Citation
Catecholamine and angiotensin II interactions in primary cultures of brain cells

Material Information

Title:
Catecholamine and angiotensin II interactions in primary cultures of brain cells
Creator:
Richards-Sumners, Elaine Mary, 1957-
Publication Date:
Language:
English
Physical Description:
viii, 100 leaves : ill. ; 29 cm.

Subjects

Subjects / Keywords:
Brain ( jstor )
Catecholamines ( jstor )
Cultured cells ( jstor )
Incubation ( jstor )
Insulin ( jstor )
Neuroglia ( jstor )
Neurons ( jstor )
Norepinephrine ( jstor )
Rats ( jstor )
Receptors ( jstor )
Angiotensin II -- physiology ( mesh )
Brain Chemistry ( mesh )
Catecholamines -- physiology ( mesh )
Cells, Cultured ( mesh )
Dissertations, Academic -- Physiology -- UF ( mesh )
Physiology thesis Ph.D ( mesh )
Genre:
bibliography ( marcgt )
non-fiction ( marcgt )

Notes

Thesis:
Thesis (Ph.D.)--University of Florida, 1988.
Bibliography:
Bibliography: leaves 90-99.
General Note:
Typescript.
General Note:
Vita.
Statement of Responsibility:
by Elaine Mary Richards-Sumners.

Record Information

Source Institution:
University of Florida
Holding Location:
University of Florida
Rights Management:
The University of Florida George A. Smathers Libraries respect the intellectual property rights of others and do not claim any copyright interest in this item. This item may be protected by copyright but is made available here under a claim of fair use (17 U.S.C. §107) for non-profit research and educational purposes. Users of this work have responsibility for determining copyright status prior to reusing, publishing or reproducing this item for purposes other than what is allowed by fair use or other copyright exemptions. Any reuse of this item in excess of fair use or other copyright exemptions requires permission of the copyright holder. The Smathers Libraries would like to learn more about this item and invite individuals or organizations to contact the RDS coordinator (ufdissertations@uflib.ufl.edu) with any additional information they can provide.
Resource Identifier:
023098377 ( ALEPH )
20460805 ( OCLC )

Downloads

This item has the following downloads:


Full Text










CATECHOLAMINE AND ANGIOTENSIN II
INTERACTIONS IN PRIMARY CULTURES OF
BRAIN CELLS















By
ELAINE MARY RICHARDS-SUMNERS




















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

1988














ACKNOWLEDGEMENTS

I would like to thank the Chairman of my committee, Dr M. Ian Phillips, for giving me the opportunity to pursue a graduate degree and the freedom to do the project of my choice. Ian speaks and writes very lucidly about science, and all students in his lab benefit from his tuition in this area. His enthusiastic support throughout the years I have been in his lab have been gratefully received, too.

Collectively, I would like to thank the other members of my committee, Drs. Fregly, Raizada, and Shiverick, for their hard work on my behalf. I will never forget their involvement and support.

There are some other people to whom I am very grateful for the use of their facilities and/or techniques. These are Dr. P. Klein in Pathology, Dr. G. Shaw in Neuroscience, Dr. S. Baker and Dr. T. Muther in Pharmacology.

The technical help of Ms. J. Perez, Mr. J. Hogenesch, Ms. S. Fuentes and Mr. J. Neal is also greatly appreciated. Big Kimura, who has never become cross with any of my millions of interruptions, deserves a medal!

The help with the use of word perfect given me by Kevin Fortin is gratefully acknowledged too.

The biggest thanks of all, however, go to my family for all that they have put up with in my graduate school years.








ii














TABLE OF CONTENTS

PAGE

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

KEY TO ABBREVIATIONS . . . . . . . . . . v

ABSTRACT . . . . . . . . . . . . . vii

CHAPTERS

I. INTRODUCTION: INTERACTIONS BETWEEN CATECHOLAMINES AND
ANGIOTENSIN II . . . . . . . . . 1

Colocalization of Catecholamines and Angiotensin II in
the Central Nervous System . . . . . . I
Effects of Catecholamines and Angiotensin II which are
Sim ilar . . . . . . . . . . . 5
The Effects of Angiotensin II on Catecholamines. . . . 5
The Effects of Catecholamines on Angiotensin II. . . . 8
The Cell Culture Model. . .. .......... 12
The Effects of Angiotensin II on Catecholamines in Cell
Culture . . . . . .. . . . . 13
The Effects of Catecholamines on Angiotensin II in Cell
Culture . . . . . . . . . . . 13
General Hypothesis. . . . . . . . . . 14

II. CHARACTERIZATION OF m2-ADRENOCEPTORS ON
NEURONAL AND GLIAL CELLS IN CULTURE . . 16

Introduction . . . . . . . . . . . 16
M ethods . . . . . . . . . . . . 18
R esults . . . . . . . . . . . . 21
D iscussion . . . . . . . . . . . 31

III. LIGHT MICROSCOPIC AUTORADIOGRAPHY OF [3H]YOHIMBINE BINDING SITES IN NEURONS IN CULTURE 35

Introduction . . . . . . . . . . . 35
M ethods . . . . . . . . . . . . 36
R esults . . . . . . . . . . . . 37
Discussion . . . . . . . . . . . 46

IV. ADRENERGIC MECHANISMS MEDIATING ANGIOTENSIN II
RELEASE ........... ......................... 47

Introduction . . . . . . . . . . . 47

iii








M ethods . . . . . . . . . . . . 48
R esults . . . . . . . . . . . . 51
D iscussion . . . . . . . . . . . 61

V. EFFECT OF VARIOUS ADRENERGIC DRUGS ON THE RENIN
CONCENTRATION OF BRAIN CELLS IN CULTURE . 66

Introduction . . . . . . . . . . . 66
M ethods . . . . . . . . . . . . 66
R esults . . . . . . . . . . . . 69
Discussion . . . . . . . . . . . 69

VI. REGULATION OF Oc2-ADRENERGIC RECEPTORS BY PEPTIDES 71

Introduction . . . . . . . . . . . 71
M ethods . . . . . . . . . . . . 72
R esults . . . . . . . . . . . . 74
D iscussion . . . . . . . . . . . 79

VII. GENERAL CONCLUSIONS . . . . . . . . . 82

REFERENCES ........ ... ......................... 90

BIOGRAPHICAL SKETCH . . . . . . . . . . 100


































iv















KEY TO ABBREVIATIONS


ACh . . . . . . . acetylcholine

ARC . . . . . . . cytosine arabinoside

AVP . . . . . . . arginine vasopressin

Bmax .......... ..... maximum number of binding sites

BST . . . . . . . bed nucleus of the stria terminalis

cAMP . . . . . . . cyclic adenosine monophosphate

CeA . . . . . . . central nucleus of the amygdala

COMT . . . . . . . catechol-O-methyl-transferase

CSF . . . . . . . cerebrospinal fluid

DMEM . . . . . . . Dulbecco's modified Eagle's medium

DMI . . . . . . . desmethylimipramine

DNase I . . . . . . deoxyribonuclease I

GFAP . . . . . . . glial fibrillary acidic protein

Gi . . . . . . . . inhibitory GTP binding protein

Gs . . . . . . stimulatory GTP binding protein

Gpp(NH)p . . . . . . 5'-guanylylimidodiphosphate

GTP . . . . . . . guanosine triphosphate

IC50 . . . . . . inhibitory constant50

kI . . . . . . . association constant

k-I . . . . . . . dissociation constant

KD . . . . . . . equilibrium dissociation constant

Ki . . . . . . . inhibitory dissociation constant

kobs . . . . . . . observed association constant

v










LC . . . . . . . locus ceruleus

LHA . . . . . . . lateral hypothalamic area

LS . . . . . . . lateral septum

MAO . . . . . . . monoamine oxidase

ME . . . . . . . medain eminence

MPO ... ......... median preoptic nucleus

NTS . . . . . . . nucleus tractus solitarius

OB . . . . . . . . olfactory bulb

OVLT . . . . . . organon vasculosum lamina terminalis

PBI . . . . . . . parabrachial nucleus-lateral

PBS . . . . . . . phosphate buffered saline

PDHS . . . . . . plasma derived horse serum

PF . . . . . . . parafascicular nucleus

PMSF . . . . . . phenylmethylsulfonyl fluoride

PP . . . . . . . .peripeduncular nucleus

PVH . . . . . . . paraventricular nucleus of the
hypothalamus

RIA . . . . . . . radioimmunoassay

SFO . . . . . . . .subfornical organ

st . . . . . . . .stria terminalis

Tris . . . . . . . .(Tris[hydroxymethyl]-aminomethane
hydrochloride)

VTA . . . . . . . ventral tegmental area

ZI . . . . . . . zona incerta

5HT . . . . . . . .serotonin

60HDA . . . . . . .6-hydroxydopamine




vi












Abstract of Dissertation Presented to the Graduate School of the University of Florida in Partial Fulfillment of the
Requirements of the Degree of Doctor of Philosophy CATECHOLAMINE AND ANGIOTENSIN II INTERACTIONS IN PRIMARY CULTURES OF BRAIN CELLS

By

ELAINE MARY RICHARDS-SUMNERS

DECEMBER 1988

Chairman: Dr. M.I. Phillips
Major Department: Physiology

Alpha2-adrenergic receptors were characterized in neurons and glia in cell culture. The receptors were present in both cell types, with similar affinities, but glia in culture contain ten times more receptors than neurons. The neuronal alpha2-receptors were visualized by light microscopic autoradiography. Antagonism of alpha2-adrenoceptors induced the release of the neuropeptide angiotensin II from neuronal cultures, but not from glial cultures, even though they also contain angiotensin II. The material released by alpha2-antagonism from neuronal cultures was found to migrate with synthetic angiotensin II on high pressure liquid chromatography, and thus was authentic angiotensin II. Norepinephrine also induced release of angiotensin II from neuronal cultures, which could not be blocked by alphaI-adrenergic antagonists. Beta-adrenoceptor agonists induced the release of angiotensin II. These data suggest that angiotensin II is released from two populations of cells, one responding to betaagonists and the other to alpha2-antagonists. Stimulation of release of angiotensin II by yohimbine did not result in an increase in the neuronal cell



vii








content of renin, suggesting that renin is not an important regulator of the availability of angiotensin II in brain cells in culture.

Insulin, a neuroactive peptide, regulated the number of alpha2adrenoceptors in glial cultures, but angiotensin II had no effect on alpha2adrenoceptors in neurons or glia.

A model describing these findings in relation to other known interactions of angiotensin II and catecholamines in brain cell culture is discussed.









































viii














CHAPTER I
INTRODUCTION

Interactions between Catecholamines and Anpiotensin II

Phylogenetically, there is a very old interaction between catecholamines and angiotensin II (Carroll and Opdyke, 1982). With the recent acceptance of tissue renin-angiotensin systems, separate from blood-borne angiotensin II, interactions with catecholamine systems have been discovered in various tissues. They are apparent in vascular beds, including the mesenteric arteries (Nakamura et al., 1986a, b) and vena cava (Gothert and Kollecker, 1986), cultured smooth muscle cells (Dzau, 1986), the vas deferens (Trachte, Stein and Peach, 1987) and in the brain, the area of particular interest to us. This introductory chapter will review what is known about these interactions between catecholamines and angiotensin II. It will include a discussion of the many advances in our knowledge of these interactions in the brain which have been made possible by the use of tissue culture techniques. These cultures of one-day old rat brains enriched either for neurons or glia have greatly facilitated these studies. Colocalization of Catecholamines and Anpiotensin II in the Central Nervous
System

In the following section, discrete localizations of angiotensin II are taken from Weyhenmeyer and Phillips, (1982) and Lind et al. (1985). The

catecholamine localizations are from Lindvall and Bjorkland (1983) and Moore and Bloom (1979).

The overlap between the anatomical localization of the catecholamines and angiotensin II systems is very striking (Fig 1). The catecholamine and angiotensin II systems are predominantly located in areas of the brain











2






















A
C E R E S U M C O T E X U L B U S O L F A C T O R IU M



S BUNDLE kTI TERMINALIS


SEPTU

AS

A2 FOREBRAIN VENTRAL SUNDL BUNDLE

Al A5 HYPOT ALAMUS

B















-- ME



















Figure 1. Diagram showing the high degree of overlap between the anatomical
localization of the norepinephrine system and the angiotensin II system in the rat brain. A) Diagram showing the origin and pathways of norepinephrine
containing tracts in the rat brain. Shaded areas are norepinephrine terminal
fields. From Chemical Neurobiology, ed. H. F. Bradford. B) Diagram showing
the distribution of angiotensin II cells (e) and tracts (-) in the rat brain. Redrawn from Lind, Swanson and Ganten, 1984.








3

associated with homeostatic control functions of the body, for example, the hypothalamus and brain stem. Angiotensin II containing fibres project in the

medial forebrain bundle, in the stria terminalis, in the periventricular regions (or ventral catecholamine bundle) and in the reticular formation. All of these pathways also contain catecholamine cell projections. However, a colocalization of fibre tracts is not indicative of functional interactions unless we can demonstrate overlap of terminal fields, either with other terminal fields or with cell bodies, of the two systems. These kinds of functional colocalization are seen in the paraventricular and supraoptic nuclei of the hypothalamus, which are terminal fields for catecholamine projections and are the major concentrations of angiotensin II cell bodies in the brain. They also contain many angiotensin II receptors (Mendelsohn et al., 1984). There are cell bodies which contain angiotensin II in the nuclei at both ends of the stria terminalis, at least one of which is a catecholamine terminal field. Similar types of colocalization can be demonstrated for many projection fields of the catecholamine system.

Summarised in table 1, taken from Lindvall and Bjorkland (1983) are the different projections from the noradrenergic lateral tegmental and dorsal medullary neurons. Every area with the possible exceptions of the dorsal raphe nucleus and the ventromedial hypothalamic nucleus, has either cell bodies or fibres containing angiotensin II.

This high degree of overlap with angiotensin II containing areas is not so apparent for the A6 and A4 (locus coeruleus) catecholamine projection areas as this projection is mostly to the cerebral cortex. However, the septal regions do have significant input from these norepinephrine cell groups and have angiotensin II cells and fibres. In the spinal cord the area of overlap between angiotensin II and catecholamines is the interomediolateral columns, which are








4

the preganglionic sympathetic nerve cell bodies. Another area of overlap which is worth specifically mentioning is the subfornical organ (SFO). This circumventricular organ located in the wall of the third ventrical is a major receptor- and cell body-containing area for angiotensin II. It is also rich in catecholamine terminals.

Table 1.1. Origins of different projections from the noradrenergic lateral
tegmental and dorsal medullary neurons.

TERMINAL AREA ORIGIN

Spinal cord A5, A7
NTS and dorsal vagal complex Al, A2
Spinal sensory trigeminal nucleus Al, A5
Locus ceruleus Al, A2, A7
Parabrachial nucleus Al
Dorsal raphe nucleus Al
Medial preoptic area Al, A2
Paraventricular hypothalamic nucleus Al, A2
Supraoptic nucleus Al, A2
Dorsomedial hypothalamic nucleus Al, A2
Ventromedial hypothalamic nucleus Al, A2
Anterior, lateral and posterior hypothalamic areas Al, A2 Median eminence Al, A2
Arcuate nucleus Al, A2
Septum Al, A2
Amygdala Al, A2

= areas without obvious angiotensin II content.
From Lindvall and Bjorkland, 1983.

There are, naturally, angiotensin II containing areas which do not share catecholamine inputs, e.g., substantia innominata, and catecholamine projection areas which have very sparse angiotensin II staining, e.g., cerebral cortex. However, the extensive nature of the colocalization suggests that the opportunity for interactions is present.

The results of ultrastructural studies confirm this view, though very few of these studies have been performed. It has been demonstrated that Ang II and arginine vasopressin (AVP) colocalize in paraventricular cells of the hypothalamus, (Hoffman et al., 1982). These AVP-angiotensin II-containing cells







5

are ringed by synaptic boutons from catecholamine cells (Sladek and McNeill, 1980).

In summary, catecholamines and angiotensin II colocalize extensively throughout the brain and spinal cord. The colocalizations are such that synaptic interactions could occur, i.e. terminal fields overlap, permitting axoaxonal synapses, or cell bodies and terminal fields overlap, allowing axodendritic or axo-somatic synaptic connections. Thus, on purely anatomical

grounds, interactions between catecholamines and angiotensin II are entirely possible.

Effects of Angiotensin II and Catecholamines which are Similar

Angiotensin II injected into the lateral ventricles causes drinking which is mediated through receptors located on the organum vasculosum of the lamina terminalis (OVLT), a circumventricular organ lying in the tip of the optic recess of the third cerebroventricle, (Buggy et. al., 1975). The OVLT also colocalizes angiotensin II and catecholamines. Similarly, norepinephrine injected into the brain causes drinking (Leibowitz, 1975a, b). Angiotensin II given into the brain causes increases in blood pressure, (Bickerton and Buckley, 1961, Severs et. al., 1970). Norepinephrine injected into certain sites of the brain also causes an increase in blood pressure, (Struyker-Boudier et. al., 1975), although at other sites it causes a decrease in blood pressure. These two strikingly similar actions of angiotensin II and catecholamines could be interpreted to mean that catecholamines act through releasing angiotensin II or vice versa. This possibility will be developed further later in this dissertation.



The Effects of Angiotensin II on Catecholamines

The effects of angiotensin II on catecholamine turnover. Angiotensin II

injected centrally induces increased turnover (or utilization) of norepinephrine








6

in brain regions whose function is to regulate blood pressure. Dopamine utilization in the same regions was unaffected. The regions examined contained angiotensin II receptors and catecholamines, and those affected were hypothalamus, SFO, Al, locus coeruleus and raphe magnus, (Sumners and Phillips, 1983). Fuxe et al. (1980) showed that central injection of renin, the enzyme which cleaves angiotensin I from angiotensinogen, increases dopamine turnover in the external layer of the median eminence, while decreasing norepinephrine turnover in the magnocellular paraventricular nucleus and the dorsomedial hypothalamic nucleus. The renin injection also increased

epinephrine turnover in the DCMO, a block of brainstem tissue containing the nucleus tractus solitarius, nucleus commisuralis and the dorsal motor nucleus of the vagus, sites of angiotensin II-catecholamine colocalization. The effects are specifically mediated by angiotensin II as they can be blocked by captopril, SQ 14225, an inhibitor of the conversion of angiotensin I to angiotensin II (Fuxe et al., 1980).

The effects of angiotensin II on the uptake and release of catecholamines.

In brain tissue Ang II affects both the release and re-uptake of catecholamines. Angiotensin II can evoke release of [3H]-dopamine from slices of striatum exposed to [3H]-tyrosine without affecting synthesis of [3H]-dopamine, an effect blocked by angiotensin II receptor blockers (Simmonet and Giorguieff-Chesselet, 1979). Angiotensin II does not seem to be able to cause spontaneous norepinephrine release from brain tissue, although it affects electrically stimulated norepinephrine release. At low doses, 0.1 and 1 nM it enhanced, and at high doses, 1 pM, it decreased the electrically evoked release of norepinephrine (Schacht, 1984). Captopril and HOE-498 diacid, two different converting enzyme inhibitors at 1 pM, were able to decrease the stimulated

outflow. Furthermore, the diacid had the same effect on release as clonidine,








7

an x2-agonist, and it could partially inhibit the effects of rauwolscine, an x2antagonist (Schacht, 1984). This strengthens the argument for interaction between the two systems.

Similarly, in a dose-dependent fashion, angiotensin II can increase the potassium-stimulated norepinephrine release from rabbit hypothalamus, an effect blocked by saralasin, an angiotensin II receptor antagonist (Garcia-Sevilla et al., 1979).

The effect of angiotensin II on electrically stimulated release of norepinephrine has been examined in more detail in a study by Meldrum et al., (1984). Angiotensin II increased the overflow of norepinephrine from the A2 region of rats, but not the hypothalamus. This stimulation was blunted in

animals on a low sodium diet. Wistar rats were more sensitive to the effect of angiotensin II on norepinephrine overflow from the A2 region than SpragueDawley rats, and a low sodium diet completely blocked the effect rather than just blunting it. Thus angiotensin II and catecholamines interact in the A2 region of the brain in a sodium-sensitive manner (Meldrum et al., 1984). The

A2 area contains angiotensin II receptors, angiotensin II cells and catecholamines, and is important in the control of blood pressure. These data suggest that these angiotensin II-norepinephrine interactions which are dependent on the sodium status of the rat, could be important in blood pressure homeostasis.

The in vitro findings discusssed above have also been demonstrated in vivo. Ventriculocisternal infusion of angiotensin II in rabbits caused an increase in the concentration of norepinephrine in the cerebrospinal fluid (CSF) which was highly correlated with the increase in blood pressure due to angiotensin II (Chevillard et al., 1979).








8

The effects of angiotensin II on norepinephrine uptake are more complex. It has been shown that Ang II partially inhibits norepinephrine uptake by brain tisssue, with no effect on release, after one hour of Ang II exposure (Palaic and Khairallah, 1967). But recent studies of neuronal cell culture have

demonstrated a biphasic effect. Shortly after Ang II exposure, enhancement of norepinephrine uptake was observed, while after longer exposure, inhibition was seen. This study did not examine release (Sumners and Raizada, 1986). The

previously cited study only examined one-time point, so the effects of angiotensin II on the reuptake of norepinephrine are still unclear. Peripherally, angiotensin II increases the amount of norepinephrine released spontaneously from sympathetic nerves as well as increasing release to sympathetic stimulation (Zimmerman and Gisslen, 1968, Hughes and Roth, 1971).

The effects of anpiotensin II on synthesis of catecholamines. In

sympathetically innervated tissues an acceleration of catecholamine biosynthesis is seen to angiotensin II treatment (Boadle-Biber et al. 1969, Davila and

Khairallah, 1971). These effects and alterations in uptake and release caused by angiotensin II (Reit, 1972; Feldberg and Lewis, 1964; Starke, 1971; Palaic and Khairallah, 1967) are probably interrelated as increased release would be expected to lead to increased synthesis. It is probable that these effects would be observed in brain tissue should the studies be performed. The Effects of Catecholamines on Angiotensin II

The effects of catecholamine blockade on responses to angiotensin II.

Angiotensin II delivered to the brain causes a pressor response (Bickerton and Buckley, 1961). This blood pressure increase can be attenuated by oantagonists (Camacho and Phillips, 1981), implying that it is partly dependent upon catecholamines. Furthermore, the central dipsogenic response and central pressor response to angiotensin II can be attenuated by the destruction of








9

catecholamines in the brain by the catecholamine neurotoxin 6-hydroxydopamine (6-OHDA) (Gordon et al., 1979, Gordon et al, 1985). Drinking was restored in

6-OHDA treated animals when transplants of norepinephrine producing brain areas from fetal rats were made into their basal forebrains (McRae-Dequeurce et al., 1986). In addition, clonidine, an c2-receptor agonist, can inhibit drinking to angiotensin II through an Oc2-adrenergic receptor-mediated mechanism, (Fregly et al., 1984). These findings clearly suggest that the full central pressor and dipsogenic responses of angiotensin II are dependent on intact catecholamine systems.

Regulation of the central angiotensin II system by catecholamines

Indications that the central angiotensin II system is regulated by catecholamines are revealed in several interesting studies. Initially, a correlation between angiotensinogen concentration and norepinephrine concentration in brain tissue, but no such correlation of dopamine and angiotensinogen concentrations, was observed (Printz et al., 1979). When rats were treated with reserpine, a catecholamine-depleting drug, for 24 hours, angiotensinogen was increased in the septum, periaqueductal gray, nucleus stria terminalis, medial basal hypothalamic nuclei, hippocampus and ventral tegmental areas of the brain. The levels of angiotensinogen had returned to pretreatment levels 96 hours after reserpine, even though catecholamines were still depleted. The increase in angiotensinogen at 24 hours after reserpine could be annuled by adrenalectomy, which prevented the rise in corticosteroids which follows reserpine treatment. However, corticosteroid levels regulate norepinephrine levels and thus the authors postulated that the angiotensinogen concentration changes were possibly due to alterations in norepinephrine levels, secondary to corticosteroid changes (Printz et al. 1983). These hypotheses have still not been tested.








10

It is difficult to interpret these studies because interference with the catecholamine system, for example by reserpine, causes changes in fluid balance, by such factors as the reserpine-induced diarrhea, and the necessity for intact catecholamine systems for full expression of drinking behaviours, etc. Since the brain angiotensin system is thought to regulate fluid balance, it would be expected that these fluid balance changes alone could alter the central angiotensin II system directly, without any influence of catecholamine depletion. A series of studies which addressed this problem has clarified the situation to some extent. The approach taken was to change catecholamine

neurotransmission in the brain and/or periphery with various drugs and examine the effects on the angiotensin II and catecholamine systems. The results of these studies show that there is an interaction between the systems but do not permit exact mechanisms to be illuminated (Mikulic et al, 1986, Kurnjek et al, 1986, Trolliet et al., 1986 and Basso et al, 1984). It will probably be necessary to examine regulation of mRNA for all components of both of the systems before we fully appreciate these interactions (Basso, personal communication).

The effects of catecholamines on anpiotensin II release. It has recently

been shown that isolated mesenteric arteries treated with iosproterenol, a #receptor agonist, release angiotensin II (Nakamura et al., 1986a, b). The release of angiotensin II caused by isoproterenol could be blocked by propranolol, a #blocker, suggesting the effect was mediated by #-receptors. The basal release

of angiotensin II was unaffected by /-blockade, but was sensitive to captopril, a converting enzyme inhibitor. This shows that there are two mechanisms governing the release of angiotensin II from mesenteric arteries. These are

basal release which is unaffected by #-receptors and stimulated angiotensin II release, which is under /-catecholaminergic control.








11

Gothert and Kollecker (1986) performed a similar study in the inferior vena cava of rats. They were able to suggest from experiments which

manipulated the tissue slices, that the 02-receptors, whose activation led to angiotensin II release, were located on the subendothelium, most probably on smooth muscle cells. The angiotensin II receptors whose activation stimulated norepinephrine release were probably located on the sympathetic nerve terminal.

In a related study, the existence of a similar mechanism for release of angiotensin II in the kidney was examined. The authors interpret their data as showing that #-mediated angiotensin II release does not occur, but on close examination of this paper, it is possible to draw the opposite conclusion (Rump and Majewski, 1987). In their study high, but not low, doses of captopril, a converting enzyme inhibitor, could block the 8-adrenoceptor mediated increase in norepinephrine release from the isolated kidney. Saralasin, an angiotensin II receptor blocker, had no effect on the fl-adrenoceptor mediated release. The

authors could not explain the effects of high doses of captopril, and saralasin is a very difficult drug to use as it has agonist as well as antagonistic properties. As a whole these considerations mean that it is not impossible for a #-receptor mediated release of angiotensin II to occur in the kidney, which in turn induces increased release of norepinephrine.

Thus, in three areas examined, there is evidence which supports the hypothesis that f-receptor stimulation causes angiotensin II release. A suggested mechanism is via 8-adrenergic stimulation of converting enzyme activity (Gothert and Kollecker, 1986), although this has not been proven. An examination of the control of angiotensin release from rat hypothalamus by #adrenoceptors has recently been made. In this study, performed in vivo with push-pull cannulae in rats, there was no evidence that fl-adrenergic stimulation led to increased angiotensin II release (Brosnihan et al., 1988).







12

Cell Culture Model.

Apart from the few studies mentioned above, nothing else was known about catecholamine-angiotensin II interactions until brain cell culture methodology was applied to the problem. In these cultures it is possible to

look at direct interactions between these systems without secondary influences. Thus a great deal of new information about the interactions of angiotensin II with catecholamines has been elucidated, as well as other factors which regulate the Ang II system in the brain.

Characterization of the Cell Culture System.

Initial studies showed that all components of the renin-angiotensin system are present in brain cell cultures. They contain renin, (Hermann et al., 1987), angiotensinogen (Hermann et al., 1988a), angiotensin I, (Hermann et al., 1988a, b), and angiotensin II (Raizada et al., 1984b, Hermann et al., 1988a, b). Converting enzyme activity has not been directly measured, but must be present because of the production of angiotensin II. The cells incorporate [3H]isoleucine into immunoprecipitable angiotensin II (Hermann et al 1988a). They have specific receptors for angiotensin II (Raizada et al., 1981, Raizada et al., 1987).

Many components of the catecholamine system are present in these cultures. The catecholamines dopamine, norepinephrine, and epinephrine, can be measured and in some cases visualized, implying that there are synthetic enzymes present (Sumners et al., 1983a). The catecholamine degrading enzymes, catechol-o-methyl transferase (COMT) and monoamine oxidase (MAO) are present (Sumners et al., 1987). There are ocI-receptors linked to

phosphatidylinositide hydrolysis (Feldstein, 1986), #-receptors stimulation of which results in cAMP accumulation (Baker et al. 1986). Thus brain cell

cultures are a good model in which to examine these two systems' interactions.








13

The Effects of Angiotensin II on Catecholamines.

Angiotensin II treatment of neuronal cell cultures resulted in an increase in the cell and media content of catecholamines (Sumners et al., 1983b). This implied that both synthesis and release of catecholamines was enhanced by angiotensin II. Later studies with labelled tyrosine showed that Ang II stimulated the conversion of dopamine to norepinephrine, implying that angiotensin II stimulated the enzyme dopamine f-hydroxylase, (McLean, Raizada and Sumners, unpublished data). Furthermore angiotensin II enhanced the activity of MAO, but not COMT, the catecholamine degrading enzyme, (Sumners et al., 1987), suggesting that it first increased release of catecholamines. However, release of catecholamines from cell cultures has never been directly examined.

Angiotensin II has significant effects on norepinephrine uptake mechanisms, which are somewhat difficult to interpret without appreciation of the status of release of catecholamines under the same conditions. Shortly after exposure to angiotensin II, norepinephrine uptake is enhanced, whereas after longer periods of time, uptake is decreased (Sumners and Raizada, 1986).

With this multiplicity of effects of angiotensin II on the catecholamine system, it was expected that catecholamine receptors would also be affected by angiotensin II. This was not true, however, of ocl-adrenoceptors, nor of freceptors (unpublished data).

Catecholamine Effects on Angiotensin II in Cell Culture

The effect of norepinephrine on angiotensin II receptors. Norepinephrine has been shown to be an important regulator of angiotensin II receptors. Adding norepinephrine to neuronal cultures resulted in a decrease in the number of angiotensin II receptors which is mediated by oc-adrenergic receptors (Sumners, Watkins and Raizada, 1986). Decreasing norepinephrine levels with o-








14

methyl-p-tyrosine resulted in an increase in the number of angiotensin II receptors (Sumners and Raizada, 1984).

The effects of catecholamines on other aspects of the angiotensin II system. Knowledge of other influences of catecholamines on the

angiotensin II system, for example on metabolism, are very limited, mainly because they have not been examined.

General Hypothesis.

From this survey of the literature dealing with catecholamine-angiotensin II interactions, it is obvious that some big areas of potential interaction have not been examined. For example, the possibility that angiotensin II could be

released from brain cell cultures by catecholaminergic mechanisms, as in the periphery. Furthermore, norepinephrine acts on three types of receptor, Ol, c2 and P. The existence of m2-receptors in these cultures had not been examined, nor had their role, if any, in the interaction between catecholamine and angiotensin II been tested. Alpha2-receptors colocalize with angiotensin II in the brain. In two seperate studies, m2-receptor distributions were mapped in the brain. c2-receptor densities were highest in areas which stain heavily for angiotensin II, e.g. the norepinephrine terminal fields, LC and NTS, interomediolateral cell column of the spinal cord, substantia gelatinosa of the trigeminal nucleus, paraventricular and arcuate nuclei of the hypothalamus, amygaloid nuclei, anterior olfactory nuclei, lateral septum, dorsal parabrachial nucleus, and bed nucleus of the stria terminalis (Unnerstall et al., 1984, Boyajian et al., 1987).

Interest in m2-receptors in these interactions was generated for several reasons. Firstly, c2-receptors influence norepinephrine release, as does angiotensin II. Could there be a common mechanism of action? Alpha2receptors also modulate the release of other substances, for example, serotonin,








15

acetyl-choline and endorphin (Frankhuyzen and Mulder, 1980, Gothert and Huth, 1980, Vizi and Knoll, 1971). Could Oc2-receptors modulate angiotensin II release? If c2-receptors did influence release of angiotensin II, would synthesis of angiotensin II also be modulated?

These questions led to the establishment of three specific aims 1. Do neurons and glia in cell culture possess m2-adrenoceptors?

2. Can m2-receptors modulate angiotensin II release? If so, can effects on

angiotensin II metabolism, specifically on renin concentration, also be seen?

3. Can a common mechanism of angiotensin II and m2-receptors be discovered which would explain how angiotensin II and m2-receptors modulate NE release?














CHAPTER TWO
CHARACTERIZATION OF Oc2-ADRENOCEPTORS ON NEURONAL AND GLIAL CELLS IN CULTURE.


Introduction

The discovery of presynaptic inhibitory autoreceptors which modulate the release of norepinephrine was one of the factors which solidified the view of two subtypes of alpha-adrenergic receptors, the ao- and the a2-receptors (e.g., Langer, 1977; Starke, 1977). Since the earlier theories of presynaptic inhibitory a2-receptors it has become apparent that this subclass of receptors is not exclusively presynaptic. For example, a2-receptors are found localized on vascular smooth muscle cells and pancreatic islets, and are also located postsynaptically in the central nervous system (Langer et al., 1985). The actual postsynaptic location of Ce2-receptors in the central nervous system is uncertain, but the implication is that they are present on postsynaptic cell bodies and dendrites, and mediate functions such as bradycardia and decreasing blood pressure (Langer et al., 1985). However, it has been established that astrocytic glial cells contain al-adrenergic (H6sli et al., 1982; Hirata et al., 1983; Murphy et al., 1987) and fl-adrenergic receptors (McCarthy, 1983; Baker et al., 1986), and one report has suggested that mouse cortical astrocytes contain specific a2adrenergic receptors (Ebersolt et al., 1981). The aim of the present study was, using brain cell culture techniques, to determine whether specific a2-receptors exist on both neuronal and glial cells and, if so, to compare their properties.

The development of differential cell culture techniques has permitted the production in our laboratories of -95% pure astrocytic glial cultures or 85-90% 16








17

pure neuronal cultures (the remaining cells being of glial origin) from one-dayold rat brains. This ability allows the properties of the different cell types to be examined in ways which are impossible in whole brain. Furthermore, the

results are easier to interpret as the cells are not transformed as many cell lines are, and thus more nearly approximate the in vivo situation. Recent studies from our laboratories have described some of the characteristics of these cultured cells. The glial cultures contain large flat cells in a confluent monolayer (Raizada et al., 1982; Raizada et al., 1987). These cultures, when

stained by antibodies against glial fibrillary acidic protein, a specific astrocyte marker, revealed that more than 95% of the cells were astrocytes of both the

polygonal and process-bearing varieties (Clarke et al., 1984). The neuronal cultures, when stained with an antibody against neuronal specific enolase, showed that 70-80% of the cells were neurons (Raizada, 1983). The remainder were astrocytic glia. The neuronal cultures contain ai-receptors (Feldstein et al., 1986) which are linked to inositol phospholipid hydrolysis (Gonzales et al., 1985), f#1-receptors which are linked to cAMP production (Baker et al., 1986), norepinephrine and dopamine (Sumners et al., 1983a), epinephrine (Sumners, unpublished data), sodium-dependent norepinephrine and dopamine uptake sites (Sumners and Raizada, 1986; Sumners et al., 1987), monoamine oxidase (MAO) and catecholamine-O-methyltransferase (COMT) activities (Sumners et al., 1987). Glial cultures also contain many of these components, e.g., fl-adrenergic receptors (Baker et al., 1986), norepinephrine and dopamine (10-20 times less than neurons) (Sumners et al., 1983a), cel-adrenergic receptors (Masters et al., 1988), sodium-independent NE uptake sites (Sumners and Raizada, 1986), and MAO and COMT activities (Hansson and Sellstr~m, 1983).








18

Obviously lacking was information about a2-adrenergic receptors to complete the picture of the catecholamine systems within these cell types. In this study we have biochemically demonstrated the presence of a2-adrenergic receptors in membranes from both neuronal and glial cells in culture, as determined by the characteristics of [3H]-yohimbine binding. This is in contrast to the classical view of exclusive association of a2-receptors presynaptically with neurons in the central nervous system. These studies provide a basis for further regulatory and functional experiments concerning central a2adrenoceptors.

Methods

Preparation of Neuronal Cultures

Neuronal cultures were prepared exactly as described (Sumners et al., 1987) previously. Brains were dissected free from one day old Sprague-Dawley rats and placed in an isotonic salt solution containing 100U penicillin G potassium, 100 jg streptomycin, and 0.25 jg amphotericin B (Fungizone) per ml, pH 7.4.

All blood vessels and pia mater were removed. The brain was chopped into

approximately 2 mm pieces, suspended in 20 ml of 0.25% (wt/vol) trypsin in isotonic salt solution (pH 7.4) and placed in a shaking water bath for 6 mins. at 370 C to dissociate the cells. After this time, 160 pg deoxyribonuclease I (DNase I) were added and the cells were shaken for a further 6 mins. Following this, the cells were suspended in Dulbecco's modified Eagle's medium (DMEM) containing 10% plasma-derived horse serum (PDHS), centrifuged at 1000 g for 10 mins. and washed with 50 ml DMEM containing 10% PDHS. Cells

were resuspended in DMEM containing 10% PDHS and were plated at a density of 18x106 cells/100-mm Falcon culture dish for binding studies, and at 3x106 cells/35-mm dish for the autoradiography experiments and the immunofluorescent








19

staining. The cells were incubated for 3 days at 370 C in a humidified

incubator with 5% C02-95% air. On day 3, cells were treated with 10 pM cytosine arabinoside (ARC) prepared in DMEM containing 10% PDHS. This

treatment results in an inhibition of cell multiplications, the majority of which are non-neuronal in origin, and provides cultures enriched in neuronal cells. After 3 days, this medium was replaced with DMEM containing 10% PDHS, and the cells allowed to grow for a further 9-10 days before being used in experiments.

Preparation of Glial Cultures

Glial cell cultures were prepared as previously described (Clarke et al., 1984). The method was as above except that 18x106 cells suspended in DMEM containing 10% fetal bovine serum (FBS) were plated in 100 mm diameter Falcon tissue culture dishes. After 3 days, the medium was replaced by 10 mls fresh DMEM containing 10% FBS. After a further 3 days, cells were passaged at a concentration of lx106 cells/100 mm diameter culture dish. The cells were then allowed to grow and divide for a further 20 days prior to use, at which time they were confluent.

Preparation of Membranes from Neuronal and Glial Cultures

Cells were scraped from the 100-mm culture dishes with the aid of a rubber policeman and placed on ice in a centrifuge tube. They were spun for 10 mins. at 40 C at 1000 g and the supernatant discarded. The pellet was

resuspended in phosphate buffered saline (PBS) and either used or frozen in liquid nitrogen until ready for use, preliminary experiments having shown that there was no significant difference between binding in fresh and frozen cells.

On the day of the binding experiment, the cells were thawed on ice and then resuspended in 20 volumes of ice-cold 50 mM Tris-HCl, pH 7.0 with 0.1%








20

ascorbic acid (binding buffer). They were homogenized using a polytron (setting 2.5 for 30 s.), then centrifuged at 50,000 g for 10 minutes at 40 C. The

supernatant was discarded, the pellet resuspended in 20 volumes of binding buffer and the centrifugation step repeated. Following this, the pellet was resuspended in 1 ml of binding buffer and the protein content of an aliquot determined by the method of Lowry et al. (1951), using bovine serum albumin as the standard, while the remainder were kept on ice. [3H-Yohimbine Binding Assays

Binding was performed in glass test tubes to which was added 50 pl of diluted [3H]-yohimbine (specific activity 80-90 Ci/mmol, New England Nuclear), 50 pl of various concentrations of drugs and buffer to bring the volume up to 200 pl. Fifty pl of the appropriately diluted membrane suspension containing 300 pg protein (neuronal cultures) or 175 pg protein (glial cultures) were added to start the reaction. Binding was performed at 250 C with shaking for 5 mins. and the reaction stopped by rapid filtration under reduced pressure through Whatman GF/B filters. The filters were rinsed 3 times with 3 ml of ice-cold binding buffer, transferred to vials containing 10 ml of Liquiscint, left for 12 hours and then counted in an LKB 1215 liquid scintillation counter at an efficiency of 40-50% for tritium.

Non-specific binding was determined in triplicate samples containing 100 pM yohimbine, and specific binding was defined as total minus non-specific binding. There was no significant change in the specific binding whether 100 pM yohimbine or 1 pM yohimbine was used as the cold displacer. The level of specific binding obtained with neuronal membranes was 40-50% of total, and with glial membranes was 50-80% of total binding.








21

Kinetic and Pharmacolopical Analyses of f3HI-Yohimbine Binding

Association curves were generated by allowing the binding reaction to proceed for various lengths of time before terminating the reaction by filtration. Dissociation curves were derived by allowing the binding to reach equilibrium, adding yohimbine to a final concentration of 100 pM and filtering at different time points following that addition. Scatchard analysis of saturation experiments were performed on data generated by following binding at concentrations of [3H]-yohimbine from 1 to 24 nM. Competition curves were generated by incubating membranes from neuronal or glial cultures with 10 nM [3H]-yohimbine in the absence or presence of various concentrations of agonist or antagonist.

Results

Kinetic Analysis of [3HI-Yohimbine Binding in Neuronal and Glial Cultures

Binding of [3H]-yohimbine to membranes prepared from either neuronal or glial cultures was, in both cases, linear with respect to protein concentration. However, it was apparent that glial membranes exhibit higher specific binding of [3H]-yohimbine compared with neuronal culture membranes on a per-milligramprotein basis. The binding of [3H]-yohimbine at 25*C was rapid, reaching

equilibrium by 3 mins. in both neuronal and glial membranes (see representative expts. in Fig. 2), and also in both cases remaining stable for at least 20 mins (data not shown). Thus, in subsequent experiments an incubation period of 5 mins. was used. The binding of [3H]-yohimbine to neuronal and glial membranes was also reversible, and the pattern of dissociation in each case can be seen from the representative experiments in Fig. 3. From the association curves the apparent rate constants for association (Kobs) were calculated by linear regression analysis of








22

In Beg
( (Beq Bt)



where Beq is the specific binding at equilibrium and Bt is the specific binding at each time point. From the dissociation curves, the rate constants for dissociation (K-1) were calculated by linear regression analysis of



In ( Bt
BO



versus time, where B0 is the amount of [3H]-yohimbine specifically bound at equilibrium, and Bt is the specific binding at various times after the addition of unlabelled yohimbine. The association rate constants (KI) for [3H]-yohimbine binding to membranes from both neuronal and glial cultures were derived from the equation



Kobs K-1
K1 [[3H]-YOH]



where K_1 was the mean dissociation constant calculated for that particular culture.

For neuronal cultures, K, was 3.199 t 0.92 x 106M-IS-1 (n = 4 expts.), and for glial cultures K, was 0.962 t 0.418 x 106M~1S-1 (n = 3 expts.). For

neuronal cultures, K_1 was 12.96 2.8 x 10-3S-1 (n = 3 expts.) and for glial cultures, K_1 was 17.4 3 x 10-3S-1 (n = 4 expts.). The equilibrium

dissociation constants determined from the ratios of K_1/K were 4.05 nM for neuronal cultures and 18.06 nM for glial cultures. These data are summarized in Table 2.2.








23

[3H1-Yohimbine Binding in Neuronal and Glial Cultures: Saturation Analyses

Saturation experiments were performed with 1 to 24 nM of [3H]-yohimbine. Fig. 4 is a representative experiment which shows that in neuronal cultures, specific binding reached a plateau between 7.5 and 15 nM [3H]-yohimbine. Scatchard (1949) analysis of the data from this experiment using a computer assisted linear regression analysis showed an apparent KD of 12.59 nM and a maximum number of binding sites (Bmax) of 0.196 pmol/mg protein (see Fig. 4 inset). This experiment has been repeated 10 times with similar results and mean KD and Bmax values are given in Table 2.2.

In glial cultures the specific binding of [3H]-yohimbine increased with increasing concentrations of [3H]-yohimbine up to 24 nM, as seen in the representative experiment in Fig. 5. With concentrations of [3H]-yohimbine greater than 24 nM, large variations in specific binding were observed, and so these data were not included in our analyses. Scatchard analysis of the data shown in Fig. 5 revealed a KD of 16.5 nM and a Bmax of 2.18 pmol/mg protein for [3H]-yohimbine binding (see Fig. 5 inset). This experiment has been

repeated 10 times with similar results, and mean KD and Bmax values are shown in Table 2.2. Thus, the KD is similar in neuronal and glial cultures, but the Bmax is more than an order of magnitude higher in the glia (Table 2.2). Pharmacological Properties of the [3H-Yohimbine Binding Sites in Neuronal and
Glial Cultures

The pharmacological properties of the [3H]-yohimbine binding site were examined by comparing the ability of various catecholaminergic drugs to compete for the binding. This was performed by adding increasing concentrations of the catecholaminergic drugs to the incubation as described in the Methods, and IC50 values were calculated by finding the concentration of drug which displaced 50% of the specifically bound [3H]-yohimbine.







24







A B

3 0 e600 ~
CL


0.


2E 6- e 400
-0


5- 5C

3- 31: -4
.0 0%2 X2

>10- 2.,200- r 2
I) C 0

0 30 60 90 0 30 60 90 120
TIME (secs) TIME (seCs)
of I 1 0 1 1 1
60 120 180 240 300 60 120 180 240 300

TIME (secs) TIME (secs)





Figure 2. Representative experiments showing the binding of [3H]-yohimbine to membranes prepared from neuronal or glial cultures as a function of incubation time. Membranes prepared form neuronal (panel A) or glial (panel B) cultures were incubated at 250C in a final volume of 250 pul 50 mM Tris HCl/0.1% ascorbic acid (pH 7.0) containing 10 nM [3H]-yohimbine, in the absence or presence of 100 pM yohimbine. At the indicated time points, the free [3H]yohimbine was removed from the membrane bound radioactivity by washing the membranes three times with ice-cold buffer, and filtration under reduced pressure. Non-specific binding was subtracted from the total binding and the data were presented on the y-axis as fmol [3H]-yohimbine specifically bound per mg protein. Each point is a mean of triplicate determinaltions. These experiments were repeated 4 times (neuronal cultures) and 3 times (glial
cultures) with similar findings.
Insets. The inset graphs in panels A and B are kinetic analyses of the
association data from neuronal and glial cultures, respectively. Data are plotted according to a first-order rate equation, and the apparent association rate
constants (Kobs) were calculated as detailed in the Results section.







25






40 A 800TIME (sece) TIME (secs)
0 60 180 300 0 60 180 300
0 0


-30 0 600 -2
0 -4
-2-4

E
400
6 20
E

-4.
0
.0
~io1- 20001I 0 0
0 60 120 180 240 300 0 60 120 180 240 300
TIME (secs) TIME (secs)



Figure 3. Representative experiments showing the dissociation of bound [3H]yohimbine from membranes prepared from neuronal or glial cultures.
Membranes from neuronal (panel A) or glial (panel B) cultures were incubated with lOnM [3H]-yohimbine for 5 mins as detailed in the legend to figure 2. After this time, unlabelled yohimbine was added to each reaction tube so that a final concentration of 100 pM of the antagonist was attained. At the indicated time points the membranes form triplicate samples were collected by filtration under reduced pressure, and were washed three times with ice-cold buffer to remove any unbound radioactivity. Data are presented as fmol. [3H]yohimbine remaining bound per mg protein at each time point. The data given at each time point are mean values of triplicate determinations. These experiments were repeated 3 times (neuronal cultures) and 4 times (glial cultures) with similar findings.
Insets: The inset graphs in panels A and B are kinetic analyses of the dissociation data from neuronal and glial cultures, respectively. Data are plotted according to a first-order rate equation, and the dissociation rate constants (K_1) were calculated as detailed in the Results section.








26






15
B/F
10

5


0 40 80 120 160 250

200

150

0 100.0



0 4 8 12 16
[3H]-YOH nM







Figure 4. Binding of [3H]-yohimbine to membranes from neuronal cultures as a function of yohimbine concentration
Neuronal membranes were incubated with increasing concentrations of
[3H]-yohimbine in the absence (A) and the presence (0) of 100 pM yohimbine. After incubation for 5 min at 250C, membrane bound radioactivity was collected on glass fiber filters as described in the methods. Specific binding (*) was determined from the difference between binding in the absence and presence of 100 pM yohimbine. Data are presented as fmoles [3H]-yohimbine bound per mg protein, being the mean of triplicate determinations. This is a representative experiment which was repeated ten times with similar findings.
Inset: The above specific binding data were analyzed accordin to Scatchard (1949) to determine the apparent dissociation constant (KD) for [ H]yohimbine binding and the maximum number of binding sites (Bmax). B = specifically bound [3H]-yohimbine in fmoles per mg protein. F = free [3H]yohimbine concentration in nM.








27









B/F
10

1.5
5

00
0.
12 3
E
B


0
*0
EC




0.5






0 12 4
[3H]-YOH nM









Figure 5. Binding of [3H]-yohimbine to membranes from glial cultures as a
function of yohimbine concentration.
This experiment was performed as described for figure 4 except that membranes from glial cultures were used instead of membranes from neuronal cultures.
(A) total [3H]-yohimbine bound; (0) non-specifically bound [3H]-yohimbine; (.) specifically bound [3H]-yohimbine. This is a representative experiment which was repeated ten times with similar findings.
Inset: Scatchard analysis of the above specific binding data. B = specifically bound [3H]-yohimbine in pmoles per mg protein. F = free [3H]yohimbine in nM.








28

These IC50 values are summarized in Table 2.1 for both neuronal and glial cultures, and the potency series show that in both cases the [3H]-yohimbine binding is displaced most strongly by drugs which are recognized to be selective for C2-adrenoreceptors. For example, yohimbine and rauwolscine (a2-adrenergic antagonists) are much better displacers of the binding of [3H]-yohimbine than prazosin or corynanthine which are selective ai-antagonists. The same was generally true of the agonists, with clonidine and naphazoline having good displacing ability in glial and neuronal cultures. In neuronal cultures, clonidine had a high affinity for the [3H]-yohimbine binding site, but naphazoline was apparently not so effective. The reason for this is unclear and does not fit well with the rest of the competition data. Overall, these data suggest that the [3H]-yohimbine binding site has the characteristics of an U2-adrenergic receptor.

Calculation of Ki values from the competition of [3H]-yohimbine binding with unlabelled yohimbine, using the following equation, Ki= IC50
I 1 + [L]/KD

gave values of 1.4 nM for neuronal cultures and 16.3 nM for glial cultures. These values are shown in Table 2.2 in comparison with the KD values obtained from both Scatchard and kinetic analyses.

Effects of Gpp(NH)p on r3Hl-Yohimbine Binding in Neuronal and Glial Cultures

Alpha2-adrenergic receptor affinity for agonists is regulated by, among other things, GTP. To determine whether the a2-adrenergic receptors in

neuronal and glial cultures were also sensitive to GTP, agonist and antagonist competition curves were performed in the presence and absence of 50 pM 5'guanylyl-imidodiphosphate (Gpp(NH)p), a non-hydrolyzable analog of GTP. Representative experiments are shown in Fig. 6. In membranes from both cell types, Gpp(NH)p caused a decrease in the ability of the agonist NE to








29



TABLE 2.1. Displacement of [3HI-Yohimbine Binding By Agonists and Antagonists



Treatment Neuronal cultures Glial cultures

IC5O (pM) IC50 (pM)


Antagonists

Rauwolscine 0.0338 0.026 (4) 0.023 0.004 (5)

Yohimbine 0.0025 0.0004 (6) 0.024 0.002 (9)

Prazosin 1.07 (2) 0.74 (2)

Corynanthine 1.78 0.72 (3) 4.36 2.5 (5)

Propranolol 5.67 2.2 (3) -Sulpiride 15.41 1.98 (3) 15.75 (2)

Aponists

Naphazoline 2.69 (2) 0.03 0.016 (3)

Clonidine 0.019 (2) 0.73 0.45 (3)

Norepinephrine 4.787 0.71 (12) 5.81 1.22 (12)

Epinephrine 14.3 8.7 (3) 3.2 1.4 (3)

Dopamine 6.25 (2) 6.25 1.25 (4)

Phenylephrine 8.5 (2) 17.81 7.35 (4)



Data are means SEM of the number of experiments indicated in the parentheses. -- = not tested.








30



TABLE 2.2. [3H]-vohimbine binding in membranes from neuronal
and glial cultures: comparison of K, and Bmax values





Neuronal cultures Glial cultures


KD (nM)A 13.73 1.35 (10) 18.42 2.34 (10)

Bmax 0.143 0.018 (10) 1.60 0.33 (10)
(pmol/mg protein)

KD (nM)B 4.05 (4) 18.06 (3)

Ki 1.40 (2) 16.34 (4)


A -KD obtained from saturation experiments/Scatchard analyses. B -KD obtained from kinetic analyses. Ki -obtained by Cheng-Prusoff transformation of competition data with unlabelled yohimbine.

Data are means SEM, and figures in parentheses are the numbers of experiments.



compete for the [3H]-yohimbine binding site, as seen by the right hand shift of the displacement curves, but did not affect the affinity of the antagonist, yohimbine, for the site. In neuronal cultures, NE competition curve IC50 values were shifted from 4.9 pM to 70.8 pM (n = 2 expts), while in glial cultures the shift was from 6.7 + 2 pM to 118 + 9.3 pM (n = 4 expts). Representative experiments are shown in Fig. 6a,b. For the antagonist, yohimbine, the ICSO values were 2.78 vs 1.75 nM (n = 2 expts) in neuronal cultures and 23.1 + 1.2 vs 22.6 + 1.4 (n = 4 expts) in glial cultures, with and without Gpp(NH)p,








31

respectively. Representative experiments are shown in Fig. 6a, b. GTP was

also capable of causing a similar shift but the magnitude of the change was less than that seen with Gpp(NH)p, probably because the GTP was hydrolyzed more rapidly than Gpp(NH)p. The affinity of clonidine, another a2-agonist, was

affected similarly by GTP (results not shown).

Discussion

In this study we have determined the characteristics of a2-adrenergic receptors in neuronal and glial cells in culture. By differential culture techniques, the preparation of -95% pure astrocytic glial cell cultures and 8590% pure neuronal cultures is possible. We have identified binding sites for [3H]-yohimbine, with characteristics of a2-adrenergic receptors in both neuronal and glial cultures. The neuronal receptor has similar characteristics to those previously described for [3H]-yohimbine binding in rat brain (Rouot et al., 1982).

The a2-adrenergic receptors of the two cell types appear to have similar characteristics in general, although some features are obviously different between the cell types. For example, the glial cultures contain many more

binding sites per milligram protein than the neuronal cultures. However, Scatchard analyses revealed that the [3H]-yohimbine binding sites in glial cultures are of slightly lower affinity than in neuronal cultures. This difference in affinity was enhanced when the KD values were derived from kinetic or competition analyses (Table 2.2). These differences are not large enough to suggest two different binding sites, but the different cellular environment of each site may cause the change in affinity. The potency order for displacement of [3H]-yohimbine binding by catecholaminergic drugs was similar for neuronal and glial cultures and suggestive of an C2-adrenergic receptor type. Alpha2adrenergic receptors are associated with adenylate cyclase such that their








32











00 eo 0
100
U


75 -5
0
100 E0




V 0I
Cz
00


25

0
coU


00
12 11'10 9 8 7 6 5 4 3 12 11 10 9 8 7 6 5 4 3

[DISPLACER] (-LOG M)









Figure 6. Competition of [3H]-yohimbine binding in neuronal and glial cultures in the absence and presence of Gpp(NH)p.
Panel A: Competition of specific [3H]-yohimbine binding in membranes from neuronal cultures by norepinephrine in the absence (m) or presence (0) of 50 pM Gpp(NH)p, or by yohimbine in the absence ( or presence (0) of 50puM Gpp(NH)p. The y-axis is % specifically bound [ H]-yohimbine, 100% binding being obtained in the absence of any yohimbine. These data are means of
triplicate determinations and are representative experiments. Each of the experiments was repeated twice with similar results.
Panel B: As for panel A, but membranes from glial cultures were used
instead of those from neuronal cultures. The data are means of triplicate determinations and are representative experiments. These experiments were repeated four times with similar results.








33

stimulation inhibits the production of cAMP (Jakobs, 1979). The linkage of the a2-receptor to adenylate cyclase is via the GTP binding protein [Gi], whereas those that stimulate adenylate cyclase are linked through the Gs subunit, e.g., #-receptors. GTP regulates the affinity of the a2-receptor so that when GTP is present, agonists are less able to bind to the receptor. In both

neuronal and glial cells in culture, GTP and its nonhydrolyzable analog Gpp(NH)p were able to cause this shift in agonist affinity, again suggesting that the receptors have a2-adrenergic receptor characteristics.

The astrocytic glial cultures contain a large number of [3H]-yohimbine binding sites compared to neuronal cultures. Studies in the whole brain have suggested that the neurotoxin 6-hydroxydopamine, which destroys catecholaminecontaining neurons, does not abolish all a2-adrenoceptor binding (U'Pritchard et al., 1979). Several explanations of this finding are possible. For example, the a2-receptors are postsynaptic as opposed to presynaptic, or they are present on cells containing other neurotransmitters, e.g., 5HT, ACH (Frankhuyzen and Mulder, 1980; Gothert and Huth, 1980; Vizi, 1972). This study suggests that

perhaps it is the glial receptors which are not destroyed by the neurotoxin. However, it may also be true that glial cells in culture express a2-receptors whereas those in adult rat brain do not. Preliminary studies from our laboratory using glia made from the brains of 30-day-old rats, as distinct from newborn rats, suggest that the a2-receptor is present on these glia and of similar characteristics. However, this still does not preclude the fact that the culture technique may cause expression of the receptor.

Another major concern arising from these studies has been the possibility that neuronal cells do not possess a2-receptors. The [3H]-yohimbine binding

seen in neuronal cultures would, according to this theory, arise from the 10-15%








34

contamination of neuronal cultures with glial cells. Several findings suggest that this is probably not true. Firstly, we have demonstrated with autoradiography that [3H]-yohimbine binds to cells which have been defined with immunofluorescent techniques as neurons (see Chapter Three). Secondly, a2-antagonism causes release of the octapeptide angiotensin II from neuronal cultures but not from glial cultures, even though both cultures contain angiotensin II (see Chapter Four).

We were also concerned that the site might be a norepinephrine uptake site in glia. To examine this possibility, we performed competition curves of the catecholamine uptake blockers maprotiline and desmethylimipramine (DMI) against [3H]-yohimbine binding. The IC50 for both drugs was approximately 10 MM. At higher concentrations, the membranes clumped and precipitated out of solution. However, as specific C2-adrenergic antagonists and agonists were much more effective at displacing the binding, it seems likely that the binding site for [3H]-yohimbine is an a2-adrenergic receptor rather than a norepinephrine uptake site.

In summary, these experiments show that cultures of -90% pure glia and 85-90% neurons both contain C2-adrenergic receptors and suggest that these receptors are present on both cell types. Glial receptors are of slightly lower affinity but much higher capacity than neuronal a2-receptors. These findings, being the first biochemical characterization of a2-receptors on glia, may help explain some of the difficulty in localizing C'2-receptors to pre- or post-synaptic sites in the brain. It also fits well with some known physiology of cultured glial cells, that the inhibition of f-adrenergic-stimulated cAMP accumulation is mediated by ce-adrenoceptors (McCarthy and deVellis, 1978; VanCalker, 1980), in particular U2-adrenoceptors (Evans et al., 1984). Thus glia and neurons in culture may represent a good model system in which to examine Q2-adrenergic receptor regulation.














CHAPTER III
LIGHT MICROSCOPIC AUTORADIOGRAPHY OF [3H]-YOHIMBINE BINDING SITES IN NEURONS IN CULTURE


Introduction

A concern which arose from our studies of [3H]-yohimbine binding sites in neuronal and glial cultures was that neurons in culture might not possess OC2adrenergic receptors. This concern arose because neuronal cultures contain between 10 and 20% glia, and glia in culture contain 10 times as many binding sites for 3H-yohimbine. Thus, glial contamination of neuronal culture could, theoretically, account for the Oc2-binding seen in neuronal cultures.

The best method we could think of, to resolve this issue, was to demonstrate by autoradiography that cells with neuronal morphology possessed Oc2-binding sites. This was complicated by the fact that [3H]-yohimbine is very lipophilic and easily becomes trapped in cells. For this reason it is not possible to perform binding studies directly in the culture dishes, instead of with membranes, because the level of non-specific binding is very high, usually more than 10% of the ligand concentration. However, by reducing the concentration of [3H]-yohimbine and by carefully examining the non-specific binding plates, so that, on examination of total binding plates, patterns of silver grains unique to total binding plates could be recognised, we achieved this goal. Some pictures of immunocytochemically stained cells are included also, to demonstrate the distinctive morphologies of neuronal and glial cell types in culture.







35








36

Methods

Light Microscopic Autoradiograohy of [2H1-Yohimbine Binding in Neuronal
Cultures

Neuronal cultures grown on 35-mm dishes were washed twice with PBS, pH 7.0, and were then incubated with 1 ml PBS containing 1 nM [3H]-yohimbine for 5 mins at 220C. For the non-specific binding dishes, parallel incubations were made using 100 juM yohimbine added to the reaction mixture. Following the incubation, the reaction mixture was removed, dishes were placed on ice and the cells washed twice with ice-cold PBS. The cells were then fixed for 30 mins at 40C with 3.7% glutaraldehyde. After fixation the cells were washed twice with PBS, dehydrated through graded ethanol solutions (30-100%, 5 mins at each concentration) and dessicated at room temperature for 2 hours. In the dark, dishes were coated with Ilford K-5 nuclear emulsion, diluted 1:1 with distilled water containing 2% glycerol, according to Rogers (1979). After 2 mins, excess emulsion was removed, and the dishes were kept at 40C for 20 mins, followed by 60 mins at room temperature. The dishes were dessicated

overnight and then stored in light-tight boxes in the refrigerator until development, usually 3-4 months later. For development, Kodak D-19 (diluted 1:1 with water) was added to each dish for 2 mins. After this dishes were

washed with distilled water and then fixed for 5 mins with sodium thiosulfate. The dishes were finally washed for several minutes with distilled water, and examined by phase-contrast microscopy (Zeiss). Immunofluorescent Staining of Neuronal Cultures

Growth media were removed from neuronal cultures grown on 35-mm dishes, which were then washed three times with PBS, pH 7.4. Following this, cells were fixed and made permeable by the addition of 2 ml of 100% methanol at -200C methanol (100%) to each dish for 7.5 mins. After removal of the

methanol, 2 ml PBS were added to each dish. After 5 mins the PBS was








37

removed and one drop (approx. 60 p1) of either the monoclonal neurofilament antibody NE-14 (1:20 dilution (Shaw et al, 1985) or glial fibrillary acidic protein antibody (GFAP: 1:20 dilution) was added to the center of each dish. The

dishes were incubated at 370C for 30 mins and then the antibody was removed by three washes with PBS (pH 7.4). Next, one drop of a secondary antibody (goat anti-mouse antibody conjugated to fluorescein (FITC)) was added to the same area of each dish as the primary antibody, and the dishes underwent a further incubation of 370C for 30 mins. After this, cells were washed three times with PBS (pH 7.4), and 2 drops of mounting media (90% glycerol with 1 mg/ml paraphenylenediamine in 20 mM Tris HCl pH 8.0) were added to the stained area followed by a coverslip. The cells were viewed using an inverted fluorescence microscope (Zeiss).

Results

Prior studies using antibodies against neuron specific enolase determined that our neuronal cultures contain approximately 70-80% neurons, with the remainder of the cells being glia (Raizada et al, 1983). Considering this fact, and also that glia contain many specific binding sites for [3H]-yohimbine, it was important to determine whether the neuronal cells actually contain [3H]yohimbine binding sites. We approached this problem by using light microscopic autoradiographic analysis of [3H]-yohimbine binding in neuronal cultures.

Fig. 7 contains three representative autoradiograms showing the binding of [3H]-yohimbine to neuronal cultures. In figs. 7a and 7b the arrows indicate the position of silver grains in the emulsion overlying cells with neuronal features. This specific association of silver grains with neuronal cells was observed at the level of cell bodies, neurites and axon terminals. Fig 7c contains an autoradiogram taken from a dish in which unlabelled yohimbine was added to the incubation mixture. No specific localization of silver grains to neuronal







38



















41 "
























Figure 7. Light microscopic autoradiographs showing [3H]-yohimbine binding in neuronal cultures.
A. Autoradiographic analyses of [3H]1-yohimbine binding in neuronal
cultures were performed as detailed in the Methods section.
The arrows indicate high densities of silver grains (shown in black) associated with cells of neuronal morphology. In these cases the silver grains
were associated specifically with the neuronal cell bodies and neurites. Bars
10 A.








39




























Fiue7 ih miscpi au ad-rah shoin [3]yhmie bin gi
AO





.44~







lot 4 10


















neuronal cultures.
B. Autoradiograph of neuronal cultures incubated with [3H]-yohimbine. Arrows indicate high densities of silver grains (shown in black) associated with cells of neuronal morphology. In this case the silver grains were associated specifically with the neuronal cell bodies and neurites. Bars 10 p.







40






























Fiur .Lgh /irsoi uoaigah swng[H-hibnbidign
na utre






Apr





I '14



















Figure 7. Light microscopic autoradiographs showing [3H]-yohimbine binding in neuronal cultures.
C. Autoradiograph of neuronal cultures incubated with [3H]-yohimbine in the presence of 100 pM unlabelled yohimbine. No specific localization of silver grains to cells of neuronal morphology was noticed. Bars 10 p.








41

cells can be observed in this autoradiogram, nor in other dishes treated in the same way.

To confirm that the cells to which silver grains are specifically localized are in fact neurons, immunofluorescent staining of our neuronal cultures was performed using antibodies against neuronal and glial proteins. With the use of the monoclonal neurofilament antibody NE-14, we have obtained specific staining of neuronal cells in our neuronal cultures. Figs. 8a and 8b show phase contrast and fluorescent micrographs, respectively, taken from the same field of neuronal cultures stained with NE-14. The arrow indicates specific immunofluorescent

staining associated with neuronal cells. These cells are similar morphologically to those observed with the specific silver grain localization in figs. 8a and 8b. Not all of the neuronal cells in fig. 8b appear stained because this antibody recognizes only heavy type neurofilaments, which only appear at late stages of neuronal development (Shaw et al., 1985). In our neuronal cultures the neuronal cells actually overlie the glial cells, and this can be clearly observed in figs. 8c and 8d. These figures are representative phase contrast and fluorescence micrographs taken from neuronal cultures stained with anti-GFAP. In the phase contrast picture (8c) the neuronal cells are clearly evident, whereas cells of typical glial morphology are not easily observed. However, in the fluorescence micrograph in fig. 8d, stained glial cells are easily seen. These cells bear no morphological relation to the neuronal cells which specifically localize silver grains representing [3H]-yohimbine binding sites in figs. 7a and 7b.

In addition, immunofluorescent staining analyses using the GFAP and NE14 antibodies, and other antisera to neurofilament proteins and as NF-l and

DA2B1 (Shaw et al., 1985) have enabled us to show that our neuronal cultures in fact contain 85-90% neuronal cells, a greater number than originally estimated (Raizada et al., 1983).








42













k.5.;






















Figure 8. Immunofluorescent staining of neuronal cultures.
Immunofluorescent staining procedures using antibodies to a neurofilament protein (NE-14) antibody) and GFAP were performed as detailed in the Methods section.
A, C are phase-contrast micrographs of cell fields from neuronal culture dishes. Cells and cell-bundles of typical neuronal morphology are easily observed (arrows). Cells with glial morphology are not easily seen, because they underlie the neuronal cells and are thus out of focus in most cases. B is a fluorescence micrograph from the same cell field as in A, showing the NE-14/FITC staining. This staining is clearly associated with the neurofilaments in the cells which have neuronal morphology. D is a fluorescence micrograph from the same cell field as in C, showing the GFAP/FITC staining. This staining is only associated with cells of typical glial morphology, and not with the neuronal cells. The GFAP/FITC- stained astrocytes are mostly not observed in the phase-contrast micrograph C.








43
















































Figure 8. Immunofluorescent staining of neuronal cultures. B. Fluorescent micrograph from the same cell field as in A, showing the NE14/FITC staining. The staining is clearly associated with the neurofilaments in the cells which have neuronal morphology. Bars 10 u.








44



























flAl



















Figure 8. Immunofluorescent staining of neuronal cultures. C. Phase-contrast micrograph of a cell field from a neuronal culture dish. Cells and cell-bundles of typical neuronal morphology are easily observed (arrows). Cells with glial morphology are not easily seen, because they underlie the neuronal cells and are thus out of focus in most cases. Bars 10 p.








45














































Figure 8 Immunofluorescent staining of neuronal cultures. D is a fluorescence micrograph from the same cell field as in C, showing the GFAP/FITC staining. This staining is only associated with cells of typical astrocytic glial morphology, and not with the neuronal cells. The GFAP/FITCstained astrocytes are mostly not observed in the phase-contrast micrograph C. Bars 10 pu.








46

Discussion

Neuronal cells in culture possess Oc2-binding sites, because silver grains localize to cells with neuronal morphology. The grains could be seen associated with cell bodies, fibre tracts and junctions between fibre tracts. At the level

of magnification used it was impossible to determine whether the grains that were associated with cell bodies were on dendrites or not. The silver grains

associated with fibres were often seen in varicosities, which in peripheral sympathetic nerves at least, are sites of synapses at intervals along the axon. This suggests that a similar pattern of synapses occurs centrally.

In conclusion, these data support the suggestive findings of the characterization studies, that there are indeed neuronal OC2-adrenergic receptors. Thus both glia and neurons possess Oc2-adrenergic receptors when in culture.














CHAPTER IV
ADRENERGIC MECHANISMS MEDIATING ANGIOTENSIN II RELEASE.


Introduction

Angiotensin II has been known to have specific actions on the central nervous system since the cross-perfusion studies of Bickerton and Buckley in dogs (1961). It is now well accepted that there is a brain renin-angiotensin system, independent of the peripheral system (e.g. Campbell et al., 1984, Phillips and Stenstrom, 1985, Phillips, 1987). However, the nature of the action of angiotensin II is not well characterized. It has recently been demonstrated that the release of angiotensin II from brain tissue and from brain cell cultures can be stimulated by depolarizing conditions (Meyer and Weyhenmeyer, 1986, Schiavone et al., 1986), suggesting that angiotensin II may be a neurotransmitter. It has also been postulated that angiotensin II acts as a neuromodulator in the brain because of its interactions with neurotransmitters, especially the catecholamines. This study deals with the ability of catecholamines to control the release of angiotensin II from neurons in culture. We were interested in this because of the demonstration in brain cell culture and in vivo and in vitro of close interactions between catecholamines and angiotensin II (e.g. Garcia-Sevilla et al., 1979; Huang et al., 1987; Meldrum et al., 1984; Sumners and Phillips, 1983; Sumners and Raizada, 1986; Schacht, 1984). In neuronal cell cultures, norepinephrine, acting on o1-receptors, modulates the expression of angiotensin II receptors (Sumners and Raizada, 1984; Sumners, Watkins and Raizada, 1986). Additionally, angiotensin II influences both the uptake and metabolism of norepinephrine in neuronal cultures (Sumners and


47








48

Raizada, 1986; Sumners et al, 1987). It was of interest to us to discover whether x2-adrenergic receptors could modulate angiotensin II release. This question was asked because of the clear interactions between the two systems and because m2-receptors modulate the release of a number of different transmitters, including norepinephrine, serotonin, and acetylcholine (Frankhuyzen and Mulder, 1980, Gothert and Huth, 1980, Vizi and Knoll, 1971). Thus, 02adrenergic receptors might interact with angiotensin II in the same fashion.

We demonstrate here that release of angiotensin II from neuronal, but not astrocytic glial, cultures is stimulated by catecholamines. Our data suggest that two mechanisms exist for the catecholaminergic modulation of angiotensin II release, one which involves o2- and the other fl-adrenergic receptors.

Methods

Preparation of Neuronal Cultures

Cultures were prepared as described in Chapter Two. Protocol for Angiotensin II Release Experiments

Ten neuronal or ten glial dishes were used for each experimental group, and the procedure for a control release experiment was as follows. The growth media were aspirated and the cells washed three times with 10 ml phosphate buffered saline (PBS), pH 7.4 per wash. Four ml PBS containing 0.75 mM CaCl2, 0.75 mM MgCl2 and 33 mM glucose (release buffer) were added to each dish and the cells were incubated for 5 min. at 37*C in an incubation bath. After 5 min. the medium was removed and placed in a plastic test tube containing 0.2 ml of concentrated HCl, mixed and stored in the refrigerator. This incubation procedure was repeated for a further three 5-min. periods. For drug treatments the protocol was exactly the same except that during the second five-minute incubation period the cells were exposed to different adrenergic agonists or antagonists.








49

In the first set of experiments, we examined whether yohimbine (50 MM) could alter angiotensin II release from either neuronal or astrocytic glial cultures. This was carried out as above with yohimbine being included in the second incubation period.

In the second set of experiments, we determined the effects of increasing concentrations of yohimbine (0.1, 1.0, 10.0 and 50 M) on angiotensin II release from neuronal cultures.

In the third set of experiments, we examined the effects of increasing concentrations of norepinephrine (0.01, 0.1, 1 and 50 MM) on angiotensin II release from neuronal cultures.

In the fourth set of experiments, the effects of norepinephrine (50 pM) on angiotensin II release in neuronal cultures were examined in the absence and presence of prazosin (0.1 10 MM).

In the fifth series of experiments, we examined the effects of norepinephrine (50 MM) on angiotensin II release in neuronal cultures in the absence or presence of DL-propranolol (1 100 MM).

In the sixth set of experiments, we determined the effects of the fadrenergic agonist, isoproterenol, (1 and 100 MM) on angiotensin II release from neuronal cultures.

Drug Solutions and Incubations

Catecholamines agonists and antagonists were dissolved as follows. Norepinephrine and isoproterenol were dissolved in release buffer containing 10 MM L-ascorbic acid to an initial dilution of 10 mM. These solutions were then diluted to the required concentrations using release buffer containing 10 MM L-ascorbic acid. In experiments where norepinephrine or isoproterenol was used, the control incubations were always performed with release buffer containing 10 MM L-ascorbic acid. Both prazosin and yohimbine were initially








50

dissolved in distilled water to a concentration of 1 mM, and DL-propranolol in release buffer to a concentration of 10 mM. All of these antagonists were

diluted to the required concentrations using release buffer. All drugs were prepared fresh for each experiment.

Preparation of Incubation Media and Cell Samoles for Radioimmunoassav

Prior to the radioimmunological determination of angiotensin II in the incubation media, the fractions obtained from the four incubation steps were purified and concentrated on octadecasilyl-silica cartridges (Sep Pak C18) according to a procedure described earlier (Phillips and Stenstrom, 1985). The purified samples then underwent radioimmunoassay for angiotensin II, as detailed previously (Hermann et al., 1988b, Phillips and Stenstrom, 1985). After the fourth incubation step, angiotensin II was extracted from neuronal and glial cells by the addition of 1 ml of 1 M acetic acid to each dish. Cells were

removed from each dish, combined in a plastic tube and boiled for five min., homogenized ultrasonically for 10 sec and then centrifuged at 2,700 g for ten minutes. The supernatant was lyophilized, dissolved in 0.5 ml of 0.05 M Tris/HC buffer, pH 7.4, spun at 10,000 x g for 2 min. at room temperature and 0.05 ml of the supernatants were subjected to radioimmunological measurement of angiotensin II, as detailed by us previously. HPLC Analysis of Released Material

To confirm the authenticity of the released material as angiotensin II, the following procedure was followed. After direct radioimmunological measurement of angiotensin II in the buffer from the incubation periods, the remainder of the control samples as well as the yohimbine-treated samples from different experiments were pooled separately, frozen and lyophilized. The dry residues were dissolved in 1.0 ml HPLC solvent A (methanol:H20:1 M ammonium acetate buffer, pH 5.4; 30:69:1, vol/vol), centrifuged at 10,000 x g for 2 min at room








51

temperature and subjected to HPLC analysis as detailed previously (Hermann et al., 1988b, Phillips and Stenstrom, 1985.). About 0.9 ml of the reconstituted samples was injected into the HPLC and 1-ml fractions were collected. The

solvent was removed from these fractions in a stream of air at 40*C. The

dried samples were dissolved in 0.5 ml distilled water, frozen and lyophilized. The freeze-dried samples were dissolved in 0.5 ml 0.05 M Tris/HCl buffer, pH 7.4, and 0.1 ml aliquots from each fraction were taken for the radioimmunological measurement of angiotensin II, as detailed previously (Hermann et al. 1988b).

Results are presented as means SEM of the amounts of angiotensin II released, expressed as a percentage of the total angiotensin II content (total angiotensin II = that angiotensin II measured in the cell extracts plus that in the media of each release period).

Results

Effects of Yohimbine on Ang II Release from Cultured Brain Cells

In neuronal cultures the mean baseline secretion of angiotensin II-like material when cells were incubated with buffer alone was 43.65 7.44 pg per 5 min. incubation period (n = 14 expts.; 52 individual determinations). In the first set of experiments, neuronal cultures were treated with yohimbine (50 pM, 5 min.) as detailed in the methods. This treatment resulted in significantly

increased release of angiotensin 11-like material, i.e. 37.87 11.7% (n = 3 expts.) of the total neuronal angiotensin II was released in the second incubation period compared with 5.15 1.32% (n = 3 expts.) in the first (buffer-treated) incubation period. This is illustrated in Figure 9, which shows a representative experiment where neuronal cultures were treated with yohimbine in the second incubation period. It can be seen clearly that addition of yohimbine is accompanied by a large increase in the release of angiotensin I-like material.







52













60
0

0-

W 40




wu20



z
< 0
0 5 10 15 20

INCUBATION TIME (mins)






Figure 9. Representative experiments showing the effects of yohimbine on the release of angiotensin Il-like material from neuronal and astrocytic glial
cultures.
Release of angiotensin II-like material was analyzed during four incubation
periods of five minutes each, as detailed in the Methods. The first, third and fourth incubation periods were with buffer alone, and the second period was with yohimbine (50 uM). (e), neuronal cultures; (0), astrocytic glial cultures.
Angiotensin II release during each period is expressed as a percentage of the
total (cell plus incubation media) angiotensin II content.







53










*
0 60





C 4020


Z

< 0

CON .1 1 10 50

[YOH] uM






Figure 10. Effects of different concentrations of yohimbine on the release of angiotensin-II like material from neuronal cultures.
Release of angiotensin II-like material was analyzed during four incubation periods of five minutes each, as detailed in the Methods. In control (CON)
incubations, cells were exposed to buffer alone for each incubation period. In yohimbine (YOH) -treated cultures, cells were exposed to buffer for periods one, three and four, and to YOH (0.1 50 juM) in period two only. In all cases, the data shown represent the summed angiotensin II released during periods two and three, expressed as a percentage of the total (cell plus incubation media) angiotensin II content. Data are means SEM of four individual experiments (40 total dishes per data point). (*) Significantly different from control release.








54









fmol Ang Il/ fraction (11-Ang 11 % MeOH
101

CONTROL

50
50- 5



00




100
100.

YOHIMBINE

-50




0
01
0 10 20 30 40 50
Time (min)








Figure 11. HPLC analysis of angiotensin II-like material released from neuronal cultures.
Incubation media containing the released angiotensin II-like material were prepared as detailed in the Methods, and injected into the HPLC, which utilized a reverse-phase C18 column and a methanol gradient for separations. Upper panel: Chromatogram obtained from yohimbine (YOH 50 uM)-treated cells. In both cases, a peak is observed at 26-27 mins. retention time, similar to the retention time of authentic Ile5-angiotensin II (arrow). This peak is larger in the yohimbine treated group.








55

After removal of the yohimbine, release of angiotensin II-like material returns to the same levels as in the first period by the fourth (buffer-treated) incubation period. However, it is apparent that a significant amount of

angiotensin II-like material is released during the third incubation period immediately following the yohimbine treatment period. For this reason, we have combined the amounts of angiotensin II released in incubation periods two and three in the later dose-response studies, and the results are presented as such. In astrocytic glial cultures, the basal release of angiotensin II-like material was 21.76 5.7 pg per 5 min. incubation period (n = 8 expts.; 24 individual determinations) when cells were incubated with buffer alone. Treatment of astrocytic glial cultures with yohimbine (50 pM, 5 min.) as detailed in the methods caused no significant changes in the release of angiotensin II-like material from these cells, compared with buffer treatments (n = 5 experiments). This is also illustrated in Fig. 9, which also shows a representative experiment where glial cultures were treated with yohimbine during the second incubation period.

In the second set of experiments, neuronal cultures were incubated with different concentrations of yohimbine (0.1 50 uM, 5 min.), as detailed in the methods. This resulted in a concentration-dependent increase in the release of angiotensin II-like material compared with identical cultures treated with buffer (i.e. controls). From Fig. 10 it can be seen clearly that a significant increase in the release of angiotensin II-like material can be obtained with 1 pM or higher concentrations of yohimbine, and that the effect starts to plateau around 50 pM of the oc2-antagonist.

The released angiotensin I-like material was subjected to HPLC analysis as detailed previously in order to establish its authenticity. Fig. I1 (upper panel) shows a chromatogram obtained from the injection of released material from








56

control (buffer-treated) cells into the HPLC, and a small peak of angiotensin II can be seen at 26-27 min. retention time (note: the arrow indicates the retention time for authentic Ile-angiotensin II). It is also apparent that some other unidentified substances cross-react with the angiotensin II antibody. Fig. I1 (lower panel) shows a chromatogram obtained from the injection of released material from yohimbine-treated (50 pM, 5 min.) cells into the HPLC. Again, a

peak of angiotensin II can be seen at 26-27 min. retention time, the same as for authentic Ile5-angiotensin II (arrow). In addition, it can be seen that the peak is larger, i.e. there is a greater amount of angiotensin II released, in the yohimbine-treated compared with control cells (Fig. 11).



Effects of Catecholamine Agonists on Angiotensin II Release from Neuronal
Cultures

To determine whether oC2-adrenergic blockade by yohimbine was acting directly to release angiotensin II, or indirectly via stimulating norepinephrine release, the third set of experiments was performed. In this set of experiments, neuronal cultures were incubated with varying concentrations of norepinephrine as detailed in the methods, and release of angiotensin II-like material was determined. Initial experiments showed that norepinephrine was able to cause increased release of angiotensin II-like material from neuronal cultures, compared with buffer treatment. As with the yohimbine treatments, most of

the angiotensin II is released during the second incubation period when norepinephrine is present. However, a significant amount of angiotensin II is released in the third incubation period before control levels of release are attained once more. Therefore, in the dose-response studies with

norepinephrine we have combined the amounts of angiotensin II released in incubation periods two and three, and the results are presented as such. Incubation of neuronal cultures with norepinephrine (0.01 50 pM, 5 min.)








57

resulted in concentration-dependent increases in the release of angiotensin IIlike material (see Fig. 12) compared with identical cultures treated with buffer (i.e. controls). From Fig. 12 it can be seen that a significant increase in

angiotensin II release is obtained with as little as 10 nM norepinephrine, and that a maximal effect is reached between 100 nM and 1 uM norepinephrine. At

the higher concentration of 50 puM norepinephrine, the response is less, but is not significantly different from that obtained with 1 pM norepinephrine.

In order to determine which type of adrenergic receptor was being activated by norepinephrine to stimulate release of angiotensin I-like material, the fourth and fifth sets of experiments were performed. In the fourth set of

experiments, neuronal cultures were incubated with norepinephrine (50 pM, 5 min.) in the absence or presence of the cr-antagonist prazosin (1 or 10 MM, 5 min.), as detailed in the methods, and release of angiotensin II-like material was analyzed. The results in Fig. 13 clearly show that prazosin does not inhibit the norepinephrine-stimulated increase in angiotensin II release from neuronal cultures, either at 1 or 10 pM. The concentrations of prazosin used here are

sufficient to cause complete blockade of ocl-adrenergic receptors and to completely inhibit norepinephrine-induced inositol phospholipid hydrolysis in neuronal cultures (Feldstein et al., 1986, Gonzales et al., 1987). However, these experiments can not be considered to be completely conclusive, as prazosin may not have been able to completely block the effects of this concentration of norepinephrine.

In the fifth set of experiments, neuronal cultures were incubated with norepinephrine (50 pM, 5 min.) in the absence or presence of the general 8antagonist DL-propranolol (1-100 pM, 5 min.). However, DL-propranolol exhibits significant cross-reactivity with our angiotensin II antibody, and so the results







58













'060



0X 4-



2j 0


20
w



0/
CON 0.01 0.1 1 50

[NEI uM





Figure 12. Effects of different concentrations of norepinephrine on the release of angiotensin II-like material from neuronal cultures.
Release of angiotensin II-like material was analyzed during four incubation periods of five minutes each, as detailed in the Methods. In control (CON) incubations, cells were exposed to buffer alone for each incubation period. In norepinephrine (NE)-treated cultures, cells were exposed to buffer for periods one, three and four, and to NE (0.01 50 puM) in period two only. In all cases, the data shown represent the summed angiotensin II released during periods two and three, expressed as a percentage of the total (cell plus incubation media) content of angiotensin II. Data are means SEM of four individual experiments (40 total dishes per data point). (*) significantly different from control values.








59






80


|
0
60

w *
0X)


<40





~20
20




0
CON A B C




Figure 13. Effects of prazosin on norepinephrine-stimulated increases in the release of angiotensin II-like material from neuronal cultures.
Release of angiotensin II-like material was analyzed during four incubation periods of five minutes each, as detailed in the Methods. In control (CON) incubation periods, cells were exposed to buffer alone for each incubation period. In drug-treated cultures, cells were exposed to buffer for periods one, three and four, and to norepinephrine (NE), or norepinephrine plus prazosin, in period two only. A, Norepinephrine (50 pM); B, NE (50 pM) plus prazosin (1 pM), C, NE (50 pM) plus prazosin (10 pM). In all cases, the data shown
represent the summed angiotensin II released during periods two and three, expressed as a percentage of the total (cell plus incubation media) angiotensin II content. Data are means SEM of three individual experiments (30 total dishes per data point). (*) Significantly different from control values.








60











~30
0






LU)




1--1

W0



Z
< 0 1
CON / 1 100

[ISOPI uM




Figure 14. Effects of isoproterenol on the release of angiotensin II-like material from neuronal cultures.
Release of angiotensin II-like material was analyzed during four incubation periods of five minutes each, as detailed in the Methods. In control (CON) incubations, cells were exposed to buffer alone for each incubation period. In isoproterenol (ISOP)-treated cultures, cells were exposed to buffer for periods one, three and four, and to ISOP (1 and 100 pM) in period two only. In all
cases the data shown represent the summed angiotensin II released during incubation periods two and three, expressed as a percentage of the total (cell plus incubation media) angiotensin II content. Data are means SEM of two to four experiments (20 40 dishes per data point). (*) Significantly different from control release.








61

from this series of studies were meaningless. In an attempt to circumvent this problem, we performed the sixth set of experiments, in which the effects of a pure fl-adrenergic agonist, isoproterenol, were tested on angiotensin II release. Incubation of neuronal cultures with isoproterenol (1 or 100 pUM, 5 min.) resulted in increases in the release of angiotensin 11-like material (see Figure 14) compared with identical cultures treated with buffer (controls). From Figure 14 it can be seen that 100 uM isoproterenol induces an increase of -17% in the release of angiotensin II-like material. This dose of isoproterenol elicits a large increase in #-receptor-mediated cyclic AMP production in neuronal cultures (Baker et al., 1986).

Discussion

In the present study we have determined that catecholaminergic systems can influence the release of angiotensin II-like material from neuronal, but not from astrocytic glial cultures. Firstly, we have shown that release of angiotensin II-like material from neuronal cultures is stimulated by blockade of m2-adrenergic receptors with yohimbine. This effect of yohimbine is quick,

starting within a five-minute incubation period, and is also dose-dependent. The discovery that OC2-adrenergic receptors are able to modulate the release of neuronal Ang II-like material is not an unreasonable finding, because these receptors have previously been shown to modulate the release of a number of different neuroactive substances, e.g. norepinephrine, serotonin and acetylcholine (Frankhuyzen and Mulder, 1980, Gothert and Huth, 1980 and Vizi and Knoll, 1971). In addition, there is good anatomical evidence for the co-localization of m2-adrenergic receptors with angiotensin I-immunoreactive material in the central nervous system (Boyajian et al, 1987, Unnerstall et al., 1984). Thus, our studies suggest that c2-adrenergic receptors are able to modulate release of neuronal angiotensin II-like material. This modulation by OC2-adrenergic








62

receptors appears to be unique to neurons because no effects of yohimbine were observed in astrocytic glia. This is perhaps surprising because cultured astrocytes contain angiotensin II-like material (Hermann et al., 1988b) and also large numbers of specific oc2-adrenergic receptors (Richards et al., 1986).

In an effort to determine whether the release of angiotensin II following stimulation of m2-adrenergic receptors was direct or indirect, the effects of catecholamines and catecholamine agonists upon the release of neuronal angiotensin II-like material were determined. As stated previously, &2adrenergic receptors modulate release of neuronal norepinephrine, and the m2blocker yohimbine should be expected to increase norepinephrine release (Langer, 1981). Therefore, perhaps the effects of yohimbine on the release of angiotensin II-like material are mediated via release of norepinephrine, and this is a reasonable suggestion because our cultures contain (Sumners et al., 1983a), and are able to release, catecholamines. For example, in preliminary experiments we have shown that in neuronal cultures preloaded with L-[3H]norepinephrine (0.2 MM, 60 min, 37*C) in the presence of 100 pM pargyline, A23187 (1 uM) stimulates release of 22.4 5.2% (n = 9 dishes) of the total [3H]-norepinephrine, compared with 2.7 1.9% (n = 9 dishes) of the total [3H]-norepinephrine in the absence of this ionophore (Sumners, unpublished observations). We have determined that incubation of neuronal cultures with norepinephrine induces a dose-dependent increase in the release of angiotensin I-like material, an effect not inhibited by the ocl-adrenergic antagonist, prazosin. Additionally, the pure 6-adrenergic antagonist,

isoproterenol, induced small but significant increases in the release of neuronal angiotensin I-like material. Again, no release of angiotensin II occurred from astrocytic glial cultures (data not shown). These results might suggest that yohimbine induces increased release of norepinephrine, which then acts at fl-








63

adrenergic receptors to increase release of angiotensin I-like material. However, we were not able to confirm this mechanism with the use of fadrenergic blockers because these drugs interfere with our radioimmunoassay for angiotensin II-like material. Also, if this were the mechanism for catecholamine regulation of angiotensin II release, it might be expected that a fl-adrenergic agonist, such as isoproterenol, would be a far more potent stimulator of angiotensin II release than norepinephrine. This is not the case, as observed from our results. Therefore, other possibilities exist for the regulation of release of angiotensin II-like material from neurons. One is the existence of distinct populations of neurons, one which releases angiotensin II in response to o2-antagonism and another which releases angiotensin II following #-adrenergic stimulation. This may be a more likely explanation as the different treatments (i.e., with isoproterenol or norepinephrine) did not induce the same level of release of angiotensin II, although it may just be difficult to compare doses of the different adrenergic drugs.

Although we have clearly shown that catecholamine systems are able to regulate the release of angiotensin I-like material from cultured neurons, other studies on the effects of catecholamines on angiotensin II release have given varying results. For example, neither isoproterenol nor DL-propranolol alter the release of angiotensin II-like material from the anterior hypothalami of rats implanted with push-pull cannulae (Brosnihan et al., 1988). However, the doses of isoproterenol used were lower than in the present study, and our culture preparation is not restricted to the anterior hypothalamus. These data would, in fact, support our above suggestion that selected populations of neurons respond to 6-adrenergic stimulation by releasing angiotensin II. Other investigations have shown that Ang II is released from vascular (subendothelial)








64

tissue by f-adrenergic stimulation (Gothert and Kollecker, 1986, Nakamura et al, 1986a), data which support our current findings.

Questions remain as to the physiologic role of regulation of neuronal angiotensin II release by catecholamine systems. In vascular tissue the angiotensin II released by fi-adrenergic stimulation facilitated neurotransmission at the sympathetic nerve terminal (Nakamura et al, 1986b). The ability of

angiotensin II to facilitate sympathetic neurotransmission is well established (e.g. Langer, 1981, Starke et al., 1970, Story and Ziogas, 1987, Zimmerman and Whitmore, 1967). Our current findings suggest that a similar mechanism may exist in the central nervous system to enhance norepinephrine neurotransmission. This suggestion is supported by the fact that angiotensin II facilitates both the electrically evoked and potassium-induced release of norepinephrine in the brain (Garcia-Sevilla et al., 1979, Huang et al., 1987, Meldrum et al., 1984, Schacht, 1984) and that the angiotensinergic pathway linking the subfornical organ to the nucleus medianus (Lind et al., 1984, Lind et al., 1985, Weyhenmeyer and Phillips, 1982) responds to the application of norepinephrine with excitation (Graham et al., 1985). A possible explanation is that norepinephrine increases the release of angiotensin II, which in turn leads to neurophysiological excitation, angiotensin II having been shown to stimulate these cells directly (Phillips and Felix, 1978).

It is also interesting to speculate on how (or whether) adrenergic regulation of release of angiotensin II-like material from neurons is related to release stimulated by depolarization. It has been reported that potassiuminduced depolarizations induce release of angiotensin II-like material from hypothalamus/neurohypophyseal slices (Schiavone et al., 1986) and from mixed (neurons plus glia) brain cell cultures (Meyer and Weyhenmeyer, 1986). Perhaps








65

depolarization-induced release of angiotensin II-like material is dependent upon prior release of norepinephrine.

Overall the current findings strengthen our growing conviction that there are extremely close ties between the angiotensin II and catecholamine systems in the brain as modeled in neuronal cell cultures. Incubation of neuronal cultures with Ang II increases the norepinephrine content of both cells and media (Sumners et al., 1983b). It has a biphasic effect on norepinephrine

uptake, enhancing after short exposure and inhibiting after longer exposure the neuronal uptake of norepinephrine (Sumners and Raizada, 1986). Norepinephrine

applied to the cells causes an oc1-adrenergic-mediated down-regulation of angiotensin II receptors (Sumners, Watkins and Raizada, 1986). This

downregulation could be secondary to the increased angiotensin II release demonstrated in this study. The data suggest that norepinephrine-induced release of angiotensin II is a requisite for a feedback system in neurons by which angiotensin II enhances the neuronal effects of norepinephrine, which in turn through m2- and fi-adrenergic receptors increases angiotensin II actions.

In summary, our results show that the release of angiotensin II-like material from neuronal cultures is modulated by catecholamines, mediated by either m2- or fi-adrenergic receptors or both. The exact mechanisms involved, topography of the different catecholamine-induced effects, and relation to depolarization-induced release have yet to be determined.














CHAPTER V
EFFECT OF VARIOUS ADRENERGIC DRUGS ON THE RENIN
CONCENTRATION OF BRAIN CELLS IN CULTURE


Introduction

We have demonstrated that adrenergic drugs could influence the release of angiotensin II from neurons but not glia in culture. We suspected that the ability of these drugs to release angiotensin II might induce increased synthetic demands, which would be revealed by increased renin concentrations in these cells. Renin is the enzyme which is responsible for cleaving angiotensin I from angiotensinogen. It is the rate limiting step in the formation of angiotensin II in the blood, and therefore, we felt, might be tightly regulated in cultured cells. We expected that those drugs which increased the release of angiotensin II from neuronal cultures would also increase the concentration of renin in these cultures.

Methods

Sample Preparation

Renin, the enzyme which converts angiotensinogen to angiotensin I is measured by adding excess angiotensinogen (substrate) to the tissue preparation being studied, allowing the reaction to proceed to completion and then measuring the angiotensin I generated by RIA.

The experiment was initiated by incubating the cells with catecholamine solutions, (0.1 to 50 MM yohimbine, 0.1 pM to ImM norepinephrine, 10 to 50 pM clonidine, or with vehicle (0.1% ascorbic acid for 4 or 24 h.)). Cells were washed three times with PBS (pH 7.2) and were scraped from the dish in I ml PBS with a rubber policeman. Material from five 100 mm dishes were 66








67

pooled for one analysis. Glial, neuronal and mixed cultures were used in these studies. Mixed cultures are cultures which were not treated with cytosine arabinoside, and therefore contain about 50% neurons and 50% glia. The cells were centrifuged at 2000 g and 40C for 5 mins., the supernatant discarded and the pellet stored at -200C until analysis. For analysis, cells were redissolved in 1 ml of ice-cold 0.9% saline containing 0.1 % Triton X-100 and homogenized by polytron (setting 10, 10s). The homogenate was centrifuged at 30,000 g and 40C for 15 min and the supernatant collected and stored at -200C until use. The pellet was dissolved in 1 ml 1.0 N NaOH for the measurement of protein by the method of Lowry et al., (1951).

Renin Analysis

Four hundred pl of cell extract were reacted with 200 p1 of nephrectomised sheep plasma (courtesy Dr. C.E. Wood), 1200 pl of NaPO4 buffer, pH 6.5 and 100 pi of an inhibitor cocktail (95 mg PMSF in 5 ml of 95% ethanol, plus 5 pl #-mercaptoethanol) for 0-12 h at 370C in a shaking bath. Blank

incubations were performed using 400 pl 0.9% saline with 0.1% Triton X-100 instead of cell extract.

At 0, 1, 2, 4, 6, and 12 h. a 200 pl aliquot was removed from each blank and sample tube and added to 200 pl of Tris buffer, pH 7.0. The tubes were placed in a boiling water bath for 5 mins. and the resulting precipitate pelleted by centrifugation at 3500 g for 15 min. The supernatants were stored at -200C until used in a radioimmunoassay.

A specific anti-angiotensin I antibody from our lab was used. This

antibody had less than 0.01% cross-reactivity with Ang II and less than 0.001% with angiotensin III. The assay sensitivity is 3 pg angiotensin I per sample with a final dilution of antibody to 1:500,000. Angiotensin I levels are expressed as pg/mg cell protein generated per hour.








68




Table 5.1 The effects of adrenergic drugs on renin concentrations in brain cell culture.

TREATMENT NEURONAL GLIAL MIXED
pg AI/mg prot./h pg Al/mg prot./h pg Al/mg prot./h YOHIMBINE

control 0.529 (2) 0.549 (2) 1.483 0.3 (4)

0.1 pM 4h 0.653 (2) 0.861 (1) 1.871 (2)

1.0 pM 4h 0.586 (2) 1.368 (1) 1.469 0.3 (4)

10.0 pM 4h 0.773 (2) 0.597 (2) 1.412 0.2 (4)

50.0 pM 4h ND 0.672 (2) 2.05 (2)

100.0 pM 4h ND 0.546 (1) ND

50.0 pM 24h ND 0.555 (1) ND

NOREPINEPHRINE

control ND 1.926 (1) 0.406 (1)

1.0 pM 4h ND 1.037 (1) 0.320 (1)

10.0 pM 4h ND 0.684 (1) 0.361 (1)

100.0 pM 4h ND 1.376 (1) ND

1.0 mM 4h ND 1.547 (1) ND

CLONIDINE

control 0.524 (1) ND ND

10.0 pM 4h 0.377 (1) ND ND

50.0 pM 4h 0.616 (1) ND ND

ND = not determined

number in brackets =number of times the determination was measured.








69

Results

Renin was present in all cultures examined. The control levels were 0.527

0.03 pg angiotensin I /mg protein/h, (n = 3) for neurons; 1.008 0.46 pg angiotensin I/mg protein/h, (n = 3) for glia and 1.267 0.32 pg angiotensin I/mg protein/h, (n = 5) for mixed cultures. However neither yohimbine, norepinephrine, nor clonidine had any effect on these levels when the cells were treated for 4 or 24 h. (see table 5.1).

Discussion

Yohimbine, which increases the release of angiotensin II from neuronal cells was unable to induce any change in the concentration of renin in these cells. Similarly, there was no change in either mixed or glial cells by this treatment. These cell types were also tested for changes in renin concentration because the exact cellular location of renin has not been established, and we were anxious not to miss an effect if there was one. The treatment times were longer than those used to induce release of angiotensin II from cultures because we wanted to allow enough time for protein synthesis to occur, assuming renin synthesis would be stimulated. However, it may be possible that the effect was overlooked by these time courses.

Several interesting questions are raised by these findings. Thus, although yohimbine induces an increase in the release of angiotensin II from neurons, it does not stimulate the cells to synthesize more renin. This means either that renin is not limiting in these cells, so that they can easily increase synthesis of angiotensin II without additional renin, or that there is an effect which we were unable to see with current techniques. It seems probable that the cells

would be synthesizing more angiotensin II, because the release of angiotensin II is high. It is usually thought, at least for neurotransmitters, that approximately 10% of the total content of neurotransmitter can be released readily. As we








70

see higher levels of angiotensin II release than this, we expect that synthesis of angiotensin II is occurring. This synthesis is apparently not dependent upon increased renin levels.

It is possible that if renin is not limiting in these cells, then another factor may be. It could be that synthesis of angiotensinogen is a more important regulator in neuronal cultures, something that remains to be studied.

Thus, in summary, although renin concentration was not altered in cultured brain cells by drugs known to enhance the release of angiotensin II from these cells, important information about the regulation of the synthesis of angiotensin can be inferred from these negative findings.














CHAPTER VI
REGULATION OF Oc2-ADRENERGIC RECEPTORS BY PEPTIDES Introduction

Angiotensin II has many interactions with catecholamines which are fully described in Chapter One. Insulin, another peptide hormone also has interactions with catecholamines. Recently, it has become clear that insulin has actions in the brain, where the classical view was that the brain was independent of insulin. However, the insulin which acts in the brain is probably synthesized in the brain, a possibility that helps to reconcile the classical and new theories.

Previous studies have shown that insulin and its receptors are present on neurons (Boyd and Raizada, 1983, Boyd et al., 1985) and astrocytic glial cells (Clarke et al., 1984) in culture, as well as in membranes prepared from the brain of the rat (Havrankova et al., 1978a, Havrankova et al., 1978b, Havrankova et al., 1983). Insulin can be released by depolarization (Clarke et al., 1986) and causes changes in the rate of firing of nigrostriatal dopaminergic neurons (Saller and Chiodo, 1980) and hippocampal neurons (Palovcik et al., 1984). It also affects the dopaminergic system of the olfactory bulbs (Barbaccia et al., 1982), and at high doses stimulates release of dopamine, norepinephrine and epinephrine from hypothalamic slices (Sauter et al., 1983).

Insulin inhibits total bioamine uptake into neuronal cultures (Boyd et al., 1985). Ninety-five percent of specific uptake of norepinephrine into neurons can be inhibited by insulin. The specific inhibitor of norepinephrine uptake, maprotiline, inhibits high affinity binding of insulin at concentrations used to



71








72

inhibit norepinephrine uptake, suggesting that maprotiline and insulin may act at the same receptors (Boyd et al., 1986).

Insulin and angiotensin II have interactions with catecholamines. However, these interactions are not always the same. In the case of uptake of norepinephrine, the actions of insulin and angiotensin II are opposite. Insulin

inhibits uptake of norepinephrine into neurons (Boyd et al, 1986), whereas angiotensin II stimulates uptake, at least initially (Sumners and Raizada, 1986). But in the case of release of catecholamine their actions are similar. Angiotensin II increases the release of norepinephrine (Schacht, 1984, GarciaSevilla et al., 1979, Meldrum et al., 1984, Chevillard et al., 1979) as does insulin (Sauter at al., 1983).

The actions of these two peptides appear to be the modulation of the amount of catecholamines present in the synaptic cleft, or in other words the modulation of noradrenergic transmission. Since Oc2-adrenergic receptors also strongly influence the level of noradrenergic transmission, we became interested in whether insulin and angiotensin II could interact with x2-receptors, and if they could, were the interactions of angiotensin II and insulin with m2-receptors the same?

Methods

Glial and neuronal cultures were prepared as described previously in Chapter Two. Insulin was dissolved in a small volume of 0.1 N NaOH and

diluted to the correct concentration in buffer. It was then sterilized by filtration and added to the medium of the cells. For the time-course experiments 167 nM insulin was added to the cells, which were removed from the dishes 24 h. later, as previously described in Chapter Two. For 48 h or

longer incubations, the cells received insulin every 24 h. to maintain its concentration. Control cells received sterile vehicle instead of insulin and a








73

control group and an insulin-treated group were always assayed together at each time measurements were made as we have previously determined that each group of glial cells demonstrates a slightly different level of [3H]-yohimbine binding. For concentration curves we used 24 h. time intervals and 16.7, 167, and 1670 nM insulin. Scatchard analysis was achieved by binding insulin-treated (1670 nM for 48 h.) and control membranes with 1.0 to 24 nM [3H]-yohimbine and then analyzing the saturation data by the method of Scatchard.

Angiotensin II was added to either neuronal or glial cultures at 10 nM, 100 nM, 1 uM or 10 pM for either 1, 2, 4, 24, or 4 and 24 h. Cells were then removed from the dishes as described in Chapter Two and frozen until assay. The Oc2-adrenoceptor binding assay was performed exactly as described in Chapter Two except that 100 pM norepinephrine was used as the cold displacer instead of 100 pM yohimbine in the norepinephrine competition curves.

Cells from more than one batch were usually pooled to obtain enough protein for these experiments, and if this were necessary, then each batch was divided so that each treatment group had equal numbers of plates from each batch. From experiments in which norepinephrine competition curves were generated, it was possible to look for changes in total binding at one concentration of ligand, and changes in the ability of norepinephrine to displace the binding. However, changes in total binding might have been missed by examination of binding at only one concentration of ligand. Therefore saturation experiments followed by Scatchard analysis were also performed. It was important to examine whether angiotensin II altered the ability of norepinephrine to displace the binding of [3H]-yohimbine because the m2adrenoceptor can exist in one of two affinity states. The affinity state of the receptor is regulated by the presence or absence of GTP. Yohimbine, an

antagonist, binds to both affinity states of the x2-receptor equally.








74

Norepinephrine, an agonist, binds only to the high affinity state of the receptor. Thus, changes in affinity state would be missed if binding of [3H]yohimbine was the only criterion measured. Changes in the ability of norepinephrine to displace the bound [3H]-yohimbine would be indicative of changes in the affinity of the receptor.

Results

Effects of Insulin on Oc2-Adrenerpic Binding

Insulin at a concentration of 167 nM for 24, 48 or 72 h. causes a decrease in [3H]-yohimbine binding. The decreases, expressed as a percentage of control, were 73 8%, (n = 3), 66 9, (n = 3) and 60 8% (n = 1) at 24, 48 and 72 hours respectively (fig 15). The dose-dependency of the decrease is shown in fig 16. After 24 h of treatment with insulin at 16.7 nM the binding was 88 7% of control, at 167 nM, 85 3%, and at 1.67 gM, 82 5%. After 48 hours of treatment with 167 nM insulin, the binding was further decreased to 68 7%. There was quite high variability in the decreases in the binding of [3H]yohimbine between different groups of glia, but decreases were always observed.

Saturation experiments followed by Scatchard analyses revealed that insulin (1670 nM for 48 hours) induced a 52% decrease in the maximal number of [3H]yohimbine binding sites compared with control cells. In addition, treatment with insulin is associated with a slight decrease in the KD of the binding site (table 6.1). The decrease in binding is not the result of a competition by insulin for [3H]-yohimbine binding sites, because even at the highest dose of insulin used in the time and competition curves there was no displacement of [3H]-yohimbine binding when insulin was added to the binding buffer.








75










1004

z
Z5
z 90

M
-j
0
I 80
z 0 0
-70


00
Z 60
LU

~50



24 48 72


TIME / HOURS
















Figure 15. Time course for the decrease in [3H]-yohimbine binding caused by insulin. 167 nM insulin was added to the cells and binding performed at the indicated times after treatment. Each point represents the mean SEM of
three experiments in which six and six non-specific binding tubes were used per data point, except the 72-h time point, which is of one experiment.








76










1004


z

go
z

0
S80
z 0
LL 70
0
w 0
i- 60
z
w

a. 50



16.7 167 1670


DOSE INSULIN / nM

















Figure 16. Dose response curve for the decrease in [3H]-yohimbine binding following treatment of cells for 24 hours with insulin.
Each point represents the mean SEM of six total and six non-specific binding tubes at each point.








77

Table 6.1. Effects of insulin on the number and affinity of f2H1-Yohimbine binding sites in glial cultures

KD Bmax
(nM) (pmoles/mg protein)

Control 21.3 2.83

Insulin 15.1 1.35
(1670 nM for 48 hours)

Data represent mean values from two independent experiments in which a control and an insulin-treated group were assayed concurrently. Effects of Angiotensin II on _C2-Binding

In glial cultures angiotensin II had no effect on total binding of [3H]yohimbine. Nor did it have any effect on the ability of norepinephrine to displace the binding, suggesting that angiotensin does not change the affinity of the receptor for agonists (n = 5 experiments).

In neuronal cultures no significant changes in total [3H]-yohimbine binding were observed with any dose or time of treatment with angiotensin II (n=28 experiments). In two separate experiments neuronal cells were treated with angiotensin II, then saturation experiments, followed by Scatchard analysis, performed with the membranes made from these cells. In the control cells the

KD was 16.95 nM (n=2) and the Bmax was 187 fmoles/mg protein (n=2) and in the angiotensin II-treated membranes (1 jpM 4 and 24 h) the KD was 17.8 nM (n=2) and the Bmax was 175 fmoles/mg protein (n=2). Thus no differences in total binding were seen by this alternate approach. These data are summarized in table 6.2.

Angiotensin II was added to the binding tubes at several different concentrations to see whether it could displace binding. There was no significant difference between binding in the presence or absence of any dose of angiotensin II tested, therefore angiotensin II does not appear to bind to the








78

site to which [3H]-yohimbine binds. Thus, neither treatment with angiotensin II nor the presence of angiotensin II affects total [3H]-yohimbine binding.



Table 6.2. The effect of angiotensin II on the number and affinity of IaHlvohimbine binding sites in neuronal cultures

KD Bmax
(nM) (pmoles/mg protein)

Control 16.9 0.187

Angiotensin II 17.8 0.175
(1 jpM 4 and 24 hours)


Data represent mean values from two independent experiments in which a control and an angiotensin II-treated group were assayed concurrently.


There was no significant effect of angiotensin II on the ability of norepinephrine to displace [3H]-yohimbine binding, at any dose or time period tested. However angiotensin II treatment at 1 uM for 4 and 4 and 24 h decreased the Hill slope from 0.55 to 0.42. The R values for this fit of the Hill curve were 0.94 for the control and 0.94 for the angiotensin I-treated cells. The IC50 values were 1.43 pM for control and 0.97 pM for the angiotensintreated cells. These data are combined from seven experiments, and were analyzed with a computer program specifically designed to analyze data from competition curves. A similar effect on the IC50 values were seen after treatment with 1 MM angiotensin II for 1 h, n=3 experiments, R=0.97 in both cases. However, after treatment with 1 pM angiotensin II for 2 h, the Hill slope was changed from 0.67 (control) to 0.78, with fits of R=0.93, and the IC50 from 1.8 to 2.6 pM (n=4) in control and treated cells respectively. No doseresponsiveness to the changes could be discerned. Thus, any changes which may occur are not consistent, nor always in the same direction. Possibly the variability associated with neuronal Oc2-binding masks these changes.








79

Discussion

As mentioned in the introduction, insulin has many interactions with monoamines. One such interaction, that of inhibition of norepinephrine uptake, would tend to increase the amount of norepinephrine present in synapses. As a population of Oc2-adrenergic receptors regulates the amount of norepinephrine in the synaptic cleft by affecting synthesis and release of norepinephrine, it might be expected that insulin and cx2-adrenergic receptors would act in opposition in the control of this function. For example, insulin would increase

norepinephrine in the cleft, which would feed back to increase m2adrenoceptors, which would in turn tend to return norepinephrine levels to normal. The present results do not support this hypothesis, as insulin downregulated cc2-adrenergic receptors.

Several possible explanations of this phenomenon exist. Firstly, insulin

may increase synaptic levels of norepinephrine not only by decreasing uptake but also by decreasing the level of "inhibitory" autoreceptors, suggesting that insulin is subserving an important physiological pathway which has redundant systems. In other words, it seems that insulin can increase the amount of norepinephrine in the synaptic cleft by two mechanisms, inhibition of uptake and increased release via decreased Oc2-adrenergic receptors. This suggests that the neuromodulatory role of insulin in the control of norepinephrine is very important physiologically. Secondly, insulin may act at a post-Oc2-adrenergic receptor site to increase the efficacy of 02-adrenergic stimulation which, while maintaining homeostasis, would tend to decrease c2-receptor binding. Thirdly,

the Oc2-adrenergic receptor we have examined is on glial cells in culture and may not have the same function as the neuronal Oc2-adrenergic receptor.

Angiotensin II has no significant effects on m2-adrenoceptor binding. It does not affect total high or low affinity binding. It is possible that an effect of








80

angiotensin II on Oc2-adrenergic binding is present, but that the variablilty in the binding masks this effect. The variability arises from several causes. Firstly, the level of specific binding in neurons is only 30-40% and secondly the number of counts bound is low. Thirdly, the specific activity of the ligand,

[3H]-yohimbine is 70-90 Ci/mmol, which is about 20 fold lower than some iodinated ligands (for example [1251]-angiotensin II), which also leads to errors in the measurement of binding. Thus, it is possible that with a better ligand, a more consistent effect of angiotensin II on m2-receptor binding would be observed.

This inability of angiotensin II to cause consistent effects on m2-receptors was somewhat surprising as we expected angiotensin II, with its well-known ability to enhance the stimulated release of norepinephrine from brain tissue to also enjoy some interaction with m2-receptors.. It suggests that the actions of angiotensin II actions do not share a common pathway with m2-adrenoceptors, and that the enhancement of release itself does not affect c2-adrenoceptors. We have shown that drugs which affect synaptic levels of norepinephrine do alter x2-binding; for example clonidine, which decreases norepinephrine release, decreased cx2-binding (in both neurons (n=2) and glia (n=1)). Norepinephrine increased the binding as did desmethylimipramine, a norepinephrine uptake inhibitor (n=2), both of which increase synaptic norepinephrine levels. Therefore, we expected angiotensin II to change Oc2-binding. However angiotensin II receptors are modulated by cxl-adrenergic receptors in a negative manner, in other words when norepinephrine levels increase angiotensin receptor binding goes down, (Sumners, Watkins and Raizada, 1986b), and we suggest that this may be a mechanism by which the norepinephrine-angiotensin II interaction is regulated.








81

Although both insulin and angiotensin II have strong interactions with catecholamines, only insulin strongly influences the Oc2-adrenoceptor. So while both peptides induced increased release of catecholamines from brain tissue, and exhibited other interactions with catecholamines as well, it was not possible to generalize about their effects on m2-receptors.














CHAPTER VII
GENERAL CONCLUSIONS

We have been able to demonstrate the presence of Oc2-adrenergic receptors on both neurons and glia in cell culture. These cultures were derived from one-day-old rat brains. The neuronal cultures are 85-90% neuronal cells, the remainder being astrocytic glial cells. The glial cultures contain more than 95% astrocytic glia, in various forms from polygonal to process-bearing. The remainder are primarily oligodendrocytes. There is very little evidence of fibroblasts in the cultures. We have demonstrated that in neuronal culture, the neurons reach high stages of development, because some of the neurons stain with antibodies against the heavy type of neurofilament, which only appears during later stages of neuronal development (Chapter Three). The light and medium types of neurofilament were also present in the neuronal cultures (data not shown).

Using [3H]-yohimbine as the ligand, we have been able to characterize m2adrenoceptors on both cell types. The neuronal and glial cx2-adrenoceptor have similar affinity, but the glial cultures have a much greater capacity, approximately ten times as many binding sites as neurons. The receptors have the pharmacological characteristics of Oc2-adrenoceptors as OC2-selective drugs are the most effective displacers of the binding. Neither dopamine antagonists nor uptake blockers, such as desmethylimipramine, bind with high affinity to the site in either neuronal or glial culture, reassuring us that were indeed examining c2-adrenoreceptors.

These data may help explain why Oc2-adrenoceptor binding can still be demonstrated following lesion of catecholaminergic neurons (U'Pritchard et al., 82








83

1979), whereon, in the classical view, pre-synaptic Oc2-receptors reside. The residual binding may represent glial binding, or it may represent binding to m2adrenoceptors mediating the release of other types of neuroactive substances. It has been thought for some time that Oc2-adrenoceptors control the release not only of norepinephrine (through inhibitory autoreceptors), but also the release of serotonin and acetylcholine (Frankhuyzen and Mulder, 1980; Gothert and Huth, 1980; Vizi, 1972).

We have been interested for some time in understanding how the release of the neuropeptide, angiotensin II, was controlled. Phylogenetically,

angiotensin II has a very old history of interactions with catecholamines (Carroll and Opdyke, 1982), which occur centrally as well as peripherally. Therefore, we postulated that Oc2-adrenoceptors located presynaptically on angiotensin II neurons might control the release of angiotensin II. To this end, we established that blockade of c2-adrenoceptors in neuronal culture induced a dose-responsive increase in the release of angiotensin II, analogous to the increased overflow of norepinephrine seen following blockade of inhibitory autoreceptors (presynaptic m2-adrenergic receptors on norepinephrine containing neurons). Glial cultures did not exhibit angiotensin II release in response to m2-adrenoceptor blockade, although they have many Oc2-adrenoceptors and contain angiotensin II (Hermann et al., 1988b).

We postulated that increased release of norepinephrine, induced by blockade of Oc2-adrenergic receptors, might have induced this response in neuronal cultures. Indeed, we saw increased release of angiotensin II from neuronal cultures when norepinephrine was applied to the cultures. The release of angiotensin II increased with increasing doses of norepinephrine until a dose of 50 pM norepinephrine, was administered. At this dose, the release started to decline, possibly due to interaction with Oc2-adrenergic receptors. The release








84

of angiotensin II induced by norepinephrine could not be blocked by prazosin, a selective ocI-adrenergic antagonist, which suggested that this response was mediated by fl-adrenergic receptors. Following unsuccessful attempts to use #adrenergic antagonists to investigate this possibility, predominantly because of high cross-reactivity with the anti-angiotensin II antibody, we were able to demonstrate P-receptor-mediated release of angiotensin II with the #-receptor agonist, isoproterenol. However, the levels of angiotensin II released were

lower than with norepinephrine or Oc2-adrenergic antagonists. We therefore suspect that there are two populations of neurons, one of which releases angiotensin II following f-receptor stimulation, as has been demonstrated in the periphery (Nakamura et al., 1986a, b, Gothert and Kollecker, 1986), and another which releases angiotensin II in response to Oc2-adrenergic blockade. However, this is not the only possible explanation of the data.

The amount of angiotensin II released from the neuronal cultures was quite high; therefore, we expected that synthesis of angiotensin II should be stimulated in order to make new angiotensin II available for release. We postulated that the renin concentration might increase in cultures following treatments which stimulated release of angiotensin II, as renin is an important enzyme in the synthesis of angiotensin II. However, we were unable to

demonstrate that this occurred with Oc2-adrenergic blockers. This suggests that renin is not an important regulatory step in angiotensin II synthesis in either neurons or glia in culture.

In the final experiment in this study, we examined whether any relationship between Oc2-adrenergic receptors and two neuropeptides known to interact with catecholamines could be demonstrated. Insulin, a catecholamine

neuromodulator, induced a time- and dose-dependent decrease in OC2-adrenergic receptor binding in glia, via a decrease in receptor number. However







85

angiotensin II, a peptide with the ability to affect synthesis and release of norepinephrine, as do O2-adrenergic receptors, demonstrated no such effect. Angiotensin II did not significantly affect glial or neuronal OC2-adrenoceptors in any way; i.e., neither total nor high affinity (agonist) binding was altered. Thus, insulin and angiotensin II interact with catecholaminergic systems differently with respect to Oc2-adrenergic receptors.

If all of the data from this study, along with some previously known interactions, are put together into a model of angiotensin II-catecholamine interactions, then fig. 17 results. Examination of this model shows that most reactions feed forward, in other words catecholamines enhance angiotensin actions and vice versa. The break in this cycle occurs through the ability of catecholamines to regulate angiotensin II receptors.

This controlling step is important for several reasons. It is the only inhibitory interaction between angiotensin II and catecholamines that has been elucidated so far, giving it great weight. Secondly, this mechanism for ojadrenergic receptor-mediated decreases in angiotensin II receptor binding is not present in the spontaneously hypertensive rat (Raizada, Muther and Sumners, 1984a), a genetic model of hypertension. So it is possible that the lack of this negative interactions is responsible for, or contributes to, the appearance of hypertension.

The model described in fig. 17 also illustrates future avenues of research which could be explored. The ability of angiotensin II and Oc2-adrenergic receptor blockade to enhance release of norepinephrine should be fully examined in culture. Complete understanding of the mechanisms for release of norepinephrine may enhance our understanding of the interactions between the two systems.







86















oc2-RECEPTOR BLOCKADE

9 (+)

ocl-RECEPTOR INCREASED NOREPINEPHRINE RELEASE STIMULATION I

P-RECEPTOR STIMULATION INCREASED ANGIOTENSIN II RELEASE( ANGIOTENSIN II RECE PTOR STIMULATION

I












Figure 17. Model depicting the interactions of catecholamines and angiotensin
II in neuronal cell cultures.








87

How well do these results correlate with the effects of angiotensin II in the brain? There is good evidence that injections of norepinephrine into sensitive parts of the brain induces drinking (Leibowitz, 1975a). The mechanism for this could be the release of angiotensin II, as very small amounts of angiotensin II (pmol) injected into the brain induces drinking (Phillips, 1987). There was some debate about the importance of the catecholamines in drinking, because much higher amounts of norepinephrine than angiotensin II had to be given to induce it. However, several factors must be taken into consideration which suggest that catecholamines may be more important than they appear on first consideration of the findings. Firstly, the receptor affinity of the catecholamines is probably three orders of magnitude lower than that of angiotensin II (Kd of approximately I aM for catecholamines and 0.1-1 nM for angiotensin II). Secondly, as catecholamines have many receptors, uptake sites etc., in the brain (to carry out functions unrelated to drinking) one would expect it to be more difficult to achieve effective concentrations of catecholamines at the receptors relevent to drinking by central injection.

It has been difficult to determine whether isoproterenol given into the brain induces drinking. Initial reports suggested that it did (Liebowitz, 1975a), but on further examination of the phenomenon it was suggested that the effect of isoproterenol was due to its escape from the brain to the periphery. Isoproterenol is a very potent stimulus to renal release of renin, which results in peripheral formation of angiotensin II which is dipsogensic. Radioactive tracer studies, combined with blockade of sympathetic outflow (Fisher et al., 1973) suggested that leakage of isoproterenol to the periphery or sympathetic activation of the juxtaglomerular apparatus to release renin were factors necessary for "central" isoproterenol-induced drinking. Despite this evidence, Fitzsimons (1979) states, "However, the complete dismissal of a possible central








88

mechanism for P-adrenergic drinking in the rat is not justified." Leibowitz

(1975b), using much smaller doses of catecholamines found a drinking response to epinephrine and norepinephrine which was dependent on both oc- and fiadrenoceptors. Further, although it does not apply to our studies, isoproterenol induced drinking, independent of renal influences, is much more reliable in other animals, eg. dog, cat, pigeon. In fact, the pigeon drinks to central

isoproterenol with a lower threshold than to peripheral isoproterenol (Fitzsimons, 1979). There is also good evidence that injection of norepinephrine into the brain (at some sites) induces increased blood pressure (Struyker-Boudier et al., 1975) as does angiotensin II (Severs et al., 1970). Therefore, the effects of norepinephrine may be mediated by release of angiotensin II. Thus, it may

be that angiotensin released by norepinephrine in a 8-receptor dependent fashion is important in the drinking and blood pressure responses to administered norepinephrine.

Norepinephrine inhibits the release of arginine vasopressin from the posterior pituitary (Share, 1983). Since angiotensin II increases the release of arginine vasopressin from the pituitary (Yamamoto, 1978), it is clear that norepinephrine does not act in the same fashion as angiotensin II in this situation. Therefore, not all of the effects of norepinephrine can be mediated by angiotensin II. The neuronal circuitary is probably unique for each system and the generalization of our findings to all catecholaminergic neurotransmission would be foolhardy. It is more probable, considering these different comparisons of the actions of angiotensin II and norepinephrine, that some populations of catecholaminergic cells interact with angiotensin II in the way these data suggest, whereas other populations do not. One likely candidate for interactions such as we describe would be the angiotensinergic connections which originate in the subfornical organ and terminate in the median preoptic








89

nucleus. The cells receiving input in the median preoptic nucleus are stimulated by electrical stimulation of the subfornical organ or by application of angiotensin II. Both of these types of stimulation are enhanced by pressure injection of norepinephrine (Graham et al., 1985). While not the only

interpretation of the results, it is tempting to suggest that, in accordance with our data, norepinephrine is inducing increased angiotensin II release and thus, facilitating the stimulation.

A candidate for an area where there is no such interaction is the anterior hypothalamus, where it has been established that neither antagonism nor stimulation of fl-receptors affects spontaneous angiotensin II release (Brosnihan et al., 1988), whereas, in cultures from whole brain we found an interaction.

In summary significant data concerning the control of angiotensin II release by norepinephrine from brain cells in culture have been obtained. Increased release of angiotensin II does not lead to increased synthesis of renin, an important enzyme in the angiotensin II synthetic pathway. Angiotensin II does not significantly affect Oc2-adrenergic receptors in neuronal or glial cultures, although they share common activities for example, release of norepinephrine. This suggests that other regulatory steps are more important. Alpha2-adrenergic receptors were characterized in both neuronal and glial cultures (the first direct characterization in cultures from rat brain) and visualized by autoradiography in neuronal cultures.














REFERENCES

Baker, S.P., Sumners, C., Pitha, J. and Raizada, M.K. Characteristics of the
beta-adrenoceptor from neuronal and glial cells in primary cultures of rat
brain. J. Neurochem. 47; 1318-1326, 1986.

Barbaccia, M., Chuang, D., and Costa, E. Is insulin a neuromodulator? Adv.
Biochem. Psychopharmacol. 33: 511-518, 1982.

Basso, N., Kurnjek, M.L., Mikulic, L., Cannata, M.A., and Taquini, A.C.,
Effect of intraventricular 6-hydroxydopamine (60HDA) on the
angiotensinogen concentration of the brain. Clin. Exp. Hypert-Theory and
Practice A6 (10 and 11): 1773-1776, 1984.

Bickerton, R.K. and Buckley, J.P. Evidence for a central mechanism in
angiotensin-induced hypertension. Proc. Soc. Exp. Biol. Med. 106: 834-836,
1961.

Boadle-Biber, M.C., Hughes, J. and Roth, R.H. Acceleration of catecholamine
biosynthesis in sympathetically innervated tissues by angiotensin-II-amide.
Nature (Lond) 222: 987-988, 1969.

Boyajian, C.L., Loughlin, S.E. and Leslie, F.M. Anatomical evidence for
alpha-2 adrenoceptor heterogeneity: Differential autoradiographic distributions of [3H]-rauwolscine and [3H]-idazoxan in rat brain. J.
Pharmacol. exp. Therap. 241: 1079-1091, 1987.

Boyd, F.T., Jr., Clarke, D.W., Muther, T.F. and Raizada, M.K. Insulin
receptors and modulation of norepinephrine uptake by insulin in neuronal
cells in culture. J. Biol. Chem. 260: 15880-15884, 1985.

Boyd, F.T., Jr., Clarke, D.W., and Raizada, M.K. Insulin inhibits specific
norepinephrine uptake in neuronal cultures from rat brain. Brain Res.
398:1-5, 1986.

Boyd, F.T., Jr. and Raizada, M.K. Effects of insulin and tunicamycin on
neuronal insulin receptors in culture. Am. J. Physiol. 245: C283-C287,
1983.

Brosnihan, K.B., Schiavone, M.T., Sprunger, A.E., Chappell, M.C., Rizzo, M.
and Ferrario, C.M. In vivo release of angiotensin II from the rat
hypothalamus. Hypert. 11 (Suppl I): 1158-1162, 1988.

Buggy, J., Fisher, A.E., Hoffmann, W.E., Johnson, A.K. and Phillips, M.I.
Ventricular obstruction: Effect on drinking induced by intracranial
injections of angiotensin. Sci. 190: 72-74, 1975.



90








91

Camacho, A. and Phillips, M.I. Separation of drinking and pressor responses
to central angiotensin by monoamines. Am. J. Physiol. 240: R106-R1 3,
1981.

Campbell, D.J., Bouhnik, J., Menard, J., and Corvol, P. Identity of
angiotensinogen precursors of rat brain and liver. Nature 308: 206-208,
1984.

Carroll, R.G. and Opdyke, D.F. Evolution of angiotensin II-induced
catecholamine release. Am. J. Physiol. 243: R65-R69, 1982.

Cheng, Y.C. and Prusoff, W.H. Relationship between the inhibition constant
(Ki) and the concentration of inhibitor which causes 50 percent inhibition
of an enzymatic reaction. Biochem. Pharmacol. 22: 3099-3108, 1973.

Chevillard, C., Duchene, N., Pasquier, R. and Alexandre, J-M. Relation of
the centrally evoked pressor effect of angiotensin II to central
noradrenaline in the rabbit. Eur. J. Pharmacol. 58: 203-206, 1979.

Clarke, D.W., Boyd, F.T., Jr., Kappy, M.S. and Raizada, M.K. Insulin binds to
specific receptors and stimulates 2-deoxy-D-glucose uptake in cultured glial
cells from rat brain. J. Biol. Chem. 259: 11672-11675, 1984.

Clarke, D.W., Boyd, F.T., Jr., and Raizada, M.K. Insulin is released from
neuronal cultures from the rat brain. J. Neurochem. 47: 831-836, 1986.

Davila, D. and Khairallah, P.A. Angiotensin II and biosynthesis of
norepinephrine. Arch. Int. Pharmacodyn. 193: 307-314, 1971.

Ebersolt, C., Perez, M. adnd Bokaert, J. oc and Oc2-adrenergic receptors in
mouse brain astrocytes from primary cultures. J. Neurosci. Res. 6: 643652, 1981.

Dzau, V.J. Significance of the vascular renin-angiotensin pathway.
Hypertension 8: 553-559, 1986.

Evans, T., McCarthy, K.D. and Harden, T.K. Regulation of cyclic AMP
accumulation by peptide hormone receptors in immunocytochemically
defined astroglial cells. J. Neurochem. 43: 131-138, 1984.

Feldberg, W., and Lewis, G.P. The action of peptides on the adrenal
medulla. Release of adrenaline by bradykinin and angiotensin. J. Physiol.
171: 98-108, 1964.

Feldstein, J.B., Gonzales, R.A., Baker, S.P., Sumners, C., Crews, F.T. and
Raizada, M.K. Decreased alpha I-adrenergic receptor mediated inositide hydrolysis in neurons from hypertensive rat brains. Am. J. Physiol. 251:
C230-C237, 1980.

Feldstein, J.B., Pacitti, A.J., Sumners, C. and Raizada, M.K. Alpha1adrenergic receptors in neuronal cultures from rat brain: Increased expression in the spontaneously hypertensive rat. J. Neurochem. 47: 11901198, 1986.








92

Fisher, A.E. Relationships between cholinergic and other dipsogens in the
central mediation of thirst. In: Neuropsychology of Thirst: New Findings and Advances in Concepts. Ed. Epstein, A.N., Kissileff, H. R. and Stellar,
E. Washington. Winston and Sons, 1973, pp 243-78.

Fitzsimons, J.T. Pharmacology of Drinking. In: The Physiology of Thirst
and Sodium Appetite. Ed. Fitzsimons, J.T. Cambridge. Cambridge Univ.
Press, 1979, p 433.

Frankhuyzen A.L. and Mulder, A.H. Noradrenaline inhibits depolarizationinduced H-serotonin release from slices of rat hippocampus. Eur. J.
Pharmacol. 63: 179-182, 1980.

Fregly, M.J., Rowland, N.E. and Greenleaf, J.E. Clonidine antagonism of
angiotensin-related drinking: a central site of action. Brain Res. 298: 321327, 1984.

Fuxe, K., Andersson, K., Locatelli, V., Mutt, V., Lundberg, J., Hokfelt, T.,
Agnati, L.F., Eneroth, P. and Bolme, P. Neuropeptides and central catecholamine systems: Interactions in neuroendocrine and central cardiovascular regulation. In: Neural Peptides and Neuronal
Communication. Ed. Costa, E. and Trabucchi, M., Raven Press, New York,
1980, pp37-50.

Garcia-Sevilla, J.A., Dubocovich, M.L. and Langer, S.Z. Angiotensin II
facilitates the potassium-evoked release of 3H-noradrenaline from the
rabbit hypothalamus. Eur. J. Pharmacol. 56: 173-176, 1979.

Gonzales, R.A., Crews, F.T., Sumners, C. and Raizada, M.K. Norepinephrine
regulation of alphaI-stimulated phosphoinositide hydrolysis in primary
neuronal cultures. J. Pharmacol. Exp. Therap. 242: 764-771, 1987.

Gonzales, R.A., Feldstein, J.B., Crews, F.T., Raizada, M.K. Receptor-mediated
inositide hydrolysis is a neuronal response: Comparison of primary
neuronal and glial cultures. Brain Res. 345: 350-355, 1985.

Gordon, F.J., Brody, M.J., Fink, G.D., Buggy, J., and Johnson, A.K. Role of
central catecholamines in the control of blood pressure and drinking
behavior. Brain Res. 178: 161-173, 1979.

Gordon, F.J., Brody, M.J. and Johnson, A.K. Regional depletion of central
nervous system catecholamines: Effects on Blood Pressure and Drinking
Behavior. Brain Res. 345: 285-297, 1985.

Gothert, M. and Huth, H. Alpha-adrenoceptor-mediated modulation of 5hydroxytryptamine release from rat brain cortex slices. NaunynSchmiedeberg's Arch. Pharmacol. 313, 21-26, 1980.

Gothert, M. and Kollecker, P. Subendothelial #2-adrenoceptors in the rat
vena cava: Facilitation of noradrenaline release via local stimulation of angiotensin II synthesis. Naunyn-Schmiedeberg's Arch. Pharmacol. 334:
156-165, 1986.




Full Text

PAGE 1

CATECHOLAMINE AND ANGIOTENSIN II INTERACTIONS IN PRIMARY CUL TURES OF BRAIN CELLS By ELAINE MARY RICHARDS-SUMNERS 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 1988

PAGE 2

ACKNOWLEDGEMENTS I would like to thank the Chairman of my committee, Dr M. Ian Phillips, for giving me the opportunity to pursue a graduate degree and the freedom to do the project of my choice. and all students in his lab Ian speaks and writes very lucidly about science, benefit from his tuition in this area. His enthusiastic support throughout the years I have been in his lab have been gratefully received, too. Collectively, I would like to thank the other members of my committee, Ors. Fregly, Raizada, and Shiverick, for their hard work on my behalf. I will never forget their involvement and support. There are some other people to whom l am very grateful for the use of their facilities and/or techniques. These are Dr. P. Klein in Pathology, Dr. G. Shaw in Neuroscience, Dr. S. Baker and Dr. T. Muther in Pharmacology. The technical help of Ms. J. Perez, Mr. J. Hogenesch, Ms. S Fuentes and Mr. J. Neal is also greatly appreciated. Big Kimura, who has never become cross with any of my millions of interruptions, deserves a medal! The help with the use of word perfect given me by Kevin Fortin is gratefully acknowledged too. The biggest thanks of all, however, go to my family for all that they have put up with in my graduate school years. 11

PAGE 3

ACKNOWLEDGEMENTS KEY TO ABBREVIATIONS ABSTRACT CHAPTERS TABLE OF CONTENTS PAGE 11 V Vll I. INTRODUCTION: INTERACTIONS BETWEEN CATECHOLAMINES AND ANGIOTENSIN II . . . . 1 Colocalization of Catecholamines and Angiotensin II in the Central Nervous System . . Effects of Catecholamines and Angiotensin II which are Similar . . . . 5 The Effects of Angiotensin II on Catecholamines. 5 The Effects of Catecholamines on Angiotensin II. 8 The Cell Culture Model. . . . 12 The Effects of Angiotensin II on Catecholamines in Cell Culture . . . . 13 The Effects of Catecholamines on Angiotensin II in Cell Culture 13 General Hypothesis. . . . 14 II. CHARACTERIZATION OF cx:2-ADRENOCEPTORS ON NEURONAL AND GLIAL CELLS IN CULTURE Introduction Methods Results Discussion III. LIGHT MICROSCOPIC AUTORADIOGRAPHY OF [3H]16 16 18 21 31 YOHIMBINE BINDING SITES IN NEURONS IN CULTURE 35 Introduction Methods Results Discussion IV ADRENERGIC MECHANISMS MEDIA TING ANGIOTENSIN II 35 36 37 46 RELEASE 47 Introduction 47 111

PAGE 4

Methods Results Discussion V. EFFECT OF VARIOUS ADRENERGIC DRUGS ON THE RENIN 48 51 61 CONCENTRATION OF BRAIN CELLS IN CULTURE 66 Introduction Methods Results Discussion 66 66 69 69 VI. REGULATION OF cx:2-ADRENERGIC RECEPTORS BY PEPTIDES 71 Introduction Methods Results Discussion VII. GENERAL CONCLUSIONS REFERENCES BIOGRAPHICAL SKETCH IV 71 72 74 79 82 90 100

PAGE 5

ACh ARC AVP Bmax BST cAMP CeA COMT CSF DMEM DMI DNase I GFAP G 1 Gpp(NH)p GTP kobs KEY TO ABBREVIATIONS V acetylcholine cytosine arabinoside arginine vasopressin maximum number of binding sites bed nucleus of the stria terminalis cyclic adenosine monophosphate central nucleus of the amygdala catechol-0-methyl-transferase cerebrospinal fluid Dulbecco's modified Eagle's medium desmethylimipramine deoxyribonuclease I glial fibrillary acidic protein inhibitory GTP binding protein stimulatory GTP binding protein 5'-guanylylimidodiphosphate guanosine triphosphate inhibitory constant50 association constant dissociation constant equilibrium dissociation constant inhibitory dissociation constant observed association constant

PAGE 6

LC LHA LS MAO ME NTS OB. OVLT PB1 PBS PDHS PF PMSF pp PVH RIA SFO st Tris VTA ZI 5HT 60HDA VI locus ceruleus lateral hypothalamic area lateral septum monoamine oxidase medain eminence median preoptic nucleus nucleus tractus solitarius olfactory bulb organon vasculosum lamina terminalis parabrachial nucleus-lateral phosphate buffered saline plasma derived horse serum parafascicular nucleus phenylmethylsulfonyl fluoride peripeduncular nucleus para ventricular hypothalamus radioimmunoassay subfornical organ stria terminalis nucleus of the (Tris[hydroxymethyl]-aminomethane hydrochloride) ventral tegmental area zona incerta serotonin 6-hydroxydopamine

PAGE 7

Abstract of Dissertation Presented to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements of the Degree of Doctor of Philosophy CATECHOLAMINE AND ANGIOTENSIN II INTERACTIONS IN PRIMARY CUL TURES OF BRAIN CELLS By ELAINE MARY RICHARDS-SUMNERS DECEMBER 1988 Chairman : Dr. M.I. Phillips Major Department: Physiology Alpha2-adrenergic receptors were characterized in neurons and glia in cell culture. The receptors were present in both cell types, with similar affinities, but glia in culture contain ten times more receptors than neurons. The neuronal alpha2-receptors were visualized by light microscopic autoradiography. Antagonism of alpha2-adrenoceptors induced the release of the neuropeptide angiotensin II from neuronal cultures, but not from glial cultures, even though they also contain angiotensin II. The material released by alpha2-antagonism from neuronal cultures was found to migrate with synthetic angiotensin II on high pressure liquid chromatography, and thus was authentic angiotensin II. Norepinephrine also induced release of angiotensin II from neuronal cultures, which could not be blocked by alpha 1-adrenergic antagonists. Beta-adrenoceptor agonists induced the release of angiotensin II. These data suggest that angiotensin II is released from two populations of cells, one responding to beta-agonists and the other to alpha2-antagonists. Stimulation of release of angiotensin II by yohimbine did not result in an increase in the neuronal cell Vll

PAGE 8

content of renin, suggesting that renin is not an important regulator of the availability of angiotensin II in brain cells in culture. Insulin, a neuroactive peptide, regulated the number of alpha2adrenoceptors in glial cultures, but angiotensin II had no effect on alpha2adrenoceptors in neurons or glia. A model describing these findings in relation to other known interactions of angiotensin II and catecholamines in brain cell culture is discussed. viii

PAGE 9

CHAPTER I INTRODUCTION Interactions between Catecholamines and Angiotensin II Phylogenetically, there is a very old interaction between catecholamines and angiotensin II (Carroll and Opdyke, 1982). With the recent acceptance of tissue renin-angiotensin systems, separate from blood-borne angiotensin II, interactions with catecholamine systems have been discovered in various tissues. They are apparent in vascular beds, including the mesenteric arteries (Nakamura et al., 1986a, b) and vena cava (Gothert and Kollecker, 1986), cultured smooth muscle cells (Dzau, 1986), the vas def erens (Trachte, Stein and Peach, 1987) and in the brain, the area of particular interest to us This introductory chapter will review what is known about these interactions between catecholamines and angiotensin II. It will include a discussion of the many advances in our knowledge of these interactions in the brain which have been made possible by the use of tissue culture techniques. These cultures of one-day old rat brains enriched either for neurons or glia have greatly facilitated these studies. Colocalization of Catecholamines and Angiotensin II in the Central Nervous System In the following section, discrete localizations of angiotensin II are taken from Weyhenmeyer and Phillips, (1982) and Lind et al. (1985). The catecholamine localizations are from Lindvall and Bjorkland (1983) and Moore and Bloom (I 979). The overlap between the anatomical localization of the catecholamines and angiotensin II systems is very striking (Fig I). The catecholamine and angiotensin II systems are predominantly located in areas of the brain

PAGE 10

B Figure I. Diagram showing the high degree of overlap between the anatomical localization of the norepinephrine system and the angiotensin II system in the rat brain. A) Diagram showing the origin and pathways of norepinephrine containing tracts in the rat brain. Shaded areas are norepinephrine terminal fields. From Chemical Neurobiology ed. H F. Bradford. B) Diagram showing the distribution of angiotensin II cells ( ) and tracts (-) in the rat brain. Redrawn from Lind, Swanson and Ganten, 1984. 2

PAGE 11

associated with homeostatic control functions of the body, for example, the hypothalamus and brain stem. Angiotensin II containing fibres project in the medial forebrain bundle, in the stria terminalis, in the periventricular regions ( or ventral catecholamine bundle) and in the reticular formation. All of these pathways also contain catecholamine cell projections. However, a colocalization of fibre tracts is not indicative of functional interactions unless we can demonstrate overlap of terminal fields, either with other terminal fields or with cell bodies, of the two systems. These kinds of functional colocalization are seen in the paraventricular and supraoptic nuclei of the hypothalamus, which are terminal fields for catecholamine projections and are the major concentrations of angiotensin II cell bodies in the brain. They also contain many angiotensin II receptors (Mendelsohn et al., 1984). There are cell bodies which contain angiotensin II in the nuclei at both ends of the stria terminalis, at least one of which is a catecholamine terminal field. Similar types of colocalization can be demonstrated for many projection fields of the catecholamine system Summarised in table l, taken from Lindvall and Bjorkland (1983) are the different projections from the noradrenergic lateral tegmental and dorsal medullary neurons. Every area with the possible exceptions of the dorsal raphe nucleus and the ventromedial hypothalamic nucleus, has either cell bodies or fibres containing angiotensin II. This high degree of overlap with angiotensin II containing areas is not so apparent for the A6 and A4 (locus coeruleus) catecholamine projection areas as this projection is mostly to the cerebral cortex However, the septa! regions do have significant input from these norepinephrine cell groups and have angiotensin II cells and fibres. In the spinal cord the area of overlap between angiotensin II and catecholamines is the interomediolateral columns, which are 3

PAGE 12

the preganglionic sympathetic nerve cell bodies. Another area of overlap which is worth specifically mentioning is the subfornical organ (SFO) This circumventricular organ located in the wall of the third ventrical is a major receptor-and cell body-containing area for angiotensin II catecholamine terminals. It is also rich in Table 1.1, Origins of different projections from the noradrenergic lateral tegmental and dorsal medullary neurons. TERMINAL AREA Spinal cord NTS and dorsal vagal complex Spinal sensory trigeminal nucleus Locus ceruleus Parabrachial nucleus Dorsal raphe nucleus Medial preoptic area Paraventricular hypothalamic nucleus Supraoptic nucleus Dorsomedial hypothalamic nucleus Ventromedial hypothalamic nucleus Anterior, lateral and posterior hypothalamic areas Median eminence Arcuate nucleus Septum Amygdala = areas without obvious angiotensin II content. From Lindvall and Bjorkland, 1983. ORIGIN A5, A7 Al, A2 Al, A5 Al, A2, A7 Al Al Al, A2 Al, A2 Al, A2 Al, A2 Al, A2 Al, A2 Al, A2 Al, A2 Al, A2 Al, A2 There are, naturally, angiotensin II containing areas which do not share catecholamine inputs, e g substantia innominata, and catecholamine projection areas which have very sparse angiotensin II staining, e.g., cerebral cortex. However, the extensive nature of the colocalization suggests that the opportunity for interactions is present. The results of ultrastructural studies confirm this view, though very few of these studies have been performed. It has been demonstrated that Ang II and arginine vasopressin (A VP) colocalize in paraventricular cells of the hypothalamus, (Hoffman et al., 1982). These A VP-angiotensin II-containing cells 4

PAGE 13

are ringed by synaptic boutons from catecholamine cells (Sladek and McNeill, 1980). In summary, catecholamines and angiotensin II co localize extensively throughout the brain and spinal cord. The co localizations are such that synaptic interactions could occur, i.e. terminal fields overlap, permitting axoaxonal synapses, or cell bodies and terminal fields overlap, allowing axo-dendritic or axo-somatic synaptic connections. Thus, on purely anatomical grounds, interactions between catecholamines and angiotensin II are entirely possible. Effects of Angiotensin II and Catecholamines which are Similar Angiotensin II injected into the lateral ventricles causes drinking which is mediated through receptors located on the organum vasculosum of the lamina terminalis (OVL T), a circumventricular organ lying in the tip of the optic recess of the third cerebroventricle, (Buggy et. al., 1975). The OVLT also colocalizes angiotensin II and catecholamines. Similarly, norepinephrine injected into the brain causes drinking (Leibowitz, 1975a, b ). Angiotensin II given into the brain causes increases in blood pressure, (Bickerton and Buckley, 1961, Severs et. al., 1970). Norepinephrine injected into certain sites of the brain also causes an increase in blood pressure, (Struyker-Boudier et. al., 1975), although at other sites it causes a decrease in blood pressure. These two strikingly similar actions of angiotensin II and catecholamines could be interpreted to mean that catecholamines act through releasing angiotensin II or vice versa. This possibility will be developed further later in this dissertation. The Effects of Angiotensin II on Catecholamines The effects of angiotensin II on catecholamine turnover. Angiotensin II injected centrally induces increased turnover (or utilization) of norepinephrine 5

PAGE 14

in brain regions whose function is to regulate blood pressure. Dopamine utilization in the same regions was unaffected. The regions examined contained angiotensin II receptors and catecholamines, and those affected were hypothalamus, SFO, A 1, locus coeruleus and raphe magnus, (Sumners and Phillips, 1983). Fuxe et al. (1980) showed that central injection of renin, the enzyme which cleaves angiotensin I from angiotensinogen increases dopamine turnover in the external layer of the median eminence, while decreasing norepinephrine turnover in the magnocellular paraventricular nucleus and the dorsomedial hypothalamic nucleus The renin injection also increased epinephrine turnover in the DCMO, a block of brainstem tissue containing the nucleus tractus solitarius, nucleus commisuralis and the dorsal motor nucleus of the vagus, sites of angiotensin II-catecholamine colocalization. The effects are specifically mediated by angiotensin II as they can be blocked by captopril, SQ 14225, an inhibitor of the conversion of angiotensin I to angiotensin II (Fuxe et al., 1980). The effects of angiotensin II on the uptake and release of catecholamines. In brain tissue Ang II affects both the release and re-uptake of catecholamines. Angiotensin II can evoke release of ( 3H]-dopamine from slices of stria tum exposed to [3H]-tyrosine without affecting synthesis of (3H]-dopamine, an effect blocked by angiotensin II receptor blockers (Simmonet and Giorguieff-Chesselet, 1979). Angiotensin II does not seem to be able to cause spontaneous norepinephrine release from brain tissue, although it affects electrically stimulated norepinephrine release. At low doses, 0 1 and nM it enhanced, and at high doses, 1 M, it decreased the electrically evoked release of norepinephrine (Schacht, 1984). Captopril and HOE-498 diacid, two different converting enzyme inhibitors at 1 M, were able to decrease the stimulated outflow. Furthermore, the diacid had the same effect on release as clonidine, 6

PAGE 15

an oc2-agonist, and it could partially inhibit the effects of rauwolscine, an oc2-antagonist (Schacht, 1984). This strengthens the argument for interaction between the two systems. Similarly, in a dose-dependent fashion, angiotensin II can increase the potassium-stimulated norepinephrine release from rabbit hypothalamus, an effect blocked by saralasin, an angiotensin II receptor antagonist (Garcia-Sevilla et al., 1979). The effect of angiotensin II on electrically stimulated release of norepinephrine has been examined in more detail in a study by Meldrum et al., (l 984). Angiotensin II increased the overflow of norepinephrine from the A2 region of rats, but not the hypothalamus. This stimulation was blunted in animals on a low sodium diet. Wistar rats were more sensitive to the effect of angiotensin II on norepinephrine overflow from the A2 region than Sprague Dawley rats, and a low sodium diet completely blocked the effect rather than just blunting it. Thus angiotensin II and catecholamines interact in the A2 region of the brain in a sodium-sensitive manner (Meldrum et al., 1984). The A2 area contains angiotensin II receptors, angiotensin II cells and catecholamines, and is important in the control of blood pressure. These data suggest that these angiotensin II-norepinephrine interactions which are dependent on the sodium status of the rat, could be important in blood pressure homeostasis. The in vitro findings discusssed above have also been demonstrated in vivo. Ventriculocisternal infusion of angiotensin II in rabbits caused an increase in the concentration of norepinephrine in the cerebrospinal fluid (CSF) which was highly correlated with the increase in blood pressure due to angiotensin II (Chevillard et al., 1979). 7

PAGE 16

The effects of angiotensin II on norepinephrine uptake are more complex. It has been shown that Ang II partially inhibits norepinephrine uptake by brain tisssue, with no effect on release, after one hour of Ang II exposure (Palaic and Khairallah, 1967). But recent studies of neuronal cell culture have demonstrated a biphasic effect. Shortly after Ang II exposure, enhancement of norepinephrine uptake was observed, while after longer exposure, inhibition was seen. This study did not examine release (Sumners and Raizada, 1986). The previously cited study only examined one-time point, so the effects of angiotensin II on the reuptake of norepinephrine are still unclear. Peripherally, angiotensin II increases the amount of norepinephrine released spontaneously from sympathetic nerves as well as increasing release to sympathetic stimulation (Zimmerman and Gisslen, 1968, Hughes and Roth, 1971 ). The effects of angiotensin II on synthesis of catecholamines In sympathetically innervated tissues an acceleration of catecholamine biosynthesis is seen to angiotensin II treatment (BoadleBiber et al. 1969, Davila and Khairallah, 1971 ). These effects and alterations in uptake and release caused by angiotensin II (Reit, 1972; Feld berg and Lewis, 1964; Starke, 1971; Palaic and Khairallah, 1967) are probably interrelated as increased release would be expected to lead to increased synthesis. It is probable that these effects would be observed in brain tissue should the studies be performed. The Effects of Catecholamines on Angiotensin II The effects of catecholamine blockade on responses to angiotensin II Angiotensin II delivered to the brain causes a pressor response (Bickerton and Buckley, 1961 ). This blood pressure increase can be attenuated by ex-antagonists (Camacho and Phillips, 1981 ), implying that it is partly dependent upon catecholamines. Furthermore, the central dipsogenic response and central pressor response to angiotensin II can be attenuated by the destruction of 8

PAGE 17

catecholamines in the brain by the catecholamine neurotoxin 6-hydroxydopamine (6-0HDA) (Gordon et al. 1979, Gordon et al, 1985). Drinking was restored in 6-0HDA treated animals when transplants of norepinephrine producing brain areas from fetal rats were made into their basal forebrains (McRae-Dequeurce et al., 1986). In addition, clonidine, an cx2-receptor agonist, can inhibit drinking to angiotensin II through an cx2-adrenergic receptor-mediated mechanism, (Fregly et al., 1984). These findings clearly suggest that the full central pressor and dipsogenic responses of angiotensin II are dependent on intact catecholamine systems Regulation of the central angiotensin II system by catecholamines Indications that the central angiotensin II system is regulated by catecholamines are revealed in several interesting studies. correlation between angiotensinogen concentration and Initially, a norepinephrine concentration in brain tissue, but no such correlation of dopamine and angiotensinogen concentrations, was observed (Printz et al., 1979). When rats were treated with reserpine, a catecholamine-depleting drug, for 24 hours, angiotensinogen was increased in the septum, periaqueductal gray nucleus stria terminalis, medial basal hypothalamic nuclei, hippocampus and ventral tegmental areas of the brain. The levels of angiotensinogen had returned to pretreatment levels 96 hours after reserpine, even though catecholamines were still depleted The increase in angiotensinogen at 24 hours after reserpine could be annuled by adrenalectomy, which prevented the rise in corticosteroids which follows reserpine treatment. However, corticosteroid levels regulate norepinephrine levels and thus the authors postulated that the angiotensinogen concentration changes were possibly due to alterations in norepinephrine levels, secondary to corticosteroid changes (Printz et al. 1983 ). These hypotheses have still not been tested. 9

PAGE 18

It is difficult to interpret these studies because interference with the catecholamine system, for example by reserpine, causes changes in fluid balance, by such factors as the reserpine-induced diarrhea, and the necessity for intact catecholamine systems for full expression of drinking behaviours, etc. Since the brain angiotensin system is thought to regulate fluid balance, it would be expected that these fluid balance changes alone could alter the central angiotensin II system directly, without any influence of catecholamine depletion. A series of studies which addressed this problem has clarified the situation to some extent. The approach taken was to change catecholamine neurotransmission in the brain and/or periphery with various drugs and examine the effects on the angiotensin II and catecholamine systems. The results of these studies show that there is an interaction between the systems but do not permit exact mechanisms to be illuminated (Mikulic et al, 1986, Kurnjek et al, 1986, Trolliet et al., 1986 and Basso et al, 1984 ). It will probably be necessary to examine regulation of mRNA for all components of both of the systems before we fully appreciate these interactions (Basso, personal communication). The effects of catecholamines on angiotensin II release. It has recently been shown that isolated mesenteric arteries treated with iosproterenol, a /3-receptor agonist, release angiotensin II (Nakamura et al., 1986a, b). The release of angiotensin II caused by isoproterenol could be blocked by propranolol, a /3blocker, suggesting the effect was mediated by /3-receptors. The basal release of angiotensin II was unaffected by /3-blockade, but was sensitive to captopril, a converting enzyme inhibitor. This shows that there are two mechanisms governing the release of angiotensin II from mesenteric arteries. These are basal release which is unaffected by /3-receptors and stimulated angiotensin II release, which is under /3-catecholaminergic control. IO

PAGE 19

Gothert and Kollecker ( 1986) performed a similar study in the inferior vena cava of rats. They were able to suggest from experiments which manipulated the tissue slices, that the .82-receptors, whose activation led to angiotensin II release, were located on the subendothelium, most probably on smooth muscle cells. The angiotensin II receptors whose activation stimulated norepinephrine release were probably located on the sympathetic nerve terminal. In a related study, the existence of a similar mechanism for release of angiotensin II in the kidney was examined. The authors interpret their data as showing that ,8-mediated angiotensin II release does not occur, but on close examination of this paper, it is possible to draw the opposite conclusion (Rump and Majewski, 1987). In their study high, but not low, doses of captopril, a converting enzyme inhibitor, could block the ,8-adrenoceptor mediated increase in norepinephrine release from the isolated kidney. Saralasin, an angiotensin II receptor blocker, had no effect on the ,8-adrenoceptor mediated release. The authors could not explain the effects of high doses of captopril, and saralasin is a very difficult drug to use as it has agonist as well as antagonistic properties. As a whole these considerations mean that it is not impossible for a .B-receptor mediated release of angiotensin II to occur in the kidney, which in turn induces increased release of norepinephrine. Thus, in three areas examined, there is evidence which supports the hypothesis that ,8-receptor stimulation causes angiotensin II release. A suggested mechanism is via ,8-adrenergic stimulation of converting enzyme activity (Gothert and Kollecker, 1986), although this has not been proven. An examination of the control of angiotensin release from rat hypothalamus by {1adrenoceptors has recently been made. In this study, performed in vivo with push-pull cannulae in rats, there was no evidence that ,8-adrenergic stimulation led to increased angiotensin II release (Brosnihan et al., 1988). 11

PAGE 20

Cell Culture Model. Apart from the few studies mentioned above, nothing else was known about catecholamine-angiotensin II interactions until brain cell culture methodology was applied to the problem. In these cultures it is possible to look at direct interactions between these systems without secondary influences. Thus a great deal of new information about the interactions of angiotensin II with catecholamines has been elucidated, as well as other factors which regulate the Ang II system in the brain. Characterization of the Cell Culture System. Initial studies showed that all components of the renin-angiotensin system are present in brain cell cultures They contain renin, (Hermann et al., 1987), angiotensinogen (Hermann et al., 1988a), angiotensin I, (Hermann et al., 1988a, b), and angiotensin II (Raizada et al., 1984b, Hermann et al., 1988a, b). Converting enzyme activity has not been directly measured, but must be present because of the production of angiotensin II. The cells incorporate [ 3H] isoleucine into immunoprecipitable angiotensin II (Hermann et al 1988a). They have specific receptors for angiotensin II (Raizada et al., 1981, Raizada et al., 1987). Many components of the catecholamine system are present in these cultures. The catecholamines dopamine, norepinephrine, and epinephrine, can be measured and in some cases visualized, implying that there are synthetic enzymes present (Sumners et al., 1983a). The catecholamine degrading enzymes, catechol-o-methyl transferase (COMT) and monoamine oxidase (MAO) are present (Sumners et al., 1987). There are ex 1-receptors linked to phosphatidylinositide hydrolysis (Feldstein, 1986), P-receptors stimulation of which results in cAMP accumulation (Baker et al. 1986). Thus brain cell cultures are a good model in which to examine these two systems' interactions. 12

PAGE 21

The Effects of Angiotensin II on Catecholamines. Angiotensin II treatment of neuronal cell cultures resulted in an increase in the cell and media content of catecholamines (Sumners et al., 1983b). This implied that both synthesis and angiotensin II. Later studies release of catecholamines was with labelled tyrosine showed enhanced by that Ang II stimulated the conversion of dopamine to norepinephrine, implying that angiotensin II stimulated the enzyme dopamine ,8-hydroxylase, (McLean, Raizada and Sumners, unpublished data). Furthermore angiotensin II enhanced the activity of MAO, but not COMT, the catecholamine degrading enzyme, (Sumners et al., 1987), suggesting that it first increased release of catecholamines. However, release of catecholamines from cell cultures has never been directly examined. Angiotensin II has significant effects on norepinephrine uptake mechanisms, which are somewhat difficult to interpret without appreciation of the status of release of catecholamines under the same conditions. Shortly after exposure to angiotensin II, norepinephrine uptake is enhanced, whereas after longer periods of time, uptake is decreased (Sumners and Raizada, 1986). With this multiplicity of effects of angiotensin II on the catecholamine system, it was expected that catecholamine receptors would also be affected by angiotensin II. This was not true, however, of ex 1-adrenoceptors, nor of ,8-receptors (unpublished data). Catecholamine Effects on Angiotensin II in Cell Culture The effect of norepinephrine on angiotensin II receptors. N orepinephrine has been shown to be an important regulator of angiotensin II receptors Adding norepinephrine to neuronal cultures resulted in a decrease in the number of angiotensin II receptors which is mediated by ex l -adrenergic receptors (Sumners, Watkins and Raizada, 1986). Decreasing norepinephrine levels with ex-13

PAGE 22

methyl-p-tyrosine resulted in an increase in the number of angiotensin II receptors (Sumners and Raizada, 1984). The effects of catecholamines on other aspects of the angiotensin II system. Knowledge of other influences of catecholamines on the angiotensin II system, for example on metabolism, are very limited, mainly because they have not been examined. General Hypothesis. From this survey of the literature dealing with catecholamine-angiotensin II interactions, it is obvious that some big areas of potential interaction have not been examined. For example, the possibility that angiotensin II could be released from brain cell cultures by catecholaminergic mechanisms, as in the periphery. Furthermore, norepinephrine acts on three types of receptor, ex 1, oc:2 and p. nor had The existence of oc:2-receptors in these cultures had not been examined, their role, if any, m the interaction between catecholamine and angiotensin II been tested. Alpha2-receptors colocalize with angiotensin II in the brain In two seperate studies, oc:2-receptor distributions were mapped in the brain. oc:2-receptor densities were highest in areas which stain heavily for angiotensin II, e.g. the norepinephrine terminal fields, LC and NTS, interomediolateral cell column of the spinal cord, substantia gelatinosa of the trigeminal nucleus, paraventricular and arcuate nuclei of the hypothalamus, amygaloid nuclei, anterior nucleus, and bed nucleus Boyajian et al., 1987). olfactory nuclei, lateral septum, dorsal parabrachial of the stria terminalis (Unnerstall et al., 1984, Interest in oc:2-receptors in these interactions was generated for several reasons. Firstly, oc:2-receptors influence norepinephrine release, as does angiotensin II. Could there be a common mechanism of action? receptors also modulate the release of other substances, for example, serotonin, 14

PAGE 23

acetyl-choline and endorphin (Frankhuyzen and Mulder, 1980, Gothert and Huth, 1980, Vizi and Knoll, 1971 ). Could cx:2-receptors modulate angiotensin II release? If cx:2-receptors did influence release of angiotensin II, would synthesis of angiotensin II also be modulated? These questions led to the establishment of three specific aims 1. Do neurons and glia in cell culture possess cx:2-adrenoceptors? 2. Can cx:2-receptors modulate angiotensin II release? If so, can effects on angiotensin II metabolism, specifically on renin concentration, also be seen? 3. Can a common mechanism of angiotensin II and cx:2-receptors be discovered which would explain how angiotensin II and cx:2-receptors modulate NE release? 15

PAGE 24

CHAPTER TWO CHARACTERIZATION OF cx:2-ADRENOCEPTORS ON NEURONAL AND GLIAL CELLS IN CULTURE. Introduction The discovery of presynaptic inhibitory autoreceptors which modulate the release of norepinephrine was one of the factors which solidified the view of two subtypes of alpha-adrenergic receptors, the oiand the orreceptors (e.g., Langer, 1977; Starke, 1977). Since the earlier theories of presynaptic inhibitory o2-receptors it has become apparent that this subclass of receptors is not exclusively presynaptic. For example, a2-receptors are found localized on vascular smooth muscle cells and pancreatic islets, and are also located postsynaptically in the central nervous system (Langer et al., 1985). The actual postsynaptic location of a2-receptors in the central nervous system is uncertain, but the implication is that they are present on postsynaptic cell bodies and dendrites, and mediate functions such as bradycardia and decreasing blood pressure (Langer et al., 1985). However, it has been established that astrocytic glial cells contain a1-adrenergic (Hosli et al., 1982; Hirata et al., 1983; Murphy et al., 1987) and .B-adrenergic receptors (McCarthy, 1983; Baker et al., 1986), and one report has suggested that mouse cortical astrocytes contain specific a2 adrenergic receptors (Ebersolt et al., 1981 ). The aim of the present study was, using brain cell culture techniques, to determine whether specific a2-receptors exist on both neuronal and glial cells and, if so, to compare their properties. The development of differential cell culture techniques has permitted the production in our laboratories of -95% pure astrocytic glial cultures or 85-90% 16

PAGE 25

pure neuronal cultures (the remaining cells being of glial origin) from one-dayold rat brains. This ability allows the properties of the different cell types to be examined in ways which are impossible in whole brain. Furthermore, the results are easier to interpret as the cells are not transformed as many cell lines are, and thus more nearly approximate the in vivo situation. Recent studies from our laboratories have described some of the characteristics of these cultured cells. The glial cultures contain large flat cells in a confluent monolayer (Raizada et al., 1982; Raizada et al., 1987). These cultures, when stained by antibodies against glial fibrillary acidic protein, a specific astrocyte marker, revealed that more than 95% of the cells were astrocytes of both the polygonal and process-bearing varieties (Clarke et al., 1984). The neuronal cultures, when stained with an antibody against neuronal specific enolase, showed that 70-80% of the cells were neurons (Raizada, 1983 ). The remainder were astrocytic glia. The neuronal cultures contain o1-receptors (Feldstein et al., 1986) which are linked to inositol phospholipid hydrolysis (Gonzales et al., 1985), .Bi-receptors which are linked to cAMP production (Baker et al., 1986), norepinephrine and dopamine (Sumners et al., 1983a), epinephrine (Sumners, unpublished data), sodium-dependent norepinephrine and dopamine uptake sites (Sumners and Raizada, 1986; Sumners et al., 1987), monoamine oxidase (MAO) and catecholamine-0-methyltransferase (COMT) activities (Sumners et al., 1987). Glial cultures also contain many of these components, e.g., ,8-adrenergic receptors (Baker et al., 1986), norepinephrine and dopamine ( I 0-20 times less than neurons) (Sumners et al., 1983a), oi-adrenergic receptors (Masters et al., 1988), sodium-independent NE uptake sites (Sumners and Raizada, 1986), and MAO and COMT activities (Hansson and Sellstrom, 1983). 17

PAGE 26

Obviously lacking was information about a2-adrenergic receptors to complete the picture of the catecholamine systems within these cell types. In this study we have biochemically demonstrated the presence of a2-adrenergic receptors in membranes from both neuronal and glial cells in culture, as determined by the characteristics of [3H]-yohimbine binding. This is in contrast to the classical view of exclusive association of a2-receptors presynaptically with neurons in the central nervous system. These studies provide a basis for further regulatory and functional experiments concerning adrenoceptors. Methods Preparation of Neuronal Cultures Neuronal cultures were prepared exactly as described (Sumners et al., 1987) previously. Brains were dissected free from one day old Sprague-Dawley rats and placed in an isotonic salt solution containing I OOU penicillin G potassium, 100 g streptomycin, and 0.25 g amphotericin B (Fungizone) per ml, pH 7.4. All blood vessels and pia mater were removed. The brain was chopped into approximately 2 mm pieces, suspended in 20 ml of 0 25% (wt/vol) trypsin in isotonic salt solution (pH 7.4) and placed in a shaking water bath for 6 mins. at 37 C to dissociate the cells. After this time, 160 g deoxyribonuclease I (DNase I) were added and the cells were shaken for a further 6 mins. Following this, the cells were suspended in Dulbecco's modified Eagle's medium (DMEM) containing 10% plasma-derived horse serum (PDHS), centrifuged at 1000 g for 10 mins. and washed with 50 ml DMEM containing 10% PDHS. Cells were resuspended in DMEM containing 10% PDHS and were plated at a density of l 8x 106 cells/ 100-mm Falcon culture dish for binding studies, and at 3x 106 cells/35-mm dish for the autoradiography experiments and the immunofluorescent 18

PAGE 27

staining. The cells were incubated for 3 days at 37 C in a humidified incubator with 5% C02-95% air. On day 3, cells were treated with 10 M cytosine arabinoside (ARC) prepared in DMEM containing 10% PDHS. This treatment results in an inhibition of cell multiplications, the majority of which are non-neuronal in origin, and provides cultures enriched in neuronal cells. After 3 days, this medium was replaced with DMEM containing 10% PDHS, and the cells allowed to grow for a further 9-10 days before being used in experiments. Preparation of Glial Cultures Glial cell cultures were prepared as previously described (Clarke et al., 1984). The method was as above except that 18xI06 cells suspended in DMEM containing 10% fetal bovine serum (FBS) were plated in 100 mm diameter Falcon tissue culture dishes. After 3 days, the medium was replaced by IO mis fresh DMEM containing 10% FBS. After a further 3 days, cells were passaged at a concentration of Ix I 06 cells/ 100 mm diameter culture dish. The cells were then allowed to grow and divide for a further 20 days prior to use, at which time they were confluent. Preparation of Membranes from Neuronal and Glial Cultures Cells were scraped from the 100-mm culture dishes with the aid of a rubber policeman and placed on ice in a centrifuge tube. They were spun for 10 mins. at 4 C at 1000 g and the supernatant discarded. The pellet was resuspended in phosphate buffered saline (PBS) and either used or frozen in liquid nitrogen until ready for use, preliminary experiments having shown that there was no significant difference between binding in fresh and frozen cells. On the day of the binding experiment, the cells were thawed on ice and then resuspended in 20 volumes of ice-cold 50 mM Tris-HCI, pH 7 .0 with 0 1 % 19

PAGE 28

20 ascorbic acid (binding buffer). They were homogenized using a polytron (setting 2.5 for 30 s.), then centrifuged at 50,000 g for IO minutes at 4 C. The supernatant was discarded, the pellet resuspended in 20 volumes of binding buffer and the centrifugation step repeated. Following this, the pellet was resuspended in 1 ml of binding buff er and the protein content of an aliquot determined by the method of Lowry et al. (1951 ), using bovine serum albumin as the standard, while the remainder were kept on ice. [3HJ-Yohimbine Binding Assays Binding was performed in glass test tubes to which was added 50 l of diluted [3H]-yohimbine (specific activity 80-90 Ci/mmol, New England Nuclear), 50 l of various concentrations of drugs and buff er to bring the volume up to 200 l. Fifty l of the appropriately diluted membrane suspension containing 300 g protein (neuronal cultures) or 175 g protein (glial cultures) were added to start the reaction. Binding was performed at 25 C with shaking for 5 mins and the reaction stopped by rapid filtration under reduced pressure through Whatman GF /B filters. The filters were rinsed 3 times with 3 ml of ice-cold binding buffer, transferred to vials containing 10 ml of Liquiscint, left for 12 hours and then counted in an LKB 1215 liquid scintillation counter at an efficiency of 40-50% for tritium. Non-specific binding was determined in triplicate samples containing 100 M yohimbine, and specific binding was defined as total minus non-specific binding. There was no significant change in the specific binding whether 100 M yohimbine or I M yohimbine was used as the cold displacer. The level of specific binding obtained with neuronal membranes was 40-50% of total, and with glial membranes was 50-80% of total binding.

PAGE 29

21 Kinetic and Pharmacological Analyses of [3H]-Yohimbine Binding Association curves were generated by allowing the binding reaction to proceed for various lengths of time before terminating the reaction by filtration. Dissociation curves were derived by allowing the binding to reach equilibrium, adding yohimbine to a final concentration of I 00 M and filtering at different time points following that addition. Scatchard analysis of saturation experiments were performed on data generated by following binding at concentrations of [3H]-yohimbine from to 24 nM. Competition curves were generated by incubating membranes from neuronal or glial cultures with 10 nM [ 3H]-yohimbine in the absence or presence of various concentrations of agonist or antagonist. Results Kinetic Analysis of [3H]-Yohimbine Binding in Neuronal and Glial Cultures Binding of ( 3H]-yohimbine to membranes prepared from either neuronal or glial cultures was, in both cases, linear with respect to protein concentration. However, it was apparent that glial membranes exhibit higher specific binding of [ 3H]-yohimbine compared with neuronal culture membranes on a per-milligram-protein basis. The binding of [3H]-yohimbine at 25C was rapid, reaching equilibrium by 3 mins. in both neuronal and glial membranes (see representative expts. in Fig. 2), and also in both cases remaining stable for at least 20 mins (data not shown). Thus, in subsequent experiments an incubation period of 5 mins. was used. The binding of ( 3H]-yohimbine to neuronal and glial membranes was also reversible, and the pattern of dissociation in each case can be seen from the representative experiments in Fig. 3. From the association curves the apparent rate constants for association (Kobs) were calculated by linear regression analysis of

PAGE 30

22 In ( Bea ) (Beq -Bt) where Beq is the specific binding at equilibrium and Bt is the specific binding at each time point. From the dissociation curves, the rate constants for dissociation (K_1 ) were calculated by linear regression analysis of In ( :~ ) versus time, where B0 is the amount of [3H)-yohimbine specifically bound at equilibrium, and Bt is the specific binding at various times after the addition of unlabelled yohimbine. The association rate constants (K1 ) for [3H)-yohimbine binding to membranes from both neuronal and glial cultures were derived from the equation Kobs -K-1 Ki = [[3H]-YOH] where K_1 was the mean dissociation constant calculated for that particular culture. For neuronal cultures, K1 was 3.199 0.92 x I06M-1s-1 (n = 4 expts.), and for glial cultures K1 was 0.962 0.418 x I06M-1s-1 (n = 3 expts.). For neuronal cultures, K_1 was 12.96 2.8 x 10-3s-1 (n = 3 expts.) and for glial cultures, K_1 was 17.4 3 x 10-3s-1 (n = 4 expts.). The equilibrium dissociation constants determined from the ratios of K_1/K1 were 4.05 nM for neuronal cultures and 18.06 nM for glial cultures. These data are summarized in Table 2.2.

PAGE 31

23 ( 3 HJ-Yohimbine Binding in Neuronal and Glial Cultures: Saturation Analyses Saturation experiments were performed with I to 24 nM of [ 3 H]-yohimbine. Fig. 4 is a representative experiment which shows that in neuronal cultures, specific binding reached a plateau between 7.5 and 15 nM [ 3 H]-yohimbine. Scatchard (1949) analysis of the data from this experiment using a computer assisted linear regression analysis showed an apparent Ko of 12.59 nM and a maximum number of binding sites (Bmax) of 0.196 pmol/mg protein (see Fig. 4 inset). This experiment has been repeated IO times with similar results and mean K0 and Bmax values are given in Table 2.2. In glial cultures the specific binding of [ 3H]-yohimbine increased with increasing concentrations of [ 3H]-yohimbine up to 24 nM, as seen in the representative experiment in Fig. 5. With concentrations of [ 3H]-yohimbine greater than 24 nM, large variations in specific binding were observed, and so these data were not included in our analyses. Scatchard analysis of the data shown in Fig. 5 revealed a K0 of 16.5 nM and a Bmax of 2.18 pmol/mg protein for [ 3H]-yohimbine binding (see Fig. 5 inset). This experiment has been repeated 10 times with similar results, and mean K0 and Bmax values are shown in Table 2.2. Thus, the K0 is similar in neuronal and glial cultures, but the Bmax is more than an order of magnitude higher in the glia (Table 2.2). Pharmacological Properties of the (3H]-Yohimbine Binding Sites in Neuronal and Glial Cultures The pharmacological properties of the [3H]-yohimbine binding site were examined by comparing the ability of various catecholaminergic drugs to compete for the binding. This was performed by adding increasing concentrations of the catecholaminergic drugs to the incubation as described in the Methods, and IC60 values were calculated by finding the concentration of drug which displaced 50% of the specifically bound [ 3 H]-yohimbine.

PAGE 32

24 A B -C: .!? 30 600 0 ... a. O> E ., -7 0 20 400 "O 5 5 C: :::, iii 4 iii 4 0 .0 I I CT g3 J: 3 ID 0 ~2 :s 2 > 10 CT CT I GI GI 1 ...... 1 J: (') C .E ...... -0 0 0 30 60 90 0 30 60 90 120 TIME (aeca) TIME (aecal 0 0 60 120 180 240 300 60 120 180 240 300 TIME (secs) TIME (secs) Figure 2 Representative experiments show i ng the binding of [3H]-yohimbine to membranes prepared from neuronal or glial cultures as a function of incubation time. Membranes prepared form neuronal (panel A) or glial (panel B) cultures were incubated at 25c in a final volume of 250 l 50 mM Tris HCl/0.1 % ascorbic acid (pH 7 0) containing 10 nM [3H]-yohimbine, in the absence or prese n ce of 100 M y ohimbine. At the indicated time points, the free [3H] y ohimbine was removed from the membrane bound radioacti v it y by washing the membranes three times with ice-cold buffer, and filtration under reduced pressure Non-specific binding was subtracted from the total binding and the data were presented on the y-axis as fmol [3H]-yohimbine specifically bound per mg protein. Each point is a mean of triplicate determinaltions These e x periments were repeated 4 times (neuronal cultures) and 3 times (glial cultures) with similar findings Insets. The inset graphs in panels A and B are kinetic association data from neuronal and glial cultures, respectively. according to a first-order rate equation, and the apparent constants (Kobs) were calculated as detailed in the Results section. analyses of the Data are plotted association rate

PAGE 33

25 40 A 800 B TIME 60 180 300 #-o -2 CD C a, CD -.E 0 ... -4 a. C) E -15 E ..... "C C ::J 0 .D :I: 0 >-I ...... :I: (') ....... 0 0 60 120 180 0 240 300 0 60 300 TIME (secs> TIME (secs) Figure 3. Representative experiments showing the dissociation of bound [3H]yohimbine from membranes prepared from neuronal or glial cultures. Membranes from neuronal (panel A) or glial (panel B) cultures were incubated with lOnM [3H]-yohimbine for 5 mins as detailed in the legend to figure 2. After this time, unlabelled yohimbine was added to each reaction tube so that a final concentration of 100 M of the antagonist was attained. At the indicated time points the membranes form triplicate samples were collected by filtration under reduced pressure, and were washed three times with ice-cold buffer to remove any unbound radioactivity. Data are presented as fmol. [3H] yohimbine remaining bound per mg protein at each time point. The data given at each time point are mean values of triplicate determinations. These experiments were repeated 3 times (neuronal cultures) and 4 times (glial cultures) with similar findings. Insets: The inset graphs in panels A and B are kinetic analyses of the dissociation data from neuronal and glial cultures, respectively. Data are plotted according to a first-order rate equation, and the dissociation rate constants (K_ 1) were calculated as detailed in the Results section.

PAGE 34

26 B/F 40 80 120 160 C Q) 25 B 0 ... a. 200 ..... 15 E 150 0 = "C C 5 100 .D ::I: 0 >50 I I M 4 8 12 16 [ 3H]-YOH nM Figure 4. Binding of [3H]-yohimbine to membranes from neuronal cultures as a function of yohimbine cncentration Neuronal membranes were incubated with increasing concentrations of [3H]-yohimbine in the absence () and the presence (O) of 100 M yohimbine. After incubation for 5 min at 25C, membrane bound radioactivity was collected on glass fiber filters as described in the methods. Specific binding ( ) was determined from the difference between binding in the absence and presence of 100 M yohimbine Data are presented as fmoles (3H]-yohimbine bound per mg protein, being the mean of triplicate determinations. This is a representative experiment which was repeated ten times with similar findings. Inset: The above specific binding data were analyzed accordin~ to Scatchard ( 1949) to determine the apparent dissociation constant (Ko) for [ H] yohimbine binding and the maximum number of binding sites (Bmax). B = specifically bound [3H]-yohimbine in fmoles per mg protein F = free [3H] yohimbine concentration in nM.

PAGE 35

27 8/F 1.51 c "iii 0 ... 0. O> E 8 -15 E 1. 0. -"O C: :::, 0 .0 ::c 0 >-I 0.5 'f M o~ae:::::::::::::=-~---~8.------~11,-2---.1t.::a~--~~--~4 [ 3H]-YOH nM Figure 5. Binding of [3H]-yohimbine to membranes from glial cultures as a function of yohimbine concentration. This experiment was performed as described for figure 4 except that membranes from glial cultures were used instead of membranes from neuronal cultures. (A) total [3H]-yohimbine bound; (O) non-specifically bound [3H]-yohimbine; () specifically bound [3H]-yohimbine. This is a representative experiment which was repeated ten times with similar findings. Inset: Scatchard analysis of the above specific binding data. B = specifically bound [3H]-yohimbine in pmoles per mg protein. F = free [3H]yohimbine in nM.

PAGE 36

28 These IC60 values are summarized in Table 2.1 for both neuronal and glial cultures, and the potency series show that in both cases the [3H]-yohimbine binding is displaced most strongly by drugs which are recognized to be selective for o:2-adrenoreceptors. For example, yohimbine and rauwolscine (o:2-adrenergic antagonists) are much better displacers of the binding of [3H]-yohimbine than prazosin or corynanthine which are selective o:1-antagonists. The same was generally true of the agonists, with clonidine and naphazoline having good displacing ability in glial and neuronal cultures. In neuronal cultures, clonidine had a high affinity for the [3H]-yohimbine binding site, but naphazoline was apparently not so effective. The reason for this is unclear and does not fit well with the rest of the competition data. Overall, these data suggest that the [ 3H]-yohimbine binding site has the characteristics of an o:2-adrenergic receptor. Calculation of Ki values from the competition of [3H]-yohimbine binding with unlabelled yohimbine, using the following equation, IC6o Ki = l + [L]/Ko gave values of 1.4 nM for neuronal cultures and 16.3 nM for glial cultures. These values are shown in Table 2.2 in comparison with the K0 values obtained from both Scatchard and kinetic analyses. Effects of Gpp(NH)p on [3H]-Yohimbine Binding in Neuronal and Glial Cultures Alpha2-adrenergic receptor affinity for agonists is regulated by, among other things, GTP. To determine whether the o:2-adrenergic receptors in neuronal and glial cultures were also sensitive to GTP, agonist and antagonist competition curves were performed in the presence and absence of 50 M 5' guanylyl-imidodiphosphate (Gpp(NH)p), a non-hydrolyzable analog of GTP. Representative experiments are shown in Fig. 6. In membranes from both cell types, Gpp(NH)p caused a decrease in the ability of the agonist NE to

PAGE 37

Treatment Antagonists Rauwolscine Yohimbine Prazosin Corynanthine Propranolol Sulpiride Agonists Naphazoline Clonidine Norepinephrine Epinephrine Dopamine Phenylephrine TABLE 2.1, Displacement of [3H)-Yohimbine Binding By Agonists and Antagonists Neuronal cultures Glial cultures IC60 (M) IC60 (M) 0.0338 0.026 (4) 0.023 0 004 (5) 0 0025 0 0004 (6) 0.024 0 002 (9) 1.07 (2) 0.74 (2) 1.78 0.72 (3) 4 36 2.5 (5) 5.67 2.2 (3) 15.41 1.98 (3) 15.75 (2) 2.69 (2) 0.03 0.016 (3) 0.019 (2) 0.73 0.45 (3) 4.787 0.71 (12) 5.81 1.22 (12) 14.3 8.7 (3) 3.2 1.4 (3) 6 25 (2) 6.25 1.25 (4) 8.5 (2) 17.81 7 35 (4) Data are means SEM of the number of experiments indicated in the parentheses. --= not tested. 29

PAGE 38

K0 (nM)A Bmax TABLE 2.2. [3H]-yohimbine binding in membranes from neuronal and glial cultures: comparison of Kd and Bmax values Neuronal cultures G lial cultures 13.73 1.35 (IO) 18.42 2 34 (IO) 0.143 0.018 (IO) 1.60 0.33 (IO) (pmol/mg protein) K0 (nM)B 4.05 (4) 18.06 K l 1.40 (2) 16.34 A -K0 obtained from saturation experiments/Scatchard analyses. B -K0 obtained from kinetic analyses. (3) (4) 30 Ki -obtained by Cheng-Prusoff transformation of competition data with unlabelled yohimbine. Data are means SEM, and figures in parentheses are the numbers of experiments. compete for the [3H]-yohimbine binding site, as seen by the right hand shift of the displacement curves, but did not affect the affinity of the antagonist, yohimbine, for the site. In neuronal cultures, NE competition curv e IC60 values were shifted from 4 9 M to 70 8 M (n = 2 expts), while in glial cultures the shift was from 6.7 2 M to 118 9.3 M (n = 4 expts). Representative experiments are shown in Fig. 6a,b. For the antagonist, yohimbine, the IC60 values were 2 78 vs 1.75 nM (n = 2 expts) in neuronal cultures and 23.1 1.2 vs 22.6 1.4 (n = 4 expts) in glial cultures, with and without Gpp(NH)p,

PAGE 39

respectively Representative experiments are shown in Fig. 6a, b. GTP was also capable of causing a similar shift but the magnitude of the change was less than that seen with Gpp(NH)p, probably because the GTP was hydrolyzed more rapidly than Gpp(NH)p. The affinity of clonidine another a:2-agonist, was affected similarly by GTP (results not shown) Discussion In this study we have determined the characteristics of a:2-adrenergic receptors in neuronal and glial cells in culture. By differential culture techniques, the preparation of -95% pure astrocytic glial cell cultures and 85-90% pure neuronal cultures is possible. We have identified binding sites for [ 3H]-yohimbine, with characteristics of a:2-adrenergic receptors in both neuronal and glial cultures. The neuronal receptor has similar characteristics to those previously described for [3H]-yohimbine binding in rat brain (Rouot et al., 1982). The a:2-adrenergic receptors of the two cell types appear to have similar characteristics in general, although some features are obviously different between the cell types. For example, the glial cultures contain many more binding sites per milligram protein than the neuronal cultures However, 31 Scatchard analyses revealed that the [3HJ-yohimbine binding sites in glial cultures are of slightly lower affinity than in neuronal cultures. This difference in affinity was enhanced when the K0 values were derived from kinetic or competition analyses (Table 2.2) These differences are not large enough to suggest two different binding sites, but the different cellular environment of each site may cause the change in affinity. The potency order for displacement of [3H]-yohimbine binding by catecholaminergic drugs was similar for neuronal and glial cultures and suggestive of an a:2-adrenergic receptor type. Alpha2 -adrenergic receptors are associated with adenylate cyclase such that their

PAGE 40

100 "cij -0 -75 -;:R., 0 "C C: ::, 0 0 0 32 0 D 0 50 .0 D I 0 > I 25 ,......, I (') ....... 0 12 11 10 9 8 7 6 5 4 3 12 11 10 9 8 7 6 5 4 [DISPLACER] (-LOG M) Figure 6. Competition of [3H]-yohimbine binding in neuronal and glial cultures in the absence and presence of Gpp(NH)p. Panel A: Competition of specific [3H]-yohimbine binding in membranes from neuronal cultures by norepinephrine in the absence () or presence (o) of 50 M Gpp(NH)p, or by yohimbine in the absence ( or presence (0) of 50M Gpp(NH)p. The y-axis is % specifically bound [ H]-y ohimbine, 100% binding being obtained in the absence of an y yohimbine These data are means of triplicate determinations and are representative e x periments Each of the e x periments was repeated twice with similar results. Panel B: As for panel A, but membranes from glial cultures were used instead of those from neuronal cultures. The data are means of triplicate determinations and are representative experiments. These experiments were repeated four times with similar results.

PAGE 41

33 stimulation inhibits the production of cAMP (Jakobs, 1979). The linkage of the a2-receptor whereas those to adenylate cyclase is via the GTP binding protein [Gi], that stimulate adenylate cyclase are linked through the G8 subunit, e.g., ,8-receptors. GTP regulates the affinity of the a2-receptor so that when GTP is present, agonists are less able to bind to the receptor. In both neuronal and glial cells in culture, GTP and its nonhydrolyzable analog Gpp(NH)p were able to cause this shift in agonist affinity, again suggesting that the receptors have a2-adrenergic receptor characteristics. The astrocytic glial cultures contain a large number of [3H]-yohimbine binding sites compared to neuronal cultures. Studies in the whole brain have suggested that the neurotoxin 6-hydroxydopamine, which destroys catecholamine containing neurons, does not abolish all a2-adrenoceptor binding (U'Pritchard et al., 1979). Several explanations of this finding are possible. For example, the a2-receptors are postsynaptic as opposed to presynaptic, or they are present on cells containing other neurotransmitters, e.g., 5HT, ACH (Frankhuyzen and Mulder, 1980; Gothert and Huth, 1980; Vizi, 1972). This study suggests that perhaps it is the glial receptors which are not destroyed by the neurotoxin. However, it may also be true that glial cells in culture express a2-receptors whereas those in adult rat brain do not. Preliminary studies from our laboratory using glia made from the brains of 30-day-old rats, as distinct from newborn rats, suggest that the a2-receptor is present on these glia and of similar characteristics. However, this still does not preclude the fact that the culture technique may cause expression of the receptor. Another major concern arising from these studies has been the possibility that neuronal cells do not possess a2-receptors. The [3H]-yohimbine binding seen in neuronal cultures would, according to this theory, arise from the 10-15%

PAGE 42

34 contamination of neuronal cultures with glial cells. Several findings suggest that this is probably not true. Firstly, we have demonstrated with autoradiography that [3HJ-yohimbine binds to cells which have been defined with immunofluorescent techniques as neurons (see Chapter Three). Secondly, a2-antagonism causes release of the octapeptide angiotensin II from neuronal cultures but not from glial cultures, even though both cultures contain angiotensin II (see Chapter Four). We were also concerned that the site might be a norepinephrine uptake site in glia. To examine this possibility, we performed competition curves of the catecholamine uptake blockers maprotiline and desmethylimipramine (DMI) against [3HJ-yohimbine binding. The IC60 for both drugs was approximately 10 M. At higher concentrations, the membranes clumped and precipitated out of solution. However, as specific a2-adrenergic antagonists and agonists were much more effective at displacing the binding, it seems likely that the binding site for [3HJ-yohimbine is an a2-adrenergic receptor rather than a norepinephrine uptake site. In summary, these experiments show that cultures of -90% pure glia and 85 90% neurons both contain a2-adrenergic receptors and suggest that these receptors are present on both cell types. Glial receptors are of slightly lower affinity but much higher capacity than neuronal a2-receptors. These findings, being the first biochemical characterization of a2-receptors on glia, may help explain some of the difficulty in localizing a2-receptors to pre-or post-synaptic sites in the brain. It also fits well with some known physiology of cultured glial cells, that the inhibition of ,9-adrenergic-stimulated cAMP accumulation is mediated by a-adrenoceptors (McCarthy and deVellis, 1978; VanCalker, 1980), in particular a2-adrenoceptors (Evans et al. 1984 ). Thus glia and neurons in culture may represent a good model system in which to examine a2-adrenergic receptor regulation.

PAGE 43

CHAPTER III LIGHT MICROSCOPIC AUTORADIOGRAPHY OF [3H]-YOHIMBINE BINDING SITES IN NEURONS IN CULTURE Introduction A concern which arose from our studies of [3H]-yohimbine binding sites in neuronal and glial cultures was that neurons in culture might not possess cx:2adrenergic receptors. This concern arose because neuronal cultures contain between l O and 20% glia, and glia in culture contain IO times as many binding sites for 3H-yohimbine. Thus, glial contamination of neuronal culture could theoretically, account for the cx2-binding seen in neuronal cultures. The best method we could think of, to resolve this issue, was to demonstrate by autoradiography that cells with neuronal morphology possessed cx2-binding sites. This was complicated by the fact that [3H]-yohimbine is very lipophilic and easily becomes trapped in cells. For this reason it is not possible to perform binding studies directly in the culture dishes, instead of with membranes, because the level of non-specific binding is very high, usually more than 10% of the ligand concentration. However, by reducing the concentration of [ 3H]-yohimbine and by carefully examining the non-specific binding plates, so that, on examination of total binding plates, patterns of silver grains unique to total binding plates could be recognised, we achieved this goal. Some pictures of immunocytochemically stained cells are included also, to demonstrate the distinctive morphologies of neuronal and glial cell types in culture. 35

PAGE 44

Methods Light Microscopic Autoradiography of rlH]-Yohimbine Binding in Neuronal Cultures Neuronal cultures grown on 35-mm dishes were washed twice with PBS, pH 7.0, and were then incubated with 1 ml PBS containing 1 nM [3H]-yohimbine for 5 mins at 22c. For the non-specific binding dishes, parallel incubations were made using 100 M yohimbine added to the reaction mixture. Following the incubation, the reaction mixture was removed, dishes were placed on ice and the cells washed twice with ice-cold PBS. mins at 4C with 3.7% glutaraldehyde. The cells were then fixed for 30 After fixation the cells were washed twice with PBS dehydrated through graded ethanol solutions (30-100%, 5 mins at each concentration) and dessicated at room temperature for 2 hours. In the dark, dishes were coated with Ilford K-5 nuclear emulsion, diluted 1:1 with distilled water containing 2% glycerol, according to Rogers (I 979). After 2 mins, excess emulsion was removed, and the dishes were kept at 4C for 20 mins, followed by 60 mins at room temperature. The dishes were dessicated overnight and then stored in light-tight boxes in the refrigerator until development, usually 3-4 months later. For development, Kodak D-19 (diluted 1:1 with water) was added to each dish for 2 mins. After this dishes were washed with distilled water and then fixed for 5 mins with sodium thiosulfate The dishes were finally washed for several minutes with distilled water, and examined by phase-contrast microscopy (Zeiss). Immunofluorescent Staining of Neuronal Cultures Growth media were removed from neuronal cultures grown on 35-mm dishes, which were then washed three times with PBS, pH 7.4. Following this, cells were fixed and made permeable by the addition of 2 ml of 100% methanol at -20c methanol (100%) to each dish for 7.5 mins. After removal of the methanol, 2 ml PBS were added to each dish. After 5 mins the PBS was 36

PAGE 45

37 removed and one drop (approx. 60 l) of either the monoclonal neurofilament antibody NE-14 (1:20 dilution (Shaw et al, 1985) or glial fibrillary acidic protein antibody (GFAP: 1:20 dilution) was added to the center of each dish. The dishes were incubated at 37c for 30 mins and then the antibody was removed by three washes with PBS (pH 7.4). Next, one drop of a secondary antibody (goat anti-mouse antibody conjugated to fluorescein (FITC)) was added to the same area of each dish as the primary antibody, and the dishes underwent a further incubation of 37c for 30 mins. After this, cells were washed three times with PBS (pH 7.4), and 2 drops of mounting media (90% glycerol with l mg/ml paraphenylenediamine in 20 mM Tris HCl pH 8.0) were added to the stained area followed by a coverslip. The cells were viewed using an inverted fluorescence microscope (Zeiss). Results Prior studies using antibodies against neuron specific enolase determined that our neuronal cultures contain approximately 70-80% neurons, with the remainder of the cells being glia (Raizada et al, 1983). Considering this fact, and also that glia contain many specific binding sites for (3H]-yohimbine, it was important to determine whether the neuronal cells actually contain ( 3H]yohimbine binding sites. We approached this problem by using light microscopic autoradiographic analysis of [3H]-yohimbine binding in neuronal cultures. Fig. 7 contains three representative autoradiograms showing the binding of [ 3H]-yohimbine to neuronal cultures. In figs 7a and 7b the arrows indicate the position of silver grains in the emulsion overlying cells with neuronal features. This specific association of silver grains with neuronal cells was observed at the level of cell bodies, neurites and axon terminals. Fig 7c contains an autoradiogram taken from a dish in which unlabelled yohimbine was added to the incubation mixture. No specific localization of silver grains to neuronal

PAGE 46

Figure 7. Light microscopic autoradiographs showing [3H]-yohimbine binding in neuronal cultures. A. Autoradiographic analyses of [3H]-yohimbine binding in neuronal cultures were performed as detailed in the Methods section. The arrows indicate high densities of silver grains (shown in black) associated with cells of neuronal morphology. In these cases the silver grains were associated specifically with the neuronal cell bodies and neurites. Bars 10 38

PAGE 47

39 Figure 7 Light microscopic autoradiographs showing [3H]-yohimbine binding in neuronal cultures B. Autoradiograph of neuronal cultures incubated with [3H]-yohimbine. Arrows indicate high densities of silver grains (shown in black) associated with cells of neuronal morphology. In this case the silver grains were associated specifically with the neuronal cell bodies and neurites. Bars IO

PAGE 48

40 Figure 7. Light microscopic autoradiographs showing [3H]-yohimbine binding in neuronal cultures. C. Autoradiograph of neuronal cultures incubated with [3H]-yohimbine in the presence of 100 M unlabelled yohimbine. No specific localization of silver grains to cells of neuronal morphology was noticed Bars IO

PAGE 49

cells can be observed in this autoradiogram, nor in other dishes treated in the same way. 41 To confirm that the cells to which silver grains are specifically localized are in fact neurons, immunofluorescent staining of our neuronal cultures was performed using antibodies against neuronal and glial proteins. With the use of the monoclonal neurofilament antibody NE-14, we have obtained specific staining of neuronal cells in our neuronal cultures. Figs. 8a and 8b show phase contrast and fluorescent micrographs, respectively, taken from the same field of neuronal cultures stained with NE-14. The arrow indicates specific immunofluorescent staining associated with neuronal cells. These cells are similar morphologically to those observed with the specific silver grain localization in figs. 8a and 8b. Not all of the neuronal cells in fig. 8b appear stained because this antibody recognizes only heavy type neurofilaments, which only appear at late stages of neuronal development (Shaw et al., 1985). In our neuronal cultures the neuronal cells actually overlie the glial cells, and this can be clearly observed in figs. 8c and 8d. These figures are representative phase contrast and fluorescence micrographs taken from neuronal cultures stained with anti-GFAP. In the phase contrast picture (8c) the neuronal cells are clearly evident, whereas cells of typical glial morphology are not easily observed. However, in the fluorescence micrograph in fig. 8d, stained glial cells are easily seen. These cells bear no morphological relation to the neuronal cells which specifically localize silver grains representing [3H]-yohimbine binding sites in figs. 7a and 7b. In addition, immunofluorescent staining analyses using the GFAP and NE-14 antibodies, and other antisera to neurofilament proteins and as NF-1 and DA2Bl (Shaw et al., 1985) have enabled us to show that our neuronal cultures in fact contain 85-90% neuronal cells, a greater number than originally estimated (Raizada et al., 1983 ).

PAGE 50

42 Figure 8. Immunofluorescent staining of neuronal cultures. Immunofluorescent staining procedures using antibodies to a neurofilament protein (NE-14) antibody) and GFAP were performed as detailed in the Methods section. A, C are phase-contrast micrographs of cell fields from neuronal culture dishes. Cells and cell-bundles of typical neuronal morphology are easily observed (arrows). Cells with glial morphology are not easily seen, because they underlie the neuronal cells and are thus out of focus in most cases. B is a fluorescence micrograph from the same cell field as in A, showing the NE-14/FITC staining. This staining is clearly associated with the neurofilaments in the cells which have neuronal morphology. D is a fluorescence micrograph from the same cell field as in C, showing the GF AP /FITC staining. This staining is only associated with cells of typical glial morphology, and not with the neuronal cells. The GFAP/FITC-stained astrocytes are mostly not observed in the phase-contrast micrograph C.

PAGE 51

43 Figure 8. Immunofluorescent staining of neuronal cultures. B. Fluorescent micrograph from the same cell field as in A, showing the NE-14/FITC staining. The staining is clearly associated with the neurofilaments in the cells which have neuronal morphology. Bars 10

PAGE 52

44 Figure 8. Immunofluorescent staining of neuronal cultures. C. Phase-contrast micrograph of a cell field from a neuronal culture dish. Cells and cell-bundles of typical neuronal morphology are easily observed (arrows). Cells with glial morphology are not easily seen, because they underlie the neuronal cells and are thus out of focus in most cases. Bars l O

PAGE 53

45 Figure 8 Immunofluorescent staining of neuronal cultures. D is a fluorescence micrograph from the same cell field as in C, showing the GF AP /FITC staining. This staining is only associated with cells of typical astrocytic glial morphology, and not with the neuronal cells. The GFAP/FITCstained astrocytes are mostly not observed in the phase-contrast micrograph C Bars 10

PAGE 54

Discussion Neuronal cells in culture possess oc:2-binding sites, because silver grains localize to cells with neuronal morphology. The grains could be seen associated with cell bodies, fibre tracts and junctions between fibre tracts. At the level of magnification used it was impossible to determine whether the grains that were associated with cell bodies were on dendrites or not. The silver grains associated with fibres were often seen in varicosities, which in peripheral sympathetic nerves at least are sites of synapses at intervals along the axon. This suggests that a similar pattern of synapses occurs centrally. In conclusion, these data support the suggestive findings of the characterization studies, that there are indeed neuronal oc:2-adrenergic receptors. Thus both glia and neurons possess oc:2-adrenergic receptors when in culture. 46

PAGE 55

CHAPTER IV ADRENERGIC MECHANISMS MEDIA TING ANGIOTENSIN II RELEASE. Introduction Angiotensin II has been known to have specific actions on the central nervous system since the cross-perfusion studies of Bickerton and Buckley in dogs (1961). It is now well accepted that there is a brain renin-angiotensin system, independent of the peripheral system (e.g. Campbell et al., 1984, Phillips and Stenstrom, 1985, Phillips, 1987). However, the nature of the action of angiotensin II is not well characterized. It has recently been demonstrated that the release of angiotensin II from brain tissue and from brain cell cultures can be stimulated by depolarizing conditions (Meyer and Weyhenmeyer, 1986, Schiavone et al., 1986), suggesting that angiotensin II may be a neurotransmitter. It has also been postulated that angiotensin II acts as a neuromodulator in the brain because of its interactions with neurotransmitters, especially the catecholamines. This study deals with the ability of catecholamines to control the release of angiotensin II from neurons in culture. We were interested in this because of the demonstration in brain cell culture and in vivo and in vitro of close interactions between catecholamines and angiotensin II (e.g Garcia-Sevilla et al., 1979; Huang et al., 1987; Meldrum et al., 1984; Sumners and Phillips, 1983; Sumners and Raizada, 1986; Schacht, 1984) In neuronal cell cultures, norepinephrine, acting on oc:1-recept ors, modulates the expression of angiotensin II receptors (Sumners and Raizada, 1984; Sumners, Watkins and Raizada, 1986). Additionally, angiotensin II influences both the uptake and metabolism of norepinephrine in neuronal cultures (Sumners and 47

PAGE 56

Raizada, 1986; Sumners et al, 1987). It was of interest to us to discover whether cx2-adrenergic receptors could modulate angiotensin II release. This question was asked because of the clear interactions between the two systems and because cx2-receptors modulate the release of a number of different transmitters, including norepinephrine, serotonin, and acetylcholine (Frankhuyzen and Mulder, 1980 Gothert and Huth, 1980, Vizi and Knoll, 1971 ). Thus, 'r adrenergic receptors might interact with angiotensin II in the same fashion. 48 We demonstrate here that release of angiotensin II from neuronal, but not astrocytic glial, cultures is stimulated by catecholamines. Our data suggest that two mechanisms exist for the catecholaminergic modulation of angiotensin II release, one which involves cx2 and the other .B-adrenergic receptors. Methods Preparation of Neuronal Cultures Cultures were prepared as described in Chapter Two. Protocol for Angiotensin II Release Experiments Ten neuronal or ten glial dishes were used for each experimental group, and the procedure for a control release experiment was as follows. The growth media were aspirated and the cells washed three times with IO ml phosphate buffered saline (PBS), pH 7.4 per wash. Four ml PBS containing 0.75 mM CaC12 0.75 mM MgC12 and 33 mM glucose (release buffer) were added to each dish and the cells were incubated for 5 min at 37C in an incubation bath. After 5 min the medium was removed and placed in a plastic test tube containing 0.2 ml of concentrated HCl, mixed and stored in the refrigerator. This incubation procedure was repeated for a further three 5-min. periods. For d r ug treatments the protocol was exactly the same except that during the second five-minute incubation period the cells were exposed to different adrenergic agonists or antagonists.

PAGE 57

49 In the first set of experiments, we examined whether yohimbine (50 M) could alter angiotensin II release from either neuronal or astrocytic glial cultures This was carried out as above with yohimbine being included in the second incubation period In the second set of experiments, we determined the effects of increasing concentrations of yohimbine (0.1, 1 0, 10.0 and 50 M) on angiotensin II release from neuronal cultures In the third set of experiments, we examined the effects of increasing concentrations of norepinephrine (0. 0 l, 0.1, 1 and 50 M) on angiotensin II release from neuronal cultures. In the fourth set of experiments, the effects of norepinephrine (50 M) on angiotensin II release in neuronal cultures were examined in the absence and presence of prazosin (0.1 -10 M). In the fifth series of experiments, we examined the effects of norepinephrine (50 M) on angiotensin II release in neuronal cultures in the absence or presence of DL-propranolol (1 -100 M). In the sixth set of experiments, we determined the effects of the /3adrenergic agonist, isoproterenol, (1 and 100 M) on angiotensin II release from neuronal cultures. Drug Solutions and Incubations Catecholamines agonists and antagonists were dissolved as follows. Norepinephrine and isoproterenol were dissolved in release buff er containing 10 M L-ascorbic acid to an initial dilution of 10 mM. These solutions were then diluted to the required concentrations using release buff er containing 10 M L-ascorbic acid. In experiments where norepinephrine or isoproterenol was used, the control incubations were always performed with release buff er containing 10 M L-ascorbic acid Both prazosin and yohimbine were initially

PAGE 58

dissolved in distilled water to a concentration of I mM, and DL-propranolol in release buffer to a concentration of IO mM. All of these antagonists were diluted to the required concentrations using release buffer. All drugs were prepared fresh for each experiment. Preparation of Incubation Media and Cell Samples for Radioimmunoassay Prior to the radioimmunological determination of angiotensin II in the incubation media, the fractions obtained from the four incubation steps were purified and concentrated on octadecasilyl-silica cartridges (Sep Pak C18 ) according to a procedure described earlier (Phillips and Stenstrom, 1985). The purified samples then underwent radioimmunoassay for angiotensin II, as detailed previously (Hermann et al., 1988b, Phillips and Stenstrom, 1985). After the fourth incubation step, angiotensin II was extracted from neuronal and glial cells by the addition of 1 ml of 1 M acetic acid to each dish. Cells were removed from each dish, combined in a plastic tube and boiled for five min., homogenized ultrasonically for 10 sec and then centrifuged at 2,700 g for ten minutes. The supernatant was lyophilized, dissolved in 0.5 ml of 0.05 M Tris/HCI buffer, pH 7.4, spun at 10,000 x g for 2 min. at room temperature and 0.05 ml of the supernatants were subjected to radioimmunological measurement of angiotensin II, as detailed by us previously. HPLC Analysis of Released Material To confirm the authenticity of the released material as angiotensin II, the following procedure was followed. After direct radioimmunological measurement of angiotensin II in the buffer from the incubation periods, the remainder of the control samples as well as the yohimbine-treated samples from different experiments were pooled separately, frozen and lyophilized. The dry residues were dissolved in 1.0 ml HPLC solvent A (methanol:H20: I M ammonium acetate buffer, pH 5.4; 30:69:1, vol/vol), centrifuged at 10,000 x g for 2 min at room 50

PAGE 59

temperature and subjected to HPLC analysis as detailed previously (Hermann et al., 1988b, Phillips and Stenstrom, 1985.). About 0.9 ml of the reconstituted samples was injected into the HPLC and I-ml fractions were collected. The solvent was removed from these fractions in a stream of air at 40C. The dried samples were dissolved in 0.5 ml distilled water, frozen and lyophilized. The freeze-dried samples were dissolved in 0.5 ml 0.05 M Tris/HCI buff er, pH 7.4, and 0.1 ml aliquots from each fraction were taken for the radioimmunological measurement of angiotensin II, as detailed previously (Hermann et al. 1988b). Results are presented as means SEM of the amounts of angiotensin II released, expressed as a percentage of the total angiotensin II content (total angiotensin II = that angiotensin II measured in the cell extracts plus that in the media of each release period). Results Effects of Yohimbine on Ang II Release from Cultured Brain Cells In neuronal cultures the mean baseline secretion of angiotensin II-like material when cells were incubated with buff er alone was 43.65 7 .44 pg per 5 min. incubation period (n = 14 expts.; 52 individual determinations). In the first set of experiments, neuronal cultures were treated with yohimbine (50 M, 5 min.) as detailed in the methods. This treatment resulted in significantly increased release of angiotensin II-like material, i.e. 37 .87 11. 7% (n = 3 expts.) of the total neuronal angiotensin II was released in the second incubation period compared with 5.15 1.32% (n = 3 expts.) in the first (buffer-treated) incubation period. This is illustrated in Figure 9, which shows a representative experiment where neuronal cultures were treated with yohimbine in the second incubation period. It can be seen clearly that addition of yohimbine is accompanied by a large increase in the release of angiotensin II-like material. 51

PAGE 60

-..J ?fl. -w en <( w ..J w a: (!) z <( 52 60 40 20 0 0 5 10 15 20 Figure 9. release of cultures. INCUBATION TIME (mins) Representative experiments showing the effects of yohimbine on the angiotensin II-like material from neuronal and astrocytic glial Release of angiotensin II-like material was analyzed during four incubation periods of five minutes each, as detailed in the Methods The first, third and fourth incubation periods were with buff er alone, and the second period was with yohimbine (50 M) (), neuronal cultures; (0), astrocytic glial cultures. Angiotensin II release during each period is expressed as a percentage of the total (cell plus incubation media) angiotens i n II content.

PAGE 61

53 .-.. ...J 60 cfl. ---w en 40 <( w ...J -w --a: 20 C, z <( 0 I CON 0.1 1 10 50 [YOH] uM Figure 10. Effects of different concentrations of yohimbine on the release of angiotensin-II like material from neuronal cultures. Release of angiotensin II-like material was analyzed during four incubation periods of five minutes each, as detailed in the Methods. In control (CON) incubations, cells were exposed to buff er alone for each incubation period. In yohimbine (YOH) -treated cultures, cells were exposed to buff er for periods one, three and four, and to YOH (0.1 -50 M) in period two only. In all cases, the data shown represent the summed angiotensin II released during periods two and three, expressed as a percentage of the total (cell plus incubation media) angiotensin II content. Data are means SEM of four individual experiments (40 total dishes per data point). (*) Significantly different from control release

PAGE 62

fmd Ang II/ fraction 10 CONTROL 50~ 10 YOHIMBINE 50 Ule5)-Ang II %MeOH -'.100 50 100 ~50 0 ----------00!,---~----,1""'0-~-~2c'=o:---~---::3a:-'.o.--~--4-,-'.o---~--50 Time (min) Figure 11. HPLC analysis of angiotensin II-like material released from neuronal cultures. Incubation media containing the released angiotensin II-like material were prepared as detailed in the Methods, and injected into the HPLC, which utilized a reverse-phase C1g column and a methanol gradient for separations. Upper panel: Chromatogram obtained from yohimbine (YOH 50 M)-treated cells. In both cases, a peak is observed at 26-27 mins. retention time, similar to the retention time of authentic Ile5-angiotensin II (arrow). This peak is larger in the yohimbine treated group. 54

PAGE 63

55 After removal of the yohimbine, release of angiotensin II-like material returns to the same levels as in the first period by the fourth (buffer-treated) incubation period. However, it is apparent that a significant amount of angiotensin II-like material is released during the third incubation period immediately following the yohimbine treatment period. For this reason, we have combined the amounts of angiotensin II released in incubation periods two and three in the later dose-response studies, and the results are presented as such. In astrocytic glial cultures, the basal release of angiotensin II-like material was 21.76 5.7 pg per 5 min. incubation period (n = 8 expts.; 24 individual determinations) when cells were incubated with buffer alone. Treatment of astrocytic glial cultures with yohimbine (50 M, 5 min.) as detailed in the methods caused no significant changes in the release of angiotensin II-like material from these cells, compared with buffer treatments (n = 5 experiments). This is also illustrated in Fig. 9, which also shows a representative experiment where glial cultures were treated with yohimbine during the second incubation period. In the second set of experiments, neuronal cultures were incubated with different concentrations of yohimbine (0.1 -50 M, 5 min.), as detailed in the methods. This resulted in a concentration-dependent increase in the release of angiotensin II-like material compared with identical cultures treated with buffer (i.e. controls). From Fig. 10 it can be seen clearly that a significant increase in the release of angiotensin II-like material can be obtained with 1 M or higher concentrations of yohimbine, and that the effect starts to plateau around 50 M of the cx:2-antagonist. The released angiotensin II-like material was subjected to HPLC analysis as detailed previously in order to establish its authenticity. Fig. 11 (upper panel) shows a chromatogram obtained from the injection of released material from

PAGE 64

control (buffer-treated) cells into the HPLC, and a small peak of angiotensin II can be seen at 26-27 min. retention time (note: the arrow indicates the retention time for authentic Ile5-angiotensin II). It is also apparent that some other unidentified substances cross-react with the angiotensin II antibody. Fig 11 (lower panel) shows a chromatogram obtained from the injection of released material from yohimbine-treated (50 M, 5 min.) cells into the HPLC. Again, a peak of angiotensin II can be seen at 26-27 min. retention time, the same as for authentic Ile5-angiotensin II (arrow). In addition, it can be seen that the peak is larger, i.e. there is a greater amount of angiotensin II released, in the yohimbine-treated compared with control cells (Fig. 11 ). Effects of Catecholamine Agonists on Angiotensin II Release from Neuronal Cultures To determine whether oc2-adrenergic blockade by yohimbine was acting directly to release angiotensin II, or indirectly via stimulating norepinephrine release, the third set of experiments was performed. In this set of experiments, neuronal cultures were incubated with varying concentrations of norepinephrine as detailed in the methods, and release of angiotensin II-like material was determined. Initial experiments showed that norepinephrine was able to cause increased release of angiotensin II-like material from neuronal cultures, compared with buff er treatment. As with the yohimbine treatments, most of the angiotensin II is released during the second incubation period when norepinephrine is present. However, a significant amount of angiotensin II is released in the third incubation period before control levels of release are attained once more. Therefore, in the dose-response studies with norepinephrine we have combined the amounts of angiotensin II released in incubation periods two and three, and the results are presented as such. Incubation of neuronal cultures with norepinephrine (0.01 50 M, 5 min.) 56

PAGE 65

57 resulted in concentration-dependent increases in the release of angiotens i n IIlike material (see Fig 12) compared with identical cultures treated with buffer (i.e. controls). From Fig. 12 it can be seen that a significant increase in angiotensin II release is obtained with as little as 10 nM norepinephrine, and that a maximal effect is reached between 100 nM and l M norepinephrine. At the higher concentration of 50 M norepinephrine, the response is less, but is not significantly different from that obtained with I M norepinephrine. In order to determine which type of adrenergic receptor was being activated by norepinephrine to stimulate release of angiotensin II-like material the fourth and fifth sets of experiments were performed. In the fourth set of experiments, neuronal cultures were incubated with norepinephrine (50 M, 5 min.) in the absence or presence of the cx:1-antagonist prazosin (1 or 10 M, 5 min.), as detailed in the methods, and release of angiotensin II-like materia l was analyzed The results in Fig. 13 clearly show that prazosin does not inhibi t the norepinephrine-stimulated increase in angiotensin II release from neuronal cultures, either at I or 10 M. The concentrations of prazosin used here are sufficient to cause complete blockade of cxi-adrenergic receptors and to completely inhibit norepinephrine-induced inositol phospholipid hydrolysis in neuronal cultures (Feldstein et al., 1986, Gonzales et al., 1987). However, these experiments can not be considered to be completely conclusive, as prazosin may not have been able to completely block the effects of this concentration of norepinephrine. In the fifth set of experiments, neuronal cultures were incubated with norepinephrine (50 M, 5 min ) in the absence or presence of the general /3-antagonist DL-propranolol (1-100 M, 5 min.). However, DL-propranolol exhibits significant cross-reactivity with our angiotensin II antibody, and so the results

PAGE 66

'?fl. -w en <( w ..J w a: (!) z <( 60 40 20 CON 0.01 0.1 1 50 [NE] uM 58 Figure 12. Effects of different concentrations of norepinephrine on the release of angiotensin II-like material from neuronal cultures Release of angiotensin II-like material was analyzed during four incubation periods of five minutes each, as detailed in the Methods. In control (CON) incubations, cells were exposed to buff er alone for each incubation period. In norepinephrine (NE)-treated cultures, cells were exposed to buffer for periods one three and four, and to NE (0.01 50 M) in period two only. In all cases, the data shown represent the summed angiotensin II released during periods two and three, expressed as a percentage of the total (cell plus incubation media) content of angiotensin II. Data are means SEM of four individual experiments ( 40 total dishes per data point) (*) significantly different from control values.

PAGE 67

w en 80 60
PAGE 68

';fl ....,_.. w20 en <( w _. 10 C, 60 0---~-,.~/__.__L.-~-----CON 1 100 [ISOP] uM Figure 14. Effects of isoproterenol on the release of angiotensin II-like material from neuronal cultures. Release of angiotensin II-like material was analyzed during four incubation periods of five minutes each, as detailed in the Methods. In control (CON) incubations, cells were exposed to buff er alone for each incubation period. In isoproterenol (ISOP)-treated cultures, cells were exposed to buff er for periods one, three and four, and to ISOP (I and 100 M) in period two only. In all cases the data shown represent the summed angiotensin II released during incubation periods two and three, expressed as a percentage of the total (cell plus incubation media) angiotensin II content. Data are means SEM of two to four experiments (20 40 dishes per data point). (*) Significantly different from control release.

PAGE 69

61 from this series of studies were meaningless. In an attempt to circumvent this problem, we performed the sixth set of experiments, in which the effects of a pure .B-adrenergic agonist, isoproterenol, were tested on angiotensin II release Incubation of neuronal cultures with isoproterenol (I or I 00 M, 5 min.) resulted in increases in the release of angiotensin II-like material (see Figure 14) compared with identical cultures treated with buffer (controls). From Figure 14 it can be seen that 100 M isoproterenol induces an increase of -17% in the release of angiotensin II-like material. This dose of isoproterenol elicits a large increase in .B-receptor-mediated cyclic AMP production in neuronal cultures (Baker et al., 1986) Discussion In the present study we have determined that catecholaminergic systems can influence the release of angiotensin II-like material from neuronal, but not from astrocytic glial cultures. Firstly, we have shown that release of angiotensin II-like material from neuronal cultures is stimulated by blockade of cx:2-adrenergic receptors with yohimbine This effect of yohimbine is quick starting within a five-minute incubation period, and is also dose-dependent. The discovery that cx:2-adrenergic receptors are able to modulate the release of neuronal Ang II-like material is not an unreasonable finding, because these receptors have previously been shown to modulate the release of a number of different neuroactive substances, e .g. norepinephrine, serotonin and acetylcholine (Frankhuyzen and Mulder, 1980, Gothert and Huth, 1980 and Vizi and Knoll, 1971 ) In addition, there is good anatomical evidence for the co-localization of cx:2-adrenergic receptors with angiotensin II-immunoreactive material in the central nervous system (Boyajian et al 1987 Unnerstall et al. 1984). Thus, our studies suggest that cx:2-adrenergic receptors are able to modulate release of neuronal angiotensin II-like material. This modulation by cx:2-adrenergic

PAGE 70

62 receptors appears to be unique to neurons because no effects of yohimbine were observed in astrocytic glia. This is perhaps surprising because cultured astrocytes contain angiotensin II-like material (Hermann et al., 1988b) and also large numbers of specific cx:2-adrenergic receptors (Richards et al., 1986). In an effort to determine whether the release of angiotensin II following stimulation of cx:2-adrenergic receptors was direct or indirect, the effects of catecholamines and catecholamine agonists upon the release of neuronal angiotensin II-like material were determined. As stated previously, cx:2 adrenergic receptors modulate release of neuronal norepinephrine, and the cx:2 -blocker yohimbine should be expected to increase norepinephrine release (Langer, 1981 ). Therefore, perhaps the effects of yohimbine on the release of angiotensin II-like material are mediated via release of norepinephrine, and this is a reasonable suggestion because our cultures contain (Sumners et al., 1983a), and are able to release, catecholamines. For example, in preliminary experiments we have shown that in neuronal cultures preloaded with L-[3H]norepinephrine (0.2 M, 60 min, 37C) in the presence of 100 M pargyline, A23 l 87 (I M) stimulates release of 22.4 5.2% (n = 9 dishes) of the total [ 3H]-norepinephrine, compared with 2.7 1.9% (n = 9 dishes) of the total [ 3H]-norepinephrine in the absence of this ionophore (Sumners, unpublished observations). norepinephrine We have determined that incubation of neuronal cultures with induces a dose-dependent angiotensin II-like material, antagonist, prazosin. an effect Additionally, not the increase in the release of inhibited by the cx:1-adrenergic pure ,8-adrenergic antagonist, isoproterenol, induced small but significant increases in the release of neuronal angiotensin II-like material. Again, no release of angiotensin II occurred from astrocytic glial cultures (data not shown). These results might suggest that yohimbine induces increased release of norepinephrine, which then acts at ,8-

PAGE 71

63 adrenergic receptors to increase release of angiotensin II-like material. However, we were not able to confirm this mechanism with the use of /3-adrenergic blockers because these drugs interfere with our radioimmunoassay for angiotensin II-like material. Also, if this were the mechanism for catecholamine regulation of angiotensin II release, it might be expected that a /3-adrenergic agonist, such as isoproterenol, would be a far more potent stimulator of angiotensin II release than norepinephrine. This is not the case, as observed from our results. Therefore, other possibilities exist for the regulation of release of angiotensin II-like material from neurons. One is the existence of distinct populations of neurons, one which releases angiotensin II in response to cx:2-antagonism and another which releases angiotensin II following /3-adrenergic stimulation. This may be a more likely explanation as the different treatments (i.e., with isoproterenol or norepinephrine) did not induce the same level of release of angiotensin II, although it may just be difficult to compare doses of the different adrenergic drugs. Although we have clearly shown that catecholamine systems are able to regulate the release of angiotensin II-like material from cultured neurons, other studies on the effects of catecholamines on angiotensin II release have given varying results. For example, neither isoproterenol nor DL-propranolol alter the release of angiotensin II-like material from the anterior hypothalami of rats implanted with push-pull cannulae (Brosnihan et al., 1988). However, the doses of isoproterenol used were lower than in the present study, and our culture preparation is not restricted to the anterior hypothalamus. These data would, in fact, respond support our above suggestion that selected to ,B-adrenergic stimulation by releasing populations of neurons angiotensin II. Other investigations have shown that Ang II is released from vascular (subendothelial)

PAGE 72

64 tissue by ,8-adrenergic stimulation (Gothert and Kollecker, 1986, Nakamura et al, 1986a), data which support our current findings. Questions remain as to the physiologic role of regulation of neuronal angiotensin II release by catecholamine systems. In vascular tissue the angiotensin II released by ,8-adrenergic stimulation facilitated neurotransmission at the sympathetic nerve terminal (Nakamura et al, 1986b ). The ability of angiotensin II to facilitate sympathetic neurotransmission is well established (e.g Langer, 1981, Starke et al., 1970, Story and Ziogas, 1987, Zimmerman and Whitmore, 1967). exist in the neurotransmission. facilitates both norepinephrine in Meldrum et al., Our current findings suggest that a similar mechanism may central nervous system to enhance norepinephrine This suggestion is supported by the fact that angiotensin II the electrically evoked and potassium-induced release of the brain (Garcia-Sevilla et al., 1979, Huang et al., 1987, 1984 Schacht, 1984) and that the angiotensinergic pathway linking the subfornical organ to the nucleus medianus (Lind et al., 1984, Lind et al., 1985, Weyhenmeyer and Phillips, 1982) responds to the application of norepinephrine with excitation (Graham et al., 1985). A possible explanation is that norepinephrine increases the release of angiotensin II, which in turn leads to neurophysiological excitation, angiotensin II having been shown to stimulate these cells directly (Phillips and Felix, 1978). It is also interesting to speculate on how (or whether) adrenergic regulation of release of angiotensin II-like material from neurons is related to release stimulated by depolarization. It has been reported that potassiuminduced depolarizations induce release of angiotensin II-like material from hypothalamus/neurohypophyseal slices (Schiavone et al., 1986) and from mixed (neurons plus glia) brain cell cultures (Meyer and Weyhenmeyer, 1986) Perhaps

PAGE 73

65 depolarization-induced release of angiotensin II-like material is dependent upon prior release of norepinephrine. Overall the current findings strengthen our growing conviction that there are extremely close ties between the angiotensin II and catecholamine systems in the brain as modeled in neuronal cell cultures. Incubation of neuronal cultures with Ang II increases the norepinephrine content of both cells and media (Sumners et al., 1983b ). It has a biphasic effect on norepinephrine uptake, enhancing after short exposure and inhibiting after longer exposure the neuronal uptake of norepinephrine (Sumners and Raizada, 1986). Norepinephrine applied to the cells causes an cxcadrenergic-mediated down-regulation of angiotensin II receptors (Sumners, Watkins and Raizada, 1986). This downregulation could be secondary to the increased angiotensin II release demonstrated in this study. The data suggest that norepinephrine-induced release of angiotensin II is a requisite for a feedback system in neurons by which angiotensin II enhances the neuronal effects of norepinephrine which in turn through cx2 and ,8-adrenergic receptors increases angiotensin II actions. In summary, our results show that the release of angiotensin II-like material from neuronal cultures is modulated by catecholamines, mediated by either cx2 or ,8-adrenergic receptors or both. The exact mechanisms involved, topography of the different catecholamine-induced effects, and relation to depolarization-induced release have yet to be determined.

PAGE 74

CHAPTER V EFFECT OF VARIOUS ADRENERGIC DRUGS ON THE RENIN CONCENTRATION OF BRAIN CELLS IN CULTURE Introduction We have demonstrated that adrenergic drugs could influence the release of angiotensin II from neurons but not glia in culture. We suspected that the ability of these drugs to release angiotensin II might induce increased synthetic demands, which would be revealed by increased renin concentrations in these cells. Renin is the enzyme which is responsible for cleaving angiotensin I from angiotensinogen. It is the rate limiting step in the formation of angiotensin II in the blood, and therefore, we felt, might be tightly regulated in cultured cells. We expected that those drugs which increased the release of angiotensin II from neuronal cultures would also increase the concentration of renin in these cultures. Methods Sample Preparation Renin, the enzyme which converts angiotensinogen to angiotensin I is measured by adding excess angiotensinogen (substrate) to the tissue preparation being studied, allowing the reaction to proceed to completion and then measuring the angiotensin I generated by RIA. The experiment was initiated by incubating the cells with catecholamine solutions, (0.1 to 50 M yohimbine, 0 1 M to l mM norepinephrine, IO to 50 M clonidine, or with vehicle (0.1 % ascorbic acid for 4 or 24 h )). Cells were washed three times with PBS (pH 7.2) and were scraped from the dish in 1 ml PBS with a rubber policeman. Material from five 100 mm dishes were 66

PAGE 75

pooled for one analysis. Glial, neuronal and mixed cultures were used in these studies. Mixed cultures are cultures which were not treated with cytosine arabinoside, and therefore contain about 50% neurons and 50% glia. The cells were centrifuged at 2000 g and 4C for 5 mins., the supernatant discarded and the pellet stored at -20c until analysis. For analysis, cells were redissolved in I ml of ice-cold 0.9% saline containing 0.1 % Triton X-100 and homogenized by polytron (setting 10, 10s). The homogenate was centrifuged at 30,000 g and 4C for 15 min and the supernatant collected and stored at -20c until use. The pellet was dissolved in I ml 1.0 N NaOH for the measurement of protein by the method of Lowry et al., (1951). Renin Analysis Four hundred l of cell extract were reacted with 200 l of nephrectomised sheep plasma (courtesy Dr. C.E. Wood), 1200 l of NaP04 buffer, pH 6.5 and 100 l of an inhibitor cocktail (95 mg PMSF in 5 ml of 95% ethanol, plus 5 l P-mercaptoethanol) for 0-12 h at 37C in a shaking bath. Blank incubations were performed using 400 l 0.9% saline with 0.1% Triton X-100 instead of cell extract. At 0, 1, 2, 4, 6, and 12 h. a 200 l aliquot was removed from each blank and sample tube and added to 200 l of Tris buffer, pH 7.0. The tubes were placed in a boiling water bath for 5 mins. and the resulting precipitate pelleted by centrifugation at 3500 g for 15 min. The supernatants were stored at -20c until used in a radioimmunoassay. A specific anti-angiotensin I antibody from our lab was used. This antibody had less than 0 01% cross-reactivity with Ang II and less than 0.001% with angiotensin III. The assay sensitivity is 3 pg angiotensin I per sample with a final dilution of antibody to 1:500,000. expressed as pg/mg cell protein generated per hour. Angiotensin I levels are 67

PAGE 76

Table 5.1 The effects of adrenergic drugs on renin concentrations in brain cell culture. TREATMENT YOHIMBINE control 0.1 M 4h 1.0 M 4h 10.0 M 4h 50.0 M 4h 100.0 M 4h 50.0 M 24h NOREPINEPHRINE control 1.0 M 4h 10.0 M 4h 100.0 M 4h 1.0 mM 4h CLONIDINE control 10. 0 M 4h 50.0 M 4h ND = not determined NEURONAL pg AI/mg prot./h 0.529 (2) 0.653 (2) 0.586 (2) 0.773 (2) ND ND ND ND ND ND ND ND 0 524 (I) 0.377 (I) 0 616 (I) GLIAL pg AI/mg prot./h 0.549 (2) 0 .861 (I) 1.368 (I) 0.597 (2) 0 672 (2) 0.546 (I) 0.555 (I) 1.926 (I) 1.037 (I) 0 684 (I) 1.376 (I) 1.547 (I) ND ND ND MIXED pg Al/mg prot./h 1.483 0.3 (4) 1.871 (2) 1.469 0 3 (4) 1.412 0 2 (4) 2.05 ND ND 0.406 (I) 0.320 (I) 0.361 (I) ND ND ND ND ND (2) number in brackets =number of times the determination was measured. 68

PAGE 77

69 Results Renin was present in all cultures examined. The control levels were 0.527 0.03 pg angiotensin I /mg protein/h, (n = 3) for neurons; 1.008 0.46 pg angiotensin I/mg protein/h, (n = 3) for glia and 1.267 0.32 pg angiotensin I/mg protein/h, (n = 5) for mixed cultures. However neither yohimbine, norepinephrine, nor clonidine had any effect on these levels when the cells were treated for 4 or 24 h. (see table 5.1 ). Discussion Yohimbine, which increases the release of angiotensin II from neuronal cells was unable to induce any change in the concentration of renin in these cells. Similarly, there was no change in either mixed or glial cells by this treatment. These cell types were also tested for changes in renin concentration because the exact cellular location of renin has not been established, and we were anxious not to miss an effect if there was one. The treatment times were longer than those used to induce release of angiotensin II from cultures because we wanted to allow enough time for protein synthesis to occur, assuming renin synthesis would be stimulated. However, it may be possible that the effect was overlooked by these time courses. Several interesting questions are raised by these findings. Thus, although yohimbine induces an increase in the release of angiotensin II from neurons, it does not stimulate the cells to synthesize more renin. This means either that renin is not limiting in these cells, so that they can easily increase synthesis of angiotensin II without additional renin, or that there is an effect which we were unable to see with current techniques. It seems probable that the cells would be synthesizing more angiotensin II, because the release of angiotensin II is high. It is usually thought, at least for neurotransmitters, that approximately 10% of the total content of neurotransmitter can be released readily. As we

PAGE 78

see higher levels of angiotensin II release than this, we expect that synthesis of angiotensin II is occurring. This synthesis is apparently not dependent upon increased renin levels. It is possible that if renin is not limiting in these cells, then another factor may be. It could be that synthesis of angiotensinogen is a more important regulator in neuronal cultures, something that remains to be studied. Thus, in summary, although renin concentration was not altered in cultured brain cells by drugs known to enhance the release of angiotensin II from these cells, important information about the regulation of the synthesis of angiotensin can be inf erred from these negative findings. 70

PAGE 79

CHAPTER VI REGULATION OF cx:2-ADRENERGIC RECEPTORS BY PEPTIDES Introduction Angiotensin II has many interactions with catecholamines which are fully described in Chapter One. Insulin, another peptide hormone also has interactions with catecholamines. Recently, it has become clear that insulin has actions in the brain, where the classical view was that the brain was independent of insulin. However, the insulin which acts in the brain is probably synthesized in the brain, a possibility that helps to reconcile the classical and new theories. Previous studies have shown that insulin and its receptors are present on neurons (Boyd and Raizada, 1983, Boyd et al., 1985) and astrocytic glial cells (Clarke et al., 1984) in culture, as well as in membranes prepared from the brain of the rat (Havrankova et al., 1978a, Havrankova et al., 1978b, Havrankova et al., 1983). Insulin can be released by depolarization (Clarke et al., 1986) and causes changes in the rate of firing of nigrostriatal dopaminergic neurons (Saller and Chiodo, 1980) and hippocampal neurons (Palovcik et al., 1984 ). It also affects the dopaminergic system of the olfactory bulbs (Barbaccia et al., 1982), and at high doses stimulates release of dopamine norepinephrine and epinephrine from hypothalamic slices (Sauter et al., 1983). Insulin inhibits total bioamine uptake into neuronal cultures (Boyd et al., 1985). Ninety-five percent of specific uptake of norepinephrine into neurons can be inhibited by insulin. The specific inhibitor of norepinephrine uptake, maprotiline, inhibits high affinity binding of insulin at concentrations used to 71

PAGE 80

72 inhibit norepinephrine uptake, suggesting that maprotiline and insulin may act at the same receptors (Boyd et al., 1986). Insulin and angiotensin II have interactions with catecholamines. However, these interactions are not always the same. In the case of uptake of norepinephrine, the actions of insulin and angiotensin II are opposite. Insulin inhibits uptake of norepinephrine into neurons (Boyd et al, 1986), whereas angiotensin II stimulates uptake, at least initially (Sumners and Raizada, 1986). But in the case of release of catecholamine their actions are similar. Angiotensin II increases the release of norepinephrine (Schacht, 1984, GarciaSevilla et al., 1979, Meldrum et al., 1984, Chevillard et al., 1979) as does insulin (Sauter at al., 1983). The actions of these two peptides appear to be the modulation of the amount of catecholamines present in the synaptic cleft, or in other words the modulation of noradrenergic transmission. Since oc2-adrenergic receptors also strongly influence the level of noradrenergic transmission, we became interested in whether insulin and angiotensin II could interact with oc2-receptors, and if they could, were the interactions of angiotensin II and insulin with oc2-receptors the same? Methods Glial and neuronal cultures were prepared as described previously in Chapter Two. Insulin was dissolved in a small volume of 0.1 N NaOH and diluted to the correct concentration in buffer. It was then sterilized by filtration and added to the medium of the cells. For the time-course experiments 167 nM insulin was added to the cells, which were removed from the dishes 24 h. later, as previously described in Chapter Two. For 48 h or longer incubations, the cells received insulin every 24 h. to maintain its concentration. Control cells received sterile vehicle instead of insulin and a

PAGE 81

control group and an insulin-treated group were always assayed together at each time measurements were made as we have previously determined that each group of glial cells demonstrates a slightly different level of [3H]-yohimbine binding. For concentration curves we used 24 h. time intervals and 16. 7, 167, and 1670 nM insulin. Scatchard analysis was achieved by binding insulin-treated (1670 nM for 48 h.) and control membranes with 1.0 to 24 nM [3H]-yohimbine and then analyzing the saturation data by the method of Scatchard. Angiotensin II was added to either neuronal or glial cultures at 10 nM, 100 nM, 1 M or 10 M for either 1, 2, 4, 24, or 4 and 24 h. Cells were then removed from the dishes as described in Chapter Two and frozen until assay. The cx:2-adrenoceptor binding assay was performed exactly as described in Chapter Two except that 100 M norepinephrine was used as the cold displacer instead of 100 M yohimbine in the norepinephrine competition curves. Cells from more than one batch were usually pooled to obtain enough protein for these experiments, and if this were necessary, then each batch was divided so that each treatment group had equal numbers of plates from each batch. From experiments in which norepinephrine competition curves were generated, it was possible to look for changes in total binding at one concentration of ligand, and changes in the ability of norepinephrine to displace the binding. However, changes in total binding might have been missed by examination of binding at only one concentration of ligand. Therefore saturation experiments followed by Scatchard analysis were also performed. It was important to examine whether angiotensin II altered the ability of norepinephrine to displace the binding of (3H]-yohimbine because the cx:2-adrenoceptor can exist in one of two affinity states. The affinity state of the receptor is regulated by the presence or absence of GTP. Yohimbine, an antagonist, binds to both affinity states of the cx:2-receptor equally. 73

PAGE 82

74 Norepinephrine, an agonist, binds only to the high affinity state of the receptor. Thus, changes in affinity state would be missed if binding of [3H] yohimbine was the only criterion measured. Changes in the ability of norepinephrine to displace the bound [3H]-yohimbine would be indicative of changes in the affinity of the receptor. Results Effects of Insulin on
PAGE 83

100 (!) z i5 90 z i!i ...J 0 a: 80 z 0 0 70 w (!) 60 z w 0 a: w Cl. 50 24 48 72 TIME/ HOURS Figure 15. Time course for the decrease in [3H]-yohimbine binding caused by insulin 167 nM insulin was added to the cells and binding performed at the indicated times after treatment. Each point represents the mean SEM of three experiments in which six and six non-specific binding tubes were used per data point, except the 72-h time point, which is of one experiment. 75

PAGE 84

10 C) z 0 z ffi ...J 0 80 a: .... z 0 u u. 70 0 w (.!) 60 z w u a: w 50 a. 16.7 167 1670 DOSE INSULIN / nM Figure 16. Dose response curve for the decrease in [3H]-yohimbine binding following treatment of cells for 24 hours with insulin. Each point represents the mean SEM of six total and six non-specific binding tubes at each point. 76

PAGE 85

77 Table 6.1. Effects of i nsulin on the number and affinity of r1HJ-yohimbine binding sites in glial cultures Control Insulin (1670 nM for 48 hours) Ko (nM) 21.3 15.1 Bmax (pmoles/mg protein) 2.83 1.35 Data represent mean values from two independent experiments in which a control and an insulin-treated group were assayed concurrently. Effects of Angiotensin II on
PAGE 86

78 site to which [3H]-yohimbine binds. Thus, neither treatment with angiotensin II nor the presence of angiotensin II affects total [3H]-yohimbine binding. Table 6.2. The effect of angiotensin II on the number and affinity of r1HJ yohimbine binding sites in neuronal cultures Control Angiotensin II (1 M 4 and 24 hours) Ko (nM) 16.9 17.8 Bmax (pmoles/mg protein) 0.187 0.175 Data represent mean values from two independent experiments in which a control and an angiotensin II-treated group were assayed concurrently. There was no significant effect of angiotensin II on the ability of norepinephrine to displace [3H]-yohimbine binding, at any dose or time period tested. However angiotensin II treatment at I M for 4 and 4 and 24 h decreased the Hill slope from 0.55 to 0.42. The R values for this fit of the Hill curve were 0.94 for the control and 0.94 for the angiotensin II-treated cells. The IC50 values were 1.43 M for control and 0.97 M for the angiotensintreated cells. These data are combined from seven experiments, and were analyzed with a computer program specifically designed to analyze data from competition curves. A similar effect on the IC50 values were seen after treatment with I M angiotensin II for I h, n=3 experiments, R=0.97 in both cases. However, after treatment with I M angiotensin II for 2 h, the Hill slope was changed from 0.67 (control) to 0.78, with fits of R=0.93, and the IC50 from 1.8 to 2.6 M (n=4) in control and treated cells respectively. No doseresponsiveness to the changes could be discerned. Thus, any changes which may occur are not consistent, nor always in the same direction. Possibly the variability associated with neuronal cx:2-binding masks these changes.

PAGE 87

Discussion As mentioned in the introduction, insulin has many interactions with monoamines. One such interaction, that of inhibition of norepinephrine uptake, would tend to increase the amount of norepinephrine present in synapses. As a population of oc:2-adrenergic receptors regulates the amount of norepinephrine in the synaptic cleft by affecting synthesis and release of norepinephrine, it might be expected that insulin and oc:2-adrenergic receptors would act in opposition in the control of this function. For example, insulin would increase norepinephrine in the cleft, which would feed back to increase oc:2adrenoceptors, which would in turn tend to return norepinephrine levels to normal. The present results do not support this hypothesis, as insulin downregulated oc:2-adrenergic receptors. Several possible explanations of this phenomenon exist. Firstly, insulin may increase synaptic levels of norepinephrine not only by decreasing uptake but also by decreasing the level of "inhibitory" autoreceptors, suggesting that insulin is subserving an important physiological pathway which has redundant systems. In other words, it seems that insulin can increase the amount of norepinephrine in the synaptic cleft by two mechanisms, inhibition of uptake and increased release via decreased oc:2-adrenergic receptors. This suggests that the neuromodulatory role of insulin in the control of norepinephrine is very important physiologically. Secondly, insulin may act at a post-oc:2-adrenergic receptor site to increase the efficacy of oc:2-adrenergic stimulation which, while maintaining homeostasis, would tend to decrease oc:2-receptor binding. Thirdly, the oc:2-adrenergic receptor we have examined is on glial cells in culture and may not have the same function as the neuronal oc:2-adrenergic receptor. Angiotensin II has no significant effects on oc:2-adrenoceptor binding. It does not affect total high or low affinity binding. It is possible that an effect of 79

PAGE 88

angiotensin II on cx:2-adrenergic binding is present, but that the variablilty in the binding masks this effect. The variability arises from several causes. Firstly, the level of specific binding in neurons is only 30-40% and secondly the number of counts bound is low. Thirdly, the specific activity of the ligand, [3H]-yohimbine is 70-90 Ci/mmol, which is about 20 fold lower than some iodinated ligands (for example [ 1251]-angiotensin II), which also leads to errors in the measurement of binding. Thus, it is possible that with a better ligand, a more consistent effect of angiotensin II on cx:2-receptor binding would be observed. This inability of angiotensin II to cause consistent effects on cx:2-receptors was somewhat surprising as we expected angiotensin II, with its well-known ability to enhance the stimulated release of norepinephrine from brain tissue to also enjoy some interaction with cx:2-receptors.. It suggests that the actions of angiotensin II actions do not share a common pathway with cx:2-adrenoceptors, and that the enhancement of release itself does not affect cx:2-adrenoceptors. We have shown that drugs which affect synaptic levels of norepinephrine do alter cx:2-binding; for example clonidine, which decreases norepinephrine release, decreased cx:2-binding (in both neurons (n=2) and glia (n=l)). Norepinephrine increased the binding as did desmethylimipramine, a norepinephrine uptake inhibitor (n=2), both of which increase synaptic norepinephrine levels Therefore, we expected angiotensin II to change cx:2binding. However angiotensin II receptors are modulated by ex 1 -adrenergic receptors in a negative manner, in other words when norepinephrine levels increase angiotensin receptor binding goes down, (Sumners, Watkins and Raizada, 1986b), and we suggest that this may be a mechanism by which the norepinephrine-angiotensin II interaction is regulated. 80

PAGE 89

Although both insulin and angiotensin II have strong interactions with catecholamines, only insulin strongly influences the cx:2-adrenoceptor. So while both peptides induced increased release of catecholamines from brain tissue, and exhibited other interactions with catecholamines as well, it was not possible to generalize about their effects on cx:2-receptors. 81

PAGE 90

CHAPTER VII GENERAL CONCLUSIONS We have been able to demonstrate the presence of cx:2-adrenergic receptors on both neurons and glia in cell culture. These cultures were derived from one-dayold rat brains. The neuronal cultures are 85-90% neuronal cells, the remainder being astrocytic glial cells. The glial cultures contain more than 95% astrocytic glia, in various forms from polygonal to process-bearing. The remainder are primarily oligodendrocytes. There is very little evidence of fibroblasts in the cultures. We have demonstrated that in neuronal culture, the neurons reach high stages of development, because some of the neurons stain with antibodies against the heavy type of neurofilament, which only appears during later stages of neuronal development (Chapter Three). The light and medium types of neurofilament were also present in the neuronal cultures (data not shown) Using [3H]-yohimbine as the ligand, we have been able to characterize cx:2-adrenoceptors on both cell types The neuronal and glial cx:2-adrenoceptor have similar affinity, but the glial cultures have a much greater capacity, approximately ten times as many binding sites as neurons The receptors ha v e the pharmacological characteristics of cx:2-adrenoceptors as cx:2-selective drugs are the most effective displacers of the binding. Neither dopamine antagonists nor uptake blockers, such as desmethylimipramine, bind with high affinity to the site in either neuronal or glial culture, reassuring us that were indeed examining cx:2-adrenoreceptors These data may help explain why cx:2-adrenoceptor binding can still be demonstrated following lesion of catecholaminergic neurons (U'Pritchard et al., 82

PAGE 91

83 1979), whereon, in the classical view, pre-synaptic cx2-receptors reside. The residual binding may represent glial binding, or it may represent binding to cx2adrenoceptors mediating the release of other types of neuroactive substances. It has been thought for some time that cx2-adrenoceptors control the release not only of norepinephrine (through inhibitory autoreceptors), but also the release of serotonin and acetylcholine (Frankhuyzen and Mulder, 1980; Gothert and Huth, 1980; Vizi, 1972). We have been interested for some time in understanding how the release of the neuropeptide, angiotensin II, was controlled. Phylogenetically, angiotensin II has a very old history of interactions with catecholamines (Carroll and Opdyke, 1982), which occur centrally as well as peripherally. Therefore, we postulated that cx2-adrenoceptors located presynaptically on angiotensin II neurons might control the release of angiotensin II. To this end, we established that blockade of cx2-adrenoceptors in neuronal culture induced a dose-responsive increase in the release of angiotensin II, analogous to the increased overflow of norepinephrine seen following blockade of inhibitory autoreceptors (presynaptic cx2-adrenergic receptors on norepinephrine containing neurons). Glial cultures did not exhibit angiotensin II release in response to cx2-adrenoceptor blockade, although they have many cx2-adrenoceptors and contain angiotensin II (Hermann et al., 1988b ). We postulated that increased release of norepinephrine, induced by blockade of cx2-adrenergic receptors, might have induced this response in neuronal cultures. Indeed, we saw increased release of angiotensin II from neuronal cultures when norepinephrine was applied to the cultures. The release of angiotensin II increased with increasing doses of norepinephrine until a dose of 50 M norepinephrine, was administered. At this dose, the release started to decline, possibly due to interaction with cx2-adrenergic receptors. The release

PAGE 92

84 of angiotensin II induced by norepinephrine could not be blocked by prazosin, a selective ex 1-adrenergic antagonist, which suggested that this response was mediated by fi-adrenergic receptors. Following unsuccessful attempts to use fi adrenergic antagonists to investigate this possibility, predominantly because of high cross-reactivity with the anti-angiotensin II antibody, we were able to demonstrate ,B-receptor-mediated release of angiotensin II with the fi-receptor agonist, isoproterenol. However, the levels of angiotensin II released were lower than with norepinephrine or cx2-adrenergic antagonists. We therefore suspect that there are two populations of neurons, one of which releases angiotensin II following fi-receptor stimulation, as has been demonstrated in the periphery (Nakamura et al., 1986a, b, Gothert and Kollecker, 1986), and another which releases angiotensin II in response to cx2-adrenergic blockade. this is not the only possible explanation of the data. However, The amount of angiotensin II released from the neuronal cultures was quite high; therefore, we expected that synthesis of angiotensin II should be stimulated in order to make new angiotensin II available for release. We postulated that the renin concentration might increase in cultures following treatments which stimulated release of angiotensin II, as renin is an important enzyme in the synthesis of angiotensin II. However, we were unable to demonstrate that this occurred with cx2-adrenergic blockers. This suggests that renin is not an important regulatory step in angiotensin II synthesis in either neurons or glia in culture. In the final experiment in this study, we examined whether any relationship between cx2-adrenergic receptors and two neuropeptides known to interact with catecholamines could be demonstrated. Insulin, a catecholamine neuromodulator, induced a time-and dose-dependent decrease in cx2-adrenergic receptor binding in glia, via a decrease in receptor number. However

PAGE 93

angiotensin II, a peptide with the ability to affect synthesis and release of norepinephrine, as do 0:2-adrenergic receptors, demonstrated no such effect. Angiotensin II did not significantly affect glial or neuronal 0:2-adrenoceptors in any way; i.e., neither total nor high affinity (agonist) binding was altered. Thus, insulin and angiotensin II interact with catecholaminergic systems differently with respect to 0:2-adrenergic receptors. If all of the data from this study, along with some previously known interactions, are put together into a model of angiotensin II-catecholamine interactions, then fig. 17 results. Examination of this model shows that most reactions feed forward, in other words catecholamines enhance angiotensin 85 actions and vice versa. The break in this cycle occurs through the ability of catecholamines to regulate angiotensin II receptors. This controlling step is important for several reasons. It is the only inhibitory interaction between angiotensin II and catecholamines that has been elucidated so far, giving it great weight. Secondly, this mechanism for 0:1 adrenergic receptor-mediated decreases in angiotensin II receptor binding is not present in the spontaneously hypertensive rat (Raizada, Muther and Sumners, 1984a), a genetic model of hypertension. So it is possible that the lack of this negative interactions is responsible for, or contributes to, the appearance of hypertension. The model described in fig. 17 also illustrates future avenues of research which could be explored. The ability of angiotensin II and 0:2-adrenergic receptor blockade to enhance release of norepinephrine should be fully examined in culture. Complete understanding of the mechanisms for release of norepinephrine may enhance our understanding of the interactions between the two systems.

PAGE 94

oc:1-RECEPTOR STIMULATION (-) oc:2-RECEPTOR BLOCKADE INCREASED NOREPINEPHRINE RELEASE ,8-RECEPTOR STIMULATION INCREASED ANGIOTENSIN II RELEASE ANGIOTENSIN II REClPTOR STIMULATION 86 (+) (+) Figure 17. Model depicting the interactions of catecholamines and angiotensin II in neuronal cell cultures.

PAGE 95

87 How well do these results correlate with the effects of angiotensin II in the brain? There is good evidence that injections of norepinephrine into sensitive parts of the brain induces drinking (Leibowitz, 1975a). The mechanism for this could be the release of angiotensin II, as very small amounts of angiotensin II (pmol) injected into the brain induces drinking (Phillips, 1987) There was some debate about the importance of the catecholamines in drinking, because much higher amounts of norepinephrine than angiotensin II had to be given to induce it. However, several factors must be taken into consideration which suggest that catecholamines may be more important than they appear on first consideration of the findings. Firstly, the receptor affinity of the catecholamines is probably three orders of magnitude lower than that of angiotensin II (Kd of approximately I M for catecholamines and 0.1-1 nM for angiotensin II). Secondly, as catecholamines have many receptors, uptake sites etc., in the brain (to carry out functions unrelated to drinking) one would expect it to be more difficult to achieve effective concentrations of catecholamines at the receptors relevent to drinking by central injection. It has been difficult to determine whether isoproterenol given into the brain induces drinking. Initial reports suggested that it did (Liebowitz, 1975a), but on further examination of the phenomenon it was suggested that the effect of isoproterenol was due to its escape from the brain to the periphery. Isoproterenol is a very potent stimulus to renal release of renin, which results in peripheral formation of angiotensin II which is dipsogensic. Radioactive tracer studies, combined with blockade of sympathetic outflow (Fisher et al., 1973) suggested that leakage of isoproterenol to the periphery or sympathetic activation of the juxtaglomerular apparatus to release renin were factors necessary for "central" isoproterenol-induced drinking. Despite this evidence, Fitzsimons ( 1979) states, "However the complete dismissal of a possible central

PAGE 96

88 mechanism for P-adrenergic drinking in the rat is not justified." Leibowitz (197 5b ), using much smaller doses of catecholamines found a drinking response to epinephrine and norepinephrine which was dependent on both ex-and P adrenoceptors. Further, although it does not apply to our studies, isoproterenol induced drinking, independent of renal influences, is much more reliable in other animals, eg dog, cat, pigeon. In fact, the pigeon drinks to central isoproterenol with a lower threshold than to peripheral isoproterenol (Fitzsimons, 1979). There is also good evidence that injection of norepinephrine into the brain (at some sites) induces increased blood pressure (Struyker-Boudier et al., 1975) as does angiotensin II (Severs et al., 1970). Therefore, the effects of norepinephrine may be mediated by release of angiotensin II. Thus, it may be that angiotensin released by norepinephrine in a P-receptor dependent fashion is important in the drinking and blood pressure responses to administered norepinephrine. Norepinephrine inhibits the release of arginine vasopressin from the posterior pituitary (Share, 1983). Since angiotensin II increases the release of arginine vasopressin from the pituitary (Yamamoto, 1978), it is clear that norepinephrine does not act in the same fashion as angiotensin II in this situation. Therefore, not all of the effects of norepinephrine can be mediated by angiotensin II. The neuronal circuitary is probably unique for each system and the generalization of our findings to all catecholaminergic neurotransmission would be foolhardy. It is more probable, considering these different comparisons of the actions of angiotensin II and norepinephrine, that some populations of catecholaminergic cells interact with angiotensin II in the way these data suggest, whereas other populations do not. One likely candidate for interactions such as we describe would be the angiotensinergic connections which originate in the subfornical organ and terminate in the median preoptic

PAGE 97

nucleus. The cells receiving input in the median preoptic nucleus are stimulated by electrical stimulation of the subfornical organ or by application of angiotensin II. Both of these types of stimulation are enhanced by pressure injection of norepinephrine (Graham et al., 1985). While not the only interpretation of the results, it is tempting to suggest that in accordance with our data, norepinephrine is inducing increased angiotensin II release and thus, facilitating the stimulation. A candidate for an area where there is no such interaction is the anterior hypothalamus, where it has been established that neither antagonism nor stimulation of ,8-receptors affects spontaneous angiotensin II release (Brosnihan et al., 1988), whereas, in cultures from whole brain we found an interaction 89 In summary significant data concerning the control of angiotensin II release by norepinephrine from brain cells in culture have been obtained. Increased release of angiotensin II does not lead to increased synthesis of renin, an important enzyme in the angiotensin II synthetic pathway. Angiotensin II does not significantly affect cx:2-adrenergic receptors in neuronal or glial cultures, although they share common activities for example, release of norepinephrine. This suggests that other regulatory steps are more important. Alpha2-adrenergic receptors were characterized in both neuronal and glial cultures (the first direct characterization in cultures from rat brain) and visualized by autoradiography in neuronal cultures

PAGE 98

REFERENCES Baker, S.P., Sumners, C., Pitha, J. and Raizada, M.K. Characteristics of the beta-adrenoceptor from neuronal and glial cells in primary cultures of rat brain. J. Neurochem. 47; 1318-1326, 1986. Barbaccia, M., Chuang, D., and Costa, E. Is insulin a neuromodulator? Adv. Biochem. Psychopharmacol. 33: 511-518, 1982. Basso, N., Kurnjek, M.L., Mikulic, L., Cannata, M.A., and Taquini, A.C., Effect of intraventricular 6-hydroxydopamine (60HDA) on the angiotensinogen concentration of the brain. Clio. Exp. Hypert-Theory and Practice A6 (IO and 11): 1773-1776, 1984 Bickerton, R.K. and Buckley, J.P. Evidence for a central mechanism in angiotensin-induced hypertension. Proc. Soc. Exp. Biol. Med. 106: 834-836, 1961. Boadle-Biber, M.C., Hughes, J. and Roth, R.H. Acceleration of catecholamine biosynthesis in sympathetically innervated tissues by angiotensin-II-amide. Nature (Lond) 222: 987-988, 1969. Boyajian, C.L., Loughlin, S.E. and Leslie, F.M. Anatomical evidence for alpha-2 adrenoceptor heterogeneity: Differential autoradiographic distributions of [3H]-rauwolscine and [3H]-idazoxan in rat brain. J. Pharmacol. exp. Therap. 241: 1079-1091, 1987. Boyd, F.T., Jr., Clarke, D.W., Muther, T.F. and Raizada, M.K. Insulin receptors and modulation of norepinephrine uptake by insulin in neuronal cells in culture. J. Biol. Chem. 260: 15880-15884, 1985. Boyd, F.T., Jr., Clarke, D.W., and Raizada, M.K. Insulin inhibits specific norepinephrine uptake in neuronal cultures from rat brain. Brain Res. 398: 1-5, 1986. Boyd, F.T., Jr. and Raizada, M.K. Effects of insulin and tunicamycin on neuronal insulin receptors in culture. Am. J. Physiol. 245: C283-C287, 1983. Brosnihan, K.B., Schiavone, M.T., Sprunger, A.E., Chappell, M.C., Rizzo, M. and Ferrario, C.M. In vivo release of angiotensin II from the rat hypothalamus. Hypert. 11 (Suppl I): 1158-1162, 1988. Buggy, J., Fisher, A.E., Hoffmann, W.E., Johnson, A .K. and Phillips, M.1. Ventricular obstruction: Effect on drinking induced by intracranial injections of angiotensin. Sci. 190: 72-74, 1975. 90

PAGE 99

91 Camacho, A. and Phillips, M.I. Separation of drinking and pressor responses to central angiotensin by monoamines. Am. J Physiol. 240: Rl06-Rl 13, 1981. Campbell D J., Bouhnik, J Menard, J., and Corvol, P. Identity of angiotensinogen precursors of rat brain and liver. Nature 308: 206-208, 1984 Carroll, R.G. and Opdyke, D.F. Evolution of angiotensin II induced catecholamine release Am. J. Physiol. 243: R65-R69, 1982. Cheng Y.C. and Prusoff, W.H. Relationship between the inhibition constant (Ki) and the concentration of inhibitor which causes 50 percent inhibition of an enzymatic reaction. Biochem Pharmacol. 22: 3099-3108, 1973. Chevillard, C., Duchene, N., Pasquier, R and Alexandre, J-M. Relation of the centrally evoked pressor effect of angiotensin II to central noradrenaline in the rabbit. Eur. J. Pharmacol. 58: 203-206, 1979 Clarke, D.W., Boyd, F.T., Jr., Kappy, M .S. and Raizada, M.K. Insulin binds to specific receptors and stimulates 2-deoxy-D-glucose uptake in cultured glial cells from rat brain. J. Biol. Chem. 259: 11672-11675, 1984. Clarke, D.W Boyd, F.T. Jr., and Raizada, M.K. Insulin is released from neuronal cultures from the rat brain. J. Neurochem. 47: 831-836, 1986. Davila D. and Khairallah, P.A Angiotensin II and biosynthesis of norepinephrine. Arch. Int Pharmacodyn. 193: 307-314, 1971. Ebersolt, C., Perez, M. adnd Bokaert J. oc:1 and oc:2-adrenergic receptors in mouse brain astrocytes from primary cultures. J. Neurosci. Res. 6: 643652, 1981. Dzau V.J. Significance of the vascular renin-angiotensin pathway. Hypertension 8: 553-559, 1986. Evans, T., McCarthy, K.D. and Harden, T.K. Regulation of cyclic AMP accumulation by peptide hormone receptors in immunocytochemically defined astroglial cells. J Neurochem. 43: 131-138, 1984. Feldberg, W., and Lewis, G.P. The action of peptides on the adrenal medulla. Release of adrenaline by bradykinin and angiotensin. J. Physiol. 171: 98-108, 1964. Feldstein, J B., Gonzales, R A ., Baker, S P., Sumners, C., Crews, F T. and Raizada, M.K. Decreased alpha 1-adrenergic receptor mediated inositide hydrolysis in neurons from hypertensive rat brains. Am. J Physiol. 251: C230-C237, 1980 Feldstein J.B., Pacitti, A.J., Sumners, C. and Raizada, M.K. Alpha 1adrenergic receptors in neuronal cultures from rat brain: Increased expression in the spontaneously hypertensive rat J. Neurochem 47: 11901198, 1986.

PAGE 100

92 Fisher, A.E Relationships between cholinergic and other dipsogens in the central mediation of thirst. In: Neuropsychology of Thirst: New Findings and Advances in Concepts. Ed. Epstein, A.N., Kissileff, H. R. and Stellar, E. Washington. Winston and Sons, 1973, pp 24378. Fitzsimons, J.T. Pharmacology of Drinking. and Sodium Appetite. Ed. Fitzsimons, J.T. Press, 1979, p 433. In: The Physiology of Thirst Cambridge. Cambridge Univ. Frankhuyzen A.L. and Mulder, A.H. Noradrenaline inhibits depolarization-induced JH-serotonin release from slices of rat hippocampus. Eur. J. Pharmacol. 63: 179-182, 1980. Fregly, M.J., Rowland, N E. and Greenleaf, J.E. Clonidine antagonism of angiotensin-related drinking: a central site of action. Brain Res. 298: 321327, 1984. Fuxe, K., Andersson, K., Locatelli, V., Mutt, V., Lundberg, J., Hokfelt, T., Agnati, L.F., Eneroth, P. and Bolme, P. Neuropeptides and central catecholamine systems: Interactions in neuroendocrine and central cardiovascular regulation. In: Neural Peptides and Neuronal Communication. Ed. Costa, E. and Trabucchi, M., Raven Press, New York, 1980, pp37-50. Garcia-Sevilla, J.A., Dubocovich M.L. and Langer, S.Z. Angiotensin II facilitates the potassium-evoked release of 3H-noradrenaline from the rabbit hypothalamus Eur. J Pharmacol. 56: 173-176, 1979. Gonzales, R.A., Crews, F T., Sumners, C. and Raizada, M.K. Norepinephrine regulation of alpha 1-stimulated phosphoinositide hydrolysis in primary neuronal cultures J. Pharmacol. Exp. Therap. 242: 764771, 1987. Gonzales, R.A., Feldstein, J.B., Crews, F.T., Raizada, M .K. Receptor-mediated inositide hydrolysis is a neuronal response: Comparison of primary neuronal and glial cultures. Brain Res 345: 350-355, 1985. Gordon, F.J., Brody, M J Fink, G.D., Buggy, J and Johnson, A.K. Role of central catecholamines in the control of blood pressure and drinking behavior. Brain Res. 178: 161-173, 1979. Gordon, F.J., Brody, M.J. and Johnson, A .K. Regional depletion of central nervous system catecholamines: Effects on Blood Pressure and Drinking Behavior. Brain Res. 345: 285-297, 1985. Gothert, M. and Huth, H Alpha-adrenoceptor-mediated modulation of 5-hydroxytryptamine release from rat brain cortex slices. Naunyn-Schmiedeberg's Arch. Pharmacol. 313, 21-26, 1980. Gothert, M. and Kollecker, P. Subendothelial .82-adrenoceptors in the rat vena cava: Facilitation of noradrenaline release via local stimulation of angiotensin II synthesis. Naunyn-Schmiedeberg's Arch. Pharmacol. 334 : 156-165, 1986.

PAGE 101

93 Graham, C.A., Nelson, D.O and Johnson, A.K. Noradrenergic modulation of subfornical organ input to the nucleus medianus Soc. Neurosci. Abs. 11:242.7, 1985 Hansson, E. and Sellstrom A. MAO, COMT and GABA-T activities m primary astroglia cultures. J. Neurochem. 40: 220-225, 1983. Havrankova, J., Roth, J., and Brownstein, M.J. Insulin receptors are widely distributed in the central nervous system of the rat. Nature (Lond) 272: 827-829, 1978a. Havrankova, J., Roth, J., and Brownstein, M.J., Insulin receptors in brain. In: Advances in Metabolic Disorders, Vol. IO, CNS Regulation of Carbohydrate Metabolism, Szabo, A.J., ed. Academic Press, New York, I 983, pp259-268. Havrankova J., Schmechel, D., Roth, J., and Brownstein, M.J. Identification of insulin in rat brain. Proc. Natl. Acad. Sci U.S.A 75: 5737-5741, 1978b. Hermann, K., Phillips, M.I., Hilgenf eldt, U. and Raizada, M.K. Biosynthesis of angiotensinogen and angiotensins by brain cells in primary culture J. Neurochem. 51: 398-405, 1988a. Hermann, K., Raizada, M.K. Sumners, C. and Phillips, M.I. Presence of renin in primary neuronal and glial cells from rat brain. Brain Res. 437: 205213, 1987. Hermann, K., Raizada, M.K. Sumners, C., and Phillips, M.I. Immunocytochemical and biochemical characterization of angiotensin I and II in cultured neuronal and glial cells from rat brain. Neuroendocrinology 47: 125-132, 1988b. Hirata, H., Slater, N.T. and Kimelberg, H.K. cx-adrenergic receptor-mediated depolarization of rat neocortical astrocytes in primary culture. Brain Res. 270: 358-362, 1983. Hoffman, D L., Knipp, L., Schrag, D., Nilaver, G., Valiquette, G Kilcoyne, M.M. and Zimmerman, E.A. Angiotensin-immunoreactivity in vasopressin cells in rat hypothalamus and its relative deficiency in homozygous Brattleboro rats. Annals N.Y. Acad. Sci. 394:135-144, 1982. Hosli, E. and Hosli, L. Evidence for the existence of ex-and .B-adrenoceptors on neurons and glial cells of rat central nervous system--An autoradiographic study. Neuroscience 7: 2873-2881, 1982. Huang, Y., Rogers, J and Henderson, G. Effects of angiotensin II on [3H] noradrenaline release and phosphatidylinositol hydrolysis in the parietal cortex and locus coeruleus of the rat. J. Neurochem. 49: 1541-1549, 1987 Hughes, J. and Roth, R.H. Evidence that angiotensin enhances transmitter release during sympathetic nerve stimulation. Br J. Pharmacol. 41: 239-255, 1971.

PAGE 102

94 Jakobs, K.H. Inhibition of adenylate cyclase by hormones and neurotransmitters. Mol. Cell. Endocrinol. 16: 147-156, 1979. Kurnjek, M.L. Trolliet, M.R., Mikulic. L.P. and Basso, N. The effect of methyldopa on the renin-angiotensin and noradrenergic sympathetic systems in normal rats. J. Hypertens. 4 (Suppl 6): S450-S452, 1986. Langer, S.Z. Presynaptic receptors and their role in the regulation of transmitter release. Br. J. Pharmacol. 60: 481-497, 1977. Langer, S.Z. Presynaptic regulation of the release of catecholamines. Pharmacol. Rev. 32: 337-362, 1981. Langer, S.Z., Duval, N. and Massingham, R. Pharmacologic and therapeutic significance of oc:-adrenoceptor subtypes. J. Cardiovasc. Pharmacol. 7 (Suppl 8), SI-S8, I 985. Leibowitz, S.F. Pattern of drinking and feeding produced by hypothalamic norepinephrine injection in the satiated rat. Physiol. and Behav. 14: 73142, 1975a. Leibowitz, S.F. Ingestion in the satiated rat: Role of alpha and beta receptors in mediating effects of hypothalamic adrenergic stimulation. Physiol. and Behav. 14: 742-54, 1975b. Lind, R.W., Swanson, L.W. and Ganten, D. Angiotensin II immunoreactivity in the neural afferents and efferents of the subfornical organ of the rat. Brain Res. 321: 209-215, 1984. Lind, R.W., Swanson, L.W. and Ganten, D. immunoreactive cells and fibres in the Neuroendocrinol. 40: 2-24, 1985. Organization of angiotensin II rat central nervous system. Lindvall, 0. and Bjorkland, A. Dopamine and norepinephrine-containing neuron systems: Their anatomy in the rat brain. In, Chemical Neuroanatomy, Ed P.C. Emson, Raven Press, New York, 1983, pp 229-256. Lowry, O.H., Rosebrough, N.J., Farr, A.L. and Randall, R.J. Protein measurement with folin-phenol reagent. J. Biol. Chem. 193: 265-275, 1951. Majewski, H., Hedler, L., Schurr, C., and Starke, K. Modulation of noradrenaline release in the pithed rabbit: A role for angiotensin II. J. Cardiovasc Pharmacol. 6: 888-896, 1984. Masters, B A., Pacitti, A.J., Baker, S.P., Pitha, J., Sumners, C. and Raizada, M.K. Distinct populations of oc: 1-adrenergic receptors on fetal rat neuronal and glial cells in primary culture. The Physiologist: In press, 1988. McCarthy, K D. An autoradiographic analysis of .B-adrenergic receptors on immunocytochemically defined astroglia J. Pharmacol. Exp. Therap. 226: 282-290, 1983. McCarthy, K.D. and deVellis, J. Alpha-adrenergic receptor modulation of beta-adrenergic, adenosine aand prostaglandin E 1 increased adenosine 3':5'-

PAGE 103

cyclic monophosphate levels in primary cultures of glia Nucleotide Res. 4: 15-26 1978. J. Cyclic 95 McRae-Degueurce A., Bellin, S.I., Landas, S.K. and Johnson, A.K. Fetal noradrenergic transplants into amine-depleted basal forebrain nuclei restore drinking to angiotensin. Brain Res. 374: 162-166, 1986. Meldrum, M.J Xue, C-S. Badino, L., and Westfall, T C. Angiotensin facilitation of noradrenergic neurotransm1ss1on in central tissues of the rat: Effects of sodium restriction. J Cardiovasc Pharmacol. 6 : 989-995 1984. Mendelsohn, F.A.0., Quirion, R. Saavedra, J M., Aguilera, G and Catt, K.J. Autoradiographic localization of angiotensin II receptors in rat brain. Proc. Natl. Acad. Sci. USA 81: 1575-1579, 1984. Meyer, J.M. and Weyhenmeyer, J.A. Chemically-evoked release angiotensin-like peptide from fetal rat brain cells in culture. Physiol. 251: Cl 7-C22, 1986 of an Am. J. Mikulic, L., Kurnjek, M.L., Trolliet, M.R., Ruiz, P., Basso, N. and Taquini, A.C. Effect of guanethidine on the development of two-kidney, two-clip renal hypertension. Hypertension 8 (Suppl. I): I 123-1127, 1986. Murphy, S. and Pearce, B, Functional receptors for neurotransmitters on astroglial cells. Neuroscience 22 : 381-394, 1987. Nakamura, M., Jackson, release of angiotensin 250: H 144-H 148, 1986a. E.K. and II from Inagami, mesenteric T. P-Adrenoceptormediated arteries. Am. J Physiol. Role of vascular Angiotens i n II Nakamura, M Jackson E.K. and Inagami T released by P-adrenergic stimulation in 8(Suppl IO): Sl-S5, 1986b. rats. J. Cardiov. Pharmacol. Moore, R.Y. and Bloom, F.E. Central catecholamine systems: Anatomy and physiology of the norepinephrine and epinephrine systems. Ann. Rev. Neurosci. 2: 113-168 1979. Palaic D. and Khairallah, P.A. Effect of angiotensin on uptake and release of norepinephrine by brain. Biochem. Pharmacol. 16: 2291-2298 1967. Palovcik, R.A., Phillips, M I., Kappy, M.S and Raizada, M K Insulin inhibits pyramidal neurons in hippocampal slices. Brain Res. 309: 187-191, 1984. Phillips M I. Functions of angiotensin in the central nervous system. Ann. Rev. Physiol. 49: 4 13-435, 1987 Phillips M.I. and Felix, D. Specific angiotensin II receptive neurons in the cat subfornical organ. Brain Res. 109: 531-540, 1976. Phillips, M.I. and Stenstrom, B. Angiotensin II in rat brain comigrates with authentic angiotensin II in high pressure liquid chromatography. Circ Res. 56: 212-219, 1985.

PAGE 104

96 Printz, M.P., Wallis, C.J., Lewicki, J.A. and Fallon, J.H. Correlation between CNS angiotensinogen and central catecholamine pathways. In: Catecholamines: Basic and Clinical Frontiers. Vol.2 Ed: Usdin E., Pergamon Press, New York, 1979, pp 1419-1421. Printz, M.P., Hawkins, R.L. Wallis, C.J., Chen, F.M. Steroid hormones as feedback regulators of brain angiotensinogen and catecholamines. Chest 83: Suppl.2 308-11, 1983. Raizada, M.K. Localization of insulin-like immunoreactivity in the neurons from primary cultures of rat brain. Exp. Cell Res. 143: 351-357, 1983. Raizada, M.K., Muther, T.F and Sumners, C. Increased angiotensin II receptors in neuronal cultures from hypertensive rat brain. Am. J. Physiol. 24 7: C364-C3 72, 1984a. Raizada, M.K., Phillips, M.I., Crews, F.T., and Sumners, C. angiotensin II receptors in glial cells cultured from rat brain. Acad. Sci. 84: 4655-4659, 1987. Distinct Proc. Natl. Raizada, M.K., Stamler, J.F., Quinlan, J.T., Landas, S. and Phillips, M.I. Identification of insulin receptor-containing cells in primary cultures of rat brain. Cell. Mol. Neurobiol. 2: 47-52, 1982. Raizada, M.K., Stenstrom, B., Phillips, M.I. and Sumners, C. Angiotensin II in neuronal cultures from brains of normotensive and hypertensive rats. Am. J. Physiol. 247: Cll5-Cll9, 1984b. Raizada, M.K., Yang, J.W., Phillips, M.I. and R.E. Fellows. Rat brain cells in primary culture: Characterization of angiotensin II binding sites. Brain Res. 207: 343-355, 1981. Reit, E. Actions of angiotensin II in the adrenal medulla and autonomic ganglia. Fed. Proc. 31: 1338-1343, 1972. Richards, E.M., Raizada, M.K., Phillips, M.I. and Sumners, C. Characterization of alpha2-adrenergic receptors in rat neuronal and glial cell cultures. The Physiologist 29: 58.11, 1986. Rogers, A.W. Techniques of Autoradiography. New York. Elsevier, 1979, pp 368-370. Rouot, R., Quennedey, M.C. and Schwartz, J. Characteristics of the [3H]yohimbine binding on rat brain cx:2-adrenoceptors. Naunyn-Schmiedeberg's Arch. Pharmacol. 321: 253-259, 1982. Rump, C. and Majewski, H. ,B-adrenoceptor facilitation of norepinephrine release is not dependent on local angiotensin II formation in the rat isolated kidney. J. Pharmacol Exp. Therap. 243: 1107-1112, 1987. Saller, C.F. and Chiodo, L.A. Glucose haloperidol-induced increases in the firing neurons. Science 210: 1269-1271, 1980. suppresses basal firing and rate of central dopaminergic

PAGE 105

Sauter, A., Goldstein, M., Engel, J. and Ueta, K. Effect of insulin on central catecholamines. Brain Res. 260: 330-333, 1983. Scatchard, G. The attraction of proteins for small molecules and ions. Ann. N.Y. Acad. Sci. 51: 660-672, 1.949. Schacht, U. Effects of angiotensin II and angiotensin converting enzyme inhibitors on electrically stimulated noradrenaline release from superfused rat brain slices. Clin. and Exper. Hypert. Theory and Practice A6 (10 and II) 1847-51, 1984. 97 Schiavone, M.T., Brosnihan, K.B., Husain, A., and Ferrario, C.M. Basal and potassium evoked release of angiotensin II from the rat hypothalamus. Brain Res. 397: 193-196, 1986. Severs, W.B., Summy-Long, J., Taylor, J.A. and Connor, J.D. A central effect of angiotensin: Release of pituitary pressor material. J. Pharmacol. exp. Therap. 174: 27-34, 1970. Share, L. Centrally Acting Humoral Factors in the Control of Vasopressin Release. The Neurohypophysis: Structure, Function and Control. Progress in Brain Res. 60: Eds. Cross, B.A. and Leng, G., 1983, pp425-435. Shaw, G., Banker, G.A. and Weber, K. An immunofluorescence study of neurofilament protein expression by developing hipppocampal neurons in tissue culture. Europ. J. Cell Biol. 39: 205-216, 1985. Simmonet, G. and Giorguieff-Chesselet, M.F. Stimulating effect of angiotensin II on the spontaneous release of newly synthesized [3H] dopamine in rat striatal slices. Neurosci. Lett. 15: 153-8, 1979. Sladek, J.R. and McNeill, T.H. Simultaneous monoamine histofluorescence and neuropeptide immunocytochemistry. IV. Verification of catecholamine neurophysin interactions through single section analysis. Cell Tiss. Res. 210: 181-189, 1980. Starke, K. Action of angiotensin on uptake, release and metabolism of I4c_ noradrenaline by isolated rabbit hearts. Eur. J. Pharmacol. 14: 112-123, 1971. Starke, K. Regulation of noradrenaline release by presynaptic receptor systems. Rev. Physiol. Biochem. Pharmacol. 77: 1-24, 1977. Starke, K., Werner, U., Hellerforth, R. and Schumann, H.J. Influence of peptides on the output of noradrenaline from isolated rabbit hearts. Europ. J. Pharmacol. 9: 136-140, 1970. Story, D.F. and Ziogas, J. Interaction of angiotensin with noradrenergic neuroeffector transmission. Trends Pharmacol. Sci. 8: 269-271, 1987. Struyker-Boudier, H., Smeets, G., Brouwer, G. and van Rossum, J.M. Central nervous system alpha-adrenergic mechanisms and cardiovascular regulation in rats. Arch. Int. Pharmacodyn. 213:285-293, 1975.

PAGE 106

98 Sumners, C. Muther, T.F. and Raizada, M.K. Altered norepinephrine uptake in neuronal cultures from spontaneously hypertensive rat brain. Am. J. Physiol. 248: C488-C497, 1985. Sumners, C. and Phillips, M.I., Central injection of angiotensin II alters catecholamine activity in rat brain. Am. J. Physiol. 244: R257-R263, 1983. Sumners, C., Phillips, M.I. and Raizada, M.K. Rat brain cells in primary culture: visualization and measurement of catecholamines. Brain Res. 264:267-275, 1983a. Sumners, C., Phillips, M.I., and Raizada, M.K. Angiotensin II stimulates changes in the norepinephrine content of primary cultures of rat brain. Neurosci. Lett. 36: 305-309, 1983b. Sumners, C. and Raizada, M.K. Catecholamine-angiotensin II receptor interaction in primary cultures of rat brain. Am. J. Physiol. 246: C502C509, 1984. Sumners, C. and Raizada, M.K. Angiotensin II stimulates norepinephrine uptake in hypothalamus/brainstem neuronal cultures. Am. J. Physiol. 250: C236-C244, 1986. Sumners, C., Shalit, S.L., Kalberg, C.J. and Raizada, M.K. Norepinephrine metabolism in neuronal cultures is stimulated by angiotensin II. Am. J. Physiol. 252: C650-C656, 1987. Sumners, C., Watkins, L.L., and Raizada, M.K. Alpha 1-adrenergic receptormediated downregulation of angiotensin II receptors in neuronal cultures. J. Neurochem. 47: 1117-1126, 1986. Trachte, G.J., Stein, E. and Peach, M.J. Alpha adrenergic receptors mediate angiotensin-induced prostaglandin production in the rabbit isolated vas deferens. J. Pharmacol. Exp. Therap. 240: 433-440, 1987. Trolliet, M.R., Kurnjek, M.L., Mikulic, L., Ruiz, P., Grinspon, D. and Basso, N. The central and peripheral renin-angiotensin system in reserpinetreated rats. Hypertension 8 (Suppl I): 175-178, 1986. Unnerstall, J.R., Kopajtic, T. and Kuhar, M.J. Distribution of cx2-agonist binding sites in the rat and human central nervous system: Analysis of some functional, anatomic correlates of the pharmacologic effects of clonidine and related adrenergic agents. Brain Res. Rev. 7: 69-101, 1984. U'Pritchard, D.C., Bechtel, W.D., Rouot, B M. and Snyder, S.H. Multiple apparent alpha-noradrenergic receptor binding sites in rat brain: Effect of 6-hydroxydopamine. Mol. Pharmacol. 16: 47-60, 1979. Van Calker, D., Muller, M. and Hamprecht, B. Regulation by secretin, vasoactive intestinal peptide and somatostatin of cyclic AMP accumulation in cultured brain cells. Proc. Natl. Acad. Sci. (USA) 77: 6907-6911, 1980.

PAGE 107

Vizi, E.S. Stimulation by inhibition of (Na+-K+-Mg2+)-activated ATPase of acetylcholine release in cortical slices from rat brain. J. Physiol. 226: 95-117, 1972. Vizi, E.S. and Knoll, J. The effects of sympathetic nerve stimulation and 99 guanethidine on parasympathetic neuroeffector transmission: The inhibition of acetylcholine release. J. Pharm. Pharmac. 23: 918-925, 1971. Weyhenmeyer, J.A. and Phillips, M.I. Immunocytochemical localization of angiotensin in the CNS of Wistar kyoto and spontaneously hypertensive rats. Hypertension 4: 514-523, 1982. Yamamoto, M., Share, L. and Shade, R.E. Effect of ventriculo-cisternal perfusion with angiotensin II and indomethacin on the plasma vasopressin concentration Neuroendocrinol. 25: 166-173, 1978. Zimmerman, B.G. and Gisslen, J. Pattern of renal vasoconstriction and transmitter release during sympathetic stimulation in the presence of angiotensin and cocaine. J. Pharmacol. Exp. Therap. 163: 320-329, 1968. Zimmerman, B.G. and Whitmore, L. Effect of angiotensin and phenoxybenzamine on release of norepinephrine in vessels during sympathetic nerve stimulation. Int. J. Neuropharmacol.6: 27-38, 1967.

PAGE 108

BIOGRAPHICAL SKETCH Elaine Richards-Sumners was born in England on November 9th, 1957. After attending Havant Grammar School and attaining nine 'O' levels and three 'A' levels, she entered the University of Southampton in 1976. Mrs Sumners left Southampton in 1979, with a Bachelor of Science (Hons) degree in biology, second class, lower division. In 1980 she married Colin Sumners and they moved to Florida in January of 1981. Mrs Sumners entered graduate school later that year in physiology Upon the birth of their daughter, Lucy, in 1985, Mrs Sumners took a year out of graduate school. Plans for a postdoctoral position at the University of Florida will be followed upon graduation. 100

PAGE 109

I certify that I have read this study and that in my opm1on it conforms to acceptable standards of scholarly presentation and is fully adequate, in scope and quality, as a dissertation for the degree of Doctor of Philosophy. M. Ian Phillips, D.Sc., Chairman Professor of Physiology I certify that I have read this study and that in my opinion it conforms to acceptable standards of scholarly presentation and is fully adequate, in scope and quality, as a dissertation for the degree of Doctor of Philosophy. I certify that I have read this study and that in my opm1on it conforms to acceptable standards of scholarly presentation and is fully adequate, in scope and quality, as a dissertation for the degree of Doctor of Philosophy. Mohan K. Raizada, Ph.D. Professor of Physiology I certify that I have read this study and that in my opinion it conforms to acceptable standards of scholarly presentation and is fully adequate, in scope and quality, as a dissertation for the degree of Doctor of Philosophy. /&dJ;,i1.) ~ t&t0 0 Kathleen T. Shiverick, Ph.D. Professor of Pharmacology and Therapeutics

PAGE 110

This dissertation was submitted to the Graduate Faculty of the College of Medicine and to the Graduate School and was accepted as partial fulfillment for the requirements for the degree of Doctor of Philosophy. December, 1988. Dean, College of Medicine