine triphosphate-evoked currents in cultured
neurones dissociated from rat parasympathetic cardiac ganglia. J. Physiol. 434,239-256.
Franklin, P.H., Tripp, E.D., Zhang, G., Gale, K. and Murray, T.F. (1988). Adenosine
receptor activation blocks seizures induced by bicuculline methiodide-induced seizures
in rat prepiriform cortex. J.Phannacol.Exp.Ther. 251,1229-1236.
Friel D.D. (1988): An ATP-sensitive conductance in single smooth muscle cells from the
rat vas deferens J. Physiol. 401:361-380.
Galindo A. Krnjevic K. and Schwartz S. (1967) Micro-iontophoretic studies on neurones
in the cuneate nucleus. J. Physiol. 192,359-377
Gordon, J.L. (1986). Extracellular ATP: effects, sources and fate.
Biochem.J. 233,309-319.
69
Greco, N., Tandon, N.N., Jacjson B.W. and Jamieson, G.A. (1992) Low structural
specificity for nucleoside triphosphates as antagonists of ADP-induced platelet activation.
J. Biol. Chem. 267,2966-2970.
Harden, T.K., Stephens, L., Hawkins, P.T. and Downes, C.P. (1987) Turkey
erythrocyte membranes as a model for regulation of phospholipase C by guanine
nucleotides. J. Biol . chem. 262, 9057-9061
Hedgvust, P. and Fredholm, B.B. (1976). Effects of adenosine on adrenergic
neurotransmission; prejunctional inhibition and postjunctional enhancement. Naunyn
schmiedeberg's arch. pharmol. 293,217-223.
Hille B. (1984) External Ca2+ ions shift voltage-dependent gating. Ionic channels of
excitable membranes. pp.316-320
Holton, F. A. and Holton, P. (1954) The capillary dilator substances in dry powders of
spinal roots; a possible role of adenosine triphosphate in chemical transmission
from nerve endings. J.Physiol. 126,124-140
Hourani, S.M.O. and Chown, J.A. (1989) The effects of some possible inhibitors of
ectonucleotidases on the breakdown ad pharmacological effects of ATP in the guinea-pig
urinary bladder. Gen. Pharmacol. 20,413-416
Hourani, S.M.O., Cusack, N.J. and Welford, L.A. (1985) L-AMP-PCP, an ATP receptor
agonist in guinea-pig bladder, is inactive on taenia coli. Eur.J.Pharmacol. 108, 197-200
Hourani, S.M.O., Loizou, G.D. and Cusack, N.J. (1986) Pharmacological effects of
L-AMP-PCP on ATP receptors in smooth muscle. Eur.J.Pharmacol. 131,99-103
Hoyle, C.H.V. and Burnstock, G. (1990) Suramin antagonizes responses to P2-pur­
inoceptor agonists and purinergic nerve stimulation in the guinea-pig bladder and taenia
coli. Br.J.Pharmacol. 99,617-621
Igusa, J. (1988) Adenosine 5'-triphosphate actviates acetylcholine receptor channels in
cultured Xenopus myotocal muscle cells. J. Physiol. 504,169-185
files, P. and Norenberg, W. (1993). Neuronal ATP receptors and their mechanism of
action. Trends Phannacol.Sci. 14,50-54.
Inoue, K. and Nakazawa, K. (1992) ATP receptor-operated Ca2+ influx and
catecholamine release from neuronal cells. News Physio.Sci. 7,56-59
Jahr, C.E. and Jessell, T.M. (1983). ATP excites a subpopulation of rat dorsal horn
neurones. Nature 304,730-733.
70
Kolb, H-A and Wake lam, M.J.O. (1983) Transmitter-like action of ATP on patched
membranes of cultured myoblasts and myotubes. Nature 306,621-623
Krishtal, 0.A., Marchenko, S.M. and Obukhov, A.G. (1988). Cationic channels activated
by extracellular ATP in rat sensory neurons. Neurosci. 27,995-1000.
Labella, F.S., Havlicek, V. and Pinsky C.(1979) Opaite-like excitatory effects of steroid
sulfates and calcium-complexing agents given cerebroventricularly. Brain Research
160,295-305
Li, C., Robert W., Oeoples S.L. and Weight, F.F. (1993) Zn2+ potentiates exvitatory
action of ATP on mammalian neurons. Proc.Natl.Acad.Sci. 90,8264-8267
Linden, J. (1991) Structure and function of Al adenosine receptors FASEB J. 5,2668-2676
Lustig, K.D., Shiau, A.K., Brake, A.J. and Julius, D (1993) Expression cloning of an
ATP receptor from mouse neuroblastoma cells. Proc.Natl.Acad.Sci. 90(11) 5113-5117
Magurire and Satchell (1979) Inhibitory effects of adenosine nucleotide analogs on the
isolated guinea-pig taenia coli. J.Pharmacol.Exp.Ther. 195,540-548.
Mahaut-Smith, M.P., Sage, S.O. and Rink, T.J. (1990) Receptor-activated single channels
in intact human platelets. J.Bio. Chem. 265,10479-10483.
Mantegazza, P. and Naimzada, K.M. (1967a) Observations on an isolated preparation
of guinea pig urinary bladder stimulated through the hypogastric nerves.
Eur.J. Pharmacol. 1,396-401
Mantegazza, P. and Naimzada, K.M. (1967b) Responses of guinea-pig vas deferens
and urinary bladder to hypogastric nerve stimulation in vivo
Eur.J.Pharrnacol. 1.402-406
Mclaren, G.J. Lambrecht, G, Mutschler, E., Baumert, H.G., Sneddon, P. Kennedy C.
(1994) Investigation of the actions of PPADs, a novel P2x-purinoceptor antagonist, in the
guinea-pig isolated vas deferens. Br.J.Pharmacol. 111,913-917
Murray, T.F., Gale, K., Zhong, P. and Miller, L.P. (1992) Amino acid neurotransmitter
interactions in " area tempestas" : an epileptogenic trigger zone in the deep prepiriform
cortex. Neurotransmitters in Epilepsy pp. 229-234
Nakazawa, K., Fujimori, K. Takanaka, A. and Inoue, K. (1990a) reversible and
seletive antagomism by suramin of ATP activated inword current in PC12
phaeochromocytoma cells Br.J.Phannacol. 101,224-226
71
Nakazawa, K., Fujimori, K., Takanaka, A. and Inoue, K. (1990b) An ATP-activated
conductance in pheochromocytoma cells and its suppression by extracellular calcium.
J. Physiol. 428,257-272
Needham, L., Cusack, N.J., Pearson, J.D. and Gordon, J.L.(1987). Characteristics of
the P2y purinoceptor that mediates prostacycline production by pig aortic endothelial cells.
Eur.J. Pharmacol. 134,199-209.
O'Connor, S.E. (1991). Further subclassification of ATP receptors based on agonist
studies. Trends Pharamaco.Sci. 12,137-141.
O'Connor, S.E., Wood, B.E. and Leff, P. (1990) Relative agonist potencies of methylene
substitute ATP analogues for P2x-purinoceptors in the rabbit ear artery. Br.J . Pharmacol.
100, 341P.
Paddle, B.M. and Burnstock, G. (1974) Blood vessels 11:110-119
Pearson, J.D. and Carter, T.D. (1990) Effects of extracellular ATP on the release of
vasoactive mediators from endothelium. Biological actions of extracellular ATP. ed. by
Dubyak, G. R. and Fedan, J. S. Ann. N. Y. Acad. Sci. 603,267-274.
Phillis, J.W. and Wu, P.H. (1981) The role of adenosine and its
necleotides in central synaptic transmission, Prog.Neurobiol. 16,187.
Piredda, S. and Gale, K. (1985) A crucial epileptogenic site in the deep prepiriform
cortex. Nature 317,623-625.
Rankumaar, V., Stiles, G.L. Beaven, M.A. and Ali, H. (1993) The A3 adenosine receptor
is the unique adenosine receptor which facilitates release of allergic mediators in mast
cells.J. Biol. Chem. 268,16887-16890
Richardsom, P.J. and Brown, S.J. (1987) ATP release from affinity-purified rat
cholinergic nerve terminals. J.Neurochem. 48, 622-630
Robinson, P.M. Melean, J.R. and Burnstock, G.(1971) Ultrastructure identification of
non-adrenergic inhibitory nerve fibers. J.Pharmacol.Exp.Ther. 179,149-160
Rozengurt,E., Heppel, L.A. and Friedberg, I. (1977) Effect of exogenous ATP on the
permeability properties of transformed cultures of mouse cell lines. J.Biol. Chem.
252,4584-4590
Satchell, D. G. and Maguire, M.H. (1975). Inhibitory effects of adenine nucleotide analogs
on the isolated guinea-pig taenia coli. J.Pharmacol.Exp.Ther. 195(3) 540-548.
72
Shen, K.Z. and North, R.A. (1993) Excitation of rat locus coeruleus neurons by
adenosine 5'-triphosphate: ionic mechanism and receptor characterization.
J. Neurosci. 13, 894-899
Silinsky, E.M. Gerzanich,V. and Vanner, S.M. (1992) ATP mediates excitatory
synaptic transmission in mammalian neurones. Br.J. Pharmacol. 106,762-763
Smith, R.M. and Martell, A.E. (1975a) Critical stability constant. vol.2 pp270
Smith, R.M. and Martell, A.E. (1975b) Critical stability constant. vol.2 pp283
Smith, R.M. and Martell, A.E. (1975c) Critical stability constant. vol.2 pp78­
79
Sternweis, P. and Smrcka, A.V. (1992) G-protein regulation of phospholipase C.
Trends Pharmacol.Sci. 17:502-506
Tatham, P.E.R., Cusack, N.J. and Gomperts, B.D. (1988) Characterization of ATP'
receptor that mediates permeabilitation of rat mast cells Eur.J. Pharmacol. 147,13-21
Tokimasa, T. and Akasu, T. (1990). ATP regulates muscarine-sensitive potassium
current in dissociated bull-frog primary afferent neurones. J.Physiol. 426,241-264.
Tschopl, M., Harms, L. Norenberg, W. and Il les, P. (1992) Excitatory effects of
adenosine 5'-triphosphate on rat locus coeruleus neurones Eur.J. Pharmacol. 213,71­
77
Valera, S., Hussy, N., Evans, R.J., Adaml, N., North, A., Surprenent A. and Nuell,
G. (1994) A new class of ligand-gated ion channel defined by P2x receptor for
extracellular ATP. Nature 371,516-519
Vizi, E.S., Sperlagh, B. and Baranyi, M. (1992) Evidence that ATP, released
from the postsynaptic site by noradrenaline, is guinea-pig vas deferens: cascade
transmission. Neurosci. 50,455-465
Von Kugelgen, I, and Starke, K. (1990). Evidence for two separate vasoconstrictionmediating nucleotide receptors, both distinct from the Pa- receptor, in rabbit basilar
artery: a receptor for pyrimidine nucleotides and a receptor for purine nucleotides
Naunyn-schmiedeberg's Arch. Pharmacol. 341,538-546.
Webb, T.E., Simon, J., Krishek, B.J., Bateson,T.G., Smart, B.F., King, B.F.,
Burnstock, G. and Barnard, E.A. (1993) Cloning and functional expression of a brain
G-protein coupled ATP receptor. FEBS lett. 324,219-225
73
Welford, L.A., Cusack, N.J. and Hourani, S.M.O. (1987) The structure-activity
relationships of ectonucleotidases and of excitatory P2-purinoceptors: evidence that
dephosphorylation of ATP analogues reduces pharmacological potency.
Eur.J. Pharmacol. 141,123-130
Williams, M. and Cusack, N. (1990) Neuromodulatory roles of purine nucleosides
and nucleotides : their receptors and ligands. Neurotransmissions VI (1) RBI Published
by research biochemicals incorporated.
Zhang, G., Franklin, P.H. and Murray, T.F. (1992). Manipulation of endogenous
adenosine in the rat prepiriform cortex modulates seizure susceptibility. J.
Pharmacol.Exp.Ther. 264(3) 1415-1424.
Ziganshin, A.U., Hoyle, C.H.V., Bo, X., Lambrecht, G., Mutschor, E., Baumert,
H.G. and Burnstock, G. (1993) PPADs selectively antagonizes P2x-purinoceptor­
mediated responses in the rabbit urinary bladder. Br.J. Pharmacol. 110,1491-1495
Ziganshin, A.U., Hoyle, C.H.V., Lambrecht, G., Mutschler, E., Baumert, H.G. and
Burnstock, G. (1994) Selective antagonism by PPADs at Pa-purinoceptors in rabbit
isolated blood vessels. Br.J. Pharmacol. 111,923-929
Zimmermann, H. (1994) Signalling via ATP in nervous system Trends Neurosci.
17,420-426