Opioid-coupled adenylyl cyclase and regulation of proenkephalin expression

MISSING IMAGE

Material Information

Title:
Opioid-coupled adenylyl cyclase and regulation of proenkephalin expression
Physical Description:
x, 125 leaves : ill. ; 29 cm.
Language:
English
Creator:
Marckel, Don Roger, 1962-
Publication Date:

Subjects

Subjects / Keywords:
Research   ( mesh )
Adenylate Cyclase   ( mesh )
Receptors, Opioid   ( mesh )
Enkephalins   ( nal )
Department of Pharmacology and Therapeutics thesis Ph.D   ( mesh )
Dissertations, Academic -- College of Medicine -- Department of Pharmacology and Therapeutics -- UF   ( mesh )
Genre:
bibliography   ( marcgt )
non-fiction   ( marcgt )

Notes

Thesis:
Thesis (Ph.D.)--University of Florida, 1990.
Bibliography:
Bibliography: leaves 110-124.
Statement of Responsibility:
by Don Roger Marckel Jr.
General Note:
Typescript.
General Note:
Vita.

Record Information

Source Institution:
University of Florida
Rights Management:
All applicable rights reserved by the source institution and holding location.
Resource Identifier:
aleph - 001581948
oclc - 25072923
notis - AHK5866
sobekcm - AA00004766_00001
System ID:
AA00004766:00001

Full Text









OPIOID-COUPLED ADENYLYL CYCLASE AND
REGULATION OF PROENKEPHALIN EXPRESSION













By
DON ROGER MARCKEL JR.














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

UNIVERSITY OF FLORIDA
1990















This dissertation is dedicated to my parents, Don and Viola, my sister,
Sonie and my friend, Beth.





*-














ACKNOWLEDGMENTS


I enjoyed my stay in Gainesville and have made many friends that I will
miss dearly. I will always remember Kevin Krajniak, Dave Sherry, Kelly
Standifer, Denise Bottiglieri, Lynne Fleming, Mary Pacheco, Chris Konkoy, Phil
Lograsso, Jeff Harris, Carolyn Bolden and John Olson. I am indebted to Bob
Raulli and Dave Kroll for their support through the good and bad times. Many
thanks to my "runner" in Gainesville, Jeff Lawrence, while I was in North
Carolina, without which I would not have made any of the dissertation
deadlines. I would also like to thank Chris Kalberg for reawakening my
competitive drive and helping me in my athletic accomplishments. I sincerely
thank my committee, Drs. Steven Childers, Stephen Baker, Fulton Crews, Colin
Sumners, and Thomas Rowe. My gratitude goes to Drs. Richard Lock and Paul
E. Kroeger for teaching me the molecular techniques utilized in these studies
and to Dr. Thomas Rowe for help in understanding molecular biology. Special
thanks are given to Debbie Otero for teaching me how to culture cell lines and
to Dr. Colin Sumners' laboratory for teaching me how to prepare primary
neuronal cultures. I also thank Carolyn Holcomb and Jeff Lipscomb for their
help on the C6 project and additional thanks to Jeff for being a great landlord.

Also, I wish to thank Tammy Sexton for her assistance in Steve Childers'
laboratory in North Carolina. I am especially indebted to Beth Harrell for her
support and understanding during the last eight months of finishing this
dissertation.














TABLE OF CONTENTS

DAGE
ACKNOWLEDGMENTS............................................................................ iii

LIST OF TABLES............................................................................... vi

LIST OF FIGURES...................................................................................... vii

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

CHAPTERS

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

General Introduction................................................................... 1
G-Proteins: Structure and Function......................................... 3
Receptor-Adenylyl Cyclase Coupling Mechanisms................ 6
Opioid-lnhibited Adenylyl Cyclase in Brain Membranes........ 9
Opioid-lnhibited Adenylyl Cyclase in Cell Culture................... 12
Opioid-Stimulation of Adenylyl Cyclase................................... 15
Opioid and Cyclic AMP Regulation of Proenkephalin mRNA. 16
Aims.......................................................................................... 20

2 MATERIALS AND METHODS........................................................ 22

Cell Culture............................................................................. 22
Cyclic AMP Assays in Intact Cells.................................. ...... 24
Adenylyl Cyclase Assay............................................................ 25
Analysis of RNA...................................................................... 26
Treatment of Cell Cultures with Drugs..................................... 27
Opioid Receptor Binding.......................................................... 28

3 OPIOID REGULATION OF ADENYLYL CYCLASE IN RAT
C6 GLIOMA CELLS............................................ .................... 29

Introduction.............................................................................. 29
Results......................................................................................... 31
Discussion.............................................. ................................. 44








4 OPIOID REGULATION OF ADENYLYL CYCLASE IN RAT
PRIMARY NEURONAL CULTURES......................... ............ 47

Introduction......................................... ........ ............................. 47
R esults.............................................................................................. 48
D iscussion....................................................................................... 56

5 OPIOID REGULATION OF PROENKEPHALIN EXPRESSION... 61

Introduction......................................... ........ ............................. 61
Results................................................ ......... ............................. 63
D iscussion....................................................................................... 93

6 SUMMARY AND CONCLUSIONS ................................................... 96

APPENDIX Opioid Dissociation Constants at Opioid Binding Sites. 109

REFERENC ES...................................................................................... 110

BIOGRAPHICAL SKETCH........................................................................ 125














LIST OF TABLES


page
3-1 Opioid receptor binding and adenylyl cyclase in
C6 glioma cells..................................................................... 32

4-1 Effect of opioid agonists on forskolin-stimulated
adenylyl cyclase activity in glial culture membranes............... 57

5-1 Densitometric quantitation for expression of proenkephalin
mRNA in forebrain neuronal cultures..................................... 66

5-2 Densitometric quantitation for time course of dynorphin A1-13
and isoproterenol regulation of proenkephalin mRNA levels
in forebrain neuronal cultures.................................................... 74

5-3 Densitometric quantitation for effect of naloxone on dynorphin
Al-13 regulation of proenkephalin mRNA levels in forebrain
neuronal cultures.................................................................. 76

5-4 Densitometric quantitation for effect of pertussis toxin on
dynorphin Al-13-induced changes in proenkephalin mRNA
levels in forebrain neuronal cultures......................................... 79

5-5 Densitometric quantitation for effect of dynorphin A1-13 on
forskolin- and dibutyryl cyclic AMP-induction of
proenkephalin mRNA levels in forebrain neuronal cultures.. 81

5-6 Densitometric quantitation for effect of dynorphin A1-13 on
proenkephalin mRNA levels in primary glial cultures............. 86

5-7 Densitometric quantitation for effect of D-Ala enk and naloxone
on isoproterenol-induction of proenkephalin mRNA levels
in C6 glioma cells.................................... ............................. 88














LIST OF FIGURES


3-1 Effect of D-Ala enk on adenylyl cyclase activity in C6
gliom a cells......................................................................................... 33

3-2 Effect of D-Ala enk on cyclic AMP accumulation in C6
glioma cells.................................. ................. ................................. 34

3-3 Effect of pertussis toxin on basal and D-Ala enk-regulated
cyclic AMP accumulation in C6 glioma cells................................ 36

3-4 Effect of PMA-pretreatment on basal and isoproterenol-
stimulated cyclic AMP accumulation in C6 glioma cells............. 37

3-5 Effect of PMA-pretreatment on opioid regulation of cyclic
AMP accumulation in C6 glioma cells.............................................. 38

3-6 Selectivity of opioid stimulation of cyclic AMP accumulation
in PMA-pretreated C6 glioma cells.................................... ........ 40

3-7 Comparison of D-Ala enk and DAGO to stimulate cyclic AMP
accumulation in C6 glioma cells........................................ ............ 41

3-8 Effect of naloxone on DAGO-stimulated cyclic AMP
accumulation in PMA-pretreated C6 glioma cells.......................... 42

3-9 Effect of naloxone on cyclic AMP accumulation in PMA-
pretreated C6 glioma cells.......................................................... 43

4-1 Effect of D-Ala enk and isoproterenol on cyclic AMP
accumulation in primary neuronal cultures from forebrain,
midbrain and hindbrain of neonatal rat..................................... 49

4-2 Effect of pertussis toxin on D-Ala enk- and isoproterenol-
regulated cyclic AMP accumulation in forebrain
neuronal cultures.................................................. ...................... 51

4-3 Effect of various opioid agonists on cyclic AMP accumulation
in forebrain neuronal cultures.................................................. 52








4-4 Effect of morphine on adenylyl cyclase activity in membranes
from forebrain neuronal cultures.................................................... 55

4-5 Effect of dynorphin A1-13 on adenylyl cyclase activity in
membranes forebrain neuronal cultures.................................. 57

5-1 Northern blot analysis of proenkephalin and alpha-tubulin
mRNA in primary neuronal cultures and rat brain........................ 64

5-2 Expression of proenkephalin mRNA in forebrain neuronal
cultures................................................................................ ................... 65

5-3 Time course of forskolin induction of proenkephalin mRNA
levels in forebrain neuronal cultures.............................. .......... 69

5-4 Effect of dibutyryl cyclic AMP and forskolin on proenkephalin
mRNA levels in forebrain neuronal cultures............................. 71

5-5 Time course of dynorphin A1-13 and isoproterenol regulation of
proenkephalin mRNA levels in forebrain neuronal cultures........ 73

5-6 Effect of naloxone on dynorphin A1-13 regulation of
proenkephalin mRNA levels in forebrain neuronal cultures......... 75

5-7 Effect of pertussis toxin on dynorphin Al-13-induced changes
in proenkephalin mRNA levels in forebrain neuronal cultures... 78

5-8 Effect of dynorphin A1-13 on forskolin- and dibutyryl cyclic AMP-
induction of proenkephalin mRNA levels in
forebrain neuronal cultures............................................................ 80

5-9 Effect of cycloheximide on dynorphin A1-13 regulation of
proenkephalin mRNA levels in forebrain neuronal cultures....... 83

5-10 Effect of dynorphin Ai-13 on proenkephalin mRNA levels in
primary glial cultures............................................ .................... 84

5-11 Effect of D-Ala enk and naloxone on isoproterenol-induction
of proenkephalin mRNA levels in C6 glioma cells..................... 87

5-12 Effect of D-Ala enk on adenylyl cyclase activity in isolated
membranes from NG108-15 cells................................................ 90

5-13 Effect of D-Ala enk, cycloheximide and pertussis toxin on
proenkephalin mRNA levels in NG108-15 cells........................ 92














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

OPIOID-COUPLED ADENYLYL CYCLASE AND
REGULATION OF PROENKEPHALIN EXPRESSION

By

Don Roger Marckel Jr.

December 1990

Chairman: Steven R. Childers, Ph.D.
Major Department: Pharmacology and Therapeutics
Opioid receptors are coupled to adenylyl cyclase through G-proteins and,
in most systems, inhibit cyclic AMP synthesis. The goals of the present studies
were to characterize: 1) a novel opioid stimulation of adenylyl cyclase in rat C6
glioma cells; 2) the relationship between opioid-inhibited adenylyl cyclase and
regulation of proenkephalin expression in primary neuronal cultures.
Opioids produced a biphasic response of cyclic AMP accumulation in C6
glioma cells, with low (nM) concentrations of opioid agonists stimulating and
higher (ipM) concentrations inhibiting cyclic AMP accumulation. In cells
pretreated with a phorbol ester to eliminate the inhibitory component of the
opioid response, opioid agonists stimulated cyclic AMP accumulation two-fold
over basal. This response was mediated via mu-like opioid receptors. The
opioid antagonist naloxone antagonized the stimulatory effect by decreasing
agonist potency. However, naloxone also stimulated cyclic AMP accumulation
and may be a partial agonist in this system.









The role of opioid-inhibited adenylyl cyclase was examined in primary
neuronal cultures from neonatal rat forebrain. The opioid agonist dynorphin

A1-13 inhibited cyclic AMP accumulation in intact cells by 60% and directly
inhibited adenylyl cyclase activity in isolated membranes by 15%. This action
was attenuated by naloxone and was sensitive to pertussis toxin.
In these cultures, proenkephalin mRNA expression was inducible by
dibutyryl cyclic AMP and forskolin. Dynorphin decreased both basal and
forskolin-stimulated expression, producing maximal effect between 2 and 8
hours. The inhibitory action of dynorphin was blocked by naloxone and by
pertussis toxin. Dynorphin was more effective in reducing forskolin-induced
proenkephalin mRNA than dibutyryl cyclic AMP-induced message. In cultures
treated with cycloheximide, dynorphin also decreased proenkephalin mRNA
levels, thus suggesting that dynorphin decreased proenkephalin transcription.
These results indicate that opioids inhibit adenylyl cyclase and repress
the expression of the proenkephalin gene in primary neuronal cultures. The
inhibition of adenylyl cyclase may be a mechanism by which opioids regulate
proenkephalin mRNA expression. This effect may be a negative feedback
mechanism for opioids to regulate their own synthesis. In addition, these results
demonstrate a function for adenylyl cyclase in brain which could regulate
neurotransmitter expression and play an important role in long-term neuronal
homeostasis.














CHAPTER 1
INTRODUCTION

General Introduction
Opioids affect both central and peripheral systems to produce such
diverse effects as analgesia and changes in gastrointestinal motility. These
processes begin with the binding of opioids to membrane receptors (Pert &
Snyder, 1973; Simon et al., 1973; Terenius, 1973), which are coupled to
specific second messenger systems like ion channels and adenylyl cyclase.
These systems are tightly coupled to further biochemical reactions, i.e. ion
transport and protein phosphorylation, respectively. The further cellular
responses caused by these second messenger systems are complex and are
only now becoming understood. However, for opioids, inhibition of neuro-
transmitter release, hyperpolarization, and inhibition of smooth muscle
contraction are well-known physiological end-points.
A primary focus of opioid research has been the mechanism of opioid
tolerance. Most neurotransmitter systems, like the B-adrenergic system,
undergo down-regulation of receptors during chronic exposure to agonists
(Hausdorff et al., 1990). Opioid receptors undergo similar changes in cell
culture (Law et al., 1982), but most studies have agreed that chronic morphine
administration in vivo, under conditions which produce dramatic tolerance to
many physiological effects of opioid agonists, does not significantly decrease
either opioid receptor number or affinity. Therefore, the molecular basis of
opioid tolerance must involve events which occur after the receptor binding
step.









The two second messenger systems which have been associated with
opioid receptors are adenylyl cyclase and receptor-coupled ion channels,
although one preliminary report of opioid-stimulated phosphoinositol turnover
has recently appeared (Periyasamy & Hoss, 1990). Opioid receptors are
coupled to potassium channels (Werz & MacDonald, 1983), calcium channels
(Hescheler et al., 1987) and adenylyl cyclase through G-proteins (Blume et al.,
1979). Electrophysiological evidence in rat brain suggests that most effects of
opioids on calcium and potassium channels are mediated directly through G-
proteins and do not involve a diffusible second messenger like cyclic AMP.
Therefore, a major question involves the biological role of receptor-coupled
adenylyl cyclase. With the finding of cyclic AMP-modulation of gene
transcription rates, adenylyl cyclase has come to be viewed as a relatively long-
term effect (compared to ion channels) of neurotransmitter actions. It is now
known that cyclic AMP increases proenkephalin mRNA levels by increasing
gene transcription rates (see below). Other studies showed that chronic
morphine treatment, which does not induce either down-regulation or
desensitization of opioid receptors, decreased proenkephalin mRNA levels in
adult rat striatum (Uhl et al., 1988). Therefore, one of the possible functions of
receptor-coupled adenylyl cyclase is to regulate proenkephalin transcription.
This dissertation explores the regulation of adenylyl cyclase by opioids in
cell culture systems, specifically rat C6 glioma cells and rat primary neuronal
cultures. It also explores the long-term actions of opioid regulation of cyclic AMP
to modulate levels of proenkephalin mRNA in various cells.


G-Proteins: Structure and Function
Neurotransmitter and hormone receptors coupled to adenylyl cyclase are
divided into two categories: stimulatory, such as B-adrenergic, glucagon,








adenosine A2 and dopamine D1; and inhibitory, which includes a2-adrenergic,

adenosine A1, dopamine D2 and opioids. Receptor-mediated stimulation or
inhibition of adenylyl cyclase requires the presence of guanine nucleotides,
specifically GTP (Rodbell, 1980). Nonhydrolyzable GTP analogs (such as
Gpp(NH)p and GTPyS) do not support agonist-induced inhibition of adenylyl
cyclase (Childers & Snyder, 1979). Guanine nucleotides also regulate agonist
binding to receptors. Agonist dissociation rates are increased in the presence of
GTP, thus decreasing agonist affinity. Moreover, GTP decreases binding of
agonists, but not of antagonists (Childers & Snyder, 1980). Thus, the receptor
exists in two different affinity states for agonists depending on the presence of
guanine nucleotides, while antagonists bind with a single high affinity to both
states (De Lean et al., 1980; Kent et al., 1980). These effects of GTP, both on
coupling receptors with adenylyl cyclase as well as regulating agonist binding
to receptor sites, are unifying actions that occur with all adenylyl cyclase-
coupled receptors. In addition, guanine nucleotides couple receptors with other
second messenger systems, including ion channels and phosphoinositol
turnover (Johnson, 1989).
The effects of guanine nucleotides are mediated through specific GTP-
binding proteins called G-proteins (Rodbell, 1980; Gilman, 1984). This family of
proteins are heterotrimers of subunit composition a, B, and y. The different G-
proteins are distinguished by their alpha subunits, which vary in their molecular
weights. The a subunits bind and hydrolyze GTP and have considerable
parallels in primary structure (Manning & Gilman, 1983).
Initially, the family of G proteins was relatively small, with only three
members: transducin, from retinal rod outer segment cells (Fung & Stryer,
1980); Gs, for stimulation of adenylyl cyclase; and Gi, for inhibition of adenylyl
cyclase. The proteins Gs and Gi were distinguished by the difference in









molecular weight of the alpha subunits (42 vs 41 kDa). However, large scale
purifications (Sternweis & Robishaw, 1984; Neer et al., 1984) resulted in the
discovery of the alpha subunit (39 kDa) of another GTP binding protein, dubbed
ao. The amino acid sequence of ao is highly homologous with both aj and as
(Tsai et al., 1987). While the function of Go is not yet clear, solubilized
muscarinic cholinergic receptors were reconstituted with ao to provide guanine
nucleotide regulation of binding (Florio & Sternweis, 1985). Purified a2-
adrenergic receptors were also reconstituted with ao to produce a2-stimulated

GTPase activity (Cerione et al., 1986a). Opioid receptors may be coupled to a
G-protein linked calcium channel in NG108-15 cells by Go (Hescheler et al.,
1987).
Examination of the cDNA sequences for the alpha subunits has revealed
several highly conserved regions which may contain the GTP binding site (Itoh
et al., 1986). Other regions, which contain much of the variability between
subtypes, may be responsible for selectivity of coupling with effector systems or
with receptors. Further molecular studies have revealed three different genes
for the ai protein which vary only slightly (Jones & Reed, 1987; Itoh et al., 1988).
Four forms of as proteins have also been found; however, these arise from
alternative splicing mechanisms of one gene (Bray et al., 1986).
Cholera toxin and pertussis toxin (islet-activating protein) interact with
specific alpha subunits by catalyzing ADP-ribosylation of the proteins (Katada &
Ui, 1982; Northup et al., 1983). Cholera toxin specifically ribosylates as and
pertussis toxin ribosylates ai and ao. The alpha subunit of transducin is
ADP-ribosylated by both cholera toxin and pertussis toxin (Manning et al.,
1984). In these reactions, the toxin catalyzes the transfer of an ADP moiety from
NAD to the G-protein alpha subunit with covalent reaction between the ribose
portion of ADP and a specific amino acid residue of the alpha subunit. This






5


reaction produces essentially irreversible modification of G-protein function.
Paradoxically, the functional consequences of toxin reactions with both Gs and
Gi proteins lead to the same result: an increase in stimulated adenylyl cyclase
activity and an increase in intracellular cyclic AMP. Cholera toxin-mediated
ribosylation blocks the as GTPase, which is normally responsible for inactivation
of adenylyl cyclase (Northup et al, 1983). Thus, cholera toxin is an irreversible
stimulator of the enzyme. On the other hand, pertussis toxin inactivates the
inhibitory function of the ai subunit. Thus, pertussis toxin stimulates adenylyl
cyclase by removing the inhibitory component of the cycle (Katada & Ui, 1982).
An oncogene product, ras-21, binds and hydrolyzes GTP, but is much
smaller than as or ai (21 kDa vs. approx. 40 kDa) (McGrath et al.,1984). There
are currently more than 20 GTP-binding proteins in the 21 kDa family. Although
considerable structural homology exists between these proteins and the alpha
subunits of other G-proteins, ras-21 cannot substitute for as or ai in regulating
adenylyl cyclase (Helmreich & Pfeuffer, 1985).
The other two subunits of G-proteins, 6 and y, were once thought to be
identical in all G-proteins. The gamma subunit is small (molecular weight 5-8
kDa) and its function in receptor-adenylyl cyclase coupling is not yet known
(Gilman, 1984; Helmreich & Pfeuffer, 1985). Analysis of gamma subunits by 2-D
peptide mapping has revealed differences between transducin gamma subunits
compared to those of other G-proteins (Hildebrandt et al., 1985). Also, recent
studies have suggested the presence of two types of nonretinal gamma
subunits (Evans et al., 1987). Two distinct beta subunits, 62 and 13 (35 and 36
kDa, respectively), have been cloned and shown to be encoded by separate
genes (Amatruda et al., 1988). The beta subunit apparently binds to the catalytic
unit of adenylyl cyclase to inactivate the enzyme (Northup et al., 1983). Another
group (Logothetis et al., 1987) has suggested that By functions directly to open









atrial K+ channels, but this was shown to be an indirect action mediated through
phospholipase A2 (Kim et al., 1989).
The identification of several homologous members among a class of
GTP-binding proteins implies that several second messenger systems are
coupled to receptors via G-proteins. In mast cells, Nakamura and Ui (1985)
showed attenuation of receptor-mediated stimulation of phospholipase A2 and

arachidonic acid release by treatment with pertussis toxin. In other cells,
receptor-mediated changes in phosphoinositide turnover have been associated
with G-protein coupling mechanisms (Nishizuka, 1984). Guanine nucleotides
stimulate phosphoinositide turnover (Cockcroft & Gomperts, 1985; Gonzales &
Crews, 1985; Litosch et al., 1985) in isolated membranes, and regulate binding
of agonists to receptors which stimulate phosphoinositide turnover (Evans et al.,
1985). The muscarinic-cholinergic receptor is coupled via G-proteins to K+
channels in the heart (Breiweiser & Szabo, 1985; Pfaffinger et al., 1985). Other
studies showed direct effects of alpha subunits on ion channels. The alpha
subunit of Gk, a Gi-like protein (Codina et al., 1987), opened atrial potassium
channels and as opened cardiac calcium channels (Yatani et al., 1987). Also,
ao may couple opioid receptors to calcium channels in NG108-15 cells
(Hescheler et al., 1987).


Receptor-Adenvlvl Cyclase Coupling Mechanisms
Experiments with purified G-proteins and genetic mutants of S-49
lymphoma cells have provided models for the actions of agonists in stimulating
and inhibiting adenylyl cyclase through Gs and Gi proteins. These models have
been tested by experiments which reconstitute purified receptors, adenylyl
cyclase and G-proteins into phospholipid vesicles (Cerione et al., 1984).
Current data suggest that the Ga subunits are responsible for the actual









coupling of receptors to adenylyl cyclase and that the By subunits are important

for inactivating the alpha subunit. Under unstimulated conditions, as binds GDP;
in this state the aBy subunits are bound together and adenylyl cyclase is
inhibited. The receptor is in the high affinity binding state, bound to asBy, while
the ligand binding site is vacant. In the presence of the agonist, binding is
facilitated because the receptor is present in the high affinity state. The binding
of the agonist causes intracellular GTP to displace GDP on as. The binding of

GTP to as dissociates the Gs complex, thus dissociating By from adenylyl
cyclase and stimulating the enzyme. The stimulation is terminated in two ways.
First, binding of GTP to as shifts the receptor from a high affinity state into a low
affinity state, thus facilitating dissociation of the agonist from the receptor.
Second, the as GTPase activity hydrolyzes GTP, thus regenerating the GDP-
bound form of as and causing re-association of the Gs subunits. This
reassociation returns the adenylyl cyclase to its inactive (basal) state and the
receptor back to its high affinity state. The crucial role of GTPase in terminating
adenylyl cyclase stimulation explains the actions of cholera toxin: By
ribosylating as and inactivating GTPase activity in this subunit, the stimulation of
adenylyl cyclase is not terminated, and the result is persistent stimulation of
adenylyl cyclase.
The mechanism of receptor-mediated inhibition of adenylyl cyclase is not
as clear. There are many obvious parallels with the stimulatory cycle, with the
respective G-proteins containing parallel structures and functions. However,
one difference between Gs- and Gi-mediated systems is the action of sodium,
which is required for receptor-inhibited adenylyl cyclase and mimics guanine
nucleotides in inhibiting agonist binding to Gi-coupled receptors (Jakobs, 1979).
Gilman and colleagues (Cerione et al., 1986b) suggested that a mechanism of
subunit dissociation followed by mass action is responsible for adenylyl cyclase









inactivation. In this theory, binding of the agonist to the receptor promotes GTP
binding to ai and dissociation of the Gi subunits in a manner precisely
analogous to the Gs cycle described above. In this case, however, dissociation
of the Gs subunits produces an excess of free By. These subunits would then be
free to associate with free as subunits which had been formed previously
through the actions of the stimulatory hormones and receptors. The reaction of
free as with the By subunits liberated by the inhibitory cycle would tend, by
simple mass action, to negate the actions of the stimulatory cycle and therefore
cause inhibition of adenylyl cyclase. This principle is best described by the
following reversible reaction:


as B-C -- as + By + C

INHIBITED STIMULATED


where an excess of free By (formed from dissociation of Gi) would tend to push

the reaction towards the left and result in inhibited adenylyl cyclase.
Unfortunately, there are several drawbacks to this scheme of adenylyl
cyclase inhibition. First, it is not clear that the G-protein subunits are sufficiently
mobile in the lipid milieu of cell membranes to allow mass action principles to
dominate their actions. Second, there are several cases where receptor-
mediated inhibition of adenylyl cyclase occurs in the absence of as proteins
(Jakobs & Schultz, 1983; Childers & LaRiviere, 1984), a situation which clearly
could not occur if as were necessary for the actions of Gi. An alternative
possibility is that ai inhibits adenylyl cyclase by direct interactions with the
enzyme itself. Katada et al. (1986) found a direct action of ai on the catalytic unit
of adenylyl cyclase, but other groups have not confirmed these results (Cerione
et al., 1986a, 1987). Finally, results in brain membranes (Hatta et al., 1986)









suggest that stimulation and inhibition of adenylyl cyclase occur during
exchange of GTP between ai and as. At the present time, most of these models
come from purified proteins in artificial systems, and precise molecular
interactions of these proteins in normal brain membranes may be quite different.


Opioid-lnhibited Adenvlvl Cyclase in Brain Membranes

Chou et al. (1971) reported inhibition of adenylyl cyclase by opioids in
brain membranes. Later, Collier and Roy (1974) reported that morphine
inhibited PGE-stimulated adenylyl cyclase in rat striatal membranes. Several
other papers appeared during this early period substantiating the role of opioid
agonists in inhibiting brain adenylyl cyclase (Minneman & Iversen, 1976;
Wilkening et al., 1976; Tsang et al., 1978), and at least one paper appeared
showing stimulation of adenylyl cyclase by opioids (Puri et al., 1975). However,
several groups were unable to reproduce these findings (Tell et al., 1975; Van
Inwegen et al., 1975). The reasons for these discrepancies lay in the lack of
information about the requirements for receptor-coupled adenylyl cyclase,
especially the necessity for GTP. As a result, most studies focused on
neuroblastoma cells (see below), in vivo changes in cyclic nucleotide
metabolism (Wolleman, 1981), on brain slice experiments. In the latter type of
experiment, several groups (Minneman & Iversen, 1976; Havemann &
Kushinsky, 1978; Barchfeld et al., 1982) have described opioid inhibition of
cyclic AMP levels in striatal slice preparations.
With the realization that guanine nucleotides were involved in receptor-
adenylyl cyclase coupling, more reproducible results in brain membranes were
obtained. Law et al. (1981) and Cooper et al. (1982) showed that opioid-
inhibited activity in brain, like in cell culture, was GTP-dependent. Although
some disagreement on sodium dependence of opioid inhibition was reported









(Law et al., 1981; Cooper et al., 1982), careful removal of sodium from all
components of the adenylyl cyclase assay revealed that opioid-inhibited
adenylyl cyclase was sodium-dependent in rat brain as well as cell culture
(Childers, 1988). The pharmacological properties of opioid-inhibited adenylyl
cyclase in rat brain were similar to a delta receptor response, since opioid
peptides were more potent than opioid alkaloids. However, this activity
remained difficult to quantitate since the actual level of inhibition by opioid
agonists was small, averaging around 20%.
The properties of opioid-inhibited adenylyl cyclase in rat brain have been
examined in low pH pretreated membranes (Childers et al., 1986; Childers,
1988), which enhances inhibition of adenylyl cyclase while abolishing
stimulation of adenylyl cyclase in rat brain (Childers & LaRiviere, 1984). Like all
other receptor-inhibited activities, this reaction required guanine nucleotides. As
in NG108-15 cells, sodium was required for opioid inhibition, and low (<5mM)
concentrations of Mg2+ optimized opioid inhibition. Kinetic studies showed that
opioid inhibition of adenylyl cyclase was noncompetitive, decreasing Vmax of
the enzyme without affecting Km of the enzyme for ATP. Regional distribution
showed that opioid inhibition occurred primarily in striatum, frontal cortex, and
amygdala, with small inhibition in other regions. This distribution was similar,
but not identical to that of classical delta receptor binding sites. The
pharmacological profile of opioid-inhibited adenylyl cyclase in striatum did not
follow any known receptor binding site. For example, the agonist profile
resembled delta receptors, with enkephalin and enkephalin analogs more
potent than mu agonists, but other opioid peptides such as B-endorphin and
dynorphin were equipotent to enkephalin. These data suggested that multiple
opioid receptors might be coupled to adenylyl cyclase in rat brain. Detailed
studies in rat thalamus (Childers, 1988) have revealed the possible presence of









mu-inhibited activity, since inhibition by DAGO was highest in that region, and
since inhibition by DAGO and DPDPE were partially additive. Also, the mu
antagonist naloxone was more potent in blocking DAGO inhibition than DPDPE
inhibition. Finally, kappa-inhibited adenylyl cyclase has been demonstrated in
guinea pig cerebellum, a tissue which lacks other opioid receptors subtypes
(Konkoy & Childers, 1989).
The relationship between classical opioid receptor binding sites and
opioid-inhibited adenylyl cyclase is not straight forward. Studies which have
examined opioid-inhibited adenylyl cyclase in cells, in which a large proportion
of the opioid receptors have been enzymatically blocked, have demonstrated
little effect on the opioid activity (Pasternak & Snyder, 1975; Fantozzi et al.,
1981; Nijssen et al, submitted). It seems possible that the opioid receptor
coupled to adenylyl cyclase in brain membranes does not correspond to any of
the traditional receptor binding sites determined by binding studies and
physiological experiments. It is important to note that receptor binding
experiments are normally conducted under conditions that would not allow any
significant coupling between receptors and adenylyl cyclase. To examine this
relationship, low affinity delta sites have been characterized in brain
membranes with GTP and sodium. These sites were identified by the ability of
gM concentrations of DPDPE to displace [3H]-naloxone binding (Childers et al.,
1987). In striatal membranes, these sites correlated well with opioid-inhibited
adenylyl cyclase by their sensitivity to phospholipase A2 and B-CNA, and by the
ability of naloxone and DPDPE to protect against the effects of B-CNA.
Therefore, low affinity binding sites, which presumably exist in intact cells with
physiological levels of GTP and sodium, are coupled to adenylyl cyclase.
Interestingly, when opioid effects are determined in neurons by
electrophysiological experiments, agonist affinities are similar to those









observed for low affinity binding sites and opioid-inhibited adenylyl cyclase
(Duggan & North, 1983).


Opioid-lnhibited Adenvlyl Cyclase in Cell Culture
While the initial discovery of opioid-inhibited adenylyl cyclase was made
in brain membranes, the most progress in this area occurred in neuroblastoma
cells. In these cultured cells, opioid inhibition of adenylyl cyclase was more
pronounced than that of brain, partly because of the biochemical simplicity of
the transformed cells compared to brain. One neuroblastoma x glioma hybrid
cell line, the NG108-15 cells, contained high levels of opioid receptor binding
sites, compared to nonhybrid neuroblastoma cells (Hamprecht, 1977). Further
studies showed that the pharmacological characteristics of these sites
corresponded to those of delta receptors (Chang et al., 1978). Although most
studies in cell culture have utilized NG108-15 cells, other experiments have
established that other transformed cell types, including N4TG1, N1E-115, and
N18TG2 cells, also contained delta receptors, although at much lower density
(Gilbert & Richelson, 1983).
Early experiments with NG108-15 cells showed that opioid agonists
inhibited PGE1-stimulated adenylyl cyclase (Sharma et al., 1975a, 1975b,

1977; Traber et al., 1975). This response was mediated by classical opioid
receptors since inhibition was blocked by naloxone. Since these cells
contained delta receptors, enkephalin analogs were more potent in inhibiting
adenylyl cyclase than opioid alkaloids; nevertheless, morphine and related
alkaloids inhibited adenylyl cyclase at micromolar concentrations in a
stereospecific manner (Sharma et al., 1975a). Opioid inhibition of adenylyl
cyclase in NG108-15 cells occurred not only for PGE1-stimulated activity but
also for basal adenylyl cyclase as well as adenosine-stimulated and cholera









toxin-stimulated adenylyl cyclase (Propst & Hamprecht, 1981). Like other
inhibitory receptor systems, opioid inhibition of adenylyl cyclase required both
sodium and GTP (Blume et al., 1979). Also, incubation of NG108-15 cells with
pertussis toxin abolished opioid inhibition of adenylyl cyclase (Hsia et al.,
1984), suggesting that Gi (or Go) proteins were required for inhibitory activity.
Initial studies of the effect of chronic morphine exposure on NG108-15
cells resulted in no observed desensitization or down-regulation of opioid
receptors (Sharma et al., 1975a, 1975b). However, later studies utilized either
the nonselective opioid agonist etorphine or opioid peptides, both of which
produced down-regulation of opioid receptors and desensitization (Law et al.,
1982, 1983b).
The relationship between receptor occupancy and efficacy of opioid
inhibition of adenylyl cyclase was explored by blocking receptor binding sites
with the irreversible antagonist B-chlornaltrexamine (B-CNA) (Fantozzi et al.,
1981). Interestingly, blockade of 95% of opioid receptor binding sites did not
alter the inhibition of adenylyl cyclase by opioid agonists, suggesting the
presence of a large population of spare receptors in the NG108-15 cells. In
detailed receptor binding studies by Law et al. (1985), sodium and GTP
produced three distinct agonist binding states of opioid receptors in NG108-15
cells whose agonist association rates were functions of receptor occupancy.
These experiments directly demonstrated a decrease in opioid agonist affinity
with respect to receptor occupancy during inhibition of adenylyl cyclase by
opioid agonists.
In NG108-15 cells, the relationship between receptor occupancy and
inhibition of adenylyl cyclase is complex. Costa et al. (1985) demonstrated
several agonist affinity states in NG108-15 cell membranes, and showed that
the sites coupled to adenylyl cyclase did not correspond to the high affinity delta








sites identified in binding studies. These studies point to the complicated
relationship which must occur between receptor binding sites and second
messenger systems, even in a simplified cell system like NG108-15 cells.
The glioma parent line of NG108-15 cells, C6 glioma cells, contain 8-
adrenergic receptors but, under normal conditions, no detectable opioid
receptors. However, when C6 cells were treated with desmethylimipramine for
24 hours, B-adrenergic receptors were down-regulated and opioid receptor
binding sites were detected (Tocque et al., 1984). Furthermore, as opioid
receptor sites appeared, opioid agonists inhibited cyclic AMP levels, consistent
with the appearance of opioid-inhibited adenylyl cyclase. The pharmacological
identity of these receptors has not yet been precisely identified, but preliminary
results suggested that they were not classical delta receptor sites.
Since NG108-15 cells (and the other cell types mentioned above)
contain only delta opioid receptors, one question is whether other opioid
receptor subtypes are also negatively coupled to adenylyl cyclase. One
approach to this question was to locate other transformed cell types containing
different opioid receptor subtypes. Yu et al. (1986) located a human
neuroblastoma cell line, SK-N-SH, with both mu and delta receptor binding
sites. Frey and Kebabian (1984) identified mu receptors in the pituitary tumor
7315c. In both cell types, opioid agonists inhibited adenylyl cyclase, with
morphine and other mu agonists being more potent than enkephalin and other
delta analogs. These results suggested that mu receptors as well as delta
receptors were negatively coupled to adenylyl cyclase in transformed cells.
A study of mouse embryonic neuronal cultures showed Leu-enkephalin-
inhibited adenylyl cyclase in striatal cultures but not in cerebral cortex or
mesencephalon cultures (Chneiweiss et al., 1988). In this study, the mu and
delta agonists, DAGO and D-Thr, Leu-enkephalin, Thr5 (DTLET), respectively,









both inhibited adenylyl cyclase by 30% with very similar affinities. Naloxone
blocked both agonist responses. Interestingly, the DAGO and DTLET responses
were not additive, suggesting that the receptors mediating the responses were
either the same or they were located on the same population of cells.


Opioid Stimulation of Adenvlvl Cyclase
In addition to the work investigating opioid-inhibited adenylyl cyclase,
there have been reports of opioid-stimulated adenylyl cyclase in various
tissues. Opioids have been demonstrated to exert an excitatory effect on
cultured cardiac myocytes (Laurent et al., 1986). Studies to investigate the
mechanism of these actions revealed that opioids stimulated calcium uptake
(Laurent et al. 1986). In addition, opioids stimulated adenylyl cyclase activity
and naloxone was able to block this action.
Studies in fetal mouse spinal cord-ganglion explants demonstrated a
biphasic nature of opioids on the action potential in many of these cells (Shen
and Crain, 1989). High concentrations of opioid agonists (0.1 to 10 gLM)
shortened the action potential, whereas lower concentrations (1-10 nM)
prolonged the action potential. Pretreatment of cultures with pertussis toxin
reduced the number of cells showing shortening of the action potential and
increased the number of cells showing prolongation of the action potential,
suggesting a role of G-proteins in both actions. Intracellular dialysis of these
cells with an inhibitor of protein kinase A blocked the prolongation of the action
potential induced by low concentrations of DADLE, but did not attenuate the
shortening induced by higher concentrations of the opioid (Chen et al., 1988).
Studies to examine the opioid regulation of adenylyl cyclase in these cells
demonstrated that in the presence of forskolin, the delta and kappa opioid
agonists, DTLET and U-50488H, respectively, inhibited adenylyl cyclase









(Makman et al., 1988). This effect was lost by incubating the cultures with
pertussis toxin. The mu agonist, levorphanol, had no significant effect on
forskolin-stimulated adenylyl cyclase. However, in the absence of forskolin,
levorphanol produced a small (23%) but significant stimulation of adenylyl
cyclase. The opioid antagonist, naloxone, also stimulated adenylyl cyclase to
the same level. Upon treatment with pertussis toxin, the levorphanol effect was
increased twofold, but the naloxone effect was unchanged from control cultures,
suggesting that the levorphanol was acting on a receptor coupled to a G-
protein, but that the naloxone effect might be a nonspecific effect. This dual
modulation of action potential duration and adenylyl cyclase may play a role in
the development of tolerance and the "hyperexcitability properties associated
with addiction" (Crain & Shen, 1990).


Opioid and Cyclic AMP Regulation of Proenkephalin mRNA
Enkephalin pentapeptides are synthesized from a high molecular weight
precursor (proenkephalin) whose gene was cloned and sequenced in 1982
(Noda et al., 1982). Regulation of proenkephalin mRNA was first studied in
adrenal chromaffin cells, which not only store enkephalin peptides, but also
release enkephalins with catecholamines in response to acetylcholine
(Kilpatrick et al., 1981). Nicotinic receptor stimulation, in addition to releasing
enkephalin, increased proenkephalin mRNA and peptide content. This effect
was cyclic AMP-dependent, in that 8-Bromo cyclic AMP (but not 8-Bromo cyclic
GMP) increased proenkephalin mRNA (Quach et al., 1984).
Two groups simultaneously reported the mapping of DNA regions which
were necessary for cyclic AMP responsiveness in the rat somatostatin gene
(Montminy et al., 1986) and the proenkephalin gene (Comb et al., 1986). A
palindromic DNA sequence of TGACGTCA was found to have only one base









substitution. This highly conserved sequence, called the cyclic AMP responsive
element (CRE), has been mapped on the human vasoactive intestinal peptide
(Tsukada et al., 1987), the rat tyrosine hydroxylase (Lewis et al., 1987) and the
human glycoprotein hormone alpha subunit genes (Delegeane et al., 1987;
Silver et al., 1987).
Transfection of PC12 cells with a fusion gene containing the somatostatin
CRE promoter and the chloramphenicol transferase gene confered cyclic AMP
responsiveness to CAT activity (Montminy et al., 1986). When this fusion gene
was transfected into the mutant PC12 line A126-1B2, which is deficient in cyclic
AMP-dependent protein kinase 2, cyclic AMP responsiveness was greatly
reduced, suggesting that protein kinase 2 activity was required. Montminy and
Bilezikjian (1987) later characterized a nuclear protein which bound selectively
to the CRE in the somatostatin gene. This 43K CRE binding protein (CREB) was
found to be phosphorylated in vitro by the catalytic subunit of cyclic AMP-
dependent protein kinase. When PC12 cells were stimulated with forskolin,
phosphorylation increased 3 to 4-fold. Gonzalez and Montminy (1989) have
also shown that this protein is phosphorylated by cAMP-dependent protein
kinase at the serine-133 position and that this phosphorylation is necessary for
induction of transcription.
Further analysis of the proenkephalin gene has shown that there are two
different promoter elements, ENKCRE-1 (TGGCGTA) and ENKCRE-2
(TGCGTCA) in the promoter region (Comb et al., 1988). ENKCRE-2 was most
analogous to the CRE of the VIP, somatostatin and TH genes (Hyman et al.,
1988). Indeed, ENKCRE-2 was shown to bind the same trans-acting factor by
cotransfection experiments and in DNAse I footprint assays as these other
elements. The ENKCRE-1 binds a different trans-acting factor than the other
genes, indicating it has a separate function.









The relationship between ENKCRE-1 and ENKCRE-2 is complex.
Mutational analysis (Comb et al., 1988) showed that a deletion in ENKCRE-1
(A97) lowered response to cyclic AMP activators 10-fold, but still gave a small
residual 2 to 3-fold induction in response to either cyclic AMP or phorbol esters.
ENKCRE-2 alone, without ENKCRE-1, is capable of augmenting transcription 2-
3 fold; however, ENKCRE-1 is inactive in the absence of ENKCRE-2. The trans-
acting DNA-binding protein which regulates each site may also be different.
ENKCRE-2 binds CREB, while another nuclear factor (ENKTF-1) binds to
ENKCRE-1, but not ENKCRE-2. This factor may act synergistically with
ENKCRE-2 to regulate transcription. Other previously described transcription
factors (AP-1 and AP-4) also bind to overlapping regions of ENKCRE-2. In
addition, AP-2 was shown to bind to a region downstream of ENKCRE-2. AP-2,
a 52 kd protein, was shown to act synergistically with ENKCRE-2 to confer
maximal response to cyclic AMP and phorbol esters. Imagawa et al., (1987),
using the human metallothionein IIA gene control region, showed that AP-2
activity increased after treatment of cells with phorbol ester or forskolin, whereas
AP-1 activity only increased with phorbol.
The regulation of proenkephalin mRNA by cyclic AMP in nontransformed
cells has been examined by several groups. Vilijn et al. (1988) demonstrated
regional variation in proenkephalin mRNA in cultured neurons and astrocytes.
Melner et al. (1990) showed that proenkephalin mRNA levels in cultured
astrocytes was increased following treatment of cells with either isoproterenol or
8-(4-chlorophenyl thio)adenosine 3'-5' cyclic monophosphate and that both
agents stimulated secretion of unprocessed proenkephalin into the culture
medium.
Other studies have explored whether opioids themselves regulate
proenkephalin mRNA synthesis. One study examined the effect of








administration of the opioid antagonist naltrexone to rats for 8 days (Tempel et
al., 1990). Striatal proenkephalin mRNA content was increased 12-fold by
chronic antagonist treatment, with only small changes in the hippocampus and
hypothalamus and no change in frontal cortex. This suggested the possibility
that opioid agonists would decrease proenkephalin mRNA in cells with
autoreceptors. However, a study in NG108-15 cells, which have an abundance
of opioid receptors and also synthesize proenkephalin mRNA, showed that
etorphine treatment transiently decreased proenkephalin mRNA levels, but if
treated for 5 days, enhanced proenkephalin mRNA levels (Schwartz, 1988).
Treatment of these cells with forskolin for 24 hours also increased
proenkephalin mRNA, but this was not additive with the etorphine increase. The
author suggested that in these cells, opioid treatment was affecting
proenkephalin mRNA levels through a mechanism other than cyclic AMP, since
forskolin also increased proenkephalin mRNA levels and etorphine treatment
did not alter the cyclic AMP content of the cells.
Another study (Uhl et al., 1988) examined proenkephalin mRNA levels in
the neostriatum of rats which had been made tolerant to morphine by
implantation of morphine base pellets. After 5 days, proenkephalin mRNA
content was reduced to 66% (1-globin mRNA and somatostatin mRNA was
unchanged). Rats which were made tolerant and then put into withdrawal by
removal of the pellets, showed a reduced proenkephalin mRNA level (77%)
after 3 days. The peptide levels were not significantly altered after the 5 days of
morphine, however, after 3 days of withdrawal, peptide levels were reduced by
36%. There were no significant changes in somatostatin peptide levels. At
present, no reports have systematically explored the relationship between
proenkephalin mRNA synthesis and opioid-inhibited adenylyl cyclase. The









working hypothesis for this project is that in neurons, opioids regulate their own
synthesis through cyclic AMP regulation.


Aims
Opioid regulation of adenylyl cyclase has been well characterized in rat
brain and the NG108-15 cell line. In addition, opioid regulation of
proenkephalin mRNA is being studied in both of these systems. However,
mechanistic studies in whole animals are not always feasible, and the effects of
opioids on both adenylyl cyclase and proenkephalin mRNA regulation in
NG108-15 cells vary significantly from those seen in rat brain. Our goal was to
study the relationship between opioid regulation of adenylyl cyclase and
proenkephalin mRNA in a system which more closely resembled rat brain, yet
offered the flexibility of a cell line. Therefore, rat primary neuronal cultures were
chosen because these are nontransformed cells from rat brain which could be
studied in vitro. In the course of these studies, a novel biphasic cyclic AMP
response to opioids was observed in C6 glioma cells. Therefore, this project
consists of three separate specific aims:


1. Opioid effects on cyclic AMP in C6 glioma cells. In these cells, opioid
receptors can stimulate, as well as inhibit, adenylyl cyclase.


2. Opioid effects on cyclic AMP in primary neuronal cultures. These experiments
characterize the opioid action on adenylyl cyclase in the culture system utilized
in specific aim number 3.






21


3. Regulation of proenkephalin mRNA levels in cell cultures by opioids. These
experiments explore the hypothesis that opioid-inhibited adenylyl cyclase can
act as a negative feedback regulatory system of enkephalin synthesis.














CHAPTER 2
MATERIALS AND METHODS


Cell Culture
Rat C6 glioma cells obtained from ATCC at passage number 39 were
maintained in Dulbecco's Modified Eagles Medium (DMEM; Hazelton
Biologicals) plus 100 units/ml penicillin, 0.1 mg/ml streptomycin and 2.5 mg/ml
amphotericin B and 10% Serum Plus (Hazelton Biologicals) or 2% fetal calf
serum/10% horse serum (Hazelton Biologicals) in a humidified atmosphere of
5% C02/95% air at 37. Cells were subcultured weekly into T75 Falcon flasks
for propagation and 100 mm plastic culture dishes for experimentation. Cells
were detached from the growing surface by removal of media, washed once
with EGTA solution (137 mM NaCI, 5.6 mM glucose, 5 mM HEPES, 5 mM KCI,
and 1 mM EGTA, pH 7.4), then incubated in EGTA solution at 370 for 10 min.
Suspended cells were then utilized for either propagation or experimentation.
The mouse neuroblastoma x rat glioma hybrid cell line, NG108-15,
obtained as a kind gift from the laboratory of Dr. Jean Bidlack, was maintained
in DMEM supplemented with 5% heat-inactivated fetal calf serum, 25 mM
glucose, 0.1 mM hypoxanthine, 16 iM thymidine, 1 liM aminopterin, 100
units/ml penicillin, 0.1 mg/ml streptomycin and 2.5 mg/ml amphotericin B in a
humidified atmosphere of 5% C02, 95% air at 370. Cells were plated in 100 mm
Corning culture dishes for experimentation or in Corning T75 tissue culture
flasks for propagation.









For growth of rat neuronal cultures, the method of Sumners et al. (1983)
was used. One-day old rats were euthanized with pentobarbital, the brains
removed and placed in an isotonic salt solution containing 100 units/ml
penicillin, 0.1 mg/ml streptomycin and 2.5 mg/ml amphotericin B, pH 7.4. Brains
were dissected, cleaned of pia, gently minced and suspended in 0.25% trypsin
for 10 min. at 37. DNAse I (160 I.g) was added to the suspension for the final 5
min. of incubation. Cells were diluted with DMEM containing 10% horse serum
and antibiotics. After centrifugation at 200 x g for 4 min., cells were resuspended
in fresh DMEM with 10% horse serum and strained though sterile cheesecloth.
Cells were then plated at 3 x 106 cells/ml on tissue culture plates coated with
poly-L-lysine and incubated in a humidified atmosphere of 5% CO2/95% air at
370. After 3 days, 10 gM cytosine arabinoside was added to fresh media and
incubated for 2 days to decrease the glial population to approximately 20-30%
of the surviving cells. The neuronal-enriched cultures were then incubated in
media without cytosine arabinoside for an additional 10-15 days. For
experiments with opioid compounds, cultures were switched to serum-free
DMEM one hour prior to drug additions.
For glial cultures, the tissue was harvested using the same method. The
cells were plated on 100 mm Corning tissue culture plates at a density of 1.8 x
106 cells/ml (10 ml/plate) and grown for 7 days with one change of media after 3

days. After 7 days in culture, the plates were washed free of serum-containing
media with solution D and the cells were detached by incubating with trypsin
(0.25%, 3 ml/plate) for 5 min. DMEM with 10% fetal bovine serum was added
and detached cells were pooled and centrifuged at 200 x g for 5 min. The cell
pellet was resuspended and plated in fresh DMEM with 10% fetal bovine serum
at a density of 100,000 cells/ml.










Cyclic AMP Assays in Intact Cells
To measure stimulation and inhibition of adenylyl cyclase in intact cells,
levels of cyclic AMP formed in the whole cell was assayed by the cyclic AMP
binding protein method (Brown et al., 1971). For C6 glioma cells, 100 mm plates
were rinsed with warm EGTA buffer (described above) and incubated at room
temperature with 5 ml EGTA buffer until cells were lifted off plate (5-10 min).
Cells were pooled and centrifuged at 200 x g for 3 min. Cells were then
resuspended in EGTA buffer with 1 mM MgCI2 and 10 mM theophylline and
added to tubes containing various drug additions and buffer to a final volume of
100 gl. All tubes were incubated 10 min at 370 with agitation and then immersed
in boiling water for 2 min. Tubes were placed on ice and 400 1l cyclic AMP
buffer (50 mM Tris, 1 mM dithiothriotol, 8 mM theophylline, pH 7.4) is added.
Samples were centrifuged at 1000 x g for 15 min. Aliquots of the supernatant
were added to 400 g assay tubes containing [3H]-cAMP (25,000 cpm), 2 gg
cyclic AMP dependent protein kinase, and buffer up to 200 gIl total. Assay tubes
were incubated on ice for 70 min., then 100 Cl of hydroxylapatite suspension
(1:5 v/v in H20) was added and tubes were centrifuged at 10,000 x g for 5 min.
The resulting pellet, containing bound [3H]-cAMP, was resuspended in 0.5 ml of
1N HCI and radioactivity determined by liquid scintillation spectrophotometry
after addition of 5 ml of liquid scintillation fluid. Quantitation of cyclic AMP is
obtained from standard curves constructed using six concentrations of cyclic
AMP from 0.2 to 10 pmoles.
For assay of cyclic AMP content in neuronal or glial cultures, cultures in
12 well plates were washed once with HBSS and then incubated for 10 min at
370 in HBSS with 100 IM IBMX. Cultures were then incubated an additional 10
min. at 370 after the addition of the indicated drugs. The reaction was terminated









by removal of the HBSS and addition of 0.5 ml ethanol to lyse the cells. The
ethanol was removed to 1.5 ml microfuge tubes and evaporated in vacuo. The
extract was then reconstituted in cyclic AMP buffer and aliquots were assayed
for cyclic AMP content in the same manner as above. Protein on plates was
determined by the method of Lowry et al. (1957) after dissolution in 1N NaOH.
Values from cyclic AMP assay were then normalized to pmoles/min/mg protein.




Adenylyl Cyclase Assay
Adenylyl cyclase activity was measured by the method of Salomon
(1976). C6 glioma cells, NG108-15 cells or primary neuronal cultures scraped
from 100 mm plates in Tris buffer (50 mM Tris, 3 mM MgSO4, pH 7.4), were

homogenized in a Potter-Elvehjem homogenizer. After centrifugation at 48,000
x g for 10 min., the pellet was resuspended in adenylyl cyclase buffer (50 mM
Tris-HCI, 1 mM EGTA, 5 mM MgCI2, pH 7.4). Membranes (20-100 gg) were
added to tubes containing 5 mM creatine phosphate, 5 units creatinine
phosphokinase, 10 mM theophylline, 0.1 mg/ml BSA, and 50 gM each of cyclic
AMP, ATP and GTP together with various drug additions to a final volume of
100 pl. Reactions were initiated by addition of [a-32P] ATP (2 gCi) and tubes

were incubated at 300 for 10 min. and stopped by the addition of 100 pl of stop
solution (2% SDS, 45 mM ATP and 1.3 mM cyclic AMP), then immersed in
boiling water for 2 min. Enzyme blanks were tubes with membranes and
cocktail added in the presence of stop solution. Tubes were allowed to cool to
room temperature, then 50 p.l [3H] cyclic AMP (10,000 cpm / 50 .l) was added to
each tube. Cyclic AMP was then separated from other nucleotides by sequential
chromatography through Dowex AG 50-X8 and alumina columns. Radioactivity
was determined by liquid scintillation spectrophotometry in both [32P] and [3H]









channels. Recovery of cyclic AMP from the columns was determined by the [3H]
cpm in each sample, and all results were corrected for differential recovery (60-
70%).




Analysis of RNA
Total RNA was extracted by the method of Chomczynski and Sacchi
(1986). Briefly, for brain tissue, rats were decapitated, brain regions dissected,
and tissue placed in 1 ml of Solution A (4 M guanidinium isothiocyonate, 25
mM sodium citrate, pH 7; 0.5% sarkosysl, 0.1 mM 8-mercaptoethanol) and
homogenized briefly with a Polytron tissue homogenizer. For primary cell
cultures, plates were rinsed with ice-cold PBS without divalent cations, then 300
PI1 solution A was added to each plate. Plates were scraped and solution
removed to sterile centrifuge tubes. For NG108-15 cells, plates were.rinsed
twice with ice-cold PBS without divalent cations and the cells collected by
scraping in 1 ml PBS and centrifuged for 30 sec in a microfuge. The cell pellet
was then resuspended by vortexing in 300 .I1 of solution B (140 mM NaCI, 1.5
mM MgCI2, 10 mM Tris (pH 7.8), 1 mM DTT, 0.5% nonidet P-40), placed on ice
for 5 min and then centrifuged in a microfuge for 90 sec. Supernatant was
transferred to fresh tubes and 300 p.l solution A added. Total RNA was isolated
by addition of 0.1 volumes of 2 M sodium acetate (pH 4.0), followed by one
phenol/chloroform extraction, two isopropanol precipitations and one ethanol
wash. In the case of brain RNA, O.D. readings (260/280 nm) were used to
normalize amount loaded. For Northern blots, RNA was denatured in formamide
and formaldehyde and stained with ethidium bromide. Samples were
fractionated by size by electrophoresis through a 1.5% agarose gel under 30-40
volts. RNA was blotted onto Zeta probe nylon membranes (Bio Rad) for >8









hours and baked at 800 in vacuo for 30 min. For analysis of proenkephalin
mRNA, blots were prehybridized in 4 x SSPE, 1% SDS, 5 x Denhardt's (2%
each of polyvinylpyrrolidone, bovine serum albumin and Ficoll 400), 50%
formamide and 500pg/ml denatured salmon sperm DNA at 650 for 1-4 hours.
Blots were hybridized in the same solution with a 950-base riboprobe
(synthesized from pYSEA1, kindly provided by Dr. Steven Sabol) for 20 hours
at 650. Blots were then washed once in 2 x SSC and 0.1% SDS and four times
in 0.1 x SSC and 0.1% SDS at 700 for 20 min each. Blots were exposed to X-
ray film with intensifying screens at -700 for 1-3 days. Where levels of
proenkephalin mRNA were reported as PE units, bands were analyzed by
densitometric scanning and proenkephalin mRNA levels expressed relative to
alpha-tubulin expression and all samples on one blot normalized to the control
lane expressed as 1 PE unit.
For analysis of alpha-tubulin, blots were stripped by washing in 0.1 x
SSC and 1% SDS at 950 for 1 hour. Blots were then prehybridized in 5 x SSC,
20 mM Na2HPO4 (pH 7), 7% SDS, 10 x Denhardt's solution, and 100 Ig/ml
denatured salmon sperm DNA at 500 for 1-2 hours. Blots were hybridized in the
same solution with a 5'-phosphorylated 30-base oligonucleotide probe (New
England Nuclear) to alpha-tubulin at 500 for 4 hours. Blots were washed twice
in 3 x SSC, 25 mM Na2HPO4, 5% SDS and once in 1 x SSC, 1% SDS at 500
for 20 min each. Blots were then exposed to X-ray film with intensifying screens
for 4-7 days at -700. Autoradiographs were analyzed on an LKB densitometer.




Treatment of Cell Cultures with Drugs
C6 glioma cells grown on 100 mm plates with 10 ml media as described
above were treated with 1 iM PMA suspended in 10% acetone/90% HBSS.









Control cells were treated with 10% acetone/90% HBSS. After 1 hour, the
plates were washed with HEPES/EGTA buffer and the cells harvested for
experimentation as described above.
Primary neuronal cultures were equilibrated in serum-free DMEM for one
hour prior to treatment with opioids or harvest for adenylyl cyclase
measurements. For mRNA experiments, drugs were added at the indicated
times directly to the culture media. Total RNA was then isolated as described
above.




Opioid Receptor Binding
Opioid receptor binding was conducted in membranes from C6 glioma
cells, NG108-15 cells and primary neuronal cultures by modifications of
techniques developed for NG108-15 cells (Blume, 1978; Chang et al., 1978)
and brain (Childers & Snyder, 1979). Briefly, cells were removed from plates by
scraping, homogenized and membranes washed twice by centrifugation at
20,000 x g and resuspension in 50 mM Tris buffer (pH 7.4). After washing,
membrane pellet was resuspended in Tris buffer and dispensed into assay
tubes. Binding was determined in triplicate with specific binding calculated as
the difference between tubes with and without 10 IM levallorphan.














CHAPTER 3
OPIOID REGULATION OF ADENYLYL CYCLASE
IN RAT C6 GLIOMA CELLS


Introduction
Opioids are known to inhibit adenylyl cyclase in mammalian brain (Chou
et al., 1971) and several cell lines (Sharma et al., 1975a; Frey & Kebabian,
1984; Yu et al., 1986). The inhibition in rat brain has been characterized as a
delta response, because the enkephalin peptide analogues are more potent
than the opioid alkaloids (Law et al., 1981). However, the regional distribution of
this inhibition does not correspond well with the classical delta receptors
identified in binding studies (Childers, 1988). Also, studies utilizing specific
irreversible agonists and antagonists have shown that high-affinity delta
receptor binding sites, measured in the absence of sodium and GTP, cannot be
correlated with opioid-inhibited adenylyl cyclase in brain membranes (Nijssen
et al., submitted). Other studies have more closely examined the relationship of
opioid-inhibited adenylyl cyclase and opioid receptor binding. Childers (1988)
showed that although mu agonists inhibited adenylyl cyclase, they were
probably acting through the same receptors as the delta agonists. Studies in
NG108-15 cells, the cell line most commonly utilized for examination of opioid
coupled adenylyl cyclase, have also demonstrated delta receptors negatively-
coupled to adenylyl cyclase (Chang et al., 1978). However, as in rat brain, the
delta receptor identified in binding studies on NG108-15 cells has been shown
to be heterogeneous by detailed analysis of the effects of selective inactivation
of receptors on opioid-regulated adenylyl cyclase (Law et al., 1983a).









In addition to the effects on adenylyl cyclase, opioids have been shown
to block Ca2+ channels in NG108-15 cells that were differentiated with dibutyryl
cyclic AMP (Tsunoo et al., 1986; Shimahara & Icard-Liepkalns, 1987). This
action of opioids was shown to be mediated through a pertussis toxin-sensitive
G-protein, possibly Go (Hescheler et al., 1987). The opioid inhibition of adenylyl
cyclase in NG108-15 cells has been shown to be coupled through Gi2
(McKenzie & Milligan, 1990). This suggests that opioids, and other
neurotransmitters, may affect more than one second messenger in a single cell
through the same receptor, with one acting immediately, like an ion channel,
and the other producing more long term regulation of the cells function, like
cyclic AMP or phosphoinositol regulation of gene transcription. Makman et al.
(1988) have suggested that opioids stimulate adenylyl cyclase in spinal cord
dorsal root ganglion cells.
Although they represent one of the parent cell lines of NG108-15
neuroblastoma x glioma hybrid cells, C6 glioma cells have traditionally been
thought to lack any detectable opioid receptor sites (Klee & Nirenberg, 1974).
Indeed, for this reason, C6 glioma cells have been used as recipient cells in
opioid receptor expression studies. However, Tocque et al. (1984) showed that
pretreatment of these cells for 1 hour with 50 gM desmethylimipramine (DMI),
which produced desensitization of 1-adrenergic receptors, induced the
expression of opioid receptors and opioid-inhibited adenylyl cyclase. The
present study was designed to characterize the specificity of the opioid
receptors and opioid-inhibited adenylyl cyclase in these cells. Instead, these
results demonstrated a biphasic response of cyclic AMP accumulation in C6
glioma cells to opioid agonists, and suggests that under certain conditions
opioids can stimulate adenylyl cyclase.









Results
Effect of desmethylimipramine treatment on opioid receptors and
adenvlvl cvclase in C6 glioma cells. Initial experiments attempted to repeat the
findings of Tocque et al. (1984) to demonstrate DMI-induced opioid receptors in
C6 glioma cells. Cells were treated for 24 hours with 50 g.M DMI, as previously
reported. Unfortunately, there was no effect of DMI treatment on either opioid
binding sites measured with either the opioid antagonist [3H]-naloxone or the
opioid agonist [3H]-D-Ala2, Met5-enkephalinamide (Table 3-1). In fact, specific
opioid receptor binding was insignificant in both control and DMI-treated cells,
less than 17% of nonspecific binding in each case. However, when cyclic AMP
accumulation was examined, 10 pIM D-Ala2, Met5-enkephalinamide (D-Ala enk)
inhibited accumulation by 20% in treated and 18% in untreated C6 glioma cells.
Experiments in isolated membranes from untreated cells, assaying adenylyl
cyclase directly, confirmed these results: D-Ala enk inhibited 20-30% of basal
activity, with an IC50 of 40 nM (Figure 3-1). The finding of an opioid response in
the absence of detectable opioid receptor binding sites suggested that this was
either a non-specific response to D-Ala enk or that the opioid receptors were
present in such a low number that they were essentially undetectable in binding
assays.
Effect of D-Ala enk on cyclic AMP accumulation in intact C6 glioma cells.
To confirm the opioid nature of the D-ala enk effect on cyclic AMP accumulation,
the remaining experiments focused on the opioid-regulation of adenylyl cyclase
in non-DMI-treated cells. The dose response of cyclic AMP accumulation to D-
Ala enk is shown in Figure 3-2. Interestingly, in intact cells, D-ala enk produced
a biphasic response, with low (nM) concentrations of D-Ala enk producing
modest stimulation (30-50%) and IM concentrations inhibiting cyclic AMP
accumulation by 20-30%.














Table 3-1 Opioid receptor binding and adenylyl cyclase in C6 glioma cells.


A. Receptor Binding:


Total Binding, cpm


Non-Specific Binding,cpm


[3H]-Naloxone
Control
DMI

[3H]-D-Ala enk
Control
DMI


2086 98
2133 99


1422 53
1573 68


1888 121
2001 112


1201 72
1312 84


B. Adenylyl Cyclase:

Treatment Control DMI-Treated


Basal 19.6 + 0.6 15.7 0.8
D-Ala enk 15.7 0.5 12.9 0.7
D-Ala enk+Naloxone 18.5 0.9 15.3 1.2
(values in pmoles/min/mg)

C6 glioma cells were incubated for 24 hours in the presence or absence of 50
IM desmethylimipramine. Binding was performed on membranes with 2 nM
each of [3H]-Naloxone or [3H]-D-Ala enk. Adenylyl cyclase was assayed with 10
gM D-Ala enk and 1 g.M Naloxone. Values are the mean of triplicate
determinations S.D., n=3.


Ligand






33













110


100
Ch


-a 90-
0 9


80


70
0 .0001 .001 .01 .1 1 10 100
Conc. D-Ala enk (pM)




Figure 3-1 Effect of D-Ala enk on adenylyl cyclase activity in C6 glioma cells.

C6 glioma cell membranes were incubated with the indicated
concentrations of D-Ala enk for 10 min at 370. Determination of adenylyl cyclase
activity was performed as described under Material and Methods. Values are
the mean + S.D., n=3.


















140

130

120

n 110

I 100 T

0( 90

80

70
0 .001 .01 .1 1 10 100
Conc. D-Ala enk (IM)




Figure 3-2 Effect of D-Ala enk on cyclic AMP accumulation in C6 glioma cells.

C6 glioma cell suspensions were incubated with the indicated
concentrations of D-Ala enk for 10 min at 370. Determination of cyclic AMP
content was performed as described under Material and Methods. Values are
the mean + S.D., n=3.









Effect of inactivation of Gi/Go on ooioid regulation of cyclic AMP
accumulation. The biphasic nature of the opioid effect was investigated further
by attempting to isolate the stimulatory component of the opioid response. Two
methods were used to inactivate receptor inhibition of adenylyl cyclase. First,
pertussis toxin (islet-activating protein) was used to inactivate the Gi subunit and
block the inhibitory component of the response (Figure 3-3). D-Ala enk (10 glM)
inhibited cyclic AMP accumulation in untreated cells by 23%. However, when
C6 glioma cells were treated with 50 ng/ml pertussis toxin (islet-activating
protein) for 18 hours, this inhibition was lost. Instead, D-Ala enk stimulated
cyclic AMP accumulation by 53% above basal. Although several experiments
showed that pertussis toxin could block the inhibitory component of the D-Ala
enk effect, the response of these cells to pertussis toxin was extremely variable.
However, the ability of pertussis toxin to abolish the opioid inhibition of cyclic
AMP accumulation indicated that this action of opioids was mediated through a
Gi/oprotein.
Another method to inactivate Gj/Go utilized phorbol esters. Jakobs et al
(1987) reported that the phorbol ester, phorbol 12-myristate 13-acetate (PMA),
could increase stimulation of adenylyl cyclase in platelets by phosphorylating
and inactivating Gi. Thus, phorbol esters may be used in a manner analogous to
pertussis toxin. Figure 3-4 shows the response of C6 glioma cells to
pretreatment with 1 M PMA for 1 hour. The phorbol treatment decreased basal
cyclic AMP levels by 25-50%. This effect proved to be very reproducible and
was used as a control to evaluate the effectiveness of the PMA treatment. This
treatment also increased isoproterenol stimulation: in control cells,
isoproterenol stimulated 90%, while in PMA-treated cells, isoproterenol-
stimulated adenylyl cyclase by 250%. The response to opioids was also altered
(Figure 3-5). The inhibition (20%) observed at high opioid concentrations in the




















10.0-




5.0-


0.0-


[ Basal
F D-Ala enk


Control


Pertussis toxin


Figure 3-3 Effect of pertussis toxin on basal and D-Ala enk-regulated cyclic
AMP accumulation in C6 glioma cells.

C6 glioma cells were incubated with 50 ng/ml pertussis toxin for 18
hours. Cells were harvested and suspensions incubated for 10 min at 370 with
or without 10 pM D-Ala enk (B). Determination of cyclic AMP content was
performed as described under Material and Methods. Values are the mean +
S.D., n=3.


















5.0-


4.0-


0.0 ---


i Basal
E Isoproterenol


-"-


Control PMA


Figure 3-4 Effect of PMA-pretreatment on basal and isoproterenol-stimulated
cyclic AMP accumulation in C6 glioma cells.

C6 glioma cells were incubated with 1 I.M PMA or vehicle for 1 hour at
370. Cells were harvested and suspensions incubated for 10 min at 370 with or
without isoproterenol (10 iM). Determination of cyclic AMP content was
performed as described under Material and Methods. Values are the mean
S.D., n=3.






38











130
-0 Control
120 -*- PMA Treated

110

C 100-

CO
T 90

2 80-

70

60

50 I... ...... ....
0 .01 .1 1 10
Conc. D-Ala enk (p.M)



Figure 3-5 Effect of PMA-pretreatment on opioid regulation of cyclic AMP
accumulation in C6 glioma cells.

C6 glioma cells were incubated with 1 lgM PMA or vehicle for 1 hour at
370. Cells were harvested and suspensions incubated for 10 min at 370 with the
indicated concentrations of D-Ala enk. Determination of cyclic AMP content was
performed as described under Material and Methods. Values are the mean +
S.D., n=3.








control cells was lost by PMA-pretreatment, and D-Ala enk stimulated cyclic
AMP accumulation by 60% in the PMA-treated cells. Since the phorbol effect
was more reproducible than the pertussis toxin effect, the phorbol treatment was
selected for further experiments to characterize this response.
Specificity of opioid stimulation of cyclic AMP accumulation. To identify
which type of opioid receptor was mediating the stimulation, several different
opioid agonists were screened for activity in PMA-treated cells (Figure 3-6). The
delta peptides, D-Pen2,5 enkephalin and D-Ser2, Leus-enkephalin-Thr6
(DPDPE and DSLET, respectively), were relatively ineffective. However, the mu
agonist D-Ala2, enkephalin-Gly5-ol (DAGO) was as effective as D-Ala enk in
stimulating cyclic AMP accumulation (110-130%). The non-selective ligand, 8-
endorphin, although it did stimulate cyclic AMP accumulation by 80%, was not
as effective as either D-Ala enk or DAGO, suggesting that it was acting through
its mu component. When D-Ala enk and DAGO were directly compared, DAGO
was found to be equipotent with D-Ala enk (Figure 3-7) at stimulating cyclic
AMP accumulation in C6 glioma cells (EC50 values of -10nM). This suggested
that the stimulation in PMA-treated C6 glioma cells may be a mu response.
Effect of the opioid antagonist naloxone on opioid-stimulation of cyclic
AMP accumulation. Other experiments tested the opioid nature of this response
with specific opioid antagonists. The DAGO response was not completely
blocked by naloxone; however, 0.1 nM naloxone shifted the EC50 of DAGO to
the right by a factor of 80 (Figure 3-8). Interestingly, naloxone alone increased
the basal cyclic AMP accumulation in C6 glioma cells at relatively low
concentrations (Figure 3-9). Naloxone was able to stimulate by >100%, similar
to the stimulation produced by agonists. Naloxone was a potent stimulator of
cyclic AMP levels, with an EC50 of 5 nM, which was similar to that of the
agonists D-Ala enk and DAGO. This suggests that naloxone may be a mixed


















250


200


0 150


(D 100


50


0 I
D-alaenk DSLET DPDPE DAGO B-end
Agonist




Figure 3-6 Selectivity of opioid stimulation of cyclic AMP accumulation in
PMA-pretreated C6 glioma cells.

C6 glioma cells were incubated with 1 gM PMA for 1 hour at 370. Cells
were harvested and suspensions incubated for 10 min at 370 with the indicated
agonists (1 gM). Determination of cyclic AMP content was performed as
described under Material and Methods. Values are the mean S.D., n=3.




















D-Ala enk
--- DAGO


.001


Conc. Opioid Peptide (gM)


Figure 3-7 Comparison of D-Ala enk and DAGO to stimulate cyclic AMP
accumulation in C6 glioma cells.

C6 glioma cells were incubated with 1 IrM PMA for 1 hour at 370. Cells
were harvested and suspensions incubated for 10 min at 370 with the indicated
concentrations of either D-Ala enk or DAGO. Determination of cyclic AMP
content was performed as described under Material and Methods. Values are
the mean + S.D., n=3.


250-


200-


150-


100-


50-


0


"I """'" """"' I


B



















160 ----- DAGO
0 with Naloxone


140
(n

120

a,
100-


80 -
0 .001 .01 .1 1 10

Conc. DAGO (gM)



Figure 3-8 Effect of naloxone on DAGO-stimulated cyclic AMP accumulation
in PMA-pretreated C6 glioma cells.

C6 glioma cells were incubated with 1 gM PMA for 1 hour at 370. Cells
were harvested and suspensions incubated for 10 min at 370 with the indicated
concentrations of DAGO and 0.1 nM naloxone. Determination of cyclic AMP
content was performed as described under Material and Methods. Values are
the mean S.D., n=3.






43











220

200

180

160

S140



100 *-

80 -- -.. .-* 1 f i
0 .1 1 10 100 1000 10000
Conc. Naloxone (nM)



Figure 3-9 Effect of naloxone on cyclic AMP accumulation in PMA-pretreated
C6 glioma cells.

C6 glioma cells were incubated with 1 gM PMA for 1 hour at 370. Cells
were harvested and suspensions incubated for 10 min at 370 with the indicated
concentrations of naloxone. Determination of cyclic AMP content was performed
as described under Material and Methods. Values are the mean S.D., n=3.









agonist-antagonist in these cells, as has been suggested in other cell culture
systems (Makman et al., 1988). Together with the nM response to DAGO and D-
Ala enk, these data suggest that this is a very efficiently coupled system.


Discussion
This study initially started to examine the regulation of opioid receptors
and opioid-inhibited adenylyl cyclase in a unique system which did not normally
express opioid receptors at a detectable level. However, the data obtained from
both opioid binding and adenylyl cyclase studies quickly redirected this study to
examine the opioid-stimulation of adenylyl cyclase in the C6 glioma cells. The
absence of detectable opioid binding sites suggested that either the opioid
effects on cyclic AMP accumulation were a non-specific action or that the
number or the affinity of the receptors on these cells were not high enough to be
detected in the binding assay utilized. The former was ruled out due to the high
potencies of opioids in regulating cyclic AMP accumulation in these cells.

With the lack of dectectable binding sites, the biphasic nature of the
opioid regulation of cyclic AMP accumulation was investigated. The ability of
pertussis toxin to block the inhibitory component suggested that this effect of
opioids was mediated through a G/o protein. However, the action of pertussis
toxin was slow, unpredictable and expensive, and another method of
inactivating the inhibitory component of the opioid response was found when
Jakobs et al. (1987) reported that PMA inactivated Gi by phosphorylation in
platelets. PMA treatment of C6 glioma cells also inactivated Gi in a manner
analagus to pertussis toxin. This was demonstrated by the increase in
isoproterenol stimulation of cyclic AMP accumulation in cells treated with PMA.
Pretreatment of C6 cells with PMA eliminated the inhibitory action of the opioids
and increased the stimulation of cyclic AMP accumulation.









The agonist specificity profile of the stimulation of cyclic AMP
accumulation in PMA-treated cells suggested that this was a mu-type response
because DAGO was equipotent with D-Ala enk while the delta-selective
peptides (DPDPE and DSLET) were ineffective. 13-endorphin, which binds to all
opioid receptor types equally well, did stimulate cyclic AMP accumulation, but
not as well as either DAGO or D-Ala enk, suggesting that it was acting through
its mu component.
The action of the opioid antagonist naloxone on cyclic AMP accumulation
proved to be unexpected. The inability of naloxone to completely block the
opioid response was troublesome until the partial agonism of naloxone in this
system was realized. The finding of agonist-like properties for naloxone is not
unique however. Makman et al. (1988) found that naloxone (2 4M) stimulated
adenylyl cyclase activity in dorsal root ganglion (DRG) explants to the same
level as the opioid agonist levorphanol. Interestingly, levorphanol had dual
effect on adenylyl cyclase in these cultures with inhibition of forskolin-stimulated
adenylyl cyclase activity and stimulation of basal adenylyl cyclase. This
stimulatory effect of opioids was enhanced by treatment of the cultures with
pertussis toxin or with morphine. These results are consistent with the opioid
effects observed in C6 glioma cells.
The presence of opioid-stimulated adenylyl cyclase in rat C6 glioma cells
suggests that opioids may dually regulate adenylyl cyclase in other tissues.
Other researchers have reported excitatory effects of opioid in the spinal cord.
Pohl et al. (1989) have demonstrated that mu-opioid agonists cause a
naloxone-reversible enhancement of capsaicin-induced release of substance P
from primary afferent fibers. Also, Xu et al. (1989) found a bimodal response of
evoked enkephalin release to opioids in enteric ganglia. In this preparation, low
(nM) concentrations of opioids enhanced enkephalin release while higher






46


concentrations decreased release. While the mechanism of this bimodal effect

is not clear, cyclic AMP may be involved since treatment of these cells with

forskolin mimicked low concentrations of opioids in stimulating enkephalin

release.














CHAPTER 4
OPIOID REGULATION OF ADENYLYL CYCLASE
IN RAT PRIMARY NEURONAL CULTURES


Introduction
The previous study examined opioid modulation of adenylyl cyclase in a
transformed cell line. This system offered a simple model in which individual
components could be manipulated. However, the opioid-coupled adenylyl
cyclase differed significantly from rat brain. The presence of dual modulation of
adenylyl cyclase by opioids in C6 glioma cells was intriguing, but it presented a
complex model for answering the question of opioid-coupled adenylyl cyclase
regulation of proenkephalin expression. The development of primary neuronal
cultures has offered accessibility to isolated cells as in transformed cell culture
systems, with the additional advantage of having nontransformed cells.
Moreover, glial and neuronal cells can be studied separately.
The presence of opioid binding sites in embryonic and neonatal mouse
and rat brain has been well documented. Mu and kappa opioid receptors are
both present at birth and reach adult levels at similar times (Leslie et al., 1982;
Barr et al., 1985; Spain et al., 1985; Tavani et al., 1985; Petrillo et al., 1987).
However, delta opioid receptors do not develop until later (Milligan et al., 1987).
Not surprisingly then, only mu and kappa opioid binding sites were found on rat
embryonic striatal neuronal cultures (Vayesse et al., 1990). This study also
demonstrated that no opioid binding sites existed on cultured astrocytes.
Chneiwiess et al., (1988) demonstrated opioid-inhibited adenylyl cyclase in









membranes from mouse embryonic striatal neuronal cultures. Interestingly,
there was no discrimination between delta and mu agonists.
Rat primary neuronal cultures were chosen to study for two reasons. First,
these are differentiated cells and therefore the responses observed should
resemble those in rat brain more closely than a transformed cell line. Second,
the ability to manipulate the individual components each experiment is greatly
enhanced in cell culture versus intact rat brain. This chapter demonstrates
opioid-regulation of adenylyl cyclase in primary rat neuronal cultures, which is
the first portion of the aims of this dissertation.


Results
Opioid effects on cyclic AMP accumulation in primary neuronal cultures
derived from forebrain. midbrain and hindbrain. The adenylyl cyclase response
to isoproterenol and the opioid peptide, D-Ala enk, was compared between
cultures derived from forebrain, midbrain and hindbrain of neonatal rat. As
demonstrated in Figure 4-1, isoproterenol (10 pM) stimulated cyclic AMP
accumulation two-, ten-, and threefold in cultures from forebrain, midbrain, and
hindbrain, respectively. This indicated that receptor-stimulated adenylyl cyclase
was functional in all cultures. The opioid agonist, D-Ala enk (10 giM), inhibited
cyclic AMP accumulation in forebrain cultures by 70% and by 50% in hindbrain
cultures. However, the activity seen in hindbrain cultures was highly variable.
Because the regions in hindbrain which possess opioid-inhibited adenylyl
cyclase in adult brain are discrete nuclei in the brain stem, one explanation for
this variability was the uneven distribution of cells from these nuclei in hindbrain
cultures. Primary neuronal cultures from midbrain did not show any activity of
the opioid (93% of basal). Due to the large inhibition seen in the forebrain
cultures, this region was selected for further study. The finding that the largest


















100-


IE Basal
* D-Ala-Enk
H Isoproterenol


FOREBRAIN


MIDBRAIN


HINDBRAIN


NEURONAL CULTURE REGION





Figure 4-1 Effect of D-Ala enk and isoproterenol on cyclic AMP accumulation
in primary neuronal cultures from forebrain, midbrain and hindbrain of neonatal
rat.

Neuronal cultures derived from neonatal rat forebrain, midbrain and
hindbrain were prepared as described under Material and Methods Cyclic AMP
accumulation was determined by incubating cultures in HBSS buffer containing
10 gM D-Ala enk or 10 gM isoproterenol at 370 for 10 min. Extraction and
determination of cyclic AMP content was performed as described under Material
and Methods. Values are the mean S.D., n=3.


0 +--









inhibition of adenylyl cyclase by opioids was in the forebrain cultures was
expected, since this area includes the striatum which contains the largest
amount of opioid-inhibited adenylyl cyclase in adult brain (Childers, 1988).
If this action was mediated through a G-protein-linked receptor negatively
coupled to adenylyl cyclase, then pretreatment of the cultures with pertussis
toxin should abolish the opioid inhibition of cyclic AMP accumulation.
Incubation of cultures with 100 ng/ml pertussis toxin for 18 hours (Figure 4-2)
increased isoproterenol-stimulated cyclic AMP accumulation. The ability of D-

Ala enk to inhibit cyclic AMP accumulation in forebrain neuronal cultures (68%)
was abolished in pertussis toxin-treated cultures (114%). This suggested that
opioid receptors were negatively coupled to adenylyl cyclase in forebrain
neuronal cultures through a Gi protein as in NG108-15 cells (Hsia et al., 1984)
and rat striatum (Law et al., 1982; Cooper et al., 1982; Childers et al., 1986).
Opioid regulation of cyclic AMP accumulation in primary rat neuronal
cultures. To characterize the opioid receptor type which mediated this
response, the cyclic AMP accumulation response in intact forebrain neuronal
cultures to various opioid agonists were examined. Although the agonists were
utilized at large concentrations (10 gM), the effects of opioids on forskolin-
stimulated cyclic AMP accumulation varied greatly in neuronal cultures (Figure
4-3). The endogenous opioid peptides, dynorphin A and dynorphin A1-13,
inhibited cyclic AMP accumulation by 40% and 60%, respectively. DAGO and
morphine stimulated cyclic AMP accumulation in forebrain cultures (50 and
75%, respectively). Naloxone (10 gM) was able to reverse both the stimulated
and the inhibited cyclic AMP accumulation in intact cells. The delta agonist,
DPDPE, stimulated cyclic AMP accumulation by 40%; however, this effect was
unaffected by 10 gM naloxone, suggesting that this could be a non-specific
effect. In fact, studies of opioid receptors in embryonic neuronal cultures have















SControl
E Pertussis toxin


"-i


Basal


D-Ala-Enkephalinamide Isoproterenol


Figure 4-2 Effect of pertussis toxin on D-Ala enk- and isoproterenol-regulated
cyclic AMP accumulation in forebrain neuronal cultures.
Neuronal cultures derived from neonatal forebrain were incubated with
100 ng/ml pertussis toxin for 18 hours. Cyclic AMP accumulation was
determined by incubating cultures in HBSS buffer containing 10 AIM D-Ala enk
or 10 pM isoproterenol at 370 for 10 min. Extraction and determination of cyclic
AMP content was performed as described under Material and Methods. Values
are the mean S.D., n=3.


T


o---















SOpioid
U With 10pM Naloxone


TT


BASAL DYN 1-13 DYN A DPEN


DAGO MORPHINE


Figure 4-3 Effect of various opioid agonists on cyclic AMP accumulation in
forebrain neuronal cultures.

Neuronal cultures derived from neonatal rat forebrain were prepared as
described under Material and Methods Cyclic AMP accumulation was
determined by incubating cultures in HBSS buffer containing 1 RM forskolin and
10 gIM of the indicated opioid agonist and 10 giM naloxone at 370 for 10 min.
Extraction and determination of cyclic AMP content was performed as described
under Material and Methods. Values are the mean S.D., n=3.


200-


150-



100-









failed to find any delta binding sites in these model systems (Vayesse et al.,
1990).
The diverse actions of morphine and dynorphin were surprising since at
10 iM, both agonists should activate all opioid receptor types. However, these
compounds produced markedly different effects on cyclic AMP accumulation in
these cultures. These two actions of opioids on cyclic AMP accumulation were
explored separately to further characterize the opioid regulation of adenylyl
cyclase in these cultures.
Effect of morphine on adenylyl cvclase. To determine whether the ability
of morphine to stimulate cyclic AMP accumulation was a nonspecific effect or
possibly an indirect action of mu opioids on adenylyl cyclase through a different
second messenger system, morphine's action on adenylyl cyclase activity in
isolated membranes from forebrain neuronal cultures was investigated. As
shown in Figure 4-4, morphine did not affect adenylyl cyclase activity at any
concentration examined. This suggested that morphine had no direct action on
adenylyl cyclase in these cultures and the stimulation of cyclic AMP
accumulation observed was probably mediated by another system and
indirectly affected cyclic AMP accumulation in these cultures.
Effect of opioid agonists on adenylyl cvclase activity in primary neuronal.
To demonstrate that the dynorphin regulation of cyclic AMP accumulation was a
direct effect on adenylyl cyclase, dynorphin A1-13 activity in membranes
prepared from forebrain neuronal cultures was examined. Dynorphin A1-13
inhibited forskolin-stimulated adenylyl cyclase activity by 15% in a dose
dependent manner in membranes from primary neuronal cultures (Figure 4-5).
This indicated that the opioid-inhibition of adenylyl cyclase in these cultures
was a direct action on the enzyme and not mediated through another second
messenger system. It is interesting to note that dynorphin A1-13 in these cultures

















30-


25


20


15


Fhi~-


0


.01


100


Morphine (gpM)






Figure 4-4 Effect of morphine on adenylyl cyclase activity in membranes from
forebrain neuronal cultures.
Adenylyl cyclase activity was determined by incubating membranes from
forebrain neuronal cultures in Tris-Mg buffer containing the indicated
concentration of morphine at 370 for 10 min. The standard cyclase components
and the determination of the cyclic AMP content were as described under
Materials and Methods. Values are the mean S.D., n=3.


'''I r -


"-II



















75


E 70


o
OL 65-


C-

55-


50
5 0 -- -. .... .... .... ....
0 .01 .1 1 10 100

Concentration Dynorphin A1-13 (iM)







Figure 4-5 Effect of dynorphin A1-13 on adenylyl cyclase activity in forebrain
neuronal cultures.

Adenylyl cyclase activity was determined by incubating membranes from
forebrain neuronal cultures in Tris-Mg buffer containing the indicated
concentrations of dynorphin A at 370 for 10 min. The standard cyclase
components and the determination of the cyclic AMP content were as described
under Materials and Methods. Values are the mean S.D., n=3.









displays an ICso of ~1 i.M, which is nearly identical to the ICso reported for
dynorphin A1-13 inhibited adenylyl cyclase in striatal membranes (0.7 RM)
(Childers, 1988).
Oioiid regulation of adenylvl cyclase in primary glial cultures. The
possibility that the opioid activity could be mediated through the small glial
population in the neuronal cultures was explored by examining the ability of
opioid agonists to regulate adenylyl cyclase in membranes from primary glial
cultures. As shown in Table 4-1, none of the opioid agonists examined (which
included kappa-, delta- and mu-agonists) showed any significant effect on
adenylyl cyclase activity in glial culture membranes. This was not surprising as
most studies of opioid receptors have failed to find any opioid binding in
membranes from primary glial cultures (McCarthy & deVellis, 1978; Van Calker
& Hamprecht, 1980). These findings suggested that the opioid inhibition of
adenylyl cyclase was on the neuronal constituent of the cultures.


Discussion
This study demonstrated that opioids regulate adenylyl cyclase in
forebrain neuronal cultures. Cyclic AMP accumulation was inhibited by opioid
agonists, particularly dynorphin A1-13 (Figure 4-3). This action was naloxone
reversible and pertussis toxin-sensitive (Figures 4-4 & 4-2, respectively),
suggesting that this was a classical opioid receptor coupled to adenylyl cyclase
via a G/Go protein. To further substantiate that this action was due to a receptor
coupled directly to adenylyl cyclase, the ability of dynorphin A1-13 to inhibit
adenylyl cyclase activity in membranes of these cultures was examined.
Dynorphin A1-13 inhibited activity in membranes by 15% (Figure 4-6), indicating
that the action seen in intact cells was direct and not an effect of other second
messenger systems on adenylyl cyclase. This is comparable to the inhibition













Table 4-1 Effect of opioid agonists on forskolin-stimulated adenylyl cyclase
activity in glial culture membranes.


Percent of Forskolin-Stimulated
Adenylyl Cyclase Activity


Forskolin (1I.M)


+ D-Ala enk
+ DADLE
+ DPDPE
+ Dyn 1-13
+ U 50,448
+ Morphine


100.0 5.1
92.8 13.1
98.3 3.3
96.7 6.9
92.0 10.7
103.6 5.3
95.6 2.4


Adenylyl cyclase activity was determined by incubating membranes from
primary glial cultures in Tris-Mg buffer containing 10 g.M of the indicated opioid
agonists at 370 for 10 min. The standard cyclase components and the
determination of the cyclic AMP content were as described under Materials and
Methods. Values are the mean S.D., n=3.


Agonist
(10 PM)









seen in membranes from rat striatum (20-30%) and NG108-15 cells (35-45%).
The lower activity of the opioids observed in the primary cultures could be due
to the heterogeneous nature of the cultures.
The primary neuronal culture method utilized in this study yields cultures
that are approximately 70-80% neurons, with the remaining cells being glial
(Raizada et al., 1984). To establish whether the opioid action on adenylyl
cyclase was due to the glial component of these cultures, primary glial cultures
were prepared and examined for opioid activity on adenylyl cyclase. These
cultures displayed no opioid action on adenylyl cyclase activity of any opioid
agonist examined, including delta-, mu-, and kappa-selective agonists. This
suggested that the actions of opioids on adenylyl cyclase were on the neuronal
population of these cultures. Moreover, this finding was in agreement with
binding studies which found no opioid binding sites on glial cultures (Vayesse
et al.,1990). However, one report of opioid binding sites in cultured glial cells
from embryonic chick brain has been reported (Maderspach & Solomonia,
1988). However, in that study, the equilibrium binding for [3H]-naloxone was not
saturable so that the significance of those results are not clear.
The mu agonists, DAGO and morphine, stimulated cyclic AMP
accumulation in forebrain cultures. In intact cultures, morphine stimulated cyclic
AMP accumulation 11-fold over basal. However, when this action was
examined in membranes, morphine did not exhibit any action on adenylyl
cyclase activity. There are several possible explanations for activity seen in
intact cells which do not appear in membranes. Mu receptors could be coupled
to an ion channel or to phosphoinositol turnover, both of which could indirectly
increase cyclic AMP accumulation in intact cells but not would not effect any
changes in membranes. In addition, the opioids could modulate one cell which
could alter the release of another neurotransmitter which could regulate









adenylyl cyclase. Because this study was done in the presence of the
phosphodiesterase inhibitor, IBMX, the opioids are probably not increasing
cyclic AMP by decreasing the activity of phosphodiesterase.
Although the opioid stimulation of adenylyl cyclase in this culture system
appears to be an indirect effect, there have been other reports of excitatory
effects of opioids. Our own data in the C6 glioma cell line suggests that, at least
in this transformed cell line, opioids can stimulate adenylyl cyclase. Makman et
al. (1988) in spinal cord-dorsal root ganglion explants, established the
presence of opioid-stimulated adenylyl cyclase in non-transformed neural cells.
Interestingly, the opioid regulation of adenylyl cyclase in these explants was
biphasic in a manner similar to that seen in C6 glioma cells. A study in guinea

pig myenteric plexus demonstrated that low concentrations (0.1-10 nM) of
dynorphin A enhanced electrically induced Met-enkephalin release, whereas
100 nM dynorphin A decreased this induced release (Xu et al., 1989). These
studies indicate that although opioids have traditionally been thought to be
purely inhibitory, they possess excitatory properties in diverse tissue.
Although in adult rat brain, delta-inhibited adenylyl cyclase is the most
prominent, delta receptors are the last opioid receptors to develop (Petrillo et
al., 1987). In fact, delta receptors were reported to be almost undetectable at
birth (Milligan et al., 1987). Indeed, binding studies in primary neuronal cultures
of embryonic rat brain demonstrated mu and kappa binding sites but no
detectable delta sites (Vayesse et al., 1990). Therefore, the lack of delta opioid
activity in these cultures was not surprising.
The lack of activity by mu agonists in isolated membranes from these
cells may suggest that mu receptors are not directly coupled to adenylyl cyclase
in these cells, or it may suggest that the mu-inhibited adenylyl cyclase is too low
to detect in these heterogeneous cells. The latter suggestion is supported by the









finding that the only significant opioid inhibition occurred with dynorphin.
Although at low concentrations, dynorphin selectively binds to kappa receptors,
the pM concentrations of dynorphin used in these assays would bind to all
opioid receptor types. Kappa opioid inhibition of adenylyl cyclase has been
demonstrated in guinea pig cerebellum (Konkoy & Childers, 1989), a tissue
devoid of any other opioid receptor subtypes. However, the potency of
dynorphin A in these cultures was the same as that determined in rat striatum
(Childers, 1988), and 20 times less potent than the IC50 value in kappa-
inhibited adenylyl cyclase in guinea pig cerebellum (Konkoy & Childers, 1989).
In fact, in adult rat brain, true kappa-selective inhibition of adenylyl cyclase
cannot be detected (Childers, 1988). Regardless of its receptor specificity, the
demonstration of dynorphin-inhibited adenylyl cyclase establishes the presence
of opioid-regulated adenylyl cyclase in these cultures. This establishes that
these cultures are suitable for study of opioid-regulation of proenkephalin
mRNA via adenylyl cyclase regulation.














CHAPTER 5
OPIOID REGULATION OF PROENKEPHALIN EXPRESSION


Introduction
The previous study established the presence of opioid-inhibited adenylyl

cyclase in primary neuronal cultures from neonatal rat forebrain. Although the
observed effect was small, it was comparable to that seen in rat brain and other

primary culture systems (Childers, 1988; Chneiweiss et al., 1988). One

unsolved question is the biological functions of opioid-inhibited adenylyl
cyclase in neurons. Although Crain et al. (1990) have suggested that
electrophysiological effects of opioids may be mediated through cyclic AMP, a
number of other studies have shown that opioid receptors are coupled directly
to ion channels through G-proteins, and these actions are not mediated through
cyclic AMP (Gross et al., 1990). North et al. (1986) suggested that opioid-
inhibited adenylyl cyclase may be important in long-term regulatory functions in
neurons.

The biological effect of opioid inhibition of adenylyl cyclase is suspected
to be inhibition of protein kinase A, which would decrease the phosphorylation
state of protein kinase A substrates. Nestler et al. (1988) have identified a
protein in rat brain whose phosphorylation level is decreased following opioid
treatment. Other experiments (Fleming & Childers, submitted) have identified at
least two proteins whose phosphorylation is attenuated by opioid-inhibited
adenylyl cyclase in rat brain membranes. The cyclic AMP responsive element
binding protein (CREB) for the somatostatin gene has been shown to be









phosphorylated by protein kinase A (Montminy & Bilezikjian, 1987) and protein
kinase A activity is required for transcriptional regulation of eukaryotic genes by
cyclic AMP (Montminy et al., 1986).
Interestingly, there have been at least 10 CREB cDNAs cloned or
isolated, and each appears to come from distinct genes (Habner, 1990). All
exhibit structural similarities, but all proteins do not bind to the CREs of the
same genes. Thus, the genes which are regulated by cyclic AMP can vary
between cells, depending on the CREB protein that is expressed in each
particular cell. This family of DNA binding proteins is very similar to fos/jun
family of proto-oncogenes. Both exhibit a leucine zipper domain which allows
dimerization of the proteins. Indeed, the fos/jun family is now known to activate
transcription by binding to the phorbol-ester responsive element (TRE), of which
the core sequence differs from that of the CRE by only one base deletion
(Deutsch et al., 1988). These two families of DNA-binding proteins provide
insight to the cell-specific control of expression of the multitude of genes whose
regulation is mediated by the cyclic AMP and diacylglycerol messengers.
Expression of proenkephalin mRNA has been demonstrated in rat
embryonic striatal cultures (Schwartz & Simantov, 1988) and in glial cultures
(Vilijn et al., 1988; Melner et al., 1990). Opioids have been shown to increase
proenkephalin mRNA in NG108-15 cells after 3 days (Schwartz, 1988).
However, there has been no published study of the relationship between
opioid-inhibited adenylyl cyclase and the cyclic AMP regulation of
proenkephalin expression. In this study this relationship is examined in rat
primary neuronal cultures from neonatal brain. This system was chosen
because 1) these are non-transformed cells where individual treatment
conditions can be manipulated and the hypothesis tested directly; 2) these cells
contain opioid-inhibited adenylyl cyclase; and 3) these cultures express









proenkephalin mRNA at a detectable level. This study demonstrates that
opioids regulate proenkephalin expression and this action is related to
inhibition of adenylyl cyclase.


Results
Proenkephalin and alpha-tubulin mRNA in primary neuronal cultures.
The riboprobe utilized in this study has been characterized in C6 glioma cells
and NG108-15 cells (Yoshikawa & Sabol, 1986a, 1986b). The specific
hybridization to proenkephalin mRNA is illustrated by Northern blot analysis in
Figure 5-1. Lanes 2 and 3 are total RNA isolated from hypothalamus and
striatum from rat brain. The lower bands are proenkephalin mRNA and the
upper bands are alpha-tubulin mRNA (hybridized with an oligonucleotide
probe). The proenkephalin expression in striatum was 20-fold higher than that
in hypothalamus, as has been reported (Pittius et al., 1985) The presence of
proenkephalin mRNA in forebrain neuronal cultures was demonstrated in lane
1 (Figure 5-1). Proenkephalin mRNA from primary neuronal cultures migrated
as a single band, -1.5 kb, the same size as that from rat brain. In addition, the
alpha-tubulin mRNA (upper band) migrated the same (-1.8 kb) as that observed
in rat brain. This indicated that the proenkephalin mRNA from neonatal rat
primary neuronal cultures was the similar to that in rat brain.
Expression of proenkephalin mRNA in primary neuronal cultures. The
level of proenkephalin mRNA in primary forebrain cultures was determined in
cultures after the two day treatment with 10 g.M cytosine arabinoside to remove
most of the glial cells. Proenkephalin mRNA levels gradually increased during
the time of incubation (Figure 5-2). The densitometric values for Figure 5-2 are
presented in Table 5-1. This increase after 12 days in culture was probably due
to proliferation of the remaining glial cells in the cultures. However, there was





























a-tubulin I a-tubulin
PE PE


1 2 3


Figure 5-1 Northern blot analysis of proenkephalin and alpha-tubulin mRNA
in primary neuronal cultures and rat brain.

Primary rat forebrain neuronal cultures were prepared as described in
Materials and Methods. Total cellular RNA was extracted from cultures and rat
brain regions as described (Chomoczynski and Sacchi, 1984), resolved by
electrophoresis on a 1.5% agarose denaturing gel, blotted onto Zeta-Probe
membrane (Bio-Rad) and hybridized with a 950-base riboprobe to
proenkephalin and a 30-base oligonucleotide probe to alpha-tubulin. Lanes are
total RNA from: 1) forebrain neuronal culture, 2) hypothalamus and 3) striatum
from rat brain. Rat brain RNA was quantitated by O.D. 260 nm and 5 ig loaded
per lane.


























a-tubulin


SPE


2 5 8 10 12 14 18 20 days



Figure 5-2 Expression of proenkephalin mRNA in forebrain neuronal cultures.

Primary rat forebrain neuronal cultures were prepared as described in
Materials and Methods. After replacing media to that without cytosine
arabinoside, cultures were maintained for the indicated times. Total cellular
RNA was extracted as described (Chomoczynski and Sacchi, 1984), resolved
by electrophoresis on a 1.5% agarose denaturing gel, blotted onto Zeta-Probe
membrane (Bio-Rad) and hybridized sequentially with a 950-base riboprobe to
proenkephalin and a.30 base oligonucleotide probe to alpha-tubulin. Similar
results were observed in another preparation of primary neuronal cultures.
















Table 5-1 Densitometric quantitation for expression of proenkephalin mRNA
in forebrain neuronal cultures.


Days after removal PE units Percent of day two
of cytosine arabinoside


2 2.0 100
5 1.8 90
8 3.2 160
10 3.3 170
12 3.0 150
16 3.8 190
18 4.1 210
20 4.6 230


The values listed above were calculated from the densitometric values obtained
from the northern blot in Figure 5-2 and represent proenkephalin mRNA
absorbance relative to alpha-tubulin mRNA absorbance. The percent values
were calculated with day 2 equal to 100%.









relatively little change between 8 and 16 days of incubation. Due to the
consistent expression prior to 2 weeks, all experiments were conducted on the
cultures 10-15 days after removal of the cytosine arabinoside.
Effect of cyclic AMP on proenkephalin mRNA levels in forebrain neuronal
cultures. Proenkephalin transcription is known to be induced by cyclic AMP
through activation of protein kinase A (Grove et al., 1987). The forebrain
neuronal cultures used in the present study have been shown to display
receptor-coupled adenylyl cyclase (Chapter 4). To ensure that these cells
contained the necessary machinery to regulate gene expression through cyclic
AMP and that the proenkephalin mRNA was cyclic AMP sensitive, forebrain
neuronal cultures were incubated with 1 p.M forskolin for various times. This
concentration of forskolin stimulated cyclic AMP in these cells by approximately
2-fold (not shown). Proenkephalin message levels increased two-fold after 4
hours, and four-fold in 24 hours, indicating that the proenkephalin mRNA in
these cultures was inducible by cyclic AMP (Figure 5-3). Cultures incubated for
20 hours with different concentrations of dibutyryl cyclic AMP, an analogue of
cyclic AMP which penetrates intact cells, showed a dose-dependent increase in
proenkephalin message (Figure 5-4). Interestingly, 600 pIM dibutyryl cyclic AMP
was required to significantly (4.3-fold) induce proenkephalin expression. This
was intriguing, as this is the threshold concentration to differentiate NG108-15
cells (Hamprecht, 1985). Forskolin induction was also dose-dependent, with 1
pM stimulating proenkephalin mRNA >five-fold over control levels in 20 hours.
These data correspond with previous studies on the induction of proenkephalin
mRNA in NG108-15 and C6 glioma cells (Yoshikawa & Sabol, 1986a,b)
Effect of dvnorphin Al-1 and naloxone on proenkeDhalin mRNA levels.
Previous studies (Chapter 4) demonstrated the presence of dynorphin-inhibited
and isoproterenol-stimulated adenylyl cyclase. To examine receptor-mediated








Figure 5-3 Time course of forskolin induction of proenkephalin mRNA levels
in forebrain neuronal cultures.

Forskolin (1pIM) was added to media on primary neuronal cultures and
incubated for the indicated times. Total cellular RNA was extracted as described
(Chomoczynski and Sacchi, 1984), resolved by electrophoresis on a 1.5%
agarose denaturing gel, blotted onto Zeta-Probe membrane (Bio-Rad) and
hybridized with a 950-base riboprobe to proenkephalin. Densitometric
quantitation for Northern blot shown in B is illustrated in A. Results are from one
representative experiment which was repeated twice.



















C:


8 4.0
CO
0
/.)
< 2.0


0 2 4 8 24
Hours


0 2 4 8 24 Hours








Figure 5-4 Effect of dibutyryl cyclic AMP and forskolin on proenkephalin
mRNA levels in forebrain neuronal cultures.

Dibutyryl cyclic AMP or forskolin were added to media on primary
neuronal cultures in the indicated concentrations and incubated for 20 hours.
Total cellular RNA was extracted as described (Chomoczynski and Sacchi,
1984), resolved by electrophoresis on a 1.5% agarose denaturing gel, blotted
onto Zeta-Probe membrane (Bio-Rad) and hybridized with a 950-base
riboprobe to proenkephalin. Densitometric quantitation of Northern blots shown
in C for dibutyryl cyclic AMP (A) and forskolin (B). Results are from one
representative experiment which was repeated twice.




















3.0



2.0-


0 6 60 600
Dibutyryl cyclic AMP (gM)


0 6 60 600
Dibutyryl cyclic AMP


0 .01


Forskolin (pIM)


0.01 0.1
Forskolin


-5.0


-4.0
(c


-3.0 C

o
0
-2.0 .


-1.0


0.0


.1 1


1 (pM)









changes in proenkephalin, the action of the opioid agonist dynorphin Ai-13 (1
J.M) and isoproterenol (10 p.M) on proenkephalin mRNA levels at various times
was examined (Figure 5-5). Dynorphin A1-13 decreased proenkephalin
expression by 30-38% between 1 and 4 hours (Table 5-2). However, by 8 hours
the proenkephalin message levels had rebounded to control levels.
Isoproterenol induced proenkephalin mRNA by 65% at one hour, but this action
was transient, with a drop to below basal levels at 4 and 8 hours. Interestingly,
neither agonist was able to affect proenkephalin mRNA levels at 30 min. This
indicated that the mechanism for cyclic AMP induction of proenkephalin mRNA
required more than 30 min. to affect transcription rates. The transient nature of
both of these effects suggested that the cells tightly regulated receptor-coupled
adenylyl cyclase by decreasing the ability of agonists to regulate adenylyl
cyclase activity during prolonged exposure. Indeed, NG108-15 cells desensitize
to opioids within 4 hours (Vachon et al., 1987) and B-adrenergic receptors
desensitize within 1 hour (Hausdorff et al., 1990).
To establish that the opioid regulation of proenkephalin mRNA
was mediated through an opioid receptor, the ability of naloxone to block the
dynorphin-decrease in proenkephalin mRNA levels was examined. Figure 5-6
shows that incubation of cultures with dynorphin Al-13 (1 gM) for 4 hours
decreased proenkephalin mRNA in from both basal and forskolin stimulated
levels (86% and 81%, respectively). The densitometric values for Figure 5-6 are
presented in Table 5-3. Naloxone (1 pM) was able to block the action of
dynorphin A1-13 on proenkephalin mRNA (103%). This suggested that
dynorphin A1-13 was regulating proenkephalin expression through a classical
opioid receptor. Although the action of dynorphin A1-13 was not very
pronounced in this experiment, it was very reproducible and typically produced
more pronounced effects. Interestingly, no effect of the opioid was observed in



























ca-tubulin


PE


1 2 3 4 5 6 7 8 9 10 11





Figure 5-5 Time course of dynorphin A1-13 and isoproterenol regulation of
proenkephalin mRNA levels in forebrain neuronal cultures.

Dynorphin A1-13 (1LM) or isoproterenol (10gM) were added to media on
primary neuronal cultures incubated for the indicated time. Total cellular RNA
was extracted as described (Chomoczynski and Sacchi, 1984), resolved by
electrophoresis on a 1.5% agarose denaturing gel, blotted onto Zeta-Probe
membrane (Bio-Rad) and hybridized with a 950-base riboprobe to
preproenkephalin and a 30-base oligonucleotide probe to alpha-tubulin. Lane
1) control; isoproterenol (lanes 2, 4, 6, 8, 10) and dynorphin A1-13 (lanes 3, 5, 7,
9, 11) at 0.5, 1, 2, 4, and 8 hours, respectively. Results are from one
representative experiment which was repeated twice.

















Table 5-2 Densitometric quantitation for time course of dynorphin A1-13 and
isoproterenol regulation of proenkephalin mRNA levels in forebrain neuronal
cultures.


Hours Isoproterenol Dynorphin Aa-13


0.5 0.90 1.00
1 1.65 0.67
2 1.20 0.70
4 0.59 0.62
8 0.68 1.26


The densitometric values listed above were calculated from the northern blot in
Figure 5-5 and represent proenkephalin mRNA absorbance relative to alpha-
tubulin mRNA absorbance and normalized where basal expression is equal to
1.00.





































Dynorphin A-13 -
Naloxone
Forskolin


+ + + +


- +


- +


- + + +


Figure 5-6 Effect of naloxone on dynorphin A1-13 regulation of proenkephalin
mRNA levels in forebrain neuronal cultures.

Dynorphin A1-13, naloxone and forskolin (1 giM of each) were added to
media on primary neuronal cultures incubated for 4 hours. Total cellular RNA
was extracted as described (Chomoczynski and Sacchi, 1984), resolved by
electrophoresis on a 1.5% agarose denaturing gel, blotted onto Zeta-Probe
membrane (Bio-Rad) and hybridized with a 950-base riboprobe to
proenkephalin and a 30-base oligonucleotide probe to alpha-tubulin. Results
are from one representative experiment which was repeated twice.


a-tubulin



PE


AOKI

















Table 5-3 Densitometric quantitation for effect of naloxone on dynorphin
A1-13 regulation of proenkephalin mRNA levels in forebrain neuronal cultures..


Treatment PE units Percent change
relative to basal


Basal 1.00 0

Dynorphin A1-13 0.86 14

Dynorphin A1-13
with Naloxone 1.03 3

Forskolin 1.70 70 (0)

Forskolin
with Dynorphin A1-13 1.57 57(19)

Forskolin
with Dynorphin A1-13
and Naloxone 1.75 75 (3)



The densitometric values listed above were calculated from the northern blot in
Figure 5-6 and represent proenkephalin mRNA absorbance relative to alpha-
tubulin mRNA absorbance and normalized where basal expression is equal to
1.00. The values in parentheses were corrected to forskolin-stimulated mRNA
level equal to 100%.









cultures which were not switched to fresh media prior to experimentation (not
shown). This suggests the possibility that there could be endogenous
enkephalin in the media which would cause the cells to become desensitized to
the opioid action.
Effect of pertussis toxin on dvnorphin Al.-regulated proenkephalin

mRNA levels. To investigate whether dynorphin A was acting through the
receptor which is coupled to adenylyl cyclase through a Gio-protein, forebrain
cultures were incubated for 16 hours with 50 ng/ml pertussis toxin. As shown in
Figure 5-7, dynorphin A1-13 (1 pM for 4 hours) decreased proenkephalin mRNA
levels by 53% in untreated forebrain cultures, but this effect was attenuated in
pertussis toxin treated cultures (90%), indicating that the opioid was acting
through a receptor coupled to a G/Go protein. The densitometric values for
Figure 5-7 are presented in Table 5-4.
Effect of dvnorphin A1 on forskolin- and dibutvryl cyclic AMP-induction
of proenkephalin mRNA levels. If the opioid induced changes in proenkephalin
expression were mediated through cyclic AMP, then treatment of cultures with
dibutyryl cyclic AMP should bypass the opioid receptor and the opioid should
be less effective at decreasing dibutyryl cyclic AMP-induction than forskolin-
induction. Forebrain neuronal cultures were treated with either forskolin (1 I.M)
or dibutyryl cyclic AMP (600 gM) for 4 hours with or without dynorphin A1-13
(Figure 5-8). The opioid was effective at decreasing proenkephalin mRNA
levels in the presence of forskolin but not dibutyryl cyclic AMP (Table 5-5).
Dynorphin A1-13 (1 iM) was able to decrease the forskolin-induction by 58%,
whereas the dibutyryl cyclic AMP-induction was only decreased by 4%. This
indicated that the action of the opioid was mediated through a reduction in
cyclic AMP in the cells.




















.. ( a-tubulin


PE


Dynorphin A-13 + +
Pertussis toxin + +




Figure 5-7 Effect of pertussis toxin on dynorphin A-.13-induced changes in
proenkephalin mRNA levels in forebrain neuronal cultures.

Primary forebrain neuronal cultures were treated for 16 hours with
50ng/ml pertussis toxin. Dynorphin A-s13 (1 pM) was added for four hours. Total
cellular RNA was extracted as described (Chomoczynski and Sacchi, 1984),
resolved by electrophoresis on a 1.5% agarose denaturing gel, blotted onto
Zeta-Probe membrane (Bio-Rad) and hybridized with a 950-base riboprobe to
proenkephalin and a 30-base oligonucleotide probe to alpha-tubulin. Results
are from one representative experiment which was repeated once.


















Table 5-4 Densitometric quantitation for effect of pertussis toxin on dynorphin
A1-13-induced changes in proenkephalin mRNA levels in forebrain neuronal
cultures.


Treatment PE units Percent change
relative to basal


Control
Basal 1.00
Dynorphin A1-13 0.60 40

Pertussis toxin
Basal 0.95
Dynorphin A1-13 0.86 10


The densitometric values listed above were calculated from the northern blot in
Figure 5-7 and represent proenkephalin mRNA absorbance relative to alpha-
tubulin mRNA absorbance and normalized where basal expression is equal to
1.00.






















- eagle


Forskolin
Dibutyryl cAMP
Dynorphin A1-13


- + + + +
- + + -


- + +


- +


Figure 5-8 Effect of dynorphin A1-13 on forskolin- and dibutyryl cyclic AMP-
induction of proenkephalin mRNA levels in forebrain neuronal cultures.

Dynorphin A1-13 (1M), forskolin (1 gM) and dibutyryl cyclic AMP
(600.M) were added to media on primary neuronal cultures and incubated for 4
hours. Pertussis toxin (50ng/ml) was added to media 16 hours prior to start of
experiment. Total cellular RNA was extracted as described (Chomoczynski and
Sacchi, 1984), resolved by electrophoresis on a 1.5% agarose denaturing gel,
blotted onto Zeta-Probe membrane (Bio-Rad) and hybridized with a 950-base
riboprobe to proenkephalin and a 30-base oligonucleotide probe to alpha-
tubulin. Results are from one representative experiment which was repeated
twice.


a-tubulin



PE


PTX

















Table 5-5 Densitometric quantitation for effect of dynorphin A1-13 on
forskolin- and dibutyryl cyclic AMP-induction of proenkephalin mRNA levels in
forebrain neuronal cultures.


Treatment PE units Percent decrease of
induction with
dynorphin As-13

Basal 1.00

Forskolin
1.80
Dynorphin A1-13 1.27 58
Dibutyryl cAMP
1.73
Dynorphin A1-13 1.75 4


The densitometric values listed above were calculated from the northern blot in
Figure 5-8 and represent proenkephalin mRNA absorbance relative to alpha-
tubulin mRNA absorbance and normalized where basal expression is equal to
1.00.








Effect of cvcloheximide on dvnorphin Aaj1 regulation of proenkephalin
mRNA levels. To investigate whether newly synthesized proteins were involved
in the opioid-induced changes in proenkephalin mRNA, forebrain cultures were
incubated with 5 gg/ml cycloheximide, a protein synthesis inhibitor. As shown in
Figure 5-9, cycloheximide caused a two-fold increase in proenkephalin mRNA
levels by 4 hours. The effect on alpha tubulin mRNA levels was similar. When
dynorphin A1-13 was added with the cycloheximide, proenkephalin mRNA was
decreased by 40-60% at all time points. The opioid treatment did not alter the
accumulation of alpha-tubulin. Naloxone (1 gM) was able to block this action of
the opioid (88-94% of cycloheximide-treated levels). This demonstrated that the
opioid agonist, dynorphin A1-13, selectively decreased the level of the
proenkephalin mRNA and that new synthesis of proteins was not necessary for
this action. This suggests that the opioid action was a direct action on
proenkephalin expression and not mediated via nonspecific gene regulators.
Effect of opioid on proenkephalin mRNA in primary glial cultures. To
determine whether these changes in proenkephalin expression in primary
forebrain neuronal cultures were due to the glial population of the cultures,
primary glial cultures were examined for opioid regulation of proenkephalin
expression. As shown in Figure 5-10, the opioid peptide dynorphin A1-13
produced a very modest effect (a 3-9% decrease) on proenkephalin mRNA
levels in these cultures (Table 5-6). This suggests that the larger effects
observed in the neuronal cultures were on the neuronal and not the glial portion
of the population of the cultures.
Effect of opioids on proenkephalin mRNA in C6 glioma cells. The ability
of cyclic AMP to induce proenkephalin mRNA levels is illustrated by the action
of isoproterenol (10 gM) on C6 glioma cells (Figure 5-11). Although opioids








Figure 5-9 Effect of cycloheximide on dynorphin A1-13 regulation of
proenkephalin mRNA levels in forebrain neuronal cultures.

Dynorphin A1-13 (1 pM) and cycloheximide (5pg/ml) were added to
primary neuronal cultures equilibrated for one hour in serum-free DMEM and
incubated for the indicated times. Total cellular RNA was extracted as described
(Chomoczynski and Sacchi, 1984), resolved by electrophoresis on a 1.5%
agarose denaturing gel, blotted onto Zeta-Probe membrane (Bio-Rad) and
hybridized with a 950-base riboprobe to proenkephalin and a 30-base
oligonucleotide probe (NEN) to alpha-tubulin. A) Densitometric quantitation of
Northern blots shown in B) Forebrain neuronal cultures were treated for the
indicated times with dynorphin A1-13 and/or naloxone. Results are from one
representative experiment which was repeated twice.















CHX
Dyn 1-13
Dyn 1-13 + Naloxone
Naloxone


IT
I i^


Hours


,n(IiiI0.


Dynorphin A1-13
Naloxone


4 4 2 8 2 4 8 2 4 8
.- + + + + + +
+ + + +


S
U-C-


1.50


1.00-


0.50-


n n n .


a-tub



PE


Hours






















W IF a-tubulin


PE


Dynorphin A1-13 + +
Forskolin + +




Figure 5-10 Effect of dynorphin A1-13 on proenkephalin mRNA levels in
primary glial cultures.

Dynorphin A1-13 (1 gM) and forskolin (1 gM) were added to media on
primary glial cultures and incubated for 4 hours. Total cellular RNA was
extracted as described (Chomoczynski and Sacchi, 1984), resolved by
electrophoresis on a 1.5% agarose denaturing gel, blotted onto Zeta-Probe
membrane (Bio-Rad) and hybridized with a 950-base riboprobe to
proenkephalin and a 30-base oligonucleotide probe to alpha-tubulin. Results
are from one representative experiment which was repeated twice.













Table 5-6 Densitometric quantitation for effect of dynorphin Al-13 on
proenkephalin mRNA levels in primary glial cultures.



Treatment PE units Percent change
relative to basal


Basal 1.00
Dynorphin A1-13 0.91 9

Forskolin 5.44 544
+ Dynorphin A1-13 5.23 523


The densitometric values listed above were calculated from the northern blot in
Figure 5-10 and represent proenkephalin mRNA absorbance relative to alpha-
tubulin mRNA absorbance and normalized where basal expression is equal to
1.00.






















J


Isoproterenol
D-Ala enk
Naloxone


- + + + +
- + + -
_- + +


Figure 5-11 Effect of D-Ala enk and naloxone on isoproterenol-induction of
proenkephalin mRNA levels in C6 glioma cells.
D-Ala enk (1I M), naloxone (1 l.M) and isoproterenol (10 gM) were added
to media on C6 glioma cells and incubated for 4 hours. Total cellular RNA was
extracted as described (Chomoczynski and Sacchi, 1984), resolved by
electrophoresis on a 1.5% agarose denaturing gel, blotted onto Zeta-Probe
membrane (Bio-Rad) and hybridized with a 950-base riboprobe to
proenkephalin and a 30-base oligonucleotide probe to alpha-tubulin. Results
are from one representative experiment which was repeated twice.


a-tubulin


PE


rm
















Table 5-7 Densitometric quantitation for effect of D-Ala enk and naloxone on
isoproterenol-induction of proenkephalin mRNA levels in C6 glioma cells.



Treatment PE units Percent of isoproterenol-
induction

Basal 1.00

Isoproterenol 5.57 100
+ D-Ala enk 4.13 75
+ D-Ala enk
& Naloxone 5.78 103
+ Naloxone 5.52 99


The densitometric values listed above were calculated from the northem blot in
Figure 5-10 and represent proenkephalin mRNA absorbance relative to alpha-
tubulin mRNA absorbance and normalized where basal expression is equal to
1.00.








produced a biphasic response of adenylyl cyclase in C6 glioma cells, IRM
concentrations in untreated cells consistently inhibited cyclic AMP
accumulation. When these cells were treated with D-Ala enk (1 uM) for 4 hours,
proenkephalin mRNA levels were decreased by 25% (Table 5-7). Naloxone (1

IM) blocked the effect of D-Ala enk but did not alter proenkephalin mRNA levels
alone. This indicated that at this concentration of D-Ala enk, the opioid was
inhibiting its own synthesis.
Effect of opioids on proenkephalin mRNA in NG108-15 cells. Opioid

peptides inhibit adenylyl cyclase by 30% in the neuroblastoma x glioma hybrid
cell, NG108-15 (Figure 5-12). These cells also express proenkephalin mRNA
(Yoshikawa & Sabol, 1984), although at a relatively low abundance. To
examine whether the same action of opioids on proenkephalin mRNA levels in
primary neuronal cultures would repeat itself in this cell line, NG108-15 cells
were treated with cycloheximide and D-Ala enk. Cycloheximide, as in primary
cultures, caused an accumlation of proenkephalin and alpha-tubulin mRNA
through 8 hours (Figure 5-13). The addition of D-Ala enk decreased the
proenkephalin mRNA by 40% at 4 and 8 hours. Interestingly, D-Ala enk did not
affect mRNA levels at 2 hours. This may be a reflection of a quicker onset of
action of cycloheximide than dynorphin on regulating proenkephalin
expression. Pretreatment of cells with pertussis toxin (50 ng/ml) did not alter the
accumulation of proenkephalin mRNA observed with cycloheximide, however,
the effect of the opioid peptide was lost. This action of opioids was the same as
that observed in primary neuronal cultures. This indicates that the same
mechanism for opioids to decrease their own synthesis occurs in both systems.






90












20
E



S15

E



0 .01 .1 1 10 100 1000
D-Ala enkephalinamide (p.M)


Figure 5-12 Effect of D-Ala enk on adenylyl cyclase activity in isolated
membranes from NG108-15 cells.

Adenylyl cyclase activity was determined by incubating membranes from
NG108-15 cells in Tris-Mg buffer containing the indicated concentration of D-
Ala enk at 370 for 10 min. The standard cyclase components and the
determination of the cyclic AMP content were as described under Materials and
Methods. Values are the mean S.D., n=3.