Regulation of Alpha2A-Adrenergic receptor expression in cultured rat astroglia

MISSING IMAGE

Material Information

Title:
Regulation of Alpha2A-Adrenergic receptor expression in cultured rat astroglia by Michael A. Reutter
Physical Description:
vii, 130 leaves : ill. (some col.) ; 29 cm.
Language:
English
Creator:
Reutter, Michael A., 1959-
Publication Date:

Subjects

Subjects / Keywords:
Research   ( mesh )
Astrocytes   ( mesh )
Protein Kinase C   ( mesh )
Cyclic AMP   ( mesh )
Receptors, Adrenergic, alpha-2 -- biosynthesis   ( mesh )
Receptors, Adrenergic, alpha-2 -- metabolism   ( mesh )
Cells, Cultured   ( mesh )
RNA, Messenger   ( mesh )
Epinephrine   ( mesh )
Angiotensin II   ( mesh )
Rats, Sprague-Dawley   ( mesh )
Gene Expression Regulation   ( mesh )
Transcription, Genetic   ( mesh )
Signal Transduction   ( mesh )
Department of Physiology thesis Ph.D   ( mesh )
Dissertations, Academic -- College of Medicine -- Department of Physiology -- UF   ( mesh )
Genre:
bibliography   ( marcgt )
non-fiction   ( marcgt )

Notes

Thesis:
Thesis (Ph.D.)--University of Florida, 1997.
Bibliography:
Bibliography: leaves 110-129.
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 - 002284741
oclc - 48839961
notis - ALN7867
System ID:
AA00011813:00001


This item is only available as the following downloads:


Full Text











REGULATION OF ALPHAA,-ADRENERGIC RECEPTOR EXPRESSION
IN CULTURED RAT ASTROGLIA












By

MICHAEL A. REUTTER


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

























This work is dedicated to the memory of my parents, Allen and Janet Reutter.














ACKNOWLEDGMENTS


I would like to extend my heartfelt gratitude to my advisor Dr. Colin Sumners for

his advice, guidance and financial support throughout the course of this work. His

understanding and patience helped me persevere through times when it would have been

easy to quit. I would also like to express my appreciation for the help of my committee

members, Dr. Stephen Baker, Dr. Mohan Raizada and Dr. Satya Kalra.

My sincere thanks go to all of the members of the Sumners lab; to Dr. Elaine

Richards for her technical advice and knowledge which made my work much easier, and

for her kind and thoughtful attitude which made the lab a pleasant place to work; to Drs.

Xian-Cheng Huang and Ujjwala Shenoy for their encouragement and light-hearted

demeanor; and to Jennifer Moore and former lab member Tammy Gault for the countless

dishes of astroglia.

Financial support was gratefully received from the University of Florida, Center for

Neurobiological Sciences, and the American Heart Association, Florida Affiliate.

Finally, I wish to thank my grandmother, Alice, my sisters, Kim and Deb, and my

niece, Chandra, for their constant encouragement and love.















TABLE OF CONTENTS




ACKNOWLEDGMENTS ................................. .......... iii

ABSTRACT ................................... ...... ......... vi

CHAPTERS

1 INTRODUCTION ............................................. 1
Characterization of Alpha-2 Adrenergic Receptors .................. .... 2
Regulation of a2A-AR Function and Expression ............. ........... 9
R ationale ................. ................. ...... ......... 17

2 MATERIALS AND METHODS .................................... 20
Materials .................................. .. .. ......... 20
Methods ........ .................................. 21

3 REGULATION OF a2A-ADRENERGIC RECEPTOR MRNA IN CULTURED RAT
ASTROGLIA: ROLE OF CYCLIC AMP AND PROTEIN KINASE C ..... 31
Introduction ...... ........ ................................. 31
Results ......... ............................................ 33
Discussion .......................... ....... .. ........... 39

4 REGULATION OF a2A-ADRENERGIC RECEPTOR MRNA IN CULTURED RAT
ASTROGLIA: ROLE OF EPINEPHRINE AND ANGIOTENSIN II ....... 60
Introduction ................................................. 60
Results .......... ............ .......... .... 62
Discussion ..................... ......... ........ ........... 67

5 REGULATION OF a2A-ADRENERGIC RECEPTOR NUMBER BY
EPINEPHRINE ................ ................... .......... 83
Introduction ............................................... 83
Results ........ ................................. ......... 86
Discussion ...................................... ........... 90









6 CONCLUSIONS AND SUMMARY ......................... .. 100

REFERENCES .......: ........................................... 110

BIOGRAPHICAL SKETCH ....... .................. ...... 130














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

REGULATION OF ALPHA2A-ADRENERGIC RECEPTOR EXPRESSION
IN CULTURED RAT ASTROGLIA

By

Michael A. Reutter

August, 1997

Chairperson: Colin Sumners, Ph.D.
Major Department: Physiology

Epinephrine (Epi) mediates various physiological effects via a2^-adrenergic

receptors (a2A-AR). Studies in mice with a point-mutation in the gene for a2A-AR have

shown that these receptors are responsible for the centrally-mediated depressor effects of

a2-AR agonists. These studies underscore the importance of understanding the basic

cellular mechanisms involved in the expression of a2A-AR, of which little is known. In this

study we have utilized astroglia cultured from the hypothalamus and brain stem of adult

Sprague-Dawley rats as a model system in which to study factors which regulate a2A-AR

expression. These cells contain a homogeneous population of a2A-AR.

We initially investigated regulation of ao2-AR mRNA in cultured astroglia as a

result of increases in intracellular cAMP accumulation and by protein kinase C (PKC)

activation. Treatment of astroglial cultures with forskolin (FSK), an adenylyl cyclase









activator, or with phorbol 12-myristate 13-acetate (PMA), a PKC agonist, caused time-

and concentration-dependent decreases in the levels of a -4.0 kb a2A-AR mRNA

transcript. These results were mimicked by treatment of the cells with cAMP analogues or

agents which activate PKC. The decreases in a2A-AR mRNA levels caused by FSK and

PMA treatment appear to be the result of decreases in transcription of the a2A-AR gene

and are not due to decreases in a2A-AR mRNA degradation rate. In addition, FSK and

PMA treatment of cultured astroglia leads to decreased numbers of a2-AR. These

observations suggest that a2A-AR are regulated by cAMP and PKC.

Epi can cause accumulation of cAMP or activation of PKC by stimulating a,- and

P-adrenergic receptors (a,-AR; P-AR). Therefore we also studied the effect of Epi on

c2A-AR expression. Our studies showed that Epi elicits a dose- and time-dependent

decrease in steady state levels of c2A-AR mRNA and number of azA-AR, effects which are

mediated by a,-AR and P-AR. Taken together, these results indicate that expression of

a2A-AR is regulated in a heterologous manner by Epi, via a,-AR and P-AR, and their

intracellular signaling pathways














CHAPTER 1
INTRODUCTION


The catecholamines epinephrine (Epi) and norepinephrine (NE) mediate a variety

of physiological responses via several subtypes of adrenergic receptors. These

catecholamines play a role in control of functions such as glycogenolysis, lipolysis, platelet

aggregation, vasoconstriction, myocardial contractile force and rate, and bronchodilation.

Within the central nervous system (CNS) Epi and NE act as neurotransmitters and

contribute to control of sympathetic nervous system outflow, state of wakefulness and

attention, modulation of pain afferents, and control of feeding behavior. In addition,

perturbations of catecholaminergic systems may play a role in pathological conditions such

as hypertension, pheochromocytoma, multiple system atrophy with sympathetic nervous

system degeneration, obesity, Parkinson's disease, Alzheimer's disease, chronic pain, and

addiction. Investigators have tried since the turn of the century to elucidate the

mechanisms by which Epi and NE exert their effect on responsive tissue, and to

understand how adrenergic receptors are involved in those responses.











Characterization of Alpha-2 Adrenergic Receptors

Evidence for Multiple Alpha Adrenergic Receptor Subtypes


The receptor concept was initially proposed in 1878 as an explanation for the

effect of certain drugs on salivary flow (Langley, 1878). This concept was extended to

suggest that Epi and NE acted at more that one "receptive mechanism," because ergot

alkaloids produced differential blockade of catecholamine-induced effects (Dale, 1906).

This ultimately led to the hypothesis that cellular responsiveness was proportional to the

number of receptors occupied by the catecholamine (Clark, 1937). Then, in 1948,

Ahlquist proposed that the effects of catecholamines were mediated by two different

adrenergic receptors, which he termed alpha and beta. He based his conclusions on

observations that responses in a variety of tissues fit into two groups based on the rank-

order potency of a series of structurally related catecholamines. Those responses

insensitive to blockade by ergot alkaloids were mediated by beta-adrenergic receptors (P-

AR), while those sensitive to blockade were mediated by alpha-adrenergic receptors (a-

AR). These studies provided the basis for further studies to delineate the pharmacology,

and eventually, the molecular biology of adrenergic receptors.

Further evidence for the pharmacological classification of adrenergic receptors as

alpha and beta was provided with antagonists which blocked certain responses.

Dichloroisoprenaline was the first agent found to block responses associated with P-AR,

but not a-AR (Powell and Slater, 1957; Moran and Perkins, 1958). In addition the a-AR

antagonists phentolamine and phenoxybenzamine increased NE overflow resulting from











nerve stimulation (Brown and Gillespie, 1957). These a-AR antagonists were initially

thought to inhibit neuronal uptake ofNE, resulting in increased overflow ofNE after

nerve stimulation (Thoenen et al., 1964; Langer, 1970). However, Langer (1970) also

showed that phenoxybenzamine increased the amount of NE released per stimulus. The

stimulus-dependent increase in NE release caused by phenoxybenzamine was also coupled

to an increase in release of dopamine P-hydroxylase (DePotter et al., 1971; Johnson et al.,

1971), an enzyme which catalyzes the synthesis of NE from dopamine and is coreleased

with NE. These results were confirmed for both phenoxybenzamine and phentolamine at

concentrations which did not inhibit reuptake, but enhanced neurotransmitter overflow

elicited by nerve stimulation (Starke et al., 1971). These results suggested that a-AR

antagonists caused an increase in neurotransmitter release without effecting reuptake.

In contrast to a-AR antagonists, a-AR agonists were shown to decrease

neurotransmitter release. Clonidine, an a-AR agonist, was found to decrease NE release

(Anden et al., 1970). Because NE is also an a-AR agonist, these results suggested that

NE may act to inhibit its own release. This led to the concept ofpresynaptic regulation of

NE release (Farnebo and Hamberger, 1971; Kirpekar and Puig, 1971; Langer et al., 1971;

Starke, 1971). A presynaptic a-AR was believed to mediate a feedback mechanism

whereby NE could act to inhibit its own release (Kirpekar and Puig, 1971).

The pharmacology of a-AR had previously been shown to differ among different

tissues suggesting the tissues may express different a-AR. The initial hypothesis that

multiple a-AR existed was based on observations that different structural requirements

were needed for antagonists and agonists to interact with a-AR in the rat vas deferens and











rabbit intestine (van Rossum, 1965). Delbare and Schmitt (1973) suggested that the

differing structural requirements for drug interaction were due to different receptors,

which they termed a,-AR and a2-AR by analogy with the earlier named iP-AR and P2-AR

(Lands et al., 1967). Pharmacological studies utilizing phenoxybenzamine (Dubocovich

and Langer, 1974) or clonidine (Starke et al., 1974) implied that pre- and postsynaptic a-

AR existed and were also different. This resulted in the hypothesis that a,-AR were

located postsynaptically while a,-AR were located presynaptically (Langer, 1974).

Subsequently, Berthelsen and Pettinger (1977) suggested that the differences between a,-

AR and a2-AR were functional and not anatomic. They proposed that the receptor which

mediated vasoconstriction be designated a,, while the receptor which mediated inhibition

ofneurotransmitter release from sympathetic nerve terminals be designated a2. Finally,

radioligand binding assays confirmed the existence of distinct a,-AR and a2-AR based on

the rank orders of potency of various agonists and antagonists at these receptor subtypes

(Fain and Garcia-Sainz, 1980), and hence provided a pharmacologic definition of these

receptors instead of anatomic or functional definitions.

Developing along with the concept of multiple adrenergic receptors, was the

concept that the different adrenergic receptor subtypes mediated different biochemical

responses in the cell. Epi was known to act at a-AR and P-AR to inhibit and stimulate

cyclic AMP synthesis, respectively (Robison et al., 1967; Turtle and Kipnis, 1967). In

many tissues which contained a-AR, cyclic AMP synthesis was not inhibited with a-AR

agonists (Robison et al., 1970), raising questions regarding the biochemical pathways

activated by a-AR. Questions regarding the biochemical pathways associated with a-AR










stimulation were answered after division ofa-AR into a,-AR and a2-AR subtypes

(Delbare and Schmitt, 1973). Stimulation of a,-AR resulted in phosphoinositide (PI)

turnover and increased intracellular calcium (Ca") concentrations, while stimulation of a2-

AR was associated with inhibition of adenylyl cyclase activity and attenuation of cAMP

accumulation (Wikberg, 1979; Fain and Garcia-Sainz, 1980). The ability of a2-AR to

inhibit basal and stimulated adenylyl cyclase activity was demonstrated in a variety of

systems including: human platelets (Jakobs et al., 1976); rabbit platelets (Tsai and

Lefkowitz, 1978); rat pancreatic islets (Katada and Ui, 1981); hamster adipocytes

(Aktories et al., 1979); and NG108-15 cells (Sabol and Nirenberg, 1979). Thus, the

existence of ai-AR and a2-AR as separate subtypes had been defined pharmacologically

and biochemically.


Evidence for Multiple Alpha-2 Adrenergic Receptor Subtypes


The initial subclassification of a2-AR was based on pharmacological studies which

showed the differential ability of the a,-AR antagonist prazosin to inhibit binding of the

specific a2-AR antagonists ['H]-yohimbine and [3H]-rauwolscine to a variety of tissues

and cell lines. Prazosin bound to the platelet ac-AR with low affinity (KI = 200-300 nM)

and to the a2-AR from neonatal rat lung with high affinity (K, = 5-10 nM; Bylund, 1985;

Nahorski et al., 1985). Based on these differences two laboratories independently

identified a2-AR in the human platelet and neonatal rat lung as a2A-AR and a,-AR,

respectively (Bylund, 1985; Nahorski et al., 1985). This was followed by evidence that

the HT29 cell line contained only a2A-AR, while the NG108-15 cell line contained only











a2B-AR (Bylund et al., 1988). While a2-AR were initially subclassified based on

differential abilities of antagonists to inhibit radioligand binding, functional studies

confirmed that both subtypes were able to inhibit cAMP synthesis (Bylund and Ray-

Prenger, 1989).

The a2-AR in the opossum kidney (OK) cell line showed a pharmacological profile

intermediate between the a2A-AR and a2B-AR suggesting a third subtype of a2-AR

(Murphy and Bylund, 1988). This receptor bound prazosin with an affinity similar to the

a, subtype, but bound rauwolscine with a higher affinity. Further studies demonstrated

that the a2-AR in the OK cell line and in OK tissue was a third subtype, which was named

the a2c-AR (Blaxall et al., 1991). This subtype was also shown to exist in the human

retinoblastoma cell line, Y79 (Gleason and Hieble, 1992).

Studies in the bovine pineal, rat submaxillary gland and in a rat pancreatic islet

tumor cell line, RINm5F, suggested the possibility of a fourth a2-AR subtype (Simonneaux

et al., 1991; Michel et al., 1989b; Remaury and Paris, 1992). Like the a2A-AR this

subtype (named am-AR) had low affinity for prazosin, but unlike the a2A-AR, it had low

affinity for rauwolscine and yohimbine. However, subclassification of this receptor was

controversial. Some radioligand binding studies in the rat CNS referred to this receptor as

am (MacKinnon et al., 1992), while others referred to is as a2A (Uhlen et al., 1992).

Molecular cloning studies later showed that this receptor is a species homologue of the

human a2A-AR and is not a distinct subtype (Link et al., 1992).

The first a2-AR cloned was the a2A-AR from human platelets (Kobilka et al.,

1987b). Subsequently, genes for the human a2-AR and a2c-AR were identified (Regan et











al., 1988; Lomasney et al., 1990). These clones were identified as a,2C10, a2C2 and aC4

(a2A-AR, a2B-AR, a2c-AR, respectively) based on their chromosomal locations (Kobilka et

al., 1987; Regan et al., 1988; Lomasney et al., 1990). Similarly, three genes for a2-AR

have been identified in the rat (Voigt et al., 1991; Lanier et al., 1991; Chalberg et al.,

1990; Zeng et al., 1990) and in the mouse (Chruscinski et al., 1992; Link et al., 1992).

The pharmacology of the cloned a2-AR closely matched the pharmacology of a2-AR

subtypes identified in tissues and cell lines (Bylund et al., 1994). The mouse and rat

homologues of the human a2B-AR and a2c-AR have pharmacology similar to the human

clones, but the azA-AR differs in that the mouse and rat clones have low affinity for

yohimbine and rauwolscine compared to the human clone. This difference in

pharmacology appears to be due to a single amino acid substitution in the mouse a2c-AR

(Link et al., 1992).

Because only three genes encode for a2-AR in the human, rat and mouse, the

designation of subtypes across species as a2A, a2, and a2c appears to be appropriate. The

subtype originally identified pharmacologically as a2 (Michel et al., 1989) appears to be a

species homologue of the human a^,-AR, based on cloning studies (Link et al., 1992).

Therefore, throughout this manuscript the rat a,-AR will be referred to as a2A-AR, a2B-

AR, and a2c-AR.

The existence of several subtypes of a2-AR implied that each subtype may mediate

specific physiological responses. The lack of subtype-selective agonists or antagonists has

precluded the pharmacological study of subtype-specific responses. Immunohistochemical

and in situ hybridization techniques, however, have been used to identify cellular and











tissue localization of az-AR subtypes, and have led to hypotheses for the physiological

role of each of the subtypes. Within the CNS all three subtypes have been identified by in

situ hybridization. Messenger RNA for as-AR has been detected in the thalamic nuclei,

but not in the rest of the CNS (Nicholas et al., 1993; Scheinin et al., 1994). Expression of

a2A-AR and a~2-AR in the CNS is more widespread. C2c-AR mRNA has been identified in

the basal ganglia, olfactory tubercle, hippocampus, and cerebral cortex, while a~-AR

mRNA has been detected in the cerebral cortex, septum, hypothalamus, hippocampus and

amygdala (Nicholas et al., 1993; Scheinin et al., 1994). Within the largest noradrenergic

cell group in the brain, the locus ceruleus, only a,2-AR have been identified (Nicholas et

al., 1993; Scheinin et al., 1994). This location implies that a2,-AR are involved in

mediating decreases in sympathetic nervous system outflow (Correa-Sales et al., 1992).

In addition, a2A-AR mRNA is expressed in CNS regions involved in cardiovascular control

such as the nucleus tractus solitarius, rostral ventrolateral medulla and the

intermediolateral cell column (Nicholas et al., 1993; Scheinin et al., 1994). Messenger

RNA for a2A-AR is also present in the dorsal root ganglia and spinal cord (Lawhead et al.,

1992; Nicholas et al., 1993; Stafford-Smith et al., 1994), suggesting that a2A-AR are

involved in the antinociceptive effects of a2-AR agonists. Immunohistochemical studies

have found that a2A-AR are located in brainstem and spinal cord regions involved in

autonomic functions (Rosin et al., 1993), lending additional evidence to the hypothesis

that a2A-AR are involved in cardiovascular control.

Recently, studies in mice with mutations or knockouts of a2-AR genes have

provided additional evidence for the subtype-specific role of a,-AR in cardiovascular











function. In studies of knockout mice, a2-AR were demonstrated to be the a2-AR

subtype on peripheral resistance vessels which mediates vasoconstriction (Link et al.,

1996). Within the CNS, a2A-AR were shown to be involved in the long-lasting

hypotensive response to a2-AR agonists (MacMillan et al., 1996).


Regulation of a2A-AR Function and Expression

Terminology


The responsiveness of cells to hormone or neurotransmitter stimulation is

regulated by changes in the functional state of the receptors for the hormone or

neurotransmitter, as well as the number of receptors on the cell surface. Receptor

responsiveness to a stimulus has been demonstrated to decrease in the continued presence

of that stimulation, a phenomenon known as desensitization. Examples of receptor

desensitization include photoadaptation (Liebman and Pugh, 1980) and tolerance to

pharmacologic agents (Rubin and Rosen 1975; Sibley and Lefkowitz, 1985). The

widespread occurrence of desensitization across receptor types suggests its importance in

cellular processes. In addition, desensitization appears to act as an intracellular feedback

phenomenon turning off cellular response to a single stimulus or reducing responsiveness

in general to other stimuli.

Desensitization of G-protein coupled receptors refers to a variety of processes

which render the receptor incapable of interacting with its G-protein, resulting in

decreased cellular responsiveness to hormones or neurotransmitters. For the purpose of











this manuscript, time-dependent processes involved in desensitization will be defined as

follows:

Uncoupling- rapid (seconds to minutes) impairment of the ability of the

receptor to interact with its G-protein (Liggett et al., 1992;

Hausdorffet al., 1990; Dohlman et al., 1991)

Sequestration movement of the receptor away from the cell surface to an

intracellular location over a time frame of minutes to hours

(Perkins et al., 1991)

Downregulation the long-term (hours to days) decrease in receptor number

as a result of decreased receptor synthesis or increased

receptor degradation (Bouvier et al., 1989; Collins et al.,

1990; Lohse, 1993)

In addition, receptor desensitization can be divided based on the causative stimulus:

Homologous result of agonist binding to its receptor

Heterologous result of activation of intracellular pathways by other ligands

and does not require agonist occupancy of the receptor

The causative and time-dependent processes of receptor desensitization are

complementary, but very different. For example, agonist-dependent uncoupling of the P-

AR as a result of P-AR kinase (PARK) phosphorylation is rapid, homologous and results

in loss of receptor function (Benovic et al., 1990, 1991). In contrast, the cAMP-

dependent downregulation of P-AR is slow, heterologous and results in decreased

receptor number (Bouvier et al., 1989).











In addition to desensitization, G-protein-mediated pathways can be sensitized.

Such sensitization appears to be the result ofheterologous interaction of G-protein-

mediated pathways (Hoffman et al., 1986; Hadcock et al., 1990, 1991; Sakaue and

Hoffman, 1991; Morris et al, 1991). By analogy to the processes defined above for

desensitization, sensitization appears to involve short-term (minutes to hours) increases in

responsiveness of receptor-effector pathways, as well as long-term (hours to days)

upregulation of components of receptor-effector pathways. For the purposes of this

manuscript, the term "regulation" will be used to collectively refer to processes involved in

the sensitization or desensitization of G-protein-coupled receptors and their signal

transduction pathways. The terms "upregulation" and "downregulation" will be used to

specifically refer to increases and decreases, respectively, in the amount of receptor.

One family of receptors that are regulated in this fashion are the P-AR, which

transduce an Epi or NE signal into a stimulation of the intracellular enzyme adenylyl

cyclase (Perkins and Moore, 1973; Tate et al., 1991). The P.-AR subtype was the first G-

protein-coupled hormone or neurotransmitter receptor purified and cloned (Dixon et al.,

1986), and therefore it has been the model for studying G-protein coupled receptor

regulation (Sibley and Lefkowitz, 1985). The cellular and molecular mechanisms of

homologous and heterologous regulation have been largely defined for the P,-AR

(reviewed by Hadcock and Malbon, 1993; Lohse, 1993; Hein and Kobilka, 1995). The

P,-AR can be phosphorylated by PARK (Benovic et al., 1986, 1989, 1990), protein kinase

A (PKA; Blake et al., 1987; Clark et al., 1989), or protein kinase C (PKC; Pitcher et al.,

1992; Bouvier et al., 1987; Johnson et al., 1990), resulting in uncoupling of the receptor











from its effector pathway. The P2-AR can also be reversibly sequestered to intracellular

compartments as a result of agonist binding (vonZastrow and Kobilka, 1992, 1994;

vonZastrow et al, 1993). Finally, the P2-AR can be downregulated via agonist-dependent

and independent events (Bouvier et al., 1989; Collins et al., 1990). These mechanisms are

summarized in Table 1-1, and have provided the basis for studies on the regulation of

other G-protein-coupled receptors.




Table 1-1. Mechanisms of P2-AR Regulation


Mechanism Specifcity

Uncoupling

PARK/P-arrestin Homologous
PKA/PKC Heterologous

Sequestration Homologous

Downregulation
S mRNA

Decreased Transcription Heterologous

Increased mRNA Degradation Heterologous
S Receptor Degradation

Agonist-Specific Homologous

PKA/PKC-Mediated Heterologous
From Lohse, 1993











Homologous Regulation of a2A-AR


Study of the cellular and molecular mechanisms involved in an-AR regulation has

lagged behind that of the P2-AR. Human platelets were the first model system used for

studying a2-AR regulation. Cooper et al. (1978) demonstrated that the platelet a2-AR

was subject to homologous regulation, because incubation of platelets with Epi resulted in

desensitization of az-AR via decreased function and decreased number. This inverse

relationship between increased catecholamine concentration and decreased platelet a,-AR

function and number was shown to occur in pathophysiological conditions such as

pheochromocytoma, a catecholamine-producing tumor (Davies et al., 1981; Brodde and

Bock, 1984). The reverse was also shown to occur in multiple system atrophy with

sympathetic nervous system degeneration where decreased catecholamine concentrations

led to increased platelet a2-AR expression (Davies et al., 1981). a2-AR on platelets from

dogs and rabbits were also regulated similarly (Meyers et al., 1983; Hamilton et al., 1985;

Deighton et al., 1988), indicating that similar regulatory mechanisms operated across

species lines. The cellular mechanisms of this regulation involved guanine nucleotides

which reduced the high affinity state of the receptor, indicating involvement of a guanine

nucleotide binding protein (Michel et al., 1980; Schloos et al., 1987). In addition,

phosphorylation of the inhibitory G protein (Gi) by protein kinase C (PKC) resulted in

suppression of the inhibitory adenylyl cyclase pathway associated with a,-AR in human

platelets (Jakobs et al., 1985; Katada et al., 1985; Watanabe et al., 1985), suggesting that

the a2-AR signaling pathway was also subject to heterologous regulation (see later











section). The a,-AR on platelets was subsequently cloned and was classified as the a2,

subtype (Kobilka et al., 1987). Like the P2-AR, the a2A-AR have become the model for

study of a2-AR regulation.

Additional information regarding homologous regulation of a2-AR has come from

the use of cell lines and transfected cells as model systems. The a,A-AR is coupled to

inhibition ofadenylyl cyclase in virtually every system studied. Therefore, modulation of

cellular cAMP levels is generally used as a measure of the functional state of a2A-AR. In

the human colonic adenocarcinoma cell line HT29, activation of a2A-AR results in

inhibition of adenylyl cyclase activity (Turner et al., 1985), a response which is

desensitized in the continued presence of a2-AR agonists (Jones et al., 1987; Jones and

Bylund, 1988). This response is notable within 5 min ofagonist exposure and is readily

reversible, even after 60 min of NE incubation (Jones et al., 1990). However, after 18 h

of NE incubation, a2A-AR receptor numbers decreased (Jones et al., 1990). Similarly,

experiments in transfected Chinese hamster ovary (CHO) cells demonstrated that agonist

exposure resulted in decreased high-affinity binding (Eason and Liggett, 1992), which is

suggestive of uncoupling of the a2A-AR from its G-protein (Hausdorffet al., 1990). This

was followed by sequestration of a,2-AR and a long-term decrease in the amount of Gi

and a2A-AR (Eason and Liggett, 1992). These studies demonstrated that a2A-AR can be

desensitized via short- and long-term changes, but did not demonstrate the mechanisms

associated with those changes.

The molecular mechanisms for these changes in c,,-AR sensitivity had been

proposed to include phosphorylation of the receptor by BARK, because PARK could










phosphorylate the purified receptor in a reconstituted system (Benovic et al., 1987).

Studies in transfected Chinese hamster fibroblasts (CHF) provided the first evidence in

whole cells that a2A-AR were phosphorylated by PARK in an agonist-dependent fashion

(Liggett et al., 1992). In a manner similar to the P2-AR (Benovic et al., 1990, 1991),

agonist-dependent desensitization of a,-AR occurred rapidly after agonist exposure, was

accompanied by phosphorylation of the receptor, and was dependent on agonist

occupancy of the receptor (Liggett et al., 1992). These events led to sequestration of 2A-

AR in CHF, but not to a long-term decrease in a2A-AR number (Liggett et al., 1992) as

was seen in CHO and HT29 cells (Jones et al., 1990; Eason and Liggett, 1992). Instead, a

decrease in G, was noted, further desensitizing the c2A-AR-mediated inhibition of adenylyl

cyclase (Liggett et al., 1992). The different regulatory results observed in HT29 cells

(Jones et al., 1990), CHO cells (Eason and Liggett, 1992) and CHF (Liggett et al., 1992)

suggest that a2A-AR may be regulated in a cell- or tissue-specific manner.


Heterologous Regulation of a2A-AR


As mentioned previously, the platelet model provided the first evidence of

heterologous regulation of the a2A-AR signaling pathway. Platelets treated with the

phorbol ester, 12-O-tetradecanoylphorbol-13-acetate did not exhibit an Epi-induced

inhibition ofadenylyl cyclase activity (Jakobs et al., 1985). This effect of phorbol ester

was due to activation of PKC and subsequent phosphorylation ofG, (Watanabe et al.,

1985; Katada et al., 1985). It appeared that the phosphorylated form of G, could not be

activated by GTP and therefore could not inhibit adenylyl cyclase activity. Similarly, in










NG108-15 cells, phorbol esters decrease responsiveness of aB-AR without changing a2,-

AR number, suggesting action at the level of G, (Convents et al., 1989). This inhibitory

response to phorbol esters was also shown to occur with the G,-linked response to

bradykinin (Convents et al., 1989), providing further evidence of the heterologous nature

of PKC regulation of G,-mediated pathways. PKC activation has not resulted in

desensitization of a2A-AR or their associated signaling components in HT29 cells or

pancreatic P-cell lines (Sakaue and Hoffman, 1991; Hamamdzic et al., 1995).

The first demonstration that a2-AR themselves were regulated in a heterologous

fashion was made in studies with cultured astroglia. Treatment of cultured astroglia with

insulin resulted in downregulation of a2-AR (Richards et al., 1987) although the specific

a2-AR subtype was not known at that time. These results were unique in that they were

also the first demonstration of a2-AR regulation in non-transfected, non-transformed cells

containing transcriptional machinery. Studies in HT29 cells also showed that insulin

downregulated a2A-AR, and that other growth factors produced similar decreases

(Devedjian et al., 1991). The downregulatory effect of insulin and growth factors was

shown to be due to decreased transcription of the a2A-AR gene resulting in decreases

levels of a2A-AR mRNA (Devedjian et al., 1991).

Because of the inhibitory coupling of a2A-AR to adenylyl cyclase, the question

arose of whether cAMP may regulate a2A-AR. Studies in HT29 cells showed that a2A-AR

mRNA and protein levels increased after exposure of the cells to drugs which stimulate

adenylyl cyclase and result in accumulation ofintracellular cAMP (Sakaue and Hoffman,

1991). The increase in a2A-AR expression occurred by transiently stabilizing the a2A-AR










mRNA transcript and by increasing transcription rate (Sakaue and Hoffman, 1991). In

order to elucidate the transcriptional mechanisms involved in a2A-AR expression, JEG-3

cells were transfected with a reporter gene coupled to the a2A-AR promoter. Cyclic AMP

increased a2A-AR gene expression via a PKA-mediated increase in activity of the 5'

promoter region of the a2A-AR gene. These results suggested the involvement of a cyclic

AMP responsive element (CRE) in the promoter region. Analysis of the promoter region

for the human a2A-AR did not, however, reveal the presence of a CRE (Handy et al.,

1992). Similarly, a CRE was not reported for the rat a2A-AR promoter (Handy et al.,

1995). It is possible that the actions of PKA are mediated by nuclear factors which are

not classically associated with cyclic AMP accumulation.


Rationale


Three subtypes ofa2-adrenergic receptors (a2-AR) have been identified by

pharmacology and molecular cloning (Bylund et al., 1994). All three subtypes can be

activated by the endogenous catecholamines epinephrine (Epi) and norepinephrine, and

inhibit the accumulation of cellular cyclic AMP via inhibition of adenylyl cyclase (Limbird,

1988). The physiological function of the different subtypes appears to be determined by

their pattern of cellular and tissue localizations. For example, activation of a2A-AR in the

central nervous system appears to mediate decreases in blood pressure (MacMillan et al.,

1996), while activation of a2B-AR on resistance vessels increases blood pressure (Link et

al., 1996). The differences in cell and tissue distribution of a2-AR may also contribute to

differing modes of regulation of expression of each subtype. Regulation of central a2A-AR










represents an important way through which adrenergic signaling is modulated. Up-

regulation of a2A-AR could potentiate the inhibitory actions of a-AR agonists, whereas

down-regulation may lead to increased activity of adrenergic signaling pathways. An

understanding of the regulation of expression of a2-AR in a variety of cell type and tissues

may lead to novel methods to affect changes in a2A-AR expression without changing the

expression or function of a-AR or a2c-AR.

Information regarding a2A-AR regulation has come largely from the use of cell

lines and transfected cells as model systems. Experiments in such transformed cells allow

study of receptor regulation within a relatively homogeneous model. Astroglial cells

cultured directly from rat brain (cultured astroglia) have been shown to contain a relatively

homogeneous population of a2A-AR (see Chap. 3). In addition cultured astroglia contain

P-AR (McCarthy, 1983; Baker et al., 1986), a,-AR (Hirata et al., 1983; Murphy and

Pearce, 1987) and a2-AR (Richards et al., 1989) that are functionally coupled to signal

transduction pathways which include modulation of cAMP accumulation (P-AR and a2-

AR: Baker et al., 1986; Atkinson and Minnemann, 1991, 1992) and increases in

phosphoinositide (PI) hydrolysis (a,-AR: Wilson and Minnemann, 1990, 1991). One of

the products of PI hydrolysis is diacylglycerol which rapidly activates protein kinase C

(PKC). Therefore, because cultured astroglia contain functional a,-AR, a2-AR and P-AR

they provide an excellent model system in which to study regulation of a2A-AR by a

variety of pathways. No studies have investigated the role of heterologous regulation of

a2A-AR by pathways associated with a,-AR and P-AR. In addition, the regulatory

mechanisms for aZA-AR on astroglia are largely unknown. Therefore in this study, I have









19

investigated the role of these receptors and their signaling pathways in regulating

expression of a2A-AR.

My specific objectives are as follows:

I. Characterize the regulation of a2A-AR mRNA levels in astroglia by cAMP and

PKC

II. Characterize the regulation of a2A-AR mRNA levels in cultured astroglia by

endogenous ligands which cause cAMP accumulation or activate PKC.

III. Investigate the role of epinephrine, cAMP accumulation and PKC activation on

a2A-AR number and the ability of a2-AR agonists to inhibit cAMP production in

astroglia.














CHAPTER 2
MATERIALS AND METHODS


Materials


Adult Sprague-Dawley rats were from our breeding colony, which originated from

Charles River Farms, Wilmington, MA. Dulbecco's modified Eagles medium (DMEM)

and fetal bovine serum (FBS) were purchased from GIBCO, Grand Island, NY. [a-

32P]dCTP (3,000 Ci/mmol), [a-32P]UTP (3,000 Ci/mmol), ['H]-MK 912 (76 Ci/mmol),

cyclic AMP radioimmunoassay kits and GeneScreen nylon membrane were purchased

from Dupont-New England Nuclear, Boston, MA. Agarose and nonidet P-40 were

purchased from ICN Biomedicals, Inc., Aurora, OH. Forskolin (FSK), dibutyryl cyclic

AMP (db-cAMP), phorbol 12-myristate 13-acetate (PMA), phorbol 12,13-dibutyrate

(PDB), mezerein (Mez), 4a-phorbol, actinomycin D, (-)-epinephrine bitartrate (Epi),

prazosin, yohimbine, s-(-)-propranolol, angiotensin II (Ang II), and N-

tris(hydroxymethyl)methyl-2-aminoethanesulfonic acid (TES) were purchased from Sigma

Chemical, St. Louis, MO. 1,9-Dideoxyforskolin (DDF) and H-89 were purchased from

CalBiochem, La Jolla, CA. Sp-adenosine 3',5'-cyclic monophosphothioate triethylamine

(Sp-cAMPS), and UK 14304 were purchased from Research Biochemicals International,

Natick, MA. Random Primer DNA labeling kits were purchased from GIBCO, BRL.

RQI RNase-free DNase (lU/pl), proteinase K, ribonucleotides and restriction enzymes









21

were purchased from Promega Corp., Madison, WI. Plasmids (pGEM7) containing RG20

(a2A-AR) cDNA were kindly provided by Dr. Steven Lanier, Medical Univ. of South

Carolina. Plasmids (pUC19) containing Cu-Zn superoxide dismutase (SOD) cDNA were

a gift from Dr. Harry Nick, Univ. of Florida, Gainesville, FL. All other chemicals and

reagents were of molecular biology grade and were purchased from Fisher Scientific,

Pittsburgh, PA.


Methods

Astroglial Cultures


Astroglial cells were cultured from the hypothalamus and brain stem of adult rats

as described previously (Zelezna et al., 1992). Trypsin-dissociated brain cells, suspended

in DMEM containing 10% FBS, were plated in poly-L-lysine coated 100-mm culture

dishes (18 x 106 cells/dish) and were grown for 14-21 days in a humidified incubator at

37*C in 5% CO2-95% air. After this time, cells were dissociated from dishes using trypsin

and related in 100-mm culture dishes for nuclear runoff and binding experiments, and in

35-mm culture dishes for RNA isolation and cyclic AMP analysis. Astroglia were grown

in a humidified incubator at 370C in 5% CO2-95% air for 13-15 days before being used in

experiments. At that time the cultures consisted of>95% astroglia, as determined by

immunofluorescent staining with antibodies against glial fibrillary acidic protein (Sumners

et al., 1991).











Neuronal Cultures


Neuronal cultures were prepared from whole brains of newborn Sprague-Dawley

rats as described previously (Sumners et al, 1991). Trypsin-dissociated cells were

resuspended in DMEM containing 10% plasma-derived horse serum (PDHS) and were

plated in poly-L-lysine coated 35-mm culture dishes (3.0 x 106 cells per dish). Cells were

grown for 3 days in a humidified incubator at 37'C in 5% CO2-95% air, and were then

treated for 2 days with 10 pM cytosine arabinoside in fresh DMEM/10% PDHS. After

this, the medium with the cytosine arabinoside was removed and replaced with fresh

DMEM/10% PDHS, and the cells were allowed to grow for 13-15 days before use. Such

cultures consisted of approximately 90% neurons and 10% astrocytes (Sumners et al,

1991).


Drug Treatment


FSK, DDF, PMA, 4a-phorbol, Mez, PDB, and UK 14304 were dissolved in

dimethyl sulfoxide (DMSO) and diluted with filter sterilized water. Sp-cAMPS, db-

cAMP, yohimbine, and propranolol were dissolved in water. Prazosin was dissolved in

ethanol and diluted with filter sterilized water. Epinephrine was dissolved in 10 mM

ascorbate in phosphate buffered saline (PBS). Final dilutions were made into the

DMEM/FBS in the culture dish. Final dilutions for the cAMP experiments were made

into the DMEM without FBS in the culture dish.











Preparation of Cultured Astroglia for Light Microscopy


Cultured astroglia were prepared for light microscopy exactly as described (Ciotti

et al., 1989). Cultured astroglia were treated with control vehicle (0.5% DMSO or 100

pM ascorbate in PBS), 10 pM FSK, 500 nM PMA, or 100 pM Epi for 4 or 24 h. Dishes

of cultured astroglia were washed with ice-cold PBS and fixed with 2.5% glutaraldehyde

in PBS for 30 min. Glutaraldehyde was then removed and dishes were washed with PBS

for 2 min. PBS was then removed and the cells were covered with glia stain (0.5%

Coomassie brilliant blue in 50% isopropanol containing 1% formic acid) for 1 min. Glia

stain was removed and the dishes were washed twice with ice-cold PBS. Plastic

coverslips were mounted with Gel/Mount.

A Zeiss Axioskop microscope with phase contrast illumination was used to

observe the stained cells. Photographs were taken to document changes in cell

morphology. Phase-contrast light was used for all the photographs. A Minolta X-570

camera was used with Kodak Gold 100 ISO film. Shutter speed was determined by use of

the camera's automatic control. Magnification was 200X at the film surface.


RNA Preparation and Northern Blot Analysis


Levels of a2A-AR mRNA in astroglial cultures were determined with the use of a

2.2 kb fragment of the rat RG20 (a2A-AR) cDNA. This was produced by EcoR I and

Hind III digestion ofpGEM7, into which the RG20 cDNA had been cloned. This

fragment contained the entire RG20 coding region plus about 700 bp of the 3' untranslated









24

region. Levels of a2A-AR mRNA were normalized with the use of a 625 bp fragment of

the rat SOD cDNA. This fragment was produced by EcoR I digestion of pUC19, into

which the SOD cDNA had been cloned. The a2A-AR and SOD fragments were "p_

labeled by a random primer labeling system according to the manufacturers directions with

[a-32P]dCTP. a2A-AR and SOD mRNA was measured by Northern blot analysis of total

RNA extracted from astroglial cultures. Total cellular RNA was prepared by the

guanidinium-acid-phenol method (Chomczynski and Sacchi, 1987), and 20 Pg aliquots

were fractionated by 1% agarose-20 mM formaldehyde gel electrophoresis. Total RNA

was then transferred overnight by capillary action to GeneScreen nylon membrane and

crosslinked to the membrane by exposure to shortwave UV radiation (12,000 PJ/cm2)

using a Stratalinker (Stratagene, La Jolla, CA). Membranes were prehybridized overnight

at 45C in hybridization buffer containing 50% formamide, 5% dextran sulfate, 5X

sodium chloride-sodium phosphate-EDTA (SSPE; IX = 150 mM sodium chloride, 10 mM

sodium phosphate, 1 mM EDTA), 5X Denhardt's reagent (IX = 0.02% ficoll, 0.02%

polyvinylpyrrolidone, 0.02% BSA), 1% sodium dodecyl sulfate (SDS), and 100 pg/ml

denatured salmon sperm. Membranes were hybridized for 18 hours in fresh hybridization

buffer (45C) containing denatured "P-labeled a2A-AR and SOD probes. After

hybridization, membranes were washed twice (30 minutes each) in 2X SSPE/0.2% SDS at

50C and twice (15 minutes and 5 minutes) in 0.1X SSPE at 60C, and exposed to Kodak

XAR film for 24 hours. Membranes were hybridized simultaneously with a2A-AR and

SOD probes since signals for either message were no different when the probes were used

simultaneously or sequentially.











Using these procedures we were able to detect an approximately 4 kb a2A-AR

mRNA and a 0.7 kb SOD mRNA, consistent with those seen in other studies (Lanier et

al., 1991; Hamamdzic et al., 1995). Quantitation of the bands corresponding to a2A-AR

and SOD mRNA was achieved by arbitrary counting using a Molecular Dynamics

Phosphorimaging System. Normalization of the a2A-AR mRNA data was accomplished by

dividing the values obtained for each a2.-AR band by the value for the corresponding

SOD band. All a2A-AR mRNA data that are presented were normalized against SOD and

expressed as a percent of control values.


Nuclear Runoff Transcription Assay


Astroglial cultures were treated with 10 pM FSK, 500 nM PMA or control

solution (0.5% DMSO) for 1, 4, or 24 hours at 370C. Nuclei were then isolated from two

100-mm culture dishes from each treatment group as described previously (Greenberg and

Ziff, 1984). Dishes were washed twice in cold phosphate-buffered saline, astroglia were

removed with the aid of a rubber policeman, and pelleted at 500 g for 5 min The pellet

was resuspended in 4 ml NP40 lysis buffer (10 mM tris-HCI pH 74, 10 mM NaCI, 5 mM

MgCI2, 0.5% nonidet P-40), incubated on ice for 5 min and centrifuged at 500 g for 5 min.

The nuclear pellet was washed with NP40 lysis buffer and centrifuged at 500 g for 5 min.

The nuclei were resuspended in storage buffer (50 mM tris-HCI pH 8.0, 40% glycerol, 5

mM MgCI2, 0.1 mM EDTA, 0.1 mM DTT), frozen in an acetone/dry ice bath and stored

at -80C.











The runoff transcription was performed as described (Greenberg, 1988, Celano,

1989) with some modifications. Frozen nuclei (2.5-3 x 106/100 pl) were thawed on ice in

the presence of 100 pi reaction buffer (10 mM tris-HCI pH 8.0, 5 mM MgCI2, 300 mM

KCI, 5 mM DTT, 1 mM each ATP, CTP, GTP) and 10 p1 [a-2P]UTP (3,000 Ci/mmol).

The nuclei were incubated for 15 min at room temperature. DNA was digested by

addition of20U RNase-free DNase and 12 pl 20 mM CaCI2 and incubated for 5 min at

room temperature. Protein degradation was initiated by incubating the mixture for 30 min

at 37C with 27 pl SET (5% SDS, 50 mM EDTA, 10 mM tris-HCI pH 7.4) and 2 pl 10

mg/ml proteinase K. Yeast tRNA (5 pl of 10 mg/ml) was added and the newly

transcribed, labeled RNA was extracted by the guanidinium-acid-phenol method

(Chomczynski and Sacchi, 1987). The radiolabeled RNA was dissolved in TES solution

(10 mM TES pH 7.4, 10 mM EDTA, 0.2% SDS) to 2-3 x 106 cpm/ml within each

experiment, and was mixed with an equal volume of TES solution containing 600 mM

NaCI (final concentration 300 mM). Two ml of the combined solution was hybridized at

65C for 36 hr with 5 pg ofpGEM7 plasmid immobilized to a nylon membrane (as a

control) and to 5 pg immobilized plasmid containing inserts of the RG20 cDNA or SOD

cDNA. The membranes were washed twice (15 min each) in 2X SSC/0.2% SDS (IX

SSC = 150 mM NaCI, 15 mM sodium citrate pH 7.0) at 65C and once (5 min) in 0.1X

SSC at 65*C and were then exposed to Kodak XAR film with intensifying screens at -

80C for 2 weeks.











a2A-AR mRNA Stability


The stability of a2A-AR mRNA in control, FSK-treated and PMA-treated astroglia

was measured by incubating astroglia in 10 pg/ml actinomycin D to block transcription.

Astroglia were collected at various times after FSK or PMA treatment, total cellular RNA

was extracted, and a2A-AR mRNA levels were assessed by Northern blot analysis as

described above. The a2A-AR mRNA degradation rate (half-life) was calculated after

quantitation of the bands as described above using a Molecular Dynamics

Phosphorimaging System.


Binding Studies


Membranes from astroglial cultures were prepared as described (Richards et al.,

1989). Astroglia in 100 mm culture dishes were washed twice with ice-cold phosphate-

buffered saline (PBS, pH 7.2) and removed from the dishes with the aid of a rubber

policeman. Cells were centrifuged at 500 g for 5 min at 40C and the supernatant

discarded. The pellet was resuspended in ice-cold PBS, frozen in an acetone/dry ice bath

and stored at -800C.

On the day of the experiment, the cells were thawed on ice and resuspended in 3

ml of ice-cold binding buffer (33 mM tris-CI, pH 7.5, 1.0 mM EDTA, 0.1% ascorbic acid).

They were then homogenized using a Tekmar Tissuemizer (setting 50 for 30 s) and

centrifuged at 500 g for 5 min at 4C. The supernatant was centrifuged at 50,000 g for 10

min at 4*C. The resulting pellet was resuspended in 500 pl binding buffer containing 140









28

mM NaCI and the protein content of an aliquot determined by the method of Lowry et al.

(1951) using bovine serum albumin as the standard. The membrane preparation was kept

on ice until the binding assay was performed.

Radioligand binding was performed essentially as described (Uhlen and Wikberg,

1991) by incubating 10 pg of the membrane protein in 500 pl binding buffer containing

140 mM NaCI and ['H]-MK 912 for one hour at 250C. Nonspecific binding was

determined in the presence of 100 iM Epi. The assay was terminated by addition of 4 ml

ice-cold binding buffer with 140 mM NaCI and rapid filtration through Whatman GF/B

filter paper using a cell harvester (Brandel, Gaithersburg, MD). The membranes were

rapidly washed three times with 3 ml ice-cold binding buffer containing 140 mM NaCI.

The filters were transferred to scintillation vials containing 5 ml Liquiscint and counted in

an LKB RackBeta liquid scintillation counter at an efficiency of approximately 50% for

'H. All assays were performed in triplicate.


Cyclic AMP Extraction and Analysis


Thirty minutes prior to treatment, growth media were removed from the astroglial

cultures and replaced with fresh DMEM without FBS. Groups of 3 dishes were then

treated with the appropriate drugs or controls for various times by adding the solutions to

the media in the dishes. In experiments utilizing isobutyl methyl xanthine (IBMX), the

serum-free media were removed by aspiration and replaced with DMEM containing 2 mM

IBMX for 2 minutes before addition of FSK or PMA. In the experiments using clonidine,

the growth media were removed by aspiration and replaced with serum-free DMEM with









29

various concentrations of clonidine (0-10 pM) for 2 min before addition of FSK or control

solution (0.5% DMSO). All treatments were stopped by rapid aspiration of the media

followed by the addition of 0.5 ml 6% trichloroacetic acid (TCA) to each dish.

Extraction of cellular cyclic AMP was performed as previously described

(Scammell et al. 1995). The cells plus 6% TCA were removed from the dish with the aid

of a rubber policeman, transferred to glass tubes and centrifuged at 2,500 g at 40C for 15

minutes. The resulting pellets were dissolved in 1 ml of 1.0 M NaOH for protein

determination (Lowry et al., 1951). The supernatants were each extracted four times with

a total of 10 ml water-saturated ethyl ether. The ether phase was discarded and the

samples were evaporated to dryness using a SAVANT evaporator-concentrator and then

reconstituted in 0.5 ml 50 mM sodium acetate buffer (pH 6.2). Samples were stored

frozen at -20C until assayed for cyclic AMP.

The levels of cyclic AMP in cell samples were analyzed in duplicate by [125I]-cyclic

AMP radioimmunoassay kit according to the directions of the kit manufacturer (Dupont,

NEN). Samples were counted in a Beckman 5500 gamma counter. Sample values were

obtained by interpolation from the standard curve, corrected for the dilution of the original

sample, and expressed as pmol cyclic AMP per mg protein. The cross-reactivity of the

antibody with cyclic GMP and ATP was less than 0.01% each.


Statistics


Each data point was obtained from four 35-mm culture dishes for the mRNA

experiments and three 35-mm culture dishes for the cyclic AMP experiments. Data points









30
for the binding experiments were obtained from triplicate samples. Each experiment was

repeated at least three times. The data on all graphs and tables are presented as mean

SEM of the number of experiments indicated in the legend. Comparison between groups

was made with the use of ANOVA followed by the Newman-Keuls test to assess

statistical significance of the individual means. The results were considered statistically

significant at p < 0.05. Statistical analyses were performed using Sigma Stat software

(Jandel, San Rafael, CA).














CHAPTER 3
REGULATION OF a2A-ADRENERGIC RECEPTOR MRNA IN CULTURED RAT
ASTROGLIA: ROLE OF CYCLIC AMP AND PROTEIN KINASE C


Introduction


The expression and function of G-protein coupled receptors are highly regulated.

The mechanisms involved in this regulation are complex and are not well known for most -

G-protein coupled receptors. The P2-AR have been the model for study of G-protein

coupled receptors and their regulation (Sibley and Lefkowitz, 1985; Dohlman et al.,

1991), and have provided the basis for studies of other receptors. The regulation of 2-

AR can occur over time-frames of seconds to days. The short-term (seconds to minutes)

changes in receptor function are typified by phosphorylation of the receptor and

uncoupling from its signal transduction pathways (Hausdorffet al., 1990). In contrast,

long-term (hours to days) regulation involves changes in the steady-state levels of receptor

mRNA and protein (Bouvier et al., 1989). Expression and function of P2-AR can be

modulated by stimulation of 2-AR themselves, or by stimulation of other G-protein

coupled receptors, tyrosine kinase receptors or steroid receptors (reviewed by Lohse,

1993).

Like the P2-AR, the mechanisms involved in regulation of a2A-AR expression and

functional coupling to inhibition of adenylyl cyclase are complex. However, relatively









32

little is known about the homologous regulation of a2A-AR. The agonist-occupied a2A-AR

is a substrate for phosphorylation by PARK (Benovic et al., 1987; Liggett et al., 1992;

Kurose and Lefkowitz, 1994), and therefore is subject to homologous regulation at the

receptor level. Long-term agonist exposure causes a decrease in a2A-AR number (Jones et

al., 1990; Eason and Liggett, 1992).

The a2A-AR gene has been reported to be subject to heterologous regulation. This

regulation appears to vary depending on the cell model used in the study. Insulin and

growth factors were shown to decrease transcription of a2A-AR mRNA in the HT29 cell

line (Devedjian et al., 1991), but were without effect in pancreatic P-cell lines from the rat

or hamster (Hamamdzic et al., 1995). Glucocorticoids did not change expression of a2A-

AR in HT29 cells, but increased steady-state levels of mRNA in pancreatic P-cell lines

(Hamamdzic et al., 1995). Activation of PKC with phorbol esters did not effect levels of

a2A-AR mRNA in HT29 cells (Sakaue and Hoffman, 1991) or in pancreatic P-cell lines

(Hamamdzic et al., 1995). PKA activation as a result of increased cAMP accumulation

led to increased expression of the a2A-AR gene in HT29 cells (Sakaue and Hoffman,

1991). Taken together, these results suggest that a2A-AR gene expression may be

regulated in a manner specific to the cell line used in the study. However, regulation of

a2A-AR in cell lines may not be representative of regulation in non-transformed, non-

transfected cells. Studies of ac2-AR regulation in non-transfected, non-transformed cells,

however, are limited.

Therefore, in the present study we have used astroglial cells cultured directly from

adult rat brain as a model system to study a2A-AR regulation. These non-transformed cells











have been shown to contain high levels of a,-AR protein (Richards et al. 1989) and aA-

AR mRNA, with very little or no a2c-AR mRNA. Our studies indicate that both increases

in cellular cyclic AMP levels and PKC activation decrease steady state levels of a2A-AR

mRNA via decreased transcription of the a2A-AR gene.


Results


Northern analysis of total RNA from astroglial and neuronal cultures and from

various rat brain regions revealed a ~4.0 kb transcript that hybridized specifically with the

a2A-AR (RG20) cDNA probe, indicating that this probe recognized the same transcript in

the cultures and in the brain regions studied (Fig. 3-1). The mRNA detected in the present

study was of similar size and distribution as the transcript reported in other studies (Lanier

et al, 1991; Zeng and Lynch, 1991; Hamamdzic et al., 1995). For example, RNA isolated

from cortex, hypothalamus, and brain stem appeared to contain more azA-AR mRNA than

striatum or cerebellum. This is in good agreement with previous studies which found that

a2c-AR, rather than a2A-AR, predominate in the striatum and cerebellum (Zeng and Lynch,

1991; Nicholas et al, 1993). When northern blots of total RNA from neuronal cultures

were hybridized with an a2c-AR probe (RG10; Lanier et al, 1991) a transcript of much

smaller size (~2.5 kb) was recognized (data not shown). In addition, when northern blots

of total RNA from cultured astroglia were hybridized with the RG0I probe, no transcript

was recognized (data not shown), suggesting that these cultures contain no a2-AR.

Overall, these data indicate that the RG20 probe used here specifically recognizes the a2A-

AR subtype.










Effects of Forskolin on aA-AR mRNA Levels


Previous studies showed that cyclic AMP regulates a2A-AR mRNA levels in HT29

cells (Sakaue and Hoffman, 1991) Therefore, in the first series of experiments, we

investigated the effects of an adenylyl cyclase activator forskolin (FSK) on steady state

a2A-AR mRNA levels in astroglia as a function of incubation time and concentration.

Treatment of astroglial cultures with FSK resulted in a time-dependent decrease in steady

state a2A-AR mRNA levels (Fig. 3-2). Reduction of a2A-AR mRNA levels was apparent

after 1 hour and was significantly different from control at 2 hours. A 90% decrease in the

a2A-AR mRNA was seen within 3 hours. Levels of a2A-AR mRNA gradually increased

toward 50% of control values at 48 and 72 hours. The FSK-induced decrease in a2A-AR

mRNA level was also concentration dependent (Fig. 3-3). A trend toward reduction of

a2A-AR mRNA levels was seen at 100 nM FSK after 4 hours, while 1 pM FSK

significantly reduced the levels of this message to about 20% of control. The FSK analog

DDF (10 pM), which does not activate adenylyl cyclase did not induce any change in

steady state a2A-AR mRNA levels (Fig. 3-3A).

FSK has been shown to stimulate adenylyl cyclase resulting in an increase in

intracellular cyclic AMP. Therefore, in the next set of experiments, we assessed the effect

of two membrane permeable, non-hydrolyzable cyclic AMP analogs on steady state a2A-

AR mRNA levels in order to indicate whether the action of FSK on a2A-AR mRNA levels

was mediated by increased cyclic AMP levels. Figure 3-4 shows that both db-cAMP and

Sp-cAMPS induced decreases in a2A-AR mRNA levels after 4 hours. Levels of aZ,-AR










mRNA were significantly reduced (p<0.05 compared to control) by 100 pM db-cAMP,

while 1 mM db-cAMP produced a reduction to about 10% of control values after 4 hours.

Levels of a2A-AR mRNA were decreased 60% with 100 pM Sp-cAMPS treatment. The

effects of db-cAMP and Sp-cAMPS on steady state a2A-AR mRNA levels were similar to

those observed with FSK (Fig. 3-4B), suggesting a common mechanism of action through

increases in cyclic AMP levels.


Effects of PMA on a2A-AR mRNA Levels


PKC activation was reported to suppress the intracellular signaling pathway

associated with the a2A-AR in platelets (Jakobs et al., 1985; Katada et al., 1985; Watanabe

et al., 1985), and to be without effect on a2A-AR regulation in cell lines (Sakaue and

Hoffman, 1991; Hamamdzic et al., 1995). Therefore we studied the effects of a PKC

activator, PMA, on steady state a2A-AR mRNA levels in astroglial cultures. PMA induced

a time-dependent decrease in c2A-AR mRNA levels (Fig. 3-5). The effect of PMA (500

nM) on a2A-AR mRNA levels was noticeable at 1 hour and decreased to about 10% of

control at 3 hours. This effect began to wane at 24 hours and levels of a2-AR mRNA

approached control values at 48 and 72 hours after treatment. Four hour treatment with

10 nM PMA produced a significant (p<0.05) depression (75% of control) in levels of a2A-

AR mRNA, while 50 nM PMA reduced message levels to 10% of control levels (Fig. 3-

6). The phorbol ester 4a-phorbol, which does not activate PKC, did not cause any change

in a2^-AR mRNA levels (Fig. 3-6A).









36
Since PMA is known to cause an activation of PKC, we next treated the cells with

two different PKC agonists, PDB and Mez, Both PDB and Mez decreased steady state

a2A-AR mRNA levels after a 4 hour treatment (Fig. 3-7). Levels of a A-AR mRNA were

reduced 80-90% after 50 nM treatment with either drug. PDB and Mez produced

reductions in a2A-AR mRNA levels similar to those observed with PMA (Fig. 3-7B).

In a further set of experiments we tested the combined effects of FSK and PMA on

c2A-AR mRNA levels. Astroglial cultures were treated with concentrations of FSK (100

nM) and PMA (10 nM) which produced approximately half-maximal decreases in a2A-AR

mRNA levels (42% and 45% of control levels, respectively) after 4 hours. When FSK and

PMA were applied together, they did not produce significantly greater decreases in levels

of a2A-AR mRNA than the decreases produced when either drug was applied individually

(Fig. 3-8).


Effect of FSK and PMA on Astroglial Morphology


Treatment of cultured astroglia with FSK or PMA resulted in stellation of the cells

(Fig. 3-9). This morphological change was notable within 1 h of treatment and reversed

48-72 h after drug treatment (not shown). FSK treatment produced higher levels of

astroglial stellation which lasted longer (Fig. 3-9A) than PMA treatment (Fig. 3-9B).

Combined treatment with FSK and PMA produced morphological changes similar to those

seen with FSK treatment alone (not shown). These morphological changes were initially

thought to contribute to the decreases in levels of a2A-AR mRNA. However, as will be

seen in the next chapter, Epi produced reductions in levels of c2A-AR mRNA similar to










those obtained with FSK and PMA, but with no accompanying morphological change

(Fig. 4-4). Furthermore, the stellation produced by FSK or PMA is reversed 48-72 h after

treatment, at which time a2A-AR mRNA levels are still significantly reduced (Fig. 3-2 and

3-5). For these reasons, it is unlikely that the observed changes in aA-AR mRNA are due

to cell stellation.


Effect of FSK and PMA on a2A-AR mRNA Gene Expression and Stability


Two possible mechanisms could contribute to the observed decrease in steady

state a2A-AR mRNA levels: decreased transcription of the gene or decreased stability of

the transcript. The effect of FSK and PMA on the rate of transcription was assessed using

nuclear runoff assays. The representative blots presented in Fig. 3-10A demonstrate that

both FSK and PMA decreased the rate of transcription of a2A-AR mRNA by 50% within 1

hour of treatment. Larger (-65%) decreases were detected following 24 hour treatment

with FSK or PMA. These results followed the pattern of decrease ofa2A-AR mRNA

steady state levels. This experiment was repeated three times with similar findings. In the

results shown in Fig. 3-10, and in the repeat experiments, neither FSK nor PMA

treatments significantly changed SOD mRNA transcription.

Stability of aA-AR mRNA was investigated by treating astroglial cultures for 30

min with actinomycin D at a final concentration of 10 mg/ml to inhibit mRNA

transcription. The cells were then treated with 10 pM FSK, 500 nM PMA or control

solution (0.5% DMSO). Total RNA was extracted hourly for 5 hours and northern blot

analysis performed as described in the Chapter 2. The amount of RNA recovered did not










vary over the course of each experiment, nor did the intensity of the SOD signal. The

results indicate that FSK decreased a2A-AR mRNA half-life by 36.7 0.1% compared to

control values (Fig. 3-12), although the decrease was not statistically significant compared

to control values (n=3; p<0.05). PMA, however, significantly (n=3; p < 0.05) increased

a2A-AR mRNA half-life by 80.0 0.1% compared with control values. The control half-

life was 3.0 0.4 hours (n=6).


Effect of FSK and PMA on Cyclic AMP Accumulation


Experiments with cyclic AMP analogues suggested that increases in intracellular

cyclic AMP mediated the effects of FSK. Previous studies also suggested that PKC can

directly activate adenylyl cyclase and result in intracellular cyclic AMP accumulation

(Kitten et al., 1994; Zhou et al., 1994). Therefore we measured cyclic AMP accumulation

in cultured astroglia in response to FSK or PMA treatments.

In the first set of experiments we used a phosphodiesterase inhibitor, IBMX, so

that we could detect changes in cAMP synthesis. As expected, FSK caused a time-

dependent increase in cyclic AMP accumulation that was significantly (p<0.05) higher

than control levels at all times observed (Table 3-1). In contrast, PMA treatment did not

significantly increase cyclic AMP levels above control values, suggesting that in cultured

astroglia PMA does not effect cAMP synthesis via action at adenylyl cyclase.

Experiments in the absence of IBMX were carried out to determine whether PMA could

increase cAMP levels by inhibition of phosphodiesterase activity. PMA treatment did not

change cAMP accumulation, but FSK treatment produced time-dependent increases in









39
cAMP levels (Table 3-2). In order to determine the duration of the FSK-induced increase

in cAMP levels we extended FSK treatments to cover the same time-frame used for the

mRNA studies. These results showed that cAMP levels peaked 30 min after FSK

treatment, and returned to control levels within 1 h (Fig. 3-12). This suggests that the

effect of FSK on a2A-AR gene expression is rapidly transduced and does not require long-

term elevations of cAMP levels to cause long-term decreases in levels of c2A-AR mRNA.


Discussion


In this study, we have demonstrated that both FSK and PMA treatments decrease

steady state levels of a2A-AR mRNA in astroglial cultures. This decrease appears to be

the result ofintracellular cyclic AMP accumulation and PKC activation, respectively.

These conclusions are supported by observations that (a) treatment of the cultures with

the cyclic AMP analogs, db-cAMP or Sp-cAMPS, elicited decreases in a2A-AR mRNA

levels similar to those observed with FSK, (b) treatment of the cultures with FSK caused

increases in intracellular cyclic AMP concentrations, and (c) treatment of the cultures with

the PKC activators, PDB or Mez, caused decreases in a2A-AR mRNA levels similar to

those seen with PMA. Two possible mechanisms could contribute to these decreases in

steady state mRNA levels: decreased transcription of the gene or decreased stability of

the transcript. Our studies indicate that both FSK and PMA treatments caused a decrease

in the transcription rate of the 2A-AR gene.

In HT29 human colonic adenocarcinoma cells, cyclic AMP was shown to

transiently increase a2^-AR mRNA levels (Sakaue and Hoffman, 1991). In the present











study, our data indicate that increases in cellular cyclic AMP levels induce a decrease in

a2A-AR mRNA levels in astroglia cultured from rat brain. These different results may

reflect differences in the lineage and culture conditions of the two cell types. HT29 cells

are a human colonic adenocarcinoma cell line, while astroglia used in this study were not

transformed and were cultured directly from adult rat brain. It is possible that different

regulatory processes operate in transformed cells cultured from the human periphery than

in non-tumor cells cultured from the rat CNS. In addition to lineage, the two cells types

are maintained in culture differently prior to experimental treatment. For example, HT29

cells were maintained in frequently-changed growth medium containing 7% FBS and were

used just after becoming confluent. Astroglia used in the present study were maintained in

growth medium containing 10% FBS until four or five days post-confluency, and the

growth medium was not changed during the astroglial culture period (13-15 days). The

degree of confluency and frequency of growth medium change in HT29 cells may play a

role in the regulatory processes seen in these cells. HT29 cells have been shown to

increase expression of a2A-AR mRNA and protein with increasing cell density (Paris et al,

1985; Sakaue and Hoffman, 1994). In addition, expression of a2A-AR in HT29 cells is

reported to decrease with increasing FBS concentration in the growth medium; an effect

which was mimicked by insulin and growth factors (Devedjian et al, 1991). Perhaps the

cyclic AMP-mediated increase in a2A-AR mRNA seen by Sakaue and Hoffman (1991) was

due to inhibition of the actions of frequent growth medium changes with fresh FBS. In the

present study, astroglia were cultured for two weeks without changing the FBS-containing

growth medium. Changing the growth media on cultured astroglia resulted in decreases in











a2A-AR mRNA levels regardless of FBS content (data not shown). Therefore, culture

conditions used in the present study may minimize interactions due to frequent growth

medium changes.

Using a chloramphenicol acetyl transferase reporter system in JEG-3 cells Sakaue

and Hoffman (1991) found that the 5'-flanking region of the human c2A-AR gene

conferred responsiveness to cyclic AMP and PKA. Analysis of the human a2A-AR

promoter region, however, did not reveal a cyclic AMP responsive element (CRE; Handy

and Gavras 1992). Similarly, a CRE was suggested but not reported to be present in the

promoter region of the rat a2A-AR gene (Handy et al., 1995). In our studies, FSK

treatment caused a decrease in transcription of the a2A-AR gene as measured by nuclear

runoff assay, suggesting the possibility that the gene is regulated differently in cultured

astroglia than in HT29 or transfected JEG-3 cells.

Studies have suggested that phorbol esters do not change transcription of the a2A-

AR gene in HT29 cells or pancreatic p-cells (Sakaue and Hoffman, 1991; Hamamdzic et

al., 1995). In the present study, PMA decreases transcription and increases the stability of

a2A-AR mRNA in cultured astroglia. Decreased transcription appears to play a greater

role, since the net effect of PMA treatment was a decrease in a2A-AR mRNA steady state

levels. In addition, elements which may confer responsiveness to PKC have not been

reported for the human or rat a2A-AR promoter regions, suggesting that PKC activation in

astroglia may decrease transcription via some other mechanism.

Previous studies have shown that activated PKC can directly stimulate adenylyl

cyclase, resulting in intracellular cyclic AMP accumulation (Kitten et al., 1994; Zhou et









42

al., 1994). The present study showed that activation of PKC with PMA did not result in

intracellular accumulation of cyclic AMP. These results, along with the nearly identical

time course of action of PMA and FSK, suggested that the effects of FSK and PMA may

converge at some common pathway. We found, however, that the combined effect of

PMA and FSK was not strictly additive (Fig. 3-8), arguing against convergence on a

common pathway. This raises the possibility that the FSK- and PMA-mediated decreases

in rat a2A-AR gene transcription observed in the present study and the FSK-mediated

increase in transcription of the human ac2-AR gene seen by Sakaue and Hoffman (1991)

may be due to changes in activity at competing promoter regions. It is possible that PKA

and PKC can modulate transcription factors which may compete for the same promoter

region of the a2A-AR gene. The suppression of a2A-AR gene expression would then be

dependent on the relative activities of the effected transcription factors.

The human and the rat a2A-AR gene contain common motifs in the 2 kb region

upstream of the protein coding region. These include a TATA-box, a GC-box, a

palindromic region, and several GGAGG repeats (Handy and Gavras, 1992; Handy et al.,

1995). The GC-box is a consensus binding site for the nuclear factor Spl (Dynan and

Tjian, 1983) which acts to enhance transcription. In the rat a2A-AR promoter, however,

the region overlapping the GC-box was shown to repress gene expression (Handy et al.,

1995). This region contains several GGAGG repeats and a related CGAGG sequence

(Handy et al., 1995) which have been postulated to bind nuclear factors which can repress

gene expression (Giovane et al., 1994; Macleod et al., 1992). Handy et al. (1995) showed

that several nuclear factors, including Spl, bind to the GC-box region. In addition,









43
phosphorylation of Spl by PKC appears to inactivate this transcription factor and lead to

decreased DNA binding (Borellini et al., 1990; Leggett et al., 1995). These studies, in

light of our results, suggests that cyclic AMP accumulation or PKC activation could result

in changes in nuclear factor binding such that transcription of the ca2-AR gene is

repressed. Since the effects of cyclic AMP are mediated at the nuclear level by PKA

(Riabowol et al, 1988), this kinase may be involved in the changes in transcription we

observed.

The long-lasting decrease in levels of a2A-AR mRNA after FSK or PMA treatment

is not unusual. Incubation ofDDT, MF-2 hamster vas deferens cells with agents which

increase cAMP accumulation leads to a 60% reduction in levels of P2-AR mRNA which

persists for at least 72 h (Hadcock and Malbon, 1988; Hadcock et al., 1989). Similarly,

insulin induces a 90% reduction in levels of a2A-AR mRNA which is transcriptionally

mediated and lasts 48 h in HT29 cells (Devedjian et al., 1991). In the present study, FSK

and PMA also decrease a2A-AR gene expression. In the case of FSK (Fig. 3-12) and Epi

(see Chap. 4, Fig. 4-2) these long-term effects on a2A-AR mRNA are mediated by short-

term accumulation of cAMP or short-term stimulation of P-AR and c -AR. These results

suggest that the long-term depression of levels of a2A-AR mRNA are mediated by long-

term changes in cellular effectors distal to the second messenger pathways. This could

include persistent changes in the expression or activity of nuclear trans-acting factors.

One question that arises from the present studies is the identity of endogenous

factors which may modulate a2A-AR mRNA via changes in cyclic AMP or PKC activity.

Our studies indicate that Epi and angiotensin II (Ang II) treatments cause decreases in











a2A-AR mRNA levels in cultured astroglia similar to those reported here (Chap. 4).

Norepinephrine applied to cultures of astroglia acts at P-AR and az,-AR to modulate

cyclic AMP formation, and at a,-AR and a2A-AR to cause phosphoinositide (PI)

hydrolysis (Atkinson and Minneman, 1991, 1992; Wilson and Minneman, 1990, 1991).

Previous studies from this lab have shown that Ang II also increases PI hydrolysis in

cultures of astroglia (Sumners et al., 1991). One of the products of PI hydrolysis is

diacylglycerol, which rapidly activates PKC (Nishizuka, 1986). Thus, it appears that

endogenous ligands may cause accumulation of cyclic AMP and/or activation of PKC and

therefore also regulate a2A-AR mRNA levels. This hypothesis will be investigated in

Chapter 4.

In summary, the present results show that both FSK and PMA treatments decrease

2A-AR mRNA levels in astroglial cultures. These effects appear to be mediated via two

different pathways: increased cyclic AMP concentrations and PKC activation. The

implications of these findings for understanding a2A-AR regulation and the physiological

role of a2A-AR on astroglia will be interesting to explore.











Table 3-1. Effect of FSK and PMA on cyclic AMP accumulation in the presence
of IBMX


Time (min)

Treatment 5 10 20 30

Conl 33.5 47.5 67.2 78.5
S5.5 7.3 68 8.5

0 M FSK 1,460 2,964 5,522 6,691
SFS 301 576 1,014 1,038

Conl 36.7 42.2 61.7 76.5
S4.2 1.7 4.6 6.8

500 nMPMA 45.6 60.7 74.6 89.6
500 nM PMA
11.0 8.2 6.2 10.6


Values (pmol cyclic AMP/mg protein) are the means SE of 4 experiments. Astroglial
cultures were incubated at 370C in DMEM containing 2 mM IBMX with the above for
the times indicated. After incubations, cellular cyclic AMP was extracted and analyzed as
detailed in Materials and Methods. significantly different (p < 0.05) from control.



Table 3-2. Effect of FSK and PMA on cyclic AMP accumulation in the absence of
IBMX

Time (min)

Treatment 5 10 30

Control 11.5 0.1 11.1 0.7 10.5 1.4

10 pM Forskolin 446 119' 947 347' 1,178 475'

500 nM PMA 11.0 0.9 10.2 0.9 10.9 0.7

Values (pmol cyclic AMP/mg protein) are the means SE of 3 experiments. Astroglial
cultures were incubated at 37C in DMEM with the above for the times indicated. After
incubations, cellular cyclic AMP was extracted as detailed in Materials and Methods.
*significantly different (p < 0.05) from control.






























Q,4 *W


I :_

< U
2


Figure 3-1. Representative northern blot showing relative steady-state levels of the 4.0
kb aC2-AR mRNA in 20 pg total RNA isolated from cultures of astroglia and neurons, and
from different brain areas. AG, astroglial cultures; NCult, neuronal cultures; Ctx, cortex;
Str, striatum; Hyp, hypothalamus; Cer, cerebellum; BS, brain stem.


4.0 kb >
RG20 W












4.0 kb lo
RG 20



Cu/Zn SOD ....


+ + + + + + +
Time(h) 1 2 3 4 24 48 72


B

0


2;
0
oI4
0

z


E
-<:


1 2 3 4 24 48 72


Time (h)


Figure 3-2. Effect of FSK on steady-state levels of a2A-AR mRNA in cultured
astroglia as a function of treatment time. Cultured astroglia were incubated in their
growth media in the absence or presence of FSK (10 pM) for the indicated time periods,
followed by analysis of ca2-AR mRNA levels as detailed in Chapter 2. (A) Representative
northern blot showing 4.0 kb a2A-AR (RG20) mRNA and 0.7 kb Cu/Zn SOD mRNA in
each treatment situation. (-), control; (+), Epi-treated for the indicated time in h. (B)
Quantification of 2A-AR mRNA data normalized against Cu/Zn SOD mRNA. Data are
means SE from 3 independent experiments and are presented as % of control levels
(100%). Control data did not vary significantly with time and are plotted on the y-axis.
*significantly different (p < 0.05) from control.










4.0 kb .
RG20

0.7 kb
Cu/Zn SOD


Forskolin


9 8 7 6 5


[Forskolin] -log M

Figure 3-3. Effect of FSK on steady-state levels of a2A-AR mRNA in cultured
astroglia as a function of concentration. Cultured astroglia were incubated in their
growth media with vehicle (0.5% DMSO in water) or the indicated concentrations of FSK
for 4 hours, followed by analysis of aA-AR mRNA levels as detailed in Chapter 2. (A)
Representative northern blot showing 4.0 kb a2A-AR (RG20) mRNA and 0.7 kb Cu/Zn
SOD mRNA in each treatment situation. Dideoxyforskolin (DDF) is included as a
negative control. (B) Quantification of a2A-AR mRNA data normalized against Cu/Zn
SOD mRNA. Data are means SE from 4 independent experiments and are presented as
% of control levels (100%). Control data are plotted on the y-axis. *significantly
different (p < 0.05) from control.


g 83gg g-









A
4.0 kb I
RG20



0.7kb 4
Cu/Zn SOD


C 0 0 0
U- 0

L Sp-cAMPS db-cAMP


B
0
o
5S
0



0
E<

E
<,
,<
a


- "

t Sp-cAMPS db-cAMP


Figure 3-4. Effect of the cyclic AMP analogs Sp-cAMPS and db-cAMP on steady-
state levels of a2A-AR mRNA in cultured astroglia. Cultured astroglia were treated
with vehicle (water) or the indicated concentrations of Sp-cAMPS or db-cAMP for 4
hours, followed by analysis of a2A-AR mRNA levels as detailed in Chapter 2. (A)
Representative northern blot showing 4.0 kb a2A-AR (RG20) mRNA and 0.7 kb Cu/Zn
SOD mRNA in each treatment situation. FSK (250 nM) was included as a positive
control. (B) Quantification of a2A-AR mRNA data normalized against Cu/Zn SOD
mRNA. Data are means SE from 4 independent experiments and are presented as % of
control levels (100%). *significantly different (p < 0.05) from control.












4.0 kb 1> gghii U



0.7 kb
Cu/Zn SOD
+ -+ + + + +
Time(h) 1 2 3 4 24 48 72


24 48 72


Time (hrs)

Figure 3-5. Effect of PMA on steady-state levels of a2A-AR mRNA in cultured
astroglia as a function of treatment time. Cultured astroglia were incubated in their
growth media in the absence or presence of PMA (500 nM) for the indicated time
periods, followed by analysis of a2A-AR mRNA levels as detailed in Chapter 2. (A)
Representative northern blot showing 4.0 kb a2A-AR (RG20) mRNA and 0.7 kb Cu/Zn
SOD mRNA in each treatment situation. (-), control; (+), Epi-treated for the indicated
time in h. (B) Quantification of a2A-AR mRNA data normalized against Cu/Zn SOD
mRNA. Data are means SE from 3 independent experiments and are presented as % of
control levels (100%). Control data did not vary significantly with time and are plotted on
the y-axis. *significantly different (p < 0.05) from control.










4.0 kb
RG20

0.7 kb
Cu/ZnSOD


B



I-
0
U
0


5
cl
E


Of


"gI5-

,.-,ma om



0 o I

PMA


9 8 7 6


[PMA] -log M


Figure 3-6. Effect of PMA on steady-state levels of a2A-AR mRNA in cultured
astroglia as a function of concentration. Cultured astroglia were incubated with vehicle
(0.5% DMSO in water) or the indicated concentrations of PMA for 4 hours, followed by
analysis of a2A-AR mRNA levels as detailed in Chapter 2. (A) Representative northern
blot showing 4.0 kb a2A-AR (RG20) mRNA and 0.7 kb Cu/Zn SOD mRNA in each
treatment situation. 4a-phorbol is included as a negative control. (B) Quantification of
a2A-AR mRNA data normalized against Cu/Zn SOD mRNA. Data are means SE from 4
independent experiments and are presented as % of control levels (100%). Control data
are plotted on the y-axis. *significantly different (p < 0.05) from control.












I-eW


-. -F IU 9 C ic Ir


68e6l46O6e*0


0


C-

100

80

60

40

20

0


0 0 0 0 0 0 0 0 0
PD M z 00 rei
W)

PDB Mezerein


-0000 -0000
-- ..O0 -'r~ O

PDB Mezerein


Figure 3-7. Effect of the PKC agonists PDB and MEZ on steady-state levels of
a2A-AR mRNA in cultured astroglia. Cultured astroglia were treated with vehicle
(DMSO in water) or the indicated concentrations of PDB or MEZ for 4 hours, followed
by analysis of a2A-AR mRNA levels as detailed in Chapter 2. (A) Representative northern
blot showing 4.0 kb a2A-AR (RG20) mRNA and 0.7 kb Cu/Zn SOD mRNA in each
treatment situation. PMA (20 nM) was included as a positive control. (B) Quantification
of a2A-AR mRNA data normalized against Cu/Zn SOD mRNA. Data are means SE
from 4 independent experiments and are presented as % of control levels (100%).
*significantly different (p < 0.05) from control.


4.0 kb .
RG20


0.7 kb
Cu/Zn SOD>


o
I^-*
B -
0
U






o


a5












4.0 kb
RG20


0.7 kb
Cu/Zn SOD


0 o
o +
C 0-


100 r


-


40 F-


Figure 3-8. Effect of combined FSK and PMA treatment on steady-state levels of
a2A-AR mRNA in cultured astroglia. Cultured astroglia were treated with vehicle
(DMSO in water), 100 nM FSK, 10 nM PMA and a combination of 100 nM FSK and 10
nM PMA for four hours, followed by analysis of a2A-AR mRNA levels as detailed in
Chapter 2. (A) Representative northern blot showing 4.0 kb a2A-AR (RG20) mRNA and
0.7 kb Cu/Zn SOD mRNA in each treatment situation. (B) Quantification of a2A-AR
mRNA data normalized against Cu/Zn SOD mRNA. Data are means SE from 2
independent experiments and are presented as % of control levels (100%). *significantly
different (p < 0.05) from control.


;t~P

~,~





































b












Figure 3-9. Effect of FSK and PMA on morphology of cultured astroglia.
A. Cultured astroglia were incubated with control solution or FSK (10
pM) for 4 or 24 hours at 37C, followed by preparation of the cells for
light microscopy as detailed in Chapter 2. (a) Micrograph of cultured
astroglia after 4 h treatment with control solution. (b) Micrograph of
cultured astroglia after 4 h treatment with FSK. (c) Micrograph of
cultured astroglia after 24 h treatment with FSK. All micrographs are
shown enlarged 700X.


4ff

























b























Figure 3-9. Continued
B. Cultured astroglia were incubated with control solution or PMA (500
nM) for 4 or 24 hours at 370C, followed by preparation of the cells for
light microscopy as detailed in Chapter 2. (a) Micrograph of cultured
astroglia after 4 h treatment with control solution. (b) Micrograph of
cultured astroglia after 4 h treatment with PMA. (c) Micrograph of
cultured astroglia after 24 h treatment with PMA. All micrographs are
shown enlarged 700X.

















S pGEM
1 h RG20

SOD

Con FSK PMA

TE
4 h RG20

SOD

Con FSK PMA

TE

24 h RG20
SOD




Figure 3-10. Effect of FSK and PMA on transcription rate of a2A-AR mRNA.
A. Cultured astroglia were treated with control solution (DMSO in water), 10 iM
FSK or 500 nM PMA for the times indicated, followed by nuclear runoff assay as
described in Chapter 2. Shown here are representative blots for each treatment at
1,4 and 24 hours. TE (10 mM tris-HCI pH 8.0/1 mM EDTA) and pGEM7
plasmid were included as negative controls. Similar results were obtained in 3
repeats of these experiments (2 for the 4 h treatment).
pGEM pGEM7 plasmid (1 hour point)
TE (4 and 24 hour points)
aA-AR pGEM7 plasmid containing the RG20 (a2A-AR) cDNA
SOD pUC19 plasmid containing the SOD cDNA


















S 120


1001


80 -



40 4-



2 20 -



Z3 0 1 2 3 4 24

Time (hrs)







Figure 3-10. (continued)
B. Cultured astroglia were treated with control solution (DMSO in water), 10 PM
FSK (0) or 500 nM PMA (0) for the times indicated. Quantification ofa2A-AR
mRNA runoff data normalized against Cu/Zn SOD mRNA runoff data. Data are
means SE of 3 independent experiments (2 for the 4 h point) and are presented
as % of control levels (100%). Control data are plotted on the y-axis.
*significantly different (p<0.05) from control.






















100





-







10 I I I I
1 2 3 4 5

Time



Figure 3-11. Effect of FSK and PMA on degradation rate of a2A-AR mRNA.
Cultured astroglia were treated with control solution (DMSO in water;@), 10 PM FSK
(0) or 500 nM PMA (A) for the times indicated, followed by analysis of a2A-AR mRNA
levels as detailed in Chapter 2. (A) Representative northern blot showing 4.0 kb a2A-AR
(RG20) mRNA and 0.7 kb Cu/Zn SOD mRNA in each treatment situation. (B)
Quantification of a2A-AR mRNA data normalized against Cu/Zn SOD mRNA. Data are
means from 3 independent experiments and are presented as % of control levels (100%).


















1400

1200 -

1000

5 800

E 600

400

O 200


0 1 2 3 4 24 48 72

Time (h)




Figure 3-12. Effect of FSK on cAMP levels as a function of time. Cultured astroglia
were treated with vehicle control (0.5% DMSO; 0) or 10 pM FSK (U) for the times
indicated. Following this the media were removed, the dishes were washed twice with ice-
cold PBS, and cellular cAMP was extracted and analyzed as detailed in Chapter 2. Values
are the means SE of 4 experiments. significantly different (p<0.05) compared to
control levels.














CHAPTER 4
REGULATION OF a2A-ADRENERGIC RECEPTOR MRNA IN CULTURED RAT
ASTROGLIA: ROLE OF EPINEPHRINE AND ANGIOTENSIN II


Introduction


Three subtypes of a2-adrenergic receptors (a2-AR), the a2A-AR, a2B-AR, and a2C-

AR, have been identified using pharmacological and molecular cloning approaches

(Bylund et al., 1994). All three subtypes can be activated by the endogenous

catecholamines epinephrine (Epi) and norepinephrine (NE), and inhibit the accumulation

of cellular cyclic AMP via inhibition ofadenylyl cyclase (Limbird, 1988). The

physiological functions of the different subtypes appear to be determined by their pattern

of cellular and tissue localizations For example, activation of a2-AR in the central

nervous system appears to mediate decreases in blood pressure (MacMillan et al., 1996),

while activation of a2B-AR on resistance vessels increases blood pressure (Link et al.,

1996). The differences in cell and tissue distribution of a2-AR may also contribute to

differing modes of regulation of expression of each subtype. Regulation of central a2A-AR

represents an important way through which adrenergic signaling is modulated. Up-

regulation of a2A-AR could potentiate the inhibitory actions of ac-AR agonists, whereas

down-regulation may lead to increased activity of adrenergic signaling pathways. An

understanding of the regulation of expression of a,-AR in a variety of cell types and











tissues may lead to novel methods to affect changes in the function of aA-AR without

changing the expression or function of am-AR or a2c-AR.

Information regarding a2A-AR regulation has come largely from the use of cell

lines and transfected cells as model systems (Thomas and Hoffman, 1986; Convents et al.,

1989; Jones et al., 1990; Sakaue and Hoffman, 1991). Experiments in such transformed

cells allow study of receptor regulation within a relatively homogeneous model. However,

regulatory processes in transformed cells may not be representative of physiological

regulation, and little is known concerning the basic cellular mechanisms of regulation of

a2A-AR in non-transformed, non-transfected cells. Astroglial cells cultured directly from

rat brain are not transformed or transfected and have been shown to contain a2-AR

(Richards et al., 1989) which are predominantly of the a2A-AR subtype (Chap. 3). In

addition, cultured astroglia also contain p-AR (McCarthy, 1983; Baker et al., 1986) and

a,-AR (Hirata et al., 1983; Murphy and Pearce, 1987). These receptors are functionally

coupled to signal transduction pathways which include modulation of cyclic AMP

accumulation (P-AR and a2-AR: Baker et al., 1986; Atkinson and Minnemann, 1991,

1992) and increases in phosphoinositide (PI) hydrolysis (a,-AR: Wilson and Minnemann,

1990, 1991). PI hydrolysis leads to increased intracellular calcium and activation of

protein kinase C (PKC) via diacylglycerol (Berridge, 1984; Nishizuka, 1986).

Consequently, cultured astroglia provide a good model system for studying the

homologous (via a2-AR) or heterologous (via ac-AR or P-AR) regulation of a2-AR.

Using these cells we have recently determined that stimulation of PKC activity with

phorbol 12-myristate 13-acetate (PMA) elicits a significant decrease in the steady state











levels of a2A-AR mRNA (Chap. 3). Similarly, treatment of cultured astroglia with

forskolin (FSK), which increases intracellular cyclic AMP, also causes a decrease in a2A-

AR mRNA levels (Chap. 3). Both effects are due to reduced transcription of the a2A-AR

gene. These data suggest that endogenous ligands which stimulate PKC activity or

increase intracellular cyclic AMP levels will modulate levels of a2A-AR mRNA in cultured

astroglia. Thus, in the present study we have investigated the effects of Epi, which

stimulates a,-, a2- and P-AR on regulation of a2A-AR mRNA levels in cultured astroglia.

We have demonstrated that Epi treatment elicits a reduction in a2A-AR mRNA, similar to

the effects of PMA and FSK. This effect is mediated by combined activation of a,-AR

and P-AR, but not a2-AR. In addition, we showed that angiotensin II (Ang II) treatment

of cultured astroglia caused a decrease in levels of a2A-AR mRNA. This effect was

mediated via Ang II type 1 (AT,) receptor stimulation, which has been shown to result in

PKC activation in cultured astroglia (Sumners et al., 1994). Overall, these data suggest

that levels of a2A-AR mRNA in astroglia undergo heterologous regulation via AT,, a,-AR

and P-AR, and that the intracellular pathways involved include activation of PKC and

increases in cyclic AMP.


Results

Effects of Epinephrine on a2A-AR mRNA Levels


Epinephrine (Epi) can act via a,-AR, a2-AR, or p-AR located on the plasma

membrane of target cells. It can act via P-AR to increase cyclic AMP accumulation or via

a,-AR to increase PKC activity (Bylund et al., 1994). Considering that increases in









63

cellular cyclic AMP accumulation or PKC activation led to decreased steady state levels of

a2A-AR mRNA via decreases in transcription of the a2A-AR gene (Chap. 3), we have

tested the effects of Epi on a2A-AR mRNA levels. Treatment of cultured astroglia with

Epi (100 pM) resulted in a time-dependent decrease in steady state levels of a2A-AR

mRNA (Fig. 4-1). The reduction of a,A-AR mRNA levels was significantly different from

control levels within 2 h of Epi treatment and reached 10% of control within 4 h. Levels

of a2A-AR mRNA remained about 10% of control levels through 24 h and then gradually

increased toward 50% of control values at 48 and 72 h. Levels of a2A-AR mRNA were

not significantly different between cultures treated with control vehicle and untreated

cultures (data not shown). The Epi-induced decreases in a2A-AR mRNA levels were

similar in magnitude and timing to the decreases produced by FSK or PMA treatment

(Figs. 3-2, 3-5). In order to determine whether the effect of Epi was due to prolonged

exposure of cultured astroglia to this catecholamine, we conducted a set of pulse-chase

experiments. Cultured astroglia were treated with Epi (100 pM) for 10 min, at which time

the media was removed and replaced with untreated (Epi-free) conditioned media. The

resulting decreases in levels of a2A-AR mRNA were similar in magnitude and timing of

onset as those seen with prolonged Epi treatment (Fig. 4-2). Recovery of a,2-AR mRNA

levels, however occurred more rapidly, reaching control levels 72 h after Epi treatment.

The Epi-induced decrease in aZ,-AR mRNA levels was also concentration-dependent,

coinciding with increases in cellular cyclic AMP (Fig. 4-3), a second messenger associated

with P-AR stimulation. Levels of a2A-AR mRNA were significantly lower than control

with 10 nM Epi treatment, while 100 nM Epi elicited a maximal effect and reduced the









64

levels to about 10% of control values. The adrenergic receptor agonist NE elicited similar

concentration-dependent decreases in levels of a2A-AR mRNA (Fig. 4-3B).


Effect of Epi on Astroglial Morphology


Treatment of cultured astroglia with FSK or PMA resulted in stellation of the cells

(Fig. 3-9). This morphological change was thought to contribute to the decreases in

steady-state levels of a2A-AR mRNA. Similar decreases in the levels of a2A-AR mRNA

were seen with Epi treatment, but without the accompanying stellation (Fig. 4-4).

Similarly, Ang II treatment caused decreases in the levels of az,-AR mRNA (Fig. 4-12),

but did not change the morphology of the cultured astroglia (not shown). Once again,

these data suggest that the observed changes in levels of a2A-AR mRNA are not due to

cell stellation.


Effect of Pertussis Toxin (PTX) on the Epi-Induced Decrease of a2A-AR mRNA
Levels


It is well known that a2-AR inhibit adenylyl cyclase via G,. In order to determine

whether the Epi-induced decrease of levels of a2A-AR mRNA was mediated through G,-

coupled pathways we pretreated cultured astroglia with PTX (200 ng/ml, 24 h), which

inactivates both G, and GO (Milligan, 1988). Treatment of cultured astroglia with PTX did

not reverse the Epi-induced decrease in levels of a2A-AR mRNA (Fig. 4-5), suggesting

that Epi did not act solely via a G,-coupled receptor. PTX treatment alone did not alter

levels of a2A-AR mRNA.












Effect of Adrenergic Receptor Antagonists on the Epi-induced Decrease of a2A-AR
mRNA Levels


Because Epi can act via p-AR, a2-AR or at-AR we assessed the effects of various

adrenergic receptor antagonists on the Epi-induced decreases in a2A-AR mRNA levels.

Previous results suggested that activation of P-AR or a,-AR may lead to decreased

expression of a2A-AR mRNA (Chap. 3). In addition, treatment of cultured astroglia with

the P-AR agonist isoproterenol produced decreases in levels of a2A-AR mRNA (data not

shown). However, treatment of cultured astroglia with the P-AR antagonist propranolol

(Prop; Fig. 4-6) or the a,-AR antagonist prazosin (Praz: Fig. 4-7) alone did not inhibit the

effect of Epi, suggesting that Epi did not exert its effect via either receptor exclusively. In

addition, this effect of Epi was not inhibited by the a2-AR antagonists rauwolscine (not

shown) or yohimbine (Yoh; Fig. 4-8). Because earlier studies had implicated signal

transduction pathways associated with a,-AR and p-AR stimulation (Chap. 3), we

suspected that the Epi-induced effect was mediated via these two receptors. Therefore,

we treated cultured astroglia with a combination of Prop and Praz, and found that

together they completely suppressed the Epi-induced decrease in levels of a2A-AR mRNA

(Fig. 4-9). In a further set of experiments we treated cultured astroglia with Epi plus the

combined a,-AR and 1-AR antagonist labetalol. Labetalol also reversed the Epi-induced

decrease in levels of a2A-AR mRNA (Fig. 4-10) providing additional evidence that Epi

exerted its effect via both a,-AR and P-AR. To further delineate the involvement of the

signaling pathways associated with a,-AR and p-AR in the Epi-induced effect we treated









66

cultured astroglia with the PKA inhibitor H-89 in combination with Praz. Praz and H-89

treatment reversed the Epi-induced decrease in levels of a2A-AR mRNA (Fig. 4-11),

providing further evidence for the involvement of signaling pathways associated with a,-

AR and P-AR.


Effect of Angiotensin II on a2A-AR mRNA Levels


Angiotensin II (Ang II) has been reported to have a negative regulatory effect on

a2-AR in the nucleus tractus solitarius (Fior et al., 1994), a region which expresses mRNA

for a2A-AR and a2c-AR (Nicholas et al., 1993; Scheinin et al., 1994). This effect of Ang II

was mediated via the Ang II type 1 (AT,) receptor, a receptor which is coupled to PI

hydrolysis and the subsequent activation of PKC (Sumners et al., 1996). Ang II can also

act via AT, receptors in cultured astroglia to stimulate PKC activity (Sumners et al.,

1994). Because we had previously shown that activation of PKC decreased levels of a,-

AR mRNA in cultured astroglia (Chap. 3), we investigated the role of Ang II in regulating

levels of a2A-AR mRNA.

Treatment of cultured astroglia with Ang II for 4 h decreased steady state levels of

a2A-AR mRNA (Fig. 4-12). This effect was blocked by the AT, receptor antagonist

losartan and not by the Ang II type 2 receptor antagonist PD 123319 (Fig. 4-13). These

results indicated that Ang II exerted its effect via the AT, receptor, suggesting

involvement of PKC in the decrease of a2A-AR mRNA levels.











Discussion


The goal of this study was to investigate the regulation of steady-state levels of

a2A-AR mRNA by ligands which act via cell-surface receptors to increase cellular cyclic

AMP accumulation or PKC activation. These investigations were prompted by earlier

studies which showed that increases in cellular cyclic AMP accumulation or PKC

activation decreased steady-state levels ofa2A-AR mRNA (Chap. 3). The agents we

tested were Epi and Ang II. Epi can act via ac-AR to stimulate PI hydrolysis and

subsequently, PKC activation, or via a2-AR and P-AR to decrease or increase cAMP

accumulation, respectively (Bylund et al., 1994). Ang II can act via AT, receptors to

stimulate PI hydrolysis and subsequently, PKC activation (Sumners et al., 1996)

In the present study, we have shown that Epi and Ang II treatment of cultured

astroglia elicits a decrease in the levels of a2A-AR mRNA similar to the decrease caused by

FSK or PMA treatment. The Epi-induced decrease was mediated by activation of P-AR

and a,-AR, while Ang II treatment decreases levels of a2A-AR mRNA via AT, receptor

stimulation.

Experiments in HT29 cells and cultured astroglia indicate that steady state levels of

a2A-AR mRNA are regulated in a heterologous fashion at the level of gene transcription.

Cyclic AMP causes an increase in transcription of the a2A-AR gene in HT29 cells (Sakaue

and Hoffman, 1991) and a decrease in cultured astroglia (Chap. 3). In addition, PKC

activation decreases transcription in cultured astroglia (Chap. 3) but is without effect in

HT29 cells (Sakaue and Hoffman, 1991). Changes in degradation rate of the a2^-AR











transcript do not appear to play a major role in steady state levels of a2A-AR mRNA in

either cell type. The opposite effect of cyclic AMP on the regulation of a,,-AR mRNA

observed in cultured astroglia compared to transformed cells may represent cell-, tissue-,

or specifies-specific regulation of this gene.

The regulatory effects of cAMP and PKC in cultured astroglia were also implied

after treatment with Epi. Concentration-dependent increases in cellular cAMP

accompanied Epi treatment of cultured astroglia, suggesting stimulation of P-AR.

However, the Epi-induced decrease in levels of a2A-AR mRNA could not be blocked with

the P-AR antagonist propranolol alone. Similarly, aI-AR and a2A-AR antagonists alone

were not able to block the Epi-induced decrease in levels ~2A-AR mRNA. Further

evidence that a2A-AR were not involved was provided by the inability of PTX to suppress

the Epi-induced effect.

We also treated cultured astroglia with a combination of Praz and Prop because

activation of signal transduction pathways associated with a,-AR and p-AR produce

decreases in levels of mRNA similar to those seen with Epi treatment. Complete reversal

of the Epi-induced effect was seen with combined Praz and Prop treatment. Similarly, the

action of Epi was inhibited by treatment of cultured astroglia with other agents which

specifically antagonize a,-AR and P-AR or their intracellular pathways. Combined with

the results of our previous study showing that cAMP and PKC decrease levels of a2A-AR

mRNA (Chap. 3), the present results suggest that Epi can lead to decreased transcription

on the a2A-AR gene via activation of a,-AR and p-AR and their intracellular signaling









69

pathways. These results also suggest that levels of a2A-AR mRNA are not regulated in a

homologous fashion in cultured astroglia by stimulation of a2A-AR.











4.0 kb
RG20


0.7 kb .
Cu/Zn SOD


+ +


+ + + + +


Time (h) 2


B

0
U




0
E


a


4 8 12 24 48 72


* *


4 8 12 24 48 72


Time (h)

Figure 4-1. Effect of Epi on steady-state levels of a2A-AR mRNA in cultured
astroglia as a function of treatment time. Cultured astroglia were incubated with
control solution (100 pM ascorbate in PBS) or Epi (100 pM) for the indicated time
periods at 37C, followed by analysis of a2A-AR mRNA levels as detailed in Chapter 2.
(A) Representative Northern blot showing 4.0 kb a2A-AR (RG20) mRNA and 0.7 kb
Cu/Zn SOD mRNA in each treatment situation. (-), control; (+), Epi-treated for the
indicated time in h. (B) Quantification of a2A-AR mRNA data normalized against Cu/Zn
SOD mRNA. Data are means SE from 4 independent experiments and are presented as
% of control levels (100%). Control data are plotted on the y-axis. *significantly
different (p < 0.05) from control.


.- I I -4if









A /l


4.0 kb
RG20



Cu/Zn SOD
UT + + + + + +
Time (h) 2 4 8 12 24 48 72


S120

100(




60

40
S\ *
4 20

S 0
4 8 12 24 48 72
Time (h)

Figure 4-2. Pulse-Chase effect of Epi on steady-state levels of a2A-AR mRNA in
cultured astroglia as a function of treatment time. Cultured astroglia were incubated
with control solution or Epi (100 pM) for 10 min at 37"C. Following this the media were
removed and replaced with conditioned (Epi-free) media for the indicated time periods,
followed by analysis of a2A-AR mRNA levels as detailed in Materials and Methods. (A)
Representative Northern blot showing 4.0 kb c2A-AR (RG20) mRNA and 0.7 kb Cu/Zn
SOD mRNA in each treatment situation. (-), control; (+), Epi-treated for the indicated
time in h. Cultured astroglia which did not have the media removed and replaced are
labeled untouched (UT). (B) Quantification ofa2A-AR mRNA data normalized against
Cu/Zn SOD mRNA. Data are means SE from 4 independent experiments and are
presented as % of control levels (100%). Control data are plotted on the y-axis.
*significantly different (p < 0.05) from control.











4.0 kb
RG20

0.7 kb
Cu/Zn SOD


B 0 120




80
o
o 80

60

40

A 20

0
*


9 8 7 6 5 4

[Epil -log M


120

100 '
80
5-
80

60 0

40



0 "-
0


Figure 4-3. Effect of Epi or NE on steady-state levels of a2A-AR mRNA in cultured
astroglia as a function of concentration. Cultured astroglia were incubated with control
solution or the indicated concentrations of Epi (@) or NE (*) for 4 h at 370C, followed
by analysis of a2A-AR mRNA levels as detailed in Materials and Methods. (A)
Representative Northern blot showing 4.0 kb a2A-AR (RG20) mRNA and 0.7 kb Cu/Zn
SOD mRNA in each treatment situation with Epi. (B) Quantification of a2-AR mRNA
data normalized against Cu/Zn SOD mRNA. Also included the levels of cellular cyclic
AMP measured as a function of Epi concentration (0). Data are means SE from 4
independent experiments in each case (1 experiment for NE) and are presented as % of
control levels (100%) for mRNA data and as pmol/mg protein for cyclic AMP data.
Control data are plotted on the y-axis. *significantly different (p < 0.05) from control.













A











B
























Figure 4-4. Effect of Epi on morphology of cultured astroglia. Cultured astroglia
were incubated with control solution or Epi (100 pM) for 4 or 24 hours at 370C,
followed by preparation of the cells for light microscopy as detailed in Chapter 2. (A)
Micrograph of cultured astroglia after 4 h treatment with control solution. (B)
Micrograph of cultured astroglia after 4 h treatment with Epi. (C) Micrograph of
cultured astroglia after 24 h treatment with Epi. All micrographs are shown enlarged
700X.






















120 -





o

60



E 40


20 -

0
Control Epi PTX Epi + PTX







Figure 4-5. Effect of pertussis toxin on Epi-induced decrease of a2A-AR mRNA
levels. Cultured astroglia were treated with control solution (water) or with 200 ng/ml
PTX for 24 h at 37"C. Cell were then treated with control solution (100 pM ascorbate in
PBS) or Epi (10 nM) for 4 hours Analysis of a2A-AR mRNA levels was then completed
as detailed in Chapter 2. Shown is quantification of a,-AR mRNA data normalized
against Cu/Zn SOD mRNA Data are from 1 experiment and are presented as % of
control levels (100%).












4.0 kb
RG20

0.7 kb
Cu/Zn SOD
Prop (-log M)
Epi (10 nM)


100


80


60


40


20 H


Prop (-log M)

Epi (10 nM)


5 8 7 6 5
+ + + + +


8 7 6 5


+ + + + +


Figure 4-6. Effect of the P-AR antagonist propranolol on Epi-induced decrease of
atA-AR mRNA levels. Cultured astroglia were treated for 15 min with the indicated
concentrations of propranolol (Prop) followed by 10 nM Epi treatment for 4 hours.
Analysis of a2A-AR mRNA levels was then completed as detailed in Chapter 2. (A)
Representative Northern blot showing 4.0 kb a2A-AR (RG20) mRNA and 0.7 kb Cu/Zn
SOD mRNA in each treatment situation. (B) Quantification of a2A-AR mRNA data
normalized against Cu/Zn SOD mRNA. Data are means SE from 4 independent
experiments and are presented as % of control levels (100%). *significantly different (p <
0.05) from control.











4.0 kb >
RG20

0.7 kb
Cu/Zn SOD
Praz (-log M)
Epi (10 nM)


140 -

120 -

100

80

60

40

20


Praz (-log M)

Epi (10 nM)


5 7 6 5
- + + + +


8 7 6 5

+ + + + +


Figure 4-7. Effect of the a,-AR antagonist prazosin on Epi-induced decrease of
a2A-AR mRNA levels. Cultured astroglia were treated for 15 min with the indicated
concentrations of prazosin (Praz) followed by 10 nM Epi treatment for 4 hours. Analysis
of a2-AR mRNA levels was then completed as detailed in Chapter 2. (A) Representative
Northern blot showing 4.0 kb aA-AR (RG20) mRNA and 0.7 kb Cu/Zn SOD mRNA in
each treatment situation. (B) Quantification of a2A-AR mRNA data normalized against
Cu/Zn SOD mRNA Data are means SE from 4 independent experiments and are
presented as % of control levels (100%). *significantly different (p < 0.05) from control.










4.0 kb >
RG20


0.7 kb >
Cu/Zn SOD
Yoh (-log M)
Epi (10 nM)


* *
Tr -


8 7 6 5


+ + + + +


Figure 4-8. Effect of the a2-AR antagonist yohimbine on Epi-induced decrease of
aA-AR mRNA levels. Cultured astroglia were treated for 15 min with the indicated
concentrations of yohimbine (Yoh) followed by 10 nM Epi treatment for 4 hours.
Analysis of a2A-AR mRNA levels was then completed as detailed in Chapter 2. (A)
Representative Northern blot showing 4.0 kb a2A-AR (RG20) mRNA and 0.7 kb Cu/Zn
SOD mRNA in each treatment situation. (B) Quantification of a2A-AR mRNA data
normalized against Cu/Zn SOD mRNA. Data are means SE from 4 independent
experiments and are presented as % of control levels (100%). *significantly different (p <
0.05) from control.


w W M'%


- -W- a f

5 8 7 6 5
+ + + + +


140 -

120

100 -


60 -

40

20 -


0 I
Yoh (-log M)

Epi (10 nM)


* r*m ~r* ~















0 0
U&


120 -

100

80 -

60 -


01


**


II

9^ ^


Figure 4-9. Epi-induced decrease of a2A-AR mRNA levels: Effect of combined
propranolol and prazosin treatment. Cultured astroglia were treated with control
solution (0.01% methanol in water), or with a combination of Prop and Praz (1 pM each)
for 15 min at 370C. Cells were then treated with control solution (100 pM ascorbate in
PBS) or Epi (10 nM) for 4 h This was followed by analysis of a,A-AR mRNA levels as
detailed in Materials and Methods. (A) Representative Northern blot showing 4.0 kb
a2A-AR (RG20) mRNA and 0.7 kb Cu/Zn SOD mRNA in each treatment situation. (B)
Quantification of a,-AR mRNA data normalized against Cu/Zn SOD mRNA. Data are
means SE from 3 independent experiments and are presented as % of control levels
(100%). *significantly different (p < 0.05) from control.


4.0 kb
RG20


0.7kb
Cu/Zn SOD











4.0 kb >
RG20

0.7 kb >
Cu/Zn SOD


o +


120

100


0 -


Figure 4-10. Effect of the P-AR and a,-AR antagonist labetalol on Epi-induced
decrease of a2A-AR mRNA levels. Cultured astroglia were treated with control solution
or labetalol (1 pM) for 15 min at 37C. Cells were then treated with control solution (100
pM ascorbate in PBS) or Epi (10 nM) for 4 hours. Analysis of a2,-AR mRNA levels was
then completed as detailed in Chapter 2. (A) Representative Northern blot showing 4.0 kb
a2A-AR (RG20) mRNA and 0.7 kb Cu/Zn SOD mRNA in each treatment situation. (B)
Quantification of a2A-AR mRNA data normalized against Cu/Zn SOD mRNA. Data are
means SE from 4 independent experiments and are presented as % of control levels
(100%). *significantly different (p < 0.05) from control.

















100


80 -


60


40 -




20 -

0

% xe








Figure 4-11. Effect of the protein kinase A inhibitor H-89 and prazosin on Epi-
induced decrease of aZA-AR mRNA levels. Cultured astroglia were treated with control
solution (0.01% methanol) or with H-89 (250 nM) and prazosin (1 pM) for 15 min at
37C. Cells were then treated with control solution (100 pM ascorbate in PBS) or with
Epi (10 nM) for 4 hours. Analysis of a2A-AR mRNA levels was then completed as
detailed in Chapter 2. Shown is quantification of a2A-AR mRNA data normalized against
Cu/Zn SOD mRNA. Data are from 1 experiment and are presented as % of control levels
(100%).











4.0kb .
RG20

0.7 kb >
Cu/Zn SOD




120

I-V
1004
80
o
'5 80

60

40

o20
'.


41400 ag mw -
::MW OW aWTo



- u -I


10 9 8 7 6 5


[Ang II] -log M



Figure 4-12. Effect of Ang I on steady-state levels of aSA-AR mRNA in cultured
astroglia as a function of concentration. Cultured astroglia were incubated with control
solution (water) or the indicated concentrations of Ang II for 4 hours, followed by analysis
of a2-AR mRNA levels as detailed in Chapter 2. (A) Representative Northern blot
showing 4.0 kb a2A-AR (RG20) mRNA and 0.7 kb Cu/Zn SOD mRNA in each treatment
situation. (B) Quantification ofa2A-AR mRNA data normalized against Cu/Zn SOD
mRNA. Data are means SE from 4 independent experiments and are presented as % of
control levels (100%). Control data are plotted on the y-axis. *significantly different (p <
0.05) from control.










4.0 kb )
RG20

0.7kb
Cu/Zn SOD


120 1

100 -


5 .


T +



- T


///I
xO


Figure 4-13. Effect of Losartan and PD 123319 on Ang H-induced decrease of
a2A-AR mRNA levels. Cultured astroglia were treated with control solution (water) or
with Losartan or PD 123319 (1 pM each) for 15 min at 37C. Cells were then treated
with control solution (water) or Ang II (100 nM) for 4 hours. Analysis of a2A-AR mRNA
levels was completed as detailed in Chapter 2. (A) Representative Northern blot showing
4.0 kb azu-AR (RG20) mRNA and 0.7 kb Cu/Zn SOD mRNA in each treatment situation
(B) Quantification of a2A-AR mRNA data normalized against Cu/Zn SOD mRNA. Data
are means SE from 4 independent experiments and are presented as % of control levels
(100%). *significantly different (p < 0.05) from control.


Coc;~P


losti~~l














CHAPTER 5
REGULATION OF a2A-ADRENERGIC RECEPTOR NUMBER BY EPINEPHRINE


Introduction


Studies have shown that a variety of cellular mechanisms can lead to changes in the

number or sensitivity of receptors on the cell surface. These changes can involve

uncoupling of the receptor from its G protein, sequestration of the receptor away from the

cell surface, degradation of existing receptors, and reduced synthesis of new receptors.

The most rapid way of uncoupling a G protein-coupled receptor from its G protein

is via agonist-dependent phosphorylation of the receptor by G protein coupled receptor

kinases. The most studied members of this kinase family are the P-adrenergic receptor

kinases (PARK), which have been shown to phosphorylate P2-AR (Benovic et al., 1986).

PARK phosphorylates only the agonist occupied 32-AR (Benovic et al., 1986) leading to

enhanced affinity of the receptor for P-arrestin (Lohse et al., 1990). Binding of P-arrestin

inhibits the receptor from interacting with its stimulatory G protein, effectively uncoupling

the receptor from its signal transduction pathway (Lohse et al., 1992). This type of

homologous regulation by PARK has also been described for G,-coupled receptors such as

the D,-dopamine receptor (Bates et al., 1991), the m2-muscarinic receptor (Richardson

and Hosey, 1992), the 5-HTi1-serotonin receptor (Pleus and Bylund, 1992), and the a2A-

AR (Liggett et al., 1992).











In addition to PARK-mediated homologous desensitization, G protein-coupled

receptors can be subject to heterologous desensitization via effector kinases such as PKA

or PKC. Phosphorylation of the P,-AR by PKA (Benovic et al., 1985, 1987; Bouvier et

al., 1987, 1989) or PKC (Pitcher et al., 1992; Bouvier et al., 1991) are not dependent on

agonist occupancy of the receptor. Activation of these kinases by any pathway can lead to

12-AR phosphorylation (Benovic et al., 1985). These mechanisms appear to occur more

slowly (minutes to hours) than homologous desensitization (seconds to minutes). PKA

and PKC preferentially phosphorylate receptor sites in the third intracellular loop thought

to be responsible for interaction with the G protein (Pitcher et al., 1992; Okamoto et al.,

1991; Bouvier et al., 1987; Johnson et al., 1990). Therefore, these kinases appear to

desensitize the receptor via uncoupling it from its G protein.

Also occurring over a time-frame of minutes to hours is the agonist-induced

translocation of P3-AR to intracellular compartment (Perkins et al, 1991). The molecular

mechanisms for this sequestration are largely unknown, but appear to involve site in the

cytoplasmic tail of the P2-AR (Hausdorffet al., 1990). Such sequestration also occurs for

the a2A-AR (Liggett et al., 1992; Eason and Liggett, 1992).

Long-term agonist exposure leads not only to sequestration of G protein-coupled

receptors, but also to downregulation of receptor number. Similarly, activation of PKA or

PKC can lead to changes in the amount of a receptor at the cell surface. These processes

are slow (hours to days) and can involve increased receptor degradation and decreased

receptor synthesis. Long-term agonist exposure can elicit homologous downregulation of

P2-AR via increased degradation of the receptor (Bouvier et al., 1985). Similarly, a,2-AR









85

undergo agonist-induced downregulation (Eason and Liggett, 1992). Degradation of P2-

AR can also be enhanced by PKA or PKC-mediated phosphorylation (Bouvier et al.,

1985; Pitcher et al., 1992; Johnson et al., 1990).

Downregulation of receptor number can also occur as a result of decreased steady-

state levels of receptor mRNA and subsequent decreased de novo synthesis of receptors.

Decreased gene transcription and increased mRNA degradation can contribute to

decreases in levels of mRNA. For example, reduction of P2-AR mRNA leads to

downregulation of the number of P2-AR (Hadcock and Malbon, 1988; Bouvier et al.,

1989). This reduction in mRNA levels is mediated by decreased stability of the transcript

and not by decreased transcription rate (Hadcock et al., 1989). These decreases in levels

of mRNA are the result of increased levels of cellular cAMP (Hadcock and Malbon, 1988;

Hadcock et al., 1989; Bouvier et al., 1989), implying activation of PKA because this

kinase mediates the intracellular effects of cAMP (Riabolwol et al., 1988).

We have shown similar heterologous downregulation of steady-state levels of a,-

AR mRNA in cultured astroglia. Accumulation of cAMP and activation of PKC lead to

decreased levels of a2A-AR mRNA via decreased transcription and not via increased

degradation of the transcript (Chap. 3). The effect on mRNA levels could be mimicked by

Epi treatment of the cultured astroglia. Epi elicited these changes via ac-AR and P-AR

providing further evidence for the heterologous nature of this regulation. Therefore, we

investigated whether treatment of cultured astroglia with Epi, or with agents that cause

activation of PKC or accumulation of cAMP lead to decreases in the number of a2A-AR.











We found that the a2A-AR number is subject to heterologous regulation, and that the

effect of Epi on a,A-AR number may have a homologous component.


Results

Effects of FSK, PMA and Epi on a,-AR Specific Binding in Cultured Astroglia


In the present study we investigated whether Epi and agents that increased PKC

activity or intracellular cyclic AMP levels cause a decrease in a2-AR specific binding,

consistent with the observed effects on aZA-AR mRNA. Binding ofa2-AR was analyzed

using ['H]-MK 912 as the radioactive ligand. Binding of [3H]-MK 912 to membranes

prepared from untreated cultured astroglia was linear with respect to protein

concentrations from 2-25 pg. At 25 C the binding of ['H]-MK 912 reached equilibrium

within 30 min and remained stable for at least 2 h. Thus, in subsequent experiments, 10

pg of protein were used during an incubation time of 1 h.

Incubation of cultured astroglia with FSK (10 pM; 2-72 h) or PMA (500 nM; 2-72

h), agents that increase intracellular cyclic AMP levels or increase PKC activity

respectively, elicited a time-dependent decrease in the level of specific ['H]-MK 912

binding (Fig. 5-1). Similar changes in binding were obtained when using either 0.2 nM

[3H]-MK 912 (Fig. 5-1) or 5 nM [3H]-MK 912 (data not shown). For example, binding of

['H]-MK 912 was significantly reduced (p < 0.05 compared to control) 8 h after PMA

treatment and 16 h after FSK treatment, and reached approximately 60% of control levels

24 h after FSK or PMA treatment. At 48 and 72 h after PMA treatment [3H]-MK 912

binding had returned to approximately 80% of control levels, while membranes from FSK-











treated astroglia exhibited a further decrease in specific binding. In order to determine

saturation binding parameters, membranes prepared from cultured astroglia treated with

control solution (0.5% DMSO), FSK (10 pM) or PMA (500 nM) for 24 h were incubated

with increasing concentrations of[3H]-MK 912 (0.1-10 nM) at 25*C for I h. Figure 5-2A

depicts a representative experiment showing that specific binding of['H]-MK 912

increased with increasing concentration of radiolabeled ligand and was saturable in each

case. The data also indicate that the level of binding was less in FSK- or PMA-treated

cells compared with controls. Scatchard analysis of this data (Fig. 5-2B) revealed that the

B, values for [3H]-MK 912 binding were 2.92, 1.65, and 1.23 pmol/mg protein for the

control, FSK, and PMA-treated groups, respectively. The KD-values did not change

among the control, FSK, or PMA-treated groups (0.55, 0.58 and 0.53 nM, respectively).

This experiment was repeated 4 times with similar results for the KD and B, values

(Table 5-1), indicating that FSK and PMA pretreatment significantly (p<0.05 compared to

control) decreased the number of a2A-AR without changing their affinity.

Similar saturation binding assays were conducted with membranes from cultured

astroglia treated with either control solution (100 pM ascorbate in PBS) or Epi (10 pM)

for 24 h. Figure 5-3A shows a representative saturation experiment which demonstrates

that [3H]-MK 912 specific binding increased with increasing concentrations ofradiolabeled

ligand and was saturable in both control and Epi-treated cells. Scatchard analysis of the

saturation binding data revealed that Epi treatment produced nearly an 85% decrease in

the number of [3H]-MK 912 binding sites (compared to control) without effecting the

affinity of those sites (Fig. 5-3B; Table 5-1). The data presented in Table 5-1 also indicate









88

that Epi-treatment produced larger decreases (-85%) in B, values than FSK (-35%) or

PMA (-55%) treatment, suggesting that activation of both pathways may be necessary to

maximally reduce [3H]-MK 912 binding.

In order to determine if activation of both pathways reduced ['H]-MK 912 binding

to the same extent as Epi, saturation binding assays were conducted with membranes from

cultured astroglia treated with either control solution (1% DMSO) or a combination of

FSK (10 gM) and PMA (500 nM). The results of a representative saturation experiment

are shown in Figure 5-4A, and demonstrate that specific binding of [H]-MK 912

increased with increasing concentration of radiolabeled ligand and was saturable in both

cases. The data also indicate that the level of binding was less in the cells treated with

FSK and PMA than in the controls. Scatchard analysis of the saturation binding data (Fig.

5-4B) revealed that the B, values for ['H]-MK 912 binding were 2.70 and 1.44 pmol/mg

protein for the control and FSK/PMA-treated groups respectively. The B_, value for the

FSK/PMA-treated cells was similar to that seen in cells treated with either drug

individually. The KD values did not vary between the control and FSK/PMA-treated

groups (0.81 and 0.80 nM, respectively). This experiment was repeated twice with similar

results for the KD and B,, values (Table 5-1). These results suggest that increases in

intracellular cyclic AMP and PKC activation are not sufficient to account for all of the

decreases in ['H]-MK 912 binding observed with Epi treatment.









89

TABLE 5-1. Summary of the K, and B.x values obtained for I[HI-MK 912 specific
binding to cultured astroglia following FSK, PMA, and Epi
treatments

KD (nM) Bm, (fmol/mg protein)
Control 0.69 0.11 2.73 0.20
Forskolin (10 pM; 24 h) 0.86 0.23 1.75 0.24*
PMA (500 nM; 24 h) 0.57 0.04 1.17 0.11*

Control 0.51 0.01 2.96 0.27
Epinephrine (10 pM; 24 h) 0.42 0.10 0.50 0.07*

[3H]-MK 912 specific binding was determined in control-, FSK-, PMA-, and Epi- treated
cultured astroglia as detailed in Materials and Methods. Data are mean SEM values of 4
experiments. Significantly different from controls (p < 0.05)


Effect of Downregulating aZA-AR on cAMP Accumulation


The a2-AR are coupled to multiple signaling pathways (Enkvist et al., 1996). For

example, activation of a2-AR by NE or Epi can lead to activation of potassium channels

(Aghajanian and VanDerMaelen, 1982), inhibition of calcium channels (Holz et al., 1986),

mobilization ofintracellular calcium stores (Michel et al., 1989), and stimulation of PI

hydrolysis (Wilson and Minneman, 1991; Cotecchia et al., 1990). In most cells, activation

of a2-AR has been shown to cause a reduction in cellular cAMP levels as a result of

inhibition of adenylyl cyclase (Limbird, 1988). Considering this, we tested whether the

aA-AR-mediated decrease in cAMP levels in cultured astroglia was modified in cells

where the levels of a2A-AR had been downregulated. In the first series of experiment, we

tested whether activation of a2A-AR caused a decrease in cAMP levels in cultured

astroglia. Incubation of cultured astroglia with the a2-AR agonist clonidine (1 nM 10









90

pM; 20 min) resulted in no significant changes in basal levels of cellular cAMP (data not

shown). By contrast, in cultures that were treated with FSK (10 pM) for 20 min to raise

intracellular cAMP levels, clonidine (1 nM 10 pM) caused a decrease in those levels of

cAMP (Fig. 5-5). In the next series of experiments, we tested whether pretreatment of

cultured astroglia with 10 pM Epi for 24 h, to decrease the number of a2A-AR present,

would reduce the inhibitory effects of clonidine on cellular cAMP levels. The data

presented in Figure 5-5 also show that incubation of Epi-pretreated cells with clonidine (1

nM 10 MM; 20 min) caused a decrease in FSK-stimulated cAMP levels. However, this

decrease was not significantly different than the decrease obtained with clonidine in cells

not pretreated with Epi (Fig. 5-5). These data suggest that the reduction in a2A-AR

produced by Epi pretreatment does not include an,-AR that are coupled to a fall in cellular

cAMP levels.


Discussion


Downregulation of G-protein coupled receptors can occur as a result of decreased

receptor synthesis or increased receptor degradation (Bouvier et al., 1989; Collins et al.,

1990; Lohse, 1993). These processes can be initiated by agonist binding to the receptor

(homologous regulation) or by activation of intracellular signaling pathways (heterologous

regulation). Previous studies (see chaps. 3 and 4) in cultured astroglia showed that levels

of a2A-AR mRNA were downregulated in a heterologous fashion via activation of ca-AR

or P-AR, or via activation of their intracellular signaling pathways with PMA or FSK.

The decreased expression of a2A-AR mRNA in response to PMA, FSK or Epi treatment











suggested that these agents may lead to decreased de novo synthesis of aZA-AR. The

results presented here demonstrate that FSK, PMA and Epi all elicit decreases in the

number of a2A-AR, but those decreases differ depending on the agent. In addition,

downregulation of a2A-AR number with Epi led to changes in the ability of an a2A-AR

agonist to inhibit cAMP accumulation.

Treatment of cultured astroglia with FSK, PMA or Epi induced similar decreases

in the levels ofa2A-AR mRNA. Changes in cell surface receptor expression can be the

combined result of changes in de novo synthesis of receptors and changes in degradation

of existing receptors. Given that Epi, FSK and PMA produce similar heterologous

decreases in levels of a2A-AR mRNA, and that steady state levels of mRNA influence de

novo synthesis rates, it is possible that all three agents decrease de novo synthesis to an

equal extent. Minimal levels of a2A-AR mRNA may reduce de novo synthesis of receptors

to minimal levels allowing for disappearance of existing receptors in accordance with

receptor half-life. We found that 24 h FSK or PMA treatment reduced a2A-AR number

approximately 40-50% suggesting that the half-life of a2A-AR in cultured astroglia may be

similar to that observed (26 h) in HT29 cells (Paris et al., 1987). Epi treatment, however,

reduced a2^-AR number to a greater extent that FSK or PMA treatment. We initially

thought that the difference was due to activation of both the cyclic AMP and PKC

signaling pathways by Epi. However, activation of both of these pathways by combined

FSK/PMA treatment did not decrease a,A-AR number more than either treatment alone

(Fig. 5-4, Table 5-1). This suggests that Epi, FSK and PMA may mediate different

changes in the degradation of existing a2A-AR.









92

Phosphorylation of the 3,-AR by cyclic AMP dependent protein kinase (PKA) and

PKC has been shown to play a role in 2-AR downregulation (Bouvier et al., 1987, 1989).

However, consensus sequences for PKA or PKC-mediated phosphorylation have not been

reported for either the human (Eason and Liggett, 1992) or the rat a2A-AR (Lanier et al.,

1991). PARK plays a role in homologous desensitization of P,-AR (Benovic et al., 1986)

and the human a2A-AR (Benovic et al., 1987). The amino acid sequence LEESSSS in the

third cytoplasmic loop of the human a2A-AR has been identified as the substrate for PARK

phosphorylation (Oronato et al., 1991). Removal of this site eliminates agonist-dependent

phosphorylation and desensitization of the a2A-AR (Liggett et al., 1992). The rat a2A-AR

contains the same sequence in the third cytoplasmic loop (Lanier et al., 1991), suggesting

that the rat a2A-AR is subject to homologous regulation. Such homologous regulation

could explain the differences we observed in a2A-AR number in the Epi-treated cells

compared to the FSK- or PMA-treated cells. Epi treatment of cultured astroglia could

result in combined heterologous decreases in a2A-AR synthesis via a,-AR and P-AR and

their signaling pathways, and homologous increases in a2A-AR phosphorylation,

sequestration, and degradation. Additional studies are needed to further delineate the

effect of these two modes of regulation on a2A-AR expression.

Because Epi induced large decreases in a2A-AR number, we hypothesized that a2A-

AR responsiveness would be decreased after treatment of cultured astroglia with Epi. We

found, however, that the a2-AR agonist clonidine inhibited FSK-stimulated cAMP levels

to the same extent in cells pretreated with Epi and in cells which were not treated with

Epi. The reasons for this lack of change are unclear. Perhaps the downregulation of a2A-











AR in this system may not be reflected in a change in the cAMP pathway, but in some

other signal transduction pathway. A variety of alternative signaling pathways have been

associated with a2-AR stimulation, including activation of potassium channels (Aghajanian

and VanDerMaelen, 1982), inhibition of calcium channels (Holz et al., 1986), mobilization

of intracellular calcium stores (Michel et al., 1989), and stimulation of PI hydrolysis

(Wilson and Minneman, 1991; Cotecchia et al., 1990). Several of these mechanisms have

been shown to operate in cultured astroglia (Enkvist et al., 1996). Perhaps the number of

a2A-AR remaining after downregulation is adequate to elicit the observed suppression of

cAMP accumulation, suggesting that a large receptor reserve exists. In addition, the

limiting factor for a2A-AR-mediated inhibition of adenylyl cyclase may not be the receptor,

but the amount of G, (Neubig et al., 1985). In some cell systems, treatment with agents

that increase cellular cAMP leads to increased expression of G (Hadcock et al., 1990),

and therefore sensitization of the inhibitory pathway associated with a2A-AR. Another

explanation for this observation in Epi-pretreated cultured astroglia is that clonidine is

acting via so-called imidazoline receptors. However, the signal transduction mechanisms

associated with imidazoline sites are uncertain (Regunathan and Reis, 1996), but are

suggested to not involve cAMP (Regunathan et al., 1991). Therefore, it is unlikely that

these sites are involved in the increased responsiveness to clonidine observed after Epi-

pretreatment.

Our results indicate that a2A-AR expression is regulated in a heterologous manner

by Epi via P-AR and a,-AR. These results are novel in that they not only imply a negative

regulatory role of cyclic AMP (via p-AR) on a2A-AR expression, by also a negative effect




Full Text
xml version 1.0 encoding UTF-8
REPORT xmlns http:www.fcla.edudlsmddaitss xmlns:xsi http:www.w3.org2001XMLSchema-instance xsi:schemaLocation http:www.fcla.edudlsmddaitssdaitssReport.xsd
INGEST IEID EQ8FG62WZ_R7575H INGEST_TIME 2012-09-24T14:40:51Z PACKAGE AA00011813_00001
AGREEMENT_INFO ACCOUNT UF PROJECT UFDC
FILES